摘要
采用微波消解结合电感耦合等离子体-质谱(inductively coupled plasma-mass spectrometry,ICP-MS)测定了宁夏和青海2个地区共180个枸杞样品中44种微量元素的含量。通过元素含量进行过滤,将具有显著性差异的9种元素(Sb、La、Tb、Lu、Al、Sc、V、Cr、Se)进行主成分分析,前2个主成分可以解释64.2%的变量,2个产地的枸杞样品基本可以分开。以9种元素为基础,应用偏最小二乘判别分析(partial least squares discrimination analysis,PLS-DA)和反向传输人工神经网络(back propogation artificial neural network,BP-ANN)2种算法分别建立宁夏枸杞和青海枸杞的判别模型。结果显示:在PLS-DA模型中,全部样品建模时,模型的灵敏度和特异性分别为100%和97.5%,75%的枸杞样品建模,模型的灵敏度和特异性分别为98.6%和98.4%,模型对25%样品预测的准确性达到100%;在BP-ANN模型中,全部样品建模和75%的枸杞样品建模,模型的灵敏度和特异性均为100%,模型对25%样品的预测的准确性达到100%,得出BP-ANN模型的灵敏度和特异性优于PLS-DA模型。应用ICP-MS测定枸杞中多种元素含量,结合化学计量学方法可以快速判别宁夏枸杞和青海枸杞。
Microwave digestion and inductively coupled plasma mass spectrometry(ICP-MS)were used to determine 44 trace elements in 180 Lycium barbarum L.samples from Ningxia and Qinghai provinces.Nine elements(Sb,La,Tb,Lu,Al,Sc,V,Cr and Se)with significant differences were selected for PCA by element content screening.The results showed that the first two main components could explain 64.2%of the variable,meanwhile the L.barbarum L.samples could be basically distinguished from Ningxia and Qinghai.Based on nine elements with significant differences,the discriminant models of L.barbarum L.from Ningxia and Qinghai were established by partial least squares discriminant analysis(PLS-DA)and back propagation artificial neural network(BP-ANN).In the PLS-DA model,when 100%L.barbarum L.samples were used,the sensitivity and specificity of the model were 100%and 97.5%,respectively.When 75%L.barbarum L.samples were used,the sensitivity and specificity of the model were 98.6%and 98.4%,respectively,and the accuracy of the model was 100%for predicting the remaining 25%L.barbarum L.samples.In the BP-ANN model,when 100%and 75%L.barbarum L.samples were used,the specificity and sensitivity of the model were both 100%.The accuracy of the model was 100%for predicting the remaining 25%L.barbarum L.samples.The sensitivity and specificity of BP-ANN model were better than PLS-DA model.The results showed that the determination of multiple elements in L.barbarum L.by ICP-MS combined with chemometrics could quickly identify L.barbarum L.from Ningxia and Qinghai.
作者
连思雨
谢瑜杰
张紫娟
范春林
王明林
陈辉
LIAN Siyu;XIE Yujie;ZHANG Zijuan;FAN Chunlin;WANG Minglin;CHEN Hui(Chinese Academy of Inspection and Quarantine,Beijing 100176,China;College of Food Science and Engineering,Shandong Agricultural University,Tai’an 271018,China)
出处
《食品与发酵工业》
CAS
CSCD
北大核心
2020年第13期250-254,共5页
Food and Fermentation Industries
基金
特色高值农产品产地判别技术研究(2017YFF0211302)。
