Spray freeze-drying (SFD), which includes spray-freezing into droplets and dynamic vacuum drying, presents a promising alternative approach to manufacture dried pharmaceuticals more efficiently than conventional vial freeze-drying. Without reliable predictive models for the SFD conditions of interest, any respective process development still relies on empirical approaches. In this work, we propose an improved modeling framework to describe the fast freezing (<1 s) that sub-millimeter droplets undergo in the present SFD process. The modeled freezing rate accounts for both the kinetics of ice growth and droplet heat transfer mechanisms. Computational fluid dynamics (CFD) simulations and experiments on bulk spray-freezing are combined to refine and validate the proposed reduced-order model. While this study is limited to water-sucrose solutions, the present modeling approach can be extended to other pharmaceutical excipients. For the cooling rates of interest, model results indicate that droplets with initial sucrose concentration higher than 20% w/w will transit to a glassy state before completion of crystallization and, consequently, devitrification is expected during post spray-freezing manipulation of the bulk material. In practice, such compact model does not only allow quantification of process parameters that cannot be measured in real time but also enable the choice of optimal spraying conditions for production of free-flowing, high-quality frozen droplets that meet the target product profile.
Crystallization, Freeze-drying, In silico modeling, Lyophilization, Mechanistic modeling, Microsphere(s), Powder technology(s), Spray freeze-drying, Sucrose