Search for dark matter candidates produced in Z(ℓℓ)+ET

miss events with the ATLAS detector at the LHC

Puzzle of Dark Matter ConferenceDESY, Hamburg, Germany

October 29-31, 2018

Kayla McLean

Searching for dark matter in the ATLAS detector

• The LHC has been delivering 13 TeV proton-proton collisions to the ATLAS detector since 2015• Analysis of the full 2015-2018 Run 2 dataset (149 fb-1) has started and is ongoing• This talk will cover results with the 2015+2016 dataset with integrated luminosity = 36.1 fb-1

1

Invisible dark matter will manifest in the missing transverse momentum, ET

miss

10/30/18 Kayla McLean (UVic)

Signal models

• The dark matter models studied are coordinated by the LHC Dark Matter Working Group• For this particular analysis we study “mono-Z” events with Z(ℓℓ)+ ET

miss final states (ℓℓ = ee or µµ)

2

s-channel simplified models

“Standard” benchmark DM model in Run 2

Two-Higgs-doublet + pseudoscalar model

Newer benchmark, more theoretically complete

t-channel simplified models

To be studied

10/30/18 Kayla McLean (UVic)

Event selection

• Event selection criteria are optimized to isolate potential signal events with large ETmiss recoiling

against Z→ee or Z→µµ, while also reducing backgrounds

3

Selection criteria Background reduced

Exactly one ℓℓ (= ee or μμ) pair with opposite charge

Veto events with 3rd lepton (e or μ) with pT > 7 GeV WZ Z mass window: 76 < mℓℓ < 106 GeV WW/Wt/tt/Z(ττ)

ETmiss > 90 GeV Z+jets

ΔR(ℓℓ) < 1.8 Z+jets, WW/Wt/tt/Z(ττ) Δφ(Z,ET

miss) > 2.7 Z+jets, WW/Wt/tt/Z(ττ)| pT(ℓℓ) - | ET

miss + pT(jets)| |/ pT(ℓℓ) < 0.2 Z+jets

ETmiss/HT > 0.6 (HT = pT(jets) + pT(ℓ1) + pT(ℓ2)) Z+jets

b-jet veto tt

10/30/18 Kayla McLean (UVic)

Backgrounds

4

Background Source Estimation

ZZ ZZ → ℓℓvv, irreducible MC

WZ WZ → ℓvℓ+ℓ-

ℓ from W not reconstructed Data (yield), MC (shape)

Z+jets Z(ee) / Z(μμ) + jetsjets mis-measured as fake ETmiss Data (yield), MC (shape)

W/top WW / Wt / tt / Z(ττ) → ℓ+vℓ-vℓℓ do not come from a Z Data

W+jets W(ℓv) + jetsℓ mis-identified from a jet Data

ttV/ttVV/VVV (V=Z,W) e.g. ttW → (ℓ+vb)(q1q2b)(ℓ-v) MC

• Standard Model background processes also produce Z(ℓℓ)+ETmiss, mimicking the DM signal of interest

% in signal region

10/30/18 Kayla McLean (UVic)

ETmiss in signal region

5

PLB 776 (2017) 318arXiv:1708.09624 [hep-ex]

• Since no significant excess is observed, we set limits using the CLs method

10/30/18 Kayla McLean (UVic)

• Perform statistical analysis on signal region

• Inputs: observed data, background estimates, all sources of systematic errors

• Small overall excess in µµchannel => worse observed limits compared to expected

[GeV]AZ’m

0 100 200 300 400 500 600 700 800 900

[GeV

]χ

m

0

50

100

150

200

250

300

350

400

χ

= 2mAZ’m

Pertu

rbat

ive lim

it

)σ1±Expected limit (Observed limitRelic density

-1 = 13 TeV, 36.1 fbs = 1.0

χ = 0.25, gqAxial-vector, Dirac, g

µµee+

ATLAS Preliminary )σ1±Expected limit (Observed limitRelic density

Simplified model mass exclusion limits (axial-vector)

6

PLB 776 (2017) 318arXiv:1708.09624 [hep-ex]

ATLAS-CONF-2018-051

LO NLO

10/30/18 Kayla McLean (UVic)

Simplified model mass exclusion limits (vector)

7

PLB 776 (2017) 318arXiv:1708.09624 [hep-ex]

[GeV]VZ’m

0 100 200 300 400 500 600 700 800 900

[GeV

]χ

m

0

50

100

150

200

250

300

350

400

χ

= 2mVZ’m

)σ1±Expected limit (Observed limitRelic density

-1 = 13 TeV, 36.1 fbs = 1.0

χ = 0.25, gqVector, Dirac, g

µµee+

ATLAS Preliminary )σ1±Expected limit (Observed limitRelic density

ATLAS-CONF-2018-051

LO NLO

10/30/18 Kayla McLean (UVic)

Mono-Z(ℓℓ) limits compared to direct detection expts

8

Vector (LO)Axial-vector (LO) PLB 776 (2017) 318

arXiv:1708.09624 [hep-ex]

10/30/18 Kayla McLean (UVic)

Mono-Z(ℓℓ) limits compared to other ATLAS searches

9

ATLAS-CONF-2018-051

0 0.5 1 1.5 2 2.5 3 3.5 [TeV]

VZ'm

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

[TeV

]c

m

Preliminary ATLAS

= 1c

= 0, gl

= 0.25, gq

gVector mediator, Dirac DM

All limits at 95% CL

Dijet

PRD 96, 052004 (2017)

-1 = 13 TeV, 37.0 fbsDijet

Dije

t TLA

PRL 121 (2018) 0818016

-1 = 13 TeV, 29.3 fbsDijet TLA

Dije

t + IS

R

ATLAS-CONF-2016-070

-1 = 13 TeV, 15.5 fbsDijet + ISR

resonance

ttEPJC 78 (2018) 565

-1 = 13 TeV, 36.1 fbs resonancett

Dibjet

PRD 98 (2018) 032016

-1 = 13 TeV, 36.1 fbsDibjet

+jetmissTE JHEP 1801 (2018) 126

-1 = 13 TeV, 36.1 fbs

+jetmissTE

g+missTE

Eur. Phys. J. C 77 (2017) 393

-1 = 13 TeV, 36.1 fbs

g+missTE

+Z(ll)missTE PLB 776 (2017) 318

-1 = 13 TeV, 36.1 fbs

+Z(ll)missTE

+V(qq')missTE

ATLAS-CONF-2018-005

-1 = 13 TeV, 36.1 fbs

+V(qq')missTE

VZ'

= mc m´2

= 0.122hcW

Thermal Relic

0 0.5 1 1.5 2 2.5 3 3.5 [TeV]

AZ'm

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

[TeV

]c

m

Preliminary ATLAS

= 1c

= 0, gl

= 0.25, gq

gAxial-vector mediator, Dirac DM

All limits at 95% CL

Pertu

rbat

ive U

nitar

ity Dije

t

PRD 96, 052004 (2017)

-1 = 13 TeV, 37.0 fbsDijet

Dije

t TLA

PRL 121 (2018) 0818016

-1 = 13 TeV, 29.3 fbsDijet TLA

Dije

t + IS

R

ATLAS-CONF-2016-070

-1 = 13 TeV, 15.5 fbsDijet + ISR

resonance

ttEPJC 78 (2018) 565

-1 = 13 TeV, 36.1 fbs resonancett

Dibjet

PRD 98 (2018) 032016

-1 = 13 TeV, 36.1 fbsDibjet

+jetmissTE

JHEP 1801 (2018) 126

-1 = 13 TeV, 36.1 fbs

+jetmissTE

g+missTE Eur. Phys. J. C 77 (2017) 393

-1 = 13 TeV, 36.1 fbs

g+missTE

+Z(ll)missTE PLB 776 (2017) 318

-1 = 13 TeV, 36.1 fbs

+Z(ll)missTE

+V(qq')missTE

ATLAS-CONF-2018-005

-1 = 13 TeV, 36.1 fbs

+V(qq')missTE

AZ'

= mc m´2

= 0.122hcW

Thermal R

elic

Vector (NLO)Axial-vector (NLO)

10/30/18 Kayla McLean (UVic)

2HDM+a model limits

10

• Mono-Z(ℓℓ) one of the most sensitive channels in the 2HDM+a model studied

100 150 200 250 300 350 400 [GeV]am

200

400

600

800

1000

1200

1400

1600

1800

2000

[GeV

]A

m

PreliminaryATLAS -1, 36.1 fb = 13 TeVs

Limits at 95% CLObservedExpected

2HDM+a, Dirac DM = 1c

= 10 GeV, gcm = 1b = 0.35, tanqsin

±H = mH = mAm

= 600 GeVAm

h + ma = mAm

a = mAm

> 20%a/mG

+Z(ll)missTE

PLB 776 (2017) 318+Z(ll)miss

TE

)b+h(bmissTE

PRL 119 (2017) 181804)b+h(bmiss

TE

)gg+h(missTE

PRD 96 (2017) 112004)gg+h(miss

TE

)q+Z(qmissTE

arXiv:1807.11471)q+Z(qmiss

TE

-1=7,8 TeV;4.7,20.3 fbsh(inv)

JHEP 11 (2015) 206, -1=7,8 TeV;4.7,20.3 fbsh(inv)

100 150 200 250 300 350 400 450 500 550 [GeV]am

1

bta

n PreliminaryATLAS -1, 36.1 fb = 13 TeVs

Limits at 95% CLObservedExpected

2HDM+a, Dirac DM = 1c

= 10 GeV, gcm

= 0.35qsin = 600 GeV±H = mH = mAm

= 1btan

> 20%a/mG

tttt

JHEP 09 (2017) 088tttt

+Z(ll)missTE

PLB 776 (2017) 318+Z(ll)miss

TE

)b+h(bmissTE

PRL 119 (2017) 181804)b+h(bmiss

TE

)gg+h(missTE

PRD 96 (2017) 112004)gg+h(miss

TE

t+tmissTE

t+tmissTE

EPJC 78 (2018) 18JHEP 06 (2018) 108

-1=7,8 TeV;4.7,20.3 fbsh(inv)

JHEP 11 (2015) 206, -1=7,8 TeV;4.7,20.3 fbsh(inv)

ATLAS-CONF-2018-051

10/30/18 Kayla McLean (UVic)

Conclusions

• Overview of the analysis and results with 36.1 fb-1 (2015+16) has been presented

• Work is ongoing in the mono-Z(ℓℓ) analysis towards unblinding the full dataset = 149 fb-1

• More DM models to be studied • In addition to simplified models, pursue models with diagrams unique to mono-Z

(2HDM+a, t-channel, …)• Signal region optimization

• New object-based ETmiss significance – better discriminating power for events with fake ET

miss

(see Dilia’s mono-H(bb) talk later today)• New background estimation techniques being studied

• Zγ data-driven estimate of ZZ background• γ+jet data-driven estimate of Z+jet background

• More potential for discovery than ever before!

1110/30/18 Kayla McLean (UVic)

Backup

Invisible Higgs limits

• Look for deviations in SM BR(H→ZZ→4v) = 1.06x10-3 = 0.1%

• At most the branching ratio is 67% or else we would have seen something… at the 95% confidence level

• Small data excess in μμ channel => worse observed upper limits compared to expected

13

PLB 776 (2017) 318arXiv:1708.09624 [hep-ex]

Excl.

10/30/18 Kayla McLean (UVic)