Thermal Corrections in Crystal Structure Prediction
In a systematic assessment of 14 molecular crystals using B86bPBE-XDM density functional theory, the experimental crystal structure of 4-hydroxythiophene-2-carbonitrile was correctly identified as the lowest-energy polymorph only when a quasi-harmonic estimate of vibrational free-energy contributions was included. In contrast, for the other 11 rigid, planar compounds and 3 co-crystals in the study, the experimental structure was correctly predicted based on lattice energy alone, without the need for thermal corrections [1].
| Evidence Dimension | Requirement of vibrational free-energy (thermal) corrections for correct crystal energy ranking in DFT calculations |
|---|---|
| Target Compound Data | Requires thermal correction for correct experimental structure identification |
| Comparator Or Baseline | 12 of 14 compounds in the same study did NOT require thermal correction; experimental structure correctly predicted from lattice energy alone |
| Quantified Difference | Target compound is an exception (1 of 14); thermal correction is mandatory for accurate energy ranking |
| Conditions | B86bPBE-XDM density functional combined with exchange-hole dipole moment (XDM) dispersion model; 14 compounds from CSP blind tests |
Why This Matters
For computational chemists performing crystal structure prediction, selecting this compound necessitates a more computationally intensive protocol (quasi-harmonic free-energy calculations), whereas its analogs may be reliably modeled with simpler lattice-energy approaches, directly impacting resource allocation and workflow design.
- [1] Journal of Chemical Theory and Computation. 2018, 14 (2), 572-583. DOI: 10.1021/acs.jctc.6b00679. View Source
