Searching for Supersymmetry at the LHC
Jeffrey D. Richman (UC Santa Barbara)
Seminar, Department of Physics, University of New Mexico,
Albuquerque, November 6, 2014
Curnet
LHC ring: 2 separate magnetic “highways”
CMS Experiment,
(Cessy, France)
Proton beam
Proton beam
LHCb Experiment
Alice Experiment
ATLAS Experiment
Meyrin, Switzerland
• 9300 magnets, including 1232 15-meter dipoles.
• Radio-frequency EM cavity devices to accelerate beams (8/beam; 40 MHz)
Outline
• Basics of supersymmetry
• The Higgs, SUSY, and
“naturalness”
• Overview of SUSY search
strategies.
• Examples: Searching for
gluinos, EWKinos, scalar
quarks, dark matter,...
• Prospects for Run 2
• Conclusions
Drawing courtesy Sergio Cittolin
CMS PUBLIC SUSY RESULTS
https://twiki.cern.ch/twiki/bin/vie
w/CMSPublic/PhysicsResultsSU
S
Most results use 19.5 fb-1 (√s =8 TeV).
2
Supersymmetry transformations
• SUSY xf’s map fermionic and bosonic degrees of
freedom onto each other.
• Q = generator of SUSY transformation
Q must be fermionic in
character!
boson (J=0 or J=1)
fermion (J=1/2)
• The charges (interaction couplings) are unchanged.
• Doubles the numbers of degrees of freedom in the
particle spectrum (but CPT did that too!)
• Unlike CPT, don’t see SUSY partners with same
masses as SM  if SUSY exists, it must be broken.
Particle content of the SM
Quarks: spin-1/2
Leptons: spin-1/2
-
c
t
up
charm
top
electron
muon
tau
d
s
b
ne
nm
nt
down
strange
bottom
electron
neutrino
u
e
g
gluon (8)
EM force
muon
neutrino
t
-
tau
neutrino
Higgs boson: spin-0
Gauge bosons: spin-1
Strong force
m
-
and field vacuum expectation value
Weak force
g
Z
photon
Z boson
0
W
+
W
W bosons
-
H
Higgs boson
Simple (naïve) SUSY spectrum
scalar quark
scalar lepton
squarks: spin-0
u
up
squark
d
down
squark
c
charm
squark
sleptons: spin-0
e
t
top
squark
-
m
-
selectron
smuon
s
b
ne
nm
strange
squark
bottom
squark
electron
sneutrino
muon
sneutrino
t
-
stau
nt
tau
sneutrino
Gauginos: spin-1/2 “-ino”  J=1/2 Higgsino: spin-1/2
Strong force
g
gluino (8)
EM force
g
photino
Weak force
Z
0
Zino
W
+
Wino
W
-
H
Higgsino
Scalar SUSY particles and chiral multiplets
• The SM is a chiral theory: the L and R chiral
projections of the fermion fields have different
EW interaction quantum numbers.
– L projections are SU(2)L doublets
– R projections are SU(2)L singlets
• Each chiral projection of a SM fermion has SUSY
scalar partner (preserving degrees of freedom).
t
partner of the R-handed e-; has J=0
SUSY spectrum in EW sector (MSSM)
Particle
-
Z
g
H
h
A
Total
J
Degrees of
freedom
Particle
J
Degrees of
freedom
1
3
1/2
2
1
3
1/2
1
3
1
2
0
1
0
1
0
Particle
J
Degrees of
freedom
1/2
2
2
1/2
2
1/2
2
1/2
2
1/2
2
1/2
2
1/2
2
1/2
2
1/2
2
1/2
2
1
1/2
2
1/2
2
0
1
1/2
2
c10
c 20
c 30
c 40
1/2
2
0
1
16
Total
Z W0
g B
H
h
Total
16
Gauginos = SUSY partners of SM gauge bosons
Higgsinos = SUSY partners of higgs bosons
Neutralinos = mix of neutral gauginos and higgsinos
Charginos = mix of charged gauginos and higgsinos
EWKinos = term that denotes neutralinos or charginos
Mixing
16
If lightest neutralino is LSP, then
can be dark matter candidate.
The gluino ( g) is special: because of
color, it cannot mix with any other
particles.
Why are we still looking for SUSY?
Hierarchy problem Unification of couplings
~1018
Atoms:
4.6%
GeV
Planck scale
(quantum gravity )
SM
(no SUSY)
Dark
energy:
71.4%
Separation of scales
is stabilized by SUSY.
~102 GeV
Electroweak scale
(unstable in SM)
Dark matter
Minimal
SUSY
Dark
matter:
24%
SUSY provides dark
matter candidate particle
WM
(Lightest Supersymmetric
AP
Particle); in MSSM this is
neutralino.
The Higgs and the Gauge Hierarchy Problem
• Evidence is very strong that the new particle
discovered at m ≈ 125 GeV is a/the Higgs boson,
with the quantum numbers JPC = 0++ (scalar).
• Assuming that it is an elementary scalar particle,
the Higgs mass is subject to enormous shifts due
to short distance quantum corrections.
• These corrections pull the Higgs mass up to a high
physical scale, e.g., the Planck scale!
f = SM fermion,
e.g., top quark
S = SUSY scalar
partner, e.g., top squark
SUSY amplitudes tame the quantum corrections!
Scalar particles and fine tuning
• Fundamental scalar fields have the problem of
quadratic divergences to the scalar mass
squared. These arise from loop-corrections to
the mass, which are generically for spin 0:
dominant at low mt
In the Standard Model:
“Natural SUSY endures”: the current fashion
Only part of the SUSY spectrum an be constrained by
naturalness considerations.
M. Papucci, J.T. Ruderman, and A. Weiler http://arxiv.org/abs/1110.6926
“Natural SUSY endures”: the current fashion
Only part of the SUSY spectrum an be constrained by
naturalness considerations.
M. Papucci, J.T. Ruderman, and A. Weiler http://arxiv.org/abs/1110.6926
Focus of SUSY searches
SUSY breaking
• SUSY, if it exists, is clearly a broken symmetry because
partners with masses equal to the SM particles would
already have been found.
• SUSY breaking is an complex subject with various
scenarios; occurs in “hidden sector”; transmitted to
MSSM particles via...
– gravity mediation heavy gravitino (G), couplings ≈gravity
– gauge mediation very light gravitino (eV range); is LSP!
• Whatever the breaking mechanism, SUSY particles still
have the same SM gauge couplings as their ordinary
SM partners. Key point when thinking about decay
modes. Your intuition for the SUSY Particle Date Book is
good!
Installing muon readout electronics
Superconducting solenoid:
B-field parallel to beam axis
Endcap detectors
Central barrel detectors
Spring – summer 2014
Proton-proton collisions
proton
proton
u
u
“partons”
gluon
u
u
u
u
d
d
• Parton momentum transverse to the beam is negligible.
• Parton momentum along the beam direction (z) is large
and unknown, except on a statistical basis.
å
p
i
x
0
i=colliding partons
å
i=colliding partons
p
i
y
0
momentum
transverse
to beam is zero
å
i=colliding partons
pzi ¹ 0
Higgs boson mass peaks
H ® ZZ ®
*
+ - +
Events / 3 GeV
30
20
10
0
- H ® gg
200
Data
mH = 126 GeV
Zg *, ZZ
Z+X
CMS preliminary
100
800
s = 7 TeV: L = 5.1 fb-1
s = 8 TeV: L = 19.6 fb-1
400
m4l [GeV]
• Discovery was based on searches for narrow peaks at a consistent
mass in multiple search channels.
• The determination of the event yield is greatly simplified.
• The masses are Lorentz-invariant quantities. Any reference frame is OK!
An example of a “natural” SUSY model (“NM3”)
Dark
matter:
24%
Dark
energy:
71.4%
WMAP
An example of a “natural” SUSY model (“NM3”)
Dark
matter:
24%
Dark
energy:
71.4%
mh = 125 GeV
WMAP
An example of a “natural” SUSY model (“NM3”)
Strongly interacting sector:
• Particles with color (QCD) charge.
• Large production cross section
if mass is not too high.
Dark
energy:
71.4%
WMAP
Dark
matter:
24%
An example of a “natural” SUSY model (“NM3”)
Electroweak sector:
• EM/weak interactions
• Major impact on
decays of colored
particles
• Generally low
production cross
sections, but may
have lower masses.
• LSP lives here!
Dark
matter:
24%
Dark
energy:
71.4%
WMAP
Lightest Supersymmetric
Particle (LSP)
10
~~
tt*
q= u,d,s,c L,R
LPCC SUSY s WG
+ c~ c~
Stop pair
production
Electroweak
production
± 0 ~~
c~ c~ qq*
Strong production
1
10-1
10-2
~+~-
10-3 l l
g~~g
105
103
102
10
LHC SUSY Cross Section Working Group
104
10-4
200
400
600
800
1000
1200
1400
1600
arXiv:1206.2892
SUSY sparticle mass [GeV]
https://twiki.cern.ch/twiki/bin/view/LHCPhysics/SUSYCrossSections
NLO(-NLL)
s(pp® and
SUSY)
[pb]
Big themes: production
decays
#events in 20 fb-1 8TeV LHC data
https://twiki.cern.ch/twiki/bin/view/LHCPhysics/SUSYCrossSections
6
10
~~
tt*
q= u,d,s,c L,R
LPCC SUSY s WG
+ c~ c~
Stop pair
production
Electroweak
production
± 0 ~~
c~ c~ qq*
Strong production
1
10-1
10-2
~+~-
10-3 l l
g~~g
Exploring 𝜎 = ~1 fb  ~1 pb
105
103
102
10
LHC SUSY Cross Section Working Group
104
10-4
200
400
600
800
1000
1200
1400
1600
arXiv:1206.2892
SUSY sparticle mass [GeV]
https://twiki.cern.ch/twiki/bin/view/LHCPhysics/SUSYCrossSections
NLO(-NLL)
s(pp® and
SUSY)
[pb]
Big themes: production
decays
#events in 20 fb-1 8TeV LHC data
https://twiki.cern.ch/twiki/bin/view/LHCPhysics/SUSYCrossSections
7
Measurements of SM processes in CMS
Production Cross Section,
105
104
103
102
10
1
10-1
Feb 2014
Z+n-jets Z
W + ³ 1 jet
W + ³ 2 jets
Z+ ³ 1 jet
Di-boson
Typical SUSY cross sections: ~few fb to hundreds of fb
CMS Preliminary
7 TeV CMS measurement (L £ 5.0 fb-1)
8 TeV CMS measurement (L £ 19.6 fb-1)
7 TeV Theory prediction
0.1 pb = 100 fb
8 TeV Theory prediction
Z+ ³ 2 jets
Top-quark Higgs
W + W+γ
³ 3 jets
production
Wg
tt production
Z+ ³ 3 jets Zg
production
W + ³ 4 jets
WW
tt +1 jet
1t (t - chan)
Z+ ³ 4 jets
tW gg ® H
WZ
tt + 2 jets
ZZ
tt + 3 jets 1t (s - chan)
tt g qqH (VBF)
VH
ttZ
ttH
CMS 95%CL limit
gg®
W ³1j ³2j ³3j ³4j Z ³1j ³2j ³3j ³4j Wg Zg WW WZ ZZ EW WVg tt 1j 2j 3j tt-ch tW ts-ch ttg ttZ ggH VBF VH ttH
qqH
WW qqll
Th. DsH in exp. D s
W+
W n-jets
s [pb]
Gluino decay
• SUSY preserves the gauge symmetries, so the SUSY
partners of the gluons must also transform according
to the 8-dimensional representation of SU(3)C.
• Fundamental vertex for
has same
coupling strength as that for
.
SM
g
u
u
SUSY
g
u
uL , uR
g
u
uL , uR
Gluino pair production and decay to stop
t
g
g
t
g
g
t
t
c
0
1
missing
momentum c 0
1
t
t
Signature: 4 top quarks + pTmiss
LSPs are unobserved: cannot reconstruct mass peaks!
Direct pair production of top squarks
Example: direct stop production with decay to neutralinos or
charginos.
t
g
g
g
t
t
b
c
c
t
0
1
0
1
b
Sensitivity of the searches will depend strongly on the neutralino mass.
The channel with
has sensitivity to lower stop mass.
Anatomy of a background: ttbar
Extremely common background in NP searches.
• looks like low-mass SUSY!
• large real MET from neutrinos in leptonic decays.
• high jet multiplicity, including b jets.
W
b
p
1. EVENT ENVIRONMENT
• Effects of pileup:
isolation, jets, MET,
vertices
• Underlying event.
t
b
t
-
3. DECAY CHAIN
• W polarization
• Final-state radiation
• Decay branching fractions
n
+
+
Each 2-body system
shown in 2-body rest
frame.
p
2. PRODUCTION
• pT distributions of t and tbar
(affected by parton distribution
functions, QCD renorm &
factorization scales)
• Effect of initial-state radiation
• Spin correlations of t and tbar
W
-
n
Most “SUSY-like” process in SM: pp  tt
Candidate event
for process
pp ® tt
t ® bW + ; W + ® e+n e
t ® bW - ; W - ® q1q2
Missing momentum vector
from neutrino
0
Data analysis
for SUSY searches
c2
SIGNAL: Characterize kinematic
distributions using simulated
(MC) samples.
BACKGROUND: Characterize
kinematic distributions using
simulated (MC) samples; validate
simulation using control samples
in data.
DEFINE SIGNAL REGIONS
(without looking at this
neighborhood of the event
sample).
Use SIGNAL MC to determine
SELECTION EFFICIENCY
PREDICT BACKGROUND
contribution in SIGNAL regions,
ideally using CONTROL REGIONS.
Extrapolation factors from MC,
with uncertainties determined
from agreement in other control
regions.
What could
possibly
go wrong?
SUSY Search Challenges
1. Presence of LSP in each SUSY decay chain (R-parity
conserving models) no observable mass peak for
parent particle.
2. Looking for excess yields in the tails of distributions of
kinematic variables, such as missing transverse
momentum.
3. The background determination is much more
complicated, since one cannot simply use mass
sidebands.
4. Many potential problems. Robust and complementary
approaches to background determinations are
extremely valuable.
What can possibly go wrong?
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Yield in signal region can be biased by tuning selection cuts on the data.
People often stop looking for mistakes when they obtain a “desirable” result.
Background shape or normalization can be estimated incorrectly.
Understanding the background in one kinematic region does not necessarily
mean that you understand it in another region.
Shapes used in fit may not adequate to describe the data. (Especially
worrisome in multidimensional fits!)
Trigger efficiencies may not be acccounted for & can bias sig. or backgrounds.
Systematic errors may be underestimated or incomplete. Assumptions may be
wrong!
Correlations may not taken into account.
Backgrounds peaking under signal may not correctly determined.
Signal significance may not be estimated correctly.
Signal can created artificially as “reflection” of another peaking process.
Changes in experimental conditions may not fully taken into account.
Average of many bad measurements might not give a good measurement.
Bugs in the program!
Advisor in a hurry. Need to finish thesis! No time to look for more problems.
A superposition of several of the above effects!
The penta-quark phenomenon: 2002-2005
Slide courtesy of R. Schumacher
Photoproduction on Deuteron Q+
CLAS-d1
LEPS-C
CLAS g11
SAPHIR
Photoproduction on Proton pKs0
Photoproduction on Proton nK+K-p+
CLAS-d2
LEPS-d2
LEPS-d
CLAS-p
DIANA
Exclusive K + (N) → pKs0
Hermes
Inclusive lepton + D, A → p Ks0
SPHINX
JINR
SVD2
p + A → pKs + X
0
BELLE
BaBar
ZEUS nBC
p + p → pKs0 + S+
HyperCP
SVD2
COSY-TOF HERA-B
Other Q+ Upper Limits
BES J,Y
CDF
FOCUS
WA89
ALEPH, Z
p + p (or A) → X
--
NA49/CERN
+X
Inclusive Q + + → p K+
Inclusive
Q0c
→
D(*) -
HERA-B
ALEPH
WA89
ZEUS
STAR/RHIC
ZEUS
H1/HERA
p
E690
ZEUS
ALEPH FOCUS
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
2002
2003
2004
From Particles and Nuclei International Conference, Santa Fe, 2005
1
2
3
4
5
6
7
8
2005
9
10
11
12
SUSY Search Strategies
1. Large missing momentum transverse to the beam axis.
2. High mass scales (but difficult to isolate decay products
of a specific SUSY parent particle)
3. Long decay chains  large number of jets and isolated
leptons. But mass splittings are unknown, so don’t
know energy distributions of these objects.
4. Natural SUSY  3rd generation quarks (t, b)  b-jet
tagging is extremely important.
5. In models with Gauge Mediated SUSY breaking, can
have high pT photons.
6. In some scenarios, can have Higgs bosons produced in
the decay chains.
CMS SUSY Results (ICHEP 2014)
~ ~0
q® q c
~ ~0
g ® qq c
~ ~0
g ® bb c
~ ~0
g ® tt c
~ ~ ~0
g ® t( t ® t c )
~ ~± ~0
g ® qq(c ® Wc )
~ ~ ~± ~0
g ® b( b ® t( c ® Wc ))
1
x = 0.25
x = 0.05
600
x = 0.50
x = 0.75
x = 0.95
x = 0.50
x = 0.95
xx==0.05
0.50
800
x = 0.20
x = 0.50
1000
R-parity violating
SUSY
1200
Summary of CMS SUSY Results* in SMS framework
m(mother)-m(LSP)=200 GeV
400
SUS 13-019 L=19.5 /fb
SUS-14-011 SUS-13-019 L=19.3 19.5 /fb
SUS-13-007 SUS-13-013 L=19.4 19.5 /fb
SUS-13-008 SUS-13-013 L=19.5 /fb
SUS-13-013 L=19.5 /fb
SUS-13-008 SUS-13-013 L=19.5 /fb
SUS-13-019 L=19.5 /fb
SUS-14-011 L=19.5 /fb
SUS-13-011 L=19.5 /fb
SUS-13-014 L=19.5 /fb
SUS-13-024 SUS-13-004 L=19.5 /fb
SUS-13-024 SUS-13-004 L=19.5 /fb
SUS-13-018 L=19.4 /fb
SUS-13-008 SUS-13-013 L=19.5 /fb
SUS-13-008 L=19.5 /fb
SUS-13-006 L=19.5 /fb
SUS-13-006 L=19.5 /fb
SUS-14-002 L=19.5 /fb
SUS-13-006 L=19.5 /fb
SUS-14-002 L=19.5 /fb
SUS-14-002 L=19.5 /fb
SUS-13-006 L=19.5 /fb
SUS-13-006 L=19.5 /fb
2
~ 0~ ± ~ 0~ 0
c c ® ll nc c
2
~ +~ - + - ~ 0~ 0
c c ® l l nnc c
~ 0~ 0 ~ 0~ 0
c c ®ZZc c
~ ± 2~ 0 2 ~ 0 ~ 0
c c ®WZ c c
2
~ 0~ 0
~ 0~ 0
c c ®HZ c c
2 2
~ 0~ 0
~ ±~ 0
c c ®HWc c
~ 0~ 0
~ 0 ~ ±2
c c ® lt nc c
2
~ 0~ 0
~ 0~ ±
c c ® t tt n c c
233
200
SUS-12-027 L=9.2 /fb
SUS-12-027 L=9.2 /fb
SUS-12-027 L=9.2 /fb
SUS-12-027 L=9.2 /fb
SUS-12-027 L=9.2 /fb
EXO-12-049 L=19.5 /fb
EXO-12-049 L=19.5 /fb
SUS-13-013 L=19.5 /fb
SUS-12-027 L=9.2 /fb
SUS-12-027 L=9.2 /fb
SUS-12-027 L=9.2 /fb
SUS-12-027 L=9.2 /fb
SUS-12-027 L=9.2 /fb
SUS-12-027 L=9.2 /fb
SUS-12-027 L=9.2 /fb
SUS-13-003 L=19.5 9.2 /fb
SUS-12-027 L=9.2 /fb
SUS-13-003 L=19.5 9.2 /fb
SUS-13-003 L=19.5 /fb
SUS-13-006 L=19.5 /fb
R
~
g ® qlln l
122
~
g ® qlln l
123
~
g ® qlln l
233
~
g ® qbt m l '
231
~
g ® qbt m l '
233
~
g ® qqb l '
113/223
~
g ® qqq l '
112
~
g ® tbs l '
323
~
g ® qqqq l '
112
~
q ® qlln l
122
~
q ® qlln l
123
~
q ® qlln l
233
~
q ® qbt m l '
231
~
q ® qbt m l '
233
~
q ® qqqq l '
112
R
~
t ® men t l
122
R
~
t ® m tn t l
123
~R
t ® m tn t l
233
R
~
t ® tbt m l '
0
~ ~0
l®lc
~ ~0
b ®b c
~ ~0
b ® tW c
~ ~0
b ® bZ c
1
~ ~0
t ®t c
~ ~+ ~0
t ® b( c ® Wc )
~ ~0 ~0
t ® t b c ( c ® H G)
~ ~ ~0
t ® (t ® t c ) Z
~2 ~1 ~ 01
t ® (t ® t c ) H
2
gluino pair
production
squark pair production
top squark pair production
bottom squark pair production
*Observed limits, theory uncertainties not included
Only a selection of available mass limits
Probe *up to* the quoted mass limit
neutralino/chargino
pair production
slepton pair production
gluino production
squark
stop
sbottom
EWK gauginos
slepton
RPV
ICHEP 2014
m(LSP)=0 GeV
CMS Preliminary
For decays with intermediate mass,
mintermediate = x×mmother+(1-x)×mlsp
1400
1800
1600
Mass scales [GeV]
Generic hadronic SUSY search using MHT
SUS-13-012
https://twiki.cern.ch/twiki/bin/view/CMSPublic/PhysicsResultsSUS13012
• Signature: Jets + MHT; events with leptons are vetoed
– Jets: ≥3 jets with pT > 50 GeV, no b-tagging.
– Veto event if MHT vector is ≈aligned with any of 3 leading jets.
• Bin data in
HT =
å
pTj
j= jets
– HT
– missing HT (MHT)
HT = HT = pTj
j= jets
– Jet multiplicity (3—5, 6—7, ≥8 jets)
å
• Background estimation: largely data driven.
• ttbar with Wl 𝜈
• W  l 𝜈 + jets
• ttbar with W𝜏 (h) 𝜈
• W  𝜏 (h) 𝜈 + jets
Control sample: Singlelepton + jets + MHT
• Z𝜈𝜈 + jets
• QCD multijet events
MHT ~ aligned with high pT jet.
Control samples:
𝛾 + jets,
Z(𝜇𝜇)+jets
Control sample:
Multijets with re-balance and
smear procedure
23
Distribution in bins of N(jets), HT , and HT
CMS SUS-13-012
HT : 0.8-1.0 TeV
HT : 1.0-1.2 TeV
24
Statistical interlude
• Consider the bin with
– N(observed) = 9 events
– N(background) = 0.8 ± 1.7 events
See CMS PAS SUS-13-012,
Table 1, p. 10
Njets: 6-7
HT: 500-800 GeV
MHT>450 GeV
• First, let’s ignore the uncertainty on the background. What
is the probability for a Poisson with μ=0.8 to fluctuate to at
least 9 events?
– Prob( n≥9 | μ =0.8 ) = 1.8 × 10-7
Have we discovered new physics?
• NO! The uncertainty is crucial!
– Prob( n≥9 | μ = 0.8 ± 1.7) ≈ 0.15
• This example highlights the importance of quantifying the
uncertainties on the SM backgrounds.
25
Search for generic jets and MET: results
• Simplified model exclusion plots
CMS SUS-13-012
1. Compute excluded cross section for each model in param space
2. Compare to reference cross section to see if model excluded
• Assume 100% branching fraction for stated process!
mLSP (GeV)
CMS Preliminary, 19.5 fb-1, s = 8 TeV
0
1
pp ® ~q ~q, ~
q ® q ~c NLO+NLL exclusion
R
Observed ± 1 stheory
Expected ± 1 sexperiment
L
~ ~
~
q +~
q (~u,d,~
s,c)
one light ~q
(L + R)
600
500
400
300
200
100
1400
10
1
10-1
10-2
10-3
10
1
10-1
-3
10-2
10
95% C.L. upper limit on cross section (pb)
m~q (GeV)
1200
decreasing production
cross section
0
1
Signal efficiency
increases away
from diagonal
 excluded 𝜎 falls
1000
m~g (GeV)
CMS Preliminary, 19.5 fb-1, s = 8 TeV
800
pp ® ~g ~g, ~g ® q q ~c NLO+NLL exclusion
Observed ± 1 stheory
Expected ± 1 sexperiment
600
Gluino pair
production
300 400 500 600 700 800 900 1000
1200
800
600
400
200
400
1000
mLSP (GeV)
95% C.L. upper limit on cross section (pb)26
Search for generic jets and MET: results
• Simplified model exclusion plots
CMS PAS SUS-13-012
1. Compute excluded cross section for each model in param space
2. Compare to reference cross section to see if model excluded
• Assume 100% branching fraction for stated process!
mLSP (GeV)
CMS Preliminary, 19.5 fb-1, s = 8 TeV
0
1
R
1400
pp ® ~q ~q, ~
q ® q ~c NLO+NLL exclusion
Observed ± 1 stheory
Expected ± 1 sexperiment
L
~ ~
~
q +~
q (~u,d,~
s,c)
Squark pair
production
(L + R)
600
500
400
300
one light ~q
m~q (GeV)
1200
10
1
10-1
10-2
10-3
10
1
10-1
-3
10-2
10
95% C.L. upper limit on crossAs
section
(pb) weak
before,
200
0
1
1000
Observed limit
w/theoretical
Th’y cross section fall-off
uncertainty
100
800
m~g (GeV)
CMS Preliminary, 19.5 fb-1, s = 8 TeV
pp ® ~g ~g, ~g ® q q ~c NLO+NLL exclusion
Observed ± 1 stheory
Expected ± 1 sexperiment
600
Gluino pair production
Expected limit
w/experimental
uncertainty
300 400 500 600 700 800 900 1000
1200
800
600
400
200
400
1000
mLSP (GeV)
upper limit
on cross
section (pb)27
limit 95%
with C.L.
reduced
number
of squarks!
SUSY search for b-jets, 1 lepton, and MET
https://twiki.cern.ch/twiki/bin/view/CMSPublic/PhysicsResultsSUS13007
CMS PAS SUS-13-007
Lepton spectrum method: for leptons produced in W decay,
the lepton spectrum can be used to measure the 𝜈 spectrum.
Events/50 GeV
CMS Preliminary
s = 8 TeV
-1
19.4 fb
Makes use of W helicity fractions in tbW
Data
Sum predicted
2
Single Tau
10
Dilepton
(m~ ,m
g
)=(1100 GeV,100 GeV)
LSP
HT>500 GeV
Njet ³ 6, Nb³ 2
10
1
...and W helicity fractions in W+jets
Normalized
residuals
10-1
3
2
1
0
-1
-2
-3
200 300 400 500 600 700 800 900 1000
ET [GeV]
200 300 400
500 600
700 800 900 1000
ET [GeV]
41
MSSM parameter count
Sector of MSSM
Number of
parameters
Standard Model parameters
18
1 Higgs parameter, analogous to Higgs mass in SM
1
Gaugino/higgsino sector
5
Gaugino/higgsino sector – CP violating phases
3
Squark and slepton masses
21
Mixing angles to define squark and slepton mass
eigenstates
36
CP violating phases
40
Total
124
In cMSSM, SUSY is described by just 4 continuous real params + 1 sign.
Formulating SUSY – a short history
Gluino pair production and decay to stop
t
g
g
t
g
g
t
t
c
0
1
c
0
1
unobserved!
t
t
Signature: 4 top quarks + pTmiss
LSPs are unobserved: cannot reconstruct mass peaks!
Results for gluino pair production with g ® tt c
900
800
700
600
500
400
300
(g
lu
in
-m
(L
SP
)=
2
m
(to
p)
SUS-13-012 0-lep
-1
19.5 fb
SUS-13-008 3-lep (3l+b) 19.5 fb
-1
-1
-1
-1
SUS-13-013 2-lep (SS+b) 19.5 fb
SUS-13-016 2-lep (OS+b) 19.7 fb
SUS-13-007 1-lep (n jets³ 6) 19.3 fb
SUS-14-011 0+1+2-lep (razor) 19.3 fb
( ET+HT)
~~
~
~0
g
g
production,
g
®
t
t
c
1
o)
800
1000
Observed
SUSY
Observed -1 stheory
Expected
CMS Preliminary
s = 8 TeV
ICHEP 2014
m
600
-1
1200
1400
1600
gluino mass [GeV]
M (g)
Mass of pair-produced particle
Decreasing cross
section
Maximum
gluino mass
limit at minimum
LSP mass.
200
Increasing mass
splitting
increasing MET
Mass of
LSP
100
0
M (c )
0
1
0
1
LSP mass [GeV]
The famous neutralino dilepton cascade
Opposite-sign, same flavor leptons
p
c
0
2
c10
c10
n
p
n
The c 2 can be produced in any process, not just direct EW
production.
0
m + - (max) = (mc 0 - m )(m - mc 0 ) / m
2
2
2
2
2
1
0
Search for SUSY
in opposite sign dileptons
c2
m + - > 20 GeV, N(jets) ³ 2 (pT > 40 GeV)
19.4 fb-1 (8 TeV)
events / 10 GeV
104
SIGNAL SAMPLE
Opposite sign,
same flavor leptons
(ee, μμ)
103
Data
DY(ee,mm)
DY(tt)
tt
Single-top
WW,WZ,ZZ
Other SM
105
104
102
50
100
150
200
Emiss
T
250
[GeV]
300
350
400
Data/MC
OF Leptons
50 100 150miss200 250 300 350 400
ET [GeV]
1
0
0
10
SF Leptons
1
20
Dominated by
ttbar production
102
10
Data
DY(ee,mm)
DY(tt)
tt
Single-top
WW,WZ,ZZ
Other SM
CONTROL SAMPLE
Opposite sign, opp.
flavor leptons (eμ)
103
Data/MC
105
19.4 fb-1 (8 TeV)
CMS Preliminary
events / 10 GeV
CMS Preliminary
RSF/OF = 1.00 ± 0.04 (central)
1 RSF/OF = 1.11 ± 0.07 (forward)
50 100 150miss200 250 300 350 400
20
ET [GeV]
1
0
0
50
100
150
200
Emiss
T
250
[GeV]
300
350
400
0
Search for SUSY
in opposite sign dileptons
c2
Fit opposite sign dilepton mass distribution to shapes from (1) Flavor Symmetric
(FS) background, Drell Yan, and signal.
SIGNAL
SAMPLE
(e+e-, μ+μevents)
180
160
140
120
Data
Fit
FS
DY
Signal
100
60
40
20
100
150
200
250
50
180
Data
160
Fit
140
CONTROL SAMPLE
(eμ events)
120
100
80
60
40
20
300
50
3
Signalmcontribution
(2.4σ local signifiance).
ll [GeV]
Pull
Pull
50
100
150
mll [GeV]
200
250
300
19.4 fb-1 (8 TeV)
OF Central Leptons
80
3
2
1
0
-1
-2
-3
CMS Preliminary
200
SF Central Leptons
Events / 5 GeV
19.4 fb-1 (8 TeV)
Events / 5 GeV
CMS Preliminary
200
2
1
0
-1
-2
-3
50
100
150
200
250
300
200
250
300
mll [GeV]
100
150
mll [GeV]
0
Search for cSUSY
in opposite-sign dileptons
2
19.4 fb-1 (8 TeV)
180
Data
160
Background
140
DY (data-driven)
120
Sys. Å Stat.
40
20
Data/Bgnd
50
100
2
150
200
250
mll [GeV]
1
50
100
150
200
250
mll [GeV]
2.6σ excess (not significant)
DY (data-driven)
events / 5 GeV
80
Sys. Å Stat.
40
20
300
0
300
≥1 forward lepton
1.6<|ηlep|<2.4
60
Forward Leptons
60
0
Background
Data/Bgnd
80
19.4 fb-1 (8 TeV)
Data
100
Central leptons
|ηlep|<1.4
100
0
CMS Preliminary
Central Leptons
events / 5 GeV
200
CMS Preliminary
50
100
2
150
200
250
300
200
250
300
mll [GeV]
1
0
50
100
0.3σ excess
150
mll [GeV]
Events / 5 GeV
0
Search for SUSY
in opposite sign dileptons
c2
19.4 fb-1 (8 TeV)
CMSPreliminary
Data
+
DY+jets (e e-,m +m -)
DY+jets (tt)
WW,WZ,ZZ
250
Madgraph tt
single-top
200
Other SM
Scaling Uncert.
JEC & Pileup Uncert.
m~b = 225 GeV m ~2 = 150 GeV
c
0
m~b = 350 GeV m ~2 = 275 GeV
c
m~b = 400 GeV m ~02 = 150 GeV
150
c
100
0
Central Signal Region
ee+m m
50
50
100
150
200
250
300
mll [GeV]
Data / MC
2
1.5
1
0.5
0
Data / Background Only
Data / Background + Signal
Data / Background + Signal
Data / Background + Signal
“Direct” pair production of light stops
Example: direct stop production with decay to neutralinos or
charginos.
t
g
g
g
t
t
b
c
c
t
0
1
0
1
b
Sensitivity of the searches will depend strongly on the neutralino mass.
The channel with
has sensitivity to lower stop mass.
Search for direct stop production:
SUS-13-011
pp ® t1t1
https://twiki.cern.ch/twiki/bin/view/CMSPublic/PhysicsResultsSUS13011 http://arxiv.org/abs/1308.1586
4-body
m(t1 ) < m( c10 )
t ® bW * c10
violates assumed
spectrum
(off-shell W)
3-body
2-body
m(t ) ³ m(t) + m( c10 )
t ® bc1+ ® bW +(*) c10
has extra mass parameter
+m(b)
Search for direct stop production:
SUS-13-011
pp ® t1t1
https://twiki.cern.ch/twiki/bin/view/CMSPublic/PhysicsResultsSUS13011 http://arxiv.org/abs/1308.1586
t ® bW * c10
m(t1 ) < m( c10 )
violates
assumed
Same
final
spectrum
(off-shell
W)
state
but different
kinematics!
m(t ) ³ m(t) + m( c10 )
t ® bc1+ ® bW (*) c10
+m(b)
32
Search for direct stop production:
SUS-13-011
pp ® t1t1
https://twiki.cern.ch/twiki/bin/view/CMSPublic/PhysicsResultsSUS13011 http://arxiv.org/abs/1308.1586
MT > 120 GeV
MET > 100 GeV
t ® t c10 (´100)
t ® bW * c10
(off-shell W)
m(t ) ³ m(t) + m( c10 )
Use BDTs
for separate
modes &
regions.
MT > 120 GeV
MET > 100 GeV
t ® bc1+ (´100)
+m(b)
33
Results from searches for top squarks
LSP mass [GeV]
700
600
500
400
300
200
100
0
Expected
Observed
~~
~
0
0
t-t production, t® t ~c1 / c ~c1
200
-m
~
mt
300
0
~c
=
1
0
~c
mW
-m
~
mt
1
=
mt
400
500
SUS-13-015 (hadronic stop) 19.4 fb -1
1
SUS-14-011 0-lep (Razor) + 1-lep (MVA) 19.3 fb -1
0
~
SUS-13-009 (monojet stop) 19.7 fb -1 ( t® c ~c )
SUS-14-011 0-lep + 1-lep + 2-lep (Razor) 19.3 fb -1
SUS-13-011 1-lep (MVA) 19.5 fb -1
CMS Preliminary
s = 8 TeV
ICHEP 2014
100
600
700
800
stop mass [GeV]
Search for dark matter at the LHC
c10
c10
q
h, H
c10
q
q
c10
q
q
q
c10
q
q
c10
c10
Z
c10
q
• Above: direct dark matter detection
processes: doesn’t have to be SUSY!
• Use crossing to get
• How to see
?
q
Initial-state
radiation
mediator on-shell
or off shell
Signature for dark matter at the LHC
c
Signature: Jet or photon from
initial-state radiation (ISR) + large
missing transverse momentum
0
1
p
miss
T
c
n
Jet
Dominant background:
Z(νν) + jets
Z
p
miss
T
0
1
n
...don’t want to rely on MC modeling of initial-state radiation
There are lots of monojet events in the data!
Monojet
pTmiss = 914 GeV
We can predict the contribution from Z + 1 jet
Monojet
Z®m m
+
å
i=e,m ,t
B(Z ® n in i )
B(Z ® m m )
+
-
»6
-
Searching for dark matter: monojet search results
• MET > 250 GeV
• 1 central jet, pT>110
GeV.
• 2nd softer jet allowed,
not back-to-back.
• Remarkable that QCD
is so well controlled.
• Veto events with leptons
• Zνν is dominant
background; predict
from Zμμ control
sample.
Searching for dark matter: monojet search results
c-Nucleon Cross Section [cm2]
c-Nucleon Cross Section [cm2]
10-36
10-37
10-38
10-39
10-40
10-41
10-42
10-43
10-44
10-45
10-46
10-36
10-37
10-38
10-39
10-40
10-41
10-42
10-43
1
1
CMS, 90% CL, 7 TeV, 5.0 fb-1
SIMP
+
3
10
W
-
CMS
-
W
12
+
12
0
LE 2
P P 20
5
be W
r-K W
IceCu
S u pe
C OU
L2
102
m 5
(cg g c)(qg mg q)
CMS, 90% CL, 8 TeV, 19.7 fb-1
Axial-vector operator
Spin Dependent
10
2011
2 0 12
m
CMS
(cg c)(qg m q)
12
L2
Vector
P P 20
S II
superCDMS
N100
20 1 3
10
3
Mc [GeV]
CMS, 90% CL, 7 TeV, 5.0 fb-1
eN T
CMS, 90% CL, 8 TeV, 19.7 fb-1
CoG
SIMPLE
C OU
C DM
102
LU X
X EN O
CMS, 90% CL, 8 TeV, 19.7 fb-1
10
Spin Independent
Scalar
a
cca (G )2
s mn
3
4L
CDMSlite
10-44
10-45
10-46
Mc [GeV]
Relative parton luminosities: 13 TeV vs. 8 TeV
1350 GeV gluinos: x30
1000 GeV gluinos: x20
750 GeV squarks: x9
350 GeV X+-X0: x3
top pairs: x4
gg ® X
qg ® X
å qi qi ® X
Sensitivity projections for selected searches
0
...is cthere
a message here?
2