⚛️ high-energy experiments

Multimuon Events from Cosmic Rays with the ALICE Detector

This paper analyzes 62.5 days of cosmic ray data collected by the ALICE experiment between 2015 and 2018 to study multimuon events, finding that while current hadronic interaction models struggle to fully describe the data, the QGSJET-II-04 model best reproduces the observed muon multiplicity distribution and high-multiplicity event rates under the assumption of a pure iron primary composition.

Original author: Mario Sitta, for the ALICE Collaboration

Published 2026-11-04

📖 4 min read🧠 Deep dive

CC BY 4.0

Original author: Mario Sitta, for the ALICE Collaboration

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 author. For technical accuracy, refer to the original paper. Read full disclaimer

High above the Earth, a constant rain of invisible particles falls from the edge of the universe. These are cosmic rays, mostly atomic nuclei traveling at nearly the speed of light, that strike the atmosphere and collide with air molecules. When they hit, they shatter into a cascade of secondary particles, creating a shower that spreads out across the sky. Among these fragments are muons, unstable cousins of the electron that can travel deep underground before decaying. Scientists have long used these muons as messengers to understand the nature of the original cosmic rays and the violent physics that occurs when they hit the atmosphere. However, a persistent mystery has lingered: the standard models used to predict how these showers develop often fail to explain the rare, massive bursts of muons that arrive together in large groups. Understanding these events is crucial because they reveal whether our current theories of particle physics are complete or if something fundamental is missing from our picture of the universe.

At the Large Hadron Collider in Switzerland, the ALICE experiment is primarily designed to smash heavy ions together to recreate the conditions of the early universe. Yet, between 2015 and 2018, while the collider was not running, the detector sat quietly underground, acting as a giant net for cosmic rays. Buried 52 meters beneath the surface, the experiment is shielded by tons of rock that block out almost all other particles, allowing only the most energetic muons to reach the sensitive tracking chambers inside. During this period, the team collected data for 62.5 days of real observation time. They focused specifically on "multimuon events," which are rare moments when more than four muons arrive at the detector simultaneously. By analyzing thousands of these events, the researchers aimed to see if the existing computer models of cosmic ray showers could accurately predict how often these large groups appear and how many muons they contain.

The researchers first looked at the distribution of muon counts, examining events with between four and fifty muons. They compared their real-world measurements against three different computer simulations, each based on a different theory of how particles interact in the atmosphere. To test these theories, the simulations assumed two extreme possibilities for the composition of the incoming cosmic rays: either they are all light protons or they are all heavy iron nuclei. The results showed a clear divide. The models based on light protons consistently predicted far fewer muon groups than what the detector actually saw. However, when the simulations assumed the cosmic rays were made of heavy iron, one specific model, known as QGSJET-II-04, matched the real data remarkably well. The other two models, even when assuming heavy iron, still predicted too few muons, failing to reproduce the observed pattern. This suggests that the standard theories of particle interaction struggle to describe these showers unless the incoming cosmic rays are heavy and the interaction model is specifically tuned to handle them.

The team also investigated the most extreme events, those containing more than one hundred muons, which they termed high muon multiplicity events. In their 62.5 days of data, they recorded thirteen such events, with the largest single burst containing 287 muons. This translates to a rate of roughly one such massive event every 4.8 days. When they compared this observed rate to their simulations, the findings reinforced the earlier conclusion. The model assuming a heavy iron composition and using the QGSJET-II-04 interaction theory predicted a rate that aligned closely with the actual measurements. In contrast, the model assuming light protons predicted a rate that was far too low to match reality. The other two interaction models also fell short, with one predicting a rate so low it was statistically incompatible with the data. The estimated energy of the cosmic rays causing these massive bursts is around 100 quadrillion electron volts, an immense amount of energy concentrated in a single particle.

Ultimately, the study provides a strong indication that the cosmic rays responsible for these rare, high-multiplicity muon bursts are likely composed of heavy elements like iron, rather than light protons. While the data does not definitively prove the composition of all cosmic rays, it strongly suggests that the current standard models of particle physics require a heavy composition to correctly describe the behavior of these atmospheric showers. The success of the QGSJET-II-04 model in matching the data, while others failed, highlights that our understanding of how high-energy particles interact in the atmosphere is still evolving. The ALICE detector, sitting deep underground, has successfully captured these fleeting moments from the cosmos, offering a clear view that challenges and refines our theoretical maps of the subatomic world.

Technical Summary: Multimuon Events from Cosmic Rays with the ALICE Detector

Problem Statement
The paper addresses the challenge of characterizing high-multiplicity muon events generated by cosmic rays interacting with the Earth's atmosphere. While collider detectors have historically been used to study atmospheric muons (pioneered by LEP experiments), previous models struggled to accurately describe the rate of high-multiplicity muon-bundle events, even under assumptions of high flux and pure iron spectra. The ALICE experiment at the CERN Large Hadron Collider (LHC) conducted a dedicated cosmic-ray data-taking campaign (2015–2018) to investigate these phenomena, specifically focusing on events with more than four detected muons. The study aims to compare experimental data against Monte Carlo simulations using leading hadronic interaction models to determine which models and primary cosmic-ray compositions best reproduce the observed muon multiplicity distributions and event rates.

Methodology
The analysis utilized data collected during LHC Run 2 (2015–2018) over 62.5 days of live time, with the ALICE magnetic field active (0.5 T) and no circulating beams.

Detector Configuration: The study relied on the ALICE Time Projection Chamber (TPC), the primary tracking device, which offers an effective detection area of approximately 17 m² for cosmic muons after track length requirements. The TPC is located 52 m underground with 28 m of rock overburden, establishing an energy threshold of 16 GeV for vertical muons and absorbing electromagnetic and hadronic shower components.
Trigger and Selection: A dedicated trigger based on the Time Of Flight (TOF) detector required a back-to-back coincidence between upper and lower sectors (±3 coincidence), achieving 100% efficiency for events with $N_\mu \ge 4$ .
Reconstruction: A specific algorithm matched "up" and "down" track segments reconstructed separately by the TPC (due to its design for inward-tracking from the interaction point) into single muon tracks. Selection criteria included a zenith angle limit ( $0 < \theta < 50^\circ$ ), a minimum of 50 clusters per track, a momentum threshold of 0.5 GeV/ $c$ , and a parallelism condition ( $\cos \Delta\Psi > 0.990$ ) for multimuon events.
Simulation: Simulated events equivalent to 62.5 days of live time were generated using CORSIKA as the event generator. Three hadronic interaction models were tested: QGSJET-II-04, EPOS-LHC, and SIBYLL 2.3d. Two extreme primary cosmic-ray compositions were assumed: pure proton (light) and pure iron (heavy). The primary energy spectrum followed a power law ( $E^{-\gamma}$ ) with a spectral index of 2.70 below the knee ( $3 \times 10^{15}$ eV) and 3.03 above, covering the range $10^{14} < E < 10^{18}$ eV.

Key Contributions and Results
The study analyzed the Muon Multiplicity Distribution (MMD) in the range $4 < N_\mu < 50$ and the rate of High Muon Multiplicity (HMM) events ( $N_\mu > 100$ ).

Muon Multiplicity Distribution (MMD):
- The measured MMD shows a smooth distribution up to $N_\mu \approx 50\text{--}60$ , followed by 13 events with multiplicities exceeding 100.
- Comparisons between data and simulations reveal that QGSJET-II-04 is the only model capable of reproducing the measured MMD across the entire investigated interval, but only when assuming a heavy (pure iron) primary composition.
- Neither SIBYLL 2.3d nor EPOS-LHC could describe the distribution, even with a heavy composition, as both models exhibited a significant muon deficit.
- The analysis suggests that lower multiplicities correspond to average primary energies of $\approx 4 \times 10^{15}$ eV, increasing to $> 6 \times 10^{16}$ eV for multiplicities above 50.
Rate of High Muon Multiplicity (HMM) Events:
- During the 62.5-day period, 13 HMM events ( $N_\mu > 100$ ) were recorded, yielding a rate of $(2.4 \pm 0.7) \times 10^{-6}$ Hz (approximately one event every 4.8 days). The highest multiplicity event contained 287 muons.
- To minimize statistical fluctuations, simulations were extended to a 1-year equivalent live time, focusing on primaries in the $10^{16} < E < 10^{18}$ eV range.
- Model Comparison:
  - QGSJET-II-04 (Iron): Predicted a rate compatible with the data within uncertainties.
  - SIBYLL 2.3d: Predicted a lower rate but remained compatible with the data.
  - EPOS-LHC: Predicted a rate outside the $2\sigma$ uncertainty of the measured data.
  - All models predicted rates for pure proton compositions that were below the measured data.
- The estimated average energy for HMM events is approximately $10^{17}$ eV.

Significance and Claims
The paper concludes that the observed muon multiplicity distribution and the rate of high-multiplicity events are best explained by a scenario where primary cosmic rays have a heavy composition (dominated by iron) across the studied energy range. While the models generally struggle to precisely describe the trend of primary proton and iron fluxes as a function of energy, QGSJET-II-04 is identified as the sole model that reasonably reproduces both the MMD shape and the HMM event rate under the assumption of a heavy composition. The results from Run 2 are consistent with previous Run 1 measurements but with significantly reduced uncertainty, reinforcing the need for heavy primary compositions to explain the high-multiplicity muon flux observed by ALICE.

More like this