betaFeedback

WorksPhysics and Astronomy

Bulk spin-orbit torque-driven spin Hall nano-oscillators using PtBi alloys

Abstract

Spin-orbit-torque-driven auto-oscillations in spin Hall nano-oscillators (SHNOs) offer a transformative pathway toward energy-efficient, nanoscale microwave devices for next-generation neuromorphic computing and high-frequency technologies. A key requirement for achieving robust, sustained oscillations is reducing the threshold current (IthI_{\text{th}}), strongly governed by spin Hall efficiency (θSHθ_{\text{SH}}). However, conventional strategies to enhance θSHθ_{\text{SH}} face trade-offs, including high longitudinal resistivity, interfacial effects, and symmetry-breaking torques that limit performance. Here, we demonstrate a substantial enhancement of the bulk spin Hall effect in PtBi alloys, achieving over a threefold increase in θSHθ_{\text{SH}}, from 0.07 in pure Pt to 0.24 in Pt94.0_{94.0}Bi6.0_{6.0} and 0.19 in Pt91.3_{91.3}Bi8.7_{8.7}, as extracted from DC-bias spin-torque ferromagnetic resonance. The enhanced θSHθ_{\text{SH}} originates from bulk-dominated, extrinsic side-jump scattering across all PtBi compositions. Correspondingly, we observe a 42% and 32% reduction in IthI_{\text{th}} in 100 nm SHNOs based on Co40_{40}Fe40_{40}B20_{20}(3 nm)/Pt94.0_{94.0}Bi6.0_{6.0}(4 nm) and Co40_{40}Fe40_{40}B20_{20}(3 nm)/Pt91.3_{91.3}Bi8.7_{8.7}(4 nm), respectively. Structural characterization reveals reduced Pt crystallinity, along with emergence of preferred crystallographic orientations upon introducing higher Bi concentrations. Together, these results position PtBi alloys as a compelling alternative to conventional 5dd transition metals, enabling enhanced θSHθ_{\text{SH}} and significantly lower IthI_{\text{th}}, thus opening new avenues for energy-efficient neuromorphic computing and magnetic random access memory.

Introduction

The efficient control of magnetization via spin-orbit torque (SOT) [1], primarily generated by the Spin Hall effect (SHE) [2, 3], lies at the core of modern spintronics [4]. It plays a key role in magnetization switching [5, 6, 7, 8], in the generation and control of propagating spin waves [9, 10], in spin Hall nano-oscillators (SHNOs) [11, 12], and in the movement of domain walls and skyrmions [13, 14, 15]. In a heavy metal (HM) with large spin-orbit coupling, the SHE generates a sizeable transverse spin current density, js\mathbf{j}_{\mathrm{s}}, from a longitudinal charge current density, jc\mathbf{j}_{\mathrm{c}}. The generated js\mathbf{j}_{\mathrm{s}} can exert different torques on the magnetization of an adjacent ferromagnet (FM) layer, where the damping-like torque, τDL\tau_{\mathrm{DL}}, is the component collinear with the usual damping, τα\tau_{\alpha}, of the FM layer [16]. If τDL\tau_{\mathrm{DL}} balances τα\tau_{\alpha}, the threshold condition for FM auto-oscillations (AO), with a corresponding threshold current (IthI_{\mathrm{th}}), is reached in the SHNO [17, 18].

Over the past decade, SOT-driven nanoconstriction-based SHNOs have garnered significant attention thanks to their straightforward nano-fabrication, direct optical and gate access to the auto-oscillating FM region, and a rich magneto-dynamical behavior [19]. SHNOs have shown tremendous promise for mutual synchronization in chains [20, 21] and arrays [22, 23, 24], paving the way for applications in neuromorphic computing [25, 26, 27], Ising machines [28], magnonic conduits [10] and high frequency GHz technologies [29].

To reduce IthI_{\mathrm{th}}, a higher conversion of jc\mathbf{j}_{\mathrm{c}} to js\mathbf{j}_{\mathrm{s}}, defined as the spin Hall efficiency, θSH\theta_{\mathrm{SH}}, is required. θSH\theta_{\mathrm{SH}} can be expressed as js=ℏ2eθSH(jc×σ^)\mathbf{j}_{\mathrm{s}} = \frac{\hbar}{2e}\theta_{\mathrm{SH}}(\mathbf{j}_{\mathrm{c}} \times\hat{\boldsymbol{\sigma}}), where σ^\hat{\boldsymbol{\sigma}} is the polarization of the spin current, ee is the elementary charge, and ℏ\hbar is the reduced Planck constant, and the most commonly used HM Pt has a moderate θSH≈0.05\theta_{\mathrm{SH}} \approx0.05-0.09, from its intrinsic SHE [30]. Ion-implantation [31, 32, 33, 34] or sputtering [35, 36, 37] of non-metallic lighter impurities, such as S, O, N, P can also be used to tune the θSH\theta_{\mathrm{SH}} of Pt, Ta, W. However, such tuning requires careful materials engineering and often comes with a trade-off in the form of increased longitudinal resistivity (ρxx\rho_{\mathrm{xx}}), as observed in TaN, which can exhibit values as high as 3000 μΩ⋅cm\mu\Omega\cdot\mathrm{cm} [37]. In contrast, metallic impurities introduced by co-sputtering with the HM, neither require complex engineering nor introduce drastic changes in ρxx\rho_{\mathrm{xx}}. A better trade-off between θSH\theta_{\mathrm{SH}} and ρxx\rho_{\mathrm{xx}}, is crucial for low-power consumption in SOT magnetic random access memory (SOT-MRAM) [34, 38]. In this context, Pt, Ta, and W based alloys have been explored (θSH\theta_{\mathrm{SH}} … values in parentheses), such as Pt28_{28}Cu72_{72} (0.07) [39], Pt45_{45}Pd55_{55} (0.06) [40], Pt92_{92}Bi8_8 (0.10) [41]; W100−x_{100-x}Tax_x (-0.35 to -0.62) [42], and Cu100−x_{100-x}Tax_x (-0.04 to -0.09) [37]. Within this framework, Hayashi et al. showed that Pt100−x_{100-x}Bix_x/Co bilayers with substantial Bi concentrations (x=25x = 25–50) exhibit markedly improved θSH\theta_{\mathrm{SH}} and reduced switching current densities relative to pure Pt, identifying PtBi as a strong candidate for energy-efficient magnetization switching [43]. Complementing this, Münzenberg et al. demonstrated that Pt92_{92}Bi8_8-based hetero-structures can act as high-performance THz spintronic emitters, delivering broader bandwidths and higher central frequencies than conventional bilayer systems [44]. In addition, at higher Bi concentrations—such as in stoichiometric PtBi2_2—this material has attracted attention as a topological system, exhibiting giant three-dimensional Rashba-like spin splitting (αR≈4.36\alpha_{\mathrm{R}} \approx4.36 eV·Å) [45] and surface-bound superconducting gaps reaching up to 20 meV [46]. Despite these advances, a unified understanding of the reciprocal connection between θSH\theta_{\mathrm{SH}} and IthI_{\mathrm{th}} in Pt100−x_{100-x}Bix_x alloys—via both ST-FMR and direct AO measurements—remains unexplored.

Here, we report on a dramatic increase in θSH\theta_{\mathrm{SH}} by more than three times when alloying Pt with Bi, and a corresponding decrease of IthI_{\mathrm{th}} by 42% in SHNOs based on Pt94.0_{94.0}Bi6.0_{6.0} and Co40_{40}Fe40_{40}B20_{20}. We deposited Pt100−x_{100-x}Bix_x alloy thin films using electron-beam co-evaporation and systematically varied the Bi concentrations across x=0.0x = 0.0, 3.9, 6.0, and 8.7 (hereafter denoted as Pure Pt, Pt96.1_{96.1}Bi3.9_{3.9}, Pt94.0_{94.0}Bi6.0_{6.0}, and Pt91.3_{91.3}Bi8.7_{8.7}, respectively). Structural characterization by grazing-incidence X-ray diffraction (GIXRD) reveals a progressive loss of crystallinity with increasing Bi content, suggesting enhanced scattering contributions. Cross-sectional transmission electron microscopy (TEM) indicates a reduction in crystallinity of Pt, and a uniform distribution of Bi in Pt. Linewidth analysis of DC-biased spin-torque ferromagnetic resonance (ST-FMR) measurements vs. current yields a sharp increase in θSH\theta_{\mathrm{SH}} from 0.07 ±\pm 0.01 in pure Pt to 0.24 ±\pm 0.02 and 0.19 ±\pm 0.01 in the Pt94.0_{94.0}Bi6.0_{6.0} and Pt91.3_{91.3}Bi8.7_{8.7} alloys, respectively. To rule out artifacts and confirm the bulk origin of SOT, we demonstrate a clear sin⁡2ϕcos⁡ϕ\sin2\phi\cos\phi in-plane angular dependence of the SOT for all PtBi alloys. Thanks to the increase in θSH\theta_{\mathrm{SH}}, we observe a 42% and 32% reduction in IthI_{\mathrm{th}} for Pt94.0_{94.0}Bi6.0_{6.0} and Pt91.3_{91.3}Bi8.7_{8.7} based 100 nm SHNOs. Our findings of a higher θSH\theta_{\mathrm{SH}} driven by bulk-extrinsic SHE, accompanied by a reduced IthI_{\mathrm{th}}, establish Bi alloying in Pt as a robust strategy for engineering energy-efficient SHNOs, opening new avenues beyond conventional 5d5d transition metals.

Results and Discussion

Structural Characterization via GIXRD and TEM

Figure 1a shows Grazing incidence X-ray Diffraction (GIXRD) patterns of Pt100−x_{100-x}Bix_x thin films (see Experimental Section). Reflections from the (111), (200), (220), and (311) planes confirm a polycrystalline face-centered cubic (fcc) phase in pure Pt and Pt96.1_{96.1}Bi3.9_{3.9} [34, 47]. Increasing the Bi content reduces the overall reflection intensities, particularly that of the (220) peak, indicating a reduction in crystallinity [34, 48, 49]. Peak broadening becomes more pronounced in Pt94.0_{94.0}Bi6.0_{6.0} and Pt91.3_{91.3}Bi8.7_{8.7}. The intensity ratio I(111)/I(220)I_{(111)}/I_{(220)} rises from 0.39 (Pure Pt) to 1.65 (6% Bi), then falls to 1.07 at 8.7% Bi. Similarly, the lattice parameter increases from 3.953 to 3.966 Å (0–6% Bi), followed by a slight decrease to 3.959 Å at 8.7% Bi. This non-monotonic trend suggests a shift in preferred crystallographic orientation, likely due to strain-induced texturing at higher Bi concentrations, consistent with previous report [43].

Figure 1. (a) GIXRD patterns of the pure Pt stack (black) and PtBi alloy stacks with compositions Pt96.1_{96.1}Bi3.9_{3.9} (red), Pt94.0_{94.0}Bi6.0_{6.0} (blue), and Pt91.3_{91.3}Bi8.7_{8.7} (green). (b,c) Low-resolution cross-sectional bright-field TEM images of (b) the pure Pt stack (HR-Si/Ta(2.4 nm)/Pt(4.0 nm)/Co40_{40}Fe40_{40}B20_{20}(3.0 nm) and (c) the Pt94.0_{94.0}Bi6.0_{6.0} stack (HR-Si/Ta(2.4 nm)/Pt94.0_{94.0}Bi6.0_{6.0}(4.0 nm)/Co40_{40}Fe40_{40}B20_{20}(3.0 nm)) both showing uniform, well-defined layers with comparable total thicknesses. Both layers are capped with MgO(2.0 nm)/SiO2_2(3.0 nm). (d,e) High-resolution TEM images of the same stacks. (d) The pure Pt stack shows clear lattice fringes and sharp Bragg spots (inset FFT), indicative of a polycrystalline Pt structure. (e) The Pt94.0_{94.0}Bi6.0_{6.0} stack displays reduced fringe contrast and weaker Bragg spots (inset), suggesting lower crystallinity or a change in preferred orientation. (f,g) EDS elemental maps of Pt and Bi from HAADF-STEM for the pure Pt and Pt94.0_{94.0}Bi6.0_{6.0} stacks, respectively. Bi is uniformly distributed within the Pt94.0_{94.0}Bi6.0_{6.0} layer, with slight enrichment at the PtBi/Co40_{40}Fe40_{40}B20_{20} interface. Additional HAADF-STEM images and full elemental maps are provided in Supporting Information S1 and S2.

Figure 1b,Figure 1c display low-resolution cross-sectional bright-field TEM images of the pure Pt stack and the Pt94.0_{94.0}Bi6.0_{6.0} stack, respectively. Both stacks exhibit well-defined, uniform layers structures with comparable total thicknesses. Figure 1d,Figure 1e present high-resolution bright-field TEM images of the same samples. In the pure Pt sample (Figure 1d), clear lattice fringes are visible, and the fast Fourier transform (FFT) analysis (inset) shows distinct Bragg spots and a clear ring, indicating a polycrystalline structure of the Pt layer [31, 34, 50]. In contrast, the Pt94.0_{94.0}Bi6.0_{6.0} sample (Figure 1e) exhibits weaker lattice fringes and diminished Bragg spots (inset), suggesting reduced crystallinity or a shift in preferred crystallographic orientation, consistent with the GIXRD results. Figure 1f,Figure 1g provide Energy-dispersive X-ray spectroscopy (EDS) elemental maps from High-angle annular dark field scanning transmission electron microscopy (HAADF-STEM) for the same samples. In the Pt94.0_{94.0}Bi6.0_{6.0} stack (Figure 1g, right panel), Bi is uniformly distributed throughout the target Pt layer, with no sign of clustering. A modest enrichment of Bi is also observed at the Pt94.0_{94.0}Bi6.0_{6.0}/Co40_{40}Fe40_{40}B20_{20} interface. The impact of both bulk and interfacial Bi distributions on the SHE in Pt will be addressed in subsequent sections.

Spin-torque ferromagnetic resonance measurements

To evaluate how Bi doping modulates the SHE in Pt, we performed spin-torque ferromagnetic resonance (ST-FMR) measurements [16] (see Experimental section for details). Figure 2a illustrates the schematic of the ST-FMR measurement and Figure 2b shows the experimental setup. An in-plane microwave current IrfI_{\mathrm{rf}} with a power of 4 dBm was applied along the x-axis of the HM/FM bilayer via a ground-source-ground (GSG) coplanar waveguide (CPW). The external magnetic field HextH_{\mathrm{ext}} was swept ±500\pm500 mT at an angle ϕ=70∘\phi= 70^\circ between IrfI_{\mathrm{rf}} and HextH_{\mathrm{ext}}. The charge current density jc\mathbf{j}_{\mathrm{c}} (x-axis) driven by IrfI_{\mathrm{rf}} generates a spin current density js\mathbf{j}_{\mathrm{s}} along the z-axis with spin polarization σ^\hat{\boldsymbol{\sigma}} along the y-axis. This js\mathbf{j}_{\mathrm{s}} exerts an in-plane τDL\boldsymbol{\tau}_{\mathrm{DL}} on the magnetization m\mathbf{m} of the FM layer. Simultaneously, the Oersted field, hrfh_{\mathrm{rf}}, generated by IrfI_{\mathrm{rf}} acts as an out-of-plane Oersted field torque, τOF\boldsymbol{\tau}_{\mathrm{OF}}. Interfacial effects, rather than bulk mechanisms, can also induce out-of-plane field-like torques (τFL\boldsymbol{\tau}_{\mathrm{FL}}) arising from the Rashba effect, particularly in thinner FM layers (≲1\lesssim1 nm) interfaced with HM [51, 52], which may overlap with τOF\boldsymbol{\tau}_{\mathrm{OF}}. When τFL\boldsymbol{\tau}_{\mathrm{FL}} is small, as in bulk SHE heterostructures, the combined torques (τDL\boldsymbol{\tau}_{\mathrm{DL}} and τOF\boldsymbol{\tau}_{\mathrm{OF}}) govern the magnetization precession. The precession of m\mathbf{m} around the effective field Heff\mathbf{H}_{\mathrm{eff}} induces a time-dependent resistance variation via the anisotropic magnetoresistance (AMR), ΔR∝cos⁡2ϕ\Delta R \propto\cos^2 \phi (see Supporting Information S3). Mixing with IrfI_{\mathrm{rf}}, the varying ΔR\Delta R generates a rectified ST-FMR voltage, VdcV_{\mathrm{dc}}. The IrfI_{\mathrm{rf}} was amplitude modulated at 98.76 Hz, serving as a reference signal for lock-in detection (see Figure 2b).

Figure 2. a) Schematic of the SHE and the microscopic mechanism of the ST-FMR excitation, depicting the torques acting on the FM magnetization. b) The ST-FMR measurement technique and detection principle along with the micro-bar device connected to a GSG-CPW. A zoom-in SEM image of the red square shows the micro-bar. c) ST-FMR voltage (VdcV_{\mathrm{dc}}) at f=8f = 8 GHz of Pt94.0_{94.0}Bi6.0_{6.0}, with separate symmetric and antisymmetric components fit to Eq. (1). d) ΔH\Delta H vs ff for f=6f = 6–20 GHz, with solid black lines being linear fits to Equation (2). e) ff vs. HRH_{\mathrm{R}} for all samples evaluated in this study. Solid lines in (e) represent fits to the Kittel Equation (3).

Figure 2c shows the ST-FMR voltage spectrum measured at f=8f = 8 GHz and Idc=0I_{\mathrm{dc}} = 0 mA for Pt94.0_{94.0}Bi6.0_{6.0}. The measured VdcV_{\mathrm{dc}} is defined as [16]:

Vdc=SFsym(Hext)+AFasym(Hext),(1)V_{\mathrm{dc}} = S F_{\mathrm{sym}}(H_{\mathrm{ext}}) + A F_{\mathrm{asym}}(H_{\mathrm{ext}}), \tag*{(1)}

where,

Fsym(Hext)=(ΔH)2(μ0Hext−μ0HR)2+(ΔH)2F_{\mathrm{sym}}(H_{\mathrm{ext}}) = \frac{(\Delta H)^2}{(\mu_0 H_{\mathrm{ext}}-\mu_0 H_{\mathrm{R}})^2+(\Delta H)^2}

is the symmetric (Lorentzian) component, and

Fasym(Hext)=ΔH(μ0Hext−μ0HR)(μ0Hext−μ0HR)2+(ΔH)2F_{\mathrm{asym}}(H_{\mathrm{ext}}) = \frac{\Delta H(\mu_0 H_{\mathrm{ext}}-\mu_0 H_{\mathrm{R}})}{(\mu_0 H_{\mathrm{ext}}-\mu_0 H_{\mathrm{R}})^2+(\Delta H)^2}

is the antisymmetric (dispersive) component. The parameters SS and AA are the weight factors corresponding to the symmetric and antisymmetric components, respectively. ΔH\Delta H and HRH_{\mathrm{R}} are the linewidth (half-width-at-half-maximum) and resonance field of the ST-FMR spectra, respectively. Strikingly, a very high symmetric component SFsym(Hext)SF_{\mathrm{sym}}(H_{\mathrm{ext}}) confirming the high τDL\tau_{\mathrm{DL}} is observed for Pt94.0_{94.0}Bi6.0_{6.0}. Furthermore, in-plane angular ST-FMR measurements reveal that both the SS and AA weight factors exhibit a clear sin⁡2ϕcos⁡ϕ\sin2\phi\cos\phi dependence across all Pt100−x_{100-x}Bix_x compositions (see Supporting Information S4 for details). This angular dependence is a characteristic of a conventional bulk-SOT origin, free from torque symmetry-breaking, and rules out significant contributions from experimental artifacts [34, 53, 54].

Next, we extract the Gilbert damping parameter α\alpha by plotting ΔH\Delta H as a function of ff (see Figure 2d), using the relation

ΔH=ΔH0+2παγμ0f,(2)\Delta H = \Delta H_0 + \frac{2\pi\alpha}{\gamma\mu_0}f, \tag*{(2)}

where ΔH0\Delta H_0 denotes the ff-independent inhomogeneous linewidth broadening [55], and γ\gamma is the gyromagnetic ratio. A systematic increase in α\alpha is observed from 0.014±0.00050.014 \pm0.0005 (pure Pt) to 0.020±0.00070.020 \pm0.0007, 0.023±0.0010.023 \pm0.001 and 0.029±0.0020.029 \pm0.002 for Pt96.1_{96.1}Bi3.9_{3.9}, Pt94.0_{94.0}Bi6.0_{6.0}, and Pt91.3_{91.3}Bi8.7_{8.7}, respectively. This trend reflects enhanced spin current generation, potentially arising from spin pumping [41] and/or an increase in SHE [32, 33]. The extracted ΔH0\Delta H_0 remains low and positive (0.2–2.1 mT), indicating excellent film quality [31, 32]. Figure 2e depicts the frequency dependence of the resonance field, HRH_{\mathrm{R}} using the Kittel formula:

f=γμ02π(HR+HK)(HR+HK+Meff),(3)f = \frac{\gamma\mu_0}{2\pi}\sqrt{(H_{\mathrm{R}} + H_{\mathrm{K}})(H_{\mathrm{R}} + H_{\mathrm{K}} + M_{\mathrm{eff}})}, \tag*{(3)}

where, HkH_{\mathrm{k}} is in-plane magnetic anisotropy field and MeffM_{\mathrm{eff}} is effective demagnetization [56]. The extracted MeffM_{\mathrm{eff}} values—1141.4 mT (pure Pt), 1268.8 mT (Pt96.1_{96.1}Bi3.9_{3.9}), 1268.4 mT (Pt94.0_{94.0}Bi6.0_{6.0}), and 1088.8 mT (Pt91.3_{91.3}Bi8.7_{8.7})—along with HKH_{\mathrm{K}} values of 4–12 mT, confirm no significant variation in magnetic properties upon Bi incorporation.

To quantify θSH\theta_{\mathrm{SH}}, we use DC-bias ST-FMR [16], where a direct current IdcI_{\mathrm{dc}} is applied simultaneously with IrfI_{\mathrm{rf}} (see Figure 2b) under an external magnetic field HextH_{\mathrm{ext}}, to modulate the linewidth ΔH\Delta H. Figure 3a–Figure 3d show the change in linewidth, δ(μ0ΔH)\delta(\mu_{0}\Delta H), as a function of IdcI_{\mathrm{dc}} at f=8f = 8 GHz, measured at an angle ϕ=70∘\phi= 70^\circ with an input power of 4 dBm. Here, ϕ\phi is the angle between the combined current direction (Irf+IdcI_{\mathrm{rf}} + I_{\mathrm{dc}}) and the external field HextH_{\mathrm{ext}}. The angle ϕ=70∘\phi= 70^\circ was chosen to maximize the linewidth modulation since sin⁡ϕ\sin\phi reaches near its peak between 70–75°, and the AMR also exhibits significant variation within the 45°–75° range [18, 57] fruitful for a clean ST-FMR spectrum (see Equation (4)). The change in linewidth is defined as δ(μ0ΔH)=μ0ΔH∣Idc−μ0ΔH∣Idc=0\delta(\mu_{0}\Delta H) = \left.\mu_{0}\Delta H\right|_{I_{\mathrm{dc}}} - \left.\mu_{0}\Delta H\right|_{I_{\mathrm{dc}}=0}, which varies linearly with IdcI_{\mathrm{dc}}. We then define the effective current-dependent damping parameter, αeff\alpha_{\mathrm{eff}} given as αeff=γ2πfδ(μ0ΔH)\alpha_{\mathrm{eff}} = \frac{\gamma}{2\pi f}\delta(\mu_{0}\Delta H). Reversing the polarity of HextH_{\mathrm{ext}} from (+Hext+H_{\mathrm{ext}}) ϕ=70∘\phi=70^\circ to (−Hext-H_{\mathrm{ext}}) at ϕ=250∘\phi=250^\circ reverses the magnetization direction of Co40_{40}Fe40_{40}B20_{20}, which correspondingly reverses the slope of δ(μ0ΔH)/Idc\delta(\mu_{0}\Delta H)/I_{\mathrm{dc}}, confirming the conventional SHE induced SOT [58]. Using a parallel resistor model (see Supporting Information S5 for details on calculating ρxx\rho_{\mathrm{xx}}), we estimate the current density in the HM layer, jdc,HMj_{\mathrm{dc,HM}}, and extract θSH\theta_{\mathrm{SH}} from the slope ΔαeffΔjdc,HM\frac{\Delta\alpha_{\mathrm{eff}}}{\Delta j_{\mathrm{dc,HM}}} according to [16]:

Figure 3. Current-induced modulation of linewidth, δ(μ0ΔH)\delta(\mu_{0}\Delta H) as a function of IdcI_{\mathrm{dc}} at f=8f = 8 GHz for a) Pure Pt, b) Pt96.1_{96.1}Bi3.9_{3.9}, c) Pt94.0_{94.0}Bi6.0_{6.0}, and d) Pt91.3_{91.3}Bi8.7_{8.7}. Open circles correspond to positive magnetic field sweep (+Hext+H_{\mathrm{ext}}), while filled circles correspond to negative sweep (−Hext-H_{\mathrm{ext}}). The solid black lines are linear fits. The increasing slope of δ(μ0ΔH)/Idc\delta(\mu_{0}\Delta H)/I_{\mathrm{dc}} with Bi impurity in Pt, indicates enhanced θSH\theta_{\mathrm{SH}}.

θSH=[2eℏ(H0+Meff2)μ0Mst∣ΔαeffΔjdc,HM∣]/sin⁡ϕ,(4)\theta_{\mathrm{SH}} = \left[\frac{2e}{\hbar}\left(H_{0}+\frac{M_{\mathrm{eff}}}{2}\right)\mu_{0}M_{\mathrm{s}}t\left|\frac{\Delta\alpha_{\mathrm{eff}}}{\Delta j_{\mathrm{dc,HM}}}\right|\right] \big/ \sin\phi, \tag*{(4)}

where MsM_{\mathrm{s}} and tt are the saturation magnetization and thickness of Co40_{40}Fe40_{40}B20_{20}, respectively, and MeffM_{\mathrm{eff}} is the effective demagnetization obtained from Kittel fitting [55, 59] (see Figure 2e). We observe a slight increase in θSH\theta_{\mathrm{SH}} from 0.07±0.010.07 \pm0.01 (Pure Pt) to 0.12±0.020.12 \pm0.02 (Pt96.1_{96.1}Bi3.9_{3.9}), with a small doping of 3.9%. However, increasing the Bi concentration to 6.0% and 8.7% yields a remarkable enhancement in θSH\theta_{\mathrm{SH}} to 0.24±0.020.24 \pm0.02 (Pt94.0_{94.0}Bi6.0_{6.0}) and 0.19±0.010.19 \pm0.01 (Pt91.3_{91.3}Bi8.7_{8.7}). In particular, the highest θSH\theta_{\mathrm{SH}} occurs at higher Bi concentrations where Pt loses its crystalline texture, as seen in Figure 1. This supports the correlation between reduced crystallinity and enhanced θSH\theta_{\mathrm{SH}} in 5d5d transition metals. Recent experimental studies by Parkin et al. [38] and Fukuma et al. [34] have similarly demonstrated that θSH\theta_{\mathrm{SH}} peaks near the crystalline-amorphous boundary, where extrinsic scattering from impurities is maximized, which we will explore in later sections.

Spin Hall nano-oscillators

We fabricated 100 nm wide SHNOs (see Experimental Section for details) and measured AO on all Pt100−xBix/Co40Fe40B20\mathrm{Pt}_{100-x}\mathrm{Bi}_{x}/\mathrm{Co}_{40}\mathrm{Fe}_{40}\mathrm{B}_{20} hetero-structures to observe the effect of a higher θSH\theta_{\mathrm{SH}} on IthI_{\mathrm{th}}. A schematic of the setup is shown in Figure 4a. A direct current, IdcI_{\mathrm{dc}} (along the x-axis) was applied through a Bias Tee, while the resulting microwave signals were extracted through the high-frequency port, amplified using a low-noise amplifier (LNA), and analyzed with a spectrum analyzer (SA). A moderate out-of-plane field (θ=84∘\theta=84^{\circ}) was selected to achieve a weak negative nonlinearity [60]. The in-plane angle, ϕ=20∘\phi=20^{\circ}, was chosen to ensure adequate electrical sensitivity of the signal (ϕ\phi is the angle between the y-axis and HextH_{\mathrm{ext}}).

Figure 4. a) Schematic of the SHNO and measurement setup. b) SEM image of the GSG-based CPW with SHNO. c) Enlarged SEM of the 100 nm wide SHNO (black scale bars in b,c). PSD from the 100 nm SHNO as a function of IdcI_{\mathrm{dc}} at μ0Hext=480\mu_{0}H_{\mathrm{ext}} = 480 mT, ϕ=20∘\phi= 20^{\circ}, θ=84∘\theta= 84^{\circ} for d) Pure Pt, e) Pt96.1Bi3.9\mathrm{Pt}_{96.1}\mathrm{Bi}_{3.9}, f) Pt94.0Bi6.0\mathrm{Pt}_{94.0}\mathrm{Bi}_{6.0}, and g) Pt91.3Bi8.7\mathrm{Pt}_{91.3}\mathrm{Bi}_{8.7}. Color bars indicate peak power (dB above noise floor). The white arrows mark respective IthI_{\mathrm{th}}.

Figure 4b shows the scanning electron microscope (SEM) image of the device under test, and Figure 4c presents the enlarged SEM image of the 100 nm wide SHNO. The actual measured width ww is in good agreement with the nominal design value. To understand the reciprocal relationship between the enhancement in θSH\theta_{\mathrm{SH}} and the reduction in IthI_{\mathrm{th}}, we performed current-dependent AO and extracted the Power Spectral Density (PSD) generated by the 100 nm wide SHNO under μ0Hext=480\mu_{0}H_{\mathrm{ext}} = 480 mT (ϕ=20∘\phi=20^{\circ} and θ=84∘\theta=84^{\circ}) for all samples, as shown in Figure 4d-Figure 4g. A non-monotonic current dependence of the AO frequency is observed, attributed to a localized AO “bullet” mode promoted by negative nonlinearity, N\mathcal{N}, typical for a FM layer with its magnetization far from the film normal [60]. The IthI_{\mathrm{th}} is 1.30 mA for pure Pt, decreasing to 0.95 mA for Pt96.1_{96.1}Bi3.9_{3.9}. At higher Bi concentrations, IthI_{\mathrm{th}} drops significantly to 0.75 mA (42% reduction) for Pt94.0_{94.0}Bi6.0_{6.0} and 0.89 mA (32% reduction) for Pt91.3_{91.3}Bi8.7_{8.7}. Concurrently, the peak and integrated output power increase while the linewidth decreases at higher Bi concentrations compared to pure Pt. See Supporting Information S6 for AMR, S7 for AO at other fields, S8 for IthI_{\mathrm{th}} extraction, and S9 for peak power, output power, and linewidth.

Reciprocity and mechanism of spin Hall effect in PtBi alloys

Building on the structural analysis and the observed reciprocity between ST-FMR and AO, several key parameters—ρxx\rho_{xx}, θSH\theta_{\mathrm{SH}}, and IthI_{\mathrm{th}}—are analyzed across all Pt100−x_{100-x}Bix_x alloys (see Figure 5a-Figure 5c). For pure Pt and 3.1% Bi doping, θSH\theta_{\mathrm{SH}} increases from 0.07 to 0.12, with a corresponding rise in ρxx\rho_{xx} from 65 to 270 μΩ⋅cm\mu\Omega\cdot\mathrm{cm}, reducing IthI_{\mathrm{th}} from 1.30 mA to 0.97 mA. At 6.0% Bi, ρxx\rho_{xx} rises marginally to 288 μΩ⋅cm\mu\Omega\cdot\mathrm{cm}, while θSH\theta_{\mathrm{SH}} significantly rises to 0.24, yielding the lowest IthI_{\mathrm{th}} of 0.75 mA. At 8.7% Bi, ρxx\rho_{xx} further increases to 301 μΩ⋅cm\mu\Omega\cdot\mathrm{cm}, but θSH\theta_{\mathrm{SH}} saturates and decreases to 0.19, causing IthI_{\mathrm{th}} to slightly rise to 0.89 mA. This interplay between θSH\theta_{\mathrm{SH}} and ρxx\rho_{xx} is critical for minimizing power consumption, expressed by the factor, ρxx/θSH2\rho_{xx}/\theta_{\mathrm{SH}}^{2} relevant for SOT-MRAM applications (Supporting Information S5) [34, 38]. The output power is visibly higher at 6% and 8.7% Bi impurity in Pt compared to pure Pt (Figure 4d–Figure 4g). Notably, linewidth narrows to ≈25\approx25 MHz at these higher Bi concentrations, likely due to increased ρxx\rho_{xx} redirecting current through Co40_{40}Fe40_{40}B20_{20} enhancing output power and reducing linewidth (Supporting Information S9). Furthermore, despite the increase in α\alpha with Bi doping, the enhanced τDL\tau_{\mathrm{DL}} from the SHE compensates for τα\tau_{\alpha}, leading to a reduction in IthI_{\mathrm{th}}. This reduction would likely have been more significant if α\alpha had remained constant or decreased (see Figure 2d), complicating the direct assessment of SOT effects. The actual effect of torque on the AO is thus likely underestimated, and FM layer with lower α\alpha, such as GdFeCo [55] could further enhance the reduction in IthI_{\mathrm{th}}.

Figure 5. a) ρxx\rho_{xx}, b) θSH\theta_{\mathrm{SH}}, and c) IthI_{\mathrm{th}} versus Bi content xx in Pt100−xBix\mathrm{Pt}_{100-x}\mathrm{Bi}_{x} (wt.%). Error bars in (b) represent standard error from equation 2 and 4; in (c), mean deviation (MD) from IthI_{\mathrm{th}} datasets (Supporting Information S7 for datasets; see Experimental Section for MD details). θSH\theta_{\mathrm{SH}} plotted against d) σxx\sigma_{xx} and e) ρxx\rho_{xx}. Inset in (d) shows bulk SHE schematic in PtBi alloy. f) ρimpSH\rho_{\mathrm{imp}}^{\mathrm{SH}} as a function of ρimp2\rho_{\mathrm{imp}}^{2} with linear fits as a solid line.

To elucidate the mechanism underlying the SHE, we examine the electrical conductivity σxx\sigma_{xx}. Figure 5d shows θSH\theta_{\mathrm{SH}} vs. σxx\sigma_{xx}, with values lying in the bad-metal/dirty-metal regime (σxx≲104 Ω−1cm−1\sigma_{xx} \lesssim10^{4}\ \Omega^{-1}\mathrm{cm}^{-1}). In this regime, a linear dependence of θSH\theta_{\mathrm{SH}} on σxx\sigma_{xx} indicates a dominant intrinsic SHE contribution in Pt as demonstrated by Casanova et al. [61]. However, no clear linear trend is observed, suggesting coexistence of intrinsic and extrinsic side-jump scattering mechanisms [34, 48, 61]. Moreover, the increase of θSH\theta_{\mathrm{SH}} with ρxx\rho_{xx} (Figure 5e) supports a dominant role of side-jump scattering and/or disorder-enhanced intrinsic contributions [37]. In metals with strong spin-orbit coupling, the anomalous Hall effect (AHE) scales with ρxx\rho_{xx} either linearly (extrinsic skew scattering) or quadratically (extrinsic side-jump scattering oralsevol intrinsic mechanisms) [62]. Recent studies have established that SHE shares this scaling behavior [63]. The observed quadratic dependence of θSH\theta_{\mathrm{SH}} on ρxx\rho_{xx} hints at the contribution of bulk-extrinsic side-jump scattering (Figure 5e) and rules out significant interfacial Bi effects on SHE, as confirmed by TEM and consistent with bulk alloy characteristics. To further confirm the mechanism, the intrinsic SHE contribution was subtracted, revealing ρimpSH∝ρimp2\rho_{\mathrm{imp}}^{\mathrm{SH}} \propto\rho_{\mathrm{imp}}^{2} (Figure 5f), indicative of extrinsic side-jump scattering dominance [32, 33, 39, 48, 56, 64]. Here, ρimpSH\rho_{\mathrm{imp}}^{\mathrm{SH}} denotes the impurity-induced spin Hall resistivity, and ρimp\rho_{\mathrm{imp}} the resistivity due to impurities (See Supporting Information S10 for details on mechanism of SHE, including the exclusion of skew scattering as a contributing mechanism, and the extraction method for ρimpSH\rho_{\mathrm{imp}}^{\mathrm{SH}} and ρimp\rho_{\mathrm{imp}}). Therefore, increasing Bi concentration in Pt enhances extrinsic side-jump scattering, substantially boosting θSH\theta_{\mathrm{SH}} and reducing the threshold current IthI_{\mathrm{th}} necessary to drive robust AO, thereby enabling energy-efficient SHNOs.

Conclusion

In summary, we demonstrate that Bi doping in Pt enhances θSH\theta_{\mathrm{SH}} by over threefold and substantially reduces the threshold current IthI_{\mathrm{th}} in SHNOs, enabling a superior ρxx\rho_{xx} trade-off crucial for low-power spintronic applications. The concurrent enhancement of θSH\theta_{\mathrm{SH}} and reduction in IthI_{\mathrm{th}} underscores the reciprocal relationship between the spin Hall effect and auto-oscillations, as revealed by complementary ST-FMR and AO measurements. Furthermore, angular-dependent ST-FMR confirms a sin⁡2ϕcos⁡ϕ\sin2\phi\cos\phi symmetry across all Pt100−xBix\mathrm{Pt}_{100-x}\mathrm{Bi}_x alloys, indicative of a bulk-origin, conventional spin–orbit torque mechanism. GIXRD analysis confirms that increased Bi doping in Pt leads to progressive reduction in crystallinity, significant peak broadening, and preferred crystallographic orientation, indicating structural modifications in the Pt100−xBix\mathrm{Pt}_{100-x}\mathrm{Bi}_x alloys. These findings are corroborated by cross-sectional TEM, which reveals a uniform distribution of Bi throughout the Pt layer and a concurrent loss of crystallinity of Pt. The non-linear dependence of θSH\theta_{\mathrm{SH}} on σxx\sigma_{xx} and ρxx\rho_{xx} further reveals a ρimpSH∝ρimp2\rho_{\mathrm{imp}}^{\mathrm{SH}} \propto\rho_{\mathrm{imp}}^2 scaling, confirming extrinsic side-jump scattering as the dominant mechanism behind the enhanced SHE. Collectively, our findings establish a framework for energy-efficient spintronics beyond conventional 5d transition metals. By demonstrating Bi alloying with Pt as an effective strategy for enhancing charge-to-spin conversion, our work offers design principles for next-generation neuromorphic computing, SOT-MRAM, Ising machines, and GHz spintronic devices.

Experimental Section

Thin film preparation

High-resistance silicon (HR-Si) wafers (>10,000 Ω⋅cm> 10{,}000\ \Omega\cdot\mathrm{cm}) were used as substrates. The Pt100−xBix(4.0 nm)/Co40Fe40B20(3.0 nm)\mathrm{Pt}_{100-x}\mathrm{Bi}_x(4.0\ \mathrm{nm})/\mathrm{Co}_{40}\mathrm{Fe}_{40}\mathrm{B}_{20}(3.0\ \mathrm{nm}) were grown on HR-Si/Ta(2.4 nm), and further protected with a capping layer of MgO(2.0 nm)/SiO2_2(3.0 nm). Ta and Co40Fe40B20\mathrm{Co}_{40}\mathrm{Fe}_{40}\mathrm{B}_{20} were sputtered at 11 W with 0.25 sccm Ar flow at a base pressure of 4×10−84 \times10^{-8} mbar, rising to 4×10−34 \times10^{-3} mbar during deposition. Other layers were deposited via electron beam evaporation. Pt100−xBix\mathrm{Pt}_{100-x}\mathrm{Bi}_x was synthesized in situ by co-evaporation, with deposition rates monitored using a quartz crystal microbalance (QCM). The Bi source was stabilized before Pt adjustment. The deposition rate ratio required to achieve the desired Pt100−xBix\mathrm{Pt}_{100-x}\mathrm{Bi}_x composition was calculated as: vavb=MaρbMbρa⋅r\frac{v_a}{v_b} = \frac{M_a \rho_b}{M_b \rho_a} \cdot r, where vav_a and vbv_b are the deposition rates, MaM_a and MbM_b are the molar masses, ρa\rho_a and ρb\rho_b are the material densities, and rr is the target molar ratio [44].

Grazing Incidence X-ray Diffraction (GIXRD)

GIXRD measurements were used to determine the full-width-at-half-maximum (FWHM) of the diffraction peaks. Data were acquired using a Mat:Nordic SAXSLAB instrument with a Rigaku 003 microfocusing Cu X-ray source (parallel beam from a two-bounce monochromator) and two Dectris detectors: Pilatus3 300K R (orthogonal) and 100K (goniometer circle). The beam path was evacuated to suppress air scattering, and a 1∘1^\circ incidence angle was employed to maximized signal intensity. Two-dimensional diffraction images were processed using SAXSGUI. FWHM values and lattice parameters were determined using TOPAS v6 (Bruker AXS, 2016), applying a pure platinum (Fm3m) structural model and a pseudo-Voigt (PVII) profile. Peak broadening was analyzed separately. Exposure time was 3h per sample. Instrumental broadening was assessed using LaB6 powder.

Cross-sectional high-resolution TEM and HAADF-STEM imaging

TEM measurements were carried out using a JEOL monochromatic ARM200F microscope, which is equipped with a Schottky field emission gun, a double-Wien monochromator, a probe aberration corrector, an image aberration corrector, a Gatan image filter (GIF) Continuum, a Gatan Oneview camera, as well as a double silicon drift detector (SDD) for energy dispersive X-ray spectroscopy (EDS). The microscope was operated at 200 kV for the TEM measurements. TEM specimens were prepared by using a FEI Versa 3D focus ion beam–scanning electron microscope (FIB-SEM). After depositing a protection layer containing Pt and C using first the electron beam and then the Ga ion beam in the FIB-SEM, a lamella of the material was cut out using an ion beam at 30 kV and 1 nA. After transferring the lamella to a Cu TEM grid, the lamella was gradually thinned down by the ion beam. The thinning process was performed first at 30 kV and with a gradually decreasing beam current from 1nA to 100 pA. Then, the gentle polishing of the specimen was carried out with an ion beam energy of 5 kV and 2 kV to minimize ion beam effect.

Device Fabrication

ST-FMR bars (14×4 μm214 \times4\ \mu\mathrm{m}^{2}) and SHNOs with width of 100 nm were fabricated via electron-beam lithography (Raith EBPG 5200), followed by Ar-ion milling. The GSG-CPW contact pads were subsequently fabricated by mask-less ultraviolet lithography (Heidelberg Instruments MLA 150) and a lift-off technique, followed by Cu(800 nm)/Pt(20 nm) electrodes, deposited by DC magnetron sputtering. Detailed fabrication process can be found in Ref. [65].

ST-FMR and Auto-oscillation (AO) measurements

For ST-FMR, signal generator (RSSMB 100 A) was used to generate IrfI_{\mathrm{rf}} and the ST-FMR voltage VdcV_{\mathrm{dc}} was detected using a lock-in amplifier (SR830). For AO, IdcI_{\mathrm{dc}} was applied using Keithley 2400 sourcemeter and AO signal was then amplified by a low noise Amplifier (LNF LNR4 14B), and observed using a spectrum analyser (RSFSV) with a resolution bandwidth of 1 MHz. All measurements were performed at room temperature.

Statistical Analysis

The mean deviation (MD) quantifies dispersion in a dataset x1,x2,…,xnx_{1}, x_{2}, \ldots, x_{n} and is computed with respect to the mean (xˉ\bar{x}) or median (MM). The mean deviation about the mean is: MDxˉ=1n∑i=1n∣xi−xˉ∣MD_{\bar{x}} = \frac{1}{n}\sum_{i=1}^{n}|x_{i}-\bar{x}|, where xˉ\bar{x} is the arithmetic mean, xˉ=1n∑i=1nxi\bar{x} = \frac{1}{n}\sum_{i=1}^{n}x_{i}. Here, ∣xi−xˉ∣|x_{i}-\bar{x}| represents the absolute deviation from the mean.

Supporting Information Supporting Information is available from the corresponding authors upon request.

Acknowledgements Financial support from the Horizon 2020 research and innovation programme (Grant Nos. 899559 “SpinAge” and 835068 “TOPSPIN”), and the Swedish Research Council (VR Grant No. 2024-01943), is gratefully acknowledged. Financial support from the Swedish Research Council (VR) and the Swedish Foundation for Strategic Research (SSF) for access to ARTEMI, the Swedish National Infrastructure in Advanced Electron Microscopy (Grant Nos. 2021-00171 and RIF21-0026), is also acknowledged. This work was performed in part at the Chalmers Material Analysis Laboratory (CMAL). We thank Dr. Michał Strach (CMAL) for his assistance and insightful discussions regarding the GIXRD measurements.

Conflict of Interest

The authors declare no competing interests.

Authors’ contributions

US, AK, TSP, MM and JÅ initiated the idea and planned the study. HH prepared the thin films with input from TSP and AK. AK fabricated the devices. US performed all electrical auto-oscillations, spin-orbit torque measurements, and analysis with input from AK and MR. LJZ and ABY performed TEM measurements with input from EO. JÅ and MM coordinated and supervised the project. All authors contributed to the data analysis and interpretation of the results and co-wrote the manuscript.

Data Availability Statement

The data that support the findings of this study are available from the corresponding authors upon reasonable request.

References

  1. [1]Qiming Shao, Peng Li, Luqiao Liu, Hyunsoo Yang, Shunsuke Fukami, Armin Razavi, Hao Wu, Kang Wang, Frank Freimuth, Yuriy Mokrousov, et al. Roadmap of spin–orbit torques. IEEE Transactions on Magnetics, 57(7):1–39, 2021.
  2. [2]JE Hirsch. Spin hall effect. Physical review letters, 83(9):1834, 1999.
  3. [3]Mikhail I Dyakonov and VI Perel. Current-induced spin orientation of electrons in semiconductors. Physics Letters A, 35(6):459–460, 1971.DOI
  4. [4]B. Dieny, I. L. Prejbeanu, K. Garello, P. Gambardella, P. Freitas, et al. Opportunities and challenges for spintronics in the microelectronics industry. Nat. Electron., 3(8):446–459, Aug 2020.DOI
  5. [5]Luqiao Liu, Chi-Feng Pai, Y. Li, H. W. Tseng, D. C. Ralph, and R. A. Buhrman. Spin-Torque Switching with the Giant Spin Hall Effect of Tantalum. Science, 336:555–558, 2012.DOI
  6. [6]Guoqiang Yu, Pramey Upadhyaya, Yabin Fan, Juan G Alzate, Wanjun Jiang, Kin L Wong, So Takei, Scott A Bender, Li-Te Chang, Ying Jiang, et al. Switching of perpendicular magnetization by spin–orbit torques in the absence of external magnetic fields. Nat. Nanotechnol., 9(7):548–554, 2014.DOI
  7. [7]Ioan Mihai Miron, Kevin Garello, Gilles Gaudin, Pierre-Jean Zermatten, Marius V Costache, Stéphane Auffret, Sébastien Bandiera, Bernard Rodmacq, Alain Schuhl, and Pietro Gambardella. Perpendicular switching of a single ferromagnetic layer induced by in-plane current injection. Nature, 476(7359):189–193, 2011.
  8. [8]K Cai, G Talmelli, K Fan, S Van Beek, V Kateel, M Gupta, MG Monteiro, M Ben Chroud, G Jayakumar, A Trovato, et al. First demonstration of field-free perpendicular sot-mram for ultrafast and high-density embedded memories. In 2022 International Electron Devices Meeting (IEDM), pages 36–2. IEEE, 2022.DOI
  9. [9]Himanshu Fulara, Mohammad Zahedinejad, Roman Khymyn, AA Awad, Shreyas Muralidhar, Mykola Dvornik, and Johan Åkerman. Spin-orbit torque–driven propagating spin waves. Science advances, 5(9):eaax8467, 2019.DOI
  10. [10]Akash Kumar, Avinash Kumar Chaurasiya, Victor H González, Nilamani Behera, Ademir Alemán, Roman Khymyn, Ahmad A Awad, and Johan Åkerman. Spin-wave-mediated mutual synchronization and phase tuning in spin hall nano-oscillators. Nature Physics, 21:245–252, 2025.DOI
  11. [11]VE Demidov, S Urazhdin, A Zholud, AV Sadovnikov, and SO Demokritov. Nanoconstriction-based spin-hall nano-oscillator. Applied Physics Letters, 105(17), 2014.DOI
  12. [12]Akash Kumar, Artem Litvinenko, Nilamani Behera, Ahmad A Awad, Roman Khymyn, and Johan Åkerman. Mutual synchronization in spin-torque and spin hall nano-oscillators. Nanomagnets as Dynamical Systems: Physics and Applications, pages 143–182, 2024.DOI
  13. [13]Satoru Emori, Uwe Bauer, Sung-Min Ahn, Eduardo Martinez, and Geoffrey SD Beach. Current-driven dynamics of chiral ferromagnetic domain walls. Nature materials, 12(7):611–616, 2013.DOI
  14. [14]Nirel Bernstein, Hang Li, Benjamin Assouline, Yong-Chang Lau, Igor Rozhansky, Wen-hong Wang, and Amir Capua. Spin-torque skyrmion resonance in a frustrated magnet. Nat. Commun., 16(1):1–9, 2025.DOI
  15. [15]AV Khvalkovskiy, V Cros, D Apalkov, V Nikitin, M Krounbi, KA Zvezdin, A Anane, J Grollier, and A Fert. Matching domain-wall configuration and spin-orbit torques for efficient domain-wall motion. Phys. Rev. B, 87(2):020402, 2013.DOI
  16. [16]Luqiao Liu, Takahiro Moriyama, D_C Ralph, and R_A Buhrman. Spin-torque ferromagnetic resonance induced by the spin hall effect. Physical review letters, 106(3):036601, 2011.pubmed.ncbi.nlm.nih.gov/21405285
  17. [17]RH Liu, WL Lim, and S Urazhdin. Spectral characteristics of the microwave emission by the spin hall nano-oscillator. Physical review letters, 110(14):147601, 2013.DOI
  18. [18]Shinya Kasai, Kouta Kondou, Hiroaki Sukegawa, Seiji Mitani, Kazuhito Tsukagoshi, and Yoshichika Otani. Modulation of effective damping constant using spin hall effect. Applied Physics Letters, 104(9), 2014.DOI
  19. [19]Tingsu Chen, Randy K Dumas, Anders Eklund, Pranaba K Muduli, Afshin Houshang, Ahmad A Awad, Philipp Dürrenfeld, B Gunnar Malm, Ana Rusu, and Johan Åkerman. Spin-torque and spin-hall nano-oscillators. Proceedings of the IEEE, 104(10):1919–1945, 2016.arxiv.org/abs/1512.03162
  20. [20]AA Awad, Ph Dürrenfeld, A Houshang, M Dvornik, Ezio Iacocca, RK Dumas, and Johan Åkerman. Long-range mutual synchronization of spin hall nano-oscillators. Nature Physics, 13(3):292–299, 2017.DOI
  21. [21]Akash Kumar, Himanshu Fulara, Roman Khymyn, Artem Litvinenko, Mohammad Za-hedinejad, Mona Rajabali, Xiaotian Zhao, Nilamani Behera, Afshin Houshang, Ahmad A Awad, et al. Robust mutual synchronization in long spin hall nano-oscillator chains. Nano Letters, 23(14):6720–6726, 2023.DOI
  22. [22]Mohammad Zahedinejad, Ahmad A Awad, Shreyas Muralidhar, Roman Khymyn, Himanshu Fulara, Hamid Mazraati, Mykola Dvornik, and Johan Åkerman. Two-dimensional mutually synchronized spin hall nano-oscillator arrays for neuromorphic computing. Nature nanotechnology, 15(1):47–52, 2020.DOI
  23. [23]Nilamani Behera, Avinash Kumar Chaurasiya, Victor H González, Artem Litvinenko, Lakhan Bainsla, Akash Kumar, Roman Khymyn, Ahmad A Awad, Himanshu Fulara, and Johan Åkerman. Ultra-low current 10 nm spin hall nano-oscillators. Advanced Materials, 36(5):2305002, 2024.arxiv.org/abs/2305.06779
  24. [24]Nilamani Behera, Avinash Kumar Chaurasiya, Akash Kumar, Roman Khymyn, Artem Litvinenko, Lakhan Bainsla, Ahmad A Awad, and Johan Åkerman. Ultra-large mutually synchronized networks of 10 nm spin hall nano-oscillators. arXiv preprint arXiv:2501.18321, 2025.arxiv.org/abs/2501.18321
  25. [25]Mohammad Zahedinejad, Himanshu Fulara, Roman Khymyn, Afshin Houshang, Mykola Dvornik, Shunsuke Fukami, Shun Kanai, Hideo Ohno, and Johan Åkerman. Memristive control of mutual spin hall nano-oscillator synchronization for neuromorphic computing. Nature materials, 21(1):81–87, 2022.DOI
  26. [26]Jacob Torrejon, Mathieu Riou, Flavio Abreu Araujo, Sumito Tsunegi, Guru Khalsa, Damien Querlioz, Paolo Bortolotti, Vincent Cros, Kay Yakushiji, Akio Fukushima, et al. Neuromorphic computing with nanoscale spintronic oscillators. Nature, 547(7664):428–431, 2017.
  27. [27]Julie Grollier, Damien Querlioz, KY Camsari, Karin Everschor-Sitte, Shunsuke Fukami, and Mark D Stiles. Neuromorphic spintronics. Nature electronics, 3(7):360–370, 2020.DOI
  28. [28]Afshin Houshang, Mohammad Zahedinejad, Shreyas Muralidhar, Jakub Chęciński, Roman Khymyn, Mona Rajabali, Himanshu Fulara, Ahmad A Awad, Mykola Dvornik, and Johan Åkerman. Phase-binarized spin hall nano-oscillator arrays: Towards spin hall ising machines. Physical Review Applied, 17(1):014003, 2022.DOI
  29. [29]Jong-Guk Choi, Jaehyeon Park, Min-Gu Kang, Doyoon Kim, Jae-Sung Rieh, Kyung-Jin Lee, Kab-Jin Kim, and Byong-Guk Park. Voltage-driven gigahertz frequency tuning of spin hall nano-oscillators. Nature communications, 13(1):3783, 2022.DOI
  30. [30]Yi Wang, Praveen Deorani, Xuepeng Qiu, Jae Hyun Kwon, and Hyunsoo Yang. Determination of intrinsic spin hall angle in pt. Applied Physics Letters, 105(15), 2014.DOI
  31. [31]Utkarsh Shashank, Rohit Medwal, Taiga Shibata, Razia Nongjai, Joseph Vimal Vas, Martial Duchamp, Kandasami Asokan, Rajdeep Singh Rawat, Hironori Asada, Surbhi Gupta, et al. Enhanced spin hall effect in s-implanted pt. Advanced Quantum Technologies, 4(1):2000112, 2021.DOI
  32. [32]Utkarsh Shashank, Rohit Medwal, Yoji Nakamura, John Rex Mohan, Razia Nongjai, Asokan Kandasami, Rajdeep Singh Rawat, Hironori Asada, Surbhi Gupta, and Yasuhiro Fukuma. Highly dose dependent damping-like spin–orbit torque efficiency in o-implanted pt. Applied Physics Letters, 118(25), 2021.DOI
  33. [33]Utkarsh Shashank, Yoji Nakamura, Yu Kusaba, Takafumi Tomoda, Razia Nongjai, Asokan Kandasami, Rohit Medwal, Rajdeep Singh Rawat, Hironori Asada, Surbhi Gupta, et al. Disentanglement of intrinsic and extrinsic side-jump scattering induced spin hall effect in n-implanted pt. Phys. Rev. B, 107(6):064402, 2023.DOI
  34. [34]Utkarsh Shashank, Takafumi Tomoda, Arun Jacob Mathew, Garima Vashisht, Koki Imai, Yu Kusaba, Chung-Li Dong, Chi-Liang Chen, Yoichi Horibe, Manabu Ishimaru, et al. Giant spin-orbit torque induced by spin hall effect in amorphous pt (p) alloys. NPG Asia Materials, 17(1):15, 2025.DOI
  35. [35]Liupeng Yang, Yining Fei, Kaiyuan Zhou, Lina Chen, Qingwei Fu, Liyuan Li, Chunjie Yan, Haotian Li, Youwei Du, and Ronghua Liu. Maximizing spin–orbit torque efficiency of ta (o) /py via modulating oxygen-induced interface orbital hybridization. Applied Physics Letters, 118(3), 2021.
  36. [36]Kai-Uwe Demasius, Timothy Phung, Weifeng Zhang, Brian P Hughes, See-Hun Yang, Andrew Kellock, Wei Han, Aakash Pushp, and Stuart SP Parkin. Enhanced spin–orbit torques by oxygen incorporation in tungsten films. Nature communications, 7(1):10644, 2016.
  37. [37]Tian-Yue Chen, Chun-Te Wu, Hung-Wei Yen, and Chi-Feng Pai. Tunable spin-orbit torque in cu-ta binary alloy heterostructures. Physical Review B, 96(10):104434, 2017.DOI
  38. [38]Peng Wang, Andrea Migliorini, See-Hun Yang, Jae-Chun Jeon, Ilya Kostanovsky, Holger Meyerheim, Hyeon Han, Hakan Deniz, and Stuart SP Parkin. Giant spin hall effect and spin–orbit torques in 5d transition metal–aluminum alloys from extrinsic scattering. Advanced Materials, 34(23):2109406, 2022.
  39. [39]Rajagopalan Ramaswamy, Yi Wang, Mehrdad Elyasi, M Motapothula, T Venkatesan, Xuepeng Qiu, and Hyunsoo Yang. Extrinsic spin hall effect in cu 1- x pt x. Physical Review Applied, 8(2):024034, 2017.DOI
  40. [40]X Zhou, M Tang, XL Fan, XP Qiu, and SM Zhou. Disentanglement of bulk and interfacial spin hall effect in ferromagnet/normal metal interface. Phys. Rev. B, 94(14):144427, 2016.DOI
  41. [41]Caiyun Hong, Lichuan Jin, Huaiwu Zhang, Mingming Li, Yiheng Rao, Bo Ma, Jialu Li, Zhiyong Zhong, and Qinghui Yang. Giant inverse spin hall effect in bi doped ptbi alloy. Advanced Electronic Materials, 4(8):1700632, 2018.DOI
  42. [42]Nilamani Behera, Himanshu Fulara, Lakhan Bainsla, Akash Kumar, Mohammad Za-hedinejad, Afshin Houshang, and Johan Åkerman. Energy-efficient W₁₀₀₋ₓTaₓ/Co-Fe-B/MgO spin Hall nano-oscillators. Phys. Rev. Appl., 18(2):024017, 2022.DOI
  43. [43]Zhendong Chi, Yong-Chang Lau, Masashi Kawaguchi, and Masamitsu Hayashi. Charge–spin conversion in ptl- xbix alloys for spin–orbit torque switching. APL Materials, 9(6), 2021.
  44. [44]Tristan Joachim Winkel, Tahereh Sadat Parvini, Finn-Frederik Stiewe, Jakob Walowski, Farshad Moradi, and Markus Münzenberg. Comparative analysis of thz signal emission from bilayer heterostructures: Wideband and high-frequency thz signal advantage of ptbi-based emitter. Applied Physics Letters, 124(1), 2024.DOI
  45. [45]Ya Feng, Qi Jiang, Baolje Feng, Meng Yang, Tao Xu, Wenjing Liu, Xiufu Yang, Masashi Arita, Eike F Schwier, Kenya Shimada, et al. Rashba-like spin splitting along three momentum directions in trigonal layered ptbi2. Nat. Commun., 10(1):4765, 2019.
  46. [46]Sebastian Schimmel, Yanina Fasano, Sven Hoffmann, Julia Besproswanny, Laura Teresa Corredor Bohorquez, Joaquín Puig, Bat-Chen Elshalem, Beena Kalisky, Grigory Shipunov, Danny Baumann, et al. Surface superconductivity in the topological weyl semimetal t-ptbi2. Nat. Commun., 15(1):9895, 2024.DOI
  47. [47]Yong-Chang Lau, Takeshi Seki, and Koki Takanashi. Highly fcc-textured pt–al alloy films grown on mgo (001) showing enhanced spin hall efficiency. APL Materials, 9(8), 2021.DOI
  48. [48]Utkarsh Shashank, Yu Kusaba, Junnosuke Nakamura, Arun Jacob Mathew, Koki Imai, Shinya Senba, Hironori Asada, and Yasuhiro Fukuma. Charge–spin interconversion in nitrogen sputtered pt via extrinsic spin hall effect. Journal of Physics: Condensed Matter, 36(32):325802, 2024.DOI
  49. [49]A Hecq, JP Delrue, M Hecq, and T Robert. Sputtering deposition, xps and x-ray diffraction characterization of hard nitrogen-platinum thin films. Journal of Materials Science, 16:407–412, 1981.DOI
  50. [50]WL Peng, JY Zhang, GN Feng, XL Xu, C Yang, YL Jia, and GH Yu. Tunable damping-like and field-like spin-orbit-torque in pt/co/hfo2 films via interfacial charge transfer. Applied Physics Letters, 115(17), 2019.DOI
  51. [51]Xin Fan, Halise Celik, Jun Wu, Chaoying Ni, Kyung-Jin Lee, Virginia O Lorenz, and John Q Xiao. Quantifying interface and bulk contributions to spin-orbit torque in magnetic bilayers. Nature communications, 5(1):3042, 2014.DOI
  52. [52]Tianxiang Nan, Satoru Emori, Carl T Boone, Xinjun Wang, Trevor M Oxholm, John G Jones, Brandon M Howe, Gail J Brown, and Nian X Sun. Comparison of spin-orbit torques and spin pumping across nife/pt and nife/cu/pt interfaces. Physical Review B, 91(21):214416, 2015.DOI
  53. [53]Michael Harder, Yongsheng Gui, and Can-Ming Hu. Electrical detection of magnetization dynamics via spin rectification effects. Physics Reports, 661:1–59, 2016.DOI
  54. [54]John Rex Mohan, Utkarsh Shashank, Angshuman Deka, Takayasu Hanashima, Rohit Medwal, Surbhi Gupta, Rajdeep Singh Rawat, Hironori Asada, and Yasuhiro Fukuma. Observation of out-of-plane antidamping torque at the platinum/permalloy interface. ACS Applied Materials & Interfaces, 2025.DOI
  55. [55]Lakhan Bainsla, Akash Kumar, Ahmad A Awad, Chunlei Wang, Mohammad Zahedinejad, Nilamani Behera, Himanshu Fulara, Roman Khymyn, Afshin Houshang, Jonas Weissenrieder, et al. Ultrathin ferrimagnetic gdfeо films with low damping. Advanced Functional Materials, 32(23):2111693, 2022.DOI
  56. [56]Akash Kumar, Rajin Bansal, Sujeet Chaudhary, and Pranaba Kishor Muduli. Large spin current generation by the spin hall effect in mixed crystalline phase ta thin films. Phys. Rev. B, 98(10):104403, 2018.DOI
  57. [57]Yi Wang, Rajagopalan Ramaswamy, and Hyunsoo Yang. Fmr-related phenomena in spintronic devices. Journal of Physics D: Applied Physics, 51(27):273002, 2018.DOI
  58. [58]Kouta Kondou, Hua Chen, Takahiro Tomita, Muhammad Ikhlas, Tomoya Higo, Allan H MacDonald, Satoru Nakatsuji, and YoshiChika Otani. Giant field-like torque by the out-of-plane magnetic spin hall effect in a topological antiferromagnet. Nature communications, 12(1):6491, 2021.DOI
  59. [59]C. Kittel. On the theory of ferromagnetic resonance absorption. Phys. Rev., 73:155, Jan 1948.DOI
  60. [60]Mykola Dvornik, Ahmad A Awad, and Johan Åkerman. Origin of magnetization auto-oscillations in constriction-based spin hall nano-oscillators. Physical Review Applied, 9(1):014017, 2018.DOI
  61. [61]Edurne Sagasta, Yasutomo Omori, Miren Isasa, Martin Gradhånd, Luis E Hueso, Yasuhiro Niimi, YoshiChika Otani, and Félix Casanova. Tuning the spin hall effect of pt from the moderately dirty to the superclean regime. Physical Review B, 94(6):060412, 2016.DOI
  62. [62]Qiang Hao and Gang Xiao. Giant spin hall effect and magnetotransport in a ta/cofeb/mgo layered structure: A temperature dependence study. Physical Review B, 91(22):224413, 2015.DOI
  63. [63]Jairo Sinova, Sergio O. Valenzuela, J. Wunderlich, C. H. Back, and T. Jungwirth. Spin Hall effects. Rev. Mod. Phys., 87(4):1213–1260, October 2015.DOI
  64. [64]R Asomoza, A Fert, and R Reich. Gadolinium-heavy rare earth alloys: preparation, metallographic study and extraordinary hall effect. Journal of the Less Common Metals, 90(2):177–201, 1983.DOI
  65. [65]Akash Kumar, Mona Rajabali, Victor Hugo González, Mohammad Zahedinejad, Afshin Houshang, and Johan Åkerman. Fabrication of voltage-gated spin Hall nano-oscillators. Nanoscale, 14:1432–1439, 2022.DOI

Paper details

vlnc:W_9pwyk9fe
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents