[Research on quality parameters of scutellariae radix (formula particles) using on-line NIR in pilot with different extraction process].

The on-line monitor for the changes in the content of baicalin in Scutellariae Radix formula particles during the extraction process was conducted by using near infrared spectroscopy (NIR). High performance liquid chromatography (HPLC) was used as a reference method. Kennard-Stone (KS) was used to divide sample sets, so as to compare different pretreatment methods. The synergy interval partial least squares (SiPLS) was used to screen out modeling wave band to establish partial least-squares models. The relative error method was applied to predict forecast set samples of Scutellariae Radix in three extraction phases. The results showed that the model established by Savitzky-Golay smoothing with 11 points (SG11 points) was the best, with the root mean square with cross validation (RMSECV), root mean square error of correction (RMSEC) and root mean square error of prediction (RMSEP) of baicalin of 0.092 7, 0.134 4 and 0.114 8, respectively, the determination coefficient R2 of greater than 0.99, and the relative error of baicalin content of less than 5%. This indicates that the on-line near infrared reflectance spectroscopy could be applied in on-line monitor and quality control of the extraction process of Scutellariae Radix formula particles.