Predictive Models for the Binary Diffusion Coefficient at Infinite Dilution in Polar and Nonpolar Fluids

Abstract

Experimental diffusivities are scarcely available, though their knowledge is essential to model rate-controlled processes. In this work various machine learning models to estimate diffusivities in polar and nonpolar solvents (except water and supercritical CO2) were developed. Such models were trained on a database of 90 polar systems (1431 points) and 154 nonpolar systems (1129 points) with data on 20 properties. Five machine learning algorithms were evaluated: multilinear regression, k-nearest neighbors, decision tree, and two ensemble methods (random forest and gradient boosted). For both polar and nonpolar data, the best results were found using the gradient boosted algorithm. The model for polar systems contains 6 variables/parameters (temperature, solvent viscosity, solute molar mass, solute critical pressure, solvent molar mass, and solvent Lennard-Jones energy constant) and showed an average deviation (AARD) of 5.07%. The nonpolar model requires five variables/parameters (the same of polar systems except the Lennard-Jones constant) and presents AARD = 5.86%. These results were compared with four classic models, including the 2-parameter correlation of Magalhães et al. (AARD = 5.19/6.19% for polar/nonpolar) and the predictive Wilke-Chang equation (AARD = 40.92/29.19%). Nonetheless Magalhães et al. requires two parameters per system that must be previously fitted to data. The developed models are coded and provided as command line program.

Diffusion of Squalene in Nonaqueous Solvents

“…Machine learning studies of diffusion have focused on solutes in polar, 19 nonpolar, 19 and supercritical CO 2 18 solvents. The gradient boosted ML algorithm gave the best agreement of 2.58% for 1476 D values in CO 2 , 18 5.07% for 430 D values in polar solvents, 19 and 5.86% for 342 D values in nonpolar solvents. 19 Our D values in the alkanes and cyclohexanes are candidates for the solute-nonpolar solvent data set.…”

“… 1 , 13 Our diffusion constants for squalene also may be useful for checking molecular dynamics (MD) computer codes 16 , 17 and machine learning (ML) diffusion constant predictions. 18 , 19 …”

“…Two other MSE fits have been made that may be useful for estimating the viscosities of lipid droplets and the diffusion constants needed to design the supercritical and pressurized extractions by which squalene can be obtained from plant sources . These separations have become more important because international regulations reduced the supply of squalene’s traditional source, shark liver oil. , Our diffusion constants for squalene also may be useful for checking molecular dynamics (MD) computer codes , and machine learning (ML) diffusion constant predictions. , …”

“…1,13 Our diffusion constants for squalene also may be useful for checking molecular dynamics (MD) computer codes 16,17 and machine learning (ML) diffusion constant predictions. 18,19 ■ EXPERIMENTAL METHODS Chemicals and Sample Preparation. In this and the following sections, n-C i is used for the n-alkanes and n-C i C 6 H 11 is used for the cyclohexanes.…”

A Theoretical Analysis on a Multi-Bed Pervaporation Membrane Reactor during Levulinic Acid Esterification Using the Computational Fluid Dynamic Method

“…The ϕM i was therefore obtained for each component according to the Equation (16): where is a molar fraction, in each step, determined for three components, using the i-th component and varying the j factor [ 28 ]. The temperature dependence on the viscosity was hence evaluated for each single component present in the reaction system, based on the empirical expression retrieved from the CHEMCAD database [ 29 ]. …”