Using unitarity to perform precision topmeasurements

Strong tW scattering at the LHCarXiv:1510.xxxx

Jeff Asaf Dror 1, Marco Farina 2, Javi Serra 3, and Ennio Salvioni4

1Department of Physics, LEPP, Cornell University, Ithaca, NY 14853, USA

2New High Energy Theory Center, Rutgers University, Piscataway, NJ 08854, USA

3Department of Physics, University of California, Davis, Davis CA 95616, USA

4Dipartimento di Fisica e Astronomia, Universita di Padova and INFN, Sezione diPadova, Via Marzolo 8, I-35131 Padova, Italy

1 / 15

Standard Model - still standing

The Standard Model has beenthoroughly tested by previous expts

LHC can have big impact onparameters relating t, h

If no new particles found at LHC,measuring SM parameters is ourfuture

Excellent way to probe SMconsistency

Interestingly, outliers in the 3rdgeneration

Gfitter,

1209.2716

2 / 15

������Unitarity

If SM parameters deviate from theory ñ amplitudes grow withenergy (think WW ÑWW )

Such processes can be searched for at LHC

Can use this feature to put constraints on SM couplings

Ñ Idea used to measure yt1211.3736: Farina, Grojean, Maltoni, Salvioni, Thamm

1211.0499: Biswas, Gabrielli, Mele

Case study: tW Ñ tW to probe ZtLtL,ZtRtR couplings

Currently measured through ttZ8TeV 95% direct limits (∆i ” pgZtiti ´ g

SMZtitiq{g

SMZtiti):

´3 À ∆L À 1 ´6 À ∆R À 4

Weak bounds!

Stronger indirect limits but can be avoided depending on modeldetails

3 / 15

tW Ñ tW at LHC

tW Ñ tW scattering (electroweak (EW) process):

t

W

Z ,γ,h

t

W

t

W

b

W

t

LHC

q j

g t

t

W

t

W

highenergy

forwardjet

Sensitive to gZtRtR , gZtLtL , gWtb, gWWZ , ...

Poorly constrained

4 / 15

Amplitudes can EXPLODE!

Quick calculation Ñ M „ s for gZtRtR , gZtLtL ‰ SM values.

SM,σ∝1/s

ΔL=+1,σ∝s

ΔR=+1,σ∝s

0.5 1.0 1.5 2.01

10

100

1000

104

∆i ” pgZtiti ´ gSMZtitiq{g

SMZtiti

5 / 15

Existing ttW search

Same-sign `

CMS performed a search for ttW CMS, 1406.7830

Only included Opg2`ns gwq contributions (σQCD) in ttWestimations, e.g.,

u

d

d

W

g tt

EW process contributes at Opgsg3wqSmall in SM but can be large if ∆L,∆R ‰ 0

Expected=39.7, Observed=36

Can put limit on size of σEW and hence (∆L,∆R)

6 / 15

8TeV limits

ttWttZ

-6 -4 -2 0 2 4 6-6

-4

-2

0

2

4

6

DLD

R

8 TeV, 68 and 95% CL, stat + sys

´4 À ∆R À 3

Simulate EW signal asfunction of coupling deviations(applying CMS cuts). Find,

σEW p∆L,∆Rq

Madgraph, Pythia, PGS

Right: compare new ttWlimits with traditional ttZsearch at 8TeV

Significantly better limits evenwithout dedicated search!

7 / 15

13TeV dedicated search

At 8 TeV existing search happened to be sensitive to signal

Would like to have dedicated search

Unitarity effects become more pronounced at higher energies

Goal:

1 Simulate backgrounds and signal2 Find optimal cuts for L “ 300fb´1, 13TeV.3 Produce projected limits

8TeV 13TeV(not to scale)

8 / 15

Simulation and Backgrounds

Simulate all bkg at 8TeV

Matched at LORescale to NLO values

Compare distributions to those found by CMS

Use the CMS search to calibrate monte carlo for 13TeV(important for tricky backgrounds such as misID`, misIDQ)

Dominant bkgs:

pttW qQCD , ttZ, misIDQ, tth, pW˘W˘jnq, misID`

electron given η - depen-dent misidentified chargeprob

Simulated prob. of misIDb as ` with smeared pT

Curtin, Galloway, Wacker, 1306.5695

9 / 15

Optimization

Dominant backgrounds (pttW qQCD and MisID`) are reducible!

ttWQCD

MisIDℓttWEW(ΔR=+1)

0 50 100 150 200 250 3000.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

pTℓ

Probability

ttWQCD

MisIDℓttWEW(ΔR=+1)

0 1 2 3 4 50.00

0.05

0.10

0.15

0.20

0.25

0.30

ηFJ

Probability

Best cuts: CMS ttW cuts +

p`1T ą 100GeV, {ET ą 50GeV, m`` ą 125GeV

|ηFJ | ą 1.75 , dηFJ1,FJ2 ą 2

10 / 15

13 TeV limits

Assuming Nobs “ Nbkg we can get 13TeV projected limitsUsed likelihood and assumed 50% systematic on misID`

-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5-2

-1

0

1

2

3

´1 À ∆R À 1

Rontsch, Schulze: 1404.1005

11 / 15

Higher Dimensional Operators (HDO)

Above SM couplings, what about HDO?

L Ąicp1qL

Λ2H:DµHqLγ

µqL `icp3qL

Λ2H:σiDµHqLγ

µσiqL

`icRΛ2

H:DµHtRγµtR

where ε ” iσ2.

δgZtLtLgSMZtLtL

“cp3qL ´ c

p1qL

`

1´ 43s

2W

˘

v2

Λ2,

δgZbLbLgSMZbLbL

“

�����

��cp1qL ` c

p3qL

`

1´ 23s

2W

˘

v2

Λ2

δgWtLbL

gSMWtLbL

“ cp3qL

v2

Λ2,

δgZtRtRgSMZtRtR

“cR43s

2W

v2

Λ2

12 / 15

HDO limits

-0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0

-0.4

-0.2

0.0

0.2

0.4

0.6

Rontsch, Schulze: 1404.1005

13 / 15

What else you got?

Many opportuni-ties for future study

bW Ñ tZ

Probed in tZjσ „ sMainly sensitive to ∆L(c

p1qL )

bW Ñ th

Probed in thjσ „ sAlready used to measure yt

tZ Ñ th .

Probed in tthjσ „

?s

Sensitive to ∆L,∆R, yt

...

14 / 15

Conclusions

We can use unitary to measure parameters in the SM

Considered tW Ñ tW as a case study

Improved measurement at 8TeVSignificant improvements on measurement can be made for13TeV analysis

NLO analysis? Better optimization? Combining channels?

How far we can push this idea?

Let’s see whats out there!

15 / 15