A new in-situ combustion strategy, the top down process, is currently under detailed laboratory study. The process, aimed at overcoming some of the problems that have restricted the successful application of in-situ combustion in oil sand and heavy oil formations, involves the stable propagation of a combustion front from the top to the bottom of a reservoir, exploiting gravity drainage of the mobilized oil to a lower horizontal well.

Operational parameters that have been investigated and presented here include: air injection flux, degree of pre-heating, internal steam flood pre-heating and injection of normal air versus injection of oxygen enriched air.

To compliment the experimental investigation, the thermal numerical simulator STARS has been applied to the in-situ combustion process by incorporating reaction kinetics for Athabasca oil sand. A successful history match of an experimental test is presented accompanied by a discussion of application of the model to field scale.


In-situ combustion has long been recognized as having the potential for being an economical thermal oil recovery process in heavy oil and oil sand deposits. The energy required to supply heat to the reservoir compares quite favourably with steam. The estimated cost1 to place 1 GJ of energy in a 7 MPa reservoir is $2.6-$4.4 using steam and $ 1.0 for in-situ combustion using air (assuming $2/GJ fuel cost, capital cost not included). In-situ combustion is not compromised by large heat losses to overburden and underburden in thin formations or by high heat losses from the well bore to the overburden in deep formations as is the case with steam injection. Also in-situ combustion theoretically has important applications in reservoirs containing bottom water and as a follow up process to waterflooded and steamflooded formations.

Previous in-situ combustion field projects, however, have been less successful than steam, primarily because of the difficulty in controlling the combustion front advancement. The customary in-situ combustion operation of the past involved the injection of an oxygen containing gas into a central vertical injection well surrounded by a number of vertical production wells (typically as part of a larger pattern of injection and production wells). Combustion was initiated near the injection well and horizontally propagated radially outwards, aiming to drive the mobilized oil towards the production wells. The problem frequently encountered was that the combustion fronts tended to advance erratically with the vertical sweep constrained by gravity override of the displacing gas and the areal sweep reduced by preferential flow to one well of the pattern. Injected oxygen, overriding the combustion zone, created problems at the production end and the overriding hot steam and combustion gases did little to heat the formation ahead of the burn zone. The displacement geometry of the process requires that the mobilized oil be displaced ahead of the combustion front into the colder immobile oil, increasing oil saturation and further reducing mobility, with the limited producibility of the vertical production wells unable to alleviate the situation.

P. 487

This content is only available via PDF.
You can access this article if you purchase or spend a download.