The Dark Matter Hunter Guide

Testing weak scale DM

from the sky from the underworld from CERN

DM

DM

SM

SM

indirect Hand cosmology?L

?

DM

SM

DM

SM

direct

?

SM

SM

DM

DM

collider

?

DM at colliders

DM at colliders, what signal?

Safe concrete expectations are well-known and trivial:

• DM is probably stable thanks to a Z2 symmetry: DM produced in pairs.

• DM behaves like ν: DM carries away missing transverse momentum 6pT .

• Maybe DM comes alone giving /pT j and /pTγ from initial state radiation.

• Maybe DM comes with other particles giving better signals.

It would be wise to stop here.

Anti-pedagogical presentation

The rest is a list of disconnected speculations and their signals

DM

...

Ultra-light

Particle

Ultra-heavy

...

...

Asymmetric

Rehating

Thermal relic

Freeze-in

Freeze-out & decay

...

...

Scalar singlet

Kalzua-Klein

SUSY

Minimal DM

Mirror world

...

...

Gravitino

Q-balls

Neutralino

Sneutrino

Axino

...

...

Wino

Bino

Higgsino

...

? ? ? ? ?

Like reading a phone book/cleaning stains from a jaguar...

Politically Correct Dark Matter

According to the ideology that dominated past decades, the Higgs mass has a

hierarchy problem solved by many new particles at the weak scale.

The most popular solutions are the supersymmetric sparticles. SUSY ruins B,L

conservations, so theorists add a new Z2 symmetry (R-parity, KK parity):

SM→ SM new→ −new

which makes the lightest new particle stable: DM candidate if neutral!

WIMP miracle: the thermal abundance of a weak particle can reproduce ΩDM!!

“Neutralino” is often used as a synonymous of “Dark Matter”!!!

Big signals! DM is the last step of a decay chain that starts with g/q production

g → gq → g`χ→ g``N

Many authors proposed kinematical variables to reconstruct the masses.

But next LHC was turned on and nothing like this has been seen so far.

Dark Matter in the CMSSM

ΩDM suggested neutralino annihilations via sleptons up to a few crazy regions.

The ‘bulk’ region got excluded leaving the tail, the nose: only special mech-

anisms can give ΩDM: ˜ co-annihilations, H,A resonance, h,H,A at large

tanβ, t co-annihilations, well-tempered B/H (excluded by Xenon for µ > 0), h

resonance (excluded by LHC, M3 > 3mh). Like dissecting the spherical cow.

Well-tempered neutralinos

If M <∼ TeV winos and higgsinos annihilate too much, binos annihilate too little.Like in the 3 bear fable, the observed thermal ΩDM is obtained by mixing them.

Wino/bino (M1 ' M2) is not detectable. Wino/higgsino (M2 ' µ) is lessplausible. Wino/bino (M1 ' |µ|, green strip) has been disfavoured by Xenon(dark region) if µ > 0; cancellations are possible for µ < 0

10-46

10-45

10-45

10-44

100 100030 300

100

1000

30

300

Μ in GeV

M1

inGe

V

well tempered binohiggsino, tan Β = 10

mDM =m

h 2m

DM =mZ 2

Stop co-annihilation

A light stop can lead to correct thermal ΩDM if ∆M = Mt −MDM ≈ 30 GeV

σvcosmo =3

8e−2∆M/T

︸ ︷︷ ︸co−annihilation

× 7 g43

216πm2t1︸ ︷︷ ︸

σ(tt∗→gg)v

≈ 2.3× 10−26 cm3s−1

0 500 1000 1500 20000

10

20

30

40

50

Neutralino DM mass MDM in GeV

DM

=m

t 1-

MD

Min

GeV

CMS razorhadronic onlyour analysis

DMlin

e

t 1®

NbΝ

t 1®

Nt

t 1®

Nc

t 1®

NbW

t1

LSP

150 200 250 300 350 400 450 5000

100

200

300

400

Lightest stop mass in GeV

Neu

tral

ino

mas

sin

GeV

Good sensitivity at LHC thanks to big σ(pp→ t+ t∗+ ISR jets) from QCD.

SUSY, SUSY, SUSY

One-letter extensions of the MSSM:

AMSSM, BMSSM, CMSSM, DMSSM, EMSSM, FMSSM, GMSSM, HMSSM,

IMSSM, KMSSM, MMSSM, NMSSM, OMSSM, PMSSM, QMSSM,

RMSSM, SMSSM, TMSSM, UMSSM, VMSSM, XMSSM, YMSSM, ZMSSM

According to google, the JMSSM and the LMSSM have not yet been proposed.

SUSY covers a lot of possibilities... The LSP could decay into the graviton.

Charged tracks and secondary vertices if the LSP is charged or coloured and τ

is slow enough. If very slow, LSP could decay while the beam is off...

Etc etc... Furthermore, natural scenarios alternative to SUSY are sometimes

considered, especially universal extra dimensions and Little Higgs.

Etc etc... In models with a light dark sector, long dark decay chains can give

things like ‘muon jets’ when/if dark particles decay into light SM particles

Is nature natural?

The good possibility is in trouble

Present data indicate do not confirm the usual naturalness ideology

The bad possibility is that the Higgs is light due to ant**pic reasons. Then,

one would expect that H is the only light scalar, so weak-scale DM must be a

fermion. This lead to consider ‘split SUSY’ i.e. neutralino/wino/higgsino DM.

The ugly possibility is that quadratic divergences should be ignored. Then

the SM satisfies ‘finite naturalness’. To preserve this property, new physics

motivated by data, such as DM, must be not much above the weak scale.

Consider: scalar/fermion DM and DM with/without SM gauge interactions.

Minimal Dark MatterAdd only one electroweak multiplet containing a neutral DM particle.Assume that it has only gauge interactions. In one case it is automatic: afermion weak 5-plet cannot couple to SM particles, so DM stability arises asan accidental symmetry, just like baryon number within the SM.

The neutral component gets lighter by ≈ 166 MeV. Compute annihilation crosssections and predict the DM mass: TeV. Finite naturalness arises at two loops.

Quantum numbers DM could DM mass mDM± −mDM Finite naturalness σSI inSU(2)L U(1)Y Spin decay into in TeV in MeV bound in TeV 10−46 cm2

2 1/2 0 EL 0.54 350 0.4×√

∆ (2.3± 0.3) 10−2

2 1/2 1/2 EH 1.1 341 1.9×√

∆ (2.5± 0.8) 10−2

3 0 0 HH∗ 2.0→ 2.5 166 0.22×√

∆ 0.60± 0.043 0 1/2 LH 2.4→ 2.7 166 1.0×

√∆ 0.60± 0.04

3 1 0 HH,LL 1.6→ ? 540 0.22×√

∆ 0.06± 0.023 1 1/2 LH 1.9→ ? 526 1.0×

√∆ 0.06± 0.02

4 1/2 0 HHH∗ 2.4→ ? 353 0.14×√

∆ 1.7± 0.14 1/2 1/2 (LHH∗) 2.4→ ? 347 0.6×

√∆ 1.7± 0.1

4 3/2 0 HHH 2.9→ ? 729 0.14×√

∆ 0.08± 0.044 3/2 1/2 (LHH) 2.6→ ? 712 0.6×

√∆ 0.08± 0.04

5 0 0 (HHH∗H∗) 5.0→ 9.4 166 0.10×√

∆ 5.4± 0.45 0 1/2 stable 4.4→ 10 166 0.4×

√∆ 5.4± 0.4

7 0 0 stable 8→ 25 166 0.06×√

∆ 22± 2

Sommerfeld correctionsNon-relativistic DM DM annihilations are enhanced if M >∼MW/α

2F

3F

3S

5F

1 102 30.4 40.6 60.8 80.00

0.05

0.10

0.15

0.20

DM mass in TeV

WD

Mh2

Wino/MDM searches

Γ(DM± → DM0π±) = (n2 − 1)/44 cm = 0.977 Γ(DM±) ∆M = 166 MeVCo

llision

Detector1 cm

B Å Tesla

Slow DM+ straigh track RelativisticcurvedΠ+ track

Trigger on initial state radiation and missing energy, LHC better than LEP

Phenomenology: EWinos

Heavy Higgsinos:

Bino LSP : c±c0 → Wh+MET c+c- → WW+MET

Wino LSP: Dm~170 MeV → 10cm stubs (trig. on ISR+MET )

Singlet Scalar DM

L = LSM +(∂µS)2

2− m

2S

2S2 − λHSS2|H|2 − λS

4S4

102 103

10-46

10-45

10-44

10-43

DM mass in GeV

Σ SIin

cm2

scalar DM singlet

Excluded by Xenon 2012

DM thermal abundance

Finite naturalness

D =1

D »100

Λ =1

σSI =λ2HSm

4Nf

2

πM2M4h

δm2 = −2λ2HS

(4π)2M2(ln

M2

µ2− 1) < ∆×M2

h

Singlet Fermion DM

L = LSM +(∂µS)2

2+ ψi/∂ψ−m

2S

2S2− λS

4S4−λHSS2|H|2 +

y

2Sψψ+

Mψ

2ψψ+ h.c.

Communicate via S. Thermal annihilations are p-wave, so σSI is often too large.

102 103

10-47

10-46

10-45

10-44

10-43

10-42

DM Mass in GeV

Σ SIinc

m2

Fermion DM singlet HmS=300 GeVL

Excluded by Xenon 2012D=1

D=100

Finite Naturalness

DM thermal

abundance

σSI =y2 sin2 2α

8π

m4Nf

2

v2

(1

M2h

− 1

M2S

)2

δm2 =6y2 sin2α

(4π)2M2(ln

M2

µ2− 1

3)

DM and Higgs decays

Consider scalar (S) or fermion (F ) or vector (V ) DM coupled to the Higgs as

rS2m2

S

VhSS + rf

mf

Vhff + rV

2m2V

VhVµVµ

where r = 1 if DM gets mass only from 〈h〉 = V . Invisible Higgs decays are agreat signal for M < Mh/2: BRinv<∼19− 28% at 95% CL constrains σSI

0.0 0.2 0.4 0.6 0.8 1.0

0

5

10

15

20

Invisible Higgs BR

DΧ

2

SM + BRinv

SM+BRinv+

+BRgg+BRΓΓ

1Σ

2Σ

3Σ

4ΣXenon100

0 10 20 30 40 50 60 70

10-12

10-11

10-10

10-9

10-8

10-7

DM mass in GeV

Spin

-In

dep

enden

tΣ

SIin

pb

scalar

fermion

vector

Effective operator description?

Assume that the physics that couples DM to quarks is so heavy that can be

integrated out leaving effective operators of the form

1

Λ2[ΨDMγµΨDM][ΨSMγ

µΨSM]

General framework where everything is computed in terms of Λ and M , e.g.

ΩDM

ΩexpDM

=(Λ/700 GeV)4

(M/150 GeV)2σSI ≈ 5 10−39 cm2

(M

mN +M

)2 (700 GeV

Λ

)4

CMS and ATLAS j /ET and γ /ET searches imply Λ > 700 GeV for M Λ.

For any collider the limit will be Λ √s. So the effective approximation is not

valid. What LHC would really see is the particle that mediates the operator.

Detectable Dark Matter below a TeV?

DM above a TeV is too heavy for LHC and for δm2h. DM below a TeV with

weak gauge interactions annihilates too much leaving a too low ΩDM, unless:

• Extra solution at M < MW such that too large σ(DM DM → W+W−) is

kinematically suppressed. Not fully excluded by LEP. E.g. ‘inert doublet’

• Mix interacting (M v) with singlets (M → 0): get any intermediate M .

• DM as singlet + extra coupling e.g. binoDM-lepton-slepton Yukawa in SUSY

works if sleptons are around or below the LEP bound. Small extra couplings

can be resonantly enhanced, e.g. DM DM → A→ bb in SUSY if MA = 2M .

DM at colliders: summary

Simplest scenario: only DM is produced, an initial state jet allows to see

6pT + soft jet

Plausible scenario: DM could be the lightest of a new set of particles (like

in SUSY). LHC dominantly produces heavier colored particles (gluino, squarks)

that decay down to DM. The signal depends a lot on the decay chains

6pT + hard jets or leptons or...

Possible scenario: the lightest sparticle is charged or colored and decays into

“gravitinos” or “axino” DM with life time τ >∼m

charged tracks, decays after the collisions

etc etc etc but nothing seen in data so far

Direct DM detection

Direct DM detection: key parameter

σSI = spin-independent DM-nucleon cross section

allows to compare experiments: DM/nucleus cross section σN = A2σSI.

DM

N

Z

N

DM

tree, vector

ΣSI »Α2 mN

2

MZ4

DM

N

h

N

DM

tree, scalar

ΣSI »Α2 mN

4

Mh6

DM

N

DM

N

W W

DM±

N

loop

ΣSI »Α4 mN

4

MW6

The vector effect vanishes if DM is real (e.g. a Majorana fermion).

Experimental progress

ææ æ

æ æææ ææ æ

æ

æææ

ææ

ææ

çç

ç

ç

çç

1985 1990 1995 2000 2005 2010 2015 202010-48

10-46

10-44

10-42

10-40

10-38

10-36 cosmology Σ » 1T0MPl

vector: Σ » Α2mN2 MZ

4

Higgs: Σ » Α2mN4 mh

6

loop: Σ » Α4mN4 H4ΠL2MW

6

atmospheric Ν background

year

ΣSI

incm

2fo

rMD

M»

MZ

History of direct DM searches

DA

MA

Xen

on

CD

MS

DM must be neutral under the γ, g and almost neutral under the Z

‘Anomaly-free’ Dark Matter

Ignoring experiments that claim ‘anomalous’ results, this is the present status:

101 102 103

10-45

10-44

10-43

10-42

DM mass in GeV

ΣSI

incm

2

Spin-independent DM detection: bounds at 90%CL

CDMS2009

Xenon2011

Xenon2012

[Below a few GeV is ≈ terra incognita]

Sub GeV DM

When nuclear recoil ER ∼ EDM(mN/M) becomes too small, EDM = M2 v

2 ∼50 eV(M/100 MeV) can lead to (a) e ionization; (b) e excitation; (c) molecular

dissociation giving individual e or γ or ion or phonons. First bound: 4

FDM=Ha meêqL2

FDM=Ha meêqLFDM=Ha meêqL2

1 10 100 10310-37

10-36

10-35

10-34

10-33

10-32

10-31

10-30

Dark Matter Mass @MeVD

se@cm

2 DExcluded byXENON10data

1electron

2electrons3electrons

Hidden-Photonmodels freeze-in

FIG. 3: 90% CL exclusion limits on e for candidates witha DM form-factor FDM(q) of (↵me/q) (red/lower line), cor-responding to DM with an electric dipole moment, and(↵me/q)2 (blue/upper line), corresponding to DM scatteringthrough a very light mediator. Dashed lines and bands are asin Fig. 2. The pale blue region shows the previously allowedparameter space for DM coupled through a very light hiddenphoton (FDM = (↵me/q)2), with the gray strip indicating the“freeze-in” region (taken from [2]).

models in which the DM candidate is a fermion coupledto the visible sector through a kinetically mixed “hid-den photon” with O(MeV-GeV) mass, and satisfying allpreviously known constraints (from [2]; see also [3] [25]).

Fig. 3 shows the exclusion limits in the mDM-e planefor DM candidates whose interaction with electrons isenhanced at small momentum-transfers by a DM form-factor, FDM. The red (lower) curves correspond toFDM = (↵me/q), or DM scattering through an elec-tric dipole moment, and the blue (upper) curves toFDM = (↵me/q)2, or DM scattering though a very light( keV) scalar or vector mediator. Bounds set by 1-,2-, and 3-electron rates are shown by dashed lines, andthe central limits by dark lines. The bands illustratethe theoretical uncertainty. Both form-factors suppressthe relative rate of events with larger energy deposition,and so reduce the fraction (and hence the importance) ofevents containing multiple electrons. The pale blue re-gion shows the parameter space for DM coupled througha very light hidden-photon mediator, and satisfying allpreviously known constraints, with the gray strip show-ing where the correct abundance is achieved through“freeze-in” (from [2]). These regions should be comparedto the blue exclusion curve.DISCUSSION. The results above demonstrate, for thefirst time, the ability of direct detection experiments toprobe DM masses far below a GeV. It is encouraging thatwith only 15 kg-days of data, and no attempt to controlsingle-electron backgrounds, the XENON10 experimentplaces meaningful bounds down to masses of a few MeV.

It should be emphasized that this analysis lacks theability to distinguish signal from background. One

promising method is the expected annual modulation ofthe signal. As discussed in [2], additional discriminationmay be possible via the collection of individual photons,phonons [24], or ions, although at present such technolo-gies have yet to be established.

Independently, this type of search could be signifi-cantly improved with a better understanding of few-electron backgrounds. A quantitative background es-timate was not made in [10], making background sub-traction impossible. Single-electron ionization signalshave been studied, and potential causes discussed, byXENON10 [9], ZEPLIN-II [7], and ZEPLIN-III [8]. Pos-sible sources include photo-dissociation of negativelycharged impurities, spontaneous emission of electronsthat have become trapped in the potential barrier at theliquid-gas interface, and field emission in the region ofthe cathode. The former two processes would not beexpected to produce true two- or three-electron events,although single electron events may overlap in time, giv-ing the appearance of an isolated, double-electron event.With a dedicated study, these backgrounds could bequantitatively estimated and reduced.

With larger targets and longer exposure times, ongo-ing and upcoming direct detection experiments such asXENON100, XENON1T, LUX, and CDMS, should beable to improve on the sensitivity reported here. Suchimprovements may require optimizations of the trigger-ing thresholds, and will strongly benefit from additionalstudies of the backgrounds.

Acknowledgements

R.E. acknowledges support from NSF grant PHY-0969739. J.M. is supported by a Simons PostdoctoralFellowship. T.V. is supported in part by a grant fromthe Israel Science Foundation, the US-Israel BinationalScience Foundation and the EU-FP7 Marie Curie, CIGfellowship.

rouven.essig@stonybrook.edu† aaronm@physik.uzh.ch‡ jmardon@stanford.edu§ pfs@llnl.gov¶ tomerv@post.tau.ac.il

[1] G. Bertone, D. Hooper and J. Silk, Phys. Rept. 405(2005) 279 [arXiv:hep-ph/0404175].

[2] R. Essig, J. Mardon and T. Volansky, Phys. Rev. D 85(2012) 076007 [arXiv:1108.5383].

[3] T. Lin, H. B. Yu and K. M. Zurek, [arXiv:1111.0293].[4] J. L. Feng and J. Kumar, Phys. Rev. Lett. 101 (2008)

231301 [arXiv:0803.4196].[5] X. Chu, T. Hambye and M. H. G. Tytgat,

[arXiv:1112.0493].[6] P. W. Graham, D. E. Kaplan, S. Rajendran and

M. T. Walters, [arXiv:1203.2531].[7] B. Edwards et al., Astropart. Phys. 30 (2008) 54

[arXiv:0708.0768].

Ultra-light DM

If DM is so light that its scatterings are too soft even for atomic physics, DM can

still be detected via quantum interference! Experimentalists are able of splitting

the wave functions of nucleons or atoms by a small distance (nm to cm).

Quantum interference is lost if one of them interacts with DM. Experiments

are done with a few atoms for a few seconds, so the bound is NA weaker 6

101 102 103 104 105 106 107 10810-30

10-28

10-26

10-24

10-22

10-20

10-18

10-16

10-14

10-12

10-30

10-28

10-26

10-24

10-22

10-20

10-18

10-16

10-14

10-12

mDM HeVL

sHcm

2 L OTIMA-6: 200 kmHdecoherenceLHphase shiftLNanosphere: 200 kmHdecoherenceLHphase shiftL

AGIS

XQC

Heating+Halo

Ly-a

CMB+

LSS

200 km HrocketL

30 km HballoonLSea level

Hen0

C70

PFNS10

OTIMA-4

OTIMA-6

OTIMA-8

AGIS

Nanosphere

SIMP

101 102 103 104 105 106 107 10810-30

10-28

10-26

10-24

10-22

10-20

10-18

10-16

10-14

10-12

10-30

10-28

10-26

10-24

10-22

10-20

10-18

10-16

10-14

10-12

mDM HeVLFIG. 5. The sensitivity of several existing and proposed superposition experiments to the spin-independentnucleon scattering cross-section of dark matter, compared with existing constraints. (a) Gray shaded regions arerobustly excluded by the X-ray Quantum Calorimetry experiment (“XQC” [9]) and galactic heating and halo stability arguments(“Heating+Halo” [29]). Hatched regions are incompatible with thermal DM models due to observations of the cosmic microwavebackground with large scale structure data (“CMB+LSS” [30]) and the Lyman-↵ forest (“Ly-↵” [31]). Solid colored lines boundregions where DM would cause decoherence in three proposed experiments: a satellite-based atom interferometer (“AGIS” [24]),optically-trapped silicon nanospheres (“Nanosphere” [19]), and the OTIMA interferometer with clusters of gold of mass 106

amu (“OTIMA-6” [17]). A successful AGIS satellite would set new exclusion limits on DM where its sensitivity dips below thehalo heating/stability bound for mDM 1 keV. On the other hand, the OTIMA and nanosphere experiments would be shieldedfrom DM by the atmosphere if operated at sea level, so exclusion regions illustrate the sensitivity at an altitude of 200 km. Thedarker regions bordered by colored dashed lines indicates where the coherent phase shift due to the DM wind could be observedwithout being overwhelmed by decoherence. (b) On top of the existing exclusions (now black dotted lines), the colored linesgive the lower limits on the sensitivities of existing interferometers with helium atoms (“He” [32]), cold neutrons (“n0” [33]),fullerenes (“C70” [34]), and the large organic molecule C60[C12F25]10 (“PFNS10” [18]). Also shown are sensitivities for theAGIS satellite, the OTIMA interferometer with three choices of gold cluster mass (104, 106, and 108 amu, although the last isnot feasible for an Earth-bound experiment), and the nanosphere experiment. The border is defined by an e-fold suppressionof the interference fringes: = 1e. Sensitivity increases dramatically for larger masses. When operated within the Earth’satmosphere, there is a potential to detect DM only where the sensitivity dips below the dashed-dotted line corresponding tothe degree of shielding at the relevant altitudes. None of the existing experiments do so. For reverence, strongly interactingmassive particle (“SIMP” [35]) models lie in the black band.

the Earth’s surface at about 10−27cm2. This can be cir-cumvented by placing the experiment on a high-altitudeballoon (∼30 km altitude; ∼10−25cm2), a sounding rocket(∼200 km altitude; ∼10−18cm2), or on a satellite. I donot know if anyone has studied the possibility of produc-ing macroscopic quantum superpositions on a balloon orrocket, but spaceborne experiments are both feasible andcompelling for independent reasons [17, 38, 39].

Figure 5 shows the potential reach of several existingmatter interferometers [18, 32–34] in the absence of atmo-spheric shielding. The separation vector x is assumedto point into the DM wind. The e↵ects of rotating xwith respect to the wind are order unity and are depicted

in figure 4. Modern experiments often use multiple grat-ings with many slits to overcome diculties with beamcoherence and tiny de Broglie wavelengths [40, 41], sothe matter is not described by a simple superpositionof two spatially separated wavepackets. But the interfer-ometers still require good coherence over distances whichspan multiple slits, so it is reasonable to estimate theirsensitivity by taking x to be the period of the relevantgrating. (Only the results for small mDM will depend onthe choice of x; for larger masses, which are in the shortwavelength limit, any scattering event results in completedecoherence independent of the spatial separation.)

To demonstrate the potential of future experiments to

Indirect DM detection

Indirect signals of Dark Matter

DM DM annihilations in our galaxy might give detectable γ, e+, p, d.

Measurements of charged cosmic rays

p e+/(e+ + e−) e+ + e−

ó óó

óó

ó

óóóóóó

ó

ó

ó

óó

óóó

ó

óóóóóó

ò

ò

ò

ò

òò ò ò

ò

ô

ôô

ôô ô ô ô ô

í

í

í

ç

ç

ò

ò

òò

÷

÷ ÷÷

÷

ææææ

æ

æ

æ

ææ

æ

æ

æ

10-1 100 101 102 10310-6

10-5

10-4

10-3

10-2

10-1

p kinetic energy T in GeV

pfl

ux

in1

m2se

csr

GeV

BESS 95+97BESS 98BESS 99BESS 00Wizard-MASS 91CAPRICE 94CAPRICE 98AMS-01 98PAMELA 08

æææ

æ

æææ

ææ

ææ

ææ

æææææ

ææ

ææ

æææææ

æææææææææææ

æææææ

æææææææææææ

à

à

à

àà

ààà

à

à

à

à

ààà

à

à

10-1 1 10 102 1031%

10%

3%

30%

Energy in GeV

Posit

ron

fracti

on

AMS 2013

PAMELA 2009

ææ

æ

æ

æ

æææ

ææ

ææ

ææææ

ææ

æææ

æææ æ æ æ æ æ æ æ æ æ æ æ æ æ æ

ææ

æ æ

à

à

à

à

à

à

à

à

ì

ì

ì

ìì

ì

ì

ì

ì

ò

ò

ò

ò

òò

ò

òò ò

ò

ò

ò

ò

ò

ò

ò

ò

ò

ò

ò

10 102 103 104

0.01

0.003

0.03

Energy in GeV

E3

He-

+e+

LGeV

2c

m2se

c

FERMI

HESS08

HESS09

ATIC08

Explaining the e+ excess

Due to astrophysics? Maybe pulsars or primary e+? Due to DM?

e+ spectrum reproduced if DM annihilates into leptons with σv ∼ 103σvcosmo

æææ

æ

æææ

ææ

ææ

ææ

æææææ

ææ

ææ

æææææ

æææææææææææ

ææ

æææææææ

æææææææ

à

à

à

àà

àà

àà

à

à

à

ààà

à

à

10 102 103 104

1%

10%

3%

30%

Positron energy in GeV

Posi

tron

frac

tion

DM DM ® VV ® 4Μ with M = 1 TeV

background?

AMS 2013

PAMELA 2009

1000 10000300 300020

30

40

50

60

DM mass in GeV

Χ2

DM annihilation fit to e+ after AMS

e+e-

Μ+ Μ-Τ+Τ-4e

4Μ

4Τ

qW

Z

b

t

h

AMS hint of a flattening favours M ∼ TeV

Not seen in p: leptonic modes again. Not seen in γ: needs quasi-constant ρ(r).

Explaining the e± excesses

Hints of drops in e+/(e+ + e−) and in e+ + e− but at different energies

ææ

ææ

æææ

ææ

æ

ææææææææ

æææææ

æææææ æ æ æ æ æ æ æ æ æ æ ææ

æ æ

à

à

à

à

à

à

à

à

ì

ì

ìì

ìì

ì

ì

ì

ò

ò

òò

ò ò

ò

ò ò ò

ò

ò

ò

ò

ò

ò

ò

ò

ò

ò

ò

10 102 103 104

0.01

0.003

0.03

Energy in GeV

E3

He-+

e+

LGeV

2cm

2se

c

DM DM ® VV ® 4Μ with M = 1 TeV

FERMI

HESS08

HESS09

1000 10000300 300055

60

65

70

75

80

DM mass in GeV

Χ2

DM annihilation fit to e+ and e

++e- after AMS, FERMI

4e

4Μ

4ΤΜ+ Μ- Τ+Τ-

AMS can clarify measuring very precisely the e+ + e− spectrum

Advertisement

You want to compute all signatures of your DM model in positrons, electrons,

neutrinos, gamma rays... but you dont want to mess around with astrophysics?

The Poor Particle Physicist Cookbookfor Dark Matter Indirect Direction

www.marcocirelli.net/PPPC4DMID.html

We provide ingredients and recipes

for computing signals of TeV-scale

Dark Matter annihilations and decays

in the Galaxy and beyond.

arXiv:1012.4515

Dark Matter: what it is?

Why DM should be a weak scale particle, if new physics must not be there?

Dark Matter: how heavy?

DM exists, but so far we have seen only its gravity

Decades of theoretical work restricted the DM mass to a range of 100 orders

of magnitude. We do not even know if DM is astro-physics or particle physics

10-30 10-20 10-10 1 1010 1020 1030eV

10-10 1 1010 1020 1030 1040kg

Solar mass

Planck scale

weak scaleTnow

Primordialblack hole

Ultra-light scalars, axion Νs

thermalparticles

DM as ultra-heavy objects (MACHO)?

Dead stars, planets, Black Holes... must be either non-baryonic (mirror world?)

or made before BBN (primordial BH?). DM ‘particles’ are lighter than small

galaxies so M < 105M where M = 2 1033 g is the solar mass. Microlensing

surveys imply that MACHO Milky Way fraction is < 20% around M ∼M.

10-7 10-6 10-5 10-4 10-3 10-2 10-1 1 100%

10%

20%

30%

40%

50%

60%

MACHO mass M in solar masses

fracti

onfo

fhalo

mass

cons

isting

ofM

ACHO

s

MACHO

coll, 2000

EROS coll., 2006

95% excluded

OGLE-3

2011

95% excl.

EROS-1 + MACHO, 1998

95% excl

DM as Primordial Black Holes (PBH)?

PBH are not predicted by standard cosmology because primordial fluctuationshave small amplitude δk ∼ 10−5 1. Allowed mass range:

10−13M <∼ M <∼ 10−7MA BH cannot be too light be-

cause it emits photons evapo-

rating in a time ∼ G2NM

3.

Non-observation of micro-

lensing nor of X-ray emission

from matter falling into BH.

1015 1020 1025 1030 1035 1040 1045 105010-4

10-3

10-2

10-1

110-20 10-15 10-10 10-5 1 105 1010 1015

MACHO or PBH mass M in grams

fracti

onfo

fDM

MACHO or PBH mass M in solar masses

PBH

evap

orated

byno

w

evap

oratio

ntod

ay

X-raysspoilCMB

allowedPBH DMwindow

GRBlens

microlensing

accumulationin GC

disfavored bygalaxy sizes

binaries

radiolensing

Axions as ultra-light scalar DM

Practical summary: the axion a is a well-motivated particle with

ma =mπfπ/fa√

(1 +mu/md)(1 +md/mu +md/ms)≈ 0.6 meV

1010 GeV

fa

gaγγ =αem

2πfa(

∑q2

T2− 2

3

4 +mu/md +mu/ms

1 +mu/md +mu/ms)

ΩDM ≈√ma

eV

(a∗

1011 GeV

)2

Axion searches

shining-through-wall

a*=0.01 fa

a*=0.1 fa

a*= fa

10-8 10-6 10-4 10-2 1 102 104

10-20

10-18

10-16

10-14

10-12

10-10

10-8

10-6

axion mass in eV

axio

nco

uplin

gg a

ΓΓin

1G

eV

Τa =

TUniverse

Bound froma ®

ΓΓ

axion

0

models

83

Supernovae, stars and helioscopes

DM a ® Γ

conversion

ADMXADMX near future:Sensitivity with dilution-refrigerator cooling"

10-10

10-16

10-15

10-14

10-13

1 10 100 1000

1 10 100

Axio

n C

oupl

ing

|gaaa |

(GeV

-1)

Axion Mass (µeV)

ADMX Achieved and Projected Sensitivity

Cavity Frequency (GHz)

"Hadronic" Coupling

Minimum Coupling

Axion Co

ld Dark

Matter

ADMX Published

Limits

2013Target

2014Target

2015Target

10 GHz R&D

500 MHz R&D

Non RF-cavity TechniquesTo

o Mu

ch D

ark

Matt

er

Supernova Bound

Dilution refrigerator cooled detectors allow scanning at or below “DFSZ” sensitivity at fractional dark-matter halo density."This defines a definitive QCD dark-matter axion search!

IDM23jul12 LJR 22

New ideas for axion detection

DM axions passing through a magnetic

fields make an electric field with ω = ma. If

a metallic surface is also present, an electro-

magnetic wave is emitted perpendicularly to

it. Using a spherical surface, such waves

can focused in the centre, and detected.

Detectors sensitive to powers of 10−25 W

would start to probe the axion strip for all

masses down to ≈ 1/m. [1212.2970]

EBL

xionOptical

CAST+Sumico

ALPS

Haloscopes

10-21W

10-23W

10-25W

1 Hz

10-2Hz

10-4Hz

Standard ALP CDM

Suppressed Therm

al mass

Axionmodels

-9 -6 -3 0 3-19

-16

-13

-10

-7

Log10 mf @eVD

Log 10g@Ge

V-1 D

Figure 3: The allowed parameter space for axion-like particle dark matter (ALP CDM) is shownin various shades of red (for details see Ref. [3]). The various colored regions are excluded byexperiments and astrophysical observations that do not require HP dark matter (for reviewssee [1,2]). The lines correspond to the sensitivity of a dish antenna (1m2 dish in a 5 T magneticfield) with a detector sensitive to 1021, 1023 and 1025 W (green, from top to bottom) and1, 0.01 and 104 photons per second (blue, from top to bottom).

(2.24) we therefore find,

g, sens =3.6 108

GeV

0@ 5 Tq

h|B|||2i

1A

Pdet

1023 W

12 m

eV

0.3 GeV/cm3

DM,halo

12

1 m2

Adish

12

,

(3.4)and

g, sens =4.6 106

GeV

0@ 5 Tq

h|B|||2i

1A

R,det

1 Hz

12 m

eV

32

0.3 GeV/cm3

DM,halo

12

1 m2

Adish

12

, (3.5)

where h|B|||2i is the average value of the magnetic field squared parallel to the antenna dish.

10

Axions with f ∼MPl could be detected as oscillating neutron electric dipole or

by angular momentum loss of rotating black holes due to axion emission...

Axion DM already observed!?

Three big claims from Sikivie and others:

1) Axions interact form a coherent Bose-Einstein condensate;

2) This leads to caustics in the DM galactic densities ρ(r) at special radii;

3) Such caustics are supported by data.

Step 1) is based on interactions rates linear in the axion couplings, either

gravity or a small quartic. This is derived as a consequence of the large axion

occupation numbers, such that short-time axion scatterings would not conserve

energy. I don’t understand what is the sense of this.

Axion coherence should hold at most for times t ∼ 1/maβ2.

Conclusions / last slide

1) DM exists.

2) We no longer believe we know where to look for it.

3) LHC starts now to be sensitive to electroweak multiplets.

4) Watch axions