Classification and Quantitation of Edible Oils by Vibrational Spectroscopy: Lessons from the Interface of Application and Analysis

Apr 12, 2022
Facebook X LinkedIn Email
Login  Register
About This Webinar
Analyzing edible oils provides a relevant and challenging application for analytical method development. Olive oil is considered one of the most frequently adulterated foods available, to the point that it has been estimated that global olive oil sales exceed global olive oil production.

Additionally, determining early onset degradation is essential to assessing the market value of edible oils. Karl Booksh’s research group at the University of Delaware has been interested in developing Raman, near-infrared (NIR), and infrared (IR) spectroscopic methods to rapidly classify edible oils by type, detection of adulteration, and quantitation of quality indices such as peroxide value (PV). But this application is challenging. All classes of oils studied consist of the same triglycerides in various proportions. The key to successful analyses lies in understanding the interaction between information content within the various vibrational spectroscopic methods and the capabilities of multivariate data analysis tools such as chemometrics and machine learning to relate the embedded spectroscopic to the desired information about the oils.

The team analyzed a truly inhomogeneous collection of 100 edible oil samples — spanning 19 edible oil types, naturally aged to varying degrees. The findings showed that most literature fails to account for the natural variations among brands and between seasons during analyses; this lack of accounting greatly exaggerates the probability of success with simple analyses. Even so, successful classification and quantitation are possible within certain bounds.

Booksh presents the interactions between spectroscopic method information content, addressing NIR and IR versus Raman, cell path length, and data fusion. He also presents differential success across multiple applications and shares advanced machine learning techniques for intrinsic property determination, explaining why some techniques work better for edible oil analysis than others.

***This presentation premiered during the 2022 Photonics Spectra Spectroscopy Conference. For more information on Photonics Media conferences, visit  

About the presenter:
Karl BookshKarl S. Booksh, Ph.D., is a professor in the department of chemistry at the University of Delaware (UD). He holds a doctorate of philosophy in analytical chemistry from the University of Washington (UW). He began postdoctoral work with the support of a National Science Foundation (NSF) postdoctoral fellowship in 1996 and a Camille and Henry Dreyfus Foundation fellowship. Booksh has received various honors for his work in analytical chemistry, including an Elsevier Chemometrics Award and an NSF CAREER Award. He serves, or has served, on the editorial advisory boards of the Journal of Chemometrics, Talanta, Analytica Chimica Acta, and Spectroscopy. Booksh is a member and past president of the Society for Applied Spectroscopy (SAS) and a member of the Coblentz Society. He is also a past advisory board member for the NSF’s Committee on Equal Opportunity in Science and Engineering and past chair of the American Chemical Society Committee on Chemists with Disabilities.
spectroscopyfoodedible oil analysisvibrational spectroscopy
We use cookies to improve user experience and analyze our website traffic as stated in our Privacy Policy. By using this website, you agree to the use of cookies unless you have disabled them.