高性能COF材料的高通量筛选策略:己烷异构体分离

High-throughput computational screening strategy for high-performance COF materials:separation of hexane isomers

己烷异构体中双支链异构体的分离可以提高汽油的辛烷值从而减少发动机的爆震现象。针对传统的蒸馏方法耗能高和新型吸附剂金属有机框架成本高、工作能力低、稳定性差的缺点,采用高通量计算筛选方法研究了688种共价有机框架(COFs)对己烷异构体的分离性能。首先计算了所有COF的几何结构描述符,通过限制孔径(PLD)6.2~15A的范围筛选出209个可容纳所有己烷异构体的COF,再利用巨正则Monte Carlo(GCMC)方法模拟433 K下上述COF对己烷异构体的吸附解吸过程。对再生能力R>80%且吸附性能分值(APS)最高的COF进行排序,筛选出具有最高APS值的COF-DL2292-fold,APS值为23.36 mol/kg,R为99.38%。分析了6个几何结构描述符与APS的相关性,发现对于COF来说较高的孔隙率(VF)、较高的孔隙体积(PV)、较低的密度(ρ)可提高COF的APS值。最后基于PV、VF、ρ利用决策树算法设计出高APS值COF的筛选路径,研究工作对今后设计用于己烷异构体吸附分离的COF具有一定的指导意义。

Separation of double-branched isomers from hexane isomers can increase the octane number of gasoline and reduce engine knocking.Aiming at the disadvantages of high energy consumption of traditional distillation methods and the disadvantages of high cost,low working capacity and poor stability of new MOF adsorbents,the separation performance of 688 covalent organic frameworks(COFs)for hexane isomers was investigated by high-throughput computational screening method.Firstly,the geometric properties of all COFs were calculated,and 209 COFs that can accommodate all hexane isomers were selected through the range of 6.2—15Åof pore limiting diameter(PLD).Then the adsorption and desorption processes of hexane isomers by the above COFs at 433 K were simulated by grand canonical Monte Carlo(GCMC)method.The COFs with regeneration capacity(R)>80%and the highest adsorbent performance score(APS)were sorted,and the COF-DL2292-fold with the highest APS value was selected.Its APS value is 23.36 mol/kg and R is 99.38%.The correlation between six geometric properties and APS was analyzed.It was found that the APS value of COF can be improved by higher void fraction(VF),higher pore volume(PV)and lower density(ρ).Finally,based on PV,VF andρ,the screening path of high APS value COF is designed by using decision tree algorithm,which has certain guiding significance for the design of COF for adsorption and separation of hexane isomers in the future.