Hydrodynamic effects on mixing and competitive reactions in laboratory reactors

Abstract Small glass laboratory reactors are commonly used to determine the kinetics of the individual reaction steps required to synthesize the active pharmaceutical ingredients contained in the typical drug dosage form. These reactors are typically assumed by the user to be perfectly well mixed, and have a homogeneous content at all times. In this work a factorial experimental design scheme was used to carry out semi-batch reaction experiments aimed at determining the reactors’ micromixing effectiveness as a function five independent variables, i.e., agitation speed, impeller type, number of baffles, feed location and liquid level. A parallel, competitive fast reaction scheme was used to quantify the mixing effects by experimentally determining the yield, X S of one of the reaction products. In addition, computational fluid dynamics (CFD) was used to predict the flow field in the reactor as a function of the independent variables, and relate it to the reaction yield. The main effects of each of the independent variables on X S , as well as the two-way interactions between the variables, were obtained. Impeller speed, impeller type, and, under specific conditions, the number of baffles were found to affect significantly the mixing effectiveness. The local value of the energy dissipation rate at the feed point was calculated with CFD and found to correlate with X S . Additional experiments were conducted to investigate the effect of scale-up in laboratory units. The results indicate that mixing effectiveness decreases appreciably when the laboratory reactor scale changes from 2000 to 4000 mL. These results could be of significant importance in the evaluation of laboratory data for the organic synthesis of pharmaceuticals.