In pharmaceutical freeze-drying processes, the freezing step is a key-step because it fixes the morphology of the frozen material and, by the way, the final morphology of the freeze-dried material. It is the most difficult step to control, as it concerns the nucleation processes and it is also a damaging step with respect to the biological activity loss of pharmaceutical proteins.In this paper,
... [Show full abstract] we have reported original data obtained with a BSA (bovine serum albumin) based formulation used to stabilize pharmaceutical proteins during their freeze-drying process. Morphological characteristics of the ice crystals just after the freezing step were determined for different freezing protocols.Mean diameters and size distributions of the ice crystals have been determined by a direct optical microscopy method in cold chamber for different freezing parameters: size and type of vials (moulded or tubing), height of filling, vial right sectional surface area, shelf loading temperature. It was observed that the ice crystals sizes distributions depended not only on the freezing rates (well known result) but also on the vial size and type and also on the filling height. It was observed that moulded vials or higher values of filling height reduced the morphology heterogeneities resulting from large supercooling degrees and gave more homogeneous ice crystal size distributions. This behaviour was explained by the decrease of the supercooling effects which generally led to larger and non-homogenous distributions.Moreover, a convenient annealing treatment could homogenize and increase notably the mean ice crystals sizes.Water vapour mass transfer resistances data during the sublimation step were also determined and correlated to the experimental values of the mean ice crystals diameters which affected directly the dried layer permeability. These data were interpreted by the mass transfer laws of molecular diffusion theory (Knudsen regime).