1
|
Mohan M, Jetti KD, Smith MD, Demerdash ON, Kidder MK, Smith JC. Accurate Machine Learning for Predicting the Viscosities of Deep Eutectic Solvents. J Chem Theory Comput 2024; 20:3911-3926. [PMID: 38387055 DOI: 10.1021/acs.jctc.3c01163] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/24/2024]
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 R2 (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.
Collapse
Affiliation(s)
- Mood Mohan
- Biosciences Division and Center for Molecular Biophysics, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, United States
| | - Karuna Devi Jetti
- Department of Biotechnology, GIS, GITAM,Vishakhapatnam, Andhra Pradesh 530045, India
| | - Micholas Dean Smith
- Biosciences Division and Center for Molecular Biophysics, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, United States
- Department of Biochemistry and Cellular and Molecular Biology, University of Tennessee, Knoxville, Tennessee 37996, United States
| | - Omar N Demerdash
- Biosciences Division and Center for Molecular Biophysics, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, United States
| | - Michelle K Kidder
- Manufacturing Science Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831-6201, United States
| | - Jeremy C Smith
- Biosciences Division and Center for Molecular Biophysics, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, United States
- Department of Biochemistry and Cellular and Molecular Biology, University of Tennessee, Knoxville, Tennessee 37996, United States
| |
Collapse
|
2
|
Schieppati D, Mohan M, Blais B, Fattahi K, Patience GS, Simmons BA, Singh S, Boffito DC. Characterization of the acoustic cavitation in ionic liquids in a horn-type ultrasound reactor. ULTRASONICS SONOCHEMISTRY 2024; 102:106721. [PMID: 38103370 PMCID: PMC10765111 DOI: 10.1016/j.ultsonch.2023.106721] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 05/19/2023] [Revised: 12/07/2023] [Accepted: 12/08/2023] [Indexed: 12/19/2023]
Abstract
Most ultrasound-based processes root in empirical approaches. Because nearly all advances have been conducted in aqueous systems, there exists a paucity of information on sonoprocessing in other solvents, particularly ionic liquids (ILs). In this work, we modelled an ultrasonic horn-type sonoreactor and investigated the effects of ultrasound power, sonotrode immersion depth, and solvent's thermodynamic properties on acoustic cavitation in nine imidazolium-based and three pyrrolidinium-based ILs. The model accounts for bubbles, acoustic impedance mismatch at interfaces, and treats the ILs as incompressible, Newtonian, and saturated with argon. Following a statistical analysis of the simulation results, we determined that viscosity and ultrasound input power are the most significant variables affecting the intensity of the acoustic pressure field (P), the volume of cavitation zones (V), and the magnitude of the maximum acoustic streaming surface velocity (u). V and u increase with the increase of ultrasound input power and the decrease in viscosity, whereas the magnitude of negative P decreases as ultrasound power and viscosity increase. Probe immersion depth positively correlates with V, but its impact on P and u is insignificant. 1-alkyl-3-methylimidazolium-based ILs yielded the largest V and the fastest acoustic jets - 0.77 cm3 and 24.4 m s-1 for 1-ethyl-3-methylimidazolium chloride at 60 W. 1-methyl-3-(3-sulfopropyl)-imidazolium-based ILs generated the smallest V and lowest u - 0.17 cm3 and 1.7 m s-1 for 1-methyl-3-(3-sulfopropyl)-imidazolium p-toluene sulfonate at 20 W. Sonochemiluminescence experiments validated the model.
Collapse
Affiliation(s)
- Dalma Schieppati
- Department of Chemical Engineering, École Polytechnique Montréal, C.P. 6079, Succ. CV, Montréal H3C 3A7, Québec, Canada
| | - Mood Mohan
- Deconstruction Division, Joint BioEnergy Institute, 5885 Hollis Street, Emeryville, CA 94608, USA; Bioscience Division and Center for Molecular Biophysics, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA
| | - Bruno Blais
- Department of Chemical Engineering, École Polytechnique Montréal, C.P. 6079, Succ. CV, Montréal H3C 3A7, Québec, Canada
| | - Kobra Fattahi
- Department of Chemical Engineering, École Polytechnique Montréal, C.P. 6079, Succ. CV, Montréal H3C 3A7, Québec, Canada
| | - Gregory S Patience
- Department of Chemical Engineering, École Polytechnique Montréal, C.P. 6079, Succ. CV, Montréal H3C 3A7, Québec, Canada
| | - Blake A Simmons
- Deconstruction Division, Joint BioEnergy Institute, 5885 Hollis Street, Emeryville, CA 94608, USA; Biological Systems and Engineering Division, Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA 94720, USA
| | - Seema Singh
- Deconstruction Division, Joint BioEnergy Institute, 5885 Hollis Street, Emeryville, CA 94608, USA
| | - Daria C Boffito
- Department of Chemical Engineering, École Polytechnique Montréal, C.P. 6079, Succ. CV, Montréal H3C 3A7, Québec, Canada.
| |
Collapse
|