On-Line Monitoring the Extract Process of Fu-Fang Shuanghua Oral Solution Using Near Infrared Spectroscopy and Different Pls Algorithms
Qian Kang,Qingguo Ru,Yan Liu,Lingyan Xu,Jia Liu,Yifei Wang,Yewen Zhang,Hui Li,Qing Zhang,Qing Wu
DOI: https://doi.org/10.1016/j.saa.2015.07.098
2016-01-01
Abstract:An on-line near infrared (NIR) spectroscopy monitoring method with an appropriate multivariate calibration method was developed for the extraction process of Fu-fang Shuanghua oral solution (FSOS). On-line NIR spectra were collected through two fiber optic probes, which were designed to transmit NIR radiation by a 2mm flange. Partial least squares (PLS), interval PLS (iPLS) and synergy interval PLS (siPLS) algorithms were used comparatively for building the calibration regression models. During the extraction process, the feasibility of NIR spectroscopy was employed to determine the concentrations of chlorogenic acid (CA) content, total phenolic acids contents (TPC), total flavonoids contents (TFC) and soluble solid contents (SSC). High performance liquid chromatography (HPLC), ultraviolet spectrophotometric method (UV) and loss on drying methods were employed as reference methods. Experiment results showed that the performance of siPLS model is the best compared with PLS and iPLS. The calibration models for AC, TPC, TFC and SSC had high values of determination coefficients of (R(2)) (0.9948, 0.9992, 0.9950 and 0.9832) and low root mean square error of cross validation (RMSECV) (0.0113, 0.0341, 0.1787 and 1.2158), which indicate a good correlation between reference values and NIR predicted values. The overall results show that the on line detection method could be feasible in real application and would be of great value for monitoring the mixed decoction process of FSOS and other Chinese patent medicines.