Exciton interactions enable spin control for bright spin LEDs

Spin light-emitting diodes, devices that translate the orientation of an electron’s spin into the polarization of the light they emit, have long been touted as building blocks for next-generation displays, spin-photonic circuits and even spin-based communication channels. Yet their practical development has been hampered by a stubborn physical problem: at room temperature, the spin polarization injected into the light-emitting layer collapses far too quickly, long before the device can produce light that is both bright and strongly polarized. Now, researchers writing in Nature Photonics report a hybrid chiral perovskite architecture that confronts this problem head-on, achieving a record combination of brightness and spin polarization and pushing the endurance of these devices to levels that could finally make them technologically interesting.

The central performance metric for a spin light-emitting diode is a quantity the researchers call asymmetric electroluminescence brightness, denoted B_CP-EL. It is defined as the product of the electroluminescence dissymmetry factor, which quantifies the degree of circular polarization of the emitted light, and the luminance, which measures how bright the device actually is. Because these two figures of merit historically traded off against one another, the field has been stuck at values of roughly 10 to 1,000 candela per square meter. Pushing luminance higher typically means pumping in more electrical excitation, but higher excitation densities accelerate the very spin relaxation that polarizes the emission in the first place, so the dissymmetry factor collapses. The result is a ceiling that has constrained spin-LED development for years.

The team behind the new study traced this ceiling to the microscopic kinetics of spin relaxation, and in particular to the role of exciton–exciton interactions. In perovskite emitters, the optically active quasiparticles are excitons, bound pairs of electrons and holes. At low excitation densities, spin flips are relatively rare and the initial degree of spin polarization injected from a magnetic or chiral contact dominates the polarization of the emitted light. But as the density of excitons rises, exciton–exciton scattering events provide an efficient channel through which angular momentum is exchanged and spins are flipped. The spin-flip rate climbs steeply with excitation density, which is precisely the regime a bright LED must operate in. Understanding and controlling this crossover between the two regimes, the researchers show, is the key to unlocking high brightness without sacrificing spin polarization.

Their solution is an elegant piece of materials engineering: a hybrid chiral perovskite heterostructure in which achiral light-emitting regions are spatially distributed and separated by a wide-bandgap chiral spin injector. The chiral layers perform double duty. First, their handedness provides the spin-selective transport that generates polarized charge injection in the first place, exploiting the mechanism by which electrons of one spin orientation pass more readily through a chiral potential than electrons of the opposite orientation. Second, the wide bandgap keeps the chiral injector electronically distinct from the emitting regions, so that excitons form and recombine primarily in the achiral emitters while the chiral layers modulate the effective strength of exciton–exciton interactions that drive spin flips. By distributing the emitters, the design dilutes the excitation density experienced locally, suppressing the rapid rise of the spin-flip rate that would otherwise destroy the polarization at high current.

The measured results are striking. The heterostructure extends the spin-relaxation time into the nanosecond regime at room temperature, an order-of-magnitude regime shift compared with the picosecond-scale relaxation that plagues conventional spin emitters, while simultaneously preserving a photoluminescence quantum efficiency of 78 percent. That combination matters because a long spin lifetime is useless if the material emits light inefficiently; the photoluminescence quantum efficiency is a direct measure of how many excitations convert into photons rather than being lost to non-radiative decay. Achieving both simultaneously has been a long-standing challenge, since the heavy-atom compositions and strong spin-orbit coupling often used to manipulate spins tend to come at the cost of radiative efficiency.

Fabricated into complete light-emitting diodes, the hybrid structure delivers an asymmetric electroluminescence brightness of 13,084 candela per square meter, more than an order of magnitude beyond the previous envelope of 10 to 1,000 candela per square meter. The devices achieve a maximum electroluminescence dissymmetry factor of 0.2, meaning that roughly sixty percent of the emitted photons carry one circular polarization versus forty percent carrying the opposite handedness. For a room-temperature, electrically driven device, that level of polarization control is remarkable, and it opens the door to applications ranging from polarization-encoded displays to optical communication schemes in which information is carried in the spin degree of freedom of photons rather than in their intensity alone.

Stability, often the Achilles heel of perovskite optoelectronics, also fared well in the study. The researchers report an extrapolated half-lifetime exceeding 5,000 hours at an initial luminance of 100 candela per square meter. While extrapolated operational lifetimes must always be interpreted with care, the figure suggests that the heterostructure is not merely a laboratory curiosity but a platform with genuine potential for long-lived devices. Combined with the solution-processability that perovskites are known for, the durability results strengthen the case for chiral perovskite spin-LEDs as candidates for scalable manufacturing.

Beyond the headline numbers, the study makes a conceptual contribution through its kinetic analysis of the emission polarization. By modeling the coupled dynamics of exciton populations and spin relaxation, the researchers reveal a crossover in the dominant determinant of emission polarization. At low excitation densities, the polarization of the emitted light is set primarily by the initial spin polarization delivered by the injector; the spins simply have time to retain their orientation before recombining. At high excitation densities, by contrast, the spin-flip rate itself becomes the controlling parameter, because exciton–exciton interactions accelerate relaxation faster than recombination can harvest the polarization. This mechanistic insight reframes how spin-LED performance should be engineered: rather than maximizing injection polarization alone, device designers should focus on materials architectures that flatten the density dependence of the spin-flip rate, exactly what the distributed emitter–chiral injector geometry accomplishes.

The implications extend well beyond displays. Circularly polarized electroluminescence is a sought-after resource in quantum information science, where the angular momentum of photons encodes qubit states, and in spin photonics, where polarized light couples to the spin states of carriers in semiconducting channels. A room-temperature source of bright, spin-polarized light that operates on simple electrical injection could serve as an interface between electronic spin logic and optical interconnects, or as a compact source for chiral sensing and circularly polarized organic and perovskite photonics. The demonstration that exciton–exciton interactions, conventionally viewed as a loss channel for spin information, can be deliberately manipulated through heterostructure design provides a general design principle that could be exported to other material systems, including organic semiconductors, quantum dots and two-dimensional chiral crystals.

Challenges, of course, remain. A dissymmetry factor of 0.2, while impressive for a bright device, still falls short of the near-unity values needed for some quantum applications, and further work will be needed to push polarization higher without compromising luminance. The extrapolated lifetime will need to be validated under accelerated aging protocols, and integration into full-color display architectures will require equivalent performance from emitters across the visible spectrum. Nevertheless, the study represents a decisive step forward, demonstrating that the long-standing brightness–polarization trade-off in spin light-emitting diodes is not an intrinsic limitation but a controllable consequence of exciton kinetics. With the spin-relaxation bottleneck now relaxed at room temperature, bright spin-LEDs have moved from a theoretical aspiration to an engineering problem, one whose solution may reshape how future displays and quantum light sources are built.