Abstract
Ebihara et al. had reproduced hydrogen desorption curves by McNabb-Foster diffusion equation with two types of hydrogen traps together with interstitial hydrogen. These curves were subjected to peak separation through Gaussian distribution for two contrasting cases; (a) sufficient amount of hydrogen with 150°C (423K) main peak and280°C (553K) sub-peak simulating soaking in NH4SCN solution, and (b) insufficient amount of hydrogen with 280°C (553K) main peak and 150°C (423K) sub-peak simulating exposure to CCT cycles. The amounts of trapped hydrogen by Gaussian separation were well consistent to those by McNabb-Foster equation in those two cases. Further, Gaussian distribution could separate interstitial hydrogen separated as a unique peak. This peak corresponding to interstitial hydrogen can be understood by random walk model where hydrogen atoms jump between interstitial sites to trap sites randomly. Distribution of hydrogen trapped by 150°C (423K) peak in the case (b) was slightly different from Gaussian distribution through the effect of 280°C (553K) trap, but it would not matter, considering the possible error in TDS measurement.
