Network modeling as a tool to predict three-phase gas injection in heterogeneous wettability porous media

The objective of the present paper is two-fold: to demonstrate the effect of small scale wettability heterogeneities on gas injection efficiency, and to develop a tool to predict this impact for various patterns and spatial distributions. To this end, an experimental investigation in transparent glass micromodels is performed and a theoretical simulator is developed. Different heterogeneity patterns are considered for the same oil-wet over water-wet surface ratio. A special technique to obtain oil-wet patches in a water-wet matrix by selective silane grafting on the glass surface is developed and described. The simulator properly incorporates imbibition–drainage events and flow through films as visualized in the experiments. Predictive efficiency of the simulator is tested by comparing the numerical results to the experimentally obtained phase distributions and saturations. In water/oil displacement, drainage and imbibition mechanisms occur according to the local wettability. Three-phase gas injection is dominated by drainage mechanisms. Oil flows through wetting films in the oil-wet regions and through spreading films on water in the water-wet regions. Both experiments and simulations show that the size and distribution of wettability heterogeneities strongly affect the microscopic and macroscopic behavior during gas injection processes. More specifically the effect of wettability heterogeneities on: (1) displacement mechanisms, (2) sweep efficiency, and (3) fluid distribution in three-phase gas injection is clearly demonstrated and successfully described by the network simulator.