基于聚类分析的白茶高效液相色谱指纹图谱研究

采用高效液相色谱法对白茶样品中儿茶素类和生物碱类成分进行分析，建立白茶的特征性指纹图谱，为白茶的质量评价提供参考依据。以福鼎、政和等地的多种类白茶为样品，经提取溶剂、流动相、波长、流速、柱温等条件筛选，确定白茶指纹图谱高效液相色谱条件为:用70%甲醇在70 ℃条件下进行提取，色谱柱为Agilent Eclipse Plus C₁₈（4.6 mm×250 mm，5 μm），以乙腈-0.5%乙酸为流动相进行梯度洗脱，检测波长258 nm，柱温35 ℃，进样量为10 μL；咖啡碱作为参照峰，计算共有峰相对保留时间和相对峰面积；选取中药色谱指纹图谱相似度评价系统（2012版）进行相似度评价；采用系统聚类分析法（Hierarchical cluster analysis，HCA）对其进行聚类分析。建立了白茶高效液相色谱法的指纹图谱，确认了28个共有峰；相似度评价结果显示，12批不同品种白茶相似度在0.98以上，表明不同年份不同产地的白茶样品成分组成一致；对没食子酸（Gallic acid，GA）、咖啡因（Caffeine，CAF）、表儿茶素（Epicatechin，EC）、表儿茶素没食子酸酯（Epicatechin gallate，ECG）、表没食子儿茶素没食子酸酯（Epigallocatechin gallate，EGCG）、表没食子儿茶素（Epigallocatechin，EGC）六个成分进行了峰指认；系统聚类分析结果表明，存放时间对白茶成分的组成及含量有一定的影响，聚类分析将测定的白茶样品分为两类，2013、2014年为一类，2017~2020年为一类。该方法方便、快捷、简单、准确可靠，可为白茶的质量控制提供参考依据。

HPLC Fingerprints of White Tea Based on Cluster Analysis

The catechins and alkaloids in white tea samples were analyzed by high performance liquid chromatography (HPLC), and the characteristic fingerprint of white tea was established to provide reference for the quality evaluation of white tea. Taking many kinds of white tea from Fuding, Zhenghe and other places as samples, through the selection of extraction solvent, mobile phase, wavelength, flow rate and column temperature, the HPLC conditions of white tea fingerprint were determined as follows: 70% methanol, extracted at 70 ℃, the chromatographic column was Agilent Eclipse Plus C₁₈ (4.6 mm×250 mm, 5 μm), the gradient elution was carried out with acetonitrile ?0.5% acetic acid as the mobile phase, the detection wavelength was 258 nm, the column temperature was 35 ℃, and the injection volume was 10 μL. Caffeine as a reference peak, the relative retention time and relative peak area of the common peak were calculated. The similarity evaluation system of chromatographic fingerprint of traditional Chinese medicine(version 2012) was selected to evaluate the similarity, and the systematic cluster analysis method (HCA) was used for cluster analysis.The fingerprint of white tea was established, and 28 common peaks were identified in the fingerprint. The similarity evaluation results showed that the similarity of 12 different varieties of white tea was above 0.98, which indicated that the composition of white tea samples from different years and different habitats were consistent. Gallic acid (GA), caffeine (CAF), epicatechin (EC), epigallocatechin gallate (ECG), epigallocatechin gallate (EGCG) and epigallocatechin (EGC) were identified by peak identification. The results of hierarchical cluster analysis showed that the storage time had a certain effect on the composition and content of white tea, and these white tea samples were divided into two categories: 2013, 2014 and 2017~2020. The method was convenient, fast, simple, accurate and reliable. It could provide a reliable evaluation method for quality control of white tea.