| Pesticides and veterinary drugs play a vital role in the production of agricultural products.However,due to the abuse,misuse and non-compliance with withdrawal period of these chemicals,a lot of agricultural products contain excessive agricultural drug residues,threatening human health and social stability.Therefore,it is of great significance to develop a rapid detection technology for drug residues in agricultural products.Raman spectra technique,which conduct the qualitative and quantitative measurement through the collection and analysis of molecular vibration and rotation,has the advantages of no pretreatment,simple operation,short time-consuming and environmental harmlessness.In this paper,the main agricultural products such as apple and pork were taken as the research object,and the simultaneous,rapid,non-destructive,real-time and efficient detection methods of agricultural drug residues were studied based on the surface enhanced Raman scattering(SERS)and Raman hyperspectral imaging technology.The main contents and results were summarized as follows:1.Combing with SERS technology,and taking silver colloid as surface strengthening agent,a rapid and efficient detection method for detecting pesticide residues in fruit and vegetable was developed.It takes short determination time and needs no sample pretreatment,the overall detection time is less than 10 min and the detection limit is lower than the national standard limit.The results of 50 times detection showed good stability for acetamiprid,chlorpyrifos and carbendazim which are common pesticide used for apple production,the RSD of the three pesticide signal intensities were all less than 5.5%.The universal applicability of the proposed method was verified by the detection results of other fruits and vegetables such as pear,oranges,mango,spinach and rape.It was found that after the spectral signals were averaged and processed by the Savitzky-Golay smoothing and then corrected fluorescence background by adaptive iteratively reweighted penalized least squares method,the established multiple linear regression model showed the optimal prediction effect,the correlation coefficients of prediction(Rp)was 0.969,0.975 and 0.972,respectively,for acetamiprid,chlorpyrifos and carbendazim,and the respective root mean square error of prediction(RMSEP)was 1.065,1.153 and 0.899.2.In view of the realistic demand of the simultaneous detection of mixed pesticide residues in fruits and vegetables,aiming to solve the problem of SERS signals overlap in the process of mixed pesticide detection,the feasibility of using SMA to decompose the SERS signals was studied,the mutual effect of SERS signals of different pesticides was also analyzed.The results showed that SMA could effectively extract the Raman spectra of single component from the Raman signals of mixture of acetamiprid,chlorpyrifos and carbendazim,and its characteristic peak position and signal intensity were consistent with pure product.The interplay of the components in the mixture was weak at low concentrations,and the model based on the pure pesticide had a good ability to predict the concentration of the corresponding component in the mixed pesticide.The Rp was 0.893,0.926 and 0.938,respectively,for acetamiprid,chlorpyrifos and carbendazim,and the respective RMSEP was 1.065,1.153 and 0.899.The rapid and effective decomposition of the SERS signals of mixed pesticide by SMA increases the speed and accuracy of the detection of pesticide residues.3.Based on SERS technology,using silver colloid as surface strengthening agent,a rapid and efficient detection method for detecting veterinary drug residues in pork,liver and other animal origin agricultural products was developed.The technique requires simple sample pretreatment to extract the analyte and has a quick and easy quantitative process after the addition of the surface enhancer.When salbutamol and sulfadimidine were taken as research object,the obtained multivariate linear model showed good prediction effect.The RP for salbutamol and sulfadimidine was 0.974 and 0.985,the RMSEP was 0.487 mg/L and 0.413 mg/L.The contents of salbutamol and sulfadimidine in pork,chicken,liver and kidney were detected.The detection limits were 0.0075-0.05 mg/kg and 0.005-0.025 mg/kg.The Rp was all greater than 0.951 and 0.957,respectively.4.A simultaneous,rapid,non-destructive and real-time detection techniques for determining the residues of ofloxacin,chloramphenicol and sulfadimidine in pork was developed based on Raman hyperspectral imaging system.The visualization of veterinary drug distribution in sample was achieved through the study of signal extraction and analysis method.After the Raman hyperspectral image of the unbroken pork sample was collected under the optimized conditions,the background correction was carried out on the the selected characteristic band images of the analytes,then the visualized distribution of analyte can be obtained by selecting an appropriate thresholds to realize the binarization of the images.The quantification of drug residues was operated by pixel ratio method,in the determination of pure matter,the detection limits of vancomycin,chloramphenicol and sulfadimidine in pork were 0.05%,0.04%and 0.05%,respectively,and the respective RP was 0.974,0.9775 and 0.982.Using the model based on pure drug to predict the content of each component in the mixture could still acquire favorable results,the RP for ofloxacin,chloramphenicol and sulfamethazine was 0.978,0.986 and 0.984,respectively.This paper provides a new technology and method for the simultaneous,fast,non-destructive,real-time and efficient detection of agricultural drug residues in agricultural products.The interested information can be collected rapidly without sample pretreatment or only with simple pretreatment,then the related accuracy and stable models were developed for the residues prediction of pesticides and veterinary drug,providing new ideas and new technologies for the safety detection of agricultural products. |
PDF Full Text Request