Issue link: https://www.e-digitaleditions.com/i/1000772
32 July 2018 Tablets & Capsules The total mass flowrate through the blender also affects the material's RTD. Changing the overall mass flowrate to manipulate the RTD is uncommon, however, since mass flowrate changes imply an unsteady continuous process, and the mass flowrate of a process is g e n e r a l l y p r e d e t e r m i n e d b a s e d o n p r o d u c t i o n requirements and other nontechnical considerations. A blender's RTD is a critical measurable attribute that aids in the understanding and design of a blending operation [7]. While the effects of process and design parameters and operating conditions on a blender's RTD have been well examined, the effects of material properties on a continuous blender's RTD remain largely unexplored. T&C References 1. Colin F. Harwood, Kenneth Walanski, Erdmann Luebcke, and Carl Swanstrom, "The performance of continuous mixers for dry powders," Powder Technology, Vol. 11, No. 3, pages 289-296. 2. Ralf Weinekötter and Lothar Reh, "Continuous mixing of fine particles," Particle & Particle Systems Characterization, Vol. 12, No. 1, pages 46-53. 3. B. Laurent and J. Bridgwater, "Continuous mixing of solids," Chemical Engineering & Technology, Vol. 23, No. 1, pages 16-18. 4. Sarang Oka, Abhishek Sahay, Wei Meng, and Fernando J. Muzzio, "Diminished segregation in continuous powder mixing," Powder Technology, Vol. 309, pages 79-88. 5. William E. Engisch and Fernando J. Muzzio, "Feedrate deviations caused by hopper refill of loss-in-weight feeders," Powder Technology, Vol. 283, pages 389-400. 6. Aditya U. Vanarase and Fernando J. Muzzio, "Effect of operating conditions and design parameters in a continuous powder mixer," Powder Technology, Vol. 208, No. 1, pages 26-36. 7. Sarang S. Oka, M. Sebastian Escotet-Espinoza, Ravendra Singh, James V. Scicolone, Douglas B. Hausner, Marianthi Ierapetritou, and Fernando J. Muzzio, "Design of an Integrated Continuous Manufacturing System," chapter in Continuous Manufacturing of Pharmaceuticals, Edited by Peter Kleinebudde, Johannes Khinast, and Jukka Rantanen, Wiley-VCH, pages 405-446. Fernando J. Muzzio is director of the Center for Structured Organic Particulate Systems (C-SOPS) (http://www.csops.org) and distinguished professor of chemical and biochemical engineering at Rutgers University, Piscataway, NJ. He can be reached at 848 445 3357 (fjmuzzio@yahoo.com). Sarang Oka is a former postdoctoral associate and graduate student at Rutgers University. He is currently a process development engineer in the Drug Product Continuous Manufacturing group at Hovione. He can be reached at 609 918 2422 (soka@hovione.com). time. Standard deviation and MCV are qualitatively similar; both are a measure of the width of the RTD. Alternately, the tracer can also be introduced as a step function (an instantaneous change in the tracer concentration at the blender inlet). The system response (the tracer concentration at the blender outlet) would then be measured as a function of time and would depend on the blender's RTD. Tracer selection Using a tracer to measure a unit operation's RTD is widely practiced, but the importance of selecting an appropriate tracer material is often overlooked. The tracer must have flow properties as similar as possible to the overall blend but must also be chemically distinguishable from the rest of the material. The tracer must not disturb the material stream's flow behavior and should travel through the blender at the same rate as the rest of the material. Figure 4a shows a blender's RTD when the physical properties of the tracer and the blend were similar. The overall throughput was 20 kilograms per hour (kg/h), and the impeller speed was 350 rpm. As the figure shows, the tracer material was fully flushed from the system within 10 minutes. However, the same blender under identical processing conditions exhibited a very different RTD when using a tracer material with a much higher bulk density than the blend. As shown in Figure 4b, the tracer had not fully exited the blender after 30 minutes of operation, resulting in incorrect RTD characterization. Adjusting a blender's residence time distribution As previously mentioned, you can often adjust or tune a blender to achieve the desired RTD for your applica- tion. Changing the blender's impeller speed is a common way to manipulate the RTD. Increasing the impeller speed decreases the mass holdup but also results in an increased MCV. For example, two RTD values for pure semi-fine acetaminophen running through a blender at 18 kg/h in steady state are shown in Figure 5. The tracer material was caffeine, introduced as a pulse at the blender inlet. The RTD on the left in the figure was obtained with the impeller speed at 350 rpm, while the RTD on the right was obtained with the impeller speed at 400 rpm. The disparity between the RTD shapes is discernible in the figure; operating at higher rpm results in a larger MCV value (0.406 at 350 rpm and 0.483 at 400 rpm). An alternate strategy for changing a material's RTD is to manipulate blender design variables. This can be achieved by altering the blender's incline angle or the angle of the weir at the blender's outlet, but the RTD is most commonly adjusted by changing the impeller blade orientation [6]. Many blenders are available with adjustable impeller blades that you can orient to either push the material fully forward or backward or at an intermediate angle to manipulate axial mixing. Increasing the number of blades pushing material backward will generate an RTD with a larger MCV.