Prediction of CO2 absorption by physical solvents using a chemoinformatics-based machine learning model

Autor: Zhien Zhang, Hao Li, Dan Yan, Eric Lichtfouse
Přispěvatelé: The University of Texas at Austin, Tsinghua University [Beijing] (THU), Ohio State University [Columbus] (OSU), Centre européen de recherche et d'enseignement des géosciences de l'environnement (CEREGE), Institut de Recherche pour le Développement (IRD)-Institut National de la Recherche Agronomique (INRA)-Aix Marseille Université (AMU)-Collège de France (CdF (institution))-Institut national des sciences de l'Univers (INSU - CNRS)-Centre National de la Recherche Scientifique (CNRS), Aix Marseille Université (AMU)-Institut national des sciences de l'Univers (INSU - CNRS)-Collège de France (CdF (institution))-Institut de Recherche pour le Développement (IRD)-Centre National de la Recherche Scientifique (CNRS)-Institut National de la Recherche Agronomique (INRA), Tsinghua University [Beijing], Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD)-Collège de France (CdF)-Institut national des sciences de l'Univers (INSU - CNRS)-Aix Marseille Université (AMU)-Institut National de la Recherche Agronomique (INRA)
Rok vydání: 2019
Předmět:
Materials science
[SDE.MCG]Environmental Sciences/Global Changes
02 engineering and technology
010501 environmental sciences
Mole fraction
Machine learning
computer.software_genre
chemoinformatics
01 natural sciences
chemistry.chemical_compound
Acetone
Environmental Chemistry
[CHIM.COOR]Chemical Sciences/Coordination chemistry
Solubility
0105 earth and related environmental sciences
Artificial neural network
[CHIM.ORGA]Chemical Sciences/Organic chemistry
[SDE.IE]Environmental Sciences/Environmental Engineering
business.industry
physical solvent
solubility
[SPI.FLUID]Engineering Sciences [physics]/Reactive fluid environment
prediction
chemical descriptors
021001 nanoscience & nanotechnology
[CHIM.THEO]Chemical Sciences/Theoretical and/or physical chemistry
machine learning
chemistry
greenhouse gas
13. Climate action
Cheminformatics
co2
Methanol
Artificial intelligence
Absorption (chemistry)
0210 nano-technology
business
absorption
Ethylene glycol
computer
Zdroj: Environmental Chemistry Letters
Environmental Chemistry Letters, 2019, 17 (3), pp.1397-1404. ⟨10.1007/s10311-019-00874-0⟩
Environmental Chemistry Letters, Springer Verlag, 2019, 17 (3), pp.1397-1404. ⟨10.1007/s10311-019-00874-0⟩
ISSN: 1610-3661
1610-3653
Popis: International audience; The rising atmospheric CO2 level is partly responsible for global warming. Despite numerous warnings from scientists during the past years, nations are reacting too slowly, and thus, we will probably reach a situation needing rapid and effective techniques to reduce atmospheric CO2. Therefore, advanced engineering methods are particularly important to decrease the greenhouse effect, for instance, by capturing CO2 using solvents. Experimental testing of many solvents under different conditions is necessary but time-consuming. Alternatively, modeling CO2 capture by solvents using a nonlinear fitting machine learning is a rapid way to select potential solvents, prior to experimentation. Previous predictive machine learning models were mainly designed for blended solutions in water using the solution concentration as the main input of the model, which was not able to predict CO2 solubility in different types of physical solvents. To address this issue, here, we developed a new descriptor-based chemoinformatics model for predicting CO2 solubility in physical solvents in the form of mole fraction. The input factors include organic structural and bond information, thermodynamic properties, and experimental conditions. We studied the solvents from 823 data, including methanol (165 data), ethanol (138), n-propanol (98), n-butanol (64), n-pentanol (59), ethylene glycol (52), propylene glycol (54), acetone (51), 2-butanone (49), ethylene glycol monomethyl ether (46 data), and ethylene glycol monoethyl ether (47), using artificial neural networks as the machine learning model. Results show that our descriptor-based model predicts the CO2 absorption in physical solvents with generally higher accuracy and low root-mean-squared errors. Our findings show that using a set of simple but effective chemoinformatics-based descriptors, intrinsic relationships between the general properties of physical solvents and their CO2 solubility can be precisely fitted with machine learning.
Databáze: OpenAIRE
Externí odkaz: