Positronium is a pure leptonic atom in which an electron and a positron are bound by electromagnetic forces. Its ordinary annihilation channels are precisely predicted: para-positronium decays into two photons, ortho-positronium into three. If an invisible particle (for example, a dark photon or an axion-like state) appears in the decay, part of the energy is carried away unnoticed, and the event looks like missing energy. Previous searches, including the ETH Zurich experiment with a 4π BGO calorimeter, reached a limit of 4.3×10⁻⁷ on the probability of invisible decay. The KAPAE II work is part of efforts to improve this limit by an order of magnitude, but it stands out by its approach: the detector is designed specifically to search for missing energy rather than to study visible decays, and it operates deep underground at cryogenic temperatures.
The KAPAE II detector is built around an array of 25 BGO (bismuth germanate) crystals measuring 30×30×150 mm, surrounding a central region with a sodium-22 source and a trigger made of a 375 µm thick PEN film. The film registers the positron, while the BGO crystals register gamma rays with energies of 511 and 1275 keV. Geant4 simulations were used to optimize the geometry and trigger thickness: a compromise was sought between positron detection efficiency and minimal distortion of the gamma radiation. Light from each crystal was read out by a 16-pixel silicon photomultiplier (SiPM) array with a preamplifier, and the data were digitized at 62.5 MHz. The detector is placed in a cryogenic chamber with copper and lead shielding and installed in the Yemi underground laboratory at a depth of 1000 meters, where the muon flux is suppressed by more than 330,000 times.
After cooling to −37.5 °C, the detector's energy resolution improved: at 511 keV it was 11.7% (compared with 14.8% at room temperature), and at 1275 keV it was 6.5%. For a single crystal with a diffuse surface, a resolution of 10.8% was achieved at 662 keV, noticeably better than the typical 15% for crystals in the ETH Zurich experiment. The main result is the suppression of background events: the positron-gamma coincidence rate at the surface was on the order of 1 Hz with bursts up to 65 Hz, and underground it fell below 3×10⁻⁴ Hz, that is, by more than three orders of magnitude. After correcting for optical light leakage between neighboring crystals, the total energy spectrum agrees well with Geant4 simulations, confirming the detector's readiness to search for invisible decays.
This work differs from others in the archive in that it brings a full cycle to practical implementation: from geometry modeling to stable operation underground at cryogenic temperature. Unlike experiments with segmented detectors, KAPAE II uses a compact calorimeter with a reduced number of channels, which lowers the probability of false events with missing energy. The setup does not merely test theoretical predictions; it creates a tool capable of improving the existing limit on the probability of invisible positronium decay by an order of magnitude. This widens the search window for light dark-sector particles that are invisible in other experiments.
Building on the achieved sensitivity, data collection is planned to set a new limit on the probability of invisible positronium decay at the level of 10⁻⁸. Further improvement may come from longer data collection, optimized event selection, and more accurate background modeling. It may also be possible to extend the search to partially invisible decays, where a single photon with anomalous energy is registered. In the longer term, the technique could be transferred to other light exotic states predicted by hidden-sector models.
The KAPAE II detector will influence elementary particle physics, especially searches for light dark-sector particles such as dark photons and axion-like particles. Its results will be useful for refining models beyond the Standard Model and for planning future positronium experiments.
The immediate step is a long data collection run in the underground laboratory and analysis of events with missing energy. Then comes comparison with simulations and setting a new limit on the probability of invisible decay.
The search for invisible positronium decays is directly linked to one of the main unsolved problems in particle physics: the nature of dark matter. If dark matter consists of light particles that interact weakly with ordinary matter, they could show up in precisely such rare decays. Improving the limit by an order of magnitude will either reveal a signal or substantially constrain the class of hidden-sector models.
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 work describes the construction and underground installation of the KAPAE II detector for hunting invisible positronium decays. The geometry of a 25-crystal BGO calorimeter with a PEN trigger was optimized, achieving an energy resolution of 11.7% at 511 keV and background suppression by more than three orders of magnitude.
Where its strength is
A full cycle from simulation to a working installation with record energy resolution and deep background suppression, which opens the path to a sensitivity of 10⁻⁸.
Where it could go next
Worth trying
Why this may matter
Reaching a sensitivity of 10⁻⁸ will either reveal new particles or place strong constraints on models with invisible positronium decay channels, which matters for physics beyond the Standard Model.
Nearby works in our archive
The archive is sparse around this work: the closest review is only 0.55, against a usual 0.64. It looks like few have drilled nearby.
Positronium is the lightest known atom: its mass is 2000 times smaller than that of a hydrogen atom. The lifetime of ortho-positronium in vacuum is about 142 nanoseconds, but in the detector it annihilates faster because of interactions with matter.
The idea of invisible particles carrying away energy echoes the "dark matter" of science fiction, for example in Liu Cixin's novel The Three-Body Problem, where an invisible substance affects physical processes. In a real experiment, however, the goal is a rigorous search for specific candidates, not fictional substances.