Di-Higgs production and Higgs self-coupling in ATLAS at HL-LHC - - PowerPoint PPT Presentation

▶
di higgs production and higgs self coupling in atlas at
SMART_READER_LITE
LIVE PREVIEW

Di-Higgs production and Higgs self-coupling in ATLAS at HL-LHC - - PowerPoint PPT Presentation

Mar 06, 2024 137 likes •438 views

Di-Higgs production and Higgs self-coupling in ATLAS at HL-LHC Petar Bokan on behalf of the ATLAS collaboration HL/HE LHC Meeting, Fermilab 4-6 April 2018 Overview o Higgs self-coupling o Di-Higgs production at the LHC o Run-2 results o

slide-1
SLIDE 1

Di-Higgs production and Higgs self-coupling in ATLAS at HL-LHC

Petar Bokan

  • n behalf of the ATLAS collaboration

HL/HE LHC Meeting, Fermilab 4-6 April 2018

slide-2
SLIDE 2

Overview

  • Higgs self-coupling
  • Di-Higgs production at the LHC
  • Run-2 results
  • Di-Higgs prospects at the HL-LHC

− hh → b¯ bb¯ b − hh → b¯ bγγ − hh → b¯ bτ +τ −

2/22

slide-3
SLIDE 3

Higgs potential

  • Important to measure the shape of the Higgs potential

V (φ) = −1 2µ2φ2 + 1 4λφ4

Expanding about minimum: V (φ) → V (v + h)

V = V0 + λv2h2 + λvh3 + 1 4λh4 + ... = V0+

1 2m2 hh2 + m2

h

2v2 vh3 + 1 4 m2

h

2v2 h4 +...

mass term hh-production hhh-production

Standard Model (SM): v = µ √ λ = 246 GeV λ = m2

h

2v2 ≈ 0.13

3/22

slide-4
SLIDE 4

SM Higgs boson pair production at the LHC

  • SM Higgs boson pair production (gluon-gluon fusion - ggF):

h h h h h

4/22

Higgs boson self-coupling Higgs-fermion Yukawa coupling

slide-5
SLIDE 5

SM Higgs boson pair production at the LHC

  • SM Higgs boson pair production (gluon-gluon fusion - ggF):

h h h h h

LO QCD NNLO QCD NLO QCD NLO QCD

qq/gg → t¯ tHH q¯ q → ZHH q¯ q′ → WHH qq′ → HHqq′ gg → HH

MH = 125 GeV

σ(pp → HH + X) [fb]

√s [TeV] 100 75 50 25 8 1000 100 10 1 0.1

Production cross-section small

− two massive final state particles − destructive interference

production mode Cross-section

(14 TeV)

gluon-gluon fusion ∼ 40 fb vector boson fusion ∼ 2 fb Higgs-strahlung ∼ 1 fb t¯ thh ∼ 1 fb

4/22

Higgs boson self-coupling Higgs-fermion Yukawa coupling

arXiv:1212.5581 arXiv:1610.07922

slide-6
SLIDE 6

SM Higgs boson pair production at the LHC

  • SM hh-production ∼ 1000× smaller compared to h-production
  • Current LHC dataset won’t be large enough to reach the sensitivity

5/22

arXiv:1712.08677

Single Higgs boson production

slide-7
SLIDE 7

BSM Higgs boson pair production

Sensitivities to BSM hh-production interesting already at LHC. Non-resonant enhancements:

  • Modified Yukawa/self-coupling
  • New couplings

6/22

Absent in SM

slide-8
SLIDE 8

BSM Higgs boson pair production

Sensitivities to BSM hh-production interesting already at LHC. Non-resonant enhancements:

  • Modified Yukawa/self-coupling
  • New couplings

Resonant Higgs boson pair production

Benchmark BSM hypotheses:

  • Randall-Sundrum graviton

G → hh (spin=2)

  • Heavy Higgs H → hh (spin=0)

X h h

6/22

Absent in SM Resonant production

slide-9
SLIDE 9

Di-Higgs final states

Di-Higgs decay modes and relative branching fractions:

bb WW ττ ZZ bb WW ττ ZZ γγ γγ

0.26% 0.10% 0.028% 0.012% 0.00052% 3.1% 1.1% 0.33% 0.070% 25% 4.6% 34% 7.3% 2.7% 0.39%

The most sensitive channels to the SM hh: hh → b¯ bb¯ b: the highest branching

fraction, large multijet background

hh → b¯ bτ +τ −: relatively large

branching fraction, cleaner final state

hh → b¯ bγγ: small branching fraction,

clean signal extraction due to the narrow h → γγ mass peak

7/22

10.23731/CYRM-2017-002

slide-10
SLIDE 10

Di-Higgs final states

Di-Higgs decay modes and relative branching fractions:

bb WW ττ ZZ bb WW ττ ZZ γγ γγ

0.26% 0.10% 0.028% 0.012% 0.00052% 3.1% 1.1% 0.33% 0.070% 25% 4.6% 34% 7.3% 2.7% 0.39%

  • ther channels being considered:

bbWW, 4W and WWγγ

7/22

10.23731/CYRM-2017-002

slide-11
SLIDE 11

Di-Higgs final states

Di-Higgs decay modes and relative branching fractions:

bb WW ττ ZZ bb WW ττ ZZ γγ γγ

0.26% 0.10% 0.028% 0.012% 0.00052% 3.1% 1.1% 0.33% 0.070% 25% 4.6% 34% 7.3% 2.7% 0.39%

feasibility studies: bbZZ, WWττ and 4τ

7/22

10.23731/CYRM-2017-002

slide-12
SLIDE 12

Di-Higgs final states

Di-Higgs decay modes and relative branching fractions:

bb WW ττ ZZ bb WW ττ ZZ γγ γγ

0.26% 0.10% 0.028% 0.012% 0.00052% 3.1% 1.1% 0.33% 0.070% 25% 4.6% 34% 7.3% 2.7% 0.39%

  • ther channels being considered:

bbWW, 4W and WWγγ feasibility studies: bbZZ, WWττ and 4τ dedicated boosted analyses, VBF-hh investigated

The most sensitive channels to the SM hh: hh → b¯ bb¯ b: the highest branching

fraction, large multijet background

hh → b¯ bτ +τ −: relatively large

branching fraction, cleaner final state

hh → b¯ bγγ: small branching fraction,

clean signal extraction due to the narrow h → γγ mass peak

7/22

10.23731/CYRM-2017-002

slide-13
SLIDE 13

SM Higgs pair production, Run-2 Results

  • Observed (expected) 95% C.L. limit on σ/σSM (Run-2 published results):

channel bbbb bbWW bbττ bbγγ WWγγ ATLAS 13 (21)

  • 117 (161)

747 (386) CMS 342 (308) 79 (89) 28 (25) 19 (17)

  • 2.3-3.2 fb−1

13.3 fb−1 27.5-35.9 fb−1

  • ATLAS publications using the 2015 + 2016 dataset expected.
  • In the context of the HL-LHC prospects studies this is important for those analyses

which perform an extrapolation of the Run-2 result.

  • Possible statistical combination.

ATLAS b¯ bb¯ b: Preliminary ATLAS b¯ bγγ: ATLAS-CONF-2016-004 ATLAS W W γγ: ATLAS-CONF-2016-071 CMS b¯ bb¯ b: PAS HIG-16-002 CMS b¯ bW W : PAS-HIG-17-006 CMS b¯ bττ: Phys. Lett. B 778 (2018) 101 CMS b¯ bγγ: PAS-HIG-17-008 8/22

slide-14
SLIDE 14

SM hh HL-LHC prospects

Two alternative approaches: (1) extrapolation of the Run-2 results → √s = 14 TeV,

  • Ldt = 3000 fb−1

(2) 14 TeV samples with the upgraded detector geometry, upgrade performance functions

9/22

slide-15
SLIDE 15

Run-2 resolved hh → b¯ bb¯ b

  • Background:

∼ 90% multijet and ∼ 10% t¯ t

  • Data-driven estimation of the

multijet background → 2b + 2j events model 4b

[GeV]

4j

m

200 400 600 800 1000 1200 1400

Data / Bkgd

0.5 1 1.5

Events / 100 GeV

1 −

10 1 10

2

10

3

10

4

10

5

10

6

10

7

10 Data Multijet t Hadronic t t Semi-leptonic t Scalar (280 GeV) 100 × SM HH =1)

Pl

M (800 GeV, k/

KK

G =2)

Pl

M (1200 GeV, k/

KK

G Stat+Syst Uncertainty

Preliminary ATLAS

Resolved Signal Region, 2016

  • 1

= 13 TeV, 24.3 fb s

[GeV]

lead 2j

m

60 80 100 120 140 160 180 200

[GeV]

subl 2j

m

60 80 100 120 140 160 180 200

2

Events / 25 GeV 20 40 60 80 100 120 140 160 180 200 220

Preliminary ATLAS

Resolved, 2016

  • 1

= 13 TeV, 24.3 fb s

Signal Region Control Region Sideband Region

  • The reweighting is performed

using one-dimensional distributions iteratively

  • t¯

t normalization from data

10/22

slide-16
SLIDE 16

b b p p b b

h h

SM hh → b¯ bb¯ b HL-LHC prospects

ATL-PHYS-PUB-2016-024 extrapolation of the previous Run-2 result:

  • Ldt = 10.1 →
  • Ldt = 3000 fb−1

Signal and background distributions scaled by f = Ldt|target/ Ldt|current All distributions are scaled by 1.18 to account for an increase in cross-section. Normalizations fixed to the best Run-2 fit values.

11/22

slide-17
SLIDE 17

Extrapolated sensitivity

500 1000 1500 2000 2500 3000

SM

σ / σ 95% C.L. exclusion limit on

2 4 6 8 10 12 14 16 18 20 ATLAS Internal

= 14 TeV s No systematic uncertainties Current systematic uncertainties

]

  • 1

Integrated Luminosity [fb

500 1000 1500 2000 2500 3000

Limits w. no Syst. Limits w. Syst.

1 2 3 Preliminary ATLAS

  • 1

= 13 TeV, 2016, 10.1 fb s

systematic uncertainties in units of signal strength

Source ∆µ Luminosity 0.05 Jet Energy 0.09 b-tagging 0.34 Theoretical 0.10 Multijet 1.85 t¯ t 2.83

  • Extrapolation of the 95% C.L. exclusion limit:

without systematics: σ/σSM = 1.5 with current level of systematics: σ/σSM = 5.2

12/22

ATL-PHYS-PUB-2016-024

slide-18
SLIDE 18

Background uncertainty reduction

Background uncertainty scale relative to current level 0.2 0.4 0.6 0.8 1

SM

σ / σ 1.5 2 2.5 3 3.5 4 4.5 5

Expected 95% C.L. limit L 1/ ∝ Expected 95% C.L. limit, background uncertainties Expected 95% C.L. limit, statistical uncertainties only

ATLAS Internal

  • 1

= 14 TeV, L = 3000 fb s

Preliminary ATLAS

  • 1

= 13 TeV, 2016, 10.1 fb s

  • Significant improvements in (data-driven) background modeling

possible with larger dataset

13/22

ATL-PHYS-PUB-2016-024

slide-19
SLIDE 19

Limits on Higgs self-coupling (Pixel TDR)

Updated in respect to ATL-PHYS-PUB-2016-024

  • extrapolated using a full 2015 + 2016 dataset and
  • includes improved ITk b-tagging expected efficiency

20 − 15 − 10 − 5 − 5 10 15 20

SM HHH

λ /

HHH

λ 20 40 60 80 100 120 140 160 180 200 [fb]

Non-resonant prediction Expected Limit (95% CL) σ 1 ± Expected σ 2 ± Expected Baseline, no systematic uncertainties

  • 1

= 14 TeV, L = 3000 fb s

ATLAS Simulation Internal σ pp→ HH→ bbbb 20 − 15 − 10 − 5 − 5 10 15 20

SM HHH

λ /

HHH

λ 50 100 150 200 250 300 350 400 450 [fb]

Non-resonant prediction Expected Limit (95% CL) σ 1 ± Expected σ 2 ± Expected Baseline, current systematic uncertainties

  • 1

= 14 TeV, L = 3000 fb s

ATLAS Simulation Internal σ pp→ HH→ bbbb

  • Extrapolation of the 95% C.L. exclusion limit:

without systematics: 0.2 < λhhh/λSM

hhh < 7.0

with systematics: −3.5 < λhhh/λSM

hhh < 11.0

14/22

slide-20
SLIDE 20

Minimum jet pT thresholds (TDAQ TDR)

Updated in respect to ATL-PHYS-PUB-2016-024

  • extrapolated using a full 2015 + 2016 dataset and
  • includes improved ITk b-tagging expected efficiency

30 40 50 60 70 80 90 100 110 [GeV]

T

Minimum offline jet p 1.5 2 2.5 3 3.5 4

SM

σ / σ 95% C.L. exclusion limit on

ATLAS Internal

  • 1

= 14 TeV, L = 3000 fb s

(a) No Systematics

30 40 50 60 70 80 90 100 110 [GeV]

T

Minimum offline jet p 2 4 6 8 10 12 14 16 18

SM

σ / σ 95% C.L. exclusion limit on

ATLAS Internal

  • 1

= 14 TeV, L = 3000 fb s

(b) With Systematics

  • Non-resonant hh → 4b σ/σSM 95% exclusion limit as a function of the

minimum offline jet pT

  • 2j35_b60_2j35 trigger most important for Run-2 SM hh

(efficient for 85% of signal)

15/22

slide-21
SLIDE 21

SM hh → b¯ bγγ HL-LHC prospects

ATL-PHYS-PUB-2017-001, Pixel TDR The study is based on √s = 14 TeV Monte Carlo (MC) simulations.

The final state particles at truth level are smeared according to the expected detector resolutions assuming a pile-up scenario with 200 overlapping events (< µ >= 200). The expected efficiencies and fake rates for identifying b-jets and photons are used.

16/22

slide-22
SLIDE 22

Background composition

  • Main backgrounds arise from processes with multiple jets and photons:

− Processes with a single Higgs boson − Continuum background (b¯ bγγ, c¯ cγγ, jjγγ, b¯ bjγ, c¯ cjγ, b¯ bjj)

  • Other backgrounds include Z(b¯

b)γγ, t¯ t and t¯ tγ processes.

[GeV]

γ γ

m 100 105 110 115 120 125 130 135 140 145 150 Events / 2.5 GeV 10 20 30 40 50 60 70

Di-photon invariant mass distribution after the selection except for mbb cut

  • Significance (Pixel TDR): 1.5σ

(based on improved b-tagging performance and photon energy resolution)

  • ATL-PHYS-PUB-2017-001: 1.05σ

17/22

slide-23
SLIDE 23

Limits on Higgs self-coupling

  • Result without systematics (Pixel TDR): 0.2 < λhhh/λSM

hhh < 6.9

(based on improved b-tagging performance and photon energy resolution)

  • ATL-PHYS-PUB-2017-001: −0.8 < λhhh/λSM

hhh < 7.7

18/22

slide-24
SLIDE 24

SM hh → b¯ bτ +τ − HL-LHC prospects

ATL-PHYS-PUB-2015-046 The study is based on √s = 14 TeV Monte Carlo (MC) simulations.

The final state particles at truth level are smeared according to the expected detector resolutions assuming a pile-up scenario with 140 overlapping events (< µ >= 140). The expected efficiencies and fake rates for identifying b-jets and τs are used. All di-τ final states considered.

Results with systematics: 0.6σ −4.0 < λhhh/λSM

hhh < 12

19/22

slide-25
SLIDE 25

Single lepton trigger (TDAQ TDR)

  • SM hh → b¯

bτ +

lepτ − had Run-2 result extrapolation based study (w/o syst)

20/22

slide-26
SLIDE 26

Summary table

channel λhhh/λSM

hhh allowed interval

significance @ 95% C.L. hh → b¯ bb¯ b current syst [-3.5,11.0] hh → b¯ bγγ w/o syst [0.2,6.9] 1.5σ hh → b¯ bτ +τ − syst [-4.0,12.0] 0.6σ

  • Very conservative estimations!

21/22

slide-27
SLIDE 27

Conclusion and Outlook

  • Other ggF channels and the VBF category for the most sensitive channels could

contribute to overall sensitivity

  • Statistical uncertainty dominant for all Run-2 analyses
  • Main systematic uncertainties: b-tagging, τ-identification, ...
  • Background modeling uncertainties can be reduced with an increased amount of

data.

  • Triggering stays the limiting factor (topological triggers could be helpful). Inner

detector upgrades important for hh

  • Hoping for updated results soon. This will provide more realistic estimations and

better understanding of the needed detector performance.

Thank you for your attention!

22/22

slide-28
SLIDE 28

backup slides

22/22

slide-29
SLIDE 29

Minimum jet pT thresholds (TDAQ TDR)

Updated in respect to ATL-PHYS-PUB-2016-024

  • extrapolated using a full 2015 + 2016 dataset and
  • includes improved ITk b-tagging expected efficiency

40 50 60 70 80 90 100 Threshold [GeV]

T

Minimum offline jet p 20 − 15 − 10 − 5 − 5 10 15 20

HHH

λ

HHH

λ 68% C.L. interval for

HHH

λ 95% C.L. interval for

ATLAS

  • 1

= 14 TeV, L = 3000 fb s

(a) No Systematics

40 50 60 70 80 90 100 Threshold [GeV]

T

Minimum offline jet p 20 − 15 − 10 − 5 − 5 10 15 20

HHH

λ

HHH

λ 68% C.L. interval for

HHH

λ 95% C.L. interval for

ATLAS

  • 1

= 14 TeV, L = 3000 fb s

(b) With Systematics

  • Allowed intervals for the λhhh parameter assuming the SM as function of

the minimum offline jet pT .

  • 2j35_b60_2j35 trigger most important for Run-2 SM hh

(efficient for 85% of signal)

23/22