Scholarlink

<p>This study introduces a hierarchical strategy designed to differentiate, classify, and quantify adulterants—namely, sugarcane molasses, inverted sugar, and corn syrup—in commercial honey samples analyzed through TD-NMR. The sample preparation process was simplified to involve only heating, ensuring crystallization was avoided. The DD-SIMCA, a one-class classifier, successfully distinguished between pure and adulterated samples, achieving a specificity of 100.00%. The next step involved predicting the type of adulterant using PLS-DA analysis, which led to classification accuracy exceeding 97.0% across all three adulterants. Furthermore, the PLS model quantified the adulterant levels in the samples, showing RMSEV values below 0.5 (% w w<sup>−1</sup>) and an RPD greater than 6.38 for each adulterant. The elliptic joint confidence region (EJCR) test was conducted to assess the method’s feasibility and reliability. Consequently, the proposed approach highlights the effectiveness of TD-NMR combined with chemometric methods as a valuable tool for verifying the authenticity of commercial honey and preventing fraudulent adulterations.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Hierarchical Identification to Quantification Method to Detect Adulterants in Honey by Time Domain Nuclear Magnetic Resonance (TD-NMR)

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月5日

查看全文
  • Leonardo Francisco Rafael Lemes,
  • Vitória Sofia da Paz,
  • Jéssica Thaís de Lima,
  • Leice Milla Ribeiro de Novais,
  • Caroline Da Ros Montes D’Oca,
  • Frederico Luis Felipe Soares

摘要

This study introduces a hierarchical strategy designed to differentiate, classify, and quantify adulterants—namely, sugarcane molasses, inverted sugar, and corn syrup—in commercial honey samples analyzed through TD-NMR. The sample preparation process was simplified to involve only heating, ensuring crystallization was avoided. The DD-SIMCA, a one-class classifier, successfully distinguished between pure and adulterated samples, achieving a specificity of 100.00%. The next step involved predicting the type of adulterant using PLS-DA analysis, which led to classification accuracy exceeding 97.0% across all three adulterants. Furthermore, the PLS model quantified the adulterant levels in the samples, showing RMSEV values below 0.5 (% w w−1) and an RPD greater than 6.38 for each adulterant. The elliptic joint confidence region (EJCR) test was conducted to assess the method’s feasibility and reliability. Consequently, the proposed approach highlights the effectiveness of TD-NMR combined with chemometric methods as a valuable tool for verifying the authenticity of commercial honey and preventing fraudulent adulterations.