In-Situ Combustion Process Study With a Combined Experimental/Analytical Approach
- L.R. Sibbald (U. of Calgary) | R.G. Moore (U. of Calgary) | D.W. Bennion (U. of Calgary)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Engineering
- Publication Date
- August 1991
- Document Type
- Journal Paper
- 295 - 302
- 1991. Society of Petroleum Engineers
- 2.4.3 Sand/Solids Control, 1.6.9 Coring, Fishing, 4.3.4 Scale, 4.6 Natural Gas, 5.8.5 Oil Sand, Oil Shale, Bitumen
- 2 in the last 30 days
- 314 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 35.00|
Analysis of combustion-tube data produced from experiments performed underrealistic reservoir conditions is currently the most valid method ofinvestigation the in-situ combustion process. This paper describes an approachthat uses a differential moving-frame descriptive model to analyze data fromstabilized combustion processes produced in combustion-tube experiments. Theapproach is applied to a set of dry-air combustion runs. This study revealedconsistent values for the fuel-combustion-reaction-kinetics parameters, showedthat oxygen consumption was not confined to the high-temperature (300 degreesC) combustion zone, and gave insight into the relationship between injected-gasflux and the distribution of energy liberation rate with implications forprocess stability.
The complexity of in-situ combustion is familiar to any engineer orresearcher who has evaluated field or laboratory performance of this recoveryprocess. The goal of this study was to investigate small-scale relationshipsamong the mechanisms involved in an experimentally observed dry forward in-situcombustion process without large reliance on assumptions concerning thesemechanisms. In the approach taken, a moving-frame descriptive analytical modelof the high-temperature region >300 degrees C) was applied to results froma series of combustion-tube experiments. The model was not designed to predictcombustion behavior, but to find a method to characterize stable combustion,particularly in the high-temperature region. This would allow thehigh-temperature zone to be decoupled from the multiphase fluid flow andlow-temperature-oxidation (LTO) reaction mechanisms of the process. This typeof analysis aids the understanding of in-situ combustion and, it is hoped, thedevelopment of improved numerical simulators.
In-situ combustion processes involve the movement of a combustion front andassociated zones through a rock matrix containing oil. Fig. 1 shows the regionstypically associated with dry, steady-state forward combustion-frontpropagation. These regions serve as a framework on which to build mechanismdescriptions.
The significance of the zones in Fig. 1 is qualitative only. Much conjectureand few reliable data exist on how the small-scale processes of mass and heattransfer and the associated reaction regimes relate to the zones. For example,many studies, have assumed that the fuel for combustion inhigh-temperature-oxidation (HTO) reactions is formed primarily through thermalcracking reactions of crude oil constituents. Evidence exists, however, thatthermal cracking alone may not be sufficient to deposit fuel in a combustionpropagation and that LTO processes are also important in this process. Thereare also important unknowns concerning the combustion zone, such as therelationships of gas-phase oxygen content, hydrocarbon fuel concentration, andtemperature.
For this study, the combustion zone was the focus of the theoreticalinvestigation. Not a great deal is known about the small-scale characteristicsof the region that drives the in-situ combustion process beyond theexperimentally observed typical temperature range. The high-temperature zone isamenable to analysis given that the region is strongly dominated by thereaction regime of HTO and flow in the region involves gas only (single-phaseflow). Steady-state. combustion-front propagation data from combustion-tubetests, such as those performed in this study's experimental program, provideinformation about the combustion zone, including apparent fuel atomichydrogen/'carbon ratios, molar CO2/CO product gas ratios, and temperatureprofiles.
Combustion-Tube System. Combustion experiments were performed in a newlydeveloped system that allowed the use of both unconsolidated and consolidatedcore materials. The system consisted of a 1.1-m-long, 5-cm-diameter,adiabatic-type combustion tube with high-quality insulation and 14 guard heaterzones. The apparatus allowed elemental combustion simulations to be performedwith a thin-walled tube (to minimize wall-heating effects) under relevantprocess pressure conditions. Packing techniques developed allowed an overburdenpressure to be imposed on the outside of the combustion tube at a levelsubstantially higher than the internal tube pore pressure in the core.Stressing of the core ensured pack integrity (no shifting of matrix material)and process uniformity (minimizing the possibility of wag flow effects) duringexperiments, and allowed consolidated core sections to be used.
Experiments were performed on cores that were preflushed with brine and thensaturated with crude oil. All tests were performed with the combustion zoneadvancing vertically downward. Inert gas was injected to develop an initial gassaturation, and the combustion process was started by switching to airinjection after the injection region was preheated to a 400 degrees C ignitiontemperature. Complete details on development of the combustion-tube apparatusand the experimental methods are reported elsewhere.
Test Program. Table 1 lists the combustion-tube tests run in the study. Theprogram primarily examined the effects of changes in injected-air flux and corepack type (unconsolidated/consolidated). Standard Berea sandstone core materialwas used. Test A was an apparatus-commissioning test with a resaturatedAthabasca oil-sand core. Tests B and C used a heavy oil with unconsolidated andoriginal consolidated Berea material, respectively. Test D (Berea consolidated)and Test E (Berea unconsolidated) used Athabasca crude oil. Test E was splitinto two experiments, Tests E1 and E2, which used different injected-air fluxesto produce two stabilized combustion periods. All the tests, except for Test C,produced stabilized combustion process behavior.
The common characteristic of stabilized combustion tests is the observationof a steadily moving, self-sustaining process after an initial ignition period.This process is characterized by essentially constant combustion-front velocityand exit-gas composition for a given injected gas rate and composition. Fig. 2shows the typical temperature-front/tube-position relationship (for Test A)exhibited by stable processes and illustrates the essentially constant anduniform process-propagation velocity characteristic of this behavior.
Table 2 lists data from the tests that exhibited stabilized combustion. Thecombustion-zone velocity, apparent fuel hydrogen/carbon ratio, and CO2 and COconcentrations in the produced gas were used in the model described later. Thestabilized parameters are typical of those determined for corresponding crudeoils in the original combustion-tube apparatus located in the samelaboratory.
|File Size||667 KB||Number of Pages||8|