The Effect of Rock Wettability on Oil-Water Relative Permeability Relationships

Abstract

Oil-water relative permeabilities measured by a steady-state method are given for a broad range of rock wetting conditions. The data show that the degree to which a porous medium is wetted preferentially by oil or water significantly affects the measurement of flow properties and the calculation of reservoir waterflood oil recovery performance. Introduction Calculations of reservoir waterflood performance are frequently based upon oil-water relative permeability relationships measured on cores in the laboratory. Inherent assumptions in such applications of laboratory data are thatthe test samples are representative of the reservoir or some part thereof, andthe core handling and test procedure or conditions do not prevent obtaining representative flow relationships. Of the many factors that can influence the validity of these assumptions, the degree to which the reservoir wetting condition is reproduced in the laboratory flow tests is perhaps the most difficult to assess. Numerous investigators have reported upon the factors that can cause wetting conditions in laboratory core tests to be different from those in the reservoir. Welge perhaps was among the first to recognize that restored-state test procedures may not provide flow characteristics representative of the reservoir. Studies by other researchers have revealed that the important factors that can contribute to changes in core-wetting behavior can be divided into two general categories: those influencing the core-wetting condition before testing, and those influencing them during testing. The factors grouped in the first category arethe well coring fluid,the techniques used in handling, packaging, and preserving cores andthe laboratory packaging, and preserving cores andthe laboratory procedures for cleaning and preparing the cores. procedures for cleaning and preparing the cores. In the second category aretest temperature,test fluids, andthe test technique. These published research efforts have demonstrated the difficulty of retaining reservoir wetting properties in a core sample during laboratory testing. However because of the varied procedures used to detect changes in core-wetting properties, those studies do not show plainly the influence of these changes on core flow behavior. Thus, with perhaps few exceptions, it has not been made clear how these changes affect calculated predictions of reservoir waterflood performance. Our purpose here is to present performance. Our purpose here is to present laboratory experimental flow data that show that the preferential wetting characteristics prevailing in a preferential wetting characteristics prevailing in a core during testing have a marked and qualitatively predictable effect on oil-water relative permeability predictable effect on oil-water relative permeability relationships. Through the use of data covering a range of wetting conditions, we shall demonstrate the significant effect of rock wettability on calculations of reservoir waterflood performance. Test Materials and Procedures The data presented in this paper were obtained on a fired sample of Torpedo outcrop sandstone, 3/4 in. in diameter and 1 3/4 in. long. Additional data were obtained on unfired samples of Berea and Torpedo sandstone and on one fired Berea core, but because of the general similarity of results obtained, data from those tests are not included. The purpose of firing the core (in an electric furnace at 1,600 degrees F for 6 hours) was to stabilize any clay minerals present in the rock pore space and to provide an internal rock surface of as near constant properties as possible. JPT P. 873