(333e) Effect of Agitation Rate on Breakage Kinetics of Urea Crystals in Agitated Slurries | AIChE

(333e) Effect of Agitation Rate on Breakage Kinetics of Urea Crystals in Agitated Slurries


Hill, P. - Presenter, Mississippi State University
Shah, P., Mississippi State University
Particle breakage can occur in stirred vessels such as solution crystallizers and often occurs when the crystals are acicular, or have high aspect ratios. These acicular crystal are frequently produced when crystallizing pharmaceuticals or fine chemicals. Crystal breakage is important because it can produce many fines that can act as secondary nuclei, or that can plug the filter downstream of the crystallizer. Producing secondary nuclei may or may not be desired, but plugging the filter is not desired. To better operate crystallizers, it is necessary to be able to model the breakage of high aspect ratio particles. Therefore, this work investigates the breakage of high aspect ratio particles at various agitation rates.

Previous researchers [1-2] modeled the specific rate of breakage of high aspect ratio crystals as being proportional to the cube of the agitation rate. In one case experiments were only performed at one agitation rate [1], while there were not experimental comparisons for the other case [2]. Previous researchers used simplified models where each parent particle breaks into two child particles. In these cases the parent particles were modeled as rectangular parallelepipeds where the width ad depth were the same and the width was the minor axis, while the length was the major axis. Some researchers modeled the breakage as occurring perpendicular to the long axis of each crystal, while others modeled breakage as occurring parallel to the long axis of each crystal. It was also determined that there is probably both types of breakage occurring.

This investigation used urea as the experimental compound. Urea crystals were grown in water to produce high aspect ratio crystals for the breakage experiments. Crystals were sieved to narrow the size distribution used for the breakage experiments and to reduce fine particles. Before starting the breakage experiments, the initial parent crystal were characterized to determine the length and width of each crystal so that the aspect ratio of each crystal could be determined.

Breakage experiments were conducted in an agitated vessel with the urea crystals suspended in hexane. Hexane was used because urea is practically insoluble in hexane, and this isolates the mechanism of breakage by preventing growth and dissolution. Initial experiments were conducted in a lab scale vessel at 1250 rpm and crystals were characterized after 1, 5, 10 and 15 minutes of breakage. The agitation rate was chosen to match the impeller tip speed in industrial crystallizers.

Results from this research are reported as the average aspect ratio as a function of crystal length for various size intervals. Average aspect ratios were studied both as a function of the major axis and as a function of the minor axis. Our results show that the residence time for breakage has little effect on the average aspect ratio for each size interval, and that the aspect ratio appears to be a stronger function of child particle size. A comparison of the models with experimental results show that: 1) the simplified model with breakage perpendicular to the major axis cannot explain the aspect ratio distribution results, 2) the simplified model with breakage along the minor axis cannot explain the results, and 3) that a combination of the two models cannot explain the resulting aspect ratio distributions.

[1] Kazuhiro Sato, Hidetada Nagai, Kazuhiro Hasegawa, Kunihiko Tomori, H.J.M. Kramer, P.J. Jansens, Two-dimensional population balance model with breakage of high aspect ratio crystals for batch crystallization, Chemical Engineering Science, 63 (2008) 3271 – 3278.

[2] Botund Szilagyi and Bela G. Lakatos, Model-based analysis of stirred cooling crystallizer of high aspect ratio crystals with linear and nonlinear breakage, Computers & Chemical Engineering, 98 (2017) 180-196.