Abstract: Window factor analysis(WFA) is an important method used to resolve the overlapping chromatographic peaks by processing HPLC-DAD data. The selection of window is one of the key factors to affect the accuracy of WFA. In this paper, the results obtained by WFA are used as the initial estimation, and then the estimated value is adaptively optimized by an alternative iteration procedure. In this way, a method named iterative fitting windows factor analysis(IFWFA) is proposed. It can correctly select the window for using WFA. Coupled with HPLC-DAD, the proposed method was used to determine tanshinone Ⅰ, cryptotanshinone and tanshinone ⅡA of Salvia miltiorrhiza, and the results show that their recoveries are 94.16%, 91.57% and 104.48%, respectively. Computer simulation and sample analysis experiments both demonstrate that the method is effective for improvement of analytical accuracy.

Key words: Analysis of Chinese Medicine, HPLC-DAD, Iterative fitting, Window factor analysis