5.1 Diboson Resonances
137
While the mass reach by this search is higher, the limits at about 2 TeV are about
an order of magnitude weaker than the ones from dilepton final states, which can be
attributed to larger backgrounds and the inability to reconstruct the resonance mass
in the γ + p
miss
T
final state.
These disadvantages can be alleviated by considering hadronic Z and W decays.
These result in larger signal efficiencies than leptonic channels due to the larger
branching fractions, and therefore in a larger reach in resonance mass. Jet substructure
methods can help to suppress the reducible backgrounds from γ +jet production, and
the reconstruction of the resonance mass helps to improve the sensitivity compared
to the γ + p
miss
T
final state. The first search to consider V tagging in Z γ resonances has
been performed by ATLAS on 3.2 fb
−1 of 13 TeV data [780]. The search includes the
Z → and Z → qq channels, taking advantage of the better signal-to-background
ratio in the Z → channel for low resonance masses and the higher sensitivity of
the Z → qq channel at high masses. The V tagger uses the trimmed jet mass, n trk
and a p T -dependent selection on D 2 to identify Z → qq decays, merged into largeR jets with p T > 200 GeV. The reconstructed Z γ mass distribution is fit with the
power-law function of (5.1) to model the smoothly falling background. The hadronic
channel extends the reach of the analysis to 2.75 TeV, where the observed upper limit
on σ ( pp → X )B(X → Z γ ) is about 10 fb. An analysis by CMS using 35.9 fb
−1 also
considers leptonic and hadronic Z decays [781]. Dilepton final states are selected
using mini-isolation for the lepton identification to retain signal efficiency at high Z γ
masses. Three different categories of large-R jets are used in this search: untagged, τ 21
tagged and b tagged. Jets in the untagged category have p T > 200 GeV and a pruned
jet mass 75 < m jet < 105 GeV. Jets in the τ 21 -tagged category pass a requirement of
τ 21 < 0.45 in addition to the requirements of the untagged category, and jets in the
b-tagged category are required to have a b-tagged subjet in addition to the τ 21 -tagged
category. These three categories are made mutually exclusive by first testing for a
b-tagged jet, then for a τ 21 -tagged jet and lastly for an untagged jet. The untagged
category improves the sensitivity at intermediate and high masses compared to a
previous CMS analysis considering only b- and τ 21 -tagged jets [782]. The b-tagged
category improves the sensitivity by exploiting Z → bb decays with smaller SM
backgrounds. The product of acceptance times signal efficiency increases from 7 to
9% in the untagged category and from 3 to 9% in the τ 21 -tagged category for signal
masses between 0.65 and 4 TeV. In all three categories, the background shape in the
reconstructed Z γ distribution can be described by a power-law function with two free
parameters controlling the shape. The leptonic channels have better sensitivity for
Z γ resonance masses below 2 TeV, while above this value the hadronic channels are
more sensitive. A combination results in limits on σ ( pp → X )B(X → Z γ ) ranging
from 50 to 0.3 fb for masses between 0.35 and 4 TeV, which considerably improves
the limits by the ATLAS search on the same amount of data considering only leptonic
final states [778].
The hadronic V decay channels are the target of a search by ATLAS using
36.1 fb
−1 of 13 TeV data [783]. Similar to the CMS search, this search makes use
of subjet b-tagging and categorises events according to the requirements on subjet
b tagging, D 2 and trimmed jet mass. In addition to these three categories, a fourth
137
While the mass reach by this search is higher, the limits at about 2 TeV are about
an order of magnitude weaker than the ones from dilepton final states, which can be
attributed to larger backgrounds and the inability to reconstruct the resonance mass
in the γ + p
miss
T
final state.
These disadvantages can be alleviated by considering hadronic Z and W decays.
These result in larger signal efficiencies than leptonic channels due to the larger
branching fractions, and therefore in a larger reach in resonance mass. Jet substructure
methods can help to suppress the reducible backgrounds from γ +jet production, and
the reconstruction of the resonance mass helps to improve the sensitivity compared
to the γ + p
miss
T
final state. The first search to consider V tagging in Z γ resonances has
been performed by ATLAS on 3.2 fb
−1 of 13 TeV data [780]. The search includes the
Z → and Z → qq channels, taking advantage of the better signal-to-background
ratio in the Z → channel for low resonance masses and the higher sensitivity of
the Z → qq channel at high masses. The V tagger uses the trimmed jet mass, n trk
and a p T -dependent selection on D 2 to identify Z → qq decays, merged into largeR jets with p T > 200 GeV. The reconstructed Z γ mass distribution is fit with the
power-law function of (5.1) to model the smoothly falling background. The hadronic
channel extends the reach of the analysis to 2.75 TeV, where the observed upper limit
on σ ( pp → X )B(X → Z γ ) is about 10 fb. An analysis by CMS using 35.9 fb
−1 also
considers leptonic and hadronic Z decays [781]. Dilepton final states are selected
using mini-isolation for the lepton identification to retain signal efficiency at high Z γ
masses. Three different categories of large-R jets are used in this search: untagged, τ 21
tagged and b tagged. Jets in the untagged category have p T > 200 GeV and a pruned
jet mass 75 < m jet < 105 GeV. Jets in the τ 21 -tagged category pass a requirement of
τ 21 < 0.45 in addition to the requirements of the untagged category, and jets in the
b-tagged category are required to have a b-tagged subjet in addition to the τ 21 -tagged
category. These three categories are made mutually exclusive by first testing for a
b-tagged jet, then for a τ 21 -tagged jet and lastly for an untagged jet. The untagged
category improves the sensitivity at intermediate and high masses compared to a
previous CMS analysis considering only b- and τ 21 -tagged jets [782]. The b-tagged
category improves the sensitivity by exploiting Z → bb decays with smaller SM
backgrounds. The product of acceptance times signal efficiency increases from 7 to
9% in the untagged category and from 3 to 9% in the τ 21 -tagged category for signal
masses between 0.65 and 4 TeV. In all three categories, the background shape in the
reconstructed Z γ distribution can be described by a power-law function with two free
parameters controlling the shape. The leptonic channels have better sensitivity for
Z γ resonance masses below 2 TeV, while above this value the hadronic channels are
more sensitive. A combination results in limits on σ ( pp → X )B(X → Z γ ) ranging
from 50 to 0.3 fb for masses between 0.35 and 4 TeV, which considerably improves
the limits by the ATLAS search on the same amount of data considering only leptonic
final states [778].
The hadronic V decay channels are the target of a search by ATLAS using
36.1 fb
−1 of 13 TeV data [783]. Similar to the CMS search, this search makes use
of subjet b-tagging and categorises events according to the requirements on subjet
b tagging, D 2 and trimmed jet mass. In addition to these three categories, a fourth