Abstract This work proposes a novel method for determining the composition of mixtures of natural polyphenolic compounds: caffeic acid, gallic acid, ellagic acid, rosmarinic acid and quercitrin in plants. The method is based on the formation of colored spots by these compounds upon reaction with nano-oxides of Al2O3, ZnO, MgO, CeO2, TiO2 and MoO3 impregnated on filter paper and constituting a colorimetric sensor array (CSA). The image of the colored spots was analyzed and the intensity of the blue colour (BCI) component has shown maximum sensitivity in relation to phenolic compounds. The inverse of BCI was linearly correlated with the logarithm of the individual phenolic compound concentrations. Chemometric analysis by partial least squares regression (PLSR) of 1/BCI values for 24 synthetic mixtures of the 5 phenolic compounds measured with the colorimetric sensor array has demonstrated good correlation between the actual and the predicted concentration of quercitrin. For the other phenolic compounds, the colors measured with the colorimetric sensor array were greatly influenced by the concentrations of the other components in the mixture. The method was applied to the determination of quercitrin in medicinal teas and the results were compared to those obtained by HPLC. The discussion of the results emphasizes possible interferences in the tea samples. While further optimization of the colorimetric sensor array-based method appears necessary, tailored to the particular targeted application in real samples, the proposed method for polyphenol determination has advantages that include simplicity, low cost, and portability.

中文翻译：

---

用于测定具有抗氧化特性的植物多酚的基于金属纳米氧化物的比色传感器阵列

摘要 本工作提出了一种测定植物中天然多酚化合物混合物组成的新方法：咖啡酸、没食子酸、鞣花酸、迷迭香酸和槲皮素。该方法基于这些化合物与浸渍在滤纸上的 Al2O3、ZnO、MgO、CeO2、TiO2 和 MoO3 的纳米氧化物反应后形成有色斑点，并构成比色传感器阵列 (CSA)。分析了彩色斑点的图像，蓝色 (BCI) 成分的强度显示出与酚类化合物相关的最大灵敏度。BCI 的倒数与单个酚类化合物浓度的对数呈线性相关。使用比色传感器阵列测量的 5 种酚类化合物的 24 种合成混合物的 1/BCI 值的偏最小二乘回归 (PLSR) 化学计量分析表明，槲皮素的实际浓度和预测浓度之间具有良好的相关性。对于其他酚类化合物，使用比色传感器阵列测量的颜色受混合物中其他成分浓度的影响很大。将该方法应用于药茶中槲皮苷的测定，并与高效液相色谱法得到的结果进行比较。结果的讨论强调了茶样品中可能存在的干扰。虽然进一步优化基于比色传感器阵列的方法似乎是必要的，但需要针对实际样品中的特定目标应用进行定制，