Optimisation of olefin epoxidation catalysts with the application of high-throughput and genetic algorithms assisted by artificial neural networks (softcomputing techniques)

An olefin epoxidation Ti catalyst has been optimised by means of high-throughput experimentation involving materials synthesis, postsynthesis treatments, and catalytic testing. Softcomputing techniques for advanced experimental design have been used. The variables explored in the hydrothermal synthesis of Ti-silicate-based catalysts were: concentration of OH−, titanium, and surfactant. The probe reaction employed for the optimisation was the solvent-free epoxidation of cyclohexene, with tert-butylhydroperoxide as oxidant. The different catalyst groups detected by clustering analysis were studied by XRD and UV diffuse reflectance. Ti-mesoporous MCM-41 and MCM-48 molecular sieves were among the most active catalysts. The best performing catalysts were tested for epoxidation of different linear olefins.