目的 应用近红外(NIR)光谱技术和化学计量学方法建立红参药材质量评价的新方法。方法 对红参同时进行原料定性鉴别和水分含量检测。将样品NIR光谱同标准光谱库相比较,计算相似度匹配值。以传统的干燥失重法(LOD)为参照方法,应用相关光谱法进行波长选择和多重散射校正(MSC)方法进行光谱预处理,采用偏最小二乘(PLS)回归方法建立NIR光谱预测水分含量的校正模型。结果 NIR光谱库可正确区别红参和伪品;最优PLS校正模型相关系数为0.9997。结论 此方法快速、准确,可用于中药生产企业的原料药材质量控制。

Objective Near-infrared (NIR) reflectance spectroscopy was used to develop a fast quality assessment method by simultaneous material identification and moisture quantification of Radix Ginseng Rubra (RGR). Methods The sample was identified by the comparison of its spectrum with a standard NIR spectral library. Similarity measurement was used as the discriminating parameter. The moisture content of sample was quantified by a partial least square (PLS) calibration model, correlative spectrum calculation was used for wavelength selection and multiplicative signal correction (MSC) was applied for pretreatment in the calibration model. Primary reference data were obtained using the traditional loss on drying (LOD) method. Results The NIR library can distinguish RGR from the counterfeit successfully. The optimized eight-factor PLS calibration model of NIR spectra has a high correlation coefficient (R=(0.999 7).) Conclusion The proposed method is rapid, accurate and can be used routinely in the traditional Chinese materia medica manufacturers for quality control of raw materials.

国家“十五”重大科技攻关计划资助(2001BA701A45);浙江省科技厅重大科技计划(021103549)