Loading...

Development of Chemometric Methods Combined with Isotope Ratio Mass Spectrometry (IRMS) for Isotope Pattern Recognition and Adulteration Detection in Foods

Ghiasi, Ali | 2020

1091 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 52743 (03)
  4. University: Sharif University of Technology
  5. Department: Chemistry
  6. Advisor(s): Parastar Shahri, Hadi
  7. Abstract:
  8. The purpose of this study was to investigate the capability of Isotopic Ratio Mass Spectrometry (IRMS) in the detection of four common natural plant frauds in saffron, namely Marigold flower, Safflower, Rubia and Saffron Style. For this purpose, first natural saffron isotopes were analyzed and used to investigate fingerprints of carbon-13 and nitrogen-15 isotopes by Elemental Analyzer Isotopic Ratio Mass Spectrometry (EA-IRMS) and carbon-13 isotopic ratio of saffron components by Gas Chromatography Isotopic Ratio Mass Spectrometry (GC-C-IRMS). Then, according to the isotopic behavior of the original saffron, they were sampled to form a mixture of saffron that is representative of the original saffron. The mentioned frauds were mixed with the original saffron at five different weight percentages (5, 10, 15, 25 and 35%) and analyzed by EA-IRMS and GC-C-IRMS. Then,the isotopic behavior of saffron with fabricated samples were modeled using appropriate preprocessing and modeling of linear discriminant analysis (LDA), quadratic (curvature) discriminant analysis (QDA) and partial least squares-discriminant analysis (PLS-DA).Finally, the behavior of the original Saffron samples was modeled as a one class using the supervised pattern recognition modeling of Data Driven Soft Indipendant Modeling of Class Analogy (DD-SIMCA) and the fabricated cheat samples were introduced to the model for prediction. The QDA model showed the ability to predict made frauds even at 5% levels with 100% accuracy. The accuracy and precision obtained for modeling the separation of cheats from the original saffron were 92.7% and 95.7% for the LDA model and 92.9% and 80.0% for the PLS-DA model, respectively. The QDA Cmodel also predicts fully-fake market samples and a mixture of frauds correctly and correctly put them in the fraud class. DD-SIMCA single-class model with 95% confidence level was able to differentiate between the original and cheat samples significantly from their isotopic behavior in the analysis of specific compounds of saffron by GC-C-IRMS
  9. Keywords:
  10. Saffron ; Gas Chromatography ; Chemometrics Method ; Adulteration in Foods ; Isotopic Ratio Mass Spectrometry (IRMS)

 Digital Object List

 Bookmark

...see more