Spin-orbit torques (SOTs) have great potential for enabling ultrafast energy-efficient memory and logic technologies. However, a successful SOT technology requires energy-efficient, high-endurance, and integration-friendly spin-current generators. From this point of view, Pt-based spin Hall metals are advantageous over the epitaxial complex oxides SrIrO3 or SrRuO3 (e.g. incompatibility with Si/SiO2 substrate and quick degradation in ambient condition) and topological insulators (giant resistivity, considerable roughness, and thermally instability). However, the dampinglike spin-orbit torque efficiency for pure Pt films is relatively low (ξDL =0.5-0.22) depending on the resistivity of the Pt. It is of great scientific and technological importance to advance the physics understanding about how, why, and to what limit ξDL would be enhanced. In this talk, I will report on an unambiguous determination of the dominant intrinsic spin Hall effect in Pt [1], how the intrinsic spin Hall conductivity of Pt is affected by carrier lifetime, strain, interruption of crystal order [2,3], and how and to what limit the spin Hall ratio and dampinglike torque efficiency of Pt-based heavy metals can be enhanced. Immediate benefits of these efforts include three optimal low-resistivity high-spin-Hall-ratio heavy metals with the power consumption of only 1% of that for the topological insulator Bi1-xSex (ξDL =18.6, ρxx = 13000 µΩ cm) based devices. We find negligible interfacial Rashba-Edelstein effect in our samples studied here. 

[1] L. Zhu et al., Sci Adv. 5, eaav8025 (2019).
[2] L. Zhu et al., Phys. Rev. Applied 11, 061004 (2019).
[3] L. Zhu and R.A. Buhrman, Phys. Rev. Applied 12, 051002 (2019).