This paper discusses the recent patent application filed by EnCana Corporation on the use of air injection to improve the performance of steam assisted gravity drainage (SAGD). EnCana's SAGD/Air Injection process employs standard SAGD well-pair infrastructure. It optimizes between the ability of steam to preheat the reservoir during SAGD and the superior (follow-up) oil displacement efficiency of in-situ combustion. Operationally, air injection is initiated after thermal communication has been established between well-pairs with steam. One interesting feature of this operating strategy is that down-hole bulk separation of oil and gas occurs which facilitates (a) efficient monitoring and control of the combustion, (b) design of surface facilities, and (c) corrosion mitigation. Laboratory combustion tube tests are presented that confirm the ability to initiate and sustain combustion, as well as mobilize residual oil saturation to steam, within a SAGD chamber. These experiments were initialized at oil saturations and conditions representative of those in a steam chamber. The residual oil saturations were determined from a full-hole core taken in the vicinity of a mature SAGD well-pair at Foster Creek. Numerical simulations of post-SAGD air injection are presented that suggest the ability to displace and produce oil banks between well-pairs and that recovery factor can be increased up to 8% of the original oil-in-place over conventional SAGD. The simulations show oil production rates and recovery factor are expected to increase with higher air injection rates. However, instantaneous air-oil ratios, which are indicative of operating costs, also increase. Thus there is an optimum continuous air injection rate that maximizes profitability. Simulations further indicate that it is possible to recycle flue gases in the injection stream without affecting oil recovery. Introduction Steam injection, to date, has been the most successful 'in situ heavy oil and bitumen recovery method. In 2004 there were several commercial cyclic-steam stimulation (CSS) projects in Alberta, Canada, producing a combined average of 27.7 103 m3/d bitumen1. In the same year, Alberta's bitumen production by steam-assisted gravity drainage (SAGD) averaged 11.1 103 m3/d. Bitumen production attributable to both processes has grown substantially since then. These achievements are due to technological innovations that have overcome several geological and reservoir challenges associated with steam injection. And, it is expected that in situ bitumen producers will continue to seek enhancements with the goal of reducing steam-oil ratio (SOR). Two historical developments suggested the SAGD optimization process proposed in this paper. First: From 1979 to 1984, BP Canada tested air injection as a follow-up process to fracture-assisted cyclic steam stimulation in the Clearwater formation at Wolf Lake2. It was expected that only 17% of the bitumen could be recovered using CSS and that combustion as a follow-up process would increase the recovery factor. Bitumen recovery factor at the pilot increased from 15% with CSS to a cumulative of almost 30% with in-situ combustion. The equivalent SOR was reduced from approximately 6.2 to 2.3. It should be noted that air injection was successful in this application because the bitumen was first preheated and mobilized with steam. Second: During SAGD operations, there comes a point where the cumulative SOR begins to increase indicating that it is no longer economic to continue steam injection. Injecting a non-condensable gas at this stage, to utilize the existing heat energy in place, can prolong oil production3. This significantly reduces the operating costs compared to continued steam injection. This paper describes the potential for implementing air injection as a follow-up to SAGD. It is shown that under controlled conditions, air injection promotes superior volumetric sweep efficiency and recovery factor in comparison with continuous steaming or with using methane as blow-down gas.

The Underground Test Facility is designed to test the application of the steam-assisted gravity drainage process for the in situ recovery of bitumen, using horizontal wells drilled from tunnels below the pay zone.

RESERVOIR ENGINEERING Mobility control in dynamic gravity segregation flow systems BEN 1. NZEKWU BP Resources Canada Limited and DOUGLAS W. BENNION Hycal Energy Research Laboratories Limited ABSTRACT Experiments were carried out in a vertical sandpack to study the effect of gravity segregation on liquid displacement by a miscible liquid, a gas and a combination of the two. The advantages of using a foaming agent during simultaneous gas-liquid injection were also investigated. A gamma-ray attenuation technique was adapted to track the shapes of the moving displacement front and monitor in situ saturation profiles.The results show that high gas mobility and low density lead to early gas breakthrough and poor sweep efficiency. Even in miscible liquid-liquid displacements a density difference of 0.079 glcm3 leads to gravity override or underride by the displacing liquid. Dur-ing simultaneous gas-liquid injection the presence of gas increases the severity of the override/underride effect and reduces the displacement efficiency ofthe miscible liquid. In situ generation of foam is an effective method of controlling the mobility of the gas and improving the conformance of the displacement front. Introduction The purpose of this study was to investigate the characteristics of gravity segregating systems and the use of foaming agents to con-trol the mobility of air and water injected simultaneously into a liquid-saturated porous medium.One of the major factors responsible for poor conformance and low recovery efficiency of several displacement processes(l) is gravity segregation caused by a density difference between displac-ing and displaced fluids. For example gas-driven processes such as C02-flooding, alternating water-gas injection, steam injection and in situ combustion are affected by varying degrees of segregated flow behaviour.Several laboratory experiments and numerical studies of fluid flow in porous media have confirmed the importance of a density difference in the occurrence of segregation. Slower rates of segre-gation were observed in brine-liquid systems than in brine-air systems(2). With miscible liquids the length of the mixed zone was influenced by gravity segregation, and a greater reduction in dis-placement efficiency was observed when the mixed zone was narrow compared to the height of the experimental model (3). The presence of a broad transition zone tended to nullify the influence of a difference in density(4-5). Gravity segregation is generally

Data collected at 10 temperature observation wells were analyzed to evaluate the implementation of a combined process of cyclic steam stimulation followed by pressure-uplblowdown combustion for the in-situ recovery of bitumen. The temperature profiles at four observation wells, along the northeast/southwest fracture plane (on-trend direction), confirmed the height and orientation of the vertical fractures during cyclic steam stimulation. Six off-trend wells indicated that the heat transfer away from the fracture plane was predominantly conductive, with varying amounts of convective flow. During the in-situ combustion phase, frontal temperatures > 2, 200°F [1200°C] were detected. Data also indicated that the vertical location of the combustion front can be controlled by either gravity or mobility effects. The off-trend heating was improved as a result of higher convective heat fluxes. Reservoir GeologyThe Clearwater formation, at a depth of about 1,475 ft [450 m], is an unconsolidated sand with several upward-coarsening trends called Sands CI, C2, and C3. Sand Cl has a net pay thickness of 13 ft [4.0 m], a gross thickness of 19 ft [5.9 m], a porosity of 32%, and a bitumen saturation of 58 %. This zone contains an occasional interbedded tight streak. Sands Cl and C2 are separated by 12.5 ft [3.8 m] of shale, which frequently is separated into two shale layers by 2.6 ft [0.8 m] of sand. Sand Cl is overlain by 16 ft [4.9 m] of shale that forms an effective seal with the Lower Grand Rapids formation.Sand C2 has a net pay thickness of9 ft [2.8 m], a gross thickness of 14.8 ft [4.5 m], a porosity of 29%, and a bitumen saturation of 68 %. Horizontal permeabilities are high, but vertical permeability is reduced by up to two tight streaks. Sands C2 and C3 are separated by a thin shale that varies in thickness from 0 to 3 ft [0 to 1 m].Sand C3 has a net pay thickness of 53 ft [16.2 m], a gross thickness of 65 ft [19.8 m], a porosity of33%, and a bitumen saturation of 65 %. Horizontal permeability is high (1 to 3 darcies) but vertical permeability is reduced by frequent clay laminae and up to three indurated ferroan calcite-cemented layers dispersed within the sand. The lower 9 to 12 ft [3 to 4 m] consists of silts and clays, and the perforated intervals were selected several feet above this zone. The basal shale is more than 10 ft [3 m] thick and forms an effective seal between the Clearwater and McMurray formations.The bitumen has a dead-oil viscosity of 100,000 cp [100 000 mPa' s] at the original reservoir temperature of 60°F [15°C]. Field data have shown that the bitumen can be produced when the viscosity is reduced to about 100 cp [100 mPa . s] by heating to 212 OF [l00°C].

A new recovery process which uses concentric coiled tubing has been developed to improve production capabilities in heavy oil. The process, called Single A . t d G 't D' PatPend (SW Well -Steam SSIS e ravi y ramage -SAGD) relies on continuous injection of high quality steam into the toe of a horizontal well while simultaneously producing mobilized oil from the same wellbore. The well is therefore dual completed with both, production tubulars and Insulated Concentric Coiled TubingPatPend (ICCT).Having completed an extensive study into the design and fabrication of ICCT, a number of single well, SAGD installations have already been completed and are in production. This paper will review this process and will document production results from some of these early wells.