Experimental Study of Coinjection of Steam With Air or Other Coinjections Into Asphalt Ridge Tar Sands

Abstract

Abstract Coinjection of steam and nitrogen, carbon dioxide, or air at low pressures (400 psi) (2760 kPa) into Utah tar sands was studied. Slight enhancement of steam drive was observed with all three (if no combustion occurred with the air) and significant improvement occurred when the tar spontaneously ignited during air injection. Little difference in recovery between simultaneous and alternating modes of injection was observed. Introduction The thermal recovery processes of in situ combustion and steam injection have long been extensively studied by the petroleum industry for application in so-called "conventional" heavy oil reservoirs (Cf 1–11) much work on the application of these processes to tar sand deposits has also been done, with the primary focus having been on the deposits in Canada and Venezuela. In the last few years the Laramie Energy Technology Center, U.S. Dept. of Energy, and its contractors have conducted a set of theoretical, laboratory field experiments directed at understanding the application of these processes to the tar sand deposits in Utah. Both combustion processes and steam injection have processes and steam injection have been demonstrated to be capable of producing hydrocarbons from Utah tar sand deposits. As a follow up the use of gas phase co-additives - carbon dioxide, nitrogen and air - was studied. A constraint based on projected operating conditions similar to those of LETC-DOE was adopted - a maximum operating pressure of 400 psi (2758 kPa) based on the deposit depth of about 400 ft (122m). EXPERIMENTAL PROCEDURE The apparatus used was the same as was used by Watts, et al in a previous study of steam injection into tar sands, but modified to accept the coinjectants. Two high pressure cylindrical steel tubes were used 3.25 in. (0.0826m) I.D. by either 32.25 in (0.8192m) or 72.25 in. (1.8352m) long with a 0.25 in. (0.0064m) wall thickness, whereas only the shorter tube was used in the earlier experiments. All runs were conducted adiabatically except the carbon dioxide-only and nitrogen-only runs, which were heated to 442 degrees F (228 degrees C) and 472 degrees F (245 degrees C), respectively, throughout the run. These were to investigate the performances due to non-condensible gases at steam temperature. RESULTS Twenty-one runs were conducted in the short tube model to compare the performance of a steam/ co-injectant process with a straight steam displacement process and to attempt to describe the factors which determine the performance and the mechanisms which are believed to contribute to the effectiveness of the process. Because of the complex nature of the recovery mechanisms, which may involve steam distillation, thermal expansion of fluids, reduction of flood viscosity, gas drive, and miscible displacement every effort was made to isolate as many process variables as possible. First, the operating temperature and pressure for all the runs were chosen to be 447 degrees F (231 degrees C) and 400 psi (2758 kPa), respectively. Second, the steam flow rate for all the runs was set at or close to 2.75 lbm/hr (3.46 × 10 - 4 kg/sec). Third, gas injectior rates were either 0.1 or 0.5 scf/min (.0028 or.014 m3/sec) for all non-reactive gas concurrent runs and 0.05 scf/min (.0014m3/sec) for all reactive gas (air) runs. [0.05 scf/min is equivalent to an air flux of 51 scf/hr-ft2 (0.0044m3/ sec m2).] A. Steam with Nitrogen Five short tube runs were made injecting steam only. These results compared favorably with those of Watts and served to confirm those results. Four runs were conducted to test the possibility of using concurrent injection of steam with possibility of using concurrent injection of steam with nitrogen. Results of this series are summarized in Table I. The relationship between oil recovery and pore volumes of injected steam is shown in Figure 1. pore volumes of injected steam is shown in Figure 1. P. 313