Open Access
Stress conditions for the propagation of discrete compaction bands in porous sandstone
Author(s) -
Tembe Sheryl,
Baud Patrick,
Wong Tengfong
Publication year - 2008
Publication title -
journal of geophysical research: solid earth
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.67
H-Index - 298
eISSN - 2156-2202
pISSN - 0148-0227
DOI - 10.1029/2007jb005439
Subject(s) - compaction , geology , scaling , porosity , cataclastic rock , deformation bands , brittleness , geotechnical engineering , quartz , rock mechanics , mineralogy , mechanics , materials science , geometry , composite material , fault (geology) , seismology , physics , paleontology , microstructure , mathematics
Compaction bands are a compactant failure mode in porous rock, forming thin tabular structures normal to the maximum compressive stress with negligible shear offset. We investigated the conditions involved in the development of compaction bands in sandstone, including the influence of composition and the geometric attributes of the bands across a range of length scales. To extend beyond existing laboratory data on the relatively pure quartz Bentheim sandstone, a suite of triaxial experiments were conducted on Diemelstadt and Bleurswiller arkoses. Mechanical data and microstructural observations demonstrate that compaction bands can develop in compositionally heterogeneous rock and are the dominant failure mode in the transitional regime from brittle faulting to cataclastic flow. Synthesis of field and laboratory data on band dimensions in five sandstones over four orders of magnitude revealed a quadratic scaling relation between the thickness and length of compaction bands, wherein thickness is proportional to the square root of the band length. Using an anticrack/antidislocation fracture mechanics model, we obtained a scaling relation in which the stress level is inversely proportional to band thickness. We show that this relation provides a mechanical basis for interpreting discrepancies between laboratory and field data. Together, the laboratory and field data constrain the critical strain energy release rate in the model to be on the order of 2–80 kJ/m 2 , comparable with laboratory measurements.