Rapid Detection Of Pesticides In Rice Using Surface-enhanced Raman Spectroscopy Technology

Rice is the most important food among the five grains and is the main food crop in our country.However,pesticide residue in rice is one of the critical hazards that causes safety issues and poses a serious threat to consumers' health and their life safety.Conventional pesticide detection methods have disadvantages such as long time,cumbersome detection steps,and expensive equipment,which cannot meet rapid on-site detection requirements.Therefore,the establishment of rapid and sensitive pesticide detection methods is in great demand to ensure rice safety and promote the development of the rice industry.In recent years,surface-enhanced Raman spectroscopy(SERS)as a simple,fast,and sensitive detection technology has been widely used in the field of food safety detection.However,SERS detection systems for pesticides usually have disadvantages,such as complex preparation processes of signal probes and enrichment probes.Therefore,the current work carried out with the objective of rapid detection of pesticides in rice based on SERS technology coupled chemometrics.The main study contents are as follows:1.Study on rapid detection of chlorpyrifos pesticide residues in rice based on SERS technology.Aiming at the complicated preparation of signal probe and enrichment probe of SERS detection system for pesticides,this study took chlorpyrifos pesticide as the research object,and carried out the rapid detection of pesticide residues in rice based on SERS technology combined with chemometrics method.Firstly,synthesized the flower-like zinc oxide nanoparticles by wet chemical method and loaded silver nanoparticles on the surface,the silver-zinc oxide nano flowers(Ag@Zn O NFs)were used as the SERS enhancement substrate.The synthetic Ag@Zn O NFs substrate was mixed with chlorpyrifos pesticides at different concentrations(0.01-1000 ?g/m L)to collect SERS spectra.Then the genetic algorithm-partial least squares algorithm(GA-PLS),ant colony optimization algorithm-partial least squares algorithm(ACO-PLS),variable combination population analysis-partial least squares algorithm(VCPA-PLS)and competitive adaptive reweighted sampling-partial least squares model(CARS-PLS)were used to construct the quantitative model between SERS spectrum and chlorpyrifos concentration,respectively.The results showed that all the four models could achieve the quantitative determination of chlorpyrifos concentration,and the CARS-PLS model had the best prediction result,with the prediction set correlation coefficient(Rp)of 0.9881 in the concentration range of 0.01-1000 ?g/m L,and the detection limit was0.01 ?g/m L.Finally,in the actual quantitative determination of chlorpyrifos in rice poisoning,the method was compared with high performance liquid chromatography(HPLC).There was no significant difference in the detection results obtained by the two methods(P>0.05),and relative standard deviation(RSD)<11%.The results showed that the method could be used for quantitative determination of chlorpyrifos in rice.The detection method of chlorpyrifos in rice based on SERS technology in this study met the requirements stipulated by the national standard,and the detection speed was greatly improved.2.Research on simultaneous detection of two pesticides in rice based on SERS technology.In response to more than one pesticide in rice,this study took chlorpyrifos and carbendazim pesticides as the research objects,and carried out a research on the simultaneous detection of multiple pesticides in rice based on SERS technology.Firstly,silver nanoflower(Ag NFS)with rough surface was synthesized by one-step synthesis method as SERS reinforced substrate.The synthesized Ag NFS substrates were mixed with chlorpyrifos and carbendazim pesticides of different concentrations(0.01-1000 ?g/m L),respectively,and SERS spectra were collected.Then,the standard normal variate transformation(SNV)algorithm was used to preprocess the collected SERS spectra to eliminate the signal interference such as baseline drift and background noise.CARS-PLS,uninformative variable elimination-partial least squares algorithm(UVE-PLS)and interval combination population analysis-partial least squares algorithm(ICPA-PLS)algorithm were used to build the quantitative relationship model between SERS spectrum and the concentration of chlorpyrifos and carbendazim.The results showed that the ICPA-PLS model had the best prediction results for chlorpyrifos pesticide,with Rp=0.9669,root mean square error of the prediction(RMSEP)= 0.3517 ?g/m L,and RPD = 3.74;while UVE-PLS model had the best prediction results for carbendazim pesticide,Rp =0.9970,RMSEP=0.1346 ?g/m L,RPD=9.76.Finally,the experimental method was applied to detect chlorpyrifos and carbendazim in actual rice samples and compared with the measured data by HPLC.The t-test results showed no significant difference between the two methods(P>0.05).The study showed that the SERS method established in this chapter could realize the simultaneous quantitative detection of chlorpyrifos and carbendazim pesticides in rice.3.Rapid identification of pesticides in rice based on SERS technology.Given of the phenomenon that the actual detection process of rice often only needs to quickly determine whether the pesticide residues exceed the standard,this study took chlorpyrifos carbendazim and thiobencarb pesticides as the research object,and carried out a research on the rapid discrimination of pesticides in rice based on SERS technology.First of all,according to the national standard of the maximum residual limits(MRL)of chlorpyrifos,carbendazim and thiobencarb in rice,set chlorpyrifos,carbendazim and thiobencarb,exceeded and did not exceed the pesticide standards.Ag NFs substrates were mixed with different concentrations(0.01-1000 ?g/m L)of chlorpyrifos,carbendazim and thiobencarb pesticides,and SERS spectra were collected.Then three qualitative discriminant models,K-nearest neighbor model(KNN),linear discriminant analysis(LDA)and partial least squares discriminant analysis(PLSDA)were established.The results showed that the LDA model had the best discriminant results for chlorpyrifos and carbendazim,and the discriminant rates of the training set and the prediction set reached 100%.The PLSDA model had the best discriminant results for thiobencarb pesticide,and the discriminant rates of the training set and the prediction set were 95.19% and 96.54% respectively.Finally,the actual rice samples were detected by SERS method and HPLC method,respectively,and the accuracy rate of established SERS method reached 100%.The study showed that the method in this chapter could realize the qualitative discrimination of three pesticides of chlorpyrifos,carbendazim and thiobencarb exceeded the standard of rice poisoning,which provided a new idea for the rapid classification of pesticides in rice.