sensitivity to new physics scenarios in invisible higgs

Sensitivity to new physics scenarios in invisibleHiggs boson decays at CLIC

K. Mekala, A.F. Zarnecki

in collaboration with M. Iglicki and B. Grzadkowski

Faculty of Physics, University of Warsaw

Work carried out in the framework of the CLICdp collaboration

Research supported by

31.10.2019International Workshop on Future Linear Colliders

LCWS 2019K.Mekala, A.F. Zarnecki Faculty of Physics, University of Warsaw

Sensitivity to new physics scenarios in invisible Higgs boson decays at CLIC 1 / 22

Compact Linear Collider (CLIC)

First stage @ 380 GeV⇒ focus on studying Higgs boson and top-quark properties

Higgs production

0 1000 2000 3000

[GeV]s

1−10

1

10

210

HX

) [fb

]→ - e+

(eσ

eνeνH

-e+He

ZH

ZHH

Htt

eνeνHH

Higgs couplings to SM particles tested at % levelWhat about couplings (decays) to BSM states?

K.Mekala, A.F. Zarnecki Faculty of Physics, University of Warsaw


Signal

SM(-like) Higgs boson decay to invisible states (Dark Matter ?)

Signature of invisible Higgs decay:

two jets consistent with hadronic Z decay higher statistics

missing energy-momentum consistent with production ofinvisible massive state of 125 GeV



Background processes considered



Simulation framework

event samples generated with Whizard 2.7.0

Non-Higgs background: qq, ll , qqqq, qqll , qqlν, qqννSM Higgs boson production: H + qq, H + ll , H + νν

(with 100% SM decays)Signal: H + qq production with Higgs defined as stable

CLIC energy spectra for 380 GeV

CLIC integrated luminosity of 1000 fb−1 (unpolarised)

detector simulation and event reconstruction with Delphes,using modified1 CLICdet Stage1 cards

Two jets reconstructed with VLC algorithm (R = 1.5, β = γ = 1)

1required to make Higgs invisible in the detectorK.Mekala, A.F. Zarnecki Faculty of Physics, University of Warsaw


Signature of e+e− → HZ → jj + inv

Two-jet events without electrons, muons, or isolated photons...



Preselection

Preselection cuts were used to select events with proper signatureand kinematics consistent with invisible Higgs boson decay:

Remove events with isolated electrons, muons or photonswith energy above 2 GeV, 3 GeV and 5 GeV respectively

Energy “lost” in jet clustering below 10 GeV

Two-jet topology: y23 < 0.01 and y34 < 0.001

Jet invariant mass: 80 < Mjj < 100 GeV (Z mass)

Dijet emision angle: | cos Θjj | < 0.8 (Z direction)



Preselection cut example

Di-jet invariant mass distribution with preselection cut indicated

40 50 60 70 80 90 100 110 120 130 140, GeVdijetm

2−10

1−10

1

10

210

Norm

aliz

ed e

vents

num

ber

No H

SMH+qqinvH



Preselection

Efficiency of preselections cuts

Event class EfficiencyNon-higgs background 0.37%

including qqνν 23%qqlν 0.68%qq 0.087%

SM Higgs decays 1.70%including H + νν 4.60%

H + qq invisible decays 47.0%



Preselection

Recoil mass distribution after preselection cutsFor 1000 fb−1 collected at 380 GeV assuming BR(H→inv) = 10%

50 100 150 200 250 300recoil mass, GeV

10

210

310

410

# e

vent

s

SM background

inv→H



Selection

Final event selection based on the multivariate analysis.Variables used as input for Boosted Decision Tree (BDT):

1 αjj – angle between two jets in LAB frame2 mjj – dijet invariant mass3 mmiss – missing mass4 Ejj – dijet energy5 pmiss

t – missing transverse momentum



Selection

0.4− 0.3− 0.2− 0.1− 0 0.1 0.2BDT response

0

1

2

3

4

5

6

7dx

/ (1

/N)

dN

Signal (test sample)

Background (test sample)

Signal (training sample)

Background (training sample)

Kolmogorov-Smirnov test: signal (background) probability = 0.004 (0.002)

U/O

-flo

w (

S,B

): (

0.0,

0.0

)% /

(0.0

, 0.0

)%

TMVA overtraining check for classifier: BDT

Highest significance for invisible Higgs decays for BDT cut ∼ 0.06K.Mekala, A.F. Zarnecki Faculty of Physics, University of Warsaw


Results

95% C.L. limit expected for 1000 fb−1 collected at 380 GeV:

BR(H → inv) < 0.86%

Assuming no excess above predicted SM background is observedCLICdp preliminary

Result consistent with the old study:BR(H → inv) < 0.94% expected for 500 fb−1 collected at 350 GeVM. A. Thomson, The European Physical Journal C, 76(2):72



Interpretation

In Higgs-portal models, new scalars fields φ coupling to darkmatter particles can mix with the SM Higgs field h resulting in twomass eigenstates:(

h1h2

)=

(cosα sinα− sinα cosα

)(hφ

)

If α� 1, h1 is SM-like (the observed 125 GeV state),but it can also decay invisibly via φ component (BR ∼ sin2 α)

If h2 is also light, it can be produced in e+e− collisions in the sameway as the SM-like Higgs boson; invisible decays dominate.

We consider Vector-fermion dark matter model (VFDM) [arXiv:1710.01853]



Interpretation

Limit on the invisible decays of the 125 GeV Higgs boson (h1)can be interpreted in terms of the VFDM mixing angle limits.

0.01

0.1

0 10 20 30 40 50 60

95

%C

Llim

iton

sin(α

)

Excluded - gHZZ

Excluded - CMS

Excluded - CLIC

DM [GeV]m

Based on WHIZARD calculations assuming gX = 1.K.Mekala, A.F. Zarnecki Faculty of Physics, University of Warsaw


Limits on new scalar production

Same approach can be used to search for production of h2 statein the process e+e− → Z h2 → qq + inv

50 100 150 200 250 300recoil mass, GeV

1

10

210

310

410

# e

vent

s

SM background = 150 GeVSM = 200 GeVSM = 250 GeVSM




Expected limits on the h2 production cross section, relative to SM,for 1000 fb−1 at 380 GeV assuming BR(h2 → inv) ≈ 100%

0

2

4

6

8

10

12

14

16

120 140 160 180 200 220 240 260 280Cro

ssse

ctio

nlim

itσ(e

+e

- ->

ZH

')/σ

SM

[%]

mh2[GeV]

CLICdp preliminary




Expected limits on the h2 production cross section, relative to SM,for 2500 fb−1 at 1500 GeV assuming BR(h2 → inv) ≈ 100%

1

10

100

1000

200 400 600 800 1000 1200

Cross

sectionlim

itσ(e

+e- ->ZH')/σ

SM[%

]

mh2 [GeV]

combinedpnegppos

CLICdp preliminary

pneg:P(e−)=−80%2000 fb−1

ppos:P(e−)=+80%500 fb−1



The VFDM model

Expected limits on the production cross section can be translatedwithin the VFDM model into limits on the mixing angle α.

0.1

1

200 400 600 800 1000 1200

95%

CLlim

itonsin(α)

mh2 [GeV]

CLIC 1.5 TeV

CLIC

380 GeV



Conclusions

1 Search for invisible Higgs boson decays based on theWhizard event generation and fast simulation withDelphes.

2 CLIC running at 380 GeV can constrain the invisible decays ofthe SM Higgs boson to below 1%.

3 Results consistent with the previous study based on fullsimulation.

4 The study can be extended to search for extra scalars at CLICoperating at 380 GeV and 1.5 TeV.

5 Cross section limits can be translated to the limits on newphysics model parameters.



References

H. Abramowicz et al.

Higgs physics at the CLIC electron-positron linear collider.

Eur. Phys. J., C77(7):475, 2017.

A. Ahmed, M. Duch, B. Grzadkowski, and M. Iglicki.

Multi-component dark matter: the vector and fermion case.

The European Physical Journal C, 78(11):905, Nov 2018.

D.Azevedo, M.Duch, B.Grzadkowski, D.Huang, M.Iglicki, and R.Santos.

Testing scalar versus vector dark matter.

Phys. Rev., D99(1):015017, 2019.

M. A. Thomson.

Model-independent measurement of the e+e− → HZ cross section at afuture e+e−-linear collider using hadronic Z decays.

The European Physical Journal C, 76(2):72, 2016.



BACKUP

Results of the model-independent fit to Higgs boson measurement at CLIC.

Parameter Relative precision

350GeV + 1.4TeV + 3TeV

1ab−1 + 2.5ab−1 + 5ab−1

gHZZ 0.6 % 0.6 % 0.6 %

gHWW 1.0 % 0.6 % 0.6 %

gHbb 2.1 % 0.7 % 0.7 %

gHcc 4.4 % 1.9 % 1.4 %

gHττ 3.1 % 1.4 % 1.0 %

gHµµ − 12.1 % 5.7 %

gHtt − 3.0 % 3.0 %

g†Hgg 2.6 % 1.4 % 1.0 %

g†Hγ γ − 4.8 % 2.3 %

g†HZγ − 13.3 % 6.7 %

ΓH 4.7 % 2.6 % 2.5 %

co

up

ling

re

lativ

e t

o S

M

0.8

0.9

1

1.1

1.2

1%

5%

HΓ

µ

c τ b t W Z g γ

γZ H

CLICdp

model independent

November 2018

-1350 GeV, 1 ab-1+ 1.4 TeV, 2.5 ab

-1+ 3 TeV, 5 ab

H. Abramowicz et al. Eur. Phys. J., C77(7):475, 2017.



BACKUP

Preselection cuts on jet clustering resuts

2–3 separation (− log10 y23)

1 2 3 4 5 6 7 8 9

23y

10-log

3−10

2−10

1−10

1

10

210

310

Norm

aliz

ed e

vents

num

ber

No H

SMH+qqinvH

3–4 separation (− log10 y34)

1 2 3 4 5 6 7 8 9

34y

10-log

3−10

2−10

1−10

1

10

210

310

Norm

aliz

ed e

vents

num

ber

No H

SMH+qqinvH



BACKUP

Preselection cuts on di-jet final state (Z boson)

Di-jet invariant mass

40 50 60 70 80 90 100 110 120 130 140, GeVdijetm

2−10

1−10

1

10

210

Norm

aliz

ed e

vents

num

ber

No H

SMH+qqinvH

Di-jet emission angle

1− 0.8− 0.6− 0.4− 0.2− 0 0.2 0.4 0.6 0.8 1)θcos(

1−10

1

10

210

310

410

Norm

aliz

ed e

vents

num

ber

No H

SMH+qqinvH



BACKUP

Input variables for multivariate analysis, for invisible decays of 125 GeV Higgs

0.5 1 1.5 2 2.5 3

data.Angle

0

0.5

1

1.5

2

2.5

3

3.5

0.06

81

/ (1

/N)

dN

SignalBackground

U/O

-flo

w (

S,B

): (

0.0,

0.0

)% /

(0.0

, 0.0

)%

Input variable: data.Angle

100− 50− 0 50 100 150 200 250 300

P4_miss.M()

0

0.002

0.004

0.006

0.008

0.01

0.012

0.014

0.016

10.6

/

(1/N

) dN

U/O

-flo

w (

S,B

): (

0.0,

0.0

)% /

(0.0

, 0.0

)%

Input variable: P4_miss.M()

100 120 140 160 180 200 220

P4_dijet.Energy()

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

3.56

/

(1/N

) dN

U/O

-flo

w (

S,B

): (

0.0,

0.0

)% /

(0.0

, 0.0

)%

Input variable: P4_dijet.Energy()

20 40 60 80 100 120 140 160 180 200

P4_miss.Pt()

0

0.002

0.004

0.006

0.008

0.01

0.012

0.014

0.016

0.018

0.02

4.99

/

(1/N

) dN

U/O

-flo

w (

S,B

): (

0.0,

0.0

)% /

(0.0

, 0.0

)%

Input variable: P4_miss.Pt()

82 84 86 88 90 92 94 96 98 100

P4_dijet.M()

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.51

2 /

(1/N

) dN

U/O

-flo

w (

S,B

): (

0.0,

0.0

)% /

(0.0

, 0.0

)%

Input variable: P4_dijet.M()



BACKUP

Signal significance as a function of the BDT cutassuming BR(H→inv) = 1%

0.4− 0.3− 0.2− 0.1− 0 0.1 0.2

Cut value applied on BDT output

0

0.2

0.4

0.6

0.8

1

Eff

icie

ncy

(P

uri

ty)

Signal efficiency

Background efficiency

Signal puritySignal efficiency*purity

S+BS/

For 386 signal and 129601 background isS+Bevents the maximum S/

1.9085 when cutting at 0.0574

Cut efficiencies and optimal cut value

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

Sig

nif

ican

ce



BACKUP

Final state Efficiency Npre

Background

qqνν 23,00% 72135

qqlν 0,68% 37588

qq 0,087% 19234

qqll 0,043% 593

qqqq 0,0010% 51

In total: 0,37% 129601

SM Higgs decays

HSM + νν 4,60% 2515

HSM + ll 0,017% 3

HSM + qq 0,0057% 47

In total: 1,70% 2565

Invisible Higgs boson decays

Hinv + qq 47,00% 38557



sensitivity to new physics scenarios in invisible higgs

Documents