This paper aims to demonstrate the application of the generalized Tendency Kinetic Modeling to the systematic modeling and optimization of Asymmetric Catalytic Hydrogenation Reactions. The developed methodology achieves the rapid identification of all plausible model structures along with the identification of their respective parameters, based on statistically informative experimental data. The informative data are acquired by online monitoring of the reactions. Experimental design methods (DoE) are used to plan the necessary experiments.

The industrial example of a catalytic asymmetric hydrogenation reaction is the system on focus, which is very important and frequently encountered in the production of pharmaceuticals. The performance of such process is sensitive to the selection of the proper conditions that favor the maximum yield to the desired product. The asymmetric catalytic hydrogenation of a pharmaceutical compound is the specific experimental system selected. This process will be optimized by directing the selectivity of the asymmetric reaction towards the desired product and maximizing the conversion of the reactant. This is achieved by selecting the optimal reaction conditions, such as temperature, pressure, catalyst loading and initial reactant concentration. The experiments are performed both in the ChemSCAN, a small scale reactor system with 8 parallel reactors of 20mL each, and in a larger scale reactor vessel of 1000mL controlled by the Büchi gas press-flow controller (bpc). The reaction is monitored using Raman spectroscopy. Appropriate chemometric models are developed and they are used on-line to estimate the reactant and product concentrations versus time. The experimental work for this project is performed at the laboratories of Sepracor, Inc.