RESUMO
This paper presents a new Process Analytical Technology based on the use of an infrared camera and a mathematical model to estimate the ice crystal size distribution obtained at the end of the freezing stage of a vial freeze-drying process. Both empirical laws and first-principle based equations, already presented in the Literature, may be used to this purpose, if the temperature gradient in the frozen product and the freezing front rate are obtained from the analysis of the thermal images. The resistance of the dried product to vapor flux may be then calculated from the distribution of the ice crystal diameters, thus enabling the use of a one-dimensional model for process simulation and optimization. Freeze-drying of 5% and 10% w/w aqueous sucrose solutions, and of 5% w/w aqueous mannitol solutions, were considered as case study. The results were validated comparing the calculated diameters of the pores of the dried cake, corresponding to the ice crystal diameters, with the results experimentally obtained from the analysis of the SEM images, and comparing the values of drying duration and maximum product temperature calculated with the mathematical model with those measured experimentally. Results evidences the effectiveness of the proposed system for process monitoring.