Application of composite membranes to determine relative permeabilities from displacement experiments

Relative permeabilities are important characteristics of multiphase flow in porous media. Displacement experiments for relative permeabilities are usually interpreted by the JBN method neglecting capillary pressure. The experiments therefore have to be run at higher rates than those experienced during reservoir exploitation. Another disadvantage is that the relative permeabilities can only be determined for the usually small saturation interval outside the shock. We have developed a method to interpret displacement experiments with the capillary pressure included. The data needed are in situ measurements of saturations and phase pressures. The experiments can then be run at low flow rates, and relative permeabilities can be determined for all saturations. When using a short core, or if the saturation profile is spread out too much, the measurements can be affected by the end effect. We investigate using a combined water-wet and oil-wet membrane at the core outlet end to eliminate the end effect. With the membrane, the steady-state saturation profile is fairly uniform, and calculations of steady-state relative permeabilities are improved. However, because of the geometry of the membrane, the phases are separated at the core outlet. The separation results in increased flow resistance, and increased pressure in the core. Calculations of relative permeabilities for the traveling wave from pressure measurements after breakthrough will therefore be erroneous.