Answers are grounded in this paper and cite their sources.
Computation of vibrational free energies of materials beyond the harmonic approximation is important for describing stability, elasticity, and phase transitions, yet the accuracy and reliability of variational and perturbative treatments is not always known. Here we compare free-energy perturbation theory (FEPT) and variational perturbation theory (VPT) for two classical anharmonic systems: the one-dimensional quartic oscillator, for which the free energy is analytically accessible, and the aluminum X-point vibrational Hamiltonian which provides a phonon subspace model for measuring applicability to condensed matter systems. Using Monte Carlo evaluation of cumulant expansions, we show that FEPT is useful only in the weakly anharmonic regime and does not improve systematically with perturbative order, whereas VPT remains accurate over a much broader range of anharmonicity. By relating the one-dimensional oscillator anharmonicity to a force-based anharmonicity measure used for real materials, we show that the regime in which VPT becomes necessary overlaps with that of materials of current interest, including perovskites.
No takes yet. Share an insight, caveat, or question.
Mathis et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: