⚛️ phenomenology

Identifying the origin of the 146-GeV excess at the LHC

This paper proposes three new production channels ( $H\gamma$ , $Hj$ , and $Hjj$) within the two-Higgs-doublet model to distinguish between Higgs-mixing and leptophilic origins of the 146-GeV CMS excess, demonstrating that the $H\gamma$ channel alone can achieve a 7.1 $\sigma$ significance at the high-luminosity LHC, rising to 7.6 $\sigma$ when combined with the other channels.

Original authors: P. Uttayarat, J. Julio, R. Primulando

Published 2026-08-20

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: P. Uttayarat, J. Julio, R. Primulando

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

For the last decade, the world of particle physics has rested on a foundation built around a single, famous discovery: the Higgs boson. Found in 2012, this particle is the final piece of the Standard Model, the rulebook that describes how the universe's most basic building blocks interact. It explains why other particles have mass. Since its discovery, scientists have spent years measuring its behavior, and so far, it has acted exactly as the rulebook predicted. However, the universe is rarely so tidy. The Standard Model does not explain everything, and physicists have long suspected that there are hidden particles or forces waiting to be found. One of the most intriguing possibilities involves a phenomenon called lepton flavor violation. In the world of neutrinos, particles can change their identity as they travel, but for the heavier, charged cousins of the neutrino—the electron and the muon—such a change has never been seen. If a particle were found that could turn an electron into a muon, or vice versa, it would shatter the current understanding of physics and point toward a deeper, more complex reality.

Recently, a team at the Large Hadron Collider, the massive particle smasher near Geneva, reported a tantalizing hint of such a discovery. While searching for new particles in the debris of high-speed collisions, they noticed an unusual bump in the data. Specifically, they found more pairs of electrons and muons appearing together than the background noise should allow. This excess centered around a mass of 146 gigaelectronvolts, a specific weight for a new particle. While the signal was not strong enough to be declared a discovery, it was significant enough to spark intense curiosity. The question became: what is causing this excess? Is it a new particle that is secretly mixing with the known Higgs boson, or is it something entirely different, a particle that only talks to leptons and ignores the rest of the universe?

A new study by physicists from Thailand and Indonesia sets out to solve this mystery before the next major run of the collider ends. The researchers did not look for the particle again; instead, they asked how we could tell which of the two leading theories is correct. They proposed a strategy to look at the company the new particle keeps. If the particle is a "mixing" type, it is produced by the same powerful forces that create the known Higgs, forces that involve the collision of gluons, the particles that bind atomic nuclei together. If it is a "leptophilic" type, meaning it loves leptons, it is produced directly by the collision of electrons and muons that are already floating inside the protons being smashed. These two origins leave different footprints in the detector.

To distinguish between these two possibilities, the authors simulated what would happen if the Large Hadron Collider were upgraded to its full high-luminosity potential, collecting a vast amount of data. They focused on three specific scenarios where the new particle is produced alongside other objects. First, they looked for the particle appearing with a photon, a particle of light. In the mixing scenario, the light would likely be emitted by the final electron or muon after the collision. In the leptophilic scenario, the light could be emitted by the electron or muon before the collision even happens. By carefully measuring the energy and direction of the light relative to the electron and muon, the researchers found that the two theories would produce very different patterns. Their simulations showed that with enough data, this single channel could distinguish between the mixing and leptophilic origins with a statistical significance of 7.1, a level of certainty that would definitively separate the two scenarios.

They also examined two other ways the particle could appear: accompanied by a single jet of debris, or by two jets moving in opposite directions. The mixing scenario, which relies on the strong nuclear force, is much more likely to produce these jets because the colliding particles are heavy and energetic. The leptophilic scenario, relying on the lighter leptons, rarely produces such energetic jets. The study found that while these jet channels alone were not powerful enough to settle the debate, they provided valuable supporting evidence. When the researchers combined the results from the light channel and the two jet channels, the ability to tell the difference became overwhelming.

The conclusion of the study is clear and robust. If the 146-GeV excess persists when the collider reaches its full data-taking capacity, the high-luminosity machine will be able to determine its origin with high confidence. The simulations suggest that by combining all three search methods, scientists could distinguish between the mixing and leptophilic origins with a level of certainty that leaves almost no room for doubt. Even when accounting for the inevitable imperfections and uncertainties in real-world measurements, the signal remains strong enough to solve the puzzle. This work does not prove the particle exists, but it provides a clear roadmap for how to find out what it is, turning a mysterious bump in the data into a potential gateway to new physics.

Technical Summary: Identifying the Origin of the 146-GeV Excess at the LHC

Problem Statement
The CMS Collaboration reported a local excess of approximately 3.8 standard deviations in the electron-muon ( $e\mu$ ) final state at an invariant mass of 146 GeV, based on 138 fb $^{-1}$ of LHC Run-2 data. This signal suggests the existence of a new scalar resonance ( $H$ ) with a production cross-section times branching ratio of $\sigma \times \text{BR}(H \to e\mu) \approx 3.89$ fb. Two competing theoretical frameworks have been proposed to explain this excess:

The Mixing Scenario: The new scalar mixes with the Standard Model (SM) Higgs boson, allowing production via gluon fusion (ggF) and vector-boson fusion (VBF). This requires a small mixing angle ( $\sin \alpha \lesssim 0.2$ ) and specific Yukawa couplings to leptons.
The Leptophilic Scenario: The new scalar has no mixing with the SM Higgs ( $\alpha = 0$ ) and couples exclusively to leptons. Production occurs via the annihilation of sea electrons and muons within the proton (lepton PDFs), requiring large lepton-flavor-violating (LFV) couplings ( $Y_{e\mu}, Y_{\mu e} \sim \mathcal{O}(1)$ ).

While both scenarios can accommodate the observed excess while satisfying low-energy constraints (such as $\mu \to e\gamma$ and LEP limits), they predict distinct initial states and associated production signatures. The primary challenge is distinguishing between these two origins to determine the nature of the resonance.

Methodology
The authors analyze the potential of the High-Luminosity LHC (HL-LHC) with an integrated luminosity of 3000 fb $^{-1}$ to differentiate these scenarios. They employ the Two-Higgs-Doublet Model (2HDM) as an explicit framework, working in the Higgs basis where the 146-GeV scalar is identified as the heavier CP-even state ( $H$ ).

The study utilizes Monte Carlo simulations (MadGraph5, Pythia8, Delphes 3) at $\sqrt{s} = 14$ TeV to model signal and background events. Key methodological steps include:

Signal Normalization: Parameters are tuned such that the total $\sigma \times \text{BR}(H \to e\mu)$ matches the CMS excess value (3.89 fb) at 13 TeV.
Event Selection: Events are required to contain exactly one oppositely charged $e\mu$ pair with $p_T > 25$ GeV and specific isolation criteria. Backgrounds from top quarks are suppressed via $b$ -jet vetoes.
Discriminating Channels: Three associated production channels are proposed to exploit the differences in initial states (hadronic vs. leptonic):
1. $H\gamma$ (Associated Photon): Distinguishes between Initial State Radiation (ISR) from leptons (Leptophilic) and Final State Radiation (FSR) from leptons (Mixing). Two signal regions (SR A and SR B) are defined based on whether the photon is associated with the $e\mu$ mass or the full $e\mu\gamma$ mass.
2. $Hj$ (Associated Jet): Targets hard ISR jets characteristic of the ggF production in the Mixing scenario, which are suppressed in the Leptophilic case.
3. $Hjj$ (VBF-inspired): Targets the characteristic topology of Vector Boson Fusion (two forward jets with large rapidity separation and high dijet mass), a dominant mode for the Mixing scenario but absent in the Leptophilic scenario (where jets arise only from minimum bias).
Statistical Analysis: The separation power is quantified using a Poisson likelihood-ratio test statistic ( $Z$ ), comparing the expected yields of the Mixing hypothesis ( $\mu_1$ ) against the Leptophilic hypothesis ( $\mu_0$ ) as the null hypothesis.

Key Contributions

Proposal of Discriminating Channels: The paper introduces $H\gamma$ , $Hj$, and $Hjj$ as specific search channels designed to test the initial-state nature of the 146-GeV resonance.
Signal Region Optimization: Specifically for the $H\gamma$ channel, the authors define two complementary signal regions (SR A and SR B) and a transverse momentum cut ( $p_T(e\mu\gamma) < 10$ GeV) to maximize the separation between ISR-dominated (Leptophilic) and FSR-dominated (Mixing) events.
Comprehensive HL-LHC Projection: The study provides a quantitative assessment of the HL-LHC's capability to resolve the origin of the excess, accounting for statistical uncertainties and estimating the impact of systematic uncertainties.

Results
The analysis yields the following significance ( $Z$ ) values for distinguishing the Mixing scenario from the Leptophilic scenario at 3000 fb $^{-1}$ :

$H\gamma$ Channel: The combination of SR A and SR B provides a significance of $Z = 7.1$ . This channel alone offers strong discriminating power, as the Leptophilic scenario yields significantly more events in the signal regions compared to the Mixing scenario.
$Hj$ Channel: This channel shows low discriminating power ( $Z = 0.6$ ) due to large Standard Model backgrounds (primarily $t\bar{t}$ ) and accidental jets from pile-up in the Leptophilic case.
$Hjj$ Channel: The VBF-inspired channel provides moderate discrimination ( $Z = 2.8$ ) by leveraging the topological differences in jet production, though the Leptophilic yield remains non-negligible due to minimum bias jets mimicking VBF topology.
Combined Significance: Combining all three channels yields a total significance of $Z = 7.6$ .
Systematic Uncertainties: When a conservative 20% uncorrelated systematic uncertainty is applied to the expected yields, the combined significance reduces to $Z = 5.4$ .

Significance and Claims
The paper claims that if the 146-GeV excess persists through the end of LHC Run 3, the HL-LHC has the potential to definitively identify its origin. The authors assert that the proposed $H\gamma$ channel is particularly powerful, capable of distinguishing the mixing and leptophilic scenarios with a significance of 7.1 on its own. By combining all proposed channels, the total discriminating power reaches 7.6 (statistical only) or 5.4 (including 20% systematics).

The authors conclude that these results demonstrate the HL-LHC's great potential to "conclusively unravel the true nature of this excess." They emphasize that the distinction relies on the fundamental difference in production mechanisms: hadronic initial states (ggF/VBF) for the mixing scenario versus leptonic initial states (lepton PDFs) for the leptophilic scenario. The study does not claim to discover new physics but rather provides a roadmap for testing the specific nature of the reported CMS anomaly.

More like this