⚛️ phenomenology

Probing the Axion--Photon--Dark Photon Interaction with LHC Mono-Photons and Fermi-LAT Gamma-Ray Lines

This paper investigates the axion-photon-dark photon interaction by deriving improved LHC mono-photon constraints that surpass LEP limits and projecting HL-LHC sensitivity, while demonstrating that Fermi-LAT gamma-ray line data provides complementary and often stronger constraints for heavy axion dark matter, collectively excluding the thermal relic scenario across the studied mass range.

Original authors: Chuan-Ren Chen, Takaaki Nomura, Van Que Tran

Published 2026-10-16

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Chuan-Ren Chen, Takaaki Nomura, Van Que Tran

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

The universe is filled with invisible threads connecting the known world to the unknown. Physicists have long suspected that the matter holding galaxies together—dark matter—is not made of the same stuff as stars or planets, but of a hidden sector of particles that rarely interact with ordinary light. Two leading candidates for these hidden particles are the axion, a ghostly particle originally proposed to solve a puzzle about why the universe prefers matter over antimatter, and the dark photon, a heavy cousin of the ordinary photon that might carry forces within this hidden realm. While these particles have never been directly seen, their existence could be revealed if they interact with each other in a specific way, creating a bridge between the visible and invisible worlds. If such a bridge exists, it would allow these hidden particles to occasionally transform into flashes of light that our most powerful telescopes and particle colliders could catch.

A team of researchers has now taken a fresh look at data from the Large Hadron Collider, the world's most powerful particle smasher, to hunt for signs of this specific interaction. They focused on a scenario where a collision produces a single, high-energy flash of light accompanied by a large amount of missing energy, a signature that would indicate a hidden particle has escaped detection. By carefully re-examining records from the CMS experiment at the collider, the scientists were able to set new, much stricter limits on how strongly axions and dark photons can talk to each other. Their analysis shows that for dark photons with masses up to 2.3 thousand times heavier than a proton, the interaction between these particles must be incredibly weak, far weaker than previously thought possible. This finding improves upon the best previous limits from older experiments by a factor of three or more, effectively closing the door on many theories that predicted these particles would interact more strongly.

The researchers did not stop at the current data. They also projected what would happen if the collider were upgraded to its future High-Luminosity version, which will smash particles together with even greater intensity. Their simulations suggest that this future machine could push the search for these interactions to even heavier masses, reaching up to three thousand times the mass of a proton. However, there is a limit to how heavy these particles can be before the mathematical tools used to describe them break down. The team found that their search remains valid up to a mass of about 2.3 thousand times the proton mass, beyond which the theory itself would need to be rewritten. This boundary is crucial because it tells us exactly where the current understanding of physics ends and where new, undiscovered physics must begin.

The study also turned its attention to the possibility that the axion itself is the dark matter filling the universe. If this is true, axions would not just be produced in collisions but would also be constantly annihilating each other in the vast spaces between stars, turning into pure energy in the form of gamma rays. The researchers compared their collider results with data from the Fermi-LAT space telescope, which scans the sky for these specific gamma-ray signals. They discovered a fascinating division of labor between the two types of experiments. For lighter axions, the collider is the better detective, able to rule out their existence more effectively. But as the axion gets heavier, approaching the mass of the dark photon, the space telescope becomes the superior tool. In fact, for axions near the heaviest masses considered, the telescope's data is more than seven times more sensitive than the collider's best current results.

When the researchers combined the limits from both the particle collider and the space telescope, they found that a thermal relic axion—a version of the particle that would have been created in the early universe and survived to the present day—is completely ruled out across the entire mass range they studied. This means that if axions are indeed the dark matter, they cannot be the simple, standard version that physicists have been looking for; they must be something more complex or exist under different conditions than the models predicted. The work highlights how different tools in the physicist's arsenal, from the massive machines on Earth to the telescopes in orbit, must work together to map the invisible landscape of the universe. By ruling out these specific possibilities, the researchers have narrowed the search, guiding future experiments toward the most promising places to look for the hidden particles that shape our cosmos.

Technical Summary: Probing the Axion–Photon–Dark Photon Interaction with LHC Mono-Photons and Fermi-LAT Gamma-Ray Lines

Problem Statement
This paper investigates a specific interaction vertex linking an axion ( $a$ ), a photon ( $\gamma$ ), and a dark photon ( $\gamma'$ ), mediated by a dimension-five effective operator. While axions and dark photons are individually well-motivated extensions to the Standard Model (SM), their three-way coupling ( $g_{a\gamma'\gamma}$ ) arises in ultraviolet completions involving heavy fermions charged under both the SM and a hidden sector. Previous studies have constrained this coupling at lepton colliders (LEP II, ILC, FCC-ee, CEPC) for dark photon masses below the collider energy. However, the parameter space for heavier dark photons ( $m_{\gamma'} \gtrsim 250$ GeV) remains largely unexplored by lepton machines. Furthermore, if the axion constitutes dark matter (DM), this same vertex induces $s$ -wave annihilation into monochromatic photons, offering a complementary probe via indirect detection. The paper aims to constrain this interaction using LHC mono-photon data and to compare these limits with Fermi-LAT gamma-ray line searches, specifically addressing the regime where the effective theory approaches its validity limit.

Methodology
The authors employ a multi-faceted approach combining collider phenomenology and indirect detection analysis:

Model Implementation: The study utilizes an effective Lagrangian containing the axion-photon-dark photon vertex ( $g_{a\gamma'\gamma}$ ) and kinetic mixing ( $\epsilon$ ). The analysis assumes the regime where the dark photon decays dominantly via $\gamma' \to a\gamma$ , suppressing kinetic-mixing-induced decays to SM fermions.
Collider Simulation (LHC):
- Signal Generation: The process $pp \to a a \gamma$ is simulated at Leading Order (LO) using MadGraph5, followed by parton showering (Pythia8) and fast detector simulation (Delphes with CMS card).
- Full Width Treatment: Crucially, the authors do not use the narrow-width approximation (NWA). They treat the dark photon as a propagator with a mass-dependent width ( $\Gamma_{\gamma'} \propto g_{a\gamma'\gamma}^2 m_{\gamma'}^3$ ). This is essential for high-mass scenarios where the width becomes comparable to the mass, causing the cross-section to saturate rather than scale purely with coupling squared.
- Validation: The simulation chain is validated against CMS's published benchmarks for electroweak DM and simplified DM models, showing agreement within $\sim 4-8\%$ .
- Analysis: The analysis recasts CMS mono-photon search data ( $\sqrt{s}=13$ TeV, $101$ fb $^{-1}$ ) using a multi-bin $\text{CL}_s$ statistical method. Signal regions (SR) are defined by high transverse energy photons ( $E_T^\gamma > 225$ GeV), large missing transverse momentum ( $E_T^{\text{miss}} > 200$ GeV), and specific isolation/kinematic cuts.
Indirect Detection (Fermi-LAT):
- The paper calculates the thermally averaged annihilation cross-sections $\langle \sigma v \rangle$ for $aa \to \gamma\gamma$ , $\gamma Z$ , and $ZZ$ .
- These are compared against Fermi-LAT Pass 8 gamma-ray line search limits (NFW profile, R41 region) to derive constraints on $g_{a\gamma'\gamma}$ assuming the axion is the sole DM component.
Projections: The study projects limits to the High-Luminosity LHC (HL-LHC, $\sqrt{s}=14$ TeV, $3000$ fb $^{-1}$ ) and compares them with existing LEP constraints and projected ILC sensitivities.

Key Contributions and Results

LHC Mono-Photon Constraints:
- The analysis derives a 95% CL exclusion on $g_{a\gamma'\gamma}$ reaching $\simeq 4.8 \times 10^{-4}$ GeV $^{-1}$ for dark photon masses $m_{\gamma'} \lesssim 100$ GeV.
- Sensitivity extends up to $m_{\gamma'} \simeq 2.3$ TeV. Beyond this mass, the required coupling to produce a signal exceeds the validity limit of the effective field theory ( $g_{a\gamma'\gamma} m_{\gamma'} > 4\pi$ ), and the signal rate saturates below the exclusion threshold.
- These results improve upon existing LEP mono-photon constraints by a factor of 3 or more across the overlapping mass range.
- HL-LHC Projection: The projected HL-LHC sensitivity improves the coupling limit by a factor of $\sim 2.7$ and extends the reach to $m_{\gamma'} \simeq 3$ TeV.
Complementarity with Indirect Detection:
- The paper identifies a mass-dependent complementarity between collider and indirect detection. For a fixed $m_{\gamma'} = 300$ $m_{γ^{'}} = 300$ GeV:
  - Light Axions ( $m_a \lesssim 160$ GeV): The LHC mono-photon search provides stronger constraints.
  - Heavy Axions ( $m_a \gtrsim 160$ GeV): Fermi-LAT gamma-ray line searches become more constraining. This is particularly significant near the kinematic endpoint ( $m_a \approx m_{\gamma'}$ ), where the on-shell collider signal is kinematically suppressed (soft photons), but the off-shell contribution (far off-shell $\gamma'$ ) and the indirect detection signal (scaling as $m_a^6$ ) remain robust.
- At $m_a \approx 298.6$ GeV, Fermi-LAT constraints are $\sim 7.5$ times stronger than the current LHC limit and $\sim 4.6$ times stronger than the HL-LHC projection.
Thermal Relic Exclusion:
- By combining the collider limits and Fermi-LAT constraints, the authors demonstrate that a thermal relic axion (where $\langle \sigma v \rangle_{\text{tot}} \approx 3 \times 10^{-26}$ cm $^3$ s $^{-1}$ ) is excluded across the entire axion mass range considered for the benchmark $m_{\gamma'} = 300$ GeV.
- The exclusion is driven by the fact that the axion's annihilation into monochromatic photons ( $\gamma\gamma$ ) carries a large fraction of the total rate (due to the absence of tree-level annihilation into SM fermions), leading to predicted gamma-ray fluxes that significantly exceed Fermi-LAT limits.

Significance
The paper claims to provide the first constraints on the axion–photon–dark photon interaction in the heavy dark photon mass regime ( $m_{\gamma'} > 250$ GeV) using hadron collider data. It highlights the necessity of treating the dark photon width non-perturbatively near the validity limit of the effective theory. Furthermore, it establishes that while lepton colliders are superior for light dark photons, the LHC and HL-LHC are the unique probes for heavy dark photons. Finally, the work demonstrates that the combination of direct collider searches and indirect gamma-ray line searches can completely exclude the parameter space for a thermal relic axion in this specific model, a conclusion that neither method could reach independently across the full mass spectrum.

Technical Summary: Probing the Axion–Photon–Dark Photon Interaction with LHC Mono-Photons and Fermi-LAT Gamma-Ray Lines

More like this