Full text of "Handbook of technical gas-analysis"

See other formats

HANDBOOK

TECHNICAL GAS-ANALYSIS.

CLEMENS WINKLER, PH.D.,

PROFESSOR OF CHEMISTRY AT THE FREIBERG MINING ACADEMY.

SECOND ENGLISH EDITION.

Translated from the Third, greatly enlarged German Edition,
with some additions,

GEORGE LUNGE, PH.D.,

IMIOFESSOR OF TECHNICAL CHEMISTRY AT THE FEDERAL POLYTECHNIC SCHOOL, ZURICH.

LONDON:

GURNEY AND JACKSON, 1 PATERNOSTER ROW

(SUCCESSORS TO JOHN VAN VOORST).

MDCCCCII.

ALEBK t FLAMMAM.

FEINTED BT TAYLOK AND FRANCIS,
RED LION COURT, FLEET STREET

TRANSLATOR'S PREFACE TO THE FIRST EDITION.

EVERY one who has to make gas-analyses for technical purposes
is aware that Professor CLEMENS WINKLER is the founder of
technical gas-analysis as a distinct branch of analytical Chemistry.
A few such processes were, of course, previously known and
practised ; but Winkler was the first to draw attention to the
importance of this subject, to invent suitable apparatus, and to
elaborate a complete system of qualitative and quantitative
technical gas-analysis *, containing a vast number of new obser-
vations and methods, along with a very complete description of
the work already done in the same direction by others.

The field first opened out by Winkler has been very successfully
cultivated by other chemists ; and it is now quite usual, at any
rate in Germany, to perform technical gas-analyses, not merely
in chemical works, but for testing the efficiency of steam-boiler
furnaces and such purposes. In England some of these pro-
cesses have also been introduced ; but they are not as yet known
and appreciated to the same extent as abroad. Hence it may not
he unwelcome to English chemists to have a translation of a
short treatise, just published by Winkler, which is primarily
intended for leaching purposes — that is, for the use of teachers
and students in public laboratories —but which likewise serves
as a guide and a handy book to other chemists wishing to make
themselves acquainted with the subject. This treatise is not
intended, as was its predecessor, to furnish a complete enumera-
tion of all apparatus hitherto proposed for technical gas-analysis,
but merely to give representative examples of each kind of
apparatus, embracing all the divisions of this branch of Chemistry.
It may be confidently said that a person who has mastered the
processes and apparatus described in this book will at once

* CL. WINKLER, ' Anleitung zur chemischen Untersuchimg der Industrie-
Gase,' Freiberg, 1877-79 (2 vols.).

104785

comprehend and manage any other gas-analytical process or
apparatus he may meet with or require for his special purpose.
The scope of this book does not in any way embrace the methods of
gas-analysis practised for purely scientific purposes, for instance,
all those in which mercury is employed for confining the gases ;
but it will, for all that, have great interest for scientific chemists.

The selection which the Author has made from the large mass
of material now accumulated was evidently, to a certain extent,
dictated by special circumstances. German sources were mainly
used by him, as these far more than sufficed for the purpose
which he had in view — that of furnishing a sufficient number of
illustrations for all parts of his field. The Translator has been
under a strong temptation to supplement the book by some other
examples of apparatus ; but this proved unmanageable, as the
present treatise would thus have lost its character, as indicated
above, and as then, with greater pretensions, it might perhaps
have been more open than it is at present to the objection that
the treatment of the subject was not sufficiently exhaustive. The
Translator has therefore contented himself with adding a few
notes where they seemed to be specially called for, and with
describing two apparatus of his own construction which have
been found very useful just for industrial purposes, and which
seemed to supply a want. All the additions be has made are
marked, the text being in other respects a faithful rendering of
the German original.

The Translator must acknowledge the most valuable services of
Dr. Atkinson in looking over the proofs and improving the style
of the translation.

All the apparatus mentioned in this book can be supplied by
Messrs. Mawson and Swan, of Newcastle-upon-Tyne, or by any
other dealers in chemical apparatus.

It is hoped, then, that English chemists, gas-managers, engineers,
factory inspectors, and others interested in technical gas-analysis,
will receive this work with favour, and that it will be as widely
employed and as useful as Winkler's works have been in his own
country.

Zurich, August 1885.

TRANSLATOR'S PREFACE TO THE
SECOND ENGLISH EDITION.

THE first edition of the German original of this book appeared
in 1884, a second followed in 1892, and a third was published
towards the end of 1901. In the meantime the first English edition
became exhausted, so that a new issue would have been called
for in any case. Professor Winkler kindly consented to allow
the Translator to do his work from the proof-sheets of the third
German edition, so that this present second English edition
corresponds to the third edition of the original.

Although the scope of this work has remained primarily to furnish
a help to the teacher and the student of technical gas-analysis, it
has been greatly enlarged, as is proved by the fact that the number
of pages has been increased by one-half. It does not even now
purport to give a complete enumeration of all processes and
apparatus proposed for technical gas-analysis, but it now embraces
all the more important of these, including the valuable additions
made to that part of technical analysis by Professor Hempel ;
and it will be found a sufficient guide and help in most cases to
the practical and manufacturing chemist, as well as to the student.
As before, only methods practically tried and approved by the
Author have been recorded in this book.

The Translator has again introduced a few remarks of his own
in the text, and has also made some additions describing his own
apparatus more fully than is done in the original ; but he has
taken care to mark everything in such a way that the reader can
never be uncertain as regards Professor Winkler's work and the
portions added by the Translator.

Zurich, March 1902.

CONTENTS.

Page

PREFACE TO THE FIRST EDITION iii

PREFACE TO THE SECOND ENGLISH EDITION v

INTRODUCTION. General Remarks . 1

CHAPTER I.

ON TAKING SAMPLES OF GASES 5

1. Aspirating-tubes , 5

2. Aspirating apparatus, pumps, bottles, &c 11

3. Vessels for collecting, keeping, and carrying Samples of Gases. 20

•

CHAPTER II.

ON THE MEASUREMENT OF GASES 23

GENERAL REMARKS, CORRECTIONS FOR TEMPERATURE AND PRESSURE. 23

Reduction instrument , 26

I. Direct Volumetrical Estimation 29

A. Measuring in Gas-burettes, tyc 29

Nitrometer 33

Gas-volumeter 41

B. Measuring in Gas-meters 45

II. Estimation by Titration 48

A. Titrating the absorbable constituent while measuring the

total volume of the gas 49

B. Estimation of the absorbable constituents when the non-

absorbable residue of gas is measured 50

Vlll CONTENTS.

Page

III. Gravimetrical Estimation 51

A. Gravimetrical Analysis 51

B. Estimation of Specific Gravity 51

Schilling's apparatus 52

Lux's gas-balance 54

IV. Arrangement and Fittings of the Laboratory 56

- CHAPTER III.

APPARATUS AND METHODS FOR CARRYING OUT THE

ANALYSIS OF GASES 59

I. ESTIMATION OF SOLID AND LIQUID ADMIXTURES 59

Dust, 59. Soot, 61. Naphthalene, 61. Water, 62. Mercury,
62. Sulphuric acid, 62. Hydrocarbon vapours, 63.
Benzene, 63. Ferrocarbonyl, 64. Nitroglycerine, 64.

II. ESTIMATION OF GASES BY ABSORPTION 65

1. Direct Gas- volumetric Estimation 65

A. Absorbing agents for Gases 65

(a) Absorbents for Carbon dioxide 65

(6) „ „ heavy Hydrocarbons 66

Fuming sulphuric acid 66

Bromine water 67

Phosphorus 68

Pyrogallol 70

Copper 72

(d) Absorbents for Carbon monoxide 73

(e) „ „ Nitrogen 75

B. Estimation of Gases by means of Apparatus combining

the functions of Absorbing and Measuring 75

(a) Winkler's Gas-burette 75

(6) Honigmann's Gas-burette , 81

C. Estimation by means of Apparatus with separate parts

for Measuring and Absorption 86

(a) Orsat's Apparatus 87

CONTENTS. ix

Pagfl

(6) Apparatus for estimating Carbon dioxide in Gaseous

Mixtures containing relatively little of it 91

(d) Hempel's Apparatus 93

Gas-burette 93

Gas-pipettes 96

Arrangement and Manipulation of Hampers Appa-
ratus 99

2. Estimation by Titration. 102
A. Estimation by Titration of the Absorbable Constituent

with Measurement of the Total Volume of the Gas ... 103

Hesse's Apparatus 103

R Titration of the Absorbable Constituent, measuring the

Unabsorbed Residue at the same time 107

(a) lleich's Apparatus 107

(6) The Minimetrical Method 112

ring in minute Quantities 116

:>. Estimation by Weight 125

III. ESTIMATION OF GASES BY COMBUSTION 129

1. General Remarks on the Combustion of Gases 129

2. Methods of Combustion 131

A. Combustion by Explosion 131

Hempel's explosion-pipette 131

B. Combustion by means of gently -heated Palladium 139

Palladium-asbestos 140

Manipulation 141

Lunge's Modification of the Orsat Apparatus 146

C. Combustion by means of red-hot Platinum 149

(«) Coquillion's Grisoumeter 150

(6) Cl. Winkler's Apparatus 151

(r) ditto for the Examination of Coal-pit Air containing

Methane 156

(d) Drehschmidt's Platinum-Capillary 160

D. Combustion of Gases-by means of hot Copper Oxide .... 164
Estimation of very small quantities of methane and other

combustible gases 164

X CONTENTS.

' Page

APPENDIX.

1. International Atomic Weights 171

2. Litre-iv eights of Gases and Vapours 172

3. Solubility of Gases in Water 173

4. Changes of Volume when Gazes are burnt in Oxyyen 174

5. Heat of Combustion of Solid, Liquid, and Gaseous Bodies .... 175

6. Standard Solutions for Technical Gas-analuses 170

7. Table for Reducing Volumes of Gases to the Normal State .... 177

ALPHABETICAL INDEX 187

INTRODUCTION,

GENERAL REMARKS.

THE chemical examination of gaseous mixtures, for the purpose of
quantitatively estimating their constituents, is usually effected
by measuring, not by weighing, the latter, owing to the general
physical behaviour of gases : gas-analysis being a volumetric
process, and hence also called gasometry, or gasometric or gas-
vo'umetric analysis.

Consequently the results of gas-analyses are not usually ex-
pressed in per cent, by weight, but in per cent, by volume. In
exceptional cases some of the gaseous constituents are estimated
by weighing ; but, even then, the weight is reduced to the corre-
sponding volume from the well-known weight of a litre of the gas
in question.

Since the volumes of gases are essentially influenced by moisture,
pressure, and temperature, they are measured when saturated with
moisture and under the existing conditions of atmospheric pres-
sure and temperature, as observed at the time by means of the
barometer and thermometer. The volume found in this way
(uncorrected volume] is afterwards reduced to the normal volume -,
that is, from the volume actually observed it is calculated what
volume the gas would occupy in a perfectly dry state at the
normal barometric pressure of 760 millims. and at the normal
temperature of 0° C. (corrected or reduced volume) . This correc-
tion may be omitted if the analyses are very quickly performed
or if they do not require any considerable degree of exactness.

INTRODUCTION.

The analytical process followed in the examination of gases
generally consists in transforming one constituent after the other
into a compound of a different state of aggregation. From the
contraction of volume thus produced, the volume of the special
constituent in question can be deduced directly or indirectly.
This can be done : —

(1) By direct absorption. — For instance,, carbon dioxide is taken
up by a solution of potassium hydroxide, oxygen by moist phos-
phorus, carbon monoxide by cuprous chloride. They are thus
dissolved out, which causes a decrease of the volume of gas
originally employed to the extent of their own volume.

(2) By combustion. — Hydrogen is burned with oxygen, forming
water. In this process two volumes of hydrogen unite with one
volume of oxygen; both gases vanish as such, and a contraction
takes place to the extent of three volumes. Hence the volume
of the hydrogen originally present is found on multiplying by
2 the contraction of volume observed.

(3) By combustion and subsequent absorption of the products. —
Certain gases cannot be directly absorbed, nor are they trans-
formed by combustion into compounds condensing of their own
accord, but these compounds are capable of being absorbed.
Thus methane is burned into water and gaseous carbon dioxide,
which is absorbed by a solution of potassium hydroxide; 1 vol. of
methane and 2 vols. of oxygen (altogether = 3 vols.) in this process
yield 1 vol. of gaseous carbon dioxide. The contraction produced
by absorbing the latter is 3 — 1 = 2 vols. From this we see that
the volume of the methane originally present in the gas can be
found in three ways : —

(o) .By dividing by 2 the contraction accompanying the
combustion.

(b) By absorbing the carbon dioxide formed in combustion,
whose volume is equal to that of the methane.

Gaseous constituents which do not lose their gaseous state,

INTRODUCTION. 3

either by absorption or by combustion, or by combustion and
absorption combined, are measured directly in the state of gas ;
that is, they form the residue remaining at the close of the operation
of gas-analysis. This case refers only to one gas, viz. nitrogen*.

In order to arrive at results which satisfy practical requirements,
without laying claim to the utmost attainable degree of accuracy,
technical gas-analysis must first and foremost aim at working
by the simplest possible means and with the least possible waste
of time. Scientific investigations are not tied to time and hour ;
but where the question is that of practically controlling the
working of some technical process, it is often necessary to get
quickly an idea of this from time to time; or it may be instan-
taneously wanted, even if that idea should be only a rough one.
Analytical results, which the manager of the works can only
receive from the chemist after the lapse of days or weeks, are in
most cases entirely useless to him, let them be ever so accurate.
This must be steadily borne in mind when working out methods
of gas-analysis ; and, fortunately, the progress made during the
last few years has shown that, although the procedure has been
simplified, the accuracy of gas-analyses has steadily increased.

For measuring the gases we employ measuring -vessels of suitable
construction, gauged and divided according to the metrical system,
within which vessels the gases are confined. As confining-liquid
we always employ pure water whenever practicable. Mercury
should be avoided as far as possible ; glycerine and fatty oils,
which do not offer the least advantage, but many inconveniences,
must be entirely avoided. If gases very soluble in water have to
be treated, they are either confined and measured between glass
taps, avoiding any liquid ; or the soluble part of the gases is first
estimated by absorbing it by means of a chemically active solvent
of known strength, and only the unabsorbed gases are subjected
to volumetric analyses. In such cases the absorbable gas is
estimated by titration. In order to avoid any delay by unneces-
sary calculations, the strength of the standard solutions used for

* Argon and its congeners are never separated from nitrogen in technical
analysis. — Translator.

4 INTRODUCTION.

titration should be made to correspond with the volume-weight
of the absorbable gas ; so that a standard solution is considered
normal if a certain measure of it is capable of absorbing exactly
one volume of the gas in question, when corrected for pressure
and temperature.

The gravimetrical estimation of gases is also sometimes per-
formed, in such cases where one of the constituents of a gaseous
mixture, which is present only in small quantity, has to be
estimated. This supposes that the gas in question can be trans-
formed into a compound of constant composition, capable of
being weighed.

Hence the estimation of the volume of a gas can take place : —

(a) By direct measuring •

(b) By titration ;

The absorption of gases is carried on either within the measuring-
apparatus, or preferably outside the same in special absorbing-
vessels. Combustions of gases are always made outside the mea-
suring-vessels. Combustions by explosion, whether with or
without addition of oxy hydrogen gas, should be avoided, if
possible. Care must be taken that during the analytical opera-
tions pressure and temperature suffer no essential changes ; the
laboratory in particular and the confining and absorbing liquids
should have the same temperature; the influence of draughts,
radiant heat, and of other external agencies which alter the volume
of gases must be excluded from the apparatus.

OF THE

UNIVERSITY

CHAPTER I.

ON TAKING SAMPLES OF GASES.

SAMPLES of gases may be taken in various ways according to
circumstances, but it is usually done by means of an aspirator.
Previously to collecting the gas, care must be taken to remove the
air completely from the connecting -tubes and 'other intermediary
apparatus. This can be done by interposing in the connecting-
tube, immediately before its junction with the collecting vessel, a
T-shaped branch whose lateral arm is joined to a small india-
rubber aspirating-pump (see below). By means of this pump it
is easy to remove the air between the place whence the sample
is taken and the collecting-vessel, and to fill the tubing with the
gas under examination ; so that, on the commencement of sampling,
only the latter can get into the collecting-vessel. If the gas is
under pressure, so that it issues of its own accord, the employ-
ment of an aspirating-pump is evidently unnecessary.

1. Aspirating -tubes.

In order to take a sample of gas from any place, such as a
furnace, a flue, a chimney, &c., an aspirating -tube is introduced
into that place in the shape of a tube open at both ends, the outside
end being connected with the collecting-apparatus by means of an
india-rubber tube. It is of very little use to provide the aspirating-
tube with several branches, or with a slit as shown in fig. 1, with
the idea of getting a better average sample in this manner.
Although it is possible, by providing a movable cleaning-rod, to
prevent any soot or dust from stopping up the slit, still such a con-
trivance does not ensure a thoroughly trustworthy average sample,
because the rate of speed of a gaseous current passing through
a flue &c. is not the same throughout, and, owing to friction, is

ON TAKING SAMPLES OF GASES.

considerably less in the neighbourhood of the walls. Moreover,
the gas is more quickly sucked in at that end of the slit which is

Fig. 1.

nearest to the aspirator. Up to the present we do not know of
any process for taking from a moving current of gas such a sample
that it may be said to represent a perfectly correct average of
the whole bulk of the gas. But it is possible to come very near
the truth by producing a strong primary current from the place

Fig. 2.

ASPI RATING-TUBES. 7

of sampling, and branching off a small secondary current from the
former by means of a T-pipe, the latter current forming the average
sample.

In cases where the composition of the gaseous mixture is
subject to frequent and sudden variations (for instance, in furnace-
gases, especially from periodically charged fire-places), it is
preferable to refrain from collecting an average sample, and in
the place of this to take a large number of special samples of
the gas, each of which must be analysed separately.

The simplest and safest sampling can be performed with gases
stored up in quantity, because these generally get mixed of their
own accord by diffusion within the gas-holder*.

The selection of the place from which the sample is taken is
sometimes a matter of importance. Thus, for instance, the
efficiency of an apparatus for absorbing acid gases cannot be
accurately ascertained, unless the samples of gas to be examined
are taken before entering the chimney which forms the final part
of the apparatus, since the gas may be diluted within the chimney
by air entering from without.

The material of the aspirating-tube must be calculated to resist
the prevailing temperature, and not to exert any chemical action
upon the gas.

Wherever it is possible glass tubes are employed for this purpose,
because they are easily constructed, inserted, and cleaned, and
because they are neither acted upon by, nor do they act upon, the
gases. If the temperature admits of it, the glass tube is simply
fixed by means of a perforated cork or caoutchouc stopper, for
instance in sampling the gases of pyrites-burners or vitriol-
chambers (fig. 2) . In such cases it is usually sufficient simply to
bore a hole in the lead ; but if greater durability and tightness
are desired, a small piece of lead tubing may be soldered on to
the hole.

A simple hole is also sufficient to admit of a cork and
glass tube being introduced into the masonry of a chimney or
flue. But it is preferable, especially where many samples of gas
have to be taken at various times, to cement, once for all, a

* It should be observed, however, that this automatic mixture of gases by
diffusion cannot be depended upon to take place expeditiously. Occasionally
the various layers of gases within a large gas-holder show decided deviating
composition for some time after filling the holder. — Translator.

O ON TAKING SAMPLES OF GASES.

porcelain socket-pipe into the hole made in the masonry by means
of common clay or of fire-clay, and to tightly insert the cork with
the glass tube into the socket.

Porcelain \aspir citing -tubes are employed, if the temperature of
the place where the gas is to be sampled is high enough to soften
glass. The porcelain tube should be of such a length that it
projects a good deal beyond the outside of the masonry ; in case
of need the projecting part may be filled with a narrow-mesh wire-
gauze, which usually suffices for cooling the gas passing through.
If the gas is charged with soot or dust, the projecting part is
filled with asbestos or glass-wool in order to retain the solid

Fig. 3.

particles (fig. 3). Porcelain tubes should be gradually heated up,
to prevent their cracking by the heat ; unglazed earthenware pipes,
which are sometimes employed instead, are certainly less sensitive
to changes of temperature, but are not gas-tight, and on this
account cannot be recommended.

A spir citing-tubes made of metal (iron, brass, copper, silver, pla-
tinum) have the advantage of not being fragile, and may be
employed wherever the temperature is not high enough to cause
the metal to fuse, or to allow the gases to diffuse, or to exert a
chemical action upon them. But an inconvenient property of
metals is their great conductivity of:' heat. Corks inserted into
them may be charred; india-rubber tubes, joined to them, generally

ASPIRATING-TUBES. 9

stick fast and soften, or melt altogether. Nevertheless metallic
aspirating-tubes cannot be dispensed with in many cases, and it
may hence become necessary, to avoid the drawbacks just men-
tioned, to provide them with cold-water jackets. In order to
cool the whole length of tubing, the following system may be
adopted : —

Three copper tubes of different width, the metal of a thickness
of 1 or 2 millims., are connected in the way shown in fig. 4.
The innermost tube a is 5 millims. wide, and forms the aspirating-
tube proper; it is surrounded by the second tube b, 12 millims.
wide, which is soldered up tight at one end, the other end towards
A being* left open. This tube has a side-branch d, provided with

Fig. 4.

a stopcock, for admitting the cooling water. The outer jacket is
formed by the tube c, 20 rnillims. wide, which at the end A is
soldered to the tube «, at the end near B to the tube b. The
tube c is also provided with a branch e, through which the cooling
water, which has been heated on its way through the tubes b and c,
is run off again. The length of the tube AB may vary according
to circumstances ; usually 0'6 to 0'7 metre (about 2 feet) will
suffice. The inlet and outlet for the cooling water should be made
wide enough to admit of a rapid flow of water, and so prevent any
formation of steam.

In order to employ this apparatus for withdrawing gases from a
heated furnace, a hole is made in a suitable place in the furnace-
wall, about 3 centims. (say IJinch) wide. The stopcock^ is con-
nected by an india-rubber tube with a water-pipe ; it is then
opened, and, as soon as the water issues at e, the end A is intro-
duced through the hole into the furnace. The joint is at once
made tight by a wet mixture of fire-clay and common clay. The
end a is now connected with the reservoir for the gas and the

ON TAKING SAMPLES OF GASES.

aspirator, in order to withdraw a sample of the gas. Water should
be caused to flow rapidly through the tubes up to the moment
the apparatus is taken out of the furnace.

A similar, but simplified combination of pipes has been recom-
mended by Drehschmidt (fig. 5). The aspirating-tube a, 4 or
5 millims. wide, is surrounded by a jacket b, closed at both

Fig. 5.
d

Fig. 6.

ends, into which cold water is introduced by the supply-pipe c,
running away continuously at d and thus cooling the inner pipe.
The whole is made of copper and the joints are brazed.

Very hot gases should be sucked
off slowly and with careful cool-
ing, because their constituents
may be in a state of dissociation.
When examining gaseous mix-
tures, whose dissociation has par-
tially become permanent in con-
sequence of violent cooling, very
erroneous conclusions may be
arrived at; in such cases espe-
cially we may expect to find
carbon monoxide coexisting with
oxygen.

It has also been attempted to
cool the gases by immediate con-
tact with water, in the manner
shown in fig. 6. The copper tube,
6 to 8 millims. wide, is U-shaped.
The part E C, which is intro-
duced into the hot gas, is provided with a number of fine cuts,
0, 0' 0", 0"', made by a saw, for admitting the gas. m n is a
copper disk, by which the tube is fastened to the outside of the wall
of the furnace. In order to start the apparatus the stopcock A is
opened, and water is thereby admitted, which runs through the
bent tube A C B, and is carried off by an india-rubber tube into

ASPIKATING-PUMPS.

a gas-holder filled with water, where the water and the gas carried
away with it are separated. At first a little water squirts out
through the saw-cuts, but soon the tube C D B begins to act as a
siphon, and, if the cock A. is set rightly, gas is aspirated through
the saw-cuts and collected in the gas-holder.

As in this process the gas is brought into intimate contact with
a large quantity of water, it cannot be avoided that certain gases,
e. g. carbon dioxide, are absorbed to a considerable extent. Hence
this manner of sampling can only be employed in a restricted
number of cases ; but it may do good service where it is only
a question of finding the relative proportions of gases possessing a
slight solubility in water, such as oxygen and nitrogen.

2. Aspirating Apparatus.

In the last-mentioned case the conducting-tube forms at the
same time the aspirator, but usually a special aspirating apparatus
is employed in taking samples of gases. As such we may use, for

Fig. 7.

instance, inaia-rubber aspirating- and force-pumps (hand or foot
blowers) of simple construction, as met with in commerce in various
sizes (fig. 7). They consist of a stout vessel A, with cylindrical
ends, which are stopped up with turned and perforated wooden
bungs, provided inside with a very simple kind of valve (viz.,
leather clacks with a pasteboard strengthening). India-rubber
tubes of differing length are attached to these bungs ; the shorter
piece 0, about 20 centims. long, forms the aspirating-tube, the
longer piece £, about 40 centims. long, the discharging-tube. On
compressing the vessel A by hand or by foot, its gaseous contents
are forced out through b ; when the pressure is relaxed the elastic

12 ON TAKING SAMPLES OF GASES.

vessel resumes its former shape, and is thereby filled with a new
supply of gas through a. By a continued alternation of these
two manipulations considerable quantities of gas may be aspirated
and forced away within a short time, say 12 to 18 litres per
minute, and the valves close tightly enough to overcome a pressure
of several metres of water. This contrivance is extremely con-
venient for filling a bottle, a tube, or any vessel whatsoever with
the gas to be examined. In this case no confining (luting) liquid
is required, but there must be an ample supply of the gas in
question, for the air previously present can only be assumed as
having been entirely replaced by the gas provided the five-fold
volume of the latter has passed through the blower.

Where high-pressure steam can be had, gases may be aspirated
continually or for a great length of time by means of a steam-jet
aspirator (fig. 8). A strong glass tube, about 3 centims. wide, or

in lieu of this a metallic tube of a length of 20 to 25 centims., is
drawn out at one end to an orifice. of 6 millims. width ; a steam-
pipe is fixed in its longitudinal axis in such a way that its point,
tapering to 2 millims. bore, ends about 12 millims. behind the
orifice of the outer tube. Near this point the steam-pipe is kept
in its place by a ferrule «, made of wood or metal ; at the other
end it is tightly fixed in the cork b, which, in its second per-
foration, carries the tube e, through which the gas is aspirated.
This cork, to make the juncture firmer, is covered with a layer
of cement c, and the whole is confined in a metal ferrule d. The
aspirator should be joined to the steam-pipe g by an india-rubber
tube with hemp lining/, since ordinary india-rubber tubing does
not resist steam-pressure.

Apart from these "dry" aspirators, a. considerable number are
constructed with a water-luting.

It is frequently necessary to aspirate a gas continuously for a
long time, whether in order to measure its volume in a gas-meter,
or to take a smaller sample from it, or to absorb one of its

BUNSEN'S WATER-AIR PUMP.

constituents present in very small quantity. In such cases it is

customary to apply that kind of aspira-

tor in which the gas is carried away by

a jet of water, and whose efficiency may

be sufficient to overcome the atmo-
spheric pressure. Very many such

apparatus have been constructed, of

which only a few of the best are here

described.

Bunsen's water- air pump* (fig. 9)

consists essentially of a cylindrical

glass vessel A, into whose contracted

upper end a narrower glass tube is sol-
dered, communicating on one side with

the glass vessel B, and on the other

side reaching nearly down to the lower

contraction of A, where it ends in a

fine orifice. To the lower end of A is

joined a lead pipe b, 8 millims. wide,
10 to 12 metres long, and bent up at
the lower end so that some water is
retained here. The side- branch a is
connected with a water- reservoir or
with the service-pipe ; the flow of water,
which need not take place under pres-
sure, can be once for all set to a certain
rate by means of a screw-clamp, and
completely shut off by another. If
water is run in through a, the lead pipe
b is filled with a column of water balancing the weight of the atmo-
sphere, and the jet of water following this carries air along through
Cj in order to yield it up only at the lower end of the lead pipe.
If c remains open, the air is continuously and strongly sucked in, so
long as the flow of water is not interrupted. If, however, c, or a
space communicating with c, is closed, a vacuum is produced, corre-
sponding to the Toricellian vacuum of the water-barometer formed
by the apparatus. The vessel B is not essential for the purpose
of aspiration ; its object is principally to retain any liquid carried
along mechanically, and to admit of discharging the latter through
* Really invented by H. Spreng'eL — Translator.

ON TAKING SAMPLES OF GASES.

/from time to time. The tube d is connected with a mercurial
pressure-gauge, which indicates the progress of the evacuation ; e is
the continuation of the aspirating-tube c, and is connected with
the space which is to be evacuated, or from which a sample of gas
is to be taken. Bunsen's pump requires no head of water, but a
considerable length of downward pipe as above-mentioned ; if,
however, the purpose is not that of complete evacuation, but merely
of aspirating gases, the downward tube may be shortened down to
1 metre, or even less than that. The long lead tube b may then
be replaced by an india-rubber tube, closed at the lower end by a
bent glass tube.

The water-jet pumps as constructed by Arzberger and Zul-
kowsky, H. Fischer, Korting Brothers, Th. Schorer, and others
work well and do not require any height of fall for the waste water ;
but, on the other hand, they require a head of 5 to 10 metres
of water for feeding. Their construction, which has been very

Fig. 10.

much varied, is apparent from fig. 10. The water enters at A,
issues from the conical tube «, 1 millim. bore, carries along the
air entering through J3, passes the contracted part b, and runs

WATER-JET PUMPS.

off at C. The three tube-ends A, B, and C are connected with
the corresponding pipes by elastic tubing ; the neck D is held
in a support. A small vacuum-meter communicating with B
indicates the degree of lessened pressure.

O£ a simpler kind but hardly less efficient are the various water-
jet pumps made of glass, which are found in commerce. These can
be connected with any water-tap by means of thick india-rubber
tubing ; they are easily moved about and are, moreover, deep. To
these belongs Finkener's aspirator (fig. 11), where the water enters

Fig. 11. Fig. 12.

from the service-pipe through the tube a, which is drawn out to a
point ; runs through the tube c, which is bell-shaped at the top,,
contracted in the middle, and again widened at the bottom end ;
and aspirates air through b, which forms a frothy mixture with the
water issuing at c. In order to diminish the fragility of the
apparatus, it is usual to make the lower, tapering tube separate
and connect it with the upper part by an elastic joint. Very
efficient also are Geissler's aspirating -tube*, which can be understood
from fig. 12 without special explanation.

ON TAKING SAMPLES OF GASES.

Another kind of apparatus admits of both aspirating and col-
lecting the gases, sometimes also of measuring them, or rather
that part which is not sensibly soluble in water. In many cases
the analytical apparatus itself, such as the gas-burette or the gauged
collecting-bottle, is employed as an aspirator, by being filled with
water, which is made either to run off within the space containing

the gas to be examined or after connecting the apparatus with the
aspirating-tube.

If somewhat large quantities of gases have to be collected, an
aspirating -bottle, as represented in fig. 13, may be employed. This
bottle A is placed on a wooden stool ; its india-rubber cork is
provided with a glass stopcock a and a tube b, reaching nearly
down to the bottom, and on the outside is connected by means of
an elastic tube with a straight glass tube of sufficient length to act

ASPIRATORS.

14.

as a siphon, capable of drawing off all the water contained in the
bottle. The connecting elastic tube can be closed by a screw-
clamp, which also permits the outflow into the vessel B to be
regulated. Before taking the sample the aspirating-bottle is filled
with water by so altering the levels, with the aid of the siphon,
that no air-bubbles remain, and the water ultimately rises to the
top of the stopcock a. Now by means of the pump c all the air
is removed froni the gas-conducting pipe, after suitably setting
the three-way tap ; the gas-pipe is connected with A, and the gas is
aspirated by allowing the water to flow out. Such an arrange-
ment can, for instance, serve in those cases where a reduced
sample is to be taken from a main current of gas, aspirated during
the whole course of work by means of a water-jet pump. This
reduced current should be removed quite as continuously and
collected in bottle A.

Robert Muencke's double aspirator (fig. 14) is very convenient,
especially where volumes of gas ap-
proximately equal have frequently
to be aspirated. Two bronzed
cast-iron pillars support in suitable
bearings a steel spindle, to which
are at tached, in opposite directions,
two cylindrical glass vessels of
kriown volume, communicating by
means of a stopcock, which also
regulates the outflow. A simple
spring arrangement, attached to
the front part of the spindle or
the upper part of the front pillar,
serves for fixing the glass vessels
in a perpendicular position. Each
glass cylinder is closed with a brass
plate affixed with screws, pierced by
a bent brass tube reaching nearly
down to the bottom of the vessel,
and on the outside connected with

an elbow-piece, to which the two pieces of tubing are attached
which communicate with the stopcock fixed to the ground plate.
This stopcock is marked for the vessels A and B, and is bored in
such a way that, if turned as in the figure, it brings the upper
vessel A into communication with the apparatus through which gas

ON TAKING SAMPLES OF GASES.

is to be aspirated^ and at the same time connects the lower vessel B
with the atmosphere. When the upper vessel is run off, the button
of the spring-arrangement is pressed, the cylinders shifted round the
spindle by 180°, and the lower stopcock turned to the same extent.
In this position the vessel B is connected with the gas-apparatus
and A with the atmosphere.
Thus this double aspirator per-
mits of almost continuous action
without any change of tubing.
A very convenient form of
zinc aspirators is shown in
fig. 15. The vessel A, con-
taining 10 to 15 litres, is placed
in a wooden stand ; it ends at
the top in a stopcock Z>, and at
the bottom in a slightly tapering
tube, provided with the tap c,
and bearing a thin brass tube d,
through which the water can
flow off regularly without air-
bubbles entering the vessel.
The side-branch a, also provided
with a tap, serves for filling with
water. The aspirator should be
filled with water of the same
temperature as the laboratory ;
or, if the water is taken from a
service-pipe, it must be allowed
sufficient time to acquire that
temperature. This is indispens-
able if the aspirator is to serve
at the same time for measuring
the volume aspirated, for which
purpose it is very well adapted.
When employing it for such
measurements, the tap b is con-
nected by means of an air-tight
screw-socket with a glass T-
piece, the upper limb of which ends in a small mercurial pressure-
gauge, whilst the side limb is connected with the aspira ting-pipe.
By opening the taps b and c the gas is aspirated and the water

AUTOMATIC ASPIRATOR.

which runs off is collected in a litre-flask placed below. The
moment the water in the latter has reached the mark, the tap c
is closed, but b is only closed the moment the pressure-gauge
has come to the level again. When this happens a volume of gas
exactly equal to thatsofthe water run off has been aspirated.

For aspirating and at the same time measuring large volumes
of gas, the automatic aspirator of J. Bonny * can be employed.
The essential feature of this apparatus (fig. 1 6) is a metal vessel A,

Fig. 10.

containing on the inside a siphon whose shorter funnel-shaped
limb reaches down to the lower part of the vessel, whilst
the longer limb is carried through its bottom and dips into the
vessel B, in which is water at a constant level. Through the
india-rubber tube w, connected with the water-service, the
vessel A can be filled with water. The tube g serves for the
entrance and exit of the gas ; it communicates with the bottles
C and D, the first of which serves as a water-lute, the second

* Sold by Cornelius Heinz & Co., Aachen: price 75 marks.

20 ON TAKING SAMPLES OF GASES.

for containing the absorbing liquid through which the aspirated
gas is to be passed. The apparatus begins to work as soon as
the tap of the service-pipe is opened and water enters into A.
Whilst this vessel is filling the gas contained in it escapes
through the bottle (7; but as soon as the water-level is up to
the bend of the siphon, this begins to act and the water flows
into the vessel B. If the supply of water through w is regulated
so that it is less than the outflow into B, the level of water in A
must sink, and the gas will be aspirated through the absorbing-
bottle D, from which it passes into A. But as soon as the water
has run off to the end of the shorter limb of the siphon, the latter
ceases to act, and only starts working again when the vessel A has
been again filled with the water which continuously flows through
w. The volume of gas aspirated every time the siphon acts is equal
to the contents of the vessel A between the highest and lowest
levels, which has been gauged once for all : the number of times is
registered by the indicator E, which moves every time the vessel
A is raised. The latter is hung from the top of a portable box by
means of a spiral spring, which is compressed when A is filled and
extended as A empties. This contrivance causes the differences
of level between A and B to be equalized.

3. Vessels for collecting, keeping, and carrying Samples
of Gases.

Unless unavoidable, a sample of gas should not in any case be
kept for any length of time, but ought to be transferred at once
to the analytical apparatus — such, for instance, as a gas-burette or
an absorption-bottle — in order to be instantly analyzed. A rule
to be observed in cases where it is unavoidable to employ water-
luting is this : to bring the water merely into superficial and
momentary contact with the gas, but never to pass the gas through
the water itself, as is done in a pneumatic trough. Otherwise
the solvent action of the water, which is entirely different towards
different gaseous substances, would unavoidably alter the com-
position of the gas to a sensible extent.

If, however, the collection of the gas in a separate vessel for the
purpose of keeping it for some time or transporting it to some
distance cannot be avoided, care must be taken not merely to
exclude the air completely from it, but also to entirely remove
the water employed in taking the sample, as this would otherwise

COLLECTING-VESSELS. 21

exercise a solvent action upon some of the constituents of the gas.
This holds good for all cases in which the collecting-vessel itself is
used as an aspirator, by "filling it with water and causing the gas to
be aspirated by the outflow of the water. If the sampling takes
place without contact with water, by pumping the gas by means of
an india-rubber pump into the dry collecting-vessel or by aspirating
it through the same by means of an aspirator, this must be con-
tinued long enough to ensure the complete expulsion of all air.

India-rubber collecting-vessels should, as a rule, be avoided,
because many gases are diffused through their walls, even if thick
or impregnated with grease. This is the case to a very consider-
able extent with sulphur dioxide and hydrogen ; whilst, for
instance, mixtures of oxygen, nitrogen, carbon dioxide, and carbon
monoxide (that is, the gases produced by combustion of fuel) can
be kept unchanged in such vessels for several hours, but never till
the next day.

Glass collecting-vessels, which are usually employed in the shape
of tubes, are only absolutely and permanently tight provided they
terminate in capillary ends which are sealed by the lamp after in-
troducing the gas. If the enclosed gas is afterwards to be trans-
ferred to a gas-burette, narrow india-rubber tubes are attached to
both ends; these are filled with water and closed by means of
glass rods or pinch -cocks, after which the sealed ends may be
broken within the india-rubber tubes by external pressure. In
most cases, however, it is sufficient to close such collecting-tubes
from the outset with india-rubber stoppers, or tubes stopped by
means of glass rods or pinch-cocks (figs. 17 and 18).

In this case the end of the tube, provided with the india-rubber
tube, is connected with the gas-burette, previously filled with
water ; the other end is made to dip into a vessel also filled with
water ; the cork &c. is opened below the water, and the water con-
tained in the burette is run off till the gas has been carried into
the burette, water taking its place in the collecting-tube.

Zinc collecting-vessels are especially employed for containing and
transporting larger volumes of gases ; and they have been found
to answer very well in all cases in which the metal does not act
upon the gas. The best form is that shown in fig. 19. The vessel
is 50 centimetres long, or 60 centimetres including the conical
ends, and has a diameter of 16 centimetres, so that it holds 10
litres of gas. Both ends have necks of 15 millimetres width, which

ON TAKING SAMPLES OF GASES.

can be tightly closed by soft india-rubber corks or by the well-
known porcelain knobs with india-rubber padding, which are
pressed down by a hinge and lever. The vessel is hung from three
thin brass chains, fitted at the top in a ring, and can thus be con-
veniently carried by hand, even when rilled with water, in order
to take a sample of gas in the proper place. If the outflow is to
be slow or capable of being regulated, the solid corks are replaced

Fig. 19.

Fig. 18

Fig. 17.

by others fitted with glass tubes and provided with screw pinch-
cocks. Such vessels are employed in large numbers for taking
samples of pit gases in the Saxon coal-pits, and sending them for
analysis to the laboratory of the Freiberg Mining Academy.

ON THE MEASUREMENT OF GASES. 23

CHAPTER II.
ON THE MEASUREMENT OF GASES.

General Remarks. Corrections.

THE volume of a gas can be found directly or indirectly. It is
estimated, either

1st, volumetrically ;
2nd, by titration ; or
3rd, gravimetrically.

The quantity found is in all cases expressed in per cent, by
volume.

Every gas has the tendency to expand and fill any space offered
to it ; it possesses a definite expanding power, which, in the state
of rest, is exerted as a permanent pressure, the amount of which
is called the tension or elastic force of the gas. All gases under
the same (ordinary) circumstances possess the same tension ; they
all are subjected to the same law as to their expansion and
contraction.

The tension, and therewith the volume, of gases depends upon —

1st, the pressure ;
2nd, the temperature ;
3rd, the state of moisture.

We measure gases in their condition at the time at which the
measurement is made — that is, at the atmospheric pressure as in-
dicated by the barometer and at the temperature as indicated by the
thermometer; and finally, since we work with water as the confining
liquid, always in a state of complete saturation with moisture.
Hence the conditions under which gases are measured may be
very different, and may vary during the analysis even from one

24 O.V THE MEASUREMENT OF GASES.

observation to another. Every such change, unless duly taken
into account, might cause very considerable errors. Hence it is
indispensable, in many cases, to make a correction, consisting in
reducing the volume of gas, which is observed in varying but
known conditions, to that volume which it would possess at the
normal barometric pressure of 760 millimetres, at the normal
temperature of 0° C., and in the dry state. By general consent
this is regarded as the normal state of a gas.

The reduction of the volume of a gas to the normal state is made
by aid of a formula derived from the following observations : —

1. Pressure. — According to Boyle's law, the volume of a gas
is in inverse ratio to the pressure upon it. Hence, if

V0= the volume at normal pressure sought,

V = the volume at the barometric pressure B,

B = the state of the barometer at the time of the observation,

we shall have

V -

• 760'

2. Temperature. — The expansion by heat of a gas is ^f 3- of its
volume at 0° for each degree Centigrade.

Hence, if a gas measures 273 cub. centims. at 0°, it will measure^
273+1 cub. centims. at 1°, and at t° 273 + t cub. centims. If,
therefore,

V0= the volume of the gas at the normal temperature,

V = the volume of the gas at the temperature t,

t = the degree of temperature at the time of observation,

we shall have

^7 273 Vx273

y __ \l r\-i* _____

273 + / 273 + T

3. Stale of Moisture. — When a gas is saturated with moisture
by contact with water, it always takes up the same quantity of
water in the same conditions. This water is itself transformed
into the gaseous state ; it therefore exerts a certain pressure, and
this pressure, the tension of aqueous vapour, increases with the tem-
perature, owing to the increased formation and expansion of that
vapour. That tension, expressed in millimetres of mercurial pres-
sure/,has been determined experimentally (compare Appendix), and
must be deducted from the observed barometric pressure (B — /).

REDUCING THE VOLUME TO NORMAL STATE. 25

From the preceding considerations we deduce the following
formula, which embraces all corrections : —

0x760

Suppose a gas, saturated with moisture, to occupy a volume of
1000 cub. centims. at 738 millims. barometric pressure and 20° C. ;
its volume in the dry state, at normal pressure and temperature,
will be

The reduction of the volumes of gases to the normal state may
be omitted in analytical estimations which are rapidly performed,
as material changes of pressure and temperature are not then to be
expected ; so also in cases in which only approximately correct
results are required.

When a gas is estimated by titration or by gavimetric analysis,
its volume is found at once in the corrected state. If one of
the gaseous constituents has been estimated, say, by titration,
and another volumetrically, it may be desirable to calculate the
former for the volume which it would occupy at the then existing
barometric pressure and temperature, and in a state of saturation
with moisture.

The following formula serves for reducing the volume of a gas
from the normal state to that which it would occupy at a different
barometric pressure and temperature, and in a state of complete
saturation ivith moisture : —

If V = the volume of the gas at the barometric pressure B and

the temperature t, saturated with moisture,
V0= the volume at 760 millims. pressure, at 0° C., and in the

dry state.
we have

V0(273 + f)760
273(B-/)

The observation of the atmospheric pressure is best made by
means of Bunsen's siphon barometer (fig. 20), which is provided
with a millimetre-scale etched on each of its limbs, and is held
in a vertical position by the aid of a stand. The reading is
made with the telescope of a cathetometer (fig. 21), which is

ON THE MEASUREMENT OF GASES.

placed at a distance of 2 or 3 metres. The sum of the readings
on both limbs indicates the barometric pressure. In certain cases
it is sufficient to employ a small aneroid barometer.

Fig-. 20.

The temperature is observed by means of a small thermometer,
divided into tenths of a degree, which is loosely placed in the
shorter limb of the barometer.

An apparatus for the expeditious reduction of the volumes of
gases to the normal state without the necessity of observing the
thermometer and barometer was first proposed by U. Kreusler

APPARATUS FOR REDUCING VOLUMES.

(Ber. der deutsch. chem, Ges. 1884, xvii. p. 29) and afterwards
constructed in a more convenient shape simultaneously by the
author (Cl. Winkler, ibid. 1885, xviii. p. 2533) and the translator
(G. Lunge, Chemische Industrie, 1885, p. 163). It is shown in

Fig. 22.

fig. :22.

An iron stand with two arms carries
two perpendicular glass tubes, con-
nected at the bottom by a thick india-
rubber tube ; one of these is the
measuring-tube, the other the level-
tube. The measuring-tube, A, is en-
larged into a bulb at the top and is
closed by a small, slightly greased, and
absolutely tight glass tap *. It holds
exactly 100 c.c. from the tap to the zero
mark ; the division marked on the
cylindrical part extends from the zero
point to 5 c.c. upwards and 25 c.c.
below, so that from 95 to 125 c.c. can
be read off accurately to 0' 1 c.c. These
two extreme values would correspond
to 100 c.c. air under normal conditions,
saturated with moisture, when brought
to 800 mm. B and (ft on the one side,
or to 700 mm. B and 30° t on the other
side, and thus embrace all values oc-
curring under ordinary circumstances.
Tube A is held vertically in the lower
arm of the stand, the division being
completely in view.

The level-tube B is open at the top,
which is protected by a dust-cover.
It is held in the lower arm of the stand

and can be moved up or down by means of a screw-clamp. It
need not hold more than 30 c.c.

In order to set the apparatus once for all for permanent use, a

* Experience has shown that no ordinary tap holds tight in the long run
the means of attaining this end will be discussed later on, when describing the
•yas-volumeter. — Translator.

28 ON THE MEASUREMENT OF GASES.

few drops of water* are introduced into tube A, an approximately
sufficient quantity oE mercury is poured in, the whole is placed in
a cool room, together with a barometer and thermometer, and
after a few hours, or better the next day, the state of both the
barometer and thermometer is accurately ascertained. According
to the formula :

100 X (270+0x760
278 x(B-/)

it is calculated what volume 100 c.c. of air, assumed to be in the
normal state, would occupy under the actually existing conditions.
The tap being left open, the level-tube is raised or lowered to the
point where the mercury level indicates precisely the calculated
volume, arid the tap is now closed. The volume of air thus con-
fined increases or decreases with every external change of pressure
and temperature exactly in the same ratio as another gaseous
volume, present in the same room and intended to be measured, so
that the normal volume of the latter can be calculated by simple
proportion, after having brought the mercury in both limbs of the
apparatus to the same level and read off the volume indicated on
tube A. For if we call

V the observed volume of air in the tube at the ruling baro-

metric pressure and temperature,
V0 the same in the normal state (constantly =100),

VI the volume of the gas to be examined at the ruling pressure

and temperature,
Vo1 the same in the normal state,
we have the proportion :

V : V0 = V1 : Vo1.

Compare later on the mechanical reduction by Lunge's gas-
volumeter.

Gr. Lunge has also modified this instrument so as to yield the
reduced volume by a simple multiplying operation, and he has
described the preparation of such reduction-tubes in a fit state for

* In those cases where the gas to be measured is sure to be in the dry state,
e. (j. the nitric oxide given off in the analysis of nitrous and nitric compounds by
means of the nitrometer, the reduction instrument may be adapted to this
special use by putting in a drop of concentrated sulphuric acid, in lieu of water,
and calculating accordingly. — Translator.

GAS-BURETTES. 29

carriage to a distance (Chem. Zeit. 1888, p. 821 ; Zeitsch. f. angew.
Chemie, 1890, p. 227) *.

Another correction apparatus is the Gas-baroscope, constructed
by J. Bodlander, on the principle of gravi metrically estimating the
gas (Zeitsch. f. angew. Chemie, 1894, p. 425).

An approximate correction, for cases where no great accuracy
is required, can be at once made by ascertaining the difference
between the volume of a gas in the normal state and that which it
possesses under average local conditions of pressure and tempera-
ture. Thus the yearly average of barometric pressure at Freiberg
is 725*6 mm., the mean temperature is 7° 0. 1 c.c. of gas in the
normal state, if saturated with moisture, would under these average
conditions occupy 1 '085 c.c., and an approximate correction would
be effected by dividing the read-off volume of gas by the above
figure. But we must consider that the temperature of the labo-
ratory is usually above the annual mean, and it is preferable to
make the calculation accordingly. Allowing a mean pressure of
725*6 mm. and a temperature of 20°, the correction factor would
be 1*135. The real average of observations made in the Freiberg
laboratory corresponds to the factor 1*118 f.

I. Direct Volumetrical Estimation.

A. Measuring in Gas-burettes (Nitrometer, Ureometer,
Gas-volumeters).

For measuring small volumes of gases, from 0*1 to 100 c.c., we
employ gas-burettes of various construction. These are cylin-
drical glass tubes,, usually graduated in cubic centimetres, which
can be closed at the top and bottom by glass cocks or pinch-cocks,
or hydraulically sealed, and whose division begins or ends at the
upper tap. When the graduation is not required to extend the
entire length of the tube, the upper portion is usually enlarged
into a bulb or a wider cylinder, in order to shorten the tube, which
may be useful for practical purposes.

It is unnecessary to say that gas-burettes, as well as all other
apparatus serving for gas-analysis, must be correctly gauged and

* These modifications, as well as the original instrument, have become obsolete
by the construction of the gas-volumeter. — Translator.

f It is evident that errors up to 10 per cent, may be caused by this method,
which consequently can serve only for very rough approximations.— Translator,

30 OX THE MEASUREMENT OF GASES.

divided. This may be controlled by the usual methods, or through
the Physico-chemical Institute of Dr. Saner, Dr. Gockel & Co.,
Berlin W., Wilhelmstrasse 49.

In order to protect its contents from the disturbing influence of
the outer temperature, the measuring-tube is frequently surrounded
by a water-jacket, formed by a wide glass tube closed at top and
oottom. This may be provided with a strip of opaque glass, placed
behind the graduation, so that the marks (which are blackened in
this case) appear on a white background. In the great majority
of cases the use of a water-jacket is quite unnecessary, since the
water which serves as confining liquid causes a sufficient equaliza-
tion of temperature *.

The gas-burette may be put in communication with a second
glass vessel, the level-tube or level-bottle, containing the confining
liquid (i. e. water), and sometimes the absorbing liquid. This liquid
serves either for confining the gas or transporting it into special
absorption-vessels, as well as for regulating the pressure, which
must be the same at each reading. The readings are usually made
at the pressure of the atmosphere, sometimes adding the pressure
of a given, and always equal, column of water.

Pure water is the best confining liquid. The advantages fre-
quently sought to be obtained by employing saline solutions,
petroleum, glycerine, or oils, are entirely illusory ; for gases,
which are relatively easily absorbed by water, are also taken up by
those other liquids to such an extent that the diminution of errors
obtained in this \vay is out of proportion to the inconveniences
incurred. Gniewosz and Walficz (Zsch. f. physik. Chem. i. p. 70)
have shown that the absorption coefficient of petroleum for oxygen
and other gases greatly exceeds that of water, although Fajans
(Chern. Zeit. 1893, p. 1002) held the contrary. The above object is
attained much more conveniently and simply by taking a definite
quantity of the gas confined in the dry state, and estimating there-
from the constituents which are easily soluble in water, whilst
only the portion which is not absorbed is received into a burette
filled with water.

There are additional reasons for filling the gas-burette — that is,

* This, of course, holds good to a much greater extent as regards the mercury
contents of the nitrometer and gas-volumeter, wherefore a water-jacket, which
would greatly hamper the manipulation of these instruments, is altogether
unnecessary in this case. — Translator.

WATER AS CONFINING LIQUID. 31

the measuring-vessel proper — with nothing but water if possible.
To introduce the absorbing liquids into the burette, as is done in
the older kinds of apparatus, involves errors, as those liquids
(e.g. solution of caustic potash or sulphuric acid) have a degree of
viscosity quite different from that of water ; they adhere much
more to the glass and require much longer time for running down
and collecting at the bottom.

Even in the case of pure water, the running down of the con-
fining liquid must be waited for before taking a reading. Without
this, errors amounting to ^ per cent, and upwards may be made.
Although the state of the surface of the glass considerably influ-
ences the degree of adhesion, still the running together of the
liquid in well-cleaned burettes takes place with sufficient regularity.
The time required for this depends, of course, on the length of path
which the liquid has to flow down in the burette. If, say, there is
only 10 c.c. of gas in the burette, the level of the liquid will be
constant in half a minute; but if there is 100 c.c. present it will
take five or six minutes before all the water has run down. In
very accurate analyses, or when estimating a very slight amount
of gas from a mixture by absorptiometrical methods, this circum-
stance must be taken into account; but generally it is sufficient
to wait a couple of minutes before each reading, and to keep the
gas in the meanwhile at a slight underpressure, before the levels
are equalized and the reading is taken. The error will then rarely
exceed 0*1 c.c. It is indispensable that the inner surface of the
burette should be clean, especially free from greasy matters, which
is secured by rinsing it with caustic potash solution, or preferably
with alcohol.

The reading itself is taken at the lower concavity of the meniscus
of the liquid (fig. 23), where the coincidence with one of the
marks of the graduation is clearly recognized. Exact
readings are taken by means of a magnifying-glass, or
preferably, with great precision and certainty, through
the telescope of a cathetometer (fig. 21, p. 26), such as
serves for barometrical and thermometrical observations.
This telescope slides up and down a triangular brass
column j and can be easily adjusted in any place by a
rack and pinion. The observations with it are best made
from a distance of 2 or 3 metres.

We shall here describe several apparatus which, .although

ON THE MEASUREMENT OF GASES.

Fig. 24.

Fig. 2-5.

Fiir. 27.

not primarily constructed for gas-analysis proper, can be equally
well applied for this purpose, and which in any case are closely
related to our subject, as they serve for the rapid estimation of

NITROMETER.

many substances by the measurement of the gaseous products
of: their decomposition. These are G. Lunge's Nitrometer, Urea-
meter 3 and Gas-volumeter *.

The Nitrometer in its original shape f, which is also that best
adapted for gas-analysis, is shown in fig. 24 (p. ,32). A is the
" measuring- tube/7 a kind of gas-burette, fitted at the top with
a three-way cock. These cocks, originally constructed, after many
laborious attempts, by Cl. Winkler and therefore quite erroneously
designated as " Geissler taps," are shown separately in figs. 25 to
30. Figs. 25 to 27 show the original Winkler (or Geissler) shape,
where the tap possesses an axial bore, curving sideways so as to
issue at right angles with an ordinary cross bore. Another shape
of three-way tap has been constructed by Messrs. Greiner &
Friedrichs, of Stiitzerbach, and is shown in figs. 28 to 30. These

Fio-. 28. Fio-. 29.

Fig. 30.

taps, in lieu of the axial and cross bores, possess two slanting
bores, and are more easily manipulated and kept tight than the

* The Translator does not in this case, as will be readily excused, simply
render the German original into English, but describes the various methods
in his own way.

f Ber. d. deutsch. chem. Ges. 1878, xi. p. 434 ; Chem. Ind. 1886, p. 273.

34 ON THE MEASUREMENT OF GASES.

former construction, wherefore they are preferred by Lunge for
his instruments (comp. Zsch. f. analyt. Ch. 1887, p. 49, and Ber,
d. deutsch. chem. Ges. 1888, xxi. p. 376).

The tap is surmounted by a cylindrical funnel or cup, visible in
fig. 24. Lest .the plug of the tap should be thrown out when
shaking the instrument, it may be fixed to the narrow part of
the funnel by a loop of fine iron or platinum wire (not copper
wire, which may be acted upon by the mercury unavoidably coming
into contact with it).

The measuring-tube is graduated, the zero-point being the
upper end adjoining the tap ; from this the graduation is con^
tinued downwards to 50 c.c., each ^ c.c. being marked. The
tube is continued about 6 inches below the graduation, and is then
tapered off, in order to be joined, by means of a strong india^
rubber tube, to a plain . cylindrical tube B, the (t level- tube/*
Both A and B are held in strong clamps ; that belonging to A is.
preferably a strong spring-clamp, so that the measuring-tube can
be taken out and readjusted in a moment.

The nitrometer can, of course, be filled with water, or with a
solution of a salt, or glycerine, or oil ; but the latter substances are
always objectionable (comp. p. 30), and it is decidedly preferable
to fill the apparatus with mercury for nearly every use it is,
put to.

Mercury is, of course, indispensable where it has to serve as
a reagent in the process ; and we shall first of all describe the
use of the nitrometer in this instance, where the evolution of gas.
takes place within the measuring-tube A itself, and afterwards,
those cases where the gas is evolved in an outside vessel and is.
merely measured in the tube A.

The original use of the nitrometer was for testing the " nitrous
ritriol" of sulphuric-acid works and similar substances by Crum/s.
process, i. e. shaking up with strong sulphuric acid and mercury..
The same process applies to the analysis of gaseous mixtures con-
taining the oxides of nitrogen, which must be first absorbed by
strong sulphuric acid and then submitted to analysis in the-
nitrometer, where the whole of their nitrogen is liberated in the
state of nitric oxide, and can thus be accurately estimated. This,
process is also very well adapted for the quick and accurate
analysis of solid and liquid compounds of nitrous and nitric acid,,
as the nitrates and nitrites of soda and potash (in the case of the

NITROMETER. 35

nitrites the whole of the nitrogen, including the nitrate, is indi-
cated), pyroxyline, nitro-gly cerine , dynamite, &c. The last-named
substances are dissolved in water, or, where this does not act, in
sulphuric acid in the cup of the nitrometer itself.

The acid contained in the nitrometer should never be diluted to
a greater extent than 2 parts of aqueous liquid to 3 parts of
strongest sulphuric acid. The assertion which has been made
that strong sulphuric acid must He diluted up to this point, or
even further, because it keeps nitric oxide in solution, is quite
erroneous.'

The manipulation of this apparatus is as follows : — It is filled
with mercury so far that, on raising the tube B, and keeping the
tap in the position shown in fig. 25 or 28, the mercury stands right in
the tap of the tube A, and about 2 inches up in the tube B. The
tap is now closed (fig. 30) so that its vents do not communicate
either with the inside or the outside of the tube A (comp. fig. 30) , and
a certain volume of nitrous vitriol (from 0*5 to 5 c.c., according to
strength) is poured into the cup ; the tube B is lowered, and the
tap cautiously opened so as to assume the position seen in fig. 25
or 28, and shut briskly when all the acid has run out except a
small drop, but no air has as yet entered. The cup is now
rinsed by pouring about 3 c.c. of strong pure sulphuric acid
into it ; this is drawn into the tube A, and this rinsing
repeated with another 2 or 3 c.c. of pure acid, always avoiding
the entrance of the smallest bubble of air into the tube A.
The tube A is nOw taken out of the clamp and the evolution of
gas started by inclining it several times almost to the horizontal,
and suddenly righting it again, so that the mercury and the acid
are well mixed and shaken for one or two minutes till no more
gas is given off. The tubes are so placed that the mercury in B
is as much higher than that in A as is required for balancing the
acid in A ; this will take 1 millim. of mercury for 6J millims. of
acid/ After the gas has assumed the temperature of the room and
all froth has subsided, which will take about 10 to 15 minutes, the
volume of the gas is read off, and also a thermometer hung up
close by and a barometer, or else the volume of air in the reducing-
apparatus (fig. 22, p. 27) or that to be described below. In order
to check the level, open the tap, when the level of the mercury in
A should not change. If the mercury rises, too much pressure has
been given, and the reading must be increased a little, say O'l c.c.
If it sinks, the contrary takes place, that is, always in the

36 ON THE MEASUREMENT OF GASES.

opposite sense to the change of level. Another plan is, to put a
little acid into the cup before opening the tap. This will be
drawn in if the pressure is too low,, or raised if it is too high.
With adroit manipulation the experiment can then soon be
corrected. After finishing it, lower the graduated tube A, lest on
opening the tap any air should enter ; open the tap, raise the tube
B, thus forcing the gas and all acid into the cup, and put the tap
in the position seen in fig. 26, so that the acid ilows out and into
a vessel placed below ; the last of it is drawn out by blotting-
paper. With the Greiner-Friedrichs tap the acid is not forced
back into the cup, but straight out of the tube by means of the
position shown in fig. 29. The nitrometer is then ready for the
next experiment. A test must always be made to ascertain
whether the glass tap is air-tight ; it will hardly remain so
without greasing it occasionally with vaseline, care being taken
that no grease gets into the bore.

This process is not interfered with by the presence of chlorides
or of a small quantity of organic substance, but it is by sulphurous
acid, the best test for which is its smell. To remove it, the acid
is stirred up with a very slight quantity of powdered potassium
permanganate ; any great excess of this makes the process very
troublesome and inaccurate. Each c.c. of gas, reduced to 0° and
760 millims., is equal to 0!627 mg. N, or 1*343 mg. NO, or 1-701
mg. N2O3, or 2*820 mg. NO3H, or 3'805 mg. NaNO3.

Where it is desirable to liberate and measure a larger volume of
nitric oxide than is practicable in the ordinary nitrometer, another
form of this apparatus can be employed. This is shown in fig. 31.
In this case the measuring-tube is provided with a strong bulb,
holding nearly 100 c.c. ; the cylindrical part below this is gra-
duated from 100 to 140 c.c. The level-tube is provided with a
similar bulb at the bottom, to receive the mercury forced out of
the measuring-tube ; it may also be provided with a graduation
similar to that of the latter, for the purpose of facilitating the
level of the liquids.

The manipulation with nitrate of soda or similar substances,
containing a large proportion of nitrogen acids, is as follows : —
Such a quantity of substance is taken that it will in any case give
off more than 100 c.c., but less than 140 c.c., of nitric oxide at
the existing temperature and pressure. In the case of commercial
nitrate of soda, for instance, it will amount to about 0'35 gramme.

NITROMETER.

Put the sample, finely ground if solid, into the tube up to the
mark, cork the tube, weigh it, pour the con-
tents into the cup of the nitrometer, taking
care that the substance settles as much as possible
upon the bottom of the cup, and re-weigh the
small tube. The three-way cock nfust have
been made to communicate neither above, nor
below, nor sideways. In the case of solid
nitre and the like, about O5 c.c. water is run in,
and when the nitre is nearly or quite dissolved
the solution is drawn into the measuring-tube
by cautiously opening the tap, the levelling-
tube being lowered, the cup is washed with,
at most, 0'5 c.c. water, and 15 c.c. con-
centrated pure sulphuric acid run in. The
operation is in other respects performed as
described above.

The nitrometer should first be tested as to
whether it really contains exactly 100 c.c. at
the mark ICO ; for instance, by inverting it, filling in mercury to
the mark 100, running it off, and weighing. It should weigh 1396
grammes, reduced to 0°. If there is a difference, this must be
allowed for in every reading.

As an apparatus for gas-analysis proper it is, in most cases, best
to employ the nitrometer fig. 24, p. 32. It is quite evident that
it will fulfil all the functions of Hempel's gas-burette, by attach-
ing to the side-opening of the three-way cock the various pipettes
described later on, or similar pipettes on a smaller scale, partially
filled with mercury, as described by Lunge (Berichte der deutschen
chemischen Gesellschaft, vol. xiv. pp. 21, 92) and by Hempel in
more recent publications. The nitrometer enjoys a great advan-
tage over HempePs burette in being filled with mercury, by
admitting of more accurate readings of level, and by being adapted
to the examination of gases partially soluble in water. In some
cases it may serve, like Bunted gas-burette (comp. later), without
any absorption-pipettes, but far more conveniently than Bunte's
burette, namely, by introducing the reagents through the cup and
the three-TV ay cock ; but this can only be done, either if only one
of the constituents has to be estimated (for instance, carbon di-
oxide), or where the reagent required for estimating a second

38 ON THE MEASUREMENT OF GASES.

constituent does not interfere with the first, as when we first
estimate carbon dioxide by means of caustic potash and sub-
sequently oxygen by pyrogallol.

The nitrometer can be very well employed for collecting, mea-
suring, and analyzing the gases dissolved in water or other liquids,
by attaching to the side-tube of the three-way cock a flask filled
with the liquid to be tested. This is connected with the cock by
an india-rubber stopper, a short elbow-tube, and a short stout piece
of india-rubber tubing. The flask is filled very nearly to the top ;
when the stopper is pressed down the liquid will enter into the
tubing and fill all the space up to the tap, which is first adjusted
like fig. 26 or 30, then like fig. 27 or 29. The liquid is now heated
till the gas is expelled, and this is collected in the measuring-tube,
the level-tube being lowered as much as possible, thus facilitating
the expulsion of the gases by aspiration. When all the gas has been
expelled, the tap is put as in fig. 26 or 30 ; the gas, after cooling,
is measured, and is analyzed by submitting it to various absorbents,
as described above.

The nitrometer is also a very convenient apparatus for the
volumetrical analysis of a great many substances, namely, for all
cases of analytical operations in which a definite quantity of a gas
is liberated which is not soluble to a very considerable extent in the
liquid from which it is liberated, and which does not act upon
mercury. Sometimes the operation can be carried on within the
measuring-tube itself, and this is even preferable when only small
quantities of gas have to be estimated. In this case the nitrometer
is treated as described for the analysis of nitrate of soda and
similar substances, only it is not possible, of course, to use the
form of apparatus shown in fig. 31, but that shown in fig. 24
(p. 32) . The measuring-tube is filled with mercury up to the tap,
the latter is closed, the level-tube is lowered, the substance to be
tested is introduced exactly like the nitrate of soda, without allow-
ing any air to enter, the decomposing reagent is then introduced
in a similar way, and the operation is finished by agitating the
tube, levelling the mercury, and reading off the volume of gas.

Another use of the nitrometer is that where the chemical re-
action docs not take place within the measuring-tube itself, but
outside *. Two classes of instruments serve for this purpose, viz.,

* Lunge, Ber. d. deutsch. chem. Ges. 188-5, xviii. p. 2030 ; Zeitsch. f. angew.
Chemie, 1890, p. 8.

NITROMETER.

Fig. 32.

that provided with a side-flask (for aqueous liquids) and that pro-
vided with an agitating- vessel (for reactions with mercury). The
former is shown in fig. 32. The side-flasl$ or decomposition- flask
is provided with an inner tube
fused on to its bottom, or (less
conveniently, because it is more
liable to breakage) simply placed
inside the flask so as to lean
against its side in an upright
position. The flask is attached to
the lateral opening of the nitro-
meter-tap exactly as that de-
scribed above, which serves for
estimating the gases dissolved in
water. This arrangement is the
most convenient one for most
purposes ; the reaction then takes
place outside the nitrometer, and
the latter only serves for mea-
suring the gas liberated, not
directly, as in most cases the bulk
of the gas will remain within the
decomposition-flask, but by the
displacement of an equal volume
of air from the flask, tubes, &c.
In this case it is not necessary to
fill the nitrometer immediately up
to the tap, that is to the zero point;
it is possible to start with 1*0 c.c.,
or any other point below the tap,
which obviates the danger of any
mercury running over into the
decomposition-flask when care-
lessly opening the tap. It is
hardly necessary to say that the volume of air left in the nitro-
meter before the operation must be exactly read off and deducted
from the final reading. Special nitrometers are also made for
this purpose, with a tap possessing only the curved axial bore, and
not surmounted by a cup ; the division begins a short distance
below the tap, which facilitates the reading. These instrument

40 ON THE MEASUREMENT OF GASES,

serve, for instance, for the analysis of ammonium salts and the
estimation of urea in urine by means of brominated soda ; in the
latter case they are called ureometers. They may also be used for
the estimation of carbon dioxide and all the analytical operations
which can be carried out by means of such an estimation ; but this
is best done by the special means to be described below, which
avoid the error caused by the solubility of carbon dioxide in the
liquid.

By far the most important uses of this shape of nitrometer
are those where hydrogen peroxide is employed, in order to act
upon substances containing " active " oxygen which is set free
in the gaseous state and can be accurately measured by the
nitrometer. Hydrogen peroxide itself can be analyzed by acting
upon it with an excess of potassium permanganate, and the latter
by acting with an excess of hydrogen peroxide, the reaction being
in both cases :

2KMnO4 + 5H2O2 + 3H2SO4= K2SO4 + 2MnSO4 + 8H2O + 10 O.

By dividing the oxygen liberated by 2 we obtain that belonging
to the H2O2 if KMnO4 has been in excess, and vice versa. In a
similar manner manganese-ore (essentially MnO2), hypochlorites
(bleaching-powder, &c.), ferricyanides, and all other substances
reacting with H2O2 can be estimated.

The operation is performed in the following way : — The solution
containing the constituent to be estimated (say, a solution of
potassium permanganate), or the substance in the shape of a very
fine powder (say, manganese-ore), is placed in the outer space of
the decomposition-flask, together with such other reagents as may
be necessary, e. g. sulphuric acid in both just-mentioned cases.
Now a sufficient quantity of hydrogen peroxide is run into the
inner tube. Meanwhile the cork of the decomposition-flask must
have been attached to the nitrometer-tap by means of a short
stout elastic tube, which will allow the flask to hang on the
nitrometer without any special support ; comp. fig. 33. The tap
should be turned as in fig. 26 or 28. The cork is now pressed
tightly down into the flask, the tap turned as in fig. 27 or 29,
the flask inclined so that the contents of the tube mix with the
liquid outside, and shaken till no more gas is given off, which
takes about a minute. The mercury-levels are adjusted and the
volume of gas in the measuring-tube is read off. The reduction

GAS-VOLUMETER.

to normal pressure and temperature is effected as prescribed above,
p. 26 or 35, or mechanically by the gas- volumeter (see below).

The third method of using the nitrometer is that where a special
" agitating-vessel " is used for the reaction of nitrates upon
mercury in the presence of sulphuric acid ; but as this is nearly
always carried out with a gas -volumeter, we shall first proceed to
describe this instrument.

Fig. 33.

The gas- volumeter * realizes the idea of doing away with all
calculations required for reducing a volume of gas to the normal
state, by effecting this reduction by a mechanical operation, carried
oat in a minimum of time and with a maximum of accuracy.

This is brought about by combining the correction-apparatus,
fig. 22, p. 27, with a gas-burette, in such manner that both the
constant volume of gas contained in the former and the gas
contained in the burette are at one and the same time compressed
to the volume of 100 c.c. in the former, and therefore equal to
the corrected volume in the latter. This is performed by com-
bining the three tubes, A, B, C, fig. 34, as follows : — They are all

* Lunge, Berl. Berichte, 1890, xxiii. p. 440, 1892, xxv. p. 3157 ; Zeitsch. f.
angew. Ch. 1890, p. 139, 1891, p. 410, 1892, p. 077.

ON THE MEASUREMENT OF GASES.

joined by very strong elastic tubing to a three-way pipe D, and
they slide upwards and downwards in strong clips. A is the
measuring-tube or gas-burette, B the reduction-tube, C the level-
tube (all these parts are here shown in their simplest form ; they
have, however, been considerably improved in shape) . A is either

Fig-. .34.

made to hold 50 c.c., divided in Ol c.c.; or 140 or 150 c.c., the
upper 90 or 100 c.c. being formed as a bulb and the graduation
being only at 90 or 100 c.c. and reaching down to the bottom ; or
else the tube has a bulb in the middle, being graduated from 0 to
30, and again from 100 to 150, so as to admit of measuring either

GAS-VOLUMETER. 43

small or large volumes of gas without unduly lengthening the
tube. B is made exactly like tube A in the reduction-apparatus,
fig. 22, p. 27, and is filled with exactly 100 c.c. air, calculated for
760 mm. pressure and 0°, precisely as stated in that place. But
it should be added that this air must be either saturated with
moisture, by previously introducing a few drops of water, or else
completely dried, by means of a drop of concentrated sulphuric
acid. In the first case the instrument is best adapted for the
measurement of moist gases, in the second for that of dry gases.

A sufficient quantity of mercury should be previously poured in
through the level-tube C, and by means of the latter the mercury
is driven up in A till it reaches the tap, whereupon the gas to be
measured is either evolved in or carried over into A. In order to
measure it in a state reduced to normal conditions, the three tubes
are so adjusted that the mercury in B stands at 100 c.c., and
at exactly the same level as the mercury in A. It is quite evident
that, by doing this, not merely the air contained in B, but also the
gas contained in A has been compressed to the point corresponding
to 760 mm. pressure and 0° temperature. Thus the reduction of
the gas in A to normal conditions is effected without observing
the barometer or thermometer, and this is facilitated by special
constructions of clamps and stands (comp. Lunge's ' Sulphuric
Acid and Alkali/ 2nd edition, vol. ii. p. 113).

The reduction-tube, B, must of course be depended upon not
merely to hold exactly 100 c.c. air of 0° and 760 mm., but also
to keep this volume entirely unchanged, in spite of the frequent
higher or lower pressures to which it is necessarily subjected in
the course of the analytical operation. Even the best ordinary
glass tap cannot be expected to stand these changes of pressure
for any length of time, although it may and should do so for a
few hours. Therefore the inventor first recommended to shape
he upper end of B so that it could be sealed off at the lamp.
As during this operation, in case of unskilful work, an error may
be introduced, he substituted for it a special kind of tap, sealed
with mercury, as shown in fig. 35. Experience has shown that
the volume of air in a reduction-tube provided with this arrange-
ment remains unchanged for years ; and it is perfectly easy at
any time to re-open and close the reduction-tube, if it has got out
of order by some extraneous cause. The plug is greased with
vaseline, and is kept tight by the mercury at the top even against

ON THE MEASUREMENT OF GASES.

Fig. 35.

strong pressure from either side, especially when kept down on
the top by the cork, as shown in the figure.

The gas-volumeter might be used like any
ordinary nitrometer, by introducing the substance
directly into the burette A and developing the
gas within the same. This is, however, not very
convenient ; the proper use of the gas-volumeter
is either for operations carried out in a side-
flask (decomposing flask), as described p. 39 et seq.;
or, for the analysis of nitrous or nitric com-
pounds by the mercury method, by combination
with two other parts, shown at E and F, fig. 34,
p. 42.

Concerning the former case we have only to
remark that, of course, during the operation of
decomposing permanganate or manganese dioxide
or hypochlorites by hydrogen peroxide, &c., the
level-tube is so placed that no pressure is exerted
on the gas ; when the reaction ceases, the mercury
in the gas-burette and that in the level-tube is
placed exactly at the same level, the top of the
burette is closed and only then the mechanical
reduction of the gas -contained in the gas-burette is carried out as
described on p. 43.

In the latter case, the mercury in A is first driven up till it just
issues from the side-tube d, which is then closed by a short india-
rubber tube and glass rod, the tap e being also closed.

Precisely in the same way we treat the agitating-vessel E, by
raising its own level-tube F, so that the mercury is retained in
the side-tube a by means of the elastic tube and glass rod b. We
now introduce the nitrous vitriol, or solution of sodium nitrate, or
the nitroglycerine, nitrocellulose, &c. into the cup c, and transfer
it into the bulb E by cautiously lowering F, just as described on
p. 35. Of course solid nitrate has to be previously dissolved in
the cup c by means of a few drops of water, solid gun-cotton &c.
by means of strong sulphuric acid, and so forth, and this is
followed by a sufficient quantity of sulphuric acid for performing
the reaction within E. When this is over and the gas has cooled
down, it is transferred to A for the purpose of being measured.
E is placed in the position shown in the figure, so that the small

GAS-METERS. 45

tubes a and d are on the same level. The bits of glass rod are
removed and a is pushed into the small piece of elastic tubing
fixed on d until the glass tubes touch. Now level-tube C is
lowered and F is raised (as shown in the figure), tap c is opened and
so is tap e, but quite cautiously. The gas will now be transferred
from E into A ; at the moment when it has all come over, and
when the acid has entered into the bore of e, but before it has got
inside of A, tap e is closed. Now the gas in A is compressed by
raising tube C to the point where the mercury stands at 100° in B
and at the same level in A, as described on p. 43, and the final
reading is taken in A.

The readings are greatly facilitated by a brass straight-edge,
provided with a spirit-level, as described by Lunge (Berl. Ber.
1891, xxiv. p. 3948).

Fig. 35 a. Fig. 35 b.

The principle of the gas-volumeter has been also applied to the
exact estimation of carbon dioxide in carbonates, and to that of
carbon in iron and steel (Lunge & Marchlewski, Zeitschr. f. angew.
Chernie, 1891, pp. 229 & 412), but we cannot discuss this here.
We will show only the decomposition -flasks constructed for
that purpose, figs. 35 a & 35 b, which admit of heating the
contents. Fig. 35 a avoids the use of cork or india-rubber, but
is more fragile than fig. 35 b.

B. Measuring in Gas-meters.

Gas-meters serve for measuring somewhat large or indefinitely
large volumes of gas, and are only rarely used in gas-analysis.
They are mostly used in those cases in which a compound present
in minute quantity in some gas has to be estimated by absorption ;
the meter is then interposed between the absorbing-vessel and an

ON THE MEASUREMENT OF GASES.

aspirator, e. g. a water-jet pump. Hence only that portion of
the gas is measured which is not absorbed, whilst the absorbable
portion is mostly estimated either by titration or gravimetrically.

A gas-meter may also be employed for finding the volume of
the bulk of a gaseous current from which an average sample is to
be taken.

We distinguish between wet and dry gas-meters according to
whether the gas is measured with or without the aid of a con-
fining liquid. Only the former are employed in gas-analyses.

The wet or hydraulic gas-meter (figs. 36 and 37) consists of a

Fig. 30.

Fig. 37.

cylindrical sheet-iron vessel, resting horizontally on a base, filled
to a little above half its height with liquid (water or glycerine of
spec, gravity 1'14), in which moves, about a horizontal spindle, a
drum divided by diaphragms into several chambers of exactly
equal capacity. There are usually four such chambers, each of
them provided with an opening near the spindle for the entrance
of the gas, and an outlet-opening situated in the periphery of the
drum, through which the gas passes into the outer case and thence
into the service-pipes. The movement of the drum, produced by
the gas passing through, is indicated by a dial arrangement so
constructed that it registers both entire and fractional revolutions
of the drum. Since the capacity of the drum is known, the
volume of the gas passing through can be read off directly upon
the dials.

GAS-METER3. 47

In the case of the gas-meter shown in the diagrams the luting
liquid is filled in by the plug d ; the gas enters at a and escapes
through b, after having passed through the drum in the direction
indicated by an arrow. A second exit is provided by the tap c}.
which is used in case the gas is to be admitted to two sets of pipes-
at the same time.

For gas-analyses the smallest descriptions of meters, known as-
experimental gas-meters, are used, as is the practice at the gas-works
themselves for photometrical purposes. At the Berlin gas-works
these are 36 centimetres high and 33 centimetres long ; they pass,
a maximum of 500, a minimum of 10, litres of gas per hour.
Their indications may deviate from the truth by as much as
1 per cent., but the error is usually not above 0*1 per cent. Such
experimental gas-meters are not officially gauged; but the makers
never send them out if they show greater deviation than £ per
cent, on passing 200 litres of gas.

Where the same kind of work frequently recurs, consisting in
the estimation of a constituent of a gas occurring in minute
quantities, e. g. ammonia in illuminating-gas, it is preferable to
work always under the same conditions, and therefore also to
employ the same volume of gas for every estimation. In such
cases the outlet of the gas is regulated by means of a tap provided
with a micrometer-screw. But as the quantity of gas to be employed
is usually large and the time required for passing it through is
considerable, it is desirable to possess a gas-meter which is auto-
matically stopped after a certain quantity of gas has passed through.
Tieftrunk (Verh. d. Ver. z. Beford. d. Gewerbfl. 1876, xxxix.
5th appendix.) has described such an automatically stopped gas-
meter where, after the passage of 100 litres of gas, the index
uncouples a lever and thus shuts the tap.

Gas-meters are never altogether reliable ; but they give service-
able approximate figures, especially if merely the number of
revolutions is noticed, as shown by the dials, without looking for
the absolute volume of the gas passed. Such restricted, but all
the more correct, observations are made by means of gas-meters
with arbitrarily divided dials, as used in physiological laboratories,
and supplied by L. A. Riedinger, of Augsburg. These meters,
pass a maximum of 500 or 600 litres per hour. Their dial is
provided with two hands, one of which (the smaller) is fixed
to the spindle of the drum and moves along with it, indicating.

48 ON THE MEASUREMENT OF GASES.

the smaller divisions. This hand must make 100 revolutions
before the second (larger) hand has completed one. The contents
of the drum is 2'5 litres; and this volume corresponds to one
revolution of the smaller, or ^ J^ of a revolution of the larger
hand. The dial has two circular divisions. The outer circle is
•divided into 100 parts, numbered from 5 to 5 ; an entire revolution
of the large hand indicates 250 litres, j^ of it 2'5 litres. The
inner circle is divided in JD-, J^, 2 ^ the divisions being marked
by various lengths. The fifth part of the smallest division, corre-
sponding to J./50 of the inner circle, or 2 cub. centims., can be
read off with certainty.

Every gas-meter should be checked by gauging. This can be
done by passing through it varying quantities of air at a constant
temperature by means of a large aspirator provided with a
pressure-gauge, the water which runs off being collected in litre-
flasks. The volume of the water run off is equal to that of the
air employed, if the pressure-gauge indicates an equilibrium both
.at the beginning and at the end of the experiments.

II. Estimation by Titration.

Sometimes a gas can be quantitatively estimated by a reaction
which takes place on its contact with unstable absorbents, and is
manifested by the formation of a precipitate, a change of colour,
and the like. Wherever possible this estimation is made by titra-
tion, and this is best done by means of solutions standardized,
not with reference to the weight, but to the volume of the gas in
question.

A normal solution is that of which 1 c.c. corresponds to exactly
1 c.c. of the gas to be absorbed, assumed to be in the normal
state, i. e. at a pressure of 760 millims. of mercury, at 0° C., and
in the dry state. A decinormal solution is one of which 1 c.c.
corresponds to O'l c.c. of the gas. Where a gas is not estimated
directly, but by re-titration, two standard liquids are required,
which, if normal, are of course equivalent ; if from practical
reasons one or the other of these cannot well be brought to the
precisely normal state, it is sufficient to obtain an exact measure
of their mutual quantitative value.

The two following methods may be employed for estimating a
gas by titration : —

TITRATING THE ABSORBED CONSTITUENT.

Fig. 38.

A. Titrating the absorbable constituent while measuring the
total volume of the gas.

In this case the gas to be analyzed is generally measured in a
flask of known capacity, bearing a mark in its neck, to which
the caoutchouc cork which serves for closing it is pressed down.
This cork has two perforations, usually closed with pieces of glass
rod ; but they have also to receive the delivery-tubes necessary
for filling the flask, and the ends of the pipette or burette used in
titration (fig. 38). By gently loosening the glass rods in question
it is easily possible to do away with any excess of pressure in
the flask, or to allow the gas, displaced by liquids running
into the flask, to escape without any actual opening of the
vessel. If any of the constituents
of the gas confined in the flask are
to be removed by absorption, for the
purpose of being estimated, an ex-
actly measured volume of the normal
solution of the absorbent, in excess
of the necessary quantity, is intro-
duced by means of the pipette;
whilst, at the same time, an equal
volume of gas is allowed to escape
by loosening the glass-rod stopping
in the aforesaid manner. The latter
volume is, of course, deducted from
the originally employed volume of
the gas. After the gas has been
thoroughly brought into contact with
the absorbent by agitating the flask,
the excess of the absorbent is esti-
mated by re-titrating ; the difference
between the two volumes of liquid,
if normal solutions have been em-
ployed, at once indicates the volume
of the absorbed gas in the normal state.

On the same principle are founded those methods by which the
gas under examination is measured in a gas-meter, and is after-
wards passed through an absorbing-vessel charged with a measured
excess of standard absorbing-liquid.

50 0\ THE MEASUREMENT OF GASES.

B. Estimation of the absorbable constituent when the non-
absorbable residue of gas is measured.

In this case the gas under examination first passes through an
apparatus containing a known volume of titrated absorbing-liquid
(normal solution), and after that through the measuring-apparatus
which indicates the volume of the non-absorbable portion of the
gas. The sum of both amounts, that found by titration and that
measured directly, corresponds to the total volume of gas employed.

The process pursued may be either that described under A,
namely, employing a measured excess of the absorbent and re-
titrating ; or else the quantity of the absorbent is limited, but the
gas is passed through till a visible reaction, for instance a change
of colour, proves that the absorbent has been completely used up.
In the former case the titration is an indirect, in the latter a
direct one.

The volume of that part of the gas which is not absorbed is found
by a measuring-apparatus attached to the absorption-vessel, and
either connected with an aspirating arrangement, or itself acting
as such. According to the bulk of the volume of gas to be
measured, and to the accuracy to be attained, we employ as a
measuring-apparatus either a gas-meter, or a water-aspirator, or an
india-rubber pump, which pumps at each stroke approximately
equal volumes of gas. If the estimation of the absorbable portion
is effected by retitration (that is, if a known excess of the absorbent,
is employed), the experiment may be continued till the non-
absorbable portion has reached a definite volume : this can be
measured either by a gas-meter which shuts oft1 automatically, or
by £i aspirator filled with a known quantity of water, to be run
off Cv_ ipletely. In that case the non-absorbable portion of the
gas is a constant, the absorbable portion a variable magnitude.

If the titration is to be direct, the volume of the ab-
sorbable gas is given by the volume of the normal solution
employed, whilst that of the unabsorbed portion is variable, and
is found by a gas-meter, by the number of strokes of a pump, or
by collecting in a graduated cylinder the water that runs out of
an aspirator.

In the processes mentioned under A and B, the absorbed gas
is measured in the normal state, but that which is not absorbed is
taken at the then existing pressure and temperature of the atmo^

ESTIMATION OF SPECIFIC GRAVITY. 51

sphere,, and in the moist state. If the result is to be correct both
volumes must be reduced to like conditions, but it is immaterial
whether the unconnected or the corrected volumes are chosen for
the purpose.

III. Gravimetrical Estimations.

A. Gravimetrical Analysis.

Finding the volume of a gas by estimating its weight presupposes
its previous absorption and transformation into a solid or liquid
compound, capable of being weighed. This kind of estimation is
but rarely employed ; principally for treating gaseous constituents
present in minute quantities. These are absorbed, and the volume
of the gases measured exactly as described under Sect. II., A and B,
for estimating gases by titration ; where it is not sufficient to
determine the increase of weight of the absorbent, the compound
absorbed must be transformed into an insoluble precipitate in
order to be ultimately weighed.

B. Estimation of Specific Gravity.

In many cases the specific gravity of gaseous mixtures admits of
drawing a conclusion as to their composition. In the manufacture
of illuminating-gas, for instance, where very different products are
formed in various stages of the process, that estimation is
universally performed. It can also be made available for judging
the quality of furnace-gases, of pyrites-kiln gases, and similar
cases. In technical examination the following two methods are
principally employed : —

a. Estimation of tlie specific gravity of a gas by measuring its velocity
wlitn issuing from an orifice.

The weight of two gases which under equal conditions issue
from an orifice is approximately in the same proportion as the
squares of the time of the outflow. If a gas of specific gravity s
possesses the outflowing-tirne t and another of specific gravity s}
the outflow ing- time tx, the relation between the outflow in ^-time
and the specific gravities is

*! ?L
* ~ t* *

E 2

ON THE MEASUREMENT OF GASES

If the unit of comparison is atmospheric air with the specific
gravity s = l, the specific gravity of the other gas is found
by the formula 2

8, = ,f.

This principle was first employed by Bunsen (Gasometrische
Methoden/ 2nd edition, p. 184)
for estimating the specific Fig. 39.

gravity of gases. H. N. Schil-
ling (Handb. d. Steinkohleugas-
Beleuchtung, 3rd edition, p. 100)
has since constructed a con-
venient apparatus for the esti-
mation of the specific gravity, in
the first instance of coal-gas, but
equally applicable to all other
gases or gaseous mixtures spa-
ringly soluble in water.

Schilling's apparatus, as shown
in fig. 89, consists of a cylindrical
glass tube, A, 40 mm. wide in-
side, and 450 mm. long. Its
upper end is cemented into a
brass cover through which passes
the inlet-pipe a, and which in its
centre carries the outlet-pipe b. A
thermometer also passes through
the cover. The inlet-pipe a is a
brass tube, 3 mm. wide, turning
outside in a right angle and pro-
vided with a stopcock ; it is
connected with the source of
the gas by an elastic tube.
The outlet-pipe b is 12 mm.
wide, and is closed at the top by
a piece of platinum foil. In the centre of this foil a small
orifice is made by means of a very fine needle, and is afterwards
hammered out ; and this forms the orifice for the issue of the gas.
Tube b has a tap, shutting off the connection between the
cylinder and the orifice. B B is a wider cylinder (125 mm. wide),

filled with so much water that this reaches nearly to the top when
the inner cylinder, filled with air or gas, is immersed in it. This
height of water is shown by a mark in the glass. The inner
cylinder has two marks, c and C1} running all round, 300 mm.
distant from each other, while Ci is 60 mm. distant from the
bottom of the cylinder A. This cylinder is open at the bottom
and is provided with a metal foot serving as guide.

In order to determine the outflowing-time, and thereby the
specific gravity, of a gas by means of this apparatus, we must first
know what time is required for a volume of air confined in
cylinder A between marks c and GI to issue through the orifice of
the platinum-foil. Fill cylinder B with water up to the mark
and then introduce A3 which is open at the bottom and filled with
atmospheric air, in a vertical position, until the water reaches a
little below mark c±. Now the top of b is opened and sufficient
air is allowed to escape through the upper orifice, until the water
in A stands exactly at mark c±. At this moment a seconds-watch
or seconds-pendulum is observed, and the air is allowed to issue
from b until the water has reached the upper mark c, which will
take place in about 4 minutes. The requisite time is exactly
observed, and the number of seconds noted.

The procedure is exactly as described if it is desired to determine
the specific gravity of the gas to be examined. Cylinder A is filled
with the gas through tap a, after having temporarily lifted it up
in the confining water ; it is emptied again through tap b, and this
filling and emptying is several times repeated until all the air
previously contained in the cylinder is completely removed. Then
it is adjusted to mark GI and the gas is allowed to issue as before
through the orifice in the platinum-foil until the water has risen
to mark c, noting again the number of seconds required for that
purpose.

Suppose we have worked first with air, then with carbon dioxide,
and we have found the time of outflow —

With air = 285 seconds (/),

„ carbon dioxide = 360 „ (^i) .

According to the above formula there is :

tf 129600
Si = "f = 81225" =1"y(>-

54 ON THE MEASUREMENT OF GASES.

This is the specific gravity of pure carbon dioxide : hence the gas
employed contained 100 per cent. CO2.

Suppose we have to examine in the same manner a mixture of
air and carbon dioxide. The outflowing-times are :

Air ..................... 285 seconds (/),

Gaseous mixture ...... 305 „ (ti) .

_/!2_ 93025
Sl ~ ^ ~ 81225 '

If we call d the difference between the specific gravity of carbon
dioxide and air, d^ that between the specific gravities of the
gaseous mixture examined and air, the expression

must give the percentage of carbon dioxide in the gaseous mixture,

or n our ca.se :

(M45-1-000) 100

= T596-Too(r~ = 2i'3 per cent- c°2 ^ volume-

b. Estimation of the specific gravity of a gas by direct weighing ivith
employment of a gas-balance (densimetric method of gas-analysis).

The direct weighing of gases is performed by the gas-balance of
Friedrich Lux, of Ludwigshafen-am-Rhein (fig. 40). It possesses
a beam, one side of which is formed by a gas-pipe which admits of
filling, through the axis, a glass ball of 2 litres capacity surround-
ing the pipe. The other branch of the beam is finished off as an
index, with balance- weight, and pointing to a scale. The gas to
be weighed can be continuously passed into the glass ball by an
elastic tube and continuously extracted in the same way. When
filling the ball with air, it takes a certain position, which, by
moving the balance-weight, can be made to coincide with point 1
of the scale. If a heavier gas enters the ball, this side of the beam
will become heavier and sink; if a lighter gas enters, this side of
the beam will rise. The index must always make the inverse
motion, and the difference of weight which has been produced
can be read off directly on the scale.

Lux's gas-balance was, like Schilling's apparatus, destined in
the first instance for controlling the specific gravity of illumi-

LUX S GAS-BALANCE.

nating-gas, and practically serves almost exclusively for this
purpose. There is, however, no reason why it should not be
employed for ascertaining the quality of other gaseous mixtures,
as fuel-gases, roasting-gases, lime-kiln gases, and thus controlling
the state of work. In all these cases the gases should be con-
tinuously passed through the glass ball, and as the proper condi-
tions of work are at the outset known to correspond to a certain
state of the index, any greater deviations from these conditions
would be at once observed.

Fig. 40.

This principle has been utilized, especially for the control of
furnace fires, by special instruments called " dasymeters " or
*' econometers " (for instance, Arndt's econometer, sold by Jos.
Wilkes, 41 Holzmarkt, Cologne). In this case the gases must
be first deprived of soot and moisture and cooled down to the
ordinary temperature before entering into the apparatus, which is
usually graduated so as to show directly the percentage of CO3
in the gases. This system offers, in fact, a very good check upon
the way in which the stoker serves the fire, and is frequently
employed in factories for that purpose.

Lux has also proposed a densimetric method of gas-analysis
He first finds the specific gravity of a gaseous mixture (^),
then removes a certain constituent (s2) by a suitable absorbent,
arid ascertains the specific gravity of the residual gas (s3) by a

56 ON THE MEASUREMENT OF GASES.

second gas-balance. In this case the proportion of the constituent
removed by absorption is : —

This might be followed by absorbing a second constituent
and examining the residual gas by a third gas-balance, and so on.
This method does not seem to have acquired much practical
application .

IV. Arrangement and Fittings of the Laboratory.

A person who has to carry out gas-analysis for technical pur-
poses is required, in very many cases, to work in anything but a
properly fitted-up laboratory. He may be compelled to put up his
apparatus and take samples of gases in the most various places —
at furnaces, flues, and chimneys, in open yards, in the field, or even
below ground, and, if possible, to perform the analyses in these
same places. It is evident that, in working at such temporary
stations, the accuracy of the results may be seriously impaired by
unfavourable circumstances, since it is sometimes quite impossible
to avoid disturbing influences.

It is different in the laboratory. Here all arrangements can
and must be provided which make it possible to work quickly and
conveniently, as well as accurately, and this should apply even to
temporary conditions of working, at least to a certain extent.

The laboratory should be a room exposed in the least possible
degree to variations of temperature. Its walls ought to be thick
and not too much exposed to the sun ; the windows should give
a good light and be as nearly as practicable turned towards the
.north. If the room must be heated, this is best done by a stove
(like the American stoves), which is lighted the night before and
then kept continually burning, so that the room and the objects
contained in it are equally warmed throughout, and the action of
radiant heat during the day is as much as possible avoided.
Hempel proposes to heat the room by means of a sheet-iron pipe,
placed on the floor in the middle of the room, which rises at one
end near the wall and passes out near the ceiling. Two gas-
burners suffice for heating the opposite ends, and a small gas-jet in
the vertical part of the pipe serves for aspirating air through it.
That portion of the pipe which is less than hand-warm is preferably

FITTINGS OF THE LABORATORY.

Fig. 41.

made of paste-board, which resists the chemical action of the
gases better than sheet-iron. This suffices for a room of 2000
cubic feet capacity.

The temperature of the apparatus, the reagents, the water, and
the absorbing-liquids must be the same as that of the laboratory ;
they should therefore be kept in the laboratory, and not in a
different room. In order
to ensure a supply of water
of equable temperature for
confining the gases, brack-
ets are fixed against the
wall at suitable distances,
about 5 feet above the
working - bench, on which
are glass bottles or ja-
panned tin vessels kept con-
stantly full of distilled or
pure well-water. They are
provided at the bottom with
a neck, fitted with an india-
rubber stopper with a some-
what wide glass tube bent
downwards at a right angle,
to which is attached an india-
rubber tube banging down
upon the working- bench,
with a glass end and a strong
pinch-cock. When out of
use the glass end is placed
in a small glass cup, attached
to the wall so as to avoid any
swaging of the tube and any
dropping of water.

The iv or king -benches are
provided with drawers for
keeping the requisite stores
of india-rubber, glass, capil-
lary and combustion tubing
of different bores, of T-

58 ON THE MEASUREMENT OF GASES.

pieces and junctions, pinch-cocks, test-papers, and so forth. Some
small tables are surrounded with a bevelled wooden ledge ; they are
made to shelve towards the centre, and a circular hole is made in
the deepest place, in which a glass funnel and down-pipe are fixed.
The top of the table is afterwards covered with thin sheet-lead ;
above the funnel a hole is cut in the lead, and the latter is evenly
laid against the inner side of the funnel; it is also turned over the
wooden ledge. Such tables serve for working with corrosive
liquids, which, in this case, may be run off straight away, and
entirely removed by rinsing the table with water.

The laboratory must be provided with a water-supply for filling
the stock-vessels, gas-holders, and aspirators, and for feeding the
water-jet pumps, as well as with a sink for cleaning the apparatus.
There should also be a gas-pipe, with branches and taps of
different bores at suitable places. The smaller taps supply the gas
for heating the combustion-capillaries, the larger ones for the
combustion-furnace. In the laboratory of the Freiberg Mining
Academy there is also a pipe-service for the gaseous mixtures
intended for the students' practice; these mixtures are kept in
large gas-holders of 150 litres capacity, and are conducted to the
various working-places. There is likewise a pipe for supplying
compressed air for use in the combustion- and absorption-
apparatus. The air of the laboratory itself sometimes contains
a sensible quantity of illuminating-gas *, and should not be used
for rinsing out the apparatus.

The barometer, thermometer, correction-apparatus, and catheto-
meter must also be suitably placed ; the last preferably on a
brick pillar.

Besides measuring- and absorbing-apparatus of various construc-
tions, stock-vessels for reagents, absorbing-liquids, and standard
solutions, there must be a sufficient selection of burettes, pipettes,
litre-flasks, graduated cylinders, and generally all the apparatus
required for volumetric analyses. Standard solutions, which are
frequently used or which easily undergo change, are best supplied
from a burette with feeding arrangements, fixed with its stock-
bottle in an assigned place. The diagram (fig. 41) will make this
arrangement clear without further description.

There should also be a galvanic battery or other source of
electricity (e.g., a storage-battery) and an induction-coil.
* This surely ought not to happen ! — Translator.

SOLID AND LIQUID ADMIXTURES. 59

CHAPTER III.

APPARATUS AND METHODS FOR CARRYING
OUT THE ANALYSIS OF GASES.

I. ESTIMATION OF SOLID AND LIQUID ADMIXTURES.

THE gases to be examined,, especially those which occur in
the practice of factories, do not always consist of gases alone.
They frequently contain solid or liquid substances mechanically
carried along, which can be retained by rest, nitration, or washing.
The liquid admixtures are always accompanied by vapours of the
same substance.

Although in many cases the presence of such substances in a
gas does not sensibly influence its volume, and therewith the
results of gasvolumetric analysis, it may be desirable to remove
them and at the same time to estimate their quantity. Both
of these functions are usually performed at the same time as
taking the samples of the gas, of course followed by measuring
the gas drawn off. If, as is frequently the case, it is necessary to
employ a comparatively large volume of gas merely for the purpose
here mentioned this is measured by a gas-meter, or else by
means of an aspirator based on the outflowing of water. The
meter or aspirator always forms the final portion of the analytical
apparatus.

Solid admixtures frequently, e. g. in smoke or furnace-gases,
consist merely of soot ; in other cases of dust of extremely varying
composition — e. g., minerals, metals, colouring-matters, fibres,
flour, coal, gunpowder. In the flue-dust from metallurgical
operations there are found the oxides, sulphides, sulphates,
chlorides, sometimes the iodides of various metals.

The quantity of dust contained in a gas may vary between great

60 ON APPARATUS AND METHODS.

limits. Thus Fodor found in the street-air of Budapest, 15 feet
above the street level : —

in winter 0*00024 gram dust per cubic metre;

„ spring 0*00035 „ „ „

„ summer 0*00055 „ ,, ,,

„ autumn 0*C0043 „ „ „

The air of Paris, according to Tissandier, contained after a
week's dry weather 0*0230, after heavy rain 0'0060 gram
dust per cubic metre.

Hesse found per cubic metre air from a living-room and nursery
0*0016 gram, from the rag-picking shop of a paperworks 0*0229
gram, from the cleaning-room of a foundry 0*1000 gram dust.

Stapff found in 1 cubic metre air from the St. Gothard Tunnel,
during the time it wras constructed, from 0*375 to 0*873 gram
dust. Stockmann found in 1 cubic metre blast-furnace gas 1*900
grams ; Theisen, in the same, before washing, 3*340 grams, after
washing 0*010 gram dust. Scheurer-Kestner found in chimney-
gases from a coal fire, when strongly firing 0*22C9 gram, when
damping the fire 0*9649 gram carbon as soot. Krause found in
1 cubic metre air of a phosphorus-matches factory, 0*004 to
0*005 gram phosphorus. F. Blum has published a pamphlet
(Frankfurt-a.-M., 1900) on the causes of lead intoxication, in
which a method for estimating the lead dust contained in the air
is described (p. 9).

Sometimes very large quantities of air have to be employed for
estimating the dust, especially wrheie not merely its quantity has to
be ascertained, but also sufficient material must be collected for
a microscopical and chemical examination, for the purpose of
determining the hygienic properties, or the value or the inflam-
mability of the dust. The latter, as is well known, exerts great
influence in the case of explosions in coal-pits and flour-mills.

The retention of solid substances mixed with a gas in the shape of
dust is performed by filtration. Even very small particles, such
as occur in smoke, of 0*OC02 to 0*C03 mm. diameter* can be retained
by means of a suitable filtering-medium, with sufficient filtering-
surface and not too rapid a current of gas. Carded cotton- wool
is very efficient : where this is not possible to use, as with acid
gases, we may take gun-cotton or soft, curly glass-wool. This
material is placed in an ordinary calcium- chloride tube, and is

SOLID AND LIQUID ADMIXTURES. 61

dried by exposing the tube in an air- or water-box at 100° to a
current of dry air, until the weight remains constant. This tube
is interposed between the place where the gas is withdrawn and
the aspirator or meter, a suitable volume of gas (up to 1 cubic
metre per 24 hours) is drawn through it more or less quickly, the
tube is dried again at 100°, and the increase of weight ascertained.
If the collected dust, which is principally found at the entrance,
is to be chemically examined, this can be done by the ordinary
analytical methods. O. Brunck describes filtering-tabes for
collecting the dust from coal-pits exposed to fire-damp, which are
provided with ground-on glass caps, and admit of weighing the
dust with its natural moisture. They are carried about in boxes
lined with cork slabs, of such size that the caps cannot fall
off.

In order to ascertain the quantity of soot in furnace-gases
(chimney-smoke), a known volume of smoke is drawn through a
tube of refractory glass, containing an asbestos layer, 20 cm. long.
The soot is afterwards burned in a current of oxygen, and the
carbon dioxide absorbed in potash bulbs, as in elementary analysis,
of course interposing a calcium-chloride tube before the potash
bulbs.

P. Fritzsche (Zsch. d. Ver. deutscher Ingenieure, 1897, p. 885)
describes a colorimetric test for the smoke -density of chimney-
yases caused by soot. It is founded on the more or less pro-
nounced grey colour of a filtering-medium, consisting of cellulose
fibre, afterwards shaken up with a certain volume of water, and
comparing it with the colour of paper tinted by Indian ink.

H. Wislicenus (Zsch. f. angew. Chem. 1901, p. 689) makes an
approximate examination of forest air suspected of being con-
taminated by soot, by exposing to it frames covered with thin
calico, and comparing the degree of blackening produced after a
certain time.

In gaseous heating- and illuminating-materials we always find
naphthalene-vapours, which up to the present cannot be estimated
with any degree of accuracy. But, according to Tieftrunck, their
quantity is closely connected with that of the ammonia contained
in the gas, and both rise and fall at the same time, although in an
unknown relation. Hence the estimation of ammonia in such
.gases affords a rough idea of the naphthalene they contain.

Liquid admixtures in gases occur mostly in the shape of vapour ;

62 ON APPARATUS AND METHODS.

but if the sample has been taken in the hot state, the liquid may
be partially condensed by cooling. This condensation is never
sufficiently complete to admit of an estimation of the substance;
it should be always combined with an absorption- or washing-
process in order to ascertain the total quantity of the substance
in question.

Water is estimated by absorption in a weighed calcium-chloride
tube. If the gas contains ammonia, the best drying-agent is
that employed by Stas, and recently again recommended by
C. Frenzel (Zsch. f. Elektrochemie, 1900, p. 486), which is
obtained by heating a mixture of 3 parts of finely divided copper
and 1 part of potassium nitrate in an iron crucible to a strong
red heat.

Mercury (of which Janda found 0'00875 gram per cubic metre
in the principal chimney of the Idria quicksilver works) is found
by interposing a weighed tube filled with gold-foil, and re-weighing
after the passage of the gas.

Sulphuric acid, occurring as such or as sulphur trioxide, together
with sulphur dioxide, in roasting-kiln gases, is found by estimating
the total acids (comp, below) and subtracting the sulphur dioxide
titrated in another portion.

The estimation of carbon disulphide in air has been described
by Biehringer (Dingler's Pol. Journal, cclxxvi. p. 78), Schmitz-
Dumotit (Chem. Zeit. 1897, pp. 487 & 510), Goldberg (Zeitsch. f.
angew. Ch. 1899, p. 75), but their methods do not seem to have
found much application in gas-analysis. Nor do we know of any
practical employment of Eiloart's assertion (Chem. News, lii.
p. 184), according to which the vapour of carbon disulphide can
be removed from a gaseous current by iodine, or else by absorption
with linseed-oil and volumetrical estimation. In illuminating-gas
carbon disulphide never occurs alone, but always together with
mustard-oil, mercaptane, arid other organic sulphur compounds.
These may be either transformed altogether into the easily
absorbable hydrogen sulphide (comp. this),, or we may content
ourselves with ascertaining the total sulphur in the gas (comp.
this), including the hydrogen sulphide.

The quantity of tar contained in a gas can be retained and
weighed by interposing a glass tube filled with loose cotton-wool.
More accurate is its estimation by means of an apparatus
constructed by Tieftrunck, in which the gas is thoroughly washed

HYDROCARBONS. BENZENE. 63

with alcohol of 25 to 30 per cent, by weight. The alcoholic
solution is allowed to evaporate in a weighed vessel at the ordinary
temperature; to the weight of the residue one-third is added,
which, according to experience, corresponds to the light tar-oils
evaporating at the same time, especially benzene and toluene.
The latter, as well as other low-boiling hydrocarbons formed in
the destructive distillation of coal, are estimated by absorbing
them in fuming sulphuric acid, as will be described in the case of
ethylene.

Hempel & Dennis (Ber. d. deutsch. chem. Ges. 1891, xxiv.
p. 1162) have worked out a method for estimating the hydrocarbon
vapours in coal-gas. They first pass the gas into a gas-burette,
where it is measured, and from this into a gas-pipette filled with
mercury, say^ an explosion-pipette (comp. later on), where it is
shaken up for three minutes with 1 c.c. of absolute alcohol. The
gas, thus freed from hydrocarbon vapours, is carried back into
the burette, and from this, in order to remove the alcohol vapours,
into a second mercury pipette, where it is shaken up for three
minutes with 1 c.c. of water, after which the contraction of
volume is noted. Both alcohol and water should first be saturated
with coal-gas, Comp. on this subject, F. Fischer, Zsch. f. angew.
Ch. 1897, p. 319.

In order to estimate benzene occurring together with ethylene
and other hydrocarbons absorbed by fuming sulphuric acid,
Drehschmidt and, later on, Harbeck & Lunge (Zsch. f. anorgan.
Chem. 1898, xvi. p. 26 ; comp. also Lunge & Akunoff, ibid. 1900,
xxiv. p. 191) have tried to utilize the property of ethylene
(first observed by P. van Wilde) to combine with hydrogen in
the presence of platinum-black, by which it is transformed
into non-absorbable ethane. But as this process does not
take place in the presence of carbon monoxide (which forms
a compound with platinum), and as the gases in question nearly
always contain that substance, this method is only exceptionally
applicable.

Harbeck & Lunge (Zsch. anorg. Ch. xvi. p. 16) describe a gravi-
metric estimation of benzene, founded on transforming the benzene
vapour contained in the gas by means of a mixture of fuming nitric
acid and sulphuric acid into dinitrobenzene, which can be weighed
as such. This method requires about 10 litres of gas and a somewhat
complicated apparatus^ so that it is not very convenient for ordinary

64 ON APPARATUS AND METHODS.

use, but very suitable for controlling other methods. O. Pfeiffer
(Lunge's Cheraisch-technische Untersuchungsmethoden, vol. ii.
p. 595) has slightly simplified that method.

Haber & Oechelhauser (Ber. d. deutsch. chem. Ges. xxix.
p. 2700; comp. Haber, Journ. f. Gasbeleucht. 1900) absorb first
benzene and ethylene at the same time by means of fuming
sulphuric acid, and in another sample of the gas they titrate
ethylene by bromine vapour, benzene vapour being indifferent to
bromine vapour. Comp. infra, p. 67.

E. M tiller (Journ. f. Gasbeleucht. 1898, p. 433) absorbs benzene,
according to Bunte's proposal, by cooled paraffin-oil,, spec. grav.
0-88-0-89, boiling at 360°. The gas, dried by calcium chloride,
is passed through four absorbing- vessels, cooled with ice and
salt, placed in series and connected so that glass touches
glass*, in a slow current, say 2 c.c. per second. The absorbed
portion is found by re-weighing the absorbers, after having taken
the temperature of the room ; the non-absorbed portion is
measured in a gas-meter. This process is employed at coke-
works for the estimation of the benzene contained in the gases,
which on the large scale is recovered by a precisely similar
process.

Ferro-carbonyl, which occurs in slight quantities in water-gas
(Roscoe & Scudder found 2*40 grams Fe in 1 cb. metre of gas),
is estimated by passing a known volume of water-gas through a
refractory glass tube, heated to red heat; here metallic iron is
deposited as a dark mirror. Another portion of the iron is carried
away by the gaseous current in the shape of dust, and is retained
by a cotton-wool plug placed in the end of the tube. The iron is
then dissolved in dilute sulphuric acid and titrated. In a similar
way nickel carbonyl might be estimated.

Nitroglycerine in the shape of vesicular dust is formed during
blasting-operations with dynamite and causes headache and other
troubles. It may be absorbed from the air by alcohol and
estimated by evaporation at ordinary temperatures ; but practical
experience on this point is still lacking.

* According to my experience this precaution does not prevent the benzene
from being- partly absorbed by the india-rubber joints. — Translator.

ABSORBENTS FOR CARBON DIOXIDE. 65

IT. ESTIMATION OF GASES BY ABSORPTION.

1. Direct Gasvolumetric Estimation.
A. Absorbing-agents for Gases.

The gasvolumetric estimation of a gas by absorption is an
estimation by difference. It is performed by taking out of a
known volume of gas the absorbable gaseous constituent by means
of a suitable reagent, measuring the residual gas and subtracting
its volume from the original volume of gas.

The absorbing-agents are nearly always employed in solution,
frequently in a somewhat concentrated state, especially when they
have to be used over and over again. The continuous employment
of the same absorbing-liquid nearly up to exhaustion is to be
recommended, because even those gases which are not absorbable
by a chemical reaction are mechanically dissolved to a sensible
extent in aqueous liquids. Consequently, wrhen employing a
freshly prepared absorbing-liquid, the percentage of the absorbable
gas is found rather too high, and this error is only absent when
the liquid has been saturated with the mechanically dissolved
gases.

We quote here the following absorbents for various gases and
their preparation.

a. Absorbents for Carbon dioxide.

Carbon dioxide is easily and rapidly eliminated by a solution
of potassium hydroxide. Dissolve 250 grams good commercial
caustic potash, but not specially purified by alcohol, in water and
dilute to 1 litre. 1 c.c. of this potash liquor usually contains
about 0'21 gram real KOH and consequently absorbs 0*083 gram
= 42 c.c. CO2. The absorption is always finished in one minute,
usually in a shorter time; it is quite unnecessary to allow ten
minutes for it, as has been recommended.

For some purposes, e. g. for use in.Bunte's burette, the liquor
need not possess the above concentration. In other cases, e. g.
with Orsat's apparatus, it^may be used in a more concentrated
state ; but the higher its concentration, the greater its viscosity
and its action on the glass vessels.

A solution of caustic potash serves also for absorbing other acid
gases, as chlorine, hydrogen chloride, hydrogen sulphide, sulphur

66 ON APPARATUS AND METHODS.

dioxide, &c. Caustic soda has the same action, but acts more
strongly on glass, and is therefore less to be recommended.

b. Absorbents for heavy Hydrocarbons.

The heavy hydrocarbons to be taken into consideration in
technical analysis belong to the following classes : — Olefins, C7!H2n,
especially etliylene, C2H4, small quantities of propylene, C->H6, and
butylene, C4H8 ; then the hydrocarbons of the series C,?PI2n_2, of
which acetylene, C2H2, is the principal representative; and lastly
the benzenoid hydrocarbons, CnHj«_8, principally benzene, C6H6,
and toluene, C7H8. As regards the occurrence of these hydro-
carbons in coal-gas as the illuminating constituents proper, they
and their determination have lost much of their importance through
the introduction of the Welsbach light. They are estimated
absorptiometrically by means of the following reagents : —

1. Fuming sulphuric acid, usually of spec. grav. 1*938, and
containing about 24 per cent, free SO3. — Below 15° crystals of pyro-
sulphuric acid are formed. This agent absorbs all the heavy
hydrocarbons to be considered here, if agitated with them during
five minutes. Ethylene is in this case converted into eth ionic
acid, C2H6S2O7, acetylene into acetylenesulphuric acid, G^^SOi,
benzene into benzenesulphonic acid, CCH6SO3. J. Schroeter (Ber.
d. deutsch. chem. Ges. 1898, xxxi. 2189) states that fuming
sulphuric acid with acetylene does not form acetylenesulphuric
acid, but methionic acid, CH4S2O6 ; but this is not the case, at
least not under the conditions present in gas-analytical operations,
for this would involve the formation of carbon monoxide, which
does not take place. In contradiction to that statement, the
acetylene is completely absorbed, as confirmed by Kiiorri & Arndt
(Verh. Gewerbfleiss, 1900, p. 166).

The absorption is carried out in a simple Hempel's gas-pipette
(comp. this), of course being very careful in filling it ; it is closed
by a small glass rod, enlarged at one end, or a glass cap, in order
to prevent the attraction of moisture. These need not be taken
off during use. Hempel recommends to provide this pipette with
an additional bulb above the ordinary ones, filled with bits of
glass, in order to enlarge the absorbing surface and to render
any agitation unnecessary. After absorption the acid vapours
contained in the gas must be removed from it by means of a
potash pipette.

ABSORBENTS FOR HEAVY HYDROCARBONS. 67

Worstall (J. Amer. Chem. Soc. xxi. p. 245) has observed that
fuming sulphuric acid on prolonged contact absorbs a little methane
and ethane; but there is no sensible error caused by this,, if the
time of absorption is not extended over a quarter of an hour.

By this agent only the totality of heavy hydrocarbons present in
a gas can be estimated, but this is done with sufficient accuracy.
The attempts at separating the single members have not met
with much success. Concerning benzene, comp. supra, p. 63.
Eritzche (Zsch. f. angew. Chem. 1896, p. 456) proposes to estimate
ethylene by itself by diluting its solution in fuming sulphuric acid
with water and distilling off the ethylic alcohol formed. He also
states that butylene and ethylene can be separated by means of
sulphuric acid, spec. grav. 1-620=70 per cent.H2SO4, which dis-
solves only butylene, not ethylene.

According to E. St. Claire Deville (Journ. des Usines k Gaz,
1889, p. 13), the benzene contained in a sample of gas is to some
extent taken up by the confining water and by the potash solution
employed for absorbing the carbon dioxide, and the same holds
good of other gases and vapours (comp. p. 65). In the case of
coal-gas analyses the error thus caused is usually eliminated by
first shaking up all the liquids with coal-gas ; but it can be directly
ascertained by first estimating CO2 by absorption plus the C6H6
taken up at the same time, and in a second sample titrating the
CO2, which shows the amount of absorbed benzene-vapour by
difference.

2. Bromine water. — Saturated bromine water is diluted with
two volumes of water, so that the liquid contains about 1 per cent,
bromine. It gives off sufficient bromine vapour to impart to the
gas in contact with it a yellow colour. It is kept in a composite
Hempel's pipette, provided with a water seal. It quickly trans-
forms ethylene and its homologues into bromides, without the
necessity of agitation.- As Treadwell & Stokes (Ber. d. deutsch.
chem. Ges. 1888, xxi. p. 3131) and Haber & Oechelhauser (ibid.
1896, xxix. p. 2700) have found, the absorption is complete, but the
employment of bromine water always involves the removal of
the bromine vapours from the residual gas by a potash pipette.

Acetylene behaves like ethylene. Benzene the author has found
(Zsch. f. analyt. Chem. 1889, p. 285) to be but slowly and incom-
pletely removed. Haber & Oechelhauser (loc. cit.) ascertained
that this removal takes place not by chemical action, but

68 OX APPARATUS AND METHODS.

mechanically ; when benzene vapour and bromine vapour are in
contact during two minutes in diffused daylight, no bromine is
consumed. Hence it is impossible to carry out the gas-volumetrical
separation of benzene from the other heavy hydrocarbons by this
agent; but the quantity of benzene can be ascertained by first
absorbing the totality of heavy hydrocarbons by means of fuming
sulphuric acid, then treating a second sample of gas with bromine
water, and determining the consumption of bromine by means of
potassium iodide and sodium thiosulphate.

c. Absorbents for Oxygen.

Only a few of the numerous agents proposed for the absorptio-
metric estimation of oxygen have in the long run proved satisfactory.
Thus, e. g., chromium protochloride, proposed by von der Pfirdten
(Ann. Chem. ccxxviii. p. 112), obtained by dissolving chromium
acetate in hydrochloric acid, is certainly efficient, but more
troublesome to prepare and to employ than other agents. De
Koninck (Zsch. f. angew. Chem. 1890, p. 727) proposed an
alkaline solution of ferrous tartrate, but this is less effective than
other agents.

The following substances can be recommended, as thoroughly
tested : —

1. Phosphorus. — It is moulded into thin sticks by melting it in
a glass cylinder under warm water so as to form a layer 10 or
15 cm. deep, dipping into this a glass tube of 2 or 3 millimetres
bore, closing this at the top with the finger and quickly transferring
it into a vessel filled with cold water. As the phosphorus
solidifies, its volume shrinks so that the stick can be easily pushed
out under water, especially if the glass tube is slightly conical.
With a little practice a large number of thin phosphorus sticks are
prepared in this manner, and these are ultimately cut into smaller
pieces under water. Phosphorus can also be obtained in this shape
from the dealers in chemicals.

The phosphorus is placed in a suitable vessel, e. g. a HempePs
tubulated pipette, completely covered with water and protected
against light. The water serves as a seal ; if it is driven out by the
gas to be examined, the latter comes into contact with the moist
phosphorus and the absorption of oxygen begins at once, with
formation of white clouds of phosphorous acid, which render the
gas opaque for some time without influencing its volume. If

ABSORBENTS TOR OXYGEN. 69

the absorption takes place in a dark room, a bright light is
produced whose vanishing, as well as the clearing away of the
cloud, marks the end of the process. About two or at most three
minutes' quiet contact of the gas with moist phosphorus is ordinarily
sufficient for the absorption of oxygen. 1 gram phosphorus when
transformed into phosphorous acid takes up 0'77 gram = 538 c.c.
oxygen ; hence the stock of phosphorus contained in one of the
absorbing-vessels generally lasts for years. The water covering
the phosphorus, which is gradually transformed into a solution of
phosphorous and phosphoric acid, should be renewed from time
to time.

Certain circumstances influence the absorption of oxygen by
phosphorus, viz. :—

(a) The temperature. At 18° or 20° the absorption proceeds
satisfactorily; between 12° and 17° it is very slow and it ceases
almost entirely at 7°. Hence the phosphorus pipettes during the
cold season should be brought to a medium temperature before use.

(b) The partial pressure of oxygen. Pure oxygen at the pressure
of an atmosphere is not absorbed by phosphorus at a temperature
not exceeding 23°. The absorption begins only when the gas has
been reduced by means of the air-pump to about 75 per cent, of
the initial pressure; it may then set in with extreme violence,
up to the production of scintillations and to the melting of the
phosphorus. If, therefore, a gas rich in oxygen, e. g. commercial
compressed oxygen itself, has to be examined, the gas should be
diluted with its own volume of pure nitrogen, which may be taken
out of a phosphorus pipette filled with air.

(c) The presence of certain gases and vapours retards or even stops
in a hitherto unexplained way the oxidizing action of oxygen 011
phosphorus. Perhaps this phenomenon is of a similar kind to
the " paralyzing " action of minute quantities of hydrogen sulphide,
carbon disulphide, and other foreign substances upon the catalytic
action of platinum and of organic ferments, as observed by Bredig
and Miiller von Berneck (Zsch. f. physik. Chem. 1899, p. 324).

Among the substances interfering with the absorption of oxygen
by phosphorus are, according to Davy, Graham, and Yogel,
hydrogen phosphide, hydrogen sulphide, sulphur dioxide, carbon
disulphide, iodine, bromine, chlorine, nitrogen peroxide, ethylene,
acetylene, ether, alcohol, petroleum, oil of turpentine, eupion,
creosote, benzene, tar, and many essential oils. Ho\v strong this

70 ON APPARATUS AND METHODS.

influence may be is shown by the fact that as little as : Q^Q vol. PH3,
5-J-Q vol. C2H4, 44144 vol. oil of turpentine suffices for making
phosphorus and oxygen indifferent to each other. Hence the
application of phosphorus as an absorbent for oxygen is restricted,
and is excluded in all cases where the presence of such disturbing
substances is to be assumed with any degree of certainty. But,
according to experiments made by O. Brunek at Freiberg, these
substances can in most practically occurring cases be removed by
a previous treatment of the gas with fuming sulphuric acid, so
that this enables the analyses of illuminating-gas and that of fire-
damp to be made by that means. This proves the incorrectness
of the assumption hitherto held, according to which methane and
ethane belong to the class of interfering substances, as these arc
not removed by the treatment with fuming sulphuric acid.

The method in question renders excellent service in the
examination of air, of chimney-gas, vitriol-chamber gases, &c. ;
for, generally speaking, phosphorus is superior as. to certainty and
speed of action to every other reagent for the absorption of oxygen.
Lindemann (Zsch, f. analyt. Chem. 1879, p. 158) has greatly
facilitated the practical use of this reagent by the construction of
a special absorbing-apparatus.

(d) The presence of combustible gases. E. Baumann (Ber. d.
deutsch. chem. Ges. 1883, xvi. p. 2146) and Leeds (Chem. News,
xlviii. p. 25) state that carbon monoxide in the presence of oxygen
in contact with moist phosphorus is partially oxidized into carbon
dioxide. This has been contradicted by Rernsen & Reiser (Amer.
Chem. Journ. 1883, p. 454) ; but Baumann maintains his position
(Berichte, 1884, xvii. p. 283). Boussingault has also shown
(Compt. rend. Iviii. p. 777) that during the slow combustion of
phosphorus in gases containing oxygen a small portion of any
combustible gases present, as carbon monoxide or hydrogen,
vanishes together with the oxygen ; but this simultaneous com-
bustion is comparatively slow and, at least in technical gas-analysis,
causes no sensible error.

2. Alkaline solution of pyrogallol. — An aqueous solution of pyro-
gallol in contact with air changes but very slowly, but on addition
of an alkali it rapidly absorbs oxygen and takes first a red ?
afterwards a deep brown colour. According to Liebig (Ann.
Chem. Pharm. Ixxvii. p. 107) 1 gram pyrogallol, with addition
of potash solution, absorbs 189'8 c.c. oxygen ; according to

ABSORBENTS FOR OXYGEN. 71

Doebereiner (Gilb. Ann. Ixxii. p. 203, Ixxiv. p. 410) on addition of
ammonia 266 c.c. oxygen. The latter statement agrees with
experiments made by P. Mann in the author's laboratory, where
1 gram pyrogallol dissolved in 20 c.c. caustic potash solution of
spec. grav. 1*166 absorbed 265'2 to 278'7 c.c. (on the average
268'9 c.c.) oxygen.

This behaviour of pyrogallol was first utilized for the eudio-
metric estimation of the oxygen in atmospheric air by Chevreul
in 1820, and was further investigated by Liebig. Weyl and
Zeitler (Ann. Chem. Pharm. ccv. p. 255) showed that the
absorbing action of pyrogallol is a function of the alkalinity of
the solution, but that in too highly concentrated solutions of
potassium hydrate the absorbing power is weakened, probably by
partial decomposition of the pyrogallol. A solution of KOH of spec,
grav. T05 was found suitable, 1'50 was too strong. The author's
experiments have shown that a solution of caustic potash of spec,
grav. T166, as employed for the absorption of carbon dioxide, is
very suitable indeed, if 50 grams pyrogallol are added to 1 litre.
1 c.c. of this solution absorbs 13 c.c. oxygen. Caustic potash
purified by alcohol should not be employed in this case. The
absorption of oxygen takes place more slowly than that of carbon
dioxide, but it is usually complete within three minutes, if the gas
and liquor are brought into very intimate contact and if the
temperature does not fall below 15°. The solution is kept in a
composite gas-pipette.

Boussingault (Compt. rend. Ivii. p. 885) and, later on, Calvert
and Cloez (ibid. pp. 870 & 875) have shown that during the
oxidation of the alkaline solution of pyrogallol a small quantity of
carbon monoxide may be formed. The quantity of this gas is not
constant, but is dependent upon the energy of the absorbing
process. Pure oxygen yields more carbon monoxide than if
diluted with nitrogen or otherwise ; the formation of: carbon
monoxide is also favoured by the concentration of the absorbent.
From 100 vols. pure oxygen Boussingault obtained 3*4 — 1'02 —
0-40—0-06, Calvert 1'99— 4'00, Cloez 3'50 ; from 100 vols.
oxygen mixed with various proportions of nitrogen, Boussingault
obtained 0'40, Cloez 2'59 vols. carbon monoxide. Consequently
Boussingault states that, when applying this absorbent to the
examination of atmospheric air, it may happen that the volume
of oxygen is found'0'1 or 0'2, or even 0'4 per cent, below the truth.

72 ON APPARATUS AND METHODS.

Vivian 13. Lewes (Journ. Soc. Cliem. Ind. 1891, p. 407) recommends
to employ the solution not more than four times, as it only then
begins to yield carbon monoxide. He also recommends to keep it
for twelve hours before use, but he gives no reason for this.
Contrary to the afore-mentioned statements, Poleck (Zeitsch. f.
analyt. Chem. lcS69, p. 451), when specially examining this source
of error in researches on the composition of air, could not find
even traces of carbon monoxide to be formed by the employment
of pyrogallol, and he therefore recommends this method as per-
fectly reliable in the case of moderate percentages of oxygen.
The same observation is made in technical gas-analysis; at all
events the quantity of carbon monoxide evolved is too small to
sensibly influence the results of determinations of oxygen.

The alkaline solution of pyrogallol of course equally absorbs
carbon dioxide, and this gas must therefore be previously removed
before commencing the estimation of oxygen.

3. Copper (ammoniacal cuprous oxide)- — Those metals which
form soluble ammonia compounds, as copper, zinc, and cadmium,
in contact with ammonia and oxygen are transformed into
the respective compounds with absorption of oxygen. Lassaignc
and later on Hernpel (Gas-analytische Methoden, 1900, p. 142)
have applied this behaviour in a very convenient manner for
the estimation of oxygen. Copper is preferred to the other
metals, because it dissolves without the evolution of hydrogen, and
because it can be employed in the shape of thin wire-gauze offering
a large absorbing-surface. A tubulated gas-pipette is charged
with small coils of such wire-gauze and with a mixture of equal
volumes of a saturated solution of commercial ammonium car-
bonate ana of liquor ammonige, of spec. grav. 0'96. If a gas
containing oxygen is introduced into such a pipette, the oxygen
is absorbed without any agitation in less than five minutes.
Probably at first a compound cf ammonia Avith cuprous oxide is
formed which absorbs a further quantity of oxygen and thus yields
a compound of ammonia with cupric oxide, which, in contact with
the copper present in excels, is re- transformed into the cuprous
compound. This would mean that 1 gram copper can absorb
177 c.c. oxygen.

Copper moistened with liquor ammonite absorbs oxygen much
more quickly than the alkaline solution of pyrogallol, and more
conveniently, as there is no necessity for agitation. Its efficiency
is nearly equal to that of phosphorus, but it is superior to the

ABSORBENTS FOR CARBON MONOXIDE. 73

latter as being absolutely harmless and as being active down to
— 7° C. But its use is restricted by the fact that it absorbs equally
well carbon monoxide, which is present in many gaseous mixtures
where oxygen must be determined. It also absorbs ethylene and
acetylene, the latter with formation of red, explosive copper acetylide.
Before employing it, carbon dioxide must of course be removed.

(1. Absorbents for Carbon monoxide.

The general absorbing-agent for carbon monoxide is a solution
of cuprous chloride, which absorbs it with formation of carbonyl-

cuprous chloride, Cu2 ^A • It may be employed both as a

solution in hydrochloric acid or in ammonia; but the latter is now
generally preferred, because the carbon monoxide absorbed by it in
the presence of an excess of ammonia is gradually used up for the
formation of ammonium carbonate, and at the same time metallic
copper is deposited on the sides of the absorbing-vessel. This re-
action is formulated thus :

Cu2 (Clo.CO) + 4NH, + 2H2O = 2Cu + 2NH4C1 + (NHJ 2CO3.

By this reaction the absorbed carbon monoxide is removed again,
and on the other hand the free copper protects the solution against
oxidation and reduces any cupric chloride formed to cuprous
chloride.

A very suitabb solution of cuprous chloride, sufficiently
ammoniacal, with but slight vapour-tension, is prepared as
follows : — 250 grams ammonium chloride is dissolved in 750 c.c.
water in a bottle provided with a good india-rubber cork and 200
grams cuprous chloride is added. The latter on frequent agitation
dissolves, leaving a little cupric oxychloride behind, forming a
brown liquid which keeps for an indefinite time, especially if a
copper spiral is inserted, reaching from top to bottom. In contact
with air the solution forms a precipitate of green cupric oxy-
chloride. In order to make it ready for use, it is mixed with one-
third its volume of liquor ammonite, spec. grav. 0'910. It is
usually kept in Hem pel pipettes with a water seal, provided at the
lowest point of the connecting-.tube with a short branch tube,
fitted with a pinch-cock, to facilitate the charging. This is per-
formed by connecting the open end of the pinch-cock tube with
an india-rubber tube reaching above the top of the pipette,
putting a funnel into the top, and pouring in first 50 c.c. liquor

74 ON APPARATUS AND METHODS.

ammonias and then 150 c.c. of the stock solution of cuprous
chloride, whereupon the charging-tube is taken off and the outer
end of the pinch-cock tube is closed by a bit of glass rod.

1 c.c. of this ammoniacal cuprous chloride solution absorbs
16 c.c. CO. But as this gas is held so loosely that the com-
bination is destroyed to a slight extent even by a decrease of
pressure, as found by Tamm (Jernkontorets Annaler, vol. xxxv.)
and Drehschmidt (Berichte, 1887, xx. p. 2752), the latter recom-
mends (ibid. 1888, xxi. p. 2158) to employ two pipettes in series —
the first, which has to receive the bulk of the carbon monoxide,
charged with a cuprous chloride solution used several times
previously ; the second, which has to absorb the small remainder of
carbon monoxide, charged with a fresh, very active, solution.
These two pipettes should be provided with labels of different
colour, to prevent mistakes.

The ammoniacal cuprous chloride absorbs also carbon dioxide,
heavy hydrocarbons (especially ethylene), and oxygen, all of which
must be first removed from the gases before estimating the carbon
monoxide.

Sometimes it is sufficient merely to prove the presence of carbon
monoxide by a qualitative reaction}iu such cases where there is too
little of it for a gas volu metrical estimation, and where its com-
bustion to carbon dioxide and gravimetrical estimation in this,
shape is not possible on account of the presence of other carbon
compounds. This is, for instance, the case with the air of heated
rooms and of certain coal-pits. In such cases a colorimetrical test
can be applied (Ch. Winkler, Zeitsch. f. analyt. Chemie, 1889,
p. 275) by absorbing the carbon monoxide in a suitable solution
of cuprous chloride, diluting and adding sodium-palladium pro to-
chloride. The absorbent is prepared by dissolving 100 grams
CuoClo in 1 litre of nearly saturated sodium chloride solution.
This solution is colourless or but slightly brownish ; it forms a
precipitate of green cupric oxychloride in contact with air, but
keeps unchanged in a bottle closed with an india-rubber cork and
provided with a copper spiral reaching from top to bottom. If the
gas is slowly passed through this solution, or agitated with it in a
closed bottle for some little time, most of the carbon monoxide is
absorbed. A portion of the liquid is put into a test-tube, diluted
with three or four volumes of water, without troubling about the
white precipitate of cuprous chloride (this is indispensable !), and

a drop of a solution of sodium-palladium protochloride is added.
In the presence of the slightest quantity of carbon monoxide a
black cloud of finely divided palladium is formed. If the test is
always performed under exactly similar circumstances, the depth
of the black colour admits of approximately guessing the quantity
of carbon monoxide. Thus O'Ol c.c. of CO = (K)000125 gram
can be found. The presence of other gases does not materially
influence the reliability or sensibility of this reaction."

e. Absorbents for Nitrogen.

These may be used for the isolation of argon and its congeners.
Hempel (Zeitsch. f. anorgan. Chem. 1899, xxi. p. 19) has proved
that nitrogen is absorbed at a red heat by a mixture of 1 part (by
weight) magnesium powder, 5 parts freshly ignited calcium oxide,
and 0'25 part sodium. 1 gram of this mixture during an hour
absorbed 52 c.c. nitrogen.

B. Estimation of Gases by means of Apparatus combining the
functions of Absorption and Measuring.

A description of these apparatus and methods might be con-
sidered superfluous, as regards the apparatus described below
under a, because they have been replaced by better ones of late
years. Nevertheless this description will not be omitted, both
because these apparatus and methods give an idea of the
gradual evolution of technical gas-analysis, and because they
are persistently employed in many places up to this day, in spite
of all progress made elsewhere. Moreover, their manipulation is
an excellent help in teaching, as they make the student quickly
familiar with the physical principles to be applied in the measure-
ment of gases.

a. WINKLER'S GAS-BURETTE.

The apparatus described below*, constructed by the author in
1872, consists of two communicating tubes, the measuring-tube A
(fig. 42) and the level-tube B, held by the clamps of an iron stand,
and connected at their lower extremities by an india-rubber
T-piece d, whose free branch is usually closed by a pinch-cock.
The rneasuring-tubc A serves for receiving the gas, and is at
its bottom provided with a double-bored tap, a, of peculiar

ON APPARATUS AND METHODS.

construction as described
above (p. 33; see figs. 2c, 26,
27). The top of the mea-
suring-tube is closed by a
simple glass tap, b.

The measuring-tube holds
about 100 c.c. between the
two taps. It is exactly mea-
sured once for'1 all, and the
amount etched on the cube.
The tube is, moreover, divided
from the bottom upwards into
tenths of cubic centimetres,
including the contracted pieces
adjoining the taps, of wlr'ch
the lower one occupies about
a quarter of the total length
of the tube, and is intended
for measuring small volumes ;
whilst the upper one should be
as short as possible, to prevent
any liquid from adhering to
it.

The level-tube B serves for
receiving the absorbing-liquid.
It is closed at the top by an
india-rubber cork, carrying a
bent tube e, with an india-
ruhhcr tube attached to it.
The lateral outlet-tap c with
india-rubber tube, which in-
creases the liability of the
apparatus to fracture, is not
indispensable and may be Jeft
out.

The stand is provided with a
movable holder for the tubes, so
that these may be placed at will either in a vertical or a horizontal
position. Jf there is no suitable working-bench, the apparatus
is placed on a lead-covered wooden basin C, provided for the
(mostly alkaline) absorbing-liquids and rinsings.

WINKLEE S GAS-BURETTE.

Manipulation. — Open the tap b, and by means of the tap a put
the measuring-tube A in communication with the gas to be
analyzed ; by means of an india-rubber pump or an aspirator,

Fig. 43.

cause a current of the gas to traverse the measuring- tube till all
air has been driven out. According to whether tliis is done by
pressure or by aspiration, either the tap a or the tap b is closed first

78 ON APPARATUS AND METHODS.

in order to be sure that the sample of gas is under atmospheric
pressure. The tap a is placed so that the inner end of its longi-
tudinal bore is turned downwards.

The level-tube B is now filled with the absorbing-liquid ; the
air enclosed below the tap a is expelled by a momentary opening
of the pinch-cock attached to that tap ; and now, since the gas
and the liquid are only separated by the plug, the absorption
may begin. For this purpose the plug is turned 90 degrees, thus
maldng a connection between the two tubes. The absorbing-
liquid at once begins to enter the measuring-tube ; by blowing
into the india-rubber tube attached to the level-tube B it is forced
up a little, and the tap a is turned back into its first position. By
alternately placing the tubes vertically and horizontally (fig. 43),
the gas and the liquid are brought into intimate contact, and this
alternate movement causes the absorbable portion of the gas to be
absorbed quickly. If, on again opening the tap «, no more liquid
enters into the measuring-tube the absorption isrcomplete. The
only thing remaining is to produce the same level of liquid in the
communicating- tubes, which is done by opening the lateral tap c,
or equally well by the pinch-cock d, leaving, of course, the tap a
open in the meanwhile. The volume of liquid entering into A is
equal to that of the gas absorbed, and is converted into per cent.
by volume on multiplication by 100, and division by the contents
of the measuring-tube.

After each estimation the apparatus is thoroughly rinsed with
water ; the taps are dried with blotting-paper, and are again
greased, slightly but equally, all over. Whilst the apparatus is
out of use the plugs of the taps should be taken out, as they
frequently stick very fast when left in.

Applications : —

(a) For estimating carbon dioxide in mixtures of that gas and air
or in the gases resulting from combustion (chimney -g as e s] } from
blast-furnaces, or lime- kilns ; in the gas for saturating the lime in
sugar-refining fyc. The absorbent is a moderately strong solution
of caustic potash.

(b) For estimating oxygen in the atmospheric air. The absorp-
tion is effected by means of an alkaline solution of pyrogallol. In
order to avoid an excess of this somewhat costly reagent, a suffi-
cient quantity of a concentrated aqueous solution of pyrogallol is
first poured into the level-tube and brought close to the tap a} and
then a solution of caustic potash is run in on the top of it.

MODIFIED WINKLER S GAS-BURETTE.

Fig. 44.

(c) For the examination of commercial liquid carbon dioxide and
of natural sources of gaseous carbon dioxide. A. Lange (Chem.
Tnd. 1900,, p. 530) has modified Winkler's gas-burette for the
above-mentioned purposes, as shown in fig. 44. Tube A, holding
100 c.c., is at top con-
tracted into a tube holding
5 c.c. and divided. It is
connected by means of an
elastic joint with tube B,
but there is no branch tube
serving as an outlet. The
elastic tube, put on the bent
tube c, is at the bottom
continued into a glass tube
reaching into the 250 c.c.
bottle D, fixed on the stand.
The apparatus is charged
by pouring into B sufficient
caustic potash solution of
spec. grav. 1'297 to fill A
and B rather more than
halfway up. Now the
cork, with tube c and the
elastic tube, is put on B, and
by means of another elastic
tube, put on b, air is blown
into A, until the level of
the liquid is below tap a,
which is then closed. B is
now filled with potash so-
lution as well as c and the
elastic tube. D is also
filled with the same solution,
and by opening tap b it is
ascertained that the bore of
this tap is also filled. The
apparatus is now ready for
use. By turning the three-
way tap a 90 degrees, the
gas can be passed into A.

80 ON APPARATUS AND METHODS.

When b has been closed and a opened, the potash solution, in
consequence of the absorption of COo, will now from D into -B and A.
After the absorption has been finished and b has been opened, the
solution flows back into D and the level is automatically restored in
a, whereupon the apparatus is again ready for work. There is no
trouble about handling the potash solution, 'and upwards of 400
tests can be made without renewing it.

The examination of liquid carbon dioxide is carried out as
follows : — The iron bottle containing it is placed in an upright
position; a coupling-piece is tightly screwed on and an elastic
tube is drawn over its free end. The valve of the bottle is
cautiously opened and is regulated so as to yield a regular,
moderate stream of gaseous carbon dioxide. Then the elastic tube
is joined on to tap a, which has been turned so as to admit the
carbon dioxide into A; the air escapes through the open tap b. After
one minute tube A is filled with carbon dioxide, and this can be
continued till needle-shaped crystals of potassium hydrocarbon ate
appear in the contracted part of A. Then b is closed, the elastic
tube is taken oft', whereupon the pressure within A becomes equal
to that of the outer air, and a is turned 90 degrees, so that A
communicates with B. The potash liquor at once rises in A, and
by inclining the apparatus, ultimately to the horizontal, the
absorption in A is accelerated without forming a vacuum. At
last the apparatus is moved upwards and downwards, then fixed
in the vertical position and the volume is read off, after lifting up
bottle D and levelling the liquid in D and A. Or else a correc-
tion table is made which admits of taking the reading without
moving bottle D. There should be no greater difference between
two tests than 0*05 p. cent. As the contracted part of A is divided
into yV c.c., the readings can be made to O'Ol c.c.

In order to take a sample of the liquid contents of an iron
bottle, this is placed horizontally on a stool so that the coupling-
joint of the valve points upwards. By slowly opening the valve
it is generally possible to produce a suitable, moderate stream of
COo, but small quantities of solid CO2are always ejected. With
some valves the adjusting of the stream is very troublesome; it
issues in jerks and sometimes stops, but with the slightest touch
of the valve it becomes so violent that the elastic tubes are thrown
off. In such cases a reducing- valve can be interposed, but the
gas must be allowed to issue long enough to drive out all the air

HONIGMANN S GAS-BURETTE.

from the valves : in this case the results agree completely with
those obtained directly from the contents of the bottle.

Holste (Zeitsch. f. d. ges. Kohlensaure-Industrie, 1897, p. 462)
has described a method for calculating the true percentage of air
in the carbon dioxide from the figures obtained for the gaseous
and the liquid portion.

(d) Examination of liquid chlorine and strony chlorine gas •
estimation of carbon dioxide in electrolytic chlorine. The process
is carried on as in (c), but the absorbent is a concentrated solution
of ferrous chloride, which absorbs the chlorine rapidly and in
quantity, leaving air and carbon dioxide behind. In a second
burette the gas is treated with caustic-potash solution, which
absorbs chlorine and carbon dioxide, leaving the air alone behind.
The carbon dioxide is found by difference. Probably it is best to
first saturate the ferrous-chloride solution with carbon dioxide,
his process is analogous to Lange's as above described, but has not
yet been introduced into factories.

b. HONIGMANN'S GAS-BURETTE.

The burette A (fig. 45) consists of a measuring-tube, tapering
at both ends, closed at the top by a simple tap a ;
whilst the bottom end b is left open and is
merely provided with a piece of stout india-
rubber tubing. The zero-mark of the tube,
which is divided into 4 cubic centimetres, is near
the bottom, and it holds up to the tap exactly 100
cub. centim. The absorbing-liquid is contained
in the glass jar C ; the elastic tube allows the
burette to be plunged down to any depth.

Manipulation.— rThis burette is specially in-
tended for estimating the percentage of carbon
dioxide in the gases employed for carbonating the
ammoniacal solution of sodium chloride in the
manufacture of sodium carbonate by the am-
monia process. Gas is aspirated through it till
all air has been expelled ; the tap a is closed
and the burette is immersed in the glass jar C,
tilled with a solution of caustic potash exactly to
the zero-mark. The tap a is now opened for a
moment, in order to equalize the pressure within
and without, and thus exactly 100 cub. centirn.

Fij?. 45.

82 ON APPARATUS AND METHODS.

of gas are confined in the burette. The absorption of the carbon
dioxide is started by immersing the burette somewhat lower, so
that its inside is wetted with the potash solution, and then pulling
it out so far that the end of the elastic tube remains within the liquid ;
but the burette itself is raised over the edge of the glass jar and
can be moved about and downwards. The solution of caustic
potash at once begins to enter, and after agitating a few times the
absorption is complete. The burette is now again immersed in the
liquid, so far that the inner and outer levels are the same, and the
reading is taken, yielding directly the percentage of carbon dioxide
by volume. This apparatus cannot produce absolutely accurate
results; but its construction and manipulation are very simple, and
the results are obtained in a few moments. After each absorption
the burette and elastic tube must be most carefully rinsed with
water.

Application : —

For estimating carbon dioxide in mixtures of that gas and air,
in the gases from lime-kilns, for saturating the lime in sugar-
refining, fyc.

c. BUNTE'S GAS-BUBETTE.

The measuring-tube A (fig. 46) carries a funnel t, provided with
a mark, and is closed at the top by the three-way cock a (comp.
p. 33) and below by the plain tap b, ending in an almost capillary
tube. The space between is rather more than 110 c.c., and is
divided into fifths of a cubic centimetre. The mark 100 coincides
with the plug of the upper tap a ; the zero-point is 6 or 8 centi-
metres above the tap b, and the division is carried 10 c.c. beyond
this. The gas in this burette is always measured at the atmo-
spheric pressure, plus the pressure of the column of water contained
in the funnel up to the mark.

This tube is fixed to an iron stand by means of an easily opened
clamp ; a second arm of this stand carries the funnel B, which can
be connected by an elastic tube, about 3 millimetres wide, with
the capillary bottom end of the burette.

There is, moreover, a small glass or porcelain cup C for holding
the absorbent, and two aspirating-bottles, whose construction is
evident from the diagram. The bottle D serves for forcing water
into the burette or withdrawing it therefrom. In both cases the
rubber end n is put upon the tip of the burette at b • whilst at the

BUNTE'S GAS-BURETTE.

Fig. 46.

84 ON APPARATUS AND METHODS.

same time air is blown by the mouth into the tube m, so that
during the fixing of the tube water is always running out of n,
and no air-bubble can be enclosed. This precaution must never be
omitted. If larger quantities of liquid have to be withdrawn from
the burette, the bottle E may serve for this; it is attached directly
to the bottom of the burette, after having been evacuated of air by
means of a water-jet pump.

Bleier (Berichte, 1895, xxviii. p. 2423) has proposed some
modifications of the Bunte burette, for which we must refer to
the original.

Manipulation. — The burette is filled with water through the
funnel B, till it begins to enter the funnel at the top of the burette.
The taps are now closed and the india-rubber tube is detached
from the bottom of the burette. The longitudinal bore of the
tap a is now connected with the tube conveying the gas, already
filled with the same, and the gas is aspirated by running water
out of the bottom end b. Rather more than 100 c.c., say about
105 c.c., of gas is allowed to enter the burette, and the exact
adjustment to the zero-mark is made as follows :— By means of the
bottle D sufficient water is forced into the burette to compress the
gas to about 95 c.c. ; then b is closed, the bottle D is taken off,
and by cautiously turning the tap b the water is run out again,
exactly to the mark for 0. The gas is still under a plus pressure,
and now, by a last operation, that pressure has to be established
at which every reading-off has to take place in the case of this
apparatus. For this purpose the funnel t is filled with water up to
the mark, and, by momentarily opening the tap #, the excess of
gas is made to escape through the water. The burette now con-
tains exactly 100 c.c. of gas at the pressure of the atmosphere,
plus the pressure of the column of water standing in the funnel t.

Of course the burette may just as well be filled by drawing the
gas through it by means of an india-rubber pump or any other
aspirator till all air has been expelled, and then forcing water into
it by means of the bottle D, adjusting it to the point 0, filling the
funnel t with water up to the mark, and by a momentary opening
of the tap a raising the surplus pressure.

In order to absorb any one of the constituents of the gas, a suit-
able absorbing-liquid must be introduced into the burette. First
the water contained in the same is drawn off by means of the bottle
D up to the tap b} the latter is closed, and the end of the burette
is dipped into the cup C containing -the absorbing-liquid. If the

BUNTE'S GAS-BURETTE. 85

tap b is now opened again, a volume of the absorbing-liquid, almost
equal to that of the water drawn off, enters the burette, and
rises in it almost up to the zero-point, but not quite, owing
to its higher specific gravity. In any case the quantity of liquid
thus introduced suffices for removing the absorbable constituent
of the gas, and in order to effect this it is only necessary to
bring the gas and the liquid into intimate contact. For this
purpose the burette, after closing the tap b, is taken hold of by the
funnel-end, closing the latter by the hand, and is moved up and down
in short, but not violent, jerks. When the absorption is complete
the tip of the burette is again dipped into the cup C, and the tap
b is opened, whereupon liquid enters in the place of the absorbed
gas. If, on repeating the operations just described, the liquid
remains at the same level, the reading may be taken. First, how-
ever, the gas has to be put under the proper pressure by running
water into the burette out of the funnel t (thus also rinsing its
sides), and then, whilst the tap a is left open between the funnel
and the tube, filling the funnel with water up to the mark.

Since the adhesion of the absorbing-liquids differs from that of
water, it is preferable to remove those liquids by water and to
repeat the reading. Both taps are opened, whilst water is run
into the funnel t in a steady stream, and this rinsing of the burette
is continued till the original reaction of the liquid has ceased.
Gas cannot be lost in this way ; therefore, after the water con-
tained in the burette has been drawn off, a different reagent can be
introduced in order to absorb another of the gaseous constituents.
In the same way a third and fourth gaseous constituent can be
removed and volumetrically estimated by rinsing out and intro-
ducing suitable absorbents.

But as this manipulation requires the use of a large quantity of
water, by which some of the gas may be dissolved, it is best to
draw off most of the absorbing-liquid by suction and wash by means
of a few c.c. of water, which is again sucked off, repeating this as
often as may be necessary.

Applications : —

(a) Estimation of carbon dioxide in a mixture of that gas and
air, or in the gases resulting from combustion, from blast-furnaces,
lime-kilns, gas-producers, fyc. The absorbent in this case is a
moderately strong solution of caustic potash.

(b) Estimation of oxygen in atmospheric air. The absorbent

86 ON APPARATUS AND METHODS.

is an alkaline solution of pyrogallol. Not to waste the latter, a
concentrated aqueous solution of pyrogallol is first introduced into
the burette, and this is followed by a strong solution of caustic
potash.

(c) Estimation of carbon dioxide, oxygen, and nitrogen in mix-
tures of air and carbon dioxide, or in chimney -yases. The carbon
dioxide is absorbed by a solution of potash ; this is washed out and
the oxygen absorbed by a strongly alkaline solution of pyrogallol.
After again washing out, the nitrogen remains behind.

(d) Estimation of carbon dioxide, oxygen, carbon monoxide, and
nitrogen in blast-furnaces or producer-gases. Carbon dioxide and
oxygen are absorbed as under (c), and afterwards carbon monoxide
by a solution of cuprous chloride in hydrochloric acid. This is
washed out first with dilute hydrochloric acid, then with pure
water, and the remaining nitrogen is ultimately measured.

C. Estimation by means of Apparatus ivith separate Parts for
Measuring and Absorption.

The absorption of a gaseous constituent is frequently not carried
out in the measuring-tube itself, but in a separate vessel, which
serves for holding the absorbing-liquid and for bringing the gas
into contact with it after being measured. When the absorption
has been finished, the remaining gas is again carried over into the
measuring-tube and its volume is read off. The volume of the
gas absorbed follows from the difference of the two readings.
This process admits of thoroughly utilizing the absorbent and dis-
penses with washing out the measuring-tube after each estimation.
In this way hundreds of measurings can be carried out without
necessitating any essential intermediary operation, and before
cleaning and refilling the apparatus.

The measuring and absorbing vessels must in this case be
capable of being connected with each other in a permanent or a
temporary way. The connection is usually made by a narrow
capillary tube whose contents scarcely amount to -^ c.c. ; the
quantity of air contained in it, which becomes mixed with the gas
under examination, is hence so slight that it does not sensibly
influence the result. In special cases this capillary tube may be
filled with water in order to drive out the air.

The first apparatus of this kind was constructed by C. Scheibler;
it served for estimating the carbon dioxide in the saturation-gases

ORSAT'S APPARATUS. 87

of sugar-works. This apparatus, like some others, for instance
that of M. Liebig, did very good service in its time, but has now
been superseded by others of simpler description.

«. Ons AX'S APPARATUS * .
The measuring-tube A (fig. 47) contains, from the zero-mark

Fig. 47.

at its bottom to the upper capillary end, exactly 100 c.c., but
its graduation (in J- c.c.) only extends to 40 c.c. and ceases where

* This apparatus is an imitation of that constructed by Schlosing and Holland,
which is not so well known j both depend upon a principle first applied by Regnault
and Reiset. It has met with extraordinary approbation, and has undergone
many modifications, e. g. by Salleron, Aron, Ferd. Fischer, Rob. Muencke,
E. Tomson, Corn. Heinz, H. Petrzilka, O. Bleier, P. Fuchs, H. Fahlenkamp, and
others. The author prefers the form now constructed by Rob. Muencke of Berlin,
and therefore describes that only.

88 ON APPARATUS AND METHODS.

the tube is enlarged. In order to protect the gas contained in
this burette from the influence of the changes of external tempe-
rature, the tube is surrounded by a water-jacket, closed at top and
bottom by india-rubber stoppers and provided with a white back-
ground of opaque glass, upon which the black divisions of the
burette are plainly visible. The bottom of the burette is connected
by an elastic tube with a level-bottle B filled two-thirds with
water; the top end is connected with a glass capillary r, bent at
a right angle and ending in the three-way cock h. This tube is
protected against breaking by a wooden frame, and carries at a
right angle three glass taps h' h", h'", each provided with a capil-
lary tube and connected by india-rubber joints with the three
U-shaped absorption-vessels O, Cn, C'n, filled with bundles of glass
tubes. The first of these is filled with a solution of caustic
potash, the second with an alkaline solution of pyrogallol, the
third with a concentrated solution of cuprous chloride in hydro-
chloric acid. In order to keep this solution in an unchanged
state, it is left in constant contact with copper spirals, introduced
into the glass tubes with which the vessel C"' is filled.

In lieu of the easily broken glass taps Naef (Chem. Ind. 1885,
p. 289) recommends india-rubber tubes with globular glass valves
(so-called Bunsen valves) ; Olschewsky prefers ordinary pinch-
cocks. Lunge has added to the three ordinary absorbing-tubes
a contrivance for burning hydrogen (Chem. Zeit. 1882, p. 262 ;
comp. the chapter on hydrogen). Thorner (Chem. Zeit. 1891,
p. 768) and Hankus (Oesterr. Zeitsch. f. Berg- u. Hiittenwesen,
1899, p. 81) have enlarged the Orsat apparatus so as to comprise
all contrivances for combustions, including an apparatus for
evolving oxyhydrogen gas, an explosion pipette, and an induction-
coil. This seems going too far in the direction of complicating
the apparatus.

The above-mentioned liquids serve for absorbing carbon dioxide,
oxygen, and carbon monoxide respectively. The whole apparatus,
as may be seen from the description, is specially intended for
analyzing combustion-gases. Oxygen can also be absorbed by
moist phosphorus in the shape of thin sticks ; comp. p. 68. In
this case the vessel C" contains no glass tubes, but ends at the top
in a small neck, closed by a soft india-rubber cork, through which
thin sticks of phosphorus may be introduced into the water con-
tained in the vessel till it is full.

The absorbing-vessels are filled with water rather more than

halfway up ; and this is then drawn up to the mark made in the
capillary neck by opening the connecting-tap and running off the
water contained in the burette A, for which purpose the level-bottle
B must be lowered. In order to protect the absorbing-liquids
against the action of the air, the outer ends of the vessels are closed
by small balls of very thin india-rubber. The apparatus is fixed
in a portable wooden box, closed on both sides by sliding doors.

Manipulation. — Raise the level-bottle at the top, open the tap h,
and allow the burette A to fill with water up to the capillary
part. Connect the outer end of the capillary tube with the tube
through which the gas is to be led, and the lower end of the three-
way cock h with an india-rubber pump by which the air is removed
from the conducting-tube. Now aspirate the gas by lowering the
level-bottle B and turning the tap h through 90 degrees. Run off
the water a little below the zero-mark, close the tap h, compress
the gas by raising the level-bottle B till the water rises above zero,
squeeze the connecting india-rubber tube close to the joint by
means of the fingers or a pinch-cock, and then, after lowering the
level-bottle B, allow the excess of water to run out to zero by
cautiously loosening the elastic tube. Last of all the tap h is
opened for an instant in order to produce a pressure equal to that
of the atmosphere, whereupon exactly 100 c.c. of gas will be con-
fined within the burette.

Now the absorption begins, first of all that of the carbon dioxide
by conveying the gas into the U-tube C1. This is done by raising
the level-bottle B, and at the same time opening the tap h1. The
absorption is hastened by causing the gas to travel several times
from C' to A and back, alternately lowering and raising the
level-bottle and leaving the tap h1 open all the time. At last the
level of the liquid in C1 is adjusted to the mark, and the tap h' is
closed. Now the reading can be taken, after raising the level-
bottle B till its contents are at the same level as the water
within the burette. The decrease of volume found indicates
directly the percentage by volume of carbon dioxide. In exactly
the same way the oxygen is absorbed in C", and then the carbon
monoxide in C'" ; the unabsorbed residue represents the nitrogen.
If the oxygen is absorbed by moist phosphorus, it is unnecessary
to convey the gas backwards and forwards as above described ; the
fumes of phosphorous acid which form during the absorption and
cling for a long time to the gas need not be taken into account.
In order to save the trouble of carrying the gas backwards and

90 ON APPARATUS AND METHODS.

forwards from the burette into the absorbing-tubes, Namias
(Stahl u. Eisen, 1890, p. 788) and Le Docte (Chem. Zeit. 1900,
p. 375) have constructed automatically moving contrivances.
Cario (Germ. Pat. No. 98667 ; Chem. Zeit. 1898, p. 977) describes
an automatically acting apparatus, slightly deviating from Orsat's.
L. Kaufmann & Co. (Chem. Zeit. 1931, Rep. p. 26) describe,
under the name of " Ados/' an apparatus for automatically
analyzing furnace-gases and continually registering the results.

Application : —

Estimation of carbon dioxide, oxygen, carbon monoxide, and
nitrogen in artificial mixtures of gases, in gases from blast-furnaces,
reverberatory-furnaces, or other combustion-gases.

The absorbing liquids are : —

For carbon dioxide : a solution of caustic potash of
specific gravity 1 *20.

For oxygen : the same solution, to which 15 or 25 grams
of pyrogallol have been added for each apparatus, or, in
lieu of that, phosphorus and water.

For carbon monoxide : ammoniacal solution of cuprous
chloride ; comp. p. 73.

For controlling the efficiency of furnace-fires it is usually
sufficient to estimate the carbon dioxide in the chimney-gases.
The calculation of the loss of heat by the chimney-gases is dealt
with in Lunge's ' Taschenbuch ftir die Sodaindustrie ' &c., 3rd ed.
p. 130, and by F. Fischer in Lunge's ' Chem. techn. Untersuchungs-
methoden/ Berlin, 1899, vol. i. p. 216.

An apparatus not exactly serving for gas-analysis, but con-
structed for an approximate estimation of the action of a
furnace-tire, is Hempel's gas lanthorn (Chem. Ind. 1896, p. 98),
the principle of which is the alteration in the size of a flame
corresponding with the percentage of oxygen in the gas by which
it is fed.

F. Fischer (Dingl. Journ. cclviii. p. 28) has employed the
Orsat apparatus for estimating the total acids of pyrites -kiln
gases, as well as their percentage of oxygen. For this purpose he
charges the apparatus with petroleum in lieu of water. [This
causes some inconveniences, therefore other apparatus is preferable
for the above purpose ; comp. below. — Translator. ~\

APPARATUS FOR ESTIMATING CARBON DIOXIDE.

b. APPARATUS FOB ESTIMATING CABBON DIOXIDE IN GASEOUS MIXTURES

CONTAINING RELATIVELY LITTLE OF IT.

The following simple apparatus, which is on the same prin-
ciple as Orsat's, is convenient for estimating relatively small quan-
tities of carbon dioxide, such as occur in pit-gases and the like,
which may, even if amounting only to a few per cent., con-
siderably interfere with respiration.

The measuring-tube A (fig. 48) is closed at the top by a three-
way cock, at the bottom by a simple tap. Its capacity is 100 c.c.;
the principal portion of this is contained in the globular part; the

Fig. 48.

narrow cylindrical part only contains 5 c.c., and is divided into
tenths of a cubic centimetre. The lower end of the burette is con-
nected by a narrow elastic tube with the level-bottle C, containing
pure water; from the upper part a glass capillary tube leads to the
absorbing-vessel B} which is filled with a solution of caustic potash
up to^the mark. The burette is filled with the gas to be examined
through the pinch-cock attachment of the three-way cock ; other-
wise the apparatus is handled exactly as Orsat's. The bottom tap
is necessary, because in this case the level of the water must be

92 ON APPARATUS AND METHODS.

very finely adjusted, which is difficult to do without a tap. When
the liquids contained in the communicating vessels A and C have
been brought to the same level, the tap is closed and the reading
is made.

Application : —

Estimation of carbon dioxide in artificially prepared mixtures of
the same with air, in the gases from coal-pits, wells, caves, subsml,
tombs, from respiration, in chimney-gases poor in CO2, fyc.

c. LINDEMANN'S APPARATUS FOR ESTIMATING OXYGEN *.

In a similar way oxygen can be estimated in many gaseous
mixtures by means of moist phosphorus as absorbent. Other
gases do not interfere, unless they are absorbed by water or
unless they disturb the reaction between oxygen and phosphorus
(comp. p. 69). Carbon dioxide in particular is all but indifferent
in this case, a fact which is frequently of importance.

The apparatus is shown in fig. 49. The measuring-tube A

Fig. 49.

has a three-way cock at the top, but no tap at the bottom. It
* Modified by the author.

HEMPEI/S APPARATUS. 93

contains altogether 100 c.c., 75 c.c. of this in the globular and
25 c.c. in the cylindrical part, which is divided into tenths of a
cubic centimetre. The level-bottle C contains water, the absorbing-
vessel B thin sticks of phosphorus and water up to the mark. The
gas is introduced through the pinch-cock arrangement connected
with the three-way cock ; otherwise the manipulation is exactly as
with Orsat's apparatus.
Applications : —

(a) Estimation of oxygen in atmospheric air (whether containing
C02 or not] ; in the air from graves, from respiration, Weldon's
oxydizers, Bessemer converters, vitriol-chambers, fyc.

(b) Estimation of the proportion between oxygen and nitrogen in
unabsorbed residues of gases, such as are left on treating gaseous mix-
tures with alkaline liquids, for instance in gas from pyrites-kilns,
from making sulphuric anhydride, from the Deacon process, &c.

d. HEMPEL'S APPARATUS.

Very important for absorptiometrical gas-analysis have been
the improvements introduced by W. Hempel (' Ueber technische
Gasanalyse/ 1877; ' Gasanalytische Methoden/ 3rd eel. 1900,
p. 29). He employs " gas-pipettes " as first constructed by
Ettling & Doyere, suitably modified, each serving for retaining
a certain constituent of the gas, and easily attached to the gas-
burette by means of a connecting capillary or removed in the
same way. This secures the advantage that the measuring of
a gas and its treatment with one or more absorbents can be
separately performed at leisure and in a very efficient way.
Consequently his methods secure a degree of accuracy otherwise
unattainable in technical gas-analyses and with water as a con-
fining liquid.

The simple gas-burette (fig. 50) consists of two communicating
cylindrical glass tubes, 1-5 cm. wide, 65 to 68 cm. long. The
measuring-tube A ends at the top in a capillary, 1 mm. wide and
3 cm. long, upon which a piece of india-rubber tube, 5 cm. long,
is drawn and fixed gas-tight by means of copper wire covered with
silk. This tube, as well as that required for closing the gas-
pipettes, should be the best thick-walled black india-rubber
tubing, 2 mm. bore and 6 mm. outer diameter. Closely above
the capillary the elastic tube carries a small pinch-cock, 5 cm.
long, which is taken off when not required. From this pinch-cock

ON APPARATUS AND METHODS.

arrangement down to the bottom mark, 3 or 4 cm. above the foot of
the burette, the tube holds 100 c.c., divided into I c.c., and showing
on one side the figures 0 to 100, on the other from 100 to 0.
Tube A is cemented into a base made of thin cast-iron or of

Fig. 50.

polished black wood, so that the lower contracted end is carried
out sideways at a right angle through this base. The level-tube B,
which is open at the top, possesses a similar base ; its bottom
end is connected with that of the burette A by an india-rubber
tube, which is conveniently intercepted in the middle by a short

HEMPEI/S APPARATUS.

piece of glass tubing, and which renders it possible to place the
tube B at a higher or lower level.

In lieu of this simple burette, one with a water-jacket may be
used, as shown in fig. 51. This jacket is 3 cm. wide, and serves
for keeping the temperature of the gas constant ; it possesses two
branches, at top and bottom, through which, if needful, water may
be run in a constant stream. The burette is fixed in the jacket

Fig. 51.

by means of india-rubber corks. In the great majority of cases,
there is no necessity for employing such a water-jacket.

For the examination of gases which cannot be confined over
water, on account of its too easy absorption of some of their con-
stituents, Hempel uses what is known as the modified Winkler's
gas-burette (fig. 52). This is closed at the bottom by a three-way
cock c, and at the top by the simple glass tap d or by a pinch-cock ;

96 ON APPARATUS AND METHODS.

the space between these contains 100 c. c., divided into fifths. Before
introducing the gas the measuring- tube must be completely dried^
which is conveniently done by rinsing it first with alcohol, then with
ether, then blowing a rapid current of air through it. The filling is
effected by passing the gas through it until all air is driven out, for
which purpose the pinch-cock attached to c is connected with the
source of the gas, and the tap d with the aspirator, or vice versa.
Otherwise the arrangement and manipulation are the same as with
Hempel's burette.

Gas-pipettes. — The pipette shown in fig. 50 at C is what is called
the simple absorption-pipette. It consists of two glass bulbs a and b,
.fixed on a wooden or iron stand and communicating by a bent tube.
The former of these is connected with a siphon- shaped capillary
tube, projecting a few centimetres beyond the wooden stand and
ending in a piece of india-rubber tubing. This is usually closed
by a short glass rod, but by a pinch-cock when the pipette is in
use. The india-rubber tubes, both of the pipette and the burette,
should be of the best thick material, and must be fastened on by
means of thin wire; otherwise leakages and other troubles may
occur.

The bulb a is completely filled with the absorbing-liquid, which
reaches into the siphon-bend of the capillary tube, whilst the bulb b
remains nearly empty. The filling takes place by pouring the
liquid into the wide tube attached to b, and sucking the air out
of a through the capillary tube. The reagent contained in the
pipette is marked on the stand by a label.

Bulb a should hold 200 c.c., bulb b 150 c.c.; if too small, the
pipette must be rejected. The capillary connected with a is closed
by an elastic tube and pinch-cock during use, and when out of
use by a piece of glass rod. The latter is put in while the pinch-
cock is closed, and is removed subsequently; otherwise air
will be forced into the capillary and the thin column of liquid
will be broken. Should this happen, the capillary must be emptied
by sucking for a moment at «5and be filled again by blowing air in
the opposite direction. The end connected with b is closed by a cork
when out of use, which must be removed before using the pipette.

A pipette for fuming oil of vitriol (comp. p. 66) is shown in
fig. 53. Here the small bulb d above a is filled by the glass-
blower with small bits of glass, which enlarges the absorbing
surface and renders agitation unnecessary. When not in use,

HEMPEL S APPARATUS.

the ends of this pipette are closed by small glass caps, which may
be made quite tight by means of small india-rubber rings.

The simple tubulated absorption-pipette (fig. 54) is arranged in
the same way as fig. 50, but the part a is made cylindrical and
has a neck at the bottom. Through this the pipette can be filled
with solid reagents, for instance phosphorus and water, whereupon
the neck is closed by a soft india-rubber cork and the pipette is
placed in the proper position.

With great ingenuity Hempel has arranged similar pipettes for
liberating and keeping gases. Thus we obtain, for instance, a
hydrogen pipette, by passing through a hole in the india-rubber

Fijr. 53.

Fig. 54.

stopper a glass rod, tapering upwards, upon which is fixed a zin
cylinder perforated in the centre, the vessel b being filled with
dilute sulphuric acid. As soon as the stopping of the capillary is
removed, the acid gets to the zinc and hydrogen is given off.
This is allowed to continue till all air has been expelled ; on
replacing the stopper the acid is forced back into the bulb b by
the liberated gas and the evolution of hydrogen ceases, while a
stock of the gas remains in the vessel a. In a similar way it
would be possible to evolve] carbon dioxide from marble and
hydrochloric acid, or nitric oxide from copper and nitric acid.

The composite absorption-pipette (fig. 55) is used for keeping
absorbents which suffer change in contact with atmospheric air,
like alkaline solution of pyrogallol, or a solution of cuprous chloride
in hydrochloric acid, or those which give off irritating vapours, such

OX APPARATUS AND METHODS.

as bromine-water. Here the pair of bulbs, a and b, forming the
absorbing arrangement proper, are connected with a second pair of
bulbs, c and d, which are converted into a water-joint by pouring in
a little water. Such pipettes must be filled through the capillary
tube fused on to the bulb a, by connecting its india-rubber end

Fig. 55.

Fig-. 56.

with a funnel-tube of about a metre in length (3 feet), through
which the absorbing-liquid is poured. If the liquid is to be filled
through a short funnel, this can be done by attaching to the outer
end of the bulb d an india-rubber tube with pinch-cock, and
withdrawing the air contained in the apparatus by now and then
sucking it out.

It is very convenient to attach to the lowest point of the con-
necting-tube between a and b a short branch, closed by a pinch-
cock or glass rod, for the purpose of charging- the pipette (fig. 56).
After doing so, the pinch-cock is replaced by a bit of glass rod.
The stand, if made of wood, is cut out accordingly.

Exactly in the same way the pipette for solid reagents, fig. 54,
can be combined with a second pair of bulbs forming a water-lute.

Arrangement and Manipulation of HempeVs Apparatus.

The general arrangement is shown in fig. 50, p. 94. The
measuring-tube A and the capillary of the pipette C, after pro-
viding both with pinch-cocks as shown, are connected by the
glass capillary E, made from a tube 18 cm. long, 6 mm. outside
diameter, and 1 mm. bore^ by bending it on each side into a right
angle, with limbs 4 or 4^ cm. long, the ends being rounded off.
The pipette is placed on a wooden bench, 46'5 cm. high, 37'5 cm.
wide, and 10*0 cm. broad.

Manipulation. — Remove the connecting capillary tube E, lift
up the level-tube B, previously filled with water, with the right
hand, and with the left open the pinch-cock of the burette A, till
it is fu^and the water begins to run out. Now connect the india-
rubber end of the pinch-cock with the aspirating- tube, already
filled with the gas, place the level-tube on the floor of the room,
and open the tap again, whereupon the water flows back into the
level-tube and the gas is drawn into the burette, Allow a little
more than 100 c.c., of gas to enter/ compress this by raising the
level-tube till the water has risen in the burette above the zero-
mark, compress the connecting-tube close to the joint with the
fingers, place the level-tube lower again, andtby cautiously loosen-
ing the elastic tube, allow the water to run out until the zero-mark
has been just reached. Then, the connecting-tube being still com-
pressed, open for a moment the pinch-cock of the burette, so that
the confined gas may be freed from the surplus pressure and assume
that of the atmosphere. In this way it is possible to get exactly
100 c.c. of gas into the burette, as will be seen on bringing the
water to the same level in both tubes. For exact measurements
it is necessary to leave some time for the water to run down (p. 31) ;
and in this case it is better not to employ exactly 100 c.c., but
such a somewhat smaller volume as may be convenient.

When the gas has been measured, we proceed to absorb such of

100 ON APPARATUS AND METHODS.

its constituents as are susceptible of the process. Connect the
burette A by interposing the capillary tube E with the pipette C}
move the pinch-cock up or down so as to leave the passage open,
raise the level-tube with the right hand and at the same time open
the pinch-cock of the pipette with the left hand. The gas now
travels from the pipette into the bulb a, driving its liquid con-
tents into the bulb b. When this has been accomplished, close
both pinch-cocks and take the pipette off. Cause the absorption to
proceed bv gently moving the pipette about, or gently (not violently)
shaking up its contents ; the absorption is generally finished in
about two minutes, but often much sooner, for instance, with carbon
dioxide. Now connect the pipette again with the capillary tube E,
place the level- tube on the floor, and, by cautiously opening both
pinch-cocks, cause the gas to re-enter the burette ; the absorbing-
liquid should at last be just allowed admittance to the end limb of
the capillary belonging to the pipette, but not to the connecting
capillary, and still less to the burette itself. With some liquids
inclined to frothing, such as the alkaline solution of pyrogallol,
this cannot always be avoided ; if, in consequence of this, the india-
rubber joints should become so slippery that the capillary tube
will not hold fast, but slips off, the joints should be washed with
water (the pinch-cocks being closed) and their ends moistened
with a little dilute acetic acid, introduced into the end of the
elastic tube.

As soon as the water in the burette has closed up, the con-
necting capillary is taken off,. the level- tube is cautiously raised
so high that both levels coincide (as in fig. 52, p. 95), and
the reading is taken after waiting two minutes for the water to
flow down. First the measuring-tube is raised with the right
hand, then the level-tube with the left, and the level of the two
liquids placed in the same plane as the eye of the manipulator.
The pipette should be previously taken off, the pinchcock is re-
moved^ and the open tubes closed with their glass rod arrangements.
In the same way a second, third, &c. constituent of the gas can
be absorbed and estimated, each time employing a different
pipette.

Applications : —

(a) Estimation of carbon dioxide in mixtures of the same with air,
or in gases of chimneys, blast-furnaces, lime-kilns, gas-producers', fyc.,

101

employing a simple absorption-pipette filled with solution of
caustic potash.

(b) Estimation of oxygen in atmospheric air, employing either a
composite absorption-pipette filled with a concentrated alkaline
solution of pyrogallol, or a tubulated absorption-pipette filled with
ammonia and small rolls of copper wire gauze, or (also in presence
of carbon dioxide) a tubulated absorption-pipette filled with
thin sticks of phosphorus and water.

(c) Estimation of ammonia, nitrous acid, nitric oxide, nitrous
oxide, chlorine, hydrochloric acid, hydrogen sulphide, sulphur
dioxide, by employing a modified Winkler's burette and simple
absorption-pipettes, to be filled for estimating : —

Ammonia with dilute sulphuric acid.

Nitrous acid .... with concentrated sulphuric acid, or with a
solution of potassium permanganate acidu-
lated with sulphuric acid.

Nitric oxide .... with concentrated solution of ferrous sul-
phate, or with potassium permanganate
acidulated with sulphuric acid, or with a
concentrated alkaline solution of potas-
sium sulphite (Divers, Chem. Zeit. 1898,
p. 1036).

Nitrous oxide. . . . with alcohol (this is only approximate ;
comp. Lunge, Bericht, 1881, xiv. p. 2188).

Chlorine with solution of caustic potash, or in the

presence of carbon dioxide by a solution
of ferrous chloride or potassium iodide.

Hydrochloric acid . with solution of potash.

Hydrogen sulphide „ „ „

Sulphur dioxide . . with solution of caustic potash or solution
of iodine.

(d) Estimation of carbon dioxide, oxygen, and nitrogen in chimney
or lime-kiln gases, §c., by absorbing, that is to say, measuring
one after another : —

1st, Carbon dioxide by solution of caustic potash.

2nd, Oxygen by alkaline pyrogallol, or copper and ammonia, or
phosphorus and water.

3rd, Nitrogen as residue.

102 ON APPARATUS AND METHODS.

(e) Estimation of carbon dioxide, oxygen, carbon monoxide,
and nitrogen in chimney-, blast-furnace-, generator-gases, fyc., by
absorbing : —

1st, Carbon dioxide by solution of caustic potash.

2nd, Oxygen by alkaline pyrogallol, or copper and ammonia, or

phosphorus and water.
3rd, Carbon monoxide by cuprous chloride and hydrochloric

acid.
4th, Nitrogen as residue.

(f) Estimation of acetylene, oxygen, and non- absorb able con-
stituents in crude acetylene, by successive absorption : —

1st, of acetylene by fuming sulphuric acid and remeasuring in

a gas- burette charged with fresh water.
2nd, of oxygen by alkaline solution of pyrogallol.
3rd, measuring the non-absorbed remainder, consisting of

hydrogen, methane, and nitrogen.

(g) Estimating carbon dioxide, ethylene (propylene, butylene),
benzene, oxygen, and carbon monoxide in illuminating -gas, gene-
rator-gas, fyc., by ^absorbing : —

1st, Carbon monoxide by solution of caustic potash.

2nd, Ethylene (propylene, butylene) and benzene, by fuming

sulphuric acid, subsequently removing the acid vapours

by employing the potash pipette.
3rd, Oxygen by alkaline pyrogallol, or copper and ammonia, or

phosphorus and water.
4tn, Carbon monoxide by ammoniacal cuprous chloride in two

pipettes (p. 74).
5th, Hydrogen -\

Methane (. remaining unabsorbed.

Nitrogen J

2. Estimation by Titration.

We have discussed this in a general way, p. 48. The com-
position of the standard liquids is indicated by a table in the
Appendix.

HESSK S APPARATUS.

103

A. Estimation by Titration of the Absorbable Constituent with
Measurement of the Total Volume of the G«?.

HESSE'S APPARATUS.

A conical absorption-bottle of strong white glass (fig. 57),
holding from 500 to 600 c.c., or
more or less, according to special
requirements, is furnished with a
mark in the neck, and is exactly
measured up to this point; the
volume is etched upon the glass.
A doubly -perforated, tightly closing,
india-rubber cork can be inserted
to that mark, the perforations, unless
closed by glass rods, serving to
introduce inlet and outlet pipes, or
the points of pipettes and burettes,
which are conveniently made 8 or
10 cm. long.

The standard solutions required,
when working in the laboratory,
are most conveniently measured by
means of stationary burettes with
inlet and outlet arrangement and
floats ; for outdoor work they are
carried in special bottles, as shown
in fig, 58, which, according to cir-
cumstances, should be provided with guard-tubes &c., and the
smaller of which can be easily replenished from the larger one by
means of the siphon.

Manipulation. — In order to take the sample of gas the conical
absorption-bottle is filled with water, a portion of which is then
displaced by the gas to be examined, whereupon the india-rubber
cork, already provided with its glass-rod stoppers,' is put in and
pressed down to the mark. If the employment of water must be
avoided, for instance, in taking a sample of air contained in the soil,
as shown in fig. 59, the india-rubber cork, provided with an inlet-
and an outlet-tube, is put into the dry empty bottle, and the gas
is drawn into it by means of a caoutchouc pump. When the
filling is complete, the end of the inlet-pipe is^ drawn out of the

104

ON APPARATUS AND METHODS.

FIST. 58.

Fig. 59,

cork, the hole in the latter is quickly closed by a glass roa, ana
the same is done with the smaller outlet-pipe, turned towards the
pump.

The abforbable constituent^ the gas^is now^estimated by means

105-

of a standard solution, employed in excess, which is run in from
a burette or pipette. The point of this is introduced into one of
the holes in the cork, one of the glass rods being removed for this
purpose and the other one being loosened in case of need. After
this the pipette is taken out and the glass rod quickly put in again.
During this operation a volume of gas escapes equal to the
volume of the standard solution introduced, which must be de-
ducted from the contents of the absorption-bottle.

The gas and the liquid are now left in contact, with frequent
gentle shaking of the bottle, till it is certain that the absorption i&
complete. In the meantime the strength of the absorbing solution
is checked, and, after taking out the cork, the excess of the
absorbent is measured by means of a second standard solution,
which, if possible, is made equal to the first, volume for volume.
When employing normal solutions, the difference found corre-
sponds to the volume of the absorbed constituent of the gas in
cubic centimetres ; and from this and from the total volume of
gas employed the percentage is found by a simple calculation, not
omitting the correction mentioned above.

This method is especially adapted for estimating small per-
centages, and yields satisfactory results.

Applications : —

1. Estimation of carbon dioxide in atmospheric air, in the
expired air, in air taken from rooms, pits, caves, walls, subsoil,
tombs, in coal-gas, fyc. — Titrated baryta- water is employed for the
absorption, normal oxalic acid for retitrating, and phenolphthalein
as indicator. The baryta-water is too changeable to be made
permanently normal, and is therefore employed empirically, but
approximately normal. Oxalic acid, which does not at all attack
the barium carbonate formed, or at least only very slowly, cannot
be replaced by any other acid. The phenolphtalein is employed
in a dilute alcoholic solution, of which only a few drops are added,
just sufficient to produce a distinct pink colour.

Example : —

Barometric pressure (B), 726 millims.
Temperature (/), 21° C.

Titre of oxalic acid : normal (1 c.c. = 1 c.c. carbon dioxide) .
Titre of baryta solution : empirical (1 c.c. = 0*88 c.c. normal
oxalic acid = 0*88 c.c. carbon dioxide).

106

ON APPARATUS AND METHODS.

Contents of absorption-bottle 618 c.c.
Baryta employed 10 c.c.

Hence : —

Air employed 608 c.c.

10 c.c. baryta- water require 8*8 c.c. oxalic acid and 1 c.c. CO2.
Required for retitra ting 6*0 c.c. „ „

Difference 2'8 c.c. „ „

Hence we have found in
608-0 c.c. air of 726 millims. B, 21° t, moist :

2*8 c.c. carbon dioxide of 760 millims. B, 0° t, dry or corrected.
525-5 c.c. air of 760 millims. B, 0° t, dry :

2-8 c.c. C02
Percentage found : 0'53 vols. CO2 per cent.

In estimating very slight percentages, for instance the carbon
dioxide contained in normal atmospheric air, it is convenient to
work with decinormal solutions. The amount of carbon dioxide is
frequently expressed not in per cent. , but in ten-thousandths. The
air mentioned in the above-given example would have contained
53 ten-thousandths. It is also usual, and very properly so, to refer
the percentage to a litre ; that is, to express the amount in
thousandths — in this case 5'3 c.c. per litre.

2. Estimation of hydrogen chloride in the gases from salt-cake
furnaces, hydrochloric-acid condensers, calcining -furnaces for the
extraction of copper by the wet process, fyc., employing a normal
silver solution for absorption, a normal solution of ammonium sul-
phocyanide for retitrating, and a solution of iron-alum as indicator.
This process can also be modified in this way, that the hydrogen
chloride is absorbed by a measured volume of solution of caustic
potash, which is afterwards acidulated with nitric acid and
titrated by Volhard's method, as just described (calculation like
that given on p. 105)*.

Cyanhydric acid can be estimated in a similar way.

3. Estimation of chlorine in the gases from chlorine-stills, from
Deacon's process, in the air of bleaching -powder chambers, fyc. — The
absorption is caused by a normal solution of arsenious acid in

* When employing sodium carbonate as absorbent, the hydrogen chloride
absorbed can be titrated by normal silver solution, potassium chromate serving
as indicator. Even a considerable quantity of sodium carbonate in excess does
not interfere with this reaction. — Translator.

107

sodium bicarbonate ; the excess employed is re-estimated by nor-
mal iodine solution, clear starch solution serving as indicator.

For estimating chlorine along with hydrogen chloride a second
volume of gas is employed, a solution of arsenious acid in sodium
carbonate being the absorbent ; this is afterwards acidulated with
nitric acid, and the total HC1 — that is, that originally present in
addition to that formed from the chlorine — is titrated as in No. 2,
with silver solution and ammonium sulphocyanide*. In calcu-
lating it must be noted that each volume of chlorine produces two
volumes of hydrogen chloride. Hence, in order to find the volume
of the hydrogen chloride originally present, twice the volume of
the free chlorine found must be deducted from the total volume of
HClf.

4. Estimation of sulphur dioxide in the gases of pyrites- kilns
and chimneys, ultramarine furnaces, glass-houses, fyc. — Absorption
is produced by a solution of sodium carbonate of arbitrary, but
not unnecessarily high, strength; a little clear starch solution is
added, and normal iodine solution run in till the blue colour
appears (calculation as on p. 105) .

B. Titraiion of the Absorbable Constituent, measuring the
Unabsorbed Residue at the same time.

a. REICH'S APPARATUS.

The absorption takes place in the three-necked bottle A (fig. 60),
holding about a litre, which is filled to about half with the absorb-
ing-liquid through the middle neck, otherwise closed by a caout-
chouc cork. One of the side necks is provided with the inlet-pipe,
drawn out to a point and bent at the end, or else provided with
many pin-hole outlets, and closed by the pinch-cock q • the other

* Or without acidifying with silver solution, the silver arseniate serving as
indicator. — Translator.

t In this case, as in the preceding No. 2, it is far more important to estimate
the weight of HC1 and 01 than the volume, both for practical purposes and to
satisfy legal requirements. It is therefore preferable to deviate here from the
rule otherwise observed in gas-analysis, by omitting the calculation into volumes
of HOI and 01 respectively, and by employing, not the " normal " solutions
otherwise meant in this book, i. e. such as indicate 1 c.c. of gas per 1 c.c. of the
reairent, but the u decinormal " solutions of ordinary titration. or else solutions
indicating per c.c. O001 grain, or parts of a grain, as the case may be.--
Translator.

108

ON APPARATUS AND METHODS.

Fur. CO.

side neck serves for the outlet-pipe, which is connected with~/the
tin aspirator B, or a glass-bottle aspirator, like fig. 13, p. 16.
Below the outlet-pipe of the latter, closed with a tap h, a glass
jar C is placed, which is divided into cubic centimetres, and serves
for holding and measuring to
half a litre the water which runs
out.

Manipulation. — Fill the ab -
sorbing- vessel A rather more than
half, the aspirator B entirely, with
water, put all corks tightly in,
close the pinch-cock q, and try
whether the apparatus is quite
tight. This is done by opening h ;
if the flow of water, continuous
at first, soon changes into slow
dropping, and at last ceases en-
tirely, there is no leakage in the
apparatus.

In order to test a gas, a suit-
able volume of absorbing-liquid
is introduced into the vessel A by
means of a pipette ; if necessary,
an indieator is also added, and the
middle cork is again put in tightly.

The aspirating-pipe is now filled up to the pinch-cock q by means
of a small india-rubber pump, and water is run off through the
tap h till the liquid standing in the inlet-pipe has just been forced
down to its point, or until a single bubble of gas has issued.
This is done in order to bring the air contained in the vessel A to
the same pressure as that prevailing during the observation. The
water which runs out is poured away, and the empty jar Cis again
put under the aspirator.

The measurement is effected by opening the pinch-cock q entirely,
and afterwards the tap h so far that the gas is just aspirated. The
gas is now passed through the vessel A in a slow stream, shaking
from time to time, till the indicator shows that the reaction is
finished. At this moment both taps are closed, and the experiment
is complete. Of course a second one may follow immediately,
after adding a fresh quantity of the absorbent ; the empty in g,

109

cleaning, and refilling of the vessel A need only be attended to at
intervals.

The quantity of water run into the cylinder C is measured.
Its volume is that of the unabsorbed residue gas, that of the
gaseous constituent absorbed' following from the quantity and
strength of the standard solution employed. This calculation is
made as follows : —

If we call the volume of the employed normal solution n c.c.,
that of the water which runs out m c.c., there would be, apart
from all the corrections : —

n = the volume of the gaseous constituent absorbed.

m — the volume of the unabsorbed residue of gas.

n + m = the total volume of gas employed for testing.

The percentage (by volume) of the constituent found by titration

, , T 100 x n
to the total volume of the gas tested would be - —

n + m

For accurate estimations we have to consider that
n means a corrected volume of gas;
m means an unconnected volume of gas.

Hence, in order to get an accurate result, m must be corrected
"by means of the formula given on p. 24, or by the table contained
in the Appendix, or by the help of the apparatus described on
p. 26, before making the calculation.

Applications : — 0

1. Estimation of sulphur dioxide in pyrites-kiln gases. — Add
.a little clear starch solution to the water contained in the absorb-
ing-bottle, and by means of a pipette a suitable volume of normal
iodine solution, and draw the gas to be tested through the
liquid till the latter is only quite faintly blue. It is not con-
venient to decolorize the liquid entirely, because the experiment
is thus very easily overdone ; should this be the case, the liquid
must be coloured faintly blue by adding one or more drops of
iodine solution before commencing a new test. Sometimes, espe-
cially when testing poor gases, it is advisable to add a little sodium
bicarbonate to the absorbing-liquid ; but in this case the bottle
should be freshly charged each time, because otherwise CO2 might
be given off and cause an error by increasing the volume of the
unabsorbed gas.

Example : —

Barometer (B), 732 millims.

110 ON APPARATUS AND METHODS.

Thermometer (t), 18° C.

litre of the normal iodine solution : 1 c.c. = l c.c. of sulphur
dioxide.

Iodine solution employed (n) 25 c.c.

Water run out (m) 295 c.c.

The percentage of SO2 will be found as follows : —

(a) Neglecting all corrections, it is

100 xn 100x25 _ _

- =- 777^ =/'81 per cent, by volume.

n + m 25 + 295

(b) Employing all corrections, we have to consider that

n= 25 c.c. at 760 millims. B, 0° t, dry.
w = 295 c.c. at 732 millims. B, 18° t, moist; or
w = 296-97 c.c. at 760 millims. B, 0° t, dry.
From this follows the corrected formula : —

100 x n 100 x 25

- = ^- ogn n^ = 874 per cent, by volume.
n + m 25+260-97

(c) An approximate correction is obtained by putting in the
volume m as directly read off, but reducing the volume n according
to the average pressure and temperature of the locality. For
instance,, at Freiburg, according to observations made during a
year, 1 c.c. is in ordinary conditions on the average actually
equal to 1*118 c.c. Hence we shall get an approximately correct
result by putting into the formula the value ft X 1*118 in lieu of
n, thus : —

100x25x1-1 18 _ *

"(25 x 1-118) +295 ~
2. Estimation of total acids in pyrites-kiln gases and analogous
gases. — Since these gases may contain a considerable percentage of
sulphur trioxide, which escapes the iodometrical estimation, Lunge
(Zeitsch. f. angew. Chem. 1890, p. 563) recommends expressing
the value of such gases not merely by their percentage of sulphur
dioxide but by that of total acids (SO2 + SO3). In such cases the
best absorbent is a standard solution of potassium or sodium
hydroxide, of which a suitable quantify is added to the water
contained in A (fig. 60). An alcoholic solution of phenolphthalein

* It is evident that considerable errors may still remain when employing this
" approximate " correction, unless at least the average temperature of the locality
is replaced by that usually prevailing at the special place where the testing-
takes place, for instance the space close to the sampling-hole in the burner-
pipe. — Translator.

Ill

(1 : 1000) serves as indicator, a few drops of which suffice to
stain the liquid a vivid red. The gas is not aspirated through it
continuously, but in small portions at a time, agitating each time-
about half a minute to secure complete absorption. Any arsenious
acid present is kept out by interposing a small glass tube filled
with asbestos. When the alkali is approximately neutralized the
red colour turns pale ; the point when the last shade of red has
vanished is easily noticed even in the dusk or with artificial light
when employing a white paper as background. This point
marks with phenolphthalein the formation of normal sulphite and
sulphate (Na2SO3 and Na2SO4) ; other indicators are not admis-
sible, as they .yield different results for sulphurous and sulphuric
acid.

If HC1 is present, it can be estimated in the liquid by titrating
the total acids by Volhard's method with silver nitrate and re-
titrating with ammonium sulphocyanide, or as described on p. 106
in the footnote.

Apart from the total acids, the percentage ofc' SO2 by itself can
be estimated in another sample of gas, as described under No. 1,,
p. 109 ; that of SO3 is found by-difference.

Example. The reduction apparatus, p. 26, shows 113*2 c.c.

(a) Estimation of sulphur dioxide.

Each c.c. of iodine solution =1 c.c. S02.

Employed 25 c.c. iodine solution (ri).

Water flowed out (w) = 320 c.c. = 282 c.c. corrected.

100 xn 100x25

8-23 vol. per cent. SO2.

(b) Estimation of total acids.
Each c.c. of caustic soda solution = 1 c.c. SO2.
Employed n c.c. soda solution = 25 c.c.
Water flowed out (m)=295 c.c. = 261 c.c. corrected.
100 xn 100x25

S°2 aS S°2 + S°3'

25 + 261

On subtracting the percentage founcj^at (a) from (b), we find
the quantity of SO3 present expressed i$ yplume percent. SO3 :

874— 8-23 = 0-51 vol. per cent. SO2 as SO3.
Hence the sulphur in the kiln-gas is present to the extent of
94*17 per cent, in the form of SO2.
5*83 „ „ „ SO3.

112

ON APPARATUS AND METHODS,

3. Estimation of nitrous acid in the gases of vitriol- chambers,
G ay -Lussac columns, §c. — The absorbent is a solution of potassium
permanganate, which is made decinormal, as the amounts in
question are only small. Before putting this into the absorbing-
vessel, the latter is rather more than half filled with dilute sul-
phuric acid. The end of the reaction is shown by the decolori-
zation of the liquid. The absorption takes place slowly and some-
times incompletely.
Example : —

Barometer (B), 728 millims.

Temperature (/), 22°.

Titre of the potassium permanganate : 1 c.c. =0-1 c.c. N2O3.

Permanganate employed 2*5 c.c. ; n = Q-25.

Water run out (w) = 410 c.c.; or, corrected, =35-61 c.c.
Hence : —

100x0-25

,- =0-0/06 per cent, by volume *.

Fig.Gl.

b. THE MINIMETRICAL METHOD.

The principle of this method was enunciated by Dr. "R. Angus
Smith ; it has beeji improved by G. Lunge (Zur Frage der Venti-
lation, 1877) and later on by Lunge & Zeckendorff. Its original
shape is shown in fig. 61.

The conical flask a (fig. 61)
serves as absorbing-vessel ; its
contents, up to a mark in the
neck, should be about 125 c.c., and
should be measured exactly and
etched upon the vessel. Its double-
perforated caoutchouc stopper,
which reaches down to the mark,
carries an inlet-pipe b} reaching
down to the bottom, and an outlet-
pipe ending just below the stop-
per. The tube b is continued into
a wider one, serving to receive

* In this case also it may be preferred to express the results in milligrams
per litre, or the like, and to choose the standard liquids accordingly (comp.
footnote to p. 107). Moreover it is for the most part not advisable to estimate
.the N203 by permanganate (comp. footnote top. 123).— Translator.

THE MINIMETR1CAL METHOD. 113

a caoutchouc valve which opens only inwards. Such a valve is
made by sliding a bit of black, strong, elastic tubing on a smooth
round piece of wood, and making in it a clean sharp longitudinal
cut,, about 2 centimetres long. The tubing is taken off from the
wood, is closed at the bottom end with a piece of glass rod, and
at the top a glass tube open at both ends is inserted, which is
carried through the perforated cork.

The tube c is connected with the pear-shaped india-rubber ball
(finger-pump) d by means of about 30 centimetres of the best
black, strong india-rubber tubing. This tube is also provided with
a longitudinal slit of 2 centimetres, forming a valve which, when
the finger-pump is compressed, can only open outwards, but
which when the pressure is relaxed closes immediately and spon-
taneously. Consequently the compressed ball, when expanding
owing to its elasticity, must receive the air necessary to refill
it through the valve b. Thus, by compressing the finger-pump
with the hand, the air contained in it is forced out of the valve c,
and by relaxing the pressure an equal volume of air is drawn
through the valve l> and through an absorbent contained in a.

The pump d serves not merely for pumping, but also as a
measuring-apparatus. We choose for this purpose a number 1
English red syringe, provided with a mouthpiece made of bone,
such as are found at all shops selling surgical instruments. These
syringes really hold 28 c c. ; when compressing them with the
hand, 23 c.c are pretty constantly forced out each time. Hence
on testing a gas it. is only necessary to count the number of times
of working the pump, . and to multiply this by 23, in order to
ascertain the volume of the aspirated gas, minus that retained
by the absorbing-liquid.

The shape ado'pted later on by Lunge and Zeckendorff (Zeitsch.
f. angew. Chem. 1888, p. 396) for their method as here described
is shown in fig. 62. The ball B here holds 70 c.c.

Manipulation. — The tube b (fig. 61) is connected by elastic tubing
with the place from which the gas is to be taken, or else the
observer takes the apparatus directly into the atmosphere to be
tested, and, first by eight or ten compressions of the finger-pump,
completely fills the apparatus with the gas in question. The cork
is raised for a moment, and a known volume of absorbing-liquid
is put into the vessel «, along with an indicator, if necessary,
whereupon the cork is again firmly pressed into the neck of the

114

ON APPARATUS AND METHODS.

flask. The volume of the absorbing-liquid employed must be
deducted from the volume of gas contained in the flask a at the
commencement of testing. The gas is now Fig. 62.

gently shaken up with the liquid, but with-
out wetting the upper part of the flask or
the cork ; then the finger-pump is com-
pressed to aspirate another volume of gas,
he liquid is shaken up again, the pump
is compressed again, and this is continued,
always counting the workings of the pump,
till the indicator shows the end of the
reaction.

The result is calculated in the same way
as indicated for Reich's apparatus, but
leaving out all corrections, as this method
can in no case claim more than an ap-
proximate degree of exactness.

If ft = the volume of the gas absorbed
( = the volume of normal solution
employed) ;

m = the volume of the unabsorbed
gas (equal to the contents of the
absorbing-vessel, less the volume of the absorbing-liquid,
but adding the number of workings of the pump, multiplied
by 23) ;
n + m = ihe total volume of gas tested;

the gas contains — - per cent, by volume of the absorbable-

71 *|~ Til

constituent.

Lunge and ZeckendorfFs method is best adapted for the rapid,,
if only approximate, estimation of small percentages. The appa-
ratus is compact, simple, and cheap. Its employment by the
original minimetrical method of estimation (for which it was de-
signed by Dr. R. A. Smith, and introduced into Germany by the
Translator in a somewhat modified form) — in which the final re-
action consists in producing a certain degree of turbidity within
the absorbing-liquid — cannot be recommended, as the degree of
accuracy obtained in this manner is quite insufficient.

Applications : —

1. Estimation of carbon dioxide in atmospheric air, in expired

THE MINIMETRICAL METHOD. 115

air, in the air of rooms, coal-pits, caves, walls, subsoil, tombs, §c.,
employing titrated baryta-water for absorption and phenolphtha-
lein as indicator. An alcoholic solution of the latter is added in
only just sufficient proportion to produce a distinct pink colour.
After each working of the finger-pump the flask should be shaken
for 25 or 30 seconds ; otherwise the absorption is incomplete.
Example : —

Titre of the baryta- water empirical, but approximately deci-
normal ; 1 c.c. = 0*104 c.c. carbon dioxide.

Total contents of the absorbing-bottle 128 c.c.

Baryta-water employed 25 c.c.; rc = 0-104x25= 2' 60 c.c. C02.

Air contained in the absorbing-bottle 128 — 25—103 c.c.

Required for decolorization : 19 syringes full at 23 c.c. each

100 xn 100x2-6 _^

= 437 : hence m = o40 c.c., or - — = = 0'47 per cent.

n+m 2-6 + 540

by volume.

2. Estimation of carbon dioxide in air fyc., by means of sodium
carbonate and phenolphthalein as indicator. — This is the method
worked out by Lunge and Zeckendorff (loc. cit.). It is founded
on the fact that a solution of sodium carbonate, stained red by
phenolphthalein, is decolorized when sodium bicarbonate has
been formed. A decinormal solution of sodium carbonate, con-
taining 5'30 grams Na2CO3 per litre, is stained red by O'l gram
phenolphthalein. Before using it 2 c.c. is diluted to 100 c.c.
with distilled water, freed from CO2 by boiling. For each test
10 c.c. of this dilute solution is employed, and the air to be
examined is blown through by means of the finger-pump until the
pink colour changes into faint yellow. With highly contaminated
air 2 to 4 charges of the finger-pump are sufficient ; pure air
requires 30 to 40 charges. The percentage of CO2 in the air
cannot be calculated from the data as above, but is found by the
following, empirically ascertained table * : —

* Fuch's (Lelimann, Prakt. Hygiene, 1900, p. 149) has shown these em-
pirically found results to be correct. But he prefers employing solutions of
twice the strength, i. e. 4 c.c. of the first solution diluted to 100 c.c. The
results agree to about ^ of their value.— Translator

116

ON APPARATUS AND METHODS.

Number
of
charges.

C02 in air
vol. per cent.

Number
of
charges.

CO2 in air
vol. per. cent. ;

Number
of
charges.

C02 in air
vol. per cent.

0-300

10 •

0-090

0-064

0-250

0-087

0-062

0-210

0-083

0-058

0-180

0-080

0-054

0-155

0-077

0-051

0-135

0-074

0-049

0-115

0-071

0-018

0-100

0069

0-042

0-066

0-038

3. Estimation of hydrogen chloride in the air of alkali-works, in
the flues and chimneys of salt-cake furnaces, in the gases from
copper -calcining furnaces, fyc., employing normal solution of
caustic potash as absorbent and a little methyl-orange as indicator.
The end of the reaction is shown by the colour changing from
light yellow into pink. When testing air very poor in HC1,
employ a decinormal solution. Manipulation and calculation as
in No. 1.

4. Estimation of total acids in dilute pyrites-kiln oases, acid smoke,
chimney- gases, fyc. is performed as in No. 2, p. 110.

5. Estimation of sulphur dioxide in dilute burner-gas, chimney -
gas, metal-smelting gas, fyc., employing normal iodine solution
as absorbent. An addition of clear starch-solution is convenient
but not indispensable. The absorption takes place easily and
quickly, without long shaking. Calculation as in No. 1, p. 114.

c. APPARATUS FOB ESTIMATING SINGLE CONSTITUENTS OCCURRING IN
MINUTE QUANTITIES.

In these cases the gas has to be brought into contact with a
suitable absorbent in the most intimate^ lasting, and repeating way.
Many apparatus have been constructed for this purpose, of which
we quote the following : —

(1) Winkler's Absorption-coil (fig. 63) consists of a spiral glass
tube A, resting on three glass feet,, and filled with the absorbing-
liquid nearly to the bulb E. Into its bottom is sealed the inlet-tube
B, provided with a bulb D and a pointed end F. From the latter
the gas issues in small bubbles, like a string of beads, along the
coils of A, and leaves the coil only after a comparatively long time
at Cj. The coil must rise gently and quite evenly ; otherwise the

ESTIMATING MINUTE QUANTITIES OF SINGLE CONSTITUENTS. 117

small bubbles unite into large ones, which lessens the contact
between the gas and the liquid. This is neglected in most of the
coils found in the trade, wherefore the following suitable dimen-
sions are given for two different sizes of coils (in millimetres) : —

Size 1. Size 2.

Width of A 22 7-5

„ „« 10 4-5

„ :, C and G! 6*5 4'5

Diameter of bulb D 35 15

„ „ E 60 30

„ coilA 200 80

Height from foot to bulb E .170 80

Fig. G3.

Good absorption- coils are excellent, especially for such cases
where the object is less the estimation than the complete removal
of a constituent, e. g. carbon dioxide from air. In such cases
size 1 is always employed.

Kyll (Chem. Zeit. 1896, p. 1006) describes a modification of
this apparatus.

(2) The Ten-bulb tube seen in fig. 64 has a very good effect, and
is recommended by Lunge (Zeitsch. f. angew. Chem. 1890, p. 567)
in preference to most other apparatus of its kind. It is very con-
venient for the purpose of estimating the absorbed constituent
volumetrically or gravimetrically.

(3) Very efficient also are the absorbing-flasks devised by Volhard
(Ann. Chem. clxxvi. p. 282), and improved by Fresenius (Zeitsch.

118

ON APPARATUS AND METHODS

f. analyt. Chem. 1875, p. 332) by the addition of another bnlb
in the lateral tube, as shown in fig. 65. The flasks are made

Fig. G4.

about 11 cm. high, and 7 cm. wide at the bottom; the mouth,
2*5 cm. wide, is either closed by a simple perforated india-rubber
cork with glass tube, or with a ground-in glass stopper. About
25 to 50 c.c. liquid is poured into the flask, which liquid under
the pressure of the gas partly enters into the lateral tube, partly
remains in a thin layer at the bottom. The absorption is still
more certain when two or three of these receivers
are employed. They have the advantage that
after finishing the absorption the liquid can be
titrated in the same flask, and cannot be forced
back by atmospheric pressure (no more than in
the ten-bulb apparatus, fig. 64).

(4) Drehschmidt's Absorbing -cylinder, fig. 66. —
The central tube, fixed in the india-rubber cork,
carries at the bottom a closed glass bulb, with
pin-holes in the upper part, by which the gas is
thoroughly divided.

General Arrangement. — The gas is aspirated by
means of air-pump S (fig. 67) ; it passes first
through the flasks A and Ax, filled with the ab-
sorbing-liquid, or in their place through a ten-bulb tube, and

ESTIMATING MINUTE QUANTITIES OF SINGLE CONSTITUENTS. 119

then into the gas-meter, G, where the non-absorbed gas is mea-
sured. Pinch-cock h serves for regulating the flow of gas so

Fig. 67.

that the suction is only just sufficient to carry the gas through
the absorbing-liquid.

In lieu of the air-pump and gas-meter an aspirator (d, fig. 68)
can be employed, and the water run out from it during the
experiment can be measured or weighed. This must be done
especially in such cases where a gas-meter would be acted upon
by acid gases or stopped up by tarry matters. Crude coal-gas,
for instance, is first passed, by tap «, through bottle b, where
ammonia &c. is retained by water; c serves for retaining hydrogen
sulphide, tar, &c., for which purpose the enlarged outlet pipe is
filled with cotton-wool. The aspirator d is best made to hold a
certain volume, say 20 litres ; its upper tap may be provided with
a micrometer-screw for regulating the flow of water, but this
can be efficiently performed by means of a screw pinch-cock placed
on the connecting-tube.

Manipulation. — The absorbent is always employed in the shape
of a standard solution, which may conveniently be a normal

120

OX APPARATUS AND METHODS.

solution, and in measured excess, which, after a certain quantity of
gas has passed through, is remeasured by means of another suit-
able standard solution. The difference, corresponding to the

Ffc. 68.

volume of standard solution saturated, shows the volume of the
absorbable gaseous constituent =n, while that which is not ab-
sorbed, m, is expressed by the quantity of water run out of the
aspirator. For exact measurements the latter must be reduced to
the standard pressure and temperature ; otherwise the calculation

of percentage is given by the formula — — . The velocity of the

current of gas passed through the absorbent is adapted to the
absorbability of the gas to be estimated, and to the capacity of
the absorbing- apparatus. Hence it may vary from 10 to 50 litres
per hour or more.

ESTIMATION OF AMMONIA IN COAL-GAS, ETC. 121

Applications : —

1. Estimation of ammonia in raw or purified illuminating -gas, in
the gases from coke-ovens, from ammonia-soda works, fyc. — The
absorbent is normal sulphuric acid, which is retitrated with normal
solution of potash ; as indicator we employ methyl-orange or
hematoxylene. The measured absorbing-liquid is put in the
cylinder b (fig. 68), and, for testing illuminating- gas, c is charged
with a solution of lead acetate to retain the hydrogen sulphide
present. If tar is also present, the gas before entering the aspi-
rator is passed through cotton-wool contained in the top part of c
For estimating the ammonia contained in crude illuminating-
gas and in coke-oven gases, 20 litres, or for purified illumina-
ting-gas 100 litres, of gas is a sufficient quantity to be tested. In
the latter case a gas-meter with an automatic shut-off arrange-
ment (p. 47) is most convenient. Since the ammonia is very
easily absorbed by the acid, the gas may be passed through
rapidly, say from 15 to 20 litres per hour ; with Drehschmidt's
absorbers (p. 118) even 60 or 70 litres per hour is allowable.

Example : —
Estimation of ammonia in purified illuminating -gas.

Barometer (B), 730 millims.

Temperature (t), 18° C.

Normal sulphuric acid employed . . . 20-00 c.c.

Normal potash solution used ..... 17*38 „

Difference (n) ................ 2-62 „

Gas passed through the meter (m), 100 litres.
The same, corrected volume, 88,216 c.c.

Percentage of ammonia following from these data : —
a. Neglecting corrections :
100 n 100x2-62

2.02 + 100,000

ns :

- bv

b. Employing the corrections :
100 n 100x2-62

2 62 + 88,216

Such small amounts are not usually expressed in per cent, by
volume, but in grammes per 100 litres of gas. The gas in question
would have contained 2'26 grammes NH3 in 100 litres.

122 ON APPARATUS AND METHODS.

2. Estimation of cyanogen and hydrogen cyanide in coal-gas. —
Drehschmidt (Journ. f. Gasbeleucht. 1892, pp. 221,268) has utilized
the fact that cyanogen * and hydrocyanic acid, even in the pre-
sence of carbon dioxide and hydrogen sulphide, are retained by
solution of caustic potash, containing freshly precipitated ferrous
hydrate, with formation of ferrocyanide, for the estimation of
both. For each test he employs 100 litres of gas and passes it with
a velocity of 60 or 80 litres per hour through two absorbing-
cylinders with bulb ends (fig. 66, p. 118) : the first charged with
,15 c.c. of a solution of ferrous sulphate (1 :10) and 15 c.c. caustic-
potash solution (1:3); the second 5 c.c. ferrous-sulphate solution,
5 c.c. caustic-potash solution and 20 c.c. water. The caustic-
potash solution is made from pure commercial caustic potash, free
from chloride, 1 part in 3 water; this solution has a specific
gravity of 178 and contains 20 per cent. KOH. An approximately
equivalent sulphuric acid is obtained by diluting pure concen-
trated sulphuric acid with four times its weight of water. The
relation of both liquids towards each other is fixed by titration.

After finishing the absorption, mix the contents of both re-
ceivers, dilute to 250 c.c., and filter though a dry filter. 200 c.c.
of the clear filtrate = 80 litres gas = 16 c.c. potash solution is
neutralized with the above-mentioned sulphuric acid ; add 2 grams
ammonium sulphate, 15 grams mercuric oxide (this is much more
than necessary), and a few drops ammonia, heat to boiling and
continue this gently for a quarter of an hour. After cooling
dilute to 300 c.c. and filter again through a dry filter. Of the
last filtrate pour 250 c.c. = 66'66 litres gas into a 300 c.c. flask,
add 6 or 8 c.c. liquor ammonise spec. grav. 0'91 and 7 grams zinc
dust, shake well up, add 2 c.c. caustic-potash solution (1 : 3), fill
up to the mark, shake up again, and filter through a double filter.
Titrate 200 c.c. of the filtrate =44*44 litres gas by adding 10 c.c.
decinormal silver solution, acidulating with nitric acid, and re-
measuring the excess of silver by decinormal ammonium sulpho-
cyanide and iron-alum as indicator. 1 c.c. decinormal silver
nitr ate = 0 002584 grams cyanogen.

Freiberg municipal gas, thus examined, required for 44'44 litres
gas 1'44 c.c. = 0'003721 grams cyanogen, and hence contained 8*37
grams cyanogen in 100 cubic metres.

* Comp. on the detection of free cyanogen in coal-gas, Kunz-Krause, Zeitschr.
f. angew. Chem. 1901, p. 652.

ESTIMATION OF NITROGEN THIOXIDE. 123

3. Estimation of nitrogen trioxide in vitriol- chamber gases fyc. —
The absorbent is concentrated sulphuric acid, 25 c.c. of which is
placed in each of the two Volhard-Fresenius flasks (fig. 65, p. 118)
employed; 10 litres or more of gas is slowly drawn through by
means of an aspirator with a pressure-gauge attached to it,
measuring the outflowing water. The liquids of both flasks are
mixed and the N2O3 absorbed estimated by one or other of the
following methods : —

(«) Part of the acid is placed in a glass-tap burette and is slowly
run into a measured volume of standard potassium permanganate
solution, strongly diluted with water of 40°, agitating all the time,
until the colour has been discharged. The reaction is :

5 N2O3 + 4 KMnO4 + 6 H2 SO4= 10 HN O3 + 2 K2S O4
+ 4MnSO4+ H2O.

From the volume of acid required and that of the standard
permanganate employed, we calculate the number of c.c. perman-
ganate (n) which would have been decolorized by the whole 50 c.c.
acid, m being represented by the volume of water run out of the
aspirator.

(b) A measured volume of acid is run into a known volume of
permanganate solution, and the excess of the latter retitrated by
hydrogen peroxide.

(c) Or else the nitrogen is estimated gas-volumetrically by the
nitrometer method, p. 33, thus converting it into nitric oxide,
1 c.c. of which is 0-0016993 grams N2O3.*

4. Estimation of chlorine takes place by absorption in a solution
of arsenious acid in sodium carbonate and retitration with iodine
solution and starch.

5. Estimation of hydrogen chloride in roasting -gases, chimney-
gases, saltcake furnace-gases, exit-gases of hydrochloric-acid con-
densers, fyc., by absorption in standard alkaline solution and re-
titration with standard acid. If other acids are present, the HC1

* It is well known that nitrogen trioxide is almost entirely dissociated into
nitric oxide and peroxide when in the state of vapour. By the above described
methods we learn how much of the equivalent mixture of NO-|-NO2 is present
in the gas. But as in most cases the chamber-gases do not contain precisely
such a mixture, but an excess of either NO or NO2, the processes quoted in the
text under (a) and (b) are not to be recommended, as they would give quite a
wrong idea of the real state of the chambers. The process (c), i. e. the esti-
mation of total nitrogen acids by means of the nitrometer, is the only one which
should be employed. — Translator.

124

ON APPARATUS AND METHODS.

may be titrated in the same liquid by means of silver nitrate and
ammonium sulphocyanide (p. 106) *.

If chlorine is present together with hydrogen chloride, the gas is
passed through a solution of arsenious acid in sodium bicarbonate.
In one part of the liquid the chlorine is estimated by retitration
with iodine, in another the total HC1 as in No. 6 f. Since 1 vol.
chlorine furnishes 2 vols. HCl, the total volume of HCl found
must be diminished by twice the volume of free chlorine, to find
the HCl originally present.

6. Estimation of total acids in poor gases from the manufacture
of sulphuric and hydrochloric acid, sulphite cellulose, ultramarine,
glass, superphosphate, from roasting -furnaces, hop-driers, brick-
works, fyc. — For very poor gases the simple Reich's apparatus
(p. 107) does not suffice for retaining all the acids with certainty.
It is preferable to use a ten-bulb tube (p. 118), charged with a
known volume of standard alkali and connected with a 10 or
15 litre water-aspirator. The gas is slowly drawn through the
liquid, which is then washed into a beaker and retitrated with
standard acid and methylorange. 1 c.cm. of decinormal soda-
solution (4-CGO grams NaOH per litre) shows 0-0040 gr. SO3j
0-0049 H2SO4, 0-0032 SO2, 0*00365 HCl, 0'0020 HP, &c. The
result is expressed in grams total acid per cubic metre of the gas,
taken at 0° C. and 760 mm. pressure %. The " total acids " are
reduced to one special compound — e. g., in the manufacture of
sulphuric acid they are expressed in terms of SO3, although com-
prising SO2, SO3, H2SO4, and the acids of nitrogen ; in the manu-
facture of saltcake and hydrochloric acid in terms of HCl. In
Prussia and Saxony (comp. Chemische Industrie, 1898, p. 535)
the total acidity of gases from the manufacture of sulphuric acid
or sulphur trioxide is limited to a maximum of 5 grams SOS per
cubic metre, except in factories working with blende, where the
utmost limit is 8 grams, always taking the sample before the gases
enter the chimney. In Great Britain the limit is 4 grains SO-.

* Or "by silver r.itiate ar.d potassium chroirate in the neutralized liquids
(see footnote to p. 1Q6).— Translator.

f In this case no addition of potassium chromate is necessary for titrating
the neutralized liquid with silver nitrate, as the sodium arseniate is quite as
good an indicator. — Translator.

| In Great Britain it is expressed in grains per cubic foot. 1 grain per cubic
fcot = 2'287 grams per cubic metre ; 1 gram per cubic metre = 0*4372 grain per
cubic foot. — Translator.

ESTIMATION OF GASES BY WEIGHT. 125

per cubic foot at 60° F. and 29 inches pressure = 9* 15 SO3 grammes
per cubic metre at 150<5 C. and 760 mm. pressure.

Example : —

Estimation, of total acidity of the exit-gases of the manufacture
of sulphuric anhydride.

Employed decinormal sodium hydrate solution

(diluted in ten-bulb tube to 125 c.c.) . . 50'00 c.c.
Employed in retitration : decinormal acid . . 38'05 „
Neutralized by total acids ....... 11*95 „

= 1 1-95 x 0-0040 = 0-0478 gram SO3.
Water run out of the aspirator, 11-320 litres.
Gas-reduction apparatus shows 111*3 c.c.

Hence corrected volume of gas employed =10' 170 litres.
1 cub. met. of gas contains total acids (S02, SO3, and H2SO4,
in terms of SO3) :

0-04780 X 1000

= 47° grams'

3. Estimation by Weight.

The estimation of gases by converting them into compounds
capable of being weighed is only made in exceptional cases, espe-
cially those in which the constituent to be estimated is present in
very slight quantity } and where volumetric methods cannot be
employed. The construction and manipulation of the absorbing-
apparatus are the same as described on p. 116 et seq., and the calcu-
lation of the results is generally made in the same way as is there
indicated.

Applications : —

1. Estimation of hydrogen sulphide, carbon disulp hide, and acetylene
in illuminating-gas. — The current of gas, to be measured by a
meter or an aspirator, before entering these passes through two
Volhard's absorbing- apparatus (fig. 65, p. 118), each of them
containing 25 c.c. of a concentrated ammoniacal solution of silver
nitrate, then through a combustion-tube of about 25 centimetres
length, filled with platinized asbestos * and heated to an incipient
dark red; finally, again through two Volhard's apparatus, each
containing 20 c.c. of ammoniacal silver solution. For greater
security three absorbing- vessels may be employed before and behind

* See the preparation of this later on (p. 140).

126 ON APPARATUS AND METHODS.

the combustion-tube, in lieu of two. For each test 100 litres of
gas should be employed, and 10 to 12 hours should be allowed
for passing them through.

The contents of the two receivers placed in front of the com-
bustion-tube after some time assume first a whitish, then a darker
turbidity, caused by the precipitation of silver acetylide and
sulphide. These receivers absorb the acetylene and the hydrogen
sulphide.

Carbon disulphide and other sulphur compounds present in
coal-gas, on passing through the combustion-tube and coming
in contact with the hot platinized asbestos, are changed into
hydrogen sulphide, which is absorbed in the following receivers,,
and causes a blackish-brown precipitate of silver sulphide.

After finishing the operation, the contents of the first two
receivers on the one hand, and those of the last two receivers on
the other, are united ; each of the two precipitates is filtered
and carefully washed with water. The precipitate originally con-
tained in the first receivers is covered on the filter with dilute
hydrochloric acid, which process must be performed cautiously,
keeping the funnel covered with a watch-glass. Acetylene is given
off with slight effervescence, the precipitate being converted into a
mixture of silver chloride and silver sulphide. After washing it,
the silver chloride is extracted by a little dilute ammonia, re-preci-
pitated by saturating the filtrate with nitric acid, and wreighed in
the usual manner. From the weight of this precipitate that of the
acetylene may be deduced by means of the formula (founded on the
research by E. H. Keiser, Amer. Chemical Journ. xiv. p. 285) : —
C2Ag2 + 2HC1= 2AgCl -f C2H2.

1 gram AgCl corresponds to 0*0907 gram acetylene^ 78*12 c.c.
in the normal state.

The silver sulphide, which is insoluble in ammonia and has
remained on the filter, corresponds to the hydrogen sulphide
originally present. Examination has proved that it contains no
free silver ; hence the precipitate, after burning the filter, can be
at once converted into metallic silver by igniting in a current of
hydrogen.

1 gram of silver corresponds to 0*1486 gram S, or 0*1579 gram
H2S, or 103*78 c.c. H2S in the normal state.

The silver sulphide found in the receivers placed behind the
combustion - tube has been produced from the other sulphur

ESTIMATION OF GASES BY WEIGHT. 127

compounds present in illuminating-gas, as carbon disulphide, phenyl
tulphocyanide, &c. It is converted in the same way into metallic
silver, which is weighed and calculated as carbon disulphide, which
is the predominant compound. 1 gr. of silver corresponds to
0-1486 gr. S, or 0'1764 gr. CS2, or 52'12 c.c. CS2 in the form of
gas in the normal state.

It is not usual to express the percentage of H2S and CS2 in
coal-gas by volumes, or as weights of these compounds, but merely
to indicate the number of grams of sulphur contained in 100
cubic metres of gas (in England in grains per cubic foot), that is
the total sulphur contained in the illuminating -gas. This is generally
done by burning a known volume of the gas, receiving the products
of combustion in a solution of potassium carbonate containing a
little bromine, and precipitating the sulphuric acid formed by
barium chloride 1 gr. BaSO4 = 0'1373 gr. S. Drehschmidt (Cheni.
Zeit. 1887, p. 1382) and F.Fischer (Zeitsch. f. angew. Chem. 1897,
p. 302) describe special apparatus for this purpose. Since the
gases in question occur in coal-gas only in minute quantities,
their volumes need not be counted when calculating the results,
the unabsorbed gas measured in the meter or aspirator being
assumed as equal to the total volume of gas tested.
Example : —

Barometer (B), 733 millims.

Temperature (t), 18° C.

Volume of gas employed, 107 litres.

The same corrected, 94,787 c.c.

Found by weighing : — •
AgCl=0'3190 gr. = 24-92 c.c. acetylene.
Ag a =0-0111 gr.= 1"15 „ hydrogen sulphide.
Ag b = 0*3888 gr. — 20'26 „ carbon disulphide.

Total sulphur: —
Silver 0 = 0-0111 gr. = 0'001647 gr. S.

„ 6 = 0-3888 „ =0-057765 „
100 cubic metres of gas contain 62'68 grams sulphur *.

Expressed in per cent, by volume :—

Acetylene 0'0262J p. c.

Hydrogen sulphide 0-00121 „
Carbon disulphide 0'02126 „

* 1 gram per cubic metre=0>4372 grains per cubic foot.-- Translator.

128 ON APPARATUS AND METHODS.

2. Estimation of sulphuretted and phosphoretted hydrogen in crude
acetylene. — The sulphur in acetylene-gas exists mostly in the shape
of organic sulphur compounds, which have been separated from it
by Knorre and Arndt (Verh. Gewerbfleiss, 1900, p. 155). It is,
however, admissible to express them in terms of H2S or in grams
S per cubic metre. Rossel & Landriset (Zeitsch. f. angew. Chem.
1901, p. 77) found in acetylene varying quantities of sulphur,
dependent both upon the quality of the calcium carbide and
the process employed for evolving the acetylene. They found per
100 cub. metres from 21 to 111 grams S=0'01466-0'07746
vol. per cent, of H2S ; and in 100 cub. metres a maximum of
73 grams phosphorus.

Lunge & Cedercrentz (Zeitsch. f. angew. Chem. 1897, p. 651)
liave indicated the following process for estimating both impurities
at the same time : — A known volume of the gas is slowly passed
through a ten-bulb tube (p. 118), charged with 2 or 3 per cent,
solution of sodium hypochlorite. The liquid is washed into a
graduated flask, and in one half of it the sulphuric acid is gravi-
metrically estimated as barium sulphate (1 gr. BaSO4 = 0'1373 gr.
S = 0'1459 gr. H2S = 95-86 c. c. H2S), in the other half the
phosphoric acid as magnesium pyrophosphate (1 gr. Mg3P3O7=
0-2784 gr. P=0'3055 H3P = 200'91 c. c. H3P).

3. Detection and approximate estimation of very small quantities
of sulphur dioxide and sulphuric acid in air, suspected of being
contaminated with acid smoke. — Ost (Chem. Zeit. 1896, p. 170) and
H. Wislicenus (Zeitsch. f. angew. Chem. 1901, p. 689) chemically
fix the acid contained in the suspected air of forests &c. by
exposing to it for a long time wooden frames, of a superficial area
of one square metre, covered with loose cotton tissue, impregnated
with barium carbonate by moistening with baryta- water. This
gives an idea of the quality of soot present, and later on, by
incineration and estimation of the sulphate contained in the ash,
the quantity of sulphur acids present in the air. For the con-
clusions to be drawn from this process (which is not yet fully
worked out) we must refer to the original.

ESTIMATION BY COMBUSTION. 129

III. ESTIMATION OF GASES BY COMBUSTION.
1. General Remarks on the Combustion of Gases

Those constituents of a gaseous mixture which cannot be
estimated by absorption, owing to the want of a suitable reagent,
are, if possible, transformed by combustion with oxygen into
compounds capable of being condensed or absorbed. Hereby
both the combustible gas and the oxygen are removed, causing a
contraction of volume from which the volume of the combustible
gas can be deduced, as the combustion always takes place in
definite proportions of volumes.

A further contraction of volume, which stands in definite
proportion to the volume of the gas burned and thereby admits
of estimating the latter, is produced by absorbing any carbon
dioxide formed during the combustion.

The oxygen required for this purpose is only exceptionally
employed in the pure state in technical gas-analysis, mostly in
the shape of air, and of course always in moderate excess.

Only three gases need be considered which cannot be estimated
absorptiometrically : —

Hydrogen, to be burned by oxygen to liquid water.
Methane, to be burned by oxygen to liquid water and gaseous,
but absorbable, carbon dioxide.

Nitrogen (plus Argon, &c.), not combustible, remaining at the
end of the analysis as gaseous and directly measurable residue.

The hydrogen, both free and as a constituent of methane,
yields liquid water, because the gas to be analyzed is previously
saturated with moisture.

Suppose we have, as described p. 1025 successively removed by
absorption carbon dioxide (by caustic potash), heavy hydrocarbons
(by fuming sulphuric acid), oxygen (by alkaline pyrogallol, &c.),
and carbon monoxide (by ammoniacal cuprous chloride), we find a
non- absorb able remainder of gases which we measure. We then
transfer it to a HempePs or B ante's burette, and employ it entirely
or partially for the estimation of hydrogen, methane, and nitrogen,
or some of these gases. First we must add a known volume of
oxygen or air, enough to suffice for the combustion. In order to
ascertain this, we assume the gaseous remainder to consist entirely
of combustible gas, neglecting the nitrogen, and taking the oxygen
in air roughly =20 per cent, by volume.

130 ON APPARATUS AND METHODS.

Hydrogen, when burning, acts thus : H2 -1- O = H2O ; hence
3 vols. gas (2 vols. H -f 1 vol. O) = 0 vol. liquid water.

The contraction, K, is = 3 vols., and of these 2 vols. had been
previously present as hydrogen ; therefore

TT ~K i

H = —_vols.

If the gas to be burned is sure to contain no methane, we must
add to each 2 vols. of it 1 vol. oxygen or 5 vols. air, i. e. for each
c.c. of gas 2'5 c.c. air.

Methane gives the following reaction :

Hence 2 vols. CH4-f 4 vols. O = 6 vols. gas furnish on com-
bustion 2 vols. CO2+ 0 vol. liquid water.

The contraction K, occurring on combustion, is :

K = 6- 2 = 4 vols. ; hence CH4 = ? vols.

If the carbon dioxide, whose volume is equal to that of the
methane burned, is afterwards removed by absorption in caustic
potash, the above 6 vols. gas (2 vols. CH4 + 4 vols. O) vanish
altogether, and the total contraction Kx is =6 vols. or CH4 =

Hence to each 2 vols. of a gaseous remainder containing
methane we must add before combustion (without troubling about
any hydrogen or nitrogen present) 4 vols. oxygen or 20 vols. air;
but this is a minimum quantity, which is best somewhat exceeded.

Besides hydrogen and methane it would be possible to estimate
other combustible gases by this method, applying the contractions
shown in Table 5 of the Appendix. But these gases are more
suitably estimated absorptio metrically, as far as this is possible ;
combustion is at most only applied to the detection of small
quantities, which cannot be directly measured by weighing or
titrating the products of combustion.

Just as hydrogen can be burned by oxygen and thus determined,
vice versa the percentage of oxygen in a gaseous mixture can be
ascertained by mixing it with a measured excess of hydrogen,
effecting the combustion, and calculating the oxygen from the
contraction.

COMBUSTION BY EXPLOSION. 131

2. Methods of Combustion.
A. Combustion by Explosion.

The inflammation of an explosible gaseous mixture, suitably
confined, by the electric spark, for the purpose of estimating
one or the other of the gases taking part in the explosion by
means of the subsequent contraction, is the oldest of all combus-
tion methods for gases. It was applied by Volta to eudiometry :
i. e.j the deflagration of a measured volume of air with a measured
excess of hydrogen in a measuring-tube luted by mercury (the
eudiometer), and ascertaining the contraction which takes place.
This method has since been generally introduced into the exact
analysis of gases, and has been greatly improved by Bunsen
(Gassmetrische Methoden, 2nd edition, 1877) and by Hempel
(Gasanalytische Methoden, 3rd edition, 1890, p. 114). It now
serves both for the estimation of hydrogen and of methane.

The explosion method is undoubtedly attractive by its neat
character, but it is not free from drawbacks. Not every gaseous
mixture can be straightway brought to explosion ; it is sometimes
necessary to add electrolytic oxyhydrogen gas, or, in the presence
of oxygen in excess, pure hydrogen. Nor can the simultaneous
combustion of a little nitrogen be always avoided; according to
Armand Gautier (Chem. Zeit. 1900, p. 586) this collateral
reaction is not constant, but varies with the other conditions.
Moreover, the explosion method requires mercury as a confining
liquid, as well as a somewhat bulky apparatus of galvanic battery,
oxyhydrogen generator, and induction-coil, which does not make it
appear very convenient for technical analysis in a general way.
Seger's eudiometer with water-seal and india-rubber taps (Thon-
industrie-Zeitung, 1878, nos. 25 & 26) has never become popular ;
and Hempel, who had formerly tried to make the explosion
method more suitable for technical analysis by employing an
explosion-pipette, provided with electrodes for the evolution of
oxyhydrogen gas and charged with caustic- potash solution (Neue
Methodeu zur Analyse der Gase, 1880, p. 156), has abandoned
this himself. There is no better prospect for the combination of
the explosion arrangement with an Orsat's apparatus, proposed
by Thorner (Chem. Zeit. 1891, p. 763).

Many of the difficulties cited have been removed by Hempers
new explosion-pipette charged with mercury, as shown in fig. 69.

K 2

]32

ON APPARATUS AND METHODS.

It consists of two strong tubulated glass bulbs, a and b, mounted
on stands and at the bottom connected by an elastic tube covered
with canvas. The explosion-bulb a is contracted at the top, like
an ordinary gas-pipette, into a siphon-like capillary, closed by a
pinch-cock or glass rod, and at the bottom carries a glass tap h
which is connected with the level-bulb b by the aforesaid elastic
tube. At c two thin platinum wires are sealed into the contracted
part of the bulb a, which wires are 2 millimetres distant from

Fig. 69.

each other, so that an induction-spark can pass between them.
For this purpose, the outer ends of these platinum wires are turned
into loops and are connected by silk- covered copper spirals with
the induction-apparatus J, which receives its current from the
dipping-battery T or any other source of electricity. Both bulbs
of the pipette are rather more than half filled with mercury ; if

COMBUSTION BY EXPLOSION.

133

b is lifted, tap h being open, a is filled witli mercury up to the
capillary,, and is kept in this state by closing tap h.

Manipulation. — A suitable volume of the gas to be burned is
roughly measured off in the measuring-tube A of a Hempers
burette ; the level-tube B is placed on the floor, the water in the
burette is allowed two minutes to flow down, and the exact
reading is now taken. Tube B is again lowered and the pinch-
cock of A is opened until the water has descended nearly to the
bottom mark and a corresponding quantity of air has entered
into the burette. After waiting again for two minutes, the
second reading is taken and the volume of the gaseous mixture
thus ascertained. Since 2 vols. hydrogen required 5 vols.
air, 100 c.c. of the mixture should not contain more than

lOOx 2
— = — = 28'57 c.c. of combustible gas, but of course this utmost

limit should never be attempted, and only about 25 c.c. of
combustible gas should be employed in the case of hydrogen.

In the case of methane, 2 vols. require 20 vols. air; hence
100 vols. in the burette ought not to contain more than

— ^-^ = 9'09 c.c. of combustible gas.

<w<w

If the gaseous remainder to be analyzed contains too much

Fig. 70.

nitrogen to explode in a mixture
with air, a sufficient amount of
pure hydrogen must be added.
This is best kept in stock in a
Hem pel's simple hydrogen-pipette,
fig. 70. It is like an absorption-
pipette for solid reagents (pp. 54,
97) , into the bottom neck of which,
a, a perforated zinc cylinder e
has been introduced by means
of a central glass rod passing
through the cork. Bulb b con-
tains dilute sulphuric acid. After
all air has been expelled from
the apparatus, the capillary is
opened, whereupon hydrogen
issues from it and is carried over H
into the gas-burette in the well-
known way. If the capillary is closed again, the hydrogen

134

ON APPARATUS AND METHODS.

expended is renewed and forces the acid out from the cylinder
into bulb b.

Hempel has also constructed a composite hydrogen-pipette,
fig. 71. Two bulbs, a and a^ are connected; a is filled with
pure zinc, mixed with cuttings of platinum foil, and is closed by
an indiarubber-covered glass rod put into the bottom neck e.
b contains dilute sulphuric acid (1 : 10), introduced through the
capillary by means of a long funnel-tube, during which process
bulbs b and c are filled with hydrogen. At last a little mercury
is poured into d, but for ordinary purposes this can be replaced
by water.

Fig. 71.

The gas given off in these pipettes is never absolutely pure
hydrogen ; it contains a slight amount of air, but this does not
affect its use.

As soon as the mixture of the combustible gas with air, and in
case of need also with a measured quantity of hydrogen, has been
made, the explosion can be effected. The explosion-pipette C
(fig. 69) is placed on a stand D, bulb a is filled with mercury
by lifting b, and tap h is closed. The capillary of the pipette is
connected by means of the capillary E with tube A of the gas-
burette, tap h is opened, and by lifting the level-tube B, the
pinch-cocks being opened, the gaseous mixture is transferred
into the explosion-bulb «, whereupon the taps are again closed.
Before closing tap h, it is best to lower bulb b and thus to
produce a partial vacuum in a ; but if the volume of gas in a is
not large and it is not highly explosive, tap h may even be left

ESTIMATION OF HYDROGEN BY EXPLOSION. 135

open. Now the battery T is put in motion, the current is closed,
and at once the explosion occurs with a flash, the mercury being
agitated and covered with a film. The gas is then re- transferred
from bulb a into the burette A, and after the water has run down
the contraction is ascertained.

Applications : —

1. Estimation of hydrogen in the absence of other gases. — In
order to learn the manipulation of the method, 20 or 25 c.c.
hydrogen from the hydrogen-pipette is transferred into a gas-
burette, air is admitted nearly up to 100 c.c., both readings being
accurately made, the mixture is transferred into the explosion-
pipette, the current is closed, the gas re-transferred into tiie burette
and the contraction read off.

Example : —

Hydrogen employed 20'4 c.c.

Hydrogen + air 96*2 „

Hence air alone 75'8 „

Containing oxygen 15'2 „

Oxygen required by theory 10'2 „

Excess of oxygen 5'0 ?,

Volume of gas after explosion ... 65*9 „

Contraction 96'2— 65'9 . =30'3 ,

Found : —

30-3 x 2

= 20-20 c.c. hydrogen.

2. Estimation of hydrogen in the presence of other gases, but in
the absence of methane, e. g. in non-carburetted water-gas. — Carbon
dioxide and monoxide are successively removed and estimated
(p. 102), a portion of the gaseous remainder is mixed with at least
2J times its volume of air, the mixture having been measured is
introduced into the explosion-pipette, and the experiment finished
as above.

Example : Examination of water-gas.

Volume employed 99'8 c.c.

After treatment by potash 957 c.c.

Contraction 4'1 „ =4'12 p. c. CO2.

After two treatments by ammo- ]
niacal cuprous chloride J

Contraction ... 957-56*0 = 397 „ =3978 p. c. CO.

136 ON APPARATUS AND METHODS.

Estimation of hydrogen. — Since the gaseous remainder = 39'7
is too large to admit of adding a sufficient quantity of air, only a
portion of it is employed, viz. 24'2c.c., corresponding to 43'13c.c.
of the original gas.

Gas employed 24*2 c.c.

Gas + air 98*3 „

Air alone 74*1 „

Containing oxygen 14'8 „

„ nitrogen 59'3 „

Volume of gas after explosion 65*9 „

Contraction 98'3 — 65'9 = 32'4 „

Corresponds to hydrogen (of the gas)... 21'6 „ =50'08 p. c.

„ „ oxygen (from air) 1O8 „

„ ,, non-combustible gas... 65 '9 „

Estimation of nitrogen. — The nitrogen contained in the gas is
found by the difference between the volume of non-absorbed gas
and the volume of hydrogen found by combustion.

Non-absorbed gas ( = 43-13 c.c. of the original gas) . . . 24*2 c.c.

Hydrogen contained therein 21-6 „

Remainder 2-6 „

= 6'02 per cent, nitrogen.

Result : — Carbon dioxide 4'12 per cent, by volume.

Carbon monoxide ... 39'78 „ „

Hydrogen 5O08 „

Nitrogen . 6'02

100-00

3. Estimation of hydrogen and methane occurring together,
e. g. in coal-gar, producer-gas, coke-oven gas, &c. — The absorbable
gases are successively removed and estimated as shown on p. 102,
8 to 15 c.c. of the remainder (according to whether there is more
methane or more hydrogen present) is transferred into a Hempel's
burette and measured, air is added nearly up to 100 c.c. and
measured again, the mixture is transferred into the explosion-
pipette and after explosion the contraction is ascertained. Now
the gas is treated in the caustic-potash pipette and the contraction,

ESTIMATION OF HYDROGEN AND METHANE.

137

caused by the removal of the CO2, is ascertained. The latter allows
of calculating the methane; by doubling the volume of the latter
we learn the contraction caused by the combustion of methane,
and by deducting this from the total contraction we find the con-
traction caused by the hydrogen, which we multiply by §, in order
to find the hydrogen.

A check test should be made to see whether sufficient air had been
employed for combustion, by transferring the last remainder of gas
to a phosphorus or pygrogallol pipette, which ought always to
show that an excess of oxygen has been present.

Example : Analysis of coal-gas.

Gas employed 997 c.c.

After treatment by potash 95 '9 c.c.

Contraction 3!8 „ =3*81 per cent, carbon

After treatment by fuming sul- dioxide,

phuric acid and removal of

the acid vapours by potash . 91*2 ,,

Contraction 4' 7 „ =4'71 per cent, heavy

After treatment by alkaline hydrocarbons.

pyrogallol 90'6 „

Contraction O'G „ =O60 per cent, oxygen.

After treatment by ammoniacal

cuprous chloride 80' 7 „

Contraction 9-9 „ =9'93 per cent, carbon

Non-absorbed gas 80' 7 „ monoxide.

Estimation of hydrogen and methane.

Non-absorbable gas employed

( = 15-07 of original gas) 12*2 c.c.

Gas plus air 99'0 c.c.

Hence air alone 86*8 „

Containing oxygen 17'4 „

„ nitrogen 69*4. „

Volume after explosion 79*0 „

Total contraction 99-0-79'G ... 20'0 „

After treatment with potash 74'4 „

Contraction (COo) 4-6 „ = 4'6 c.c. CH4.

138

ON APPARATUS AND METHODS.

Contraction caused by combus-
tion of methane 4'6 x 2 . .

Contraction caused by combus-
tion of hydrogen 20'0-9'2 ...

Estimation of nitrogen.
Unabsorbed gas employed ( =
15'07 c.c. of original gas, as

before)

Containing methane . . . 4'6 c.c.
„ hydrogen ... 7'2 „

leaving a remainder of ...

Results : —

Carbon dioxide

Heavy hydrocarbons

Oxygen

Carbon monoxide
Methane
Hydrogen
Nitrogen

9'2 c.c. = 4-6 c.c. CH4.

= 30-52 per cent,
methane.

10-8

12-2 c.c.

11-8
0-4

10-8x2

= 7'2 c.c.

47'78 per cent,
hydrogen.

= 2'65 per cent,
nitrogen.

3'81 per cent, by volume,
4-71

100-00

4. Estimation of methane in the absence of hydrogen, e. g. in
fire-damp. — To the mixture of methane and air a measured volume
of pure hydrogen must be added from a hydrogen-pipette ; if
there is too little oxygen present, more air (measured) is added,
the whole is transferred to the explosion-pipette and the current
closed. After this the gas is carried back into the gas-burette,
measured, and the carbon dioxide formed (whose volume is equal
to that of the original methane) is estimated by means of the
caustic-potash pipette. This is safer than calculating the methane
from the contraction after explosion, as the hydrogen from the
pipette-is never pure.

COMBUSTION BY PALLADIUM. 139

Example : —

Gas employed 85*1 c.c.

G as + hydrogen 95'4 „

Hydrogen alone 103 ,,

Gas after explosion 70*5 „

After absorption by caustic potash 65*7 ,,

Contraction ( = CO2) 4'8 .,

Found: — 4'8 c c. = 5*63 per cent, methane.

B. Combustion by means of gently -heated Palladium.

Several metals of the platinum group, as platinum, indium, and
especially palladium, have the property of causing the combustion
of various gases by oxygen at a temperature below the point of
inflammation. This property is exhibited to the greatest extent, the
finer the state of division and consequently the greater the surface
offered by those metals to the gases. Especially easy and com-
plete is the combustion of hydrogen, if mixed with a sufficient
quantity of air and carried over gently heated, finely divided palla-
dium. Under the same conditions carbon monoxide, ethyleue, and
benzene are burned with a little more difficulty, but without giving
any trouble. Methane, however, whose temperature of inflamma-
tion is very high (about 790 °), remains unchanged at moderately
low temperatures. From this follows the possibility of estimating
the more easily burning gases in the presence of methane, on the
principle of fractional combustion ; and this is especially practicable
for estimating hydrogen in the presence of methane, which two
gases form the combustible remainder from the absorptiometrical
analysis of gaseous mixtures.

The first to apply fractional combustion was W. Henry (Annals
of Philosophy, xxv. 428), who employed spongy platinum heated to
177° C. Bunte (Berl. Berichte, 1878, xi. 1123) employed mode-
rately heated palladium wire ; Hempel (ibid. xii. 1006) superficially
oxidized palladium sponge at a temperature of 100° C. The latter
subsequently (Gasanal. Methoden, 1900, p. 159) proposed to utilize
the property of finely divided palladium to retain hydrogen by
occlusion, for retaining that gas without any addition of air. Tho
author himself a long time ago proposed palladium-asbestos as an
agent for fractional combustion, and up to this day he prefers that
method as worked out by himself in the following manner.

The combustion-apparatus consists of a short glass "capillary

140

ON APPARATUS AND METHODS.

tube, bent at each end in a right angle, into which a fibre of
asbestos, impregnated with finely divided palladium, has been
loosely introduced, so that it does not impede the passage of a
current of gas.

Palladium-asbestos is prepared in the following way: — Dissolve
1 gram palladium in aqua regia, evaporate the solution to dry-
ness on a water-bath, so as to remove any adhering hydrogen
chloride as completely as possible, and dissolve the palladium
chloride thus produced in a very little water. To this add a
few cubic centimetres of a cold saturated solution of sodium
formiate and sufficient sodium carbonate to produce a strongly
alkaline reaction. Now introduce 1 gram of very soft, long-fibred
asbestos, which, if any unnecessary excess of water has been
avoided, absorbs the whole liquid and forms with it a thick paste.
This is dried at a gentle heat, by which process black, finely
divided palladium is uniformly precipitated upon the asbestos-fibre.
In order to make the palladium adhere,, the asbestos thus prepared
must be heated en a water-bath till completely dry, then soaked
in a little warm water, put into a glass funnel, and freed from all ad-
hering salts by thorough washing, without removing any palladium.
After drying, the substance exhibits a dark grey colour, having a
slight tendency to stain the fingers, and contains 50 per ceut. palla-
dium. It possesses a very high degree of chemical activity ; in
the perfectly dry state it can cause the combination of hydrogen
and oxygen even at the ordinary temperature, but in order to
secure this result it is always employed in the heated state. The
same process is employed for producing platinum-asbestos, required
for other purposes, but it is sufficient to make this with from
10 to 25 per cent, of platinum.

For the preparation of the capillary combustion-tubes we employ
capillary glass-tubing of about 1 millim. bore and 6 millim.
outside diameter, cut in pieces 16 or 18 centimetres long. The
asbestos-fibre must be introduced into them before bending off the
end, in the following way : a few loose fibres of the palladium-
asbestos are laid alongside each other on smooth filtering-paper iip
to a length of 4 centimetres ; they arc moistened with a few drops
of water, and, by sliding the finger over them, are twisted into a
fine straight thread,, which in the moist state has the thickness of
stout sewing-cotton. This thread is grasped at one end with the
nippers, and, without bending or nicking, is slid from above into

COMBUSTION BY PALLADIUM. 141

the capillary tube, which is held vertically. This is then, filled
with water by means of the washing-bottle, and by jerking or by
drawing off the water at the ends the asbestos-thread is brought
into the centre of the tube. This is now allowed to dry in a warm
place ; the two ends are bent off at right angles for a length of
3*5 or 4 centimetres, and the edges rounded off with the lamp.

The measuring-apparatus (fig. 72) is a Hempel's burette, A, with
a simple absorbing-pipette, C. The latter is filled with water and
has a brass tube, G, fixed to the back of the stand, but movable
in different directions and ending in a small steatite burner. This
tube is connected with the gas-supply by an elastic tube, and
serves for producing a small gas-flame F*. By means of this,
the capillary combustion-tube E, placed between A and C, can

Fig. 72.

be heated at will ; if the heating is to be interrupted, the flame
need not be put out, but the tube b need only be moved a little
backwards.

Manipulation. — The volume of the combustible gas contained
in the burette A is read off; it should in no case exceed 25 c.c.

* Of course this gas-jet may be conveniently replaced by the small spirit-
lamp attached to Lunge-Orsat's apparatus, fig. 73, p. 147. — Translator.

142 ON APPARATUS AND METHODS.

The level-tube is placed on the floor of the room, and by opening
the pinch-cock^ enough air is admitted to bring up the total volume
of the confined gases nearly, but not quite, to 100 c.c. When
all the water has run together, the volume is carefully read off.
The capillary tube E is now interposed between the burette A and
the pipette (7, and heated for one or two minutes by means of the
small gas-jet F. The heating should be only slight, and should
in no case rise till the tube is at a visible red heat, still less till
it softens. The combustion may now begin. The level-tube is
elevated, the pinch-cocks are opened, and the gaseous mixture is
conveyed in a low stream through the heated palladium-asbestos
into the pipette C. The end of the asbestos- thread first meeting the
gaseous current begins to glow visibly, and this glowing frequently
reappears when conveying the gas back into the burette in the
same way. During the whole operation the gas-jet is left burning
under the capillary t ube ; otherwise care must be taken lest the
gas should pass too quickly and lest any drops of water should get
into the heated part of the capillary tube, which would thereby be
sure to crack. With easily combustible gases the combustion is
usually finished by two passages forward and backward; but in
any case it is necessary to be convinced that no further decrease
of volume takes place by another passage. The residue of gas
ultimately obtained is measured, and the contraction which has
taken place is thus found. From this the quantity of the gas
burned is calculated either directly, or after removing any carbon
dioxide formed by the combustion, and estimating the decrease
of volume thus produced.

In this way hydrogen can be burned most easily and quickly ;
carbon monoxide is burned a little less easily, but still quite
conveniently ; ethylene, acetylene, and benzene more slowlv,
and only at a stronger heat. Methane (marsh-gas) is not burned
at all ; even in presence of considerable excess of easily combus-
tible gases there is no methane, or at most extremely slight traces
of it, burned along with them. An explosion has never been
observed to take place.

Applications : —

1. Estimation of hydrogen in the absence of other gases. — In
order to practise the manipulation of this method^put 20 or 25 c.c.
of hydrogen (taken from a hydrogen-pipette, p. 133) into the burette,
admit air up to nearly 100 c.c.? and notice both amounts after

COMBUSTION BY PALLADIUM. 143

careful reading off. Then carry out the combustion as described,,
and by another reading tind the contraction produced. Since
the water produced is condensed, there is 1 vol. of oxygen ab-
stracted for each 2 vols. of hydrogen ; and thus the volume of the
hydrogen originally present is found on multiplying the contraction
observed by two-thirds. Since the hydrogen employed is never
quite pure, the yield will be always a little deficient.

Example : —

Hydrogen employed =22'8 c.c.

Hydrogen -f air =98'0 „

Hence air alone =75-2,,

Oxygen contained in the same ... = 16'3 „

Oxygen required by theory = 11*4 „

Excess of oxygen employed = 4'9 „

Volume of gas after combustion . = 64'0 „

Contraction . =34*0 ,.

Found : —

34x2

= 22*66 c.c. hydrogen.

2. Estimation of hydrogen in the presence of other gases, for
instance in water-gas, producer -gas, coal-gas. — The following gases,
if present, are removed and estimated first by absorption, in the
order given : carbon dioxide, ethylene (propylerie, butylene), ben-
zene, oxygen, carbon monoxide (seep. 102) ; the remaining gas or
a measured portion of it is mixed with a quantity of air, sufficient
in any case for burning the hydrogen present, and the mixture
passed over heated palladium-asbestos. The gaseous mixture now
left can only contain, as belonging to the original gas, methane
and nitrogen, mixed with the remainder of the atmospheric air,
that is nitrogen and oxygen, whose volume is known.

Example : —

Estimation of a heating -gas containing nitrogen, produced by
working coke- gas producers with air and steam.
A^olume of gas employed, 97'7 c.c.

A. Estimation of the absorbable gases.
After absorption by potash 87*5 c.c.

Decrease of volume . . . 12'0 „ =12*8 volume p. c. C O2.

144 ON APPARATUS AND METHODS.

After absorption by cuprous j

chloride J

Decrease of volume 17'1 ,, = 17*46 p. c. CO.

Unabsorbed residue 68'6 },

B. Estimation of hydrogen.

Since the volume of the non-absorbable gaseous residue is too
large to admit of the admixture of a sufficient quantity of air for
burning the hydrogen within the confined space of the burette,
only a portion of it is employed for continuing the analysis.

Unabsorbed gas employed^

(equal to 86*44 per cent. r _ 0
v-i, • • i i r>59'3 c.c.

of the original volume off

gas) J

Gas + air 98-8 „

Air alone : 39*5 ,,

In this : Oxygen 8*2 „

Nitrogen 31'3 „

Volume after combustion ... 80*5 „'

Contraction 18'3 „

Corresponding to : —

Hydrogen (from the original! 12.2 c.c. = 14.43 cent fc

mixture) ,

volume.

Oxygen (from the air) 6'1 „

Non-combustible residue . . . 8O5

C. Estimation of nitrogen.

The amount of nitrogen in the gas is found by deducting from
the non-combustible residue that which was left from the air
employed for combustion.

Non-combustible residue 80-5 c.c.

Containing oxygen of the air (8'2 — 6'1) = 2*1
„ nitrogen „ 31'3

Altogether 33'4 c.c.

The difference is nitrogen 47'1 „ =55-77 per

cent, by volume.

ESTIMATION OF OXYGEN. 145

Final result : —

Carbon dioxide 12'28

Carbon monoxide 17'46

Hydrogen 14'43

Nitrogen 55'77

99-94

3. Estimation of oxygen in atmospheric air and other suitable
mixtures of gases. — Add to the gas measured off in the burette a
volume of hydrogen exceeding twice the possible percentage of
oxygen, and let the combustion take place in the capillary tube.
Since two volumes of hydrogen vanish for each volume of oxygen,
the contraction observed, divided by 3, yields the proportion of
oxygen.

Example : —

Air employed 66'7 c. c.

Air + hydrogen 99*2 „

Hydrogen added : 32*5 „

Hydrogen required by theory 27*6 „

Hydrogen in excess 4*9 „

Volume after combustion 57*8 „

Contraction 41*4 „

Found : —
41'4

= 13'8 c.c. = 20'69 per cent, by volume of oxygen.

4. Estimation of carbon monoxide in chimney -gases, blast-furnace
yases, fire-damp, fyc. — The carbon dioxide is first estimated by
absorption, a measured excess of air is then added to the un-
absorbed residue, or to a measured portion of it, and the combustion
is made by the capillary tube. The equation is

CO + O =CO2,
2 vols. + 1 vol. = 2 vols. ;

hence the volume of the air to be added must be at least 2J times
that of the carbon monoxide present. The contraction taking
place after the combustion must be multiplied by 2; but it is
more accurate to absorb the CO2 produced by combustion in the
potash-pipette, and to multiply the total decrease of volume by f .

146 ON APPARATUS AND METHODS.

The combustion of carbon monoxide to carbon dioxide by means
of palladium- (or platinum-) asbestos is especially useful for esti-
mating very slight quantities of carbon monoxide, such as may
occur in the air of inhabited rooms or (frequently along with
marsh -gas) in the " fire-damp" of coal-pits. But in the latter
case the carbon dioxide cannot be estimated volumetrically, but
must be titrated. For this purpose we employ the apparatus
described and figured for the estimation of methane (see below) ;
but, in lieu of the combustion-tube charged with copper oxide, a
tube of equal size, charged with platinum-asbestos, is employed,
and heated just to an incipient red heat. In this case carbon
monoxide only, and no methane, is burned. The CO2 formed is
absorbed by titrated baryta-water (as in the case quoted), and the
excess of the latter retitrated with normal oxalic acid. After this
treatment the gas can be tested for methane.

Of course the gas should be free from organic dust and tarry
matters, which would equally furnish carbon dioxide on combustion.
If such impurities are present, the gas should be filtered through
cotton-wool and washed with concentrated caustic-potash solution ;
but even then a qualitative test for carbon monoxide should be
made first, as prescribed p. 74.

LUNGE'S MODIFICATION OF THE ORSAT APPARATUS*.

Fig. 73 shows this apparatus, which, in addition to all the
essential parts of an ordinary Orsat apparatus, contains a con-
trivance for burning hydrogen &c. by means of heated palladium-
asbestos, a is the gas-burette ; b, c3 and d are the usual U- tubes
for absorbing carbon dioxide, oxygen, and carbon monoxide ; k is
the ordinary three-way cock ; e is a glass tap, to which is fused a
capillary tube bent twice at a right angle. This is tightly joined
by apiece of stout india-rubber tubing to the combustion-capillary
tube/, which contains a thread of palladium-asbestos, made accord-
ing to Wmkler's description (p. 140) . The U-tube h is exactly
similar to the vessels b, c, and d} and is filled with water up to a
mark in its capillary neck. A very small spirit-lamp, gy is fixed
with its thin stem in a spring-clamp which, by means of the pivot-
wire i, turns in a socket fastened to the wooden box containing the
apparatus. The dotted U-tube to be seen at the left side is partly

* Added by the Translator.

LUNGE S MODIFICATION OF THE ORSAT APPARATUS.

147

filled with cotton- wool, and serves for retaining any tarry matters.
(Such a contrivance is generally found connected with the ordinary
Orsat apparatus.)

Fig. 73.

Manipulation. — After absorbing carbon dioxide, oxygen, and
carbon monoxide in the manner described p. 89 (any ethylene
present would be absorbed along with the carbon monoxide by
the acid solution of cuprous chloride) , air is admitted through the
three-way cock k to the gaseous residue contained in the burette
0, till the total volume as nearly as possible comes up to 100 c.c.
The air added will allow of the burniug of a quantity of hydrogen
corresponding to two-fifths of its volume (i. e. twice the volume of

148 ON APPARATUS AND METHODS.

oxygen contained in the air). This suffices for ordinary producer-
gas ; but when analyzing ' ' water-gas/' or similar mixtures con-
taining a rather considerable quantity of hydrogen, a smaller
quantity of gas must be employed for analysis, or else oxygen is
introduced in lieu of atmospheric air. After reading off the total
volume, the spirit-lamp g is lighted and turned so that it heats
the capillary / very gently ; then the level-bottle is raised, the
tap e is opened, and the gas is passed through the capillary / into
the receiver h and back again into the burette. One end of the
palladium-asbestos should become red-hot during this operation.
The volume of gas is read off and the passage through f is re-
peated ; if, which is usually not the case, a further contraction is
now observed, the passage through /must be repeated once more.
The residual gas is now finally measured, and two-thirds of the
diminution in volume calculated as hydrogen (compare p. 143).

Application : —

Estimation of hydrogen along with carbon dioxide, oxygen, and
carbon monoxide in producer-gas ^water-gas, and similar mixtures. —
The advantage of this apparatus is that it is much more portable
than Hempel's burette with its appendages, and that the analysis
can be performed in any place and very quickly. Ethylene
and other heavy hydrocarbons would be absorbed along with
carbon monoxide ; but they occur in such gases in quantities so
small that they may be safely neglected, or rather calculated as
carbon monoxide. I£ they had to be accounted for in another
way, a second test should be made, leaving out the operation
with cuprous chloride ; this time the gas, after absorbing CO2 and
O in the usual way, is at once mixed with an excess of air and
burnt by the palladium-asbestos. By measuring the contraction
produced, then absorbing the CO2 formed in the receiver b filled
with caustic potash, and measuring the new diminution of volume,
we obtain another estimation of the combustible gases carbon
monoxide, hydrogen, and ethylene (if present) in this way. If the
first contraction be diminished by one half of the second contrac-
tion (that is, that taking place by absorption of the CQ2 formed
in combustion), two-thirds of the difference represent the hydrogen;
the carbon monoxide corresponds to the second contraction,
according to the following formulae : —

£*?vols. CO + x vol. O yield 2#vols. CO2.
Syvols. H + y vol. O yield (condensed) H2O.

COMBUSTION BY RED-HOT PLATINUM. 149

Hence : —

First contraction =A =
Second contraction = B =

It follows from this that

Carbon monoxide = B.

Hydrogen ......... = 2

If the numbers thus obtained closely agree with those found by
the first test, made in the ordinary way, as described before, we
may conclude that no heavy hydrocarbons are present ; indeed
we must expect to find rather less CO2 than theory requires, as
part of it will be absorbed by the water contained in the appa-
ratus (in order to diminish this error, the analysis should be per-
formed as rapidly as possible). If, therefore, the CO2 found is in
excess of that required on the assumption that only CO and H
were present, we must conclude that heavy hydrocarbons were
present, and equations might be given including these as well;
but there is no sufficiently accurate method of carrying out this
estimation by means of technical gas-analysis, working over water.
Ethylene &c. may also be previously absorbed by bromine-water
(compare p. 67) and estimated in this way.

It is unnecessary to say that any methane present will be left
in the unconsumed remainder of gas ; it may be estimated by any
of the methods described below.

C. Combustion by means of red-hot Platinum.

While palladium, both in the compact state and, even better,
when finely divided, causes the combustion of hydrogen, carbon
monoxide, and heavy hydrocarbons, mixed with air, at a gentle heat,
without drawing methane into the action, methane itself is burned
easily and without explosion if mixed with sufficient air and
brought into contact with palladium at a bright red heat. But
as palladium has not much tenacity at high temperatures, and
especially thin palladium wires easily break when getting too hot,
and as in this case the principal function of the metal is a trans-
mission of heat, it is preferable to employ platinum, which is hardly
less active and much more durable.

150

ON APPARATUS AND METHODS.

a. COQUILLION'S GBISOUMETER,

Coquillion showed (Compt. rend. 1877, clxxxiv. p. 458) that a
mixture of methane and air in contact with red-hot platinum or
palladium burns perfectly without explosion. He utilized that
observation for the examination of fire-damp (" grisou ") by means
of an instrument called a Grisoumeter, which is shown in fig. 74.
A is a measuring-tube, ending at the top in a T-piece with two taps.
It contains from these to the zero-point 25 c.c., the bottom part

Fig. 74.

being divided. The lower end is connected with the level-bottle
F, filled with water, and is thereby charged and emptied like
Orsat's apparatus (p. 87). Through its two taps the burette
can be connected either with the reservoir containing the sample
of gas, or with the combustion-vessel B, which is hydraulically
sealed by C. If, after combustion, the carbon dioxide is to be
removed from the gas and measured, the apparatus is provided
with an absorption-vessel D (fig. 75), charged with caustic-potash
solution, and is then called a Carburometer. The thimble-shaped
glass vessel B is closed by an india-rubber cork, pierced by two
strong brass pins provided with screw-clamps, and connected
inside the vessel by a spiral of thin platinum or palladium wire^

COQUILLION'S GRISOUMETER.

151

which can be made red-hot by passing an electric current through
it.

Manipulation. — A is filled with water by raising F, connection
is made with the cylinder containing the sample of gas, the latter
is opened under water by removing its cork, and by lowering F ths
gas is transferred into A, where it is drawn in up to the zero-mark
in the well-known manner. Now the current is closed and the
gas is passed over the red-hot platinum spiral in B by means of

Fig. 75.

raising the level-bottle F, repeating this several times. After
cooling the contraction is noted, half of which corresponds to the
methane present. If there is too little oxygen present, more air
(measured) must be admitted first.

The combustion is easy and rapid, but the cooling takes a long
time and the results are only approximate. Small percentages of
methane in the air of coal-pits, whose control is very important,
cannot be estimated by the grisoumeter.

b. CL. WINKLER'S APPARATUS.

Several chemists have applied Coquillion's principle with more
suitable apparatus, as Mertens (Zeitsch. f. analyt. Chem. 1887

152 ON APPARATUS AND METHODS.

p. 42),Thorner (ibid. 1889, p. 642), Jeller (Zeitsch. f. angew. Chem.
1896, p. 692), and the author (Zeitsch. f. analyt. Chem. 1889,
p. 286). The latter's apparatus is shown in fig. 76.

Fig. 76.

In a Hempel's tubulated gas-pipette two brass electrodes are
introduced, 175 mm. long, 5. mm. thick, not varnished. At the
bottom they have holes for the current- wires, at the top incisions
in which the two ends of a platinum spiral are fixed by small
screws.- The spiral consists of platinum wire, 0'35 mm. thick,
made by coiling it six times over a steel needle T3 mm. thick
and leaving 1 cm. at the ends for being fixed in the above-named
incision. Previously the two electrodes are coupled by a twice-
perforated cork (not shown in the figure) which reaches halfway
up and prevents them from moving. The electrodes ought to be
2 or 2'5 cm. distant from the top of the pipette. The pipette is
then completely filled with water and is kept closed in the usual
manner.

Dennis & Hopkins (Zeitsch. f. anorgan. Chem. 1899, xix. p. 179)
fill the pipette with mercury, with slightly modified electrodes.

Manipulation. — The gaseous remainder, freed from absorbable
constituents and hydrogen, and containing only methane and

153

nitrogen, is measured in a Hempel's pipette and mixed with a
measured excess of air.

The burette is by means of an ordinary glass capillary connected
with the pipette and the current closed. The level-tube of the
burette is lifted up with the left hand, one of the pinch-cocks is
opened entirely, the other one partly with the right hand, and
thus the gas is slowly transferred to the pipette. As soon as the
water has sunk below the spiral, this becomes red-hot. Now the
entrance of the gas must be interrupted for a moment and the
remainder of the gas introduced very gradually, in which case the
combustion always takes place quietly and without any danger.
If, however, the gas is passed in very quickly, or if it is first put
into the pipette and the current closed subsequently, an explosion
may occur which throws out the cork containing the electrodes
and the water out of the side bulb.

The thickness of the wire and the number of coils (i. e. its
length) must correspond to the strength of the current. The
above given dimensions refer to a current from two small Grove
elements. If the wire is too thin, it fuses; if it is too thick, it
does not get hot enough, but it is not difficult to hit the proper
medium.

The combustion is finished within one minute. The current
is shut off, the pipette (the upper part of which gets rather hot) is
allowed to cool down, the gas is re- transferred into the burette,
the carbon dioxide is removed by means of a caustic-potash pipette
and the total contraction noted. By dividing the latter by 3 the
volume of the methane is found.

Application : —

Estimation of methane in natural gas, in f( blowers " of coal-pits,
in marsh-gas, coal-gas, producer-gas, fyc. — Remove by absorption
successively carbon dioxide, heavy hydrocarbons, oxygen, carbon
monoxide (p. 102), then hydrogen by combustion with air
and palladium-asbestos (p. 135), and burn the methane as described
above.

In natural gas, ethane, propane, and other hydrocarbons of
the series C»H2n+a frequently occur in very small quantities.
These are burned at the same time as methane, but with different
contractions.

OP THE

UN i I/CD £>••*...

154 ON APPARATUS AND METHODS.

EXAMPLE : Examination of natural gas.
Employed 99'8 c.c.

A. Absorbable constituents.

After treatment with potash. . . 99-6 c.c.

Contraction ......... O2 „ =0*20 per cent, carbon

After treatment with fuming dioxide.

sulphuric acid and taking
away the acid vapours by
potash ........................... 99-3 „

Contraction ......... 0'3 ,, =0'30 per cent, heavy

After treatment with alkaline hydrocarbons.

pyrogallol ..................... 98'8 „

Contraction ......... 0*5 ,, =0'50per cent. oxygen-

After two treatments with

ammoniacal cuprous chloride 98*8 „

Contraction ......... 0*0 „ =0*00 per cent, carbon

monoxide.

B. Hydrogen. — Since natural gas contains very little hydrogen,
the latter does not require much air for burning by means of
palladium-asbestos, and hence most of the non-absorbable re-
mainder can be employed for this estimation.

99-8x78-2
Employed unabsorbed gas -

t/O o

= 78-99 c.c of original gas ... 78'2 c.c.

After addition of air .................. 99'2 „

Air alone .............................. 21*0 „

Containing oxygen .................. 4*2 ,,

„ nitrogen .................. 16'8 „

After burning with palladium as-

bestos .............................. 96-5 „

Contraction 99-2— 96-5 ...... =2'7

„

Corresponding to : —

Hydrogen in the gas .................. 1'8 n = 2'28 per cent.

Oxygen from air ..................... O9 „ hydrogen.

Gaseous remainder 96'5 .

155

C. Estimation of methane. — The gaseous remainder is considered
to be pure methane ; no more than 9 c.c. of it, with ten times its
volume of air, is burned by means of electrically glowing platinum.
By dividing the contraction thus produced by 2, the percentage
of methane is found ; but it is more correct to remove the carbon
dioxide by caustic potash and divide the total contraction by 3
(p. 138).

78*99 x 8*8

Gas employed after treatment as under A and B : — —

9o*o

= 7*20 c.c. of the original volume. 8*8 c.c.

After addition of air 99'2 „

Air alone 90*4 ,,

Containing oxygen 18*1 „

„ nitrogen 72'3 „

After combustion 85'6 „

Contraction 99-2 — 85-6 13*6 „

After treatment with potash 78'8 „

Contraction 6*8 „

Total contraction 99-2 -78-8 . 20'4 „

20*4
Methane in gas = — — 6'8 „ =94*44 per cent.

methane.
Oxygen from air 1 3*6 )}

D. Nitrogen is found by difference.

Final results : —

Carbon dioxide 0'20 per cent, by volume,

Heavy hydrocarbons 0'30 „ „

Oxygen 0'50 ,, „

Hydrogen 2'28 „ „

Methane 94'44 „ „

Nitrogen 2'28

100-00

156

ON APPARATUS AND METHODS.

c. CL. WINKLER'S APPABATUS FOB THE EXAMINATION OF COAL-PIT AIR

CONTAINING METHANE.

It is frequently assumed that the prevention of danger from
fire-damp in coal-pits need only extend to ascertaining whether
the atmosphere of the pit contains enough methane to make it

Fig. 77.

inflammable or explosive, and various apparatus have been con-
structed with that object. This is, however, a mistake. The
mining engineer must try to prevent any accumulation of fire-
damp before the percentage of methane has reached the lower limit

157

of explosiveness. By the examination for methane, both in the
branch current and in the principal current of air issuing from
the pit, he must carefully establish the average composition of
the pit-air, as it changes with the progress of working the coal-
seams. In all these cases it is necessary to determine compara-
tively small quantities of methane, such as could not possibly be
read off in a gas-burette with any degree of exactness. The
following process leads to the desired end in a simple manner. It
consists in burning the methane, consumed in a large volume of
pit-air, by means of electrically glowing platinum and afterwards
estimating the carbon dioxide by titration. This process has been
thoroughly tested in the laboratory of the Freiberg Mining
Academy ; and it has been established there that a stream of in-
duction-sparks, even of considerable length, cannot replace the
electrically glowing platinum.

All the operations of measuring, burning, and titrating are
carried on in the conical flask A, fig. 77, which is turned upside
down, as shown in the figure, during the combustion. On its
neck it has a circular mark up to which it is ordinarily closed by a
twice -perforated india-rubber cork, with glass-rod stoppers. The
contents of the flask up to this mark are ascertained by weighing
or measuring and are noted on the glass by etching; it ought to
hold about two litres, but in the case of gases containing much
methane one litre is sufficient.

When the flask has to serve for a combustion, its stopper is taken
out under water and replaced by an india-rubber cork k, with
electrode e. It possesses a second perforation, closed by a short
glass rod, for the purpose of introducing by means of a pipette
a certain volume of water, say 10 c.c. This water prevents
during the combustion the contact of the gas with the india-rubber,
which might produce considerable errors ; when turning the
flask upside down, it forms the protecting layer w. Its volume
must be known, as well as that of the electrode e, and these
amounts must be deducted from the contents of flask A.

Lest the flask should get hot during the combustion, it is im-
mersed in a beaker filled with water and prevented from rising up
by means of the adjustable iron holder H, as shown in the figure.
If the beaker is replaced by a tin vessel, the holder can be fixed
to its side. Wires d and dlf which transmit the electric current,
should be at least 1 mm. thick and insulated with gutta-percha.

158

ON APPARATUS AND METHODS.

The electrode e, shown half-size in fig. 78, has been constructed
by O. Brunck *. It is made of brass and must not
be varnished, to avoid any organic substance. Its Fl"- 78<
parallel arms a and al at the top form an open ring ^

b which carries, by means of screws, the platinum
spiral c and at the same time protects this against
a shock. In their lower part the two arms are
insulated by a strong strip of india-rubber, and at
the bottom they form together a cylindrical part d
which passes gas-tight through the central opening
of the cork of flask A, so that the insulating strip
of india-rubber does not project above the pro-
tecting layer of water w. The insulating strip is
continued down to the end, being thinner there ;
at the bottom the holes e and et are drilled into the
two arms ; into these holes the current- wires are
introduced and held fast by screws / and flm
Spiral c consists of platinum wire, 0*35 mm. thick ;
the total length of wire within the screw-clamps is
7 cm. In order to bring this platinum spiral to a
bright red heat without any fear of fusing it, a cur-
rent of 8 or 9 amperes should be applied, e. g., by two
large Bunsen elements placed in series, or by two
storage-cells.

Manipulation. — Flask A is filled with distilled water,
carried into the coal-pit, and the water run out on
the spot where the sample of gas is to be taken.
The flask is closed by its twice-perforated cork and
taken into the laboratory. If the sample had been
^aken in another vessel, e. g. the zinc vessel described
p. 23 , the flask A is filled from this in the labora-
tory, taking care to let the inlet-tube end at the
highest point of the flask, previously filled with water and inverted
under water, so that the gas comes into the least possible contact
with the water.

The flask A, filled with the gas to be examined, is now closed by
the other cork, provided with the electrode e, effecting the exchange
of corks under water of the temperature of the room. The pro-

* It is sold by Louis Jentzscb, Silberniarmstrasse ], Freiberg in Saxony.

e e1

WINKLER'S APPARATUS FOR COAL-PIT AIR. 151>

tecting water w is put in, the current-wires d and dl are attached,
the flask is placed under the water contained in B and fixed bv
holder H. Now the current is closed and the platinum spiral
kept at a bright red heat for half an hour, in order to burn the
methane completely by the oxygen always present in excess.
Then the current is interrupted, the electrode-cork is replaced by
the ordinary cork, and the carbon dioxide titrated as described
pp. 49 & 105. As a rule, the baryta-water can be run in from the
burette without raising the cork. The volume of gas employed
must be reduced to normal conditions.

Application as above described for pit-air and other non-inflam-
able mixtures, containing relatively little methane. Heavy hydro-
carbons and carbon monoxide must be absent. Carbon dioxide is
rarely absent ; it is estimated by titration in another sample of gas
by Hesse's method (p. 103), and deducted from the total CO2 found
after combustion.

EXAMPLE. — Pit-air. The reduction apparatus (p. 26) shows
112-8 c.c.

A. Estimation of carbon dioxide.

Oxalic acid (gas normal) 1 c.c.^1 c.c. CO2.

Baryta-water titrated 1 „ = 1 '03 „ oxalic acid.

= 1-03 „ CO2.

Contents of absorption-bottle 622 „

Baryta-water employed 10 „ = 10'3 c.c. CO2..

Gas analyzed 622-10 =612 „

, , /100x612v
Corrected/— -^J 542 „

Oxalic acid required for retitrating . 8'5 „
Oxalic acid equivalent to 10 c.c.

baryta- water 10*3 „

Difference 10-3-8-5 =1-8 „ =0'33p.c. CO2,

B. Estimation of methane.

Contents of absorption-bottle... 2000 c.c.

less protecting water 10

contents of electrode 6
baryta-water added
after combustion . . .20

36 c.c.

160 ON APPARATUS AND METHODS.

Gas really employed for test ... 1964 c.c.

Corrected 1741 „

20 c.c. baryta- water required... 20'6 „ normal oxalic acid.

Employed for retitrating 4*3 „ „ „

Difference 16'3 „

= 0-93 p.c. CO3p. vol.

Deduct C02 found sub A. . . 0'33 p.c. "„
Leaving as methane 0'60 p.c. CH4.

d. DREHSCHMIDT'S PLATINUM-CAPILLABY.

The just described method requires an electric current which,
if it has to be specially produced, makes the apparatus quite as
complicated as that required for the explosion method. This is
avoided by the following method.

Already Orsat (' Note sur 1' Analyse industrielle des Gaz/ Paris,
1887) noticed that methane, mixed with air or even with pure
oxygen, can be burned without loss or danger in a capillary tube
made of platinum. This observation was taken up and enlarged
by Drehschmidt (Berl. Berichte, 1888, xxi. p. 3242). The possi-
bility of employing pure oxygen, and therefore avoiding the
•dilution with nitrogen, admits of employing a larger volume of
gas. The oxygen need not be absolutely pure, especially from
nitrogen ; it can be made as usual from potassium chlorate, and
after careful washing kept for use in a gas-holder.

Drehsphmidt's capillary is a platinum tube, 200 mm. long, 2 mm.
thick, 0*7 mm. bore, and at both .ends soldered to brass connections.
In order to avoid explosions, it is nearly filled up all along by two
or three thin platinum wires.

The above length is necessary, because otherwise the ends
become too hot. This can be avoided, as shown by the author, bv
cooling the ends with water, fig. 79.

Fig. 79.

The platinum tube proper, p, which must not be made by
soldering, but bored or drawn, is 2*5 or 3 mm. wide, 0'7 mm. bore,

DHEHSCHMIDT'S PLATINUM-CAPILLARY.

1G1

and only 100 mm. long, and is filled with several thin platinum wires
in such manner that the gas passes through without sensible
resistance. To its ends are soldered the copper elbows Jc arid klt
5 mm. outside diameter, 1 or 2 mm. bore, equally filled with thin
platinum or copper wire. These are surrounded by the copper or
brass jackets w and ivl} tubulated at top, 50 mm. long, 25 mm.
wide, which receive the cooling water. The whole, in order to be
protected against bending &c., rests by means of the washers s
and sl} on a fork movable on. a stand, and is heated by a gas-
burner with fan-shaped top *.

Fig. 80.

Manipulation. — The gaseous remainder, containing nothing but
methane and nitrogen, is transferred into the gas-burette A, fig. 80,
a sufficient volume of oxygen is added, the platinum-capillary E
interposed between the burette and the gas-pipette C (filled with
water), the capillary heated to a bright red heat by means of the gas-
burner F; and by opening one pinch-cock entirely, and regulating
the other with the right hand, the gas is carried into the pipette
and back again in a moderately rapid stream. This operation is

* The capillary and burner are sold by Dr. Robert Muencke, Luisenstrasse 58,
Berlin N.W.

162 ON APPARATUS AND METHODS.

repeated once, or at most twice. The capillary is allowed to cool,
the contraction is measured, the carbon dioxide removed and the
measuring repeated. The methane originally present corresponds
to a third of the total contraction observed.

In the case of gases containing very much nitrogen, it is pre-
ferable to avoid unnecessary dilution by employing pure oxygen
in the previous estimation of hydrogen (by combustion with
palladium-asbestos, p. 142) *.

EXAMPLE : Analysis of producer -gas.

Gas employed 99'7 c.c.

After treatment with caustic

potash 93-8 „

Difference 5'9 „ = 5'92 p.c. CO2.

After treatment with fuming
sulphuric acid and removal of

acid vapours by potash 93 '7 „

Difference O'l „ = 0 10 p.c. heavy hy-
drocarbons.
After treatment with alkaline

pyrogallol 93*7 ,,

Difference O'O „ = O'OO p.c. oxygen.

After two treatments with ammo-

niacal cuprous chloride 71*5 „

Difference 22'2 „ = 22-27 p.c. CO.

Non-absorbable gaseous remainder 71*5 „

Hydrogen. — Producer-gas contains rarely above 10 p.c. of hydro-
gen and 5 p.c. of methane f, and the gaseous remainder would
therefore require at most 5 + 10 = 15 c.c. oxygen. When em-

* It is evident that a Drehschmidt or "Winkler platinum-capillary can be-
attached to the Great-Lunge apparatus (fig. 73, p. 147) in place of the palladium-
asbestos capillary, e, thus enabling that portable apparatus to be used for the
estimation of methane. If no regard is taken of the heavy hydrocarbons (which
is unnecessary witli producer-gas or water-gas), that apparatus is thus capable of
performing the entire analysis. — Translator.

t Dowson gas and similar " semi-water gas " contains sometimes not much less
than 20 p. c. hydrogen, but both this and ordinary producer-gas hardly ever
contain 5 p.c. methane. — Translator.

ANALYSIS BY THE PLATINUM-CAPILLARY. 163

ploying the oxygen in the pure state the whole of this remainder
can be used for the following operations.

Non-absorbable gas 7T5 c.c.

After addition of oxygen 94'8 „

Oxygen alone 23'3 „

After burning with palladium-asbestos. . . 84*0 „

Contraction 94-8 -84-0 =10*8 „

Hydrogen in the gas 10 x § 7'2 „ = 7'22 p.c. H.

Oxygen corresponding to this 3'6 „

Gaseous remainder 84*0 „

Containing oxygen 19'7 „

As this oxygen more than suffices for the combustion of the
methane, the gaseous remainder is at once burned in the platinum

capillary, leaving behind 78*2 c.c.

Contraction 84-0-78-2 5'8 „

After absorption by potash 75*3 „

Difference 2'9 „

Total contraction 84-0 — 75'3 8'7 „

Corresponding to methane 2*9 „ = 2'91p.c.CH4.

» oxygen 5'8 „

Nitrogen.

The iion-absorbable gas =71 '5 c.c.

Contains hydrogen 7'2 c.c.

methane 2*9

together 1O1 „

leaving a remainder of 61'4 „ = 61*58 p. c. N.

Final result : —

Carbon dioxide 5*92 p. c. by volum ?.

Heavy hydrocarbons O10 „ „

Carbon monoxide 22'27 „ „

Hydrogen 7'22 ,, „

Methane 2'9l „ „

Nitrogen 61'58 „ „

100-00

164 ON APPARATUS AND METHODS.

2. Estimation of nitrous oxide by burning with hydrogen. — If a
mixture of nitrous oxide and excess of hydrogen is passed through
a (moderately) heated tube, the reaction is : N2O + 2H = 2N + H2O.
This means that 2 vols. N2O 4- 2 vols. H furnish 2 vols. N (the
H2O being condensed to liquid water), and that the contraction
is = J, or = the volume of the nitrous oxide present. This
method might be applied to estimating N2O in gases containing
no other constituents acting upon hydrogen, as oxygen, nitric
oxide, &c., and it might be carried out by means of a moderately
heated platinum-capillary.

3. Estimation of nitric oxide by burning with hydrogen. — This
can be carried out (Knorre & Arndt, Berl. Ber. 1899, xxxii.
p. 2136; 1900, xxxiii. p. 32) as just proposed for N2O, but the passage
of the gas must be slow, to avoid the formation of ammonia. The
reaction is: NO + 2H = N-fH2O ; hence 2 vols. NO + 2 vols. H =
1 vol. N; the contraction is therefore 4 — 1 — 3, and the volume
of nitric oxide is found by multiplying the contraction by f .

D. Combustion of Gases by means of hot Copper Oxide.

This method has been worked out by Fresenius (Zeitschr. f.
anal. Chem. 1864, p. 339) at a time when there were no simple
gas- analytical methods extant for the estimation of combustible
gases. The same principle is even now useful, and indeed indis-
pensable, for the detection of minimal percentages of combustible
gas ; and it has been employed with full success for many years
past by the author, and in other laboratories, for transforming the
methane of pit-air into carbon dioxide, to be estimated by titration.
It can be applied to indefinitely large quantities of gas, and cor-
respondingly small percentages of combustible gas can be thus
estimated. The method will be best understood by the description
of the following apparatus, fig. 81, which has been thoroughly tested
for the examination of pit-air for methane.

The gas is brought into the laboratory in a zinc vessel A*. It
can be forced from this into the combustion apparatus by a
stream of water from the raised vessel B. The gas first enters the

* Hankus (Oesterr. Zsch. f. Berg- und Hiittenwesen, 1897, p. 548) counsels the
employment of glass vessels only, because the organic substance attached
to the inside of the metal vessels might cause the methane to be transferred
into C02. This is contrary to the chemical behaviour of methane and to the
experience obtained at Freiberg.

COMBUSTION BY HOT COPPER OXIDE.

165

166 . ON APPARATUS AND METHODS.

absorbing-worm K, which, in case of need, can also be connected
with the gas-holder L, filled with air,, or a tube service for com-
pressed air ; this worm is filled with concentrated solution of
caustic potash (of about 1/26 specific gravity), and serves for
retaining every trace of carbon dioxide. From this the gas
passes through the drying-bottle S, filled with concentrated
sulphuric acid, into the combustion-tube,, filled with a layer of
granulated copper oxide, 25 centimetres (10 inches) long, and
heated to a bright red heat in the furnace J^by means of a four-
fold burner provided with a contrivance for regulating the supply
of gas and air at the same time. The combustion-tube is sur-
rounded by wire gauze for three-fourths of its circumference,
the gauze being tied fast with wire loops in distances of a few
centimetres (about an inch), without pulling the wire loops too
tightly ; it is then covered on the outside with a thick paste of
finely ground fire-clay mixed with a little pipe-clay (the whole
being made into a paste by means of commercial solution of
silicate of soda, diluted with four volumes of water), so that only
the upper fourth, not covered with wire gauze, remains free. The
paste is laid on with a brush in three coats, waiting each time for
the former coat to dry. Combustion-tubes thus guarded, when
otherwise carefully heated up and cooled down, are very durable
and often remain serviceable for months. In lieu of a glass tube
a tube of drawn copper might be used, of twice the length of the
former, so that the projecting ends remain cool.

From here the gas gets into the absorbing-vessels W and W ,
and at last into the aspirator N, surmounted by the mercurial
pressure-gauge M, whilst the lighter flask 0 is placed below the
outlet-tube in order to measure the water run out.

Manipulation. — Before making a gas combustion, the copper-
oxide filling of the combustion-tube must be ignited in a current
of air, until clear baryta-water is no longer affected by the issuing
gas, even after passing for some time. This air (as well as that
required at the end for sweeping out the apparatus) should be
taken outside the laboratory, since the laboratory air is mostly
contaminated with a little coal-gas which would be detected by the
precipitate formed in the baryta-water after passing through the
red-hot copper oxide. When the contents of the tube have been
sufficiently ignited and all vessels are filled with air free from carbon
dioxide, the combustion can begin. The gas-holder L or the air.

COMBUSTION BY HOT COPPER OXIDE. 167

service must first be shut off by the pinch-cock s, and the pressure
gauge M be made to show an equilibrium.

Suppose we have to examine a sample of air taken from a coal-
pit infested with fire-damp, and carried to the laboratory in the
well-closed tin-plate cylinder A. First of all the solid corks of
this cylinder have to be replaced by simply perforated india-rubber
corks, carrying a bent glass tube with a pinch-cock stopper.
This change is made by first dipping one end, then the other end,
of the cylinder under water, replacing the solid cork by the per-
forated one below the surface of the water, and then introducing
the bent glass tube with its pinch-cock stopper into the per-
foration of the cork. Since the collecting-vessel had been filled
below ground (that is, under higher atmospheric pressure), on
opening it a portion of the gas bubbles out through the water,
which is the best sign that the stoppings had been tight.

When the tin cylinder A has been arranged in this way it is
hung in its stand, the elastic tube s' is attached to the pinch- cock/?,
and p' is connected with the outlet-pipe of the vessel B, after
having been filled with water. The pinch-cocks p and p' may now
be permanently opened, and are for this purpose slid over the
adjoining glass tubes. The issue o£ the gas and the regulation of
its current is exclusively performed by the screw pinch-cock s .
Before opening the same, the receivers W and W are charged
with 25 c.c. each of approximately normal baryta-water and a
drop of phenolphthalein ; this quantity is measured off by a
burette provided with a float and divided in ^ c.c., and the corks
are only opened for an instant. When the receivers are again
connected, a little water is run out of the aspirator N3 till the
baryta- water rises in the bulbs of the receivers ; the water run out
so far is poured away, and now the current of the gas to be burned
is started by gently opening the screw-tap s'. The pressure-gauge
M at once begins to rise ; the outlet-cock of the aspirator N there-
fore is opened so far as to show a slight minus pressure, which is
maintained throughout the combustion. The current of the gas
is so regulated that about 200 bubbles per minute pass through
the washing-bottle S, and thus the litre-flask O is filled once in
40 minutes. The water running out is each time poured into
the vessel B, closing the aspirator-tap in the meanwhile ; the
empty litre-flask is at once put in the old place, and each litre-
full is noticed.

168 . ON APPARATUS AND METHODS.

Once started, tlie experiment requires but little supervision,
The contents of the absorbers are now and then shaken up ; in
the meantime the state of the barometer and thermometer is
noticed, and the baryta-water is standardized by means of normal
oxalic-acid solution. Here also a burette with a float is employed,
All measurings have to be made very carefully throughout.

After a little time the baryta-water contained in the vessel W
becomes turbid, and a distinct precipitate gradually collects in it j
but the contents of W ought to remain clear, or at most slightly
opalescent. The volume of gas to be employed for combustion
must be adapted to the quantity of the barium carbonate preci-
pitated ; 3 or 4 of the 10 litres contained in the cylinder A will
be usually consumed, rarely more than 6. In finishing the ex-
periment, the flask O is run full up to its mark for the last time,
the tap of the aspirator is closed, and the current of the gas is
continued till the pressure-gauge M is in equilibrium. The pinch-
cock s' is then closed at once, arid the volume of gas passed
through the apparatus is now exactly equal to the volume of the
water run out.

Now follows the sweeping-out. The aspirator-tap is once more
opened, and air is drawn through the apparatus, by opening the
pinch-cock s, till the gas remaining in the vessels K and S has
been swept out. This may be said to be done with certainty after
at most two litres of water are run out : and the apparatus is now
fully in trim for a second combustion. The last operation is to
titrate the contents of the receivers W and W by normal oxalic
acid. The diminution of the standard of the baryta-water imme-
diately indicates the volume, expressed in cubic centimetres, of the
carbon dioxide formed, and at the same time that of the methane
originally present.

The calculation is as follows : —

If n = the volume of methane found,

m = that of the aspirated gas (i. e. the run-out water) in

the corrected state,
n + m = the volume of the gas employed for testing,

, . 100«

the amount of methane is - per cent.

This method, when carefully carried out, yields very accurate
results, and admits of estimating the smallest quantities of
methane.

COMBUSTION BY HOT COPPER OXIDE. 169

Application : —

Estimation of methane in fire-damp and in the up-cast currents
of coal-pits and other gaseous mixtures containing very little
methane ; estimation of all carbon compounds, combustible to carbon
dioxide, as carbon monoxide, hydrocarbons, coal-gas, empyreumatic
products, benzene, carbon disulphide, carbon oxysulphide when
present in small quantities in the air of dwelling -houses, factories,
dry ing -stoves, heating-apparatus, fyc.

EXAMPLE : —

Up-cast current of a coal-pit.
Barometer (B), 726 millims.
Temperature (t\ 23°.
Titre of oxalic acid normal :

1 C.C. = i C.C. UU2 = I C.C.

Titre of baryta-water empirical :

1 c.c. = 0'97 c,c. normal oxalic acid.

Volume of gas employed = 4 litres

or corrected (m) 3422*5 c.c.

Baryta-water employed : —

receiver 1, 25*0 c.c.
„ 2, 25-0 „

50*0 „ = 48'5 c.c. normal oxalic acid.
Oxalic acid employed for re-titration : —

receiver 1, 13*9 c.c.
,, 2, 23-5 „

37-4 c.c.

Difference (w) = iri c.c.
Found :

100/i 100x11*1
^M, = 11-1 + 3422-5 =°'323 Per cent- methane by Volume'

Of course the process as described may be modified in various
directions, according to the object aimed at. Thus the gas may
be transported in a glass vessel, or brought into the apparatus
directly from the place where it is generated. Where carbon
disulphide is present it may be removed by interposing between

170 ON APPARATUS AND METHODS.

the copper oxide and the baryta-water a heated layer of lead chro-
mate. Where there are several compounds present which furnish
carbon dioxide on combustion, and the object is to estimate the
total contamination caused by them, it is necessary to have an idea
of their relative quantities. If, for instance, the contamination of a
room by coal-gas is to be estimated, we must consider that the
average composition of coal-gas is as follows, for 100 vols. of gas : —

4 vols. ethylene gas, burned to 8 vols. CO2.

1 „ benzene vapour ,, „ 6 „ ,,

8 .,, carbon monoxide, „ ^8 ,, „

35 „ methane, ,, „ 35 „ „

48 vols. burned to 57 vols. CO2.

Hence 1 c.c. of carbon dioxide found corresponds to 1*75 c.c.
coal-gas.

171

APPENDIX.

1. International Atomic Weights, as fixed for 1902 by the
German Chemical Society.

O = 16
H= 1-008

H=l

0=15-88

0 = 16
H= 1-008

H=l

0=15-88

Aluminium Al
Antimony Sb

27-1
120

26-9
119-1

Xeodymium
Neon

143-6
20

142-5
19-9

Argon A

39-9

39-6

Nickel ...

58-7

58-3

Arsenic As

75-0

74-4

Niobium

IVb

93-3

Barium Ba

137-4

136-4

Nitrogen

14-04

13-93

Beryllium Be
Bismuth Bi

9-1

208-5

9-03
206-9

Osmium
Oxvgen

191
16

189-6

15-88

Boron B

10-9

.Palladium

106

105-2

Bromine Br

79-96

79-36

Phosphorus

31-0

30-77

112-4

111-6

Platinum

194-8

193-3

Caesium Cs

133

132

Potassium

39-15

38-86

Calcium .... Ca

40-1

39-8

Praseodymium ..

140-5

139-4

Carbon C

12-00

11-91

Rhodium

103-0

102-2

Cerium Ce

140

139.

Rubidium

85-4

84-76

Chlorine . .. Cl

35-45

35-18

Ruthenium

101-7

100-9

Chromium Cr

52*1

51-7

Samarium

150

148-9

Cobalt Co

59-0

58-56

Scandium

44-1

43-8

Copper Cu

63-6 -

63-1

Selerium

79-1

78-5

Erbium Er

166

164-8

Silicon

28-4

28-2

Fluorine F

18'9

Silver

Aff

107-93

107-12

Gadolinium Gd
Gallium Ga

156
70

155
69-5

Sodium ".
Strontium

23-05

87-6

22-88
86-94

Germanium Ge

71-5

Sulphur

32-06

31-83

Gold : Au

197-2

195-7

Titntalium

T»

183

181-6

Helium He

Tellurium

127-6

126-6

Hydrogen H

1-008

Thallium

201-1

202-6

Indium ... . In

114

113-1

Theriuin

232-5

2:30-8

Iodine I

126-85

125-90

Thulium

171

170

Iridium . Ir

193-0

191-5

Tin ...

118-5

117-6

Iron Fe

55-9

55-5

Titanium ..

43-1

47-7

Krypton Kr

81-8

81-2

Uranium

239-5

237-7

Lan thamum La
Lead Pb
Lithium Li

138
206-9
7-03

137

205-35
6-98

Vanadium
Wolfram! uin
Xenon

Va
W

51-2
184
128

50-8
182-6
127

Magnesium Mg

24-36

24*18

Ytterbium

173

172

Manganese Mn

55-0

54-6

Yttrium

88-3

Mercury Ha

200-3

198-8

Zinc

65-4

64-9

Molybdenum. ... Mo

96-0

95-3

Zirconium

90'7

90'0

172

TECHNICAL GAS- ANALYSIS.

2. Litre-weights of Gases and Vapours.

Name of the gas.

Molecular
formula.

Acetylene C2H2

Air (atmospheric)

Ammonia H3N

Antimoniuretted hydrogen i H:jSb

Arseniuretted hydrogen HHAs

Benzene | C(JH6

Bromine Br0

Butane C4H10

Butylene i C4H8

Carbon monoxide | CO

Carbon dioxide CO2

Carbon disulphide CS0

Carbon oxychloride < CO"C12

Carbon oxysulphide ; COS

Chlorine i C12

Cyanogen (CN)2

Ethane C.,HG

Ethylene C^H4

Fluorine F2

Hydrogen H2

Hydrogen bromide HB

Hydrogen chloride HC1

Hydrogen cyanide HCN

Hydrogen fluoride HF

Hydrogen iodide HI

Hydrogen sulphide H>S

Methane CHt

Nitrogen N2

Nitrogen protoxide N2O

Nitric oxide NO

Nitrogen trioxide (hypothetical) N2O.<

Nitric peroxide * i NO2

Nitrosyl chloride NOC1

Oxygen O,

Phosphoretted hydrogen | H3P

Propane ' C3HS

Propylene C3H6

Silicon tetrafluoride - SiF4

Sulphur dioxide ' SO2

Water H,O

1 litre of the

gas in the

normal state

weighs.

grams

1-16148

1-29315

0-76163

5-6040

3-48801

3-48563

7-14259

2-59161

2-50355

1-25058

1-96519

3-40098

4-41799

2-68250

3-16742

2-32653

1-34058

1-25103

1-70684

0-08955

3-61607

1-62848

1-2091

0-89820

5-71067

1-52147

0-71506

1-25461

1-96923

1-34192

3-38051

2-05054

2-92563

1-42923

1-52058

1-96727

1-87654

4-68083

2-86115

0-80458

* This compound, as actually existing at temperatures not too much above the
ordinary temperature, is a mixture of molecules of N2O4 and NO2. The calculation
has been made on the base of the formula NO2.

APPENDIX. 173

3. Solubility of Gases in Water.

1 vol. water of 20° C. absorbs the following volumes of gases,
reduced to 0° C. and 760 mm. barometric pressure : —

Air, atmospheric 0*01869

Ammonia 054*0

Carbon monoxide 0*02319

Carbon dioxide 0*90140

Chlorine ., 2*15650

Ethane 0*04724

Ethylene 0*14880

Hydrogen 0*01819

Hydrogen chloride 445*0

Hydrogen sulphide 2*90530

Methane 0*03308

Nitrogen 0*01542

Nitrogen protoxide 0*06700

Nitrogen oxide 0*04706

Oxygen 0*03103

Propylene 0*22050

Sulphur dioxide , 39*374

174

TECHNICAL GAS-ANALYSIS.

f lifil

* H- * -,, * „, H. . •: * H. *

i " li|'H

"3 t- t~ O •* 70 C5 00 ; O (71 -H

i Jij^i

j „ , "^ *"

•3- -° 03 us *• . o v« /' *o '<* to <# co • ->e

ll|||||

o3 °° O w

•SOOOOOOO |OCO
p>

e*i ^
0 *•> ' 0

o> J ^2-J
| ^ *3 S §

rg •* O=l OCT 00 €(» ^H •* O " <M CO «0

"* £ i §

=1 |l

•

^ IT* t^» ^O ^ *O O^ CO CO CO (71 T— *

•Q •* <71 00 00 (71 TH •* | \« « ffl

• S ' t» ,_, ^i,

§ " 1 Ml

"3 (71 ?3 O CO • O ^T" C"! •* CO- CO

X!
| 1

tn r—

S 1

I"5 i

o5 S <3^ •

-§ <71 (71 (71 (71 (71 (71 (71 <71 (71 (71 C1

§1

&, ft wf K o ^ w, - w" w. ^

o d Q" o o d o ffl o o" o

<M
O
0

r2 :

: ° :
: i 2 c

1 I s j I § i r • i 1 1

^S ST;^ s-^^J Is

§§^-^^5553^2

•**j r*i r*i ^n o M M i^l ?j r i r i

APPENDIX.

175

5. Heat of Combustion of Solid, Liquid, and Gaseous Bodies,

for 1 kilogram of substance, expressed in calories, one of which
= the heating of 1 kilogram water from 0° to 1° C.

1 kilog. substance.

Burning to

Gives off
calories.

Observer.

Acetylene C H

2CO,4-H2O

11945-0

Thoinsen.

(liquid)
2CO2+H2O

11,529-6

Arsenic

(steam)
AS2O3

1,030-5

Benzene, C0HG

6C02+3H20

(liquid)"
6CO.,+3H O

10,330-7
9915-3

Bismuth..

(steam)
BLO.>

95-5

Woods.

Calcium, CaO ..

CaO

3,284-0

Favre and Silbermann.

Carbon: wood-charcoal
sugar-coal

C02

2,473-0
8,080-0
8,039-8

gas-retort coal
blast-furnace graphite
natural graphite
diamond.

8,047-3
7,762-3
7,7966
7 770-1

Carbon monoxide, CO
Carbon disulphide, CS2
Copper

2SO,+CO2
Cu2O "

2,441-7
3,400-0
321-3

Thomsen.
Favre and Silbermann.
Thomsen

CuO

593-6

Joule

Ethane, C H6

9CO2+3H O

12 444-4

Thomsen

(liquid)
2CO2+3H2O

11,364-3

Ethylene, C2H4

(steam)"
2CO2+2H O

11967-1

(liquid)
2CO2+2H O

11 185-9

Hydrogen

(steam)
H O (liquid)

34 180-0

H O (steam)

28 780-0

Iron ....

FeO

1 352 6

Favre and Silbermann

Fee3%

1,582-0
2 028 0

» »

Lead

PbO3

243-0

Thomsen.

Magnesi um

MgO

6,077-5

Manganese

MnO

I 724-0

MnO2

2113-0

Mpr*cury

He O

1055

HgO

153-3

Methane, CHt

CO2+2H2O

13,345-6

Nitrogen

(liquid)
CO2+2H2O

(steam)
N0O

11,995-6
— 654-3

1 541-1

— 143-^

176

TECHNICAL GAS-ANALYSIS.

Table (continued).

I kilog. substance.

Burning to

Gives off
calories.

Observer.

Nitrogen monoxide N O

- 564-3

Thorn sen

Nitric oxide NO

NO2

652-3

Phosphorus

P2O-

5,964-5

" i

Potassium

K O

1 745-0

Woods

Propane

3CO +4H0O

12,125-0

Thomson.

(liquid)
3CO0+4H,O

11,136-3

Propylene, C3HG

(steam)
3CO,4-3H2O

1 1 ,790-4

(liquid)
3CO2+3H O

1 1 ,019-0

Silicon

(steam)
SiO2

7,830-0

Silver

Ag,O

27-3

Sodium .

Na,O

3,293-0

Woods.

2 221-3

Thomson

inonoclinic

2,241-4

Sulphuretted hydrogen

SO24-H7O

2,741-0

Favre and Silbermann.

j> »i
Tin

(liquid)
S02+H20
(steam)
SnO

2,457-0
573-6

j> )>
Andrews.

SnO2

1,147'0

Z^c

ZnO

1 314-3

Thomsen.

6. Standard Solutions for Technical Gas-analyses.

1 vol. gas at 760 I
millims. and 0°, Formula,
dry.

Indicated by 1 vol. normal
solution, containing per litre

j grains

2*5075 Potassium hydroxide

KOH

Carbon monoxide...
» ?> •••

Carbon dioxide
Chlorine . . .

CO
C02

5-6296 Oxalic acid, crystallized...
14*0943 Barium hydroxide, cryst.

5-6296 Oxalic acid, crystallized...
14-0943 Barium hydroxide, cryst.

4*4216 Arsenious acid dissolved

C2H204, 2H20
Ba(OH)2, 8H20

C2H204, 2H20
Ba(OH)2, 8H2O

in sodium bicarbo-
nate

As Oo

11*3353 Iodine, dissolved in po-

tassium iodide

APPENDIX. 177

Table (continued).

^r.-,s? k^ is^^-;^.

! grams
Hydrogen cbloride...j JIC'l 4 '8215 Silver, dissolved in nitric

acid Ag

,, ,, ,, 3'4028 Ammonium sulphocya-

nide "... CNS, NITl

,, ,, ... ,, 2'o075 Potassiuui hydroxide... IvOIl

.MVthane Gil l 5-6296 Oxalic acid, -erystalized C.JT.,0^ 2H.O

, 14-0943 Barium hydroxide cryst. Ba(OH),, 8H,0

Nitrogen trioxide ... N .,(")., 5 -(5230 Potassium permanga-
nate KMnOt

Nitrogen oxide NO 4'2406

Sulphur dioxide ... SO2 11 '3353 Iodine in Potassium

iodide

5-0166 Potassium hydroxide... KOH

7. Talk for Reducing Volumes of Gases to the Normal Slate.
By Professor Dr. LEO LIEBERMANN.*

[Communicated by permission of the Author.]
Instructions for Use.

Suppose the volume of a gas to have been found =26'2 c.c. at
742 mm. barometric pressure,, 18° C. temperature, saturated with
moisture. In order to reduce it to the -normal state (760 mm.,
0° C., dr^), we proceed as follows : —

1st. Look out the degree 18 (columns 1 and 4) and deduct the
tension of aqueous vapour given, =15*3 mm., from the observed
pressure = 742'0 :

742-0-15-3 =

2nd. Now find the volume which 1 vol. of the gas would have
at the pressure of 726'7 mm. by looking out seriatim the figures
7, 2, 6, and 7 in column 2 at the 'temperature 18°, and placing the
numerical values, to be found opposite those figures, in the same
column, multiplying them seriatim by 100, 10, 1,0'1 ; whereupon
they are added up, thus :—

* For permanent laboratory use G. Lunge has published tables, to bo
hung on the wall, for reducing gas-volumes to normal conditions and for gas-
volumetric analysis (Brunswick, 1897).

178

TECHNICAL GAS-ANALYSIS.

0-0086408 X 100 = 0-80-1-08
0-0024688 x 10 =0-024088
0-0074064<x 1 =0-0074004
0-0080408 x 0-1 =0-00086408

3rd.

0-89703848
The corrected volume of a cubic centimetre is lastly

multiplied by the number of the c.c. previously found ; that is in
the present case :

0-89703848 x 26-2 = 23-502 c.c.

Temperature

°C.

Pressure in

ITlillilBS.

mercury.

Volume at 0° and
7GO mm.

Tension of aqueous vapour
in millimetres of mercury
for degrees C.

0 1 0-0013157

2 0-0026315

0 3 0-0039473

4 0-0052631

5 0-0065789

0° =4-5

6 0-0078946

7 0-0092104

8 0-0105262

9 0-0118420

1 0-0013109

2 0-0026219

3 0-0039328

4 0-0052438

0-0065548

l°=4-9

0-0078657

0-0091767

8 0-0104876

0-0117986

0-0013061

2 0-0026123

0-0039184

0-0052246

0-0065307 2° -5-2

0-0078369

0-0091430

0-0104492

0-0117553

APPENDIX.

Table (continued).

179

Temperature

Pressure in
millims.
mercury.

Volume at 0° and
760 mm.

0-0013013
0-0020020
0-0039039
0-0052053
0-0065066
0-0078079
0-0091093
0-0104106
0-0117119

Tension of aqueous vapour

in millimetres of mercury

for degrees C.

3° =5-6

0-0012965
0-0025930
0-0038895
0-0051860
0-0064825
0-0077790
0-0090755
0-0103720
0-0116685

0-0012916
0-0025833
0-0038750
0-0051667
0-0064584
0-0077501
0-0090418
0-0103335
0-01 16252

4° =6-0

5° =6-5

0-0012868
0-0025737
0-0038606
0-0051474
0-0064343
0-0077212
0-0090080
0-0102949
0-0115818

6° =6-9

180

TECHNICAL GAS-ANALYSIS.

Table (continued).

Temperature

0°C.

Pressure in t .-.0 , Tension of aqueous vapour
millims. ^ ohl™ at ° and i in millimetres of mercurv
mcrcury. for degrees C.

0-0012828

0-0025056

0-0038484

0-0051312

0-0064140

7° = 7-4

0-0076968

0-0089796

0-0102624

0-0115452

0-0012783

0-0025566

8 3 0-0038349

8 4

0-0051132

8 5

0-0063915

8° = 8-0

8 6 0-0076698

8 7 0-0089481

8 8 0-0102264

9 0-0115047

0-0012737

9 2 0-0025474

3 0-0038211

4 0-0 .'50948

5 0-0063685

8° = 8-5

6 0-0076422

7 0-0089159

8 0-0101896

9 0-0114633

10 1

0-0012692

10 2 0-0025384

10 3 0-0038076

4 0-0050768

] 0 5 0*0063460

10° = !)•!

10 6 0-0076152

10 7 0-0088844

10 8 0-0101536

10 9

00114228

ATTEND IX.

Table (continued).

181

Temperature

U°C.

11
11
11
11
11
11
11
11
Jl

Pressure in
milliius.
mercury.

Volume at 0° and
760 mm.

0-0012648
0-0025296
0-0037944
0-0050592
0-0063240
0-0075888
0-0088536
0-0101184
0-0113832

Tension of aqueous vapour

in millimetres of mercury

for degrees C.

11° = 9-

12
12
12
12
12
12
12
12

13
13
13-
13
13
13
13
13
13

14
14
14
14
14
14
14
14
14

U-U012603
0-0025206
0-0037809
0-0»'50412
0 0063015
0-0075618
0-0088221
0-0100824
0-0113427

0-0012559
0-0025118
00037677
0-0050236
0-0062795
0-0075354
0-0087913
0-0100472
0-0113031

0-0012516
0-0025032
0-0037548
0-0050064
0-0062580
0-0075096
0-0087612
0-0100128
0-0112644

12° = 10-4

13° = ll'l

14° = 11-9

18.2

TECHNICAL GAS-ANALYSIS.

Table (continued}.

Temperature
0°C.

Pressure in
Hiillims.
mercury.

TT i Tension of aqueous \i
Volume at 0° and in niillimetros of mer
'60mm- for degrees 0.

0-001:2472

0-0024944

0-0037416

0-0049888

0-0062360 15° = 12-7

0-0074832

0-0087304

0-0099776

0-0112248

0-0012429

<•)

0-0024858

0-0037287

0-0049716

0-0062145 16° = 13-5

0-0074574

0-0087003

0-0099432

0-0111861

0-0012386

0-0024772

0-0037158

0-0049544

0-0061930 17° = 14-4

0-0074316

0-0086702

0-0099088

0-0111474

0-0012344

0-0024688

0-0037032

0-0049376

0-0061720 18° = 15-3

0-0074064

0-0086408

0-0098752

0-0111096

APPENDIX.

Table (continued).

185

Temperature
°0.

Pressure in
millims.
mercury.

Volume at 0° and
760 mm.

Tension of aqueous vapour
in millimetres of mercury
for degrees C.

0-0012301

0-0024602

0-0036903

0-0049204

0-0061505

19° = 16-3

G 0-0073806

7 0-0086107

8 0-0098408

9 0-0110709

20
20
20
20
20
20
20
20
20

21
21
21
21
21
21
21
21
21

22
22
22
22
22
22
22
22
22

0-0012259
0-0024518
0-0036777
0-0019036
0-0061295
0-0073554
0-0085813
0-0098122
0-0110331

0-0012218
0-0024436
0-0036654
0-0048872
0-0061090
0-0073308
0-0085526
0-0097744
0-0109962

00012176
0-0024352
0-0036528
0-0048704
0-0060880
0-0073056
0-0085232
0-0097408
0-0109584

20° = 17-4

21°= 18-5

22° = 19-6

184

T E C H N 1 C A L G A S- A N A L Y S 1 S .

Table (continued).

Pressure in

Temperature ^^^

mercury.

-jr , Tension of aqueous vapour
Volume at 0° and jn miUimet^8 of mer£ry

for degrees C.

23 1

0-0012135

23 2

0-0024270

23 3

0-0036405

23 4

0-0048540

23 5

0-0060675

23° = 20-9

23 6

0-0072810

0-0084945

0-0097080

0-0109215

24 1

0-0012094

24 2

0-0024 J 88

24 3

0-0036282

24 4

0-0048376

0-0060470

24° = 22-2

0-0072564

0-0084658

0*0096752

0-0108846

25 1

0-0012054

25 2

0-0024108

25 3

0-0036162

25 4

0-0048216

25 5

0-0060270 25° = 23'5

0-0072324

0-0084378

0-0096432

0-0108486

0-0012013

0-0024026

0-0036039

0-0048052

26 5

0-0060065 26° =25-0

2G 6

0-0072078

0-0084091

0-0096104

0-0108117

APPENDIX.

Table (continued).

185

Temperature

27
27
27

27
27

28
28
28
28
28
28
28
28
28

29
29
29
29
29
29
29
29
29

30
30
30
30

30
30
30
30
30

Pressure in
millims.
mercury.

Volume at 0° and
760 mm.

Tension of aqueous vapour
in millimetres of mercury
for degrees C.

0-0011973

0-0023946

0-0035919

0-0047892

0-0059865

27° = 26-5

0-0071838

0-0083811

0-0095784

00107757

0-0011933

0-0023860

0-0035799

0-0047732

0-0059665

28° = 28-1

0-0071598

0-0083531

0-0095464

0-0107397

0-0011894

0-0023788

0-0035682

0-0047576

0-0059470

29° = 29-8

0-0071364

0-0083258

0-0095152

0-0107046

0-0011855

0-0023710

0-0035565

0-0047420

00059275

30° = 31-6

0-0071130

0-0082985

0-0094840

0-0106695

ALPHABETICAL INDEX.

Abaorbable constituents, titrating them
while measuring the total volume of
the gas, 49 ; measuring the non-absorb-
able residue, 50.

Absorbents for carbon dioxide, 65 ; for
heavy hydrocarbons, 66; for oxygen, 68 ;
for carbon monoxide, 73 ; for nitrogen,
75.

Absorbing agents for gases, 65 ; previous
saturation with gases, 65; use in Bunte'n
burette, 84.

Absorbing-coils, 117; flasks, 117; cylin-
ders, 118 ; pipettes, 96.

Absorption processes, 2, 4, 65 ; apparatus
for, 75.

Acetylene, (Mi, 67, 102, 125, 128.

Arid smoke, 110, 116, 124, 128.

Acids, total, iu roasting-gases, 110, 116,
124.

Agitating-vessels for nitrometer, 41, 44.

Air, removal from connecting-pipes, 5.

Ammonia, estimation byHempel'.s bin'ette,
101 ; by Drehschmidt's apparatus, 121.

Ammonium salts, 40.

Analytical processes for gases, 2.

Aqueous vapour, tension, 178.

Argon, 75.

Arndt's econometer, 55.

Arrangement of laboratory, 56.

Aspirating-bottles, 16.

Aspirating-tubes, 5 ; of glass, 7 ; of por-
celain, 8 ; of metal, 8 ; cooled by water,
0, 13 ; by pumps, 11 ; by steam, 12.

Aspirators, 11 ; for steam, 12 ; water, 13 ;
Muencke's, 17; of zinc, 18; Bonny's
automatic, 19; DrehschrnidtV, 119.

Atomic weights, 171.

Average sample, 5.

Barometer, 25.

Benzene, estimation of vapours, 63 ; ab-
sorbents for, 66 ; estimation, 67 ; by
bromine, 67 ; by Hempel's burette, 102.

Blast-furnace gases, s. Producer-gas.

Bleaching-powder, estimation, 40.
Bromine water as absorbing agent, 67.
Bunsen's water-air pump, 13.
Bunte's gas-burette, 82.
Burette, s. Gas-burette.

Capillary, palladium, 140 ; platinum, 160.

Carbon in iron and steel, -45.

Carbon dioxide, liquid, 79.

Carbon dioxide, 40, 45 ; absorbents for,
65 ; estimation by Winlder's burette,
78 ; in electrolytical chlorine, 81 ; by
Honigmanu's burette, 81 ; by Bunte's
burette, 85, 86, 87; by Scheibler's
apparatus, 86 ; by Orsat's apparatus,
88, 80, 90, 91 ; by Hempel's burette,
100, 101 : by Hesse's apparatus, 105 ; by
Lunge & Zeckendorff's method, 114,
115.

Carbon disulphide, 62, 125.

Carbon monoxide, formation in absorption
of oxygen by pyrogallol, 71 ; absorption
by cuprous chloride, 73 ; detection in
small quantities by palladium chloride,
74 ; estimation by Bunte's burette, 86 ;
by Orsat's apparatus, 89, 90 ; by
Hempel's burette, 102 : by combustion
with palladium, 145.

Cathetoraeter, 26, 31.

Chimney-gases, soot in, 61 ; CO in, 145.

Chlorine, 81, 101, 107, 123; electrolytic,
81, 107; liquid, 81.

Chromium protochloride, for absorbing
oxygen, 68.

Coal-gas, 137, 143, 153.

Coal-pit gases, methane in, 153, 156, 159,
169.

Collecting gases, 16 ; in liquids, 38 ;
vessels for, 20 ; of india-rubber, 21 ;
of glass, 21 ; of zinc, 21.

Combustion, 2, 4 ; estimation by, 129 ; by
explosion, 131 ; by palladium, 139,
146; by red-hot platinum, 149, 151,
156, 160; by copper oxide, 165; changes
of volume by, 174; heat of, 175.

188

INDEX.

Confining-liquids, 3, 30, 34.
Connecting-tubes, removal of air from, 5.
Contraction of volumes, 2; tables for,

174.

Copper for absorbing oxygen, 72.
Copper oxide, combustion by, 164.
Coquillion's grisou meter, 150.
Corrected volumes, 1, 24; approximated,

29; tables of, 177.
Correction apparatus, 20.
Cuprous chloride as absorbent for carbon

monoxide, 73, 90.
Cyanhydric acid, 10(5.

Dasy meter, loo.

Deacon process gases, 106.

Decomposition- flaak for nitrometer, 31).

Dissociation, 10.

Drehsehmidt's aspirating-tube, 10 ; ab-
sorbing-apparatus, 118 ; platinum-
capillary, 160.

Dust, 59.

Dynamite, 35.

Econometer, 58.
Ethane, 67.
Ethylene, 6(5, 67, 102.
Eudiometry, 131.
Experimental gas-meters, 47.
Explosion, estimation by, 131.
Explosion-pipette, 131.

Ferricyanides, 40.

Ferro-carbonyl, 64.

Ferrous tartrate for absorbing oxygen,
68.

Filtering gases, 59.

Finkener's water-jet pump, 15.

Fire-damp, 138, 145, 146, 169.

Fittings of laboratory, 56.

Flue-dust, 59,

Foot-blowers, 11.

Furnace-gases, soot in, 61 ; s. Carbon di-
oxide, &c.

Fuming sulphuric acid for absorbing
hydrocarbons, 66 ; pipette for, 96.

Gases, volume of, 23 (s. Eeduction &
Volumes); litre-weights, 172; dissolved j
in liquids, collection of, 38 ; solubility j
in water, 173; change of volume in ;
combustion, 174 ; table for reducing to
normal state, 177.

Gas-analysis, expression of results, 1 ;
technical, 3; standard solutions for,
176.

Gas-balance, Lux's, 54.

Gas-burette, as collector of gases, L6 ; for
measuring gases, 29 ; with jacket, 30 ;
Winkler's, 75 ; modification by Langc
for liquid carbon dioxide, 79 ; Honig-
mann's, 81 ; Bunte's, 82; Lindemann's,
92 ; Hempel's, 93.

Gas-meters, 45; wet, 46; experimental,
47 ; arbitrarily divided, 47 ; auto-
matically stopped, 47 ; gauging, 48.

Gas-pipettes, 96.

Gas-volumeter, 41.

Gasometry, 1.

Gasvolumetric analysis, 1, 38.

Geissler water-jet pump, 15 ; three-way
tap, 33.

Glycerine as confining-liquicl, 30, 35.

Graviraetrical estimation of gases, 4, 51.

Gravity, specific, estimation of, 51.

Greiuer & Friedrichs's three-way tap, 33,

Grisoumeter, 150.

II.

Hand-blowers, 1 1 .

Heat of combustion, table, .175.

Hempel's gas-burette, 93 ; pipettes, 96 ;

manipulation, 99.

Hesse's apparatus for titration, 103.
Honigmann's gas-burette, 81.
Hydrocarbon vapours, 63.
Hydrocarbons, heavy, absorbents for,

66.
Hydrochloric acid, 101, 106, 111, 123;

together with chlorine, 107, 124.
Hydrogen, combustion of, 130, 133, 135,

'137; by palladium, 142, 143; by

Lunge-Orsat's apparatus, 146.
Hydrogen chloride, ,s. Hydrochloric acid.
Hydrogen peroxide, analysis by means

of, 40.

Hydrogen pipette, 97, 133, 134.
Hydrogen sulphide, 101, 125, 128.
Hypochlorites, 40.

India-rubber aspirators and pumps, 11;
collecting-vessels, 21.

Kiln-gases, 107, 109, 110.
Koerting's water-jet pump, 14.

Laboratory, arrangement and fittings,

56.

Level-bottle and tube, 30.
Lime-kiln gases, s. Carbon dioxide.
Lindemann's apparatus for estimating

oxygen, 92.

INDEX.

189

Liquid admixtures in gases, 59, 61.

Litre-weights of gases, 172.

Lunge's modification of Winkier's bu-
rette for liquid carbon dioxide, 79.

Lunge's nitrometer, 33 ; gas-volumeter,
41 ; mercury-sealed tap, 43 ; straight-
edge, 43 ; minimetric method (with
Zeckendorflh, 112; analysis of acety-
lene (with Cedercrentz), 128 ; modi-
fication of Orsat's apparatus, 146.

Lux's gas-balance, 54-.

Manganese ore, estimation, 40.

Measuring-vessels, 3.

Meniscus, 30.

Methane, absorption by sulphuric acid,
C>7: estimation. 2; combustion, 130,
133,130, 138; by grisouuieter, 150;
by Winkier's apparatus, 156 ; in
coal-pit air, 156 ; by platinum-
capillary, 163 ; by copper oxide, 169.

Minimetrical method, 112.

Moisture, influence on volumes of gases,

Naphthalene vapours, estimation, 61.
Natural gas, 154.
Nickel carbouyl, 04.
Nitrates, estimation, 34.
Nitric oxide, 101, 104.
Nitrites, 34.

Nitrogen, absorbent for, 75.
Nitrogen oxides, 34.
Nitrogen trioxide, 101, 112, 123.
Nitroglycerine, 35, 64.
Nitrometer, 33.

Nitrous acid, 101, 102, 112, 123.
Nitrous oxide, 101. *6Sr /£?-
Nitrous vitriol, 34.

Normal solutions for gas-analysis, 176.
Normal volume of gases, 20, 43. 48 ;
tables for reducing gases, 177.

Oil as confining-liquid, 30. 35.

Olefins, 66.

Orsat apparatus, 87.

Orsat-Lunge apparatus, 140.

Oxygen required for combustion of
gases, 129.

Oxygen, absorbents for, 08 (phosphorus,
68; pyrogallol, 70); estimation in air,
78; by Bunte's burette, 85 ; by Orsat's
apparatus, 88, 89, 90 ; by Lindemann's
apparatus, 92; by Hempel's burette,
101,102; by combustion with hydro-
gen, 130; 'by means of palladium,
145.

Palladium, heated, for combustion of
gases, 139.

Palladium-asbestos, 139.

Palladium chloride as reagent for carbon

monoxide, 74.

I Permanganate of potassium, 40, 112.
! Petroleum as contining-liquid, 30.
| Phosphoretted hydrogen, 128.
', Phosphorus for absorbing oxygen, 68.
| Pipettes, s. Gas-pipettes.
I Pit-gases, s. Coal-pit gases.

Platinum, heated, for combustion of gases,

149.
j Platinum-asbestos, 140.

Platinum-capillary, Drehschmidt's, 160.

Potassium hydroxide for absorbing

carbon dioxide and other gases, 65.
1 Pressure, correction for, 25, 43, 177.

Producer-gas, 143, 147, 153, 102.

Propylene, 66.

Pyrites-kiln gases, 107, 109, 110.

Pyrogallol for absorbing oxygen, 71.

Pyroxyline, 35.

Reading of volumes, 31.
Reduced volumes, 26 ; tables of, 177.
Reduction -apparatus, 26.
Reduction- tube in gas-volumeter, 43.
Reich's apparatus, 107.
Respiration-gases, s. Carbon dioxide.
Running down of water in measuring-
tubes, 31.

Saline solutions as confming-liquids, 30,

35.
SaUcake-furnace gases, s. Hydrochloric

acid.
Samples of gases, collecting, 11 ; vessels

for, 20.

Sampling gases, 5 ; selection of place, 7.
Schilling's apparatus for estimating the

specific gravity, 52.
Side-flask for nitrometer, 39.
Smoke, 61.

Sodium hydroxide as absorbent, 66.
Solid admixtures in gases, 59, 60.
Solubility of gases in water, 173.
Soot, 59, 60.
Specific gravity of gases, estimation 51 ;

tables, 172.
Sprengel pump, 13.

Standard solutions for gas-analysis, 170.
Steam-jet aspirators, 12.
Straight-edge with spirit-level, 45.
Sulphur in gas, 127.
Sulphur acids, total, 110, 110, 124, 128.
Sulphur dioxide, 107, 109, 116, 128.

190

INDEX.

Sulphur trioxide, 110, 111, 124.
Sulphuretted hydrogen, 101, 125, 128.
Sulphuric acid, estimation in gases, 62.
Sulphuric acid, fuming, as absorbent, 60

Tap, three-way, 33.

Tar, estimation in gas, 62.

Technical gas-analysis, 3.

Temperature, reductions for, 23, 24, 25,

26, 43 ; tables, 177.
Ten-bulb tube, 117.
Tension of aqueous vapour, 178.
Three-way tap, 33.
Titrating absorbable constituents, 49, 50 ;

standard solutions, 176.
Titration, estimation of gases by, 48.

Urea, 40.
Ureometer, 41.

TJ.

Vapour, aqueous tensions, 178.
Vitriol-chamber gases, 112, 123.

Volhard-Fresenius absorbing-flask, 117.

Volumes, expression by, 1 ; normal, re-
duction to, 1 ; tables for reducing, 177;
influenced by temperature and pressure
and moisture, 23, 24 ; formula for
correcting, 26 ; changes by combustion,
174.

Water, not to pass gases through, 20; as
confming-liquid, 30, 34 ; supply for
laboratory, 57, 58.

Water, estimation in gases, 62.

Water-gas, 135, 143, 147.

Water-jacket for burettes. 30.

Water-jet pumps, 13.

Weight, estimation by, 125.

Winkler's gas-burette, 79 ; modified, 95 ;
absorption-coil, 116 ; apparatus for
combustion with platinum, 154 ; for
coal-pit air, 156.

Working-benches, 57.

Zinc aspirators, 18.

THE END.

PRINTED BY TAYLOR AND FRANCIS
UED LION COURT, FLEET STREET.

Y£ 94080

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 "Handbook of technical gas-analysis"

See other formats