科研之友
微信
新浪微博
Facebook
分享链接
摘要
The aim of this study was to explore the feasibility of near infrared (NIR) spectroscopy for the quantitative analysis of Compound E Jiao oral liquid, a popular oral TCM preparation, through intact glass vials. Transmission spectra were acquired with a sample holder tailored to the instrument's sample compartment. Total flavonoids, total saponins and soluble solid contents were considered as quality indicators of this oral preparation and calibration models were built using partial least-squares regression (PLSR) algorithm. The sample set was comprised of production samples as well as laboratory samples prepared by a concentrating-diluting method so as to extend the concentration range beyond the production variation. The developed models gave root mean square error of prediction (RMSEP) of 0.084 mg mL(-1) for total flavonoids, 0.014 mg mL(-1) for total saponins and 0.21% for soluble solid contents. To better evaluate the prediction performance, an independent set of ten production samples was introduced for the external validation. The models are compared with those built with the transflectance spectra collected in the destructive way. The results of paired t-test and F-test at a significance level of 0.05 revealed that the two sets of models exhibited comparable model performance. The overall results demonstrate that the proposed non-destructive method can be used as an advantageous alternative in the quality inspection of Compound E Jiao oral liquid production.
- 出版日期2015-5
- 单位浙江大学
