Searches for new resonances decaying to leptons,photons or jets with CMS

Diego Beghin, on behalf of the CMS collaboration

Université Libre de Bruxelles (ULB/IIHE)

EPS-HEP 2019 - Ghent, July 11

1 / 16

Why search for new resonances?

We know our understanding of physics is incomplete. There areopen questions such as

What is the nature of dark matter? Dark energy?How to solve the incompatibility between the Standard Model(SM) and General Relativity at high energies.

New physics theories have been proposed to explain these and otherquestions.

e.g. supersymmetry (SUSY), grand unification models, modelswith extra dimensions.Looking for new resonances predicted by these models is agood way to test them.It’s also important to have model-independent searches, wenever know what new physics might be out there.

2 / 16

New resonance searches at CMS

Great variety of recent searches for new resonances in CMS.Included in this presentation (clickable links!) :

Final states with jetsHigh-mass dijet: CMS PAS EXO-19-012, JHEP 08 (2018) 130.Boosted two-parton searches: PRD 99 (2019) 012005, CMSEXO-17-027, CMS PAS EXO-18-012.Pair-produced 3-jet resonances: PRD 99 (2019) 012010.

Leptonic final statesDilepton: CMS PAS EXO-19-019.Multilepton: CMS PAS EXO-19-002.Lepton and missing energy (in backup slides) : JHEP 06(2018) 128, PLB 792 (2019) 107.

Diphoton final state (in backup) : PRD 98 (2018) 092001.

3 / 16

Jet resonance searches

Searches for new resonances in two-parton (qq, qg, gg) final states have beenperformed for invariant masses all the way from ∼10 GeV to almost 8 TeV.Different trigger strategies are needed for different mass ranges.

10 GeV

Boosted searchesEXO-17-027

EXO-17-024PAS EXO-18-012

500 GeV

Scouting data EXO-16-056

1.6 TeV

Single Jet and HT triggers PAS EXO-19-012

8 TeV

4 / 16

High mass dijet searchesNew paper with full Run2 data: CMS EXO-19-012

Trigger: HT and single-jet.

Wide jet construction: ∆R < 1.1.

QCD multijets bg suppressed by |∆η| < 1.1.

QCD modeled by dσdmjj

= P0(1−x)P1

xP2+P3ln(x) where x =mjj√s.

New data-driven method (based on |∆η| side-bands):enhances sensitivity to broad resonances:

NPredictionSR = R · NData

CRhigh

R = C ·NSimulation

SRNSimulation

CRhigh

C =RData

aux.

RSimulationaux.

= p0 + p1 ·(mjj√s

)3

Raux. =NCRmiddle

NCRhigh

5 / 16

High mass dijet searches: results

Limits considerably improved since JHEP 08 (2018) 130.

Excluded mass ranges (TeV) at 95% CL:

Model Old NewString resonances <7.7 <7.9Scalar diquarks <7.2 <7.5Axigluons/colorons <6.1 <6.6Excited quarks <6.0 <6.3Color-octet scalars <3.4 <3.7SM-like W’ <3.3 3.6SM-like Z’ <2.7 <2.9, 3.1-3.3RS gravitons <1.8, 1.9-2.5 <2.6DM mediators <2.6 2.8

Decays to qq̄+DM Decays to qq̄ only

narrow resonances

broad resonances

6 / 16

Intermediate mass searches

JHEP 08 (2018) 130 - CMSEXO-16-056 - has results for0.49 TeV < mjj < 1.6 TeV, withscouting data.

Limited eventreconstruction, calo-jetsonly.

Trigger: HT > 250 GeV.

2016 data, only 27 fb−1

because of a loss of triggerefficiency in part of thedata.

QCD background fit withone extra parameter:dσdmjj

= P0(1−x)P1

xP2+P3ln(x)+P4ln2(x).

Otherwise same strategy asthe high-mass search.

500 1000 1500 2000 2500 3000 3500Z' mass [GeV]

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45' qgC

oupl

ing

CMS

(13 TeV)-1 & 36 fb-127 fb

95% CL upper limits

/ 2Med > MDMm

= 0DMm

ObservedExpected 1 std. deviation± 2 std. deviation±

←→Lowmass

Highmass

7 / 16

Low mass (boosted) searches

For dijet masses lower than ∼500 GeV, it’s no longer possible to trigger directly on the jets wewish to study. Even the scouting trigger thresholds are too high.

Instead, we look for boosted events, with initial state radiation (ISR).The two partons recoil against a high pT photon or jet.In practice, search for one large-radius jet with a two-pronged substructure.Soft-drop mass algorithm to remove soft and wide-angle contributions to the jet.

(GeV)Am50 100 150 200 250 300 350

qAg

2

4

6

8

10

12

14 (13 TeV)-135.9 fb

CMS Observed

1 s.d.±Expected

2 s.d.±Expected

AK8← →CA15

Observed

1 s.d.±Expected

2 s.d.±Expected

EXO-17-024PRD 99 (2019) 012005

pseudoscalar coupling to bb̄

10 210

Z' mass (GeV)

1−10

1

cou

plin

g st

reng

thq

g'

Observed limit

Expected limit

68% Expected

95% Expected

Indirect constraint: Z

ΥIndirect constraint:

UA2

CDF: Run 1

CDF: Run 2

CMS ISR Jet: Run 2

: Run 2γATLAS ISR Jet &

95% CL Upper limits

(13 TeV)-135.9 fb

CMS

EXO-17-027Submitted to PRL

Z’→ qq̄

(GeV)SDZ' m60 70 80 100 200 300 400

qg'

0.050.06

0.1

0.2

0.3

0.40.50.6

1 13 TeV

CMSPreliminary

Observed

1 s.d.±Expected

2 s.d.±Expected

(2017)-1 (2016) + 41.1 fb-135.9 fb

, R=0.8T

anti-k ← →

(2017)-141.1 fbCA, R=1.5

Observed

1 s.d.±Expected

2 s.d.±Expected

PAS EXO-18-012

leptophobic Z’

8 / 16

Pair-produced three-jet resonances

PRD 99 (2019) 012010 - CMS-EXO-17-030

Theoretical model: RPV gluinos (g̃).

QCD multijets bg and combinatorial noise.Discriminating variables:

Mass asymmetry between the two jettripletsDalitz variables, which exploit topologicaldifferences between signal and bg"Delta cut", which exploits the fact thatthe bg mass scales with the scalar jet pTsum, while the signal mass doesn’t.

QCD+combinatorial bg fit to a smooth function.In higher mass regions, it’s the same as in thedijet searches.

More details in Kin Ho Lo’s presentation.

x1200 1400 1600 1800 2000 2200

Trip

lets

/ 40

GeV

1

10

210

310

DataBackground fit

= 1500 GeVg~ m

Region 4

(13 TeV)-135.9 fb

CMS

[GeV]jjjM1200 1400 1600 1800 2000 2200

data

σda

ta -

fit

2−02

Resonance Mass [GeV]200 400 600 800 1000 1200 1400 1600 1800 2000

[pb]

Β × σ

3−10

2−10

1−10

1

10

210

310

410

Region 1

Region 2

Region 3

Region 4

95% CL upper limitsObservedExpected

1 std. deviation± 2 std. deviation±

qqq)→g~, g~g~→Theory (ppTheory uncertainty

(13 TeV)-135.9 fb

CMSmg̃ > 1500 GeV

9 / 16

Dilepton searchesJHEP 06 (2018) 120 - CMS-EXO-16-047: Z’ → ee/µµ, with 2016 data.CMS PAS EXO-18-006 added 2017 data to the ee final state.

Selections as model independent as possible.Triggers: single electron and single muon.Electrons: HEEP ID, optimized for high energies.Muons: isolation, dedicated ID when pT > 200 GeV.

Backgrounds with real leptons estimated from simulation, normalized to data.QCD multijet background estimated by a data-driven method.

80100 200 300 1000 2000

Eve

nts

/ GeV

5−10

4−10

3−10

2−10

1−101

10

210

310

410

510

610 Data−µ+µ →*/Z γ

ττ, tW, WW, WZ, ZZ, ttJets

(M = 3 TeV)ψZ'

(13 TeV)-136.3 fb

CMS

) [GeV]−µ+µm(80100 200 300 1000 2000

Bkg

) / B

kg−

(Dat

a 1−0.5−

00.5

1

M [GeV]1000 2000 3000 4000 5000

] ZB⋅σ

] Z' /

[B⋅σ[

8−10

7−10

6−10

5−10

4−10CMSPreliminary

Obs. 95% CL limit

Exp. 95% CL limit, median

Exp. (68%)

Exp. (95%)

SSMZ'

ψZ'

)-µ+µ (13 TeV, -1 (13 TeV, ee) + 36.3 fb-177.3 fb

ee (2017) µµ (2016)

Limits (combined): mZ ′SSM

> 4.7 TeV, mZ ′ψ> 4.1 TeV

10 / 16

Multilepton searchesCMS PAS EXO-19-002

Resonant model tested: new light(pseudo)scalars, produced in association with tt̄:gg → tt̄φ→ bqq̄ b̄l−ν l ′+l ′−

Events with 3+ leptons are selected.

Trigger: single electron or muon.

Two mass ranges: 15-75 GeV and 108-340 GeV.

Instead of cutting, inclusive binning of thefollowing discriminating variables:

ST (scalar pT sum of leptons and jets +ZET ).number of b-jets (0, 1+).number of leptons (3, 4+).

Irreducible background estimated from simulation,non-prompt bg from data-driven methods.

Analysis variable: "attractor mass" of theopposite sign same flavor (OSSF) fermions.

20 GeV for the low mass region: M20OSSF

300 GeV in the high mass region: M300OSSF

20 30 40 50 60 70

1−10

1

10

210

310

410

Eve

nts

Data MisIDWZ ZZConversion ZttRare (20)PSφtt

(70)Sφtt Uncertainty

CMS Preliminary (13 TeV)-1137 fb

ee)=0.05→φBR(×2t

g

<400 GeVTS3L(ee), 1B

20 30 40 50 60 70

(GeV)20OSSFDielectron M

00.5

11.5

2

Obs

/Exp

150 200 250 300 350

1−10

1

10

210

310

410

Eve

nts

Data MisIDWZ ZZConversion ZttRare (125)PSφtt

(300)Sφtt Uncertainty

CMS Preliminary (13 TeV)-1137 fb

)=0.05µµ→φBR(×2t

g

>800 GeVTS), 1Bµµ3L(

150 200 250 300 350

(GeV)300OSSFDimuon M

00.5

11.5

2O

bs/E

xp

11 / 16

Multilepton searches: results

One of the first analysiswith full Run2 results.

Assuming gt = 1,branching ratio limitson new scalar(pseudoscalar) bosons→ ee/µµ:< 0.003 (0.03) for thelow mass range.< 0.04 (0.03) for thehigh mass range.

Check out Maxi’s talkfor the non-resonantpart!

50 100 150 200 250 300

Mass (GeV)φ

4−10

3−10

2−10

1−10

1

10

ee] (

pb)

→ φ B

R[

×] φt[tσ

BR=0.05)×2t

(g] with unc. φt[tσ

95% CL upper limits

Observed

Expected

1 s.d.±Expected

2 s.d.±Expected

CMS Preliminary (13 TeV)-1137 fb

φee), Scalar →(φtt

50 100 150 200 250 300

Mass (GeV)φ

4−10

3−10

2−10

1−10

1

10

] (pb

)µµ

→ φ B

R[

×] φt[tσ

BR=0.05)×2t

(g] with unc. φt[tσ

95% CL upper limits

Observed

Expected

1 s.d.±Expected

2 s.d.±Expected

CMS Preliminary (13 TeV)-1137 fb

φ), Pseudoscalar µµ→(φtt

50 100 150 200 250 300

Mass (GeV)φ

5−10

4−10

3−10

2−10

1−10

1

ee]

→ φ B

R[

× 2 tg

95% CL upper limits

Observed

Expected

1 s.d.±Expected

2 s.d.±Expected

CMS Preliminary (13 TeV)-1137 fb

φee), Scalar →(φtt

50 100 150 200 250 300

Mass (GeV)φ

5−10

4−10

3−10

2−10

1−10

1

]µµ→ φ

BR

[× 2 tg

95% CL upper limits

Observed

Expected

1 s.d.±Expected

2 s.d.±Expected

CMS Preliminary (13 TeV)-1137 fb

φ), Pseudoscalar µµ→(φtt

scalar→ ee

pseudoscalar→ µµ

12 / 16

Summary

Many resonance searches in CMS. Too many to summarize in one slide!11 papers covered in this talk, others could not be included, such as:

PRD 98 (2018) 112014 - CMS-EXO-17-021: Search for pair-producedresonances decaying to quark pairs.JHEP 05 (2018) 148 - CMS-EXO-17-011: Search for a heavy right-handed Wboson and a heavy neutrino in events with two same-flavor leptons and two jets.JHEP 04 (2018) 073 - CMS-EXO-16-058: Search for lepton-flavor violatingdecays of heavy resonances and quantum black holes to eµ final states.Many SUSY searches, covered in dedicated talks.

Still a productive research area in CMS, more results with full Run2 data are comingsoon.

New techniques used to improve sensitivity and broaden scope of searches.Scouting is becoming ever more important to extend the mass reach at lowerresonance masses.

Stay tuned for the many other resonance results yet to come!

13 / 16

Backup

14 / 16

Lepton + @@ET

Two recent CMS papers about searches in l +ZET final states:JHEP 06 (2018) 128 - CMS-EXO-16-033, where l = e, µ.PLB 792 (2019) 107 - CMS-EXO-17-008, where l = τ .

Similar analysis strategies:e/µ analysis: single e/µ/γ triggerτ analysis: cross-trigger requiring a hadronic τ (τh) andZET .e/µ are reconstructed in a similar way as in the dilepton search.Taus: only well-reconstructed and identified τh are considered.Final analysis variable: transverse mass MT (l ,ZET ).

(GeV)W'M1000 2000 3000 4000 5000

(fb

)Β × σ

1−10

1

10

210

310

410

510

610 SSM W' LOσ

SSM W' NNLOσ

=0,05 TeV NNLOµ, 2kk Wσ

= 10 TeV NNLOµ, 2kk Wσ

95% CL limit

Median expected limit

1 s.d.±

2 s.d.±

miss

Te+p

(13 TeV)-135.9 fb

CMS

(GeV)W'M1000 2000 3000 4000 5000

(fb

)Β × σ

1−10

1

10

210

310

410

510

610 SSM W' LOσ

SSM W' NNLOσ

=0,05 TeV NNLOµ, 2kk Wσ

= 10 TeV NNLOµ, 2kk Wσ

95% CL limit

Median expected limit

1 s.d.±

2 s.d.±

miss

T+pµ

(13 TeV)-135.9 fb

CMS

(GeV)W'M1000 2000 3000 4000 5000

(fb

)Β σ

1−10

1

10

210

310

410

510

610 SSM W' NNLOσ

SSM W' LOσ

Observed 95% CL limit

Expected 95% CL limit

1 s.d.±

2 s.d.±

missT

+phτ

(13 TeV)-135.9 fb

CMS

e

MW ′SSM

> 5.0 TeV

µ

MW ′SSM

> 4.9 TeV

τ

MW ′SSM

> 4.0 TeV

15 / 16

Diphoton searchesLatest results from PRD 98 (2018) 092001 -CMS-EXO-17-017.

Trigger: double photon with pT > 60 GeV.

Photon ID criteria to suppress electron andjet misID.

Diphoton vertex reconstructed by MVAalgorithm.

Dominant background: prompt SM diphoton,fit to a function:f (mγγ) = m

a+b log(mγγ/GeV)γγ

Eve

nts

/ 20

GeV

1

10

210

310

DataFit model 1 std dev±

EBEB

(GeV)γγm

500 1000 1500 2000 2500

stat

σ(d

ata-

fit)/

-2

0

2

(13 TeV)-135.9 fb

CMS

(GeV)Gm210×5 310 310×2 310×3

) (f

b)γγ

→ G

→

(pp

σ

-110

1

10, J=20.01=κ∼

Expected limit

1 std dev± 2 std dev±

Observed limit

(LO)γγ→G

(13 TeV)-135.9 fb

CMS

(GeV)Gm210×5 310 310×2 310×3

) (f

b)γγ

→ G

→

(pp

σ

-110

1

10, J=20.2=κ∼

Expected limit

1 std dev± 2 std dev±

Observed limit

(LO)γγ→G

(13 TeV)-135.9 fb

CMS

Spin-2RS gravitons

mG > 2.3 TeV

mG > 4.6 TeV

16 / 16