暨南大学化学学院李丹研究小组提出了MOF-5类似物化学空间的机器学习辅助探索增强C₂H₆/C₂H₄分离。2025年3月11日出版的《德国应用化学》杂志发表了这项成果。

在此，该课题组提出了一种机器学习辅助的分子模拟策略来探索pcu-MOFs与MOF-5等垂直的C₂H₆/C₂H₄分离能力。极限梯度增强（XGBoost）算法对C₂H₆/C₂H₄选择性和C₂H₆摄取具有较高的预测精度，其中亨利系数比（S0）和C₂H₆的亨利系数（K(C₂H₆)）是预测C₂H₆的关键因素。该课题组进一步合成了性能最好的MOF，命名为A-66，并通过实验验证了它具有较大的C₂H₆吸附量和优异的C₂H₆/C₂H₄分离性能。这项工作为探索MOF-5类似物的化学空间和从C₂H₆/C₂H₄混合物中高效纯化C₂H₄提供了有价值的策略。

研究人员表示，采用C₂H₆选择性吸附剂进行吸附分离，可直接产生高纯度的C₂H₄，是一种具有替代低温精馏潜力的节能分离方法。虽然已经报道了许多C₂H₆选择性MOFs，但开发具有大C₂H₆吸附容量和高C₂H₆/C₂H₄选择性的MOFs仍然具有挑战性。

附：英文原文

Title: Machine Learning-Assisted Exploration of Chemical Space of MOF-5 Analogues for Enhanced C2H6/C2H4 Separation

Author: Ying Wang, Zhi-Jie Jiang, Weigang Lu, Dan Li

Issue&Volume: 2025-03-11

Abstract: Adsorptive separation using C2H6-selective adsorbents can produce high-purity C2H4 directly, making it an energy-efficient separation method with the potential to replace cryogenic distillation. While many C2H6-selective MOFs have been reported, developing MOFs with both large C2H6 adsorption capacity and high C2H6/C2H4 selectivity remains challenging. Herein, we present a machine learning-assisted molecular simulation strategy to explore the C2H6/C2H4 separation capability of pcu-MOFs isoreticular to MOF-5. The eXtreme Gradient Boosting (XGBoost) algorithm showed high accuracy in predicting the C2H6/C2H4 selectivity and C2H6 uptake, where Henry coefficient ratio (S0) and Henry coefficient of C2H6 (K(C2H6)) were identified as key factors. We further synthesized the top-performing MOF termed A-66 and experimentally verified its large C2H6 adsorption capacity and excellent C2H6/C2H4 separation performance. This work provides a valuable strategy for exploring the chemical space of MOF-5 analogues and identifying promising candidates for the efficient purification of C2H4 from C2H6/C2H4 mixtures.

DOI: 10.1002/anie.202500783

Source: https://onlinelibrary.wiley.com/doi/10.1002/anie.202500783