Montemerlo Antonella E, Azcarate Silvana M, Camiña José M, Messina Germán
Instituto de Química de San Luis, "Dr Roberto A. Olsina" (INQUISAL-CONICET-UNSL), San Luis, 5700, Argentina.
Facultad Ciencias Exactas y Naturales, Universidad Nacional de La Pampa, Santa Rosa, 6300, La Pampa, Argentina.
Anal Methods. 2024 Jun 13;16(23):3774-3783. doi: 10.1039/d4ay00536h.
This article explores the application of chemometric tools including multivariate curve resolution with alternating least squares for the simultaneous determination of gallic and caffeic acids on the surface of a glassy carbon electrode without additional modification. Gallic and caffeic acids are primary polyphenols, the most abundant in red wines produced in Argentina, and are often used as quality markers for them. These polyphenols significantly contribute to the organoleptic properties of wines from this origin, but their electrochemical signals overlap significantly, making simultaneous quantification challenging without additional experiments such as electrode modification or alternative analytical techniques beyond differential pulse voltammetry. This study successfully quantified these compounds in complex mixtures by generating second-order data from differential pulse voltammetry experiments conducted at various potential steps and subsequently applying multivariate curve resolution with alternating least squares. The use of constraints during optimization prevented rotational ambiguities common in this modeling, leading to unique results in validation samples. The limits of detection (LOD) found for gallic and caffeic acids were 1.6 and 7.6 mg L, which are in excellent agreement with the expected concentrations of these compounds in red wines. The concentration ranges analyzed showed a linear dependency (between the LOD and 300 mg L) with the signals estimated by the model for both analytes. Advantages such as simplicity, low cost, and high speed, as well as not requiring electrode modification, combined with excellent results obtained for real samples, make it a promising alternative for polyphenol analysis in the wine industry.
本文探讨了化学计量学工具的应用，包括采用交替最小二乘法的多元曲线分辨法，用于在无需额外修饰的玻碳电极表面同时测定没食子酸和咖啡酸。没食子酸和咖啡酸是主要的多酚类物质，在阿根廷生产的红葡萄酒中含量最为丰富，常被用作红葡萄酒的品质指标。这些多酚类物质对源自该产地的葡萄酒的感官特性有显著贡献，但其电化学信号严重重叠，若不进行诸如电极修饰或采用除差分脉冲伏安法之外的其他分析技术等额外实验，同时进行定量分析具有挑战性。本研究通过在不同电位步长下进行差分脉冲伏安实验生成二阶数据，随后应用交替最小二乘法的多元曲线分辨法，成功对复杂混合物中的这些化合物进行了定量分析。优化过程中使用约束条件避免了该建模中常见的旋转模糊性问题，从而在验证样品中得到了唯一结果。没食子酸和咖啡酸的检测限（LOD）分别为1.6和7.6 mg/L，与红葡萄酒中这些化合物的预期浓度高度吻合。所分析的浓度范围（在检测限和300 mg/L之间）与模型估算的两种分析物信号呈线性相关。该方法具有简单、低成本、速度快以及无需电极修饰等优点，同时在实际样品分析中获得了优异结果，使其成为葡萄酒行业多酚分析的一种有前景的替代方法。
