Imaging and Measurement of Pore‐Scale Interfacial Curvature to Determine Capillary Pressure Simultaneously With Relative Permeability

Abstract

There are a number of challenges associated with the determination of relative permeability and capillary pressure. It is difficult to measure both parameters simultaneously on the same sample using conventional methods. Instead, separate measurements are made on different samples, usually with different flooding protocols. Hence, it is not certain that the pore structure and displacement processes used to determine relative permeability are the same as those when capillary pressure was measured. Moreover, at present, we do not use pore‐scale information from high‐resolution imaging to inform multiphase flow properties directly. We introduce a method using pore‐scale imaging to determine capillary pressure from local interfacial curvature. This, in combination with pressure drop measurements, allows both relative permeabilities and capillary pressure to be determined during steady state coinjection of two phases through the core. A steady state waterflood experiment was performed in a Bentheimer sandstone, where decalin and brine were simultaneously injected through the core at increasing brine fractional flows from 0 to 1. The local saturation and the curvature of the oil‐brine interface were determined. Using the Young‐Laplace law, the curvature was related to a local capillary pressure. There was a detectable gradient in both saturation and capillary pressure along the flow direction. The relative permeability was determined from the experimentally measured pressure drop and average saturation obtained by imaging. An analytical correction to the brine relative permeability could be made using the capillary pressure gradient. The results for both relative permeability and capillary pressure are consistent with previous literature measurements on larger samples.

Pore Network Model Predictions of Darcy‐Scale Multiphase Flow Heterogeneity Validated by Experiments

“…As demonstrated by the sensitivity analysis ( Figure S2 in the SI), pore network model descriptions of capillary pressure are much less dependent on image processing uncertainty than relative permeability. This is also significant because sub-core scale estimates of capillary pressure characteristics can be validated by ganglia-curvature based measurements (Herring et al, 2017;Garing et al, 2017;Lin et al, 2018Lin et al, , 2018Jackson et al, 2019), whereas estimates of relative permeability across similar size domains cannot.…”

“…Different acquisition settings were not tested as these will be highly dependent on different micro-CT scanner hardware. While contact angle and interfacial tension are very important parameters in the network model simulation, these properties were well constrained from previous experimental studies with similar rock-fluid pairs (Lin et al, 2018;Jackson et al, 2019). For other rock-fluid pairs, there are extensive contact angle and interfacial tension datasets available in literature (e.g.…”

“…Micro-CT has been a valuable tool for experimental characterization of pore space geometry, (Lin et al, 2016) mineralogy (Lai et al, 2015;Menke et al, 2015;Beckingham et al, 2017;Al-Khulaifi et al, 2019), wettability (Iglauer et al, 2012;Bartels et al, 2019;Lin et al, 2019), residual trapping (Herring et al, 2013;Chaudhary et al, 2013;Al-Menhali & Krevor, 2016;Øren et al, 2019), and curvature-based capillary pressure (Armstrong et al, 2012;Herring et al, 2017;Garing et al, 2017;Lin et al, 2018;T. Li et al, 2018).…”

“…Increased availability of affordable high-power computational and data management resources have enabled micro-CT imaging to increasingly be used to image centimeter-scale samples with voxel resolutions less than 10 micrometers (Lin et al, 2018;Jackson et al, 2019;Øren et al, 2019).…”

Relative Permeability Scaling From Pore‐Scale Flow Regimes

“…Multiphase Darcy's law has been rigorously formulated by upscaling the Stoke's equation at the pore scale (e.g., Auriault, 1987;Daly & Roose, 2015;Hassanizadeh & Gray, 1980;Hornung, 1997;Lasseux et al, 2008;Kjelstrup et al, 2018;Whitaker, 1986), and since the microscopic details of the flow are usually neglected, its classical interpretation has been recently questioned. In fact, although the inherent instability of multiphase flows (Ling et al, 2017) may render the analysis of pore-scale distribution of the flowing phases challenging, recurring features in the topology of the nonwetting and wetting phase at the pore scale have been revealed by X-ray microtomography and high-performance computing (Armstrong et al, 2014(Armstrong et al, , 2012Blunt et al, 2013;Berg et al, 2013;Cueto-Felgueroso & Juanes, 2012;Gao et al, 2017;Garing et al, 2017;Li et al, 2005Li et al, , 2018bLin et al, 2018;Prodanovic et al, 2006Prodanovic et al, , 2007Reynolds et al, 2017;Tallakstad et al, 2009;Tahmasebi et al, 2017;Zarikos et al, 2018). Experiments and simulations show that the nonwetting phase can become disconnected flowing in the form of individual ganglia or even remaining immobilized (trapped) in the porous matrix.…”