Maximizing similarity: Using correlation coefficients to calibrate kinetic parameters in population balance models.

Area of Science:

• Chemical Engineering
• Process Systems Engineering
• Crystallization Technology

Background:

• Crystallization is vital for chemical and pharmaceutical purification.
• Model-based design, particularly Population Balance Models (PBMs), optimizes crystallization processes.
• Inline particle monitoring offers qualitative insights but faces challenges in direct kinetic parameter estimation for PBMs.

Purpose of the Study:

• To develop and validate a novel approach for kinetic parameter estimation in PBMs using inline particle monitoring data.
• To overcome limitations of direct data interpretation from inline monitoring tools.
• To compare the novel method's performance against classical and naive estimation techniques.

Main Methods:

• Developed a novel approach utilizing offline product size data and correlation-based information from inline particle monitoring.
• Compared the novel method with a classical approach (solute concentration and product size data) and a naive approach (direct comparison of inline data with simulations).
• Evaluated prediction capabilities using two in-silico case studies, employing Pearson's correlation coefficient.

Main Results:

• The novel correlation-based technique demonstrated precision and predictive capability comparable to the classical approach.
• The method proved effective even with noisy data and complex systems involving agglomeration and deagglomeration.
• Pearson's correlation coefficient yielded optimal results in the tested scenarios.

Conclusions:

• The proposed method offers a viable alternative for PBM kinetic parameter estimation, leveraging inline particle monitoring data.
• This approach effectively bypasses experimental and data interpretation challenges associated with direct use of inline monitoring data.
• The findings support the practical application of this technique, unlocking the potential of inline measurements for PBM development.