Development of a High-Throughput Kinetics Protocol and Application to an Aza-Michael Reaction

High-throughput experimentation (HTE) has become integral to the pharmaceutical industry with most major pharmaceutical companies investing in automation and high-throughput screening technologies. Testing hundreds of reactions in parallel has distinct advantages; however, one clear disadvantage is that performing a reaction on micromolar scale is not always indicative of the reaction's performance on multikilogram scale. Additionally, a great deal of information is lost by looking at a single time point. Valuable data around intermediates, over-reaction, catalyst induction periods, and so forth are invisible to a typical HTE workflow, which involves analyzing reactions at a single time point (e.g., 18 h). We envisioned a workflow in which time courses for each well of a high-throughput screen were collected. With this change in strategy, it could then become possible to complete high-throughput screening, select reaction conditions, gather kinetic information, and successfully build a kinetic model in less than 1 week. A kinetic model consisting of scale-independent parameters allows for virtual reaction optimization where the input concentrations, catalyst loading, and temperature can all be simulated and adjusted to understand their impact on yield or quality in a matter of seconds. A case study is presented with a transition metal salt/TMSCl-catalyzed aza-Michael reaction to showcase the performance and robustness of the high-throughput kinetic platform. A reaction progress kinetic analysis approach is utilized to quickly screen the rates of 48 catalyst/solvent combinations and create a mechanistic model. The first-principles kinetic model provides support for a proposed mechanism of dual activation by TMSCl.

Automated summary

High-throughput experimentation is crucial in the pharmaceutical industry, but traditional workflows have limitations in capturing valuable data and providing accurate predictions. This study proposes a new workflow that collects time course data for each reaction, enabling faster screening, selection of reaction conditions, and building of kinetic models. The case study demonstrates the performance and robustness of this high-throughput kinetic platform.