Spin–orbit torque (SOT) switching of a nanomagnet whose in-plane easy axis is collinear with the charge current (the 'type-x' geometry) was demonstrated by Fukami, Anekawa, Zhang and Ohno in a three-terminal Ta/CoFeB/MgO device, and its comparison with the established perpendicular (type-z) and transverse-in-plane (type-y) geometries exposed a set of unresolved mechanism questions: the measured ratio of zero-field critical current densities between type-x and type-y devices is about 4.3, far below the value of 12.6 predicted by macrospin dynamics containing only the Slonczewski-like (damping-like) torque, which was attributed to an unmeasured field-like torque component; the effective spin Hall angle extracted from switching thresholds differs strongly between geometries (about −0.22 in type-x, −0.25 in type-z, −0.08 in type-y); the partitioning of the total torque between the bulk spin Hall effect of the heavy metal and interfacial or nonlocal mechanisms is unsettled; the pulse-width dependence of the critical current that should discriminate SOT-dominated from spin-transfer-torque-like switching was unmapped in the collinear regime; and deterministic type-x switching required an external symmetry-breaking field. This problem asks for the quantitative microscopic account of these torques in the collinear geometry. Later literature has largely mapped the pulse-width dependence and has demonstrated field-free type-x switching by easy-axis canting, but the central mechanism core — the magnitude and origin of the field-like component, the bulk-versus-interfacial partition, and the reconciliation of method- and geometry-dependent spin Hall angle extractions for these stacks — remains open, with domain-mediated (non-coherent) reversal competing with the field-like-torque explanation of the reduced thresholds.

Background

A current flowing in a heavy metal with strong spin–orbit coupling generates a transverse spin current, an effect known as the spin Hall effect; when this spin current is absorbed by an adjacent ferromagnetic layer it exerts a torque on the magnetization, called a spin–orbit torque (SOT). In a three-terminal magnetic tunnel junction (MTJ — a nanopillar of two ferromagnets separated by an MgO tunnel barrier whose resistance depends on their relative magnetization), a current in a Ta channel beneath the free layer can switch the free-layer magnetization without sending current through the tunnel barrier. Two torque components are conventionally distinguished: a damping-like (Slonczewski-like) torque that behaves like a spin-transfer torque (STT — the torque exerted when a spin-polarized current crosses a ferromagnet) with spin polarization fixed by the heavy-metal interface, and a field-like torque that acts like an effective magnetic field. SOT switching was first demonstrated for perpendicular magnetization (Miron et al., Nature 2011; Liu et al., Science 2012), and the symmetry and magnitude of both torque components were measured by magnetotransport methods such as harmonic analysis of the anomalous Hall response (the anomalous Hall effect is the transverse voltage of a ferromagnet, which tracks its magnetization; the harmonics it shows under an alternating current reveal the current-induced effective fields) (Garello et al., Nature Nanotechnology 2013; Kim et al., Nature Materials 2013, for Ta|CoFeB|MgO). The torque efficiency is commonly reported as an effective spin Hall angle, θSHeff\theta_{SH}^{eff}, the fraction of charge current converted into spin current absorbed by the ferromagnet, extracted either from switching thresholds using macrospin modeling (macrospin = the assumption of spatially uniform magnetization dynamics) or from effective-field measurements. Before 2016 all devices fell into two geometries relative to the current along xx: the easy axis out of plane ('type-z', which requires an in-plane symmetry-breaking field for deterministic switching and often reverses by incoherent domain nucleation) or in-plane and orthogonal to the current ('type-y', whose dynamics resemble conventional spin-transfer-torque precessional switching). Fukami et al. (Nature Nanotechnology 2016) completed the line-up with the third geometry, 'type-x', in which the in-plane easy axis lies along the current, so the spin polarization of the damping-like torque is orthogonal to the easy axis. They demonstrated deterministic bipolar switching (with the switching direction set by the sign of a small out-of-plane field, μ0Hz=∓15\mu_0 H_z = \mp 15 mT) at a zero-field critical current density JC0≈4.3×1011J_C^0 \approx 4.3 \times 10^{11} A m−2^{-2}, and compared type-x, type-y and type-z devices on comparable Ta/CoFeB/MgO stacks. Three quantitative puzzles emerged from that comparison: (i) the measured ratio JC0(x)/JC0(y)≈4.3J_C^0(x)/J_C^0(y) \approx 4.3 is about a factor of three below the macrospin prediction of 12.6 that includes only the damping-like torque, which the authors attributed to an unmeasured field-like torque lowering the thresholds in type-x (and similarly type-z); (ii) θSHeff\theta_{SH}^{eff} extracted from switching thresholds differs markedly between geometries (about −0.22-0.22 for type-x, −0.25-0.25 for type-z, −0.08-0.08 for type-y), so the extracted torque efficiency is not a geometry-independent material parameter; and (iii) the microscopic origin of the total SOT in such stacks — bulk spin Hall effect in the heavy metal versus interfacial spin–orbit coupling or nonlocal effects — was, and remains, contested. In addition, the paper noted that the pulse-width dependence of the critical current, predicted to differ between SOT-dominated and spin-transfer-torque-like switching, was masked in type-z by incoherent domain processes, and that the need for an external symmetry-breaking field in type-x is a practical drawback whose structural removal was undemonstrated.

Problem Statement

Provide a quantitative, experimentally verified microscopic account of spin–orbit torque switching in the collinear easy-axis (type-x) geometry of heavy-metal/ferromagnet heterostructures, of the kind demonstrated in Ta/CoFeB/MgO three-terminal devices, so that the type-x platform functions as a definitive probe of SOT physics. Concretely: (1) determine the magnitude, microscopic origin (interfacial versus bulk), and stack dependence of the field-like torque component in these stacks, and establish whether it — rather than domain-mediated (non-coherent) reversal, thermal activation, or modeling error — explains why the measured zero-field critical current densities of type-x (and type-z) devices fall well below macrospin predictions that include only the Slonczewski-like (damping-like) torque, quantitatively reconciling the measured type-x/type-y threshold ratio of about 4.3 with the predicted 12.6; (2) quantitatively partition the total SOT between the bulk spin Hall effect of the heavy metal and interfacial or nonlocal mechanisms for these stacks; (3) consolidate the pulse-width dependence of the critical current density now mapped across the full pulse-width range in the collinear (in-plane) regime, where domain processes are not expected to mask the intrinsic dynamics (Ref. 1), and assess how it discriminates SOT-dominated switching from spin-transfer-torque-like switching; (4) assess the structural symmetry breaking (for example a canted easy axis) that has been demonstrated to remove the external symmetry-breaking magnetic field for deterministic type-x switching, and check that it is consistent with the mechanism account; and (5) reconcile the effective spin Hall angles extracted from switching thresholds in different geometries with those extracted from other observables such as effective-field measurements, so that the reported torque efficiency becomes a well-defined, geometry-independent material parameter. The quantitative account sought applies to the Ta/CoFeB/MgO-type stack class of the collinear (type-x) geometry as a class rather than to a single device, and it must be quantitatively consistent with that class's switching-threshold, effective-field and geometry-comparison data — including the pulse-width dependence of the critical current density now mapped across the collinear (in-plane) regime and the established field-free type-x switching by easy-axis engineering (Refs. 6, 8, 10, 16, 17, 20). Together these items serve one common objective: a closed, quantitative picture of which torque components, of which microscopic origin, set the switching thresholds, dynamics and field requirements of the collinear geometry.

References

  1. S. Fukami, T. Anekawa, C. Zhang, H. Ohno, 'A spin–orbit torque switching scheme with collinear magnetic easy axis and current configuration', Nature Nanotechnology 11, 621–625 (2016). DOI: 10.1038/nnano.2016.29, https://doi.org/10.1038/nnano.2016.29
  2. I. M. Miron, K. Garello, G. Gaudin, P.-J. Zermatten, M. V. Costache, S. Auffret, S. Bandiera, B. Rodmacq, A. Schuhl, P. Gambardella, 'Perpendicular switching of a single ferromagnetic layer induced by in-plane current injection', Nature 476, 189–193 (2011). DOI: 10.1038/nature10309, https://doi.org/10.1038/nature10309
  3. L. Liu, C.-F. Pai, Y. Li, H. W. Tseng, D. C. Ralph, R. A. Buhrman, 'Spin-torque switching with the giant spin Hall effect of tantalum', Science 336, 555–558 (2012). DOI: 10.1126/science.1218197, https://doi.org/10.1126/science.1218197
  4. K. Garello, I. M. Miron, C. O. Avci, F. Freimuth, Y. Mokrousov, S. Blügel, S. Auffret, O. Boulle, G. Gaudin, P. Gambardella, 'Symmetry and magnitude of spin–orbit torques in ferromagnetic heterostructures', Nature Nanotechnology 8, 587–593 (2013). DOI: 10.1038/nnano.2013.145, https://doi.org/10.1038/nnano.2013.145
  5. J. Kim, J. Sinha, M. Hayashi, M. Yamanouchi, S. Fukami, T. Suzuki, S. Mitani, H. Ohno, 'Layer thickness dependence of the current-induced effective field vector in Ta|CoFeB|MgO', Nature Materials 12, 240–245 (2013). DOI: 10.1038/nmat3522, https://doi.org/10.1038/nmat3522
  6. Y. Takahashi, Y. Takeuchi, C. Zhang, B. Jinnai, S. Fukami, H. Ohno, 'Spin–orbit torque-induced switching of in-plane magnetized elliptic nanodot arrays with various easy-axis directions measured by differential planar Hall resistance', Applied Physics Letters 114, 012410 (2019). DOI: 10.1063/1.5075542, https://doi.org/10.1063/1.5075542
  7. T. Taniguchi, 'Switching induced by spin Hall effect in an in-plane magnetized ferromagnet with the easy axis parallel to the current', Physical Review B 102, 104435 (2020). DOI: 10.1103/PhysRevB.102.104435, https://doi.org/10.1103/PhysRevB.102.104435
  8. E. Grimaldi, V. Krizakova, G. Sala, F. Yasin, S. Couet, G. S. Kar, K. Garello, P. Gambardella, 'Single-shot dynamics of spin–orbit torque and spin transfer torque switching in three-terminal magnetic tunnel junctions', Nature Nanotechnology 15, 111–117 (2020). DOI: 10.1038/s41565-019-0607-7, https://doi.org/10.1038/s41565-019-0607-7
  9. Y.-T. Liu, T.-Y. Chen, T.-H. Lo, T.-Y. Tsai, S.-Y. Yang, Y.-J. Chang, J.-H. Wei, C.-F. Pai, 'Determination of spin-orbit-torque efficiencies in heterostructures with in-plane magnetic anisotropy', Physical Review Applied 13, 044032 (2020). DOI: 10.1103/PhysRevApplied.13.044032, https://doi.org/10.1103/PhysRevApplied.13.044032
  10. Y. Shiokawa, E. Komura, Y. Ishitani, A. Tsumita, K. Suda, K. Hamanaka, T. Taniguchi, T. Sasaki, 'Dependency of high-speed write properties on external magnetic field in spin–orbit torque in-plane magnetoresistance devices', Applied Physics Express 14, 013001 (2021). DOI: 10.35848/1882-0786/abd598, https://doi.org/10.35848/1882-0786/abd598
  11. Q. Shao, P. Li, L. Liu, H. Yang, S. Fukami, A. Razavi, H. Wu, K. L. Wang, F. Freimuth, Y. Mokrousov, M. Stiles, S. Emori, A. Hoffmann, J. Åkerman, K. Roy, J.-P. Wang, S.-H. Yang, K. Garello, W. Zhang, 'Roadmap of spin-orbit torques', IEEE Transactions on Magnetics 57, 1–39 (2021). DOI: 10.1109/TMAG.2021.3078583, https://doi.org/10.1109/TMAG.2021.3078583
  12. S. Dutta, A. Bose, A. Tulapurkar, R. Buhrman, D. Ralph, 'Interfacial and bulk spin Hall contributions to fieldlike spin-orbit torque generated by iridium', Physical Review B 103, 184416 (2021). DOI: 10.1103/PhysRevB.103.184416, https://doi.org/10.1103/PhysRevB.103.184416
  13. H. Wu, J. Zhang, B. Cui, S. A. Razavi, X. Che, Q. Pan, D. Wu, G. Yu, X. Han, K. L. Wang, 'Field-free approaches for deterministic spin–orbit torque switching of the perpendicular magnet', Materials Futures 1, 022201 (2022). DOI: 10.1088/2752-5724/ac6577, https://doi.org/10.1088/2752-5724/ac6577
  14. C. Safranski, J. Z. Sun, A. D. Kent, 'A perspective on electrical generation of spin current for magnetic random access memories', Applied Physics Letters 120, 160502 (2022). DOI: 10.1063/5.0084551, https://doi.org/10.1063/5.0084551
  15. Y. Zhuo, W. Cai, D. Zhu, H. Zhang, A. Du, K. Cao, J. Yin, Y. Huang, K. Shi, W. Zhao, 'Mechanism of field-like torque in spin-orbit torque switching of perpendicular magnetic tunnel junction', Science China Physics, Mechanics & Astronomy 65, 107511 (2022). DOI: 10.1007/s11433-022-1975-4, https://doi.org/10.1007/s11433-022-1975-4
  16. Y.-T. Liu, Y.-H. Huang, C.-C. Huang, Y.-C. Li, C.-L. Cheng, C.-F. Pai, 'Field-Free Type-x Spin-Orbit-Torque Switching by Easy-Axis Engineering', Physical Review Applied 18, 034019 (2022). DOI: 10.1103/PhysRevApplied.18.034019, https://doi.org/10.1103/PhysRevApplied.18.034019
  17. H. Ye, Z. Yan, M. Wang, Z.-H. Wang, 'Critical current for field-free switching of the in-plane magnetization in the three-terminal magnetic tunnel junction', Science China Information Sciences 68, 209404 (2025). DOI: 10.1007/s11432-025-4455-2, https://doi.org/10.1007/s11432-025-4455-2
  18. H. Zhang, Z. Cui, B. Jiang, Y. Wang, C. Bi, 'Subnanosecond in-plane magnetization switching induced by fieldlike spin-orbit torques from ferromagnets', Physical Review Applied 23, 044041 (2025). DOI: 10.1103/PhysRevApplied.23.044041, https://doi.org/10.1103/PhysRevApplied.23.044041
  19. A. Khan, S. Sharma, T. de Oliveira Schneider, M. Meinert, 'Uniaxial magnetic anisotropy and type-x/y current-induced magnetization switching in oblique-angle-deposited Ta/CoFeB/Pt and W/CoFeB/Pt heterostructures', Physical Review Applied 25, 044089 (2026). DOI: 10.1103/qkty-8dgb, https://doi.org/10.1103/qkty-8dgb (preprint arXiv:2510.14540, https://arxiv.org/abs/2510.14540)
  20. H. Wang, Z. Ye, L. Zhu, K. Cai, 'Practical design rules for easy-axis engineering in field-free in-plane SOT-MRAM cells', Applied Physics Letters 128, 192405 (2026). DOI: 10.1063/5.0328334, https://doi.org/10.1063/5.0328334