Falling film melt crystallization (I): Model development, experimental validation of crystal layer growth and impurity distribution process

• Published in: Chemical engineering science 2012-12, Vol.84, p.120-133
• Main Authors: Jiang, Xiaobin, Hou, Baohong, He, Gaohong, Wang, Jingkang
• Format: Article
• Language: English
• Subjects: Applied sciences; Chemical engineering; chemical industry; Coefficients; cooling; Crystal Layer Growth; Crystallization; Crystals; Diffusion; Exact sciences and technology; Falling; heat; Impurities; Mathematical models; Melt Crystallization; Melts (crystal growth); Parameter identification; Phosphoric acid; Separations; Simulation; sweating
• ISSN: 0009-2509; 1873-4405
• DOI: 10.1016/j.ces.2012.08.026

This paper was concerned with the model development and experimental validation of the detailed crystal layer growth and multi-ions impurity distribution process in the falling film melt crystallization (FFMC) model. The phosphoric acid (PA) was separated and purified by FFMC to obtain a hyperpure phosphoric acid (HPA), which was a vital electronic chemical in IT industry. To establish a valid model, which offered an easy and convenient path of the simulation, dynamic heat and mass balance, approaches were adopted to describe the variation of crystal layer growth rate along the crystallizer. An impurity balance approach was adopted to describe the change of distribution coefficient for multi-ion impurity. A criterion was proposed to determine the formation of branched-porous (B–P) structure. The model was validated by experimental results with various equipments and operational conditions and a good agreement was obtained. The effective distribution coefficient Keff for multi-ion impurities were less than 0.2 (Na+), 0.25 (Fe3+) and 0.35 (Ca2+) with proper operation conditions. The resulting model was directly exploited to understand crystal layer growth and impurity distribution behaviors in FFMC from laboratory to industrial scale. More significantly, the model proposed a method for the separation effect evaluation and the key operational conditions (feed rate and cooling rate) determination which could readily develop optimal crystal layer growth route during industrial crystallization. In addition, the model was a vital base to describe the subsequent purification step of FFMC: sweating process. ▸ A dynamic model was developed for the layer growth and impurity separation process. ▸ The separation effect of branched-porous crystal layer was well simulated by the model. ▸ The feed rate and cooling rate were the key operation parameters. ▸ The critical length had a linear dependence with the feed rate. ▸ The model displayed the industrialization capacity of FFMC for hyperpure product.