Transferring the Lyophilization Process from One Freeze-Dryer to Another
Transferring the Lyophilization Process from One Freeze-Dryer to Another
Dr. Thomas A. Jennings, Phase-Technologies
Formulation: In order to simplify the discussion it will be assumed that the formulation is reproducible from batch to batch [1] unless this is true, then validation of the lyophilization process would be extremely difficult and it is very doubtful that the process could be optimized. So in order to avoid such a lengthy discussion, I will just assume that the formulation is reproducible from batch to batch with respect to the quantities of components and constituents.
But for a formulation to be truly reproducible it must not only be reproducible with regards to its components and constituents [2], but also with regards to the thermal properties [1]. If the transfer of the process occurs at the same facilities then one can assume that the nature of the formulation will not be a factor in the transfer of the process. However, if the process is being transferred to a dryer in another facility, it may be safe to assume that the components and constituents of the formulation will be the same but there could be a difference in the thermal properties that could impact not only the cosmetic properties of the cake but even the potency of the dried product [1].
One problem could arise from a difference in the degree of supercooling [2] of the process water. The problem arises from the effect of the supercooling of the water in the formulation and its effect on the lyophilization process and quality of the final product which has been previously discussed elsewhere [3]. For example., assume that the intent is to transfer the process from facility A to a facility B. In facility A, the mean degree of supercooling of the formulation was found to be 12 oC while in facility B the formulation had a degree of supercooling of only 3 0C when tested in the same lot of glass vials. By using vials from the same lot, one can assume that there is no change in the frequency distribution of the heat transfer coefficient (Co ) [3,4].
Complete paper can be viewed online.
Votes:32