Effects of Nucleation and Crystal Growth Rates on Crystal Size Distribution for Seeded Batch Potash Alum Crystallization Process

Abstract

The driving force of the cooling crystallization process is supersaturation, where the supersaturation level during the crystallization process is crucial to grow the crystal sufficiently. Nucleation and crystal growth rates are two concurrent phenomena occurring during crystallization. Both are supersaturation functions that determine the growth of seed crystals and the formation of fine crystals. Trade-offs between nucleation and crystal growth are essential for achieving the large size of seed crystals with the minimum number of fine crystals. Thus, the objective of this study is to analyze the effects of nucleation and crystal growth rates on final product quality, which is crystal size distribution (CSD). Modeling of the crystallization process using a potash alum case study is highlighted and simulated using Matlab software. Then, the effects of nucleation rate, crystal growth rate, and both nucleation and crystal growth rates on CSD are evaluated using local sensitivity analysis based on the one-factor-at-a-time (OFAT) method. Based on simulation results for all strategies, a low combined rate delivers the best performance of the final CSD compared to others. Its primary peak has a mean crystal size of 455 µm with 0.0078 m3/m volume distribution. This means that the grown seed crystals are large with high volume distribution compared to the nominal strategy, which is at the mean crystal size of 415 µm and 0.00434 m3/m. Meanwhile, the secondary peak has the mean crystal size of 65 µm, 0.00028 m3/m in volume distribution. This corroborates the least number of fine crystals at the considerably small size compared to nominal’s (0.00151 m3/m, 35 µm). Overall, the low nucleation and crystal growth rates strategy provides useful insights into designing temperature profiles during the linear cooling crystallization process, whereby achievable supersaturation levels in obtaining large crystals with fewer crystal fines are provided via simulation.

Size-Dependent Growth Modeling Method for Batch Crystallization of Carbamazepine from Variable Seed Crystal Size Distributions

“…35 More recently Adnan et al studied the effect of varying seed sieving strategy and its effects on product CSD, and Long et al conducted an uncertainty analysis on seed CSD revealing that the seed mean size can have significant effect on product CSD. 36,37 Still a more systematic study on varying seed CSD in regard to both mean size and amount of fines with a focus on developing a crystallization model to ensure model robustness as well as to evaluate the effect of seed CSD on product CSD would be beneficial.…”

“…A number of groups have sought to use PBMs to control product CSDs by varying supersaturation and seed loading in the solution. − However, the effects of variation in seed CSD on the product CSD have largely remained unexplored in crystallization modeling, save for a few publications. − Aamir et al modeled the control of product CSD by varying the seed recipe, which demonstrated capability to predict product CSD with variable seed CSD despite the fact that the model was not trained using variable seed CSD and did not consider nucleation . Hsu and Ward also considered the effect of seed CSD on the product, yet their model only considered the broadness of the seed CSD and not the overall size of the crystals nor was their work focused on evaluating model accuracy with varying seed CSD .…”