SUPERSYMMETRY: WHERE DO WE STAND?

Matthew Reece Harvard University At LHCP, Barcelona, May 16, 2013

WHY SUPERSYMMETRY?
Naturalness Gauge Dark

Coupling Unication

Matter

Recent experimental results make this look shakier than before....

(This is a review talk; apologies for omissions and idiosyncracies)

SUSY Spectrum, 1984 From Lawrence Halls talk at SavasFest

W boson near the top of the spectrum.... 1984 was a utopian year for SUSY. Times have changed!
Text
Over 3 decades of susy: seismic shifts!

NATURAL SUSY, 1984

125 GEV HIGGS AND SUSY

on mal per this g. mass ggs lecpercale d at mixggs recund 13]. corubcts,

125 GEV HIGGS AND SUSY

Very An Observation seems sane, but heavy enough that many models dont.
For mt . 5 TeV, a Higgs mass of mh 125 GeV t t places strong constraints on tan and the stop parameters. Although we will use FeynHiggs for all the plots in h h + MSSM: this section, it is useful to keep in mind the approximate h h one-loop formula for the Higgs mass,
2 2 m2 = m h Z c2

the Higgs couplings are SM-like. This limit also avoids constraints from direct searches for H/A [2123]. For tan we will set a benchmark value of 30 and consider a range of values in some cases.

II.

IMPLICATIONS FOR WEAK-SCALE MSSM PARAMETERS interesting! Light enough that SUSY still

3m4 t + 2 2 4 v

log

2 MS h m2 t

more: Haber, Hemping, Hoang, Ellis, Ridol, Zwirner, Casas, Espinosa, Quiros, Riotto, Carena, Wagner, Degrassi, Heinemeyer, Hollik, Slavich, Weiglein

2 Xt + 2 MS

1h

Haber, Hemping 91 2 Xt (1) 2 12MS

on mal per this g. mass ggs lecpercale d at mixggs recund 13]. corubcts,

125 GEV HIGGS AND SUSY

Very An Observation seems sane, but heavy enough that many models dont.
For mt . 5 TeV, a Higgs mass of mh 125 GeV t t places strong constraints on tan and the stop parameters. Although we will use FeynHiggs for all the plots in h h + MSSM: this section, it is useful to keep in mind the approximate h h one-loop formula for the Higgs mass,
2 2 m2 = m h Z c2

the Higgs couplings are SM-like. This limit also avoids constraints from direct searches for H/A [2123]. For tan we will set a benchmark value of 30 and consider a range of values in some cases.

II.

IMPLICATIONS FOR WEAK-SCALE MSSM PARAMETERS interesting! Light enough that SUSY still

3m4 t + 2 2 4 v

log

2 MS h m2 t

more: Haber, Hemping, Hoang, Ellis, Ridol, Zwirner, Casas, Espinosa, Quiros, Riotto, Carena, Wagner, Degrassi, Heinemeyer, Hollik, Slavich, Weiglein

Haber, Hemping 91 Tree-level bound: 90 GeV t

2 Xt + 2 MS

1h

2 Xt 2 12MS

(1)

on mal per this g. mass ggs lecpercale d at mixggs recund 13]. corubcts,

125 GEV HIGGS AND SUSY

Very An Observation seems sane, but heavy enough that many models dont.
For mt . 5 TeV, a Higgs mass of mh 125 GeV t t places strong constraints on tan and the stop parameters. Although we will use FeynHiggs for all the plots in h h + MSSM: this section, it is useful to keep in mind the approximate h h one-loop formula for the Higgs mass, mh = mZ c2 3m4 t + 2 2 4 v
t

the Higgs couplings are SM-like. This limit also avoids constraints from direct searches for H/A [2123]. For tan we will set a benchmark value of 30 and consider a range of values in some cases.

II.

IMPLICATIONS FOR WEAK-SCALE MSSM PARAMETERS interesting! Light enough that SUSY still

Logarithmic growth with stop mass 2 2 2

log
2 MS h m2 t 2 Xt + 2 MS

1h

more: Haber, Hemping, Hoang, Ellis, Ridol, Zwirner, Casas, Espinosa, Quiros, Riotto, Carena, Wagner, Degrassi, Heinemeyer, Hollik, Slavich, Weiglein

Haber, Hemping 91 2 Xt (1) 2 12MS

on mal per this g. mass ggs lecpercale d at mixggs recund 13]. corubcts,

125 GEV HIGGS AND SUSY

Very An Observation seems sane, but heavy enough that many models dont.
For mt . 5 TeV, a Higgs mass of mh 125 GeV t t places strong constraints on tan and the stop parameters. Although we will use FeynHiggs for all the plots in h h + MSSM: this section, it is useful to keep in mind the approximate Polynomial growth with h h one-loop formula for the Higgs mass, Xt, a mixing
II. IMPLICATIONS FOR WEAK-SCALE MSSM PARAMETERS interesting! Light enough that SUSY still

the Higgs couplings are SM-like. This limit also avoids constraints from direct searches for H/A [2123]. For tan we will set a benchmark value of 30 and consider a range of values in some cases.

between left- and right- handed stops.

2 2 m2 = m h Z c2

3m4 t + 2 2 4 v

log

2 MS h m2 t

more: Haber, Hemping, Hoang, Ellis, Ridol, Zwirner, Casas, Espinosa, Quiros, Riotto, Carena, Wagner, Degrassi, Heinemeyer, Hollik, Slavich, Weiglein

2 Xt + 2 MS

1h

Haber, Hemping 91 2 Xt (1) 2 12MS

a 125 GeV Higgs can be even larger [4244].

and corresponds to mh = 123 127 GeV, and d lines indicate the same range of Higgs masses mt = 172 174 GeV. (The central value in all 135 will always be mh = 125 GeV at mt = 173.2 om all this, we conclude that to be able to get 130 GeV, we must have

around the curve corresponds to mh =123-127 GeV. Finally, the dashed lines correspond to varying mt from 172-174.

high-scale SUSY
tan 3. 5

3.0 2.5 2.0

MS TeV

125

(2)

mh GeV

an absolute lower bound on tan just from the 120 lifting the ss measurement. We also nd that the Higgs cally ceases to depend on tan for tan beyond Higgs mass 115 for the rest of the paper we will take tan = 30 needs ~ 5 city. to 10 TeV 110 tan , the Higgs mass is then a function of Xt Shown in g. 2 are contours of constant mh vs scalar Xt . We see that for large M , we want S 105 Xt MS

1.5 1.0 0.5

masses

or few TeV trilinear 0.0 Higgs-stop-stop coupling

6 4 2 0 2 Xt TeV 4 6

2, 1.5, 5 or 3 10 3, 1 1.7 .5 20 MS TeV

50 (3) 100

ee that the smallest the A-terms and the SUSY.absolutely 6. Higgs mass as a function of MS , with Xt = 0. The FIG. 2. Contours of constant mh in the MS vs. Xt plane, be are n band is the output of FeynHiggs together with with its astan = 30 and mQ = mU . The solid/dashed lines and bands are as in g. 1. ated blue line GeV represents 1-loop gray renorXt | uncertainty. 1000 GeV, TheM 500 . (4) S zation group evolution in the Standard Model matched he MSSM atto Mexamine estimates S . The blue interesting the bands limits give in the plane of errors m the top mass in between 172plots and 174 GeV (darkerIII. IMPLICATIONS FOR THE SUSY alvarying stop masses. Shown g. 3 are of the BREAKING SCALE d) the renormalization scale between Here mt /2 and 2mt of and constant Xt in the m vs. mt 2Meade, 1 plane. t P. Draper, P. MR, D. Shih 11; similar work by many others ter band). s of Xt < 0 and Xt > 0 were chosen to satisfy

In the MSSM, a 125 GeV Higgs requires large quantum corrections, with multi-TeV SUSY-breaking parameters, reintroducing (part of) the hierarchy.

An Observation
t h h

NATURALNESS
t h t h + h h

Higgs potential quantum corrections h h to the mass2 term. Direct searches constrain them:
t

-2|H|2+|H|4: large

2 mHu

3 2 2 2 2 y m + m + | A | . log t t t t L R 8 2 TeV

n Fig. 1. Weve already observed that the one at left is problematic: its a nal line, so we dont want to include it when we compute a loop amplitude. In shows up as unpleasant s12...1 factors in the amplitudes were trying (n 1) t of, which we are currently removing by hand. e diagram with one gluon connected at one end is shown on the right in Fig. 1. Consider the diagrams in Fig. 1. Weve already observed that the one at left is problema at the other end. As a result, it has the structure:
4

Either the stop is light, or Higgs potential is nely-tuned.

Two stops (LH/RH), one sbottom (LH) should be below of an external so we dont want to include it when weal.) compute a loop am about 500 renormalization - 700 GeV (e.g. line, 1110.6926 Papucci et
1

THE MSSM IS UNNATURAL

In the MSSM, a 125 GeV Higgs mass requires heavy stops / large A-terms, but those directly undermine the naturalness argument for SUSY.
Higgs Mass vs. Fine Tuning
3000 2500
Suspect FeynHiggs

Lightest Stop Mass

3000

1000

2000

100

500 75

1500 1000 500

1000

500

Tuning contours (Hall/ 2500 Pinner/Ruderman 2000 1112.2703) for low-scale 1500 mediation, = 10 TeV.
Suspect FeynHiggs
2000 1500

2500

m t @GeVD

m t @GeVD

50 25 200

1000
500 750

1000

200

D mh
-2

0 -4

Xt m t
0

500 Always at least a factor of m 0 100 - 4 tuning. -2 0 2 4

100 300

t1

Xt m t

DICHOTOMY
Higgs at 125 GeV MSSM, tuned with heavy scalars

Beyond MSSM, natural

robust experimental connection

Stop search; Higgs sector

(rates, decays)

Models?
(NMSSM, D-terms, compositeness....)

Gluino search

Top-down theory

NATURAL SUSY

NATURAL SUSY
Have to complicate the MSSM in two ways: 1. Raise the Higgs mass to 125 GeV. Typically new tree-level interactions. 2. Explain lack of squark signals. Usually splitting 3 1st/2nd gen from third. Example: U(2) avor models (e.g.
1206.1327 by Barbieri, Buttazzo, Sala, Straub, less minimal avor violation)

or hide the decays, so all squarks can be light: e.g. R-parity violation (Barbier et al. review hep-ph/0406039, MFV RPV by Csaki, Grossman, Heidenreich), stealth supersymmetry (Fan, MR,
Ruderman )

125 GEV, NATURALLY

The Higgs mass could be raised to 125 GeV by beyondMSSM tree-level interactions (quartic terms). NMSSM / Fat Higgs / lambdaSUSY
W = SHu Hd + f (S )

New D-terms: SU(2) SU(2)

works best with lowscale compositeness: higher-dim operators around the corner? Look for more Higgses!

SU(2)

Z bosons at a few TeV?

DIRECT STOP LIMITS

2013 update: ATLAS and CMS are aggressively pursuing the direct signatures of naturalness. No hints so far.

TARGETING STOPS
NEXT STEPS
Probe the scalar nature through spin correlations or rapidity differences (Z. Han, A. Katz, D. Krohn, MR, 1205.5808)

Allow for asymmetric decays

(Graesser, Shelton 1212.4495)
dilep m (our cuts)
T2

t ! t t

l+jets m (ATLAS cuts)

T

# events / 5 GeV (LHC7 5 fb-1)

# events / 10 GeV (LHC7 5 fb )

102

102

tt ~ ~ tR (220) tB(0) ~ ~0 tR (220) th (0)

Dileptonic mT2
(Kilic/Tweedie 1211.6106)

10

-1

10

10-1 0 20 40 60 80 100 120 140

mT2 (GeV)

50

100

150

200

250

mT (GeV)

and more, for instance: Plehn et al 1102.0557 & 1205.2696; Bai et al handed stops decaying into either a et massless neutralino orKaplan a massless higgsino-like neu1203.4813; Alves al. bino-like 1205.5805; et al. 1205.5816, ....
FIG. 1: The distribution of dileptonic mT 2 (left) and semileptonic mT (right) for 220 GeV right background is also shown. Events have tralino. For comparison, the distribution for the SM tt

section, ensures that the quadratic divergence cancels, but there is still a r to the higgs mass. Given that other scalars have already been argued to

NATURALNESS AND we can use this correction to estimate the natural range for the mass of b GLUINOS m 3TeV.
bR

We need the stop to be relatively light for naturalness of a Finally, q L , tR also being relatively light scalars, suer from their own na light Higgs. But the stop is itself a scalar eld, and can get with mass corrections dominated by the diagrams in gure 4: quadratic corrections!
g t t t t g t t

Large corrections come from the gluino, which hence FIG. 4. Stop mass correction should be light (below about 1.5 TeV). As a color octet, the gluino has a large production cross section at the LHC.

GLUINOS
Gluino mass bounds are now above a TeV; e.g., 1.3 TeV if gluino decays through stops.
LSP mass [GeV]

LSP mass [GeV]

0 ~ ~ ~ g-g production, g t t 1
800 700 600 500 400 300
lu i no )

1000 900 800 700 600 500 400 300

0 ~ ~ ~ g-g production, g b b 1

CMS Preliminary s = 8 TeV Moriond 2013

Observed Observed -1 SUSY theory Expected
)= 2 SP (L (to p) m

SUS-12-024 0-lep ( ET +HT ) 19.4 fb-1 SUS-13-007 1-lep (n

jets

6) 19.4 fb-1

SUS-12-017 2-lep (SS+b) 10.5 fb -1 SUS-12-026 (MultiLepton) 9.2 fb -1

CMS Preliminary s = 8 TeV Moriond 2013

ly al rb fo n de d i

Observed Observed -1 SUSY theory Expected

(g

-m

n ki

em

ic at

200 100 0 500 600 700 800 900 1000 1100 1200 1300 1400 1500 gluino mass [GeV]

SUS-12-024 0-lep ( ET +HT ) 19.4 fb-1

200 100
SUS-12-028 0-lep ( T ) 11.7 fb-1

0 500 600 700 800 900 1000 1100 1200 1300 1400 1500 gluino mass [GeV]

NATURAL SUSY: SUMMARY

Requires

more complicated model-building: new Higgs interactions, possible avor problems / new avor structures predict signals -- look for them!

those

Standard

decay modes of stops, sbottoms, gluinos are being ruled out to uncomfortably high masses. Look for higgsinos! bounds, for now... models could alter decays enough to evade

RPV, stealth, other

Are

we complicating the models so much that theyre less appealing than tuning?

UNNATURAL SUSY

MSSM WITH LARGE A-TERMS

The least-tuned corner of the MSSM has large At. This doesnt happen in General Gauge Mediation, but can happen in extended models that add Yukawa mediation: new couplings of messengers to matter.
5000 3000 2000 1000 700 500 250

Mass HGeVL

------ ---- - - - - - - -- - - - - - - -- - - -- - - - - - - - --- -- - - - --- - - 0 + { c c g q t b

I.8 I.9 I.10 I.11 I.12 I.13 I.14 I.15 I.9' I.13' II.1 II.2 II.3 II.7

5000 3000 2000 1000 700 500 250

- -

100

- -

100

Evans/Shih 1303.0228: spectra of some models. Keep searching for stops and/or gluinos; slepton NLSPs.

I.8 I.9 I.10 I.11 I.12 I.13 I.14 I.15 I.9' I.13' II.1 II.2 II.3 II.7

Model

SEMI-SPLIT SUSY
Many models predict mgaugino
g2 mscalar . 2 16

Tuned EWSB. But: solves most of hierarchy problem (Planck down to ~100 TeV). Gauge coupling unication works. SUSY dark matter also possible. Helps avor/CP problems. Taken seriously early on by James Wells: hep-ph/0306127. Followed by Arkani-Hamed / Dimopoulos split SUSY, others....

(ignoring soft masses), the Higgsino threshold increases the wino and bino masses such that the gluino/wino ratio is reduced to roughly a factor of six. An interesting limit occurs if the Higgses are mildly sequestered from Whid such that Planck-suppressed couplings to supersymmetry breaking are absent, but the -term comes from Hu Hd W0 . In such a limit, the threshold correction suppresses the wino mass, and in fact at leading order in B/2

HIGGS MASS IN SPLIT MODELS

140

Higgs mass mh HGeVL

130

120
tan b = 50

tan b = 4

110
tan b = 2 tan b = 1

100

105

106

MscHGeVL

107

108

109

FIG. 2. Here the Higgs mass predicted as a function the scalar et masses and tan . Arkani-Hamed etwe al show 1212.6971; also see Acharya/Kane et al, of Arvanitaki al, Hall/Nomura

POTENTIAL SIGNALS
q g
0

q
g q
FIG. 10.

The gluino remains the best bet, possibly with a somewhat displaced vertex. Also, neutralino dark matter could give signals in direct or indirect detection experiments.
(9)

nteresting immediate observation: the fact that gluinos decay at all Arkani-Hamed et al 1212.6971 a decay can be quite long, with
c 10 m PeV m . (9)

q a decay can be quite long, with m 4 TeV 5 q c 10 5 m . FIG.PeV 10. mg

orders 5 of magnitude of the gluino mass m a 4few ll imply a scale 5within TeV q

WHY THE HIGH SCALE?

Why couldnt the whole spectrum have been lighter, both semi-split and natural? (1TeV scalars, 1 GeV gauginos) One possibility: moduli, scalar elds interacting with m m3/2 and gravitational strength, tend to have mass m3 decay width 2
MPl

Coherent moduli oscillations ruin cosmology unless they decay early enough for BBN: But 100 TeV soft scalar masses imply tuned EWSB!
Treheat p MPl 10 MeV ) m 100 TeV

NONTHERMAL DARK MATTER

Considering moduli cosmology motivates pairing semisplit SUSY with nonthermal dark matter generated through moduli decay.
see: Moroi/Randall hep-ph/9906527; J. Kaplan hep-ph/0601262; Gelmini/ Gondolo hep-ph/0602230, Acharya/Kane/Kuik 1006.3272, others....

For given h v i, DM abundance is enhanced by a factor of Tfreezeout/TRH. Ideal for light wino DM, with large annihilation rate.

IN WINO VERITAS?
Both thermal and nonthermal wino DM are in some trouble from observations of the gamma-ray sky:
1.00 0.50 0.20
WDM h2

h=1

102 1 0.01 BBN 10-4

3000

TR @GeVD

0.10 0.05 0.02 0.01

h2

Wino thermal relic HESS line HEinastoL Fermi line HNFWL

- Fermi dwarf 4 yrs - Hooper et. al. GC HNFWL - Hooper et. al. GC HEinastoL
150 200 300

100

m @GeVD
500 700

1000

1500 2000

10-6 100

W0

m @GeVD W0

500

1000

Constraints on the relic abundance of Wino component in dark matter. The red solid curve is the thermal relic of ulated in [12]. The rest curves and points are from different indirect detection searches: purple squares: Fermi line een lled circles: HESS line search; purple dash-dotted line: Fermi photon continuum search in the satellite galaxies Way; green dotted (blue dashed): continuum constraints from galactic center assuming NFW (Einasto) proles.

Hard not to overproduce DM without even heavier moduli, RPV, or more complex cosmology.
Preliminary work in progress, J. Fan and MR.

plications for the cosmological history

WHATS NEXT?
If

SUSY is right, could well be beyond the MSSM. If SUSY is natural, it must be beyond MSSM. to keep pushing stop and gluino searches, also broadening to RPV, etc, to really rule out naturalness. split SUSY: scalars at ~100 to ~1000 TeV? Now some tension with dark matter / moduli constraints. Add RPV? looking for hard-to-nd but theoretically motivated options: displaced gluinos, light higgsino, pure higgsino DM.... hoping for more surprises!

Important

Mildly

Keep

Still