Experiments were conducted to study the feasibility of using propane as a steam additive to accelerate oil production and improve steam injectivity in the Hamaca field, Venezuela. The experiments utilized a vertical injection cell into which a mixture of sand, oil and water was tamped. The Hamaca oil sample had an oil gravity of 8°API and a viscosity of 25,000 cp at 50°C. The injection cell was placed inside a vacuum jacket, set at the reservoir temperature of 50°C. Superheated steam at 170°C was injected at 3. 5 ml/min (cold-water equivalent) simultaneously with propane at the top of the cell. The cell outlet pressure was maintained at 50 psig. Four propane:steam mass ratios were used, namely, 0:100 (pure steam), 2. 5:100, 5:100, and 10:100. Produced liquid samples from the bottom of the cell were collected, treated to break emulsion, and analyzed to determine oil and water volumes, and density and viscosity of the oil. The oil was subjected to SARA analysis to determine the degree of in-situ oil upgrading. Composition of the produced gas was determined using a gas chromatograph. Each run lasted three hours. Experimental results indicated the following. First, with steam-propane injection, start of oil production was accelerated by 17% compared to that with pure steam injection. In the field, this could translate into significant gains in discounted revenues and reduction in steam injection costs. Second, steam injectivity with propane as an additive was up to three times higher than that for pure steam injection. Third, oil production acceleration and injectivity increase were practically the same for runs with propane as a steam additive (irrespective of the propane:steam mass ratios). Propane appears to be a viable steam additive at propane:steam mass ratios as low as 2. 5:100. Introduction Several studies have been carried out to test the effect of injecting steam along with other gaseous additives. In this section, previous experiences with the combined use of steam and gaseous additives are presented. Redford (1982)1 conducted experiments to study the effect of adding carbon dioxide, ethane and/or naphtha in combination with steam. Results showed that the addition of carbon dioxide or ethane improved oil recovery. Further recovery was reached when naphtha was added. Harding et al. (1983)2 presented both experimental and simulation results suggesting that the co-injection of carbon dioxide or flue gas with steam yielded higher recoveries when compared to pure steam injection. Stone and Malcolm (1985)3 performed several tests to study the benefit of injecting carbon dioxide along with steam. Higher production rates were obtained for the case of steam-carbon dioxide injection. Good agreement was found between the experimental data and numerical simulation results. Stone and Ivory (1987)4 carried out further investigations using the model from Stone and Malcolm. 3 This time, experiments with CO2 presoak and CO2 co-injection with a solvent were conducted. They found that under certain conditions, carbon dioxide pre-soaking increased oil recovery above the conventional CO2-steam injection. Nasr et al. (1987)5 presented results of experiments conducted to test the effect of injecting CO2, N2 and flue gas with steam. Both continuous and cyclic injection were tested. The addition of gasses increased bitumen recovery. The use of CO2 resulted in higher oil recoveries when compared to that with N2 and flue gas injection. Frauenfeld et al. (1988)6 presented results showing that for oils without an initial gas saturation, co-injection of CO2 with steam was capable of improving oil recovery over that obtained with steam alone. On the other hand, when an initial non-zero gas saturation was present, co-injection of CO2 was not beneficial.

Cold Heavy Oil Production with Sand (CHOPS) has been widely and successfully applied for the last three decades in the Heavy Oil Belt region that straddles the provinces of Alberta and Saskatchewan in Canada. As its name suggests, the method relies on continuous production of sand to improve the recovery of oil from the reservoir. In CHOPS, a significant pressure drawdown around the wellbore is created by using progressive cavity pumps, which causes the loosely consolidated formation to fail, creating increased permeability channels, usually called wormholes, through which, a slurry-like mixture of sand, oil and water flows. Many attempts have been made to use conventional numerical reservoir simulators to model the CHOPS process. However, many of the commercial finite-difference reservoir simulators do not incorporate capabilities to model the complex geomechanical processes responsible for the failure of poorly consolidated formations in CHOPS. To circumvent these limitations, several approaches have been proposed. The most common relies on explicitly defining high permeability channels that radiate from the producing wells in an attempt to mimic wormholes created during CHOPS production. In this paper, we present a different, more rigorous approach that relies on the coupling of a finite-element geomechanical simulator with a finite-difference reservoir simulator. In the coupling process, the geomechanical simulator uses the pressure gradients calculated by the reservoir simulator to determine changes in the stress regime of the reservoir. In the case of CHOPS, these changes cause failure in the loosely consolidated formation, which in turn induces sand production with a corresponding increase in porosity and permeability. The new porosity and permeability values in the affected gridblocks are then fed back to the reservoir simulator, which is now capable of incorporating the effects of formation failure into fluid flow calculations. This process is then repeated at user-controlled intervals during the course of the simulation. The methodology has been validated by successfully history matching the production data from a section of a heavy oil field operated by Husky Energy in Western Canada. In this paper we compile the data integration efforts to create a coupled geomechanical model and the results of the history match.

TX 75083-3836, U.S.A., fax 01-972-952-9435. * AbstractStudies have been conducted to evaluate the feasibility of steam-propane injection for the Hamaca heavy oilfield (8°API) and for the Duri intermediate oilfield (20°API). The experiments involved injecting steam simultaneously with propane at the top of a vertical cell containing a mixture of sand, oil and water. Superheated steam was injected at 170ºC and 3.5 ml/min (Hamaca) and 260°C and 5.5 ml/min (Duri), with the cell outlet pressure at 50 psig (Hamaca) and 500 psig (Duri). Runs were made with propane:steam mass ratios (PSR's) ranging from 0:100 to 10:100. With steam-propane injection (PSR 0.05): (i) start of oil production is accelerated by 23% (Hamaca) and 30% (Duri), (ii) steam injectivity is up to three times higher, and (iii) steam front velocity is higher, indicating greater partitioning of the distilled fractions into the propane stream that appears to act as an efficient carrier gas.A simulation study was conducted using a 1D 48 grid-cell model with a ten pseudo-component oil model for the Hamaca oil. Satisfactory history matches of the experimental runs were obtained; C 7 -C 10 appear to play a significant role during steam-propane injection. A reservoir simulation study was performed using a 9 × 5 × 10 3D Cartesian model representing one-eighth of a 10-acre 5-spot pattern. Five-year forecast runs were made with well steam injection rate of 600 BPDCWE. Production is enhanced with steam-propane injection (PSR 0.05) as follows: (i) start of oil production is accelerated by 20%, (ii) oil production peak with steam-propane injection (1100 STB/D) is significantly higher than that with pure steam injection (690 STB/D), (iii) oil production acceleration increases with propane content, (iv) oil recovery is about * previously with PDVSA-Intevep three times that with pure steam injection, indicating improved sweep/displacement efficiency with steam-propane injection.

Unconventiona estimated world hat the volum recovery factorTo help unlock fracturing in ho geometry (not state in the dra geomechanical fractured tight g Traditionally, t system to a pla presented in th orientation and pressure deplet are ignored.