(536f) Scale-up of Drying of Supported Catalysts in a Fluidized Bed

Drying of supported catalysts is an essential unit operation in the manufacturing of catalysts. Several different types of dryers are used for catalyst drying include belt dryers, rotary kiln dryers, and fluidized bed dryers. In fluidized bed drying, the heated gas flows through a bed of solids and fluidizes the bed, promoting high rates of heat and mass transfer between the gas and solids. Fluidized bed dryers allow for simultaneous spray impregnation and drying of supported catalysts, while promoting high moisture content uniformity through fluidization. Although several studies have been done on fluidized bed drying, general scale-up equations for fluidized bed drying of catalysts are lacking, as drying in a fluidized bed depends on the material properties and fluidization conditions. Thus, it is difficult to choose the operating conditions to optimize the drying process due to several phenomena occurring simultaneously, including heat and mass transfer, drying and fluidization. In general, one would like to have sufficient understanding of the process so that experiments done on a laboratory scale could be used to predict behavior at larger scales.

Experiments were carried out in a Glatt GPCG-1 fluidized bed dryer. Wetted catalyst supports were used as the experimental material. The rate of drying can be described by the change in moisture content of the bed over time. Experiments were done for different temperatures, gas flowrates and bed loadings. Model equations for predicting the moisture content of the bed were implemented using reference drying curves. We found that the model was generally capable of predicting the moisture content of the bed and the rate of drying for various process conditions. Moreover, for different fluidization conditions, we found that the model was capable of predicting the time to dry to 0% moisture content for cases with a delay in fluidization. We found that the model worked best when the majority of the drying took place in the unhindered drying period where the rate was roughly constant. However even when drying took place in the hindered drying period, the model was still able to make reasonably good predictions. We also found that there was better agreement between the model and experiments for high bed loadings where the number of transfer units was large. This work provides quantitative findings on how process conditions affect the rate of drying for a bed of catalyst supports with corresponding scale up equations.