HYDROGEN INTO ATOMS 215 



When equilibrium exists in the gas phase, we may place (0 = and thus 

 find 



a = ^^' = ^'k,. (41) 



a2M2 or 



Combining (41) and (40) gives 



(«-3 



This is the equation which was derived in Part II and which was found 

 to be in excellent agreement with the results of the experiments. In fact, 

 we may look upon this equation as having been proved correct by the 

 experiments. 



When we compare this equation with (17), which was calculated by the 

 ist hypothesis, we see that the two become identical if we place 6 = i. 

 We see by (18) that this condition will be fulfilled if Vi is very large, com- 

 pared to ai}.ii and (o. We may thus conclude that the first hypothesis is 

 entirely in agreement with the experiments if we assume that Vi, the 

 rate of evaporation of hydrogen from the surface, is so high that only a 

 small fraction of the surface remains covered. 



If we compare Equation 42 with 33 and 34, we see that they become 

 identical if we take 6 = i while /?i and 1^2 are taken to be very large. 

 This last requirement means the velocity with which adjacent hydrogen 

 atoms or molecules interact or dissociate must be very large compared 

 to that at which they evaporate. This is equivalent to assuming that 

 equilibrium exists between the atoms and molecules on the surface. Thus 

 we see that the second hypothesis is also in accord with the experiments 

 if the rate of evaporation is assumed to be so high that only a small frac- 

 tion of the surface is covered. 



The experiments thus do not enable us to decide between the two 

 hypotheses. In fact, it is also possible by means of slightly different 

 hypotheses from those described here to obtain the same equation. 



But all these hypotheses have this in common : they lead to the con- 

 clusion that only a small fraction of the surface is covered with hydrogen 

 atoms or molecules, and that the reaction is not dependent upon phenomena 

 occurring within the surface of the metal. 



This conclusion that only a small fraction of the surface is covered 

 with hydrogen is confirmed by other facts. 



We have seen that the accommodation coefficient of hydrogen in contact 

 with tungsten at relatively low temperatures is about 19%, whereas at 



