RESUMO
OBJECTIVE: To develop an on-line analysis method based on near infrared spectroscopy (NIRS) for monitoring the percolation process of compound Kushen injection traditional Chinese medicines. METHODS: Samples and correlative spectra were collected with online equipment in percolation process. The contents of total alkaloids and total solids measured with standard methods were used as reference values. Multivariate calibration models based on partial least squares (PLS) algorithm were developed to correlate the spectra and the reference values. RESULTS: The correlation coefficients of the calibration models for total alkaloids and total solids were 0.9979 and 0.9960 with root mean square errors (RMSECs) of 0.0721 g · L-1 and 0.0811%, respectively. External validation samples collected on-line were used to evaluate the quantitative models. The correlation coefficients between the predicted value and the reference value were 0.9915 and 0.9990 with root mean square errors of prediction (RMSEPs) of 0.1950 g· L-1 and 0.0681%, respectively. CONCLUSION: The presented method is rapid, accurate and reliable, which can be used for monitoring the percolation process of compound Kushen injection.