Searches for exotic physics with unconventional signatures at ATLAS and CMS - Cristiano Sebastiani
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Searches for exotic physics
with unconventional
signatures at ATLAS and CMS
Cristiano Sebastiani
1
Exotic among exotics?
Search for new BSM physics at LHC with exotic signatures:
• Standard decays
ATLAS and CMS are not
• Unconventional signatures: long time-of-flight, anomalous energy optimised for this kind of signals
deposits, displaced secondary vertices…
• Detector-stable particles
Unusual and unique signatures are extremely challenging to probe:
NON-COLLISION
TRIGGER RECONSTRUCTION
BACKGROUND
Anomalous signatures Object identification and Unconventional
not associated with reconstruction signatures have
standard activity in the algorithms are to be unconventional
detector require the updated to include non- backgrounds, from
development of standard tracks and detector noise to non-
dedicated triggers! energy deposits collision physics events
!2
Unconventional signatures
New particles can be long-lived:
observed lifetime is governed by an
exponential defined by the proper lifetime cτ
decay probability
disappearing tracks cτ = 5 cm, βγ ∼ 40
calorimeter 60%
multitrack
out of the detector 1%
late photons
muon chambers 13%
vertices in ID
displaced leptons, emerging jets
tracker 25%
lepton-jets
prompt 1%
trackless,
displaced jets,
delayed jets
(meta) stable charged
particles distance
multitrack vertices in MS
analysis
featured
in backup !3
CMS-PAS-EXO-19-021
Displaced jet
Inclusive search for long-lived particles decaying into jets (covering lifetime between ~1 mm to ~1 m)
•Dedicated di-jet trigger to lower HT
threshold from ~1 TeV to 0.4 TeV
•Select 1 displaced secondary
vertex (SV) within tracker and 2 Secondary Vertex
displaced jets
•MVA selection to discriminate signal CMS high-level trigger
from huge QCD background:
select displaced tracks
online and tag displaced-
•Minor backgrounds from nuclear jets with the HLT system by
interactions, heavy flavour decays, counting the number of
randomly crossing tracks prompt/displaced tracks
associated with the jets
HT: scalar sum of jet pT for jets with pT > 40 GeV and |η| < 2.5 !4
CMS-PAS-EXO-19-021
Displaced jet
Data-driven background estimate based on 3 dimensional
ABCD method: -1
95.9 fb (13 TeV)
selection 1: # prompt tracks of first jet
CMS-PAS-EXO-19-021
Displaced jet
Exclusion limits at 95% CL: for a Higgs portal model with two long-lived scalars, each decaying to a quark-antiquark
pair, Higgs branching fractions larger than 1% can be excluded for mean proper decay lengths between 1mm and 1m
-1 -1
132 fb (13 TeV) 132 fb (13 TeV)
107 107
σ/ σSM(gg→H)
σ/ σSM(gg→H)
95% CL upper limits 95% CL upper limits
10 6 CMS 10 6 CMS
Preliminary Median expected Preliminary Median expected
5 5
10 10
Observed q
10 4
gg → H, mH = 125 GeV 104 gg → H, mH = 125 GeV Observed
mS = 15 GeV
10 3 H → SS, S → dd 103 H → SS, S → bb mS = 40 GeV p S q
mS = 40 GeV
H
102 102 mS = 55 GeV
mS = 55 GeV
10 10 p S q
1 1
q
10−1 10−1
10−2 10−2
10−3 10−3
10−4 10−4
1 10 102 103 1 10 102 103
cτ0 [mm] cτ0 [mm]
!6
ATLAS-CONF-2021-005
Displaced Vh 4b
Dedicated searches:
Search for exotic decays of the Higgs boson to pairs of long- in calorimeter: 0.1 ≤ cτ ≤ 10 m
lived neutral particles, each decaying to a bottom quark pair: in muon system: cτ ≤ 100 m
EPJC 79 (2019) 481
•Trigger on ZH production: target lower masses and be
sensitive in this unprobed phase space
•Dedicated LLP reconstruction in tracker region with custom
Large Radius Tracking algorithm
2 2 2
κH S + μH S
Higgs portal and mixing Prompt search
(no displaced techniques)
with ‘dark’ scalars 50 μm ≤ cτ ≤ 2 mm
JHEP 10 (2018) 031
For scalar masses above 10 GeV —> bb decay mode favoured
Uncovered spot:
LLP decays inside inner detector
!7
ATLAS-CONF-2021-005
Displaced Vh 4b
Zero events observed in signal region and consistent with background estimate
95% CL upper limit on Β (H → aa)
1
−1
10
This analysis
ATLAS Preliminary complements ATLAS
−2
10 s = 13 TeV prompt and dedicated
m H = 125 GeV
searches
ma = 15 GeV
−3
10 ma = 35 GeV
CR+(MS1+MS2)
ma = 55 GeV
ma = 5 GeV
Prompt
−4 Prompt
10 ma = 8 GeV
ma = 20 GeV ma = 15 GeV
ma = 30 GeV ma = 25 GeV
−5 ma = 60 GeV ma = 40 GeV
10
−4 −3 −2 −1 2 3
10 10 10 10 1 10 10 10
Mean proper lifetime cτa [m]
Most stringent 95% CL upper limits on BR(H->ss) for masses below 40 GeV
!8
CMS-PAS-EXO-19-013
Displaced vertices
Search for long-lived particles decaying into jets (covering lifetime between ~0.1 mm to ~100 mm)
•Jet trigger with HT 1050 GeV
beam
•Select 2 displaced vertices events with spot
more than 4 jets
•Customised track and vertex
reconstruction for LLPs decaying inside
the beam pipe (no association with jets)
•Target pairs of LLP: distance between the Displaced vertex
within beam pipe
two DV as main discriminating variable
•Target tiny lifetimes: displacement of the
vertex within the beam pipe ( 40 GeV and |η| < 2.5
!9
CMS-PAS-EXO-19-013
Displaced vertices 101 fb-1 (13 TeV)
Events/0.1 mm
104 CMS Data
Background prediction purely data-driven: build background template
Preliminary Multijet signals,
3 m = 800 GeV, σ = 0.3 fb:
10
for dVV distribution in 2DVs events from dBV distribution in 1DV events cτ = 0.3 mm
cτ = 1.0 mm
cτ = 10 mm
102
y 10
1
Signal region: 2DVs with 10−1
dVV - transverse distance
track multiplicity >=5 0 0.5 1 1.5 2 2.5 3 3.5 4
between the two DVs dBV (mm)
101 fb-1 (13 TeV)
dVV V 0.3
Events/0.1 mm
ΔϕVV dBV CMS Preliminary
0.25 ≥5-track x ≥5-track
B x Data
0.2 Background template
dBV
V 0.15
dBV - transverse distance 0.1
between the DV and the
beam spot 0.05
0
0 0.5 1 1.5 2 2.5 3 3.5 4
!10 dVV (mm)CMS-PAS-EXO-19-013
Displaced vertices
The search is sensitive to various models with pair produced LLP decays into two or more jets in
the final state. Zero events observes, results are interpreted in RPV SUSY models.
p t̄
-1
6
Fall 2020 132-140 fb (13 TeV)
10 b̄
95% CL upper limit on σ (fb)
gluino is long-lived 0
˜ , g̃
s̄
5
CMS Preliminary t̃
10
pp→~
g~g, ~
0
g →qq∼
χ Delayed jets (1906.06441)
0
t̃ s̄
1 ˜ , g̃
q Displaced jets (EXO-19-021)
4 m~g = 2400 GeV b̄
10
q m∼0 = 100 GeV Displaced vertices (EXO-19-013) p t̄
p !
g χ
1
10
3 ~~
!01
χ Approx. NNLO+NNLL σ(pp → g g)
more interpretations
!01
shown in backup
χ
p 102
!
g q
q
10 p d¯
1 t̃ d¯
10−1
t̃⇤ d
large jet activity
10−2 −1 2 3 4 5 6 p d
10 1 10 10 10 10 10 10
cτ0 (mm)
Exclusion limits complementary to the displaced-jets and delayed-jet (j.physletb.2019.134876) searches
!11ATLAS-CONF-2021-015
Disappearing tracks
Search for long-lived particle decaying within the inner tracker
• MET trigger to select events with
jets and ‘disappearing
tracks’ ( due to suppressed
interaction or low-pT)
• ‘Disappearing track’: 4-hit pixel
tracks with no hits in the silicon Targets very compressed SUSY scenarios (…and various DM models)
traker (SCT) and < 5 GeV of
energy deposits in calo dR < 0.2 undetectable
]
± ± 0
• Rare SM backgrounds from χ̃ → π χ
]
]
charged lepton scattering and
leaves hits MET
combinatorial fakes Electroweak (EWK)
production Strong production
!12ATLAS-CONF-2021-015 Disappearing tracks
Exclusion limits at 95% CL for the EWK channel fo the AMSB model, neutralino mass vs lifetime
Exclusion for 660 GeV wino, best limit for Exclusion for 210 GeV Higgsino, best limit for
wino-like neutralino (460 GeV in early Run2) higgsino-like neutralino (155 GeV in early Run2)
!13CERN-EP-2021-041
Stopped particles
Search for late decays to hadronic jets from long-lived particles
• dedicated jet+MET trigger, recorded in empty bunches to
reduce background SUSY gluino R-hadrons as
benchmark (gluino as LLP)
• Non collision background that can produce energetic jets
in out-of-time bunch crossings: cosmic-ray muons, beam
induced background (BIB), cavern background and
calorimeter noise if LLP long-lived enough, some will stop
inside the detector before decaying
p
X
Empty Paired Empty
p
calo
LLP produced here
!14CERN-EP-2021-041
Stopped particles
Search for late decays to hadronic jets from long-lived particles
• dedicated jet+MET trigger, recorded in empty bunches to
reduce background SUSY gluino R-hadrons as
benchmark (gluino as LLP)
• Non collision background that can produce energetic jets
in out-of-time bunch crossings: cosmic-ray muons, beam
induced background (BIB), cavern background and
calorimeter noise if LLP long-lived enough, some will stop
inside the detector before decaying
p p
decay time
Paired X Empty Paired
p p
calo
LLP produced here
LLP decays here
!15CERN-EP-2021-041
Stopped particles
Fully data-driven background estimate with inputs from the background enriched control regions for cosmic-ray
muons and BIB background. No excess found and 95%CL limits are presented.
signal region, 2018 data |η|Conclusions
• New searches for long-lived particles in events with jets at ATLAS and CMS using full Run-2
dataset have been presented, mainly focusing on decays inside the inner tracking system and
calorimeters
• Several New Physics models imply the presence of unconventional signatures at colliders, which
can be very challenging and require new ideas and techniques
• Experiments increasingly focus on unconventional signals and long-lived particles with ambitious
research programs… but there is still much to be done
• Great effort in developing new tools and strategies to improve identification and reconstruction of
long-lived particles pushing the detectors beyond their limits
• Run-3 and HL-LHC programmes offer a unique opportunity to plan, innovate and create new
unconventional searches yet to be explored
!17Prospects CERN-ESU-004
CERN-LPCC-2018-05
2 2 2
κH S + μH S
Higgs portal and mixing with ‘dark’ scalars
!18Backup !19
ATLAS-CONF-2021-015 Disappearing tracks
Exclusion limits at 95% CL for the strong channel fo the AMSB model, neutralino mass vs lifetime
!20CERN-EP-2020-205 Displaced leptons
Search for two leptons with high impact parameter
~~ ~ ~
l l → l G l G, l ∈ [e, µ, τ]
5
10
Lifetime [ns]
Expected limits
104 ATLAS
-1 Observed limits
s=13 TeV, 139 fb
co-NLSPL,R
10 3 All limits at 95% CL
~
eL,R
2 ∼
µ
10 L,R
∼τ
101
1,2
Exclusion up to 830 GeV for 0.1 ns, up to 200
GeV for 10 ns lifetime (previous limits at 90 GeV)
100
10−1
10−2
−3
10
100 200 300 400 500 600 700 800 900
~
m (l) [GeV]
!21j.physletb.2019.134876
Delayed jet + MET
Search for delayed decays to hadronic jets of long-lived particles and missing transverse momentum
• exploits ECAL energy
deposits timestamp to
define tjet
• Main QCD background
highly suppressed by out-
of-time (within same bunch
crossing) jet cuts
• ~0 expected background in
SR; high signal efficiency of
out-of-time decays events
!22j.physletb.2019.134876
Delayed jet + MET
!23ATL-PHYS-PUB-2020-007
Displaced jet recast
Reinterpretation of the search low electromagnetic fraction jet analysis preserved with RECAST to exclude
new models not included in the paper: Stealth SUSY, Baryogenesis, dark photon
The analysis allows to extend displaced jet exclusion from dark sector scenarios to lower lifetimes
!24ATLAS-CONF-2021-005
Displaced Vh 4b
Event selection and LLP reconstruction:
•Select displaced jets events with small charged fraction
and few tracks associated to the primary vertex ATL-PHYS-PUB-2017-014
• Reconstruct large d0 tracks with Large Radius Tracking
(LRT) algorithm
•Build displaced vertices (DV) with LRT tracks
•Match DV with displaced jet
Control region < 2 DV:
Signal region >= 2 DV:
Derive per-jet efficiency
Predict background from DV
map pT (jet) vs b-tag score
probability efficiency map
in 0 DV and 1 DV events
!25ATLAS-CONF-2021-015
Disappearing tracks
Background tracklet pT-shape estimated via control regions:
•Fakes shape directly from CR
•e/mu/had shape by applying transfer factors and smearing
functions to CR events
Final background estimate:
3.10 ± 0.74 (4 observed)
(strong production results in backup)
!26You can also read
