The effect of wettability heterogeneity on capillary pressure and relative permeability

Abstract

The effect of wettability on fluid flow properties in porous media has been extensively studied, and is still a subject of highly active investigation. Most of the work has focused on cores of homogeneous wettability. Little attention has been paid to wettability heterogeneity effects at the core or pore scale. In a previous paper, we reported on a series of centrifuge experiments performed to study the effect of wettability heterogeneity on capillary pressure. An experimental technique, named cyclic aging, was developed to create regions of different wetting in the same core sample. In this paper, the work is extended to study the effect of wettability heterogeneity on both capillary pressure and relative permeability curves using centrifuge, continuous injection and steady state techniques. The experimental procedure consists of three steps: (1) the core plug is fully saturated with brine and subsequently a drainage experiment is performed targeting an initial oil saturation S oi, (2) after aging, oil is displaced by water to residual oil saturation S or, and (3) oil is injected targeting higher initial oil saturation. In the secondary drainage experiment (step 3), oil first displaces water from the pores exposed to crude oil in primary drainage (step 1) and then enters fresh pores not exposed to crude oil before. In our previous study, it was found that wettability heterogeneity caused a step change in capillary pressure which correlated very well with the saturation at which wettability contrast was expected. However, the height of the step could not be explained by wettability contrast and/or water trapping alone. An experimental artifact caused by the centrifuge technique made the step higher than expected. The experimental artifact was the result of the nonuniform saturation profile developed across the core at the end of the centrifuge experiment. In this study, new techniques were used which resulted in a uniform saturation profile along the core sample during the primary drainage experiment. It was found that in this case the step in the capillary pressure is determined by wettability contrast and water trapping. It was also found that the relative permeability curve changes its characteristics when oil accesses the pores not previously exposed to crude oil. The results of this study show that (1) only the part of the pore space exposed to crude oil undergoes wettability changes on both core scale and pore scale, and (2) ignoring wettability heterogeneity can lead to large errors in the estimated two phase flow functions with important consequences with respect to fluid flow in porous media.