近红外光谱测定人参与西洋参的主要皂甙总量  

Determination of Total Main Ginsenosides Contents in American Ginseng and Chinese Ginseng Using Near Infrared Spectroscopy

作　　者：黄亚伟[1] 王加华[1] Jacqueline J. Shan Lei Ling 韩东海[1] 

机构地区：[1]中国农业大学食品科学与营养工程学院,北京100083 [2]Afexa Life Sciences Inc. Edmonton,T6N 1G1,Canada

出　　处：《分析化学》2011年第3期377-381,共5页Chinese Journal of Analytical Chemistry

基　　金：中加合作研究项目(No.200904810610008)资助

摘　　要：采用近红外光谱测定人参与西洋参的主要皂甙总量。采集人参与西洋参的漫反射光谱,分别对光谱进行正交信号校正(OSC)与常规预处理,建立了对应的偏最小二乘(PLS)回归模型。与常规最优预处理方法相比,OSC能很好地消除人参与西洋参的品种差异,显著提高了光谱与皂甙含量的相关系数,同时降低了PLS建模因子数,提高了模型的稳健性与预测能力。所建立的OSC-PLS模型的预测集相关系数为0.97,预测标准差为0.519,相对分析误差为4.07。结果表明:近红外光谱可用于检测人参与西洋参主要皂甙总量,实现原料的现场快速筛查。Potential of NIR spectroscopy to determine total main ginsenosides contents in Chinese ginseng and American ginseng was studied.Diffuse reflective spectra of all samples were collected.Then the raw spectra were preprocessed with orthogonal signal correction（OSC） and routine methods respectively.After that,the corresponding partial least-squares（PLS） regression models were developed.Comparing with the optimal routine spectral preprocessing method,OSC can eliminate the sample species differences perfectly and greatly enhance the correlation between the spectra and the analyte of interest.Meanwhile,OSC decreased the PLS factors for modeling,which made the model more robust.The correlation coefficient of prediction（Rp）,root mean standard error of prediction（RMSEP） and ratio of deviation to performance（RPD） of the OSC-PLS model finally developed were 0.97,0.519 and 4.07 respectively.The result indicated that NIR spectroscopy could be used in practice to fast determine the total main ginsenosides contents in Chinese ginseng and American ginseng and fulfill the screen of raw materials on site.

关 键 词：近红外光谱 正交信号校正 人参 西洋参 皂甙 

分 类 号：S567.5[农业科学—中草药栽培] O657.33[农业科学—作物学]