Abstract: The soil samples （10 g） that had been freeze-dried and sieved were taken, and 10 g of diatomaceous earth was added. The mixture was mixed uniformly and 50 µL of surrogate standard solution including 2-fluorobiphenyl and 4,4′′-triphenyl-d₁₄ with mass concentration of 10 mg·L⁻¹ was added. The mixture was placed in the pressurized fluid extractor, static extraction was performed twice with the mixed solution of n-hexane and acetone at volume ratio of 1∶1 as extractant at temperature of 100 ℃, pressure of 10.34 MPa, and extraction time of 8 min. The extract was concentrated to near dryness by nitrogen at 30 ℃, and made its volume up to 1.0 mL with n-hexane. 0.5 g of copper powder was added to the activated magnesium silicate cartridge, followed by the addition of the aforementioned solution to be purified, and the mixture was allowed to stand for 5 min. The concentration device was washed by 5 mL of the mixed solution of isopropanol and n-hexane at volume ratio of 2∶8, and the washing solution was transferred to the magnesium silicate cartridge together. Elution was carried out sequentially with 10 mL of the mixed solution of acetone and n-hexane at volume ratio of 1∶9 and 8 mL of the mixed solution of isopropanol and n-hexane at volume ratio of 2∶8. The eluate was collected and concentrated to 0.80 mL by nitrogen at 30 ℃. 50 µL of mixed isotope internal standard solution including naphthalene-d₈, phenanthrene-d₁₀, chrysene-d₁₂, perylene-d₁₂, and decachlorobiphenyl with mass concentration of 10 mg·L⁻¹ was added, and the mixture was made its volume up to 1.0 mL with n-hexane. The 64 semi-volatile organic compounds in the solution was determined by gas chromatography-tandem mass spectrometry. In chromatographic analysis, HP-5MS column was used as the stationary phase for separation under the column heating program. In mass spectrometric analysis, electron impact ion source was used for ionization, multiple reaction monitoring mode was used for detection, and internal standard method was used for quantification. As shown by the results, linear relationships between the ratios of the corresponding quantitative ion pair response values to the internal standard quantitative ion pair response values and the ratios of mass concentrations of 64 targets to their internal standards were kept in the range of 20‒1 000 µg·L⁻¹, with detection limits （3.143s） in the range of 0.19‒2.89 μg·kg⁻¹. Test for recovery was made by the standard addition method, giving results in the range of 63.4%‒125%, with RSDs （n=6） of the determined values in the range of 1.4%‒14%. This method was applied to the analysis of 4 soil samples, and 64 targets were all detected, with the detection amounts in the range of 3.41‒1 606 952 μg·kg⁻¹.