A General Theory of Gas Production in SAGD Operations

Abstract

Abstract The small gas/oil ratio (GOR) commonly measured in SAGD projects has not previously been adequately explained, and various phenomena such as "microfingering" have been proposed to account for its presence. It is shown that the production of gases can be entirely explained by gases dissolved in the produced fluids at the temperature and pressure conditions of the SAGD steam chamber. Although methane is produced, in part, via bitumen, there is a significant contribution from methane dissolved in water as well. Other gases, such as carbon dioxide and hydrogen sulphide, are primarily produced by virtue of their solubility in water at the pertaining temperature and pressure. This result is a consequence of the asymptotic Henry's Law behaviour of gases in water as the critical point of water is approached. This asymptotic behaviour is shown to govern at temperatures well below the critical point, and within the temperature range of SAGD steam zones. The theoretical foundation of this work permits the estimation of gas-water equilibrium constants for the major produced gases of importance in SAGD, and thus an ultimate understanding of gas effects in the steam zone. Introduction The production of hydrogen sulphide and carbon dioxide, together with other minor gases in thermal recovery processes such as Steam Assisted Gravity Drainage (SAGD), is a common observation. The process that gives rise to these gases is a high temperature hydrolysis of aliphatic sulphur linkages in the bitumen, dubbed "aquathermolysis" by Hyne et al.(1–3) Typically, the amount of hydrogen sulphide produced per tonne of bitumen varies between six and 75 l. Considerably more carbon dioxide is produced, usually in the range 900 - 10,000 l per tonne. However, Hyne and co-workers have not dealt with the question of whether steam or steam condensate affects these reactions, and their experiments permit no conclusions in this regard. In some of our own experiments, the procedure of Hyne et al. was changed in that a stainless steel reaction vessel was used. In order to avoid loss of hydrogen sulphide to the steel, it was necessary to suppress the hydrogen sulphide in the gas phase. The experiment was arranged such that the reaction vessel was completely filled to eliminate a headspace, and an overpressure of 50,000 kPa was applied by means of helium gas. Thus, while Hyne's experiments at 240 °CDATA [C were conducted at the steam saturation saturation pressure of approximately 3,500 kPa, in our case, a gas phase was effectively prevented from forming. Our results were within the range reported by Hyne, suggesting strongly that the steam condensate, rather than the steam itself, is the reagent in the aquathermolysis. The question thus arises about the location within the steam zone where this reaction occurs. Elementary considerations of chemical kinetics would suggest that the steam front and the fluid drainage zone are the only regions of the SAGD steam chamber where the reaction is possible, these being the only regions where both steam condensate and bitumen are present in high saturations.