Scholarlink

<p>Six food dyes were monitored in different food matrices within minimum experimental efforts. Sunset yellow, allura red, tartrazine, carmoisine, brilliant blue, and brilliant black were accurately detected in food samples down to 5–10&#xa0;µg/g using second-order multivariate calibration after a simple sample cleanup method. Using the acid/base characteristics of dyes, second-order data was generated from absorbance-pH measurements following a batch-experiment approach. The experiment was conducted in the spectral domain of 330–700&#xa0;nm and over pH 2.0–12. Unfolded partial least squares-residual bilinearization U-PLS/RB outperformed parallel factor analysis PARAFAC for dye quantification, which is attributed to the rank deficiency of pH-spectral data. Figures of merit of the proposed analytical method were estimated using matrix-based standards over the range 10–200&#xa0;µg/g. The efficacy of U-PLS/RBL in resolving dye signals from other food ingredients is credited with its success. The rank deficiency in the gathered matrices (for test samples), linear dependency in the pH dimension, and intense spectrum overlapping in the spectral mode all contributed to the modest performance of PARAFAC calibration. The performance of U-PLS/RBL is compared to that of other published approaches. The proposed analytical method was used to monitor dyes in highly consumed candies.</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

Spectroscopic Detection of Colouring Agents in Food Matrices Using Second Order Multivariate Calibration

  • 期刊论文

  • 发表日期：

    2026年6月3日

查看全文
  • Khaled Al-Zawahreh,
  • Ahmad B. Albadarin

摘要

Six food dyes were monitored in different food matrices within minimum experimental efforts. Sunset yellow, allura red, tartrazine, carmoisine, brilliant blue, and brilliant black were accurately detected in food samples down to 5–10 µg/g using second-order multivariate calibration after a simple sample cleanup method. Using the acid/base characteristics of dyes, second-order data was generated from absorbance-pH measurements following a batch-experiment approach. The experiment was conducted in the spectral domain of 330–700 nm and over pH 2.0–12. Unfolded partial least squares-residual bilinearization U-PLS/RB outperformed parallel factor analysis PARAFAC for dye quantification, which is attributed to the rank deficiency of pH-spectral data. Figures of merit of the proposed analytical method were estimated using matrix-based standards over the range 10–200 µg/g. The efficacy of U-PLS/RBL in resolving dye signals from other food ingredients is credited with its success. The rank deficiency in the gathered matrices (for test samples), linear dependency in the pH dimension, and intense spectrum overlapping in the spectral mode all contributed to the modest performance of PARAFAC calibration. The performance of U-PLS/RBL is compared to that of other published approaches. The proposed analytical method was used to monitor dyes in highly consumed candies.