Sdudy On The Detection Of Pesticide Residues By Silver Nanowire Based On Surface-enhanced Raman Spectroscopy

As the problem of food safety is becoming the focus of society, the detection of pesticide residue has become a research hotspot to a higher degree. This paper based on the technology of surface-enhanced Raman spectroscopy(SERS), selecting organophosphorus pesticide(phosmet and dimethoate) as the research object. The study combined with chemometrics algorithms, focusing on the research of silver nano material in the application of pesticide residue detection. Using different SERS substrate, acquisition phosmet, dimethoate, and mixing pesticides(phosmet and dimethoate mixed) surface enhanced Raman spectroscopy.Spectral data acquired were qualitative and quantitative analyzed. The main results of the thesis were introduced as follows:(1) Silver nanowire(Ag NWs) was used as SERS substrates to get SERS of phosmet,dimethoate pesticide solution samples. Combined with the partial least square(PLS), use different pretreatment methods to establish a model for SERS spectra of standard samples.Deal the spectra of the standard phosmet sample with the first derivative data preprocessing and get the best effect. As showed above, the correlation coefficient of prediction(Rp) was0.994, and the root mean square error of prediction(RMSEP) was 1.068 mg/L. Deal the spectra of dimethoate with smoothing process and the first derivative preconditioning to get the best effect. The Rp was 0.956, and the RMSEP was 3.293 mg/L.(2) Ag NWs was used to get the enhanced spectroscopy of pesticide residued on the surface of orange, and the pesticide included phosmet and dimethoate. Developed quantitative models by PLS regression combined with different data preprocessing methods and compared the advantages and disadvantages of the model. The results showed that the model built by PLS combined with the second derivatives data preprocessing was ideal for phosmet which the Rp was 0.930 and the RMSEP was 3.169 mg/L. It’s optical to achieve the model combined with the multiplicative scatter correction(MSC) was ideal for dimethoate with the Rp was0.957 and the RMSEP was 2.969 mg/L. In order to ensure each characteristic peak of the pesticides to be analyzed, so we selected bands according to the location of the characteristic peak of phosmet and dimethoate. Analysis of spectral selected were conducted after the filter processing. The results showed that the model combined with the MSC data preprocessing was ideal for phosmet, the Rp was 0.911, the RMSEP was 4.317 mg/L; the model combined with the second derivatives was ideal for dimethoate which the Rp was 0.853, the RMSEP was5.159 mg/L.(3) The research object of mixing pesticides(phosmet and dimethoate mixed) on the surface of navel orange were involved to get further results. Although interfered each other of the two pesticides, good effects were got with Ag NWs on the pesticides. The best prediction model was achieved by PLS with the Rp of 0.954 and the RMSEP of 4.822 mg/L using the second derivative data preprocessing. According to the location of the characteristic peak of phosmet and dimethoate, the spectral feature bands were selected. The results showed that the model combined with the MSC data preprocessing method was ideal for phosmet with the Rp was 0.899 and the RMSEP was 6.621 mg/L respectively. All the same, the model combined with the second derivatives was ideal for dimethoate which the Rp was 0.911 and the RMSEP was 7.369 mg/L.The Klarite substrates were prepared for obtaining the SERS of mixing pesticides on the surface of navel orange. The original spectral data of the pesticide were through pretreatment by the different pretreatment methods and PLS. The results showed that it is optimal to apply the model combined with the smoothing and first derivative data preprocessing with the Rp was 0.964 and the RMSEP was 4.599 mg/L respectively. According to the location of the characteristic peak, the selection bands were selected. The results showed that the model combined with the MSC data preprocessing method was ideal for phosmet. The best model had the Rp was 0.899 and the RMSEP was 6.964 mg/L. It is ideal for the dimethoate model combined with the baseline method that the Rp was 0.851 and the RMSEP was 9.897 mg/L...