(169a) Online Kinetic Analysis of Reactions In a Microreactor System

Authors: 
Moore, J. S., Massachusetts Institute of Technology
Smith, C. D., Massachusetts Institute of Technology
Jensen, K. F., Massachusetts Institute of Technology
Heublein, N., Technical University of Munich


Multicomponent reactions (MCRs) are important to drug discovery by affording complex products in only a single step.  By linking two of these MCRs, a Petasis boronic acid-Mannich reaction and an Ugi reaction, six different components can be incorporated in a relatively short time.  The kinetics of each reaction and their side reactions, both in isolation and in series, were investigated in an automated silicon microreactor system with online UPLC analysis.  The use of silicon microreactors allowed for precise temperature control and rapid changes between setpoints due to the high thermal conductivity of silicon.  In addition, the small volume minimized the reagent cost and waste production from these studies.  Furthermore, the use of automation significantly increased reaction efficiency and allowed for minimal operator intervention.  Online analysis via UPLC allowed for quantification of a number of reaction components, including monitoring the formation of side products that were unknown prior to experimentation, which would not be possible with other some online techniques, such as IR.

These kinetic data were then used to perform a model-based optimization of these reactions in series.  The optimum from this model was then used as the initial condition for an experimental optimization to test the model prediction and adjust the optimal conditions if necessary.  This experimental optimization was performed with a conjugate gradient algorithm that determined the search direction using gradient information calculated from a factorial design of experiments around the initial condition.  A line search was then executed with an Armijo-type technique.  The efficiency of this method was then compared to an optimization performed with no a priori knowledge of the system kinetics using the same experimental optimization procedure.