(789c) Estimation of Kinetic Parameters From in-Situ Measurements During Batch Cooling Crystallization | AIChE

(789c) Estimation of Kinetic Parameters From in-Situ Measurements During Batch Cooling Crystallization


Li, H. - Presenter, Georgia Institute of Technology
Kawajiri, Y., Georgia Institute of Technology
Grover, M., Georgia Institute of Technology
Rousseau, R. W., Georgia Institute of Technology

Recently many studies have focused on using in-situ measurements to monitor and control crystal size distribution (CSD) [1, 2, 3]. Our objective in this study is to model crystallization kinetics and design a control strategy to produce a desired CSD using information from in-situ measurements to estimate crystallization kinetics. The measurements include focused beam reflectance measurement (FBRM) for the solid phase and attenuated total reflectance-Fourier transform infrared spectroscopy (ATR-FTIR) for the liquid phase.

Using an empirical approach we proposed previously [4], CSD can be estimated from the chord length distribution (CLD) measured by FBRM. In the approach, crystals of different size ranges are prepared by sieving and their characteristic CLDs were measured. The matrix formulated from the characteristic CLDs is based on the linear FBRM model, which associated CSD with CLD. For the ATR-FTIR spectrum, the ratio of peak heights produced by solute and solvent was used to estimate the solute concentration.

Different temperature profiles and initial concentrations were used in batch cooling crystallizations of paracetamol from ethanol, in order to determine the influence of these variables on crystallization kinetics. Key findings from these experiments were (1) the occurrence of nucleation during cooling, as opposed to nucleation under isothermal conditions at the same supersaturation, increased the number of crystals produced; and (2) growth kinetics were highly sensitive to temperature.

A population-balance model was constructed to find the kinetic parameters that fit the experimental data. The method of moments was applied to the data to obtain ranges of the model parameters, which were then used as starting points for the optimization of the full population-balance model. With the resulting model and the fitted parameters, an optimal operation profile was determined and confirmed by subsequent experiments.


[1] N. Kail, W. Marquardt, and H. Briesen. Estimation of particle size distributions from focused beam reflectance measurements based on an optical model. Chemical Engineering Science, 64(5): 984-1000, 2009.

[2] J. Worlitschek and M. Mazzotti. Model-based optimization of particle size distribution in batch-cooling crystallization of paracetamol. Crystal Growth & Design, 4(5):891{903, 2004.

[3] A. N. Saleemi, C. D. Rielly, and Z. K.Nagy. Comparative investigation of supersaturation and automated direct nucleation control of crystal size distributions using ATR-UV/vis spectroscopy and FBRM. Crystal Growth & Design, 12(4): 1972-1807

[4] H. Li, M. A. Grover, Y. Kawajiri, and R. W. Rousseau. Development of an empirical method relating crystal size distributions and FBRM measurements. Chemical Engineering Science, 89:142-151, 2013.