ABSTRACT
Because the red and bright color of corolla is the main indicator for the quality assessment of good safflower,the dyed safflower is sometimes found at the herbal market,what is influence on this herb quality and efficacy. A total of 127 safflower samples was therefore collected from different cultivating areas and herbal markets in China to develop a rapid method to identify the dyed safflower. Near-infrared spectroscopy(NIRS) combined with characteristic identification,high performance liquid chromatography(HPLC),principal component analysis(PCA) and partial least squares regression analysis(PLS) were employed to differentiate safflower from dyed safflower samples,and further quantify the levels of the 6 dyes,i.e. tartrazine,carmine,sunset yellow,azorubine,acid red 73 and orange Ⅱ in the dyed safflower. The results indicated that the 50 safflower samples and 77 dyed safflower samples were located at different regions in PCA cluster diagram by NIR spectra. Tartrazine,carmineand and sunset yellow were found in the 77 dyed safflower samples with the amounts of 0. 60-3. 66,0. 11-1. 37,0. 10-0. 71 mg·g-1,respectively. It indicated that the three dyes were the common and main dyes in the dyed safflower. However,azorubine,acid red 73 and orange Ⅱ were not detected in all herb samples. A total of 62 dyed safflower samples were chosen as calibration samples to develop the model for estimating the amount of dyes in dyed safflower. The estimating accuracy was verified by another 15 dyed safflower samples. The values of tartrazine,carmine and sunset yellow in dyed safflower samples were compared between the NIRS and HPLC methods. Each value of mean absolute difference(MAD) was less than 5%. The correlation coefficients of tartrazine,carmineand and sunset yellow were 0. 970,0. 975,0. 971,respectively. It indicated the data quantified by NIRS and HPLC were consistence. It is concluded that NIRS can not only differentiate safflower from dyed safflower,but also quantify the amount of the dyes. NIRS is suitable for rapidly identify the quality of safflower.