164
5 Direct Searches for New Physics
fourth power of this coupling. Interference effects with SM single top-quark production resulting in the same final state, pp → W bq, need to be taken into account.
ATLAS [911] and CMS [912] have performed searches in this final state, where SM
backgrounds are suppressed by requiring low jet multiplicity and the presence of a
forward jet. The best limits are obtained by the ATLAS analysis, which is based on
a data sample four times larger than the one used in the CMS analysis. The upper
limits on the cross section times branching fraction are between 0.08 and 0.11 pb for
m VLQ = 1.8 and 1 TeV, respectively.
Two searches have been performed by CMS targeting the single production of
B, both based on 13 TeV data with 35.9 fb
−1 . A search performed in the allhadronic final state is optimised for the B → Hb channel, where the H → bb
decay is reconstructed by a H -tagged large-R jet [913]. The H jets are defined
by the pruned jet mass in the range 105–135 GeV and by two b-tagged subjets,
obtained with the soft drop algorithm. Events in the signal region require an H jet
balanced by a high- p T b-tagged small-R jet. Trigger requirements lead to a selection
of H T > 950 GeV, calculated from all small-R jets with p T > 30 GeV. Events are
sorted into four categories, based on the presence of a forward jet and the value
of H T . The low-mass category with H T < 1250 GeV shows higher sensitivity for
signals with m VLQ < 1500 GeV, whereas the multijet background is reduced in the
high-mass category with H T > 1250 GeV, resulting in a better sensitivity for signals
with m VLQ > 1500 GeV. The main background in this search is multijet production,
with only 5–7% from tt production. Other SM processes give negligible contributions. The multijet background is estimated from three sideband regions, obtained
by requiring only one b-tagged subjet and/or changing the soft drop jet mass to
75 < m jet < 105 GeV or m jet > 135 GeV. For the method to work, the subjet btagging has to be uncorrelated from the soft drop jet mass, which has been verified
using simulation. The analysis excludes cross sections times branching fractions
above 0.07 and 0.4 pb for m VLQ = 1.8 and 1 TeV, respectively. The limits worsen by
factors between 1.3 for m VLQ = 1 TeV and 2.1 for m VLQ = 1.8 TeV, when increasing
the relative width of the B from 1% (narrow width approximation) to 30%.
A dedicated search for B/ X → W t is carried out in the +jets channel [914],
where the lepton can originate either from the t → W b → νb decay, or from the W
boson from the B/ X decay. The analysis selects leptons with p T > 55 GeV, which
are identified with a two-dimensional isolation requirement in order to achieve high
selection efficiency for decays of boosted t quarks (see Sect. 4.2.3). The analysis
uses W and t tagging, based on the soft drop jet mass, τ 21 , τ 32 and subjet b tagging.
Selected events are attributed to five categories, defined by the presence of a t tag,
a W tag, two, one or no b-tagged small-R jets. In the t tag category, the VLQ mass
is reconstructed from the four-vectors of the t-tagged jet, the lepton and p
miss
T . In all
other categories, it is reconstructed using combinations of small-R jets, where the
best combination is chosen based on a χ
2 estimator. The data sample is divided into
a signal region with a forward jet and a control region without one. The background
distribution in the reconstructed VLQ mass in the signal region is estimated from
the corresponding distribution in the control region. This allows for a data-driven
background estimation of all SM backgrounds in this search. Residual differences
5 Direct Searches for New Physics
fourth power of this coupling. Interference effects with SM single top-quark production resulting in the same final state, pp → W bq, need to be taken into account.
ATLAS [911] and CMS [912] have performed searches in this final state, where SM
backgrounds are suppressed by requiring low jet multiplicity and the presence of a
forward jet. The best limits are obtained by the ATLAS analysis, which is based on
a data sample four times larger than the one used in the CMS analysis. The upper
limits on the cross section times branching fraction are between 0.08 and 0.11 pb for
m VLQ = 1.8 and 1 TeV, respectively.
Two searches have been performed by CMS targeting the single production of
B, both based on 13 TeV data with 35.9 fb
−1 . A search performed in the allhadronic final state is optimised for the B → Hb channel, where the H → bb
decay is reconstructed by a H -tagged large-R jet [913]. The H jets are defined
by the pruned jet mass in the range 105–135 GeV and by two b-tagged subjets,
obtained with the soft drop algorithm. Events in the signal region require an H jet
balanced by a high- p T b-tagged small-R jet. Trigger requirements lead to a selection
of H T > 950 GeV, calculated from all small-R jets with p T > 30 GeV. Events are
sorted into four categories, based on the presence of a forward jet and the value
of H T . The low-mass category with H T < 1250 GeV shows higher sensitivity for
signals with m VLQ < 1500 GeV, whereas the multijet background is reduced in the
high-mass category with H T > 1250 GeV, resulting in a better sensitivity for signals
with m VLQ > 1500 GeV. The main background in this search is multijet production,
with only 5–7% from tt production. Other SM processes give negligible contributions. The multijet background is estimated from three sideband regions, obtained
by requiring only one b-tagged subjet and/or changing the soft drop jet mass to
75 < m jet < 105 GeV or m jet > 135 GeV. For the method to work, the subjet btagging has to be uncorrelated from the soft drop jet mass, which has been verified
using simulation. The analysis excludes cross sections times branching fractions
above 0.07 and 0.4 pb for m VLQ = 1.8 and 1 TeV, respectively. The limits worsen by
factors between 1.3 for m VLQ = 1 TeV and 2.1 for m VLQ = 1.8 TeV, when increasing
the relative width of the B from 1% (narrow width approximation) to 30%.
A dedicated search for B/ X → W t is carried out in the +jets channel [914],
where the lepton can originate either from the t → W b → νb decay, or from the W
boson from the B/ X decay. The analysis selects leptons with p T > 55 GeV, which
are identified with a two-dimensional isolation requirement in order to achieve high
selection efficiency for decays of boosted t quarks (see Sect. 4.2.3). The analysis
uses W and t tagging, based on the soft drop jet mass, τ 21 , τ 32 and subjet b tagging.
Selected events are attributed to five categories, defined by the presence of a t tag,
a W tag, two, one or no b-tagged small-R jets. In the t tag category, the VLQ mass
is reconstructed from the four-vectors of the t-tagged jet, the lepton and p
miss
T . In all
other categories, it is reconstructed using combinations of small-R jets, where the
best combination is chosen based on a χ
2 estimator. The data sample is divided into
a signal region with a forward jet and a control region without one. The background
distribution in the reconstructed VLQ mass in the signal region is estimated from
the corresponding distribution in the control region. This allows for a data-driven
background estimation of all SM backgrounds in this search. Residual differences