⚛️ nuclear experiments

On the importance of cosmic-ray background in the Atomki anomaly

This paper presents Geant4 simulations demonstrating that cosmic-ray muon backgrounds in Atomki's e+e- pair spectrometers can produce statistically significant excesses in opening-angle distributions that closely mimic the reported 17 MeV boson anomaly, suggesting that cosmic rays may be a dominant source of the observed signal rates.

Original authors: Hicham Benmansour, Gianluigi Boca, Gianluca Cavoto, Marco Chiappini, Elia G. Grandoni, Luca Galli, Giovanni Gallucci, Angela Papa, Francesco Renga, Antoine Venturini, Cecilia Voena

Published 2026-09-17

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Hicham Benmansour, Gianluigi Boca, Gianluca Cavoto, Marco Chiappini, Elia G. Grandoni, Luca Galli, Giovanni Gallucci, Angela Papa, Francesco Renga, Antoine Venturini, Cecilia Voena

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the subatomic world, particles are constantly colliding, breaking apart, and reassembling in ways that reveal the fundamental rules of nature. Sometimes, these collisions produce unexpected results: a tiny bump in a graph where physics predicts a smooth line. Such anomalies can be the first whisper of a new force or a previously unknown particle, a discovery that would rewrite textbooks. For several years, a team of physicists at the Atomki institute in Hungary has reported seeing exactly this kind of bump. They observed pairs of electrons and positrons (the antimatter twin of the electron) flying apart at a specific angle after a nuclear reaction, suggesting the brief existence of a new, light particle they named X17. This claim has sparked intense debate, as independent experiments have struggled to confirm it, leaving the scientific community in a state of cautious curiosity. The question remains: is this a glimpse of new physics, or is it a trick of the experimental setup?

A new study by researchers from Italy and Switzerland suggests that the answer might lie not in new particles, but in the sky above. The team used powerful computer simulations to recreate the exact geometry of the Hungarian experiments, focusing on a background noise that is often overlooked: cosmic rays. These are high-energy particles, mostly protons and atomic nuclei, that rain down on Earth from deep space. When they hit the atmosphere, they create showers of secondary particles, including muons, which can penetrate deep underground and pass right through detectors. The researchers wanted to see if these passing muons could accidentally mimic the signal of a new particle, creating a false alarm that looks exactly like the one reported by Atomki.

To do this, the team built a detailed digital twin of the Hungarian detectors using a software package called Geant4, which is designed to track how particles move through matter. They modeled two different versions of the Atomki apparatus: a five-arm setup used in 2016 and a six-arm version used later. These machines consist of telescopes arranged in a circle around a target, designed to catch pairs of particles flying out in opposite directions. The researchers then simulated millions of cosmic muons raining down on these virtual detectors, tracking exactly where they hit and how much energy they deposited. They were looking for a specific pattern: two arms of the detector firing at the same time, with the particles flying apart at an angle of roughly 140 degrees, which is the signature of the alleged new particle.

The results were striking. In the simulations, cosmic muons passing through the detectors naturally created a pile-up of events at that exact 140-degree angle. This happened because of the specific way the detector arms were arranged. When a muon passed through two non-adjacent arms of the six-arm detector, the geometry of the setup and the way the energy was shared between the detectors conspired to make the opening angle look like 140 degrees. This effect was strongest when the total energy of the two hits fell within the specific range where the new particle was expected to appear. The simulation showed that these cosmic events also tended to have a balanced energy split between the two arms, another key feature of the reported signal. When the researchers looked at the five-arm detector, they found a similar effect, with cosmic muons creating a significant excess of events at the same angle and energy.

The study went a step further to see how this background compared to the actual signal the Hungarian team was trying to measure. They calculated the expected rate of the real nuclear reaction and compared it to the rate of cosmic muon hits, adjusting the numbers to match the typical running conditions of the Hungarian experiment. They found that in the critical energy window where the anomaly was reported, the number of cosmic muon events was actually larger than the number of real nuclear events. In fact, the cosmic background accounted for more than half of all the events in that specific region. This means that if the method used to subtract the cosmic background was even slightly off, it could easily leave behind a "ghost" signal that looks like a new particle. The researchers noted that the Hungarian team relies on taking data when the particle beam is turned off to estimate this background, but their simulations suggest that small errors in this subtraction process could create the very bump that was interpreted as new physics.

The paper does not claim to have disproven the existence of the new particle, nor does it say the Hungarian team made a mistake. Instead, it highlights a critical vulnerability in this type of measurement. The study demonstrates that the specific arrangement of the detector arms, combined with the constant rain of cosmic rays, creates a natural pattern that perfectly overlaps with the signal of interest. The researchers argue that the features of the reported anomaly—its specific angle, its energy range, and how it changes when the energy is slightly different—are all things that cosmic muons can reproduce on their own. This suggests that the origin of the anomaly might be a subtle interplay between the detector's geometry and the cosmic background, rather than a new fundamental particle.

Ultimately, the work serves as a call for greater precision in how these experiments handle background noise. The authors suggest that future searches for such anomalies need to be even more rigorous in distinguishing between real signals and cosmic impostors. They recommend using active shields to block cosmic rays and collecting much more data with the beam turned off to understand the background with higher certainty. While the question of the X17 particle remains open, this study provides a compelling alternative explanation: that the universe's constant shower of cosmic rays, passing unnoticed through the detector, might have been the one creating the illusion all along.

Technical Summary: On the importance of cosmic-ray background in the Atomki anomaly

Problem Statement
The Atomki group has reported anomalous excesses in the $e^+e^-$ opening-angle distribution ( $\theta_{open}$ ) from nuclear transitions in $^8$ Be and $^4$ He, interpreted as evidence for a new 17 MeV boson ( $X17$ ). These excesses appear as bumps above the expected Internal Pair Conversion (IPC) background near $\theta_{open} \simeq 140^\circ$ (for $^8$ Be) and $\simeq 115^\circ$ (for $^4$ He). While independent experiments have yielded conflicting results, a critical gap in the Atomki analyses is the lack of a fully efficient cosmic-ray veto. The existing analyses rely on beam-off data for cosmic background subtraction. This paper investigates whether cosmic-ray muons, passing through the specific geometries of the Atomki spectrometers, can naturally produce coincident events that mimic the topological features (angular position, energy asymmetry, and energy sum) of the reported $X17$ signal.

Methodology
The authors performed detailed Geant4 11.1 simulations of the cosmic-ray muon background for both the five-arm (2016) and six-arm (2021) pair spectrometers operated by the Atomki group.

Geometry Implementation: Full detector geometries were modeled, including position-sensitive detectors (Multi-Wire Proportional Chambers for the five-arm setup; Double-Sided Silicon Strip Detectors for the six-arm setup) and plastic scintillator calorimeters. Passive elements like the beam pipe and target assembly were largely omitted, while active volumes and scintillator dimensions were implemented based on published specifications.
Muon Generation: Cosmic muons were sampled from the Gaisser parametrization (100 MeV to 100 GeV) and propagated through the detector. The simulation tracked energy deposits in sensitive volumes to reconstruct "hits."
Event Selection: The analysis focused on two-arm coincidences ( $N_{hit}=2$ ). Key observables included the scalar energy sum ( $E_{sum}$ ), energy asymmetry ( $E_{asym}$ ), and the reconstructed opening angle ( $\theta_{open}$ ).
Normalization: For the five-arm setup, the cosmic background was normalized to typical Atomki running conditions ( $I_p = 1.0 \, \mu\text{A}$ , $\sim300$ hours) and compared against IPC events generated using the Zhang–Miller model for the $^7\text{Li}(p, e^+e^-)^8\text{Be}$ reaction.

Key Results

Six-Arm Spectrometer ( $^8$ Be and $^4$ He):
- $^8$ Be Window ( $E_{sum} \in [16, 20]$ MeV): The simulation reveals a statistically significant peak in the $\theta_{open}$ distribution near $140^\circ$ for symmetric pairs ( $|E_{asym}| < 0.5$ ). This peak arises from "next-to-adjacent" arm pairs ( $\Delta\phi = 120^\circ$ ) where the muon traverses the scintillators. The peak vanishes for asymmetric pairs and is suppressed in lower energy windows ( $E_{sum} \in [12, 16]$ MeV).
- $^4$ He Window ( $E_{sum} \in [18, 22]$ MeV): A distinct peak emerges near $120^\circ$ , consistent with the Atomki observation. The simulation attributes this to a specific next-to-adjacent topology at higher energies. The target offset (25 mm) in the $^4$ He experiment shifts the distribution, creating a broad enhancement in the $110^\circ$ – $140^\circ$ region.
- Correlation: A strong negative correlation exists between $E_{sum}$ and $\theta_{open}$ . Muons traversing the scintillators centrally deposit energy over a long path length (high $E_{sum}$ ) but produce hit centroids that are displaced, reducing the reconstructed opening angle below the nominal azimuthal separation. Conversely, muons clipping the outer edges of the arm pair deposit less energy (low $E_{sum}$ ) but produce increased $\theta_{open}$ , above the nominal separation.
Five-Arm Spectrometer ( $^8$ Be):
- A similar excess near $140^\circ$ is reproduced in the signal window ( $E_{sum} \in [16, 20]$ MeV), driven by the $\Delta\phi = 150^\circ$ arm configuration.
- Quantitative Impact: When normalized to 300 hours of data, the cosmic background contribution in the signal window is found to be comparable to, or even exceeding, the IPC rate. The ratio of cosmic-to-IPC events in the signal window is above unity.
- Background Suppression: In the background window ( $E_{sum} \in [12, 16]$ MeV), the angular excess is significantly reduced (by a factor of 2 to 5), mirroring the behavior reported by Atomki where the excess disappears outside the resonance energy.

Significance and Claims
The paper does not claim to definitively prove that the Atomki anomaly is caused by cosmic rays, nor does it settle the question of the $X17$ boson's existence. Instead, the authors argue that their results highlight the critical importance of a robust cosmic-ray treatment in these specific measurements.

The primary significance of the work lies in demonstrating that:

Topological Mimicry: The discrete azimuthal arrangement of the detector arms and the dimensions of the plastic scintillators naturally produce angular peaks in the cosmic background that align with the reported signal positions ( $140^\circ$ and $120^\circ$ ).
Feature Reproduction: The cosmic background reproduces key characteristics of the reported excesses, including the dependence on energy asymmetry (suppression for asymmetric pairs) and the disappearance of the peak in off-resonance energy windows.
Sensitivity to Subtraction: Because the cosmic contribution constitutes a large fraction (potentially >50%) of the total event rate in the signal window, the paper concludes that the Atomki results are highly sensitive to the accuracy of the cosmic background subtraction. Small errors in normalizing beam-off data (due to gain drifts or statistical fluctuations) could generate spurious residuals of the magnitude reported by Atomki.

The authors call for detailed descriptions of beam-off studies supporting the search for anomalies in the $(\theta_{open}, E_{sum})$ distribution and suggest that future measurements should employ active cosmic vetoes or extended beam-off statistics to disentangle potential new physics from geometric cosmic-ray artifacts.

More like this