This thesis work is focused on the modeling of inter- and intra-tablet coating variability of pan coated tablets. Tablets are coated for a number of reasons such as controlling the bioavailability and release profile of the drug (functional coatings), ensuring product identification and aesthetics, masking odor and taste and protecting the tablet core. Due to the critical nature of functional film coating, significant variations in coating between tablet-to-tablet (inter-tablet coating variation) and between different regions of a tablet, such as the cap and the band of a biconvex tablet (intra-tablet coating variation) will adversely affect product efficacy. Therefore, modeling the process is an important tool towards predicting and controlling variability and can help eliminate some of the problems caused by poor coating uniformity.;The thesis work uses first principles analysis, Discrete Element Method (DEM) simulations, and experiments to determine the variables that control coating uniformity. The parameters that can potentially affect inter-tablet coating variability and studied in this work are the pan speed, the tablet load, coefficient of friction, and spray zone size and location. The parameters that can potentially affect intra-tablet coating variability and studied in this work are the pan speed, the tablet load, tablet aspect ratio (and sphericity), and the effect of baffles. DEM models used to study intra-tablet coating variability are restricted to simple glued sphere particles instead of modeling true geometric primitives such as cylinders as this requires the implementation of prohibitively difficult contact detection algorithms and development of new force models.;The combination of DEM simulations, experiments and analysis provides a comprehensive framework for the understanding of the processes that control coating variability and serves as a platform from which more complex models of coating processes can be developed and implemented.;The thesis work investigates inter-tablet coating variability, specifically, tablet residence times within the spray zone. DEM computer simulations, experiments, and analytical investigations are performed to measure the residence time per pass, the circulation time, and appearance frequency of spherical shaped tablets for a range of pan speeds and tablet loads. In addition, the fractional residence time, defined as the ratio of time spent by a tablet in the spray zone to the total coating time, is measured. The average fractional residence time (averaged over all the tablets in the bed) is found to be equal to the ratio of the time-averaged number of tablets exposed to the spray to the total number of tablets in the pan, a result that is consistent with analyses. The average fractional residence time is observed to be independent of pan speed and total coating time. Furthermore, the fractional residence time is shown to be related to the residence time per pass and circulation time per pass. Circulation time per pass for a tablet is defined as the average time between successive appearances in the spray zone and residence time per pass is defined as the average time spent in the spray zone per pass. Appearance frequency is defined as the number of appearances a tablet makes in the spray zone per pan rotation. Simulations and analyses show that appearance frequency decreases with increasing pan speed. These various measures of residence time are all related, but from the standpoint of developing an analytical model for coating variability, fractional residence time is a more useful and intuitive parameter as it determines the fraction of total run time that a tablet spends in the spray.;To study the coefficient of variation of the coating mass distribution, the variation in tablet residence times is studied, as both quantities are directly related. The DEM simulations indicate that the coefficient of variation of tablet residence times, and subsequently, of coating mass, decreases with time following a power law relation. The theoretical model demonstrates that the coefficient of variation of residence time for a randomly mixed tablet bed is inversely proportional to the square root of the number of coating "trials." DEM simulations show that during each pan revolution, tablets in the spray zone remain in a quasi-segregated state from tablets located outside the spray zone for some time period termed Delta tseg. Increasing the pan's Froude number (to ensure the tablet bed operates in the well-mixed rolling regime), the spanwise-to-streamwise spray zone aspect ratio, and the tablet-tablet and tablet-pan friction coefficient all act to decrease Deltatseg, leading to more uniform residence times and less inter-tablet coating variability for a given operating time. (Abstract shortened by UMI.)
展开▼
