Oxidation kinetic studies of oils derived from unmodified and genetically modified vegetables using pressurized differential scanning calorimetry and nuclear magnetic resonance spectroscopy

A. Adhvaryu, S. Z. Erhan, Z. S. Liu, J. M. Perez

Research output: Contribution to journalArticle

181 Scopus citations


Evaluation of oxidative stability of a series of vegetable oils and oils derived from genetically modified vegetables were carded out using pressure differential scanning calorimetry (PDSC). The purpose of including the genetically modified oils along with other oils were to illustrate the effect of high oleic and linoleic content on the thermal and oxidative behavior of these oils. Kinetic and thermodynamic parameters were computed and variation of results explained in terms of structural data derived from quantitative 1H and 13C NMR spectroscopy. For a variety of vegetable oil samples used in the study, log b (program rate of heating; i.e. 1, 5, 10, 15 and 20°C/min) was linearly related (R2 = 0.99) to the reciprocal of absolute temperature corresponding to maximum oxidation rate (peak height temperature). From the resulting slope it was possible to compute activation energy (E(a)) for oxidation reaction and various other kinetic parameters, e.g. rate constant (k), Arrhenius frequency factor (Z) and half-life period (t 1/2). The presence of C-C unsaturation in the fatty acid (FA) chain, their nature and relative abundance, affect thermal and oxidative stability of the oil and subsequently their kinetic and thermodynamic parameters. Quantitative analysis of the NMR spectra yielded various other structural parameters that were correlated with start (T(S)) and onset (T(O)) temperature of vegetable oil oxidation, and certain important kinetic parameters (E(a) and k). This is a novel approach, where statistical models were developed as a predictive tool for quick assessment of oxidative and thermodynamic data. The correlations developed have an adjusted R2 of 0.922 and higher using 3 or 4 NMR derived predictor variables. These correlations revealed that in addition to nature and abundance of C=C, relative abundance of other structural parameters (e.g. bis-allylic methylene group, allylic-CH2, α-CH2 to C=O, etc.) influence oxidation and kinetic data.

Original languageEnglish (US)
Pages (from-to)87-97
Number of pages11
JournalThermochimica Acta
Issue number1-2
Publication statusPublished - Dec 1 2000


All Science Journal Classification (ASJC) codes

  • Instrumentation
  • Condensed Matter Physics
  • Physical and Theoretical Chemistry

Cite this