Dimitrios G. Hatzignatiou

Pore-Scale Study of the Impact of Fracture and Wettability on Two-Phase Flow Properties of Rock

PORE-SCALE STUDY OF THE IMPACT OF FRACTURE AND WETTABILITY ON T H E TWO-PHASE FLOW PROPERTIES OF ROCK DMITRIY SILIN, JONATHAN AJO-FRANKLIN, JOHAN OLAV HELLAND, ESPEN JETTESTUEN, AND DIMITRIOS G. HATZIGNATIOU Abstract. Fractures and wettability are among other factors that can strongly affect the two- phase flow properties of porous media. Maximal-inscribed spheres (MIS) and finite-difference flow simulations on computer-generated structures mimicking micro-CT images of fractured rock suggest the character of the capillary pressure and relative permeability curves modification by natural or induced fracture and wettability alteration. 1. Introduction The presence of two or more different immiscible fluids (e.g., water, oil, gas) makes the pore space saturation and fluid flow multiphase. Any oil or gas recovery operation has to deal with two or three-phase flow. The same is true for subsurface injection, for example, for the purpose of CO 2 geologic sequestration. Capillary pressure and relative permeability curves are conventional concepts characterizing the capability of porous materials to store the fluids and the possibility of fluid flow and migration [16]. These curves can be determined from core laboratory experiments. Recently, this traditional approach has been complemented by digital rock methods [18, 21, 22] including pore-network modeling [2, 4, 6–8, 14, 19] and simulations on 2D and 3D micro-tomography data [9, 12, 15, 17]. The method of Maximal Inscribed Spheres (MIS) [23, 24] uses digitized micro-CT data as input information for evaluation of fluid distribution in capillary equilibrium. This method estimates two-phase fluid occupation of the pores, computes capillary pressure and relative permeability curves. The results are in agreement with experimental data [25, 27, 28]. In this study, the MIS method is employed to estimate the impact of natural fractures and wettability on capillary pressure and relative permeability curves. Micro-CT imaging of fractured rock can be difficult. In this study, we generate (via computer) a synthetic material with angular grains. We use micro CT data for samples of Bentheimer sandstone to validate the procedure. After model validation, a fracture is simulated by parting sand grains whose centers are on difference sides of the fracture plane. 2. The method of maximal inscribed spheres The idea of the MIS method of is simple [23, 27, 28]. In capillary equilibrium, the curvature of the fluid-fluid interfaces is determined by the capillary pressure, which is the pressure difference between wetting and nonwetting fluids. For a given radius of curvature, R, the nonwetting fluid saturation is estimated by evaluating the relative pore volume that can be covered by spheres that can be inscribed in the pores, and whose radii are greater than or equal to R. Such a calculation assumes a zero contact angle at the interface between the solid and the two fluids. In order to account for a non-zero contact angle, the MIS computation is modified. Namely, each inscribed Date : August 17, 2012. Key words and phrases. Pore-scale flow modeling, maximal inscribed spheres, wettability, fracture.

Mature Oilfields Screening for CO2 EOR Utilization and Storage – A Case Study

The potential use of CO2 as an Enhanced Oil Recovery (EOR) agent (either miscible or immiscible CO2-flooding) to increase oil recovery from five mature oil reservoirs (fields A through E) followed by CO2-storage has been evaluated. Primary screening of the five mature fields’ potential for EOR by CO2-flooding has been carried out by comparing the reservoir description, oil composition/ properties and reservoir conditions with literature proposed criteria. In the binary screening, the same parameters were compared with data from oil reservoirs where miscible CO2-flooding has been carried out. Based on the primary and secondary screenings, three reservoirs (C, D and E) are expected to give miscible CO2-flooding, and due to their relative homogeneous nature CO2 injection for EOR purposes appears to be promising. Of these reservoirs, reservoir D is the most interesting one since the volume of the remaining oil is the highest, similar to reservoir E, and it also has the largest degree of pressure depletion compared to reservoirs C and E. A fourth reservoir (B) is also expected to give miscible CO2-flooding, but CO2-flooding is not recommended in this field because of its heterogeneous nature and wax deposition tendency reported to have caused severe production problems. In the fifth reservoir (A) the pressure is too low to achieve miscibility between the injected CO2 and reservoir crude oil at reservoir conditions; in addition, this reservoir is heterogeneous and the risk of asphaltene deposition is expected to be high. For the three most promising fields, the Sword software was also utilized to identify the two most promising assets with respect to application of existing EOR processes and make preliminary evaluations based on the existing data. Finally, the CO2-storage potential for the two most promising assets, fields C and D, was assessed for CO2 EOR followed by CO2 storage processes in the oil-bearing formation region. This was based on the requirement of a safe CO2 storage with the reservoir unit re-pressurized back to original average field pressure. Enhancement of stored CO2 volumes could be achieved by increasing the final field abandonment pressure to levels higher than the original ones and/or utilizing the pore volume of the underlying aquifers for CO2 storage.

When Size Matters - Polymer Injectivity in Chalk Matrix

This paper investigates the injectivity of polymers in chalk matrix by experimental methods. Different polymers with variable molecular weight and structure was injected into a chalk core plug while monitoring the differential pressure. Based on the experimental results in this study, we found that it was possible to inject polymers with low molecular weight and low viscosity through chalk matrix core plugs, without causing major plugging of the rock.

Associative Polymers as Enhanced Oil Recovery Agents in Oil-wet Formations - A Laboratory Approach

Summary Associative polymers recently tested for their EOR potential in water-wet systems displayed a good potential for reducing residual oil saturation in polymer-flooded cores. In this work, an oil-wet porous medium was used to investigate these observations. A low molecular weight associative polymer was tested as a displacing agent and its ability to increase oil recovery on chemically treated oil-wet Berea cores was evaluated. Linear coreflood experiments were performed using filtered associative polymer solution as the EOR agent at standard pressure and 60°C temperature. Results from the polymer floods conducted at an established waterflood residual oil saturation (Sorw) yielded increased oil recoveries, i.e., reduced residual oil saturations, Sor, in the formation. The observed incremental oil production was a function of the injected associative polymer treatment volume; Sor decreased with increased injected associative polymer volume. It should be noted that at laboratory conditions it is often hard to establish and also distinguish a 100% water-cut; in other words, true residual oil saturation, Sorw, is often difficult to be established during water injection. Oil production profile can be discussed based on fractional flow theory, which defines the true Sorw at 100% water-cut. Whenever the produced water-cut is not precisely 100%, oil saturation in the formation is higher than the true Sorw; polymer injection with an improved mobility ratio compared to the water injection one results in an additional oil production, which could be misinterpreted as a reduction in the residual oil saturation, i.e., enhance oil production. Although this accelerated oil production is an attractive possibility (mobility control), it is not an EOR process. Our results are in agreement with previously reported observations in water-wet media related to the EOR nature of the injected associative polymer as opposed to the traditional mobility control of other, either synthetic or organic, polymers. The same results showed that the polymer mobility reduction is highly affected by the injected polymer velocity at the lower spectrum of velocity values and a correlation for the velocity dependent mobility reduction was developed. Finally, during the injection of the associative polymer, a column of oil-polymer emulsion was formed gradually in the separator which caused some difficulties and introduced uncertainties in the separator’s fluids level readings, and thus eventually in the fluids saturation evaluation. Resistivity data obtained in real time were used to correct for the overestimated values of oil production during polymer injection attributed to the formation of the oil/water emulsion.

Simulation of Three‑Phase Capillary Pressure Curves Directly in 2D Rock Images

Pore-scale modeling of three-phase capillary pressure in realistic pore geometries could contribute to an increased knowledge of three-phase displacement mechanisms and also provide support to time-consuming and challenging core-scale laboratory measurements. In this work we have developed a semi-analytical model for computing three-phase capillary pressure curves and the corresponding three-phase fluid configurations in uniformly-wet rock images encountered during tertiary gas invasion. The fluid configurations and favorable entry pressure are determined based on free energy minimization by combining all physically allowed gas-oil, gas-water, and oil-water arc menisci in various ways. The model is shown to reproduce all threephase displacements and capillary entry pressures that previously have been derived in idealized angular tubes for gas invasion at uniform water-wet conditions. These single-pore displacement mechanisms include (i) gas invasion into pores occupied by oil and water leading to simultaneous displacement of the three fluids, (ii) simultaneous invasion of bulk gas and surrounding oil into water filled pores, and finally (iii) the pure two-phase fluid displacements in which gas invades pores occupied by either water or oil. The proposed novel semi-analytical model is validated against existing analytical solutions developed in a star-shape pore space, and subsequently employed on an SEM image of Bentheim sandstone to simulate three-phase fluid configurations and capillary pressure curves at uniform water-wet conditions and different spreading coefficents. The simulated fluid configurations for the different spreading coefficients show similar oil layer behaviour as previously published experimental three-phase fluid configurations obtained by computed microtomography in Bentheim sandstone. The computed saturation paths indicate that three-phase oil-water capillary pressure is a function of the water saturation only, whereas the three-phase gas-oil capillary pressure appears to be a function of two saturations. This is explained by the three-phase displacements occurring in the majority of the simulations, in which gas-water interfaces form immediately during gas invasion into oil- and water-saturated pore shapes.