MLPROP is an open-access, web-based interface for predicting thermophysical properties of pure substances and mixtures by machine learning models. Integrating GRAPPA, UNIFAC 2.0, mod. UNIFAC 2.0, and HANNA, it enables accurate and accessible predictions of vapor pressures, activity coefficients, and vapor-liquid equilibria without requiring users to download or install any software.

Abstract

Machine learning (ML) enables the development of powerful methods for predicting thermophysical properties, but technical barriers like cumbersome implementation in established workflows hinder their application in practice. With MLPROP, we provide a web interface to predict thermophysical properties with advanced ML methods. MLPROP includes models for predicting the vapor pressure of pure components (GRAPPA), activity coefficients and vapor-liquid equilibria in binary mixtures (UNIFAC 2.0, mod. UNIFAC 2.0, and HANNA), and a routine to fit NRTL parameters to the model predictions. MLPROP will be continuously updated and extended and is accessible via https://ml-prop.mv.rptu.de/. The source code of all models is available as open source, which allows integration into existing workflows.