Peter Fisher - MIT

Dark matter, electrons andanti-protons

Peter FisherOct. 29, 2005

Peter Fisher - MIT

Maybe one of the lastantiprotons observed at theBevatron

Peter Fisher - MIT

Outline

•The electron, positron, antiproton’sroles in the search for dark matter

•AMS-01- a first try

•A search for dark matter

•The AMS-02 - a bridge too far?

Peter Fisher - MIT

Origins of Dark Matter

Fritz Zwicky (1933) applied Virial Theoremto motions of “nebulea” (galaxies) in theComa cluster and showed the motionimplied gravitational potential far in excessof the amount of matter measured based onthe light output of the nebulea.

€

G t( ) =r p i •

r r i

i∑

€

dG t( )dt

= 0 = 2 T + V

€

MComa >35

Rv 2

G

€

Mgalaxy ~MComa

1000= 4.5 ×1010MSun

L ~ 8.5 ×107LS

→γ = 500

R~600 kpc,characteristic radius

v2=5 x 1015cm2/s, time& space averagedvelocity

Peter Fisher - MIT

Contemporarypicture:

Halo surroundingbaryonic disk

•May be largevariations of DMdensity

•May be bulkmotion of DM inhalo

•May be satellitehalos

Peter Fisher - MIT

Three methods of looking for evidence of dark matter

Nuclear recoilAnnihilation inour galaxy

Capture followedby annihilation

Recoiling nucleus with ~10 keVin matter Excess electrons, positrons ,

gammas or antiprotons incosmic rays

Neutrinos from the Earth, Sun orgalactic center

Peter Fisher - MIT

SuperKAMANDAICECube

AMS,HEAT,GLAST,

CAPRICE,PAMELA

CDMSCRESSTZEPLIN

Experiments

Flux integratedover lifetime of

galaxy

Flux in local3kpc now

Flux at Earth nowSampling

Partial suppressionfor Majorana

Majorana notsuppressed

Majoranasuppressed by

N2

Majorana/Diracsuppression

(g2g2q/M4)(g2g2

B/M4) g2g2B/M4g2g2

q/M4Rate

n2n2nDensity

Process

Capture andAnnihilation

AnnihilationScattering

g

gq

M2g gB

M2

+

Peter Fisher - MIT

Three methods of looking for evidence of dark matter

Nuclear recoilAnnihilation inour galaxy

Capture followedby annihilation

Recoiling nucleus with ~10 keVin matter

Excess electrons, positrons ,gammas or antiprotons incosmic rays

Neutrinos from the Earth, Sun orgalactic center

Peter Fisher - MIT

DM neutral current interactionsχ, p’ Neutral current:

Jµ=ψ(p’)(CVγµ-CAγµγ5)ψ(p)

J0=CVψ(p’)ψ(p)

€

dσdTR

=GF2Mc2

8πv 2N 2 exp −MTRR

2

3h2

J= CAψ(p’) S ψ(p)

VectorAxial-vector

p 0

Quarkcontent

€

dσdTR

=2GF

2

πh2TRmax µ2λ2J J +1( )

qΣTq3Δq

Nuclearphysics

Nuclearspin

Z0

χ, p Jµ

Dirac, scalarDirac,ScalarMajorana

Peter Fisher - MIT

DM neutral current interactionsDirac (Spin Independent, SI)•Elastic cross section proportional to N2 (~1600 for Ge)•Independent of nuclear spin•Simple nuclear physicsMajorana (Spin Dependent, SD)•No enhancement from coherence•Proportional to J(J+1)•Complicated nuclear physics, QCDIf a recoil signal is observed, do not know if it is from SI orSD, sSI~N2sSD

The annihilation channel is more complex (depends onneutral scalars) and Majorana is not obviously suppressedrelative to Dirac.

Peter Fisher - MIT

γ

e+/e-p/pχDM

χDM

W+

W-

q

q

Fragmentation anddecay

Typical processes:

€

νDM + νDM → W + + W −

νDM + νDM → Z 0 + Z 0

νDM + νDM → b + b

Peter Fisher - MIT

Galprop - I.Moskolenko and A.Strong - cosmic raypropagationproblem fit to all CRknown data

Green’s functions -expected flux onEarth for uniformmonoenergeticsource of electrons

Notnormalized!

Propagation

Peter Fisher - MIT

Charged particles followmagnetic field lines

€

rL =p

0.3 GeVT −m

B≈ 7AU

€

v = c

€

v|| = c3

Peter Fisher - MIT

€

r B o

€

r ′ B

Magnetic turbulence - averagevariation of magnetic field:

€

η =

r ′ B

r B o +

r ′ B ≈10−4

Mean time between scatteringfrom inhomogenieties:

€

τ s =1

ηωL

≈10y

Peter Fisher - MIT

Electron lifetime determined by time τo topropagate one Xo=65 g/cm2 in hydrogen

For a proton, the relevant scale isλN=51g/cm2

€

v = c

€

v|| = c3

1 proton/cm3 in ISM Xo=1.3 x 1013 kpc

τo=45 My

30 GeV electron:v=c, gives averagevelocity along fieldc/31/2

€

⇒

€

⇒

Peter Fisher - MIT

Number of scatterings: N=τo/τs

Random walk diffusion distance

€

d = v|| τ s N =13c 2τ sτ o ≈ 3kpc

Advanceeach step RMS

number ofsteps

Diffusioncoefficient

Peter Fisher - MIT

Integrated positron signalabove 8 GeV for 100 GeV(solid line) and 30 GeV(dotted line). The Earth islocated at 8.5 kpc radius.

Earth

Contribution of DMoutside of plane ofgalaxy difficult tounderstand - magneticfield structure notwell known

Peter Fisher - MIT

Propagation makes amess!

During 3 kpc transit,DM annihilationproducts go through~1 Xo or 1 λN ofmaterial

e+/e- spectrum atpoint ofannihilation

e+/e-

spectrumat Earth

Peter Fisher - MIT

“Typical” signal - neutralino annihilation

Moral - in cosmic rays everything looks like

€

dNdE

∝ E −(2.5 to 3.2)

Peter Fisher - MIT

For a search for dark matter annihilation products, oneshould measure the spectra of

•Electrons - sharp cutoff from prompt decays in somemodels

•Positrons - low background rate, high signal rate, sharpcutoff from prompt decays in some models

•Antiprotons - low background rate, moderate signal rate

•Photons - directionality, but possibly large backgrounds

Ideally, one would have a single detector for all four andthen carry out a simultaneous fit to a series of models.

HEAT and AMS-01 have begun with electrons and positrons.

There are interesting results from HEAT…

Peter Fisher - MIT

This first problem is that thepropagation (especially solarmodulation) is not wellunderstood…

…the second is that HEATruns out of sensitivity at ~50GeV…

Fit with expected (smooth)normalization with signal (bump)gives 55 times higher relicdensity than observed

Peter Fisher - MIT

AMS-01 - June 1998

•~100 h data taking at 400 km

•200 M triggers

•6 plane silicon tracker in 1500 Gfield

•4 plane TOF system

•Aerogel thershold Cerenkovcounter (~5 GeV)

Peter Fisher - MIT

AMS Collaboration

Peter Fisher - MIT

AMS-01 had no particle ID above a few GeV, so

•Select clean Z=-1 events

•Compute all Z=-1 backgrounds from expected CR source

•Use PYTHIA to compute signal spectra for electrons andanitprotons from W, Z and b decays at center of massenergies from 100-1000 GeV (Note: we do not yet considerSUSY models; our limits will be applicable for any processwhich has ZZ, WW or bb final states.

•GALPROP finds signal spectra at Earth

•Signal and background fit to data

How well can this work?

Peter Fisher - MIT

Mar.2008

(Hah!)

FlewFlewStatus

8-240.4-1.51FOM forDM e-

3,000360170MDR (GV)

26,00023945Exposure(h)

0.50.140.05Aperture(m2-str)

AMS-02AMS-01HEAT

€

FOM = 0.1 to 0.01( ) Ω0.05m2 − str

τ45h

Peter Fisher - MIT

AMS-01MDRHeat

MDR

HighestHEATDatapoint

Preliminary AMS-01 Z=-1selection:

•Downward going

•|Q|=1 from both trackerand TOF

•Well fit track with 4 hits

•Not docked to MIR, not overSAA

•Good match between TOFand track

A major difficulty is mis-reconstructed protons in theZ=-1 signal. This backgroundis calculated by Monte Carlo(200 M events)

Peter Fisher - MIT

The primary background is mis-reconstructed protons (shown in red), whichdominates for p>50 GeV. Owing to this, thefit is largely determined by the slope atlower energy rather than the spectral cutoffat higher energy.

Fit for the normalization of•Expected electrons•Expected antiprotons•DM signal (electrons, antiprotons)from W+W- fragmentation

Troublehere!

Peter Fisher - MIT

Not a result! Expectedsensitivity will lie in thisrange

Taking the 90% c.l.upper limit for the DMnormalization from thefit gives the exclusionplot shown. The limitis in an interestingregion.

Clearly, more work isneeded to understandthe instrumentaleffects.

Need something better!

Peter Fisher - MIT

Improvement requires•Much higher statistics - 3 years inorbit, 3x larger aperture•Higher momentum reach - 7x strongermagnet, two more tracker planes•Particle ID

AMS-02

See: http://ams.cern.ch

Peter Fisher - MIT

AMS-02 is under construction. All major subsystems arecomplete or will be complete by Jan. 2006, at which timeintegration will start.

20 layer Xe:CO2transition Radiation

Detector

0.7 Tsuperconductingmagnet

8 plane silicontracker

Peter Fisher - MIT

AMS is currently in the NASA manifest for UF-4 which will bein 18 flights. The President’s space exploration visionforesees retiring the shuttle in 2010. You do the math…

…all options are being explored.

Peter Fisher - MIT

Summary•Elucidation of the dark matter problem will eventuallyrequire detection or very stringent limits on annihilation inour galaxy

•Detecting the decay products is a tough problem:

•High statistics requires a large detector (~1 m2-str),long duration (months)

•Backgrounds are high

•Propagation effects are tricky to assess

•Complex instrument needed for electrons, positrons,protons and anti-protons

For myself, I’m thinking about nuclear recoils