Mathematical modelling and FTIR spectroscopic imaging of pharmaceutical tablet dissolution

• Main Author: Kimber, James A.
• Other Authors: Kazarian, Sergei ; Stepanek, Frantisek
• Published: Imperial College London 2012
• Subjects: 615.19
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.550910

The process of pharmaceutical tablet dissolution is a vital stage in the delivery of active pharmaceutical ingredients (APIs). The constituent components and their spatial arrangement within the tablet determine the release characteristics of the API. It is therefore important to understand and characterise the various processes and component interactions that occur during tablet dissolution. Computational simulations of tablet dissolution can be used to obtain parametric sensitivities and optimise formulations so that the desired API release profile is achieved. This thesis describes the methods behind modelling the behaviour of non-swelling and swelling tablets, the mathematical validation of the models, parametric studies and the experiments which were used to obtain parameters and verify the models. The experimental method used in this work is Fourier Transform Infrared (FTIR) spectroscopic imaging, which, when using an attenuated total reflection (ATR) accessory and flow cell, enable chemical and spatial information to be obtained from the tablet as it dissolves. UV/Visible spectroscopy was also used to obtain drug release information. The non-swelling model discretised a tablet over a Cartesian grid and solved the mass transfer equations (dissolution and diffusion) to obtain drug release profiles. Two parametric studies were conducted where the particle size distribution and mass fractions were varied in one, and the API diffusivity, saturated concentration and mass fraction in the other to see what effect these had on drug release, demonstrating the importance of the choice of excipient and the impact of particle size on release variability. For experimental validation, tablets containing different quantities of polyethylene glycol and nicotinamide were dissolved and imaged, and optimisation was used to obtain the pure component saturated concentrations. The model was then tested against a different tablet to demonstrate the predictive capability of the model. The swelling model discretised a tablet into small cylindrical particles, whose size was proportional to the mass of components within them and whose motion was determined using the Discrete Element Method (DEM). As water diffused into polymer particles, they could expand, resulting in macroscopic swelling. The DEM model of a swelling and dissolving tablet was validated against a numerically exact model of the same tablet and parametric studies were conducted into the effect of polymer disentanglement threshold, polymer equilibrium water fraction and polymer dissolution rate. The model was also optimised against a dissolving tablet containing HPMC to obtain parameters for this excipient. To conclude, both models were implemented, validated and found to accurately describe the dissolution kinetics of both swelling and non-swelling tablets.