The kinetics of freeze-drying and temperature changes in the product core have been established for milk and water frozen in vials. The influence of heating plate temperature, working pressure and initial water content on the sublimation kinetics have been studied. The operating pressure is paramount since it causes dehydration speeds to vary by a factor of 1–2. Through the use of a simple model of the ‘uniformly retreating ice front’ type, it has been possible to identify the transfer parameters of the model. Correlations with the operating pressure have been established. The results show a working pressure optimum at 200 Pa.

Frozen suspensions of Streptococcus thermophilus were freeze-dried in a vacuum or a fluidized adsorbent bed at atmospheric pressure. Optimum operating conditions for each process were defined. For the duration of processing and survival rate of bacteria, in each case vacuum freeze-drying seemed more satisfactory than atmospheric pressure freeze-drying. The use of reconstituted skimmed milk as a suspension medium provides good protection for S. thermophilus.

Freeze-drying experiments were carried out for milk and water in a vial, and the influence of the heating plate temperature was examined. A simple simulation based on a 'uniformly retreating ice front' type model was performed. A numerical model enables the changes in both water content and temperature of these products to be described satisfactorily during the experiment. Three transfer parameters (water vapour diffusivity in the dry layer, the external mass transfer coefficient, and the contact resistance to the heat transfer) were determined. The contact resistance between the product and the heating plate is the most important resistance to heat transfer, even controlling the dehydration kinetics.