Evolution of crystallographic preferred orientations of ice sheared to high strains by equal-channel angular pressing

Abstract. Plastic deformation of polycrystalline ice 1 h induces crystallographic preferred orientations (CPOs), which give rise to anisotropy in the viscosity of ice, thereby exerting a strong influence on the flow of glaciers and ice sheets. The development of CPOs is governed by two pivotal mechanisms: recrystallization dominated by subgrain/lattice rotation and by strain-induced grain boundary migration (GBM). To examine the impact of strain on the transition of the dominant mechanism, synthetic ice (doped with ∼1 vol.% graphite) was deformed using equal-channel angular pressing technique, enabling multiple passes to accumulate substantial shear strains. Nominal shear strains up to 6.2, equivalent to a nominal von Mises strain of ε′ ≈ 3.6, were achieved in samples at a temperature of −5 °C. Cryo-electron backscatter diffraction analysis reveals a primary cluster of crystal c axes perpendicular to the shear plane in all samples, accompanied by a secondary cluster of c axes at an oblique angle to the primary cluster antithetic to the shear direction. With increasing strain, the primary c-axis cluster strengthens, while the secondary cluster weakens. The angle between the clusters remains within the range of 45° to 60°. The c-axis clusters are elongated perpendicular to the shear direction, with this elongation intensifying as strain increases. Subsequent annealing of the highest-strain sample reveals the same CPO patterns as observed prior to annealing, albeit slightly weaker. A synthesis of various experimental data suggest that the CPO pattern, including the orientation of the secondary cluster, results from a balance of two competing mechanisms: lattice rotation due to dislocation slip, which fortifies the primary cluster while rotating and weakening the secondary one, and grain growth by strain-induced GBM, which reinforces both clusters while rotating the secondary cluster in the opposite direction. As strain increases, GBM contributes progressively less. This investigation supports the previous hypothesis that a single cluster of c axes could be generated in high-strain experiments, while further refining our comprehension of CPO development in ice.