Properties of non-FCC hard-sphere solids predicted by density functional theory

TL;DR

This study uses density functional theory models to analyze the properties of various non-FCC hard-sphere crystal phases, revealing phase stability and structural behaviors near close packing.

Contribution

It compares multiple density functional models to predict properties of non-FCC hard-sphere solids and assesses their physical plausibility and phase stability.

Findings

FCC phase shows vanishing Lindemann parameter near close packing

Multiple solid structures identified for BCC above certain density

Simple cubic phase stability varies across models

Abstract

The free energies of the FCC, BCC, HCP and Simple Cubic phases for hard spheres are calculated as a function of density using the Fundamental Measure Theory models of Rosenfeld et al (PRE 55, 4245 (1997)), Tarazona (PRL 84, 694 (2001)) and Roth et al (J. Phys.: Cond. Matt. 14, 12063 (2002)) in the Gaussian approximation. For the FCC phase, the present work confirms the vanishing of the Lindemann parameter (i.e. vanishing of the width of the Gaussians) near close packing for all three models and the results for the HCP phase are nearly identical. For the BCC phase and for packing fractions above $\eta \sim 0.56$, all three theories show multiple solid structures differing in the widths of the Gaussians. In all three cases, one of these structures shows the expected vanishing of the Lindemann parameter at close packing, but this physical structure is only thermodynamically favored over the unphysical structures in the Tarazona theory and even then, some unphysical behavior persists at lower densities. The simple cubic phase is stabilized in the model of Rosenfeld et al. for a range of densities and in the Tarazona model only very near close-packing.