Probing Type-I 2HDM light Higgs in the top-pair-associated diphoton channel

Motivated by the 95 GeV diphoton excess, this study evaluates the discovery potential of a light Higgs boson in the Type-I Two-Higgs-Doublet Model via the top-pair-associated diphoton channel, demonstrating that while direct searches constrain the mixing angle $\alpha$, future colliders like the HL-LHC, HE-LHC, and FCC-hh can probe significant regions of the parameter space, particularly for $\sin(\beta-\alpha)$ values away from zero, with the FCC-hh offering the most comprehensive sensitivity.

The Gist

The Mystery: A Ghostly Glitch at 95 GeV

Imagine the Standard Model of particle physics as a perfectly written instruction manual for how the universe works. For years, scientists have been checking this manual against reality, and everything has matched up perfectly—except for one tiny, persistent glitch.

Back in 2003, and again recently in 2018, 2023, and 2024, experiments at the world's biggest particle accelerators (the LHC) noticed a "ghost." They saw a faint signal of two light particles (photons) appearing together with a specific energy weight of about 95 GeV.

Think of it like this: You are listening to a radio station playing a clear song (the known 125 GeV Higgs boson). Suddenly, you hear a faint, staticky whisper of a second song playing underneath it. It's not loud enough to be a confirmed discovery yet (it's only about a 2-3% chance of being a fluke), but it's there, and it keeps coming back.

The authors of this paper ask: "What if this ghost is real? Could it be a new particle predicted by a specific theory called the Type-I Two-Higgs-Doublet Model (2HDM-I)?"

The Theory: The "Double-Decker" House

The Standard Model says there is only one "Higgs field" (like a single floor in a house) that gives particles their mass. The 2HDM-I theory suggests the house has a second floor.

• The Ground Floor: This is the familiar Higgs boson we found in 2012 (125 GeV).
• The Second Floor: This theory predicts a lighter, hidden Higgs boson (the 95 GeV ghost) and some other heavy particles (charged Higgses) that we haven't seen yet.

In this specific "Type-I" version of the theory, the rules of the house are strict: only one floor interacts with the "heavy" particles (like quarks), while the other handles the "light" ones. This specific setup makes it easier for the 95 GeV ghost to hide in plain sight and decay into two photons, which is exactly what the experiments are seeing.

The Investigation: The Top-Quark Trap

To catch this ghost, the scientists needed a better net. Usually, looking for a new particle is like trying to find a needle in a haystack made of other needles. The "haystack" here is the background noise of the universe (Standard Model processes).

The authors proposed a clever strategy: The Top-Pair Trap.

Imagine the Top Quark as a very heavy, loud bouncer at a club. When two Top Quarks are created, they are so heavy and energetic that they leave a very distinct trail. By looking for the 95 GeV ghost only when it is created alongside a pair of these Top Quarks, the scientists can filter out most of the background noise.

It's like looking for a specific rare coin, but only looking for it inside a specific, heavy, locked safe (the Top Quark pair). If you find the coin inside that safe, you know it's real because the safe is so hard to open that random noise rarely gets in.

The Simulation: Building a Virtual Future

Since we can't build a bigger machine tomorrow, the authors used supercomputers to run Monte Carlo simulations. They built a "virtual universe" to test how well different future colliders could find this ghost.

They tested three different "magnifying glasses":

1. HL-LHC (High-Luminosity LHC): The current machine, upgraded to run longer and brighter (like a brighter flashlight).
2. HE-LHC (High-Energy LHC): A machine that crashes particles together much harder (like a stronger hammer).
3. FCC-hh (Future Circular Collider): A massive, 100 TeV machine (the "Super Hammer").

The Results:

• The Flashlight (HL-LHC): With enough time (3 years of data), the current machine can find the ghost if it's not too shy. It can confirm the signal if the "brightness" (cross-section) is above a certain threshold.
• The Stronger Hammer (HE-LHC): This machine is so powerful it can find the ghost even if it's hiding deeper in the shadows.
• The Super Hammer (FCC-hh): This machine is so powerful it can find the ghost almost anywhere, even in the darkest corners of the theory.

The Catch: The "Alignment" Blind Spot

There is one tricky part. The theory has a "knob" called $\sin(\beta - \alpha)$.

• If you turn the knob to a certain position, the 95 GeV ghost loves to turn into two photons (the signal we see).
• If you turn the knob to the "alignment" position (where $\sin(\beta - \alpha) \approx 0$), the ghost becomes invisible to our photon detectors. It stops turning into photons and starts turning into other things (like bottom quarks) that are much harder to see because they are buried in noise.

The paper concludes that while future colliders can find the ghost in most scenarios, if the ghost is in this "alignment" position, even the Super Hammer might miss it in the photon channel. We would need to look for it in different ways (like looking for it in the bottom-quark channel).

The Bottom Line

This paper is a roadmap for the next 10–20 years of particle physics. It says:

1. The 95 GeV signal is plausible within this specific theory (Type-I 2HDM).
2. We can find it. If the signal is real, the upgraded LHC, the High-Energy LHC, or the Future Circular Collider will almost certainly confirm it by looking for it alongside Top Quarks.
3. We need to be careful. If the signal is hiding in a specific "alignment" corner of the theory, we might need to change our search strategy to catch it.

In short: The ghost might be real, and we have the tools to catch it, provided we look in the right place and use the right net.

Technical Summary

1. Problem Statement

The paper addresses the persistent experimental anomalies observed at the LHC and previous colliders regarding a potential 95 GeV scalar particle:

• Experimental Context: A local excess was first noted at LEP (~98 GeV) and recently reinforced by CMS and ATLAS in the diphoton ($h \to \gamma\gamma$) channel around 95.3–95.4 GeV. CMS reported a local significance of ~2.9$\sigma$, and ATLAS reported ~1.7$\sigma$.
• Theoretical Gap: While many Beyond the Standard Model (BSM) scenarios (e.g., singlet-dominated scalars) have been proposed to explain this, Type-I Two-Higgs-Doublet Models (2HDM-I) featuring a doublet-dominated light scalar have received less attention.
• Specific Challenge: The 2HDM-I must simultaneously explain the 95 GeV excess, satisfy the observed 125 GeV Higgs properties, and comply with stringent constraints from flavor physics (e.g., $B \to X_s\gamma$) and electroweak precision observables. Furthermore, the paper investigates whether the top-pair-associated production channel ($pp \to t\bar{t}h$) offers sufficient sensitivity to discover or constrain this light Higgs at future colliders.

2. Methodology

The authors employed a multi-stage approach combining theoretical parameter scans with full Monte Carlo (MC) simulations:

• Model Framework:

  • 2HDM-I: A model where only one Higgs doublet ($\Phi_2$) couples to all fermions (up-type, down-type, and leptons), naturally suppressing Flavor-Changing Neutral Currents (FCNCs).
  • Parameters: The study scans seven input parameters: masses of the neutral scalars ($m_h=95$ GeV, $m_H=125$ GeV, $m_A$), charged Higgs mass ($m_{H^\pm}$), mixing parameter $m_{12}^2$, mixing angle $\alpha$, and $\tan\beta$.
  • Constraints: The parameter space was filtered using:
    • Flavor Physics: $B \to X_s\gamma$, $B_s \to \mu^+\mu^-$, $B_d \to \mu^+\mu^-$.
    • Direct Searches: LEP and LHC limits on Higgs production.
    • 125 GeV Higgs Data: Compatibility with SM-like Higgs couplings (using HiggsSignal).
    • Theoretical Limits: Unitarity, vacuum stability, and perturbativity.
    • Muon $g-2$: Consistency with both the 2020 (WP20) and 2025 (WP25) White Paper predictions for the muon anomalous magnetic moment.

• Simulation & Analysis:

  • Tools: SARAH and SPheno for spectrum calculation; MadGraph5_aMC@NLO for event generation; PYTHIA for parton showering/hadronization; Delphes for detector simulation; MadAnalysis5 for analysis.
  • Signal Process: $pp \to t\bar{t}h(\to \gamma\gamma)$, where $t \to Wb$.
  • Backgrounds: Resonant ($t\bar{t}H_{125}$, $tjH_{125}$) and non-resonant ($t\bar{t}\gamma\gamma$, $b\bar{b}\gamma\gamma$, $t\bar{t}\gamma$, $Wjj\gamma\gamma$).
  • Colliders Simulated:
    • HL-LHC: 14 TeV, $\mathcal{L} = 3 \text{ ab}^{-1}$.
    • HE-LHC: 27 TeV, $\mathcal{L} = 10 \text{ ab}^{-1}$.
    • FCC-hh: 100 TeV, $\mathcal{L} = 30 \text{ ab}^{-1}$.
  • Selection Strategy: A series of kinematic cuts (Basic, Mass, Energy) were optimized to isolate the 95 GeV resonance peak in the diphoton invariant mass ($M_{\gamma\gamma}$) while suppressing backgrounds.

3. Key Contributions

• Parameter Space Viability: Demonstrated that the 2HDM-I can successfully accommodate the 95 GeV diphoton excess while satisfying all current experimental constraints, provided specific regions of $\alpha$ and $\tan\beta$ are selected.
• Role of the Top-Associated Channel: Established that the $t\bar{t}h$ channel is a viable discovery mode for a 95 GeV light Higgs, offering a cleaner signature than inclusive production due to reduced QCD backgrounds and the presence of $b$-jets and leptons for tagging.
• Muon $g-2$ Correlation: Analyzed the impact of the updated muon $g-2$ measurements (WP25 vs. WP20) on the 2HDM-I parameter space, showing that the model can satisfy the WP25 result across a wide range of $\tan\beta$, whereas WP20 prefers lower $\tan\beta$.
• Discovery Potential Quantification: Provided the first comprehensive MC-based sensitivity projections for the 95 GeV Higgs in the $t\bar{t}h \to t\bar{t}\gamma\gamma$ channel across three generations of future colliders.

4. Key Results

• Parameter Constraints:

  • Mixing Angle ($\alpha$): Direct Higgs searches strongly exclude $\alpha \lesssim 0.95$.
  • $\tan\beta$: Flavor constraints (specifically $B \to X_s\gamma$) require $\tan\beta \gtrsim 2.6$.
  • Signal Strength: The observed 95 GeV excess ($\mu_{\gamma\gamma} \approx 0.18 - 0.33$) can be explained within $1\sigma$ for a wide range $0.95 \lesssim \alpha \lesssim 1.55$.
  • Branching Ratio: $Br(h \to \gamma\gamma)$ is highly sensitive to $\sin(\beta-\alpha)$. It drops to $\sim 10^{-6}$ near the alignment limit ($\sin(\beta-\alpha) \approx 0$) but reaches $\sim 10^{-1}$ for $\alpha \approx 1.55$.

• Cross Sections:

  • Maximum signal cross sections ($pp \to t\bar{t}h \to t\bar{t}\gamma\gamma$) for surviving samples:
    • HL-LHC (14 TeV): $\sim 0.44$ fb.
    • HE-LHC (27 TeV): $\sim 1.42$ fb.
    • FCC-hh (100 TeV): $\sim 17.21$ fb.

• Discovery Significance (Statistical $S$):

  • Thresholds for 5$\sigma$ Discovery:
    • HL-LHC ($3 \text{ ab}^{-1}$): Requires $\sigma \geq 0.3$ fb.
    • HE-LHC ($10 \text{ ab}^{-1}$): Requires $\sigma \geq 0.67$ fb.
    • FCC-hh ($30 \text{ ab}^{-1}$): Requires $\sigma \geq 2.36$ fb.
  • Coverage of Parameter Space:
    • HL-LHC: Can probe $\sin(\beta-\alpha) \gtrsim 0.4$ (5$\sigma$) and $\gtrsim 0.25$ (2$\sigma$).
    • HE-LHC: Sensitivity improves to $\sin(\beta-\alpha) \gtrsim 0.25$ (5$\sigma$) and $\gtrsim 0.15$ (2$\sigma$).
    • FCC-hh: Can cover even extreme regions, $\sin(\beta-\alpha) \gtrsim 0.1$ or $\lesssim -0.05$ at 5$\sigma$.
  • Limitation: Regions where $\sin(\beta-\alpha) \approx 0$ remain unprobed even at FCC-hh due to the suppression of the $hWW$ coupling, which drastically reduces the diphoton decay rate.

5. Significance

• Validation of 2HDM-I: The study confirms that a doublet-dominated light scalar in the Type-I 2HDM is a robust candidate for the 95 GeV excess, provided the mixing angle $\alpha$ is sufficiently large.
• Future Collider Roadmap: The paper provides concrete benchmarks for future experiments. It suggests that while the HL-LHC may only partially probe the viable parameter space, the HE-LHC and especially the FCC-hh have the potential to fully cover the 2HDM-I explanation for the 95 GeV excess via the $t\bar{t}h$ channel.
• Strategic Insight: It highlights a critical "blind spot" in the diphoton channel (the alignment limit $\sin(\beta-\alpha) \approx 0$). If the 95 GeV particle exists in this region, the diphoton channel will fail to detect it, necessitating searches in alternative channels like $h \to b\bar{b}$ or $h \to \tau^+\tau^-$.
• Methodological Benchmark: The rigorous application of full MC simulations with realistic detector effects sets a standard for evaluating BSM signals in complex final states involving top quarks and photons.