Raman-Chemometric Framework for Rapid Authentication of Edible Oils in Processed Foods Using a Validated One Step Sampling Technique

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Authentication of cooking oils in processed foods is essential for food safety, quality control, and regulatory compliance, yet routine analysis remains constrained by solvent-intensive extraction procedures and limited applicability to complex food matrices. This study presents a proof-of-concept Raman–chemometric framework for direct oil authentication in processed foods using minimal, solvent-free sample handling. Potato chips were selected as a representative fried food matrix, and five commonly used edible oils (sunflower, soybean, groundnut, palm, and vanaspati) were analyzed in both pure form and corresponding chip matrices to enable systematic cross-matrix evaluation. A one-step tissue paper blotting method was employed for rapid oil recovery, followed by Raman spectroscopic analysis. Fatty-acid-associated Raman bands were identified and systematically combined into chemically interpretable inter-peak intensity ratios reflecting variations in saturation and ester content. A statistically grounded two-stage marker selection workflow integrating Random Forest importance ranking with non-parametric Kruskal–Wallis testing was applied, followed by using one-way analysis of variance and coefficient of variation analysis. Five robust ratiometric markers (I₁₆₅₂/₁₇₄₂, I₁₆₅₂/₁₄₃₄, I₁₇₄₂/₁₂₅₉, I₁₆₅₂/₁₂₅₉, and I₁₂₉₆/₁₄₃₄) showed strong association with saturated versus unsaturated fatty acid profiles. Multivariate analysis based on these markers revealed pronounced separation among oil types (91% explained variance; F-values up to 22,629; p < 0.001) and statistically significant discrimination within chip matrices (77% explained variance), despite attenuation effects from the food matrix. Multivariate analysis of variance confirmed robust separation (Wilks’ Λ = 0.0307; p < 0.0001). Overall, this framework establishes an interpretable, extraction-free, and scalable foundation for high-throughput oil authenticity screening in processed foods.

Article activity feed