May 4, 1917.

THE COLLIERY GUARDIAN.

851

THE STANDARDISATION OF BENZOL
WORKS TESTS.

At the meeting of the Northern section of the Coke
Oven Managers’ Association, at Darlington on
Saturday last, a paper on the above subject was read
by Mr. Thomas B. Smith, of Messrs. Pease, and
Partners Limited, Bowden Close.

Mr. Smith stated that when he undertook the task
of preparing a paper upon that subject, he did not
fully appreciate the magnitude of that field of work.
The number of important sections rendered it a very
difficult matter to choose one which could claim pre-
ference in importance to the others. • For example,
there were:—(1) The formation of a standard test for
the distillation of benzol works products, in which, at
the present time, retorts and distilling flasks of various
sizes were both used, with the thermometer sometimes
in the liquid and sometimes in the vapour. (2) The
calculation of the “ daily make ” of crude benzol at a
standard strength. In many cases the “ actual make ”
was calculated to terms of 65 per cent, at 120 degs.
Cent, by multiplying by a percentage at 120 degs.
Cent., and dividing by 65. That method was obviously
wrong, since in many instances he had known of a still
being charged with low-strength benzol, the amount of
which had been reported to the Ministry of Munitions
in terms of 65’s, which yielded more gallons of saleable
products than were, reported to be in the still. If they
took a plant making 7,000 gals, of crude benzol per
week (testing 50 per cent, at 120 degs. Cent.), this
would have to be reported to the Ministry of Muni-
tions as 5,385 gals, of 65’s. Now, if the 7,000 gals,
were rectified, about 5,000 gals, of saleable products
were obtained, excluding naphthalene. That amounted
to 92-85 per cent, recovery on the “ reported make ”—
a figure very much too high, and very misleading,
covering up actual losses on the rectification plant,
which, if found out, could probably be stopped. (3)
Lastly, they had the crying need for a standard method
for the testing of the gas before and after scrubbers
for its benzol content. His paper was devoted to the
latter section.

In the testing of coal gas for its content *of the
vapours of benzol and its homologues, they rarely
came across two chemists who carried out the test in
exactly the same manner, there being so many methods
in existence, and each being divided into hundreds of
modifications. Very few methods were in use for
drawing the sample of gas from the main that would
bear a strict scientific examination. Perhaps the most
scientific method in use was to bend a pipe into the
form of a Pitot tube, and fix it in the centre of the
main facing the flow of gas, and as near to the entrance
of the scrubbers as possible. This pipe could then be
led to any convenient distance; but it was advisable
to keep within 10 yds. The end of the pipe was bent,
and immersed in water to a depth just insufficient to
seal it. By that means, the gas could travel through
the pipe at the same speed as that in the main, and
retained a minimum pressure of several inches of
water, which was more than sufficient to overcome the
resistance of the absorption bottles. The pipe was
drilled a few inches above the water seal, and a small
glass Pitot tube was inserted directly connected to the
apparatus. Where the gas passing through the meter
was under pressure or suction, a manometer should be
attached, and the volume of gas corrected to normal
pressure by the usual formula. By that means a good
average sample of gas was continually passing through
the pipe at a constant velocity to within a few inches
of the absorption bottles. It was not an uncommon
practice to leave an absorption test on until 100 litres,
or other pre-arranged quantity, of gas had passed
through the apparatus, the length of time required
varying for each test, and mainly depending upon the
gas pressure. That was unworthy of the chemical pro-
fession, and was one of the greatest mistakes it was
possible to make. A regular flow of gas should pass
throughout the test, which, in the case of a plant work-
ing on the eight-hour shift system, and where it could
be assumed that an equal number of ovens was drawn
per shift, should remain on for eight hours or for a
simple multiple. of that period. Where possible,
perhaps it would be advisable to keep the test on over
a complete coking period.

Each of the principal makers of benzol plant had a
standard method of their own for the absorption of
the hydrocarbons, all these methods having a few
advantages, together with many disadvantages. How-
ever, most of these, if carried out with extreme care
by a competent person, would give very constant
figures, and show fairly accurately the percentage
efficiency of the scrubbers, although they could not
generally be relied upon when the actual content in
the gas of aromatic hydrocarbons was required.
Muller’s test, or “ the increase in weight of oleum
petroleum,” was one of these. This was the test
favoured most in Germany, and, consequently, it was
used on many German-built plants in this country. It
gave very constant figures, but these could not be said
to agree with the actual make of benzol per diem.
However, they fluctuated directly with the latter, and
the test could be relied upon to give intimation if
excessive amounts of aromatic hydrocarbons were left
in the return gas.

The various methods of absorption amTfor the subse-
quent distillation of the absorbent were, in many cases,
hopelessly misapplied. Creosote oil, oleic acid, and
nitrobenzene were suggested as absorbents. Now, to
work successfully on that principle, similar treatment
must be aimed at to that given to the gas on the prac-
tical scale; hence it was obvious that the most suitable
of the three absorbents was the first. It was impor-
tant, however, that the creosote oil should be specially
prepared. Wash oil should be distilled, and the frac-
tion distilling between 240 and 270 degs. Cent, should
be taken. This should be washed with acid and soda,
to remove tar acids, pyridine bases, etc., and allowed
to stand in ice in order to freeze out the naphthalene.
Four bottles should be used, each being of sufficient

capacity to hold 100 c.c. of absorbent, of a pattern
offering a maximum surface of contact. The gas
should first be dried by passing it through two glass
towers containing glass wool (previously damped with
water) and calcium chloride respectively, after which
it should be allowed to flow through the four absorp-
tion bottles (each surrounded with ice) to the meter,
at the rate of eight to ten litres per hour. It was
during the subsequent distillation of the absorbent that
so many chemists relaxed their scientific watchfulness.
He found many distilling the absorbent to about 200
degs. Cent., and then re-distilling the distillate to
120 degs. Cent.; others to 140, 160, or 180 degs. Cent.
The point was, which was correct? Personally, he
said “ None.” If they rectified their crude benzol,
they obtained a fair amount of heavy solvent naphtha
(90 per cent, at 175 degs. Cent.), together with raw
heavy naphtha (90 per cent, at 190 degs. Cent.). These
substances re-distilled up to 160 or 180 degs. Cent.,
and were, therefore, not counted in the yield of crude
benzol. He left it for others to say how much xylol
and light solvent naphthas were also left out by those
using the lower temperatures.

If the content in the gas of the heavier homologues
of benzol were constant, these would, perhaps, be of
some use as comparative tests; but who could guar-
antee them ? His contention was that they would vary
directly with the benzol and toluol. Consequently,
such results were worthless.

As an alternative, he described a method which he
had found to give every satisfaction. The absorbent
was distilled until 50 c.c. of distillate had been
obtained. A 50 c.c. flask was fitted with a miniature
fractionating column and condenser, the latter being
moistened with benzol and allowed to drain; 50 c.c. of
a representative sample of the day’s make of crude
benzol (made on the plant the same day) were placed
in the flask, and distilled up to 160 degs. Cent., correc-
tion being made for the' barometric pressure. The
yield was noted, and the experiment was repeated,
using the 50 c.c. of absorbent distillate. For example,
let them suppose that 100 litres of gas had passed
through the absorbent, which, on distillation and
re-distillation of the distillate up to 160 degs. Cent.,
yielded 3 c.c.; and also that 50 c.c. of crude benzol,
when distilled to 160 degs. Cent., yielded 37-5 c.c.
Then, if 37-5 c.c. (100 per cent, at 160 degs. Cent.)
represented 50 c.c. crude benzol, 3 c.c. (100 per cent,
at 160 degs. Cent.) would represent 4 c.c. crude benzol.
Therefore, there would be 4 c.c. of crude benzol of the
same strength as that made on the plant of the same
day in the absorbent, which represented 100 litres of
gas. To prove the constancy of the figures obtained
by the creosote test and Miiller’s test, he carried out
the two tests simultaneously on the main, and got very
similar results.

He next turned his attention from the mechanical
to the chemical activities of benzol and its homologues.
In making use of chemical reactions, they had a great
factor in their favour. The vapours of the hydro-
carbons were so diluted with the gas—their molecules
were so considerably spread out—that they got what
they might call “ molecular contact,” wherein a much
more perfect contact took place than if the liquid
hydrocarbons were merely poured into the reacting
agent. Benzene, being the nucleus and the first of
the aromatic series, was the most difficult to get to
react chemically, toluene, xylene, etc., in which one or
more of the hydrogen atoms were replaced by alkyl
radicals, being much more easily reacted upon. Conse-
quently, what would easily react upon benzene, would
generally more easily react upon its homologues.
Therefore, he had proceeded to experiment with those
reagents known to react with benzene that were the
most easily obtainable. If benzene -vyere gradually
brought into contact with hot concentrated sulphuric
acid, a process of sulphonation took place, in which the
benzene was converted into its sulphonates with the
liberation of a molecule of water.

His next experiments were made with a mixture of
nitric and sulphuric acids. When benzene was added
to strong nitric acid in the cold, a very vigorous
reaction occurred. Nitrobenzene (oil of mirbane) was
formed, but a molecule of water was liberated for every
molecule of nitrobenzene formed. The water soon
diluted the acid, giving it oxidising instead of
nitrating properties; consequently, a large bulk of
concentrated sulphuric acid should be used to act as a-
dehydrating agent.

Hsing two glass drying towers, containing cotton or
glass wool and calcium chloride respectively, he passed
the gas through four absorption bottles containing
(Nos. 1 and 2) a mixture of nitric and sulphuric acids,
in the proportions of 70 per cent, sulphuric and 30 per
cent, nitric, (No. 3) distilled water, and (No. 4)
approximately twice-normal caustic soda solution, and,
finally, though a IT-tube containing calcium chloride,
to the meter at the rate of 10 litres per hour. When
the gases passed through the first and second bottles,
nitration took place, but nitric acid fumes were given
off. These were absorbed in the second and third
bottles, and any moisture was removed by the U-tube.
The four absorption bottles and the IJ-tube were
weighed before and after the experiment, and the
increase in weight calculated to grammes per unit
volume. If, by any chance, the temperature of the
acids should rise, the higher nitration compounds were
formed, such as ortho-, para-, and meta-dinitrobenzene
and the corresponding dinitro and trinitrotoluene and
xylene compounds; but, since they were working on
the increase in weight, these would not affect the
accuracy of the test. In consequence, the necessity
for ice contact, etc., was dispensed with. The nitra-
tion products would all be found in the first bottle,
floating in a layer at the top ; and the fact that none
was found in the second, proved the great efficiency of
the method. Although this method suited one plant
very well, it might require slightly modifying on
others, e.g., where much ammonia had been left in the
gas, a small absorption bottle containing approxi-
mately normal sulphuric acid should be placed between
the two towers to extract it.

In many cases, the results obtained were reported
as grammes or cubic centimetres per unit volume of
gas. That was a quite correct method, but gave very
little information as to the amount of benzol to be
expected from the carbonisation of a definite weight of
coal. Others made.use of the old assertion that a ton
of coal yielded 10,000 cu. ft. of gas. That might be
correct in a particular instance, but it was obvious that
the amount would vary not only with the class of coal
used, but also with the temperature of carbonisation.
Personally, he was in favour of separate estimation of
the yield of gas at each plant. That, of course, was
an easy matter where a meter was installed, but not so
otherwise. However, the method and formulae pro-
posed by Dr. Pole would be found to give every satis-
faction. Where the plant was working normally on
the same class of fuel, a test every month would
suffice. That would enable results to be reported
which would give information as to actual yields to
be expected.

DISCUSSION.

Mr. L. Crawford (Messrs. Thomas Ness and Com-
pany Limited, Littleburn) stated that he had carried
out numerous tests with the new method put forward
by Mr. Smith. He was inclined to disagree with him
as to the best place from which samples of the gas to
be tested should be taken. Mr. Smith favoured the
centre of the main as the point of maximum velocity.
From that point to the bottom, the velocity decreased,
and, consequently, the suspended matter gravitated to
the bottom side. In that way, he thought Mr. Smith
would get higher results—that was, with mechanic-
ally-carried matter—than the actual average content
of the gas. The speaker’s experiments, carried out
with producer gas, showed that the best point was
about one-third from the periforse, which gave the
mean value. He did not agree that there was bound
to be a certain amount of haphazard working when
the “increase in weight of oleum petroleum” and
other absorption methods were used for the determi-
nation of the amount of benzol. Speaking as a
chemist, there was no need for that. It was the
easiest matter in the world to acquire a constant pres-
sure. He maintained that there was no difficulty in
the creosote oil and nitrobenzene tests. When the
results became abnormal, it was easy to find the diffi-
culty. With regard to the reporting of results, no
works manager returned his benzol in anything but
gallons per ton. Determination of the weight neces-
sarily involved an assumption of the specific gravity,
but it had always to be taken back into units of
volume. Dr. Pole, mentioned by the author, was a
very eminent personage about 1850, but the speaker
did not think that the formula could be applied to
every-day plant. One could not say now, “ I am
going to get so many thousand feet of gas and so many
gallons of benzol.” He had spent a great deal of
time and had had much experience in benzol tests,
and his experience was unfortunate. The results of
the experiments had nearly always been unsatisfac-
tory. Mr. Smith’s method was delightfully simple,
and for that reason he had been suspicious.

Mr. G. H. Johnson (North-Eastern Steel Company
Limited) said that he had invariably found, in his
creosote oil tests in connection with blast furnace gas,
that he used very much more than the quantity men-
tioned by Mr. Smith. He had always used from 200
to 250 c.c. and from 15 to 20 cu. ft. of gas. The results
of tests w’ith oil as it came to the works were low.

Mr. Smith, replying on the discussion, quite agreed
with Mr. Crawford that the centre would probably
not be the best place for taking average samples of
the gas. He, had tried the tube screwed in at the
half-way distance, one-third along, and three-quarters,
but found that about four-tenths from the centre gave
results which agreed most nearly with those obtained
on the plant. He himself had always sworn by the
creosote oil test, upheld by Mr. Crawford, until now.
He had, however, taken specially prepared nitro-
benzene, and, after 24 hours, had found nitric acid in
the bottles, showing that, although the creosote oil
best was efficient, it did not give 100 per cent, efficiency.
The reason that creosote oil, used as it came into the
works, gave poorer results than oil specially prepared,
was that it was cruder, and that the presence of various
foreign bodies reduced the absorption power. The
standardisation of the benzol test merely opened up
the question of the standardisation of all coke oven
works tests. Nothing had as yet been done by that or
any other section, and he suggested that they should
appoint a sub-committee to investigate the matter and
report. They had brains in the North of England,
and there was not the slightest need why they should
always have to rely on university men and professors
for their developments. The country was crying out
for more benzol, and it was their duty to do all possible
to increase the output.

School of Mines Diplomas.—In connection with the
diploma of associateship, the governors of the Imperial
College of Science and Technology have recently considered
the conditions to be fulfilled in the case of students of the
Royal School of Mines whose associateship courses of study
have been interrupted by their undertaking service with the
Forces of the Crown, or other approved war work. Each
case will be considered on its merits, a curtailment being
possible, but it is considered inadvisable to make any such
curtailment in the work of the first and second years.

American Coal for Allied Vessels.—A great increase in
the consumption of American coal will be called for by the
use of merchant vessels that will have to accompany the
American fleet as colliers, or be used in the transportation
of coal from the Hampton Roads ports to the various points
where the navy has storage plants. It is likely, also, that
the United States Government, says Black Diamond, will
be called upon to supply a considerable quantity of coal to
some of the European Allied countries, notably France and
Italy, which countries are in great need of coal at the
moment. Furthermore, the French and English vessels
that have been patrolling the Atlantic on that side, can
now go into American ports for bunkers, without
violating neutrality laws.