Explicitly stable Fundamental Measure Theory models for classical density functional theory

TL;DR

This paper revisits tensorial Fundamental Measure Theory models in classical DFT, demonstrating that explicit stability can be achieved at the expense of low-density accuracy, with implications for dense system studies.

Contribution

It introduces explicitly stable FMT models that balance stability and accuracy, providing a new approach within classical DFT for hard spheres.

Findings

Stable functional is competitive with White Bear model

Explicit stability achieved at low-density accuracy cost

Functional performs well for dense systems

Abstract

The derivation of the state of the art tensorial versions of Fundamental Measure Theory (a form of classical Density Functional Theory for hard spheres) are re-examined in the light of the recently introduced concept of global stability of the density functional based on its boundedness (Lutsko and Lam, Phys. Rev. E 98, 012604 (2018)). It is shown that within the present paradigm, explicitly stability of the functional can be achieved only at the cost of giving up accuracy at low densities. It is argued that this is an acceptable trade-off since the main value of DFT lies in the study of dense systems. Explicit calculations for a wide variety of systems shows that a proposed explicitly stable functional is competitive in all ways with the popular White Bear model while sharing some of its weaknesses when applied to non-close-packed solids.