Abstract: Brine wastewater deep well injection in China is still at the stage of access demonstration. The transport of liquid waste in deep reservoirs involves variable density drive, water-rock chemical reactions and variable-parameter synergistic physical-chemical interactions. How to quantitatively characterize these processes is one of the key scientific and technological challenges to be solved in deep well injection. Consequently, this paper selected a certain injection zone in Ordos Basin as prototype, used PFLOTRAN to construct a variable-density and variable-parameter reactive solute transport model to predict and evaluate the migration, transformation and evolution of reservoir minerals, as well as the change of reservoir properties after the brine wastewater injection to the deep geological body. As the simulation results showed, the injection of brine wastewater into the injection zone contributes to secondary minerals precipitation, such as anhydrite, calcite and fluorite; the geological storage capacity decreases by 1.51% approximately in the simulated 100 years; and the contamination area is around 2.57 km². This model provides a simulation tool to quantitatively characterize the migration and transformation process of brine wastewater components after the deep well injection.

Key words: deep well injection, variable parameters, variable-density flow, reactive transport modeling, PFLOTRAN