[Paper Review] Searching for Low-Mass Resonances Decaying into $W$ Bosons

Guglielmo Coloretti, Andreas Crivellin+2 more|Zurich Open Repository and Archive (University of Zurich)|Feb 14, 2023

Particle physics theoretical and experimental studies4 citations

TL;DR

This paper reanalyzes ATLAS and CMS data from the LHC's Run 2 to search for low-mass new scalar resonances decaying into W bosons via gluon fusion. It finds a local significance of ≳2.5σ for a 95 GeV scalar boson in the WW→ℓνqq final state, with a preferred cross section of ≈0.5 pb, strengthening hints from other channels and suggesting a possible new Higgs-like particle beyond the Standard Model.

ABSTRACT

In this article, we recast and combine the CMS and ATLAS analyses of the Standard Model Higgs boson decaying to a pair of $W$ bosons in order to search for low-mass resonances in this channel. We provide limits on the corresponding cross section assuming direct production via gluon fusion. For the whole range of masses we consider (90$\,$GeV to 200$\,$GeV), the observed limit on the cross section turns out to be weaker than the expected one. Furthermore, at $\approx95\,$GeV the limit is weakest and a new scalar decaying into a pair of $W$ bosons (which subsequently decay leptonically) with a cross section $\approx0.5\,$pb is preferred over the Standard Model hypothesis by $\gtrsim 2.5\,σ$. In light of the excesses in the $γγ$, $τ^+τ^-$ and $b\bar b$ channels at similar masses, this strengthens the case for such a new Higgs boson. Furthermore, this analysis also gives room for the scalar candidate at 151$\,$GeV decaying into $W$ bosons.

Motivation & Objective

• To search for low-mass new scalar resonances decaying into W boson pairs using LHC Run 2 data.
• To test whether existing hints for a 95 GeV scalar in other decay channels are supported by its decay into WW final states.
• To combine ATLAS and CMS analyses of SM Higgs boson decays into WW to constrain new physics in the 90–200 GeV mass range.
• To assess the viability of a new neutral scalar as the neutral component of an SU(2)L triplet with hypercharge 0.
• To evaluate whether the observed signal excess at 95 GeV is consistent with direct production via gluon fusion and subsequent leptonic decays of W bosons.

Proposed method

• Reconstructed the SM Higgs boson decay process gg→h→WW*→ℓ⁺ℓ⁻νν̄ using MadGraph5aMC@NLO, Pythia8.3, and Delphes for fast simulation.
• Simulated 1 million events per mass point (90–200 GeV) for a new scalar H produced via gluon fusion and decaying into WW* with subsequent leptonic decays.
• Validated the simulation by comparing the transverse mass (mT) distribution to ATLAS data for the SM Higgs boson.
• Performed a global fit combining ATLAS and CMS data, allowing background refitting in CMS categories to improve sensitivity.
• Used profile-likelihood ratio test statistics to compute local and global significances, accounting for look-elsewhere effects.
• Evaluated the significance of the signal with and without background refitting, particularly in high-pT2 regions where efficiency is low.

Experimental results

Research questions

• RQ1 Is there evidence for a new scalar resonance decaying into W bosons in the 90–200 GeV mass range in the combined ATLAS and CMS Run 2 data?
• RQ2 Does the observed signal excess at ≈95 GeV in the WW→ℓνqq final state persist when accounting for background uncertainties and data combination?
• RQ3 Can the 95 GeV signal excess be interpreted as a new scalar boson produced via gluon fusion and decaying into W pairs, consistent with the SM Higgs-like behavior?
• RQ4 How does the significance of the signal compare to the expected limit, and what is the impact of background refitting on the significance estimate?
• RQ5 Is the observed signal consistent with a neutral scalar from an SU(2)L triplet model that couples exclusively to W bosons and explains the CDF II W-boson mass anomaly?

Key findings

• The observed cross-section limit is weaker than the expected limit across the entire 90–200 GeV mass range, indicating a preference for a new physics signal.
• At approximately 95 GeV, the signal significance reaches ≳2.5σ locally, with a preferred cross section of ≈0.5 pb for a new scalar decaying into W bosons.
• The combined fit of ATLAS and CMS data yields a global significance of 2.6σ (refit) and 2.8σ (no refit), with the 95 GeV signal being the most prominent.
• The CMS analysis shows a significant tension in the high-pT2 region, where the signal significance drops due to low efficiency, but the combined fit remains consistent with a positive signal.
• The 150 GeV case shows a smaller but still non-zero signal preference (2.2σ global significance), though it is less compelling than at 95 GeV.
• The results are consistent with a new scalar as the neutral component of an SU(2)L triplet with hypercharge 0, which could explain the CDF II W-boson mass measurement via its vacuum expectation value.