Abstract
Predictions by the International Energy Agency (IEA, 2018) showed that as the world's energy system is being decarbonised, the demand of fossil fuels, especially from the petrochemicals and transportation sectors which do not have any alternative, will continue to grow for the next two decades (i.e. to 2040). As a result, fully understood designs and operations of advanced technologies, such as the toe-to-heel air injection (THAI) heavy oil and bitumen upgrading and production process, are needed to develop the vast reserves of the virtually untapped bitumen/tar sand/heavy oil deposits. A new in-situ combustion type processes scaling procedure, which involved decreasing the reaction frequency factors, as proposed and detailed in Rabiu Ado (2017), was used to carry out a comparative study between the predictions by the two different types of validated kinetics schemes using computer modelling group (CMG) thermal reservoir simulator, the STARS. It is found that both models P and G, though each having different kinetics schemes and parameters, provided very similar qualitative predictions of the oil saturation and temperature profiles. However, model G was found to predict an unphysically high fuel concentration (206-330 kg m(-3)) at relatively low temperature (i.e. 165 to 252 degrees C) at the base of the reservoir. This is not surprising since the field scale kinetics parameters for model G where not obtained in a systematic way, rather, they were obtained via trial and error. Additionally, model G's kinetics scheme is heavily dependent on the stoichiometric coefficient of the reaction's product. As a consequence, it is concluded that model P provided a more realistic prediction of the physicochemical processes, and hence, it should be used for future field studies.