【作者】 胡钢亮；

【导师】 吕秀阳；

【作者基本信息】 浙江大学 ， 生物化工， 2003， 硕士

【摘要】 中药是我国的民族瑰宝，现正面临着迫切的现代化和国际化难题，解决这些难题，必须参考和借鉴其他学科成熟的技术和工艺。近红外光谱技术在农业、食品、石化等领域中得到了成功的应用，本文就近红外光谱技术在中药领域中的应用进行了初步的探索。 中药材的质量直接决定了中药成药的质量。长期以来，中药材质量判别依赖于老药工经验判别法，缺乏一种科学的仪器分析方法，是制约中药现代化的一大障碍。本文利用漫反射近红外光谱技术，通过将中药材人工定性判别结果数字化，与近红外光谱数据进行关联，建立校正模型，从而实现对未知中药材样本的预测判别。本文以中药材油菜花粉为例，对该方法在中药材质量判别中的应用进行了研究，所建立的校正模型能较好地预测油菜花粉未知样本的质量，预测准确率达到了92.3％，表明此方法是可行的，判别结果是准确的，同时，这种基于近红外光谱技术的中药材质量判别方法具有良好的普遍意义，可以应用于其他中药材的质量判别。 中药制药过程尚缺乏有效的快速分析手段。本文利用近红外投射光谱直接定量测定银杏提取物中总黄酮和总内酯的含量，所建立的校正模型对预测集样本的总黄酮和总内酯含量的复相关系数分别为0.998和0.986，相对标准偏差(RSD)分别为2.44％和3.19％，表明尽管存在强烈的溶剂吸收干扰，但是仍然可以利用近红外透射光谱快速、准确地测定中药提取液中有效成分的含量。这种基于近红外光谱的中药提取液快速分析技术有望应用于中药提取过程的即时分析和大批量产品的检测。 中药制药过程中大量存在着粉末、浸膏以及无规则固体状态的分析对象，传统的分析技术无法直接测定这些样品。本文利用漫反射近红外光谱技术，直接测定了银杏提取物固体粉末中银杏总黄酮的含量，通过所建立的校正模型对未知样本的检验，其相关系数为0.973，相对标准偏差为1.13％。结果表明，应用漫反射近红外光谱技术直接测定中药粉末中有效成分的含量是完全可行的，并具有快速高效、结果准确的优点。—— 通过以上研究表明，近红外光谱技术在中药领域中具有良好的应用前景，借助于近红外光谱技术，可以克服中药领域中一些久悬未决的难题，有助于中药的现代化和国际化进程。但是，近红外光谱技术在中药领域中的应用研究目前仍然停留在实验室初期阶段，在中药工业中应用这项技术，还需要持续深入的研究。更多还原

【Abstract】 Traditional Chinese Medicine (TCM) is a great treasure of China, while TCM is facing serious problems to become a kind of modern medicine accepted by the world. Those mature technologies in other areas should be used into TCM to settle these problems. Near Infrared Spectroscopy (NIRS) is a kind of successful technology which is widely used in lots of areas, such as agriculture, food industry and petrochemical engineering. The applications of NIRS in TCM are studied in this paper.The quality of TCM is directly due to the quality of TCM materials, while there is not a kind of modern instrumental analysis method to judge the quality of TCM materials, so we have to judge the quality of TCM materials by experienced workers. In this paper, a news method to classify the quality of TCM materials based on NIRS is reported. By digitalization of the qualitative results of Man-made Quality Judgment and correlation with NIRS data, a model can be established with calibration of this method to the Brassica Campestris pollen is made. The prediction veracity is 92.3%. This result shows that the classify method of TCM materials based on NIRS is viable with good catholicity.Instant analysis by NIRS in TCM process is discussed in this paper. The total flavones and total lactones in Gingko Biloba Extraction (GEB) liquor are determined by NIRS with Partial Least Square (PLS). The multicorrelation coefficients of the model to the total flavones and total lactones of the prediction samples are 0.985 and 0.971, the RSD are 2.44% and 3.19%. The result shows that the method proposed can be used to simultaneously analyze the total flavones and total lactones in GEB liquor with high accuracy under strong absorption of menstruum, and the application of NIRS in TCM process to the area of instant analysis and on-line control is possible.These are a lot of samples in TCM are powder, concentrate and atactic solid. Such kinds of samples cannot be directly analyzed by normal analysis method. In this paper, the total flavones in GEB powder directly determined by NIRS is studied. The multicorrelation coefficients of the prediction samples are 0.973, the RSD is 1.13%.inThis result shows that NIRS can be used to directly determine the TCM with high speed and good veracity.In a word, the application of NIRS in TCM has a good potential and it will help to settle some problems in the modernization of TCM. But the research of NIRS’ application in TCM is still in laboratory and a lot of researches should to be done to realize the application of NIRS in TCM industry.更多还原