Research Of Kinetic Modeling And Corresponding Experiments For Batch Cooling Crystallization From Solution

Crystallization, as a solid-liquid separation process, is widely used in the chemical industry. The transfer of the solute molecules from the continuous phase to the solid phase is governed by nucleation and growth rate, which are of importance for the design and performance of the industrial crystallizers. In order to obtain the kinetic rates, the appropriate crystallization kinetic model is expected to develop. The various fundamental mechanisms involved in crystallization, however, are so complex and unpredictable that the mathematic description of crystallization process is relatively difficult, which slows the further application of the crystallization separation craft. The progress of the kinetics study on crystallization from solution was firstly reviewed, and then the research of kinetic models for batch crystallizers was carried out in this thesis using the test apparatus set up in our laboratory.For seeded batch crystallization process, combining with Beer-Lambert law, ΔL law and the population balance theory, a kinetic model relating to transmittance was deduced by optical methods. The potassium nitrate-water system was tested and the parameters of nucleation and crystal growth were acquired simultaneously by correlating the kinetic model with the dynamic transmittance, the relative supersaturation and concentration, which agreed well with those in the literature.Compared with seeded crystallization process, unseeded one has more complex kinetic mechanism and less attention in the literature. A volume of nuclei will spontaneously occur in a relatively short interval of crystallization without seeding, and then it can be assumed that the secondary nucleation should be followed on within obvious decrease of supersaturation. Responding to this characteristic of unseeded crystallization process, it was considered that the generation rate of new nucleus was approximately equal to secondary nucleation rate after primary nucleation, and a modified kinetic model similar to the seeded crystallization model was proposed for potassium nitrate-water system. The parameters in the model, obtained employing the sectionalized nonlinear regression method, appeared to be in good agreement with those published in the literature.For improving the precision of kinetic parameter estimation, it is important to identify the nucleation and crystal growth stages during the crystallization procedure. In this thesis, two new zero dimension variables, and, concerning the processidentification were defined after analyzing the above models for batch crystallization process. Based on the fact that and decreased within nucleation process and increased within crystal growth process of potassium nitrate aqueous solution, coupled with the dynamic transmittance and concentration data, the nucleation and crystal growth stages could be quantitatively identified, and the result agreed well with that of linear growth rate model.The crystallization process of organic system is being one of the most interesting research focuses. Accordingly, the crystallization kinetics of xylene mixtures was investigated in this thesis. Considering the characteristics reported for xylene system, seeding policy was performed in the first separation stage and experimental concentrations within the crystallization process were measured using gas chromatography. The nucleation and crystal growth kinetic parameters were evaluated directly from the concentration data, and the results calculated showed that the values of also exhibited an essential tendency as that of potassium nitrate-water system.