Pore scale study of interfacial areas at drainage and imbibition in granular media Maša Prodanović 1 , Dorthe Wildenschild 2 , Elena Rodriguez Pin 1 , and Steven L. Bryant 1 1 Center for Petroleum and Geosystems Engineering The University of Texas at Austin 2 School of Chemical, Biological, and Environmental Engineering, Oregon State University American Geophysical Union Fall Meeting San Francisco, CA, Dec 14, 2009
25
Embed
Pore scale study of interfacial areas at drainage and imbibition in granular media Maša Prodanović 1, Dorthe Wildenschild 2, Elena Rodriguez Pin 1, and.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Pore scale study of interfacial areas at drainage and imbibition in granular media
Maša Prodanović1, Dorthe Wildenschild2, Elena Rodriguez Pin1, and Steven L. Bryant1
1Center for Petroleum and Geosystems EngineeringThe University of Texas at Austin
2School of Chemical, Biological, and Environmental Engineering, Oregon State University
American Geophysical Union Fall MeetingSan Francisco, CA, Dec 14, 2009
Support & computational resources
“US Department of Agriculture, grant "Quantifying the mechanisms of pathogen retention in unsaturated soils“ (MP, ER, SLB)
National Science Foundation (EAR 337711 and EAR 0610108) (DW)
Texas Advanced Computing Center (TACC)
Outline Motivation:
Evaluate role of fluid-fluid and fluid-solid interfaces interfacial areas and triple contact in pore scale displacements (hypothesized to be missing link in Pc-Sw relationships)
equation, (const. capillary pressure Pc and interfacial tension σ)
o Model slow displacement as a sequence of const. curvature interfaces
Imaged by D. WIldenschild
Progressive quasi-static algorithm (PQS) Drainage
Initialize with a planar front Solve evolution PDE with slightly
compressible curvature model for F until steady state:
Iterate increment curvature Find steady state of prescribed
curvature model
Imbibition starts from drainage endpoint and decrements curvature
Zero contact angle: wall BC
M. Prodanović and S. L. Bryant. A level set method for determining critical curvatures for drainage and imbibition. Journal of Colloid and Interface Science, 304 (2006) 442
Simple 2D example for LSMPQS
Haines
jump
drainage (controlled by throats) imbibition (controlled by pores)
Simulation steps (alternating red and green colors). All <= 2% rel.abs.err.
Melrose
criterionM. Prodanović and S. L. Bryant. A level set method for determining critical curvatures for drainage and imbibition. Journal of Colloid and Interface Science, 304 (2006) 442
Textbook example3D packing of equal spheres
Experimental vs. simulated system
Imaged volume
7mm (420 voxels)
LSMPQS simulated volume
200 voxels (3.4mm)
Drainage Pc-Sw comparison: Beads
Experimental setup cross-section
420x420
LSMPQS simulation cross-section
200x200
In all samples, we get LSMPQS curves higher than experimental because we picked inner, tighter subsample.
Simulations in larger samples on the way.
Drainage Pc-Sw: Small beads
Experimental setup cross-section
420x420
LSMPQS simulation cross-section
200x200
Drainage Pc-Sw: Tuff
Tuff grains are larger, boundary/size effects especially severe: this probably affects residual wetting phase.
LSMPQS initial drainage
Tuff is tough… Coarsened image simulation (larger volume)
currently going on We possibly have resolution effects: films
cannot be resolved
dx=2.8µm
LSMPQS SimulationEfficiency (M70) and interfacial area contribution (HG93)
Both theories comparable for the most part in range [0.6,0.8]
Morrow (‘70) Hassanizadeh & Gray ’93.
XCMT ExperimentsEfficiency (M70) and interfacial area contribution (HG93)
Contributions/Ed larger than Pc? Observed by Pyrak-Nolte et al., WRR, 2008 differences in Pc (local, global)? resolution?
Morrow (‘70) Hassanizadeh & Gray ’93.
I plot the change in interfacial area term as a fraction of capillary pressure so we can compare it directly to the efficiency (one is integral form of the other).
Triple contact line measurement
medial axis thinning used to extract length Lc from segmented images (simulation or experiment)
Dimensionless specific length
Initial study done on monodisperse packing
D
LcLc =
R2D
bD b
Lc Lc= R
V V bbD 3
VV =
R
Movie: CL advancement during drainage in a simple
pore
Resolution effects
dx=0.04 dx=0.08
Resolution effects If known that pendular rings not resolved,
double the computed result
Conclusions
Contribution of change interfacial areas to capillary pressure in Hassanizadeh & Gray ’93 computed for the first time in 3D Shown to be sizable in both simulation and
experiments In some experiments even larger
needs investigation Pyrak-Nolte et al. show similar in 2D
Preliminary contact line measurements Objective to determine role in colloid retention in