Summary
Oils that are potential candidates for in situ combustion recovery processes are often screened by means of their oxidation characteristics: in particular, the kinetics of the ignition process and the transition from low-temperature to high-temperature oxidation through what is known as the "negative temperature gradient region." These characteristics are readily studied in ramped-temperature oxidation tests, which involve the controlled heating of recombined, oil-saturated cores in a one-dimensional plug flow reactor under a flowing stream of air (or oxygen-containing gas). The purpose of these tests is to study the global oxidation behavior and reaction kinetics under controlled conditions, with the end purpose of providing realistic data for incorporation into a numerical simulator which can be used to predict field performance.
A ramped-temperature oxidation apparatus was used to conduct a detailed, two-year parametric study of the oxidation characteristics of Athabasca Oil Sands bitumen. The text matrix involved various levels of pressure, gas injection rate, oxygen content of the injected gas, and maximum ramptemperature. This paper details the principal findings for the 45-test study;especially the need to maintain high reaction temperatures >380°C) in order to mobilize and produce heavy oils under conditions of dry in situ combustion. Design considerations and operational guidelines for successful field projectsarising from the results of this study are also discussed.
Introduction
In order to successfully exploit the vast potential of processes based on the injection of air or an oxygen-containing gas for the recovery of conventional and heavy oils, it is necessary to understand the nature of the oxidation reactions which are involved. The traditional definition of in situ combustion, which is based on the high-temperature combustion of a coke-like fuel, does not explain the combustion behavior which is observed in many field projects or even in laboratory combustion tube experiments. For this reason, a number of experiments have been developed which concentrate on the global oxidation kinetics. These studies normally involve exposing the crude oil to a programmed rate of heating while in contact with the oxidizing gas. The oxidation kinetics are then observed using effluent gas analysis techniques,1–7 and differential thermal techniques such as the differential thermal analysis (DTA) work of Vossoughi et al.,8 the pressurized differential scanning calorimetry (PDSC) studies of Phillips et al.9 and Belkharchouche and Hughes,10 and the accelerating rate calorimetry (ARC) technique of Yannimaras et al.11
Previous investigations of the oxidation reactions which occur during in situ combustion processes have shown the existence of at least two temperature ranges over which the oxygen uptake rates are significant. 2,4-7While Kisler and Shallcross have reported that the light (40.2°API) Australian oil which they studied exhibited at least three temperature ranges over which localized maxima in the oxygen uptake rate were observed, the majority of heavy oils for which oxidation data have been reported show only two distinct local maxima in the oxidation rates. For convenience, the two temperature ranges where elevated oxygen uptake or energy generation rates are observed are denoted as the low-temperature oxidation (LTO) and high-temperature combustion(HTC) regions. For heavy oils, the range of temperatures associated with the low-temperature oxidation region is roughly from 150 to 300°C, while the high-temperature combustion region generally corresponds to reaction temperatures in the range from 380 to 800°C. The transition temperature range which falls between the low-temperature oxidation and high-temperature combustion regions is characterized by reduced oxygen uptake and energy generation rates. The lower temperature portion of this transition range in which the oxygen uptake and energy generation rates decrease with increasing temperature is the "negative temperature gradient region" (NTGR).
This behavior is illustrated in Fig. 1, which is the temperature history for a test involving a heater temperature of 350°C (near the upper end of the NTGR). This test, which was previously described by Moore et al.,12 shows that a distinct low-temperature reaction zone formed when the temperature was approximately 140°C and it propagated through the core for a short period of time as the heater continued its ramp towards the setpoint maximum temperature of 350°C At the end of the propagation period, the centerline temperatures remained very close to the heater temperature as the latter was increased over the temperature interval from 280 to 330°C It is apparent from the small temperature differences between all of the centerline locations and the heater that energy generation over this temperature interval was very low. A high-temperature reaction zone started to form when the temperature at the first thermocouple location attained 355°C.
Fig. 2 provides the oxygen uptake history for the same test, and the data show that there were also two distinct periods of high oxygen uptake rates. The first period corresponds to the time that the lower-temperature reaction zone propagated through the core, and it is apparent that the prime mode of oxygen uptake is by reactions which do not result in the formation of carbon oxides. These reactions have been denoted as LTO reactions, although it should be noted that hydrogen conversion to water (which is normally classified as a combustion reaction) is included as a LTO reaction. Oxygen uptake rates associated with the second period correspond to the propagation of the high-temperature reaction zone. At these higher temperatures, oxygen consumption is primarily associated with the formation of carbon oxides. Oxygen uptake by LTO reactions is also significant, but this reflects the inclusion of hydrogen conversion to water as a LTO reaction. In essence, the oxidation reactions associated with the high-temperature propagating reaction zone are those which are normally designated as high-temperature combustion, in that the primary products are carbon oxides and water.
Numerical simulation of n-dodecane spray combustion based on OpenFOAM
