The development of micro-sample electrical impedance analysis for freeze-drying process development

Background: Since primary drying in freeze drying cycle is the longest step, it is desired to use optimal process conditions for efficient drying by running a process under the critical temperatures (e.g., Tg’, Teu, Tc’). Multiple techniques and instruments (e.g., differential scanning calorimetry, dynamic mechanical analysis, Lyotherm, freeze-drying microscopy) are often used to measure the critical temperature during freeze-drying process development. But often, it is difficult to predict these critical temperatures as most data varies between the equipment and depends on the analyst's judgment and, as a result, a large volume of valuable products (i.e. proteins) are utilised in the process. Many studies showed that electrical impedance-based techniques are sensitive for studying formulation. There is a need to develop an impedance-based analytical strategy for freeze-drying process development to overcome these issues. Aim: The purpose of this thesis is (1) to develop new technologies for the freeze-drying process development using micro-scale sample volume using electrical impedance analysis, (2) remove operator's subjectivity and allow accurate determination of critical temperature and prediction of freezing/drying rate, (3) study dielectric relaxation of ice and sugar solutions using interdigitate electrode and evaluate using electrical circuit elements. Method: The initial stage of the project involved a comprehensive study of water/ice using the broadband frequency range (10 Hz to 1 MHz) on a broadband dielectric spectroscopy (NovoControl - Concept 40) using gold interdigitated electrodes on a glass substrate (Micrux – ED-IDE1-AU). The next step involved the selection of relevant frequencies suitable for freeze-drying application and using data fitting software to study the dielectric relaxation process. The later stage of the project focused on developing impedance-enabled freeze-drying microscopy (Z-FDM) to operate using frequency identified from the BDS study. In Z-FDM development, Lyostat5 freeze-drying microscope (Biopharma Process Systems Ltd) with an ISX3-min impedance analyser (ScioSpec) were integrated to analyse the sample using gold interdigitated electrodes (IDE) on a glass substrate (Micrux – ED-IDE1-AU). Mathematical modelling and MATLAB image analysis were used to study critical formulation temperature and sublimation kinetics. Result: Two impedance-based micro-sample analytical methods were developed for freeze-drying process development (BDS & Z-FDM). BDS feasibility study indicates the device's sensitivity with sample volume as low as 0.1 L. Furthermore, the dielectric response of water from an IDE is comparable with well-established through-vial impedance spectroscopy. Initial Z-FDM results demonstrate that the use of IDE does not affect the optical ability under the microscope and does not bear thermal mass. The data fitting method using RelaxIS software was established to interpret broadband spectrum with an appropriate electrical circuit model. A novel technique of predicting glass transition temperature was developed using Broadband dielectric spectroscopy. A method was developed on Z-FDM to perform a micro-scale freeze-drying study that has application to study ice nucleation, ice growth, ice solidification and confirmation of the end of the sublimation process. A new image analysis technique using MATLAB was developed to predict drying rate. The frequency of 1.6 kHz by Z-FDM shows potential for freeze-drying cycle development (freezing and drying). Furthermore, the Z-FDM is applied to accurately determine the collapse temperature and remove subjectivity and operator error. Conclusion: Using IDE, this project developed two novel methods using micro-sample electrical impedance analysis for freeze-drying process development. Both BDS and Z-FDM remove the subjectivity in determining the critical temperature of the formulation, predicting T_g^', ice growth rate and drying rate.