2015 GSA Annual Meeting in Baltimore, Maryland, USA (1-4 November 2015)

Paper No. 326-10
Presentation Time: 4:00 PM

AN EXPERIMENTAL STUDY ON EVOLUTION OF FRACTURE PERMEABILITY OF ULTRAMAFIC ROCKS DURING SERPENTINIZATION REACTIONS


FAROUGH, Aida, Geosciences, Virginia Tech, Blacksburg, VA 24061, MOORE, Diane E., U. S. Geological Survey, 345 Middlefield Rd. MS 977, Menlo Park, CA 94025, LOCKNER, David A., U.S. Geological Survey, 345 Middlefield Road MS 977, Menlo Park, CA 94025 and LOWELL, Robert, Department of Geosciences, Virginia Tech, 4044 Derring Hall, Blacksburg, VA 24061-0420, afarough@vt.edu

Serpentinization reactions, during which olivine and pyroxene minerals are replaced by serpentine, magnetite, brucite and talc, affect chemical and physical properties of the host rock, such as permeability. To advance our understanding of the evolution of permeability accompanying serpentinization reactions, we performed a series of flow-through experiments (using de-ionized water) at a temperature of 260˚C and effective pressure of 30 MPa on cylindrical cores of ultramafic rocks containing a single through-going tensile fracture to estimate effective permeability. A 7.5 mm thick layer of the same rock, crushed and sieved (0.18-1.0 mm) was placed on the inlet end of the sample to produce a reactive heated reservoir for the pore fluid before entering the fracture. Multiple peridotite samples were tested, to investigate the effect of mineral assemblage on fluid-rock interaction and permeability. We also performed flow through experiments at different effective pressures (10-20-30 MPa) on intact sample to estimate matrix permeability and it's pressure dependency. Using both effective and matrix permeability we were able to estimate fracture permeability and it's evolution as serpentinization reactions progressed.

The initial effective permeability of the samples varied between 10-(15-18)m2, and it decreased by about 2 orders of magnitude in 7-10 days, showing that serpentinization reactions result in an initially rapid decrease in permeability. Assuming flow between parallel plates, we find the fracture permeability decreases by approximately 2 orders of magnitude during the experiments as a result of a combination of factors including mineral dissolution/precipitation and pressure solution of asperities. Fracture permeability is about 9 orders of magnitude higher than intact permeability. This suggests a preferential flow path and EMP analysis and SEM imaging show serpentine phases precipitated along the walls of the tensile fracture as the main flow path in the sample. Presence of Mg, Fe and Si (10-(3-5) molar) in the final pore fluid sample collected from all experiments is consistent with occurrence of serpentinization reactions.