Low cycle fatigue behavior of magnesium matrix nanocomposite at ambient and elevated temperatures

AZ31B alloy reinforced by 1.5 vol.% nano-sized Al2O3 particles has been subjected to fully-reversed strain-controlled uniaxial tension-compression cyclic loading at room temperature, 100 °C, and 200 °C. A combination of mechanical and magnetic stir casting methods followed by a hot-extrusion process was used to fabricate the composite material. Cyclic and fatigue behaviors of the composite were studied at the strain ranges of 0.8%–2.5%. The experimental results of the fatigue lives were used to assess and compare the life prediction capabilities of different existing strain-based and energy-based fatigue models. The results exhibited that the presence of the nano-sized reinforcing particles leads to a homogenously fine microstructure and slightly changes the texture of the composite extrusion, compared to the typical microstructure and texture of monolithic AZ31B extrusion. The cyclic behavior of the composite is altered from an asymmetric shape at room temperature to a symmetric one at 200 °C, resulting in reduction of mean stress. Unlike the room-temperature behavior, twinning and detwinning are not governing plastic deformation mechanisms at the elevated temperatures, especially at 200 °C. Cyclic softening occurs by increasing the temperature, in a manner similar to the monotonic tensile tests. Due to the strain-controlled loading and increasing the composite ductility at the elevated temperatures, the fatigue lives are comparable at the different temperatures. Finally, considering the results at all the temperatures, Jahed-Varvani (JV) as an energy-based model shows a more promising life prediction. •A homogenous fine microstructure with a weakened basal texture was observed.•Composite exhibits asymmetric and symmetric loops at room temperature and 200 °C.•Cyclic softening happens and mean stress decreases at the elevated temperatures.•Composite shows an improved fatigue life at the higher temperatures.•Jahed-Varvani model is able to offer a more desirable prediction of fatigue life.