Introduction

• Lattice preferred orientation (LPO) in mantle peridotite
• • caused by dislocation creep with a dominant slip system
  • indication of dynamic flow in the mantle
  • causing seismic anisotropy
  • • What does the decrease in seismic anisotropy with depth indicate?
    • • transition in LPO pattern with pressure?
      • transition in LPO pattern with temperature?
      • decrease in strain rate with depth?
• LPO types of peridotite
• • A-Type: olivine a-axis parallel to lineation, b-axis normal to the foliation
  • • Slip in the [100] direction on the (010) plane: a-slip
  • B-Type: olivine c-axis parallel to lineation, b-axis normal to the foliation
  • • Slip in the [001] direction on the (010) plane: c-slip
  • AG-Type: a- and c-axes on the foliation, , b-axis normal to the foliation
  • • What kind of slip?

• What conditions are these olivine fabrics reflect?
• • If temperature, the mobility of dislocations produced by a- and c-slips has different temperature dependence.
• Dislocation recovery experiment
• • Dislocations are annihilated by coalescence of dislocations with opposite sign
  • Measurement of annihilation rate by comparing the initial and final dislocation density after annealing under quasi-hydrostatic conditions.
  • Dislocation annihilation rate ∝ dislocation mobility

Experimental procedure

• Production of dislocations in a- and c-slip systems
• • Deformation of olivine single crystals in a shear geometry of the [100] or [001] direction on the (010) plane
  • at 3 GPa and 1600 K

• Annealing for dislocation annihilation
• • Ambient pressure and temperatures of 1460 to 1760 K
  • 20 min to 24.5 h
  • fO2 near the Ni-NiO buffer
• Dislocation density measurement
• • oxidation decoration
  • SEM

Results

• Produced dislocations
  
• • a-slip
  • • • Edge dislocation
      • Burgers vector [100]
      • Elongation in [001]
    • a-dislocations
      
  • c-slip
  • • • Screw dislocation
      • Burgers vector [001]
      • Elongation in [001]
    • a-dislocations

• Annihilation rates
• • Absolute values of dislocation annihilation rate of a- and c-dislocations: comparable
  • • within 0.3 log scale
  • Temperature dependence: identical
  • • Activation energy
    • • a-dislocation: 400+-20 kJ/mol
      • c-dislocation: 400+-30 kJ/mol

Discussion

• Mechanism of dislocation motion
• • Smaller activation energy than obtained by deformation experiments (500 kJ/mol)
  • • different mechanism of dislocation motion between deformation and recovery experiments
  • Identical activation energy to that of Si diffusion (410+-30 kJ/mol)
  • • Dislocation motion: diffusion-controlled
    • • climb control
  • Comparable annihilation rate and identical activation energy between a- and c-dislocations
  • • very small geometry factor
    • independence of dislocation character
    • • edge or screw
    • supporting diffusion-control mechanism
• Fabric transition of the mantle olivine
• • Temperature conditions causes no fabric transition between A-type and B-type
  • • identical temperature dependence of the responsible dislocations
  • Fabric transition by strain rate
  • • Deformation experiment:
    • • very high creep rate
      • glide control
      • anisotropy to produce A- or  B-type fabrics
      • No AG-type fabrics
    • Recovery experiment:
    • • corresponding to low strain rates
      • climb control
      • less isotropic
      • • little difference between a- and c-slips
        • could produce AG-type fabric

• Possible fabric transition due to lowering strain rate with increasing depth
• • glide control -> climb control
  • A-type or B-type fabric -> AG-type fabric
  • AG-type fabric will be dominant in the oceanic asthenosphere

Wang, L., S. Blaha, Z. Pinter, R. Farla, T. Kawazoe, N. Miyajima, K. Michibayashi and T. Katsura, Temperature dependence of [100](010) and [001](010) dislocation mobility in natural olivine, Earth Planet. Sci. Lett. 441, 81-90, 2016.