The kinetics of lysozyme crystallization under seeded isothermal batch conditions was followed by measurement of the decline in solution concentration versus time. Kinetics were measured for five different values of the seed crystal mass. The data were analyzed using a recently proposed mathematical model. For each seed mass, the model fit the kinetic data well. Growth rate constants determined using the model were approximately constant over a sixfold increase in the seed crystal mass, and fell well within the range of values reported in the literature, but obtained using entirely different experimental techniques. These results confirmed the utility of the proposed model. The proposed model can be used to analyze crystallization kinetics using absorbance measurements only, without the need to characterize the crystal size, thus avoiding the need for expensive laser light scattering and digital microscopy instrumentation. Thus, the model offers a low-cost straightforward method to analyze and simulate the effects of changes in operating parameters such as the seed crystal mass, solution volume, initial protein concentration, pH, temperature, salt concentration, and time.

Bulk protein crystallization, unlike small molecule crystallization, has found very limited use in biopharmaceutical manufacture. Most work in this area targets obtaining single large crystals for molecular structure determination by crystallography. Design and optimization of bulk crystallization for protein recovery and purification is much less common, and requires a mathematical model for analysis of laboratory data suitable for scale-up purposes. Traditionally, the crystal size distribution and method of moments is used to characterize the crystallization process. A simpler method is presented in this paper that utilizes the desupersaturation curve. The method uses an approach that does not require expensive instrumentation or characterization of the seed crystal size distribution. The method is extended to allow determination of both the mass deposition rate constant and the growth rate order from a single desuperaturation curve. Experimental data for the bulk crystallization of ovalbumin are used to validate the method. The rate constants and rate order obtained using the new method compare well with literature values. Scale-up is illustrated by prediction of the impact of changes in seed mass on protein crystallization. This new method offers a straightforward and low-cost alternative to traditional methods for the analysis and scale-up of protein crystallization data.