Relationship between NIR Spectrum and Polymer Particle Size for Monitoring Emulsion Polymerization Processes

Emulsion polymerization is widely used in industrial applications, such as paints and resins, due to its versatility and ability to create materials with varied properties. To enhance these processes, in-line and online monitoring techniques are crucial, particularly for tracking the polymer particle size during reactions. Near-infrared (NIR) spectroscopy combined with fiber optic sensors has demonstrated potential for this purpose; however, direct measurement is not feasible, which requires the development of calibration models to correlate spectral data with desired properties. The present study includes experimental data from the synthesis of 11 homopolymers and 3 copolymers of methyl methacrylate (MMA) and styrene (St) and analyzes their NIR spectra. The spectral region between 9000 and 14,000 cm–1 showed high sensitivity to particle size changes during emulsion polymerization. Calibration models using Principal Component Regression (PCR), Partial Least Squares regression (PLS), and Artificial Neural Networks (ANN) with spectral ranges of 14,000–4000 cm–1 and 14,000–9000 cm–1 were developed. Results indicated that the ANN model with the 14,000–4000 cm–1 range performed best, demonstrating that a robust calibration model for monitoring particle size across various formulations and conditions was achievable.