208 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1909. 
The essentially adiabatic expansion of air can be effected in two 
different ways. 
(a) Air compressed to the pressure p, may be expanded without 
doing available exterior work. It passes from the compression tank 
to the liquefaction tank by the way of a narrow orifice. This is the 
manner of expansion adopted by Linde in his liquid-air machines. 
The lowering of temperature obtained under such conditions is only 
appreciable if the difference between the pressures p, and p, is con- 
siderable. In Linde’s apparatus” gaseous air cooled to about —100° 
©. is expanded from a pressure of 200 to 40 atmospheres; the lique- 
fied part of the gas at about —140° C. passes into a regenerator where 
it cools the air compressed at 200 atmospheres; it is then led into a 
pump which brings it up to this latter pressure. A second auxiliary 
pump draws air from the atmosphere to take the place of that part 
which has been liquefied. In the industrial machines the gases com- 
. pressed to 200 atmospheres, before passing into the economizer where 
the gas at —140° circulates, are cooled by liquid ammonia. 
Under such conditions, in machines which produce 50 liters per hour 
the yield of liquid air is about half a liter per horsepower-hour. 
(6) The second method of air expansion consists in utilizing the 
exterior work which the gas is capable of doing when it passes from 
pressure p, to p,. This mode of expansion with utilizable exterior 
work is the basis of the processes of G. Claude for the production 
of liquid air.” Air compressed to a maximum pressure of 30 or 40 
atmospheres passes first to an economizer, where it is cooled down as 
in Linde’s apparatus by unliquefied gas. It is then expanded in the 
cylinder of a motor whose energy can be utilized in the original 
compression of the air. In course of time a partial liquefaction of 
the air occurs in the evlinder of the auxiliary motor. The lubrica- 
tion of this cylinder is accomplished by means of a petroleum dis- 
tillate having a specific gravity of 0.675 (automobile gasoline), 
which, at the low temperature at which the motor operates, attains 
a sirupy consistency comparable to industrial lubricants. 
Applied in this form the process of G. Claude gives only unsatis- 
factory results. The expansion of the air, occurring at temperatures 
of —175° to —180° by the gas expanding in the auxiliary motor, 
takes place under unfavorable conditions. The appearance of liquid 
air in this auxiliary cylinder is likely to produce a peculiar water- 
hammer effect and is accompanied by a large increase in friction, that 
@¥Wor a description of Linde’s machine, see E. Mathias: “La préparation 
industrielle et les principales applications des gaz liquéfiés.” Revue générale 
des Sciences, vol, 12, 1901. 
®’ Readers desirous of obtaining the details of Claude’s processes will find an 
account of them in the following: G. Claude: Air liquide, Oxygéne, Azote. 
Paris, H. Dunod and E. Pinat, 1909. 
