|
Designing formulations for Drug Delivery Systems (DDS), which deliver the molecule according to a particular desired profile, is an important area in pharmaceutics, requiring time-intensive experimental research. A good computer simulation of the dissolution process would be an excellent tool permitting to reduce the in vitro (and hence in vivo) testing required.
Modern delivery systems present themselves as a large range of complex systems consisting of both drug and excipient molecules. The process of dissolution is conditioned not only by the physical-chemical properties of all the components plus the interaction of these properties, but also by the geometrical form of the DDS and the USP paddle apparatus settings. In recent years, the Monte Carlo method has come to be used in simulating many complex systems and microstructural evolution in materials. We are approaching the simulation of the behaviour of such complex systems as DDS dissolving in the USP apparatus, by Monte Carlo and Cellular Automata based methods. Linear and non-linear mathematical equations have extensively been used in describing drug diffusion. However, the kinds of systems we are looking at here typically show much more complex behaviour, and have not many simplifying properties as to permit the use of the field of non-linear dynamics. The whole picture is made even more complicated by the presence of one or more moving boundaries, as the components dissolve. The method used in this study consists of representing a tablet (or a tablet's section in the case of 2D simulations) as a grid of sites, representing different states of the system (crystalline solid, dissolving or leaking solid, solvent in the immediate boundary to the tablet, stirred solvent), housing different kinds of particles (drug and excipient), which dissolve according to a number of rules based on the situation in the neighbouring sites. Different types of neighbourhoods can be considered. A Monte Carlo chain is run for a number of time steps (generations) and the emerging macroscopic properties of the whole system are examined at the end. Finally, the simulations are to be calibrated with reference to experimental data. |