Non-destructive Raman spectroscopy as a tool for measuring ASTA color values and Sudan I content in paprika powder
| Autor: | Nils Kristian Afseth, Arsenio Muñoz de la Peña, Olga Monago-Maraña, Carl Emil Eskildsen, Teresa Galeano-Díaz, Jens Petter Wold |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2019 |
| Předmět: |
Sudan I
Analytical chemistry Color Food Contamination Naphthols Spectrum Analysis Raman 01 natural sciences Fluorescence Analytical Chemistry Root mean square chemistry.chemical_compound symbols.namesake 0404 agricultural biotechnology Non destructive Partial least squares regression Least-Squares Analysis Spectroscopy Mathematics 010401 analytical chemistry Discriminant Analysis Signal Processing Computer-Assisted 04 agricultural and veterinary sciences General Medicine 040401 food science 0104 chemical sciences chemistry Content (measure theory) symbols Powders Raman spectroscopy Capsicum Food Analysis Food Science |
| Zdroj: | Food Chemistry |
| Popis: | The aim of this study was developing a non-destructive method for the determination of color in paprika powder as well as for detecting possible adulteration with Sudan I. Non-destructive Raman spectroscopy was applied directly to paprika powder employing a laser excitation of 785 nm for the first time. The fluorescence background was estimated, by fitting a polynomial to each spectrum, and then subtracted. After preprocessing the spectra, some peaks were clearly identified as characteristic from pigments present in paprika. The preprocessed Raman spectra were correlated with the ASTA color values of paprika by partial least squares regression (PLSR). Twenty-five paprika samples were adulterated with Sudan I at different levels and the PLSR model was also obtained. The coefficients of determination (R2) were 0.945 and 0.982 for ASTA and Sudan I concentration, respectively, and the root mean square errors of prediction (RMSEP) were 8.8 ASTA values and 0.91 mg/g, respectively. Finally, different approaches were applied to discriminate between adulterated and non-adulterated samples. Best results were obtained for partial least squares – discriminant analysis (PLS-DA), allowing a good discrimination when the adulteration with Sudan I was higher than 0.5%. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|