model-independent (?) search for supersymmetryhorvath/talks/2011/susy_gensearch.pdf · cms-meeting,...

Model-Independent (?) Search forSupersymmetry

CMS-meeting, Budapest-Debrecen, 2011.06.06

Dezso Horváth

KFKI Research Institute for Particle and Nuclear Physics, Budapest

and Institute of Nuclear Research (ATOMKI), Debrecen

Horváth Dezso: Model-Independent SUSY Search Budapest-Debrecen, 2011.06.06 1. fólia – p. 1/36

Outline

Supersymmetry (SUSY).

SUSY models: mSUGRA, cMSSM, GMSB, ...

Benchmark points of CMS.

Exclusion in the SUSY parameter space.

Simplified models.

New search strategy?

Literature:S.P. Martin: A Supersymmetry Primer, hep-ph/9709356, Version 5, 2008

D.S.M. Alves et al., The LHC New Physics Working Group: Simplified Models for LHCNew Physics Searches, arXiv:1105.2838v1 [hep-ph] 13 May 2011

The LHC New Physics Working Group: Signatures of New Physics at the LHC,http://lhcnewphysics.org/

With the support of OTKA Grant NK-81447


Problems of the Standard Model – 1

3 independent (?) components:U(1)Y ⊗ SU(2)L ⊗ SU(3)C

Gravitation? S = 2 graviton?

Asymmetries: right ⇔ left World ⇔ Antiworld

Artificial mass creation: Higgs-field ad hoc

Many fundamental particles:8 + 3 + 1 + 1 = 13 bosons3 × 2 × (2 + 3 × 2) = 48 fermions

Charge quantization: Qe = Qp, Qd = Qe/3

Why the 3 fermion families?Originally: Who needs the muon??

Nucleon spin: how 1/2 produced?


Problems of the Standard Model – 219 free parameters (too many ??):

3 couplings: α, ΘW , ΛQCD; 2 Higgs: MH , λ

9 fermion masses: 3 × Mℓ, 6 × Mq

4 parameters of the CKM matrix: Θ1, Θ2, Θ3, δ

QCD-vacuum: Θ

Mν > 0 ⇒ +7 parameters

Gravitational mass of the Universe:

4% ordinary matter (stars, gas, dust, ν)23% invisible dark matter73% mysterious dark energy

Naturalness (hierarchy):The mass of the Higgs boson quadratically divergesdue to radiative corrections. Cancelled if fermions andbosons exist in pairs.


Coupling constants

[GeV])µ(10

log2 4 6 8 10 12 14 16 18

0

10

20

30

40

50

60 Standard Model)µ(-1

1α

)µ(-12α

)µ(-13α

αi: Local SU(i) couplings

They almost meet at µ ∼ 1013 − 1016 GeV

Do they unite at high energy?


Supersymmetry (SUSY)Fermions and bosons in pairs:

Q|F>= |B>; Q|B>= |F> mB = mF

Identical particles, just spins different

Broken at low energy, no partners: much larger mass?

SUSY Zoo

Chiral multiplets Gauge multiplets

S=1/2 S=0 S=1 S=1/2

quark: qL, qR squark: q1, q2 photon: γ photino (bino): γ(b)

lepton: ℓL, ℓR slepton: ℓ1, ℓ2 weak W± wino: W±

bosons Z zino: Z

higgsino: Φ, Φ′ Higgs: Φ,Φ′ gluon: g gluino: g

Scalar fermion: sfermion, boson’s partner: bosino


SUSY, cont’d.

2 Higgs doublets ⇒ masses to upper and lower fermionsmL = mR, but mL 6= mR

8 Higgs fields ⇒ 5 Higgs bosons: h0,H0,A0,H±

Higgs-parameters: tanβ = v1/v2, masses

SUSY’s quantum number: R parity R = (−1)3B−L+2S

R = +1 particle, R = −1 SUSY partner (sparticle)

Parity-like: R2 = +1

If R conserved, lightest sparticle (LSP) stableR parity may not be much violated: we would see

Neutral LSP can be dark matter (ρ ≈ 0, 1 GeV/cm3!!)


SUSY: coupling constants

Unification!Bend at low energies: SUSY enters with many new

particles ⇒ more loop correctionsHorváth Dezso: Model-Independent SUSY Search Budapest-Debrecen, 2011.06.06 8. fólia – p. 8/36

Minimal Supersymmetric SM (MSSM)Electroweak symmetry breaking

⇓MSSM-fermions mix into ⇒ mass eigenstates

{Electroweak gauginos + higgsinos} ⇒{charginos and neutralinos }

{γ, W±, Z; h0, H0, H±} ⇒ {χ±

1 , χ±

2 ; χ01, χ

02, χ

03, χ

04}

mass grows with index

Lightest SUSY particle (LSP) depends on model, e.g.mSUGRA: χ0

1 or GMSB: gravitino (G)

SUSY breaking ⇒ many (> 100) new parametersmasses, couplings, mixing angles

Lots of model variants, huge parameter space, differentconstraints.


SM and MSSM: menagerie

Almost 50 % discovered already!!(We see half (–1) of all SUSY particles :-)


CMSSM, mSUGRA

Constrained MSSM or Minimal Supergravity Model

Many simplification constraints (boundary conditions),105 ⇒ 5 or 6 parameters, e.g. in mSUGRA:

m1/2: fermion masses at the Grand Unification energy

(GUE ∼ 1014 − 1015 GeV)

m0: boson masses at GUE

A0: SUSY-breaking triple (X–Y–Higgs) couplings at GUE

tanβ = v1/v2: vacuum exp. values upper/lower Higgs fields

mA: mass of a Higgs boson

µ: mixing parameter of the higgsinos (sign ±)


SUSY models


Many-many alternative models


Experimental limits, constraints

No SUSY phenomenon observed, the data limit theparameter space

LEP measurements: Higgs sectorMass of SM Higgs from direct searchesMH > 114.4 GeV; H ∼ h0

Fitting electroweak dataSearch for neutral Higgs bosons (h and A)

BR(b→sγ) measurements at B-factories

Anomalous magnetic moment of the muon (BNL)

WMAP (Wilkinson Microwave Anisotropy Probe):density of dark matter (DM), indirect

Direct searches for DM with ν-detectors


MSSM mass spectrum: preconceptionsEven if we remain sceptic it is worthwhile to know what do most of

the model constructors think (after S.P. Martin)

R parity is barely violated

LSP: χ01

or gravitino

Gluino mass M3 ≡ m(g) ≫ m(χ01),m(χ0

2),m(χ±

1 )

m(ui) ∼ m(di) ∼ m(ci) ∼ m(si) ≫ m(ℓi)

m(ui) ∼ m(di) ∼ m(ci) ∼ m(si) > (0, 6MSUGRA . . . 0, 8GMSB)m(g)

m(uL) ≥ m(uR) . . .m(sL) ≥ m(sR) andm(eL) ≥ m(eR),m(µL) ≥ m(µR) as M2

L ∼ M2R + 0, 5m2

1/2.

t1, b1 lightest squarks and τ1 lightest charged slepton (mixing, Higgscoupling)

m(h0) . 150 GeV ≪ m(A),m(H±),m(H0)


The missing MSSM menagerieKind spin R parity gauge eigenstate mass eigenstate

Higgs bosons 0 +1 H01,H0

2,H+

1 ,H−2 h0,H0,A0,H±

uL, uR, dL, dR same

squark 0 -1 sL, sR, cL, cR same

tL, tR, bL, bR t1, t2, b1, b2

eL, eR, νe same

slepton 0 -1 µL, µR, νµ same

τL, τR, ντ τ1, τ2, ντ

neutralino 1/2 -1 B0, W0, H01, H0

2χ0

1, χ0

2, χ0

3, χ0

4

chargino 1/2 -1 W±, H+1 , H−

2 χ±1 , χ±

2

gluino 1/2 -1 g same

goldstino 1/2 -1 G same

gravitino 3/2


SUSY search

Production in pairs, decay to other SUSY particle(if R conserved)

Lightest (LSP) stable, neutral, not observable

Signal: missing energy

Tipical SUSY decays (LSP = χ01):

squark: q→q + g; q + χ01

slepton: ℓ→ℓ + χ01

gluino: g→q + q + χ01; g + χ0

1

wino: W→e + νe + χ01


SUSY search at LHC

Look for excess above SM expectation: multi-jet eventswith high Emiss

T .

If found, study it, how real.

If real, make a prearranged SUSY selection.

Look for specific features (e.g. long-lived sleptons).

Look for b-jets, lepton pairs, τ ’s.

Determine SUSY parameters with global fitting, searchfor cut-off.


SUSY testing points of CMS

SUSY benchmarkpoints

MSUGRA!Look wherethere is light

LM: Low Mass

HM: High Mass

Favorit: LM1

LM0 added later+ LM0 200 160 10 + -400


mSUGRA: test points and constraints

Blue bandis allowed by WMAP

LM1 left, LM0 added later...


Stop search in LM1

Stop final state:

b-jet + ℓ± + Emiss

Stop signal:

Minv(ℓ±b) cutoff

at Emiss < 180 GeV

Identification:t → 3 hadron jets

b-tag, M (W), M (t)


CMS strategies for discovery

αT search for early discovery in (forced) 2-jet events(ET (J1) > ET (J2)):

Cut αT = ET (J2)MT (J1,J2)

= ET (J2)√(ET (J1)+ET (J2))2−(px(J1)+px(J2))2−(py(J1)+py(J2))2

Exclusive 2-jet, inclusive 3-jet search

Jets + 6HT for > 2 jets, inclusiveScalar mom. sum: HT =

∑i |pT

(Ji)|;Missing transverse mom.:MHT = 6HT = | −

∑i pT

(Ji)|Razor search: test kinematic consistencyfor pair production of heavy particlesTwo jets (inv. mass MR) + 0 or 1 lepton


SUSY search byCMS

αT is good against hadronicbackground

Cut: αT > 0.55

CMS Collaboration,Phys.Lett.B698:196-218,2011.

2010 data, 35 pb−1


LHC, 7 TeV: 2010 data set vs. 2011

In 2010LHC: 46.4 pb−1, CMS: 42.5 pb−1

In March-May 2011LHC: 666 pb−1, CMS: 604 pb−1

In June 2011 luminosity per day ≫ total in 20102 June: 1092 + 1092 bunches, 1042 collisions in CMS and ATLAS

LHC is like Formula 1: boring without collisions


SUSY search: 2010 data, 35...36 pb−1

CMS Collaboration,Phys.Lett.B698:196-218,2011.

CMS Analysis NoteSUS-10-005-pas

LM0 and LM1 are excluded by 2010 CMS data


SUSY search: two more analyses

CMS Analysis Note SUS-10-005-pasBayes and CLs = CLs+b/CLb

CMS Analysis Note SUS-10-009-pasTwo jets (of inv. mass MR)


SUSY search: a newαT analysis

tanβ = 3 10 50

αT > 0.55

CMS Analysis NoteSUS-11-001-pas: Further

interpretation of the search forsupersymmetry based on αT


ATLAS, all-hadronic, 2010 data

More agressive search, not too sensitive to tanβ and A0

Search for squarks and gluinos using final states with jets and missingtransverse momentum with the ATLAS detector in sqrt(s) = 7 TeV

proton-proton collisions.ATLAS, arxiv 1102.5290, Feb. 2011


What and where to look for?

Even if SUSY is valid, MSSM or cMSSM may not be.If we find new physics, how can we tell it is SUSY?

Simplified models ⇒ easier interpretation

LHC inverse problem:Given model and parameters ⇒ prediction of reactions

But experiment works the other way around:We have to tell which model from the data.

SUSY: Cascade decays are model-dependentSimplified models give reactions with few particles ⇒dependence on few masses and cross sections with

relatively wide allowed intervals ⇒ characteristic for severalmodels


OSET: On-Shell Effective TheoryCMS + theory, 2007–2008

Off-shell particles: hard to identify, missing energy harder todetermine

Assume simple production and simple decay of newparticle, analyze decay spectra, find corresponding

deviations from SM.

Bridge between LHC phenomena and Lagrangian of newphysics

http://tools.marmoset-mc.net/osetology_wordpress/Case study: gluino production and decay


OSET: On-Shell Effective Theory

Amplitudes (cross-sections and branching ratios) free parameters

N.Arkani-Hamed et al: MARMOSET, hep-ph/0703088


Simplified Models

Few on-shell particles, simple topology and decaysNot model-independent, but possibly associated with

several models.Possible new physics on well understood SM-base

What can we learn of such analysis?

Boundaries of search sensitivity, both for data analysisand for new theories.

Characterizing new physics signals: what models canbe associated?

Limits on more general models: from possiblecross-sections.


LHC New Physics Working GroupTheorists helping ATLAS and CMS to find benchmark

points to check

http://lhcnewphysics.org/

Example: m(g) < m(qi), LSP = χ0

direct decay, g→qq′χ0

1-step cascade (W), g→qq′χ±→qq′W±χ0

1-step cascade (Z), g→qq′χ′0→qq′Z0χ0

Z better than W, does not decay to neutrínos.Hadronic decay dominates in W and Z,

no lepton, missing transverse energy is for LSP

Signature: 4 jets + large missing energy


Phenomenological MSSMRandom space points in (105 →) 19-parameter pMSSM(1st and 2nd generation sfermions assumed degenerate)

10 (real) sfermion masses

3 gaugino masses

3 trilinear couplings)

µ, tanβ, MA.

Masses: 50-100 GeV ... 1-3 TeV, 1 < tanβ < 60

Experimental and theoretical constraints applied

So far (mSUGRA, GMSB, ...) overlooked phenomena couldemerge

C.F.Berger, J.S.Gainer, J.L.Hewett, T.G.Rizzo:Supersymmetry Without Prejudice, JHEP 0902:023,2009.


Exclusion with simplified modelsSearch for new physics at CMS with jets and missing momentum,

CMS PAS SUS-10-005, April 2011

Pure hadronic events: no neutrino, missing momentum from LSP

gg → 4 jets + LSPs qq → 2 jets + LSPs95 % exclusion curves for production of gluino and squark pairs

Maximal cross-sections ⇒ eliminate modelsSelection: maximal missing transverse momentum


Conclusion

mSUGRA does not seem to be supported by experimentaldata (g-2, LEP, WMAP, LHC, ...)

More general models needed

Simplified approaches: OSET, pMSSM, simple reactions.

Helps to find new, characteristic reactions.

Adjust theory to data, not the other way around.


model-independent (?) search for supersymmetryhorvath/talks/2011/susy_gensearch.pdf · cms-meeting,...

Documents