Photon-Fusion Production of W Boson Pairs

• The paper details photon-fusion production via quasi-real photon exchange, emphasizing tree-level t-/u-channel and quartic gauge-boson interactions.
• The study outlines experimental strategies, using isolated lepton selection and rapidity-gap requirements to isolate the exclusive W+W- signal.
• The analysis presents precise cross section measurements and sensitivity to anomalous quartic gauge couplings, providing a benchmark for electroweak precision tests.

Photon-fusion production of $W$ boson pairs refers to the process in which two quasi-real photons, radiated by the incoming protons in high-energy proton-proton ($pp$) collisions, interact to produce a $W^+W^-$ pair. This mechanism, fundamentally electroweak, provides a unique probe of Standard Model (SM) gauge structure—especially the triple and quartic gauge-boson self-interactions, including the $\gamma WW$ and $\gamma\gamma WW$ vertices. While subdominant to the leading Drell–Yan $q\bar q\to W^+W^-$ process in inclusive $WW$ production, photon-induced $WW$ represents a rare channel whose clean experimental signature and sensitivity to new-physics effects (notably anomalous quartic gauge couplings, aQGCs) are essential at the precision frontier of the LHC.

1. Theoretical Framework of Photon-Fusion $WW$ Production

Photon-fusion $W^+W^-$ production in $pp$ collisions is modeled as $pp\to p^{(*)}(\gamma\gamma\to W^+W^-)p^{(*)}$, where $p^{(*)}$ denotes either an intact proton (elastic emission) or a proton that dissociates into a hadronic system (inelastic emission). The process proceeds via three tree-level topologies: - $t$-channel and $u$-channel $W$-exchange diagrams: characterize the SM $\gamma WW$ trilinear coupling. - Quartic $\gamma\gamma W^+W^-$ vertex: genuine four-gauge-boson interaction, central for aQGC studies.

The Equivalent Photon Approximation (EPA), both in collinear and $k_T$-factorized formulations, is used to describe the emission spectrum of quasi-real photons from high-energy protons. The unintegrated photon fluxes $f_\gamma^{(el)}(x,q_T^2)$ (elastic) are governed by proton electromagnetic form factors, while $f_\gamma^{(in)}(x,q_T^2)$ (inelastic) rely on deep inelastic structure functions $F_2$ and $F_L$. The dominant contribution at LHC energies comes from the inelastic–inelastic ($in$–$in$) channel.

The inclusive (total) cross section is schematically:

$$\sigma_{tot}(pp \to p^{(*)}W^+W^-p^{(*)}) = \sum_{i,j=el,in}\int dx_1\,dx_2\, f_\gamma^{(i)}(x_1) f_\gamma^{(j)}(x_2) \hat{\sigma}_{\gamma\gamma\to W^+W^-}(\hat{s}=x_1x_2s)$$

where $\hat{\sigma}$ is the hard $\gamma\gamma\to W^+W^-$ partonic cross section, calculable using the gauge-boson self-interaction vertices.

2. Experimental Strategies and Event Topology

Photon-fusion $WW$ events are experimentally selected by identifying final states characterized by: - Two oppositely charged, isolated leptons ($e^\pm\mu^\mp$), arising from $W$ decays. - Minimal additional hadronic activity near the interaction vertex—enforced by a strict zero-track requirement to isolate events without additional charged particles, enhancing purity against Drell–Yan and other backgrounds. - Lepton $p_T$ and $\eta$ thresholds reflecting detector acceptance (for instance, leading $p_T>24$ GeV, subleading $p_T>15$ GeV, $|\eta_e|<2.5$, $|\eta_\mu|<2.4$ in CMS at 13 TeV (Collaboration, 29 Jan 2026)). - Additional kinematic selections, including dilepton mass $m_{e\mu}>20$ GeV and acoplanarity $A>0.015$, help suppress backgrounds.

The fiducial cross section, $\sigma_{fid}$, is defined in the generator-level phase space matching these selection criteria.

CMS and ATLAS analyses both mandate no extra tracks ($N_{trk}=0$) at the dilepton vertex, which suppresses non-exclusive production and ensures a high-purity $\gamma\gamma\to W^+W^-$ sample (Collaboration, 2020, Collaboration, 29 Jan 2026). In exclusive measurements using dedicated forward proton spectrometers, detection of both outgoing protons further ensures exclusivity and enables full event kinematics reconstruction (Baldenegro et al., 2020).

3. Cross Section Measurements and Differential Properties

Comprehensive measurements at $\sqrt{s}=13$ TeV using the CMS and ATLAS detectors have provided the first observation of photon-fusion $W^+W^-$ at the LHC:

• CMS (138 fb$^{-1}$, 2016–2018):
  • Inclusive cross section: $\sigma_{tot}=643^{+82}_{-78}$ fb
  • Fiducial cross section: $\sigma_{fid}=3.96^{+0.53}_{-0.51}$ fb
  • Standard Model predictions: $631\pm126$ fb (total), $3.87\pm0.77$ fb (fiducial)
  • Both measurements are consistent with the SM (Collaboration, 29 Jan 2026).
• ATLAS (139 fb$^{-1}$): $$\sigma_{fid}=3.13\pm0.31(\text{stat})\pm0.28(\text{syst})$$ fb (Collaboration, 2020).

These cross sections represent about 1–2% of the inclusive $W^+W^-$ rate at central rapidities and low $p_T$, but contribute 10% or more of the total at high $p_T$ ($>200$ GeV) and high invariant mass ($M_{WW}$ above 800 GeV) (Luszczak, 2014, Luszczak et al., 2014, Bierweiler et al., 2012). The differential spectra are characterized by:

• A relatively flat rapidity distribution for the $W$ bosons over $|y_W|<2.5$.
• Harder $p_T$ spectra than $q\bar q$-initiated modes—photon fusion dominates the $p_T$ and $M_{WW}$ tails.

Cross section uncertainties are dominated by the modeling of photon fluxes, experimental efficiencies, and the treatment of rapidity-gap survival factors, especially in the presence of proton dissociation and pileup (Collaboration, 29 Jan 2026, Łuszczak et al., 2020).

4. Elastic, Inelastic, and Exclusive Production: Modeling and Rapidity Gap Survival

Photon emission can occur elastically or with proton dissociation:

• Elastic–elastic: both protons remain intact.
• Elastic–inelastic (or inelastic–elastic): one proton remains intact, the other dissociates.
• Inelastic–inelastic: both protons dissociate.
• Central exclusive production: both protons remain intact with no additional hadronic activity; can be directly tagged with forward proton detectors (Baldenegro et al., 2020).

Inelastic photon emission dominates the total $\gamma\gamma\to W^+W^-$ cross section at LHC energies, but these events often produce hadronic remnants that populate the forward detector regions. Imposing large central rapidity gaps (absence of charged particles in $|\eta|<2.5$) suppresses inelastic channels:

• Rapidity-gap survival factors $S_{R,SD}$ (single dissociation) and $S_{R,DD}$ (double dissociation) quantify the signal loss due to secondary hadron activity.
• Approximate factorization holds: $S_{R,DD}\approx S_{R,SD}^2$ (Łuszczak et al., 2020, Szczurek et al., 2019).
• For $|\eta|<2.5$ at 13 TeV, $S_{R,SD}\approx0.80$, $S_{R,DD}\approx0.65$; the overall taming of the cross section due to rapidity-gap requirements is in the $20$–$30$% range.
• Proton-dissociative fluxes rely critically on up-to-date structure function parametrizations (ALLM97, LUX-like, MNSZ2017), with resulting cross section uncertainties at the $\pm$20% level (Luszczak et al., 2018, Szczurek et al., 2019).

Forward proton tagging allows for the clean isolation of exclusive $\gamma\gamma\to W^+W^-$ processes, reducing background and enhancing the sensitivity to anomalous couplings (Baldenegro et al., 2020).

5. Sensitivity to Anomalous Quartic Gauge Couplings and Effective Field Theory Interpretation

Photon-fusion $W^+W^-$ is uniquely sensitive to quartic gauge boson couplings, especially the $\gamma\gamma WW$ vertex. Constraints are set in the context of both dimension-6 (operators $a_0^W$, $a_C^W$) and dimension-8 (operators $f_{M,i}/\Lambda^4$, $f_{T,i}/\Lambda^4$) effective field theory (EFT) frameworks:

• In CMS at 13 TeV, stringent bounds are placed using profile likelihood scans and reweighting of signal templates (Collaboration, 29 Jan 2026):
  • $f_{T5}/\Lambda^4\in[-0.16,+0.13]$ TeV$^{-4}$ (most stringent among the $f_T$ series)
  • $f_{M2}/\Lambda^4\in[-0.49,+0.50]$ TeV$^{-4}$
  • First-time constraints for CP-odd operators (e.g., $f_{\tilde{T}2}/\Lambda^4\in[-0.03,+0.03]$ TeV$^{-4}$)
• Central exclusive measurements (with forward proton detectors) provide complementary and competitive bounds on $a_0^W$, $a_C^W$ down to $3.7\times10^{-7}$ GeV$^{-2}$ and $9.2\times10^{-7}$ GeV$^{-2}$, respectively, at 14 TeV and 300 fb$^{-1}$ integrated luminosity (Baldenegro et al., 2020).
• Sensitivity to aQGCs increases with $W^+W^-$ mass and $p_T$ due to the quadratic and quartic scaling of EFT contributions relative to the SM, making the high-mass kinematic tails crucial for new-physics searches.

Modern studies now incorporate, for the first time, CP-odd operator constraints in this channel at the LHC (Collaboration, 29 Jan 2026). These results are central to global EFT fits of the electroweak sector.

6. Numerical Summary and Key Phenomenological Features

A collation of integrated and differential cross-section benchmarks, uncertainties, and topology contributions is given below (at $\sqrt{s}=13$ TeV):

Channel Topology	Cross Section (pb)	Relative Fraction (%)
$q\bar q\to WW$ (total)	$\sim$80–120	Dominant
$\gamma\gamma$-fusion	1.3–1.8	1–2 in inclusive, up to 10–30 in tails
Elastic–elastic	0.27	$\sim$15–20 of $\gamma\gamma$
Inel–inel	1.1	$>$60 of $\gamma\gamma$

Differentially:

• $d\sigma/dM_{WW}$ falls steeply; by $M_{WW}\sim1$ TeV, $\gamma\gamma$ and $q\bar q$ become comparable.
• $d\sigma/dp_T^W$ peaks at $p_T\sim30$–$50$ GeV for $\gamma\gamma$, but the fraction rises to 10% at $p_T>200$ GeV (Luszczak, 2014, Luszczak et al., 2014, Bierweiler et al., 2012, Luszczak et al., 2018).
• Polarization fractions are stable, with transversely polarized $W$ pairs dominating.

At very high $p_T$ and $M_{WW}$, omission of this channel significantly biases precision SM and BSM studies (Bierweiler et al., 2012).

7. Implications for Electroweak Precision Physics and New-Physics Searches

Photon-fusion $WW$ production, though subleading in total rate, is pivotal for:

• Directly probing the SM quartic $\gamma\gamma WW$ coupling;
• Setting stringent constraints on dimension-6 and dimension-8 EFT coefficients for aQGCs, notably in previously unprobed CP-odd directions (Collaboration, 29 Jan 2026);
• Complementing other di-boson ($VV$) and vector boson scattering (VBS) observables in global electroweak fits.

Accurate modeling, including inelastic photon fluxes, rapidity-gap survival, and full differential kinematics, is mandatory for exploiting the full sensitivity of LHC data to SM and BSM structures (Łuszczak et al., 2020, Luszczak et al., 2018, Szczurek et al., 2019, Baldenegro et al., 2020).

Systematic uncertainties are predominantly theory-driven, particularly from the photon PDF/structure function modeling, but recent collider measurements now calibrate many aspects experimentally (Collaboration, 2020, Collaboration, 29 Jan 2026).

In conclusion, photon-fusion $W$ boson pair production forms a precision electroweak benchmark, offers a uniquely clean window on gauge-boson self-interactions, and delivers leading sensitivity to anomalous quartic couplings at the LHC (Collaboration, 29 Jan 2026, Baldenegro et al., 2020, Collaboration, 2020).