Low-mass doubly-charged Higgs bosons at LHC

This paper proposes a novel search strategy using multivariate analysis to identify low-mass doubly-charged Higgs bosons decaying into boosted W-bosons as exotic jets and same-sign leptons, demonstrating that such particles could be directly probed using existing LHC Run 2 data to address the CDF W-boson mass anomaly.

The Gist

Imagine the Standard Model of particle physics as a massive, incredibly detailed instruction manual for how the universe works. For decades, this manual has been perfect, except for one glaring typo: it has no explanation for why neutrinos (tiny, ghostly particles) have mass.

To fix this typo, physicists proposed a "Type-II See-Saw" mechanism. Think of this like a playground see-saw: if one side goes up (heavy particles), the other side goes down (light neutrinos). This theory predicts the existence of a new, heavy particle called a Doubly-Charged Higgs Boson ($H^{\pm\pm}$).

The Mystery of the "Missing" Particle

For years, scientists at the Large Hadron Collider (LHC)—the world's biggest particle smasher—have been hunting for this specific particle. They've looked everywhere, but they've mostly ignored a specific "neighborhood": particles that are light (between 84 and 200 GeV).

Why did they ignore this neighborhood?

1. The "Needle in a Haystack" Problem: When these light particles decay, they don't shoot out hard, fast debris. Instead, they produce a messy, soft spray of particles that looks exactly like the billions of ordinary collisions happening every second at the LHC. It's like trying to find a specific, slightly damp leaf in a forest during a hurricane.
2. The "CDF Surprise": Recently, a different experiment (CDF) measured the mass of the W-boson and found it was heavier than the Standard Model predicted. This new paper suggests that if our "missing" light Higgs particles exist, they would perfectly explain this heavy W-boson measurement.

The New Strategy: The "Fat Jet" Detective Work

The authors of this paper say, "Let's stop looking for the needle in the haystack and start looking for the haystack itself."

Here is their clever new strategy, explained with analogies:

1. The "Super-Boost" Trick

Usually, particles fly apart slowly. But the authors decided to only look for the rare moments when these Higgs particles are produced with huge energy, zooming away at nearly the speed of light (a "highly Lorentz-boosted regime").

• Analogy: Imagine two people throwing a pair of heavy bowling balls at each other. Usually, they bounce off and roll slowly. But if you throw them at each other at 99% the speed of light, they don't just bounce; they smash together and create a single, massive, glowing fireball that flies off in one direction.
• In the LHC, this "fireball" looks like a single, giant, fat clump of debris called a "Fat Jet."

2. The "Exotic Jet" vs. The "Ordinary Jet"

The LHC is full of "Ordinary Jets" (garbage from normal particle collisions). The signal they want is an "Exotic Jet" (the Higgs fireball).

• The Problem: They look very similar.
• The Solution: The authors used a Machine Learning AI (specifically a "Boosted Decision Tree") to act as a super-detective. They fed the AI thousands of examples of ordinary jets and exotic jets, teaching it to spot tiny differences in the "shape" and "charge" of the debris.
• The Analogy: It's like teaching a dog to distinguish between a real apple and a wax apple. At first, they look identical. But the dog learns that the real apple has a specific texture and weight (in this case, the internal structure of the jet) that the fake one lacks.

3. The Final Clue: The "Same-Sign" Twins

The search doesn't stop at the Fat Jet. The theory says that when the Higgs decays, it also spits out two leptons (like electrons or muons) that have the same electric charge (both positive or both negative).

• The Analogy: In the normal world, if you see two people running away from an explosion, they usually have opposite charges (like a positive and negative battery terminal). Seeing two people with the same charge running away is weird and suspicious.
• The final search looks for this specific trio: One Giant Fat Jet + Two Same-Sign Leptons + Missing Energy (neutrinos).

The Results: We Might Have Already Found It!

The authors ran simulations using data the LHC has already collected (from 2015–2018, known as "Run 2").

• The Good News: They found that with their new "Fat Jet" detective method, they can spot these light Higgs particles using data that is already sitting in the computer. They don't need to wait for new data; the LHC has been collecting it all along, but no one knew how to look for it properly.
• The Conclusion: If these light Higgs particles exist to explain the CDF W-boson mystery, the LHC should be able to find them right now if they apply this new search strategy.

Summary

This paper is a call to action. It says: "We've been looking for a specific type of particle in the wrong way. By focusing on high-speed collisions and using AI to spot 'fat' particle clumps, we can finally solve the mystery of the W-boson's mass and potentially discover a new piece of the universe's puzzle using data we already have."

Technical Summary

1. Problem Statement

The Standard Model (SM) lacks a mechanism to explain neutrino masses. The Type-II Seesaw model, which extends the SM with an $SU(2)_L$ scalar triplet, offers a solution and predicts the existence of doubly-charged Higgs bosons ($H^{\pm\pm}$).

• The Gap: While experimental collaborations (ATLAS and CMS) have set stringent lower mass limits for $H^{\pm\pm}$ decaying into same-sign leptons ($\ell^\pm\ell^\pm$) or $W$-bosons, a specific mass window of 84–200 GeV remains largely unexplored for the $H^{\pm\pm} \to W^\pm W^\pm$ decay channel.
• The Motivation: Recent measurements of the $W$-boson mass by the CDF collaboration, which deviate significantly from the SM global electroweak fit, can be explained by the Type-II Seesaw model if it contains low-mass $H^{\pm\pm}$ (84–200 GeV) accompanied by slightly heavier singly-charged and neutral scalars.
• The Challenge: Conventional LHC searches for this mass range fail because:
  1. The decay products ($W$ bosons) are not sufficiently "hard" (high transverse momentum) to stand out against the overwhelming QCD and electroweak backgrounds.
  2. The signal is often drowned in SM resonances.
  3. Standard searches rely on leptons and jets that are not optimized for the specific kinematic regime of these light, boosted particles.

2. Methodology

The authors propose a novel search strategy targeting the highly Lorentz-boosted regime of $H^{\pm\pm}$ pair production ($pp \to H^{++}H^{--}$). In this regime, the decay products of the $H^{\pm\pm}$ are collimated, manifesting as a single "fat-jet" (dubbed an $H^{\pm\pm}$-jet) or a pair of adjacent same-sign leptons, accompanied by missing transverse momentum ($p_T^{miss}$) from neutrinos.

The analysis proceeds in three main stages:

A. Event Simulation and Reconstruction

• Signal Generation: Simulated $pp \to H^{++}H^{--}$ at $\sqrt{s}=13$ TeV using MadGraph5_aMC@NLO with NNPDF2.3 LO PDFs. Cross-sections are scaled by an NLO $K$-factor of 1.15.
• Decay Assumption: The study assumes exclusive prompt decays of $H^{\pm\pm} \to W^\pm W^{\pm(*)}$.
• Detector Simulation: Events are processed through PYTHIA 8.2 (showering/hadronization) and Delphes 3.4.2 (CMS detector simulation).
• Object Definition:
  • Fat-jets: Clustered using the anti-$k_T$ algorithm ($R=1.0$), pruned to remove soft wide-angle radiation, and analyzed using $N$-subjettiness ($\tau_N$).
  • Leptons: Electrons and muons required to be isolated ($|\eta| < 2.5$, $p_T > 10$ GeV).
  • Kinematics: The analysis focuses on events with high transverse momentum ($p_T(H^{\pm\pm}) > 300$ GeV), which reduces the total signal cross-section but significantly enhances the signal-to-background ratio by isolating the boosted regime.

B. Multivariate Analysis (MVA) for Jet Tagging

A critical component is distinguishing the exotic $H^{\pm\pm}$-jet from standard Model jets (QCD, $W/Z$, Higgs, Top).

• Classifier: A Boosted Decision Tree (BDT) implemented in TMVA.
• Input Variables:
  • Invariant mass ($m$)
  • $b$-tagging status
  • Jet charge ($Q_k$)
  • $N$-subjettiness ratios ($\tau_{21}, \tau_{32}, \tau_{43}$)
• Performance: The BDT achieves a separation power where the Area Under the Curve (AUC) of the ROC is $\sim 0.13$ (indicating strong separation). The variable $\tau_{21}$ is identified as the most discriminating feature.
• Efficiency: Signal efficiency drops for masses $< 100$ GeV due to the small mass difference between $H^{\pm\pm}$ and the $W$ boson, causing decay products to be too soft to pass reconstruction cuts.

C. Event Selection Strategy

The final search channel is defined as: One $H^{\pm\pm}$-jet + Two Same-Sign (SS) Leptons + Missing Transverse Momentum.
Selection cuts (S0–S3) are applied to suppress backgrounds (Diboson, Top, Drell-Yan, etc.):

1. S0: One fat-jet ($p_T > 300$ GeV), two SS leptons, $\Delta R_{\ell\ell} > 0.05$, and $m_{\ell\ell}$ cuts to remove $J/\psi$ and bremsstrahlung.
2. S1: BDT response $> 0.1$ (selects exotic jets).
3. S2: $m_{\ell\ell} < 80$ GeV (suppresses $Z \to \ell\ell$ background with charge misidentification).
4. S3: $\Delta R_{\ell\ell} < 1.2$, $p_T^{miss} > 80$ GeV, and $\Delta\phi(\ell\ell, p_T^{miss}) < 0.8$. These cuts exploit the fact that in the boosted regime, leptons and neutrinos are collimated.

3. Key Contributions

1. Novel Search Channel: The paper introduces a search for low-mass $H^{\pm\pm}$ in the $H^{\pm\pm}$-jet + SS Leptons + $p_T^{miss}$ final state, a channel previously overlooked by ATLAS and CMS.
2. Jet Substructure Technique: It demonstrates the efficacy of using jet substructure variables (specifically $N$-subjettiness and jet charge) combined with BDTs to identify boosted $W$-boson pairs originating from light scalars.
3. Reinterpretation of CDF Anomaly: It provides a concrete experimental pathway to test the Type-II Seesaw explanation for the CDF $W$-mass anomaly using existing LHC data.
4. General Applicability: The strategy is noted to be applicable to any low-mass BSM Higgs (charged or neutral) decaying into pairs of SM gauge bosons.

4. Results

• Cross-Sections: For the mass range 84–200 GeV, the pair production cross-section ranges from 1.72 pb down to 74.5 fb.
• Background Suppression: The proposed cuts effectively reduce the total SM background by orders of magnitude while retaining a significant fraction of the signal (e.g., for $m_{H^{\pm\pm}} = 150$ GeV, signal efficiency is $\sim 43\%$ after all cuts).
• Discovery Potential:
  • The authors calculate the required luminosity for a $5\sigma$ discovery and $95\%$ CL exclusion.
  • Crucial Finding: The analysis concludes that the already collected Run 2 LHC data is sufficient to probe this entire mass window (84–200 GeV) with $5\sigma$ discovery significance.
  • If the data is consistent with the SM, the existing data is also sufficient to exclude these particles at the 95% confidence level.

5. Significance

This work is significant because it addresses a "blind spot" in current LHC searches. While high-mass $H^{\pm\pm}$ are well-constrained, the low-mass region (84–200 GeV) remains open despite its theoretical relevance to the CDF $W$-mass anomaly. By leveraging the Lorentz-boosted regime and jet substructure techniques, the authors demonstrate that the LHC has the sensitivity to either discover these particles or definitively rule out the specific Type-II Seesaw parameter space favored by the CDF anomaly using data already in hand. This suggests that the CDF anomaly, if real, could be resolved or refuted by re-analyzing existing LHC datasets with this specific strategy.