Predicting CO2 capture of ionic liquids using machine learning

Abstract Ionic liquid (IL) based CO 2 capture is currently seen as a promising alternative to conventional amine-based solvents. While the possible combinations of cations and anions are numerous, it is time consuming and expensive to carry out experimental measurements for CO 2 solubilities for each new IL. Therefore, as a means to rapidly screen suitable ILs as potential solvents for CO 2 absorption, we investigate the use of machine learning (ML) based models to establish structure–property relationships between molecular structures of cations and anions and their CO 2 solubilities. Over 10,000 IL-CO 2 solubility data of 185 ILs measured at different operating temperatures and pressures were extracted from the literature. Using semi-empirically derived geometrical and charge-based molecular descriptors, good agreement with the available experimental measurements was obtained for both single decision tree (mean absolute error of 0.10) and ensemble random forest (mean absolute error of 0.04) approaches. The results were found to be more accurate than those obtained with the quantum chemistry based COSMOtherm predictions.

[1]  C. Jun,et al.  Performance of some variable selection methods when multicollinearity is present , 2005 .

[2]  Amir AghaKouchak,et al.  Water and climate: Recognize anthropogenic drought , 2015, Nature.

[3]  J. Brennecke,et al.  Anion effects on gas solubility in ionic liquids. , 2005, The journal of physical chemistry. B.

[4]  Geert Versteeg,et al.  Gas solubility of H2S and CO2 in aqueous solutions of N-methyldiethanolamine , 2007 .

[5]  Armistead G Russell,et al.  Amine-based CO2 capture technology development from the beginning of 2013-a review. , 2015, ACS applied materials & interfaces.

[6]  K. A. Kurnia,et al.  Dispelling some myths about the CO2 solubility in ionic liquids. , 2016, Physical chemistry chemical physics : PCCP.

[7]  Zhigang Lei,et al.  Gas solubility in ionic liquids. , 2014, Chemical reviews.

[8]  Joan F Brennecke,et al.  Solubility of CO2, CH4, C2H6, C2H4, O2, and N2 in 1-hexyl-3-methylpyridinium bis(trifluoromethylsulfonyl)imide: comparison to other ionic liquids. , 2007, Accounts of chemical research.

[9]  Jasper Knight,et al.  The impacts of climate change on terrestrial Earth surface systems , 2013 .

[10]  William F. Schneider,et al.  Molecular Design of High Capacity, Low Viscosity, Chemically Tunable Ionic Liquids for CO2 Capture , 2010 .

[11]  B. Kowalski,et al.  Partial least-squares regression: a tutorial , 1986 .

[12]  M. Mirarab,et al.  Intelligent Prediction of CO2 Capture in Propyl Amine Methyl Imidazole Alanine Ionic Liquid: An Artificial Neural Network Model , 2015 .

[13]  W. Goddard,et al.  UFF, a full periodic table force field for molecular mechanics and molecular dynamics simulations , 1992 .

[14]  Li Fei Ji,et al.  Substituent effects on the properties of the hemi-bonded complexes (XH2P···NH2Y)+ (X, Y=H, F, Cl, Br, NH2, CH3, OH) , 2015, Journal of Molecular Modeling.

[15]  B. Alsberg,et al.  A quantitative structure-property relationship study of the photovoltaic performance of phenothiazine dyes , 2015 .

[16]  Alireza Baghban,et al.  Estimating hydrogen sulfide solubility in ionic liquids using a machine learning approach , 2014 .

[17]  Ron Wehrens,et al.  The pls Package: Principal Component and Partial Least Squares Regression in R , 2007 .

[18]  Vishwesh Venkatraman,et al.  Quantitative structure-property relationship modelling of thermal decomposition temperatures of ionic liquids , 2016 .

[19]  Nusret Duygu Yilmazer,et al.  Large-scale virtual high-throughput screening for the identification of new battery electrolyte solvents: computing infrastructure and collective properties. , 2015, Physical chemistry chemical physics : PCCP.

[20]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[21]  Nasrullah Khan,et al.  Carbon conundrum, climate change, CO2 capture and consumptions , 2014 .

[22]  Amir H. Mohammadi,et al.  Application of decision tree learning in modelling CO2 equilibrium absorption in ionic liquids , 2017 .

[23]  R. Stine Bootstrap Prediction Intervals for Regression , 1985 .

[24]  A. Klamt,et al.  COSMO-RS as a tool for property prediction of IL mixtures—A review , 2010 .

[25]  J. Stewart Optimization of parameters for semiempirical methods V: Modification of NDDO approximations and application to 70 elements , 2007, Journal of molecular modeling.

[26]  S. Garg,et al.  Experimental data, thermodynamic and neural network modeling of CO2 solubility in aqueous sodium salt of l-phenylalanine , 2017 .

[27]  Frank Neese,et al.  Calculation of solvent shifts on electronic g-tensors with the conductor-like screening model (COSMO) and its self-consistent generalization to real solvents (direct COSMO-RS). , 2006, The journal of physical chemistry. A.

[28]  Vishwesh Venkatraman,et al.  Evolutionary de novo design of phenothiazine derivatives for dye-sensitized solar cells , 2015 .

[29]  A. Y. Meyer The size of molecules , 1987 .

[30]  J. Topliss,et al.  Chance correlations in structure-activity studies using multiple regression analysis , 1972 .

[31]  A. Tropsha,et al.  Development and validation of k-nearest-neighbor QSPR models of metabolic stability of drug candidates. , 2003, Journal of medicinal chemistry.

[32]  James D. Malley,et al.  Predictor correlation impacts machine learning algorithms: implications for genomic studies , 2009, Bioinform..

[33]  M. Klähn,et al.  What Determines CO₂ Solubility in Ionic Liquids? A Molecular Simulation Study. , 2015, The journal of physical chemistry. B.

[34]  Surya Moganty,et al.  Regular Solution Theory for Low Pressure Carbon Dioxide Solubility in Room Temperature Ionic Liquids: Ionic Liquid Solubility Parameter from Activation Energy of Viscosity , 2010 .

[35]  J. A. Lazzús,et al.  Activity coefficient models to describe isothermal vapor-liquid equilibrium of binary systems containing ionic liquids , 2010 .

[36]  J. Valderrama Myths and Realities about Existing Methods for Calculating the Melting Temperatures of Ionic Liquids , 2014 .

[37]  Nicolai Meinshausen,et al.  Quantile Regression Forests , 2006, J. Mach. Learn. Res..

[38]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[39]  S. Dai,et al.  Molecular Dynamics Simulation of Anion Effect on Solubility, Diffusivity, and Permeability of Carbon Dioxide in Ionic Liquids , 2014 .

[40]  T. Hoye,et al.  A guide to small-molecule structure assignment through computation of (1H and 13C) NMR chemical shifts , 2014, Nature Protocols.

[41]  Alireza Baghban,et al.  Prediction carbon dioxide solubility in presence of various ionic liquids using computational intelligence approaches , 2015 .

[42]  K. Hornik,et al.  Unbiased Recursive Partitioning: A Conditional Inference Framework , 2006 .

[43]  E. Batista,et al.  Applications of Ionic Liquids in the Food and Bioproducts Industries , 2016 .

[44]  J. Bara,et al.  Free Volume as the Basis of Gas Solubility and Selectivity in Imidazolium-Based Ionic Liquids , 2012 .

[45]  Wilhelm Kuckshinrichs,et al.  Worldwide innovations in the development of carbon capture technologies and the utilization of CO2 , 2012 .

[46]  Geoffrey W. Stevens,et al.  Review of solvent based carbon-dioxide capture technologies , 2015, Frontiers of Chemical Science and Engineering.

[47]  M. Rahimpour,et al.  The ability of artificial neural network in prediction of the acid gases solubility in different ionic liquids , 2015 .

[48]  Behzad Pouladi,et al.  Connectionist technique estimates H2S solubility in ionic liquids through a low parameter approach , 2015 .

[49]  Vishwesh Venkatraman,et al.  KRAKENX: software for the generation of alignment-independent 3D descriptors , 2016, Journal of Molecular Modeling.

[50]  Vishwesh Venkatraman,et al.  Artificial evolution of coumarin dyes for dye sensitized solar cells. , 2015, Physical chemistry chemical physics : PCCP.

[51]  Pawel Jajesniak,et al.  Carbon Dioxide Capture and Utilization using Biological Systems: Opportunities and Challenges , 2014 .

[52]  Thijs J. H. Vlugt,et al.  State-of-the-Art of CO2 Capture with Ionic Liquids , 2012 .

[53]  Jean-Michel Poggi,et al.  Variable selection using random forests , 2010, Pattern Recognit. Lett..

[54]  J. Brennecke,et al.  Improving carbon dioxide solubility in ionic liquids. , 2007, The journal of physical chemistry. B.

[55]  Patricia Luis,et al.  Use of monoethanolamine (MEA) for CO2 capture in a global scenario: Consequences and alternatives , 2016 .

[56]  E. J. Anthony,et al.  Carbon capture and storage update , 2014 .

[57]  F. Mjalli,et al.  Prediction of CO2 solubility in ionic liquids using the PSRK model , 2015 .

[58]  B. Van der Bruggen,et al.  Application of artificial neural networks for estimation of solubility of acid gases (H2S and CO2) in 32 commonly ionic liquid and amine solutions , 2015 .

[59]  João A. P. Coutinho,et al.  Predictive methods for the estimation of thermophysical properties of ionic liquids , 2012 .

[60]  Vishwesh Venkatraman,et al.  Computer-aided molecular design of imidazole-based absorbents for CO2 capture , 2016 .

[61]  A. Henni,et al.  Ionic liquids for CO2 capture using COSMO-RS: Effect of structure, properties and molecular interactions on solubility and selectivity , 2011 .

[62]  B. Berne,et al.  Why is the partial molar volume of CO2 so small when dissolved in a room temperature ionic liquid? Structure and dynamics of CO2 dissolved in [Bmim+] [PF6(-)]. , 2005, Journal of the American Chemical Society.

[63]  Frank Neese,et al.  The ORCA program system , 2012 .

[64]  Yu-feng Hu,et al.  The molecular characteristics dominating the solubility of gases in ionic liquids. , 2011, Chemical Society reviews.

[65]  Fadwa T. Eljack,et al.  Designing ionic liquid solvents for carbon capture using property-based visual approach , 2016, Clean Technologies and Environmental Policy.

[66]  M. A. Bustam,et al.  Solubility of CO2 in pyridinium based ionic liquids , 2012 .

[67]  Siavash Riahi,et al.  The QSPR models to predict the solubility of CO2 in ionic liquids based on least-squares support vector machines and genetic algorithm-multi linear regression , 2017 .

[68]  Kenichi Fukui,et al.  Theory of Orientation and Stereoselection , 1975 .

[69]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[70]  P. Simon,et al.  Energy applications of ionic liquids , 2014 .

[71]  M. Rahimpour,et al.  Prediction of CO2 solubility in ionic liquids with QM and UNIQUAC models , 2016 .

[72]  Amr Henni,et al.  Densities and low pressure solubilities of carbon dioxide in five promising ionic liquids , 2014 .

[73]  F. Weigend,et al.  Balanced basis sets of split valence, triple zeta valence and quadruple zeta valence quality for H to Rn: Design and assessment of accuracy. , 2005, Physical chemistry chemical physics : PCCP.

[74]  Chris Morley,et al.  Open Babel: An open chemical toolbox , 2011, J. Cheminformatics.

[75]  Farhad Gharagheizi,et al.  Ionic liquids: Prediction of melting point by molecular-based model , 2012 .

[76]  Scott D. Kahn,et al.  Current Status of Methods for Defining the Applicability Domain of (Quantitative) Structure-Activity Relationships , 2005, Alternatives to laboratory animals : ATLA.

[77]  Kamil Paduszynski,et al.  Viscosity of Ionic Liquids: An Extensive Database and a New Group Contribution Model Based on a Feed-Forward Artificial Neural Network , 2014, J. Chem. Inf. Model..

[78]  Igor V Tetko,et al.  Robustness in experimental design: A study on the reliability of selection approaches , 2013, Computational and structural biotechnology journal.