Chemistry Department, Universidade Federal de Minas Gerais, 31270-901 Belo Horizonte, MG, Brazil.
Chemistry Department, Universidade Federal de Minas Gerais, 31270-901 Belo Horizonte, MG, Brazil; Instituto Nacional de Ciência e Tecnologia em Bioanalítica (INCT Bio), 13083-970 Campinas, SP, Brazil.
Food Chem. 2022 Oct 15;391:133258. doi: 10.1016/j.foodchem.2022.133258. Epub 2022 May 20.
A recent case of contamination of some batches of a Brazilian beer brand with diethylene glycol (DEG) had great repercussion, resulting in at least seven deaths. In this article, a direct method was developed for the rapid detection of DEG in beer samples based on portable near-infrared spectroscopy combined with partial least squares discriminant analysis (PLS-DA). The discriminant model was built with 100 uncontaminated beer samples and 100 samples containing DEG in a concentration range between 10 and 1000 mg L, totalizing 200 samples of different brands and styles. The method was validated by estimating figures of merit, such as false positive and false negative rates, sensitivity, specificity, accuracy, accordance, and concordance. The decision limit (CCα) of the method was 52 mg L and the detection capability (CCβ) was 106 mg L. This method does not consume reagents/solvents and can be suitable for the beer industry quality control or forensic investigations.
最近，一些批次的巴西啤酒品牌被二甘醇（DEG）污染，引起了极大的反响，导致至少 7 人死亡。本文建立了一种基于便携式近红外光谱结合偏最小二乘判别分析（PLS-DA）的快速检测啤酒中 DEG 的直接方法。该判别模型使用 100 个未受污染的啤酒样品和 100 个含有浓度在 10 到 1000 mg L 之间的 DEG 的样品构建，总共有 200 个不同品牌和风格的样品。通过估计误报率和漏报率、灵敏度、特异性、准确性、一致性和符合率等衡量标准对方法进行了验证。该方法的决策限（CCα）为 52 mg L，检测能力（CCβ）为 106 mg L。该方法不消耗试剂/溶剂，适用于啤酒行业的质量控制或法医调查。
