betaFeedback

WorksPhysics and Astronomy

Spin current leakage and Onsager reciprocity in interfacial spin-charge interconversion

Abstract

Experimental investigations of spin-charge interconversion in magnetic bilayers comprising a ferromagnet adjacent to a topological insulator have reported scattered results on the spin-charge and charge-spin conversion efficiency. Attempting to reconcile these contradicting experimental results, we develop a phenomenological theory of spin-charge interconversion accounting for both interfacial interconversion through the spin galvanic effect, also called the Rashba-Edelstein effect, as well as bulk interconversion via the spin Hall effect. We find that the spin current leakage into the nonmagnetic metal plays a central role during the spin-to-charge and charge-to-spin conversion, leading to dissymmetric interconversion processes. In particular, spin-to-charge conversion is much less sensitive to the spin current absorption in the nonmagnetic metal than charge-to-spin conversion. This suggests that spin pumping is a more trustable technique to extract the interfacial Rashba parameter than spin-orbit torque.

Introduction

Spin-to-charge interconversion mediated by spin-orbit coupling has become a central technique in spintronics enabling the electrical manipulation of magnetization [1], as well as the generation of charge currents induced by magnetization precession [2, 3, 4]. The standard system is composed of a magnetic thin film deposited on top of a nonmagnetic metal with strong spin-orbit coupling. Whereas the original workhorse of spin-charge interconversion was NiFe/Pt [5], a broad range of materials has been explored over the years, including both metallic [6, 7, 8] and insulating ferromagnets [9, 10, 11, 12], antiferromagnets [13, 14], adjacent to strongly spin-orbit coupled materials including transition metals (Pt, W, and Ta). In these systems, the interconversion is mostly attributed to the spin Hall effect (SHE) taking place in the bulk of the heavy metal [15, 16]. In the past decade, the attention has been shifted towards materials displaying large interfacial spin-moment locking. Recent research encompasses metals with strong interfacial Rashba states [17, 18, 19] (Ag/Bi, Ag/Sb or Ag/Bi2_2O3_3), topological insulators [20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33] (Bi2_2Se3_3, etc.), transition metal dichalcogenides [34, 35, 36, 37, 38] (MoS2_2, WTe2_2), van der Waals materials [39, 40, 41, 42, 43] (for non-local interconversion, see Refs. [44, 45]), oxide heterostructures [46, 47, 48, 49, 50] (see Ref. [51] for a review), chiral metals [52] and the so-called ferroelectric Rashba semiconductors such as SrTiO3_3 and α\alpha-Ge [53, 54], whose interfacial spin-momentum locking can be controlled by switching the ferroelectric polarization [55].

In all these materials, strong spin-orbit coupling and inversion symmetry breaking at the interface result in Dirac or Rashba spin-momentum locking [56], which enables interconversion between charge and spin currents via the so-called (inverse) spin galvanic effect (SGE), also known as the Rashba-Edelstein effect [57, 58, 59]. In contrast, the SHE that converts three-dimensional (3D) spin currents into 3D charge currents, the interfacial spin galvanic effect converts a two-dimensional (2D) charge current into a 3D spin current, and vice-versa. The interconversion efficiency of SHE is quantified by the dimensionless parameters ξsc\xi_{\mathrm{sc}} and ξcs\xi_{\mathrm{cs}}. In a bulk heavy metal, Onsager reciprocity dictates that ξcs=ξsc\xi_{\mathrm{cs}} = \xi_{\mathrm{sc}}. As discussed in more detail in the present work, in the case of a magnetic bilayer, this equality holds true as long as the charge current density involved in the definition of these efficiencies is restricted to the charge current flowing in the heavy metal.

The interconversion efficiency of SGE is rather quantified in terms of charge-to-spin conversion length λsc\lambda_{\mathrm{sc}} and spin-to-charge conversion inverse length qcsq_{\mathrm{cs}}. From a theoretical standpoint, ξ\xi is related to the SHE in the bulk heavy metal by the spin transparency, i.e., the ability of the heavy metal to absorb the spin current. The figures of merit λsc\lambda_{\mathrm{sc}} and qcsq_{\mathrm{cs}} have been recently derived for a 2D Rashba or Dirac gas in Refs. [33, 60, 61, 62] using either phenomenological or microscopic approaches. These references provide the general form λsc=(αR/ℏ)(1/τp+1/τt)\lambda_{\mathrm{sc}} = (\alpha_{\mathrm{R}}/\hbar)(1/\tau_{\mathrm{p}} + 1/\tau_{\mathrm{t}}) and qcs=(αR/ℏvF2)τtq_{\mathrm{cs}} = (\alpha_{\mathrm{R}}/\hbar v_{\mathrm{F}}^2)\tau_{\mathrm{t}}, where τt\tau_{\mathrm{t}} is the tunneling time between the 2D gas and the spin current source (typically an adjacent ferromagnet, but also possibly a nonmagnetic metal through which the spin current is injected) and τp\tau_{\mathrm{p}} is the momentum relaxation time in the 2D gas. When we emphasize that the validity of these theories, as well as the very definition of λsc\lambda_{\mathrm{sc}} and qcsq_{\mathrm{cs}}, is restricted to the ideal case of 2D transport. In other words, whether these parameters are meaningful in the case of oxide heterostructures where the charge current is purely two-dimensional, it may simply not apply in other metallic systems or in weakly insulating ”topological insulators” such as Bi2_2Se3_3. As a matter of fact, experiments in topological insulators report large variations in spin-charge interconversion efficiency. In particular, it is often observed that spin-to-charge and charge-to-spin conversion efficiencies massively differ from each other. Charge-to-spin conversion in Bi-based topological materials ranges from 0.4 to more than 400 [20, 21, 23, 24, 27, 29, 30, 31] whereas spin-to-charge conversion rather ranges from 0.0001 to 0.4 [22, 23, 25, 26], which cannot be properly accounted for using the simple formulas given above. This set of results contrasts sharply with the metallic case, where the reported efficiencies, ξcs\xi_{\mathrm{cs}} and ξsc\xi_{\mathrm{sc}}, are usually of the same order.

In an attempt to clarify the situation, we develop a phenomenological model for spin-charge interconversion in a metallic bilayer where interfacial SGE and bulk SHE coexist. We show that whereas spin-to-charge and charge-to-spin conversion are intrinsically asymmetric, although not breaking Onsager reciprocity, the leakage of the spin current in the neighboring nonmagnetic metal plays a crucial role when the conductivity of the nonmagnetic metal becomes large.

Drift-diffusion theory

We consider the system represented in Figure 1 and composed of three elements: a 2D gas embedded between a ferromagnet F and a nonmagnetic metal N, of thicknesses dFd_{\mathrm{F}} and dNd_{\mathrm{N}}, respectively. The 2D gas can be thought of as the surface state of a heavy metal (e.g., Co/Pt, Ag/Bi) or that of a weakly conductive topological insulator (e.g., Bi2_2Se3_3). For simplicity, we assume that the spin-charge conversion in the 2D gas takes place through the SGE (or Rashba-Edelstein) effect, with an effective coupling αR\alpha_{\mathrm{R}} (eV.m), and in the nonmagnetic metal, it is induced by SHE with an angle θH\theta_{\mathrm{H}} (unitless).

Charge-to-spin and spin-to-charge conversion in a metallic bilayer

Figure 1. (Color online). (a) Charge-to-spin and (b) spin-to-charge conversion in a metallic bilayer composed of a 2D gas embedded between a ferromagnet and a nonmagnetic metal.

Jc2D=σ2DE−e2N2DαRℏz×μ2D,(1)\mathbf{J}_{\mathrm{c}}^{\mathrm{2D}}=\sigma_{\mathrm{2D}}\mathbf{E}-e^{2}\mathcal{N}_{\mathrm{2D}}\frac{\alpha_{\mathrm{R}}}{\hbar}\mathbf{z}\times\boldsymbol{\mu}_{\mathrm{2D}}, \tag*{(1)}

where the second term is the SGE that converts a nonequilibrium spin density into a charge current. Here, μ2D=μ2Dy\boldsymbol{\mu}_{\mathrm{2D}}=\mu_{\mathrm{2D}}\mathbf{y} (in the units of V) is the spin chemical potential, N2D\mathcal{N}_{\mathrm{2D}} (eV−1⋅^{-1}\cdotm2^{2}) and σ2D\sigma_{\mathrm{2D}} (Ω−1\Omega^{-1}) are the density of states and conductivity of the 2D gas. The spin continuity equation reads

∂tμ2D=αRℏz×E−μ2Dτsf−μ2D−μFiτF−μ2D−μNiτN.(2)\partial_t\boldsymbol{\mu}_{\mathrm{2D}}=\frac{\alpha_{\mathrm{R}}}{\hbar}\mathbf{z}\times\mathbf{E}-\frac{\boldsymbol{\mu}_{\mathrm{2D}}}{\tau_{\mathrm{sf}}}-\frac{\boldsymbol{\mu}_{\mathrm{2D}}-\boldsymbol{\mu}_{\mathrm{F}}^{i}}{\tau_{\mathrm{F}}}-\frac{\boldsymbol{\mu}_{\mathrm{2D}}-\boldsymbol{\mu}_{\mathrm{N}}^{i}}{\tau_{\mathrm{N}}}. \tag*{(2)}

The first term is the inverse SGE, i.e., the generation of a spin density induced by the flow charge current, with αR\alpha_{\mathrm{R}} being the so-called Rashba parameter. A formal derivation of the spin diffusion equation in a Rashba gas can be found in Refs. [63, 64, 65]. The second term accounts for the spin relaxation in the gas (that may be anisotropic, although for the sake of simplicity this is neglected in the present study) and the last two terms account for spin injection in the neighboring layers: 1/τF(N)1/\tau_{\mathrm{F(N)}} is the tunneling rate between the 2D gas and the ferromagnet (nonmagnetic metal), and μF(N)i\boldsymbol{\mu}_{\mathrm{F(N)}}^{i} is the interfacial spin chemical potential in the ferromagnet (nonmagnetic metal). These tunneling rates are related to the interfacial spin currents flowing between the 2D gas and the ferromagnetic (JsF,iJ_{\mathrm{s}}^{\mathrm{F},i}) and nonmagnetic metals (JsN,iJ_{\mathrm{s}}^{\mathrm{N},i}) through the boundary conditions

JsN,i=GN(μNi−μ2D),(3)\mathbf{J}_{\mathrm{s}}^{\mathrm{N},i}=G_{\mathrm{N}}\left(\boldsymbol{\mu}_{\mathrm{N}}^{i}-\boldsymbol{\mu}_{\mathrm{2D}}\right), \tag*{(3)}

where GN=e2N2D/τNG_{\mathrm{N}}=e^{2}\mathcal{N}_{\mathrm{2D}}/\tau_{\mathrm{N}} (Ω−1⋅m−2\Omega^{-1}\cdot\mathrm{m}^{-2}) is the interfacial conductance between the 2D gas. Notice that without loss of generality, the spin current is expressed in the same units as the charge current for simplicity. Similarly,

JsF,i=2GT↑↓μ2D−Js0,(4)\mathbf{J}_{\mathrm{s}}^{\mathrm{F},i}=2G_{\mathrm{T}\uparrow\downarrow}\boldsymbol{\mu}_{\mathrm{2D}}-\mathbf{J}_{\mathrm{s}}^{0}, \tag*{(4)}

where Js0\mathbf{J}_{\mathrm{s}}^{0} is due to spin pumping (if any), and 2GT↑↓=e2N2D/τF2G_{\mathrm{T}\uparrow\downarrow}=e^{2}\mathcal{N}_{\mathrm{2D}}/\tau_{\mathrm{F}} is the spin mixing conductance (Ω−1⋅m−2\Omega^{-1}\cdot\mathrm{m}^{-2}), assuming that the magnetization is perpendicular to the spin density in the 2D gas (which is valid in both spin-to-charge and charge-to-spin interconversion scenarios), i.e., μFi=0\boldsymbol{\mu}_{\mathrm{F}}^{i}=0.

In the nonmagnetic metal, the (3D) charge and spin currents, JcN\mathbf{J}_{\mathrm{c}}^{\mathrm{N}} and JsN\mathbf{J}_{\mathrm{s}}^{\mathrm{N}}, are defined

JcN=σNE+θHσN∇×μN,(5)\mathbf{J}_{\mathrm{c}}^{\mathrm{N}}=\sigma_{\mathrm{N}}\mathbf{E}+\theta_{\mathrm{H}}\sigma_{\mathrm{N}}\boldsymbol{\nabla}\times\boldsymbol{\mu}_{\mathrm{N}}, \tag*{(5)}
JsN=−σN∇μN−θHσNy×E.(6)\mathbf{J}_{\mathrm{s}}^{\mathrm{N}}=-\sigma_{\mathrm{N}}\boldsymbol{\nabla}\boldsymbol{\mu}_{\mathrm{N}}-\theta_{\mathrm{H}}\sigma_{\mathrm{N}}\mathbf{y}\times\mathbf{E}. \tag*{(6)}

Here, one recognizes the SHE and inverse SHE, σN\sigma_{\mathrm{N}} (Ω−1⋅\Omega^{-1}\cdotm−1^{-1}) being the conductivity of the metal. In the scenario studied in the present work, the spin polarization remains aligned along y and therefore all the equations above can be simply projected onto this direction. Applying the two boundary conditions JsN(z=0)=GN(μNi−μ2D)\mathbf{J}_{\mathrm{s}}^{\mathrm{N}}(z=0)=G_{\mathrm{N}}(\boldsymbol{\mu}_{\mathrm{N}}^{i}-\boldsymbol{\mu}_{\mathrm{2D}}) and JsN(z=−dN)=0\mathbf{J}_{\mathrm{s}}^{\mathrm{N}}(z=-d_{\mathrm{N}})=0, we obtain

μN(z)=11+ηN(ηNμ2D+λ~Nθ~HE)cosh⁡z+dNλNcosh⁡dNλN+λ~Nθ~HEtanh⁡dNλNsinh⁡zλNcosh⁡dNλN−1,(7)\mu_{\mathrm{N}}(z)=\frac{1}{1+\eta_{\mathrm{N}}}\left(\eta_{\mathrm{N}}\mu_{\mathrm{2D}}+\tilde{\lambda}_{\mathrm{N}}\tilde{\theta}_{\mathrm{H}}E\right)\frac{\cosh\frac{z+d_{\mathrm{N}}}{\lambda_{\mathrm{N}}}}{\cosh\frac{d_{\mathrm{N}}}{\lambda_{\mathrm{N}}}}+\tilde{\lambda}_{\mathrm{N}}\tilde{\theta}_{\mathrm{H}}E\frac{\tanh\frac{d_{\mathrm{N}}}{\lambda_{\mathrm{N}}}\sinh\frac{z}{\lambda_{\mathrm{N}}}}{\cosh\frac{d_{\mathrm{N}}}{\lambda_{\mathrm{N}}}-1}, \tag*{(7)}

where θ~H=θH(1−cosh⁡−1dNλN)\tilde{\theta}_{H}=\theta_{H}\left(1-\cosh^{-1}\frac{d_{N}}{\lambda_{N}}\right), λ~N=λN/tanh⁡dNλN\tilde{\lambda}_{N}=\lambda_{N}/\tanh\frac{d_{N}}{\lambda_{N}}, and ηN=λ~NGN/σN\eta_{N}=\tilde{\lambda}_{N}G_{N}/\sigma_{N}. This relation does not depend on the scenario (spin-to-charge or charge-to-spin conversion) and shows that the nonmagnetic metal always acts like a spin sink since (for E=0E=0), ηN→+∞⇒μNi=μ2D\eta_{N}\to+\infty\Rightarrow\mu_{N}^{i}=\mu_{2D} (perfect spin insulator) and ηN→0⇒μNi=0\eta_{N}\to0\Rightarrow\mu_{N}^{i}=0 (perfect spin sink). Let us now compute the spin-charge interconversion efficiency in the cases depicted in Figure 1.

Charge to spin conversion

An electric field E\mathbf{E} is applied along x\mathbf{x} and generates a spin accumulation, polarized along y\mathbf{y}, throughout the structure [see Eq. (7)]. This spin accumulation injects a spin current JsF,iJ_{s}^{F,i} into the ferromagnet. In the absence of spin pumping, JsF,i=2G↑↓μ2DJ_{s}^{F,i}=2G_{\uparrow\downarrow}\mu_{2D} [Eq. (4)]. The spin chemical potential μ2D\mu_{2D} can be obtained from the spin continuity equation, Eq. (2), posing ∂tμ2D=0\partial_{t}\mu_{2D}=0 and injecting μNi=μN(0)\mu_{N}^{i}=\mu_{N}(0) [Eq. (7)]. We get

μ2D=αRℏ+1τNλ~Nθ~H1+ηN1τsf+1τF+1τN11+ηNE.(8)\mu_{2D}=\frac{\frac{\alpha_{R}}{\hbar}+\frac{1}{\tau_{N}}\frac{\tilde{\lambda}_{N}\tilde{\theta}_{H}}{1+\eta_{N}}}{\frac{1}{\tau_{sf}}+\frac{1}{\tau_{F}}+\frac{1}{\tau_{N}}\frac{1}{1+\eta_{N}}}E. \tag*{(8)}

which implies that

JsF,i=2G↑↓αRℏ+1τNλ~Nθ~H1+ηN1τsf+1τF+1τN11+ηNE.(9)J_{s}^{F,i}=2G_{\uparrow\downarrow}\frac{\frac{\alpha_{R}}{\hbar}+\frac{1}{\tau_{N}}\frac{\tilde{\lambda}_{N}\tilde{\theta}_{H}}{1+\eta_{N}}}{\frac{1}{\tau_{sf}}+\frac{1}{\tau_{F}}+\frac{1}{\tau_{N}}\frac{1}{1+\eta_{N}}}E. \tag*{(9)}

To obtain, the unitless charge-to-spin interconversion efficiency, we divide the spin current by the current flowing in both the nonmagnetic metal and 2D gas, Jc=σNdN+σ2DdN+t2DEJ_{c}=\frac{\sigma_{N}d_{N}+\sigma_{2D}}{d_{N}+t_{2D}}E, where t2Dt_{2D} is the effective thickness of the 2D gas. We then obtain,

ξcs=∣JsF,iJc∣=2G↑↓dN+t2DσNdN+σ2DαRℏ+1τNλ~Nθ~H1+ηN1τsf+1τF+1τN11+ηN.(10)\xi_{cs}=\left|\frac{J_{s}^{F,i}}{J_{c}}\right|=2G_{\uparrow\downarrow}\frac{d_{N}+t_{2D}}{\sigma_{N}d_{N}+\sigma_{2D}}\frac{\frac{\alpha_{R}}{\hbar}+\frac{1}{\tau_{N}}\frac{\tilde{\lambda}_{N}\tilde{\theta}_{H}}{1+\eta_{N}}}{\frac{1}{\tau_{sf}}+\frac{1}{\tau_{F}}+\frac{1}{\tau_{N}}\frac{1}{1+\eta_{N}}}. \tag*{(10)}

One immediately notices the competition between the spin relaxation inside the 2D gas (∼1/τsf\sim1/\tau_{sf}), the spin absorption in the nonmagnetic metal (∼1/τN\sim1/\tau_{N}) and in the ferromagnet (∼1/τF∝G↑↓\sim1/\tau_{F}\propto G_{\uparrow\downarrow}). This competition will be analyzed in more details in the next section. For now, we emphasize that other renormalizations can be adopted. For instance, assuming that the charge-to-spin conversion only occurs in the 2D gas (θH=0\theta_{H}=0), one can normalize the spin current to the charge current that effectively flows along the interface, Jc2D=σ2DEJ_{c}^{2D}=\sigma_{2D}E, which yields,

qcs=∣JsF,iJc2D∣=αRℏ2G↑↓σ2D11τsf+1τF+1τN11+ηN.(11)q_{cs}=\left|\frac{J_{s}^{F,i}}{J_{c}^{2D}}\right|=\frac{\alpha_{R}}{\hbar}\frac{2G_{\uparrow\downarrow}}{\sigma_{2D}}\frac{1}{\frac{1}{\tau_{sf}}+\frac{1}{\tau_{F}}+\frac{1}{\tau_{N}}\frac{1}{1+\eta_{N}}}. \tag*{(11)}

This expression extends the ones derived in Refs. [33, 60, 61, 62] beyond the model Rashba or Dirac gas, and includes the effect of the adjacent spin sink, providing a more accurate expression for the figure of merit of interfacial spin-to-charge conversion.

Spin to charge conversion

A spin current JsF,i=−(e2N2D)/τFμ2D+Js0J_{s}^{F,i}=-(e^{2}N_{2D})/\tau_{F}\mu_{2D}+J_{s}^{0} is pumped out of the ferromagnet and converted into a charge current composed of Jc2D=e2N2D(αR/ℏ)μ2DJ_{c}^{2D}=e^{2}N_{2D}(\alpha_{R}/\hbar)\mu_{2D} and JcN=−(1/dN)∫−dN0dzθHσN∂zμNJ_{c}^{N}=-(1/d_{N})\int_{-d_{N}}^{0}dz\theta_{H}\sigma_{N}\partial_{z}\mu_{N}. The spin continuity equation, Eq. (2), gives

μ2D=Js0e2N2D11τsf+1τF+1τN11+ηN.(12)\mu_{2D}=\frac{J_{s}^{0}}{e^{2}N_{2D}}\frac{1}{\frac{1}{\tau_{sf}}+\frac{1}{\tau_{F}}+\frac{1}{\tau_{N}}\frac{1}{1+\eta_{N}}}. \tag*{(12)}

Therefore, we deduce the charge current flowing in the 2D gas and in the nonmagnetic metal,

Jc2D=αRℏ11τsf+1τF+1τN11+ηNJs0,(13)J_{c}^{2D}=\frac{\alpha_{R}}{\hbar}\frac{1}{\frac{1}{\tau_{sf}}+\frac{1}{\tau_{F}}+\frac{1}{\tau_{N}}\frac{1}{1+\eta_{N}}}J_{s}^{0}, \tag*{(13)}
JcN=θ~Hλ~NdN11τsf+1τF+1τN11+ηNJs0.(14)J_{c}^{N}=\frac{\tilde{\theta}_{H}\tilde{\lambda}_{N}}{d_{N}}\frac{1}{\frac{1}{\tau_{sf}}+\frac{1}{\tau_{F}}+\frac{1}{\tau_{N}}\frac{1}{1+\eta_{N}}}J_{s}^{0}. \tag*{(14)}

The total current that is pumped through the system is therefore Icpump=w(JcNdN+Jc2D)I_{c}^{\mathrm{pump}}=w(J_{c}^{N}d_{N}+J_{c}^{2D}). In other words, if the effective {curly braces} {nonmagnetic+2D gas}\{\text{nonmagnetic}+2D\ \text{gas}\} thickness is dN+t2Dd_{N}+t_{2D}, the effective current density pumped out of the system is Jcpump=(JcNdN+Jc2D)/(dN+t2D)J_{c}^{\mathrm{pump}}=(J_{c}^{N}d_{N}+J_{c}^{2D})/(d_{N}+t_{2D}). Finally, the spin-to-charge efficiency reads

ξsc=∣JcpumpJsF,i∣=1dN+t2DαRℏ+1τNλ~Nθ~H1+ηN1τsf+1τF+1τN11+ηN.(15)\xi_{sc}=\left|\frac{J_{c}^{\mathrm{pump}}}{J_{s}^{F,i}}\right|=\frac{1}{d_{N}+t_{2D}}\frac{\frac{\alpha_{R}}{\hbar}+\frac{1}{\tau_{N}}\frac{\tilde{\lambda}_{N}\tilde{\theta}_{H}}{1+\eta_{N}}}{\frac{1}{\tau_{sf}}+\frac{1}{\tau_{F}}+\frac{1}{\tau_{N}}\frac{1}{1+\eta_{N}}}. \tag*{(15)}

Comparing Eqs. (10) and (15), one can notice that - besides an obvious geometrical factor - these expressions differ on the role played by the spin mixing conductance, G↑↓∝1/τFG_{\uparrow\downarrow}\propto1/\tau_{F}, so that in general ξcs≠ξsc\xi_{cs}\ne\xi_{sc}. Notice though that this inequality does not mean that Onsager reciprocity is broken at all. As a matter of fact, Onsager reciprocity applies to the response tensor of generalized currents to thermodynamical forces, or, in other words, to the connection between current densities and chemical potential gradients (for a discussion on Onsager reciprocity applies to spin pumping and spin torque, see Ref. [66]). Since the spin-charge interconversion efficiencies are defined as the ratio between two current densities, they do not fulfill Onsager reciprocity.

In fact, two limits illustrate the difference between spin-to-charge and charge-to-spin conversion in metallic bilayers. In the absence of interfacial spin-charge interconversion, GN→+∞G_{N}\to+\infty, t2D→0t_{2D}\to0, σ2D→0\sigma_{2D}\to0 and τsf→+∞\tau_{sf}\to+\infty, we obtain

ξsc=λ~NdNθ~H≈ξcs=θ~H.(16)\xi_{sc}=\frac{\tilde{\lambda}_{N}}{d_{N}}\tilde{\theta}_{H}\approx\xi_{cs}=\tilde{\theta}_{H}. \tag*{(16)}

These expressions are valid in the limit 2G↑↓λ~N/σN≫12G_{\uparrow\downarrow}\tilde{\lambda}_{N}/\sigma_{N}\gg1. Finally, assuming only interfacial spin-to-charge conversion is allowed, the charge current is only pumped in the 2D gas and the spin-to-charge conversion length reads

λsc=αRℏ11τsf+1τN11+ηN.(17)\lambda_{sc}=\frac{\alpha_{R}}{\hbar}\frac{1}{\frac{1}{\tau_{sf}}+\frac{1}{\tau_{N}}\frac{1}{1+\eta_{N}}}. \tag*{(17)}

which qualitatively agrees with the expressions derived in Refs. 33, 60–62. The difference is that in Eq. (17), the momentum relaxation time τp\tau_p is replaced by the spin-flip time in the 2D gas τsf\tau_{\mathrm{sf}}, and the tunneling time τt\tau_t is replaced by τN\tau_N, accounting for the backflow spin current in the nonmagnetic metal.

Influence of spin current leakage

Phenomenological parameters

Besides the conventional electronic properties of the metallic layers (conductivity, spin diffusion length, spin Hall angle), the phenomenological theory developed in the previous section is controlled by two parameters: the coupling between the ferromagnetic layer and the 2D gas, 1/τF=G↑↓/e2NF1/\tau_F = G_{\uparrow\downarrow}/e^2N_F, and the coupling between the 2D gas and the nonmagnetic metal, 1/τN=GN/e2NN1/\tau_N = G_N/e^2N_N. Qualitatively, the interfacial areal conductance between two metallic layers is about e2/hA≈1014−1015 Ω−1 m−2e^2/hA \approx10^{14} - 10^{15}\ \Omega^{-1}\,\mathrm{m}^{-2} (see, e.g., Refs. 67 and 68), AA being the area of a unit cell at the interface. Assuming an interfacial density of state of about 1/(eV⋅A)1/(\mathrm{eV}\cdot A), one arrives at a coupling rate of 1/τN,F≈1015 s−11/\tau_{N,F} \approx10^{15}\ \mathrm{s}^{-1}, which seems reasonable for direct coupling between metallic states and remains one to two orders of magnitude larger than the typical spin relaxation time in metals.

Interfacial interconversion

Let us first consider the case where the spin-charge interconversion is solely due to spin-momentum locking at the interface. In this case, the adjacent metallic layer does not participate to the conversion itself but only in the spin absorption. To model this case, we set the following parameters, GN=G↑↓=4×1014 Ω−1m−2G_N = G_{\uparrow\downarrow} = 4 \times10^{14}\ \Omega^{-1}\mathrm{m}^{-2}. In Figure 2(a), we report the dependence of the interconversion efficiencies, ξsc\xi_{\mathrm{sc}} (red) and ξcs\xi_{\mathrm{cs}} (blue), as a function of the nonmagnetic metal conductivity σN\sigma_N for various values of the spin diffusion length λN\lambda_N. Notice that the figure is given in logarithmic scale. Clearly, the spin-to-charge and charge-to-spin conversion efficiencies are markedly different in the limit of weakly conducting nonmagnetic metal (σN≈104 Ω−1m−1\sigma_N \approx10^4\ \Omega^{-1}\mathrm{m}^{-1}), providing an explanation for the experimental values reported in bilayers involving topological insulators (Bi2_2Se3_3, etc.), as pointed out above. Upon increasing the conductivity of the nonmagnetic metal, the charge-to-spin efficiency (blue) is progressively reduced whereas the spin-to-charge efficiency (red) is only weakly affected. In the case of a conducting nonmagnetic metal (σN≈106 Ω−1m−1\sigma_N \approx10^6\ \Omega^{-1}\mathrm{m}^{-1}), both efficiencies are comparable, as mentioned in the previous section, corroborating the experimental data collected using a conducting heavy metal substrate such as Pt. Reducing the spin diffusion length reduces the interconversion efficiency, as the spin information is lost in the nonmagnetic metal.

Spin-to-charge and charge-to-spin conversion efficiency and spin accumulation profile

Figure 2. (Color online) (a) Spin-to-charge (red) and charge-to-spin (blue) conversion efficiency as a function of the conductivity of the nonmagnetic metal for different values of the spin diffusion length λN\lambda_N. (b) Spin accumulation profile in the nonmagnetic metal for different values of the conductivity σN\sigma_N. In these calculations, we set τsf=10−14 s\tau_{\mathrm{sf}} = 10^{-14}\ \mathrm{s}, σ2D=2×10−4 Ω−1\sigma_{2D} = 2 \times10^{-4}\ \Omega^{-1}, dN=5 nmd_N = 5\ \mathrm{nm} and τ2D=1 nm\tau_{2D} = 1\ \mathrm{nm}.

In the model proposed above, the interfacial spin chemical potential due to spin pumping, μNcs\mu_N^{\mathrm{cs}}, and that due to electrical injection, μNcs\mu_N^{\mathrm{cs}} are related by

μNsc(z)Js0=μNcs(z)(e2NFαR/ℏ)E(18)\frac{\mu_N^{\mathrm{sc}}(z)}{J_s^0} = \frac{\mu_N^{\mathrm{cs}}(z)}{(e^2N_F\alpha_R/\hbar)\mathcal{E}} \tag*{(18)}
=ηN1+ηNcosh⁡z+dNλNcosh⁡dNλN11τsf+1τN+1τF11+ηN.(19)= \frac{\eta_N}{1+\eta_N}\frac{\cosh\frac{z+d_N}{\lambda_N}}{\cosh\frac{d_N}{\lambda_N}}\frac{1}{\frac{1}{\tau_{\mathrm{sf}}}+\frac{1}{\tau_N}+\frac{1}{\tau_F}\frac{1}{1+\eta_N}}. \tag*{(19)}

The profile of the spin chemical potential is reported in Figure 2(b) for λN=1 nm\lambda_N = 1\ \mathrm{nm} and for various conductivities σN\sigma_N. The larger the conductivity the smaller the interfacial spin chemical potential, suggesting that interfacial spin-charge interconversion is enhanced when the adjacent layer is a good spin insulator.

Interfacial versus bulk interconversion

Let us now turn on the spin Hall angle of the nonmagnetic layer. Figure 3 reports the interconversion efficiencies as a function of the nonmagnetic metal conductivity for various magnitudes of the spin Hall angle and for low [Figure 3(a)], medium [Figure 3(b)] and strong value of the Rashba term [Figure 3(c)]. On the right panel, the contribution of SGE (solid) and SHE (dashed) to spin-to-charge (red) and charge-to-spin conversions (blue). In the case of vanishingly weak Rashba strength [Figure 3(a,Figure 3b)], the interconversion is dominated by the (inverse) SHE and steadily increases with θH\theta_H. The spin-to-charge interconversion efficiency ξsc\xi_{\mathrm{sc}} (red lines) increases with the conductivity of the normal metal as more output charge current is allowed to flow. On the other hand, the charge-to-spin interconversion efficiency ξcs\xi_{\mathrm{cs}} (blue lines) decreases in the limit of highly conductive normal metal as the shunting increases. To understand this feature, Fig. 3(b) shows that the contributions of the SGE and ISGE systematically decrease when increasing the conductivity of the nonmagnetic metal, as relatively less current flows inside the 2D gas. Notice that the effect is more dramatic for the spin-to-charge than for the charge-to-spin conversion. The contribution of the SHE is more subtle. Whereas its contribution to the charge-to-spin interconversion increases steadily (dashed red), its contribution to spin-to-charge interconversion first increases, reaches a maximum, and then decreases (dashed blue). This behavior is associated with the competition between the enhanced SHE and the shunting effect.

Spin-to-charge and charge-to-spin conversion efficiencies and their corresponding contributions

Figure 3. (Color online). (Left panels) Spin-to-charge (red) and charge-to-spin (blue) conversion efficiencies for (a) weak, (c) intermediate, and (e) strong interfacial Rashba spin-orbit parameters for various values of the spin Hall angle. (Right panels) Corresponding contributions of the spin Hall effect (SHE), inverse spin Hall effect (ISHE), interfacial spin galvanic effect (SGE), and inverse spin galvanic effect (ISGE), for θH=10%\theta_{\mathrm{H}} = 10\%. In all these calculations, we set λN=1\lambda_{\mathrm{N}} = 1 nm.

Upon increasing the strength of the Rashba parameter [Figure 3(c) and (e)], the spin-to-charge and charge-to-spin conversion efficiencies become less sensitive to the spin diffusion length. Concomitantly, the spin-to-charge conversion (red lines) becomes progressively independent of the nonmagnetic metal conductivity [this is particularly clear in panel (e)], while the charge-to-spin conversion

keeps on decreasing. In other words, in the case of dominant interfacial spin-orbit coupling, the spin-to-charge conversion is much less sensitive to the spin current leakage than the charge-to-spin conversion. The decomposition of the conversion efficiencies in terms of SGE and SHE confirms this trend.

Extracting interfacial spin-charge interconversion coefficient from experiments

The spin-charge interconversion coefficients ξsc\xi_{\mathrm{sc}} and ξcs\xi_{\mathrm{cs}} are often used to extract the interfacial Rashba parameter, αR\alpha_{\mathrm{R}}. The extraction procedure requires the knowledge of basic transport properties such as conductivity, interfacial conductances, and spin diffusion lengths, but also relies on the scenario adopted. Assuming purely 2D transport, the Rashba parameter associated with charge-to-spin conversion reads

αR2Dℏ=ξcs2GT↑↓σ2DdN+t2D(1τsf+1τF),(20)\frac{\alpha_{\mathrm{R}}^{2\mathrm{D}}}{\hbar} = \frac{\xi_{\mathrm{cs}}}{2G_{\mathrm{T}\uparrow\downarrow}} \frac{\sigma_{\mathrm{2D}}}{d_{\mathrm{N}} + t_{\mathrm{2D}}} \left(\frac{1}{\tau_{\mathrm{sf}}} + \frac{1}{\tau_{\mathrm{F}}}\right), \tag*{(20)}

and the parameter associated with spin-to-charge conversion reads

αR2Dℏ=ξscτsf(dN+t2D).(21)\frac{\alpha_{\mathrm{R}}^{2\mathrm{D}}}{\hbar} = \frac{\xi_{\mathrm{sc}}}{\tau_{\mathrm{sf}}}(d_{\mathrm{N}} + t_{\mathrm{2D}}). \tag*{(21)}

Now, assuming that transport is allowed in the neighboring metal, we obtain

αRℏ=ξcs2GT↑↓σ2D+σNdNdN+t2D(1τsf+1τF+1τN11+ηN)−1τNλ~Nθ~N1+ηN,(22)\frac{\alpha_{\mathrm{R}}}{\hbar} = \frac{\xi_{\mathrm{cs}}}{2G_{\mathrm{T}\uparrow\downarrow}} \frac{\sigma_{\mathrm{2D}} + \sigma_{\mathrm{N}}d_{\mathrm{N}}}{d_{\mathrm{N}} + t_{\mathrm{2D}}} \left(\frac{1}{\tau_{\mathrm{sf}}} + \frac{1}{\tau_{\mathrm{F}}} + \frac{1}{\tau_{\mathrm{N}}}\frac{1}{1+\eta_{\mathrm{N}}}\right) - \frac{1}{\tau_{\mathrm{N}}}\frac{\tilde{\lambda}_{\mathrm{N}}\tilde{\theta}_{\mathrm{N}}}{1+\eta_{\mathrm{N}}}, \tag*{(22)}

for charge-to-spin conversion and

αRℏ=ξsc(dN+t2D)(1τsf+1τN11+ηN)−1τNλ~Nθ~N1+ηN(23)\frac{\alpha_{\mathrm{R}}}{\hbar} = \xi_{\mathrm{sc}}(d_{\mathrm{N}} + t_{\mathrm{2D}})\left(\frac{1}{\tau_{\mathrm{sf}}} + \frac{1}{\tau_{\mathrm{N}}}\frac{1}{1+\eta_{\mathrm{N}}}\right) - \frac{1}{\tau_{\mathrm{N}}}\frac{\tilde{\lambda}_{\mathrm{N}}\tilde{\theta}_{\mathrm{N}}}{1+\eta_{\mathrm{N}}} \tag*{(23)}

for spin-to-charge conversion. To assess how spin current leakage affects the extraction of the interfacial Rashba parameter, Figure 4 shows the ratio αR/αR2D\alpha_{\mathrm{R}}/\alpha_{\mathrm{R}}^{2\mathrm{D}} as a function of the nonmagnetic metal conductivity for (a) spin-to-charge and (b) charge-to-spin conversion experiments. For these simulations, we assumed ξsc=ξcs=50%\xi_{\mathrm{sc}}=\xi_{\mathrm{cs}}=50\%. Clearly, in the limit of an insulating nonmagnetic metal (σN→0\sigma_{\mathrm{N}}\to0), αR→αR2D\alpha_{\mathrm{R}}\to\alpha_{\mathrm{R}}^{2\mathrm{D}} and in the limit of highly conducting nonmagnetic metal, the extracted value of αR\alpha_{\mathrm{R}} increases substantially. Interestingly, spin-to-charge and charge-to-spin conversion lead to very different behaviors. When ξsc\xi_{\mathrm{sc}} is used to quantify the Rashba parameter, the value of αR\alpha_{\mathrm{R}} saturates when ηN≈1\eta_{\mathrm{N}}\approx1. The maximum value obtained in (a) is about

Ratios between actual and nominal Rashba parameters

Figure 4. (Color online) Ratios between the actual value of the Rashba parameter in a magnetic bilayer and its nominal when only interfacial transport is assumed, for (a) spin-to-charge and (b) charge-to-spin conversion experiments.

αRαR2D→1+(τsfτN)(1−η~Nξsc(dN+t2D))(24)\frac{\alpha_{\mathrm{R}}}{\alpha_{\mathrm{R}}^{2\mathrm{D}}}\to1+\left(\frac{\tau_{\mathrm{sf}}}{\tau_{\mathrm{N}}}\right)\left(1-\frac{\widetilde{\eta}_{\mathrm{N}}}{\xi_{\mathrm{sc}}(d_{\mathrm{N}}+t_{2\mathrm{D}})}\right) \tag*{(24)}

which yields a saturation value of the order of 2.2 for our set of parameters. In contrast, using the charge-to-spin conversion efficiency ξcs\xi_{\mathrm{cs}}, the extracted parameter is proportional to the conductivity of the nonmagnetic metal,

αRαR2D→σNdNσ2D1τsf+1τF+1τN11+ηN1τsf+1τF(25)\frac{\alpha_{\mathrm{R}}}{\alpha_{\mathrm{R}}^{2\mathrm{D}}}\to\frac{\sigma_{\mathrm{N}}d_{\mathrm{N}}}{\sigma_{2\mathrm{D}}}\frac{\frac{1}{\tau_{\mathrm{sf}}}+\frac{1}{\tau_{\mathrm{F}}}+\frac{1}{\tau_{\mathrm{N}}}\frac{1}{1+\eta_{\mathrm{N}}}}{\frac{1}{\tau_{\mathrm{sf}}}+\frac{1}{\tau_{\mathrm{F}}}} \tag*{(25)}

Based on these observations, one can draw several conclusions. First, neglecting the potential spin current leakage and its associated SHE can lead to major discrepancies between the extracted interfacial Rashba parameter and the real one. A critical knob in the present theory is the spin transparency ηN\eta_{\mathrm{N}}. The proper estimate of interfacial spin-charge interconversion requires the accurate determination of this coefficient, i.e., spin relaxation length and interfacial conductance. Finally, it is remarkable that the Rashba parameter extracted from spin-to-charge conversion experiments is much less sensitive to spin current leakage than when using charge-to-spin conversion experiments. This result suggests that spin-pumping is a better tool than spin-orbit torque for the extraction of the interfacial Rashba parameter.

Conclusion

In order to solve controversial puzzles in spin-charge interconversion experiments, we have developed a phenomenological model that accounts for both interfacial SGE and bulk SHE effect. We find that when interfacial spin-charge interconversion is present, the leakage of spin current into the adjacent nonmagnetic metal substantially affects the overall spin-charge conversion efficiency, even in the absence of SHE. Most interestingly, the spin-to-charge and charge-to-spin conversions are affected differently: converting a charge current into a spin current is much more sensitive to the spin current absorption in the nonmagnetic metal due to the backflow associated with spin relaxation. In contrast, the charge current induced by a spin current is much more robust against spin current leakage.

This observation opens two interesting avenues. First, it clarifies the recent experimental reports on interconversion in topological insulators and explains why charge-to-spin conversion efficiencies tend to be orders of magnitude larger than spin-to-charge conversion efficiencies [20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33]. In addition, it also indicates that the value of the interfacial Rashba parameter extracted from spin-pumping experiments are, a priori, much more trustable than values extracted from spin-orbit torque experiments.

Acknowledgments

A.M. acknowledges support from the ANR ORION project, Grant No. ANR-20-CE30-0022-01 of the French Agence Nationale de la Recherche as well as the ANR MNEMOSYN project, Grant No. ANR-21-GRF1-0005. S.S. acknowledges support from the National Natural Science Foundation of China, Grant No. 12104032.

References

  1. [1]A. Manchon, J. Zelevný, M. Miron, T. Jungwirth, J. Sinova, A. Thiaville, K. Garello, and P. Gambardella, Current-induced spin-orbit torques in ferromagnetic and antiferromagnetic systems, Review of Modern Physics 91, 035004 (2019).DOI
  2. [2]W. Han, Y. C. Otani, and S. Maekawa, Quantum materials for spin and charge conversion, npj Quantum Materials 27, 3 (2018).DOI
  3. [3]T. Nan, D. C. Ralph, E. Y. Tsymbal, and A. Manchon, Emerging materials for spin-charge interconversion, APL Materials 9, 120401 (2021).DOI
  4. [4]K. Kondou and Y. Otani, Emergence of spin–charge conversion functionalities due to spatial and time-reversal asymmetries and chiral symmetry, Frontiers in Physics 11, 10.3389/fphy.2023.1140286 (2023).DOI
  5. [5]E. Saitoh, M. Ueda, H. Miyajima, and G. Tatara, Conversion of spin current into charge current at room temperature: Inverse spin-hall effect, Applied Physics Letters 88, 182509 (2006).DOI
  6. [6]O. Mosendz, J. E. Pearson, F. Y. Fradin, G. E. W. Bauer, S. D. Bader, and A. Hoffmann, Quantifying spin hall angles from spin pumping: Experiments and theory, Phys. Rev. Lett. 104, 046601 (2010).DOI
  7. [7]J.-C. Rojas-Sánchez, N. Reyren, P. Laczkowski, W. Savero, J.-P. Attane, C. Deranlot, M. Jamet, J.-M. George, L. Vila, and H. Jaffrès, Spin pumping and in- verse spin hall effect in platinum: The essential role of spin-memory loss at metallic interfaces, Physical Review Letters 112, 106602 (2014).pubmed.ncbi.nlm.nih.gov/24679318
  8. [8]V. T. Pham, L. Via, G. Zahnd, A. Marty, W. Savero-Torres, M. Jamet, and J. P. Attané, Ferromagnetic/nonmagnetic nanostructures for the electrical measurement of the spin hall effect, Nano Letters 16, 6755 (2016).pubmed.ncbi.nlm.nih.gov/27712075
  9. [9]B. Heinrich, C. Burrowes, E. Montoya, B. Kardasz, E. Girt, Y.-Y. Song, Y. Sun, and M. Wu, Spin pumping at the magnetic insulator (y)/normal metal (au) interfaces, Phys. Rev. Lett. 107, 066604 (2011).
  10. [10]C. Hahn, G. de Loubens, O. Klein, M. Viret, V. V. Naletov, and J. B. Youssef, Comparative measurements of inverse spin hall effects and magnetoresistance in yig/pt and yig/ta, Physical Review B 87, 174417 (2013).
  11. [11]C. Hahn, G. D. Loubens, M. Viret, O. Klein, V. V. Naletov, and J. B. Youssef, Detection of microwave spin pumping using the inverse spin hall effect, Physical Review Letters 111, 217204 (2013).DOI
  12. [12]C. O. Avci, A. Quindeau, C.-F. Pai, M. Mann, L. Caretta, A. S. Tang, M. C. Onbasli, C. A. Ross, and G. S. D. Beach, Current-induced switching in a magnetic insulator, Nature Materials 16, 309 (2017).DOI
  13. [13]P. Vaidya, S. A. Morley, J. V. Tol, Y. Liu, R. Cheng, A. Brataas, D. Lederman, and E. Barco, Subterahertz spin pumping from an insulating antiferromagnet, Science 368, 160 (2020).DOI
  14. [14]J. Li, C. B. Wilson, R. Cheng, M. Lohmann, M. Kavand, and W. Yuan, Spin current from sub-terahertz-generated antiferromagnetic magnons, Nature 578, 70 (2020).DOI
  15. [15]A. Hoffmann, Spin hall effects in metals, IEEE Transactions on Magnetics 49, 5172 (2013).DOI
  16. [16]J. Sinova, S. O. Valenzuela, J. Wunderlich, C. H. Back, and T. Jungwirth, Spin hall effect, Review of Modern Physics 87, 1213 (2015).arxiv.org/abs/1411.3249
  17. [17]J. C. Rojas-Sánchez, L. Vila, G. Desfonds, S. Gambarelli, J.-P. Attane, J. M. D. Teresa, C. Magén, and A. Fert, Spin-to-charge conversion using rashba coupling at the interface between non-magnetic materials., Nature Communications 4, 2944 (2013).
  18. [18]W. Zhang, M. B. Jungfleisch, W. Jiang, J. E. Pearson, and A. Hoffmann, Spin pumping and inverse rashba-edelstein effect in nife/ag/bi and nife/ag/sb, Journal of Applied Physics 117, 17C727 (2015).
  19. [19]S. Karube, K. Kondou, and Y. C. Otani, Experimental observation of spin-to-charge current conversion at non-magnetic metal/bi2o3 interfaces, Applied Physics Express 9, 033001 (2016).DOI
  20. [20]A. R. Mellnik, J. S. Lee, A. Richeardella, J. L. Grab, P. J. Mintun, M. H. Fischer, A. Vaezi, A. Manchon, E. a. Kim, N. Samarth, and D. C. Ralph, Spin-transfer torque generated by a topological insulator, Nature 511, 449 (2014).DOI
  21. [21]Y. Fan, P. Upadhyaya, X. Kou, M. Lang, S. Takei, Z. Wang, J. Tang, L. He, L.-T. Chang, M. Montazeri, G. Yu, W. Jiang, T. Nie, R. N. Schwartz, Y. Tserkovnyak, and K. L. Wang, Magnetization switching through giant spin-orbit torque in a magnetically doped topological insulator heterostructure., Nature Materials 13, 699 (2014).
  22. [22]P. Deorani, J. Son, K. Banerjee, N. Koirala, M. Brahlek, S. Oh, and H. Yang, Observation of inverse spin hall effect in bismuth selenide, Physical Review B 90, 094403 (2014).DOI
  23. [23]Y. Shiomi, K. Nomura, Y. Kajiwara, K. Eto, M. Novak, K. Segawa, Y. Ando, and E. Saitoh, Spin-electricity conversion induced by spin injection into topological insulators, Physical Review Letters 113, 196601 (2014).pubmed.ncbi.nlm.nih.gov/25415913
  24. [24]Y. Wang, P. Deorani, K. Banerjee, N. Koirala, M. Brahlek, S. Oh, and H. Yang, Topological surface states originated spin-orbit torques in bi2se3, Physical Review Letters 114, 257202 (2015).arxiv.org/abs/1505.07937
  25. [25]M. Jamali, J. S. Lee, J. S. Jeong, F. Mahfouzi, Y. Lv, Z. Zhao, B. Nikolic, K. A. Mkhoyan, N. Samarth, and J.-P. Wang, Giant spin pumping and inverse spin hall effect in the presence of surface spin-orbit coupling of topological insulator bi2se3, Nano Letters 15, 7126 (2015).
  26. [26]H. Wang, J. Kally, J. S. Lee, T. Liu, H. Chang, D. R. Hickey, K. A. Mkhoyan, M. Wu, A. Richardella, and N. Samarth, Surface-state-dominated spin-charge current conversion in topological-insulator ferromagnetic-insulator heterostructures, Physical Review Letters 117, 076601 (2016).
  27. [27]K. Kondou, R. Yoshimi, A. Tsukazaki, Y. Fukuma, J. Matsuno, K. S. Takahashi, M. Kawasaki, Y. Tokura, and Y. Otani, Fermi level dependent charge-to-spin current conversion by dirac surface state of topological insulators, Nature Physics 12, 1027 (2016).
  28. [28]J. C. Rojas-Sánchez, S. Oyarzún, Y. Fu, A. Marty, C. Vergnaud, S. Gambarelli, L. Vila, M. Jamet, Y. Ohtsubo, A. Taleb-Ibrahimi, P. L. Fèvre, F. Bertran, N. Reyren, J. M. George, and A. Fert, Spin to charge conversion at room temperature by spin pumping into a new type of topological insulator: α-sn films, Physical Review Letters 116, 096602 (2016).
  29. [29]Y. Wang, D. Zhu, Y. Wu, Y. Yang, J. Yu, R. Ramaswamy, R. Mishra, S. Shi, M. Elyasi, K. Jeong Teo, Y. Wu, and H. Yang, Room temperature magnetization switching in topological insulator-ferromagnet heterostructures by spin-orbit torques, Nature Communications 8, 1364 (2017).DOI
  30. [30]D. Mahendra, R. Grassi, J. Y. Chen, M. Jamali, D. R. Hickey, D. Zhang, Z. Zhao, H. Li, P. Quarterman, Y. Lv, M. Li, A. Manchon, K. A. Mkhoyan, T. Low, and J. P. Wang, Room-temperature high spin-orbit torque due to quantum confinement in sputtered bixse(1−x)films, Nature Materials 17, 800 (2018).
  31. [31]H. Wu, P. Zhang, P. Deng, Q. Lan, Q. Pan, S. A. Razavi, X. Che, L. Huang, B. Dai, K. Wong, X. Han, and K. L. Wang, Room-temperature spin-orbit torque from topological surface states, Physical Review Letters 123, 207205 (2019).DOI
  32. [32]F. Bonell, M. Goto, G. Sauthier, J. F. Sierra, A. I. Figueroa, M. V. Costache, S. Miwa, Y. Suzuki, and S. O. Valenzuela, Control of spin-orbit torques by interface engineering in topological insulator heterostructures, Nano Letters 20, 5893 (2020).DOI
  33. [33]H. He, L. Tai, H. Wu, D. Wu, A. Razavi, T. A. Gosavi, E. S. Walker, K. Oguz, C. C. Lin, K. Wong, Y. Liu, B. Dai, and K. L. Wang, Conversion between spin and charge currents in topological-insulator/nonmagnetic-metal systems, Physical Review B 104, L220407 (2021).DOI
  34. [34]Q. Shao, G. Yu, Y. wen Lan, Y. Shi, M.-Y. Li, C. Zheng, X. Zhu, L.-J. Li, P. K. Amiri, and K. L. Wang, Strong rashba-edelstein effect-induced spin-orbit torques in monolayer transition metal dichalcogenide/ferromagnet bilayers, Nano Letters 16, 7514 (2016).
  35. [35]D. MacNeill, G. M. Stiehl, M. H. D. Guimaraes, R. A. Buhrman, J. Park, and D. C. Ralph, Control of spin-orbit torques through crystal symmetry in wte2/ferromagnet bilayers, Nature Physics 13, 300 (2017).DOI
  36. [36]S. Shi, S. Liang, Z. Zhu, K. Cai, S. D. Pollard, Y. Wang, J. Wang, Q. Wang, P. He, J. Yu, G. Eda, G. Liang, and H. Yang, All-electric magnetization switching and dzyaloshinskii–moriya interaction in wte2/ferromagnet heterostructures, Nature Nanotechnology 14, 945 (2019).DOI
  37. [37]F. Xue, C. Rohmann, J. Li, V. Amin, and P. Haney, Unconventional spin-orbit torque in transition metal dichalcogenide-ferromagnet bilayers from first-principles calculations, Physical Review B 102, 014401 (2020).DOI
  38. [38]Q. Xie, W. Lin, S. Sarkar, X. Shu, S. Chen, L. Liu, T. Zhao, C. Zhou, H. Wang, J. Zhou, S. Gradečak, and J. Chen, Field-free magnetization switching induced by the unconventional spin-orbit torque from wte2, APL Materials 9, 051114 (2021).DOI
  39. [39]W. S. Torres, J. F. Sierra, L. A. Benítez, F. Bonell, M. V. Costache, and S. O. Valenzuela, Spin precession and spin hall effect in monolayer graphene/pt nanostructures, 2D Materials 4, 041008 (2017).arxiv.org/abs/1709.01854
  40. [40]C. K. Safeer, N. Ontoso, J. Ingla-Aynés, F. Herling, V. T. Pham, A. Kurzmann, K. Ensslin, A. Chuvilin, I. Robredo, M. G. Vergniory, F. D. Juan, L. E. Hueso, M. R. Calvo, and F. Casanova, Large multidirectional spin-to-charge conversion in low-symmetry semimetal mote2 at room temperature, Nano Letters 19, 8758 (2019).DOI
  41. [41]L. A. Benítez, W. S. Torres, J. F. Sierra, M. Timmermans, J. H. Garcia, S. Roche, M. V. Costache, and S. O. Valenzuela, Tunable room-temperature spin galvanic and spin hall effects in van der waals heterostructures, Nature Materials 19, 170 (2020).DOI
  42. [42]I. Shin, W. J. Cho, E. S. An, S. Park, H. W. Jeong, S. Jang, Y. J. Baek, S. Y. Park, D. H. Yang, J. H. Seo, G. Y. Kim, M. N. Ali, S. Y. Choi, H. W. Lee, J. S. Kim, S. D. Kim, and G. H. Lee, Spin-orbit torque switching in an all-van der waals heterostructure, Advanced Materials 34, 2101730 (2022).arxiv.org/abs/2102.09300
  43. [43]Q. Xie, W. Lin, J. Liang, H. Zhou, M. Waqar, M. Lin, S. L. Teo, H. Chen, X. Lu, X. Shu, L. Liu, S. Chen, C. Zhou, J. Chai, P. Yang, K. P. Loh, J. Wang, W. Jiang, A. Manchon, H. Yang, and J. Chen, Rashba–edelstein effect in the h-bn van der waals interface for magnetization switching, Advanced Materials 34, 2109449 (2022).DOI
  44. [44]W. Y. Choi, I. C. Arango, V. T. Pham, D. C. Vaz, H. Yang, I. Groen, C. C. Lin, E. S. Kabir, K. Oguz, P. Debashis, J. J. Plombon, H. Li, D. E. Nikonov, A. Chuvilin, L. E. Hueso, I. A. Young, and F. Casanova, All-electrical spin-to-charge conversion in sputtered bixse1-x, Nano Letters 22, 7992 (2022).DOI
  45. [45]N. Ontoso, C. K. Safeer, F. Herling, J. Ingla-Aynés, H. Yang, Z. Chi, B. Martin-Garcia, I. Robredo, M. G. Vergniory, F. D. Juan, M. R. Calvo, L. E. Hueso, and F. Casanova, Unconventional charge-to-spin conversion in graphene/mote2 van der waals heterostructures, Physical Review Applied 19, 014053 (2023).DOI
  46. [46]E. Lesne, Y. Fu, S. Oyarzun, J. C. Rojas-Sánchez, D. C. Vaz, H. Naganuma, G. Sicoli, J. P. Attané, M. Jamet, E. Jacquet, J. M. George, A. Barthélémy, H. Jaffrès, A. Fert, M. Bibes, and L. Vila, Highly efficient and tunable spin-to-charge conversion through rashba coupling at oxide interfaces, Nature Materials 15, 1261 (2016).DOI
  47. [47]Y. Wang, R. Ramaswamy, M. Motapothula, K. Narayanapanpillai, D. Zhu, J. Yu, T. Venkatesan, and H. Yang, Room-temperature giant charge-to-spin conversion at the strio3-laalo3 oxide interface, Nano Letters 17, 7659 (2017).DOI
  48. [48]D. C. Vaz, P. Noël, A. Johansson, B. Göbel, F. Y. Bruno, G. Singh, S. Mckeown-walker, F. Trier, L. M. Vicente-arche, A. Sander, S. Valencia, P. Bruneel, M. Vivek, M. Gabay, N. Bergeal, F. Baumberger, H. Okuno, A. Barthélémy, A. Fert, L. Vila, I. Mertig, J. philippe Attané, and M. Bibes, Mapping spin-charge conversion to the band structure in a topological oxide two-dimensional electron gas, Nature Materials 18, 1187 (2019).DOI
  49. [49]F. Trier, D. C. Vaz, P. Bruneel, P. Noe, A. Fert, L. Vila, J. philippe Attane, A. Barthelemy, M. Gabay, H. Jaffres, and M. Bibes, Electric-field control of spin current generation and detection in ferromagnet-free srtio3 - based nanodevices, Nano Letters 20, 395 (2020).DOI
  50. [50]F. Gallego, F. Trier, S. Mallik, J. Bréhin, S. Varotto, L. M. Vicente-Arche, T. Gosavy, C. C. Lin, J. R. Coudevylle, L. Iglesias, F. Casanova, I. Young, L. Vila, J. P. Attané, and M. Bibes, All-electrical detection of the spin-charge conversion in nanodevices based on srtio3 2-d electron gases, Advanced Functional Materials , 2307474 (2023).
  51. [51]F. Trier, P. Noël, J. V. Kim, J. P. Attané, L. Vila, and M. Bibes, Oxide spin-orbitronics: spin–charge interconversion and topological spin textures, Nature Reviews Materials 7, 258 (2022).DOI
  52. [52]F. Calavalle, M. Suárez-Rodríguez, B. Martín-García, A. Johansson, D. C. Vaz, H. Yang, I. V. Maznichenko, S. Ostanin, A. Mateo-Alonso, A. Chuvilin, I. Mertig, M. Gobbi, F. Casanova, and L. E. Hueso, Gate-tuneable and chirality-dependent charge-to-spin conversion in tellurium nanowires, Nature Materials 21, 526 (2022).DOI
  53. [53]P. Noël, F. Trier, L. M. V. Arche, J. Bréhin, D. C. Vaz, V. Garcia, S. Fusil, A. Barthélémy, L. Vila, M. Bibes, and J. philippe Attané, Non-volatile electric control of spin–charge conversion in a srtio 3 rashba system, Nature 580, 483 (2020).DOI
  54. [54]S. Varotto, L. Nessi, S. Cecchi, J. Slawińska, P. Noël, S. Petrò, F. Fagiani, A. Novati, M. Cantoni, D. Petti, E. Albisetti, M. Costa, R. Calarco, M. B. Nardelli, M. Bibes, S. Picozzi, J. P. Attané, L. Vila, R. Bertacco, and C. Rinaldi, Room-temperature ferroelectric switching of spin-to-charge conversion in germanium telluride, Nature Electronics 4, 740 (2021).DOI
  55. [55]C. Rinaldi, S. Varotto, M. Asa, J. Sławinska, J. Fujii, G. Vinai, S. Cecchi, D. D. Sante, R. Calarco, I. Vobornik, G. Panaccione, S. Picozzi, and R. Bertacco, Ferroelectric control of the spin texture in gete, Nano Letters 18, 2751 (2018).DOI
  56. [56]G. Bihlmayer, P. Noël, D. V. Vyalikh, E. V. Chulkov, and A. Manchon, Rashba-like physics in condensed matter, Nature Reviews Physics 4, 642 (2022).DOI
  57. [57]E. L. Ivchenko and G. E. Pikus, New photogalvanic effect in gyrotropic crystals, Pis’ma Zh. Eksp. Teor. Fiz 27, 604 (1978).
  58. [58]F. Vasko, Spin splitting in the spectrum of two-dimensional electrons due to the surface potential, Pis’ma Zh. Eksp. Teor. Fiz 30, 574 (1979).
  59. [59]V. M. Edelstein, Spin polarization of conduction electrons induced by electric current in two-dimensional asymmetric electron systems, Solid State Communications 73, 233 (1990).DOI
  60. [60]S. Zhang and A. Fert, Conversion between spin and charge currents with topological insulators, Physical Review B 94, 184423 (2016).
  61. [61]R. Dey, N. Prasad, L. F. Register, and S. K. Banerjee, Conversion of spin current into charge current in a topological insulator: Role of the interface, Physical Review B 97, 174406 (2018).DOI
  62. [62]H. Isshiki, P. Muduli, J. Kim, K. Kondou, and Y. Otani, Phenomenological model for the direct and inverse edelstein effects, Physical Review B 102, 184411 (2020).DOI
  63. [63]E. Mishchenko, A. Shytov, and B. Halperin, Spin current and polarization in impure two-dimensional electron systems with spin-orbit coupling, Phys. Rev. Lett. 93, 226602 (2004).pubmed.ncbi.nlm.nih.gov/15601109
  64. [64]a. a. Burkov, Theory of spin-charge-coupled transport in a two-dimensional electron gas with rashba spin-orbit interactions, Phys. Rev. B 70, 155308 (2004).DOI
  65. [65]W.-G. Wang, M. Li, S. Hageman, and C. L. Chien, Electric-field-assisted switching in magnetic tunnel junctions, Nat Mater 11, 64 (2012), 10.1038/nmat3171.DOI
  66. [66]A. Brataas, Y. Tserkovnyak, G. E. W. Bauer, and P. J. Kelly, Spin pumping and spin transfer (Oxford University Press, New York, 2012).arxiv.org/abs/1108.0385

Paper details

vlnc:W_kkvvba6r
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 8

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