Several different versions of the transfer functions evolved from classical Mattax and Kyte's dimensionless group and Aranofsky et al.'s abstract relationship were tested. Another transfer function derived analytically based on power law relationship between recoverable oil and time was also included in the testing process. The exponents in all these transfer functions reflect the strength and type of the transfer. The recovery curves obtained from the spontaneous imbibition of different aqueous phases (brine and surfactant) into cylindrical rock samples saturated with different types of oil were used to fit the transfer functions. The exponents yielding the best fit to experimental data were obtained and correlated to the effective parameters such as the viscosity of oil, matrix permeability, IFT, matrix size, and wettability using multivariable regression analysis. The correlations developed were analyzed for the rock and oil types, and IFT. It was observed that the exponential relationships were more suitable for synthetic and processed oil samples whereas the power law transfer functions were more applicable for crude oil cases. It is hoped that the analysis provided in this paper would facilitate the selection of proper transfer function type for performance estimation of naturally fractured reservoirs. Introduction Reliable description of matrix-fracture interaction is the key to an accurate simulation of naturally fractured reservoirs if matrix is heavily contributing to the production. During the injection of any material for EOR purpose, the injectant will preferably flow in the fracture. Matrix oil will be recovered by an interaction between the fracture and matrix. Obviously, the same type of transfer function is not expected to be applicable to every rock, fluid and process type. Suitable transfer functions need to be determined for particular rock and fluid properties. This entails a critical analysis and testing of matrix-fracture transfer functions that would eventually lead to a classification of them for different fluid, rock and process types. Oil in the matrix is typically recovered by capillary imbibition if the matrix is water wet and enough amount of water is supplied in the fracture. For such systems, the accuracy of the dual porosity model depends on the accurate description of matrix fracture transfer function1. A general description of matrix-fracture transfer function is difficult to propose due to complexity of the phenomenon and significant number of parameters involved in the process. In this paper, several different types of matrix-fracture interaction functions were tested for different rock and fluid properties. After an extensive review of the previously proposed matrix-fracture transfer functions, the exponents that control the rate of spontaneous imbibition in the exponential and power law transfer functions were correlated to matrix and fluid properties. Experimental data obtained using a wide variety of oil, rock and aqueous phase types were used for this purpose.

The Solvent Aided Process (SAP), described previously in literature, is an improvement to SAGD that promises to enhance the economics of bitumen/heavy oil recovery projects and reduce their impact on the environment. In SAP, a small amount of hydrocarbon solvent (such as a low molecular weight alkane) is introduced as an additive to the injected steam during SAGD. The viscosity of the oil thus is reduced due to solvent dilution in addition to heating. SAP can significantly improve the energy efficiency of SAGD, thus reducing the heat requirement. Cenovus's field trials of SAP, discussed elsewhere, have shown the practical upside of this process. Modeling predicts that the higher the amount of solvent used in SAP, the better is the performance (rates, energy intensity) of the recovery process. Besides rate of Bitumen production, economics of SAP depend on the availability and cost of solvent. Although the existing literature has discussed deterministic variations in solvent input, it is largely silent on how much solvent is the right amount of solvent in SAP.This paper contains discussion of using optimal amount of solvent with steam in SAP. The discussion is based on modeling and compares performance of the scheme under various solvent injection strategies. It explores effect of temporal variation in the concentration of solvent at the vapor-liquid interface as well as of pulsed solvent injection on the performance of the process.

Surfactant-steam process (SSP) is a novel and potentially cost-effective process that utilizes a small amount of surfactant coinjected with steam to enhance the oil recovery of steam assisted gravity drainage (SAGD) well pairs. The mechanism of this process involves interfacial tension (IFT) reduction, reservoir rock wettability alteration, oil relative permeability enhancement, and in-situ emulsification. SSP is expected to result in oil rate acceleration, steam-to-oil ratio (SOR) reduction and enhanced ultimate oil recovery factor. Analogous enhancement is expected if the SSP is combined with other steam-based or steam-solvent processes. This paper provides an introduction to this concept and presents a unique protocol that has been developed for screening surfactants for co-injection with steam in SAGD process. In particular, the paper presents a scientific approach to surfactant selection for SSP applications, describes the conditions in which the surfactants needs to be deployed within the reservoir, and also predicts the potential synergies if use of different classes of surfactants is made. Novel experimental design on different aspects of surfactant-steam phase behavior indicates the optimum surfactant concentrations for field trial applications. Lab testing of selected surfactants on typical Canadian oilsands sand packs shows an improved incremental oil recovery factor (RF) in the range of 6 to 16% (for different tested surfactants) compared to a SAGD base case. SSP simulations were conducted for one of the surfactants that were tested in the lab. The simulation results indicate that this particular surfactant on average accelerated the oil rate by 15% in the first 30 months of SSP operation, increased the ultimate oil RF by 10%, and reduced the cumulative steam-to-oil ratio (CSOR) by almost 11% relative to a SAGD base case. In addition, a sensitivity analysis was conducted to investigate the effect of surfactant concentration co-injected with steam. The simulation results suggest that there is an optimum concentration for a given surfactant that needs to be explored through lab investigations and field trials. It is evident that once the SSP is successfully developed, the use of surfactant promises to improve environmental performance and project economics of the in-situ oilsands recovery.

This experimental study was designed to provide a detailed understanding of the blocking mechanism of heavy oil-in-water emulsions injected into a porous medium. The process in mind is one where a created emulsion will break near the well bore or at some pre-determined distance from it, in order to provide an effective, stable plug against, for instance, water or gas coning.Well-characterized heavy oil-in-water emulsions were injected into micro-models and their behavior was recorded in visualization experiments. The effect of droplet-to-pore size ratio, droplet stability, and surfactant type and concentration were studied. It was observed that blockage took place because of size exclusion. Droplets may coalesce and produce a larger droplet due to the local high shear rate or to surfactant adsorption on the porous medium. Also, emulsion droplet size distribution, emulsion viscosity and oil/water interfacial tension increased as the surfactant content decreased, resulting in higher capillary pressure across the trapped droplets.The effect of oil type, rock permeability, injection velocity, and wettability alteration were also studied. The experiments showed that an oil-in-water emulsion was effective in sealing unconsolidated cores for long periods of time. Emulsions carrying more viscous oils could resist higher pressures. Also, conditioning the medium with pre-flush solutions predictably affected the depth to which an emulsion may penetrate into a porous medium. Surfactant and alkaline based pre-flush solutions may enhance an emulsion penetration depth significantly. However, the emulsion may break down and emplace at a desired depth as a result of using low pH solutions. PETROLEUM SOCIETY CANADIAN INSTITUTE OF MINING, METALLURGY & PETROLEUM 2A novel sealant that uses heavy oil-in-water emulsion to block the near well bore matrix has been developed. Stable reduction in permeability to other fluids was observed as the plug withstood 42,500 kPa/m (about 1,800 psi/ft) pressure gradients. Criteria are defined for field application of this blockage phenomenon.

No abstract