Lattice gas with random-site energies and theory of novel amorphous metal hydride phase

A lattice gas with random-site energies is investigated as a model for hydrogen in amorphous metals. The author's recent theory for calculating the chemical potential in a system with many competing interactions is modified to include the random-site energies. Results are qualitatively different from those recently presented by Griessen using simple mean-field (MF) theory. Whereas MF theory predicts no phase separation above a critical value of the site energy width $\Delta$, the present model gives a finite critical temperature $\frac{T_c\alpha 1}{\Delta }$ for large $\Delta$. It also predicts the critical concentration to decrease proportionally to $\frac{1}{{\Delta }^2}$ and yields a closed-loop, retrograde-solubility phase diagram. Thus, analogous to binary liquids with orientation-dependent interactions, there is a maximum concentration above which no phase separation occurs. For interactions and spread in site energy expected for Pd-based amorphous hydrides, the critical temperature is predicted to be approximately 200—250 K, which may be detectable by heat-capacity or spectroscopic techniques even if it is too low for pressure-versus-composition studies.