Visualizing the Dynamics of Fluid-Bed Dryers
A team at the University of Saskatchewan is applying tomography to the study of fluid bed dryers in a bid to make proper operation of the equipment easier and clearer.
Operating fluid bed dryers today still involves more art than it does science. “Basically, an operator looks through a sight glass and manually controls the process,” says Todd Pugsley, professor of chemical engineering at the University of Saskatchewan (Saskatoon).
Scale-up is even more problematic, he says, since what happens at the benchtop level is very different from what happens during full-scale production. “Patterns within the vessels change,” he says. And those changes ultimately affect product quality; larger granules can be segregated or finer granules entrained. Highly potent drugs or protein-based therapies are at greatest risk for quality problems.
Pugsley and his team have been studying the gas-particle and particle-particle interactions of fluid bed drying since 1998. They’re using new sensors, tomography and high-speed photography to capture the dynamic behavior of particles in the dryer. “Tomography’s nonintrusive, and you get a clear picture of what’s going on,” Pugsley says.
|Fiber optics and high-speed photography pinpointed the problematic zone (shown here): the base of the product container, at the interface of the container wall and horizontal orifice plate. Graphic courtesy of Glatt.|
A team including Pugsley and his colleagues, as well as Merck Frosst Canada, used electrical capacitance tomography to determine the optimal flow behavior to improve drying and minimize fines. Then they used models to calibrate the sensor and confirmed results with Xray tomography.
In a previous project, they had used a piezoelectric sensor to measure pressure fluctuation within a dryer. They replaced the product bowl’s viewing window with an acrylic insert threaded so that the transducer could be mounted within it. They then collected data for both a dry and a wet bed of granules, with the wet bed moving from 33% wt. to 4% wt. moisture.
Chaos analysis, in the form of the S-statistic method, developed by J.R. van Ommen of the Technical University of Delft, was then used to analyze data. Hydrodynamics varied depending on the amount of moisture present. “When you take wet granules in a fluidized bed and start to dry them, the behavior of gases and the solid flow patterns change significantly,” says Pugsley. Ultimately, knowledge gained will lead to automatic control systems, he says.
Not only drug manufacturers are refining visual hydrodynamic measurement techniques. The fluidized bed equipment maker Glatt used fiber optics and high-speed photography to characterize the Wurster coating process and pinpoint problem zones, to optimize it for tablet coating. Research ultimately led to the design of a new insert (see "Insert Improves Wurster Coating for Tablets, Particles") that improves coating uniformity.
Academic research promises to raise the level of process understanding. “Manufacturers will be able to show FDA that they understand how changes in gas velocity and changes in the humidity of incoming air affect the process, as well as product quality,” says Pugsley. Research in this subject is also ongoing at the University of Calgary, the City College of New York, and Rutgers University.