Goldstini and the Z-peaked ATLAS excess Alberto Mariotti Based on - - PowerPoint PPT Presentation

Goldstini and the Z-peaked ATLAS excess

Alberto Mariotti

03 - 09 - 2015 GGI - Gearing up for LHC13 Based on arXiv:1506.08803 with S. Liew, K. Mawatari, K. Sakurai, M. Vereecken, and on arXiv:1102.2386 with R. Argurio,Z. Komargodski, and on arXiv:1112.5058 with also G. Ferretti, K. De Causmaecker

SUSY after LHC8

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• SUSY under pressure by negative LHC8 results
• However still main candidate for BSM physics
• Bet: SUSY beyond Minimality
• Investigate un-explored SUSY scenarios !!!
• Keep eyes open on possible ahead indications from LHC8

ATLAS Z-peaked excess Non-minimal SUSY

ATLAS Z-peaked excess (1503.03290)

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

Channel SR-Z ee SR-Z µµ SR-Z same-flavour combined Observed events 16 13 29 Expected background events 4.2 ± 1.6 6.4 ± 2.2 10.6 ± 3.2 Flavour-symmetric backgrounds 2.8 ± 1.4 3.3 ± 1.6 6.0 ± 2.6 Z/γ+jets (jet-smearing) 0.05 ± 0.04 0.02+0.03

−0.02

0.07 ± 0.05 Rare top 0.18 ± 0.06 0.17 ± 0.06 0.35 ± 0.12 WZ/ZZ diboson 1.2 ± 0.5 1.7 ± 0.6 2.9 ± 1.0 Fake leptons 0.1+0.7

−0.1

1.2+1.3

−1.2

1.3+1.7

−1.3

• 2 SF leptons + at least 2-jets + MET
• Minimal cuts (object selection, isolation, etc …)
• MET> 225 GeV, H_T > 600 GeV
• Different categories based on m_ll
• Excess over SM bkg in Z-mass region (81 — 101 GeV)

ATLAS Z-peaked excess (1503.03290)

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

Channel SR-Z ee SR-Z µµ SR-Z same-flavour combined Observed events 16 13 29 Expected background events 4.2 ± 1.6 6.4 ± 2.2 10.6 ± 3.2 Flavour-symmetric backgrounds 2.8 ± 1.4 3.3 ± 1.6 6.0 ± 2.6 Z/γ+jets (jet-smearing) 0.05 ± 0.04 0.02+0.03

−0.02

0.07 ± 0.05 Rare top 0.18 ± 0.06 0.17 ± 0.06 0.35 ± 0.12 WZ/ZZ diboson 1.2 ± 0.5 1.7 ± 0.6 2.9 ± 1.0 Fake leptons 0.1+0.7

−0.1

1.2+1.3

−1.2

1.3+1.7

−1.3

!!! Significance 3.0 sigma !!!

• 2 SF leptons + at least 2-jets + MET
• Minimal cuts (object selection, isolation, etc …)
• MET> 225 GeV, H_T > 600 GeV
• Different categories based on m_ll
• Excess over SM bkg in Z-mass region (81 — 101 GeV)

˜ χ0

1,2

˜ G ˜ g Z q q

ATLAS Z-peaked excess (1503.03290)

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• ATLAS interpretation in GMSB scenario
• Gluino pair production
• Higgsino decaying to Goldstino + Z-boson

BP1 BP2 900 GeV 700 GeV 600 GeV 200 GeV ∼ 0 GeV ∼ 0 GeV

m˜

g

m˜

χ0

m ˜

G

200 400 600 800 1000 1200

m˜

χ0

1 [GeV]

600 700 800 900 1000 1100 1200 1300 1400

m˜

g [GeV]

Allanach, Kvellestad, Raklev 2015

m˜

g < m˜ χ0

1

5 10 15 20 25 30

Expected ATLAS onZ signal events

GMSB interpretation

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• GMSB interpretation conflict with JETS+MET searches
• Hadronic decay of Z-boson contribute significantly
• Also analogous CMS search impose stringent constraints
• ATLAS benchmark points are already excluded
• B. Allanach, A. Raklev, A. Kvellastad

200 400 600 800 1000 1200

m˜

χ0

1 [GeV]

600 700 800 900 1000 1100 1200 1300 1400

m˜

g [GeV]

Allanach, Kvellestad, Raklev 2015

m˜

g < m˜ χ0

1

5 10 15 20 25 30

Expected ATLAS onZ signal events

100 200 300 400 500 600 700 800 900

∆m [GeV]

0.0 0.2 0.4 0.6 0.8 1.0

Gluino branching ratio

Allanach, Kvellestad, Raklev 2015

˜ χ0

1,2 g

˜ χ±

1 t b

˜ χ0

1,2 t ¯

t ˜ χ0

1,2 b¯

b

GMSB interpretation

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• GMSB interpretation conflict with JETS+MET searches
• Hadronic decay of Z-boson contribute significantly
• Also analogous CMS search impose stringent constraints
• ATLAS benchmark points are already excluded
• Also decay chain not viable in the MSSM
• B. Allanach, A. Raklev, A. Kvellastad

Recap: ATLAS Z-peaked excess

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• NMSSM
• Split SUSY
• Composite models
• MSSM with light squarks, bino, higgsino

Alternative Proposed Explanations

• G. Barenboim, J. Bernabeu, et al.
• B. Allanach, A. Raklev, A. Kvellastad
• U. Ellwanger
• N. Vignaroli
• A. Kobakhidze, A. Saavedra, L. Wu , J. Min Yang
• J. Cao, L. Shang, J. Min Yang, Y. Zhang
• M. Cahill-Rowley, J.L. Hewett, A. Ismail, T.G. Rizzo
• X. Lu, S.Shirai, T. Terada

J.H. Collins, J. Asaf Dror, M. Farina

Recap: ATLAS Z-peaked excess

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• JETS+MET
• Analogous CMS search
• Two Jets + 2 leptons + MET
• No H_T cut

Main LHC8 Constraints

arXiv:1502.06031

• NMSSM
• Split SUSY
• Composite models
• MSSM with light squarks, bino, higgsino

Alternative Proposed Explanations

arXiv:1405.7875

• JETS+MET
• Analogous CMS search
• Two Jets + 2 leptons + MET
• No H_T cut

Main LHC8 Constraints

Recap: ATLAS Z-peaked excess

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• NMSSM
• Split SUSY
• Composite models
• MSSM with light squarks, bino, higgsino

Alternative Proposed Explanations

Spectrum compression Different selection cuts

Gauge Mediation interpretation

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• Extra fermionic light modes (Goldstini) in the spectrum
• Mass computable perturbatively
• Coupling structure dictated by supersymmetry
• Exotic phenomenology at collider
• Accommodate the ATLAS excess!!!

Gauge Mediation with multiple susy br sectors

• S. Liew, A. M., K. Mawatari, K. Sakurai, M. Vereecken

SUSY breaking paradigm

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

Susy Sector MSSM Mediation

• Hidden sector with spontaneous SUSY breaking
• Mediation transmits SUSY breaking to MSSM
• Two main scenarios: Gravity or Gauge Mediation

F

Gauge Mediation

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

Susy Sector MSSM Gauge Inter.

• Susy breaking mediated by SM gauge interactions
• Soft terms are loop suppressed
• M typical supersymmetric hidden sector scale
• Low supersymmetry breaking scale F
• Calculability, address susy flavour problem …

msoft = α 4π F M

Goldstino

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• What do we know about hidden sector?
• Spontaneous susy breaking Massless Goldstino
• Eaten via superHiggs mechanism:

Gauge Mediation

m3/2 ' F MP l

Gravity mediation

m3/2 ⌧ msoft

m3/2 ∼ msoft

G

Goldstino

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

Gauge Mediation

m3/2 ⌧ msoft

• Tipically gravitino mass of order eV
• Light Gravitino equivalent to Goldstino in HE processes
• Goldstino interaction with MSSM fixed by supersymmetry
• What do we know about hidden sector?
• Spontaneous susy breaking Massless Goldstino
• Eaten via superHiggs mechanism: m3/2 '

F MP l

G

Goldstino Interactions

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

Goldstino can be described as constrained superfield

X2

NL = 0

⇒ XNL = G2 2F + √ 2θG + θ2F

Reproduce both soft terms and Goldstino couplings

Goldstino Lagrangian

Reproduce Volkov-Akulov lagrangian

L ⊃ Z d2θmλ 2F XNLW2 = 1 2mλλλ + imλ √ 2F ✓ GλD − i 2λσµ¯ σνGFµν ◆ + . . . L ⊃ Z d4θ m2

Q

F 2 X†

NLXNLQ†Q = m2 Qq†q +

m2

Q

F

• GGψq† + ¯

G ¯ ψq

• + . . .

Z.Komargodski and N.Seiberg ‘09

Multiple SUSY br sectors

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• Every sector has its own goldstino field
• One combination is the true goldstino eaten by the superHiggs
• The others are extra fermionic particles (Pseudo-Goldstini)

What changes if there are multiple susy breaking sectors?

• Suppose having two decoupled susy breaking sector
• What are the pheno consequences?

K.Benakli and C.Moura ’07 C.Cheung, Y.Nomura and J.Thaler ‘10

• Total susy breaking scale
• Total soft terms are sum of two sector contributes

Goldstino and Pseudo-Goldstino

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

Susy Sector 1 MSSM Gauge Gauge Susy Sector 2 F1, G1 F2, G2

G = 1 F (F1G1 + F2G2) mG = m3/2

True-Goldstino Pseudo-Goldstino

G0 = 1 F (−F2G1 + F1G2) mG0 =???

F = q F 2

1 + F 2 2

mλ = m(1)

λ

+ m(2)

λ

Pseudo-Goldstino mass

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• Contribution from gravity is negligible
• Pseudogoldstino can get mass via radiative corrections
• What is the typical size of ?

Gauge Mediation

mG0

Gravity Mediation

• PseudoGoldstino get mass

mG0 = 2m3/2

Cheung, Nomura, Thaler ’10

• R. Argurio, A.M., Z. Komargodski ’11

PGld mass computation

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• Mass matrix for the Goldstini (one zero eigenvalue)

M12 G1 G2

L ⊃ G1 G2 ✓ −F2/F1 M12 M12 M12 −F1/F2 M12 ◆ ✓ G1 G2 ◆ mG0 = ✓F1 F2 + F2 F1 ◆ M12

• Physical PGld mass
• One has to compute radiative corrections to

Pseudo-Goldstino couplings

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

General Gauge Mediation (GGM) Lagrangian

• B and C are GGM functions encoding susy breaking sector
• Goldstino coupling reads
• Goldstino coupling in model independent formalism

1 g2 LGGM = 1 2C0D2 − iC1/2λσµ∂µ¯ λ − 1 4C1F 2

µν − 1

2

• B1/2λ2 + h.c.
• 1

g2 LGold = iB1/2 √ 2f ✓ GλD − i 2λσµ¯ σνGFµν ◆ + 1 √ 2f (C0 − C1/2)Gσµ∂µ¯ λD + i √ 2f (C1 − C1/2)Gσν∂µ¯ λF µν + . . .

P.Meade, N.Seiberg and D.Shih ’08

Pseudo-Goldstino mass

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

Lagrangian for two goldstini coupled only via MSSM gauge fields

• PseudoGoldstino mass

G1 G2 G1 G2 +

L = L(1)

GGM + L(2) GGM + L(1) G1 + L(2) G2

• GGM parameterization of SUSY breaking sectors
• Leading contribution at order g4
• B and C are GGM functions encoding susy breaking sectors
• Expressions checked in toy models (MGM)

m ˜

G0 = g4

2 ✓ 1 F 2

1

+ 1 F 2

2

◆ Z d4p (2π)4 B(1)

1/2

⇣ C(2) − 4C(2)

1/2 + 3C(2) 1

⌘ + 1 ↔ 2

Pseudo-Goldstino mass

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• Typical value of PGLD mass is around weak scale
• For instance in Minimal Gauge Mediation with same scales
• Can be enhanced if susy breaking scales are different
• For heavy soft terms we can reach hundreds of GeV
• Validity limit set by backreaction of sector 1 on sector 2

F1 ⇠ F2 ⇠ F ) m ˜

G0 '

g4 (16π2)3 F M ' g2 (16π2)2 msoft

F1 F2 ⇠ F ) m ˜

G0 '

g2 (16π2)2 ✓F1 F2 ◆ msoft

PGLD in Gauge Mediation

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• Extra light fermionic degrees of freedom w.r.t. to MSSM
• PGLD get mass radiatively
• Their couplings is fixed by soft terms of the different sectors
• Can lead to distinctive phenomenology

More SUSY breaking sectors Gauge Mediation

• f SUSY breaking

Pseudo-Goldstini

• is the total neutralino mass
• determines PGLD coupling with neutralino
• It depends on contributions from the two sectors

Kχ

mχ = m(1)

χ + m(2) χ

PGLD Lagrangian for two sectors

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

Simplified model for Goldstino and PseudoGoldstino and the MSSM

L = LMSSM + Lkin(G) + Lkin(G0) + mG0G0G0+ + mχ 2 √ 2F Gσµ¯ σνχFµν + Kχ mχ 2 √ 2F G0σµ¯ σνχFµν + . . .

• If susy decay chains will end into massive PGLD

Kχ = −m(1)

χ

mχ F2 F1 + m(2)

χ

mχ F1 F2

Kχ 1

• If PGLD coupling are enhanced susy decays end into PGLD
• Phenomenology determined by
• LOSP: lightest observable supersymmetry particle
• NLSP: Pseudo-Goldstino
• True Goldstino is the LSP but does not play a role at LHC
• At LHC the MET is carried by PGLD (massive particle)
• New signatures can be realized

PGLD phenomenology at LHC

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess N.Craig, J.March-Russell and M.McCullough ’10 J.Thaler and Z.Thomas ’11; G.Ferretti, A.M., K.Mawatari and C.Petersson ’13 R.Argurio, K.De Causmaecker, G.Ferretti, A.M., K.Mawatari and Y.Takaesu ’12 K.i.Hikasa, T.Liu, L.Wang and J.M.Yang '14

• Possibility of hiding signals of Gauge Mediated SUSY

Back to ATLAS excess

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

˜ χ0

1,2

˜ G′ ˜ g Z g

• Simplified model with Gluino, Higgsino and Pseudo-Goldstino
• Other sparticles decoupled
• Pseudo-Goldstino carries missing energy
• Different phenomenology varying PGLD mass

Can it be the Pseudo-Goldstino of gauge mediation?

Susy decay channels: gluino

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• We fix the Neutralino (Higgsino) mass such that one loop

gluino decay process dominates

• Squarks at O(TeV) scale
• Robust result as soon as Neutralino is mostly Higgsino

˜ t t ˜ Hu ˜ g g

m˜

g − m˜ χ0

1,2 = 200 GeV

⇒ BR[˜ g → g ˜ χ0

1,2] ≥ 85%

Susy decay channels: Higgsino

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• Higgsino decay determined by effective Lagrangian

L ˜

G0 = i

˜ yi

γ

2 √ 2F ˜ G0σµ¯ σν ˜ χ0

i Aµν + i ˜

yi

ZT

2 √ 2F ˜ G0σµ¯ σν ˜ χ0

i Zµν

+ ˜ yi

ZLmZ

√ 2F ¯ ˜ χ0

i ¯

σµ ˜ G0Zµ + ˜ yi

h

√ 2F ˜ χ0

i ˜

G0h

• Couplings depends on susy breaking soft terms
• We can find configurations such that the decay into PGLD

and Z is dominant

˜ y

Susy decay channels: Higgsino

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• Higgsino decay determined by effective Lagrangian

L ˜

G0 = i

˜ yi

γ

2 √ 2F ˜ G0σµ¯ σν ˜ χ0

i Aµν + i ˜

yi

ZT

2 √ 2F ˜ G0σµ¯ σν ˜ χ0

i Zµν

+ ˜ yi

ZLmZ

√ 2F ¯ ˜ χ0

i ¯

σµ ˜ G0Zµ + ˜ yi

h

√ 2F ˜ χ0

i ˜

G0h

• Couplings depends on susy breaking soft terms
• We can find configurations such that the decay into PGLD

and Z is dominant

˜ y

300 400 500 600 700 800

m ~

G’ [GeV]

0.2 0.4 0.6 0.8 1

Neutralino branching ratios

Z ~ G’ h ~ G’ γ/Z/h ~ G

m~

χ1

0 ≈ 800 GeV

BR[˜ χ0 → Z ˜ G0] ≥ 80%

Example with

Susy production and decay chain

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

Main SUSY channel is gluino pair production

• Decay chain leads to two Z plus jets plus MET

Differences with ATLAS model

• First step of decay only one jet (gluon)
• Missing energy carried by PGLD (massive particle)

p p ˜ g ˜ χ0

i

˜ g ˜ χ0

j

g Z ˜ G′ ˜ G′ Z g

Simulation and main constraints

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

p p ˜ g ˜ χ0

i

˜ g ˜ χ0

j

g Z ˜ G′ ˜ G′ Z g

• Model simulated in MadGraph5+Pythia
• FastLim/Atom for detector simulator
• LHC constraints from
• ATLAS multi-jet
• CMS 2-leptons plus Jets + MET
• For FIT:
• For EXCLUSION at 95%CL:

Fit and exclusion parameterized by R factor

R ≡ NSUSY/(Nobs − NSM) Ri ≡ N i

SUSY/N UL,i BSM

arXiv:1502.06031 arXiv:1405.7875 M.Papucci, K.Sakurai, A.Weiler and L.Zeune '14 J.Alwall, M.Herquet, F.Maltoni, O.Mattelaer and T.Stelzer '11

Results for benchmark scenarios

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

ATLAS fit and CMS constraints ATLAS fit and ATLAS constraints

As GMSB with one SUSY br sector

m˜

χ0

1,2 = m˜

g − 200 GeV

m0

˜ G = 0

Results for benchmark scenarios

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

Very compressed scenario

m˜

χ0

1,2 = m˜

g − 200 GeV

m ˜

G0 = m˜ χ0 − 100 GeV

ATLAS fit and CMS constraints ATLAS fit and ATLAS constraints

Results for benchmark scenarios

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

OPTIMAL CASE

m˜

χ0

1,2 = m˜

g − 200 GeV

ATLAS fit and CMS constraints ATLAS fit and ATLAS constraints

m ˜

G0 = m˜ χ0 − 200 GeV

Diff distributions for best fit point

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

m˜

g = 1 TeV

m˜

χ0 = 800 GeV

m ˜

G0 = 600 GeV

[GeV]

miss T

E 200 250 300 350 400 450 500 Events / 25 GeV 2 4 6 8 10 12

Data (ATLAS SR-Z) Signal + BG BG

Jet Multiplicity 2 3 4 5 6 7 8 9 Events 2 4 6 8 10 12

Data (ATLAS SR-Z) Signal + BG BG

Statistics is poor but shape is consistent Jet multiplicity differs from scenario with 3-body gluino decay and fit better the ATLAS excess data

˜ g → qq ˜ χ0

Comments

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• ATLAS Jets + MET search evaded thanks to compression
• CMS constraint evaded because of different H_T cut
• Statistics is low so it is premature to draw strong conclusions
• Jet distribution hints at one-loop 2-bd gluino decay
• SUSY decay chain is robust in PGLD model
• More parameters to fit the excess

Comparison with other SUSY explanations

Conclusions

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• SUSY is under pressure after LHC8
• However:
• SUSY can be realized in non-minimal way
• SUSY can be already hidden in the existing LHC8 data
• Scenarios with multiple susy breaking sectors present rich

phenomenology

• Goldstini: extra light fermionic degrees of freedom with

computable couplings

• A simplified goldstini model can fit the Z-peaked ATLAS

excess without conflicting with other LHC constrains

Goldstini

Conclusions

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

Thank you very much for your attention! LHC13 will soon give some new excitement !!!

Backup slides

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

2 4 6 8 10 2 3 4 5 6 7 8 Events Jet Multiplicity GMSB P1 P2 data-background

• Jet Multiplicity distribution for GMSB ATLAS model and for

NMSSM compressed possibility

Ellwanger arXiv:1504.02244

Backup slides

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• pt_lep1 > 25, pt_lep2 > 10
• eta_lep < 2.4
• d_0 differs from electron to muon (more stringent to avoid cosmic)
• N_lep > 1
• pt_jet > 35, eta_j < 2.5
• N_jet > 1
• MET > 225
• H_T = p_T(j)+p_T(lep) > 600
• 81 < m_ll < 101
• also delta_phi (jet_1,2 , MET) < 0.4 to suppress DY

ATLAS

• main background is ttbar
• DY Z contributes with mismisured jet as MET

Backup slides

Alberto Mariotti (VUB) 03-09-2015 Goldstini vs ATLAS excess

• pt_lep1,2 > 20
• eta_lep < 2.4
• N_lep > 1
• pt_jet > 40, eta_j < 3.0
• N_jet > 1
• 81 < m_ll < 101
• bin on
• MET (100-200,200-300,>300)
• N_jet > 1, N_jet > 2

CMS