Clapeyron Neural Networks for Single-Species Vapor-Liquid Equilibria

TL;DR

This paper introduces Clapeyron Neural Networks that incorporate thermodynamic principles into graph neural networks to improve predictions of vapor-liquid equilibrium properties, especially in data-scarce situations.

Contribution

The paper develops thermodynamics-informed GNNs based on the Clapeyron equation, enhancing prediction accuracy and thermodynamic consistency over traditional data-driven models.

Findings

Improved accuracy in predicting vapor pressure, molar volumes, and enthalpy of vaporization.

Enhanced thermodynamic consistency compared to purely data-driven models.

Significant benefits in data-scarce scenarios for chemical engineering applications.

Abstract

Machine learning (ML) approaches have shown promising results for predicting molecular properties relevant for chemical process design. However, they are often limited by scarce experimental property data and lack thermodynamic consistency. As such, thermodynamics-informed ML, i.e., incorporating thermodynamic relations into the loss function as regularization term for training, has been proposed. We herein transfer the concept of thermodynamics-informed graph neural networks (GNNs) from the Gibbs-Duhem to the Clapeyron equation, predicting several pure component properties in a multi-task manner, namely: vapor pressure, liquid molar volume, vapor molar volume and enthalpy of vaporization. We find improved prediction accuracy of the Clapeyron-GNN compared to the single-task learning setting, and improved approximation of the Clapeyron equation compared to the purely data-driven multi-task learning setting. In fact, we observe the largest improvement in prediction accuracy for the properties with the lowest availability of data, making our model promising for practical application in data scarce scenarios of chemical engineering practice.