Simulation and 3-D Visualization of Crystallization Process Based on Cellular Automata

Crystallization is widely used for separation and purification. During this process, the different growth conditions, such as temperature and supersaturation, would result in different size and morphology, which not only affect the product quality, but also affect the subsequent process operations. Through the simulation of the crystallization process, a better understanding of crystallization process dynamics can be extracted. The population balance (PB) model has been used to simulate industrial crystallization processes, while the multi-dimensional population balance (MPB) model can be used to obtain the information of size and morphology distribution. However, the morphological distribution information provided by numerical solution of MPB is macroscopic, and the specific morphology of a crystal particle cannot be displayed at the microscopic level, so the difference among crystals morphology cannot be clearly displayed. Cellular automata (CA) is a method that simulates a process by considering simple cell changes and cell-to-cell interactions. It can be employed to conduct cross-scale simulation, i.e., both macro and micro. In this work, a CA method is introduced into an MPB model to simulate the batch cooling crystallization process from solution. The rule of crystal growth comes from classical diffusion theory, and whether the solution is transformed into a crystal is realized by the Monte Carlo method. The crystallization of potassium dihydrogen phosphate (KDP) is taken as a case-study and the results is verified with the results in the literature. Through CA simulation, not only the macroscopic crystal size and morphology distribution information can be obtained, but also the crystal morphology in microscopic can be exhibited. At the same time, the crystallization process can be visualized in 3-D environment, with size and morphology distribution being presented intuitively. In addition, the calculation time is reduced under certain accurate conditions. This result provides a theoretical reference for modeling, analysis and controlling of the crystallization process in the future. Copyright (C) 2022 The Authors.