Effect of scale of operation on CSD dynamics in evaporative crystallizers

Abstract

Experimental measurements of the crystal-size distribution ( CSD) clearly indicate a pronounced difference in the dynamic behavior of a 20-and 970-L continuous crystallizer that produces ammonium sulfate. The difference in their circulation time offers a probable explanation for this phenomenon. It causes different supersaturation profiles in the two crystallizers, which leads to internal fines dissolution in the large crystallizer. This contributes to the observed oscillations in the 970-L crystallizer as opposed to the first-order responses in the 20-L crystallizer. To numerically study the effect of the supersaturation profile a dynamic model, from which the MSMPR (mixed suspension mixed product removal) assumption is omitted, is developed. Calculated supersaturation profiles differ considerably for the 20-L, the 970-L and an imaginary 50,000-L continuous evaporative crystallizer. Coincident with changes in the supersaturation profiles, the numerical solution of the model indicates the tendency of large crystallizers to oscillate and supports this suggested explanation. I. Experimental IntroductionIndustrial crystallizers are known to exhibit large variations in the crystal-size distribution (CSD) they produce. These variations most probably result from unstable behavior of the crystallization process itself, although the influence of unwanted changes in process conditions cannot be fully excluded. Laboratory crystallizers, on the other hand, usually demonstrate stable operation that is illustrated by rapidly decaying CSD-transients (Randolph and Larson, 1988). Relatively few experimental studies have been devoted to the dynamic behavior of crystallizers in general, particularly to the difference in dynamic behavior between the crystallizers of different scale.It was shown previously (Jager et al., 1990a,b) that CSD transients in a 20-L crystallizer rapidly approach steady-state conditions. These results can be simulated well with an MSMPR model including secondary nucleation kinetics combined with attrition. In this article, this analysis is made to a larger-scale of operation. Also discussed are CSD transients measured in a 970-L crystallizer, which was operated with and without fines removal and whose results are compared with CSD transients measured in the 20-L crystallizer, operated at the same con- 182February 1991 ditions using the same ammonium sulfate solur ion. Finally, differences between the two crystallizers are explained. Experimental ProcedureThe 20-L crystallizer is a continuous, evaporative, drafttube crystallizer, operated at 50°C. The crystalliier has a propeller stirrer of 100-mm-diameter operated at 700 rpm. Other dimensions, together with a process scheme, are given by . The slurry circulation time is approximately 2 seconds, equally divided between the inner and the outer tube. In this article, the crystallizer startup, at the operating conditions of the 970-L crystallizer, will be discussed. Sieve analysis of filtered and dried samples, using 20 Veco microprecision sieves, was used to ...

Importance of macromixing in batch cooling crystallization

“…Turnover time usually increases with size of agitated tanks, from 0.2 s in a 6.3 L crystallizer, to 60-70 s in a 50 m3 unit (Garside, 1985;Heffels, 1986;Jager et al, 1991). Hence, the results suggest that the influence of spatial variations is low, not only in small laboratory crystallizers, but also in large industrial units.…”

“…In a continuous evaporative crystallizer, the solution may actually become undersaturated in the heat exchanger. Small crystals may dissolve completely, resulting in an oscillating product mean size especially in large units (Jager et al, 1991). This suggests that our conclusions concerning the influence of macromixing in a batch cooling crystallizer should not be extrapolated uncritically to batch evaporation crystallizers.…”

“…Kratz and Hoyer (1989) suggest a scale-up method based on a loop model. Jager et al (1991) modeled the effect of scale of operation on CSD dynamics, and the results were used to explain oscillatory behavior of a continuous evaporation unit.…”

In Situ Ultrasonic Spectroscopy Study of the Nucleation and Growth of Copper Sulfate Pentahydrate Batch Crystallized from Supersaturated Aqueous Solutions

“…The decrease in solid concentration could reflect partial dissolution of attrition product, possibly as a result of variations of solubility with crystal size. 20 A similar phenomenon was reported during evaporative crystallization of ammonium sulfate 21 and was also ascribed to the dissolution of fines during the process.…”

Modeling, simulation and kinetic parameter estimation in batch crystallization processes

“…

IntroductionThe onset of instability in industrial continuous crystallization processes in response to small disturbances has previously been shown to be determined primarily by the strong nonlinear dependence of nucleation rates on the relative supersaturation driving force (Buyevich et al, 1991;Jager et al, 1991). Consequently, accurate kinetic modeling of crystallization kinetics, especially nucleation, is a prerequisite to the optimal design and control of crystallization processes.

…”