Full text of "American chemical journal"

See other formats

; -^^

^'■mt'.^.

Iti ^ i*"i.

><">.> ^:

UNIVERSITY OF ILLINOIS
LIBRARY

1)A0.% ^^

Book Volume

v^ ^

K '^

r*^t' V'.

4i ^ r^.- . -^^

•^ ^i .^-

CENTRAL CIRCULATION AND BOOKSTACKS

The person borrowing- this maf-en-al Jo .^

the latest j3afe stamped below. You may

eo%&v.„"r5r:st%u'*^^<~'-

Of ..Hnols and ore pZc, ' 7^:.; '';6S':r;?^ **' ''^ '""'
Law and Procedure. "' "''"°" ^"""'"''

TO RENEW, CALL (217) 333-8400

J'::!::!:;!!!!::^^^ Urbana-Champalgn

JW 2 7 2000
JAN 1 6 2001

When

below previous due date.

renewing by phone, write new due dat

e
L162

AMERICAN ^^^^

ChemicalJournal

IRA REMSEN

EDITOR

CHARLBS A. ROUILLKR

ASSISTANT EDITOR

Vol. XLVI.— July-December, 191 i

BALTIMORE: THE EDITOR

EscHKNBACH Printing Company, Printers UT
E ASTON, Pa.

CONTENTS OF VOL. XLVI

No. I.
Contributions from the Chemical Laboratory of Cornell Uni-
versity ■

Telrachlorgallein and Some of Its Derivatives. By W. R.

Orndorff and T. G. Delbridge i

A Study of th^ Conductivity and Dissociation of Organic
Acids in Aqueous Solution between Zero and Thirtv-
FrvE Degrees. By E. P. Wightman and Harry C. Jones . 56

REVIEWS.

Jahrbuch der Chemie . . . . . . . -113

Das chemische Gleichgewicht auf Grund mechanischer Vorstel-

lungen . . . . . . .114

Das Acetylen, seine Eigenschaften, seine Herstellung und Ver-

wendung . . . • "S
NOTE.

International Association of Chemical Societies . . . .116

No. 2.
Differences of Potential between Cadmium and Alcoholic

Solutions op Some of Its Salts. By Frederick H. Getraan 117
Conductivity and Viscosity in Mixed Solvents Containing

Glycerol. By J. Sam Guy and Harry C. Jones . . 131

The Reaction between Organic Magnesium Compounds and

Cinnamylidene Esters. By Grace Potter Reynolds . .198

REVIEWS.
Trattato di Chimica Inorganica Generale e applicata all' Industria 212
An Introduction to Thermodynamics for Engineering Students . 212

Introduction to General Chemistry . . . . . .213

A Laboratory Manual of Inorganic Chemistry . . . .214

Chemistry for Beginners . . . . . . . -215

A Course in Qualitative Chemical Analysis of Inorganic Substances 215
Dizionario di Merceologia e di Chimica Applicata . . . .216

A Short Hand-Book of Oil Analysis . . . . .216

No. 3.
The Reaction between Unsaturated Compounds and Organic
Zinc Compounds. By E. P. Kohler, G. L. Heritage and A. L.
Macleod . . . . . . . . . .217

2i6'V56

iv Contents

Anhydrous Formic Acid. By James B. Garner, Blair Saxton and

H. O. Parker 236

Tim Conductivities, Temperature Coefficients of Conduc-

TITITY AND DISSOCIATION OF CERTAIN ELECTROLYTES. By H.

H. Hosford and Harry C. Jones .240

The Viscosity and Fluidity of Suspensions of Finely-Divided

Solids in Liquids. By Eugene C. Bingham and T. C. Durham 278
Twe Relation of Heat of Vaporization to Other Constants
AT THE Boiling Temperature of Some Liquids at Atmos-
pheric Pressure. By Jack P. Montgomery .... 298

REVIEWS.
Zerkleinerungsvorrichtungen und Mahlanlagen
Allen's Commercial Organic Analysis .....

Die Direkte Einfiihrung von Substituenten in den Benzolkern
Elektrische Doppelbrechung der Kohlenstoflfverbindungen
The Electrical Nature of Matter and Radioactivity
Traitd de Chemie Generale .......

Qualitative Chemical Analysis, a Laboratory Guide

308
308
309
311
312
313
3I4

No. 4.
On the Color Changes Occurring in the Blue Flowers of the
Wild Chicory, Cichorium Intybus. By Joseph H. Kastle and
R. L. Haden ......... 315

Contribution from the Kent Chemical Laboratory of the
UNirERSiTY OF Chicago:

Stereoisomeric Chlorimido Ketones. By Peter P. Peterson 325

C»NWiIBUTlONS from THE SHEFFIELD LABORATORY OF YaLE UnI-
▼ERSITY:

CXCIII. — Researches on Pyrimidines: The Condensation
of Ethyl Formate and Diethyl Oxalate with Some Pyr-
imidine-Thioflycollates. By Treat B. Johnson and
Norman A. Shepard ...... 345

Re»HCTiON OP Mercuric Chloride by Phosphorous Acid and
THE Law of Mass Action. By James B. Garner, John E.
Foglesong and Roger Wilson . . . . . . -361

Thc Conductivity, Temperature Coefficients of Conductivity
AND Dissociation of Certain Electrolytes in Aqueous
Solution from 0° to 35°. Probable Inductive Action in
Solution, and Evidence for the Complexity of the Ion.
By L. G. Winston and Harry C. Jones ..... 368

REVIEWS.
Uebcr Katalyse .......... 314

Osnovi Physicheskoi Chemie . . . . . -414

Alcoholic Fermentation ........ 414

Contents v

The Fats 415

An Introduction to Bacteriological and Enzyme Chemistry . -415

Die Schwelteere, ihre Gewinnung und Verarbeitung . . .416

No. 5.
On Chlorimidoquinones. By Lemuel Charles Raiford -417

Contributions from the Sheffield Laboratory of Yale Uni-
versity:

CXCIV. — On Hydantoins: The Action of Acylthioncarha-
mates, Acyldithiocarbamates and Acylimidodithiocar-
J'onates on a-Amino Acidd 2-Thiohydantoin. By
Henry L. Wheeler, Ben H. Nicolet and Treat B. John
son ........

Unsaturated ^-Ketonic Acids. By E. P. Kohler

The BrominaTion of Phenol. By J. G. Dinwiddle and J. H

Kastle

A Study of Orthoaminoparasulphobenzoic Acid and Its Deriv-
atives, WITH Special Reference to their Fluorescence
By Joseph H. Kastle and R. L. Haden ....
The Synthesis of Fumaric and Maleic Acids from the Acetylene

DiiodidES. By Edward H. Keiser and LeRoy McMaster
The Nitrile of Fumaric Acid. By Edward H. Keiser and J. J
Kessler .........

456
474

502

50?
518

523

OBITUARY.
Albert Ladenburg . . . . . .528

REVIEWS.

Traits Complet d 'Analyse Chimique Appliqu^e aux Essais Industriels 529

New Ideas on Inorganic Chemistry . . . 530

Kapillarchemie .......... 533

The Chemistry of the Coal-Tar Dyes ...... 534

Die Konstitution der Chinaalkaloide ...... 535

No. 6.
Contributions from the Sheffield Laboratory op Yale Uni-
versity:

CXCV. — Researches on Pyrimidines: The Condensation of
Urea and Guanidine with Esters of Allylmalonic and
Some Alkyl-Suhstituted Allylmalonic Acids. By Treat
B. Johnson and Arthur J. Hill ..... 537
Contributions from the Chemical Laboratory of Harvard Uni-
versity :

i,3,5-Triiod-2-Brom-4,6-Dinitrobenzene and Some of Its

Derivatives. By C. Loring Jackson and H. E. Bigelow 549

vi Contents

The Conductivity op Certain Salts in Methyl and Ethyl Alco-
hols AT High Dilutions. By H. R. Kreider and Harry C.
Jones 574

A Study op the Hydrogen Electrode op the Calomel Elec-
trode and of Contact Potential. By N. E. Loomis and
S. F. Acree 585

The Application op the Hydrogen Electrode to the Measure-
ment OP the Hydrolysis op Aniline Hydrochloride, and
the Ionization op Acetic Acid in the Presence of Neutral
Salts. By N. E. Loomis and S. F. Acree . . .621

On Difficulties in the Use of the Hydrogen Electrode in the
Measurement op the Concentration of Hydrogen Ions in
THE Presence of Organic Compounds. By L. J. Desha and
S. F. Acree 638

Reduction of Mercuric Chloride by Phosphorous Acid and the

Law op Mass Action. By James B. Garner . . . 648

REVIEWS.

Radiumnormalmasse und deren Verwendung bei Radioaktiven

Messungen .......... 648

An Experimental Course of Physical Chemistry .... 649

Hydrocarbures Alcools et Ethers de la Serie Grasse . . . 649

Organic Chemistry for the Laboratory ...... 650

Index . . . . . . . . . . .651

Vol. XLVI July, 191 i No. i

AMERICAN

CHEMICALJOURNAL

[Contributions from the Chemical Laboratory of Cornell University]

TETRACHLORGALLEIN AND SOME OF ITS DERIVA-
TIVES

By W. R. Orndorff and T. G. Delbridge

[second PAPER^]
In the first paper it was stated that the crude tetrachlor-
gallein was completely soluble in acetone.^ It was thought,
therefore, that it might be possible to use acetone for the purpose
of purifying this product. The method was first tried on
some tetrachlorgallein which had crystallized out of 94 per
cent, ethyl alcohol on long standing. This material,^ which
had a light yellowish gray color, was dissolved in freshly
distilled acetone, the solution concentrated by distillation
and allowed to stand overnight. The next day it was found
that almost colorless crystals had separated from the red
solution. These were filtered off on a Buchner funnel, washed
with acetone, thoroughly drained and then ground in an agate
mortar until they had only a faint odor of acetone. A por-
tion was placed in a vacuum desiccator over phosphorus
pentoxide and the desiccator evacuated. The product, which
was almost white at first and well crystallized, became pink
and then red, while the loss of weight was considerable. An-
other portion of the white crystals from acetone, when allowed

1 See This Journal, 42, 183 (1909) for the first paper.

2 This Journal, 42, 208.

3 This material was probably the colorless hydrate, see page 26.

2 Orndorff and Delbridge

to stand in the air, became deep red. The same color is pro-
duced more rapidly by adding water to the crystals.

Both the colorless crystals and the red product obtained from
them gave the characteristic reactions of tetrachlorgallein, dis-
solving in dilute sodium carbonate solution with a red color and
in excess of sodium hydroxide with a blue color. Both products
dissolve in alcohol or acetone with a pink color which turns
red on the addition of water.

A marked difference between the colorless crystals and the
red product is the fact that the crystals are fairly soluble in ether
while the red compound is practically insoluble in this solvent.

A small portion of the colorless crystals was placed in a
platinum boat and heated to iio° in a current of carbon
dioxide in the apparatus elsewhere described.^ The out-
flowing carbon dioxide was passed through Hugershoff gas
wash bottles containing distilled water. The water was then
found to contain acetone, as shown by its odor, by the forma-
tion of large amounts of iodoform on treatment with am-
monium hydroxide and iodine-potassium iodide solution,
and finally by the formation of acetone phenylhydrazone
on treatment with phenylhydrazine hydrochloride and sodium
acetate. The substance in the boat came to constant weight
after seven hours' heating at iio° and there was no further
loss on heating for one hour at 156°. The loss of weight was
18.90 per cent, and the fact that no moisture condensed on
the cold part of the tube indicated that this loss was due
entirely to acetone, past experience having shown that even
0.5 per cent, of water in a compound leads to the deposition
of moisture on this part of the tube. This loss of weight
agrees closely with that required by the formula C20H8O7CI4.
2CH3COCH3 for which the calculated loss (acetone) is 18.79
per cent.'- The chlorine content of the original white crystals
is also in accord with the above formula:

Calculated for Found

Ca.H.^OrCl,. zCH.jCOCH:, Chlorine

Chlorine Percent.

Per cent. I II

22.95 22.83 22.75^

1 This Journal, 41, 403.

-' International Atomic Weights for 1911 have been used in all calculations.
3 All chlorine determinations were made by the method described in This Jour-
nal, 41, 393.

Tetrachlorgallein and Some of Its Derivatives 3

The colorless crystals become intensely red when kept
in vacuo over phosphorus pentoxide for several days. The
loss of acetone is gradual and continuous, however, and no
tendency to remain at constant weight was noted although
the substance was kept in vacuo for over 75 days. It should
be mentioned that the red color was most intense when the
loss in weight corresponded with that calculated for one mole-
cule of acetone. From this point the color became fainter
until finally the compound had very little color.

Further work on these colorless crystals showed the correct-
ness of the above formula and hence we shall call the compound

Tetrachlorgallein Diacetonate

In making the second lot of this compound, v/e used the
crude tetrachlorgallein, thoroughly washed first with dilute
hydrochloric acid, then with v/ater, and air-dried at room
temperature. Fifty grams of this material were dissolved in»
freshly distilled acetone which had been previously dried over
calcium chloride, and the solution filtered and concentrated.
On standing, a yellowish white crystalline compound separ-
ated. This was filtered off and crystallized three times from
dry acetone. It was thus obtained as an almost colorless,
beautifully crystallized product. It was filtered off with suction
on a Buchner funnel and transferred to a porcelain dish,
thoroughly mixed with pure acetone and again filtered. This
treatment was then repeated and the product drained as dry
as possible on the Buchner funnel. This material, which
was practically white, was placed in a desiccator and the air
pumped out as quickly as possible. Fresh air was then ad-
mitted, the material stirred thoroughly and the desiccator
again evacuated. This process v^^as repeated eight times,
after which the material had no odor of acetone, but was
still crystalline and almost white, having only the faintest
pink tinge. It v^^as transferred to a tightly-stoppered bottle
and kept in a cool place. Some of this substance was first
heated to 110° in a current of carbon dioxide in the apparatus
described.* It came to almost constant weight after eleven hours

' This Journal, 41, 403.

4 Orndorfj and Delbridge

and much acetone was evolved but no water. The temperature
in the apparatus was then raised to 157° and the substance
came to constant weight in four hours, the results being as
follows :

Loss at 110°

Loss at 157'

Grain Gram Per cent. Gram Per cent.

0.4090 0.0765 18.70 0.0768 18.78

0.4643 0.0867 18.67 0.0869 18.72

Calculated for C2oH807Cl,.2CH3COCH3 18 . 79

The product heated to 157° had a faint pink color and was
anhydrous tetrachlorgallein, as shown by the following chlorine
determinations :

Substance
Gram

Tenth-normal
silver nitrate

CO.

Chlorine
Per cent.

0 . 3090

24.50

28.12

O.I914

I5I3

Calculated for C20H3O7CI,

28.03
28.26

In order to determine whether these low results were due
to loss of chlorine or hydrochloric acid at 157°, two more
determinations were made, this time on the original substance.
Although this material had been kept in a cool place, it seemed
to be losing acetone. Hence the loss on heating was again
determined and found to be 18.25, 18.34, 18-23 and 18.32
per cent.: mean, 18.28 per cent. Taking this mean as the
acetone content, the original substance was analyzed for chlorine
by both the gravimetric and volumetric methods, calcu-
lating the chlorine for tetrachlorgallein:

Tenth-
Tetra- normal

chlorgal- silver Silver

Substance lein nitrate Chlorine chloride Chlorine

Gram Gram cc. Per cent. Gram Per cent.

I 0.3134 0.2561 20.25 28.04 0.2904 28.05

II 0.3400 0.2778 21.96 28.03 0.3139 27.95

Calculated for CaoHgOyCl^ 28 . 26

These results agree very closely with those obtained on
the material heated to 157° and prove that the substance
loses neither chlorine nor hydrochloric acid at this temperature.
Later, after we had discovered the modification of tetrachlor-

Tetrachlorgallein and Some of Its Derivatives.

gallein which we have called the carbinolcarboxylic acid^
and which contains only 27.28 per cent, of chlorine, we con-
cluded that these low results on chlorine were due to the pres-
ence of a very small amount of this carbinolcarboxylic acid
in the diacetonate. The formation of this carbinolcarboxylic
acid was doubtless due to the condensation of moisture on
the diacetonate when its temperature was lowered by the
rapid evaporation of the mechanically held acetone.'^

The third lot of diacetonate was made from the same crude
tetrachlorgallein as the first two, but the product had already
been crystallized once from acetone. Twenty grams of this
material were dissolved in 500 cc. of boiling acetone which
had been previously carefully dried. The solution, which was
.claret-red, was filtered, concentrated to 300 cc, chilled rapidly
and stirred occasionally. The white crystalline product
which separated after a few minutes was filtered off with suc-
tion, washed with anhydrous acetone and dried in the same ■
manner and with the same precautions used in making the
second lot of the diacetonate^, the object being to obtain the
compound as free as possible from acetone held mechanically.
This product, which was practically white, having only a sug-
gestion of a faint pink tinge, was immediately bottled and
analyzed.

Substance Loss at 157° Acetone

Gram Gram Per cent.

0.4606 0.0868 18.85

0.5014 0.0945 18.85

Calculated for C3oHACl4.2CH3COCH3 i8 . 79

The chlorine determinations were made on the product
heated to constant weight at 157° in carbon dioxide. This
compound had only a slight pink tinge and is anhydrous
tetrachlorgallein.

Substance
Gram

Tenth-normal

silver nitrate

cc.

Chlorine
Per cent

0.2975
0.2700

23 56

21.47

Calculated for C^oHgO^Cl,

28.08
28.20
28.26

See page 37.
Page 3.
Page 3.

6 Orndorff and Delhridge

From the close agreement of all the results, we were con-
vinced that the diacetonate is a definite compound. Tetra-
chlorphthalic acid itself crystallizes with two molecules of
acetone* and a tetrachlorgallein hydrochloride containing
acetone has also been prepared.- It was therefore thought
necessary to determine the amount of acetone in the diacetonate,
in other words, to show that the loss of weight on heating to
157° was due entirely to loss of acetone. After experimenting
with various methods for determining acetone, we finally
adopted that of Messinger^ as modified by Collischonn.^ The
procedure is as follows: The acetone solution is made up to
about 250 cc. and 20 cc. of normal sodium hydroxide added.
Tenth-normal iodine solution is then run in from a burette,
drop by drop with constant shaking, until about 20 per cent,
excess is present. The flask is then closed with a stopper,
shaken for four minutes and acidified vvdth five cc. of 6-normal
hydrochloric acid. A measured excess of tenth-normal sodium
thiosulphate solution and four cc. of starch solution (one gram
of arrowroot starch in 500 cc. of boiling water) are added
and the excess of thiosulphate titrated. The thiosulphate
solution was standardized with pure iodine and the iodine
solution was titrated against the thiosulphate under exactly
the conditions given above except that no acetone was used.
Acetone has a molecular weight of 58.048 and is equivalent to
six atoms of iodine. Hence, each cc. of tenth-normal iodine so-
lution is equivalent to 0.0009675 gram of acetone. The
method was first tested with vpure acetone. About 60 cc.
of distilled water in an Erlenmeyer flask covered with a watch-
glass were carefully weighed, a little more than one cc. of
acetone (from the bisulphite compound, Kahlbaum's) added
and the flask containing the solution weighed at once. The
increase in weight was i . 148 grams, which represents the amount
of acetone in the solution. This solution was then diluted
to one liter and measured portions analyzed with the following
results :

' This Journal. 41, 410.

2 This Journal, 42, 212. Heller and Langkopf have also made a ".Salzsaures
Galleinacetonat." Z. Farb. Ind., 6, 267.

3 Bar. d. chem. Ges., 21, 3366 (1888).
* Z. anal. Chem., 29, 562 (1890).

Tetrachlorgallein and Some of Its Derivatives

Acetone solu-
tion used
cc.

Tenth-normal

iodine Acetone
cc. Grams per liter

20.49
15.06

24.17
17.86

1. 141
1. 147

17-35

20.36

I -135

15-99

18.99

Acetone actually present

1. 149

1. 148

The determination of the acetone in the tetrachlorgallein
diacetonate was carried out as follows : About 0.6 gram of the
substance in a platinum boat was weighed in a glass-stoppered
weighing tube. The boat was then placed in the apparatus
already described^ and two Hugershoff gas wash bottles
attached. These were filled with distilled water and packed
in ice. The substance in the boat was then heated for one
hour at 157° while a slow current of carbon dioxide was passed
over the compound to drive out the acetone. The contents
of the first wash bottle were then transferred to a 500 cc.
measuring flask, the bottle thoroughly washed out with dis-
tilled water, the washings added to the measuring flask and
the solution finally made up to 500 cc. Fifty cc. portions
were titrated and the mean of three determinations was taken.
The contents of the second wash bottle were also titrated,
giving a slight addition. The boat with its contents was
weighed again after the heating and the loss of weight de-
termined. The results obtained were as follows:

Substance Loss at 157° Acetone

Gram Gram Per cent.

I 0.5814 0.1088 18.71

II 0.6625 0.1244 18.78

Calculated for aoH807Cl,.2CH3COCH3 1 8 . 79

The solution in the first Hugershoff bottle was made up to
500 cc. and three 50 cc. portions titrated, (a), (b) and (c) :

Tenth-normal iodine solution

First bottle

Second bottle

Acetone

cc.

Per cent

(a)

11.24

0.55

18.80

(b)

II . 19

0.55

18.71

(c)

II .22

0.55

18.75

Mean

18.75

This Journal, 41, 403.

Orndorff and Delhridge

Tenth-normal iodine solution

First bottle

Second bottle

Acetone

cc.

I'er

cent.

(a)

12.84

0.41

ih)

12.82

0.41

(c)

12.88

0.41

Mean

Calculated for C20H8O7CI,

.2CH3COCH3

Hence the loss of weight at 157° is due entirely to loss of
acetone and the compound is properly called tetrachlorgallein
diacetonate.

The process of preparing the diacetonate has been described
in some detail because a number of later attempts to make
this substance from the crude tetrachlorgallein, as well as
from the anhydrous product, resulted in failure. Mr. T. R.
Briggs found that the diacetonate can readily be obtained
from the light-colored tetrachlorgallein hydrate which crys-
tallizes out of methyl alcohol,^ Analyses of this product
made by Mr. Briggs show that it is the colorless hydrate
containing a small amount of the red hydrate :'

Substance
Gram

Loss at 158'=
Gram

0.0107

Water
Per cent.

3-79

Calculated for C20H8O7CI1.H2O 3 . 47

Substance
(dry)
Gram

0.2714

Tenth-normal
silver nitrate

21-59
Calculated for CooHgO^CI,

Chlorine
Per cent.

28.21

28. 26

About 20 grams of this mixture of hydrates were dissolved
in 500 cc. of redistilled anhydrou acestone, and the solution
filtered and concentrated by distillation to 100-150 cc, when
crystals began to separate. These were filtered off, washed
with acetone and recrystallized from anhydrous acetone. The
crystals thus obtained, which were almost pure white, lost
acetone on standing in the air, and took up water, forming the

1 This Journal, 42, 210.

2 See pages 24' and 26.

Teirachlorgallein and Some of Its Derivatives 9

red hydrate. Analyses, by Mr. Briggs, of the white cr>'stals
gave the following results ;

Substance Loss at 158° Acetone

Gram Gram Per cent.

0.3956 0.0738 18.66

Calculated for aoH807Cl4.2CH3COCH3 i8 . 79

Substance Tenth-normal

(heated to 158°) silver nitrate Chlorine

Gram cc. Per cent.

0.3218 25.50 28.10

Calculated for CooHgO^Cl, 28.26

In order to observ^e the conduct of the diacetonate on stand-
ing in the air, a weighed amount of the substance was spread
out on a watch-glass and protected from dust. Weighings
were made at inter\'als until constant weight was reached,
64 days being required. The actual loss was 14.48 per cent.
The resulting product retained the crystalline form of the*
colorless diacetonate but had a red color. A portion of this
colored substance was then heated to 157° until it came to
constant weight. The loss was 4.41 per cent., of which 0.62
per cent, was due to loss of acetone,^ and 3.79 per cent, to
loss of water. Furthermore, the substance, after being heated
to 157°, weighed 0.6 per cent, more than the calculated weight
of the tetrachlorgallein in the original diacetonate, proving
the formation of some of the carbinolcarboxylic acid described
below. ^ Hence the diacetonate gradually loses most of its
acetone in the air, taking up somewhat more than one molecule
of water to form the colored tetrachlorgallein hydrate and a
small quantity of the colorless tetrachlorgalleincarbinol-
carboxylic acid.

The formation of the diacetonate may be represented as
follows :

1 Determined as in the case of the diacetonate; see page 7.

2 Page 37.

Orndorf} and Delhridge

Tetrachlorgallein and Some of Its Derivatives 1 1

This formula for the diacetonate is in accord with all the
facts known. It represents the compound as a colorless
substance and if we suppose that the first molecule of acetone
it loses is the one which joins the quinoid grouping, the re-
sulting product would be

OH ^ OH

HOr

^\C(CH3)3

This has the quinoid structure and would be colored. It
will be remembered that the colorless diacetonate gradually
loses acetone and becomes colored when placed in a desiccator
over phosphorus pentoxide and that the color is deepest red
when the diacetonate has lost one molecule of acetone. From
this point the color becomes fainter and the product which
results from heating the diacetonate to 157° and which no
longer contains any acetone has almost no color.' It is prob-
ably the lactoid form of tetrachlorgallein :

1 The same product results when the colorless or the colored hydrate is heated
to 157 ».

Orndorff and Delbridge
OH ^ OH

HOi

Tetrachlorgallein (anhydrous)

In this connection it may be well to call attention to the
fact that the diacetonate of tetrachlorphthalic acid loses acetone
when a current of air dried with calcium chloride is passed
over it and the product left is the anhydrous tetrachlorphthalic
acid.^ On standing in the air or on boiling with water, this
diacetonate loses all of its acetone and takes up a half -molecule
of water, forming tetrachlorphthalic acid, C6Cl4(COOH)2.
o-sH^O.^

The diacetonate of tetrachlorgallein takes up water when
exposed to the air and forms the colored hydrate and a small
amount of the carbinol acid :

2 This Journal, 41, 410.
2/6«/.. 41, 411.

Tetrachlorgallein and Some of Its Derivatives 13

.0-C(CH3)2

OH Q OH

+ 3H2O

/^\C(CH,),
C^O-^

OH Q^ OH

HO,

OH
C]/ \,C^OH

CI
Colored Hydrate

HOf

\0H

ar N,c^oH

4-4CH3COCH,

;ci

Carbinol Acid

14 Orndorff and Delbridge

It is probable that here, too, the first molecule of acetone
lost is the one connected with the quinoid grouping. This
explains in a very satisfactory manner why, when the second
molecule of acetone is replaced by water, the main product
of the reaction is the colored hydrate.

Tetrachlorgallein Eiherate

Twenty grams of the pure tetrachlorgallein diacetonate
were shaken vigorously for two minutes with 300 cc. of anhy-
drous ether, which had been previously dried with phosphorus
pentoxide and distilled. The greater part of the substance
dissolved, forming a reddish solution. This was filtered
at once through a folded filter and the filtrate, which was
perfectly clear, allowed to stand in a stoppered flask. Within
two minutes a perfectly white, beautifully crystalline pre-
cipitate began to separate and in the course of fifteen minutes
about 15 grams of the material had accumulated. The mixture
was allowed to stand overnight at room temperature in the
tightly-stoppered flask, and the crystals filtered off the next
day, washed thoroughly with anhydrous ether and stirred in
a crystallizing dish until the ether held mechanically by the
crystals had entirely evaporated. They were then placed
in a bottle, tightly corked and kept in the ice box. A weighed
portion was heated to 100° for half an hour in a stream of
carbon dioxide. There was no condensation of moisture
on the cold part of the tube (F) such as is always noted when
compounds containing water are heated in the apparatus,^
but the odor of ether was very marked and persisted to the end
of the heating. The loss in weight of the compound at this
temperature was 9.7 per cent. The outflowing carbon dioxide
was passed through HugershofiF gas wash bottles containing
distilled water to remove any acetone which might have been
present in the gas. The water, however, gave only an ex-
tremely minute quantity of iodoform when tested for acetone
and a quantitative determination of the amount present,
carried out by the method described under tetrachlorgallein
diacetonate,- gave only 0.034 P^r cent., an amount so small

' See This Journal, 41, 404.
2 See page 7.

Teirachlorgallein and Some of Its Derivatives 1 5

as to be entirely negligible. The substance had acquired
a faint pink color after the half -hour's heating to 100° and
further heating to 157° for an hour and a half did not increase
this color, though the total loss of weight at this temperature
amounted to 10.4 per cent. A second lot of the etherate
was made, using the same quantity of the diacetonate and
the same precautions detailed above, only the product was
simply stirred in the crystallizing dish until it showed no
further tendency to form lumps. It was then bottled and
analyzed. The following are the results obtained on the
two products :

Substance Loss at 157° Ether
Gram Gram Per cent.

0.5766 0.0602 10.4

0.8529 0.0970 I I. 4

Calculated for C^oHACl^.QHioO 12.9

The compound loses its ether very rapidly indeed and
for this reason cannot be weighed in an open dish. The re-
sults, however, show that it contains approximately one
molecule of ether.

A chlorine determination was made on the product obtained
by heating the etherate to 157°.

Substance

Tenth-normal

(heated to 157°)

silver nitrate

Chlorine

Gram

cc.

Per cent.

0.2594 20.30 27.75

Calculated for a mixture of equal parts of
the anhydrous tetrachlorgallein and the
carbinolcarboxylic acid 27.77

It is evident from this result that the compound heated
to 157° is a mixture of the anhydrous tetrachlorgallein, which
contains 28.26 per cent, chlorine, and the carbinolcarboxylic
acid^ containing 27.28 per cent. The formation of the latter
is presumably due to the condensation of moisture on the
compound cooled by the rapid evaporation of the ether. It
will be remembered that the diacetonate also gave low results for
chlorine due to the same cause. ^

About three grams of the same etherate (II), which lost

1 See page 37.

2 See page 4.

l6 Orndorff and Delbridge

1 1.4 per cent, in weight when heated to 157°, turned pink
when exposed to the air and came to constant weight after
six days with a loss of only 7.2 per cent. There was no furtlier
loss of weight, although the product stood in the air for two
months after this. Hence it seems probable that all the ether
had evaporated and that the difference, 4.2 per cent., be-
tween the two losses in weight (i 1.4 — 7.2) is due to absorption
of water to form a mixture of hydrate and carbinol acid, as
in the case of the diacetonate.^ While the product from the
latter, however, is red, that from the etherate has only a faint
pink tinge. This difference is due to the fact that the di-
acetonate combines with water to form the colored hydrate
and a small amount of the carbinol acid while the etherate
gives the colorless hydrate- and the carbinol acid mixed with
sufficient of the colored hydrate to produce the faint pink color.
Analyses^ of this etherate (II), which had stood in a museum
bottle for over a year, gave the following results:

Substance

Loss at 157°

Water

Gram

Per cent.

0.2539

0 . 0099

3-90

0.2612

0 . 0098

3-75

0.2677

0.0103

3.85

Calculated for C^oHgO^Cl,

.H3O 3.47

Substance

Tenth-normal

(dry)

silver nitrate

Chlorine

Gram

cc.

l\'r cent.

0 . 2440

19.20

27.90

0.2514

1975

27.86

0.2574

20.21

27.85

Calculated for a mixture of equal parts of
anhydrous tetrachlorgallein and the car-
binolcarboxylic acid 27 . 77

As the etherate is a colorless compound and is formed
from the colorless diacetonate, it probably has the lactoid
structure, like the anhydrous tetrachlorgallein resulting from
heating the diacetonate to 157°. The ether must be repre-
sented as very loosely combined since it is so readily given
off. The following structural formula is suggested for the
compound :

> See page 9.

2 See page 26.

' These analyses were made by Mr. T. R. Briggs.

Tetrachlorgallein and Some of Its Derivatives
OH ^ OH

HOi

iOH

!C1

.0:

AH5

^C,H,

Tetrachlorgallein Etherate

That the ether attaches itself to the carbonyl group of the
phthalic acid residue and not to the quinoid grouping seems
very likely, first, from the fact that the etherate loses ether
and takes up water, when exposed to the air, to form a mix-
ture of the colorless hydrate and the carbinol acid' and, secondly,
from the fact that the methyl ester of tetrachlorgallein forms
a colored etherate- which must have the following formula:

Etherate of Methyl Ester of Tetrachlorgallein (colored)

1 See page 37.

2 This Journal, 42, 237.

Orndorf} and Delbridge

Hence the ether must attach itself to the carbonyl group in
the phthalic acid residue and not to the quinoid grouping,
otherwise this etherate would be colorless.

Tetrachlorgallein etherate loses ether and takes up water
when exposed to the air, forming a mixture of the colorless
hydrate and the carbinol acid :

iOH

+ 3HA

CI
Colorless Hydrate

Tetrachlorgallein and Some of Its Derivatives 19

HOi

+ 2(CH,)20.

CI
Carbinol Acid

Tetrachlorgallein Hydrochloride ^

It has been shown that the hydrochloride of tetrachlor- '
gallein can be readily made by passing dry hydrochloric acid
gas into a concentrated solution of tetrachlorgallein in dry
acetone. The product thus obtained, however, contains
acetone and it was found to be impossible to accurately de-
termine the amount, probably owing to the action of the
hydrochloric acid on the acetone. The anhydrous tetrachlor-
gallein, made by heating the pure diacetonate to 157° for the
purpose of determining the amount of acetone it contained,^
was found to take up dry hydrochloric acid gas very readily,
becoming first red, then dark red and finally green. This
green color, which was also observed in the case of the hy-
drochloride containing acetone, is a surface color, as the powd-
ered material is red. The method of preparing the hydrochlor-
ide was as follows: A weighed amount of the anhydrous
tetrachlorgallein in a boat was placed in a glass tube and
hydrochloric acid gas, dried first with 99 per cent, sulphuric
acid and then with phosphorus pentoxide, was passed over
it for several hours. The boat containing the hydrochloride
was then removed and weighed in a tightly-stoppered weighing
tube. This process was repeated until the substance came

1 This Journal, 42, 212.
- See page 5.

20 Orndorff and Delhridge

to constant weight, which required about sixteen hours"
exposure. The amount of hydrochloric acid absorbed by the
tetrachlorgallein is then given directly by the increase in
weight.

TeUachlor- Time

Gain in

Hydrochloric

gallein exposed

weight

acid

Experinient

Gram Hours

Gram

Per cent.

0.4412 20

0.0326

6.88

0 . 4809 20

0.0352

6.82

III

0.4752 16

0.0351

6.88

0.5838 16

0.0436

6.94

0.4752 18

0.0350

6.86

0.4286 72

0.0320

6.95

Calculated for C^o

H,0,C1,.HC1

6.77

In order to determine whether the increase of weight was
due entirely to the addition of hydrochloric acid, the tetra-
chlorgallein hydrochloride obtained in Experiments III and
IV was placed in a tube and heated to 157° in a stream of
carbon dioxide which was afterward passed through two
Hugershoff gas wash bottles containing distilled water. The
hydrochloric acid in this water was then determined by ti-
tration, with the following results :

Tetrachlorgallein Tenth-normal Hydrochloric

hydrochloride potassium hydroxide acid

Gram cc. Per cent.

Illa 0.5103 9.54 6.82

IVa 0.6274 11-65 6.77

Calculated for C^oHgO^Cl,. HCl 6.77

The material left in the boat was anhydrous tetrachlor-
gallein, its weight in both cases being exactly the same as
that of the original tetrachlorgallein used in preparing the
hydrochloride. Indeed, the anhydrous tetrachlorgallein used
in Experiment V was actually that obtained in III;: by heating
the hydrochloride to 157° to drive off the hydrochloric acid.
These results show that anhydrous tetrachlorgallein is con-
verted by the action of dry hydrochloric acid gas into tetra-
chlorgallein hydrochloride which, when heated to 157°, loses
hydrochloric acid quantitatively with the formation of the
original tetrachlorgallein. The hydrochloride is also decom-
posed quantitatively by hot water with the fonnation of

Tetrachlorgallein and Some of Its Derivatives 21

hydrochloric acid and the red tetrachlorgallein hydrate.^
As the tetrachlorgallein hydrochloride is intensely colored,
it must be a quinoid compound, as represented in the formulas
given for this substance in the first article.^

Tetrachlorgallein Hydrate (Colored)

This substance, made from the pure monosodium salt by
decomposing it with hydrochloric acid, has already been de-
scribed.^ It is insoluble in chloroform, benzene or ether,
but dissolves readily in dry acetone, forming a practically
colorless solution. Benzene and ether cause no precipitation
when added to this solution but chloroform gives, on standing,
a white precipitate of the colorless hydrate.* The addition
of these solvents to the acetone solution produces no color.
When diluted slightly with distilled water, however, the
acetone solution becomes pink and turns cherry-red on further
dilution, depositing a white precipitate of the carbinolcarbox^/lic '
acid. If the slightest trace of acid, even of acetic acid, be
added to the water used to dilute the acetone solution, no
color results. Methyl and ethyl alcohols, also dissolve the
hydrate with a faint pink color, and dilution with distilled
water gives a red color unless the water is previously acidified.
Further dilution with water precipitates the carbinolcarboxylic
acid.

It was thought desirable to determine whether the blue
tetrasodium salt of tetrachlorgallein would give the same red
hydrate^ as the monosodium salt. Twelve grams of the pure
monosodium salt were therefore slowly added to an actively
boiling solution of 2.5 grams of sodium hydroxide in 2.5 liters
of water, in a Jena glass flask. By keeping the solution boiling
vigorously, the air was excluded and oxidation prevented.
After solution was complete, an excess of dilute hydrochloric
acid was added to the boiling solution. The precipitate
formed had the same color and appearance as the red hydrate

' See page 22.

' This Journal, 42, 258.

3 Ibid., 42, 220. See also p. 255 for its structural formula.

* See page 26.

6 This Journal, 42, 220.

22 Orndorff and Delbridge

from the monosodium salt. It was filtered off, washed free
from chlorides with hot water, dried in the air and analyzed.

Substance

Loss at 157"

Water

Gram

Per cent.

0.3345

O.OI2I

3-62

0.2966

0.0108

3 64

0.3568

0.0127

356

Calculated for C^oHgO^d.-Hp

3-47

Substance

Tenth-normal

(dry)

silver nitrate

Chlorine

Gram

cc.

Per cent.

0.2830

22.51

28.21

0.2899

23.00

28.13

Calculated for CoHgO^Cl,

28.26

Hence this product is also the colored hydrate. Its proper-
ties, too, show that it is identical with the red hydrate made
from the monosodium salt.

As has already been stated,* the hydrochloric acid salt
of tetrachlorgallein also yields the red hydrate. This hydro-
chloride was decomposed by boiling with water for thirty
minutes, and the reddish precipitate was filtered off, washed
free from hydrochloric acid, dried in the air and analyzed.

Substance
Gram

Loss at 157°
Gram

Water
Per cent.

0.5375

0.0207

3-85

0.3061

0.0122

Calculated for C20H3O7CI,

.H3O

3 99

3-47

Substance
(dry)
Gram

Tenth-normal

silver nitrate

cc.

Chlorine
Per cent.

0.2934

23 23

28.08

0.2554

20.33

28.23

Calculated for C20H8O7CI, 28.26

The properties of this product, as well as the analyses,
show clearly that it is the colored tetrachlorgallein hydrate.

In determining the number of acetyl groups in tetrachlor-
gallein tetracetate- by the modified Wenzel method, a product
was obtained which seemed to be the red hydrate. This mate-

1 Page 2 1 .

2 This Journal, 42, 227.

Tetrachlorgallein and Some of Its Derivatives 23

rial was therefore investigated. Ten grams of the pure acetate
were treated with 200 cc. of sulphuric acid (2 parts of pure
concentrated sulphuric acid to one of distilled water) and
the mixture heated in a boiling water bath for thirty minutes.
The solution was then poured into cold water and the resulting
precipitate washed by decantation six times with 800 cc. of
distilled water. It was then filtered off, dried and placed
in the apparatus used in determining acetyl groups. Thirty
cc. of sulphuric acid were added and the mixture heated for
a half-hour. Then 30 cc. of distilled water and 250 cc. of
phosphoric acid mixture were added and the solution distilled
in steam for one hour. No acetic acid was found in the dis-
tillate. Hence the saponification of the acetate was complete
in the first operation. The mixture was poured into distilled
water and the precipitate allowed to settle. The supernatant
liquid was filtered and tested for hydrochloric acid but was
found to contain none, so that no decomposition of the tetra-
chlorgallein had taken place. The precipitate was thoroughly
washed with hot water and dried in the air. It was red and
resembled the colored tetrachlorgallein hydrate in its chem-
ical and physical properties, although it was distinctly darker
in color than the red hydrate precipitated from the alkaline
solution with hydrochloric acid.^ Analyses, however, showed
that this compound is really the colored hydrate containing a
small quantity of the carbinolcarboxylic acid :

Substance
Gram

Loss at 157° Water
Gram Per cent.

0.2728

0.0092 3.37

0.3019

0.0103 3.41

Calculated for C^oHgO^Cl^.H^O 3 . 47

Substance
(dry)
Gram

Tenth-normal
silver nitrate Chlorine
cc. Per cent.

0.2915

23.07 28.06

0.2729

21.56 28.01

Calculated for C.oHgO^Cl, 28 . 26

Some of the pure tetrachlorgallein diacetonate was dried
at 170° and crystallized from 94 per cent, alcohol which was

Orndorff and Delbridge

free from both acid and alkali. The crystals appeared almost
colorless when observed with the magnifying glass in the
highly colored solution. They were filtered off, washed with
alcohol and with ether and dried in the air. The product was
then reddish brown in color and gave all the reactions for
tetrachlorgallein. It contained no ethyl ester, as it gave no
color when treated with ether. ^ When heated to 1 5 7 ° in carbon
dioxide for 12 hours, it gave off water but no alcohol or ether.

Substance
Gram

Loss at 157°
Gram

Water
Per cent.

0.4429

O.OI5I

341

0.2947

0 . 0099

Calculated for C,oHs07Cl,.H30

3.36

3-47

Substance
(dry)
Gram

Tenth-normal
silver nitrate

CO.

Chlorine
Per cent.

0.2848

22.56

28.09

0.2761

21 .91

Calculated for CooHgO^Cl^

28.13
28.26

This substance is therefore the colored hydrate.

In order to obtain crystals for examination, some of the
pure colored tetrachlorgallein hydrate was dissolved in 94
per cent, alcohol, and the solution filtered, heated to boiling
and diluted with water very slowly. Fairly good crystals
were thus obtained. They were submitted to Professor Gill,
of the Mineralogical Department, who reports as follows :

Tetrachlorgallein Hydrate (Colored). — "The crystals occur
as minute rhomboidal plates, the acute plane angle of the
rhomboid measuring about 67°. Extinction takes place
nearly parallel and normal to the shorter side of the rhomboid,
less optical elasticity being the normal to this side. In con-
vergent polarized light one axis appears eccentric near the
edge of the field. From this optical behavior the crystals
must be triclinic. They show a slight brownish red color and
weak pleochroism. Possibly, if seen on edge, the color and
pleochroism would be stronger. In mass the crystals are
distinctly and strongly colored."

When a hot solution of the tetrachlorgallein diacetonate

' The ethyl ester, like the methyl ester, dissolves in ether with a red color.

Teirachlorgallein and Some of Its Derivatives

in acetone is poured into boiling water, a reddish violet product
results. This was investigated and found to be a mixture
of the red hydrate and the carbinolcarboxylic acid. Analyses
of the air-dried material gave the following results :

Substance
Gram

0.3668 0.0137 3.73

0.4364 0.0162 3.71

Calculated for C2oH807a,.H20 3 . 47

No. II was afterwards heated for two hours at 175° and
then two hours more at 202 ° but lost no further weight.

Substance
(dry)
Gram

Tenth-normal
silver nitrate

2351

21 .66

I o . 3000

II 0.2760

III 0.2524 19.73

Calculated for a mixture of equal parts of
anhydrous tetrachlorgallein and the car-
binolcarboxylic acid

Chlorine
Per cent.

27.79

27.83

27.72

7-77

Analysis III was made by the Pringsheim method.

A product very similar to this one was also obtained by
the following method: 1.5 grams of sodium hydroxide purified
by alcohol were dissolved in 900 cc. of distilled water, and the
solution filtered and cooled to 0°. Five grams of pure tetra-
chlorgallein diacetonate were then added with vigorous stirring.
The solution was filtered as rapidly as possible and acidified
at 3° with dilute acetic acid. A dirty reddish precipitate
was thrown down. This was filtered off, washed thoroughly
and dried at room temperature. It was then analyzed :

Substance
Gram

0.3809

Loss at 157°
Gram

Water
Per cent.

0.0168 4.41

Calculated for C2oH807Cl,.H.O 3 . 47

Substance
(dry)
Gram

Tenth-normal
silver nitrate

Chiorine
Per cent.

0.2845 22.28 27

Calculated for a mixture of equal parts of
anhydrous tetrachlorgallein and the car-
binolcarboxylic acid 27.77

26 Orndorff and Delbridge

This product, like the one made by saponifying the te trace tate
with cold concentrated sulphuric acid and pouring the mixture
into cold water/ is somewhat soluble in ether while the red
hydrate is practically insoluble in this solvent. When the
above alkaline solution was precipitated by a mineral acid
instead of acetic acid' the red hydrate insoluble in ether was
obtained under all conditions.

The carbinolcarboxylic acid also gave a product similar
to the last two described. Fifteen grams of the pure compound
were dissolved in 500 cc. of acetone, the solution filtered,
heated to boiling and the acetone distilled off by passing in
steam. A light reddish crystalline precipitate separated.
This was filtered off and washed with water. It dried very
quickly in the air. The following analyses show that it is a
mixture of the red hydrate and the carbinolcarboxylic acid:

Substance
Gram

Loss at 157° Water
Gram Per cent.

0.4218

O.OI5I 358

Calculated for C.,oH Ad-i-Hp 3 • 47

Substance
(dry)
Gram

Tenth-normal

silver nitrate Chlorine
cc. Per cent.

0.2777 21.88 27.94

Calculated for a mixture of three parts of
anhydrous tetrachlorgallein and one of the
carbinolcarboxylic acid 28.01

Tetrachlorgallein Hydrate (Colorless)

This product was obtained in an attempt to prepare the
carbinolcarboxylic acid in a crystallized condition. Sixteen
grams of the pure, air-dried red tetrachlorgallein hydrate
were dissolved in 250 cc. of freshly distilled acetone, the reddish
brown solution faltered, cooled to 25° and about 20 cc. of dis-
tilled water added. No precipitate was formed. A slow
stream of moist carbon dioxide, free from mineral acid, was
then passed through the solution to remove some of the acetone
and prevent oxidation. After several days silvery white
needles crystallized out. This product was at first thought

' See Thls Journal, 42, 229.
2 See page 37.

Tetrachlorgallein and Some of Its Derivatives

to be the carbinol acid. Analyses/ however, showed that it
was a colorless hydrate of tetrachlorgallein.

Substance
Gram

Loss at 157°
Gram

Water
Per cent.

0.5398

0.0192

3-56

0.5491

0 . 0204

Calculated for C2oH80,Cl,.H20

3.72
3-47

Substance
(dry)
Gram

Tenth-normal

silver nitrate

cc.

Chlorine
Per cent.

0.2675

Calculated for C^oH^O^Cl,

28.29
28.26

Another sample of the same product was dried to constant
weight at 157° in a stream of dry hydrochloric acid gas and
then heated in carbon dioxide at the same temperature until
it ceased to lose weight.^

Substance
Gram

o. 1609

Loss at 157°
Gram

o . 0056
Calculated for C^oHACl^.H^O

Water
Pe» cent.

3.4«
3-47

The chlorine in this product dried at 157° was then determined :

Tenth-normal
silver nitrate

0.1543

12.36 28.40

Calculated for C^oH^O^Cl,, 28.26

The crystals of the colorless hydrate were examined by
Professor Gill who reports as follows: "Long colorless needles,
the larger ones somewhat flattened. The end face makes
an angle of about 80° with the long direction. Extinction
angle about 35° to 38°, less optical elasticity lying in the
direction of the arrow I / • The crystals were not large enough
to give good optical figures in converged polarized light, but
it could be seen that the substance is biaxial. It is probably
triclinic, as all needles show inclined extinction."

This colorless hydrate is present in the crude tetrachlor-
gallein together with the red hydrate, and as it is less soluble

1 These analyses were made by Mr. E. H. Nichols.

2 These analyses were made by Mr. E. F. Hitch.

28 Orndorff and Delbridge

in methyl alcoliol than the red hydrate it crystallizes out
first, thus explaining the very light red color of this product.^
The. later fractions from methyl alcohol grow steadily deeper
red in color and finally a dark red crystalline meal is obtained,
identical with the product crystallized from ethyl alcohol.^
It is quite probable that the material from which the diacetonate^
was first obtained and which had a light yellowish gray color
consisted very largely of this colorless hydrate. When the
crude tetrachlorgallein is extracted with methyl alcohol,
for the purpose of removing the zinc chloride and impurities,
it is found that the undissolved material grows steadily lighter
in color as the extraction proceeds, owing to the fact that the
red hydrate is more soluble in methyl alcohol than the colorless
hydrate.

The colorless hydrate exhibits almost the same chemical
properties as the red compound. It loses one molecule of
water at 157°, forming the anhydrous tetrachlorgallein, as
shown by the analyses given above. This product, like the
one tiaobned by heating the red hydrate and also the diacetonate
to 157°, has only a faint color and is probably a lactoid form
of tetrachlorgallein.* The colorless hydrate dissolves in dilute
sodium carbonate solution with a red color and in excess
of sodium hydroxide with the characteristic blue color given
by all the forms of tetrachlorgallein (the hydrates, the anhyd-
rous compound and the carbinol acid). When these solutions
are treated with mineral acids, the red hydrate is precipitated.
When exposed to dry hydrochloric acid gas, the colorless
hydrate turns red at once and slowly absorbs hydrochloric
acid, at the same time losing water, to form the same hydrochloride
that the anhydrous tetrachlorgallein does.^ The absorption
of the hydrochloric acid takes place much more slowly than
in the case of the red hydrate, 240 hours being required before
constant weight is attained. This difference in conduct
of the two hydrates is due to the difference in structure. The

1 See page 8 for the analyses.

2 See page 24.

3 See page 1.

* See page 12.

« Compare this conduct with that of the red hydrate, see page 30.

Tetrachlorgallein and Some of Its Derivatives 29

f^ed hydrate is already a quinoid compound and can absorb
hydrochloric acid directly to form the hydrochloride of the
hydrate, while the colorless hydrate must first lose water and
be transformed into the quinoid condition before it can unite
with hydrochloric acid, forming the hydrochloride of the an-
hydrous tetrachlorgallein. Experimental data:^

Tetrachlor-

gallein

Anhydrous

Tenth-normal

Hydrochloric

hydrate

tetrachlor-

sodium

acid ab-

Water

(colorless)

gallein

hydroxide

sorbed

lost

Gram

cc.

Per cent.

0.1765 0.1699 3.04 6:28 3.74

Calculated for the formation of
CooHsO^d.-HCl from aoHACl^.H^O
with loss of H2O 7.01 3-47

With dry ammonia gas the colorless hydrate acts exactly
like the colored hydrate,^ losing water and forming at once
the colored tetrammonium salt, as shown by the following
experiment made by Mr. K. F. Hitch:

Tetrachlor-
gallein Tenth-normal

hydrate Gain in hydrochloric Ammonia Gain in Water

(colorless) weight acid absorbed weight lost

Gram Gram cc. Per cent. Per cent. Per cent.

0.2074 0.0199 15-93 1308 9.60 3.48

Calculated for the formation of

C,oHACl4(NHJ, from

C2oH807Cl,.H20 with loss of

H2O 13. II 9.64 3.47

As the colorless hydrate must have the lactoid structure, it
follows that the transformation of the lactoid to the quinoid
condition can take place in solids as well as in solution.

When the colorless hydrate is dissolved in acetone or methyl
alcohol and the solution is poured into a large quantity of
acidified water, the carbinol acid is precipitated.^ If the
acetone solution of the colorless hydrate be concentrated by

> This experiment was made by Mr. E. F. Hitch. See page 30 for details of the
method.

2 See page 33 for details of the method.

3 See page 37.

30 Orndorff and Delbridge

distilling off a sufficient amount of the acetone, the diacetonate^
crystallizes out.

As this hydrate is a colorless compound, it must have the
lactoid structure and the following formula is suggested for
it:

OH ,, OH

Tetrachlorgalleiu Hydrate (colorless)

The influence of the chlorine in the phthalic acid part of the
compound on the stability of a molecule containing a carbon
atom united to two or three hydroxyl groups shows itself very
clearly in these hydrates and in the carbinolcarboxylic acid.^

Action of Dry Hydrochloric Acid Gas on the Red Tetrachlorgallein
Hydrate
The air-dried red hydrate made from the blue tetrasodium
salt of tetrachlorgallein^ was placed in a stream of hydrochloric
acid gas dried with phosphorus pentoxide. It became much
darker red in color and absorbed slightly more than one molecule
of hydrochloric acid :

Tetrachlorgallein

Gain in

Hydrochloric

hydrate

weight

acid

Gram

Per cent.

0.4296 0.0330 7 13

Calculated for C^oHACl-Hp.HCl 6 . 38

See page 8.

! See also the article on tetrachlorphthalic acid. — This Journal, 41, 411.
' See page 21.

Tetrachlorgallein and Some of Its Derivatives 3 1

This hydrochloride was then heated at 157° in carbon
dioxide and lost both water and hydrochloric acid:

Total loss

Hydrochloric

ibstance

at 157°

acid

Water

Gram

Per cent.i

0.4626 0.0481 0.0330 0.015 1 3.51

Calculated for CaoHgO^Cl^. H2O 3.47

Hence the colored hydrate takes up one molecule of hydro-
chloric acid to form a hydrochloride but without loss of water. In
this respect it differs very markedly from the colorless hydrate
which forms the hydrochloride only by the loss of a molecule
of water. ^

The anhydrous tetrachlorgallein obtained by heating some
of this hydrochloride to 157° also absorbs approximately
one molecule of hydrochloric acid :^

Anhydrous tetra-

Hydrochloric

chlorgallein

Gain in weight

acid

Gram.

Gram

Per cent.

0.3439 0.0267 7.20

Calculated for C2oH807Cl,.HCl 6.77

The formation of the hydrochloride of the red hydrate may
be regarded as a confirmation of the formula given for the
colored tetrachlorgallein hydrate. If this formula be com-
pared with that assigned to tetrachlorgallein hydrochloride,*
it will be seen that it admits of the formation of a hydrochloride
of the colored hydrate without the loss of water :

1 Per cent, of the original hydrate used.

2 See page 29.

3 Compare page 19.

* This Journal, 42, 255 and 258.

32 Orndorff and Delbridge

OH ^ OH

HO;

/\/

C]/ \C^OH + HCl

Ci' 'CI ^^

CI
Red Hydrate

Hydrochloride of Red Tetrachlorgallein Hydrate

Action of Dry Ammonia on the Red Tetrachlorgallein Hydrate
Teirammonium Salt of Tetrachlorgallein. — The colored hy-
drate combines very readily with dry ammonia gas, forming
a tetrammonium salt and at the same time losing a molecule
of water. This was shown by determining first the increase
in weight (which represents the weight of ammonia absorbed

Tetrachlorgallein and Some of Its Derivatives 33

by the hydrate minus the weight of the water lost), secondly,
by determining the amount of ammonia actually present
in the ammonium salt. The colored hydrate used was that
obtained from the blue sodium salt and was some of the same
product made use of in the preparation of the hydrochloride.^
A weighed portion of this red hydrate in a boat was kept in
a stream of dry ammonia gas, at room temperature, until
it came to constant weight, the boat and its contents being
quickly transferred, before weighing, to a tube with a tightly
fitting stopper. This tube was then opened under water
containing a measured volume of tenth-normal hydrochloric
acid, the mixture stirred to ensure complete decomposition
of the ammonium salt and the excess of hydrochloric acid ti-
trated with tenth-normal ammonia solution, the red tetra-
chlorgallein hydrate set free acting as the indicator. The
amount of water lost by the red hydrate used is then found
by subtracting the gain in weight from the amount of
ammonia actually present in the ammonium salt :

Substance
Gram

Gain in
weight
Gram

Tentn-
nonnal hy-
drochloric
acid

CO.

Ammonia
absorbed
Per cent.

Gain in
weight
Per cent.

Water

lost
Per cent.

0.2278

0.0215

17.19

12.85

9-44

341

0.2472

0.0231

18.35

12 .64

9-34

3 30

■ Calculated for the formation of

C3oHACl,(NHJ, from C,,U,0,C\,.U,0
with loss of H2O 13- II 964 3.47

In Analysis II a trace of ammonia was lost owing to the
stopper of the weighing tube becoming loose before it was
completely immersed in the hydrochloric acid solution. The
formation of the ammonium salt is complete in about half an
hour. It is a bluish black compound which dissolves in water
with a deep purple color and is extremely easily oxidized
either in the solid form or in solution. It loses most of its
ammonia so readily that it can be weighed only in a closed
tube. When acids are added to its solution in water, the red
tetrachlorgallein hydrate is precipitated. The same tetram-
monium salt was also obtained by exposing the carbinol-

1 See pages 21 and 30.

34 Orndorff and Delbridge

carboxylic acid and the colorless hydrate to the action of
dry ammonia gas. ^ Furthermore, anhydrous tetrachlorgallein,
made by heating the pure diacetonate to constant weight at
157°, also absorbs ammonia with the evolution of considerable
heat and a marked increase in volume, forming the same
bluish black tetrammonium salt. The same precautions
in weighing were observed to prevent loss of ammonia.

Tetrachlorgallein Gain in weight Ammonia absorbed

Gram Gram Per cent.

0.3293 0.0440 13.36

0.4159 0.0559 1344

Calculated for formation of C2oH407Cl4(NH4)4 13.58

This salt was placed in a Hempel desiccator with concen-
trated sulphuric acid above it and allowed to stand in the
hope that it would come to constant weight. The loss was
very rapid at first, amounting, after two hours, to two-thirds
of the original ammonia. The rate of loss gradually became
less but constant weight was never attained, though there
was evidence of the formation of a mono ammonium salt
which loses some of its ammonia very slowly.

The structure of the tetrammonium salt is best represented
as follows :

ONH4 p. ONH4

NH,0|

CI
Tetrammonium Salt of Tetrachlorgallein
(colored)
1 See pages 29 and 48.

Tetrachlorgallein and Some of Its Derivatives 35

Monoammonium Salt of Tetrachlorgallein
Preparation. — Five and two-tenths grams of the colored
tetrachlorgallein hydrate were dissolved in 300 cc. of absolute
alcohol by the aid of gentle heat, the solution filtered and cooled
to about 35°, and 11.3 cc. of 0.731 normal aqueous ammonia
were added drop by drop from a burette, the solution being
shaken constantly. The weight of anhydrous ammonia
corresponding to the above amount of aqueous ammonia is
0.14 gram while the amount required to combine with the
5.2 grams of the hydrate to form the monoammonium salt
is 0.17 gram. Hence the tetrachlorgallein hydrate was in
excess. The. alcoholic solution turned a deep red on the
addition of the ammonia but no precipitate formed until
all the ammonia had been added and the solution had been
agitated for several minutes. Then very fine dark needles
began to appear and the crystallization was complete after ^
about 90 minutes. The crystals were filtered off, washed
with 94 per cent, alcohol, spread out on filter paper and let
stand in the air for 40 hours. They were then analyzed as
follows: A weighed amount in a platinum boat was placed
in the drying apparatus^ and hydrogen, purified by concen-
trated sulphuric acid, 10 per cent, caustic potash solution,
calcium chloride and phosphorus pentoxide in succession,
was passed over it. The other end of the drying apparatus
was connected with a Hugershoff gas wash bottle containing
a measured amount of standard hydrochloric acid properly
diluted. The boat was then heated to 160° until the substance
lost no further weight, which required about three hours.
The total loss in weight was due entirely to loss of ammonia
and of water, as the solution in the Hugershoff wash bottle
was tested for alcohol by means of the iodoform reaction
but none was present. Considerable water was given off
and condensed on the cold part of the tube. The ammonia
in the wash bottle was determined by titrating the excess of
standard hydrochloric acid with tenth-normal sodium hydroxide
solution and cochineal. The difference between the total loss
and the weight of ammonia found by titration is evidently

' This Journal, 41, 403.

36 Orndorff and Delbridge

due to water. The dry residue was then placed in a measured
amount of tenth-normal hydrochloric acid and the mixture
heated to ensure the complete decomposition of the ammonium
salt. It was then cooled, transferred to a porcelain dish
and the excess of hydrochloric acid titrated with tenth-normal
sodium hydroxide solution, the colored tetrachlorgallein
hydrate set free acting as the indicator.

Tenth-normal hydrochloric acid

For volatilized For ammonia
Substance Loss at 160° ammonia in residue

Gram Gram cc. cc.

I 0.2701 0.0334 2.50 2.05

II 0.2894 0.0360 2.73 2.07

Ammonia Ammonia in Total Water

Loss at 160° volatilized residue ammonia lost

Per cent. Per cent. Per cent. Per cent. Per cent.

I 12.37 1.58 1.29 2.87 10.79

II 12.44 I-6l 1.22 2.83 10.83

Calculated for C2oH707Cl,(NH J. 3. 5H2O 2.93 10.84

These analyses show that this compound is the mono-
ammonium salt of tetrachlorgallein having the formula
C2oH707Cl4(NH4).3.5H20. All the water and over half the
ammonia are driven off at 160°. Since the pure colored tetra-
chlorgallein hydrate, which had been previously analyzed, was
used in making the salt, chlorine determinations would be
superfluous.

Properties. — The compound is beautifully crystallized, con-
sisting, as seen under the microscope, of slender prisms, which
are red in transmitted light and almost black in reflected
light. In bulk, the salt is nearly black with a faint bronzy
luster. It dissolves in water, giving a red solution, but part
of the salt is hydrolyzed and some of the red tetrachlorgallein
hydrate is precipitated as in the case of the sodium and potas-
sium salts. ^ The compound does not lose ammonia in the air
at room temperature, as shown by the fact that the product
analyzed had stood for 40 hours. It dissolves slightly in cold
absolute alcohol with a pink color. When heated with ab-
solute alcohol, it dissolves rather easily but with decomposition,

' This Joxjrnal, 42, 215 and 218.

Tetrachlorgallein and Some of Its Derivatives 37

ammonia being evolved. It is insoluble in cold acetone and
is partially decomposed by boiling acetone, with the loss of
ammonia. In ether, benzene and other neutral organic
solvents, it is insoluble. Its structure is similar to that of
the monosodium and monopotassium salts of tetrachlorgallein
and is given by the formula on page 259 in the first article.'

Tetrachlorgalleincarbinolcarboxylic A cid
/(0H)2

C,H

/C(OH) / >0

\C(0H)3 \0H)2

This compound was first obtained by dissolving the pure
diacetonate in hot acetone, cooling and diluting the cold
solution with cold water. The colorless acetone solution
became red and then, on the addition of more water, an almost
colorless precipitate was thrown down. The mixture was
divided into two parts. One was filtered, the precipitate
washed with cold water and dried at room temperature. The
other portion was boiled for one hour after the addition of
a little acetic acid, filtered hot and the precipitate, after thor-
ough washing with hot water, dried at 25 °. The two products
looked exactly alike and were very nearly white. The product
which had been boiled was analyzed.

Substance Loss at 157° Water

Gram Gram Per cent.

0.3278 0.0129 3.94

0.3796 0.0148 3.90

Calculated for C2oHio08Cl,.H20 3.35

A chlorine determination on the substance dried at 157°
gave the following result :

Tenth-normal

Substance

silver nitrate

Chlorine

Gram

cc.

Per cent.

0.3145 24.26 27.35

Calculated for CaoHioOgCl^ 27 . 28

See This Journal, 42.

Orndorff and Delbridge

The compound, dried at 157°, therefore contains one molecule
of water more than the anhydrous tetrachlorgallein, CjoHgOyCl^.
In order to drive off this water, the substance dried at 157°
was heated in carbon dioxide to 203° but lost very little weight
even at this high temperature, less than 0.4 per cent. The
extra molecule of water which the dried substance contains
must therefore be represented as water of constitution. This
compound dissolves in excess of caustic alkalies with a blue
color and gives all the characteristic reactions of the red
tetrachlorgallein hydrate, but while this hydrate is colored
and loses a molecule of water when heated to 157°, giving
the anhydrous tetrachlorgallein, the carbinol acid is white
and also loses a molecule of water at 157°, but the resulting
product, though colorless, still differs from the anhydrous
tetrachlorgallein by one molecule of water which can not be
driven off even at 203°. The compound is therefore the
carbinolcarboxylic acid :

Tetrachlorgallein and Some of Its Derivatives 39

OH ^ OH

Tetrachlorgalleincarbinolcarboxylic Acid

The molecule of water which is given off at 157° is repre-
sented in the second formula as present in the same form as
in the colored tetrachlorgallein hydrate and for the same
reasons.^

A second preparation was made from another lot of the
diacetonate, which was first recrystallized from acetone and
then dissolved in pure dry acetone. This solution was filtered,
cooled and poured slowly and with constant stirring into
2.5 liters of distilled water containing a trace of hydrochloric
acid and previously cooled to 13°. A faintly pink precipitate
was at once thrown down. The mixture was then heated
to boiling for thirty minutes by passing in steam, the solution
at once filtered, and the precipitate washed with distilled
water and dried in the air at room temperature. The product
dried to constant weight at 157° was then analyzed for chlorine.

Substance
Gram

Tenth-normal

silver nitrate

cc.

Chlorine
Per cent.

0.2875
0 . 2402

.22 . 14
18.43

Calculated for C20H

oOsCl,

27.31
27.21
27.28

See This Jour

NAL,

42, 255

40 Orndorff and Delhridge

Some of the diacetonate made by Mr. Briggs* was dissolved
in acetone and the pale red solution added to a large excess
of cold water acidified with hydrochloric acid. The precipitate
was filtered off, washed thoroughly with water and dried to
constant weight in a water oven. It was almost colorless,
having a faint pink tinge due to the presence of a minute
quantity of the colored hydrate. Analysis •?

Substance

Tenth-normal

(dry)

silver nitrate

Chlorine

Gram

cc.

Per cent.

0.3091 23.81 27.32

Calculated for C2oH,o08Cl4 27 . 28

In purifying the crude tetrachlorgallein by recrystallization
from methyl alcohol, it was noted that the methyl alcohol
solution, when poured into a large quantity of cold water,
gave a very light colored product resembling that obtained in a
similar manner from the acetone solution. Analyses by Mr.
Briggs show that this product is also the carbinol acid :

Substance Tenth-normal

(dried at 158°) silver nitrate Chlorine

Gram cc. Per cent.

0.2543 19.59 27.31

O.25II 19.38 27.37

Calculated for CjoHioOgCl^ 27.28

When the ethyl alcohol or glacial acetic acid solution of
tetrachlorgallein is diluted with water, the precipitate formed
is also very light colored and is probably the carbinol acid.

The carbinol acid was also made by Mr. Briggs from the
tetrachlorgallein tetracetate. About four grams of the pure
tetracetate were dissolved by heating on a water bath in 50 cc.
of pure concentrated sulphuric acid in an atmosphere of carbon
dioxide. The deep red solution thus obtained was cooled and
poured slowly, with constant stirring, into 100 cc. of methyl
alcohol kept cold in a freezing mixture. This solution was then
poured into a large volume of cold water, the precipitate
washed thoroughly and dried in the air. Analyses gave the
following results :

1 See page 8.

2 Made by Mr. Briggs.

Tetrachlorgallein and Some of Its Derivatives

Substance
Gram

Loss at 158°
Gram

Water
Per oent.

0 . 2406

0.0138

5-74

0.2471

Calculated for C^oHioOsCl^.H^O
Calculated for C,,U,,0,C\,.2}ilO

5-75
3-35
6.49

Substance

(dried at 158°)

Gram

Tenth-normal

silver nitrate

cc.

Chlorine
Per cent.

0.2268

17-45

27.28

0.2329

17-93
Calculated for C.oHioOgCl^

27.30
27.28

When the tetracetate is hydrolyzed with concentrated
sulphuric acid and the resulting solution poured directly
into ice water, a reddish precipitate is formed consisting of the
red hydrate and a small amount of the colorless carbinol
acid, as the following analyses by Mr. Briggs show •}

Substance
Gram

Loss at 158°
Gram

Water
Per cent.

0.2592
0.2626

0.0095
0 . 0094

Calculated for C2oH807Cl,.H20

3 66

3-59
3-47

Substance
(dried at 158°)
Gram

Tenth-normal

silver nitrate

cc.

Chlorine
Per cent.

0.2497
0.2351

19 75
18.54
Calculated for C2oH«07Cl,

28.05
27.97

28.26

The carbinol acid is also obtained when an acetone solution
of the red hydrate is poured into a large quantity of ice water
acidified with hydrochloric acid. Analysis by Mr. Briggs:

Substance

(dry)
Gram

0.2164

Tenth-normal
silver nitrate

16.63

Calculated for C^oH^oOgCl^

Chlorine
Per cent.

27.25

27.28

The anhydrou? tetrachorgallein made by heating the di-
acetonate to 158° also gives the carbinol acid when its acetone
solution is poured into a large volume of ice water. ^

The light colored hydrate obtained by crystallizing the

1 See also page 23.
1 See page 48.

42 Orndorff and Delbridge

crude tetrachlorgallein several times from methyl alcohol,
and which is essentially the colorless hydrate containing a
trace of the colored hydrate, also gives the carbinol acid when
it is dissolved in acetone and the solution is poured into cold
water acidified with hydrochloric acid, as shown by the follow-
ing analysis made by Mr. Briggs :

Substance Tenth-normal

(dry) silver nitrate Chlorine

Gram cc. Per cent.

0.2II2 16.28 27.33

All the forms, therefore, of tetrachlorgallein (the colorless
hydrate, the colored hydrate and the anhydrous product),
as well as the colorless diacetonate, give the carbinol acid
when precipitated from acetone solution by water.

Many attempts were made to obtain the carbinol acid in
crystals but without success, as it was found that the colorless
hydrate or a mixture of the two hydrates crystallized out in
every case. A well crystallized product having almost no
color and obtained by crystallizing the carbinol acid from
dilute acetone was analyzed by Mr. Briggs:

Substance
Gram

Loss at 157° Water
Gram Per cent.

0.2625

0 . 0096 3 . 66
Calculated for CaoHgOjCl^.HaO 3 . 47

Substance
(dry)
Gram

Tenth-normal

silver nitrate Chlorine
cc. Per cent.

0.2525

19.99 28.08

Calculated for C^oH ACI4 28 . 26

These results show that the compound is the colorless hydrate.
The same product was obtained from methyl and ethyl alcohols
in attempts to crystallize the carbinol acid from these solvents,
though in this case it was more highly colored and hence con-
tained more of the colored hydrate. The colored hydrate,
itself, when crystallized from any of these solvents, is partially
converted into the colorless hydrate, which crystallizes out
first in the characteristic white needles. Owing to this easy
conversion of the colored into the colorless hydrate, it is doubtful

Tetrachlorgallein and Some of Its Derivatives

whether the red hydrate has ever been prepared entirely free
from its colorless isomer. To get the colorless hydrate pure
is a comparatively easy matter, the presence of the slightest
trace of the red isomer being shown at once by the color of
the product.

The transformation of the carbinol acid into the colorless
hydrate can be readily understood from the following equation :

CI
Carbinol Acid

HO,

CI
Colorless Hydrate

+ H2O.

Orndorfj and Delbridge

A certain amount of the carbinol acid is converted into the
colored hydrate at the same time, the amount depending upon
the boiling point of the solvent. For example, the crystals
from acetone are almost colorless while those from methyl
and especially from ethyl alcohol are more highly colored.
This conversion is represented as follows :

CI
Carbinol Acid

CI
Colored Hydrate

Tetrachlorgallein and Some of Its Derivatives

When heated to 158°, the carbinol acid loses water from a
different part of the molecule and gives the anhydrous carbinol
acid:

OH ^ OH

CI
Carbinol Acid

Carbinol Acid (anhydrous)

Attempts were made to crystallize the anhydrous carbinol
acid from dry acetone and from absolute methyl alcohol but
analyses showed that the products obtained were mixtures
of the hydrates and the carbinol acid.

46 Orndorff and Delbridge

Action of Hydrochloric Acid on Tetrachlorgalleincarbinolcarb-
oxylic Acid
It was found that the colorless carbinolcarboxylic acid,
dried at 157°, takes up hydrochloric acid, when placed
in an atmosphere of the dry gas at room temperature, to form
a colored hydrochloride and it seemed probable that it would
give the same product that the red tetrachlorgallein hydrate
does^ and hence would lose a molecule of water and of hydro-
chloric acid on heating to 157°. Some of the pure carbinol-
carboxylic acid was therefore dried to constant weight at
158° and put in an atmosphere of dry hydrochloric acid gas
at room temperature. It turned red at once and gradually
gained weight until it had absorbed one molecule of hydro-
chloric acid, after which it began to lose weight. This loss
proved to be due to water but was so slow that the red hydro-
chloride was heated to 157° in an atmosphere of carbon dioxide.
Both water and hydrochloric acid were evolved. The latter
was absorbed in distilled water in a Hugershoff gas wash bottle
and the amount determined by titration with standard sodium
hydroxide solution. The residue dried at 157° was then
weighed and shown to be anhydrous tetrachlorgallein by
analysis :

Tetrachlorgal-

leincarbinol- Anhydrous Tenth-normal Hydrochloric

carboxyHc tetrachlor- sodium acid ab- Water

acid gallein hydroxide sorbed lost

Gram Gram cc. Per cent. Per cent.

0.4024 0.3897 7.97 7.22 3.16

0.6417 0.6205 12.80 7.27 3.30

Calculated for formation of CjoHgOjCl^.I-lCl

from C20H loOgCl, 7.01 3.47

The chlorine in the material dried at 157° was then de-
termined :

Tenth-normal
Substance silver nitrate Chlorine

Gram cc. Per cent.

0.2740 21.70 28.08 .

Calculated for CjoHgO^Cl^ 28 . 26
These results show that the dried, colorless carbinolcarb-
oxylic acid, which loses no weight even when heated to 203°,
takes up hydrochloric acid to form a colored hydrochloride

Tetrachlorgallein and Some of Its Derivatives 47

and this, when heated to 157°, loses both hydrochloric acid
and water, giving the anhydrous tetrachlorgallein. This
furnishes a most striking proof of the structure of the carbinol-
carboxylic acid.^ It is quite likely that the hydrochloride
of the colored hydrate is formed here, as the carbinol acid gains
in weight until it has absorbed an amount equivalent to one
molecule of hydrochloric acid, after which it slowly loses weight
due to the loss of water. The reaction may be represented
as follows :

OH ^ OH

CI
Carbinol Acid
(anhydrous)

OH ^ OH

CI
Hydrochloride of the Colored Hydrate
1 See page 39.

48 Orndorff and Delbridge

Action of Dry Ammonia on Tetrachlorgalleincarhinolcarhoxylic
Acid

The colorless carbinol acid, dried at 157°, takes up ammonia
at ordinary temperatures, when placed in an atmosphere
of the dry gas, forming the same bluish black tetrammonium
salt that the hydrates do,^ and loses a molecule of water. The
carbinol acid used was made from the anhydrous tetrachlor-
gallein obtained by heating the diacetonate to constant weight
at 157°. This product was dissolved in acetone, the solution
cooled to 8° and diluted with a large volume of cold distilled
water containing a little hydrochloric acid. The precipitated
carbinol acid was suspended in hot alcohol, filtered, washed
with alcohol, dried in the air and then heated to constant
weight at 157°. A weighed quantity was then exposed to the
action of dry ammonia at ordinary temperature until it came
to constant weight, for which one hour was sufficient. The
gain in weight was then obtained directly and the ammonia
absorbed was determined by decomposing the tetrammonium
salt with a measured amount of standard hydrochloric acid
and titrating the excess of acid with standard ammonia solu-
tion, the red tetrachlorgallein hydrate set free^ acting as the
indicator. The water lost in the formation of the tetrammon-
ium salt was then obtained by subtracting the gain in weight
from the amount of ammonia absorbed.

Carbinol acid

(dried at 157°)

Gram

Gain in
weight
Gram

Tenth-normal
hydrochloric
acid
cc.

Gain in
weight
Per cent.

Ammonia
absorbed
Per cent.

Water

lost

Per cent.

0.2262

0.0203

15 -95

8.97

12.01

3 04

0.3120

0.0291

22.60

9-33

12.34

3.01

0.2208

0.0226

17.18

10.24

1325

3.01

Calculated

for conversion of

C^oH.oO^Cl, into C30H AC1,(NHJ,

^ 9.64

13 II

3-47

Hence the colorless tetrachlorgalleincarbinolcarboxylic acid,
which had been dried to constant weight at 157,° loses water
of constitution at room temperature in an atmosphere of dry
ammonia and absorbs ammonia, forming the colored tetra-

1 See pages 29 and 32.

2 See page 33.

Tetrachlorgallein and Some of Its Derivatives 49

chlorgallein tetrammonium salt. A comparison of the struct-
ural formula for this salt^ with the one given for the carbinol-
carboxylic acid^ shows that its formation from the carbinol
acid, dried at 157°, is possible only by the loss of a molecule
of water of constitution.

Potassium Salt made from the Tetrachlorgalleincarbinolcarboxylic
Acid

A solution of 12 grams of the air-dried carbinol acid in
600 cc. of 94 per cent, alcohol was heated to boiling in a flask
connected with a reflux condenser. Alcohol vapor was passed
into this solution through a tube reaching to the bottom
of the flask. By this means the mixture was kept constantly
agitated and all the air was expelled from the flask, thus pre-
venting oxidation. A filtered solution of 3 grams of pure
potassium acetate in 100 cc. of 94 per cent, alcohol was then ^
slowly added to the boiling liquid through the condenser.
The potassium salt was filtered off after boiling for 30 minutes,
boiled again with 600 cc. of fresh 94 per cent, alcohol, filtered,
washed with alcohol, dried in the air several days and analyzed.
It contained water of crystallization which was determined
by heating the salt to 157° for three hours in a stream of pure
hydrogen in the drying apparatus.^ An examination of the
vapors given off showed that they contained no alcohol.
The loss at 157° was, therefore, due to water, considerable
moisture condensing on the cold part of the drying tube.

Substance Loss at 157° Water

Gram Gram Per cent.

0.3073 0.0327 10.64

0.3000 0.0320 10.67

Calculated for C2oH707Cl,K.3.5H20 lo . 46

The chlorine in the material dried at 157° was determined
by the modified lime method^ and the potassium by decom-
posing a weighed quantity with a measured amount of standard
hydrochloric acid and titrating the excess of the hydrochloric

» Page 34.

2 See page 39.

3 This Journal, 41, 403.
* Ibid., il, 397.

50 Orndorff and Delbridge

acid with standard alkali, the red tetrachlorgallein hydrate
set free acting as the indicator.^

Substance
Gram

Tenth-normal

silver nitrate

cc.

Chlorine
Per cent.

0.2739

20. 13

26.07

0.2519

18.53

Calculated for C^oH^O^Cl.K

26.09
26.26

Substance
Gram

Tenth-normal

hydrochloric acid

cc.

Potassium
Per cent.

0.4894

8.86

7.09

0.6625

12.00
Calculated for C2,

»H,0,C1,K

7.09

7-25

This compound is, therefore, the monopotassium salt of
tetrachlorgallein and its properties are also identical with
those of the potassium salt made from the anhydrous tetra-
chlorgallein and from the red hydrate.^ The crystals, which
are dark brown, dissolve in water with a red color, at the same
time undergoing partial hydrolysis with precipitation of a small
amount of the colored tetrachlorgallein hydrate. Acids pre-
cipitate the red hydrate from the aqueous solution completely.
With regard to the crystallography of the salt, Professor
Gill reports as follows: "The substance occurs in minute
columnar crystals 0.002 to 0.004 mm. in thickness, and o.oi
to 0.02 mm. long. They show strong pleochroism, changing
from nearly colorless to deep purplish brown on rotation in
transmitted polarized light. Interference colors between
crossed Nicols are of lower first order, indicating fairly strong
double refraction, probably 0.060 to 0.080. An extinction
angle of 20° to 25° proves the substance to be either mono-
clinic or triclinic in crystallization but no optical figure could
be obtained, on account of the very small size of the crystals."

The structural formula for this colored monopotassium
salt of tetrachlorgallein has already been given. ^

Baeyer,* after his very thorough investigation of the con-

> See This Journal, 42, 216.
2 Ibid.. 42, 217 to 219.
3/6id.,42, 259.
* Ann. Chem. (Liebig), 364-, 152.

Tetrachlorgallein and Some of Its Derivatives 51

nection between color and chemical composition in the case
of the derivatives of triphenylcarbinol, finds that all carhinols
are colorless; color appears only in consequence of the elimination
of water from the molecule. This conclusion is strikingly con-
firmed by the conversion of the colorless tetrachlorgallein-
carbinolcarboxylic acid into the three colored compounds,
the hydrochloride, the ammonium salt and the potassium
salt.^ By means of these compounds it is possible to convert
the colorless carbinol acid into the colored hydrate, for the
hydrochloride gives this hydrate when boiled with water
and the two salts give the same product when their aqueous
solutions are decomposed with acids. It is not necessary to
isolate the compounds. Thus if the carbinol acid is dissolved
in caustic alkali and acid is added to the solution, the red
hydrate is precipitated. This may then be reconverted into
the colorless carbinol acid by dissolving in acetone and pre- .
cipitating with cold water. The anhydrous tetrachlorgallein
may be obtained from the carbinolcarboxylic acid most readily
by treating the latter at room temperature with dry hydro-
chloric acid gas until it ceases to gain weight and then heating
the product to 158°.

Action of Acetic Anhydride on Tetrachlorgalleincarbinolcarhoxylic
Acid

It was thought possible that the presence of the five alcoholic
hydroxyl groups in the carbinol acid might be shown by
making a pentacetate. Five grams of the pure white carbinol
acid were therefore boiled with 30 cc. of acetic anhydride
for one hour. The solution had a dark brown color but showed
no fluorescence. Fifty cc. of methyl alcohol were then added
to the cold solution and the mixture heated to boiling to con-
vert the excess of acetic anhydride into methyl acetate. The
solution became almost solid from the separation of the tetra-
chlorgallein acetate. More methyl alcohol was therefore
added and the mixture boiled. The tetracetate, which was
almost pure white and beautifully crystallized, dissolved

1 Although the air-dried potassium salt contains water of crystallization, still
this is entirely driven off at 157° while the carbinol acid itself is stable at 203°.

52 Orndorff and Delbridge

only to a slight extent. The crystals were filtered ofiF, washed
with methyl alcohol and dried. They were found to be in-
soluble in sodium carbonate solution even on heating. This
fact indicated that the product was not an acid and hence
that it was not the pentacetate of the carbinolcarboxylic acid.
The white crude acetate melted at 25o°-252° (uncor.) but
became pink much below this temperature, thus indicating
that the compound was the tetrachlorgallein tetracetate.
It was, therefore, purified by crystallization from benzene.^
The crystals were then pure white and melted sharply at 260°
(cor.), the melting point of the tetracetate, and also showed
the same marked contraction at 150°- 160° noted in the case
of the tetracetate made from the anhydrous tetrachlorgallein.
A chlorine determination on the substance dried to constant
weight at 100° in an atmosphere of carbon dioxide gave
further proof that this product really was tetrachlorgallein
tetracetate :

Tenth-normal
Substance silver nitrate Chlorine

Gram cc. Per cent.

0.3692 21.97 21.10

Calculated for C2oH407Cl4(C2H30)4 21.17

The carbinol acid then gives tetrachlorgallein tetracetate,
a colorless product having the lacioid structure,^ when heated,
with acetic anhydride, thus showing that the elimination of
water from the carbinol acid does not necessarily give a colored
compound. The product formed from the carbinol acid by
the loss of water must have the quinoid structure in order
that a colored compound may result.

Preparation of Tetrachlorgallein Hydrate

As it seemed probable that the long and tedious heating of
the tetrachlorphthalic acid to 100° to convert it into the
anhydride might be avoided, Mr. Briggs, at our suggestion,
endeavored to make tetrachlorgallein hydrate directly from
the tetrachlorphthalic acid, using 155 grams of the acid, 135
grams of pyrogallol and 70 grams of freshly fused zinc chloride.

1 This Journal, 42, 226.
2/6ti..42, 267.

Tetrachlorgallein and Some of Its Derivatives 53

The substances were intimately mixed in a two-liter balloon
flask and heated to 200° in a bath of potassium sulphate
and sulphuric acid. Carbon dioxide was passed into the
flask during the heating to prevent oxidation. Large quantities
of steam were evolved and the reacton proceeded just as
smoothly as when the anhydride was used. The yield seemed
to be as large and the product just as pure as that made from
the anhydride, in fact, most of the tetrachlorgallein hydrate
used by Mr. Briggs in this investigation was made by this
method and purified by extracting the zinc chloride with
methyl alcohol and crystallizing the residue from methyl
alcohol.

An interesting experiment illustrating the sensitiveness
of tetrachlorgallein to alkalies and the solubility of glass in
water can be performed as follows: A very dilute solution of
any form of tetrachlorgallein is used as indicator. Some
finely powdered, easily fusible glass is placed in a test tube
and washed with distilled water. Pure water, preferably
conductivity water, containing no alkali is then added and a
few drops of the indicator run in. In a few minutes the
supernatant liquid and the powdered glass are colored an
intense purplish blue, showing that the water has dissolved
a certain quantity of alkali from the glass. Indeed, so sensi-
tive is the tetrachlorgallein to alkalies that it has been found
advisable to use Jena-glass flasks in working with solutions
of the substance and to avoid the use of soft glass entirely.
A solution of any form of tetrachlorgallein may also be used
to show the alkalinity of the blood. If a drop of such a solu-
tion be placed on the skin, it turns first red and then, more
slowly, blue, owing to the formation of sodium salts. If
the spot be treated with acids, it turns red but soon becomes
blue again after the acid, is washed off.

SUMMARY

The results of this investigation may be briefly stated as
follows :

I. Tetrachlorgallein combines with two molecules of acetone
to form a colorless diacetonate. This loses acetone when

54 Orndorff and Delhridge

kept in a desiccator over phosphorus pentoxide and becomes
colored, the color being deepest red when one molecule of
acetone has evaporated. In the air, the diacetonate loses
acetone and takes up water, turning red and forming the
colored hydrate. A structural formula for the compound,
in accord with these facts, has been suggested, page lo.

2. A colorless etherate of tetrachlorgallein has been made.
This loses ether and takes up water when exposed to the
air, forming a mixture of the colorless hydrate and the colorless
carbinolcarboxylic acid. A structural formula, in accord
with this conduct, has been suggested for the compound,
page 17.

3. Anhydrous tetrachlorgallein absorbs hydrochloric acid gas
very readily, forming a red hydrochloride, which, when heated
to 157°, loses all its hydrochloric acid and goes back to the
original tetrachlorgallein. Here again the transformation from
the lactoid to the quinoid condition and back again takes
place in solids.^

4. Two hydrates of, tetrachlorgallein have been made,
one colorless and the other colored. To the colorless product
the lactoid formula has been assigned, while the quinoid formula
best represents the conduct of the red modification.^ The red
hydrate takes up a molecule of hydrochloric acid, without
losing water, to form the hydrochloride of the hydrate, while the
colorless hydrate loses a molecule of water and takes up hydro-
chloric acid, forming the same red hydrochloride as the an-
hydrous tetrachlorgallein does, thus strikingly confirming the
structural formulas given the two isomers. With ammonia both
hydrates give the same tetrammonium salt as the anhydrous
tetrachlorgallein does and both lose water, as the formulas
require. Both hydrates, when heated, yield the same anhy-
drous tetrachlorgallein, which, from its slight color, is prob-
ably the lactoid modification (page 12).

5. A colored tetrammonium salt and also a colored mono-
ammonium salt have been prepared from the red hydrate
by the action of ammonia. The colorless hydrate gives the

' This Journal, 42, 256, 258 and 265.
2 Page 30 and This Journal, 42, 255.

Tetrachlorgallein and Some of Its Derivatives 55

same colored tetrammonium salt when brought into contact
with dry ammonia gas, again showing that the transformation
of the lactoid into the quinoid condition can take place in
solids.

6. Tetrachlorgalleincarbinolcarboxylic acid, a colorless com-
pound, has been made and studied. It is formed when any
form of tetrachlorgallein (the colorless hydrate, the red hydrate
or the anhydrous product) or the diacetonate is dissolved in
acetone or methyl alcohol and the solution poured into a large
excess of acidified water. The anhydrous carbinol acid,
obtained by heating the acid to constant weight at 158°,
still contains one molecule of water (of constitution) more than
the anhydrous tetrachlorgallein and it retains this water
even when heated to 203°. Yet it combines very readily,
even in the cold, with dry hydrochloric acid to form the red
hydrochloride and with dry ammonia to form the colored
tetrammonium salt and at the same time loses a molecule of
water. This confirms, in a very striking manner, Baeyer's
statement that all carhinols are colorless and that color only
appears in consequence of the loss of water.

7. A colored monopotassium salt of tetrachlorgallein has
been made from the colorless carbinol acid. Here again a
molecule of water splits off from the carbinol acid in the forma-
tion of this colored salt.

8. The colorless carbinol acid splits off a molecule of water
also when heated with acetic anhydride, forming the colorless
tetrachlorgallein te trace tate, a lactoid compound, thus showing
that the elimination of water from the carbinol acid does
not necessarily give a colored compound. The product re-
sulting from the loss of water must have the quinoid structure
in order that a colored compound may be formed.

9. A simpler method for the preparation of tetrachlor-
gallein hydrate, involving the use of tetrachlorphthalic acid
instead of the anhydride, has been found.

Cornell University,
May, 1911

A STUDY OF THE CONDUCTIVITY AND DISSOCIATION
OF ORGANIC ACIDS IN AQUEOUS SOLU-
TION BETWEEN ZERO AND
THIRTY-FIVE DEGREES

By E. p. Wightman and Harry C. Jones
HISTORICAL

White and Jones/ in their work on the conductivity and
dissociation of organic acids in aqueous solution, give a survey
of the data obtained up to that time, and a large amount of
the earlier work is discussed in full. It is not necessary to
repeat all of this discussion, but a short summary of the results
as a whole will be given as an introduction to this work.

One of the first things noticed by early workers was the
increase in molecular conductivity with rise in temperature
and increasing dilution. It was shown also that this increase
is, for most electrolytes in dilute solutions, a parabolic function
of the temperature, and the following interpolation formula
was deduced and employed by Euler •}

X = a + ht — ct^

in which k is the molecular conductivity at the temperature
t, a is the known conductivity at some other temperature,
and h and c are constants depending upon the nature of the
electrolytes. Extensive use has been made of this formula.

As the dilution increases the rate of increase in conductivity
becomes less, and in some cases there is a maximum value
of conductivity, as was shown by Grotian,^ Jahn,* Schaller,^
and the later workers. As a matter of fact, the maximum
occurs for nearly all strong electrolytes at dilutions at which
the conductivity can be measured directly.

This is not the case with most of the organic acids, but
indirect methods were devised by Ostwald" and by White

1 This Journal, 44, 159 (1910).

2 Z. physik. Chem.. 21, 257 (1896).

3 Pogg. Ann.. 164, 215 (1875),

■» Z. physik. Chem.. 16, 72 (1895).

(■ Ibid.. 26, 497 (1898).

o/6id., 1, 74,97 (1887); 2, 840 (1888).

Conductivity and Dissociation of Organic Acids 57

and Jones/ based upon Kohlrausch's law of the independent
migration velocities of ions, by means of which /jl^ for the
acids could be determined.

Ostwald showed that there is a constant difference between
fi^ for a given dilution, say thirty-second normal, of the sodium
salt of any acid, and the n^ value of the same, which is found
at about ten hundred and twenty -fourth normal. By means
of this constant difference he calculated the /x^ values of a
large number of sodium salts without direct measurement,
and from these it was easy to determine the }i^ values of
the acids in question, by subtracting the value for the migra-
tion velocity of the sodium ion, and adding the corresponding
constant for hydrogen.

A second method suggested by Ostwald was based upon
the fact that the velocities of anions of acids containing over
twelve atoms in the anion are dependent upon the number
of these atoms present — ions with the largest number of atoms
having the smallest velocity.

The method of White and Jones (for monobasic acids) is
based upon the direct measurement of the fi^ value of the
sodium salt of the acid and will be discussed later. The
H^ values of the dibasic acids were determined by a graphic
method similar to the second method used by Ostwald.

Just as the molecular conductivity increases at a diminishing
rate with dilution, so also it increases at a diminishing
rate with rise in temperature, as was brought out by Schaller-
and by Noyes.^

Another important fact is the decrease in dissociation with
rise in temperature, first noticed by Arrhenius^ in the case of
phosphoric and hypophosphoric acids, later brought out
by Schaller^ and a number of others,® and finally thoroughly

1 This Journal, 44, 159 (1910).

2 Z. physik. Chem., 25, 497 (1898).

3 J. Am. Chem, Soc, 26, 134 (1904); 30, 335 (1908); 31, 987 (1909).
* Z. physik. Chem., 4, 96 (1889).

^Ibid.. 26, 497 (1898).

" Jones and West: This Journal, 34, 357 (1905); Jones and Jacobson: Ibid., 40,
355 (1908); Jones and Clover; /6!d., 43, 187 (1910); White and Jones: Ibid., 44, 159
(1910).

58 Wightman and Jones

established by the work of Noyes' at higher temperatures.
No entirely satisfactory explanation of this decrease in dis-
sociation has been given; but the results of Noyes in his first
work^ show that the dissociations of the two salts, sodium
chloride and potassium chloride, are nearly identical at all
temperatures and concentrations; and he says: "This gives
support to the idea that decrease of conductivity and of cal-
culated dissociation with rise in temperature is due to a physical
cause (probably in some way to the electrical charges on the
ions) and not to specific chemical affinity." More will be said
about this in another connection.

It was pointed out by the later workers (Schaller^ was one
of the first) that the temperature coefficients of conductivity
increase with dilution and decrease with rise in temperature
for acids, and increase with temperature for neutral salts.
Amino acids are an exception to this, as was shown by White
and Jones.* They explain the increase with rise in tempera-
ture as "probably due to the formation of inner salts having
both acidic and basic groups, which break up with rise in
temperature." The decrease of the temperature coefficients
for nearly all other acids is explained by them in terms of the
theory of hydration.^

It was found by White and Jones that the Ostwald dilution
law® holds very well for dilute, weak organic acids, with the
exception of the amino acids. The law is expressed thus :

where a = -^is the dissociation, V is the volume, and K is

a constant. The law is easily deduced from the gas laws and
those of osmotic pressure. The discrepancies in the tempera-
ture coefficients in the case of amino acids were explained as
stated above, viz., as due to the breaking down of the inner

salts.

■ J. Am. Chem. Soc, 26, 134 (1904); 30, 355 (1908); 31, 987 (1909).

^ Ibid.. 26, 134 (1904).

3 Z. physik. Chem., 25, 497 (1898).

* This Journal. 44, 159 (1910).

5 Ibid.. 40, 402 (1908).

"Z. physik. Chem., 2, 36 (1888); 3, 170 (1889). Jahn: Ibid., 23, 545 (1900).

Conductivity and Dissociation of Organic Acids 59

Strong acids (and also other strong electrolytes) do not
conform to the Ostwald law, and a large number of empirical
formulae have been suggested/ all of which hold fairly well
for specific cases, but only a few of which are of general appli-
cation. These are discussed very thoroughly by Noyes,^
who says of the following formulae :
A„-A

Aq-A

K (Kohlrausch)
K (Barmwater)
K (Van't Hoff)

ib^ = K (Rudolphi)

"The Kohlrausch formula expresses the results for both salts
(potassium and sodium chlorides) at all temperatures without
great error, and the same is true of the Barmwater formula
except at the highest temperature, where the deviation with
both salts is large. The van't Hoff and Rudolphi formulas
do not accord at all with the observed values at 306°, the
deviations in the case of the latter being especially large;
while at the lower temperatures, 140°, 218° and 281°, the
van't Hoff formula is far less satisfactory than those of Kohl-
rausch and Barmwater. On the whole, therefore, the simple
Kohlrausch formula furnishes the best representation of the
results and the Barmwater next best."

In terms of the Ostwald formula, i. e., using the same nota-
tion, these would be :

— a
\ = K (Kohlrausch)

^ajV

K (Barmwater)

1 Wied Ann.. 26, 200 (1885); 60, 394 (1893). MacGregory: Ibid., 61, 133 (1894).
Barmwater: Z. physik. Chem., 28, 134, 428 (1899). Sabat: Ibid., 41, 224 (1902).
Muller: Compt. rend., 128, 505 (1899). Kohlrausch: Sitz. preus. Akad., 44, 1002
(1900). Rudolphi: Z. physik. Chem., 17, 385 (1895). Van't Hoff: Ibid., 18, 300
(1895). Kohlrausch: Ibid., 18, 662 (1895). Starch: Ibid., 19, 13 (1896). Bancroft:
Ibid., 31, 188 (1899). Jahn: Ibid., 37, 499 (1901); 41, 265, 288 (1902). Nernst: Ibid.,
38, 493 (1901).

2 J. Am. Chem. Soc, 26, 162 (1904).

6o Wightman and Jones

Noyes^ says also, concerning strong electrolytes: "This
principle has now received a further confirmation through
the demonstration of the fact that certain purely empirical
laws relating to the ionization of salts in water still continue
to be valid, even when the physical condition of that solvent
is greatly altered by a large change in the temperature. This
principle is that the ionization of salts, strong acids and bases
is a phenomenon primarily determined not by specific chemical
affinities, but by electrical forces arising from charges on the ions;
that it is not affected (except in a secondary degree) by chem-
ical mass action, but is regulated by certain general, compara-
tively simple laws, fairly well established empirically but of
unknown theoretical significance, and that, therefore, it is
a phenomenon quite distinct in almost all its respects from
the phenomenon of dissociation ordinarily exhibited by chem-
ical substances, including that of the ionization of weak acids
and bases."

He distinguishes between ordinary unchanged molecules,
which he calls "chemical molecules," and a loosely united
ionized molecule, or "electrical molecule."

Walden- in 1891 and 1892 showed, in his work on di-, tri-,
and tetracarboxylic acids, that the general tendency of such
organic acids is to dissociate, up to quite high dilutions, like
monobasic acids. In the case of tribasic acids he mentions
the three possibilities :

A'^Hg = A"'H2 + H (r)

A'^'Ha = A'"H + H + H {2)

A'"H3 = A'" + H -f- H + H (i)

of which only the first takes place at ordinary dilutions.

The same thing was found to be the case by Walker,^ namely,
that dibasic acids behave just like monobasic acids within
the limits of dilutions at which he worked. Pimelic acid,

+ —

for instance, splits into H and OOC(CH2)5COOH, and not

into H, H, and OOC(CH2)5COO.

1 J. Am. Chem. Soc, 30, 335 (1903).

2 Z. physik. Chem.. 8, 434 (1891); 10, 563 (1892).

3 J. Chem. Soc, 61, 696 (1892).

Conductivity and Dissociation of Organic Acids 6i

EXPERIMENTAL
A pparatus

Burettes and flasks (200 cc, 250 cc, 500 cc, and 1000 cc.)
were all calibrated by the method of Morse and Blalock^ for
a temperature of 20°. The 200 cc. flasks were also calibrated
by weight, and the results were found to agree very closely.
The time necessary to drain all pipettes and burettes was
determined and properly taken into account in the measure-
ments. The thermometers were also carefully standardized
against a Reichsanstalt thermometer.

At first a Wheatstone bridge was used for making the con-
ductivity measurements, and this was calibrated by the method
of Strouhal and Barus.^ Later a very fine Kohlrausch slide-
wire bridge was obtained, by means of which it was possible
to read distances on the slide-wire corresponding to tenths
of a millimeter (the total length of the wire was five meters) .

The resistance box employed in the later work was one
that had been standardized by the U. S. Bureau of vStandards.
The one first used was later compared with this one, and was
found to be accurate to well within the limits of experimental
error.

Three thermostats were employed to keep the cells at
constant temperature; one for 0°, similar to that described
by Jones and Jacobson;^ one for 15° and 25°, a galvanized
tub containing 25 or 30 liters of water, and in the bottom of
which was placed a lead coil through which cold water was
passed under constant pressure; a third for 35°, differing from
the latter only in not having a coil in the bottom. They were
both kept constantly stirred by propellers driven by a hot-air
engine. In this way it was possible to keep the temperature
constant to within o°.02.

At first the thermostats were regulated by hand, and this
was found to be sufficient, provided they were continually
watched. It was found, however, that thermoregulators
save both time and labor, so that finally these were installed.

' This Journal, 16, 479.

2Wied. Ann., 10, 326. Kohlrausch and Holborn: " Leitvennogen der Electro-
lyte." p. 45 (1898).

3 This Journal, 40, 355 (1908).

62 Wightman and Jones

They were of the general type used in this laboratory, so
need not be described here.

The cells resembled those used by Jones and Bingham/
with platinum-plate electrodes, attached to glass tubes con-
taining mercury, the tubes being sealed into ground-glass
stoppers. As many as eight cells were employed with constants
ranging from about 330 to about 10 in vSiemens' units. A
cell of special type,^ having a very low constant, was used for
obtaining the conductivity of the water.

In order to get a sharp reading in the cells, electrodes were
covered with a fine coating of platinum black in the usual
manner.

Mention only will be made of the viscosity apparatus,
which was of the form used for such work in this laboratory,*
consisting of a picnometer, viscometer, and a stopwatch.

Reagents

Water for making up all solutions and for the final puri-
fication of the acids was obtained by the method of Jones
and Mackay.*

Standard Acid. — Two methods were made use of for stand-
ardizing sulphuric acid; namely, the barium sulphate method,
and a check method which consists in standardizing against
a solution of sodium hydroxide, which, in turn, has been
titrated against very thoroughly standardized hydrochloric
acid. Both methods gave practically identical results.

As to the first one, the acid was made up to approximate
strength (about 0.15 N) from pure concentrated sulphuric
acid. Three 50 cc. portions of the dilute solution were then
run into fairly large beakers and further diluted, and then
heated nearly to boiling. A hot, dilute solution of barium
chloride, containing a slight excess of the salt, was then poured
gradually down the side of one of the beakers containing
sulphuric acid, to which had been added a few drops of hy-
drochloric acid, and which was kept stirred all the while.

1 This Journal, 34, 493 (1905).
i Ibid., 46, 282 (1911).

3 Ostwald-Luther: Physik.-chem. Mess., 2nd Ed., p. 260. Z. physik. Chem, 61,
651 (1908).

4 This Journal, 19, 91. Z. physik. Chem., 22, 237.

Conductivity and Dissociation of Organic Acids 63

In this way a complete precipitation takes place almost at
once. Nevertheless, the beakers were allowed to stand on
a warm sand bath for half an hour. The precipitate was
collected in a Gooch crucible on a layer of purified asbestos.

Standard Alkali. — In order to be able to use the sodium
hydroxide both for titration purposes and for preparing the
sodium salt solutions of the acids, it was necessary to have
an aqueous solution of the alkali, as free as possible from
carbonates and other impurities. To prepare such a solution
the method of H. W. Cowles, Jr.,* is an excellent one.

One hundered grams of sodium hydroxide, purified from
alcohol, was dissolved in 100 grams of conductivity water
(obtained as above described) and the concentrated solution
was allowed to stand in a closed vessel for about a week. By
that time practically all the carbonate, etc., was precipitated
and there was left a perfectly clear supernatant solution of
sodium hydroxide, portions of which were pipetted out and
diluted to the proper strength with conductivity water. The
dilute solution was then standardized by means of the stand-
ard sulphuric acid, and otherwise. When thus prepared,
the solution is perfectly free from carbonate, as is shown by
the fact that it does not give a precipitate of barium carbonate
with barium hydroxide, and that when titrated with indicators,
both those that are sensitive and those that are not sensitive
to carbonates, the results are practically the same.

Organic Acids. — Kahlbaum's so-called pure acids were almost
exclusively employed, and before using them they were all still
further purified by one method or another, according to the
nature of the acid. Their purity was tested by means of their
melting or boiling points, and by titration. No acids were
used whose purity could not thus be established.

Sodium Salts of Organic Acids. — These salts, for the most
part, were prepared by titrating a solution of the acid (usually
about N/128) with a standard solution of sodium hydroxide
exactly to neutrality, using a drop of phenolphthalein as
indicator. Alizarin is also a good indicator and was used
in later work because it is less sensitive to carbonic acid-

1 J. Am. Chem. Soc. 30, 1192 (1908).

64 Wightman and Jones

In a few cases the purified sodium salts were weighed out
and made up to the proper strength.

The potassium chloride used for obtaining cell constant^;
was Kahlbaum's purest. To insure its purity it was pre-
cipitated from a saturated solution by hydrochloric acid,
and then recrystallized three times from conductivity water.
Finally, after drying in an oven for some time, it was
heated at a moderate temperature for ten or fifteen minutes
in a porcelain dish over a Bunsen flame.
Procedure

The molecular conductivities and temperature coefficients
were calculated in the usual manner.

For percentage temperature coefficients Schaller's^ equation,

o._ ^h — f't

was employed.

The equations given above for dissociation and for dis-
sociation constants

/^oo ii-a)V

were employed in calculating these values for weak mono-
basic acids. Constants for the strong acids were calculated
by a method which will be described later.

Cell Constants. — The usual method, already described by
White and Jones,^ was followed, a 0.02 N solution of potassium
chloride being used for the cells in which the electrodes were
fairly wide apart, and a 0.002 N solution for those with the
platinum plates close together. The value //50 = 129.7 for
the conductivity of the 0.02 N solution at 25° was taken from
Kohlrausch, and the value /(500 = 137.9 for the 0.002 N
solution was found by direct measurement.

The cells were standardized about once a month, but very
little change in the constants was found to take place, even
in the course of the whole year. Great care was always taken
not to change the position of the electrodes, which would of
course alter the values of the constants. The following table

1 Z. physik. Chem.. 2, 561 (1888).
« This Journal. 42, 527 (1909).

Conductivity and Dissociation of Organic Acids

is typical of the manner in which the results were tabulated,
W being the resistance in the rheostat, b the distance on the
wire from the point of contact to one end of the wire, and K
the cell constants.

Table I. — Cell Constants

328.

Cell

Solution

VIII

0.02 N

100

5590

328

140

475

328

150

458

328

VII

0.02 N

471

184

459

184

447

184

0.02 N

458

131

445

131

434

131

0.02 N

454

442

430

0.02 N

481

465

450

III

0.002 N

200

445

210

433

220

421

0.002 N

100

443

420

120

469

0.002 N

505

240

470

241

460

245

0.0005 N

451

381

439

381

427

381

0.002 N

250

555

138

260

546

138

270

536

138

0.002 N

250

511

138

260

501

138

270

492

138

0.0005 N

340

473

143

350

466

143

370

452

143

0.0005 N

160

495

143

170

479

143

180

465

143

184.97

131-52

86.38

72.30

44.10

21.94

11.243

2.38]

138.66

143 -44

66 Wightman and Jones

Three readings were taken for each cell, and these agree
to within two or three hundredths of a per cent. All the
later work, in which the Kohlrausch slide- wire bridge was
used, was equally as accurate. In one case cell constants
were taken two days in succession, using entirely new solu-
tions on the second day, and here again the agreement was
practically perfect.

fx for the Sodium Salts of Organic Acids

Whenever it was possible, the /x^ values were obtained
by direct measurement, as was done by White and Jones.
Such values are given in the following table :

Table II. — [x^ for Sodium Salts
Sodium Trichloracetaie

V 0° 15° 25°
1024 41-96 64.75 82.45

i^< = 41 .96 + I .38/ + 0.00952^2
Sodium Cyanacetate

V 0° 15° 25°

2048 44 65 65.43 86.80

iC/ = 44.65 + 1 .52/ + 0.00668/2

Sodium a- Br om propionate

V 0° 15° 25° 35°
1024 42.10 65.04 84.26 105.4
2048 44-94* 69.83 89.61* 108.2*
4096 46.63 70.38 90.40 I I I. 2

iC/ = 44.94 + I .74/ + 0.002/2

Sodium a ,^-Dihrompropionate

V 0° 15° 25° 35°
2048 41.56 64.89 83.24 103.08

jftTj = 41 .56 + 1 .44/ + 0.009/-

Sodium ^-lodpropionate

V 0° 15° 25° 35°
2048 41-54 63.70 81.16 102.8

Kt = 41.54 + i.iSt + 0.0168/-
S odium Levulinate

V 0° 15° 25° 35'»

2048 38.47 59-11 75.13 92.94

Ar< = 38.47 + 1.242/ + 0.00898/2

35'

lOI.

35'

106.

Conductivity and Dissociation of Organic Acids 67

Sodium a-Bromhutyrate

0° 15° 25°

35°

1024

2048

4096

41.50 64.16 81.90
42.46* 65.07 82.53*
43.34 66.33 84.32

Kt = 42.46 + 1.32/ + 0.0115^2

lOI .0

102.6*

103.4

Sodium Hydroxyisohutyrate

0° 15° 25°

35°

2048

40.44 62.36 79 42
Kt = 40.44 + 1 .36/ + 0.00779^^

Sodium Isovalerate

97-74

0° 15° 25°

35°

2048

32.31 50.15 63.87

Kt = 32.31 + 1 . 06^ + 0 . 008 1 1-

79-33

Sodium Caprylate

0° 15° 25°

35°

2048

42.67 61.85 7761
Kt = 42.67 + i.oggt + 0.0120^2

Sodium Benzilate

95-77

0° 12° 25°

35°

1024
2048

4096

35 I 50.5 69.5
36.3 52 -2 71.5
35-8 51.7 70.8
Kt = 363 + 1.17^ + 0.0095/2

Sodium Chlorhenzoate

86.2
88.9
88.1

0° 15° 25°

35°

2048

38.03 58.97 75.47

K, = 38.03 + 1.30/ + 0.0078/2
Sodium p-Nitrobenzoate.

93-18

0° 12° 25°

35°

1024

2048
4096

38.85 55.48 75.71
39.78 56.28 76.48
38.91 5501 75.86
Kt = 39-7^ + I Hi + 0.0133^'

Sodium 1 ,2 ,4-Dinitrobenzoate

93 30
95 80

94 00

V 0°

15° 25° 35°

2048 37 - 80 58 - 25 74 - 77 I 92 • 90 (by titration)
^ ^' ^ ^ '^ '^ ( 92.83 (from dry salt)

-^/ = 37-80 + 1 .24/ -f 0.0095/2

68 Wightman and Jones

Sodium 1 ,3,5-Dinitrobenzoate

V 0° 15° 25° 35°

2048 37.83 58. 30^74. 60 92.93 (by titration)
Kt\= 3783 + 1.24^ + 0.0095^2

Sodium 1 ,3,5-Dinitrohenzoate

V 0° 15° 25° 35°

1024 36-74'!r56-53 7i-8i 87.70]

2048 37.46 5756 7313 89.64 M solution of dry salt);

4096 37.98 58.10^^74.60 91.70 J

Sodmm i ,2 ,4-Dihydroxybenzoate

V 0° 15° 25° 35°

2048 39.64 60.72 77.49 95-11

Kt = 3964 + 1.337^ + 0.00708^2

Sodium 1 ,2 ,^-Dihydroxyhenzoate

V 0° 15° 25° 35"
2048 39.36 60.58 77.52 95.62

K-t = 3936 + 1-324^ + 0.0081/^

Sodium p-Sulphamidohenzoate

V 0° 15° 25° 35°
2048 39.30 60.23 76.57 94.00

-K'< = 3930 + 1.31^ + 0.0072^2

In nearly every one of the above cases a concentration
of N/2048 alone was used, since it was found by White and
Jones that the n^ values of the sodium salts generally occur
at this concentration.

The fact that the conductivity of a salt made up by titration
and that of the same salt made up from the dry solid agree
shows that the titration method is reliable.

It is seen, as would be expected from previous work, that
the n^ values of the sodium salts of isomeric acids are prac-
tically identical, as, for example, sodium 1,2,4- ^^^ Ij3>5"
dinitrobenzoates, and sodium 1,2,4- ^^^ 1,2,5-dihydroxy-
benzoates.

Further, the pL^ values of the strong acids calculated from
the sodium salts are identical with the maximum values
of the acids themselves. Trichloracetic and 1,2,4-dinitro-
benzoic acids are examples illustrating this point.

Conductivity and Dissociation of Organic Acids 69

Moreover, we see from the tables that those salts with the
largest number of atoms in the anion have the highest }i^
values. The curves expressing the p.^ values for these acids
bring this point out still more strikingly.

-"oo ^^^'^^^ of ^^^ Acids

The method by which these values are calculated has already
been referred to. Essentially, it is merely a subtracting of
the migration velocity value of the sodium ion from the p.^
value of the sodium salt, and an addition of the migration
velocity value of the hydrogen ion. In actual practice, the
following equation was used :

/loo (acid) = /<oo (HCl) + /^oo (Na salt of acid) — pcc (NaCl)

iVs the values of conductivity between 0° and 25° were,
throughout the present work, obtained almost exclusively
at 15°, the equations given by White and Jones for calculating
the /i^ values of sodium chloride and of hydrochloric acid were
made use of in order to obtain the p^ values of the acids
worked with at this temperature.

For sodium chloride at 15°,

/«oo = (63-04 + 2.04^ + 0.00823^-) = 95.49
and for hydrochloric acid at 15°,

/«oo = (2454 + 6.06^ — 0.00776^2) = 334.5

In the table given below are presented the fi^ values of
all the acids with which we worked (except dichlorphthalic,
tetrachlorphthalic, and meconic acids).

The values for the acids marked with an asterisk were
not obtained by means of the method stated above, since
they are dibasic and their sodium salts do not yield a maxi-
mum value of conductivity at any dilution up to N/4096,
which is the highest dilution with which we worked.

A curve, in which the ordinates represent /i^ values and
abscissas the number of atoms, was plotted for a large number
of monobasic acids, and by placing the dibasic acids in their

Wighiman and Jones

proper position on this curve (according to the number of
atoms present) their ji^ values can be obtained (see Fig. I).

l8 20 22

Number of Atoms
Fig. I — Limiting Conductivities

Table III. — //qo Values of the Acids

Acid /iooO° /loo 12° /ioolS" /loo 25°

Trichloracetic 224 .8 ... 303

Cyanacetic 271 .1 ... 304

Benzilic 218.7 280.5

«-Brompropionic 229.0 ... 308

a,/?-Dibrompropionic 223.9 ... 303

^-lodpropionic 223 .9 ... 302

Levulinic 220.9 . . . 298

a-Brombutyric 230. 7 . . . 304

Hydroxyisobutyric 222 .8 . . . 301

Isovaleric 214.7 ... 289

Caprvlic 225 .1 ... 300

/-Tartaric* 221 .0 ... 298

Thiodiglycolic* 221 .6 ... 300

355
360

344
363
356
354
348
357
352
337
350
350
351

*i«35
406
410
392
415
407
406
396
407
401
383

399
399
401

Conductivity and Dissociation of Organic Acids

Table III. — (Continued)
Acid /<„ 0° /i^ 12° ^

Tricarballylic* 219.9 • •

/)-Nitrobenzoic 222 .2 284

1,2,4-Dinitrobenzoic 200.0

1,3,5-Dmitrobenzoic 220.2

o-Chlorbenzoic 220.4

1,2,4-Dihydroxybenzoic 222 .0

1,2,5-Dihydroxybenzoic 221.8

/j-SuIphamidobenzoic 221. 7

Benzenesulphonic 228.0

/>-Toluenesulphonic 210.6 269

m-Nitrobenzenesulphonic. . . . 204 .5
1,2,4-Nitrotoluenesulphonic 200.5 ..

Camphoric* 218.3 279

Uric* 221 .0 . .

Cyanuric

RESULTS
Trichloracetic Acid, CCI3COOH
The acid is very hygroscopic, so it was not dried thoroughly
after recrystallization from water, but was made up to
approximate strength and then standardized by titration.

Ostwald^ determined the conductivity of the acid at 25°
and comparison of his results with our own at the same temp-
erature shows a fairly close agreement, especially at the higher
dilutions :

Ostwald

296

347

6 396.

6 . ,

349

7 399-

297

347

9 396.

297

347

4 396.

301

34H

7 397-

299

350

7 399

• 299

350

7 399

■ 299

349

8 39»

■ 309

359

0 410

.7 . .

332

7 379

.16° 2 75

223

5 369

. 276

31B

4 361

.8 ..

344

5 392

. 298

350

0 399
405

tiv 25°

From Tables IV and VI

,1V 25°

3230

322.46

128

341.0

344 90

512

353-7

353 96

1024

356.0

355-94

The relation of trichloracetic acid to acetic acid, mono-
chloracetic acid, and other acetic acid derivatives, was brought
out with sufficient clearness by Ostwald, and need not be dis-
cussed here. The acid itself, however, is of especial interest,
because it is a very strong electrolyte (very nearly as strong
as hydrochloric acid) .

We see from Table VI that the acid is entirely dissociated
at N/2048. The n^ value {n^ = 358) given by Ostwald

» Z. physik. Chem., 3, 177 (1889).

Wightman and Jones

is greater than any of the conductivity vakies given by him,
and, therefore, his percentage dissociation values do not reach
a maximum; but we find that fi^ , as obtained from the sodium
salt, agrees ver}'^ closely with the maximum value for conduc-
tivity of the acid at 25° (at N/2048).

Table IV. — Molecular Conductivity

0°

15°

25°

35°

193.02

256.24

298.40

334-67

208.75

277.67

322.46

363.69

128

221.73

297.62

344-90

389-83

512

223.65

302.33

353-96

403 -45

1024

224.77

303 • 94

355-94

406.44

2048

221 .52

300 . 2 1

349-57

397-46

Table V. — Temperature

Coefficienis

0°-15

15°

-25°

25°

-35°

Cond.

Per Cond

Per Cond.

Per

units

cent. units

cent.

units

cent.

425

2 . 20 4.22

1.65

I .22

4.60

2.20 4.48

I. 61

1.28

128

506

2.28 4.73

1-59

1.30

512

5-25

2-35 516

1.70

I .40

1024

5.28

2.35 5 20

I. 71

1.42

2048

5 24

2-37 4-94

1.65

1-37

/<co

Values (measured directly)

0°

15°

25°

35°

1024

224.77

303 • 94

355-94

406.44

/^oo

Values (from sodium, salt)

0°

15°

25°

35°

1024

223.67

303 . 86

355-65

405-91

Table VI.— Percentage

Dissociation

0°

15°

25°

35°

85.87

84-31

83-83

82.34

92.87

91.36

90.59

89.88

128

98.65

97.92

96.90

95 91

312

99 50

99-47

99-44

99.26

1024

100.00

Cyanaceiic Acid, CH2(CN)C00H
The acid was easily purified by recrystallization from water,
and was dried over sulphuric acid in a vacuum desiccator.
The pure, dry acid has a melting point of 65°.

Conductivity and Dissociation of Organic Acids

It is of interest to note the effect of the cyanogen group
when introduced into acetic acid, and also to compare its effect
with that of other groups.

Acetic Acid

Monochloracetic Acid

Monobromacetic Acid

White and Jones'

,a-y25° a 25°

Ostwald2

,iv2S°

a 25°

^v 25°

a 25°

4 342

1.20

8.699

68.7

18.95

128

17. II

127

122.3

33-7

512

33 24

205

199.2

550

1024

45 87

249

241.2

66.6

2048

63.00

A'3

= 0

184

' 1

C =

= 15-5

= 13-8

Phenylacetic Acid

Cyanacetic

Acid

White and Jones'*

From Tables VII and IX

^L■v25°

a 25°

ti-v

25°

a 25°

16.54

I4I5

405

106

29 58

128

27.96

8.01

178

49.68

512

52.39

14 -97

259

72.12

1024

71.63

20.52

291

80.99

2048

95 50

27.36

314

87.30

K =0.545

^ = 36.

The strong negative character and influence of the cyanogen
group is strongly brought out by the fact that the constant
for cyanacetic acid is but Uttle less than that for monobrom-
acetic acid, and that it is very nearly two hundred times as
great as that for acetic acid.

Table VII.-

— Molecular Conductivity

0°

15°

25°

35"

38.27

51.66

59-53

66.15

68.70

92.26

106.47

118.79

128

114.23

154.10

178.86

199.67

512

164.90

223.37

259 64

293.00

1024

187.49

252.59

291.56

332.00

2048

199.90

2 70 . 40

314-30

356-60

1 This Journal, 44, 165 (1910).

2 Z. physik. Chem., 3, 178 (1889).

' Note. — K, throughout this work, means constant X lo*.
* This Journal, 44, 168 (1910).

Wightman and Jones

Table VIII. -

-Temperatm

■e Coeffi

dents

0°-15°

15°-

25°

-35°

Cond.

Per

Cond.

Per

nd.

units cent.

units

cent.

units

cent.

0.893 2

•33

0.787

152

662

I . ir

I .67 2

1.42

1-54

I. 16

128

2.66 2

•33

2.48

1. 61

I. 16

512

3 . 90 2

■36

3^63

1.62

I. 17

1024

4-35 2

•32

3^97

157

I 39

2048

4.70 2

■35

4-39

1.62

I 35

Ta6/e /X.-

-Percentage Dissociation

0°

)°

35°

16.86

•

16.13

30.26

28.97

128

50.31

48.70

512

72.63

71.46

1024

82.58

80.98

2048

88.04

86.97

Table X.-

-Di

SSOCl

ation Constants

10'

0°

15°

25°

35°

128

512

1024

2048

Benzilic or Diphenylglycolic Acid, (C6H5)2C(OH)COOH
The ordinary method of recrystallization was used for

purifying this acid (melting point, 150°).

Here we have the effect of both hydroxyl and phenyl groups,

and the result is an acid (ATjeo = 9.20) which is fifty times

as strong as acetic acid {K^rp = 0.184)^ and six times as strong

as glycolic acid (isTjso = 1-52).^

Table XI. — Molecular Conductivity

0°

12°

25°

35°

128

63.8

81.7

IOI.5

115. 0

512

106.4

138.3

1695

186.9

1024

133^6

169.8

208.4

237.1

2048

1523

193.0

233 7

260.6

This Journal, 44, 168 (1910).

Ostwald: Z. physik. Chem., 3, 183 (1889).

Conductivity and Dissociation of Organic Acids 75

Table XII. — Temperature Coefficients

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

128

I 52

2.38

1.50

1.84

1-35

0.93

512

2.66

2.50

2.24

1.62

1.94

1.04

1024

3.02

2.26

2.97

1-75

2.87

1.23

2048

3-39

2.23

313

1.62

2.69

I 03

Table XIII .

— Percentage

Dissociation

0°

12°

25°

35°

128 .

29.17

29.08

29-45

29.27

512

48.64

49.24

48.60

47-57

1024

61.08

60.44

60.46

60.20

2048

69.63

68.71

67.81

64.82

Table XIV.—

Dissociation Constants X

10'

0°

12°

25°

35°

128

9.38

9.10

9.60

9.46

512

9.00

932

8.97

8.43

1024

936

9.02

8.89

2048

7.80

7-37

6.97

5-83

a-Brompropionic Acid, CH3.CHBr.COOH

The acid was purified by distillation and was then a clear
colorless liquid, boiling at 205°. The calculated quantity
for making up a N/32 solution was accurately weighed out
into a weighing bottle and then poured into a flask and diluted.
A portion of this dilute solution, enough to make a N/1024
solution, was titrated in order to form the sodium salt, and
also to be sure of its normality.

Table XV. — Molecular Conductivity

0°

15°

25°

35°

38.00

49-38

58.86

61.5

128

77.10

100.00

114. 4

125.9

512

124.7

164. I

186.8

206.7

1024

151-7

200.8

229.5

257.0

2048

I7I-5

227.4

262.0

295-3

Wightman and Jones

Table XVI.-

■—Temperature Coefficients

C-IS"

15°-

25°

-35°

Cond. Per

Cond.

Per

Cond.

Per

units cent.

units

cent.

xinits

cent.

0 . 76 2 . 00

0.65

I-3I

0.56

I .00

128

I 53 1-99

1.44

1.05

0.92

512

2 . 63 2 . I I

2.27

1-38

1-99

1.07

1024

3.27 2.16

2.87

1-43

2.75

I. 17

2048

3.70 2.16

3 46

1.52

3.33

1.27

Table XVII.

— Percentage

Dissociation

0°

15°

25°

35°

16.60

15 -99

15-38

14.81

128

33 67

32.37

31-47

30.33

512

54-45

5313

50.21

49-79

1024

66.25

65.01

62.98

61 .90

2048

7752

73.62

72.06

71.12

Table XVIII.—

Dissociation Constants X

10'

0°

15°

25°

35°

10.3

10.2

8.7

8.0

128

134

13.2

11-3

10.3

512

12.7

13 I

9-9

9.6

1024

12.7

13.5

10.6

9.8

2048

13. 1

II. 9

II. 4

8.4

B-Iodpropionic Acid, CH2ICH2COOH

Several recrystallizations were necessary in order to purify
this acid, as it decomposes on standing in the presence of
light, and the acid as it came from Kahlbaum was found to
be quite impure. Its melting point when pure is 85°.

There is also quite a rapid decomposition of the acid when
its solution is placed in the cells in the presence of the plati-
num electrodes, especially at 35°. This made it necessary
to refill tlie cells with fresh solution after the measurements
were taken at each temperature.

The decomposition just spoken of may be an explanation
of the increase in its temperature coefficients between 25°
and 35°, as compared with those from i5°-25°.

Conductivity and Dissociation of Organic Acids
Table XIX. — Molecular Conductivity

(

15°

25°

35°

8.42

9-73

II . 12

16.81

19-37

21.98

128

31.86

36.67

41.69

512

59-47

68.42

78.04

1024

78,67

91-05

104.24

2048

102.87

118.35

135-40

Tabic :

\^X.—

Temperature

Coefficients

0°-l

5°

15°-:

'5°

'-35°

Cond.

Per

Cond.

Per Cond.

Per

units

cent.

units

cent. units

cent.

0. 141

2.23

O.I3I

1.56 0

-139

1-43

0.283

2.25

0.256

1.52 0

.261

1-35

128

0.538

2.26

0.482

I. 51 0

.502

1-37

512

I .007

2.27

0.895

I. 51 0

.964

1. 41

1024

1-337

2.28

1.238

1-57 I

-319

1-45

2048

1-755

2.29

1-548

I. 51 I

-705

1-44

Table

XXI.

— Percentage Dissociation

0°

15°

25°

35°

2 . 836

2.781

2.752

2-744

5-657

5-552

5-478

5-425

128

10.71

10.56

10.37

10.29

512

19.97

19.64

19-35

19.26

1024

26. 38

25.98

25-75

25-73

2048

34-46

33-98

33-47

33-42

Table X^

:ii.—

Dissociation Constants X 10*

0°

15°

25°

35°

1 .04

1 .00

0.97

1 .04

1 .02

0.99

0.97

128

1 .00

0.97

0.94

0.93

512

0.97

' 0.94

0.91

0.90

1024

0.92

0.89

0.87

2048

0.89

0.85

0.82

Levulinic or ^-Acetylpropionic Acid, CHjCOCCHJjCOOH

The melting point of the acid is 32°. 5-33°, so that it was
purified by solidifying it, and pressing on a porous plate.

The relation of levulinic acid to those just preceding is
brought out by the following tables :

Wightman and Jones

Propionic Acid

a-Brnmpropionic Acid

White and Jones*

From Tables XV and XVII

/'r 25°

a 25°

,iTj 25°

a 25"

3 70

1.05

7-44

2.10

58^86

1538

128

14-57

4.12

114.40

31-47

512

28.40

8.02

186.40

50.21

1024

38.94

II .00

229.50

62.98

2048

53-47

15.10

262.00

72.06

K =

= 0.138

K ==

10.4

U ^'

fi-Iodpropionic Acid
From Tables XIX and XXI

Levulinic Acid
From Tables XXIII and XXV

128

512

1024

2048

-73
-37
.67
.42
-05
•35
K

5
10

25
33
0.977

-752

-478

-37

-35

•75

•47

857

37
85
24
K =

-39
-79
.48
•44
31
.02

0.243

As we would expect, propionic acid has the lowest values
for conductivity, for dissociation and for the constant; and
levulinic acid, with an acetyl group in place of one of the
hydrogens of propionic acid, has the next higher values. The
bromine substitution product is seen to be one hundred times
as strong as that containing iodine, which, in turn, is over
four times as strong as the acetyl derivative.

Table XXIII. — Molecular Conductivity

0°

15°

25°

35°

2-939

4. 114

4.851

5-539

5-85

8.24

9.71

11 . 10

128

11-57

16. 13

19.08

21.84

512

22.06

30.78

36.37

41.68

1024

29.81

41.92

49-85

56 -99

2048

3941

56.31

66.24

76.53

« This Journal, 44, 165 (1910).

Conductivity and Dissociation of Organic Acids

Table XXIV.

— Temperature Coefficients

0°-I5

15°-

25°

-35°

Cond.

Per

Cond.

Per Cond.

Per

imits

cent.

units

cent. units

cent.

0.0783

2.66

0.0737

1.79 0

0688

1.42

0. 160

2.74

0.147

1.78 0

139

1-43

128

0.304

2.63

0.295

1.83 0

276

1-45

512

0.581

2.63

0.559

1.82 0

531

1 .46

1024

0.807

2.71

0.793

1.89 0

714

1-43

2048

I . 126

2.86

0.993

1 .76 I

029

1.55

Tabic XXV.

— Percentage

Dissociation

0°

15°

25°

35°

1-33

1.38

1-39

1 .40

2.65

2.76

2.79

2.80

128

5 24

5-41

5-48

5 50

512

9 99

10.32

10.44

10.50

1024

13 50

14.06

14-31

14.36

2048

17.84

18.89

19.02

19.28

Table XXVI.—

Dissociation Constants X j

ro"

0°

15°

25°

35°

0.224

0.234

0.245

0.249

0.225

0.238

0.250

0.252

128

0.226

0.235

0.248

0.250

512

0.217

0.226

0.238

0.241

1024

0.206

0.219

0.233

0.235

2048

0. 189

0.209

0.218

0.225

a-Brombutyric Acid, CHgCHjCHBrCOOH
This acid is very similar in almost every respect to a-brom-
propionic acid; it is a liquid, boiling at 215°, which was purified
by distillation, and was weighed out and made up to proper nor-
mality in the usual way. When we compare its conductivity
and dissociation with a-brompropionic acid we see that the
conductivity, dissociation, and constant of the latter, though
somewhat smaller, are yet not very different.

Table XXVII. — Molecular Conductivity

0°

15°

25°

42.75

54 70

61 .0

,128

84.94

109.5

122.8

512

133 7

173.2

195-2

1024

160.6

209.3

239.8

2048

180.7

238.0

2750

35°
66.42

134-3
214.9

266.0
305 -9

Wightman and Jones

Table XXVIII. — Temperature Coefficients

Cond.
units

Per
cent.

Cond.
units

Per Cond.
cent. units

Per
cent.

32
128
512

1024

2048

0.793
I .64
2.63

3-25
3.82

I
I
2
2

86
93
97
02
II

0.630

1-33
2.20

3 05
3 70

I. 15 0
I. 21 I

1.27 I
I .46 2

1-55 3

•542
15

97
62

0.889
0.937
I. 01
1.09
I . 12

Table XXIX.

—Percentage Dissociation

0°

15°

25°

35°

32
128
512

1024

2048

18
36
57
69

•53
82

-97
.61

•32

17.99
36.00
56-92
68.82
78.26

17.06

34-35
54 • 59
67-08
76.91

16.30

32 -97
52.74
65-30
75 09

Table XXX. — Dissociation Constants X

10'

0°

15°

25°

35°

I3-I

13.2

II .0

10. 1

128
512

16.8
15.6

17.2
16.4

14.0
12.8

12.7
II 5

1024

2048

15-6
13-8

17.0
16.6

13.2
12.5

12.0
II .1

HydroxyisobtUyric Acid, (CH3)2C(OH)COOH

Hydroxyisobutyric acid sublimes at 50° and the fresh
sublimate melts at 79°.

Ostwald obtained the conductivity of this acid at 25°,
and for the sake of comparison his values are given below :

Ostwald'

From Tab

les X

XXI an

d XXXII

^v 25°

«25°

ot25°

a 25°

20.05

5.65

20. 19

5-75

128

38.86

10.95

39- 18

II. 15

512

73-49

20.70

73-16

20.82

1024

99 52

28.05

97.00

27.61

K =

1.06

K =

= 1.06

The conductivities agree to within five- tenths of a per cent.,
except at N/2048. The dissociation given by Ostwald is less

1 Z. physik. Chem.. 3, 197 (1889).

Conductivity and Dissociation of Organic Acids

(except at N/204S), since he used the value /.i^ = 355, whereas
our own value of //^ is 352.6. The mean value for the constant
in both cases, however, is the same.

As to the relation of hydroxyisobutyric acid to butyric
acid, its constant {K = 1.06) is a little less than seven times as
great as that of butyric acid {K = 0.163). a-Brombutyric
acid has a constant {K = 12.7), about twelve times as large
as the hydroxyl derivative.

Table XXXI.

— Mo lee ular Conductivity

0°

15°

25°

35°

6.075

8-553

10.147

11. 576

12. II

17.04

20.19

22.97

128

23 50

33 04

39.18

44 65

512

44.06

61.74

73-16

83.41

1024

58.80

81.95

97.00

II 1 . 60

2048

76.78

106 . 95

126.20

144.07

Table XXXIL

— Temperature Coefficients

0°-15°

15°-

-25°

25'

-.35°

Cond. Per

Cond.

Per Cond.

Per

units cent.

units

cent. units

cent.

0.165 2

.72

0.159

1.86 0

143

I. 41

0.329 2

0.315

1.85 0

278

1.38

128

0.636 2

0.614

1.86 0

547

1.40

512

I . 18 2

I. 14

1.85 I

I . 40

1024

1-54 2

I. 51

1.84 I

1.50

2048

2 .01 2

1.93

1.80 I

1.41

Table XXXIII

. — Percentage Dissociatio n

0°

15°

25°

35°

2.75

2.838

2.89

5-47

5.653

5-75

5-74

128

10.62

10.96

1 1. 15

II. 15

512

19.92

20.48

20.82

20.84

1024

25-58

27.19

27.61

27.88

2048

34 70

35.48

35 92

36.00

Table XXXIV.—

Dissociation Constants X

10*

0°

15°

25°

35°

0.97

1.05

1.08

0.99

1.06

I . 10

1 . 10

128

0.99

1. 17

1.09

512

0.97

1.03

1.07

1024

0.94

0.99

1.03

1.08

2048

0.90

0.95

0.98

0.99

Wightman and Jones

Isovaleric Acid, (CHgj^CH.CH^.COOH

This acid, boiling at 176°, was fractionally distilled and
diluted in the usual manner for nonvolatile liquids. Its re-
lation to the other acids of the aliphatic series is discussed
under caprylic acid.

Table XXXV. — Molecular Conductivity

0°

15°

25°

35«'

2-474

3.229

3.666

4.044

5 052

6.591

7-493

6.262:

128

9.832

12.92

14.69

16. 19

512

19.023

24.81

28.13

31 03

1024

26.264

34.22

38.49

42.98

2048

34.804

45.28

51-69

58.02

Table XXXVI. -

—Temperature Coefficients

0°-15°

15°-

25°

-35°

Cond. ]

Per

Cond.

Per Cond.

Per

units cent.

units

cent. units

cent.

0 . 0503 2

•03

0.0473

1-35 0

0378

1.03

0.1026 2

.04

0 . 0902

1.36 0

0769

1.03

128

0 . 2044 2

.07

0.1767

1.37 0

1506

1-03

512

0.3856 2

•03

0.3318

1-34 0

291

1.03

1024

0.530 2

.02

0.427

1.25 0

449

I. 16

2048

0.699 2

.06

0.640

1 .41 0

634

I .22

Table XXXVII

. — Percentage Dissociation

0°

15°

25°

35°

I 15

1. 117

1.09

I .06

2-35

2.279

2.22

2.16

128

4-59

4.467

4 36

4.22

512

8.86

8.579

8.34

8.10

1024

12.23

11.83

II .42

II .22

2048

16.21

15.66

15-33

15-14

128

512

1024

2048

Table XXXVIII. — Dissociation Constants X 10*

0° 15° 25° 35°

0.167 0.154 0.150 0.142

0.177 0.162 0.158 0.149

0.172 0.159 0.155 0.145

0.168 0.154 0.148 0.139

0.166 0.151 0.144 0.138

0-153

138

0.136

O.I32'

Conduciivity and Dissociation of Organic Acids 83

Caprylic Acid, CH3(CH2)eCOOH

The easiest method of purifying this acid, which melts at
16. °5, was found to be to soHdify it and press out the solid
on a porous plate. Being a nonvolatile liquid at ordinary
temperatures, solutions of it were made up in a manner similar
to that employed for a-brompropionic and a-brombutyric
acids.

Table XXXIX

— Molecular Conductivity

0°

15°

25°

35°

512

27.79

31.07

1024

24 -39

32:76

37 84

42 -35

2048

32.84

44.08

51.08

56.89

Table XL.-

-Temperature

Coefficients

0°-15°

15°-

-25° 25 =

-35°

Cond. Per

Cond.

Per Cond.

Per

units cent.

units

cent. units

cent.

512

0.328

I. 18

1024

O.55S 2.29

0.508

1-55 0.453

I. 17

2048

0.749 2.28

0.700

1.58 0.571

I . 12

Tabic XLL-

—Percentage Dissociation

0°

15°

25°

35 »

512

7.96

7.80

1024

10.84

10.89

10.84

10.64

2048

14.60

1465

14-63

14.29

Table XLIL — Dissociation Constants X 10*
V 0° 15° 25° 35°

512 ... ... o. 134 0.129

1024 0.129 0.130 0.129 0.124

2048 0.122 0.123 0.123 0.116

We would expect, since caprylic acid is one of the higher
aliphatic acids, that its conductivity and dissociation would
be less than that of those aliphatic acids with less complex
molecules, and such is the case, as was first shown by Ostwald,^
and as can be seen by comparison of its values with the values
in the following tables :

1 Z. physik. Chem., 3. 176 (1889).

WighUnan and Jones

512
1024

2048

Acetic Acid
White and Jones*

« 25°

9.21

12 . 71

17-45
0.175

33 24

45 87

63.00

K =

Propionic Acid
White and Jones

Hv 25°
28.40

38 • 94

53-47
K =

a 25°

8.00

II .00

15.10

0.133

n-Butyric Arid
White and Jones

n 25°
29.86
41 .22
59.20

K =

8,
II ,
16
0.163

512
1024
2048

Valeric Acid
Ostwald^

Hv 25°

157
30.4
41.9

K =

« 25°

4-44
8.59
11.83
0.161

Isovaleric Acid

From Tables

XXXV and XXXVII

w25°

28.13

38.49
51-69

K =

a 25°

8.34
II .42

15-33
0.149

Caprylic Acid

From Tables

XXXIX and XLI

/iz;25° a 25°

27.79 7 96

37.84 10.84
51.08 14.63

K = 0.129

\-Tartaric Acid, H00C(CH0H)2C00H

This acid was recrystallized several times from water, and
finally dried at 105° in a hot-air bath. The N/8 solution of
the pure acid was tested as to its optical activity with a
polariscope. The amount of rotation indicated that the
acid was pure.

The constant of this acid {K^^o = 10.7) at 25°, as compared
with that obtained by White and Jones for racemic acid
(/^250 = 10.8), indicates a close relationship in chemical activity
as well as in structure. Ostwald's values of conductivity,
dissociation, etc., of both d- and /-tartaric acids are given
below, and it is seen that though his values of conductivity
are not very concordant with our own, the values for the
constant are not very different.

d-Tartaric Acid

-Tartaric Acid

Ostwald

Oslwald

w25°

a 25°

A- 25°

nv 25 °

a 25°

A- ,5c

57.60

16.20

9.8

57-6

16. 19

9-7

128

106.2

2985

9-9

105.6

29.7

9-7

512

184.5

51-80

10.9

183.2

51-5

10.7

1024

236.0

66.0

12.7

234.0

65.8

12.3

2048

291 .0

Si. 8
K = 9.7

17.9

289.5

81.4

K = 9.7

17.4

1 This Journal. 44

165 (1910).

■■^ Z. physik. Chem.. 3, 175 (1889).

Conductivity and Dissociation of Organic Acids

We notice that the constants given by Ostwald increase
rapidly with dilution. The same is true of our own, and is
no doubt due to the fact that the acid is dibasic and the second
hydrogen begins to dissociate at about N/128. Ostwald 's
figure for the constant, K^^° = 9.7, includes only the values
at N/32 and N/128. If all the values except that at N/2048
are averaged, then his constant would be K^^° = 10.6, which
agrees closely with our own, K^^ = 10.7, averaged in a similar
manner.

Tabic XLIIL-

—Molecular

Conduct

'vity

0°

15°

35°

15 64

22.58

31.12

34.18

49 03

67 65

128

62.81

90. 12

107

123-5

512

109.3

156.8

186

213.0

1024

136.0

192.0

229

261.6

2048

171-7

241.0

285

325.5 '

Table XLIV.-

-Temperature Coeffic

ients

0°-15°

15°-25

°-35°

Cond. Per

Cond.

Per

Cond.

Per

units cent.

units

cent.

lits

cent.

0 . 463 2 . 96

0.435

1-93

419

1-55

0 . 990 2 . 94

0.969

1.98

893

1-52

128

1.89 3.00

1-73

1.92

1-47

512

3.17 2.90

3.01

1.92

1.40

1024

3-73 2 . 75

3-74

1-95

1 .40

2048

4.62 2.69

4-44

1.84

4.09

1 .40

/ioo Values (from graph)
0° 15° 25° 35°

221.0 298.8 350.0 399-9

K^^c = [221.0 + (5-28 X 15) — (0.00486 X 225)] = 299.1

Table XLV.-

—Percentage

Dissociation

0°

15°

25°

35°

7.08

7-56

7.69

7-78

15-47

16.41

16.78

16.91

128

28.42

30.16

30.68

30.88

512

49.46

52.48

53 40

53-25

1024

61.54

64.26

65 -54

65.40

2048

77.69

80.66

81.54

81.38

■86 Wighbnan and Jones

Table XLVL-

—Dissociation Constants X 10*

0°

15°

25°

35"

6.7

7-7

8.0

8.2

8.9

10. 1

10.6

10.8

128

8.8

10.2

10.6

10.8

512

9-5

II 3

12.0

II. 8

1024

9.6

II -3

12.2

12 . 1

2048

13.2

16.4

17.6

17 4

Thiodiglycolic Acid, S(CH2COOH)2

Thiodiglycolic acid (^25° = 4-8), as Ostwald^ showed, k
weaker than digly colic acid {,K^^° = ii.o), the introduction of
sulphur in the place of oxygen decreasing the dissociation.
Both acids, however, are stronger than either glycolic acid
(K25° = 1.52) or thioglycolic acid (^25° = 2.25), which bear
the opposite relation to each other as compared with diglycolic
and thiodiglycolic acids.

Ostwald* calculated the constants of the latter by a formula
somewhat different from that representing the ordinary
dilution law :

{i — a)V'

From this equation he obtained the value /Cjo" == 4-8 for
thiodiglycolic acid, whereas we found the value K'js" = 6.51
for the same acid, calculated by means of the ordinary
dilution law. A glance at Table L will show that the
constants thus obtained are fairly good up to N/2048, where
evidently the second hydrogen begins to dissociate appreciably.

If the above equation is used we obtain the values :

K2S°

K^"

128

607

257

113

512
1024
2048

504
338

255

from which it is readily seen that the ordinary law can be
better applied to our values.

1 Z. physik. Chem,, 3, 187 (1889).

Conductivity and Dissociation of Organic Acids

Table XLVII. — Molecular Conductivity

15°

25° 35°

21 .40

25.00 28.16

39 38

46.27 52.18

128

72.42

84 .80 96 . 00

512

127.47

148.93 169.03

1024

119

164.00

191.30 216.13

2048

152

207.38

242.65 275.70

Table

XLVIII.

— Temperature Coefficients

0°

-15°

15°

-25° 25°-35°

Cond.

Per

Cond.

Per Cond. Per

units

cent.

units

cent. units cent.

0.38

2.42

0.36

1.68 0.316 1.26

0.70

2-43

0.69

1-75 0.591 1.28

128

I-3I

2.48

1.24

I. 71 I. 12 1.32

512

2.28

2.44

2.15

1.68 2.01 1.35

1024

2.94

2.45

2.73

1.67 2.48 1.30

2048

3.68

2.42

3-53

1.70 3.31 1.36

fXao Values {from

graph)

0°

15°

25° 35°

2048

21.6

300.2

350.1 401.0

Kt = [221.6 + (532 X 15) — (0.00544 X 225)] = 300.2
Table XLIX. — Percentage Dissociation

V 0° 15° 25°

7.09

713

7.12

7.02

13 03

13 II

1317

13.01

128

23 83

24.12

24.14

23 -94

512

42. 12

42 -45

42 -39

42 15

1024

54 14

34.61

54 46

53 90

2048

68.70

69.06

69. 10

68.76

Table L.-

—Dissociation Constants X 10*

0°

15°

25°

35°

6.77

6.85

6.82

6.63

6. 10

6.18

6.24

6.08

128

5.83

5-99

6.00

589

512

5-99

6. II

6.09

6.00

1024

6.24

6.33

6.36

6.16

2048

7.36

7-53

7-54

7-39

Tricarballylic Acid, HOOC.CH(CH2COOH)2
The acid was recrystallized in the usual manner. It melts
at 165°.

VVightman and Jones

Walden^ and Walker^ both obtained the conductivity of
this acid at 25°. Their values are given in the following
tables :

Walden

Walker

V nv 25°
32 28.33

128 54-8

512 102.7

1024 1390

K - 2.2

33-4

1336

534 0

1068.0

29.1

55-6
103.0
135-8
24

The fact that the constants given in Table LIV agree so
well seems to indicate that the acid dissociates all the way
up to N/2048 as if it were a monobasic acid.

Table LI. — Molecular Conductivity

0°

15°

35°

8.26

11-73

16.24

16.39

23-41

32.38

128

31.82

45-13

62.28

512

59-35

83-65

115 38

1024

78.79

110.53

131

152.40

2048

103.03

143-90

170

196.65

Table LIL-

-Temperature Coefficii

mts

0°-15°

15°

-25°

25°-

-35°

Cond. Per

Cond.

Per

Cond.

Per

units cent.

units

cent.

units

cent.

0.231 2.79

0.231

1.97

218

1-55

0 . 468 2 . 86

0.461

1-97

436

1-56

128

0.887 2.79

0.885

1.96

830

1-54

512

1.63 2.74

1.63

1-95

1-54

1024

2.12 2.68

2 . II

1. 91

1-57

2048

2.73 2.65

2.70

1.87

I-5I

/^oo Values (determined graphically)
0° 15° 25° 35°

219.9 296.7 347.6 396.8

= [219.9 X (5.22 X 15) — (0.00438 X 225)] = 297.2

Z. physik. Chem., 10, 563 (1892).
J. Chem. Soc. •!, 707 (1892).

Conductivity and Dissociation of Organic Acids
Table LIII. — Percentage Dissociation

from

0°

graph

equation

25°

35°

3.76

3-95

4.04

4.09

7-45

7.89

7.88

8.07

8.16

128

1447

15.21

15.18

15-53

15-69

512

26.99

28.19

28.14

28.77

29.06

1024

35-83

37-25

37.18

37-88

38.39

2048

46.85

48.49

48.41

49-15

49-54

Table LIV.

— Dissociation Consta

15°

nts X 10*

from

0°

graph

equation

25°

35°

1.84

2.03

2.13

2.18

1.87

2. II

2 . II

2 .21

2.27

128

1. 91

2.13

2. 12

2.23

2.28

512

1-95

2.16

2.15

2.27

2-33

1024

1-95

2.16

2.15

2.25

2-34

2048

2.02

2.25

2.22

2.32

2.38

p-Nitrobenzoic Acid, OjNCeH.COOH

The acid was purified by dissolving in alcohol and precipi-
tating by adding a large quantity of water and stirring rapidly.
The melting point was found to be 240°.

The constant for /j-nitrobenzoic acid (^25° = 4.14) is much
greater, in fact, nearly seven times greater, than that for
benzoic acid (^25° = 0.686).

0°

12°

25'

35°

512

1024

2048

79.1

99-9
126.8

104.0

131. 5
165 9

128
163
205

4
4

148.3
187.4

235-4

TableLVL-

-Temperature

Coefficients

0°-12

12°

-25

25°

-35°

Good.

units

Per

cent.

Cond.
units

Per
cent.

Cond.
units

Per
cent.

512
1024
2048

2.08
2.63
3.26

2.63
2.63
2.57

1.92
2.46
3-04

1.85

1.87

1-83

1-95
2.40
3.00

I 51
1-47
I .46

90 Wightman and Jones

Table LVII. — Percentage Dissociation

12° 25°

35 •

512

1024
2048

35-59
44-93
55-47

36.55 36.86
46.20 46.73
58.30 58.73

37.12
46.86

58.87

Table LVII I.-

—Dissociation Constants X 10*

0°

12° 25°

35°

512

1024
2048

3.84
358
3-43

4. II 4.30
3-87 4-00
3.98 4.08

4.28

4-03
4. II

1,2,4-Dinitrobenzoic Acid, C6H3(N02)2COOH
The acid was purified by the ordinary method of recrystal-
lization and melted at 179°.

The introduction of two nitro groups into the ortho and meta
positions of benzoic acid seems to produce a marked effect
on its conductivity, and forms an acid which can be classed
among the strong electrolytes. Its constants will be spoken
of later in connection with those of the other strong acids.

Table LIX.-
0°

— Molecular Conductivity

15° 25°

35°

32
128
512

1024

2048

166.51
199.23
214-97
218.60
220.00

212. 12
262.30
288.23
293.40
297.30

238
299
334
343
347

54
83
50
55
91

260 . 00

336.35
379.00
391.02
396.83

Table LX.-

0°-l5°

-Temperature Coefficie

15°-25°

nts

-35°

Cond. Per
units cent.

Cond.
units

Per
cent.

Cond.
units

Per
cent.

32
128
512

1024

2048

3.04 1.83
4.21 2. II

4.88 2.27
5-05 2.31
5-15 2.34

2.64

3-75
463
4.87
5.06

1.25

1-43
1. 61
1.66
1.70

2.15
3.65

4-55
4.80
4.89

0.90
I .22
1.36
1.40
I. 41

/'oo
0°

Values (found)

15° 25°

35°

048

220.00

297.30

347-91

396.83

fi^ Values {calculated from sodium

0° 15° 25°

salt)
35°

048

220.20

297.31

347-97

396 . 90

Conductivity and Dissociation of Organic Acids

Tabic LXI. — Percentage Dissociation

0° 15° 25°

35°

75.68 71-34 68.39

65-49

128

90.55 88.21 85.96

84.74

512

97 -70 96 . 93 95 - 90

95-47

1024

99.35 98.67 98.50

98.49

2048

100.00 100.00 100.00
1,3,5-Dinitrobenzoic Acid, CgHgCNO-OoCOOH

100.00

The acid was precipitated from alcoholic solution by means
of water. It melted at 205°.

Although the conductivity and dissociation of this acid
are fairly high, there is a wide difference between the values
for this acid and the corresponding values for the preceding
acid. The conductivity and dissociation of the 1,2,4-acid
reach a maximum at N/2048, whereas the 1,3,5-acid is only
78.7 per cent, dissociated at the same dilution.

It should be noted that the n^ values of the two acids,
however, are practically identical. Moreover, in the case
of the 1,2,4-acid, the maximum value of conductivity as meas-
ured directly agrees with that obtained from the sodium
salt. The fact that the values of the sodium salt prepared
by titration agree with those of the salt made up by weight
has already been mentioned.

Table LXII.~
0°

—Molecular

15°

Conductivity

25°

35°

512

1024

2048

122.28
147.86
167.63

171-4
205-4
231.9

203.6
244.0

273-5

233-5
280.0
3130

Table LXIIL-

0°-15°

—Temperature Coefficients

I5°-25°

25°

-35°

Cond. Per
units cent.

Cond.
units

Per Cond.
cent. units

Per
cent.

512

1024
2048

3.23 2.64
3.84 2.60
4.23 2.52

3.22

3.87
4.16

1.88 2
1.88 3
1-79 3

99
59
95

1-47
1-47
1-44

512

1024

2048

Table LXIV.-

0°
55-52
67.14
76.12

—Percentage

12°

57.62

69.06

77.96

Dissociation

25°

58.60
70.23
78.72

35°
58.83
70.54
78.86

92 Wightman and Jones

Table LXV.-

-Di

issociation Constants X 10*

0°

15°

25°

350

512

13-5

15 3

16.2

16.4

1024

13-4

151

16.2

16.5

2048

II. 9

13-5

14.2

14.4

0-Chlorhenzoic Acid, QH.Cl.COOH
The acid was purified by the precipitation of the alcoholic
solution with water, and was dried in an air bath at 105°.
The slightest friction causes the dry acid to become highly
charged electrically, so that special precautions must be taken
not to agitate it, any more than necessary, before weighing, and
it was found best to use a closed bottle for this purpose.

The values found by Ostwald^ for 25° are given below
because they differ quite appreciably from our own at the
higher dilutions :

V /cu 25° «25° V iJv 25° « 25 °

128 119. 4 33.5 512 197.0 55.3

256 156. I 43.8 1024 238.7 67.1

K = 13.2

After noticing this disagreement, a second determination
was made with a N/512 solution of the acid and a N/2048
solution of the salt at 25°, in order to see whether there had
not been some error in making up the first solutions. The
following is the result of the second measurement as compared
with the first ;

First determination Second determination

25° 25°

Acid 512 19405 193-82

Sodium salt. . 2048 75-47 75- 56

The difference in the two determinations of conductivity,
both for the acid and the sodium salt, is less than two-tenths
of one per cent.

The value of K as found by ourselves {K = 13.6) is fairly
close to that found by Ostwald.

As compared with the nitro substitution product (K = 4.14)
it is seen from the tables that the chlorine substitution prod-
uct {K = 13.6) has much the larger values both of conductivity

1 Z. physik. Chem., 3, 255 (1889).

Conductivity and Dissociation of Organic Acids

and dissociation, and the latter has a constant, at 25°, over
four times as large as the former.

o-Chlorbenzoic Acid
Table LXVI. — Molecular Conductivity

0°

15°

25°

35°

123

85.20

107.08

118. 91

128.39

256

109.00

138.40

154.12

167. 12

512

134-81

172.70

194 05

211 .86

1024

158.72

205.64

232.91

256.43

2048

178.00

233.29

266.52

296.94

Table LXVIL-

—Temperature

Coefficients

0°-15°

15°-25°

'-35 °

Cond. Per

Cond. Per Cond.

Per

units cent.

units cent. units

cent.

128

I .46 I

I. 18 I

II 0

0.80

256

I .96 I

.80

1-57 I

14 I

0 . 84

512

2.52 I

2.13 I

24 I

0.92

1024

313 I

•97

2.73 I

33 2

I. 01

2048

3.68 2

.07

3-32 I

42 3

I. 14

Table LXVIII

— Percentage L

lissociatio}

0°

15°

25°

35°

128

38.66

35-94

34-12

3232

256

49-45

46-45

44.22

42 . 06

512

61.16

57 96

55-67

53 • o^

1024

72.00

69.01

66.82

64 -54

2048

80.76

78.29

76 46

74-74

Table LXIX. — Dissociation Coy

istants X ^

ro*

0°

15°

25°

35°

128

19.0

15.8

13.8

12 . 1

256

18.9

15-7

13-7

II. 9

512

18.8

15-6

13-7

II. 9

1024

18. 1

15.0

13. I

115

2048

16.6

13-8

12 . I

10.8

1,2,4-Dthydroxybenzoic Acid, C6H3(OH)XOOH
The acid was precipitated from alcohol by water.
composed at 213°.

Table LXX. — Molecular Conductivity

V 0° 15° 25°

128 44-74 65.11 79.27

512 80.73 116.40 140.15

1024 103.30 147-77 177-20

2048 127.65 180.58 215.81

It de-

92.14
162 .02
203.58
248.28

94 Wightman and Jones

Table LXXL-

—Temperature Coefficients

0°-15»

15°.

-25°

25°

-35°

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

128

1.36

2.04

1.42

2.17

1.28

I .62

512

2 . 38

2.95

2. 38

2.04

2.19

1.56

1024

2.97

2.87

2.94

1.99

2.64

I .46

2048

3-53

2.77

352

1.92

3-25

1.50

Table LXXII

. — Percentage Dissociation

0°

15°

25°

35°

128

20. 16

21.71

22.70

23 • 10

512

36.37

38.82

40.12

40.62

1024

46.54

49.28

50.73

51.04

2048

57-51

60.22

61.79

62.24

Table LXXIIL—

■Dissociation Constants X 10*

0°

15°

25°

35°

128

3.98

4.. 62

5-21

5.42

512

4.06

4.81

5-25

5-43

1024

3.94

4.66

5 08

5.12

2048

3.80

4-45

4.88

5.01

1 ,2,yDihydroxybenzoic Acid, C6H3(OH)2COOH

The acid was purified like the 1,2,4 compound. It melted at
200°.

It was found to decompose rapidly in the presence of the
platinum electrodes, giving a yellow solution; therefore,
measurements had to be made as quickly as possible, and a
fresh solution had to be introduced into the cells after each
reading. This decomposition accounts, no doubt, for the
disagreement with Ostwald's values.

Like the two dinitrobenzoic acids given above, these twa
acids, 1,2,4- and 1,2,5-dihydroxybenzoic acids, have fi^
values which agree, though the conductivities and dissocia-
tions of the two are quite different. The constants of the
1,2,5-acid (^^25° =" 12.8) are more than twice as great as those
of the 1,2,4-acid (^25° = 5.11). Ostwald^ has given a very
complete discussion of the relation of the various hydroxy-
benzoic acids to each other, so that reference only need be
made to them here.

J Z. physik. Chem., 3, 247-51 (1889).

Conductivity and Dissociation of Organic Acids
Table LXXIV . — Molecular Conductivity

0°

15°

25°

35°

128
512

1024

2048

66.18
114.49
141.50
184.36

95 50
163.00
200 . 68
252.38

113.96
191.90
234.70
290.83

131.22

219-43
267.72
328.42

Table LXXV.-

—Temperature Coefficients

0°-15°

15°-25'

°-35°

Cond. Per
units cent.

Cond.
units

Per Cond.
cent, units

Per
cent.

128
512

1024

2048

1-95 2

3 23 2
3-95 2

4-54 2

•94
■83
■79
.46

1.85
2.89

3 40
3.86

1-93 I
1-77 2
1.70 3
1-52 3

73
75
30
76

1-52

1-43
1. 41
1.29

Table LXXVI.

— Percentage

Dissociation

0°

15°

25°

35°

128

512

1024

2048

2985
5163
63.82

83 15

3187
54-39
66.97

84.22

32.49
54-71
66.91
82.92

32.83
54 89
66.98
82.17

Table LXXVIL-

-Dissociation

Constants X ro

0°

15°

25°

35°

128
512

1024

2048

9-9
10.8
10.6
23 -9

II. 7
12.7

13-3
22.0

12.2
12.9
13.2
19.7

12.5
13.0

13-3
18.5

y-Sulphaniidobenzoic Acid, HoNOjS.CeH^.COOH

The acid was purified by crystallization. When dry it
decomposes at 280°.

We notice here the difference between the amino acids,
such as metanilic, sulphanilic, and 0- and /^-aminobenzoic
acids^ and the acid amides containing another acid radical.
Whereas the temperature coefficients increase with rise in
temperature and are very large in the former case, in the latter
the temperature coefficients are perfectly normal. The amino
group neutralizes the sulpho group but has no effect, prac-
tically, on the carboxyl group.

' This Journal, 44, 189 (1910).

96 Wighiman and Jones

The strength of />-sulphamidobenzoic acid as compared
Vv'ith that of the amino acids is seen in the following list of
constants :

0-Aminobenzoic acid

K = 0.0671

/>-Aminobenzoic acid

K = 0.0714

Metanilic acid

K = I . 99

/j-Sulphamidobenzoic acid

K = 2.96

Sulphanilic acid

^ = 6.55

Table LX XV 1 1 1. —Molecular Conductivity

0°

15°

25°

35°

512

67.79

96.00

113.02

128.03

1024

90.60

124.82

146 . 94

167.17

2048

114-55

157-37

185. II

210.05

Table LXXIX

. — Temperature Coefficients

0°-15°

25°-25

25 =

'-35°

Cond. Per

Cond.

Per Cond.

Per

units cent.

units

cent. units

cent.

512

1.88 2.77

1.70

1-77 I 50

I. 17

1024

2.28 2.52

2.21

1.77 2.02

I .21

2048

2.85 2.49

2-77
/.too Values

1.76 2.49

I. 19

V 0°

15°

25° 35°

048 221.70 :

299 29 349-77 39- -o

Table LXXX

. — Percentage

Dissociation

0°

15°

25°

35°

512

30.59

32.06

32.31

32.17

1024

40.88

41.69

42.01

2048

51.68

52.56

52.92

52.79

Table LXXXL-

-Dissociation Constants X 10*

0°

15°

25°

35°

512

2.63

2.96

3.01

2.98

1024

2.76

2.91

2.97

2048

2.70

2.84

2.91

2.88

Benzenesulphonic Acid, CgHg.SOgH
The acid was recrystallized from water, but not dried, since

the acid is very hygroscopic. It was standardized in the

same way as trichloracetic acid.

It should be noted that the temperature coefficients of strong

acids like benzenesulphonic, trichloracetic acids, etc., are

much larger than those of the weaker acids, and the increase

Conductivity and Dissociation of Organic Acids

in the

temperature coefficients with

dilution, as well

as the

decrease with temperature

is very much less rapid.

Table LXXXII

— Molecular Conductivity

0°

15°

25°

35°

204.57

27538

321.07

366.1

210.23

281.69

326.55

370.1

128

222. 14

300.43

350.47

399 8

512

226.92

305 81

356.38

407.0

1024

228.00

308 . 97

359 03

410.3

2048

226.83

305 71

354-22

407.1

Table LXXXIII

— Temperature Coefficients

0°-15°

15°-

25° 25°-

-35°

Cond. Per

Cond.

Per Cond.

Per

units cent.

units

cent. units

cent.

4.72 2.31

4-57

1.66 4.50

I .40

4.76 2.27

4-49

1-59 4-35

1-33

128

5-22 2.35

5 00

I 67 4.93

I. 41

512

5.26 2.32

505

I . 65 5 . 06

1.42

1024

540 2.37

5.01

1.62 5.13

1-43

2048

5.26 2.32

485
oQ Vahies

1-59 5 29

I 49

V 0°

15°

25° 35°

2048 228.00

308 . 97

359.03 410.3

Table LXXXIV

— Percentage Dissociation

0°

15°

25°

35°

89.72

89- 13

89 -43

89.23

92.21

91.17

90.95

90.20

128

97-43

97.24

97.62

97-44

512

99 53

98.98

99.26

99.20

1024

100.00

p-Toluenesulphonic Acid, H3C.CeH4.SO3H
This acid is similar to benzenesulphonic acid. It has a
somewhat lower conductivity, although its percentage dissocia-
tion is a little greater (the increase to a maximum, in the
case of benzenesulphonic acid, is at a slower rate).

Table LXXXV. — Molecular Conductivity

0°

12°

25°

35°

203.0

258.5

317.3

361-4

128

208.4

267.0

328.2

374-2

512

210.0

269.0

331.7

376.4

1024

210.6

269.7

332 . 7

379-3

2048

206.7

266.4

327-7

372.3

Wightman and Jones

Table LXXXVI. — Temperature Coefficients

0°

-12"*

'-25 »

25'

-35°

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

4.62

2.28

452

1-75

4.41

I 39

128

4.88

2-34

4.76

1.76

4.60

1.40

512

4.92

2.35

4.82

1.79

4.66

1. 41

1024

4-93

2-34

4.84

1.80

4.67

1.40

2048

4-97

2.40

4.76

1.79

4.46

1.36

j«oo Values

0°

12°

25°

35°

2048 210.6

269.7

332.7

379-3

Table LXXXV II. —Percentage Dissociation

12°

35°

95 85

95-30

128

99.00

98.68

512

99-74

99.26

1024

100

100.00

100

100.00

m-Nitrobenzenesulphonic Acid, O2N.CgH4.SO3H
This acid is similar to the two preceding acids, except that
it is not quite so hygroscopic and could be more easily handled.
Its conductivity is a little less than that of benzenesulphonic
acid, but its dissociation is greater.

Table LXXXV III. — Molecular Conductivity

0°

16°

25°

35°

195-9

262.9

307.1

350.0

128

200.5

269. I

313-8

357-2

512

202.0

272.9

320.4

367.2

1024

204.3

275-5

323-5

369 4

2048

204.3

274.6

321.5

368.3

Table LXXXIX. — Temperature Coefficients

0°-16°

16°-25° 25

-35°

Cond. Per

Cond.

Per. Cond.

Per

units cent.

units

cent. units

cent.

4.47 2.28

4.42

1.68 4.29

I .40

128

4.57 2.28

4-47

1.66 4.34

1-38

512

4-73 2.34

4-75

1.74 4.68

1.46

1024

4-73 2.31

4.20

1-74 4-59

1.42

2048

4-69 2.30

4.69
//oo Values

I. 71 4-68

1-45

V 0'

16°

25° 35°

1024 204.5

275-5

223.5 369-4

Conductivity and Dissociation of Organic Acids

Table XC-
0°

—Percentage Dissociation

16° 25°

35°

32
128
512

1024

95-60

97-84

98.58

100.00

95.43 94.92

97.68 97.00

99.09 99-04

100.00 100.00

94-75

96.69

99.40

100.00

1 ,2,4-Nitrotolu£nesulphonic Acid, CeH3.CH3.NO2.SO3H
The acid is very much like the preceding. It is not quite
as strong as nitrobenzenesulphonic acid, as would be expected.

Table XCL-

—Molecular Conductivity

0°

15°

25°

35°

176.9

240.9

275.6

312.6

193.0

264.1

303-6

344-2

128

198.4

272.0

312.4

354-7

512

199.9

274-3

315-6

358.6

1024

200.5

276.5

318.4

361.9

2048

199.7

274-5

314-8

357.6

Table XCIL-

—Temperature Coefficients

0°-15°

15°-2i

25'»-35°

Cond. Per

Cond.

Per Cond.

Per

units. cent.

units

cent. units

cent.

4 . 00 2 . 26

3-86

I . 60 3 . 70

1-34

4-44 2.30

4-31

1.63 4.06

1-34

128

4.60 2.32

4 49

1.65 4.23

1-35

512

4-65 2.33

4 59

1.67 430

1-36

1024

4-74 2.37

4.67

1-69 4-35

1-37

2048

4-67 2.34

4-48
//oo Values

1.63 4.28

1-36

V 0°

15°

25° 35°

1024 200.5

276.5

318.4 361.9

Table XCIII.

— Percentage

Dissociation

0°

15°

25°

35°

88.22

87.12

86.94

86.38

t^: 32

96.27

95-52

95-35

95-13

128

98.97

98.37

98.13

98.02

512

99.62

99.22

99.10

99.08

1024

100.00

4,3-Dichlorphthalic Acid, C6H2Cl2(COOH)2
The acid was purified by crystallization from water.
The conductivity of this acid as compared with that of

loo Wightman and Jones

o-phthalic acid^ is about 1.5 times as great. This acid evi-
dently dissociates as a dibasic acid, that is, both hydrogens
are dissociated, since its conductivity is too great for that
of a monobasic acid. One can see from this that it is not
possible to find the n^ values of the acid by the graphic
method, because this method is applicable only to acids dis-
sociating as monobasic acids.

Nor was it possible to obtain fi^ for the sodium salt with
the cells we were using. Further work will be carried
out later with cells of the cylindrical type on this and the
two following acids in order to obtain their a^ values.

Table XCIV.

— Molecular

Conductivity

0°

15°

35°

128

194.24

353 00

286,

,82

318.34

512

238.55

314.66

359

398.42

1024

263.80

348.53

397-

440.78

2048

288.09

378.49

436 03

48747

Table XCV.-

-Temperature Coefficients

0°-15°

15°-;

25°

°-35°

Cond. Per

Cond.

Per

Cond.

Per

units cent.

units

cent.

units

cent.

128

3.91 2.01

1-38

1-33

I. 10

512

5.07 2.13

4.44

I. 41

I. 10

1024

565 2.14

4.86

1-39

I . 10

2048 .

6.03 2.09

5-75

1-52

I. 18

Tetrachlorphthalic Acid, CeCl4(COOH)2
This acid was an exceptionally pure one obtained from

Professor W. R. Omdorff and T. G. Delbridge, of Cornell,

who had carried out an investigation on its composition and

method of purification. '^

Tetrachlorphthalic acid is quite similar, in general, to

dichlorphthalic acid, though its conductivity is much greater.

Table XCVL — Molecular Conductivity

0°

15°

25°

35°

512

296.8

386.5

441 -3

492.7

1024

328.6

432.7

495-9

555.2

2048

356.0

469.2

539-8

605.0

> This Journal, 44, 187 (1910).
^ Ibid... 41, 393 (1909).

Conductivity and Dissociation of Organic Acids
Table XCVII. — Temperature Coefficients

Cond.
units

Per
cent.

Cond.
units

Per
cent.

Cond.
units

Per
cent.

512
1024
2048

5-98
6.94

7-55

2 .02
2. II
2 . 12

548
6.32
7.06

1.42
I .46
1.50

5-14

5-93
6.52

I. 17
1.20
I. 21

Meconic Acid, (HO)C5H02(COOH)2 + 3H2O
The acid was purified by precipitation from alcohol with

water.

Table XCVIII. — Molecular Conductivity

32
128
512

1024
2048

0°

347-8
412.8

435-9
442.1

Table XCIX.-

0°-15°

15° 25
358.6 412
463 2 536

553-6 645
586.8 686
597-3 700

—Temperature Coefjic

15°-25°

8
4
4

ier

25'

35°
146.4
598.9
729-5
778.0
802.7

-35°

32
128
512

1024

2048

Cond. Per
units cent.

7.69 2.21

9-39 2.27

10.06 2.36
10.35 2.34

Cond. Per
units cent.

5-42 1.58
7-32 1.44
9.18 1.65

9 94 1-69
10.28 1.72

Cond.
units

4.86
6.25
8.41
9.18
10.26

Per
cent.

I. 18

I. 17

1-33
1-34
1-47

It is the strongest of all the acids with which we worked.
It has the highest percentage temperature coefficients and a
conductivity approaching closely to that of sulphuric acid,
although it does not give a maximum conductivity.

Sulphuric

Acid

Meconic Acid

Jones and West'

From Table XCVIII

Iiv25°

Cond. units

/iT,25°

Cond. units

491.4

4.22

412.8

4.86

128

589-4

S.68

536.4

6.25

512

675-2

7.78

645-4

8.41

1024

686.2

9.18

2048

709.9

10.45

700. 1

10.26

This Journal, 34, 414 (1905)

I02 Wightman and Jones

Camphoric Acid, CioHigO^

The acid was purified like meconic acid.

It is stronger than the acids of the aliphatic series, but is
much weaker than those of the aromatic series. It dissociates,
as can be seen from Table CII, like a weak monobasic acid,
though it titrates as a dibasic acid.

Table C. — Molecular Conductivity

0° 12°

25°

35°

512

24.94 33.05 .

38.17

42 -57

1024

34.05 4527

52.12

57-99

2048

45.10 59.54

68.15

76.12

Table CI. — Temperature Coefficients

o°-i2° 12°-

-25° 25°

'-35°

Cond. Per Cond.

Per Cond.

Per

units cent. units

cent. units

cent.

512

0.541 2.17 0.512

1.55 0.440

I 15

1024

0.748 2.20 0.685

1-57 0.587

113

2048

0.963 2.14 0.861

1.46 0.797

I. 17

/foo Values {from curve)

0° 12° 25'

35°

218.3 279.8 344

•5 392.5

K,.==

[218.3 + (5-2 X 12) — (0.00609 X 144)] = 279.8

Table CII. — Percentage 1

Dissociation

0° 15°

25°

35°

512

11.43 ii-Si

11.08

10.85

1024

15.60 16.18

15-13

14.78

2048

20.66 21.28

19.78

19.40

Table CIII. — Dissociation Constants X 10*

0° 15°

25°

35°

512

0.288 0.309

0.270

0.259

1024

0.282 0.305

0.264

0.250

2048

0.263 0.289

0.238

0.228

Uric Acid, CsH.N^Og

The acid was crystallized from water, a small quantity at
a time.

Its slight solubility made it impossible to determine its
conductivity at any dilution lower than N/2048. It is seen

Conductivity and Dissociation of Organic Acids 103

to be a very weak acid, its constant at 25° (K = 0.015 1) being
only about one-fifth that of />-aminobenzoic acid (K = 0.0714).

Table CIV. — Molecular Conductivity

V 0° 15° 25° 35°

2048 8.34 14 85 18.92 22.77

Table CV. — Temperature Coefficients

Cond. Per Cond. Per Cond. Per

V xmits cent. units cent. tinits cent.

2048 0.434 5.20 0.404 2.72 0.385 2.03

j"oo Values (determined graphically)

V 0° 15° 25° 35°

2048 221.0 298.8 350.0 399 9

i^i6 = [221.0 + (5.28 X 15) — (0.00486 X 225)] = 299.1

Table CVI. — Percentage Dissociation

V 0° 15° 25° 35°
2048 3.77 4.97 5.41 5.71

Table CVII. — Dissociation Constants X 10*

V 0° 15° 25° 35°
2048 0.0072 0.0127 0.0151 0.0169

Cyanuric Acid, ^o/JJ^^^NnH + 2H2O

The acid was purified by recrystallization from water.
Its conductivity is so small that it required resistances of
over 10,000 ohms to be thrown into the circuit in order to
obtain the minimum on the bridge at 35° and measurements
at temperatures below this were so inaccurate that they are
not given.

Constants could not be obtained with the use of the Ostwald
dilution law.

Table CVII I. — Molecular Conductivity

V 35° V 35°
128 1.46 1024 3.52
512 2.78 2048 4.67

Table CIX. — Percentage Dissociation

V 35° V 35°
128 0.360 1024 0.869
512 0.686 2048 I. 15

I04 Wightvian and Jones

The following acids were tested but were found to be either
too insoluble or they decomposed too rapidly :

Thioacetic (decomposed)
Tribromacetic (decomposed)
Phenylbromacetic (decomposed)
Sebacic (insoluble)
/)-Brombenzoic (insoluble)
/?-Chlorbenzoic (insoluble)
Isophthalic (insoluble)
Brompalmitic (insoluble)
Benzenesulphinic (insoluble)

a-Naphthylaminesulphonic acid and gluconic acid were both
tested, but it was found that the solutions had not been
standardized properly so they were, for the time being, discarded.

DISCUSSION OF RKSUIvTS

The temperature coefficients of conductivity, expressed in
conductivity units, for all the above acids show an increase
with dilution of the solution and a decrease with rise in tem-
perature. The rate of increase and decrease is, of course,
larger for some acids than for others. The general rule seems
to be that the stronger the acid, the less rapid is both the
increase with dilution and the decrease with rise in temperature.
The very fact that the dissociation of strong acids is so large
for all dilutions readily accounts for the small change in the
coefficients with dilution.

We know that the temperature coefiicients of conductivity
for even strong mineral acids decrease, to a small extent, with
rise in temperature;^ and also that most of the strong mineral
acids are more or less hydrated when in aqueous solution.
It is evident that if an acid were not hydrated or only very
slightly hydrated (as is the case with most of the weak organic
acids) , then the decrease in temperature coefficients with rise in
temperature would be more rapid, since the complex, hydrated
ions could not lose much water, and in consequence, the
conductivity could not thus be greatly increased. In order
to make this clearer, we have collected in Table CX the values
of the temperature coefficients of conductivity for a number

1 This Journal, 34, 418 (1905).

Conductivity and Dissociation of Organic Acids 105

Decrease

0°-15°

15°-25°

250-35°

Per cent.

0.43

0.40

0.39

II. 6

0.56

0.51

0.45

19.6

0.61

0.59

0.51

16.4

0.72

0.69

0.62

139

0.81

0.79

0.71

12.4

0.75

0.69

0.59

21.3

of weak and practically nonhydrated acids, expressed in con-
ductivity units, for the dilution N/1024 between 0° and

35°.

Table ex.

Acid. Kqo

Uric 0.0072

Caprylic o. 125

Propionic* o. 133

Acetic* o. 175

Levulinic o . 220

Camphoric 0.285

In Table CXI are found the similar values for some of the
hydrated acids. There are, in addition to the four strong
acids in this table, also two weaker acids — racemic and citric
acids — and these show the same decrease in the coefficients
as the strong acids.

Table CXI

Water of
cryst.
Acid H2O

Meconic 3

Benzenesulphonic i

/)-Toluenesulphonic 4

; ,2 ,4-Nitrotoluenesulphonic 2 . 5

0°-15° 15°-25° 2S''-35'

Racemic,* K.

Citric,* Kq = 6.85 I

0.06

9-94

5.26

5 -05

4.92

4.84

4-74

4.67

376

356

3.64

r J 1

3-59

! J 1

Decrease
Per cent.

8.8
3-8

5-3
8.2
8.0
71

This indicates that the strength of the acid has very little
to do with the gradual decrease in temperature coefficients,
but that this decrease is caused by hydration; and the above
relation applies to the strong acids only because they are
the acids which are most hydrated.

Another interesting point that should be brought out in
connection with the temperature coefficients expressed in
conductivity units is that the strong acids have the much
larger coefficients. This is shown by the following table,
which contains the conductivity units of the N/2048 solutions
of a number of acids between 0° and 15 °, the acids being arranged
according to their strength :

1 White and Jones: This Journal, 44, 159 (1910).

io6 Wightman and Jones

Table CXII

Cond. units

0°-15° K

Uric O

Isovaleric o

Levulinic i

/?-Iodpropionic i

/)-Sulphamidobenzoic 2

1,2,4-Dihydroxybenzoic. . . 3

a-Brompropionic 3

a-Brombutyric 3

Cyanacetic 4

/j-Toluenesulphonic 4

Meconic 10

434 0.0072

699 0.125

13 0.220

76 0.977

95 2 . 70

53 3-95

70 12.4

82 14.98

70 38 . 7

97 (completely dissociated)

White and Jones found that the percentage temperature
coefficients for very nearly all the acids with which they
worked showed a slight decrease with increasing dilution,
and a much more marked decrease with rise in temperature.
We find that the latter decrease is shown by all the acids
with which we worked, but there are a number of exceptions
to the decrease in temperature coefficients with dilution, prob-
ably due to hydrolysis.

All the percentage temperature coefficients of all the acids
except uric acid are small, and of the same order of magnitude —
not greater than 2.8 per cent, and not less than 2.0 per cent.
from o°-i5° — and decrease to i. 7-1.0 per cent, from 25 "-35°.
That this decrease^ is regular and of the same order of magni-
tude throughout would seem to indicate that some constant
factor, such as the viscosity of the medium, plays the most
important role here, and that the viscosity decreases with
increasing dilution and rising temperature at the same diminish-
ing rate. That the latter is true is seen from the following data,
showing the viscosity of thiodiglycolic acid in aqueous solution :

Table CXIII

Viscosity Viscosity Viscosity

at 15° at 25° at 35°

Water^ 0.01134 0.00891 0.00720

N/2048 solution ... 0.00896

N/ 8 solution 0.01188 0.00927 0.00750

1 For a full discussion of the decrease in percentage temperature coeflScients see
Jones and West: This Journal. 34, 418 (1905).

2 Thorpe and Rogers: Phil. Trans., 186, 307 (1894); 189, 71 (1897).

Conductivity and Dissociation of Organic Acids 107

That the conductivity of organic acids is a parabolic function
of the temperature was brought out by Schaller and other
earlier workers, and by the work of White and Jones, as we
have already had occasion to mention in the introduction
to this paper. White and Jones plotted several curves, using
conductivities as ordinates and temperatures as abscissae,
and calculated the conductivity of a number of acids by means
of the formula

Ht = Ho + at — bt^
and found this law to hold in every case.

The curves in Figs. II-V and the values given in Table
CXIV are also all in perfect accord with the law, which seems
to hold whether the acid is strong or weak, monobasic, dibasic or
tribasic. The observed and calculated values in the last
two columns of the table are seen to agree very closely.

Trichloracetic 221

a-Brompropionic. . . 124

/3-Iodpropionic 23

Levulinic 11

Table CXIV

a b t Obs.

7 5.24 0.0127 i5°297.6
7 2.84 0.0141 I5°i64.i
79 0.526 0.0004015° 31.86
57 0.318 0.00070 15° 16.13
7 2.81 0.0140 I5°i73.2
500.6840.0022915° 33.04
85 0.224 0.0012215° 12.92
31 2.37 0.00606 15° 127.47
82 0.927 0.0016215° 45.13
3 3.43 0.00750 15° 171. 4

Calc.

297-5
164. I

31 59
16.18
172.7

33 24
12.94

127.50
45 36

172 .0

ct- Brombutyric 133

Hydroxy isobutyric 23

Isovaleric 9

Thiodiglycolic 93

Tricarballylic 31

1,3,5-Dinitrobenzoic 122
1,2,5-Dihydroxy-

benzoic 114

/j-Sulphamidobenzoic 90

Benzenesulphonic . . 222

/>-Toluenesulphonic 208
1 ,2 ,4-Nitrotoluene-

sulphonic 198

Dichlorphthalic . ... 194

The dissociation of a large number of the acids decreases
steadily with rise in temperature. In other cases the per-
centage dissociation apparently reaches a maximum at some
temperature between 15° and 35°, usually around 25°. There
are still others whose percentage dissociation increases all

5 3.35 0.00983 15° 163.0 162.5

602.41 0.00660 15° 124.8 125.2

1 5.27 0.00563 i5°3oo. 4 300.0
4 4.93 0.00549 12° 267.0 266.8

4 4.79 0.00945 16° 272 .0 272.6

2 4.10 0.01576 i5°253.o 252.2

io8

Wightman and Jones

iflttitionpuoj

Xiiatjonpuoj

Conductivity and Dissociation of Organic Acids 109

i-3

3i3

■r:M>

III

2 o

111

Wightman and Jones

the way up to 35°, though at a diminishing rate. The latter
phenomenon is not unusual, Euler,^ Schaller,^ and White and
Jones, ^ all having observed the same increase in dissociation
for more than a dozen acids.

No adequate explanation, however, has been offered, and
the problem is a complex one, since there are several factors
which can influence the dissociation. Decrease in the as-
sociation and in the dielectric constant of the solvent with
rise in temperature are two well-known causes for the decrease
in the dissociating power; and rise in temperature alone,
which produces a greater instability in the molecules and a larger
ionic mobility, would tend to cause an increase in the dissociation.
The nature of the dissolved substance itself very likely plays
some part, and when all these factors are taken into account,
it seems practically impossible to predict beforehand what
the result would be.

The following table contains the dissociation constants for all
except the strong acids, calculated by means of Ostwald's law :

Table CXV.—D'i

Acid 0

Cyanacetic 38

Benzilic 9

a-Brompropionic 12

/?-Iodopropionic o

Levulinic o

a-Brombutyric 14

Hydroxyisobutyric .... o

Isovaleric o

Caprylic o

/-Tartaric 9

Thiodiglycolic 6

Tricarballylic i

/?-Nitrobenzoic 3

1,3,5-Dimtrobenzoic. . . 13

o-Chlorbenzoic 13

1,2,4-Dihydroxybenzoic 3
1,2,5-Dihydroxybenzoic 10

/>-Sulphamidobenzoic 2

Camphoric o

Uric o

1 Z. physik. Chem.. 21, 247 (1896).

■^Ibid., 26, 497 (1898).

3 This Journal, 44, 196 (1910).

issocic

ition Constants X lo*

15°

25°

35°

37-5

36.3

34-8

9.15(12°)

9.20

8.93

12.4

10.4

9.2

977

0.945

0.917

0.910

220

0.230

0.243

0.246

16. 1

12.7

II 5

1.04

1.06

1.07

167

0.153

0.149

0. 141

125

0. 126

0.129

0.123

■AS

10. 1

10.7

•38

6.50

6.51

6.36

.92

2.14

2.24

2.30

.62

3-99(12°)

4.14

•4

15.2

16.2

16.4

•7

15.5

13.6

II. 9

•95

4.64

5. II

525

•4

12.6

12.8

12.9

.70

2.90

2.96

2 94

.285

0.307(12°

) 0.267

0.254

.0072

0.0127

O.OI5I

0.0169

Conductivity and Dissociation of Organic Acids iii

A more general, though empirical, equation for calculating
the constants was suggested by Storch,^ and employed by
Bancroft,^ Noyes,^ and others, viz.,

K = C^/C^

in which C- = — denotes the volume concentration of the

I — a
dissociated portion, and C^ = — r^ — , the volume concentra-
tion of the undissociated part; K and n are both functions
of the electrolyte. By using C- as values along the ordinates
and C^ as values along the abscissae, a curve can be plotted
from which it is posible to obtain n. The values for n found
by previous workers, for a large number of salts and acids,
vary from 1.36 to 1.55.

We have attempted by means of this method to obtain
constants for the strong acids, but the equation did not seem
to be applicable to this case.

Further work on the organic acids, and also on the organic
bases, both in aqueous and in nonaqueous solutions, and
through a wider range of temperature, is being carried out in
this laboratory.

SUMMARY

Several conclusions reached by White and Jones have been
confirmed by the present work and are repeated here in quota-
tion marks.

1. The temperature coefficients of conductivity, expressed
in conductivity units, increase rapidly with dilution, and
decrease rapidly with rise in temperature for weak organic
acids— when not hydrated. When the acids are hydrated,
the temperature coefficients of conductivity are larger, and
their increase with dilution and decrease with rise in temp-
erature both take place at a slower rate.

2. Organic acids with the largest constants also have the
largest temperature coefficients of conductivity expressed
in conductivity units.

1 Z. physik. Chem., 19, 13 (1896).

^ Ibid., 31, 188 (1889).

3 J. Am. Chem. vSoc, 26, 137 (1906); 30, 333 (1908).

112 Wightman and Jones

3. "The percentage temperature coefficients of conduc-
tivity of the organic acids are generally small and of the same
order of magnitude, and decrease with rise in temperature
and with increase in dilution."

4. "The conductivity of most of the organic acids is a
parabolic function of the temperature, as proved by comparing
observed values with those calculated from interpolation
formulae."

5. "There is no general statement possible concerning
the change in dissociation of the organic acids with change
in temperature. Maxima occur with several between 25° and
35°, while in other cases maxima are indicated at slightly
higher temperatures than those at which measurements were
made. The dissociation of several acids decreases regularly
from o°."

6. The strong organic acids do not obey the Ostwald dilu-
tion law, and, therefore, dissociation constants cannot be
obtained for them in the ordinary way.

7. The migration velocities of the anions of organic acids
are a function of the number of atoms present in the anion,
and p.^ values for dibasic acids may be found by means of
this principle.

8. Most dibasic organic acids dissociate like monobasic
acids.

9. "Isomeric acids do not behave similarly as regards
change in their dissociation."

10. "The migration velocities of isomeric ions are identical."

11. "The behavior of the organic acids with respect to the
change in their dissociation with the temperature is not in
accord with the hypothesis of Thomson and Nernst, which
connects dissociating power and dielectric constants, or at
least the influence of some other unknown force is suggested."

Johns Hopkins Universttv.
May, 19 n

REVIEWS

Jahrbuch der ChemiE. Bericht iiber die wichtigsteu Fortschritte der
reinen und angewandteu Chemie. Unter Mitwirkung von O. Aschan,
Helsingfors, H. BeckurTS, Braunschweig, M. Dei.bruck, Berlin, J.
M. Eder, Wien, P. FriedlAnder, Wieu, C. Haeussermann, Lud-
wigsburg, A. Herzfeld, Berlin, K. A. Hofmann, Miinchen, G.
KEPPEivER, Hannover, E. Knecht, Manchester, J. LEWkowiTSCH,
London, A. Morgan, Hohenheim, B. Neumann, Darmstadt, M.
NiERENSTEiN, Bristol, P. Sackur, Breslau, K. Spiro, Strassburg
i. E., herausgegeben von Richard Meyer, Braunschweig. XIX
Jahrgang. 1909. Braunschweig: Druck und Verlag von Friedrich
Vieweg und Sohn. 1910. S. xii + 608.

We take pleasure in welcoming this annual guest, as we have
derived much pleasure and profit from such an examination
as is possible in the case of a Jahrbuch. The preface is a model
of brevity and may be quoted in full. It is this: "Diesmal
hat weder die Anordnung des StoiTes, noch die Liste der Mit-
arbeiter eine Anderung erfahren."

This notice might be made equally brief. Thus — the Jahrbuch
seems to be as good as usual. That would suffice. But we
are tempted to refer to a few points that have attracted at-
tention in the perusal. One whose chief interests lie in the
field of pure chemistry is more likely to turn to the reports
on applied chemistry for in these he is likely to read of things
of which his ordinary daily reading has not told him.

New to the writer is the fact recorded under acetylene that
this gas is now being used dissolved in acetone under pressures
of several atmospheres. This was proposed several years ago
but did not at first prove satisfactory. The "acetylen dissous"
comes into the market in steel flasks. The pressure employed
is 12 atmospheres.

Another interesting fact, new to the writer, but no doubt
well-known to the industrial chemist, is that the Sicilian
sulphur market is more and more seriously disturbed by
American competition. This is true of refined as well as of
crude sulphur. Last year an American refinery capable of
producing 100,000 tons a year was put up in Marseilles. This
quantity is sufficient to meet the demands of the French vine-
yards. The state of the market is such that, according to the
writer in the Jahrbuch, "die Bildung eines Schwefeltrusts
nur eine Frage der Zeit sei." The American production is
increasing slowly: 1902, 7,443 tons; 1903, 35,098 tons; 1904,
193,492 tons; 1905, 215,000 tons; 1906, 294,000 tons; 1907,
307,806 tons; 1908, 312,670 tons.

But let this suffice. Any one interested in the progress of

114 Reviews

chemistry will find the Jahrbuch full of valuable matter — and
most of it readable. I. r.

Das chemische Gleichgewicht auf Grund mechanischer Vor-
STELLUNGEN. Vou H. V. JxjPTNER, O.6. Professor an der k. k.
technischen Hochschule in Wien. Mit 60 Figuren im Text. Leipzig
und Berlin: Druck und Verlag von B. G. Teubner. 1910. S. iv + 367.
Preis, geh., M. 11; geb., M. 12.50.

After the great researches upon chemical equilibria at high
temperatures carried on some 50 years ago by Henri Sainte
Claire-Deville and his followers the attention of physical
chemists was diverted into other channels, and it is only
within the last ten years that a world-wide interest has been
manifested once more in this great problem, the solution of
which will permit a deeper insight into the laws, and perhaps
the causes, of chemical affinity. On account of the importance
of this difficult problem any work which contributes toward an
appreciation of the way in which it is being attacked, and of
the significant results which have already been obtained, must
be most heartily welcomed.

This comprehensive volume of von Jiiptner's, containing a
large number of numerical tables and other important data,
appears on hasty inspection to provide such a resume of the
subject of chemical equilibrium as has long been desired.
The reader, however, who examines the work more critically
meets only disappointment. The tables which are included
in the book have apparently been collected without system,
their source is only rarely given, and in many cases they con-
tain old and uncorrected data which have been superseded
as the result of more recent investigation. In those cases
where the tables have been recalculated, the reviewer has
found, even in a hasty inspection, numerous errors or mis-
prints. The book therefore must fail of its purpose, since it is
impossible to place reliance upon the figures given, or to trace
their source.

Many statements in the work are misleading; for example,
that "solubility is in most cases a linear function of the tem-
perature" or that the "dilution law when applied to good
electrolytes shows a poorer agreement between calculation and
observation than in the case of half-electrolytes, but neverthe-
less a satisfactory one." In fact, the dilution law predicts
the constancy of a certain quantity which in some cases actually
changes a millionfold during a hundredfold change in con-
centration. To regard this as a satisfactory agreement between
fact and theory would seem somewhat too complacent.

The author adopts in his theoretical discussion of the sub-

Reviews 115

ject the point of view of Nernst, and nearly all of the recent
data which are included in the book are those obtained in the
important investigations of the Physico-Chemical Institute in
Berlin. Thus we find in the index fifty-seven references to the
work of Nernst; but only three to Haber, one to Bronsted, and
none to Bodenstein.

The first few chapters deal with the laws of gases and with
the phenomenon of vapor pressure. Then dissociation of
solids, fusion, solution, and the transition from one solid phase
to another are successively discussed. Finally gas reactions
and the reactions of solutions are considered, and the book
ends with some technical applications of the laws of chemical
equilibrium to the metallurgy of iron. This forms, indeed,
by far the most satisfactory chapter of the book, and will
serve well to show to the technical chemist, to whom in part
the book is addressed, the numerous ways in which modern
physical chemistry may be applied in industrial problems.

Gilbert N. I/EWis.
Das ACETYI.EN, SEINE ElGENSCHAFTEN, SEINE HERSTEr.I.UNG UND

Verwendung. Unter Mitwirkung von Dr. Anton Levy-Ludwig,
Berlin; Dr.-Ing. Armin Schulze, Berlin; Ing. Ai^fred Schneider,
Chemnitz; Dr. Paul Wolff, Berlin, von Prof. Dr. J. H. Vogel,
Berlin. Mit 137 Figuren im Text. Chemische Technologie in
Einzeldarstellungen, herausgegeben von Ferd. Fischer. Leipzig:
Verlag von Otto Spamer. S. viii + 294. Preis, geh., M. 15; geb.,
M. 16.50.

One need not read further than the second line to know that
this whole series is going to be excellent. The fact that Fischer
of the Jahresbericht is editor-in-chief is alone sufficient to
warrant the purchase of the monographs by every technical
school.

The present volumes deal minutely with the physical,
chemical and hygienic properties of acetylene, with its analysis
and technical preparation. Its various uses and applications
as an illuminant, both stationary and portable, as a heating
agent for the kitchen and laboratory, for autogenous welding
and the cutting of metals and lastly as a raw material from
which to make chemical products are carefully and thoroughly
discussed. Two chapters on its use in the chemical labora-
tory and on the laws and official regulations for its handling
and use complete this thoroughly satisfactory book.

A valuable feature of the treatise for Americans is the ex-
tent to which American work and inventions have been rec-
ognized and considered — an encouraging symptom in German
writers. a. h. gill.

1 1 6 Note

NOTE

INTERNATIONAI. ASSOCIATION OF CHEMICAL SOCIETIES

At a meeting of delegates from the Chemical Society of
London, the Deutsche Chemische Gesellschaft and the
Societe Chimique de France, held at Paris April 25-26, an
International Association of Chemical Societies was organized
with the object of establishing a common bond between the
world's chemical societies and of taking up questions of
general and international interest in chemistry. Any chemical
society is eligible for membership, but only one from each
country can be represented in the Governing Council which
shall be composed of three delegates from each country. The
present Council consists of the delegates from the three found-
ing societies, and representatives of other societies can be
admitted to the Council only on a two-third's majority vote
of the members voting, votes by correspondence being per-
mitted. The Council shall elect, at the beginning of each
meeting, a President who shall take ofl&ce at the end of the
meeting and who, with the other two delegates from his
country, acting as Vice-President and General Secretar3%
respectively, shall constitute an Executive Committee or
"Bureau." The functions of the association shall be the
appointment of commissions to investigate problems submitted
by the Council, the publication of reports in the journals of
affiliated societies or in any other form which the Council
may select, and the holding of congresses. General expenses
shall be borne by the affiliated societies in proportion to the
number of their members, but other expenses shall not be
assessed against the various societies except on their individual
consent. Amendments to the by-laws can be made only on
a two- third's majority vote of the Council.

The next meeting will be held in Berlin, April 13, 1912,
under the presidency of Prof. Ostwald.

Vol. XLVI August, 191 i No. 2

AMERICAN

CHEMICALJOURNAL

DIFFERENCES OF POTENTIAL BETWEEN CADMIUM

AND ALCOHOLIC SOLUTIONS OF SOME OF ITS

SALTS

By Frederick H. Getman

In 1899 Kahlenberg^ published an account of some experi-
ments on the dififerences of potential established when metals
are immersed in nonaqueous solutions of their salts. In this
investigation he studied the behavior of silver, zinc, magne-
sium, cadmium, thallium, lead, copper, antimony, bismuth
and iron in solutions of salts of these metals in nearly thirty
organic solvents.

Previous to this work of Kahlenberg, Jones^ had shown
that the solution pressure of silver immersed in a solution of
silver nitrate in ethyl alcohol is only about one-twentieth of
the solution pressure of silver in an aqueous solution of the
same salt. He also measured the potential differences be-
tween silver and solutions of silver nitrate in methyl alcohol
and acetone. The highest electromotive force measured in
acetone was 0.182 volt, while in methyl alcohol and ethyl
alcohol the maximum values were 0.123 and 0.130 volt,
respectively. Jones concludes from these experiments that

J Kahlenberg: J. Phys. Chem., 3, 379 (1899).
2 Jones: Z. physik. Chem.. 14, 346 (1894).

ii8 Geiman

the solution pressure of the metal varies from one solvent
to another. Later, Jones and Smith* studied the system

Zn— o. I N ZnCl2 in C2H5OH— o. i N ZnCl2 in HjO— Zn

and concluded that the solution pressure of zinc in ethyl alco-
hol is io~* times its solution pressure in water. In 1893
Campetti" made some measurements of the potential differ-
ences between cadmium and nonaqueous solutions of its
salts.

In all of the investigations cited no attempt was made to
determine the change in electromotive force with the concen-
tration of the electrolyte.

It has seemed of interest to study the cadmium electrode
rather more closely than has been done hitherto and to deter-
mine, if possible, the solution pressure of this metal in several
nonaqueous solvents.

Method of Experimentation

The differences of potential were measured by means of
the familiar compensation method of Poggendorf, using as a
zero instrument a Lippmann electrometer of improved form
in which the mercury and sulphuric acid are permanently
enclosed. With this instrument it was possible to make meas-
urements correct to one millivolt except where the poor con-
ductivity of the solutions or other disturbing factors rendered
the readings less trustworthy. A chloride accumulator was
used as the source of potential against which the experimental
cell was balanced. The electromotive force of this cell was
determined before and after each series of measurements by
means of a Weston standard element, the electromotive force
of which was 1.0197 volt at 20°, this voltage being based on
the assumption that the electromotive force of a Clark cell
at 15° is 1 .434 volts.

One half of the experimental cell consisted of a normal
calomel electrode prepared according to the directions given
in Ostwald-Luther's Physiko-Chemische Messungen (p. 381)

• Jones and Smith: This Journal. 23, 397 (1900).

2 Campetti: Atti accad. Torino. 29, 61 and 228 (1893).

Differences between Cadmium, Etc. 119

and was assumed to have an electromotive force of 0.56
volt, the positive sign indicating that the electrode was posi-
tive to the solution. The cadmium electrode and containing
vessel composed the other half of the experimental cell.

The cadmium electrode was a rod of pure cadmium obtained
from Kahlbaum, a copper wire being soldered to the top of
the electrode in order to establish connection with the circuit.
The cadmium and calomel electrodes were connected through
an intermediate vessel containing some of the solution to be
studied. The experimental cell was suspended by means of
adjustable supports within a bath of water, the tempera-
ture of which was maintained at 20° with a possible variation
of ±o°.2. At this temperature the electromotive force of
the calomel electrode is 0.5612, the temperature coefficient
being 0.0006 volt per degree.

The methyl and ethyl alcohols used were obtained from
Kahlbaum and were subjected to treatment with lime for
several weeks before they were required, when they were
distilled into dry receivers which could be tightly stoppered.
Merck's cadmium iodide and Kahlbaum's anhydrous cad-
mium chloride were the solutes employed, each being kept
in a dry atmosphere for several days before making up the
solutions.

The solutions were prepared by direct weighing of the so-
lute and then making the solution up to a volume of 50 cc.
in a calibrated measuring flask. In several cases the dilute
solutions were prepared by dilution of the more concentrated
solutions.

The general arrangement of the voltaic combination was
as follows:

Cd— sol. Cd salt— sol. Cd salt— N KCl— HgCl— Hg

After each series of measurements on a solution of one con-
centration the experimental cell was taken apart and the
cadmium electrode washed and then rubbed with a fresh
piece of emery cloth to insure removal of any oxide or other
substance which might interfere with the contact between metal

I20 Getman

and solution. The electrode vessel and connecting vessel
were thoroughly washed and dried, while the capillary tube
connecting the calomel electrode with the other half of the
experimental cell was washed and fresh N KCl forced through
the tube.

The electromotive force developed at the junction of the
two solutions, being very small, was disregarded in all of the
calculations.

Results

The results of the measurements are given in Tables I to
V, in which m denotes the molar concentration of the solu-
tion and TT the electromotive force in millivolts, the prefixed
sign indicating the potential of the cadmium electrode with
reference to the solution. The data recorded in the tables
are represented graphically in Fig. I.

The potentials given are the mean of a series of observa-
tions and in several cases represent the average of two en-
tirely independent series made upon solutions of the same
concentration.

It will be observed that the potential of the cadmium elec-
trode in alcoholic solution becomes more strongly negative
as the concentration increases.

In the case of the potential differences measured in solu-
tions of cadmium iodide in methyl alcohol we are at a loss to
account for the irregularities observed, especially with solu-
tions of 0.08 and o. 10 molar concentration. When cadmium
is immersed in solutions of cadmium iodide in ethyl alcohol
of concentration less than 0.06 molar there is a reversal of
polarity of the system and the current flows outside of the
cell from cadmium to mercury instead of from mercury to
cadmium, as in methyl alcohol and aqueous solutions. The
curves C and D show the change in potential of the cadmium
electrode when immersed in aqueous solutions of cadmium
iodide and cadmium chloride, respectively.

Differences between Cadmium, Etc.

•S 80

200

o.is

0.20

0.0$ O.IO

Molecular Concentration
?ig. I. — A. Cadmium iodide in methyl alcohol. B. Cadmium iodide in ethyl alcohol.
C. Cadmium iodide in water. D. Cadmium chloride in water.

Getman
Table I — Cadmium Iodide in Methyl Alcohol

Temperature m n

20°. O O.OI — 82.6

19°. 9 0.02 — 85.6

19°. 9 0.03 — 84.2

20°. 2 0.05 — 87.3

20°. o 0.06 — 91.7

20°. o 0.08 lOI.O

20°. I o.io — 95.0

19°. 9 0.15 —III. 9

20°. I 0.20 121. O

Table II — Cadmium Iodide in Ethyl Alcohol

Temperature m r

20°. o 0.02 +18.6

19'

20^
20*

0.04
0.06
0.08

o. 10

+ 10.5

+ i.o

- 8.3
—17.8

Table III — Cadmium Chloride in Methyl Alcohol

Temperature

19°. 9

sat. sol,
sat. sol.

-113. o
-II5-6

Mean — 114 -3

Table IV-

-Cadmium Iodide in

Water

Temperature

20°. 2

O.OI

— 192.3

19°. 8

0.03

— 190.7

20°. 2

0.05

190.2

20°. 0

0.075

— 188.2

19°. 8

O.IO

— 186. I

20°. 0

0.15

184.2

20°. 2

0.20

— 180. I

Table V — Cadmium Chloride in Water

Temperature m

20°. O

19^
20'
19'
19'

O.OI

—182.0

0.075

—180.5

O.IO

—179.4

0.15

—177-4

0.20

— 176.0

Differences between Cadmium, Etc. 123

It will be observed that in aqueous solution the potential of the
cadmium electrode becomes less strongly negative as the concen-
tration increases, or in other words, its behavior is exactly
opposite to that in the alcohols. It is of interest to note
that the slope of the curve A is approximately the mean of
the slopes of the curves B and C, which is what we should
expect from other investigations on the physical properties
of methyl alcohol in comparison with the same properties of
water and ethyl alcohol.

The potentials of the cadmium electrode in aqueous solu-
tions of cadmium chloride have been studied by Labend-
zinski.^ For the combination

Cd— o. 2 N CdCla— o. 2 N CdCl2 — N KCl— HgCl— Hg

he finds an electromotive force of 0.738 volt, while for the
same combination we find o. 737 volt.

Owing to its slight solubility in both methyl and ethyl
alcohols, only two series of measurements were made on cad-
mium chloride in methyl alcohol. In a paper on the solu-
bility of some inorganic salts in organic solvents Lobry de
Bruyn^ gives the following solubilities of cadmium chloride

at 15°. 5:

100 grams absolute methyl alcohol dissolve 1.71 grams
cadmium chloride ;

100 grams absolute ethyl alcohol dissolve 1.52 grams cad-
mium chloride.

In selecting cadmium chloride for this investigation it
was assumed that these figures referred to the anhydrous
salt, but it would appear that they must have been obtained
with the hydrated salt, CdCl2.2H,0. The measurements on
cadmium chloride in methyl alcohol were confined to the
saturated solution.

Discussion of Results

The electromotive force of the experimental cell is given
by the equation

E = 7: + 0.5612

> Labendzinski: Z. Elek. Chem.. 10, 77 (1904).

2 Lobry de Bruyn: Z. physik. Chem.. 10, 783 (1892).

124 Getman

or the potential of the cadmium electrode, referred to the
solution, is

X = E — 0.5612

Applying Nemst's theory of the voltaic element we have as
an expression for the osmotic work in transferring ions from
the pressure P to the pressure p,

IP
which on integration gives

RT log, J

The corresponding electrical work is nzF where n is the
valence of the ions and F is the quantity of electricity carried
by I gram-equivalent of ions or i Faraday = 96580 coulombs.
Equating electrical and osmotic work we have

nnF = jRT log, —
P
or

RT , P , .

- = ;^log.- (I)

Since osmotic pressure is proportional to the concentration
we may substitute the ionic concentration, c, for p in equation
(i), when we have

RT , P , .

Except in the case of completely ionized electrolytes, c
must be replaced by wa where m is the molar concentration
of the solution and a is the degree of ionization. Assuming
that the solution pressure remains constant for any one sol-
vent, we may obtain from (2) an equation expressing the re-
lation between the electrode potentials for two different ionic
concentrations, c, and c,. Letting ttj and ^Tj denote the elec-

Differences between Cadmium, Etc. 125

trode potentials corresponding to ionic concentrations c^
and C2, we have

'^^ = '^^ + S^^°S^c^ ^^)

Inspection of (3) shows that if the electrode potential is
positive it will be diminished by dilution, whereas if it is nega-
tive it will increase with dilution. In aqueous solutions of
cadmium iodide and cadmium chloride the potential of the
electrode increases (becomes more strongly negative) with dilution
and consequently the electrode should be negative with refer-
ence to the solution. In alcoholic solutions the electrode po-
tential diminishes (becomes more strongly positive) with dilution
and therefore the electrode should be positive with reference
to the solution. The experiments show that the cadmium
electrode is negative in aqueous solutions, the potential differ-
ence increasing with dilution, whereas the cadmium electrode
in alcoholic solution is negative, the potential difference de-
creasing with decrease in concentration and eventually re-
versing to a positive potential in the dilute solutions. The
cadmium electrode can acquire a positive potential only
when the combined electromotive force of the system is less
than 0.5612 volt. The behavior of the cadmium electrode
in alcoholic solution might be explained by assuming that

the value of the term — = log^ - in equation (3) is positive

and increases with dilution. In the more concentrated solu-
tions the value of this term would be positive but numerically less
than 7:^, which is negative, and therefore the value of rij would be
negative also ; as the concentration diminishes, the value of

RT c

—f:i log, — would increase and eventually would become greater

nr Cj

than TTi when the value of TTa would be positive. In order that the

RT c

numerical value of — = losf. — may increase we must as-
nF ° Ci

sume C2>Ci or W2a2>^i«i-

A consequence of this assumption is that the degree of

ionization must increase with dilution more rapidly than the

dilution increases. This assumption is highly improbable.

126 Getman

In his investigation of the differences of potential between
metals and nonaqueous solutions of their salts, Kahlenberg^
found several cases of reversal of electrode potential, notably
in the combinations involving zinc. In the systems

Zn— ZnClj in Quinoline— ZnClj in Water— Zn
and

Zn— ZnClj in Ethyl Alcohol— ZnClj in Water— Zn

the zinc electrode in the nonaqueous solution was positive.

Commenting on this, Kahlenberg says: "This is very
unusual and it is to my knowledge the first case found where
zinc is positive toward the solution in which it is immersed."

And again: "I desire once more to call special attention to
the fact that while the electromotive force at the contact
of a metal and a solution of one of its salts varies with the
nature of the solvent, this variation may di^er not only as
to degree but also as to sense or direction from the potential
that the metal exhibits toward, for example, an aqueous
solution of the salt." In view of the improbable assumptions
involved in any explanation based on the osmotic theory of
the voltaic cell it seems unwise at the present time to extend
this theory to nonaqueous solutions.

Solution Pressure

The value of the solution pressure of a metal may be cal-
culated by means of equation (i),

RT , P

t: = —=, log, -

nF p

Letting i? = 8.32 X lo' absolute units, T = 293° abso-
lute, n = 2, and F = 96580 coulombs, and multiplying by
2 .3026, the logarithmic modulus, we have

, P

z = 0.029 log —
P
and

Differences between Cadmium, Etc. 127

The osmotic pressure of the ions is given by the equation

ma X 22.4 X —
273

As the complete expression for P at 20° we have

logP =

0.029

+ log 24.04 ma

(4)

The calculation of the solution pressure by means of (4)
involves the degree of ionization, a. The values of a for
aqueous solutions of cadmium iodide and chloride obtained
by the conductivity and the freezing-point methods are not
in good agreement, as the following tables (VI and VII) show :

Table VI — Cadmium Iodide in Water at 18°

m A o» o* (f. p. method)

I.O

15-4

0.13

0.5

18.3

0.15

0.2

24.2

0.20

O.I

31 0

0.26

0.05

40.1

0.33

0.2

53-9

0.45

O.OI

65.6

0.55

Table VII — Cadmium Chloride in Water at 18°

m A a^ a' (f. p. method)

1.0 22.4 ....

0.5 30.8

0.2 41.2 ....

O.I 50.0 ....

005 59.0 0.51 0.61

0.02 73.0 0.63

o.oi

83.0

0.72

With the more dilute solutions the two methods give con-
cordant results and we may venture to employ the values
of a for these concentrations in calculating the solution pres-
sure. The results are given in the following table:

1 Kohlrausch and Holborn: Leitvermogen der Elektrolyte.

2 Jones: Z. physik. Chem., 11, 544 (1893).

128

Getman

Table VIII

O.OI

5.64 X 10^ atmos.

0.05

1.68 X io« atmos.

0. 10

1 .64 X 10' atmos.

O.OI

3.45 X 10^ atmos.

0.05

1.07 X 10" atmos.

Cd in aqueous Cdlj

Cd in aqueous CdClj

Comparing these values with the value, P = 2.7 X 10'
atmospheres, given by Neumann^ for normal solutions of the
chloride, nitrate and sulphate of cadmium, the agreement is
perhaps as close as could be expected when we consider the
variation among themselves in the above values.

If the Nemst theory is applicable to voltaic combinations
in nonaqueous solvents, the value of the solution pressure
may be calculated by means of the equation

log P = — ^ + log p
^ 0.029 ^ ^ ^

provided the value of p can be determined. In order to cal-
culate p the degree of ionization of the solution must be known.
The conductivities of cadmium iodide in methyl and ethyl
alcohols have been measured by Zelinsky and Krapiwin^
and Jones and Carroll,^ their results being given in Table IX
and plotted in Fig. II, together with the values for the con-
ductivity of cadmium iodide and cadmium chloride in aqueous
solution. The values of the conductivity are very low and
increase but slightly with dilution. While a cannot be ob-
tained by the conductivity method, yet we may infer, from
the preceding table and the curves in Fig. II, that the degree
of ionization is very small.

Table IX— Cadmium Iodide at

23°

Methyl alcohol

Ethyl alcohol

/■i/

13.07

2.29

13-59

2.30

14.16

2.32

128

15.01

2-39

256

15 -44

2.66

1 Neumann: Z. physik. Chem., 14, 193 (1894).

2 Zelinsky and Krapiwin: Ibid.. 21, 35 (1896).

3 Jones and Carroll: This Journal. 28, 329 (1902).

Differences between Cadmium, Etc.

129

o .1 0.02 o.OJ Molecular Concentration 0.005

250

100 150 200

Volume
Fig. II. — A. Cadmium iodide in methyl alcohol. B. Cadmium iodide in ethyl alcohol.
C. Cadmium iodide in water. D. Cadmium chloride in -water.

I30 Getman

It having been shown by Jones* that the boiling-point
method may be employed to measure the ionization in non-
aqueous solutions, an attempt was made to determine a
for cadmium iodide in both methyl and ethyl alcohols. The
small value of the boiling-point constant for methyl alcohol
renders the results in this solvent rather unreliable, but it
was found that cadmium iodide is partially polymerized in
each of the two solvents. The molecular weight of cadmium
iodide in methyl alcohol was found to be 702 . 5 (calculated
for Cdlj, 366.24), while in ethyl alcohol it was found to be
687.8. These values are affected by the same errors which
we ordinarily associate with molecular-weight determina-
tions by the boiling-point method.

It is evident that at the present time we have no data upon
which to base a calculation of the solution pressure of cad-
tnium in the alcohols and furthermore it is doubtful, as has
been suggested above, whether the Nemst theory can be ap-
plied to nonaqueous solutions, at least without modification.
In this connection Kahlenberg^ says : " If the theory of electro-
lytic solution tension be held at all, there seems to be no es-
cape from the conclusion that the solution tension varies for
different solvents and mixtures of solvents; and hence the
dissolved substances — other than the simple ions of the metal
in question — may exert an influence in determining the differ-
ence of potential between the metal and the solution. If
the solution tension changes not only with the solvent, but
also with the dissolved substance present, it is questionable,
to say the least, whether the hypothesis of an electrolytic
solution tension is helpful at all."

It is our intention in the near future to extend the work
of which this paper gives a preliminary account.

Bryn Mawr College,
May, 1911.

» Jones: Z. physik. Chem., 31, 114 (1899).
2 Kahlenberg: Loc. cit.

CONDUCTIVITY AND VISCOSITY IN MIXED SOLVENTS
CONTAINING GLYCEROL

By J. Sam Guy and Harry C. Jones
INTRODUCTION

Jones and Schmidt,^ in a previous paper published from
this laboratory, gave a detailed historical sketch of the
work of Jones with Lindsay,- Carroll,^ Bassett,* Bingham,'
Rouiller,^ McMaster,^ Veazey,^ and Mahin,^ dealing with
the relations existing between conductivity and viscosity
of a large number of electrolytes in binary mixtures of methyl
alcohol, ethyl alcohol, acetone, and water. Schmidt worked
with binary mixtures, and introduced glycerol as one of the
solvents.

The results of these investigations have been to show that
curves representing fluidity and conductivity have, in general,*
the same form, whether they show maxima or minima as the
composition of the mixture is changed.

A fuller discussion of the results and conclusions drawn
from the first seven of these investigations has been published
as Monograph No. 80 of the Carnegie Institution of Washing-
ton (1907). In all of these publications, due credit has been
given to previous workers in this field, hence mention of
their results need be made only in so far as they bear upon
points of interest in this investigation.

The work of Jones and Veazey^** included a study of the
conductivities and fluidities of cupric chloride and potassium
sulphocyanate in mixtures of the same general composition
as those used by Jones and Bingham.** Copper chloride

» This Journal, 42, 37 (1909).

2 Ibid.. 28, 329 (1902).

3 Ibid.. 32, 521 (1904).
* Ibid.. 3i, 409 (1904).
s/6irf.,34, 481 (1905).
6/6td.. 36, 427 (1906).

7 Ibid.. 36, 325 (1906).

8 Z. physik. Chem., 61, 641 (1908).
»/6irf.,69, 389 (1909).

10 This Journal. 37, 405 (1907).
» Ibid., 34, 481 (1905).

132 Guy and Jones

gave results that were about normal, i. e., the curves repre-
senting conductivity and fluidity were very similar.

One of the most interesting points brought out in the in-
vestigation of Jones and Veazey was the fact that in certain
of the mixtures the solution of potassium sulphocyanate gave
a viscosity that was less than that of the pure solvent.

Euler* had noted that certain salts had the power to lower
the viscosity of water, and explained this fact by the aid
of the "electrostriction theory" of Drude and Nemst,^ ac-
cording to which there exists about every ion, by virtue of its
charge, a strong electrostatic field, which causes a strong
compression of the liquid in this field.

Wagner and Miihlenbein^ showed that Euler's reasoning
could not hold, since the viscosity of a liquid could be lowered
by the addition of certain nonelectrolytes whose viscosity
was even greater than that of the solvent. In a word, the
effect could not be due to any phenomenon specific to ions,
since the molecules could produce the same change.

Jones and Veazey* offer a possible explanation of this phe-
nomenon. A careful study of all the viscosity data avail-
able showed that only certain salts of potassium, rubidium,
and caesium had the power of lowering the viscosity of water.
The work of Thorpe and Rodger^ had indicated that, in all
probability, viscosity was a direct function of the skin friction
of the ultimate particles present. This being the case, it
is not surprising that some salts of the above named metals
do not produce this effect, since it is clear that viscosity is
an additive property of both the ions present. The one
might tend to decrease, the other to increase the viscosity,
and the final results would depend upon whether or not the
sum of these two opposing influences was positive or nega-
tive. These same three metals occupy the maxima on the
well-known atomic volume curve of Lothar Meyer. ^ This, of
course, means that these metals have very large atomic volumes.

' Z. physik. Cliem.. 26, 536 (1898).

2 Ibid.. 15, 79 (1894).

3 Ibid.. 48, 867 (1903).

" This Journal, 37, 405 (1907).

s Phil. Trans.. 186, A, 307 (1894).

6 Ann. Chem. (Liebig). Suppl.. 7, 354 (1870).

Conductivity and Viscosity in Mixed Solvents 133

With these facts at hand, Jones and Veazey offer the fol-
lowing simple explanation as to how any substance may
lower the viscosity of the solvent in which it is dissolved.
If the atomic volume of the added electrolyte is larger than
the molecular aggregates of the solvent, then the relative
amount of skin friction in a given volume of solution would be
decreased, and hence, according to the hypothesis of Thorpe
and Rodger,^ the viscosity, which is a direct function of the
skin friction, would be decreased.

Jones and Veazey use the same reasoning to account for an
increase in viscosity when water and alcohol are mixed. Parts
of these liquids, as shown by the method of Ramsay and
Shields,^ exist, when pure, in a highly associated condition.
Jones and Lindsay,' in measuring the conductivities in such a
mixture, had noted a minimum conductivity in a mixture
containing fifty per cent, of each solvent. In a word, at this
point the conductivity was less than that in either solvent
independently.

They offer the following explanation. Jones and Murray*
showed that when two highly associated liquids, which in
terms of the hypothesis of Dutoit and Aston^ would have
strong dissociating powers, are mixed, the one breaks down
the molecular association of the other. This decrease in
association would lessen the power of the solvent to dissociate
a given electrolyte into its ions, and thus decrease the conduc-
tivity. Jones and Murray actually found that the molecular
weights of water, formic acid and acetic acid, when mixed
in pairs, showed smaller values than in the pure homogeneous
condition. This change in the molecular aggregation would
increase the skin friction and thus increase the viscosity.

This lowering of viscosity is of importance as bearing upon
some facts established in this investigation, and these will
be discussed later.

It is well known that in a strongly dissociating solvent

1 Loc. cit.

2 Z. physik. Chem., 12, 433 (1893).

3 This Journal. 28, 329 (1902).
*Ibid., 30, 193 (1903).

5 Compt. rend.. 126, 240 (1897).

134 ^'"-y ^^'^ Jones

the conductivity of a ternary electrolyte is, in general, larger
than that of a binary one in the same solvent — since there
is a larger number of ions present. Jones and Veazey*
were able to show that potassium sulphocyanate in ethyl
alcohol gave a larger molecular conductivity than copper
chloride, while in aqueous solution the reverse was true. This,
in the opinion of Jones and Veazey, was due to the fact that
ethyl alcohol, being a relatively weak dissociating agent, had,
at moderate dilutions, the power of breaking copper chloride
down into only two ions. This fact will be referred to again
under the discussion of the results obtained in this investiga-
tion.

Cattaneo^ measured the conductivities of a few salts in
glycerol and found values much smaller than in water. Schett-
ner^ and Arrhenius* measured the viscosities of glycerol and
mixtures of this solvent with water and with other nonaqueous
solvents. By far the larger part of the work, with glycerol
as a solvent, has been done by Jones and Schmidt. The
present investigation is a continuation of their work.

Jones and Schmidt have shown that glycerol is an excellent
solvent and, in all probability, a fairly good dissociating solvent,
since it has a dielectric constant of 16.5 at 1 8°, and an association
factor of 1.8 at the same temperature. With such a dielectric
constant and association factor glycerol, according to the
Thompson^-Nemst^ and Dutoit and Aston^ hypotheses, should
have a dissociating power approximately equal to that of
ethyl alcohol. Jones and Schmidt believed that the extremely
small conductivities shown by solutions of electrolytes in
glycerol were due to the high viscosity of this solvent.

With these facts before us, an attempt was made to study
the relative ionic velocities of electrolytes in glycerol. The
apparatus used for this purpose was that devised by Jones and

1 Loc. cit.

2 Rend R. Accad. Lincei, [5] 8, II, 112 (1893).

3 Wien. Ber., 77, II 682 (1878).

* Z. physik. Chem., 1, 285 (1887).
s Phil. Mag., 36, 320 (1893).

6 Z. physik. Chem., 13, 531 (1894).

7 Loc. cit.

Conductivity and Viscosity in Mixed Solvents 135

Bassett,* and used subsequently in this laboratory. ^ A normal
solution of copper chloride in such an apparatus was subjected
to a current of 120 volts for forty-eight hours, and only a few
milligrams of silver were deposited in the voltameter. Al-
though no final data concerning the migration velocities
were obtained, yet the above experiment was sufficient to
show that the movement of the ions in solutions of glycerol
must be extremely slow as compared with the movement of
ions in water and the alcohols, etc.

Jones and Getman* had measured the amount of solvation of
glycerol in aqueous solution. This work has been repeated
and was found to contain an error, probably in the strength
of the solution.

The following table shows that the amount of solvation is
extremely slight even in the most dilute solutions.

Table A

N A A/»» Waol. W'glyc W^water

0.2 0.383 I. 91 25.1600 0.4603 24.6997

0.4 0.773 1-93 25.2150 0.9206 24.2944

0.8 1.627 2.03 25.4925 I. 8413 23.6512

1.2 2.528 2.10 25.6300 2.7619 22.8681

1.6 3.482 2.18 25.9025 3.6826 22.2199

2.0 4.451 2.22 26.0650 4.6032 21.4618

2.4 5.764 2.34 26.2450 55238 20.7212 17. II 1.86 1.64

2.8 6.986 2.46 26.4375 6.4445 19.9930 20.03 1-86 1.95

In this table A^ is the normality of the solutions, A the observed
lowering of the freezing point corrected for the separation of
ice, A/m the molecular lowering of the freezing point, W^^^
the weight of 25 cc. of solution, W^giyc. the weight of glycerol
in 25 cc. of solution, t^^vater the weight of water contained
in 25 cc. of solution, L the theoretical molecular lowering
of the freezing point referred to 1000 grams of solvent, and
L' the observed corrected lowering on the same basis. It
is seen that the observed and theoretical molecular lowefings

» This Journal. 82, 429 (1904).

2 Ibid., 30, 427 (1906).

3 Ibid.. 31, 303 (1904).

Cor.
Per cent. L

I.20 1.86
2.82 1.86

1.89
1.88

5-39 1-86

8.52 1.86

II . 12 I .86

14.15 1.86

1.92
1.92

I 94
1.90

136 Guy and Jones

are nearly the same, indicating that the substance does not
show any marked hydration in the solutions worked with.

EXPERIMENTAI,

Apparatus

In this investigation the Kohlrausch method of measuring
conductivity has been employed, the improved Kohlrausch
slide- wire bridge, resistance box, induction coil, and telephone
receiver being used. The entire apparatus was made and
carefully calibrated by Leeds, Northrup and Co., Philadelphia,
and, in addition, the standard resistance was checked- by the
United States Bureau of Standards, Washington, D. C. The
new form of bridge is a great improvement over the ordinary
Wheatstone bridge, both in convenience and accuracy. By
means of such a bridge readings may be checked, under
favorable conditions, to one-tenth of a millimeter.

The conductivity cells were of the same type as those de-
scribed by Jones and Schmidt^ and Jones and Klreider.^ Such
cells, as has been stated, have very small constants, and
hence are well adapted to measuring the conductivity of
solutions with high resistances. In every case the cell con-
stants were determined by means of a fiftieth-normal potas-
sium chloride solution, and checks made at frequent intervals
showing only slight variations in the cell constants through-
out the entire investigation. The molecular conductivity
of the fiftieth-normal potassium chloride solution was taken
as 129.7 reciprocal Siemens units at 25°.

The constant temperature baths were regulated by elec-
trically-controlled regulators, devised by Reid,^ and were
kept within o°.02 of the desired temperature. The ther-
mometers were carefully standardized by means of a certifi-
cated Reichsanstalt instrument. All flasks, burettes, and
other apparatus were carefully calibrated, by weighing, to
hold aliquot parts of the true liter at 20°.

* Loc. cil.

2 This Journal, 45, 295 (1911).

3 Ibid., 41, 148 (1909).

Conductivity and Viscosity in Mixed Solvents 137

Solutions

For the work at 25°, 35°, and 45°, solutions were made up
at 30°, while for the higher temperature work, the solutions
were made up at 50°. In all cases the mother solution was
made by direct weighing of the carefully dried, anhydrous
salt, and from this the N/50 and N/ioo solutions were made
by dilution. These solutions then served as the mother
solutions for the N/200 and N/400, from which, in turn,
the N/800 and N/1600 solutions were made. The highest
dilution was made by diluting the N/400 solution four times.

Measurements were not made at dilutions higher than
sixteen hundred, on account of the extremely high resistance
and consequent difficulty in making the readings. In pure
glycerol measurements were made at intervals of 5° from
25° to 75°, while in the mixed solvents they were made
only at 25°, 35°, and 45°.

Solvents

Glycerol. — The glycerol used was Kahlbaum's best double-
distilled product, and had a mean specific conductivity of
about 0.9 X io~' at 25°. Schmidt had showed that redis-
tillation did not essentially improve the glycerol. Its specific
gravity showed that it contained about 0.02 of a per cent, of
water. The two lots obtained from Kahlbaum showed some-
what different viscosities, as is indicated in the experimental
results.

Water. — The water was purified by the method of Jones
and Mackay,^ with the modification as mentioned by Schmidt,
and had a mean specific conductivity of 1.5 X lo-e at 25°.

Ethyl and Methyl Alcohols. — The ethyl alcohol was puri-
fied by several distillations from the very best quality of
lime, and block-tin condensers were always used. It had
a mean conductivity of 1.8 X io~' at 25°. The methyl alcohol
was first distilled from a small amount of dilute sulphuric
acid and then several times from lime. It had a mean specific
conductivity of 2.0 X io~' at 25°.

I This Journal. 17, 83 (1895).

138 Guy and Jones

Salts

In all cases, Kahlbaum's purest articles were used, and
these were recrystallized at least three times from conduc-
tivity water, carefully dried at 125°, and the solutions made
by direct weighing.

Viscosity

The viscosity measurements were made by means of the
Ostwald viscosimeter as modified by Jones and Veazey,* and
the size of the capillary so regulated as to be best adapted
to glycerol measurements. The method of calibration has
been discussed in detail by Schmidt. ^ Viscosity was cal-
culated from the formula

X = -^

in which rj is the viscosity coefficient for the liquid in question,
7)^ that of water, S the specific gravity of the liquid, t the time
of flow of the same, S^ the specific gravity of water at the given
temperature, and t^ the time of flow of the water. Fluidity
was calculated from the formula

V
where 6 represents the fluidity. The values of jjq are taken
from the work of Thorpe and Rodger,- being 0.00891 at 25°,
0.00720 at 35°, 0.00598 at 45°, 0.005057 at 55°, 0.004355 at
65°, and 0.003786 at 75°.

Temperature Coefficients

The temperature coefficients, both in per cent, and in conduc-
tivity units, have been calculated, the latter being simply
the actual increase in molecular conductivity per degree rise
in temperature, while the former were calculated from the
formula

Temp, coejj. of Uy = 5 • ^-— ^^

" '^ fi^ 25° 10

» Z. physik. Chem.. 61, 641 (1908).
2 Loc. cU.

Conductivity and Viscosity in Mixed Solvents 139

The temperature coefficients of fluidity were calculated
in the same way.

Viscosity measurements were made only with the tenth-
normal solutions, since at higher dilutions the difiference
between the viscosity of the solution and that of the solvent
was very slight.

Table I — Molecular Conductivity of Potassium Nitrate in

Glycerol

2t 25°,

55°. 45°

tiv25<'

//^35°

/<t,45°

0.337

0.681

1.248

0.368

0.754

1.384

100

0.373

0.769

I. 419

200

0.397

0.818

1-509

400

0.397

0.818

1.510

800

0.412

0.845

I 569

1600

0.431

0.900

1-739

Table II — Temperature Coefficients

Per cent.

Cond. units

^25

"-SS" 35°-45°

25°-35°

35°-45°

1020 0

■0833

0.0344

0.0567

1050 0

•0835

0.0386

0 . 0630

100

I061 0

.0847

0.0396

0.0650

200

1060 0

.0845

0 . 042 I

0.0691

400

1060 0

.0846

0 . 042 I

0.0692

800

105 1 0

.0857

0.0433

0.0724

1600

1084 0

•0932

0.0469

0.0839

e III—

Molecular Conductivity of Potassium Chloride

Glycerol at 23°,

55°, 45°

/^25<'

lxv^S°

/iv45°

0.385

0.772

I-413

0.405

0.841

I. 516

100

0.412

0.844

1-538

200

0.415

0.850

1-545

400

0.439

0.852

I-57I

800

0.443

0.870

1.623

1600

0.536

0.915

1.630

140

Table

Guy and

Jones

Table IV — Temperature Coefficients

Per cent.

Cond. units

25''-35» 35 "-45°

25°-35''

35''-4S''

0.1005 0.0830

0.0387

0.0641

0.1074 0.0804

0.0436

0.0675

100

0.1048 0.0822

0.0432

0 . 0694

200

0.1047 0.0818

0.0435

0.0695

400

0.0941 0.0844

0.0413

0.0719

800

0 . 0964 0 . 0865

0.0427

0.0753

1600

0.0708 0.0781

0.0379

0.0715

e V — Molecular Condiictivity

of Potassium

Bromide in

Glycerol at 23°,

35\ 45"

Pv2S-

H-viS"

Pv^S"

0.366

0.752

1.376

0.369

0.752

1.396

100

0.384

0.778

1-434

200

0.385

0.782

1-435

400

0.386

0.801

1.527

800

0.390

0.821

1.578

1600

0.413

0.877

1.667

Table VI — Temperature Coefficients

Per cent.

Cond. units

25°-3S'' 35°-45°

25°-35°

35°-45''

0.1054 0.0829

0.0386

0.0624

O.IO4I 0.0857

0 . 0383

0.0644

100

0.1028 0.0843

0.0394

0.0656

200

0.1031 0.0835

0.0397

0.0653

400

0 . 1080 0 . 0906

0.0415

0.0726

800

O.I 104 0.0922

0.0431

0.0757

1600

O.II23 0.0901

0 . 0464

0.0790

e VII-

Molecular Cojiductivity of Sodium

Chloride in

Glycerol at 25°,

35^ 45°

/.„25»

fCVSS"

ttv iS"

0.328

0.666

1.223

0.351

0.7II

I 319

100

0.353

0.720

1.350

200

0.372

0.753

1.409

400

0.375

0.765

I. 421

800

0.391

0.806

1.588

1600

0.395

0.825

1.629

Conductivity and Viscosity in Mixed Solvents 141

Table VIII — Temperature Coefficients

Per cent.

Cond. units

25°-35° 35°-45°

25°-35°

35°-45°

0.1030 0.0838

0.0338

0.0557

0.1024 0.0855

0 . 0360

0 . 0608

100

0.1038 0.0872

0.0367

0 . 0630

200

0.1024 0.0871

0.0381

0.0656

400

0.1040 0.0856

0 . 0390

0.0656

800

O.I061 0.0970

0.0415

0.0782

600

0.1090 0.0974

0 . 0430

0 . 0804

IX — Molecular Conductivity

of Sodium Iodide in Glycerol

at 25°, 33'

,45''

liv2S°

^v55°

,iv 45 °

0.342

0.690

1.265

0.364

0.737

I. 361

100

0.366

0.745

1.372

200

0.379

0.761

1-397

400

0.397

0.786

1.452

800

0.388

0.760

1. 418

1600

0.447

0.840

1-557

Table X — Temperature Coefficients

Per cent.

Cond. units

25°-35° 35°-45°

25°-35°

35°-45°

0.1047 0.0833

0.0348

0.0575

0. 102 I 0.0846

0.0373

0.0624

100

0.1035 0.0841

0.0379

0.0627

200

O.IOI9 0.0836

0.0382

0.0636

400

0.0978 0.0847

0.0389

0 . 0666

800

0.0959 0.0865

0.0372

0.0658

600

0.0879 0.0853

0.0393

0.0717

XI — Molecular Conductivity of Sodium

Bromide in

Glycerol at 25°,

55°, 45°

/V25°

/ir35°

^vi5°

0.318

0.646

I . 192

0.331

0.678

I .260

100

0.332

0.682

1.293

200

0.359

0.734

1.367

400

0.363

0.754

I .410

800

0.379

0.784

1.465

1600

0.384

0.791

1-515

142

Guy and Jones
Table XII — Temperature Coefficients

Per cent.

Cond. units

25 "-as" 35°-45°

25°-35°

35°^5°

0.1034 0.0846

0.0328 0.0546

0.1046 0.0864

0.0347 0.0582

100

0.1054 0.0884

0.0350 0.06 I I

200

0.1042 0.0868

0.0375 0.0633

400

0.1077 0.0870

0.0391 0.0656

800

0.1067 0.0869

0.0405 0.0681

1600

0.1068 0.0913

0.0407 0.0724

Table XIII-

—Molecular Conductivity of Sodium

Nitrate

Glycerol at 25°,

35°, 45°

/'t;25°

fiv35° rt

45°

0.303

0.617 I

129

0.331

0.677 I

239

100

0.338

0.707 I

284

200

0.355

0.735 I

362

400

0.358

0.737 I

378

800

0.372

0.766 I

412

1600

0.386

0 . 796 I

544

Table XIV — Temperature Coefficients

Per cent.

Cond. units

25 "-as" 35°-45»

25°-35°

35°-45°

0.1033 0.0828

0.0314 0.0512

0.1046 0.0830

0.0346 0.0562

100

0.1096 0.0818

0.0369 0.0577

200

0.1070 0.0851

0.0380 0.0627

400

0.1058 0.0870

0.0379 0.0641

800

0.1058 0.0843

0.0394 0.0646

1600

0.1062 0.0940

0.0410 0.0748

Table XV—

Molecular Conductivity of Ammonium

Chloride

Glycerol at 25°,

35°, 45°

I'v25°

Hv 35 " iiv

45°

0.393

0.801 I

452

O.4II

0.849 I

543

100

0.426

0.879 I

605

200

0.427

0.889 I

623

400

0.432

0.889 I

639

800

0.440

0.931 I

696

1600

0.442

0 . 948 I

709

Conductivity and Viscosity in Mixed Solvents 143

Table XVI — Temperature Coefficients

Per cent.

Cond.

units

25°-3S°

3S°-45°

25°-35°

35°-45°

0.1038

0.0812

0.0408

0.0651

0. 1065

0.0808

0.0438

0 . 0694

100

0. 1063

0.0827

0.0453

0.0726

200

0. 1080

0.0825

0.0462

0.0734

400

0.1057

0 . 0844

0.0457

0.0750

800

O.III3

0.0822

0.0491

0.0765

1600

0.1123

0 . 0803

0.0506

0.0761

Table XVII — Molecular Conductivity of Ammonium Bromide
in Glycerol at 25°, 35°, 43°

/'7j25°

P„35°

w45°

0.373

758

1 391

0.391

802

490

100

0.397

824

531

200

0.422

878

632

400

0.430

889

642

800

0.444

.926

694

1600

0.492

034

1.864

Table XVIII— Temperature Coefficients

Per cent.

Cond.

units

25°-35° 35 "-45°

25°-35°

35°-45°

0.1032 0.0835

0.0385

0.0633

O.IO5I 0.0850

O.O4II

0.0688

100

0.1075 0.0856

0.0427

0.0707

200

0.1075 0.0862

0.0456

0.0754

400

0.1069 0.0847

0.0459

0.0753

800

0.1092 0.0829

0.0482

0.0768

1600

0.II02 0.0803

0.0542

(

3.0830

Table XIX— Molecular Conductivity of Ammonium Nitrate in
Glycerol at 23°, 33°, 45°

/xa,25<'

/.viS"

45°

0345

0.696

1.272

0.379

0.778

440

100

0.392

0.805

488

200

0.407

0.840

547

400

0.417

0.869

594

800

0.396

0.825

579

600

0.437

0.917

651

144

Guy and Jones

Table XX — Temperatur

e Coefficients

Per cent.

Cond. units

25°-35°

35 "^5°

25''-35''

35°-45°

0.1020

0.0832

0.0351

0.0576

0.1053

0.0851

0.0399

0 . 0662

100

0. 1058

0.0850

0.0413

0.0683

200

0 . 1063

0 . 0844

0.0433

0.0707

400

0. 1084

0.0835

0.0452

0.0725

800

0. 1084

0.0914

0.0429

0.0754

1600

0.1095

0 . 0802

0 . 0480

0.0734

Table XXI — Molecular Conductivity of Barium Chloride in
Glycerol at 25°, 35°, 45°

liv2S°

^v25°

45°

0.315

664

I .221

0.432

915

695

100

0.464

978

803

200

0.502

056

951

400

0.520

lOI

994

800

0.561

197

230

1600

0.565

332

368

Table XXII— Temperature Coefficients

Per cent.

Cond.

units

25°-35° 35°H15»

25°-35°

35°-45°

O.IIO8 0.0839

0.0349

0.0557

O.III5 0.0853

0.0483

0.0780

100

O.I 108 0.0844

0.0514

0.0825

200

O.I 103 0.0852

0.0554

0.0895

400

O.III6 O.081I

0.0581

0 . 0893

800

O.H34 0.0863

0.0636

0.1033

1600

0.1358 0.0778

0.0767

).io36

Table XXIII — Molecular Conductivity of Barium Bromide in
Glycerol at 25°, 35°, 43°

l^v 25°

M,35°

Pn,^5°

0.330

0.696

1-314

0.396

0.832

1.566

100

0.426

0.900

1.698

200

0.443

0.938

1-774

400

0.474

I .001

1.896

800

0.520

I. 127

2. 115

1600

0.530

I 157

2.200

Conductivity and Viscosity in Mixed Solvents
Table XXIV — Temperature Coefficients

[45

Per cent.

Cond

units

25°-35°

35»-45°

25°-35°

35°-45°

0. I 109

0.0888

0.0366

0.0618

0. IIOI

0.0882

0 . 0436

0.0734

100

0. III2

0.0887

0.0474

0.0798

200

0. III7

0.0894

0.0495

0.0836

400

0. III2

0.0894

0.0527

0.0895

800

0. I 160

0.0876

0.0607

0.0988

1600

O.II80

0 . 0900

0.0627

0 . 1043

e XXV

— Molecular Conductivity of Barium Nitrate

Glycerol at 25°,

35°, 45°

Iiv25°

^v35°

/<T,45°

246

0.517

0.959

347

0.738

367

100

368

0.792

479

200

401

0.871

634

400

414

0.904

719

800

456

0.988

871

1600

462

0.991

897

Table XXVI— Temper

T.ture Coefficients

Per cent.

Cond.

units

25 "-35"

35°-45°

25 °-35 "

35°-45°

0. IIOI

0.0854

0.0271

0 . 0442

0. II26

0.0852

0.0391

0.0629

100

O.II52

0.0867

0.0424

0.0687

200

0 . 1 1 70

0.0876

0.0470

0.0763

400

O.II68

0.0901

0 . 0490

0.0815

800

0. I 166

0.0893

0.0532

0.0883

1600

O.II45

0.0914

0.0529

) . 0906

Table XXVII — Molecular Conductivity of Calcium Bromide in
Glycerol at 25°, 35°, 45°

to2S»

p„35°

45°

0.245

0.519

0.972

0.324

0.687

298

100

0.340

0.729

374

200

0.373

0.803

514

400

0.386

0.833

556

800

0.395

0.882

721

1600

0.408

0.909

743

146

Guy and Jones

Table XXV III— Temperature Coefficients

Per cent.

Cond.

units

25°-35°

35°-45°

25 "-350

35°-45°

0. III8

0.0873

0.0274

00453

0. 1 120

0.0888

0.0363

0.061 I

100

0. I 144

0.0883

0.0389

0 . 0645

200

O.II52

0.0886

0 . 0430

O.O7II

400

O.II57

0.0891

0.0447

0.0723

800

0.1233

0.0951

0.0487

0.0839

1600

0.1226

0.0918

0.0501

0.0834

Table XXIX — Molecular Conductivity of Strontium Bromide
in Glycerol at 25°, 35°, 45°

Table

!<v 25°

;/„35°

45°

0.264

0.556

I 054

0.340

0.717

362

100

0.365

0.776

468

200

0.388

0.831

581

400

0.391

0.876

659

800

0.409

0.886

681

1600

0.428

0.924

758

Table XXX— Temper

%ture Coefficients

Per cent.

Cond.

units

0-35° 35°-45°

25°-35°

35°-45°

1 106 0

0895

0.0292

0 . 0498

III8 0

0899

0.0377

0 0645

100

II26 0

0892

0.041 I

0.0692

200

II33 0

0903

0.0443

0.0750

400

I189 0

0893

0 . 0485

0.0783

800

1 166 0

0895

0.0477

0.0795

1600

II62 0

0902

0 . 0496

0.0834

e XXXI—

Molecular Conductivity of Strontium Nitrate

in Glycerol

at 25

°, 35°, 45°

ft-v 25°

,<r35°

/it. 45°

0.235

0.501

0934

0.323

0.687

I .292

100

0.349

0.744

1-394

200

0.392

0.833

1-563

400

0.401

0.872

1.686

800

O.4II

0.891

1 .671

1600

0.449

0.945

759

Conductivity and Viscosity in Mixed Solvents 147

Table XXXII — Temperature Coefficients

25 "-JS" 35°-45»

25 "-35°

35°-45°

O.I I 27 0.0864

0.0266

0.0433

O.II2I 0.0885

0 . 0364

0 . 0605

100

O.II3I 0.0871

0.0395

0.0650

200

O.II2I 0.0876

0.0441

0.0730

400

O.II73 0.0933

0.0471

0.0814

800

O.I I 70 0.0874

0 . 0480

0.0780

1600

O.I 102 0.0861

0 . 0496

0.0814

e XXXIII— Molecular Conductivity of Cobalt Chloridt

Glycerol at 25°,

J5°, 45°'

/<r25<'

W35°

,„;45°

0.270

0.546

1.003

0.369

0.744

373

100

0.391

0.784

450

200

0.455

O.9II

691

400

0.473

0.959

779

800

0.497

1.005

856

1600

0.519

I .040

I .920

Table XXXIV— Temperature Coefficients

Per cent.

Cond.

units

100
200
400
800
1600

25°-35°

o. 1023

O.IOI5

o. 1004

0.1004

o. 1027
o. 1022
o. 1002

35°-45°
0.0836

o . 0846
o . 0849

0.0857
0.0855

o . 0847
o . 0846

25°-35°
0.0276
0.0375
0.0393

o . 0456
o . 0486
0.0508
0.0521

35°-45''

0.0457
0.0629

O . 0666

0.0780
0.0820
0.0851

o . 0880

Table XXXV-

100

200

400

800

1600

-Molecular Conductivity of Cobalt Bromide in
Glycerol at 25°, 35°, 45°

uv 35 ° ud 45 "

Mv 25°

0.364

0.460

0.468

0.514

0.533

0.552

0.564

0.744
0.932

0.953
I 045
1.076
I. 103
I .091

.370
.702

•743
.911

■967
031
.005

148

Guy and Jones

Table XXXVI — Temperature Coefficienis

Per cent.

Cond

units

°-35° 35°-#5°

25 "-35"

35 "--tS"

1043 0

0841

0.0380

0.0626

1026 0

0826

0.0472

0.0770

100

1036 0

0829

0.0485

0 . 0790

200

1032 0

0827

0.0531

0 . 0866

400

102 1 0

0827

00543

0.0891

800

0998 0

0841

0.0551

0.0928

600

0934 0

0837

0.0527

0.0914

XXXVII

— Molecular Conductivity of Potassium Chlot

in Glycerol at 3^

°. 65°, 75°-

/<^55°

Mv65°

fv 75°

2.391

3 • 755

5.601

601

124

6.176

100

707

252

6.300

200

734

341

6.489

4CX)

738

470

6.691

800

817

562

6.862

1600

940

693

6.891

Table

XXXVIII-

-Temp

erature Coefficients

r cent.

Cond.

units

55°-65°

65°-75°

55 "-65°

65 °-75 "

0.0570

0.0491

0.1364

0.1846

0.0586

0.0497

0.1523

0.2052

100

0.0571

0 . 0482

0.1545

0 . 2048

200

0.0588

0 . 0496

0. 1607

0.2148

400

0.0632

0.0499

0.1732

0.2221

800

0.0623

0 . 0504

0.1745

0.2300

1600

0.0596

0.0470

0.1753

0.2198

Table XXXIX — Molecular Conductivity of Potassium Bromide
in Glycerol at 53°, 63°, 73°

^.^55°

,,-,65"

/'v75"

2.293

3.619

5-332

2-453

906

4.786

100

2-557

062

6.080

200

2.606

122

6.154

400

2.680

275

6.317

800

2.705

286

6.408

600

2.770

400

6.897

Conductivity and Viscosity in Mixed Solvents

149

Table XL — Temperature Coefficients

Per cent.

Cond.

units

55°-65°

65°-75°

55 °-65 °

65°-75°

0.0576

0.0473

0.1326

O.I713

0.0592

0.0481

0.1453

0.1880

100

0.0587

0 . 0496

0.1505

0.20l8

200

0.0572

0.0493

O.I516

0.2032

400

0.0594

0.0477

0.1595

0 . 2042

800

0.0584

0 . 0496

O.I581

0.2122

1600

0.0588

0.0568

0. 1630

0 . 2497

Table XLI — Molecular Conductivity of Sodium Bromide in
Glycerol at 55°, 65°, 75°

HvSS"

65°

fin)

75°

2.006

153

4 763

2.203

500

262

100

2.299

656

504

200

2.325

683

566

400

2.397

715

753

800

2.438

760

864

1600

2.493

965

938

Table XLII — Temperatu

re Coefficients

Per cent.

Cond.

units

5S<»-65° 65"»-75°

S5''-6S°

65°-75<»

0.0570 0.0510

O.II47

0. 161O

0.0588 0.0503

0.1297

0. 1762

100

0.0590 0.0505

0.1357

0.1848

200

0.0584 0.051 I

0.1358

0.1883

400

0.0550 0.0548

O.I318

0.2038

800

0.0542 0.0559

0.1322

0.2104

1600

0.0590 0.0497

0.1472

(

5.1973

Table XLIII-

-Molecular Conductivity of Sodium Iodide
Glycerol at 55°, 65°, 75°

lxv5S°

65°

75°

2. lOI

3 300

4.878

2.246

568

407

100

2.347

731

590

200

2.377

756

604

400

2.441

865

822

800

2.410

833

745

600

2.591

263

415

ISO

Guy and Jones

Table XLIV — Temperature Coefficients

Per cent.

Cond.

units

55°-65°

65 °-75 °

ss^-es"

65 "-75°

0.0570

0.0478

0.1199

0.1578

0.0588

0.0515

0.1322

0.1839

100

0.0589

0 . 0498

0.1384

0.1859

200

0.0581

0.0492

0.1379

0.1848

400

0.0584

0.0506

0. 1424

0.1957

800

0.0591

0.0498

0.1423

0. I912

1600

0.0644

0 . 0644

0. 1672

0.2152

Table XLV — Molecular Conductivity of Ammonium Chloride
in Glycerol at 33°, 65°, 73°

/<,,55°

fiv 65°

tiv 75°

2.785

313

6.285

2.863

498

6.593

100

3.109

821

7 033

200

3 144

789

7.018

400

3 146

858

7. 162

800

3252

051

7.409

1600

3.224

015

7-351

Table XLV I — Temperature Coefficients

Per cent.

Cond.

units

SS^-eS" 65 "-75°

55°-65°

65 °-75 °

0.0545 0.0457

0.1528

0.1972

0.0571 0.0466

0.1635

0.2095

100

0.0550 0.0459

O.I7I2

0.2212

200

0.0523 0.0465

0.1645

0.2229

400

0.0544 0.0465

O.I712

0.2304

800

0.0553 0.0466

0.1799

0.2358

600

0.0557 0.0465

O.I79I

0.2336

Table XLV 11 — Molecular Condtictivity of Ammonium Nitrate
in Glycerol at 55°, 65°, 75°

Pv55»

mf'S"

^75°

2.558

3 942

5-873

2.766

4.250

6.310

100

2.907

4 458

6.772

200

2.947

. 4580

6.844

400

3015

4.661

6.956

800

3 103

4-754

7.107

600

3 194

4-923

8.372

Conductivity and Viscosity in Mixed Solvents
Table XLVIII — Temperature Coefficients

Per cent.

Cond.

units

55°-65°

65°-75°

55°-65°

65°-75°

0.0541

0 . 0489

0.1384

0. I93I

0.0536

0.0485

0. 1484

0 . 2060

100

0.0533

0.0519

O.I55I

0.2314

200

0.0554

0.0494

0.1633

0.2264

400

0.0545

0 . 0492

0. 1646

0.2295

800

0.0532

0.0494

O.165I

02353

1600

0.0541

0 0497

0.1729

0 . 2449

t5i

Table XLIX — Molecular Conductivity of Barium Nitrate in
Glycerol at 55°, 65°, 75°

liv5S°

65°

75°

2.262

3 565

5 300

856

480

725

100

106

906

304

200

362

269

858

400

555

629

555

800

757

987

046

1600

942

236

466

Table L — Temperature Coefficients

Per cent.

Cond.

units

55°-65° 65°-75°

55 =

-65°

65°-75°

0.0576 0

0486

0. 1

303

0.1735

0.0569 0

0499

0. 1

624

0.2245

100

0.0579 0

0491

0. 1

800

0.2394

200

0.0567 0

0491

0.1

907

0.2589

400

0.0579 0

0519

0.2074

0.2926

800

0.0593 0

.0511

0.2230

0.3059

1600

0.0581 0

•0517

0.2294

0.3230

e LI — Molecular Conductivity

of Strontium

Chloride in

Glycerol at 55°,

65°, 75°

nvSS"

Hv(>5°

/<^,75°

2.243

3 576

5-378

727

312

442

100

900

610

880

200

lOI

946

423

400

314

257

855

800

389

400

078

1600

645

750

780

152

Guy and Jones

Table LII — Temperature Coefficients

lo
50
100
200
400
800
1600

Per cent.

55°-65°
0.0594
0.0581
0.0589
0.0592
0.0587
0.0593
0.0577

65°-75°

o . 0503

0.0493

o . 0492

0.0501

o . 0494

0.0495
0.0527

55°-65°
01333
0.1585
O.I7IO
0.1845

0.1943
0.20II
0.2105

65 "-75°

o. 1802

0.2130
0.2270
0.2477
0.2598
0.2678
0.3030

Table LIII-

-Molecular Conductivity of Cobalt Chloride in
Glycerol at 55°, 65°, 75°

tiv 55°

65°

tiv

75°

1.789

778

4. 102

2.373

686

447

100

2.610

074

024

200

2.890

513

687

400

3.104

864

236

800

3.286

178

750

1600

3-471

503

247

Table LIV — Temperatu

re Coefficients

Per cent.

Cond.

units

SS^-eS" 65 "-75°

55°-65°

65°-75°

0.0553 0.0476

0 0989

0.1324

0.0553 0.0477

O.I313

O.I761

100

0.0560 0.0478

0. 1464

0.1950

200

0.0561 0.0481

0.1623

0.2174

400

0.0566 0.0487

0.1760

0.2372

800

0.0575 0.0496

0.1892

0.2572

1600

0.0585 0.0497

0.2032

0.2744

Table LV-

-Molecular Conductivity of Cobalt Bromide in
Glycerol at 55°, 63°, 73""

Pi, 55°

.«i

65°

75°

2.340

3.676

5.462

905

561

841

100

952

628

954

200

229

068

584

400

338

242

904

800

429

420

549

1600

400

399

112

Conductivity and Viscosity in Mixed Solvents 153

Table LVI — Temperature Coefficients

10
50
100
200
400
800
1600

55°-65°
0.0571
0.0571
0.0568
0.0569
0.0572
0.0582
0.0588

65°-75°
0.0485

o . 0499

0.0503

o . 0496

0.05II
0.0596

0.0508

55°-65°
0.1336

o. 1656
o. 1676

0.1839

o. 1904
o. 1991

0.1999

0.1786
0.2280
0.2326
0.2516
0.2662
0.3129
0.2713

Table LVII — Molecular Conductivity of Potassium Chloride in

50
100
200
400
800
1600

Glycerol at 25*^

fv 25°
0.385
0.405
0.412

0.415
0.439
0443
0.536

35 , 45'

fv 35°
0.772
0.841
0.844
0.850
0.852
0.870
0.915

■413
.516
■538

•545
■571
.623
.630

Table LVII I — Temperature Coefficients

10
50
100
200
400
800
1600

25°-35°

o . 1006
o. 1074
o. 1049
o. 1047
o . 0948
o . 0962
0.0707

35°-45°
0.0830
0.0804
0.0822
O.081S
0.0842
0.0865
0.0781

25°-35°
0.0387

o . 0436

0.0432
0.0435
0.0413
0.0427
0.0379

35°-^5°
0.0641
0.0675
0.0694

o . 0695

0.0719
0.0753
0.0715

Table LIX — Molecular Conductivity of Potassium Chloride
of 75 Per cent. Glycerol with Water at 25°,

- a Mixture
u 45°

10
50
100
200
400
800
1600

t'v25°

5-33
578
5-86
6.07
6.38
6.61
6.51

,«X; 35 °

8.29

!tv 45°
II .92

154

Guy and Jones

Table LX

' — Temperature Coefficients

Per cent.

Cond. units

25 °-35 °

35°-45°

25 "-as"

35 "-45°

0.0554

0 . 0438

0.296

0.363

0.0556

0 . 0449

0.322

0.404

100

0.0549

0.0441

0.322

0.401

200

0.0548

0 . 0460

0.332

0.432

400

0.0550

0 . 0463

0.351

0.458

800

0.0553

0.0427

0.366

0.439

1600

0.0558

0 . 0469

0.364

0.477

Table LXI — Molecular Conductivity of Potassium Chloride
in a Mixture of 50 Per cent. Glycerol with Water at 25°,
35°> 45°

/<i-25°

(l-U

35°

liV

45°

23-55

40.28

100

200

400

800

1600

Table LXII—

Temperatur

e Coefficients

Per cent.

Cond

units

25°-35°

35°-45°

25°-35°

35°-i5°

0.0341

0.0275

0.804

0.869

0.0345

0.0294

0.872

0.995

100

0.0336

0.0316

0.873

I 095

200

0.0338

0 . 0300

0.887

1.056

400

0.0344

0.0294

0.983

I. 128

800

0.0344

0.0282

1.007

I . 109

1600

0.0345

0.0273

I 055

121

Table LXIII — Molecular Conductivity of Potassium Chloride
in a Mixture of 25 Per cent. Glycerol with Water at 25°,
33°, 45°

^'v

25°

Mv35-

Hv45°

5981

74 52

90. 16

98.63

100

101.08

200

103.36

400

112.24

800

116.68

1600

121.32

Conductivity and Viscosity in Mixed Solvents
Table LXIV — Temperature Coefficients

155

Per cent.

Con(

i. units

25°-35°

35°-45°

25°-35°

35 "-45°

0.0246

0.0212

I. 471

1.564

0.0258

0 . 0204

689

100

0.0244

0.0216

626

814

200

0.0253

0.02 1 1

721

802

400

0.0243

0.0206

817

920

800

0.0238

0.0213

845

038

1600

0.0253

0.0226

999

234

Table

LXV — Molecular Conductivity of Potassium Chloride
in Water at 25°, 33°, 43°

w25°

^v^i

tiv^S

120.4

143

0 166.7

129.7

154

5 181

100

132.0

158

5 184

200

1353

161

6 189

400

137-7

165

4 193

800

138. 1

165

8 194

1600

140.3

169

3 197

Table LXVI—

Temperature

Coefficients

Per cent.

Cond. units

25°-35°

35 °-45 °

25°-35° 35°-45°

0.0188

0.0158

2.26 2.37

0.0192

0.0171

48 2.67

100

0.0200

0.0166

65 2.62

200

0.0195

0.0171

63 2.77

400

0.0201

0.0171

77 2.84

800

0.0201

0.0174

77 2.90

1600

0.0206

0.0169

90 2 A

Table LXVII — Molecular Conductivity of Potassium Chloride
in a Mixture of 73 Per cent. Glycerol zvith Ethyl Alcohol at
25" ^ '^

°, 35°, 45°

liv 25°

,-^35°

,.x,45°

I. 21

2.05

3-26

I 31

2.25

100

1-35

2-34

200

1. 41

2.43

400

1-53

2.63

800

1-54

2.67

1600

1-59

2.72

156 Guy and Jones

Table LXVIII — Temperature Coefficients

Per cent.

Cond. units

25°-35°

35°-45°

25°-35° 35°-45°

0.0694

0.0590

0.084 O.I2I

0.0717

0.0596

0.094 0.134

100

0.0733

0.0577

0.099 0.135

200

0.0723

0 . 0605

0.102 0.147

400

0.0719

0 . 0605

o.iio 0.159

800

0.0733

0.0599

O.II3 0.160

1600

0.0710

0.0588

O.II3 0.160

Table LXIX-

—Molecular Conductivity

of Potassium Chlot

in a Mixture of 50

Pet

cent. Glycerol with Ethyl Alcoho

25°, 35°

, 45°

flV

25°

liv

35° A<^45°

48 6.29

21 7

100

63 7

200

94 8

400

56 9

800

76 9

1600

84 9

Table LX.

K—

Temperature Coefficients

r cent.

Cond. units

25°-35°

35°-45°

25°-35° 35°-45°

0.0459

0 . 0404

O.I4I O.18I

0.0471

0.0420

0.167 0.217

100

0 . 0500

0.0396

0.187 0.223

200

0.0451

0 . 0409

0.185 0.243

400

0 . 0490

0.0413

0.216 0.271

800

0.0491

0 . 042 I

0.224 0.295

1600

0 . 048 I

0.0431

0.222 0

295

Table LXXI — Molecular Conductivity of Potassium Chloride
in a Mixture of 25 Per cent. Glycerol with Ethyl Alcohol^at
25°, 35°, 45°

25°

A.^35°

fiv45-

7.26

931

11.94

10.78

13.61

100

12.15

1539

200

13.02

16.61

400

15-31

I915

800

15.68

20.28

1600

16.31

21.06

Conductivity and Viscosity in Mixed Solvents 157

Table LXXII — Temperature Coefficients

10
50
100
200
400
800
[600

25°-35°
0.0281
0.0297

o . 0308
o . 0306

0.0352
0.0320
0.0319

35°-45°
0.0282
0.0262
0.0267
0.0276
0.0251
0.0293
0.0281

25 "-35°
0.205
0.247
0.286
0.305
0.399
0.380

0-394

35 "-45°
0.263
0.283
0.324

0.359
0.384
0.460
0.475

Table LXXIII — Molecular Conductivity of Potassium Chloride
in a Mixture of 75 Per cent. Glycerol with Methyl Alcohol

at 25°,

10
50
100
200
400
800
1600

35 , 45'

Pv 25°
2 .22
2.41
2.47
2.58
2.78
2.83
2.83

350
58

93
07
21
52
64
62

/iv 45°

5-43
591
6. II
6.38
6.88
7.07
6.99

Table LXXIV — Temperature Coefficients

10
50
100
200
400
800
1600

25°-35°
O . 06 1 2
0.0630

o . 0640

0.0632
0.0625
0.0639
0.0632

35°-45°

0.0517

0.0505

o . 0500

0.0515
0.0522
0.0524
0.0515

25°-35°
0.136
0.152

o. 160

0.163

0.174
O.181
0.179

35°-45»
0.185

o. 198

0.204
0.217
0.236
0.243
0.237

Table LXXV — Molecular Conductivity of Potassium Chloride
in a Mixture of 50 Per cent. Glycerol with Methyl Alcohol
at 25^ 35°, 45°

;<r25°

flV

35°

«,45°

8.10

II .09

14-54

9.24

16.71

100

9-59

17.48

200

10.05

18.22

400

II .04

20. 17

800

II .20

20.41

600

11.38

20.64

158

Guy and Jones

Table LXXVI-

—Temperature Coefficients

Per ceni

Cond.

units

25»-35°

35 "-45°

25 "-as"

35 "^5 °

0.0369

0.0311

0.299

0.345

0.0378

0.0311

0.351

0.396

100

0.0374

0.0326

0.358

0.431

200

0.0375

0.0323

0.372

0.445

400

0.0376

0.0324

0.416

0.497

800

0.0371

0.0330

0.414

0.507

1600

0.0365

0.0321

0.425

0.501

Table LXXVII — Molecular Conductivity of Potassium Chloride
in a Mixture of 25 Per cent. Glycerol with Methyl Alcohol
at 25°, 35°, 45°

Table

liV

25°

J5°

45°

21.76

26.55

31. II

-75

100

-36

200

-30

400

-51

8no

•85

1600

33-99

42.05

49-55

Table LXXVIII—

Temperature Coefficients

Per cent.

Cond

units

25°-35'' 35°-45°

-35°

35°-45°

0220 0

.0172

479

0.456

0218 0

.0200

560

0.630

100

0221 0

•0199

608

0.671

200

0230 0

.0197

662

0.696

400

0231 0

.0193

718

0.732

800

.0227 0

.0200

755

0.815

1600

.0237 0

.0179

806

0.750

LXXIX-

—Molecular

Conductivity

Sodium Ni

in Glycerol

at 25

\ 35°,

45°

t,v25°

Mv35°

^v45°

0.303

0.617

I. 129

0.331

0.677

239

100

0.338

0.707

284

200

0.355

0.735

362

400

0.358

0.737

378

800

0.372

0.766

412

1600

386

796

544

Conductivity and Viscosity in Mixed Solvents
Table LXXX — Temperature Coefficients

Per cent.

Cond.

units

25°-35°

35°-45°

25°-35°

35°-45°

0.1033

0.0828

0.0314

0.0512

0. 1046

0.0830

0.0346

0.0562

100

0. 1096

0.0816

0.0369

0.0577

200

0.1070

0.0853

0.0380

0.0627

400

0.1058

0.0869

0.0379

0.0641

800

0. 1058

0 . 0843

0.0394

0 . 0646

600

0. 1062

0.0939

0.0410

0.0748

t59

Table LXXXI — Molecular Conductivity of Sodium Nitrate
in a Mixture of 73 Per cent. Glycerol with Water at 25°,
35°, 45°

flV

25°

35°

m.^s"

4.88

10.80

12.03

100

12.33

200

12.58

400

13.65

800

14.20

1600

6.37

14 34

Table LXXXII

—Tempe

rat

ure Coefficients

Per cent.

Cond.

units

25°-35<'

35°-^5°

25°-35°

35°^5°

0.0529

0 . 0448

0.258

0.334

0.0561

0.0434

0.302 .

0.364

100

0.0549

0 . 0460

0.299

0.389

200

0.0541

0.0449

0.305

0.390

400

0.0534

0.0459

0.326

0.430

800

0.0538

0.0455

0.341

0.445

1600

0.0531

0.0471

0.338

0.459

Table LXXXIII — Molecular Conductivity of Sodium Nitrate
in a Mixture of 50 Per cent. Glycerol with Water at 25°,
35°, 45°

liv25°

ftv

35°

flV

45°

18.87

2541

33 03

20.60

100

21 .26

200

21 .46

400

21 .69

800

23 -73

1600

24-53

i6o

Guy and Jones

Table LXXXIV — Temperature Coefficients

Per cent.

Cond. units

25°-35°

35°-45°

25°-35° 35°-45°

0.0348

0.0298

0.654 0.762

0.0350

0.0298

0.724 0.824

100

0.0352

0.0297

0.753 0.856

200

0.0367

0.0295

0.788 0.864

400

0.0365

0.0294

0.794 0.879

800

0.0338

0.0329

0.801 I 043

1600

0.0329

0.0341

0.804 I. 112

Table LXXXV — Molecular Conductivity of Sodium Nitrate
in a Mixture of 25 Per cent. Glycerol with Water at 25°,
35°, 45°

Hv^S"

W35°

//-1, 45

48.19

60.40

7381

52.

64.

80.77

100

53-

68.

82.75

200

54-

69.

84.41

400

55-

69.

86.03

800

60.

75-

93.20

1600

62.

77-

96.30

Table LXXXVI-

-Temperature Coefficients

Per cent.

Cond.

units

25°-35''

35°-45°

°-35°

35°-45''

•0253

0.0222

221

I 341

0244

0.0244

273

1-587

100

0272

0.0213

460

I 450

200

0267

0.0221

471

1-523

400

0264

0.0233

449

I .629

800

0254

0.0236

526

1.785

600

0254

0.0235

587

1.840

LXXXV 11— Molecular Conductivi

ty of Sodium Nii

in Water at 25°

. 35°,

45°

Iiv25°

Pi, 35°

/.v45°

94-7

II3-4

133 -2

103

125

147

100

104

•7

127

149

200

107

130

•5

153

400

113

•7

135

•3

159

800

113

135

160

1600

116

142

169

Conductivity and Viscosity in Mixed Solvents 16 1

Table LXXXVIII — Temperature Coefficients

Per cent.

Cond.

units

25 "-35°

35*>-45°

25°-35°

35°-45°

0.0198

0.0175

1.87

1.98

0 . 0204

0.0180

2. 12

2.25

100

0.0212

0.0176

2.23

2.25

200

0.02II

0.0174

2.27

400

0.0190

0.0179

2.16

2-43

800

0.0201

0.0179

2.28

2-43

600

0.0230

0.0190

2.66

2.71

Table LXXXIX — Molecular Conductivity of Sodium Nitrate

in a Mixture of 7^ Per
at 23°, 35°, 45'

cent. Glycerol with Ethyl Alcohol

liX

25°

35° KT,

45 »

1.02

1.77 2.79

•17

I 99 3

100

.20

2.09 3

200

.26

2.19 3

400

•38

2.37 3

800

•39

2-43 3

1600

I 39

2.42 3

Table XC— Temperature Coefficients

Per cent.

Cond. units

25°-35°

350-45°

25 0-35° 35°-45''

0.0736

0.0576

0.075 0.102

0.0701

0.0605

0.082 0.121

100

0.0742

0.0576

0.089 0.121

200

0.0739

0 . 0602

0.093 0.132

400

0.0721

0.0582

0.099 0.138

800

0.0746

0.0579

0.104 0.141

600

0.0742

0.0600

0.103 0.145

XCI-

-Molecular Conductivity of Sodium Nitrate

Mixture of 50 Per

cent. Glycerol with Ethyl Alcoho

'°. 35°

, 45°

Hv25°

Mv35° juv45°

3 08

4.49 6.20

41 7.58

100

74 8.07

200

00 8.44

400

67 9.49

800

95 9- 78

1600

14 10.

Table a.l.1 — Moiecuiar Lonauctivtty oj ::^oatum isitrate tn
a " ' " "' . . - ~

l62

Guy and Jones

Table XCII-

-Temperature Coefficients

Per cent.

Cond. units

25°-35°

35 "-45°

25 "-as" 35 "--15°

0.0457

0.0381

O.I4I O.I7I

0 . 0470

0 . 0400

0.173 0.217

100

0.0475

0 . 0406

0.185 0.233

200

0.0478

0 . 0406

0.196 0.244

400

0.0475

0.0422

0.215 0.282

800

0.0478

0.0393

0.225 0.273

1600

0.0487

0.0426

0.234 0.304

Table XCIII — Molecular Conductivity of Sodium Nitrate in
a Mixture of 25 Per cent. Glycerol with Ethyl Alcohol at
25°, 35°, 43°

250

/iv

35°

i5°

7.36

11.74

5,0

100

200

400

800

1600

Table XCIV—

Temperatuf

e Coefficients

Per cent.

Cond.

units

25°-35°

35°-45°

25°-35°

35°-45°

0.0284

0.0243

0.209

0.229

0.0288

0.0246

0.281

0.309

lop

0.0292

0.0269

0.308

0.368

200

0.0290

0.0270

0.335

0.402

400

0 0305

0.0268

0.396

0.449

800

0.0288

0.0261

0.397

0.468

1600

0 . 0305

0.0244

0.436

436

Table XCV — Molecular Conductivity of Sodium Nitrate in
a Mixture of 75 Per cent. Glycerol with Methyl Alcohol at
25°, 35°, 45°

P^25°

TO 35°

M.45°

1.86

2.99

4-54

2.07

342

531

100

2.17

358

5-43

200

2.24

364

5.62

400

2.41

3-99

6.02

800

2.53

4.08

6.24

600

2.49

413

6.26

Conductivity and Viscosity in Mixed Solvents 163
Table XCVI — Temperature Coefficients

25°-35°

35 "-45°

25 "-35° 35 "-45°

0.0603

0.0519

O.II3 0.155

0.0652

0.0552

0.135 0.189

100

0.0650

0.0521

O.I4I 0.185

200

0,0714

0.0544

0.160 0.198

400

0 . 0654

0.0510

0.158 0.203

800

0.0613

0.0532

0.155 0.216

1600

0.0658

0.0515

0.164 0.213

e XCVII — Molecular Conductivity of Sodium Nitrate ii

Mixture

of 50 Per cent. Glycerol with Methyl Alcoko

25°, 35°

43°

liv2S°

/<^35° /,^45«'

7 35

10.02 1325

11.88 15

100

12.53 16

200

13.22 17

400

14.46 19

800

14.87 19

1600

15.08 19

Table XCVIII-

—Temperature Coefficients

Per cent.

Cond. units

100
200
400

800

1600

25 °-35°
0.0363
0.0368
0.0377
0.0378
0.0385
0.0383

o . 0390

350-450
0.0315
0.0320
0.0314
0.0304
0.0318
0.0316
0.0293

25°-35°
0.267
0.320

o 344
0.363
0.402
0.412
0.428

35°-45°

0.323
0.381

o 394
0.431
0.460
0.470
0.449

Table XCIX — Molecular Conductivity of Sodium Nitrate in a
Mixture of 25 Per cent. Glycerol with Mrthyl Alcohol at

, 35'

10
50
100
200
400
800
1600

I'v 25°
20.77
2571
2759
28.81
30.06

33"
34 00

/,v45°

30 • 59

1 64

Guy and Jones
Table C — Temperature Coefficients

Per cent.

Cond.

units

25 "-as"

35 "-45°

25 "-35°

35 0-45 »

0.0214

0.0210

0.445

0.537

0.0223

0.0198

0.564

0.612

100

0.0224

0.0198

0.622

0.650

200

0.0220

0.0196

0.646

0.692

400

0.0225

0.0205

0.682

0.757

800

0.0218

0.0192

0.731

0.778

600

0.0230

O.OI9I

0.782

0.796

Table CI — Molecular Conductivity of Ammonium Bromide in
Glycerol at 25°, 35°, 45°

/<^25°

Piv 35° p.

45 »

0.373

•758 I

391

0.391

802 I

490

100

0.397

824 I

531

200

0.422

878 I

632

400

0.430

.889 I

642

800

0.444

.926 I

694

1600

0.492

•034 I

864

Table CII — Temperature Coefficients

Per cent.

Cond. units

25°-35° 35°-45°

25 "-35°

35°-45<'

0.1032 0.0838

0.0385 0.0633

O.IO5I 0.0850

0.041 I 0.0688

100

0.1075 0.0850

0.0427 0.0707

200

0.1080 0.0862

0.0456 0.0754

400

0.1069 0.0847

0.0459 0.0753

800

0.1085 0.0829

0 . 0482 0 . 0768

600

O.I 106 0.0802

0.0542 0.0830

Table cm — Molecular Conductivity of Ammonium Bromide
in a Mixture of 75 Per cent. Glycerol with Water at 25°,
35°, 45°

25°

Pv35°

45°

5-53

8.48

12.28

9.14

100

9.25

200

9-54

400

10.28

800

10.81

1600

II .20

Conductivity and Viscosity in Mixed Solvents 165

Table CIV — Temperature Coefficients

10
50
100
200
400
800
1600

Cond. units

2S°-35°
0.0536

o . 0546
o 0548

0.0546
0.0553

^-0555
0.0538

3S°-A5°

o . 0448
o 0450

0.0429

o . 0446
o . 0446

0.0429

o . 0420

25°-35°
0.295

0.323
0.328

0.337
0.366
0.386
0.391

35°-45<'
0.380
0.412
0.405
0.429

0.459
0.464
0.468

Table CV — Molecular Conductivity of Ammonium Bromide in
a Mixture of 50 Per cent. Glycerol with Water at 25°,

35'

45'

10
50

100

200

400

800
1600

24-31

25 -74
26.62
27.01
27.86
30.20
32.58

111) 35°
32.58
34-54
35-61
36.12
37 32
40 -54
43 00

42.06

Table CVI — Temperature Coefficients

10
50
100
200
400
800
[600

250-35°
0.0340
0.0341
o . 0340

0.0334
0.0339
0.0342
0.0325

35°-45°
0.0291
0.0290
0.0282
0.0285
0.0278
0.0288
0.0275

25°-35°
0.827
0.880
0.899
O.9II
0.946
I 034
I .042

35°-45°
0.948
1.005
I .004
1.032
I 055

I. 179
I. 179

Table CVI I — Molecular Conductivity of Ammonium Bromide in
a Mixture of 25 Per cent. Glycerol with Water at 25°,

35'

10
50
100
200
400
800
1600

Hv 25°
61.45

liV

35°

76 . 93

/<t,45°
92.72
101.38
103.56
104.52
106.74
108.68
107.96

1 66

Guy and Jones
Table CVIII — Temperature Coefficients

Per cent.

Cond.

units

25°-35°

35°-^5°

25 "-35°

SS'-tS"

0249

0205

1.548

1-579

0254

0215

688

795

100

0255

0218

722

866

200

0251

0204

748

772

400

0245

0212

739

866

800

0258

0209

853

886

1600

0255

0205

824

836

Table CIX — Molecular Conductivity of Ammonium Bromide
in Water at 25°, J5°, 45°

li-v 2S°

^Lv3t

w45

122.7

148

173.2

131-4

158

185

100

133-5

159

187

200

135-3

163

191

400

138.2

[66

195

800

142.0

[70

199

1600

147.2

172

205

Table CX — Temperature

Coefficie

nts

Per cent

Cond.

units

25°-35°

35°-45''

25 "-35°

35°-45°

0.0212

0.0165

2.59

2.46

0.0202

0.0174

2.68

2.76

100

0.0199

0.0174

2.59

2-77

200

0 . 02 I I

0.0168

2.85

2.73

400

0.0205

0.0170

2.84

2.91

800

0.0202

O.OI7I

2.87

2.86

1600

0.0180

0.0183

2-57

3-^

'■7

Table CXI — Molecular Conductivity of Ammonium Bromide
in a Mixture of 7^ Per cent. Glycerol with Ethyl Alcohol

at 25°,

i5°

45'

m25°

/<v35°

,.^45°

1.32

2.25

3-55

2-55

100

2-59

200

2.77

400

2.62

800

2.85

1600

2.82

Conductivity and Viscosity in Mixed Solvents

[67

Table CXII — Temperature Coefficients

Per cent.

Cond.

units

25°-35°

35°-4S°

25°-35°

35 "-45°

0 . 0704

0.0577

0.093

0.130

0.0783

0.0558

0. 112

0. 142

100

0.0724

0.0582

0. 109

0.152

200

0.0721

0 . 0560

0. 116

0.154

400

0.0755

0.0610

0. 117

O.161

800

0.0721

0.0568

0. 120

0. 161

600

0.0699

0.0591

O.II5

0.168

Table CXIII — Molecular Conductivity of Ammoniu^n Bromide
in a Mixture of 50 Per cent. Glycerol with Ethyl Alcohol at
25°, 35°, 45°

100

200

400

800

1600

35°

43
28
76
90
06
48
59

Table CXIV — Temperature Coefficients

50
100
200
400
800
1600

Cond. units

25 "-SS"
0.0472

o . 0462

0.0516

0.0475
0.0495

o . 0490

0.0489

35 "-45°
0.0397
0.0396
0.0369
0.0401
0.0399
0.0393
0.0389

25°-35°
0.174

o. 198

0.231
0.222
0.234
0.246
0.249

35°-45°
0.216
0.249
0.246
0.282
0.282
0.297
0.292

Table CXV — Molecular Conductivity of Ammonium Bromide
in a Mixture of 25 Per cent. Glycerol with Ethyl Alcohol

25'',

10
50
100
200
400
800
1600

35 , 45'

liv 25°

8.51

10.54

II 45
12.50
12.94
13.92
14.38

35°

10.85

liv 45°

13 39
17.37
18.59
20.41
21.23
23.07
24.05

i68

Guy and Jones

Table CXVI — Temperature Coefficients

Per cent.

Cond.

units

25 "-35°

35°-45°

25 "-as"

35°-45°

0.0263

0.0237

0.234

0.254

0.0322

0.0247

0.340

0.343

100

0.0292

0.0255

0.336

0.378

200

0.0298

0.0259

0.373

0.418

400

0.0303

0.0258

0.393

0.436

800

0 . 0302

0.0273

0.420

0.495

600

0.0314

0.0271

0.453

0.514

Table CXVI I — Molecular Conductivity of Ammonium Bromide
in Ethyl Alcohol at 25°, 35°, 45°

V /jj/ 25 ° Hi) 35 ° Hi) i5°

21.6

30.9

100

35-5

200

39-8

400

47.2

800

510

1600

54-5

Table CXVIII-

-Temperature Coefficients

Per cent.

Cond.

units

25 "-35''

35°-45°

25°-35°

35°-45°

0.0156

0.0119

0.260

0.230

0.0149

0.0130

0.350

0.360

100

0.0157

0.0137

0.420

0.440

200

0.0165

0.0144

0.490

0.510

400

0.0160

0.0180

0.550

0.720

800

0.0179

0.0154

0.660

0.680

600

0.0178

0.0173

0.680

0.810

Table CXIX — Molecular Conductivity of Ammonium Bromide
in a Mixture of 7^ Per cent. Glycerol with Methyl Alcohol

25°, J5°

45°

;.^25<'

,.^35°

av^S"

2.50

4.00

6.04

2.70

100

2.87

200

2.94

400

2.94

800

3 05

1600

3.06

Conductivity and Viscosity in Mixed Solvents 169
Table CXX — Temperature Coefficients

Per cent.

Cond

units

V 25°-35°

35°-45°

25°-35°

35°-45»

10 0.0600

0.0510

0.150

0.204

50 0.0636

0.0563

0. 172

0.249

100 0.0637

0.0502

0.183

0.231

200 0.0629

O.O5II

0.185

0.244

400 0.0633

0.0501

0.186

0.243

800 0 . 0642

0 . 0499

0.196

0.252

1600 0.0631

0.0520

0.193

0.263

le CXX I — Molecular Conductivity of Ammonium Br or

in a Mixture of 50

Per cent. Glycerol with

Methyl Ale

at 25°, 35°, 45°

V fiv

25°

liv

35°

Mv*5°

10 9

16.86

50 10

1933

100 II

21.13

200 II

21.03

400 II

21.59

800 12

22.30

1600 12

22.90

Table CXXII-

-Temperatui

-e Coefficients

Per cent.

Cond

units

V 25 "-as"

35°-45°

25°-35°

35°-45°

10 0.0348

0.0293

0.337

0.383

50 0.0345

0 . 0308

0.379

0.455

100 0.0362

0.0368

O.4II

0.569

200 0 . 0368

0 . 0308

0.432

0.497

400 0.0372

0.0314

0.444

0.516

800 0.0398

0.0311

0.478

0.530

1600 0.0383

0.0310

0.485

0.542

Table CXXIII — Molecular Conductivity of Ammonium Bromide
in a Mixture of 25 Per cent. Glycerol with Methyl Alcohol
at 25°, 35°, 43°

,.^25°

Mv25°

/'^45°

26.0

31 4

37-3

100

200

400

800

600

170

Guy and Jones
Table CXXIV — Temperature Coefficients

Per cent.

Cond.

units

25°-35°

35°-A5°

25°-35''

35 •'-45°

0 . 0204

0.0187

0.540

0.590

0.0214

0.0202

0.640

0.740

100

0.0213

0.0194

0.700

0.760

200

0.0218

0.0209

0.740

0.860

400

0.0217

0.0202

0.760

0.860

800

0.0218

0.0198

0.790

0.880

1600

0.0220

0.0193

0.840

0.910

Table CXXV — Molecular Conductivity of Ammonium Bromide
in Methyl Alcohol at 25°, 55°, 45°

V nv 25°

/M,35° Mv^S"

10 59- I

4 730

50 74

9 91

100 79

3 99

200 83

I 105

400 89

5 III

800 90

102

2 117

1600 93

105

0 118

Table CXXVI-

-Teviperatur

? Coefficients

Per cent.

Cond. units •

V 25»-35°

35°-45°

25»-35° 35°-45''

10 0.0107

O.OII6

0 . 630 0 . 760

50 0.01 16

0.0106

0.870 0.880

100 0.0136

0.0102

1.080 0.920

200 0.0130

0.0123

1.080 I . 160

400 0.0103

0.0132

0 . 920 I . 300

800 0.0125

0.0148

I. 130 I. 510

1600 0.0124

0.0126

I. 160 1.330

Table CXXV II — Molecular Conductivity of Strontiuni Chloride
in Glycerol at 25°, 55°, 45°

w;25°

f.^35°

l^v^S°

0.322

0.664

1.252

0.403

0.840

558

100

0.426

0.900

650

200

0.452

0.958

777

400

0.475

1.008

866

800

0.483

1.037

934

1600

0.507

I 075

994

Conductivity and Viscosity in Mixed Solvents 171
Table CXXVIII — Temperature Coefficients

Per cent.

Cond.

units

25°-35°

3S°-4,5°

25°-35°

35°-A5°

0. 1062

0.0885

0 . 0342

0.0588

0 . 1084

0.0855

0.0437

0.0718

100

0. III2

0.0833

0.0474

0.0750

200

0. III8

0.0854

0 . 0506

0.0819

400

0. I 107

0.0851

0.0533

0.0858

800

O.I 150

0.0863

0.0554

0.0897

600

0. IIOI

0.0853

0.0568

0.0919

Table CXXIX — Molecular Conductivity of Strontium Chloride
in a Mixture of 7^ Per cent. Glycerol with Water at 25°,
35°, 45°

ftv

25°

liv

35°

A,„45°

5-85

1338

16.08

100

16.78

200

18.07

400

19.99

800

21.23

1600

22.46

Table CXXX-

—Temperatu

re Coefficients

Per cent.

Cond.

units

25 "-as"

35°-45°

25°-35°

35°-45°

0.0560

0 . 0465

0.328

0.425

0.0565

0 . 0476

0.392

0.526

100

0.0571

0 . 0466

0.416

0.533

200

0.0590

0.0467

0.458

0.573

400

0.0588

0 . 0465

0.502

0.636

800

0.0581

0.0458

0.535

0.667

1600

0.0581

0.0458

0565

0.709

Table CXXXI — Molecular Conductivity of Strontium Chloride
in a Mixture of 50 Per cent. Glycerol with Water at 25°,
35°, 45°

10
50
100
200
400
800
1600

35
36
38

19
84

42
42

03
84

35"

172

Guy and Jones

Table CXXXII — Temperature Coefficients

Per cent.

Cond.

units

25°-35°

35 0-45°

25°-35°

35°^5<»

0.0315

0.0337

1.030

1.292

0.0358

0.0315

I. 177

419

100

0.0379

0.0312

1-398

523

200

0.0373

0.0307

1-375

554

400

0.0356

0.0315

1-385

657

800

0.0354

0.0315

1.485

790

600

0.0378

00343

I .624

031

Table CXXXIII — Molecular Conductivity of Strontium
Chloride in a Mixture of 25 Per cent. Glycerol with Water
at 25°, 55°, 45°

W,25°

Mv35° ,1^45°

79-7

100

5 122.3

92.2

117

3 1442

100

97-7

122

9 152.2

200

102.3

129

3 159-8

400

103.8

133

0 163.0

800

107. 1

135

6 168.7

1600

109.2

137

3 170.2

Table CXXXIV-

—Temperature Coefficients

Per cent

Cond. units

sso-ss"

35°-45°

25°-3S° 35°-45°

0.0261

0.0216

2.08 2.18

0.0271

0.0229

2.51 2.69

100

0.0258

0.0239

2.52 2.93

200

0.0261

0.0235

2.70 3 05

400

0.0281

0.0224

2.92 3.00

800

0.0266

0.0243

2-85 3-31

1600

0.0258

0.0238

2.81 3.29

Table CXXXV — Molecular Conductivity of Strontium Chloride
in Water at 23°, 33°, 43°

tiv 25°

100
200
400

800

1600

175
199

207

215

224

230
235-9

fiv 35°
210.6
249.1

252.5
262 .7
274.8
279.0
285.6

liv 45°
247.8
285
299
310

323
332

342 -9

Conductivity and Viscosity in Mixed Solvents 175

Table CXXXVI — Temperature Coefficients

Per cent.

Cond.

units

25°-35°

35 "-^S"

25 "-35°

35°-45°

0.0201

0.0180

3-53

3 72

0.0250

0.0149

5 00

100

0.0214

0.0146

450

200

0.0219

O.OI8I

4-73

400

0.0224

O.OI7I

5 03

800

0.0209

0.0193

4.82

600

0.0210

0.0200

4-97

Table CXXXVII — Comparison
Ammonium Bromide from 25
and Water

of Temperature Coefficients of
^ to 35° in Mixtures of Glycerol

100 per cent.

75 per cent.

50 per cent.

25 per cent.

0 per cent

0.1032

0.0536

0.0340

0.0249

0.0212

O.IO5I

0.0546

0.0341

0.0254

0.0204

100

0.1075

0.0548

0.0340

0.0255

0.0199

200

0. 1080

0.0546

0.0334

0.0251

0.02II

400

0. 1069

0.0553

0.0339

0.0245

0.0205

800

0.1085

0.0555

0.0342

0.0258

0.0202

1600

O.I 106

0.0538

0.0325

0.0255

0.0180

Table CXXXVIII — Comparison of Temperature Coefficients of
Ammonium Bromide from 25° to 35° in Mixtures of Glycerol
and Ethyl Alcohol

100 per cent.

75 per cent.

50 per cent.

25 per cent.

0 per cent.

0.1032

0.0704

0.0472

0.0263

0.0156

O.IO5I

0.0783

0.0462

0.0322

0.0149

100

0.1075

0.0724

0.0516

0.0292

0.0157

200

0.1080

0.0721

0.0475

0.0298

0.0165

400

0. 1069

0.0755

0.0495

0.0303

0.0160

800

0. 1085

0.0721

0.0490

0.0302

0.0179

1600

0. 1 106

0.0699

0 . 0489

0.0314

0.0178

Table CXXXIX — Comparison of Temperature Coefficients of
Ammonium Bromide from 25° to 35° in Mixtures of Glycerol
and Methyl Alcohol

100 per cent.

75 per cent.

50 per cent.

25 per cent.

0 per cent.

0.1032

0 . 0600

0.0348

0.0204

0.0107

O.IO5I

0 . 0636

0.0345

0.0214

O.OI16

100

0. 1080

0.0637

0.0362

0.0213

0.0136

200

0.1075

0.0629

0 . 0368

0.0218

0.0130

400

0. 1069

0.0633

0.0372

0.0217

0.0103

800

0.1085

0 . 0642

0.0398

0.0218

0.0125

600

0. I 106

0.0631

0.0383

0.0220

0.0124

174

Guy and Jones

Table CXL — Comparison of Temperature Coefficients of
Sodium Nitrate from 25° to 35° in Mixtures of Glycerol
and Water

V 100 per cent. 75

per cent. 50 per cent. 25 per cent

;. 0 per cent.

lo 0.1033 0

.0529 0.0348 0.0253

0.0198

50 0.1046 0

.0561 0.0350 0.0244

0 . 0204

100 0.1096 0

.0549 0.0352 0.0272

0.0212

200 0.1070 0

.0541 0.0367 0.0267

0.02II

400 0.1058 0

•0534 0.0365 0.0264

0.0190

800 0.1058 0

.0538 0.0338 0.0254

0.0201

600 0.1062 0

.0531 0.0329 0.0254

0.0230

ble CXLI — Comparison of Temperature Coefficients of

Sodium Nitrate from 25° to 33° in Mixtures

of Glycerol

and Ethyl Alcohol

V 100 per cent.

75 per cent. 50 per cent. :

25 per cent.

10 0.1033

0.0736 0.0457

0.0284

50 0.1046

0.0701 0.0470

0.0288

100 0.1070

0.0742 0.0475

0.0292

200 0.1096

0.0739 0.0478

0.0290

400 0.1058

0.0721 0.0475

0.0305

800 0.1058

0.0746 0.0478

0.0288

1600 0.1062

0.0742 0.0487

0.0305

Table CXLII — Comparison of Temperature Coefficients of
Sodium Nitrate from 25° to 35° in Mixtures of Glycerol
. nd Methyl Alcohol

100 per cent.

75 per cent.

50 per cent.

25 per cent.

0.1033

0 . 0603

0.0363

0.0214

0. 1046

0.0652

0.0368

0.0223

100

0. 1070

0.0650

0.0377

0.0224

200

0. 1096

0.0714

0.0378

0.0220

400

0. 1058

0.0654

0.0385

0.0225

800

0. 1058

0.0613

0.0383

0.0218

1600

0. 1062

0.0658

0.0390

0.0230

Table CXLIII — Comparison of Temperature Coefficients of
Potassium Chloride from 25° to 35° in Mixtures of Glycerol
and Water

100 per cent.

75 per cent.

50 per cent.

25 per cent.

0 per cent.

0. 1006

0.0554

0.0341

0.0246

0.0188

0.1074

0.0556

0.0345

0.0258

0.0192

100

0. 1049

0.0549

0.0336

0.0244

0 . 0200

200

0. 1047

0.0548

0.0338

0.0253

0.0195

400

0.0948

0.0550

0.0344

0.0243

0.020I

800

0.0962

0.0553

0.0344

0.0238

0.0201

1600

0.0707

0.0558

0.0345

0.0253

0.0206

Conductivity and Viscosity in Mixed Solvents

Table CXLIV — Comparison of Temperature Coefficients of
Potassium Chloride from 25° to 55° in Mixtures of Glycerol
and Ethyl Alcohol

100 per cent.

75 per cent.

50 per cent.

25 per cent.

0. 1006

0 . 0694

0.0459

0.0281

0.1074

0.0717

0.0471

0.0297

100

0. 1049

0.0733

0 . 0500

0.0308

200

0. 1047

0.0723

0.0451

0 . 0306

400

0 . 0948

0.0719

0 . 0490

0.0352

800

0 . 0962

0.0733

0.0491

0.0320

1600

0.0707

0.0710

0 . 048 I

0.0319

Table CXLV — Comparison of Temperature Coefficients of
Potassium Chloride from 25° to J5° in Mixtures of Glycerol
and Methyl Alcohol

10
50
100
200
400
800
[600

100 per cent. 75 per cent. 50 per cent. 25 per cent.

O. ioo6
o. 1074
o. 1049
o. 1047
o . 0948

0.0962
0.0707

O . 06 I 2
0.0630

o . 0640

0.0632
0.0625
0.0639
0.0632

0.0369
0.0378
0.0374
0.0375
0.0376
0.0371
0.0365

0.0220
0.0218
0.0221
0.0230
0.0231
0.0227
0.0237

Table CXLV I — Comparison of Temperature Coefficients of Stron-
tium Chloride from 25° to 35° in Mixtures of Glycerol and
Water

V 100 per cent. 75 per cent.

50 per cent.

25 per cent.

0 per cent.

lo 0.1062 0.0560

0.0315

0.0261

0.0201

50 0.1084 0.0565

0.0358

0.0271

0.0250

100 O.III2 0.0571

0.0379

0.0258

0.0214

200 O.I 1 18 0.0590

0.0373

0.0261

0.0219

400 O.I 107 0.0588

0.0356

0.0281

0.0224

800 O.I 150 0.0581

0.0354

0.0266

0 . 0209

1600 o.iioi 0.0581

0.0378

0.0258

0.0210

The last figure in all tables of

"per cent,

," "temperature coef-

ficients" should be disregarded.

176

Guy and Jones

Table CXLVII — Viscosities and Fluidities of Solutions in Glycerol at 25°, 35°, 45^

Temp. coef.

Salt

1J25°

V35°

7? 45°

e 25°

fl35°

0 45°

25°-35°

35°-45°

KCl

6.362

2.836

1.399

0.1571

0.3527

0.7147

0.124

0.103

KBr

6.197

2.760

1.376

0.1613

0.3623

0.7264

0.124

0.101

KNO3

6.065

2.734

1.353

0.1648

0.3659

0.7391

0.122

0.099

NaCl

6.716

2.920

1.445

0.1613

0.3429

0.7143

0.124

0.106

NaBr

6.439

2.865

1.400

0.1553

0.3490

0.7143

0.124

0.106

Nal

6.303

2.822

1.409

0.1586

0.3543

0.7105

0.124

0.101

NaNOa

6.288

2.803

1.405

0.1590

0.3546

0.7117

0.123

0.101

NH4CI

6.142

2.741

1.360

0.1628

0.3649

0.7357

0.124

0.101

NHiBr

5.970

2.681

1.329

0.1672

0.3729

0.7524

0.123

0.102

NH4NO3

6.306

2.800

1.408

0.1587

0.3572

0.7097

0.124

0.099

BaCl2

7.447

3.288

1.626

0.1343

0.3041

0.6150

0.126

0.102

BaBra

7.100

3.199

1.571

0.1409

0.3126

0.6366

0.122

0.103

Ba(N03)2

7.212

3.182

1.571

0.1387

0.3143

0.6516

0.126

0.107

SrCl2

7.336

3.224

1.589

0.1363

0.3104

0.6291

0.127

0.103

SrBr2

7.337

3.219

1.574

0.1365

0.3107

0.6354

0.127

0.104

Sr(N03)2

7.640

3.335

1.640

0.1308

0.2998

0.6098

0.129

0.106

CaBr2

7.674

3.373

1.630

0.1303

0.2964

0.6135

0.127

0.106

Ca(N03)2

7.411

3.278

1.617

0.1350

0.3050

0.6184

0.125

0.103

Solvent

6.067

2.761

1.352

0.1648

0.3683

0.7396

0.124

0.101

Table CXLVIIJ— Viscosities and Fluidities of Solutions

in Glycerol at 55°. 65°, 75'

Temp

. coef.

Salt

J? 55°

ri 65°

J? 75'

e 55'

° 965°

e 15°

55°-65°

65°-75°

KCl

0.7435

0.4353 0.2648 1.345 2.297

3.776

0.071

0.064

KBr

0.7475

0.4353 0.2709 1.338 2.297

3.692

0.065

0.061

NaBr

0.7664

0.4439 0.2689 1.305 2.253

3.719

0.072

0.065

NH4CI

0,7366

0.4269 0.2613 1.357 2.342

3.827

0.072

0.063

NH4NO3

0.7284

0.4254 0.2618 1.373 2.351

3.819

0.071

0.062

C0CI2

0.8225

0.4762 0.2884 1.215 2.099

3.467

0.073

0.065

SrCl2

0.8536

0.4932 0.2981 1.172 2.028

3.355

0.073

0.065

Solvent

0.7415

0.4288 0.2620 1.350 2.331

3.817

0.072

0.063

Table CXLIX— Viscosities and Fluidities of Solutions ■

in Glycerol at 55°, 65°. 75°

J? 55°

)?65°

IJ 75° 9 55°

0 65°

6 75° "

Temp. coef.

Salt

55°-65°

65°-75°

KCl

0.6387

0.3781

0.2334 1.565

2.645

4.283

0.0689

0.0619

NH4CI

0.6457

0.3805

0.2318 1.548

2.628

4.313

0.0697

0.0641

NH4NO3

0.6251

0.3701

0.2291 1.599

2.702

4.365

0.0689

0.0616

Nal

0.6524

0.3827

0.2340 1.532

2.613

4.273

0.0705

0.0635

Ba(N03)2

0.7080

0.4159

0.2544 1.412

2.404

3.931

0.0702

0.0635

CoBr2
Solvent

0.7388 0.4292 0.2638 1.353 2.329 3.789 0.0721 0.0629
0.6370 0.3732 0.2309 1.569 2.678 4.329 0.0706 0.0616

Table CL — Viscosities and Fluidities of Solutions in Mixtures of Glycerol with Water

■°. 35°. 45°

■ Glycerol

Salt

)?25°

5J35°

rj45°

e 25°

e 35°

Temp. coef.

9 45° 25°-35° 35°-45°

A ■ ■'■■

' ' *■• KCl

6.362

2.836

1.399

0.1571

0.3527

0.7147 0.124 0.103

NHiBr

5.970

2.681

1.329

0.1672

0.3729

0.7524 0.123 0.102

NaNOs

6.288

2.803

1.405

0.1590

0.3546

0.7117 0.123 0.101

SrClj

7.336

3.224

1.589

0.1363

0.3104

0.6291 0.127 0.103

Solvent

6.067

2.761

1.352

0.1648

0.3683

0.7396 0.124 0.101

Conductivity and Viscosity in Mixed Solvents.

[77

In 75 Per cent. Glycerol with Water

KCl

0.3394

0.2003 0.1293

2.943

4.993

7.733

0,0698

0.0549

NH^Br

0.3278

0.1932 0.1249

3.035

5.176

8.008

0.0699

0.0547

NaNOa

0.3274

0.1947 0.1233

3.054

5.137

8.111

0.0682

0.0558

SrClz

0.3642

0.2179 0.1326

2.746

4.696

7.543

0.0713

0.0606

Solvent

0.3169

0.1884 0.1186

3.156

5.307

8.431

0.0681

0.0586

In 50 Per cent. Glycerol with Water

KCl

0.06481

0.04385 0.03187

15.27

22.82

31.37

0.0422

0.0347

NH^Br

0.06085

0.04251 0.03102

16.43

23.52

32.05

0.0431

0.0321

NaNOa

0.06333

0.04372 0.03216

15.79

22.87

31.10

0.0447

0 . 0363

SrClz

0.06607

0.04563 0.03335

15.13

21.90

29.99

0.0379

0.0369

Solvent

0.06109

0.04233 0.03114

16.37

23.63

32.10

0.0438

0.0358

In 25 Per cent. Glycerol with Water

KCl

0.02054

0.01546 0.01246

48.68

64.67

80.25

0.0328

0.0242

NH4Br

0 . 02046

0.01552 0.01226

48.88

64.50

81.56

0.0320

0.0264

NaNOa

0.02086

0.01556 0.01235

47.95

64.28

80.96

0 . 0340

0.0245

SrClz

0.02145

0.01614 0.01277

46.62

61.97

78,31

0.0329

0.0263

Solvent

0.01946

0.01466 0.01171

51.38
In Water

68.22

85.45

0.0327

0.0253

KCl

0.00902

0.00729 0.00608

110.8

137.0

164.6

0.0243

0.0201

NH4Br

0.00894

0.00722 0.00609

112.0

138.6

164.1

0.0246

0.0199

NaNOa

0.00903

0.00732 0.00608

110.8

136.6

164.4

0.0236

0.0202

SrCla

0.00927

0.00749 0.00628

107.9

133.5

159.4

0.0237

0.0194

Solvent

0.00891

0.00720 0.00598

112.2

138.9

167.2

0.0237

0.0204

Table CLI — Viscosities and Fluidities of Solutions in Mixtures of Glycerol with Ethyl

Alcohol at 25°, 35°, 45°

In 7S Per cent. Glycerol with Ethyl Alcohol

Temp. coef.

Salt

25°

35° 45°

25°

35°

45°

25°-35°35°-45°

KCl

1.123

0.5942 0.3387

0.8904

1.683

2.952

0.0890 0.0754

NH4Br

1.085

0.5762 0.3291

0.9214

1.736

3.039

0.0885 0.0751

NaNOa

1.171

0.6185 0.3509

0.8547

1 .635

2.850

0.0900 0.0762

Solvent

1.029

0.5404 0.3111

0,9720

1.830

3.215

0.0912 0.0759

In 50 Per cent. <

Glycerol with Ethyl Alcohol

KCl

0.2175

0.1377 0.08840

4.598

7.381

11.31

0.0605 0.0533

NH4Br

0.2163

0.1325 0.08668

4.731

7.550

11.54

0.0595 0.0528

NaNOa

0.2213

0.1360 0.08906

4.523

7.353

11.23

0.0620 0.0527

Solvent

0.2123

0.1351 0,08723

4.712

7.402

11.46

0.0600 0.0529

In 25 Per cent.

Glycerol u

■ilh Ethyl Alcohol

KCl

0.04473

0.03263 0.02487

22.36

30.66

40.21

0,0371 0.0311

NH4Br

0.04396

0.03227 0.02442

22.75

31.01

40.94

0.0369 0.0326

NaNOa

0.04464

0.03276 0.02481

22.40

30.52

40.31

0.0362 0.0320

Solvent

0.04184

0.03061 0.02303

23.90

32.77

43,42

0.0371 0.0324

Ethyl Alcohol

NH4Br

0.01216

0.009526 0.007979

' 86.13

105.1

125.3

0.0219 0.0193

Solvent 0.01068

0.008683 0.007292

: 93.70

115.2

137.7

0.0227 0.0191

178

Guy and Jones

-Viscosities and Fluidities of Solutions in Mixtures of Glycerol with Methyl
Alcohol at 25°, 35°, 45°
In 75 Per cent. Glycerol with Methyl Alcohol

Temp. coef.

Salt

25° 35° 45°

25°

35°

45°

25°-35°

35°-45°

KCl

0.6308 0.3512 0.2129

1.585

2.850

4.696

0.0797

0.0659

NHiBr

0.5999 0.3347 0.20U

1.666

2.987

4.973

0.0793

0.0665

NaNOs

0.6362 0.3590 0.2122

1 .572

2.786

4.713

0.0771

0.0689

Solvent

0.6242 0.3519 0.2087

1.609

2.842

4.792

0.0763

0.0681

In 50 Per cent. Glycerol with Methyl Alcohol

KCl

0.09521 0.06367 0.04474

10.51

15.70

22.35

0 . 0494

0.0423

NHiBr

0.09225 0.06300 0.04361

10.84

15.87

22.93

0.0464

0.0444

NaNOs

0.09717 0.06502 0.04574

10.29

15.74

21.87

0.0496

0.0436

Solvent

0.09657 0.06512 0.04446

10.35

15.35

22.50

0.0484

0.0468

In 25 Per cent. Glycerol with Methyl Alcohol

KCl

0.02083 0.01631 0.0131

48,02

61.32

76.31

0.0276

0.0244

NH4Br

0.02064 0.01610 0.0130

48.46

62.11

76.01

0.0261

0.0223

NaNOa

0.02098 0.01627 0.0130

47.75

61.48

76.46

0.0287

0.0243

Solvent

0.01886 0.01481 0.0U9

53.01

67.53

83.71

0.0274

0.0240

In Methyl Alcohol
NHiBr 0.006254 0.005410 0.004745 159.9 184.8 211.
.Solvent 0.005842 0.005066 0.004469 171.2 197.4 223.

0.0155 0.0143
0.0157 0.0139

Table CLIII — Table Showing Viscosities and Fluidities of Substances which were Found
to Lower the Viscosity of Pure Glycerol at 25°, 35°, and 45°

Temp. coef.

Salt

25°

35°

45°

25°

35°

45°

25 "-35°

35°-45°.

NaNOs

0.10

5.367

2.425

1.222

0.1863

0.4125

0.8186

0.121

0,100

NH4Br

O.iO

5.206

2.329

1.187

0.1929

0.4264

0.8423

0.121

0,098

NH^Br
NH4I

0.50
0.10

5.071
5.108

2.324
2.320

1.189
1.165

0.1972
0.1957

0.4302
0.4308

0.8409
0,8583

0.118
0.118

0.096
0.098

NH4I
RbBr

0.50
0.10

4.605
5.183

2.157
2.332

1.080
1 .176

0.2173
0.1975

0.4745
0.4288

0.9259
0.8502

0,118
0,117

0,096
0,098

RbBr

0.50

4.768

2.183

1.112

0.2098

0.4583

0.8998

0,118

0.096

Solvent

5.298

2.366

1.198

0.1888

0.4226

0.8347

0,118

0,097

DISCUSSION OF RESULTS

A rise in temperature causes an increase in conductivity,
which may be due to either or to both of the following causes :
First, an increase in the number of the ions present, and
second, an increase in the velocity of the ions. That the
number of the ions does not generally increase with rise in
temperature has been shown by direct measurement of the
degree of dissociation by means of the conductivity method.
This is in accord with the theory of Dutoit and Aston, ^ which

Conductivity and Viscosity in Mixed Solvents 179

makes the dissociating power of a solvent a function of its
own association. The degree of association of a solvent
has been shown by the method of Ramsay and Shields^ to
decrease with rise in temperature; hence, its power to dis-
sociate an electrolyte into its ions has been diminished. It is,
however, true that the theory of Dutoit and Aston is only an
approximation.

The increase in velocity of the ions with rise in temperature
must then be the one conditioning cause of the increase in
conductivity. This change in velocity of the ions may be
due to either or to both of the following causes : First, change
in the viscosity of the medium through which the ions move;
second, as Jones^ and his coworkers have shown, to the change
in complexity of the solvates which surround the ion.

In no other solvent is the change in conductivity with
change in temperature so pronounced as in the one which
chiefly concerns this investigation, viz., glycerol. The chief
cause of this change is largely the change in the viscosity of
the solution, while we believe that there is some evidence
brought out in this investigation that indicates the presence
of glycerolates.

Tables I to XXXVI, inclusive, give the molecular conductivi-
ties at 25°, 35° and 45° of all the electrolytes which we have
studied in pure glycerol as a solvent. It is seen that in all cases
the values for n^ are extremely small, but show, in general,
a regular increase, both with increased dilution and with rise
in temperature.

Associated with each table of conductivity is a table giving
the temperature coefficients of conductivity, both in per cent,
and in conductivity units. Since the latter show the actual
increase in conductivity per degree rise in temperature, a dis-
cussion of these data will bring out the most interesting points
of this part of the work.

Although the temperature coeflficients of conductivity,
when expressed in conductivity units, show, in general, a
regular increase with increased dilution, yet this is much

» Loc. cit.

- This Journal, 36, 445 (1906).

i8o Guy and Jones

more marked with ternary than with binary electrolytes.
This fact has been observed by Jones^ for aqueous solutions
in a discussion of the work of West.^

Results of the present investigation show that in glycerol
the temperature coefficients of conductivity of any given
substance, at high dilution, are larger than at lower dilution,
and that the relative increase is greater with salts of barium,
strontium, calcium, and cobalt than with salts of sodium,
potassium and ammonium. These facts may be explained in
terms of the theory of solvation. That solvation takes place
in aqueous solution has been shown beyond reasonable doubt by
Jones and his coworkers; and, indeed, Jones and Strong have
obtained abundant spectroscopic evidence for solvates in
glycerol as a solvent.

If there is solvation, then, according to the mass law, in
the more dilute solutions, where the amount of solvate per
ion is greatest, we should expect to find the most complex
solvates. Any change in temperature would produce the
greatest effect where the solvation was greatest, that is, in the
most dilute solutions. Again, this change in solvation should
be more apparent in those salts which have the greater power
of combining with the solvent, or, in the case of water, with
those salts that have the largest number of molecules of water
of crystallization.

We cannot, of course, say that salts of barium, strontium,
calcium, and cobalt possess a power of combining with glycerol
similar to that which they manifest towards water, but it
is not surprising to find solvation more marked with these
salts than with salts that have very slight hydrating power,
such as the salts of sodium, potassium and ammonium.

It is also true that salts of approximately the same hy-
drating power show, in glycerol, temperature coefficients of
the same order of magnitude.

The molecular conducti vnties at low dilutions in nearly
every case are smaller for ternary than for binary electrolytes,
while at higher dilutions the reverse is true without excep-

1 Loc. cit.

2 Tins Journal, 34, 357 (1905).

Conductivity and Viscosity in Mixed Solvents i8i

tion. This may be due to the fact that glycerol is only a
fair dissociating agent, resembling methyl and ethyl alcohols,
and has, at moderate concentrations, the power of producing
only two ions from a ternary electrolyte, or at least dissociating
a ternary electrolyte only to a moderate extent.

We should expect to find the ternary electrolytes yielding
more ions at higher dilutions, and, hence, showing a greater
molecular conductivity than binary electrolytes under the
same conditions. That this is true may be best shown by
comparing the molecular conductivities of several of the binary
and ternary electrolytes used.

Salt

MvlO

1^ 1600

KNO3

0.337

0.431

KBr

0.366

0.413

NaCl

0.328

0.395

BaBr^

0.330

0.530

Ba(N03)3

0.246

0.462

Ca(N03)3

0.283

0.472

SrCl^

0.322

0.507

In the above table the molecular conductivities of several
typical salts at 25° are compared at volumes 10 and 1600,
respectively. These data confirm the above statement, that
while at low dilutions a ternary electrolyte usually has the
smaller molecular conductivity, at higher dilutions the re-
verse is usually true.

Tables XXXVII to LVI give the molecular conductivities and
temperature coefficients of conductivity of all the salts studied
at 55°, 65° and 75°. The same general relations hold at these
temperatures as at the lower temperatures, viz., a regular
increase in conductivity with increased dilution and rise in
temperature; and a more marked increase, or a larger tem-
perature coefficient, with those salts which in aqueous solutions
possess the greatest power of hydration. The same reasoning
employed above for the lower temperatures is applicable
here.

Tables LVII to GXXXVI, inclusive, contain the data for the
molecular conductivity and temperature coefficients of con-
ductivity, expressed both in per cent, and in conductivity units,

l82

Guy and Jones

for potassium chloride, sodium nitrate, ammonium bromide, and
strontium chloride in the various mixtures of glycerol with
water, methyl alcohol, and ethyl alcohol. The results are
plotted in Figures I to X, inclusive.

t- ISO

Per cent. Glycerol
Fig. I — Coaductivity of Potassium Chloride in Glycerol-Water at 25 '

Conductivity aiid Viscosity in Mixed Solvents. 183

50 25 o

Per cent. Glycerol
Fig. II— Conductivity of Potassium Chloride in Glycerol-Etbyl Alcohol at 25"=

i84

Guy and Jones

Per cent. Glycerol
Fig. Ill — Conductivity of Potassium Chloride in Glycerol -Methyl Alcohol at 25 '

Conductivity and Viscosity in Mixed Solvents

t85

Per cent. Glycerol
Fig. IV — Conductivity of Sodium Nitrate i

Glycerol-Water at 25 '

1 86

Guy and Jones

T
50

Per cent. Glycerol
Fig. V-Conductivity of Sodium Nitrate in Glycerol-Ethyl Alcohol at 25 »

Conductivity and Viscosity in Mixed Solvents 187

Per cent. Glycerol
Fig. VI — Conductivity of Sodium Nitrate in Glycerol-Metbyl Alcohol at 25 »

x88

Guy and Jones

50 25 o

Per cent. Glycerol
Fig. VII — Conductivity of Ammonium Bromide in Glycerol-Water at 25*

Conductivity and Viscosity in Mixed Solvents 189

Per cent. Glycerol
Fig. VIII — Conductivity ofjAmmonium Bromide in Glycerol-Ethyl Alcohol at 25*

t90

Guy and Jones

00 75 so

Per cent. Glycerol^
Fig. IX — Conductivity of Ammonium Bromide in Glycerol-Methyl Alcohol at 25*

Conductivity and Viscosity in Mixed Solvents 191

50 35

Per cent. Glycerol
Fig- X— ConductiTity of Strontium Chloride in Glycerol- Water at 25 '

192 Guy and Jones

These curves show that the conductivities in such mixtures
do not follow the law of averages, but are always less. In
every case there is a marked sagging of the curves, but in
no instance was a minimum obtained. This deviation from the
law of averages has been explained by the work of Jones with
Lindsay and Murray, which has been discussed elsewhere in
this paper. When glycerol is mixed with water, or with either
of the alcohols, it is clear that the properties of the mixture
are not additive, the one solvent tending to lessen the asso-
ciation of the other; and, hence, their combined power of dis-
sociating electrolytes is less than would be expected if
there were no such lowering of each other's association.

Potassium chloride and sodium nitrate are nearly insoluble
in the alcohols, and yet curves expressing the conductivities
of these salts in mixtures of the alcohols with glycerol are
strikingly similar to those of ammonium bromide. This
seems to indicate that the deviation from the law of aver-
ages is due largely to the change in association of the glycerol.

Tables CXXXVII to CXLVI, inclusive, give a comparison of
the percentage temperature coefficients of conductivity from 25°
to 35° of all the salts we have studied in mixed solvents.
In pure glycerol these values are very large, being from ten
to eleven per cent, per degree rise in temperature. They
decrease very rapidly with the addition of either water or the
alcohols. The temperature coefficients also decrease very
rapidly with rise in temperature.

VISCOSITIES AND FLUIDITIES

Table CXLVII includes the viscosities and fluidities of the
eighteen electrolytes whose conductivities we have studied.
Measurements were made only with the tenth-normal solu-
tions, since, at higher dilutions, the difference in viscosity
between the solution and solvent is hardly large enough to
be detected, much less measured. In nearly every case the
viscosity of the solution is greater than that of the solvent.
Ammonium bromide was found to be an exception to this
rule, and will be discussed more fully. The temperature
coefficients of fluidity are very large and almost equal to the

Conditctivity and Viscosity in Mixed Solvents 193

temperature coefficients of conductivity. That the former
are larger than the latter is not surprising, since rise in tem-
perature would decrease the dissociation and thus decrease
the conductivity, which would, at least in part, offset the
increase in conductivity caused by increase in fluidity.

The ternary electrolytes show a much greater increase in
viscosity than the binary electrolytes. It will be recalled that
the salts which show the greatest increase in viscosity are
those in which the solvation seemed to be the greatest.

This increase in viscosity of the ternary over the binary
electrolytes may be due to several causes. There may be a
greater number of ions present, which, since the viscosity is a
function of the skin friction, would increase the viscosity;
or the molecules of the solvent, combined as solvates, may be
so attached to the molecule of the solute as to hinder its move-
ment. It is not supposed that in any case of solvation th^
molecules of the solvent are so held as to form a complex
chemical molecule, since this would, of course, decrease the
skin friction and thus lessen the viscosity of the solution.

The fact that solutions of ternary electrolytes show greater
viscosities than solutions of binary electrolytes may be a
conditioning factor in the small molecular conductivity shown
by them in the more concentrated solutions. It is, how-
ever, hardly possible that this coald account entirely for the
phenomenon, since there is probably less actual dissociation of
a ternary than of a binary electrolyte in the most concentrated
solutions.

It is probable, then, that the large viscosity of the ternary
electrolytes in glycerol is due to a summation of at least two
effects: The small atomic volumes of barium, strontium,
calcium and cobalt, and possibly to some factor caused by
solvation of the ions or molecules of the electrolytes, which, as
stated above, would probably be greater with the salts of these
metals than with salts of sodium, potassium and ammonium.

Tables CXLVIII and CXL,IX give the corresponding viscosity
data at 55°, 65° and 75°. The same general relations seem
to hold at the higher as at the lower temperatures. It was
found necessary to give these results in two tables, since

194 ^^y ^^^ Jones

the specific viscosity of the two samples of glycerol used in
this part of the work differed to some extent. There was only
a small difference in the specific conductivity of the two speci-
mens used. This difference in viscosity may be due to some
polymerization of the glycerol. The temperature coeffi-
cients of fluidity at these higher temperatures are very similar
to those of conductivity at the same temperatures.

From the data obtained, we are justified in concluding
that curves representing change in conductivity and change in
fluidity with rise in temperature are very similar to one an-
other. In a word, conductivity seems to follow fluidity quite
closely over the range of temperature from 25° to 75°.

The fact that glycerol has such a very large temperature
coefficient of viscosity presents the possibility of throwing
some light upon the relation between viscosity and reaction
velocity. It has long been felt that the viscosity of the medium
in which the reaction is taking place must be taken into con-
sideration, and if the velocity of some reaction could be fol-
lowed, using glycerol as a solvent, it is highly probable that
interesting results would be obtained. Glycerol, being such
an excellent solvent, seems well adapted to such work.

The viscosities and fluidities of solutions in the various
mixtures of glycerol with the alcohols and with water are
given in Tables CL to CLII, inclusive. Measurements
were made only with the tenth-normal solutions, since the
viscosities of the more dilute solutions differ very slightly
from that of the solvent in each case. Curves representing
the change in fluidity with concentration of glycerol are given
in Figure XI. These curves are, in general, strikingly analogous
to the curves representing the conductivities in the same
mixtures, though it is seen that the increase in fluidity is more
rapid than the increase in conductivity. The viscosities of
the solutions are in nearly every case greater than that of the
pure solvent.

NEGATIVE VISCOSITY COEFFICIENTS

One of the most interesting points brought out in this
investigation is the fact that certain salts have been found to

Conductivity and Viscosity in Mixed Solvents 195

CH3OH

H20

C2H6OH

r
50 25

Per cent. Glycerol
Fig. XI— Fluidity of Glycerol Mixtures at 25 '

196 Guy and Jones

lower the viscosity of glycerol. The fact that certain electro-
lytes have the power to lower the viscosity of water has been
known for some time, and the various theories put forward
to explain such phenomena have been discussed elsewhere in
this paper. Jones and Veazey^ were the first to offer an
apparently satisfactory explanation, the large atomic vol-
umes of the metals whose salts produced such a change being
the key to the phenomenon. The presence of elements with
large atomic volumes, as has been stated, would decrease,
the amount of skin friction in a given volume of solution,
and, thus, in terms of the theory of Thorpe and Rodger, would
decrease the viscosity. Jones and Veazey pointed out that
only salts of potassium, rubidium, and caesium produce
a decrease in the viscosity of water, and that these
salts do so in a direct ratio to their respective atomic vol-
umes. Schmidt^ had noted that the increase in viscosity
of solutions in p;lycerol over that of the pure solvent was
in an inverse ratio to the atomic volumes of the metals whose
salts he studied ; but in no case did he find a negative viscosity
coefiicient in pure ''glycerol.

The results showing negative viscosities in glycerol are
given in Table CLIII. From this table it can be seen that one-
tenth gram-molecute of rubidium bromide lowers the vis-
cosity of glycerol about two per cent., while one-half gram-
molecule lowers the viscosity of the solvent over eight per
cent.

This lowering of the viscosity of glycerol by a salt of rubidium
is analogous to the lowering of the viscosity of water produced
by the same salt. The explanation of this phenomenon
may be sought for in the theory of Jones and Veazey, as elab-
orated in the introduction to this paper, i. e., the large atomic
volume of rubidium.

Ammonium bromide and ammonium iodide produce the
same effect on the viscosity of glycerol, as is seen in Table
CLIII. It is clear that we can not speak of the atomic volume
of ammonium, since we know of it neither in the "atomic"
nor the "free" condition. It is, however, well known that

Conductivity and Viscosity in Mixed Solvents 197

ammonium is closely analogous chemically to potassium,
caesium and rubidium, and it is not surprising to find it ex-
hibiting the same physical behavior, such as the effect on the
viscosity of a solvent.

Summary of Conclusions Drawn from this Investigation

(i) Glycerol forms mixtures with water, ethyl alcohol, and
methyl alcohol whose properties are not additive. This is in
agreement with the work of Jones and Schmidt.

(2) Curv^es representing fluidity and conductivity are very-
similar to one another over the range of temperature from 25°
to 75°.

(3) Salts which have the highest power of solvation show
the greatest temperature coefficients of conductivity, and these
are greater in the more dilute solutions.

(4) In mixed solvents containing glycerol, with water, ethyl
and methyl alcohols, the curves representing conductivity and
fluidity are strikingly analogous.

(5) The molecular conductivities of ternary electrolytes in
glycerol at lew dilutions are usually smaller than those of
binary electrolytes under the same conditions, while at high
dilutions the reverse is generally true.

(6) While the majority of the salts studied increase the viscos-
ity of glycerol, yet certain salts of rubidium and ammonium
lower the viscosity of glycerol.

(7) Some evidence for the existence of glycerolates has been
given.

Work in glycerol as a solvent is now being continued in this
laboratory by two investigators and we intend to study very
thoroughly the physical chemistry of this solvent.

Johns Hopkins University,
May, 1911.

THE REACTION BETWEEN ORGANIC MAGNESIUM
COMPOUNDS AND CINNAMYLIDENE ESTERS

III. REACTIONS WITH THE ISOMERIC METHYL ESTERS OF CIN-
NAMYUDENACETIC ACID*
Bv Grace Potter Reynolds

In continuing the investigation of cinnamylidene esters
for the purpose of determining the influence of the second
ethylene linkage on the system C : C.C : O, the next ester in
the series to be studied was methyl cinnamylidenacetate.
The addition reactions previously investigated with esters of
svibstituted cinnamylidenacetic acids show that the charac-
ter of the products formed is determined by the substituent
in the a:-position to the carboxyl group. Methyl cinnamyl-
idenemalonate^ and ethyl a-cyancinnamylidenacetate^ react
readily with both aromatic and aliphatic magnesium compounds,
giving only 1,4-addition products, the strongly negative car-
boxyl and cyanogen groups in the a-position preventing the
replacement of the alkoxyl group of the esters. The addi-
tion of the Grignard reagent to the methyl ester of a-phenyl-
cinnamylidenacetic acid/ however, takes place only in boil-
ing ether and the product in this case is not an ester formed
by 1,4-addition but a ketone or tertiary alcohol formed by
replacement of the alkoxyl group and subsequent addition,
either 1,2 or 1,4, to the product first formed.

In the study of the esters of cinnamylidenacetic acid de-
scribed in this paper complications were anticipated because
of the presence in the esters of a mobile hydrogen atom in the
ft-position to the carboxymethyl group. With both aro-
matic and aliphatic magnesium compounds, complex sub-
stances were formed which interfered to a great extent with
the isolation of the primary products of the reactions. By
using a large excess of the Grignard product and keeping the re-

' Second paper. Rcimer ;\inl Reynolds: This Jouknai.. 40, 428.

- Rcimer: Ibid., 38, 227.

■■' >.r;ic;e<>d: Ihid.. 44, MX.

^ kciiiK-r and Keynnkls: Loc. cit.

Reaction between Organic Magnesium, Etc. 199

action mixture at a very low temperature, however, a min-
imum yield of complex product was obtained.

The isomeric esters react readily at — 10° with phenyl-
magnesium bromide; the final product in both cases is the
same substance, formed by replacement of alkoxyl with phenyl
and subsequent i ,4-addition to the ketone first formed. The
behavior of the two esters, however, is not identical. Ex-
amination of the products obtained when equal quantities of
the two esters were allowed to react under exactly similar
conditions with phenylmagnesium bromide indicated that
the esters show a difference in the ease with which the reac-
tion takes place. From the products obtained with the
ester of the stable acid one-third of the original ester was re-
covered unchanged, whereas all the ester of the alio acid had
reacted.

The course of these reactions may be compared with that«
of a-phenylcinnamylidenacetic acid ester and phenylmag-
nesium bromide :

QHsCH : CHCH : CHCOOCH3 + C^H^MgBr — ^

[C^H^CH : CHCH : CHCOC«HJ

[C„H,CH : CHCH : CHCOC^HJ + CgH^MgEr -^-

CeHjCH : CHCH(C«H5)CH2COC6H5

CeH-CH : CHCH : CCOOCH3 + C^HsMgEr — ^

(CeH,)

[CeHjCH : CHCH : C(C,H5)COCeH5]

[C«H,CH : CHCH : C(C6H5)COC,H.J + CeH^MgBr — ^

C,H,CH : CHCH(C«H,)CH(CeH,)C0C«H3

The character of the product formed is the same, as might be
expected, since the cinnamylidenacetic acid ester differs
from the a-phenyl acid ester only in that it has the more strongly
positive hydrogen atom in the a-position in place of a phenyl
group.

The reaction between the ester of the stable acid and ethyl-
magnesium bromide takes place very violently even at — 15°;
the sole primary product is a tertiary alcohol formed by re-
placement of methoxyl with ethyl and subsequent 1 ,2 -addi-
tion of ethylmagnesium bromide to the resulting ethyl ketone.

200 Reynolds

From the reaction between the ester of the stable acid
and benzylmagnesium bromide three primary products were
isolated. Their formation may be expressed thus:

QHgCHiCHCHiCHCOOCHg + C.H^CHjMgBr -^

[QHjCH : CHCH : CHCOCH^C.HJ

This unsaturated ketone at once reacts with benzylmagne-
sium bromide to form both 1,2- and 1,4-addition products:

[CeHsCH : CHCH : CHCOCH^CeHJ + CeHgCH^MgEr — >

/CHjCgHj
C^HsCH : CHCH : CHC(-OH

\CHX,H3

[CeHjCH : CHCH : CHCOCH^CgHJ + CsH^CH^MgEr -^

CgHgCH : CHCH(CH2C6H5)CH2COCH,C6H5

The third product is an ester formed by 1,4-addition to the
original ester:

CeHjCH : CHCH : CHCOOCH3 + CeHgCH^MgBr -^

CeHjCH : CHCH(CH2CeH5)CH2COOCH3

When 1,4-addition was established in this case, the reac-
tions with both phenylmagnesium bromide and ethylmag-
nesium bromide were repeated with greatest care in an attempt
to find any trace of 1,4-addition product that might have
been lost in earlier experiments. If any 1,4-ester was formed
in either case, it could not be isolated by the method of separa-
tion adopted.

EXPERIMENTAL

For the preparation of the large quantities of cinnamyliden-
acetic acid needed in this work, the Doebner method,* as modi-
fied by Hinrichsen,^ was first used, but was abandoned be-
cause a poor yield of cinnamylidenacetic acid was invariably
obtained. The method of Liebermann^ was adopted; it was
fovmd possible, however, to carry out the reaction with much
larger quantities of material than were recommended by
lyiebermann. The procedure was as follows: To 64 grams

' Her. d. chem. Ges., 33, 2140.

- Ann Chem. (I.iebig). 336, 197.

- Her. a ehcm. Cius., 28, 1441,

Reaction between Organic Magnesium, Etc. 201

of quinoline in a tall beaker 100 grams of cinnamylidene-
malonic acid^ were added quickly and with vigorous stirring.
When the quinoline salt had formed, the beaker was heated
at 170° in an oil bath for about half an hour until evolution
of carbon dioxide ceased. The oily product was poured,
while still warm, into iced hydrochloric acid; the crude acid
separated in the solid condition on standing. It was dis-
solved in sodium carbonate and the sodium salt of cinnamyl-
idenacetic acid was purified by extracting the solution with
ether. The acid was precipitated from a dilute solution of
the purified salt. When the slightly yellow product was re-
crystallized from benzene, the first crop of crystals was the
alio acid; the mother liquor contained a mixture of the alio
and stable acids. A 50 per cent, yield of pure acids was ob-
tained. After several recrystallizations from benzene both
of the isomers were absolutely colorless.

The stable acid used for the preparation of the ester was ob-
tained by the isomerization of the alio acid in the sunlight.
The mixture of the two isomers invariably obtained in the
preparation of the alio acid was also used for this reaction.
It was found very difficult to separate the stable acid in the
pure condition, however, if the large quantity of iodine recom-
mended by Liebermann was used for the isomerization, but
the acid could be separated easily in pure condition by using
a small quantity of iodine as catalyst. Accordingly, for
the reaction, 22 grams of pure alio acid were dissolved in 275
cc. of benzene, to which o . 25 gram of iodine was added. Isom-
erization took place almost instantaneously in the bright
sunlight; the trace of iodine was volatilized during the process
of filtering the acid.

The ester of the alio acid was prepared by heating the acid
on a water bath for one hour with 4 parts by weight of a 3
per cent, solution of hydrochloric acid in methyl alcohol.
The oily product was poured into iced sodium carbonate, the
solution extracted with ether and the ethereal extract care-
fully dried over calcium chloride. The lemon-yellow liquid

' Ber. d. chem. Ges., 28, 1439. Ann. Chem. (I.iebig), 306, 2S2.

202 Reynolds

left after removal of the ether was used without further
purification.

Analysis :

O.I 66 1 gram substance gave 0.4666 gram CO2 and 0.1015
gram HjO.

Calctilated for

Cl2H,202

Found

76.59

76.61

6.38

6.78

The ester is a liquid at ordinary temperature, and solidifies
at about — 15° to a white crystalline mass. It is partially
changed into the ester of the stable acid by distillation under
diminished pressure.

The ester of the stable acid was prepared by heating the
acid on a water bath for i . 5 hours with 6 parts, by weight,
of a 3 per cent, solution of hydrochloric acid in methyl alco-
hol. On cooling, the alcohol solution deposited a quantita-
tive yield of solid ester.

Since both of the esters of cinnamylidenacetic acid decom-
pose on standing, in the Grignard reaction with the ester of
the alio acid it was necessary to use freshly prepared ma-
terial; and the ester of the stable acid was always redistilled
or recrystallized before adding it to the organic magnesium
compounds.

The first reactions between the esters and the Grignard
reagent were carried out in an atmosphere of carefully dried
hydrogen. Subsequent work, however, showed that this
extreme precaution was not necessary.

Reaction with Phenyltnagnesium Bromide

To a solution of phenylmagnesium bromide containing
2 . 5 molecules of magnesium to one molecule of ester, 28 grams
of the ester of the stable acid were added. The orange-colored
Grignard product was decomposed with iced hydrochloric
acid and the solution treated in the usual way. From the
oily product left after removal of the ether, two fractions
were separated by distillation under diminished pressure;
the lower-boiling fraction was identified as the original ester —

Reaction between Organic Magnesium, Etc. 203

9 grams were recovered unchanged. The higher-boiling frac-
tion, distilling at 285°-290° (18 mm.), solidified in contact
with alcohol; yield, 25.5 grams. The solid crystallizes from
alcohol in fine, white needles melting at 93°.

Analysis :

0.1530 gram substance gave 0.4950 gram COj and 0.0876
gram 11,0 .

Calculated for

C23H20O Found

C 88.46 88.23

H 6.41 6.36

The substance is /?-phenyl-^-benzalpropylphenyl ketone,
previously obtained by the action of phenylmagnesium bro-
mide on cinnamylidenacetophenone }

QHsCH : CHCH : CHCOCeHg + QHgMgBr — >

QHsCH : CHCHCC^HJCHaCOCeHs

1 ,1 ,3,5-Tetraphenyl-4-pentene-J-ol,

CgHsCH : CHCHCCeHJCHjCCCeHJ^OH.— Thirteen and one-
half grams of solid /?-phenyl-;'-benzalpropylphenyl ketone were
added slowly to a solution of phenylmagnesium bromide con-
taining 2 molecules of reagent to one molecule of ketone.
After allowing the reaction mixture to stand at room tempera-
ture for an hour, it was decomposed with iced hydrochloric
acid. The solid which appeared at once in the ether layer
was separated and washed with ether. A quantitative yield
of tertiary alcohol was obtained. It was purified by recrys-
tallization from alcohol, from which it separated in fine, white
needles melting at 134°. The substance is very soluble in ace-
tone, chloroform and benzene, less soluble in ether, methyl
and ethyl alcohols.

Analysis :

0.1858 gram substance gave 0.6078 gram CO2 and 0.1093
gram Hfi.

Calculated for

C29H26O Found

C 89.23 89.21

H 6.66 6.53

' Kohler: Ber. d. chem. Ges., 38, 1204.

204 Reynolds

The Grignard reaction showed that the substance is a
hydroxyl compound. When the solid was added to a solu-
tion of ethylmagnesium bromide, evolution of gas took place
and on decomposition of the magnesium derivatives, all the
original material was recovered unchanged.

Attempts to eliminate one molecule of water from this
tertiary alcohol were unsuccessful.

The structure of the tertiary alcohol was established by
oxidation. Four grams of the solid were dissolved in acetone
and treated with finely powdered potassium permanganate
at room temperature. The mixture was allowed to stand
overnight and the excess of permanganate was then decol-
orized with an acetone solution of sulphur dioxide. The
oxides of manganese were separated by filtering with suction
and washed with hot acetone. The acetone solution deposited
a crystalline solid, which was carefully washed with cold alcohol
and purified by recrystallization from the same solvent. It
separated in long, iridescent needles melting at 157°; 1.5
grams were obtained from this solution.

Analysis :

0.1549 gram substance gave 0.4752 gram COg and 0.0800
gram Hp.

Calculated for

Found

83.66

5-73

Subsequent extraction of the oxides of manganese with hot
water gave a solution from which one gram of benzoic acid
was separated. The oxides of manganese were finally dis-
solved in sulphuric acid to which sulphur dioxide was added,
and the solution extracted with ether. On evaporation of
the ether 0.6 gram of the same solid that separated from the
acetone solution was obtained. Analysis and reactions showed
that this substance is the lactone of 7--hydroxytriphenylbutyric
acid, formed by loss of one molecule of water from the /--hydroxy
acid which is a primary product of oxidation.

Calculated for

C22H.8O2

84.07

5-73

Reaction between Organic Magnesium, Etc. 205

C.H^CH : CHCH(C,H5)CH2C^Oh' + 4O — >

C^H^COOH + (C,H5)2C— CHaCHCC^Hs)— CO

I 1 + H,0

The lactone is very soluble in acetone, chloroform and
benzene, much less soluble in ether, methyl and ethyl alcohol.
It does not dissolve in sodium carbonate and dissolves readily
in potassium hydroxide. The lactone is precipitated un-
changed from its solution in potassium hydroxide, even when
the reaction is carried out in a freezing mixture.

With a view of determining whether the isomeric methyl
esters of cinnamylidenacetic acid show a difference in their
action with organic magnesium compounds, reactions with
the two esters and phenylmagnesium bromide were carried
out at the same time under conditions as nearly alike as pos-
sible. A much larger excess of Grignard reagent was prepared
than in the previous reaction with phenylmagnesium bro-
mide, 4 molecules of reagent to one molecule of ester being
used. There was no apparent difiference in the action of
the two esters, both magnesium derivatives soon acquiring
a bright, orange color. The Grignard products were decom-
posed and treated in the usual way. As the ether extract
from the two solutions evaporated, both deposited crystals
of the same ketone, ^-phenyl-^'-benzalpropylphenyl ketone.
The product from the reaction with the ester of the alio acid,
however, crystallized with greater ease and gave a much
larger crop of ketone crystals. From 10 grams of this ester
10 grams of ketone were obtained — a 60 per cent, yield; no
other product could be isolated.

Portions of the products from the two reactions were saponi-
fied with alcoholic potassium hydroxide. No acid was found
in the solution obtained from the product of the reaction with
the ester of the alio acid. An acid, however, was isolated
from the other solution. Accordingly, the whole of the prod-
uct from the reaction with the ester of the stable acid was

2o6 Reynolds

saponified and stable cinnamylidenacetic acid, equivalent to
one-third of the ester used, was separated. No trace of any
other acid could be found. From the 6 grams of the ester
of the stable acid that had reacted, 7.5 grams of /?-phenyl-;--
benzalpropylphenyl ketone were obtained — a 75 per cent,
yield.

Reaction with Benzylmagnesium Bromide

An ether solution of 14 grams of the ester of the stable acid
was allowed to react with a solution of benzylmagnesium bro-
mide in the same solvent. After a few grams of the ester
had been added, the magnesium mixture assumed a dark
green color, and finally when 14 grams had been added an
oil began to separate from the ether. When this first ap-
peared the reaction was stopped and the Grignard product was
decomposed at once with iced hydrochloric acid; the solution
was extracted and the ether extract washed and dried in the
usual way. The liquid left after evaporation of the ether
was treated with alcoholic potassium hydroxide, and after
allowing the mixture to stand overnight, the alcohol was dis-
tilled off and the residue treated with water.

The portion of the product insoluble in water was extracted
with ether, the ethereal solution washed and dried over cal-
cium chloride, and after removal of the solvent the yellow
liquid was fractioned under diminished pressure; 2 grams
of a very mobile, lemon-yellow liquid distilled at 200° (10
mm.) and 12.5 grams of a viscous, lemon-yellow liquid dis-
tilled at 265 (15 mm.). There were several grams of residue
left in the Claisen flask. The higher boiling fraction is
/3-benzyl-^-benzalpropylbenzyl ketone,

CeH,CH : CHCH(CH2CeH5)CH2COCH2C6H5

Analysis :

0.1566 gram substance gave 0.5071 gram CO2 and 0.0969
gram HjO.

Calculated for

CzsHjiO

Found

88.23

88.31

7.06

6.87

Reaction between Organic Magnesium, Etc. 207

Dibrom-^-henzyl-y-benzalpropylhenzyl Ketone,
CeH5CHBrCHBrCH(CH2CeH6)CH2COCH2CeH5.— Two grams of
the ketone were dissolved in carbon bisulphide and the solu-
tion cooled in a freezing mixture and treated with a solution
of bromine in the same solvent. The calculated quantity of
bromine was instantly decolorized. Evaporation of the
solvent in a current of air deposited a white solid, which was
purified by recrystallization from a mixture of ligroin and
chloroform. It melts at 165°. 5 and is very soluble in ace-
tone and chloroform, not so soluble in ether, ligroin and ethyl
alcohol.

Analysis :

0.1393 gram substance gave 0.3063 gram CO2 and 0.0608
gram H2O.

Calculated for

C25H240Br2

Found

60.00

59 96

4.80

4.84

Sufficient material was accumulated for purification of "the
lower-boiling fraction. This was a,a-dibenzyl-^-benzalcro-
tonyl alcohol,

C9H5CH : CHCH : CHC(CH2CeH5),OH

Analysis :

0.1496 gram substance gave 0.4826 gram CO2 and 0.0932
gram HjO.

Calculated for

C25H24O Found

C 88.23 87.98

H 7.06 6.92

The Grignard reaction showed that the substance is a hy-
droxyl compound. There was a steady evolution of gas
when the liquid was treated with ethylmagnesium bromide.

Tetrabrom-a,a-dihenzyl-d'henzalcrotonyl Alcohol,
CeHsCHBrCHBrCHBrCHBrCCCHjCeHJjOH.— Two grams of
tertiary alcohol were dissolved in chloroform and treated with
a chloroform solution of bromine as long as instantaneous
decolorization took place. As the solvent evaporated, fine

2o8 Reynolds

needles appeared, which were filtered and washed with hot
methyl alcohol. The product thus treated melts at 227°
with evolution of gas and charring. It was impossible to
purify it further by recrystallization, for the melting point
of the solid was lowered ten degrees by one recrystallization
from a mixture of cold chloroform and methyl alcohol. The
carefully washed solid was therefore used for analyses. It
is very soluble in acetone, chloroform and ethyl alcohol, less
soluble in ether and ligroin.
Analysis :

I. 0.1613 gram substance gave 0.2772 gram CO2 and
0.0538 gram H2O.

II. o. 1538 gram substance gave o. 2630 gram CO, and 0.0519
gram HjO.

Calculated for

Found

CasHsiOBr*

45-45

46.86

46.63

3-63

3 70

3-74

^-Benzyl-y-benzalbutyric A cid,
CeH^CH : CHCH(CH2CeH5)CH2COOH. — On acidifying the
aqueous solution obtained after saponification of the original
product with alcoholic potassium hydroxide, 1.2 grams of
an oily acid separated which solidified at once in contact with
alcohol. It was recrystallized from a mixture of acetone
and ligroin, from which it separated in heavy, iridescent
plates melting at 144°.

Analysis :

0.1500 gram substance gave 0.4447 gram CO, and 0.0910
gram H2O.

Calculated for

CisHisOz

Found

81.20

80.85

6.76

6.74

The acid is very soluble in acetone, methyl and ethyl alco-
hol, not so soluble in ligroin and chloroform.

Reaction between Organic Magnesium, Etc. 209

Molecular weight determined in boiling acetone (K = 1710)-

Solvent
Grams

Substance
Gram

Elevation of Molecular
boiling point weight

41.8

0. I 198
0 . 4605

0.019 258
0.071 265

0.7219
0.9874

0.109 271
O.I5I 267

Average 265
Calculated for CigHj.Og 266

The acid was dissolved in a solution of sodium carbonate and
oxidized with potassium permanganate; benzoic acid and
benzylsuccinic acid were the oxidation products isolated.
The benzylsuccinic acid was identified by comparison with a
specimen on hand. The reaction, represented by the follow-
ing equation, establishes the structure of the substance:

QHsCH : CHCH(CH,C6H5)CH2COOH + 40 =

C,h;COOH + C,H5CH3CH(COOH)CH2COOH

Methyl ^-Benzyl-y-henzalbxUyrate,
CgHsCH : CHCH(CH2CeH5)CH2COOCH3.— The ester was pre-
pared by treating the acid with a saturated solution of hydro-
chloric acid in methyl alcohol. After allowing the mixture
to stand overnight, it was poured upon iced sodium carbonate
and the solution extracted with ether. The solid which
separated on evaporation of the ether was purified by recrys-
tallization from methyl alcohol. It melts at 66°.

Analysis :

0.1554 gram substance gave 0.4640 gram CO2 and 0.0984
gram HjO.

Calculated for

C19H20O2 Found

C 81.42 Si. 43

H 7.14 703

Reaction with Ethylmagnesium Bromide

An ethereal solution of 10 grams of the ester of the stable
acid was added very slowly to an ethereal solution of ethyl-
magnesium bromide containing 4 molecules of magnesium
to one molecule of ester. A very vigorous reaction took place
even at — 10° and the magnesium mixture soon acquired a

2IO Reynolds

brilliant green color, indicating the formation of secondary-
products. When all the ester had reacted, the Grignard
product was decomposed with iced hydrochloric acid, the solu-
tion extracted with ether, the ether extract washed with
sodium carbonate and dried over calcium chloride. On evap-
oration of the solvent a yellow liquid was obtained. This
was treated with alcoholic potassium hydroxide and after
allowing the mixture to stand several hours, the alcohol was
distilled off and the residue treated with water. The por-
tion insoluble in water was extracted with ether and the ethe-
real extract washed and dried. After removal of the solvent
the liquid was distilled under reduced pressure. One gram of
pale yellow liquid distilled at 169° (10 mm.), leaving several
grams of dark residue in the Claisen flask.

On acidifying the aqueous solution, 3 . 5 grams of acid
separated; it was purified by recrystallization from a mixture
of acetone and ligroin, from which it separated in fine, color-
less needles melting at 179°. Analyses indicated that it was
a secondary product.

As the treatment with potassium hydroxide might have
been too vigorous, the reaction was repeated and the product
was not saponified but was at once distilled under diminished
pressure. Two fractions were separated; the lower-boiling
one was a pale yellow, mobile liquid distilling at 169° (10
mm.). It is a,a-diethyl-^-benzalcrotonyl alcohol, formed by
replacement of the methoxyl group of the ester and subse-
quent 1, 2 -addition of a second molecule of ethylmagnesium
bromide to the unsaturated ketone thus formed.

CeHjCH : CHCH : CHCOOCH3 + C^HgMgBr — ^

[CeH^CH : CHCH : CHCOC^HJ
[C^HgCH : CHCH : CHCOC^HJ -f- C^H^MgEr — >

CjHgCH : CHCH : CHC(C2H5)20H
Analysis :

0.1634 gram substance gave 0.4979 gram CO2 and 0.1324
gram H2O.

Calculated for

C16H20O Found

C 83.33 83 10

H 9.25 9.00

Reaction between Organic Magnesium, Etc. 211

The liquid was sho\vn to be a hydroxyl compound by treat-
ment with ethylmagnesium bromide. There was a steady
evolution of gas and on decomposition of the magnesium
derivative the original material was entirely recovered.

Several attempts were made to eliminate one molecule of
water from this tertiary alcohol. Ten grams of the liquid
were heated with 3 molecules of acetic anhydride for three
hours over a free flame. The tertiary alcohol was recovered
unchanged from the solution. No reaction, moreover, took
place when the liquid was heated for four hours at 215° in a
sealed tube with excess of acetic anhydride, and attempts
to split off water by the use of oxalic acid were likewise un-
successful.

The higher-boiling fraction, separated by distillation under
diminished pressure, was a heavy, viscous, yellow liquid
boiling at 2 78°-285° (10 mm.). After standing overnight
the liquid deposited fine, white needles melting at 136°. The
solid was purified by recrystallization from ethyl alcohol,
after carefully washing it free from the yellow liquid with
ether. Analyses indicated that this solid is also a secondary
product; it is an ester, from which, by saponification with
potassium hydroxide, an acid melting at 230° was obtained.
It was not studied further and its structure was not estab-
lished. The higher-boiling fraction is doubtless a mixture
of secondary products, which would be anticipated in the
Grignard reaction with methyl cinnamylidenacetate.^

In the reaction between this ester and ethylmagnesium
bromide, the proportion of a:,a-diethyl-^-benzalcrotonyl alco-
hol and the higher-boiling products is dependent upon the
conditions under which the reaction is allowed to take place.
If the magnesium mixture is kept at — io° and the ester is
added very slowly, a 50-55 per cent, yield of the tertiary
alcohol may be separated and a comparatively small yield
of secondary products. If, however, the temperature of
the reaction mixture is allowed to rise a much larger propor-
tion of secondary products is obtained.

Chemical Laboratory
Barnard College
May. 19U
• Kohler: This Journal, 84, 568.

REVIEWS.

Trattato di Chimica Inorganica Generai:,e e applicata alV In-
DUSTRiA. DoTT. Ettore Moi,inari, Professore di Chimica indus-
triale alia Society d'Incoraggiamento d' Arti e Mestieri e di mer-
ceologia all 'University Commerciale Luigi Bocconi in Milano. 280
Incisioni nel Testo, 1 Tavola in Cromolitografia e 2 Tavole in Fototipia.
Terza edizione riveduta e ampliata. Milano: Ulrico Hoepli. 1911.
pp. xvii + 923. Price, L. 16.

To this, the first volume of a work, of which the second vol-
ume dealing with organic chemistry was reviewed last year/
the same unqualified praise is due for the surprising amount
of useful information and sound science compressed into a
limited space. Thus, for example, under "air," in thirty-two
pages, is given a comprehensive account of liquid-air machines
and their practical operation, with data as to cost, power re-
quired, and efiiciency, etc., Linde's, Claude's and Pictet's
systems, fractionation of the liquid and utilization of the oxy-
gen and nitrogen, methods of fixing the latter as practiced in
Norway, America, Switzerland, Germany, Italy, France and
England, with a summary of recent inventions and patents,
action of calcium cyanamide as a fertilizer, besides records of
atmospheric pressure and a brief account of the properties
of the rare gases. Commercial objects and practical require-
ments are kept constantly in view; current prices and locali-
ties and quantity of production are quoted throughout, and
the information is brought down to about 19 10. There are
numerous clear illustrations of apparatus and an index. For
the chemist who is interested in practical applications the
book can be highly recommended. b.b. turner.

An Introduction to Thermodynamics for Engineering Students.
ByJohn Mir.LS. Ginn & Co. pp.136. Price, |2. 00.

In studying any subject with a mathematical basis two
phases of the mental process involved may be distinguished,
as distinct as the ability to read and speak a foreign language.
To comprehend the fundamental principles, when their demon-
stration is presented, and the power to reproduce that demon-
stration is rightly placed first; experience shows, however,
that real mastery of the subject is only reached when the
student can readily apply those principles to concrete instances,
with definite data. This is peculiarly the case with such
an abstruse subject as thermodynamics, and its applications
to chemical problems, so that while the above-named book,
as its title implies, deals solely with questions of interest to

1 This Journal, 44, 202.

Reviews 213

engineers, expansion of gases, properties of steam and water
flow of gases, etc., it would greatly advantage most students
of physical chemistry to work through a number of the simple
problems with which the book abounds. The text is usually
well written and the explanations clear, while many tables
and diagrams give the facts a reality and readily intelligible
appearance. The use of Fahrenheit temperatures, British
thermal units, foot pounds, etc., which is justified by the pur-
pose of the book, will only provide a little extra arithmetical
exercise for the chemist. The author is to be heartily con-
gratulated on the production of a useful little book.

B. B. Turner.

Introduction to General Chemistry. By John Tappan Stoddard,
Ph.D., Professor of Chemistry in Smith College. New York : The
Macmillan Company. 1910. pp. xviii -|- 432. Price, $1.60.

The first agreeable thing about this new text is the size of
the volume. It contains httle more than one-half the num-
ber of words contained in similar texts. And yet, the sub-
ject matter enclosed by its covers is sufficient with a good
teacher to keep a college class very busy for a year. Of course
one is curious to know how the author can have accomplished
his purpose within such limits. It is stated that the book
"is designed as an introduction to advanced study, providing
a foundation which shall be both broad and thorough." Al-
though we might expect from such a program a crowding of
the page, ease and dignity seem to stamp the work. All the
essentials are introduced and yet there is plenty of room.
In explanation of how this is possible and how the high purpose
of the author has been successfully carried out, we note first
that great importance is attached to all those generalizations
which, after taking root in the history of the science, have
surrounded themselves with a hedge of succinct and very
convenient phraseology. Then, when the newer and less
universally accepted theories, physical or mathematical,
are introduced, they are presented simply, with no defense,
and with the chemical bearing exclusively emphasized. The
size of the book is affected somewhat by the omission of all
pictures. The author is strongly of the opinion that illustra-
tions are unnecessary and at times misleading and he has
bravely decided to omit them all.

As reflecting the attitude of the teacher, we read in the
preface: "Above all, I have tried to help the student to enter
into the spirit of chemistry and to acquire the scientific point
of view." One path towards this purpose is by free use of
the quantitative method; and there are many references in

214 Reviews

the text to Stoddard's own little book of quantitative ex-
periments for beginners which was reviewed in This Journal.*
Thus the important laws of chemical combination are pre-
sented from the student's experimental results and the quan-
titative relations are emphasized; but not without the data
for illustration. Dr. Minot has just said at Minneapolis,
and well said : ' ' Mathematics cannot give any comprehensive

expression of complex relations For our accuracy it

is necessary often to have a number of data presented to our
consciousness at the same time." So, for the student's as-
similation to proceed with an accuracy of which mathematics
is incapable, the statement of the quantitative law must always
be accompanied by the illustrative data.

Again, the plan and spirit of the book are reflected by the
statement: "So far as is possible, the progress is from the
familiar to the unfamiliar, from the known to the unknown."
This has been a guiding thought with the author and has
strongly regulated the choice of material and the order of
its presentation. The advantage of following the periodic
system is considered minor to that of proceeding from the
familiar and important. This gives an arrangement some-
what unfamiliar to the older chemist. Sulphuric acid is made
the basis for the study of all acids and is introduced very
early; and in general the substance, element or compound
which is most important is for that reason presented first.

The language is clear, and it is convincing because it is
simple. A fev/ errors exist, but only such as may be easily
corrected in the future editions. The mechanical construc-
tion of the book is Macmillan's best. a. j. h.

A Laboratory Manuai, of Inorganic Chemistry. By Eugene C.
Bingham, Ph.D. (Johns Hopkins), Professor of Chemistry, Richmond
College, Richmond, Virginia, and George F. White, Ph.D. (Johns
Hopkins), Associate Professor of Chemistry, Richmond College,
Richmond, Virginia. New York : John Wilay & Sons ; London :
Chapman and Hall, Ltd. 1911. pp. viii + 147.' Price, $1.00.

The first section of the book is called "Inorganic Prepara-
tions," but comprises nothing more than the laboratory work
on nonmetals usually included in the first year's work in chem-
istry. The second section, called "Quahtative Analysis,"
is a combination of the study of reactions of the single metals
with a scheme of systematic qualitative analysis, closing with
the analytical detection of acids. The third section is a brief
exposition of principles of quantitative analysis. The authors
say in the preface: "The Principles of Physical Chemistry

1 Vol. 42, 373-

Reviews 215

have been freely introduced whenever they seemed necessary
to the understanding of the subject at hand, but with as lit-
tle technical language as possible."

The tendency of the day is to introduce more or less physical
chemistry into class and laboratory work in the second col-
lege year, and this tendency meets with general approval.
Whether it is advisable in the first year seems questionable.
The authors have employed these methods sparingly and in
a very simple way; yet whether, for example, the help to the
student in balancing equations (pages 66-69) by graphic
representation of the change in electric charge, will balance
the confusion caused in the beginner's mind by such formulas
as P^Clg-^ + Cl2° = P^Clr^ or H^S-^ + UN'O, -^ H^S^O, +
N-0 + H2O (in which the lower figures denote the number
of atoms in the molecule, the upper figures and symbols the
number and sign of the charges) seems again questionable.

This criticism may not be valid, and if it is valid it applies
to only a small part of this book, which as a whole is clear,
logical, well written and gives an excellent course of study,
although the authors acknowledge that it is not feasible to
complete the work outlined in one college year. E. r.

Chemistry for Beginners. I. Inorganic Bv Edward Hart,
Ph.D., Professor of Chemistry, Lafayette College, Easton, Pa. Fifth
Edition, Revised and Enlarged, with 78 Illustrations and 2 Plates.
Easton, Pa.: The Chemical Publishing Co. 1910. pp. viii -{- 214.

Professor Plart says in the preface: "I have tried to bear
constantly in mind that the large majority of those studying
chemistry are not likely to become professional chemists,
and have therefore taken pains to enlarge upon those topics
which all educated persons should understand, such as water
purification, fertilizers, the concentration of ores, the roasting
of ores, assaying, the iron blast furnace, steel manufacture,
etc." A perusal of the volume leads to the belief that the
author has performed his task successfully. E. R.

A Course in Qualitative Chemicai, Analysis of Inorganic Sub-
stances, with Explanatory Notes. By Olin Freeman Tower,
Ph.D., Hurlbut Professor of Chemistry in Adelbert College of
Western Reserve University. Second Edition, Revised. Philadelphia:
P. Blakiston's Son & Co. 191 1. pp. xiii -f- 84.

A favorable review of the first edition of this book has already
appeared.^ Evidently the book has been well received by
the public as a second edition is so soon printed. E. R.

> This Journal, 46, 414.

2i6 Reviews

DiziONARio Di MERCEOI.OGIA K Di Chimica Appucata alia conos-
cenza dei prodotti delle cave e miniere, del suolo e dell'industria, con
specialeriguardoaiprodottialitnentari,chimiciefarmaceutici. Pel Prof.
DOTT. VlTTORio V1LI.AVKCCHIA, Direttore dei laboratori chimici delle
Gabelle, con la collaborazione dei Dott. Guido Fabris, Dott. Guido
Rossi, Dott. Arnaldo Bianchi. Terza edizione, completamente
riveduta, corretta ed ampliata. Vol. I, L,ettere A-M. Milano : Ulrico
Hoepli. 191 1, pp. xiv -\- 779. Price, L. 15.

For purposes of reference and as a supplement to other
sources of information, this alphabetical list of commercial
substances should be very useful. The large number of
separate headings (this volume appears to contain about
2000) makes possible the inclusion of most raw materials,
textiles, pharmaceutical substances, drugs, foodstuffs, metals
and minerals, dyes, etc., and short articles on such varying
topics as dentifrices and densimetry, weights and measures
and jujubes. The English, French, German and Spanish
equivalents are given for each substance, together with many
facts of interest for their practical and commercial bearing.
Thus, under Menthol, a typical article, in about 600 words
we are told its properties and tests, uses, derivatives (with
their trade names), production in kilograms, price, and the
(Italian) tariff. Indigo receives 10 columns (about 4,000
words), including a short account of its synthesis; Cement
receives 8 columns; and Cotton twelve. Emphasis is placed
on commercial qualities and brands, but purely chemical in-
formation is necessarily restricted. The second volume is
promised shortly. b. b. turner.

A Short Hand-Book of Oii, Anai.ysis. By Augustus H. Gill, S.B.,
Ph.D., Professor of Technical Analysis at the Massachusetts Institute
of Technology, Boston, Mass. Sixth edition, revised and enlarged.
Philadelphia and London : J. B. Lippincott & Co. 1911. pp. 188.

The sixth edition of Gill's "Oil Analysis," recently issued,
following soon after the preceding edition, indicates a brisk
demand in the field of analysis that this book occupies.

Some additional tests appear in this edition in the line of
the expansive use of mineral oils and vegetable oils that have
recently come into the trade, notably the carbon test for
auto oils and the gasoline test. More space is given to greases
and miscellaneous oils, such as transformer and turbine lubri-
cants and reduced oils. Acheson's oildag is mentioned.
Some changes in the tables and useful additions, especially
the formulas to change the readings of one viscosimeter to
another, are noted. This book is indispensable in the field it
covers. c. f. mabkry.

Vol. XIrVI September, 191 i No. 3

AMERICAN

CHEMICALJOURNAL

THE REACTION BETWEEN UNSATURATED COM-
POUNDS AND ORGANIC ZINC COMPOUNDS

[second communication]

By E. p. Kohler, G. L. Heritage and A. L. Macleod

In the first paper* on this subject we published the results
that were obtained by treating a variety of unsaturated com-
pounds with zinc and bromacetic esters. We showed that
a number of unsaturated ketones that form only saturated
compounds when treated with Grignard reagents give un-
saturated ;9-hydroxy esters when they react with zinc and
bromacetic ester; and that ketones like benzaldesoxybenzoin
and benzalacetomesitylene, which have an exceptionally un-
reactive carbonyl group, do not react with the zinc derivatives
at all, although they give an excellent yield of 1,4-addition
product with Grignard reagents. From these results we
concluded that the mode of addition of organic zinc com-
pounds derived from a-brom esters is different from that of
any Grignard reagent that has hitherto been used.

Our later experiments show that this conclusion should
have been confined to the zinc derivatives of bromacetic es-
ters. For while zinc bromacetates appear to form only 1,2-
addition products with all kinds of unsaturated compounds,

1 This Journal, 43, 475.

2i8 Kohler, Heritage and Macleod

the zinc compounds obtained from other a-brom esters behave
differently. Our work on the subject is not nearly completed,
but in order to avoid misconceptions we think it best to indi-
cate the character of the results by publishing an account of
our experiments with one unsaturated ketone — benzalaceto-
phenone.

The product obtained by heating a benzene solution of benzal-
acetophenone and an a-brom ester may contain three classes
of substances : zinc compounds that give unsaturated hydroxy
esters when treated with an acid, zinc compounds that, on
treatment with acid, give saturated ke tonic esters, and un-
saturated lactones. The first class of zinc compounds is
evidently the result of 1,2 -addition to carbonyl:

CgHg.CHiCH.CO.CeH^ + RCHBrCOjR + Zn =

CeHj.CH : CH.CCCeHJOZnBr CeHs.CH : CH.C(CoH5)OH

R.CH.CO2R R.CH.CO2R

More than one compound of this class may be formed be-
cause these substances contain both double linkages and dis-
similar asymmetric carbon atoms. Stereoisomeric esters are
frequently obtained.

The second class might be formed either by 1,2-addition
to the ethylene linkage, or by 1,4-addition to the conjugated
system C :C.C :0:

I. CeHj.CH : CH.CO.CeH5 + RCHBrCO^R + Zn =

C9H5.CH.CH(ZnBr)COC8H5 CeH5.CH.CH3.CO.CeH5

R.CH.CO2R RCH.COjR

II. CeHj.CH : CH.CO.CeH5 -f RCHBrCO^R -i- Zn =
CeH5.CH.CH : C(CeH5)0ZnBr

R.CH.CO,R
CeH5.CH.CH : C(CeH5)0H CeH5.CH.CH2.CO.CeH5

RCH.CO2R R.CH.COjR

The character of the unsaturated lactones proves that the
second interpretation is correct. When these lactones are

Reaction between Unsaturated Compounds, Etc. 219

treated with sodium carbonate they give the same ketonic
acids that result from the hydrolysis of the esters obtained
from the second class of zinc compounds. They are there-
fore ^-lactones, as represented by the formula

CgH5.Cxl.Crl : C.CgXlg

O . '

I
RCH CO

The relative amount of lactones formed from a given brom
ester varies with the conditions — concentration, length of
heating, and other factors that cannot be sharply defined.
As the lactones can be separated from the benzene solution
containing all the products of the reaction without treating
this solution either with water or acids, they must be formed
directly from a zinc compound. They, therefore, supply
excellent proof that the second class of zinc compounds is
due to 1,4-addition:

CeHj.CH : CH.CO.CeH5 + BrZnCH.COjR =

CgHs.CH.CH = C.CgHg CeH5.CH.CH : C.CeHs

I I — > I I + ROZnBr.

R.CH.CO2R OZnBr RCH. CO. O

Stereoisomerism is possible also in the case of the second
class of zinc compounds and in that of the lactones.

It is clear, now, that the mode of addition of organic zinc
compounds, derived from a-brom esters, to unsaturated com-
pounds containing the conjugated system C : CO : O is pre-
cisely the same as that of organic magnesium derivatives to
the same class of substances. The final result depends upon
the character of the unsaturated compound as well as that
of the ester from which the metallic derivative is made. In
most cases both 1,4- and 1,2-addition take place in the same
reaction, and in order to ascertain the factors that influence
the mode of addition it is necessary to find some method for

220 Kohler, Heritage and Macleod

estimating the relative amounts of 1,2- and i ,4-addition prod-
ucts in a mixture.

The oxidation method used with the corresponding products
obtained from Grignard reagents was found inapplicable,
because the lactones are almost as easily oxidized by perman-
ganate as the unsaturated hydroxy esters, and the method
failed even when, by working in very dilute solution, lactone
formation was prevented, because the ke tonic esters are
oxidized with sufficient rapidity to vitiate the results. An
attempt to determine the relative amounts by titration with
a standard solution of bromine gave no better results. The
hydroxy esters rapidly combine with bromine at — 20°, and
while the pure ketonic esters do not react with bromine at
this temperature, in a mixture of the two both are attacked —
probably because a small quantity of hydrobromic acid acts
as a catalyst in the substitution of bromine in the a-position
in ketones.

The best method that we have been able to find up to this
time consists in heating the mixture, containing all of the
products, with a solution of sodium carbonate in a current
of steam. By this procedure the unsaturated lactone and
ketonic ester are gradually dissolved as sodium salt of the
ketonic acid. The hydroxy ester is also hydrolyzed, but the
process takes place very slowly and in most cases the re-
sulting acid is decomposed almost as fast as it is formed. The
acid obtained by acidifying the resulting solution was dried,
weighed and regarded as ketonic acid due to 1,4-addition.
The method does not give accurate results for two reasons:
the ketonic acid is decomposed, very slowly and in variable
quantities, into unsaturated compounds and fatty acids, and
small, variable quantities of hydroxy acid remain undecom-
posed and are weighed as ketonic acid. These errors are in
opposite directions, and from our whole experience with the
method we estimate that our error does not exceed 10 per cent.
The following results, obtained with benzalacetophenone,
show how the results vary with the character of the brom
ester :

Reaction between Unsaturated Compounds, Etc. 221

Per cent, of
1,2-addition

Per cent, of
1.4-addition

Br.CH^.CCCHg

100

BrCHCCHgjCO^CHg

BrCH(C2H5)C02CH3

BrC(CH3)3CO,CH3

100

BrCHCCOjCHs)^

100

The significance of these results for the interpretation of un-
saturation will be discussed in a later paper dealing with a
greater variety of unsaturated compounds.

EXPERIMENTAIv PART

The reactions between unsaturated compound, brom ester,
and zinc were, in the main, carried out as described in our first
paper. In a few cases we met with unexpected difficulties,
owing to the peculiar inactivity of the brom ester. Thus,
when we used a fresh supply of methyl bromacetate we found
it quite impossible to start a reaction either with saturated or
unsaturated ketones. The preparation contained no hydro-
bromic acid, was free from water and alcohol, had a constant
boiling point, and gave correct results on analysis — seemed
in fact, to be an unusually pure substance. We met with the
same difficulty while using brommalonic ester that we had
prepared ourselves. In these cases it was necessary to use a
catalyst. We found that none of the methods devised for
getting more active magnesium in the preparation of Grig-
nard reagents could be used with zinc; but that it is easy to
start the reaction by adding a very small quantity of some
organic copper compound that is soluble in benzene. We used
copper acetacetic ester, and we have found that even when
it is possible to start a slow reaction without it, better yields
are sometimes obtained by adding a small quantity of copper
derivative than by protracted boiling. With nitro com-
pounds, saturated or unsaturated, the reaction fails com-
pletely even in the presence of a catalyst.

Our treatment of the product of the reaction depended
upon the object of the experiment. When the reaction was
used for the purpose of determining the relative amounts of
1,2- and 1 ,4-addition, the benzene solution was poured di-

222 Kohler, Heritage and Macleod

rectly into sodium carbonate and distilled with steam as long
as anything went into solution. In case the object was to
isolate the substances formed, the benzene solution was usually
poured into iced acid, the benzene layer washed thoroughly
and dried, and the benzene evaporated in a draught. The
residue generally contained a number of substances because
in most cases spatial as well as structural isomers were formed.
The isolation of all of these substances is frequently difficult
even though each one crystallizes well when pure. Great
care must be taken to remove all free acid before concentra-
ting the solution, because in the presence of acid the solid
unsaturated hydroxy esters rapidly pass into viscous liquids
that interfere with the crystallization of everything present.

Experiments with Bromacetic Ester
We repeated the reaction between bromacetic ester and
benzalacetophenone described in the first paper, but took
special precautions to detect any 1,4-addition product that
might be formed. Thus, in one experiment in which more
than 200 grams of benzalacetophenone were used, we sepa-
rated the solid hydroxy ester as far as possible by crystalliza-
tion, and then oxidized the accumulated residues with potas-
sium permanganate at a temperature that was never allowed
to rise above — 10°. Parallel tests showed that in acetone
solution the hydroxy ester is oxidized fairly rapidly, while
methyl benzoylphenylbutyrate — the ketonic ester that would
be formed by 1,4-addition — is not attacked at this tempera-
ture. We examined the oxidation products with care, but
could not find a trace either of the ketonic ester or of the cor-
responding acid. We then repeated the experiment, but
heated the residue with sodium carbonate in a current of
steam. This also gave a negative result. Finally we de-
stroyed the hydroxy ester by heating. Methyl hydroxy-
phenylbenzalbutyrate decomposes into benzalacetophenone
and methyl acetate below 180°; the isomeric ketonic ester
does not change below 260°. We heated to 200° until methyl
acetate ceased to distil, and found the residue free from ke-
tonic ester. From these results we are compelled to conclude
that no 1,4-addition takes place with methyl bromacetate.

Reaction between Unsaturated Compounds, Etc. 223

y-Benzal-^-hydroxy-^-phenylbutyric A cid,
CgHs.CH : CH.C(OH)C6H5.— When the methyl ester obtained

CH2.COOH
in the experiments described above is treated with aqueous
or alcoholic potassium hydroxide it is rapidly decomposed
into benzalacetophenone and potassium acetate. By heating
with sodium carbonate in a current of steam it is possible,
however, to hydrolyze it with a yield of more than 70 per
cent, of the corresponding acid. This was purified by crys-
tallization from ether and ligroin. It is readily soluble in
alcohol and ether, very sparingly in ligroin. From ether it
separates in colorless needles that melt with decomposition
at about 147°, when heated rapidly in a capillary tube. When
heated slowly the acid loses both carbon dioxide and water
far below the melting point.

Analysis :

0.13 16 gram substance gave 0.3652 gram CO, and 0.0683
gram HjO.

Calculated for

C17H16O3 Found

C 76.1 75 9

H 5-9 5-8

^-Phenylcinnamylidenacetic Acid, CeHj.CH :CH.C(C6H5) :CH.
COOH. — While working with methyl bromacetate and benzal-
acetophenone we occasionally obtained a considerable quan-
tity of an exceedingly viscous oil instead of the crystalline
hydroxy ester. We found that the oil is formed whenever
the ester is warmed in the presence of a small quantity of
hydrochloric acid. The oil was hydrolyzed by boiling with
alcoholic potassium hydroxide. It yielded a solid acid that
crystallized from ether in fine, very pale yellow needles, melting
at i45°-i46°.

Analysis :

0.1413 gram substance gave 0.4235 gram CO2 and 0.0770
gram HjO.

Calculated for

C,7H„02 Found

C 81.6 81.7

H 5.6 6.1

224 Kohler, Heritage and Macleod

Experiments with Brompropionic Ester
The reaction with ethyl a-brompropionate was easily started
and it proceeded rapidly, all of the zinc disappearing after
boiling for less than an hour. The resulting benzene solution
was poured into a saturated solution of sodium carbonate
and heated in a rapid current of steam. At intervals of 5
hours the mixture was cooled, extracted with ethex", the aqueous
layer removed, and the ethereal solution heated with sodium
carbonate as before. This was repeated as long as organic
acid was obtained by acidifying the carbonate solution. The
crude acid obtained by acidifying all the solutions in this
way represented 46 per cent, of the benzalacetophenone
used in the experiment. The last fractions melted between
130° and 140° and gave off a small quantity of gas when
heated above 150°. They therefore probably contained a
small quantity of hydroxy acid, but the only substance that
could be isolated in pure form was a ketonic acid melting at
149°.

The separation of the products of the reaction proved diffi-
cult, the best results being obtained as follows: The benzene
solution was diluted with ether and the mixture shaken with
dilute hydrochloric acid. The ethereal layer was then separa-
ted and poured into a solution of sodium carbonate and heated
in a current of steam for an hour. The solvents and excess
of brom ester were carried off by the steam and any lactones
formed in the reaction went into solution as sodium salts of
ketonic acids, while the esters in the main remained unchanged,
because they are very slowly saponified by sodium carbonate.
The mixture was shaken with ether, the aqueous layer acidified,
and the ethereal layer evaporated. The aqueous layer yielded
two acids (4.5 grams from 42 grams of benzalacetophenone)
that were separated by recrystallization from ether and ligroin.
When pure, these melted at 149° (2 grams) and 107°.

The ethereal solution deposited a mixture of esters that
was separated by repeated crystallization from methyl alco-
hol. The principal product was an ester melting at 107°,
the remainder consisting mainly of an isomeric ester melting
at 81°.

Reaction between Unsaturated Compounds, Etc. 225

Ethyl y-Benzal-^-phenyl-^-hydroxy-a-meihylhutyrate, CgHg-CH :
CH.C(OH)(C6H5)CH(CH3)COOC2H5, the ester melting at 107°,
is readily soluble in alcohol and ether, insoluble in ligroin.
It crystallizes in fine needles.

Analysis :

0.15 1 8 gram substance gave 0.4302 gram CO2 and 0.0996
gram HjO.

Calculated for
C20H22O3

Found

77-4
7-1

77-3
7-3

The ester is oxidized slowly by a cold solution of potassium
permanganate in acetone, decomposes into benzalacetophe-
none and ethyl butyrate when heated above the melting point,
gives benzalacetophenone and potassium propionate when
warmed with alcoholic potassium hydroxide, and instantly
decolorizes a solution of bromine at the ordinary temperature.
It is therefore an unsaturated hydroxy ester as formulated.

The ester melting at 81° was obtained only in small quanti-
ties. It separates slowly from alcohol in large transparent
prisms readily soluble in all common organic solvents except
ligroin.

Analysis :

0.1584 gram substance gave 0.4527 gram CO2 and 0.1012
gram HjO.

Calculated for
C20H22O3

Found

77-4

78.0

7.2

The analysis shows that the ester is isomeric with the ester
melting at 107°, and its properties indicate that it is another
hydroxy ester. It reduces permanganate, decolorizes bro-
mine, and is decomposed into benzalacetophenone and potas-
sium propionate by treatment with alcoholic potassium hy-
droxide. It is therefore a stereoisomer of the ester melting at

y-SlHItU'^- phenyl- a-methylbutyric Acid, CgHg.CO.CHj.
CH(C6H5)CH(CH3)COOH.— As stated above, the acid obtained
in largest quantity by heating the product of the reaction

226 Kohler, Heritage and Macleod

with sodium carbonate melts at 149°. This acid separates
from a mixture of ether and ligroin in fine feathery needles.

Analysis :

0.1557 gram substance gave 0.4383 gram CO2 and 0.0918
gram HjO.

Calculated for

CigHjsOs Found

C 76.6 76.7

H 6.4 6.6

The acid can be heated far above its melting point without
causing decomposition. It does not combine with bromine
and is not decomposed by boiling with alcoholic potassium
hydroxide. It is therefore not an unsaturated hydroxy acid
formed by addition to carbonyl, but a saturated ketonic acid
formed by i ,4-addition. Like other ketonic acids of this char-
acter, it slowly reduces permanganate at the ordinary tem-
perature.

The acid melting at 105° is likewise a ketonic acid because
it has the same chemical properties as the acid melting at 149°.
The analysis shows that it is isomeric with that acid; it is
therefore a stereoisomer, possible because the ketonic acids
have 2 dissimilar asymmetric carbon atoms. The acid crys-
tallizes in very fine needles readily soluble in alcohol and
ether, sparingly in ligroin.

Analysis :

0.1345 gram substance gave 0.3800 gram COj and 0.0777

gram HjO.

Calculated for
CisHjsOa

Found

C
H

76,6
6.4

77.0
6.5

Methyl j-Benzoyl- /9 -phenyl- a -methylbutyrate, CeHj.CO.CHj.

OnM^ CH(CeH5)CH(CH3)COOCH3.— This ester was prepared by

fi^ ^ saturating the solution of the omr melting at 149° in methyl

i/).b/^ alcohol with hydrogen chloride, and purifying the product

' by recrystallization from methyl alcohol. It crystallizes in

fine needles melting at 68°.

Analysis :

Reaction between Unsaturated Compounds, Etc. 227

0.1309 gram substance gave 0.3720 gram CO2 and 0.0813
gram HjO.

Calculated for

C19H20O3 Found

C 770 77.5

H 6.7 6.9

The corresponding ethyl ester was prepared in the same way.
It crystallizes in needles melting at 41°.

Analysis :

0.1358 gram substance gave 0.3849 gram COj and 0.0858
gram HgO.

Calculated for

C20H22O3 Found

C 77-4 77-3

H 71 71

The low melting point of this ester explains why it cannot
be isolated from the product of the reaction between benzal-
acetophenone, a-brompropionic ester and zinc, even though
it composes a large part of this product. As shown by the
above results, this reaction gives at least 5, and possibly 6,
substances: two stereoisomeric hydrox)' esters, one or two
ketonic esters, and one or two lactones. The presence of
lactones is inferred from the relatively large quantity of acid
obtained by boiling the product with a solution of sodium
carbonate for a short time.

Experiments with Brombutyric Ester

The reaction with methyl a-brombutyrate goes so slowly
that we found it advisable to use a small quantity of copper-
acetacetic ester as catalyst. For the purpose of determining
the relative amounts of 1,2- and 1,4-addition, the benzene solu-
tion was poured into sodium carbonate and heated as described
under the experiments with brompropionic ester. The quantity
of crude ketonic acid obtained by this procedure represented
50 per cent, of the benzalacetophenone used. The acid was
entirely free from hydroxy acid. It was purified by recrys-
tallization from alcohol.

228 Kohler, Heritage and Macleod

y-Benzoyl-^-phenyl- a-ethylhutyric acid,

CeH6.CO.CH2.CH(C6H5)CH(C2H5)COOH,
crystallizes in needles melting at i8i°.

Analysis :

0.1355 gram substance gave 0.3809 gram COj and 0.0825
gram HjO.

Calculated for

C,9H20O3

Found

77.0

76.8

6.7

6.8

The acid does not combine with bromine, and does not de-
compose when heated above its melting point, but, like other
ketonic acids previously described, slowly reduces perman-
ganate at the ordinary temperature.

The methyl ester was made by saturating a solution of the
acid in methyl alcohol with hydrogen chloride. It separates
from methyl alcohol in fine needles melting at 95°.

Analysis :

0.1425 gram substance gave 0.4045 gram CO2 and 0.0909
gram HgO.

Calculated for
C20H22O3

Found

77-4
71

77-3
7.2

Methyl y-Benzal-^-phenyl-^-hydroxy-a-ethylbutyrate,
CeHg.CH :CH.C(C6H5)(OH).CH(C2H5)COOCH3.— The hydroxy
ester is easily obtained by decomposing the zinc compounds
in the usual way. It crystallizes in needles melting at 117°.

Analysis :

0.1324 gram substance gave 0.3757 gram CO2 and 0.0836
gram Ufi.

Calculated for
C20H22O3

Found

77-4
71

77-4
7.2

The ester combines with bromine, reduces permanagnate,
and above its melting point decomposes cleanly into benzal-
acetophenone and methyl butyrate.

Reaction between Unsaturated Compounds, Etc. 229

Experiments with Bromisobutyric Ester

In order to get a complete reaction between benzalaceto-
phenone, ethyl bromisobutyrate, and zinc, it is necessary to
use an excess of at least ten per cent, both of ester and zinc
because a part of these substances is lost in secondary proc-
esses. The final result of the reaction in this case depends,
to a considerable extent, upon the concentration and the
amount of heating. In concentrated solutions, in which the
reaction proceeds rapidly after it is once started, the principal
product is an unsaturated lactone. In dilute solution, in
which the substances react more slowly, the principal product
is, usually, a ketonic ester; but if the dilute solution is boiled
for a long time the ketonic ester steadily disappears and a
relatively larger amount of lactone is obtained. The lactone
is, therefore, formed at the expense of the zinc derivative of
the ester:

C,H5.CH.CH:C<

I ^OZnBr ~

(CH3)X— CO.OC2H5

QHs.CH.CHtC.CeHg

I I + Zn(OC3H5)Br

(CH3)2C— CO— O

The most convenient way of getting pure products from
the reaction is as follows: The mixture is poured into acid
and washed as usual. The benzene layer is then poured
into water and distilled in a rapid current of steam for an
hour. This removes the benzene, any unchanged materials,
and most of the secondary products. The residue is dissolved
in ether and this solution extracted with sodium carbonate
to save a variable quantity of acid that is formed from the
lactone during distillation. The ethereal solution, on evapora-
tion, deposits a mixture of lactone and acid contaminated
with a small quantity of yellow oil. The solid is washed with
methyl alcohol that has been cooled in a freezing mixture
and then recrystallized from the same solvent. The lactone
separates first and, from concentrated solution, almost com-
pletely, while the ketonic ester remains in solution until most

230 Kohler, Heritage and Macleod

of the solvent has evaporated if crystallization is not started
by the addition of a small quantity of solid. The lactone is
very sparingly soluble in ligroin, moderately in ether and cold
alcohol, readily in acetone. It crystallizes in long colorless
needles melting at 97°.

Analysis :

0.1744 gram substance gave 0.5228 gram CO2 and o. ioo8
gram HjO.

Calculated for

CigHisOo Found

C 82.0 81.7

H 6.5 6.4

The substance reduces a solution of potassium perman-
ganate in acetone in the cold and combines with bromine be-
low 0°. The lactone ring is slowly opened by boiling with
water, more rapidly by boiling with sodium carbonate, im-
mediately by solution in alcoholic potassium hydroxide.

y-Benzoyl-^-phenyl-a,a-dimethylbutyric Acid, CeHg.CO.CHa-
CH(C6H5)C(CH3)2COOH, the acid obtained from the lactone,
was purified by crystallization from methyl alcohol. It
separates in fine white needles melting at i59°-i6o°.

Analysis :

0.1142 gram substance gave 0.3228 gram COj and 0.0678
gram HjO.

Calculated for

C,9H20O3

Found

77.0

6.8

77-1
6.6

The acid is sparingly soluble in ether and chloroform, readily
in alcohol and acetone. Its solution in sodium carbonate
does not reduce potassium permanganate in the cold, but the
color rapidly disappears on warming. The acid does not re-
act with bromine at the ordinary temperature, but by adding
bromine to a hot solution of the acid in chloroform or carbon
tetrachloride it is easy to replace one of the hydrogen atoms
in combination with the carbon atom that adjoins the car-
bony 1 group.

Ethyl y-Benzoyl-^-phenyl-a, a-dimethylbutyrate,
CeH5.CO.CH2CH(CeH5)C(CH3)2COOC2H5.— The ester was sep-

Calculated for

C2,H2403

Found

77.8

77-8

7-4

Reaction between Unsaturated Compounds, Etc. 231

arated from the mother liquor after removing the lactone. It
was also made from the acid by saturating its solution in abso-
lute alcohol with hydrogen chloride, and from the lactone by
the same procedure. It is readily soluble in ether and alco-
hol, sparingly in ligroin. It separates from methyl alcohol
in needles melting at 83°.

Analysis :

0.15 1 2 gram substance gave 0.4317 gram CO2 and 0.1003
gram HjO.

C
H

The methyl ester, made and purified like the ethyl ester,
crystallizes in needles melting at 92°. As it is less soluble
and crystallizes much more readily than the ethyl ester it is
a much more convenient substance to work with.

Analysis :

o . 1 2 1 1 gram substance gave o . 3449 gram CO, and o . 0790
gram H2O.

C
H

Y-Brom-y-benzoyl-^-phenyl-a,a-dimethylbutyr'ic acid,
C6H5.CO.CHBrCH(CeH5)C(CH3)2COOH,
the acid obtained by adding the calculated amount of bromine
to a hot solution of the ketonic acid in chloroform or carbon
tetrachloride, is so unstable that we found it exceedingly
difficult to get a pure product. By dissolving it in acetone
at the ordinary temperature, adding an equal volume of car-
bon tetrachloride to this solution, and then cooling in a freezing
mixture, we finally obtained long, colorless needles that gave
consistent analytical results. The acid begins to decompose
slowly below the melting point, which is consequently incon-
stant; but on rapid heating in a capillary tube, the substance
seems to melt and decompose fairly sharply at 186°.

Analysis :

Calculated for

C20H22O3

Found

77-4

77-6

7.2

232 Kohler, Heritage and Macleod

0.1 6 10 gram substance gave 0.3607 gram CO2 and 0.0765
gram HjO.

Calculated for
CisHigOsBr Found

C 60 . 9 6 1 . I

H 51 5-3

The esters of the acid were easily made by adding bromine
to a solution of the corresponding ketonic esters in carbon
tetrachloride. It is necessary to start the reaction by heat-
ing, but after that the bromine disappears as fast as it is
added, until the solution contains one molecule of bromine
per molecule of ester. The solution in carbon tetrachloride,
on evaporation, deposits the esters in solid form, and they are
readily purified by recrystallization from methyl alcohol.

The methyl ester crystallizes in large prisms, moderately
soluble in ether and cold alcohol, readily soluble in boiling
alcohol. The melting point is 125°.

Analysis :

0.1457 gram substance gave 0.3286 gram CO2 and 0.0720
gram H^O. 0/,/)^^^

Calculated for ASJ jg>Vi/^

c 61.9 y^» ^ ' 61.6

H 5-4 5-5

The ethyl ester crystallizes in fine needles or thin plates
melting at 131°.

Analysis :

0.1489 gram substance gave 0.3431 gram COj and 0.0802
gram HjO.

Calculated for
CziHgaOaBr

Found

62.5

62.8

5-7

6.0

y-Benzoyl-^-phenyl-a, a-dimethylh utyrolactone,
CgHs . CH . CH . CO . CeHg.— Brombenzoylphenyldimethylbutyric

(CH3)2C — CO

acid, when finely powdered, dissolves readily in dilute sodium

Reaction between Unsaturated Compounds, Etc. 233

carbonate, but the solution stays clear only a very short time.
A faint milkiness appears in a few minutes and in the course
of an hour the liquid becomes filled with a network of fine needles
that represent practically all of the acid dissolved. The solid
consists of two stereoisomeric lactones formed as represented
by the equation

C^Hg.CH.CHBr.CO.CeHs CeH5.CH.CH.CO.CeH5

O + NaBr.

1
(CH3)2C.COONa (CH3)2C— CO

The lactones were separated as follows: The solid was fil-
tered, washed thoroughly with boiling water, dried, and ex-
tracted with boiling methyl alcohol. This extracted, mainly,
the lower-melting lactone, which had a constant melting point
of 113° after a few recrystallizations from the same solvent.
The residue, left after extraction with methyl alcohol, was re-
crystallized from acetone until the melting point remained
constant at 173°. Both lactones crystallize in colorless
needles.
Analyses :

I. 0.13 13 gram substance (low melting) gave 03726 gram
CO2 and 0.0741 gram HjO.

II. 0.1442 gram substance (high melting) gave 0.4090
gram COj and 0.0815 gram Hfi.

Calculated for
CxsHisOa

Found

C
H

77.6

6.1

77-4
6.3

The lactones are not aflfected by hot sodium carbonate, but
they dissolve readily in cold alcoholic potassium hydroxide.
If the pale yellow solutions obtained in this way are allowed
to stand for some time they give only complex products when
acidified; but if they are acidified immediately they give the
corresponding hydroxy acid. The same two acids are formed
from each lactone: a small quantity of a high-melting, spar-
ingly soluble acid that loses water so readily that it was im-

234 Kohler, Heritage and Macleod

possible to get it free from the high-melting lactone, and a
low-melting acid that was purified by recrystallization from
ether. It separates in needles that lose water below the
melting point, but on rapid heating melt fairly sharply at 126°.
When heated to 120° for several hours the acid passes com-
pletely into the low-melting lactone. The results of an analy-
sis show that this acid is ■jr-benzoyl--jr-hydroxy-^-phenyl-a,a-
dimethylbutyric acid,

C6H5.CO.CHOH.CH(C6H5)C(CH3)2COOH.

Analysis :

0. 1410 gram

substance gave 03793 gram CO2 and 0.0815

gram HjO.

Calculated for

C19H20O4 Found

73 I 73-4
6.3 6.4

Experiments with Brommalonic Ester

Methyl Brommalonate , BrCH(C02CH3)2.— As the ethyl ester
of brommalonic acid gave only oily products we prepared the
methyl ester by gradually adding the calculated quantity
of bromine to methyl malonate and fractionating the product
under diminished pressure. The reaction starts slowly, but,
once started, proceeds rapidly at the ordinary temperature.
The yield is unsatisfactory. The best result — about 70 per
cent, of the calculated amount of pure product — was ob-
tained when the crude product was heated under diminished
pressure and finally distilled without preliminary washing.
The ester is a colorless, mobile liquid boiling at 145° (22 mm.).

Analysis :

o. 1582 gram substance gave o. 1410 gram AgBr.

Calculated for
CsHyO^Br

Fotind

37 92

38.03

Methyl y-Benzoyl-^-phenylethylmalonate,
C6H5.CO.CH2.CH(C6H5)CH(C02CH3)2.— The product obtained
from methyl brommalonate, benzalacetophenone and zinc

Reaction between Unsaturated Compounds, Etc. 235

partially solidified when the benzene solution was evaporated.
The solid was easily purified by a few recrystallizations from
methyl alcohol. It is sparingly soluble in ligroin, moderately
in ether and methyl alcohol, readily in alcohol, chloroform,
and acetone. From methyl alcohol it separates in large,
colorless needles melting at 107°.

Analysis :

0.1 163 gram substance gave 0.3017 gram COj and 0.0652
gram Ufi.

Calculated for

C20H20O5 Found

C 70.6 70.8

H 5.9 6.23

On hydrolysis with alcoholic potassium hydroxide the ester
gave a dibasic acid that readily lost carbon dioxide when heated
to 170°. The resulting monobasic acid, after recrystalliza-
tion from ether, melted at 156°. It was identified as ^--benzoyl-
/?-phenylbutyric acid by comparison with a specimen on hand.
The solid is, therefore, the methyl ester of the dibasic acid
obtained by Vorlander^ by condensing benzalacetophenone
and sodium malonic ester and hydrolyzing the product with
alcoholic potassium hydroxide. Only about one-third of the
product was obtained in solid form, the remainder consisting
of a viscous liquid that decomposed when distilled under di-
minished pressure. This liquid is apparently a lac tonic ester
formed from the zinc compound in the usual way. When
boiled with sodium carbonate it slowly goes into solution
and is precipitated as an oily acid when the solution is acidified.
With alcoholic potassium hydroxide it gives the same acid
that is obtained from the solid ester.

Very little, if any, 1,2 -addition product is formed in the re-
action between benzalacetophenone, brommalonic ester and
zinc. We have prepared the unsaturated hydroxy ester that
would be formed by 1,2 -addition in another way, and know
that it is very rapidly decomposed into benzalacetophenone
and malonate by boiling with sodium carbonate. As the en-
tire product from the zinc compound gave only a ver>' small

• Ann. Chem. (Liebig), 294, 332.

236 Garner, Saxton and Parker

quantity of benzalacetophenone when boiled with sodium
carbonate the amount of hydroxy ester present is certainly
small. Moreover, since the ketonic ester formed by 1,4-addi-
tion is slowly decomposed by the treatment, it is quite possible
that the small quantity of unsaturated ketone obtained was
derived from this source.

ANHYDROUS FORMIC ACID*

By James B. Garner, Blair Saxton and H. O. Parker

[preliminary paper]

For some time past several students, working in the labora-
tory here, have been studying the absorption spectra of solu-
tions of the various indicators. In connection with this
work results were obtained which necessitated a study of solu-
tions of these substances in some other solvent than water.
Formic acid was selected for the study on account of the fact
that all its physical constants were supposed to be more nearly
like those of water than those of any other ordinary sol-
vents.

The examination of the text-books, books of reference,
and the literature, disclosed the following: (i) That there
are no well defined, accurately described methods of prepara-
tion of anhydrous formic acid. (2) That there is very little
agreement in books or among authors as to the values of the
various physical constants.

Methods of Preparation which have been Used

Fractional crystallization was the method used by Peters-
son,^ Bannow,^ Hartwig,^ Sapojnikoff,^ Tessarin,^ and other
investigators.

• This paper was read before the Section of Physical and Inorganic Chemistry of
the Americn Chemical Society, at Indianapolis, Indiana.

2 J. prakt. Chem., 24, 296 (1881).

3 Ber. d. chem. Ges., 9, 4.

* J. Chem. Soc, M, 1308.
5/6td., 66, 66.

« Z. physik. Chem., 19, 251.

Anhydrous Formic Acid 237

Kahlbaum,* Schiflf,^ and the Chemische Fabrik, Griinau,''
attempted to prepare the pure acid by fractional distillation
under diminished pressure with sulphuric and metaphosphoric
acids and acid salts as dehydrating reagents. The maximum
concentration obtained by any of these investigators was
97 to 98 per cent.

Richardson and Allaire^ prepared what they regarded as
the pure anhydrous acid by repeated distillation over dry
lead formate.

Physical Constants which are Given

The following selection will clearly show the disagree-
ments which exist as to the value of the various physical
constants :

(i) In reference to melting point:

Melting Point Investigator Reference

8°.o BerthoUet

4°.o Bannow Ber. d. chem. Ges., 9, 4

8°. 43 Petersson and Ekstrand /6t(i. , 13, 1880

Note: No description of method of preparation given.

J. prakt. Chem., 24, 296
J. Chem. Soc, 66, 66
Ibid., 87, 1436

7°.45

Petersson

8°. 39

Sapojnikoff

8°. GO

Miss Homfray

i°.65,6°.5,

6°. 6, 6°. 9,

and 7°. I

Tessarin

Z.physik.Chem., 19, 251

(2) The following data with reference to specific gravity
measurements at 20° C. clearly illustrate the situation:

Sp. Gr. at 20° Investigator Reference

1 .219 J. Traube Ber. d. chem. Ges., 19, 884

1 .2213 Richardson and Allaire This Journal, 19, 149

1.2205 Miss Homfray J. Chem. Soc, 1905, 1436

1.223 Tessarin Z. physik. Chem., 19, 251

1.235 Liebig Smithsonian Tables

(3) The following table shows the uncertainty which exists
as to the true boiling point of anhydrous formic acid:

1 Ber. d. chem. Ges , 16, 2480.

2 Ibid., 19, 561.

3 J. Chem. Soc, 1908, 598.

< This Journal, 19, 149 (1897).

238 Garner, Saxton and Parker

Temp.

Pressure

Investigator

Reference

100°

760

Schiff

100°

760

Favre and Silber-

100°

760

mann
Landolt

Smithsonian Tables

100°

760

Zander

100°

760

Roscoe J

101°. 0

760

Tessarin Z. physik. Chem., 19,

251

100°

760

Kopp

100°

760

Kahlbaum Ber. d. chem. Ges., 16,

2480

100°

763 -5

Schiff Ibid., 19, 561

98°

748

Liebig

99°

748

Landolt

100°

749

Petersson and Ek-

strand Ber. d. chem. Ges., 13,

1880

Our first task therefore was the preparation of the pure
substance and our second task that of the accurate determina-
tion of the physical constants.

Method of Preparation

When method of Richardson and Allaire, namely, that of
repeated distillation from dry lead formate, was used
the acid obtained had the melting point 7°. 91. The meth-
ods of Kalhbaum, Schiff, and the Chemische Fabrik, Griinau,
were also used. It was found that phosphorus pentoxide and
sulphuric acid act destructively. Formic acid decomposes
violently into carbon monoxide and water under all the vary-
ing pressures. The yield is very poor. The acid thus obtained
has a melting point of 8°.

The laborious method of fractional recrystallization gave
us an acid melting at 8°. 34. Distillation under diminished
pressure over anhydrous nickel sulphate yielded an acid
melting at 8 ° . 2 1 . The acid obtained by distillation under
diminished pressure over anhydrous sodium sulphate melted
at 8°. 27. Distillation under diminished pressure, 120 mm.
at 50°, over anhydrous copper sulphate gave us an acid melt-
ing constantly at 8°. 35. This last method of preparation
we regard as the most satisfactory and efficient. Beginning
with the acid having a melting point of 7°. 26, the first dis-

at 8°, the second in one

Anhydrous Formic Acid

239

having a melting point of 8°. 21, and the third distillation
yields an acid melting at 8°. 31, while the fourth, fifth, and
sixth distillations all give an acid which melts at 8°. 35.

The analyses of the acid melting at 8°. 35 show it to be the
pure anhydrous acid.

I. 0.4276 gram substance gave 0.4103 gram CO2 and o. 1704
gram HjO.

II. o . 95 1 2 gram substance gave o . 9048 gram CO2 and o . 395 7
gram H2O.

Calculated

26.09

26.18

25 -95

438

4.46

465

Physical Constants

(a) The following values of the specific gravity were ob-
tained :

10°

15°
20°

1.2322
I . 2360
I .2200

1.2322
1.2260
I. 2199

25°

<
40°

I. 2139
1.2078
I .2019
I 1957

I. 2139
1.2079
I .2019
I. 1956

Viscos

ity measurements resulted as

follows :

C. G. S. units

Gartemneister's values
(Smithsonian tables)

10°,
15°

0.0226
0.02002

0.0231

20°

25°
30°

K
40°

Boilin

0.01793
0.01625
0.01474
0.01343
O.OI218

g-point determinations were :

0.0184
0.0149
0.0125

Temp.

Pressure
mm.

99°-7

120
741

240 Hosford and Jones

id) Surface tension, refractive index, specific heat, and
specific conductivity measurements are being made.

(e) Experiments are in progress which have for their pur-
poses: (i) The determination of the conditions — time, tem-
perature, friction, and presence of solid state — which influ-
ence the spontaneous crystallization of supercooled anhydrous
formic acid. (2) The determination of the range of the
metastable state. We have found that formic acid may be
readily cooled to 14°. 35 below its melting point in ordinary
glass vessels. Even with vigorous stirring, it supercools
to 5°. 8, or 2°. 55 below the melting point.

It is the intention of the authors to continue the work
with anhydrous formic acid to determine the following :

(i) The variation of solubility of electrolytes with the tem-
perature.

(2) The relationship between dissociation constant and
heat of dissociation.

(3) The relationship between molecular conductivities of
salts and increasing dilution.

(4) The re'ationship between dielectric constant and dis-
sociating power.

(5) The relationship between association factor and con-
ductivity; and

(6) The re ationship between viscosity and conductivity.

Peck Chemical Laboratory

Wabash College

Crawfordsville. Ind.

THE CONDUCTIVITIES, TEMPERATURE COEFFI-
CIENTS OF CONDUCTIVITY AND DISSOCIA-
TION OF CERTAIN ELECTROLYTES

By H. H. Hosford and Harry C. Jones
HISTORICAL REVIEW

Volta, at the end of the eighteenth century, distinguished
two classes of conductors of the then recently discovered
galvanism. The first class comprised those substances, such
as metals, which conduct without chemical change; while

Conductivity and Dissociation of Certain Electrolytes 241

conductors of the second class were decomposed by the passage
of the current. A few years later, by electrolyzing conduc-
tors of the second class, Davy isolated the previously unknown
metals of the alkalies and the alkaline earths. Faraday,^ in
1832, published his laws showing the relation between the
amount of electricity passed through the electrolyte and the
amount of the electrolyte decomposed.

Measurements of the resistance of solutions of electrol3'tes
were soon made by many investigators. Of these early re-
searches those of Hankel,- Becquerel,^ Horsford,* Wiedeman,^
Becker,^ and Beetz^ may be especially noted. Brief dis-
cussions of these and other investigations can be found in
Wiedemann's book.^

The earlier methods were very imperfect. The continuous
current was used, causing polarization of the electrodes, ex-
cept in some special cases, as when the metal of the salt was
used for the electrodes. The standard method now used
practically eliminates polarization by using the alternating
current. This method was first developed and used by Kohl-
rausch and his coworkers in a series of researches® that were
far more comprehensive than any preceding investigations.

The dissociation theory of Arrhenius^" imparted new life to
conductivity measurements of electrolytes as affording a
basis for the accurate determination of the degree of ioniza-
tion.

Following Kohlrausch, many investigations in this field
have been carried out, but in most cases with some special
object in view which has limited the scope of the work. The
researches were concerned with a few substances only, or were
confined to a narrow range of temperature and concentra-
tion.

1 Exp. Researches, III, Ser. No. 373 (1832).

2 Pogg. Ann., 69, 255 (1846).

3 Ann. chim. phys., [3] 17, 365 (1864).
■• Pogg. Ann., 70, 238 (184.7).
^Ibid., 99, 225 (1856).

6 Ann. Chem. (Liebig), 73, 1 (1850); 76, 94 (1851).
' Pogg. Ann., 117, 1 (1862).

* G. Wiedemann: Die Lehre von der Elektricitat, Band 1 (Braunschweig, 1882).
9 For brief discussions and references to original publications see Wiedemann:
Loc. cil.

10 Z. physik. Chem., 1, 631 (1881). Scientific Memoirs, Series IV. p. 47.

242 H OS ford and Jones

PURPOSE OF THIS INVESTIGATION

It has seemed desirable to secure conductivity data relative
to all the substances in more common use by the chemist,
and under the conditions of temperature and dilution at which
they are usually employed in chemical work. With this end
in view, a systematic study of the electrical conductivities and
allied relations of acids, bases and salts in aqueous as well as
in nonaqueous solutions, and at various temperatures and
concentrations, has been in progress in this laboratory for ten
years. Six papers^ dealing solely with aqueous solutions
have been published and other investigations are in progress.

The work herein described was undertaken as a continua-
tion of that already carried out on the general problem. It
includes the determination of the electrical conductivities,
temperature coefficients of conductivity and percentage dis-
sociation of a number of inorganic salts at dilutions ranging
from N/2 to N/4096. Some of the measurements were made
over a range of temperature from 0° to 35°, and a part from
35° to 65°.

PREPARATION OF MATERIAI,

The salts used were the purest available. In nearly all
cases Kahlbaum's chemicals were employed. These were re-
crystallized from one to five times, the final crystallizations
in all cases being made from water of special purity or so-called
"conductivity water." Any deviations from this general
procedure are stated in connection with the experimental
data under each salt.

The water used in making up the solutions was prepared
by a modification of the method of Jones and Mackay,^ i. e., by
subjecting the distilled water of the laboratory to three ad-
ditional distillations: first in the presence of potassium di-
chromate and sulphuric acid to oxidize organic matter and re-
tain ammonia, and twice with barium hydroxide to absorb
carbon dioxide. The conductivity of such water varies from

> Jones and Douglas: This Journal, 20, 428 (1901). Jones and West: Ibid., 34>
357 (1905). Jones and Jacobson: Ibid.. 40, 355 (1908). Clover and Jones: Ibid., 43.
187 (1910). White and Jones: Ibid., 44, 159 (1910). West and Jones: Ibid., 44,
508 (1910).

* Z. physik. Chem., 14, 317 (1894). This Journal, 19, 91 (1897).

Conductivity and Dissociation of Certain Electrolytes 243

i.o to 1.5 X io~*. The correction of the molecular conduc-
tivity due to this cause is negligible in the greater concentra-
tions, but was calculated and applied to the conductivity
values obtained for the dilute solutions.

APPARATUS AND METHOD

The Kohlrausch method was used in this investigation.
In the work from 35° to 65° a slide- wire bridge of the usual
type was employed, while measurements from 0° to 35° were
made by means of an improved slide-wire bridge made by
Leeds and Northrup, the wire being about five meters long.
The bridges and resistance coils were standardized by Leeds
and Northrup and also by means of resistances which had been
corrected by the U. S. Bureau of Standards.

The conductivity cells were of the type used and described
by Clover and Jones^ and Jones and West.^ The constants
of these cells were determined at short intervals. In connec-
tion with the work from 0° to 35° the following method of
determining the constants was employed. A 0.02 N solution
of carefully purified potassium chloride was prepared, using
water of special purity. The molecular conductivity of this
solution at 25° was assumed to have Kohlrausch's value of
129.7, and this solution was used to determine the constants
of the cells designed for concentrated solutions. A 0.002 N
solution of potassium chloride was also prepared, and its
molecular conductivity found by means of a cell whose con-
stant had been determined as explained above. This 0.002 N
solution was then used in finding the constants of the cells
intended for the more dilute solutions.

In connection with the work from 0° to 35°, a slightly dift'er-
ent plan was adopted. Solutions of potassium chloride of
0.02 N and 0.002 N concentration were prepared and used as de-
scribed; but a fixed value of 136.5 at 25° was taken for the
molecular conductivity of the 0.002 N solution. This value is
based on repeated measurements made in this laboratory.

So far as possible the initial or mother solution of each salt

1 This Journal. 43, 192 (1910).
''Ibid., 44, 510 (1910).

244 Hosford and Jones

was prepared by direct weighing of the properly purified sub-
stance. If this, was impracticable a mother solution of con-
venient strength was made up and standardized by analysis.
From the mother solution the various concentrations were
prepared by dilution. In the case of the work from o° to 35°,
solutions were made up at 20° and were used without correc-
tions at the various temperatures at which measurements
were made, the correction being less than the known experi-
mental error. When working from 35° to 65° solutions were
prepared at 50°, and a factor was employed in the reduction
of all measurements made at 35° and 65° to correct for the
change in concentration due to change in volume. The cor-
rection factor for the molecular conductivity of solutions
made at 50° and used at 35° is 0.994; for those made at 50°
and used at 65° the value is i .0076. The burettes and meas-
uring flasks used in making up solutions were carefully cali-
brated for the temperature at which they were to be used.

For the work at 0° an ice bath was employed in which the
cells were supported so as to be immersed as deeply as possi-
ble in the crushed ice and water. A shallow vessel filled with
ice and water covered the ice baths when in use. Connections
were made through openings closed with perforated stoppers
carrying the conducting wires. The baths for higher tem-
peratures were properly sheathed with asbestos cement, and
in the case of the 50° and 65° baths efficient covers were pro-
vided to retain the heat. Hot-air engines were used to stir
the baths. It was found easy to keep the temperature of the
baths constant to within o°.02 or o°.03 by hand regulation,
and this method was adopted.

From two to four independent measurements of the con-
ductivity were made for each concentration at each tempera-
ture. If there was not close agreement in the results, or if any
abnormally large errors were suspected, the measurements
were repeated. So far as possible our results were compared
with those obtained by other workers. In most cases there is
reasonable agreement. When wide discrepancies appeared
our work was duplicated.

Concentrations are indicated under the heading V, or

Conductivity and Dissociation of Certain Electrolytes 245

the number of liters which would contain one gram-
molecular weight of the salt. Molecular conductivities are
expressed in Siemens' units. The temperature coefficients
and dissociation were calculated in the usual way. On ac-
count of hydrolysis or other causes, the maximum value of
the molecular conductivity (j^^) was not found for certain
salts at the greatest dilution worked with. In such cases the
dissociation was not calculated.

Ammonium Aluminium Sulphate, NH4Al(S04)2.i2H20

The mother solution was standardized by determining
aluminium as aluminium oxide.

Table I. — Molecular Conductivity

12°.

25°

35 =

80.0

110.9

143 I

168.8

102 .5

143

185

220

128

130. 1

182

238

284

512

162 .2

230

304

365

1024

181. 0

257

342

415

2048

201.8

288

386

485

4096

224.1

322

437

540

Table II. — Temperature Coefficients

0°-12°.5 12°. 5-25° 25°-35'

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

2.47

309

2.58

2-33

2-57

1.80

3-39

2.37

1.88

128

4-49

2.46

I 93

512

589

2-55

2 .02

1024

6.79

2.64

2. 12

2046

7.86

2.73

2.57

4096

8.38

2.60

2-35

Ammonium Chromium Sulphate (Violet Variety),
NHA(SO,)2.i2H20

The mother solution was standardized in the same manner
as in the case of the potassium salt.

246

Hosford and Jones

Table III. — Molecular Conductivity

12*

25°

35°

77-5

106.4

1373

162.7

123.2

1595

188.3

100

140.3

182.2

216.0

128

129

183.0

240.2

285.9

512

165

238.0

321 .0

385 -9

1024

187

272.0

372.0

455-7

2048

211

310.7

428.5

530.0

4096

240

355-6

492.2

617.0

Table IV. — Temperature Coefficients

0°-12''.5

P^r

12°. 5

-25°

25°-35°

md.

nd.

Per

nd. Per

units cent.

units

cent.

units cent.

2.31 2.98

2.47

2.32

54 1-85

74 3

2.35

88 I. 81

16 3

2.39

38 1.86

128

28 3

2.50

57 1-90

512

80 3

2.79

49 2.02

1024

80 3

2.94

37 2.25

2048

90 3

303

15 2.37

4096

307

48 2 . 54

Ammoniurn Chromium Sulphate {Green Variety)

The mother solution was prepared by heating a portion of
the mother solution of the violet variety to 70° for about
seven hours in a stoppered bottle.

Table V. — Molecular Conductivity

0°

12°.

25°

103.6

133-2

162.9

185-3

119. 7

155

190.6

219

136.4

178

220.8

255

128

172.3

228

288.1

336

512

202.6

274

355-7

423

1024

215.6

294

386.2

471

2048

222.0

313

414.0

518

4096

234-4

328

458.1

593

Conductivity and Dissociation of Certain Electrolytes 247
Table VI. — Temperature Coefficients

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

2.37

2.29

2.38

I 79

2.24

1.38

2.26

1.82

2-45

1. 91

128

2.61

2.09

512

2.83

2.37

1024

2.92

2.50

2048

3-37

2.57

4096

3.21

316

Ammonium Copper Sulphate, (NH4)2Cu(S04)2.6H20

The mother solution was standardized by determining the
sulphuric acid as barium sulphate, and the copper was also
determined as copper oxide.

Table VIL — Molecular Conductivity

12°

25°

35"

106.3

146

190.4

225.7

122.7

169

220.7

262.2

153 -5

213

280.2

334-3

128

187.8

262

346 -7

412.6

512

221.6

312

411. 7

495-7

1024

236.0

333

442.6

532-5

2048

246.4

347

463.6

560.0

4096

259 4

367

494.0

597-3

Table VIII. — Temperature Coefficients

0°-12''.5 12°. 5-25° 25''-35'=

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

3-22

3 03

3 50

2.39

3-53

1.85

2.39

1.88

2.48

I 93

128

2.57

1.90

512

2-55

2.04

1024

2.62

2.03

2048

2.66

2.08

4096

2.76

2.09

248

Hosford and Jones

Sodium Ferrocyanide, Na4Fe(CN)Q. 12H2O

The mother solution was made up by direct weighing of the
anhydrous salt.

Table IX. — Molecular Conductivity

0°

12°. 5

35°

136.7

194.9

259.2

313-4

151-3

215-5

287.0

347-7

167

238.5

318

386.2

128

203

289.6

385

464-5

512

234

334-1

446

543-2

1024

253

361.7

482

581.2

2048

266

380.3

504

612.0

4096

275

398.1

527

632.2

Table X. — Temperature Coefficients

0°-12°.5 12°. 5-25° 25°-35"^

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

4.66

3-41

5-14

2.64

5.42

2.09

5-72

2.65

2 . 12

6.40

2.68

2.13

128

7.70

2.66

2.04

512

8.98

2.69

2.17

1024

9.66

2.67

2.05

2048

9.90

2.60

2.14

4096

10.32

2-59

2.00

Table XL — Percentage Disssociation

°.5

25°

5°

49.58

48.96

49.18

49-57

54-45

60.43

128

73.21

512

84.69

1024

91-52

2048

95-62

4096

100

100.00

100

Potassium Sodium Sulphate, KNaSO^

The mother solution was standardized by determining
sulphuric acid as barium sulphate.

Conductivity and Dissociation of Certain Electrolytes 249

Table XII. — Molecular Conductivity

12°

25°

88.4

122.5

159 0

189.6

146.6

170.6

209. 1

113

158. I

207.2

249.7

128

179.0

236.1

284.5

512

135

189.6

250.8

301 .0

1024

140

197. I

259.2

3132

2048

140

198.2

261 .4

316.2

4096

144

202.6

267.6

322.1

Table XIII.-

—Temperature Coeffi

dents

0°-12°.5

12°. 5-

25°

25°-35°

Cond. Per

Cond.

Per

Cond. Per

units cent.

units

cent.

units cent.

2.73 3.09

2.92

2.38

3 . 06 I . 92

04 4

.92

1.30

85 I 79

68 3

•93

2.49

25 2.05

128

02 3

-57

2^55

84 2.05

512

32 3

.90

2.58

02 2.00

1024

50 3

•97

2.52

40 2 . 08

2048

58 3

.06

2.55

48 2 . 10

4096

66 3

.20

2.56

45 2.04

Table XIV.-

-Percentage Dissociation

0°

12°. 5

25°

35°

61.26

60.46

5942

58.88

66.60

72.

63 75

78.31

78.

77 43

128

89.26

88.

88.23

512

93-97

93-

93 72

1024

97-57

97-

96.86

2048

97.64

97-

97.68

4096

100.

100.00

100

Potassium Aluminium Sulphate, KAl(S04)2.i2H20
The mother solution was standardized by determining
aluminium as aluminium oxide.

Table XV. — Molecular Conductivity

12°.

35°

78.9

108.9

140.3

165 3

lOI

140

182

215

128

127

177

232

283

512

158

223

294

358

1024

177

250

332

402

2048

197

281

378

470

4096

218

314

425

528

250

Hosford and Jones

Table XVI. — Temperature Coefficients

0°-12°.S 12°. 5-25° 25°-35°

Cond. Per

Cond.

Per

Cond. Per

units cent.

units

cent. units cent.

2.40 3.04

2-51

2.30 2.50 1.78

17 313

■31

2-35 3

35 1-84

128

01 3.14

•42

2.49 5

08 2 . 18

512

19 3.27

.69

2.54 6

34 2.15

1024

81 327

■57

2.62 7

01 2. II

2048

74 3 41

■73

2.74 9

16 2.42

4096

67 351

.86

2.82 10

33 2.43

Potassium Nickel

Sulphate, K^NiCSOJ^.eH^O

The mother solution

was standardized by determin

sulphuric

acid as barium

sulphate

and also by determin

nickel as

oxide.

Table XVII.-

—Molecular Conductivity

0°

12°. 5

25° 35°

122.6

170.7

221.9 265.3

155

•4

217.0

283

-8 339

-7

128

187

•5

263.0

344

.8 414

. I

219

309 -3

407

-7 490

•7

1024

235

•5

3312

437

•I 527

. I

2048

249

•5

349-9

463

.0 560

. I

4096

268

367 -9

487

-4 588

. I

Table XVIII.-

—Temperature Coefficients

0°-12°.5

12°. i

-25°

25°-35°

Cond. F

nd.

Per

nd. Per

units cent.

units

cent. units cent.

385 314

4. 10

2 . 40 4 . 34 I . 96

4-93 3

■32

2-45 5

59 I 97

128

6.04 3

■54

2.48 6

93 2.01

512

7i8 3

.87

2.54 8

30 2 . 04

1024

7.66 3

■47

2.56 9

00 2 . 06

2048

8.03 3

■05

2.59 9

71 2.09

4096

8.57 3

■56

2.60 10

07 2.07

Table XIX.-

-Percentage Dissociation

0°

12°. 5

25° 35°

47.01

46.40

45-53 45-11

58.

23 57-

]

71-

70.

74 70.

84.

83-

65 83.

1024

90.

89.

68 89.

2048

95-

94-

99 95-

4096

100.

Conductivity and Dissociation of Certain Electrolytes 251

Potassium Chromium Sulphate {Violet Variety),
KCr(SO,)2.i2H20

The mothei

solution

was

stanc

lar

dize

!d by

det

srmining

chromium as chromic oxide and also

by determining sulphuric

acid as barium sulphate.

Table XX.-

-Molecular Conductivity

0°

12°. 5

25° 35»

75-8

105.0

135 3 1594

873

121 .2

157

3 185

99 0

138. 1

179

6 211

128

127.0

179 5

236

7 279

512

161 . 1

232.0

311

5 374

1024

186.6

271 .6

369

6 443

2048

2133

3142

428

8 520

4096

245.8

364.8

500.1 613.9

Table XXI.-

-Temperature Coefficients

0°-12°.5

c'^

12°. 5

-25°

c'^

25°-35°

nd. P

nd.

Per

nd. Per

units cent.

units

cent. units cent.

2.34 3 09

2.42

2.31 2.41 1.78

71 3

38 2

80 I . 78

13 3

40 3

17 1.77

128

20 3

55 4

32 1.82

512

67 3

74 6

30 2 . 02

1024

80 3

89 7

42 2.01

2048

07 3

92 9

18 2.14

4096

52 3

2.28

Potassium Chromium Sulphate (Green Variety)
The mother solution was prepared by heating a portion of
the mother solution of the violet variety to 70° for about seven
hours in a stoppered bottle.

Table XXI L — Molecular Conductivity

0°

12°

25°

35°

lOI .0

130

158.4

179.6

119

154

188. I

213.2

137

179

2195

249 -3

128

177

234

290.6

333 ■ 5

512

210

283

359 1

426.6

1024

229

310

399 6

479.0

2048

247

339

441 -3

539 I

4096

273

379

500.3

616.2

252

Hosford and Jones

Table XXIII. — Temperature Coefficients

128

512

1024

2048

4096

Cond.
units

2-33
78

2.31

2-33
2.41

2-55
2 .76
2.83

Cond.
units

7 . 40 3 . 00
"50 3- II

73
22

50
05
10

14 2.40
967 2.55

Per
cent.

1.74
1.77
1.80
I .92
2.13
2.28

Cond.

Per
cent.

34
33
36
48
88

22
32

Calcium Formate, Ca(00CH)2
The mother solution was standardized by determining
calcium as the sulphate.

Table XXIV. — Molecular Conductivity

12°

25°

58.4

•7

107. I

128.6

. 2

•4

1245

149 -7

■4

115

■3

1531

184.7

128

131

174-3

211 .6

512

•7

135

■5

181. 9

223.5

2048

lOI

•4

144

190.4

230.6

4096

lOI

■3

145

•4

190.6

229.2

Table XXV.-

-Tem

perature Coeffic

ients

0°-12''.5

12°

.5-25°

25°-35°

md. F

and.

Per

nd. Per

units cent.

units

cent.

units cent.

86 3.19

2.03

2.49

15 2.01

18 3

2.41

2-55

52 2.02

70 3

3.02

2.62

16 2.06

128

12 3

3-45

2.63

73 2.14

512

18 3

371

2.74

16 2.29

2048

46 3

3.66

2.53

02 2 . 1 1

4096

53 3

362

2.49

86 2 . 03

Table XXVI.-

—Percentage Dissociation

0°

12°. 5

25°

35°

57 65

56.19

56.11

66.

64.

80.

128

90.

512

93-

2048

100.

100

4096

100.

100

Conductivity and Dissociation of Certain Electrolytes 253

Calcium Chr ornate, CaCrO^

The mother solution was standardized by titrating with
ferrous ammonium alum.

Table XXVII. — Molecular Conductivity

0°

120

35°

57-7

80.9

105.8

125.4

64.6

118

•5

140.9

72.2

lOI

•4

133

. I

158.2

128

91.2

126

167

•5

200.8

512

106.7

150

198

•7

239 -5

1024

III .6

157

208

253 -3

2048

114. 4

160

214

264.0

4096

116. 1

162

•5

216

. I

261.6

Table XXVIII.

— Temperature Coefficients

0°-12°.5

12°. 5-

-25°

25°-35°

nd. F

ond.

Per

)nd. Per

units cent.

units

cent.

units cent.

1.85 3.21

I 99

2 .46

96 1-85

06 3

•25

2.49

24 1.89

33 3

■54

2.51

51 1.89

128

86 3

•25

2.56

33 1-99

512

46 3

.90

2.60

08 2 . 05

1024

66 3

. 12

2.62

45 2.13

2048

71 3

.26

2.65

00 2.34

4096

71 3

.29

2.64

55 2. 1 1

Table XXIX-

—Percentag

e Dissociatic

0°

12°. 5

25°

35°

49.70

49.78

48.96

47 94

55-

54-

62.

61.

128

78.

77-

512

92.

1024

96.

2048

98.

100

4096

100

100.

Zinc Nitrate, Zn(N03)2.6H20

The mother solution was standardized by determining zinc
as zinc oxide.

2 54

Hosford and Jones

Table XXX. — Molecular Conductivity

12°

25°

35°

80.6

no. 8

146.6

171.2

87.6

121 .2

157-2

188.5

100. 0

139.2

182. 1

219.0

128

no. 4

154- 1

202.6

243 -5

512

114. 1

164.9

210. 1

254 -3

1024

117. 1

165.0

216.6

261.3

2048

120.4

169.2

222.4

270.2

4096

124.4

1750

229. 1

279.4

Table XXXI.-

—Temperature Coefficients

0'>-l2°.S

(

12°.

5-25°

25°-35°

ond. Per

2ond.

Per

)nd. Per

units cent.

units

cent.

units cent.

2 . 42 3 , 00

2.86

2.58

2 . 46 I . 68

69 3

.88

2.38

13 I 99

34 3

•43

2.46

69 2.03

128

50 3

.88

2.52

09 2 . 02

512

06 3

,62

2.20

42 2.10

1024

83 3

■13

2.50

47 2.06

2048

90 3

.26

2.52

78 2.15

4096

05 3

■33

2.47

03 2 . 20

Table XXXII.

— Percentage Dissociation

12°. 5

25°

35°

64.79

6331

63 -99

61.27

70.

69.

80.

59-

128

88.

512

1024

94-

2048

96.

4096

100.

100

Zinc Acetate, Zn(C2H302)2

The mother solution was standardized by determining zinc

as zinc oxide.

Table XXXIII

— Molecular Conductivity

0°

12°. 5

25°

35°

27.8

38.0

48.0

550

■7

52.2

66.6

77.2

•5

78.6

103.0

122.4

128

100.7

134 -2

162. 1

512

II3-7

153 -2

185.5

1024

116. 1

156.7

191.6

2048

120.8

163.2

200. 1

4096

121

•3

163

•4

201

• I

Conductivity and Dissociation of Certain Electrolytes 255
Table XXXIV. — Temperature Coefficients

128

512

1024

2048

4096

Cond.
units

0.81
I. 16
1.85
2-45
2.81
2 .90

cent.
2 .91

Cond.
units

0.80

I 15

I 95

2.68
3 16

3-25

Per

cent.

2 . II

2 .20
2.48

2.66

2.78
2.79

3.01 3.61 3.39 2.81 3.69

3.00 3.58 3.37 2.78

Per

cent.

I .46

I 59
1.88
2.08
2 . II
2.23
2.26
77 2.31

Cond.
units

0.70

I
I
2
3

Table XXXV. — Percentage Dissociation

128

512

1024

2048

4096

100

Lead Acetate, Pb(C2H302)2.3H20

The mother solution was standardized by determining
lead^as lead sulphate.

Table XXXVI. — Molecular Conductivity

0°

12°. 5

25°

35°

II .2

16.4

22 . I

27.0

23 3

31.2

37.8

41.4

54 9

66.2

128

66.3

87.1

104.2

512

92.7

123. I

146.2

1024

108.2

139 I

167.2

2048

119. 4

156.8

189. 1

4096

124.6

165.5

198.7

256 H OS ford and Jones

Table XXXVII. — Temperature Coefficients

0»-12«'.5 120.5-25" 25''-35"

Cond.

Per

2ond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

0.41

3.66

0.46

2.81

0.49

2.22

0.58

0.63

2.70

0.66

2. 12

I .01

1.08

2.61

I 13

2.06

128

I 59

1.66

2.50

I. 71

1.96

512

2.19

2-43

2.62

2.31

1.88

1024

2.70

2.47

2.28

2.81

2.02

2048

2.81

2.99

2.50

3 23

2.06

4096

2.94

327

2.62

332

215

Table XXXVIII.— Percei

itage Dissociation

12". 5

25°

35°

13 16

13 -35

1024

2048

4096

100

A mmonium A luminium Sulphate , NH4Al(S04)2.i2H20
The mother solution was standardized by determining sul-
phuric acid as barium sulphate.

Table XXXIX. — Molecular Conductivity

35°

65°

168.8

203

236.5

202.3

247

288.0

261.5

325

384.8

128

284.8

347

426.3

512

365 9

477

573-5

2048

485 -8

643

8315

Table XL.— Temperature C

Coefficients

35 °-50

50°-65°

Cond

Per

Cond.

Per

units.

cent.

units

cent.

2.31

1-37

2.20

1.08

3.01

1.48

2.70

1.09

4.29

1.64

3 93

I .21

128

4.18

I 47

525

I-5I

512

7-44

2.03

6.40

1-34

2048

10.49

2.16

]

2.56

1-95

Conductivity and Dissociation of Certain Electrolytes 257

Disodium Phosphate, HNajPO^.iaHjO
The mother solution was standardized by determining the
phosphoric acid as magnesium pyrophosphate.

Table XLI. — Molecular Conductivitv and Dissociation

t^v

l>-v

l^v

141

8 61.8

184. 1

61.5

228.0

60.6

176

8 770

228.2

76.3

287.9

76.6

128

206

5 90.0

269.0

89.8

334-4

88.9

512

224

3 97-8

292.7

97.8

376.1

100. 0

2048

229

5 100. 0

299 -3

100. 0

(355 -4)

ble XLIL-

-Temperature Coefficients

35°-50°

S0°-65°

Cond.

Per

Cond.

P«:

units

cent.

units

cent.

2.82

1.99

2.93

•59

3-43

1.94

398

•74

128

417

2.02

436

.62

456

2.03

5.56

.90

2048

465

2.03

4.65

Sodium Tetraborate, Na2B407.5H20
The mother solution was standardized as the anhydrous
salt.

Table XLIIL — Molecular Conductivity and Dissociation

141

3 70.9

182.8

67.6

2313

64.4

157

I 78.8

204.0

75-5

256.2

71 3

128

172

4 86.5

224.1

82.9

281.6

78.4

512

186

7 93-6

247.8

91.7

316.7

88.1

2048

199

4 100. 0

270.3

100. 0

359-3

100. 0

Tab

le XLIV.-

-Temperature Coefficients

35°-50°

50°-65°

Cond.

Per

Cond.

units

cent.

units

cent.

2.77

1.96

3 23

•76

313

1.99

3-48

]

3-45

2.00

383

-71

,12

4.01

215

4-59

-85

2048

4-73

2.37

5-93

.19

258

Hosford and Jones

Potassium Aluminium Sulphate, KAl(S04)2.i2H20
The mother solution was standardized by determining sul-
phuric acid as barium sulphate.

Table XLV. — Molecular Conductivity

35°

50°

65°

142.3

172.5

196. I

165.3

207.5

240.6

215.7

255.1

317.4

128

283.7

356.9

426.2

512

358.3

446.9

557-1

2048

470.0

626.4

796.4

Table XLVL—

Temperature Coefficients

35°-50° 50

°-65°

Cond.

Per Cond.

Per

units

cent. units

cent.

2.01

I. 41 1.57

0.87

2.81

1.70 2.21

1.06

2.63

1.22 4.15

1.63

128

4.88

1.72 4.62

1.29

512

5 91

1.65 7-35

I .64

2048

10.42

2.22 11.33

I. 81

Potassium Sulphocyanate, KCNS
The mother solution was prepared by direct weighing.
Table XLV II. — Molecular Conductivity and Dissociation

l^v

liV

„

fiV

„

127

6 79.2

160.2

77.6

191 . I

76.2

132

9 82.4

166.7

80.8

201.8

80.4

142

3 88.3

179.6

87.0

219.6

87.5

128

149

3 92.6

190.0

92.1

232.4

92.6

512

153

7 95.4

192.6

93.3

239.3

95-4

2048

161

2 100. 0

206.4

100. 0

250.9

100. 0

rabl

e XLV III.-

-Temperature Coefficients

35°-50

50°-65°

Cond.

Per

Cond.

Per

units

cent.

units

cent.

2.17

1.70

2.06

2.25

I .69

2.34

2.49

1.75

2.67

2.74

1.84

2.83

2.60

1.69

3. II

2048

3.01

1.86

2.97

Conductivity and Dissociation of Certain Electrolytes 259

Monopotassium Phosphate, H2KPO4

The mother solution was standardized by determining phos-
phoric acid as magnesium pyrophosphate.

Table XLIX. — Molecular Conductivity and Dissociation

35° 50° 65°

128

512

2048

310
380
424
452
471

63.8
78.1
87.2
93 o
96.9

391.6
481.2
537-6
573 I
599-9

8192 486.4 100. o 621.4 lOO.O

l^v

477.2

61.

588.4

75-

661.2

84.

708.2

90.

740.9

779-4

100.

Table L. — Temperature Coefficients

Cond.

Per

Cond.

Per

imits

cent.

units

cent.

5-41

1-74

5-71

I .46

6-75

1.78

7-15

I 49

128

7-55

1.78

8.24

I 53

512

8.05

1.78

9.01

1-57

2048

8.56

I. 81

9.40

1.58

8192

9.00

1.82

10.53

1.69

Potassium Acetate, KCjHjOj

The mother solution was standardized by determining
potassium as the sulphate.

Table LL

128

512

2048

-Molecular Conductivity and Dissociation

35° 50° 65°

l>-v

94-4
102 .7
112 .0
118.7
125.2
123-3

/<!/

V-V

75-40

125-6

78-84

142 . 1

72.

82.03

131. 6

82.61

160.4

81.

89.46

147.0

92.28

180.8

91-

94-81

154-6

97 05

184-5

93-

100.00

159-3

100.00

194.9

98.

157-7

197.0

100.

26o

Hosjord and Jones

Table LI I. -

-Temperature

Coefficients

35'

-50°

50°-65

Cond.
units

Per
cent.

Cond.
units

Per
cent.

4
8

32
128

2.08

1-93
2-33

2.40

2.20

1.88
2.08
2.02

I . 10
1.92
2.25
2.CX)

0.88
1.46

1-53
1.29

512
2048

2.27
2.29

1. 81
1.86

2.37
2.62

I 49
1.66

Calcium Chloride, CaCl2.6H20

The mother solution was standardized by determining cal-
cium as carbonate and chlorine by Mohr's method.

Table LIII. — Molecular Conductivity and Dissociation

35° 50° 65°

l^v

t^v

189

I 63.41

237-7

62.22

290.4

61.16

208

I 69.78

258.5

67.67

318.7

67.12

242

0 81.16

306.5

80.24

378.5

79.72

128

267

I 89.57

340.8

89. 21

418.9

88.22

512

283

5 95 07

362.4

94.87

452.5

95 30

048

298

2 100. 0

382.0

100.00

474.8

[OO.OO

Table LIV . — Temperature Coefficients

35°-50°

50°-65°

Cond.

Per

Cond.

Per

units

cent.

units

cent.

3 24

I. 71

3-51

336

1.62

4.01

4 30

1.78

4.80

4.91

1.84

5-21

5 26

1.86

6.01

2048

5-59

1.88

6. 19

Magnesium Chloride, MgCl2.6H20

The mother solution was standardized by determining
magnesium as magnesium pyrophosphate and chlorine by
Mohr's method.

Conductivity and Dissociation of Certain Electrolytes 261

Table LV. — Molecular Conductivity and Dissociation

35° 50° 65°

V Hv

!tv

X fly

4 179

62 . 17

228.0

09 280.6

60.27

8 196

67 -95

249.7

91 303.8

65 25

32 231

80,08

294.7

97 364 -8

78.35

128 249

86.37

311. 8

55 401.6

86.25

512 269

91 .20

348.3

33 433 • I

93.02

048 289

100.00

373-2

100

00 465 . 6

100.00

ableLVL —

Temperature

Coefficients

35°-5C

50°-65°

Cond.

Per

Cond.

Per

units

cent.

units

cent.

3.21

1.79

351

1-54

3-55

I. 81

3-61

1-45

4.21

1.82

4.67

1-58

128

413

1-65

5-93

1.90

512

5 23

1.94

5.65

1.62

2048

5.60

1.94

6.16

1.65

Manganese Sulphate, MnS04.4H20
The mother solution was standardized by determining
manganese as manganous pyrophosphate and sulphuric acid
as barium sulphate.

Table LVII. — Molecular Conductivity

V 35° 50° 65°

4 78.0 88.0 108.3

8 92.6 112. 8 130.0

32 128.5 156.4 181. 8

128 166.7 204.1 241.9

512 219.4 277.5 338.7

2048 246.0 326.7 404.6

Table LVII L-

-Tempet

ature Coefficients

35°-50

50°-65°

Cond.

Per

Cond.

Per

units

cent.

units

cent.

0.67

0.86

1-35

I 53

1-35

1.46

I 15

1 .04

1,86

1-45

1 .69

1.08

128

2.49

1.49

2.52

1.24

512

3.87

1.76

4.08

1-47

2048

5.38

2.19

5 19

1-59

262

Hosford and Jones

Ferric Chloride, VeC\^.6¥i.^0
The mother solution was standardized by determining iron
as ferric oxide.

Table LIX. — Molecular Conductivity and Dissociation

V l^v

^2/

a f-v

4 214

16.92

269.5

71 327.0

1939

8 276

21.83

346 -9

32 424

33 52

515-8

128 827

65.29

1037.6

75 1512.5

89.71

512 1050

82.94

1405.4

48 1685.9

100.00

048 1266

100.00

1487 -5

100

00 1673.6

^able LX.-

-Temperature Coefficients

35°-

SO"

50°-65°

Cond.

Per

Cond.

Per

units

cent.

units

cent.

3.68

1.72

383

1.42

4.69

1.70

6.07

I 43

128

14.00

1 .69

31-7

3-85

512

23.6

2.25

18.7

1-33

2048

14-7

1.60

12.4

0.83

Chromium Sulphate {Green Variety)
The mother solution was standardized by determining
chromium as chromic oxide.

Table LXI. — Molecular Conductivity

65°

128

160

189.6

183

227

262.9

302

354

■4

417-4

128

433

522

• 7

606.0

512

673

811

. I

977-3

2048

961

1207

1534-7

Table LXII.

—

Temperature Coefficients

35°-

50°

"-es"

Cond

Per

Cond.

Per

units

cent.

units

cent.

2. 12

1.65

I 97

1.23

2 95

1. 61

2-34

1.03

3 49

1. 16

4.20

I. 19

128

5 92

1-34

5-55

1.06

512

9.19

I 37

11.08

1-37

2048

16.45

I. 71

21.79

1.80

Conductivity and Dissociation of Certain Electrolytes 263

Nickel Nitrate, Ni(N03)2.6H20
The mother solution was standardized by determining
nickel as nickel oxide.

Table LXIII. — Molecular Conductivity and Dissociation

128

512

2048

200.8

61.

216.8

65.

260. I

79-

289.7

89.

314.2

95-

3293

100.

252

276

330
369

399
420

60. 1
65.6
78.6

87.9

95-2

1 00.0

306.6

343 • 5
402.4
453-2
494.8
516.0

59-4
66.6
78.0
87.8

95-9
100. o

Table LXIV. — Temperature Coefficients

128

512

2048

Cond.
units

44
97
68
30
70
05

Per
cent.
I
I

I
I
I
1

Cond.
units

3 61
4.48
4.81
5.60

6-34
6.40

cent.

I 43

Nickel Sulphate, NiS04.6H20
The mother solution was standardized by determining
sulphuric acid as barium sulphate.

Table LXV . — Molecular Conductivity

128

512

2048

35
78
93
127
171
219
264

95
115
158
215

278

341

65
III

135
187
259

339

425

Table LXV I. — Temperature Coefficients

512
2048

Per
cent.

44
60

Cond.
units
I .09

I 35
1.97

2.95

4 05

5 63

.14
•17

■25
•37
•45
•65

264

Hosford and Jones

Cobalt Sulphate, CoSO^.yH^O
The mother solution was standardized by determining sul-
phuric acid as barium sulphate.

Table LXVII. — Molecular Conductivity

350

50°

65°

80.0

112. 7

94 9

117

137-5

129. 1

160

189.6

128

172.5

203

256.6

512

229.8

290

346.0

2048

264.8

340

421.6

Table LXVIIL-

-Temperature Coefficients

35°-50

50 "-es"

Cond.

Per

Cond.

Per

units

cent.

units

cent.

I .04

1.30

I. 14

I. 19

1.49

1-57

1-35

I 15

2.06

1-59

1.97

1.23

128

2.06

1,19

3-55

1.74

512

4.06

1.76

3-49

1.20

2048

503

1.90

5 42

I 59

Copper Sulphate, CuS04.5H,0
The mother solution was standardized by determining copper
> copper oxide and sulphuric acid as barium sulphate.

Table LXIX. — Molecular Conductivity

V 35° 50° 65

4 75-6 93-8 107

8 90.5 109. I 124

32 126. I 152.7 173

128 170.7 210.3 247

512 222.7 279.1 337

2048 266.3 343-3 422

Table LXX. — Temperature Coefficients

4
8

32
128
512

2048

Cond.

units

. 21
.24

-77
.64
.76
-13

Per
cent.
.60

Cond.
luiits
0.91
I .02
I. 41

2-45
3 91
5-29

Per
cent.
0.97

0.93
0.92
I. 16
I .40

1-54

Conductivity and Dissociation of Certain Electrolytes 265

DISCUSSION OF RESULTS
Salts Studied from 0° to 33°

The conductivity data (Tables XXIV to XXXVIII) for
the salts derived from diacid bases are of the same general
character. This is best seen by drawing curv^es. Fig. I shows

•I

g 100 -
c3

12 .5°

32 128

512 1024

Concentration
Fig. I— Zinc acetate

2048

conductivity-concentration curves for zinc acetate, which
may be taken as an example of these salts. The diagrams for
the other members of the group are similar, except that the
cun/e is much flattened in the cases of lead acetate and calcium
chromate. This is apparent in Fig. II, which gives the
conductivity-concentration curves for lead acetate.

Lead acetate presents a number of exceptions. In concen-
trated solutions it has the smallest molecular conductivity
and dissociation of any of the salts studied. At infinite
dilution, however, the conductivity is nearly the same at all
temperatures as for zinc acetate. Again, lead acetate shows
the smallest temperature coefficients of conductivity at high
temperatures, and the most rapid increase with dilution of
any of the salts brought within the scope of this investigation.

It has been known that dissociation seems to be nearly
independent of temperature over the range of temperature at

266

Hosjord and Jones

which our work was done, but in general decreases as the tem-
perature rises. This is true of all but four of the salts included
in this investigation.

The four apparent exceptions are potassium acetate, cal-
cium formate, lead acetate and sodium ferrocyanide. Potas-
sium acetate shows no well-marked change in the dissocia-
tion with temperature. Dissociation apparently increases as
the temperature rises in the cases of calcium formate and
sodium ferrocyanide. The same is true of lead acetate in
concentrated solutions. It is intended to study farther the
dissociation of these apparently exceptional salts.

— n — 1~

32 128

1024

Concentration
Fig. II.— I<ead acetate

2048

The complex salt sodium ferrocyanide has high conduc-
tivity and large temperature coefficients of conductivity. The
percentage coefficients are remarkably constant at the various
dilutions. The value for /^^ was hardly reached at the high-
est dilution used, which is probably due, in part at least, to
the breaking down into simpler ions in the more dilute solutions.
Hydrolysis also probably takes place.

The remaining salts studied over the lower range in tem-
perature are double salts and all are sulphates. Ammonium
copper sulphate and potassium nickel sulphate yield very
similar conductivity data. In the stronger solutions the

Conductivity and Dissociation of Certain Electrolytes 267

nickel salt shows the greater conductivity, but at higher dilu-
tion the molecular conductivity and the temperature coeffi-
cients of these two salts are almost identical. Conductivity
in these cases is probably somewhat affected by hydrolysis.

A comparison of the data for the violet and green varie-
ties of ammonium chrome alum shows that in the more con-
centrated solutions the green variety has much higher con-
ductivity. At higher dilutions the reverse is true; the violet
conducts better than the green form. The same general
relation is shown in the case of potassium chrome alum, though
H^ for the green variety does not actually fall below the fjt^
value for the violet form at the highest dilution employed.
This relation between the two varieties appears in Fig. III.

I
4300

32 128 SI2

2048

Concentration

Fig. III. — Ammonium chromium sulphate-

— X X X Violet variety

-(^)—C'y-Gy~ Green variety

The violet form of the chromium salts is transformed into
the green variety by heating the solid salt or its solution for
some time at about 70°. The relation between the two forms

268 Hosford and Jones

has been the subject of many investigations. Monti' ob-
served that the change was accompanied by an increase in
the conductivity of the solution. Recoura,^ in an elaborate
investigation, explained the change as hydrolytic, resulting
in the formation of free sulphuric acid and a basic salt. This
conclusion was confirmed by Whitney.^ Jones and Mackay^
showed by conductivity measurements that the transforma-
tion was continuous as the temperature rose slowly from
37°.5to90°.

In the light of the conclusions of these workers, the explana-
tion of the relative conductivity of the two forms of ammonium
chrome alum would seem to be that the green modification,
by virtue of the hydrolysis caused by heating, has the higher
initial conductivity, but when diluted it is incapable of further
hydrolysis to the same extent as appears to occur in the case
of the normal violet variety.

Salts Studied from 35° to 65°

Before considering in detail the salts studied from 35° to
65° some general relations should be discussed. It was found
by Jones and Ota^ and by Jones and Knight^ that concen-
trated solutions of certain salts in water often show abnormally
great depressions of the freezing point of the solvent. It was
also shown to be true in many cases that the molecular lower-
ing of the freezing point increased from a certain concentra-
tion both with dilution and with increased concentration.
The subject was further studied by Jones and Chambers'
and by Jones and Getman.* It was found that the molecular
conductivities of solutions of the substances which showed a
minimum in the value of the molecular lowering were per-
fectly normal at all concentrations.

1 Z. anorg. Chem., 12, 75 (1896).

2 Ann. chim. phys., [7] 4, 494 (1895).

3 Z. physik. Chem., 20, 40 (1896).

4 This Journal, 19, 103 (1897).
s/6id., 22. 5 (1899).

^Ibid., 22, 110 (1899).

T/fcid, 23, 89 (1900).

s Z. physik. Chem., 46, 244 (1903).

Conductivity and Dissociation of Certain Electrolytes 269

To account for the facts, Jones^ offered the suggestion that
the molecules of the dissolved substance form complex com-
pounds or hydrates with a portion of the water, thus virtually
increasing the concentration of the solution. The freezing
point is thus abnormally depressed. It was also pointed out
that substances which give these abnormal results are often
hygroscopic and that, when dehydrated, they readily com-
bine with water. Jones and his assistants have collected a
large amount of evidence,^ by several independent methods,
which supports the theory of hydration. A method^ was de-
veloped by which the approximate composition of the hy-
drates of many substances was calculated.

It was pointed out by Jones* that the breaking down of the
hydrated molecules, or of the hydrated ions, by a rise in
temperature would diminish the mass of the ion and thus in-
crease the conductivity. The more complex the hydrates
the greater would be the change in hydration and, conse-
quently, the greater the change in conductivity. Therefore
"we should expect to find those ions with the largest hydrating
power having the largest temperature coefficients of conduc-
tivity."^ An examination of the experimental results of Jones
and West® led to the following conclusions :

1. The temperature coefficients of conductivity of aqueous
solutions of electrolytes are greater, the greater the hydra-
ting power of the electrolyte.

2. The temperature coefficients of conductivity of aqueous
solutions of electrolytes are of the same order of magnitude
for those substances having, approximately, the same hydra-
ting power.

3 ■ The temperature coefficients of conductivity, for any
given substance, increase with the dilution of the solution,
and the increase is greatest for those substances with large
hydrating power.'

1 This Journal, 23, 103 (1900).

2 See Hydrates in Aqueous Solution; Carnegie Institution of Washington, Publica-
tion No. 80.

* Ibid., pp. 28-145.

■• This Journal, 36, 445 (1906).

5 Loc. cit., p. 447.

e This Journal, 34, 357 (1905).

■'Ibid., 36, 450 (1906).

270

Hosford and Jones

Similar results were obtained by Jones and Clover.'
The composition of the hydrates formed by some of the
substances which were brought within the scope of this inves-
tigation has been approximately determined by Jones^ and
his assistants. In general, the hydrating power may be taken
as roughly proportional to the amount of water of crystallization.
The substances named in Table LXXI crystallize with little
or no water, and have slight hydrating power. They are
seen to have small temperature coeflEicients of conductivity.
The substances named in Table LXXI I have large hydrating
power and also have large temperature coefficients of conduc-
tivity.

Table LXXI. — Substances with Slight Hydrating Power

Temperature coefficients in conductivity units

Temp, range

KCNS

2.17

2048 3.01

35°-50°

KCHgO^

2.o8

2048 2

35°-50°

Ca(bOCH),

215

2048 4

25°-35°

Zn(aH30,)2

0.70

2048 3

25°-35°

Pb(C3H30,)3.3H30

0.49

2048 3

25°-35°

CaCrO,.2H,0

1.96

2048 5 . 00

25°-35°

Table LXXI I.-

—Substances with Large Hydrating

Power

emperature

coefficients in conductivity units

Temp, range

Ni(N03)2.6H20

3-44

204S 6.05

35°-50°

CaCU.eH^O

2048 5

35°-50°

MgCl2.6H,0

2048 5

35°-50°

Zn(N03)2.6H20

2048 4

25°-35°

FeClg.eHaO

2048 14

35°-50°

KNa^PO^.iaHp

2048 4

35°-50°

H,KPO,

2048 8

35°-50°

Na-^BA-SH^O

2048 4

35°-50°

The values used in these tables are not strictly compara-
ble, since the concentrations and the ranges of temperatures
at which the temperature coefficients were determined are
not the same throughout, but the agreement is sufficiently
close to warrant their use in showing the general relations.

1 This Journal. 43, 215 (1906).

2 Hydrates in Aqueous Solution; Carnegie Institution of Washington, No. 60.

Conductivity and Dissociation of Certain Electrolytes 271

Our results confirm the conclusion of Jones cited above, and
are in perfect accord with the theory of hydration advanced
by him.

The sulphates which we have studied in this investigation
are omitted from Tables LXXI and LXXII, because, as shown
by Jones and his coworkers,^ the sulphates usually show ab-
normal results. In general, sulphates have very small tem-
perature coefficients of conductivity, and appear to have
small hydrating power in solution. There is evidence that
some sulphates, at least, are polymerized in concentrated
solutions.

Sodium tetraborate gives normal conductivity results at
35°, but at higher temperatures the increase in conductivity
with dilution is exceptionally rapid. The salt also has large
temperature coefficients of conductivity, and is undoubtedly
hydrated in solution. Boric acid being little dissociated and,
therefore, a weak acid, the sodium salt would certainly undergo"
hydrolysis. By assuming both hydration and hydrolysis to
take place, the behavior of the salt is easily accounted for.
At the lower temperatures the hydrolysis, due to increasing
dilution, is balanced by the increasing complexity of the
fairly stable hydrates. As the temperature rises the hydrates
break down, while hydrolysis continues unchecked so that
the conductivity increases rapidly.

Calcium and magnesium chlorides give almost identical
results, except that the molecular conductivity of the cal-
cium salt is about ten conductivity units above that of magne-
sium chloride. The latter salt shows greater increase in the
value of /i^ with dilution at 56°, which accords with its greater
hydration, as shown by Jones and Bassett.^

Nickel nitrate also shows large temperature coefficients.
It is known^ to possess marked power of hydration.

The conductivity data for the sulphates of nickel, cobalt,
copper, and manganese are remarkably similar in every re-

' Hydrates in Aqueous Solution; Carnegie Institution of Washington, Publica-
tion No. 60, pp. 80, 136, 148.

- This Journ.il, 33, 555 (1905),

3 Jones: Hydrates in Aqueous .Solution; Carnegie Institution of Washington,
Publication No. 60, p. 78.

272

Hosford and Jones

spect. There is little indication of an approach to //^ at the
highest dilution employed. The temperature coefficients are
not large in concentrated solutions, but increase rapidly with
dilution. This behavior is probably due, in part at least, to
the polymerization of the sulphates in concentrated solu-
tions.

The molecular conductivity of the quaternary electrolytes,
chromium sulphate and ferric chloride, increases very rapidly
with dilution and also, in dilute solutions, with rise in tem-
perature. Hydrolysis undoubtedly plays a prominent part.
In the case of ferric chloride distinct precipitation, due to
hydrolysis, occurred at 65° in the N/8 and N/32 solutions.
The conductivity curve of this salt is interesting (Fig. IV).

2048

1024 %

J 600 1500 1400 1300 1200 1 1 00 1000 goo 800 yoo 600 $00 400 joo 200

Conductivity

Fig. IV.— Ferric chloride

The abrupt bend in the curve for 50° and 65° at the N/512
concentration indicates that the cause of the increasing con-
ductivity— presumably hydrolysis — is rapidly becoming less
effective. It would seem that under these conditions of tem-
perature and dilution the hydrolysis of the salt is very
great.

The ammonium and potassium alums were studied through
both ranges of temperature (o°-35° and 35°-65°), and the

Conductivity and Dissociation of Certain E lecirolytes 273

values of fx^ recorded for 35° were deduced from all the read-
ings made at this temperature. Fig. V shows the conduc-

Conductivity

it
B 8

tivity-temperature curves for potassium alum through the
entire range of temperature. In strong solutions, at ordinary

274 H OS ford and Jones

temperatures, molecular conductivity is nearly a linear func-
tion of temperature ; but at greater dilutions the curve in para-
bolic.^ All of the salts studied in this investigation yield
conductivity-temperature cur\-es of this same general charac-
ter.

The condition of double salts, when in solution, presents a
problem of interest. Investigators have sought for evidence
which would decide whether such salts, when dissolved, break
down into their constituent salts, which then dissociate in the
usual way; or whether they ionize to some extent as salts of
complex acids. Four investigations^ bearing on the general
problem have been carried out in this laboratory. Jones and
Mackay compared the conductivity of certain alums with the
sum of the conductivities of the constituent salts. They
found the conductivity of the alums in dilute solutions to be
almost the same as the sum of the conductivities of the com-
ponents. In concentrated solutions the alums were found to
have a conductivity less than the sum of the conductivities of
the components, and the difference increased with the concen-
tration. The difference was greater than that observed when
mixtures of sulphates incapable of forming double salts were
compared. Similar methods were used by the other workers.
The general conclusion from these researches was that the
double salts in moderately concentrated solutions are not
wholly broken down into the simple salts.

In these investigations the conductivities were measured
at 25°. As we now have at hand conductivity data over a
considerable range in temperature, it appears to be of interest
to apply the method of Jones and Mackay at other tempera-
tures.

The following is the table of Jones and Mackay^ giving the
comparisons for potassium alum at 25°:

1 Jones and Jacobson: This Journal, 40, 402 (1908). White and Jones: Ibid.,
44, 199 (1910).

2 Jones and Mackay: Ibid., 19, 83 (1897). Jones and Ota: Ibid., 22, 5 (1899).
Jones and Knight: Ibid., 22, 110 (1899). Jones and Caldwell: Ibid., 25, 349 (1901).

^ Ibid.. 34, 357 (1905).

Conductivity and Dissociation of Certain Electrolytes 275

LX XIII. —Potassium

Alum, 23°

(Jones

and Ma

Diff.

V K2SO4 Al2(S04)3

Sum/2

KAISO4

Per cent

5 172.7 108.0

140.3

133-9

—4-5

8 183.3 124.2

153-7

149.2

—30

20 205.1 158. I

181. 6

178.3

—1-7

40 220,3 185.7

203.0

202.5

— 0.2

200 252.4 290.4

271.4

269.0

-^.8

400 262 . 2 342 . 6

302.4

305-2

+ 0.9

This may be compared with Table LXXIV, which gives the
corresponding relations at 0°, 35°, and 65°. The values of
the conductivity of aluminum sulphate were kindly furnished
by Miss L. G. Winston. The conductivity values of potassium
sulphate are taken from the work of Jones and West.^

Table LXXIV. — Potassium Alum

Diff.

K2S0

A]2(S04)3

Sum/2

KAISO4

Difif.

Per cent

lOI .9

65.2

83.6

78.9

—4-7

—5-6

117. 9

89-5

103-7

lOI . 2

—2.5

—2.4

128

131-9

121. 8

126.9

127.6

+ 0.7

+ 0.5

512

142.7

164. I

153-4

. 35°

158.8

+ 5-4

+ 3-5

220.3

137.2

178.8

165.3

—13-5

—7-5

259-7

197. I

228.4

215-7

—12.7

—5-5

128

296.9

274.1

285.5

283.7

— 1.8

— 0.6

512

319.6

388.1

353-9
65°

358.3

+ 4-4

+ 1.2

332-8

188.4

260,6

240.6

— 20.0

—7-7

400.0

264.6

332.3

317.4

—14.9

—4-5

218

456.2

387.6

421.9

426.2

+ 4-3

— 1 .0

512

500.7

581.6

541 -I

557-1

+ 16,0

+ 2.9

Our values for "difference" in per cent, are of the same
order of magnitude as those obtained by Jones and Mackay,
and confirm their conclusions. It is noticeable that the per-
centage differences are nearly the same at the various tem-
peratures. This may be regarded as evidence that the break-
ing down of potassium alum in solution is little affected by
temperature, which, from other evidence, is known to be
true of dissociation in general.

1 This Journal. 34, 357 (1905).

276

Hosford and Jones

In Table LXXV, from the work of Jones and Caldwell,'
the conductivity of the double salt, potassium nickel sulphate,
is compared with the sum of the conductivities of the com-
ponents, all measurements being made at 25°. Table LXXVI
shows the same relations for this salt at 0° and 35°. The
values for the conductivity of nickel sulphate given in Table
LXXVI are taken from the work of Jones and Jacobson.^

Table LXXV.-

-Potassium Nickel Sulphate, 25° {Jones and

Caldwell

Di£f.

KzSO^

NiS04

Sum

K2Ni(SO«)2

Diff.

Per cent

182.4

77-9

260.3

219-5

—40.8

-15-6

220.3

109.0

329-3

291 .6

—37-7

—II. 4

237 -9

122.8

360.7

323 ■ 7

—37-0

—10.3

400

262.2

173 I

435-3

400.2

—35-1

— 8.0

800

273.0

194.8

467.8

438.0

—29.0

— 6.2

Table LXXVL-

—Potassium Nickel Sulphate

0°

Diff.

K2SO4

NiSOi

Sum

KzNi (304)2

Difif.

Per cent

101.9

40.4

142-3

122.6

—19.7

-13-8

117. 9

54-8

172.7

155-4

—17-3

— 10.0

128

131-9

73-9

205.8

187-5

-18.3

-8.9

512

142.7

93-1

235-8

219.6

—16.2

— 6.9

1024

145-0

100.4

245-4

35°

235-5

— 9 9

— 4.0

219.8

90.9

310.7

268.3

—42.4

—13-7

256.9

123.0

379-9

339-7

— ^40.2

— 10.6

128

291.0

168.4

459-4

414. 1

—45-3

— 9-8

512

318.4

213-5

531-9

490.7

—41.2

— 7-7

1024

325-0

234.6

559 6

527-1

—32-5

— 5-8

The percentage " dififerences " at 0° and 35° agree closely
with those found by Jones and Caldwell at 25°, showing that
the relations which they established as holding at 25° are also
true at higher and lower temperatures. Our results also ac-
cord with the general law that dissociation is nearly independ-
ent of temperature.

' This Journal, 26. 349 (1901).
2 Ibid., 40, 390 (1908).

Conductivity and Dissociation of Certain Electrolytes 277

SUMMARY

1. The molecular conductivities of fifteen inorganic salts
from 0° to 35°, and of sixteen inorganic salts from 35° to 65°,
were measured by the Kohlrausch method. The tempera-
ture coefficients of conductivity, both in conductivity units
and in percentages, were calculated for these salts through the
ranges of temperature above stated. The percentage disso-
ciations were also calculated in all cases where the data were
sufficient.

2. Jones and his coworkers^ have shown that the ions of
an electrolyte are hydrated in aqueous solutions, and that the
complexes break down with rise in temperature, thus increas-
ing the conductivity. If this is true, substances of large
hydrating power should have large temperature coefficients
of conductivity. Jones- showed this to be true for the sub-
stances studied by Jones and West.^ The substances which
we have studied show the same relations and our results are
in perfect accord with the theory of hydration.

3. Hydrolysis is evidently a frequent cause of abnormally
great conductivity. It is increased both by dilution and by
rise in temperature.

4. Another probable cause of abnormally rapid increase in
conductivity is decrease in polymerization. There is evidence
that sulphates are polymerized in concentrated solutions.

5. Observers have found an increase in the conductivity
of a solution of a chromium salt when it is changed from the
violet to the green variety. Our results show that while the
conductivity is increased in concentrated solutions by this
change, the increase is relatively less at higher dilutions.
The conductivity of the green variety may even fall below
that of the violet variety. This would appear to show that
the green variety is not as susceptible to hydrolysis by dilu-
tion as is the normal violet form.

• Hydates in Aqueous Solution; Carnegie Institution of Washington, Publication
No. 60.

2 This Journal, 36, 445 (1906).

3 Ibid.. 34, 357 (1905).

278 Bingham and Durham

6. Jones and his coworkers* found that the conductivities
of alums and other double salts were less than the sum of the
conductivities of the constituent salts. They inferred that
double salts exist as such to some extent in concentrated
solutions. Their work was done at 25°. We have made
similar comparisons at other temperatures, and find that the
relations pointed out by them as holding at 25° also manifest
themselves from 0° to 65°. In addition, our results show
that the breaking down or dissociation of double salts, like
dissociation in general, is little affected by temperature.

The following general relations, established by previous
investigators, are true of the salts which we have studied:

7. The temperature-conductivity curves for concentrated
solutions are nearly straight lines; at higher dilutions the
curves are often parabolic.

8. The percentage temperature coefficients increase with
dilution, but decrease with temperature. Temperature co-
efficients in conductivity units increase with dilution.

9. Dissociation decreases with temperature. Four salts
among those studied seem to be exceptions to the rule.

Work along the above lines will be extended to all of the
more common electrolytes, organic and inorganic.

Johns Hopkins University
May, 1911

THE VISCOSITY AND FLUIDITY OF SUSPENSIONS OF
FINELY-DIVIDED SOLIDS IN LIQUIDS

By Eugene C. Bingham and T. C. Durham
[TWELFTH COMMUNICATION BEARING ON THIS SUBJECT]

It was pointed out in a recent paper ^ that there is very-
little literature upon the subject of the viscosity of suspensions.
It has long been observed that dust must be carefully ex-
cluded from liquids whose viscosities are to be measured by
the method of Poiseuille. Among more recent workers may

> Jones and Mackay: This Journal, 19, S3 (1897). Jones and Ota: Ibid., 22, 5
(1899). Jones and Knight: Ibid., 22, 110 (1899). Jones and Caldwell: Ibid.. 26,
349 (1901).

2 Bingham and White: J. Am. Chem. Sec., 33, 1257 (1911).

Suspensions of Solids in Liquids 279

be mentioned Thorpe and Rodger/ who took great pains in
this matter. On the other hand, Friedlander^ noted that
suspensions of rosin in water which were highly opalescent
affected the viscosity scarely any. Bose^ notes that a sus-
pension of finely divided quartz particles in a mixture of
bromoform and water resembles an emulsion, but that the
viscosity of the mixture, when dilute, is but little greater
than that of the bromoform and water mixture alone. The
apparent contradiction is probably explained by the fact that
dust is composed partly of lint which is of such large dimen-
sions that it is unable to pass through the capillary and hence
forms an obstruction at the entrance to the latter. On the
other hand, the particles of rosin in the second case must
be exceedingly small, for we have observed that they remain
suspended even after standing for months at room tempera-
ture. But the concentration of the suspension investigated
by Friedlander was very small, being less than one-tenth of
one per cent. That the presence of a solid, whose fluidity
must be regarded as practically zero, should not lower the
fluidity of a liquid in which it might happen to be sus-
pended, even although the concentration might be small,
seemed so very remarkable that it appeared worth while to
make it the subject of special study. Especially is this true
since only with a knowledge of these effects can a theory of
paints and lubricants be established. The work of Schwedoff*
on the rigidity of liquids may be referred to in this connection.
As materials for study, it seemed desirable to use substances
which could be obtained in a very finely divided condition;
but the materials must be insoluble in the liquids employed,
so as to avoid the complications which would arise if partial
solution took place. It was, of course, desirable to select
substances which would stay in suspension for a considerable
time, hence large differences in specific gravity between the
liquid and solid were undesirable. The rate of settling, how-
ever, is by no means dependent solely upon the difference in

1 Phil. Trans., 186, A, 411, 414, 415, 444 (1894).

- Z. physik. Chem., 38, 430 (1901).

3 Physik. Z., 8, 347 (1907).

•* Rapport au Congrfes de Physique, 1, 478 (1900).

28o Bingham and Durham

specific gravities, or upon the fineness of subdivision of the
solid, but it is dependent very largely upon circumstances
which would at first seem to be quite extraneous. These
conditions will be discussed later.

The Richmond deposits of infusorial earth are composed
principally of silica and the material is very light. When
mixed with a considerable quantity of distilled water, a por-
tion of the material will remain in suspension for a very long
time, not becoming clear after several weeks. It was observed
that tap water or distilled water containing a very little salt,
such as sodium, potassium, or ammonium chlorides, would
cause this suspension to coagulate and settle rapidly. A
considerable amount of material was elutriated, the portion
remaining in suspension after fifteen minutes being siphoned
off and evaporated to dryness. This material was pulver-
ized and bolted through a 200-mesh sieve and preserved in
the dry condition. The material was cream-colored. It
darkened somewhat on heating to a high temperature, pre-
sumably due to the presence of organic matter.

A supply of the finest English china clay was obtained
from the Chesapeake Pottery Co., of Baltimore. On drying,
the material nearly all passed through a 200-mesh sieve, with-
out other preliminary treatment. It seemed impossible to
elutriate the material satisfactorily by suspending it in
pure water, as it coagulated and settled en masse. But it
was noticed that the effect of potassium chloride and am-
monium chloride on the clay suspension was exactly the
opposite of that on the infusorial earth, the coagulation
being much less marked.

The graphite sold under the name of "Aquadag" for lubri-
cation was obtained from the International Acheson Graphite
Company, of Niagara Falls, N. Y. It is a suspension in water
which does not settle and hence is suitable for our purposes.

As media for the suspensions, water and alcohol seemed to
be the most suitable liquids for our present purposes. The
water was redistilled from both chromic acid and barium hy-
droxide and was quite dust-free. The alcohol was dehy-

Suspensions of Solids in Liquids 281

drated with lime, distilled, and, after adding a little metallic
sodium, redistilled through a Glinsky distilling head.

It seemed unwise to attempt to use the apparatus referred
to in a recent paper, after the method of Poiseuille. With a
capillary tube of about one- tenth mm. diameter there would
be danger of permanently plugging the tube. A tube of larger
bore could be used if it were sufficiently long and the
pressure not too high. The Ostwald type of viscometer
semed best suited for our present purposes on account of its
extreme simplicity, but it was observed by one of us several
years ago that when this type of viscometer was calibrated
with one liquid it would not give perfectly satisfactory values
for the viscosities of other liquids whose viscosities are known.
The same thing has been noted by other workers, among whom
we may cite Alexander Findlay. Nevertheless viscosities can
now be measured with great accuracy, as has been abundantly
proved by the work of Thorpe and Rodger and others. Al-
though we could not hope to attain such a high degree of
accuracy with the solid suspensions, it seemed worth while
to make a study of the use of the Ostwald viscometer.

An imported viscometer was tested with water, alcohol,
and ether over a series of temperatures. Care was taken to
have the capillary uniformly vertical, and the temperature of
the bath was kept constant to within a tenth of a degree.
Impurities were carefully excluded from the apparatus. The
time was recorded with a high grade stop-watch which was
tested for its reliability. The results are given in the follow-
ing tables :

'.e I— The

Viscosity

of Water as

Measured by

Mr. Adrian

Thomas with the Ostwald Viscometer ' ' A

Time in

Values

seconds

by other

Percentage

Temp.

Average

Viscosity

observers

difference

lO.O

79 2

0.01303

0.01307

0.30

15.0

69.2

O.OII34

0.01 140

—0.52

25.0

550

Standard

0 . 00895

30.0

49 2

0 . 00804

0.00801

+ 0.37

36.0

44.2

0.00727

0 . 00706

+ 2.08

40.0

40.9

0.00667

0.00655

+ 1 79

45 0

37.8

0.00613

0.00599

+ 2.44

50.0

35 0

0.00567

0.00551

+ 2.64

282 Bingham and Durham

In this and the following tables, the time given is the aver-
age of two or more observations.

Table II — The Viscosity of Ethyl Alcohol as Measured by Mr.
Adrian Thomas with the Ostwald Viscometer "A "

Temp.

Time in
seconds
Average

Viscosity

Viscosity
by other
observers

Percentage
difference

8.0

II5-5

O.OI515

0.01513

+ 0.1

9.0

112. 3

0.01470

O.OI481

lO.O

1 10. 3

0.01442

0.01449

15.0

100.8

O.OI3IO

0.01320

20.0

92 .0

O.OII9I

O.OII92

25.0

84.2

0.01085

O.OIO9I

350

71. 1

0 . 00904

0 . 00908

450

60.2

0.00758

0.00762

—^

Table Ill-

-The Viscosity of Ethyl

Ether as Measured by il

Adrian Thomas with the Ostwald Viscometer

"A"

Temp.

Time in
seconds
Average

Viscosity

Viscosity
by other
observers

Percentage
difference

8.0

24.9

0 . 003002

0 . 002640

+ 16.0

10.0

245

0.002915

0.002585

+ 10.0

15.0

23.8

0 . 002804

0 . 002465

+ 130

20.0

23-5

0.002751

0.002345

+ 16.0

25.0

22.5

0.002613

0.002232

+ 12.0

30.0

21.8

0.002519

0.002120

+ i6.o

We are indebted to Mr. Adrian Thomas for these measure-
ments but quite similar results have been obtained by one of
the authors of this paper, working with this type of viscometer.

From the above tables it is clear that the results are not
satisfactory. It may be properly objected that the visco-
meter used was unsuited to the measurement of the viscosity
of ether, but it must be noted that the deviations for water
are of the same kind, although not so great. In other words,
the deviations here shown are characteristic of this form of
viscometer. The time of flow is too small, in all of the meas-
urements, for accurate work, since the time can be estimated
only to two-tenths of a second. In the case of ether it is
possible that the velocity of the liquid in the capillary was so

Suspensions of Solids in Liquids 283

great that the motion became turbulent, instead of linear,
as should be the case. But by far the greatest source of error
with this form of apparatus arises from the fact that no cor-
rection is made for the kinetic energy imparted to the liquid
on entering the capillary. That this correction is indeed
adequate to explain the discrepancies observed may be in-
ferred from the formula used by Thorpe and Rodger in calcu-
lating absolute viscosities,

■q = TtrHp/Slv — pv/Sitlt

where r is the radius of the capillary, i the time of flow, p the
difference of pressure in dynes per sq. cm., I the length of the
capillary, v the volume of liquid passing through the tube
and p the density of the liquid. The second term is the cor-
rection term for kinetic energy. In the apparatus in ques-
tion the volume of liquid is about four cc. and the length of the
capillary is about seven cm. This gives a correction of 0.007,
which is more than sufhcient, but we must recall that no cor-
rection was included when the instrument was standardized.

An attempt was now made to construct an apparatus with
which the correction might be reduced to much smaller dimen-
sions by using a longer and smaller capillary. The volume
of liquid used in the Ostwald viscometer, 4 cc, is larger than
is necessary since, if the volume can be read to one one-thou-
sandth cc, an accuracy of one-tenth of one per cent, can be
obtained with only one cc. of liquid. Thus the time of flow
can be reduced.

In the new apparatus "B" made by Mr. Adrian Thomas,
the volume of liquid was 2.3 cc. and the length of the capillary
18.4 cm. The results obtained are given in the following
tables :

284

Bingham and Durham

Table IV-

-The Viscosity of Water

as Measured by

Mr. Adn

Thomas -with Viscometer "B"

Temp.

Time in
seconds
Average

Viscosity

Viscosity
by other
observers

Percentage
difference

25.0

630.4

0.00892

0.00895

—0.3

30.0

565 -4

0 . 00802

0.00801

+ 0.1

350

509 I

0.00724

0.00725

0. I

40.0

462.5

0.00657

0.00655

+ 0.3

45 0

424 -3

0.00601

0.00599

+ 0.3

50.0

391 -5

0.00553

0.00551

+ 0.3

550

360.1

0 . 00506

0 . 00508

0.2

60.0

334 I

0 . 00468

0 . 00470

0.2

65.0

3II-4

0.00435

0.00437

0. 2

70.0

291.3

0 . 00406

0 . 00407

0.2

750

274.0

0.00381

0.0

80.0

258.2

0 . 00358

0.00357

+ 0.3

85.0

242.8

0.00335

0.00336

—0.3

90.0

230.4

0.00317

0.0

95 0

219.0

0 . 00300

0.0

Table V—

■The Viscosity

' of Ethyl

Ether as Measured by R

Adrian Thomas with Viscometer "B"

Temp.

Time in
seconds
Average

Viscosity

Viscosity
by other
observers

Percentage
difference

1-7

278.6

0.00291

0.00284

+ 2.1

2 .0

276.8

0.00290

0.00281

+ 3-1

269.1

0.00280

0.00272

+ 2.5

10. 0

257.6

0 . 00266

0.00259

+ 2.6

15.0

246.5

0.00253

0 . 00246

+ 1.0

20.0

236.0

0.00240

0.00234

+ 2.2

25.0

227.0

0.00229

0.00223

+ 2.1

30.0

217.6

0.00219

0.00212

+ 2.8

The agreement is very much better with this viscometer,
being satisfactory in the case of water.

Another viscometer was then made for use in measuring
the viscosities of solid suspensions, a capillary of larger diam-
eter than the above but of considerably greater length, 40.8
cm., being used. The values for the instrument with pure
liquids are as follows :

Suspensions of Solids in Liquids

285

Table VI-

-The Fluidity of Water at Different Temperatures,
as Measured with Viscometer "C"

Temp.

30.0
350

40.0
45 o

50.0
550

60.0

65.0
70.0

750
80.0
85.0

90.0

Time in
seconds
Average

225
204

185
170
156

134
125

104
99

Fluidity
125.8

139 5
153 8
168.0
182.9
197.7
213. 1
229. 1
245.6
261 .7
278.7
294.6
311 7

Fluidity

observed by

Thorpe and

Rodger

125
138
153
167
.182
197

213
229
246
263
280
298
316

Percentage
difference

+ 0.3
+ 0.4
4-0.5
+ 0
+ 0

4
3
o

-O. I

-0.2

-0.7
0.6

-I .2

-1.6

Table VII-

-The Fluidity of Phenol at Different Temperatures,
as Measured with Viscometer "C"

Temp.

45 o
50.0
550
60.0
65.0
70.0
750
80.0
850
90.0

Time in
seconds
Average

1080.4

925.2

797.6

696.5

616.2

548.7
492.0

443 I
403.0
370.0

Fluidity

24 -93
29.22

34 01
39.12
44.40
50.10
56.11
62.63
69.17
75.71

Fluidity
observed
by Bingham Percentage
and White difference

—0.4
0.0
0.0

0.0

0.2

+ 0.3
—0.6

—1.3

The volume of liquid in viscometer " C" was 8 . 15 cc. When
we note that this large volume, in the case of water at high
temperatures, flowed through the capillary in less than 100
seconds, we are not surprised that the agreement is unsatis-
factory. The agreement at the lower temperatures seemed
sufficient for the preliminary investigation at hand, where
it was entirely uncertain what conditions would arise. It
was, however, certain that the fluidities to be measured would

286

Bingham and Durham

be no greater than those of water. When measuring the vis-
cosity of suspensions it was advantageous to use a compara-
tively large volume in order to minimize the error due to the
liquid not draining properly.

The suspensions were made up by weight and the volumes
calculated therefrom. The percentages in the following
tables are expressed in terms of volume of solid to volume of
mixture, on the supposition that the volume of the mixture
is the sum of the volumes of each of the components singly.
The specific gravities were taken from the tables of Landolt,
Bomstein and Meyerhoffer.

The viscometer was kept at a constant temperature in a
water bath which was vigorously stirred, the temperature
being read on a thermometer divided to one one-hundredths
of a degree and calibrated at the German Reichsanstalt. The
volume of the liquid in the viscometer was kept constant by
causing any excess to overflow the viscometer before the
measurement began. A mark was etched on the lower part
of the instrument to indicate the exact volume necessary.
Complete mixing of the liquid was insured by blowing air
through the mixture. By using air which was pure except
for the vapor of the liquid in the viscometer, appreciable error
from evaporation was avoided.

Table VIII — The Fluidity of an Approximately 3.1 Per Cent.
Suspension of Infusorial Earth in Water

Volume

Fluidity

per cent

Viscometer

Temp.

solid

Time

"C"

"B"

25.0

318

309 3

87.79

30.0

317

281

97-7

350

3-17

257

106

107.2

40.0

3 16

234

117

450

315

215

127

128.8

50.0

3 14

199

138

550

314

186

148

150.9

60.0

313

172

160

65.0

3.12

163

169

1774

70.0

3.12

153

180

750

143

193

Suspensions of Solids in Liquids

287

Table IX — The Fluidity of an Approximately 6.4. Per Cent.
Suspension of Infusorial Earth in Water as Measured
with Viscometer "C"

Temp.

Volume
per cent, of solid

Time

Fluidity

30.0

6.46

421-5

62.1

350

6.44

384.2

69.0

450

6.41

325 -5

81. I

550

6.39

280.0

94-8

65.0

6.37

246.0

109.0

Table X — The Fluidity of an Approximately g . 3 Per Cent.
Suspension of Infusorial Earth in Water as Measured with
Viscometer "C"

Temp.

per cent, of solid

Time

Fluidity

25.0

9-37

760.0

34.08

30.0

936

703

350

9-34

662

450

9 30

588

550

9.27

492

65.0

9 23

458

750

9.19

442

Table XI — The Fluidity of an Approximately 11. 6 Per Cent.
Suspension of Infusorial Earth in Water Measured with
Viscometer "C"

Temp.

Voltmie
per cent, of solid

Time

Fluidity

30.0
350
65.0

11.63

II 59
11.48

1702.8
15250
1440.0

147
17.0
17.2

Grouping the fluidities together and including the values
of the fluidity of pure water given by Thorpe and Rodger,
we have :

288

Bingham and Durham

Table XII — The Fluidities of Suspensions of Infusorial Earth
in Water at Different Temperatures and Concentrations

Approx

. per

cent, earth by volume

0.9

3.1

6.4

9.3

11.6

Per cent

earth by weight

1.96

6.75

13.24

18.57

22.53

Temp,

Fluidity

25.0

112. 0

87.8

34-1

30.0

125

117

•7

97-5

350

138

106.8

40.0

153

117-3

450

167

127.6

50.0

182

138.3

550

197

148.5

60.0

213

160. 1

65.0

229

169.9

109

70.0

246

180.8

750

263

193-7

Table XIII — The Fluidities of Suspensions of English China
Clay in Water at Different Temperatures and Concentra-
tions as Measured with Viscometer "C"

ight

Voliune

per cent.

Temp.

of clay

Time

Fluidity

30.0

2-15

0.580

259-7

104.0

35-0

0.576

239

113. 8

40.0

573

213

128.3

50.0

566

181

152.7

65.0

558

147

190.6

30.0

468

327

51-2

35-0

465

298

90.9

45-0

458

248

109.9

55-0

452

216

126.2

65.0

446

189

145-3

30.0

630

415

30.0

794

1794

13-76

Suspensions of Solids in Liquids

289

Table XIV — The Fluidities of Suspensions of Infusorial Earth
in Ethyl Alcohol at Different Temperatures and Concen-
trations, as Measured with Viscometer "C," and the Fluidity
of Pure Alcohol by Thorpe and Rodger

Weight

Volume

per cent.

Temp.

of earth

Time

Fluidity

30.0

3.00

1.082

376.4

94 I

35 0

I 075

344-7

103 4

45 0

1.062

296.2

121 .9

55 0

1.048

254-1

144.0

65.0

I 034

220.0

168.5

30.0

15-82

6.237

667.8

48.6

350

6.200

594-8

54 9

45 0

6. 121

544-4

60.7

55 0

6.045

501.2

66.7

30.0

0.0

lOI.O

40.0

0.0

120.7

50.0

0.0

143 -7

60.0

0.0

169.4

70.0

0.0

198 -3

Table XV — The Fluidities of Suspensions of Graphite in Water
at Different Temperatures and Concentrations as Measured
with Viscometer "C"

Volume

Weight

per cent.

Temp.

of graphite

Time

Fluidity

30.0

0.396

0.852

242.6

116. 8

35-0

395

218

129.8

45-0

394

182

156.3

55-0

392

154

184.9

65.0

390

133

215-5

30.0

048

2.236

278

100.9

35 0

046

250

113-4

45-0

042

209

1350

55-0

037

174

161. 7

65.0

032

148

192 . 1

The results are shown graphically in Figs. I-IV. It was
to be expected that any relation would be most easily ob-
served by plotting the results in terms of volume percentages.
The results show that the relations are extremely simple. It

290

Bingham and Durham

I 2 34 56 7 8 Q 10 II 12 13 14i
Percentaoe Volume of Earth
Fig. I — The Fluidity of Suspensions of Infusorial Earth in Water at
Different Temperatures and Concentrations

Suspensions of Solids in Liquids

291

is evident that finely divided substances in suspension depress
the fluidity of the liquids in which they are suspended by
amounts which are directly proportional to the volume of the
solid. But the depression is such that a zero value of fluidity
is reached at a definite but comparatively small percentage,
by volume, of the solid. This composition where the fluidity

240

■

230

220

210

•-V

ZOO

IQO

180

■\r-

170

■\\

160

I w

150

140

' \\\

■S' '^30

• \*a \

•| 120

k, no

\\|

100
90

■ \l\

- \\\\

' WW

\\\

■ , , .\ .

12345%

Percentage Volume of Clay
Fig. II — The Fluidity of Suspensions of China Clay in Water at Different
Temperatures and Concentrations

becomes zero is the same for all temperatures, as all of the
diagrams indicate. When we note that the temperature
coefficient of fluidity of an oil is very great but that a sus-
pension of the critical concentration has a zero temperature
coefficient, it becomes evident that a mixture like "aquadag"
has a great advantage as a lubricant where there are consid-

292

Bingham and Durham

erable temperature fluctuations. Furthermore, the zero of
fluidity is apparently not dependent upon the particular
apparatus used, as is indicated by Table VIII, where it is
shown that two different viscometers give duplicating values.
This composition seems to demarcate viscous from plastic flow.
This is apparently the first time that the limits of viscous
flow and plastic flow have been sharply diS^erentiated, although
Maxwell clearly defined them. This is probably to be accounted
for by the fact that the viscosity curve is not a simple curve,

170
160

150

140

130

120

■v.

\ \

100
90

\^.

( \X\\

\\v\

^^.

■

^^^\

^^^

I 23 456 7 8 9 10 II 12 13'f,
Percentage Volume of Earth.

Fig. Ill — ^The Fluidity of Suspensions of Infusorial Earth in Ethyl Alcohol at Different
Temperatures and Concentrations

whose meaning is immediately apparent. It is to be noted
that the mixture having zero fluidity is not stiff, as might
be supposed; it will not even maintain its own shape in the
containing bottle. But it is difficult to work with mixtures
which even approach this critical composition, for they drain
very badly. A portion of the mixture evidently reaches the
critical composition and sticks to the walls of the viscometer
and does not flow, as would a viscous liquid.

Suspensions of Solids in Liquids

293

That the law enunciated above, that the lowering of the
fluidity is proportional to the percentage by volume of the
suspended solid, is true as a first approximation may be proved
by calculating the fluidity of a 1.04 per cent, suspension of
graphite in water, from the known fluidity of water and the

I 23456%

Percentage Volume of Graphite.
Fig. IV — The Fluidity of Suspensions of Graphite-Aquadag in Water ;
Temperatures and Concentrations

experimentally determined fact that a 5.5 per cent, mixture

has a zero fluidity, by means of the formula 9^ = f i — ^)?<-'

where c is the volume concentration of the unknown suspen-
sion, K the volume concentration of the suspension of zero
fluidity, and 9^ is the fluidity of the pure solvent at the given

294 Bingham and Durham

temperature. The observed and calculated values are given
in the following table :

Table XVI — Calculation of the Fluidity of Graphite Suspen-
sions on the Assumption that a 5.5 Per Cent. Suspension
has Zero Fluidity

Volume

Fluidity

Temp.

per cent.

observed

calculated

30.0

1.048

100.9

IOI.9

350

I .046

II3-4

112. 4

45 0

1.042

1350

1358

550

1.032

161. 7

160.5

65.0

1.032

192. 1

186.3

In viscous flow we must think of the flow as being in the
liquid between the solid particles, the particles of solid in no
way contributing to that flow. The results given above show
that not only is this true, but it is evident that the particles
of solid are able to extend their influence somewhat beyond
the actual space which they fill as a compact solid. Thus, in
a mixture of zero fluidity it seems probable that the particles
of solid are able to exert their forces of cohesion throughout
the whole mass of liquid. When the spheres of action thus
fill the whole space, there is no longer opportunity for true
viscous flow.

As to the size of the individual particles, we seem forced
to the conclusion that the zero of fluidity is not primarily
dependent upon the size of the particles of the solid in the dry
condition. The particles of clay, and particularly the parti-
cles of graphite, were very small and we might suppose that
the fluidity would be correspondingly high; however, it was
much lower than for the infusorial earth whose particles were
rather coarse. A reason for this behavior may be inferred
from the fact that mixtures of clay in water coagulate readily,
so that the particles are really of large dimensions. More-
over, we have observed that the amount of coagulation and
the fluidity are greatly affected by small amounts of impuri-
ties.

Dr. Acheson has shown that various substances, as tannin
and extract of straw, will promote the suspension of graphite,

Suspensions of Solids in Liquids 295

while acids will coagulate it. So far as is known to the authors,
no one has related viscosity to these phenomena; yet, as shown
in a recent paper, anything which causes a mixture to separate
out will decrease its fluidity. That viscosity measurements
afiford a delicate means of detecting these changes is indicated
by the following table :

Table XVII — The Effect of Adding Small Amounts of Sub-
stances to Suspensions at 30° .0

Per cent, solid Fluidity Substance added Resultant fluidity

3. 17 infus. earth

97-5

0 . 03 gram KCl

83.6^

6.46 infus. earth

62.1

o.oi gram KCl

53-2^

9.36 infus. earth

36.3

0.05 gram KCl

22.7

0.58 clay

104.0

0.02 gram KCl

106.5

2 . 63 clay

41-5

0 . 02 gram KCl

65.8

0. 396 graphite

116. 8

0.02 gram KCl
I drop 90 per

116. 9

0.396 graphite

116. 8

cent, acetic
acid

64.5^

As to the cause of these extraordinary effects, there ap-
pears to be no easy explanation. The table shows that the
effect cannot be due to the fluidity of the added substance.
If the coagulation were due merely to the presence of an elec-
trolyte, then acids and bases would not act in opposite man-
ner. Furthermore, potassium chloride would not act in ex-
actly opposite manners upon infusorial earth and clay, coagu-
lating the formed and delaying the coagulation of the latter.
An instructive analogy may be obtained from the behavior
of a suspension of colophony in water, made by dissolving
the colophony in alcohol and adding this to water. On ex-
posing it to the air the alcohol may be evaporated off. This
suspension is like the permanent suspension of infusorial
earth in that it will remain permanently suspended. Like it,
too, it may be coagulated by acids. On the addition of alka-
lis the coagulation is destroyed. This action is so easily
observed and so sharply defined that it may be used to test

' Two grams more of potassium chloride raised the fluidity to 86.5.

2 Three-hundredths gram of potassium hydroxide raised the fluidity to 58 . 3 at
first but the value then steadily fell, presumably due to the action of the alkali upon
the clay. On adding hydrochloric acid the fluidity fell greatly.

3 Alkali nearly restored the fluidity to its former value.

296 Bingham and Durham

for hydrogen or hydroxyl ions, one drop of half-normal acid
or base giving a very decided change at the neutral point in
50 to 100 cc. of liquid. The explanation in this case is evi-
dent. The particles of the acid anhydride, being insoluble
in water, adhere together, resulting in coagulation. As soon
as an alkali is added there is formed the salt of the acid, which
is soluble in water, so that the particles no longer adhere, a
layer of the salt or soap surrounding each particle. The
action of acids and alkalis here described is quite general,
but whether the explanation of the above case can be extended
to explain the suspension of such chemically inert substances
as graphite, clay, infusorial earth, rouge, etc., seems rather far-
fetched. Of course, a thin layer of a resinous substance on
the surface of the particles would be all that would be neces-
sary.

We have tried several colloids, as gelatin, agar-agar and
arsenious sulphide, and found that they do aid suspension
when in the right proportions. A suspension of colophony
in water does not exert much action upon the suspension of
other solids, but the effect of a rosin soap made by adding an
alkali to the suspension of colophony is very marked. Other
alkaline substances, like sodium oleate, sodium carbonate
and dilute ammonia, greatly promote suspension. A weak
acid like boric acid is not pronounced in its influence, while
strong acids and acid salts like sodium acid sulphate produce
immediate coagulation.

From the above, it is evident that the zero of fluidity can
be changed by the presence of small amounts of impurities.
It would be therefore unwarranted to connect the value of
this composition with the specific nature of the substance.
It is to be noted in this connection that graphite requires
many times as much water or oil to make it into a paste as is
required for a substance like white lead. We have made
measurements to show that the composition corresponding
to the zero of fluidity is considerably higher for the white lead
than for the graphite. Whether there is a simple relation be-
tween the composition corresponding to zero fluidity and the
amount of liquid necessary to make a paste of a given con-

Suspensions of Solids in Liquids 297

sistency we are not prepared to state. It would seem to be
necessary to first study the nature of plastic flow. That the
two viscometers "B" and "C" gave fairly duplicating results,
evi^n though the hydrostatic pressures were somewhat different,
does not prove conclusively that the zero of fluidity is in-
dependent of the pressure. Further work will be devoted to
these points.

CONCLUSIONS

1. The viscosity of dilute suspensions can be easily measured
by the capillary-tube method. The fluidity decreases rapidly
as the volume concentration of the solid increases.

2. The decrease in the fluidity is directly proportional to
the volume concentration of the solid.

3. At a definite concentration a zero fluidity is reached
which apparently demarcates viscous from plastic flow.

4. This composition is independent of the temperature
and of the dimensions of the particular apparatus employed
in the viscosity measurements.

5. This composition is not the same for different substances
suspended in the same liquid, nor is it necessarily the same
for one substance suspended in different liquids.

6. The fluidity and the stability of suspensions are greatly
altered by small amounts of impurities which ai)pear to affect
the cohesion between the particles. Foremost in their effects
are hydrogen and hydroxyl ions. But their action is ex-
actly opposite, for the former coagulate suspensions, while
the latter promote them.

7. The much-used " Ostwald viscometer" is open to a serious
defect from the theoretical point of view in that it fails to
take into consideration the varying losses of kinetic energy.
In practice, it has been found that the rate of transpiration
must be very great in order to secure entirely satisfactory re-
sults.

Richmond College

Richmond, Va.

THE RELATION OF HEAT OF VAPORIZATION TO
OTHER CONSTANTS AT THE BOILING TEM-
PERATURE OF SOME LIQUIDS AT AT-
MOSPHERIC PREvSSURE

liY Jack P. Montgomery

Pictet* showed, by the application of the theory of heat
to the study of volatile liquids, that at the boiling point, under
fixed pressure, cohesion is constant, and that at the same
temperature and pressure the latent heat of vaporization is
inversely proportional to the molecular weight.

Trouton^ observed that molecular heats of vaporization are
directly proportional to the absolute temperature of the
boiling point.

Schiff^ determined a large number of heats of vaporization.
He found that, in general, heat of vaporization decreases
as the molecular weight increases, and increases as the abso-
lute temperature of the boiling point. Neither of these rela-
tions alone could be formulated, but by combining the two a
confirmation of Trouton's law was had. We have the formula

MX ^

where M is the molecular M^eight, A the heat of vaporization,
and T the boiling temperature on the absolute scale. Schiff
found for a large number of substances 20.7 as average
value for C.

Linebarger,'' in discussing the above average found by
Schiff^, calls attention to the fact that the greatest divergences
from this number are found in the case of water, the alcohols,
and certain organic acids. Water and most of the alcohols
are associated liquids, but in the state of vapor the molecules
are simple, as is shown by the normal vapor density. When
such liquids boil there occurs a dissociation of complex mole-
cules into simple ones and this uses up heat in addition to that

» Compt. rend., 82, 260.

2 Phil. Mag., 18, 84.

3 Ann Chcm. (Liebig). 234, M3.
" -Vm. I. .Soi., 49, M'.O.

Relation of Heat of Vaporization, Etc. 299

required for vaporization. Consequently these substances
give a value for C much higher than is expected. In the case
of some organic acids, on the other hand, the liquid contains
simple molecules while the vapor contains associated mole-
cules. In such cases the value for C is low. There are other
organic acids which show association in both liquid and vapor
and these also give widely varying results. Linebarger calls
attention to the fact that the above classes of substances
were omitted by Schiff in obtaining his average 20. 7.

Dudley^ has shown, in the case of certain esters of formic,
acetic, propionic, butyric, and valeric acids, that the heat of
vaporization, for each series of esters of the same acid, in a
unit volume of vapor at a given temperature and pressure
is proportional to the density and to the absolute boiling tem-
perature. The same regularity is shown in the case of methyl,
ethyl, propyl and butyl esters of given acids. In Dudley's
paper the attempt is made to show that the acid radicals de-
termine heat of vaporization in esters. From the heats of
vaporization of certain esters he calculates that of the acids.
But his conclusions are not in accord with later experimental
evidence. Linebarger's observation that the heats of vapor-
ization of organic acids are abnormal makes it impossible to
calculate their vaporization constants.

Walden^ has found that heat of vaporization is proportional
to specific cohesion at the boiling temperature and that molec-
ular cohesion is proportional to the absolute boiling tempera-
ture. These observations do not apply to substances under-
going association of molecules in the liquid or vapor.

In addition to those named, a large number of investiga-
tors have made contributions to the subject by the careful
determination of various constants. The literature abounds
in reports of the determination of heats of vaporization, boil-
ing points, densities, etc., and a general acknowledgment
must be made here.

It has been noted above that only those substances having
simple molecules in both liquid and vapor are comparable
when heat of vaporization is concerned. In this paper, giv-

' J. Am. Chem. Soc, 17, 969.
2 Z. physikal. Chem.. 86, 129.

300

Montgomery

ing some relations the writer has found to exist between
heat of vaporization and some other constants, substances
having associated moleclues in either liquid or vapor are, for
the most part, purposely excluded.

I. With liquids at their respective boiling temperatures,
under atmospheric pressure, the volume of a unit weight of
vapor produced is directly proportional to the heat of vapor-
ization. We have the equation

when V is the volume of i gram of vapor at the boiling tem-
perature, V the volume of i gram of the liquid at that tem-
perature, and /I the heat of vaporization.

In the following table the heats of vaporization are taken
from the Landolt-Bomstein and the Castell-Evans tables.
The volumes of the vapors were calculated from data derived
from the same sources, correction being made for tempera-
ture and according to the equation of van der Waals. The
volumes of the liquids were calculated from data given in
various journals and books, correction being made for tem-
perature where necessary.

V —

Benzene

369 -5

93-5

3 952

Ether

333

3 879

Carbon bisulphide

337

894

Ethyl acetate

320

860

Chloroform

228

902

Mercury

240

879

Carbon tetrachloride

179

881

Phosphorus trichloride

206

Ethylene oxide

536

138

Diethylamine

368

•9

Ethylbenzene

300

•4

933

Mesitylene

281

•3

Cymene

254

838

Methyl iodide

181

939

Ethyl iodide

180

851

The above substances were included in the table because
they are representative of widely different classes, but the

Relation of Heat of Vaporization, Etc. 301

same relation was found for a large number of substances.
The average of the values for C in the above is 3 . 89. Esters
of the fatty acids show a great degree of regularity, the value
for C being in most cases from 3.81 to 3 . 90.

Acetone, methylal, sulphur dioxide, tin tetrachloride and
bromine give values for C from 3.71 to 3 . 79. That is, a given
quantity of heat does not produce as great a volume of vapor
as would be expected from the above average. This is readily
explained by the fact that those substances are slightly asso-
ciated in the liquid state. Water and a number of the alco-
hols give values for C lower than the average, ranging from
3 . 83 in the case of amyl alcohol to 2 . 95 for ethyl alcohol,
the divergence of the values depending upon the degree of
association in the liquid state.

The organic acids give varying values for C from 3.56 to
5.68, these values being the complex result of the degree of
association in both liquid and vapor.

The relation outlined above may also be stated thus: The
weight of a unit volume of vapor at the condensation tem-
perature is inversely proportional to the heat of vaporization.
That is, equal volumes of vapor of unassociated substances
at their condensation temperatures produce equal quantities
of heat in condensing to liquids at the same temperature.
For the fifteen substances given above we have an average of 257
calories for one liter of vapor condensing.

It may be well to note here that the relation is an abso-
lute one and is independent of any theory. It will be seen
in the sequel, however, that it is in full accord with the kinetic
molecular hypothesis.

2. In vapors at their condensation temperatures the heats
of vaporization are directly proportional to the squares of
the velocities of the molecules. The velocity of the molecules
may be calculated by substituting the known values in the

formula 5 = \i , where 5 is the velocity, p the pres-

1(2 73 WW

sure, V the gram-molecular volume, T the condensation tempera-
ture (absolute) , and m n the gram-molecular weight. The
formula is derived from the kinetic molecular hypothesis

302 Montgomery

and a discussion of it may be found in Walker's " Physical
Chemistry,'' chapter X.

The above relation is expressed by the formula _ = C,

where S is the velocity of the molecules, }. the heat of vaporiza-
tion, and C a constant.

The constants for some substances are shown in the follow-
ing table:

5
(Meters per

econd)

Ether

100.9

86.00

118.2

Benzene

106

93 5

125.0

Carbon bisulphide

lOI

86.7

118. 6

Ethyl acetate

83 05

117 3

Chloroform

58.5

118.0

Mercury

62.0

117.9

Carbon tetrachloride

46.35

118.2

Ethylene oxide

127

138.6

118.0

Diethylamine

104

91.0

120.4

As previously noted, Trouton found that the product of
molecular weight and heat of vaporization is proportional
to the absolute temperature of the boiling point. As shown
above, the square of the velocity of the molecules is propor-
tional to heat of vaporization. It therefore foUov/s that
the product of molecular weight and the square of the velocity
is proportional to the absolute temperature of the boiling
point. We have the equation

This equation is true for the substances given in the last
table and for other substances not associated. We may say,
then, that the molecular kinetic energy of a vapor at the
condensation point is proportional to the absolute tempera-
ture.

Trouton points out that the molecular heat of vaporization is
proportional to the absolute temperature of the boiling point,
and Walden has shown that the molecular cohesion is pro-
portional to the same constant, so that we have molecular
kinetic energy of the vapor, molecular cohesion of the liquid,

Relation of Heat of Vaporization, Etc.

303

and molecular heat of vaporization mutually proportional,
and for a given substance specific cohesion in the liquid is
proportional to the square of the velocity of the molecules
of the vapor at the transition temperature.

3. The quantity of heat required to vaporize a unit volume

of a liquid at the boiling temperature is directly proportional

to the relative number of molecules, in the unit volume of

liquid, and to the absolute temperature. We have the equation

where H is the heat required to vaporize a unit volume of
liquid, n the relative number of molecules in that volume,
and T the absolute temperature at the boiling point. If ^ve
choose I cc. as the unit volume, H is the product of the density,
d, and the heat of vaporization, ^; w is found by dividing
the density, d, by the molecular weight m.

The following table shows the relation for several substances :

= H

d/M = n

Benzene

76.67

0.01052

353 0

20.646

Chloroform

2>1

O.OII78

334 0

20.877

Ethyl acetate

0.00956

350.0

20 . 900

Methyl butyrate

0.00819

3750

20.624

Carbon bisulphide

102

O.O161

319.0

20. 140

Carbon tetrachloride

0.00973

349-7

20.410

Phosphorus trichloride

0.0108

349 0

20.457

Diethylamine

0.00917

329.0

20.495

4. In the case of isomers we have the equation —

= C.

This relation for several sets of isomers is shown below :

Methyl butyrate
Ethyl propionate

Propyl bromide
Isopropyl bromide

Ethyl acetate
Methyl propionate
Propyl formate

Ethylene chloride
Ethylidene chloride

d (at boiling point)

0.8190

3750

0.002184

0.8165

372.0

0.002192

1.287

344 0

0.003743

I. 241

333 0

0.003727

0 . 8424

350.0

0 . 002405

0.8489

353 0

0 . 002406

0.8435

351 0

0.002414

I. 165

356.6

0.003268

1.093

3330

0.003284

304

Montgomery

Kopp's observation that the densities of isomeric compounds
at the boiling temperature are equal has been found not strictly-
true. The relation noted above is, in the case of many pairs
of isomeric compounds, quite close.

5. The temperature of a boiling liquid is an inverse func-
tion of the relative size of the molecules and a direct function
of molecular weight and of tlie square root of the number of
atoms in the molecule.

We have the equation -z-z — ~ = C in which b has the same

value as in the equation of van der Waals, T is the absolute
temperature, M is the molecular weight, and n the number
of atoms in the molecule.

In the following table the relation is given for some sub-
stances, but there are a number of exceptions. The constants
given in this table may be criticized as being very divergent,
but it should be remembered that the term b depends in value
upon critical temperature and critical pressure, and that data
for critical pressure given by different investigators some-
times vary largely. For benzene, as an example, the value of
C in the above equation is 0.00558 or 0.00669 according to
the data used. With such divergence for one substance it is
not surprising to find an equal and sometimes a greater di-
vergence in the value of the constants when a number of
substances are compared :

b (detd. by

Sajoschewsky)

Ether

0.005745

307.6

74.08

0.00618

Benzene

0 . 005 1 2 I

353 0

048

0.00669

Chloroform

0 . 004450

335

119

358

0.00558

Ethyl acetate

0.005551

350.0

064

0 . 00590

Ethyl chloride

0 . 003966

285.5

0.00622

Ethyl formate

0.004729

326.5

048

0.00627

Methyl acetate

0.003997

b (detd. by
Nadejdine)

3293

74.048

0.00620

Ethyl acetate

0 . 006034

350.0

88.064

0 . 00642

Ethyl butyrate

0.008567

392.0

116.096

0 . 00647

Ethylene chloride

0.004850

3570

98.932

0.00619

Ethylidene chloride

0 . 004790

333 I

932

0.00571

Relation of Heat of Vaporization, Etc.

305

b (detd. by

Nadejdine)

Isobutyl acetate

0.008185

389 -5

116.096

0 . 00620

Methyl acetate

0.004870

3293

74.048

0 . 00654

Isoamylene

0.006274

294.0

70.08

0 . 00680

Methyl formate

0.003602

305 -5

60.032

0 . 00648

Methyl propionate

0 . 006070

352.5

88.064

0.00650

Methylethyl ether

0.004364

6 (detd. by
Young)

284.0

60 . 064

0.00596

Ether

0.00601 I

307.6

74.08

0.00645

Ethyl formate

0.004971

326.5

74.048

0 . 0066 I

Brombenzene

0 . 006880

428.0

157.0

0.00542

Chlorbenzene

0 . 006496

405.0

112.49

0.00676

Carbon tetrachloride 0 . 005663

349 0

153.8

0.00575

Diisopropyl

0.007443

331 0

86.112

0.00641

Heptane

0.009201

371-4

100. 128

0.00746

Hexane

0.007747

344 0

86.112

0 . 00748

Octane

O.OIO571

398.5

114. 144

0.00725

Pentane

0.006518

311. 0

72.096

0 . 0068 I

Isopentane

0 . 006408

303.0

72.096

0.00654

Ethyl isobutyrate

0.008410

b (detd. by

Vincent and

Chappuis)

383 0

116.096

0 . 0062 1

Ethylamine

0.003122

291.0

45.096

0.00637

Diethylamine

0.005598

329.0

73.128

0.00631

Dimethylamine

0.003565

280.2

45.096

0 . 00702

Dipropylamine

0.008124

383 0

101 . 16

0.00657

Methylamine

0.002722

267.0

31.08

0 . 00886

Propylamine

0.004496

322 .0

59.112

0.00680

Triethylamine

0 . 008 I 76

362.0

lOI . 16

0.00625

Trimethylamine

0.004841
b (detd. by

276.5

59.112

0.00629

Ethylene

0.002540

168.0

28.032

0.00626

Cyanogen

0.002900

252.0

52.08

0.00716

CONCLUSION

Heat of vaporization has been variously explained, some
claiming that the heat was required merely to cause the mole-
cules to occupy a greater space in the gaseous condition,
and others holding that the heat is responsible not only for
greater volume but for intramolecular work as well. While

3o6 Montgomery

a complete explanation is not offered in this paper, some re-
marks may be of value.

When a substance passes from the liquid to the gaseous
condition a part of the heat is of course used in overcoming
cohesion and giving the molecules sufficient kinetic energy to
balance the atmospheric pressure. There is no question as
to this point. In considering the matter of intramolecular
work, however, there is a variety of opinions.

Pictet showed that at the boiling point cohesion is con-
stant. Walden found that molecular cohesion is propor-
tional to absolute temperature. Trouton showed that molec-
ular heat of Vaporization is proportional to absolute tempera-
ture. It has been shown in this paper that heat of vaporiza-
tion is directly proportional to the velocity of the molecules
in the gaseous condition at the transition temperature, and
that molecular kinetic energy is proportional to absolute
temperature. We have, then, molecular cohesion, molecular
kinetic energy, and molecular heat of vaporization mutually
proportional. Since we believe that velocity of molecules
and molecular kinetic energy are independent of molecular
complexity, and since we have no evidence that cohesion
is affected by molecular complexity, we must conclude that
molecular heat of vaporization is independent of molecular
complexity.

It has been shown in this paper that the weight of a unit
volume of a vapor at the transition temperature is inversely
proportional to the heat of vaporization. Since the weight
of a volume of a gas depends upon the pressure, temperature,
and molecular weight it follows that molecular complexity
is here of no effect. It has also been shown above that in a
unit volume of liquid at the transition temperature, the heat
required to vaporize it is proportional to the number of mole-
cules and to the absolute temperature. Again, it is seen that
complexity of molecules has no effect upon heat of vaporiza-
tion.

Since, therefore, various molecules of different degrees of
complexity show the same general behavior where heat of
vaporization is involved, we may conclude that none of the

Relation of Heat of Vaporization, Etc. 307

heat of vaporization is required for intramolecular work,
but that all is used in overcoming cohesion and in giving the
molecules sufficient kinetic energy to balance the pressure.

In determining the boiling temperature, however, the size,
weight, and complexity of molecules are all important fac-
tors. A liquid made up of heavy, small, and complex mole-
cules will boil at a higher temperature than one made up of
molecules lighter, larger, or less complex.

SUMMARY

1. The volume of a unit weight of a vapor produced from
a boiling liquid is proportional to the heat of vaporization.
In other words, equal quantities of heat applied to liquids
at their transition temperatures will produce equal volumes
of vapor at these temperatures.

2. In vapors at their transition temperatures the heat of
vaporization is proportional to the square of the velocity of
the molecules and the molecular kinetic energy is proportional
to the absolute temperature.

3. The quantity of heat required to vaporize a unit volume
of a liquid at the transition temperature is directly propor-
tional to the number of molecules in a unit volume and to the
absolute temperature.

4. In the case of isomers the absolute temperature of the
boiling point is proportional to the density of the liquid at
the boiling point.

5. The absolute temperature of the boiling point is inversely
proportional to the relative size of the molecules and directly
proportional to the molecular weight and to the square root
of the number of atoms in the molecule.

6. The conclusion is reached that the boiling point of a liquid
is a function of size, weight, and complexity of molecules,
but that heat of vaporization is not concerned in molecular
complexity, except association, or in intramolecular work.

L.*BORATORY OF THE MISSISSIPPI

A. &. M. College.

REVIEWS

ZERKI.KINERUNGSVORRICHTUNGEN UND MaHLANI^AGEN. Votl CaRL

Naske, Zivilingenieur. Mit 257 Figuren im Text. Chemische Tech-
nologic in Einzeldarstellungen, herausgegeben von Ferdinand
Fischer. Leipzig: Verlag von Otto Spamer. 1911. S. x + 235.
Preis, geh., M. 13.50; geb., M. 15.

This book has been written especially for the industrial
chemist but will be found of value for the engineer. The
author has made use of prominent authorities and catalogues
for obtaining descriptions of breaking and comminuting
machines of all kinds and for all purposes, together with other
necessary apparatus for handling the ground products. With
the limits set the description of any one machine is necessarily
not detailed, but brief and to the point.

The following list indicates the range of subjects covered:
Breakers, rolls, Chili mills, cone mills, impact machines or
disintegrators, stamps of gravity, steam and hydraulic types,
grinders, roller mills, ball mills, tube mills, screens of bar,
trommel and shaking types, air separators of various types,
dust collection by settling chambers, dry filters (bag houses),
centrifugal action, water spray and wet filters, storage and
packing products, individual plants for phosphate, super-
phosphate, paints, sal ammoniac, hyposulphite, dolomite,
common salt, clay for earthenware, cement, road material,
drugs and spices.

Some machines of doubtful value are bound to be included
in a book of this kind but the author has done his work so well
that the reader may well overlook any minor fault of this

nature. Charles E. IvOckb.

Ai^len's Commerciai, Organic Analysis. Vol. IV. Resins, India-
Rubber, Rubber Substitutes and Gutta-Percha, Hydrocarbons of
Essential Oils, Ketones of Essential Oils, Volatile or Essential Oils,
Special Characters of Essential Oils, Tables of Essential Oils. By the
editors and the follovi^ing contributors: M. BENNETT Blacki,er, E.
W. Lewis, T. Martin Lowry, Ernest C. Parry, Henry Leffmann,
Charles H. LaWax,l. Fourth edition. Entirely rewritten. Edited
by W. A. Davis, B.Sc, A. C.G.I. , and Samuel S. Sadtler, S.B.
Philadelphia: P. Blakiston's Son & Co. 1911. pp. viii + 466. Price,
?5-oo.

The chapter upon resins (103 pp.) gives a clear, concise and
satisfactory account of the more important of these substances,
together with the oleoresins and balsams. Due acknowledg-
ment is made of the debt chemists owe to Tshirch and Diete-
rich.

The chapter on rubber (58 pp.) is more valuable for informa-

Reviews 309

tion as to manufactured products and their uses than for
analytical directions. Weber is very closely followed. The
remainder of the book is devoted to the essential oils, their
origin, properties and adulterants being given. Careful at-
tention has been paid to the recent methods and literature.

The availabiUty of the work would be much increased by a
more complete index, including a list of authors.

The volume may be commended to all those interested. No
technical chemist's library is complete without one or more
volumes of "Allen." a. h. Gill.

Die Direkte Einfuhrung von Substituenten in den Benzoi^kern.
Ein Beitrag zur Losung des Substitutionsproblems in aromatischen
Verbindungen, Kritische Literaturiibersicht und experimentelle
Untersuchungen. Von Dr. A. F. H01.1.EMAN, ord. Professor der
Chemie an der Universitat Amsterdam. Mit zahlreichen Figuren.
Leipzig: Veit und Comp. 1910. S. vi -f 516. Preis: geh., M. 20;
geb., M. 23.

In the preface of this valuable book Professor Holleman
gives his reasons for publishing it :

"A review of the experimental work and theories on the
direct introduction of substituents into the benzene nucleus
has to my knowledge not yet been published. As I have been
engaged for several years in working on this problem I resolved
to undertake such a review and to incorporate my own inves-
tigations at the proper places. For my own work it was neces-
sary to study over a thousand papers, as I was desirous, as
far as possible, of giving a complete review of the literature
of the subject. In doing this I have not merely compiled
the work done, but have also subjected the papers to criti-
cism, so far as this was possible without repeating the ex-
periments therein described." It is certainly a matter for
congratulation that this task has been undertaken by such a
competent authority in this field as Professor Holleman, and
all chemists owe him a debt of gratitude for the excellent
manner in which the work has been done.

Anyone who has worked in this field of organic chemistry
or who is familiar with the literature of the subject will in-
dorse this statement of the author, also found in the preface:

"A cursory inspection of this book will show at once that
even in the case of many quite simple reactions it has not yet
been determined with sufficient accuracy qualitatively what
isomers are formed under certain definite conditions, let alone
the relative quantities of these isomers which are formed
simultaneously. The conviction forces itself upon us that a
complete revision of all work on the direct introduction of

3IO Reviews *

substituents into the benzene molecule which already con-
tains one or more groups, not only qualitatively but espec-
ially quantitatively, is essential if this problem is to be seriously
attacked."

This is shown very clearly in an article just published by
the author and some of his pupils^ entitled "Quantitative
Investigations on the Nitrating of Aniline," in which he shows
that the conclusions stated in this book regarding the nitra-
tion of aniline and its derivatives are not correct. He had
assumed, from the work of others, that the direct nitration
(that is, the direct replacement of hydrogen of benzene by
the group NOj) of aniline sulphate gave only the vieta product,
while the indirect nitration (through the intermediate forma-
tion of phenylnitramine) led to para and ortho derivatives.
The investigation above referred to, however, proves that
this hypothesis is not correct. For, in nitrating aniline in
sulphuric acid, in which case the aniline is undoubtedly nitra-
ted as sulphate, there is formed, in addition to the nieta com-
pound, a considerable quantity of /jamnitraniline also, and
only very little of the ortho isomer. On the basis of the re-
sults of this investigation Professor Holleman now formu-
lates the reaction as follows :

1. In the nitration of aniline and its derivatives, as well
as in the introduction of halogens, the main product is the
para compound, together with certain quantities of the ortho
isomer.

2. Two circumstances may modify this result: (a) the
formation of sulphate and the accumulation of acid residues
in the group NH2, whereby the metanitro product is formed;
(b) indirect substitution whereby mainly the orthonitro com-
pound results.

The first chapter of the book takes up the direct introduc-
tion of one substituent into the unsubstituted benzene nucleus.
The second is devoted to the quantitative estimation of iso-
mers in the presence of one another and the determination
of the series to which the bisubstitution products belong. In
the third chapter the direct introduction of a second sub-
stituent into a monosubstituted benzene is considered, and in
the fourth chapter the author discusses the results and gives
some theoretical observations. In the fifth chapter the intro-
duction of a third substituent into a disubstituted benzene is
taken up, and the sixth is devoted to a discussion of the re-
sults, both qualitative and quantitative. The seventh and
last chapter is given up to a discussion of the melting-point
curves of binary mixtures and the quantitative analysis or

» Ber. d. Chem. Ges.. 44, 704 (1911).

Reviews 311

mixtures of three isomers by means of the melting-point figure
of ternary mixtures.

In chapter six the author develops his own views on the
mechanism of substitution. Starting with Kekule's benzene
formula, he assumes that substitution in the benzene ring is
preceded by the formation of an addition product. For ex-
ample, if the compound CeHjX is nitrated one or more of the
three addition products

HO X

Hs .H

«>A„

„/\<"

\0H

«\/C,

H NO,

Para.

Ortho.

Meta.

first formed lose water and the less stable ring with two double
bonds goes back to the more stable benzene ring. Which of
these three addition products will be formed, that is, which type
of substitution product will result, will be determined by the
velocity of the reaction. If the group X accelerates the reac-
tion, substitution will take place in the para and ortho posi-
tions and exclusively in these positions when X exerts a large
accelerating influence. If the accelerating influence of X is
not so great, then certain quantities of the meta compound
may be formed. For example, phenol gives on nitration
only para- and orthonitrophenol, while toluene yields some of
the meta compound in addition to the other two. If X has a
retarding influence then the addition at the double bond
2,3 predominates, the consequence of which is that the meta
compound will be the main product.

For anyone who works in this field of organic chemistry
this book will be absolutely indispensable and even the aver-
age chemist will find here material that will fully repay all the
time he devotes to it. w. r. o.

ECBKTRISCHE DOPPKLBRECHUNG DKR KOHLENSTOFFVERBINDUNGEN.

Von Dr. Phil. Richard I^eiser, aus Wien. Mit 15 als Anhang
gedruckten Abbildungen. Abhandlungen der Deutschen Bunsen
Gesellschaft fiir angewaudte physikalische Chemie. Nr. 4. Halle
a. S. : Druck und Verlag von Wilhelm Knapp. 1910. pp. 71. Price,
M. 3.60.
In this monograph the author gives an account of a series

312 Reviews

of interesting experiments on the relation between the elec-
trical double refraction and the chemical constitution of or-
ganic compounds. He bases his method of experimentation
upon the well known observation made by Kerr in 1875 that
if a layer of carbon disulphide is placed between two crossed
Nicol prisms and is subjected to electric stress the plane of
polarization is rotated. The experimental arrangements are
too complicated to admit of a brief description, hence it must
suffice to state that an optical system was devised which per-
mitted the detection of a phase difference of 3 X lo""'' ^. With
this apparatus the "Kerr Constants" of electrical double re-
fraction of one hundred and fifty organic compounds were
measured.

The constants of a few well known compounds are as follows :

Propyl alcohol — 78 . o

Isopropyl alcohol + 73 o

Benzene 1 2 . i

Toluene 24.3

o-Xylene 41.2

w-Xylene 24 . 4

/^-Xylene 22.6

Chloroform — 1 00. o

Carbon tetrachloride 2 . 3

These examples would seem to justify the author's claim:
"Es diirfte sogar, nach den bisherigen Erfahrungen zu urteilen
keine andere physikalische Eigenschaft so charakteristisch
fiir die chemische Konstitution und daher so geeignet sein,
daraus Schliisse auf die Konstitution unbekannter Verbind-
ungen zu ziehen, als die elektrische Doppelbrechung." The
author has also sought to develop a new concept of the nature
of electrical double refraction which differs fundamentally
from the theory which is generally accepted.

The method described in this monograph, notwithstanding
the somewhat elaborate and complex , apparatus required,
should prove of great value in helping to establish the consti-
tutions of many compounds which at the present time are

unsettled. Frederick H. Gbtman.

Thb electrical Nature of Matter and Radioactivity. By
Harry C. Jones, Professor of Physical Chemistry in the Johns Hop-
kins University. .Second edition, completely revised. New York:
D. Van Nostrand Co. 1910. pp. ix -f- 210. Price, |2.oo.
The first edition of this book has already been reviewed in

This Journal.^ The aim of this edition has been "to bring

* Vol. 36, p. 614.

Reviews 313

it up-to-date as far as matters of fundamental importance
are concerned," otherwise it hardly needs an introduction to
the chemical public.

There is surely no topic upon which the physical chemist
is so often questioned by the laity as that which is included
by the title of this book. In spite, however, of the general
interest and fundamental scientific importance of these mat-
ters, but few have any direct experimental knowledge of them.
This volume fills, therefore, a very general need for informa-
tion which shall be concise, nonmathematical, and which shall
explain clearly into just what sort of a thing the atom has
evolved nowadays. The ability of the author to do this in
a manner within the grasp of those without special training,
and yet thoroughly scientific, is well known, so that this new
edition will be welcomed by many.

In view of the accumulated evidence in favor of 225 as the
atomic weight of radium, it seems rather strange that the
long argument in favor of the higher value is still retained.
There are occasional evidences of revision, such as the refer-
ence to work done in 1905 as "quite recent." On p. 95 the*
term "hardness" is used in a sense familiar only to workers
with the X-ray tube. On p. 190 "radium" is printed where
evidently "sodium" is intended, and on p. 197 we find "hydro-
gen sulphate." The liquefaction of helium occurred too late
to be used in correcting the text on pp. 137 and 154 or else
was neglected. A few such errors are, however, almost in-
evitable, and the volume remains, doubtless, the best semi-
popular text on the subject. joel h. hildebrand.

TraiTe; de Chimie G6ne;rai.e. Par W. Nernst, Professeur k I'Univer-
site et Directeurde I'lnstitut de Chimie Physique de I'Universite de
Berlin. Ouvrage traduit sur la 6e edition allemande par A.
CORVISY, Professeur Agreg^ des Sciences Physiques au Lycee Gay-
Lnssac, Professeur Suppleant k I'^^cole de Medecine et de Pharmacie
de Limoges. Premiere Partie: Proprietes Centrales des Corps- Atome
et Molecule. Paris: Librairie Scientifique A. Hermann et Fils.
1911. pp. II + 510. Price, Fr. 12.

The increasing number of German books being translated
into French, and vice versa, gives gratifying evidence of the
growing feeling of good will and mutual appreciation between
the two nations. In his preface, Prof. Corvisy, the excellence
of whose work as a translator is already known through his
translation of Ladenburg's History of Chemistry, pays a grace-
ful tribute to Prof. Nernst's book which, since its first appear-
ance, has been a source of so much help and inspiration to all
physical chemists. The volume under review covers the
first 18 chapters of the "Theoretische Chemie." The original

314

Reviews

text of the 6th German edition has been closely followed,
only a few changes in the interest of historical accuracy hav-
ing been made at the suggestion of Prof. Nernst himself.
French chemists are to be congratulated on having such an
excellent translation placed at their disposal. c. a. r.

QUAUTATIVK ChKMICAI, ANAI.YSIS, A LaBORATORV GUIDB. BY WIL-
FRED Weld AY Scott, A.M., Chief Chemist, Baldwin Locomotive
Works, formerly Professor of Chemistry, Morningside College.
Illustrated. New York: D. Van Nostrand Co. 1910. pp. xi + 165.
Price, $1.50.

In this well-balanced manual the author gives 11 pages of
the introduction to the application of physical chemistry to
analysis, and a !• w pages to laboratory directions.

The greater part of the book is a systematic study of bases
and acids; the order observed with each element is a brief
mention of the important compounds, special tests, pre-
liminary reactions, group separations, discussion of details of
analytical methods, and a number of questions. A chapter
on systematic analysis follows. The next chapter contains
tables of reactions of acids and bases, followed by directions
for preparing reagents and by a table of solubilities.

Professor Scott's book is admirable. The analytical meth-
ods employed are without exception the best. The discus-
sions of methods are clear and thorough. E. R.

Vol. XI^VI October, 1911 No. 4

AMERICAN

CHEMICALJOURNAL

ON THE COLOR CHANGES OCCURRING IN THE BLUE

FLOWERS OF THE WILD CHICORY, CICHOR-

lUM INTYBUS

By Joseph H. Kastle and R. L. Haden

Two varieties of the wild chicory, Cichorium intyhus, are
abundantly distributed over certain parts of Virginia and
the District of Columbia, a more common blue variety and
a rarer albino variety having white flowers. Several years
ago the attention of one of us (Kastle) was called to the fact
that the blue flowers of the wild chicory turned white and
then brown, these changes occurring with great rapidity.
The blue chicory occurs very abundantly in the immediate
vicinity of the University of Virginia, so that an excellent op-
portunity presented itself for a closer inquiry into the cause
of these interesting color changes exhibited by this blue flower.

A closer study of this phenomenon has revealed the fol-
lowing facts: The flowers of the blue chicory open in the
morning, usually between seven and eight o'clock. When
perfectly fresh the flowers have a light blue color and occa-
sionally show a lavender or pinkish coat. Thus the flowers
gathered by Mr. Haden on the morning of July 19 were dis-
tinctly pink near the base of the corolla. Even when left un-
disturbed upon the plant these blue flowers soon begin to

3i6 Kastle and Haden

bleach and gradually lose their blue color, this change usually
beginning at the outermost tips of the petals (corolla) and grad-
ually extending in to the point at which the corolla is attached
to the stem. The blue color nearly always persists in the
stamens and pistil long after it has disappeared from the
corolla. The composite flower gradually closes during the
day and turns brown, so that after several days it is reduced
to a nodule of brown, dry vegetable material. The rapidity
with which these changes take place in the undisturbed flower
seems to depend very largely on atmospheric conditions. On
hot, moist days the flowers have frequently bleached and
closed and have turned brown before noon. In other in-
stances and under other atmospheric conditions they may re-
tain their blue color until two or three o'clock in the afternoon,
but always before the close of the day on which they bloomed ;
these blue flowers first turn white and close and are converted
into a brownish mass of vegetable matter. When the flowers
are pulled they pass through precisely the same changes of
color and the same phenomena are observed as occur in those
flowers which have been left undisturbed upon the plant.
We shall show in this communication that these changes can
be brought about even in vitro. It frequently happens that
in bleaching the blue flowers become distinctly pink.

In the blue chicory, then, we have a flower which in the course
of a few hours shows a most striking succession of color changes,
namely, from blue to pink, or from blue to white, and then
brown. As a matter of fact, this phenomenon is so striking
that it has to be seen in order to be fully appreciated. It is
well known, of course, that a great variety of flowers gradually
alter in color as they fade and ultimately become brown.
There are very few, however, which show a distinct succession
of color changes during their life cycle, and so far as we are
aware there are none, other than the blue chicory, in which
these color changes are complete in a few hours.

Like other varieties of anthocyanin, the blue pigment of
these flowers becomes bright carmine-red or rose-red on the
addition of hydrochloric acid and yellowish green on the addi-
tion of alkali. We have made use of this reaction in follow-

Color Changes in Blue Flowers of Wild Chicory 317

ing the color changes occurring in this flower. In our eff'orts
to arrive at the cause of the color changes occurring in the blue
chicory we have studied the conduct of the flowers in light
and darkness and also in chloroform vapor and in hydrogen.
Secondly, we have studied the conduct of the flower pigment
towards acids and alkalis and also towards the oxidases, per-
oxidases and various oxidizing agents. Lastly, we have ob-
served that the flowers of both the blue and white chicory
contain a powerful oxidase which in our opinion is responsi-
ble for the destruction of the flower pigment and the turning
brown of the flower. The results of these observations are
given in the following.

In the first place, it should be observed, in passing, that
like all biochemical processes these color changes in the flower
are dependent on a certain degree of moisture. With reason-
able care these blue flowers can be pressed and dried between
the folds of bibulous paper without losing their blue color.
When perfectly dry they retain their blue color for months,
indicating that water is essential to the bleaching. It has
also been observed that the flowers of the blue chicory turn
white somewhat more rapidly in the dark than in the light,
so that exposure to sunlight cannot be the cause of this phe-
nomenon.

Among other things, it occurred to us to test the behavior
of the freshly gathered blue flowers in chloroform vapor. A
stem of the freshly gathered plant, bearing a number of fully
bloomed blue flowers, was placed in a small beaker of water
and the whole placed in a tall desiccator containing some
absorbent cotton saturated with chloroform. The first ef-
fect of the chloroform was to cause the flower to open out
straight and full like a daisy. The blue petals began to show
signs of bleaching almost immediately, and in twelve minutes
after placing in the chloroform vapor the petals were nearly
white. In twenty minutes the blue color had disappeared
from the petals completely and only the stamens and pistil
were blue. In twenty-five minutes the petals presented a
shriveled appearance and were translucent and waxy looking.
In forty-five minutes they were light brown in color and much

3i8 Kastle and Haden

shriveled, and after one hour and ten minutes no trace of blue
color was visible anywhere in the flower except in the stamens
and pistil.

A second test of the effect of chloroform vapor on the blue
flowers gave the following results. Some of the flowers were
put in the chloroform vapor at 8.32 a. m. The petals of
many of these flowers began to shrivel almost immediately
and to turn pink. Some few, however, retained their original
form and bleached quickly as described in the foregoing para-
graph. In nine minutes many of the flowers had bleached
and after thirteen minutes exposure they were removed from
the chloroform vapor and the petals macerated with a small
amount of o. i N hydrochloric acid. These gave only a faint
pink color with the acid, indicating that only a little of the
flower pigment remained, about as much, in fact, as remained
in the petals of the flower which had bleached and faded
normally in the air. In this connection it is to be noted that
the fresh blue flowers give a deep carmine-red solution on
maceration with o.i N hydrochloric acid. By way of com-
parison vv^ith the foregoing experiments in chloroform vapor,
another stem bearing fresh blue flowers was placed in water
and put in a desiccator without chloroform. After twelve
minutes the flowers were unchanged and blue. At the end
of forty-five minutes they were still blue. In one hour and
fifteen minutes they had begun to turn white, this change
first showing itself on the outermost tips of the corollas. After
three hours, however, there was still some blue color to be
observed near the central portions of the corollas. It was
also observed that in passing to white these blue flowers
tended to become pink. It is evident, therefore, that these
color changes in the blue chicory are greatly accelerated by
chloroform. This is not difficult to understand when it is
borne in mind that these color changes are in some way asso-
ciated with the gradual death of the flower. Hence, anything
like chloroform which would tend to poison the flower ought
to hasten these color changes.

We have also studied the conduct of the blue flower in
hydrogen. A stem of the plant bearing blue flowers was

Color Changes in Blue Flowers of Wild Chicory 319

placed in a small beaker of water and the whole set under a
bell jar through which a current of hydrogen was passing.
After forty minutes the flowers were observed to have a slight
pink color. At the end of an hour and ten minutes they ap-
peared slightly shriveled and somewhat more pink. After
one hour and forty minutes the flowers were still blue, but
showed also a pink tint. At the end of six and a half hours
the petals were found to be considerably shriveled, translucent
and waxy in appearance, and almost entirely pink. The
stamens and pistil were plum colored and very little, if any,
of the blue color was left. A number of the petals, which were
sticky to the feel, were macerated with i cc. o . i N hydro-
chloric acid and water. A rose-red solution was thus ob-
tained, showing the presence of considerable amounts of the
flower pigment.

On the other hand, the blue flowers which had been left ex-
posed to the air during this interval, to serve as a control on
the hydrogen experiment, had long since bleached, withered
and turned brown. The petals of these flowers were found
to contain only traces of the flower pigment. In a second
observation some of the freshly gathered flowers were placed
in hydrogen at 8.30 a. m. on July 21, and an equal number
of flowers were left exposed to the air by way of comparison.
After half an hour the flowers in hydrogen had become de-
cidedly pink and a few of the petals showed slight evidences
of bleaching at the tip. All of the flowers which had been
left exposed to the air had bleached greatly, the color other-
wise remaining being blue. All of them had withered some-
what, and one was closed entirely. One flower also was en-
tirely white. At the end of an hour the flowers in hydrogen
had shown no further change except that they had become
somewhat more pink in color. The flowers which had been
kept in the air were all withered by this time and all of them
had become almost white, the only blue color remaining being
in the stamens and pistil and at the point of attachment of
the corolla to the stem. At 10 a. m. the flowers in hydrogen
were all open except one and all were pink, although some blue
color remained, giving a sort of lavender tint. In the air all

320 Kastle and Haden

of the flowers were closed except one and all were white.
The contrast in color and general appearance between these
two lots of flowers at this time was very striking. The flow-
ers in hydrogen stood until lo a. m., July 23rd. At this time
three of the flowers were found to be open, somewhat shriveled
and deep pink in color. Three of the flowers had bleached
and one was closed. The bleached flowers gave only traces
of pink on maceration with o. i N hydrochloric acid, whereas
the pink flowers gave a rose-red solution, indicating the pres-
ence of considerable amounts of flower pigment. The flow-
ers which had been kept in the air during this time had changed
to small, dark brown masses. It is clear, therefore, that hy-
drogen greatly retards those processes resulting in the de-
struction of the flower pigment in the flowers of the wild
chicory. In this respect it exercises a protective and conserva-
tive action on the coloring matter contained in these flowers.
As a matter of fact, the conduct of the flower in hydrogen,
as compared with air, naturally suggests that in some way
oxygen is concerned in these changes. It was thought at one
tim.e that the bleaching observed in this flower was due to a
migration of the coloring matter from the corolla to the stamens
and pistil. It is certain that these parts of the flower remain
blue long after the corolla has become perfectly white. In
order to test this point, a quantity of the fresh blue petals
were cut off near the base of the flower. These were used as
follows :

(i) Some of the fresh blue petals were placed in chloroform
vapor. They bleached very rapidly, becoming pink during
the process.

(2) Some of the fresh blue petals were pressed and dried be-
tween the folds of fine filter paper. These retained their blue
color and have kept their color unchanged for over a month.

(3) Some of the cut petals were allowed to remain in the
air, in the light. These gradually bleached, much less rapidly,
however, than those in chloroform vapor.

(4) Some of the cut petals were allowed to remain in the
air, in the dark. These also bleached at about the same rate
as those in the air, in the light.

Color Changes in Blue Flowers of Wild Chicory 321

In none of these experiments was there any evidence of any
migration of the flower pigment from one part of the corolla
to the other. It is clear, therefore, that the bleaching of
the blue chicory is not due to a migration of the coloring
matter from the corolla to the stamens and pistil, but rather
to some chemical change within the tissues of the corolla
itself. Reference has already been made to the fact that the
flower pigment turns bright carmine-red on the addition of a
small amount of o.iN hydrochloric acid. We have also
pointed out that the blue flowers of the chicory tend to turn
pink under certain conditions. It therefore occurred to us
that possibly changes in the reaction of the cell contents in
the pigment cells might be responsible for the changes of
color observed in the blue flower. With a view to testing this
point a quantity of the blue petals were macerated in a porce-
lain mortar with i cc. o.i N hydrochloric acid. To the red
mass 10 cc. of water was added and the mixture filtered. A
clear carmine-red solution was thus obtained which has been
found to keep unaltered for a number of days. To some of
this red solution o. i N sodium hydroxide was gradually
added. After a certain amount had been added a beautiful
blue solution was obtained, exactly matching in tint the color
of the fresh blue flower. On the addition of still further
amounts of o . i N sodium hydroxide the solution gradually
became lighter in color, until finally a perfectly colorless solu-
tion was obtained, corresponding, of course, to the colorless
condition of the flower after it has bleached. On the addi-
tion of still more o . i N sodium hydroxide the colorless solu-
tion became bright yellowish green. If now more hydrochloric
acid is added the solution again becomes carmine-red and on
this solution the succession of colors with alkali could be
again obtained. The red, blue and colorless solutions of the
flower pigment have all been found to be acid to phenol-
phthalein and even the yellowish-green solutions do not always
give an alkali reaction with this indicator. In other words,
it would seem that in the variety of anthocyanin contained in
the flower of the blue chicory we have an indicator capable
of indicating various amounts of acid. It is also clear that the

322 Kastle and Haden

particular color displayed by the flower of the blue chicory,
whether pink, blue or white, will depend on the acidity of the
cell contents of the pigment cells. If they contain relatively
large amounts of acid the flower will be pink ; if smaller amounts,
it will be blue; and if still smaller amounts, white. As a mat-
ter of fact, it not infrequently happens that certain corollas
of this flower show all three colors simultaneously, distributed
over different areas of the corolla. In this connection it is
interesting to note that the buds of the blue weed, Echium
boreale, L., are pink in color, whereas the corolla of the full-
blown flower is blue. In all likelihood this pronounced differ-
ence in color is due merely to a difference in acidity.

On the other hand, it has been observed that the bleached
petals of the blue chicory, especially after they have begun to
shrivel and turn brown, contain very little of the flower pig-
ment. On maceration with o. i N hydrochloric acid such
petals give only traces of pink. It would seem, therefore,
that another agent whose action has resulted in the com-
plete destruction of the flower pigment has been at work in
the bleaching and browning of these blue flowers. That such
is the case is indicated by the following:

We have already seen that the conduct of the blue flowers
in hydrogen is such as to indicate that the bleaching of the
flower and the destruction of the flower pigment are due to
oxidation. In this connection it is interesting to note that
on maceration with water the blue flowers yield a red solution.
This color is soon lost, however, and the solution turns brown.
Such aqueous extracts of the blue flower give the guaiacum
reaction directly, indicating the presence of an oxidase.

The rapid oxidation of the flower pigment by an oxidase
would also explain the results which we have obtained with
acetic and malic acid. When the blue flowers are macerated
with I cc. o. I N acetic or malic acid and the macerated mass
diluted with water, rose-red solutions are obtained, but by
the time these are filtered it is found that the filtrates have
lost their rose-red color and show only traces of anthoc3^anin
on the addition of dilute hydrochloric acid. In this connec-
tion it has been found that extracts of the blue flower with

Color Changes in Blue Flowers of Wild Chicory 323

these weak acids blue guaiacum, whereas extracts of the
flower in o . I N hydrochloric acid, in which the flower pigment
is stable, do not give the guaiacum reaction. In other words,
the oxidase has been destroyed by the stronger acid and hence
the flower pigment escapes oxidation. The weaker acids,
however, do not destroy the oxidase and hence in such ex-
tracts, as also in water, the flower pigment is destroyed by
oxidation. In the light of these results it is only logical to
ascribe the destruction of the flower pigment in this blue
flower to the action of an oxidase. That such is the case has
been proven in the following manner:

, A number of the blue petals were macerated with i cc.
o . I N hydrochloric acid and a small quantity of distilled water
added. The bright rose-red solution was then filtered and
o. I N sodium hydroxide added until the solution became
bluish green in color. Three tubes were then prepared, each
containing 3 cc. of this solution. These were labeled (i), (2)
and (3), respectively. To (i) i cc. of distilled water was
added, to (2) i cc. of a boiled aqueous extract of the blue
flower, and to (3) i cc. of an active (unboiled) aqueous extract
of the blue flower. The three tubes were then allowed to
stand fifteen minutes at ordinary temperature. At the end
of this time i cc. of o. i N hydrochloric acid was added to each
tube, with the following results :
(i) became rose-red in color;

(2) became rose-red in color ;

(3) remained yellowish brown and showed no pink color
at all.

This proves conclusively that the flower pigment of this
blue flower is destroyed by an aqueous extract of the flower
itself.

If now the destruction of the flower pigment is due to an
oxidase contained in the flower, then it is more than likely
that the coloring matter of the flower would be destroyed
by plant oxidases from other sources, and in fact by all of those
oxidizing agents which blue guaiacum directly. In order
to test this point the following experiment was carried
out. A number of the blue petals were macerated in a porce-

324 Kastle and Haden

lain mortar with i cc. o. i N hydrochloric acid, in order to ex-
tract the flower pigment and at the same time to destroy
the oxidase of the flower. About 10 cc. of water was added
to the macerated mass and the mixture filtered. Then o. i N
sodium hydroxide was added gradually to the rose-red fil-
trate until the solution became colorless or faintly blue. Three
tubes were then prepared, each containing 3 cc. of this solu-
tion of the flower pigment. These were labeled (i), (2) and
(3), respectively. To tube (i) there was added i cc. of water;
to (2) I cc. of a boiled, aqueous extract of the peel of the potato;
to (3) I cc. of an active (unboiled) aqueous extract of the
potato peel. These tubes were then allowed to stand at
ordinary temperature for fifteen minutes, at the end of which
time I cc. of o. I N hydrochloric acid was added to each, with
the following results :

(1) became rose- red ;

(2) became rose- red ;

(3) remained yellowish brown and failed to show any trace
of pink color.

This experiment goes to show conclusively that the flower
pigment of the blue chicory is oxidized by the oxidase of the
potato.

The colorless solution of the flower pigment is also readily
oxidized and destroyed by the following substances: potas-
sium ferricyanide, potassium permanganate, quinone, dilute
solution of iodine, and by a o . 3 per cent, solution of hydrogen
peroxide under the influence of various catalysts, such as
platinum black, manganese dioxide and lead peroxide. As
is well known, all of these substances blue guaiacum
directly. It is more than likely, therefore, that any sub-
stance or mixture which gives the guaiacum reaction will
oxidize and destroy the flower pigment of the blue chicory.

In this connection it is interesting to note that the flowers
of the white chicory also contain a powerful oxidase. The
white flowers contain no anthocyanin, however, and hence do
not exhibit the color changes shown by the blue variety.
They wither rapidly, however, and turn brown as is the case

Stereoisomeric Chlorimido Ketones 325

with the blue variety, and in these changes the oxidase is un-
doubtedly concerned.

SUMMARY

The flowers of the blue chicory fade and wither with great
rapidity. In so doing they exhibit certain marked changes of
color, changing from blue to pink, from blue or pink to white
and finally to brown, with practically complete destruction of
the flower pigment.

It has been conclusively shown that these changes of color
are due in part to variations in the amount of acid contained
in the pigment cells and in part to the action of an oxidase^
contained in the flower which completely oxidizes and de-
stroys the flower pigment.

It has also been shown that other plant oxidases and various
other oxidizing agents can afifect the oxidation of the flower
pigment into nonchromogenic products. •

University of Virginia
July, 1911

[Contribution from the Kent Chemical Laboratory of the University of Chicago]

STEREOISOMERIC CHLORIMIDO KETONES^

By Peter P. Peterson

In 1903 Stieglitz and Earle,^ in the course of their work
on the theory of the "Beckmann Rearrangement," discovered
the first representatives of a new group of stereoisomeric

1 The word oxidase is here used in its usual signification, viz., as being an un-
stable, biologically active substance capable of oxidizing guaiacum, aloin and phenol-
phthalin, etc., directly. Quite recently Miss Wheldale (P. Roy. Soc, B, 84, 121-124)
has advanced the view that the browning of many plant extracts is due to the oxida-
tion of pyrocatechinol under the influence of air and a peroxidase, and that the plant
oxidase is in reality a mixture of pyrocatechinol, or the product of its autoxidation,
and a peroxidase; in other words, that the power of plant extracts to blue guaiacum
is due to an oxidation accomplished by the autoxidation product of pyrocatechinol
under the activating influence of a peroxidase. It is scarcely necessary to add that
this hypothesis regarding the nature of the plant oxidase in no way affects our view
regarding the destruction of the anthocyanin in the bleaching of the flowers of the
blue chicory. Our findings are that the anthocyanin in this flower is destroyed during
its bleaching by an oxidase which is present or developed within the flower itself.
This, of course, postulates nothing regarding the nature of the oxidase.

2 See a preliminary paper by J. Stieglitz and P. Peterson: Ber. d. chem. Ges., 43,
782 (1910).

3 This Journal, 30, 399 (1903).

326 Peterson

nitrogen derivatives, in which the stereoisomerism depends
on the "syn" and "anti" positions of a chlorimide group in
chlorimido esters :

R— C— OR R— C— OR

II II

NCI CIN

Later Stieglitz and Hale^ obtained a second pair of such
stereoisomeric chlorimido esters and showed under what con-
ditions the labile form can be converted into the stable variety.
Finally Hilpert,^ by preparing five other pairs of stereoiso-
mers and determining the conditions for the transformation
of the members of a pair reversibly into each other, estab-
lished their existence on a sufficiently broad basis. These
isomers are interesting, partly because of the simple charac-
ter of the radical (C1+) attached to the nitrogen atom, and
partly because of the light they shed on the question of the
' ' B eckmann Rearrangement . " ^

The stereoisomers heretofore prepared, have in all cases been
chlorimido esters. They made probable the existence of
stereoisomerism also for chlorimido ketones,

R— C— R' R— C— R'

II and II

NCI CIN

comparable with the stereoisomeric ketoximes and hydrazones
of Hantzsch and Werner and of the former's collaborators.
The isolation of such stereoisomers seemed desirable, not only
as confirming the conclusions reached in the work on the
chlorimido esters, but also because it was thought their prepara-
tion would settle, experimentally, important questions connected
with the "Beckmann Rearrangement." In the rearrange-
ment of ketoximes, R — C( : NOH)R', under the influence of
phosphorus pentachloride, many chemists are inclined to as-
sume^ the formation of such chlorimides, R — C(:NCl) — R',

' Unpublished results.

2 This Journal, 40, 155 (1908).

• See a later paper, to be published by Stieglitz, covering unpublished results with
Hale and Eckstein.

* Cf. Hantzsch: Ber. d. chem. Ges., 36, 3579 (1902).

Stereoisomeric Chlorimido Ketones 327

as intermediate products, the first products isolated in the
reaction being arylimidoacyl chlorides. These are supposed
to be formed according to the scheme

R— C— R' CI— C— R'

II — ^ II

CIN R— N

According to tlie theory advanced by Stieglitz, such chlor-
imides would not be intermediate products; the rearrangement
is supposed to be due to the anhydrizing power of phosphorus
pentachloride and the other reagents used to accomplish the
rearrangement, an intermediate product, containing uni-
valent nitrogen, being considered to be the actual substance
undergoing the rearrangement. For instance, we would have }

I
R— C~R' + HCl :z±: R— C— CI ^

II I

NOH HNOH

I
H2O — ^ R— C— CI — > R— C— CI

I II

N NR'

At the suggestion and under the direction of Professor Stieg-
litz, I undertook the problem of isolating, if possible, such

• In confirmation of these views, Stieglitz and Reddick have recently found that
triphenylmethylhydroxylainine, (C6H.5)3C.NHOH, whose structure corresponds to
that of the assumed intermediate addition product with hydrogen chloride, undergoes
the "Beckmann rearrangement" with lemarkable ease and smoothness under the
influence of a dehydrating agent (phosphorus pentachloride was used). This result is
the more noteworthy as it forms a striking contrast to the fact that hitherto all at-
tempts to effect a rearrangement of chlorimidobenzophenone, (C6H5)2C : NCl, have
been unsuccessful (Stieglitz and Watkins). G. Schroeter has recently (Ber. d. chem.
Ges., 44, 1201 (1911)) brought valuable experimental evidence supporting the above
theory. Experiments have been undertaken in this laboratory toward establishing a
connection between the rearrangement and the electrical charges on the atoms involved.
It is thought that, in the final instance, the significance of the r61e of the univalent
nitrogen atom will be found to lie in such electrical relations. In a later paper these
views will be discussed more elaborately and due consideration given to the confirma-
tive experimental work of Schroeter, as well as to the criticisms of J. Stieglitz,
Montagne (Ibid., 43, 2014 (1910)) and others.

and

R-

1
-C— CI

HNOH

328 Peterson

stereoisomeric chlorimido ketones. I have found three pairs
of stereoisomers, viz., stereoisomers of:
(a) Chlorimido-/>-chlorbenzophenone,

CeH— C— CeH.Cl CeH— C— CsH.Cl

II and II

NCI CIN

(6) Chlorimido-/j-methoxybenzophenone,

CeH-C-C«H,0CH3 CeH-C-CeH,0CH3

II and II

NCI CIN

and (c) Chlorimido-/'-chlor-/)-methoxybenzophenone,

ClCeH,— C— C,H,0CH3 ClCeH— C-CeH,0CH3

II and II

NCI CIN

In each case the structural identity of the a and /? forms was
established by converting each, by dry hydrogen chloride, into
the hydrochloride of an imidobenzophenone, which gave v/ith
water the same ketone; a mixture of the ketone obtained
from the a and that obtained from the /? varieties melted at
the same point as either separately. The reactions are:

R— C— R' R— C— R'

II + HCl -^ II + CI,

NCI NH2CI

and

R— C— R' R— C— R'

II + H,0 — > II + NH.Cl

NH2CI O

That the two forms of a pair are not crystal or physical
modifications is shown by the facts that each form has its
own characteristic solubility, each form is recovered un-
changed from its solutions, from its liquid form, when fused,
and from its vapors — even in the presence of added crystals
of the other form.^ Further, each form persists even up to a
temperature of 100°, and a mixture of approximately equal
quantities of the two forms invariably melts lower than either
form alone. One determination of the molecular weight

1 Vide Stieglitz and Earle and Hilpert: Loc. cit.

Stereoisomeric Chlorimido Ketones 329

of the higher melting of a pair of stereoisomers was made,
namely of the a form of chlorimido-/?-methoxybenzophenone;
the higher melting form should, in case of polymerization of
either form, be the polymer. The molecular weight deter-
mination showed it to have the normal molecular weight
for a nonpolymerized body.^

The change of one stereoisomer into another was observed
only once; the lower melting (^) form of chlorimido-/5-chlor-
benzophenone was found, after being kept for three months
during a very hot summer, to have become completely con-
verted, spontaneously, into the higher melting, stable or a form.
Efforts to induce the same change by artificial means, e. g.,
by treatment of the two forms with chlorine, were not successful,
the stereoisomers being unusually persistent, as compared with
the chlorimido esters.- It may be recalled here that Hantzsch
also found the stereoisomeric hydrazones of these ketones to
be extremely stable, resisting transformation into each other
by the agents which are usually successful.^

The stereoisomeric chlorimido ketones, thus far, have not
given the slightest indication of suffering the " Beckmann
rearrangement" which they were supposed to undergo so
readily by those who had assumed them as intermediate
products in the action of phosphorus pentachloride on ketox-
imes. They can be heated to 100° a short time without under-
going any such rearrangem.ent. It is quite obvious, then, that
the view that they form such intermediate products is now
wholly untenable.

EXPERIMENTAL PART
Chlorimidohenzophenone , {C^^^C : NCI. — ^The first chlori-

1 Stieglitz and Earle and Hilpert carried out molecular weight determinations for
both stereoisomers of a number of pairs of stereoisomeric chlorimido esters and in every
case found the molecular weight of each form normal.

- Vide Stieglitz, Hale and Hilpert.

3 By way of the return to the imidohydrochloride of the chlorimido-/>-chlorbenzo-
phenone, the a form was, by subsequent conversion back into a chlorimide, converted
into a mixture of the a and ft forms — this process being an aid in the preparation of
pure material (the hydrochloride of the imidobenzophenone) for the formation of the
chlorimides. As the higher melting a form is always obtained in large excess, the circle
enables one to use the material over and over again to accumulate the lower melting ^
modification. These reactions, while proving the structural identity of the stereoiso-
mers, do not accomplish any direct change of one stereoisomer into the other.

330 Peterson

mide prepared was that of benzophenone. Here, of course, no
stereoisomers were expected, but the work was done for the
purpose of developing a good method for preparing chlor-
imido ketones. Benzophenone was prepared according to
the Friedel and Crafts^ reaction and converted into benzo-
phenone dichloride by means of phosphorus pentachloride
according to Kekule and Franchimont's' directions. Benzo-
phenone imidohydrochloride was prepared from the dichloride
with the aid of urethane by the method of Hantzsch and
Kraft, ^ a yield of forty- two per cent, of the theoretical being
obtained. The best yield was obtained when three mole-
cules of urethane were heated with one molecule of benzophenone
dichloride. The reaction is a very slow one when carried
out at 140°, but the product turns brown, indicating a decom-
position, if the temperature is raised much higher. At this
temperature the mass gives off gases (carbon dioxide and
alkyl chloride) and crystallizes very slowly, and at least three
hours are required for completion of the action. Longer heat-
ing does not increase the yield.

The chlorimide of benzophenone was prepared from the
imidohydrochloride by treatment of the base with hypo-
chlorous acid, much in the same way as the chlorimido esters
are prepared. The action is represented by the equation
(CeH5)2C : NH + HOCl — > (C,H5)2C : NCI + H,0

The hypochlorous acid was prepared from sodium carbonate
according to the directions of Erlenmeyer and Lipp.* A solu-
tion of two equivalents of the acid was prepared and one
equivalent of potassium bicarbonate added to it. Then, while
the solution was at freezing temperature, the solid imidohydro-
chloride was added. For instance, 8.5 grams of the hydro-
chloride was added to a mixture of 4 grams of potas-
sium bicarbonate and of hypochlorous acid (prepared from 1 7
grams of dry sodium carbonate dissolved in 250 cc. of water
and II .5 grams of chlorine).^ The solution became milky as

1 Ann. chim phys., [6] 31, 510.

2 Ber. d. chem. Ges., 6, 909 (1873).

3 Ibid., 24, 3516 (1891).

4 Ann. Chem. (Liebig), 219, 185 11883).

6 Graebe: Ber. d. chem. Ges., 35, 2750 (1902).

Stereoisomeric Chlorimido Ketones 331

the imidohydrochloride dissolved, and a solid gradually crys-
tallized out. After about ten minutes, during which the
mixture was shaken constantly in the freezing bath, chloro-
form was added to dissolve the crystals. The two layers were
separated by means of a separatory funnel, the stoppers of
which were carefully greased to prevent any scratching.
This precaution was deemed necessary because some nitrogen
trichloride v/as liable to be produced in the reaction and the
pungent odor of the mixture indicated its presence.

On evaporation of the chloroform, a yellowish crystalline
mass was left. This mass was purified by crystallization
from chloroform and ligroin and by recrystallization from warm
ligroin. A yield of chlorimide representing seventy per cent,
of the theoretical was thus obtained. The melting point of
the crystals was 37°. The chlorimidobenzophenone thus ob-
tained, when treated with potassium iodide in acid solution,
sets iodine free quantitatively according to the equation

iC,U,),C : NCI + 2HI = (CeH5)2C : NH,C1 + 2I

This was used as a means of analysis, the iodine set free being
titrated with sodium thiosulphate.

o. 1872 gram substance gave 0.0308 gram chlorine.

Calculated for

C13H10NCI

Found

16.44

16.46

The chlorimide (0.4 gram) was dissolved in 25 cc. of ligroin,
and dry hydrogen chloride was passed into the solution ; a fine
white precipitate resulted which analysis proved to be the
regenerated imidohydrochloride.* The yield was quantita-
tive. Analysis of it, by the silver nitrate method, gave the
following results :

0.0776 gram substance gave 0.0128 gram chlorine.

Calculated for
C13H12NCI

Found

16.30

16.49

CI
Treatment of the imidohydrochloride with hot water gave ben-

1 Vide the equation, page 328.

33 2 Peierson

zophenone quantitatively.^ That the substance thus obtained
was the ketone was proved by the fact that it melted at 46°,
the melting point of synthetically prepared benzophenone,
and when it was mixed with the synthetic product its melt-
ing point was not depressed.

Hydrochloride of Imido-p-chlorbenzophenone,
CgHj— C( : NH2CI)— CeH^Cl.— Chlorimido-/>-chlorbenzophenone
was prepared in the same way as chlorimidobenzophenone,
starting from /j-chlorbenzophenone. This ketone was prepared
according to Hantzsch and Kraft's- directions, and when treated
with phosphorus pentachloride according to Overton's^ method
it gave /'-chlorbenzophenone dichloride. Fractionation of
the dichloride was carried out in a Briihl apparatus, the pure
substance distilling at 192° under 12 nun. pressure. A yield
of 88 per cent, of the theoretical was thus obtained.

To prepare /?-chlorbenzophenone imidohydrochloride one
molecule of /?-chlorbenzophenone dichloride was mixed with
three molecules of ure thane and the whole heated to 140°-
160° on the metal bath until the entire mass solidified. This
was then extracted four times with benzene to free it from
unchanged urethane and /^-chlorbenzophenone which is formed.
Recrystallization, by dissolving the crude residue in chloro-
form and precipitating with ligroin, gave the pure white hydro-
chloride. Owing to its difficult solubility in chloroform, it
was found impractical to use this method for the purification
of large quantities of material. I found that the imidoh3''dro-
chloride did not need to be purified before conversion into the
chlorimide. More nitrogen trichloride was formed, but this
did not interfere with the work, and is not dangerous if the
stopcocks are kept well greased with vaseline. The chlor-
imide could then be converted back into the imidohydrochlor-
ide quantitatively by the use of hydrogen chloride. This was
found to be an easier method of preparing the pure imido-
chloride than the method of precipitation spoken of above,
and it was used exclusively after the first preparation. A
yield of 53 per cent, of the theoretical was obtained of the

1 Vide the equation, page 328.

2 Loc. cit.

3 Ber. d. chem. Ges., 26, 28 (1893).

Stereoisomeric Chlorimido Ketones 333

crude material, and this was reduced to 42 per cent, in pass-
ing through the chlorimide. Analysis of the pure substance
gave the following results :

0.1 1 72 gram substance gave 0.0167 gram CI, and 0.1224
gram substance gave 0.0175 gram CI.

Calculated for Found

CsHnNCy I II

lonizable CI 1409 1424 1430

Stereoisomeric Forms of Chlorimido-p-chlorbenzophenone,
CfiHs— C— CeH^Cl CgH— C— CeH.Cl

II and II .— a- and /?-/)-Chlor-

NCl CIN

benzophenone chlorimide were prepared from /j-chlorbenzo-
phenone imidohydrochloride by treatment with hypochlorous
acid. A solution of two equivalents of hypochlorous acid
was prepared as described above and the solid imidohydro-
chloride added to it. When the crude salt, containing am-
m.onium chloride, was used much effervescence occurred.
Great care had to be taken in handling the product, owing to
the formation of nitrogen trichloride. No explosions ever did
occur, but chemists trying the method must work very cau-
tiously. After the imidohydrochloride was added, the flask
was shaken vigorously in the freezing mixture for about ten
minutes. The mixture was then extracted with very little
chloroform and the chloroform solution drawn off in a
separatory funnel. The chloroform was evaporated by a blast
of air, a semisolid mass being left. This was extracted with
warm ligroin. A solution of a mixture of the two forms of
the chlorimide v\ras thus obtained (about 80 per cent, of the
theoretical yield). The two forms were separated as follows:
Crystals obtained by cooling the ligroin extract were dissolved
in as little chloroform as possible and about four volumes of
ligroin added. This precipitated a mass of crystals which
melted at ioi°-i02°. Recrystallization from warm ligroin
raised their melting point to 104°. This form of the com-
pound is called the a form. It appears in thin plates.

' Only the ionizable chlorine is determined, the method used being the titration
of the salt with silver nitrate.

334 Peterson

The chloroform-ligroin filtrate was then cooled to about
— io° for a few minutes. Crystals melting at 46°-48° sep-
arated out. On repeated crystallization of this product from
ligroin its melting point was finally raised to 55°. This form
of the compound is called the ^ form. It appears as thin
prisms or needles. Analysis of the two varieties by the potas-
sium iodide method gave the following results:

o. 1592 gram of the a form gave 0.0225 gram active chlorine
and 0.1 190 gram gave 0.0168 gram active chlorine.

Calculated^ for Found

C,3H9NCl2 I II

Active CI 14.18 14- 14 14.12

o. 1274 gram of the ,5 form gave 0.018 17 gram active chlorine
and 0.0940 gram gave 0.0136 gram active chlorine.

Calculated for
CJ3H9NCI,

Found

112

14.18

14.26

14 -45

Active CI

The identity of the structure of the two compounds was
proved by converting them both into the imidohydrochloride
and then into the ketone as follows:^

Three-tenths gram of the pure a form was dissolved in about
25 cc. of ligroin and dry hydrogen chloride passed into the
solution to saturation. The imidohydrochloride was thus
precipitated. The ligroin was decanted and the last traces
blown off by a blast of air. Then the salt was treated with
warm water. This gave 0.223 gram (89 per cent, of the
theoretical yield) of the ketone, melting at 75°. 5-76°. Mix-
tures of this substance with the synthetically prepared ketone
and with the ketone obtained from the /? form melted at the
same temperature.

Three-tenths gram of the pure /? form was treated in the
same way and gave 0.230 gram of the same ketone as was
obtained from the a form.

' Only one of the two chlorine atoms is determined, it alone being active toward
the hydroiodic acid which was used in the analysis.

- In the preliminary report by Stieglitz and Peterson, as the result of errors in
calculation 14. 19 and 13.97 per cent, was reported.

3 See page 328.

Stereoisomeric Chlorimido Ketones 335

These experiments show conclusively that the substances
are not structural isomers and are the same except in the con-
figuration of the radical ( : NCI) .

The following experiments were carried out to exclude the
possibility of crystal isomerism. One- tenth gram of the a
form was heated to 120° and cooled slowly. At 85° solidifica-
tion took place very rapidly. One-tenth gram of the /? form
was heated to 60°, then cooled to 45° and kept at that tem-
perature for some time. No crystals separated out until
inoculated with a or /? crystals. The melting point was then
found to be 54°; 55° is the melting point of the purest /? form.
A trace of the /? form probably was in the a material and
caused crystallization. The melting point shows that the
substance persisted in the /? form. One-tenth gram of each
form was then brought into solution, separately, so that each
solution was saturated at 25°. Cooling of the solutions then
gave only the crystals from which the solutions were originally
made, as melting point determinations proved. The crys-
tals were again brought into solution, the solutions super-
cooled and inoculated with crystals of the other variety.
Again, nothing but the crystals from which the solutions
were originally made were obtained, as proved by their melt- ^

ing points and habit. /O^ ^/l/W^

Hydrochloride of tttttimido-p-methoxybenzophenone, yi,(o'7h

C6H5C(:NH2C1)C6H^0CH3.— /'-Methoxybenzophenone was pre-
pared according to the method of Gattermann, Ehrhardt, and
Maisch.* The first preparation was made from anisyl chlor-
ide and benzene, but the yield was small. Benzoyl chloride
and anisole gave, by the same treatment, almost a quantitative
yield. This method was therefore used in subsequent prepara-
tions. It was found best to dissolve one equivalent of benzoyl
chloride in ten times its volume of carbon disulphide and to
add aluminium chloride to the mixture in about the same
weight as the benzoyl chloride. A little more than one equiv-
alent of anisole was then allowed to run slowly from a dropping
funnel into the mixture. The reaction was complete as soon
as all the anisole had been added. The carbon disulphide

> Ber. d. chem. Ges.. 23, 1204 (1890).

336 Peterson

was then distilled oflF, the aluminium chloride decomposed
with water acidulated with hydrochloric acid, and the ketone
extracted with ether. Recrystallization from high-boiling
ligroin was the best method found for the purification. The
pure substance melted at 62 °.

The ketone dichloride was obtained by the Hantzsch and
Kraft^ method, except that it was not purified. It was found
that purification was not necessary before treatment with
urethane. The excess of phosphorus pentachloride and the
phosphorus oxychloride were distilled off in a vacuum at
about 100°. The residue was treated directly with urethane.
This gave an impure hydrochloride which could be readily
purified. For instance, 20 grams of ^-methoxybenzophenone
and 20 grams of phosphorus pentachloride were heated in a
flask on a metal bath to 140°- 160° for three hours. The
phosphorus oxychloride was then distilled off in a vacuum.
The residue was treated with 18 grams of urethane and the
temperature again raised to 160° for several hours. The
viscous mass thus obtained was extracted several times with
benzene to remove any urethane, ketone and dichloride. A
white solid remained. The crude yield was 17 grams, or 73
per cent, of the theoretical. The substance melted above
170° with decomposition. When treated with warm water
it gave a substance melting at 55°-58°, which, when mixed
with synthetically prepared ketone (melting at 62°), melted
a little higher than before. These facts indicated that the
substance was impure imidohydrochloride of /?-methoxybenzo-
phenone. The hydrochloride was obtained in a pure state
by converting the crude salt into the corresponding chlor-
imide, which by treatment with hydrogen chloride in ligroin
solution forms the imidohydrochloride again. The analysis
of the salt, purified in this way, gave the following results :

o. 2390 gram substance gave 0.0339 gram chlorine.

Loc. cit.

Calculated for
CuHi40NCl

Found

14.48

14.19

Stereoisomeric Chlorimido Ketones 337

The salt is decomposed into the ketone (melting at 62°)
and ammonium chloride on treatment with water.

Stereoisomeric Chlorimides of p-Methoxybenzophenone,
C„H— C-QH.OCHa CeH-C-CeH,0CH3

li and II .—The chlori-

NCl CIN

mides of /j-methoxybenzophenone were obtained from the im-
pure imidohydrochloride by treatment with hypochlorous acid.
A solution of hypochlorous acid was prepared in the same way
as described for the preparation of benzophenone chlorimide,
and the solid imidohydrochloride of /j-methoxybenzophenone
was added to the solution. When three equivalents of hypo-
chlorous acid were used the product was gummy and hard to
handle. When only two equivalents were used, the product
was crystalline and gave a higher yield of the low-melting
variety of the chlorimide, the one which is the more difficult
to isolate. Use of the pure imidohydrochloride gave a still
more satisfactory result. The rapid evaporation of the chloro-
form used for extracting the chlorimides from the hypochlorous
acid solution also seemed to give a larger yield of the low-
melting variety. In one preparation 10 grams of imidohydro-
chloride was treated with the equivalent of two molecules of
hypochlorous acid. The chlorimides were extracted with chloro-
form and the chloroform evaporated rapidly by a blast of
air. The mass was fractionated by dissolving it in as little
chloroform as possible and reprecipitating part of the solid
with about four volumes of ligroin. This gave a substance
consisting almost entirely of the a variety. After several
precipitations from chloroform by ligroin and recrystalliza-
tions from warm ligroin alone, the true melting point of the
a variety was found to be 90°. These crystals were thin, six-
sided plates. The first chloroform-ligroin filtrate from the
precipitate of the a form was cooled to — 10° and the beaker
scratched. Fine white crystals came out; they were found
to melt at 40^-42 °. These crystals were dissolved in ligroin
in the proportion of i gram of crystals to 25 cc. of ligroin and
the solution set in the ice box for two days in a stoppered
bottle. One very large asymmetric crystal of the /? form and

338 Peterson

several flat ones of the a variety appeared and could be
easily separated mechanically. The melting point of the
large crystal was found to be 54°, which was taken as the true
melting point of the /? form. A mixture of the two forms
melted at 40°, which is far below the melting point of either
form alone. Analysis gave the following results:

o. 1249 gram of the a form gave 0.0177 gram chlorine and
o. 1792 gram gave 0.0254 gram chlorine.

Calculated for Found

C14H12ONCI I II

CI 1444 14- 15 14- 19

o. 1363 gram of the /? variety gave 0.0198 gram of chlorine.

Calculated for
CuHiaONCl Found

CI 14.44 14-52

Conversion of the chlorimides back into the imidohydro-
chloride and the ketone, similar to that carried out with the
stereoisomers of chlorimido-/>-chlorobenzophenone, was effected
as follows: 0.335 gram of the a form was dissolved in chloro-
form and ligroin and dry hydrogen chloride passed into the
solution to saturation. A yellow, sticky mass which smelled
strongly of chlorine separated out. The filtrate was evapo-
rated and gave but a small residue. Treatment of the mass
with warm water dissolved it, and when the solution cooled a
white crystalline solid, ^-methoxybenzophenone, separated
out. After about two hours the crystals were collected on a
filter and dried on a clay plate. A yield of 0.263 gram of the
ketone, or 83 per. cent, of the theoretical, was obtained. Its
melting point was found to be 59°. 5-60°. Then 0.259 gram
of the /? form was subjected to the same treatment. A yield
of o . 200 gram of the same ketone, or 89 per cent, of the theo-
retical, was obtained. The ketone obtained from the /? form,
and mixtures of this ketone with the preparation obtained
from the a form and with synthetically prepared /^-methoxy-
benzophenone, all melted at 59^-60°. There can be no ques-
tion, therefore, of the identity of the structure of the two
chlorimides.

Stereoisomeric Chlorimido Ketones 339

Attempts to convert either variety of the chlorimide di-
rectly^ into the other were made but without success. First
1 . 5 grams of the substance was heated to 100° for three hours.
Its melting point was thereby lowered from 90° to 85 "-87°.
A mixture of the product with some pure a form melted at
88°-89°. Consequently very little transformation, if any,
took place even at 100°. Then o. i gram of the /5 form was
heated in the same manner; the melting point of the substance
was lowered from 54° to 47°. A very small proportion of
a form mixed with the /? form lowers the melting point of the
latter to 40°. There could have been very little change,
therefore, if any, of the /? into the a form. Finally o . i gram
of the substance was heated for three hours in an atmosphere
of dry chlorine^ in a sealed tube. The mass became gummy,
and after having been crystallized from chloroform, decom-
posed slightly at 175° but did not melt. Some transforma-
tion must have occurred, but not the transformation into the
a form. The investigation of this reaction will be continued.

The molecular weight of the a form was determined by the
cryoscopic method: 0.0903 gram substance dissolved in
1 1 . 58 grams benzene lowered the freezing point 0^.1675 ; o . 4791
gram in 11.58 grams benzene lowered the freezing point 0° 835.

Calculated for Found

C14H12NOCI I II

Mol. wt. 242.5 246.8 262.6

The theoretical constant, 53, for the lowering of the freezing
point of 100 grams of benzene by one gram molecule of solute
was used, as in previous determinations with chlorimido
esters. The experimental constant, 49, would give the weights
226.5 ^•iid 238.

p-Chlor-p-methoxybenzophenone, ClCgH^ — CO — CgH^OCHg. —
The method used for the preparation of /'-methoxybenzophe-
none was also used for the preparation of /5-chlor-/>-methoxy-
benzophenone. Thirty- two grams of dry aluminium chloride

> Indirectly the two forms may be converted into each other by treatment with
hydrogen chloride and subsequent change of the regenerated hydrochloride of the
imidomethoxybenzophenone into the chlorimides by means of hypochlorous acid.

- Chlorine effects the reversible change of the labile stereoisomers of the chlor-
imido esters into the stable forms.

34© Peterson

was pulverized in a hot mortar and added to a mixture of
I GO cc. of carbon disulphide and 32 grams of freshly distilled
/?-chlorbenzoyl chloride.^ The mixture was then cooled to
0° and kept at that temperature during the addition of twenty
grams of anisole. The anisole was added very slowly, only
one or two grams at a time, and the mixture was well shaken
between the additions. Hydrogen chloride escaped very
regularly. When the action ceased at this temperature, the
mixture was allowed to warm up to room temperature. The
carbon disulphide was distilled off and a mixture of ice and
water, acidulated with hydrochloric acid, was added to the resi-
due to decompose the aluminium chloride. During this opera-
tion the flask was kept in an ice-salt mixture. The gummy
mass resulting was extracted with benzene, and the solution
shaken five times with sodium hydroxide to remove any
/'-chlorbenzoic acid. The benzene was then evaporated
and the residue crystallized from hot alcohol as follows: The
whole mass was dissolved in the least possible quantity of
boiling alcohol and then the solution was cooled. Thirty- two
and a half grams of crystals separated out. The filtrate was
added to an equal volume of cold water. A further precipita-
tion occurred. This fraction was then crystallized from
alcohol. The entire yield was 37 grams, or 82 per cent, of
the theoretical. The melting point is 125°. Analysis of the
substance gave the following results:

0.2323 gram substance gave 0.5797 gram COj and 0.0980
gram HjO.

Calculated for
CuHnOzCl Found

C 68.18 68.06

H 4.49 4 69

Hydrochloride of Imido - p - chlor-p-methoxybenzophenone,
ClCeH,— C(:NH2C1)— CeH.OCHg.— This was prepared from
the ketone as follows: /?-Chlor-/?-methoxybenzophenone was
heated with one equivalent of phosphorus pentachloride on the
metal bath to 160° for three hours. The phosphorus oxy-
chloride was then distilled off in a vacuum at a temperature

* This was prepared from Kahlbaum's />-chLlorbenzoic acid, melting at 237 ".

Stereoisomeric Chlorimido Ketones 341

of about 100°. Without further purification this crude
product was dissolved in chloroform, and ammonia, dried over
sodium hydroxide, run into the solution to saturation. A
fine white mass, which was mainly ammonium chloride, separa-
ted out. This was filtered off and dry hydrogen chloride
run into the filtrate to saturation. A yellowish precipitate
was thrown down. The chloroform was decanted and the
precipitate washed several times with chloroform and finally
with ligroin. It was then dried in a vacuum over paraffin.
The substance thus obtained gave low values for chlorine
content for the hydrochloride of imido-/'-chlor-/>-methoxyben-
zophenone, but its behavior leaves no doubt that it consists
chiefly of this salt, in an impure condition. From 10 grams
of />-chlor-/?-methoxybenzophenone o . 5 gram of the crude
product was obtained. Water decomposes it into ammonium
chloride and /'-chlor-/?-methoxybenzophenone (melting poinlj
122°, which was raised somewhat by the admixture of the
pure ketone, melting at 125°).

Treatment of the crude salt with hypochlorous acid con-
verts it into the corresponding chlorimide, which is described
below. It was expected that the latter would be converted
by hydrogen chloride back into the hydrochloride of the imide,
but contrary to expectations and to the behavior of other
chlorimides, the chlorimide of />-chlor-/)-methoxybenzophenone
did not form a pure imidohydrochloride when treated with
hydrogen chloride. The following analysis was obtained for
the crude product prepared by the action of ammonia on the
ketone dichloride :

0.2816 gram substance gave 0.0296 gram chlorine.

Calculated for
C,4Hi30NCl2

Found

12.57

10.54

The behavior of the salt towards water agrees with the
structure assigned to it, and the preparation of pure chlorimides
from it (see below) confirms this structure. The preparation
of the hydrochloride by this method is considered interesting,
as it is the first instance of the formation of an imidohydro-
chloride from the aromatic ketone dichlorides by the direct

342 Peterson

action of ammonia. Hantzsch showed that under the most
varied conditions benzophenone dichloride would not form
the imido hydrochloride with ammonia, and had to have re-
course to the use of urethane with the dichloride in order to
prepare the salt. I obtained exactly the same results on re-
peating Hantzsch's work with benzophenone dichloride.

Stereoisomeric Chlorimides of p-Chlor-p-methoxybenzophenone,
ClCoH— C— CeH.OCHg ClCeH— C— CeH.OCHg

II and II .—The crude

NCI CIN

hydrochloride of imido-/?-chlor-/>-methoxybenzophenone (5
grams) obtained in the previous experiment was treated with
the equivalent of two molecules of hypochlorous acid in the
same way that the other chlorimides were prepared. This
gave 3 . 5 grams of a mixture of a- and /9-chlorimides, a yield of
70 per cent, of the theoretical. The product was extracted
with about 25 cc. of cold ligroin (40°-6o°). The solution
was cooled to — 20° and gave i .4 grams of a substance melt-
ing at 5o°-52°. This was again dissolved in ligroin and the
solution set in the ice box in such a manner as to allow a very
slow evaporation of the ligroin. Large crystals, mixed with
some flat plates of the a variety, came out. The large crys-
tals were recrystallized several times from ligroin before a
substance was obtained which gave a sharp melting point,
65°, the highest melting point obtained for this form, the /?
variety. The residue from the first ligroin extraction was
recrystallized from boiling ligroin (boiling at 70°-8o°) several
times and finally gave the pure a form with a melting point
of 94°. 5. Melting points of the two varieties and of mixtures
of them were taken simultaneously and gave the following
results: ^ form, 64°; a form, 94°; mixture, 50^-52°.

Analysis of the two varieties showed the following: o. 1778
gram of the a form gave 0.0227 gram active chlorine, and
o. 1131 gram of the /? form gave 0.01442 gram active chlorine.

Calculated' for Found

CnHuONCla a 0

Active CI 1 2 . 66 1 2 . 75 1 2 . 75

1 Only the chlorine in the (: NCI) group is indicated by the analysis, which was
made with the aid of potassium iodide.

Siereoisomeric Chlorimido Ketones 343

The same kind of evidence as was used with the other
stereoisomeric chlorimides was brought to exclude crystal and
structural isomerism. After melting, each substance crystal-
lizes again in its own characteristic form, even when it has
been heated several degrees above the melting point of the
higher-melting form. When the substances are mixed with
each other the melting point is decidedly depressed, as shown
above. And, finally, each substance crystallizes in its own
form from a solution of a mixture of the two. These facts
exclude crystal isomerism.

The stereoisomeric chlorimidoketones were converted back
into the ketone in the usual way in order to exclude the possi-
bility of structural isomerism. The results follow: 0.266
gram of the a form was dissolved in chloroform and the solu-
tion saturated with dry hydrogen chloride. This treatment
produced a white precipitate. Ligroin was then added to,
complete the precipitation. The chloroform-ligroin mixture
was decanted and the last traces of the solvent blown off in
a blast of air. The residual salt was dissolved in hot
water and gave 0.216 gram of ^-chlor-/)-methoxybenzophe-
none (melting at 124 "-124°. 5). This is a yield of 92 per
cent, of the theoretical. A mixture of the product with the
ketone obtained in the same way from the /? form melted at
the same point and a mixture with synthetically prepared
ketone (melting at 125°) melted at 125°. When 0.173 gram
of the /? form v/as treated by the same process o. 134 gram,
or 90 per cent, of the theoretical weight of ketone (melting at
125°), was obtained.

Attempts to prepare chlorimido-/?-nitrobenzophenone and
chlorimido-/?-methoxy-o-chlorbenzophenone had to be given
up as a result of difficulties experienced in preparing the cor-
responding imidohydrochlorides of the two ketones. A
chlorimide of />-methylbenzophenone was prepared but not
investigated further when it was found that the lower melting
of the two stereoisomers, if present at all, formed an oil diffi-
cult to purify. A crystalline compound, presumably the
higher melting stereoisomer, was not specially purified and
melted between 35° and 45°; it contained 14.62 per cent.

344 ■ Peterson

of chlorine, while a compound CH3CeH,C( : NCl)C6H5 would
demand 15 44 per cent. There is no doubt that such a chlor-
imide may be prepared, but since the main object of this
investigation was to isolate well-defined stereoisomeric com-
pounds, the product was not further investigated. The oil,
according to the analytical results, was not a pure compound,
but contained some chlorimide.

o-Chlor-p-niethoxybenzophenone, ClCgH4.CO.CgH40CH3. — In
the attempt to prepare o-chlor-/7-methoxybenzophenone chlor-
imide, o-chlor-/?-methoxybenzophenone was obtained from
o-chlorbenzoyl chloride and anisole. Since it has not been de-
scribed in the literature, its preparation Avill be briefly dis-
cussed here. Twenty-five grams of o-chlorbenzoyl chloride
was dissolved in 100 cc. of carbon disulphide and 25 grams
of dry powdered aluminium chloride was added to the solution.
The mixture was then cooled in a freezing mixture and the
equivalent of one molecule of anisole allowed to drop into it
slowly from a dropping funnel. Hydrogen chloride was given
off regularly and a crystalline body separated out. The car-
bon disulphide was then distilled off and the aluminium chlor-
ide decomposed by water acidulated with hydrochloric acid.
The residue was extracted with ether and the extract well
washed with sodium hydroxide to free it from any o-benzoyl
chloride. The ether was then distilled off and the residue
fractionated in a Eruhl apparatus at reduced pressure. The
pure substance distilled at 250° under 50 mm. pressure.
Analysis of it shows the following results:

0.1553 gram substance gave 0.3865 gram CO2 and 0.0680
gram HjO.

Calculated for
CkHiiOjCI Found

C 68.18 67.90

H 4.49 4 90

In conclusion, I wish to express my gratitude to Professor
Stieglitz for his kind guidance and for the inspiration he has
given me.

[Contributions from the Sheffield Laboratory of Yale University]

CXCIII— RESEARCHES ON PYRIMIDINES
THE CONDENSATION OF ETHYL FORMATE AND DI-
ETHYL OXALATE WITH SOME PYRIMIDINE-
THIOGLYCOLLATES
[fifty-third paper]

By Treat B. Johnson and Norman A. Shepard

This paper is a contribution to our knowledge of the chem-
ical behavior of the thioglycolUde grouping - — S.CHj.CO — .
The study of this grouping is one of a projected series on or-
ganic sulphur combinations which has been planned for this
laboratory.

In a previous publication, Johnson and Guest^ have shown
that ethyl formate condenses in ether with ethyl benzylthio-
glycollate (I), in the presence of sodium, forming the sodiunf
salt of ethyl a-benzylmercapto-5-hydroxyacrylate (II). This
ester (II) condensed, in an alkaline solution, with pseudo-
ethylthiourea, NH : C(SC2H5)NH2, giving a representative
of a new class of mercaptopyrimidines in which a mercapto
group is linked to the 5 -position of the pyrimidine ring, viz.,
2 -'etliylmercapto-s - benzylmercapto - 6 - oxypyrimidine (III) .
The action of diethyl oxalate on this thioglycoUate (I) was not
examined.

HCOOC2H5

+ '
CeHsCH^SCHXOOC^Hg — ^ HOCH :C(SC7H7)COOC2H5 — ^
I II

NH— CO

I I
C2H5S.C C.SC7H7.

II II
N— -CH

III

Recently Hinsberg^ has observed that the methylene group-
ings in diethyl thiodiglycoUate (IV) are also reactive towards

1 This Journal, 42, 271 (1909).

2 Bex. d. chem. Ges., 43, 901 (1909).

346 Johnson and Shepard

esters, in the presence of sodium. The thio compound con-
denses, for example, with diethyl oxalate, giving the cyclic
compound a, a'- dicarbethoxy- ^,^' - diketotetrahydrothiophene

(V).

(COOC2H5)2 CO— CO

+ I I

C,H500CCH2 CH2COOC2H5 — > CH^OOCCH CHCOOCoH..

s s

IV V

These ketone esters (VI and VII), which are formed by the
condensation of thiogly collates, RS.CHjCOOCoH^, with ethyl
formate and diethyl oxalate, respectively, represent new types
of sulphur compounds. They are of special interest because

OHC.CH(SR)COOC2H5 C2H500C.COCH(SR)COOC2H5

or or

HOCH :C(SR)COOC2H5 CJ-IsOOCCCOH) :C(SR)COOC2H5

VI VII

of their possible use for further syntheses and their in-
vestigation is now in progress. We shall give a description
of the chemical behavior of some of these condensation prod-
ucts in later papers.

It became apparent, after much preliminary experimental
work, that the success of our research depended on the em-
ployment of thioglycoUates which possessed certain charac-
teristic properties. ThioglycoUates, RS.CH2COOC2H5, in which
R was substituted by alkyl groups (CJi^, C2H5, etc.), were un-
suitable because of the firm union between the alkyl group
and the sulphur atom. Ethyl benzoyl thiogly collate,^

CeH5CO.SCH2COOC2H5,

and other acyl thioglycoUates were then tried, but this class
of esters was found to be too unstable because of the easy re-
placement of the acyl radicals by the action of alkali during
the condensations. Furthermore, all the thioglycoUates of
both these types which were examined gave, on condensation
with ethyl formate and diethyl oxalate, products which were

1 wheeler and Johnson: This Journal, 2S, 198.

Researches on Pyrimidines 347

oils or low-melting solids possessing very unfavorable proper-
ties. They could not be purified by distillation, and more-
over the yields were so small in most cases that they could
not be prepared easily in sufficient amounts for synthetical
work.

Efforts have therefore necessarily been made to find an or-
ganic grouping to substitute for R which would confer a crys-
talline character on the compounds under examination, and
also could be removed easily from the sulphur by hydrolysis.
We now find, after an examination of several groupings, that
a 6-oxypyrimidine radical (VIII) fulfils all these conditions.

NH— CO

— C CH (R=-H, CH3, etc.)

II II

N CR

VIII

The only pyrimidinethioglycollate hitherto described in
the literature is ethyl 6-oxypyrimidine- 2 -thiogly collate (XI),
which Wheeler and Liddle^ prepared by the action of ethyl
chloracetate on the sodium salt of 2-thiouracil. We now find
that this ester condenses with ethyl formate and diethyl oxalate
in alcohol and in the presence of sodium ethylate, forming
almost quantitatively ethyl 6-oxypyrimidine-2-(a:-thio-/?-hy-
droxyacrylate) (XII) and diethyl 6-oxypyrimidine- 2 -oxal thio-
gly collate (X), respectively. We have also prepared ethyl
4-methyl-6-oxypyrimidine-2-thioglycollate (XIV) by the action
of ethyl chloracetate on 2-thio-4-methyluracil.^ This new
thioglycoUate condensed as smoothly with ethyl formate and
diethyl oxalate as the ester (XI) and formed ethyl 4-methyl-6-
oxypyrimidine-2-(a-thio-^-hydroxyacrylate) (XV), and di-
ethyl 4-methyl-6-oxypyrimidine-2-oxalthioglycollate (XVI),
respectively. These four /?-ketone esters (X, XII, XV and
XVI) are crystalline compounds melting above 100°, and all
possess physical properties which are very desirable for the

' This Journal, 40, 547.

2 List: Ann. Chem. (Liebig), 238, 12 (1886). Wheeler and McFarland: This
Journal. 42, 105 (1909).

348 Johnson and Shepard

study of the chemical behavior of these types of sulphur com-
pounds.

Ethyl 6-oxypyrimidine-2-(a:-thio-/?-hydroxyacrylate) (XII)
condensed normally with thiourea, forming a representative
of a new type of dipyrimidines, viz., 2-(2-thio-6-oxypyrimi-
dine-5-mercapto)-6-oxypyrimidine (IX). A compound con-
taining the same percentage of nitrogen and having nearly
the same decomposition point as the dipyrimidine (IX) was
also obtained by the condensation of the diethyl 6-oxypyrimidine
(X) with thiourea. The yield, however, was so small that
the identity of the two condensation products could not be
accurately established. Ethyl 4-methyl-6-oxypyrimidine-2-
(a-thio-/?-hydroxyacrylate) (XV) condensed with thiourea,
giving 2-(2-thio-6-oxypyrimidine-5-mercapto)-4-methyl-6-oxy-
pyrimidine (XVIII), while diethyl 4-methyl-6-oxypyrimidine-
2 -oxalthiogly collate (XVI) reacted to form a dipyrimidine of
unknown constitution. We have previously assigned to it
the formula of a tricyclic compound (XVII). Work on the
determination of the constitution of these condensation prod-
ucts of thiourea with the pyrimidine-2-oxalthioglycollates
(X and XVI) is now in progress and the results will be pub-
lished in a later paper.

Especially interesting was the behavior of diethyl 6-oxy-
pyrimidine-2 -oxalthioglycoUate (X) when digested with con-
centrated hydrochloric acid. Uracil was not formed as ex-
pected, but the ester underwent hydrolysis, giving an excellent
yield of 6-oxypyrimidine-2-thiopyruvic acid (XIII). 6-Oxy-
pyrimidine-2-thioglycollic acid undergoes hydrolysis under
the same conditions, and even when boiled with water, giving
uracil quantitatively.' Thiopyruvic acid, HSCHjCOCOOH,
has not been described and this pyrimidine (XIII) should be
of value for further interesting syntheses.

The various transformations described above are repre-
sented by the following structural formulas:

1 wheeler and Liddle: Loc. cil.

Researches on Pyrimidines

349

ffi

f~5

- o

ffi

< ^

^ _

t/2

^ "^^

= o— :25

rz:=o-

= o — o

G=^-

ffi o

K ffi

ffi

s ^

K b
o o

w
12;= b-

b
o
o
■o

Xfi

I
o— o

b
o
o
o

C/5

§ f

A-

m o ffi ffi

i /

3= /

n—n
K O

350 Johnson and Shepard

EXPERIMENTAI^ PART

Ethyl 6-Oxypyrimidine-2-thioglycollate,
NH— CO

I I

C2H5OOCCH2S.C CH.— This pyrimidine, which has been

II II
N CH

described in a previous paper/ is easily obtained by the action
of ethyl chloracetate on thiouracil in the presence of alkali.
To insure, however, a complete reaction it is necessary that
the thiouracil be first completely converted into its sodium salt.
Suspension of the pyrimidine in alcohol containing the re-
quired amount of sodium is not sufficient. Since both the
pyrimidine and its sodium salt are practically insoluble in
alcohol it is necessary to thoroughly shake the mixture and
finally to warm it on the steam bath for a few minutes before
adding the ethyl chloracetate. Fifty-two grams of 2-thio-
uracil were converted into its sodium salt by treatment with
9.6 grams of sodium dissolved in 310 cc. of absolute alcohol.
Fifty-two grams of ethyl chloracetate were then added, in
small portions, and the mixture digested on the steam bath
until the solution gave an acid reaction, and then allowed to
stand for 8-10 hours. The solution was then filtered to sepa-
rate the sodium chloride which had deposited, and finally con-
centrated on the steam bath, when we obtained the thiogly-
coUate. This was washed with cold water to remove traces of
salt, and then purified by crystallization from alcohol or hot
water. It separated in beautiful, prismatic crystals, which
melted at 155° to a clear oil.

The solid material removed by filtration, which was chiefly
sodium chloride, was triturated with water, when a crystalline
substance was obtained insoluble in this solvent. It was also
practically insoluble in alcohol but dissolved in boiling water
and separated, on cooling, in glistening plates. The com-
pound contained no inorganic material and after decoloriza-
tion in hot water with bone charcoal it did not melt below
300°. It was also insoluble in dilute hydrochloric acid and

1 Wheeler and Liddie: Loc. cit.

Researches on Pyrimidines 351

was not altered by digestion with this reagent. 2-Thiouracil
decomposes at 340°. The compound contained sulphur,
and nitrogen determinations agreed with the calculated value
for thioureaacrylic acid. Because of its stability, in the pres-
ence of boiling hydrochloric acid, it was assigned the consti-
tution of the trans modification of this acid, viz., trans-^-
thioureaacrylic acid,

H.C.COOH

Calculated for Found

C4H6O2N2S C4H4ON2S I II

N 1917 21.87 18.78 1919

Ethyl 6-Oxypyrimidine-2-{a-thio-^-hydroxyacrylate),

NH— CO

I I

CHsOOC.C.S.C CH.— Seventy-four hundredths of a gram

II II II

HOCH N CH

of sodium was dissolved in 20 cc. of anhydrous ethyl alcohol
and 3.4 grams of ethyl 6-oxypyrimidine-2-thioglycollate dis-
solved in the cold solution. Two and four-tenths grams (2
mols.) of ethyl formate were then poured into the solution and
the mixture allowed to stand at ordinary temperature for 3
days. No solid material separated and a transparent yellow
solution was obtained. This was cooled to 0°, acidified with
hydrochloric acid and the insoluble sodium chloride filtered
off. When the alcohol filtrate was evaporated at ordinary
temperature the above pyrimidine was obtained as a crystal-
line solid. A small amount of the same compound also separa-
ted with the sodium chloride and was recovered by trituration
with water to dissolve the salt. This compound is very solu-
ble in alcohol and difficultly soluble in cold water. It crys-
tallizes from absolute alcohol in well developed prisms which
melt at 138-140° to an oil with slight effervescence. A mix-
ture of this pyrimidine and unaltered thiogl3^collate (melting
at 155°) melted at 120-130°. The yield was good. Analyses
(Kjeldahl) :

352 Johnson and Shepard

Calculated for
C9H,oO«N2S

Found

II 57

11.03

II . 16

Diethyl 6-Oxypyrimidine-2-oxaUhioglycollate,

NH— CO

I I

C2H5OOC.CH.S.C CH.— This pyrimidine was prepared by

I II II

C2H5OOC.CO N CH

dissolving 150 grains of ethyl 6-oxy pyrimidine- 2- thiogly col-
late and 15.4 grams of diethyl oxalate in 58 . o cc. of cold abso-
lute alcohol containing 3.3 grams of sodium in the form of
sodium ethylate, and then allowing the solution to stand at
ordinary temperature for several days (i week). The solu-
tion was then acidified with hydrochloric acid, when this
pyrimidine was deposited mixed with sodium chloride. After
trituration with cold water to remove the salt it was obtained
as a colorless crystalline solid which crystallized from hot
alcohol in stout blocks. They melted at 171° to an oil and
the yield was 17.5 grams. Analyses (Kjeldahl) :

^^jUi) £/VUi>lZOk Calculated for Found

N ^ '^ 8.9 8.4 8.68

When the acid alcoholic filtrate (above) was allowed to
evaporate at ordinary temperature more of this same pyrimi-
dine was obtained, but mixed with another substance which
did not contain sulphur. The latter compound dissolved in
boiling water and separated on cooling in corpuscular crys-
tals which did not melt or decompose below 300°. It was
identified as uracil; it contained 24.5 per cent, of nitrogen,
while the calculated is 25 per cent.

Action of Hydrochloric Acid on Diethyl 6-0xypyrimidine-2-
oxalthioglycollate
6-Oxypyrimidine-2-thiopyruvic A cid,
NH— CO

HOOC.CO.CH2S.C CH.— One gram of the above thiogly-

II II
N CH

Researches on Pyrimidines 353

collate was dissolved in 20 cc. of concentrated hydrochloric
acid and the solution evaporated to dryness on the steam
bath. The residue was redissolved in 15 cc. of acid and evap-
orated again, when we obtained an oily product which imme-
diately solidified on triturating with alcohol. The crude sub-
stance decomposed at i8o°-i82° with effervescence, leaving
a charred residue. It was purified by recrystallization from
alcohol and separated in colorless crystals which began to
darken at 190°-! 95° and then decomposed at 20o°-2oi°
with violent effervescence. It gave a strong test for sulphur.
6-Oxypyrimidine-2-thioglycollate^ melts at 178° and is de-
composed when boiled with water, giving uracil. It also crys-
tallizes with one molecule of water. Our compound was an-
hydrous. The nitrogen determinations were high for the
pyruvic acid, but this was very probably due to a trace of
uracil which was present as impurity. The experiment vms
repeated and the same compound, melting at 2oo°-2oi°, was
obtained. Analyses (Kjeldahl) :

Action of Alkali on Diethyl 6-Oxypyrimidine-2-oxalthiogly-
collate. — ^Two grams of the pyrimidine and three molecular
proportions of potassium hydroxide (i gram) were dissolved
in 5 cc. of water and the solution heated at 100° for one hour.
After standing for 10-12 hours the solution was acidified with
hydrochloric acid and evaporated to dryness. The residue
was dissolved in boiling water and cooled, when uracil separa-
ted in characteristic, corpuscular crystals melting above 320°.
Analyses (Kjeldahl) :

Calculated for
C7H6O4N2S

Found

13 I

13 6

13-7

Wheeler and Liddle: Loc. cil

Calculated for
C4H4O2N2

Found

25.00

24 52

24.6

354 Johnson and Shepard

Condensation of Thiourea with Ethyl 6-OxyPyrimidine-2-(a-thio-
p-hydroxyacrylate)

2-(2-Thio-6-oxypyrimidine-ymercapto)-6-oxypyrimidine,

NH— CO NH— CO

i I I i

CS C— S— C CH.— Two-tenths of a gram of sodium (2

I II II II

NH— CH N CH

mols.) were dissolved in 10 cc. of absolute alcohol and 0.9
gram of the formyl compound dissolved in the solution. Forty-
two hundredths gram of finely pulverized thiourea was then
added and the mixture heated on the steam bath for about one
hour. The excess of alcohol was then expelled and a residue
obtained which dissolved immediately in cold water. On
acidifying the solution with hydrochloric acid no precipitate
was obtained. An excess of a saturated mercury chloride
solution was mixed with the solution and a small volume of
dilute sodium hydroxide solution added, when we obtained a
yellow, insoluble mercury salt. This was filtered off, washed
with water, suspended in cold water and the mercury pre-
cipitated as mercury sulphide with hydrogen sulphide. After
filtering from the mercury sulphide and concentrating the fil-
trate we obtained the above pyrimidine. It was very solu-
ble in alcohol and hot water. It separated from a saturated
aqueous solution in needles which shriveled when heated in
a capillary tube and then decomposed with efferv^escence
from 285°-295°, according to the rate of heating. It gave a
strong test for sulphur. Analysis (Kjeldahl) :

Calculated for
C8H6O2N4S2 Found

N 22.04 22.11

Condensation of Thiourea with Diethyl 6-0 xy pyrimidine- 2-
oxalthiogly collate. — Five grams of this pyrimidine were con-
densed with 2 . 4 grams of thiourea (2 mols.) by digestion in
35 cc. of alcohol with 0.72 gram of sodium in the form of
sodium ethylate. After heating on the steam bath for two
hours the alcohol was removed by evaporation and the resi-

Researches on Pyrimidines 355

due dissolved in water. On adding hydrochloric acid to this
solution, and cooling to 0°, nothing separated. An excess of
mercury chloride solution was then added, when we obtained
an immediate precipitate of a dense yellow mercury salt.
This was washed with water and decomposed in the usual
manner with hydrogen sulphide and the aqueous filtrate, after
separation of mercury sulphide, evaporated to dryness. We
obtained a very small quantity of a crystalline product which
was very soluble in water and hot alcohol. When recrys-
tallized from 95 per cent, alcohol it separated in needlelike
prisms which decomposed with effervescence at about 295°-
298° when heated rapidly. The compound contained sul-
phur and behaved in every respect like the pyrimidine obtained
by condensation of thiourea with ethyl 6-oxypyrimidine-2-(a-
thio-/?-hydroxyacrylate) . It contained the same percentage
of nitrogen and a mixture of the two compounds decomposed
with effervescence at 290°-294°. Owing, however, to the
indefinite melting point and the small quantity of this com-
pound we were not able to establish accurately the identity
of the two compounds. Its examination will be taken up
again later. Analysis (Kjeldahl) :

Calculated for

CSH6O2N4S2 Found

N 22.04 21.85

Ethyl 4-M eihyl-6-oxypyrinfidine-2-th ioglycollate,
NH— CO

I I

C2H5OOC.CH2S— C CH.— The 2-thio-4-methyl-6-oxy-pyr-

II II

N CCH3

imidine which was used in this preparation was made by
condensing thiourea with ethyl acetacetate in alcoholic
solution.^ Sixty-five grams of this thiopyrimidine were con-
verted into its sodium salt by boiling for one-half an hour
with 325 cc. of alcohol in which were previously dissolved
10.5 grams of sodium. Fifty-five and eight- tenths grams of
ethyl chloracetate were then slowly added and the mixture

1 Wheeler and McFarland: This Journal, 42, 105.

356 Johnson and Shepard

finally heated on the steam bath until the reaction was com-
plete (1.5 hours). The chief amount of the above pyrimidine
deposited, on cooling, with the sodium chloride and was
separated from this impurity by trituration with cold water.
The rest of the compound was obtained after evaporation of
the alcohol filtrate. The pyrimidine is soluble in boiling ben-
zene, alcohol and water. It was purified for analysis by re-
crystallization from 95 per cent, alcohol and separates from
this solvent in colorless needles which melt at i45°-i46°.
The yield was 65 grams. Analyses (Kjeldahl) :

/QJli<^ir'X JJ-^ t^alculatedfor Found

N 12.28 11.92 11.96

4-Methyl-6-oxypyrimidine-2-thioglycoll'ic A cid,

NH— CO

CH. — ^This acid was obtained in the form

II II

N C.CH3

of its dipotassium salt by saponification of five grams of the
preceding thioglycoUate with 2 . 4 grams of potassium hydrox-
ide in alcoholic solution. After heating on the steam bath
for I hour and then cooling the salt separated in colorless
crystals. Analysis (Kjeldahl) :

Calculated for

CjHaOaNzSKa Found

N 10. 1 9.62

The free acid was obtained by dissolving its potassium salt
in cold water and then adding a slight excess of hydrochloric
acid. It separated as a crystalline solid and crystallized
from hot water in colorless prisms which melted at 192°-
197° with efi'ervescence. Analyses (Kjeldahl) :

Calculated for
C7H8O3N2S

Found

14.0

13-78

13 -77

Researches on Pyrimidines 357

Ethyl 4-Methyl-6-oxypyrimidine-2-(a-thio-^-hydroxyacrylate) ,

NH— CO

I I

C2H5OOC.C.S.C CH.— Two grams of sodium and 10

11 II II

HO.CH N CCH3

grams of ethyl 4-methyl-6-oxypyrimidme- 2 -thiogly collate
were dissolved in 40 cc. of cold absolute alcohol and then
6.5 grams of ethyl formate (2 mols.) were mixed with the
solution. After allowing to stand at ordinary temperature
for two days the solution was acidified with hydrochloric
acid, filtered and then allowed to evaporate spontaneously
in the air. We obtained a yellow, viscous substance which
did not solidify until after long trituration with cold water.
It was very soluble in 95 per cent, alcohol and separated,
after the first crystallization, in colorless prisms melting at
75°-85°. After two further recrystallizations from this
same solvent the melting point remained constant at 106°.
Analyses (Kjeldahl) :

Calculated for

C10H12O4N2S

Found

10.9

10.89

lO.i

In a second experiment 4.0 grams of sodium were first dis-
solved in 80 cc. of absolute alcohol and then 20 grams of the
thioglycoUate and 13 grams of ethyl formate dissolved in the
solution. The mixture was then allowed to stand, in a corked
flask, for three weeks, when the excess of alcohol was removed
by evaporation in a vacuum over concentrated sulphuric acid.
We obtained the sodium salt of the condensation product as a
lemon-yellow powder which became pasty on exposure to the
air. Three grams of the salt were dissolved in a few cc. of
water, cooled to 0° and hydrochloric acid added to the solu-
tion, when we obtained a light yellow oil which showed no
signs of solidifying. The oil was separated and washed re-
peatedly with cold water and finally cooled to 0°. There was
still no evidence of the oil solidifying after standing for several
hours. The water was finally decanted and the oil dissolved
in 95 per cent, alcohol and precipitated again by diluting the

358 Johnson and Shepard

alcohol with water. After standing a few minutes it then
solidified and melted at about 85°. On recrystallizing from
dilute alcohol the melting point was finally constant at
io6°-io8°.

The remainder of the above sodium salt was dissolved in
cold water and the solution acidified with hydrochloric acid.
The pyrimidine separated first as an oil which finally crys-
tallized on long standing. It was purified by crystallization
from alcohol and melted at io6°-io8°. A small quantity
of a difficultly soluble compound was separated here. It
contained no sulphur, melted above 290° and was identified
as 4-methyluracil.

Attempt to Reduce Ethyl 4-Methyl-6-oxypyrimidine-2-{a-thio-
(S-hydroxyacrylate) with Sodium Amalgam. — ^Two and eight-
tenths grams of the pyrimidine were dissolved in 75 cc. of 95
per cent, alcohol and a large excess of 2 . 5 per cent, sodium
amalgam added to the solution. A few cc. of water were
then added, the reaction allowed to proceed at ordinary tem-
perature for a few hours and the mixture finally heated for
four hours on the steam bath. The excess of alcohol was
then removed by evaporation, the solution neutralized with
hydrochloric acid, decanted from mercury and finally evap-
orated to dryness. We obtained a crystalline residue which
was easily soluble in cold water. This was dissolved in dilute
acetic acid and the solution allowed to stand, when colorless
prisms separated which melted without further purification
at about i92°-i93° with effervescence. It crystalHzed from
hot water in prisms melting at 193°- 197° and was identified
as 4-methyl-6-oxypyrimidine-2-thioglycollic acid,

NH— CO

I I
HOOCCH2S.C CH

II II

N CCH3

the formyl group of the original pyrimidine having been re-
moved by the action of the alkali. Analysis (Kjeldahl) ;

Calculated for
C7H8O3N2S

Foxrnd

14.0

13.81

Researches on Pyrimidines 359

Condensation of Thiourea with Ethyl 4-Methyl-6-oxypyrimidine-
2-(a-thio-^-hydroxyacrylate)

2- (2-Thio-6-oxypvrimidine-3-mercapto) -4-methyl-6 - oxypyrimi-

NH— CO 'nH— CO

II II
dine, CS C — S — C CH. — Seven grams of the sodium

I II II II

NH— CH N CCH3

salt of ethyl 4-methyl-6-oxypyrimidine-2-(a-thio-/?-hydroxy-
acrylate) were dissolved in 25 cc. of alcohol and condensed
with 2 . 7 grams of thiourea in a manner similar to that de-
scribed in the previous experiments. After heating on the
steam bath for four hours, the alcohol was evaporated
and the residue dissolved in cold water. On acidifying the
solution with hydrochloric acid an orange powder separated.
This substance melted at 145°- 146° and was identified as the
original thiogly collate. The filtrate was treated with an ex-
cess of mercury chloride solution, when an orange mercury
salt deposited. This was decomposed in the usual manner
with hydrogen sulphide, the mercury sulphide filtered off,
and the filtrate evaporated to dryness. We obtained a crys-
talline substance, mixed with a little oil, which was purified
by washing with cold alcohol and benzene to remove the oil
and then crystallized from hot water. It separated in char-
acteristic clusters of prisms which did not melt below 300°.
Analysis (Kjeldahl) :

Calculated for
C9H8O2N4S2.H2O C9H8O2N4S2 Found

N 19.5 20.9 19.38

Diethyl 4-Methyl-6-oxypyrimidine-2-oxalthioglycollate,

NH— CO

I 1

C2H5OOC.CH.S.C CH.— Ten grams of ethyl 4-methyl-6-

I II II

C2H5OOC.CO N CCH3

oxypyrimidine-2 -thiogly collate were dissolved in 40 cc. of
anhydrous ethyl alcohol and condensed with 9.6 grams of
diethyl oxalate in the presence of sodium ethylate. After al-
lowing the mixture to stand for 10 days the solution was al-

360 Johnson and Shepard

lowed to evaporate spontaneously to remove the excess of
alcohol and then diluted with water and carefully acidified
with hydrochloric acid. The above pyrimidine separated as
a colorless, crystalline solid. It deposited from boiling alco-
hol in beautiful stout blocks which melted at i39°-i40°.
When mixed with unaltered material the melting point was
lowered to 120°. Analyses (Kjeldahl) :

Calculated for Found

CHieOeNaS I II

N 8.54 8.66 8.51

When the acid filtrate (above) was allowed to stand a small
quantity of 4-methyluracil separated.

Condensation of Thiourea with Diethyl 4- Methyl-6-oxy py-
rimidine-2-oxalthioglycollate. — Seven- tenths of a gram of so-
dium (2 mols.) was dissolved in 25 cc. of absolute alcohol and
5 . o grams of the above pyrimidine and 2 . 4 grams of thiourea
dissolved in the solution. After allowing to stand for two
days the solution was heated on the steam bath for two
hours. A bright red, jellylike solution was obtained which de-
posited, on cooling, a small quantity of yellow, ragged prisms
(thiourea). These were filtered off, the excess of alcohol
removed by evaporation at 100°, and the product obtained
dissolved in water. On cooling, a little unaltered pyrimidine
separated, which melted at 145° after being crystallized from
alcohol. The filtrate was treated with mercury chloride
and sodium hydroxide, when a dense mercury salt deposited.
This was decomposed as in the preceding experiments, the
mercury sulphide filtered off, and the filtrate then evaporated
to dryness. We obtained a crystalline substance which was
purified by repeated crystallizations from 95 per cent, alcohol.
It separated in hexagonal tables which shriveled at 140° and
then melted at 164°- 165°. Nitrogen determinations agreed
with the calculated value for the two isomeric tricyclic de-
rivatives

NH— CO N— CO NH— CO N— CCH3

II II I II II II

CS C— S C CH or CS C S C CH

I II I II I II II

NH— C— CO— N— CCH3 NH— C — CO— N— CO

Reduction of Mercuric Chloride, Etc. 361

The study of these interesting compounds will be taken up
later. Analyses (Kjeldahl) :

Calculated for Found

C.oHsOaNiSz I II

19.0 18.94 18.84

New Haven, Conn.
July 20, 1911

REDUCTION OF MERCURIC CHLORIDE BY PHOS-
PHOROUS ACID AND THE LAW OF MASS AC-
TION

By James B. Garner, John E. Foglesong and Roger Wilson

In very recent years the velocities of a large number of
chemical reactions have been studied in order to determine
their conformity to the law of mass action. Many of these
confirm the theory that the velocity of a reaction is a func-
tion of the number of molecules taking part in it, yet quite a
number of notable exceptions have been discovered. In the
greatest number of cases, reactions have been investigated
which do not involve more than two or three molecules, since
the chances of two or three molecules coming together in
such a way as to produce chemical action are much greater
than those of four or a larger number coming into a similar re-
lation. It is true that many chemical reactions are repre-
sented as taking place between more than three molecules,
yet a careful study of the velocity of these reactions in terms
of the law of mass action reveals the fact that in many cases
the reaction takes place in steps, or is complicated by side
reactions until the usual simple mathematical expressions
will not yield the results expected from the theory. In this
way, however, the study of the velocity of reactions has been
highly beneficial, because more light has been thrown upon
the mechanism of the reactions themselves than could have
been done by any other means.

The first reaction of the fourth order to be investigated
was the interaction of bromic and hydriodic acids by Ost-

362 Garner, Foglesong and Wilson

wald.^ He found that the values of K obtained from the
equation

K = ^ ( ^ ^

t \a{a — x))

were not constant, but continually decreased with the
progress of the reaction. He did not attempt to find a
solution for this behavior, however, as his purpose was to find
the influence of certain acids upon the velocity of the reaction.
This same reaction was later studied by Magnanini,^ then by
Noyes,^ and finally by Judson and Walker.* It was shown
by Noyes and by Judson and Walker that the reaction was
one of the fourth order. Since that time a few reactions of
this order have been investigated. Such are the action of
heat upon potassium chlorate,^ the reaction between chromic
acid and phosphorous acid,*' and the action of bromine upon
benzene.^

These have shown that in general the number of molecules
interacting determines the order of the reaction. The more
favorable the conditions, the more nearly do the facts con-
firm the theory.

It has been noticed in the work in this laboratory that the
action of phosphorous acid upon mercuric chloride proceeds
very slowly in dilute aqueous solutions and at ordinary tem-
peratures. It has therefore occurred to us that it might be
of interest to determine the extent of the reaction, the effect
of different temperatures and concentrations, and the order
of the reaction.

It has been shown that in aqueous solutions at ordinary
temperatures, mercurous chloride is precipitated as a white,
granular precipitate and orthophosphoric acid and hydro-
chloric acid are formed. The reaction has been expressed
by the equation

2HgCl2 + H3PO3 + HP = 2HgCl + H3PO4 + 2HCI

1 Z, physikal. Chem.. 2, 127 (1888).

2 Gazz. chim. ital., 20, 390.

3 Z. physikal. Chem.. 19, 599 (1896).
■» J. Chem. Soc. 73, 410 (1898).

s J. vScobai: Z. physikal. Chem., 44, 319 (1903).

6 G. Viard: Compt. rend., 124, 148 (1897).

■> L. Bruner: Z. physikal. Chem., 41, 513 (1902).

Reduction of Mercuric Chloride, Etc. 363

At higher temperatures the velocity of the reaction is greatly-
increased, and in addition to the precipitate of mercurous
chloride there is a deposit of metallic mercury. It is now
our plan to show that the above equation is an exact state-
ment of the facts for aqueous solutions of varied concentra-
tions of mercuric chloride and phosphorous acid at 25° and
30°.

Methods of Work

At first it was our endeavor to determine the temperature
and concentrations at which the change took place to the best
advantage, and to find the point of equilibrium, should there
be one. It was found that a solution of phosphorous acid
containing about 3.9 grams per liter and a solution of mer-
curic chloride containing about 26 grams per liter best served
the purpose for preliminary experiments. Mercuric chloride
is too difficultly soluble for more concentrated solutions. At
temperatures below 25° the velocity of the reaction was too
small to permit of accurate determination. At 25° and 30°
the velocities could be very accurately measured. For ex-
ample, 100 cc. of a solution of phosphorous acid were mixed
with 100 cc. of a solution of mercuric chloride in a 500 cc.
flask provided with an aspirator, and put in a thermostat
at a temperature of 30°. At intervals of 5 and 10 hours, 10
cc. were drawn from the flask and titrated with a o . 05 N solu-
tion of sodium hydroxide, with methyl orange as indicator.
In 545 hours, 16.2 cc. of sodium hydroxide were required to
neutralize 10 cc. of the mixture; thus the reaction was prac-
tically complete in that time. This indicated that there was no
equilibrium point.

For the accurate determinations, 500 cc. of each solution
were mixed in a two-liter flask provided with an aspirator,
and placed in the thermostat at the temperatures indicated
in the tables. Fifty cc. were used for each titration. Methyl
orange was the indicator used.

Theory and Formulae
Since it is obvious that in this reaction more than one sub-
stance changes concentration, we can therefore eliminate

364 Garner, Foglesong and Wilson

the possibility of its being of the first order. Our supposi-
tion is that it is of the second, third, or fourth order. If it is
of the second order, since we have used equivalent solutions,
it must conform to the equation

|=/f(a-.)- (I)

^=7(S(^) W

If the reaction is of the third order, it would be expressed by
the equation

'j^ = K(a-.y (III)

^~ 2tl(a~xy a'\

(IV)

Finally, if it is of the fourth order, it would be expressed
by the equation

f^=K(a-.y (V)

3tl(a~xy a^j

(VI)

The calculations were made upon the following assumption:
that the increase in acidity was dependent upon the concen-
tration of the mercuric chloride solution. Hence, by taking
the concentration of the mercuric chloride in 25 cc. and from
the equation

2HgCl2 = 2HCI
541 . 8 : cone, of HgClj : : 72 . 9 : aj,

the total increase can be found. Equating this with the hy-
droxide used in the titrations gives the value in cc. of the hydrox-
ide. This number is then represented by (a) and the amount
changed during any interval of time, equal to the increase in
the amount of hydroxide required, is represented by (x).

Reduction of Mercuric Chloride, Etc.

365

The progress of the reaction was found by subtracting the
number of cc. of hydroxide equivalent to 25 cc. of the phos-
phorous acid solution from the amount necessary to neutralize
50 cc. of the mixture at the time of titration. Substitutions
were then made in equations II, IV, and VI, above.

Table I— {30° C.)

Mercuric chloride, 26 . 75 grams per liter
Phosphorous acid, 3.981 grams per liter

(hours)

a —

- X

K^ X 104

K^ X 100

K* X 109

48.66

39-91

8.75

434

205

471

48.66

12,82

357

168

434

48.66

18.42

304

169

458

48.66

18.61

294

159

446

48.66

18.92

348

160

444

48.66

22.37

202

132

442

48.66

24.17

219

138

439

48.66

25.16

231

145

474

48.66

28.17

174

124

453

48.66

29.56

168

126

488

164

48.66

32.33

120

lOI

480

165

48.66

32.57

122

103

466

The 24.33 cc. of base used in titration are equivalent to
the phosphorous acid contained in 50 cc. of the mixture.

Tablen—{25° C.)

Mercuric chloride, 26.677 grams per liter
Phosphorous acid, i . 99 grams per liter

(hours)

a — X

K^ X 108

K* X 10^

5-33

29.56

23-57

5 99

41.09

13 53

20.0

29.56

19.92

9.64

34 39

14 63

127-5

29.56

12.07

17 49

22.43

13.85

228.0

29.56

19 53

20.03

19.28

13.92

491.0

29.56

8.00

21.56

14.78

13.67

The 14. 78 cc. of base used in titration are equivalent to the
phosphorous acid contained in 50 cc. of the mixture.

366 Garner, Foglesong and Wilson

Table III~{25° C.)

Mercuric chloride, 26 . 4 grams per liter
Phosphorous acid, 0.9378 gram per liter

Time

(hours)

0 — X

K^ X 106

K^ X 107

5-2

13.2

7.61

5-59

I I . 084

11.758

20.0

13.2

5 19

7.91

7.846

II. 197

79.0

132

2.22

10.98

5 507

16.939

The 6 . 60 cc. of base used in titration are equivalent to the
phosphorous acid contained in 50 cc. of the mixture.

Table IV ~{2 5"" C.)

Mercuric chloride, 26.677 grams per liter
Phosphorous acid, 5.971 grams per liter

Time

(hours)

0 — X

K^ X 106

K* X 10'

38.9

61.32

42.26

19.06

2)1 1^

7.60

90.0

61.32

35 08

26.24

30.3

6.97

181. 8

61.32

28.27

33 05

27.0

7-32

248.0

61.32

2571

35-6i

25 I

732

The 42.40 cc. of base used in titration are equivalent to
the phosphorous acid contained in 50 cc. of the mixture.

Table V—{30° C.)

Mercuric chloride, 20.01 grams per liter
Phosphorous acid, 3 . 693 grams per liter

Time

(hours)

a — X

/C3 X 108

K^ X 1

5.66

48.41

43.88

4-58

8.03

1.77

10.16

6.61

I 59

20.33

10.75

1. 61

80.0

18.15

1.72

126.33

21.57

1.77

174 -66

23.57

1.72

249 . 66

25.41

1.68

441.08

27.86

1-73

The 29.51 cc. of base used in titrations i to 3 are equiv-
alent to the phosphorous acid in 50 cc. of the mixture.

The 27.42 cc. of base used in titrations 4 to 8 are equiva-
lent to the phosphorous acid in 50 cc. of mixture.

Reduction of Mercuric Chloride, Etc.

Table VI— (30° C.)

Mercuric chloride, 13.33 grams per liter
Phosphorous acid, 3 . 693 grams per liter

367

Time
(hours)

0 — X

K3 X 10«

K* X IC

516

32.48

2734

5 14

33-73

4.84

22.16

6.95

46.0

10.60

77-75

12.85

121.83

15 19

168.25

16.64

218.5

18.17

The 29.51 cc. of base used in titration are equivalent to the
phosphorous acid in 50 cc. of the mixture.

Table VII-

-Oo° C.)

Mercuric chloride,

6 . 669 grams per liter

Phosphorous acid.

3 • 693 grams per liter

Time

(hours)

0 X

X A-3 X 106

K* X 1

5 90

16.12

12.63

3-57 20.2

1-45

10. 0

16. 12

5.22 22.6

1. 41

22.0

16. 12

6.37 15-0

1.26

38.66

14.02

5-79 12.4

1 .22

131.08

14.02

8.12 9 . 02

1. 14

171-75

16.12

10.89 9.74

1.29

235-33

16.12

11.65 927

1. 14

The 29.51 cc. of base used in titration are equivalent to the
phosphorous acid contained in 50 cc. of the mixture.

Table VIII— (30° C.)

Mercuric chloride, 3 . 333 grams per liter
Phosphorous acid, 3 . 693 grams per liter

Time
(hours)

a — x

K3 X 106

K* X 107

5-0

8.07

6.61

1-35

76.0

10.57

9 9

8.07

5-76

2.31

73-79

II .2

21.75

8.07

4-85

3.21

61.52

10.5

77-3

8.07

3-31

4.76

44.08

10.96

237.0

8.07

2.18

6.89

43-25

10.5

The 29.51 cc. of base used in titration are equivalent to the
phosphorous acid contained in 50 cc. of the mixture.

368 Winston and Jones

CONCLUSIONS

A careful examination of the results of this study of the
interaction of mercuric chloride and phosphorous acid justifies,
we believe, the following conclusions:

(i) That at temperatures of 25° and 30° and with the
varied concentrations, the products of the reaction are mer-
curous chloride, orthophosphoric acid, and hydrochloric
acid. No metallic mercury is deposited.

(2) That there is no equilibrium point, but the reaction
goes to completion and the equation

2HgCl2 + H3PO3 + H2O — > H3PO4 + 2HgCl + 2HCI

adequately expresses the facts.

(3) That since the values for the second and third order
equations decrease rapidly, and those of the fourth order are
reasonably constant, this reaction is obviously one of the
fourth order. Therefore the velocity constant of the change
is expressed by the equation

St lia~xy a^j

Peck Chemical Laboratory

Wabash College

Crawfordsville, Ind.

THE CONDUCTIVITY, TEMPERATURE COEFFICIENTS
OF CONDUCTIVITY AND DISSOCIATION OF CER-
TAIN ELECTROLYTES IN AQUEOUS SOLU-
TION FROM 0° TO 35°. PROBABLE
INDUCTIVE ACTION IN SOLU-
TION, AND EVIDENCE
FOR THE COMPLEX-
ITY OF THE ION

By L. G. Winston and Harry C. Jones
INTRODUCTION

This paper forms one of a series dealing with the conduc-
tivity of electrolytes in aqueous solution. In it we shall take

Conductivity, Temperature Coefficients, Etc. 369

up for consideration the conductivity, temperature coefficients
of conductivity, and percentage dissociation of certain salts,
and shall show how these results confirm those already ob-
tained, and point out some new relations. The work is part
of an investigation which has been carried on in this labora-
tory for a dozen years or more. The importance of such an
investigation is obvious, since chemistry is a branch of the
science of solutions, and one of the very best methods of study-
ing solutions is the conductivity method.

HISTORICAL

Electrochemical theories were advanced as early as 1807
by Davy and by Berzelius. Berzelius was among the first
to call attention to the electrically charged atom. Faraday
appeared later, giving to the world the laws which bear his
name. His work has stood the test of time. His law show-
ing the relation between the quantity of electricity and amount
of decomposition holds rigidly to-day, and in the light of the
electron theory takes on a new meaning. In the years 1853
to 1859 Hittorf determined the relative velocities of the ions
of many salts. He pointed out a relation between chemical
activity and conductivity, and also called attention to the
analogy existing between solutions and gases. This latter
problem was taken up later by Raoult, Ostwald, van't Hoff,
and others. The laws of Raoult, dealing with the lowering of
the freezing point and vapor pressure of liquids, and Ostwald 's
dilution law are well known. Van't Hoff, in 1887, working on
osmotic pressure, found certain solutions that behaved ab-
normally. Arrhenius, attempting to explain their behavior,
pointed out the fact that salts and analogous substances
break down into ions. Thus was given to the world the theory
of electrolytic dissociation. Its truth is attested on every
hand. Facts once inexplicable become wonderfully clear
and lend confirmation to the theory. Many workers have
appeared in the field since Arrhenius. The most important of
these, perhaps, is Sir J. J. Thomson, whose brilliant experi-
ments have well-nigh revolutionized our conception of mat-
ter.

370 Winston and Jones

The result of the work already done may be summarized
briefly as follows: The conductivity of electrolytes in solu-
tion is dependent primarily on two things, viz., the number
of ions and their velocity. These two factors may be affected
by various others. The most important of these is tempera-
ture. The effect of rise in temperature is chiefly to increase
the velocity of the ions. The number of ions would not be greatly
affected unless they were complex. In addition to the effect
of temperature on the number and velocity of ions in solu-
tion, there are still other factors which, for convenience, may
be divided into three classes :

1. Those dependent upon the solute.

2. Those dependent upon the solvent.

3. Those dependent upon the combination of the solvent
with the solute.

In class I — factors dependent upon the solute — mention
should be made first of all of the effect of valence. This would
determine largely the number of ions capable of entering into
solution. As is well known, the conductivities of binary,
ternary and quaternary compounds are found to vary con-
siderably. Factors affecting the velocity of the ion would
be the atomic weights and atomic volumes of the elements ex-
isting in the compound. We would naturally expect that the
velocity would be an inverse function of the atomic weight
and atomic volume. Experimentally, however, this has not
been found to be true. Jones and Pearce^ found that those
elements which have the smallest atomic volumes have the
greatest hydrating power. This would tend to diminish
their velocity.

As to the factors dependent upon the solvent, the most im-
portant are its viscosity, its dielectric constant and its asso-
ciation.

In class 3 should be placed the concentration of the solu-
tion and the power of the solute and solvent to form solvates
with one another.

The conductivity of solutions has been studied from each of

1 This Journal. 38, 737 (1907).

Conductivity, Temperature Coefficients, Etc. 371

these standpoints, and much valuable data have been ac-
cumulated.

The efifect of temperature has been worked out carefully
by Jones and his coworkers, West,^ Jacobson,^ Clover,^ West,*
White,^ Wightman,® and Hosford.^ Conductivity always
increases with rise in temperature from 0° to 65°, while dis-
sociation usually decreases slightly. The decrease in disso-
ciation would tend to diminish the number of ions, and thus to
lessen the conductivity, but this effect is more than offset
by the increased velocity of the ions due to rise in tempera-
ture. This decrease in dissociation may be accounted for in
two ways. It may be due to a decrease in the association of
the solvent, which would tend to decrease the dissociation of
the dissolved substance; or it may be due to the fact that a
rise in temperature diminishes the dielectric constant of the
solvent and consequently its dissociating power, since, accord-
ing to the Thompson-Nemst hypothesis, a substance having
a high dielectric constant has great dissociating power. While,
as just shown, the effect of temperature is to diminish the
number of ions present, its effect on the velocity of ions is
just the reverse. Rise in temperature increases the velocity
of ions in two ways: First, it diminishes the viscosity of the
solvent. Second, rise in temperature" would decrease the
complexity of the hydrates formed. This also would tend
to increase the velocity of the ions. At all events, the de-
crease in the number of ions seems to be more than compen-
sated for by the increase in their velocity, and the general
effect of rise in temperature is, therefore, to increase the con-
ductivity.

The most important factor in its effect on conductivity
with rise in temperature is hydration. That the dissolved
substance combines with some of the solvent to form solvates
seems now to be an undisputed fact, the existence of hydrates

1 This Journal, 34, 357 (1905).

2 Ibid.. 40, 355 (1908).

3 Ibid.. 43, 187 (1910).
* Ibid.. 44, 508 (1910).
5 Ibid., 44, 159 (1910).
>^ Ibid.. 46, 56 (1911).

7 Ibid..i», 240 (1911).

372 Winston and Jones

in solution being shown by several independent lines of evi-
dence.* The close connection between hydration and water
of crystallization has also been established in this laboratory.

Important relations between amount of hydration and
temperature coefficients of conductivity have been pointed
out. Jones and his coworkers, Bingham, McMaster, Rouil-
ler,- Veazey,^ Guy,^ Davis, Reinhart, Mahin,^ Schmidt,* and
Kreider'' have made important observations on the effect of
viscosity on the conductivity of electrolytes.

The work in this laboratory has been extended to non-
aqueous solutions. Apparatus has been improved, the range
of temperature has been extended, old sources of error have
been eliminated, and the conductivities of hundreds of com-
pounds have been added to those already measured.

The problem has been undertaken in this laboratory of
measuring the conductivity of all of the more common acids,
bases and salts in aqueous solution, from o° to 65°, and of
calculating the dissociation whenever possible.

This work will be pushed forward as rapidly and carefully
as possible.

One fact, overlooked thus far in the consideration of the
conductivity of electrolytes, is the probable inductive action*
of the ion on the unionized molecule. In the solution of a
salt there is every condition necessary for inductive action.
There are the charged ions, the neutral molecules and the
dielectric or solvent. Ordinary electrical induction in con-
ductors, as is well known, takes place as follows: A charged
body brought near to a neutral body, but separated from it
by a dielectric, causes a separation of the electricity in the
neutral body, drawing the opposite kind nearest to itself and
repelling the like charge to the side farthest from 'itself. If,
while the charged body is still near, the repelled charge in the

' Publication No. 60 of the Carnegie Institution of Washington.
- Publication No. 80 of the Carnegie Institution of Washington.
■"' This Journal. 41, 433 (1909).
^ Ibid.. 46, 131 (1911).
^ Ibid.. 41, 433 (1909).
<^ Ibid., 42, 37 (1909).
^ Ibid., 45, 282 (1911).

* See paper in This Journal (46, 547, (1911)) by L. G. Winston, to whom this
entire suggestion and its applications in tliis and the preceding paper are due.

Conductivity y Temperature Coefficients, Etc. 373

conductor is removed by contact with some other body, on
the removal of the charged body the once neutral body would
be left charged with the opposite kind of electricity. The
ion, a charged body, acting through the water (a dielectric)
on an unionized molecule, would produce just such an efifect.
Several results may follow from this. First, a positive ion
brought near to a neutral molecule, but separated from it by
the nonconducting water, would cause a separation of the
electricity in the molecule; the negative will be drawn near
to the ion and the positive repelled. Suppose, for instance,
that the repelled charge is not removed, the charged ion
would simply attach itself to the molecule, and as a charged
system move through the solution. Moreover, this charged
system could play the part of the original ion and, acting
through the water, in a similar way draw other molecules to
itself. There would be a limit, of course, to the number of
molecules which could thus be attached. This, no doubt,
would be a function of the valency of the ion.

If, on the other hand, the repelled charge is removed and
the inducing ion then moves off, the once neutral molecule
would be left charged with a sign opposite to that on the in-
ducing ion, and moving through the solution would be able
to attract other molecules or oppositely charged ions to itself.
This, of course, would give rise to a great complexity of ions
and molecules. The velocity of the ions would thus be greatly
affected, because their masses would be greatly increased.
This may in a measure account for the apparent discrepancy
between the dissociation as found by the freezing point method
and that found by conductivity, since by this inductive ac-
tion there would be brought about a change in the number of
particles which would probably affect the dissociation as
found by the freezing point method.

The effect on conductivity, on the other hand, would be
due rather to a change in the velocity of the ions. The com-
plex ions would tend to move more slowly than the individual
ion, thus making the conductivity measurements of dissocia-
tion too low. The change in the number of particles would
not be so apparent in the case of conductivity because, when,

374 Winston and Jones

by means of induction, an ion attaches itself to a neutral
molecule, it would still give rise to a charged system, and
would not thereby reduce the number of charged particles
in solution. The breaking up of these moving systems by
heat would show itself in increased temperature coefficients.

Jones and Pearce* have shown that the dissociation as
measured by the conductivity method is less than that cal-
culated from the freezing point lowering. Conditions were
chosen such that the number of ions, velocity of ions, hydra-
tion and viscosity were the same in both cases. It was found
by them that the greater the dilution, the greater the differ-
ence in dissociation as measured by the two methods. This
is due to the fact that the complexity of the hydrate is greater,
the greater the dilution.

Evidence seems to be accumulating in many directions
that the ions in solution are complex. Some interesting re-
lations are brought out in connection with the various dilu-
tion laws, to which sufficient attention has not as yet been
directed, which apparently point to the complexity of mole-
cules in solution. Ostwald's law,

(i~a)V

has been found to apply to weakly dissociated electrolytes,
but not at all to strong electrolytes. Moreover, various dilu-
tion laws have been formulated which apply to strong elec-
trolytes but are extremely unsatisfactory when it is attempted
to apply them to weaker electrolytes.

The question naturally arises, why this difference? The
thought has suggested itself that it may be due to the com-
plexity of the molecule — one dilution law applying to solu-
tions containing molecules of a certain complexity, while an-
other would apply to solutions containing molecules of a
different order of complexity. Of the many dilution laws
for strong electrolytes only two will be considered, viz., that
of Rudolphi and that of van't Hoff . The Rudolphi formula is

1 This Journal. 38, 743 (1907).

Conductivity, Temperature Coefficients, Etc. 375

= K

Van't Hoff's is

(i — a)^|v

(i~ayv~

Since the Ostwald law,

= /iC

(i-a)l^

applies to weakly dissociated electrolytes, in solutions to which
it applies there are very few ions. If the Rudolphi formula
is applied to a solution, a certain volume, Vi, is obtained,
corresponding to a definite value for a and for K. If now,
retaining the same values as before for a and for K, the Ost-
wald formula is applied to the same solution, there is obtained
a volume V which is the square root of the volume obtained
by the Rudolphi formula. In other words, there is found
the relation Vl^j/F = i, a relation which would indicate
complexity of the molecule in solutions to which the Rudolphi
formula applies. Treating the van't Hoff formula in the
same way, i. e., comparing the volume obtained by the use
of the van't Hoff formula with a certain solution, for a definite
value of a and of K, with the volume obtained by the use of
the Ostwald formula for the same solution, keeping a and K
the same as before, there is found the relation

V_ __ a
V,~ I — a

where V represents the volume when the Ostwald law was
applied and V^ the volume obtained when the van't Hoff law
was used. Now if V/Vy = constant, the molecule would be
simple in each case ; but on examining the formula it is readily
seen that the relation is not a constant one, but that it is a
function of the dissociation. This would indicate complexity
of the molecule in solutions to which the van't Hoff law applies.
The interesting fact about this last relation is that the degree of
complexity varies with the dissociation, i. e., with the number of
ions present; just exactly what has been referred to above as

376 Winston and Jones

the probable result of inductive action. Let us now turn to
the consideration of the data in hand.^

EXPERIMENTAL

The well known Kohlrausch method was used to deter-
mine the conductivities. A Kohlrausch slide wire bridge
was employed with an induction coil and telephone receiver.
The cells used were of the type designed by Jones and Bing-
ham.^ The cell constants were redetermined at regular,
short intervals. The measurements were made at o°, i2°.5,
25°, and 35°. Three separate readings were taken for each
solution at each temperature, different resistances being used
for each reading. The average of the conductivities obtained
by using each of these readings was taken to be the correct
conductivity.

The flasks and burettes were carefully calibrated at 20°
by the method of Morse and Blalock.^

Solutions

Kahlbaum's "chemically pure" materials were taken as
the starting point in almost every case. These were purified,
whenever practicable, by crystallization. A solution some-
what more concentrated than the most concentrated solution
to be used was made up. Its strength was determined by
volumetric or gravimetric methods, and the solutions pre-
pared from it as a mother solution. This solution was made
by direct weighing whenever it was possible, and in the
measurements given below this method was always used un-
less otherwise stated.

Water

The water used in making the solutions was prepared ac-
cording to the method of Jones and Mackay,^ which has been
employed in this laboratory for many years. This method is

' It should be stated that all of the above relations have been worked out entirely
independently by L. G. Winston.

2 This JotiRNAL, 34, 493 (1903).

3 Ibid., 16, 479 (1894).
* Ibid., 19, 91 (1897).

Conductivity, Temperature Coefficients, Etc. t^TJ

too well known to need discussion here. The water thus ob-
tained had a conductivity of about 0.9 to 1.3 X io~^ at 0°.

Discussion of Results

The following salts have been classified, approximately,
according to the position of the metal in the Periodic System.
The ammonium, potassium and sodium compounds would,
therefore, be first in order. These are, therefore, grouped to-
gether for consideration. A careful examination of the re-
sults for these compounds will show some points of interest.

(i) The difference in the conductivities of the binary,
ternary and quaternary salts is quite evident. The conduc-
tivity of ammonium nitrate, potassium acetate, and potas-
sium permanganate, between 0° and 35°, ranges from 46 at
0°, in the most concentrated solution of potassium acetate,
to 163.62 at 35° in the most dilute solution of ammonium
nitrate. The conductivity of those compounds which are
not binary, viz., ammonium sulphate, acid ammonium sul-
phate, dipotassium phosphate, sodium sulphate, and borax,
at 35° in the most dilute solutions is, in every case, above
200, and for acid ammonium sulphate is considerably above
500.

The very high values for the temperature coefficients of
conductivity, expressed in conductivity units, in the case of
the four sulphates is very noticeable. The highest values
are 5, 6, and 7+ in the case of sodium sulphate, ammonium
sulphate and acid ammonium sulphate, respectively; while
for the other salts under consideration in this group, the
temperature coefficients in conductivity units is 4 + . This
is probably due to the fact that sulphates show a tendency
towards polymerization.

The very largest temperature coefficient of conductivity
of this group belongs to acid ammonium sulphate. It is
7.96. This is doubtless accounted for by the fact that this
salt breaks up into very complex ions.

In the case of potassium acetate and potassium perman-
ganate, it is somewhat peculiar that the temperature coeffi-
cients of conductivity in per cent, are in both cases, from 0°

378

Winston and Jones

through 25°, larger than those measured in conductivity
units.

It is also striking that in the case of acid ammonium sul-
phate the temperature coefficients of conductivity decrease
with rise in temperature.

In dealing with the following data the percentage dissocia-
tion is not discussed for the individual salts, but by means of
curves which are given after the data their points of differ-
ence are brought out.

Ammonium Nitrate
Table I — Conductivity

lOI

119

113

135

123

146

128

152

512

lOI

132

157

1024

102

134

159

2048

103

134

160

4096

105

5I-

137

163.62

Table II — Temperature Coefficients

o°-

-12°. 5

12°.

5-25°

25°-35°

Cond.

Per

Cond.

Per

Cond. Per

V units

cent.

units

cent.

units cent.

2 I

2.81

I. 81

2.29

1.80 1.77

8 I

2.47

2-33

2.77

2.17 I

32 2

2.97

2.31

2.45

2.34 I

128 2

3.00

2.40

2.44

2.45 I

512 2

3.02

2.50

2.47

2.48 I

1024 2

2.99

2-55

2.49

2.50 I

2048 2

2.99

2.52

2.44

2.56 I

4096 2

305

2-59

2.46

2.58 1.87

Table HI-

-Percentage

Dissociation

0°

12°. 5

25°

35°

76.5

74.8

73-6

730

79-9

82.2

82.6

89.4

893

90.0

128

93-3

93-2

93-5

512

96.1

96.2

96.3

1024

97.2

97-5

2048

98.0

97.8

98.0

4096

)0.0

>o.o

lOO.O

Conductivity, Temperature Coefficients, Etc.

379

Ammonium

Sulphate

Table IV —Conductivity

0°

12°. 5

25°

35°

82.37

I I 2 . 09

145 09

170.72

136.28

179

213

115

160.26

210

254

128

130.

182.65

241

291

512

139

195 -77

259

313

1024

143-

202.31

267

322

2048

150.

209.74

275

337

4096

150

211.55

280

340

Table V—

Temperature Coefficients

0°-12°.5

12°. 5

-25°

25°-35°

Cond. Per

Cond.

Per

Cond. Per

V units cent.

units

cent.

units cent.

2 2

38 2.89

2.64

2.36

56 I . 76

8 3

06 3

. 12

2-54

36 1.87

32 3

60 3

. 12

2-53

39 2.08

128 4

14 3

.16

2-57

03 2 . 08

512 4

49 3

.21

2.60

38 2.08

1024 4

68 3

-25

2.58

49 2.05

2048 4

73 3

.14

2.53

15 2.23

4096 4

89 3

-25

2.58

95 2.12

ra6/e y/-

-Percewtoge

Dissociation

0°

12°. 5

25°

35°

54-6

52.9

51-6

50.1

6=:

128

512

-7

1024

2048

IOC

4096

100

.4 ad A

mmonium

Sulphate

Table VII — Conductivity

0°

12°. 5

25°

35°

155-26

186.49

211.99

226.06

183.4

223

258

277

223 -5

279

322

349

128

265.2

339

404

444

512

289.7

378

463

522

1024

295.2

386

483

547

2048

303-4

400

496

573

4096

304.2

401

)

497

576

38o

Winston and Jones

Table VIII — Temperature Coefficients

0°-12°.5 12°. 5-25° 25°-35<

Cond. Per

Cond.

Per

Cond.

Per

units cent.

units

cent.

units.

cent.

50 I. 61

2.04

I .09

0.66

04 2 . 20

I .22

48 2 .00

1.23

128

90 2.22

1-54

512

08 2 . 44

1.80

1024

33 2.48

2 .00

2048

73 2.55

I 94

4096

81 2.57

1.89

Table IX—

■Percentage

Dissociation

0°

12°. 5

25°

35°

51.0

46.4

42.7

39 2

60.3

55-7

48.1

73-5

69.6

60.6

87.1

84.4

77.1

95-2

94 2

90.5

1024

97 0

96.3

94 9

2048

99-7

99 6

99 4

4096

100. 0

100

100. 0

Sodium Sulphate

Table X — Conductivity

0°

12°. 5

25°

68.49

97 54

129.13

156

III

146.40

178

132

176.76

215

128

107

152

203 10

247

512

117

166

221 .21

269

1024

119

169

226.34

276

2048

125

176

235 -35

287

4096

127

181

243.42

294

Table XI—

Temperature Coefficients

0"'-12°.5

12°. 5

-25°

25°

-35°

nd. Per

Cond.

Per

nd.

Per

lits cent.

units

cent.

units

cent.

32 3-39

2.53

2-59

2.14

63 3

2.51

2.17

05 3

2.65

2.17

128

59 3

2.66

2.16

512

90 3

2.65

2.18

1024

00 3

2.68

2.19

2048

01 3

2.69

2.20

4096

2.72

Conductivity, Temperature Coefficients, Etc.

381

Table XII — Percentage Dissociation

12°

53-6

53-7

53 I

53-2

61 .4

60. 1

73 I

72.6

128

84.0

834

512

91 .6

90.9

1024

93 4

93 0

0 .

2048

97 0

96.7

4096

100. 0

lOO.O

Borax

100. 0

Table XIII — Conductivity

0°

12°. 5

25°

35°

57 99

83-76

113-54

139-83

125

154

128

104

141

174

512

112

152

187

1024

153

189

2048

119

161

198

4096

122

163

202

Table XIV-

-Temperature Coefficients

0°

-12°. 5

12°. 5

-25°

25°-35°

Cond.

Per

Cond.

Per

Cond. Per

V units

cent.

units

cent.

units cent.

16 2

3-55

2.38

2.84

63 2.32

32 2

2.62

2.83

91 2.32

128 2

2.95

2.82

28 2.32

512 2

3.18

2.83

60 2.37

1024 2

3.21

2.83

60 2.35

2048 2

3-33

2-79

71 2.30

4096 2

3-34

2.73

87 2.36

Ta&Ze XF-

-Percentage

Dissociation

0°

12°. 5

25°

35°

67.8

68.5

69.2

69.0

128

512

1024

2048

4096

382

Winston and Jones

Potassium Acetate
Table XVI — Conductivity

12°. 5

•25°

105

117

128

102

123

512

105

126

1024

106

129

2048

108

129

4096

108

129

Table XVII — Temperature Coefficients

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

1.32

2.86

1.66

2.64

1.65

1.98

1. 71

2-55

1-74

I 97

1.90

2.58

2 .02

2.07

128

1.98

2.54

2 .09

2.05

512

2.02

2.53

2.17

2.06

1024

2 .06

2-54

2.23

2.09

2048

2 . II

2.57

2.14

I 97

4096

2.14

2.61

2.13

I 96

Table XVIII — Percentage Dissociation

12°

77.8

76.6

76.9

128

512

1024

2048

100

4096

100

Potassium Permanganate

The strength of the mother solution was determined volu-
metrically by means of potassium tetroxalate.

Conductivity, Temperature Coefficients, Etc.
Table XIX — Conductivity

383

°.5

35°

59-34

80.17

104.36

124.74

113

136

128

119

142

512

117

141

1024

113

137

2048

133

4096

III

133

Table XX—

■Temperature Coefficienti

0°

-12°. 5

12°. 5-

-25°

25°-35°

Cond.

Per

Cond.

Per

Cond. Per

V units

cent.

units

cent.

units cent.

8 I

2.81

I 94

2.42

04 I . 96

32 I

2.93

2.13

2-45

24 I 97

128 I

2.95

2.23

2.44

31 1-94

512 I

2.96

2.14

2-35

36 2 . 00

1024 I

3.02

I 99

2.24

.31 2.03

2048 I

2.87

1.94

2.24

22 2.00

4096 2

323

1. 91

2.17

22 1.99

Table XXI-

-Percentage Dissociation

0°

12°. 5

25°

35°

88.8

87-7

87.5

87.6

128

100

lOO

100

512

1024

2048

4096

Dipotassium Phosphate

This salt was precipitated by magnesia mixture and the
phosphoric acid thus determined.

Table XXII — Conductivity

)°

12°

63.01

86.82

113.04

138.16

109

143

174

127

167

203

128

102

142

188

230

512

107

150

199

239

1024

109

152

200

242

2048

157

206

242

4096

107

154

201

250

384

Winston and Jones

Table XXIII — Temperature Coefficients

O^-W.S 12°. 5-25° 25'

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

3 03

2. 10

2.42

2-51

2.22

2.50

2.21

2.53

2.16

128

2-57

2.27

512

2.57

2.03

1024

2.54

2.10

2048

2.50

1.77

4096

2-43

2.42

Table XXIV-

-Percentage Dissociation

0°

12°. 5

570

55-3

71.7

83.0

92.8

97 6

1024

99 0

2048

100. 0

100

4096

[OO

The group consisting of strontium acetate and magnesium
bromide, nitrate, formate and acetate will be considered next.

There is nothing special to note in the case of strontium
acetate. It is readily hydrolyzed, and any irregularities
might easily be attributed to this fact. Attention might be
called, however, to the increase in percentage dissociation
with rise in temperature.

It is interesting in considering the data of the four mag-
nesium compounds to discover, if possible, the effect of the
different anions. Of course, the water of crystallization
would also be a factor. This is the same, however, in the
case of the bromide and nitrate, and any difference in the
conductivity of these two compounds may correctly be attri-
buted to the different anions.

On examining the data for these substances, it is readily
seen that the conductivity of magnesium bromide is decidedly
greater than that of magnesium nitrate. Its temperature
coefficient of conductivity is also larger. This would point

Conductivity, Temperature Coefficients, Etc.

385

to some difference in the anions either as to velocity or com-
plexity. Apart from their remarkable similarity, magnesium
acetate and formate present nothing of special interest.

Strontium Acetate

The strontium was precipitated and weighed as the car-
bonate.

Table XXV— Conductivity

0°

12°. 5

25"

35°

34-94

49.26

66.52

81. II

80.19

106

129.99

100.20

135

164.88

128

117.19

157

193 -44

512

128.09

170

209.22

1024

131.09

177

218.24

2048

139.01

180

219.77

4096

139.60

184

224.75

Table XXVI — Temperature Coefficients

0''-12°.5 12°. 5-25° IS'-SS'

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent:

115

3-29

1.38

2.80

1.46

2.20

2.67

2.15

2.79

2.19

128

2.77

2.27

512

2.63

2.30

1024

2.82

2.30

2048

2.36

2.21

4096

3-34

3 41

3-59

2.57

403

2.19

Table XXVII-

-Percentage Dissociation

12°. s

35-7

35-3

128

1024

2048

4096

100

[GO

386

Winston and Jones

Magnesium Bromide

The magnesium was prcipitated as ammonium magnesium
phosphate, and weighed as the pyrophosphate.

Table XXVIII — Conductivity

104

132

162

130

170

206

104

147

194

235

128

113

159

211

257

512

118

167

223

270

1024

122

173

230

279

2048

127

179

238

289

4096

130

185

244

305

Table XXIX — Temperature Coefficients

0°-12°.5 12°. 5-25° 25°-35°

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

2.91

2.31

2.22

2-93

2.20

2.49

2.08

2.56

2. II

128

2.60

2.14

512

2.64

2. 12

1024

2.65

2. ID

2048

2.63

2.13

4096

2.59

2.49

Table XXX-

-Percentage

Dissociation

0°

12°. 5

58.3

56.2

1024

2048

4096

100

]

Magnesium Nitrate
The magnesium was weighed as the pyrophosphate.

Condtictivity, Temperature Coefficients, Etc.

387

Table XXXI — Conductivity

0°

12°

25°

35°

88.91

123.42

160.86

191.88

101.55

141

187.10

223.24

128

[10.78

155

204.72

247.66

512

[19.01

165

220.89

265.33

1024

[20.68

170

224.49

272.30

2048

[23-34

173

229.70

280.09

4096

[22.89

173

229.58

277-54

Table XXXII-

—Temperature Coefficients

0"=

-12°. 5

12°. 5

-25°

25°-35°

Cond.

Per

Cond.

Per

Cond. Per

V units

cent.

units

cent.

units cent.

8 2

3.10

2 99

2.42

3.10 1.93

32 3

2-54

3-6i 1.93

128 3

2.28

4.29 2.10

512 3

2.66

4.44 2.01

1024 3

2-55

4.78 2.12

2048 3

2.61

5.04 2. II

4096 4

2.57

4.80 2.09

''able XXXIIl

— Percentage Dissociation

0°

12°. 5

25°

35°

72.1

71. 1

70.0

68.5

81.7

81.5

128

89.5

89.1

512

95-4

96.2

1024

98.0

99 7

2048

100

99-7

100. 0

100

4096

)0.0

99.96 99

Magnesium Formate
The magnesium was weighed as the pyrophosphate.
Table XXXIV — Conductivity

[)°

25°

35°

37-33

52.53

69.24

83-25

109

132

106

141

172

128

122

164

200

512

123

167

205

1024

133

176

209

2048

138

184

226

4096

138

182

223

388

Winston and Jones

Table XXXV — Temperature Coefficients

0''-12''.5 12". 5-25° 25°-35'

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

1.22

327

1-33

2-53

I .40

2.02

2.48

2. 10

2-51

2.69

2.16

128

2.89

2.74

2.21

512

2.82

2.84

2.24

1024

319

2.52

I. 91

2048

3 31

2.66

2.25

4096

3 32

2-54

2.20

Table XXXVI-

-Percentage Dissociation

12°. 5

38.4

37-9

1024

2048

100

[OO

4096

100

).o

99. c

Magnesium Acetate
The magnesium was determined as in the preceding salt.
Table XXXVII — Conductivity

12'=

54 50

72.50

109

119

146

128

103

139

172

512

153

189

1024

116

158

201

2048

121

164

203

4096

121

165

203

Table X}

(XVII

/ — Temperature Coefficients

-12°

12°. 5

-25°

25°-

-35°

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

1.36

362

I 44

2.64

1.64

2.26

1.63

2.76

2.01

2.24

2.16

2.85

2.69

2.25

128

2.58

2.80

3-28

2.35

512

2.81

2.83

3.61

2.35

1024

2.91

2.89

4.28

2.69

2048

3.00

2.86

3-84

2-33

4096

2.94

2.8'

383

2.32

Conductiviiy, Temperature Coefficients, Etc.
Table XXXIX — Percentage Dissociation

V 0" 12°. 5 25° 35°

389

128

512

1024

2048

4096

100

The next group taken up for study consists of cadmium
chloride, cadmium bromide, cadmium iodide and lead chlor-
ide. Attention should be called to the fact that cadmium
iodide, having no water of crystallization, has just about the
same temperature coefficients of conductivity as cadmium
bromide and cadmium chloride, both of which have water of
crystallization. Apparent increase of percentage dissocia-
tion with rise in temperature is unusual, and is quite notice-
able in the case of cadmium iodide.

Lead chloride has no water of crystallization but, like
cadmium iodide, has high temperature coefficients of conduc-
tivity. There must be some factor operative here affecting
temperature coefficients just as hydration does, but which,
from the nature of the case, cannot be due to hydrates.

Cadmium Chloride

Silver nitrate was used to precipitate the halogen in cad-
mium chloride, bromide and iodide.

Table XL — Conductivity

12°

33 65

46.21

60 15

71.92

118

142

128

122

162

195

512

106

148

197

236

1024

113

159

212

258

2048

121

166

221

269

4096

121

172

232

282

390

Winston and Jones

Table XLI — Temperature Coefficients

0°-12°.5 12°. 5-25° 25°-35°

Cond. Per

Cond.

Per

units cent.

units

cent.

00 2 . 97

I. I 2 2.42

1.96

24 2

1.47 2.42

I 93

97 3

2.26 2.50

2.02

128

77 3

3 15 2.54

2.06

512

38 3

3 . 94 2 . 66

I 99

1024

67 3

4.23 2.65

2.17

2048

60 2

441 2.65

2. 10

4096

14 3

4.74 2.62

2. 12

Table XLII-

—Percentage Dissociation

0°

12°. 5 25°

27.8

26.7 25.9

2 34-2

3 511

2 69.9

9 85.1

1024

4 916

2048

100

2 95-4

4096

100

0 100. 0

100

Cadmium Bromide

Table XLI 1 1— Conductivity

0°

12°. 5 25°

35°

28.63

40.59 53 40

53 36 70

82.06 109

132

128

113 57 151

184

512

lOI

143.25 190

232

024

156.85 208

252

2048

121

170.89 227

275

4096

123

174.05 232

280

Table XLIV-

-Temperature Coefficients

0°-12°.5

12°. 5-25°

25°-

-35

nd. Per

Cond. Per

Cond.

Per

lits cent.

units cent.

units

cent.

96 3-35

1.02 2.51

2.07

24 3

1.37 2.56

94 3

2.18 2.66

128

70 3

3.01 2.65

512

55 3

3.78 2,64

1024

69 3

4.13 2.62

2048

97 3

4,52 2.64

4096

4.6;

, 2.6;

Conductivity, Temperature Coefficients, Etc.

391

Table XLV — Percentage Dissociation

0°

2°. 5

23.1

23.3 23.0

23.0

30.6 30.3

30.2

47 I 47 I

47-3

128

65.2 65.1

65.6

512

82.3 82.1

82.9

1024

90.1 89.8

90.0

2048

98 . 2 97 . 9

98.0

4096

100. 0

100. 0 100. 0

100. 0

Cadmitim Iodide

Table XLV I— Conductivity

0°

12°. 5 25°

35°

20.45

29.76 39.84

48.41

85 48

23 81

lOI

128

36 127

157

512

127

74 172

211

1024

140

03 188

231

2048

109

157

20 209

256

4096

[18.78

170

69 224

271

Table XLV II-

—Temperature Coefficients

0°

-12°. 5

12°. 5-25°

25°-35°

Cond.

Per

Cond. Per

V units

cent.

units cent.

4 0

3 67

0.81 2.72

86 2.16

8 0

3-78

I. 01 2.82

10 2.27

32 I

4.01

1.78 3.01

97 2.42

128 2

3 90

2.72 2.91

00 2 . 36

512 3

3-73

3.62 2.83

90 2 . 26

1024 3

371

3.82 3.71

24 2.25

2048 3

3-54

4.20 3.67

67 2.23

4096 4

3-49

4-34 2.54

63 2 . 06

Table XLV III

— Percentage Dissociati

0°

12°. 5 25°

35°

17.2

17.4 17.7

17.8

20.5

21.0 21

33-2

34 7 36

128

52.8

54-7 56

512

73-3'

74-8 76

1024

81.0

82.6 83

2048

91 7

92 . 1 93

4096

0.0 IC

»o

392

Winston and Jones

Lead Chloride

The lead was precipitated by means of sulphuric acid and
weighed as lead sulphate.

Table XLIX—

Conductivity

0°

12°. 5

25°

35°

104.41

144.76

188.71

224

•76

128

116.27

161.56

211.43

252

•17

512

133 10

186.16

246.31

293

1024

136.89

191.98

253 96

306

■43

2048

138.88

195.16

258 -49

312

4096

144.70

204 . 36

270.26

327

.80

Table L—

Temperature Coefficients

0°-12°.5

12°. 5

-25°

25°-

-35°

Cond. Per

Cond.

Per

Cond.

Per

V units cent.

units

cent.

units

cent.

64 3

23 3.09

3 52

2-43

3 61

I. 91

128 3

63 3 12

3 99

2.47

4.07

I 93

512 4

25 3 19

4.81

2.58

4.67

1.90

1024 4

41 3.22

4.96

2.58

525

2.07

2048 4

70 3 . 24

507

2.60

5.36

2.07

4096 4

77 3-30

527

2.58

5-75

2.13

Table LI-

■Percentage

Dissociation

0°

12°. 5

25°

35°

72.2

69.8

68.6

128

80.4

78.2

76.9

512

92 .0

91. I

89.4

1024

94 6

94 0

93-5

2048

96.0

95 6

95-2

4096

100. 0

100

100. 0

The aluminium and chromium compounds will be taken up
next for discussion. In these compounds we should expect
to find strong resemblances. These are very apparent.
Chromium and aluminium compounds, with respect to their
conductivities, are in a class by themselves. Their very
large conductivities and their exceedingly large temperature
coefficients must attract attention. Their very large conduc-
tivities are due mainly to the great number of ions into which
they are capable of ionizing and to hydrolysis. Judging from

Conductivity, Temperature Coefficients, Etc.

393

their water of crystallization and from freezing point lowerings/
they must be hydrated to an enormous extent. Their large tem-
perature coefficients of conductivity would also indicate this to be
the fact. The change in conductivity, both with rise in tempera-
ture and with dilution, is much more gradual in the case of
the aluminium salts than with those of chromium. The ex-
tremely small percentage dissociation in concentrated solu-
tions, in the case of chromium sulphate and aluminium sul-
phate, is worthy of notice. This is probably connected with
the fact that sulphates, especially in concentrated solution,
undergo marked polymerization.

Aluminium Chloride

The aluminium was determined by precipitating the hy-
droxide and weighing as the oxide AljOg. This was done
also in the case of aluminium nitrate and aluminium sulphate.

Table LI I — Conductivity

0°

12°. 5

25°

35°

105 90

147.40

193-51

232.54

120

168.23

220

266.58

142

200 . 06

265

322.18

128

162

231.08

308

377.28

512

176

252.75

341

42 I . 06

1024

184

266.73

360

446 . 95

2048

193

279.49

381

472.46

4096

199

290.06

398

499.92

Table LIII-

-Temperature Coefficients

o°-

12°. 5

-25°

25°-3S°

Cond.

Per

Cond.

Per

ond. Per

units

cent.

units

cent.

units cent.

3 14

3 69

2.50

90 2 . 02

4.21

2.50

57 2.07

5-21

2.60

71 2.15

128

6.22

2.69

85 2.22

512

7.08

2.80

98 2 . 34

1024

7-51

2.82

64 2.40

2048

8.16

2.92

10 2.39

4096

8.70

3.0c

)

II 2.54

1 Jones and Getman: This Journal. 31, 303 (1904). Publication No. 60, Car-
negie Institution of Washington.

394

Winston and Jones

Table LIV — Percentage Dissociation

0°

12°

53-2

50.8

48.5

46 -5

60.4

715

128

81.7

512

88.8

1024

92.8

2048

97 2

4096

100. 0

100

Aluminium Nitrate
Table L V — Conductivity

12°

102.82

139.22

180.52

216.54

115

158.84

206

248

136

188.54

247

299

128

156

217.14

287

349

512

166

234.81

313

384

1024

173

247.08

332

410

2048

179

255.68

345

428

4096

187

272.12

372

462

Table LVI-

-Temperature Coefficients

-12°. 5

12°. 5-

25°

25°-35°

Cond.

Per

Cond.

Per

Cond. Per

V units

cent.

units

cent.

units cent.

4 2

2.83

330

2.37

3.60 1.99

8 3

2.42

19 2.03

32 4

2.51

23 2. II

128 4

2.58

25 2.18

512 5

2.67

17 2.28

1024 5

2.77

86 2.36

2048 6

2.83

37 2.39

4096 6

4-53

8.19

2-95

932 2.45

Table LV 1 1-

— Pcrcewtoge

Dissociation

0°

12°. 5

25°

35°

54-7

512

48.5

46.8

55.6

66.6

128

77.1

512

84.1

1024

893

2048

92.9

4096

)0.0

Conductivity, Temperature Coefficients, Etc.

395

Aluminium Sulphate
Table LVIII — Conductivity

51 90

71.81

107

65.21

89.81

114

132

89.50

123.63

158

183

128

121.87

169.38

219

266

512

164.08

230.86

301

358

1024

191 95

271.31

359

433

2048

222.31

317.20

425

518

4096

262.35

378.23

514

634

Table LIX — Temperature Coefficients

0°-12°.5 12°. 5-25° 25°-35<=

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

3 06

1.65

2.30

1-53

1.66

2.19

1-57

2.23

1. 61

128

2.34

2.16

512

2.43

I 79

1024

2-59

2.07

2048

2.72

2.19

4096

2.87

2.35

Table LX-

-Percentage

Dissociation

0°

12°. 5

19.8

128

1024

2048

4096

100

]

Chromium Chloride

The chromium was weighed as the oxide CfjOg in the case
of both chromium chloride and chromium sulphate.

396

Winston and Jones

Table LXI — Conductivity

86.30

116

153

199

104-53

138 83

184

243

130.03

182.75

245

319

128

162.34

231.28

313

393

512

188.46

272.50

372

465

1024

200.21

294 -55

403

504

2048

214.48

316.60

434

543

4096

229.73

341 14

467

580

Table LXII-

-Temperature Coefficients

0°

-12°. 5

12°. 5

-25°

25 "-as"

Cond.

Per

Cond.

Per

Cond. Per

V units

cent.

units

cent.

units. cent.

•4 2.45

2.84

2.91

2.49

58 2.99

8 2

2.62

94 3

32 4

2.73

42 3

128 5

2.84

02 2

512 6

2-93

28 3

1024 7

2.96

07 2

2048 8

2.98

87 2

4096 8 91

2.97

26 2.41

Table LXIII-

—Percentage Dissociatio

12". 5

25 *»

35°

37.6

32.8

34-3

45-5

42.0

56.6

550

128

70.7

67.9

512

82.1

80.2

1024

87.2

86.9

2048

93 3

93 6

4096

100

100. 0

Chromium Sulphate
Table LXIV — Conductivity

12°

58.14

78.48

99.64

116. 41

103

130

151

1 20

158

197

230

128

169

225

283

338

512

215

292

376

472

1024

240

329

459

561

2048

293

405

534

708

4096

315

445

598

808

Conductivity, Temperature Coefficients, Etc.
Table LXV — Temperature Coefficients

0°-12''.5 12°. 5-25° 25°-35°

397

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

2.80

I .69

2.15

1.68

1.69

2.65

2 . 12

2.05

2. 10

I. 61

2.54

2.46

1-55

3 30

1.67

128

2.67

4.64

2.06

5-51

1.94

512

2.87

4.69

1.60

9 59

2.55

1024

2.98

10.39

315

10. 10

2 .22

2048

3 06

10.31

2.54

17.36

325

4096

329

12.26

2.75

20.98

351

Table LXVI-

-Percentage

' Dissociation

0°

12°. 5

18.4

17.6

24.7

38.2

53-6

68.3

1024

76.2

2048

93 0

4096

100. 0

100

In the next group will be considered manganous sulphate,
silver nitrate, copper sulphate and cobalt bromide. Man-
ganous sulphate calls for no comment. The data obtained
for silver nitrate are remarkably similar to those obtained for
ammonium nitrate. It apparently behaves as any other
ordinary, unhydrated, binary compound. It differs from
ammonium nitrate in that its percentage dissociation, appar-
ently decreasing with rise in temperature from 0° to 25°, in-
creases somewhat at 35°.

The data for copper sulphate resemble strikingly those ob-
tained for manganous sulphate, cadmium bromide and cad-
mium iodide. At ordinary temperatures manganous sulphate
and copper sulphate have the same amount of water of crys-
tallization. That their temperature coefficients should be
approximately the same is not surprising; but that the tem-
perature coefficients of cadmium chloride and cadmium bro-
mide, crystallizing with less water, and cadmium iodide,
crystallizing with no water, should be the same is surprising.

398

Winston and Jones

The temperature coefficients of conductivity of cobalt
bromide indicate much hydration, as would be expected from
its water of crystallization.

Manganous Sulphate
The manganese was weighed as the pyrophosphate.
Table LXVII — Conductivity

109

129

128

III

147

176

512

138

184

222

1024

[07

152

202

245

2048

[16

165

221

268

4096

[24

177 56

238

289

Table LXVIII-

—Temperature Coeffii

zients

0°

-12°. 5

12°. 5-

25°

-35°

Cond.

Per

Cond.

Per

Cond.

Per

V units

cent.

units

cent.

units

cent.

4 I . 16

3 II

1.23

2.38

I 19

1.77

8 1.38

2.40

1-43

I 79

32 I 91

2.47

2.05

1.88

128 2.58

2.54

2.89

1.96

512 3.26

2.64

3-8i

2.06

1024 3.62

2.66

4.28

2. II

2048 3 93

2.71

4.70

2. 12

4096 4.25

3 42

485

2-73

512

215

Table LXIX-

-Percentag

? Dissociation

0°

12°. 5

25°

35"

29.9

29.2

28.2

7-3

34-6

47 0

128

62.9

512

78.1

1024

85.8

2048

93 I

4096

)0.0

100

Conductivity, Temperature Coefficients, Etc.

399

Silver Nitrate
The silver was weighed as the chloride.
Table LXX — Conductivity

0°

°.5

25°

35 »

5143

70.55

91.63

109 95

99.80

120

III .20

133

128

119. 14

142

512

125.23

148

2048

126.81

151

4096

129.68

153

Table LXXI — Temperature Coefficients

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

units

cent.

I 53

2.98

I .69

2.40

I 83

2.00

I 65

2.06

1.88

2.07

2.19

128

2.02

2.25

2-35

512

2.06

2.42

2-35

2048

2-15

2.41

2.44

4096

2.24

2-45

2.36

Table LXXIl — Percentage Dissociation

128

512

2048

4096

71
77
86
92
95
97
100

70
76
85
91
96

97
100

35
71
78
86
93
97
98
100

Cobalt Bromide

This salt was precipitated by means of silver nitrate, and
the bromine determined from the weight of silver bromide
obtained.

400

Winston and Jones

Table LXXIII — Conductivity

120

87.82

120

155 60

196.30

131

171

204

105

147

193

233

128

115

162

214

259

512

119

169

224

273

1024

120

173

231

281

2048

124

174

234

282

4096

125

177

236

289

Table LXXIV — Temperature Coefficients

0°-12°.5 12°. 5-25° 25°-35=

Cond.

Per

Cond.

Per

Cond.

Per

units

cent.

units

cent.

imits

cent.

2-59

2-95

2.83

2-35

4.07

2.62

2.44

1.94

2.50

2.07

128

2.56

215

512

2.60

2.18

1024

2.68

2.14

2048

2.73

2.07

4096

2.65

2.22

Table LXXV — Percentage Dissociation

0 12 =

70.0 67.6

65.7

67.8

7 73

I 82

128

3 92

512

2 95

1024

3 97

2048

8 98

4096

100. 0 100. 0

100. 0

Copper S

ulphate

The sulphuric acid in this salt was precipitated and weighed
as barium sulphate.

Conductivity, . Temperature Coefficients, Etc.

401

Table LXXVI — Conductivity

128

512

1024

2048
4096

30
42
57
76

97
105
113
119

59
80
108
138
150
161
171

55
77
105
143
184
202
217
231

35°

65 15
91 . 16
124.94
1 70 . 60
221 .08
245 05
264.44
281.42

Table LXXVI I — Temperature Coefficients

0°-12°.5 12". 5-25° 25''-35°

128

512

1024

2048

4096

Cond.
units

0.96

1.36

1.86

2-54
3.28
3.60
387
415

Per

cent.

.22

•25
■30

•35
.40

.48

Cond.
units

1 .04
1.44

2 .01
2 .76
3.68
4.14

4-47
4.82

Per
cent.

2.47

2-43
2.50

2-54
2.65
2.74
2.76
2.82

Cond.
units

1 .00

I 38

I 93
2.74
3 61
425
4.67
5.02

Per

cent.

1.82
1.79

I 83
I. 91

1-95
2. 10

2.15
2.17

Table LXXVI II — Percentage Dissociation

128

512

1024

2048

4096

25.2

35-5
48.0

64 -5
82.1
88.8

95 I

100. o

12°. 5
24.6

34-7
47.1
63.6
81.2
88.2
94.6
100. o

25°
23.8

33-4
45-7
61 .9
80.0
87.6
94.1
100. o

35°
23.2
32.4
44-4
60.6
78.6
87.1
94.0
100. o

The conductivity values obtained for uranyl sulphate
and uranyl acetate do not agree satisfactorily with those ob-
tained by West.^ His solutions were evidently standardized
on a different basis. It should be noticed that the tempera-
ture coefficients, in conductivity units, of uranyl sulphate de-
crease with rise in temperature through V = 512. After this
dilution they increase, as in the case of the other uranyl salts.
The percentage dissociation of uranyl acetate apparently

» This Journal, 44, 537 (1910).

402

Winston and Jones

increases with rise in temperature through V = 128. The
more dilute solutions show a decrease with rise in tempera-
ture. This may be seen in the curve for uranyl acetate which
follows.

Uranyl Chloride

The uranium in uranyl chloride, nitrate, sulphate and acetate
was precipitated by means of ammonium hydroxide and
weighed as the oxide UgOg.

Table LXXIX — Conductivity

= .5

25°

35°

101.45

139.09

180.45

214.70

157

.64

206.01

246

133

186

•56

246. 12

297

128

148

209

279.00

339

512

155

220

296.56

360

1024

161

231

311.92

383

2048

168

242

328.24

405

4096

174.98

254.22

348.16

433

Table LXXX-

—Temperature Coefficients

(

)°-12°.5

12°. 5

-25°

25 °-35 °

Cond.

Per

Cond.

Per

Cond. Per

V units

cent.

units

cent.

units cent.

4 301

2.97

331

2.38

3 43 I 90

8 3

2.46

05 1-97

32 4

2-55

17 2.10

128 4

2.64

04 2.17

512 5

2-75

39 2.i6

1024 5

2.78

20 2.31

2048 5

2.82

77 2.37

4096 6 . 34

2.96

55 2.46

Table LXXXI

— Percentage Dissociation

12°. 5

25°

35°

58.0

54-7

51-8

49 5

62.0

59-2

73 4

70.7

128

82.5

80.2

512

86.8

85.2

1024

91 .0

89.6

2048

95 5

94-3

4096

0.0

100. 0

100

Conductivity, Temperature Coefficients, Etc.

403

Uranyl Nitrate
Table LXXXI I— Conductivity

12°.

74 91

102.01

132.91

158.84

83-

114

150

181

97-

136

180

219

128

153

207

254

512

116

166

224

277

1024

123

177

241

298

2048

128

187

255

317

4096

136

200

274

343 09

Table LXX XIII— Temperature Coefficients

°-12°.5

12°. 5

-25°

25°-35°

Cond.

Per

Cond.

Per

Cond. Per

V units

cent.

units

cent.

units cent.

4 2

2.90

2.47

2.42

2.59 1-95

8 2

2.68

06 2 . 03

32 3

2.60

87 2.14

128 3

2.67

63 2.23

512 4

2.80

24 2.33

1024 4

2.87

72 2.37

2048 4

2.92

21 2.43

4096 5

2.97

6 . 86 2 . 50

able LXXXI V — Percentage Dissociation

0°

12°. 5

25°

35°

54-8

51.0

48.4

46 -3

61.0

71. 1

128

80.5

512

85.0

1024

90.0

2048

94 2

4096

100. 0

100

100. 0

Uranyl Sulphate

Table LXX XV— Conductivity

0°

12°. 5

25°

35°

78.13

99 77

120.82

136.43

100

129

156

176

128

166

203

229

512

157

•54

207

257

295

1024

175

.68

235

296

343

2048

191

.68

260

■77

332

391

4096

203

■3

285

373

446

404

Winston and Jones

Table LXXXVI — Temperature Coefficients

0°-12°.5 12°. 5-25" 25°-3S°

Cond.

Per

Cond.

Per

■"

Cond.

Per

units

cent.

units

cent.

units

cent.

2.22

1.68

1.29

2.30

1.68

128

2.37

I 74

.512

2.56

1. 91

1024

2.72

2.10

2048

2.89

2.20

4096

3.22

7.09

2.49

I 95

Table LXX XV 11— Percentage Dissociation

0°

12°. 5

25°

38.4

32.3

1024

76.

2048

87-

4096

100. 0

100.

Uranyl Acetate

Table LXXXV in—Conductivity

0°

12°. 5

25°

35°

30.59

42 -75

56.53

128

108

512

IIO

129

1024

I20

141

2048

103

131

154

4096

83 -75

113. 81

145.10

170

ible LXXXIX

— Temperature Coefficients

-12°. 5

12°. 5-

-25°

25°-

-35

nd.

Per

Cond.

Per

Cond.

Per

its

cent.

units

cent.

units

cent.

3.12

I.IO

2-57

2.05

3 10

2.49

128

2.97

2-34

512

2.83

2.26

1024

2.81

2.16

73'

2048

2.80

2.17

4096

)

2.2c

)

Conductivity, Temperature Coefficients, Etc. 405

Table XC — Percentage Dissociation

12°. 5

25°

35°

36 5

37.6

39 0

40.0

48.4

48.8

50.8

128

62.1

63.0

63-7

512

75-6

76.1

75-7

1024

833

83.0

82.8

2048

91.1

90.8

90.6

4096

100

100. 0

PLATES

So far little or nothing has been said in regard to the per-
centage dissociation of the salts studied. Attention will be
called to these by means of curves. The curves of ten of the
thirty salts showed the percentage dissociation to be almost
a linear function of rise in temperature. Plotting percentage
dissociation as ordinates against rise in temperature as ab-
scissae, for each dilution, in ten cases out of the thirty, cur\-es.
were obtained resembling the one for aluminium sulphate,
Fig. I. The other 20 salts all showed variations in the maxima
or minima. Some of these are very slight. Diagrams of the
most striking variations follow. The salts giving curves
showing the percentage dissociation to be a linear function
of rise in temperature were acid ammonium sulphate, alu-
minium nitrate, aluminium chloride, aluminium sulphate,
uranyl chloride, uranyl sulphate, uranyl nitrate, chromium
sulphate, cadmium chloride and manganous sulphate. The
others showed more or less variations, the most striking being
here represented.

Fig. II is very interesting, showing in the case of cadmium
iodide the increase in percentage dissociation with rise of
temperature from F = 8 to 1/ = 2048.

From the curve. Fig. Ill, it is easily seen that the percentage
dissociation of chromium chloride increases decidedly with rise
in temperature between 25° and 35°. The increase becomes
less and less as the dilution increases.

Uranyl acetate shows an increase in percentage dissociation
in the more concentrated solutions, but at greater dilutions
gives a falling curve (Fig. IV).

4o6

Winston and Jones

■ISO

3048

13-5

Temperature
Fig- I — Aluminium Sulphat

Conductivity, Temperature Coefficients, Etc. 40?

2048

^ ,o..

I2S

T "*'

12.5 25

Temperature
Fig. II— Cadmium Iodide

'4o8

Winslon and Jones

2048

4, 12^

12-5 25

Temperature
Fig. Ill — Chromiiun Chloride

Conductivity, Temperature Coefficients, Etc. 409

•^ 512

12.5

Temperature
Fig. IV— Uranyl Acetate

4IO

Winston and Jones

12.5 25

Temperature

Fig. V — Silver Nitrate

12.5 25

Temperature
Fig. VI — Magnesium Nitrate

Conductivity, Temperature Coefficients, Etc.

411

Silver nitrate shows a decided increase in percentage disso-
ciation with rise in temperature from F = 4 to V = 2048.
See Fig. V.

The curves representing magnesium nitrate and magnesium
bromide (Figs. VI and VII) show a remarkable resemblance.
The maxima at higher dilutions are pronounced.

•I So :^

•2

13.5 25

Temperature
Pig. VII — Magnesium Bromide

An examination of the curves raised the question, what pro-
duces this variation? The apparent increase in percentage
dissociation with rise in temperature would, naturally, be
thought to be due to hydration. When there is little or no
hydration the question becomes more difficult to answer. If,
however, the ions are assumed to be complex, rise in tem-
perature would bring about greater dissociation, and the effect

412 Winston and Jones

would be just the same as if hydrates had been present. It is
difficult to differentiate the two factors. That hydrates exist is
not doubted. The complexity of the ions is not so well estab-
lished, so that we shall present arguments only for the latter.
If the change were a gradual increase, it might be attributed
easily to hydration, but a change from an increase to a de-
crease in dissociation could not be accounted for in this way;
whereas complex ions once dissociated might reach a state
where recombination would take place. Moreover, the amount
of hydration has been found to depend on the amount of water
of crystallization. In several of the preceding salts, notably
in the case of cadmium compounds, cadmium iodide, which
has no water of crystallization, is found to have temperature
coefficients of conductivity equal in magnitude to those of
cadmium chloride, cadmium bromide and copper sulphate,
all of which have water of crystallization. Lead chloride,
also, which has no water of crystallization, has temperature
coefficients which compare well with those of substances
that are much hydrated, i. e., copper sulphate and cobalt
bromide. This would indicate that there must be some other
factor present producing the same effect as hydration.

SUMMARY

1. In the main the results obtained in the case of the thirty
salts studied tend to confirm the earlier results.

2. Without exception conductivity increases with rise in
temperature and with dilution.

3. The temperature coefficients of conductivity expressed
in conductivity units, with two exceptions, increase with rise
in temperature, while the temperature coefficients expressed
in percentage decrease.

4. Salts greatly hydrated have large temperature coeffi-
cients. The amount of hydration, judged by the tempera-
ture coefficients, seems closely related to the water of crys-
tallization.

5. The apparent exceptions to the results earlier obtained,
viz., an increase in percentage dissociation with rise in tem-
perature and a large temperature coefficient when there is no

Reviews 413

reason to expect large hydration, point, in the opinion of the
authors, strongly to the view advanced above, that inductive
action takes place through the solvent between charged ions
and neutral molecules, and that this gives rise to complex
molecules and ions in solution.

After sufficient work has been done in this field, it is hoped
to bring together all of the condvictivity and dissociation data
obtained in this laboratory and to publish them as a mono-
graph. However, before this is done it is intended to repeat
the work with every substance, where the repetition is not
already complete, starting with new material, repurifying,
restandardizing and remeasuring the conductivity.

Work will be continued in this laboratory along the lines
indicated above, probably for at least the next ten years,
and there will be five investigators working here in this field
during the next academic year. *

Johns Hopkins University
May, 1911

REVIEWS

Ueber KaTaIvVSE. Von Wii^helm Ostwai^d. Rede gehalten am 12
December, 1909, bei Empfang des Nobelpreises ftir Chemie. Zweite
Auflage. Leipzig: Akademische Verlagsgesellschaft m. b. H. 1911.
Price, M. 1.50.

It is not surprising that a second edition of this interesting
lecture has been called for. The circumstances, the subject,
the lecturer form a strong combination. Every chemist
will find it profitable to read the lecture, and not the least
interesting part is that which deals with the author himself.
He does not apologize for withdrawing from his activity in
experimental work but offers an explanation. Lothar Meyer
had warned him against too close application and possible
physical exhaustion, but Ostwald took the ground that it
is much more important to keep at the work, even to the
point of complete exhaustion, than to spare the individual
in the hope that he may be able to contribute more by living
longer.

The author's references to Berzelius and his discussion
of catalysis show once more what is well known to those
who have carefully studied the history of any branch of science,

414 Reviews

that the great original is often held responsible for ideas which
he never entertained.

The phenomena of catalysis are among the most interesting
and important of all chemical phenomena, and while this
lecture furnishes anything but a systematic treatise on the
subject its reading will no doubt increase the interest of the
reader. i. r.

OSNOVI Physicheskoi Chemie (The Elements of Physical Chemistry).
By Prof. Harry C. Jones. Translated by Prof. E. V. Biron, I. I.
Zhukoff and A. V. Sopozhnikoff. St. Petersburg : A. S. Suvorin.
191 1. Price, 5 Rub. 75 kop.

The importance of Russian as a medium of scientific com-
munications is now well established, and chemistry in par-
ticular is represented in that language by an extensive literature
of original work and translations. The above translation
of the fourth English edition of a well-known and popular
work appears to be careful and well written. The printing
and illustrations are clear and the paper is good. b. b. t.

AivCOHoivic Fermentation. By Arthur Harden, Ph.D., D.Sc,
F.R.S., Head of the Biochemical Department, Lister Institute,
Chelsea. Monographs on Biochemistry, edited by R. H. Aders
PiviMMER and F. G. Hopkins. London, New York, Bombay and
Calcutta : Longmans, Green & Co. igir. pp. ix + 128. Price, $1.25.

It is not too much to say that the subject of alcoholic fer-
mentation is one of the most interesting and important in the
whole field of biochemistry, and since the discovery of zymase
by Buchner in i8g6, our knowledge of the process has made
rapid progress. In the historical introduction the author
reviews the gradual development of our knowledge of alcoholic
fermentation in a most interesting and stimulating manner.
This is followed by a chapter on zymase and its properties, in
which the masterly researches of Buchner and others are
brought to the attention of the reader. Other valuable and
suggestive chapters in the monograph are those on thef unction
of phosphates in alcoholic fermentation, the co-enzyme of yeast
juice, and the by-products of alcoholic fermentation. The value
of the monograph is further enhanced by a comprehensive bibli-
ography containing two hundred and seventy- two (272) refer-
ences to the extensive literature of this subject. In the opinion
of the writer, the value of this monograph lies not only in the
fact that it brings before the reader an accurate and com-
prehensive picture of our present knowledge of alcoholic
fermentation, but also in that it will stimulate further inves-
tigation in this promising field of biochemical research.

Joseph H. Kastle.

Reviews 415

The Fats. By J. B. Leathes, M.A., M.B., F.R.C.S., Professor of
Pathological Chemistry in the University of Toronto. Monographs
on Biochemistry, edited by R. H. Aders Pummer and F. G. Hopkins.
London, New York, Bombay and Calcutta : Longmans, Green &
Compan}^ 1910. pp. ix -|- 138.

According to the author the object of this monograph
is twofold : first, to acquaint the physiologist with the recent
advances in the chemistry of fats; and, second, to enable the
chemist, familiar with the chemistry of fats, to look at the
subject from a standpoint of the biologist. In its present
shape the book is such as to indicate that the author has
abundantly attained his end. The first three chapters are
devoted to a brief description of the fats and fatty acids,
to the extraction and methods of estimating fats in animal
tissues, to the physical properties of fats, and the methods
employed in their separation and identification. While
not easy reading, these three chapters contain a wealth of
physical and chemical data which one would scarcely expect
to find in a volume of only 138 pages, and which will prove
immensely valuable and convenient to those interested in
this line of work. The last chapter of the monograph, en»
titled "The Physiology of Fats," is intensely interesting and
is written in admirable style. Section C of this chapter,
especially, gives us much to think about, and the subjects
therein considered are such as to raise the fats to a level of
biological significance hitherto attained only by the proteins.
Hitherto we have been accustomed to look upon the fats
solely as convenient and efficient forms of reserve material.
When we recall, however, that 50 milligrams of cerebronic
acid, which contains no nitrogen, will neutralize from six
to nine hundred lethal doses of tetanus toxin, we begin to
look at the fats in an altogether new and significant light.

Joseph H. Kastle.

An Introduction to Bacteriologicai. and Enzyme Chemistry.
By Gii,bert J. FowtER, D.Sc, F. I. C, Lecturer in Bacteriological
Chemistry in the Public Health Department, Victoria University of
Manchester, Examiner in Biological Chemistry to the Institute of
Chemistry of Great Britain and Ireland. New York : Longmans,
Green & Co. London : Edward Arnold, pp. vi + 328. Price, $2.10.

The processes ordinarily included under the head of bac-
teriological and enzyme chemistry are assuming a continually
increasing importance in their economic and scientific aspects,
and advances in this field of work rapidly make their way into
our domestic, civic and industrial life. From the time of
Pasteur's epoch-making researches upon beer, wine and vinegar,
the importance of the part played by the organized and un-

41 6 Reviews

organized ferments has come more and more to be recognized
and utilized in the arts and manufactures. As the author points
out, however, in his preface, we have no elementary book serv-
ing as a reasonably comprehensive and accurate guide to the
vast literature of this subject. In the writing of this very readable
book the author has conferred a boon upon English-speaking
students seeking a knowledge of the chemistry of bacterio-
logical and zymotic processes; and, in my opinion, the chemist,
the bacteriologist, the student of agriculture, and the sanitary
engineer will all find in it something of interest and of value.

Joseph H. Kastle.

Die Schwelteere, ihre Gewinnung und Verarbeitung. Von
Dr. W. Scheithauer, Direktor. Mit 70 Figuren im Text. (Chemische
Technologic in Einzeldarstellungen, herausgegeben von Ferdinand
Fischer.) Leipzig : Verlag von Otto Spamer. 1911. S. viii + 192.
Preis geheftet, M. 8.75; gebunden, M. 10.

This deals very fully with the distil ation of brown coal
and bituminous shales, more particularly in Thuringia, Messel
and Scotland. Not only is the apparatus described but
also the method of working yields and treatment of by-prod-
ucts. A chapter follows upon candle-making from the
wax produced and two chapters upon the chemical composition
of the distillates and the laboratory tests. The book opens
with a short historical sketch and closes with some statistics.

The volume is one of Fischer's noted Technological Series
and may be cordially recommended to all interested.

A. H. Gill.

Vol. XLVI November, 191 i No. 5

AMERICAN

CHEMICALJOURNAL

ON CHLORIMIDOQUINONES

By Lemuel Charles Raiford •

In 1890 Hantzsch and Werner' formulated the theory
that the isomerism of nitrogen compounds of the same com-
position and structure which contain a doubly bound nitro-
gen atom is of the same nature as the stereoisomeric ethyl-
ene derivatives, like maleic and fumaric acids. In accord-
ance with the theory we should have the two configurations

X— C— Y X— C— Y

li and II

Z— N N— Z

The objections of Victor Meyer^ that the differences be-
tween the isomeric benziloximes amd dioximes might de-
pend on peculiarities of the hydroxylamine molecule were
met by Hantzsch^ by the preparation of stereoisomeric phenyl-
hydrazones of anisylphenylketone ; but he was unable to ex-
tend* the proof beyond these two closely related groups of
compounds, the oximes and the hydrazones, his efforts to
prepare stereoisomers in which Z in the above formula is repre-
sented by such simple radicals as CH3, C^H^ etc., being un-

1 Ber. d. chem. Ges., 23, 11 and 1243 (1890).
2/6t(/., 1«, 503 (1883); 21, 784. 3510 (1888).
a/6id.,a4, 3525 (1891).
* Hantzsch: "Stereochemie," p. 141.

41 8 Raijord

successful. The first representatives of such stereoisomers,
other than oximes and hydrazones, were prepared in 1903
by Stieglitz znd Earle^ in the form of a pair of stereoisomeric
chlorimido esters in which Z in the configuration given is repre-
sented by a single chlorine atom. Stieglitz and Hale^ pre-
pared a second pair of stereoisomers in 1904, and Hilpert,^
working with Stieglitz in 1907, added five pairs of such com-
pounds to the list. The type is now characterized on a broad
and firm basis.*

In order to investigate the question of the occurrence of
stereoisomeric chlorimides in classes of compounds other
than the acid esters, Professor Stieglitz suggested to me to
determine whether stereoisomerism can be observed in the case
of chlorimidoquinones, comparable with that of the quinone
oximes, discussed and investigated by Kehrmann.^ Although
six chlorimidoquinones whose structure would admit of the
occurrence of such stereoisomerism were prepared and studied,
and though the structure was varied to cover all the possible
types, instances of stereoisomerism were not observ^ed, not
even in the case of the chlorimide of 2-chlor-5-methylqui-
none, whose oxime, according to Kehrmann,® has the most
favorable structure for the occurrence and persistence of this
form of stereoisomerism. Stieglitz and Hale found that the
labile form of stereoisomeric chlorimidonitrobenzoates is very
readily converted into the stable form by the action of chlorine
and Hilpert confirmed this observation for the chlorimido
esters which he studied. Whether the failure to obtain stereo-
isomeric chlorimidoquinones is due to the fact that chlorine
was present in the solution in which the chlorimides were
prepared, and that the method of preparation^ necessarily
involved a much greater time of contact of the reagents with
the material than in the case of the preparation of the chlor-

1 This Journal. 30, 399 (1903); 40, 37 (1908).

2 Unpublished reports.

3 This Journal. 40, 150 (1908).

■• Stereoisomeric chlorimido ketones have recently been prepared by Stieglitz
and Peterson.

s Ann.. Chem. (Liebig). 279, 27 (1894); 303, 1 (1898).

* Loc. cit.

^ The oxidation of />-aminophenols with hypochlorous acid ia acid solution.

On Chlorimidoquinones 419

imido esters, or whether one of the two possible forms is so
much the more stable that it is obtained exclusively, it is
impossible, of course, to say. It is quite a common expe-
rience that cases of stereoisomerism theoretically possible
are not readily realizable by our preparation methods.

In the study of the chlorimidoquinones for this work a num-
ber of interesting new observations were made on compounds
of the phenol and quinone series, and, in particular, some in-
correct statements and uncertain points given in the litera-
ture on the preparation and structure of such compounds
were noted and corrected. All our conclusions were care-
fully verified by experimental work. In the following paper
the most interesting of the observations made are reported.

/. 4-Chlonimdo-2-bromqmnone

The starting point in the preparation of this compound
was 4-nitrophenol, which was first brominated according ta
the method of Brunk^ and Komer.^ The resulting brom
compound was purified by crystallization of its barium salt,
from which, by subsequent treatment with hydrochloric
acid, the free phenol was obtained. The latter was finally
crystallized from water, from which it separated out in color-
less needles that melt at 112°.^ It was further identified by
the preparation of its reduction products, the corresponding
amine and its hydrochloride.

Hydrochloride of 2-Brom-4-aminophenol. — 2-Brom-4-nitro-
phenol was dissolved in the smallest possible quantity of hot
alcohol (i gram to i cc.) and to the hot solution was
added one-fourth more than the calculated amount of stan-
nous chloride, dissolved in concentrated hydrochloric acid
(i gram to i cc). During this time the flask was shaken
and the mixture kept hot in order to prevent the immediate
separation of crystals. Next, one volume of concentrated
hydrochloric acid was added and the solution set aside to
cool. Crystals of the amino hydrochloride soon separated

> Z, Chem., 1867, 204.

^ Ibid., 1868, 323.

3 The statement of Brunk that 2-brom-4-nitrophenol melts at 102° is probably
a misprint. Meldola and Streatfield (J. Chem. Soc, 7S, 681 (1898)) prepared this
compound by nitrating o-bromphenol, and they found 112" as the melting point.

420 Raiford

out. These were filtered off and recrystallized as follows:
The crude material was dissolved in warm water and the solu-
tion filtered through paper, after which one volume of con-
centrated hydrochloric acid was added to the filtrate. Upon
standing, slightly yellowish crystals of the amino hydrochlor-
ide, free from tin compounds, were deposited.

The hydrochloride thus obtained is readily soluble in water
containing a trace of acid, in alcohol, and in a solution of
sodium hydroxide. When heated to 225° the compound
begins to blacken, but does not melt. A sample dried in
vacxw over potassium hydroxide for 72 hours was analyzed
for halogen and gave the following results:

0.2913 gram substance gave 0.4311 gram AgHal.

CeHjONClBr

Found

51-42

51 56

Halogen

2-Brom-4-aminophenol. — The free amine was easily ob-
tained by treatment of a water solution of the hydrochloride
described above with ammonium carbonate solution. The
precipitated amine was filtered off at once, and dried on a clay
plate. In this condition it melted at 164° and was very
nearly pure. The compound is soluble in alcohol, chloro-
form, ether, and a solution of sodium hydroxide, but much
less soluble in benzene. It is best crystallized from the last-
named liquid, from which it separates in needles having a
faintly brownish color and melting at 165°. Analysis for bro-
mine gave the following figures :

o. 2450 gram substance gave o. 2474 gram AgBr.

CoHeONBr

Found

42.54

42.97

4-Chlor'iinido-2-bromquinone, O : CgHgBr : NCI. — A portion
of the hydrochloride of 2-brom-4-aminophenol, weighing 5
grams, was dissolved in 75 cc. of water to which a trace of
hydrochloric acid had been added, and the resulting solution
cooled to about 0°. This liquid was then allowed to flow slowly
from a tap funnel into acidulated (hydrochloric acid) solu-

On Chlorimidoquinones 421

tion of sodium hypochlorite^ which had been' cooled to 0°,
and in which pieces of ice were floating. The flask containing
the hypochlorite was kept surrounded by a mixture of ice
and water, and was shaken continuously while the amino
hydrochloride was being added. A yellow solid, having the
characteristic odor of a chlorimidoquinone, was promptly
precipitated, and after standing for a few minutes was fil-
tered off and washed several times with cold water.

When dried on a clay plate the chlorimide began to darken
after 24 hours, a change that goes on more rapidly as the
temperature rises. The crude product did not have a char-
acteristic melting point, but decomposed suddenly, after soft-
ening, when heated above 60°. The compound is soluble
in alcohol, ether and chloroform, but less readily so in ligroin.
It is best crystallized from the low-boiling (40°-6o°) fraction
of the latter liquid, because the heat required to saturate
the higher boiling fractions to a sufficient degree to give a
satisfactory yield of crystals will decompose the chlorimide.
Attempts to fractionate- the compound gave negative re-
sults. The purest product obtainable decomposed, without
melting, when heated to 60°, as already indicated.

For analysis, the method employed by Stieglitz and Earle^
in the determination of halogen in their chlorimido esters,
consisting in the liberation of iodine from hydrogen iodide,
was first tried; but concordant results could not be obtained,
because not only the chlorimide group but the quinone ring,
too, oxidizes the hydrogen iodide. Halogen was determined,
in this and all other compounds described in this paper, by
the Carius method.

0.3460 gram substance gave 0.5233 gram AgHal.

Calculated for

CuHaONClBr Found

Halogen 52.36 52.71

//. 4-Chlorimido-2-chlor-6-bromqmnone
This chlorimidoquinone was prepared in accordance with

» Graebe: Ber. d. chem. Ges.. 35, 43 and 2753 (1902).

3 Hilpert: Loc. cit.

3 This Journal, 30, 402 (1903).

422 Raijord

the method already described. The starting point was 2-chlor-
6-brom-4-nitrophenol, melting at 137°, which was obtained by
brominating 2-chlor-4-nitrophenol in glacial acetic acid solu-
tion.* A portion of this nitro compound was converted
into the corresponding amino hydrochloride^ by reduction
with stannous chloride in the manner already described. By
recrystallization of the product colorless plates, that black-
ened and decomposed without melting when heated above
225°, were obtained. Analysis for halogen gave the follow-
ing figures :

0.2614 gram substance gave 0.4789 gram AgHal.

Calculated for
CeHeONClzBr

Found

5827

58.26

Halogen

2-Chlor-6-hrom-4-aminophenol.^ — The free amine was read-
ily obtained from a water solution of the hydrochloride by
precipitation with ammonium carbonate. The mixture was
at once filtered and the residue well washed with water, and
dried on a clay plate. In this condition the substance is but
slightly colored and is otherwise practically pure. It is sol-
uble in alcohol, ether, chloroform, benzene, and a solution
of sodium hydroxide. It is best crystallized from benzene,
from which it separates in colorless, thin, elongated scales
having a melting point of 181°. Analysis for halogen gave
the following figures :

0.241 1 gram substance gave 0.3606 gram AgHal.

Calculated for
CeHiONClBr Found

Halogen 51.88 52.13

4-Chlorimido-2-chlor-6-bromquinone, O : CgHjBrCl : NCI. —
When a cold solution of the hydrochloride of 2-chlor-6-brom-
4-aminophenol was dropped slowly into a cold solution of
sodium hypochlorite in the usual way, a yellow solid having
the odor characteristic of a chlorimidoquinone was formed.

1 Ling: J. Chem. Soc, 66, 57 (1889). Clark: This Journal, 14, 563 (1892).

2 Clark iloc. cit.) found that this hydrochloride forms a double salt with tin chlor-
ide, but he did not isolate the hydrochloride of the amine.

8 This compound is probably identical with Clark's product, though he reports
neither melting point nor analysis.

On Chlorimidoquinones 423

The mixture was filtered, and the solid washed well with cold
water and dried on a clay plate. In this condition it melted
at 83°-84°. Attempts to fractionate it gave portions of ma-
terial having the same melting point. The chlorimide is sol-
uble in alcohol, ether, chloroform and ligroin. After being
crystallized from ligroin (70°-8o°), it melted at 87°-88°,
and decomposed suddenly, with charring, when heated to
about 175°. The crystals darken after a few days when
kept under ordinary conditions. Analysis for halogen gave
the results indicated below :

0.2770 gram substance gave 0.5143 gram AgHal.

Calculated for

CgHjONCbBr Found

Halogen 59 19 59

The study of a dihalogenated chlorimidotoluquinone was
next suggested, and an effort was made to obtain this by
means of 2,4,6-trichlor-m-cresol. It was hoped that this
could be nitrated in such a way as to replace by the nitro
group the chlorine atom in the para position to the hydroxyl
group. In the parallel case of 2,4,6-tribrom-?/i-cresol such
substitution takes place very readily, as observed by Claus
and Hirsch.^ No difficulty was experienced in preparing
trichlor-M-cresol — statements in the literature to the con-
trary notwithstanding — but it was then fourd impossible
to effect the desired substitution, the trichloride being very
much less reactive than the tribromide.

2,4,6-Trichlor-m.-cresol.^ — In a study of the actioij of
chlorine on the cresols, Claus and Schweitzer^ found that
m-cresol gave a dichlor compound, and that more chlorine
could not be introduced. Chandelon* studied the action of
chlorinating agents on phenol, and found that treatment
of| an aqueous solution of phenol with a solution of sodium
hypochlorite at 60°- 70° gave trichlorphenol. After re-
peating Chandelon's work and obtaining the same results, I

1 J. prakt. Chem.. [2] 39, 61 (1888).

2 The trichlortoluenol prepared by Lallemand (Jahresb., 1866, 620) and reported
by Beilstein as trichlorcresol, can hardly be identical with this compound.

3 Ber. d. chem. Ges.. 19, 929 (1886).
* Bull. soc. chim.. 38, 116 (1882).

424 Raijord

tried the method with m-cresol. An impure product was at
first obtained, but the greater portion of it turned out to be
the desired compound, 2,4,6-trichlor-m-cresol. When the
operation was carried out at room temperature instead of
60°- 70°, as in the first case, a much better yield of a purer
product was secured. This was further purified by distilla-
tion in an Anschiitz flask. The larger fraction came over
at i42°-i44° and 14 mm. as an oil which, upon cooling, crys-
tallized out in the form of colorless plates melting at 47°.

Trichlor-w-cresol is but slightly soluble in water, but is read-
ily soluble in a solution of sodium hydroxide, from which it
is completely precipitated by carbon dioxide. Alcohol,
ether, chloroform and ligroin dissolve it easily. Its satura-
ted aqueous solution is not colored violet by ferric chloride.
It was analyzed for chlorine and gave the following results:

o. 1849 gram substance gave 0.3773 gram AgCl.

Calculated for

C7H5OCI3 Found

CI 50 31 50 40

2,4,6-Trichlor-3-methylphenyl Acetate. — A portion (5 grams)
of trichlor-m-cresol was mixed with anhydrous sodium ace-
tate and slightly more than the theoretically required amount
of acetic anhydride. Action began at once, causing con-
siderable rise of temperature. The flask was warmed until
the contents were fluid, and it was then allowed to remain
on the electric heater in this condition for about ten minutes.
When the solid that formed upon cooling was mixed with
water, in order to remove sodium acetate, the acetyl deriva-
tive separated out as an oil, which was further washed with
water, dried over calcium chloride and distilled. At 273^-274°
the chief portion came over as a colorless, viscid liquid of
about the consistence of glycerol. The compound was analyzed
for chlorine and gave the following figures:

0.2107 gram substance gave 0.3578 gram AgCl.

Calculated for
C9H7O2CI3

Found

41-97

42.00

On Chlorimidoquinones 425

Several attempts were next made to nitrate 2,4,6-trichlor-
w-cresol. When a glacial acetic acid solution of the com-
pound was mixed with solid sodium nitrite/ in accordance
with Zincke's method, the unchanged trichlorcresol was re-
covered from the mixture. Shaking its ethereal solution
with silver nitrite also gave negative results. Fuming nitric
acid at the room temperature converts it into 2,6-dichlor-
toluquinone, along with resinous products.

2,6-Dichlortoluquinone. — Further proof of the structure
of 2,4,6-trichlor-m-cresol was obtained by oxidizing a por-
tion of it to quinone, which was done by the aid of dichromate
mixture. Five grams of trichlor-m-cresol was dissolved in
150 cc. glacial acetic acid, and the solution cooled to about
0°, and to this was added gradually the cold dichromate
solution. This mixture was allowed to stand for half an hour,
after which it was diluted with one volume of water. The
yellow solid that formed subsided during an hour, and was
then filtered oiSF, washed well with water, and dried on clay
plate. A yield of 72 per cent, was obtained.

This quinone is soluble in chloroform, ether, ligroin, and
alcohol. Portions crystallized from ligroin and alcohol, re-
spectively, melted at 103°, which agrees with the results ob-
tained by Claus and Schweitzer,^ when they oxidized dichlor-
w-cresol. My product is also probably identical with the
dichlortoluquinone prepared by Southworth^ from m-cresol,
though he states that he could obtain no melting point for
his compound, as it began to sublime at 100°. The sub-
stance obtained from trichlor-m-cresol was shown, both by
analysis and by the depression of its melting point when
mixed with 4-chlortoluquinone (melting at los*^), not to be an
impure sample of the latter. The mixture melted between
85° and 95°. Analysis for chlorine resulted as follows:

0.2242 gram substance gave 0.3380 gram AgCl.

Calculated for
C7H4O2CI2 Found

CI 3714 3724

» J. prakt. Chem.. [2] 61, 561 (1900).
aSer. d. chem. Ges.. 19, 931 (1886).
9 Aub. Cbem. (Uebig). IM, 270 (1873).

426 Raijord

2 ,6-Dichlortoluhydroquinone. — The dichlortoluquinone (2 . 5
grams) was mixed with 60 cc. water and the mixture
saturated with sulphur dioxide. After standing overnight
the mixture was heated to the boiling point, enough water
gradually added to dissolve the solid, and then the solution
was filtered through paper and set aside to cool. The needle-
shaped crystals that formed were collected and dissolved in hot
water and the solution saturated with sulphur dioxide. The
colorless needles that crystallized out melted at 171°, as found
by Claus and Schweitzer,^ and in close agreement with the
results of Southworth.^

After being dried in vactw over potassium hydroxide for
24 hours the hydroquinone was analyzed for chlorine.

o 3531 gram substance gave 0.5244 gram AgCl.

Calculated for
C7H6O2CI2

Found

36.76

36.70

///. 4-Chlorimido-2,6-dihrointoluquinone

The starting point in the preparation of this compound
was pure w-cresol, obtained by fractionating Kahlbaum's
product. From this material 2,4,6-tribrom-w-cresol was pre-
pared in accordance with Werner's method. ^

Pure tribrom-m-cresol obtained as above specified was next
nitrated.^ Thirty grams of the compound was dissolved in
300 cc. glacial acetic acid, the liquid cooled to i2°-i5° and
then 10 per cent, more than the calculated quantity of sod-
ium nitrite added during half an hour, while the flask was
continually shaken and the temperature kept down to that
given above. When the nitrite had all dissolved, the dark
liquid was poured with stirring into five volumes of water,
and the mixture set aside for some hours to allow the pre-
cipitate to subside. At the end of this time the yellow solid
was filtered off and dried on a clay plate. A yield of 95 per
cent, was obtained. The crude product softened at 65° and

' hoc. cit.

2 Bull. soc. chim., 46, 275 (1886).

3 Zincke: J. prakt. chem.. [2] 61, 561 (1900).

On Chlorimidoquinones 427

was completely melted at 113° (when it appeared to decom-
pose), which suggested the possible presence of two com-
pounds.' It may be stated at once that isomeric mononitro-
dibrom-w-cresols, viz.,

OH OH

Br Brr >Br

and I ,

H3 hI Jch,

Br NO2

were isolated from the mixture.

The best method of separating the compounds present
was found to consist in dissolving the dried crude product
in hot chloroform (i gram to i cc.) and treating this solution
with two volumes of ligroin (40°-6o°). Precipitation of the
high-melting isomer (para compound) took place at onc^.
After half an hour this was filterd off and crystallized from
benzene, when a compound melting at 128°, with decompo-
sition,^ was secured. Repeated crystallization from the same
solvent finally gave pale yellow plates melting at 134° with
decomposition.^ The substance so obtained has the nitro
group in the para position as respects hydroxyl, and is 2,6-
dibrom-4-nitro-w-cresol. A yield of 35 per cent, was ob-
tained.

When the filtrate from the chloroform-ligroin mixture
specified above was allowed to evaporate, it left a yellowish
red, sticky mass which, after repeated crystal-
lizations from alcohol, gave deep yellow needles melting
at 87°.^ This compound is isomeric with the nitro product
mentioned above, and was proved (see below) to have the
nitro group adjacent to hydroxyl. A yield of 40 per cent.
was obtained.

1 A mixture of the isomers, purified as described in this paper, melts at 79°-! 15°
with decomposition. Zincke reported the formation of only one isomer, in which
the bromine atom para to the hydroxyl group had been replaced.

2 Zincke: J. prakt. Chem , [2j 61, 563 (1900).

3 Claus: Ibid.. [2] 39. Si (1888).

* claus (loc. cit.) prepared this compound by brominating 6-nitro-*t-cresol, and
found 93° to be the melting point.

428 Raiford

In view of the fact that the melting points found for these
compounds did not agree with those given by other workers,
both substances were analyzed for halogen and several of
their derivatives prepared and studied. Analyses of the
nitro compounds for bromine gave the following figures:

I. The para compound, 2,6-dibrom-4-nitro-?n-cresol.

0.4668 gram substance gave 0.5662 gram AgBr. ,

Calculated for
CyHsOsNBrj

Found

51-43

51.26

II. The ortho compound, 2,4-dibrom-6-nitro-m-cresol.
0.4266 gram substance gave 0.5156 gram AgBr.

Calculated for
C7H503NBr2 Found

Br 51.43 51.42

Hydrochloride of 2,6-Dibrovi-4-amino-m.-cresol. — Five grams
of the p-m\XQ compound was dissolved in 20 cc. hot alcohol,
and to this solution was added stannous chloride dissolved in
concentrated hydrochloric acid, as before specified. The hy-
drochloride obtained upon cooling was recrystallized in the man-
ner already described, and gave colorless needles free from
tin compounds. The crystals are readily soluble in water
containing a trace of hydrochloric acid, in a solution of sod-
ium hydroxide (giving a liquid that at once becomes brown),
and in alcohol, and almost insoluble in chloroform. When heated
above 225° the substance blackens but does not melt.

2,6-Dibrom-4-amino-m-cresol. — The free amine was easily
prepared from the hydrochloride by treatment of an aqueous
solution ©f the latter with ammonium carbonate. The pre-
cipitated base, after being filtered off, washed well with water
and dried on a clay plate, melted at 175°- 176° with blacken-
ing.* It is readily soluble in alcohol, benzene, chloroform
and a solution of sodium hydroxide, less soluble in ligroin
and practically insoluble in water. Portions of the com-

> Zincke's statement (J. prakt. Chem., [2] 61, 564 (1900)) that this compound
melts at 116° is possibly a misprint, or he had the o-amino compound in his hands.
He appears to have overlooked the formation of this substance, which melts at 116°. as
Kiven below.

On Chloriviidoquinones 429

pound purified by crystallization from alcohol and by treat-
ment of a chloroform solution with an equal volume of ligroin
(40°-6o°), respectively, in both of which cases it separated
in colorless crystals, melted at 176°. It was analyzed for
nitrogen.

I. 0.4905 gram substance gave 0.0236 gram N (Kjeldahl).
II. 0.4925 gram substance gave 0.0238 gram N.

Calculated for
CyHrONBrz

Found

4.98

4.81

483

2,6-Dibrom-4-benzoylamino-m-cresol. — In order to furthet
identify the amine, a portion of it was benzoylated in ac-
cordance with the Baumann method, using 2.5 molecules
of benzoyl chloride. The product was found to be soluble
in alcohol, chloroform and benzene, less soluble in ligroin,
and insoluble in water. It was best purified by treatment
of its hot chloroform solution with one volume of ligroin (40°-
60°), when colorless crystals melting at 198° were obtained.
The pure compound was soluble in a solution of sodium hy-
droxide and from this hydrochloric acid precipitated the
original substance. These results indicated a monoben-
zoylated product, and they were confirmed by an analysis.
for nitrogen.

0.4550 gram substance gave 0.0165 gram N (Kjeldahl).

Calculated for
Ci4H„02NBr2

Found

363

3-62

3,yDibro'}n-6-'methyl-Y>-hydroxyphenylurethane, HOQHBrj-
(CH^ONHCOjCjHs.— The first efforts to prepare this com-
pound were made in accordance with Groenvik's* method,
but the yield was very small and the product could not
easily be purified. A much purer substance in satisfac-
tory yield was secured by shaking an alkaline solution of
2,6-dibrom-4-amino-w-cresol (obtained by dissolving the
amine in a solution of sodium hydroxide) with one molecule
of ethyl chlorcarbonate for about ten minutes and, after al-

1 Bull. soc. chim., 26, 177 (1876).

430 Raiford

lowing the mixture to stand for half an hour, acidifying it
with dilute hydrochloric acid. The substance thus obtained
was found to be soluble in a solution of sodium hydroxide,
from which acids precipitated it in the unchanged form.
It was easily purified by mixing with its warm chloroform
solution two volumes of ligroin (40°-6o°) and removing the
brownish solid that was precipitated at once. The urethane
melts sharply at 155°, and after being dried for 24 hours in
vacuo gave the following results when analyzed for nitrogen.
I. 0.6198 gram substance gave 0.0224 gram N (Kjeldahl).
II, 0.5752 gram substance gave 0.0217 gram N.

Calculated for
CoHnOsNBr,

Found

3 96

363

3-77

Up to this point the derivatives prepared and analyzed
indicate that the amine under consideration has the formula
assigned to it. The relative positions of the amino and the
hydroxyl groups weie determined by the behavior of the com-
pound when oxidized with the usual dichromate mixture, in
which case it gave a quinone melting at 117°, and evidently
identical with the 2,6-dibromtoluquinone that Claus and
Dreher^ obtained by oxidizing 2,4,6-tribrom-w-cresol. In
order to decide this a portion of the tribrom-w-cresol used in
the preparation of the compounds described above was oxid-
ized to quinone in accordance with the directions given by
Claus. A compound having a melting point of 117°, which
was not depressed when mixed with the quinone obtained
directly from the amine, resulted.

4-Chlorim'ido- 2,6-dibromtoluquinone, O : CgHBrjCCH^) : NCI.
— When a cold solution of the hydrochloride of 2,6-dibrom-4-
amino-w-cresol was oxidized in the usual way with an acidi-
fied solution of sodium hypochlorite, an immediate precipi-
tate of chlorimidoquinone was formed. This was filtered off

1 J. iwakt. Chem., [2] 39, 370 (1889). Claus and Dreher found that this dibrom-
toluquinone melts at 115°. Claus and Hirsch (Ibid., p. 60) report 117° as the melt-
infj point of the conx-sponding hydroquinone ; but this is probably an error, since hydro-
quinones usually melt much higher than their quinones. Reduction of our dibrom-
toluquinone with sulphur dioxide in the usual way gave a hydroquinone that crys-
tallized from water in colorless crystals and had a constant melting point of 150°.

On Chloriniidoquinones 431

and washed with water, and when dried on a clay plate it
melted at 85°-86°. A yield of 85 per cent, was obtained.
The compound is readily soluble in alcohol, chloroform, and
ligroin, but insoluble in water. Attempts to fractionate the
substance in accordance with the methods indicated above
gave crystals having the same melting point, 86°. The best
method of purification of the chlorimide is crystallization
from ligroin (40°-6o°), from which it separates in yellow
hexagonal plates melting at 86°, and having the characteris-
tic odor of this class of substances. Determination of nitro-
gen resulted as follows :

I. 0.5990 gram substance gave 0.0263 gram N (Kjeldahl).
II. o. 7515 gram substance gave 0.0334 gram N.

Calculated for

C7H40NClBr2 Found

N 4.46 4.39 4.44

•

It has been noted (p. 427) that the nitration of 2,4,6-tri-
brom-m-cresol under the conditions described in this paper
gives rise to two isomeric mononitro derivatives, one of which
(melting at 134° with decomposition) has been shown to
have the nitro group para to hydroxyl. When this had
been done it was at once suspected that the other one (melt-
ing at 87°) had the nitro group ortho to hydroxyl. The be-
havior of its derivatives described below shows that this is
the case.

Hydrochloride of 2,4-Dibrom-6-amino-m.-cresol. — Five grams
of 2,4-dibroni-6-nitro-m-cresol was dissolved in 50 cc. hot
alcohol and then reduced with stannous chloride in the usual
way. It was noted that the hydrochloride obtained is much
less readily soluble in water than is its isomer, the para com-
pound, and could not be purified in exactly the manner em-
ployed with the latter. Two m.odifiications were tried, viz.,
crystallization from alcohol and treatment of the alcoholic
solution with concentrated hydrochloric acid. In both cases
the crystals secured were free from tin compounds. Exactly
as in the case of the para compound, this hydrochloride is
but sparingly soluble in chloroform, but is readily soluble

43* Raiford

in a solution of sodium hydroxide, forming a brown liquid.
Nitrogen determinations gave the following figures:

I. 0.6942 gram substance gave 0.0292 gram N (Kjeldahl).
II. 0.5553 gram substance gave 0.0239 gram N.

Calculated for Found

CTHgONClBra I II

N 4.41 4.20 4.30

2,4-Dibrom-6-ammo-Ta-cresol. — A solution of the hydro-
chloride in very dilute hydrochloric acid was mixed with a slight
excess of a solution of ammonium carbonate, and the pre-
cipitated amine at once filtered off and washed well with
water. When dry the base was further purified by treat-
ment of its warm chloroform solution with two volumes ligroin
(40° — 60°), from which it crystallized in brownish scales
melting at ii6-°ii7°.^ Alcohol, chloroform and solutions
of sodium hydroxide dissolve the amine easily, giving in the
latter case a strongly colored liquid. Nitrogen analyses gave
the results recorded below:

I. 0.6323 gram substance gave 0.0326 gram N (Kjeldahl).

II. 0.6683 gram substance gave 0.0342 gram N.

^ i* Calculated for Found

CjHTONBra I II

N 498 515 511

2 ,4.-Dihrorn- s-mcthyl-6-henzoylaniinophenyl Benzoate. — Tlie
amine just described was further identified by treatment
with benzoyl chloride by the Baumann method, 2 . 5 mole-
cules of benzoyl chloride being used. The product was col-
lected on a filter, dried on a clay plate, and repeatedly crys-
tallized from alcohol until it was practically colorless and melted
sharply at 188°. The compound is easily soluble in chloro-
form and ether, less readily so in ligroin, and insoluble in solu-
tions of sodium hydroxide. It was regarded as the diben-
zoylated compound, and analysis for nitrogen agreed with
this view :

J It will be noted that Zincke (J. prakt. Chem., [2] •!, 564 (1900)) found 116*
as the melting point of the para com{x>und, while I found 176" for the compound
that I proved to be the para product.

On Chlorimidoquinones 433

I. 0.5070 gram substance gave 0.0156 gram N (Kjeldalil).
II. 0.5070 gram substance gave 0.0159 gram N.

Calculated for
CjiHisOjNBr.

Found

2.86

307

3,5-Dibroin-4-meihyl-o-hydroxyphenylurethane, HOCjH (CH.j) -
BrjNHCOzCjHj. — As in the case of the para compound,
an attempt was made to prepare this urethane by Groen-
vik's methods but the very small yield, taken in connection
with the difficulty experienced in obtaining a pure pro-
duct, made it advisable to try some other method. A yield
of 60 per cent, was obtained by shaking an alkaline solution
of the corresponding aminocresol with one molecule of ethyl
chlorcarbonate, and the product was not difficult to purify.
It was crystallized by miximg with its warm chloroform solu-
tion an equal volume of ligroin (40°-6o°). The ciystals
that separated were brownish and readily soluble in alcohol and
in solutions of sodium hydroxide, from which liquid they \vere
separated unchanged by the addition of acid. They melted
sharply at 169°. A portion of the compound dried in vacuo
for 24 hours was analyzed for nitrogen.

I. 0.6776 gram substance gave 0.0271 gram N (Kjeldahl).

II. 0.5376 gram substance gave 0.0201 gram N.

Calculated for Found

CioHiiOiNBro I II

N 3 96 3-99 3-73

The preparation and analyses of the derivatives of the
dibromnitro-w-cresol now under consideration indicate that
it is isomeric with the one already disposed of, and which was
shown to have the structure

BwU. »oc. chim.. 26. 177 (1876).

434 Raiford

by the ease with which it was ultimately converted into the
corresponding quinone. Direct proof that the substance
now in question has the nitro group adjacent to hydroxyl,
and has the structure

ICH3

was secured as follows :

In a study of the behavior of aminophenols, Ladenburg^
found that he could distinguish between the ortho and para
varieties by the difference in reaction of the two compounds
toward acetic anhydride. The ortho compounds give con-
densation products which he called ethenylaminophenols,
or anhydro bases,

O— QH — N = C— CH3

J 1

while the para compounds give diacetyl derivatives,

CH3CO . O . QH^NHCOCHg.

This method v/as tried on the aminocresol obtained from the
nitro compound now being studied.

2, 4-Dibrom-j-methyl-6-acetylaTnino phenyl Acetate. Three

grams of the corresponding dibromamino-5;i-cresol was mixed
with acetic anhydride (2 mol.) as above indicated. Inter-
action began immediately with the evolution of heat. The
flask was next attached to a reflux condenser and warmed
gently until the mass became liquid, after which it was boiled
for two hours. Upon cooling, a solid that melted at 207°
was formed. Repeated crystallization from alcohol gave
long, colorless, silky needles melting sharply at 216°, soluble
in chloroform and ligroin, but insoluble in solutions of sodium
hydroxide except on standing. The behavior toward alkali

» Ber. d. chem. Ges., 9, 1524 (1876).

On Chlorimidoquinones 435

pointed to a diacetyl derivative, and this indication was con-
firmed by analyses for nitrogen and bromine.

I. 0.4134 gram substance gave 0.0165 gram N (Kjeldahl).

o. 2382 gram substance gave 0.2466 gram AgBr.
II. 0.371 1 gram substance gave 0.0144 gram N.

Calculated for
C„H„03NBr.^

Found

383

3 99

3-8

43.82

44.08

That the base is nevertheless an o-aminophenol derivative
is shown conclusively by the following determination. In a
study of the rearrangements of o-aminophenylethyl carbon-
ate, Ransom,' working with Stieglitz, showed that when its
hydrochloride is dissolved in water it goes over into o-hydroxy-
phenylurethane. Upson, ^ also working with Stieglitz, studied
this rearrangement in quite a number of substances, and found
that while it is a general reaction for ortho compounds it
does not take place with the corresponding para compounds.
With these facts in view, tie dibromnitro-w cresol (melting
at 87°) in question was conv^erted into its carbonate,
and this was reduced to the amino hydrochloride, and the
latter then allowed to react with water in the usual way. Its
behavior showed conclasively tliat the dibromnitro-m-cresol
(melting at 87°) is an o-nitrophenol, and that its reduction
product is an o-aminophenol.

2,4-Dibrom-j-methyl-6-niirophenylethyl Carbonate, 02NCeH-
Br2(CH3)OC02C2H5. — Six grams of 2,4-dibrom-6-nitro-w-cresol
was dissolved in 300 cc. water containing slightly more than
the theoretically required amount of sodium hydroxide, and
to this was added i . 25 molecules of ethyl chlorcarbonate, after
which the mixture was shaken for twenty to thirty minutes,
or until the red color characteristic of the sodium salt of
the nitrocresol had disappeared. A yellow oil, that solidified
on cooling, separated out, and after standing for some time
was filtered and dried. The substance is very soluble in
alcohol, ether, chloroform, ligroin and glacial acetic acid, and

' This Journal, 23, 43 (1900). ^
a/Wd., 32, 13 (1904).

436 Rg ijord

could not be crystallized from any of these liquids. The purest
sample secured was obtained by pouring the glacial acetic
acid solution into water. The oil that separated solidified
on standing, and was then filtered off and dried. It melted

at43°-45°-

The nitrocarbonate was next reduced^ by shaking its glacial
acetic acid solution with tin^ and concentrated hydrochloric
acid until the liquid was practically colorless, after which it
was filtered through paper and the filtrate mixed with two
volumes of water and then allowed to stand overnight in
order to enable the phenolcarbonate to rearrange and form
the corresponding urethane by a migration of the carbonate
group from the phenol group to the amine group. ^ The solid
that formed was collected on a filter and dried on a clay plate.
In this condition it melted at 167°-! 69°. It was further
purified by solution in warm chloroform and treatment of
this solution with one volume of ligroin (6o''-8o°). The
crystals so obtained melted sharply at 169°, which is the
melting point of 3,5-dibrom-4-methyl-o-hydroxyphenylure-
thane already described (p. 433). A mixture of these two
products has the same melting point as either of them
separately, from which it follows that they are identical,
and that in the dibromnitro-m-cresol in question the
hydroxyl and the nitro groups are adjacent to each other.
It will be noted that this method of proving the relative
positions of the hydroxyl and nitro (amino) groups by means
of rearrangement is easier of execution and gives more con-
clusive evidence than does Ladenburg's method.

IV. The Three Chlorimidochlortoluqiiinones

In their proof of the structure of the stereoisomeric oximes
obtained from 4-chlortoluquinone, Kehrmann and Tichvinsky*
found that reduction of the oximes by means of stannous
chloride gave a chloraminocresol that is identical with the
base they obtained when they reduced the product secured

' Ransom: This Journal. 23, 14 (1900).

2 Equally satisfactory results were obtained by the use of zinc.

3 Ransom: Loc. cit.

••Ann. Chem. (Liebig), 308, 20 (1S9S).

On Chlorimidoquinones 437

by chlorinating 6-mtro-w-cresol in glacial acetic acid solu-
tion. In the latter reaction Kehrmann^ assumed that chlor-
ine took a position para to methyl, and from such a view
assigned to the oximes the stereoisomeric formulas

NOH HON

H/ \CH3 H/ ],CH3

and

Cll[ ^H CW ^H

O O

He states tliat this is the structure most favorable to the oc-
currence and persistence of this form of stereoisomerism.
My experiments in the attempts to prepare the stereoisomeric
chlorimidoquinones corresponding to these oximes involved
the preparation of a chloraminocresol from 6-nitro-m-cresol,
and it was found to have a melting point of 1 66°- 167°, which
is different from that (204°-205°) given by Kehrmann, a fact
that at once made necessary a repetition of Kehrmann's
work on the oximes. The latter were prepared (the stereoiso-
mers were not separated) and reduced according to the direc-
tions given, and the resulting hydrochloride of the chlor-
aminocresol decomposed by ammonium carbonate in the
usual way. The base obtained melted at 206 "-207°, which
is in close agreement with the observation of Kehrmann.
The acetyl derivative, the quinone and the hydroquinone
obtained from this base also had the properties reported by
Kehrmann and others^ in the literature. It was further
identified by the preparation of its dibenzoate (p. 444)-

When the chloraminocresol (melting at 2o6°-207°) was
oxidized to the chlorimidoquinone it gave a product melting at
91°, while the chloraminocresol (melting at i66°-i67°)

• Loc. cU.

2 Schniter's statement (Ber. d. chem. Ges.. 20, 2286 (1887)) that the monochlor-
inated toluquinone, melting at 105°, could not be obtained free from the higher chlor-
inated products is not confirmed by the analysis reported by him. The calculated
percentage of chJorine given by him is incorrect.

438 Raiford

obtained from the chlorine derivative of /j-nitro-m-cresol
gave a chlorimidoquinone melting at 87°. A mixture of the
two chlorimides melts at 65 "-70°, which shows that they are
not identical. If Kehrmann's assumptions as to the structure
of the chlornitro-w-cresol formed by chlorinating p-n\Xxo-m-
cresol were correct, the two chlorimides would have the same
structure and would therefore be stereoisomers of the type
sought by us. However, the low melting point obtained by
me for the chloraminocresol prepared by way of chlomitro-
w-cresol, as compared with the melting point (204°-205°)
given by Kehrmann for the chloraminocresol obtained by
reduction of his oximes (and confirmed by me) suggested
plainly that the two chlorimides, as well as the two amino
products, might well be structural isomers rather than stereo-
isomers. By subsequent investigation I was able to prove
that the chlorimides are indeed structural isomers — the chlor-
imide, melting at 91°, obtained from Kehrmann's chloramino-
cresol having the structure of 6-chlorimido-4-chlortoluquinone,

NCI

H|r >CH3

cil Ih

while that obtained from the chloraminocresol, melting at
i66°-i67°, derived from /?-nitro-w-cresol of is 6-chlorimido-2-
chlortoluquinone,

' NCI

On Chlorimidoquinones 439

melting at 87°. A third chlorimidochlortoluquinone, melt-
ing at 65°, was obtained from 2-methyl-4-amino-5-chlor-o-
cresol (from m-nitro-/j-chlortoluene) and has the struc-
ture of 4-chlorimido-2 methyl-5-chlorquinone,

cu ;h

NCI

6-Chlorimido-4-chlortoluquinone. — It is possible that when
Kehrmann chlorinated /j-nitro-m-cresol, the chlorine, to a
certain extent at least, entered the position para to the methyl
group, in the way he assumed (Kehrmann did not isolate the
chloride) ; but when I carried out the chlorination of the same
substance, following Kehrmann's directions as closely as
the brief descriptions will permit, except that I isolated and
purified all m}- products, the chlorine, as proved below, must
have persistently entered chiefly irto the position ortho to
the meth>l group, that is, between it and the hydroxyl group.
It seems possible, now that I have cleared up the facts, that
in both cases a mixture of both chlorides is formed — the pro-
portions varying according to conditions that have not yet
been determined; and that Kehrmann obtained derivatives
(amine, etc.) of one compound, while I secured those of another.
It is beyond question, howcv^er, that the derivative which I
isolated is always formed under the conditions followed by
me. In a very closely related case, the chlorination of o-nitro-
toluene, I was able to establish conclusively (see below) the
formation of the two chlorine derivatives, containing chlorine
ortho and para to the methyl group, although Janson* had
reported the formation of but one (in this case the ortho
compound).

1 Centralb., 1900, I, 1110.

440 Raiford

The proof of the structural isomerism of the three chlor-
imidochlortoluquinones, melting at 91°, 87°, and 65°, re-
spectively, consisted in showing first that not only do the
three chlorimides depress each others' melting points, but
that all their derivatives — their reduction products, the cor-
responding aminophenols and their acyl derivatives, the quin-
ones obtained by the oxidation ol the aminophenols, and the
hydroquinones obtained by the reduction of the quinones —
form different series with different melting points and other
properties. If the difference were due to space relations of
the : NCI group, this difference would vanish with reduction
of the group, and identical compounds would result. In
the second place, the structure of each of the three chlor-
imides was brought into definite relations with simpler deriv-
atives of known or readily proved structure.

The chlorimidochlortoluquinone whose melting point is
65° was brought into relation with 3-nitro-4-chlortoluene,
and found to have the structure

Hjj >jCH3

Cll! ^H

NCI

This connection was established by converting m-mtro-p-
toluidine into 'm-nitro-/5-chlortoluene by the Sandmeyer re-
action, transforming this 7«-nitro-/j-chlortoluene into the
corresponding chloraminocresol (melting at 197°-! 99° with
blackening) by electrolytic reduction and rearrangement
in concentrated sulphuric acid solution by Gattermann's
method,* and by converting this chloraminocresol directly
into the chlorimide (melting at 65°). The steps may be
summarized as follows:

» Gattermann and Kaiser: Ber. d. cbem. Ges.. 18, 2599 (1895) .

On Chlorimidoquinones

CH,

h/ \ch3

a „

NO2

^CH,

441

ci( Ih

NH, II

NCI

The chlorimidochlortoluquinone that melts at 91° was ob-
tained by the action of hypochlorous acid on the reduction
product of Kehrmann's oximes; the oximes in turn were ob-
tained by Kehrmann from so-called /j-chlortoluquinone in
which the chlorine is assumed to be in the para^ position to
the methyl group. The direct proof that such is the case
was brought by the oxidation of 4-chlor-3-amino-6-hydroxy-
toluene, prepared as just described. The same chlortolu-
quinone (melting at 105°) as that used by Kehrmann in the
preparation of his oximes was obtained. The para position
of the chlorine atom being thus established, the chlorimide
melting at 91 ° must have the structure

1 That position for chlorine was generally assumed in the chlortoluquinone from
which Kehrmann started, but we can find no direct proof for the assumption. The
chlortoluquinone is obtained by oxidation of chlortoluhydroquinone which, in turn, is
obtained by the action of hydrochloric acid on toluquinone, in accordance with Schni-
ters (Ber. d. chem. Ges., 20, 2286 (1887)) reaction. By the corresponding action of
hydrobromic acid on toluquinone, Kehrmann and Rust (Ann. Chem. (Liebig), 303,
24 (1898)), obtained a bromtoluquinone melting at 105°, and evidently identical
with the compound (melting at 106") in which Gattermann (Ber, d. chem. Ges., 27,
1931 (1894)) proved definitely that bromine goes into a position para to methyl. But
Schniter (Ber. d. chem. Ges., 20, 1317 (1887)) has shown by his work on thymoquin-
ones that the position taken by chlorine in his reaction is sometimes not the same
as that taken by bromine.

442 Raiford

O NCI

H^ \CH,

H^ ^CH,

NCI
I

O
II

But structure (I) has just been proved to be that of the chlor-
imide melting at 65°, obtained from the chloraminocresol
melting at i97°-i99°, which cannot possibly be stereoiso-
meric, but must be structurally isomeric, with the chloramino-
cresol that melts at 206 ^-207°, and which was obtained by
the reduction of the chlorimide melting at 91°. The latter,
therefore, cannot be stereoisomeric, but must be structurally
isomeric, with the chlorimide melting at 65°, and must have
structure (II), and its chloraminocresol (melting at 206°-
207°) must be 4-chlor-6-amino-3-hydroxytoluene,

NH,

This, it may be added, brings proof, also, of the correctness
of the structure assigned by Kehrmann to his stereoisomeric
oximes and their derivatives. His own proof, in the light
of the above, does not appear unassailable.

The third chlorimidochlortoluquinone (melting at 87°),
first obtained from the chlorination product of p-mtro-m-
cresol, has the structure

On Chlorimidoquinones 443

NCI

HjJ ^CH^

m l!ci

The chlortoluquinone obtained by oxidation of its reduction
product, chloraminocresol (melting at i66°-i67°), was brought
by me into relationship with o-chlor-o-nitro toluene. This
was converted into the corresponding aminocresol by elec-
trolytic reduction in sulphuric acid solution, and from this
was obtained, by oxidation, the same chlortoluquinone (melt-
ing at 55°) as from the chloraminocresol melting at i66°-
167°. The changes are indicated as follows:

NO2 NH2

CH3

OH OH

In this way the position of the chlorine atom in the chlor-
imide and its derivatives is established. The position of the
chlorimide group is determined by the preparation of the
chlorimide from /?-nitro-m-cresol :

444

ar ^CH3
hI Jh

NCI

h!; fci

Melts at i66°-i67° Melts at 87°

The experimental data on which these conclusions are based
w ill now be given :

2-Chlor-4-henzoylamino-5'methylphenyi Benzoate. — This com-
pound was prepared as a further simple derivative to identify
Kehrmann's aminocresol, melting at 2o6°-207°. A portion
of the amine was dissolved in a solution of sodium hydroxide,
and this was shaken with benzoyl chloride (3 mol.) until the
odor of the chloride had about disappeared. The solid that
separated was filtered off and repeatedly crystallized from
alcohol until it was nearly colorless. It melts at 220°. The
substance was found to be soluble in chloroform and ether,
but insoluble in a solution of sodium hydroxide. This fact
and the figures given on analysis indicate a dibenzoylated
compound :

I. 0.5562 gram substance gave 0.0195 gram N (Kjeldahl).

II. 0.3400 gram substance gave 0.0124 gram N.

Calculated for Found

CMHieOsNCl I II

N 3 83 3 50 3 63

3 4 16

6 - Chlorimido - 4 - chlortoluquinone, O : CgHgCKCHg) : NCI. —
Five grams of the hydrochloride of the aminocresol obtained

On Chlorimidoquinones 445

by reduction of Kehrmann's oximes was dissolved in 100 cc.
water to which a trace of hydrochloric acid had been added,
and the solution cooled to about 0° and then dropped slowly
into a cold acidified solution of sodium hypochlorite. The
yellow solid that formed was filtered off, washed with cold
water and dried. The crude product melted at 88°, and at-
tempts to fractionate it gave lots of crystals having the same
melting point, 91°. It is readily soluble in alcohol, chloro-
form and ligroin, from which latter it is best crystallized,
separating in the form of needles radiating from a common
center. A warm alcoholic solution of the compound was
easily reduced by stannous chloride dissolved in concentrated
hydrochloric acid, with the formation of the corresponding
chloraminocresol .

The chlorimide was analyzed for chlorine and for nitrogen
and gave the following results :

I. 0.2067 gram substance gave 0.3120 gram AgCl.

0.4753 gram substance gave 0.0336 gram N (Kjeldahl).

II. 0.2107 gram substance gave 0.3192 gram AgCl.
0-535I gram substance gave 0.0376 gram N.

C7H5ONCI2

37-32
7-37

37-32
7.06

37-45
7-03

Proof of the Structure of p-Chlortoluquinone. — For the proof
of the structure of the chlortoluquinone which Kehrmann
employed in the preparation of his stereoisomeric oximes,
and, at the same time, the determination of the position of
chlorine in the chlorimide that has just been described, a
quantity of pure 3-nitro-4-chlortoluene^ was first prepared
by the Sandmeyer reaction from 3-nitro-4-toluidine, and then
reduced electrolytically to an aminocresol by Gattermann's
method. For the latter purpose 10 grams of the pure oil
was dissolved in 100 grams of sulphuric acid (184), and to
this solution 0.5 cc. water was added. The liquid so obtained
was poured into a porous cup, and the latter placed in a beaker
-of convenient size so that the distance between the outer

1 Gattermann and Kaiser; Ber. d. Chem. Ges.. 18, 2599 (1885).

446 Raiford

wall of the porous cup and the inner wall of the beaker was
about 1.5 cm . Outside the porous cup was placed a mix-
ture of 100 grams of sulphuric acid (i . 84) and 5 grams of water.
The cathode was placed in the liquid in the porous cup and a
current of 7 . 5 volts and 1.9 to 2.1 amperes was passed for
15 hours. The liquid was then allowed to cool and was poured
into six volumes of water. Slight precipitation^ took place
at once, and the mixture stood urtil it had reached the room
temperature and the supernatant liquid was clear. This was
next filtered ofF, and the filtrate, which was deep blue in color,
was neutralized with a solution of sodium carbonate. The
chloraminocresol formed was precipitated. After filtration
the residue was washed with water, and, without further puri-
fication, was at once dissolved in very dilute hydrochloric
acid, cooled and mixed with a solution of ferric chloride.
When the mixture so obtained was distilled with steam, a quin-
one melting at 105°, and identical with that obtained by
oxidizing Kehrmann's chloraminocresol (melting at 206°-
207°), passed over. A mixture of the two substances melts
at the same temperature as either of them separately. A
yield of 60 per cent, was obtained. The reactions may be
expressed as follows :

H OH

cil Ih

NH,

3-Chlorimido-4-chlortoluquinone {melting at 65°),
6 41 3

O : C6H2C1(CH3) : NCI.— A second portion of the chloramino-
cresol, obtained from 3-nitro-4-chlortoluene by electrolytic

1 This precipitate is probably a sulphonic acid. Gattertnann found that with
o-nitrotoluene the chief product was a sulphonic acid of the corresponding amino-
cresol. Noyes and Clement (Ber. d. chem. Ges., 26, 991 (1893)) found that the sul-
phonic acid was not formed when they used sulphuric acid below a certain concentra-
tion. In my work some sulphonic acid appeared to form in every case, the amount
being smaller when halogen was in the ring.

On Chlorimidoquinones

447

reduction, was dissolved in dilute hydrochloric acid and the
solution cooled to about o° and then dropped slowly into a
cold acidified solution of sodium hypochlorite in the usual way.
A chlorimidoquinone, having the characteristic odor, and mixed
with some resinous products, was precipitated. It was col-
lected on a filter, washed with cold water and dried. At-
tempts to fractionate gave portions of solid all having the
same melting point, 65°. The compound is soluble in alco-
hol, ether, and chloroform, but is best crystallized from ligroin,
from which it separates in brownish wartlike nodules. Analy-
sis for chlorine indicates that it has the same composition
as, and is isomeric with, 6-chlorimido-4-chlortoluquinone,
melting at 91°, described above.

o. 1965 gram substance gave 0.2969 gram AgCl.

Calculated fo:
C7H5ONCI2

3732

Found

37-35

6-Chlorimtdo-2-chlortoluquinone {melting at 87°),
3 21 6

O : C6H2C1(CH3) : NCI.— The starting point in the preparation
of this compound was w-cresol, which was obtained by frac-
tionating Kahlbaum's pure product. Fifty grams of the
liquid boiling at i99°-2oo° (uncorr.) was nitrated in accord-
ance with the method of Staedel and Kolb,* and the isomeric
nitro products that resulted were separated by distillation
with steam. The para compound, which is not volatile un-
der these conditioms, and which was obtained in crude form
from the distillation residue, was further purified by succes-
sive crystallization of its sodium salt from water. These
crystals were then decomposed by treatment of their aqueous
solution with dilute sulphuric acid, and the free nitrocresol
subsequently crystallized from hot water, from which it sep-
arated in very nearly colorless needles melting at 127°- 129°.
A yield of 38 to 40 per cent, was obtained.

3 2 I 6

2-Chlor-6-nitro-m-cresol, H0.C6H2C1(CH3)N02.— Ten grams
of purified 6-nitro-m-cresol was chlorinated,- the operation

1 Ann. Chem. (Liebig). 269, 210 (1890).
^Ibid., 303, 23 (1898).

448 Raiford

being started at the room temperature, and the reaction mix-
ture was poured with stirring into six volumes of water. A
colorless solid precipitated out at once, and was filtered off
and dried on a clay plate. The substance is soluble in ben-
zene, ligroin and chloroform, very soluble in alcohol, and but
very slightly soluble in water. After being crystallized
twice from benzene it melted sharply at 133°, and this was
not changed when a portion of the material was crystallized
in the form of its sodium salt, this decomposed by acid, and
the free cresol again crystallized from benzene. Analyses*
for chlorine and for nitrogen gave the following results:

o. 2302 gram substance gave o. 1784 gram AgCl.

0.1 88 1 gram substance gave 12.25 cc. N at 21° and 752.5
mm. (uncorr.).

Calculated for

Found

I915

7-34

Hydrochloride of 2-Chlor-6-amino-m.-cresol. — Five grams of
2-chlor-6-nitro-m-cresol was dissolved in hot alcohol and re-
duced with stannous chloride in the manner already described,
forming practically colorless crystals that darkened when
heated above 225°, but did not melt when heated as high as
250°. The crystals are readily soluble in water acidulated
with hydrochloric acid, in alcohol and in a solution of sodium
hydroxide with the formation of a brown liquid. A sample
dried in vacuo for 72 hours over potassium hydroxide gave
the following results when analyzed for chlorine :

0.2674 gram substance gave 0.3966 gram AgCl.

Calculated for
CyHsOaNCl

18.90

7.46

Calculated for
C7H»ONCl2

Found

36.54

36.66

3 316

2-Chlor-6-amino-m-cresol, HO.CeH2Cl(CH3)NH2.— The free
base was easily obtained by treatment of an aqueous solu-
tion of the hydrochloride with ammonium carbonate. The
precipitate was collected on a filter at once, washed with water

' Kehnnann (loc. cil.) reports no analyses for the intermediate products through,
which he obtained proof of the structiire of his oximes.

On Chlorimidoquinones 449

and dried. In this condition the compound was nearly pure,
as was shown by the fact that crystallization did not change
the melting point. It dissolves in alcohol, benzene, chloro-
form and solutions of sodium hydroxide, but much less read-
ily in ligroin. Portions were obtained in practically colorless
crystals by crystallizing from 75 per cent, alcohol. These
melted at 1 66°- 167°, with slight darkening. That this sub-
stance could not be an impure specimen of the compound
obtained by reducing Kehrmann's oximes was shown by the
fact that a mixture of the two melts lower than either of them
separately, viz., 157°, as well as by the fact that the deriva-
tives now to be described are not identical with the corre-
sponding compounds obtained from the amine secured by the
reduction of the oximes. Analysis for chlorine gave the fol-
lowing results :

0.2982 gram substance gave 0.2716 gram AgCl.

Calculated for

CtHsONCI Found

CI 22.50 22.51

4' Acetyiammo-ymethyl-d-chlor phenyl Acetate. — In order to
further identify this aminocresol, it was converted into an
acetyl derivative by warming it with anhydrous sodium ace-
tate and acetic anhydride (1.5 mol.). After recrystalliza-
tion from benzene, long, colorless, silky needles melting at
178° were secured. A mixture of this substance with the di-
acetyl derivative (melting at 162°) obtained from 4-chlor-6-
amino-m-cresol melts at 140°- 149°. The new compound did
not dissolve in solutions of sodium hydroxide, and this be-
havior, taken in connection with the percentage of chlorine
found upon analysis, indicated a diacetyl derivative:

0.271 1 gram substance gave o. 1618 gram AgCl.

Calculated for
CnHuOgNCl Found

CI 14.67 14.75

3 116

2-Chlortoluquinone, O : CgHjClCHj : O. The new chlor-

aminocresol was further characterized by conversion of a por-
tion of it into its quinone by treatment with a cold dichromate

450 Raiford

mixture. After standing, the precipitate tliat formed was
collected on a filter and dried. A yield of 60 per cent, was
obtained. A more convenient method is to mix the cold,
slightly acid solution of the h3'drochloride of the amine with
a solution of ferric chloride, and then to distil the quinone
with steam. The quinone is appreciably soluble in water,
and is very readily soluble in alcohol, chloroform and ligroin.
Crystals deposited from a solution in the latter solvent melted
at 55°, and a mixture of the substance witli 4-chlortoluquin-
one, employed by Kehrmann in the preparation of his stereo-
isomeric oximes, melts at 46°, which shows that the com-
pound in question is a new one. It completes the list of pos-
sible chlortoluquinones, the other two being already known,
viz., /?-chlortoluquinone, melting at 105°, prepared by Schniter^
and m-chlortoluquinone, melting at 90°, prepared by Claus
and Schweitzer.- The syntheses (see below) and anaylsis for
chlorine indicate that this compound is isomeric with the sub-
stance Kehrmann used as a starting point :

0.2830 gram substance gave 0.2611 gram AgCl.

Calculated for

C7H6O2CI Found

CI 22.65 22.79

3 216

2-Chlortoluhydroquinone, HOC6H2Cl(CH3)OH. — A portion of
2-chlortoluquinone was mixed with water and the mixture
saturated with sulphur dioxide. The yellow color of the quin-
one was discharged, and after standing overnight in a cool
place the solid was nearly colorless. This was filtered off and
repeatedly recrystallized from hot water, the filtered solu-
tion in each case being decolorized by sulphur dioxide. Short,
colorless leaflets melting at 173°, that colored slightly when
exposed to the air, crystallized out. These were found to
be soluble in alcohol but much less so in chloroform, ligroin,
and benzene. In solutions of sodium hydroxide they dis-
solve, giving first a green color that rapidly changes to a dark
red. A mixture of this compound with 4-chlortoIuhydro-

1 Ber. d. chem. Ges., 80, 2286 (1887),

2 J. prakt. Chem., [2] 38, 328 (1888).

On Chlorimidoquinones 451

quinone, melting at 176°, melts at i46°-i58°, which shows
that they are not identical. Analysis for chlorine indicates
that they are isomeric.

0.3205 gram substance gave 0.2895 gram AgCl.

Calculated for

C7H7O2CI Found

CI 22.36 22.33

3 216

6-Chlorimido-2-chlortoluquinonc, O : C6H2C1(CH3) : NCI. A

portion of the new chloramino-w-cresol, melting at i66°-
167°, was dissolved in very dilute solution of hydrochloric
acid, and this liquid was cooled to about 0° and then dropped
slowly into a cold acidified solution of sodium hypochlorite in
the usual way. A yellow solid, having the odor characteristic
of a chlorimidoquinone, separated out at once. This was filtered
off, washed with cold water, and dried on a clay plate. It
melted at 86°-87°. All lots of crystals obtained in attempts
to fractionate it had the same melting point, 87°. A mix-
ture of this compound with 6-chlorimido-4-chlortoluquinone,
melting at 91° (obtained from the chloraminocresol that re-
sulted from the reduction of Kehrmann's oximes), melts at
6o°-65°. Analyses for chlorine and for nitrogen gave the
following results :

I. o. 1671 gram substance gave 0.2536 gram AgCl.

0.2050 gram substance gave 0.0146 gram N (Kjeldahl).

II. o. 1648 gram substance gave o. 2496 gram AgCl.
0.4864 gram substance gave 0.0353 gram N.

Calculated for
CtHsONCIj

Foimd

37-32

37-52

37-44

7-37

7.12

7.27

Synthesis of 2-Chlortoluquinone

Action of Chlorine on 2-Nitrotoluene. — Janson* passed chlor-
ine (i molecule) into dry 2 -nitro toluene and reports the forma-
tion of a chlor compound melting at 37° and boiling at 236°-
238°, to which he assigned the structure of 2-nitro-6-chlor-
toluene. As necessary precautions Janson states that the nitro-

1 Centralbl., 1900. I, 1110.

452 Raiford

toluene must be pure and dry, that a chlorine carrier should
be used, and that excess of chlorine must be avoided.

In accordance with these directions, loo grams of pure dry
2-nitro toluene, boiling at 218°, obtained by fractionating
Kahlbaum's product, was mixed with 2 grams of iron filings
dried at 115°, placed in a suitable vessel, and protected from
moisture by a tube containing calcium chloride. Into this
liquid the calculated amount of chlorine (obtained by the
interaction of potassium permanganate and hydrochloric
acid) was passed. The operation was begun with the liquid
at the room temperature, but 30^-40° was reached during
the course of the reaction. When the liquid had cooled it
was filtered from the excess of iron, made alkaline with a solu-
tion of sodium hydroxide and distilled with steam. The oil
that came over was separated from water, dried over calcium
chloride and cooled with ice and salt. No crystals* v/ere ob-
tained. The oil was next slowly fractionated, a Glinsky dis-
tilling tube being used, and the fraction boiling at 236°-238°
removed and cooled. Only a very small amount of solid
separated.

When the above operation was carried out it was expected
that chlorine might enter both ortho and para positions as
respects methyl, and thus give rise to the isomers 2-Ditro-6-
chlortoluene and 2-nitro-4-chlortoluene. In order to prove
that this had occurred, 30 grams of the oil (boiling at 236°-
238°) was reduced to the corresponding toluidine by means
of tin and hydrochloric acid, after which the mixture was
made alkaline and distilled with steam. The oil that came
over was dried and fractionated and the portion obtained
between 237°-245°- was next boiled for ten hours with glacial
acetic acid and the mixture poured into 500 cc. water. The
acettoluide that separated was filtered off and fractionally
crystallized as follows :

The entire mass was dissolved in boiling water and the
liquid filtered through paper and allowed to come to the

' Janson states that at this point he was able to obtain crystals.

2 4-Chlor-2-toluidine boils at 237° (Goldschmidt and Honig: Ber. d. chem. Ges..
19, 2440 (1886)), and 6-chlor-2-toluidine boils at 245° (.Wynne and Greeves: Centralbl..
1895, II. 530).

On Chloriniidoqiiinones 453

room temperature. The crystals that had formed were re-
moved and further fractionated by treatment of their warm
chloroform solution with one volume of ligroin (6o°-8o°).
Three crystallizations in this manner gave a fraction that
melted sharply at 154°.^ The aqueous filtrate that had been
set aside meanwhile was now placed in the refrigerator and
allowed to stand overnight. The solid that formed was re-
moved and crystallized from hot water, and gave colorless
needles that melted at i3o°-i3i°.^ When these were dis-
solved in hot chloroform and the solution was mixed with
enough ligroin (6o°-8o°) to cause slight precipitation, warmed
until solution occurred and then set aside to cool, the crys-
tals that formed melted at 136°-! 37°.

In order to determine whether this low-melting acettoluide
is identical with Goldschmidt and Honig's^ product, some of
the latter was prepared according to the directions given by
these chemists, the start being made with pure 4-chlortoluene.
When the acettoluide was crystallized once from water, it
melted at 130°-! 31°, as stated in the literature, but further
crystallization from chloroform and ligroin, as specified above,
raised the melting point to 136°-! 3 7°, and a mixture of this
product with the low-melting compound obtained from Jan-
son's oil melted at the same temperature as either of them
separately, which proves that they are identical. A mix-
ture of either of these with the high-melting acettoluide, melt-
ing at 154°, softens at 110°, and is entirely melted at 123°.
From this it follows that chlorination of 2 -nitro toluene gives
both 4-chlor-2-nitrotoluene and 6-chlor-2-nitrotoluene.

Electrolytic Reduction of the Mixture of Chlornitrotoluenes . —
Ten grams of the oil boiling at 236°-238°, obtained by chlori-
nating o-nitrotoluene, was dissolved in 100 grams of sulphuric
acid (1.84), and this solution diluted with 2 cc. water.
Through this a current of 7.5 volts and 2 . 2 amperes was
passed for ten hours, at the end of which time the liquid
was poured into 360 cc. water and the mixture set aside to

1 Wynne and Greeves {loc. cil.) and Janson {loc. cit.) found the melting point
of the acettoluide from o-chlor-o-toluidine to be 157°-159°.

2 Goldschmidt and Honig: Loc. cil.

3 Ibid.

454 Raijord

cool. It was next filtered, and the filtrate, which was deep
puqDle in color, was cooled to about o°, and 25 grams of pow-
dered sodium dichromate was slowly added to oxidize the
aminophenols produced to their corresponding quinones.
The resulting mixture was allowed to come to room tempera-
ture during an hour, and was then distilled with steam. The
yellow solid that passed over was dried and found to melt
at 48°-5i°, which suggested a mixture of the p- and o-chlor-
toluquinones, corresponding to the positions of chlorine al-
ready indicated by the acettoluides described.

A portion of this mixture was fractionated by crystalliza-
tion from ligrom (40°-6o°), and there was obtained a frac-
tion that separated in the form of yellow prisms melting at
105°, and was found to be identical with 4-chlortoluquinone,
a mixture of the two still melting at 105°. When mixed with
2-chlortoluquinone, melting at 55°, the melting point was
depressed exactly as stated on page 450. The combined
mother liquors were allowed to evaporate, but the crystals
obtained from them did not melt sharply, and repeated crys-
tallization did not change this.

A second portion of the dried crude quinone, melting at
48°-5i°, mentioned above, was fractionally sublimed^ at a
tem-perature of 45^-50° and a pressure of 15-17 mm. The
process was very slow, and complete separation was not se-
cured. The sublimate melted at 48°-5i° and behaved like
2-chlortoluquinone containing a trace of the isomeric 4-chlor-
toluquinone. The sublimate was next mixed with a little
water, and saturation of this with sulphur dioxide gave a
hydroquinone which, after one subsequent crystallization
from water, melted at 173° and did not depress the melting
point of 2-chlortoluhydroquinone, melting at 173°, described
on page 450. A mixture of this new product with 4-chlor-
toluhydroquinone, melting at 176°, melts at i47°-i5i°, show-
ing that the new product is not identical with 4-chlortolu-
hydroquinone.

A third portion of the dried crude quinone was reduced to
hydroquinone, and gave crystals melting at i48°-i52°, indi-

» Kempf: J. prakt. Chem., [2] 78, 203 (1909).

On Chlorimidoquinones 455

eating the presence of a mixture. This was dissolved in hot
benzene,* the liquid filtered through cotton, the filtrate al-
lowed to cool to the room temperature, and the cr}''stals im-
mediately removed. Recrystallization in this way three
times and final crystallization from water gave a product
melting at 174°-! 75°. A mixture of this with 2-chlortolu-
hydroquinone, melting at 173°, melted at i73°-i75°, while
with 4-chlortoluhydroquinone the melting point was de-
pressed to i47°-i5i°.

It is thus clear that the mixture of 2 -chlor-6-nitro toluene
and 4-chlor-6-nitrotoluene obtained by chlorinating o-nitro-
toluene, whose structures were proved on the preceding pages,
gives by electrolytic reduction and rearrangement in sulphuric
acid solution the corresponding pair of aminophenols, viz.,
i-methyl-2-chlor-6-aminophenol and i-methyl-4-chlor-6-amino-
phenol, which in turn give by oxidation the two chlortolu-
quinones, viz., /)-chlortoluquinone, melting at 105°, and
o-chlortoluquinone, melting at 55°. The latter is identical
with the chlortoluquinone, melting at 55°, obtained by oxi-
dizing the chloraminocresol, melting at i66°-i67°, produced
by the action of chlorine on /)-nitro-m-cresol, and subsequent
reduction. It is thus shown that the chlorine atom enters
into the ortho position to the methyl group in p-mtro-m-
cresol and that it has that position in all the derivatives ob-
tained from the chlorinated nitro-m-cresol, viz., the chlor-
aminocresol, melting at 1 66°- 167°,' and the chlorimido-
chlortoluquinone, melting at 87°.

SUMMARY

The most important results of the work described in this
paper may be summarized as follows :

I. (a) When />-nitro-w-cresol is chlorinated by passing
chlorine, diluted with carbon dioxide, into a glacial acetic

1 It had previously been noted that these two compounds differed in the rapidity
withjwhich crystals were deposited from benzene solution, 2-chlortoluhydroquinone
separating more quickly than its isomer.

2 The work will be continued, and attempts made to isolate directly the chlor-
aminocresols produced by the electrolytic reduction in sulphiuic acid solution as de-
scribed on page 445.

456 Wheeler, Nicolet and Johnson

acid solution of the nitro compound, chlorine takes a posi-
tion adjacent to methyl.^

{b) o-Chlortoluquinone, the last of the three possible chlor-
toluquinones, was prepared by oxidation of the aminocresol
obtained by reducing the chlorinated nitrocresol.

2. When o-nitro toluene is chlorinated in the presence of
iron, a mixture of the o- and the />-chlor-o-nitrotoluenes re-
sults.

3. When 2,4,6-tribrom-m-cresol is nitrated under certain
conditions it gives rise to two isomeric mononitro derivatives,
viz., 2,6-dibrom-4-nitro-m-cresol and 2,4-dibrom-6-nitro-w-
cresol, while 2,4,6-trichlor-m-cresol could not be nitrated under
any conditions employed in this work.

4. The chlorimidoquinones obtained as end products in
these experiments do not appear to exist in stereoisomeric
forms. Chlorimidochlortoluquinone was obtained in three
forms, melting at 91°, 87°, and 65°, respectively, but they
were proved to be structural isomers and not stereoisomers.

In conclusion, I hereby tender my sincere thanks to Pro-
fessor Stieglitz for the painstaking scrutiny with which he
has guided this work.

[Contributions from the Sheffield Laboratory of Yale University]

CXCIV.— ON HYDANTOINS

THE ACTION OF ACYLTHIONCARBAMATES, ACYLDI-

THIOCARBAMATES AND ACYLIMIDODITHIO-

CARBONATES ON a-AMINO ACIDS

2-THIOHYDANTOIN

By Hbnry L. Wheeler, Ben H. Nicolet and Treat B. Johnson

[sixth paper]

It is known that the alkyl isocyanates and alkyl isothio-
cyanates (mustard oils) combine with a-amino acids, forming
hydantoic acids, (I) and (III), which then undergo inner con-

' Taking Kehrmann's results, as well as my own, into consideration, one would
be inclined to conclude that mixtures of this compound and its isomer, with the chlor-
ine atom in a position para to the methyl group, are formed. In this work the forma-
tion of the ortho derivatives is conclusively proved.

On Hydantoins

457

densations, giving the corresponding i-alkylhydantoins^ (II)
and i-alkylthiohydantoins^ (IV), respectively. The action
of acyl isocyanates, RCONCO, and the corresponding acyl
isothiocyanates, RCONCS, on amino acids has not been in-
vestigated. The specific aim of the work described in this
paper was to synthesize some N-acyl derivatives of hydan-
toic and thiohydantoic acids, (V) and (VI), and investigate
their behavior on hydrolysis and also the reactivity of their
methylene hydrogens, =N — CHj — CO, towards aldehydes.
We shall describe also some interesting derivatives of the
pseudo forms of these acids (VII). Hydantoic acid deriva-
tives of these types have not hitherto been described.

RNH COOH

I
CO

1
NH— CHR
I

RNH COOH

NH— CHR
III

(R = C^H^, CH3, etc.)

+ H3O

CHR

RCONHCSNHCH2COOH

RCONHC

OH(SH)

^NCHjCOOH
VII

The action of amines on the four classes of acylcarbamates,
(X), (XII), (XIV), and (XV), has been investigated in this
laboratory. Ethyl acetylcarbamate (acetylurethane) , which

' Kuhn: Ber. d. chem. Ges.. 27, 2880. Paal: Ibid., 27, 975 (1894). Mouneyrat:
Ibid.. 33, 2393! Fischer: Ibid.. 33, 2370 (1900). Fischer and Mouneyrat: Ibid.. 33,
2383. Neuberg and Manasse: Ibid., 38, 2359. Neuberg and Rosenberg: Biochem.
Z., 6, 459.

^^ 2 Aschan: Ber. d. chem. Ges., 17, 420 (1884). Marckwald, Neumark and Stelz-
ner: Ibid., 24, 3278 (1891). Wheeler and Brautlecht: This Journal, 45, 446 (1911).
Brautlecht: J. Biol. Chem., 10, 139.

458 Wheeler, Nicolet and Johnson

was thoroughly investigated by Young and Clark/ reacts
with bases, giving as chief products normal ureas. Ethyl
benzoylcarbamate (VIII) reacts in a similar manner with
aniline, for example, giving benzoylphenylurea (IX) and
alcohol.^ Acylthiolcarbamates (XIV) and acyldithiocarb-

CeHgCONHCOOCjHs + CoH^NH^ =
VIII

CcHjCONHCONHCeHs + C^HjOH

IX
amates (XV) react with bases in a manner analogous to that
of the oxygen derivatives, forming acylureas and acylthio-
ureas, respectively, with evolution of mercaptans.^ The
acylthioncarbamates (XII), on the other hand, react in an
abnormal manner with bases, forming addition products
which break down with evolution of h5^drogen sulphide, giv-
ing acylpseudoureas.^

Of the three classes of acylimidocarbonates, (XI), (XIII)
and (XVI), only the acylimidothio- and acylimidodithiocar-
bonates, (XIII) and (XVI), have been studied. They both
react smoothly with amines. The monothiocarbonates give
the same acylpseudoureas as are obtained by the action of
bases on the acylthioncarbamates^ (XII). The acylimidodi-
thiocarbonates react with bases, with evolution of mercap-
tans, forming acylpseudothioureas.^ Acylimidocarbonates,
represented by formula (XI), have not been synthesized.^

/OC3H5
RCONHCOOC2H5 — > RCON : C<

\OC2H,
X XI

RCONHCSOC2H5 __ „^ „

XII ^ /^^2^5

RCON : C<

RC0NHC0SC,H3 •--- \sc,H,

XIV ^ ^

XIII
J J. Chem. vSoc, 73, 361 (1898).

2 Wheeler and Merriam: J. Am. Chem. Soc. 23, 289 (1901).

3 Wheeler and Merriam: Loc. cii.

* wheeler and Johnson: This Journal, 24, 190 (1900); 27, 218 (1902).
' Wheeler and Johnson: Loc. cii.

* Wheeler and Johnson: This Journal, 26, 408.

^ The writer desires to call attention to the fact that Mr. Lewis H. Chernoff is
now working on a method of synthesizing these compounds. The results will be pub-
lished later. — T. B. Johnson.

On Hydantoins 459

RCONHCSSC2H5 -^ RCON : C<

XV XVI

(R = C„H„ CH3. etc.)

'^ P P P

• go o 05.
Ss C O Ok

•^g :z; 'z iz; § 5

• V- o o o 3- o

.„ C/5 Oi CO s !z^

SI Q O Q I ffi

K ffi K "■ g.

000

^SC,H,

n I f -I ■ I

s. 5 X o o ss o

^ « Kg

I o o o 11 ^

- a a: k > w g

S o b f o
I z z 2 I ?

I A A A f A

ri^ 02, 02, olg 2
- Q o o IP

"8 8 8 18

en j„

S a

460 Wheeler, Nicolet and Johnson

The action of amino acids or their esters on acylcarbamates
and acylimidocarbonates (sulphur and oxygen compounds)
has not been investigated. We have now examined the be-
havior of glycocoU and its ethyl ester towards acylthion-
carbamates (XII), acyldithiocarbamates (XV) and acylimido-
dithiocarbonates (XVI). The acylthioncarbamates reacted
smoothly with this amino acid, in the presence of alkali, and
with its ester in the same manner as with amines, forming
alkyl derivatives of acylpseudohydantoic acids. The thion-
carbamates which were used in our work and the pseudo-
hydantoic acid derivatives prepared from them are tabulated
above for inspection.

Benzoylpseudoethylhydantoic acid and its ethyl ester
were converted into the normal benzoylhydantoic acid
(XVII) when heated with dilute hydrochloric acid. If,
however, concentrated hydrochloric acid was used this ben-
zoylhydantoic acid also underwent hydrolysis, forming ben-
zoic acid and hydantoin (XVIII).

CeH3CON : C<

/ ^NHCH2COOH(C2H5)

CeH.CONHCONHCH^COOH — > NH— CO— NH— CO— CH2

I 1

XVII XVIII

The acylthiohydantoic acids (VI) and their esters theo-
retically should be formed by the action of acyl isothiocyan-
ates on amino acids and their esters, respectively:

RCONCvS + NHjCHjCOOH = RCONHCSNHCH^COOH

We did not, however, employ these rhodanides for the prepara-
tion of this new class of compounds. We found that they were
formed smoothly by the action of the potassium salts and
esters of amino acids, respectively, on the acyldithiocarbamates.
The representatives of this new class of compounds, which
were prepared by the action of glycocoll and alanine on di-
thiocarbamates, are represented in the following table:

On Hydantoins 461

n n n

0 0

a ic K

0 n 0

8 si
III

000

12; :z; 5;

ffi ffi W

000

W XJi Xfi

'^ s f^

ffi ffi ffi

.?|i

^ ^2; :^

^ ^-^

K ffi K

ffi Kb

000

a 5= K

000

ffi 0 0

ffi

5^: 0 0
8 ^

n 0 0

^ 0

ffi tU ffi
000
000

2 =;»
0 SI

8 8i

ffi 0

2 ^ ^

:z;

ffi ffi s
000

www
^ 5: ^

ffi ffi ffi

ffi

n 0 0

ffi K ffi

000

S88
0 KP

8 "^

^ 0

ffi

KJ >-(

^_^

►0 0

Ln 4^

00 g

^° j.°

i°b£

.0 Jh 5-g-

•^ M 0

0 K) ''S

w vj Ol

10 \o

Vv^hen these acylthiohydantoic acids or their esters (VI)
were digested with hydrochloric acid they behaved like the
corresponding oxygen acids and were transformed into cyclic
compounds. For example, we found that benzoyl- and acetyl-

462 Wheeler, Nicolet and Johnson

thiohydantoic acids (XIX) underwent hydrolysis under these
conditions, and that the unknown 2-thiohydantoin (XX)
was formed in both cases. i-Acetyl-4-methyl thiohydantoic
acid (XXI) underwent a similar transformation, giving 2-thio-
4-methylhydantoin (XXII) :

(CH3CO)

CeHjCONH COOH(C2H5) NH CO

QH^COOH + ^^

NH CHj NH

XIX XX

CH3CONH COOH NH CO

\ \ 1

CS —^ CH3C00H + cs

NH CHCH, NH CHCH,.

XXI XXII

These nitrogen-unsubstituted thiohydantoins (XX) and
(XXII) represent new types of thiohydantoins. Their forma-
tion, in this manner, from the acyl thiohydantoic acids is re-
markable since the plain ethyl thiohydantoate,

NH2CSNHCH2COOC2H5

undergoes no condensation under similar conditions, while
hydantoic acid and its esters condense quantitatively to
hydantoin.^ Harries and Weiss write as follows regarding
the behavior of the thiohydantoic ester:

"Ein VersuchdenThiohydantoinsaureester durch Schmelzen
oder Behandeln mit Salzsaure in das neutrale Thiohydantoin
umzuwandeln fiihrte zu keinem positiven Hrfolge. "

These investigators were also unable to introduce a sul-
phur atom into hydantoin by heating it with phosphorus
trisulpuide. They also heated hydantoin with ammonium
sulphide, a reagent which was used successfully by Fischer^
for introducing sulphur in the purine series, but it was com-

J Harries and Weiss: Ann. Chem. (Liebig), 327, 355 (1903); Ber. d. chem. Ges.,
33, 3418 (1900). Bailey: This Journal. 28, 386.
2 Ann. Chem. (Liebig), 288, 159.

On Hydantoins 463

pletely decomposed under these conditions and glycocoll
was formed. The fact that the benzoyl- and acetyl thiohy-
dantoic esters condense to 2-thiohydantoin when heated with
hydrochloric acid, while ethyl thiohydantoate undergoes no
condensation under the same conditions, indicates that the acyl
derivatives first condense, giving as intermediate products
acylthiohydantoins. These, being unstable in the presence
of hydrochloric acid, then undergo hydrolysis, giving Uie
plain 2-thiohydantoins.

Diethyl benzoylimidodithiocarbonate^ reacted normally with
aminoacetic acid and its ethyl ester, giving the alkyl deriva-
tives of the pseudo forms of benzoylthiohydantoic acid and
ethyl benzoylthiohydantoate, respectively (Table III).

The acylhydantoic and acylthiohydantoic acids, (V) and
(VI), may be viewed as acylcarbamyl- and ac3dthioncarbamyl
derivatives of glycocoll corresponding to the benzoyl deriv-
ative of tliis amino acid, or hippuric acid. Since they all
contain the grouping — NH.CHj.CO — it v^^as therefore of
especial interest to examine their beha^dor towards alde-
hydes. PlochP was the first to show that hippuric acid con-
denses with aldehydes, but it was Erlenmeyer^ who correctly
explained the nature of the condensation products.

We have now made the interesting observation that ben-
zoyl- and acetyl thiohydantoic acids condense readily v^ith
benzaldehyde when boiled in acetic acid solution and in pres-
ence of sodium acetate and acetic anhydride, forming ben-
zoyl- and acetylbenzaltliiohydantoins, (XXIII) and (XXV),
respectively. On the other hand, under the same conditions,
we obtained no condensation products with the esters of these
two acids. An attempt to condense benzoylhydantoic acid
(XVII) with benzaldehyde was also unsuccessful. This latter
observation is of interest because it again illustrates the
greater tendency of the sulphur compounds to condense, as
was observed by Wheeler and Brautlecht^ in the case of the
phenylthiohydantoins.

1 Wheeler and Johnson: Loc. cit.

2 Ber. d. chem. Ges., 16, 2815 (1883).

3 Ann. Chem. (Liebig), 271, 137; 276, 1. Erlenmeyer and Stadlin: Ibid.. 337^
283, 265. Erlenmeyer and Arbenz: Ibid., 337, 302.

* This Journal, 45, 446 (1911).

4^4 Wheeler, Nicolet and Johnson

+ w

^„ U

< ? to

o o

^ ffi

o 5 o o

o o

§ .2 ^ O

8 5 ••

:2 O O

a u o

I A ^

^ U (J

When benzoylbenzalthiohydantoin (XXIII) was warmed
with alkah the benzoyl group was removed and benzalthio-
hydantoin (XXIV) was formed. This same benzal compound
was also obtamed by condensation of thiohydantoin (XX)
with benzaldehyde. Ruhemann and Stapleton^ have assigned

» J. Chem. Soc. 77, 246.

On Hydantoins

465

this same structure to a compound obtained by them by-
condensation of thiourea with phenylpropiolic acid (XXVI).
Our compound agreed apparently with their product in all
its properties except the melting point. They give the melt-
ing point as 280° on slow heating or 300° on heating rapidly.
Our compound, which was made by two methods, melted in
both cases at 258° when heated in a capillary tube slowly
or rapidly.

M I

•S

o
o

^ I

< o-

o
o

^A-

o
+

466 Wheeler, Nicolet and Johnson

EXPERIMENTAL PART

Ethyl Benzoylpseudoethylhydantoate,
QHsCON :C(OC3H5)NH.CH2COOC2H5.— Ethyl aminoacetate
and ethyl benzoyl thioncarbamate* react slowly at ordinary
temperature with evolution of hydrogen sulphide. When
equivalent quantities of the two esters were heated on the
steam bath, a condenser being used to prevent loss of the
acetate, the reaction was complete witnin 15 minutes and a
65-70 per cent, yield of the hj^dantoate was obtained. The
compound crystallizes from 95 per cent, alcohol in slender,
colorless, rectangular plates, which melt at 79^-80° to a clear
oil and decompose above 200°. The ester is very soluble in
boiling alcohol, moderately soluble in cold and soluble in hot
water. Analysis (Kjeldahl) :

Calculated for

C14H18O4N2 Found

N 10.05 9 90

Ethyl Benzoylpseudomethylhydantoate,
CgHsCON : C(OCH3)NHCH2COOC2H5.— From ethyl aminoace-
tate and methyl benzoyl thioncarbamate.' It forms colorless,
transparent prisms melting at 103°. Analyses (Kjeldahl):

Calculated for Found

CisH.eO^Nz I II

N 10.60 10.77 10.63

BcnzoylpseiAdoethylhydantoic A cid,
CeH5CON:C(OC3H5)NHCH3COOH.— One and nine-tenths
grams of aminoacetic acid were dissolved in 10 cc. of water
containing i . 5 grams of potassium hydroxide, and 5 . 3 grams
of ethyl benzo34thioncarbamate, in 10 cc. of alcohol, was
added to the solution. There was an immediate reaction with
evolution of hydrogen sulphide. After heating 2 hours on
the steam bath the solution was then concentrated to 10 cc,
cooled and acidified with acetic acid. The hydantoic acid
separated and was purified by crystallization from alcohol.
It separated in needles which melted at 161° to a turbid oil

' Wheeler and Johnson: J.oc. cit.
2 Miqucl: Loc. cit.

On Hydantoins 467

which finally became clear at 203°. While this behavior on
melting indicated a mixture, the nitrogen determinations
(Kjeldahl) agreed with the calculated value for the hydantoic
acid:

Calculated for Found

C,2Hu04N2 I II

N 11.20 11.20 II. 19

Benzoylhydantoic Acid, CeHsCONHCONHCH^COOH.— This
acid was prepared by hydrolysis of benzoylpseudoethylhy-
dantoic acid or ethyl benzoylpseudomethylhydantoate with
hydrochloric acid. There was some difficulty in choosing
conditions most favorable for the change since the benzoyl
group is removed by too long hydrolysis. The best yields,
about 60 per cent., were obtained by digesting the acid for
4 to 5 hours with equal parts of water and concentrated hy-
drochloric acid. Ethyl benzoylpseudomethylhydantoate un-
derwent only partial conversion into the hydantoic acid after,
digestion with hydrochloric acid of the above concentration
for 30 minutes. During the digestion a gas was evolved
which burned with a green flame and was identified as methyl
chloride. The hydantoic acid crystallizes in plates which
melt at 253^-254° with effervescence. It is difficultly soluble
in hot alcohol, practically insoluble in cold alcohol and water.
Analysis (Kjeldahl) :

Calculated for
C10H10O4N2 Found

N 12.15 12.28

Conversion of the Hydantoic Acid into Hydantoin. — Two
grams of the benzoylhydantoic acid were suspended in 100
cc. of concentrated hydrochloric acid, 40 cc. of water
added and the solution evaporated to dryness on the
steam bath. One and two-tenths grams of the acid
were recovered unaltered, and in the alcoholic extract of the
residue only benzoic acid and hydantoin were identified,
No benzoylhydantoin was detected.

Attempt to Condense Benzoylhydantoic Acid with Benzaldehyde.
— The hydantoic acid was recovered unaltered after long
digestion with benzaldehyde in glacial acetic acid solution in

468 Wheeler, NicolH and Johnson

the presence of acetic anhydride and anhydrous sodium
acetate. Analysis (Kjeldahl) :

Calculated for

C,oHio04N2 Found

N 12.15 12.4

Ethyl Benzoylthiohydantoate, CeHjCONHCSNHCHjCOOCjHs.
— When ethyl benzoyldithiocarbamate and ethyl aminoace-
tate were mixed in molecular proportions there was an imme-
diate reaction with evolution of mercaptan. The reaction
was complete after heating on the steam bath for about 15
minutes. We obtained a crystalline mass which was dis-
solved in alcohol, digested for about 4 hours and then cooled,
when the ester separated in needles melting at i28°-i29°
to a pale yellow oil. The compound is very soluble in boil-
ing alcohol, moderately soluble in cold and slightly soluble
in hot water. Analysis (Kjeldahl) :

Calculated for
Ci2Hi«03N2S Found

N 10.51 . 10.35

Benzoylthiohyddntoic Acid, C0H5CONHCSNHCH2COOH.—
Five grams of aminoacetic acid and 3.8 grams of potassium
hydroxide were dissolved in 30 cc. of water, and 15 grams of
ethyl benzoyldithiocarbamate (i mol.) in 30 cc. of alcohol
added to the aqueous solution. After 8 hours' digestion on
the steam bath the solution was evaporated to dryness and
the residue digested with 600 cc. of water. After cooling
the solution and acidifying with hydrochloric acid the above
hydantoic acid separated in the form of needles. It ciystal-
lizes from alcohol in needles and from \\ater in plates which
melt at 202°. The acid is very soluble in hot alcohol, mod-
erately soluble in cold and difficultly soluble in hot water.
The yield was about 75 per cent, of the calculated. Analysis
(Kjeldahl) :

Calculated for
C,oH,o03N2S Found

N 11.75 II. 41

On Hydantoins 469

NH CO

2-Thiohydantoin, CS I . — This thiohydantoin can be

NH CH2

prepared either from benzoylthiohydantoic acid or acetyl-
thiohydantoic acid (see below). Benzoylthiohydantoic acid
was digested for 10-12 hours with about five tinies its
weight of concentrated hydrochloric acid. The solution
wash ten allowed to evaporate to dryness and sufficient alco-
hol added to dissolve the residue on heating. On cooling,
the thiohydantoin separated in yellow prisms which decom-
posed slowly above 200° and finally melted at 227°. This
compound is readily soluble in warm alcohol and water and
moderately soluble in cold. Analysis (Kjeldahl) :

Calculated for

Found

C3H4ON2S

II III

24. 12

23.68

24.04 24.1

Acetylthiohydantoic acid or its eth}^ ester are botri con-
verted into this thiohydantoin by digestion with hydrochloric
acid. The method finally used for making the thiohydantoin
in large quantities was as follows: The ethyl acetylthiohy-
dantoate was prepared as described below and the solution
evaporated to dryness. The residue was then digested several
hours with concentrated hydrochloric acid and again evapo-
rated to dryness and the thiohydantoin then separated from
potassium chloride by extraction with alcohol. The yield
was 62 per cent, of the calculated.
Condensation of Benzoylthiohydantoic Acid with Benzaldehyde

i-Benzoyl-4-henzalthiohydantoin,
CjHjCON CO

-Five grams of benzoyl thiohydan-

NH — C : CHC.H,
toic acid and 3 grams of benzaldehyde were dissolved in 25
cc. of glacial acetic acid, and 3 . 5 grams of fused sodium acetate
and 4.3 grams of acetic anhydride added to the solution.
The mixture was then heated to boiling, in an oil bath, for

470 Wheeler, Nicolet and Johnson

45 minutes. The solution assumed, under these conditions,
a blood-red color. On cooling and adding water, 2 . 2 grams
of the benzoyl thiohydan to in were precipitated, and after con-
centrating the filtrate more was obtained. The yield was
3.6 grams, corresponding to 56 per cent, of the calculated.
The hydantoin is moderately soluble in boiling alcohol and
separates in rectangular plates which melt at 181° to a red
oil and then decompose when heated above 260°. Its be-
havior on melting indicated a ring structure and not that of
a hydantoic acid, as these generally melt with effervescence.
It is insoluble in water. Analysis (Kjeldahl) :

Calculated for
C,7H,202N2S C,7Hj403Nj!S Pound

N 9.10 8.59 9.07

An attempt to condense benzaldehyde with ethyl benzoyl-
thiohydantoate, under the same conditions as described in
this experiment, was unsuccessful. The ester was recovered
unaltered.

NH CO

4-Benzalthiohydantoin, CS

. — This com-

NH C : CHQHj

pound was formed when the corresponding benzoylhydantoin,
described in the preceding experiment, was dissolved in cold
10 per cent, potassium hydroxide solution. On neutralizing
the alkaline solution with acetic acid the benzalthiohydantoin
separated as an oil which soon crystallized in the form of
microscopic, yellow needles. It crystallizes from alcohol in
yellow needles melting at 258° with slight decomposition.
Analysis (Kjeldahl) :

Calculated for
CioHsONzS Found

N 13-70 13-57

Ruhemann and Stapleton^ state that this compound melts
at 280° on slow heating and 300° on heating rapidly. Our
product melted at 258° when heated rapidly, and also at this

On Hydantoins 471

same temperature after being held at 240 ^-250° for three
minutes before raising the temperature to that of its melting
point. The same benzalthiohydantoin was also obtained by-
condensation of thiohydantoin with benzaldehyde in glacial
acetic acid solution and in the presence of sodium acetate.
It melted at 258° and a mixture of this benzal compound
with some prepared from the benzoylthiohydantoin melted at
exactly the same temperature.

Benzalthiohydantoin is very stable in the presence of alkali.
One-half a gram was dissolved in 12 per cent, potassium hy-
droxide solution and this then heated at 100° for one hour.
The solution was then cooled and acidified with hydrochloric
acid, when the unaltered hydantoin deposited. It melted at
258°.

Ethyl Benzoylpseudoethylthiohydantoate,
CeHjCON : C(SC2H5)NHCH,COOC2H5.— This ester was pre-
pared by the action of ethyl aminoacetate on diethyl ben-»
zoylimidodithiocarbonate.^ They reacted at ordinary tem-
perature with evolution of mercaptan and after standing over-
night the mixture completely solidified. The yield was prac-
tically quantitative. The ester is very soluble in alcohol
and separates from saturated solutions, on cooling, in flakes
melting at 77°-78° to a clear oil. Analysis (Kjeldahl) :

Calculated for
CmHisO^NzS

Found

9-53

9.26

This ester does not add methyl iodide. Some of the com-
pound was dissolved in an excess of methyl iodide and the
solution allowed to stand for 4 days. After evaporation of
the iodide the hydantoate was recovered unaltered and melt-
ing at 77°. There was no evidence of any reaction.

Benzoylpseudoethylthiohydantoic A cid,
QHsCON : C(SC2H5)NHCH2COOH.— From aminoacetic acid
and diethyl benzoylimidodithiocarbonate. The manipula-
tion was the same as in the condensation of glycocoU with
ethyl benzoyldithiocarbamate (above). After heating on
the steam bath for 7 hours the mixture was concentrated to

' Wheeler and Merriam; Loc. cit.

472 Wheeler, Nicolet and Johnson

lo cc, washed with ether to remove any unaltered ester and
the hydantoic acid then precipitated by addition of acetic
acid. The acid was colored violet but on recrj-stallization
from boiling alcohol it separated in clusters of colorless needles
which melted at 198°. The yield was about 65 per cent, of the
calculated. The acid is readily soluble in boiling alcohol,
moderately soluble in cold, and difficultly soluble in water.
The acid undergoes slight decomposition on prolonged heat-
ing at 100°. Analyses (Kjeldahl) :

Calculated for Found

CiaHuOsNjS I II III

N 10.5 9.72 9.98 9.88

Aceiylthiohydantoic Acid, CHgCONHCSNHCHjCOOH.— This
acid was prepared by the action of aminoacetic acid on ethyl
acetyldithiocarbamate. The procedure was the same as in
the preparation of benzoylthiohydantoic acid. After heat-
ing for 22 hours the solution was concentrated and acidified
with acetic acid. On cooling, the potassium salt of the hy-
dantoic acid separated in colorless prisms which melted at
225^-227° with effervescence. Analysis (Kjeldahl) :

Calculated for
C5H7O3N2SK

Pound

13.08

13-52

In order to obtain the free acid the potassium salt was dis-
solved in warm, dilute hydrochloric acid. On cooling, the
acid separated in slender, colorless needles which melted at
205° with effervescence. The acid is very soluble in hot and
moderately soluble in cold 'water. When heated in the pres-
ence of concentrated hydrochloric acid it is converted into
thiohydantoin.

i-Acetyl-2'thio-4-henzalhydantoin,
CH3CON CO

CS I . — Acetylthiohydantoic acid was dis-

NH— C : CHCeHs
solved in 5 times its weight of glacial acetic acid, and i . 2 and
2.0 molecular proportions of benzaldehyde and sodium ace-

On Hydantoins 473

tate, respectively, added to the solution. After heating one
hour a portion of the mixture was diluted with water. No
precipitate resulted. Two molecular proportions of acetic an-
hydride were then added and the solution digested for an-
other hour. It had assumed a deep red color and on adding
water the benzalhydantoin immediately separated. The
presence of acetic anhydride is apparently necessary, in this
case, to effect a condensation. This acetylhydantoin is read-
ily soluble in boiling alcohol and insoluble in water. It crys-
tallizes from alcohol in light yellow prisms which melt at 231°
to a clear red oil. Analysis (Kjeldahl) :

Calculated for
C12H10O2N2S Found

N 11-37 11.42

An attempt to condense ethyl acetylthiohydantoate (see
below) with benzaldehyde under the above conditions was
unsuccessful.

Ethyl Acetylthiohydantoate, CH3CONHCSNHCH2COOC2H5.—
This was prepared by warming, in alcoholic solution, ethyl
aminoacetate with ethyl acetyldithiocarbamate. It crystal-
lizes in hexagonal prisms which melt at 104°- 105° to a clear
colorless oil. Analysis (Kjeldahl) :

Calculated for
C7H,203N:iS Found

N 13.71 1364

I- A cctyl-4-methylthiohydantoic A eid,
CH3CONHCSNHCH(CH3)COOH.— From alanine and ethyl
acetyldithiocarbamate. Alanine is apparently less reactive
tlian glycocoU and it was necessary to heat with the carbamate
for 40 hours before the reaction was complete. The acid is
very soluble in hot and cold water. It crystallizes in prisms
which melt at 171° to a clear colorless oil, effervescing when
heated higher. Analysis (Kjeldahl) :

Calculated for
CsHioOsNyS Found

N 14.72 14.36

474 Kohler

NH CO

1 I

2-Thio-4-methylh}dantoin, CS . — i - Acetyl - 4 -

NH CHCHg

methylthiohydantoic acid was heated with concentrated
hydrochloric acid on the steam bath for several hours. After
concentration of the solution the methylthiohydantoin was
obtained and crystallized from alcohol in flat prisms. The
compound melts at 158°- 159° to a clear, colorless oil. It is
very soluble in boiling alcohol and water. Analysis (Kjeldahl) .

Nbw Haven, Conn.
July 22. 1911

Calculated for
C4HPON2S

21.53

Found
21.31

UNSATURATED ^-KETONIC ACIDS

By E. p. Kohler

The main object of the following investigation was to study
the mode of addition to the conjugated system, C : C.C : O,
in cases in which both the addend and the unsaturated chain
are parts of the same molecule. Addition in these circum-
stances necessarily results in some kind of ring formation,
hence lactone formation, which generally takes place spon-
taneously, seemed best adapted for such an investigation.
Unsaturated (?-ketonic acids were selected for study because
they present the largest number of possibilities, since the lac-
tone ring can close in the /?, ;-, or d positions.

The ketonic acids that were used have essentially the same
chain that is responsible for the interesting properties of the
glutaconic acids. In the course of the investigation it be-
came necessary to compare the two classes of substances
and it was found that the properties of the former are quite
as remarkable as those of the latter. This part of the work
will, however, be reserved for later publication, and the pres-
ent paper will deal only with the results obtained in the study
of lactone formation.

Unsaturated ^-lactonic acids are unknown; but they can be

Unsaturated d-Ketonic Acids 475

prepared without very great diflficulty either by means of the
methods used for getting the corresponding saturated com-
pounds, or by introducing bromine into saturated ketonic
acids and eliminating hydrobromic acid from the product.
Thus when a-brombenzalacetophenone is added to methyl
sodiummalonate suspended in ether, one of the products is
an unsaturated ^-ketonic ester represented by one of the two
following formulas :

C.H.C : CHCOaH, CeH.C . CHXOCeHj

I II

CHCCOjCHj), CCCOjCHa)^

I II

The formation of the ester from a-brombenzalacetophe-
none, and its behavior on hydrolysis, are most easily inter-
preted on the assumption that the ethylene linkage is in the
position represented by formula I. I shall therefore use this
formula; but in view of the ease with which substances of
this type undergo isomeric change, and the diflficulty of defi-
nitely locating the double linkage, this choice must be regarded
as provisional.

The ester is rapidly hydrolyzed by alcoholic potassium
hydroxide. In the presence of excess of base the product
is the expected dipotassium salt, but lactone formation takes
place so easily that the corresponding acid cannot be ob-
tained. The dipotassium salt loses base when dissolved in
water, and passes into a salt of a monobasic acid that has the
composition CigHi^Og. For this lactonic acid three formulas
are possible, depending upon the way in which the lactone
ring is established.

C^H^CH CHCOC.H5 C,H5CCH2COC,Hj

I .\

o o

I \

HOjCCH CO HO2C— CH— CO

III IV

C.H.C : CHC(OH)C,H,

I
HO2CCH— CO
V

476 Kohler

Above 170°, the acid rapidly loses carbon dioxide and passes,
in the main, into two products: a saturated lactone melt-
ing at 93° and having the composition CiyHiPg, and an un-
saturated acid isomeric with the lactone. The structure
of this unsaturated acid can be deduced from the following
transformations: When a solution of the acid in methyl
alcohol is saturated with hydrogen chloride, it passes into
the methyl ester of a chlor acid that on reduction gives methyl
/--benzyl'/J-phenylbutyrate. This shows that no shifting of
the groups occurred in the process of heating — the chlorine
derivative is methyl benzoylchlorphenylbutyrate (VI), and
the acid from which it was obtained must be either benzoyl-
phenylvinylacetic acid (VII) or benzoylphenylcrotonic acid
(VIII).

CeH^CHClCHjCOCjHj C^H^Q : CHCOC^H^

CH2CO2H CH2CO2H

VI VII

C,H,CCH3C0C,H,

II
CHCOjH

VIII
The acid combines with bromine and forms a stable di-
bromide. The dibromide of benzoylphenylvinylacetic acid
would be expected to give either a bromlactone or an unsat-
urated T-'lactone when dissolved in sodium carbonate:

C^HsCBrCHBrCOCeH^ C«H,CBr CHCOCaH^

— >- NaBr+ O — >

1 I

CHXO^Na CH, — CO

C5H5C CHCOC.Hs

II I

II O -t- HBr

II I

CH — CO

The dibromide of benzoylphenylcrotonic acid, on the other
hand, would be expected to give, on similar treatment, a
bromine derivative of an unsaturated ketone:

Unsaturated d-Ketonic Acids 477

CeHjCBrCH^COC^Hs C^HsC— CH^COCeHj

I -^ NaBr + || + CO^

CHBrCO^Na CHBr

The dibromide of the acid in question readily dissolves in
sodium bicarbonate, but the clear solution almost imme-
diately becomes milky owing to the separation of an indiffer-
ent bromine compound, and in a very short time this passes
quantitatively into an unsaturated lactone. The unsaturated
acid is, therefore, benzoylphenylvinylacetic acid, as repre-
sented by formula (VII).

Benzoylphenylvinylacetic acid usually forms about 40
per cent, of the product obtained by heating the ketolactonic
acid. Its potassium salt is the only substance obtained by
dissolving the lactone melting at 93° in alcoholic potassium
hydroxide, showing that no shifting of the phenyl and ben-
zoyl groups has occurred during the formation of this lactone.
This leaves three possible formulas for the substance :

CeHjCH CHCOC^Hs . CgH^C CHXOC.H^

I o I o

i I I \

CH2 CO CH2-CO

IX X

QH5C : CHCCeH,

O OH

/
CH2CO

Benzoylphenylvinylacetic acid could be formed from a
lactone having any of these formulas, but it is little likely
that an a-hydroxy acid, such as would be obtained by opening
the ring in the ^--lactone represented by formula (IX), would
so readily lose water and pass into an unsaturated acid. More-
over, the two stereoisomeric lactones having this structure
are easily obtained by applying to benzoylphenylbutyric
acid the cpmmonest method used for getting ;'-lactones:

478 Kohler

CeH^CHCH.COCeHs CeHjCHCHBrCOCeHj

CH2CO2H CH.CO^H

CgHsCH CHCOCcHs

CHj CO

Both of the substances obtained in this way are diflferent
from the lactone in question. This must, therefore, be repre-
sented either by (X) or (XI). It was difficult to decide be-
tween these formulas because nearly all reagents that attacked
the carbonyl group also opened the lactone ring, and any
product obtained in this wa}' can be accounted for equally
well with either formula. By using great care, however, it
was found possible to add ethylmagnesium bromide to the
substance and get two stereoisomeric hydroxy lac tones. This
excludes formula (XI), and leaves only that of a /?-lactone.

The ease with which the lactone passes into an unsaturated
acid is in harmony with this conclusion. The hydroxy acid
obtained by opening the ring in a substance represented by
formula (X) belongs to the most unstable type of hydroxy
acids because it has a tertiary hydroxyl group in the /? posi-
tion and no hydrocarbon residue in the a position with refer-
ence to ketonic carbonyl; it therefore readily loses water
and passes into an unsaturated acid. This formula, finally,
is confirmed by a method of preparation of the lactone that
is essentially the same as that which has been used for getting
nearly all /?-lactones that are known .

When a solution of benzoylphenylvinylacetic acid, in glacial
acetic acid, is saturated with hydrogen bromide, the unsatu-
rated acid passes into an exceedingly unstable brom acid in
which the bromine must be in the /? position :

C^HsC : CHCOC^Hs CaHsCBrCHjCOCjHs

CH^CO.H CHjCOjH

When the brom acid is dissolved in sodium carbonate it
rapidly loses all its bromine; a small part of the product re-

Unsaturated d-Keionic Acids 479

mains in solution as sodium salt of the unsaturated acid, and
the rest separates as a lactone identical with the product ob-
tained by heating the ketolactonic acid.

There can be no doubt, therefore, that the lactone melting
at 93° is a ^-lactone, and that in the ketolactonic acid from
which it was obtained by heating the ring is also closed in
the /? position. The formation of a /^-lactone containing a
ring that is usually closed with difficulty here takes place so
easily that the ester of the dibasic acid is rapidly and com-
pletely hydrolyzed by one equivalent of base :

CgHjC : CHCOCgHs

1 + KOH + H3O -

CH(CO,CH3)2

C,H,CCH,COC,H,

I O + 2CH,0H

I \
KO2CCH— CO

This is the best method for getting the ketolactonic acid.
The product obtained in this way appears to be homogeneous,
but by very careful recrystallization from absolute ether it
is possible to separate it into two isomeric acids, the one al-
ready considered and which constitutes at least 98 per cent,
of the whole, and another that gives the same products with
nearly all reagents but crystallizes in a different form and
gives a lactone melting at 124° when heated.

The lactone melting at 124° was obtained in very small
quantity only, but its structure is established by the follow-
ing transformations:

1. The substance, like the isomeric lactone melting at 93°,
gives benzoylphenylchlorbutyric ester when its solution in
alcohol is saturated with hydrogen chloride.

2. When the ring is opened by solution in alcoholic potas-
sium hydroxide, the result is the potassium salt of the same
unsaturated acid that is obtained from the isomeric lactone.

3. With ethylmagnesium bromide the first product is a
magnesium derivative from which acids regenerate the lac-
tone; it is not possible to add the reagent without opening
the ring.

48o

Kohler

These facts show that the substance melting at 124° is the
hydroxylactone represented by formula (XI), the typical re-
actions being represented by the following equations :

I. CeH,C : CHC(OH)C„H,

CH2 — CO
CcH5CClCH2C(OH),C6H5

CH2CO2CH3

+ CH3OK + HCl =

C,H3CCICH,C0C«H,
CH2CO2CH3

+ H2O

II. CeH,C:CH-C(0H)QH3

O + C^HgMgBr =

CH2CO

CbH.C : CH— C(OMgBr)C6H5

CH,

The second lactonic acid obtained by hydrolyzing benzoyl-
phenyvinylmalonic ester is, therefore, the (?-lactonic acid
represented by formula (V). The formation of a /3- and a
^-lactone from the unsaturated ketonic acid, when all other
unsaturated acids give ^--lactones, shows that ring formation
in this case is determined entirely by the properties of the con-
jugated system C : C . C : O. The ^^-lactone is the result
of 1 ,4-addition, the <?-lactone that of 1,2-addition to this sys-
tem, and it is clear that in addition reactions in which both
addend and unsaturated chain are a part of the same mole-
cule, the mode of addition is the same as in those in which
they are brought together as separate compounds.

EXPERIMENTAL PART

In the first experiments sodium alcoholate was added in
small quantities to an alcoholic or ethereal solution contain-
ing equivalent amounts of a-brombenzalacetophenone and
ethyl malonate. The solution assumed a bright red color

Unsaturated d-Ketonic Acids 481

and sodium bromide began to separate almost immediately.
The addition of alcoholate was continued until the mixture
remained alkaline for an hour. The solution yielded no
solid product, but by boiling it with alcoholic potassium
hydroxide it was possible to get the dipotassium salt of ben-
zoylphenylvinylmalonic acid, showing that condensation had
taken place in the desired direction.

Methyl malonate was then substituted for the ethyl ester
in the hope of getting a solid product and the reaction was
carried out with the sodium derivative suspended in absolute
ether. Some solid methyl ester of the unsaturated ketonic
acid was obtained in this way, but the method is not satis-
factory. In order to complete the reaction it is necessary
to use much more than one equivalent of the sodium deriva-
tive and to boil for hours. It is difficult, therefore, to deter-
mine when the reaction is complete and the protracted boil-
ing leads to the formation of by-products that seriously in-
terfere with the isolation of the ester. The yield of dipotas-
sium salt obtained by hydrolyzing the entire product was 60-
70 per cent., that of pure solid ester small.

Much better results were obtained by starting with the
corresponding saturated ketonic ester. Vorlander^ showed
that, under the influence of sodium alcoholate, malonic ester
readily combines with benzalacetophenone both in ethereal and
alcoholic solution. He did not, however, isolate the ad-
dition product, but hydrolyzed it and analyzed the resulting
acid. I found that the amount of pure ester obtainable by
this method is small and that even when sodium malonate
and absolute ether are used the yields both of pure ester
and of acid are unsatisfactory.

Excellent results were obtained, however, by using piperi-
dine in place of sodium alcoholate as condensing agent. The
procedure is sho\vn by the following experiment: Ten grams
of piperidine were added to a solution of 208 grams of benzal-
acetophenone and 140 grams of methyl malonate in 400
cc. of methyl alcohol. The mixture was boiled continuously
for 72 hours, then distilled until most of the alcohol was re-

1 Ann. Chem. (Liebig), 294, 332.

482 Kohler

moved, and the residue allowed to cool. The solid was trans-
ferred to a Buchner funnel and thoroughly washed with
well-cooled methyl alcohol. One recrystallization from methyl
alcohol gave 244 grams of the pure methyl ester described
in a previous paper ^ — a yield of about 80 per cent. The
filtrates were combined, the solvent removed by distillation
and the piperidine by washing with acid, but the residue only
partially solidified. It was therefore hydrolyzed. It gave
30 grams of pure acid, making a total yield of over 90 per
cent.

Ethyl Y-benzoyl-^-phenylethylmalonate, CgHjCHCHoCOCeH-,

I
CHCCOoCjHJ,

was obtained both by direct condensation and by saturating

an alcoholic solution of the acid with hydrogen chloride.

It was purified by recrystallization from alcohol, from which

it separates in friable needles melting at 65°. It is readily

soluble in all common organic solvents except ligroin and

crystallizes poorly from all.

Analysis :

0.1422 gram substance gave 0.3730 gram CO2 and 0.0861
gram H2O.

Calculated for

C22H2i05 Found

C 71-9 71 5

H 6.5 6.7

The esters of benzoylphenylmalonic acid have two hydro-
gen atoms that are replaceable with bromine; but, as was
expected, one of these is much more easily replaced than the
other because substituents diminish the mobility of hydro-
gen in malonic esters. The introduction of bromine was car-
ried out in chloroform or carbon tetrachloride. The reac-
tion commenced slowly at the ordinary temperature, but after
it was well started the bromine disappeared as fast as it was
added until the solution contained a molecule of halogen per
molecule of ester. The solvent was then removed by distil-
lation under diminished pressure and the residue poured
into metliyl alcohol.

' This Journal, 46, 234.

Unsaturated o-Ketcnic Acids 483

Both the ethyl and methyl esters gave isomeric bromine
compounds. That these are not structural isomers is shown
by the fact that both form the same unsaturated ester when
hydrogen bromide is eliminated by means of reagents which,
like diethylaniline, do not cause isomeric change. They are,
therefore, stereoisomers — possible because the substances
contain two dissimilar asymmetric carbon atoms.

Ethyl y-Benzoyl-j-brom-^-phenylethyhnalonate,
CeHsCHCHBrCOCoH^

I . — The products from the ethyl ester

CH(COAH5)3
were separated with difficulty. The methyl alcoholic solu-
tion first deposited needles that, after purification, melted at
88°. The filtrates, on eva^joration, left an oil that solidified
in the course of a few weeks, and the resulting solid was finally
separated by means of ether-ligroin mixtures. It yields some
more of the needles melting at 88° and a new substance melt-
ing at 43 ° and crystallizing in large prisms or tables.

Analysis :

I. 0.1586 gram substance (88°) gave 0.3438 gram COj
and 0.0745 gram HjO.

II. 0.1762 gram substance (43°) gave 0.3835 gram CO2
and 0.0821 gram HjO.

Calculated for

C22H2305Br

Found

59 I

59-2
5-2

59-4
5-2

The two isomeric bromine compounds obtained from the
methyl ester of benzoylphenylethylmalonic acid were easily
separated by crystallization from methyl alcohol. One crys-
tallizes in needles melting at 113°, the other in plates or prisms
melting at 87°. Both are moderately soluble in cold alcohol
and ether, sparingly in ligroin.

Analysis :

I. 0.1525 gram substance (113°) gave 0.3227 gram CO2
and 0.0618 gram HjO.

II. 0.1681 gram substance (87°) gave 0.3550 gram CO2
and 0.0679 gram HjO.

484 Kohler

Calculated for
CsoH.sOjBr

Found

57-3

57-7

57.6

4-4

4-5

Methyl Y-Benzoyl-a,y-dibrom-{^-phenylethylmalonate,
QH^CHCHBrCOQH,

I . — The second reactive hydrogen atom

CBrCCO^CHJ^
in benzoylphenylethylmalonic esters can be replaced both by
boiling solutions of the monobrom substitution products,
in carbon tetrachloride, with bromine, and by exposing the
solutions to the action of direct sunlight for a long time. The
methyl ester yielded two isomeric products that were separa-
ted by crystallization from alcohol. The less soluble com-
pound separated in needles melting at 132°, while the second
product crystallized in large tables melting at 94°.

Analysis :

I. 0.1206 gram substance (132°) gave 0.2110 gram CO,
and o. 0390 gram HoO .

II. 0.1608 gram substance (94°) gave 0.2825 gram CO2
and 0.0520 gram 11,0.

Calculated for Found

C2oH,g05Br2 I II

C 48.2 47.7 47.9

H 3.6 3-6 36

Methyl Y-Benzoyl-^-phenylvinylmalonate,
CgHsC : CHCOCeHj

I . — Although the monobrom substitution

CHCCO^CHa)^
products of benzoylphenylethylmalonic esters lose hydro-
bromic acid with the greatest ease, it is, nevertheless, difficult
to get a satisfactory yield of pure unsaturated ester. The
difficulty is two-fold : the unsaturated esters very readily un-
dergo isomeric change under the influence of nearly all reagents
that can be used for eliminating the halogen acid; and small
quantities of impurities greatly interfere with the separation
of the esters in solid form. After trying many reagents under
various conditions the following procedure was finally adopted
as the most satisfactory.

Unsatiiraied o-Ketonic Acids 485

A 10 per cent, solution of potassium hydroxide in equal
parts of methyl alcohol and water is added, in small quanti-
ties, to a boiling solution of the ester in 4 times its weight
of the same alcohol. Every addition of base produces a faint
yellow color that immediately disappears on shaking. The
reaction is complete when the yellow color and an alkaline re-
action remain after vigorous shaking for a minute. The
methyl alcohol is then removed by distillation, the residue
extracted with ether, and the ethereal solution washed with
water until free from alcohol. The addition of a few crys-
tals of ester to the dried ethereal solution induces the separa-
tion of a large part of the ester in almost pure form, and the
amount may be increased by judicious addition of ligroin
to the solution. In the absence of "seed" it is necessary to
allow the solvent to evaporate and await spontaneous crys-
tallization in the resultant oil. This generally takes several
weeks and may take months.

The yield of pure solid ester obtainable in this way is 60-
70 per cent. The amount may be increased, somewhat, by
manipulating the filtrates, but it is more economical to boil
these with alcoholic potassium hydroxide and isolate the re-
sulting acid. The total yield is 85-90 per cent.

The ester is readily soluble in common organic solvents
except ligroin. It separates from ether in small flat needles,
from methyl alcohol in large transparent prisms or tables.
Its melting point is 94°.

Analysis :

0.1383 gram substance gave 0.3590 gram CO, and 0.0700
gram HjO.

C
H

Methyl benzoylphenylvinylmalonate immediately reduces
a cold solution of potassium permanganate in acetone, but
apparently does not combine with bromine. No reaction
takes place when bromine is added to its solution in carbon
tetrachloride at the ordinary temperature and in diffuse

Calculated for
CaiHisO.,

Found

71.0

5-4

70.8

5-6

486 Kokler

daylight. When the soUition is warmed or exposed to di-
rect sunlight, the bromine slowly disappears but the reaction
is accompanied by evolution of hydrobromic acid and the
resulting solid is a monosubstitution product. This was
purified by crystallization from alcohol. It separated in
prisms or tables melting at 141°.

Analysis :

0.1622 gram substance gave 03434 gram CO2 and 0.0613
gram HoO.

Calculated for
C2oH,705Br Found

C 57-6 57-7

H 4.1 4.2

xVs it is impossible to determine whether the substance is
formed by direct replacement, or by addition followed by
loss of hydrogen bromide, the location both of the bromine
and of the ethylene linkage is uncertain.

Hydrolysis of Methyl Benzoylphenylvinylmalonate. — As stated
in the introduction, the pure ester can be hydrolyzed without
loss by using exactly i equivalent of potassium hydroxide.
For this purpose the ester is dissolved in 5 times its weight
of alcohol that has been distilled from solid potassium hy-
droxide. The base is added as a 25 per cent, solution in equal
parts of alcohol and water. The solution becomes neutral on
standing for 10-12 hours in an ice chest, and most of the re-
sulting potassium salt separates in long, colorless needles.
These were washed with a small quantity of cooled acetone
and analyzed.

0.31 10 gram substance gave o.iioi gram KjSO^.

Calculated for

C18H.3O5K Found

K 15.72 15.6

For hydrolyzing the impure ester obtained from residues
it is better to use a large excess of base because it is possible
to get rid of impurities by washing the dipotassium salt,
which is sparingly soluble in excess of strong potassium hy-
droxide. The procedure is as follows: Aqueous potassium
hydroxide (2 : 3) is added in large excess to an alcoholic

Unsaturated d-Keionic Acids 487

solution of the ester. The mixture is well shaken and allowed
to stand in an ice chest for 10-12 hours. The resulting yel-
low or orange liquid is then slowly concentrated on a water
bath until most of the potassium salt has separated. The
liquid is filtered and the salt repeatedly washed with cold
acetone. Unless tiie ester contained very large quantities
of impurities the result is a colorless salt crystallizing in small,
thin plates. An analysis of the air-dried salt gave the follow-
ing results :

I. 0.4225 gram salt lost 0.0363 gram at 120°.

II. 0.61 10 gram salt lost 0.0397 gram at 120°.

Calculated for Found

Ci8H,205K2.2H20 I II

H3O 8.53 8.6 8.65

0.3862 gram anhydrous salt gave 0.1741 gram K2SO4.

Calculated for
C18H12O5K2 Found

K 20.21 20.2

The salt is stable in ordinary air, but on extremely damp
summer days it deliquesces and passes into hydrogen potas-
sium carbonate and the monopotassium salt previously de-
scribed. The same salt separates when a mixture of alco-
hol and ether is added to the strongly alkaline water solu-
tion of the dipotassium salt.

QHsCCH^COQHj

Benzoylphenylbutyrolactonic A cid,

Q . — From

solutions of either of the potassium salts obtained by hydro -
lyzing methyl benzoylphenylvinylmalonate acids precipitate
a colorless oil that slowly solidifies. The only solvents from
which the resulting solid can be obtained in well-defined
crystals are water and ether. It is sparingly soluble in boil-
ing water and separates very slowly and incompletely on cool-
ing the solution. In ether it is so readily soluble that only
large quantities or comparatively pure preparations can be
recry stall ized from it with advantage.

488 Kohler

The acid separates from water in small, lustrous pyramids
containing 2 molecules of water of crystallization, which it
loses below 100°.

I. 1 . 1066 grams substance, dried on paper, lost 0.1196
gram at 80° to 105°.

II. 1.7707 grams substance, air-dried, lost 0.1871 gram at
80° to 105°.

Calculated for Found

C18H14O5.2H2O I II

H2O 10.4 10.8 10.5

When heated slowly, the hydrous acid melts at about
100°, then loses water and resolidifies, and finally melts with
decomposition at about 170°. The acid separates from
ether in translucent plates that begin to decompose at the
same temperature as the dehydrated acid crystallized from
water.

Analysis :

I. 0.1400 gram substance (dehydrated) gave 0.3550 gram
CO2 and 0.0606 gram H^O.

II. 0.1404 gram substance (from ether) gave 0.3577 gram
CO2 and 0.0581 gram H2O.

Calculated for

Found

C.8H14O5

69.7

69.2

69 5

4-5

4.8

4.6

Diphenylhydroxycrotolactonic A cid,
CeH.C : CH-C(0H)QH3

I I . — The product of hydrolysis of the

HO,CCHCO— O

unsaturated ester contained a small quantity of a second acid
that accumulated in the ethereal filtrate. This acid could
be isolated only from the unusually pure preparations ob-
tained by hydrolyzing with one equivalent of base, and from
these only when large quantities were hydrolyzed at a time.
It crystallizes in fine needles that are extremely soluble in
ether, are decomposed by boiling water, and are apparently
changed when exposed in an impure state to air or light.

alculated for

C18H14O5

Found

69.7

69 5

4-5

4.6

Unsaturated d-Ketonic Acids 489

When heated rapidly, the substance melts with decomposi-
tion at about 170°.

Analysis :

0.1544 gram substance gave 03935 gram CO2 and 0.0639
gram HjO.

C
H

CeH^C : CHCOCeHs
y-Benzoyl-^-phenylvinylacetic Acid, \ . — The

CH2CO2H
product obtained by heating the ketolactonic acid contained
at least 4 substances, 2 of which were acid, the remainder
indifferent. The acids were isolated as follows: The lac-
tonic acid was heated to 170°- 185° in an oil bath until the
evolution of carbon dioxide ceased. The amber-yellow liquid,'
while still hot, was poured into ether in a separating funnel
and the solution extracted with sodium carbonate until free
from acid. From the solution in sodium carbonate acids
precipitated an oil that soon solidified. The solid was washed,
dried and crystallized from etlier or ether-ligroin mixtures.
From 95 to 98 per cent, of the acid product consists of an
acid which separates in small colorless prisms that melt at

i35°-

Analysis :

0.1128 gram substance gave 0.31 71 gram CO2 and 0.0542
gram HjO.

C
H

Benzoylphenylviny lace tic acid liquefies in contact with
alcohol and acetone, is sparingly soluble in carbon tetrachlor-
ide and ligroin, insoluble in water. In ether it is moderately
soluble, but both dissolves and separates very slowly. The
acid immediately reduces permanganate and combines readily
with bromine and halogen acids. An attempt was made
to determine the character of the chain by reduction to ben-

Calculated for

C,7H,403

Found

76.7

76.6

5-3

5-4

490

Kohler

zoylphenylbutyric acid, but it was impossible to confine the
action of the reducing agent to the ethylene linkage. The
acid was therefore transformed into

Methyl y-benzoyl-^-chlor-^-phenylhutyrate,
CeH^CClCH^COCeHs.

I by saturatimg its solution in methyl

CH2CO2CH3
alcohol with hydrogen chloride and purifying the product from
alcohol.

Analysis :

0.1325 gram substance gave 0.3330 gram COj and 0.0636
gram HjO.

Calculated for
CigHiyOgCl Found

C 68.2 68.6

H 5-4 5-3

The ester is moderately soluble in alcohol and ether. It
crystallizes in fine needles melting at 131°. By boiling with
zinc dust and methyl alcohol it was possible to eliminate the
halogen without attacking the carbonyl group. The product
was the methyl ester of benzoylphenylbutyric acid, identified
by comparison with a specimen on hand. This result proves
that the high temperature used in decomposing the ketolac-
tonic acid did not lead to any shifting of the groups.

■f -Benzoyl-^- phenyl- ^,-f'dihromhutyric Acid,
CeH^CBrCHBrCOCeHs^l

j "".—The addition of bromine to the

CH2CO2H
unsaturated acid was carried out in carbon tetrachloride. In
diffuse daylight combination takes place very slowly; it can
be greatly accelerated by addition of a little iodine and ex-
posure to direct sunlight. It results in two stereoisomeric
products. One of these is very sparingly soluble in carbon
tetiachloride and separates almost completely when the solu-
tion is cooled in a freezing mixture. It was washed with cold
tetrachloride and recrystallized from ethyl acetate or ace-
tone. From these solvents it separates in large colorless
plates that decompose below the melting point.

The carbon tetrachloride filtrates, on evaporation, deposi-

Unsaturated d-Ketonic Acids 491

ted fine needles that were readily soluble in all solvents ex-
cept ligroin, but could be purified by recrystallization from
alcohol-free ether. This substance also decomposes below
its melting point.
Analysis :

I. 0.1455 gram substance (plates) gave 0.2571 gram CO2
and 0.0440 gram H2O.

II. o. 1600 gram substance (needles) gave 0.2820 gram CO2 .
and 0.0482 gram, H2O. ^g^^^ ftAAd^^u^

Calculated for ?
C.7H,403Br2 ♦

Found

47-9

48.2

48.1

3-3

p. ^74

The two bromine compounds behave in the same way
when they are added to a solution of sodium bicarbonate
and they give the same product. If they are added as a suffi-.
ciently fine powder they dissolve rapidly and form a clear
solution. This shortly becomes milky, owing to the separa-
tion of a bromine compound. All efforts to isolate this failed
because it loses hydrogen bromide so easily that before the
extremely finely divided material can be filtered most of it
passes into an unsaturated lactone. This was easily puri-
fied by crystallization from alcohol. It crystallizes in color-
less or very pale yellow needles that melt at 131°.

Analysis :

0.1322 gram substance gave 0.3760 gram CO2 and 0.0540
gram HjO.

Calculated for

C.7H,203

Found

77.2

76.9

4.6

4-5

The lactone is insoluble in sodium carbonate but dissolves
readily in alcoholic potassium hydroxide. From the bright
red solution acids reprecipitate the lactone. It is therefore
probably a /--lactone, and as its formation is preceded by
that of an intermediate bromine compound, the reaction
probably takes place in the following steps:

492 Kohler

CeHgCBrCHBrCOCjHs C^HsCBr CHCOC^Hj

I
CH^CO^Na

C^H^C CHCOC^H^

II I

II o
II I

CH CO

The formation of any unsaturated lactone in this reaction
establishes the position of the bromine atoms, and therefore
that of the double linkage in the unsaturated acid.

The mother liquors from benzoylphenylvinylacetic acid,
on standing, deposited a small quantity of a second acid that
is much less soluble, crystallizes in large plates, and melts
at i8o°.

Analysis :

01375 gram substance gave 0.3842 gram CO2 and 0.0643
gram HjO.

Calculated for
CnHuOg Found

C 76 . 7 76 . 2

H 5-3 5-2

The acid is isomeric with the unsaturated acid just de-
scribed. It immediately reduces permanganate but does
not combine with bromine. Its solution in sodium carbon-
ate is colorless while that in strong potassium hydroxide is
yellow. The acid is evidently not a geometrical isomer of
the acid melting at 135°. It is obtained in very small amounts
as it constitutes less than half a per cent of the acid products.

The ethereal solution from which the acids were extracted
with sodium carbonate contained two lactones. One of these
is colorless, readily soluble in alcohol, moderately soluble
in ether; the other is lemon-yellow, readily soluble in ether,
moderately in alcohol. By adding ligroin to the dried ethereal
solution and cooling in a freezing mixture it was possible to
precipitate most of the colorless lactone. This was filtered,
washed with cooled ether until colorless and recrystallized

Unsaturated d-Ketonic Acids 493

from alcohol. It separates from hot solutions in needles,
from cold solutions in stout prisms or tables melting at 93°.

Analysis :

0.1304 gram substance gave 0.3660 gram CO2 and 0.0650
gram HjO.

Calculated for

C,7H„03

Found

76.7

76.6

5-3

5-5

As shown in the introduction, this substance is

QH.CCH^COCoH,

f-Benzoyl-^-phenyl-^-butyrolacione, I O . It is

I 1
CH2CO
much more stable than the /^-lactones that have been described
heretofore ; it is not affected by boiling either with water or
sodium carbonate and it can be heated to 200° vvithout change.
Even above this temperature it does not, like most other
/?-lactones, lose carbon dioxide but instead undergoes a com-
plex decomposition.

The lactone ring is easily opened by solution in alcoholic
potassium hydroxide. If tlie lactone is added to the well-
cooled alkaline solution in the form of a very fine powder,
and the liquid is acidified as soon as all is in solution, the sole
product is benzoylphenylvinylacetic acid. If the alkaline
solution is allowed to stand for some time before acidification,
the product may contain a considerable quantity of indiffer-
ent substances. For the purpose of isolating these, 25 grams
of lactone were dissolved in an excess of cold alcoholic pot-
ash and the solution allowed to stand in an ice chest overnight.
The solution was then poured into water and the indifferent
products extracted with ether. The ether, on distillation,
left 8.4 grams of colorless oil that was fractioned under di-
minished pressure. It was thus separated into almost equal
parts of benzaldehyde, recognized by the odor, and benzoyl-
carbinol, identified by its boiling point and by comparison
with a specimen made from w-bromacetophenone.

The formation of these substances indicates that when the

494 K oilier

lactone ring is opened the first product is, as usual, the salt
of a hydroxy acid, and that this decomposes in the alkaline
solution while the corresponding acid more easily loses Water:
C„H5C(OH)CH2COCeH,

I. I + H.O ^
CH2CO2K

CeHsCHO + QH^COCHjOH + CHjCO^K
C6H5C(OH)CH2COCeH5 CeHsC : CHCOCeHj

II. I =1 + H.,0
CH2CO2H CH2CO2H

The lactone ring is also easily opened with acids. When
the methyl alcoholic solution is saturated with hydrogen
chloride and then allowed to stand for several days, the lac-
tone passes completely into the ester of benzoylphenylchlor-
butyric acid that was described above. Similarly, when the
solution of the lactone in glacial acetic acid was saturated
with hydrogen bromide the principal product was

C6H-CBrCH2COC6H5

y-Bcnzoyl-^-phenyl-^'hrombutyric Acid, I . —

CH2CO2H
This extremely unstable acid was isolated as follows. The
solution was allowed to stand in an ice chest for a week after
it had been saturated with hydrogen bromide. It was then
poured into the ice water and shaken with enough benzene to
dissolve all of the yellow oil that separated. The benzene solu-
tion was repeatedly washed with ice water, then dried and the
benzene evaporated in a current of dry air, under diminished
pressure. The residue consisted of crystals imbedded in a yel-
low oil. The oil was absorbed in a porous plate and tlie solid
washed, on the plate, with small quantities of cold, alcohol-
free ether until it was almost colorless. It was then recrj'^s-
tallized from absolute ether, from which it separated in long
•colorless needles that decomposed without melting.

Analysis :

o. 1505 gram substance gave 0.3267 gram CO, and 0.0624
gram HjO.

Calculated for
CijH.sOsBr

Found

58.8

59-2

4-3

4-5

Unsaturated d-Ketonic Acids 495

The same acid was obtained by dissolving benzoylphenyl-
vinylacetic acid in glacial acetic acid, saturating the solution
with hydrogen bromide and treating the solution as before.
This method of preparation proves that the bromine is in the
^ position. For even if the double linkage in the unsaturated
acid had shifted, the result of the addition of hydrogen brom-
ide would still necessarily be a /?-brom derivative.

The acid is readily soluble in sodium carbonate but the
solution remains clear for but a few minutes. The faint milk-
iness that appears at first soon changes to a mass of needles
that continue to grow for several hours. The needles are
identical with the product obtained by heating the keto-
lactonic acid, thus proving, conclusively, that in this lactone
and in the ketolactonic acid from which it is obtained the
lactone rings are closed in the /? position. The only other
product formed by the action of sodiam carbonate on the^
brom acid is a small quantity of benzoylphenylvinylacetic acid.

8-Hydroxy-l3,d-diphenyl-^-heptalactone,
QH,CCH3C(C,H3)(C,H,)

O • — When an ethereal solution of

I
CH2COOH

the lactone melting at 93° is added to ethylmagnesium bro-
mide at the ordinary temperature the resulting reaction in-
volves both the carbonyl group and the lactone ring — the
product is insoluble in alcoholic potassium hydroxide. By
adding tlie lactone in the form of a fine powder and keeping
the reagent in a freezing mixture it is possible to confine the
action to the carbonyl group. The final result in this case
depends upon the subsequent treatment of the magnesium
compound. When this is decomposed with ice water before
addition of acid, the product is a substance that crystallizes:
in thick needles melting at 190°. When, on the other hand,,
the magnesium derivative is poured directly into a mixture
of ice and concentrated hydrochloric acid, the product crys-
tallizes in fine needles melting at 140°. Both substances
were purified by crystallization from methyl alcohol.
Analysis :

49-1 K older

I. 0.1291 gram substance (190°) gave 0.3655 gram CO.^
and 0.0804 gram HjO.

II. 0.1420 gram substance (140°) gave 0.4030 gram CO3
and 0.0880 gram HoO.

Calculated for Found

C19H20O3 I II

C 770 77.2 77.4

H 6.8 6.9 6.9

These substances are stereoisomeric lactones. They are
insoluble in sodium carbonate, soluble in alcoholic potassium
hydroxide. From solutions of the potassium salts obtained
from both, acids precipitate a third isomeric lactone that
melts at 150°. This was likewise purified by crystallization
from methyl alcohol, from which it separates in large lus-
trous needles.

Analysis :

0.1560 gram substance gave 0.4380 gram CO2 and 0.0960
gram HjO.

Calculated for

CigHzoOa Found

C 77.0 77.0

H 6.8 6.8

As only 2 stereoisomeric /^-lactones are possible, this third
substance must be a ^-lactone. Its formation from an acid
that has tertiary hydroxyl groups in both the /? and d posi-
tions, and similar residues in combination with the carbon
atoms that hold the hydroxyl groups, is good evidence that the
lactone ring closes more readily in the 8 than in the /? position.
€«H,CCH,C(OH)(C«H,)aH,

CeH,C(OH)CH.,C(OH) (C,H,)aH,
CH,CO„K

CeH3C(OH)CH,C(C«H5)C2H5

I
CH, CO

Unsaturated d-Ketonic Acids 497

The yellow lactone contained in small quantity in the fil-
trates from benzoylphenylbutyrolactone was purified with
great difficulty. It does not crystallize from ether and it is
so sensitive that it is changed even by boiling with alcohol.
It was therefore dissolved in ether and this solution poured
into ten times its volume of hot alcohol. After several repe-
titions of this process it was obtained in thin lemon-yellow
plates that melted sharply at 172°.

Analysis :

o. 1403 gram substance gave 0.4215 gram CO2 and 0.0632
gram Hp.

0.1106 gram substance gave 0.3325 gram CO, and 0.0500
gram 11,0.

Calculated for
C.tH.jO,

Found
I

82.3

81.9

82.0

4-9

The substance is insoluble in sodium carbonate. From
its solution in alcoholic potassium hydroxide acids precipi-
tate the unsaturated acid melting at 180°. The yellow sub-
stance is evidently, therefore, an unsaturated lactone. Since
acids neither precipitate a hydroxy acid nor regenerate the
lactone from the potassium salt, it probably is not a 7--lactone.
The ketolactonic acid could not yield an unsaturated ^-lac-
tone without a shifting of the phenyl group. The most proba-
ble formula for the yellow substance is therefore

^e^sC — CH = C — CgHg

II . I
CHCO — O

This formula accounts for the color, the instability, and the
formation of an unsaturated acid, isomeric with benzoyl-
phenyl vinylacetic acid, when the ring is opened.

CeHjC — CH=CCeH5

II I + H,0 =

CHCO — O

C^H^CCH : C(OH)C«H, C«H5CCH3COCeH,

II -^ II

CHCO,H CHCO^H

498 Kohler

d-Hydroxy-^,d-d iphenyl-d- cro tola done ,
CeH^C = CHC(0H)QH5

I I . — As stated above, the second lactonic

CH^CO— O
acid obtained by hydrolyzing benzoylphenylvinylmalonic
ester also decomposes with evolution of carbon dioxide when
heated above 170°. The product was treated exactly like
that obtained from the isomeric acid. It contains two sub-
stances in approximately equal amounts: benzoylphenyl-
vinylacetic acid and a new lactone melting at 124°. The lac-
tone was purified from methyl alcohol, from which it separates
in forms that can hardly be distinguished from those of the
lactone melting at 93°.

Analysis :

0.1493 gram substance gave 0.4189 gram CO2 and 0.0711
gram HjO.

Calculated for

C,7Hi403

Found

76.7

76.5

5-3

In all reactions in which the ring is opened the substance
gives the same products that are obtained from the lactone
melting at 93°. The behavior of the lactone toward ethyl-
magnesium bromide is, however, different. When the finely
powdered solid is dropped into a solution of the reagent that
is cooled in a freezing mixture, there is brisk evolution of gas
and the powder is transformed into a ciystalline magnesium
compound from which acids regenerate the original substance.
This is therefore a hydroxyl compound. If the temperature
of the reagent is allowed to rise the solid magnesium com-
pound disappears; but when the resulting magnesium deriva-
tives are decomposed with ice and acid they yield a mixture
of products of which none are soluble in alcoholic potassium
hydroxide. The only formula consistent with these results
is that given above.

y-Lactones

For purposes of comparison I made the ^--lactones that have
the same chain as the ^- and (^-lactones above described. The

Unsaturated d-Ketonic Acids 499

starting point was the corresponding saturated ke tonic acid,
prepared as directed by Vorlander.^

Y-Benzoyl-y-hrom-^-phenylhutyric Acid,
CeHsCHCHBrCOCeHj

] . — Benzoylphenylbutyric acid is spar-

CH2CO2H
ingly soluble both in chloroform and in carbon tetrachloride
but it reacts so readily with bromine that it is not necessary
to dissolve it. In working with large quantities it is better
to suspend the finely powdered acid in a moderate quantity
of nearly boiling carbon tetrachloride, add a little bromine
and await the initial reaction. After that, bromine may be
added freely until the color no longer disappears. Much of
the product separates in crystalline form in the course of
the reaction and most of the remainder is deposited when the
solution is cooled in a freezing mixture. The solid that sep-
arates from the boiling solution consists largely of an acid <
that melts with decomposition at about 189°, while most of
that which is deposited from the cooled solution consists of
an isomeric acid melting and decomposing at about 145°.
Both products Were recrystallized from ethyl acetate. The
higher melting acid separates in large fiat needles or plates,
the lower melting in small prisms.

Analysis :

I. 0.1660 gram substance (189°) gave 0.3601 gram COj
and o . 0692 gram HjO .

II. 0,1675 gram substance (145°) gave 0.3615 gram CO,
and 0.0685 gram 11,0 .

Calculated for

Found

CirHisOgBr

58.8

59 I

58.8

4-3

4.6

The two stereoisomeric methyl esters corresponding to these
acids were obtained both by introducing bromine into methyl
benzoylphenylbutyrate and by saturating solutions of the
brom acids in methyl alcohol with hydrogen chloride. They
were separated by recrystallization from methyl alcohol.

1 Ann. Chem. (LiebigV 294, 332.

Calculated for

Found

CsHnOgBr

59-3

59-5

4-7

j-Benzoyl-^-phenyl-y-hutyrolactone,

500 Kohler

One crystallizes in needles melting at 132°, the other in prisms
melting at 87°.
Analysis :

I. 0.1404 gram substance (132°) gave 0.3063 gram CO2
and 0.0634 gram HgO.

II. 0.1775 gram substance (87°) gave 0.3835 gram CO2
and 0.0787 gram HjO.

lated for Found

59 6
4-9
CeH.CH— CHCOQHj

O _

CH2— CO
Each of the ^-brom acids rapidly passes into 2 stereoisomeiic
lactones when dissolved in sodium carbonate. These were
separated by crystallization from methyl alcohol. The prin-
cipal product (80 per cent.) crystallizes in large tables melt-
ing at 130°, the other (18 per cent.) separates in lustrous
needles melting at 98°. Both are readily soluble in alcohol
and ether.
Analysis:

I. 0.1328 gram substance (130°) gave 0.3725 gram CO2
and 0.0625 gram HjO.

II. 0.1205 gram substance (98°) gave 0.3370 gram CO2
and 0.0585 gram HjO.

Calculated for Found

CuHmOs I II

C 76.7 76.5 76.3

H 5.3 5-3 5-4

The same lactones are obtained when either the brom
acids or their methyl esters are heated with dimethylaniline
to the boiling point of the solvent. The quantitative rela-
tions are, however, reversed, the product obtained at the
high temperature being composed mainly of the lower-melt-
ing lactone. Some of this lactone is formed when the acid
is heated by itself, but it is mixed with a large quantity of
by-products. When the lactones are obtained by these high-

Unsaturated o-Kcionic Acids 501

temperature reactions they are formed directly from the bro-
mine compound and not from an intermediate unsaturated
compound, because the unsaturated ke tonic acid obtained
by eliminating hydrogen bromide from the ^'-brom acid gives
only the /^-lactone when heated with dimethylaniline.

y-Benzoyl-y-hydroxy-^-phenylbutyric A cid,
CoH5CHCH(OH)COC,H5

I . — The hydroxy acid was obtained from

CH2CO.3H
both of the lactones by solution in alcoholic potassium hy-
droxide and subsequent acidification. It is unstable both in
the presence of bases and of acids. In alkaline solution it
probably behaves like other a-hydroxy ketones, as one of
the products identified was benzoylphenylbutyric acid:

2C6H5CHCH(OH)COCeH5 CeH^CHCH^COC.Hs

I • - I +

CH2CO2K CH2CO2K

QHsCHCOCOC^Hs

I +H2O

CH2CO2K

In the presence of acids, the substance rapidly reverts to the
lactone. The pure acid was obtained as follows :

An ethereal solution of the lactone was shaken with strong
cold sodium hydroxide until further addition of the base no
longer increased the amount of crystalline sodium salt that
separated. The salt was immediately filtered off and washed
with alcohol. The washed salt was dissolved in ice water
and cautiously acidified with cold dilute hydrochloric acid.
This precipitated a colorless solid which was removed as quickly
as possible and washed thoroughly with water. The wet
acid was dissolved in ether, the solution dried, filtered and
diluted with petroleum ether. The acid separated in long
colorless needles that melted, with decomposition, at 160°.

Analysis :

o. 1480 gram substance gave 0.3909 gram CO2 and 0.0730
gram HjO.

C,7H.oO.

71.8

72.0

5-6

5-5

502 Dinwiddie and K as tie

When the acid is allowed to remain in contact with dilute
acids it rapidly passes into the high-melting lactone; above
the melting point it loses water rapidly and gives the low-
melting lactone, which is evidently the only form that is
stable at high temperatures.

Y-Benzoyl-y-hydroxy-^-phenyl-y-heptalactone,
QH,CHCHC(OH)(C,H,)CeH,

O , was one of the products ob-

CH^CO
tained by the action of ethylmagnesium bromide on the high-
melting lactone. It was purified by crystallization from methyl
alcohol, from which it separates in needles melting at 103°.
Analysis :

0.1412 gram substance gave 03974 gram COj and 0.0865
gram Hp.

Calculated for

C19H20O3 Found

C 770 76.7

H 6.8 6.8

The substance is not attacked by boiling sodium carbonate
and it is reprecipitated when its solution in alcoholic potas-
sium hydroxide is acidified.

It is much more difficult to confine the action of the Grig-
nard reagent to the carbonyl group in the case of this ^--lac-
tone than in that of the corresponding /3-lactone. The pro-
cedure was exactly the same with both, but the ^'-lactone
always yielded a considerable quantity of material that was
insoluble in alcoholic potassium hydroxide.

Chemical Laboratory
Bryn Mawr College

THE BROMINATION OF PHENOL

By J. G. Dinwiddie and J. H. Kastle

It is an interesting fact that when bromine is added to an
aqueous solution of phenol only tribromphenol and tribrom-
phenol bromide, tetrabromcyclohexadienone, are formed. No
one has ever obtained any evidence of the formation of lower

The Bromination of Phenol 503

substitution products. In the course of some work with
phenol it was observed by one of us, Kastle, that bromine acts
readily upon phenol in such organic solvents as glacial acetic
acid and chloroform, with the formation of colorless sub-
stitution products and the evolution of hydrobromic acid gas.
It therefore occurred to him that it might be of interest to
determine just what substitution products of phenol were
produced under these conditions. Accordingly it was sug-
gested to Mr. Dinwiddle to undertake a few experiments with
this end in view. The experimental results given in the
following are his. Four solvents were employed, namely,
glacial acetic acid, chloroform, carbon tetrachloride and carbon
bisulphide. In all of the experiments a i per cent, solution
of phenol in the several solvents and about a 3 per cent, solu-
tion of bromine were used, the precise concentration of the
bromine being accurately determined in each case. A
measured quantity of the phenol solution, in all experiments
5 cc, w^as placed in a dry glass-stoppered bottle and to this
an excess of bromine in the given solvent was added. The
mixture was then allowed to stand for different intervals
of time at the end of which the bromine remaining was deter-
mined by adding a measured quantity of the solution to a
strong aqueous solution of potassium iodide, and titrating
with 0.1 N sodium thiosulphate, i cc. of which was found
to be equivalent to 0.00799 gram bromine.

The results of these determinations are given in Table I.

The calculated amounts of bromine required to convert
0.05 gram of phenol into the mono-, di- and tribromphenols,
respectively, are 0.0851 1, 0.1702 and 0.2553 gram. It is
evident, therefore, that in the organic solvents thus far in-
vestigated the bromination of phenol results in the formation
of a dibromphenol or a mixture of the mono- and tribrom-
phenols.

In order to get some further idea of the nature of the bromine
substitution product produced by the action of bromine on
phenol in these organic solvents, 0.5 gram of phenol was dis-
solved in 10 cc. of chloroform and to this solution there was
added the required amount of bromine to form the dibrom-

504

Dinwiddie and Kastle

Table I

Phenol

in Glacial

Acetic

Acid

6 a

•0

"1
0-"

0.05

0. 2099

I I

24 hrs.

0.0475

0. 1624

0.05

0.3147

((

0.1366

0. 1781

0.05

0.2801

O.1051

0.1750

0.05

0.2365

I wk.

0.0479

0.1886

Phenol in

Chloroform

0.05

0.1768

24 hrs.

0.0126

0. 1642

0.05

0.1768

0 . 0084

0.1684

0.05

0.1768

0.0104

0. 1664

4=^

0.05

0.2231

1 wk.

0.0242

0. 1989

Phenol i

in Carbon Tetrachloride

0.05

0.2613

24 hrs.

0.0975

0.1638

0.05

0.3055

0.1386

0. 1669

0.05

0.3492

0.1861

O.1631

0.05

0.2619

I wk.

0.0870

0.1749

Phenol

in Carbon Bisulphide

0.05

0.1846

24 hrs.

0.0352

0.1494

0.05

0.2308

0.0931

0.1377

0.05

0.2805

II .

O.1318

0.1487

0.05

0.2769

II .

I wk.

0. 1087

0.1682

phenol. The bromine used was dissolved in chloroform.
After standing for two days this mixture was poured into an
evaporating dish and allowed to evaporate spontaneously.
There was left an oily liquid which, when poured into water,
sank to the bottom of the vessel and on cooling in ice water
solidified to a white solid. This substance was found to melt
at 34°. A bromine determination gave 60.15 per cent,
bromine; calculated for dibromphenol, 63.49 per cent. Con-

I In the first three of the above experiments the quantity of bromine acting on
the glacial acetic acid was found to be negligible. In experiment (4), however, a cor-
rection has been made for the amount of bromine acting on the acid. Thus, in a
control experiment extending over one week 0.2791 gram of bromine in 13 cc. of acetic
acid took 20.6 cc. 0.1 N sodium thiosulphate, equivalent to 0.2365 gram of bromine.
Hence, in experiment (4) of the above series there was left only this last amount of
bromine to react with the phenol.

^ In this experiment a correction was made for the amount of bromine acting on
chloroform, after one week's exposure. This correction was determined as in the case
of the acetic acid.

The Bromination of Phenol 505

sidering the fact that the substance was not specially purified
and that, owing to its low melting point, it proved difficult
to handle, the agreement is sufficiently close to indicate that
we are dealing with a dibromphenol. The substance is prob-
ably the 2,4-dibromphenol. It is evident from these ob-
servations that in its conduct toward bromine in organic
solvents phenol is similar to many other simple aromatic
substances. A considerable number of these are converted
directly into dibrom derivatives on bromination. The pro-
cess is a simple act of substitution. On the other hand the
conversion of phenol into tribromphenol and tribromphenol
bromide, in aqueous solution, by the action of bromine, is
somewhat peculiar and anomalous. Similar changes are
shown by but few organic substances. Among those which
do react similarly may be mentioned aniline which, in aqueous
solution, is converted into tribromaniline by the action of
bromine. So far as is known at present there is no essential
difference between a solution of bromine in water and in an
organic solvent, beyond the fact that in all probability a
portion of the bromine is hydra ted in its aqueous solution.

According to Jakowkin^ bromine in its aqueous solution
cannot be ionized, so that these differences in the conduct
of bromine in water and organic solvents towards phenol are
evidently not referable to the bromine but to the phenol.

One of us, Kastle, is inclined to the opinion that the peculiar
conduct of phenol, in its aqueous solution, towards bromine is
in some way dependent on the tendency of phenol to form a
quinoid derivative. It is well known that in the presence
of water bromine converts tribromphenol into tribromphenol
bromide. This latter compound is a quinoid derivative hav-
ing the constitution

Ber. d. chem. Ges., 30, 518 (1897).

C.Br,

5o6 Dinwiddie and Kastle

Kastle and Gilbert have shown that trichlorphenol is con-
verted in the same way into trichlorphenol bromide on the
addition of bromine to an aqueous suspension of trichlor-
phenol. These changes can be readily accounted for on the
supposition that under the influence of water the tri-sub-
stitution products of phenol are converted into the corre-
sponding quinoid derivatives. Thus,

OH CO

Brr \Br Brr >Br

Br C

/\
Br H

It now we have the quinoid hydrogen in this last compound
replaced by bromine we have tribromphenol bromide and tri-
chlorphenol bromide produced by simple substitution. Thus,

CO CO

Brj^ >Br Brr ^.Br

+ Br, = I J -f HBr and

C C

Br H Br Br

CO CO

Cir ^Cl CK \CI

+ Br, = 1 J 4- HBr

C C

/\ /\

CI H CI Br

The Bromination of Phenol

507

There is no reason why phenol itself might not under certain
conditions pass to the quinoid form. In this event we would
have

OH CO

CH2

If now bromine were to replace the two quinoid hydrogens
we would have

+ 2 Br,

+ 2HBr

C.H,

C.Br,

The latter compound, being a quinoid derivative, contains two
double bonds, as shown by the formula

C.Br^

It could therefore take up two molecules of bromine thus,
CO CO

+ 2 Br,

BrHC
BrHC

jCHBr
CHBr

CBr,

C.Brj

5o8 Kastle and Haden

By loss of hydrobromic acid this last substance would pass
to tribromphenol bromide,

Brr >Br

C.Br,

Kastle has lately come to believe that while tribromphenol
bromide results from the action of bromine on tribromphenol,
it is also the first product of the bromination of phenol in water
and that the tribromphenol is formed from it only by the
action of an excess of phenol. Thus we see that when bromine
is added gradually to an aqueous solution of phenol a yellow
precipitate is always formed which, by reaction with the
excess of phenol present, becomes white. Now tribrom-
phenol bromide is yellow, whereas tribromphenol is white.
Hence it appears probable that tribromphenol bromide is
formed first whenever bromine acts on an aqueous solution
of phenol and that only through interaction of this substance
with phenol is tribromphenol produced. On the other hand,
in the absence of water and in organic solvents bromine acts
upon phenol in precisely the same way that it does upon other
aromatic substances. That is, the change is one of simple
substitution.

University of Virginia
July, 1911

A STUDY OF ORTHOAMINOPARASULPHOBENZOIC

ACID AND ITS DERIVATIVES, WITH SPECIAL

REFERENCE TO THEIR FLUORESCENCE

By Joseph H. Kastle and R. L. Haden

[second communication]
The Esters of o-Amino-p-sulphobenzoic Acid. — As pointed
out in a previous communication^ our earlier attempts to

1 This Journal, 45, 58-78.

Study of Orthoaminoparasulphohenzoic Acid 509

esterify this amino acid by the action of ethyl iodide on the
disilver salt of the acid invariably resulted in the production
of a N-ester of the formula

/COOH
QHj^NH.aHs + H,0
\SO2OH

In one experiment, however, a very small amount of a sul-
phur-yellow compound crystallizing in fine needle-shaped
crystals and melting at about 160° was obtained. Later it
was observed that this yellow substance is soluble in ether,
chloroform and alcohol. Putting to account this additional
knowledge regarding its solubility in these organic solvents the
attempt has been made to obtain the yellow compound in lar-
ger quantity. Six grams of the dry, finely pulverized disilver
salt of the amino acid was placed in a small Erlenmeyer flask
and ethyl iodide added in small portions at a time. The sub-
stances reacted almost immediately with slight rise of tem*
perature. When the reaction was apparently complete
another portion of ethyl iodide was added and the flask was
tightly closed with a cork. It was then allowed to stand
overnight at room temperature. The next day the excess of
ethyl odide was removed by a current of air, and the residue
extracted with absolute alcohol and filtered. A yellow solu-
tion was thus obtained which on the gradual addition of water
became milky and gave a crop of fine, yellow, needle-shaped
crystals. These were filtered off and dried in the air. The
yellow compound thus obtained was found to weigh 0.9 gram
and to melt at I5i°-i53°, uncorrected. The small yield of
this compound is discouraging and difficult to explain. The
fact, however, that the N-ester is always formed in this re-
action renders it probable that the esterification of the di-
silver salt of the amino acid takes place in the following stages :

/COOAg /COOAg

(i) CeH^^NH^ + QHJ = CeH3^NH3.C,H,I
\SO2OAg \SO2OAg

/COOAg /COOAg

(2) QH.^NHAHJ =CeH,f NH.C2H, -f HI
\SO2OAg \SO2OAg

5IO Kastle and Haden

/COOAg /COOH

(3) QHgfNH.C^H, + HI = QHg^NHCK, + Agl

\SO2OAg \SO2OAg

/COOH /COOK /COOAg

(4) 2CeH3f NH.QH, = C,H3f NHC3H, + CeH3f NH.C,H,

\SO2OAg \SO2OH \SO2OAg

/COOAg /COOC2H5

(5) C,H3^NH.C3H, + 2C2HJ = C,H3^NH.C3H, + 2AgI

\SO2OAg NSOPC^H,

According to this conception of the process only half of the
disilver salt of the amino acid would really go to form the tri
ester and the theoretical yield of the tri ester would be re-
duced by half. The actual yields, therefore, amounting to
from 22-36.9 per cent., would not be so bad. The above
conception of the process would also explain the constant
production of the nitrogen rhombic ester in this reaction.

Triethyl o-Amino-p-sulphobenzoate. — The compound thus
obtained by the action of ethyl iodide on the disilver salt of
the amino acid is yellow and crystallizes from a mixture of
water and alcohol in fine needle-shaped crystals, melting at
151°- 1 53°, uncorrected. The compound is practically in-
soluble in water but soluble in a great variety of organic
solvents. It contains no iodine and no water of crystalliza-
tion.

On combustion by de Roode's method 0.2062 gram of the
substance gave 0.1171 gram of water and 0.3918 gram of

/COOC,H,
carbon dioxide; calculated for CgH3;— NH.CjHj , 0.1067 gram

\SO2OC2Hj
of water and 0.3873 gram carbon dioxide. A sulphur
determination by Liebig's method on 0.1070 gram sub-
stance gave 0.0742 gram BaS04, equivalent to 0.0102

yC00C2H5
gram or 9.53 per cent, sulphur; calculated for CgHgtr-NH.CjHj ,

\SO,OC2H,
10.65 per cent.

This large error is probably due to the very small amount
of the substance employed for analysis. The yellow sub-
stance is neutral to litmus, but its alcoholic solution is acid to

Study of Orthoarninoparasulphohenzoic Acid 511

phenolphthalein and on boiling with water it also becomes
strongly acid and gradually passes into solution.

0.0925 gram of the substance boiled for eight hours with
water under an inverted condenser, required 6 cc. o.i N
sodium hydroxide for neutralization and after standing for
several days the same solution required 0.15 cc. more, making
a total of 6.15 cc; calculated for

/COOC3H, /COONa

QH3f NH.C.,H, + "^^^f C«H3^NH.C,H„

6.14 cc.

0.0562 gram of the ester, dissolved in absolute alcohol
and titrated immediately, required 1.85 cc. o.i N sodium
hydroxide for neutralization, phenolphthalein being used as
indicator; calculated for

/COOQH, , NaOH /COOC.H,

^SCPC^N. \SO.,ONa

1.86 cc. It should also be borne in mind in this connection
that the ease with which one of the ester groups is hydrolyzed or
saponified is characteristic of the sulphonic esters. It is evi-
dent, therefore, from these results that the yellow substance re-
sulting from the action of ethyl iodide on the disilver salt of the

yCOOC.Hs
amino acid is a triethyl ester of the constitution CgHj^NH.CjHj

\SO,OC,H5
which on complete hydrolysis yields the nitrogen rhombic

^COOH
ester, C6H3;;-NH.CjH5 -1- HjO, described in our first communica-

\SO,OH
tion.^

The Acid Silver Salt of o- Amino p-sulphobenzoic Acid,

f /COOH /COOAg -,

QH3^NH, + QHg^NH, + 2H,0

L \SO,OH \SO,OAg J

During the course of this investigation the attempt was made

1 This Journal, 46, 71-74.

512 Kastle and Haden

to prepare a benzoyl derivative of the amino acid by treating
the disilver salt of the acid with benzoyl chloride. In one
experiment 6 grams of the disilver salt was heated in a sealed
tube with the calculated amount of benzoyl chloride until
apparently the reaction was complete. The contents of the
tube were then treated with successive portions of ether and
acetone to remove the excess of benzoyl chloride. The residue
was then boiled with water and filtered to remove the silver
chloride and the filtrate evaporated to crystallization. A
quantity of prismatic crystals which were found to darken
in the light was thus obtained. They gave an acid reaction
to litmus and were found to contain silver.

0.1335 gram of this salt required 3.8 cc. o.i N sodium hy-

/COOH
droxide for neutralization; calculated for C5H3;;—NH^ + H,0,

^SO.Ag
3.9 cc.

0.1240 gram of the salt took 3.7 cc. o.i N ammonium sul-
phocyanide; calculated according to the above formula it
should have required 3.9 cc.

0.6770 gram of the salt, heated in the air bath at 150° for
one hour, lost 0.0363 gram, or 5.36 per cent., of water of crystal-
lization; calculated according to the above formula, 5.26
per cent.

Repeated analyses have established the composition of this
acid silver salt and in harmony with these results the best
way of preparing the salt consists in heating together, in
water, the calculated quantity of the amino acid and the
disilver salt, filtering from any undissolved disilver salt and
evaporating the filtrate to crystallization. In this way large
amounts of the acid silver salt have been prepared. Our
earlier attempts to esterify this acid silver salt resulted in the
formation of an acid substance which we now have reason to
believe consisted of a mixture of the original amino acid and
the nitrogen rhombic ester in varying proportions. These
results were obtained before we had discovered a satisfactory-
method of isolating the yellow tri ester described in the fore-
going. Having obser\^ed that absolute alcohol could be

Study of Orthoaminoparasulphohenzoic Acid 513

employed satisfactorily in extracting the yellow tri ester from
the products of the reaction of ethyl iodide on the disilver
salt of the amino acid, it occurred to us that possibly it could
also be employed to advantage in the isolation of the ester
resulting from the action of ethyl iodide on the acid silver
salt. Accordingly 1.0202 grams of the anhydrous acid silver
salt was heated in a sealed tube, on a water bath, with an
excess of ethyl iodide for one hour. On cooling large yellow
crystals were seen suspended in the excess of ethyl iodide.
The tube was then opened and the excess of ethyl iodide re-
moved by warming the tube for a few minutes on the water
bath. The contents of the tube were then extracted with hot
absolute alcohol and filtered. Water was th( n added to the
yellow alcoholic filtrate and the yellow crystcil:> thus obtained
were collected on filter paper and dried at room temperature
in the air. There was thus obtained 0.13 15 gram of a
yellow substance identical in all respects with the yellow
tri ester obtained by the action of ethyl iodide on the
disilver salt of the amino acid. Thus, it was found to melt at
1 50°- 1 53°, and when mixed with some of the tri ester which
had been made from the disilver salt, the mixture showed the
same melting point, namely, I5i°-i53°. For 0.0487 gram of
the yellow ester prepared from the acid silver salt, when dis-
solved in absolute alcohol, was required 1.6 cc. o.i N sodium
hydroxide for neutralization; calculated for

/COOC,H, ^ ^ /COOC^H^

CeH3f NH.C.H, ^^^f C.H^f NH.C.H,
^SO.OQHj ~^ \SO,ONa

1. 6 1 cc. After the first titration, the solution of the ester
in alcohol was boiled under an inverted condenser, at inter-
vals, for about ten hours. It became acid again and on the
second titration took 1.7 cc. o.i N sodium hydroxide; calcu-
lated for

/COOQH, ,. ^„ /COONa

QH,^ NH.C,H, ^^ QH3^NH.C,H,
\SO,ONa ^ ^SO.ONa

1. 6 1 cc. There is no doubt, therefore, that the yellow ester

514 Kastle and Haden

resulting from the action of ethyl iodide on the acid sih er salt
of the amino acid is the same as that resulting from the action
of the iodide on the disilver salt. The residue left after ex-
tracting with absolute alcohol the yellow ester from the products
of the reaction of ethyl iodide on the acid silver salt was then
extracted with boiling Water and filtered to remove the silver
iodide. The aqueous extract was then evaporated to small
bulk, when a mass of white crystals separated on cooling.
These v/ere collected on the filter and allowed to dry in the air
at room temperature. They were found to be acid in reaction
and to possess the characteristics of the original amino acid.
Of this substance o.iooo gram required for neutraliza-
tion 8.75 cc. 0.1 N sodium hydroxide, while o.iooo gram of
the amino acid should require 8.85 cc. It is evident, there-
fore, that two substances result from the action of ethyl iodide
on the acid silver salt of the amino acid, namely, the yellow tri
ester of this acid and the original amino acid itself. This goes to
prove, of course, that the acid silver salt is not a monohydrated

/COOH
single molecule such as CgHj^NH^ + H^O, but a dihy-

^SO.OAg
drated, mixed acid salt consisting of one molecule of the
amino acid and one of the disilver salt of the amino acid,
having the formula

r /COOH /COOAg -,

CeH3^NH, + C,H3^NH, -f- 2H,0

L \SO,OH \SO,OAg -J

The last formula enables us to account for the conduct of the
acid silver salt on esterification with ethyl iodide. Up to the
present we have obtained no evidence of the existence of such
acid esters of the amino acid as the following :

/COOC.H, /COOH

QH3^NH, " , CeH3^NH,

\SO,OH \S0,0C3H5

/COOC2H5 /COOH

C.Hj^NH.C.Hj and C«H,(-NH.C,H5
\SO,OH ^SOjOCjHs

Study of Orthoaminoparasulpkobcnzoic Acid 515

Such compounds would be of great interest as possibly aflfect-
ing the fluorescence of the parent acid and every effort will be
made to prepare them. From all that is known at present,
however, regarding the conduct of the amino acid on esterifica-
tion we must confess that we have little hope of success. In
fact, the possibility of the existence of certain of these com-
pounds appears doubtful.

As indicated in our first communication the study of 0-
amino-Zj-sulphobenzoic acid was originally undertaken with
the view of determining the influence of simple chemical
changes in constitution on the degree and character of the
fluorescence exhibited by the original compound. This has
proven far m >re difficult than was originally contemplated by
reason of the difficulty of obtaining even the simpler derivatives
of this acid. In view, therefore, of our original purpose it
seemed desirable to compare the two esters already described
with the parent amino acid as to fluorescence in various sol-
vents. These solvents were selected at random from the chemi-
cals on hand at the time and without any particular regard to
their chemical nature. The results of these observations are
given in Table I.

It is evident from these observations that of the three sub-
stances here under consideration the tri ester exhibits fluores-
cence in by far the greatest number of different solvents.
This is evidently due to the fact that this substance is soluble
in a greater number of solvents than either of the other two
compounds. The phenomenon of fluorescence is a phenom-
enon of solution, at least to a great degree, and hence if a
substance is not soluble in a given solvent it does not fluoresce
on being brought in contact therewith. Whether either of the
substances here under consideration would fluoresce in the dry
state in ultraviolet light has not as yet been determined, but
it seems hardly likely that they would do so. Another point
of interest in this connection is that in all strictly indifferent
solvents in which the compound is soluble the tint or quality
of the fluorescence is the same, whereas water and the organic
acids alter the quality of the fluorescence to some extent.

It is also evident from the above comparison that the intro-

5i6

Kastle and Haden

•5 S

> o
o

o
o

y c

tn W

O 6 ^
v- Wh C

-MOO

en :3
o >:;

D 3 fl O)

illi

O (U

5i o

O -H '",

5 (u

03 ,^ >-i
O

biD ..

a O

OJ O

?^ be OJ

y c y

(/5

o - o
q^ O cc

0) (U

3 -^ -Q
-w ^ '=«
.be-a 2

4J O

c Si

^ bo

o >.

■^ c o

oT
a

D
O
(/)
<U

O <U

:3 be

4) Cl,'

;3
a

.^« O
CO ^

ell

WucxJ

2 a c

^' O 3

Study of Orthoaminoparasulphobenzoic Acid 517

o g o .-- C C g "G o

Is I- Is |e||l§sB s ^ s.

3 •^S'oS §2f.-°„o3ScS £ §

8.1

^-5

0 c

^— s s§

I ^

2 o t^

<L» ^ 5 <u

CO JJ-a o

■| ^ § -^- ^^

^ i!<« o § ^q=l

,0 o i -^ '^ o 2 ^ ^ o

^ iz; o^ 00 ^ H :z; :

I
Sao ?i .;

r -M aj cn

bjO o g

O Sao -MO

o a-" o -^^ o

!S O ^ cH ^

0-J3 ^ 'o n -o c ■;

•^uoo pq S WW h4 wwwwj^ooo \

5i8 Reiser and McM aster

duction of an ethyl group into the compound in place of one
of the acid or amino hydrogens has the effect of greatly in-
creasing the blue tint of the fluorescence. This is seen to special
advantage in the aqueous and alcoholic solutions. Thus in a
mixture of equal volumes of water and alcohol the amino acid
shows a pinkish purple fluorescence, the nitrogen rhombic ester
pure blue and the yellow tri ester pure blue, with a yellowish
solution. Since the publication of our first communication on
this subject, we have noted from an examination of the litera-

/COOH
ture that aminoterephthalic acid, C5H3;— NHj(o) , is also

^COOH(p)
fluorescent in aqueous solution. That such is the case
has been confirmed by Mr. Lester Patton, working in this
laboratory. Mr. Patton finds that in dilute aqueous and
alkaline solutions aminoterephthalic acid shows a reddish
blue fluorescence, whereas in acetone it shows a pure blue
fluorescence. He also observed that the mono- and diethyl
esters of this acid exhibit a pure blue fluorescence in their
aqueous solution. So that, so far as fluorescence is con-
cerned, aminoterephthalic acid is apparently quite analogous
to o-amino-/j-sulphobenzoic acid.

We hope in the near future to make a more careful investiga-
tion of aminoterephthalic acid and its derivatives with the
view of comparing them, as to fluorescence, with the corre-
sponding derivatives of the acid here under consideration
and those of aminobenzene-/>-disulphonic acid which, so far
as we have been able to ascertain from the literature at present
available, has never been made.

University of Virginia
July. 1911

THE SYNTHESIS OF FUMARIC AND MALEIC ACIDS
FROM THE ACETYLENE DIIODIDES

By Edward H. Keiser and LbRoy McMaster

In 1890 one of us (Keiser*), by treating solid acetylene
diiodide in alcoholic solution with potassium cyanide and

' This Journal, 12, 99.

Synthesis of Fumaric and Maleic Acids 519

caustic potash, succeeded in making fumaric acid. Subse-
quently in 1899 Keiser^ obtained for the first time a pure
liquid isomer of the solid acetylene diiodide and showed that
these substances were geometrical isomers and at that time
expressed the view that the solid acetylene diiodide was to be
regarded as the trans or fumaric form and the liquid geo-
metrical isomer the cis or maleic form. Their configuration
was represented thus :

H— C— I H— C— I

II II

I — C— H H— C— I

Solid acetylene diiodide Liquid acetylene diiodide

(melts at 73°) (boils at 185°)

We have now established these formulas by repeating and
confirming the earlier work upon the synthesis of fumaric
acid from the solid acetylene diiodide, and, in addition, we
have transformed the liquid diiodide into maleic acid.

Synthesis of Fumaric Acid
Experiment No. i. — Solid acetylene diiodide (4. 11 grams)
was dissolved in 150 cc. absolute alcohol, 3 grams of powdered
potassium cyanide added and the mixture boiled on a water
bath under a reflux condenser for 47 hours. Three grams of
solid caustic potash were now added and the heating con-
tinued for one and one-half hours, when ammonia was no
longer given off. The solution was filtered and allowed to
evaporate spontaneously. A mixture of crystals was ob-
tained, some of which were needle-shaped. Qualitative tests
showed the presence of iodide, cyanide and fumarate in this
mixture. The fumarate was then separated and purified as
follows: The crystals were dissolved in water and silver
nitrate added until a precipitate was no longer formed. The
precipitate, consisting of silver fumarate and some silver
iodide and cyanide, was digested with dilute ammonia and the
mixture filtered. The filtrate now contained the fumarate
and some cyanide. Dilute nitric acid was now added in
slight excess. This caused the cyanide to be precipitated,
but held the fumarate in solution. After filtration the filtrate,

» This Journal, 21, 261.

520 Keiser and McM aster

containing the fumarate, was now made exactly neutral with
ammonia. A white precipitate formed. This was collected
upon a filter and again dissolved in dilute nitric acid and
precipitated by ammonia. The silver fumarate was then
collected upon a hardened filter and dried at ioo° for one
and a half days. The salt turned slightly brown and when
heated it deflagrated like gunpowder. Determinations of the
percentage of silver gave the following results:

0-I354 gram of salt gave 0.1183 gram AgCl.

0.2050 gram of salt gave 0.1792 gram AgCl.

alculated for

C4H204Ags

65 -43

65 -74

65 78

These determinations were made by dissolving the weighed
amounts of fumarate in dilute nitric acid and adding hydro-
chloric acid in slight excess. The silver chloride was filtered,
dried and weighed. The filtrate from the silver chloride,
containing the free fumaric acid, was evaporated to dryness,
when a white flaky residue was obtained. This was dried at
105°. It weighed, in the first analysis, 0.0428 gram instead
of 0.0476 gram, calculated. A small quantity of this residue
was placed between watch glasses and heated in an air bath.
At 200° it sublimed into the upper watch glass. A sample
of pure fumaric acid (Kahlbaum) acted likewise. Some small
crystals of this synthetic fumaric acid were introduced into a
melting-point capillary and heated beside a similar specimen
of the pure fumaric acid in a sulphuric acid bath. At 200°
each specimen sublimed to the upper cold part of the capillaries.

Experiment No. 2. — Ten grams of solid acetylene diiodide
and 7.5 grams potassium cyanide were heated with alcohol for
50 hours on the water bath under a reflux condenser. Eight
grams of caustic potash were then added and the boiling con-
tinued until ammonia was no longer given off. The hot
alcoholic solution was poured off" and allowed to evaporate.
The salt that separated was treated with dilute nitric acid
and silver nitrate added. The precipitate was filtered ofif
and the filtrate was exactly neutralized with ammonia. The
white precipitate of silver fumarate was redissolved and re-

Synthesis of Fumaric and Maleic Acids 521

precipitated as in the first synthesis and dried at 105°-! 10°.
It turned slightly brown on drying. A quantitative deter-
mination of silver gave this result :

0.1734 gram of salt gave 0.15 16 gram AgCl.

Calculated for

C4H204Ag2 Found

Ag 65.43 65.80

The free fumaric acid was obtained by evaporating the
filtrate from the silver chloride, as described above. It sub-
limed at 201 ° in the melting-point apparatus.

Synthesis of Maleic Acid

Liquid acetylene diiodide was prepared and purified by
freezing as described by Keiser.* Ten grams of the pure
liquid diiodide and 7 grams of potassium cyanide in absolute
alcohol were boiled, under a reflux condenser, on the water
bath for 53 hours. The mixture turned pink, then dark
brown and finally black. A strong odor of isocyanide was
noticed on opening the flask. Eight grams of solid caustic
potash were added and the boiling continued for five hours,
until the odor of ammonia had disappeared. The solution
was filtered, a black residue remaining on the filter. This
consisted chiefly of iodides and carbonaceous matter, but
contained some maleate. The yellow filtrate was allowed to
evaporate. The salts that separated were dissolved in water
and the water solution evaporated in a desiccator. The
solid matter thus obtained was treated with just enough
dilute hydrochloric acid to decompose the small amount of
carbonate present and then the solution was warmed to expel
the gas. Barium hydroxide solution was now run in until
the liquid was exactly neutral. On standing a short time a
fine silky precipitate settled out. This precipitate was very
similar in appearance and behavior to barium maleate made
in the same way from pure maleic acid (Kahlbaum). The
synthetic barium maleate was filtered off and washed with
cold water until the wash water gave no test for barium,
chlorine or iodine. The salt was then dried at 105° for eight

\^ I THis'jorRNAL, 21, 264.

52 2 Reiser and McM aster

hours. A determination of barium gave 52.26 per cent.
As the calculated value for barium in anhydrous barium
maleate, 0^1120463, is 54.65 per cent, and in the salt with
one molecule of water of crystallization, C4H204Ba.H20, is
51 per cent., it was evident that our salt was a mixture,
or rather it was not dried sufficiently to completely convert
it into the anhydrous salt. We found at this stage of the work
that Vorlander* stated that barium maleate does not lose
water at 100° but begins to do so at 110° and that the water
is completely driven off at 130°- 135°. Another sample of
the barium salt was dried at iio°-i25° for 22 hours. Deter-
minations of barium now gave the following results :
I. 0.2434 gram salt gave 0.2276 gram BaSO^.
II. 0.1739 gram salt gave 0.1610 gram BaS04.
III. 0.3502 gram salt gave 0.3279 gram BaSOi.

Calculated for Found

CiHaOiBa I II III

Ba 54.65 54.98 54.45 55.04

A second synthesis of maleic acid was now made, with the
same quantities of substances, and the same method as before.
The barium salt was prepared and purified as described above.
It is to be noted that the fumaric acid is not precipitated by
barium hydroxide under the conditions under which we made
barium maleate. The barium maleate in this synthesis was
dried at 135° for three days. The analysis gave the following
results :

0.2506 gram salt gave 0.2340 gram BaSO^.

Calculated for

C4H304Ba Found

Ba 54.65 54.9

Some of this barium maleate was tested for fumarate by
dissolving it in hot water to which a little nitric acid had been
added, then neutralizing exactly with ammonia and adding
silver nitrate. No precipitate of silver fumarate was ob-
tained. Further, the free maleic acid from the barium salt,
made in this synthesis, was prepared by dissolving the salt

J Ann. Chem. (Liebig). 380, 192.

The Nitrile of Fumaric Acid 523

in hot water and adding sulphuric acid to the point of exact
neutrality, then filtering ofif the barium sulphate and evapora-
ting the filtrate. A melting-point determination of the white
residue showed that it melted at 131°. Pure maleic acid
(Kahlbaum) melts at 130°.

These experiments show conclusively that the solid acetylene
diiodide can be converted into fumaric acid and the liquid
isomer into maleic acid. This makes it highly probable that
these compounds have the configuration given above, namely,
that the solid diiodide has the fumaric or trans form, and the
liquid diiodide has the maleic or cis form.

Chemical Laboratory

Washington University

St. Louis, Mo.

THE NITRILE OF FUMARIC ACID

By Edward H. Keiser and J. J. Kessler

In 1890 E. H. Keiser,^ starting with solid acetylene di-
iodide, C2H2I2, succeeded in making fumaric acid by means of
potassium cyanide and caustic potash. Recently E. H.
Keiser^ and L. McMaster have repeated this synthesis of
fumaric acid and have also made maleic acid synthetically
by starting with the liquid acetylene diiodide and treating
it with potassium cyanide in alcoholic solution and then
boiling with alkali. In these syntheses the nitriles of the
acids must be formed, in the one case the nitrile of fumaric
acid and in the other that of maleic acid, and these are subse-
quently saponified by the alkali. Experiments made to
isolate these nitriles were not successful. We have, there-
fore, attempted to prepare them in another way, namely, by
heating fumaramide and ammonium maleate v/ith phos-
phorus pentoxide. We have succeeded in making the fumaric
nitrile, but have not gotten the maleic nitrile. We made the
fumaramide in two ways: in the one, we started with succinic
acid and made diethyl monobromsuccinate and treated this
with ammonia, thus obtaining fumaramide; in the other, we

' This Journal, 12, 99.
2 See preceding paper.

524 Keiser and Kessler

started with fumaric acid and made the dimethyl ester, which
was then converted into the amide by means of ammonia.
Both specimens of fumaramide were identical, and when each
was heated with phosphorus pentoxide the same sublimate
was obtained in each case. This sublimate we find has the
composition of fumaric nitrile.

Fumaramide from Succinic Acid

Portions of loo grams of succinic acid and 25 grams of red
phosphorus were ground together in a mortar until a fine
uniform mixture was obtained. This mixture was placed
in a dry two-liter flask, provided with a dropping funnel and
an exit tube joined to two Woulf bottles containing water to
absorb the hydrobromic acid. Bromine was now dropped
upon the mixture, the flask being cooled with water, and no
attempt was made to hasten the action, which was quite violent
at first. For each portion of 100 grams of succinic acid 454
grams of bromine were used. The reaction mixture was then
allowed to stand 12 hours before distilling off the excess of
bromine. This was done by disconnecting one Woulf bottle
and attaching a suction pump in its place and then heating
the flask on a water bath until the excess of bromine and the
hydrobromic acid were driven off. This is essentially tlie
method of brominating succinic acid recommended b)'^ Vol-
hard^ except that the reaction mixture was allowed to stand
longer and some of the details were modified.

In one case this brominated acid was removed from the
flask and weighed. The weight was found to be 300 grams.
This included also the metaphosphoric acid which had been
formed and which cannot be separated at this point since it is
soluble in the bromsuccinyl dibromide. This bromsuccinyl
dibromide fumes strongly in the air and will, if left in contact
with the air, be completely decomposed. When it is allowed
to stand in a closed bottle a white layer of succinic anhydride
rises to the top and the hea\'y oily bromide becomes quite
clear and may be poured off from the anhydride which forms
a crust on top of the oily liquid. The amount of anhydride

1 Volbard: Aan. Cbem. (Liebig), 242, 144.

The Niirile of Fumaric Acid 525

was not greater than 5 per cent, of the total product. In
one case more bromine was used but the increase in the yield
of bromsuccinyl bromide was not sufficient to justify this
procedure.

Without opening the flask from which the excess of bromine
and hydrobromic acid have been distilled 300 grams of ab-
solute ethyl alcohol are now added drop by drop. The hydro-
bromic acid given off is absorbed in water in the two Woulf
bottles. Care should be taken that the alcohol enters into
action as it drops into the flask. If the temperature is low
this may not always be the case and a considerable excess of
alcohol will accumulate. If now the flask be shaken a very
violet reaction will occur and an enormous amount of hydro-
bromic acid will be liberated in a short time, thus making the
operation a dangerous one. After all the alcohol has been
added, which requires several hours, the reaction mixture is
again allowed to stand overnight, then the excess of alcohol
and the hydrobromic acid are removed by distillation on a
water bath, a suction pump being used to diminish the pres-
sure. Water is now added in excess to the diethyl monobrom-
succinate. This dissolves the metaphosphoric acid and any
excess of alcohol that may remain. After thorough shaking the
water solution is poured off by decantation from the ester.
This washing is repeated several times, and finally tlie oily ester
is washed with water containing a small amount of bicar-
bonate of soda to remove the last traces of phosphoric and
succinic acids.

To the ester is now added dilute ammonia, made by mixing
450 cc. of water with 50 cc. of concentrated ammonia. The
action is allowed to go on for a week, the flask being thoroughly
shaken each day. The fumaramide is formed slowly. If the
mixture could be continually shaken it would, no doubt, in-
crease the velocity of the reaction. It is best to wait until
the reaction has been entirely completed and no oily ester is left.
During the earlier part of the work several separations of the
amide were made as it was formed from day to day, but it is
difficult to remove the traces of oily ester from it. If the re-
action is allowed to proceed to completion the whole may be

526 Keiser and Kessler

easily filtered and washed with water, then with alcohol and
ether. The amide is not very soluble in alcohol or ether but
it dissolves readily in hot water. It will be observed that until
the filtration of the amide all the reactions are carried out in
one flask without removing the product that is sought and it is
believed that this procedure helps to increase the yield. From
100 grams of succinic acid 30-40 grams of fumaramide are
obtained. The theoretical yield is 96.6 grams. This method
of preparing the amide cannot be hastened by using more
concentrated ammonia or by heating the mixture, as other
products are then formed.

Fuviaric Nitrite from Fumaramide

Five grams of perfectly dry fumaramide were mixed with
12-15 grams of phosphorus pentoxide. In order to insure a
thorough mixing, the fumaramide was first thoroughly pul-
verized and the powdered pentoxide added to the fine powder
and the whole thoroughly stirred together. As water is
rapidly taken up from the air no time must be lost in placing
the mixture in a crystallizing dish, covering it with a funnel
and heating the dish on a sand bath to about 120°. At this
temperature the reaction mixture turns black and a miniature
snow storm commences in the dish and the sides of the funnel
are quickly covered with a deposit of a fine needlelike sub-
limate. Care must be taken that the funnel does not become
too hot. Cool funnels must be put on from time to time until
no more sublimate is obtained. The first funnel contains
the nitrile crystals in best condition. The last has a deposit
which sticks to the sides of the glass and which results from the
fact that the temperature has risen so high that good crystals
are not obtained. The contents of the funnels are removed
by tapping them over a sheet of paper, when the crystals
fall out. The material that sticks to the sides may be removed
hy dissolving it in ether and recovered by evaporating the
ether.

Furaaric nitrile has a pleasant, pungent, characteristic nitrile
odor. It sublimes very easily. The needlelike crystals are ex-
tremely light and if they are inhaled while floating in tlie air they

The Nitrile of Fumaric Acid 527

prove very irritating to the mucous membranes. If a few of
the needles are placed in the bottom of a test tube and heated
they sublime without melting. If, howe\er, a larger amount
is used, especially material that has been obtained by the
evaporation of the ethereal solution, then the substance melts
at 96°. It was thought at one time that two substances were
produced in this reaction, one the needlelike crystals, very
volatile and subliming without melting, and another sub-
stance less volatile and melting at 96°. Further experiments
showed that they are identical. The best condition for the
formation of the needles is to raise the temperature of the
reaction mixture rapidly. With slow heating less is formed.
Starting with 5 grams of fumaramide a yield of 0.15 gram of
fumaric nitrile was obtained.

The analysis gave the following results:

Calculated for

C2H,(CN)2 Found

C 61.5 61.5 i

H 2.6 3.1

N 35-9 370-35. 9

The melting point of the substance prepared from fumar-
amide and from succinic acid, and also of that prepared directly
from fumaric acid as described below, is 96° and the boiling
point is 186° (760 mm.). It sublimes readily even below 100°,
and it is through this property that we have been able to
isolate it, as in all cases we have used temperatures lower than
the boiling point.

Fumaric nitrile is easily soluble in water, alcohol and ether.
When heated with alkalis, ammonia is evolved. The solution
of the substance in water gives but a very slight turbidity
with silver nitrate. If, however, the solution is heated but a
short time with an alkaki, and then acidified and silver
nitrate added, a copious precipitate of silver cyanide is ob-
tained. The alkali seems to completely decompose the ni-
trile with the formation of alkali cyanide. Attempts were
made to saponify the fumaric nitrile by heating it with dilute
sulphuric acid in a sealed tube and thus to obtain fumaric
acid but they were not successful. The amounts of nitrile
used were, however, very small.

528 Obituary

Fumaric Nitrile from Fumaric Acid

Five grams of Kahlbaum's fumaric acid was converted
into the methyl ester by dissolving the acid in 35 cc. absolute
methyl alcohol, conducting a stream of dry hydrochloric
acid gas into the solution and allowing it to stand overnight.
When the pure crystallized ester was allowed to stand in a
30 per cent, solution of ammonia for several days, it was
transformed into fumaramide.

The pure white fumaramide was filtered, dried, finely pul-
verized and mixed intimately with 15 grams of phosphorus
pentoxide. The mixture was quickly put into dry test tubes
which were then placed into long combustion tubes sealed
at one end. The open ends were then sealed and the part
containing the test tubes with the reaction mixture heated
in a Carius furnace for 48 hours at 130°. Needlelike crystals
collected in the cold ends of the tubes. These crystals con-
tained nitrogen, melted at 96° and had all of the properties
of the fumaric nitrile made from succinic acid.

Chemical Laboratory

Washington Univbrsity

St. Louis, Mo.

; OBITUARY

ALBERT LADENBURG

The death of Professor Ladenburg, at Breslau, on August
15th has been announced. Professor Ladenburg was born
at Manheim, July 2, 1842. Receiving his preparatory educa-
tion in the schools of his native city, he matriculated at Heidel-
berg in 1863, where he studied under Bunsen and received his
doctor's degree in 1868. He then worked for a time with
Friedel at Paris and with Kekul6 at Ghent. With the former
he began those investigations on organic silicon compounds
which have done so much in helping to establish the analogy
between this element and carbon, while Kekul6's influence
led|him to undertake his researches on benzene compounds
which ;resulted in the proof of the equivalence of the hydrogen
atoms in benzene and that there are two pairs of hydrogen
atoms arranged symmetrically with respect to each hydrogen
atom in the ring. Equally important was his proof of the
symmetrical constitution of mesitylene. His prism formula
for benzene, although now practically universally rejected,

Reviews 529

has been of great use in the development of chemistry in that,
as early as 1868, it showed the necessity of taking into account
steric considerations in the formulation of the constitution
of chemical compounds. He next turned to synthetic inves-
tigations, especially in the field of alkaloids, succeeding, in
1886, in preparing the first artificial alkaloid, coniine. He was
awarded the Hanbury gold medal for his services in the pro-
motion of research in drug chemistry. When van't Hoff and
LeBel advanced their theory of the asymmetric carbon atom,
Ladenburg became interested in the problem of racemic com-
pounds and has since had a large share in the development
of methods for the resolution of such compounds into their
active components. In the later years of his life he devoted
himself chiefly to inorganic problems, such as the determina-
tion of the atomic weight of iodine, of the molecular weight of
ozone, etc.

Besides the large number of experimental investigations
which Ladenburg carried out or directed, he was very prolific
in compilatory and editorial work. The first edition of his
well-known "Vortrage iiber die Entwickelungsgeschichte der
Chemie" appeared in 1869, when he was but 27 years old,
and the book has gone through four editions. Largely in
collaboration with his wife, he brought out the classical works
of Pasteur, Kekul6, Wurtz and Berthelot. Under his direc-
tion, a 13-volume dictionary of chemistry was issued. Within
the last three years he has published addresses on a multitude
of natural phenomena, many of them extending far beyond
the field of chemistry.

In 1872, Ladenburg was appointed "Professor Extraordin-
arius" at Heidelberg and the following year went to Kiel as
professor of chemistry. From there he was called to Bres-
lau in 1889, where he remained until forced by ill-health, in
1909, to give up active work. During his lifetime he was
awarded many honors, and he was a member of numerous scien-
tific societies.

REVIEWS

Trait6 Complet d'Analyse CniwiQue Appuqube aux Essais In-
DUSTRiKivS. Par J. Post, Profpsseur Honoraire k I'Universite de
Goettingue, et B. Neumann, Profe.sseur k la Technische Hochschule
de Darmstadt, avec la collaboration de nombreux chimistes et
specialistes. Deuxieme Edition franraise enti^rcment refondue.
Traduite d'apres la troisieme Edition allemande et augnient^e de nom-
breuses additions par G. Chenu, lug^nieur E. P. C. et M. Pei.i,ET,
Ing^nieur I. N. A. Tome premier, quatrifeme fascicule. Avec 210

530 Reviews

figures dans le texte et 36 planches hors texte, comprenant roi photo-
graphies. Paris: Librairie Scientifique A. Hermann et Fils. 191 1.
pp. iv + 492. Price, Fr. 18.

The translation of this well-known treatise follows for the
most part very closely the German edition, but many of the
chapters have been elaborated or added to by the translators.
Of the 490 pages, 236 are devoted to the subject of Metal-
lography. The translation follows the original of P. Goerens
which has been published separately as a book and of which
there is a good English translation. Some new material
has been added to the text which will add considerably to
the value of the book, and the number of illustrations has
been increased. The subject is treated extensively and in-
cludes chapters on pyrometry, preparation and use of polishing
powders, microscopic technique and apparatus, the establish-
ment of thermal equilibrium, diagrams of many binary and
a few ternary alloys. Many of the diagrams are discussed
in detail and illustrated with photomicrographs. The dia-
grams are more up-to-date than those originally used by
Goerens in his book, although some of the earlier less accurate
work is still retained. It is unfortunate that the diagrams
are reproduced with the German wording, except in a few
cases, although a glossary of terms is given at the end of the
chapter. The translators have added much of value, par-
ticularly in the section treating of pyrometry.

Much of the second half of the book is devoted to the con-
sideration of the analytical chemistry and physical properties
of the acids, sulphuric, hydrochloric and nitric, and the im-
portant commercial salts of these acids.

There is an appendix of 30 pages on spectrum analysis
written by A. de Gramont, of Paris, and a supplement to
the first volume dealing with the analysis of coal, gas, lubricants
and steel. Much of this is too briefly treated to be of value.

H. F.

New Idbas on Inorganic Chemistry, By Dr. A. Werner, Pro-
fessor of Chemistry in the University of Zurich. Translated, with the
author's sanction, from the second German edition by Edgar
PERCY Hedley, Ph.D., A.R-Sc.I., I,ecturer and Demonstrator in
Chemistry, University of Birmingham. London, New York, Bombay
and Calcutta : Longmans, Green & Co. 1911. Pages xvi -f 268.

In 1893 Werner published his first paper* on the constitu-
tion of inorganic compounds, which shed a flood of light upon
this field and showed how many apparently unrelated facts
could be correlated and harmonized. The constitution of
the metal-ammonia compounds (metalammines) and of the

> Z. anorg. ChetD.. 9, 267-330 (1893).

Reviews 531

double halides was shown to be essentially the same as that
of ordinary salts with water of crystallization. The leading
idea was that there are two kinds of valence. The "principal"
bonds serve to hold together the metal and acid radicals
in such compounds as CaClj, FeClj, etc. When anhydrous
salts take up water of crystallization or are transformed into
metalammines by taking up ammonia or analogous compounds,
the new groups entering the molecule are held by the secondary
"coordinate" bonds. Considering the elements in general,
the principal valence may have any value from one to seven,
and it often varies for a given element. The "coordination
number" is sometimes four, but in the great majority of cases
is equal to six. The groups held by the coordinate bonds
are believed to be in closer proximity to the central atom than
are the acid radicals. Such characteristic formulas as the
following are used to indicate which groups are held by the two
kinds of bonds: [CaCH^Oja, [Fe(H,0)e] CI3, [Co(NH3)e]Cl„
[PtajK^, [Co(N03)„]K3, etc.

What has been said gives only a slight hint, an inkling, of
the many ramifications of the theory of Werner. In every
paper he has written since the one already referred to, there
has been some new idea advanced or some new compounds
for which vacant niches were waiting have been described.
In 1905 there appeared the first addition of his book. Five
years later the second, enlarged edition was published, of which
the book now under review is a translation. "Werner's
theory " has for so long a time been a by- word in chemistry that a
translation into English ought to be of great value to stu-
dents not well enough acquainted with the German language
to unravel the unusually intricate sentences with which the
original abounds. This intricacy justifies the translator in
dividing sentences and using circumlocutions, but it does not
excuse the numerous incorrect translations and misstate-
ments in the book before us.

Among the minor faults are incorrectness and inconsist-
ency in spelling, and grammatical errors. Two typical sen-
tences illustrating the last point are found on page 16. " When
the molecular composition of compounds * * * * are exam-
ined," and "the combining values * * * for hydrogen is
variable."

Three times, on pages 4 and 6, we find "Blitz," as well as
in several other places, and only once do we see "Biltz."
"Rhubidium" is used three times on pages 7 and 8, as well
as elsewhere, though the correct spelling is used now and
then. "Paladium" occurs on pages 65 and 105. In a num-
ber of places "ammonium" is used for "ammonia," while

532 Reviews

"ammin" and "ammine" are used indiscriminately. Mis-
takes like these and the more serious ones now to be mentioned
are so numerous that in the translation there are only 63 out
of 268 pages on which some correction has not been necessary.

Some examples of incorrect translation may be given, if
for no other reason than to justify the criticisms of the re-
viewer.

On page 59, in discussing the change of active into inactive
amyl alcohol, we read, "it even takes place during the prepa-
rations of the active form," the German wording being "bei
der Darstellung des Alkoholates." In the next sentence we
learn that "heat transforms active lactic acid into the inactive
form," although Werner wrote "gewohnliches Lactid." Two
lines further on we read "when aspartic acid is heated to 170-
180° with any salt of hydrochloric acid." The German is
"aktive Asparaginsaure, als salzsaures Salz mit Wasser auf
170 bis 180° erhitzt."

By the translator's apparent forgetfulness of the old terms
"carburet" and "hydruret" he entirely misses the point
of the first two sentences on "Nomenclature" (page 74) in
which Werner advises the use of the ending "-ide" in place of
"-lir." He gives as examples of the usage to be avoided the
names "Hydriir" and "Carbiir." These are unfortunately
translated "hydride" and "carbide."

On page 81, near the bottom, we read "different hydrates
(acids) and salts may be derived from the same acidic anhy-
dride of an oxide." This is hardly an intelligible translation
of "leiten sich von gewissen als Saureanhydride wirkenden
Oxyden verschiedene Hydrate (Sauren) und Salze ab."

The words (page 85) "polyhalides, i. e., compounds the pos-
itive radical of which consists of several halogen atoms"
give no very clear idea of "Verbindungen, die auf eine Valenz
der positiven Radikale nicht nur ein, sondern mehrere Halo-
genatome enthalten."

On page 198 we read "This idea explains the enormous
increase in electrical conductivity which water undergoes on
the solution of ammonia." Werner wrote "Auf Grund dieser
Anschauung wird das abnorme elektrochemische Verhalten
des Amraoniaks in wasseriger Losung, welches in der geringen
Leitfahigkeit desselben zum Ausdruck kommt, verstandlich."

Page 236: "Isomeric with the metaphosphates is a series
of salts in which four of the six metals are replaceable." which
is too free a translation of "Isomer mit den Metaphosphaten
ist eine Salzreihe in der auf sechs Metaphosphatreste vier
substituierbare einwertige Metallatorae kommen."

Reviews 533

Many more examples of this sort might be given but enough
has been said to show that the book was translated and pub-
lished in too great a hurry. Werner has written with such
evident care and precision that it is a pity that the transla-
tion should contain so many passages often utterly mislead-
ing, which, when they may not be ascribed to carelessness,
are apparently due to a misapprehension of the meaning of
the original.

We do not wish to condemn the book utterly, for in spite
of its many faults, which it will be possible to correct in a future
edition, it is the only source of information regarding Wer-
ner's ideas for those who read German with difficulty, or not

at all. C. E. Waters.

Kapii,i,archEmik. Eine Darstellung der Chemie der Kolloide und
verwandter Gebiete. Von Dr. Herbert Freundlich, Privatdozent
an der Universitat Leipzig. Leipzig : Akademische Verlagsgesell-
schaft tn. b. H. 1909. S. viii -f 591. Preis ; brosch., M. 16.50 ; geb.,
M. 17.50.

This volume gives a well-balanced treatment of the entire
field of capillary chemistry. Its contents fall under three
divisions: A. The general properties and deportment of'
surfaces of separation. B. Disperse systems. C. The tech-
nical and physiological significance of capillary chemistry.
The first two divisions comprise the main substance of the work,
to the last but eight pages are devoted. An appendix gives
in tabular form the methods of preparation and the properties
of sols in both aqueous and nonaqueous media.

Under the first division the several surfaces formed by the
phase pairs liquid — gas, solid — gas, liquid — liquid and solid^ —
liquid are treated of in detail. Chapters are devoted to
capillary electrical phenomena and to the thickness, stability
and other properties of the surface film.

Following Wo. Ostwald, the author describes as disperse
systems those in which the surfaces of contact between the
phases concerned are of very great area. In this division also
the systems are classified according to the general physical
nature of the phases that are in contact. The properties of
clouds, smokes and foams are here briefly but adequately
treated, while over 200 pages are given to a clear and most
interesting discussion of the more broadly significant sols and
gels.

A brief outline is all that is attempted of the technical bear-
ings of the subject. Its importance in such processes as dye-
ing, tanning and photography, and in bacteriology, is pointed
out.

534 Reviews

A central conception of the book, that of adsorption, is in-
troduced early, and is largely used as a unifying principle.
The approximate adsorption law, due primarily to Gibbs,
and elaborated by Milner, is expressed by the equation

u = —c/RT(da/hc)

o being the surface tension of a solution, c the concentration
and u the excess in concentration, per unit surface, in the
surface layer. From this law and the empirical relation

CTj — a2 = sc^
Oi and 02 being the tensions of solvent and of solution and s

and - being constants, the relation

u = a c"
is gained, and in its several forms is shown to have a remark-
ably broad application. This is brought out in an especially
satisfactory manner in the case of the adsorption of gases by
solids.

To write comprehensively and clearly on a subject of great
complexity and that is undergoing unusually rapid develop-
ment is a difficult task. This the author has performed with
signal success. The book is one that calls for, but richly re-
pays, careful reading; it is an excellent treatment, by a fore-
most worker, of a subject of general interest. Geo. h. burrows.

The Chemistry of the Coal-Tar Dyes. By Irving W. Fay, Ph.D.
(Berlin), Professor of Chemistry, Polytechnic Institute, Brooklyn, N.
Y. New York: D. Van Nostrand Company. 1911. pp. vi -f 467.
Price, 1:4.

This book i? intended for students and dyers who have a
good knowledge of general chemistry and some knowledge of
organic chemistry. After several introductory chapters deal-
ing with the chemistry of the benzene series as related to dye-
stuffs, the author takes up in eleven chapters the chemistry
of the various groups of dyestuffs and of a number of the dyes.
Chapters on the seven colors allowed in food in this country
and on mordants are also included, and there is a chapter con-
taining 44 laboratory experiments for the preparation of various
dyes or materials used in their manufacture. Diagrams and
formulas are used skillfully to bring out complex chemical
relations. Students and others who desire a knowledge of
the chemistry of dyestuffs will find this book very useful.

G. s. F.

Reviews 535

Die Konstitution der Chinaai^kaloide. Von Prof. Dr. Ezio
COMANDUCCI, Neapel. Deutsche Ausgabe durchgesehen von Dr. W.
Roth (Cothen). Mit 5 Texttafeln. Sammlung chemischer und
chemisch-technischer Vortrage, begriindet von F. B. Ahrens,
herausgegeben von Prof. Dr. W. Hkrz, Breslau. XVI Band, 4/7
Heft. Stuttgart: Verlag von Ferdinand Enke. 1911. pp. 141-306.

The monograph is a supplemented translation of Coman-
ducci's article on the subject published in the Nuova Enciclo-
pedia di Chimica by I. Guareschi, Vol. 6, pp. 430-505. It
contains an exhaustive discussion of those of the forty, more
or less, Cinchona alkaloids, the constitution of which has been
established with considerable certainty. The book consists
of two parts: A short general introduction of 37 pages, and
a special part of 127 pages. In the first part the experimental
data are arranged with a view to clearly demonstrate the
presence of certain nuclei, groups and Unkings in the crystal-
lized Cinchona alkaloids, regardless of the historical sequence
in which these data were obtained. In the second part the
subject is treated historically, the alkaloids being taken up
one by one and the gradual accumulation of evidence upon
which is based the generally accepted constitution of these
alkaloids being discussed. A very useful feature of the book'
is the arrangement of the numerous transformation products
of the alkaloids into five tables in a way to show at a glance
their genetic relations. From the point of view of workman-
ship the book is excellently gotten up. For those interested
in alkaloidal chemistry it will prove of great value, as it brings
together into a comparatively small volume all the results
and the whole literature on one of the most important groups
of natural bases. As a reference work the book could be
made still more valuable by supplying it with a good index
or at least a table of contents. The book is also in need of a
list of errata, of which there are quite a number, though as a
rule they are not very important. To mention only a few of
these: On page 67, top line, the ethyl ether of apocinchene
is called the ethyl ester of apocinchene; further down on the
same page, line 24, from the bottom, />-oxycinchene oughtHo
be replaced by ^-oxyapocinchene ; still further on the same
page, line 8 from the bottom, there ought to be "an einem der
zwei Kohlenstoffatomen," instead of "an diesen zwei Kohlen-
stofF atomen ; " on page 180 the formulas of hydrobromcincho-
nine and hydrochlorcinchonine are wrong; the same is true of
the formula of chinchonine hydrobromide on page 183, in the
paragraph on tautocinchonine. A much more serious error is
to be found in the opening sentences on page 178, where it is
stated that "cinchonine and cinchonidine yield one and the

536 Reviews

same desoxycinchonine." As desoxycinchonine melts at
90-92°, while desoxycinchonidine melts at 61°, the above
statement is evidently an error. In fact the nonidentity of
the two desoxy bases, in connection with the assumption
(based upon the identity of the cinchene obtained from cin-
chonine with the cinchene obtained from cinchonidine) that
the isomerism of cinchonine and cinchonidine is due solely
to the asymmetric carbon atom to which is linked the OH
group, is used by Miller and Rohde as evidence of the hydroxyl
belonging to a tertiary alcohol group, since a secondary CH . OH
group would become by reduction a CHg group, and the asym-
metry of the carbon atom would be destroyed. The book is
completely up-to-date, and should be in every well equipped
library. h. m. gokdiit.

Vol.. XLVI December, 19 1 1 No. 6

AMERICAN

CHEMICALJOURNAL

[Contributions from the Sheffield Laboratory of Yale University]

CXCV.— RESEARCHES ON PYRIMIDINES

THE CONDENSATION OF UREA AND GUANIDINE

WITH ESTERS OF ALLYLMALONIC AND SOME

ALKYL-SUBSTITUTED ALLYLMALONIC

ACIDS

[fifty-fourth paper]

By Trbat B. Johnson and Arthur J. Hill

It was stated in a previous paper^ that a derivative of
allylmalonic acid,

CH2 : CHCH^CHCCOOH)^,

in which the carboxyl groups are linked in such a manner as to
prevent lactone formation after the addition of halogens
or halogen acids to the ethylene bond, would be of value for
synthetical purposes. With the object in view of preparing
a compound of this character, we investigated the action of
diethyl allylmalonate on thiourea. Instead of reacting nor-
mally, as we expected, to form allylthiobarbituric acid (I),
the thiourea combined with this ester in a unique manner,
giving the first representative of a new class of cyclic com-
pounds, viz., /(-amino- a-keto-/?-carbethoxy-(5-methyltetrahydro-

1 Johnson and Hill: This Journal, 46, 356 (1911).

538 Johnson and Hill

hexathiazole (II). On the other hand, alkyl derivatives of
diethyl allylmalonate — diethyl diallylmalonate and diethyl
benzylallylmalonate — reacted in an entirely diffeient manner
and combined with thiourea, in the presence of sodium ethylate,
forming, apparently, /--lactone thioureas.

Because of this abnormal behavior of thiourea towards
allylmalonates, it was therefore of interest to determine whether
barbituric acid derivatives are formed by condensation of
these same esters with urea and guanidine. We now find
that urea condenses normally with diethyl allylmalonate,
diethyl diallylmalonate and diethyl benzylallylmalonate, in
the presence of sodium ethylate, forming smoothly allylbar-
bituric (III), diallylbarbituric (V) and benzylallylbarbituric
(IX) acids, respectively. AUylbarbituric (III) and diallyl-
barbituric (V) acids were hydrolyzed smoothly, when heated
with potassium hydroxide, giving the alkali salts of allyl-
malonic and diallylmalonic acids. On the other hand, all
attempts to hydrolyze benzylallylbarbituric acid (IX) smoothly
under the same conditions were unsuccessful. We also ex-
perienced this same difficulty when we attempted to hydro-
lyze benzylallylmalonylguanidine (VIII). Pure allylbenzyl-
malonic acid was not obtained in either case and the product
of hydrolysis was an oil of indefinite composition. The struc-
ture of this pyrimidine (IX) was established by the fact that
the same compound was formed by the action of allyl iodide
on the silver salt of benzylbarbituric acid (X).^

Guanidine condensed normally with diethyl allylmalonate
and diethyl diallylmalonate, in alcohol solution and in the
presence of sodium ethylate, giving allylmalonylguanidine
(IV) and diallylmalonylguanidine (XI), respectively. Both
of these pyrimidines underwent hydrolysis smoothly, when
heated with potassium h)^droxide, giving allylm.alonic and
diallylmalonic acids, respectively (potassium salts). Diethyl
benzylallylmalonate, on the other hand, condensed with
guanidine, forming allylbenzylim.inomalonuric acid (VI) or,
which is more probable, its inner cyclic salt (VII). This
latter structure (VII) is supported by the fact that the con-

1 Conrad and Guthzeit: Ber. d. chem. Ges., 16, 2846.

Researches on Pyrimidines

539

O W O

X
o

— O-

-o-

X ^

K X

^— o— ^

X
o

— a— y

tm
^

o X

X
o

8-0

540 Johnson and Hill

densation product has no melting point below 300°, while an
iminomalonuric acid would be expected to evolve carbon di-
oxide on heating and be con^'erted into an acylguanidine.
The corresponding dialkylmalonuric acids which have been
examined all behave in this manner, when heated, and are
transformed into acylureas.^

The iminomalonuric acid or the cyclic salt(?) is unstable in
the presence of acids. When it was suspended in an excess
of cold dilute hydrochloric acid it was immediately trans-
formed into the hydrochloride of benzylallylmalonylguani-
dine (VII). This salt is dissociated by water and converted
into the free base (VII), melting above 300°. The constitu-
tion of this pyrimidine was established by the fact that the
same compound was formed by the action of benzyl iodide
on the silver salt of allylmalonylguanidine (IV). Therefore
the silver salts of benzylbarbituric acid and allylmalonyl-
guanidine react with alkyl halides in a similar manner, and
the alkyl radicals attach themselves to the carbon atom in
the 5 -position of the pyrimidine ring.

The study of allylpyrimidines will be continued.

EXPKRIMENTAL PART

yAllylmalonylurea {Allylbarhituric Acid),

NH- CO

I I

CO CHCH.CH :CH,.— Three and six-tenths grams of so-

II"

NH CO

dium were dissolved in 40 cc. of absolute alcohol and 4 . 5 grams
of urea then dissolved in the solution. One molecular pro-
portion of diethyl malonate (15 grams) was then poured
into the solution and the mixture shaken vigorously, when a
magma of a colorless sodium salt was formed immediately.
This was heated at 100° for 5 hours to complete the reaction
and the alcohol then expelled by heating on the steam bath. The
sodium salt was dissolved in the least possible quantity ■; of
cold water, and dilute sulphuric acid added in slight excess.
This pyrimidine separated in flesh-colored crystals. -> It was

J Fischer and Dilthey: Ann. Chcm. (Liebig), 336, 334 (1904).

Researches on Pyrimidines 541

purified by crystallization from hot, absolute alcohol and
separated, on cooling, in almost colorless plates which melted
at 167° to an oil without effervescence. The pyrimidine is
soluble in boiling water and insoluble in benzene. The yield
of purified material was lo.o grams. Analysis (Kjeldahl) :

Calculated for Found

CjH^OsNz I II

N 16.66 16.72 16.72

Hydrolysis of Allylharhituric Acid with Potassium Hydroxide.
— Two grams of the barbituric acid were hydrolyzed by dis-
solving in 8 . o cc. of a 50 per cent, potassium hydroxide solu-
tion and then heating for 2 hours at i3o°-i50°. After acidi-
fying with hydrochloric acid the solution was then evapora-
ted at 100° to remove water and the residue triturated with
warm absolute alcohol to dissolve the allylmalonic acid. On
cooling, this acid separated. In order to identify it, it
was dissolved in dilute sodium hydroxide solution and cal-
cium chloride added, when its characteristic calcium salt
separated. Calcium determination:

0.2457 gram substance gave 0.0761 gram CaO.

Calculated for

C6H604Ca Found

Ca 21.98 22.14

Some of this calcium salt was warmed in dilute hydro-
chloric acid solution. After cooling, the solution was then
extracted with ether and the ether evaporated. We obtained
in this manner the pure allylmalonic acid, which crystallized
from benzene and melted at 103°.

NH— CO

I I

3-Allylmalonylguanidine, HN : C CHCHjCH : CH..2H,0.

I i

NH CO

— This pyrimidine was prepared by dissolving three molecu-
lar proportions of sodium (3 grams) in 45 cc. of absolute alco-
hol and adding to this solution 4 . o grams of guanidine hydro-
chloride and 8.0 grams of diethyl malonate. An immediate
reaction took place with separation of a colorless sodium salt.

542 Johnson and Hill

After heating for 5 hours to complete the reaction the excess
of alcohol was then removed by evaporation at 100°. The
salt was dissolved in a little cold water and sulphuric acid
added in only slight excess, when this pyrimiduie separated
in beautiful pink prisms. It crystallizes from acetic acid in
characteristic hexagonal tables which melt at 265°-266°.
The compound is insoluble in alcohol and benzene but soluble
in water. It contains two molecules of water of crystalliza-
tion, which it loses when heated at 120°. Water determina-
tion:

0.8600 gram substance lost 0.1441 gram H,0.

Calculated for
C7H9O2N3.2H2O Found

H2O 17.73 16.76

Nitrogen determination (Kjeldahl) :

Calculated for Found

C7H9O2N3.2H2O I II

N 20.69 20.56 20.56

5,yDiallylmalonylurea {Diallylharhituric Acid),
NH CO

I I .CH^CH : CH2

CO C<^ . — This pyrimidine was prepared by

I I \CH3CH : CH2

NH CO

condensation of urea (2.4 grams) with diethyl diallylmalonate
(10 grams) in the same manner as described in the preceding
condensation. The yield was 8 . o grams. It crystallizes from
50 per cent, alcohol in characteristic, rhombohedral crystals
which melt at 173° to a clear oil. The crystals are colorless.
The pyrimidine is very soluble in cold alcohol and in warm
water, but only moderately soluble in benzene. Nitrogen
determinations (Kjeldahl) :

alculated for
C,oH,20aN2

3und

13.46

13.26

13.2

Hydrolysis of Diallylbarbituric Acid with Potassium. Hy-
droxide.— ^This pyrimidine was saponified by dissolving it (3
grams) in 50 per cent, potassium hydroxide solution (10 cc.)

Researches on Pyrimidines 543

and then heating for 4 hours at 145°. After heating, in an
open dish, at 100° to remove ammonia the solution v^ras cooled
to 0°, acidified with hydrochloric acid, and finally extracted
with ether. The ether solution was washed with water to
remove hydrochloric acid, dried over sodiam sulphate and
the ether evaporated, when we obtained an oily product which
immediately crystallized on cooling. This substance was
chiefly diallylmalonic acid, but contaminated with some ma-
terial which was not identified. When this mixture was dis-
solved in dilute ammonia and an excess of silver nitrate
solution added the colorless, granular silver salt of diallyl-
malonic acid separated. After washing with water, alcohol,
and finally with ether, it was dried for analysis in a desiccator
over concentrated sulphuric acid. Silver determinations:
I. 0.1700 gram substance gave 0.0915 gram Ag.
II. 0.2500 gram substance gave 0.1347 gram Ag.

Calculated for Found *

C9Hio04Ag2 I II

Ag 54 27 53 82 53.80

NH CO

I I .CHjCH rCH^

5,5-Diallylmalonylguanidine,^^ '-C C\ . —

1 I \CH2CIi :CH2

NH^ — ^CO
This was prepared by condensation of guanidine (9 grams of
the hydrochloride) with diethyl diallylmalonate (22 . 7 grams)
in the presence of sodium ethylate. The yield of the crude
pyrimidine was 18.0 grams. It is soluble in acetic acid and
insoluble in water and alcohol. It crystallizes from acetic
acid in characteristic, colorless, rhombohedral prisms
which do not melt below 300°. We experienced great diffi-
culty in obtaining consistent results by the Kjeldahl method,
therefore the nitrogen was determined by combustion (Dumas) :

Calculated for
CioHiaOaNs Found

N 20.29 20.36

Hydrolysis of 5,5-Diallylmalonylguanidine with Potassium
Hydroxide. — This pyrimidine was hydrolyzed under similar

544 Johnson and Hill

conditions as those employed in the hydrolysis of 5,5-diallyl-
barbituric acid. After acidifying with hydrochloric acid the
diallylmalonic acid was extracted with ether. It was ob-
tained in colorless crystals after evaporation of the ether and
melted at 133°.

5,5-Allylbenzylmalonylurea {5-Allyl-5-henzylharbituric Acid),
NK CO

1 I /CH3C«H,

CO Q.{ . — ^This was prepared by condensa-

i I \CH2CH:CH2

NH CO

tion of urea (1.4 grams) with diethyl allylbenzylmalonate (7
grams). The yield was 6 grams. The pyrimidine crystal-
lizes from dilute alcohol in characteristic barrel-shaped prisms
which melt at 198° to an oil. It is soluble in cold ether and
absolute alcohol and insoluble in water. It did not contain
water of crystallization. Nitrogen determinations after dry-
ing the pyrimidine at 100° (Kjeldahl) :

Calculated for
CuHuOaNz

Found

10.85

10. 70

10.67

Hydrolysis of Allylbenzylbarhituric Acid with Potassium
Hydroxide. — One and four-tenths grams of this pyrimidine
were dissolved as usual in 10 cc. of 50 per cent, potassium
hydroxide solution and heated for 3 hours at 135°. Ammonia
was evolved, and also carbon dioxide on making the solution
acid. The solution was acidified with hydrochloric acid
and tlaen thoroughly extracted with ether. The ether solu-
tion was washed with water, dried over sodium sulphate
and the ether then e"s^aporated, when we obtained a thick
oil, having acid properties, which showed no signs of crystal-
lizing on long standing. Assuming that we had in hand allyl-
benzylmalonic acid we therefore dissoh^ed the product in a
slight excess of dilute ammonia and then added silver nitrate
in excess. A colorless, granular silver salt separated at once.
It was washed with water, alcohol and ether and then dried
for analysis in a desiccator over concentrated sulphuric acid.
A silver determination on this salt gave a result lower than

Researches on Pyrimidines 545

the calculated value for a disilver salt of allylbenzylmalonic
acid and indicated a mixture of this salt with the mono-
silver salt of the lactone^ of this dibasic acid. Silver de-
termination : •

o. 1076 gram substance gave 0.0400 gram Ag.

Calculated for
CisHizOtAga CsHiaOiAg Found

Ag 48,22 31.7 37.2

Because of this low result, we therefore repeated the hy-
drolysis twice under different conditions. We first heated
with 50 per cent, alkali for 4 hours at 150° and in the second
experiment for the same length of time at 170°, but even un-
der these conditions we obtained an oily acid which would
not crystallize on long standing. Furthermore, the analytical
results for silver were higher than those obtained in the first
experiment, but still were consistently much lower than the
calculated value for a disilver salt. For example, we found'
45-39. 45 14. 45-55 and 45. 58 per cent, of silver, while the
calculated value is 48. 22.

Since this compound did not undergo hydrolysis smoothly,
giving the corresponding malonic acid, as observ^ed in the two
previous cases, we therefore established its structure by its
synthesis in another manner. For example: Ten and five-
tenths grams of benzylbarbituric acid^ and 2 . 9 grams of potas-
sium hydroxide were dissolved in warm water and mixed with
an aqueous solution of silver nitrate (9.4 grams of the salt).
The silver salt of the pyrimidine deposited at once. This
was separated by filtration, washed with water, alcohol and ether
and then thoroughly dried in a desiccator over sulphuiic acid.
This dry salt was then suspended in anhydrous ether and
digested with a molecular proportion of allyl iodide for 8-9
hours. Silver iodide was foimed. The ether solution was then
filtered and the ether allowed to evaporate, when a solid sub-
stance was obtained. This was dried on a porous plate to re-
move traces of allyl iodide and finally washed with ether.
The compound crystallized from dilute alcohol in the charac-

' Johnson and Hill: Loc. cil.
2 Conrad: Loc. cil.

546 Johnson and Hill

teristic, barrel-shaped prisms of allylbenzylbarbituric acid
and melted at 197°. A mixture of this compound and the
barbituric acid derivative melted at i97°-i98°. The allyl
group therefore entered the 5-posi?tion of the pyrimidine
ring. An attempt to alkylate this pyrimidine by digestion
of the silver salt in absolute alcohol with pure distilled allyl
iodide was unsuccessful. Hydriodic acid apparently was gen-
erated, under these conditions, and decomposed the salt,
forming silver iodide and the unaltered benzylbarbituric
acid.

Condensation of Diethyl Allylhenzylmalonate with Gtianidine

A llylhenzylimidomalonuric A cid,

Nil— CO NH CO

I ] /CHjCgHg I I /CHjCgHg

HN :C C< or HN :C C<

I 1 ^CHjCH : CH2 1 I ^CHjCH : CH^

NH2 COOH NH3O.CO

— One and sixty-five hundredths grams of sodium were dis-
solved in 50 cc. of absolute alcohol and 2.37 grams of finely
pulverized guanidine hydrochloride sifted into the solution.
Seven and two-tenths grams of diethyl allylhenzylmalonate
were then added and the mixture warmed gently at 100°,
when a colorless sodium salt separated. The heating was con-
tinued for 5 hours and the excess of alcohol then removed.
This compound v;as obtained by dissolving the salt in water
and then adding a slight excess of sulphuric acid. It crys-
tallizes from acetic acid in needlelike prisms which do not
m.elt below 300°. The compound is insoluble in ether, alco-
hol, water and benzene. The yield was 9 grams. Analyses
(Kjeldahl) :

Calculated for Found

CuHiTOaNs I II III IV V

N 15-7 15-3 15-32 1565 1567 15-65

This interesting compound is especially characterized by
its behavior towards hydrochloric acid. Seven- tenths of a
gram of the compound, finely pulverized, was placed in a
dry test tube and 5 cc. of cold dilute hydrochloric acid poured

Researches on Pyrimidines 547

upon the dry powder. The compound began to undergo a
change immediately at ordinary temperature. It first melted
and deposited on the bottom of the test tube as a thick gum.
There was no apparent evolution of heat during this trans-
formation. This oily product finally completely solidified,
forming beautiful prismatic crystals. After being allowed
to stand in the acid for about 5 minutes it was filtered off and
without further treatment placed in a desiccator over sul-
phuric acid and potassium hydroxide and allowed to dry.
This substance gave a strong test for chlorine and a nitrogen
determination (Kjeldahl) indicated a basic salt of allylbenzyl-
malonylguanidine :

Calculated for
(Ci4His02N3)2HCl Found

N 15.20 15.03

This salt is completely dissociated by water. When warmed
with water, the solution gave a strong test for free hydro-
chloric acid and the pyrimidine separated as a colorless pow-
der.

5,S-Allylbenzylmalonylguanidine,

NR CO

I I /CHAH3

HN : C C <' . — This same pyrimidine was also

I I \CH2CH : CH2

Nil CO

obtained by alkylation of 5-all3dmalonylguanidine with benzyl
iodide. For example: Allylmalonylguanidine was dissolved
in water containing one molecular proportion of potassium
hydroxide and the silver salt precipitated by addition of the
required amount of silver nitrate solution. This was filtered
off, washed free from inorganic salts with water and alcohol
and then dried in a desiccator over sulphuric acid. Twelve
grams of this dry silver salt were suspended in 50 cc. of dry-
ether and digested, on the water bath, with 9.9 grams of
benzyl iodide. Silver iodide began to form almost imme-
diately and after 5 hours' digestion the reaction was com-
plete. The pyrimidine separated with the silver iodide and
when the ether was evaporated only a small amount of unal-
tered benzyl iodide was obtained. In order to obtain the

548 Johnson and Hill

pyrimidine the mixture of silver iodide and pyrimidine was
digested with acetic acid and the solution filtered and cooled.
The pyrimidine then separated as a fine, colorless powder
which did not melt below 300°. It was absolutely identical
with the product formed by the action of hydrochloric acid
on allylbenzyliminomalonuric acid. The pyrimidine is insolu-
ble in water and alcohol. Nitrogen determinations (Kjel-
dahl) :

Calculated for
C14H15O2N3

Found

16.34

16. I

16.6

Hydrolysis of Allylhenzylmalonylguanidine with Potassium
Hydroxide. — As in the case of allylbenzylbarbituric acid,
attempts to obtain pure all^dbenzylmalonic acid by hydroly-
sis of this malonylguanidine, with alkali, were unsuccessful.
Two grams of the pyrimidine were dissolved in 6 cc. of 50
per cent, potassium hydroxide solution and heated for 2.5
hours at 130°. The acid was separated as in the hydrolysis
of allylbenzylbarbituric acid, and was obtained as an oil which
would not crystallize. A silver salt was made in the usual
manner and the following result obtained for silver:

Calculated for
Ci3H,204Ag2 Found

Ag 48.22 38.48

Several attempts were then made to obtain a definite acid
by heating the pyrimidine with 50 per cent, alkali from 2 to
5 hours at 150°. In every case we failed to obtain a silver
salt that contained over 43 per cent, of silver. The results
of six different determinations on salts prepared after \'arious
hydrolyses are as follows :

Calculated for Found

Ci.^H,204Ag2 I II III IV V VI

Ag 48.22 42.56 42.66 41.16 41.03 42.70 42.83

CH, : CHCH^v

Allylbenzylmalonic Acid, >C(COOH),.— This

CeH.CH/
Was prepared by dissolving 2 grams of its ethyl ester and 3
molecular proportions of potassium hydroxide (2.7 grams)

i,j,yTriiod-2-Brom-4,6-Dinitrobenzene 549

in 50 per cent, alcohol and heating the solution for 14 hours
on the steam bath. After cooling, hydrochloric acid was
added in excess and the acid solution repeatedly extracted
with ether to remove the malonic acid. After evaporating
the excess of ether the acid was obtained as a thick oil which
did not crystallize. It was dissolved in dilute ammonia and
the silver salt prepared by addition of silver nitrate solution.
It separated in a granular condition and, after being washed
with water, alcohol and ether, was dried for analysis in a desic-
cator over sulphuric acid. Silver determination:

0.3310 gram substance gave on ignition 0.1580 gram Ag.

Calculated for

C,3H,204Ag2

Found

48.22

47-74

New Haven, Conn.

August 11, 1911

[Contributions from the Chemical Laboratory of Harvard University]

1,3,5 - TRIIOD - 2 - BROM - 4,6 - DINITROBENZENE AND
SOME OF ITS DERIVATIVES^

By C. Loring Jackson and H. E. Bigelow

In 1888, W. S. Robinson and one of us^ found that sodic
malonic ester converted tribromdinitrobenzene, CgHBr3(N02)2,
into bromdinitrophenylmalonic ester,

CeH2Br(N02)2CH(COOC2H5)2,

by the replacement of one atom of bromine by hydrogen and
another by the malonic ester residue. The further investiga-
tion of this subject brought to light many similar cases of re-
placement of halogens by hydrogen, and in the present paper
we describe our study of i,3,5-triiod-2-brom-4,6-dinitroben-
zene, C6l3Br(N02)2, which, when treated with sodic malonic
ester in the cold, is converted into the compound

CeHl2Br(N02)2

1 The work described in this paper formed part of a thesis presented to the Faculty
of Arts and Sciences of Harvard University for the degree of Doctor of Philosophy
by Harold E. Bigelow.

2 This Journal, 11, 93 (1889).

550 Jackson and Bigelow

with acetylenetetracarbonic ester, C2H2(COOC2H5)j, as the
secondary product; while, if the reaction is carried on hot,
the substituted malonic ester, C8HIBr(N02)2CH(COOC2H5)2, is
formed. This discovery proves the incorrectness of the re-
actions^ by which the formation of the acetylenetetracarbonic
ester has been explained hitherto, since these are impossible
without free malonic ester, which was excluded from these
experiments by the use of a slight excess of sodic ethylate and
could not have been set free from sodic malonic ester, as in
the earlier reactions, because no acid substance was present.
It became necessary therefore to explain the formation from
sodic malonic ester of iodmalonic ester (the mother sub-
stance of the acetylenetetracarbonic ester), and the only
hypothesis we have found for this consists in assuming that
the sodic malonic ester reacts in the enol form,

CH - (CONaOC2H5)— (COOC2H5),

and that the action consists in the addition of the I and
— C6l2Br(N02)2 radicals at the double bond, when the more
acid of these will undoubtedly attach itself to the carbon
carrying the ONa group, and this may well be the substituted
phenyl loaded with the highly negative nitro groups and
halogens. In that case we shall have the compound

COOCH.,
/
ICH
\ /OC2H,
C^ONa
Xl2Br(N02)2

On acidification the hydrogen of the hydroxyl might combine
with the substituted phenyl, giving the compounds

CeHl2Br(N02)2 and CHI(COOC2H3)2,
which, reacting with the excess of sodic malonic ester, will
give acetylenetetracarbonic ester, the two products actually
obtained. This hypothesis not only explains the formation
of the acetylenetetracarbonic ester, but also gives the first

' Jackson and Moore: This Journal, 12, 7. Jackson: Ibid., 307 (1890).

i,j,^-Triiod-2-Broni-4,6-Dinitrobenzene 55 1

explanation of the replacement of a halogen by hydrogen
under these conditions.

Of the suppositions which make up this hypothesis, perhaps
the least probable is the splitting off of the — C6l2Br(N02)2
group with the hydrogen of the hydroxyl, as this necessitates
breaking apart two atoms of carbon, one of which forms part
of a benzene ring. Some reasons have occurred to us for be-
lieving such an action might take place, but we think it un-
wise to discuss these, or other arguments in favor of our hy-
pothesis, before it has been brought to the test of experiment
by work now in progress in this laboratory.

If this hypothesis should be accepted it will also explain
the similar behavior vv^ith acetacetic ester, but we do not see
how it can be applied to the replacements of halogen by hy-
drogen under the influence of sodic alcoholates, unless it is
assumed that the intermediate addition compounds in these
cases contain quadrivalent oxygen, and are therefore analogous
to the quinhydrones as formulated by Richter.^

The work described in this paper, with that also on com-
pounds of iodine by Langmaid and one of us,- has established
the order of the halogens in regard to replacement by hydro-
gen rather than by the malonic ester radical as follows:
Chlorine has the least tendency of this sort, since only one
case^ of it has been observed, so far as we know; bromine
undergoes the replacement by hydrogen as easily as that by
the malonic ester residue, and iodine with more ease; so that
the tendency toward the hydrogen substitution increases
as the acidity of the element decreases, which is a strong
argument in favor of our hypothesis.

In preparing i,3,5-triiod-2-brom-4,6-dinitrobenzene,

Cel3Br(N02)2,

it was found that a persistent impurity of triioddibromnitro-
benzene was formed, if the triiodaniline used contained the
dark-colored substances with which it is usually contamina-
ted, and further, that these are produced either by an excess

' Ber. d. chem. Ges., 43, 3603 (1910).

- This Journal, 32, 304 (1904).

3 Jackson and Gazzolo: Ibid.. 22, 51 (1899).

552 Jackson and Bigelow

of iodine, or by trichloride of iodine in the chloride of iodine.
By making this substance with sufhcient chlorine to give a
slight amount of trichloride of iodine, and then decomposing
this by heat, a product was obtained which gave a nearly
colorless triiodaniline when passed into a solution of aniline
hydrochloride; and from this the pure i,3,5-triiod-2-brom-
4,6-dinitrobenzene melting at 292° was obtained without
difficulty.

The impurity formed, when dark-colored triiodaniline was
used, gave percentages of halogen agreeing with the formula
Cf;l3Br2N02, but this must be accepted with some reserve, as
it is not supported by analyses for carbon and nitrogen, which
were impossible with the amount of pure material at our dis-
posal after the extended fractional crystallization necessary
to obtain it. The i,3,5-triiod-2,4-dibrom-6-nitrobenzene melts
at 255° to 256° with blackening, but melting points as high
as 260° have been observed. We ascribe these differences
to variations in the conditions under which the decomposi-
tion point was observed ; in fact, much greater differences than
this have been found in similar cases in this laboratory. It
is almost impossible to separate this substance completely
from the dinitro compound by crystallization, so a better
way to obtain it is to treat the mixture with sodic malonic
ester, which acts on the dinitro but not on the mononitro
compound, after which this latter can be more easily separa-
ted from the other products by fractional crystallization.
This greater inertness of the mononitro compound is another
example of the slighter influence of an atom of bromine than of
a nitro group on halogen atoms ortho-para to it. The triiod-
dibromnitrobenzene was probably formed during the boiling
with fuming nitric acid by the destruction of a portion of the
dark impurity of the triiodbrombenzene, giving free bromine
which then replaced one atom of hydrogen in that substance,
the nitric acid subsequently converting it into the nitro com-
pound. The iodine also formed by this decomposition was con-
verted into iodic acid, which was always obtained in such
preparations. The formation of tetrabromdinitrobenzene from

i,3,5'Ti'nod-2-Byom-4,6-Dinitrobenzene 553

tribromdinitrobenzene/ and of pentabromnitrobenzene from
tetrabrombenzene^ with nitric acid are parallel cases.

The i,3,5-triiod-2-brom-4,6-dinitrobenzene, when treated
with sodic malonic ester, gave, as already stated, in the cold,
or when heated for a short time, the i,3-diiod-2-brom-4,6-
dinitrobenzene, C6Hl2Br(N02)2, melting at 187°, and in ad-
dition to this acetylenetetracarbonic ester, C2H2(COOC2H5)4,
which would be expected^ as the secondary product from such
a replacement of halogen by hydrogen. If instead the sodic
malonic ester was heated for some time with the dinitro com-
pound, the reaction went further, forming the i-iod-2-brom-
4,6-dinitrophenyl-3-m.alonic ester,

C6HIBr(N02)XH(COOC2H5)2,

melting at 107°. This agrees with the only other case known
of this reaction with an iodine compound, as the 1,3,5-triiod-
dinitrobenzene^ gave with sodic malonic ester diioddinitro-
benzene with but little ioddinitrophenylmalonic ester. The
constitution of these substances has not been determined
experimentally, but analogy with the corresponding bromine
compounds^ leaves no doubt that their structure is that given
above.

The observation that one atom of iodine is replaced by
hydrogen before any iodine has beeen replaced by the malonic
ester radical was rendered possible by the fact that iodine
shows a greater tendency toward the hydrogen substitution
than toward that by the malonic ester radical, while with
the bromine compounds the halogen seems to be replaced with
equal ease by hydrogen or by the malonic ester radical, so
that the action could not be stopped after it had run half
way. On the other hand, the reaction with sodic alcoholates
and tribromdinitrobenzene* ran in two stages, the first being
the replacement of two atoms of bromine by two etlioxyls,
followed by that of the third atom by hydrogen. As this was

1 Jackson and Wing: This Journal, 10, 283 (1888).

- Jackson and Bancroft: Ibid., 12, 289 (1890).

* Jackson and Moore: Ibid., 12, 7 (1890).

^ Jackson and Langmaid: Ibid., 32, 304 (1904).

5 Jackson and Robinson: Ibid., 11, 557 (1889).

6 Jackson and Warren: Ibid.. 13, 164 (1892).

554 Jackson and Bigeloiv

the only observation of the sort know'n before our present
■work, our opposite result was decidedly unexpected, and the new
light shed on the subject by this discovery led to the hypothe-
sis given at the beginning of this paper.

When sodic ethylate acted on tribromtrinitrobenzene
(which is the true analogue of our bromdinitro substance),
two different reactions took place simultaneously' — the replace-
ment of all three atoms of bromine by ethoxyls, giving trinitro-
triethoxybenzene, C6(OC2H5)3(N02)3, and the replacement of
two nitro groups by ethoxyls, giving tribromnitrodiethoxy-
benzene, C8Br3N02(OC2H5)2; while the 1,2,3,5-tetrabromdini-
trobenzene gave the replacement of the 2 -bromine and a nitro
group by ethoxyls.^ It was not surprising therefore that our
compound was converted into i,3,5-triiod-2-brom-6-nitio-4-
phenetole, melting at 148°, by sodic etliylate, and into the
corresponding anisole melting at 163° by sodic methylate.
It is probable that the parallel replacement of the iodine
atoms also occurred to a limited extent, as the wash waters
from the reaction product gave a test for halogen.

The effect of reducing agents on our i,3,5-triiod-2-brom-
dinitrobenzene was also tried to see if the iodine could be re-
moved by reagents which ordinarily do not produce this
effect, since Calvert and one of us^ had found that tin and
hydrochloric acid converted tribromdinitrobenzene,

CeHBr3(N02)2,
into metaphenylenediamine, while zinc and acetic acid gave
tribromphenylenediamine. Our experiment with tin and
hydrochloric acid gave no satisfactory result, and another
with stannous chloride left our dinitro compound unaffected,
which, however, we think was probably due to our failure to
find the proper conditions; but we did not repeat these ex-
periments, as zinc and acetic acid removed all three atoms of
iodine, forming 2-brom-4,6-diaminobenzene, CeH3Br(NH2)2,
melting at 92 °, and, as this mixture had not removed the bro-
mine in the earlier work, there was no question that the more

1 Jackson and Warren: This Journal, 16, 607 (1894).

2 Jackson and Calvert: Ibid., 18, 298 (1896).

3 Ibid.. 18, 467 (1896).

j,3,5-Triiod-2-Brom-4;6-Dinitrobenzene 555

efficient tin and hydrochloric acid would produce the same
effect, when the proper conditions were found. We next
tried to find a reducing agent which woulc leave the iodine
untouched. Sulphurous acid had no action, iron and acetic
acid gave the symmetiical bromphenylenediamine, and so
did the use of iron by hydrogen and sulphuretted hydrogen
recommended by Merz and Weith^ — a method which does
not deserve the neglect into which it has fallen. These
results made it probable that all acid agents would replace
the iodine by hydrogen, and, as the usual alkaline reducer,
ammonic sulphide with sulphuietted hydrogen, was inad-
missible for fear of the formation of organic sulphides, we
tried ferrous hydroxide, and with this obtained the 1,3,5-
triiod-2 -bromphenylenediamine, C6l3Br(NH2)2, which melts
at 187°, and shows basic properties, but so reduced by the
halogen atoms in the molecule that its hydrochloride,

CelsBrNH^NHgCl,

is decomposed at 100°.

The i,3,5-triiod-2-brom-6-nitro-4-phenetole was converted
by zinc and acetic acid into metaminophenol, recognized by
the decomposition point of its hydrochloride, which we found
to be 225°. Ikuta^ gives 229°. In this case, therefore, the
reduction was moie complete than in that of the dinitro com-
pound, as even the bromine has been removed, which seems
to show that a hydioxyl or ethoxyl has greater loosening effect
than an amino group. The loosening effect on the halogens
is ascribed to the amino rather than the nitro groups, as it
seems probable that the first action of the reducing agent
would be exercised on the nitro groups, and the -work of Cal-
vert and one of us^ has shown that amino groups produce
such a loosening effect on halogens.

We have also tried reduction experiments on a number of
related iodine compounds. Zinc and acetic acid remove iodine
from 1,3,5-triioddinitrobenzene. The metaphenylenediamine
was not isolated, but a quantitative determination showed

1 Z, Chem., 1869, 242.

2 This Journal, 16, 40.
^ Ibid., 18, 467 (1896).

556 Jackson and Bigelow

that in one case as much as 8i per cent, of the iodine had been
removed. 1,3,5-Triiodaniline gave no removal of iodine
with tin and hydrochloric acid, and a mere trace with zinc
and acetic acid. The corresponding tribromaniline, on the
other hand, is converted into dibromaniline in a few hours
b}^ tin and hydrochloric acid,* so that here we have a striking
difference in the behavior of the two halogens, unless, indeed,
it is due to a difference in the conditions of the experiments.
i,3,5-Triiod-2-brombenzene with zinc and acetic acid lost
the two atoms of iodine ortho to the bromine, giving p-\od-
brombenzene melting at 92 °, the removal of the iodine in this
case being much easier than would be expected. A quanti-
tative determination of the amount of halogen eliminated
showed that there was a secondary action, in which still more
of it was replaced by hydrogen. The reduction experiments
therefore confirm the inference drawn from those with sodic
malonic ester, that iodine shows a greater tendency tlian bro-
mine to be replaced by hydrogen, the only exception being
the case of the triiodaniline compared with tribromaniline. In
considering this inference it must be remembered that the
conditions under which the experiments were tried, such as
temperature, solvents, and others, weie not necessarily parallel,
since they were carried on by different men with an interval
of several years, but we think that in spite of this there are
enough agreeing obser^^ations to justify the conclusion to
which we have come. Work in this direction will be con-
tinued in this laboratory.

Sodic ethylate did not act, so far as we could find, on 1,3,5-
triiodbenzene, and only very slightly on i,3,5-triiod-2-brom-
benzene, as but a trace of sodic halide could be found. The
same reagent caused the substitution of bromine by hydro-
gen in i,3,5-trichlor-2-brombenzene- with apparent ease,
and with less ease in 1,2,3,5-tetrabrombenzene,^ so, if these
observ^ations are confirmed by further experiments, chlorine
in the 1,3,5 positions exerts the greatest loosening effect on a

> Jackson and Calvert: This Journal, 18, 468 (1896).
i2.- Jackson and Gazzolo: Ibid... 22, 50 (1899).
[,^,3 Jackson and Calvert: Ibid., 18, 298 (1896).

i,3,5-Triiod-2-Brom-4,6-Dinitrohenzene 557

bromine atom on the benzene ring, bromine less, and iodine
the least.

EXPERIMENTAL PART

In preparing triiod aniline by the method of Michael and
Norton,^ modified by Langmaid^and one of us and by Green^
and one of us, we were troubled like our predecessors with
the formation at first of a brown impure substance from which
it was hard to obtain satisfactory derivatives. After many
experiments we found that brown impurities were formed
if the chloride of iodine used contained trichloride of iodine
on the one hand, or an excess of iodine on the other, but that
with pure chloride of iodine, made by the following process,
a triiodaniline of excellent quality could be obtained with a
yield often as high as 80 per cent, of the theoretical, and
pure enough to be used directly in making bromtriiodbenzene.
The 42 grams of iodine (to be used with 10 grams of ani-
line) were treated with a rapid stream of chlorine until yellow
crystals of trichloride of iodine appeared on the sides of the
flask, which must be shaken frequently during the process, as
these yellow ciystals may be foimed while there is still much
unaltered iodine in the bottom of the flask. The crystals
formed at first usually dissolved on shaking, and the shaking,
followed by addition of chlorine, was continaed until the yel-
low crystals appeared at once on the addition of more chlor-
ine. The flask was then covered with a watch glass and could
be kept in a cool place even for some days. Just before using,
the mixture was heated gently on the steam bath for about
10 minutes, with occasional shaking, and in this way the tri-
chloride was decomposed, the chlorine escaped and the resi-
due was monochloride of iodine pure enough for our purpose.
It was heated to about 60° and passed by means of a current
of air into a solution of 10 grams of freshly distilled aniline
in 500 cc. of strong hydrochloric acid and 7 liters of water,
as described in previous papers.

1 ,3,5-Triiod-2-hrombenzene, CgHjIaBr. — Ten grams of this

1 Ber. d. chem. Ges.. 21, 1707 (1888).

2 This Journal, 29, 300 (1904).

3 Ibid., 31, 600 (1906).

558 Jackson and Bigelow

triiod aniline, mixed with 120 cc. of glacial acetic acid and 40
cc. of constant-boiling hydrobromic acid, were treated witli
two and a half grams of sodic nitrite, which was finely pow-
dered and sifted into the flask in small quantities at a time,
the flask being shaken thoroughly and cooled with running water
after each addition. This took from one to four hours (the
solid at no time having gone completely into solution), and
after all the nitrite had been added, the mixture was allowed
to stand overnight and then gently heated for a few minutes
to start the decomposition of the diazo compound, when it
was allowed to stand at ordinary temperature until the evo-
lution of nitrogen had nearly ceased, after which the heating
and standing were continued as long as any reaction was ob-
served. If the triiodaniline Was pure, yellow crystals were
obtained, the yield being 22 grams from 20 grams of triiod-
aniline, that is, 93 per cent, of the theoretical. If, on the
other hand, the dark brown triiodaniline was used, the diazo com-
pound, on heating, either went into solution or changed to an
oil which, on cooling, solidified to a black tarry mass, and
from this very little triiodbrombenzene could be obtained.
The products in this case will be discussed later. The yellow
product was purified by recrystallization from alcohol until
it showed the constant melting point 146°.

I. o. 1638 gram substance gave o. 2726 gram AgBr + Agl.^

II. o. 1481 gram substance gave o. 2467 gram AgBr + Agl.

III. o. 1776 gram substance gave o. 2952 gram AgBr + Agl.

Calculated for Found

CeHzBrlg I H HI

Brl3 86.16 85.90 85.96 85.76

The constitution of this triiodbrombenzene follows from its
preparation from symmetrical triiodaniline.

Properties of i ,j,^-Triiod-2-brombenzene. — It crystallizes from
alcohol in very light yellow needles and melts at 146°. It
is easily soluble in ether, chloroform, acetone, benzene, tol-
uene, carbonic disulfide, ethyl acetate, petroleum ether or
amyl alcohol; slightly soluble in cold, soluble in hot ethyl

1 The analyses were made by the method of Carius, and the halogens calculated
on the assumption that the mixture was AgBr + 3 Agl, which is justified by the method
of formation of the substance.

i,3,5-Triiod-2-Brom-4,6-Dinitrohenzene 559

alcohol; nearly insoluble in cold, soluble with difficulty in hot
glacial acetic acid. Alcohol is the best solvent for it. Hydro-
chloric acid, nitric acid, sulphuric acid or sodic hydroxide is
without action on it hot or cold.

As already mentioned, the product of the diazo reaction
on a brown impure triiodaniline was a black or dark red tarry
mass, from which a little triiodbrombenzene could be ob-
tained by repeated crystallizations from benzene and chloro-
form, but we did not succeed in raising its melting point
above 139°. This difference of 7° in melting point was appar-
ently caused by a very small amount of impurity, as this speci-
men gave good results on analysis (I), but it is much easier
to start from pure triiodaniline, as already advised. From
the benzene and chloroform mother liquors of the first crys-
tallization a nearly white substance melting at 82° was ob-
tained, but in too small quantity for analysis, and we had
no time to follow this line of work further.

1 ,3 ,3-Trnod-2-brom-4,6-dimtrobenzene, Cgl3Br(N02)2. — Triiod-
brombenzene, melting not lower than 139°, was heated with
fuming nitric acid of specific gravity i . 5 1 in a flask with a re-
turn condenser attached by a ground-glass joint. After heat-
ing for several hours there were crystals in the hot acid, more
of which were deposited on cooling ; the liquid was then poured
off into water, giving an additional quantity of the product,
all of which was washed with water, and crj^stallized from a
mixture of equal parts of alcohol and benzene until it showed
the constant melting point 292°, when it was dried at 100°
for analysis. This product is obtained easily by heating
with the acid for two or three hours, but, if crude triiodbrom-
benzene is used, the process runs less satisfactorily, in the way
described after the properties of the dinitro compound.

I. o. 1084 gram substance gave o. 1555 gram Agl + AgBr.

II. 0.201 1 gram substance gave 0.2862 gram Agl 4- AgBr.

Calculated for Found

C6l3Br(N02)2 I II

Br + I 7376 7403 73-47

The method of preparation determines the constitution of
the substance.

560 Jackson and Bigelow

Properties of i,s,5-Trnod-2-hrom-4,6-dinitrohenzene. — It forms
white needles sometimes a centimeter in length and melts
at 292°. Freely soluble in acetone; somewhat soluble in cold,
freely in hot benzene or chloroform; nearly insoluble in cold,
not very soluble in hot alcohol; essentially insoluble in cold,
somewhat more soluble in hot ether or tetrachloride of car-
bon ; only slightly soluble in glacial acetic acid, even when hot.
Aniline dissolves it easily without change of color. Not acted
on by hydrochloric acid, nitric acid, sulphuric acid or a solu-
tion of sodic hydroxide even when hot.

In most of our preparations of the dinitro compound crude
triiodbrombenzene was used, when it was found wise to carry
on the heating for 24 hours, and if the melting point of the
product was low, which happened not infrequently, to crys-
tallize from alcohol and benzene and boil again with fuming
nitric acid. We have devoted much time to the study of the
impurity formed under these conditions but without entirely
satisfactory results, so that all our statements on this sub-
ject must be regarded as preliminary. The principal sec-
ondary product from the action of boiling faming nitric acid
on triiodbrombenzene crystallized sometimes in short hex-
agonal prisms, at others in needles, and melted somewhat in-
definitely at 256°, although melting points as high as 260°
have been observed. As the melting was accompanied by
blackening, these differences may be due to the fact that it
is a decomposition point, not a true melting point, rather
than to the presence of impurities. The substance had been
crystallized repeatedly before the melting point was taken.
This same substance, to judge from crystalline form, melting
point and solubility, was obtained when crude triiodbrom-
dinitrobenzene was heated with sodic malonic ester. Analy-
sis I was made with a specimen direct from nitric acid, II,
III and IV with specimens from the malonic ester reaction.
I. o. 1032 gram substance gave o. 1696 gram AgBr + Agl.
II. o. 1779 gram substance gave o. 2929 gram AgBr + Agl.

III. o. 1194 gram substance gave o. 1968 gram AgBr + Agl.

IV. 0.2424 gram substance gave, on combustion, 0.1019
gram CO2.

i,j,§-Trnod-2-Brom-4,6-Dinitrobenzene 56 1

Calculated for

Found

C6Br2l3(N02)

II III

Br + I

82.28

82.29

82.46 82.53

10.93

II .46

In spite of the variation in melting points, these results
seem to determine the nature of the substance, if, as is fair to
suppose, the numerous crystallizations have removed impuri-
ties. The formation of a substance of this sort is not un-
expected, as shown in the introduction.

Properties of Triioddibromnitrobenzene, CglaBrjNOj. — -It forms
white crystals, at times short hexagonal prisms, at others
needles. It melts with blackening at 256° , but could be made
to blacken as low as 253° and to melt as high as 260° under
different conditions. It is very soluble in ether, acetone,
chloroform, ethyl acetate, benzene or toluene; nearly insolu-
ble in cold, soluble in hot methyl alcohol or glacial acetic
acid; soluble with difficulty in ethyl alcohol. The substance
prepared directly from nitric acid was found to be insoluble
in ligroin, while that extracted in the malonic ester reaction
was soluble; this is the only difference in solubility observed
by us, and we ascribe it to a difference in the boiling point of
the ligroin used in the two cases, since quite as marked a differ-
ence was found in the solubility of iodbromdinitrophenyl-
malonic ester when it was treated with low-boiling or high-
boiling ligroir. The strong acids or a solution of sodic hy-
droxide have no effect on it, even if hot. Sodic malonic ester
seems to be without action on it.

It may be worth mention that a product from the fractional
crystallization of the mixture formed by the action of boiling
nitric acid on crude triiodbrombenzene gave results on analy-
sis agreeing fairly well with the formula CgHIgBrNOj (calcu-
lated, Br + 1, 79.89; found, 78.47, 78.16), but we do not
feel inclined to accept these results, as the melting point,
258^-260°, was not sharp and lies very near that of the dibrom
compound just described, and tlie substance did not give
triiodbromdinitrobenzene by further treatment with fuming
nitric acid. Iodic acid was also found among the secondary-
products.

562 Jackson and Bigelow

J,3,5-Triiod-2-hrom-4,6-dinitrohenzene and Sodic Malonic Ester
Action in the Cold

Ten grams of the triiodbromdinitrobenzene were dissolved
in benzene dried by long standing with sodium and mixed
with a solution of sodic malonic ester made from i . 5 grams
of sodijm, 50 g/ams of absolute alcohol and 10.6 grams of
malonic ester, which amounts to 4 molecules of the ester
for each molecule of the substituted benzene. A claret color
appeared as soon as the solutions were mixed. After standing
at ordinary temperatures for three or four hours the mixture
was poured into highly diluted sulphuric acid and shaken
thoroughly. The benzene, which held most of the color, was
separated from the aqueous portion, and this at first was ex-
tracted with ether but, as the extract was only a very little
oil, which did not solidify after standing for a year, this treat-
ment was abandoned in later experiments. The principal
amount of the products was contained in the benzene solu-
tion, which was allowed to evaporate spontaneously, when it
left a solid mixed with a considerable amount of a dark red
oil. After removing the oil by pressing between filter papers
the residue was crystallized from alcohol and in this way two
substances were isolated, one slightly yellowish and melting
at 187°, the other white and melting at 75°.

1,3- Diiod - 2 - hrom-4,6-dinitrobenzene, CgHIjBr (NOj) 2- — The
substance melting at 187° was dried at 100°.

I. o. 1981 gram substance gave 0.2608 gram Agl 4- AgBr.
II. o. 1733 gram substance gave o. 2297 gram Agl 4- AgBr.

N, at 28° and

III. O.I96I

gram substance gave 10.20 cc.

758.5 mm.

Calculated for Found
C6Hl2Br(N02)2 I II

Br + I2

66.92 66.82 67.31

5.61

5 69

The compound was formed therefore by the replacement
of one atom of iodine by one of hydrogen. The position of
this iodine has not been determined experimentally, but the
analogy of the bromine compounds leaves little doubt that it
is the one between the two nitro groups.

1 ,3 ,5-Triiod-2-Brom-4,6-Dinitrohenzene 563

Properties of i,3-Diiod-2-hrom-4,6-dinitrohenzene. — It forms
straw-colored needles, often half a centimeter in length, with
abrupt terminations. It melts at 187°. It is easily soluble
in methyl alcohol, ether, acetone, chlorofrom, ethyl acetate,
benzene or toluene; easily soluble in hot alcohol, nearly insol-
uble in cold; slightly soluble in hot ligroin, essentially insol-
uble in it when cold. Alcohol is the best solvent for it, but
good results can also be obtained from a mixture of benzene
and ligroin. When this substance, dissolved in dry benzene,
was heated with sodic malonic ester, it was converted into
the iodbromdinitrophenylmalonic ester described later, which
was recognized by its melting point.

Acetylenetetracarhonic Ester

The substance melting at 75°, obtained by the action of
triiodbromdinitrobenzene and sodic malonic ester in the cold,
was easily separated from the diiod compound by crystal-
lization from alcohol, in which it is more soluble. Its melt-
ing point, 75° instead of 76°, and the fact that it contained
no nitrogen suggested that it was the acetylenetetracarhonic
ester, which was confirmed by the following analysis:

Calculated for

C2H2(COOC2H6),

Found

52.83

53-33

0.69

0.58

This ester is the secondary product to be expected in all
cases where a halogen is replaced by hydrogen by means
of sodic malonic ester. ^ It was detected in all our experi-
ments where such a replacement had occurred.

i,3,5-Triiod-2-hrom-4,6-dinitrohenzene and Sodic Malonic Ester
Action when Heated

When the reaction mixture was heated for only half an
hour, the same products were obtained as in the cold, that is,
diiodbromdinitrobenzene and acetylenetetracarhonic ester,
with occasionally a small amount of a substance melting
near 260°. Accordingly, the heating was carried on for twelve

1 This Journal. 12, 308 (1890).

564 Jackson and Bigelow

hours on the steam bath in a flask with a return condenser,
after which the product was poured into diluted sulphuric acid,
thoroughly shaken, and the layer of benzene removed and
allowed to evaporate spontaneously. The residue sometimes
crystallized at once, at others it was an oil which required
standing for days or even months before it solidified. The
solid was divided by crystallization from alcohol into two
products (in addition to acetylenetetracarbonic ester), one of
which was the substituted malonic ester described below,
while the other melted at 256° and proved to be the triioddi-
bromnitrobenzene already obtained by the action of nitric
acid on impure triiodbrombenzene and described earlier in
this paper. Its appearance here may be explained by the
supposition that the triiodbromdinitrobenzene mixed with
it had reacted to form the substance about to be described,
which could then be separated from the more inert dibrom-
nitro compound by crystallization.

i-Iod-2-brom-4,6-dinitrophenyl-3-malonic Ester,
C6HIBr(N02)2CH(COOC2H5)2.— This substance was obtained
from the oily product of the reaction of sodic malonic ester on
triiodbromdinitrobenzene which was drained away from the
solid material and allowed to stand until it crystallized; this
usually took some months, and even then the hil only partially
solidified; but all our experiments to bring about a more
rapid or complete crystallization gave unsatisfactory results.
The solid was freed from adhering oil with filter paper and re-
crystallized from alcohol until it showed the constant melt-
ing point 107°, when it was dried in vacuo.

I. o. 1502 gram substance gave o. 1 190 gram AgBr 4- Agl.
II. o. 1807 gram substance gave o. 1446 gram AgBr -1- Agl.

III. 0.1981 gram substance gave 10.30 cc. Ng at 22° and
755 mm.

Calculated for
C6HIBr(N02)2CH (COjCaHr,)^

Found
II

Br -h I

38.98

38.79

3918

5-27

5 85
The structure of this substance follows from that assigned

i,j,yTruod-2-Brom-4,6-Dinitrobenzene 565

to the diioddinitrobenzene, as it can be prepared by the action
of sodic malonic ester on this compound.

Properties of i-Jod-2-hror,i-4,6-dinitrophenyl-3-inalonic Ester.
— It forms short, thick, lemon-yellow crystals which melt at
107°. It is easily soluble in methyl alcohol, ether, acetone,
amyl alcohol, chloroform, ethyl acetate, low-boiling ligroin,
benzene or toluene; soluble in hot alcohol, slightly soluble in
cold; nearly insoluble in high-boiling ligroin. Alcohol is the
best solvent for it. Sodic h)^droxide converts it into a blood -
red soluble salt, and this change of color was of use during the
preparations in determining which portions of the product
contained this substance.

Still another product of the action of sodic malonic ester on
triiodbromdinitrobenzene was isolated by fractional crys-
tallization, but it occurred in such small quantity that it could
not be purified, especially as its decomposition point was too
indefinite to give a criterion of purity. It was yellow and
began to turn dark at 230°, growing darker till 250°, where
it melted. Specimens showing this behavior toward heat,
and crystallized until cr}'Stals and evaporated mother liquors
behaved alike, gave results on analysis differing by several
per cent.

1 ,3 ,5'Triiod-2-brom-4,6-dinitrohenzene and Sodic Ethylate. — To
a solution of ten grams of the substituted benzene in
100 cc. of dry benzene were added 100 cc. of a solution of
sodic ethylate in absolute alcohol, which contained one mole-
cule of the ethylate for each atom of halogen present. The
mixture was allowed to stand in a tightly corked flask for 7
days, during which time a considerable quantity of solid had
crystallized out on the sides of the flask. The liquid was then
allowed to evaporate spontaneously and the resulting yellow
crystals washed with water (which gave a test for halogen)
and recrystallized from alcohol until they showed the con-
stant melting point 148°, when the substance was dried at
100°.

I. 0.2330 gram substance gave o 1317 gram CO, and
0.0116 gram HjO.

II. o. 1228 gram substance gave o. 1751 gram AgBr + Agl.

566 Jackson and Bigelow

III. o. 1284 gram substance gave o. 1844 gram AgBr + Agl.

IV. 0.3027 gram substance gave 6.6 cc. N2 at 20° and
763 4 mm.

Calculated for Found

CelsBrNOzOCzHs I II III IV

C 15-38 15-41

H 0.80 0.56 ...

I3 + Br 74.00 . . 73.55 74,11 ....

N 2 .24 ... ... ... 2 .56

The substance has been formed, therefore, by the replace-
ment of a nitro group by an ethoxy group. Its constitution
follows from that of the mother substance.

Properties of the i,j,^-Trnod-2-brom-6-nitrophenetole,
C6l3Br(N02)OC2H5.— It crystallizes in very light pink, flat-
tened needles with sharp ends, which in quantity have the ap-
pearance and luster of asbestos. From methyl alcohol or
ligroin the needles are deposited in groups shaped like a dumb-
bell. It melts at 148°, and is soluble in ether, acetone, chloro-
form, tetrachloride of carbon, ethyl acetate or toluene; nearly
insoluble in cold alcohol, moderately soluble in hot; moder-
ately soluble in methyl alcohol, benzene or ligroin; soluble in
hot glacial acetic acid, very slightly soluble in cold. Ethyl
alcohol is the best solvent for it.

The fact that the wash waters from the crude phenetole
gave a test foi a halogen would indicate that a second com-
pound had been formed by the action of sodic ethylate on the
triiodbromdinitrobenzene, but a careful search for it gave no
indication of its presence.

j,3,5-Triiod-2-brom-6-nitroanisole, C6l3Br(N02)OCH3. — Ten
grams of the triiodbromdinitrobenzene dissolved in dry ben-
zene were treated with enough sodic methylate dissolved in
methyl alcohol to give one molecule of the methylate to each
atom of the halogen, and the mixture was allowed to stand
in a corked flask for three days, when it was found that a con-
siderable amount of solid had separated on the sides of the
flask, and the liquid had taken on a slight reddish tint. It
gave a slight test for halogens. It was allowed to evaporate
spontaneously and the residue, after a thorough washing with

i,3,5-Triiod-2-Brom-4,^-Dinitrobenzene 567

water, was crystallized from alcohol until it showed the con-
stant melting point 163°, when it was dried at 100°.

I. o. 1499 gram substance gave o. 2206 gram AgBr + Agl.
II. o. 2102 gram substance gave 0.3092 gram AgBr + Agl.

Calculated for Found

CalsBrNOzOCHg I II

Br + I 75.57 75.97 75.95

Properties of i,3,5-Triiod-2-hrovi-6-nitroanisole. — It forms
slightly yellowish needles, which melt at 163°. It is easily
soluble in ether, acetone, chloroform, ethyl acetate, benzene
or toluene; tolerably soluble in methyl alcohol; nearly insol-
uble in cold ethyl alcohol, more soluble when hot. A mix-
ture of equal parts of benzene and alcohol is the best solvent
for it.

Five grams of triiodbromdinitrobenzene were dissolved in
benzene and treated with enough sodic phenylate to remove
all the halogen from it. The sodic phenylate was made by treat-
ing the calculated amount of sodium with a little water and
adding the sodic hydroxide thus formed to phenol in excess.
After adding the sodium phenylate enough alcohol was added
to the mixture to just dissolve it, and the whole was allowed
to stand in a corked flask for a week, after which tests for a
halogen or nitrite gave negative results. It was next heated on
the steam bath for one hour and a half, and again tested for halo-
gen or nitrite with negative results. The solvents were then
allowed to evaporate spontaneously and the residue was
washed with water till free from phenol and inorganic matter,
when one crystallization from benzene and alcohol raised its
melting point to 280°. There can be little doubt, therefore,
that it was the unchanged triiod compound, which melts
at 292°.

Reduction Experiments

Tri'iodbroninitrophenetole with Zinc and Acetic Acid. — Two
grams of the phenetole mixed with a few cc. of alcohol were
treated with zinc and 80 per cent, acetic acid and the action
allowed to go on at ordinary temperatures. In a few hours
the substance had dissolved completely, when the zinc was

568 Jackson and Bigelow

filtered out and water added to the filtrate, which threw down
a precipitate. Extraction with ether removed most of the
product, but to be certain that no considerable amount was
lost the zinc was precipitated with sodic hydroxide not in
excess, and the zincic hydroxide extracted with alcohol, while
the filtrate from it was extracted with benzene, but very lit-
tle additional solid material was obtained by these two last
extractions. The main product left by the evaporation of
the ether was a white solid which rapidly turned dark on ex-
posure to the air and melted (crude) at about 60°. As the puri-
fication of this free amino compound would evidently require a
large amount of material, we converted it into the hydro-
chloride by saturating its solution in dry benzene with dry
hydrochloric acid gas. The precipitate formed in this way
was filtered out and dried between filter papers, as we feared
heat might decompose it.

o. 1765 gram substance gave o. 1688 gram AgCl.

Calculated for
C6H4OHNH2HCI Found

CI 24.40 23.68

The action of the zinc and acetic acid on the phenetole,
therefore, consisted, in addition to the reduction of the nitro
group, in the replacement of all the halogen atoms by hydro-
gen and the saponification of the phenetole to the phenol.
The new substance must be the in-aminophenol, the hydro-
chloride of which melts, according to Ikuta,^ at 229°. Our
crude product melted at 210°, which was raised by one crys-
tallization from dilute hydrochloric acid to 225°, supporting
the conclusion derived from the analysis.

i,3,5-Triiod-2-hrom-4,6-dinitrobenzene with Zinc and Acetic
Acid. — Five grams of the triiod compound mixed with 80
per cent, acetic acid and alcohol in equal volumes were allowed
to stand with zinc at ordinary temperatures for several hours,
but as there seemed to have been very little action, the mix-
ture was heated on the steam bath, when the solid gradually
dissolved. After it had disappeared entirely, the dark-colored
solution was diluted with water, filtered from the excess of

» This Journal, 16, 40.

i,3>5- Truod-2-Brom-4,6-Dinitrobenzene 569

zinc and. extracted with ether. The residue, after the spon-
taneous evaporation of the ether, was extracted with gaso-
lene, which yielded, on evaporation, spindle-shaped crystals
and rosettes, and these were crystallized from a mixture of
benzene and ligroin, giving spindles, and finally from alcohol
until they showed the constant melting point 92°. The crys-
tals from alcohol were prisms. This melting point, the crys-
talline form, and the solubilities of the compound, suggest
the 5-brom-m-phenylenediamine,^ which melts at 93^-94°;
and as this would be the most probable product of the reac-
tion, there can be no doubt about the nature of the substance.

The same 5-brom-m-phenylenediamine was obtained by
the action of iron and 80 per cent, acetic acid, the mixture
being heated on the steam bath for two days, after which the
iron was precipitated with sodic carbonate and sodic hydrox-
ide, and both the precipitate and filtrate extracted with ether.
The extract crystallized from alcohol showed the melting
point 92°. Tin and hydrochloric acid seemed to give the
same product, but the experiment was tried on too small a
scale to give decisive results.

Stannous chloride, on the other hand, seemed to act very
slightly on the triiodbromdinitrobenzene under the condi-
tions used by us, as, after heating a mixture of one gram of
it with alcohol. Water (equal volumes) and a freshly prepared
solution of stannous chloride for 8 hours on the steam bath,
crystals separated on cooling which weighed 0.57 gram and
melted after one crystallization from benzene and alcohol at
278°, while an additional precipitate obtained by adding
water to the alcoholic filtrate melted after one crystallization
at 260°. We inferred, perhaps somewhat rashly, that these
consisted of the unaltered substance (melting at 292°), and did
not lepeat the experiment.

Sulphurous dioxide dissolved in 50 cc. of alcohol and 10 cc.
of water had no perceptible action on the triiodbromdinitro-
benzene, and the liquid gave no test for halogen after the
experiment.

' Jackson and Gallivan: This Journal, 18, 242. Jackson and Calvert: Ibid.
486 (1896).

570 Jackson and Bigelow

In the hope of finding a milder reducing agent which would
not remove the iodine, we tried "iron by hydrogen" and sul-
phuretted hydrogen recommended by Merz and Weith.^ For
this purpose 2 grams of the triiod compound, with equal
parts of alcohol and benzene, were mixed with the iron, and a
stream of sulphuretted hydrogen passed through the mixture
for 10 hours. The solution was then filtered from the ferrous
sulphide and warmed gently to remove the excess of sulphur-
etted hydrogen, after which it was extracted with ether and
the extract crystallized first from benzene and ligroin, when
it showed the characteristic spindle forms, and then from alco-
hol, which raised its melting point to 92°, so that it is the
5-brom-m-phenylenediamine.

The experiments just described showed with tolerable
certainty that acid reducing agents removed the iodine from
the triiodbromdinitrobenzene if they acted at all, and we next
tried an alkaline agent in the hope of confining the reduction
to the nitro groups. As the usual alkaline reducer, sulphur-
etted hydrogen with ammonic sulphide, was inadmissible
because of the danger of the replacement of halogens by other
groups, we used ferrous hydroxide. Ten grams of triiod-
bromdinitrobenzene dissolved in alcohol were mixed with
five times the amount of ferrous hydroxide needed for com-
plete reduction. The ferrous hydroxide was made by dis-
solving the calculated weight of iron in hydrochloric acid,
taking care to avoid a large excess of the acid, and treating
this ferrous chloride with sodic hydroxide in a flask closed with
a cork carrying a Bunsen valve, until a very slight alkaline
reaction was obtained. If this remained after shaking, enough
ferrous chloride was added to destroy the sodic hydroxide,
as it had been found that even a very small excess of this sub-
stance reduced the yield to a considerable extent. After
adding the triiod compound to the ferrous hydroxide in the
flask its color changed quickly from light green to black. The
mixture was allowed to stand for two days with frequent
shaking and warming on the steam bath, and then the solu-
tion was filtered off. The filtrate, as was expected, yielded

• Z. Chem., 1869, 242.

i,3,5-Triiod-2-Brom-4,6-Dinitrohenzene 571

very little solid on extraction with ether and with benzene,
but the main product Was obtained from the ferroferric hy-
droxide by drying and extracting it with alcohol and also with
benzene. The combined extracts furnished a solid which,
after crystallization from a mixture of benzene and alcohol,
showed the constant melting point 187°, when it was dried
on a porous plate in a desiccator.

o. 1302 gram substance gave 0.2059 gram AgBr -- Agl.

Calculated for
C6l3Br(NH2)2

Found

81.59

81.67

I3 + Br

The redaction therefore has been confined to the nitro
groups.

Properties of i,3,5-Triiod-2-hrom-4,6-diaminohenzene. — It crys-
tallizes in short, rather thick, grayish white needles which
melt at 187°. It is soluble in ether, chloroform, benzene
or toluene; easily soluble in hot alcohol, very slightly in cold.
A mixture of benzene and alcohol is the best solvent for it.
The yield was very small, principally due, it seemed, to the
tenacity with which the ferroferric hydroxide retained the
organic substance ; we had enough of it, however, to make the
hydrochloride.

Hydrochloride of Tr-iiodbromdiaminobenzene,
C6l3BrNH2NH3Cl. — A small amount of the diamine was dis-
solved in benzene (dried over sodium) and the solution satu-
rated with dry hydrochloric acid gas. The precipitate formed
was filtered out, washed with dry benzene and dried on a
porous plate in a desiccator.

0.0643 gram substance heated to constant weight at 100°
lost 0.0039 gram.

Calculated for
CelsBrNHsNHaCl Found

HCl 6.02 6.06

The analysis made with such a small quantity can be re-
garded only as indicating the probable composition of the sub-
stance, but it shows with certainty that the number of halo-
gen atoms in the diamine has reduced its basic properties

572 Jackson and Bigelow

in a very marked degree, as the hydrochloride is decomposed
at ioo°

Triioddinitrohenzene with Zinc and Acetic Acid. — The re-
mainder of this paper contains an account of parallel experi-
ments with other triiod compounds.

Two grams of i,3,5-triiod-4,6-dinitrobenzene^ mixed with
80 per cent, acetic acid, to which an equal volume of alcohol
had been added, were allowed to react with zinc at ordinary
temperatures for several hours, when, upon extracting the
filtered and diluted solution with ether, a light colored solid
was obtained, which blackened so rapidly that there was no
chance of purifying it without work on a large scale. To
avoid this we determined the amount of iodine removed by
the reducing agent by precipitating it as argentic iodide,
and found in one experiment 47 per cent., in another 81 per
cent., of the iodine had been removed. No pains had been
taken to carry on these two reductions under the same con-
ditions, or to bring them to an end, so that the want of agree-
ment was not surprising; and no further work was done in
this line, as our object had been accomplished, which was to
show that zinc and acetic acid remove iodine from the tri-
ioddinitrobenzene.

1 ,3,3-Trnodanilme and Rediicing Agents. — On heating a
small amount of triiod aniline melting at 187° with tin and
hydrochloric acid on the steam bath for several hours no evi-
dence of a reaction could be found, as a test for iodine with
chlorine water and carbonic disulphide gave a negative re-
sult, and the undissolved solid melted at 186°, showing it
was unaltered triiodaniline.

Half a gram of the triiodaniline heated with 80 per cent,
acetic acid and zinc on the steam bath for several hours gave
a solution which on cooling deposited crystals, and these,
after purification with benzene and alcohol, melted at 185°,
showing they were unaltered triiodaniline. There was ob-
tained, however, a slight test for iodine in the liquid, so that
there was probably some action, but only a little, as most of
the triiodaniline was recovered. We do not feel that these

' Jackson and Langmaid: This Journal, 32, 304 (1904).

i,j,^-Trnod-2-Broin-4,6-Dinitrobenzene 573

experiments settle the matter, however, as it is possible other
conditions might bring about a removal of most of the iodine.

1 ,3,5-Triiod-2-hromhenzene with Zinc and Acetic Acid. — Five
grams of triiodbrombenzene mixed with 80 per cent, acetic
acid and 25 cc. of alcohol were heated with zinc on the steam
bath until the solid had disappeared completely. The solu-
tion was then filtered from the zinc and largely diluted, when
a precipitate fell which was extracted with ether, and the ex-
tract crystallized from alcohol until it showed the constant
melting point 92°, which is that of the /?-iodbrombenzene,^
CgHJBr; the action therefore had consisted in replacing the
two atoms of iodine ortho to the bromine by hydrogen.
But, although this was the principal product in this case, it
was not the only one, as shown by the following determination
of the amount of halogen removed in a similar reduction.

1.0525 grams of substance reduced with zinc and acetic
acid yielded i . 2 1 80 grams of argentic halide.
Calculating this as argentic iodide we obtain

Calculated Foixnd

2I 47-47 62.57

3I 71.04

It is obvious that the precipitate may have contained
argentic bromide also, but in any case the amount of argentic
halide obtained shows that a secondary reaction had taken
place, accompanied by a further removal of halogen.

1 ,3,5-Triiodhenzene and Sodic Ethylate. — ^Triiodbenzene^ was
dissolved in dry benzene and enough sodic ethylate in abso-
lute alcohol added to give one molecule for each atom of
iodine. After boiling for three hours no test for iodine was ob-
tained, nor was the result better after boiling for 5 more hours.
Upon cooling crystals separated which, after purification,
showed the melting point 181°, and were therefore unchanged
triiodbenzene, which melts at this temperature. The mother
liquor from these crystals was evaporated to a small vol-
ame and heated for two more hours with a strong solution of
sodic ethylate, but no test for iodine could be obtained.

1 Griess; Jahresb. d. Chem., 1866, 452. Korner: Gazz. chim. ital., 4, 339.

2 This Journal, 29, 300 (1904).

574 Kr eider and Jones

1 ,3,3-Triiod-2-brombenzene with Sodic Ethylate. — Ten grams
of the triiodbrombenzene were dissolved in dry benzene and
mixed with 250 cc. of absolute alcohol containing enough
sodic ethylate to giA^e one molecule for each atom of halogen.
No change was visible in the cold, and no test for a halogen
could be obtained, but after boiling on the steam bath for 3
hours there was a slight test and after 12 additional hours of
boiling the test for halogen was distinct. On evaporating to
dryness a somewhat tarry residue was left, from which we suc-
ceeded in isolating only unaltered triiodbrombenzene, recog-
nized by its melting point, 146°, and two determinations of
halogens. It seems therefore that sodic ethylate acts upon
the triiodbrombenzene but only to a very limited extent.

The research will be continued.

THE CONDUCTIVITY OF CERTAIN SALTS IN METHYL
AND ETHYL ALCOHOLS AT HIGH DILUTIONS

By H. R. Kreider and Harry C. Jones

This work is a continuation of that already discussed in a
previous article.^ In our earlier Work We measured the con-
ductivity of certain salts in very dilute solutions of ethyl
alcohol and methyl alcohol. The salts employed were potas-
sium iodide, ammonium bromide, potassium sulphocyanate,
lithium nitrate, sodium iodide, calcium nitrate, cobalt chlor-
ide and copper chloride. The conductivities of these salts
were measured in solutions ranging in concentration from
N/1600 to N/51200. The conductivities of these salts in more
concentrated solutions had been previously measured by Jones
and his coworkers.

In our former work a well defined maximum in conduc-
tivity with increasing dilution was noted in solutions of a
number of salts. Certain relations between these maxima
were pointed out. The ratio between /too for a certain salt
in methyl alcohol, and //oo for the same salt in ethyl alcohol

1 This Journal, 46, 282 (1911).

Conductivity of Certain Salts 575

was found, to be a constant for different binary salts. The
following values were obtained f or c == /loo methyl alcohol/^ 00
ethyl alcohol.

Binary Electrolytes

LiNOgato" 2.37

Nal at 0° 2.37

NH4Brat25° 2.44

Nal at 25° 2.17

Ternary Electrolytes

CoCl2ato° 3.68

The conductivities of other salts at the same high dilutions
have now been measured, with the object of seeing whether
the relations given above are general. The salts used were
sodium bromide, lithium bromide and cobalt bromide. The
work with potassium sulphocyanate in methyl alcohol wa5
repeated, and more satisfactory results obtained.

Most of the measurements were made at tlie dilution of
complete dissociation. The values of /x^ for the same salt
in methyl and ethyl alcohols aie then compared with each
other, and also with the value oi n^ for the same salt in
water. The apparatus employed is the same as that previously
described, with some minor improvements. The same cells,
with concentric platinum cylinders as electrodes, were used.
One cell having the electrodes closer together and a smaller
constant was also employed.

Both salts and solvents were purified with special care. A
solution of a certain concentration was then prepared as the
mother solution, and from it the other solutions were made.
Some of the solutions were as dilute as N/ 102400.

The precautions observed were the same as those previously
discussed. In addition, greater care was exercised in purify-
ing the methyl alcohol. It is very difficult to get methyl
alcohol in large quantities in such a state of purity that its
specific conductivity is low enough to obtain good results
for such dilute solutions. It was found that the conductivity
of methyl alcohol could be much reduced by treating the

576

Kreider and Jones

alcohol with a little dilute sulphuric acid. The best results
were obtained when the alcohol was in contact with the acid
not more than twelve hours. The alcohol was then distilled
from the acid and boiled with, and distilled from lime at least
twice.

The results obtained are given in the following tables:

Table I. — Conductivity of Sodium Bromide in Methyl Alcohol

800

1600

3200

6400
12800
25600
51200

Iiv0°

Iiv25°

530

77-4

582

62.3

63.8

64.1

64.6

69.1

lOI

Table II .—Conductivity of Sodium Bromide in Ethyl Alcohol

V fvO° tiv25°

8cx)

1600

3200

6400

12800

25600

51200

102400

Table III. — Conductivity of Lithium Bromide in Methyl Alcohol

V . /'r/0° Pi; 25°

800 40.9 57-2

1600 44.5 64.8

3200 45.6 68.7

6400 48 . 2 69 . 8

12800 47.8 73.7

25600 49.5 74.6

Conductivity of Certain Salts

577

Table IV. — Conductivity of Lithium Bromide in Ethyl Alcohol

1600

3200

6400

12800

25600

51200

102400

Table V. — Conductivity of Potassium Sulphocyanate in Methyl
Alcohol

1600

3200

6400
12800
25600

ItvO"

Pvis"

70.7

99 9

75-8

106.2

72.9

102.9

73-9

104.5

72.1

105.4

86.1

132.2

Table VI.-

800

1600

3200

6400

12800

25600

51200

-Conductivity of Cobalt Bromide in Methyl Alcohol

Iiv0° f'v25°

19.0 23.6

21.6 27.

24.9
29. I

33-1
36.8
38.6

659

Table VII.— Conductivity of Cobalt Bromide in Ethyl Alcohol

800

1600

3200

6400

12800

25600

51200

57-9
61.3
62. 2
64.7
66.9
70.1
96.2

I'v25°

77-3
82.0

87.4

95 I

106.7

106.6

124.0

578

Kreider and Jones

cs M N n cs

c3 !^ ^ O ^t!

^ k4 ;z; U! h4

CN rO O CA M (S C<

o ^ H -^ 1-^ cs

•^

M (S lO LO

ro ON

o o o o

O OS

1 ^^

M H W l-i

HH 6

|g 8 8

i-( 05.'^ c3

^4^

t^:z;a^

**^

f§ s^

3 3

E S

N rO fO --d-

CO O

^«

O) C4 (N CS

<S CO

8 8

^'^

Conductivity of Certain Salts 579

Discussion of Results

Table I gives the conductivities of sodium bromide in methyl
alcohol at both 0° and 25°. Ato°-we have complete dissocia-
tion probably at 6400 liters. At 25° complete dissociation
is not attained until a dilution of 12800 liters is reached.

Table II gives the conductivities of sodium bromide in ethyl
alcohol at 0° and 25°. Complete dissociation is reached at
both temperatures at a dilution of 12800 liters. At the volume
102400 there is a marked decrease in conductivity, probably
due to the great solvation at this extremely high dilution.
The same fact may be noticed in several other tables where
ethyl alcohol is used as a solvent.

Table III gives the conductivity of lithium bromide in
methyl alcohol. At 0° complete dissociation probably is
reached at 6400 liters. At 25 ° there is no maximum in conduc-
tivity, but the rate of increase is much smaller above 3200
liters than it is at greater concentrations, indicating that
the maximum is nearly reached.

Table IV gives the conductivity of lithium bromide in ethyl
alcohol. At 0° there is complete dissociation at 3200 liters.
At 25° it is complete at 6400 liters. For both temperatures at
higher dilutions the conductivity remains almost constant
up to a dilution of 102400 liters, where, at 25° there is a marked
decrease in the conductivity.

Table V gives the conductivity of potassium sulphocyanate
in methyl alcohol. This is a repetition of work previously
done. We have now obtained more concordant results. Here
complete dissociation is reached at both temperatures at
3200 liters.

Tables VI and VII give the conductivity of cobalt bromide
in both methyl alcohol and ethyl alcohol. Here there is no
maximum in conductivity. This is probably due to the fact
that cobalt bromide is much solvated.

Table VIII gives, in the right-hand columns, the ratios of
the values of /zoo for a number of salts in the following sol-
vents: Water and ethyl alcohol; methyl alcohol and water;
and methyl alcohol and ethyl alcohol. When we consider the
large magnitude of the experimental error in working at these

580 Kr eider and Jones

great dilutions it is quite probable that the relation, //oo for
one solvent//^ 00 for another solvent = c, holds where the salts
have approximately the same degree of solvation.

That there is a constant relation between the values of
H^ for different salts in different solvents we would expect.
When a certain salt in two different solvents is completely-
dissociated, we have either the same number of ions present,
or, relative to the concentration, the same number of ions
present.

When the point of complete dissociation is reached at the same
dilution in both solvents, we have the same number of ions
present in the same volume of the two solvents. When such
a point of complete dissociation exists at different dilutions
in solutions of the two solvents, the number of ions in equal
volumes of the two solutions varies directly as the concentra-
tion, and we have, relative to the concentration, the same
number of ions present.

Conductivity is a function of the number of ions present
and the velocity with which these ions move. Since at the
complete dissociation of a salt in solutions of two different
solvents the number of ions is actually the same, or relative
to the concentration the same, we can eliminate this factor —
the number of ions — and consider only the velocities with
which these ions move.

Two factors primarily determine ionic velocity. The ionic
mass and volume, and the fluidity of the solution, which is,
of coarse, the reciprocal of the viscosity. Assuming that the
ionic masses and volumes of a certain salt in two different
solvents at complete dissociation remain the same, then the
velocities of the ions ought to vary as the fluidities of the re-
spective solvents. Since the number of ions in the two sol-
vents at the same dilution of the solutions is the same, the
ionic masses and volumes being the same, the conductivities
ought to vary directly as the fluidities of the solvents ; the
ratio between the values oi [x^in the various solvents ought
to be the same as the ratio of the fluidities of these solvents.
This, however, is not the case. If the ratios between the
values of // QQ for the salts in the two solvents are not the same

Conductivity of Certain Salts 581

as the ratios between the values for the fluidities of these
solvents, the mass and probably the volumes of the solvated
ions must differ in the two different solutions.

The fluidity of methyl alcohol at 0° is 123.9 while that of
ethyl alcohol is 56 . 24. Since the fluidity of methyl alcohol
is much greater than that of ethyl alcohol, we would expect
the ions of a dissolved salt to move much faster in m,ethyl
alcohol, and, consequently, the conductivity of a solution of
a salt at any concentration would be much greater in methyl
than in ethyl alcohol. At complete dissociation, where the
number of ions is the same in both solvents, we would
expect the conductivity in methyl alcohol to be as much greater
than that in ethyl alcohol as the fluidity of the former is greater
than that of the latter. We would expect a direct relation
between fluidity and conductivity.

Jones and his coworkers have shown that there is solva-,
tion in solutions in water, methyl alcohol, ethyl alcohol and
solvents in general; and that this solvation increases with
dilution. From this we know that at complete dissociation
solvation is greater than at any dilution short of complete
dissociation.

Since a molecule of methyl alcohol is much heavier than
one of water the ion, for equal solvation, would be loaded
down more in the former solvent, and the conductivity in
methyl alcohol Would be less than in water, even if the fluidi-
ties were the same. Again, the molecule of ethyl alcohol is
heavier than the molecule of methyl alcohol. The equally
solvated ion would, therefore, be heavier in ethyl alcohol
and the conductivity less.

Table VIII ^ gives the ratios of the fluidities of three sol-
vents— water, methyl alcohol and ethyl alcohol; also the ratio
of /«oo for certain salts in these solvents at both 0° and 25°.
Water and ethyl alcohol are first compared. Their fluidities
are nearly the same, hence their ratio is nearly unity. The
ratios between the values of /zoo for four salts in these two
solvents are given. The mean of these values is 2 . 33 with a

1 The values for /i^ of the various salts in water as a solvent given in Table VIII
are taken from the work of Jones and Getman (Z. physik. Chem.. 49, 385 (1904)).

582 Kr eider and Jones

minimum of 2.26 and a maximum of 2 . 44. This mean ratio
of the values is much greater than the ratio of the fluidity
values of the two solvents, indicating that the ion in ethyl
alcohol moves much more slowly, compared with its velocity
in water, than we would expect from a comparison of the
fluidities of the solvents. The only explanation of this fact
seems to be that the ion is loaded down much more in ethyl
alcohol than in water, and, hence, its velocity diminished,
giving a much lower conductivity. The only way in which
it could be loaded down is by one or more molecules of the
solvent being united with the ion.

In the second column of Table VIII water and methyl
alcohol are compared. The fluidity of methyl alcohol is much
greater than that of water. It is 2.18 times as great at 0°.
There are four /Xco ratios. The mean is i .03. The maximum
is 1 .05 and the minimum is i .01. The fact that the fluidity
of methyl alcohol is much greater than that of water, while
the conductivity of salts in methyl alcohol is but little greater,
indicates that in methyl alcohol the ion is loaded down more
than in water, and its velocity more retarded. This, of course,
we would expect from the masses of the molecules alone, since
those of methyl alcohol are so much greater than those of
water.

In the third column, methyl alcohol and ethyl alcohol are
compared. The fluidity of methyl alcohol is 2.17 times as
great as that of ethyl alcohol at 0°; while the mean of the con-
ductivity ratios is 2 . 44. This would indicate that in ethyl
alcohol the ions are loaded down more than in methyl alcohol,
which, again, we would expect. This is the case at both 0°
and 25°. The diff'erence is, however, much less at 25° than
at 0°. At 25° the ratios of fluidity and conductivities are
more nearly equal. This would indicate that at the higher
temperatures the ions are less retarded and, accordingly,
less solvated, as Jones and Bassett^ have pointed out.

The fact that the equation

/^oo s/ficc s' = c,
where s and s' are any two of the solvents above mentioned,

• This Journal, 34, 290 (1905).

Conductivity of Certain Salts 583

holds for the values in Table VIII is in itself important. But
it further shows approximately which salts are solvated to
the same extent in any given solvent, since it holds only in
such cases. For such salts as are not solvated to the same ex-
tent in a given solvent the equation will, of course, not hold.

We have worked with one salt, cobalt chloride, which is
very hydroscopic, and which has been shown by Jones and his
coworkers to be strongly hydrated in aqueous solutions. This
salt shows a greater departure in the ratios of its conductivi-
ties from the ratios of fluidities than is shown by the other
salts which we studied. From the results in the column
fioo W///00 E we see that cobalt chloride is much more loaded
down in ethyl alcohol than it is in water. The value of the
ratio of j« 00 in the two solvents, 3 . 70, is greater than that of
the other salts, 2.33, and much greater than the value of the
fluidity ratio of the two solvents, which is 0.995.

From the column //oo M/j«oo W we see that the ions of cobalt
chloride are loaded down more in methyl alcohol than in
water, and that the ratios of the conductivities deviate more
from the ratios of fluidities than in the case of the other salts ;
but this deviation ought not to be as great as in the case of
water and ethyl alcohol, since there is not so much difference
between the masses of the molecules of methyl alcohol and
water as there is between those of ethyl alcohol and water.
Comparing the figures in this column we find that this is the
case, since 0.99, the value of the ratio of the conductivities
of cobalt chloride, is less than 103, the ratio for the other
salts, and very much less than 2.18, the ratio of the fluidities
of the two solvents.

We would expect the cobalt chloride ions to be loaded down
to the greatest extent in ethyl alcohol, and that the conduc-
tivity in ethyl alcohol would be lower than in methyl alcohol.
This difference should be greater than that of the other salts
which are less solvated. We would expect the ratio //oo M/
//oo E for this salt to show a greater departure from the ratio
of the fluidities of the two solvents than is shown by the other
salts. A comparison of the figures in column /too M//X00 E
shows this to be the case.

584 Kr eider and Jones

Although this does not give us a method for measuring
the actual degree of solvation in the different solvents, it
throws some light on the relative masses of the solvents com-
bined with any given ion in these dilute solutions.

It has been shown that water hydrates salts in more con-
centrated solutions, and that this hydration increases with
the dilution, as we would expect from the mass action of the
solvent. If in water these salts are solvated, it is evident
from the above facts that in methyl alcohol there is a larger
mass of the solvent in combination with the ion. This does
not necessarily indicate that in the latter case there are more
molecules of the solvent in combination with the ion. There
may be the same number, or even less, in combination, but
the molecules of methyl alcohol being so much heavier might
cause the difference in conductivity. In ethyl alcohol also
there is a larger molecular mass than in water or in methyl
alcohol, but here, again, the difference between the ratios
may be due to the heavier molecules of ethyl alcohol.

It is evident that this factor of solvation at complete disso-
ciation plays a role in oar present method for calculating dis-
sociation by means of the conductivity of the solution. Here
we employ the equation a = jjl^I [icx) ■ In the above discussion
/;too is shown not to be a true function of the actual number
of ions present, but is smaller than it ought to be in considera-
tion alone of the number of ions present. Furthermore, it
has been shown that there is soh^ation in more concentrated
solutions also; hence, //^ is not a function only of the number
of ions present in the solution in question. The deviations
of the values of /z^ and //oo from the true values are probably
of the same order of magnitude, if not very nearly equal,
so that the validity of the equation in calculating conduc-
tivity is probably not seriously aft'ected.

We have measured the conductivities of very dilute solu-
tions of a number of salts in methyl alcohol and ethyl alcohol.

In most cases the values of /ico were found. These values
bear a definite relation to one another.

Study of the Hydrogen Electrode 585

It was found that the ratio of the value of ^t^^ for a certain
salt in one solvent and the value of //oo for the same salt in
another solvent is nearly constant for salts which are solvated
to approximately the same extent.

In cases where one salt is solvated very much more than
another the value of ,«oo is generally less for the more solvated
salt.

We have compared the ratios of /^oo for certain salts in two
different solvents with the ratios of the fluidities of these
solvents. In all cases the former ratios show a departure
from the latter, the value of ^Uoo in the solvent which has the
greater molecular mass always being less than we would
expect from the ratio of the fluidities alone of the two solvents.
Those salts which are known to be solvated to the greatest
extent show the greatest difference between these ratios.

Johns Hopkins Univ.

June, 1911 (

A STUDY OF THE HYDROGEN ELECTRODE, OF THE

CALOMEL ELECTRODE AND OF CONTACT

POTENTIAL

By N. E. LooMis AND S. F. Agree

(We are indebted to the Carnegie Institution of Washing-
ton for aid in this work.)

I. INTRODUCTION

For several years we organic chemists have felt the need
of some direct, very rapid, and accurate method for determin-
ing the hydrogen (also hydroxyl) ion concentration of dilute
solutions. Such a method would be of special value in the
study of many organic reactions involving, for example, the
hydrolysis of salts or the saponification of esters, the reac-
tions of addition products in cases of catalysis by hydrogen
ions, and many others in which the system is gradually chang-
ing.

The methods commonly in use heretofore have presented
serious difficulties in their general application. The conduc-
tivity method, for example, which has had the widest range

586 Loo mis and Acree

of application, rapidly diminishes in accuracy with increase
in dilution of the solution, and furthermore, the measurement
of small concentrations of acids in the presence of other elec-
trolytes, especially the salts of these acids with weak bases, is
almost impossible. Special methods, such as the use of
diazoacetic ester, as suggested by Bredig and Fraenkel,^ have
been used brilliantly in some cases, but are too limited in their
range of application. Methods involving titration are of
course useless in systems in which a state of equilibrium is
established comparatively quickly, for the equilibrium is dis-
turbed as soon as any one of the components is removed.

Although colorimetric methods in the hands of Veley, Salm,
Tizard, Szyszkowski and others yield beautiful results in
many cases, it has been found that neutral salts affect the colors
so greatly in other cases that the method is useless or at least
uncertain.

The hydrogen electrode has been recognized by one of us
as a possible instrument for the solution of this problem.^
Particularly suggestive is the work of H. G. Denham,^ who
m.easured the degree of hydrolysis of several inorganic salts and of
aniline hydrochloride. He obtained results agreeing extremely
well with those determined by Bredig by the conductivity
method. Efforts to duplicate his results and other work on
similar lines in this laboratory at first met with serious diffi-
culties, but these are being gradually overcome.

In view of the extreme importance of any favorable re-
sults in this field, it seemed vv^orth while to make a careful
study of the hydrogen electrode, with special reference to its
constancy, the value of its potential in different acids, the
ease of reproduction, etc. ; in other words, to attempt to make
it a standard electrode for use in the same way that calomel
and mercurous sulphate electrodes are used. To make the
hydrogen electrode an accurate instrument for measuring

» Z. Elektrochem., 11, 525 (1905); Z. physik. Chem., 60, 202 (1907).

2 Desha: Diss., Johns Hopkins Univ., 1909. This work was begun in 1907-8,
and reported at the Christmas meeting of the American Association for the Advance-
ment of Science, in 1908. See Science, 30, 624. Lapworth has also for some time
advocated an attempt in this direction.

3 J. Chem. Soc, 93, 41 (1908).

Study of the Hydrogen Electrode 587

hydrogen ion concentrations, the measurements must be made
with a much higher degree of accuracy than has ordinarily
been done.

Early in the investigation it was realized that much of the
accuracy of the work would be dependent upon the constancy
and ease of reproduction of the calomel electrodes which were
used with the hydrogen electrode. For this reason the study
of the calomel electrode was gone into very thoroughly.

The method of approaching the problem under considera-
tion resolved itself into five lines of investigation:

1. The study of the relative efficiency of the apparatus
used by others and of newer forms devised by us to eliminate
various sources of error. The literature and our own expe-
rience in this laboratory have shown that there is still much
to be done in this line. In this connection we cannot refrain
from expressing our deep obligation to Professor H. N. Morse
for his kindness in giving us many valuable suggestions when
we needed the benefit of the rare knowledge and mechanical
skill that have enabled him to overcome such great difficul-
ties in his own researches.

2. The study of the calomel electrodes. A large number
of calomel electrodes were prepared and measured against
each other so that the value adopted for the potential of the
calomel electrode was the average of a number and not de-
pendent upon a single electrode.

3. The preparation of a large number of platinum elec-
trodes which were intercompared by the method used for the
calomel electrodes.

4. The direct comparison of the hydrogen electrode with
the calomel electrode, involving experiments to determine
the efficiency of various solutions in eliminating contact poten-
tial.

5. The application of the hydrogen electrode to the deter-
mination of the hydrogen ion concentration of various solu-
tions. The hydrolysis of aniline hydrochloride and the effect
of neutral salts upon the dissociation of acetic acid were espe-
cially studied (see the next article).

588 Loomis and Acree

II. PREVIOUS WORK

The normal calomel electrode, composed of mercury, calo-
mel and normal potassium chloride solution, was first used
and described by Ostwald.^ It is described as reproducible
to within one millivolt. He gives the potential of the electrode
as +0.5600 -f- o.ooo6(^° — 18°). This value was determined
by Rothmund- by the drop-electrode method.

A year later CoggeshalF made an extended study of calo-
mel and mercurous sulphate electrodes as to constancy, ease
of reproduction, best form of cell, effect of mechanical disturb-
ance, etc. Although this work constituted the most complete
study of standard electrodes made up to that time, it left much
to be desired. The measurements of potential were made
with a Lippman electrometer, which is a far less accurate
method than that in which a potentiometer and sensitive
galvanometer are used. He did not use the decinormal calo-
mel electrodes at all. Although finding tht mercurous sul-
phate electrodes to be on the whole more suitable for use than
the calomel electrodes, he concluded that "bei Anwendung
wohl gereinigter Chemikalien und einer Vorkehiung gegen
Erschiitterungswirkungen, wie eine solche in der partiellen
Sandfiillung gegeben ist, sind ohne Miihe Normal-Quecksilber-
Kalomel-Electroden herstellbar, deren electromotorische Kraft
von dem Normal Wert um nicht mehr als 0.0008 Volt ab-
weicht, und dies mit ausserordentlicher Konstanz."

Smale^ was the first to make an extended study of the hy-
drogen electrode and his work was principally in connection
with the oxygen-hydrogen gas element. He concluded that
the material (platinum, palladium, gold and carbon) in the
electrode used played no part in the electromotive force of the
cell, provided that it was not acted upon chemically. The
surface and size of the electrodes, above a certain limit, had no
effect.

Wilsmore^ repeated Smale's work and made allowance for

1 Ostwald-Luther: "Physiko-Chemische Messungen," 3rd Edition, p. 441.
2Z. physik. Chem., 16, 15 (1894).
^ Ibid., 17, 62 (1895).
" Ibid.. 14, 577 (1894).
'^ Ibid.. 36, 296 (1900).

Study of the Hydrogen Electrode 589

the contact potential of the solutions. In regard to the part
played by the electrode, he arrived at the same conclusions
as Smale. As his zero of potential he adopted the potential
of the hydrogen electrode toward a solution normal with
respect to hydrogen ions. On this basis he found the value
of the normal calomel electrodes to be 0.283 volt. He calcu-
lated the contact potential of a large number of pairs of solu-
tions, using data obtained by others, and studied the electro-
motive force of the hydrogen-oxygen gas battery and the
potential of a large number of metallic electrodes.

Richards^ found that the temperature coefficients increased
with dilution of the solution, and that the decinormal
calomel electrode is more uniform in its behavior than the
electrodes containing normal potassium chloride solution. He
determined the temperature coefficient of the decinormal elec-
trode to be 0.00079. In this work he noticed certain gradual
changes in the potential of the calomel cells, especially of those
containing the more concentrated salt solutions.

In a later article^ Richards and Archibald showed that this
gradual change is caused by the formation of a complex mercuric
ion by the intei action of the alkali chloride and the calomel. This
decomposition is very slightly affected by light or air but is
hastened by elevating the temperature or by increasing the
concentration of the solution of the alkali chloride. The de-
composition is almost negligible in o . i N solutions. This
work was corroborated by experiments carried on simul-
taneously by Gewecke.^

Sauer,^ in an extended study of various electrodes, includ-
ing the systems

Hg— HgCl— N KCl

Hg— HgCl— 0.1 N KCl

Hg— HgCl— N HCl

Hg— HgCl— o.i N HCl

Hg— Hg^SO — N H2SO,

Hg— Hg,S04— o. I N H,S04

concluded that the normal potassium chloride-calomel elec-

1 Z. physik. Chem., 24, 39 (1897).

2 Ibid., 40, 385 (1902).

3 Ibid., 45, 685 (1903).
*Ibid., 47, 146 (1904).

590 Loomis and Acree

trodes can be made up with a slightly greater degree of uni-
formity than can the decinormal electrodes, which were found
to be reproducible to within about 0.2 millivolt. Light was
found to have no effect upon the potential of the calomel
electrodes. By direct comparison of the normal and deci-
normal electrodes he found that if the former is assumed to
have a value of o. 560 volt at 18°, the latter will have a value
of 0.612.

Sauer's observation as to the greater uniformity of the
normal electrodes has been corroborated by Lewis and Sar-
gent,^ who have also placed emphasis upon the purification
of materials and uniform methods of preparation of the calo-
mel-m^ercury paste.

Palmaer^ determined by the drop-electrode method that
the absolute potential of the decinormal electrode at 18° is
— 0.5732 ± 0.0003. At the same temperature the value of
the normal electrode is — 0.56. This gives a difference of
0.013 between the potentials of the normal and decinormal
electrodes. By direct comparison Sauer found the difference
at 18° to be 0.052. Since the procedure of Sauer is probably
a far more accurate method for determining differences in poten-
tial than is the drop-electrode method, it has seemed best to
adopt Wilsmore's standard as the zero of potential, viz., the
potential of the hydrogen electrode toward a solution normal
with respect to hydrogen ions. If we use Sauer's value for
the difference between the potentials of the normal and deci-
normal electrodes, the value of the decinormal electrode,
according to Wilsmore's standard, becom^es — o. 283 — 0.052 =
— 0.335 at 18°. At 25° the potential of the decinormal
electrode becomes — ^0.335 — 0.0008(25° — 18°) = — ^0.3406.

Besides the work briefly reviewed above, there has appeared
an immense amount of work involving the use of the hydrogen
electrode, the calomel electrode or other standard elec-
trodes. In this connection may be mentioned the work of
Lorenz and Mohn^ on the neutral point of the hydrogen elec-

1 J. Am. Chem. Soc, 31, 362 (1909).

2 Z. physik. Chem., 59, 129 (1907).
^ Ibid.. 60, 422 (1907).

Study of the Hydrogen Electrode 591

trode; that of lyorenz and Bohi^ on the electrolytic dissocia-
tion of water; that of Lewis and Rupert^ on the chlorine
electrode; that of Naumann^ on the electromotive force of the
hydrogen-cyanogen gas element; that of Schoch^ on the oxy-
gen electrode; and many others.

III. THEORETICAL DISCUSSION

The theory of the hydrogen electrode is generally familiar.
It will only be recalled here that according to Nemst the poten-
tial of the electrode toward the solution in which it is immersed
is dependent upon the pressure of the hydrogen gas and upon
the osmotic pressure of the hydrogen ions in the solution.

In the comparison of a calomel electrode against a hydrogen
electrode in a solution whose hydrogen ion concentration
is H', we find that if 7: represents the observed electromotive
force, 7Zi the potential of the calomel electrode against a hydro-
gen electrode when immersed in a solution with unit concen-
tration of hydrogen ions, and tt, the contact potential between"
the solutions of the system, then the equation

Tt = 7Z^ ^ log,. H' + TT^

holds when the hydrogen gas is under atmospheric pressure.
From this equation, when T = (25 -f 273)° we find that

0.0591

The value of n is obtained by actual measurement with the

potentiometer, n^ is calculated from some system in which

RT
Tz^ and -7^ logio W are known and % has been previously meas-
t

ured, and n^ is calculated. The best system for determining

the value of tTj is

H2— Pt I o. I N HCl I 0.1 N KCl 1 HgCl— Hg

in which it can be measured, -7:;- log« H' = 0.0591 X logj^

1 Z. physik. Chem., 66, 733 (1909).

2 J. Am. Chem. Soc, 33, 299 (1911).

3 Z. Elektrochem., 16, 191 (1910).

•• J. Phys. Chem., 14, 665, 719 (1910).

592 Loomis and Acree

0.0922, and 712 can be calculated by some such formula as
that of Planck.

Because of the difficulty of calculating 7C2 exactly in many
cases, attempts have been made to eliminate this potential
by interposing between the two solutions in question a satura-
ted solution of some highly soluble salt, the two ions of which
have nearly the same migration velocity, such as ammonium
nitrate, potassium chloride,' and others. The use of ammo-
nium nitrate for this purpose has been advocated by Abegg
and Gumming,^ who claim that it practically eliminates the
contact potential. On this assumption it was used by Denham
in his measurements of the hydrolysis of aniline
hydrochloride and a number of inorganic salts. That it
does not do away entirely with the contact potential was
shown by Desha. Some measurements of our own in this
connection will be spoken of later.

In order to show a change of o . i per cent, in the hydrogen
ion concentration of a solution, the measurements must be
accurate to within 0.000025 volt, and an accuracy of o.ooooi
volt was striven for. The question of the effect of temperature
also comes into consideration. Since i ° makes a difference of
0.0008 volt in the potential of the decinormal calomel elec-
trode, the temperature had to be kept constant to within
about o°.oi.

IV. EXPERIMENTAL

I. Apparatus
The electromotive force measurements were all made with
a Leeds and Northrup potentiometer calibrated by the Bureau
of Standards. All measurements were made by the zero
method, that is, the potentiometer was adjusted until there
was no deflection of the galvanometer. The galvanometer
was a Leeds and Northrup special high-sensibility, short-
period instrument of the Marvin type. It had a sensibility
of 117, a period of i . 7 seconds, and a resistance of 215 ohms.

' After this article was in type we learned of the very important work of Bjerrum
on the use of a saturated solution of potassium chloride to eliminate contact potential
(Z. Blektrochem,, 17, 389; Z. physik. Chem.,53, 428). His results agree very closely
•with ours.

2Z. Blektrochem., 13, 17 (1907).

study of the Hydrogen Electrode 593

As primary standards of potential vve used two Weston
standard cells kindly loaned to us by the Bureau of Standards
and calibrated by them from time to time.

The apparatus was tested for leakage currents and ther-
mal effects by making various commutations and found to be
free from them within the limit of accuracy of the work. With
this apparatus measurements could be made to within o.ooooi
volt with a high degree of accuracy. The apparatus was tested
by measuring the electromotive force of one standard cell
against the other.

Value obtained by Bureau of Standards = i. 01892.

Value measured on the potentiometer = i .01892.

Value after correcting in accordance with calibration of
potentiometer = i. 01 892.

The experiments were all carried out at 25° C. The ther-
mometer used in the bath was compared about once a week
with a Beckmann thermometer, which was in turn compared
to within o°.oo2 with two mercury thennometers calibrated
by the Bureau of Standards to about o°.ooi.

For a constant- temperature bath there was used a glass aqua-
rium 36 X 16 X 15 inches, partially filled with oil, as illustrated
in Figs. I, 7, ya. At first an attempt was made to use
an ordinary water bath, then an oil bath immersed in a water
bath, but electric leakage currents in both cases made it
necessary to adopt the oil bath. The oil used was a light
lubricating oil, very transparent, nearly colorless and odor-
less, and free from sulphur, as was shown by the mercury
test. The heating was accomplished by an electric light,
(L) in Fig. 7, which was regulated by a relay and thermo-
regulator. A fan stirrer situated at one end of the aquarium
drove the oil down and under a glass plate placed four inches
above the bottom to the further end of the bath, where oil
rose and returned through the thermoregulator and above
the plate to the stirrer. The glass plate also served as a sup-
port for the apparatus used. Under the glass plate is a cool-
ing coil not shown in the figures. The thermoregulator was
a toluene grid of the type in use in this laboratory. The tem-
perature regulation was constant to within o°.oi.

594 Loomis and Acree

The hydrogen used for the hydrogen electrode was generated
electrolytically from ten per cent, sodium hydroxide solu-
tion, as shown in (/), Fig. 7, nickel electrodes being used.
A current of about one ampere was generally employed. To
remove the last traces of oxygen from the hydrogen, it was
passed through an electrically heated tube containing palladium
asbestos. The tube (M in Fig. 7 and A in Fig. 7a) was
made of Jena combustion tubing, 6 mm. in diameter, and was
fitted with a mercury trap at one end. It was covered with
a layer of asbestos and inserted in a close fitting brass tube
(N in Fig. 7) having an inner diameter of 9 mm. Around
the brass tube were wrapped 60 ohms of "No; 38 Nichrome"
ribbon, the layers being insulated from each other by asbes-
tos paper. When this coil, in series with 73 ohms (a 32 and
a 16 c. p. lamp in parallel) was connected to the no- volt city
circuit a temperature of 170° C. Was obtained. A slow flow of
gas through the tube had no appreciable effect upon the tem-
perature.

From the palladium asbestos tube the hydrogen passed
through a washing apparatus (B, in Figs. 7 and 7a),
which contained the same solution as that used around the
hydrogen electrode.

The general arrangement of the bath, motor, hydrogen
generator and potentiometer is shown in the accompanying
photograph. Fig. i. Since the metallic tank in which the
aquarium is placed hides the lower part of the apparatus,
a more detailed diagram of it is given later in Fig. 7.

2. Calomel Electrodes

(a) The Preparation of the Materials used in the Calomel
Electrodes. Purification of Mercury. — About thirty pounds
of mercury was purified by washing it with 3 per cent, nitric
acid for 24 hours in a modified form of Desha's mercury ap-
paratus, illustrated in Fig. 2. In this apparatus three im-
portant changes have been made in the form described by
Desha. In his apparatus the mercury, descending from the
reservoir (A), entered a trap and overflowed into the tube

1 This Journal, 41, 152 (1909).

Study of the Hydrogen Electrode 595

leading to the nitric acid. This caused a "dead" space which
perhaps slightly decreased the efficiency of the washing. In
the apparatus shown in the figure this dead space has been
eliminated by the substitution of the bulb {B), which keeps
all the mercury in circulation. A more important change has
been made, however, in the method of spraying the mercury
into the acid solution. In Desha's apparatus the mercury
streamed through holes in a glass bulb. Hildebrand^ sug-
gested the use of muslin for spraying mercury into acid. Pro-
fessor Morse and Dr. W. W. Holland in this laboratory have
improved upon muslin by using No. 21 bolting silk, which has
an extremely fine mesh. This practice has been incorporated
in this apparatus, the silk being tied with silk thread to the
glass tube at (C).

Another change consists in the substitution of a tube (£)
of 3 mm. internal diameter for the i mm. tube which Desha
used to draw the mercury from the bottom to the top of the
apparatus. This change doubled the rapidity of washing.
With the I mm. tube three minutes were required for 100 cc.
of mercury to circulate through the apparatus; only one and
a half minutes are required with the larger tube. This form of
apparatus makes possible the electrolysis of the mercury simul-
taneously with the washing, the column of mercury above
the silk being made the anode and a piece of platinum foil
introduced at {D) the cathode. The platinum cathode was
inclosed in a silk bag to prevent the deposited metal from
dropping back into the solution. After being washed about
500 times through nitric acid the mercury was rinsed with
water and allowed to stand under concentrated sulphuric
acid until used.

The mercury thus purified was distilled four times in a cur-
rent of air in an electrolyticalty heated Hulett vacuum still.
An attempt was made to determine the relative purity of the
different samples of mercury by Hulett 's^ electromotive-
force method by using the sample distilled four times as a
standard. Although by this method one part of zinc in 10**

' J. Am. Chem. Soc, 31, 933 (1909).
^ Phys. Rev., 21, 388 (1905).

596 Loomis and Acree

parts of mercury can be detected, no difference could be ob-
ser\'^ed between the samples distilled one, two, three and
four times, respectively. We understand that no differences
can be detected in the electromotive force of standard Weston
cells made in the Bureau of Standards from different samples
of mercury purified in a manner similar to ours.

Mercury distilled three times was used for the preparation
of calomel and in the calomel electrodes.

Preparation of Calomel. — ^Pure calomel was prepared by
dissolving thrice-distilled mercury in redistilled nitric acid,
an excess of mercury being present, then pouring this solution
into dilute nitric acid and precipitating the mercurous chlor-
ide by the addition of hydrochloric acid with constant stirring.
The calomel was filtered and washed thoroughly to remove
hydrochloric and nitric acids. It was then shaken with suc-
cessive portions of water for several days in a shaking ma-
chine, then with a dilute solution of potassium chloride and
finally with a o. i N potassium chloride solution made by dis-
solving 7.456 grams of ignited recrystallized potassium chlor-
ide in conductivity water and diluting the solution to one
liter. During the entire procedure free mercury was present
and the calomel was protected from the light by the use of
bottles painted black.

(b) Form of Cell. — -Some preliminary experiments were
next carried out to determine the relative value of different
forms of cells for use in comparing the hydrogen electrode
against the calomel electrode. Four different types were used
and the one finally decided upon as most efficient and best
meeting our requirements is that shown in Fig. 3. The cell
consists of a tube (A) about 2 cm. in diameter and 15 cm.
high, into the bottom of which is sealed a platinum wire with
which contact is made through a side arm (B) containing
mercury. Over the top of the cell fits a cap (C) with a ground-
glass joint (D). The cap is attached above to a reservoir
(E), through which liquid can be poured into the cell. The
side tube (F) is about one cm. in diameter. The stopcock
(//), in which the side tube of the cell terminates, serves to
prevent the diffusion of liquids into the cell ; diffusion is still

^ 5

Study of the Hydrogen Electrode 597

more prevented when (H) and accessible portions of the wide
side tubes are packed with glass or quartz wool. It terminates
in the ground joint (/) (about 22 mm. long) by which one cell
may be connected with another, as shown in the figure, or
with the hydrogen electrode apparatus.
The advantages of this form of cell are :

1. The ground-glass cap can readily be removed to allow
free access to the inside of the cell for cleaning and filling.

2. The mercury contact through the side arm prevents dis-
turbances of the calomel-mercury paste such as are likely to
occur when contact is made by a tube running down inside of
the cell.

3. The stopcock and reservoir above the cell permit the
rinsing out of the cell with fresh solution when there is any
suspicion that impurities have diffused into the side tube.

4. The stopcock at the end of the side tube in a great meas-
ure prevents diffusion.

5. The large diameter of the side tube gives a low resistance
to the cell.

6. The cell can be immersed entirely in the oil bath, only
the reservoirs and stopcocks being above the oil.

This was the form of cell adopted for comparison with the
hydrogen electrode. In order that the value used for the
potential of the standard electrode might not be dependent
upon one electrode only, a battery, Figs. 4 and 4a, of ten
cells sealed together was prepared, so arranged that the com-
parison cell could be checked against this battery. These
ten cells were of the same type described above except that the
side tube was left off and instead five electrodes were sealed
to each side of a central tube (A). This central tube was
turned up at each end and ground to fit the ground joint
(B) of the electrode (C), or (/) of the comparison electrode,
shown in Fig. 3; the comparison electrode could therefore
be directly checked against any of the ten electrodes. Any
defective electrode can be emptied, cleaned and refilled at any
time without opening the other nine.

(c) Filling the Cells. — Before the cells were filled they were
first cleaned with chromic acid and then washed thoroughly

598 Loomis and Acree

with water. The platinum wires in the bottom of the cells
were coated with mercury by the electrolysis of mercurous
nitrate solution. The cells were then filled with a strong solu-
tion of potassium hydroxide, allowed to stand 24 hours, washed
with water and treated successively with chromic acid, water
for 2 days, a solution of potassium hydroxide for 12 hours,
dilute nitric acid for 2 hours, water and finally alcohol.

In making up the cells the side arms (E) were first filled with
mercury, that washed in nitric acid being used for this pur-
pose. About 2 cc. of the mercury distilled 3 times was then
placed in the bottom of each cell and on top of tiiis about 4
cc. of the calomel-mercury paste. The apparatus was then
filled with a decinormal potassium chloride solution previously
saturated with calomel. Recrystallized and ignited potas-
sium chloride and conductivity water were used.

In the earlier experiments thick stopcock grease was used
for the ground-glass joint {D) between the cap and the cell.
The cells were then painted over entirely with a black varnish,
especial care being taken to get a good coating of paint over the
exposed edge of the ground-glass joint. The paint was intended
for the double purpose of protecting the calomel from light
and of preventing the oil from dissolving the grease in the
ground-glass joint. In spite of this precaution considerable
difficulty was at first experienced in the creeping of the oil
into the cell. Later this difficulty was obviated by the use of
sealing wax in the ground-glass joint. This accomplished the
purpose desired but was rather inconvenient to use, as the
joint had to be heated upon making up or taking down any
cell. There was, furthermore, the attendant danger of crack-
ing the apparatus, which, however, never occurred. Another
form of joint has been planned which should obviate this
difficulty. It is sketched in Fig. 5. It differs from that
shown above in having the cap fit into the top of the cell
and in having a mercury trap around the base of the joint
at {A). Sealing wax will be used to close the ground joint
at the exposed edge {B).

(d) Measurements with the Calomel Electrodes. — The bat-
tery of calomel electrodes was made up 3 times in all. During

js^r-

Study of the Hydrogen Electrode 599

the first 2 times difficulty was experienced in keeping the
oil out of the cell and individual cells had to be renewed oc-
casionally. The third time sealing wax v/as used in the joints
and this proved efficient in protecting the cell from oil. Only
the results of the third series are given in detail. The first
2 series of readings are briefly summarized.

The battery was first made up on November 28, 1910. Four
days later the maximum variation of the electrodes was 0.05
millivolt. By December 16, 2 of the 10 electrodes were 0.14
and 0.16 millivolt, respectively, from the mean of the other
eight, which differed from each other by a maximum varia-
tion of only 0.06 millivolt. These two electrodes were emp-
tied, cleaned out and made up fresh. The potential of these
two on December 19 agreed closely with that of the others,
there being a maximum variation of o. 10 millivolt. By
January 1 3 the maximum variation had increased to 0.17
millivolt. The battery was taken apart and traces of oil
were found in all the electrodes.

The battery was made up a second time on January 24,
191 1. On January 26 there was a maximum variation of
0.09 millivolt, which increased to 0.17 by February 10. The
caps of the cells were then sealed on with sealing wax. On
February 13 the maximum variation was o.io, which grad-
ually increased to 0.16 by March 7. The battery was then
taken apart and cleaned.

On March 9 the battery was made up for the third time and
the caps sealed on with sealing wax. In the following tables
the calomel electrodes are designated by the numbers i to
12. In cleaning the battery Cell No. 6 was accidentally broken
and after making up the battery it was found that the poten-
tial of No. 7 could not be read because of oil that had crept
in between the platinum wire and the mercury in the side
tube; only the potentials of i, 2, 3, 4, 5, 8, 9 and 10 are given.
The readings are expressed in hundredths of a millivolt. Cell
No. I is taken as the standard electrode and considered posi-
tive and the potentials of the other electrodes are referred to
it. A negative sign before the reading of any electrode means
that that electrode is really positive with respect to No. i.

6oo Looniis and Acree

If, for example, we write No. i : No. 9 = +2, No. 9 has a
potential of two-hundredths of a millivolt less than No. i ;
whereas if we write No. i : No. 9 = — 2, No. 9 has a potential
two-hundredths of a millivolt greater than No. i .

Date

Mar. 10

— II

— 4

— 2

—23

—5

—15

—4

— 2

—3

— 2

—5

— I

—5

—4

— I

—5

—3

—4

— I

— 2

— I

—4

—5

— I

—6

—5

— I

—3

— 2

Apr. I

—3

— 2

—3

— 2

— I

—7

—4

—5

—4

—5

—6

—5

—3

—7

—8

—4

—6

—8

— -2

—5

—4

—5

—7

—6

—4

—5

—6

— -2

—4

—5

— I

—4

—6

—7

Study of the Hydrogen Electrode 601

(made up)

—I

—3

—7

—8

— 13

— 2

—I

—4

—8

—6

— 2

—3

—6

—9

—8

— -2

—3

—6

—8

—3

— 10

—6

—4

— 12

— ^11

—7

—4

— ^11

— 10

—5

— I

— 9

—7

— I

—5

—4

(e) Conclusions Regarding the Electrodes o. i N KCl-HgCl-
Hg. — An examination of this series of readings shows that
decinormal calomel electrodes can be prepared which after
the first 4 or 5 days will vary from each other by not more
than a tenth of a millivolt, the majority being in even much
closer agreement. This agreement lasts for about 3 weeks,
after which there is a gradual increase in the maximum varia-
tion to 0.14 millivolt after 2 months. Essentially the same
facts were obser\^ed in the first 2 series of readings made
when the battery was set up on November 28, 1910, and again
on January 24, 191 1.

The average constancy of the mean of the seven electrodes
is very good. On March 13 the mean potential of the seven
was 2.1, on May 9 it was 2.3, when referred to electrode No. 1.

The constancy of the electrodes is shown by their variations
from the mean at the beginning and at the end of the ex-
periment :

Date

^ar. 13
Vlay 9

2
—6

3
4

—7

8
1

—6

9
0

—4

+ 7

—6

— 10

+ 3

—5

+ 5

+ 13 =

= Total
change

The average change in the potential of the individual elec-
trodes with respect to the mean potential was 0.07 millivolt
in 2 months. That a gradual change in the potential of all
the cells did not occur is shown by the fact that Cell No. 11,
made up on April 27, agreed so closely with the others.

6o:? Loomis and Acree

The average daily change in the potential of the electrodes
was very nearly o.oi millivolt, although larger variations
often occuned.

IvCwis has emphasized the necessity of preparing the calo-
mel-mercury paste for the calomel electrode under uniform
conditions. To test the effect of a different sample of calo-
mel, 2 electrodes were made up from calomel prepared about
6 months before that used in the battery. Before use it was
shaken out with fresh o. i N potassium chloride solution.
No difference in the potential of the electrodes due to the
change in material could be detected.

A few experiments were made to test the effect of light on
the potential of the electrodes. As a rule the cells were painted
black, but v/hen left unpainted and exposed to the electric
light of the bath and the diffused light of the room no change
in potential could be noticed.

J. Experiments upon the Constancy and Accuracy of Reproduc-
tion of the Hydrogen Electrode
In the course of the experiments with the hydrogen elec-
trode there were used 19 platinum electrodes designated by
the numbers i to 19. Of these electrodes three, Nos. 5, 6 and
7, had been prepared by Desha. Nos. 5 and 6 were sheet elec-
trodes of the same style as Nos. 1-4 and 8-19, inclusive. They
were made up of sheet platinum, i X 2 cm. in size, welded to
a piece of platinum wire i . 5 cm. long, which was sealed into
t^(e bottom of a glass tube. Contact was made with the elec-
trode by a small quantity of mercury in the bottom of the
glass tube. No. 7 was a Cottrell gauze electrode^ made by
weaving together with fine platinum wire the edges of two
platinum wire "baskets" from a broken Linneman fraction-
ating column. This gauze sphere was sealed to the end of
a piece of glass tubing, through which the hydrogen passed
from the washing apparatus. Contact was made with this
electrode by a platinum wire running through the tube of
the mercury trap of the washing apparatus. Desha had coated
the platinum gauze with gold and then with iridium.^

> Robertson: J. Phys. Chem., 11, 437. Schmidt and Finger: Ibid., 12, 406.
2 Qstwald-Luther, p. 438. Cottrell, Lewis: private communications.

Study of the Hydrogen Electrode 603

We first used 4 sheet platinum electrodes, Nos. 1-4, inclu-
sive. After the electrodes had been thoroughly cleaned,
they were platinized with a solution made of very pure plat-
inum chloride obtained from Heraeus and quite free from
iridium and other metals, which often occur in platinum
chloride. Electrodes 5, 6, and 7 were also replatinized at this
time with the same solution. In platinizing the electrodes
no special precautions were used. A potential of 2.5 volts
was generally employed and the electrodes adjusted until
there was a fairly rapid evolution of gas. The current was
commutated each 5 minutes until a good coating of platinum
black had been deposited. The behavior of the electrodes
appeared to be independent of the thickness of the platinum
coating, provided it was so thick that the electrodes did not
appear gray.

When platinized, the electrodes were washed with water
and then connected i . 5 hours as cathodes in the electrolysis
of dilute sulphuric acid. They were finally boiled several
hours in water and were then ready for use.

In the earlier experiments the following arrangement was
used for comparing the electrodes: Several electrodes were
passed through a rubber stopper which fitted the outer jacket
of a freezing-point apparatus, and this tube was filled with
the acid solution until the electrodes were about three-fourths
immersed. The hydrogen was introduced into the solution
through a central tube drawn out to a capillary. By this
arrangem.ent the hydrogen could not be bubbled directly
against the electrodes and hence they were rather slow in
coming to equilibrium. The gauze electrode No. 7 could not
be compared with the others as its shape prevented its intro-
duction through the rubber stopper. It could be compared
with any one of the other electrodes, however, and numerous
experiments proved it to have approximately tlie same poten-
tial. These measurements with the gauze electrode are dis-
cussed in the next two sections of the experimental work.

In this preliminary work two comparisons of Electrodes
I, 2, 3, 5 and 6 were made. In each case o. i N hydrochloric
acid was used as the electrolyte. In the first experiment

6o4 Looviis and Acree

the electrodes reached a constant potential after 24 hours.
Four of the five electrodes showed a maximum variation from
each other of 0.08 millivolt. Electrode i, however, varied
by 0.19 millivolt from the mean of the others. It was found
to be oily, so it was washed with ether, alcohol and water
and then replatinized with pure platinum chloride. After
cleaning it thoroughly we compared the electrodes.

In the second comparison the electrodes became constant
in potential after 28 hours. The maximum variation between
any two was o. 1 1 millivolt. Four of the five electrodes were
within o . 02 millivolt of each other.

Twelve new sheet-electrodes were obtained and were desig-
nated by the numbers 8-19. Nos. 8, 9, 12 and 13 were platin-
ized with the pure platinum chloride used for the first elec-
trodes; Nos. 14, 15, 16 and 17 were platinized with ordinary
platinum chloride; Nos. 10 and 11 were platinized first with
pure platinum chloride and then with the ordinary material,
and Nos. 18 and 19 were left bright.

In the comparison of these electrodes another form of ap-
paratus^ was used. This is shown in Fig. 6. In the actual
comparison only one-half of the apparatus was employed.
When the second half was used it contained acid of another
strength for the comparison of the potentials of the electrodes
in acid solutions of two different strengths. The bore of the
ground joint (A) which joins the two parts of the apparatus
was made of the same size as that used in the pair of calomel
cells shown in Fig. 3, so that the hydrogen electrodes might
be measured directly against a calomel electrode. The large
tubes (B) had an inner diameter of 2 . 25 inches and were 6
inches deep. This size enabled us to compare readily 8 or
more platinum electrodes (D) with each other. The hydro-
gen bubbled in through the small side tubes (C), and escaped
through the tube (E), which was bent downward at the top
to prevent the rapid diffusion of air back into the cell.

' We have never noticed any ill effects resulting from the use of rubber stoppers in
this cell. It was found to be impossible to construct a glass stopper of this size capa-
ble of holding a number of platinimi electrodes, but we have now devised another type
of apparatus in which the rubber stopper is absent. Comparisons will show whether
the rubber stopper is objectionable. We can now compare 34 hydrogen electrodes
at once.

riSUR.C <E>

Study of the Hydrogen Electrode 605

Comparison of the Electrodes. — The new electrodes, Nos. 8-
17, were not electrolyzed in sulphuric acid before comparison.
They were merely washed thoroughly with water and then
with alcohol and ether to remove any grease or oil.

In the first experiment Electrodes Nos. 2, 3, 6, 12, 13, 16
and 17 were compared with each other. The detailed meas-
urements are given in the following table. All the electrodes
were completely immersed in 0.1 N hydrochloric acid. No.
3 was taken as the comparison electrode and considered posi-
tive and the readings with it are given in the same way as in
the comparison of the calomel electrodes. The readings are
expressed in hundredths of a millivolt :

Time

May 10,

10.30 A. M.

Started

I . 00 P. M.

—8

-637

— 18

2 .00 P. M.

—4

—272

—9

— 13

—3

4.00 P. M.

—6

160

— 13

5.00 P. M.

—6

— 109

—9

—3

May II,

9 . GO A. M.

— I

— 17

—7

10.00 A. M.

— I

—5

— I

The potentials of six of the seven electrodes show a max-
imum variation of 0.06 millivolt. The maximum variation
of any electrode from the mean of all the electrodes is 0.07
millivolt; the average variation from the mean is 0.03 milli-
volt.

In the second experiment electrodes Nos. i, 5, 8, 9, 10, 11,
14 and 15 were compared with each other. No. 5 is taken as
the comparison electrode. The data of the experiment are
given in the following table :

Time 1 8 9 10 1 1 14 15

May 12, 11.50 A. M. Started

1.45 p. M. —170 —15 —21 —14 6 —347 —75

2.30 P.M. —98—10—14 5 6 —153 —9

3.40 P.M. —36 —4—4 23 —68 —3

4.40 P. M. 28 2 2 3 3 ^52 O

9.00 P. M. 2 O O 1 O 8 O

May 13, 9.30 A.M. — 3 o o — 2 o — 17 — i

12.30 P.M. 2 2 2 O I 2 2

3 . 45 P. M. 3 3 3 33 —3 3

6o6 Loomis and Acree

The maximuiii variation between any two electrodes is
0.06 millivolt. The maximum variation of any electrode
from the mean of all the electrodes is 0.05 millivolt, and the
average variation from the mean is o. 02 millivolt.

In one of tlie earlier experiments Nos. 3 and 5 were found
to have exactly the same potential. We can therefore reduce
all the results of the two above tables to the potentials which
should be given when electrode No. 3 is compared against any
of the electrodes. Considering No. 3 positive, we obtain the
figures •}

3—10 —10 3 3 3 3—5—5—3 3—1 I

The maximum variation of any electrode from the mean
is 0.095 millivolt. The mean variation from the mean is
0.030 millivolt. We shall try the experiment of connecting
diflferent electrodes as cells, or passing a current through them,
to see if the potentials can be made more nearly equal.

The experiments show that the potential of the hydrogen
electrode is easily reproduced to within o. 10 millivolt and that
the potential wliich the electrode gives is independent of the
purity of the platinum chloride used and the thickness of the
coating of platinum black, above a certain limit. To clean the
electrode, it need not be used as cathode in the electrolysis
of sulphuric acid nor boiled with water; rinsing with ether,
alcohol and water is sufficient. The electrodes may be com-
pletely immersed in the acid solution into which the hydrogen
gas is bubbled.

4. Comparison of the Hydrogen Electrode with the Calomel
Electrode

Apparatus and Method of Procedure. — The apparatus used
in the comparison of the hydrogen electrode with the calomel
electrode is shown in Figs. 7 and -ja. The arrangement of
the apparatus there is that which was employed when a solu-

^ In this calculation it is assumed that the relative potentials of Nos. 3 and 5 are
constant. This assumption is justified bj- later experiments in which Electrodes
6 and 7 in different solutions of aniline hydrochloride were found to maintain con-
stant relative potentials. We shall study this point further.

Study of the Hydrogen Electrode 607

tion was used to eliminate the contact potential. The hydro-
gen passed from the palladium asbestos tube (A) through the
washing apparatus (B) to the gauze electrode (G). In most
of the experiments a sheet electrode (H) was also used, so
that one electrode would serve as a check upon the other.
The solution in the hydrogen-electrode chamber also filled the
rest of the piece of apparatus (C). The end of (C) dipped into
the chamber (D) which contained the saturated solution for
eliminating contact potential. This solution was prevented
from diffusing back into the hydrogen-electrode chamber by
the two stopcocks on (C). In the chamber (E) was placed
a o. I N solution of potassium chloride, the same as that in
the calomel cell (F). This solution prevented any diffusion
into the calomel electrode of the solution for eliminating
contact potential.

In the earlier experiments the hydrogen electrode was com-
pared directly with the calomel electrode without the use oi
any solution for eliminating contact potential. In such experi-
ments chamber (D) served as the hydrogen-electrode chamber
and potassium chloride solution was put in (E) . The piece of
apparatus (C) was not used. The stopcocks were always
closed when measurements were not being made, and often
even during measurements. The thin film of solution around
the stopper served to conduct the current, although under
these conditions the measurements were not quite so accurate.
This procedure served to stop diffusion and the attendant
changes in potential.

After some preliminary experiments the following method of
procedure was adopted: The comparison calomel electrode
was first compared with the calomel electrodes of the battery
and then placed in position for use with the hydrogen elec-
trode. The objection might be raised to this method of pro-
cedure, especially in view of the experience of Coggeshall,
that the comparison electrode would change in potential
by being moved around. To eliminate any such source of error
the comparison calomel cell was again compared with the
battery at the completion of the experiment. This was
hardly necessary, however, as the following experiment to

6o8 Loomis and Acree

test the effect of mechanical disturbance shows. Cell No. 14,
one of the comparison electrodes, gave a voltage against
No. 7 of the battery of 0.00029 volt. No. 14 was then moved
around and put back again with the battery. The voltage
of 14 : 7 was 0.00030, a change of only o.ooooi volt. A still
more striking proof of the small effect of mechanical disturb-
ance occurred by accident. On February 3, 14 : 7 gave a
voltage of 0.00032. On the mornihg of February 4 No. 14
fell over in the oil bath, flat on its side. It was quickly picked
up, the side tube below the stopcock freed from oil (the stop-
cock was closed at the time of the accident) and immediately
compared with No. 7:14 : 7 gave a reading of o . 00030, a change
of only 0.00007 volt being caused by the accident.

The figures just given show the reason for the method of
procedure adopted. Whereas the calomel electrodes in the
battery were very nearly constant in value, the potential of
the comparison electrode fluctuated from day to day, being
generally in the neighborhood of two to three-tenths of a milli-
volt lower in potential than the cells of the battery. This is
to be explained by the constant disturbance this cell was sub-
jected to and also to the likelihood of impurities diffusing into
the cell during measurements.

Measurement of the Hydrogen Electrode against the Calomel
Electrode. — A typical experiment in which the hydrogen elec-
trode was compared with the calomel electrode is given below.
The calomel electrode is positive.

An experiment in which two electrodes were used, covering
considerably more time, is given on page 609. The gauze elec-
trode, it will be noted, reaches a constant potential much
sooner than the sheet electrode. "No. 14" is the calomel
electrode, "5" and "gauze" are the two hydrogen electrodes.

The figures given in the seventh column should theoretically
be the difference between those in the third and fifth columns.
The comparison of the observed and calculated differences
shows tlie accuracy of the measurements.

Study of the Hydrogen Electrode

609

S ^

3 £

> > >
S § S

4i. -^ OJ w to K)

y h3 _^ _^ > > >
§ § § § § g S

■1 O „. B
^ ?• St §

^
^

I3 r

wo o o
II o -i^ -i^ -f^

^ "73 ON ON Ov

o rc 00 00 00

000

4^ -t^ -j^

K> bO to

o^ a^ o\
^ ^ ^

4^ ^ On

oooocooo

000
-i^ -i^ -p-

K) to to

On 0\ On
^ OO'^

00 O Ck>

totototototototo

OnOnOnOnOnOnOnOn
OiOiCn ONOiCnCn On
^J^^MOOOO'-l'^

00000

-1^ -1^ 4^ 4^ 4^
to to to to to

On 0^ On On On
4^ 4^ 4^- Cn Cn
vO ON4^ to O

00000000

4^4^4^4^4^4^4^4^
totototototototo

ON On On On On OnCji On
4^4^O0toi-iw\oO
^JCyi4^-fi. ooO-t^-f^

II ffi

O O

§ T

O I

s Q

EL 3 „

^ fi s*

00000

4^ -F^ -f^

4i. 4^ 4^ 4^ 4^

to to to

to to to to to

0^ 0^ Os

On On On On On

ON OS ON

OJ OJ to M M

^ vo N>

VO ^ ON 00 0

000

000000

(n 0- o 2.

3-g "

= D. 2
I3* -■ B
■rt B «

000
ON-<I to

000

000
000
000
O O M

«-4 004^

000000
000000
000000

(O OJ OJ 4^ Cn Cyi

O OOnO O 4^ 00

00000000

booooooo
00000000
00000000

M O i-(Oo4^4^cyi ON
OvoOoOnOO^m

0.C '

6io

Loomis and Acree

H3-Pt,

N HCl — o. lo KCl — HgCl — Hg

lo :

14 = 0.00030

Time

E. M. F.

10.30 A. M.

Started

0.4050

0.4153

0.42547

II . 12

0.42590

0.42610

12.03 P. M.

0.42616

0.42619

Stopped

Observed E. M. F.

0.42619

Correction for calomel cell

+ 0.00030

Barometric pressure

,^0.983 ats.

;

bar. correction

+ 0.00044

E. M. F.

0.42693

rhe following table gives the summary of a number of

)eriments made in this way :

E. M F. cor-

rected for E. M. F.

Pt electrode Bar. pres.

E. M. F.

for calomel corrected

used in ats.

observed

electrode for bar.

"Gauze" 1.014

0.4265

0.4266 0.42624

No. 5 1 .014

0.4261

0.4262 0.42584(?)

No. 3 1 . 000

0.4265

0.4266 0.42660

No. 2 1 . 000

0.4263

0.4264 0.42640

No. 4 1 . 003

0.4266

0.4267 0.42662

No. I 1 . 003

0.4262

0.4267 0.42662

No. 6 0.983

0.4262

0.4265 0.42694

No. 6 1 . 007

0.4266

0.4264 0.42622

"Gauze" 1.009

0.4272

0.4267 0.42647

"Gauze" 1.002

0.4269

0.4266 0.4265

No. 5 0.99S

0.4269

0.4266 0.42665

"Gauze" 0.998

0.4270

0.4267 0.42675

"Gauze" 0.998

0.4269

0.4266 0.42665

No. 5 0.998

0.4268

0.4265 0.42655

Average

0.42656 0.42652

1 The barometric pressure is given in atmospheres; i. e., 760 mm. at 0° C. and
45° latitude. This does not include a correction of about 23.5 mm. for the vapor
tension of the solutions at 25°. This correction seems never to have been made by
others. The error involved when barometer readings are referred back to 760 mm.
partial pressure of hydrogen, as shown by the data presented in this article, is only
about 0.00001 volt for the ordinary pressures.

Study of the Hydrogen Electrode

6ii

If the very low value 0.42584 is omitted the average is
0.42657, whether the barometer correction is applied or not.
If the value 0.42584 is omitted, the mean deviation of the in-
dividual readings from the average is o.oooii, and the max-
imum deviation is 0.00037. As a working average we shall
use 0.4266. This is very close to the value 0.4270 found by
Bjerrum.

Besides the experiments given above, there were also a few
experiments in which values were obtained not agreeing with
the others, but in which it was shown that the platinum elec-
trode was at fault.

— o.i N KCl
o . 00050

Time

1 . 00 P. M.

2 .24

2 .36 P. M.

2.38 P. M.
2.41 P. M.
2.44 P. M.
2.48 P. M.
2.55 P. M.

3.06 P. M.

Electrode moved

Electrode turned around
Experiment stopped

No.

E. M. F.

Started
0.3666
0.3675

o.36r6

0.3674

0.36S2

0.368

0.3694

03659

On inspection the coating of platinum black was found
to be very thin. The electrode was replatinized with pure
platinum chloride and electrolyzed in sulphuric acid.

We then repeated the above experiment.

Hj — PtNo. 1—0.1 N HCl — o.i N KCl — No. 14
T^ • T. _ ^ ^^^.^ ga^j- pres., 1003 ats.

Time
9.00 A. M.
10.08 A. M.
10. 13 A. M
10.23 A. M.
10.29 A. M.
10.44 A. M.
10.52 A. M.

10 : 14 = 0.00047.

E. M. F.

Started
0.42524
0.42567
0.42604
0.42615
0.42619
0.42620

Time
II .01 A. M.
II. 18 A. M.
II .40 A. M.

0.42615
+ 0.00047

■ — o . 00008

0.42654 -

E. M. F.
0.42620
0.42630
0.42615

obs. E. M. F.
corr. for No. i
corr. for bar.
corr. E. M. F.

6 12 Loomis and Acree

It is seen that the value 0.42654 now given by Electrode
No. I is in good agreement with the average value of a large
number of experiments given above.

In the experimental work dealing with the comparison of
the hydrogen electrodes Electrode 4 was broken. In resealing
the platinum wire into another piece of glass the platinum
black was turned to gray by ignition. This gray electrode
was tried in one experiment :

H2 — Pt^o. 4 — 0.1 NHCl — 0.1 NKCl — No. 14

10/: 14 = 0.00014

Calomel electrode considered positive

Time E. M. F.

9.00 A. M. Started

10. 12 p. M 0.00344

10. 16 A. M. 0.00280

10.23 A. M. 0.002..;0

10.48 A. M. 0.003 6

Stopped

Electrode 4 was now platinized with ordinary platinum
chloride solution, electrolyzed in sulphuric acid, washed with
water and again used to check the above experiment. This
time a value of 0.4267 was obtained.

Discussion of Results. — In the value 0.42657 of the poten-
tial difference measured in these experiments are included
three factors, the potential of the calomel electrode, the poten-
tial of the hydrogen electrode and the contact potential of
the two solutions, decinormal hydrochloric acid and deci-
normal potassium chloride. For calculating the contact poten-
tial of two solutions various formulas have been proposed.
The first formula was that of Planck.^ When applied to two
solutions having the same concentration this formula becomes

n2S^ = 0.059 log,, :^^-^^

where Wp u^, v^ and V2 represent the migration velocities of
the anions and cations of the two solutions. Using the data
of Kohlrausch and Holbom we obtain at 25 ° the values
H+ = 352.1, K+ = 74-5. and CI" = 75-3

I Wied. Ann., 40, 561 (1891).

Study of the Hydrogen Electrode 613

Substituting these data in the equation we obtain 0.0274
as the value of the contact potential at 25 °.

There have been various modifications of this formula, for
example, that of Henderson^ and that of Lewis and Sargent,^
Bjerrum,^ in a discussion of the accuracy of Planck's and Hen-
derson's formulas, gives 0.0277 as the value of the contact
potential between o . i N potassium chloride and o . i N hydro-
chloric acid. He also noticed that in the case of some solu-
tions there was a small change in the potential at first, due to
diffusion at the planes of contact between the solutions.

Lewis and Sargent's formula has the form

RT. X,

where X^ and .^2 represent the equivalent conductivities of the
two solutions. Lewis and Rupert^ give the values 389.9 for
the equivalent conductivity of o. i N hydrochloric acid and
128.8 for o.i N potassium chloride. These data substituted
in the above equation give 0.0284. The same data when ap-
plied* to Planck's original formula give 0.0266. In the arti-
cle by Lewis and Sargent o. 0286 is the value given^ and further
evidence for this value is found in the fact that it is identically
the same as the value found by Sauer for the potential differ-
ence of the combination

Hg — HgCl — o. I N KCl — o. I N HCl — HgCl — Hg

Lewis claims to have ample proof that o . i N potassium chloride
and o . I N hydrochloric acid are equally dissociated (86 per
cent.). If the concentration of chlorine ions is the same in
each solution, of course the potential of the o . i N potassium
chloride-calomel electrode will be identical with the potential
of the o . I N hydrochloric acid-calomel electrode, and the whole

1 Z. physik. Chem.. 69, 118 (1907); 63, 325 (1908).

2 J. Am. Chem. Soc, 31, 363 (1909).

3 Z. Elektrochem., 17, 58 (1911).

^ J. Am. Chem. Soc. 33, 306 (1911),

5 By using the data given by Lewis and Sargent (J. Am. Chem. Soc, 31, 363) in
the footnote on page 365, we calculate the value 0.0293 volt instead of 0.0286 volt
given by them at the top of the page. We also find the value 0.0277 volt instead of
0.0271 volt given there.

6i4

Loomis and Acree

difference found in comparing the two electrodes will be due
to the contact potential.

We thus see that by the use of different data and formulas
the calculated values of the contact potential between o. i N
hydrochloric acid and o. i N potassium chloride vary from
0.0266 to 0.0286. As this difference is much greater than is
desirable, it was attempted to use some solution to eliminate
the contact potential.

Experiments to Determine the Efficiency of Several Solutions
for Eliminating Contact Potential. — Following the suggestion
of Abegg and Gumming, ammonium nitrate was first used.
In these experiments the arrangement of apparatus was that
shown in Fig. 7.

The summary of three experiments with the combination

H2 — Pt — o. I N HCl — saturated NH4NO3 — o. i N KCl —
HgCl - Hg

is as follows ;

Experiment
I
2
3

Bar. pres.
in ats.

0.988

I .009

I .002

Corrected for
calomel cell

0.3990

0.3998

0.3995

Average value 0.3994

Corrected for
bar. pres.

0.39931
0.39957

o . 39946

0.39944

The presence of the saturated ammonium nitrate solution
causes a difference in electrom.otive force of o. 4266 — o. 3994 =
0.0272.

The requisites of a good salt for eliminating contact poten-
tial are that it shall be very soluble and that the velocities
of the ions shall be nearly equal. Potassium chloride, while
not nearly as soluble as ammonium nitrate, has almost iden-
tical velocities for its ions. Saturated ammonium nitrate
is about II N; saturated potassium chloride at 25°, 4.12 N.
The effect of 4 . 1 2 N potassium chloride solution was next
tried. The following table is a summary of the experiments.
Two electrodes were used in each case :

Study of the Hydrogen Electrode 615

— Pt -

- o.iNHCl — :
HgCl

Bar. pres.
in ats.

I .006
I .000

Average value

saturated KCl — o.iNKCl
-Hg

E. M. F.

Experiment

Corrected
calomel cell

0 . 4002

0 . 4000

Corrected for
bar. pres.

0 40005

0. GOO5

0 . 40000

0.4001

0 . 4000

The saturated potassium chloride solution causes a dififer-
ence of 0.4266 — 0.4000 = 0.0266 volt.

It was impossible to tell from these experiments whether
it was better to use potassium chloride or ammonium nitrate,
so a series of experiments was run with diflferent concentra-
tions of acids. Here it was possible to calculate the theo-
retical difference in the electromotive force due to the known
change in hydrogen ion concentration. By comparing this
theoretical difference with the observed difiference, the rela-
tive efficiency of different solutions for eliminating contact
potential could be determined. Besides ammonium nitrate
and potassium chloride, the effect of potassium iodide, potas-
sium bromide and calcium acetate was determined.

In calculating the hydrogen ion concentration of the hydro-
chloric acid solutions used the following dissociation values
were taken:

N HCl 81.0 per cent. Determined by Sauer
. o . I N HCl 92 . 2 per cent,
o.oi N HCl 96 . 9 per cent. [ Determined by A. A. Noyes'

o . 001 N HCl 100 . o per cent.

The following table summarizes the results :

1 The Electrical Conductivity of Aqueous Solutions, Carnegie Institution Publi-
cation No. 63, page 141

6i6

Loomis and Acree

M CO
CO W

o o

1-^

00
lO

o o

i! o 2i CO CO

"^■S CO CO

CO CO

r^ t^oo oo
Tt- Th O ^<
W N cs c«

M M rj- -^
»0 lO CO CO

6 6 6 6

vO vO 00 o

CS M i-i N
lO lO CO CO

b "-r,

00 00
00 00

O O

00 00
ID lO

O O

lO lO

o o

LO lO

o o

!3.S

Study of the Hydrogen Electrode

617

»o

Tf-

o o
o o

ro ro '^ "*

0000

10 10 fO CO

0000

0 0

C< «N

r^vo

10 10

r^ r^

i^ r^

vO vo

10 >o

On on

^ '^

10 10

ro rO

iigh'OvOcOtO OOOnOn lOiO

fc"'3odo6 6606 do

•-I VO vO 00

10 10 MM

o o

t^vO 00 00

MM ON ON

o o

00 ro vO 00

10 10 MM

W M

o d

6i8 • Loomis and Acree

The results given in the above table indicate that potassium
chloride is by far the most efficient of the salts used for elimina-
ting contact potential. This conclusion is confirmed by some
results obtained with the decinormal hydrochloric acid-calo-
mel electrode.

Potential of H^—Pt—o . i N HCl—o . i N HCl—HgCl—Hg.—
Two decinormal hydrochloric acid-calomel electrodes were
prepared and measured against the hydrogen electrode in deci-
normal hydrochloric acid solution. The following corrected
values were obtained for the electromotive force of the series
Hj — Pt — o. I N HCl — o. I N HCl — HgCl — Hg:

Bar. pres.

E. M. F. corr.

in ats.

E. M. F.

for bar. press.

0.994

0.3999

0 . 40005

0.999

0 . 4003

0 . 40006

I .DIG

0.4005

0.40024

Average o . 4002 o . 400 1 2

If we accept Lewis's conclusions that the potential of the
o . I N HCl — HgCl — Hg electrode is the same as that of the
o . I N KCl — HgCl — Hg electrode, then the difference be-
tween the electromotive force of the element

Hj — Pt — o. I N HCl — o. I N KCl — HgCl — Hg
and that of the element

H2 — Pt — o. I N HCl — o. I N HCl — HgCl — Hg

should be equal to the contact potential of the system o . i N
HCl — o.iNKCl. The actual difference which we find is
0.4266 — 0.4001 = 0.0265. This value is very close to the
difference which we find between the electromotive force of
the element

H2 — Pt — o. I N HCl — o. I N KCl — HgCl — Hg

and that of the element

H2 — Pt — o. I N HCl — saturated KCl — o. i N KCl —

HgCl — Hg

viz., 0.4266 — 0.4000 = 0.0266, the change in potential due

to the presence of concentrated potassium chloride solution.

If, therefore, the assumption of Lewis in regard to the iden-

Study of the Hydrogen Electrode 619

tity in potential of the hydrochloric acid and potassium chlor-
ide electrodes is granted, we reach the conclusion from the
experiments with the hydrochloric acid electrode, as well as
from those in which potassium chloride was used to eliminate
contact potential, that a saturated solution of potassium chlor-
ide eliminates almost completely the contact potential of liquid
systems, at least those composed of potassium chloride-hydro-
chloric acid.

It will be noticed, however, that the acceptance of the above
assumption of Lewis also involves the acceptance of the value
of 86 per cent, for the dissociation of o. i N hydrochloric acid.
If this value is used the agreement of the figures obtained for
the efficiency of potassium chloride for eliminating contact
potential is by no means so good unless a corresponding de-
crease in the dissociation of the other concentrations of hydro-
chloric acid is assumed. Inasmuch as the value of 92 . 2 per
cent, for the dissociation of o. i N hydrochloric acid is the
value generally accepted, this is the value which has been
used in calculating the potential of the calomel electrode,
but the other values are also discussed.

It should be stated that the values obtained in the experi-
ments involving the hydrochloric acid-calomel electrode are
by no means as certain as those with the potassium chloride-
calomel electrode, for the reason that only two acid electrodes
were prepared. Work is now in progress on the HCl-HgCl-
Hg and the H2S04-Hg2S04-Hg electrodes.

Potential of the Decinormal Calomel Electrode

On the assumption that a concentrated solution of potas-
sium chloride entirely eliminates the contact potential of the
system o. i N HCl — ^ o. i N KCl, the value of the electrode
o . I N KCl-HgCl-Hg becomes o . 4000-0 . 059 1 [ — log (o . 1006 X
0.922)] = 0.3390. If the dissociation of o.i N hydrochloric
acid is taken as 86 per cent., the corresponding value of the
electrode becomes 0.3372. The value obtained by Sauer is
0.3406. If the contact potential of o. i N HCl — o. i N KCl
is given the value 0.0284 assigned by Lewis, or 0.0286 found
by Sauer, experimentally, and 86 per cent, is taken as the

620 Loomis and Acree

degree of dissociation of 0.1 N hydrochloric acid, the vakie
of the calomel electrode becomes 0.3355. It is evident that
we cannot draw final conclusions until we know more accurately
the per cent, of ionization of all the electrolytes concerned and
have a very accurate method to calculate contact potential,
and an experimental method to eliminate it completely. Such
measurements of the electromotive force of various systems
will perhaps be very helpful in this direction.

In the experiments in the following article the value
o . 339 is used as the potential of the system o. i N KCl —
HgCl — Hg, and the value o . 3355 is compared with this in
the discussion.

SUMMARY

This series of experiments has shown that :

1 . Calomel electrodes, o. i N KCl — HgCl — Hg, can be pre-
pared which for the first three weeks vary not more than o. 10
millivolt. With longer standing the variation slowly in-
creases.

2. Platinum electrodes can be prepared which, when used
as hydrogen electrodes in o. i N hydrochloric acid, show a varia-
tion from the mean value of less than 0.10 millivolt.

3. The electromotive force of the system

H2 — Pt — o. I N HCl — o. I N KCl — HgCl — Hg

is 0.4266.

4. Saturated potassium chloride solution eliminates almost
completely the contact potential of systems consisting of
potassium chloride and hydrochloric acid.

5. The value of the potential of the electrode o. i N KCl —
HgCl — Hg is 0.339 if the dissociation of o.i N hydrochloric
acid is 92 . 2 per cent. ; it is o. 337 if the dissociation of the acid
is 86 per cent., and 0.3355 if the contact potential is assumed
to be 0.0284. The results obtained by using these different
values are discussed in the following article and are found to
harmonize better with Lewis's data.

Johns Hopkins University
Baltimore, June 1, 1911

THE APPLICATION OF THE HYDROGEN ELECTRODE

TO THE MEASUREMENT OF THE HYDROLYSIS

OF ANILINE HYDROCHLORIDE, AND THE

IONIZATION OF ACETIC ACID IN THE

PRESENCE OF NEUTRAL SALTS

By N. E. Loomis and S. F. Acree

(We are indebted to the Carnegie Institution of Washing-
ton for aid in this work.)

We organic chemists have been nearly completely baffled
in the study of some of our reactions because of the lack of
some direct, accurate and very rapid method for determining
the concentration of hydrogen ions (also hydroxyl, chloride,
bromide, sulphide ions) in the presence of all organic com-
pounds, especially when the system is undergoing change.^
We have methods involving conductivity, catalysis, dilata-
tion, colorimetry, etc., some or all of which can be applied
reasonably well in some cases; but all of these methods may
fail utterly in special cases, especially when the solution is con-
stantly varying in composition.

When acetamide (or any ester, oxime, etc.) is hydrolyzed
in the presence of hydrochloric acid a small amount of the
salt of the amide is formed, and the concentration of this salt
is constantly diminished as the amide disappears.

CH3CONH2 + HCl + H + CI + CH3CONH3CI + CH3CONH3 +
H2O —^ CH3COOH + NH, + CI, etc.

We have no method to-day for determining the concentra-
tions of the constituents of such a system, as there are too
many unknowns in the equation. If, however, we had a di-
rect, accurate, instantaneous method for determining at any
moment the concentration of the hydrogen ions, we could
calculate the concentration of the amide salt and its ions,
and could then determine directly whether this amide salt,
or its ions, or some other constituent, is the substance directly

1 See the address of the Chairman of the Division of Organic Chemistry in Section
C of the American Association for the Advancement of Science, Baltimore, 1908.
Science, 30, 624.

622 Loomis and Acree

yielding the end products. We have the same case in the de-
composition of amides, esters, etc., by alkalis, and just as
great a need for a satisfactory method for determining the con-
centration of hydroxy 1 ions.

With these facts in mind, and with the advantage of the
experiences of Desha* in this investigation, we have taken
up again the attempt to apply the hydrogen electrode to this
problem. In the present communication we are presenting
some experiments bearing on the accuracy of the hydrogen
electrode for determining the concentration of hydrogen ions
in the presence of the organic substances aniline and acetic
acid. We have chosen these simple compounds because the
substances are stable, the constants which we wish to measure
have been accurately determined by other methods, and we
eliminate the uncertainties due to the changes in a reacting
system. The question of the rapidity attainable in measuring
the concentration of the hydrogen ions of a solution is consid-
ered in the next article by Desha and one of us. So many
more unforeseen difficulties have beset us in this work with
organic com.pounds than ever occur in work with inorganic
substances that we shall present these difficulties rather fully
for the benefit of others.

Hydrolysis of Aniline Hydrochloride
One of our first experiments to learn whether the hydrogen
electrode might be applied to organic reactions was the de-
termination of the hydrolysis of aniline hydrochloride.

The hydrolysis of this salt had been carefully determined
by Bredig^ by the conductivity method to be 2 . 63 per cent,
at 25° in N/32 solution. Denham had applied the hydrogen
electrode to this problem and determined the hydrolysis of
aniline hydrochloride in N/16, N/24 and N/32 solutions.
For the N/32 solution he obtained the value 2.58 per cent.,
a result agreeing remarkably well with the value determined
by Bredig for the same solution. There are two small points
in Denham 's work, however, in which there is a chance for
difference of opinion. In the first place the number 2.58 is

> Desha: Diss., Johns Hopkins Univ., 1909.
2Z. physik. Chem.. 13, 289 (1894).

Application of the Hydrogen Electrode 623

not the percentage hydrolysis of the N/32 solution of aniline

1 J 1 1 -J L ^ ^L U.1- u.- [H' concA X 100 ^ ,
hydrochloride, but rather the ratio ^^ — p . To ob-

tain the degree of hydrolysis this value must be divided by
0.96, the degree of dissociation of N/32 hydrochloric acid.
This raises the degree of hydrolysis to 2.69. The second
point is a much more vital one. He uses the number o . 56 as the
value of his normal calomel electrode. This value is the one
determined by Rothmund by the drop-electrode method at
18°. Denham's measurements were carried out at 25°. If
we apply the temperature factor of the normal calomel elec-
trode as determined by Richards^ we obtain for the potential
of the electrode at 25° the value 0.56 + (7 X 0.0006) =
0.564. If this value is used in the calculations instead of

^ .Li. .L- W cone] X 100 r .,, >T / IX-

0.56, the ratio f r — r^i for the N/32 solution is

^ ' [total salt] '^

found to be 3.02 instead of 2.58, and the per cent, of hy-
drolysis becomes 3. 15, a value differing quite widely from that
obtained by Bredig.

Desha attempted to repeat the experiments of Denham,
along with other experiments of his own, in this laboratory,
but had little success. For the N/32 solution Desha found

XI- X- [H' cone] X 100^ , _-. X ui J 1-- £1

the ratio ^^ — = r^ — r^5 to be 5 . 79. He was troubled chieny

[total salt]

by the decomposition of his material, the solution acquiring

a pink color after the experiment had proceeded for a time.

The aniline hydrochloride prepared by Desha had not been
recrystallized, whereas the material used by Denham was re-
peatedly recrystallized from acetone and finally washed with
ether. Thinking that the cause of the discrepancies between
the results of Denham and Desha might be due to impurities
present in Desha's salt, we prepared our aniline hydrochloride
with considerable care, especially as Dr. Denham had kindly
told one of us of his own difficulties in this connection.

The aniline was fractionally distilled twice, the fraction
boiling between 182° and 183° being used for the preparation
of the hydrochloride. It was dissolved in ether and the hy-
drochloride precipitated by passing in dry hydrochloric acid

1 Z. physik. Chem.. 24, S3 (1897).

624 Loomis and Acree

gas, the solution being kept cold by an ice bath. The white
crystals were filtered off, washed repeatedly with ether and
dried over sulphuric acid and caustic potash in vacuo. This
formed Sample I.

Of this dry salt 4.0478 grams were dissolved in 500 cc. of
conductivity water to form a N/16 solution. A portion of
this was diluted to form a N/32 solution.

The results obtained with these first two solutions are shown
in the following table. In accordance with the practice of
both Denham and Desha, saturated ammonium nitrate solu-
tion was used to eliminate the contact potential. The value
of the calomel electrode used is 0.339.

E. M. F.
Corrected for
Concentration calomel electrode

N/16 0.5066

o . 5003

N/32 0.5124

0-4943
N/32 0.5072

o . 5064

N/32 0.5129

N/32 Indefinite

As will be noted, the results are extremely discordant.
Almost invariably the gauze electrode gave a higher value
than the sheet electrode. To see if the electrodes were at
fault, they were tested, after the first experiment, in a solu-
tion of o.iN hydrochloric acid, the electrodes having first
been washed. with alcohol and ether. Both electrodes gave
the same potential, this fact showing that they were all right.
It was noted in each of these experiments, after the removal
of the electrodes, that there was oil on the surface of the solu-
tion. As no trouble of this kind was experienced with solu-
tions of acids, it is probable that the oil was present at first
in the aniline hydrochloride, or was the product of the de-
composition of some substance present in the aniline hydro-
chloride solution. As in these experiments it had been the
custom to have the sheet electrode partially out of the solu-
tion, the lower value of this electrode is probably due to its
becoming coated with this oil. In subsequent experiments

[H' cone

.] X 100

[total salt]

• 30

•95

.68

•47

•51

.64

.61

Application of the Hydrogen Electrode 625

the sheet electrode was entirely immersed, but in the other
samples of aniline hydrochloride, which were further purified
as described below, no traces of oil were found. No pink
color, described by Desha, was noticed in the solution in any
of the experiments. Constant readings were generally ob-
tained in three hours, the drift being very small after the first
hour and a half. A typical experiment with the N/32 solu-
tion is given below :

— N/32 CeHsNH^.HCl — satd. NH.NOg —
o.iNKCl — HgCl — Hg

Electromotive force

H.-

— Pt

'■gauze + 6

Time

12.06 P. M.

12. II P. M.

12.55 P. M.

1.58 P. M.

2 . 15 P. M.

3. 10 P. M.

14 : gauze 14 : 6

Started

0.5021 0.4995

o . 5058 o . 5042

0.5067 0.5055

0.5067 0.5058

0.5070 0.5062

The remainder of the aniline hydrochloride was further
purified by precipitating it from alcohol by the addition of
ether. The material was filtered off and washed with ether
to remove all traces of alcohol. The salt was dried in vaciio
over solid caustic potash and sulphuric acid. This material
constituted Sample II. The hydrolysis of N/16 and N/32
solutions of this salt was determined as before except that
experiments were also performed with potassium chloride
solution to eliminate the contact potential. The summary
of the results is given in the following tables:

E. M. F. with NHjNOs E. M. F. with KCl

Corr. for rrr/ -, ^, . Corr. for rn/ t v^ .^^^

calomel [H' cnnc.] X 100 calomel [H' cone] X 100

[total salt]

2.13
2.06

(3-18)
(3 • 10)
2.86

2.85

Cone.

electrode

[total salt]

electrode

N/16

0 . 5005

2.92

0.4999

2.99

N/16

0.5009

2.88

0.5086

0 . 500 ;

2.88

0.5095

N/32

0.5126

3 63

O.5161

0.512

3 62

0.5168

N/32

0.5189
0.5190

626 Loomis and Acree

The results obtained in the experiments in which potassium
chloride was used to eliminate the contact potential agree
much more closely with the results obtained by Bredig and by
Denham than do those in which ammonium nitrate was used.

The remainder of the aniline hydrochloride was recrys-
tallized from acetone and the product washed thoroughly with
ether. After it was dried, solutions were prepared from this
Sample III. Only potassium chloride was used in the contact
solution with this sample. The results are summarized below :

E. M. F. with KCl

Bar. pres.

Corr.i for

Corr. for [//' cone] X 100

Cone.

in ats.

calomel electrode

bar.

[total salt]

N/16

I. Oil

0.5088

0.50852

215

0.5092

0.50892

2. II

N/16

I .002

0.5091

0.50905

2.09

0.5089

0.50885

2 . II

N/32

I .Oil

0.5184

O.51812

2.94

0.5190

0.51872

2.88

N/32

I .002

0.5184

0.51835

2.91

0.5184

0-51835

2.91

The agreement between the results of the two experiments
at each concentration is very good in this series and further-
more the results agree well with those obtained with Sample
II when potassium chloride was used in the contact solution,
if the one experiment be excluded in which the values 3.18
and 3. 10 were obtained for the N/32 solution.

A typical experiment in which potassium chloride is used
in the contact solution is shown below:

■N/16 CeHsNH^.HCl (III)— satd. KCl— o. i N
KCl — HgCl — Hg (No. 14)

14 : gauze 14 : 6 6 : gauze

Started

o . 5088 o . 5048

0.5090 0.5072

0.5090 0.5079

0.5091 0.5085 0.00051

0.5091 0.5089 0.00017

1 It is worthy of note that the averages of the electromotive force, and of the per
cent, of hydrolysis, per cent, of ionization, and other factors depending upon the elec-
tromotive force, are approximately the same for long time periods whether corrected
for the barometric pressure or not. The fluctuations of the barometer here in Balti-
more are such that the pressxure averages close to 760 mm. over long time periods.

H3-

"•-gauze + 6
Time

8.42 A.

9.52 A.

10.13 A.

10.28 A.

10.58 A.

12.08 P.

Application of the Hydrogen Electrode 627

A fresh lot of aniline hydrochloride was prepared and was
extracted four times with about 200 cc. of boiling acetone.
The material remaining was well washed with ether, dried,
and used as Sample IV :

Bar. pres.

E. M. F. with KCl

Corr. for

Corr. for [H'

conc/\ X 100

Cone.

in ats.

calomel electrode

bar.

[total saW^

N/16

0.987

0.5087

0.50904

2. 10

0 . 5083

0 . 50864

2.13

N/16

0.989

0.5100

0.51028

1.99

0.5100

0.51028

1.99

N/16

I .009

0.5103

0.51007

2.01

0.5100

0.50977

2.04

N/32

0. 996

0.5182

0.51830

2.9;

0.5180

O.5181O

2.95

N/32

I .009

0.5180

0.51777

2.98

0.5179

0.51767

2.99

Sample V of the aniline hydrochloride was prepared by re-
crystallizing some of Sample IV from alcohol by the addition
of ether. It gave the following results :

Bar. pres.

. M. F. with KCl

Corr. for

Corr. for [//'

conc.'\ X 100

Cone.

in ats.

calomel electrode

bar.

[total salf]

N/16

I .009

0 . 5098

0.50957

2.06

0.5098

0.50957

2.06

N/16

0.996

0.5086

0.50870

2.13

0.5094

0.50950

2.06

N/16

0.997

0.5102

0.51028

1.99

0.5097

0.50978

2.04

N/32

0.988

0.5173

O.51761

3.01

0.5173

O.51761

3.01

N/32

0.998

0.5177

0.51775

2.99

0.5177

0.51775

2.99

N/32

0.999

0.5183

0.51832

2.93

0.5180

0.51802

2.96

Sample VI was prepared by recrystallizing the remaining
material of Sample V from alcohol by the addition of ether.
The results obtained with this material are given in the fol-
lowing table:

628

Loomis and At

cree

Bar. pres.

E. M. F. with KCl

Corr. for

[H' cone.] X 100

Cone.

inats.

calomel electrode

bar.

[total salt}

N/16

0.997

0.5096

0 . 50968

2.05

0 . 5092

0.50928

2.09

N/16

0.995

0.5100

O.51OI3

2 .01

0.5097

0.50983

2.04

N/52

0.995

0.5189

0.51903

2.85

0.5187

0.51883

2.87

N/32

1.003

0.5188

0.51872

2.88

A summary of the results obtained with the different sam-
ples of aniline hydrochloride when potassium chloride was
used as the contact solution is given in the following table:

Aniline Hydrochloride

N/16 [H' cone.] X 100 N/32 [//' cottc] X 100

Sample [total salt] Sample

II 2.13 II

2.06
HI 2 15

2 . II

2.09 III

2. II
IV 2 . 10

2.13

1 . 99 IV

1 99

2 .01
2 .04

V 2.06

2 .06
2.13
2 .06

1 99

2 .04.
VI 2.05

[total

salt]

.18)

.10)

.86

■85

•94

.88

•91

•95

.98

•99

.01

•99

•93

.96

.85

•87

.88

2.09 VI

2 .01
2 .04

2 . 07 Average 2 . 93

0.08 Max. variation from

mean o . 08

Application of the Hydrogen Electrode 629

Excluding the values in the parentheses, we have the aver-
age value of 2.07 for the N/16 solution and 2.93 for the N/32
solution. This gives us the following per cent, of hydrolysis
of N/16 and N/32 solutions of aniline hydrochloride if we use
0.944 and 0.960 as the degree of ionization of N/16 and N/32
solutions of hydrochloric acid.

2.07/0.944 = 2.19 per cent, hydrolysis of N/16 aniline
hydrochloride.

2.93/0.960 = 3.05 per cent, hydrolysis of N/32 aniline
hydrochloride.

C6H5NH3CI ±1;; C6H5NH3+ + Q-

The per cent, of ionization of aniline hydrochloride can be
calculated from conductivity data given by Bredig.^ From
these data the ionization of N/32 aniline hydrochloride is 86.6
per cent, and by extrapolation the ionization of the N/16
solution is found to be 84 . 4 per cent.

Let k^ = the ionization constant of water and k^ = the
affinity constant of aniline.

C6H5NH3+ :^ C6H5NH2 + H+

kb CC6H5NH3+ C'cgH5NH3+

Cc6HsNH3+ = CceHfiNHsCi X ionization

(i — per cent, hydrolysis/ 100) X ionization
volume

k^ __ [Ch+? X volume

kb (i — percent, hydrolysis/ 100) X ionization

For y = 16

ku, (0.0207)2

1 = -, ^—T — = 0.0000324

kb (i— 0.0219) X 0.844 X 16 '^ ^

For y = 32

k^ _ (0.0293)2

kb (1—0.0305) X 0.869 X 32

0.0000318

Average value of k^/kf^ =0.321 X io~^

Tizard^ obtained by colorimetric methods the value o. 242 X

> Z. physik. Chem., 13, 191 (1894).
2 J. Chem. Soc, 98, 2492 (1910).

630 Loomis and Acree

io~* for k^/kf^ while Bredig found 0.24 X io~^ by his conduc-
tivity measurements. Two factors enter into the explanation of
the difference between these results. The first point is that
Tizard assumes that the hydrochloric acid formed by the hy-
drolysis of the aniline hydrochloride is entirely dissociated.
As was pointed out in the discussion of Denham's work, this
assumption is not justifiable. The second factor has to do
with the calculation of our results. It was pointed out in the
discussion of Denham's work that the value assigned to the
potential of the calomel electrode plays a large part in the
value found for the hydrogen ion concentration. This indi-
cates that the greatest source of uncertainty in the deter-
mination of the hydrolysis of aniline hydrochloride is not in
any difficulty in the experimental measurements, as Desha
thought, but in the determination of the potential of the calo-
mel electrode. If the value of the decinormal calomel elec-
trode is taken as 0.3362, the figure adopted by Desha, the
ratio [//' cone] X 100/ [total salt] found for N/16 aniline
hydrochloride becomes 1.85 instead of 2.07. If Lewis's
value of the contact potential between o. i N potassium chlor-
ide and 0.1 N hydrochloric acid is adopted, viz., 0.0284, the
value of the calomel electrode becomes 0.3355 ^^^ ^^ ratio
[H' cone] X 100 /[total salt] becomes 1.81 for the N/16
and 2 .56 for the N/32 solution.
If these values are used

k^/k^ior N/16 -= 0.247 X io~*
^^/fej, for N/32 = 0.242 X io~*

This gives an average of o. 244 X io~^ for k^/k-h, which agrees
well with the values found b}^ Tizard and Bredig. It should
be noted that Bredig used the value 383 for the equivalent
conductivity of the hydrochloric acid formed by hydrolysis of
the aniline salt, whereas Lewis used 389.9. This makes no
appreciable difi'erence in k^/kf^. Bredig's equivalent conduc-
tivity should change with change in concentration according
to the isohydric principle. We intend to redetermine all the
data needed for such work.

Application of the Hydrogen Electrode 631

Experiments with Acetic Acid

To test the applicability of the hydrogen electrode to the
determination of the concentration of the hydrogen ions in
solutions containing weak organic acids, a series of experi-
ments was carried out with acetic acid.

First, two experiments were carried out with o. 5 N acetic
acid, with saturated ammonium nitrate as the contact solu-
tion.

H2 — Pt — o. 5 N CH3COOH — satd. NH4NO3 — o. I N KCl —

No. 14

This series gave the results :

E. M. F. Per cent.

Corr. for dissocia-

calomel electrode tion

0.4817 0.764

0.4822 0.748

With 0.25 N acetic acid and ammonium nitrate, the results
were as follows :

E. M. F.

Corr. for ' Per cent,

calomel electrode dissociation

0.4908 1.069

o . 4905 I . 08 I

With 0.25 N acetic acid and potassium chloride as contact
solution, the results were the following:

Per cent,
dissociation

E. M. F.

Corr. for

calomel electrode

0.4930

0.4927

0.982

0.993

The difference of approximately o . i between the per cent,
of ionization determined with ammonium nitrate and that
determined with potassium chloride is evidently due to the
change in contact potential in the two systems. Judging
from the previously described experiments with potassium
chloride, that solution is the better for eliminating contact
potential and therefore the per cent, ionization of 0.25 N

632 Loomis and Acree

acetic acid is probably nearer o . 985 than i . 080 . By con-
ductivity measurements White and Jones^ found at 25° the
dissociation of 0.5 N acetic acid to be 0.58 and of 0.25 N
acetic acid to be o . 89 (calculated by interpolation) .

Effect of Neutral Salts upon the Dissociation of Acetic Acid. —
The catalytic action of neutral salts is a problem upon which
a great deal of work has been done. The literature of this
field and the principal theories have been summarized by
Acree. ^ Besides the catalytic action of neutral salts upon
the velocity of decomposition of diacetone alcohol by alka-
lis, cane sugar inversion, etc., work has been done upon
the effect of neutral salts on the dissociation of weak acids.
In the study of this problem two methods have been used
heretofore. The conductivity method has been applied
by Arrhenius* and the colorimetric method by Brunei and
Acree^ and by Szyszkowski.^ Arrhenius studied the conduc-
tivity of acetic, formic and phosphoric acids in the presence
of a number of different salts. He came to the conclusion
that the effect of small quantities of neutral salts upon the
dissociation of the acids is much greater at the higher dilu-
tions of the acids than at the lower; whereas if the amount
of salt added is larger, the effect is nearly proportional to the
amount of salt added. The addition of o. 125 N sodium chlor-
ide and o. 5 N acetic acid increased the hydrogen ion concen-
tration by 5 per cent. A number of factors enter into conduc-
tivity measurements which make these results uncertain.
Among these factors are the change in hydration and in vis-
cosity caused by the salt; the possibility of double compounds
or complex ions between the salt and acid; the relatively
small change in conductivity due to any change in the hydro-
gen ion concentration compared to the conductivity of the
salt added ; and other factors.

Szyszkowski based his method upon the change in color of
methyl orange in the presence of weak acids caused by the
addition of neutral salts. He studied acetic and carbonic

I This Journal, 44, 159 (1910).

2/6id., 41, 457 (1909).

3 Z. physik. Chem., 1, 110; 11, 823; 31, 197 (1899).

■I This Journal, 36, 120 (1906).

5 Z. physik. Chem.. 68, 420 (1907); 63, 421 (1908); 73, 269 (1910).

Application of the Hydrogen Electrode

633

acids, using solutions varying from 0.0022 N to 0.046 N.
He interpreted his results to mean that neutral salts greatly
increase the hydrogen ion concentration of weak acids. So-
dium chloride apparently increased the ionization of acetic
acid about 23 times. It should be pointed out, however,
that Kurt Meyer and also Hantzsch have shown that some
dyes unite with salts and form still more deeply colored double
compounds. This tends to throw doubt on the validity
of Szyszkowski's conclusions until further evidence to the
contrary is presented.

It was believed that the hydrogen electrode would prove
serviceable in the study of this problem, and to that end a
series of experiments was performed.

In order that the effects of potassium chloride and ammo-
nium nitrate for eliminating contact potential might be com-
pared, each solution of acetic acid was used with both contact
solutions. The results are included in the following table : The
solutions were prepared by mixing o . 5 N acetic acid with an
equal volume of the different solutions of potassium chloride.
The concentrations given below are the concentrations after
mixing the two solutions. To make the table complete the
results obtained with 0.25 N acetic acid alone are included.
Only the corrected electromotive force readings are given:

NH4NO3 as contact soln.

E. M. F.
corr. for
Solution bar. Dissoc.

0.25 N Acetic 0.49032 1.089

0.49071 1.073
0.49071 1.073

o . 25 N Acetic -(- o . 05 N

KCl 0.49152 1. 041

0.49192 I .026
o . 25 N Acetic + o . i N

KCl 0.49271 0.992

0.49271 0.992
o . 25 N Acetic + o . 5 N

KCl 0.49445 0.925

0.49445 0.925
0.25 N Acetic + 2.06 N

KCl 0.49837 0.797

0.49837 0.797

KCl as contact soln.

E. M. F.

corr. for

bar.

Dissoc.

0 . 49308
0.49296

0.980
0.980

0.49276

0.990

0.49178

1.030

0.49178

1.030

0.48945
0.48945

I. 125
I. 125

0.48575
0.48575

1.300
1.300

634 Loomis and Acree

The results of this series of experiments are uncertain.
According to the experiments with ammonium nitrate, the
addition of a neutral salt appears to decrease the hydrogen
ion concentration ; according to the experiments with potas-
sium chloride, the hydrogen ion concentration appears to in-
crease. The difficulty evidently lies in the contact potential
of the system. We tried to carry out some experiments in
which ammonium nitrate was added to acetic acid, but these
were unsuccessful for some reason. Instead of the potential
becoming constant within about two hours it would continue
to rise, showing a decrease in the hydrogen ion concentration
of the solution. This may be due to a reduction of the ammo-
nium nitrate to ammonia by the hydrogen in the presence of
platinum black.

If we assume the potassium chloride series of results to be
the more accurate, as they have been seen to be in other ex-
periments, then the results are not dissimilar to those ob-
tained by Arrhenius. The addition of o. i N potassium chlor-
ide to 0.25 N acetic acid increases the hydrogen ion concen-
tiation about 4.5 per cent, of the original value. We shall
extend this study in several related directions.

SUMMARY

1 . If we use o . 339 as the value of the electrode o. i N KCI —
HgCl — Hg, the hydrolysis of a N/16 solution of aniline hydro-
chloride is 2.19 per cent, while that of the N/32 solution is
3.05 per cent. If we use 0.3355, calculated from the data
of Lewis, as the value of this electrode the hydrolysis be-
comes 1. 8 1 per cent, for the N/16 and 2.56 per cent, for the
N/32 solutions, values which agree excellently with those of
Bredig and Tizard. The hydrogen electrode gives us then
another instrument for studying these relations between
conductivity and hydrolysis accurately, and we shall extend
these studies to a large number of other organic salts.

2. The addition of potassium chloride to acetic acid solu-
tions slightly increases the dissociation of the acetic acid.

Johns Hopkins University
Baltimore, Md.
June 1, 1911

Application of the Hydrogen Electrode 635

BIBLIOGRAPHY

1. Abegg and Cumming: Z. Elcktrochem., 13, 17 (1907). Elimination

of Liquid Potentials.

2. Acree: This Journal, 41, 457 (1909). Studies in Catalysis.

3. Arrhenius: Z. physik. Chem., 31, 197 (1899). Change in the Strength

of Weak Acids by the Addition of Salts.

4. Arrhenius: Ibid., i, 1 10 (1887). Effect of Neutral Salts on the Saponi-

fication of Ethyl Acetate.

5. Arrhenius and Shields: Ibid., 11, 823 (1893). Electrolysis of Alkali

Salts.

6. Bancroft: Ibid., 10, 387 (1892). Oxidation Elements.

7. Barmwater: Ibid., 28, 424 (1899); 45, 557 (1903); 54, 225 (1906).

Conductivity of Mixtures of Electrolytes.

8. Bjerrum: Z. Elektrochem., 17, 58 (191 1). The ReliabiHty of Planck's

Formula in Determining Contact Potential. Ibid., 17, 389. On the
Elimination of Contact Potentials in the Measurement of Electrode
Potentials. Z. physik. Chem., 53, 428. Elimination of Contact
Potential between Two Dilute Aqueous Solutions by the Introduc-
tion of a Concentrated Solution of Potassium Chloride.

9. Bose: Z. physik. Chem., 34, 742 (1900). Electromotive Activity

of Elementary Gases.

10. Bredig: Ibid., 13, 289 (1894). Afhnity Constants of Bases.

11. Bredig: Ibid., 13, 191 (1894). Contributions to the Stoichiometry of

Ionic Mobility.

12. Bredig and Fraenkel: Z. Elektrochem., 11, 525 (1905); Z. physik.

Chem., 60, 202 (1907). Hydrogen Ion Catalysis.

13. Bronsted: Z. physik. Chem., 65, 84 (1909). The Electromotive Force

of the Hj— O2 Element.

14. Brunei and Acree: This Journal, 36, 120 (1906). On a New Method

for the Preparation of Standard Solutions.

15. Bruner: Z. physik. Chem., 32, 133 (1900). The Hydrolysis of Salt

Solutions.

16. Coggeshall: Ibid., 17, 62 (1895). Constancy of Calomel Electrodes.

17. Denham: J. Chem. Soc, 93, 41 (1908). The Electrometric Deter-

mination of the Hydrolysis of Salts

18. Desha: Diss., Johns Hopkins Univ., 1909.

19. Desha: This Journal, 41, 152 (1909). An Apparatus for the Puri-

fication of Mercury.

20. Euler: Z. physik. Chem., 32, 357 (1900). Neutral Salt Catalysis.

21. Freundlich and Makelt: Z. Elektrochem., 15, i6i (1909). Absolute

Zero of Potential.

22. Gewecke: Z. physik. Chem., 45, 685 (1903). Decomposition of Mer-

curous Chloride.

23. Goodwin: Ibid., 13, 583 (1894). Study of the Voltaic Cell.

636 Loo mis and Acree

24. Henderson: Ibid., 59, 118 (1907); 63, 325 (1908). Thermodynamics

of Liquid Elements.

25. Hildebrand: J. Am. Chem. Soc, 31, 933 (1909). Purification of

Mercury.

26. Hulett and Minchin: Phys. Rev., 21, 388 (1905). The Purification

of Mercury.

27. Jones and White: This Journal, 44, 159 (1910). Conductivity of

Organic Acids, Etc.

28. Kistiakowsky : Z. Elektrochem., 14, 113 (1908). A Method of Meas-

uring Electrode Potentials.

29. Laurie: Z. physik. Chem., 64, 615 (1909). Electromotive Force of

Iodine Concentration Cells in Water and Ethyl Alcohol.

30. Le Blanc: Ibid., 12, 351 (1893). Electromotive Force of Polarization.

31. Lewis: Ibid., 55, 449 (1906). Silver Oxides and Suboxides.

32. Lewis: Ibid., 63, 171 (1908). Calculation of Ion Concentrations from

the Electromotive Force of Concentration Elements.

33. Lewis and Rupert: J. Am. Chem. Soc, 33, 299 (1911). The Poten-

tial of the Chlorine Electrode.

34. Lewis and Sargent: Ibid., 31, 362 (1909). Potential of the Ferro-

ferricyanide Electrode.

35. Lewis and Sargent: Ibid., 31, 363 (1909). Potentials between Liquids.

36. Lorenz: Z. Elektrochem., 14, 781 (1908); 15, 157, 206, 293, 349, 661

(1909). Oxide Theory of the Oxygen Electrode.

37. Lorenz: Ibid., 15, 62, 121 (1909). Zero of Electrochemical Potential.

38. Lorenz and Bohi: Z. physik. Chem., 66, 733 (1909). Electrolytic Dis-

sociation of Water.

39. Lorenz and Mohn: Ibid., 60, 422 (1907). The Neutral Point of the

Hydrogen Electrode.

40. Loven: Ibid., 20, 593 (1896). Theory of Liquid Elements.

41. Lunden: J. chim. phys., 5, 574 (1907). Dissociation of Water.

42. Luther and Michie: Z. Elektrochem., 14, 826 (1908). Electromotive

Force of Uranyl-Urano Mixtures.

43. Maitland: Ibid., 12, 265. Concerning the Iodine Potential and the

Ferri-Ferro Potential.

44. Michaelis and Rona: Ibid., 14, 251 (1908). On the Determination of

Hydrogen Ion Concentrations by Indicators.

45. Nauman: Ibid., 16, 191 (1910). The Electromotive Force of the

Cyanogen-Hydrogen Element.

46. Nernst: Z. physik. Chem., 4, 150 (1889). Electromotive Force Ef-

fect of Ions.

47. Nernst: Ibid., 56, 544 (1906). Electromotive Force of H2 — Oj.

48. Neumann: Ibid., 14, 193 (1894). Concerning the Potential of Hydro-

gen and a Metal.

49. Ostwald- Ibid., 11, 521 (1893). Dissociation of Water Measured by

the Acid-Alkali Element.

Application of the Hydrogen Et,ectrode 637

50. Ostwald: Ostwald-Luther's " Physiko-Chemische Messungen," 3d

Edition, p. 441. Calomel Electrode.

51. Palmaer: Z. physik. Chem., 59, 129 (1907). Absolute Potential of

the Calomel Electrode.

52. Peters: Ibid., 26, 217 (1898). Oxidation and Reduction Elements

and the Influence of Complex Ions.

53. Planck: Wied. Ann., 40, 561 (1891). On the Difference of Potential

between Two Dilute Solutions of Binary Electrolytes.

54. Richards: Z. physik. Chem., 24, 39 (1897). Temperature Coefficients

of Potentials of the Calomel Electrode, Etc.

55. Richards: Ibid., 24, 53 (1897). Temperature Coefficients of Poten-

tials of the Calomel Electrode.

56. Richards and Archibald: Ibid., 40, 385 (1902). Decomposition of

Mercurous Chloride by Dissolved Chlorides.

57. Rothmund: Ibid., 15, 15 (1894). Potential Differences between

Metals and Electrolytes.

58. Salessky: Z. Electrochem., 10, 204 (1904). Concerning Indicators in

Acidimetry and Alkalimetry.

59. Salm: Ibid., 10, 341 (1904). Determination of the Hydrogen Ion Con-

centration of a Solution by the Help of Indicators.

60. Sammet: Z. physik. Chem., 53, 673 (1905). The Potential of the

Iodine Ion Electrode.

61. Sauer: Ibid., 47, 146 (1^04). Standard Electrodes.

62. Schoch: J. Am. Chem. Soc, 26, 1422 (1904). A Study of Reversi-

ble Oxidation and Reduction Reactions in Solutions.

63. Schoch: Ibid., 29, 314 (1907). The Electrolytic Deposition of Nickel-

Zinc Alloys.

64. Schoch: This Journal, 41, 232 (1909). The Behavior of the Nickel

Anode and the Phenomena of Passivity.

65. Schoch: Ibid., 41, 208 (1909). The Electromotive Force of Nickel

and the Effect of Occluded Hydrogen.

66. Schoch: J. Phys. Chem., 14, 719 (1910). Behavior of Iron and Nickel

Electrodes in Various Electrolytes.

67. Schoch: Ibid., 14, 665 (1910). The Potential of the Oxygen Elec-

trode.

68. Smale: Z. physik. Chem., 14, 577 (1894). Studies on Gas Elements.

69. Spohr: Ibid., 2, 194 (1888). Effect of Neutral Salts on Chemical

Reactions.

70. Szyszkowski: Ibid., 58, 420 (1907); 63, 421 (1908); 73, 269 (1910).

Contribution to the Knowledge of Neutral Salt Action.

71. Tizard: J. Chem. Soc, 97, 2477. The Colour Changes of Methyl-

Orange and Methyl-Red in Acid Solution. Ibid., 97, 2492 (1910).
The Hydrolysis of Aniline Salts Measured Color imetrically.

72. Tower: Z. physik. Chem., 20, 198 (1896). Potential Difference at

Ihe Contact Surface of Dilute Solutions.

73. Wilsmore: Ibid., 35, 296 (1900). Electrode Potentials.

ON DIFFICULTIES IN THE USE OF THE HYDROGEN

ELECTRODE IN THE MEASUREMENT OF THE

CONCENTRATION OF HYDROGEN IONS

IN THE PRESENCE OF ORGANIC

COMPOUNDS

By L. J. Desha and S. F. Acree

(We are indebted to the Carnegie Institution of Washing-
ton for aid in this work.)

When an oxime* is formed by the reversible reaction be-
tween a carbonyl compound and a salt of hydroxylamine, the
total amount of oxime formed is of course equal to the amount
of hydroxylamine which disappears. This quantity may be
determined in any particular reaction by titration with a stand-
ard solution of iodine. But the oxime may exist in solution
in three forms, free oxime, oxime salt and oxime cation, and
the quantity measured by the method referred to is equal
to the sum of these three. When it became highly probable
that the oxime cation is one of the substances existing in equi-
librium with the ketone (aldehyde, etc.), and hydroxylamine
salt a means of determining its concentration was sought.

As direct measurement of the concentration of this oxime
cation is out of the question the only feasible method of attack
seems to be a differential one involving the determination of
the amount of free hydrogen ions which disappear in its forma-
tion. That is, the free hydrogen ion concentration at any
time may be calculated from the known amount of acid added
as catalyzer plus that generated by the reaction

R2CO + NH2OH.HCI :;i^ R3C = N0H + HCl + H3O

and minus any disappearing in the reaction

R2C = N0H + HCl :^ R2C = N0H.HC1.

Consideration of the various known methods for estimating
the hydrogen ion concentration during the reaction, and espe-
cially when equilibrium is attained, showed that only two

' Lapworth and Barrett: J. Chem. Soc, 91, 1133; 93, 85. Johnson, Desha and
Acree: This Journal, 38, 258; 39, 300. See Desha's Dissertation, Johns Hopkins
University, 1908, for further work which will soon be published in This Journal.

Difficulties in the Use of the Hydrogen Electrode 639

seemed likely to give satisfaction under the conditions ex-
isting in this case, namely, cane sugar inversion and the poten-
tial of the hydrogen electrode. The inversion method for
determining the concentration of the hydrogen ions, when
the system is at equilibrium, has proven very satisfactory
in some experiments that will be described later in another
article.

The use of the hydrogen electrode seemed most promising.
From the work of Salm,' Friedenthal,^ Salessky,^ Schmidt
and Finger, "^ Robertson^ and many others, it was apparent
that the method may be applied with considerable accuracy
to even very dilute solutions of mineral acids. Particularly
suggestive, however, was the work of Denham,* who meas-
ured the hydrolysis of several inorganic salts and of aniline
hydrochloride by this method. The latter instance led to
the belief that small amounts of organic substances would
not materially affect the measurements.

Realizing the great importance which such a method, if per-
fected, would have not only as regards the problems in hand
but also for the study of many other organic reactions, the
matter was gone into quite thoroughly. While it may as well
be stated at once that no method of general application has
been evolved, some of the observations made may be of gen-
eral interest, and for the benefit of future workers along the
same line it seems advisable to give here a full account of the
methods employed and the difficulties encountered.

EXPERIMENTAL

The measurements were made very much in the same way
as described in the preceding articles. A Leeds and North-
rup "type K" potentiometer was used in part of the work
and in the other a simple potentiometer consisting of 5 strands
of No. 38 manganin wire on a meter slide-wire bridge. The

' Z. Elektrochem., 10, 341, Z. physik. Chem, 67, 471.

^ Z. Elektrochem., 10, 113.

■> Ibid.. 10, 204.

•• J. Phys. Chem., 12, 406.

^Itid., 11,437.

« J. Chem. Soc. 93, 41.

640 Desha and Acree

wire had a resistance of 50 ohms per meter and proved to be
very durable. The two potentiometers were intercompared.
The measurements recorded in this article are probably ac-
curate to not more than a millivolt. The hydrogen was made
in a Kipp generator by the action of pure sulphuric acid on
pure zinc and platinum foil. It was washed well with alkaline
solutions and passed over palladium asbestos at 200°-300°
to remove the oxygen. The platinum and the gauze iridium
electrodes are described in one of the preceding articles.^ Stand-
ard normal and tenth-normal calomel cells were prepared
and used as the subsidiary electrodes.

It was desired to measure, if possible, the hydrogen ion
concentration of reaction mixtures where such concentra-
tion v/ould vary with the time. Calculation of the contact
potential between such a solution and the calomel electrode
was therefore impracticable. Recourse was had to the use
of a saturated solution of ammonium nitrate between the two
electrodes. This procedure is stated by Abegg and Gumming^
to annul the contact potential nearly completely and on this
assumption was so used by Denham.^ That it does not do
so entirely is shown by the following data for a cell arranged
as follows :

H electrode —

- HCl —xN NH.NOj (satd.) -
Hg,Cl3-Hg

-O.I NKCl

dv (calc.)

I . 0038
0. 10038

O.OI
O.OOI

0 . 3400*
0.4018

0.4574
0.5120

0.0618
0.0556
0.0554

0.0552
0.0575
0.0586

The normality of the acid used is given under x; under v
is the voltage read; under dv is the difference in voltage for
two successive concentrations of acid ; dv (calc.) is the theo-

1 This Journal, 46, 602.

2 Z. Elektrochem., 13, 17.

3 Loc. cit.

* These values, when corrected for the potentionieter, become 0.339,0.401,0.456
and 0.511, respectively. Loomis obtained the values 0.3366, 0.3996, 0.4566 and
0.5126 for similar concentrations (This Journal, 46, 616).

Difficulties in the Use of the Hydrogen Electrode 641

retical difference in the potentials of two concentrations of
acid as calculated from the expression

1 ^1

T) = 0.0591 log -^

(the dissociation of the hydrochloric acid being taken into
consideration in computing Cj and c^ on the assumption that
all contact potential has been eliminated by the ammonium
nitrate solution. According to this assumption the values
in the third and fourth columns should be identical; the fact
is that they are not only different, but one series increases
as the other diminishes.^

These facts make it impossible to use this expedient in
absolute measurements but it is quite possible to apply it
(in solutions o. i N or more dilute) in the measurement of
relative differences in potential, which is all that is desired
in work of this kind; for by measuring the potential given by
a number of concentrations of acid and plotting the voltage
recorded against the known hydrogen ion concentrations,
it is possible to obtain a curve which will give accurately the
hydrogen ion concentration corresponding to any other volt-
age read imder the same conditions.

One of the first problems which we considered was the study
of the rapidity with which we may ascertain the true and final
value of the electromotive force of a given system. A very
rapid determination of the true potential is an absolute neces-
sity if we are to develop a method for measuring, at any moment,
the concentration of the hydrogen ions present in reacting
systems. Fig. I shows the form of hydrogen electrode and
cell used. The pure hydrogen enters at D, passes through
the washing solution / into the glass cell E, saturating the
hydrogen electrode / and passing out at G. The calomel

1 We planned an elaborate series of experiments in which we intended to use
potassium chloride, calcium and magnesium acetates, potassium bromide, potassiimi
iodide and other very soluble salts possessing ions moving with nearly the same veloci-
ties, but never finished them. Loomis has since studied this question and found that
a saturated solution of potassium chloride nearly completely annuls the contact
potentials studied (This Journal, 46, 614). We have recently learned that Bjerrum
has already done considerable work in this connection (Z. physik. Chem., 63, 428.
Z. Elektrochem. , 17, 389). We shall use saturated solutions of potassium bromide and of
potassium iodide with the systems Hz — Pt — HX — satd. KX — KX — HgzXj — Hg.
in which X is Br or I.

642 Desha and Acree

electrode dipping into the saturated solution of ammonium
nitrate in M completes the hydrogen-calomel cell. The fol-
lowing experiments show how quickly the true electromotive
force is established when the solutions are put into the cell
and the flow of hydrogen is started. A practically constant
potential was registered Vs^ithin ten or fifteen minutes, though
a slight drift, amounting to a few tenths of a millivolt, could
usually be noticed for one or two hours. As a specimen, the
following uncorrected data from an experiment with o. 10038 N
hydrochloric acid may be given.

Time

Voltage

Time

Voltage

11.05

(Started)

3.08

(Started)

II. 15

0.401 1

3.12

0.4018

II. 31

0.4013

3.28

0.4020

11.56

0.4015

3-59

0.4018

12.30

0 . 4020

423

0.4018

12.58

0.4017

4.40

0.4018

5.20

0.4018

The same solution was used in both instances; the current
of hydrogen had been discontinued between the two experi-
ments. The same solution was further allowed to remain in
the cell in contact with the electrode for two days, no hydrogen
being passed in. At the end of that time the hydrogen was
turned on and after running for ninety minutes the potential
was 0.4019 volt.

Similarly satisfactory results were obtained with 0.01 N
hydrochloric acid. When o.ooi N acid was used the high
resistance of the solution diminished the sensitiveness of read-
ing. This cell was measured with the Leeds and Northrup
"type K" potentiometer.

H electrode — o.ooi N HCl — satd. NH.NO., — o. i N KCl —
Hg3Cl,-Hg

Time. Voltage

1 1 . 00 A. M. (Started)

II. 12 0.5095

11.24 0.5080

11-35 0.5075

I .00 p. M. 0.5100

2.10 o . 5090

Average o . 5088

Difficulties in the Use of the Hydrogen Electrode 643

When solutions contain neutral salts (as would be the case
in nearly all reaction mixtures) the conductivity is increased
and hence more dilute acids can be read with greater accuracy
than here shown.

We then tried some experiments in which we first put the
acid solution into F, closed the stopcocks G and L, opened the
stopcocks H and K and passed the hydrogen through /, E and F
from 30 to 90 minutes in order to saturate the damp platinum
electrode / and the acid solution in F before the acid was al-
lowed to flow from F into E and over /. In this way the true
electromotive force was established much more quickly, as
the following table shows:

fi Volt

6 0.4973

35 0.4975

55 0.4969

74 o 4969 ,

137 0.4964

184 0.4964

Another similar experiment was performed in which an acid
solution filling F and E was electrolyzed overnight with / as
cathode, a 3-ampere current being used in order to saturate /
thoroughly with hydrogen. After changing the solutions
in the same way as in the preceding experiment, the following
results were obtained :

<i Volt

5 0.4555

II 0.4563

25 0.4563

46 0.4561

130 0.4554

These experiments show that a preliminary saturation of
the platinum electrode and the reacting liquid somewhat
shortens the time required to obtain equilibrium between the
electrode and the solution. After this equilibrium is once es-
tablished probably no appreciable error will be involved in

1 / = number of minutes after starting the experiment.

644 Desha and Acree

using the observed electromotive force as a measure of the
concentration of the hydrogen ions, even when the system is
changing. *

On the Hydrolysis of Aniline Hydrochloride

After this work was first suggested by one^ of us, Denham
published a very important contribution in which he described
the use of the hydrogen electrode in the measurement of the
hydrolysis of aniline hydrochloride, his results agreeing very
closely with those obtained by Bredig. Denham's results
were so valuable that we began to repeat them to see if anyone
could readily use the method and to learn the care which must
be exercised in this particular work. Instead of preparing
the hydrochloride in a state of great purity, as Denham did,
we purified the aniline carefully, dissolved it in ether and passed
in hydrogen chloride gas. The white precipitate was filtered
off, washed with ether and dried over concentrated sulphuric
acid and solid potassium hydroxide. An analysis by the
silver chloride method showed the salt to be 99 . 93 per cent,
pure. Some solutions were made by dissolving this salt in
conductivity water. Others were made by treating weighed
quantities of aniline with the required amount of hydrochloric
acid. A large number of experiments with various solu-
tions and different electrodes was made but very discordant
results were obtained. For instance, a N/16 solution of ani-
line hydrochloride gave a value 0.4563 at one time and 0.4490
at another time against a normal calomel electrode, Wils-
more having obtained the value 0.4567. Similarly, one plat-
inum electrode in a N/32 solution of aniline hydrochloride
gave a value of 0.4662 volt, whereas another electrode
gave a value o . 46 1 r , although both electrodes behaved nor-
mally in 0.01 N hydrochloric acid. It was shown very defi-
nitely, by exposing the electrodes to the air or allowing air to
enter the apparatus, that, unless the oxygen is rigidly excluded,
oxidation of the aniline takes place in the presence of the

1 We are now planning a more elaborate and accurate series to test these points
more thoroughly. — S. F. A.

2 Professor Arthur Lapworth has also advocated the use of the hydrogen electrode
in this connection.

Difficulties in the Use of the Hydrogen Electrode 645

strongly catalytic spongy platinum of the hydrogen electrode
and greatly varying electromotive forces were recorded. Sim-
ilar changes in the solutions of hydroxylamine and dimethyl-
aniline were also found. In the solution of aniline hydro-
chloride a pink color was often developed and an oil appeared,
probably from aniline remaining in the aniline hydrocliloride.
The results of all these experiments and of those of Loomis de-
scribed in the preceding articles have shown us that in order
to secure reliable data in the use of this hydrogen electrode
we must work with a perfectly pure sample of aniline hydro-
chloride and must rigidly exclude air from the cell. When
this was done by Loomis sufficiently reproducible and har-
monizing results were obtained. Hoping to minimize any
such trouble, we decided to use dimethylaniline hydrochloride,
in which the liability to oxidation is less. These solutions were
prepared by dissolving a weighed amount of redistilled di-
methylaniline in the corresponding measured volume of stand- ^
ard hydrochloric acid and diluting to the strength desired.
Repeated experiments with 0.02 N and o.oi N solutions
prepared in this way showed that in every case the potential
rose to a maximum value within 15-30 minutes and then be-
gan to fall away slowly. If the electrode was not removed,
the vessel kept closed and the hydrogen flow uninterrupted,
this falling off would usually amount to only i . 5-2 milli-
volts for even as long a period as eighteen hours. But if the
platinum electrode was removed for some time, thus being
brought into contact with the air, the highest value obtain-
able after replacing it was always from 40-120 millivolts less
than that recorded in the first instance.

So far only a single platinum-black electrode (designated
PtJ had been used with the dimethylaniline hydrochloride solu-
tion. The effect of a second electrode of the same kind (Pt,) and
of the iridium one (IrJ was now investigated. It soon became
apparent that each of these three metal electrodes gave a
different value. In a solution of 0.02 N dimethylaniline
hydrochloride, Irj gave a maximum potential of 0.5316 volt
in 27 minutes. It was removed and replaced by Pto in the
same solution; in 26 minutes this gave a maximum value of

646 Desha and Acree

0.5036 volt, falling off to 0.5028 after 28 minutes more. Ir,
was then replaced; 0.5276 volt was recorded in 7 minutes,
falling off to 0.5256 after remaining 46 minutes longer. Pt^
was now introduced and gave 0.5120 volt in 13 minutes,
Iri was introduced for the third time and gave 0.5158 volt in
17 minutes, only decreasing to 0.5148 after 77 minutes more.
Ptj gave 0.4882 and finally Ptj registered 0.4958. These re-
sults are shown in the following table; the first column
gives the order in which the electrodes were introduced ; under
"time" is given the number of minutes between the introduc-
tion of the electrode and the reading of the maximum poten-
tial in each case.

Ir, Pt2 Pt,

0.5316

0.5276

0.5158

26 0.5036

0.4882

13 0.5120

9 0.4958

Between the removal of electrode Ir^ at the close of Experi-
ment I and its introduction in Experiment 3, fifty- two min-
utes had elapsed, during which time the electrode had been kept
in a closed vessel containing some of the same o . 02 N dimethyl-
aniline hydrochloride solution saturated with hydrogen. It
was therefore in contact with the air only while being trans-
ferred from one vessel to the other; the loss in potential here
was only 4.0 millivolts (0.5316-0.5276). Between Ex-
periments 3 and 5 it was exposed to the air continuously for
97 minutes; the loss here was 120 millivolts.

The metal electrodes were naturally suspected of occasioning
these peculiar results. But when a check experiment was
carried out with 0.01 N hydrochloric acid the next day all
these quickly reached the same maximum value, differing not
more than o . 5 millivolt among themselves. On removing an
electrode and leaving it in contact with the air for an hour
the same original value was restored within 8 minutes after
replacing it and remained constant indefinitely.

Difficulties in the Use of the Hydrogen Electrode 647

Before such results had been obtained with the aniline com-
pounds certain preliminary experiments were carried out with
hydroxylamine hydrochloride solutions, preparatory to the
oxime work. After some preliminary experiments with a
N/32 solution of this salt fairly constant potentials were ob-
tained which indicated hydrolysis amounting to 6.5 to 1 1 . 8
per cent., as against 0.74 per cent, found by Winkelblech^
at the same dilution. The solutions were then analyzed for
hydroxylamine by the iodine method. Ten cc. of the original
solution required for oxidation 10.40 cc. of a certain solution
of iodine; 10 cc. of the same hydroxylamine hydrochloride
solution which had been used in an experiment and was in
contact with the platinum electrode for 85 minutes required
only 7.67 cc. of the same iodine solution. Some of the same
stock solution was used in another experiment in which it
remained in contact with the electrode for 260 minutes; 10
cc. of it then required only 4.31 cc. of iodine , indicating a de- ^
composition of 58.6 per cent. Bright platinum electrodes
produced little or no decomposition but gave low and unsteady
potentials. Knowing that tin has little decomposing action
on even free hydroxylamine,^ we attempted to use block-
tin plates on which spongy tin was deposited. The potential
given by such an electrode was practically independent of the
concentration of the acid surrounding it.

Returning to the plated electrodes, we found that by sepa-
rately saturating the electrode and the solution with hydrogen
a very constant value could be obtained within 5 to 10 min-
utes after bringing them together. In this time there could be
little decomposition of the hydroxylamine, and this was still
further retarded when free acid was present. In this way a
solution of 0.025 N hydroxylamine hydrochloride in 0.0125 N
hydrochloric acid gave a potential only o . 5-0 . 7 millivolt
lower than that given by the pure acid of the same concentra-
tion. Similarly, a solution 0.025 N with respect to acetone
and 0.0125 N with respect to hydrochloric acid gave a poten-
tial only 0.3 millivolt lower than the pure acid. But when

1 Z. physik. Chem., 36, 546.

2 Mackay: P. Nova Scotia Inst. Sci., [2] 11, 324.

648 Reviews

both hydroxylamine hydrochloride and acetone were present
the results were quite different. In several cases the maximum
potential thus obtained indicated a hydrogen ion concentra-
tion about double that which was possible if the hydroxyl-
amine and acetoxime hydrochlorides were completely hy-
drolyzed.

CONCI.US10NS

1. Some organic compounds are decomposed in the pres-
ence of the hydrogen electrode, and especially so when oxygen is
present, and the electromotive force observed may not in
every case correspond to the concentration of the hydrogen
ions present.

2. In most experiments the true electromotive force can be
ascertained within 30 minutes after starting the hydrogen
electrode. If the platinum black is saturated beforehand
the electromotive force can be measured within a millivolt
in 5 to ID minutes.

3. Ammonium nitrate does not entirely eliminate contact
potential.

Johns Hopkins University

Baltimore, Md.

June 1, 1909

REDUCTION OF MERCURIC CHLORIDE BY PHOS-
PHOROUS ACID AND THE LAW OF MAvSS ACTION

By James B. Garner

My attention has been called to an error in the interpreta-
tion of the data obtained by a study of the reaction which
occurs between mercuric chloride and phosphorous acid.^ I
have at once recognized the validity of the criticism and will,
therefore, in a subsequent number of This Journai., resubmit
my calculations based upon the experimental data given.

REVIEWS
Radiumnormai^masse und deren Verwendung bei radioaktiven
Messungen. Von E. Rutherford. Deutsch von Dr. B. Finkel-
STEin. Mit 3 Abbildungen im Text. Leipzig: Akadetnische Ver-
lagsgesellschaft m. b. H, pp. 45.
Rutherford, in this brochure, calls attention to the desira-

1 This Journal, 46, 361.

Reviews 649

bility of an international standard, in terms of which the
amount of radium present in any substance can be determined.
This is important from both the scientific and commercial
side. He suggests a form of electroscope which could be used
to detect the ;--rays that are shot off from radium. At the
radiological congress in Brussels in 19 10 the desirability of
an international unit was discussed, and a committee was ap-
pointed for this purpose. The committee voted unanimously
that the one to work out such a unit was Mme. Curie, the dis-
coverer of radium, and she has undertaken the work.

H. c. J.

An Experimentai. Course of Physical Chemistry. Part I. Stat-
ical Experiments. By James Frederick Spencer, D.Sc. (Liver-
pool), Ph.D. (Breslau), Assistant Lecturer in Chemistry, Bedford
College (University of London). London : G. Bell & Sons, Ltd.
1911. pp. xiv + 228. Price, 3s. 6d.

This little volume is a description of physical chemical
apparatus together with a list of experiments to be carried
out. The apparatus described is not always the newest or
best form. Indeed, very far from it. The Beckmann boil-,
ing-point apparatus, as sketched on page 102, is to-day of
hardly more than historical interest, and it is very difficult
to believe that the apparatus for measuring osmotic pressure
described on page 94 could give even approximate results.

Again, the surface-tension method of determining the
association factor of a liquid is rather difficult as a general
laboratory method.

Such minor defects as those referred to above can easily
be remedied in subsequent editions. The book, as a whole,
is a convenient laboratory guide, and will doubtless prove to
be a useful and valuable contribution to the literature of the
field which it aims to cover. h. c. j.

Hydrocarbures, A1.C001.S ET Ethbrs de la S6rie Grasse. Par P.
Carr6, Docteur es sciences, Professeur lil'Ecoles de Hautes Etvides
commerciales, Pr^parateur k I'Institut de Chimie appliqu^e. Paris :
Octave Doin & Fils. pp. xii -f 410. Price, Fr. 10.

This volume is one of a series covering chemistry and form-
ing a part of an extensive "encyclopedic scientifique."

The book is divided into six parts: hydrocarbons, their
halogen derivatives, alcohols, ethers, ethereal salts of mineral
acids, and derivatives of sulphur, selenium and tellurium.

The general properties and methods of preparation of each
class or series of compounds are well set forth and special in-
formation is added in the case of the more important individual

650 Reviews

compounds. A moderate number of references are given to
the literature. The treatment is very systematic and quite
thorough, considering the scope of the work. The book is
inexpensively but neatly gotten up. E. E. R.

Organic Chbmistry for thb IvAboratory. By W. A. Noyes, Ph.D.,
Professor of Chemistry in the University of Illinois, Urbana, 111.
Second Edition, Revised and Enlarged. Easton, Pa.: The Chemical
Publishing Co. ^gn. pp. xi + 291. Price, |2.

The first edition of this useful book has been enlarged by
the addition of chapters on Analysis of Organic Compounds,
on General Operations, on Ethers, on Hydroxy and Ketonic
Acids, and on Carbohydrates. Some thirty preparations
have been added to the already large number in the first edi-
tion. The material has been rearranged to bring it more
nearly in accord with the author's well-known text-book,
though it may still be used with any other text-book. New
tables are given for nitrogen, the latest atomic weight deter-
minations being used. A chapter is given at the end on the
examination and identification of organic compounds.

Each chapter is opened with a good general discussion of
the properties and various methods of preparation of the class
of compounds under consideration, and the student is further
aided toward an understanding of the principles involved by
well selected references to the literature, inserted at the be-
ginning of the directions for the preparation of each com-
pound.

The preparations given cover a wide field and involve a
great variety of operations. In this way assistance is given
the advanced worker who may find here methods adapted to
almost any use. The directions are clear and sufficient.
The book is well gotten up, though some of the drawings
leave something to be desired. E. E. R.

INDEX TO VOL. XLVI

AUTHORS
AGREE, S. F. See Desha, L. J. and Loomis, N. E.

BIGELOW, H. E. See Jackson, C. L.

Bingham, E. C. and Durham, T. C. The viscosity and fluidity of

suspensions of finely-divided

solids in liquids 278

" and White, G. F. A laboratory manual of inor-
ganic chemistry (Review) 214

Biron, E. V. See Jones, H. C.

CARRE, P. Hydrocarbures, alcools et ethers de la s^rie grasse

(Review) 649

Chenu, G. See Post, J.

Comanditcci, E. and Roth, W. Die Constitution der Chinaalka-

loide (Review) 535

Corvisy, A. See Nernst, W.

DAVIS, W. A. and Sadtler, S. S. Allen's commercial organic anal-
ysis, IV (Review) 308

Delbridge, T. G. See Orndorff, W. R.

Desha, L. J. and Acree, S. F. On difficulties in the use of the hy-
drogen electrode in the measure-
ment of the concentration of hy-
drogen ions in the presence of or-
ganic compounds 638

Dinwiddie, J. G. and Kastle, J. H. The bromination of phenol 502

Durham, T. C. See Bingham, E. C.

FAY, I. W. The chemistry of the coal-tar dyes (Review) . . 534

Finkelstein, B. See Rutherford, E.

Foglesong, J. E. See Garner, J. B.

Fowler, G. J. An introduction to bacteriological and enzyme

chemistry (Review) 415

Freundlich, H. Kapillarchemie (Review) 533

GARNER, J. B., Foglesong, J. E. and Wilson, R. Reduction of

mercuric
chloride by
phosphorous
acid and the
law of mass
action 361, 648

652 Index

" Saxton, B. and Parker, H. O. Anhydrous

formic acid.. . . 236

Getman, F. H. DiJBferences of potential between cadmium and

alcoholic solutions of some of its salts 117

Gill, A. H. A short hand-book of oil analysis (Review) 216

Guy, J. S. and Jones, H. C. Conductivity and viscosity in mixed

solvents containing glycerol 131

HADEN, R. L. See Kastle, J. H.

Harden, A. Alcoholic fermentation (Review) 414

Hart, E. Chemistry for beginners. I. Inorganic (Review) 215

Hedley, E. P. See Werner, A.

Heritage, G. L. See Kohler, E. P.

Hill, A. J. See Johnson, T. B.

Holleman, A. F. Die direkte Einfiihrung von Substituenten in

den Benzolkern (Review) 309

Hosford, H. H. and Jones, H. C. The conductivities, tempera-
ture coefficients of conductivity
and dissociation of certain elec-
trolytes 240

JACKSON, C. L and Bigelow, H. E. 2-Brom-i,3-5-triiod-4,6-dini-

trobenzene and some of its

derivatives 549

Johnson, T. B. and Hill, A. J. Researches on pyrimidines: the
condensation of urea and' guani-
dine with esters of allylmalonic
and some alkyl-substituted allyl-
malonic acids 537

" and Shepard, N . A. Researches on pyrimidines: the

condensation of ethyl formate
and diethyl oxalate with some

pyrimidinethiogly collates 345

See Wheeler, H. L.
Jones. H. C. The electrical nature of matter and radioactivity

(Review) 312

" Biron, E. V., Zhukoff, I. I. and Sopozhnikoff, A. V.

Osnovi physicheskoi Chemie (Review) 414

See Guy, J. S., Hosford, H. H., Kreider, H. R.,
Wightman, E. P. and Winston, L. G.
jUptner, H. von. Das chemische Gleichgewicht auf Grund me-

chanischer Vorstellungen (Review) 114

Index 653

KASTLE, J. H. and Haden, R. L. A study of o-ainino-/>-suIpho-
benzoic acid and its deriva-
tives, with special reference to

their fluorescence 508

" " On the color changes occurring

in the blue flowers of the wild
chicory, dehor ium intybus . . . 315

" See Dinwiddie, J. G.

Keiser, E. H. and Kessler, J. J. The nitrile of fumaric acid 523

" and McMaster, L. The synthesis of fumaric and

maleic acids from the acetylene

diiodides 518

Kessler, J. J. See Keiser, E. H.

Kohler, E. P. Unsaturated ^-ketonic acids 474

" Heritage, G. L. and MacLeod, A. L. The reaction

between unsat-
urated c o m -
pounds and or-
ganic zinc com-
pounds 217

Kreider, H. R. and Jones, H. C. The conductivity of certain salts
in methyl and ethyl alcohols at
high dilutions 574

LEATHES, J. B. The fats (Review) 415

Leiser, R. Elektrische Doppelbrechung der KohlenstoiTverbind-

ungen (Review) 311

Loomis, N. E. and Acree, S. F. A study of the hydrogen elec-
trode, of the calomel electrode

and of contact potential 585

" " The application of the hydrogen
electrode to the measurement of
the hydrolysis of aniline hydro-
chloride and the ionization of
acetic acid in the presence of
neutral salts 621

MACLEOD, A. L. See Kohler, E. P.
McMaster, L. See Keiser, E. H.

Meyer, R. Jahrbuch der Chemie, 1909 (Review) 113

Mills, J. An introduction to thermodynamics for engineering

students (Review) 212

Molinari, E. Trattato di chimica inorganica (Review) 212

Montgomery, J. P. The relation of heat of vaporization to other
constants at the boiling temperature of some
liquids at atmospheric pressure 298

654 Index

NASKE, C. Zerkleinerungsvorrichtungcn und Mahlanlagen (Re-
view) 308

Nernst, W. and Corvisy, A. Traite de chimie generale, I (Re-
view) 313

Neumann, B. See Post, J.

Nicolet, B. H. See Wheeler, H. L.

Noyes, W. A. Organic chemistry for the laboratory (Review).. . . 650

ORNDORFF, W. R. and Delbridge, T. G. Tetrachlorgallein and

some of its deriva-
tives I

Ostwald, W. Ueber Katalyse (Review) 413

PARKER, H. O. See Garner, J. B.

Pellet, M. See Post, J.

Peterson, P. P. Stereoisomeric chlorimido ketones 325

Post, J., Neumann, B., Chenu, G. and Pellet, M. Traite complet

d'analyse chim-
ique. I, 4 (Re-
view) 529

RAIFORD, C. L. On chlorimidoquinones 417

Reynolds, G. P. The reaction between organic magnesium com-
pounds and cinnamylidene esters 198

Roth, W. See Comanducci, E.

Rutherford, E. and Finkelstein, B. Radiumnormalmasse (Re-
view) 648

SADTLER, S. S. See Davis, W. A.
Saxton B. See Garner, J. B.

Scheithauer, W. Die Schwelteere (Review) 416

Scott, W. W. Qualitative chemical analysis (Review) 314

Shepard, N. A. See Johnson, T. B.
Sopozhnikoff, A. V. See Jones, H. C.

Spencer, J. F. An experimental course of physical chemistry (Re-
view) 649

Stoddard, J. T. Introduction to general chemistry (Review) 213

TOWER, O. F. A course in qualitative chemical analysis of in-
organic substances (Review) 215

VILLAVECCHIA, V. Dizionario di merceologia, A-M (Review). 216

Vogel, J. H. Das Acetylen (Review) 115

WERNER, A. and Hedley, E. P. New ideas on inorganic chem-
istry (Review) 530

Index

655

Wheeler, H. L., Nicolet, B. H. and Johnson, T. B. On hydantoins:

The action of
acylthi o n c a r-
bamates, acyl-
dithiocarbam-
ates and acyli-
midodithiocar-
bonates on a-
amino acids ;
2-Thiohydan-
toin

White, G. F. See Bingham, E. C.
Wightman, E. P. and Jones, H. C.

Wilson, R. See Garner, J. B.
Winston, L. G. and Jones, H. C.

A study of the conductivity
and dissociation of organic
acids in aqueous solution be-
tween 0° and 35°

The conductivity, temperature
coefficients of conductivity and
dissociation of certain electro-
lytes in aqueous solution from
0° to 35°. Probable inductive
action in solution, and evidence
for the complexity of the ion. .

456

368

ZHUKOFF, I. I. See Jones, H. C.

SUBJECTS
ACETIC acid, ionization in the presence of neutral salts. Loomis

and Acree 621

Acetylthiohydantoic acid, 472; ethyl ester, 473; potassium salt,

472. Wheeler, Nicolet and Johnson 472

Acylthioncarbamates, acyldithiocarbamates and acylimidodithio-
carbonates, action on a-amino acids. Wheeler, Nicolet and

Johnson 456

Allylbenzyliminomalonuric acid. Johnson and Hill 546

AUylbenzylmalonic acid. Johnson and Hill 548

AUylbenzylmalonylguanidine, 547; basic hydrochloride, 547.

Johnson and Hill 547

5,5-Allylbenzylmalonylurea. Johnson and Hill 544

Allylmalonic and some alkyl-substituted allylmalonic acids, con-
densation of urea and guanidine with esters of. Johnson and

Hill 537

5-Allylmalonylguanidine. Johnson and Hill 541

656 Index

5-Allylraalonylurea. Johnson and Hill 540

Aluminium chloride, nitrate and sulphate, conductivity and disso-
ciation. Winston and Jones 393

a-Amino acids, action of acylthioncarbamates, acyldithiocarba-
mates and acylimidodithiocarbonates on. Wheeler, Nicolet
and Johnson 456

i-Amino-3-hydroxybenzene hydrochloride. Jackson and Bigelow. 568

o-Amino-/)-sulphobenzoic acid and its derivatives, study of, with

special reference to their fluorescence. Kastle and Haden. . . . 508

Ammonium aluminium, copper and chromium sulphates, conduc-
tivity and dissociation. Hosford and Jones 245, 256

Ammonium bromide, chloride and nitrate, conductivity and vis-
cosity in mixed solvents containing glycerol. Guy and Jones 142

Ammonium nitrate, sulphate and acid sulphate, conductivity and

dissociation. Winston and Jones 378

Aniline hydrochloride, hydrolysis of. Loomis and Acree 621

BARIUM bromide, chloride and nitrate, conductivity and viscos-
ity in mixed solvents containing glycerol. Gtiy and Jones .... 144

4-Benzalthiohydantoin. Wheeler, Nicolet and Johnson 470

Benzenesulphonic acid, conductivity and dissociation. Wightman

and Jones 96

Benzilic acid, conductivity and djssociation, 74; sodium salt, 67.

Wightman and Jones 74

i-Benzoyl-4-benzalthiohydantoin. Wheeler, Nicolet Bind. Johnson. 469

Benzoylhydantoic acid. Wheeler, Nicolet and Johnson 467

Benzoylpseudoethylhydantoic acid, 466 ; ethyl ester, 466. Wheeler,

Nicolet and Johnson 466

Benzoylpseudoethylthiohydantoic acid, 471; ethyl ester, 471.

Wheeler, Nicolet and Johnson 471

Benzoylthiohydantoic acid, 468; ethyl ester, 468. Wheeler, Nicolet

and Johnson 468

/?-Benzyl-pbenzalbutyric acid, 208 ; methyl ester, 209. Reynolds. . 208

Borax, conductivity and dissociation. Winston and Jones 380

2-Brom-4-aminophenol, 420; hydrochloride, 419. Raiford 420

a-Brombutyric acid, conductivity and dissociation, 79; sodium

salt, 67. Wightman and Jones 79

2-Brom-4-chlorimidoquinone. Raiford 420

2-Brom-i,3-diiod-4,6-dinitrobenzene. Jackson s-nA Bigelow 562

7'-Brom-a,a-dimethyl-/?-phenyl-7--benzoylbutyric acid, 231; methyl

ester, 232; ethyl ester, 232. Kohlcr, Heritage and Macleod. . . 231

^-Brom-|i9-phenyl-;'-benzoylbutyric acid. Kohler 494

j--Brom-^-phenyl-;--benzoylbutyric acids, stereomeric, 499; methyl

esters, 499. Kohler 499

a-Brompropionic acid, conductivity and dissociation, 75; sodium

salt, 66. Wightni-an and Jones 75

Index

657

2-Brom-i,3,5-triiodbenzene. Jackson and Bigelow 557

2-Brom-i,3,5-triiod-4,6-diaminobenzene, 571; hydrochloride, 571.

Jackson and Bigelow 57 1

2-Brom-i,3,5-triiod-4,6-dinitrobenzene and some of its derivatives.

Jackson and Bigelow 549

2-Brom-i,3,5-triiod-6-nitroanisole. Jackson and Bigelow 566

2-Brom-i,3,5-triiod-6-nitrophenetole. Jackson and Bigelow 566

CADMIUM and alcoholic solutions of some of its salts, differences

of potential between. Getman 117

Cadmium bromide, chloride and iodide, conductivity and dissocia-
tion. Winston and Jones 389

Calcium bromide, conductivity and viscosity in mixed solvents

containing glycerol. Guy and Jones 145

Calcium chloride, conductivity and dissociation. Hosford and

Jones 260

Calcium chr ornate and formate, conductivity and dissociation.

Hosford and Jones 252

Calomel electrode, study of. Loomis and Acree 585

Camphoric acid, conductivity and dissociation of. Wightman and

Jones I02*

Caprylic acid, conductivity and dissociation, 83; sodium salt, 67.

Wightman and Jones 83

Chicory, wild, color changes occurring in the blue flowers of. Kas-

tle and Haden 315

6-Chlor-4-acetylamino-5-methylphenyl acetate. Raiford 449

2-Chlor-6-amino-j«-cresol, 448; hydrochloride, 448. Raiford 448

o-Chlorbenzoic acid, conductivity and dissociation, 92 ; sodium salt,

67. Wightman and Jones 92

2-Chlor-4-benzoylamino-5-methylphenyl benzoate. Raiford 444

2-Chlor-6-brom-4-aminophenol, 422; hydrochloride, 422. Raiford 422

2-Chlor-6-brom-4-imidoquinone. Raiford 422

Chlorimido ketones, stereoisomeric. Peterson 325

Chlorimidobenzophenone. Peterson 329

cis- and <roMi'-Chlorimido-/'-chlorbenzophenones. Peterson 333

cis- and imn5--Chlorimido-/'-chlor-/j-methoxybenzophenones.

Peterson 342

cis' and <rani^-Chlorimido-/>-methoxybenzophenones. Peterson. . . . 337

Chlorimidoquinimes. Raiford 417

2-Chlor-6-chlorimidotoluquinone. Raiford 447. 45 1

4-Chlor-6-chlorimidotoluquinone. Raiford 444

3-Chlorimido-4-toluquinone. Raiford 446

o-Chlor-/>-methoxybenzophenone. Peterson 344

/'-Chlor-/'-methoxybenzophenone. Peterson 339

2-Chlor-6-nitro-w-cresol. Raiford 447

2-Chlortoluhydroquinone. Raiford 450

658 Index

2-Chlortoluquinone. Raiford 449

Chromium chloride and sulphate, conductivity and dissociation.

Winston and Jones 395

Chromium sulphate, conductivity and dissociation. Hosford and

Jones 262

Cichorium intybus, color changes occurring in the blue flowers of

Kastle and Haden 315

Cinnamylidene esters, reaction with organic magnesium com-
pounds. Reynolds 198

Cobalt bromide, conductivity and dissociation. Winston and

Jones 399

Cobalt bromide, conductivity in methyl and ethyl alcohols at high

dilutions. Kreider and Jones 577

Cobalt bromide and chloride, conductivity and viscosity in mixed

solvents containing glycerol. Guy and Jones 147

Cobalt sulphate, conductivity and dissociation. Hosford and Jones 264
Color changes occurring in the blue flowers of the wild chicory.

Kastle and Haden 315

Complexity of the ion, evidence for. Winston and Jo7ies 368

Conductivity and dissociation of certain electrolytes. Hosford

and Jones 240

Conductivity and dissociation of certain electrolytes in aqueous

solution from 0° to 35°. Winston and Jones 368

Conductivity and viscosity in mixed solvents containing glycerol.

Guy and Jones 131

Conductivity of certain salts in methyl and ethyl alcohols at high

dilutions. Kreider and Jones 574

Contact potential, study of. Loomis and Acree 585

Copper sulphate, conductivity and dissociation. Hosford and

Jones 264

Winston and Jones 400

Cyanacetic acid, conductivity and dissociation, 72 ; sodium salt, 66.

Wightman and Jones 72

Cyanuric acid, conductivity and dissociation. Wightman and

Jones 103

5,5-DIALLYLMAI.ONYLGUANIDINE. Johnson and Hill 543

5,5-Diallylmalonylurea. Johnson and Hill 542

a,a-Dibenzyl-(J-benzalcrotonyl alcohol. Reynolds 207

2,4-Dibrom-6-acetamino-3-methylphenyl acetate. Raiford 434

2,4-Dibrom-6-amino-OT-cresol, 432; hydrochloride, 431. Raiford. . 432

2,6-Dibrom-4-amino-m-cresol, 428; hydrochloride, 428. Raiford. . 428

2,6-Dibrom-4-benzoylamino-w-cresol. Raiford 429

2,4-Dibrom-6-benzoylamino-3-methylphenyl benzoate. Raiford. 432

Dibrom-,5-benzyl-^-benzalpropylbenzyl ketone. Reynolds 207

Index

659

2,6-Dibrom-4-chlorimidotoluquinone. Raiford 430

3,5Dibrom-2-hydroxy-4-methylphenylurethane. Raiford 433

3,5-Dibrom-4-hydroxy-6-methylphenylurethane. Raiford 429

2,6-Dibrom-4-nitro-m-cresol. Raiford 427

2,4-Dibrom-6-nitro-w-cresol. Raiford 427

2,4-Dibrom-6-nitro-3-methylphenylethyl carbonate. Raiford 435

^,;--Dibrom-/?-pheayl-;--benzoylbutyric acids, stereomeric. Kohler. 490
2,(?)-Dibrom-i,3,5-triiod-(?)-nitrobenzene. Jackson and Bigelow. 561
4,5-Diclilor-o-phthalic acid, conductivity and dissociation. Wight-
man and Jones 99

2,6-Dichlortoluhydroquinone. Raiford 426

2,6-Dichlorotoluquinone. Raiford 425

a,a-Diethyl-^-benzalcrotonyI alcohol. Reynolds 210

Diethyl 2-brom-3-iod-4,6-dinitrophenylmalonate. Jackson and

Bigelow 565

Diethyl /9-brom-a-phenyl-/?-benzoylethylmalonates, stereomeric

forms. Kohler 482

Diethyl o-ethylamino-Zj-sulphobenzoate. Kastle and Hadcn 510

Diethyl oxalate, condensation with pyruxiidinethioglycollates.

Johnson and Shepard 345

Diethyl 6-oxy-4-methylpyrimidine-2-oxalthioglycollate. Johnson ,

and Shepard 359

Diethyl 6-oxyprimidine-2-oxalthioglycollate. Johnson and Shep-
ard 352

Diethyl a-phenyl-/?-benzoylethylmalonate. Kohler 482

2,4-Dihydroxybenzoic acid, conductivity and dissociation, 93;

sodium salt, 68. Wightman and Jones 93

2,5-Dihydroxybenzoic acid, conductivity and dissociation, 94;

sodium salt, 68. Wightman and Jones 94

Dimethyl brom-,5-brom-a-phenyl-/?-benzoylethylmalonates, stereo-
meric. Kohler 484

Dimethyl brommalonate. Kohler, Heritage and Macleod 234

a,a-Dimethyl-/9-phenyl-;--benzoylbutyric acid, 230; methyl ester,

231 ; ethyl ester, 230. Kohler, Heritage and Macleod 230

a,a-Dimethyl - [i - phenyl - ;- - benzoylbutyrolactones, stereomeric.

Kohler, Heritage and Macleod .232

Dimethyl a-phenyl-/?-benzoylethylmalonate. Kohler, Heritage and

Macleod 234

2,4-Dinitrobenzoic acid, conductivity and dissociation, 90; sodium

salt, 67. Wightman and Jones 90

3,5-Dinitrobenzoic acid, conductivity and dissociation, 91 ; sodium

salt, 68. Wightman and Jones 91

Dipotassium phosphate, conductivity and dissociation. Wifiston

and Jones 383

Disodium phosphate, conductivity and dissociation. Hosford and

Jones 257

66o Index

RRATA 676

o-Ethylamino-/>-sulphobenzoic acid, fluorescence of. Kastle and

Hadin 516

Ethyl benzoylpseudomethylhydantoate. Wheeler, Nicolet and

Johnson 466

Ethyl formate, condensation with pyrimidinethioglycollates.

Johnson and Shepard 345

Ethyl /?-hydroxy-a-methyl-/3-phenyl-j'-benzalbutyrates, stereo-

meric. Kohler, Heritage and Macleod 225

Ethyl 6-oxy-4-methylpyrimidine- 2 - [a-thio-/?-hydroxyacrylate ].

Johnson and Shepard 357

a-Ethyl-/?-phenyl-;--benzoylbutyric acid, 228; methyl ester, 228.

Kohler, Heritage and Macleod 228

Ethyl 6-oxypyrimidine-2-thioglycollate. Johnson and Shepard. . . 350

Ethyl 6-oxypyrimidine-2-[a-thio-/?-hydroxyacrylatc]. Johnson

and Shepard 35 1

FERRIC chloride, conductivity and dissociation. Hosford and

Jones 262

Fluorescence of o-amino-ZJ-sulphobenzoic acid and its derivatives.

Kastle and Haden . 508

Formic acid, anhydrous. Garner, Saxton and Parker 236

Fumaric acid nitrile. Keiser and Kessler 523

Fumaric and maleic acids, synthesis from the acetylene diiodides.

Keiser and McMaster 518

GLYCEROL, conductivity and viscosity in mixed solvents con-
taining. Guy and Jones 131

Guanidine, condensation with allylmalonic and alkyl-substituted

allylmalonic esters. Johnson and Hill 537

HYDANTOINS: Action of acylthioncarbamates, acyldithiocar-
bamates and acylimidodithiocarbonates on a-amino acids;
2-Thiohydantoin. Wheeler, Nicolet and Johnson 456

Hydrogen electrode, application in the measurement of the hydroly-
sis of aniline hydrochloride, and the ionization of acetic acid
in the presence of neutral salts. Loomis and Acree 621

Hydrogen electrode, difficulties in the use of, in the measurement
of hydrogen ion concentrations in the presence of organic
compounds. Desha and Acree 638

Hydrogen electrode, study of. Loomis and Acree 585

7--Hydroxy-a,a-dimethyl-/?-phen)-l-;--ben.zoylbutyric acid. Koh-
ler, Heritage and Macleod 234

^-Hydroxy-/?,5-diphenyl-a,7--butadiene-a-carboxylic acid lactone.

Kohler , . . 497

Hydroxydiphenylcrotolactonic acid . Kohler 488

Index 66 1

5-Hydroxy-/?,5-diphenyl-^-crotolactone. Kohler 498

^-Hydroxy-/9,5-diphenyl-/?-heptalactones, stereomeric. Kohler. . . . 495
Hydroxyisobutyric acid, conductivity and dissociation, 80; sodium

salt, 67. Wightman and Jozies 80

e-Hydrox5^-/?-methyl-7-,£-diphenyl-(?-pentene-/?-carboxylic lactone.

Kohler, Heritage and Macleod 229

/5-Hydroxy-/?-phen)'^l-;'-benzalbutyric acid. Kohler, Heritage and

Macleod . 223

7--Hydroxy-/3-phenyl-7--benzoylbutyric acid. Kohler 501

?--Hydroxy-/?-phenyl-;'-benzoyl-;--heptalactone. Kohler 502

7'-Hydroxy-/?-phenyl-^-benzoyl - a - propene - a - carboxylic lactone.

Kohler 49 1

7--Hydroxytriphenylbutyric lactone. Reynolds 204

IMIDO-/>-BENZOPHENONE hydrochloride. Peterson 332

Imido-^-chlor-Zj-methoxybenzophenone hydrochloride. Peterson. 340

Imido-/>-inethoxybenzophenone hydrochloride. Peterson 335

Inductive action in solution, probable. Winston and Jones 368

International Association of Chemical Societies (Note) 116

/?-Iodpropionic acid, conductivity and dissociation, 76; sodium

salt, 66. Wightman and Jones 76

Isovaleric acid, conductivity and dissociation, 82; sodium salt, 67.

Wightman and Jones 82

^-KETONIC acids, unsaturated. Kohler 474

LEAD acetate, conductivity and dissociation. Hosford and Jones. 255

Lead chloride, conductivity and dissociation. Winston and Jones . 392
Levulinic acid, conductivity and dissociation, 77; sodium salt, 66.

Wightman and Jones 77

Lithium bromide, conductivity in methyl and ethyl alcohols at

high dilutions. Kreider and Jones 576

MALEIC and f umaric acids, synthesis from the acetylene diiodides.

Keiser and McMaster 518

Magnesium acetate, formate, bromide and nitrate, conductivity

and dissociation. Winston and Jones 386

Magnesium chloride, conductivity and dissociation. Hosford and

Jones 260

Manganese sulphate, conductivity and dissociation. Hosford and

Jones 261

Winston and Jones 398

Meconicacid, conductivity and dissociation. Wightman and Jones. loi
Mercuric chloride, reduction by phosphorous acid and the law of

mass action. Garner, Foglesong and Wilson 361, 648

662 Index

4-Methyl-i-acetylthiohydantoic acid. Wheeler, Nicolet and John-
son 473

Methyl /?-chlor-/?-phenyl-7--benzoylbutyrate. Kohler 490

Methyl cinnamylidenacetates, reaction with ethyl-, phenyl- and

benzylmagnesium bromides. Reynolds 198

a-Methyl-^-phenyl-;--benzoylbutyric acids, stereomeric; methyl

ester, 226; ethyl ester, 227. Kohler, Heritage and MacLeod. . . 225

NEUTRAL salts, ionization of acetic acid in the presence of.

Loomis and Acree 62 1

Nickel nitrate and sulphate, conductivity and dissociation. Hos-

ford and Jones 263

w-Nitrobenzenesulphonic acid, conductivity and dissociation.

Wightman and Jones 98

/>-Nitrobenzoic acid, conductivity and dissociation, 89; sodium

salt, 67. Wightman and Jones 89

1,2,4-Nitrotoluenesulphonic acid, conductivity and dissociation.

Wightman and Jones 99

OBITUARY:

Ladenburg, Albert 528

Organic acids, conductivity and dissociation in aqueous solution

between 0° and 35°. Wightman and Jones 56

6-Oxy-4-methylpyrimidine-2-thioglycollic acid, 356; ethyl ester,

355. Johnson and Shepard 35^

6-Oxyprimidine-2-thiopyruvic acid. Johnson and Shepard 352

PHENOL, bromination of. Dinwiddie and Kastle 502

Phenylbenzoylbutyrolactonic acid. Kohler 487

/?-Phenyl-;'-benzoyl-/?-butyrolactone. Kohler 493

/?-Phenyl-7--benzoyl-;--butyrolactone. Kohler 500

a-Phenyl-/?-benzoylvinylacetic acid. Kohler 489

a-Phenyl-5-benzoylvinylmalonic acid, potassium salts, 486; di-
methyl ester, 484. Kohler 484

/?-Phenylcinnamylidenace tic acid. Kohler, Heritage and Maclcod . 22^
Potassium acetate and permanganate, conductivity and dissocia-
tion. Winston and Jones 382

Potassium acetate, phosphate and sulphocyanate, conductivity

and dissociation. Hosford and Jones 258

Potassium aluminium, chromium, nickel and sodium sulphates,

conductivity and dissociation. Hosford and Jones 248, 258

Potassium bromide, chloride and nitrate, conductivity and vis-
cosity in mixed solvents containing glycerol. Guy and Jones . 139
Potassium sulphocyanate, conductivity in methyl alcohol at high

dilutions. Kreider and Jones 577

Index 663

Potential, differences of, between cadmium and alcoholic solutions

of some of its salts. Getman 117

Pyrimidines, researches on: Condensation of urea and guanidine
with esters of allylmalonic and alkyl-substituted allylmalonic

acids. Johnson and Hill 537

Pyrimidinethioglycollates, condensation with ethyl formate and

diethyl oxalate. Johnson and Shepard 345

REVIEWS:

Acetylen, das. Vogel 115

Alcoholic fermentation. Harden 414

Allen's commercial organic analysis, IV. Davis and Sadtler . . 308
Analyse chimique, traits complet d'. I, 4. Post, Neumann,

Chenu and Pellet 529

Bacteriological and enzyme chemistry, an introduction to.

Fowler 415

Chemistry for beginners. I. Inorganic. Hart 215

Chimica inorganica, trattato di. Molinari 212

Chimie generale, traits de, I. Nernst and Corvisy 313

Chinaalkaloide, die Constitution der. Comanducci and Roth. . 535

Coal-tar dyes, the chemistry of the. Fay 534

Direkte Einfiihrung von Substituenten in den Benzolkern, die.

Holleman 309

Electrical nature of matter and radioactivity. Jones 312

Elektrische Doppelbrechung der Kohlenstoffverbindungen.

Leiser 311

Fats, the. Leathes 415

General chemistry, introduction to. Stoddard 213

Gleichgewicht auf Grund mechanischer Vorstellungen, das

chemische. von Jilptner 114

Hydrocarbures, alcools et ethers de la serie grasse. Carr^ . . 649
Inorganic chemistry, a laboratory manual of. Bingham and

White 214

Inorganic chemistry, new ideas on. Werner and Hedley 530

Jahrbuch der Chemie, 1909. Meyer 113

Kapillarchemie. Freiindlich 533

Katalyse, ueber. Ostwald 4^3

Merccologia, dizionario di. A-M. Villavecchia 216

Oil analysis, a short hand-book of. Gill 216

Organic chemistry for the laboratory. Noyes 650

Physical chemistry, an experimental course of. Spencer .... 649
Physicheskoi Chemie, Osnovi. Jones, Biron, Zhukoff and

Sopozhmkoff 414

Qualitative chemical analysis. Scott 314

Qualitative chemical analysis of inorganic substances, a course

in. Tower 215

664 Index

Radiumnormalmasse. Rutherford and Finkelsiein 649

Schwelteere, die. Scheithauer 416

Thermodynamics for engineering students, an introduction to.

Mills 212

Zerkleinerungsvorrichtungen und Mahlanlagen. Naske 308

SILVER nitrate, conductivity and dissociation. Winston and

Jones 399

Sodium bromide, conductivity in methyl and ethyl alcohols at high

dilutions. Kreider and Jones 576

Sodium ferrocyanide, conductivity and dissociation of. Hosjord

and Jones 248

Sodium tetraborate, conductivity and dissociation. Hosford and

Jones 257

Sodium sulphate, conductivity and dissociation. Winston and

Jones 380

Strontium acetate, conductivity and dissociation. Winston and

Jon£s 385

Sodium bromide, chloride, iodide and nitrate, conductivity and

viscosity in mixed solvents containing glycerol. Guy and

Jones 140

Strontium bromide and nitrate, conductivity and viscosity in

mixed solvents containing glycerol. Guy and Jones 146

/j-Sulphamidobenzoic acid, conductivity and dissociation, 95;

sodium salt, 68. Wightman and Jones 95

Suspensions of finely-divided solids in liquids, viscosity and fluidity.

Bingham and Durham 278

/-TARTARIC ACID, conductivity and dissociation. Wightman

and Jones 84

Tetrabrom-a,a-dibenzyl-5-benzalcrotonyl alcohol. Reynolds 207

Tetrachlorgallein and some of its derivatives. Orndorff and Del-
bridge I

Tetrachlorgallein, ammonium salt, 35; tetrammonium salt, 32, 48;
potassium salt, 49; diacetonate, 3; etherate, 14; hydrochloride,
19; colored hydrate, 21 ; colored hydrate hydrochloride, 30, 46;
colorless hydrate, 26; tetracetate, 51. Orndorff and Delhridge. 1
Tetrachlorgalleincarbinolcarboxylic acid. Orndorff and Delhridge 37
Tetrachlor-o-phthalic acid, conductivity and dissociation. Wight-
man and Jones 100

Tetraethyl acetylenetetracarboxylate. Jackson and Bigelow 563

i,i,3,5-Tetraphenyl-4-pentene-i-ol. Reynolds 203

2-Thio-i-acetyl-4-benzalhydantoin. Wheeler, Nicolet and John-
son 472

Thiodiglycolic acid, conductivity and dissociation. Wightman

and Jones 86

Index 665

2-Thiohydantoin. Wheeler, Nicolet and Johnson 456

2-Thio-4-methylhydantoin. Wheeler, Nicolet and Johnson 474

2 - [2 - Thio - 6 - oxypyrimidine - 5 - mercapto] - 6 - oxy - 4 - methyl-

pyrimidine. Johnson and Shepard 359

2 - [2 - Thio - 6 - oxypyrimidine - 5 - mercapto] - 6 - oxypyrimidine.

Johnson and Shepard 354

t/-a»j'-Thioureaacrylic acid. Johnson and Shepard 351

/>-Toluenesulphonic acid, conductivity and dissociation. Wight-
man and Jones 97

Tricarballylic acid, conductivity and dissociation. Wightman and

Jones 87

Trichloracetic acid, conductivity and dissociation, 71; sodium salt,

66. Wightman and Jones 71

2,4,6-Trichlor-w-cresol, 423; acetate, 424. Raiford 423

UNSATURATED compounds and organic zinc compounds, reac-
tion between. Kohler, Heritage and Macleod 217

Uranyl acetate, chloride, nitrate and sulphate, conductivity and

dissociation. Winston and Jones 401

Urea, condensation with esters of allylmalonic and alkyl-substitu-

ted allylmalonic acids. Johnson and Hill 537

Uric acid, conductivity and dissociation of. Wightman and Jones. 102

VAPORIZATION, heat of, relation to other constants at the boil-
ing temperature of some liquids at atmospheric pressure.
Montgomery 298

Viscosity and conductivity in mixed solvents containing glycerol.

Guy and Jones 131

Viscosity and fluidity of suspensions of finely-divided solids in

liquids. Bingham and Durham 278

ZINC acetate and nitrate, conductivity and dissociation. Hosford

and Jones 253

FORMULAS
Cj-GROUP

1 II

CHjOo. Formic acid. Garner, Saxton and Parker 236

C,-GROUP

2 II

C2H4O2. Acetic acid. Loomis and Acree 621

C2H4O4. (i) ci>-Ethene-a„5-dicarboxylic acid (maleic acid).

Keiser and McMaster 521

(2) <ra«5^-Ethene-a:,,5-dicarboxylic acid (fumaric acid).

Keiser and McMaster 510

666 Index

2 III

C2HO2CI3. Trichlormethanecarboxylic acid. Na (p. 66). Wight-
man and Jones 71

C,-GROUP

3 III

C3H3O2N. Cyanmethanecarboxylic acid. Na (p. 66). Wight-
man and Jones 72

C3H3O3N3. 2,4,6-Triketohexahydro - 1,3,5 - triazine (cyanuric

acid). + 2H2O. Wightinan and Jones 103

CjH^OjBr. a - Bromethane - a - carboxylic acid. Na (p. 66).

Wightman and Jones 75

C3H5O2I. /9-Iodethane-a-carboxylic acid. Na (p. 66). Wightman

and Jones 76

3 IV

C3H4ON2S. 2-Thio-5-ketotetrahydro-i,3-diazole. Wheeler, Nico-

let and Johnson 469

C4-GROUP

4 II

CJH2N2. Dinitrile of imwi--ethene-a,/?-dicarboxylic acid. Reiser

and Kessler 523

C4H60g. /-a,/?-Dihydroxyethane-a,/?-dicarboxylic acid. Wight-
man and Jones 84

C4Hg03. a-Hydroxy-a-methylethane-a-carboxylic acid. Na (p.

67). Wightman and Jones 80

4 III

C4H60<S. Carboxymethylmercaptoacetic acid (dithioglycollic

acid). Wightman and Jones 86

C^HjOgBr. a-Brompropane-a-carboxylic acid. Na (p. 67).

Wightman and Jones 79

4 IV

C4HSON2S. 2-Thio-5-keto-4-methyltetrahydro-i ,3-diazole.

Wheeler, Nicolet and Johnson 474

QHeOjNiS. irawi--/9-Thioureinoethane-a-carboxylic acid. John-
son and Shepard 351

C5-GROUP

5 II

CjjHgOa. 7--Ketobutane-a-carboxylic acid (levulinic acid). Na

(p. 66). Wightman and Jones 77

CjHiuOj. /?-Methylpropane-a-carboxylic acid. Na (p. 67).

Wightman and Jones 82

5 III

C5HJO3N4. 2,5,7 - Triketohexahydro - 1,3,4,6 - benztetrazole (uric

acid). Wightman and Jones 102

Index 667

C5H704Br. Dimethyl ester of brommethanedicarboxylic acid.

Kohler, Heritage and Macleod 234

5 IV

CjHgOsNjS. /?-Acetylthioureinoacetic acid. K. Wheeler, Nico-

let and Johnson 472

Cs-GROUP

6 U

CgHjN. Aminobenzene. + HCl. Loomis and Acree 621

QHgO,;. Propane-a,/?,;--tricarboxylic acid (tricarballylic acid).

Wightman and Jones 87

6 III

CgHaBrlg. 2-Brom-i,3,5-triiodbenzene. Jackson and Bigelow . . . 557

CgHgOgS. Benzenesulphonic acid. W ightman and J ones 96

C0H7ON. I - Amino - 3 - hydroxybenzene. + HCl. Jackson

and Bigelow 568

6 IV

Ci.HjNjBrla. 2 - Brom - 1,3,5 - triiod - 4,6 - diaminobenzene.

+ HCl (p. ). Jackson and Bigelow 57 1

CgH^O^NS. 3-Nitrobenzene-i-sulphonic acid. Wightman and •

Jones 98

CyH,,ONBr. 2 -Brom-4-amino- 1 -hydroxybenzene. HCl (p. 419).

Raiford 420

Cr,H,oO;jN2S. a-[/9-Acetylthioureino]propionic acid. Wheeler,

Nicolet and Johnson 473

CjOoNBrjIj. 2,(?) - Dibrom - 1,3,5 - triiod - (?) - nitrobenzene.

Jackson and Bigelow 561

CLO^NoBrlj. 2 - Brom - 1,3,5 ■ triiod - 4,6 - dinitrobenzene. Jack-
son and Bigelow 559

6 V

C^HO^NjBrlj. 2 - Brom - 1,3 - diiod - 4,6 - dinitrobenzene. Jack-
son and Bigelow 562

CgHjONClgBr. 2 - Chlor - 6 - brom - 4 - chlorimino - i - keto -1,4-

dihydrobenzene. RaiJQrd 422

C,.H.,ONClBr. 2 - Brom - 4 - chlorimino - i keto - 1,4 - dihydro -

benzene. Raiford 420

C|,H-,ONClBr. 2 - Chlor - 6 - brom - 4 - amino - i - hydroxyben -

zcne. HCl. Raiford 422

C,-GROUP

7 II

C;HjO-. 3-Hydroxy-i,4-pyrone-2,6-dicarboxylic acid (meconic

acid). + 3H2O. Wightman and Jones loi

CjHcOj. (i) 2,4-Dihydroxybenzenc-i-carboxylic acid. Na (p.

68). Wightman and Jones 93

668 Index

(2) 2,5 - Dihydroxybenzene- 1 -carboxylic acid. Na (p. 68)
Wightman and Jones 94

7 III

C7H4O2CI2. 2 ,6-Dichlor- 1 ,4-diketo-3-methyl- 1 ,4-dihydrobenzene.

Raijord 425

CjH^OgNa. (i) 2,4-Dinitrobenzene- I -carboxylic acid. Na (p.

67). Wightman and Jones 90

(2) 3,5 - Dinitrobenzene - i - carboxylic acid. Na (p. 68).

Wight-man and Jones 91

C7H5OCI3. 2,4,6-Trichlor-i-hydroxy-3-methylbenzene. Raiford . 423
C-H5O2CI. (i) 2-Chlorbenzene- 1 -carboxylic acid. Na (p. 67).

Wightman and Jones 92

(2) 3-Clilor-i,4-diketo-2-methyl-i,4-dihydrobenzene. Raiford 449
C7H5O4N. 4 - Nitrobenzene - I - carboxylic acid. Na (p. 67).

Wightman and Jones 89

CjHgOgCIa- 2,6-Dichlor-3-methyl-i,4-dihydrobenzene. Raiford . 426
C7H7O2CI. 3 - Chlor - 1,4 - dihydroxy - 2 - methyl - 1,4 - dihydro -

benzene. Raiford 450

CjHgOsNa. 2,4,6 - Triketo - 5 - [/? - propenyljhexahydro - 1,3 - di -

azine. Johnson and Hill 540

CyHgOgS. i-Methylbenzene-4-sulphonic acid. Wightman and

Jones 97

CjHgOaNj. 2 - Imino - 4,6 - diketo - 5 - [/? - propenyljhexahydro -

1,3-diazine. -f- 2H2O. Johnson and Hill 541

7 IV

CjHpNCU. (1)5- Chlor - 4 - chlorimino- 1 -keto- 2 - methyl -1,4-

dihydrobenzene. Raiford 446

(2) 2 - Chlor - 4 - chlorimino - i - keto - 3 - methyl - 1,4 - dihy -
drobenzene. Raiford 447, 45 1

(3) 2 - Chlor - 4 - chlorimino - i - keto - 5 - methyl - 1,4 - dihy -
drobenzene. Raiford 444

C7Hj03NBr2. (i) 2,4 - Dibrom - 6 - nitro - i - hydroxy - 3 - methyl-
benzene. Raiford 427

(2) 2,6 - Dibrom - 4 - nitro - i - hydroxy - 3 - methylbenzene.
Raiford 427

C^H^OgNCl. 2 - Chlor -6 - nitro - 3 - hydroxy - i - methylbenzene.

Raifrod 447

C^HpO^NgS. 2 - [/? - Carboxyl - /? - ketoethylmercapto] - 4 - keto -

3,4-dihydro-i,3-diazine. Johnson and Shepard 352

CjHyONBrg. (i) 2,4 - Dibrom - 6 - amino - i - hydroxy - 3 -

methylbenzene. HCl. Raiford 432

(2) 2,6 - Dibrom - 4 - amino - i - hydroxy - 3 - methylbenzene.

HCl. Raiford 428

C7H7O4NS. 4- Amide of benzene- i-carboxylic-4-sulphonic acid.

Na (p. 68). Wightman and Jones 95

Index 669

C7H7O5NS. (i) 2 - Aminobenzene- 1 -carboxylic-4-sulphonic acid.

Kastle and Hadin 508

(2) 2-Nitro-i-methylbenzene-4-sulphonic acid. Wight-
man and Jones 99

CjHgONCl. 2 - Chlor - 6 - amino - 3 - hydroxy - i - methylbenzene.

HCl. Raiford 448

CjHgOgNjS. 2 - Carboxymethyl -4- keto-6-methyl - 3,4 - dihydro -

1,3-diazine. K2 (p. 356). Johnson and Shepard 356

CyHjaOgNxS. Ethyl ester of ;9-acctylthioureinoacetic acid.

Wheeler, Nicolet and Johnson 473

7 V

CyHgOgNBrlj. 2 - Brom - i ,3,5 - trii d - 6 - nitro - 4 - methoxyben -

zene. Jackson and Bigelow 566

CjH^ONClBr,. 2,6 - Dibrom - 4 - chlorimino - 1 - keto - 3-niethyl-

1,4-dihydrobenzene. Raiford 430

Cs-GROUP

8 II

CgHieOa- Heptane- a-carboxylic acid. Na (p. 67). Wightman

and Jones 83*

8 III

C8H2O4CI4. 3,4,5,6 - Tetrachlorbenzene - 1,2 - dicarboxylic acid.

Wightman and Jones 100

C8H4O4CI0. 4,5-Dichlorbenzene-i,2-dicarboxylic acid. Wightman

and Jones 99

8 IV

C8H6O2N4S2. 2 - [2 - Thio - 4 - keto - 1,2,3,4 - tetrahydro -1,3-di-
azine - 5 - mercapto] - 4 - keto - 3,4 - dihydro - 1,3 - diazine.
Johnson and Shepard 354

CgHioOgNjS. Ethyl ester of 2-[carboxymethylmercapto]-4-keto-

3, 4-dihydro- 1,3-diazine. Johnson and Shepard 350

8 V

CgHjOgNBrlg. 2 - Brom - 1,3,5 - triiod - 6 - nitro - 4 - ethoxyben -

zene. Jackson and Bigelow 566

C9-GROUP

9 III

CgHjOjClg. 2,4,6-Trichlor-3-methylphenyl ester of methanecar-

boxylic acid. Raiford 424

9 IV

C9H8O2N4S2. 4-Keto-6-methyl - 2 - [2 - thio - 4 - keto - 1,2,3,4 -
tetrahydro - 1,3 - diazine - 5 - mercapto] - 3,4 - dihydro - 1,3 -
diazine. -|- H2O. Johnson and Shepard 359

670 Index

CsHioOjNjS. Ethyl ester of 2-[/?-hydroxy-a-carboxyethenyl-
mercapto]-4-keto-3,4 - dihydro - 1,3 - diazine. Johnson and
Shepard 351

CgHiiOjNS. 2-Ethylaminobenzene-i-carboxylic-4-sulphonic acid.

Kastle and Hadin 516

CgHjjOjNaS. Ethyl ester of 2-carboxyinethylmercapto-4-keto-6-

methyl-3,4-dihydro-i,3-diazine. Johnson and Shepard 355

Cio-GROUP
10 II

CjoHjgOj. 1,2,2 - Trimethyltetrahydro - R - pentene - 1,3 - dicar -

boxylic acid (camphoric acid). Wightman and Jones 102

10 III

CjoH,o04N.j. ,?-Benzoylureinoacetic acid. Wheeler, Nicokt and

Johnson 467

.CioHj.^03N2. 2,4,6 - Triketo - 5,5 - di[/? - propenyljhexahydro -

1,3-diazine. Johnson and Hill 542

CioHigOoNj. 2 - Imino - 4,6 - diketo - 5,5 - di[5 - propenyljhexa -

hydro- 1,3-diazine. Johnson and Hill 543

10 IV

CioHgOgN^Sj. Compound obtained from thiourea and diethyl
6-oxy-4-methylpyrimidine-2-oxalthioglycollate. Johnson and
Shepard 360

CioHgONjS. 2 - Thio - 5 - keto - 4 - benzaltetrahydro - 1,3 - diazole.

Wheeler, Nicolet and Johnson 470

CioHgO.NBrj. Ethyl 2,4-dibrom-6-nitro-3-methylphenyl ester of

carbonic acid. Kaiford 435

CioHioOjNaS. /?- Benzoyl thioureinoace tic acid. Wheeler, Nicolet

and Johnson 468

CioHnOjNBr^. ( I ) Ethyl ester of 3,5-dibrom-2-hydroxy-4-methyl-

phenylaminoformic acid. Raiford 433

(2) Ethyl ester of 3,5-dibrom-4-hydroxy-6-methylphenyl-
aminoformic acid. Raiford 429

CioH,204N2S. Ethyl ester of 2-[a-carboxyl-/3-hydroxyethenylmer-
capto]-4-keto-6-methyl-3,4-dihydro- 1,3-diazine. Johnson and
Shepard 357

C„-GROUP

11 IV

CiiHjjOgNBrj. 2,4-Dibrom-6-acetylaiTiino-3-methylphenyl ester of

acetic acid. Raiford 434

CiiHjjOaNCl. Acetate of 6-chlor-4-acetylamino-5-methyl-i-hy-

droxybenzene. Raiford 449

Index 671

C12-GROUP
12 II

CjjHjaOa. Isomeric methyl esters of (J-phenyl-a,;--butadiene-a-

carboxylic acid. Reynolds 200

12 III

C,2Hj404N2. [Benzoyliminoethoxymethyljaminoacetic acid.

Wheeler, Nicolet and Johnson 466

12 IV

CjaHinOjNjS. 2 - Thio - 4 - keto - i - acetyl - 4 - benzaltetrahydro -

1,3-diazole. Wheeler, Nicolet and Johnson 472

CJ2H14O3N2S. (i) Ethyl ester of /J-benzoylthioureinoacetic acid.

Wheeler, Nicolet and Johnson 468

(2) [Benzoyliminoethylmercaptomethyl]aminoacetic acid.
Wheeler, Nicolet and Johnson 47 1

CijHi^OoNaS. Diethyl ester of 2-[a,/?-dicarboxyl-/?-ketoethylmer-

capto]-4-keto-3,4-dihydro-i,3-diazine. Johnson and Shepard 352
C13-GROUP

13 II

CjjHj^O^. £-Phenyl-a-pentene-o,(?-dicarboxylic acid. Agj. John- ,

son and Hill 545

13 III

CjaHgNClj. cis- and /raw-y-Chloriminophenyl - 4 - chlorphenyl -

methanes. Peterson 333

C,3H,(,NC1. (i) Iminophenyl-4-chlorphenylmethane. + HCl.

Peterson 332

(2) Chloriminodiphenylmethane. Peterson 329

C,3Hi,i04N2. Ethyl ester of [benzoyliminomethoxymethyl ] amino-

acetic acid. Wheeler, Nicolet and Johnson 466

13 IV

C.sH^OeNjS. Diethyl ester of 4-keto-6-methyl-2-[a,;3-dicarboxyl-
/9-ketoethylmercapto]-3,4-dihydro-i,3-diazine. Johnson and
Shepard 359

Cii^HigOjNS. Diethyl ester of 2-ethylaminobenzene- 1 -carboxylic-4-

sulphonic acid. Kastle and Hadin 5 10

13 V

CijHjgOgNjBrl. Diethyl ester of 2-brom-3-iod-4,6-dinitrophenyl-

methanedicarboxylic acid. Jackson and Bigelow 565

Cj,-GROUP

14 II

C,4H,20j. Hydroxydiphenylmethanecarboxylic acid. Na (p. 67).

Wightman and Jones 74

Ci^HjjOg. Tetraethyl ester of ethane-a,«,/?,/?-tetracarboxylic acid.

Jackson and Bigelow 563

672 Index

14 III

CijH^OaCl. (i) 2-Chlorphenyl-4-methoxyphenyl ketone. Peter-
son 344

(2) 4-Chlorphenyl-4-methoxyphenyl ketone. Peterson 339

Ci^HjgON. Iminophenyl-4-methoxyphenylmethane. + HCl.

Peterson 335

Q^Hi^OaNa. 2,4,6 - Triketo - 5 - [/? " propenyl] - 5 - phenylmethyl -

hexahydro-i,3-diazine. Johnson and Hill 544

Ci4H,502N3. 2 - Imino - 4,6 - diketo - 5 - [/? - propenyl] - 5 - phenyl -
methylhexahydro - 1,3-diazine. + o.5HCl(p. 547). John-
son and Hill 547

Ci4H,703N3. £ - /veto - £ - iminoureino - 0 - phenylmethyl - a - pen -

tene-5-carboxylic acid. Johnson and Hill 546

Ci4H,804N2. Ethyl ester of [benzoyliminoethoxymethyljamino-

acetic acid. Wheeler, Nicolet and Johnson 466

14 IV

Ci4H,iONCl2. cis- and <ra«j-Chlorimino-4-chlorphenyl-4-methoxy-

phenylmethanes. Peterson 342

Ci4H,j02NBr3. 2,6-Dibrom-4-benzoylamino-i-hydroxy-3-methyl-

benzene. Raiford 429

Ci4H,20NCl. (i) Imino - 4 - chlorphenyl - 4 - methoxyphenyl-

methane. + HCl. Peterson 340

(2) cis- and /ra«i^-Chloriminophenyl-4-methoxyphenylmeth-

anes. Peterson 337

C,4HigO,N2S. Ethyl ester of [benzoyliminoethylmercaptomechyl]-

aminoacetic acid. Wheeler, Nicolet and Johnson 471

C15-GROUP

15 II

CjjHaoO. ^-Hydroxy-7--ethyl-jj-phenyl-5,i^-heptadiene. Reynolds. . 210

Ci,-GROUP
17 II
CijHjjOg. Lactone of 5-hydroxy-5,^-diphenyl-a,7--butadiene-a-

carboxylic acid. Kohler 497

CijHijOg. Lactone of ^--hydroxy-^-keto-zS'.cJ-diphenyl-a-butene-a-

carboxylic acid. Kohler 491

Cj7H,402. /?,5-Diphenyl-a,7--butadiene-a-carboxylic acid. Kohler,

Heritage and Macleod 223

Ci 7X1,403. (i) 5-Keto-5,5-diphenyl-/?-butene-a-carboxylic acid.

Kohler 489

(2) Lactone of /y-hydroxy-^-keto-yS'.^-diphenylbutane-a-car-
boxylic acid. Kohler 493

(3) Lactone of ^-hydroxy-cJ-keto-/?,5-diphenylbutane-a-car-
boxylic acid. Kohler 500

Index 673

(4) Lactone of 5,^-dihydroxy-^,5-diphenyl-/9-butene-a-car-
boxylic acid. Kohler 498

(5) Acid from /?-hydroxy-5-keto-/?,5-diphenylbutane-a,a-di-
carboxylic acid lactone. Kohler 492

CjjHjgOa. ^-Hydroxy -/?,5-diphenyl-;--butene-a-carboxylic acid.

Kohler, Heritage and Macleod 223

C,7Hjg04. ;--Hydroxy-(?-keto-/?,5 - diphenylbutane -a- carboxylic

acid. Kohler 501

17 III

CiyHj^OgBrj. Stereomeric /?,;- - dibrom - d - keto - /?,5 - diphenyl -

butane- a-carboxylic acids. Kohler 490

Cj^HijOjBr. (i) /?- Brom - 5 - keto -/?,(?- diphenylbutane - a - car -

boxylic acid. Kohler 494

(2) Stereomeric ;--brom-5-keto-/?,5 - diphenylbutane -a-car-
boxylic acids. Kohler 499

17 IV

CiyHijOaNgS. 2 - Thio - 5 - keto - i - benzoyl - 4 - benzaltetrahy -

dro-i,3-diazoIe. Wheeler, Nicolet and Johnson 469

C18-GROUP ^

18 II

CjgHi^Oj. (i) 5 - Keto - p,d - diphenyl - ^ - butene - a,a - dicar -

boxylic acid. K, K^ -h 2H2O. Kohler 486'

(2) Lactone of /?-hydroxy-(J-keto-/?,^-diphenylbutane-a,a-
dicarboxylic acid. -|- 2H2O. Kohler 487

(3) Lactone of ^,(?-dihydroxy-/?,5-diphenyl-/?-butene-a,a-di-
carboxylic acid. Kohler 488

CigHjgOj. /? - Phenylmethyl - d - phenyl - ;- - butene -a- carboxylic

acid. Reynolds 208

CigHjgOj. Stereomeric £-keto-7-,£-diphenylpentane-/3-carboxylic

acids. Kohler, Heritage and Macleod 225

18 III

CigHj^OsCl. Methyl ester of /9-chlor-(J-keto-/?,5-diphenylbutane-a-

carboxylic acid. Kohler 490

CigHjjOjBr. Methyl esters of the stereomeric ;--brom-^-keto-/?,(?-

diphenylbutane- a-carboxylic acids. Kohler 499

C19-GROUP

19 II

CjjHjgOa. Lactone of £-hydroxy-/?-methyl-;-,e-diphenyl-5-pentene-

^-carboxylic acid. Kohler, Heritage and Macleod 229

CjgH^gOj. Stereomeric lactones of 5-hydroxy-£-keto-/?-methyl-7-,£-
diphenylpentane-/?-carboxylic acid. Kohler, Heritage and
Macleod 232

674 Index

CioH^oOj. Methyl ester of /?-phenybnethyl-^-phenyl-j'-butene-

a-carboxylic acid. Reynolds 209

CidHjoO,. (i) Methyl ester of e-keto-^-.E-diphenylpentane-ZJ-car-

boxylic acid. Kohler, Heritage and Macleod 226

(2) £-Keto - [i - methyl - y.s - diphenylpentane - ,3 - carboxylic

acid. Kohler, Heritage and Macleod 230

(3) ^-Keto-5,(^-diphenylhexane-7--carboxylic acid. Kohler,
Heritage and Macleod 228

(4) Stereomeric a„/?-lactoncs of .5,(?-dihydroxy-/?,(J-diphenyl-
hexane-a-carboxylic acid. Kohler 495

(5) ayf - Lactone of y,d - dihydroxy - /i.d - diphenylhexane -
a-carboxylic acid. Kohler 502

CibHjoO^. ^-Hydroxy - e - keto - /? - methyl - ;-,£ - diphenylpentane -

/5-carboxylic acid. Kohler, Heritage and Macleod 234

19 III

CiflHjgOgBr. d - Brom - s- keto - /? - methyl - y,e - diphenylpentane-

/9-carboxylic acid. Kohler, Heritage and Macleod 231

Co-group

20 II

CaoHjgOj. Dimethyl ester of (?-keto-/?,<?-diphenyl-9-butene-a,a-di-

carboxylic acid. Kohler 484

C20H20O5. Dimethyl ester of 5-keto-,5,o-diphenylbutane-a,a-dicar-

boxylic acid. Kohler, Heritage and Macleod 234

C20H23O3. (i) Ethyl ester of £-keto-^,£-diphenylpentane-/?-car-

boxylic acid. Kohler, Heritage and Macleod 227

(2) Stereomeric ethyl esters of /--hydroxy-T-jS-diphenyl-o-pen-
tene-5-carboxylic acid. Kohler, Heritage and Macleod 225

(3) Methyl ester of £-keto-/?-methyl-;',£-diphenylpentene-/3-
carboxylic acid. Kohler, Heritage and Macleod 231

(4) Methyl ester of iJ-keto-o,C-diphenylhexane-7--carboxylic

acid. Kohler, Heritage and Macleod 228

20 III

C2UH8O7CI4. 4,5,5' - Trihydroxy - 4' - keto - i',4' - dihydroxan -
thane - 7 - [3,4,5,6 - tetrachlorbenzene - 2 - carboxylic acid]
(tetrachlorgallein). NH, + 3.5H2O (p. 35). (NH,), (pp.
32, 48). K + 3.5H2O (p. 49). Omdorff and Delbridge 32

C2oH5,07Cls. 4'-0-Hydrochloride of 4,5,5'- trihydroxy -4'- keto-
i',4' - dihydroxanthane - 7 - [3,4,5,6 -tetrachlorbenzene- 2 -
carboxylic acid] (tetrachlorgallein hydrochloride). Orndorff
and Delbridge 19

C2nH,„08Cl4. (i) 4,5,7,4',5' - Pentahydroxyxauthane - 7 -[3,4.5.6 -
tetrachlorbenzene- 1 -carboxylic acid] (tetrachlorgalleincarbi-
nolcarboxylic acid). Orndorff and Delbridge 37

Index 675

(2) 7, 7 2- Anhydride of 4,5,7,4',5'-pentahydroxy-7-[3,4,5,6-
tetrachlor - 2 - trihydroxyinethylphenyl]xanthane (colorless
tetrachlorgallein hydrate). Orndorff and Delbridge 26

(3) 4.5.5' - Trihydroxy - 4' - keto - 7 - [3,4,5,6 - tetrachlor - 2 -
trihydroxymethylphenyl] - i ',4' - dihydroxanthane (colored
tetrachlorgallein hydrate. Orndorff and Delbridge 21

CjoHiiOsCl-,. 4'-0-Hydrochloride of 4,5,5 '-trihydroxy-4 '-keto-7-
[3,4,5,6 - tetrachlor - 2 - trihydroxymethylphenyl] - i',4' -
dihydroxanthane (hydrochloride of colored tetrachlorgallein
hydrate). Orndorff and Delbridge 30. 46

CjoHijOgCl,. 4,5,7,4',5'-Pentahydroxy - 7 - [3,4.5.6 - tetrachlor -
2-trihydrox)maethylphenyl]xanthane (hydrated tetrachlor-
galleincarbinolcarboxylic acid). Orndorff' and Delbridge 37

QoHj^OsBr. Compound from dimethyl 5-keto-/?,5-diphenyl-/?-

butene-a,a-dicarboxylate and Br. Kohler 486

CjoHigOsBr. Stereomeric dimethyl esters of a,;'-dibrom-5-keto-

/?,^-diphenylbutane-a,a-dicarboxylic acid. Kohler 484

C.oHjiOgBr. Methyl ester of o-brom-£-keto-/?-methyl-r,£-diphenyl-

pentane-/3-carboxylic acid. Kohler, Heritage and Macleod. ... 232

C,i-GROUP

21 II

CjiHaPj. Ethyl ester of £-keto-^-methyl-?-,£-diphenylpentane-/9-

carboxylic acid. Kohler, Heritage and Macleod 230

21 III

CjiHjgOjBr. Ethyl ester of ^-brom-£-keto-/?-methylY,£-diphenyl-

pentane-9-carboxylic acid. Kohler, Heritage and Macleod. ... 232

21 IV

CaiHisOgNBrj. 2,4-Dibrom-6-benzoylamino-3-methylphenyl ester

of benzoic acid. Raiford 432

CaiHieOjNCl. 2-Chlor-4-benzoylamino-5-methylphenyl ester of

benzoic acid. Raiford 444

C,,-CROUP

22 II

CjzH.gOj. Lactone of 7--hydroxy-a,r,?'-triphenylpropane-a-car-

boxylic acid. Reynolds -04

C22H24OJ,. Diethyl ester of 5-keto-/9,f)-diphenylbutane-a,a-dicar-

boxylic acid. Kohler 482

22 III

CazHagOjBr. Stereomeric diethyl esters of 7'-brom-^-keto-/?,5-di-

phenylbutane-a,a-dicarboxylic acid. Kohler 482

676 Errata

C,4-GROUP

24 III

C24H18O8CI4. 7'-Carbonyl etherate of the 7-lactone of 4,5,7,4',5'-
pentahydroxyxanthane - 7 - [3,4,5,6 - tetrachlorbenzene - 2 -
carboxylic acid] (tetrachlorgallein etherate). Orndorff and
Delbridge 14

C25-GROUP

C25H24O. /? - Hydroxy - /? - phenylmethyl - a,l^ - diphenyl - y,^ -

hexadiene. Reynolds 207

25 III

CjsHj^OBr. £,(^ - Dibrom - /? - keto - d - phenylmethyl - a,l^ - di -

phenylhexane. Reynolds 207

CjsHj^OBr^. r.^.s.C - Tetrabrom - ^ - hydroxy - /9 - phenylmethyl -

a,(;;-diphenylhexane. Reynolds 207

C„-GROUP

26 III

CjcHgoOgCl^. 7,4'-Isopropylidene ether of 4,5,7,4',5 '-pentahydroxy-
xanthane-7 - [3,4,5,6 - tetrachlor-2 - trihydroxymethylbenzene
isopropylidene ether] (tetrachlorgallein diacetonate) . Orn-
dorff and Delbridge 3

C,8-GROUP

28 III

CjgHjgOiiCl^. Tetracetate of the 7-lactone of 4,5,7,4',5'-pentahy-
droxyxanthane - 7 - [3,4,5,6 - tetrachlorbenzene - i- carboxylic
acid ] (tetrachlorgallein tetracetate) . Orndorff and Delbridge . 5 1

C29-GROUP

29 II

C29H26O. a - Hydroxy - a,aj,£ - tetraphenyl - d - pentene. Rey-
nolds 203

ERRATA
''^K 225, I. 3 from the bottom, ;--benzal-/3-phenyl-a-methylbutyric

should be 7--benzoyl-5-phenyl-a-methylbutyric.
U--P'. 226, I. 5 from the bottom, ester should be acid.
1,^232, I. 22, C2oH2,02Br should be CjoHoiOgBr.
l""^ 335> I- 14 from the bottom, chlorimido-/>-methoxybenzophenone should

be imido-/>-methoxybenzophenone.
l*^. 352, I. 16 from the bottom, CiaH^O.NjS should be C,2Hi40sN2S.
W^. 356, I. 12, C2H,203N2S should be CgHijOjNaS.
\^. 491, I. II, Ci^HjPsBr should be Q^HiABrj.

EIMER & AMEND

LABORATORY SUPPLIES

205-2M THIRD AVE., NEW YORK

Largest and most complete stock of Chemical
Apparatus & C. P. Chemicals in the U. S.

Edward GuDenflN. Ph.D.

CONSULTINQ ChEHIST ChEHICAL EfiQINEER

POOD. LEGAL AND TECHNICAL EXPERT

903 POSTAL TELEGRAFH:BLDQ.

Chicaqo. ill.

fiS

SANITATS" PORCELAIN!

(gesetzl. geschiitzter Name-Registered Name)
from tbe porcelain works of

W. HALDENWANGER, Spandau

can be obtained from ail liouses that deal in

cbemicai apparatus

18 4 2 IKm^^mflmBK^^m 1 9 11

PLATINUM

CRUCIBLES DISHES

ELECTRODES

SHEET WIRE FOIL

SEND FOR OUR NEW ILLUSTRATED CATALOGUE
PLATINUM SCRAP PURCHASED OR EXCHANGED

J. Bishop ^ Co., Fiatinum Works

REFINERS AND MANUFACTURERS

MALVE,RN, PF.NNA.

Seit November 1908 erscheint:

"ION"

Zeitschrift fiir Elektronik, Atomistik,
lonologie, Radioaktivitat und Raumchemie.

In Verbindung mit hervorragenden Fachgenossen herausgegeben von
Chas. H. Wai,tkr.

Die Zeitschrift "ION," welche in ihrem ersten Bande in englischer
Sprache publiziert wurde, erscheint jetzt in deutscher Sprache, den Wfin-
schen der Mitarbeiter und Leser entsprechend.

"ION" ist die einzige Fachzeitscbrift, welche das Gesamtgebiet der
lonologie allseitig.pflegt, iind ist deshalb nicht nur fiir den Mathematiker,
Physiker, Chemiker, sondern fiir den Naturwissenschaftler fiberhaupt von
Bedeutung.

"ION" erscheint zwanglosin Banden von je 6 Heften zum Preise von
$7.50 pro Band (post frei).

Probehefte werded nicht abgegeben.

Alle AnfragenundBestellungensindzurichtenan: The Canoabury
Institute of physica-ehemical Researchas, Dept. ION.

22 Douglas Road, Canonbary

LONDON, N.

PRODUCTS

Farbwerke vorm. Meister Lucius & Bruening

Aniilae Colors Alizarine Colors

Synthetic Indigo Chemicals

Pharmaceutical and Medicinal Products

Antitoxin ' ' Behring " Tuberculin " KooJi "

Photographic Chemicals

H.A.METZ&CO.

Sole Agents for tiie United States and Canada

New York, N. Y.
Boston, nass.
Philadelphia, Pa.
Providence, R. i.
Cbicaeo, III.

Laboratories
NEWARK, N. J.

Charlotte, N. C.

Atlanta, Oa.

San Francisco, CaL

Hontreal, CanadA

Toronto, Canada

-FOUNDED 1818-

POWERS-WEIGHTMAN-
ROSENGARTEN CO.

NEW YORK PHILADELPHIA ST. LOUIS

Technical C H E M I C A LS Medicinal

COMPLETE LIST ON APPLICATION

Peroxides

Perborates
OXONE HYDRONE

Producing OXYGEN GAS 100/^ PURE Producing HYDROGEN GAS

The Roessler 4 Hasslacher Chemical Co.

100 William St., NE.WYORK.

F. Sartorius

United Worka for Scientilic
Instruments o! F. Sartorius,
A. Becker, 4 Ludwit Tesdorpt

GOTTINGEN, V. (Hannover)

Analytical Balances and Weights

Of World-Wide Reputation

'^f*f FK?*^^

Built on the b>ti> of 40 years' icientific and practical ex-
perience. Highest accuracy with shortest period of swing.
Prizes at all Exhibitions where shown.
Brussels International Exhibition, 1697: Diploma of Honour

and 500 Franc Prize for best workmanship in sensitive

balances.
Paris: Gold Medal. St. Louis: Grand Prix and Gold Medal.
Brussels Inter. Exhibition, 1910: 4 Grand Prix and Gold
Medal.

WHEN BOYING OUR INSTRUMENTS LOOK rOR THE TRADE MARK

Catalogue Gratis and Post Free

I ESCHENBACH

PRINTING COMPANY

FHntefs of Scientific Publica-
tions, Catalogues and Maga-
zines :: :: :: :: ::

EASTON

PENNSYLVANIA

sassssssssssssssi

NOW READY

FOURTH AND REVISED EDITION

ENGINEERING CHEMISTRY

BY
THOMAS B. STILLMAN, M.SC, PH.D.

X + 744 Pages. 174 Illustrations

Price, $5.00

SECOND AND REVISED EDITION

CHEMIST'S POCKET MANUAL

BY
RICHARD K. MEADE, M.S.

xii + 444 Pages. 39 Illustrations

Price, $3.00

THE CHEMICAL PUBLISHING CO.. EASTON. PA.

National Aniline and Chemical Co.

100 William Street, New York

MANUFACTURERS OF

JInlllne Colors and €oal Car Products

,.,..,... o, Dye$tuff$, €l)emical$, Gums, etc.

OFFICES AND WAREHOUSES

MEWYORI BOSTON NEW HAVEN CINCINNATI KANSAS CITY

PHILADELPHIA PROVIDENCE CHICAGO MILWAUKEE MINNEAPOLIS

Works: Sctioeiiitopf, Hartford & Hanna Company, Buffalo.

Henry Heil Chemical Co.

Headquarter* for

Chemical Apparatus
and Chemicals

210-214 South 4th Street. St. Louis, Mo.
Sole Agents for the celebrated MaX DfC VCrhoff *S Filter Paper

Write for our lartfe illustrated catalogue covering 521 patfc*
Satisfaction Guaranteed

Vacuum Drying and

Evaporating Apparatus

Passburg Patents

f Drum 1

VACUUM \ Chamber Dryers
[ R^otary J

Our apparatus have been adopted by all the
leading chemical manufacturers in this country
and Europe.

Catalogue and List of Users cheerfully
sent on request

J. P. DEVINE CO.

BUFFALO, N.Y.

SARSERT'S

Automatic
Water Stills

FOR HEATING BY GAS,
GASOLINE, and STEAM.

Write for Further Information

E. H. SAROENT & GO.

Established 1852
CHICAGO.

Importers, Makers and Dealers in Chemicals and
Apparatus of High Grade Only.

CONTENTS.

PAGE.

Contributions from the vSheffield Laboratory of Yale Uni-
versity :

CXCV. — Researches on Pyrimidines: The Condensation of
Urea and Giianidine -with Esters of Allylmalonic and
Some Alkyl-Suhstituted Allylmalonic Acids. By Treat
B. Johnson and Arthur J. Hill ..... 537
Contributions from the Chemical Laboratory op Harvard Uni-
versity:

i,3,5-Triiod-2-Jirom-^.,6-Diniirobenzcnc and Some of Its

Derivatives. By C. Loring Jackson and H. E. Bigelow 549
The Conductivity of Certain Salts in Methyl and Ethyl Alco-
hols AT High Dilutions. By H. R. Kreider and Harry C.

Jones 574

A Study of the Hydrogen Electrode of the Calomel Elec-
trode AND of Contact Potential. By N. E. Loomis and
S. F. Acree .......... 585

The Application op the Hydrogen Electrode to the Measure-
ment OF THE Hydrolysis of Aniline Hydrochloride, and
THE Ionization of Acetic Acid in the Presence of Neutral
vSalts. By N. E. Loomis and S. F. Acree . . . .621

On Difficulties in the Use of the Hydrogen Electrode in the
Measurement op the Concentration of Hydrogen Ions in
THE Presence op Organic Compounds. By L. J. Desha and *

S. F. Acree 638

Reduction op Mercuric Chloride by Phosphorous Acid and the

Law of Mass Action. By James B. Garner . . . 648

REVIEWS.
Radiumnormalmasse und dcren Vcrwcndung bei Radioaktiven

Messungen .......... 648

An Experimental Course of Physical Chemistry .... 649

Hydrocarbures Alcools et Ethers de la Serie Grasse . . • 6-i9

Organic Chemistry for the Laboratory . . . . . .650

Index . . . . . . . . . 6si

American Chemical Journal

This Journal contains original articles by American chemists; reviews
of works relating to chemistry; reports on progress in the various depart-
ments of chemistry; and items of general interest to chemists. Twelve
numbers of about 100 pages each are published annually, at regular inter-
vals, and form two volumes. A number is issued each mouth. The sub-
scription price is ^.5.00 a year; single numbers 50 cents; Canadian
subscriptions I5. 25; and foreign subscriptions $5.50 per year in advance;
this includes postage.

Fifty copies of each original communication will be furnished, if re-
quested, without charge; any larger number can be obtained at a slight ad-
ditional expense to the author.

Communications concerning subscriptions should be addressed to The
Johns Hopkins Press, Baltimore, Md. Editorial communications should
be addressed to the Assistant Editor, Charles A. RouiLLER, Johns Hop-
kins University, Baltimore, Md.

0. 0. BAKEB, President C. W. BAXER. Vln-PresideBl

PLATINUM

IN ANY FORM

standard Shapes and Sixes ef

Crucibles, Dishes, Triangles

and Manufactured Ware

of all kinds

carried in steeii

Special Forms Made to Order
Scrap Purchased

Platinum Ghloride
Iridium Chloride
Ruthenium Chloride

SEND FOR CATALOGUE

BAKER & CO., Inc.

Platinum Refiners and Manufaclyrers

CORTLAND! BL06., 30 CHURCH ST. N E WA R K
NEW YORK CITY new jersey

2^. Ar ^

^:- ^ ,;^,*^'-'

Tis^f jv.

J^'M.,^

><*^s!*-

:^.;Vx^-

UNIVERSITY OF ILUNOia-URIANA

3 0112 048311937

•C

r*«

^■W'*s^:

\t- '^^

i /"^

tic' 3

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "American chemical journal"

See other formats