Nested sampling for materials

Abstract We review the materials science applications of nested sampling (NS) method, which was originally conceived for calculating evidence in Bayesian inference. describe how NS can be adapted to sample potential energy surface (PES) atomistic systems, providing a straightforward approximation partition function and allowing evaluation thermodynamic variables at arbitrary temperatures. After an overview basic we number extensions, including using variable cells constant pressure sampling, semi-grand-canonical approach multicomponent parallelizing algorithm, visualizing results. cover range from past decade, exploring PES Lennard–Jones clusters that condensed phase systems. highlight examples information gained via promotes understanding properties through novel way PES, identifying thermodynamically relevant basins, entire pressure–temperature(–composition) diagram. Graphic abstract