Monitoring of sugars adsorption breakthrough curves with online Raman spectroscopy

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Chemical Engineering

Background:

• Adsorption processes are crucial in chemical separations.
• Monitoring adsorption breakthrough is vital for process optimization.
• Raman spectroscopy offers a non-invasive, real-time analytical technique.

Purpose of the Study:

• To develop and validate an online Raman spectroscopy method for monitoring glucose and xylose adsorption breakthrough.
• To compare univariate and multivariate (Partial Least Squares) calibration models for sugar quantification.
• To assess the impact of ethanol as a cosolvent on model performance.

Main Methods:

• Online Raman spectroscopy was applied to monitor adsorption breakthrough of glucose and xylose mixtures.
• Univariate and Partial Least Squares (PLS) calibration models were developed using spectral data.
• Models were validated against offline High-Performance Liquid Chromatography (HPLC) measurements.
• Spectral subtraction and first derivative were used as data preprocessing techniques.

Main Results:

• Accurate predictions of glucose and xylose concentrations were achieved using univariate models.
• No significant improvement in prediction accuracy was observed with PLS models compared to univariate models.
• The presence of ethanol as a cosolvent did not negatively impact the univariate model performance.
• Effective pretreatment methods (spectral subtraction and first derivative) enhanced model robustness.

Conclusions:

• Online Raman spectroscopy is a viable tool for real-time monitoring of sugar adsorption breakthrough.
• Univariate calibration models are sufficient and robust for quantifying glucose and xylose in this application.
• Pretreatment strategies are key to developing reliable spectroscopic models for complex mixtures.