Transient creep of olivine: Point-defect relaxation times

Publisher Website
Google Scholar
Abstract
The time taken for San Carlos olivine specimens to reach a new steady-state creep rate after a change in either po2 or oxide activity has been used to calculate a diffusivity for the defect species which rate-limits the re-equilibration process. For temperatures between 1200 and 1400°C, the calculated diffusivities are approximately equal to reported values for the chemical diffusion coefficient for metal vacancies (Nakamura and Schmalzried 1984). These experimental results are consistent with a model in which dislocation glide or cross-slip rate-limits the high-temperature deformation of olivine, since changes in creep rate need only involve the motion of majority defects (metal vacancies and electron holes). However, if dislocation climb rate-limits the high-temperature deformation of olivine, and if local equilibtium is attained on the experimental timescale, these experimental results demonstrate that the minority defects that control diffusion on the silicon and oxygen sublattices re-equilibrate at least as fast as the majority defects. While one of these minority defects can re-equilibrate internally by producing or destroying jog segments on dislocation lines, the other can re-equilibrate in one of two ways. First, this second minority defect can diffuse from the sample surface to the interior; in this case, an upper limit can be placed on the concentration of this defect. Second, if exchange of trivalent iron between the silicon and metal sublattices occurs, both of the minority defects can re-equilibrate internally; in this case, only the majority defects are required to diffuse from the sample surface to the interior.