Discovery prospects for photophobic axion-like particles in the $WWjj$ final state at the High-Luminosity LHC

TL;DR

The paper probes heavy photophobic axion-like particles at the HL-LHC in the $WWjj$ final state, exploiting both $s$-channel EW exchange and VBF-like production with $a\to W^+W^-$. It developes a two-step analysis—preselection to define a signal-like final state and a per-mass optimized BDT-based MVA—to separate signal from backgrounds in the $e\pm\mu^\mp+jj+\slashed{E}_T$ channel. The study delivers 2$\sigma$ and 5$\sigma$ discovery sensitivities for the ALP–$W$ coupling $g_{aWW}$ across $m_a=170$–$4000$ GeV, with a near-flat reach of $\sim0.61$ and $0.76\,\mathrm{TeV}^{-1}$ in the $260\le m_a\le1500$ GeV range, and provides model-independent fiducial cross-section thresholds for reinterpretation. The results show that the $WWjj$ channel is competitive and complementary to other photophobic ALP searches, offering direct access to the $aWW$ vertex and a robust cross-check of electroweak couplings when combined with $Z\gamma$ channels.

Abstract

We evaluate discovery prospects for photophobic axion-like particles (ALPs) in the $WWjj$ final state at the High-Luminosity LHC (HL-LHC; $\sqrt{s}=14$ TeV, $L=3$ ab$^{-1}$). In the photophobic limit ($g_{aγγ}=0$), ALPs couple to electroweak gauge bosons and are produced in association with two jets ($pp\to jj a$) via both $s$-channel electroweak exchange and vector-boson-fusion (VBF)-like topologies, followed by $a\to W^+W^-$ decay. We target the different-flavour dilepton mode $W^+W^-\to e^\pmμ^\mpν\barν$ with two jets and moderate missing transverse momentum. The analysis employs a two-step strategy: an initial preselection that defines the signal-like final state, followed by a multivariate analysis (MVA) trained on dijet and dileptonic-$WW$ kinematics from both the $s$-channel and VBF-like production mechanisms to separate signal from background; the MVA threshold is optimized independently at each $m_a$. We present $2σ$ and $5σ$ discovery sensitivities for the ALP-$W$ coupling $g_{aWW}$ across the $170$--$4000$ GeV mass range. For $260$ GeV $\le m_a \le 1500$ GeV, the $2σ$ ($5σ$) sensitivity is approximately flat around 0.61 (0.76) TeV$^{-1}$. We also report model-independent discovery thresholds for the fiducial quantity $σ(pp\to jj a)\,\mathrm{Br}(a\to W^+W^-)$ over the same mass range to enable reinterpretation for other models. These results indicate that the $WWjj$ topology offers competitive and complementary sensitivity to heavy photophobic ALPs at the HL-LHC.

Figures (13)

• Figure 1: Representative production mechanisms for heavy photophobic ALPs at $pp$ colliders: $s$-channel exchange (left) and VBF-like topology (right), yielding $pp\to jj\,a$ with $a$–$V$–$V'$ vertices ($aWW$, $aZZ$, $a\gamma Z$). The ALP subsequently decays to two $W$ bosons, defining the signal $pp\to jj\,a(\to W^\pm W^\mp)$ with $W^\pm\to \ell^\pm\bar{\nu}/\nu$.
• Figure 2: Production cross section of $pp\to jj\,a(\to W^\pm W^\mp)$ at the HL-LHC ($\sqrt{s}=14~\mathrm{TeV}$) for $g_{a\gamma\gamma}=0$ and $g_{aWW}=1~\mathrm{TeV}^{-1}$ as a function of the ALP mass $m_a$.
• Figure 3: BDT response distributions for signal (shaded) and total background (outline) at the HL-LHC with $\sqrt{s}=14~\mathrm{TeV}$, for $m_a=750~\mathrm{GeV}$ (left) and $m_a=2000~\mathrm{GeV}$ (right).
• Figure 4: Projected $2\sigma$ and $5\sigma$ discovery sensitivities on the model-independent fiducial production cross section $\sigma(pp\to jj\,a)\times \mathrm{Br}(a\to W^+W^-)$ versus $m_a$ at the HL-LHC ($\sqrt{s}=14~\mathrm{TeV}$, $\mathcal{L}=3~\mathrm{ab}^{-1}$). Assumes $g_{a\gamma\gamma}=0$ and per-mass BDT optimization. Red and blue curves correspond to $2\sigma$ and $5\sigma$, respectively.
• Figure 5: Projected discovery reach on $g_{aWW}$ for photophobic ALPs as a function of $m_a$ (170–4000 GeV) at the HL-LHC ($\sqrt{s}=14~\mathrm{TeV}$, $\mathcal{L}=3~\mathrm{ab}^{-1}$). Solid (dashed) black curves denote $2\sigma$ ($5\sigma$) significances. The projection assumes $g_{a\gamma\gamma}=0$ and uses a per–mass-point BDT optimization; significances are statistical only. For comparison, we overlay $95\%$ CL limits recast from triboson $pp \to Z (\to \nu\bar{\nu}) \gamma \gamma$ at 8 TeV with 20.3 fb$^{-1}$Craig:2018kne, and from Run-2 analyses at 13 TeV: $pp\!\to\!W^\pm W^\pm W^\mp$ with 35.9 fb$^{-1}$ and $pp\!\to\!Z(\to\ell^+\ell^-)\gamma jj$ with 140 fb$^{-1}$Aiko:2024xiv. Also shown are HL-LHC ($14~\mathrm{TeV}$, $3~\mathrm{ab}^{-1}$) $2\sigma$ projections for $pp\!\to\!W^\pm a(\to W^+W^-)$Mao:2024kgx and $pp\!\to\!jj\,a(\to Z\gamma)$Ding:2024djo.