Summary Experiments on initial stages of the steam-assisted gravity drainage (SAGD) process were carried out, using two-dimensional (2D) scaled reservoir models, to investigate production process and performance. Expansion of the initial steam chamber, its shape and area, and its temperature distributions were visualized with video and thermal-video pictures. The relationship between isotherms and steam-chamber interface was investigated to study the drainage mechanism. Temperature at the expanding steam-chamber interface was observed to remain nearly constant at close to 80°C. The effect of vertical spacing between the two horizontal wells on oil recovery was also investigated. For the Conventional SAGD case, oil production rate increased with increasing vertical spacing between the wells; however, the lead time for the gravity drainage to initiate oil production became longer. The results suggest that vertical spacing between the wells can be used as a governing factor to evaluate production rate and lead time in the initial stage of the SAGD process. Based on these experimental results, the SAGD process was modified; the lower production well was intermittently stimulated by steam injection, in conjunction with continuous steam injection in the upper horizontal injector. With the modified process (named SAGD-ISSLW), the time to generate near-breakthrough conditions between two wells was shortened, and oil production was enhanced at the rising chamber stage compared with that of the Conventional SAGD process. Introduction The SAGD process was developed by Butler and his coworkers.1,2 In Canada, the SAGD process has proven successful for recovery of bitumen, as demonstrated in the reports on the UTF projects (Phases A and B).3,4 Chung5 and Chung and Butler6,7 reported experimental results for the SAGD process with scaled and visual reservoir models. Furthermore, Chow and Butler8 reported numerical simulation results matching Chung's experimental results5 using Computer Modelling Group Ltd.'s STARS™ simulator. Recently, Mukherjee et al.9 successfully forecasted the performance for Phase B of the UTF project. Butler10 gave a review of the SAGD process. An operational problem of the SAGD process for oil sands reservoirs is the lead time required to generate a steam chamber in near-breakthrough conditions between the two horizontal wells before the production stage. In this study, we first examined characteristics of the Conventional SAGD process, especially the expansion rate of the steam chamber by gravity drainage and the effects of well spacing. It was found that by using smaller vertical spacing between the two horizontal wells, the lead time was reduced, while production rate after breakthrough became lower. As shown in this paper, results from our investigation demonstrated that a more economical SAGD operation could be achieved by a simple modification involving selective intermittent stimulation of the lower horizontal producer by steam injection. For this process, called the SAGD-ISSLW process, the lower horizontal well was modified to enable intermittent stimulation by steam injection along the well design reported by Liderth.11 As such, this well served two functions: selective intermittent steam injection, and continuous fluid production. Steam from this lower well was injected intermittently to prevent steam breakthrough. The experiments using this process were compared to those using the Conventional SAGD process. The results showed that the SAGD-ISSLW process was successful in reducing the lead time to generate the steam chamber in the initial stage. The quick generation of the steam chamber plus the intermittent steam injection provided the advantage of allowing larger vertical spacing to be set between the two horizontal wells. Intermittent steam injection also led to another advantage of enhancing the instability of the steam-chamber interface near the ceiling, and thus it could be used to control the expanding steam chamber more effectively. Experimental Apparatus and Procedures Many experiments were performed in scaled 2D reservoir models with porous packing materials to investigate steam-chamber behavior and oil-production mechanisms, with respect to heat and mass-transfer phenomena. To compare process performance, steam-injection and fluid-production rates were measured. The experimental apparatus was designed and dimensions were determined according to the scaling criteria given in Refs. 5 and 6. Major experimental conditions and the purposes for the four phases of experiments are listed in Table 1. One difference between our experiments and those of Chung and Butler6 is the process used to preheat the reservoir by circulating steam through two wells before injecting it into the reservoir. We did not use preheating in our experiments, as we believed that it would interact with the well structure and materials, and as a result, heat not only the reservoir but also both side plates of the 2D models. Fig. 1 shows a schematic of the experimental apparatus, including the reservoir model. The apparatus consisted of a water pump, steam generator, steam accumulator, 2D scaled reservoir model, production-control mechanism, visualization system, and the data-acquisition system. All components, except the data-acquisition and video-camera systems (DAS), were mounted on a flat steel table designed and built in-house. Scaled Reservoir Model. The 2D scaled reservoir models (Fig. 2) were designed to represent a vertical segment of an oil sands reservoir. The models were made from smooth and transparent acrylic-resin plates 20 mm in thickness. The transparent side plates allowed visualization of the displacement of the oil in the steam chamber. Glass beads (diameter: 0.18 to 0.25 mm, average 0.21 mm) and heavy oil were packed between the two side plates. Motor oil (COSMO #1000, molecular weight=490 g/gmol, ?=998 kg/m3) served as the heavy oil in the experiments. Viscosity of the COSMO #1000 oil and Athabasca bitumen (extracted by Suncor Inc.) was measured as a function of temperature with a rheometer (Shimadzu, RM-1), as shown in Fig. 3. Viscosity of this oil was 93 000 mPa's (or 93 Pa's) at an initial temperature of 20 to 25°C, and 120 mPa's at a steam temperature of 106°C. Thus, the viscosity of the heavy oil used in the present experiments is roughly one-fifth that of the bitumen.

No abstract

TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractExperiments and numerical simulations on initial stages of the steam-assisted gravity drainage (SAGD) process were carried out. Experimental studies used two-dimensional scaled reservoir models to investigate fluids flow characteristics in the steam chamber and production fluids. The rise and growth of the initial steam chamber were observed and visualization of micro-phenomena at inclined interface on the side of the steam chamber with a high-resolution optical-fiber scope was carried out. Very fine water droplets of 0.01 mm order in size were observed at the interface between steam and heavy oil phases. These droplets entered into heavy oil phase and created emulsion together with the heated oil, which flowed down and was produced from the production well. It was successfully demonstrated that these micro phenomena has an influence on chamber expansion rate in horizontal direction and oil production rate.The numerical simulation of the SAGD process was also performed. The thermal simulator, CMG's STARS TM , was used to simulate the experimental steam chamber growth at the initial stage. The simulation used the two-components (water and heavy oil) black oil, three-phase (water, heavy oil and steam) and three-dimensional numerical model for the physical model. The results from the history-matched numerical simulation were found to be in good agreement with those of the experiments for oil production and steam chamber shape by using the Intermediate3-Stone1 wettability model represents fluids behavior cumulating the microscopic phenomena at the chamber interface. Furthermore, a new process named Surfactant-SAGD injecting a surfactant before starting steam injection to enhance the communication between two wells and mobility of the production fluids was tested.

The experiments on initial stages of steam assisted gravity drainage(SAGD) process have been carried out using two-dimensional scaled reservoir models to investigate its production process and performance. Rising or growing process of the initial steam chamber, its shape and area, and temperature distributions have been visualized by using video and thermal-video pictures. As a drainage mechanism, the relationship between isothermal lines and chamber interface have been presented. The temperature on the interface, where the chamber was expanding, was maintained at almost constant temperature of 80 C. Furthermore, the effect of vertical well spacing between two horizontal wells on oil recovery has been investigated. For the case of usual SAGD, oil production rate increases with increasing vertical well spacing, however the leading time to start oil production by gravity drainage becomes longer. The results show that the well spacing may be a representative length for initial stage of the process. Based on these experimental results, a modified SAGD process by adding intermittent steam injection from the lower production well have been proposed. By applying the modified process, the time to generate near break-through condition between two wells was relatively shorten, and oil production was enhanced at the stage of rising chamber compared with that of usual SAGD process. Introduction There are vast heavy oil and oilsands reserves not fully exploited, because it is not easy to produce heavy oil efficiently and economically. However, steam-assisted gravity drainage (SAGD) process has been successfully applied to oilsands fields. The process has been developed by Butler and his co-workers. Their ideas was to overcome the problems associated with the highly viscous bitumen by gravity drainage in steam chambers generated by displacement of heavy oil. As shown in the reports on the UTF projects (phase A and B) in Canada, the SAGD process (see Fig. 1) has proven to be very superior process for the recovery of the bitumen due to its high recovery factor. Chung and Chung & Butler have reported experimental results for SAGD process with scaled and visual reservoir models. Furthermore, Chow & Butler have reported the numerical simulation results matching the Chung's experimental results with the STARSTM. Recently, Mukherjee have reported the successful forecast of the performance for the phase B of the UTF project. Butler gives an excellent review of the SAGD process. In this paper, the SAGD process operated by steam injection from upper well and production from lower well like that of UTF project, is hereinafter referred as "usual SAGD." A problem of the usual SAGD process for oilsands reservoirs is leading time to generate a steam chamber in near break-through condition between two horizontal-wells before production stage by rising and expanding chamber. The more economical SAGD production should be achieved by a modified process to shorten the period of initial stage and enhanced steam injection for effective usage of steam generation and production facilities. First, we have investigated characteristics of the usual SAGD process, especially expanding rate of steam chamber by gravity drainage and effects of vertical well spacing on it. It was found that by using shorter vertical spacing between two wells, leading time is reduced while production rate after break-through becomes lower. Based on experimental results, a modified process has been proposed to start oil production earlier and enhance oil production at higher rate after break-through. In this process, steam is injected from both of upper and lower wells. Then, the lower well has both functions of production and steam injection, which is similar to the single SAGD well developed and reported by Liderth. The steam is injected intermittently from the lower well, because steam injection holes and oil production holes at the well are quite close to prevent steam break-through. P. 467

This study provided a model for calculating the aquifer transmissibility, the CO2 injection rate, the inner diameter of the injection well, and the number of wells for liquid CO2 disposal in the aquifer. The possibility of disposing liquid CO2 in an aquifer just beneath the sea floor was shown, based on the equilibrium lines in the pressure and temperature map. Our study focused on the feasibility of liquid CO2 disposal below the critical temperature because CO2 can be denser in the low‐pressure range (below the critical temperature) than above the critical temperature. An aquifer about 200 m under the sea floor, at a water depth of around 500 m (700 m below the sea surface), will serve for liquid CO2 disposal. In the aquifer the absolute pressure is approximately 7.3 MPa, sea‐floor temperature is about 4–6°C, and aquifer temperature is about 15–20°C. Therefore, it can be assumed that CO2 dissolves in the aquifer water, and liquid CO2 replaces the water. This means that under the previous conditions, more CO2 can be injected into the aquifer compared to supercritical conditions. Furthermore, by forming a cap of CO2 hydrates, the sediment between the sea floor and the aquifer, prevents CO2 leakage to the sea. Even without the cap, liquid CO2 and CO2 hydrates form at the sea floor, so the CO2 exerts no large environmental impact.