---
title: Orbital Rashba-Edelstein Effect
url: https://www.emergentmind.com/topics/orbital-rashba-edelstein-effect
type: topic
---

# Orbital Rashba-Edelstein Effect

Searching arXiv for recent papers on the Orbital Rashba-Edelstein Effect and closely related topics.
arXiv search query: "orbital Rashba Edelstein effect orbitronics"
The orbital Rashba-Edelstein effect (OREE) is the generation of a homogeneous nonequilibrium orbital polarization or orbital magnetization by an applied electric field or charge current in a system with broken inversion symmetry and momentum-space orbital textures. In direct form it is commonly written as $\langle L_i\rangle=\chi^L_{ij}E_j$, while inverse formulations relate an orbital accumulation to a transverse charge current. OREE is the orbital analogue of the spin Edelstein effect, but its microscopic origin lies in orbital Rashba coupling produced by inter-orbital hybridization, crystal fields, and inversion-asymmetric hopping rather than in spin-orbit interaction alone; accordingly, several studies find that it can persist, or remain largely unchanged, when spin-orbit coupling is removed [2502.09719, 2505.21340, 2511.09511].

## 1. Definition and conceptual boundaries

The broader orbital Edelstein effect (OEE) is the linear-response generation of a homogeneous orbital polarization by a uniform charge current in crystals lacking inversion symmetry. In Rashba systems, this orbital response is commonly identified as the orbital Rashba-Edelstein effect. The underlying picture is that an electric field shifts the Fermi surface, thereby weighting an odd-in-$\mathbf{k}$ orbital texture and producing a nonzero net orbital moment. In nonmagnetic crystals with time-reversal symmetry, the equilibrium spin and orbital polarizations vanish; the Edelstein response is therefore a nonequilibrium Fermi-surface phenomenon [2505.21340, 2207.08698].

A useful distinction, developed explicitly for noncentrosymmetric conductors, is between Hall-type conversion and Rashba-Edelstein-type conversion. Hall-type conversion uses gradients of spin or orbital chemical potential and bulk Hall conductivities, whereas Rashba-Edelstein conversion is local and driven by a Rashba texture. In the macroscopic notation of inverse orbital conversion,
\[
\boldsymbol{\ell}=\eta_{c\rightarrow \ell}\,\mathbf{z}\times \mathbf{J}_c,
\qquad
\mathbf{J}_c=\eta_{\ell\rightarrow c}\,\mathbf{z}\times \boldsymbol{\ell},
\]
so direct and inverse OREE are defined as charge-to-orbital and orbital-to-charge conversion mediated by broken inversion symmetry and orbital-momentum locking [2511.09511].

This separation is important because OREE is not equivalent to the orbital Hall effect. The latter is a transverse transport response associated with Hall conductivities; the former is a local current-induced orbital accumulation. A related misconception is that OREE is merely a spin Edelstein effect rephrased in orbital language. Multiple studies instead treat the orbital channel as primary, with spin appearing through subsequent $\lambda\,\mathbf{L}\cdot\mathbf{S}$ conversion or through coexistence of orbital and spin textures [2505.21340, 1905.08279].

## 2. Microscopic origin and effective Hamiltonians

Minimal orbital Rashba couplings take several symmetry-dependent forms. In multiorbital oxide two-dimensional electron gases, a canonical expression is
\[
H_{\mathrm{OR}}=\gamma\,(k_xL_y-k_yL_x)=\gamma[(\mathbf{k}\times \mathbf{L})\cdot \hat{\mathbf z}],
\]
while interfacial metallic systems are frequently described by
\[
H_{\mathrm{OR}}=\alpha_{\mathrm{OR}}\,(\hat{\mathbf z}\times \mathbf{k})\cdot \mathbf{L}.
\]
For the LaAlO$_3$/SrTiO$_3$ (111) interface, the orbital Rashba term appears in a $t_{2g}$ $k\cdot p$ Hamiltonian as $\alpha_{\mathrm{OR}}(k_x\Lambda_5+k_y\Lambda_2)$, with additional anisotropic linear terms allowed once the symmetry is reduced from trigonal $C_{3v}$ to mirror-only $C_s$ below the tetragonal transition of SrTiO$_3$ [2207.08698, 2503.16319, 2502.09719].

The microscopic sources of these terms differ by platform but follow the same logic: inversion-asymmetric hopping mixes orbitals with different angular character. In KTaO$_3$(001), antisymmetric momentum-dependent hoppings couple Ta $5d$ orbitals, and the inclusion of $e_g$ states is essential to reproduce the large Rashba splitting of the mixed $d_{xz}/d_{yz}$ pair. In gated monolayer transition-metal dichalcogenides, an out-of-plane gate field breaks $\sigma_h$ and induces off-diagonal couplings controlled by $\gamma_1$ and $\gamma_2$, generating in-plane chiral orbital textures near $\Gamma$ and $K/K'$ valleys. In ferroelectric GeTe, the relative displacement of Ge and Te ions along [111] breaks inversion symmetry in the bulk and produces Rashba-like split bands with chiral orbital textures on the Fermi surfaces [2207.08698, 2509.25634, 2505.21340].

A foundational surface mechanism was formulated for $sp$ systems. In the BiAg$_2$ monolayer, $sp$ hybridization and a surface electric field yield an effective orbital Rashba Hamiltonian
\[
H_{\mathrm{OR}}(\mathbf{k})=\frac{\alpha_{\mathrm{OR}}}{\hbar}\,\hat{\mathbf L}\cdot(\hat{\mathbf z}\times \mathbf{k}),
\]
with
\[
\alpha_{\mathrm{OR}}=
\left.
\frac{\eta\, a\,\gamma_{sp}\,V_z(\mathbf{k})}{\Delta E_{sp}(\mathbf{k})}
\right|_{\mathbf{k}=0}.
\]
In that formulation the orbital Rashba effect is tied directly to the electric polarization of surface states, and the modern theory of orbital magnetization becomes essential for the magnitude and variation of the orbital textures near band crossings [1611.04674].

Interfacial metallic ferromagnets provide a distinct realization. In Co/Al(111), the large helical in-plane orbital texture at the interfacial Co layer originates from hybridization of Co $d_{z^2}/d_{zx}/d_{yz}$ states with Al $p_x/p_y$ surface states at $K$ under broken inversion symmetry. In that case the dominant orbital texture exists already without spin-orbit coupling, while spin-orbit coupling adds a smaller correction with higher-order winding and converts orbital accumulation into torque [2410.09957].

## 3. Symmetry, response tensors, and geometric structure

In semiclassical and Kubo formulations, OREE is a Fermi-surface response weighted by orbital texture and band velocity. For a generic observable $Q\in\{s,\ell\}$, one common expression is
\[
\chi_{ij}^{(Q)}=
e\tau\sum_n\int\frac{d^3k}{(2\pi)^3}\,
v_{n,j}(\mathbf{k})\,q_{n,i}(\mathbf{k})
\left(-\frac{\partial f_0}{\partial \varepsilon}\right),
\]
and closely related two-dimensional forms are used for oxide interfaces. Interband Kubo formulas make the geometric content explicit by linking the orbital Edelstein susceptibility to interband mixing, orbital Berry curvature, and orbital Berry-curvature dipoles [2505.21340, 2502.09719].

Crystal symmetry strongly constrains the allowed tensor structure. In ferroelectric GeTe and gated monolayer TMDs, $C_{3v}$ symmetry allows only in-plane antisymmetric components,
\[
\chi_{xy}=-\chi_{yx},
\]
with all other components vanishing in the nonmagnetic time-reversal-symmetric bulk or monolayer. Physically, a current along $\hat x$ induces an orbital polarization along $\hat y$, and vice versa. In contrast, the $C_s$ symmetry realized in tetragonal domains of LAO/STO(111) permits anisotropic orbital Rashba couplings, rotated principal axes, and off-diagonal conductivity elements even at zero magnetic field; the principal axes of the orbital response rotate by $120^\circ$ across $X$, $Y$, and $Z$ domains [2505.21340, 2509.25634, 2502.09719].

In noncentrosymmetric antiferromagnets, symmetry operates at the sublattice level. The induced local magnetization can contain staggered in-plane components and non-staggered out-of-plane components, and the spin and orbital susceptibilities can display Rashba-like or Dresselhaus-like symmetry depending on the magnetic configuration. A notable outcome is that the orbital Rashba-Edelstein response is often closer to an ideal Rashba form than its spin counterpart [1905.08279].

The geometric aspect of OREE is especially visible in systems where transport and nonlinear response share a common origin. In LAO/STO(111), anisotropic orbital Rashba coupling generates both zero-field conductivity anisotropy and a finite Berry-curvature dipole responsible for the non-linear Hall effect. Their simultaneous onset near the same temperature scale is used as evidence that anisotropic orbital textures, rather than spin alone, govern the low-energy response [2502.09719].

## 4. Electronic material platforms and comparative phenomenology

A foundational surface platform is BiAg$_2$, where first-principles and tight-binding calculations revealed chiral orbital textures without spin-orbit coupling and showed that the modern theory of orbital magnetization predicts a singular enhancement up to $1.01\,\mu_B$ near a band crossing, whereas the atom-centered approximation gives approximately $0.03\,\mu_B$. This comparison established that Berry-phase contributions are not a correction of detail but a controlling factor for orbital-response magnitudes near avoided crossings and Fermi-surface singularities [1611.04674].

Representative electronic platforms studied subsequently span oxide interfaces, ferroelectric Rashba semiconductors, and gated two-dimensional semiconductors.

| Platform | Symmetry or structure | Reported OREE-related feature |
|---|---|---|
| LaAlO$_3$/SrTiO$_3$(111) [2502.09719] | $t_{2g}$ 2DEG with $C_s$ domains | Domain-correlated current modulations of $\sim 8$–$11\%$; modulation vanishes near $35$ K; non-linear Hall response onsets near $40$ K |
| GeTe [2505.21340] | Bulk ferroelectric Rashba semiconductor, $C_{3v}$ | Orbital moment surpasses spin moment by one order of magnitude; orbital susceptibility about one order of magnitude larger than spin; OEE remains largely unchanged without SOC |
| KTaO$_3$(001) 2DEG [2207.08698] | Multiorbital Ta $5d$ interface bands | OEE dominates the total Edelstein response; predicted maxima exceed SrTiO$_3$ 2DEGs by $\sim 4\times$ for the orbital channel |
| AlO$_x$/SrTiO$_3$ [2006.14958] | Topological oxide 2DEG with Rashba-like split $t_{2g}$ pairs | Orbital Edelstein effect exceeds the spin Edelstein effect by more than one order of magnitude |
| Gated monolayer TMDs [2509.25634] | Gate-induced $D_{3h}\rightarrow C_{3v}$ symmetry lowering | $K_{\mathrm{orb}}\approx 10^{-8}$–$10^{-7}\,\mu_B\,\mathrm{V}^{-1}\mathrm{m}$; one to two orders of magnitude larger than benchmark systems |

These platforms also show that large Rashba splitting is not by itself a reliable proxy for a large orbital Edelstein response. In KTaO$_3$, the strongly split mixed $d_{xz}/d_{yz}$ band pair has $\alpha_R\approx 320\,\mathrm{meV}\cdot\text{\AA}$, yet it contributes surprisingly little to both spin and orbital Edelstein effects because its Rashba-like texture resides in a pseudo-spin basis and because compensation between quasi-degenerate components suppresses the physical spin and orbital moments. By contrast, $d_{xy}$-like bands with smaller $\alpha_R$ dominate large parts of the response [2207.08698].

Ferroelectric GeTe illustrates a different organizing principle: the orbital and spin textures both reverse when the ferroelectric polarization is reversed, so the sign of the Edelstein response is electrically switchable. In the bulk calculations, the orbital moment is up to $10\times$ larger than the spin moment around $Z$, the OEE does not change sign across the studied energy range, and the orbital response remains largely unaffected when spin-orbit coupling is switched off [2505.21340].

Gated monolayer TMDs extend OREE into a valley-structured setting. For electron doping the Edelstein response is dominated by the orbital channel, whereas for hole doping the orbital and spin contributions are comparable; for hole doping, both channels are strongly enhanced by a small amount of strain because compressive strain moves the $\Gamma$ valley toward the valence-band maximum. The reported values, $K_{\mathrm{orb}}\approx 10^{-8}$–$10^{-7}\,\mu_B\,\mathrm{V}^{-1}\mathrm{m}$, are stated to exceed previously studied systems by one to two orders of magnitude [2509.25634].

## 5. Interfacial torques, magnonics, ultrafast dynamics, and ordered states

At Co/Al interfaces, orbital Rashba textures become directly relevant for torque generation. First-principles calculations find chiral orbital textures at the interfacial Co layer with amplitudes reaching $0.6$–$0.8\,\mu_B$, an intraband orbital Edelstein susceptibility $\chi^{\mathrm{oREE}}_{xy}\approx 5.94\times 10^{-10}\,\hbar\,\mathrm{m/V}$ per atom, and a predominantly field-like torque that corresponds to $B_{\mathrm{FL}}\approx 2$–$3\,\mathrm{mT}$ for $E_x=2.5\times 10^4\,\mathrm{V/m}$. Inserting a single Pt layer reduces $\chi^{\mathrm{oREE}}_{xy}$ from $5.94$ to $1.56$ in the same units, a suppression of about $74\%$, and changes the integrated field-like torkance from $+0.37$ to $-0.08$ in units of $[e a_0]$, while a damping-like component emerges. The reported interpretation is a crossover from interfacial orbital field-like torque in Co/Al to heavy-metal-like spin damping-like torque when Pt is introduced [2503.16319].

Magnonic devices demonstrate the inverse side of the same physics. In Pt/CuO$_x$/YIG nonlocal structures with $t_{\mathrm{Pt}}=2\,\mathrm{nm}$, the first-harmonic nonlocal signal is enhanced by about $7$–$8\times$ and the second-harmonic signal by about $4$–$5\times$ relative to Pt-only electrodes, and the analysis yields $K^{D}_{\mathrm{CuOx}}\approx 0.4\,K^{I}_{\mathrm{CuOx}}$, implying that the inverse OREE is approximately $2.5\times$ stronger than the direct OREE in that device class. In Py/Cu$^\ast$ heterostructures, an orbital Rashba-Edelstein magnetoresistance with spin-Hall-magnetoresistance-like angular dependence is observed without heavy elements; fits to the Py-thickness dependence give effective spin diffusion and spin dephasing lengths in Py that are approximately larger by the factor of $2$ and $4$, respectively, than in Py/Pt bilayers [2401.01090, 2105.04495].

The ultrafast regime reveals that OREE is not confined to steady-state linear transport. For Au(001) excited by a femtosecond laser pulse with $\hbar\omega=3.0\,\mathrm{eV}$ and a $10\,\mathrm{fs}$ Lorentzian envelope, the induced $l_y(t)$ is an intrinsic nonlinear orbital Edelstein response. The orbital signal reaches a pronounced maximum at about $7.5\,\mathrm{fs}$ after the pulse maximum, shows strong beating afterward, and carries odd-harmonic content, whereas the associated longitudinal orbital current $j_x^{l_y}(t)$ oscillates at $2\omega$ [2505.02006].

OREE also generalizes to ordered states. In noncentrosymmetric antiferromagnets such as CuMnAs and Mn$_2$Au, the electrically induced local magnetization contains staggered orbital components; in CuMnAs with the moments along $c$, the staggered off-diagonal orbital response $\chi^L_{xy}$ is about $60$ times larger than the spin counterpart at dc and remains essentially unchanged when spin-orbit coupling is removed. In spin-singlet noncentrosymmetric superconductors with orbital Rashba coupling, the dissipationless magnetoelectric response driven by a supercurrent is typically orbital-dominated, and the spin Edelstein effect is usually one order of magnitude smaller than the orbital one, although it can be enhanced in selected multiband situations with larger atomic spin-orbit coupling [1905.08279, 2408.08151].

## 6. Open problems, controversies, and design principles

A persistent misconception is that OREE requires strong spin-orbit coupling or heavy elements. Several models and material calculations contradict that view directly. The orbital Edelstein effect in GeTe remains largely unchanged without spin-orbit coupling; the orbital Rashba texture in Co/Al is present already without spin-orbit coupling; and in a Rashba bilayer the OREE remains finite even for $\alpha_A=\alpha_B=0$ if there is interlayer asymmetry in effective masses and finite interlayer hopping. This suggests that inversion asymmetry and multiorbital or multilayer hybridization are the indispensable ingredients, whereas spin-orbit coupling primarily governs orbital-to-spin conversion and spin torque generation [2505.21340, 2410.09957, 2307.02872].

A second misconception is that a larger Rashba splitting automatically implies a larger orbital Edelstein response. The literature instead emphasizes multiband compensation, pseudospin structure, and Fermi-surface anisotropy. In KTaO$_3$, the most strongly Rashba-split pair contributes little; in Rashba bilayers the sign of the orbital polarization can reverse when layer localization interchanges; in LAO/STO(111) both the conductivity anisotropy and the nonlinear Hall effect have non-monotonic gate dependence; and in gated TMDs strain and valley occupation determine whether the large $\Gamma$-valley orbital response becomes active [2207.08698, 2307.02872, 2502.09719, 2509.25634].

Experimentally, direct detection of the nonequilibrium orbital polarization remains less mature than indirect transport inference. LAO/STO(111) links zero-field linear conductivity anisotropy, scanning SQUID current imaging, and nonlinear Hall transport to a common orbital Rashba origin, but the work explicitly notes that direct measurement of current-induced $\langle L\rangle$ by XMCD or NV-based magnetometry would be a compelling future validation. Similar caveats apply to many other platforms, where the orbital response is inferred from torques, magnetoresistance, or charge conversion rather than from a direct orbital probe [2502.09719].

Theoretical limitations are also recurrent. Many studies use constant relaxation-time approximations, clean-limit Kubo formulas with phenomenological broadening, or low-temperature Fermi-surface expressions. Disorder, vertex corrections, domain-wall transport, interatomic contributions to orbital moments, and explicit conductivity tensors for inverse conversion are often neglected or only parameterized. These approximations do not erase the qualitative pattern that orbital channels can dominate spin channels, but they do complicate quantitative comparison across materials and experiments [2505.21340, 2207.08698, 2503.16319, 2509.25634].

Reported design principles are correspondingly material-specific but consistent. Electrostatic gating, selective filling of $a_{1g}$ versus $e_{\pi g}$ subbands, strain engineering, and orientation engineering toward (111) oxide interfaces are proposed routes to enhance anisotropic orbital Rashba coupling and $\chi^L$ in LAO/STO. In KTaO$_3$, lifting $d_{xz}/d_{yz}$ compensation by in-plane anisotropy and exploiting heavy-element hosts strengthens interconversion. In Co/Al, maximizing interfacial charge asymmetry and avoiding Pt insertion are presented as ways to preserve OREE-driven field-like torque. A more macroscopic outlook, developed for GeTe, redefines the effective Rashba parameter through measurable susceptibilities,
\[
\bar{\alpha}_R=\mu_B\,\frac{\hbar}{e}\,\frac{\chi_s}{\Sigma_B},
\]
and concludes that Rashba-Edelstein conversion dominates over Hall-type conversion in realistic drift-diffusion modeling, with effective parameters that can be much smaller than near-$\Gamma$ Rashba fits. This suggests that experimentally anchored response coefficients, rather than band-edge splittings alone, will likely become the standard language for comparing OREE across materials [2502.09719, 2207.08698, 2503.16319, 2511.09511].

Source: https://www.emergentmind.com/topics/orbital-rashba-edelstein-effect