IPTC 13664

State of the Art Special Core Analysis Program Design and Results for Effective Reservoir Management, Dukhan Field, Qatar Jon P. Meissner, SPE, ExxonMobil Upstream Research Company (EMURC), Fred H.L. Wang, SPE, EMURC, Jim G. Kralik, SPE, EMURC, Mohamed Naguib Ab Majid, SPE, Qatar Petroleum (QP), Mohamed Ismail Bin Omar, SPE, QP, Tarak Attia, SPE, QP, and Khaled Al-Ansari, SPE, QP

Copyright 2009, International Petroleum Technology Conference This paper was prepared for presentation at the International Petroleum Technology Conference held in Doha, Qatar, 7–9 December 2009. This paper was selected for presentation by an IPTC Programme Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the International Petroleum Technology Conference and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the International Petroleum Technology Conference, its officers, or members. Papers presented at IPTC are subject to publication review by Sponsor Society Committees of IPTC. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the International Petroleum Technology Conference is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, IPTC, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax +1-972-952-9435.

Abstract A state-of-the-art special core analysis (SCAL) program for the Dukhan Arab C and Arab D carbonate reservoirs was designed to provide reliable relative permeability and capillary pressure models for use in field-wide reservoir studies. The workflow process for the design and its implementation of such a program is described with a specific focus on four key requirements: 1) measurements must be on representative rock samples (the right samples), 2) measurements must be made under conditions representative of displacement processes in the reservoir (the right conditions), 3) measurements must be made using precision equipment and techniques (the right equipment), and 4) trained and experienced technologists are needed to conduct the measurements and model the data (the right people). Results from the Arab D program are presented to demonstrate consistent saturation function models (capillary pressure and relative permeability) for simulation. The steady-state relative permeability methods, coupled with centrifuge, provide data over the range of saturation conditions encountered in the reservoir for both water-oil and gas-oil fluid pairings; this range of data coverage is significantly greater than is typically available in the industry and hence reduces uncertainty in the resultant relative permeability models. The water-oil relative permeability behaviors suggest a mixed-wet character with a preference to oil; however, one reservoir rock type (RRT) shows mixed wet character with a neutral preference for oil and water. The gas-oil relative permeability measurements are made on a unique apparatus capable of conducting the testing with reservoir fluids at reservoir conditions in the presence of irreducible water saturation. Centrifuge USBM wettability measurements were conducted on samples in the native, cleaned and restored state. Based on the measurements, it is concluded that the wettability restoration method used in this study was effective for a high-permeability Dukhan Arab D limestone core, but wettability restoration cannot be achieved for the moderately low-permeability limestone core. In general, restoration of carbonate cores to native state wettability is less successful than for siliciclastic cores. This is thought to be due to the complex pore structure of carbonates and the potential for change of pore structure through diagenesis after migration of petroleum.

Introduction The large, mature Dukhan Field is located onshore Qatar, approximately 80km west of Doha. Hydrocarbons are contained in a North-South plunging anticline approximately 70km long by 8km wide. The field was discovered in 1939 and first production occurred in 1949. The field has more than 750 well penetrations. The major oil reservoirs are the Upper Jurassic Arab C and Arab D. There are lesser amounts of oil and gas in the Middle Jurassic and non-associated gas in the Permo-Triassic Khuff formation. The Arab C and Arab D reservoirs have gone through various phases of production including natural depletion, flank water injection, gas injection processes, vertical as well as horizontal well development and gas-lift implementation to maintain production and maximize hydrocarbon recovery. Reservoir performance prediction is an integral component of an effective reservoir management program. The reliability of reservoir performance predictions is affected by the quality of the underlying relative permeability and capillary pressure models used in reservoir simulation. In order to develop reliable models, four key requirements must be employed: 1) measurements must be on rock samples representative of the reservoir (the right samples), 2) measurements must be made under conditions representative of displacement processes in the reservoir (the right conditions), 3) measurements must be made using precision equipment and techniques (the right equipment), and 4) trained and experienced technologists are needed to conduct the measurements and model the data (the right people).

2 IPTC 13664

The involvement of the right people is critical. The complexity of carbonate reservoirs requires close interaction between geoscientists and reservoir engineers to jointly plan the distribution and frequency of SCAL measurements across RRT in order to derive appropriate relative permeability and capillary pressure models for use in reservoir simulation. A RRT is a group of reservoir facies that are found to have characteristic single and multiphase flow properties; specifically, capillary pressure and relative permeability behavior can be described by a single set of models, potentially with a dependence on permeability, porosity or some combination thereof. A Reservoir Facies (ResFac) is characterized as a group of rocks having an internally consistent mineralogy pore and grain structure, this definition is typically based on visual core description, routine core analysis, petrographic thin section description and pore body/throat size distribution information. Preparation, execution and quality control of the SCAL measurements requires the focused attention of engineers, technicians and support staff, including fabricators, machinists, equipment inspection/testing specialists, and safety professionals. Finally, once the SCAL data are available, the data can be examined and modeled for use in reservoir simulation and feedback to the geologic model for adjustments in the distribution and proportion of rock types. In our experience, the approach described herein has supplied the appropriate underlying data that has enabled successful reservoir simulation matching of field production history and reservoir performance.

Background Measurement techniques for relative permeability and capillary pressure are well established in the industry. Hassler and Brunner proposed the centrifuge methods that are still widely used today for capillary pressure measurement.1 Honarpour et al. summarized the methods currently available for primary drainage capillary pressure.2 Braun and Blackwell reported a robust method for the measurement of steady-state water-oil relative permeability.3 Schafer et al. reported on improvements to the steady-state method with the inclusion of an unsteady-state displacement cycle.4 Wang et al. reported a system for steady-state gas-oil relative permeability measurements at reservoir conditions using reservoir fluids in the presence of irreducible water.5,6 The measurements of gas-oil relative permeability at reservoir conditions illustrated that the established practice of making such measurements at mild conditions with laboratory fluids may not accurately represent reservoir displacement.5,6 In contrast to the water-oil steady-state relative permeability measurements at reservoir conditions that were described over twenty years ago, to our knowledge, reliable reservoir conditions gas-oil steady-state relative permeability measurements are unique.3,5,6 In contrast to measurement techniques, a workflow for the systematic experimental design of a SCAL program to provide saturation functions for reservoir simulation is not well established in the literature. This challenge is further complicated for carbonate reservoirs where the diversity and complexity of carbonate RRTs require more targeted experimental design to provide data coverage over the RRTs necessary to populate the reservoir simulation model.

Application A state-of-the-art SCAL program includes both high accuracy measurements and the appropriate distribution of those measurements across RRTs so that consistent relative permeability and capillary pressure models can be constructed for use in simulation. Table 1 highlights the data typically required to build saturation function models for reservoir simulation throughout the life of an oil field, excluding miscible displacements. The table describes each measurement and its application to reservoir simulation. Note that the application is on the basis of defining inputs for each RRT. Note that the secondary drainage water-oil relative permeability data may not be required for simulation, unless it is expected that grid blocks will experience increasing oil saturations after imbibition. These data are however, often measured to gain insight into the wettability state of the in-situ rock. A similar situation applies to imbibition gas-oil relative permeability in that it is only required if the gas saturation decreases within a grid block. The native wettability state refers to the in-situ wettability state of the reservoir, whereas water-wet refers to the presumed initial reservoir state prior to charging with hydrocarbons. For carbonates, the best practice for relative permeability and capillary pressure measurements is to perform them on native state core material rather than restored state core materials; the restoration of carbonate core materials to the true in-situ reservoir wettability state is not a reliable technique.7 However, the best practice for primary drainage measurements is to perform them on cores in a water-wet state that is representative of conditions during the hydrocarbon charging of the reservoir. The overall workflow to obtain the data described in Table 1 is shown in Figure 1. The objective defines the RRTs to be tested and the frequency of measurements over those selected RRTs. The goal is to ensure that displacement properties (saturation functions) can be properly defined for the gridblocks in the reservoir simulation model. The objective is typically jointly defined by the SCAL project lead and the geoscientists/reservoir engineers developing/producing the asset. The reservoir engineers’ scaled-up simulation model provides insight into the character of the grid blocks for which saturation functions must be defined. When designing the SCAL program, due to the large number of ResFacs described at Dukhan, the ResFacs are grouped into preliminary lithology bins. An example for the Arab D reservoir is shown in Table 2. If selected properly, the lithology bins may approximate the final RRT classification. For the purpose of clarity the lithology bins will be referred to as RRT for the remainder of the paper. The final determination of RRT is guided by the integration of sedimentologic, geologic, petrophysical and core analysis parameters including the capillary pressure and relative permeability data. At the outset of a study, all of these factors were not known so it was not certain whether a lithology bin does in fact represents a RRT. Accordingly it is recommended that all SCAL measurements be made on samples of a single ResFac. Specifically if multiple plug samples are combined into a composite core, they should be of the same ResFac, as well as similar porosity and permeability.

IPTC 13664 3

The second step of the workflow focuses on reviewing the existing data sources (ResFac classification system and geologic information, existing routine and SCAL data, fluid property data) so that the program is defined properly with regard to the rock (lithology, vugs, fractures, anhydrite nodules/inclusions) and fluids (composition, saturation pressure, gas-oil ratio, asphaltene content). Additionally it is designed to minimize overlap where quality SCAL measurements may already exist. Next the experimental plan defines the measurement procedures, methods and laboratories to perform the work. The next major steps are to define the sample acquisition plan, ResFac determination and selection of samples. The sample acquisition plan is shown in Figure 2. It consists of visual inspection of the core followed by using a medical-grade x-ray computerized tomography unit (sometimes referred to as a "CT Scanner"), to produced two-dimensional tomographs of longitudinal slices. For carbonates, features such as vugs and flow barriers may not be visible on the core surface. The CT scans are then used in conjunction with the visual inspection to develop a plan to acquire core plugs; in general our recommendation is to acquire plugs in pairs. In a plug pair, one plug is preserved for SCAL testing, whereas another plug is collected as close as possible to the SCAL plug for use in basic property characterization/measurement (porosity, permeability, grain density). Once the plan is completed, plugs are acquired. As the plugs are acquired, any that are fractured or damaged sufficiently to preclude use, should not be considered for SCAL testing. It is often advisable to trim the SCAL plug to produce disk material for a thin section and mercury injection measurement. After plugging the individual plugs are visually inspected to make sure they meet the sampling requirements. It is recommended that the plug sample also be CT scanned (in a native state); the plug CT scan has higher fidelity than the whole core scan. The outcome of the sample acquisition plan is a set of SCAL candidate plugs that are uniform and do not contain fractures, damage, defects, or heterogeneities that are not representative of the rock type. The next step is the ResFac determination workflow shown in Figure 3. The ResFac identification typically begins with a description of the core slab. Next data on the plug samples are obtained by preparing and describing a thin section obtained from one of the end trims from each SCAL plug. Additional basic property data are acquired on the companion sample to the SCAL plug. Depending on the RRT method utilized, it may also be necessary to obtain an estimate of pore size distribution as inferred from NMR T2 measurements on the fully saturated companion core plug, and to obtain an estimate of the pore throat size distribution as obtained from mercury injection capillary pressure measurements. The mercury injection data are typically acquired using a cylindrical sample cut from a SCAL plug end trim. The selected data set is then used to determine the ResFac of each SCAL candidate plug. The sample acquisition plan and the ResFac determination workflow are aligned with key requirement (1) that measurements must be on rock samples representative of the reservoir (the right sample). Finally the samples are grouped into sets for testing according to Figure 4. Sample Set 1 consists of the rock samples selected for steady-state relative permeability; in our best practice each sample is tested for both water-oil and gas-oil relative permeability in the presence of water. This eliminates differences in rock samples when comparing the data models for gas-oil and water-oil relative permeability. Sample Set 2 consists of the samples for centrifuge measurements of gas-oil relative permeability and capillary pressure; similarly, the use of the same samples enables comparison of data sets without the variation of the rock sample. Set 3 consists of the samples for imbibition water-oil relative permeability and capillary pressure. Set 4 contains the samples on which primary drainage capillary pressure and electrical properties are measured using the porous plate method. Finally, Sample Set 5 consists of the samples selected for centrifuge wettability measurements. Sample Sets 2-5 are closely modeled after the RRT that exist in Sample Set 1 to ensure that well defined saturation functions can be modeled from the measured data. Once the selection of samples is complete, the next step in the overall workflow, as shown in Figure 1, is to conduct the measurements under rigorous quality assurance and quality control. Note that a differentiation is made between the measurements of relative permeability and capillary pressure that directly define displacement data in the simulator and the supporting measurements such as interfacial tension and wettability. While wettability, NMR T2 pore size distribution, mercury injection capillary pressure pore throat size distribution, petrographic analysis of thin sections and other characterization measurements are not direct inputs into the reservoir simulation, they do influence the development of capillary pressure and relative permeability models. Wettability data often corroborate trends seen in relative permeability and capillary pressure tests. The data are then modeled for reservoir simulation as illustrated in Figure 5. This process begins with crossplotting of the endpoint data defined by capillary pressure measurements versus rock properties. It should be noted that the trapped gas saturation is typically defined by the liquid flood out at the end of a gas-oil steady-state imbibition relative permeability test rather than a capillary pressure test. Each crossplot should utilize a labeling system to clearly the reservoir facies of each sample. The definition of the endpoints for a given RRT may either be a single value or a trend based on permeability, porosity or some function thereof. The endpoints or endpoint functions that are defined for the irreducible water saturation, the ultimate residual oil saturation by water displacement, the ultimate residual oil saturation by gas displacement, and the trapped gas saturation, enable the next step, developing data plots for each data type to be made on the basis of actual saturation and normalized saturation (see Appendix A for additional information). The plots made for each data type are

4 IPTC 13664

actually a family of plots illustrating the most common data viewing methods. Specifically, the capillary pressure is plotted with the y-axis on the basis of capillary pressure (on a constant IFT basis) and using the Leverett method.8 The use of normalized saturations for visualization aids in the development of the saturation functions in that the models used to describe capillary pressure and relative permeability in reservoir simulation are typically based on normalized saturations. The plots must include data labels for each point including the ResFac, porosity and permeability in order to facilitate the grouping of data into the most descriptive RRT model of the reservoir behavior. From the data plots, groupings of the data by RRT become evident, which are then fitted to the desired mathematical models for relative permeability and capillary pressure. Typically only the highest quality data is used in the first pass of modeling data for reservoir simulation; it is then useful to review the models and highest quality data with successively lower quality bins to asses whether the lower quality data supports the developed models. Finally, returning to the workflow described on Figure 1, after the data are incorporated into the reservoir simulation, the adequacy of the data models in predicting field behavior is assessed. In our experience, the approach described herein has been critical in supplying the appropriate underlying data that has enabled successful reservoir simulation matching of field production history and reservoir performance. Thus far the importance of the “right sample” is clear, however the conditions under which the measurement is made effectively determines whether the measurement is representative of the displacement processes in the reservoir (the right conditions). Table 3 is a concise description of these requirements. All of the measurements require the application of reservoir stress to meet best practice standards. Since the primary drainage measurement is made with the rock in a strongly water-wet state using strongly wetting and non-wetting phases, the measurement can be made at mild temperatures, pressures, and without the use of reservoir fluids; the capillary pressure can be scaled to reservoir conditions using interfacial tension measurements. The centrifuge measurements are recommended to be made using reservoir crude oil at elevated temperature. The capillary pressure data obtained by this method along with measurements of the interfacial tension at laboratory (dead oil, reservoir temperature) and reservoir conditions has proven to be an accurate representation of the reservoir behavior as evidenced by repeated history matching and simulation model interrogation success. In the 1960’s, Jersey Production Research (an ExxonMobil pre-cursor) pioneered reservoir conditions measurements of water-oil imbibition capillary pressure using live oil; including both the positive and negative portions of the imbibition curve. With the introduction of robust centrifuge technology (with dead oil) in the 1970’s, an internal evaluation of these two methods was made and it was concluded that the centrifuge method was superior due to highly reliable and repeatable measurements. The live-oil spontaneous imbibition curve measurements indicated that for centrifuge measurement, the spontaneous imbibition curve can be modeled by scaling a primary or secondary drainage curve by the total quantity of spontaneous imbibition as measured by the centrifuge method. The steady-state water-oil and gas-oil relative permeability measurements are recommended to be conducted at reservoir temperature and pressure using live reservoir fluids. Note that live fluids refer to recombining the dead oil with synthetic gas to mimic the composition and the saturation pressure of the reservoir oil. Before the start of testing the live fluids are mixed in the experimental systems at test conditions for several days to ensure that true phase equilibrium has been established. Note that interfacial tension data are measured for all fluids. Interfacial tension data are required in calculating capillary forces in porous media. The difference in interfacial tension between the laboratory fluid system and the reservoir fluid system must be accounted for when the laboratory-measured capillary pressure data are applied to reservoir engineering calculations. All interfacial tension measurements for the Dukhan Arab D system were performed using the pendant drop method. When performing relative permeability measurements it is not advisable to perform them at flow rates representative of the reservoir. At low rates representative of the reservoir, relative permeability data are strongly influenced by capillary pressure.9 Regarding key requirement (3), that measurements must be made using precision equipment and techniques (the right equipment), Braun et al. and Wang et al. describe the measurement methods used for steady-state relative permeability on water-oil and gas-oil systems, respectively.3,6 The centrifuge techniques are based on the work by Hassler and Brunner and a summary of the primary drainage porous plate techniques were described by Honarpour.1,2 People impact the process in that trained and experienced technologists are needed to conduct the measurements and model the data; this represents key requirement for success (4), the right people. Presentation of Data and Results SCAL measurements were made for five lithology based rock-type groupings (as shown in Table 2): RRT A - lime mudstone/packstone, RRT C - lime grainstone, RRT D - calcareous dolostone, RRT E - dolograinstone, and RRT F - sucrosic dolostone. Note that there were other RRT defined for the field, RRT B for example, but not within the scope of this study.

IPTC 13664 5

NMR NMR T2 measurements were conducted at overburden pressure at 100% brine saturated conditions. The NMR data shows that the pore systems of these plugs are complex. Most are multi-modal, and many display significant micro-porosity. In our interpretation, as shown by example in Figure 6, some samples display a variable pore size distribution with large moldic pores (T2 relaxation time mode of 1000 ms) resulting from grain dissolution, somewhat smaller between crystal pores (200 ms), and two types of smaller pores (30 ms and 2 ms) considered to be micro-porosity. This sample is from the dolograinstone RRT E. Other samples are dominated by micro-porosity (largely invisible at the scale of the thin section photo insert), with a T2 relaxation time mode of 40 ms, as shown by example in Figure 7. This sample is from the lime mudstone/packstone RRT A. Capillary Pressure Data Capillary pressure data provide both asymptotic endpoints for and the functions to describe the capillary forces opposing displacement of one fluid by another in the reservoir. This is described further in Tables 1 and 3. Gas-brine primary drainage capillary pressure data were obtained by the porous plate method on thirty extracted-state core plugs representing the five RRT groups, with permeability ranging from 0.01 mD to 4,000 mD; the limestone samples are plotted in Figure 8 and the dolostones in Figure 9. Note that the saturation axes are presented in terms of normalized saturations; for all of the capillary pressure, relative permeability and wettability data figures in this paper, the axes are presented on the basis of normalized saturations as defined in Appendix A. Both capillary pressure and relative permeability data are typically modeled for use in reservoir simulation on a normalized saturation basis; the wettability data are plotted on this basis for consistency with the other data plots. The capillary pressure values are reported at laboratory conditions for the nitrogen-brine fluid system. For the Arab D Dukhan samples that were tested, higher reservoir quality rock can be defined as those with greater than 10% porosity and greater than 1 mD Klinkenberg permeability. In general, low permeability and low porosity plugs tend to have higher and more variable irreducible water saturation while the higher permeability and higher porosity plugs tend to have lower and less variable irreducible water saturation. While the change in water saturation from 200 to 800 psi is generally less than three saturation units for the higher reservoir quality samples, the change in water saturation from 200 to 800 psi is greater than 10 saturation units for the lower reservoir quality samples. This emphasizes the importance of high pressure measurements for lower reservoir quality rocks to match or exceed the maximum primary drainage capillary pressure that has existed in the reservoir. Water-oil imbibition capillary pressure data were obtained by the centrifuge method on 16 native state core plugs of different rock types, with permeability ranging from 0.2 mD to 3,708 mD. The limestone samples are plotted in Figure 10 and the dolostones in Figure 11. Drainage gas-oil capillary pressure in the presence of irreducible water saturation were obtained by the centrifuge method on eight native state core plugs of different reservoir facies, with permeability ranging from 0.6 mD to 2,593 mD. The data are plotted in Figure 12. The saturation and capillary pressure values presented here were calculated at the core inlet face using a numerical technique to solve the general equation described by Hassler and Brunner.1 Water-Oil Imbibition and Secondary Drainage Relative Permeability Details of the experimental apparatus for the measurement of water-oil relative permeability were described in a prior publication.3 The steady-state method used for testing allowed accurate measurements of water-oil relative permeability over a wide range of saturation in both imbibition and secondary drainage directions. The final step of each path included an unsteady-state flood analyzed by the method of Johnson, Bossler, and Naumann.10 This method yielded additional data over the relatively large saturation interval measured during this final step. All of the composite cores in the Arab D study, except for one, show similar hysteresis behaviors in water-oil relative permeability as illustrated in Figure 13. The oil relative permeability data fall almost on the same curve in both drainage and imbibition directions (i.e., little hysteresis). The water relative permeability data, however, are very different in the two directions (i.e., significant hysteresis). These observations indicated that the tested Arab D cores tended toward mixed wet with a preference for oil. Figure 13 also includes a centrifuge water-oil imbibition relative permeability measurement on a sample of the same rock type and similar permeability and porosity. The centrifuge test measures oil relative permeability at low levels of oil saturation. The one core that showed a different wettability behavior than the others in the Arab D study is illustrated in Figure 14; the significant hysteresis for both phases indicates a mixed-wet character with no preference for oil or water. Gas-Oil Drainage and Imbibition Relative Permeability The steady-state method allows accurate measurements of gas-oil relative permeability over a wide range of saturation in both drainage and imbibition directions, as well as measurements of endpoint saturations, such as critical gas saturation and trapped gas saturation. The steady-state relative-permeability tests were performed first in the drainage direction and then in the imbibition direction. Details of the experimental apparatus for the measurement of gas-oil relative permeability were described in a prior publication.6 A key aspect of the system is the recirculation of gas with high-rate pumps. As described,

6 IPTC 13664

the system enables accurate measurements for multiphase systems that contain gas. In these situations matching reservoir conditions, specifically interfacial tensions, is critical to obtaining quality results. Typical results are shown in Figure 15. The gas relative permeability data in all seven composite cores show strong hysteresis, i.e., the gas relative permeability data are very different in the drainage direction versus in the imbibition direction. The oil relative permeability data show only slight hysteresis. In two of the composite cores from RRTs C and F, the oil relative permeability data are almost identical in both directions. For the other five composite cores (two from RRT A, one each from RRTs C, D, and E), the imbibition oil relative permeability data are somewhat higher than drainage data. Additional gas-oil drainage oil relative permeability data were obtained by the centrifuge method (also shown on Figure 15). The centrifuge test measures oil relative permeability at low levels of oil saturation. Wettability in the Native, Clean and Restored State Wettability measurements are not an input into reservoir simulation; however, they provide insight into the relative permeability and capillary pressure behavior at the reservoir’s native wettability state. Specifically, knowledge of the reservoir’s native wettability state enables one to evaluate whether the measured relative permeability and capillary pressure are consistent with that wettability state. As such, wettability measurements were made in the native state, as well as the cleaned and restored states. The core acquisition plan for the Dukhan Arab D samples focused on preservation of native wettability. The cores were cut with a bland water based mud, free of surfactants or other chemicals that would alter the core’s wettability state. Details of the wettability analysis and measurement techniques are found in Appendix B. The conclusions drawn from this study should apply to restoration of weathered cores, but will not apply to cores that have been contaminated with oil-based drilling-fluid chemicals. Removal of oil-based mud filtrate components from carbonate mineral surfaces may be much more difficult than removing adsorbed crude oil components from the mineral surfaces. Recent work on a siliciclastic samples rich in carbonate cements further supports this observation.11 After USBM testing in the native state, flow through solvent extraction was applied to the samples with the goal of achieving a strongly water-wet condition in the cleaned state. Following USBM measurements in the cleaned state, the samples were aged at reservoir temperature and crude oil in an effort to restore the wettability found in the reservoir. Table 4 compares the wettability test results from measurements in the native, cleaned, and restored states for two Dukhan Arab D core plugs. Figure 16 shows the USBM capillary pressure curves in these three states for Plug No. 264, RRT A and Plug No. 187, RRT C. For the native state cores, the USBM wettability indices showed mixed wet character. Also shown are the Amott oil index (Io) and Amott water index (Iw). These indices define the ratios of spontaneous displacement to total displacement during the oil-drive and during the water-drive cycles, respectively. Most of these samples show some spontaneous displacement in both oil-drive and water-drive cycles. In evaluating the core wettability, a mixed wet sample showing preference for water is usually indicated by a positive USBM index, a positive Iw, a zero value of Io, and a small value of displaceable saturation. On the other hand, a mixed wet sample showing preference for oil is normally indicated by a negative USBM index, a zero value of Iw, and a positive Io. The majority of reservoir cores are of mixed wettability, which is commonly indicated by a near-zero USBM wettability index, some positive value of Iw and/or Io, and a displaceable saturation greater than 0.6. For the Arab D there were no significant differences in wettability among different RRTs. The USBM wettability test results of the two core plugs after the cores had been cleaned by solvent injection show that were successfully cleaned to a water-wet condition. These tests were performed with n-tetradecane and brine. Crude oil was not used because adsorption of polar crude oil components could occur during the tests, changing the wettability. The USBM wettability results for the restored cores show that the restored cores had become mixed wet after they were aged for six weeks at the irreducible water saturation and reservoir temperature. The USBM tests on restored cores were performed with degassed crude oil and brine to preserve the wettability. Note that the wettability indices are very similar between the native and the restored states. However, the USBM capillary pressure curves of the restored cores are generally shifted to the right of the native state core data. For the low-permeability core (Plug No. 264, RRT A, Figure 16), the USBM capillary pressure curves were very different after the restoration. Figure 17 compares the steady-state water-oil relative permeability data measured on a limestone composite core (Core No. A2, RRT A), which had a moderately low permeability (ko,Swi = 68 mD). Note that the shape of the relative permeability curves changed significantly after restoration. This observation is consistent with that in the USBM capillary pressure data for Plug No. 264. Figure 18 compares the steady-state water-oil relative permeability data measured on another limestone composite core (Core No. C1, RRT C), which had a very high permeability (ko,Swi = 2439 mD). Note that the shape of the relative permeability curves after restoration was similar to that in the native state. However, the relative permeability curves were shifted to the right. One explanation was that the restored core had an initial water saturation (Swi) higher than that of the native state core, 18.5% versus 12.5% PV. We repeated the wettability restoration procedure on Core No. C1 at

IPTC 13664 7

12.5% PV of initial water saturation, and found that the relative permeability curves of the restored core and the native state core became very similar (Figure 19). Based on the wettability results described, we concluded that the wettability restoration method used in this study was effective for a high-quality Dukhan Arab D limestone core, but did not work for the moderately low-quality limestone core. In general restoration of carbonate cores to native state wettability is less apt to be successful than for siliciclastic cores.7 Relative to siliciclastics, the complex pore structure of carbonates and the change of pore structure by diagenesis after migration of petroleum are the most likely factors which influence wettability restoration. The hypothesis is that distribution of oil and water in the core during the laboratory aging is not representative of that in the reservoir. This insight results in improved models for native state relative permeability and capillary pressure. Conclusions The best practices for the design of a state-of-the-art special core analysis (SCAL) program to provide reliable relative permeability and capillary pressure models for use in reservoir simulation have been described. The workflow process for the design of such a program and its implementation was described with a specific focus on key requirements: 1) measurements must be on rock samples representative of the reservoir (the right samples), 2) measurements must be made under conditions representative of displacement processes in the reservoir (the right conditions), 3) measurements must be made using precision equipment and techniques (the right equipment), and 4) trained and experienced technologists are needed to conduct the measurements and model the data (the right people). Key conclusions include: • The NMR data shows that the pore systems of these plugs are multi-modal (micro, meso and macro pores), many with

significant micro-porosity. • In general, low permeability and low porosity plugs tend to have higher and highly variable irreducible water saturation

while the higher permeability and higher porosity plugs tend to have lower and less variable irreducible water saturation. • While the change in water saturation from 200 to 800 psi is generally less than 3 saturation units for the higher reservoir

quality samples, the change in water saturation from 200 to 800 psi is greater than 10 saturation units for the lower reservoir quality samples. This emphasizes the importance of high pressure measurements for lower reservoir quality rocks to match or exceed the maximum primary drainage capillary pressure in the reservoir.

• The steady-state relative permeability methods, coupled with centrifuge, provide data over the range of saturation conditions encountered in the reservoir for both water-oil and gas-oil fluid pairings; this range of data coverage is significantly greater than is typically available in the industry and hence offers reduced uncertainty in the resultant relative permeability models.

• The water-oil relative permeability behaviors suggest a mixed-wet character with preference to oil wet; however, one RRT shows mixed wet character with a neutral preference for oil and water.

• The gas-oil relative permeability measurements are made on a unique apparatus capable of conducting the testing with reservoir fluids at reservoir conditions in the presence of irreducible water saturation.

• The wettability restoration method used in this study was effective for a high-permeability Dukhan Arab D limestone core, but did not work for the moderately low-permeability limestone core.

• In general, restoration of carbonate cores to native state wettability is less successful than for siliciclastic cores. Relative to siliciclastics, the complex pore structure of carbonates and the potential for change of pore structure through diagenesis after migration of petroleum are the most likely factors which influence wettability restoration.

Acknowledgments The authors wish to thank Qatar Petroleum and ExxonMobil Upstream Research Company for their support and Qatar Petroleum for permission to publish this paper. The authors also would like to recognize A.M Trabelsi, M.M. Honarpour, Nizar F. Djabbarah, S.A. Dixon, and J.T. Edwards for their significant contributions. Nomenclature A1 area under the oil-drive capillary pressure curve A2 area above the water-drive capillary pressure curve CT computer-aided tomography IFT interfacial tension Io Amott oil index Iw Amott water index kg,abs absoulte permeabiliy kg,inf Klinkenberg gas permeability

8 IPTC 13664

ko,Swi effective permeability to oil at irreducible water saturation krg relative permeability to gas kro relative permeability to oil krw relative permeability to water mD Milli-Darcies NMR nuclear magnetic resonance psi pounds per square inch PV pore volume(s) RRT reservoir rock type SCAL special core analysis Sl liquid saturation Slirr irreducible liquid saturation Sln normalized liquid saturation Sorw ultimate residual oil saturation Sw water saturation Swi average initial water saturation Swirr irreducible water saturation Swmin minimum water saturation Swn normalized water saturation T2 transverse relaxation time USBM United States Bureau of Mines W USBM wettability index References 1. Hassler, G.L. and Brunner, E.M., “Measurement of Capillary Pressure in Small Core Samples,” Trans. AIME, (1945) v.160, p.114-123. 2. Honarpour, M.M., Djabbarah, N.F., Kralik, J.G.: “Expert-Based Methodology for Primary Drainage Capillary Pressure Measurements

and Modeling,” paper SPE 88709 presented at the 2004 Abu Dhabi International Petroleum Exhibition and Conference, Abu Dhabi, Oct. 10-13.

3. Braun, E.M. and Blackwell, R.J.: “A Steady-State Technique for Measuring Oil/Water Relative Permeability Curves at Reservoir Conditions,” SPE 10155, presented at the 1981 SPE Annual Technical Conference and Exhibition, San Antonio, TX, Oct. 5-7.

4. Shafer, J.L., Braun, E.M., Wood, A.C., and Wooten, J.M.: “Obtaining Relative Permeability Data using a Combination of Steady-State and Unsteady-State Waterfloods,” SCA 9009, presented at the 4th Annual Technical Conference, Society of Core Analysts, Dallas, TX, Aug 14-16, 1990.

5. Wang F.H.L., Honarpour, M.M., Djabbarah, N.F., and Haynes, F.M.: “Characterization of Multiphase Flow Properties for Tertiary Immiscible Displacement Processes in an Oil-Wet Reservoir,” SCA2006-33, presented at the 2006 International Symposium of the Society of Core Analysts, Trondheim, Norway, Sep. 12-16.

6. Wang F.H.L., Braun, E.M., Kuzan, J.D., Djabbarah, N.F., Honarpour, M.M., Chiasson, C.G., and Milligan, B.E.: “Application of Unique Methodology and Laboratory Capability for Evaluation of Hydrocarbon Recovery Processes,” IPTC 12418, presented at the 2008 International Petroleum Technology Conference, Kuala Lumpur, Malaysia, Dec. 3-5.

7. Wang, F.H.L.: “Some Aspects of Wettability Alteration, Restoration, and Preservation,” Proceedings of the 3rd International Symposium on Evaluation of Reservoir Wettability and its Effect on Oil Recovery, edited by Norman R. Morrow, University of Wyoming, Laramie, WY, Sep 21-23, 1994, p.119-122.

8. Leverett, M.C.: “Capillary Behavior in Porous Solids,” SPE 941152, presented at the 1940 AIME Tulsa Meeting, Tulsa, OK, 1940. 9. Chen, A.L. and Wood, A.C.: "Rate Effects on Water-Oil Relative Permeability," SCA2001-19, presented at the 2001 International

Symposium of the Society of Core Analysts, Edinburgh, Scotland, Sep. 17-19. 10. Johnson, E.F., Bossler, D.P., and Naumann, V.O.: “Calculation of Relative Permeability for Displacement Experiments,” Trans

AIME, (1959) v. 216, p.370-372. 11. Kelleher, H.A., Braun, E.M., Milligan, B.E., Glotzbach, R.C., Haugen, E.: “Wettability Restoration in Cores Contaminated by Fatty

Acid Emulsifiers,” SCA2007-03, presented at the 2007 International Symposium of the Society of Core Analysts, Calgary, Canada, Sep. 10-12.

12. Donaldson, E.C., Kendall, R.F., Pavelka, E.A., and Crocker, M.E.: “Equipment and Procedures for Fluid Flow and Wettability Tests of Geological Materials,” DOE/BETG/IC-79/5 S.S. DOE, Bartlesville Energy Technology Center (May 1980).

Appendix A

Normalization For all of the capillary pressure, relative permeability and wettability data plots, the axes are presented on the basis of normalized saturations. A description of the normalization methodology is included below. For the primary drainage capillary pressure the normalized saturation is defined as:

( )wirr

wwn S1

SS

−=

The value of the irreducible water saturation, Swirr, used in the normalization of each limestone curve was defined as the

IPTC 13664 9

lowest value of water saturation reached in the entire limestone sample set. The value of the irreducible water saturation, Swirr, used in the normalization of each dolostone curve was defined as the lowest value of water saturation reached in the entire dolostone sample set. For the imbibition water-oil capillary pressure the normalized saturation is defined as:

( )

( )orww

wwwn SS

SSS

−−−

=min

min

1

The value of the ultimate residual oil saturation, Sorw, used in the normalization of each limestone curve was defined as the lowest value of oil saturation reached in the entire limestone sample set. The value of the ultimate residual oil saturation, Sorw, used in the normalization of each dolostone curve was defined as the lowest value of oil saturation reached in the entire dolostone sample set. The value of the minimum water saturation, Swmin, used in the normalization of each limestone curve was defined as the minimum value of water saturation on each forced imbibition capillary pressure curve. The value of the minimum water saturation, Swmin, used in the normalization of each dolostone curve was defined as the minimum value of water saturation on each forced imbibition capillary pressure curve. For the drainage gas-oil capillary pressure the normalized saturation is defined as:

( )lirr

lln S1

SS

−=

The value of the irreducible liquid saturation, Slirr, used in the normalization of each curve was defined as the lowest value of the total liquid saturation reached in the combined limestone and dolostone data sets. For the water-oil relative permeability the normalized saturation is defined as:

( )

( )orwwi

wiwwn SS1

SSS

−−−

=

The value of the initial water saturation, Swi, used in the normalization of each curve was defined as the initial water saturation in the core at the start of the primary drainage measurement. The value of the ultimate residual oil saturation to water displacement, Sorw, used in the normalization of each curve was defined by extrapolation of the oil relative permeability curve as it asymptotically approaches zero relative permeability; note that this is an approximation for the purpose of presenting the data, the imbibition water-oil capillary pressure is used to define true ultimate residual oil saturation, Sorw. For the gas-oil relative permeability the normalized saturation is defined as:

( )( )lirr

lirrlln S1

SSS

−−

=

The value of the irreducible liquid saturation, Slirr, used in the normalization of each curve was defined by extrapolation of the oil relative permeability curve as it asymptotically approaches zero relative permeability; note that this is an approximation for the purpose of presenting the data, the drainage gas-oil capillary pressure measurements in the presence of irreducible water saturation are used to define the irreducible liquid saturation, Slirr. For the USBM wettability data the normalized saturation is defined as:

( )

( )orwwirr

wirrwwn SS1

SSS

−−−

=

The value of the irreducible water saturation, Swirr, used in the normalization of each curve was defined as the lowest water saturation reached in either the preserved, cleaned or native state testing. The value of the ultimate residual oil saturation to water displacement, Sorw, used in the normalization of each curve was defined by the lowest value of oil saturation reached in either the preserved, cleaned or native state testing.

Appendix B

Wettability The USBM wettability test, introduced by Donaldson et al., consists of an oil-drive (drainage) centrifuge cycle and a water-drive (imbibition) centrifuge cycle.12 The USBM wettability index, W, is defined as: W = log10 (A1/A2) where, A1 = The area under the oil-drive capillary pressure curve

A2 = The area above the water-drive capillary pressure curve.

10 IPTC 13664

The capillary pressure curves are constructed by plotting “average” capillary pressure—which is exactly one-half of the capillary pressure at the core inlet face—versus average saturation in the core sample. When the index, W, is much greater than zero, the core is considered mixed wet with a preference for water; and when W is much less than zero, the core is considered mixed wet with a preference for oil. A wettability index near zero (positive or negative) means that the core is mixed wet with neutral preference for oil or water. The larger the absolute value of W, the greater the wetting preference. Method In preparation for the USBM tests, the plugs were first centrifuged under nitrogen at 3200 rpm at room temperature to establish the irreducible water saturation (Swi). The void pore space was saturated with degassed crude oil. The water-displacement residual oil saturation was then established by centrifuging the core plugs under brine. Next, spontaneous oil imbibition was measured by bringing the cores at the “residual” oil saturation in contact with degassed crude oil. The forced oil-drive cycle was conducted by centrifuging the cores under degassed crude oil. Next, spontaneous water imbibition was measured. The forced water-drive cycle was conducted by centrifuging the cores under brine. Finally, the core samples were extracted to determine the final water saturation. The water volume was measured by vacuum distillation with the core inside the triaxial core holder. The oil residue and salts were removed from the core by flushing the core with solvents. Final oil volume was measured by collecting oil produced from the vacuum distillation and from the solvent flushing. The cores were dried under vacuum. The absolute air permeability and the pore volume of each plug were measured under the net confining stress. The grain volume and grain density were measured after the plugs were unloaded from the core holders. Cleaned State Cores were cleaned with the goal of returning them to a strongly water-wet condition. This was done by injecting a series of hot solvents at a low flow rate through the core samples. In this study, the temperature was kept at 150 °F and the flow rate at 2 cc/min or less. The solvent sequence used was 5 PV of tetrahydrofuran (THF), 5 PV of chloroform/methanol azeotrope (78/22 volumetric ratio), and 2.5 PV of methanol. The cores were intermittently shut in with each solvent to enhance rock/fluid contacts. At the end of the injection, the remaining methanol was blown out, and the core was dried under active vacuum at 180 °F overnight. Restored State The clean cores were than aged with crude oil in the presence of connate water at the reservoir temperature (205°F) for at least six weeks to restore the original reservoir wettability. After the aging, the crude oil in the core was replaced with fresh crude oil, and the USBM test or the water-oil relative permeability test was repeated on the restored samples. At the end of the tests, the restored cores were again vacuum-distilled and solvent-flushed to determine the final saturations.

IPTC 13664 11

Tables

Table 1. Typical Data Required for Reservoir Simulation Saturation Functions over the Lifetime of a Field

Measurement Wettability State Application to Reservoir Simulation

Primary drainage capillary pressure Water-wet Define the irreducible water saturation, in some cases define the initial water distribution

Imbibition water-oil capillary pressure Native Define the ultimate residual oil saturation endpoint to displacement by water, provide an expression for the capillary force opposing the displacement of oil by water

Imbibition and secondary drainage water-oil relative permeability

Native Define the permeability to oil and water as a function of fluid saturation, measure hysteresis behavior

Drainage gas-oil capillary pressure in the presence of irreducible water saturation

Native Define the irreducible oil saturation to displacement of oil by gas, provide an expression for the capillary force opposing displacement of oil by gas

Drainage and imbibition gas-oil relative permeability in the presence of irreducible water saturation

Native Define the permeability to oil and gas as a function of fluid saturation, determine the trapped gas saturation, measure hysteresis behavior

Table 2. Lithology Groups

Lithology Group Letter A C D E F

Lithology lime mudstone / packstone lime grainstone calcareous

dolostone dolograinstone sucrosic dolostone

Primary drainage capillary pressure 10 10 5 2 3

Imbibition water-oil capillary pressure 5 5 2 2 2

Imbibition and secondary drainage water-oil relative permeability 2 2 1 1 1

Drainage gas-oil capillary pressure in the presence of irreducible water saturation

3 2 1 1 1

Drainage and imbibition gas-oil relative permeability in the presence of irreducible water saturation

2 2 1 1 1

Limestones Dolostones

Mea

sure

men

ts

Number of Tests in Each Lithology Group

12 IPTC 13664

Table 3. SCAL Measurement Conditions

Porous plate capillary pressure/ resistivity

Centrifuge relative permeability, capillary

pressure and wettability (1)

Two Phase Plug-composite core relative

permeability (2)

Three Phase Plug-composite core relative

permeability (3)

Reservoir Stress

Reservoir Temperature – 150 F

Reservoir Pressure – –

Reservoir Fluids –

(1) centrifuge measurements: reservoir net confining stress, degassed crude oil, ambient pore pressure and 150 °F(2) water-oil steady state measurements: reservoir net confining stress, reconstituted live reservoir oil, live synthetic brine, a pore pressure 220 psi greater than the bubble point and 205 °F(3) gas-oil steady state measurements: reservoir net confining stress, reconstituted live reservoir oil, live synthetic brine, a pore pressure at the saturation pressure and 205 °F

4. Wettability Comparison

Plug No. 264 187Plug Depth (ft) 6421.6 6457.8RRT Group A CKg,inf (mD) 19.8 1129

Native StateUSBM Index, W 0.11 -0.15Amott Io 0.10 0.09Amott Iw 0.00 0.01Displaceable saturation, frac 0.702 0.767

Cleaned StateUSBM Index, W 1.17 0.66Amott Io 0.00 0.00Amott Iw 0.02 0.22Displaceable saturation, frac 0.389 0.615

Restored StateUSBM Index, W 0.08 -0.19Amott Io 0.04 0.10Amott Iw 0.07 0.03Displaceable saturation, frac 0.562 0.724

IPTC 13664 13

Figure 1. Workflow for Acquiring SCAL data for Reservoir Simulation

Reservoir Facies and Geologic Information

Fluid Property Data

Define the Objective

Review Existing Data

Existing Routine and Special Core Analysis Data

Define Experimental Plan

Model Data for Reservoir Simulation

Assess Adequacy of Data Models

Select Samples

Conduct Measurements with Rigorous QA/QC

Capillary Pressure and Relative Permeability Supporting/Supplementary measurement: IFT, Wettability

Sample Acquisition

Reservoir Facies Determination

Compile Data into a Database

Assess Data Quality and Report

Figure 2. Workflow for Sample Acquisition

CT Scan the Whole Core

Develop Plug Aquisition Plan

CT Scan the Candidate Plugs

Aquire Plugs and Visually Inspect

Preserved SCAL Plug Companion Plug

14 IPTC 13664

Figure 3. Reservoir Facies Determination

Define Reservoir Facies for Each Sample

Describe Core Slab

Obtain the Minimum Data Set

Prepare and Describe Thin Section

Measure Permeability, Porosity and Grain Density on a

Companion Sample

Potential Additional Data Required

Pore Size Distribution - NMR T2 on Fully Brine Saturated Sample

Estimate of Pore Throat Size Distribution - Mercury Injection

Capillary Pressure (typically unstressed)

Figure 4. Sample Selection Process

For Each Sample within the Set: Steady-state imbibition and

secondary drainage water-oil relative permeability

For Each Sample within the Set: Drainage gas-oil capillary pressure by

centrifuge in the presence of irreducible water saturation

For Each Sample within the Set: Imbibition water-oil capillary pressure by centrifuge

For Each Sample within the Set: Measure Primary Drainage Capillary Pressure and Electrical Properties by

the Porous Plate Method

For Each Sample within the Set:

Sample Set 4. Select one plug closely analogous to each member of Sample Set 1. Allocate remaining Samples over other RRTs

Measure Formation Factor

Centrifuge Measurements of Wettability

Sample Set 5. Select one plug closely analogous to each member of Sample Set 1.

Measure USBM wettability by centrifuge

Primary Drainage Capillary Pressure and Electrical Properties by the Porous Plate Method

Steady-state drainage and imbibition gas-oil relative permeability in the presence of irreducible water

Steady State Relative Permeability

Sample Set 1. Select the most representative samples available, composite cores may only be composed of one reservoir facies

Drainage gas-oil relative permeability by centrifuge in the presence of irreducible water saturation

Centrifuge Relative Permeability and Capillary Pressure

Sample Set 2. (Gas-Oil) Select one plug closely analogous to each member of Sample Set 1. Allocate remaining samples over other RRTs

Sample Set 3. (Water-Oil) Select one plug closely analogous to each member of Sample Set 1. Allocate remaining Samples over other RRTs

Imbibition water-oil relative permeability by centrifuge

IPTC 13664 15

Figure 5. Saturation Function Modeling Workflow Define Endpoints for Each Unique RRTPrepare crossplots of the following versus absolute permeability, porosity, and (absolute permeability/porosity)^1/2

1. Irreducible water saturation from each primary drainage capillary pressure measurement (or water saturation at a constant capillary pressure maximum)

2. Ultimate residual oil saturation to water displacement from each water-oil capillary pressure measurement

3. Residual oil saturation to gas displacement from each gas-oil capillary pressure measurement in the presence of irreducible water saturation4. Trapped gas saturation from each gas-liquid flood

Develop Screening Plots for Each Data TypeFor Water-Oil measurements:

1. Primary drainage capillary pressure

2. Primary imbibition capillary pressure3. Primary drainage relative permeability (two phase only)

4. Primary imbibition relative permeability

5. Secondary drainage relative permeability

For Gas-Oil measurements in the presence of irreducible water saturation:

1. Drainage capillary pressure2. Drainage relative permeability

3. Imbibition relative permeability

Review Data and Define RRT Families

Develop Saturation Function Models for RRT Families

Figure 6. NMR Data and Photomicrograph of Sample I-86, RRT E

0

1

2

3

0.1 1 10 100 1000 10000

T2 Relaxation Time (ms)

Incr

emen

tal P

oros

ity (%

)

Moldic Pores

Between Crystal Pores

Micro-porosity

Sample I-28Porosity = 19.6%RRT E

16 IPTC 13664

Figure 7. NMR Data and Photomicrograph of Sample I-86, RRT A

0.0

0.2

0.4

0.6

0.8

1.0

0.1 1 10 100 1000 10000

T2 Relaxation Time (ms)

Incr

emen

tal P

oros

ity (%

)

Micro-porosity

Moldic Pores

Sample I-86Porosity = 8.3 %RRT A

Figure 8. Gas-brine Drainage Capillary Pressure Data for Limestone Samples

0

100

200

300

400

500

600

700

800

0 20 40 60 80

Normalized Water Saturation, percent

Cap

illar

y Pr

essu

re, p

si

100

I-38, RRT AI-86, RRT AI-90, RRT AI-91, RRT AI-43, RRT CI-45, RRT CI-48, RRT CI-55, RRT CI-57, RRT CI-62, RRT CI-63, RRT C126, RRT C155, RRT C205, RRT C228, RRT C334, RRT C

Dashed line drawn from 0 psi to the first measured pressure

IPTC 13664 17

Figure 9. Gas-brine Drainage Capillary Pressure Data for Dolostone Samples

0

200

400

600

800

0 20 40 60 80

Normalized Water Saturation, percent

Cap

illar

y Pr

essu

re, p

si

100

I-68, RRT F 80, RRT F121, RRT FI-28, RRT E414, RRT EI-30, RRT DI-32, RRT DI-47, RRT DI-67, RRT DI-74, RRT DI-79, RRT DI-80, RRT DI-87, RRT D367, RRT D

Figure 10. Water-oil Imbibition Capillary Pressure Data for Limestone Samples

-40

-30

-20

-10

0

10

0.0 0.2 0.4 0.6 0.8 1.0Normalized Water Saturation (fraction pore volume)

Face

Cap

illar

y Pr

essu

re (p

si)

RRT A, 6417.30 ft

RRT A, 6419.55 ft

RRT A, 6432.45 ft

RRT A, 6512.50 ft

RRT C, 6445.30 ft

RRT C, 6457.70 ft

RRT C, 6459.55 ft

RRT C, 6486.45 ft

RRT C, 6488.45 ft

18 IPTC 13664

Figure 11. Water-oil Imbibition Capillary Pressure Data for Dolostone Samples

-40

-30

-20

-10

0

10

0.0 0.2 0.4 0.6 0.8 1.0Normalized Water Saturation (fraction pore volume)

Face

Cap

illar

y Pr

essu

re (p

si)

RRT D, 6503.20 ft

RRT D, 6526.55 ft

RRT D, 6514.40 ft

RRT D, 6547.65 ft

RRT D, 6518.50 ft

RRT E, 6360.65 ft

Figure 12. Gas-Oil Drainage Capillary Pressure Data

0

20

40

60

80

100

120

0.0 0.2 0.4 0.6 0.8 1.0

Normalized Liquid Saturation (frac. PV)

Face

Cap

illar

y Pr

essu

re (p

si)

RRT A, 6406.65 ft

RRT A, 6450.60 ft

RRT C, 6386.45 ft

RRT C, 6376.95 ft

RRT C, 6438.30 ft

RRT D, 6367.70 ft

RRT E, 6349.30 ft

RRT F, 6516.50 ft

IPTC 13664 19

Figure 13. Steady-State Water-oil Relative Permeability, Core No. C1, RRT C

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.2 0.4 0.6 0.8 1.0

Normalized Water Saturation (frac. PV)

Rel

ativ

e Pe

rmea

bilit

y (fr

ac. k

o,Sw

i)

1st imbibition steady-state kro1st imbibition steady-state krw1st imbibition unsteady-state kro1st imbibition unsteady-state krw2nd drainage steady-state kro2nd drainage steady-state krw2nd drainage unsteady-state kro2nd drainage unsteady-state krw1st imbibition centrifuge kro

1E-05

1E-04

1E-03

1E-02

1E-01

1E+00

0.0 0.2 0.4 0.6 0.8 1.0

Normalized Water Saturation (frac. PV)

Rel

ativ

e Pe

rmea

bilit

y (fr

ac. k

o,Sw

i)

k o,Swi = 2439 md

Figure 14. Steady-State and Centrifuge Water-Oil Relative Permeability, Core No. A2, RRT A

1E-05

1E-04

1E-03

1E-02

1E-01

1E+00

0.0

Rel

ativ

e Pe

rmea

bilit

y (fr

ac. k

o,Sw

i)

0.2 0.4 0.6 0.8 1.0

Normalized Water Saturation (frac. PV)

steady-state sample k o,Swi = 68 mDcentrifuge sample k o,Swi = 44.9 mD

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.2 0.4 0.6 0.8 1.0

Normalized Water Saturation (frac. PV)

Rel

ativ

e Pe

rmea

bilit

y (fr

ac. k

o,Sw

i)

1st imbibition steady-state kro1st imbibition steady-state krw1st imbibition usteady-state kro1st imbibition usteady-state krw2nd drainage steady-state kro2nd drainage steady-state krw2nd drainage usteady-state kro2nd drainage usteady-state krwcentrifuge imbibition kro

20 IPTC 13664

Figure 15. Steady-State Gas-Oil Relative Permeability, Core No. C2, RRT C,

Figure 16. A Comparison of USBM Wettability Data Measured for Three Wettability States

1E-05

1E-04

1E-03

1E-02

1E-01

1E+00

0.0 0.2 0.4 0.6 0.8 1.0

Normalized Liquid Saturation (frac. PV)

Rel

ativ

e Pe

rmea

bilit

y (fr

ac. k

o,S

wi)

k o,Swi = 1010 mD

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.2 0.4 0.6 0.8

Normalized Liquid Saturation (frac. PV)

Rel

ativ

e Pe

rmea

bilit

y (fr

ac. k

o,S

wi)

1.0

Drainage steady-state kroDrainage steady-state krgImbibition steady-state kroImbibition steady-state krgDrainage centrifuge kro

-10

-5

0

5

10

0.0 0.2 0.4 0.6 0.8 1.0

Normalized Water Saturation

Ave

rage

Cap

illar

y Pr

essu

re (p

si)

Native W= +0.11

Cleaned W= +1.17

Restored W= +0.08

Plug No. 264, RRT A

-10

-5

0

5

10

0.0 0.2 0.4 0.6 0.8

Normalized Water Saturation

Ave

rage

Cap

illar

y Pr

essu

re (p

si)

1.0

Native W= -0.15

Cleaned W= +0.66

Restored W= -0.19

Plug No. 187, RRT C

IPTC 13664 21

Figure 17. A Comparison of Imbibition Water-Oil Relative Permeability Data for Core No. A2 in Two Wettability States – Native and Restored, RRT A

1E-05

1E-04

1E-03

1E-02

1E-01

1E+00

0.0 0.2 0.4 0.6 0.8 1.0

Normalized Water Saturation (frac. PV)

Rel

ativ

e Pe

rmea

bilit

y (fr

ac. k

o,Sw

i)

k o,Swi = 68 mD

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.2 0.4 0.6 0.8 1.0

Normalized Water Saturation (frac. PV)

Rel

ativ

e Pe

rmea

bilit

y (fr

ac. k

o,Sw

i)

1st Imbibition kro (Restored)

1st Imbibition krw (Restored)

1st Imbibition kro (Native)

1st Imbibition krw (Native)

Figure 18. A Comparison of Imbibition Water-Oil Relative Permeability Data for Core No. C1 in Two Wettability States – Native and Restored, RRT C

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.2 0.4 0.6 0.8 1.0

Normalized Water Saturation (frac. PV)

Rel

ativ

e Pe

rmea

bilit

y (fr

ac. k

o,Sw

i)

1st Imbibition kro (Restored)1st Imbibition krw (Restored)1st Imbibition kro (Native)1st Imbibition krw (Native)

1E-05

1E-04

1E-03

1E-02

1E-01

1E+00

0.0 0.2 0.4 0.6 0.8 1.0

Normalized Water Saturation (frac. PV)

Rel

ativ

e Pe

rmea

bilit

y (fr

ac. k

o,Sw

i)

k o,Swi = 2439 mD

22 IPTC 13664

Figure 19. A Comparison of Imbibition Water-Oil Relative Permeability Data for Core No. C1 in Two Wettability States – Native and Restored, with Similar Swi, RRT C

1E-05

1E-04

1E-03

1E-02

1E-01

1E+00

0.0 0.2 0.4 0.6 0.8 1.0

Normalized Water Saturation (frac. PV)

Rel

ativ

e Pe

rmea

bilit

y (fr

ac. k

o,Sw

i)

1E-05

1E-01

2E-01

3E-01

4E-01

5E-01

6E-01

7E-01

8E-01

9E-01

1E+00

0.0 0.2 0.4 0.6 0.8 1.0

Normalized Water Saturation (frac. PV)

Rel

ativ

e Pe

rmea

bilit

y (fr

ac. k

o,Sw

i)

1st Imbibition kro (Restored)1st Imbibition krw (Restored)1st Imbibition kro (Native)1st Imbibition krw (Native)