Multifactor Multilevel Optimization of MOF Derivatives for Promoting Catalyst Fabrication.

Abstract

Designing catalysts with high performance on the heterogeneous catalysis fields has puzzled many scientists due to the multifarious repeated experiments which takes most of their time. Herein, a multifactor and multilevel experimental design with an artificial neural network has been performed to optimize the catalyst synthesis tactic. During the process, five factors within one experiment are considered to establish a neural network model to pick out the optimal synthesis condition. Excitingly, the as-synthesized catalyst according to the above strategy displays superior catalytic activity to the other similar synthesis tactics. This work not only fabricates a catalyst with extremely catalytic performance but also provides new insights into constructing catalysts with special function efficiently.