engleski

Mathematical Modelling of Drying

Mathematical modeling of drying is very difficult task due to the changing states of the moist product during drying and many different dryer types. To date, no uniform design and mathematical model for dryers exist. The most used models are so called thin-layer models. In the world there are two major opinion that exist. One group of authors claimed that that are purely empirical models, and no one expect to find physical meaning of the model parameters. On the other hand, the other group of authors states quite the oposite. The objective of this work is to point out the physical meaning of drying model parameter and to find out how the heating intensity and granulometric properties of material influences the drying kinetics and model parameters for various types of materials. Drying data were correlated with new modification of Page's thin-layer model. The modification was made in order to define phisycal meaning of one model parameter. Experiments were carried out on a laboratory scale dryers. Microwave drying of leather, paperboard, wood and two pharmaceutical powders were performed at a different microwave power levels. Convection drying data for clay, Al-Ni catalyst and one of the powder were obtained in the range of temperature between 40°C and 70 °C. Obtained results show that applied mathematical model describes drying kinetics very well. It turned out that one model parameter directly define the moment at which diffusion as the governing moisture removal mechanism starts. In other words this parameter, tk, is the time at which second, if exist, falling rate period starts. The other model parameter, n, does not depend on temperature and microwave power level, so it was supposed that its value depends on the way that heat is supplied to the material (i.e. drying method) and initial moisture content. At higher drying rates and smaler specific surface area estimated values of parameter tk were lower, so diffusion take place earlier.

diffusion; drying; pore size distribution; thin layer model

nije evidentirano