Multidimensional dynamic experiments for data-rich process development of reactions in flow

Continuous flow processing continues to garner interest in the pharmaceutical industry for the production of small-molecule drug substances given the advantages associated with improved mass and heat transfer, more facile scale-up, and smaller process footprints. However, to date, in-depth understanding and development of reactions can be hindered by limited availability of key raw materials and short timelines during the early stages of drug development. In this work, we present the benefits of nonlinear 2-dimensional dynamic experiments for characterization of flow reaction design spaces in a resource-sparing manner, expanding on recent reports around the use of transient flow experiments for reaction characterization. Experimental reaction data for a Knoevenagel condensation reaction are collected from a system operated at either steady state or under dynamic conditions varying one or two reaction parameters simultaneously in a linear or nonlinear pattern. These experimental data are used to generate a mathematical model describing the reaction across the design space tested. It was observed that a single nonlinear 2-dimensional dynamic experiment sufficiently spans conditions in the design space enabling the development of a robust model, which can drastically improve process knowledge and risk mitigation for early- or late-stage process development of pharmaceutical drug substances. The use of multidimensional dynamic flow experiments for reaction profiling and generation of an empirical surface response model for a Knoevenagel condensation reaction is described.