« Back
Real time detection of fish fraud using rapid evaporative ionisation mass spectrometry (REIMS)
Real time detection of fish fraud using rapid evaporative ionisation mass spectrometry (REIMS)
Submitted on 03 Apr 2017

Olivier Chevallier, Connor Black, Julia Balog, Sara Stead, Steven Pringle, Zoltan Takats, Christopher Elliott
Institute for Global Food Security, Queen’s University Belfast, UK; Waters Research Centre, Budapest, Hungary; Waters, Wilmslow, UK; Imperial College London, London, UK
This poster was presented at Pittcon 2017
Poster Views: 1,161
View poster »
Poster Abstract
The increasing number of reports regarding food fraud scandals has brought food authenticity and safety to the attention of regulators, industry and consumers worldwide. Ambient ionization mass spectrometry (AMS) methods have overcome a number of intrinsic constraints of traditional mass spectrometric analysis schemes, allowing in situ, real-time analysis of a wide variety of samples. REIMS has been used for the analysis of human tissue during surgery and has shown to be capable of the identification of various tissue types based on lipid fingerprinting. In this study, we present an effective, near real time method to identify fish product speciation methods using REIMS.
REIMS was operated in single stage MS and negative mode. Data was acquired using a Medimass REIMS source coupled with a Waters Xevo G2-XS QTof mass spectrometer. All specimens were sampled using a monopolar handpiece that was equipped with a smoke evacuation line that was mounted on the atmospheric interface of the mass spectrometer. Full spectral information was recalibrated, normalised, baseline subtracted and binned up to 0.1 m/z bin size. The resulting data was subjected to multivariate analysis such as principal component analysis (PCA) followed by a linear discriminant analysis (LDA) using a non-commercial prototype software developed by Waters.
Over 3000 spectra were acquired from five different authenticated species of fish; cod, coley, haddock, pollock and whiting. Spectral data were acquired between m/z 200-1200. Both PCA and LDA score plots, built using m/z 600-950, identified clear signs of fish speciation. Validation of the PCA-LDA models carried out with another batch of fish samples resulted in a 94% correct classification rate.
REIMS technology could provide a paradigm shift across authenticity applications by providing real-time, reliable, and simple method for the analysis of food products.
Report abuse »
Ask the author a question about this poster.
Ask a Question »

Creative Commons