Feali, Mostafa (U Of New South Wales) | Pinczewski, Wolf Val (U. of New South Wales) | Cinar, Yildiray (U. of New South Wales) | Arn, Christoph (U. of New South Wales) | Arns, Ji-Youn (Australian National Univ.) | Francois, Nicolas (Australian National Univ.) | Turner, Michael (Digital Core Laboratories Pty Ltd) | Senden, Tim | Knackstedt, Mark
It is now widely acknowledged that continuous oil spreading films observed in two-dimensional glass micro-model studies for strongly water wet three-phase oil, water and gas systems are also present in real porous media and result in lower tertiary gas flood residual oil saturations than for corresponding negative spreading systems which do not display oil spreading behavior. However, it has not been possible to directly confirm the presence of spreading films in real porous media in threedimensions and little is understood of the distribution of the phases within the complex geometry and topology of actual porous
media for different spreading conditions. This paper describes a preliminary study using high resolution X-ray microtomography to image the distribution of oil, water and gas after tertiary gas flooding to recover waterflood residual oil for two set of fluids, one positive spreading and the other negative spreading, for strongly water wet conditions in Bentheimer sandstone.
We show that for strongly water-wet conditions and a positive spreading system the oil phase remains connected throughout the pore space and results in a low tertiary gas flood residual oil saturation. The residual oil saturation for the corresponding negative spreading system is significantly higher and this is shown to be related to the absence of oil films in this system. The presence of films for positive spreading systems and the absence of such films for negative spreading systems is further confirmed by the computation of the Eurler characteristic for each phase.
Dehghan Khalili, Ahmad (U Of New South Wales) | Arns, Christoph Hermann (University of New South Wales) | Arns, Jiyoun (U. of New South Wales) | Hussain, Furqan (U. of New South Wales) | Cinar, Yildiray (U. of New South Wales) | Pinczewski, Wolf Val (Australian National University) | Latham, Shane (Saudi Aramco) | Funk, James Joseph
High-resolution Xray-CT images are increasingly used to numerically derive petrophysical properties of interest at the pore scale, in particular effective permeability. Current micro Xray-CT facilities typically offer a resolution of a few microns per voxel resulting in a field of view of about 5 mm3 for a 20482 CCD. At this scale the resolution is normally sufficient to resolve pore space connectivity and transport properties. For samples exhibiting heterogeneity above the field of view of such a single high resolution tomogram with resolved pore space, a second low resolution tomogram can provide a larger scale porosity
map. The problem then reduces to rock-typing the low resolution Xray-CT image, deriving viable porosity-permeability transforms from the high resolution Xray-CT image(s) for all rock types present, and upscaling of the permeability field to derive a plug-scale permeability.
In this study we characterize spatially heterogeneity using overlapping registered Xray-CT images derived at different resolutions spanning orders of magnitude in length scales. A 38mm diameter carbonate core is studied in detail and imaged at low resolution - and at high resolution by taking four 5mm diameter subsets, one of which is imaged using full length helical scanning. Fine-scale permeability transforms are derived using direct porosity-permeability relationships, random sampling of the porosity-permeability scatter-plot as function of porosity, and structural correlations combined with stochastic simulation. A range of these methods are applied at the coarse scale. We compare various upscaling methods including renormalization theory with direct solutions using a Laplace solver and report error bounds.
We find that for the heterogeneous samples permeability typically increases with scale. Conventional methods using basic averaging techniques fail to provide truthful vertical permeability due to large permeability contrasts. The most accurate upscaling technique is employing Darcy's law. A key part of the study is the establishment of porosity transforms between highresolution and low-resolution images to arrive at a calibrated porosity map to constraint permeability estimates for the whole core.
At the conclusion of flooding an oil- or gas-bearing reservoir, a significant fraction of the original hydrocarbon in place remains trapped. In addition to determining the amount of residual phase, knowledge of its microscopic distribution within the rock pore space would allow a better understanding of recovery mechanisms, and the design and implementation of improved or enhanced recovery processes. While the importance of the pore scale structure, mineralogy and wettability in dictating the residual phase distribution is widely acknowledged, little quantitative information on these properties and dependencies has been directly available. To this end, we describe an ongoing interdisciplinary study, bridging the core-, pore- and molecular scales using x-ray microtomographic imaging, petrographical imaging and wettability imaging. The experimental techniques used are reviewed, emphasizing the registration technology which enables spatial alignment and integration of 2D SEM-based information with 3D µ-CT images. Application of these techniques to visualization of pore scale oil and brine populations is presented, with a particular focus on characterizing native state carbonate plugs. In parallel, direct visualization of the alteration of rock surface chemistry at the pore- and molecular scales due to oil exposure is presented for macroporous and microporous reservoir carbonates. This interdisciplinary approach provides the foundation for more systematic development of strategies to increase recovery, in particular by tuning wettability.
The amount of residual hydrocarbon phase in a reservoir rock after flooding has obvious importance in determining the completeness of secondary recovery and the target for tertiary recovery. Further knowledge of the distribution of this trapped phase within the rock pore space would facilitate a more transparent and systematic approach to improved or enhanced recovery. In flooding experiments on reservoir core material, the core scale distribution of residual can be quantified (e.g. by magnetic resonance imaging or computerized tomography), however deeper insight into its configurations at the pore scale is necessary to better understand the underlying displacement mechanisms. Most methods previously pursued to characterize the residual oil phase at the pore scale use either idealized 2D micromodels (Lenormand et al. 1983) or destructive techniques on model liquids in rock (pore/blob casts) (Chatzis et al. 1983). The observations from these studies helped to ascribe rules for meniscus advance through connected pores in simple network models of multiphase flow. The enormous advances in x-ray micro-computerized tomography (µ-CT) over the past decade have greatly increased the scope for imaging rock pores and calculating properties from the digitized 3D images. This has allowed network models to more realistically incorporate the geometry and topology of pores in rock cores. However, the corresponding characterization of mineralogy and wettability to specify the pore scale distribution of molecular-scale surface chemistry adorning the pore walls is lacking, and thus the pertinent contact angles to be used in modeling are not known. Further, the ability to 3D visualize with µ-CT the residual phase occupancy in individual pores of rock cores is required to test the predictive power of such models and guide any augmentation of the mechanisms of pore displacement in real rocks to obtain better agreement.
Ghous, Abid (U. of New South Wales) | Knackstedt, Mark Alexander (Australia National University) | Arns, C.H. (Australia National University) | Sheppard, Adrian (Australia National University) | Kumar, Munish (Australia National University) | Sok, Rob (Australia National University) | Senden, Timothy (Australia National University) | Latham, S. (Australia National University) | Jones, A.C. (Australia National University) | Averdunk, H. (U. of New South Wales) | Pinczewski, Wolf Val
The prediction of hydrocarbon recovery is related to both the detailed pore scale structure of core material and fluid interfacial properties. An increased understanding of displacement efficiencies and overall recoveries requires an ability to characterize the pore structure of reservoir core in 3D and to observe fluid distributions at the pore scale.
Micro-CT imaging is capable of acquiring 3D images of the pore structure of sedimentary rock with resolutions down to the micron scale. This allows the 3D pore-space of many reservoir rock samples to be imaged at the pore scale. The 3D porespace of tighter clastics and carbonate core material includes a significant proportion of microporosity—pores at the submicron scale which are not directly accessible via current micro-CT capabilities. Porosity at all scales can affect fluid flow, production, recovery data and log responses. It is important to characterize pore structure and connectivity in a continuous range across over six decades of length scales (from nm to cm) to better understand these petrophysical and production properties. In this paper we describe 2D and 3D imaging studies of reservoir core via micro-CT coupled with complementary petrographic techniques (thin section, mercury intrusion) and high resolution focused ion beam (FIB) scanning electron microscopy studies of a range of reservoir core. Results are given which illustrate the importance of pore structures at varying scales in determining petrophysical properties.
Microtomography is then used to observe pore scale fluid distributions within the core material. Displacement experiments under controlled wettability conditions are undertaken. The local pore-scale fluid distributions identified via 3D tomographic imaging experiments. These results provide insight into the role of rock microstructure in determining recovery and production characteristics.
Kumar, Munish (Australian National University) | Senden, Timothy (Australian National University) | Sheppard, Adrian P. (Australian National University) | Latham, Shane (Australian National University) | Knackstedt, Mark Alexander (Australian National University) | Cinar, Yildiray (U. of New South Wales) | Pinczewski, Wolf Val (U. of New South Wales)
In this paper we describe a technique based on radio frequency plasma treatment in H2O vapour to reproducibly clean and modify the surface energy of clastic and carbonate core material allowing the establishment of well defined wettability conditions. We present micro-tomographic observations of the pore-scale fluid distributions in strongly water wet clastic and carbonate cores. We then establish mixed-wet states in the same cores using controlled hydrophobation. Micro-tomography is again used to reveal the three-dimensional geometry and topology of water and oil wet regions. The tomographic data shows that under water wet conditions at intermediate saturations larger pores are predominantly oil filled while smaller pores remain water wet. We perform displacement experiments using clastic and carbonate cores at well defined wettability conditions and report measurements of resistivity index. These methodologies may provide insight into the role of rock microstructure and surface energy variability in determining recovery and production characteristics of oil and gas reservoirs.
This paper presents the results of drainage capillary pressure and relative permeability measurements made on cores of different size -bulk volumes ranging from 0.5 to 12 cm3. The aim of the measurements was to obtain reliable experimental data which can be used to validate the predictive value of micro-CT based network models for capillary pressure and relative permeability. Micro-CT based network models typically use realistic networks constructed from the X-ray images of the rock samples representing bulk volumes of the order of 0.3 cm3.
Experimental data for drainage capillary pressure were obtained using the centrifuge technique. The results for the largest cores were comparable to data obtained on the same sample using the porous plate technique. Relative permeability data were obtained by history matching unsteady state displacement data. Homogeneous outcrop sandstones (Berea and Bentheim) and carbonate (Mt. Gambier) were used in the experiments. Air-brine and oil-brine fluid-systems were used for drainage capillary pressure and relative permeability measurements, respectively. The relative permeability data were compared with those predicted from empirical and geometry based models using capillary pressure data.
Good agreement was obtained for the drainage capillary pressure measured on all samples used. The residual saturations obtained from the cores used in the displacement experiments were also in good agreement. The models were found to predict relative permeability of oil and water with varying degrees of success. For water relative permeability, the Pirson model predicts the experimental data successfully while the Corey, Corey-Brooks/Burdine and van Genuchten/Burdine models provide the best predictions for oil relative permeability. The results demonstrate for the first time that reliable drainage capillary pressure and relative permeability measurements can be made on small sandstone and carbonate cores of size similar to that used for micro-CT-imaging.
This paper presents experimental data for co-current spontaneous imbibition into cores having bulk volumes from 0.1 to 12 cm3. Simple experiments of brine imbibing into air-filled cores were carried out. Homogeneous sandstone cores (Berea and Bentheim) and a carbonate core (Mt. Gambier) were used in the experiments. The experimental data were scaled using the scaling laws reported in the literature.
The results demonstrate that reliable experimental data of spontaneous imbibition can be obtained for the small cores of homogeneous porous rocks. Such data are of immense interest for validating the predictive value of network models based on micro-CT images of rock fragments with bulk volumes as small as 0.3 cm3. The data for cores of different sizes were satisfactorily scaled using five different methods[6, 11, 14, 15, 18]. The recovery models proposed by Ma et al. and Viksund et al. produced an excellent match for the normalized gas recovery data. Although the Li and Horne model successfully correlated the imbibed water volume as a function of time, the model failed to correlate the normalized recovery data. A comparison of the scaled data with data previously reported for water-gas systems showed excellent agreement.
Accurate prediction of spontaneous imbibition is crucial in optimization of oil and gas recovery processes, e.g., assessing water injection performance in oil fields and determining residual gas saturation in gas fields. Network models based on micro-CT images promise great potential to better understand and more accurately predict the dynamics of imbibition processes.
Advances in extracting representative networks from micro-CT images of porous materials have improved the predictive capabilities of the network models[2, 3]. The networks extracted from the images are used in models to predict multiphase flow properties such as relative permeability, capillary pressure and spontaneous imbibition process. All these predictions need to be validated using laboratory data in order to test the predictive value of the network models. The imaged rock fragments are small compared to conventional cores having bulk volumes of the order of 0.1 cm3.
A number of attempts have been made to compare limited laboratory measurements with network model predictions[1, 4, 5]. Although the results of these comparisons are encouraging, the networks used were derived using computer generated process-based reconstructions of the porous medium which have significantly different properties to the cores used in the actual experiments. Moreover, the experimental data used were obtained on conventional cores having bulk volumes of at least 10 cm3, which are several orders of magnitude larger than those of the rock fragments used to produce the micro-CT images.
Scaling of spontaneous imbibition has long been used as a predictive tool for estimating the field performance of water-wet, fractured reservoirs subjected to water flood/drive. In this approach, simple imbibition tests on small reservoir cores are scaled to estimate reservoir performance. A number of scaling approaches have been proposed and tested against experimental data [6-11]. The bulk volumes of rock samples were varied in order to study the effect of plug size on the imbibition process (Table 1). The table shows that the smallest core used in these tests had a bulk volume that is several orders of magnitude larger than the core fragments typically used for micro-CT imaging.
The aim of this paper is to provide simple and well-defined experimental data for spontaneous imbibition in order to demonstrate that reliable experimental data can be produced on small core plugs which are of comparable size to rock fragments used in micro-CT imaging. This data will be used for testing and validation of image based network models. The comparisons with scaling laws for water-air systems will also be of interest to applications in geothermal and gas reservoirs.
2. Scaling of Spontaneous Imbibition
Spontaneous imbibition of water into a matrix block is a very complex process and depends on many factors such as permeability, wettability, shape, and size of the matrix, boundary conditions, and interfacial tension and viscosities of the fluid system. Detailed reviews of hydrocarbon recovery by spontaneous imbibition have been reported[11, 12].
Olafuyi, Olalekan Adisa (University of New South Wales) | Sheppard, Adrian P. (Australia National University) | Arns, Christoph Hermann (Australia National University) | Sok, Robert Martin (U. of New South Wales) | Cinar, Yildiray (Australia National University) | Knackstedt, Mark Alexander (U. of New South Wales) | Pinczewski, Wolf Val
This paper presents comparisons between drainage capillary pressure curves computed directly from 3D micro-tomographic images (micro-CT) and laboratory measurements conducted on the same core samples. It is now possible to calculate a wide range of petrophysical and transport properties directly from micro-CT images or from equivalent network models extracted from these images. Capillary pressure is sensitive to rock microstructure and the comparisons presented are the first direct validation of image based computations. The measured data include centrifuge and mercury injection drainage capillary pressure for fired Berea, Bentheimer and Obernkirchner sandstones and unfired Mount Gambier carbonate. The measurements cover a wide range of porosities and permeabilities. The measurements are made on core samples with different diameters (2.5 cm, 1.5 cm, 1 cm and 0.5 cm) to assess the effect of up-scaling on capillary pressure measurements. The smallest diameter samples were also used to obtain the 3D micro-CT images. Good agreement is obtained between the experimental measurements and direct computations on 3D micro-CT images.
Recent advances in imaging technology now make it possible to routinely image rock microstructure in 3D at the pore scale. Coupling this with an ability to computationally predict petrophysical and multiphase flow properties directly on the 3D digitised tomographic images or on equivalent networks (digital core technology) results in a powerful tool to interpret conventionally measured core data and to extend the range of available data by examining rock fragments which cannot be tested by conventional means (sidewall cores, drill cuttings and unconsolidated or poorly consolidated rocks). A number of studies (Auzerias et al., 1996; Arns et al., 2001; Knackstedt et al., 2004) suggest that computations of permeability, formation factor and mercury injection capillary pressure on digitised image of a small rock fragments cut from a core plug are consistent with laboratory measurements performed on the same plug even though the computations and measurements are performed at significantly different scales.
Micro-CT imaging is currently limited to small sample sizes; pore scale imaging on most materials requires resolutions of 3-5 microns, and image size is limited to approximately 2000 cubed—this limits the sample sizes for imaging studies to 5mm-1cm which is significantly smaller than conventional core plug scale. Moreover, computational times usually limit the computational domain used to a smaller sub-set of the imaged volume. Conventional laboratory measurements, on the other hand, are carried on core plugs and composite cores at scales several orders of magnitude larger than that for the image based computations.
We investigate this scale effect by performing laboratory measurements at a number of different scales from the core plug scale down to a scale closer to that imaged using micro-CT. We limit the investigation to what are usually considered to be homogeneous or model rock types. These are the rock types normally used to validate image based calculations of a wide range of rock properties.
A new dynamic network model is used to investigate the effects of displacement rate and wettability on imbibition relative permeability. The network geometry and topology is representative of a Berea-type sandstone. In contrast to existing quasi-static network models where snap-off, the major pore-scale trapping mechanism in imbibition, is suppressed by contact angle alone, the dynamic model introduces displacement rate as an additional snap-off inhibiting mechanism. The network model is used to analyse the complex rate dependence of relative permeability and residual saturation displayed by laboratory measured data reported in the literature.
Bauget, Fabrice (Institut Français du Petrole) | Arns, Christoph Hermann (Australia National University) | Saadatfar, Mohammad (Australia National University) | Sheppard, Adrian (Australia National University) | Sok, Rob (U. of New South Wales) | Turner, Michael (Australia National University) | Pinczewski, Wolf Val | Knackstedt, Mark Alexander
Rock formation permeability is arguably the most important flow parameter associated with subsurface production and injection. Its importance is reected by the number of techniques (well-log evaluation and correlation, core measurement and well testing) used to estimate it. Clearly permeability should be linked to other porous media properties (e.g., surface area, porosity, pore/grain size). There have been numerous attempts over the last sixty years to establish a relationship between the permeability of a rock and other characteristic rock properties. Most empirical approaches for the prediction of permeability, which has units of length squared, propose a function of a characteristic length scale, formation factor (tortuosity) and porosity. The most widely used is the Carmen-Kozeny equation where the length scale is equated to the hydraulic radius (pore volume / pore surface area). Other length scales used include a critical pore radius associated with mercury injection experiments (Katz-Thompson), lengths associated with NMR relaxation (e.g., T2) and grain size and rock fabric measures.
To uncover the relationship between permeability and other pore scale properties requires directly measuring the geometric and transport properties of the pore system. This is now possible with 3D microtomographic imaging (Knackstedt et.al. SPE 87009, Arns et.al. SPE 90368). In this paper we describe a comprehensive study of permeability correlation across a range of rock types. We directly compute permeability, formation factor, NMR response, hydraulic radius, rock fabric and texture, pore size and capillary pressure on 3D microtomographic images of 39 porous materials including over 30 clastic and carbonate samples from a wide range of reservoirs. Subsampling enables one to generate more than 6500 \independent" samples. Empirical correlations between permeability and various length scales are tested for a range of lithotypes including unconsolidated sands, homogeneous sands, consolidated reservoir sands, limestones and reservoir carbonates. We and that the most robust length scale correlation is based on the critical pore radius. All correlations which use the Formation factor as a measure of tortuosity give good predictions. Empirical correlations for permeability based on grain size perform well for permeabilities greater than one Darcy.