Self-optimizing Bayesian for continuous flow synthesis process

The integration of artificial intelligence (AI) and chemistry has propelled the advancement of continuous flow synthesis, with program-controlled automatic process optimization being facilitated. Optimization algorithms play a pivotal role in the automated optimization process. The costs associated with optimization processes are expected to be further mitigated through increased accuracy and predictive capability of the algorithms. A self-optimizing Bayesian algorithm (SOBayesian), incorporating Gaussian process regression as a proxy model, has been devised. Adaptive strategies are implemented during the model training process, rather than on the acquisition function, to enhance the modeling efficacy. The continuous flow synthesis process of pyridinylbenzamide, an important pharmaceutical intermediate, was optimized via the Buchwald–Hartwig reaction using this algorithm. A yield of 79.1% was achieved in under 30 rounds of iterative optimization, and subsequent optimization with reduced prior data resulted in a 27.6% reduction in the number of experiments, significantly lowering experimental costs. Based on the experimental results, it was concluded that the reaction is kinetically controlled. New research ideas and strategies for optimizing similar reactions in continuous flow automated optimization are thus provided.

• Liu, R
• Wang, Z
• Yang, W
• Cao, J
• Tao, S

• School of Chemistry, State Key Laboratory of Fine Chemicals, Frontier Science Center for Smart Materials, Dalian Key Laboratory of Intelligent Chemistry, Dalian University of Technology, Dalian, 116024, China

• Request a Brochure
• Request more Info
• Request a Quote