当前位置: X-MOL 学术Food Control › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Fluorescence spectroscopy for discrimination of botrytized wines
Food Control ( IF 5.6 ) Pub Date : 2018-06-01 , DOI: 10.1016/j.foodcont.2017.12.033
Jana Sádecká , Michaela Jakubíková , Pavel Májek

Abstract Some botrytized wines with “protected designation of origin” have a high market price, thus they are prone to adulteration with cheaper alternatives. This work presents the use of fluorescence spectroscopy combined with chemometrics as a relatively fast and inexpensive tool to discriminate botrytized wines according to two classification criteria: (1) distinguishing between botrytized wines of different quality, namely four-, five-, and six butt wines, and essence; and (2) distinguishing between unadulterated and adulterated samples. Various emission and synchronous fluorescence spectra were recorded and compressed by principal component analysis (PCA) and then linear discriminant analysis (LDA) was performed. The best PCA-LDA results (the percentage of correct classification for each wine category in the prediction step) were obtained with fluorescence spectra recorded on raw samples. Regarding wines of different quality, four- and five butt wines as well as essences were 100% correctly classified, while six butt wine samples were 80% correctly classified using emission spectra excited at 390 or 460 nm as well as synchronous fluorescence spectra recorded at wavelength difference of 100 nm. Regarding unadulterated and adulterated samples, the percentages of correct classification were 60, 80, 80 and 100% for four-, five-, and six butt wines and essence, respectively, while adulterated samples were 100% correctly classified, in all cases, using synchronous fluorescence spectra recorded at wavelength difference of 100 nm.

中文翻译:

用于鉴别贵腐葡萄酒的荧光光谱法

摘要 一些具有“原产地保护名称”的贵腐酒市场价格较高,因此容易被更便宜的替代品掺假。这项工作介绍了使用荧光光谱结合化学计量学作为一种相对快速且廉价的工具来根据两个分类标准来区分贵腐葡萄酒:(1)区分不同质量的贵腐葡萄酒,即四、五和六支酒, 和本质; (2) 区分未掺假和掺假样品。通过主成分分析(PCA)记录和压缩各种发射和同步荧光光谱,然后进行线性判别分析(LDA)。最好的 PCA-LDA 结果(预测步骤中每个葡萄酒类别的正确分类百分比)是通过记录在原始样品上的荧光光谱获得的。对于不同品质的葡萄酒,4 和 5 干酒以及香精被 100% 正确分类,而使用 390 或 460 nm 激发的发射光谱以及在波长记录的同步荧光光谱,6 个干酒样品被正确分类 80%相差 100 nm。对于未掺假和掺假样品,四、五和六酒和香精的正确分类百分比分别为 60%、80%、80% 和 100%,而掺假样品在所有情况下都 100% 正确分类,在所有情况下,使用在 100 nm 波长差处记录的同步荧光光谱。
更新日期:2018-06-01
down
wechat
bug