Rapid and simultaneous determination of mixed pesticide residues in apple using SERS coupled with multivariate analysis

This study aims to apply multivariate analysis algorithms for modeling the same spectra, for simultaneous determination of pymetrozine and carbendazim residues in apple. To mitigate the impact of competitive adsorption, SERS spectra are obtained from mixed solutions of pymetrozine and carbendazim at varying concentration ratios, which are then utilized for modeling. Results suggest that the PLSR model based on full-band SNV processed spectra shows the best performance for predicting pymetrozine and carbendazim contents, with R2 p of 0.9751 and 0.9779, RMSEP of 0.0492 and 0.5531 mg/L, RPD of 6.4297 and 6.8246, respectively. This model yielded R2 p of 0.9644 and 0.9671, RMSEP of 0.0747 and 0.8247 mg/L, RPD of 5.3857 and 5.6066 for pymetrozine and carbendazim in apple, respectively. The findings suggest that the proposed approach is suitable for simultaneous detection of pymetrozine and carbendazim in apples, offering a novel avenue for monitoring food safety. • SERS was used to simultaneously detect pymetrozine and carbendazim residues in apple. • SERS intensity of pesticides changed obviously due to competitive adsorption. • Spectra of mixed solution with different concentration ratios were used for modeling. • PLSR model showed superior result for pymetrozine and carbendazim content prediction. • All models exhibited the best result by using the SNV processed spectra for modeling.