Virginia journal of science

The Virginia

' Journal of Science

Vol 111 JANUARY, 1942

CONTENTS

PAGE

Inorganic Analysis with Organic Reagents— John H. Yoe and Lan-

DON A. Sarveb . . . . . . . . . . . . . 1

Third Symposium on Organic Analytical Reagents:

Introduction— John H. Yoe . . . . . . . 8

A Progress Report on the Reactions of Organic Compounds with In¬
organic Ions — W. J. Frierson and Paher M. Simpson . . . 11

/3z.2-Oxyhenzanthrone and Bz.2-Hydroxybenzanthrone as Acid-Base

Indicators— F. H. Fish and W. H. Wrenn . 12

/3-Resorcylic Acid as a Colorimetric Reagent for Ferric Iron — Jean

L. Larner and Wm. E. Trout, Jr . . . . . . . 13

A Progress Report on the Investigation of Organic Compounds as

Reagents in Analytical Chemistry — ^F. T. Lense, C. A. Clever,

AND Edwin C. Markham . . . . . . . 14

The Salinogenic Organic Reagents— Landon Arndale Sarver . . . 16

A Progress Report on Studies in Inorganic Analysis with Organic
Reagents — Ira A. Updike, Oscar W. Clarke, Jr., and Richard

M. Irby, Jr . . . . . . 22

A Progress Report on Studies of Organic Compounds as Analytical

Reagents — John Robert Taylor................. . . . 23

Note on the Ferric Ion-Phenol Color Reaction — John Robert Taylor.. 24
Some Complex Compounds of Copper, Silver and Mercury with Etha-

nolamines — James W. Cole and M. Brooks Shreaves . 27

A Progress Report on Inorganic Analysis with Organic Reagents — E.

Louise Wallace and Alfred R. Armstrong . . . . 28

A Progress Report: The Reaction between Silver and Guanidyl Thio¬
urea Carbonate— Thomas B. Crumpler and Earl B. Claiborne .... 29
A Progress Report on Inorganic Analysis with Organic Reagents —

W. E. Clark and L. R. Stallings . . . . . . . . 31

A Selected Bibliography on Organic Analytical Reagents — ^John H. Yoe 32

^ Published by The Virginia Academy of Science
Monthly, except June, July, August and September
at

The Virginia Journal of Science

Official journal
of the

VIRGINIA ACADEMY OF SCIENCE

George W. Jeffers, President^ State Teachers College, Farmville, Va.

E. C. L. Miller, Secretary-Treasurer, Medical College of Virginia, Rich¬
mond, Va.

Sidney S. Negus, Assistant Secretary-Treasurer, Medical College of Vir¬
ginia, Richmond, Va.

COUNCIL

1941-42

Charles E. Myers .

. 1942

Earle B. Norris .

. 1942

Preston Edwards .

. 1943

Ruskin S. Freer . .

. 1943

Marcellus H. Stow .

. 1944

WoRTLEY F. Rudd . .

. .1944

H. H. Zimmerley .

. 1945

George W. Jeffers .

. 1945

H. B. Haag .

. ..1946

Marcellus H. Stow .

. 1946

EDITORIAL BOARD

Editor4n-Chief — Ruskin S. Freer, Lynchburg College, Lynchburg, Va.

Mam/aging Editor — Lt.-Col. Robert P. Carroll, Virginia Military Institute,
Lexington, Va.

S. A. Mitchell — Astronomy
C. L. Albright — Physics
Robert F. Smart — Biology
J. Stanton Pierce — Chemistry
John Alex. Ti(mER--Education
Robert A Fisher — Engineering
Chapin Jones — Forestry
Edward C. H. Lammers—

Carl C. Speidel — Medicine
R. S. Henneman — Psychology

Entered as second-class matter February 20, 1940, at the post office at
Lexington, Virginia, under the Act of March 3, 1879. Subseriptioni — $1.00
per volume to members of the Virginia Academy of Science; $2.00 per vol¬
ume to others. Published at Lexington, Virginia.

The Virginia Journal of Science

VOL. Ill JANUARY, 1942 No. 1

Inorganic Analysis with Organic Reagents^

John H. Yoe and Landon A. Sarver

The purpose of this paper is to outline and discuss briefly the
various ways in which organic compounds may be used in in¬
organic analysis. The use of organic reagents in analysis is al¬
most as old as analytical chemistry itself, yet comparatively few
workers at the present time possess more than a superficial
knowledge of their advantages and possibilities. Indeed, in the
United States at least, the growing tendency towards specializa¬
tion has seemed to limit more and more the number of those who
are trained both in organic and in analytical chemistry, with its
numerous ramifications in the physical and inorganic fields.

It is not necessary for organic reagents to be superior to in¬
organic ones as a class, in order that their study should be con¬
sidered important; it is enough that in many cases they can do
what inorganic substances cannot do, or that they can do a given
thing better. As a first consideration, the great number and the
variety of organic compounds increase the probability of finding
superior reagents for many purposes; already, more than six
hundred compounds of carbon have been recommended for use
in analysis in one way or another. It was not to be expected that
a majority of these would come into general use, but, neverthe¬
less, a considerable number have achieved conspicuous success.
For example, by conventional methods iron must be precipitated
f rom^ alkaline solution ; but cupferron does this from an acid
solution, thus effecting some otherwise difficult separations.

Organic Solvents and Wash Liquids

In the matter of solutions themselves, we are practically
limited to water among inorganic substances. Liquid ammonia
for instance, could not be considered a convenient reagent; but
in the organic field, the opportunities are greatly multiplied. We
are able to separate salts by differential solubilities in alcohols,
ethers, esters, and ketones when they defy resolution in strictly
aqueous media.

The use of organic compounds as solvents or wash liquids,
where their functions are physical rather than chemical, is very

presented at the 1941 Symposium on Organic Analytical Reagents

1

-

widespread ; yet, at the same time, it cannot be said that a great
deal of ingenuity has been shown so far in the employment of
new substances or in devising new methods for using the old.
One of the earliest recorded uses of an organic reagent in quanti¬
tative analysis was the observation of Serullas in 1831 that po¬
tassium perchlorate may be separated from sodium perchlorate
by extraction with ethanol ; and since that time a great volume
of work has been done in the study of this and other extraction
methods.

Organic liquids may be used in the laboratory for various
physical purposes : ( 1 ) those which are miscible with water may
serve for the washing of vessels and precipitates, and (2) cer¬
tain low boiling ones may be used for the drying of vessels and
precipitates; (3) when added to aqueous solutions of salts, ori-
ganic liquids frequently change the activities of some ions to a
greater extent than they do others, thus reducing the solubility of
a given salt and giving a cleaner separation of it from other
accompanying substances — e. g., the precipitation of lead sul¬
fate and calcium sulfate in the presence of ethanol; (4) in much
the same way organic liquids may serve in dry extractions;
(5) likewise, by their effect on the ionic activities in solution,
organic liquids assist in the displacement of chemical equilibria
in favor of complexes or undissociated colored molecules, and in
numerous cases they increase the sensitivity of color reactions, as
with ferric and cobaltic thiocyanates upon the addition of ethanol
or acetone; (6) reagents, especially organic ones, are frequently
employed in the form of a solution in an organic liquid, and the
same liquid may be used in the subsequent removal of an excess
of the reagent; (7) organic liquids which are immiscible with
water are often extremely useful for extracting and concentrat¬
ing some constituent into an upper or lower layer, especially in
the case of color reactions; (8) immiscible liquids may also be
used for protective layers, to reduce the rate of evaporation or
prevent atmospheric oxidation; (9) likewise, organic liquids may
be used as aids in distillation, as in the use of hydrocarbons' for
the determination of moisture; and (10) by their effect on the
surface tension of aqueous solutions, organic liquids may increase
or decrease foaming, or may aid in collecting small quantities of
precipitates at interfaces.

Several factors need to be considered in selecting an organic
liquid for a given purpose. Primarily, of course, it must perform
whatever function is desired, and normally must do so more ef¬
fectively than any other liquid. C!ost and other things, however,
must also be considered ; intensive efforts have been made to find
satisfactory substitutes for ethanol, because of excise taxes and
stringent regulations on its use. The boiling-point should be
high or low, according to whether the solvent must be removed
quantitatively at some stage of the operation or if it is important

2

to prevent undue loss by evaporation. The vapor pressure is im¬
portant not only in connection with evaporation but also in rela¬
tion to odor and toxicity, since high volatility would be extremely
disagreeable in the case of an odoriferous or poisonous substance.
Wherever possible, non-toxic compounds with pleasant odors are
to be preferred. Although toxicity in the liquid phase would be
somewhat less serious, it is nevertheless undesirable. Also related
to the vapor pressure is the flash-point, which should be as high
as possible in order to reduce the fire hazard. Especially in ex¬
traction methods, the specific gravity of the organic solvent
should differ appreciably from that of the aqueous solution, in
order to obtain satisfactory separation of the layers. In certain
cases, the refractive index and the dielectric constant may also
be of interest.

Organic Acids and Bases

Organic acids and bases are often used for the adjustment of
the hydrogen ion concentration of aqueous solutions, being super¬
ior to ammonia, the alkalies, and the mineral acids for this pur¬
pose. Both they, and their salts, are widely employed as stand¬
ards and in the preparation of buffer solutions. Since many of
them also undergo molecular rearrangement with change of color
upon dissociation, they are extremely valuable as hydrogen ion
indicators. In addition to their neutralizing function, organic
acids and bases may act in any one of several other ways: (1)
they may precipitate insoluble normal salts, such as calcium,
strontium, or barium oxalates; (2) they may tie up certain ions
as soluble complexes, as in the case of aluminum and tartaric
acid; or (3) they may serve as reducing agents, as in the reaction
between permanganate and oxalate ions, or in the use of certain
dyestuffs for the colorimetric detection and estimation of oxi¬
dants.

Organic Oxidizing and Reducing Agents

The number of organic reducing agents is enormous. By this
action, a metal or compound may be preciptated, sometimes as a
colored colloidal suspension suitable for quantitative comparison
with standards. In other cases, a soluble colored ion may result,
which is still better. Or, the organic oxidation product itself
may be colored, and serve for the detection or quantitative de¬
termination of an inorganic ion by direct means. A smaller, but
important, group of organic compounds act as oxidizing agents.
And either oxidizing or reducing substances may be employed as
oxidation-reduction indicators, whenever the two forms have
different colors. Often the color changes are reversible.

Indicators and Other Aids

The great majority of indicators employed in volumetric an¬
alysis are organic compounds; in fact the end-points in each of

3

the three types of volumetric analyses are usually determined by
use of an organic indicator. We may divide these indicators into
three classes:

1. Hydrogen ion indicators^ — for acidimetric and alkalimet-
ric titrations, and for pH determinations.

2. Indicators for oxidation-reduction titrations.

3. Indicators for precipitation titrations.

A considerable number of organic compounds act as adsorp¬
tion indicators, as with dichlorofluorescein in the titration of a
chloride by silver nitrate. Others diminish adsorption or post¬
precipitation of unwanted ions, as is the case with certain alde¬
hydes in the precipitation of copper and zinc sulfides; while
still others, notably proteins, delay the coagulation of precipi¬
tates, and act as protective colloids in colorimetric and nephelo¬
metric determinations.

Chelate Compounds

But the most interesting organic reagents of all are those
which form a ring containing a metal, the products being known
as chelate compounds or inner-complex salts. The chelate com'-
pounds were so named by Morgan because of the imagined struc¬
tural resemblance to the great claws of the lobster, crab, and
other crustaceans (Greek xnH — claw). While, in the majority
of cases, the metal is united to the rest of the compound by two
arms, or bonds, a full consideration of these substances requires
their classification into unidentate, bidentate, terdentate, and
quadridentate compounds, respectively, according to the num¬
ber of points of union involved. Compounds in which the metal
is attached to the body of the substance by one or more co¬
ordinate valences differ profoundly, in many of their important
properties, from those held together by normal or electrovalences
alone. For instance, the electrostatic bond breaks readily in
aqueous solution, giving electrically charged ions and conducting
solutions. Coordination compounds, on the other hand, do not
readily dissociate, and are usually either insoluble or extremely
slightly ionized in water; the reactions, therefore, are generally
complete and irreversible. They differ, too, from normal inor¬
ganic salts in being soluble in hydrocarbons and other organic
solvents; therefore, they may be extracted from aqueous solu¬
tions, thus being concentrated in a small volume. As a class, they
have low melting-points, while those of inorganic salts are high.
They are usually quite volatile, sometimes without decomposition,
at atmospheric pressure. The metal being determined normally
constitutes only a small proportion of the total weight of the
chelate compound. Finally, they frequently react in a very se¬
lective way with one or, at the most, a very few metallic ions;

4

and this is one of the ideal goals of analytical chemistry. This
opens up still another promising field of work, namely, the use
of organic compounds as concentrating reagents for ‘‘trace ele¬
ments'’ in spectrographic analysis and studies. Thus, by means
of organic compounds that react selectively with certain elements
to form slightly soluble precipitates, it should be possible to de¬
tect and to measure quantitatively elements at concentrations
far below the spectrographic limit, when analyses are made with¬
out previously concentrating selectively the constituent in ques¬
tion.

Primary Standards in Volumetric Analysis

Many of our primary volumetric standards are organic com¬
pounds. Few substances, if any, completely satisfy all the re¬
quirements for an ideal standard. Generally a compromise be¬
tween the various requirements must be made, although several
organic substances approach the ideal very closely. Organic
compounds offer a wide variety from which to choose ; a number
of these have been thoroughly investigated and are now extens¬
ively used. Some are not so satisfactory from the standpoint of
the ideal and yet may serve as useful standards under well-de¬
fined conditions.

The Salinogenic Reagents

As already mentioned, many organic substances possess acidic
or basic properties. Because they influence the hydrogen ion
concentration of solutions they are widely used as regulators
of pH, and in the preparation of buffers. This is by no means
their only function, however, and organic acids and bases may
also form salts or complex ions with inorganic ions or salts.
Those compounds which are capable of acting in this way, either
by replacement of hydrogen or hydroxyl ions or by direct addi¬
tion^ of uncharged molecules, may be called “salinogens” or
“salinogenic reagents”, because of this property.

The more common types of salts, which usually dissociate
largely in solution (e. g., calcium, strontium, or barium oxa¬
lates), are known as “true” salts, and when such compounds
exist at all as undissociated molecules, the component parts afe
held together by electrostatic forces. When, on the other hand,
at least some of the parts are held together by covalent forces,
and are capable of existing independently under other condi¬
tions (e.g., the pyridine-thiocyanate complexes), they may con¬
veniently be described as “coordination-complex'” salts, or simply
as “complex” salts ; they may, or may not, be “true” salts as well,
according to whether or not they dissociate in solution to produce
electrically charged ions (e. g., K2PtCl6 is an electrolyte, while
Pt(NH3)2Cl4 is not). And finally, we have the “chelate” or

5

‘‘inner-complex’’ type of salt, which we have already discussed
briefly.

In actual practice, the great majority of salt-forming reagents
which have so far been discovered are acidic in nature. Fortu¬
nately, the field is not limited to the more conventional types of
organic acids, because the carboxy-acids yield only a few salts
of analytical value; of considerably greater interest are other
groups which are capable of splitting off hydrogen ions in solu¬
tion, with subsequent replacement by metallic ions. It also fre¬
quently happens that substances which are not ordinarily con¬
sidered to be acids, do actually yield small concentrations of
hydrogen ions due to keto-enol isomerism, or other types of
dynamic isomerism, when the equilibria are disturbed by the
presence of certain metallic ions which are capable of forming
very slightly soluble or slightly ionized salts.

The most common acidic radical in organic compounds is the
hydroxyl, or OH group; it seldom splits off hydrogen ions with
any great degree of completeness, but it very frequently does
so sufficiently to yield metallic salts. When the carbonyl, or CO
group, is interposed between the hydroxyl and the organic radical
proper, it will be recognized as the essential part of the familiar
carboxyl, or COOH group; and in a similar manner, when N,
NO, SO, SO2, As, or AsO are interposed, we have the oxime (or
enolic form of the nitroso group, NOH), nitroxyl (or enolic form
of the nitro group, NO -OH), sulfinic acid (SO* OH), sulfonic
acid (S02-0H), arsinic acid (As (OH) 2); or arsonic acid
(AsO (OH) 2) groups, respectively. When viewed in this man¬
ner, the oxygenated organic acidic groups assume a much smaller
degree of complexity.

It is also a noticeable fact that replaceable hydrogen is always
attached to a non-metal, and since in the case of an organic com¬
pound the non-metal must be interposed between the hydrogen
and the organic radical, it must have a valence of at least two ;
therefore, only a few elements need to be considered. The inter¬
position of nitrogen and arsenic between the hydroxyl and the
organic radical has already been mentioned; apparently, anti¬
mony and bismuth are too basic in nature to give rise to acidic
substances, and in such examples as are known the hydroxyl
group splits off as a unit in water solution; phosphorus, on the
other hand, does give rise to the corresponding phosphonic,
phosphinic, and phosphinous acids, but these have not been in¬
vestigated as yet from the analytical point of view.

In conclusion, let us emphasize that much experimental work
is needed to establish the relationship between the molecular
structure of organic compounds and their analytical reactions.

6

When a new reaction is discovered, it is then necessary to make
an extensive investigation to determine its nature, limits of
accuracy, its sensitivity, optimum conditions for its use, the in¬
terference of various ions, etc. The vast number and variety of
organic compounds offer a most promising field for new and bet¬
ter analytical reagents and amply justify further research.

University of Virginia, Charlottesville, Virginia.

American Viscose Corporation, Roanoke, Virginia.

7

Virginia Academy of Science
Richmond, Virginia
May 1-3, 1941

Third Symposium on Organic Analytical Reagents
Introduction
John H. Yoe

Today marks the third time our group has met to discuss or¬
ganic compounds as reagents in inorganic analysis; to make
progress reports on investigations being made in the laboratories
of the various co-operating institutions; and to present certain
topics of special interest to workers in this comparatively new
held in chemical analysis. Our group now includes : The College
of William and Mary, Hampden-Sydney College, Mary Baldwin
College, Randolph-Macon College (Ashland), Tulane University,
University of North Carolina, Virginia Military Institute, Vir¬
ginia Polytechnic Institute, Washington and Lee University, and
University of Virginia. We especially welcome our sister insti¬
tution, the University of North Carolina, whose representative
is here in person to take part in today’s program. We also greet
our friends at Tulane University and regret that distance makes
it impractical for them to be with us today in person.

The number of organic compounds investigated by all ten co¬
operating institutions now totals more than 4,500 and includes a
great variety of substances from the standpoint of molecular
structure. These studies should lead to a better knowledge of
the relationship between the structure of organic molecules and
their reactivity as analytical reagents; thus the way should be
opened up for the discovery of new specific and highly sensitive
reagents in inorganic analysis. During the past year several new
and useful organic analytical reagents have been discovered and
these are now being critically studied. Reports today from the
co-operating institutions will outline the progress that has been
made in the respective laboratories since the Symposium held in
Lexington last May.

During the past year at the University of Virginia more than
500 compounds have been investigated, bringing the total num¬
ber examined in this laboratory to approximately 2,400. The
reactions of these have followed the general trend of those pre¬
viously observed; i.e., oxidizing agents, such as Ce+^, Au+^, Fe+^,
Ir+^, and VO+''^, have been the most reactive; especially Ce+^ and
Au+^ which react in a specific manner with many of the com¬
pounds.

About 40 anthraquinone derivatives were included in the
studies but were not especially reactive. Their insolubility causes
a disadvantageous precipitation of the reagent. Moreover, when

8

acting as precipitants, these anthraquinone derivatives react 'with
groups, i.e., rare earths, alkaline earths, and many of the other
metals, thus affording no specific reaction. About 30 benzoan-
throne compounds were tested, and as a class, they are as unre¬
active as any encountered in our studies.

The studies on the various oximes have been continued and
35 additional oximes were prepared and investigated, about half
of them having been synthesized for the first time. Observations
have confirmed pervious generalizations as to the reactivity of
oximes. Two of the oximes give excellent promise as analytical
reagents. Dianisalacetone oxime and cinnamalanisalacetone
oxime precipitate tungsten quantitatively in hydrochloric acid
medium, yielding bright yellow precipitates that are satisfactory
for the gravimetric determination of tungsten. The reagents ap¬
pear to be specific for the W04“^ ion.

Sodium catechol disulfonate has been studied as a possible
reagent for the detection and colorimetric determination of ferric
iron. The purple color formed in acid solution permits the detec¬
tion of 1 part of iron in 20,000,000 parts of solution; in basic
solution the bright red color formed gives a means of determin¬
ing iron at a concentration of 1 part in 50,000,000.

More than 65 substituted thioureas have been carefully
studied. Their reactions were observed and correlated as regards
the respective ions. The sensitivity of these compounds as react¬
ants with Cu+“, Bi+3, Pd+2, -and Se+^ were carefully deter¬
mined and compared with that of thiourea.

Similarly, about 40 substituted diphenylamines have been in¬
vestigated and an attempt made to correlate their reactivity with
the groups present. The substituted groups included NO2, NO,
NIIo, HO, COOH, OCIis, SO3, halogens, etc. Fairly definite con¬
clusions as regards the reactivity of diphenylamines in relation
to the groups present may be formulated. Thus, the presence of
one or two nitro groups either decreases the reactivity markedly
or destroys it entirely; whereas the presence of a hydroxyl or
amino group may alter the color of the product formed but does
not necessarily decrease the reactivity.

The reaction of 2~thio-5-keto-4-carbethoxy-l,3-dihydropyri-
midine with silver was studied in detail. This reagent gives a
brilliant purple color with silver which serves as a sensitive
colorimetric method for this element, detecting as little as 1 part
in 50,000,000 parts of solution. The optimum conditions for the
reaction were determined, including the pH, interfering ions,
temperature effect, concentration of the reagent, and stability of
the colored solution. Spectrophotometric measurements of the
transmission of solutions of the silver compound and of the or¬
ganic reagent were made and optimum conditions established
for the visual and spectrophotometric application of the reagent.

The precipitation of zirconium by means of 5-chloro-broma-

9 ’ ^

mine acid was studied in some detail. The reagent gives a volumi¬
nous, bright red precipitate, but unfortunately no means was
found for making the reaction quantitative. The reaction is, how¬
ever, valuable as a qualitative test for zirconium. Similar studies
were made with copper and again resulted in a non-quantitative
precipitation of copper.

Recently a compound has been discovered that gives an ex¬
tremely sensitive reaction with cupric ions. The sensitivity is of
the order of 1 part of copper in 20,000,000 parts of solution and
the specificity is very favorable. The reaction rate is slow but
an attempt is being made to work out a pratical method for the
colorimetric determination of small amounts of copper.

University of Virginia.

10

A Progress Report on the Reactions of Organic Compounds
with Inorganic Ions

W. J. Frierson and Palmer M. Simpson

During this school year , tests have been completed on two
hundred and fifty organic compounds. Of this number forty-nine
reacted, eleven of which were specific tests. Nine indicators were
found in this group. It is interesting to note that in the work
this year, twenty-eight anthraquinones and sixteen benzoic acids
gave no reaction at all. Anthraquinones and benzoic acids were
especially selected this year because last year three out of eigh¬
teen benzoic acids and ten out of forty anthraquinones gave
promising reactions.

Again this year, as last, the element which gave the most
reactions was gold with thirty-one reactions recorded. Others
reacting most frequently were : silver 16, copper 14, ferric iron
14, osmium 13, tungsten 11, vanadium 10, and palladium 8.

Among the indicators, six had amino groups and four were
phenols. One, thiophenanthrone dicarboxy acid, was fluorescent.
With two exceptions, all the indicators were yellow in basic solu¬
tions and colorless in acid solutions.

Of the specific tests found eight were recorded for tungsten,
two for ferric iron, and one for silver. Six of the organic com¬
pounds which gave a test for tungsten contained a nitro group.

There were no outstanding results from the standpoint of
sensitivity. However, there were some which seemed to have
possibilities. For example, dichlorobenzidine gave a green color
with gold which was sensitive to about one part in five million.
Other compounds gave sensitivity tests of about one or two parts
per million.

The two hundred and fifty compounds investigated this year
make a total of seven hundred and fifty for Hampden-Sydney. No
attempt at present has been made to correlate the structure with
the colors or precipitates observed. Next year, the work at
Hampden-Sydney will be the thorough investigation of the most
promising of the comijounds. The most promising one in the
seven hundred and fifty appears to be the test for gold using 4-
amino-3-methylphenylmorpholine which is sensitive to gold one
part in thirty million.

Hampden-Sydney College.

11

Bz.Z-Oxybenzantlirone and Bz.2“Hydroxybenzanthrone
as Acid-Base Indicators

F. H. Fish and W. H. Wrenn

It has been found that Bz.2-oxybenzanthrone and its enol
form, Bz.2-hydroxybenzanthrone, may be used as acid-base in¬
dicators. Both of these compounds are quite insoluble in water.
Six drops of a saturated solution in 48% alcohol is satisfactory
for a single titration. The color change takes place at pH 7.4.
The indicators are canary yellow in acids and delicate purple in
alkalies; dissolved in 48% alcohol both compounds yield yellow
solutions.

A solution of each indicator was used in titrations and the
precision checked using phenolphthalein and methyl orange. Ti¬
trations indicate that either Bz.2-oxybenzanthrone or Bz.2-hy-
droxybenzanthrone may be used wherever phenolphthalein is
satisfactory.

The color change of the indicators is a function of the hydro¬
gen-ion concentration and the conversion (at the end-point) of
the keto form to the enol takes place almost instantaneously.

Titration data and several curves, the ionization constant for
the indicators, and the effect of carbonates, boiling solutions, and
oxidizing agents will be published later in a full report.

Virginia Polytechnic Institute.

12

/^-Resorcylic Acid as a Colorimetric Reagent for Ferric Iron

Jean L. Larner and Wm. E. Trout, Jr.

Work has been continued on the study .of /5-resorcylic acid
(2,4-dihydroxybenzoic acid) as a colorimetric reagent for ferric
ion. The reagent forms a colorless, stable solution in water, and
is apparently unaffected by sunlight. It is readily synthesized,
and is obtainable from the Eastman Kodak Company. /3-Resor-
cylic acid forms a red complex with ferric ions, which appears to
be stable in sunlight and on standing. The transmission curve
shows a peak between 425m/x and 450m/x and rises to its highest
value in the red.

At the present time it is convenient to produce the color
standards as follows : Suitable volumes of standard iron solutions
are added to 100 ml. Nessler tubes, to which are then added 25 ml.
of buffer solution (1 part 1 N HCl to 1 part 1 N CHsCOONa) and
1 ml. of saturated ,/3-resorcylic acid solution (0.017 M), The mix¬
ture is diluted to the mark. The pH of the solution is 2.9.

The lowest concentration of iron detected under these condi¬
tions, using the Yoe Roulette Comparator, is about 1 part in
20,000,000. The favorable pH appears to be between 2.5 and 3.0.

Several salts interfere with the color formation. The inter¬
fering ions are to be studied.

We are indebted to Dr. Lyle G. Overholser for much assist¬
ance and guidance, and to Dr. John H. Yoe for suggesting the
work, and for his continued inspiration and assistance.

Mary Baldwin College.

13

A Progress Report on the Investigation of Organic Compounds
as Reagents in Analytical Chemistry

F. T. Lense, C. a. Glover, and Edwin C. Markham

Approximately 700 organic compounds were tested as pos¬
sible reagents for the metal ions. The most interesting classes
of compounds were the chalcones, ketones, and aldehydes from
which the chalcones were derived, flavanones, coumarins, sul-
fonium compounds, derivatives of sulfanilamide, various sulfonyl
compounds, and urethanes. Numerous phenols were tested and
results characteristic of phenols were obtained.

Without attempting to be specific, a summary is given below
showing the most striking characteristics of the above classes of
compounds.

Chalcones. Most compounds of this class contained one or
more hydroxyl groups and in some cases methoxy and benzoyloxy
groups in addition. The chalcones are characterized by a great
variety of reactions with metal ions. Some are extremely re¬
active while others react with only two or three metal ions. No
compound in this class was specific, except one which was oxi¬
dized by the ceric ion. In general, the chalcones are more reactive
than the ketones and the substituted benzaldehydes from which
they are made. A study is being made to determine the relation¬
ship between structure and reactivity.

Coumarins. A number of the 4-methyl-7-Ri-8-R2 and 4-
methyl-7-hydroxy-8-R2 coumarins were tested and as a class
were found to be unreactive despite the presence of the hydroxyl
group.

Pyrylium Compounds. This class of compounds, though very
reactive, was so unstable in the presence of either acid or base as
to vitiate the results.

Sulfonium Compounds. Salts of the phenyl phenacyl sulfon-
ium series in most cases reacted with one or more of these ions :
AUCI4— , IrCle"^, Pd+2, PtCl6~^, Ag+, RhCl5“2. It was interest¬
ing to note that iridium reacted with 13 of the 15 salts in this
class. Rhodium reacted with only one.

Sulfonyl Compounds. Included in this group of compounds
are the sulfones, sulfides and sulfonates. Although as a class the
sulfonyl compounds are rather unreactive, some unusual reac¬
tions were discovered. For example, six of the compounds appear
to be specific for osmium. It must be admitted that the purity of
the compounds in some cases was doubtful.

Sulfanilamides. Of the 36 N^ and N^ substituted derivatives

14

of sulfanilamide tested, 16 were found to react with the cupric
ion. In several instances, the compounds were oxidized by the
ceric ion. In only one instance did a compound of this class react
with other metal ions (gold and platinum).

Urethanes. Fourteen compounds of this type were tested and
six reacted with Cu+2, AuCb"", PtCle”^ and Pd+2. One react¬
ed with Cu and Au, while four were specific for copper. The
other three gave no tests. In all cases in which the cupric ion
reacted, a green precipitate resulted, the sensitivity being about
1 p.p.m. One serious difficulty in connection with the use of the
urethanes as reagents for colorimetric procedures lies in their
low solubility in water.

Some of the compounds tested give promise of being useful in
colorimetric analysis and are being investigated.

The authors wish to express their appreciation to Dr. R. W.
Bost and Dr. Alfred Russell who furnished many of the com¬
pounds tested.

University of North Carolina.

15

The Salinogenic Organic Reagents

Landon Arndale Sarver

Although organic substances are frequently used in analysis
as solvents, acids, bases, indicators, et cetera, the most important
and interesting ones are those that form an actual compound with
the element or group being determined.

Some of them form salts in the true sense of the term, e.g.,
calcium oxalate. Others give double salts, such as the pyridine-
thiocyanate complexes, in which ordinary valence forces . do
not come into play at all. But by far the most spectacular re¬
sults have been achieved with a class of substances that are
capable of forming closed^ rings, or chelate compounds.^ In these,
the element being determined attaches itself to the organic mole¬
cule at two or more different points ; most commonly, one of the
bonds is formed by replacement of a hydrogen atom, while the
other is the result of coordination; but both may also be of the
latter type.

A new word, salinogenic^ has been coined to designate salt-
formation of all kinds, whether the product be a true salt like
calcium oxalate, a coordination complex like Pt[NH3]2Cl4, or a
chelate compound like nickel dimethylglyoxime. It thus embraces
salt-like substances as well as true salts, and the reagents may be
either acidic or basic in nature.

In actual practice, the great majority are acidic, and the re¬
action is initiated by the replacement of a hydrogen ion by a
metal. Fortunately, the field is not limited to the more conven¬
tional types of organic acids, because the carboxylic acids yield
only a few salts of analytical value ; of much greater importance
are other groups that are capable of splitting off hydrogen ions,
even though their degrees of dissociation' may not be very great.
Some compounds that do not appear at first glance to be acids at
all, do in fact ionize to a slight extent as a result of dynamic
isomerism.

Briefly, it can be expected that hydrogen may be replaceable
if it is attached to oxygen, sulfur, or imino nitrogen. The hy¬
droxyl group is important not only in alcoholic and phenolic com¬
pounds, but also in a number of other familiar larger radicals,
where it constitutes the essential working part. Thus, the inter¬
position of carbonyl between the hydroxyl and the main body of
the molecule yields the carboxy group; while the interposition
of N, NO, SO, SO2, As or AsO give rise to the oxime or eno^ic
form of the nitroso group, nitroxyl or the enolic form of the
nitro group, sulfinic acids, sulfonic acids, arsinic acids and ar-
sonic acids, respectively. Apparently, antimony and bismuth do
not act in an analogous manner to arsenic ; but phosphorus does
give rise to organic phosphonic, phosphinic and phosphinous

16

acids ; the analytical properties of these latter have not yet been
investigated.

Mercapto compounds are frequently more highly dissociated,
than the corresponding hydroxyl compounds, but fewer examples
are known. Here, too, the interposition of other groups between
the mercapto part and the body of the molecule gives rise to
other acidic groups ; for example, the interposition of CO yields
monothiocarboxylic acids, while CS gives dithiocarboxylic acids
and CS2 produces trithiocarboxylic acids. Other theoretically
possible acids containing nitrogen, phosphorus, arsenic and sul¬
fur have not yet been investigated.

While nitrogen is frequently supposed to be an exclusively
basic element, nevertheless the imino group does split off small
quantities of hydrogen ions ; and these can be replaced by certain
metals, particularly silver and mercury. The best known ex¬
amples are rhodanine and its derivatives. Corresponding sub¬
stances, in which the nitrogen has been replaced by phosphorus
or arsenic, have not yet been investigated.

The value of an acidic reagent will usually be further en¬
hanced if, in addition to the electrovalent linkage, it can also
give rise to a coordinate bond in such a position that the metal
becomes part of a cyclic structure. The essential conditions for
this are: (1) the element that gives rise to the coordinate bond
must be so located in the molecule that the ring which is formed
will contain not less than three or four, nor more than seven
or eight atoms ; the number of such rings containing other than
four, five or six atoms is negligible; (2) the element giving rise
to the coordinate bond with the metal will be a non-metal possess¬
ing at least one unshared pair of electrons in its outer orbit;
in practice, it will be either oxygen, sulfur or nitrogen; (3) the
metal must be able to act as an acceptor, and have space free in
its outer orbits to receive a pair of electrons.

The limitation as to the size of ring that can be produced is
based upon considerations of Baeyer’s ‘‘strain theory’’. The com¬
mon elements behave as if they were regular tetrahedra, with
nuclei at the centers and valence forces diercted towards the
apexes; and the natural angle between lines drawn from the
center to two apexes will be 109° 28'. When two atoms become
united by a single covalent (or directional) bond, the valence
forces are exerted along a straight line drawn from the center of
one through their two respective coinciding apexes to the center
of the other; hence the natural angle between two such adjacent
lines will also be 109° 28'. When, on the other hand, two atoms
are joined by double covalent bonds, the two tetrahedra are con¬
sidered to have an edge in common; and the angle between the
lines representing the directions of the valence forces will be
that between the normal from the center to the common edge and
a line from the center to another apex, or 125° 16'.

17

Now, the values of the angles of equilateral triangles, squares,
pentagons, hexagons, heptagons, and octagons are 60°, 90°, 108°,
120°, 128° 34' 17", and 135°, respectively. If, then, a three-
membered ring is to be formed, the natural valence angles have
to be decreased by 49° 28' if only single bonds are involved, and
by considerably more if even one double bond is present. When
both single and double bonds exist in the same ring, it is difficult
, to say how the necessary deflections will be distributed ; and since
the values of the covalent radii ought also to be taken into ac¬
count, it is probable that not all rings of a given number of atoms
are entirely identical.

For four-membered rings, the average deflections would be
19° 28', 27° 22', and 35° 16' for structures with no double bonds,
and 1 or 2 double bonds, respectively; these values are rather
high, and comparatively few such compounds are known.

Only small deflections are required, however, for five- and
six-membered rings, and large numbers of such substances are
known. For five-membered rings, there would be decreases of
1° 28', 7° 47', 14° 6' for 0, 1 and 2 double bonds, respectively;
while for six-membered rings there would be a decrease of 5° 16'
for 3 double bonds, and increases of 10° 32' and 5° 16' for 0 and
1 double bond, respectively, with no deflection necessary for 2
double bonds.

Rings containing seven or more atoms are comparatively
rare. In order to form a seven-membered ring, increases of 19° 6',
14° 35', 10° 5', and 5° 34' for 0, 1, 2, and 3 double bonds, re-
ispectively; and with eight-membered rings, increases of 25° 32',
21° 35', 17° 38', 13° 41', and 9° 44' would be necessary for 0, 1,
2, 3, and 4 double bonds, respectively. It should be remarked,
however, that negative strain is of less importance than positive,
because it can be dissipated by the departure of certain atoms
from the plane ; and rings containing as many as thirty-two atoms
have actually been prepared.

The preceding calculations are based on the assumption that
all the atoms involved behave like regular tetrahedra ; but recent
work by Pauling^ indicates that the natural valence angles of
oxygen and sulfur are closer to 90° than to 109° 28'. The strain
theory has been very useful in explaining certain facts and pre¬
dicting the discovery of others, but in the present state of our
knowledge it must be regarded only as a useful approximation ;
no doubt it will become exact when all factors have been correctly
evaluated. Particularly, it must be remembered that in chelation
one of the bonds is formed by the replacement of hydrogen, and
that it is therefore an electrostatic, or non-directional one. Cer¬
tain critical requirements as to the length of this bond and to
the magnitude of angular strain will undoubtedly apply, but
they will not be so rigid as in the case where all the bonds are
covalent.

18

The location of the second linkage, which completes the ring,
is dependent upon the existence of suitable electronic conditions.
The coordinate bond is no different from any other covalent one,
except in the manner of its establishment; any covalent bond is
created by the sharing of a pair of electrons ; normally, each of
the atoms furnishes one of the electrons, but stable systemsl can
also be set up when one of the atoms donates a pair which the
other accepts. It happens that nitrogen, sulfur and oxygen all
possess one or more unshared pairs of electrons in their usual
states of combination, and can therefore act as donoTs if they are
so located in the molecule that they can yield rings without too
great distortion of their natural valence angles. The only re¬
maining requirement, then, is the presence of an acceptor with
space for the reception of the proffered pair of electrons.

It is now generally agreed that all the elements are formed
from the same basic building blocks, the positive protons and the
negative electrons ; part of the latter rotate at high velocities in
series or orbits at considerable distances from the central nuclei,
and while the masses of the atoms are determined chiefly by the
number of protons present, the other properties are determined
chiefly by the manner in which the rotating electrons are ar¬
ranged.

On this basis, the elements can be divided into four classes,
as follows :

I. Elements which in their normal or unexcited states have
no incomplete shells of electrons, according to Bohr's theory of
atomic structure, and hence are extremely stable (the rare
gases).

II. Elements which in their normal or unexcited states have
one incomplete shell of electrons, the outermost of which can
easily be lost or completed to an octet, with the formation of ions
of the rare gas type.

III. Elements which in their normal or unexcited states have
two incomplete shells of electrons and give upon excitation ions
that differ markedly from those of the inert gas type (the transi¬
tion elements).

IV. Elements which in their normal or unexcited states have
three incomplete shells of electrons and resemble each other in
that all have the same number of electrons in the two outermost
shells (the rare earth elements).

Of these four classes, the members of the first are completely
non-reactive, the ions of the second have no incomplete shells,
and such differences as exist in the fourth group lie so deep in
the molecules that analytical differentiation becomes practically
impossible. The elements of the second group, however, possess
numerous distinctive properties, including the following:

(1) Variable valency to a marked extent;

19

(2) Strong tendency to form colored ions;

(3) Strong series resemblance, especially with the end mem¬
bers ;

(4) Strong catalytic activity;

(5) Small covalent radii, and small atomic volumes;

(6) Very strong tendency toward the formation of coordinate
bonds.

Since a number of different requirements have to be met in
order to produce a chelate compound, it is not surprising that
some metallic ions will do so with a given reagent while others
will not. The chelate nature of a given salt can usually be es¬
tablished by a consideration of the following facts, but unfor¬
tunately the necessary studies have been made in only a compara¬
tively small number of cases. «

( 1 ) Chelate salts are characterized by high volatility and low
melting points.

(2) They are, as a rule, non-electrolytes and do not take part
in ionic reactions.

(3) They are either very insoluble in water or very slightly
ionized, but soluble in numerous organic solvents.

(4) They have low dielectric constants.

(5) In cases where the necessary data are available, the ex¬
istence of coordinate bonds and chelate rings can be established
definitely by means of a property called the parachor.

The parachor^’® permits the comparison of molecular volumes
at constant surface tension, and the establishment of additive
constants for the individual elements, for double, triple and co¬
ordinate bonds, and for rings of various sizes. If M is the mo¬
lecular weight, D the density, and y the surface tension of the
liquid :

M.yy-

- = The Parachor

D

Unfortunately, however, the constants have not been determined
generally for the metals because of the lack of suitable liquid
substances containing them ; therefore, the parachor has not yet
been of very much value in analytical research.

The considerations outlined in the preceding pages apply to
the basic salinogens as well as to the acidic ones, except that in
the former case there is no replacement of hydrogen and no
electrostatic bond; both linkages are covalent in nature.

It is impossible to enumerate even a reasonable number of
the most important salinogens and to show their structural form¬
ulae in the space available for this paper.'^ Some idea of their
possibilities in analytical research, however, can be gathered
from the fact that more than three hundred known reagents are
covered by the classification given below, and that several of
them give reactions with sensitivities of the order of one part
in a hundred million.

20

I. The Acidic Salinogens

A. Hydroxyl group.

a. Alcohols.

b. Phenols.

c. Lake-forming dyestuffs,

d. Enolizable ketones and diketones.

e. Dioximes and related compounds.

f. Acyloin oximes,

g. o-Hydroxy oximes.

h. Other oximes (including cupferron).

i. Acidic nitro compounds.

j. Nitroso and isonitroso substances.

k. Hydroxy carboxylic acids.

l. Carboxylic acids.

m. Amino acids.

n. Arsinic acids.

0. Arsonic acids.

p. Sulfinic acids.

q. Sulfonic acids.

B. Mercapto group.

a. Simple mercapto compounds.

b. Enolizable sulfur compounds.

c. Other thio compounds.

C. Imino group.

a. Simple imino substances.

b. Enolizing imino compounds.

II. Tpie Basic Salinogens

A. Substituted ammonias.

a. Amines.

b. Amides.

c. Substituted ammonium compounds.

B. Heterocyclic nitrogenous bases.

a. Pyridine and its homologues.

b. Other nitrogen ring compounds.

C. Diazonium compounds.

a. Simple diazonium compounds.

b. Diazo derivatives of arsonic and sulfonic acids.

D. Special reagents for nitrate and chlorate.

a. Substituted amines.

b. Heterocyclic ring compounds (including nitron).

, References

1. Feigl, F., Ind. Eng. Chem., Anal. Ed. 8, 401 (1936).

2. Macloed, D. B., Trans. Faraday Soc. 19, 38 (i923).

3. Morgan, G. T., and Drew, H. D. K., J. Chem. Soc. 117, 1456, (1920).

4. Pauling, L., J. Am. Chem. Soc., 53, 1367 (1931).

5. Sarver, L. A., J. Chem. Ed. 13, 511 (1936).

6. Sugden, S., J. Chem. Soc. 125, 1177 (1924).

7. Yoe, J. H., and Sarver, L. A., “Organic Analytical Reagents”. Book.
John Wiley & Sons, Inc. New York, 1941.

American Viscose Corporation,

Roanoke, Virginia,

21

A Progress Report on Studies in Inorganic
Analysis with Organic Reagents

Ira a. Updike, Oscar W. Clarke, Jr., and Richard M. Irby, Jr.

With possibly one exception, our work has not produced any
useful reagent but it has given us some interesting experience.
The compound which may prove to be useful as a reagent is 1-
hydroxy-2-carboxyanthraquinone, CisHsOs. It gives color changes
or precipitates with no less than 53 ions in the various tests with
inorganic ions at different pH values. It has been suggested that
blocking one or more of these groups by ester or ether formation
might make the resulting compound more nearly specific in its
reaction with inorganic ions. Such work has been planned.

Randolph-Macon College.

22

A Progress Report on Studies of Organic
Compounds as Analytical Reagents

John Robert Taylor

A total of 187 compounds have been examined at Washington
and Lee. Since the last Symposium report, a series of anthraqui-
none derivatives were run through the routine tests with over 70
inorganic ions at different pH values but none gave color or pre¬
cipitation reactions.

A preliminary test of a proposed structure for the ferric ion-
phenol colored complex has been made; a report of this work
appears in the following ‘‘Note'’.

Washington & Lee University.

23

Note on the Ferric Ion-Phenol Color Reaction

John Robert Taylor

In a report on the ferric chloride reaction with a number of
phenols, Wesp and Brode^ proposed for the colored substance the
anionic structure Fe[RO]6~^, where RO” represents the anion of
the phenol ROH. This structure is analogous to that proposed for
the ferrithiocyanate ion by Schlesinger and Van Valkenburgh.^
The ferrithiocyanate structure has recently been criticized by
Bent and French^ who investigated the complex by a photometric
method. An adaptation of the method of these authors was used
in the present tests to find some evidence for the Wesp and Brode
structure of the ferric phenol complex.

If the colored complex has a composition Fen[RO]m, the dis¬
sociation constant for the reaction Fen[RO]m = n[Fe+^] +
m[RO~] is:

Kc [Fe]“ • [RO]- / [Fe,(RO) J ... (1)

If Kp is the ionization constant of the phenol, then [RO] is :

[RO] = Kp.[ROH]/[H+] . (2)

Substituting this expression into equation (1) and writing the
result in logarithmic form yields :

log[Fen(Ra) J = m log[ROH] +

n log[Fe] — m log[H] + log Kp/Kc

If the concentrations of Fe+^ and H+ are now fixed, and a con¬
stant ionic strength is maintained, log [Fen(RO)in] should be a
linear function of log [ROH], provided the phenol' concentration
is large enough to be only slightly affected by reaction with ferric
ion, so that the formal concentration may be considered the same
as [ROH]. The slope of a plot of log [Fen(RO)in] versus log
[ROH] should give an indication of the value of m.

In the preliminary tests reported here, two reactive phenols
encountered during the routine spot testing were used: meta
bromophenol, and alpha naphthol trisulfonic acid. The concen¬
tration of the colored body in solutions of varying phenol con¬
centration, was measured in a visual colorimeter by matching
with a reference solution of fixed phenol concentration. The con¬
centration of ferric chloride was 0.004 M, and was the same in all
solutions. The 0.01 M hydrogen ion concentration was checked
potentiometrically, and was the same in all solutions. The total
ionic strength was adjusted in all solutions to a value of 0.3 by
addition of the required amount of sodium chloride. The inci¬
dent light was passed through a filter transmitting a 15 band
at 515 m/x.

24

[ OH 9H] 2oi

In the plot of results, Fig. 1, the concentration of the phenol
was expressed as a fraction of the phenol concentration in the
reference solution. Concentration of the colored body was ex-

Fi g . I

pressed also as a fraction of its concentration in the reference
solution, that is, as the ratio of the colorimeter scale reading to
the scale reading of the reference solution. The full line has a
slope of 1. The experimental points can be seen to fall close to
this line, rather than to the broken line of slope 6. Consequently
for both phenols, m = 1 approximately. If it is assumed that the
complex contains a single iron atom [n = 1], the structure of the
complex appears to be Fe[RO]+“, or possibly Fe[ROH]+^.

25

Summary

Preliminary tests, using a colorimetric method, indicate that
the colored ferric ion complex of m-bromophenol (and of naph-
thol-trisulfonic acid) has the structure of Fe[RO]+^ where RO~
represents the anion of the phenolic compound.

Further work is now in progress at this laboratory on similar
complexes of simpler phenols, and a more detailed report of re¬
sults will be made later.

References

1. We&p and Erode, Jour. Am. Chem. Soc., 56, 1041 (1934).

2. Schlesinger and Van Valkenburgh, Jour Am. Chem. Soc., 53, 1212 (1931).

3. Bent and French, Jour. Am. Chem. Soc., 63, 568 (1941).

Washington & Lee University.

26

Some Complex Compounds of Copper, Silver and Mercury
with Ethanolamines

James W. Cole and M. Brooks Shreaves

A systematic investigation has been made of reactions of
mono-, di- and tri-ethanolamines with salts of copper, silver and
mercury. In the case of copper, crystalline salts were obtained of
the type, [Cu(ethanolamine)in]X. The values of m were, three
for the monoethanolamine and two for the di- and tri-ethanola¬
mines, when X was the sulfate ion. When X was chloride ion,
compounds with m equal to one were also obtained. In the latter
cases chlorine and/or water appeared to be in the cation.

In aqueous solution, the value of m appeared to be influenced
by the pH of the solution since definite color changes occurred
with variation of pH. Migration of color boundaries under elec¬
trolytic conditions and complete precipitation of anions support
a structure containing ethanolamine in the cation.

Application of electronic theories of valence and manipula¬
tion of molecular scale models (Fisher-Hirschf elder) led to the
conclusion that the functional groups of the ethanolamines (N
and OH) formed chelated rings with Cu++ as a central ion. In
the solid crystalline compounds the coordination number of Cu++
toward the functional groups appears to be six.

Attempts to prepare complex ethanolamine salts with silver
and mercury ions in aqueous solutions resulted in reduction of
the ions to the free metal and formation of complex organic me¬
tallic substances of uncertain composition. Hot concentrated
solutions of the ethanolamines reduced Cu++ to Cu+ and to ele¬
mentary copper and also formed amorphous organic copper
compounds of variable composition.

University of Virginia.

27

A Progress Report on Inorganic Analysis with Organic Reagents

E. Louise Wallace and Alfred R. Armstrong

Since the Second Symposium held last May, seventy-four or¬
ganic compounds have been tested by spot-plate technique with
seventy-odd inorganic ions at different hydrogen ion concentra¬
tions. Ten compounds showed acid-base color changes and nine
gave color reactions with certain of the inorganic ions, eight re¬
acting with auric ions. None of the color reactions, however,
appear to be sufficiently sensitive to be of value in analytical
chemistry.

No attempt has been made to correlate the various structures
of the organic compounds and their reactivity with inorganic
ions. With an increasing number and variety of substances test¬
ed, correlation studies should be helpful in predicting possible
new analytical reagents.

College of William and Mary.

28

A Progress Report: The Reaction between Silver
and Guanidyl Thiourea Carbonate

Thomas B. Crumpler and Earl B. Claiborne

To date 32 compounds have been investigated. Twenty-five
of these were supplied by Dr. Yoe. The remaining seven were se¬
lected from a collection of compounds prepared by advanced
organic students in this Laboratory.

Eighteen of the compounds showed no reaction ■ with any of
the seventy-odd inorganic ions. Thirteen compounds reacted
with five or more ions but showed no promise as reagents.

The only reaction deemed worthy of further study was that
between silver and guanidyl thiourea carbonate, C5H14N8O3S2.
A yellow color is produced which can be differentiated from a
blank with a silver ion concentration as low as 0.4 p.p.m. in a
50 ml, tall form, Nessler tube. The color attains maximum de¬
velopment in 5 minutes and when stabilized with ghatti gum re¬
mains unchanged for a week. Without ghatti gum, the color
changes after 24 hours and eventually a brown precipitate settles
out. The optimum condition for the reaction is in 0.1 N sodium
hydroxide solution with an ammonium hydroxide concentration
of 0.1 N. In the absence of ammonium hydroxide, a turbidity
develops in solution with silver-ion concentrations above 5 p.p.m.
The color fails to obey Beer’s law. Temperature variations from
20-30°C show no effect on color. Matching is most sensitive in
the range of 1-12 p.p.m. of silver. Copper, mercury (both val¬
ences), lead, cadmium and bismuth interfere and should be ab¬
sent. Analyses were performed in which silver was separated
from mixtures by precipitation as chloride in hot solution, subse¬
quently centrifuging and dissolving the preciptate in ammonia.
The soluble complex can be treated with the reagent and the color
developed. The color is independent of ammonia concentration
provided it is at least 0.05 N. Duplicate analyses of National
Bureau of Standards Cast Bronze 52a yielded identical results,
namely 0.010% Ag. This is in good agreement with the average
value of 0.009% given by the certificate of analysis.

Ultramicroscopic examination of the product of the reaction
between silver and guanidyl thiourea shows the presence of col¬
loidal particles. Photomicrographs clearly indicate that the
colored product is neither silver sulfide nor a compound with
thiourea. While the possibility of its being colloidal silver^ is
not completely excluded by the present evidence, it seems likely

^In subsequent experiments, it was obseiwed that when the yellow color was developed in
the absence of ghatti gum, the brown precipitate which settled out after ten days proved to
be metallic silver. This indicates but does not prove conclusively that the yellow color first
developed is due to colloidal silver. In this cnnection see the note on “Silver”, Yoe, “Photo¬
metric Chemical Analysis”, Vol. I (Colorimetry), p. 683, John Wiley and Sons, New York.
1928.

29

that it is a colloidal suspension of a compound of silver with
guanidyl thiourea of the type :

NH2-C-NH-C-NH2

N-

\

V/

Ag

S

✓

Feigl states that the =N~H group is a silver binding group and
the adjacent sulfur offers the possibility of ring closure by co¬
ordinate linkage, to produce a stable, six-membered ring.

In a separate project which was supported by a research
grant from the American Association for the Advancement of
Science through the New Orleans Academy of Sciences, 49 amines
were investigated. These were tested with the following group of
ions: Cr+2, Fe+^, Co+^ Ni+2, Cu+^, Zn+^, RhCls""^, Pd+^ Ag+,
Cd+^, PtCle"”^, AuCU”, and Hg’+^ These ions show maximum ac¬
tivity in forming ammonia complexes. Several reactions of pos¬
sible future interest were observed. The most significant gen¬
eralizations were that amines with OH groups in the side chains
and amines with several NH2 groups (i.e., polyamines) give the
deepest colorations with copper, cobalt, etc. Tri-isopropanola-
mine and triethylene tetramine are the most sensitive reagents
for copper that were tried, being respectively twice and three
times as sensitive as ammonia. The search for other amines of
these two classes is being continued. In addition, it was found
that several slightly soluble di- and trinsubstituted amines with
butyl and amyl side chains form blue complexes with copper ; the
complexes are far more soluble in undissolved excess of the amine
than in water. This appears to offer a delicate and highly efficient
means of separating copper in the form of a highly colored com¬
plex concentrated in a small volume. A micro-colorimeter is be¬
ing developed for the further study of these reactions.

Tulane University of Louisiana.

30

A Progress Report on Inorganic Analysis with Organic Reagents

W. E. Clark and L. R. Stallings

Nineteen organic compounds have been investigated during
the past year for color reactions and for distinctive precipitates
V7hen tested on spot-plates with about seventy-five inorganic ions.
Reactions which show some possibility of practical use included
the following :

(1) 4-Amino-4'-hydroxy-diphenyl sulfide gives a lavender col¬
or with the hypovanadous ion in concentrations as low as 0.2 mg.
per ml. The color shows up in still weaker solutions if allowed
to stand for 15-20 minutes.

(2) 4-Amino-4'-hydroxy-diphenyl sulfide gives a bright yel¬
low color with palladous ions. When compared with a blank, the
color could be readily detected in a solution which contained 5 y
of Pd per ml.

(3) Furo-acetyl-2, 5-diethoxy aniline in acetone gives a yellow
color with 0.03 mg. of uranium (as uranyl ion) per ml. This is
the only intense color produced by the reagent, and consequently
it would appear to be relatively free from interference due to
other cations. It should be studied as a possible qualitative test
for uranyl ion.

(4) 2-Methoxy-4-nitro-phenyl acetate shows some possibility
of being a useful acid-base indicator. It changes from colorless
in acid to yellow in basic solutions.

Virginia Military Institute.

31

A Selected Bibliography on Organic Analytical Reagents

John H. Yoe

1. Berg, R., “Das o-Oxychinoline, Die Chemische Analyse,” Vol. 34.

Ferdinand Enke, Stuttgart, 1935.

2. Bottger, W., (editor) et al, “Newer Methods of Volumetric Analysis.”

Transl. by R. E. Oesper, Van Nostrand, New York, 1938.

3. Brennecke, E., “Neuere Massanalytische Methoden.” Ferdinand Enke,

Stuttgart, 1937. Chapter VI, p. 164.

4. British Drug Houses, “The B.D.H. Book of Reagents for ‘Spot’ Tests

and Delicate Analysis.” Fifth Edition, 96 pages. The British Drug
Houses, Ltd., London, 1936.

5. Browning, E., “Toxicity of Industrial Organic Solvents,” H. M. Sta¬

tionery Office, London, 1937; Philadelphia Book Co., Philadelphia.

6. Clark, W. M., “The Determination of Hydrogen Ions.” Third Edition.

Williams and Wilkins, Baltimore, 1928.

7. Diehl, H., Chem. Revietvs 21, 39 (1937). Diehl, H., “The Applications

of the Dioximes to Analytical Chemistry.” The G. Frederick Smith
/Chemical Co., Columbus, Ohio, 1940. 62 pages.

8. Dobbins, J. T., Markham, E. C., and Edwards, H. L., J. Chem^ Educa¬

tion 16, 94 (1939).

9. Dubsky, j. V., Chem. Listy 31, 66, 84 (1937).

10. Dubsky, J. V., Mikrochemie, Festschr. von Hans' Molisch, 1936, 59.

11. Dubsky, J. V., ibid. 23, 24 (1937).

12. Dubsky, J. V., ibid. 23, 42 (1937).

13. Dubsky, J. V., Brychta, F., and Kuras, M., Pub. faculte sci. univ.

Masaryk No. 129, 1 (1931).

14. Dubsky, J. V., and Danger, A., Chem. Obzor. 12, 27 (1937).

15. Dubsky, J. V., and Danger, A., ibid. 13, 78, 99, 123, 144 (1938).

16. Dubsky, J. V., and Okac, A., Spisy vyddvane prirodovedeokou Facultou

Masarykovy Univ. No. 83, 3 (1927).

17. Dubsky, J. V., and Trtilek, J., Chem. Obzor. 9, 68 (1934).

18. Eastman Kodak Co., Synthetic Organic Chemicals 9, No. 4, 1 (1936).

19. Falciola, P.. Industria Chiynica 6, 1251, 1356 (.1931).

20. Feigl, F., Z. angew. Chem. 39, 393 (1926).

21. Feigl, F., ibid. 44, 739 (1931).

22. Feigl, F., 1st Comm. New Intern. Assoc. Testing Materials (Zurich)

1930, Group D, 216.

23. Feigl, F., Ind. Eng. Chem., Anal. Ed. 8, 401 (1936).

24. Feigl, F., Kolloid-Z. 35, 344 (1924).

25. Feigl, F., Mikrochemie 1, 4 (1923).

26. Feigl, F., '‘Qualitative Analyse mit Hilfe von Tilpfelreaktionen.*’ Third

Edition. Leipzig, 1938. English Translation. Two volumes. Norde-
man Publishing Co., New York, 1937, 1940.

27. Feigl, F., and Gleich, H., Monatsh. 49, 385 (1928).

28. Grant, J., Ind. Chemist 7, 197, 227 (1931).

29. Grant, J., ibid. 8, 169, 217 (1932).

30. Grisollet, H., and Servigne, M., Ann. chim. anal. chim. appl. 12, 321

(1930).

31. Heller, K., Mikrochemie 8, 33 (1930).

32. Hillebrand, W. F., and Lundell, G. E. F., “Applied Inorganic Analy¬

sis.” John Wiley and Sons, New York, 1929.

33. Hopkin and Williams, “Organic Reagents for Metals.” Third Edition,

p. 41. Hopkin and Williams, Ltd., London, 1938.

34. Hustein, K. M., Am. Dyestuff Reporter 22, 442 (1933).

35. Huybrechts, M., Chimie & Industrie Special No., Ill (1931).

36. Karaoglanov, Z., and Dmitrov, M., ibid. 63, 1 (1923).

32

37. Kolthoff, I. M., “Indicators.” Transl. by N. H. Furman. John Wiley

and Sons, New York, 1928.

38. Kolthoff, I. M., “Saure-Base Indikatoren.” This constitutes the fourth

edition of “Gebrauch der Farbindikatoren.” Berlin, 1932; see also,
“Acid-Base Indicators.” Transl. by Rosenblum, New York, 1937.

39. Korolev, A., and Rostovzeva, K., Z. anal. Chem. 108, 26 (1937).

40. Krumholz, P., and Krumholz, E., Mikrochemie 19, 47 (1935).

41. Meisenheimer, J., and Thetlacker, W., In Freudenberg’s “Stereo-

chemie,” p. 1'002, Franz Deuticke, Leipzig, 1933.

42. Meisenheimer, J., and Theilacker, W., ibid. p. 1076.

43. Mellan, I., “Organic Reagents in Inorganic Analysis.” Blakiston Co.,

Philadelphia, 1941.

44. Michaelis, L., “Hydrogen Ion Concentration.” Transl. by W. A. Perl-

zweig. Williams and Wilkins, Baltimore, 1926.

45. Pregl, F., “Quantitative Organic Microanalysis.” Fourth German edi¬

tion, p. 134. Edited by H. Roth and English translation by E. B. Daw.
P. Blakiston’s Son and Co., Philadelphia, 1937.

46. Prodinger, W., Chem.-Ztg. 62, 373 (1938).

47. Prodinger, W., “Organische Fallungsmittel in der Quantitativen'

Analyse.” Stuttgart, 1937. English translation from the Second Ger¬
man edition by S. Holmes. Nordeman Publishing Co., New York,
1940.

48. Raeder, M. G., Tids. Kjemi Bergvesen 18, 131 (1938).

49. Rosenthaler, L., Mikrochemie 19, 17 (1935).

50. Rosenthaler, L., ibid. 20, 85 (1936).

51. Rosenthaler, L., ibid. 21, 215 -(1937).

52. Rosenthaler, L,, ibid. 23, 194 (1937).

53. Sarver, L. a., j. Chem. Education 13, 511 (1936).

54. Scott, W. W., “Standard Methods of Chemical Analysis.” Fifth Edition.

2 Vols. Edited by N. H. Furman. Van Nostrand Co., New York,
1939.

55. Sensi, G., and Testori, R., Ann. chim. applicata 19, 383 (1929).

56. Shemyakin, F. M., Compt. rend. acad. sci. U.S.S.R. 14, 115 (1937).

57. Smith, G. F., Chemical Co., Columbus, Ohio, publishes booklets on

analytical procedures based upon organic reagents.

58. Smith, G. F., “Ortho-Fhenanthroline/^ ibid. 35 pages.

59. Smith, G. F., “Cupferron and Neo-Cupferron,” ibid., 1938. 47 pages.

60. Snell, F. D., and Snell, C. T., “Colorimetric Methods of Analysis,”

Vol. I, Inorganic (1936) ; Vol. II, Organic and Biological (1937), Van
Nostrand, New York.

61. “Tables of Reagents for Inorganic Analysis” (Akademische Verlagsge-

sellschaft, Leipzig, 1938). This book (409 pages, in English, German,
and French) is the first report of the International Committee on
New Analytical Reactions and Reagents of the Union Internationale
de Chemie.

62. Tananaev, N. a., Ukrain. Khem. Zhur. 2, 27 (1926).

63. Tananaev, N. A., and Romaniuk, A. N., J. Applied Chem. (U.S.S.R.)

10, 1624 (1937).

64. Tougarinoff, B., J. pharm. Belg. 15, 174, 189, 205, 223 (1933).

65. Tougarinoff, B., “Les reactions organiques dans I’analyse qualitative

minerale,” Societe scientifique de Bruxelles, 1930.

66. Van Nieuwenburg, C. J., Mikrochemie 9, 199 (1931).

67. Wellings, a. W., “Adsorption Indicators.” A pamphlet published by

The British Drug Houses, Ltd., London.

68. White, W. E., Chem. Education 14, 169 (1937).

69. White, W. E., ibid. 15, 425 (1938).

70. Whitmore, W. F., and Schneider, H., Mikrochemie 17, 279 (1935).

71. Yoe, j. H., “A Laboratory Manual of Qualitative Analysis.”) John

Wiley and Sons, New York, 1938.

72. Yoe, J. H., J. Chem. Education 14, 170 (1937).

33

73. Yoe, J. H., “Photometric Chemical Analysis.’^ Vol. 1, Colorimetry

(1928); Vol. II, Nephelometry (1929). John Wiley and Sons, New
York.

74. Yoe, J. H., and Co-Workers, “Second Symposium on Organic Analytical

Reagents,’^ Va. J. Sci. 1, 121-67 (194'0).

75. Yoe, J. H., and Sarvee, L. A., “Organic Analytical Reagents.” John

Wiley and Sons, New York, 1941.

76. Zan'ko, a. M., and Bursuk, A. Y., Ber. Inst, physik. Chem., Akad.

Wiss. Ukraine S.S.R. 6, 245 (1936).

University of Virginia.

CAROLINA CULTURES

L 1 Giant Amoeba proteus (standard for laboratory study)
eels'*.

Class of 25 (including container and postage) . $2.00

Class of 50 “ “ “ “ 3.50

Class of 100 “ “ “ “ 6.00

Same price as above: Paramecium caudatum, StentoTf VortiGella^
Peranema, Volvox, Mixed Protozoa, Anguillula or ^^Vinegar
L 60 Hydra, Green or Broivn (state preference desired)

Class of 25 (including container and postage) . $1.50

Class of 50 “ “ “ “ 2.50

Class of 75 “ “ “ “ 3.25

Class of 100 “ “ “ “ 4.00

Same price as Hydra: Paramecium multimicronucleata (giant form
of paramecia, excellent for laboratory study), Euglena, Ar-
cella, Chilomonas, Daphnia, Copepods, Spirogyra, Nitella,
Elodea, Cahomba, Myriphyllum.

L 220 Planaria maculata or dorotocephala (the former or light
colored is generally preferred)

Class of 25 (including container and postage) . $1.75

Class of 50 “ “ “ “ 3.00

Class of 75 “ “ “ . 4.00

Class of 100 “ “ “ “ . 5.00

For Drosophila cultures, Tenebrio or “Meal-Worms”, Aquarium
Sets or Assortments, living Frogs, Turtles, Rats, Mice, etc., see our
regular catalogue.

We have a complete line of Preserved Specimens, Microscopic
Slides, Dissecting Instruments, etc. Our publications — Carolina
Tips and general catalogue will be sent free upon application.

Carolina Biological Supply Company

Eloii College; North CaroEna

SMITHSONIAN INSTITUTION LIBRARIES

3 9088 01379 9309

THE QUESTION?

Where is the most conven¬
ient place for me to purchase
my physiology apparatus?

THE ANSWER!

From Phipps & Bird, Inc.,
of Richmond, Virginia.

THE REASON!!

Phipps & Bird has long spe¬
cialized in the development
and manufacture of better
apparatus for physiology,
pharmacology and psychology.
All apparatus is designed to
be simple and sturdy, to meet
the requirements of student
use. Yet, manufacture is to
tolerances well within tiie re¬
quirements for the most ex¬
acting research.

Phipps and Bird are pioneers in this field. First to intro¬
duce an inexpensive, electrically-driven K3nmograph, illus¬
trated above, they have also made available, ink-writing
levers of all types to aid in eliminating tedious smoking
and shellacking.

For modern, efficient, dependable kymographs, heart and
muscle levers, pneumographs, ergographs and other physi¬
ology equipment, IT IS

Phipps and Bird, Inc.

Southern Center for Laboratory Apparatus and Chemicals

RICHMOND, VIRGINIA