Observations and Design Considerations for In Situ Combustion Projects

Abstract

Abstract An important part of any engineer's tool box is a set of simple equations thatwill allow one to quickly assess the performance potential of a given project. For many years, the model of Nelson and McNeil, as well as that of Gates and Ramey, have remained the standards for the preliminary design of in situcombustion projects. However, advances have been made in our understanding ofthe fundamental mechanisms and reaction kinetics of the combustion process, both for heavy and light oils. Combining established procedures with animproved interpretation of the mechanisms forms the basis for this paper. Theinclusion of the negative temperature gradient region, which affects thetransition to the high-temperature combustion mode, is one of the key elementsof these advances. The paper also addresses the sizing of air injection capacity and itsimportance, as well as the monitoring and analysis of gas-phase combustionproducts. In summary, the paper will provide the engineer with some realisticguidelines for estimating the oil recovery performance of an in situ combustionproject. Background In situ combustion is an enhanced recovery process with tremendous theoreticalpotential, but it has failed to perform up to expectations in many of its fieldimplementations. Two decades of research work at the University of Calgary onin situ combustion has focussed on the question "What makes in situ combustionfail?" While many mechanisms have been identified, the most critical withregard to heavy oil is the temperature of the oxidation (combustion)reactions. Figures 1 and 2 illustrate the importance of the oxidation zone temperature onthe residual oil and oil production for Athabasca bitumen (8 ° API). Figure 1shows high residual hydrocarbon concentrations for temperatures of less than300 ° C, while Figure 2 shows that no tests exhibited significant levels of oilrecovery for reaction temperatures of less than 350 ° C. These data, which werepreviously reported by Moore et al.(1), were determined by heatingpre-mixed Athabasca Oil Sands cores in a one dimensional core holder at a rateof 40 ° C/hr. while flowing an oxygen containing gas. Maximum oil recoveriesfor these ramped temperature oxidation tests were significantly lower thanthose for combustion tube tests due to the oxidation of the oil during theperiod of heating when temperatures were less than 300 ° C. However, thequalitative behaviour was similar in both types of tests. What these ramped temperature oxidation data show is that the mobilizationefficiency of air is low for oxidation temperatures of less than 300 ° C. However, reaction temperatures in excess of 350 ° C, while necessary, are not asufficient condition to mobilize oil from the region swept by the oxidationzone. It was found that elevated levels of oil recovery were associated with arapid transition of the leading edge temperatures to a value in excess of 350 °C.