It was noted that recent measurements of the diffusion of H in Pd and Pd-Fe solutions, in the vicinity of the 50K anomaly, did not agree with data which had been obtained for H migration in Pd-Fe alloys by using the magnetic after-effect technique. The discrepancy between these 2 sets of data was considered here, and it was concluded that mass diffusion data for temperatures above 100K supported the contention that, at temperatures ranging from 40 to 55K, the activation energy increased with the Fe content. However, as the temperature fell below 100K, the activation energy began to decrease rapidly with decreasing temperature, and its actual value for a given alloy was expected to be sensitive to the temperature range. The tunnelling mechanism which was responsible for this phenomenon was well-established, but it was noted that such effects could be observed in classical jump theory when the barrier energy was assumed to be a stochastic variable.

R.B.McLellan, L.Yang: Scripta Metallurgica et Materialia, 1994, 30[2], 155-7