Abstract

The prediction of high pressure phases of a chemical system is realized as a two-step process: identification of structure candidates through the global exploration of their energy landscapes over a range of different pressures, followed by a limited local optimization using ab initio methods. The application of this recipe is presented for several systems, Li₃N, Na₃N, Li_xNa_(6 − x)N₂ (x = 1,…,5), Li₂S and Na₂S. We find that at standard pressure, the optimal configurations for the binary end compositions of the alkali nitrides, Li₃N and Na₃N, exhibit the Li₃N and Li₃P structures, respectively. Among the ternary compounds, the compositions with x = 2 and x = 4 are preferred, the optimal structures being ternary variations of the Li₃N structure type. At moderately high pressures, phase transitions from Li₃N to Li₃Bi related structures are predicted for all ternary compounds, while Na₃N and Li₃N exhibit transitions from Li₃P to Li₃Bi and Li₃N via Li₃P to Li₃Bi, respectively. The analogous study of the landscapes of Li₂S and Na₂S shows that a sequence of phase transitions from the CaF₂ structure via the PbCl₂ structure to the Ni₂In structure is expected.