Spin-LED: Spin-Modulated Light Emission

• Spin-LEDs are light-emitting diodes that convert the spin state of injected carriers into circularly polarized light through spin-selective radiative recombination.
• They utilize diverse architectures including III–V semiconductors, quantum dots, organic-inorganic hybrids, and 2D materials to achieve electrical helicity switching and zero-field operation.
• These devices find applications in optical communications, quantum cryptography, and biophotonics, with ongoing research aimed at improving modulation speed, efficiency, and integration.

A spin-modulated light-emitting device, or spin light-emitting diode (spin-LED), is a light-emitting diode in which the spin state of electrically injected carriers, or of the emitting excitonic channel, is converted into a measurable property of the emitted light, most commonly its circular polarization. In the III–V formulation, spin-polarized electrons recombine with holes under semiconductor optical selection rules, so the helicity of electroluminescence directly encodes carrier spin. In broader contemporary usage, closely related devices include atomically thin heterostructures in which gate-defined singlet and triplet excitons, 2D-magnet-controlled tunneling, magnetic-order-mediated excitonic transitions, or spin–valley-locked photonic modes determine the polarization, spectrum, or intensity of the emitted light (Nishizawa et al., 2018, Joe et al., 2020, Dang et al., 2024, Hautzinger et al., 2023).

1. Spin-to-photon conversion

The defining mechanism of a spin-LED is spin-selective radiative recombination. In GaAs-based structures, the conduction-band minimum has $J=1/2$, the valence-band maximum consists of heavy-hole and light-hole states with $J=3/2$, and optical selection rules map electron spin orientation onto photon helicity. In the lateral GaAs spin-LED, recombination of a spin-up electron with a heavy-hole or light-hole state produces $\sigma^+$ circularly polarized light along the spin quantization axis, whereas spin-down electrons produce $\sigma^-$ emission. The degree of circular polarization is defined as

$P_{\text{circ}}=\frac{I^+-I^-}{I^++I^-},$

with $I^+$ and $I^-$ the intensities of the two helicity components (Nishizawa et al., 2018).

The same basic logic extends beyond bulk III–V selection rules. In p-doped InGaAs/GaAs quantum-dot spin-LEDs, the measured circular polarization is written as

$$P_c=\alpha \cdot F \cdot P_{\text{inj}}, \qquad F=\frac{\tau_s^{\text{QD}}}{\tau_s^{\text{QD}}+\tau_r},$$

so the optical output reflects the injected spin polarization, the effective selection rules, and the competition between spin relaxation and recombination inside the dot (Giba et al., 2020). In organic spin-OLEDs, the spin degree of freedom enters through exciton statistics: unpolarized injection gives a singlet:triplet ratio of $1:3$, whereas in the idealized antiparallel spin-polarized case the ratio becomes $1:1$, increasing the radiative singlet fraction (Prieto-Ruiz et al., 2016). In MoSe$J=3/2$0/WSe$J=3/2$1 heterobilayers, the relevant emitting states are interlayer excitons, and electrical control selects spin-singlet or spin-triplet channels through band filling of CB$J=3/2$2 or CB$J=3/2$3 (Joe et al., 2020). In CrSBr, by contrast, electroluminescence is modulated because exciton energies and injection efficiency are both functions of magnetic order, so spin-flip and spin-canting transitions directly tune EL energy and intensity (Qin et al., 1 Aug 2025).

2. Architectures and material platforms

Spin-LEDs are not a single device architecture but a family of emitter classes that differ in how the spin degree of freedom is generated, transported, and read out. The historically central platform is the III–V semiconductor LED with a ferromagnetic injector and a tunnel barrier. Lateral GaAs-based devices use in-plane Fe/oxide contacts and side-facet emission so that the injected spin axis is parallel to the optical axis, eliminating the need for an external magnetic field during operation (Nishizawa et al., 2018). Earlier dual-electrode InGaAs devices used anti-parallel Fe spin-injection electrodes to demonstrate electrical helicity switching and continuous polarization blending (Nishizawa et al., 2014).

Other platforms replace the conventional metallic injector. A hybrid organic/inorganic structure uses the organic ferrimagnetic semiconductor V[TCNE]$J=3/2$4 as the spin source and a GaAs/AlGaAs quantum-well LED as the optical spin polarimeter, with heavy-hole and light-hole electroluminescence providing opposite-sign signatures of the injected spin current (Fang et al., 2010). A p-doped InGaAs/GaAs quantum-dot spin-LED uses a perpendicularly magnetized CoFeB/MgO injector and positively charged dots to obtain efficient zero-field operation in remanence (Giba et al., 2020). A diluted-magnetic-semiconductor formulation uses Zn$J=3/2$5Mn$J=3/2$6Se:Cl to inject spin-polarized electrons into an InGaAs quantum dot, producing circularly polarized single photons under electrical drive (Asshoff et al., 2011).

Atomically thin implementations broaden the concept further. In MoSe$J=3/2$7/WSe$J=3/2$8 van der Waals LEDs, spin selectivity is encoded in spin–valley locking, type-II alignment, and stacking-dependent optical selection rules rather than ferromagnetic contacts (Joe et al., 2020). In CrI$J=3/2$9/hBN/WSe$\sigma^+$0, the few-layer 2D magnet and the hBN tunnel barrier act as a spin filter for holes injected into monolayer WSe$\sigma^+$1, and the emitted helicity follows the CrI$\sigma^+$2 magnetic state (Dang et al., 2024). In CrSBr tunneling LEDs, the magnetic semiconductor itself is the light-emitting layer, and the optical output is governed by magnetic-order-mediated excitonic transitions and spintronic transport (Qin et al., 1 Aug 2025). A different zero-field route uses a chiral 2D hybrid perovskite, $\sigma^+$3, as a spin-selective interface on top of an AlGaInP multiple-quantum-well LED, enabling spin accumulation in a standard III–V platform without magnetic materials or applied fields (Hautzinger et al., 2023).

Platform	Spin-control element	Representative optical signature
Lateral GaAs/InGaAs spin-LED	Dual Fe/oxide injectors in anti-parallel remanence	Electrically switched $\sigma^+$4 side-facet EL (Nishizawa et al., 2014, Nishizawa et al., 2018)
QD III–V spin-LED	Perpendicular CoFeB/MgO injector	Zero-field circularly polarized EL from p-doped QDs (Giba et al., 2020)
DMS QD spin-LED	ZnMnSe giant-Zeeman spin injector	Circularly polarized single-photon emission (Asshoff et al., 2011)
Hybrid organic/inorganic spin-LED	V[TCNE]$\sigma^+$5 ferrimagnetic semiconductor	HH/LH-resolved EL polarization tracking magnetization (Fang et al., 2010)
Spin-OLED	Parallel/antiparallel ferromagnetic electrodes	MEL via singlet/triplet population control (Prieto-Ruiz et al., 2016)
TMD van der Waals LED	Spin–valley locking and gate-controlled band filling	Electrically tuned singlet/triplet interlayer exciton emission (Joe et al., 2020)
2D magnet/TMD spin-LED	CrI$\sigma^+$6/hBN spin filter	Gate-tunable EL helicity with sign reversal (Dang et al., 2024)
2D magnetic-semiconductor LED	Spin-flip and spin-canting transitions in CrSBr	Hysteretic and continuous EL modulation (Qin et al., 1 Aug 2025)
Chiral perovskite/III–V spin-LED	CISS at a chiral perovskite/AlGaInP interface	Room-temperature zero-field CP-EL (Hautzinger et al., 2023)

This diversity corrects a common misconception: spin-LEDs are not restricted to ferromagnetic metal injectors on III–V quantum wells. Semiconductor/semiconductor interfaces, organic ferrimagnets, chiral perovskites, intrinsic spin–valley systems, and 2D magnets all appear in the literature as routes to electrically controlled spin-dependent light emission (Fang et al., 2010, Joe et al., 2020, Hautzinger et al., 2023, Dang et al., 2024).

3. Modes of spin modulation

The most direct form of spin modulation is electrical helicity switching. In the dual-electrode lateral spin-LED, two Fe stripes with anti-parallel remanent magnetizations are driven by a two-channel current source. When square currents are applied $\sigma^+$7 out of phase, the device alternates between the two spin injectors and the emitted helicity switches correspondingly. The 2014 device demonstrated switching at $\sigma^+$8, while the room-temperature 2018 device resolved helicity inversion from $\sigma^+$9 up to $\sigma^-$0 (Nishizawa et al., 2014, Nishizawa et al., 2018). The same dual-injector architecture also supports analog tuning: by adjusting the current density ratio between the two electrodes, the net circular polarization is swept continuously from approximately $\sigma^-$1 to $\sigma^-$2 while the total EL intensity remains almost constant at room temperature (Nishizawa et al., 2018). In this context, $\sigma^-$3 does not denote a fixed linear polarization mode; it denotes a statistical mixture or average of opposite circularly polarized components (Nishizawa et al., 2014, Nishizawa et al., 2018).

A second mode is remanent or zero-field operation. The CoFeB/MgO quantum-dot spin-LED uses perpendicular magnetic anisotropy to inject spins along the optical axis in remanence, so the device operates at zero applied magnetic field while still producing a large $\sigma^-$4 (Giba et al., 2020). The chiral-perovskite/III–V spin-LED removes ferromagnetism altogether: spin-polarized holes are generated by chiral-induced spin selectivity as carriers traverse $\sigma^-$5, and the sign of the emitted circular polarization flips when the enantiomer is switched (Hautzinger et al., 2023).

A third mode is gate- or field-controlled selection of emitting channels. In MoSe$\sigma^-$6/WSe$\sigma^-$7, dual gates tune the Fermi level through WSe$\sigma^-$8 valence-band states and MoSe$\sigma^-$9 conduction subbands, thereby selecting neutral singlet, charged singlet, or triplet interlayer excitons (Joe et al., 2020). In CrI$P_{\text{circ}}=\frac{I^+-I^-}{I^++I^-},$0/hBN/WSe$P_{\text{circ}}=\frac{I^+-I^-}{I^++I^-},$1, a back gate tunes the few-layer CrI$P_{\text{circ}}=\frac{I^+-I^-}{I^++I^-},$2 magnetization at fixed magnetic field, and the electroluminescence helicity can be tuned efficiently, including sign reversal, because the CrI$P_{\text{circ}}=\frac{I^+-I^-}{I^++I^-},$3 magnetic state determines the spin polarization of injected holes (Dang et al., 2024). In CrSBr, the modulation variable is the magnetic phase of the emitter itself: in-plane spin-flip transitions produce hysteretic changes in EL peak energy and intensity, while out-of-plane spin-canting yields continuous spectral and intensity modulation with strong anisotropy (Qin et al., 1 Aug 2025). Organic spin-OLEDs provide yet another modulation channel: changing the magnetic configuration of two ferromagnetic electrodes alters singlet and triplet formation probabilities and produces a magneto-electroluminescence contrast between parallel and antiparallel states (Prieto-Ruiz et al., 2016).

4. Observables, measurement protocols, and theoretical descriptors

The principal observable in most spin-LEDs is circular polarization. In III–V emitters the relevant quantity is usually

$P_{\text{circ}}=\frac{I^+-I^-}{I^++I^-},$4

while TMD literature ხშირად uses the degree of circular polarization or degree of circularly polarized photoluminescence/electroluminescence, and photonic CP-OLED work often reports

$P_{\text{circ}}=\frac{I^+-I^-}{I^++I^-},$5

Spin-OLED literature supplements these with

$P_{\text{circ}}=\frac{I^+-I^-}{I^++I^-},$6

so the optical output can be compared directly with the magnetic configuration of the electrodes (Prieto-Ruiz et al., 2016, Deng et al., 2024).

Magneto-optical spectroscopy supplies the complementary spectroscopic fingerprints. In TMD interlayer-exciton devices, Zeeman splitting is analyzed through

$P_{\text{circ}}=\frac{I^+-I^-}{I^++I^-},$7

and the sign of the effective exciton $P_{\text{circ}}=\frac{I^+-I^-}{I^++I^-},$8-factor distinguishes spin-singlet and spin-triplet emission channels (Joe et al., 2020). In QD spin-LEDs, the decomposition

$P_{\text{circ}}=\frac{I^+-I^-}{I^++I^-},$9

makes explicit that the optical output depends on injector polarization, QD charge state, and the ratio of electron spin lifetime to recombination lifetime (Giba et al., 2020). In hybrid organic/inorganic devices, the observed optical polarization is written as $I^+$0 with $I^+$1, allowing an injected spin polarization to be inferred from heavy-hole and light-hole electroluminescence at the quantum well (Fang et al., 2010).

Experimentally, the measurement chain is usually polarization-resolved EL or PL spectroscopy. The lateral GaAs spin-LED uses a quarter-wave plate set at $I^+$2 or $I^+$3, a linear polarizer fixed at $I^+$4, and a photomultiplier to reconstruct $I^+$5, $I^+$6, and time-dependent $I^+$7 during helicity switching (Nishizawa et al., 2018). The V[TCNE]$I^+$8/GaAs hybrid device resolves heavy-hole and light-hole peaks and exploits their opposite-sign spin selection rules to separate true spin injection from magnetic circular dichroism backgrounds (Fang et al., 2010). The chiral-perovskite/III–V spin-LED performs a Hanle-type test, showing that the CP-EL decreases under transverse magnetic field, which supports a genuine spin-accumulation origin rather than optical artifacts (Hautzinger et al., 2023).

5. Representative performance regimes and uses

Reported performance spans room-temperature helicity switching, zero-field operation, single-photon emission, and high-dissymmetry cavity emitters. The room-temperature lateral spin-LED with dual Fe injectors achieves $I^+$9 for the 100 nm Fe electrode and $I^-$0 for the 30 nm electrode, supports helicity switching up to $I^-$1, and tunes the net polarization from about $I^-$2 to $I^-$3 at nearly constant EL intensity (Nishizawa et al., 2018). The earlier dual-SIE InGaAs device demonstrated $I^-$4 electrical switching and $I^-$5 at $I^-$6 (Nishizawa et al., 2014). The p-doped InGaAs/GaAs quantum-dot spin-LED reaches $I^-$7 in remanence up to $I^-$8 (Giba et al., 2020). A diluted-magnetic-semiconductor single-photon spin-LED attains $I^-$9 circular polarization at $$P_c=\alpha \cdot F \cdot P_{\text{inj}}, \qquad F=\frac{\tau_s^{\text{QD}}}{\tau_s^{\text{QD}}+\tau_r},$$0 and $$P_c=\alpha \cdot F \cdot P_{\text{inj}}, \qquad F=\frac{\tau_s^{\text{QD}}}{\tau_s^{\text{QD}}+\tau_r},$$1 at $$P_c=\alpha \cdot F \cdot P_{\text{inj}}, \qquad F=\frac{\tau_s^{\text{QD}}}{\tau_s^{\text{QD}}+\tau_r},$$2, with antibunching $$P_c=\alpha \cdot F \cdot P_{\text{inj}}, \qquad F=\frac{\tau_s^{\text{QD}}}{\tau_s^{\text{QD}}+\tau_r},$$3 under pulsed operation (Asshoff et al., 2011). The chiral-perovskite/III–V MQW device yields room-temperature zero-field CP-EL with degree of polarization up to $$P_c=\alpha \cdot F \cdot P_{\text{inj}}, \qquad F=\frac{\tau_s^{\text{QD}}}{\tau_s^{\text{QD}}+\tau_r},$$4 (Hautzinger et al., 2023).

In 2D and cavity-based platforms, the relevant metrics shift somewhat. CrI$$P_c=\alpha \cdot F \cdot P_{\text{inj}}, \qquad F=\frac{\tau_s^{\text{QD}}}{\tau_s^{\text{QD}}+\tau_r},$$5/hBN/WSe$$P_c=\alpha \cdot F \cdot P_{\text{inj}}, \qquad F=\frac{\tau_s^{\text{QD}}}{\tau_s^{\text{QD}}+\tau_r},$$6 spin-LEDs exhibit electroluminescence circular polarization up to $$P_c=\alpha \cdot F \cdot P_{\text{inj}}, \qquad F=\frac{\tau_s^{\text{QD}}}{\tau_s^{\text{QD}}+\tau_r},$$7 in optimized devices and electrically induced sign reversal of helicity by gate-tuning the CrI$$P_c=\alpha \cdot F \cdot P_{\text{inj}}, \qquad F=\frac{\tau_s^{\text{QD}}}{\tau_s^{\text{QD}}+\tau_r},$$8 magnetic state (Dang et al., 2024). CrSBr tunneling LEDs show EL modulation efficiencies of $$P_c=\alpha \cdot F \cdot P_{\text{inj}}, \qquad F=\frac{\tau_s^{\text{QD}}}{\tau_s^{\text{QD}}+\tau_r},$$9 for in-plane spin-flip control and $1:3$0 for out-of-plane spin-canting control (Qin et al., 1 Aug 2025). The spin-valley-locked CP-OLED based on photonic spin–orbit coupling reports narrowband emission of $1:3$1, $1:3$2, maximum luminance near $1:3$3, and $1:3$4 up to $1:3$5 (Deng et al., 2024). Organic spin-OLEDs, although not usually framed as semiconductor spin-LEDs, have demonstrated spin-valve effects up to $1:3$6 and magneto-electroluminescence enhancement of $1:3$7 at $1:3$8 in the antiparallel state (Prieto-Ruiz et al., 2016).

These functionalities motivate a wide application range. The lateral room-temperature spin-LED literature explicitly targets monolithic circularly polarized light sources, on-chip optical interconnects, spin-encoded optical communication, 3D displays, quantum information tasks, and biophotonics (Nishizawa et al., 2018). The quantum-dot single-photon device connects spin-LEDs to quantum cryptography by mapping $1:3$9 and $1:1$0 onto $1:1$1 and $1:1$2 states (Asshoff et al., 2011). A biomedical systems paper proposes an endoscope probe that combines spin-LEDs and spin-photodiodes at $1:1$3 to exploit angle-dependent circular-polarization scattering from tissue, with synchronous detection enabled by electrical helicity switching (Nishizawa et al., 2020). Hybrid structures also repurpose the LED as a sensitive spin polarimeter, as in the V[TCNE]$1:1$4/GaAs platform where inorganic electroluminescence is used to probe spin physics in an organic ferrimagnet (Fang et al., 2010).

6. Limitations, misconceptions, and research directions

Several constraints recur across the literature. In dual-electrode lateral devices, the measured $1:1$5 ceiling is not a fundamental recombination limit: the recombination time $1:1$6 implies intrinsic modulation at order $1:1$7, while earlier work likewise noted that GHz switching should be possible if LRC constraints are reduced (Nishizawa et al., 2018, Nishizawa et al., 2014). The same lateral architecture becomes unstable for arbitrary polarization control at high current density because minority-spin EL departs from linearity and can even decrease beyond $1:1$8 (Nishizawa et al., 2018). In quantum-dot spin-LEDs, the shallow confinement of the present InGaAs/GaAs dots causes strong thermal escape, and EL becomes undetectable beyond about $1:1$9 despite the persistence of sizeable $J=3/2$00 up to $J=3/2$01 (Giba et al., 2020). CrI$J=3/2$02-based devices remain limited to low temperature, require finite out-of-plane magnetic fields for state preparation, and rely on air-sensitive materials (Dang et al., 2024). Organic spin-OLEDs still face the long-standing difficulty that OLED operating voltages exceed the low-bias regime where spin injection is usually most efficient, even though interface engineering has pushed spin-valve behavior to $J=3/2$03 (Prieto-Ruiz et al., 2016).

A second class of issues concerns interpretation and scaling. One misconception is that $J=3/2$04 must imply linear polarization; dual-injector GaAs devices explicitly rule this out by showing that the zero point is a cancellation of opposite circularly polarized contributions rather than a fixed linear mode (Nishizawa et al., 2018). Another is that a spin-LED necessarily means carrier-spin injection from a ferromagnetic metal. The literature now includes organic ferrimagnetic semiconductors, chiral perovskites, 2D magnets, spin–valley excitonic LEDs, and photonic spin–orbit CP-OLEDs, which suggests that “spin-LED” has become a broader functional category centered on electrically controlled spin-dependent light emission rather than a single materials recipe (Fang et al., 2010, Joe et al., 2020, Hautzinger et al., 2023, Deng et al., 2024). Scaling introduces its own complications: spatially resolved magnetoluminescence in organic OLEDs finds intra-device Overhauser-field variations exceeding $J=3/2$05 and spatial correlations beyond $J=3/2$06, indicating that spin response can vary significantly within a nominal pixel (Pappas et al., 2021).

The main research directions follow directly from those bottlenecks. For III–V devices, thinner transport layers, optimized tunnel barriers, deeper confinement, and faster electronics are the obvious routes to higher room-temperature polarization and faster helicity modulation (Nishizawa et al., 2018, Giba et al., 2020). For 2D and van der Waals devices, the open problems are dynamic modulation speed, zero-field operation, higher ordering temperatures, and scalable integration with photonic circuitry (Joe et al., 2020, Dang et al., 2024, Qin et al., 1 Aug 2025). For organic and photonic spin-modulated emitters, the central tasks are improving reproducibility, controlling disorder, and combining high circular polarization with high efficiency and simple fabrication (Prieto-Ruiz et al., 2016, Deng et al., 2024). Across all platforms, the field is moving from spin injection as a diagnostic toward spin modulation as a device function: helicity switching, polarization blending, singlet/triplet channel selection, spin-valley routing, and magnetic-state readout are now explicit operating modes rather than secondary signatures.