(451d) Modelling the Effect of Process Variables On the Polymorphic Transformation of L-Glutamic Acid

This work concerns the modelling of the Solution-Mediated Polymorphic Transformation (SMPT) of L-glutamic acid in aqueous solutions. L-glutamic acid exhibits two polymorphs, the metastable alpha form and the stable beta form. For downstream processing the metastable α form is preferred due to its block-like habit, instead of the β form, which forms needles with a high aspect ratio. The modelling of the SMPT of L-glutamic was conducted using Process Systems Enterprise’s (PSE) crystallization modelling software, namely gCRYSTAL. This software provides the capability for multiphase population balance modelling in a user-friendly GUI-based environment. Numerous process variables can affect the rate of the SMPT. In the case of L-glutamic acid, the effects of seed mass, seed PSD, seed purity and agitation rate on the SMPT will be examined in this work. The model utilised the experimental data of Cornel et al.[1] to evaluate the crystallization kinetics for both polymorphs. This work involved the seeded isothermal transformation experiments from the α form to the β form in aqueous solutions, saturated with respect to the α form. Therefore, the α form could only undergo dissolution, while the β form was found to exhibit secondary nucleation and crystal growth. Using experimental data for the evolution of the solution concentration and the polymorphic fraction, eight parameters including the level of β impurity and size in the α seed crystals, were employed to fit the experimental data.

The developed model of the SMPT process was able to accurately describe the effects of seed mass, PSD, purity and agitation rate on the transformation rate. The inclusion of parameters to describe the impurity of the alpha seeds employed in the experiments was validated by two blind tests, not utilised in the estimation of kinetic parameters, where a known amount of α and β seeds were added initially. The model was able to accurately describe the effect of the presence of the β seed impurity in these cases, which caused a large reduction in the time necessary for the SMPT process. The time necessary for the transformation process was found to reduce with increasing alpha seed mass, agitation rate and impurity concentration. These observations led to the conclusion that the rate limiting step of the SMPT process was the rate of secondary nucleation on the surface of the β crystals, which increased with agitation and available seed surface area. A contradictory finding for the effect of seed size was attributed to an increased impurity concentration of the β crystals in the larger α seed sizes.