Single Vial Monitoring of Pharmaceutical Freeze Drying Processes using Through Vial Impedance Spectroscopy

Currently, there is no single process analytical technology (PAT) in the market that can provide non-invasively, product and process related information which is representative of the entire batch at key stages of the freeze drying process: (i) the ice nucleation onset and the ice solidification endpoint, and the resulting ice crystal structure at the end of the freezing stage; and (ii) the true batch endpoint of ice sublimation to ensure all the ice is removed before switching to the secondary drying stage. Through Vial Impedance Spectroscopy (TVIS) employs a pair of electrodes that measure the electrical impedance of the glass vial and its contents. TVIS has been used as a PAT tool for the non-invasive monitoring of the critical events during the freeze drying process. While it has already been established previously that the TVIS parameters, 𝐹𝑃𝐸𝐴𝐾 (sensitive to product temperature) and 𝐶′(0.2𝑀𝐻𝑧) (largely sensitive to phase change) can be employed together to determine the ice nucleation onset and the ice solidification endpoint respectively, and 𝐶𝑃𝐸𝐴𝐾″ can be used for drying rate determination (largely sensitive to ice cylinder height), the present work involves the use of 𝐹𝑃𝐸𝐴𝐾 and 𝐶′(0.2𝑀𝐻𝑧) to demonstrate a direct dependency of ice nucleation temperature (or the freezing method) on the ice growth time, and subsequently the stability of the ice crystal microstructure determined using 𝐶𝑃𝐸𝐴𝐾″. Further, the present study establishes a high sensitivity of another TVIS parameter, 𝐶′(100 𝑘𝐻𝑧), towards the changes in the shape of the ice sublimation interface and subsequently the removal of the last vestiges of ice from a single vial during primary drying. This feature has been exploited for the prediction of the primary drying endpoint at any vial location by employing a mathematical methodology, which has been extended to a simple sugar solution, wherein visual imaging cannot be used to detect ice disappearance. Finally, TVIS was multiplexed with batch measurement sensors (comparative pressure measurement and pressure rise test) for a comparison between the batch endpoint and the TVIS endpoints predicted at the shelf edge and the core. It has been demonstrated none of the PAT tools can determine the true endpoint of the batch (though the TVIS method is a significant improvement over one that relies on the thermocouple response). In conclusion, the study highlights a significant potential of TVIS for delivering additional product and process related information when it is multiplexed with batch measurement tools non-invasively, such as the thermal history of the sample before primary drying and the primary drying endpoints at various locations on the shelf.