Determining Relative Permeabilities by Use of Experimental Design for Very Heterogeneous Samples

It is usually assumed when measuring relative permeabilities that the reservoir sample is homogeneous. Great progress has been made in non destructive techniques to quantitatively estimate the degree of homogeneity of the core samples: CT scan, surface minipermeameter and mini-sonic, miscible displacement, ... These methods are nowadays routinely used to scan the samples prior to any measurement and to select the most appropriate one in terms of petrophysical properties and homogeneity. Unfortunately, core samples are often heterogeneous either because it was not possible to select a more homogeneous sample or because the heterogeneity is intrinsic to the facies. A lot of attention has been paid to that question in the recent years, and the possible impact of heterogeneities at the small scale on relative permeabilities is well documented in the literature¹.