ACS Sustainable Chemistry & Engineering · 2023 · 44 citations · 78 references
EngineeringChemical EngineeringChemistryCatalysisOptimizationReaction EngineeringProcess OptimizationProcess SynthesisReaction Process DesignBayesian OptimizationCatalyst OptimizationCatalytic Process OptimizationCopper-catalyzed CouplingReaction Condition OptimizationReaction Condition ScreeningReaction Process OptimizationSustainable Strategy
Bayesian optimization is a powerful machine learning technique that is particularly well-suited for optimizing chemical reactions in the early stages of process development. It can efficiently explore vast reaction spaces and predict high-yielding reaction conditions by evaluating only a small number of experiments. In this report, we investigated the potential of Bayesian optimization as a tool to enhance the sustainability of chemical synthesis. Specifically, we focused on a real-world early-stage process development example: the C–N coupling of sterically encumbered bromo-pyrazines with amines. Our objective was to identify sustainable reaction conditions that utilize Earth-abundant copper catalysts and non-hazardous solvents. We used Bayesian optimizers with various acquisition functions. We assessed their performance and identified key features affecting the optimization results. The optimized conditions enabled the synthesis of a range of sterically encumbered pyrazines and pyridines with moderate to excellent yields.
78
Asymptotically efficient adaptive allocation rules
Tze-Leung Lai, Herbert Robbins · Advances in Applied Mathematics · 1985 · 2.5K citations