Sept 5, 2011, TAUP meeting

Maxim Pospelov University of Victoria and Perimeter Institute

Alternatives to WIMPs and alternative uses of WIMP detectors

Plan 1.  Rutherford-Marsden-Geiger centennial. First “missing mass”

problem and its resolution. 2.  The second “missing mass” problem – the nature/origin of dark

matter in the Universe is unknown. Different ideas cover the mass range from 10-25 eV to Mblack hole. Weakly-interacting massive particles and their modifications, super-weakly interacting massive particles, super-cold DM.

3.  How robust is the WIMP paradigm? Are we guaranteed an eventual WIMP discovery? Secluded WIMPs and dark forces.

4.  Super-weakly interacting massive particles. Mono-energetic energy deposition of bosonic super-WIMPs in underground detectors.

5.  Alternative uses of WIMP detectors: looking for solar axions, exotic solar neutrinos, constraining the neutrino properties.

Atomic nucleus centennial The discovery of atomic nucleus created the first missing mass problem:

A>Z Or why is Mnucleus > Z mproton ? What accounts for 50% or more of the missing mass in a nucleus of an atom?

Rutherford’s own suggestion – tightly packed A-Z electrons on top of A protons inside the nucleus – was soon shown to be incorrect via the studies of hyperfine structure of e.g. 13C. A wild theoretical suggestion was floated – a new type of particle, electrically neutral, with spin ½ and strong interaction with protons.

Resolution of the 1st missing mass problem

The issue was resolved within a few years: α  + 9Be --> α’s + very same particle theorists were talking about. The resolution of the 1st missing mass problem brought about the discovery of a new (quasi)-stable particle – the neutron – and the discovery of a new force that glues nucleons together – the strong force. Practical lesson: if you smack things hard enough, the stuff has to come out.

2nd missing mass problem – origin/nature of dark matter

In the era of precision cosmology we know that 1.  There is substantial body of

evidence for DM at different distance scales.

2.  It is 6 times more abundant than baryons and contributes ~1/4 of the total energy budget.

The 2nd missing mass problem – the origin of dark matter – is much harder to crack. We know about DM presence only via gravity, and we need some luck (e.g. accessible masses and sizable interaction rates) to find out what it is. Unlike the 1st problem, we do not know which things to smack.

The most unambiguous evidence is on CMB scales

Prominence of the third peak çè significant presence of cold-ish DM (Any candidate theory toying with modified gravity on galactic scales is forced to address its existence. DM of some sort is typically needed.)

From SPT website, U of Chicago.

Simple classification of particle DM models

At some early cosmological epoch of hot Universe, with temperature T >> DM mass, the abundance of these particles relative to a species of SM (e.g. photons) was Normal: WIMPs Very small: super-WIMPs Huge:

“super-cold” DM

Simple classification of particle DM models

At some early cosmological epoch of hot Universe, with temperature T >> DM mass, the abundance of these particles relative to a species of SM (e.g. photons) was Normal: Sizable interaction rates ensure thermal equilibrium, NDM/Nγ =1. Stability of particles on the scale tUniverse is required. Freeze-out calculation gives the required annihilation cross section for DM -> SM of order ~ 1 pbn, which points towards weak scale. These are WIMPs! [Huge number of “motivated” models.] Very small: Very tiny interaction rates (e.g. 10-10 couplings from WIMPs). Never in thermal equilibrium. Populated by thermal leakage of SM fields with sub-Hubble rate (freeze-in) or by decays of parent WIMPs. [Gravitinos, sterile neutrinos, and other “feeble” creatures] Huge: Almost non-interacting light, m< eV, particles with huge occupation numbers of lowest momentum states, e.g. NDM/Nγ ~1010. “Super-cool DM”. Must be bosonic. Axions, or other very light scalar fields.

Signatures vary considerably WIMPs: Scattering on nuclei in the underground detectors. Indirect signatures of cosmological and galactic annihilation (weak scale energy cosmic rays, small alteration of CMB, neutrinos from regions of WIMP accumulation). Collider signatures (DM pair production, invisible Higgs decay etc)! Super-WIMPs. Very model-dependent, including no detectable signatures. Possible X-rays/gamma-rays from occasional decays, can be warm-ish and influence cosmic structures. In rare exceptions, signatures from absorption in underground detectors. Super-cool DM: Very model-dependent, including no detectable signatures. Possible conversion to photons in external fields. Possible CMB signatures via iso-curvature modes etc.

My main points in this talk 1.  Thinking about different DM possibilities within/

around WIMP paradigm leads to new theoretical ideas/new experimental possibilities.

2.  Direct detection of super-WIMPs (superweakly interacting massive particles) is possible within certain classes of models and can be e.g. superior to X-ray signatures.

3.  Underground WIMP detectors – whose primary goal is to detect WIMP recoil – are also sensitive to a variety of other signals: solar axions, non-SM ν

How robust is a “conventional” WIMP?

! annv !1pbn" c

DM-SM mediators

SM states DM states

Cosmological (also galactic) annihilation

Collider WIMP pair-production

WIM

P-nu

cleu

s sca

tterin

g

Question: does sizable annihilation cross section always imply sizable scattering rate and collider DM production? If WIMP DM is what nature chose, is there a guarantee that WIMPs will be eventually discovered by something like Xenon-1000…0 ?

Two examples from SM Example 1. Consider electron-muon scattering at E ~ q ~ 1 GeV

Same order of magnitude cross sections. Therefore, one has some ground to expect that the answer is “yes”. However… Example 2. Consider proton-proton scattering at E ~ q ~ 1 GeV, and ask the question of what is the neutrino production cross section relative to scattering and the neutrino annihilation into SM states

In this example, a process of neutrino production goes into two-steps, with pions being forced to decay to neutrinos by kinematic reasons

!µµ!ee ~! µe!µe ~! ee!µµ

!pp!nn" +" +!ppµµ##

~ 30mbn >> !##!SM ,!# p!# p

13

Secluded WIMPs MP, Ritz, Voloshin; Finkbeiner and Weiner, 2007. Original model: Holdom 86

This Lagrangian describes an extra U(1)’ group (dark force), and some matter charged under it. Mixing angle κ controls the coupling to the SM.

ψ – Dirac type WIMP; Vµ – mediator particle. Two kinematic regimes can be readily identified: §  mmediator > mWIMP ψ+ + ψ- -> virtual V* -> SM states

κ has to be sizable to satisfy the constraint on cross section 2. mmediator < mWIMP ψ+ + ψ- -> on-shell V +V, followed by V -> SM states

There is almost no constraint on κ other than it has to decay before BBN. κ2 » 10-20 can do the job.

14

Two types of WIMPs Un-secluded Secluded

Ultimately discoverable Potentially well-hidden Size of mixing*coupling is set by Mixing angle can be annihilation. Cannot be too small. 10-10 or so. It is not

fixed by DM annihilation You think gravitino DM is depressing, but so can be WIMPs

15

Indirect signatures of secluded WIMPs

Annihilation into a pair of V-bosons, followed by decay create boosted decay products.

If mV is under mDM vDM ~ GeV, the following consequences are generic

(Arkani-Hamed, Finkbeiner, Slatyer, Weiner; MP and Ritz, Oct 2008) 1.  Annihilation products are dominated by electrons and positrons 2.  Antiprotons are absent and monochromatic photon fraction is

suppressed 3.  The rate of annihilation in the galaxy, <σann v>, is enhanced relative

to the cosmological <σann v> because of the long-range attractive V-mediated force in the DM sector. (Sommerfeld and resonant enhancement)

Fits the PAMELA signature. [which can of course be explained by a variety of pure astrophysical mechanisms]

16

PAMELA positron fraction seem[ed] to be “abnormal”

No surprises with antiprotons, but there is seemingly a need for a new source of positrons!

There is a “boost” factor of 100-1000 “needed” for the WIMP interpretation of PAMELA signal. E.g. SUSY neutralinos would not work, because the annihilation cross section is too small. Light dark force rectifies this problem.

17

Thinking about secluded WIMPs and dark forces have resulted in the brand new research program at the intensity frontier: searches of light mediators using colliders and fixed target experiments. Independent motivation for light force also emerges from models of MeV WIMP DM, Fayet; Boehm, 2004.

Recently, exclusion limits have become more stringent thanks to Mainz and Jlab experiments.

Such searches are motivated in their own right, independently from the DM theme and will be continued in the future. Thinking outside m0-m1/2 box might be good for you.

18

Currently all “direct DM detection” experiments search for the same thing

An  average  Dark  Ma,er                          A  more  expensive  DM  experiment  detec6on  experiment                    Diversifying physcis output of direct detection exp’s is needed !!! (Take a cue from HEP exp’s)

$$

$$$$$$

19

Scattering vs absorption

WIMP-nucleus scattering Atomic absorption of super-WIMPs

WIMP Super-WIMP electron

nucleus nucleus

Signal: ionization + phonons/light Ionization at E=msuperWIMP

d(Events)/dE d(Events)/dE

E E

Direct super-WIMP detection Detecting WIMPs is not easy. Directly detecting super-weakly interacting massive particles must be impossible? Not necessarily. The same experiments that look at the recoil of WIMPs may look at the absorption of (multi)-keV scale vectors, scalar and pseudoscalars. For example, thinking that direct WIMP detection experiments can be sensitive to 10-36 cm2 elastic cross sections, they can also be sensitive to 1/(1010 GeV) coupled axion-like keV particles: Smaller cross sections are compensated by larger DM number densities and absence of v~10-3 suppression. MP, Ritz, Voloshin, 2008

Examples of super-WIMP models 1.  Pseudoscalars and scalars in the keV range,

Lifetime and X-ray constraints on fa are very stringent. 2. Vectors, For mV<MeV, the lifetime of V is extended because of the slowness of V -> 3γ. Finite but very small couplings lead to the thermal leakage of SM states into a or V. You can have super-WIMP DM out of this.

22

Features of the axion-type models

•  The lifetime constraint is very strong:

 and γ-background constraint is even stronger (Gondolo, Raffelt). Absorption by atoms is given by Hamiltonian that remains finite in the limit vDM -> 0, and it is unmodulated. (Inelastic cross sections ~ 1/velocity) Much better calculations have recently ben performed using

proper atomic physics in Dzuba et al, 2010)

23

Counting Rates Axion-like super-WIMPs: Vector-like super-WIMPs: The rates are not seasonally modulated, contrary to some earlier

claims made earlier in astro-­‐ph/0511262. Absorption of super-WIMPs is not the source of DAMA

modulation. Some constraints on keV-scale axion-like DM have been published by CoGeNT and CDMS.

24

Summary of constraints on vector DM

Direct detection search of Vector super-WIMP is competitive with other constraints. MP, Ritz, Voloshin, 2008.

See also Postma, Redondo, 2008, for the in-depth analysis of the same model.

Very

“ap

prox

imat

e”, p

rope

r Ex

perim

enta

l ana

lysi

s is n

eede

d

Alternative uses of DM detectors You keep looking for WIMPs; you see some signal; is it clear that what you see is actually a WIMP recoil? What else other than DM are you sensitive to? 1.  Sun can emit exotic nearly massless particles (axions, “dark vectors”,

pico-charge particles etc), which lead to the ionization signal in DM detectors. (F.Avignone et al, from 1980s).

2.  Sun does emit neutrinos, but their recoil is too weak to be seen. However, SM neutrinos can oscillate into new neutrino states, that leave a more pronounced nuclear elastic recoil signal comparable to present sensitivity levels. (MP, 2011).

26

Axions from the Sun

 

5

The solar axion flux was calculated in Ref. [19] (whereCAST results were also used to constrain it in combina-tion with coupling of axions to photons). At Earth thisflux is given by

!a = 4.5! 1023!1 GeV

faN

"2

! cm!2s!1, (18)

where faN is some e"ective coupling constant to nucleonsthat can be related to the coupling of axions to quarkspins. The expected counting rates of argon, germaniumand xenon experiments are given by

RAr " 4

!106GeV

(fafaN )1/2

"4

kg!1day!1, (19)

RGe " 18

!106GeV

(fafaN )1/2

"4

kg!1day!1, (20)

RXe " 11

!106GeV

(fafaN )1/2

"4

kg!1day!1, (21)

where the following values for the K-factors are used:

KAr(14.4 keV) = 329;

KGe(14.4 keV) = 2746;

KXe(14.4 keV) = 2930.

These rates should provide the sensitivity to (fafaN )1/2

in the window between 106 and 107 GeV. Similar strengthconstraints were derived in the recent work [20], wherea # 3% annual modulation of the axion signal was ex-ploited in conjunction with DAMA results. (Unlike thesignal from WIMP dark matter that is expected to havea maximum in June, the solar axion signal is minimizedin early July.) We leave it to the experimental collabora-tions to determine the exact upper limits on solar axionsensuing from their results.If the coupling to photons is not zero, Fµ! F̃µ!a/(4fa"),

then we can calculate the counting rate, using the axionflux provided in Ref. [8]:

d!a

d#a= 6.02! 1030

#1 GeVfa!

$2!2.481a e!

"a1.205 (22)

!cm!2 s!1 keV!1.

Counting rate for the axio-electric e"ect is given by theproduct of the calculated absorption cross section andthe flux (22). For (fafa")1/2 normalized on 108 GeV, weget the counting rates plotted in Figure 4. Integrationover axion energy leads to the following total countingrates

RAr " 5.0

!108GeV

(fafa")1/2

"4

kg!1day!1, (23)

RGe " 5.2

!108GeV

(fafa")1/2

"4

kg!1day!1, (24)

RXe " 8.2

!108GeV

(fafa")1/2

"4

kg!1day!1. (25)

FIG. 4: Counting rate for the axio-electric e!ect for Ar, Geand Xe as a function of axion energy.

Comparing this to the counting rate of the CDMSexperiment [11], one can see that the equivalent of(fafa")1/2 # 108 GeV are being probed, as the count-ing rates in the window from 1.5 to 4 keV reachO(1 kg!1day!1keV!1). Similar sensitivity is achievedin the CoGent experiment [13].Finally, the axion flux can be created by the emission

of the axions due to the same interaction that leads toatomic ionization. In this case, however, the productioncross section is down by additional factor of E2

a/m2e [14],

and the sensitivity to fa in this case does not exceed 106

GeV.

V. CONCLUSIONS

QCD axions represent one of the most well-motivatedextensions of the Standard Model. Their light mass andsmall couplings allow them to be produced in the Solarinterior and escape reaching the Earth. With the prolif-eration of the low-background searches of dark matter,one should also conduct searches of solar axions. In thispaper we have calculated the cross sections relevant forthese searches, improving upon the simple scaling rela-tions that tie the axio-electric and photo-electric e"ects.Last two years has brought a significant progress in

sensitivity to any ionizing e"ects in Germanium in thewindow from 1 to 10 keV [11, 13]. Currently, the Co-Gent experiment has very low backgrounds in the windowfrom 2 to 4 keV, where the solar axion signal is expectedto peak. With acquiring more statistics, the sensitivity

5

The solar axion flux was calculated in Ref. [19] (whereCAST results were also used to constrain it in combina-tion with coupling of axions to photons). At Earth thisflux is given by

!a = 4.5! 1023!1 GeV

faN

"2

! cm!2s!1, (18)

where faN is some e"ective coupling constant to nucleonsthat can be related to the coupling of axions to quarkspins. The expected counting rates of argon, germaniumand xenon experiments are given by

RAr " 4

!106GeV

(fafaN )1/2

"4

kg!1day!1, (19)

RGe " 18

!106GeV

(fafaN )1/2

"4

kg!1day!1, (20)

RXe " 11

!106GeV

(fafaN )1/2

"4

kg!1day!1, (21)

where the following values for the K-factors are used:

KAr(14.4 keV) = 329;

KGe(14.4 keV) = 2746;

KXe(14.4 keV) = 2930.

These rates should provide the sensitivity to (fafaN )1/2

in the window between 106 and 107 GeV. Similar strengthconstraints were derived in the recent work [20], wherea # 3% annual modulation of the axion signal was ex-ploited in conjunction with DAMA results. (Unlike thesignal from WIMP dark matter that is expected to havea maximum in June, the solar axion signal is minimizedin early July.) We leave it to the experimental collabora-tions to determine the exact upper limits on solar axionsensuing from their results.If the coupling to photons is not zero, Fµ! F̃µ!a/(4fa"),

then we can calculate the counting rate, using the axionflux provided in Ref. [8]:

d!a

d#a= 6.02! 1030

#1 GeVfa!

$2!2.481a e!

"a1.205 (22)

!cm!2 s!1 keV!1.

Counting rate for the axio-electric e"ect is given by theproduct of the calculated absorption cross section andthe flux (22). For (fafa")1/2 normalized on 108 GeV, weget the counting rates plotted in Figure 4. Integrationover axion energy leads to the following total countingrates

RAr " 5.0

!108GeV

(fafa")1/2

"4

kg!1day!1, (23)

RGe " 5.2

!108GeV

(fafa")1/2

"4

kg!1day!1, (24)

RXe " 8.2

!108GeV

(fafa")1/2

"4

kg!1day!1. (25)

FIG. 4: Counting rate for the axio-electric e!ect for Ar, Geand Xe as a function of axion energy.

Comparing this to the counting rate of the CDMSexperiment [11], one can see that the equivalent of(fafa")1/2 # 108 GeV are being probed, as the count-ing rates in the window from 1.5 to 4 keV reachO(1 kg!1day!1keV!1). Similar sensitivity is achievedin the CoGent experiment [13].Finally, the axion flux can be created by the emission

of the axions due to the same interaction that leads toatomic ionization. In this case, however, the productioncross section is down by additional factor of E2

a/m2e [14],

and the sensitivity to fa in this case does not exceed 106

GeV.

V. CONCLUSIONS

QCD axions represent one of the most well-motivatedextensions of the Standard Model. Their light mass andsmall couplings allow them to be produced in the Solarinterior and escape reaching the Earth. With the prolif-eration of the low-background searches of dark matter,one should also conduct searches of solar axions. In thispaper we have calculated the cross sections relevant forthese searches, improving upon the simple scaling rela-tions that tie the axio-electric and photo-electric e"ects.Last two years has brought a significant progress in

sensitivity to any ionizing e"ects in Germanium in thewindow from 1 to 10 keV [11, 13]. Currently, the Co-Gent experiment has very low backgrounds in the windowfrom 2 to 4 keV, where the solar axion signal is expectedto peak. With acquiring more statistics, the sensitivity

Counting rates in the DM detectors can provide sensitivity to axion couplings complementary to e.g. CAST. Derevianko et al, 2010 Emission of other exotic light states and their signals at DM detectors need to be studied

27

New Baryonic Currents and “Semi-sterile” neutrinos

•  If there is a 4th neutrino, sterile under standard EW interactions, but very interactive via new baryonic currents unexpected phenomenological consequences show up:

1.  Signals at direct Dark Matter detectors at low recoil 2.  New “neutral-current-like” events at fixed targets/neutrino

beams 3.  New signatures at neutrino detectors 4.  ….

 

28

The model of “baryonic neutrino”

•  Consider a new “neutrino-like” particle coupled to baryonic currents:

At the nucleon level we have a isosinglet vector current: These properties suppress standard neutrino signals and

enhance the elastic recoil. Let us introduce an analogue of Fermi constant:

 

29

Oscillation of 8B neutrinos into νb

•  Suppose the mass matrix is such that some part of the solar neutrinos oscillate into νb.

At the Sun location we have (“+” is an appropriate mu-tau neutrino combination that participates in solar neutrino oscillations)

•  At Earth’s location one can easily have a more complicated mix:

 

30

Effective interaction and enhancement of elastic channels

How much signal you would have is given by Probability of oscillation * interaction strength Despite N being very large, say a 100 or a 1000, standard neutrino

detectors will have hard time detecting νb because nuclear excitations and deuteron breakup due to iso-singlet vector are extremely inefficient

For calculation of the neutron signal at SNO and C(4.4 MeV) signal at

Borexino, see, MP, 2011.

31

Fitting CoGeNT excess

 

const

!b

!b+const

L-shell bkg. subtracted

CoGeNT 442 days

!m2 = 1.76 ! 10!10 eV2

Ne! = 228

Ev [keVee]

even

ts/0.1

keV

ee

32.521.510.5

250

200

150

100

50

0

J. Pradler, MP, in preparation

32

Morphology of νb recoil signal •  Very similar to sub-10 GeV scale WIMPs. •  Somewhat softer at the highest recoil, hence “safer” from

strong Xe, Ge CDMS etc constraints where threshold is higher •  Has a chance of “explaining CoGeNT and/or CRESST

signals”. Can be a correct magnitude and not too bad a spectral shape.

•  Will show difference with the low-mass WIMPs if a lighter target (e.g. He) is used. Neutrinos will give more recoil on He, while WIMPs will give less.

•  What about “DAMA modulation signal”? After all, the Sun was closer to Earth in January – hence anti-modulation compared to DAMA

 

33

“Just-so” phase reversal •  If oscillation length is comparable to the Earth-Sun distance, the

phase can be reversed, and more neutrinos will arrive in July. νB Boron-8 neutrino spectrum with “just so” Δm

 

34

Fitting DAMA/Libra modulation

 

The magnitude can be fit well. The phase – not so much – 5 weeks off J. Pradler, MP, in preparation

!bDAMA 2010

!m2 = 2.52 ! 10!10 eV2

Ne! = 102

.

Ev [keVee]

modulation

amplitu

de(cpd/kg/keV

ee)

2015105

0.03

0.025

0.02

0.015

0.01

0.005

0

-0.005

-0.01

35

CRESST and Coupp

 WCaO

Ne! = 100

!m2 = 2.5! 10!10 eV2

dotted: hep neutrinossolid: 8B neutrinos

CRESST-II, 400 kg days

Er [keV]

even

ts/keV

2018161412108642

10000

1000

100

10

1

0.1

COUPP, 60 kg!1 year

Ne! = 100

!m2 = 2.5 ! 10!10 eV2

dotted: hep neutrinossolid: 8B neutrinos

Ethr [keV]

even

ts

30252015105

100000

10000

1000

100

10

1

0.1

0.01

“Baryonic neutrino” is a legitimate piece of new physics that can be searched with exactly the same instruments/types of signal.

36

Conclusions

 

1.  2nd missing mass problem – the nature of DM – is much harder than the 1st one. There is a variety of different ideas: WIMPs, super-WIMPs, super-cold DM and other possibilities I have not covered. Some of these models can be difficult to explore in experiments.

2.  WIMPs can come in different shapes and varieties. A generic modification – secluded WIMPs – may only have an indirect detection signatures. However, there is a new avenue for research emerging from the secluded WIMP and light WIMP ideas – search for light mediators at the intensity frontier.

3.  Superweakly interacting massive particles, axion-type, vector-type etc can be searched via its absorption signal in the existing DM detectors. So can be solar axions, furnishing complementary sensitivity to existing searches.

4.  New neutrino states, sterile under SM weak currents, but with enhanced interaction via baryonic current, can be considered as a legitimate search object, and mimic some/most of the current anomalies in direct DM detection.

37

Inelastic processes are suppressed

•  Even if coupling^2 is enhanced by 10000, the NCB process is just about 10% of the SM NC process at SNO:

 

38

Importance of different couplings for elastic and inelastic scattering

 

coupling   Inelas8c  sca9ering   Elas8c  sca9ering  

Isosinglet  vector  g_V^0          4  (looser)        1  (winner)  

Isovector  vector  g_V^1          2-­‐3        2  

Isosinglet  axial  g_A^0            2-­‐3    3-­‐4  

Isovector  axial  g_A^1          1  (winner)    3-­‐4  

If in SM isovector axial coupling would have been zero, there could not have been any SNO NC signal.

39

Possible avenues to search for neutrino_b and new baryonic currents

•  Hadron colliders: If GB /GF is fixed at a 100 or so, Tevatron experiments will produce an upper bound on vector mass.

•  Neutrino oscillations: Matter effects for (anti)neutrino_b can be significant. In light of latest developments in neutrino physics, the 4th one may not be an unwelcome addition.

•  Neutrino beams: Ample opportunities to produce neutrino_b in hadronic cascades and detect them using the “NC-like” scattering on nucleons.

•  Cosmology: a departure from N_neutrino = 3 is expected. Better CMB probes are forthcoming.

•  Rare decays: New precision tests of K-> pi nu nu may detect extra energy sinks.