Optimal Design Criteria for SAG Foam Processes in Heterogeneous Reservoirs

Abstract

Abstract Miscible gas injection can displace nearly all of the oil from the portions of the reservoir swept by gas. However, reservoir heterogeneity, low gas density and high gas mobility reduce sweep efficiency and drastically reduce recovery. Foam can reduce gas mobility and the effects of heterogeneity and gravity, and therefore increase sweep efficiency. Here we extend a previous study (SPE 110408) on optimal design strategy for surfactant-alternating-gas (SAG) foam processes in layered reservoirs. We develop design criteria for the optimal foam strength and slug size for a given permeability contrast between layers. We reach the following conclusions: Trends of foam strength with permeability or surfactant formulation, as measured in conventional coreflood tests at fixed injected water fraction, may not correspond to behavior in a SAG process in the field. In the cases examined, foam strength in a SAG process is much less sensitive to permeability and foam parameters than is foam strength at fixed injected water fraction. Placement of surfactant into low-permeability layers is a key challenge of SAG processes in heterogeneous reservoirs. Gas breakthrough occurs via low-permeability layers that did not receive enough surfactant. For that reason, unlike gas injection without foam, having a low-permeability layer at the top of the formation is a disadvantage because it is difficult to place foam there. Using multiple surfactant and gas slugs allows foam to redistribute surfactant in later slugs. However, this strategy suffers from poor injectivity during liquid injection, which slows the process and promotes gravity segregation. Injection of both gas and surfactant slugs at the maximum allowed injection pressure, rather than at fixed rate, gives best results. Injecting gas from only the bottom of the well offers no significant advantages in the best case, where high-permeability layers lie at top and bottom of the reservoir, and performs significantly worse if low-permeability layers lie at the top (where lack of surfactant leads to override) and bottom (where low permeability restricts injectivity) of the reservoir. A surfactant slug sized for a homogeneous reservoir is too large for a heterogeneous reservoir, because little surfactant enters lower-permeability layers: much of the injected surfactant goes to waste. A surfactant slug sized to sweep high-permeability layers and a portion of low-permeability layers performs nearly as well as one sized to sweep the entire reservoir.