文献类型：期刊文献

英文题名：Intelligent Molecular Identification Approach to High-Efficiency Solvents for Organosulfide Capture Using the Active Machine Learning Framework

作者：Chen, Yuxiang[1,2];Liu, Chuanlei[1,2];An, Yang[1,2];Lou, Yue[1,2];Zhao, Yang[1,2];Qian, Cheng[1,2];Jiang, Hao[1,2];Wu, Kongguo[1,2];Shen, Benxian[1,2];Zhang, Xianghui[3,4];Cao, Fahai[1];Wu, Di[3,4,5,6];Sun, Hui[1,2]

机构：[1]East China Univ Sci & Technol, Sch Chem Engn, Shanghai 200237, Peoples R China;[2]East China Univ Sci & Technol, Int Joint Res Ctr Green Energy Chem Engn, Shanghai 200237, Peoples R China;[3]Washington State Univ, Alexandra Navrotsky Inst Expt Thermodynam, Pullman, WA 99163 USA;[4]Washington State Univ, Gene & Linda Voiland Sch Chem Engn & Bioengn, Pullman, WA 99163 USA;[5]Washington State Univ, Dept Chem, Pullman, WA 99163 USA;[6]Washington State Univ, Mat Sci & Engn, Pullman, WA 99163 USA

年份：2023

卷号：37

期号：16

起止页码：12123

外文期刊名：ENERGY & FUELS

收录：;EI(收录号：20233414614896);WOS:【SCI-EXPANDED(收录号:WOS:001041669900001)】；

基金：This work was financially supported by the National Natural Science Foundation of China (grants 21878097 and 22178109) and the Natural Science Foundation of Shanghai (grant 21ZR1417700). D.W. acknowledges institutional funds from the Gene and Linda Voiland School of Chemical Engineering and Bioengineering and the Alexandra Navrotsky Institute for Experimental Thermodynamics at Washington State University.

语种：英文

外文关键词：Iterative methods - Machine learning

摘要：There is increasing interest in the development of intelligentstrategies for rational design and/or identification of promisingsolvent compounds for volatile, environmentally unfriendly compoundcapture. However, typical computer-aided methods require huge datasetsand/or suffer from accumulated prediction bias. Here, we constructeda computational framework by introducing a stepwise screen approachfor molecular descriptors and molecular active selection machine learningto modify the adequate chemical space iteratively. This frameworkidentifies the optimal solvent candidates by molecular similaritysearch and iterative molecular addition to the training dataset. Ina virtual screening of 126,068 compounds, 2443 solvent candidateswere successfully identified for the capture of methyl mercaptan (MeSH),one of the major organosulfides in fossil gases.