(585d) On the Optimization of the Cooling Profile in Crystallization Using the Design of Dynamic Experiments Approach

To control the size and shape of crystals during batch crystallization, one must either have an accurate crystal growth and nucleation model or run a series of experiments to find the optimal operating conditions. Since the former is a very intricate and time-consuming task, especially for new compounds, the later will be applied using a novel technique, Design of Dynamic Experiments (DoDE)(Georgakis, 2008, 2009), to determine the optimal operating conditions for a batch crystallizer that yield the desired size and shape distributions of the product crystals. The novelty of DoDE is that it is a model-free approach. Instead of using a fundamental model, the optimal operating conditions are found by performing a number of experiments in which the time dependent decision variable(s) is (are) varied according to a systematic set of dynamic signatures, often constrained between minimum and maximum allowed values. The results of the experiments are then used to calculate the optimal profile of the time-dependent decision variable(s). This method determines the optimum processing conditions while utilizing a minimum number of resources. In crystallization the most prevalent time dependent decision variables are the cooling rate and the antisolvent addition. In this presentation, we will discuss results showing the usefulness of DoDE to optimize the cooling profile of a two-liter batch crystallization unit equipped with a model predictive temperature controller to obtain the desired cooling profile(s) in each of the DoDE experiments as well as in the optimal one producing the desired distribution of crystal population characteristics. While the methodology presented here is applicable to any crystallization, the experimental data reported relate to the crystallization of potassium dihydrogen phosphate (KH2PO4), or KDP.

____________________________

Georgakis, C. (2008). Dynamic Design of Experiments for the Modeling and Optimization of Batch Process.

Georgakis, C. (2009, July 12-15). A Model-Free Methodology for the Optimization of Batch Processes: Design of Dynamic Experiments. Paper presented at the IFAC Symposium on Advanced Control of Chemical Processes 2009, Istanbul, Turkey.