ATTAINMENT OF VERY LOW TEMPERATURES. 1 5 



Quantity of air liquefied = 9.5 to 10 percent. 



This result is of considerable value in considering the phenomena 

 connected with the liquefaction of hydrogen. 



VI. The Liquefaction of Hydrogen. Preliminary 



Experiments. 



It has already been pointed out that the original experiments of 



Joule and Lord Kelvin showed that hydrogen, when allowed to expand 



freely, behaved differently from other gases, becoming heated instead 



of cooled. The magnitude of the effect depended upon the absolute 



temperature at which the experiment was performed, decreasing with 



fall of temperature. From their experiments it might be predicted that 



the effect would change sign at very low temperatures. Employing the 



formula, 



(I- 

 /f = >P — (i 



to represent the experimental results, Witkowski found 64.1 and 0.331 

 for the values of the constants a and /9 respectively, and — 79°C. for 

 the inversion point of the effect. Witkowski's prediction has recently 

 been verified experimentally by Olszewski (p. 10). 



In 1898 Dewar* succeeded in obtaining a jet of hydrogen suffi- 

 ciently cold to solidify air, and later constructed an apparatus with 

 which he obtained liquid hydrogen in quantity. In his first experiments 

 he compressed the gas into steel cylinders under 180 atmospheres, and 

 then allowed it to pass through coils cooled in solid carbonic acid, and 

 in liquid air boiling under reduced pressure, to a regenerator coil con- 

 tained within a vacuum vessel. The gas escaped through a pin valve 

 at the bottom of the regeneratory coil, and became cooled, but before 

 any liquid could be collected the opening had become blocked with 

 some solid impurity, and the experiment was at an end. Later, Dewar 

 constructed a larger apparatus, and succeeded in obtaining liquid 

 hydrogen in quantity. He has not, however, published any descrip- 

 tion of it. 



Dewar's first experiment had confirmed the prediction that hydro- 

 gen, v/hen cooled to the temperature of liquid air, became further 

 cooled on free expansion, but it did not throw any further light on the 

 behavior of this gas. Indeed it still remained possible, though perhaps 

 improbable, that it would only be necessary to cool the gas to the 

 temperature of solid carbonic acid before allowing it to enter the 

 regenerator coil of the liquefaction apparatus. Accordingly some pre- 

 liminary experiments were set on foot in which a Hampson air lique- 

 fier, modified as occasion required, was employed. 



* Proceedings Chemical Society. 



