Reaction Kinetics Modeling & Optimization

Project Status: Active | Type: First-Principles Simulation

1. Overview

This project simulates a Series Reaction (A -> B -> C) in a batch reactor to solve a classic chemical engineering optimization problem: maximizing yield under conflicting kinetic constraints.

Unlike "Black Box" machine learning models, this project uses First-Principles Modeling (Arrhenius equations & Differential Equations) to determine the optimal operating conditions.

2. The Engineering Problem

We are optimizing the production of Product B in the following pathway: $$A \xrightarrow{k_1} B \xrightarrow{k_2} C$$

• Reaction 1 (A->B): High Activation Energy ($E_a = 50,000$ J/mol). Favored by High Temp.
• Reaction 2 (B->C): Lower Activation Energy ($E_a = 40,000$ J/mol). Represents thermal degradation (Waste).
• Constraint: Batch time is fixed at 60 minutes (shift change).

3. Key Results

Using scipy.optimize and odeint, we determined the global maximum:

• Optimal Temperature: 398.6 K (125.4 °C)
• Predicted Yield: 0.78 mol/L
• Sensitivity Analysis: The process is highly sensitive to Temperature (narrow operating window) but robust to Time variations, suggesting that capital investment should focus on precise Temperature Control Units (TCUs) rather than automated timing systems.

4. Tech Stack

• Python 3.14
• Scipy (odeint, minimize_scalar): For solving differential equations and optimization.
• Numpy & Matplotlib: For vector calculus and surface mapping.

5. Usage

1. Run the Simulation: python src/reactor_simulation.py
2. Find Optimal Temp: python src/optimize_temperature.py
3. Generate Heatmap: python src/yield_heatmap.py