Predicting the Surface Tension of Deep Eutectic Solvents Using Artificial Neural Networks

Abstract

Studies on deep eutectic solvents (DESs), a new class of “green” solvents, are attracting increasing attention from researchers, as evidenced by the rapidly growing number of publications in the literature. One of the main advantages of DESs is that they are tailor-made solvents, and therefore, the number of potential DESs is extremely large. It is essential to have computational methods capable of predicting the physicochemical properties of DESs, which are needed in many industrial applications and research. Surface tension is one of the most important properties required in many applications. In this work, we report a relatively generalized artificial neural network (ANN) for predicting the surface tension of DESs. The database used can be considered comprehensive because it contains 1571 data points from 133 different DES mixtures in 520 compositions prepared from 18 ions and 63 hydrogen bond donors in a temperature range of 277–425 K. The ANN model uses molecular parameter inputs derived from the conductor-like screening model for real solvents ( S σ-profiles ). The training and testing results show that the best performing ANN architecture consisted of two hidden layers with 15 neurons each (9–15–15–1). The proposed ANN was excellent in predicting the surface tension of DESs, as R 2 values of 0.986 and 0.977 were obtained for training and testing, respectively, with an overall average absolute relative deviation of 2.20%. The proposed models represent an initiative to promote the development of robust models capable of predicting the properties of DESs based only on molecular parameters, leading to savings in investigation time and resources.

Critical Properties of Ternary Deep Eutectic Solvents Using Group Contribution with Extended Lee–Kesler Mixing Rules

“…To show how the determined critical properties and acentric factors can be used for the development of computational models for ternary DESs, we also developed in this work a similar model that is based on the density of 4995 data points (Table 4), which was established with stepwise regression and analysis of variance (ANOVA) using JMP SAS Pro software (version 16). 73,74 The algorithm resulted in the following expression in eq 19, which is detailed in Table 5.…”

“…The physical, physicochemical, and transport properties of DESs, such as boiling point, density, viscosity, surface tension, pH, vapor pressure, etc., are primarily determined by the interaction and structure of the DES’s constituents. Such properties can be tailored by modifying the HBA and HBD structure and their corresponding molar ratios. , Like any solvent, the physicochemical properties of DESs play a pivotal role in the feasibility and applicability of a specific chemical or physical process. − For example, the density of a solvent is of great importance in the separation between the aqueous and organic phases for hydrophobic DESs, , and the efficiency of solid–liquid extraction processes ( e.g ., sugar extraction). Likewise, the viscosity of a solvent affects the mass transfer from and to the solvent. , For a balanced process operation and design, careful choice of the DES and its constituent should be considered .…”

“… 2 , 11 Like any solvent, the physicochemical properties of DESs play a pivotal role in the feasibility and applicability of a specific chemical or physical process. 12 − 14 For example, the density of a solvent is of great importance in the separation between the aqueous and organic phases for hydrophobic DESs, 15 , 16 and the efficiency of solid–liquid extraction processes ( e.g ., sugar extraction 17 ). Likewise, the viscosity of a solvent affects the mass transfer from and to the solvent.…”

Multitask Neural Network for Mapping the Glass Transition and Melting Temperature Space of Homo- and Co-Polyhydroxyalkanoates Using σProfiles Molecular Inputs

“…107 Furthermore, as was aforementioned, each σ Profiles is composed of 51 points, which represent the σ Profiles values at 51 different σ values, where it has been successfully included as a molecular descriptor in our previous works for predicting deep eutectic solvents and ionic liquids. 89,90 Abranches et al 108 also used the σ Profiles of 1432 molecular solvents as molecular descriptors in ML to develop ANNs that reliably predict a wide variety of physical and chemical properties including molar mass, boiling temperature, vapor pressure, density, refractive index, and solubility.…”

“…Nevertheless, there is no defined method to select the ideal network architecture and determine the number of hidden layers and the number of neurons in each hidden layer. For instance, previous studies used either one or two hidden layers , for predicting the properties of deep eutectic solvents. Thus, the most prevalent approach in the literature is trial and error .…”