CMS has performed a broad range of searches for particles predicted by
theories of supersymmetry (SUSY), which can in principle explain the
astrophysical dark matter and the low observed value of the Higgs-boson
mass. Because of the many possible theoretical scenarios, the program of CMS
SUSY searches is designed to be as generic as possible.
Early Run 2 searches are designed to be inclusive, but they focus especially on
the pair production of gluinos, for which the production cross section for
the mass range of interest has the largest increase from 8 TeV (Run 1) to 13
TeV (Run 2). In SUSY scenarios motivated by the gauge hierarchy problem, the
gluino typically decays either into a third-generation squark-anti-quark
pair; decays to a generic first- or second-generation squark-anti-quark pair
are considered as well. Many of these scenarios produce multiple b quarks,
together with the characteristic large missing transverse momentum
associated with the lightest supersymmetric particle (LSP), assumed to be a
neutralino.
|
Figure 1 shows simplified-model diagrams for gluino pair production with
three-body gluino decays into a quark (either a light quark, a bottom quark or a top quark),
its corresponding anti-quark, and a neutralino.
Searches for these processes have been performed in all-hadronic (jets + missing transverse momentum), single-lepton (jets +e/μ + missing transverse momentum), and like-sign dilepton channels (jets +
like sign, same or opposite flavor leptons + missing transverse momentum).
|
Figure 2 shows that significant progress has already made in Run 2 for the
exclusion limits in the plane of gluino mass (x-axis) vs. neutralino mass
(y-axis). The value of the gluino mass determines the cross section, with a
very rapid, approximately power-law falloff that limits all analyses at some
high gluino mass value. The value of the neutralino mass, in particular its
splitting with respect to the gluino, determines the momentum spectrum of
final state particles, as well as the missing transverse momentum, limiting
the search sensitivity to some high LSP mass approaching the diagonal of the
plot. Limits are first placed on the production cross section times
branching fraction for each scenario; the displayed boundary curves are then
obtained by assuming 100% branching fraction of the gluino to the given
final state particles and comparing with the theoretical cross section. The
excluded region from the combination of all CMS Run 1 searches (19.5 fb^-1)
extends up to a gluino mass of about 1350 GeV (for low neutralino mass).
This value has already been surpassed by most of the individual searches in
Run 2, using only about 2.2 fb^-1.
Typical SUSY searches look for broad excesses in the extreme tails of the
distributions of missing transverse momentum and hadronic energy and
therefore require excellent detector performance, a detailed understanding
of the background shapes in these tails. The most important background
contributions are measured from data in control regions. Four searches are
performed in the all-hadronic channel, based on complementary approaches
using the kinematic variables MHT (missing HT), MT2, the razor variables,
and alphaT. These searches are also binned in a several variables, such as
HT, the number of jets, and the number of b-jets. Hadronic searches are
typically the most difficult, because they have the most complex background
composition, with contributions from QCD, W+jets, Z+jets, and ttbar. The
variables MT2 and alphaT are particularly powerful in suppressing the QCD
background. The Z + jets background is important in search bins with no
b-jets; this contribution is often determined using a combination of (Z->ll)
+ jets and photon + jets control samples. The single lepton SUSY search, in
contrast, was designed to have a simple background composition dominated by
dilepton ttbar events. This analysis uses MET, MT, and MJ, the sum of the
masses of large-radius jets. Finally, the same-sign dilepton search has very
strong suppression of SM backgrounds and requires a very careful evaluation
of the contribution of leptons from non-prompt sources.
In Run1, CMS observed a small excess
in the opposite-sign dilepton channel. This search as been performed again
in Run 2 and no excess has been observed.
|