Coreflood Study of Non-Monotonic Fractional-Flow Behavior with Foam: Implications for Surfactant-Alternating-Gas Foam EOR

Foam is able to increase gas’s sweep efficiency in Enhanced-Oil-Recovery applications. A surfactant-alternating-gas, or SAG, process is usually preferred for placing foam in the reservoir. During a SAG process, foam is generated away from the wellbore, offering both good injectivity and good mobility control at the leading edge of the foam bank. Scale-up of laboratory data for SAG to field applications remains a challenge. Direct scale-up of dynamic SAG coreflood results is unreliable because of the dominance of core-scale artifacts. Steady-state coreflood data can be scaled up using fractional-flow theory (Kibodeaux and Rossen, 1997; Rossen and Boeije, 2015). However, about half the published laboratory studies of foam fractional-flow curves report non-monotonic behavior, where at some point liquid saturation Sw increases with decreasing liquid fractional flow fw. Rossen and Bruining (2007) warn that such behavior would result in foam collapse during injection of the gas slug in a SAG process at the field scale. Here we report and analyse a series of steady-state and dynamic coreflood experiments to investigate the occurrence of non-monotonic fractional-flow behavior. These corefloods vary surfactant concentration, injected gas fraction (foam quality) and total superficial velocity and are supported by CT measurements. The CT data confirm that in these cases, as foam weakens with decreasing fw, liquid saturation increases, confirming the non-monotonic fw(Sw) behaviour. In our results, every case of non-monotonic fractional-flow behavior begins with propagation of foam from the inlet, followed by eruption of a much-stronger foam at the outlet of the core and backwards propagation of the stronger foam state to the inlet, similar to behavior reported by Apaydin and Kovscek (2001) and Simjoo et al. (2013). This suggests that there may be more than one stable local-equilibrium (LE) foam state. The initial creation of the stronger foam near the outlet is at least in part due to the capillary end effect. It is thus not clear which LE foam state controls behaviour in a SAG process in the field. In our results, the subsequent transition from a stronger- to a weaker-foam state, leading to non-monotonic fw(Sw) behavior, coincides with conditions for weaker foam (lower surfactant concentration, lower fw) and less-vigorous foam generation (lower superficial velocity); this agrees with the theory of foam propagation of Ashoori et al. (2012). We discuss the implications of these findings, if confirmed to apply generally, for design of SAG foam processes.