Direct searches for axion dark matter in the mass range around a millielectronvolt face a fundamental problem: traditional microwave cavities (such as ADMX) lose efficiency at terahertz frequencies, while optical and plasmonic schemes require either bulky mechanical tuning or excessively large volumes. This work joins recent proposals to use quantum materials and resonant structures for axion detection, but neighboring approaches (dielectric haloscopes, phonon-polariton devices, axion quasiparticles) are limited either by tuning complexity or by a scarcity of suitable materials. Here, a well-established platform of semiconductor quantum wells is proposed, where the resonant frequency is controlled by simply tilting the sample in a magnetic field.
The detector is based on a multilayer structure of GaAs/AlGaAs quantum wells grown by molecular beam epitaxy. In a strong magnetic field (up to 36 T in realistic configurations), the two-dimensional electron gas in the wells acquires an anisotropic dielectric permittivity that, at a certain cyclotron frequency, vanishes for one of the circular polarizations—this is the epsilon-near-zero (ENZ) point. The axion field, interacting with the magnetic field via the inverse Primakoff effect, induces an electric field in the medium, which is enhanced by tens to hundreds of times near ENZ. The signal is emitted from the resonator surface and focused onto a quantum-dot photodetector. The resonant frequency is tuned by rotating the sample: the normal component of the magnetic field changes the cyclotron frequency, while the parallel component controls the strength of axion-photon conversion.
For three benchmark configurations (Config. 1: 2–5 meV, Config. 2: 0.6–2 meV, Config. 3: improved low-mass), signal boost factors are calculated, ranging from 70 to 390. With an integration time of 300 days, a broad scan can achieve sensitivity to the axion-photon coupling constant $g_{aγγ}$ on the order of 5×$10^{-14}$ GeV^{-1} for masses 0.6–5 meV, surpassing current CAST limits. The optimistic Config. 3 comes close to the theoretically preferred QCD axion band (DFSZ and KSVZ models) in the 0.6–2 meV range. Simultaneously, the detector is sensitive to dark photons with kinetic mixing χ > 6×$10^{-16}$ in the same mass range.
Unlike neighboring works that use either mechanically tunable dielectric stacks (MADMAX) or exotic topological materials with limited tunability, SQW ARE relies on the mature technology of semiconductor heterostructures with record mobility. The key difference is the ability to scan mass by simply rotating the sample in existing magnets with high field homogeneity, removing the main engineering bottleneck for the meV range. This makes the approach scalable and accessible to laboratories with varying resources.
The immediate next step is to fabricate a prototype MQW structure a few millimeters thick and test the resonant response in a real magnet. In parallel, quantum-dot photodetectors can be improved (reducing dark count to microhertz) and stronger magnets (up to 50 T) with larger working volumes can be explored. Machine learning could optimize the design of multilayer structures with thousands of layers for maximum signal enhancement.
Beyond axions, the same detector, without modification, searches for dark photons in the meV range and can also serve as a platform for studying magnetoplasmonic resonances in quantum wells.
Fabricate a test MQW structure with Config. 1 or 2 parameters, calibrate the ENZ resonance using ellipsometry in a strong magnetic field, then perform a trial scan over a narrow mass range.
For condensed matter physics, an open question remains the maximum achievable mobility and uniformity of the two-dimensional electron gas in thick multilayer structures: inhomogeneity in electron density or magnetic field directly broadens the resonance and reduces sensitivity. Also unresolved is the growth of thick (centimeter-scale) MQW structures without loss of interface quality.
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
The paper proposes the SQW ARE detector based on GaAs/AlGaAs semiconductor quantum wells. In a strong magnetic field, the two-dimensional electron gas forms a magnetoplasmonic resonator with a tunable terahertz response, where axion-photon conversion is enhanced near the epsilon-near-zero point. Calculations for three configurations show that sensitivity to g_{aγγ} on the order of 5×10^{-14} GeV^{-1} can be achieved in the mass range 0.6–5 meV, closing a critical gap in the search for the QCD axion.
Where its strength is
Using a mature semiconductor platform with resonance frequency controlled by simply rotating the sample in the magnetic field eliminates the mechanical tuning problem common to other approaches. The large signal amplification (up to 390) and simultaneous sensitivity to dark photons make the detector a versatile tool.
Where it could go next
Worth trying
Nearby works in our archive
The archive is sparse around this work: the closest review is only 0.56, against a usual 0.64. It looks like few have drilled nearby.
The detector's resonant frequency is tuned by simply tilting the sample in the magnetic field: a change in angle of a few hundredths of a degree shifts the resonance by a linewidth, allowing the axion mass to be scanned without a single moving mechanical part inside the cryostat.