Computer-assisted design of asymmetric PNP ligands for ethylene tri-/tetramerization: A combined DFT and artificial neural network approach
Haonan Fan,Xiaodie Yang,Jing Ma,Biaobiao Hao,Fakhre Alam,Xumeng Huang,Aixi Wang,Tao Jiang
DOI: https://doi.org/10.1016/j.jcat.2023.01.011
IF: 7.3
2023-02-01
Journal of Catalysis
Abstract:The combination of computational and experimental sciences accelerates the design and development of molecular catalysts. A general strategy for developing ethylene oligomerization catalysts is still lacking. Consequently, herein, we proposed a widely applicable strategy for designing ethylene oligomerization catalysts. We combined density functional theory (DFT) and an artificial neural network (ANN) to establish a relation between catalyst structure and performance. The structure optimization and electronic calculation of a series of asymmetric PNP/Cr active species were conducted using DFT, and the steric and electronic descriptors were extracted to establish datasets. The catalyst prediction model was constructed using ANN and the leave-one-out cross-validation (LOOCV) method was used to verify the generalization ability of the models. The optimized ANN-based models used to predict 1-hexene and 1-octene selectivity exhibited high R2 values, which indicates satisfactory prediction accuracy of the models. We designed new PNP ligands and successfully predicted the ethylene oligomerization performance of PNP/Cr precatalysts using ANN-based models, which were verified through experiments. In addition, we found that the steric properties more significantly affect the performance of precatalysts than the electronic properties.
chemistry, physical,engineering, chemical