Pore‐Scale Experimental Investigation of Two‐Phase Flow Through Fractured Porous Media

Abstract

We present the results of a systematic pore-scale experimental investigation of two-phase oil/brine flow through a miniature water-wet, fractured sandstone core sample. X-ray microtomography is employed to generate three-dimensional fluid occupancy maps within a rough-walled fracture and its neighboring rock matrix during drainage and imbibition flow experiments. Several different imbibition flow conditions were created by changing brine flow rate, fracture aperture field, and interfacial tension between the fluids. These maps along with steady-state pressure drop data are then used to shed light on the dominant flow mechanisms and preferential flow paths through the matrix and fracture domains as well as fluid transfer between them during the imbibition processes. Depending on the fracture aperture properties and the magnitude of the local capillary pressures that are established under varying flow conditions, transport of the wetting phase across the hybrid matrix-fracture medium is governed by flow through wetting layers of brine on fracture walls (fracture layer flow), center of the fracture (fracture bulk flow), and brine-filled pores within the matrix. The hydraulic conductivity through these conduits is regulated by the medium as it identifies a combined flow path with minimum pressure drop from the inlet to the outlet of the system. The resulting balance determines the magnitude of fluid transfer experienced by the neighboring matrix and ultimate oil recovery as a result.

Energy Conversion Reveals Regime Transition of Imbibition in a Rough Fracture

“…• Movie S2 Such transition from capillary to capillary-viscous regimes occurs when the distance from injection wells decreases during water flooding in EOR (Arshadi et al, 2018;Rücker et al, 2015) and further identifies different pore-scale mechanisms of fluid-fluid interface advancements (Krevor et al, 2015). In capillary imbibition, studies on pore-scale events, including Haines jump (Haines, 1930), corner flow (Dong & Chatzis, 1995), and snap off (Roof, 1970), have shown that the pore-scale mechanisms not only impact the displacement patterns and efficiency (Chen, Guo, et al, 2018;Hu et al, 2018;Odier et al, 2017) but also contribute to the macroscopic hydraulic properties at the Darcy scale (Niessner et al, 2011).…”

Modeling Immiscible Two‐Phase Flow in Rough Fractures From Capillary to Viscous Fingering

“…Furthermore, assuming impermeable walls, the model does not consider fracture‐matrix interaction which may alter the displacement dynamics. Given that typical matrix pore sizes are significantly smaller than fracture apertures, fracture‐matrix interaction is mostly relevant for imbibition in fractured porous media (see, e.g., Arshadi et al, ; Kazmouz et al, ). In drainage configurations, the fluid‐fluid interfaces are very unlikely to enter the matrix.…”

“…For horizontal fractures, a number of experimental studies on two‐phase flow have been reported over the past two to three decades (e.g., Amundsen et al, ; Arshadi et al, ; C.‐Y. Chen & Horne, ; Y.‐F.…”

“…Chen et al, ; Ferer et al, ; Fourar et al, ; Huo & Benson, ; Hu, Wu, et al, ; Karpyn & Piri, ; Neuweiler et al, ; Persoff & Pruess, ; Pyrak‐Nolte et al, ; Reitsma & Kueper, ; Watanabe et al, ). Many of these studies have focused on pursuing suitable continuum/Darcy‐scale descriptions (i.e., capillary pressure and relative permeabililities as a function of saturation) using natural rock fractures (e.g., Arshadi et al, ; Huo & Benson, ; Reitsma & Kueper, ; Watanabe et al, ) or their transparent replicas (e.g., Nowamooz et al, ; Persoff & Pruess, ; Pyrak‐Nolte et al, ). The obtained capillary pressure and relative permeability relationships are particularly useful when one needs to predict two‐phase flow behavior at the field scale, where the fine details at the subfracture scale will be difficult to incorporate.…”

Benchmarking NAPL Redirection and Matrix Entry at Fracture Intersections Below the Water Table

“…In their lab‐scale study of excavated, naturally fractured clay, O'Hara et al () observed DNAPL within/nearby to a vertical intersection conduit, in addition to their general conclusion that invading non‐wetting phase flow occurs preferentially in larger‐aperture regions of fractures. Lab experiments involving wetting and non‐wetting fluids invading fractures and governing physical mechanisms are summarized in Arshadi et al (). Experiments targeting intersections of fractures specifically have been conducted by Dragila and Weisbrod () and Ji et al (, ).…”