SIOC English
化学学报
高级检索

化学学报 >

2020 , Vol. 78 >Issue 5: 427 - 436

DOI: https://doi.org/10.6023/A20030065

研究论文

基于机器学习和高通量计算筛选金属有机框架的甲烷/乙烷/丙烷分离性能

  • 蔡铖智 ,
  • 李丽凤 ,
  • 邓小梅 ,
  • 李树华 ,
  • 梁红 ,
  • 乔智威
展开
  • 广州大学化学化工学院 能源与催化研究所 广州 510006

收稿日期: 2020-03-13

  网络出版日期: 2020-04-16

基金资助

项目受国家自然科学基金（Nos.21978058，21676094，21576058）和广东省自然科学基金（No.2020A1515010800）资助.

收起

Machine Learning and High-throughput Computational Screening of Metal-organic Framework for Separation of Methane/ethane/propane

  • Cai Chengzhi ,
  • Li Lifeng ,
  • Deng Xiaomei ,
  • Li Shuhua ,
  • Liang Hong ,
  • Qiao Zhiwei
Expand
  • Guangzhou Key Laboratory for New Energy and Green Catalysis, School of Chemistry and Chemical Engineering, Guangzhou University, Guangzhou 510006

Received date: 2020-03-13

  Online published: 2020-04-16

Supported by

Project supported by the National Natural Science Foundation of China (Nos. 21978058, 21676094, 21576058) and the Natural Science Foundation of Guangdong Province (No. 2020A1515010800).

Fold

摘要

针对天然气中的甲烷、乙烷、丙烷（C1、C2、C3）气体分离困难的问题，本工作采用高通量计算了137953种假设的金属有机框架（Metal-organic framework，MOF）对这三种混合气体的吸附分离吸能.为了避免水蒸气的竞争吸附，首先，筛选出31399种疏水性MOF.然后，单变量分析了这些MOF的最大孔径（LCD）、孔隙率（φ）、体积比表面积（VSA）、亨利系数（K）、吸附热（Qst）、密度（ρ）共六种MOF结构/能量描述符与MOF对C1、C2、C3的选择性、吸附量及两者权衡值（Trade-off between Si/j and Ni，TSN）的关系，发现了吸附量和选择性“第二峰值”的存在；尤其对于C1、C2的分离，所有最优MOF都分布在第二峰值区间.随后采用决策树、随机森林（Random forest，RF）、支持向量机和反向传播神经网络四种机器学习算法，分别训练并预测了六种MOF描述符与性能指标的关系，结果表明RF预测效果最好.然后应用RF算法定量地分析出K、LCD和ρ三种描述符对TSNC1、TSNC2的相对重要性最高，而TSNC3的是K、Qst和ρ，根据这些描述符分别设计了吸附C1、C2、C3最优MOF的决策树模型路径.最后筛选出针对C1、C2和C3不同分离应用的18种最优MOF.本工作基于机器学习和高通量计算的研究思路和研究方法，第二峰值规律的发现以及最优设计路线的提出将有助于MOF在吸附分离领域的发展提供有力的指导和启示.

关键词： 金属有机框架; 气体分离; 分子模拟; 机器学习

本文引用格式

蔡铖智 , 李丽凤 , 邓小梅 , 李树华 , 梁红 , 乔智威 . 基于机器学习和高通量计算筛选金属有机框架的甲烷/乙烷/丙烷分离性能[J]. 化学学报, 2020 , 78(5) : 427 -436 . DOI: 10.6023/A20030065

Abstract

In this work, the separation performance of methane/ethane/propane (C1, C2 and C3) mixture in the 137953 hypothetical metal-organic frameworks (MOFs) is calculated by high throughput computational screening and multiple machine learning (ML) algorithms. First, to avoid the competitive adsorption of water vapor, 31399 hydrophobic MOFs (hMOFs) were screened out. Then, grand canonical Monte Carlo (GCMC) simulations were employed to calculate the adsorption behavior of a mixture with a mole ratio of C1:C2:C3=7:2:1 in these hMOFs, respectively. Second, the relationships among six MOF structures/energy descriptors (the largest cavity diameter (LCD), void fraction (f), volumetric surface area (VSA), Henry coefficient (K), heat of adsorption (Qst), density of MOF (ρ)) and three performance indicators of MOFs (selectivities (S), adsorption capacities (N) of C1, C2, C3 and their trade-offs (TSN)) were established. The LCDs were calculated by Zeo++software, and VSAs were calculated using RASPA software using He and N2 as probes, respectively, and Qst and K were calculated in an infinite dilution of each gas molecule in an infinite dilution state using NVT-MC method in RASPA software. Then, we found that there existed the "second peaks" of N and S in part of structure-property relationships, and all the optimal MOFs located in the range of second peaks, especially for the separation of C1 or C2. Third, the above-mentioned six MOF descriptors and three MOF performance indicators were trained, tested and predicted by four ML algorithms, including decision tree, random forest (RF), support vector machine and Back Propagation neural network. Although the predictive effect for the selectivity was very low, the introduction of TSN can significantly improve the accuracy of ML prediction, especially for RF algorithm (R=0.99). Therefore, the RF was used to quantitatively analyze the relative importance of each MOF descriptor, and found that three descriptors (K, LCD and ρ) possessed the highest importance for the separation of C1 and C2, and three other descriptors (K, Qst and ρ) for the separation of C3. Moreover, three simple and clear paths of optimal MOFs for C1, C2 and C3 adsorption were designed by the decision tree model with the descriptors. Based on those paths, there were 96%, 85%, 95% probability that we can search for high-performance MOFs, respectively. Finally, the best 18 MOFs were identified for different separation applications of C1, C2 and C3. This study reveals the second peaks and key MOF descriptors governing the adsorption of light alkane, develops quantitative structure-property relationships by ML, and identifies the best adsorbents from a large collection of MOFs for the separation of C1, C2 and C3 from natural gas.

Key words： metal-organic framework; gas separation; molecular simulation; machine learning

参考文献

[1] Schoots, K.; Rivera-Tinoco, R.; Verbong, G.; van der Zwaan, B. Int. J. Greenhouse Gas Control. 2011, 5, 1614.
[2] Wu, F. F. M.S. Thesis, Tianjin University, Tianjin, 2014 (in Chinese). (吴菲菲, 硕士论文, 天津大学, 天津, 2014.)
[3] Ravanchi, M. T.; Kaghazchi, T.; Kargari, A.; Soleimani, M. J. Taiwan Inst. Chem. Eng. 2009, 40, 511.
[4] Xie, C. L.; Fang, Y. D. Petrochem. Ind. Technol. 2005, 12, 63. (谢春雷, 方义东, 石化技术, 2005, 12, 63.)
[5] Wu, D. M.S. Thesis, Tianjin University, Tianjin, 2012 (in Chinese). (吴頔, 硕士论文, 天津大学, 天津, 2012.)
[6] Li, X. F.; Li, D. F. Petrochem. Technol. 2007, 36, 94. (李晓峰, 李东风, 石油化工, 2007, 36, 94.)
[7] Ma, Y. T.; Cong, S. G.; Hu, Y. F. Energy Chem. Ind. 2017, 38, 34. (马宇彤, 丛树阁, 胡云峰, 能源化工, 2017, 38, 34.)
[8] Zhang, H.; Liu, Y. S.; Liu, W. H.; Zhang, D. X.; Zhai, H. Chem. Ind. Eng. Prog. 2007, 26, 95. (张辉, 刘应书, 刘文海, 张德鑫, 翟晖, 化工进展, 2007, 26, 95.)
[9] Yu, Q. Q. M.S. Thesis, Beijing University of Chemical Technology, Beijing, 2016 (in Chinese). (于清泉, 硕士论文, 北京化工大学, 北京, 2016.)
[10] Li, S. Z. M.S. Thesis, Harbin Institute of Technology, Harbin, 2011 (in Chinese). (李守柱, 硕士论文, 哈尔滨工业大学, 哈尔滨, 2011.)
[11] Wu, X. J.; Zhao, P.; Fang, J. M.; Wang, J.; Liu, B. S.; Cai, W. Q. Acta Phys.-Chim. Sin. 2014, 30, 2043. (吴选军, 赵鹏, 方继敏, 王杰, 刘保顺, 蔡卫权, 物理化学学报, 2014, 30, 2043.)
[12] Zhou, J. H.; Zhao, H. L.; Hu, J.; Liu, H. L.; Hu, Y. CIESC J. 2014, 65, 1680. (周建海, 赵会玲, 胡军, 刘洪来, 胡英, 化工学报, 2014, 65, 1680.)
[13] Zhu, G. F.; Chen, L. T.; Cheng, G. H.; Zhao, J.; Yang, C.; Zhang, Y. Z.; Wang, X.; Fan, J. Acta Chim. Sinica 2019, 77, 434. (朱桂芬, 陈乐田, 程国浩, 赵娟, 杨灿, 张耀宗, 王醒, 樊静, 化学学报, 2019, 77, 434.)
[14] Fu, J.; Zhou, G. Y.; Hou, Z. Y.; Tian, H. C.; Xia, C. M.; Zhang, W.; Liu, J. T.; Wu, J. L.; Zhao, J. D.; Cang, X. L. Opt. Laser Technol. 2017, 91, 22.
[15] Liu, M. L.; Wu, Q.; Shi, H. F.; An, Z. F.; Huang, W. Acta Chim. Sinica 2018, 76, 246. (刘明丽, 吴琪, 史慧芳, 安众福, 黄维, 化学学报, 2018, 76, 246.)
[16] Cardenal, A. D.; Park, H. J.; Chalker, C. J.; Ortiz, K. G.; Powers, D. C. Chem. Commun. 2017, 53, 7377.
[17] Meng, S. Y.; Wang, M. M.; Lu, B. L.; Xue, Q. J.; Yang, Z. W. Acta Chim. Sinica 2019, 77, 1184. (孟双艳, 王明明, 吕柏霖, 薛群基, 杨志旺, 化学学报, 2019, 77, 1184.)
[18] Wu, Z. M.; Shi, Y.; Li, C. Y.; Niu, D. Y.; Chu, Q.; Xiong, W.; Li, X. Y. Acta Chim. Sinica 2019, 77, 758. (武卓敏, 石勇, 李春艳, 牛丹阳, 楚奇, 熊巍, 李新勇, 化学学报, 2019, 77, 758.)
[19] Liu, R. X.; He, X. Y.; Niu, L. T.; Lv, B. L.; Yu, F.; Zhang, Z.; Yang, Z. W. Acta Chim. Sinica 2019, 77, 653. (刘茹雪, 何小燕, 牛力同, 吕柏霖, 余菲, 张哲, 杨志旺, 化学学报, 2019, 77, 653.)
[20] Cao, L. Y.; Wang, T. T.; Wang, C. Chin. J. Chem. 2018, 36,
[21] Zou, Z.; Li, S. Q.; He, D. G.; He, X. X.; Wang, K. M.; Li, L. L.; Yang, X.; Li, H. F. J. Mater. Chem. B 2017, 5, 2126.
[22] Couck, S.; Van Assche, T. R.; Liu, Y. Y.; Baron, G. V.; Van Der Voort, P.; Denayer, J. F. Langmuir 2015, 31, 5063.
[23] Ponraj, Y. K.; Borah, B. J. Mol. Graph. Model. 2020, 97, 107574.
[24] Tang, Y. N.; Wang, S.; Zhou, X.; Wu, Y.; Xian, S. K.; Li, Z. Chem. Eng. Sci. 2020, 213, 115355.
[25] Fan, W. D.; Wang, X.; Zhang, X. R.; Liu, X. P.; Wang, Y. T.; Kang, Z. X.; Dai, F. N.; Xu, B.; Wang, R. M.; Sun, D. F. ACS Central. Sci. 2019, 5, 1261.
[26] Chen, Y. W.; Qiao, Z. W.; Lv, D. F.; Wu, H. X.; Shi, R. F.; Xia, Q. B.; Wang, H. H.; Zhou, J.; Li, Z. Ind. Eng. Chem. Res. 2017, 56, 4488.
[27] Guo, W. J.; Yu, J.; Dai, Z.; Hou, W. Z. Acta Chim. Sinica 2019, 77, 1203. (郭文娟, 于洁, 代昭, 侯伟钊, 化学学报, 2019, 77, 1203.)
[28] Wang, X.; Zhang, Y.; Chang, Z.; Huang, H.; Liu, X. T.; Xu, J. L.; Bu, X. H. Chin. J. Chem. 2019, 37, 871.
[29] Qiao, W. Z.; Song, T. Q.; Zhao, B. Chin. J. Chem. 2019, 37, 474.
[30] Chen, Z. Y.; Liu, J. W.; Cui, H.; Zhang, L.; Su, C. Y. Acta Chim. Sinica 2019, 77, 242. (陈之尧, 刘捷威, 崔浩, 张利, 苏成勇, 化学学报, 2019, 77, 242.)
[31] Zeng, J. Y.; Wang, X. S.; Zhang, X. Z.; Zhuo, R. X. Acta Chim. Sinica 2019, 77, 1156. (曾锦跃, 王小双, 张先正, 卓仁禧, 化学学报, 2019, 77, 1156.)
[32] Liu, Z. L.; Li, W.; Liu, H.; Zhuang, X. D.; Li, S. Acta Chim. Sinica 2019, 77, 323. (刘治鲁, 李炜, 刘昊, 庄旭东, 李松, 化学学报, 2019, 77, 323.)
[33] Bian, L.; Li, W.; Wei, Z. Z.; Liu, X. W.; Li, S. Acta Chim. Sinica 2018, 76, 303. (卞磊, 李炜, 魏振振, 刘晓威, 李松, 化学学报, 2018, 76, 303.)
[34] Lan, Y. S.; Han, X. H.; Tong, M. M.; Huang, H. L.; Yang, Q. Y.; Liu, D. H.; Zhao, X.; Zhong, C. L. Nat. Commun. 2018, 9, 5274.
[35] Qiao, Z. W.; Xu, Q. S.; Jiang, J. W. J. Mater. Chem. A 2018, 6, 18898.
[36] Wu, X. J.; Zheng, J.; Li, J.; Cai, W. Q. Acta Phys.-Chim. Sin. 2013, 29, 2207. (吴选军, 郑佶, 李江, 蔡卫权, 物理化学学报, 2013, 29, 2207.)
[37] Li, W.; Xia, X. X.; Cao, M.; Li, S. J. Mater. Chem. A 2019, 7, 7470.
[38] Shi, Z. N.; Yang, W. Y.; Deng, X. M.; Cai, C. Z.; Yan, Y. L.; Liang, H.; Liu, Z. L.; Qiao, Z. W. Mol. Syst. Des. Eng. 2020, DOI:10.1039/d0me00005a.
[39] Moghadam, P. Z.; Rogge, S. M. J.; Li, A.; Chow, C.-M.; Wieme, J.; Moharrami, N.; Aragones-Anglada, M.; Conduit, G.; Gomez-Gualdron, D. A.; Van Speybroeck, V.; Fairen-Jimenez, D. Matter 2019, 1, 219.
[40] Fernandez, M.; Woo, T. K.; Wilmer, C. E.; Snurr, R. Q. J. Phys. Chem. C 2013, 117, 7681.
[41] Shah, M. S.; Tsapatsis, M.; Siepmann, J. I. Angew. Chem. 2016, 128, 6042.
[42] Breiman, L. I.; Friedman, J. H.; Olshen, R. A.; Stone, C. J. Encycl. Ecol. 1984, 40, 358.
[43] Breiman, L. Mach. Learn. 2001, 45, 5.
[44] Raccuglia, P.; Elbert, K. C.; Adler, P. D. F.; Falk, C.; Wenny, M. B.; Mollo, A.; Zeller, M.; Friedler, S. A.; Schrier, J.; Norquist, A. J. Nature 2016, 533, 73.
[45] Zhang, W. G.; Goh, A. T. C. Geosci. Front. 2014, 7, 45.
[46] Wu, X. J.; Xiang, S. C.; Su, J. Q.; Cai, W. Q. J. Phys. Chem. C 2019, 123, 8550.
[47] Wang, X.; Zhang, X. R.; Zhang, K.; Wang, X. K.; Wang, Y. T.; Fan, W. D.; Dai, F. N. Inorg. Chem. Front. 2019, 6, 1152.
[48] Llewellyn, P. L.; Horcajada, P.; Maurin, G.; Devic, T.; Rosenbach, N.; Bourrelly, S.; Serre, C.; Vincent, D.; Loera-Serna, S.; Filinchuk, Y.; Férey, G. J. Am. Chem. Soc. 2009, 131, 13002.
[49] Wilmer, C. E.; Farha, O. K.; Yildirim, T.; Eryazici, I.; Krunglevi-ciute, V.; Sarjeant, A. A.; Snurr, R. Q.; Hupp, J. T. Energy Environ. Sci. 2013, 6, 1158.
[50] Wilmer, C. E.; Leaf, M.; Lee, C. Y.; Farha, O. K.; Hauser, B. G.; Hupp, J. T.; Snurr, R. Q. Nat. Chem. 2012, 4, 83.
[51] Rappé, A. K.; Casewit, C. J.; Colwell, K. S.; III, W. A. G.; Skiff, W. M. J. Am. Chem. Soc. 1992, 114, 10024.
[52] Martin, G. M.; Siepmann, J. I. J. Phys. Chem. B 1998, 102, 2569.
[53] Horn, H. W.; Swope, W. C.; Pitera, J. W.; Madura, J. D.; Head-Gordon, T. J. Chem. Phys. 2004, 120, 9665.
[54] Kadantsev, E. S.; Boyd, P. G.; Daff, T. D.; Woo, T. K. J. Phys. Chem. Lett. 2013, 4, 3056.
[55] Willems, T. F.; Rycroft, C. H.; Kazi, M.; Meza, J. C.; Haranczyk, M. Microporous Mesoporous Mater. 2012, 149, 134.
[56] Dubbeldam, D.; Calero, S.; Ellis, D. E.; Snurr, R. Q. Mol. Simul. 2015, 42, 81.
[57] Moghadam, P. Z.; Fairen-Jimenez, D.; Snurr, R. Q. J. Mater. Chem. A 2016, 4, 529.
[58] Ewald, P. P. Ann. Phys. 2006, 369, 253.
Options
文章导航

/

〈 〉