Raw material quality assessment approaches comparison in pectin production

Area of Science:

• Biotechnology and Bio-based Production
• Process Analytical Technology (PAT)

Background:

• Quality variations in biological raw materials cause significant process disturbances.
• Product properties are directly linked to the initial quality of raw materials.
• Current industrial practices often rely on less precise, rule-of-thumb decisions for quality assessment.

Purpose of the Study:

• To explore methods for evaluating quality variations in biological raw materials.
• To assess Process Analytical Technology (PAT) tools for rapid and informative quality assessment.
• To compare different quality assessment approaches using citrus peels as a case study.

Main Methods:

• Utilized near-infrared (NIR) spectroscopy coupled with multivariate data analysis (MVDA).
• Compared three distinct quality assessment approaches: expert-knowledge-based discriminant classification, unsupervised classification, and spectroscopic correlation with physicochemical variables.
• Performed quantitative comparative analysis on a single dataset.

Main Results:

• Multivariate data analysis combined with NIR spectroscopy demonstrated significant advantages for raw material characterization.
• The study quantitatively compared the performance of different classification and correlation methods.
• Citrus peels were used as a model to showcase the effectiveness of the proposed approach.

Conclusions:

• NIR spectroscopy and MVDA provide a more informative and rapid alternative to traditional quality assessment methods.
• Tuning operational conditions based on accurate raw material quality assessment ensures desired product specifications.
• This approach enhances process stability and product consistency in bio-based manufacturing.