Combined effective field theory interpretation of measurements sensitive to quartic gauge boson couplings in $pp$ collisions at $\sqrt{s}=13$ TeV with the ATLAS detector

Abstract

A combination of measurements sensitive to anomalous quartic electroweak gauge boson couplings is presented using proton-proton collision data collected by the ATLAS detector at $\sqrt{s} = 13$ TeV at the LHC. Contributing analyses include measurements of vector-boson scattering in numerous final states as well as a tri-boson measurement. The combined measurement is used to constrain anomalous electroweak boson quartic self-couplings that result from dimension-8 operators in the Éboli model using an effective field theory. Results are presented as 68% and 95% confidence level intervals parameterised by one or two Wilson coefficients, both with and without unitarity constraints applied. Theoretical bounds from unitarity and positivity are overlaid where relevant. Confidence intervals obtained from simultaneous profiled fits to all Wilson coefficients are also presented.

Figures (5)

• Figure 1: Representative Feynman diagrams for a) vector-boson scattering with triple-gauge vertices, b) vector-boson scattering with quartic-gauge vertices, c) tri-boson production with triple-gauge vertices, and d) tri-boson production with quartic-gauge vertices. The aQGCs act only on diagrams b) and d) through anomalous values or vertices forbidden in the SM.
• Figure 2: Contributions of the individual channels to the combination (top), expected and observed 68% and 95% confidence level intervals from the combined fit (middle) and the probed energy scale for two illustrative values of the Wilson coefficients (bottom). No constraints from unitarity bounds are applied. The energy scale of the observed unitarization cut-off as evaluated through a clipping procedure is overlaid on the bottom panel. No unitarization cut-off is found for the $f_{S1}$ operator where the experimental interval is always weaker than the unitarity constraint.
• Figure 3: Contributions of the individual channels to the combination (top), expected and observed 68% and 95% confidence level intervals from the combined fit (middle) and the probed energy scale for two illustrative values of the Wilson coefficients (bottom). Unitarity is preserved by setting higher dimension operator contributions to zero above a clipping threshold of 1.5 $\text{Te V}$. The energy scale of the observed unitarization cut-off as evaluated through a clipping procedure is overlaid on the bottom panel. No unitarization cut-off is found for the $f_{S1}$ coefficient where the experimental interval is always weaker than the unitarity constraint.
• Figure 4: Expected and observed 95% confidence level confidence intervals on the a) $f_{S02}$, b) $f_{M0}$ and c) $f_{T0}$ Wilson coefficients as a function of clipping threshold. Unitarity and positivity bounds are shown in green and blue respectively. Unitarized intervals are taken to be the intersection between the interpolated confidence intervals and the unitarity bound.
• Figure 5: Expected (red) and observed (black) 68% (dashed) and 95% (solid) confidence level contours when considering two of the $f_{M2}$, $f_{M3}$, $f_{M4}$, and $f_{M5}$ coefficients simultaneously alongside the relevant cross-term. The expected (greyed-red) and observed (grey) negative log-likelihood (NLL) curve for each of the operators when considering just one coefficient at a time is also shown. Unitarity constraints are not visible and positivity bounds are overlaid as a blue filled region.