HR: 09:45h
AN: S21C-08    [PDF]
TI: Relating Pore Fabric Geometry to Acoustic Wave Velocity, Anisotropy and Fluid Flow in Porous Sandstone: A Laboratory Study Using Magnetic Ferrofluid.
AU: * Benson, P
EM: p.benson@ucl.ac.uk
AF: Department of Earth Sciences, University College London Gower Street, London, WC1E 6BT United Kingdom
AU: Meredith, P
EM: p.meredith@ucl.ac.uk
AF: Department of Earth Sciences, University College London Gower Street, London, WC1E 6BT United Kingdom
AU: Platzman, E
EM: e.platzman@ucl.ac.uk
AF: Department of Earth Sciences, University College London Gower Street, London, WC1E 6BT United Kingdom
AU: White, R
EM: roy.white@ucl.ac.uk
AF: School of Earth Sciences, Birkbeck College Malet Street, London, WC1E 7HX United Kingdom
AB: Pore fabric geometry is a key feature of sedimentary rocks. Anisotropy arising from pore fabric has been commonly studied in terms of acoustic-wave (seismic) anisotropy, fluid flow (permeability) anisotropy and magnetic anisotropy (Anisotropy of Magnetic Susceptibility - AMS). However, combined approaches are relatively few, and often concentrate on grain fabric AMS. Here, we present results from an experimental study in which the AMS technique is used to determine the average 3D void space geometry in porous rock saturated with a high susceptibility magnetic ferrofluid. Using this approach, we independently show that the acoustic wave anisotropy and permeability anisotropy are well described by knowledge of the pore fabric anisotropy. We also demonstrate that pressure produces marked changes in both permeability and acoustic wave velocity, and that pore fabric is a useful tool with which to explain such changes. Measurements were made on Crab Orchard sandstone (COS) and Bentheim sandstone (BHS), chosen specifically for their contrasting strong (COS) and weak (BHS) anisotropy. COS is fine-grained and exhibits layering on a mm scale. It has a high cement content, resulting in a porosity of 4.5%. By contrast, BHS comprises 95% quartz grains in an open structure, resulting in a porosity of 22%. AMS was determined by measuring the susceptibility of ferrofluid saturated samples in 15 different orientations. A least squares ellipsoidal fit was then applied to this data to calculate the principal directions. Comparative elastic wave velocity measurements were then made in 10 degree increments around the circumferences of sets of three orthogonal cores. An equivalent 3D velocity ellipsoid was then determined, allowing for direct comparison of the velocity and AMS data. The error in using an ellipsoidal fit, rather than a fourth rank tensor, is estimated as less than 1.5%; approximately equal to the error in velocity measurement. Finally, measurements of permeability anisotropy, together with simultaneous ultrasonic velocity measurements, were made in a servo-controlled permeameter at effective pressures from 5 to 90MPa. In general, the permeability of COS parallel to bedding is some 3 times higher than that normal to bedding; whereas the permeability of BHS exhibits no discernible anisotropy. We find a strong positive correlation between the principal directions given by pore space AMS, velocity anisotropy, and permeability anisotropy. P-wave velocity anisotropy on dry samples was 19% and 5% for COS and BHS, respectively. This compares with a pore fabric anisotropy of 3.8% (COS) and 1.4% (BHS). The permeability of COS decreases from 75 to 8 mDarcy as effective pressure is increased from 5 to 90 MPa, a corresponding increase in acoustic wave velocity is also observed. BHS has a considerably higher permeability (830 mDarcy), but this changes little as pressure is increased. Our results clearly demonstrate that the overall anisotropy in these sedimentary rocks is dominated by the average pore fabric shape and orientation. Hence, the analysis of this fabric provides a good indicator of the anisotropy of other related physical properties, such as mechanical strength.
DE: 1518 Magnetic fabrics and anisotropy
DE: 5104 Fracture and flow
DE: 5112 Microstructure
DE: 5114 Permeability and porosity
DE: 7299 General or miscellaneous
SC: Seismology [S]
MN: 2003 Fall Meeting