近红外光谱技术在赤芍等中药材中定量分析应用研究

发布时间：2019-05-25 01:35

【摘要】：中药是一个复杂的多组分体系,中药的质量控制对保证临床疗效、实现中药现代化、推动中药产业的技术发展有重要的意义。因此,需要一种简便、快捷的质量分析方法,对中药材进行准确、快速的含量测定。近红外光谱技术是一种快速、无损的绿色分析方法,结合化学计量学手段,可以对中药材进行快速含量测定。本研究借鉴其在定量检测方面的应用,实现对赤芍、黄芪、栀子、大黄、独活、白芷和防风7味中药材中指标成分的含量进行快速检测的目的。主要取得了以下进展:1.利用高效液相色谱法测定赤芍中没食子酸、儿茶素、芍药内酯苷和芍药苷的含量,选择101批样品,运用偏最小二乘法建立其含量与近红外光谱之间的校正模型。结果显示,没食子酸经多元散射校正+二阶导数处理,模型校正集r为0.9065,RMSEC值为0.10;验证集r为0.9163,RMSEP值为0.09;儿茶素经多元散射校正+二阶导数处理,模型校正集r为0.9408,RMSEC值为0.44;验证集r为0.9231,RMSEP值为0.40;芍药内酯苷经多元散射校正+一阶导数处理,模型校正集r为0.9406,RMSEC值为0.13;验证集r为0.9226,RMSEP值为0.14;芍药苷经Savitzky-Golay平滑+一阶导数处理,模型校正集r为0.9674,RMSEC值为0.22;验证集r为0.9281,RMSEP值为0.22。4种成分的HPLC测定值与NIR模型测定值有较好的线性关系,说明模型的拟合能力较好,可用于快速测定大批量赤芍药材中没食子酸、儿茶素、芍药内酯苷和芍药苷的含量。2.利用高效液相色谱-质谱联用法测定黄芪中毛蕊异黄酮葡萄糖苷和黄芪甲苷的含量,选择91批样品,运用偏最小二乘法建立其含量与近红外光谱之间的校正模型。结果显示,毛蕊异黄酮葡萄糖苷经多元散射校正+ 一阶导数+Savitzky-Golay平滑处理,模型校正集r为0.8635,RMSEC值为0.019;验证集r为0.8266,RMSEP值为0.023;黄芪甲苷经二阶导数+Savitzky-Golay平滑处理,模型校正集r为0.7963,RMSEC值为0.008;验证集r为0.8548,RMSEP值为0.006。2种成分模型拟合关系良好,可以实现对黄芪中毛蕊异黄酮葡萄糖苷和黄芪甲苷的快速含量测定。3.分别使用近红外、中红外光谱分析技术对栀子苷进行定量分析。首先,利用高效液相色谱法测定栀子中栀子苷含量,以药典法为依据,选择100批样品,运用偏最小二乘法分别建立栀子苷含量与近红外光谱、中红外光谱之间的校正模型。结果显示,近红外光谱中,光谱经过二阶导数+Savitzky-Golay平滑处理,模型校正集r为0.9725,RMSEC值为0.20;验证集r为0.9606,RMSEP值为0.22;中红外光谱中,光谱经过二阶导数+标准正态变量变换处理,模型校正集r为0.9256,RMSEC值0.24;验证集r为0.9174,RMSEP值为0.24。通过实验可以发现,近红外光谱、中红外光谱均可以快速、无损、有效的对栀子中栀子苷含量进行测定。4.采用近红外光谱技术,结合药典含量测定方法,对106批大黄样品中大黄酸、大黄素、大黄酚、大黄素甲醚和芦荟大黄素的含量进行测定,并计算总含量,通过对预处理方法的考察,建立其与近红外光谱之间的最佳校正模型。结果显示,光谱经一阶导数处理后,模型校正集r为0.9856,RMSEC值为0.13;验证集r为0.9445,RMSEP值为0.23。说明采用近红外光谱技术,可以对大黄中大黄酸、大黄素、大黄酚、大黄素甲醚和芦荟大黄素的总含量进行快速测定。5.采用近红外光谱技术,结合药典含量测定方法,对97批独活样品中蛇床子素和二氢欧山芹醇当归酸酯的含量进行测定,通过对预处理方法的考察,建立其与近红外光谱之间的最佳校正模型。结果显示,蛇床子素光谱经一阶导数处理后,模型校正集r为0.9233,RMSEC值为0.16;验证集r为0.9413,RMSEP值为0.14;二氢欧山芹醇当归酸酯光谱不需经过预处理,模型校正集r为0.8315,RMSEC值为0.11;预测集r为0.8574,RMSEP值为0.10。说明采用近红外光谱技术,可以实现对独活中蛇床子素和二氢欧山芹醇当归酸酯的快速定量分析。6.采用近红外光谱技术,结合药典含量测定方法,对99批白芷样品中欧前胡素的含量进行测定,通过对预处理方法的考察,建立其与近红外光谱之间的最佳校正模型。结果显示,欧前胡素光谱经二阶导数+Savitzky-Golay平滑处理后模型最佳,校正集r为0.9755,RMSEC值为0.01;预测集r为0.9630,RMSEP值为0.01。说明通过近红外光谱技术建立了一种快速测量白芷中欧前胡素含量的新方法。7.采用近红外光谱技术,结合药典含量测定方法,对100批防风样品中升麻素苷及5-O-甲基维斯阿米醇苷的含量进行测定,并计算总含量,通过对预处理方法的考察,建立其与近红外光谱之间的最佳校正模型。结果显示,光谱经二阶导数+多元散射校正处理后,模型最佳,模型校正集r为0.9773,RMSEC值为0.05;验证集r为0.9617,RMSEP值为0.05。说明采用近红外光谱技术,可以实现对防风中升麻素苷及5-O-甲基维斯阿米醇苷总含量的快速测定。
[Abstract]:The traditional Chinese medicine is a complex multi-component system, and the quality control of the traditional Chinese medicine is of great significance for ensuring the clinical curative effect, realizing the modernization of the traditional Chinese medicine and promoting the technical development of the traditional Chinese medicine industry. Therefore, a simple and rapid quality analysis method is needed, and the accurate and rapid content determination of the traditional Chinese medicinal materials is required. The near infrared spectroscopy is a kind of fast and non-destructive green analytical method. By using the application of the method in the quantitative detection, the purpose of rapid detection of the content of the index components in the red paeony root, the yellow root, the gardenia, the rhubarb, the pubescent angelica, the dahurian angelica root and the wind-proof 7-flavor traditional Chinese medicine is realized. The following progress has been made:1. The content of gallic acid, catechin, Paeonolactone and Paeonia lactiflora in Radix Paeoniae Rubra was determined by high performance liquid chromatography, and 101 samples were selected, and the correction model between its content and near infrared spectrum was established by partial least square method. The results show that gallic acid is treated with multiple scattering correction + second order derivative, the model correction set r is 0.9065, the RMSEC value is 0.10, the verification set r is 0.9163, the RMSEP value is 0.09, the catechin is processed by the multivariate scattering correction + second order derivative, the model correction set r is 0.9408, the RMSEC value is 0.44, the verification set r is 0.9231, the RMSEP value is 0.40, the paeonolactone is processed by the multivariate scattering correction + first derivative treatment, the model correction set r is 0.9406, the RMSEC value is 0.13, the verification set r is 0.9226, the RMSEP value is 0.14, the paeonia is processed by the Savitzky-Golay smoothing + first derivative processing, the model correction set r is 0.9674, the RMSEC value is 0.22, the verification set r is 0.9281, The results showed that the value of RMSEP was 0.222.4, and the linear relationship between the measured value and the measured value of NIR model showed that the fitting ability of the model was good, and it can be used to quickly determine the content of gallic acid, catechin, Paeonolactone and Paeonia lactiflora in large batch of Radix Paeoniae Rubra. A high performance liquid chromatography-mass spectrometry (LC-MS) method was used to determine the content of the total content of the soybean isoflavone and the level of the yellow-and-green in the yellow rice, and a 91-batch sample was selected, and the correction model between the content and the near-infrared spectrum was established by using the partial least squares method. The results showed that, with the first derivative + Savitzky-Golay smoothing process, the corrected set r of the model was 0.8635, the RMSEC value was 0.019, the verification set r was 0.8266, the RMSEP value was 0.023, the second derivative of the yellow and the second derivative, the Savitzky-Golay smoothing process, the model correction set r was 0.7963, and the RMSEC value was 0.008; The results showed that the value of RMSEP was 0.8548 and the value of RMSEP was in the range of 0.006.2 species. The quantitative analysis of the gardenia was carried out by using the near-infrared and mid-infrared spectrum analysis techniques. First of all, the content of gardenia in the gardenia was determined by high performance liquid chromatography, and 100 batches of samples were selected according to the pharmacopoeia method. The correction model between the content of the gardenia and the near infrared spectrum and the mid-infrared spectrum was respectively established by the partial least square method. The results show that, in the near-infrared spectrum, the spectrum passes through the second-order derivative, the Savitzky-Golay smoothing process, the model correction set r is 0.9725, the RMSEC value is 0.20, the verification set r is 0.9606, the RMSEP value is 0.22, the mid-infrared spectrum, the spectrum passes through the second-order derivative + standard positive-state variable conversion processing, the model correction set r is 0.9256, The RMSEC value is 0.24; the validation set r is 0.9174 and the RMSEP value is 0.24. It can be found that in the near infrared spectrum, the infrared spectrum can be fast, non-destructive and effective in the determination of the content of the gardenia in the gardenia. The content of rhein, emodin, chrysophanol, emodin and aloe emodin in 106 rhubarb samples was determined by near-infrared spectroscopy, and the total content was calculated. The best correction model between the near-infrared spectrum and the near-infrared spectrum is established. The results show that, after the first derivative treatment, the model correction set r is 0.9856, the RMSEC value is 0.13, the verification set r is 0.9445, and the RMSEP value is 0.23. The total content of rhein, emodin, chrysophanol, emodin and aloe emodin in Radix et Rhizoma Rhei can be quickly determined by near-infrared spectroscopy. In this paper, the content of cnidium and dihydro-ocarvanol in 97 batches of pubescent angelica is determined by near-infrared spectroscopy, and the best correction model between them and near-infrared spectrum is established by the investigation of the pre-treatment method. The results showed that, after the first derivative treatment, the model correction set r is 0.9233, the RMSEC value is 0.16, the verification set r is 0.9413, the RMSEP value is 0.14, the dihydro-parsley alcohol angelica ester spectrum does not need to be pre-processed, the model correction set r is 0.8315, the RMSEC value is 0.11, the prediction set r is 0.8574, The RMSEP value is 0.10. In this paper, the rapid quantitative analysis of cnidium and dihydro-ocarvanol in the radix angelicae pubescentis can be realized by using the near-infrared spectroscopy. By means of near-infrared spectroscopy and the method of the method of compendial content determination, the content of the pre-determination of the content of the former in the 99 batches of Radix Angelicae Dahuricae is determined, and the optimal calibration model between the two-infrared spectrum and the near-infrared spectrum is established by the investigation of the pretreatment method. The results show that the model is best after the second derivative + Savitzky-Golay smoothing process, the correction set r is 0.9755, the RMSEC value is 0.01, the prediction set r is 0.9630, and the RMSEP value is 0.01. A new method for rapid measurement of the content of imperatorin of Radix Angelicae Dahuricae by near-infrared spectroscopy is presented. In this paper, the content of L-O-methyl and 5-O-methyl-vibriol in 100 batches of wind-proof samples was determined by near-infrared spectroscopy, and the total content was calculated. The results show that the model is the best after the second-order derivative + multi-element scattering correction process, and the model correction set r is 0.9773, the RMSEC value is 0.05, the verification set r is 0.9617, and the RMSEP value is 0.05. The method of near-infrared spectroscopy can be used to measure the total content of L-and 5-O-methyl-vibriol in the wind-proof.
【学位授予单位】：中国中医科学院
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R284.1;O657.33

【参考文献】