⚛️ phenomenology

Earth-scale searches for displaced vertices: KM3NeT meets the LHC

This paper proposes that the KM3NeT neutrino detector, specifically its ORCA component, can serve as a powerful Earth-scale search for long-lived dark particles produced in rare meson decays at the LHC, offering competitive sensitivity to existing limits without requiring new dedicated detectors.

Original authors: Ui Min, Wen Yin

Published 2026-09-07

📖 5 min read🧠 Deep dive

CC0 1.0

Original authors: Ui Min, Wen Yin

Original paper dedicated to the public domain under CC0 1.0 (http://creativecommons.org/publicdomain/zero/1.0/). ✨ 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

For decades, physicists have built massive machines to smash particles together, hoping to catch a glimpse of the universe's hidden secrets. These collisions create a shower of known particles, but they also might produce something entirely new: particles that are so shy they barely interact with anything around them. In the standard experiments, these elusive particles would zip right through the detectors without leaving a trace, appearing only as a missing piece of energy. However, if these particles are also very long-lived, they might travel a significant distance before finally decaying, or breaking apart, into visible debris. The challenge has always been how to catch them. They are too weak to be seen by the main detectors, yet they might travel too far to be caught by the small, specialized detectors built nearby.

A new approach suggests that we do not need to build a new detector to find these ghostly travelers. Instead, we can look to the deep sea. The paper proposes using a massive, existing underwater telescope designed to hunt for neutrinos—tiny, nearly massless particles that pass through the Earth as if it were transparent—as a giant trap for these long-lived particles. The idea is to treat the ocean floor as a distant screen, waiting for particles produced in a collider thousands of kilometers away to arrive and decay inside the water. If they do, they would leave a visible flash of light, a signal that the main collider experiments would have missed entirely. This strategy turns the vast distance between a particle accelerator and a deep-sea observatory into a powerful tool for discovery.

The researchers behind this study, working from Tokyo, focused on the Large Hadron Collider (LHC) in Europe and the KM3NeT/ORCA detector, a massive array of sensors currently being deployed in the Mediterranean Sea. They calculated that the LHC produces an enormous number of short-lived particles called mesons during its collisions. Occasionally, these mesons might decay into a new, hidden particle that is light and travels at nearly the speed of light. Because this hidden particle interacts so weakly with normal matter, it would pass through the Earth's crust and rock layers without stopping. The team calculated that if such a particle is produced, it could travel a significant distance from the LHC to the ORCA detector in the Mediterranean. If the particle is just the right size and speed, it would survive the journey and then decay inside the water volume of the detector, creating a visible shower of light that the sensors could record.

The study does not claim to have found these particles yet. Instead, it maps out exactly where to look and how sensitive the search could be. The researchers ran detailed simulations to see how many of these hidden particles might be produced and how many would actually reach the detector. They found that for certain types of particles, the current data from the LHC's recent run is already powerful enough to set strict limits on their existence. If the ORCA detector sees no unusual flashes of light coming from the direction of the LHC, it means these specific types of hidden particles cannot exist within a certain range of properties. This result is significant because it rivals the sensitivity of experiments that look for missing energy inside the main collider detectors, but it does so by looking for the particles' decay products far away.

The team also looked ahead to the High-Luminosity LHC, a future upgrade that will produce many more collisions. They projected that with this increased data, the sensitivity of the underwater detector would improve dramatically, potentially uncovering particles that are currently invisible to other methods. A key part of their work involved estimating the background noise—the natural flashes of light caused by cosmic rays and other natural phenomena that could mimic the signal. They determined that by using the precise timing of the LHC collisions and the specific direction of the incoming particles, they could filter out most of this noise. Even with the current, partially completed state of the detector, the method is robust enough to provide meaningful scientific constraints.

This approach represents a clever repurposing of existing infrastructure. The ORCA detector was originally built to study neutrinos, but the researchers showed that its location and size make it an accidental, yet perfect, far-away laboratory for studying long-lived particles. The geometry is crucial: the distance between the source and the detector is so great that only particles with a very specific lifetime would survive the trip and decay exactly within the detector's volume. If the decay happens too soon, the particle vanishes before reaching the sea; if it happens too late, it passes right through. The sweet spot is a distance of hundreds of kilometers, which aligns perfectly with the distance between the LHC and the Mediterranean.

The study concludes that this "Earth-scale" search is a viable and powerful strategy. It does not require building a new facility but rather using the Earth itself as a shield and a guide. By combining the immense energy of the LHC with the vast volume of the deep ocean, physicists can probe a region of particle physics that is difficult to access with traditional methods. While the paper does not announce a discovery, it establishes a clear path forward. It shows that if these long-lived particles exist, the deep sea might be the best place to find them, and that the tools to do so are already being built, waiting for the first signal to arrive from the other side of the world.

Technical Summary: Earth-scale searches for displaced vertices: KM3NeT meets the LHC

Problem and Motivation
Following the discovery of the Higgs boson, the search for physics beyond the Standard Model (BSM) continues, particularly for light, long-lived particles (LLPs). Conventional collider experiments often struggle to detect feebly interacting particles that travel macroscopic distances before decaying, as they frequently escape main detectors leaving only missing-energy signatures. While dedicated detectors (e.g., FASER, MATHUSLA) and atmospheric shower searches have been proposed, this paper investigates a novel strategy: utilizing existing large-volume detectors located hundreds of kilometers away from a collider as "far detectors." Specifically, the authors propose using the KM3NeT/ORCA neutrino detector, located approximately 382 km from CERN, to search for LLPs produced in proton-proton collisions at the LHC. The core hypothesis is that BSM particles produced in rare meson decays can survive the journey to ORCA and decay visibly within its instrumented volume, creating "geometrically long displaced vertices."

Methodology
The study employs a geometric and phenomenological framework to estimate the sensitivity of KM3NeT/ORCA to LLPs ( $X$ ) produced at the LHC.

Geometric Framework: The probability ( $P_X$ ) for a particle $X$ produced at the collider to enter the detector and decay visibly is derived based on the baseline distance ( $L$ ), the detector's effective volume ( $V_{eff}$ ), and the particle's lab-frame decay length ( $\lambda_X$ ). The detection probability is maximized when $\lambda_X \approx L$ . The authors calculate the required number of produced particles ( $N_{req}$ ) to observe a specific number of events, comparing various configurations (CERN to KM3NeT, SuperKEKB to Hyper-K, etc.). The CERN–ORCA geometry is identified as the most promising due to the favorable baseline and detector volume.
Production Mechanism: The analysis focuses on the production of $X$ via rare decays of mesons ( $M$ ) generated in inelastic $pp$ collisions at $\sqrt{s} = 13.6$ TeV (Run 3) and 14 TeV (HL-LHC). Using PYTHIA 8.317 simulations, the authors determine the transverse momentum ( $p_T$ ) and rapidity ( $y$ ) distributions for various mesons ( $\pi^0, \pi^\pm, \eta, \eta', \omega, \phi, K_S^0, K^\pm$ ). The flux of $X$ toward ORCA is calculated by integrating these distributions over the solid angle subtended by the detector.
Kinematics and Efficiency: The expected number of reconstructed decays ( $N_X^{ORCA}$ ) is computed by summing contributions from all LHC interaction points (ATLAS, CMS, ALICE, LHCb). The analysis imposes an energy cut ( $E_{vis} > 1$ GeV) to ensure detectability. For LHC Run 3, a luminosity-weighted suppression factor ( $f_{ORCA}^{Run3} \approx 0.22$ ) is applied to account for the detector's partial deployment during that period.
Background Estimation: A comprehensive background analysis is performed, considering four sources: atmospheric neutrinos ( $B_{\nu atm}$ ), atmospheric muons ( $B_{\mu atm}$ ), optical/instrumental noise ( $B_{noise}$ ), and neutrinos produced at the LHC ( $B_{\nu LHC}$ ).
- Atmospheric Neutrinos: Suppressed via directional selection (requiring arrival from the LHC direction, $\sim 1.35^\circ$ below the horizon), energy selection ( $1 < E < 5$ GeV), shower topology (vetoing muon tracks), and timing coincidence with LHC bunch crossings.
- Other Sources: Atmospheric muons and optical noise are estimated to be negligible after standard cuts. LHC-produced neutrinos are treated as a potential background, estimated via geometric scaling, though no dedicated flux calculation exists.

Key Contributions and Results

Sensitivity Projections: The paper presents 95% Confidence Level (C.L.) exclusion limits on the branching ratios of mesons decaying into $X$ $X$ (e.g., $M \to XX$ $M \to X X$ or $M \to \mu X$ $M \to μ X$ ) as a function of the Lorentz-invariant parameter $\sqrt{m_X \Gamma_X}$ $m_{X} Γ_{X}$ .
- Run 3: The projected sensitivity for LHC Run 3 data is competitive with existing limits on invisible meson decays for specific channels, particularly kaon decays. However, for $\pi^0$ decays, the paper notes that existing limits are so strong that the entire parameter region is already excluded.
- HL-LHC: Future projections with 3 ab $^{-1}$ (ATLAS/CMS) and 300 fb $^{-1}$ (LHCb) show significantly improved sensitivity, potentially surpassing current missing-energy constraints for various meson species.
Specific Channels:
- For $\pi^0 \to XX$ , existing limits are so strong that the entire parameter space is already excluded.
- For $K_S^0 \to \pi^0 X$ , the authors identify a potentially large, unexplored window. Using a conservative limit based on $K_S^0 \to \pi^0 \pi^0$ , they show that Run 3 data could set a much more stringent limit than currently available.
- The optimal sensitivity occurs at a decay length $\lambda_X \approx 382$ km, matching the CERN–ORCA baseline.
Background Handling: The study provides detailed estimates for background suppression factors. For the $K_S^0 \to XX$ benchmark, the expected background varies significantly by scenario: $B \approx 3.0$ in the realistic case and $B \approx 29.4$ in the conservative case. Consequently, the statistical significance ( $Z$ ) exceeds the 95% C.L. threshold ( $Z > 1.64$ ) only in the realistic scenario ( $Z \approx 1.77$ ), while the conservative scenario falls slightly below the nominal threshold ( $Z \approx 0.74$ ). The HL-LHC benchmark remains sensitive in both cases.

Significance and Claims
The paper claims that existing large-volume neutrino detectors can serve as effective far-detectors for LLP searches without the need for new dedicated infrastructure. This approach offers a complementary strategy to conventional collider searches and dedicated far-detector proposals.

Immediate Impact: The authors assert that LHC Run 3 data, when analyzed in conjunction with KM3NeT/ORCA, can already set competitive limits on very long-lived particles, specifically in kaon decays, provided background assumptions align with the realistic scenario.
Future Outlook: The method is presented as a scalable strategy applicable to future colliders (FCC-ee, FCC-hh, ILC). The authors advocate for the intentional design of future facilities (particularly FCC-hh) to optimize the alignment between interaction points and remote detectors, transforming the current "accidental" alignment between CERN and KM3NeT into a deliberate, optimized long-baseline LLP facility.
Modesty: The authors acknowledge limitations, noting that their efficiency estimates ( $\epsilon_{sig} = 1$ ) are idealized benchmarks. They emphasize that the angular and trigger responses to direct $X \to e^+e^-$ decays have not yet been simulated in ORCA, and the background estimates for LHC-produced neutrinos are order-of-magnitude approximations. The results are framed as sensitivity estimates for a new search strategy rather than definitive discovery claims.

More like this