Study On Methods For The Detection Of Tea Geographic Producing Area/Cadmium/Fluorine/Bifenthrin Based On Spectral Technology

The quality and safety of tea affect not only economic benefits but also people's health.Due to the influence of soil elements,fertilizer,light,precipitation,and the management level for a tea garden,the quality of tea varies significantly from place to place.In order to obtain high profits,some unscrupulous businessmen use low-grade tea to fake famous and high-quality tea products,which leads to losses for consumers and production enterprises.How to quickly identify the origin of tea is an urgent problem to be solved for the tea industry.In tea tree growth and tea processing,it is easy to be polluted by heavy metals,pesticides,and other harmful substances,leading to the body's damage.Therefore,tea safety has always been a difficult problem faced by the industry.Therefore,tea detection is in great demand,especially for the methods with the advantages of high-throughput,on-site,fast,and convenient.Update,UV-visible,fluorescence,infrared and Raman spectra have been widely used in food,medicine,chemical industry,energy,and other fields,since they have advantages of a large amount of information,fast detection speed,and convenience,people,therefore,pay more and more attention to them.Aiming at the problems of quality(origin of tea)and safety(heavy metals,agricultural residues,and fluorine)faced by the tea industry,this thesis carries out studies on the development of methods for detecting tea geographic producing area/cadmium/fluorine/bifenthrin based on spectral technology,which mainly includes the following contents.1.study on metal the element characteristics and geographical recognition method for the Zhenjiang tea.Zhenjiang tea had excellent quality and was a favorite for people.However,there are two problems in product quality and safety:(1)excessive heavy metals.(2)Zhenjiang tea was often counterfeited by tea produced by other places.In order to solve the above problems,46 samples of Zhenjiang tea(half young leaves and half old leaves)and 25 samples of non-Zhenjiang tea were collected,and 48 samples of corresponding soil were collected,among which 23 samples were Zhenjiang soil,and25 samples were non-Zhenjiang soil.The X-ray fluorescence(XRF)based high-throughput methods for rapid detection of metal elements in tea and soils were established;metal elements in the corresponding soil were used to interpret the reasons for the increase of heavy metals in tea.Based on the detection results of elements in tea,linear and nonlinear geographical recognition methods were established for the rapid recognition of Zhenjiang tea.The results showed that:(1)the contents of Be,Mg,Ni,and Ti in Zhenjiang tea were higher than those in other areas;for Zhenjiang tea,the concentrations of Cu,Mn,and Ni in tender leaf were higher than those in old leaf.The levels of Mg,Ni,Cu,and Ti in Zhenjiang soil were higher than those in other areas;(2)the qualified rates for Co,and Cr in all samples were high(>90%).The unqualified samples for Cu were mainly from Zhenjiang area,in which the tender leaf tea accounted for the majority.For Ni,the qualified rate of Zhenjiang tea was low to 63.04%),especially for the tender leaf tea,which was the lowest(43.48%);(3)the excessive Ni content in soil was one of the main reasons for the excessive Ni concentration in Zhenjiang tea.Furthermore,the content of Cu in Zhenjiang tender leaf tea was higher than that in soil,and maybe,it was because tender leaves have a higher capability to accumulate Cu.(4)for the linear PLS-DA and nonlinear SVM method,the SVM model had a higher capability for recognizing Zhenjiang tea,and the prediction accuracy for the correction set,cross-validation set,and validation set were 95.74%,82.98%,and 87.50%,respectively.2.Fast determination of bifenthrin in tea by surface-enhanced Raman scatter(SERS)technology.Tea is often polluted by bifenthrin.It is of great significance to develop rapid quantitative and qualitative analysis methods to detect bifenthrin in tea.The gold nanorods(Au NRs)were prepared as the SERS substrate,and then it was characterized.After the bifenthrin polluted tea was prepared,they were pretreated with the SPE column,and the non-target impurities were removed,and then their SERS signals were collected using Au NRs.The wavelength optimization was carried out by using four methods,synergy interval(SI),colony optimization(ACO),competitive adaptive weighted sampling(CARS),and genetic algorithm(GA).The quantitative spectral prediction model was established by partial least square(PLS)regression.The qualitative models were built by linear discriminant analysis(LDA),k-nearest neighbor(KNN),support vector machines(SVM),and backpropagation neural network(BPNN).The results showed that the average length and width of the Au NRs were 27 nm and14 nm,respectively.After the SNV pretreatment for the raw spectra,the GA method screened the spectral wavelength variables with the highest amount of information and has the highest prediction accuracy and the best robustness.The values of R_cv and R_p were0.9710 and 0.9664,respectively,and the corresponding RMSECV and RMSEP were0.2237 and 0.2524,respectively.The LOQ_min and LOQ_max were 0.15mg/kg and0.71mg/kg,respectively.The method's stability was good(RSD=5.16%),and the accuracy was high(relative error was 3.55±1.35%).It demonstrated that the developed method could be used for the rapid quantitative detection of bifenthrin in tea.For the Semi-quantitative analysis method,among the four classification models,the SVM and BPNN had the same highest accuracy of 99.12%,which met the needs for rapid screening and a rough quantity of tea samples.3.Study the development of a method for detecting Cd²⁺in tea based on optical fiber LSPR.Based on the principle that the change of surface refractive index easily shifts the local surface plasma resonance(LSPR)peak of gold nanoparticles,an optical fiber LSPR sensor for the detection of cadmium in food(including tea)was constructed.Firstly,the optical fiber end face was coupled with-SH,which was then connected Au NPs through Au-S bond;finally,the Cd²⁺nucleic acid aptamer was coupled on by Au-S bond to form the sensor.As a result,the optimum preparation process parameters were obtained,including that the concentration of MPTMS was 2.0%,the time of fiber end face connecting Au NPs was 12.0 h,the assembly concentration of Cd²⁺nucleic acid aptamer was 5.0?M,and the p H of the detection system was 7.5.There was a good linear relationship for Cd²⁺concentration range from 0.1 ng/m L to 1000 ng/m L,R²=0.972.For mercury ion,lead ion,and iron ion,the sensor showed excellent selectivity for Cd²⁺and good detection performance for tea samples.The recoveries for 100 ng/m L and 500 ng/m L Cd²⁺in tea digestion solution were 100 ng/m L and 104%,respectively,so the sensor could be used for the detection of Cd²⁺in real samples.The sensor has the advantages of small volume,low cost,and high detection sensitivity,which provided a new idea for the detection of Cd²⁺in tea and other foods.4.study on a method for the detection of F^-content in tea based on quantum dot-inner filter effect.CdTe QDs were mixed in acetonitrile to form a detection system,and F^-and curcumin were formed a complex as an inner-filter,resulting in a significant weakening of the fluorescence of quantum dots,thus realized the quantitative detection of F^-.The results showed that the quantitative detection method had an F^-concentration range of 2?mol/L?10 mmol/L,the linear relationship was good(R²=0.982),and the LOD was 0.122?mol/L.Thus,it has strong anti-interference capability against 9 kinds of common substances,such as dipotassium hydrogen phosphate,sodium sulfite,sodium oxalate,etc.The recoveries of F^-for Shushan white tea samples were 91.9%-97.3%.For the determination of actual white tea,the relative error was 4.51%,compared with the result by the national standard method.The method has the advantages of simple,high speed,high selectivity,strong anti-interference,low requirement for instruments,and can be widely used to determine F^-content in actual tea samples.In general,this study explored and established a set of integrated detection methods with both commonness and individuality for different detection components.