Determination of adulterated neem and flaxseed oil compositions by FTIR spectroscopy and multivariate regression analysis

Abstract

Natural oils are increasingly being used in the food, cosmetic, and agrochemical industries in recent years. However, natural oils with high market values are often counterfeited and adulterated with cheap, poor quality oils, with serious economic implications for the food and cosmetic industries, and potential health implications to consumers. This study reports the first combined use of Fourier transform infrared spectroscopy (FTIR) and partial-least-square (PLS) multivariate regression analysis for rapid, accurate, and low cost determination of the % compositions of two natural oils (neem oil (NO) and flaxseed oil (FO)) adulterated either with edible vegetable oil (VO) or extra virgin olive oil (EVOO). The FTIR of the calibration sample sets containing adulterated NO and FO with VO and EVOO at a concentration range of 1–99% w/w were measured and subjected to PLS multivariate regression analyses. The obtained FTIR spectra profile of the adulterated samples are highly dependent on the type of natural oil as well as the type of adulterant oil. The developed PLS models were subsequently used to determine the % compositions of independently prepared validation samples of NO and FO adulterated with VO and EVOO. The figures-of-merit of the PLS regression models were excellent and with good linearity (R2 > 0.998814). The score plots of the PLS regressions revealed interesting and useful information for pattern recognition of adulterated NO and FO samples. The PLS regression models correctly determined % compositions of adulterated NO and FO with VO with low root-mean-square-relative-percent errors (RMS%RE) of determination of 3.02%, and 4.24%, with an overall RMS%RE of 3.63%. The models also correctly determined % compositions of adulterated NO and FO with EVOO with RMS%RE of 7.13% and 2.00%, with an overall RMS%RE of 4.56%. The simplicity and high accuracy of this low-cost study presents an attractive model with potential real-world applications in quality control and quality assurance for consumer products in the food, cosmetics, and agrochemical industries.