Comparison of Laboratory and In-Situ Measurements of Waterflood Residual Oil Saturations for the Cormorant Field

Abstract

Summary. Shell Expro and Koninklijke/Shell E and P Laboratorium (KSEPL) havebeen engaged in a multidisciplinary effort to determine the waterflood residualoil saturation (ROS) in two principal reservoirs of the Cormorant oil field inthe U.K. sector of the North Sea. Data acquisition included special coring andtesting. The study, which involved new reservoir- engineering and petrophysicaltechniques, was aimed at establishing consistent ROS values. Reservoir-engineering work centered on reservoir-condition corefloods in therelative-permeability-at-reservoir-conditions (REPARC) apparatus, in whichrestoration of representative wettability conditions was attempted with theaging technique. Aging results in a consistent reduction of water-wetness ofall core samples. The study indicated that ROS values obtained on aged cores atwater throughputs of at least 5 PV represented reservoir conditions. Thepetrophysical part of the study involved ROS estimation from sponge-coreanalysis and log evaluation. Introduction Knowing the ROS to waterflood is very important both to manage waterfloodsin progress and to define possible long-term EOR targets. Shell Expro and KSEPLhave therefore been engaged in an integrated campaign to determinerepresentative ROS values in the Cormorant field. The in-situ measurementtechniques include logging and single-well tracer testing in Wells CormorantCA-28 and CA-29. Laboratory measurements included saturation measurements onpreserved whole core pieces cut from sponge core, countercurrent imbibitionexperiments, and corefloods under reservoir conditions with real reservoircrudes and brine. Reservoir-condition corefloods were performed in KSEPL's REPARC-2 equipment. REPARC-2 works on the basis of the restored-state, whichentails making a core sample water-wet by an appropriate cleaning procedure andsubsequently exposing it to reservoir crude and brine at connate watersaturation for a long enough time to establish the appropriate wettability(aging). In this paper we discuss the results of these techniques and attemptto reconcile obtained data. Sponge Coring Sponge coring is similar to conventional coring except that an aluminumliner, enclosing a porous, oil-wet polyurethane sponge is run inside the corebarrel. During the coring operation, the core enters the liner and fits tightlyinside the sponge. As the core is brought to the surface, gas comes out ofsolution inside the core, expelling water and possibly oil out of the core. This oil is normally trapped by the sponge; experience has shown that no oil islost from the liner when the core is brought to surface. The amount of oil inthe core and sponge can be measured and converted to in-situ conditions byapplying the oil FVF. PV in the core is adjusted for compaction with thestress-affection factor. Thus, in flooded zones, an estimate of in-situ Sor canbe obtained. A total of 323 ft of core was cut in Well Cormorant Well CA-29ST1 with a recovery of 263 ft (81 %) over the interval from 13,435 to 13,762 ftbelow derrick floor (BDF) in the Upper Ness, Lower Ness, Etive, Rannoch, and Broom formations. The entire core, still sealed inside the liner, was inspectedby X-ray tomography, and representative sections of the flooded and unfloodedintervals were selected for Dean-Stark analysis. The selected core intervalsand liner were cut into 0.3- to 1.0-ft pieces (whole core) and all oil andwater in both, as well as in the sponge, were removed by Dean-Stark extractionwith toluene as solvent. The volume of water was collected and measured; theoil remained in the toluene. The total weight of oil originally present in thecore plow was calculated from the difference between the weight of watercollected and the initial and dry weight of the core and sponge. Across theintervals in the Lower Ness and Etive formations, which from log data appear tobe the most completely flooded, the measured oil saturations ranged from 20 to52% and 22 to 43%, respectively. Across the interval considered to be floodedin the Lower Ness, up to 75 % of the total oil volume was found in the sponge, indicating that the majority of the zone may not be at irreducible orsaturation. An indication of Sor is obtained from 2.5 ft of core at the base ofthe flooded interval where there was practically no oil in the sponge and Sowas found to be 25% (measurement range of 20 to 30%). Across the floodedinterval in the Etive, little or no oil (4 % maximum) was found in the sponge, indicating that this zone is prodominantly at ROS. The most reliable Sorestimate is obtained prodominantly at ROS. The most reliable Sor estimate isobtained from the lowermost 4 ft of core, where So was found to be 25% (22 to28%). REPARC-2 Measurements Equipment. Fig. 1 is a schematic of the apparatus. The main parts of theequipment are oil and brine supply vessels, two core holders, productionvessels, a pressure-regulations system, differential production vessels, apressure-regulations system, differential pressure transducers, and adisplacement pump. The core holders pressure transducers, and a displacementpump. The core holders normally contain standard 1-in. cylindrical plugsbetween 2 and 5 in. long. The core plugs are embedded in epoxy resin in a monelmold. The volumes of produced fluids from the core plugs are measured inspecial production vessels. The principle of these measurements is the changingelectrical capacitance as the oil/water interface moves through the productionvessel. Core-Plug Selection and Preparation. From the results of X-ray tomography onsponge cores from Well Cormorant CA-29S1 in both Etive and Lower Nessformations, four core pieces were selected for REPARC-2 experiments. Plugs 1 in. in diameter and about 3 in. long were cut along the bedding plane from Cores 6.1.2 and 6.3.3 of the Lower Ness and Cores 12.6.3 and 13.3.6 of the Etive. Some difficulties were encountered in the experiments on the first plugof Core 13.3.6 (Plug 13.3.6.A), so two other plugs (Plugs 13.3.6.B and13.3.6.C) were taken for duplicate measurements. The coreplugs were thoroughlycleaned by Dean-Stark extraction with toluene for 24 hours, followed by asubsequent extraction with a chloroform/methanol mixture (70/30 vol%). The coreplugs were dried at 140 degrees F under light vacuum while nitrogen, humidifiedunder room conditions, flowed slowly through the stove. The dry cores werewrapped in teflon tape and centered in a monel mold. The annular space betweenthe mold and the core plug was filled with liquid epoxy resin that was thenallowed to cure at an elevated temperature (284 degrees F) for 24 hours. SPEFE P. 39

Characterization of porous media by chemical tracers using time‐of‐flight and residence time distribution concept

“…Due to chromatographic effect, the partitioning tracers reaches little late than conservative tracers at producing wellhead. This difference in arrival time of tracers at the producing well is used to estimate the remaining oil saturation . The average remaining oil saturation can be estimated using simple formula using Equation (2).…”

“…The difference in arrival times of tracer to the production well is used as parameter to determine the saturation between the injections and production well . The number of literature on the single well as well as interwell tracer is available, which focuses on the determination of residual oil saturation (Sor) using application of the injection of conservative and partitioning tracers together into subsurface …”