【作者】 赵海燕；

【导师】 魏益民；

【作者基本信息】 中国农业科学院 ， 农产品质量与食物安全， 2013， 博士

【摘要】 矿物元素指纹分析技术已被广泛应用到农产品的产地溯源中，研究农产品产地矿物元素指纹信息的成因及其在年际间的稳定性，可为农产品产地矿物元素指纹溯源技术的应用提供理论依据。本研究选用小麦作为试验的模式生物，于2008年和2009年小麦收获期在河北、河南、山东和陕西4省随机采集240份小麦籽粒样品；利用电感耦合等离子体质谱（ICP-MS）测定样品中多种矿物元素的含量。结合单因素方差分析、主成分分析（PCA）、判别分析（LDA），探讨矿物元素指纹分析技术对小麦产地溯源的可行性。于2010年和2011年小麦播种期在河北省赵县、河南省辉县和陕西省杨凌区3个地点种植10个小麦品种，进行田间试验，得到180份小麦籽粒样品。利用高分辨率电感耦合等离子体质谱（HR-ICP-MS）检测样品中55种矿物元素的含量。结合多因素方差分析，分析地域、基因型、年际及其交互作用对小麦籽粒矿物元素指纹信息的影响；通过计算各因素对各元素含量变异的方差与总变异方差的比值，解析各因素对小麦籽粒矿物元素含量变异的贡献率；筛选与地域密切相关的元素。于2010年小麦收获期在河北省和河南省随机采集61对小麦及其产地的土壤样品，利用ICP-MS和X射线荧光光谱（XRF）检测样品中55种矿物元素的含量。结合相关分析，分析表层土壤和母质土壤中矿物元素含量对小麦籽粒矿物元素指纹信息的影响；筛选与表层或母质土壤密切相关的元素。以在河北、河南、山东和陕西4省随机采集的2008/2009年度120份小麦籽粒样品为试验材料，结合PCA、LDA，检验已筛选的溯源指纹信息对非试验环境样品产地的鉴别效果。最终明确小麦籽粒矿物元素指纹信息的成因；筛选出准确、稳定的产地溯源指纹信息。主要结论如下：(1)不同地域来源的小麦籽粒矿物元素含量存在显著差异；矿物元素指纹分析技术可用于小麦产地溯源。(2)小麦籽粒中元素Ca、Mn、Zn、Rb、Sr、Mo、Cd、Cs含量与地域密切相关；元素Cu含量与基因型密切相关；元素Na、Mg、Al、Ti、V、Cr、Fe、Co、Ga、Se、Y、Zr、Sn、Eu、U含量与年际密切相关。(3)小麦籽粒中元素Cr、Mn、Ga、Rb、Sr、Zr、Cd含量与表层土壤中相应元素的总含量显著相关；元素Na、Mn、Cd、Sn、Ba含量与母质土壤中相应元素的总含量显著相关。(4)与地域密切相关的元素Ca、Mn、Zn、Rb、Sr、Mo、Cd、Cs对非试验环境小麦样品的产地鉴别效果较好，是小麦产地溯源的理想指纹信息。(5)以小麦为试验模式生物的矿物元素指纹产地溯源模型试验及其研究结果，可为农产品产地溯源技术体系的建立提供参考和借鉴。更多还原

【Abstract】 Multi-element fingerprinting technique has been successfully applied to discriminate geographicalorigins of various agricultural products. The main objective of this dissertation was to study the formingreason of multi-element origin fingerprints of agricultural products and their stability among differentyears. It could provide theoretic basis for the practical application of identifying the geographical originof agricultural products by multi-element fingerprinting technique. Wheat was chosen as theexperimental model organism in this study. A total of240wheat samples were collected randomly fromHebei, Henan, Shandong and Shaanxi provinces during the2008and2009harvest time, respectively. Theconcentrations of multi-elements in wheat kernel samples were determined by inductively coupledplasma mass spectrometry (ICP-MS). Combined with one-way analysis of variance, principal componentanalysis (PCA) and linear discriminant analysis (LDA), the feasibility of multi-element analysis inidentifying the geographical origin of wheat was discussed. Field experiments were conducted at threedifferent locations: Zhaoxian (Hebei province), Huixian (Henan province), and Yangling (Shaanxiprovince). Ten wheat varieties were cultivated on each of three agricultural fields during the2010and2011wheat seeding time, respectively.180wheat samples were obtained from the model experiment. Theconcentrations of55elements in wheat kernel samples were determined by high resolution inductivelycoupled plasma mass spectrometry (HR-ICP-MS). Combined with multi-way analysis of variance, theinfluence of location, genotype, harvest year and their interactions on multi-element fingerprints of wheatkernel were analyzed. The contributions of location, genotype and harvest year to the observed elementcontent variability were calculated by the variance component ratio σ2i/σ2total(σ2i=square sum of varianceof certain factor; σ2total=total square sum of variance). The element fingerprinting information closelyrelated to geographical origin was selected.61pairs of wheat and soil samples were randomly collectedfrom Hebei and Henan provinces during the2010harvest period. The concentrations of55elements inwheat kernel and their provenance soil samples were analyzed by HR-ICP-MS and X-ray fluorescence(XRF). The influence of multi-element contents of topsoil and parent soil samples on those of wheatkernel were analyzed, respectively. And the element fingerprinting information closely related to topsoilor parent soil was selected.1202008/2009wheat kernel samples from Hebei, Henan, Shandong andShaanxi provinces were chosen as the experimental materials to examine the origin discrimination effectof the selected fingerprinting information to wheat samples from non-experimental environment.Through above analysis, the forming reason of wheat multi-element origin fingerprints was made clear.The accurate and stable origin traceability fingerprinting information was selected.The main conclusions were as follows:(1) The multi-element contents in wheat samples from different regions were significantly different. Itwas feasible to classify wheat according to the geographic origin using multi-element fingerprintingtechnique.(2) The origin effect was observed to be strongest on Ca, Mn, Zn, Rb, Sr, Mo, Cd and Cs content variabilities; genotype effect was strongest on Cu content variability; year effect was observed to bestrongest on Na, Mg, Al, Ti, V, Cr, Fe, Co, Ga, Se, Y, Zr, Sn, Eu and U content variabilities.(3) The significant correlations existed for the element Cr, Mn, Ga, Rb, Sr, Zr and Cd contents betweenwheat kernel and topsoil and for the element Na, Mn, Cd, Sn and Ba contents between wheat kernel andparent soil.(4) The origin discrimination effect of the elements Ca, Mn, Zn, Rb, Sr, Mo, Cd and Cs to wheatsamples from non-experimental environment was good. The information contained in these8elementswas perfect for wheat traceability of geographical origin.(5) The study results of the model experiment of geographical origin traceability by multi-elementfingerprinting using wheat as model organism can provide reference for building the technical system oforigin traceability of agricultural products.更多还原