CP Sensitive Observables in VBF Production Verena Walbrecht, - - PowerPoint PPT Presentation

▶
cp sensitive observables in vbf production
SMART_READER_LITE
LIVE PREVIEW

CP Sensitive Observables in VBF Production Verena Walbrecht, - - PowerPoint PPT Presentation

Feb 27, 2023 136 likes •364 views

CP Sensitive Observables in VBF Production Verena Walbrecht, Katharina Ecker, Davide Cieri, Sandra Kortner Max Planck Institute for Physics (Werner-Heisenberg-Institut) Friday 5 th October, 2018 Setup Previous Talk: CP sensitivity of

slide-1
SLIDE 1

CP Sensitive Observables in VBF Production

Verena Walbrecht,

Katharina Ecker, Davide Cieri, Sandra Kortner

Max Planck Institute for Physics (Werner-Heisenberg-Institut)

Friday 5th October, 2018

slide-2
SLIDE 2

Setup

Previous Talk: CP sensitivity of ∆φsign

jj

in VBF (https://indico.cern.ch/event/693955/) Integrated Luminosity: L= 150 fb−1 (end of Run II) MC samples : Prod_v12 VBF Category: njets ≥ 2 and mjj > 120 GeV Test CP Invariance: CP-odd coupling parameter κAZZ For comparison: very simple signal model – only VBF for signal model – no background included – no width scaling (not needed for shape only fit) – no best prediction scaling – no systematics Only shape information

Category VBF (MG5) m4ℓ ∈[118,129] GeV 18.94 VBF-enriched 9.90 Fraction 52 %

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

2/20

slide-3
SLIDE 3

CP Sensitive Observables

3 possibilities to test CP invariance in VBF production:

  • 1. Azimuthal angle between the tagging jets in the final state: ∆φsign

jj

Previous talk: (https://indico.cern.ch/event/693955/)

  • 2. 1st optimal observable for VBF production: OO1,jjH

Nikita’s talk: (https://indico.cern.ch/event/693949/) Antoine’s talk: (https://indico.cern.ch/event/743162/) Spin-CP H → ττ:

(https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PAPERS/HIGG-2015-06/)

  • 3. 1st optimal observable for VBF production and decay: OO1,jj4ℓ

Supporting Note (Appendix P) - L= 36 fb −1:

(https://cds.cern.ch/record/2231596/files/ATL-COM-PHYS-2016-1604.pdf)

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

3/20

slide-4
SLIDE 4
  • 1. Short Reminder: ∆φ

sign jj

Definition of ∆φsign

jj

∈ [−π, π]:

if yleading jet >ysubleading jet: if ysubleading jet >yleading jet:

∆φsign

jj

= φleading jet - φsubleading jet

∆φsign

jj

= φsubleading jet - φleading jet

3 − 2 − 1 − 1 2 3

sign jj

φ ∆ 0.05 0.1 0.15 0.2 0.25 0.3 Normalized to unit area ATLAS Work in progress 4l → ZZ* → H

13 TeV > 120 GeV

jj

m 2, ≥

j

N

=1

α

=1, c

SM

κ VBF SM, 2 1 =

α

=5, c

AVV

κ =1,

SM

κ VBF, 2 1 =

α

=-5, c

AVV

κ =1,

SM

κ VBF,

Discriminates between negative and positive CP-odd coupling parameters Advantage: model independent

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

4/20

slide-5
SLIDE 5
  • 2. Optimal Observable for VBF Production: OO1,jjH

Matrix element for VBF production: sum of CP-even contribution from the SM and a CP-odd contribution:

MMix = MSM + ˜

d · MCP−odd, where ˜ d = v · κAVV 4·Λ·κSM tan(α) Squared matrix element:

|MMix|2 = |MSM|2 + ˜

d · 2ℜ (M∗

SMMCP−odd) + ˜

d2 · |MCP−odd|2 Interference term is CP-odd ⇒ can be used to measure CP invariance Parameter configuration ˜ d = 1:

Λ = 1 TeV, v = 246.22 GeV, κSM = √

2, κAVV = 22.975, cos(α) = 1/

√

2

Definition of optimal observables:

OO1,jjH = |MMix|2 − |MSM|2 − |MCP−odd|2 |MSM|2

and OO2,jjH = |MCP−odd|2

|MSM|2

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

5/20

slide-6
SLIDE 6
  • 2. Optimal Observable for VBF Production: OO1,jjH

Matrix elements are calculated with the MadGraph5 tools Option standalone_cpp: standalone C++ code for each subprocess Takes as input the four-momenta of the incoming and outcoming particles (5) and calculates the value of the matrix element: MG5 version of EFT samples from Prod_v12: MG5_aMC_v2_3_3:

import model HC_NLO_X0_UFO p p > j j x0 $$ w+ w- z / a @0

  • utput standalone_cpp PROC_SA_CPP_VBF

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

6/20

slide-7
SLIDE 7
  • 2. Optimal Observable for VBF Production: OO1,jjH

60 diagrams

⇒ sum over all possible flavour configurations ij → klH weighted by the parton

distribution functions

|MSM|2 = ∑

i,j,k,l

fi(x1)fj(x2) |MSM|2 (ij → klH) 2ℜ (M∗

SMMCP−odd) =

∑

i,j,k,l

fi(x1)fj(x2)2ℜ (M∗

SMMCP−odd) (ij → klH)

Reconstructed Bjorken x:

xreco

1

=

MjjH

√

s eyjjH and xreco

2

=

MjjH

√

s e−yjjH

Same PDF which was used for EFT samples: NNPDF23_lo_as_0130_qed

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

7/20

slide-8
SLIDE 8
  • 2. Optimal Observable for VBF Production OO1,jjH:

Our distribution:

Avv

κ 1,

O 10 − 8 − 6 − 4 − 2 − 2 4 6 8 10 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05

SM kAvv=5 kAvv=-5 SM kAvv=5 kAvv=-5 SM kAvv=5 kAvv=-5

Nikita’s distribution:

ττ-paper:

Similar distribution as shown by Nikita & in the paper T R U T H R E C O T R U T H

(https://indico.cern.ch/event/693949/) (H → ττ Paper) 10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

8/20

slide-9
SLIDE 9
  • 2. Optimal Observable for VBF Production:

OO1,jjH:

10 − 8 − 6 − 4 − 2 − 2 4 6 8 10

jjH

OO1 0.05 0.1 0.15 0.2 0.25 0.3 Normalized to unit area ATLAS Work in progress 4l → ZZ* → H

13 TeV > 120 GeV

jj

m 2, ≥

j

N

=1

α

=1, c

SM

κ VBF SM, 2 1 =

α

=5, c

AVV

κ =1,

SM

κ VBF, 2 1 =

α

=-5, c

AVV

κ =1,

SM

κ VBF,

OO2,jjH:

2 4 6 8 10 12 14 16 18 20

jjH

OO2 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Normalized to unit area ATLAS Work in progress 4l → ZZ* → H

13 TeV > 120 GeV

jj

m 2, ≥

j

N

=1

α

=1, c

SM

κ VBF SM, 2 1 =

α

=5, c

AVV

κ =1,

SM

κ VBF, 2 1 =

α

=-5, c

AVV

κ =1,

SM

κ VBF,

OO1,jjH:

Discriminates between negative and positive CP-odd coupling parameters

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

9/20

slide-10
SLIDE 10
  • 3. OO for VBF Production and Decay: OO1,jj4ℓ

BSM coupling can enter in both the production (p) and the decay (d) vertices Matrix element for VBF production and decay:

MMix = (gSM·MSM,p+gCP−odd·MCP−odd,p)·(gSM·MSM,d+gCP−odd·MCP−odd,d)

Squared matrix element:

|MMix|2 = g4

SM|MSM,p|2|MSM,d|2 + g4 CP−odd|MCP−odd,p|2|MCP−odd,d|2+

+ g3

SMgCP−odd

[ |MSM,p|2ℜ(|MSM,d|∗|MCP−odd,d|)+ + ℜ(|MSM,p|∗|MCP−odd,p|)|MSM,d|2] + + g2

SMg2 CP−odd(|MSM,p|2|MCP−odd,d|2 + |MCP−odd,p|2|MSM,d|2)+

+ g3

CP−oddgSM

[ |MCP−odd,p|2ℜ(|MSM,d|∗|MCP−odd,d|)+ + ℜ(|MSM,p|∗|MBSM,p|)|MSM,d|2] .

(1)

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

10/20

slide-11
SLIDE 11
  • 3. OO for VBF Production and Decay: OO1,jj4ℓ

Parameter configuration gSM = gCP−odd = 1:

κSM = κAVV = √

2, cos(α) = 1/

√

2

Assumption:

|MSM,p|2 = |MSM,d|2

and

|MCP−odd,p|2 = |MCP−odd,d|2

Definition of optimal observables:

OO1,jj4ℓ = |MMix|2 − |MSM|2 − |MCP−odd|2 − 2 · |MSM| · |MCP−odd| |MSM|2

and

OO2,jj4ℓ = |MCP−odd|2 |MSM|2

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

11/20

slide-12
SLIDE 12
  • 3. OO for VBF Production and Decay: OO1,jj4ℓ

Matrix elements are also calculated with MadGraph5:

import model HC_NLO_X0_UFO p p > j j x0 $$ w+ w- z / a @0, x0 > z > l+ l- l+ l- standalone_cpp PROC_SA_CPP_VBF_H4l

75 diagrams (now: 8 four-momenta as inputs)

⇒ sum over all possible flavour configurations (same as for OO1,jjH)

Same PDF which was used for EFT samples: NNPDF23_lo_as_0130_qed

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

12/20

slide-13
SLIDE 13
  • 3. OO for VBF Production and Decay OO1,jj4ℓ:

Our distribution:

Avv

κ 1,

O 4 − 2 − 2 4 6 8 10 12 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

SM kAvv=5 kAvv=-5 SM kAvv=5 kAvv=-5 SM kAvv=5 kAvv=-5

Distribution from supporting note: Similar distribution as shown in the supporting note

(Couplings and EFT supporting Note) 10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

13/20

slide-14
SLIDE 14
  • 3. Optimal Observable for VBF Production and Decay:

OO1,jj4ℓ:

1 − 0.8 − 0.6 − 0.4 − 0.2 − 0.2 0.4 0.6 0.8 1

jj4l

OO1 0.05 0.1 0.15 0.2 0.25 0.3 0.35 Normalized to unit area ATLAS Work in progress 4l → ZZ* → H

13 TeV > 120 GeV

jj

m 2, ≥

j

N

=1

α

=1, c

SM

κ VBF SM, 2 1 =

α

=5, c

AVV

κ =1,

SM

κ VBF, 2 1 =

α

=-5, c

AVV

κ =1,

SM

κ VBF,

OO2,jj4ℓ:

0 0.0002 0.0004 0.0006 0.0008 0.001 0.0012 0.0014 0.0016 0.0018 0.002

jj4l

OO2 0.2 0.4 0.6 0.8 1 1.2 1.4 Normalized to unit area ATLAS Work in progress 4l → ZZ* → H

13 TeV > 120 GeV

jj

m 2, ≥

j

N

=1

α

=1, c

SM

κ VBF SM, 2 1 =

α

=5, c

AVV

κ =1,

SM

κ VBF, 2 1 =

α

=-5, c

AVV

κ =1,

SM

κ VBF,

OO1,jj4ℓ:

Discrimination between negative and positive CP-odd coupling parameters

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

14/20

slide-15
SLIDE 15

FIT

comparison of ...

  • 1. ... shape only: ∆φ

sign jj

  • 2. ... shape only: OO1,jjH
  • 3. ... shape only: OO1,jj4ℓ
  • 4. ... rate

at a Luminosity of 150 fb−1

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

15/20

slide-16
SLIDE 16

Expected Limits (L = 150 fb−1)

SM Asimov:

Avv

κ

α

c

4 − 3 − 2 − 1 − 1 2 3 4

) λ

  • 2ln (

2 4 6 8 10 12

68% CL 95% CL

sign jj

φ ∆

jjH

OO1

jj4l

OO1 Rate

ATLAS

4l → ZZ* → H

  • 1

13 TeV, L=150 fb VBF = 1

SM

κ = 1,

Hgg

κ

Expected Limits (150 fb−1):

Observable Best Fit 68% CL 95% CL

∆φsign

jj

0.00 [-1.67, 1.66] [–,–]

OO1,jjH

0.00 [-1.77, 1.82] [–,–]

OO1,jj4ℓ

0.00 [-1.31, 1.53] [-3.28, –] Rate 0.00 [-2.01, 1.96] [-2.93, 2.89]

Expected Limits H → ττ

(36 fb−1), translated into cακAVV:

https://cds.cern.ch/record/2298266/? Observable Best Fit 1σ 2σ

OO1,jjH

0.00 [-0.50, 0.53] [1.59, 1.77]

Shape only more sensitive for 68% CL as rate?

ττ: OO more sensitive than ∆φ

sign jj

at 1σ level, for us opposite behaviour Why OO1,jj4ℓ asymmetric? Stat artefact or real physics?

d ~

  • 0.4
  • 0.2

0.2 0.4 NLL ∆ 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

=1.0) µ =0, d ~ OO Expected ( =1.0) µ =0, d ~ Expected ( jj sign φ ∆

σ 1

ATLAS

  • 1

= 8 TeV, 20.3 fb s

R u n I

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

16/20

slide-17
SLIDE 17

Expected Limits (L = 150 fb−1)

BSM Asimov cακAvv = 0.43:

Avv

κ

α

c

4 − 3 − 2 − 1 − 1 2 3 4

) λ

  • 2ln (

2 4 6 8 10 12

68% CL 95% CL

sign jj

φ ∆

jjH

OO1

jj4l

OO1

ATLAS

4l → ZZ* → H

  • 1

13 TeV, L=150 fb VBF = 1

SM

κ = 1,

Hgg

κ

Expected Limits (150 fb−1):

Observable Best Fit 68% CL 95% CL

∆φsign

jj

0.43 [-1.01, 3.17] [–,–]

OO1,jjH

0.43 [-1.25, 2.48] [–,–]

OO1,jj4ℓ

0.43 [-0.86, 2.18] [-2.56, –]

OO1,jj4ℓ most promising

For negativ coupling parameters ∆φsign

jj

has a better performance as

OO1,jjH at 68 %CL

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

17/20

slide-18
SLIDE 18

Expected Limits (L = 150 fb−1)

BSM Asimov cακAvv = − 1.45:

Avv

κ

α

c

4 − 3 − 2 − 1 − 1 2 3 4

) λ

  • 2ln (

2 4 6 8 10 12

68% CL 95% CL

sign jj

φ ∆

jjH

OO1

jj4l

OO1

ATLAS

4l → ZZ* → H

  • 1

13 TeV, L=150 fb VBF = 1

SM

κ = 1,

Hgg

κ

Expected Limits (150 fb−1):

Observable Best Fit 68% CL 95% CL

∆φsign

jj

−1.45

[–, 0.13] [–, 1.92]

OO1,jjH −1.45

[–, -0.25] [–, 2.19]

OO1,jj4ℓ −1.45

[-3.28, -0.19] [–, 1.21]

Only with OO1,jj4ℓ possible to set 68% CL limits

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

18/20

slide-19
SLIDE 19

Expected Limits (L = 150 fb−1)

BSM Asimov cακAvv = − 2.89:

Avv

κ

α

c

4 − 3 − 2 − 1 − 1 2 3 4

) λ

  • 2ln (

2 4 6 8 10 12

68% CL 95% CL

sign jj

φ ∆

jjH

OO1

jj4l

OO1

ATLAS

4l → ZZ* → H

  • 1

13 TeV, L=150 fb VBF = 1

SM

κ = 1,

Hgg

κ

Expected Limits (150 fb−1):

Observable Best Fit 68% CL 95% CL

∆φsign

jj

−2.89

[–, -0.36] [–, 1.13]

OO1,jjH −2.89

[–, -0.72] [–, 0.92]

OO1,jj4ℓ −2.89

[–, -1.25] [–, -0.06]

OO1,jj4ℓ tightest limits, but the nll flattens out for larger values (like ∆φsign

jj

) Overall OO1,jjH has the best performance

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

19/20

slide-20
SLIDE 20

Summary

Comparison of 3 possibilities to test CP invariance in VBF production:

∆φsign

jj

, OO1,jjH, OO1,jj4ℓ

All of them show a good separation between SM, negative and positive BSM coupling parameters First limits only using the shape information For SM and small BSM values: largest sensitivity is given by OO1,jj4ℓ Larger BSM values: OO1,jjH has the largest sensitivity, since OO1,jj4ℓ flattens out Not understood yet:

  • 1. At 68% CL ∆φ

sign jj

has a better performance as OO1,jjH

  • 2. Asymmetry of OO1,jj4ℓ
  • 3. Sensitivity of shape-only fit comparable to rate-only fit

Next steps: – Answer open questions – Include OO2 in the fit (strategy of CP-Analysis?) – Add ggF signal and include OO4ℓ in 0 and 1 jet category

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production

20/20

slide-21
SLIDE 21

BACKUP

10/05/2018

  • V. Walbrecht - CP Sensitive Observables in VBF Production