Rapid analysis of the in-process extract solutions of compound E Jiao oral liquid using near infrared spectroscopy and partial least-squares regression

摘要

A rapid and simultaneous quantification of the main active compounds of the in-process extract solutions is beneficial to the process monitoring as well as ensuring quality consistency of the end products during the production of traditional Chinese medicine. Here we present a near infrared (NIR) spectroscopy-based method for rapid analysis of extract solutions in the production of compound EJiao oral liquid. The partial least-squares regression (PLSR) models for four quality indicators (viz. total flavonoids, total saponins, total saccharides and soluble solid contents) were established and validated. The results showed that all of the four models exhibited satisfactory fitting and predictive capacity. The root mean squares error of prediction (RMSEP) was 0.0384 mg mL~(-1), 0.0154 mg mL~(-1), 3.80 mg mL~(-1) and 0.199% for total flavonoids, total saponins, total saccharides and soluble solid contents, respectively. This work here demonstrated that NIR spectroscopy coupled with PLSR calibration can offer a reliable and non-destructive alternative in the routine monitoring of the extraction process in the production of compound E Jiao oral liquid. The presented approach is expected to be equally applicable to the mixed decoction of other herbal medicines.