Aspects of Higgs portal DM models: light scalar singlet and intense - - PowerPoint PPT Presentation

Aspects of Higgs portal DM models: light scalar singlet and intense

• ray from hidden vector

Thomas Hambye

• Univ. of Brussels (ULB), Belgium

Firenze, 19/05/2010

γ

Higgs portal interaction

where is

L ∋ λ H†Hφ†φ

φ

• r a messenger between the SM and DM

the DM particle simplest way to couple the SM to a hidden sector where DM could lay

Part I. DAMA and/or CoGeNT: scalar DM??

in collab. with S. Andreas, C. Arina, F.-S. Ling and M. Tytgat

DAMA and/or CoGeNT ???

DAMA: CoGeNT:

could have nothing to do with DM but makes sense to look for simplest possible DM explanations of them

Possible DM annihilations to SM particles

if : only mDM ∼ 10 GeV DM DM → f ¯ f (f = b, c, s, d, u, τ, µ, e, νe,µ,τ) in 3 ways:

=

Z exchange: h exchange: BSM particle exchange e.g. squarks loops, ...

Z0

DM DM f ¯ f > > DM DM f ¯ f > > h excluded by LEP (invisible width) to be analyzed

Z0

Predictivity of the Higgs exchange scenario

Annihilation cross section:

DM DM h f ¯ f

DM DM h N N

Cross section on Nucleon: the ratio of cross sections depends only on ! mDM Yf λ λ

ghNN

if one fixes the Nucleon cross section to reproduce the DAMA and/or CoGeNT the relic density is fixed R ≡ σ(DM DM → f ¯ f) vrel σ(DM N → DM N) = fct′′(mDM, Yf, ghNN)

σ(DM DM → f ¯ f) vrel ∝ λ2 1 (s − m2

h)2 Y 2 f · fct(mDM) ∼ λ2 1

m4

h

Y 2

f · fct(mDM)

see also Burgess, Pospelov, ter Veldhuis 01’

σ(DM N → DM N) ∝ λ2 1 (t − m2

h)2 g2 hNN · fct′(mDM) ∼ λ2 1

m4

h

g2

hNN · fct′(mDM)

The simplest DM model: a scalar singlet

DM= a real scalar singlet S:

Mc Donald 94’, Burgess, Pospelov, ter Veldhuis 01’, Patt, Wilczek 06’; Barger et al 08’,...

L 1

2µSµS − 1 2µ2

S S2 − S

4 S4 −L H†H S2

assuming a symmetry

for S stability Z2, S ↔ −S ,

(SS → ¯ f f)vrel = nc 2

L

• m2

f

m4

hm3 S

(m2

S −m2 f)3/2

(SN → SN) = 2

L

• µ2

r

m4

hm2 S

f 2m2

N

R ≡

f

(SS → ¯ f f)vrel (SN → SN) =

f

ncm2

f

f 2m2

Nµ2 r

(m2

S −m2 f)3/2

mS λ = λLv

fmN ≡ N|

• q

mq¯ qq|N = ghNNv

m2

S = µ2 S + λLv2

! " # $ % & ' ( ) * "!

!"

"!

!

"!

"

"!

#

"!

$

"!

%

"!

&

+,-.-/01-2 3

4-5-!6"% 4-5-!6$! 4-5-!6''

Results for the scalar singlet

ratio predicted for f=0.3 central value ratio required to match both DAMA and

0.094 < ΩDMh2 < 0.129

intriguing result R ∼ m2

S, ...

• S. Andreas, T.H., M. Tytgat 08’

Results for the scalar singlet

5 10 15 20 1042 1041 1040 1039 mS GeV Σ0

n cm2

DAMA value of Nucleon cross section obtained once one requires 0.094 < ΩDMh2 < 0.129

• S. Andreas, C. Arina, F.-S. Ling, T.H., M. Tytgat 10’

0.2 < f < 0.4

with

Results for the scalar singlet

5 10 15 20 1042 1041 1040 1039 mS GeV Σ0

n cm2

DAMA value of Nucleon cross section obtained once one requires 0.094 < ΩDMh2 < 0.129 CoGeNT

0.2 < f < 0.4

with

Results for the scalar singlet

5 10 15 20 1042 1041 1040 1039 mS GeV Σ0

n cm2

value of Nucleon cross section obtained once one requires CoGeNT DAMA

0.094 < ΩDMh2 < 0.129

less channeling radioactivity contamination

0.2 < f < 0.4

with

Results for the scalar singlet

5 10 15 20 1042 1041 1040 1039 mS GeV Σ0

n cm2

DAMA value of Nucleon cross section obtained once one requires 0.094 < ΩDMh2 < 0.129 CoGeNT CDMS 2 events 1σ CDMS Si Xenon 10 Xenon 10 High scintil. effic. Low scintil. effic.

0.2 < f < 0.4

with

Issues

Theoretically: -value of f? Higgs exchange scenario requires 0.2 < f < 0.6

• Experimentally: -DAMA: channeling, ...
• CoGeNT: radioactivity, ...
• Xenon 10 and 100: scintillation efficiency, ...
• CRESST?
• fairly large value of is required:

|λL| ∼ 0.2 λL m2

S = µ2 S + λLv2

O(100 GeV)2 O(10 GeV)2 tuning at the % level

• why a scalar around 5-10 GeV?

ΩDM/ΩB ?

not apparent in model independant analysis such as in Fitzpatrick, Hooper, Zurek 10’

Consequences for Higgs invisible decay width

0.95 1 0.95 1 0.95 1 0.95 1 0.95 1 0.95 1

5 10 15 20 1042 1041 1040 1039 mS GeV Σ0

n cm2

0.1 0.9 0.1 0.9 0.1 0.9 0.1 0.9 0.1 0.9 0.1 0.9 0.1 0.9 0.1 0.9 0.1 0.9

5 10 15 20 1042 1041 1040 1039 mS GeV Σ0

n cm2

mH = 180 GeV mH = 120 GeV

98 % < BR(H → DMDM) < 99.5 % 60 % < BR(H → DMDM) < 90 %

DAMA and CoGeNt lead to

BR distinguishable at LHC

Indirect detection

Neutrino flux from DM annihilation in the Sun: rays from the galactic center: γ

µS (GeV)

mS (GeV) and log10 φµ [km−2 yr−1] SuperK sensitivity

EGRET NFW 0.1 0.2 0.5 1.0 2.0 5.0 10.0 20.0 1109 5109 1108 5108 1107 5107 1106 Photon energy E GeV E2 d dE GeV cm^2 s^1

hements calculés au

• ur −2

˚< b < 2 ˚et et −5 ˚< l < 5 ˚ .

FERMI NFW 0.2 0.5 1.0 2.0 5.0 10.0 1109 5109 1108 5108 1107 5107 1106 Photon energy EGeV E2 d dE GeV cm^2 s^1

• n, r < 0.5

˚ .

• S. Andreas, T.H., M. Tytgat 08’

courtesy of M. Casier and M. Tytgat

• S. Andreas, M. Tytgat, Q. Swillens 08’

Savage, Gelmini, Gondolo, Freese 08’; Savage, Freese, Gondolo, Spolyar 09’, ... Feng, Kumar, Strigari 08’

¯ p, De :

Bottino, Donato, Fornengo, Scopel 08’; Nezri, Tytgat, Vertongen 09’

Higgs exchange scenario in other scalar models

the scalar DM particle doesn’t necessarily need to be a weak singlet also applies to inert doublet model

DM = one neutral component

• f a second Higgs doublet

H0

with other neutral component heavier than (LEP invis. Z decay width) ∼ 80 GeV

A0

same predictions as for the singlet

Z → H0A0

Fermion DM with Higgs exchange

doesn’t work!

L ¯

(i / −m0)− Y √ 2 ¯ h Example: a Dirac fermion:

( ¯ → ¯ f f)vrel = nc Y 2

• 16

m2

f v2 rel

v2m4

h

(m2

−m2 f)3/2

m

(N → N) = Y 2

• 2

µ2

r

v2m4

h

f 2m2

N

R ≡ f ( ¯ → ¯ f f)vrel (N → N) = f ncm2

f

f 2m2

Nµ2 r

v2

rel

8 (m2

−m2 f)3/2

m

Annihilation: Cross section on N: extra v2

rel

m2

DM

v2

suppression P-wave and heli- city suppressed much smaller predicted DAMA and/or CoGeNT can be reproduced but give a relic abundance way too large

Part II. Intense -ray lines from hidden vector DM

γ

in collab. with C. Arina, A. Ibarra and C. Weniger

Monochromatic -ray lines: a smoking gun for DM

annihilation leads to a monochromatic -ray line

γ

γ

DM DM → γγ , γZ (not expected in astrophysics background) e.g. obtained at one loop level rather suppressed

e.g. needs for large boost factor or a TeV DM mass

But what about a -ray line from DM decay?????

γ

has been considered from gravitino decay through R-parity violation

Buchmuller, Covi, Hamagushi, Ibarra, Tran 07’; Ibarra, Tran 07’; Ishiwata, Matsumoto, Moroi 08’; Buchmuller, Ibarra, Shindou, Takayama, Tran 09’; Choi, Lopez-Fogliani, Munoz, de Austri 09’ Boudjema, Semenov, Temes 05’; Bergstrom, Ullio, 97’, 98’;Bern, Gondolo, Perelstein 97’; Bergstrom, Bringmann, Eriksson, Gustafsson 04’, 05’; Jackson, Servant, Shaughnessy, Tait, Taoso 09’, ...

• ne tree level exception: Dudas, Mambrini, Pokorski,

Romagnoni 09‘

A scenario for large -ray lines through DM decays

If DM stability results from an accidental symmetry (as proton in SM)

i.e. doesn’t result from an ad-hoc symmetry

• r from a gauge symmetry remnant subgroup

we expect higher dimensional operators destabilizing the DM to be generated by higher scale physics a dim-5 operator leads to τDM << τUniverse even if Λ ∼ MP lanck but a dim-6 operator leads to a

• ray flux of order the expe-

rimental sensitivity if

γ

Λ ∼ MGUT DM model based on accidental symmetry decaying to from dim-6 operator

γ

Eichler; Nardi, Sannino, Strumia; Chen, Takahashi,

Yanagida; Arvanitaki, Dimopoulos et al.; Bae, Kyae; Hamagushi, Shirai, Yanagida; ...

as for other cosmic rays:

• C. Arina, T.H., A. Ibarra, C. Weniger 09’

γ

Hidden vector DM

• simple viable spin-1 DM model

non-abelian global symmetry

• the stability can be “understood” only from the low-energy

point of view as for the proton in the SM

• no possible dim-5 operators but dim-6 ones which

based on the existence of a accidental custodial symmetry:

all leads to a -ray line

γ

Custodial symmetry DM stability

simplest example: a gauged SU(2) + a scalar doublet φ gets a vev

φ vφ φ =

• φ+

(η + ia0 + vφ)/ √ 2

spectrum: - 3 degenerate massive gauge bosons V : i

• one real scalar :

η mV = gφvφ 2

This lagrangian has a custodial symmetry SU(2) or equivalently a SO(3) : triplet and singlet

η =

the 3 V are stable! forbidden

i

C C

Vi → ηη, ... L = −1 4F µνaF a

µν + (Dµφ)†(Dµφ) − µ2 φφ†φ − λφ(φ†φ)2

mη =

• 2λφ vφ

(V µ

1 , V µ 2 , V µ 3 ) =

T.H. 08’

Hidden sector through the Higgs portal

L = LSM + LHidden Sector + LHiggs portal LHiggs portal = −λmφ†φH†H ∋ −λmvφ v h η

mixing

h − η

SU(2)HS

LHidden Sector = −1 4F µνaF a

µν + (Dµφ)†(Dµφ) − µ2 φφ†φ − λφ(φ†φ)2

doesn’t spoil the stability of the V µ

i

Dimension-6 operators breaking the custodial symmetry

(A) 1 Λ2 Dµφ†φ DµH†H , (B) 1 Λ2 Dµφ†φ H†DµH , (C) 1 Λ2 Dµφ†Dνφ F µνY (D) 1 Λ2 φ†F a

µν

τ a 2 φF µνY .

all give 2-body decay to or γh γη

Benchmark ηη hη hh γη Zη γh Zh 1

• 0.09
• 0.04

0.02 0.65 0.20 2

• 0.04

0.62 0.002 0.003 0.15 0.18 3

• 0.04

0.80 3 × 10−6 0.002 0.0003 0.16 Benchmark Zη γη Zh γh 1 0.19 0.81 2 0.22 0.78 3 0.23 0.77 4 0.028 0.79 0.041 0.14 × Benchmark Zη Zh γη W +W − ν¯ ν e+e− u¯ u d ¯ d 1 0.01 0.005 0.04 0.02 0.09 0.39 0.29 0.15 2 0.019 0.004 0.036 0.014 0.072 0.35 0.39 0.12 3 0.22 0.0002 0.73 0.0005 0.003 0.016 0.018 0.005

examples of branching ratios:

• perator A & B
• perator C
• perator D
• C. Arina, T.H., A. Ibarra, C. Weniger 09’

Benchmark MA gφ vφ Mη Mh sin β 1 300 GeV 0.55 1090 GeV 30 GeV 150 GeV ≈ 0 2 600 GeV 0.6 2000 GeV 30 GeV 120 GeV ≈ 0 3 14 TeV 12 2333 GeV 500 GeV 145 GeV ≈ 0 4 1550 GeV 2.1 1457 GeV 1245 GeV 153 GeV 0.25

Flux of monochromatic -rays

• 0.1

1 10 100 1000 104 109 108 107 106 105 Energy GeV E2dJdE cm2str s1GeV

• Prel. Fermi b10
• Prel. Fermi EGBG

HESS eeΓ

• 0.1

1 10 100 1000 104 1108 5108 1107 5107 1106 5106 1105 Energy GeV E2dJdE cm2str s1GeV

• 0.1

1 10 100 1000 104 1108 5108 1107 5107 1106 5106 1105 Energy GeV E2dJdE cm2str s1GeV

γ

• perator A & B
• perator C
• perator A & B
• 0.1

1 10 100 1000 104 1108 5108 1107 5107 1106 5106 1105 Energy GeV E2dJdE cm2str s1GeV

• perator D

mDM = 300 GeV mDM = 600 GeV mDM = 600 GeV mDM = 1.5 TeV Λ = 2.9 · 1015 GeV Λ = 3.7 · 1015 GeV Λ = 1.2 · 1016 GeV Λ = 1.5 · 1016 GeV ← τDM = 1.6 · 1027 sec

n 0 ≤ l ≤ 360◦, 10◦ ≤ |b| ≤ 90◦,

• C. Arina, T.H., A. Ibarra, C. Weniger 09’

Backup

Relic density

in thermal equilibrium with SM thermal bath

• annihilation freeze out (WIMP)

with at least one SM part. in final state: to two real : η

T mV : T < mV : neq.

V

∼ e−mV /T

• i

i

Ai Ai η, h

Ai η Ai Ai η

Vi Vi Vi Vi Vi Vi Vi η η η η η η η

Ai η Ai Ai η

Vi Vi Vi

i i

Ai Ai η, h

Vi Vi Vi Vi η η η η η

Ai Ai η, h W, Z W, Z

i i

η, h f ¯ f

h, η h, η h, η h, η h, η h, η h, η Vi Vi Vi Vi

η h :

with due to coupling

Vi η :

with due to coupling

λm gφ

with subsequent decay of to

SM particles via mixing

η h − η

gφ (+λφ) λm, gφ, ...

V µ

1,2,3

Relic density: additional new type of contribution

non abelian trilinear gauge couplings:

F a

µνF µνa ∋ εijk∂µAiν(Aµ j Aν k − Aν j Aµ k)

do not lead to any V decay even if trilinear (carries 3 indices)

i

=

but induces two DM to one DM particle annihilation

i j

Ak

Ai A Aj

j

η

Vi Vi Vj Vj Vk Vk Vk Vk η, h η, h

no dramatic effect for the freeze out (same order as other diagrams) from the Z case

2

=

Small Higgs portal regime

λm 10−3

depend only on with

ViVi → ηη , ViVj → Vkη dominant gφ , vφ, λφ

λφ

if also small:

σannih. ∼ g4

φ

m2

V

∼ g2

φ

v2

φ

mV ∝ g2

φ

(∝ v2

φ) mV (GeV)

gφ

λφ = 10−4

∼ 10−7

(but larger than to have thermalization with the SM bath)

mV = gφvφ 2 , mη ≃

• 2λφ vφ

1 MeV mDM 25 TeV

Small Higgs portal regime

λm 10−3

depend only on with

ViVi → ηη , ViVj → Vkη dominant gφ , vφ, λφ

λφ

if large:

∼ 10−7

(but larger than to have thermalization with the SM bath)

mV = gφvφ 2 , mη ≃

• 2λφ vφ

mV (GeV)

gφ

λφ = 10−1

Large Higgs portal regime

λm 10−3

large mixing

η − h

tor - SM mixing large hidden sec- can lead to the right even for maximal mixing

ΩDM

mV (GeV)

mη (GeV)

production at LHC of just as for the Higgs in the SM but

η

with possibly a larger mass T parameter constraint:

if mη = mh mh = mη < 154 GeV (3σ) mh = 120 GeV mη < ∼ 240 GeV (3σ) if

• r larger if non

maximal mixing

Hidden vector: direct detection

N N

Vi Vi

Log(ViN → ViN) (cm2) mV (GeV)

can saturate the experimental bound easily

η h ×

• 1

5 10 50 100 500 1000 0.20 0.10 0.05 0.02 Energy GeV eee

• 1

10 100 1000 0.050 0.020 0.010 0.005 Energy GeV E3e cm2str s1GeV2

• 0.1

1 10 100 1000 104 1108 5108 1107 5107 1106 5106 1105 Energy GeV E2dJdE cm2str s1GeV

• 1

2 5 10 20 50 100 106 105 104 0.001 0.01

• kin. Energy GeV

p p

• 1

5 10 50 100 500 1000 0.20 0.10 0.05 0.02 Energy GeV eee

• 1

10 100 1000 0.050 0.020 0.010 0.005 Energy GeV E3e cm2str s1GeV2

• 0.1

1 10 100 1000 104 1108 5108 1107 5107 1106 5106 1105 Energy GeV E2dJdE cm2str s1GeV

• 1

2 5 10 20 50 100 106 105 104 0.001 0.01

• kin. Energy GeV

p p

• perator A & B

mDM = 300 GeV Λ = 2.9 · 1015 GeV

• perator C

mDM = 1.5 TeV Λ = 1.2 · 1016 GeV

e+ e+ ¯ p ¯ p e+ + e− e+ + e− γ γ

Hidden vector: cosmic ray fluxes

Pamela: can we reproduce the positron spectrum?

from annihilation: yes

(ViVi → ηη dominant)

thanks to Gilles Vertongen

e+ e− + e+

• PAMELA 08

CAPRICE 98 CAPRICE 94 HEAT 00 AMS 98

MDM 500 GeV Boost 4.102 1 101 102 103 104 102 101 1

E GeV eee

mV = 500 GeV, mη = 1 GeV

η → µ+µ−

as in Arkani-Hamed et al. light mediator scenario

Pamela: can we get a large enough Sommerfeld boost?

mediated between 2 V is attractive:

η

i

.....

η η η η η η η

Vi Vi Vi Vi Vi Vi .....

.....

Vi Vi

Sommerfeld boost:

10-1 1 10 102 103 / 10-1 1 10 102 103 / = MDM /(MV/) Attractive potential strong TeV weak 1.3 2 3 5 10 30 100 1000

*

where we are for example with: apparently the boost has just the right size

(in agreement with which fixes the Som- merfeld coupling)

ΩDM

Cirelli, Strumia, Tamburini ‘07 mV = 500 GeV, mη = 1 GeV

but stability problem:

mη

Vi Vi

η η

δm2

η ∼

What about the non-perturbative regime of this model?

SU(2) confines automatically if

Hidden Sect.

perturbative dynamical

ΛSU(2) >> vφ

breaking scale scale

but the custodial symmetry remains exact in this case too confines: boundstates are eigenstates of the custodial sym.:

φ

‘t Hooft ‘98

• scalar state: singlet of SO(3) expected the lightest
• “charged” vector state:
• “neutral” vector state:

S ≡ φ†φ

V +

µ ≡ φ†Dµ ˜

φ V −

µ ≡ ˜

φ†Dµφ V 0

µ ≡ φ†Dµφ − ˜

φ†Dµ ˜ φ √ 2 }SO(3) triplet

stable DM candidates!

T.H., M. Tytgat, arXiv:0907.1007

Relic density in the confined regime

strongly interactive massive particle (SIMP) annihilation cross section cannot be calculated perturbatively

Vi Vi

S S

Vi

Vi

h S

...

if mixing is

S − h

large (for large )

+

( )

expected do- λm minant channel:

σannih. ∼ A Λ2

SU(2)

A = 10 − 50

confining non-abelian hidden sector coupled to the SM through the Higgs portal: perfectly viable DM candidate

mDM ≃ 20 − 120 TeV

Expected spectrum (in a similar case)

208.0 210.0 212.0 214.0 216.0 218.0 T*/GeV 0.0 0.2 0.4 0.6 0.8 1.0

m/g3

2

64

3

48

3

32

3

m*H = 120 GeV vector scalar

Kajantie, Laine. Rummukainen, Shaposhnikov ‘96

vector states e.g. expected heavier than scalar ones:

Possible effects on Electroweak Symmetry Breaking

contribution of the vev of the hidden scalar to the Higgs mass term:

LHiggs portal = −λmφ†φH†H

∋ −λmv2

φH†H

gives a contribution to the Higgs vev: gives a hint for the versus WIMP coincidence

mDM v v2 ∝ λm λH v2

φ ∝ m2 DM

see also T.H, M. Tytgat, arXiv 0707.0633, (PLB 659)