Digital Core Analysis and Pore-Network Modeling in a Mature-Field Project

For the planning of an enhanced-oil-recovery (EOR) project in a major mature oil field in east Malaysia, an extensive routine-core-analysis (RCA) and special-core-analysis (SCAL) program has been performed on unconsolidated clastic reservoir rocks. In view of the limited availability of homogeneous core plugs of suitable size for coreflooding experiments and for “conventional” SCAL laboratory investigations, a complementary analysis of petrophysical properties was performed on the basis of the acquisition of high-resolution 3D microcomputed-tomography (MCT) images.

Introduction

As part of an extensive core study for this field, the “conventional” core-analysis program was complemented by acquiring 3D images and applying digital-core-analysis (DCA) and pore-network-modeling (PNM) methods at different scales with the following main objectives:

1. Visualize the status and monitor potential changes of pore morphology, fluid distributions, and wettability during the handling and execution of laboratory experiments on unconsolidated-friable-rock samples.
2. Evaluate and model the impact of sample heterogeneity and wettability changes on petrophysical properties.
3. Provide petrophysical properties for individual distinct rock types at pore scale for heterogeneous or irregularly shaped samples that are not suitable or available for RCA/SCAL and flooding experiments.
4. Scale up drainage and imbibition predictions of subplugs from heterolithic/laminated plugs to full-plug scale and, potentially, further to reservoir scale.

However, because of the lack of preserved core material from earlier wells, and the relatively poor recovery of new conventional cores from recent infill wells in this field, alternative data-acquisition options were investigated in this pilot test on the basis of the significant improvements in, and encouraging results from, DCA and PNM technologies in recent years.

Procedures

Sample Selection. Approximately 22 m of core was recovered from the shallower reservoir sections of two wells (three cores from Well A from a depth interval between 700 and 800 m and three cores from Well B from a depth interval between 1100 and 1300 m), but only 8 m was suitable for plugging. The low-resolution medical computed-tomography images from the whole core indicated a high extent of rubble and broken and highly fractured unconsolidated sections.

From the best sections, approximately 80 standard core plugs of 1.5- and 1-in. diameter were drilled under liquid nitrogen from the frozen cores for further RCA/SCAL.

From the petrographic analysis, three major facies and rock types were identified: sandstone, heterolithics, and mudstones. The main focus was on the sandstone reservoir facies, which was further divided into massive, laminated, and bioturbated subfacies.

A first set of 11 samples with 1.5‑in. diameter and up to 2.5-in. length (that were not suitable for conventional RCA/SCAL because of their heterogeneous or fractured nature) was investigated with experiments at a commercial core-­analysis laboratory. Additional MCT scans were acquired to complement formation-damage (FD) studies; MCT was performed on 10 plugs before FD experiments, and three plugs were investigated after FD experiments to visualize potential fines and grain migration during the filtration tests.

Acquisition of 3D X-Ray MCT Images. A high-resolution and large-field X-ray MCT facility is used for image acquisition. Both the X-ray source and the detector are optimized for high resolution and maximal field of view. The sample sizes can vary from full 1.5-in. core plugs to subplugs down to 2 mm. The voxel resolution is 1/2,000 times the sample size, typically 20 µm for a 1.5-in. plug and 1 to 5 µm for 2- to 10-mm sample volumes. An acquisition time of 10 to 20 hours is typical for the 20,483 tomograms collected. This equates to the collection of 23 gigabytes of data.

For special high-resolution region-of-interest (ROI) scanning, a new generation of high-fidelity helical microtomogaphy equipment was used.

Discussion and Results

High-Resolution Scanning in Native State. In a first screening, MCT images were acquired for all 11 plugs in the as-received or quasinative state. To evaluate the porosity and to distinguish the potential fluid types in the pores, a higher-resolution scan is required. This can be achieved either by physically drilling a new subsample and scanning this subsample separately at a higher resolution or by applying a nondestructive acquisition for ROI scanning. Fig. 1 (above) is an example representing a slice from the tomogram at the plug scale, slices from the 3D image at higher resolution, and two stills from a 3D animation illustrating the distribution of the gas (red) and water in the plug imaged in the native state.

Visualization and Analysis of In-Situ Saturation and Wettability in Native- and Restored-State Samples. In one sample from each rock type, an innovative technique was used to identify where the remaining-oil saturation may be concentrated in the plug; this incorporated imaging the plug in the as-received state and imaging the same plug after cleaning, to register and to visualize the hydrocarbon-rich regions.

Fig. 2 shows an example of slices through the MCT images before and after cleaning for a plug characterized as laminated sand. In this sample, it is observed that most of the larger pores are swept by water or gas (air), and the remaining-oil saturation is mainly concentrated in the tighter zones of the sample.

Wettability Investigations With Field-Emission Scanning-Electron Microscopy (FESEM). For further investigation of the wettability status, additional FESEM images were acquired. FESEM measurements have been conducted on four samples in native- and restored-state conditions for comparison and visualization of potential wettability changes. The presence or absence of asphaltene deposits can be taken as an indicator of wettability. On the regions of pore walls with altered wettability, the adsorbed/deposited asphaltenic aggregates exhibit a characteristic nodular film of nanoparticles often of a size on the order of 10 nm, which aggregates or merges into larger features. All four samples show similar wettability behavior: (1) quartz surfaces have a patchy asphaltene deposit, (2) the illite also shows evidence of asphaltene deposits, and (3) both illite and quartz have areas devoid of asphaltene deposits.

For the simulation and modeling studies, the samples can be interpreted as being intermediate-wet, which could be confirmed with wettability studies on one sister plug. The laboratory measurements indicate an intermediate-wet sample. For the required input of contact angles, advancing, receding, and equilibrium contact angles were measured with a sessile drop of crude oil in contact with asphaltene-rich model substrates (quartz, kaolinite, illite) immersed in synthetic seawater. The measured contact angles for the substrates also confirmed the assumed intermediate-wet-state assumptions.

For a discussion of the analysis of grain and pore-throat size from MCT, please see the complete paper.

Petrophysical Properties. Once the grain and pore partitioning in the 3D tomograms is performed, the standard static petrophysical properties (porosity, electrical conductivity, and hydraulic conductivity) can be calculated from the 3D pore images. The analysis can be performed in all directions, which is very important for anisotropic reservoirs and direction-dependent tensor properties such as permeability. In the basic evaluations, three axial directions are investigated, with the z-direction perpendicular to and the x–y plane parallel to the bedding plane. In addition to hydraulic permeability values, electrical properties can be calculated directly from the MCT images.

Comparison of Conventional-RCA Results With Image-Derived Properties. The RCA results of the porosity and permeability measurements of the core plugs with conventional RCA show the same distribution and trend compared with the eight core plugs with the MCT-derived PNM properties. These results indicate that a careful selection of statistically representative sections for the imaging and PNM analysis is required to generate the true porosity and permeability that can be compared directly with the full-sample RCA results. This result has an important practical implication for the petrophysical evaluation of different rock types that are characterized by significantly different grain and pore-size distributions: A permeability/porosity trend can be obtained even from small rock fragments that are unsuitable for standard laboratory measurements, but all these reservoir properties can be derived directly from 3D MCT-image analysis.

Generation of Synthetic Drainage and Imbibition Curves for P_cand k_rel. The key challenge of this pilot study of DCA and PNM applications was the generation of synthetic capillary pressure curves (P_c) and relative permeability curves (k_rel) for a heterogeneous sample that was not suitable for conventional multiphase-flow experiments. Furthermore, the PNM simulations allow quick investigation of the sensitivities of the P_cand k_rel properties from the petrophysical and fluid input parameters, which cannot be executed experimentally.

The results of the image-generated multiphase-flow curves depend strongly on the scale of investigation, choice of plug region, and plug orientation. It is important to characterize the different types of heterogeneity to recognize the effect of fluid flow and capture the relevant flow physics at different scales by use of upscaling.

To extract the effective two-phase properties (P_c and k_rel) of the plug sample, a two-phase steady-state upscaling method was used. The steady-state upscaling method was also used to incorporate the contribution of the mud drapes.

Drainage and imbibition simulations were undertaken on a clean subset of the image. The analysis was then upscaled to incorporate the effect of the muddy or silty laminations and the denser inclusions.

Sensitivity Analysis. A sensitivity analysis was undertaken as an example to illustrate the effect of the input parameters of wettability, flooding rate, and fluid type on the prediction of capillary pressure and gas/oil relative permeability curves.

Comparison of the relative permeability for the clean subsets of a heterolithic sample and a laminated sample under strongly water-wet (WW) condition and under mixed-wet (MW) condition was used for simulation. The results are summarized in Fig. 3a. Clear differences are observed under WW and MW conditions, because of variations in the local contact angles (+/–10°) and wettability conditions (45–60% WW surfaces) while preserving the Amott index at zero.

All laminated and heterolithic samples exhibit open, clean regions of high porosity. The relative permeability is compared from two subsets of laminated sample and one heterolithic sample. The results are shown in Fig. 3b. The relative permeability shows a similar response because the simulations were conducted under the same wettability conditions for subregions of good-reservoir-quality sand within plugs.

Fig. 3c shows the results of relative permeability for the three samples (two laminated and one heterolithic) after inclusion of laminated silty regions and mud drapes as observed in the samples. Variations in results for the three samples are observed in simulation at the plug scale; these differences are primarily attributable to the plug heterogeneity (the orientation of silty laminations or mud drapes in the plugs).

These simulated k_rel and P_c curves showed similar results compared with the laboratory-measured k_rel curve from a clean sister sample, but for a direct comparison and further anchoring and calibration of the endpoints, the MCT imaging and PNM analysis of the identical plug are required.

For a discussion of the FD study, please see the complete paper.

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper OTC 24772, “Application of Digital Core Analysis and Pore-Network Modeling on the Basis of 3D Micro-CT Images for an Enhanced-Oil-Recovery Project in a Mature Oil Field in East Malaysia,” by W. Nur Safawati, Bt. W.M. Zainudin, Zahidah M. Zain, and Lutz Riepe, Petronas, prepared for the 2014 Offshore Technology Conference Asia, Kuala Lumpur, 25–28 March. The paper has not been peer reviewed. Copyright 2014 Offshore Technology Conference. Reproduced by permission.