Abstract
The first-principles calculation is attempted for Au–Pd phase diagram in the vicinity of equi-atomic composition. The obtained transition temperature is about 350 K, for which experimental data scatter significantly. The transition seems to be of a second order below 40 at.% of Pd, which is most probably caused by a fairly big contribution of the multi-body interaction energy. Also a serious doubt is cast on the existing phase diagrams near the 1:3 and 3:1 composition.