ABSTRACT
We assess the concept of alchemical transformations for predicting how a further and not-tested change in composition would change materials properties. This might help to guide ab initio calculations through multidimensional property-composition spaces. Equilibrium volumes, bulk moduli, and relative lattice stability of fcc and bcc 4d transition metals Zr, Nb, Mo, Tc, Ru, Rh, Pd, and Ag are calculated using density functional theory. Alchemical derivatives predict qualitative trends in lattice stability while equilibrium volumes and bulk moduli are predicted with less than 9% and 28% deviation, respectively. Predicted changes in equilibrium volume and bulk moduli for binary and ternary mixtures of Rh-Pd-Ag are in qualitative agreement even for predicted bulk modulus changes as large as +100% or -50%. Based on these results, it is suggested that alchemical transformations could be meaningful for enhanced sampling in the context of virtual high-throughput materials screening projects.
