Detection Of Mixed Pesticide Residues Of Prochloraz And Imazalil In Citrus Epidermis By Surface Enhanced Raman Spectroscopy

Prochloraz and Imazalil are commonly used fresh-keeping fungicides for citrus fruits,and additionally mixture of them can effectively reduce bacterial resistance and achieve better freshness.However,the high concentrations of pesticide residues in fruits and vegetables may affect the health of consumers.Consequently,in order to ensure the safety,it is important to develop different technologies to detect pesticide residues in citrus peel.At present,the commonly used detection methods of citrus pesticide residues include gas chromatography and liquid chromatography,which have the disadvantages of complex operation and low detection speed.In this paper,a detection method of mixed pesticide residues of Prochloraz and Imazalil in citrus epidermis by surface enhanced Raman spectroscopy was proposed.Taking Prochloraz and Imazalil pesticides commonly used in citrus as the research object,SERS technology combined with stoichiometry methods were applied to collect pesticide spectra in different gradients to realize the qualitative and quantitative analysis of the Prochloraz,Imazalil and the mixture of them in citrus peel.The main research contents and conclusions are as follows:（1）The optimal experimental conditions were determined for the qualitative analysis of the standard solutions of Prochloraz and Imazalil,and their detection limits were explored under the experimental conditions.The best enhancer aiming to serve as the active base was selected by comparing the enhancement effects of gold and silver sol on Prochloraz standard solution and Imazalil standard solution.For the sake of achieving the goals of qualitative analysis for the Prochloraz and Imazalil,the standard solutions of Prochloraz and Imazalil and their solid powders were tested,and the characteristic peaks were assigned and identified.With the condition that gold sol selected as the base,the most suitable volume ratio of Prochloraz and Imazalil standard solution was determined,and moreover the optimal concentration of the agglomerating agent Na Cl was determined.Through comparative analysis,the most suitable volume ratio of Prochloraz and Imazalil standard solution to the gold sol was determined to be100:100.According to the optimal ratio,1mol/L Na Cl solution was added to promote the reaction between the pesticide molecules of Prochloraz,Imazalil and gold sol,which could obviously increase the Raman signal.The detection limits of Prochloraz standard solution and Imazalil standard solution were measured combined with stoichiometry methods.The results were lower than 1mg/L and 0.5mg/L respectively,which were lower than the maximum residue limit of citrus crops pesticides stipulated by the state.（2）Quantitatively detection of Prochloraz and Imazalil standard solutions in citrus peel was operated on the basis of the above optimum experimental conditions.The SVR regression model was established after different preprocessing for the full spectrum data of Prochloraz and Imazalil in citrus peel with different concentrations.However,the results showed that the SVR regression model had an unsatisfactory prediction effect for the reason that the existence of massive invalid information in full spectrum caused interferences with the prediction of the model.Aiming to establish the quantitative regression model by exploring the appropriate characteristic peaks of Prochloraz and Imazalil,the SVR model for the original spectra of different characteristic peaks of the two pesticides was established.The obtained relatively good characteristic peaks of Prochloraz and Imazalil were 829cm^-1and 845cm^-1 respectively,which could be used to establish the subsequent regression model.On this foundation,the optimal quantitative model of Prochloraz pesticide in citrus peel was determined.Combining the optimal preprocessing method MSC with the support vector regression（SVR）optimized by particle swarm optimization（PSO）,a quantitative regression model was established.Its Rp was 0.918 and RMSEP was 1.807mg/L.Besides,the optimal quantitative model of Imazalil pesticide in citrus peel was determined.Combining the optimal preprocessing method 2ND with the support vector regression（SVR）optimized by Gray Wolf algorithm（GWO）,a quantitative regression model was established.Its Rp was 0.944and RMSEP was 2.874mg/L.（3）A quantitative analysis model of mixed Prochloraz and Imazalil pesticide in citrus peel extract based on silver sol was established.In order to achieve the purpose of qualitative analysis of mixed pesticides,the characteristic peaks of Prochloraz and Imazalil in mixed pesticides were identified.The SVR regression model was established after different preprocessing for the full spectrum data of Prochloraz and Imazalil mixed pesticides in citrus peel with different concentrations.The result showed that the SVR regression model had an ideal prediction effects.The Rp of the full spectrum after Baseline optimized was 0.860,however its RMSEP was higher than the original spectrum,and the prediction error was larger.In order to achieve better results,the SVR models established under the original spectra of different characteristic peaks were compared.The characteristic peaks of 829cm^-1and 1168cm^-1 with better modeling effect were finally selected for subsequent model establishment.After comparing the modeling effects of SVR,GWO-SVR,PSO-SVR and GA-SVR under optimal preprocessing conditions,the effect of 1ST preprocessing method combined with grey GWO-SVR model was proved to achieve the best effects,and its Rp and RMSEP were0.967 and 2.227mg/L respectively.In conclusion,the combination of SERS technology and chemometrics can realize the rapid and accurate detection for mixed pesticide residues of Prochloraz and Imazalil in citrus peel.