Autonomous Optimization of Discrete Reaction Parameters: Mono-Functionalization of a Bifunctional Substrate via a Suzuki-Miyaura Reaction

The autonomous optimization of discrete reaction parameters in organic synthesis remains unexplored, offering the potential to revolutionize synthetic methodology and expand the scope and efficiency of reaction development. Here, we introduce a fully integrated system that combines an automated synthesis robot, supercritical fluid chromatography, and Bayesian optimization to achieve the selective mono-functionalization of a bifunctional substrate via a Suzuki–Miyaura reaction. By leveraging variations in the ligand, base, and solvent, our approach enables precise tuning of the reaction parameters to achieve the targeted selectivity. Preliminary trials spanning 68 conditions identified eight critical descriptors, providing a framework for the systematic parameter characterization. This framework supported autonomous experimentation across 192 reaction conditions, comprising the initial four conditions and 47 iterative cycles, which ultimately enhanced the product yield to 49%. The generated mono-functionalized products serve as promising building blocks for organic photoelectronic applications, highlighting the far-reaching impact of autonomous, data-driven methodologies in synthetic and materials chemistry.