Any Light Particle Searches with ALPS II: first science results

The light-shining-through-a-wall experiment ALPS II at DESY in Hamburg searched for axions and similar lightweight particles in its first science campaign from February to May 2024. No evidence for the existence of such particles was found. For pseudoscalar bosons like the axion, with masses below about 0.1 meV, we achieved a limit for the di-photon coupling strength of 1.5e-9 1/GeV at a 95% confidence level. This is more than a factor of 20 improvement compared to all previous similar experiments. We also provide limits on photon interactions for scalar, vector and tensor bosons. An achievement of this first science campaign is the demonstration of stable operation and robust calibration of the complex experiment. Currently, the optical system of ALPS II is being upgraded aiming for another two orders of magnitude sensitivity increase.

💡 Research Summary

The paper reports the first scientific results of the ALPS II (Any Light Particle Search II) experiment, a light‑shining‑through‑a‑wall (LSW) setup located in the former HERA tunnel at DESY. The campaign ran from February to May 2024 and aimed to detect axion‑like particles (ALPs), hidden photons, and other weakly interacting slim particles (WISPs) by converting photons into such particles in a strong magnetic field, allowing them to pass through an opaque wall, and reconverting them back into photons on the far side.

The experimental apparatus consists of two strings of twelve straightened HERA superconducting dipole magnets, each providing a magnetic field of 5.318 T over a length of 8.826 m, for a total magnetic length of about 123 m on each side of the wall. The magnets are spaced by 0.936 m gaps, and the two strings are separated by roughly 6.29 m. A high‑power 1064 nm laser (≈25 W) is injected into the production region; its polarization can be set either parallel or perpendicular to the magnetic field to probe pseudoscalar (axion‑like) or scalar bosons, respectively.

Only the regeneration cavity (RC) behind the wall was operated in a resonant configuration for the first run, achieving a power‑build‑up factor β≈7 000. The optical layout uses three auxiliary lasers (high‑power, auxiliary, and reference) together with a local oscillator to maintain resonance and to generate a heterodyne beat note at a well‑defined offset frequency (33 free‑spectral‑range offsets). The beat signal is detected by a photodiode (PD_science) and analyzed in the frequency domain.

Data were collected for 580 000 s with the wall open (allowing a tiny fraction of the high‑power laser light to leak into the RC for calibration) and for 1 060 000 s with the wall closed (the actual search mode). The analysis compares the power measured in the heterodyne channel with the wall open (P_open) to that with the wall closed (P_γ). Because many systematic effects cancel in the ratio P_γ/P_open, the conversion probability P_{γ↔WISP} can be extracted directly.

The spectral power distribution shows a peak at zero frequency offset, but its width is far larger than expected for a genuine WISP signal, indicating that stray light around the wall dominates the observed feature. Background estimation uses nearby frequency bins and a non‑central χ² model. No excess above background is observed.

From the null result the authors set 95 % confidence level upper limits on the photon‑WISP coupling. For pseudoscalar bosons (axion‑like particles) with masses below ~0.1 meV they obtain |g_{ϕγγ}| < 1.5 × 10⁻⁹ GeV⁻¹, a factor of more than 20 improvement over previous LSW experiments such as OSQAR. Comparable limits are derived for scalar, vector (hidden photon kinetic mixing ε), and massive spin‑2 tensor bosons.

The paper also emphasizes the successful demonstration of long‑term stability of the magnet strings, the vacuum system (pressure < 10⁻⁹ mbar), and the complex laser‑frequency‑locking scheme. These achievements validate the feasibility of the full ALPS II design, which will later include a resonant production cavity in addition to the regeneration cavity, further increasing the effective magnetic length and optical power. The planned upgrades aim at an additional two orders of magnitude improvement in sensitivity, potentially reaching coupling limits near 10⁻¹¹ GeV⁻¹.

In summary, the first ALPS II run confirms that LSW experiments remain a powerful laboratory probe of ultra‑light, weakly coupled particles, and sets the most stringent laboratory limits to date on axion‑photon couplings in the sub‑meV mass range. The ongoing upgrades promise to explore previously inaccessible regions of WISP parameter space, with important implications for dark‑matter, dark‑energy, and beyond‑Standard‑Model physics.