The Liquefaction of Gases. 87
- = ST ——— rh ———_ ——tp me ee LL mt Ap a
} answered the question by experiment. Two receivers
were taken ; one was filled with air compressed to about
T H E C H E M I C A L N E Ww S. 20 atmospheres; the other receiver was exhausted by an
air-pump. The two receivers having been conne&ed by
a short pipe in which there was a stopcock, the entire
apparatus was immersed in a canful of water. Then by
opening the stopcock the compressed air was permitted
to flow from the full to the empty receiver.
The advocates of the old theory of heat would have
said that since the compressed air in this operation dilated
to twice its original volume heat must disappear to meet
the increased capacity of air in consequence of its expansion.
 But the result, indeed, entirely contradicted this.
There was no cooling effet whatever measurable by the
most delicate thermometer, a result which certainly would
1ave occurred had the compressed air been made to displace
 the atmosphere. In fact it has since been abunlantly
 proved that the rzdudion of temperature which a
:ompressed gas undergoes when allowed to expand into
‘he atmosghere is owing to work done in displacing the
atmosphere, which offers the well-known resistance of about
t5 Ibs. per 8q. inch to the escape of the gas.
Putting the fads into a brief statement, it may be taken
1s proved that when a compressed air or gas is allowed to
expand the greater the resistance, and consequently the
work done in overcoming such resistance, the greater the
resultant cooling effe@ produced. Expand a compressed
gas into a vacuum, no cold is produced; expand it in
opposing the atmosphere, some cold is produced ; expaud
it against a resistance such as that of a heavily loaded
piston so as to contribute to its motion, and the maximum
cooling effedt is produced. Sir W. Thomson, Rankine,
and Clausius have worked out the whole subjet mathematically,
 but it may be convenient to quote Clausius.
He takes typical cases to illustrate the difference on the
one hand of the expansion of a compressed gas as against
the resistance of the atmosphere; on the other hand, as
against the resistance of a loaded piston. Assuming an
initial pressure of 5 atmospheres in each case, he shows
that to bring the temperature of the expanded gas to the
initial temperature before expansion, the first case would
require 17 units of heat as against 74'g required in the
second case.
Long before the mechanical theory of heat was enunciated
 Dr. Gorrie utilised for freezing-machines the intense
cold produced by expanding a gas behind a working
piston. Though his theoretical ideas were somewhat
contused, Charles Randolph noticed and commented upon
it in the case of the compressed air-engine erected at
Govan Colliery in 1849. Air compressed to 20 to 30 lbs.
to the square inch was sent down a shaft 176 yards deep,
and along a road about 700 yards long, where it was used
for working an old steam-engine in the place of steam,
the result being that the ports of the engine cylinder were
frequently blocked up with ice.
The same phenomena can be observed in any kind of
machinery worked with compressed air. In fact I have
frequently observed the temperature of air issuing from
ihe cylinder of coal- and rock-cutting machines to register
fully many degrees below zero. Indeed, by such a process
an unlimited reducion of temperature can be produced,
provided the air or gas be subjected to sufficient compression
 before expansion, and supposing always that care be
taken to remove the heat of compression by passing the
compressed air through a surface condenser, kept cool
with water, or by the actual injection of water at the
time of its being compressed. When I first turned attendion
 to the subje@ I exercised myself considerably in
efforts to contrive a method of expanding air or gas so as
to cause it to do the maximum of work in the a& of expansion.
 But I speedily became convinced of the un.
likelihood of anyone being able to invent a more perfe®t
instrument for converting the expansion of an elastic fluid
into work than is afforded by a well constructed steamengine.
 Mallard shows us that compressed air made to
do work in an engine, at an initial pressure of 10 atmo-———



THREE methods have been employed in the liquefaction of
gases :—I. Increase of pressure at normal temperatures.
II. Abstra&ion of heat at normal pressures. III. The
combination of these two operations. The gases which
can be liquefied by pressure at normal temperatures are
limited. Natterer showed us, some years ago, that under
such conditions the enormous pressure of 3000 atmospheres
was insufficient to liquefy hydrogen and oxygen. It was
reserved for our countryman, Dr. Andrews, to demonstate
the fallacy of relying upon pressure alone for gaseous liquefaction.
 To use the words of Prof. Wurtz in his recent Fara.
day Le&ure, he has shown us that with all vapours theres 2
point at which the molecular movements caused by heat
finally gain the victory over the force of cohesion, whatever
be the pressure to which the air is subjeéted. Dr. Andrews
describes it as the * critical point; Mendelejeff names it
the * absolute boiling-point.”
It is, therefore, easy to understand why Natterer failed
and Cailletet and Pictet succeeded with their experiments.
Indeed, it would appear from these researches, and the
principles underlying them, and if it be taken for granted
that the molecules of all gases have a tendency to gravitate
 to each other independent of external pressure, that
of the two agents for effecting the liquefaction of gases
the abstra®ion of heat is more efficacious than the appli.
cation of pressure.
This is evident, and follows from the ordinary laws of
gaseous expansion. It is a well-established fact that
gases expand uniformly, and that the rate of expansion is
such that a given volume of gas at the freezing-point of
water will become double its bulk at 491° F. Therefore,
if it be followed in its path of contra&ion for temperatures
below freezing-point, by the time its temperature becomes
reduced 491°, which is the absolute zero of physicists, the
gas must necessarily be either liquefied or solidified.
The greatest artificial cold yet produced has not exceeded
 about 220° below zero F.; but it would be rash tc
assume that it is impossible to produce greater cold.
Indeed, it is not improbable that means may yet be found
to produce cold sufficient to liquefy either oxygen or hydrogen
 at ordinary pressures. The question at once
arises, what has already been done in the produéion of
extremely low temperatures, and in what direction can
such efforts be extended, But practice is dependent upon
principles, and for principles we must fall back upon the
mechanical theory of heat, thoroughly established and
accepted universally by physicists, but as yet imperfectly
understood by numbers of chemists educated in the old
school of Black and Leslie.
It is useful in the outset to recal certain experiments of
Dr. Joule; and, parenthetically, it may be observed that it
is to be regretted that no colle&ted edition of the papers of
this able man has yet been published. In his paper read
to the Royal Society in June, 1844, some very interesting
experiments with compressed air are detailed. The paper
commences by reference to the well-known fa& that compressed
 air when expanded becomes cooled, that is, when
it escapes into the atmosphere ; ‘but Dr. Joule asked himself
 this question—* Would the cooling effet be greater
or less were the same air exnanded into a vacuum ?* He

* A paper read to the Chemical Section of the Philosophical Society
of Glasgow and to the Institute of Engineers (Scotland). Full
drawings of the machinery will be found in the T7rahsactions of the
latter body, published in Glasgow.