[ 188 ] 



XVIII. Note on the Heat of Formation of Hydrogen from 

 Hydrogen Atoms. By Irving Laxgmuib, Ph.D., General 

 Electric Co., Schenectady, jV. Y* 



DR. X. BOHE, in his recent valuable and wonderfully 

 suggestive paper on the Constitution of Atoms and 

 Molecules (Phil. Mag. xxvi. p. 863, 1913), calculates the 

 heat of: formation of hydrogen molecules from the atoms to 

 be 60,000 calories per gram-molecule. He points out that 

 this value is " considerably less than the value of 130,000 

 found by Langmuir (Journ. Amer. Chem. Soc. xxxiv. p. 860, 

 1912) by measuring the heat-conduction through the gas 

 from an incandescent wire in hydrogen." 



In order that the results obtained by me may not bear 

 false evidence against the beautiful theory derived by 

 Dr. Bohiv, I take this opportunity of giving a preliminary 

 account of some more recent measurements carried out by 

 Mr. Gr. M. J. Mackay and myself, on the dissociation of 

 hydrogen. 



The method adopted was similar to that previously used, 

 namely, to determine the heat-loss from electrically heated 

 tungsten wires in hydrogen. In these experiments, how- 

 ever, very much greater care w T as used in determining the 

 temperatures of the wires and in obtaining extremely pure- 

 hydrogen. Pressures of hydrogen ranging from 1 mm. up 

 to 760 mm. of mercury were used. Still further experiments 

 which, however, have not yet been completed, were made at 

 pressures from 0*01 mm. up to 20 mm. pressure. 



The measurements at low pressures gave very interesting 

 results, namely, that at about one-tenth of an atmosphere 

 pressure of hydrogen, the total heat-loss from filaments heated 

 to very high temperatures (3000° K.) is several times as great 

 as in hydrogen at atmospheric pressure. 



By subtracting the heat-loss due to normal heat-conduction 

 from the total observed heat-loss, the heat carried by diffusion 

 of hydrogen atoms is obtained. By plotting the logarithm 

 of this quantity against the reciprocal of the absolute tempe- 

 rature, straight lines are obtained in each experiment. The 

 slopes of all these lines are practically equal, no matter what 

 pressure of the hydrogen was used (above 10 mm.). 



On the assumption that the diffusion coefficient varies with 

 the 3/2 power of the temperature, this leads to the result that 

 the heat of formation of hydrogen is about 76,000 calories 

 per gram-molecule, as against 130,000 previously found. The 

 reason for the high value of the latter figure is, that it was 

 * Communicated bv the Author. 



