近红外光谱法在参芪扶正注射液醇沉工艺质控中的应用研究

发布时间：2018-05-26 22:38

  本文选题：党参 + 黄芪　； 参考：《浙江大学》2017年硕士论文

【摘要】：随着"质量源于设计"等先进药品质控理念逐渐应用于生产实践,中药质量控制的重心也由成品检验往原料品质控制和制药过程质量控制转移。针对重要操作单元,采用过程分析技术建立有效过程监控方法,能显著提升制药过程质控水平,进而提升中药质量一致性。本文以参芪扶正注射液醇沉工艺为研究对象,开展近红外光谱与多变量数据分析技术相结合的应用研究,主要包括以下内容:1.建立黄芪一次醇沉过程在线监控模型。采用近红外光谱与多变量统计过程控制相结合的方法,建立了黄芪一次醇沉过程监控模型。通过计算各时间点处的主成分得分、HotellingT2和DModX统计量,观察过程是否处在受控范围内,据此判断过程运行是否正常。结果表明所建模型具有较好的监控能力,能及时识别过程异常。2.建立党参一次醇沉上清液多指标近红外光谱快速分析法。通过实验设计和浓缩-稀释的方法获得具有代表性的样本集。逐一优化各建模步骤,最终所得模型预测性能良好,实现了党参一次醇沉上清液中党参炔苷、总黄酮、色素、总固体等多指标快速定量。3.建立黄芪二次醇沉的在线近红外光谱分析法。建立黄芪醇沉液中毛蕊异黄酮苷、芒柄花苷、紫檀烷葡萄糖苷、异黄烷葡萄糖苷、黄芪甲苷、黄芪皂苷Ⅱ等6个化学指标的HPLC-UV-ELSD测定法。在建立近红外光谱校正模型时,通过实验设计构建了包含原料、关键工艺参数、环境温度、光谱采集方式等变异的校正集样本,采用独立外部验证样本用于模型评价,用实验设计法找出预处理策略的优化方向,调整算法参数,减少了试错法的计算次数,提高了模型对目标工艺过程的预测性能,模型复杂程度低,稳健性较好,实现了黄芪二次醇沉中总固体和6种化学成分的同时测定。4.基于准确度轮廓的黄芪二次醇沉过程多指标在线近红外光谱分析方法的验证。首先设立独立验证批次,计算模型性能指标,从总体上评价模型预测能力;然后分别从真实性、精密度、准确度、线性与范围、专属性、稳健性和不确定度几个方面考察方法性能;再以准确度轮廓作为方法有效性的决策工具,计算各模型在不同含量水平下预测误差的β-期望容许区间,选择落在误差接受限±15%以内的有效含量范围;最终获得总固体、毛蕊异黄酮苷、芒柄花苷、紫檀烷葡萄糖苷、异黄烷葡萄糖苷、黄芪甲苷和黄芪皂苷Ⅱ等指标的有效含量范围分别为:8.44-39.8%、0.541-2.26 mg/mL、0.118-0.502 mg/mL、0.220-0.940 mg/mL、0.106-0.167 mg/mL、0.137-0.320 mg/mL 和 0.484-0.879 mg/mL。在以上浓度范围内,近红外光谱在线分析方法能够实现黄芪二次醇沉过程中7个指标的准确定量,且方法稳健性良好。
[Abstract]:With the application of "quality from Design" and other advanced drug quality control concepts in production practice, the focus of quality control of traditional Chinese medicine is also transferred from finished product inspection to raw material quality control and pharmaceutical process quality control. According to the important operation unit, using process analysis technology to establish an effective process monitoring method can significantly improve the quality control level of pharmaceutical process, and then improve the consistency of quality of traditional Chinese medicine. In this paper, the alcohol precipitation technology of Shenqi Fuzheng injection was taken as the research object, and the application of near infrared spectroscopy combined with multivariate data analysis technology was carried out, mainly including the following contents: 1: 1. An online monitoring model for primary alcohol precipitation process of Astragalus membranaceus was established. The monitoring model of primary alcohol precipitation process of Astragalus membranaceus was established by combining near infrared spectroscopy with multivariable statistical process control. By calculating the statistics of Hotelling T2 and DModX at each time point, we observed whether the process was in a controlled range and judged whether the process was running normally. The results show that the model has better monitoring ability and can identify process anomalies in time. A fast near-infrared spectrum analysis method for the supernatant of primary alcohol precipitation of Codonopsis pilosula was established. The representative sample set was obtained by experimental design and concentration-dilution method. Each modeling step was optimized one by one, and the prediction performance of the model was good, and the rapid quantification of Codonosine, total flavonoids, pigment and total solids in the supernatant of primary alcohol precipitation of Codonopsis pilosula was achieved. An on-line near-infrared spectrometric analysis of secondary alcohol precipitation of Astragalus membranaceus was established. A HPLC-UV-ELSD method was established for the determination of isoflavone glycosides in Astragalus membranaceus alcohol precipitate, Astragalus saponin 鈪，

本文编号：1939237