5.5 Dark Matter and Mono-X
187
139 fb
−1 , resulting in an exclusion of m φ < 250 GeV [1082]. The +jets final state has
been analysed in a dedicated ATLAS analysis using 36.1 fb
−1 of 13 TeV data [528].
This analysis considers 16 individual signal regions, each optimised for a particular
decay topology. Hadronic decays of boosted t and W are selected by reclustering
small-R jets with an iterative approach. Large-R jets are obtained with an initial
value R = 3.0, which is then reduced to R = 2m/ p T , where m is either m t or m W .
If a large-R jet loses a large fraction of its p T in this iteration, it is discarded. The
masses of the reclustered jets help to select events with t and W jets. The obtained
upper cross section limits from a combination of all signal regions are compatible
with the limits obtained in the jets+ p
miss
T
[1081] and dilepton final states [1082],
but no exclusions can be derived from this analysis alone. In models with b quark
couplings, the upper cross section limits are about two orders of magnitude above
the predicted signal cross sections [1081], showing the challenge of this final state
at the LHC.
An analysis by CMS in search for tt + χ ¯
χ and bb + χ ¯
χ is based on 2.3 fb
−1
of 13 TeV data [1083], and combines dilepton, +jets and jets+ p
miss
T
final states. In
order to suppress the large tt and Z (→ ν ¯
ν)+jets backgrounds in the jets+ p
miss
T
final
state, a resolved top tagger is employed. In this algorithm, triplets of small-R jets
are built from which a discriminant is constructed using a BDT. The inputs are the
q/g [411] and b tagging discriminants for each jet, the opening angle between the
candidate b jet and the two jets from the W boson decay, and the output value of a
kinematic fit to m t and m W . The working point chosen in this analysis corresponds
to an efficiency of 94% for correctly identifying the tt decay with two resolved top
tags, with a misidentification rate of 48%. In the dilepton and +jets final states,
backgrounds are suppressed with angular selections between the leptons, jets and
p
miss
T . In the final combination, the analyses in the +jets and all-hadronic final states
show the best sensitivity to tt + χ ¯
χ production. The analysis has been updated using
35.9 fb
−1 [1084], resulting in excluded masses of scalar mediators coupling to top
quarks below 160 GeV for m φ = 2m χ and unity couplings, g SM = 1 and g χ = 1.
At 13 TeV, also mono-t searches have profited from jet substructure techniques.
A CMS analysis with 36 fb
−1 of data [1085] has been optimised for a scalar mediator
with flavour-changing couplings, resulting in t + χ ¯
χ production [1049, 1050]. The
same experimental signature of t + p
miss
T
is obtained by a coloured mediator, decaying to a top quark and a dark fermion, t + ψ [1086], which is also considered in this
analysis. The analysis is carried out in the J + p
miss
T
final state, where J denotes a
t-tagged large-R jet. For t tagging, CA R = 1.5 jets are used with p T > 250 GeV.
The jets are corrected for pileup effects using the PUPPI algorithm and groomed
using soft drop with z cut = 0.15 and β = 1. Jets are selected if their soft drop mass
is in the range 110–210 GeV and a b-tagged subjet is found. A BDT is used to
combine substructure information to achieve better background rejection. The input
to the BDT are τ 32 and eleven ratios of generalised ECFs v e
(β)
N
(see Sect. 2.4.2),
with varying parameters β, N and v. In addition, it was found that the variable f rec ,
which describes the deviation of the ratio of the reconstructed values of m W and m t
from its expectation and has originally been defined for the HTTv2 [234], helps to
187
139 fb
−1 , resulting in an exclusion of m φ < 250 GeV [1082]. The +jets final state has
been analysed in a dedicated ATLAS analysis using 36.1 fb
−1 of 13 TeV data [528].
This analysis considers 16 individual signal regions, each optimised for a particular
decay topology. Hadronic decays of boosted t and W are selected by reclustering
small-R jets with an iterative approach. Large-R jets are obtained with an initial
value R = 3.0, which is then reduced to R = 2m/ p T , where m is either m t or m W .
If a large-R jet loses a large fraction of its p T in this iteration, it is discarded. The
masses of the reclustered jets help to select events with t and W jets. The obtained
upper cross section limits from a combination of all signal regions are compatible
with the limits obtained in the jets+ p
miss
T
[1081] and dilepton final states [1082],
but no exclusions can be derived from this analysis alone. In models with b quark
couplings, the upper cross section limits are about two orders of magnitude above
the predicted signal cross sections [1081], showing the challenge of this final state
at the LHC.
An analysis by CMS in search for tt + χ ¯
χ and bb + χ ¯
χ is based on 2.3 fb
−1
of 13 TeV data [1083], and combines dilepton, +jets and jets+ p
miss
T
final states. In
order to suppress the large tt and Z (→ ν ¯
ν)+jets backgrounds in the jets+ p
miss
T
final
state, a resolved top tagger is employed. In this algorithm, triplets of small-R jets
are built from which a discriminant is constructed using a BDT. The inputs are the
q/g [411] and b tagging discriminants for each jet, the opening angle between the
candidate b jet and the two jets from the W boson decay, and the output value of a
kinematic fit to m t and m W . The working point chosen in this analysis corresponds
to an efficiency of 94% for correctly identifying the tt decay with two resolved top
tags, with a misidentification rate of 48%. In the dilepton and +jets final states,
backgrounds are suppressed with angular selections between the leptons, jets and
p
miss
T . In the final combination, the analyses in the +jets and all-hadronic final states
show the best sensitivity to tt + χ ¯
χ production. The analysis has been updated using
35.9 fb
−1 [1084], resulting in excluded masses of scalar mediators coupling to top
quarks below 160 GeV for m φ = 2m χ and unity couplings, g SM = 1 and g χ = 1.
At 13 TeV, also mono-t searches have profited from jet substructure techniques.
A CMS analysis with 36 fb
−1 of data [1085] has been optimised for a scalar mediator
with flavour-changing couplings, resulting in t + χ ¯
χ production [1049, 1050]. The
same experimental signature of t + p
miss
T
is obtained by a coloured mediator, decaying to a top quark and a dark fermion, t + ψ [1086], which is also considered in this
analysis. The analysis is carried out in the J + p
miss
T
final state, where J denotes a
t-tagged large-R jet. For t tagging, CA R = 1.5 jets are used with p T > 250 GeV.
The jets are corrected for pileup effects using the PUPPI algorithm and groomed
using soft drop with z cut = 0.15 and β = 1. Jets are selected if their soft drop mass
is in the range 110–210 GeV and a b-tagged subjet is found. A BDT is used to
combine substructure information to achieve better background rejection. The input
to the BDT are τ 32 and eleven ratios of generalised ECFs v e
(β)
N
(see Sect. 2.4.2),
with varying parameters β, N and v. In addition, it was found that the variable f rec ,
which describes the deviation of the ratio of the reconstructed values of m W and m t
from its expectation and has originally been defined for the HTTv2 [234], helps to