1898.] Liquid Air as an Analytic Agent. 815 



WEEKLY EVENING MEETING, 



Friday, April 1, 1898. 



Sir Edward Frankland, K.C.B. D.C.L. LL.D. F.R.S. Vice- 

 President, in the Chair. 



Professor Dewar, M.A. LL.D. F.E.S. M.BJ. 



Liquid Air as an Analytic Agent. 



The increasing importance of low-temperature research is shown 

 by the gradual development of the applications of liquid air for 

 scientific and other purposes. The much larger apparatus now used 

 in the production of the liquid enables experiments to be made on a 

 more imposing scale. 



Liquid air poured from a tin can, filled by being dipped into a 

 5-gallon jar filled with the liquid, into a large silver basin heated to red- 

 ness, remained apparently as quiescent at this high temperature as in 

 cooler vessels, and maintained a spheroidal condition, just like other 

 liquids. The temperature of the liquid air was about —190° C, or 

 83^ absolute, while the vessel in which it was placed had a temperature 

 of 800° C, or 1073° Ab. In other words, between the wall of the 

 silver vessel and the liquid air there was a difference of temperature 

 of 1000° C, 12 times the absolute temperature of the liquid. 



Liquid air can be of great service in the qualitative separation 

 of mixtures of gases. With the object of ascertaining the propor- 

 tion of any gas in air that is not condensable at about —210° C. under 

 atmospheric pressure, or is not soluble in liquid air under the same 

 conditions, a series of experiments was made with the following 

 apparatus. 



A cylindrical bulb of a capacity of 101 c.c, marked B in figure, 

 had a capillary tube sealed into it terminating in a three-way stop- 

 cock, as shown at E. The parts marked C and D consist of soda-lime 

 and sulphuric acid tubes for removing carbonic acid and water. The 

 stand marked G holds the large vacuum test-tube into which B is 

 inserted, and which contains liquid air maintained under continuous 

 exhaustion. As this low temperature had to be kept steady from 

 one to two hours, while at the same time the bulb B had to be com- 

 pletely covered with liquid air, it was necessary to arrange some means 

 of keeping up the liquid air supply without disturbing the apparatus. 

 The plan adopted is shown at H, which is a valve arrangement which 

 can be so regulated as to suck liquid air from the large vacuum 

 vessel A, and discharge it continuously along a pipe into the vacuum 

 test-tube G, the latter being kept under good exhaustion. In work- 

 ing the apparatus, the tube I is connected to a gasometer containing 

 10 cubic feet of air, so that the volume of air condensed in each 



