(671f) Determination of the Power Number of a Retreat-Blade Impeller in a Stirred Vessel Partially Baffled with Different Types of Single Baffles Used in Pharmaceutical Processing
AIChE Annual Meeting
2021
2021 Annual Meeting
North American Mixing Forum
Experimental and Computational Investigations of Mixing Processes
Monday, November 15, 2021 - 2:35pm to 3:00pm
Mixing is an essential unit operation in the pharmaceutical and biopharmaceutical industry to conduct processes such as crystallization, fermentation, cell culture, chemical reactions, and others. In many such applications, the vessel/reactors typically utilized in processing plants are a glass-lined vessel provided with a single impeller, typically a Retreat-Blade Impeller (RBI), and a single baffle placed midway between the vessel wall and the impeller shaft. Since cleanability and chemical compatibility are critical factors, all internals, including the impeller and the baffle, must be glass-coated. In addition, the impeller is typically placed near the vessel bottom to avoid colliding with the baffle located in a more central position in the vessel rather than at the wall. While the presence and position of the single baffle make the system more easily cleanable and accessible, it also reduces the baffling action typically associated with fully baffled systems provided with four wall baffles. Therefore, single-baffle systems are often classified as âpartially baffled.â The most commonly used single baffle is the âbeaver-tailâ baffle consisting of a flattened elongated pipe closed at the end and glass-coated. However, over the years, equipment manufacturers have developed different types of single baffles with greater surface areas exposed to the rotating flow generated by the impeller greater than that of the beaver-tail baffle. Such types include the finger-type baffle, D-type baffle, H-type baffle, fin-type baffle and others types less commonly found. Despite their common use, very little information is available on the hydrodynamics of vessels equipped with these baffles and especially the impeller Power Number associated with their use. Thus, the objective of this study was to experimentally determine the power dissipation and Power Number in a scaled-down RBI system, using the various types of single baffles shown in the figure below, i.e., beavertail, D-type, fin-type, finger-type, and H-type baffles in the turbulent regime at different impeller agitation speeds. Computational Fluid Dynamics (CFD) was additionally used to predict the flow field in all these systems as well as the power dissipation and hence the Power Number. The use of different baffles resulted in the establishment of different flow patterns in the vessel. Furthermore, the CFD predictions and the experimental results for the Power Numbers were found to be in very close agreement. It is expected that this knowledge can help industrial practitioners in the optimization of pharmaceutical processes.