(281a) Automated Optimization, Kinetic Modeling, and Scale-up of Flow Chemistry Processes

Moore, J. S., Massachusetts Institute of Technology
McMullen, J. P., Massachusetts Institute of Technology

Different approaches to automated optimization of microreactor systems are presented starting with “black box” methods requiring little knowledge of the reaction system and moving on to statistical approaches to determine intrinsic reaction kinetics. The resulting information can then be combined with fluid flow and heat transfer models to predict optimum performance of scaled up flow reactor systems. As an example, we present an automated, silicon microreactor system utilizing a sequential experimentation framework driven by model-based optimization feedback for online reaction rate parameter determination. Additional techniques for using flow systems to determine kinetics by batch-like time courses are also described along with in-line micro-calorimetry for measuring heats of reaction. The benefits of quickly obtaining kinetic information with an automated microreactor system are further demonstrated by successfully scaling of production by orders of magnitude. The scale-up procedure circumvents the significant fluid distribution problems integral to scale-out of microreactors by allowing increases in reactor geometry as far as possible without losing heat transfer advantages inherent in flow systems. Further increases then have to be realized by scaling out with the now smaller number of units, which mitigates the fluid distribution problems.