Merging Partial Least Squares & Raman spectroscopy to quantify oxidative stability in biodiesel

Abstract

As biodiesel is becoming more important as an alternative and cleaner biofuel in the energy transition, methodologies for monitoring its quality are necessary to guarantee the integrity of engines. One of the most relevant parameters in the quality control routine is oxidative stability, a characteristic related to the effect of oxygen in biodiesel at room temperature during storage. Quantifying the degree of oxidation in biodiesel is essential as undesirable products, such as sediments, polymers, or short-chain fatty acids, can be formed, significantly modifying its biofuel properties. In this work, for the first time, a rapid and non-destructive methodology based on Raman spectroscopy assisted by Partial Least Squares (PLS) regression was developed to quantify the oxidative stability of soy biodiesel by calibrating its Raman spectra against the induction time, monitored through the reference method. This combined strategy allows a fast and environmentally friendly methodology to quantify the induction time. With it, Root Mean Squared Error values for calibration and prediction of 0.2759 and 0.3260 h, respectively, were reached. These values were significantly lower than those reported by other spectroscopic methodologies, showing that merging the biodiesel's highly informative Raman features with the potentiality of chemometric modeling is very promising for rapid routine analyses that can be easily adapted outside the laboratory. Besides, this provides a substantial advancement in biodiesel quality control, considering that its use as a biofuel continuously grows worldwide.