Abstract: A method based on ultra performance convergence chromatography-tandem mass spectrometry (UPC²-MS/MS) was established for the determination of ten kinds of azo dyes (ADs) in Chinese medicinal materials. The analytical conditions were optimized, and the matrix effect was investigated. The samples were crushed, and extracted by N-hexane. In order to remove the terpenoids and free fatty acids, the Cleanert Alumina N (60 mg/3 mL) solid phase extraction columns were used to purify the solutions for inducing strong matrix effects. The resulting solution was separated on a Viridis HSS C18 SB column (150 mm×2.1 mm×1.8 μm) under isocratic elution, and the flow rate was 1.5 mL/min. Supercritical carbon dioxide and methanol (containing 10 mmol/L ammonium formate) were used as the mobile phases. The separated compounds were analyzed with an electrospray ionization source at positive and multiple reaction monitor modes. Methanol (containing 3% water and 0.2% formic acid) was used as the auxiliary pump mobile phase, and the flow rate was 0.3 mL/min. The external standard method was used to quantitatively analyze the ADs. Good linear relationships are obtained between the peak area and concentration in the corresponding mass concentration ranges with the correlation coefficients (r²) not less than 0.998 8. The limits of detection (LODs) and limits of quantification (LOQs) are 0.2-1.0 μg/kg (S/N=3) and 0.5-2.5 μg/kg (S/N=10), respectively. The average recoveries at three spiked levels are in the range of 92.0%-106.2% (n=9). The relative standard deviations are in the range of 1.9%-4.2% (n=9). The matrix effects in the case of the samples were weak. The developed method was successfully used to determine the ten kinds of ADs in actual samples. Sudan Ⅰ and Sudan Ⅳ were detected in one safflower sample, and their amounts were 7.2 mg/kg. Meanwhile, the secondary mass spectrometry of this sample were analyzed to avoid false positive. The research findings provided accurate data to assess the exposure risk of ADs in Chinese medicinal materials. The method is fast, simple, sensitive, high specificity, environment friendly, which can provide an effective way to determine the content of azo dyes in Chinese medicinal materials. However, this method has some limitations. The resolution ratio of the mass spectrometer used in this method is too low for qualitative analysis of unknown compounds. It's difficult to qualitative the suspect ADs particularly accurate in samples which outside the scope of the investigation.