NIRS法对栀子不同炮制品栀子苷含量的快速检测

发布时间：2018-10-31 11:45

【摘要】：以96批栀子不同炮制品为研究对象,高效液相色谱测定栀子苷含量为参考值,利用近红外光谱仪积分球漫反射测定其光谱图,建模波段取8 660~7 500,6 650~5 600和4 900~4 000cm~(-1),以标准正态变换(SNV)和二阶导数法(2nd derivative)为预处理方法,主成分数为8,采用偏最小二乘法(PLS)对83批栀子样品建立栀子苷的定量校正模型,最终以13批栀子不同炮制品对模型进行验证。结果,定量模型的内部交叉验证决定系数(R2)为0.992 85,校正均方差(RMSEC)为0.240,预测均方差(RMSEP)为0.254,内部交叉验证均方差(RMSECV)为0.386 91,RMSEP/RMSEC=1.06。模型验证得到的相对分析误差(RPD)为8.81,绝对偏差范围-0.39%~0.23%,说明模型预测性较好。通过相关系数法,优选样品装样量、扫描次数、重复次数、分辨率实验条件;并由近红外一阶导和二阶导图,除去温湿度和样品水分影响波段,结合栀子苷对照品近红外光谱图,确定建模波段。首次利用NIRS法建立栀子不同炮制品栀子苷定量校正模型,方法简单快速,模型稳定可靠、准确性高,可同时应用于不同炮制品栀子中栀子苷含量的预测。
[Abstract]:96 batches of Gardenia jasminoides (Gardenia jasminoides) processed with different processed products were used as the study object, the content of Gardenia was determined by HPLC as reference value, and the spectrogram was measured by diffuse reflectance of the integrated sphere of near infrared spectrometer. The modeling wave band was 8 660,500,500,650,50,600 and 49004000cm ~ (-1), respectively. Using standard normal transformation (SNV) and second derivative method (2nd derivative) as the pretreatment method, the main component was 8. The quantitative calibration model of gardenia was established by partial least square method (PLS) for 83 lots of gardenia samples. Finally, 13 batches of different processing products of Gardenia jasminoides were used to verify the model. The results showed that the internal cross validation determination coefficient (R2) of the quantitative model was 0.992 85, the corrected mean square error (RMSEC) was 0.240, the predicted mean square error (RMSEP) was 0.254, and the internal cross-validation mean square error (RMSECV) was 0.386 91rms EPP / RMSECC 1.06. The relative analysis error (RPD) obtained from the model verification is 8.81, and the absolute deviation range is -0.39 and 0.23, which indicates that the model is predictable. Through the correlation coefficient method, the sample loading quantity, scanning times, repetition times and resolution experimental conditions were optimized. The first and second order map of near infrared was used to remove the temperature and humidity and the influence band of sample moisture, and the modeling band was determined by combining with the near infrared spectrum of the reference substance of geniposide. For the first time, the NIRS method was used to establish a quantitative calibration model of gardenia in different processed products. The method was simple, rapid, stable, reliable and accurate, and could be used to predict the content of gardeniposide in different processed Gardenia jasminoides simultaneously.
【作者单位】：广东药科大学中药学院;国家中医药管理局中药数字化质量评价技术重点研究室;广东高校中药质量工程技术研究中心;广东省药品检验所;
【基金】：全国中药炮制技术规范研究中医药行业科研专项(201207004-7)资助
【分类号】：O657;R286.0

【相似文献】