基于太赫兹时域光谱技术的黄酮类化合物研究  

Research on Flavonoids Based on Terahertz Time-Domain Spectroscopy

作　　者：殷明[1] 王建林 黄浩亮 黄秋萍 杨萌萌 傅正平 陆亚林 YIN Ming;WANG Jian-lin;HUANG Hao-liang;HUANG Qiu-ping;YANG Meng-meng;FU Zheng-ping;LU Ya-lin(National Synchrotron Radiation Laboratory,University of Science and Technology of China,Hefei 230026,China;Hefei National Laboratory for Physical Sciences at the Microscale,University of Science and Technology of China,Hefei 230026,China;School of Chemistry and Materials Science,University of Science and Technology of China,Hefei 230026,China)

机构地区：[1]中国科学技术大学国家同步辐射实验室,安徽合肥230026 [2]中国科学技术大学合肥微尺度物质科学国家研究中心,安徽合肥230026 [3]中国科学技术大学化学与材料科学学院,安徽合肥230026

出　　处：《光谱学与光谱分析》2020年第12期3919-3924,共6页Spectroscopy and Spectral Analysis

基　　金：国家自然科学基金项目(51627901,11704373);国家重点研发计划项目(2016YFA0401004,2017YFA0402904);中科大重要方向培育基金项目(WK2340000071,WK2310000076)资助。

摘　　要：黄酮类化合物常以游离或糖苷形式广泛存在于植物体内的一大类多酚物质,具有抗氧化、抗菌、抗病毒、抑制肿瘤生长等药理作用,作为一种重要的中药活性成分,其具有较高的药用价值和开发前景。利用太赫兹时域光谱技术(THz-TDS)研究了8种常见的黄酮类化合物(黄芩素、槲皮素、柚皮素、大豆素、黄芩苷、葛根素、染料木素和天麻素)在0.2~2.5 THz波段的生物分子特性,结果表明这些黄酮类物质在太赫兹波段具有明显不同的特征吸收峰,并研究了它们在78~320 K范围内随温度变化的太赫兹吸收特性,结果显示随着温度的降低,特征吸收峰逐渐增强,并且吸收峰频率位置发生蓝移。另外,通过化学计量学方法结合太赫兹吸收谱对黄酮类物质进行定性鉴别和定量分析研究,首先利用主成分分析(PCA)提取光谱特征变量,然后将前五个主成分分量作为支持向量机(SVM)的输入变量建立分类模型,通过优化模型选择最优参数,最终得到100%的分类准确度。另外,采用偏最小二乘回归(PLSR)模型和人工神经网络(ANN)模型对淀粉中含有不同浓度含量的黄酮类物质进行定量分析,经过对比这两种方法, ANN模型得到了最高的预测精度,其中,柚皮素和大豆素在预测集中的浓度预测相关系数分别为R^2=0.994 4,R^2=0.996 4,均方根误差分别为RMSE=1.932 5, RMSE=1.544 1。综上所述,利用THz-TDS技术研究了黄酮类物质在太赫兹波段的生物分子特性,并结合化学计量学方法为黄酮类物质提供了一种快速、有效、无损的定性鉴别和定量分析方法,在中草药的检测方面具有潜在的应用价值,对其他生物分子的研究也具有较强的借鉴意义。Flavonoids are a large class of polyphenols widely distributed in plants in the form of free or glycoside,which has anti-oxidation,anti-bacterial,anti-viral,anti-tumor growth and other pharmacological effects.As an important active component of traditional Chinese medicine,they have high medicinal value and development prospect.In this paper,the biomolecular properties of eight common flavonoids,including baicalein,quercetin,naringenin,daidzein,baicalin,puerarin,genistein and gastrodin,were studied by terahertz time-domain spectroscopy(THz-TDS)in the 0.2~2.5 THz band.The results showed that these flavonoids have different characteristic absorption peaks in the terahertz band.The terahertz absorption characteristics with temperature variation in the range of 78~320 K were studied.The results showed that the characteristic absorption peaks gradually increased with the decrease of temperature,and the frequency position of absorption peak was blue-shifted.In addition,Qualitative identification and quantitative analysis of flavonoids were carried out by chemometrics combined with terahertz absorption spectra.First,the spectral characteristic variables were extracted by principal component analysis(PCA),then the first five principal components were used as input variables of support vector machine(SVM)to establish a classification model,and the optimal parameters were selected through the optimization model,and finally,the classification accuracy of 100%was obtained.In addition,the partial least squares regression(PLSR)model and the artificial neural network(ANN)model were used to analyze the flavonoids with different concentrations in starch quantitatively.By comparing the two methods,the ANN model obtained the highest prediction accuracy.The correlation coefficients of naringenin and daidzein in the prediction set were R 2=0.9944,R 2=0.9964,and the root means square error was RMSE=1.9325 and RMSE=1.5441,respectively.In summary,the biomolecular properties of flavonoids were studied by THz-TDS technology,and a rapid,ef

关 键 词：黄酮类化合物 太赫兹时域光谱 化学计量学 定性鉴别 定量分析 

分 类 号：O433.4[机械工程—光学工程]