Selective Separation And Analysis Of Pesticide Residues In Tea By Molecular Imprinting And Matrix Solid Phase Dispersion

In recent years, with the strengthen of food safety supervision and management and increasing of public healthy life, "pesticide residues" has been one of the social sensitive topics. And similarly, this problem in tea has troubled every link of the whole plant production-sale-consumption chain. It’s of great importance to verify the identity of "pesticide residues’problem" for the whole tea industry, by establishing high selective separation and analysis method for accurate qualitative and quantitative analysis of pesticide residues in fresh tea leaf, dried tea and tea soup.In this thesis, we improved the method of normal matrix solid phase dispersion (MSPD) on the sample pretreatment of fresh tea leaf and dried tea for the analysis of 23 pesticide residues, based on the deep development of GC-QQQ-MS/MS MRM method for the accurate analysis of 23 kinds of organochlorine and pyrethroid pesticides. By the combination of MIT and pesticide residue analysis, we prepared magnetic molecularly imprinted microspheres (MIMs) for the selective separation of dicofol in tea soup.Firstly, the methods of GC-ECD, GC-MS Scan and GC-MS SIM were established to analyze 23 kinds of pesticides standard mixture simultaneously. Then, the GC-QQQ-MS/MS MRM method was developed and the results showed that it realized the effective combination of accurate qualification and sensitive quantification with low baseline noise, less interference and high sensitivity. The limit of detection (LOD) of this method reached to 10-15 level and the whole results were under 10-12 level. So, this method maybe will become the mainstream method of pesticide residue determination in the future.Secondly, the normal MSPD method was improved by ultrasonic extraction before grinding and column chromatography for purification. It could effectively separate interfering substances from fresh tea leaf and dried tea matrix, such as pigment and so on. The recoveries (R) of method ranged between 62.10% to 115.83%, in addition to β-HCH, cypermethrin and fenvalerate when added with low amount. The relative standard deviation (RSD) of different adding standard recoveries was under 20.62%, the in day RSD was less than 4.91% and the different day RSD less than 7.84%. There was no interference of reagent blank during analysis. This method was proved to be applicable in the dried tea samples by comparison with the national standard method.Thirdly, the matrix effect (ME) was discussed. Because of the other chemical ingredients in fresh tea leaves, the 23 kinds of pesticides showed positive ME during GC-ECD detection, while during the GC-QQQ-MS/MS MRM detection they showed different positive ME or negative ME. To effectively eliminate ME effect, matrix calibration method (MC) was applied for internal standard quantification. Among the concentration range of 0.01mg/L to 2.00mg/L, the linear correlation coefficient (r) of analytes were all above 0.99960 with LOD under 0.00191ng and LOQ under 0.00636ng, except deltamethrin which were 0.14514ng in LOD and 0.48379ng in LOQ.Fourthly, the magnetic molecularly imprinted microspheres (MMIMs) was prepared in water using Fe3O4 nanoparticles as magnetic "core" which were modified with oleic acid in the surface. The scanning electron microscope (SEM) was used to observe the surface morphology. The X-ray diffractionit (XRD)was used to characterize the crystal structure. The FT-IR was used to analyze the imprinting bonding mechanism in polymers. The hysteresis loop confirmed they were superparamagnetic and the thermogravimetry (TG) analysis showed that they were thermal stable and constituted of inorganic and organic parts.Finally, the problem of template leaking was solved effectively by dummy template technology and Soxhlet extraction elution method. The adsorptive properties of Fe3O4@MIMs were discussed through adsorption kinetics and adsorption thermodynamics. Until 120 minutes, the adsorption could reach relative equilibrium and the adsorption capacity to dicofol nonlinearly increased with its initial concentration. By Scatchard analysis, we found that the binding sites of these polymers were not uniform and there were two different types. According to the analysis of imprinting factor and selectivity factor, the effect of imprinting-recognition was depended on not only the interaction between molecules and functional groups in cavities, but also the matching of cavities with the size and shape of molecular functional groups. At last, we applied these imprinted polymers to adsorb and detect dicofol in tea soup successfully and we verified the accuracy of this method by methodology validation.