Grain boundary sliding in San Carlos olivine: Flow law parameters and crystallographic-preferred orientation

Research output: Contribution to journalArticlepeer-review

191 Scopus citations


We performed triaxial compressive creep experiments on aggregates of San Carlos olivine to develop a flow law and to examine microstructural development in the dislocation-accommodated grain boundary sliding regime (GBS). Each experiment included load and temperature steps to determine both the stress exponent and the activation energy. Grain boundary maps, created with electron backscatter diffraction data, were used to quantify grain size distributions for each sample. Inversion of the resulting data produced the following flow law for GBS: $\dot{\varepsilon}$GBS = 104.8 0.8 (2.9 0.3/d0.7 0.1) exp[(-445 20 kJ mol-1)/RT], with , d, and $\dot{\varepsilon}$GBS in units of MPa, m, and s-1, respectively. Although relatively weak, crystallographic-preferred orientations (CPOs) have [010] maxima parallel to the compression direction along with [100] and [001] girdles perpendicular to the compression direction. CPOs and subgrain boundary misorientation axes suggest that the (010)[100] slip system contributes significantly to deformation. We propose that these experimental results are best modeled by a deformation mechanism in which strain is accomplished primarily through grain boundary sliding accommodated by the motion of dislocations. Extrapolation of our flow laws to mantle conditions suggests that GBS is likely to be the dominant deformation mechanism in both lithospheric shear zones and asthenospheric flow, and therefore strong upper mantle seismic anisotropy can not be attributed solely to the dominance of dislocation creep.

Original languageEnglish (US)
Article numberB08201
JournalJournal of Geophysical Research: Solid Earth
Issue number8
StatePublished - Aug 1 2011


Dive into the research topics of 'Grain boundary sliding in San Carlos olivine: Flow law parameters and crystallographic-preferred orientation'. Together they form a unique fingerprint.

Cite this