To dye for: No more
- Published: Oct 15, 2014
- Author: David Bradley
- Channels: Chemometrics & Informatics
Spline is fine
A combination of an iteratively cubic spline fitting baseline correction method with discriminant partial least-squares qualitative analysis can be used to pluck evidence of banned additives from food samples analysed with surface-enhanced Raman spectroscopy (SERS), quickly revealing the presence of compounds such as Sudan I dye, Rhodamine B and Malachite green.
Wei Zhang's research team at Chongqing Institute of Green and Intelligent Technology, part of the Chinese Academy of Sciences explain how spectroscopy combined with a baseline correction method and exploratory analysis methodology could be a useful technique in the field of food safety. They describe details in the journal Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy.
There are many inexpensive and previously widely available dye compounds used as brightly coloured additives to artificially enhance the appearance of a range of foods from confectionary to savoury snacks and many food and drink products. Among their number is Sudan I dye (also known as CI Solvent Yellow 14 and Solvent Orange R), a red pigment with the chemical name 1-phenylazonaphth-2-ol, which betrays it as an azo compound. Sudan I like Sudan III, and Sudan IV and other azo dyes have been banned on the basis of their putative carcinogenicity through their interaction with the enzyme cytochrome P450. Sudan I has been associated in particular with "chilli" and "curry" powder because it gives those products their expected hot orange-red colour. Likewise, rhodamine B is banned in many jurisdictions on the basis of, in its case, unproven carcinogenicity. The widely used dye malachite green, on the other hand, is also widely used in aquaculture to protect fish eggs from mould, but it too is a suspected carcinogen and thus its presence in seafood is a potential threat to human health.
Numerous analytical techniques have been developed to detect these and similar dye stuffs in food products with varying degrees of success. Ultraviolet-visible spectroscopy has been used to test spices for contamination with the Sudan dyes, for instance, and various cheminformatics approaches used to process the spectra from UV-Vis or chromatography and mass spectrometry data. The research suggests that only tiny sample quantities are needed with SERS and that this technique has the added advantage of being fast and non-destructive.
However, background fluorescence effects and baseline noise and other interference detrimentally affect Raman spectra and so the team has developed a cheminformatics approach to the analysis of these spectra that essentially allows them to clean up the data and so more easily identify contaminants in food samples. As such, the team has developed what they call a "chemometric spectra treatment, a polynomial baseline correction method based on cubic spline fitting" and investigated this for its potential to remove background noise. This cleanup they explain is then successfully followed by the application of an unsupervised exploratory analysis principal component analysis (PCA) and multivariate statistical analysis discriminant partial least squares (DPLS classification) to assess the suitability of SERS as a screening method for discerning food contaminated with banned food additives.
In the future, the team plans to develop novel SERS substrates with high sensitivity and reproducibility as well as programming the analysis for better performance. The new SERS substrates and method might then be integrated into a portable Raman spectrometer. Meanwhile, the Raman spectra database and software will be optimized to be useful in environmental pollutant and biomedical detection, Wei Zhang told SpectroscopyNOW.
Spectrochim Acta A Mol Biomol Spectrosc, 2014, 137C, 1092-1099: "Multivariate qualitative analysis of banned additives in food safety using surface enhanced Raman scattering spectroscopy"
Analyt Methods, 2014, 6, 4402-4407, "Baseline Correction for Raman Spectra Using Improved Asymmetric Least Squares"
Article by David Bradley
The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.