Computer-assisted approach to develop a new prediction method of liquid viscosity of organic compounds

Temperature-Dependent Density and Viscosity Prediction for Hydrocarbons: Machine Learning and Molecular Dynamics Simulations

Machine learning-based predictive models allow rapid and reliable prediction of material properties and facilitate innovative materials design. Base oils used in the formulation of lubricant products are complex hydrocarbons of varying sizes and structure. This study developed Gaussian process regression-based models to accurately predict the temperature-dependent density and dynamic viscosity of 305 complex hydrocarbons. In our approach, strongly correlated/collinear predictors were trimmed, important predictors were selected by least absolute shrinkage and selection operator (LASSO) regularization and prior domain knowledge, hyperparameters were systematically optimized by Bayesian optimization, and the models were interpreted. The approach provided versatile and quantitative structure-property relationship (QSPR) models with relatively simple predictors for determining the dynamic viscosity and density of complex hydrocarbons at any temperature. In addition, we developed molecular dynamics simulation-based descriptors and evaluated the feasibility and versatility of dynamic descriptors from simulations for predicting the material properties. It was found that the models developed using a comparably smaller pool of dynamic descriptors performed similarly in predicting density and viscosity to models based on many more static descriptors. The best models were shown to predict density and dynamic viscosity with coefficient of determination (R2) values of 99.6% and 97.7%, respectively, for all data sets, including a test data set of 45 molecules. Finally, partial dependency plots (PDPs), individual conditional expectation (ICE) plots, local interpretable model-agnostic explanation (LIME) values, and trimmed model R2 values were used to identify the most important static and dynamic predictors of the density and viscosity.

Shear viscosity prediction of alcohols, hydrocarbons, halogenated, carbonyl, nitrogen-containing, and sulfur compounds using the variable force fields

Viscosity of organic liquids is an important physical property in applications of printing, pharmaceuticals, oil extracting, engineering, and chemical processes. Experimental measurement is a direct but time-consuming process. Accurately predicting the viscosity with a broad range of chemical diversity is still a great challenge. In this work, a protocol named Variable Force Field (VaFF) was implemented to efficiently vary the force field parameters, especially λ_vdW, for the van der Waals term for the shear viscosity prediction of 75 organic liquid molecules with viscosity ranging from -9 to 0 in their nature logarithm and containing diverse chemical functional groups, such as alcoholic hydroxyl, carbonyl, and halogenated groups. Feature learning was applied for the viscosity prediction, and the selected features indicated that the hydrogen bonding interactions and the number of atoms and rings play important roles in the property of viscosity. The shear viscosity prediction of alcohols is very difficult owing to the existence of relative strong intermolecular hydrogen bonding interaction as reflected by density functional theory binding energies. From radial and spatial distribution functions of methanol, we found that the van der Waals related parameters λ_vdW are more crucial to the viscosity prediction than the rotation related parameters, λ_tor. With the variable λ_vdW-based all-atom optimized potentials for liquid simulations force field, a great improvement was observed in the viscosity prediction for alcohols. The simplicity and uniformity of VaFF make it an efficient tool for the prediction of viscosity and other related properties in the rational design of materials with the specific properties.

A New Machine-Learning Tool for Fast Estimation of Liquid Viscosity. Application to Cosmetic Oils

The viscosities of pure liquids are estimated at 25 °C, from their molecular structures, using three modeling approaches: group contributions, COSMO-RS σ-moment-based neural networks, and graph machines. The last two are machine-learning methods, whereby models are designed and trained from a database of viscosities of 300 molecules at 25 °C. Group contributions and graph machines make use of the 2D-structures only (the SMILES codes of the molecules), while neural networks estimations are based on a set of five descriptors: COSMO-RS σ-moments. For the first time, leave-one-out is used for graph machine selection, and it is shown that it can be replaced with the much faster virtual leave-one-out algorithm. The database covers a wide diversity of chemical structures, namely, alkanes, ethers, esters, ketones, carbonates, acids, alcohols, silanes, and siloxanes, as well as different chemical backbone, i.e., straight, branched, or cyclic chains. A comparison of the viscosities of liquids of an independent set of 22 cosmetic oils shows that the graph machine approach provides the most accurate results given the available data. The results obtained by the neural network based on sigma-moments and by the graph machines can be duplicated easily by using a demonstration tool based on the Docker technology, available for download as explained in the Supporting Information. This demonstration also allows the reader to predict, at 25 °C, the viscosity of any liquid of moderate molecular size (M < 600 Da) that contains C, H, O, or Si atoms, starting either from its SMILES code or from its σ-moments computed with the COSMOtherm software.

Application of a General Computer Algorithm Based on the Group-Additivity Method for the Calculation of Two Molecular Descriptors at Both Ends of Dilution: Liquid Viscosity and Activity Coefficient in Water at Infinite Dilution

The application of a commonly used computer algorithm based on the group-additivity method for the calculation of the liquid viscosity coefficient at 293.15 K and the activity coefficient at infinite dilution in water at 298.15 K of organic molecules is presented. The method is based on the complete breakdown of the molecules into their constituting atoms, further subdividing them by their immediate neighborhood. A fast Gauss–Seidel fitting method using experimental data from literature is applied for the calculation of the atom groups’ contributions. Plausibility tests have been carried out on each of the calculations using a ten-fold cross-validation procedure which confirms the excellent predictive quality of the method. The goodness of fit (Q²) and the standard deviation (σ) of the cross-validation calculations for the viscosity coefficient, expressed as log(η), was 0.9728 and 0.11, respectively, for 413 test molecules, and for the activity coefficient log(γ)^∞ the corresponding values were 0.9736 and 0.31, respectively, for 621 test compounds. The present approach has proven its versatility in that it enabled the simultaneous evaluation of the liquid viscosity of normal organic compounds as well as of ionic liquids.

Application of neural networks to modeling and estimating temperature-dependent liquid viscosity of organic compounds

Back-propagation neural network models for correlating and predicting the viscosity-temperature behavior of a large variety of organic liquids were developed. Experimental values for the liquid viscosity for 1229 data points from 440 compounds containing C, H, N, O, S, and all halogens have been collected from the literature. The data ranges covered are from -120 to 160 degrees C for temperature and from 0.164 (trans-2-pentene at 20 degrees C) to 1.34 x 10(5) (glycerol at -20 degrees C) mPa.s for viscosity value. After dividing the total database of 440 compounds into training (237 with 673 data points), validation (124 with 423 data points), and test (79 with 133 data points) sets, the modeling performance of two separate neural network models with different architectures, one based on a compound-specific temperature dependence and the second based on a compound-independent one, has been examined. The resulting former model showed somewhat better modeling performance than latter, and the model gave squared correlation coefficients of 0.956, 0.932, and 0.884 and root mean-squares errors of 0.122, 0.134, and 0.148 log units for the training, validation, and test sets, respectively. The input descriptors include molar refraction, critical temperature, molar magnetic susceptibility, cohesive energy, temperatures, and five kinds of indicator variables for functionalities, alcohols/phenols, nitriles, amines, amides, and aliphatic ring. The reliability of the proposed model was assessed by comparing the results against calculated viscosities by two existing group-contribution approaches, the method of van Velzen et al. and the Joback and Reid method.