VBF and Vector Boson Scattering at 13 TeV, the EFT approach

VBF and VBS atLHC Run II

RaquelGomez-Ambrosio

Introduction:

VBF/VBSmotivation

Search for newphysics: EFT

Searching fordeviations

Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT

Tools used forsimulation

Monte CarloGenerators

Next steps

Implementing EWcorrections in theMC Generators

Study of VBF/VBS in the LHC at 13 TeV(and) the EFT approach

Raquel Gomez Ambrosio1

Universita & INFN @Torino

28th Rencontres de Blois

May 31, 2016

1In collaboration with G. Passarino

1 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsOutline

1 Introduction:VBF/VBS motivation

2 Search for new physics: EFTSearching for deviationsTop-Down approachBottom-up approachDim-6 LO EFTNLO EFT

3 Tools used for simulationMonte Carlo Generators

4 Next stepsImplementing EW corrections in the MC Generators

2 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsVector Boson Fusion & Vector Boson Scattering

q

q

H

q

q

V

V

V

V

V

V

V = W + , W−, Z

Why look at this production channel? Mainly 2 reasons:

Subleading channel for H production at LHC.

Most important channel for EWSB and unitarity studies.a

aFor example, “delayed unitarity”

3 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsWhy study VBF at LHC RUN-II ?

σVBF is suppressed with respect to σggF but has many experimental advantages:

• Clean signature:

Two well defined tagged jets: large pT

Big rapidity gap between jets

• Reasonable X-Section expected for Run-II luminosity

p

p

Tagged jet 1

Tagged jet 2

Underlying Events

Underlying Events

4 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsWhich decay channel shall we study?

Experimentally: pp → ZZ → 4`

Cleanest channel: 4 leptons in the final state

Theoretically: pp →W +W− → 2`2ν

NLO EW corrections are easier to calculate (less diagrams)

H

Z

Z

e−

e+

µ+

µ−

H

W−

W +

e−

νe

µ+

νµ

5 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsFrom the theorist point of view . . .

VBF/VBS is much more than the previous 2 diagrams

q

q

q

q

q

q

H

q

q

V

V

V

V

V

V

H

H

V

V

V

?

?

?

?

Portal to triple and quartic gauge couplings and to double Higgs production

. . . and to new physics ?

6 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsImportance of NLO EW corrections

pp → ZZ → 4`

Pure EW corrections are of the order of 3.6 % (to the fiducial σ at 8-13TeV)

(Biedermann, Denner, Dittmaier, Hofer, Jager : 1601.07787)

pp →W +W− → 2`2ν

NLO EW corrections are of the order 2% (inclusive σ) and 3% (fiducial σ)

(Biedermann, Billoni, Denner, Dittmaier, Hofer, Jager, Salfelder : 1605.03419)

H

Z

Z

e−

e+

µ+

µ−

H

W−

W +

e−

νe

µ+

νµ

EW corrections are needed for sensitive RUN-2 analyses

7 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstools

8 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsThe κ framework

Back in 2012 . . .

The κ framework postulates (ad-hoc) deviations from the SM couplings anddecay widths, and proposes strategies to measure those “anomalous couplings”

If there is something we learnt from the kappa framework is:

9 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsThe κ framework

Back in 2012 . . .

The κ framework postulates (ad-hoc) deviations from the SM couplings anddecay widths, and proposes strategies to measure those “anomalous couplings”

If there is something we learnt from the kappa framework is:

9 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsNew Physics Quest’s Agenda

Future Steps

1 Include EW corrections in the Monte Carlo Generators2 Explore new scenarios → through Effective Field theory

Top- Down approach: Explore specific BSM modelsBottom- Up approach: studies the most general extensions of the SM

Bottom-Up approach

Leff = LSM︸︷︷︸dim 4

+1

Λ2

∑k

αkO(6)k︸︷︷︸

dim 6

+ . . .︸︷︷︸higher dim. operators

Follows the SM spirit: Do the same calculations with (new) Feynman rulesand a (new) renormalization

For dim-6: “Warsaw Basis” → see backup slides

10 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsEffective field theory

Top-down approach

Take a BSM model and integrate out the heavy fields. i.e: The propagatorbecomes a point.

Recall Fermi theory:

q

q q

q

W± MW →∞

q

q

q

q

11 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsEffective field theory

Top-Down approach

Advantages: Straightforward and useful for Dark Matter and SUSY

Model-dependent, therefore not very efficient

12 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsBottom-up approach

In the bottom up approach we do exactly the opposite:

q

q q

q

W±

q

q q

q

W±

W±

Observe: Unlike in the top-down, the relation is not bijective: onelow-energy operator might come from many different UV completions

13 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsHierarchy of the corrections

Implementing EFT in Vector Boson Scattering

What is LO EFT? is it more relevant than NLO-SM? and NLO-EFT?

M =MLOSM + MNLO

SM︸︷︷︸all available

QCD, EW corrections

+MLOdim=6 +MNLO

dim=6︸︷︷︸κ−frameworkand dim 6 EFT

The answer depends on the scale of new physics Λ , the only way to be sure isto calculate the contributions and include them in the analysis.

14 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsSome words on NLO EFT and renormalization

NLO EFT corrections are quite involded to calculate:

1 Need to renormalize the new Leff → The SM renormalization is notrecyclable.

2 Many new contributions will appear, through non-SM loops:

Ongoing work ... More in future presentations

15 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstools

16 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsPHANTOM & MadGraph5 aMC@NLO

PHANTOM: Monte Carlo Event Generator for Six Parton Final States

Scope: Six fermion final states: VBF, VBS, tt production . . .

Output: Unweighted events (LHA) and Cross-Sections.

Order O(α6) +O(α4α2s ) + interferences

MadGraph5 aMC@NLO: “The MC generator that will calculate any process”

Predecessors: MadGraph, MadEvent, MadGraph5, MC@NLO

Also Lowest Order in the EW sector

Highlight: Can accomodate Dim-6 EFT operators

PHANTOM: A. Ballestrero, A. Belhouari, G. Bevilacqua , V. Kashkan, E. Maina

MadGraph5 aMC@NLO: J. Alwall, R. Frederix, S. Frixione, V. Hirschi, F. Maltoni, O. Mattelaer,

H.-S. Shao, T. Stelzer, P. Torrielli, M. Zaro

17 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsImplementing EW corrections in the Monte Carlo Generators

Challenges of the NLO EW calculation. In order of appearance:

1. There are more diagrams in a NLO EW process (photon emission!)

2. Each of them is more involved to calculate (massive particles!)

3. Renormalization

Big improvements have been published recently

RECOLA: REcursive Computation of One-Loop Amplitudes

COLLIER: Complex One-Loop LIbrary in Extended Regularizations

RECOLA Ansgar Denner, Lars Hofer, Jean-Nicolas Lang, Sandro Uccirati

COLLIER Ansgar Denner, Stefan Dittmaier, Lars Hofer

18 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps



Phase space integration:

The multichannel method has to be optimized for the EW sector: The peakdistribution is much more involved than for QCD processes and in the case ofVBS, we have more particles in the final state.

19 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps



Phase space integration:

The multichannel method has to be optimized for the EW sector: The peakdistribution is much more involved than for QCD processes and in the case ofVBS, we have more particles in the final state.

19 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps



Other difficulties, well known from NLO QCD implementation:

IR divergencies matching → not so bad, many methods are known fromQCD: Antenna Substraction, qT substraction, sector decomposition . . .

Matching to parton shower, double counting → Need to write a plug-in

20 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstoolsThe Message

We live exciting times for particle physics, after LHC Run-II we will have ahuge integrated luminosity

Before the end of the summer we will have news from a new particle . . .

Or a much better understanding of statistical fluctuations.

There are new codes in the market able to provide us with all the SMcorrections needed.

The EFT framework is well established in the theoretical community andon its way to the experimental analyses.

ENJOY IT! (and don’t forget to do your calculations!)

21 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps









21 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps









21 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps









21 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps









21 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps









21 / 22



Introduction:

VBF/VBSmotivation



Top-Downapproach

Bottom-upapproach

Dim-6 LO EFT

NLO EFT



Next steps


higgstools

22 / 22



higgstools

Backup

1 / 6



higgstoolsSM EFT (bottom-up approach)

Leff = LSM︸︷︷︸dim 4

+1

Λ2

∑k

αkO(6)k︸︷︷︸

dim 6

+ . . .︸︷︷︸higher dim. operators

αk is the Wilson coefficient of the kth operator → needs to be calculated

One can build 80 dim-6 operators (compatible with SU(2)× SU(3)× U(1)

and lepton/baryon conservation)

Eqs. of motion, reduce this set to a 59-operator basis

(for one generation of particles! for three → 2499 operators)

“Warsaw Basis” → arXiv: 1008.4884

2 / 6



higgstoolsWarsaw Basis: 34 Bosonic + 25 (+5) fermionic operators

3 / 6



higgstoolsWarsaw Basis: Purely Fermionic Operators

At lowest order in VBF, these operators do not contribute. At NLO they shouldbe taken on account during the renormalization process

q

q

HV

V

q

q

HV

V

4 / 6



higgstoolsWarsaw Basis: Bosonic Operators

5 / 6



higgstoolsVector Boson Fusion Lowest Order EFT

Amplitude with dim-6 insertions:

A =∞∑

n=N

gnA(4)n + g6

∞∑n=N6

n∑l=0

gng l6A

(6)nl , with : g6 =

1√

2GFΛ2.

VBF Signal

At lowest order we don’t have to consider loops originated by dimension 6operators. Therefore only corrections to the actual SM vertices

6 / 6

VBF and Vector Boson Scattering at 13 TeV, the EFT approach

Science