基于矿物元素指纹图谱的黑龙江黄豆产地溯源

该研究探讨了矿物元素指纹分析技术对黑龙江黄豆产地溯源的可行性,筛选出判别黑龙江黄豆产地溯源的有效指标。利用电感耦合等离子体质谱仪(inductively coupled plasma mass spectrometry,ICP-MS)测定来自齐齐哈尔和北安2个地域50份黄豆样品中52种矿物元素的含量,并对数据进行了方差分析、主成分分析和判别分析。研究表明,46种矿物元素含量在地域间存在显著差异,通过逐步判别分析筛选出8项元素指标建立黄豆产地判别模型,所建立的模型对黄豆产地整体交叉检验判别率为95.7%。As、Ru、Gd含量在黄豆与土壤间呈显著正相关(P0.05),Tb含量在黄豆与土壤间呈极显著正相关(P0.01),由4种元素建立的判别模型对产地判别准确。因此,上述元素是黄豆矿物元素产地鉴别较可靠的指纹信息指标。

Origin traceability of Heilongjiang soybean using fingerprint of mineral elements

Abstract: China is a major production country of soybeans. A number of geographical indication products (such as soybeans in the Qiqihar region and soybeans in the Bei''an area) have been formed in long-term climate environments and specific geographical locations. The establishment of source traceability technology for the soybean production with geographical indication provides a strong technical support to improve the reputation of production enterprises and market competitiveness. Identification of the information to be recorded is the most important requirement for developing an effective traceability system. The feasibility analysis with mineral composition fingerprint technique for geographical origin traceability of Heilongjiang soybean and the screening of effective indicators to determine origin place of Heilongjiang soybean were performed in this research. The contents of 52 mineral elements including Na, Mg, Al, and K in 50 soybean samples from Qiqihar and Bei''an were measured using inductively coupled plasma mass spectrometry (ICP-MS), and then variance analysis (VA), principal component analysis (PCA) and discriminant analysis (DA) were performed with the data. The contents of 46 mineral elements of the soybeans significantly differed among regions, which suggested that these mineral components could be used for origin traceability of Heilongjiang soybean. Based on the scree plot and the contribution rate of variance, 8 principal components were extracted by the PCA to establish the origin discrimination model using the DA. The overall criterion rate of crossing examination of the model was 95.7%, which confirmed that the models could accurately discriminate the origin place of soybean. The contents of AI, V, Ag, Cd, Ba, Tb, Pt and Au showed effective judgment on soybean samples from Qiqihar and Bei''an. In order to further explain the relation between mineral element contents in soil and those in soybean, Pearson correlation analysis was carried out. The contents of As, Ru and Gd in soybean were positively related with those in soil, while the content of Tb in soybean was highly positively related with that in soil. The discriminant model formed by 4 mineral elements was proved to discriminate the sources of the whole samples correctly. The correct discrimination rates of the models for Qiqihar and Bei''an were 96.2% and 100% respectively. Therefore, the mineral element fingerprint technology is feasible to determine the origin place of soybean.