Accurate Machine Learning for Predicting the Viscosities of Deep Eutectic Solvents

Abstract

Deep eutectic solvents (DESs) are emerging as environmentally friendly designer solvents for mass transport and heat transfer processes in industrial applications; however, the lack of accurate tools to predict and thus control their viscosities under both a range of environmental factors and formulations hinders their general application. While DESs may serve as designer solvents, with nearly unlimited combinations, this unfortunately makes it experimentally infeasible to comprehensively measure the viscosities of all DESs of potential industrial interest. To assist in the design of DESs, we have developed several new machine learning (ML) models that accurately and rapidly predict the viscosities of a diverse group of DESs at different temperatures and molar ratios using, to date, one of the most comprehensive data sets containing the properties of over 670 DESs over a wide range of temperatures (278.15−385.25 K). Three ML models, including support vector regression (SVR), feed forward neural networks (FFNNs), and categorical boosting (CatBoost), were developed to predict DES viscosity as a function of temperature and molar ratio and contrasted with multilinear and two-factor polynomial regression baselines. Quantum chemistry-based, COSMO-RS-derived sigma profile (σ-profile) features were used as inputs for the ML models. The CatBoost model is excellent at externally predicting DES viscosity, as indicated by high R 2 (0.99) and low root-mean-square-error (RMSE) and average absolute relative deviations (AARD) (5.22%) values for the testing data sets, and 98% of the data points lie within the 15% of AARD deviations. Furthermore, SHapley additive explanation (SHAP) analysis was employed to interpret the ML results and rationalize the viscosity predictions. The result is an ML approach that accurately predicts viscosity and will aid in accelerating the design of appropriate DESs for industrial applications.

Physics-Based Machine Learning Models Predict Carbon Dioxide Solubility in Chemically Reactive Deep Eutectic Solvents

“…Further, to incorporate the chemical reactivity, we calculated the dissociation constants (p K a ) of HBAs and HBDs. The commercial package ChemAxon was utilized for the calculation of p K a values of HBA and HBD. , We have also calculated the viscosity of DESs using our in-house ML models and used this as an input feature to study the effect of viscosity in predicting solubility. First, we calculated the solubility of CO 2 in chemically reactive (DESs) using the COSMO-RS and multilinear regression (MLR) models.…”

High-Throughput Screening and Accurate Prediction of Ionic Liquid Viscosities Using Interpretable Machine Learning

“…All the ML models were developed and trained in Python 3.9.13 with the scikit-learn package. , For all the ML models, RandomizedSearchCV in scikit-learn with 10-fold cross-validation was used for optimization of the hyperparameters. , The optimized hyperparameters for SVR, FFNN, and CATBoost are reported in Table S1. In addition, the computational costs associated with each model and their practical implications for industrial applications are also discussed in Table S2.…”

Machine learning models accelerate deep eutectic solvent discovery for the recycling of lithium-ion battery cathodes

“…It is a powerful tool to accelerate molecule design and functional materials discovery and processing for applications such as catalysts, 20 pharmaceutical synthesis, 21 and pretreatment of Li-ion batteries. 22,23 ML approaches also have great performance in DESs, such as property prediction 24 and gas absorption, 25 which could unlock new opportunities for the recovery of spent LIBs through DESs. Great attention has been paid to leaching cathodes through DESs; 12 currently, the mainstream view assumes that low viscosity, high acidity, strong coordination, and reducibility of DESs might be beneficial for efficient leaching, 26 while little work has validated these hypotheses, and we are still lacking quantification of the importance of each property.…”