Accurate prediction of carbon dioxide capture by deep eutectic solvents using quantum chemistry and a neural network (original) (raw)

* Corresponding authors

a Deconstruction Division, Joint BioEnergy Institute, 5885 Hollis Street, Emeryville, California 94608, USA
E-mail: moodm@ornl.gov, mohanchauhan08@gmail.com, Seema.Rose.Singh@gmail.com, ssingh@lbl.gov

b Biosciences Division and Center for Molecular Biophysics, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, USA

c Biological Systems and Engineering Division, Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, California 94720, USA

d Department of Biochemistry and Cellular and Molecular Biology, University of Tennessee, Knoxville, Tennessee 37996, USA

e Manufacturing Science Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831-6201, USA

Abstract

Carbon dioxide (CO2) emissions from fossil fuel combustion are a significant source of greenhouse gas, contributing in a major way to global warming and climate change. Carbon dioxide capture and sequestration is gaining much attention as a potential method for controlling these greenhouse gas emissions. Among the environmentally friendly solvents, deep eutectic solvents (DESs) have demonstrated the potential capability for carbon capture. To establish a theoretical framework for DES activity, thermodynamics modeling and solubility predictions are significant factors to anticipate and understand the system behavior. Here, we combine the COSMO-RS model with machine learning techniques to predict the solubility of CO2 in various deep eutectic solvents. A comprehensive data set was established comprising 1973 CO2 solubility data points in 132 different DESs at a variety of temperatures, pressures, and DES molar ratios. This data set was then utilized for the further verification and development of the COSMO-RS model. The CO2 solubility (ln(_x_CO2)) in DESs calculated with the COSMO-RS model differs significantly from the experiment with an average absolute relative deviation (AARD) of 23.4%. A multilinear regression model was developed using the COSMO-RS predicted solubility and a temperature-pressure dependent parameter, which improved the AARD to 12%. Finally, a machine learning model using COSMO-RS-derived features was developed based on an artificial neural network algorithm. The results are in excellent agreement with the experimental CO2 solubilities, with an AARD of only 2.72%. The ML model will be a potentially useful tool for the design and selection of DESs for CO2 capture and utilization.

Graphical abstract: Accurate prediction of carbon dioxide capture by deep eutectic solvents using quantum chemistry and a neural network

This article is part of the themed collections:Machine learning and artificial neural networks: Celebrating the 2024 Nobel Prize in Physics, Machine Learning and Artificial Intelligence: A cross-journal collection and 2023 Green Chemistry Hot Articles

You have access to this article

Please wait while we load your content... Something went wrong. Try again?

Supplementary files

Article information

DOI

https://doi.org/10.1039/D2GC04425K

Article type

Paper

Submitted

22 Nov 2022

Accepted

22 Feb 2023

First published

28 Feb 2023

Download Citation

Green Chem., 2023,25, 3475-3492

Author version available

Permissions