searches for resonant signatures of production at the lhc

Jose Benitez (ATLAS, University of Iowa, U.S.) on behalf of the ATLAS and CMS collaborations

13th Recontres du Vietnam Exploring the Dark Universe

Quy Nhon, Vietnam July 23-29, 2017

Searches for resonant signatures of Dark Matter production at the LHC

J. Benitez (U.Iowa) DM Resonance searches at the LHC

introduction

• This talk focuses on the searches for the mediator in decays to quarks or leptons. • Interpretations use generic models and benchmark scenarios [arXiv:1507.00966, arXiv:1703.05703]:

• dijet searches: gq = 0.25, gl= 0, gDM = 1 • dijet + dilepton searches:

• gq = 0.1, gl=0.1, and gDM = 1 (Axial-Vector) • gq = 0.1, gl=0.01, and gDM = 1 (Vector )

• Choices lead to small widths (Γ/M < 10%) needed for interpretations of the mass distributions.

2

missing energy search mediator searchq

q’

gq

Possible searches at the LHC:

see talk by M. Queitsch

https://arxiv.org/abs/1507.00966


Dijet searches

3

Displays of dijet events recorded with real data by CMS and ATLAS


high mass search strategy

• CMS Search:

• Particle flow jets (anti-kt R=0.4), pT>30 GeV.

• Trigger: sum pT of jets>900 GeV

• ATLAS search:

• Calorimeter jets (anti-kt R=0.4), pT> 20 GeV.

• Trigger: one high pT jet > 380 GeV. Offline: pTlead> 440 GeV, pTsublead>60 GeV

• Multijet background reduction with: |Δηjj| < ~1.3

• Challenges calibrating jets with TeV momentum

• Detection resolution ~5% at high mass (qq’)

4

p p

j1

j2

η/y


Dijet mass distributions

• Reconstructed masses up to ~9 TeV ! • Background fit using empirical power law models • Model independent excess searches: multi-bin counting experiment in

~100 GeV steps (CMS) and sliding windows (ATLAS)

5

arxiv:1703.09127 CMS: EXO-16-056


http://cms-results.web.cern.ch/cms-results/public-results/preliminary-results/EXO-16-056


crossection limits

• Limits at 95 CL are computed on the σ x BR x Acceptance. • Values range between ~10

-3 pb @ 6 TeV and ~1pb @ 1 TeV

• ATLAS results computed for a “generic” model with Gaussian shape at particle level for easier model-independent interpretations.

6

arxiv:1703.09127 CMS: EXO-16-056




Dijet high mass limits

• No significant excess is observed.

• Limits on the σ x A x BR available in the references.

• Limits on resonances coupling *only* to quarks (assuming ~zero decay to DM) placed up to 3.5 TeV.

• Above 3.5 TeV and above gq~0.5 the natural width of the resonance is larger than ~10%, narrow width model not valid.

7

arxiv:1703.09127 CMS: EXO-16-056

uncertainty bands not shown.




Going lower in mass: Trigger level analysis• Dijet events with low mass cannot be recorded due to high rates from SM processes.

• Both ATLAS and CMS implemented a special readout at trigger level recording a compact data format for these events.

• Challenging jet calibration given less detector information. Also, difficult to model the background given the huge amount of events.

8

ATLAS-CONF-2016-030 CMS: EXO-16-056

https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/CONFNOTES/ATLAS-CONF-2016-030



Trigger level analysis limits

• Limits placed on the quark-coupling for masses from • 600 GeV up to 1600 GeV by CMS • 450 GeV up to 950 GeV by ATLAS , limits extend to low mass using tigher |y*| <0.3

category. • No significant excess is found

9






Low mass search: boosted dijets• Search for very low mass resonances is difficult to

trigger due to large SM background. Boosted topologies from events with initial state radiation (jet or photon) reduce the rate of events.

• Resonance reconstruction using a single large radius jet with substructure and large transverse momentum by CMS or two R=0.4 calorimeter jets by ATLAS.

10

q

q

gq

ISR

boosted resonance

ATLAS-CONF-2016-070/ CMS: EXO-17-001

Difference in mass ranges due to difference in resonance reconstruction (merged vs resolved jets).

https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/CONFNOTES/ATLAS-CONF-2016-070/



Boosted dijet search limits

• CMS uses ISR jets and search extends from 50-300 GeV. First time limits in the range 50-100 GeV

• ATLAS uses both ISR photons and jets, and complements up to ~900 GeV.

• CMS: small excess at 115 GeV (2.9σ local, 2.2σ global)

11

ISR photon

ISR jet

ISR jet



https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/CONFNOTES/ATLAS-CONF-2016-070/



wide resonance search• At large coupling values (gq~0.5) the

resonance becomes very wide and the narrow-width approximation is not valid.

• Angular distribution (χdijet) can still reveal excess of events over the SM background.

12

• Couplings excluded above gq~0.7 for masses up to ~5 TeV for case of mDM~0.

CMS:EXO-16-046 ATLAS: arxiv:1703.09127




bb resonances • Searches performed using b-tagged dijets

test possible flavor dependent coupling • b-tagging of high pT jets is challenging for

high mass resonances • Complementary searches by CMS and

ATLAS (low vs high mass) • No excess found in searches up to ~5 TeV.

13

ATLAS-CONF-2016-060

CMS: HIG-16-025


http://cms-results.web.cern.ch/cms-results/public-results/preliminary-results/HIG-16-025

Dilepton searches

14

Dilepton event displays

15

Left: display of a data event where a dimuon event is reconstructed by CMS with an invariant mas of 1.7 TeV

Right: display of a dielectron data event reconstructed by ATLAS with an invariant mass of 1.5 TeV

mass resolution

16

m(ee) (GeV)500 1000 1500 2000 2500 3000 3500

(%)

2 extr

aσ+

2 fitσ

1

1.1

1.2

1.3

1.4

1.5 ee→DY ee→ ψZ'

ee→ SSMZ'fit to DY

-1CMS unpublished, 8 TeV, 19.7 fbEB-EB

2 + Cm2S + 2m

2N

0.004±C = 1.019 1.086±S = 0.000

1.857±N = 122.657

Dimuon invariant mass (GeV)1000 1500 2000 2500 3000

Dim

uon

rela

tive

mas

s re

solu

tion

0.04

0.06

0.08

0.1

0.12

0.14

0.16

/ ndf 2χ 4.375 / 7p0 0.001836± 0.001551 p1 2.421e-06± 5.386e-05 p2 7.016e-10± -5.455e-09

/ ndf 2χ 4.375 / 7p0 0.001836± 0.001551 p1 2.421e-06± 5.386e-05 p2 7.016e-10± -5.455e-09

/ ndf 2χ 8.055 / 7

p0 0.001293± 0.01675

p1 1.749e-06± 2.575e-05

p2 5.139e-10± -2.862e-10

/ ndf 2χ 8.055 / 7

p0 0.001293± 0.01675

p1 1.749e-06± 2.575e-05

p2 5.139e-10± -2.862e-10

Tracker-only

Tracker + muon system

= 8 TeVsCMS simulation,

• Resolution on the invariant mass of dielectron (left) and dimuons (right) reconstructed by CMS up to ~3.5 TeV.

• The resolution for electrons is worse at low mass due to low energy electrons, but remains ~1% at high mass.

• For dimuons, the resolution degrades up to ~10% at high mass due to the worse muon momentum resolution.

• Similar performance with ATLAS

CMS- EXO12061


μ+μ- final state

17

• Analysis selections: • muon pT>53 GeV (CMS) , pT>30 GeV (ATLAS) • isolation requirement, • opposite charge.

Background mainly from SM Z—>μ+μ-.

ATLAS-CONF-2017-027CMS: EXO-16-031




e+e- final state

• Analysis selections:

• electron transverse energy >35 GeV (CMS), >30 GeV (ATLAS)

• sophisticated electron identification algorithms

• isolation w.r.t. tracks and calorimeter deposits

18


Background mainly from SM Z—>e+e-.




crossection limits

• Very clean search, but no excess observed above the SM background

• Limits combine both electron and muon channels

• Cross section x BR limits between 10-4 and 10-3 pb for TeV resonances

• Limits similar for both experiments,

19




Summary plots for mediator exclusions

20

Computed for benchmark scenarios:

• dijet searches: gq = 0.25, gl= 0, gDM = 1

• dijet + dilepton searches:

• gq = 0.1, gl=0.1, and gDM = 1 (Axial-Vector)

• gq = 0.1, gl=0.01, and gDM = 1 (Vector )


AV mediator exclusions (dijet)

• Limits on the mDM vs mediator mass parameter plane with no coupling to leptons.

• Mediator mass exclusion slightly improves for high mDM as the mediator has larger decay fraction to quarks.

• Nice complementarity to the exclusions from direct searches using missing energy signatures.

21

[GeV] medMediator mass M0 500 1000 1500 2000 2500 3000 3500 4000 4500

[GeV

] D

MD

ark

mat

ter m

ass

m

0

200

400

600

800

1000

1200

1400

1600

1800

2000LHCP 2017 PreliminaryCMS

Axial-vector mediatorDirac DM

= 1.0DM

g = 0.25

qg

= 0l

g Exclusion at 95% CL

Observed

Expected

[EXO-16-056])-1 (35.9 fbDijet

[EXO-17-001])-1 (35.9 fbBoosted dijet

[EXO-16-048])-1 (35.9 fbDM + j/V(qq)

[EXO-16-039])-1 (12.9 fbγDM +

[EXO-16-052])-1 (35.9 fbDM + Z(ll)

DM = 2 x mMedM

0.12≥ 2 hcΩ

LHCP 2017 PreliminaryCMS


= 1.0DM

g = 0.25

qg

= 0l

g

[https://twiki.cern.ch/twiki/pub/CMSPublic/PhysicsResultsEXO/DM_summary_plots_LHCP_2017.pdf] [https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/CombinedSummaryPlots/EXOTICS/index.html]

https://twiki.cern.ch/twiki/pub/CMSPublic/PhysicsResultsEXO/DM_summary_plots_LHCP_2017.pdf

https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/CombinedSummaryPlots/EXOTICS/index.html


AV mediator exclusions (dijet+dilepton)

• The exclusions are computed for a dark matter coupling gDM = 1.0, a quark coupling gq = 0.1 universal to all flavors, and lepton coupling gl = 0.1 [arXiv:1703.05703].

• For these couplings the mass exclusions from dileptons are higher than from the dijets

• The vertical dijet bands correspond to different dijet analyzes, and fluctuations in the crossection limits.

22


[GeV

] D

MD

ark

mat

ter m

ass

m

0

200

400

600

800

1000

1200

1400

1600

1800



= 1.0DM

g = 0.1

qg

= 0.1l

g

Exclusion at 95% CL

Observed

Expected

[EXO-16-031])-1 / 13.0 fb-1 (12.4 fbDilepton

[EXO-16-056])-1 (35.9 fbDijet


DM = 2 x mMedM

0.12≥ 2 hcΩ



= 1.0DM

g = 0.1

qg

= 0.1l

g





Vector mediator exclusions (dijet)

• Similar exclusion for Vector resonances as for Axial-Vectors

23


[GeV

] D

MD

ark

mat

ter m

ass

m

0

200

400

600

800

1000

1200

1400

1600

1800


Vector mediatorDirac DM

= 1.0DM

g = 0.25

qg

= 0l

g Exclusion at 95% CL

Observed

Expected

[EXO-16-056])-1 (35.9 fbDijet


[EXO-16-048])-1 (35.9 fbDM + j/V(qq)

[EXO-16-039])-1 (12.9 fbγDM +

[EXO-16-052])-1 (35.9 fbDM + Z(ll)

DM = 2 x mMedM

0.12≥ 2 hcΩ



= 1.0DM

g = 0.25

qg

= 0l

g





Vector mediator exclusions (dijet+dilepton)

• The exclusions are computed for a dark matter coupling gDM = 1.0, a quark coupling gq = 0.1 universal to all flavors, and lepton coupling gl = 0.01[arXiv:1703.05703].

• With this choice of couplings, the Z’ decays to leptons are reduced with respect to the decays to quarks.

• Exclusions start mainly above 2 x mDM as the decays of the mediator to DM are then suppressed.

• The vertical dijet bands correspond to different dijet analyzes, and fluctuations in the crossection limits.

24


[GeV

] D

MD

ark

mat

ter m

ass

m

0

200

400

600

800

1000

1200

1400

1600

1800



= 1.0DM

g = 0.1

qg

= 0.01l

g

Exclusion at 95% CL

Observed

Expected

[EXO-16-031])-1 / 13.0 fb-1 (12.4 fbDilepton

[EXO-16-056])-1 (35.9 fbDijet


DM = 2 x mMedM

0.12≥ 2 hcΩ



= 1.0DM

g = 0.1

qg

= 0.01l

g





Summary

• Very exciting times with large amount of new data from the LHC • Increased center-of-mass energy of 13 TeV increases possible production of

new heavy particles. • Searches in dijet and dilepton final states by ATLAS and CMS are sensitive

to mediator signatures. • Many analyses already performed with full 2016+2015 data, including

standard high mass search, and new techniques used to search at low mass.

• Dilepton searches recently started to be interpreted in the context of mediator searches.

• Parameter exclusions from dijet and dilepton searches complement those of more direct searches for missing energy.

• So far no significant excess has been observed, but much more data will be collected this year and in 2018. Stay tuned !

25

Thanks.

gq,l

searches for resonant signatures of production at the lhc

Documents