July 1, 1898.] 



SCIENCE. 



neering skill, manipulative ability and 

 loyal perseverance the present successful 

 issue might have been indefinitely delayed. 

 My thanks are also due to Mi*. J. W. Heath 

 for valuable assistance in the conduct of 

 these experiments. 



THE BOILING POINT AND DENSITY OF LIQUID 

 HYDROGEN.* 



The boiling point of liquid hydrogen at 

 atmospheric pressure has been determined 

 by a platinum resistance thermometer. 

 This was constructed of pure metal and 

 had a resistance of 5.3 ohms at 0° C, 

 which fell to about 0.1 ohm when the ther- 

 mometer was immersed in liquid hydrogen. 

 On reduction of this resistance to normal air 

 temperatures the boiling point is found to 

 be— 238.2° and— 238.9° respectively by two 

 methods, and to be — 237° by a Dickson 

 formula calculated for this thermometer (c/. 

 Phil. Mag., 1898, 45, 525). The boiling point 

 of the liquid is, therefore, about — 238° C, or 

 35° absolute, and is thus about 5° higher 

 than that obtained by Olszewski by the 

 adiabatic expansion of the compressed gas, 

 and about 8° higher than that deduced by 

 "Wroblewski from van der Waals' equation. 

 It may be inferred that the critical point of 

 hydrogen is about 50° absolute, and that the 

 critical pressure will probably not exceed 15 

 atmospheres. As molecular latent heats are 

 proportional to absolute boiling points, the 

 latent heat of liquid hydrogen will be about 

 two-fifths that of liquid oxygen. From 

 analogy it is probable that the practicable 

 lowering of temperature to be obtained by 

 evaporating liquid hydrogen under pres- 

 sures of a few mm. cannot amount to more 

 than 10-12° C, and it may be said with 

 certainty that no means are at present 

 known for approaching nearer than 20-25° 

 to the absolute zero of temperature. The 

 platinum resistance thermometer used had 



*Eead before the Chemical Society , London, on 

 June 2, 1898. 



a zero point of — 263.2 platinum degrees, 

 and when immersed in boiling liquid hy- 

 drogen indicated a temperature of — 256.8° 

 on the same scale, or 6.4 platinum degrees 

 from the point at which the metal would 

 become a perfect conductor. The effect of 

 cooling platinum from the boiling point of 

 liquid oxygen to that of liquid hydrogen is 

 to diminish its resistance to one-eleventh. 



The approximate density of liquid hydro- 

 gen at its boiling point wajS determined by 

 measuring the volume of the gas obtained 

 by evaporating 10 cc. and is slightly less 

 than 0.07, or about one-sixth that of liquid 

 marsh gas, which has a density of 0.41 and 

 is the lightest liquid at its boiling point 

 hitherto known. It is remarkable that, with 

 so low a density, liquid hydrogen is so 

 easily seen, has so well defined a meniscus, 

 and can be so readily collected and manip- 

 ulated in vacuum vessels. As hydrogen 

 occluded in palladium has a density of 0.62, 

 it follows that it must be associated with 

 the metal in some other state than that of 

 liquefaction. The atomic volume of liquid 

 hydrogen at the boiling point is about 14.3, 

 atomic volumes of liquid oxygen and nitro- 

 gen being 13.7 and 16.6, respectively, at- 

 their boiling points. The density of the 

 gas at the boiling point of liquid hydrogen 

 is 0.55, or about one-half that of air, and is 

 eight times that of the gas at ordinary tem- 

 peratures. The ratio of the density of hy- 

 drogen gas at the boiling point to that of 

 the liquid is approximately 1 : 100, as com- 

 pared with a ratio of 1 : 255 in the case of 

 oxygen. 



The specific heat of hydrogen in the 

 gaseous state and in hydrogenized palla- 

 dium is 3.4, but may very probably be 6.4 

 in the liquid substance. Such a liquid 

 would be unique in its properties, but as 

 the volume of one gram of liquid hydrogen 

 is about 14-15 cc. the specific heat per 

 unit volume must be nearly 0.5, which is 

 about that of liquid air. It is highly pro- 



