Skip to Main content Skip to Navigation

Freeze-drying of vaccines : Contribution of mathematical modelling for assessing product heterogeneity and scale-up risks

Abstract : Freeze-drying is the process of choice in pharmaceutical industry for the stabilization of heat sensitive products such as vaccines. However, due the product pre-conditioning in individual vials, this process is difficult to design and often results in batches presenting a significant heterogeneity in the quality of the final product. The main goal of this Ph.D. project was the development of a mathematical model making it possible to predict the risk of failure when designing the freeze-drying process, i.e., the percentage of "rejected vials". To this end, the work focused on the understanding and quantification of the sources responsible for heat and mass transfer variability during the process. Firstly, the vial-to-vial heat transfer variability was investigated by taking the vial bottom dimensions and the vial position on the shelf of equipment into account. The variability of geometrical dimensions observed within a batch of vials (i.e., contact area between the shelf and the vial and the mean bottom curvature depth) moderately influenced the heat transfer coefficient distribution among vials (by less than 10 %). Secondly, a original 3D mathematical model was developed in COMSOL Multiphysics to explain and predict atypical heat transfer observed in vials located at the border of the shelf during the freeze-drying process. Conduction through low-pressure water vapour appeared as the dominant mechanism explaining the additional heat transfer to border vials rather than as reported in literature radiation from the walls of the drying chamber. Furthermore, this 3D mathematical model was used to investigate the effect of the vial loading configuration and of the equipment characteristics on heat transfer variability. In a second part, mass transfer variability was quantified on a 5% sucrose solution and by focusing on two parameters, the resistance of the dried layer to mass transfer during sublimation and the characteristic desorption time. The dried layer resistance was assessed by combining complementary approaches, the pressure rise test and gravimetric methods. The estimated variability of the dried layer resistance was found to have a higher impact on the product temperature distribution than the heat transfer coefficient variability. The value and variability of characteristic desorption time was evaluated for different temperatures and made it possible to simulate moisture content heterogeneity between vials in the batch. In the last part of the work, the main quantified sources of heat and mass transfer variability were integrated in a mathematical model of freeze-drying process. This multi-vial, dynamic model was used not only to predict the evolution of product temperature and moisture content during freeze-drying for a batch of 100 vials, but also to estimate the percentage of vials that could potentially be rejected. The proposed approach, extended to a greater number of simulated vials, could be applied to calculate design spaces of the primary and secondary drying steps of freezedrying process at a known risk of failure.
Complete list of metadatas

Cited literature [126 references]  Display  Hide  Download
Contributor : Abes Star :  Contact
Submitted on : Saturday, November 16, 2019 - 1:02:32 AM
Last modification on : Friday, October 23, 2020 - 4:38:10 PM
Long-term archiving on: : Monday, February 17, 2020 - 12:39:23 PM


Version validated by the jury (STAR)


  • HAL Id : tel-02366530, version 1


Bernadette Scutella. Freeze-drying of vaccines : Contribution of mathematical modelling for assessing product heterogeneity and scale-up risks. Chemical and Process Engineering. Université Paris Saclay (COmUE), 2017. English. ⟨NNT : 2017SACLA034⟩. ⟨tel-02366530⟩



Record views


Files downloads