betaFeedback

WorksPhysics and Astronomy

Energy-efficient field-free unconventional spin-orbit torque magnetization switching dynamics in van der Waals heterostructures

Abstract

Van der Waals (vdW) heterostructure of two-dimensional (2D) quantum materials offers a promising platform for efficient control of magnetization dynamics for non-volatile spin-based devices. However, energy-efficient field-free spin-orbit torque (SOT) switching and spin dynamics experiments to understand the basic SOT phenomena in all-2D vdW heterostructures are so far lacking. Here, we demonstrate energy-efficient field-free spin-orbit torque (SOT) switching and tunable magnetization dynamics in a vdW heterostructure comprising out-of-plane magnet Fe3GaTe2 and topological Weyl semimetal TaIrTe4. We measured the non-linear second harmonic Hall signal in TaIrTe4 /Fe3GaTe2 devices to evaluate the SOT-induced magnetization dynamics, which is characterized by a large and tunable out-of-plane damping-like torque. Energy-efficient and deterministic field-free SOT magnetization switching is achieved at room temperature with a very low current density. First-principles calculations unveil the origin of the unconventional charge-spin conversion phenomena, considering the crystal symmetry and electronic structure of TaIrTe4. These results establish that all-vdW heterostructures provide a promising route to energy-efficient, field-free, and tunable SOT-based spintronic devices.

Corresponding author: Saroj P. Dash, Email: [email protected]

Introduction

In quantum materials, the interplay between spin-orbit coupling and magnetism, with additional control over the band topology, quantum geometries, and crystal symmetries can offer the potential for next-generation universal memory and computing technologies[1, 2]. Specifically, enhanced functionalities can be achieved using efficient charge-spin conversion (CSC) phenomena in such quantum materials to enable spin-orbit torque (SOT) induced magnetization switching of a ferromagnet (FM)[3]. In conventional SOT memory devices, commonly used spin-orbit materials (SOM) exhibit moderate CSC efficiency and primarily generate in-plane SOT torque components, limiting their application in switching a magnet with perpendicular magnetic anisotropy (PMA)[4].

Recently developed van der Waals (vdW) heterostructures of two-dimensional (2D) SOMs and FMs can offer a new framework to address the challenges in SOT technologies[5]. Interestingly, low crystal symmetries of vdW SOMs can generate out-of-plane SOT components, making them suitable for field-free switching of ferromagnets with PMA[6, 7, 8, 9]. Meanwhile, vdW magnets such as Fe3GeTe2\mathrm{Fe}_3\mathrm{GeTe}_2 and Fe3GaTe2\mathrm{Fe}_3\mathrm{GaTe}_2 with strong PMA are also developed, showing promises for reliable SOT device operations[10, 11, 12, 13]. Taking advantage of such available quantum materials, all-2D vdW heterostructures have been explored for field-free SOT magnetization switching[14, 15, 16, 17]. However, the SOT switching parameters are two to three orders of magnitude lower than required for energy-efficient switching.

To circumvent this issue, Weyl semimetals TaIrTe4\mathrm{TaIrTe}_4 with low crystal symmetry, large spin-orbit coupling (SOC), and large Berry curvature dipole were explored to generate a larger out-of-plane SOT component for energy-efficiency and field-free SOT switching of conventional magnets[8, 9, 18]. Therefore, all-2D vdW heterostructure combining the best vdW quantum materials with a large current-induced out-of-plane spin polarization and above room temperature vdW ferromagnet with an out-of-plane magnetization is a promising for energy-efficient non-volatile spintronic technologies. Furthermore, the investigation of magnetization dynamics in all-2D vdW heterostructures is critical for understanding the interplay between broken crystal symmetries, unconventional charge-spin conversion, and SOT-induced magnetization dynamics, ultimately enabling the design of efficient and ultrafast spintronic devices.

Here, we report tunable magnetization dynamics using harmonic measurements and demonstrate energy-efficient field-free SOT magnetization switching using the all-vdW heterostructures of TaIrTe4/Fe3GaTe2\mathrm{TaIrTe}_4/\mathrm{Fe}_3\mathrm{GaTe}_2 at room temperature. Weyl semimetal TaIrTe4\mathrm{TaIrTe}_4 with a tunable canted spin polarization combined with a vdW ferromagnet Fe3GaTe2\mathrm{Fe}_3\mathrm{GaTe}_2 with strong PMA enable the exploration of magnetization dynamics and their tunable SOT efficiency. The 2nd2^{\mathrm{nd}} harmonic measurements with detailed magnetic field and angle-dependent measurements at various temperatures reveal a large and tunable unconventional out-of-plane SOT torque in the TaIrTe4/Fe3GaTe2\mathrm{TaIrTe}_4/\mathrm{Fe}_3\mathrm{GaTe}_2 all-vdW heterostructure. The SOT components are observed to vary with temperature and correlate with the measured spin canting angle. Moreover, we observed a field-free deterministic SOT magnetization switching with a very low critical switching current density of 1.81×1010 A/m21.81 \times10^{10}\,\mathrm{A}/\mathrm{m}^{2}, demonstrating energy-efficient nonvolatile spintronic memory device. To unveil the origin of the unconventional CSC phenomena in TaIrTe4\mathrm{TaIrTe}_{4}, detailed first-principles calculations were performed considering crystal symmetry and electronic structures.

Results and discussion

We investigated TaIrTe4/Fe3GaTe2\mathrm{TaIrTe}_{4}/\mathrm{Fe}_{3}\mathrm{GaTe}_{2} vdW heterostructures (Figure 1a) due to their promising properties, anticipating that their combination could yield new phenomena such as large non-linear Hall effects and unconventional spin-orbit torque (SOT) magnetization dynamics. TaIrTe4\mathrm{TaIrTe}_{4} is a vdW topological Weyl semimetal (WSM) candidate, with a significant Berry curvature dipole and large spin splitting of the electronic bands[19]. In addition, it provides unconventional charge-spin conversion with an out-of-plane spin polarization component that can induce an out-of-plane SOT[18] on the adjacent PMA ferromagnet to induce a magnetic field-free switching. On the other hand, Fe3GaTe2\mathrm{Fe}_{3}\mathrm{GaTe}_{2} is a unique vdW topological nodal line metallic ferromagnet with strong PMA above room temperature with TcT_{\mathrm{c}} around 370 K[10]. We fabricated Hall-bar devices based on TaIrTe4/Fe3GaTe2\mathrm{TaIrTe}_{4}/\mathrm{Fe}_{3}\mathrm{GaTe}_{2} vdW heterostructures, along with individual Hall bars on TaIrTe4\mathrm{TaIrTe}_{4} and Fe3GaTe2\mathrm{Fe}_{3}\mathrm{GaTe}_{2} crystals, to characterize properties, such as the anomalous Hall effect (AHE), 2{sup}nd Harmonics measurements and SOT-driven switching experiments (details in Methods section and {xref}Supplementary Fig. S1`). Figure 1b presents a typical optical microscope image of a representative TaIrTe4_4/Fe3_3GaTe2_2 Hall-bar device.

Van der Waals heterostructure of TaIrTe4/Fe3GaTe2 and harmonic measurements on TaIrTe4

Figure 1. a. Schematic diagram of a van der Waals heterostructure of Weyl semimetal TaIrTe4\mathrm{TaIrTe}_{4} and out-of-plane ferromagnet Fe3GaTe2\mathrm{Fe}_{3}\mathrm{GaTe}_{2}. Band structure of typical type-II Weyl semimetal with two Weyl nodes. b. Optical image of representative TaIrTe4/Fe3GaTe2\mathrm{TaIrTe}_{4}/\mathrm{Fe}_{3}\mathrm{GaTe}_{2} vdW heterostructure Hall bar device with a scale bar of 5 μ\mum. c. 2nd harmonic transverse Hall voltage Vxy2ωV_{xy}^{2\omega} in response to an applied alternating current IωI^{\omega} along a-axis at different temperatures for a device with 20 nm TaIrTe4\mathrm{TaIrTe}_{4}. The inset illustrates the crystal structure of TdT_{\mathrm{d}}-TaIrTe4\mathrm{TaIrTe}_{4}, characterized by low crystal symmetry and a mirror plane along the crystallographic b-axis. d. 2nd harmonic transverse Hall voltage Vxy2ωV_{xy}^{2\omega} with temperature at an IωI^{\omega} of 0.1 mA of TaIrTe4\mathrm{TaIrTe}_{4}. Insets show the energy dispersion curve of type-II Weyl semimetal and tuning of Fermi level energy (EFE_{\mathrm{F}}) with temperature. e. 2nd harmonic voltage V2ωV^{2\omega} response measured in TaIrTe4\mathrm{TaIrTe}_{4} device as a function of angle between current applied along a-axis of TaIrTe4\mathrm{TaIrTe}_{4} (∣Iω∣=0.1|I^{\omega}| = 0.1 mA) and external magnetic field (13 T). The device is rotated in XY and ZY planes, as depicted in schematics. In the XY rotation, the device rotates such that the magnetic field aligns parallel to the sample surface and making ΦB\Phi_{\mathrm{B}} angle with a-axis of TaIrTe4\mathrm{TaIrTe}_{4}, whereas in ZY rotation, the device rotation changes magnetic field direction from a-axis of TaIrTe4\mathrm{TaIrTe}_{4} to c axis and making θB\theta_{\mathrm{B}} angle with c-axis with TaIrTe4\mathrm{TaIrTe}_{4}. The solid lines are the fits. f. Temperature dependence of shift (Δ\Delta) in the maxima or minima of V2ωV^{2\omega} vs ΦB\Phi_{\mathrm{B}} and θB\theta_{\mathrm{B}} curves. This shift is denoted as out-of-plane canting angles, as illustrated in schematics. Such shift is directly correlated to the out-of-plane spin canting angle, which is estimated to be −(27±0.76)∘-(27 \pm0.76)^\circ at room temperature.

Tunable spin texture using bilinear magnetoresistance and 2{sup}nd` harmonic Hall effect in TaIrTe4_4

TaIrTe4_4 exhibited a strong nonlinear Hall effect, characterized by a 2{sup}nd harmonic Hall voltage that nonlinearly depends on driving currents sourced along the a-axis of the crystal, perpendicular to its mirror plane at room temperature ({xref}Fig. 1c). Unlike linear Hall effects observed in systems with broken time-reversal symmetry, the nonlinear Hall effect in TaIrTe$_4$ arises from the large Berry curvature dipole in the absence of inversion-symmetry. Notably, the nonlinear Hall voltage changed sign near ∼200 K ({xref}Fig. 1c and {xref}1d), indicating temperature-induced shift in the chemical potential, consistent with the Weyl semi metallic properties of TaIrTe$_4${citep}20. The current induced spin polarization in TaIrTe$_4$ are probed using bilinear magnetoelectric resistance (BMER) technique{citep}9,21, measuring 2{sup}nd harmonic voltage while rotating the samples in XY and ZY planes (Figure 1e). In XY rotation, the magnetic field vector remains in the ab crystallographic plane sweeping azimuthal angle (ΦB\Phi_{\mathrm{B}}) with respect to the a-axis of TaIrTe4_4, whereas in ZY rotation, the field vector sweeps polar angle (θB\theta_{\mathrm{B}}) with respect to the c-axis of TaIrTe4_4 in the ac plane. Figure 1e depicts the temperature dependence of 2{sup}nd` harmonic voltage with ΦB\Phi_{\mathrm{B}} and θB\theta_{\mathrm{B}}.

The direction of resultant spin angular momentum arises due to charge-spin conversion effects in TaIrTe4_4 being equivalent to angular shift (Δ\Delta) of BMER curves measured along XY and ZY geometries. The Δ\Delta is found to be −(27 ± 0.76)° at room temperature, indicating the presence of an out-of-plane spin density induced in TaIrTe4_4. Such spin-polarization can help in generating unconventional out-of-plane SOT in adjacent ferromagnetic layer Fe3_3GaTe2_2 with PMA resulting in field-free deterministic switching. The temperature dependence of Δ\Delta (Figure 1f) suggests that the polarity and magnitude of the spin canting angle in TaIrTe4_4 are highly tunable by the position of chemical potential/Fermi level[9].

Perpendicular magnetic anisotropy of Fe3_3GaTe2_2

To verify the magnetic property and anisotropy of Fe3_3GaTe2_2, the anomalous Hall resistance RxyR_{xy} is measured at different temperatures ranging from 2 to 300 K (Figure 2a, Figure 2b). A square-shaped magnetic hysteresis loop is observed with coercivity around 100 mT and anomalous Hall resistance (RAHER_{\mathrm{AHE}}) of around 1.5 Ω at room temperature, where the latter is directly proportional to saturation magnetization (MsM_{\mathrm{s}}) of Fe3_3GaTe2_2. The RAHER_{\mathrm{AHE}} vs T curve, shown in Figure 2c, is fitted with

Magneto-transport characterization of Fe₃GaTe₂

Figure 2. Magneto-transport characterization of Fe₃GaTe₂. a,b, Anomalous Hall resistance of Fe₃GaTe₂ as a function of out-of-plane magnetic fields at 300 K and temperature dependence ranging from 2 to 300 K. c. Anomalous Hall amplitude at the saturated field as a function of temperature. Solid line is fit to extract the Curie temperature (Tc=369.14±7.73T_{\mathrm{c}} = 369.14 \pm7.73 K) d. Anomalous Hall resistance of Fe₃GaTe₂ as a function of in-plane magnetic fields at different temperatures ranging from 2 to 300 K. e. Comparison of anomalous Hall effect measurement for field swept parallel to sample plane (i.e., H⊥cH \perp c) vs perpendicular (i.e., H∥cH \parallel c) to sample plane at 2 K temperature. The anisotropic field (HKH_{\mathrm{K}}) is ~7.9 T, indicating strong perpendicular magnetic anisotropy present in Fe₃GaTe₂. f. Variation of coercive fields and anisotropic fields with temperature extracted from (RxyR_{xy} vs μ0H⊥\mu_{0}H_{\perp}) and (RxyR_{xy} vs μ0H∥\mu_{0}H_{\parallel}) measurements. g. AHE signals RxyR_{xy} with different out-of-plane angles (θ\theta) between the magnetic field and the c-axis of the sample plane at 300 K. h. Variation of AHE signals RxyR_{xy} with positive and negative DC bias currents.

Rxy(T)=Rxy(0)(1−(TTc)2)βR_{xy}(T)=R_{xy}(0)\left(1-\left(\frac{T}{T_{\mathrm{c}}}\right)^2\right)^\beta

analogues to Bloch equation for magnetization vs temperature curve to estimate Curie temperature (Tc=369.14±7.73T_{\mathrm{c}} = 369.14 \pm7.73 K)[10] and critical magnetization exponent β=0.35\beta= 0.35[22, 23, 24, 25]. Figure 2d shows the anomalous Hall resistance of Fe3_3GaTe2_2 as a function of in-plane magnetic fields at different temperatures from 2 to 300 K. A magnetic hysteresis loop is observed at all temperatures, with finite remanence and coercivity, consistent with the typical behavior of PMA magnets along their hard axis (Figure 2e).

Figure 2f shows the variation of magnetic coercivity (HcH_{\mathrm{c}}) in both field directions (i.e., H⊥cH \perp c-axis and H∥cH \parallel c-axis) and anisotropic field with temperature. The anisotropic field (HKH_{\mathrm{K}}), defined as the difference in saturation between in-plane and out-of-plane magnetic fields, reaches ~7.9 T at 2 K and ~3.8 T at 300 K. Such a high value of HKH_{\mathrm{K}} suggests that Fe₃GaTe₂ has a very high magnetic anisotropy energy density with a very strong PMA. The coercive field (HcH_{\mathrm{c}}) is also quite high along in-plane direction as compared to out-of-plane direction. Both the HcH_{\mathrm{c}} and HKH_{\mathrm{K}} decrease with an increase in temperature approaching the Curie temperature of Fe₃GaTe₂. Figure 2g illustrates AHE signals RxyR_{xy} measured at varying out-of-plane angles (θ\theta) between c-axis of sample and magnetic field. It can be noted here that the magnitude of AHE signal (RxyAHE=Rxy(+HS)−Rxy(−HS)2R_{xy}^{\mathrm{AHE}} = \frac{R_{xy}(+H_{\mathrm{S}})-R_{xy}(-H_{\mathrm{S}})}{2}) remains almost constant till ±80∘\pm80^\circ; beyond that AHE loop disappears between ±600\pm600 mT field range. Again, this indicates a strong out-of-plane magnetic anisotropy present in Fe₃GaTe₂. Figure 2h shows the variation of AHE signals RxyR_{xy} with positive and negative DC bias currents. We observed that the magnitude of anomalous Hall signal, the coercivity and saturation fields remain unchanged with positive or negative current bias varied to ±0.1\pm0.1 mA to ±1\pm1 mA, indicating the robustness of perpendicular anisotropic magnetic moment against dc current within these bias ranges.

2nd harmonic nonlinear Hall effect and spin-orbit torque induced magnetization dynamics in TaIrTe4_4/Fe3_3GaTe2_2 heterostructures

The harmonic Hall measurements are performed on TaIrTe4_4/Fe3_3GaTe2_2 heterostructures to quantitatively evaluate the non-linear effects and magnetization dynamics driven by SOT. When a sinusoidal current (IωI^{\omega}) is applied to the vdW heterostructure, composed of the spin-orbit material TaIrTe4_4 and a ferromagnet Fe3_3GaTe2_2, spin-orbit torques (τSOT\tau_{SOT}) are exerted on the magnetization (m\mathbf{m}) of the Fe3_3GaTe2_2. This effect originates from the spin accumulation at the vdW interface due to efficient charge-spin conversion in TaIrTe4_4. Typically, two mutually orthogonal torques are generated: the damping-like torque (τDL∼m×(σ×m)\tau_{DL} \sim\mathbf{m} \times(\boldsymbol{\sigma} \times\mathbf{m})) and the field-like torque (τFL∼σ×m\tau_{FL} \sim\boldsymbol{\sigma} \times\mathbf{m})[11, 26].

In these measurements, applying a sinusoidal current (IωI^\omega) with a fixed frequency of 213.3 Hz induces SOT-driven magnetization oscillation, generating harmonics in both the longitudinal and transverse resistance signals. The 1st1^{\mathrm{st}} and 2nd2^{\mathrm{nd}} harmonic signals are measured and analyzed across various angles (ΦB\Phi_B) between the in-plane magnetic field (H⊥cH_{\perp c}) and the applied sinusoidal current (IωI^\omega), as well as under varying external magnetic fields (HextH_{\mathrm{ext}}). This analysis provides information about the current-induced effective SOT fields and torques.

Since the spin Hall effect (SHE) in TaIrTe4_4 induces both in-plane and out-of-plane spin polarizations (σX,Y,Z\sigma^{X,Y,Z}), the applied IωI^\omega along the a-axis of TaIrTe4_4 generates corresponding components of the damping-like (τDLX,Y,Z\tau^{X,Y,Z}_{DL}) and field-like (τFLX,Y,Z\tau^{X,Y,Z}_{FL}) torques. The 2nd2^{\mathrm{nd}} harmonic transverse voltage generated from these current-induced effective SOT fields (HDLX,Y,ZH^{X,Y,Z}_{DL}, HFLX,Y,ZH^{X,Y,Z}_{FL}) and torques (τDLX,Y,Z\tau^{X,Y,Z}_{DL}, τFLX,Y,Z\tau^{X,Y,Z}_{FL}) in PMA ferromagnets is expressed as[27, 28],

Vxy2ω=VDLYcos⁡ΦB+VDLXsin⁡ΦB+VDLZcos⁡2ΦB+VFLYcos⁡ΦBcos⁡2ΦB+VFLXsin⁡ΦBcos⁡2ΦB+VFLZ(1)V_{xy}^{2\omega}=V_{DL}^{Y}\cos\Phi_B+V_{DL}^{X}\sin\Phi_B+V_{DL}^{Z}\cos2\Phi_B+V_{FL}^{Y}\cos\Phi_B\cos2\Phi_B+V_{FL}^{X}\sin\Phi_B\cos2\Phi_B+V_{FL}^{Z} \tag*{(1)}

Here, damping-like torque components generated by X, Y and Z spin polarization contribute to coefficients VDLX,Y,ZV_{DL}^{X,Y,Z}, and the field-like torque counterparts give rise to coefficients VFLX,Y,ZV_{FL}^{X,Y,Z}. The estimation of SOT fields (HDLX,Y,ZH_{DL}^{X,Y,Z}, HFLX,Y,ZH_{FL}^{X,Y,Z}) from coefficients (VDLX,Y,ZV_{DL}^{X,Y,Z}, VFLX,Y,ZV_{FL}^{X,Y,Z}) is detailed in Supplementary Note 6 (Eqs. S1-S7).

The 2nd2^{\mathrm{nd}} harmonic Hall signal as a function of ΦB\Phi_B at constant fields (H>>HK_K) is plotted in Figure 3b. The Vxy2ωV_{xy}^{2\omega} vs ΦB\Phi_B curve is fitted with Eq. 2 (Vxy2ω=VDLcos⁡ΦB+VFLcos⁡ΦBcos⁡2ΦBV_{xy}^{2\omega}=V_{DL}\cos\Phi_B+V_{FL}\cos\Phi_B\cos2\Phi_B) to estimate SOT in materials containing only conventional in-plane spins[29, 30]. However, Vxy2ωV_{xy}^{2\omega} vs ΦB\Phi_B curve of TaIrTe4_4/Fe3_3GaTe2_2 could not be well fitted with Eq. 2 (see Figure 3b). For proper fitting, we need to include VDLcos⁡2ΦBV_{DL}\cos2\Phi_B and VDLsin⁡ΦBV_{DL}\sin\Phi_B terms as in Eq. 1, which consider additional torque components due to current-induced out-of-plane spin canting in TaIrTe4_4. The coefficients Vxy,cos⁡ΦB2ωV_{xy,\cos\Phi_B}^{2\omega}, Vxy,sin⁡ΦB2ωV_{xy,\sin\Phi_B}^{2\omega} and Vxy,cos⁡2ΦB2ωV_{xy,\cos2\Phi_B}^{2\omega} are hyperbolic functions of the magnetic field (see Supplementary Note 6, Eqs S2-S4), implying linear function on 1/(H-HK_K) or 1/H. The values of these coefficients were extracted by fitting the experimental 2nd2^{\mathrm{nd}} harmonic transverse voltage and plotted in Figure 3c-Figure 3e. From these slopes, HDLX,Y,ZH_{DL}^{X,Y,Z} are estimated (using RAHE_{AHE}=1 Ω and RPHE_{PHE}=0.0113 Ω; see Supplementary Note 5) and plotted as a function of current densities Ja.c._{a.c.} in Figure 3g. The slope of HDLX,Y,ZH_{DL}^{X,Y,Z} vs Ja.c._{a.c.} are found out to be HDLXH_{DL}^{X}/J∼\sim (3.09 ±\pm 0.37) ×\times 10−12^{-12} T A−1^{-1}m2^2, HDLYH_{DL}^{Y}/J∼\sim (2.43 ±\pm 0.15) ×\times 10−12^{-12} T A−1^{-1}m2^2 and HDLZH_{DL}^{Z}/J∼\sim (6.78 ±\pm 0.44) ×\times 10−12^{-12} T A−1^{-1}m2^2. This analysis indicates that current-induced effective SOT fields or torques in TaIrTe4_4/Fe3_3GaTe2_2 heterostructure originated from the out-of-plane spin polarization (σZ\sigma^Z), and it is larger than its in-plane counterparts (σXY\sigma^{XY}).

Angle-dependent harmonic Hall measurements in TaIrTe4/Fe3GaTe2 heterostructure

Figure 3. Angle-dependent harmonic Hall measurements in TaIrTe4_4/Fe3_3GaTe2_2 heterostructure. a. Schematic of the TaIrTe4_4/Fe3_3GaTe2_2 heterostructure, illustrating the effects of damping-like torques (τDLXY\tau_{DL}^{XY} and τDLZ\tau_{DL}^{Z}) and field-like torques (τFL\tau_{FL}) on Fe3_3GaTe2_2 magnetization when the current is applied along the a-axis of TaIrTe4_4 layer. The 2nd harmonics Hall voltage (Vxy2ωV_{xy}^{2\omega}) measurement scheme is shown with an external in-plane magnetic field at angle ΦB\Phi_B relative to the a.c. current direction IacI_{ac}. b. Vxy2ωV_{xy}^{2\omega} vs ΦB\Phi_B of Dev1 at magnetic field 7 T and temperature 300 K. The solid lines are fitted with Eq. 1 and 2. The second panel shows Vxy2ωV_{xy}^{2\omega} vs ΦB\Phi_B for varied magnetic fields (7–12 T). c,d,e. Coefficient Vxy,cos⁡ΦB2ωV_{xy,\cos\Phi_B}^{2\omega} (cos⁡ΦB\cos\Phi_B dependent in Vxy2ωV_{xy}^{2\omega}), Vxy,sin⁡ΦB2ωV_{xy,\sin\Phi_B}^{2\omega} (sin⁡ΦB\sin\Phi_B dependent in Vxy2ωV_{xy}^{2\omega}) and Vxy,cos⁡2ΦB2ωV_{xy,\cos2\Phi_B}^{2\omega} (cos⁡2ΦB\cos2\Phi_B dependent in Vxy2ωV_{xy}^{2\omega}) as a function of 1/(H−Hk)1/(H-H_k) and 1/H1/H under different current densities Ja.c.J_{a.c.}. f. Angle sweep of Vxy2ωV_{xy}^{2\omega} at different temperatures (2–325 K) at a constant magnetic field of 10 T. Solid lines are fit to experimental data using equation 1. g. Damping-like field components (HDLXH_{DL}^{X}, HDLYH_{DL}^{Y}, HDLZH_{DL}^{Z}) as a function of current density, with linear fits estimating HDL/Ja.c.H_{DL}/J_{a.c.}. h. Temperature dependence of HDL/Ja.c.H_{DL}/J_{a.c.} for TaIrTe4_4/Fe3_3GaTe2_2 device. Insets show the energy dispersion curve of type-II Weyl semimetal and tuning of Fermi level energy (EFE_F) with temperature.

Tunable spin-orbit torque with temperature due to Fermi level tuning of TaIrTe4_4

The polarity and magnitude of spin accumulation generated by TaIrTe4_4 are influenced by the chemical potential[9], resulting in temperature dependence changes in the tilt angle (as shown in Figure 1f). A similar trend is expected in the current-induced effective SOT fields or torques. Hence, to observe the temperature dependence of SOT efficiency from the 2nd2^{\mathrm{nd}} harmonic Hall signal of TaIrTe4_4/Fe3_3GaTe2_2, angle sweep SHH measurements are conducted at different temperatures. Figure 3f illustrates the Vxy2ωV_{xy}^{2\omega} vs ΦB\Phi_B curve at 10 T across different temperatures.

Damping-like SOT effective fields (HDLX,Y,ZH_{DL}^{X,Y,Z}) are estimated using Eq. (1). HDLZ/Ja.c.H_{DL}^{Z}/J_{\mathrm{a.c.}} is highly tunable with temperature (Figure 3h), it decreased from 2 K to 100 K, reached a minimum between 100-200 K, and increased from 200 K to 325 K. This behavior aligns with the temperature dependence of current-induced spin accumulation (Figure 1f), showing large out-of-plane spin polarization at 2 K and room temperature with a minimum near 100 K. Hence, the out-of-plane damping-like torque is observed to be tunable by the chemical potential of TaIrTe4_4.

Field-dependent harmonic Hall measurements

To further validate and estimate SOT components, we measured the 1st^{\mathrm{st}} and 2nd^{\mathrm{nd}} harmonic transverse Hall resistance RxyR_{xy} signal as a function of magnetic field applied parallel to the sample surface (H⊥cH \perp c) and perpendicular to the applied current direction. In the 1st^{\mathrm{st}} harmonic RxyωR_{xy}^{\omega} vs H, a hysteresis loop with a magnetic anisotropic field HKH_K of ~ 1.5 T is observed (Figure 4a). The 2nd^{\mathrm{nd}} harmonics transverse Hall resistance signal Rxy2ωR_{xy}^{2\omega} varied with the external magnetic field applied parallel to the sample surface. The measurements are conducted with the field oriented either perpendicular (HyH_y, ΦB=90∘\Phi_B = 90^\circ) or parallel (HxH_x, ΦB=0∘\Phi_B = 0^\circ) to the direction of current (or a-axis of TaIrTe4_4). These results are displayed in Figure 4b and Figure 4c. The resistance exhibited a hyperbolic dependence on the field for ∣H∣>Hk|H|>H_k, however became discontinuous for ∣H∣<HK|H| < H_K.

Field-dependent harmonic Hall measurements in TaIrTe4/Fe3GaTe2 heterostructure

Figure 4. Field-dependent harmonic Hall measurements in TaIrTe4_4/Fe3_3GaTe2_2 heterostructure. a. 1st^{\mathrm{st}} harmonic transverse resistance (Rxy1ωR_{xy}^{1\omega}) as a function of magnetic field swept parallel to the sample surface (H⊥cH \perp c) and perpendicular to current direction, measured at 300 K on Dev 2. b,c, 2nd^{\mathrm{nd}} harmonic transverse resistance Rxy2ωR_{xy}^{2\omega} varied as a function of the external magnetic field applied along parallel to the sample surface, with HyH_y representing H⊥cH \perp c and perpendicular to the current (H⊥JH \perp J), and HxH_x representing H⊥cH \perp c and parallel to the current (H∥JH \parallel J). d,e, Dependence of the 2nd^{\mathrm{nd}} harmonic transverse resistance (Rxy2ωR_{xy}^{2\omega}) on the in-plane magnetic field (HyH_y and HxH_x) for different magnitudes of constant write current density (Ja.c.J_{\mathrm{a.c.}}). f. Extracted effective damping-like field components (HDLX,Y,ZH_{DL}^{X,Y,Z}) corresponding to spin polarization (σX,σY,σZ\sigma^{X},\sigma^{Y},\sigma^{Z}) as a function of Ja.c.J_{\mathrm{a.c.}}, obtained from fits to the 2nd^{\mathrm{nd}} harmonic signal.

Figure 4d and Figure 4e show 2{sup}nd harmonics transverse resistance (Rxy2ωR_{xy}^{2\omega}) versus HyH_y and HxH_x for different applied sinusoidal current densities (Ja.c.J_{a.c.}). The hyperbolic curvature of these plots sharpens with increasing current density. For Rxy2ωR_{xy}^{2\omega} vs HyH_y data at ΦB=90∘\Phi_B = 90^\circ, equation (1) reveals that only x and z components of the SOT fields contributing to the 2{sup}nd harmonics signal. Thus equation (1) simplifies to:

Rxy2ω=HDLXRAHE2(Hy−HK)+HDLZRPHEHy+HFLXRPHEHy+HFLZRAHE2(Hy−HK)(2)R_{xy}^{2\omega} = H_{DL}^{X}\frac{R_{AHE}}{2(H_y-H_K)} + H_{DL}^{Z}\frac{R_{PHE}}{H_y} + H_{FL}^{X}\frac{R_{PHE}}{H_y} + H_{FL}^{Z}\frac{R_{AHE}}{2(H_y-H_K)} \tag*{(2)}

From the analysis of second harmonic angular dependence data, we have concluded that HFLXH_{FL}^{X} and HFLZH_{FL}^{Z} make negligible contributions to the fitting of Vxy2ωV_{xy}^{2\omega} vs ΦB\Phi_B curve. This reduces the fitting parameter, allowing precise calculation of HDLXH_{DL}^{X} and HDLZH_{DL}^{Z}. Similarly, for Rxy2ωR_{xy}^{2\omega} vs HxH_x data (ΦB=0∘\Phi_B = 0^\circ), the extracted HDLZH_{DL}^{Z} values are used to determine HDLYH_{DL}^{Y} and HFLYH_{FL}^{Y} (see Eq. (1) and Supplementary Note 6). The extracted values of HDLX,Y,ZH_{DL}^{X,Y,Z} with different current densities Ja.c.J_{a.c.} are plotted in Figure 4f. The slopes of HDLH_{DL} vs Ja.c.J_{a.c.} are found to be: HDLX/Ja.c.∼(0.012±0.002)×10−12 TA−1m2H_{DL}^{X}/J_{a.c.} \sim(0.012 \pm0.002) \times10^{-12}\ \mathrm{T A^{-1}m^2}, HDLY/Ja.c.∼(0.013±0.0015)×10−12 TA−1m2H_{DL}^{Y}/J_{a.c.} \sim(0.013 \pm0.0015) \times10^{-12}\ \mathrm{T A^{-1}m^2} and HFLZ/Ja.c.∼(1.905±0.17)×10−12 TA−1m2H_{FL}^{Z}/J_{a.c.} \sim(1.905 \pm0.17) \times10^{-12}\ \mathrm{T A^{-1}m^2}. These findings also confirm that effective damping like field corresponding to Z spin polarization is significantly larger than those from XY polarized spins.

It should be noted that TaIrTe4_4 alone also exhibits a 2{sup}nd harmonic voltage signal as function of ΦB\Phi_B, arising from broken mirror symmetry and finite Berry curvature dipole. This signal follows a cos⁡ΦB\cos\Phi_B or sin⁡ΦB\sin\Phi_B dependence (Figure 1e). So, the HDLXH_{DL}^{X} and HDLYH_{DL}^{Y} values from the fitting of Vxy2ωV_{xy}^{2\omega} vs ΦB\Phi_B data can be overestimated (Eq. (1)). However, the estimation of Z-component damping-like field HFLZH_{FL}^{Z} using coefficient Vxy,cos⁡2ΦB2ωV_{xy,\cos2\Phi_B}^{2\omega} (Eq. (1)), central to the conclusion of second harmonic measurements, remains consistent. Furthermore, only TaIrTe4_4 also has a unique field dependence of 2{sup}nd harmonics signal as shown in Supplementary Fig. S5c, which is quite different from SOT-induced Rxy2ωR_{xy}^{2\omega} vs H curve (see Figure 4b-Figure 4e). At large magnetic field, the Rxy2ωR_{xy}^{2\omega} vs H curve from TaIrTe4_4 appears to be linear function of magnetic field; hence, to account for the 2{sup}nd harmonics field contribution of TaIrTe4_4 and thermal effects[30, 31], a linear polynomial term is included while fitting the field-dependent curves.

Field-free deterministic spin–orbit torque switching in TaIrTe4_4/Fe3_3GaTe2_2 heterostructure

SOT magnetization switching experiments are crucial for investigating magnetization switching characteristics, such as determining the critical switching current density, assessing the need for an external field to aid in switching, and identifying whether the process is deterministic or non-deterministic. A series of pulse currents (IpulseI_{\mathrm{pulse}}) applied along the a-axis in the TaIrTe4_4/Fe3_3GaTe2_2 heterostructure can induce an unconventional spin current along the z-axis, with spin polarization σZ\sigma^Z oriented along the z-axis in TaIrTe4_4[9]. This spin current generates an unconventional SOT on Fe3_3GaTe2_2, consisting of both field-like (τFL\tau_{\mathrm{FL}}) and damping-like (τDL\tau_{\mathrm{DL}}) torques, facilitating the switching of the magnetization direction M\mathbf{M}. The field-like torque τFL∼σ×m\tau_{\mathrm{FL}} \sim\boldsymbol{\sigma} \times\mathbf{m} induces the precession of M\mathbf{M} around the exchange field generated by spin polarization, while the damping-like torque τDL∼m×(σ×m)\tau_{\mathrm{DL}} \sim\mathbf{m} \times(\boldsymbol{\sigma} \times\mathbf{m}) aligns M\mathbf{M} with the spin polarization σ\boldsymbol{\sigma}, predominantly driving the magnetization switching Figure 4a[32]. Figure 4b shows the AHE at 300 K of Dev3 used for switching experiments. Figure 4c presents SOT-induced magnetization switching, measured by applying a pulsed write current (IpI_{\mathrm{p}}) along the a-axis with a pulse duration of 50 μ\mus. This is followed by a small DC read current (Ir∼500 μI_{\mathrm{r}} \sim500\,\muA) to determine the magnetization state via the Hall resistance Rxy=Vxy/IrR_{\mathrm{xy}}=V_{\mathrm{xy}}/I_{\mathrm{r}}. Due to a large unconventional SOT, fully deterministic field-free magnetization switching could be observed at room temperature with Ip=±3.5I_{\mathrm{p}}=\pm3.5 mA. Since the signal RxyR_{\mathrm{xy}} is proportional to the out-of-plane magnetization MzM_{\mathrm{z}}, the SOT RxyR_{\mathrm{xy}} signal indicates a current-induced magnetization change between +MM and -MM. Notably, deterministic SOT switching of TaIrTe4_4/Fe3_3GaTe2_2 heterostructure is observed at Hx=0H_{\mathrm{x}}=0 T, which indicates the creation of σspinZ\sigma_{\mathrm{spin}}^{Z} spin polarization in TaIrTe4_4 leading to an out-of-plane SOT component. The magnitude of the switching signal is comparable to the AHE signal magnitude with field sweep, showing a full magnetization switching.

We further investigated the impact of deterministic SOT switching on the external in-plane magnetic field parallel to the current direction (Figure 4d). The external in-plane magnetic field (HxH_x) can break the symmetry of deterministic SOT switching. As the strength of HxH_x increases, the switching mechanism transitions from being predominantly driven by the out-of-plane spin torque component (τDLz\tau_{DL}^{z}) to being influenced by the in-plane components (τDLx,y\tau_{DL}^{x,y}). We observed that a small positive HxH_x has minimal effect on the SOT switching signal, however, increasing HxH_x beyond 100 mT results in a noticeable reduction of the signal magnitude. Despite this reduction, the switching efficiency was maintained at 50%, demonstrating some robustness against the external magnetic field. In contrast, when HxH_x is applied in the negative direction, the switching efficiency drops significantly to about 50% even at -10 mT, and it nearly diminishes to ~10% at -300 mT. Interestingly, the switching polarity remains unchanged up to 100 mT, indicating the effectiveness of the out-of-plane spin polarization of TaIrTe4_4 in counteracting the external magnetic field[9, 14]. In conventional SOT, where magnetization switching is driven purely by in-plane spin current, the switching polarity typically reverses abruptly with HxH_x[22, 31]. However, this was not observed in our experiments, highlighting the larger contribution of τDLz\tau_{DL}^{z} from TaIrTe4_4 in the magnetization dynamics of Fe3_3GaTe2_2. In device 4 (data provided in Supplementary Fig. S2), we observed that the switching polarity remained unchanged even up to 200 mT when pulse current of ±4 mA was applied along the a-axis of TaIrTe4_4. However, it abruptly reversed when both the current and magnetic field of similar magnitude was applied along the b-axis of TaIrTe4_4.

Furthermore, to examine the presence of τDLz\tau_{DL}^{z} and calculate unconventional SOT driven switching efficiency, we have performed AHE loop shift measurement with bias current (see Supplementary Fig. S3)[37, 38]. The out-of-plane antidamping torque can shift the AHE hysteresis loop when a positive and negative dc bias current beyond a threshold value equivalent to switching current density is applied along the a-axis of TaIrTe4_4. Such AHE loop shift (HshiftH_{\mathrm{shift}}) is observed for compensating τDLz\tau_{DL}^{z} driven intrinsic damping in Fe3_3GaTe2_2[9, 14, 37]. The SOT efficiency (εSOT\varepsilon_{\mathrm{SOT}}) due to unconventional τDLz\tau_{DL}^{z} torque is defined by equation[38, 39, 40]

εSOT=2eMsηtFMℏHshiftJswitch(3)\varepsilon_{\mathrm{SOT}} = \frac{2eM_s\eta t_{\mathrm{FM}}}{\hbar}\frac{H_{\mathrm{shift}}}{J_{\mathrm{switch}}} \tag*{(3)}

In our device, the εSOT\varepsilon_{\mathrm{SOT}} is 1.76, with the HshiftH_{\mathrm{shift}} and JswitchJ_{\mathrm{switch}} calculated to be 2 mT and 1.81×1010 A m−21.81 \times10^{10}\,\mathrm{A\,m^{-2}}, respectively and MsM_s taken as 0.97×105 A m−10.97 \times10^{5}\,\mathrm{A\,m^{-1}} (see Supplementary Note 4). The switching efficiency parameter (η\eta), defined as the ratio of switching current-driven and magnetic field-driven AHE, is observed to be 1 (Figure 4c). Using the device parameters (εSOT=1.76\varepsilon_{\mathrm{SOT}} = 1.76 and charge conductivity of TaIrTe4_4 σc=9.4×105\sigma_c = 9.4 \times10^{5} S/m), we estimate the spin Hall conductivity in TaIrTe4_4/Fe3_3GaTe2_2 heterostructure to be σSH=ℏ/2e(εSOT⋅σc)=1.65×106 ℏ/2e (Ωm)−1\sigma_{\mathrm{SH}} = \hbar/2e(\varepsilon_{\mathrm{SOT}} \cdot\sigma_c) = 1.65 \times10^{6}\,\hbar/2e\,(\Omega\mathrm{m})^{-1}. By employing both SOT-induced magnetic switching and 2nd harmonic Hall measurements, we have established that the magnetization of Fe3_3GaTe2_2 in heterostructure with TaIrTe4_4 can be effectively manipulated with a switching current density of Jswitch∼1.81×1010J_{\mathrm{switch}} \sim1.81 \times10^{10} A/m2^2 and power density P (= Jswitch2/σcJ_{\mathrm{switch}}^2/\sigma_c) of 0.348×1015 W m−30.348 \times10^{15}\,\mathrm{W\,m^{-3}} at room temperature. The benchmarked of the spin Hall conductivity σSH\sigma_{\mathrm{SH}} and power density P of TaIrTe4_4/Fe3_3GaTe2_2 devices along with literature available on state-of-the-art SOT devices [8, 9, 14, 16, 17, 18, 30, 31, 40, 43, 45, 46, 47] are shown in Figure 4e and Supplementary Table 1.

Calculation of unconventional spin Hall effect in TaIrTe₄

Our experimental observations strongly suggest the presence of unconventional spin Hall effect in TaIrTe4\mathrm{TaIrTe}_{4}, which originates from the in-plane charge current and results in an out-of-plane spin-polarized spin current across the interface, corresponding the σZXZ\sigma^{Z}_{ZX} component of the spin Hall conductivity (SHC) tensor (IZSZ=σZXZIXCI^{S Z}_{Z} = \sigma^{Z}_{ZX} I^{C}_{X} where ISI^{S} and ICI^{C} are spin and charge currents, respectively). While similar effects were also found in another low-symmetry Weyl semimetal (TD_{\mathrm{D}}-WTe₂)[6, 7, 15, 16, 18], the symmetry constraints theoretically prohibit this configuration, and the experimental results have not been explained. Like WTe₂, TaIrTe4\mathrm{TaIrTe}_{4} has low crystal symmetry described by space group (SG) 31 (Pmn2₁), consisting of a mirror plane perpendicular to the a axis (See Figure 5a), as well as glide reflection and two-fold screw rotation, which prevents an unconventional SHC component[42].

Energy-efficient, field-free deterministic magnetization switching by spin-orbit torque in the TaIrTe4/Fe3GaTe2 heterostructure at room temperature

Figure 5. Energy-efficient, field-free deterministic magnetization switching by spin-orbit torque in the TaIrTe4_4/Fe3_3GaTe2_2 heterostructure at room temperature. a. Diagrammatic representation of TaIrTe4_4/Fe3_3GaTe2_2 heterostructure. This configuration leads to a significant out-of-plane antidamping torque (τADOOP\tau_{\mathrm{AD}}^{\mathrm{OOP}}), which is symmetric with respect to the current direction, facilitating field-free deterministic switching of the Fe3_3GaTe2_2 magnetization. b. AHE of the TaIrTe4_4/Fe3_3GaTe2_2 heterostructure device 3 with magnetic field sweep at 300 K. c. Field-free full deterministic switching achieved at 3.5 mA pulse current and 500 μ\muA current is used as reading current to measure magnetization states keeping external field zero at 300 K temperature. The current is applied along the symmetry axis (a-axis) of TaIrTe4_4. d. Current-driven magnetization switching of TaIrTe4_4/Fe3_3GaTe2_2 under different bias fields parallel to the sample surface and current (HxH_{\mathrm{x}}). e. The benchmark of SOT spin Hall conductivity vs. power consumption with state-of-the-art results[8, 9, 16, 27, 30, 33, 34, 35, 36]. Ellipse represents error and device to device variation in the calculated parameters.

To unveil the origin of the unconventional SHE, we have performed first-principles calculations of TaIrTe4\mathrm{TaIrTe}_{4} (see Methods for computational details). As shown in Figure 6, our results reveal a large spin splitting of bands near the Fermi level due to SOC, and the presence of seven spin Hall conductivity components: six conventional (σjki\sigma^{i}_{jk} with i≠j≠ki \ne j \ne k) and one unconventional component σZXZ\sigma^{Z}_{ZX}, with the latter showing a magnitude comparable to the conventional components. The unconventional SHE at the Fermi level reaches σSH=1.56×104 ℏ/2e (Ωm)−1\sigma_{\mathrm{SH}}=1.56 \times10^{4}\,\hbar/2e\,(\Omega\mathrm{m})^{-1}, in agreement with experimental values reported for TaIrTe4\mathrm{TaIrTe}_{4} (1.47–5.44)×104 ℏ/2e (Ωm)−1(1.47\text{--}5.44) \times10^{4}\,\hbar/2e\,(\Omega\mathrm{m})^{-1} [8, 9, 18], which vary depending on experimental conditions and sample characteristics. Although previous studies attributed its presence to the topological properties of the surface[8], our calculations show that a large unconventional SHE occurs even in the bulk.

Electronic structure and spin Hall conductivity calculations of TaIrTe₄

Figure 6. Electronic structure and spin Hall conductivity calculations of TaIrTe4\mathrm{TaIrTe}_{4}. a. Orthorhombic crystal unit cell of TaIrTe4\mathrm{TaIrTe}_{4}. b. Electronic structure of TaIrTe4\mathrm{TaIrTe}_{4} calculated via density functional theory along the high-symmetry lines in the Brillouin zone shown in c. d,e,f. Calculated intrinsic spin Hall conductivity, representing configurations of spin current with spin polarization along a-, b-, and c-axis, respectively. The magnitude of the unconventional spin Hall conductivity σzxz\sigma^{z}_{zx} at the Fermi level is determined mostly by the band marked as red in b.

We analyzed the crystal symmetry in more detail by directly applying symmetry operations, revealing that the relaxed structures exhibit a slight deviation from SG 31. This small structural distortion reduces the symmetry to either SG 6 (Pm) or SG 1 (P1), depending on the specified numerical precision (see Supplementary Note 7 for details). In both cases, the two-fold screw symmetry which normally prohibits unconventional SHE is absent, thus allowing for out-of-plane spin polarization of spin current. Structural distortion could further increase near the surface, potentially enhancing the generated spin accumulation. Therefore, from a symmetry perspective, the unconventional SHE component is justified, while its magnitude arises from the electronic properties, as discussed in Supplementary Materials.

Our experiments indicate even larger SHE values than previously reported and reveal an additional component, σZYZ\sigma^{Z}_{ZY}, induced by charge current along the mirror planes. This component was absent in the previous studies and does not emerge in bulk calculation, suggesting a possible role of interfacial effects. The overall enhancement of SHE could arise from the proximity to the magnet, though it is possible that the spin Hall effect in Fe3GaTe2\mathrm{Fe}_{3}\mathrm{GaTe}_{2} could also contribute to unconventional SOT[13, 41, 43, 44, 45, 46, 47, 48, 49, 50]. This highlights the unique behaviors of the TaIrTe4/Fe3GaTe2\mathrm{TaIrTe}_{4}/\mathrm{Fe}_{3}\mathrm{GaTe}_{2} heterostructure and suggests new avenues for exploring SOT in low-symmetry materials.

Summary

In summary, we demonstrated the potential of TaIrTe4/Fe3GaTe2\mathrm{TaIrTe}_{4}/\mathrm{Fe}_{3}\mathrm{GaTe}_{2} vdW heterostructures for generating a large and tunable nonlinear 2nd harmonic Hall effect, and energy-efficient deterministic field-free magnetization switching at room temperature. By leveraging the unique properties of the topological Weyl semimetal TaIrTe4\mathrm{TaIrTe}_{4} and the magnetic Fe3GaTe2\mathrm{Fe}_{3}\mathrm{GaTe}_{2} with strong PMA, our findings reveal a large non-linear Hall effect, substantial unconventional out-of-plane damping-like torque and a remarkably low switching current density, outperforming conventional systems. To unveil the origin of unconventional charge-spin conversion phenomena in TaIrTe4\mathrm{TaIrTe}_{4}, detailed first-principles calculations were performed considering crystal symmetry and its impact on the energy-dependent electronic structure and spin Hall conductivity. Finally, we measured a substantial and tunable damping-like torque and observed deterministic field-free magnetization switching at a very low current density offering a promising route to energy-efficient and external field-free spintronic technologies.

Methods

Single crystal growth: TaIrTe4_4 single crystals were synthesized by evaporating tellurium from a Ta-Ir-Te melt, with the crystal growth conducted at 850 ∘C850\,^\circ\mathrm{C} and Te condensation at 720 ∘C720\,^\circ\mathrm{C}[51]. Fe3_3GaTe2_2 single crystals were grown via a self-flux method using Fe, Ga, and Te with 99.99% purity in the molar ratio of 1:1:2 in an evacuated and sealed quartz tube. The solid reactions took place for 24 hr at 1273 K, followed by cooling to 1153 K within 1 hr and slowly cooling down to 1053 K within 100 hr[10].

Device fabrication: The van der Waals heterostructure samples were prepared by mechanically exfoliating nanolayers of TaIrTe4_4 and Fe3_3GaTe2_2 crystals on top of each other on a SiO2_2/Si wafer using the Scotch tape method inside a glove box. The top sample surface was immediately capped with a 2 nm Al2_2O3_3 layer to protect from degradation with time. For Devs1-3 nearly rectangular-shaped flakes were selected and the TaIrTe4_4/Fe3_3GaTe2_2 heterostructures were patterned to Hall-bar geometry using electron-beam lithography (EBL) and Ti (15 nm)/Au (250 nm) contacts were prepared by EBL and electron beam evaporation. For Dev4, flakes are in arbitrary shapes, therefore dry physical etching by Ar ion milling was used to fabricate well-defined Hall-bar devices.

Spin-orbit torque 2nd harmonic measurements: Spin-orbit torque was characterized using an in-plane 2nd harmonic Hall lock-in measurement technique. The Rxy1ωR_{xy}^{1\omega} and Rxy2ωR_{xy}^{2\omega} for an a.c. current IωI^\omega of 213.3 Hz were simultaneously measured while rotating the sample in the plane (azimuthal angle ϕB\phi_B) under an external field μ0Hext\mu_0 H_{\mathrm{ext}}. The harmonic measurements were conducted using a Lock-in SR830 to detect the in-phase 1st and out-of-phase 2nd harmonic voltages. The 2nd harmonic measurements in the high magnetic field range were performed with a Quantum Design cryogen-free PPMS DynaCool system, interfaced with the SR830 to record the 1st and 2nd harmonic voltages. The 1st harmonic signal is detected by putting the voltmeter in phase with the oscillator, whereas the 2nd harmonic signal is out of phase with the source signal.

Spin-orbit torque switching measurements were conducted in a vacuum cryostat with a magnetic field strength of up to 0.7 T. Electronic measurements were carried out using a Keithley 6221 current source and a Keithley 2182A nanovoltmeter. To monitor the longitudinal and transverse Hall resistances, Keithley 2182A nanovoltmeters were employed. For SOT-induced magnetization switching, the Keithley 2182A nanovoltmeters were used to observe the Hall resistances responses, while a Keithley 6221 AC source applied a pulse current of 50 microsecond (μs) through the device, followed by a DC read current of magnitude 500 μA.

Density functional theory calculations for bulk TaIrTe4_4 were performed using the Quantum Espresso package[52, 53] by employing the Perdew, Burke, and Ernzerhof (PBE) generalized gradient approximation (GGA) for exchange-correlation functional[54]. We used fully relativistic pseudopotentials and expanded the electron wave functions in a plane-wave basis with the energy cutoff of 80 Ry. We adopted an orthorhombic unit cell with the experimental lattice constants a=3.77a=3.77 Å, b=12.42b=12.42 Å, and c=13.18c=13.18 Å[55]. The atomic positions were relaxed with the force and energy convergence thresholds set to 10−310^{-3} Ry/Bohr and 10−410^{-4} Ry, respectively. The Brillouin Zone (BZ) was sampled following the Monkhorst-Pack scheme with the k-grids of 20×8×8 and adopting a Gaussian smearing of 10−310^{-3} Ry. For the post-processing analysis, we used the python package PAOFLOW, which projects the ab initio wavefunctions onto pseudo-atomic orbital (PAO) basis to construct tight-binding Hamiltonians[56, 57], further interpolated to a denser grid of 80×40×40. The charge-to-spin conversion response tensors were calculated using the approaches implemented in PAOFLOW and described in the previous works[58, 59, 60].

Acknowledgments

Authors acknowledge funding from European Commission (EU) Graphene Flagship, European Innovation Council (EIC) project 2DSPIN-TECH (No. 101135853), 2D TECH VINNOVA competence center (No. 2019-00068), Wallenberg Initiative Materials Science for Sustainability (WISE) funded by the Knut and Alice Wallenberg Foundation, EU Graphene Flagship (Core 3, No. 881603), Swedish Research Council (VR) grant (No. 2021–04821, No. 2018-07046), FLAG-ERA project 2DSOTECH (VR No. 2021-05925) and MagicTune, Carl Tryggers foundation, Graphene Center, Chalmers-Max IV collaboration grant, VR Sweden-India collaboration grant, Areas of Advance (AoA) Nano, AoA Materials Science and AoA Energy programs at Chalmers University of Technology, Dutch Research Council (NWO grant OCENW.M.22.063). The fabrication of devices was performed at Nanofabrication laboratory MyFab at Chalmers University of Technology. The calculations were carried out on the Dutch national e-infrastructure with the support of SURF Cooperative (EINF-10786) and on the Hábrók high-performance computing cluster of the University of Groningen.

Corresponding author

Correspondence to Saroj P. Dash ([email protected])

Contributions

L.P., S.P.D. conceived the idea and designed the experiments. L.P. fabricated and characterized the devices with support from B.Z., R.N., L.S., H.B. and P.R. The TaIrTe4\mathrm{TaIrTe}_{4} single crystals were grown by A.A. and M.A.H, while G.Z., H.W., and H.C. grew the Fe3GaTe2\mathrm{Fe}_{3}\mathrm{GaTe}_{2} single crystals. V.L., K.T. and J.S. performed, analyzed and described the density functional theory calculations. J.S. supervised theoretical calculations. L.P. and S.P.D. analyzed and interpreted the experimental data and wrote the manuscript, with comments from all the authors. S.P.D. coordinated and supervised the project.

Note: After preparation of this manuscript, we came across reports on magnetization switching in TaIrTe4/Fe3GaTe2\mathrm{TaIrTe}_{4}/\mathrm{Fe}_{3}\mathrm{GaTe}_{2} system[33, 61]. However, spin dynamics experiments to understand the spin-orbit torque phenomena in all-2D vdW heterostructures are so far lacking. In our manuscript, in addition to energy-efficient magnetization switching, we report a detailed understanding of unconventional and tunable SOT magnetization dynamics using 2nd2^{\mathrm{nd}} harmonic measurements in all-vdW heterostructures for the first time.

References

  1. [1]Manchon, A. et al. Current-induced spin-orbit torques in ferromagnetic and antiferromagnetic systems. Rev. Mod. Phys. 91, 035004 (2019).DOI
  2. [2]Shao, Q. et al. Roadmap of Spin–Orbit Torques. IEEE Trans. Magn. 57, 1–39 (2021).
  3. [3]Han, W., Otani, Y. & Maekawa, S. Quantum materials for spin and charge conversion. Npj Quantum Mater. 3, 27 (2018).DOI
  4. [4]Liu, L. et al. Spin-Torque Switching with the Giant Spin Hall Effect of Tantalum. Science 336, 555–558 (2012).DOI
  5. [5]Gong, C. et al. Discovery of intrinsic ferromagnetism in two-dimensional van der Waals crystals. Nature 546, 265–269 (2017).
  6. [6]Zhao, B. et al. Unconventional Charge–Spin Conversion in Weyl-Semimetal WTe₂. Adv. Mater. 32, 2000818 (2020).DOI
  7. [7]Zhao, B. et al. Observation of charge to spin conversion in Weyl semimetal WTe₂ at room temperature. Phys. Rev. Res. 2, 013286 (2020).DOI
  8. [8]Zhang, Y. et al. Room temperature field-free switching of perpendicular magnetization through spin-orbit torque originating from low-symmetry type II Weyl semimetal. Sci. Adv. 9, eadg9819 (2023).DOI
  9. [9]Liu, Y. et al. Field-free switching of perpendicular magnetization at room temperature using out-of-plane spins from TaIrTe₄. Nat. Electron. 6, 732–738 (2023).DOI
  10. [10]Zhang, G. et al. Above-room-temperature strong intrinsic ferromagnetism in 2D van der Waals Fe₃GeTe₂ with large perpendicular magnetic anisotropy. Nat. Commun. 13, 5067 (2022).DOI
  11. [11]Zhao, B., Bainsla, L., Ngaloy, R., Svedlindh, P. & Dash, S. P. Coexistence of non-trivial van der Waals magnetic orders enable field-free spin-orbit torque switching at room temperature. Preprint at http://arxiv.org/abs/2308.13408 (2023).
  12. [12]Wang, X. et al. Current-driven magnetization switching in a van der Waals ferromagnet Fe₃GeTe₂. Sci. Adv. 5, eaaw8904 (2019).
  13. [13]Zhang, K. et al. Gigantic Current Control of Coercive Field and Magnetic Memory Based on Nanometer-Thin Ferromagnetic van der Waals Fe₃GeTe₂. Adv. Mater. 33, 2004110 (2021).DOI
  14. [14]Kao, I.-H. et al. Deterministic switching of a perpendicularly polarized magnet using unconventional spin-orbit torques in WTe₂. Nat. Mater. 21, 1029–1034 (2022).DOI
  15. [15]Kajale, S. N., Nguyen, T., Hung, N. T., Li, M. & Sarkar, D. Field-free deterministic switching of all–van der Waals spin-orbit torque system above room temperature. Sci. Adv. 10, eadk8669 (2024).DOI
  16. [16]Shin, I. et al. Spin–Orbit Torque Switching in an All-Van der Waals Heterostructure. Adv. Mater. 34, 2101730 (2022).arxiv.org/abs/2102.09300
  17. [17]Pan, Z.-C. et al. Room-temperature orbit-transfer torque enabling van der Waals magnetoresistive memories. Sci. Bull. 68, 2743–2749 (2023).DOI
  18. [18]Bainsla, L. Large out-of-plane spin–orbit torque in topological Weyl semimetal TaIrTe₄. Nat. Commun. 15, 4649 (2024).
  19. [19]Koepernik, K. et al. TaIrTe₄: A ternary type-II Weyl semimetal. Phys. Rev. B 93, 201101 (2016).
  20. [22]Li, W. et al. Room-Temperature van der Waals Ferromagnet Switching by Spin-Orbit Torques. Adv. Mater. 35, 2303688 (2023).DOI
  21. [23]Herring, C. & Kittel, C. On the Theory of Spin Waves in Ferromagnetic Media. Phys. Rev. 81, 869–880 (1951).DOI
  22. [24]Sato, K. et al. Weak itinerant ferromagnetism in Heusler-type Fe₂VAl₀.₉₅. Phys Rev B 82, 104408 (2010).DOI
  23. [25]Mishra, V. et al. Investigation of spin gapless semiconductor behaviour in quaternary CoFeMnSi Heusler alloy thin films on Si (100). J. Magn. Magn. Mater. 547, 168837 (2022).DOI
  24. [26]Miron, I. M. et al. Perpendicular switching of a single ferromagnetic layer induced by in-plane current injection. Nature 476, 189–193 (2011).
  25. [27]Wu, H. et al. Room-Temperature Spin-Orbit Torque from Topological Surface States. Phys. Rev. Lett. 123, 207205 (2019).DOI
  26. [28]Bose, A. et al. Tilted spin current generated by the collinear antiferromagnet ruthenium dioxide. Nat. Electron. 5, 267–274 (2022).DOI
  27. [29]Takeuchi, Y. et al. Spin-orbit torques in high-resistivity-W/CoFeB/MgO. Appl. Phys. Lett. 112, 192408 (2018).
  28. [30]Kajale, S. N. et al. Current-induced switching of a van der Waals ferromagnet at room temperature. Nat. Commun. 15, 1485 (2024).
  29. [31]Wang, H. et al. Room temperature energy-efficient spin-orbit torque switching in two-dimensional van der Waals Fe₃GeTe₂ induced by topological insulators. Nat. Commun. 14, 5173 (2023).DOI
  30. [32]Zhu, L. Switching of Perpendicular Magnetization by Spin–Orbit Torque. Adv. Mater. 35, 2300853 (2023).DOI
  31. [33]Zhang, Y. et al. Robust Field-Free Switching Using Large Unconventional Spin-Orbit Torque in an All-Van der Waals Heterostructure. Adv. Mater. 36, 2404644 (2024).arxiv.org/abs/2405.06250
  32. [34]Wang, X. et al. Room temperature field-free switching of CoFeB/MgO heterostructure based on large-scale few-layer WTe2. Cell Rep. Phys. Sci. 4, 101468 (2023).DOI
  33. [35]Hu, S. et al. Efficient perpendicular magnetization switching by a magnetic spin Hall effect in a noncollinear antiferromagnet. Nat. Commun. 13, 4447 (2022).DOI
  34. [36]Wang, F. et al. Field-free switching of perpendicular magnetization by two-dimensional PtTe₂/WTe₂ van der Waals heterostructures with high spin Hall conductivity. Nat. Mater. 23, 768–774 (2024).DOI
  35. [37]Yu, G. et al. Switching of perpendicular magnetization by spin–orbit torques in the absence of external magnetic fields. Nat. Nanotechnol. 9, 548–554 (2014).DOI
  36. [38]Pai, C.-F., Mann, M., Tan, A. J. & Beach, G. S. D. Determination of spin torque efficiencies in heterostructures with perpendicular magnetic anisotropy. Phys. Rev. B 93, 144409 (2016).DOI
  37. [39]Nguyen, M.-H. et al. Enhancement of the anti-damping spin torque efficacy of platinum by interface modification. Appl. Phys. Lett. 106, 222402 (2015).DOI
  38. [40]Pai, C.-F., Ou, Y., Vilela-Leão, L. H., Ralph, D. C. & Buhrman, R. A. Dependence of the efficiency of spin Hall torque on the transparency of Pt/ferromagnet interfaces. Phys. Rev. B 92, 064426 (2015).
  39. [41]Highly Efficient Room-Temperature Spin-Orbit-Torque Switching in a Van der Waals Heterostructure of Topological Insulator and Ferromagnet. Advanced Science vol. 11 2400893 (2024).
  40. [42]Roy, A., Guimarães, M. H. D. & Stawińska, J. Unconventional spin Hall effects in nonmagnetic solids. Phys. Rev. Mater. 6, 045004 (2022).DOI
  41. [43]Zhang, G. et al. Room-temperature Highly-Tunable Coercivity and Highly-Efficient Multi-States Magnetization Switching by Small Current in Single 2D Ferromagnet Fe₃GaTe₂. ACS Mater. Lett. 6, 482–488 (2024).DOI
  42. [44]Deng, Y. et al. Room-Temperature Highly Efficient Nonvolatile Magnetization Switching by Current in van der Waals Fe₃GaTe₂ Devices. Nano Lett. 24, 9302–9310 (2024).DOI
  43. [45]Zhang, K. et al. Highly Efficient Nonvolatile Magnetization Switching and Multi-Level States by Current in Single Van der Waals Topological Ferromagnet Fe₃GeTe₂. Adv. Funct. Mater. 31, 2105992 (2021).DOI
  44. [46]Johansen, Ø., Rissingård, V., Sudbø, A., Linder, J. & Brataas, A. A Current Control of Magnetism in Two-Dimensional \\mathrm{Fe₃GeTe₂}. Phys Rev Lett 122, 217203 (2019).DOI
  45. [47]Cui, J., Zhang, K.-X. & Park, J.-G. All van der Waals Three-Terminal SOT-MRAM Realized by Topological Ferromagnet Fe₃GeTe₂. Adv. Electron. Mater. 10, 2400041 (2024).DOI
  46. [48]Martin, F. et al. Strong bulk spin–orbit torques quantified in the van der Waals ferromagnet Fe₃GeTe₂. Mater. Res. Lett. 11, 84–89 (2023).
  47. [49]Robertson, I. O. et al. Imaging current control of magnetization in Fe₃GeTe₂ with a widefield nitrogen-vacancy microscope. 2D Mater. 10, 015023 (2022).DOI
  48. [50]Zhang, K.-X. et al. Broken Inversion Symmetry in Van Der Waals Topological Ferromagnetic Metal Iron Germanium Telluride. Adv. Mater. 36, 2312824 (2024).DOI
  49. [51]Chareev, D. A. et al. Growth of Transition-Metal Dichalcogenides by Solvent Evaporation Technique. Cryst. Growth Des. 20, 6930–6938 (2020).
  50. [52]Giannozzi, P. et al. QUANTUM ESPRESSO: a modular and open-source software project for quantum simulations of materials. J. Phys. Condens. Matter 21, 395502 (2009).DOI
  51. [53]Giannozzi, P. et al. Advanced capabilities for materials modelling with Quantum ESPRESSO. J. Phys. Condens. Matter 29, 465901 (2017).
  52. [54]Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 77, 3865–3868 (1996).pubmed.ncbi.nlm.nih.gov/10062328
  53. [55]Mar, A., Jobic, S. & Ibers, J. A. Metal-metal vs tellurium-tellurium bonding in WTe₂ and its ternary variants TaIrTe₄ and NbIrTe₄. J. Am. Chem. Soc. 114, 8963–8971 (1992).DOI
  54. [56]Buongiorno Nardelli, M. et al. PAOFLOW: A utility to construct and operate on ab initio Hamiltonians from the projections of electronic wavefunctions on atomic orbital bases, including characterization of topological materials. Comput. Mater. Sci. 143, 462–472 (2018).
  55. [57]Cerasoli, F. T. et al. Advanced modeling of materials with PAOFLOW 2.0: New features and software design. Comput. Mater. Sci. 200, 110828 (2021).arxiv.org/abs/2107.13026
  56. [58]Stawińska, J. et al. Giant spin Hall effect in two-dimensional monochalcogenides. 2D Mater. 6, 025012 (2019).
  57. [59]Roy, A. et al. Long-range current-induced spin accumulation in chiral crystals. Npj Comput. Mater. 8, 243 (2022).DOI
  58. [60]Tenzin, K. et al. Analogs of Rashba-Edelstein effect from density functional theory. Phys. Rev. B 107, 165140 (2023).DOI
  59. [61]Li, D. et al. Room-temperature van der Waals magnetoresistive memories with data writing by orbital current in the Weyl semimetal TaIrTe₄. Phys. Rev. B 110, 035423 (2024).DOI

Paper details

vlnc:W_rp338dr7
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