Characterization and classification of PGI Moroccan Argan oils based on their FTIR fingerprints and chemical composition
| Autor: | Yvan Vander Heyden, Issam Barra, Katim Alaoui, Rabie Kamal, Mourad Kharbach, Abdelaziz Bouklouze, Mohammed Bousrabat, Mohammed Alaoui Mansouri, Yahia Cherrah |
|---|---|
| Přispěvatelé: | Faculty of Medicine and Pharmacy, Analytical Chemistry and Pharmaceutical Technology, Department of Analytical Chemistry, Applied Chemometrics and Molecular Modelling, Pharmaceutical and Pharmacological Sciences |
| Jazyk: | angličtina |
| Rok vydání: | 2017 |
| Předmět: |
spectroscopy
food.ingredient Principal component analysis Argan oil 01 natural sciences Analytical Chemistry Chemometric classification 0404 agricultural biotechnology food Partial least squares regression Peroxide value Fourier transform infrared spectroscopy Chemical composition Chromatography Protected Geographical Indication Chemistry Process Chemistry and Technology 010401 analytical chemistry 04 agricultural and veterinary sciences Linear discriminant analysis 040401 food science 0104 chemical sciences Partial Least Squares Discriminant Analysis Computer Science Applications FTIR fingerprints lipids (amino acids peptides and proteins) Fatty acid composition Software |
| Popis: | In this work Fourier Transform Infrared Spectroscopy (FTIR) was selected as a reliable, fast and non-destructive technique to record spectroscopic fingerprints of Moroccan Protected Geographical Indication (PGI) Argan oils. Classification and discrimination according to their five geographical origins (Ait-Baha, Agadir, Essaouira, Tiznit and Taroudant) was performed. A total of 120 PGI Argan oil samples were collected during four harvest seasons between 2011 and 2014.First, several physicochemical parameters were measured, i.e. free acidity, peroxide value, spectrophotometric indices, fatty acid composition, tocopherols and sterols content. Secondly, FTIR fingerprints were recorded for all samples. The data was subjected to Principal Component Analysis (PCA) for visualization and to reveal differences between samples. Classification models were developed by Partial Least Squares Discriminant Analysis (PLS-DA). Mathematical data pre-treatments were applied to improve the performance of the multivariate classification models. The results obtained, based on both the chemical composition and the spectroscopic fingerprints, indicate that PCA plots were able to distinguish the five sample classes. PLS-DA models based on either chemical composition or FTIR spectra gave a good prediction and an accurate discrimination between the samples from different regions. The proposed approach with the FTIR spectra provided reliable results to classify the Moroccan PGI Argan oils from different regions in a rapid, inexpensive way requiring no prior separation procedure. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|