Detection and quantification of pork adulteration in beef meatballs with Raman spectroscopy and near infrared spectroscopy

Abstract

One of the main halal concepts requires that food is free from pork substances. Muslim-majority countries establish halal regulations that require the screening of processed meat products, such as meatballs, are screened for adulteration with pork meat to guarantee appropriate halal certification for consumers. Currently, halal authorities rely on the analysis of DNA, protein, or fat with RT-PCR, LC-MS, or GC-FID, which are reliable but are not suitable for rapid screening of large numbers of samples. Hence, high throughout screening tools are demanded to identify suspected samples. Vibrational spectroscopy methods such as Raman spectroscopy (RS) and Near Infrared spectroscopy (NIRS) are widely studied as fast and non-destructive methods for compositional analysis of agrifood products. Therefore, the aim of this study was to evaluate their potential for screening of suspected meatball samples. To this end, different batches of pure beef meatballs and meatballs with different levels of adulteration (3, 5, 10, 50, and 100 % w/w) were prepared and scanned in backscattering (RS) and reflectance (NIRS) mode in intact and cut form. The acquired Raman spectra had dominant peaks at 1657 cm⁻¹, 1443 cm⁻¹ and 1299 cm⁻¹, which were attributed to saturated and unsaturated fat, while the dominant peaks in the NIR spectra corresponded to O–H bonds of water (1457 nm and 1934 nm). The cross-sectioned configuration was found to provide more stable classification performance compared to measurements on intact meatballs for both RS and NIRS. The accuracy of the partial least squares-discriminant analysis (PLS-DA) models for cross-sectioned samples using four latent variables ranged from 52.50 % to 85.00 % for RS and from 58.97 % to 75.00 % for NIRS. The performance of RS and NIRS shows little difference, but RS provides better insights on primary component of meat. For further research, improving the quality of Raman signal with a higher excitation wavelength laser or RS techniques that minimize fluorescence interference may improve model performance.