Integrating computational chemistry for the design of an automated nitroso species fluidic module learning

The unique nitrogen-oxygen combination located next to a specific carbonated backbone of C-nitroso derivatives is responsible for a complex reactivity that renders them not only widely effective in various processes, such as in nitroso Diels-Alder reactions, but also difficult to demystify [1]. Within this context, an automated interdisciplinary nitroso species fluidic module learning from nitrosoaldol reactions has been implemented (Figure 1). On the one hand, continuous flow technology offers accurate control of paramount local parameters such as stoichiometry and temperature as well as robustness through automation [2]. On the other hand, computational chemistry provides insights on intrinsic properties, hence saving time and resource-consuming synthetic work [3]. The module learning relies on an upstream computational study that extracts theoretical data determining the pertinence of the reaction into fluidic conditions. Experimental kinetics data are then determined in flow at different temperatures and used for dynamic feedback.