Hamburg School of Food Science, Institute of Food Chemistry, University of Hamburg, Grindelallee 117, 20146 Hamburg, Germany.
J Agric Food Chem. 2020 Dec 9;68(49):14374-14385. doi: 10.1021/acs.jafc.0c02334. Epub 2020 Jul 2.
The aim of this study was to develop a protocol for the authentication of truffles using inductively coupled plasma mass spectrometry. The price of the different truffle species varies significantly, and because the visual differentiation is difficult within the white truffles and within the black truffles, food fraud is likely to occur. Thus, in the context of this work, the elemental profiles of 59 truffle samples of five commercially relevant species were analyzed and the resulting element profiles were evaluated with chemometrics. Classification models targeting the species and the origins were validated using nested cross validation and were able to differentiate the most expensive from any other examined truffle. For the black truffles, an overall classification accuracy of 90.4% was achieved, and, most importantly, a falsification of the expensive by could be ruled out. With regard to the geographical origin, for Italy and Spain, one-versus-all classification models were calculated each to differentiate truffle samples from any other origins by 75.0 and 86.7%, respectively. The prediction was still possible according to an internal mathematical normalization scheme using only the element ratios instead of the absolute element concentrations. The established authentication protocol was successfully tested with an external sample set of five fresh truffles. Our results show the high potential of the element profile for the parallel species and origin authentication of truffles.
本研究旨在开发一种使用电感耦合等离子体质谱法鉴定松露的方法。不同松露品种的价格差异很大，而且由于白松露和黑松露内部的视觉差异难以区分，因此很可能发生食品欺诈。因此，在这项工作的背景下，分析了 59 个商业相关品种的松露样本的元素谱，并使用化学计量学对所得元素谱进行了评估。针对物种和起源的分类模型使用嵌套交叉验证进行了验证，并能够区分最昂贵的品种与任何其他检查过的品种。对于黑松露，总体分类准确率达到 90.4%，最重要的是，可以排除昂贵的 被假冒的可能性。关于地理起源，对于意大利和西班牙，每个国家都计算了一对一的所有分类模型，以分别将松露样本与任何其他产地的样本区分开来，准确率分别为 75.0%和 86.7%。根据仅使用元素比而不是绝对元素浓度的内部数学归一化方案，仍然可以进行预测。该鉴定方案已成功应用于一组五个新鲜松露的外部样本集进行了测试。我们的结果表明，元素谱在松露的平行物种和起源鉴定方面具有很高的潜力。
