204 PHENOMENA, ATOMS, AND MOLECULES 



In 1912 the value given was 130000 calories at constant volume. 

 Diffusion Coefficient of Atomic Hydrogen in Molecular Hydrogen. — 

 This quantity is found to be 



D = 2.14 10-3 T''^^ (70) 



This result is 4.2 times greater than that found previously by calculation 

 from the principles of the kinetic theory. This difference suggests that the 

 hydrogen molecules are more or less "transparent" to hydrogen atoms. 

 The effect is probably quite analogous to that of the abnormal mobility of 

 the H and OH ions in aqueous solutions. 



The Coefficients Oi and a2. — The coefficient ai is found to be constant 

 and equal to unity, while ao is apparently constant and equal to 0.68. In 

 other words, all the hydrogen atoms striking the filament are absorbed by it 

 and 68% of all the hydrogen molecules are absorbed. The velocity of the 

 reaction is thus determined practically entirely by the equilibrium con- 

 ditions within the wire. The fact that these coefficients are found to be 

 so nearly equal to vmity is excellent evidence that in the present theory we 

 are actually dealing with the fundamental factors determining the velocity 

 of the reaction. 



11. Other possible theories of the mechanism of the reaction are dis- 

 cussed, but no other is found which agrees at all well with the facts. 



12. The apparent increase in specific heat caused by the dissociation 

 is calculated. Even at temperatures as low as 2000° the effect should be 

 perceptible. The magnitude of the effects found is, according to statements 

 of Seigel, of the right order to account fully for certain anomalies in the 

 experiments of Bjerrum. Therefore, the results of the explosion method, 

 instead of conflicting with the present determinations of the degree of dis- 

 sociation, may be regarded as additional evidence in their favor. 



13. The rate at which hydrogen at very low pressures is dissociated by 

 a tungsten wire at 1200-1500° has been calculated according to the theory 

 and is found to be 8-10 times greater than the greatest observed rate at 

 which active hydrogen was deposited on glass surfaces. In view of the 

 marked fatigue effects characteristic of this adsorption of active hydrogen 

 by glass, the agreement is close enough to lend further support to the 

 theory. The quantitative evidence of the dissociation of hydrogen may 

 therefore be said to extend over a temperature range from 1200 to 

 3500° K., in which the degree of dissociation increases in the ratio 

 1 : I 70000. 



14. Experiments on the heat losses from tungsten wires in mixtures 

 of nitrogen and hydrogen yield results also in accord with the theory. See 



