journal article·2025-04Open access

Interpretable Machine‐Learning and Big Data Mining to Predict the CO2 Separation in Polymer‐MOF Mixed Matrix Membranes

Hao Wan, Yue Fang, Min Hu, Shuya Guo, Zhiqiang Sui, Xiaoshan Huang, Zili Liu, Yue Zhao, Hong Liang, Yufang Wu, Hanyu Gao, Zhiwei Qiao

Advanced Science· Volume 12· Issue 16

Abstract

Abstract Mixed matrix membranes (MMMs) are renowned for their exceptional gas separation capabilities. In this work, high‐throughput computing screening and machine learning are employed to evaluate the CO2 separation performance of 54117 MMMs composed of 9 polymers and 6013 metal–organic frameworks (MOFs). The structure‐property relationships of MMMs are analyzed for 4 binary mixtures (CO2/X, X = CH4, N2, H2, O2), and the best‐performing combinations of MOFs and polymers are found, with which the MMM performance exceeded the Robeson's upper limit. Then, a stacked ensemble regression model with high accuracy (average R2 = 0.96) is trained, demonstrating excellent extrapolation capability (R2 = 0.95) for new MMMs containing 6FDA‐DAM. Furthermore, by utilizing Shapley Additive Explanations and data segmentation, it is identified that the pore limit diameter and largest cavity diameter in MOF features and the fractional free volume and density in polymer features are of paramount importance. Two extrapolation methods are compared and found that transfer learning is better for predicting CO2 separation performance in MMMs and designing new materials with large datasets. Finally, an interactive desktop software is developed to assist researchers in rapidly and accurately calculating the CO2 separation performance of MMMs. This work presents a novel approach for the rapid evaluation of high‐quality MMMs and the efficient calculation of gas permeation rates within membranes.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Membrane Separation and Gas Transport · Metal-Organic Frameworks: Synthesis and Applications · Zeolite Catalysis and Synthesis

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Guangzhou Key Laboratory for New Energy and Green Catalysis School of Chemistry and Chemical Engineering Guangzhou University Guangzhou 510006 P. R. China

Department of Chemical and Biological Engineering Hong Kong University of Science and Technology Clear Water Bay Hong Kong 999077 P. R. China

State Key Laboratory of NBC Protection for Civilian Beijing 100191 P. R. China

Joint Institute of Guangzhou University & Institute of Corrosion Science and Technology Guangzhou University Guangzhou 510006 P. R. China

Funding

National Natural Science Foundation of China · Natural Science Foundation of Guangdong Province · China Postdoctoral Science Foundation

References (61)

Show source-provided references
1. Pachauri R. K. · 2014
2. 10.1016/j.egypro.2017.03.175210.1016/j.egypro.2017.03.1752 ↗
3. 10.1016/j.cjche.2017.07.00610.1016/j.cjche.2017.07.006 ↗
4. 10.1080/00908319093382510.1080/009083190933825 ↗
5. 10.1039/C5TA00124B10.1039/C5TA00124B ↗
6. 10.1002/ceat.20120073410.1002/ceat.201200734 ↗
7. 10.1039/C6EE00811A10.1039/C6EE00811A ↗
8. 10.1016/j.memsci.2008.04.03010.1016/j.memsci.2008.04.030 ↗
9. Fauzi A. · 2015
10. 10.1016/j.memsci.2016.02.02210.1016/j.memsci.2016.02.022 ↗
11. 10.1126/science.adi154510.1126/science.adi1545 ↗
12. 10.1007/s41980-021-00654-z10.1007/s41980-021-00654-z ↗
13. 10.1021/la800251t10.1021/la800251t ↗
14. 10.1016/S0376-7388(02)00429-510.1016/S0376-7388(02)00429-5 ↗
15. 10.1016/S1872-5805(21)60034-310.1016/S1872-5805(21)60034-3 ↗
16. 10.1039/c2ee21996d10.1039/c2ee21996d ↗
17. 10.1016/j.proeng.2016.06.49910.1016/j.proeng.2016.06.499 ↗
18. 10.1039/C8EE02582G10.1039/C8EE02582G ↗
19. 10.1021/acsami.2c0401610.1021/acsami.2c04016 ↗
20. 10.1016/S0376-7388(96)00303-110.1016/S0376-7388(96)00303-1 ↗
21. 10.1016/j.memsci.2020.11855510.1016/j.memsci.2020.118555 ↗
22. 10.1021/acssuschemeng.9b0102010.1021/acssuschemeng.9b01020 ↗
23. 10.1021/acs.jpcc.8b1064410.1021/acs.jpcc.8b10644 ↗
24. 10.1016/j.memsci.2019.11767510.1016/j.memsci.2019.117675 ↗
25. 10.1021/acsami.2c0897710.1021/acsami.2c08977 ↗
26. 10.1002/adts.20190010910.1002/adts.201900109 ↗
27. 10.1002/jcc.2578710.1002/jcc.25787 ↗
28. 10.1080/0026897001002004110.1080/00268970010020041 ↗
29. 10.1016/0010-4655(96)00016-110.1016/0010-4655(96)00016-1 ↗
30. 10.1080/08927022.2015.101008210.1080/08927022.2015.1010082 ↗
31. 10.1016/j.memsci.2005.01.00910.1016/j.memsci.2005.01.009 ↗
32. 10.1016/j.memsci.2013.01.00910.1016/j.memsci.2013.01.009 ↗
33. 10.1016/j.memsci.2003.08.03110.1016/j.memsci.2003.08.031 ↗
34. 10.1016/j.desal.2006.03.34710.1016/j.desal.2006.03.347 ↗
35. 10.1016/j.memsci.2017.04.02210.1016/j.memsci.2017.04.022 ↗
36. 10.1016/j.micromeso.2011.08.02010.1016/j.micromeso.2011.08.020 ↗
37. 10.1039/C6TA06262H10.1039/C6TA06262H ↗
38. B.Pavlyshenko presented at2018 IEEE Second Int. Conf. on Data Stream Mining & Processing (DSMP) Lviv Ukraine August2018.
39. 10.1007/s10346-009-0183-210.1007/s10346-009-0183-2 ↗
40. 10.1016/j.scs.2022.10367710.1016/j.scs.2022.103677 ↗
41. Lundberg S. M. · 2017
42. 10.1038/s42256-019-0138-910.1038/s42256-019-0138-9 ↗
43. 10.1021/jp201075u10.1021/jp201075u ↗
44. 10.1039/c2sc01097f10.1039/c2sc01097f ↗
45. 10.1021/ja210823910.1021/ja2108239 ↗
46. 10.3390/ma905035510.3390/ma9050355 ↗
47. 10.1016/j.memsci.2008.12.00610.1016/j.memsci.2008.12.006 ↗
48. 10.1016/S0376-7388(96)00303-110.1016/S0376-7388(96)00303-1 ↗
49. 10.1016/j.cej.2022.13678310.1016/j.cej.2022.136783 ↗
50. 10.1002/advs.20200219010.1002/advs.202002190 ↗
51. 10.1016/0376-7388(91)80060-J10.1016/0376-7388(91)80060-J ↗
52. 10.1021/acscentsci.2c0009910.1021/acscentsci.2c00099 ↗
53. 10.1016/j.seppur.2020.11719810.1016/j.seppur.2020.117198 ↗
54. 10.1016/j.memsci.2008.09.01010.1016/j.memsci.2008.09.010 ↗
55. Chen X. Y. · 2017
56. 10.1016/j.seppur.2018.05.01210.1016/j.seppur.2018.05.012 ↗
57. 10.1016/j.ces.2020.11549710.1016/j.ces.2020.115497 ↗
58. Alexopoulos E. C. · Hippokratia. · 2010 · 14 · 23
59. 10.1109/JPROC.2020.300455510.1109/JPROC.2020.3004555 ↗
60. 10.1021/acsami.0c0804010.1021/acsami.0c08040 ↗
61. 10.1246/bcsj.2021023710.1246/bcsj.20210237 ↗