A longstanding challenge in thermodynamics has been the development of a unified analytical expression for the free energy of matter capable of describing all thermodynamic properties. Since the 19th century, significant strides have been made in modeling fluid phases using continuous equations of state (EoSs) while the crystalline state has remained largely unexplored because of its complexity.
Erich A. Muller,
Imperial College London
This work introduces an approach that employs artificial neural networks to construct an EoS directly from comprehensive molecular simulation data. The efficacy of this method is demonstrated through application to the Mie potential, resulting in a thermodynamically consistent model seamlessly bridging fluid and crystalline phases. The proposed EoS accurately predicts metastable regions, enabling a comprehensive characterization of the phase diagram, which includes the critical and triple points. Further discussion revolves around the use of machine learning to predict transport properties (diffusion coefficients, viscosities and thermal conductivities) on how simple force fields are incapable of predicting accurately both volumetric and transport properties.
Erich A. Muller has over 30 years of accumulated experience in the molecular description of complex fluids and interfaces with particular application to bridging gaps between detailed molecular studies and industrial applications. He is a professor in thermodynamics at the department of Chemical Engineering at Imperial College London, a Fellow of the Royal Society of Chemistry, a Fellow of the Institute of Chemical Engineers and an Adjunct Professor at North Carolina State University. His research interests are in the molecular simulation of complex fluid thermodynamics (liquid crystals, asphaltenes, polymers), adsorption on nanoporous materials (gases on activated carbons and nanotubes), interfacial phenomena (vapour-liquid equilibria and surface tensions) and machine learning of thermophysical properties.