Tracking down the structural and enthalpic changes in molecular materials upon melting.

Authors

Jan Ludík, Ctirad Červinka

Published in

Physical chemistry chemical physics : PCCP. Sep 07, 2026. Epub Sep 07, 2026.

Abstract

A novel X10 benchmark set is introduced to enable an assessment of the computational performance of atomistic simulations of enthalpies of fusion for molecular materials. The fusion enthalpy, i.e. the enthalpy of melting, represents a fundamental descriptor of the solid-liquid equilibrium, and as such, the ability to predict this property is highly valuable across various fields of material design. Computational accuracy is tested for all-atom molecular-dynamics simulations, relying either on a classical generic non-polarizable force-field model (MD), or on a selected density-functional-theory (DFT) treatment of the electronic degrees of freedom in ab initio molecular dynamics (AIMD). Performance of these MD and AIMD methods is critically assessed against reference experimental data and it is evaluated in terms of the fusion enthalpies and related structural descriptors, such as bulk phase densities. A detailed interpretation of the predicted fusion enthalpies in terms of important cohesive non-covalent interactions, such as hydrogen bonding and dispersion forces, and variations of their intensity upon melting is presented. A special emphasis is laid on investigations of the finite-size artifacts that are necessarily at play in costly AIMD simulations that are feasible only for small molecular ensembles. The presented results indicate that the fusion enthalpies for neutral-molecular materials can be typically predicted within roughly 2 kJ mol^-1 from the experiment with the PBE-D3(BJ)/GTH based AIMD model, yielding a somewhat higher accuracy on average than the classical force-field based simulations. This finding is important as similar DFT theories are commonly used nowadays as a source of training data for development of machine-learning interatomic potentials.

PMID:
42704294
Bibliographic data and abstract were imported from PubMed on 07 Sep 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.