In order to solve the problem of formation time of inkjet printed documents in forensic field，the volatile components in four models of two brands black inkjet printed documents were quantitatively and qualitatively analyzed by gas chromatography-mass spectrometry（GC-MS）. The samples printed at one certain time were tracked and tested，at regular intervals the printed samples were punched by a micro punch with a diameter of 0.5 mm and the extract was analyzed by GC-MS. Retention time and qualitative ions combined with NIST mass spectrometry library were used for qualitative analysis and internal standard method to determine the concentration of each substance in the extract. The content of volatile solvent

C

（ng/mm） in the ink blot per unit le

ngth（mm） was calculated and used as a quantitative unit of aging of printed samples. The results showed that four experimental samples contained six volatile solvent components，including 2-pyrrolidone，1，5-pentanediol，2-methyl-1，3-propylene glycol，N-hydroxyethyl-2-pyrrolidone，glycerol and triethylene glycol. These volatile components began to volatilize and transfer at the moment the printing ink was entried in the paper，and tended to stabilize. Finally，the content continued to decrease to a certain value. Based on the analysis data measured over different time period，two types of exponential functions were used to analyze the diachronic change of volatile solvents and a single component and time nonlinear regression models including S-curve，Power and Logarithm model，a multi-component and time multiple linear regression model were established to predict the formation time of printed documents and compare the effect of prediction. By comparing the evaluation indexes of the models，it was concluded that the prediction effect of the multiple linear regression model is better for the ink with more solvent sets.