Monte Carlo simulations for the study of drug release from matrices with high and low diffusivity areas.

We use Monte Carlo simulations in order to study diffusion controlled drug release from matrices consisting of random mixtures of high and low diffusivity areas (random mixing), and from matrices covered by a thin film of low diffusivity (ordered mixing). We compared our results with the Weibull model for drug release and found that it provides an adequate description of the release process in all cases of random mixing and most cases of ordered mixing. We have studied the dependence of the Weibull parameters on the diffusion coefficient and, in most cases, found a rather simple linear dependence. Moreover, our results indicate that a device covered by a thin film with diffusion coefficient three orders of magnitude lower that the coefficient of the rest of the device, will release drug at constant rate for most of the release process. This last result may have considerable practical applications.