Steady-state relative permeability measurements in rough-walled fractures: The effects of wettability and aperture

Tight subsurface geosystems provide significant prospects for the recovery and storage of mass and energy. Because of the tightness of these systems, natural and engineered fractures are the main conveyance conduits transporting the fluids from the tight rock matrix to the wellbore or vice versa. During the fracturing process, most of the newly induced fractures are considered water-wet, however during injection/production, the wettability of the fractures gradually changes. These changes are mainly due to the interactions between the resident and injected fluids, or the exposure of the virgin fracture surfaces to the resident fluids during production. This alteration in wettability can affect the trapping and flow dynamics, e.g., relative permeability, of the fluids flowing in the fracture. Furthermore, fluid injection and production into and from these systems impact the fractures pore pressure and hence the effective stress they are exposed to, resulting in variations in the fracture aperture. In this study, we present a comprehensive macro-scale experimental investigation focusing on the impact of wettability and aperture on steady-state two-phase relative permeability in rough-walled fractures. Using the stead-state measurement technique, we meticulously characterized oil-brine relative permeabilities and residual trapping in water-wet, oil-wet, and mixed-wet rough-walled fractures induced in Eagle Ford shale rock samples. Additionally, we assessed the influence of fracture aperture size on these properties during both drainage and imbibition flow processes. Our results indicate significant phase interference in oil-brine flow in non-water-wet rough-walled fractures, which renders the commonly used x-curve and Corey models inadequate to represent the steady-state oil-brine relative permeabilities measured in our study, leading to substantial overestimations. Mixed-wet fractures exhibited reduced relative permeabilities compared to oil-wet fractures, which in turn demonstrated lower relative permeabilities than water-wet counterparts. The observed trends stem from evolving fluid configuration and transport dynamics within the fractures as surface wettability transitions. Furthermore, an increase in fracture aperture corresponded with an increase in relative permeability. This was attributed to a broader flow area, diminished capillarity, attenuated impact of fracture wall roughness and tortuosity effects, and a reduction in bypassing and trapping events, all synergistically contributing to the mitigation of phase interference. Similarly, an increase in fracture aperture led to a marked decrease in postwaterflooding endpoint oil saturation and post-oil flooding endpoint brine saturation. Finally, we provided generalized correlation models based on experimental results from fractures with varying conductivities and wettability, yielding a more accurate representation of fracture relative permeabilities compared to traditional models. The data and correlation models produced in this study improve our understanding and prediction of multiphase flow in subsurface fractures and offer insights into rapid production decline in unconventional shale reservoirs.