Effect of dirty-hold time on cleaning process of pharmaceutical equipment.

The work was aimed at the evaluation of a cleanliness of pharmaceutical equipments after the end of the production and subsequent cleaning process. The influence of a dirty-hold time, a time interval between the end of the production period and the beginning of the cleaning process on its efficiency and the cleanliness of the equipment has been studied. The evaluation was performed for commercial tablet antihypertensive formulation with API losartan potassium. Sampling was carried out by a wet-swabbing method from the equipments and consequently obtained samples were analytically evaluated using HPLC. In the production of the concerned pharmaceutical, it has been found that the cleaning process is properly designed and validated. Despite the concentration of losartan in swabs from the equipment was in all cases within the limits of acceptance criteria, the effect of the dirty-hold time was proved. In the equipments with long hold-time period, the monitored substance was found in substantially higher concentrations.

Optimization of the high-shear wet granulation wetting process using fuzzy logic modeling.

A fuzzy model has been developed for the optimization of high-shear wet granulation wetting on a plant scale depending on the characteristics of pharmaceutical active substance particles. The model optimized on the basis of experimental data involves a set of rules obtained from expert knowledge and full-scale process data. The skewness coefficient of particle size distribution and the tapped density of the granulated mixture were chosen as the model input variables. The output of the fuzzy ruled system is the optimal quantity of wetting liquid. In comparison to manufacturing practice, a very strong sensitivity of the optimal quantity of the added wetting liquid to the size and shape of the active substance particles has been identified by fuzzy modeling.

Kinetic models for volatile chlorinated hydrocarbons removal by zero-valent iron.

Kinetic models for removal of trichloroethylene, trichloromethane and tetrachloroethylene from water by zero-valent iron were tested. The dehalogenation reactions were modelled by first-order and power law models, pseudo-stationary models with a controlling surface reaction rate and non-stationary models without the assumption of rate controlling step. Regression analysis proved, that the first-order kinetic is not suitable for the modelling of chlorinated hydrocarbons dechlorination. On the other hand, power law models, Langmuir-Hinshelwood analogy models and general models of heterogeneous reactions are reliable for the kinetic description of dechlorination. In spite of an empirical or semi-empirical character, the power law models and models of controlling surface reaction rate can be recommended for the regression analysis owing to the their simple regression parameters interpretation.