Mathematical Modeling for the Industrial 2-Mercaptobenzothiazole Batch Production Process.

Area of Science:

• Chemical Engineering
• Process Modeling
• Industrial Chemistry

Background:

• 2-mercaptobenzothiazole (MBT) is a key chemical intermediate, crucial for the rubber industry.
• Existing models lack a first-principles basis for describing the MBT synthetic process.
• Industrial batch production of MBT requires a robust and reliable mathematical model.

Purpose of the Study:

• To formulate a reliable mathematical model for the industrial batch production of 2-mercaptobenzothiazole (MBT).
• To address challenges in parameter estimation due to limited industrial/experimental data.
• To ensure the model's generalizability and prediction performance.

Main Methods:

• Development of a mathematical model using differential and algebraic equations.
• Application of estimability analysis to reduce the parameter set for estimation.
• Utilization of a multiple-starting-point strategy with weighted least-squares for parameter estimation.
• Employing cross-validation techniques to assess model generalizability.

Main Results:

• A reduced set of estimable parameters was successfully derived.
• The non-convex parameter estimation problem was numerically solved.
• The proposed model demonstrated encouraging prediction performance on independent test data.
• Cross-validation confirmed the model's generalizability.

Conclusions:

• The developed mathematical model provides a reliable framework for MBT industrial batch production.
• The methodology effectively handles parameter estimation challenges with limited data.
• The model shows strong predictive capabilities, validated by independent data.