Answers are grounded in this paper and cite their sources.
The thermodynamic properties of molecular crystals are commonly calculated at the harmonic level. Despite being the simplest approach to incorporating vibrational effects, their calculation is expensive and fraught with difficulties. The presence of a divergence in the entropy integrand at zero frequency in combination with artifacts from the usually employed Fourier interpolation and noise in the DFT forces make finite-displacement calculations difficult to converge to an acceptable precision. There are two objectives to this work. First, we examine the sources of error in the calculation of harmonic thermodynamic properties of molecular crystals. Second, we provide a reliable protocol for their calculation with DFT, with a target precision of 0.1 kJ/mol per molecule, in the most efficient manner possible. The four sources of error (quadrature error, force constant aliasing, finite difference error, and noise in the DFT forces) are explore using a surrogate force field model. The proposed protocol involves using non-diagonal supercells with maximal reach (half the shortest sublattice vector), a dipole-dipole correction, and atomic displacements adapted to the expected noise in the forces. The protocol yields free energies at 300 K within tenths of a kJ/mol per molecule at reasonable supercell sizes.
No takes yet. Share an insight, caveat, or question.
Blancas et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: