dark matter in the universe katherine freese michigan center for theoretical physics university of...
TRANSCRIPT
Dark Matter in the Universe
Katherine FreeseMichigan Center for Theoretical PhysicsUniversity of Michigan
OUTLINE
• I) Review of Evidence for Dark Matter• II) Dark Matter Detection: Are any of three claimed
detections right?
DAMA, HEAT, gamma-rays from galactic center• Effect of mass distribution in Halo on detection:
Sagittarius stream can be a smoking gun for WIMP detection
• III) DARK STARS: dark matter in the first stars produces a new phase of stellar evolution!
The Dark Matter Problem:
• 90% of the mass in galaxies and clusters of galaxies are made of an unknown dark matter component
Known from: rotation curves,
gravitational lensing,
hot gas in clusters.
Solar System Rotation Curve
Average Speeds of the Planets
As you move out from the Sun, speeds of the planets drop.
Tyco Brahe(1546-1601)
Lost his nose in a duel, and wore a gold and silver replacement.
Studied planetary orbits.
Died of a burst bladder at a dinner with the king.
Speed is determined by Mass
The speed at distance r from the center of the galaxy is determined by the mass interior to that radius. Larger mass causes faster orbits.
€
GM(r)
95% of the matter in galaxies is unknown dark matter
• Rotation Curves of Galaxies:
EXPECTEDFROM STARS
OBSERVED:FLAT ROTATIONCURVE
Hot Gas in Clusters: The Coma Cluster
Optical Image X-ray Image from the ROSAT satellite
Without dark matter, the hot gas would evaporate.
Dark Matter Candidates
• MACHOs (massive compact halo objects)
• WIMPs
• Axions
• Neutrinos (too light)
• Primordial black holes
• WIMPzillas
• Kaluza Klein particles
The Dark Matter is NOT
• Diffuse Hot Gas (would produce x-rays)• Cool Neutral Hydrogen (see in quasar absorption
lines)• Small lumps or snowballs of hydrogen (would
evaporate)• Rocks or Dust (high metallicity)
(Hegyi and Olive 1986)
MACHOS(Massive Compact Halo
Objects) Faint starsFaint stars
Planetary Objects (Brown Dwarfs)Planetary Objects (Brown Dwarfs) Stellar Remnants:Stellar Remnants: White DwarfsWhite Dwarfs Neutron StarsNeutron Stars Black HolesBlack Holes
Is Dark Matter Made of Stars? NO!
• Faint Stars: Hubble Space Telescope
• Planetary Objects:
parallax data
microlensing experiments
Together, these objects make up less than 3% of the mass of the Milky Way.
(Graff and Freese 96)
Is Dark Matter made of Stellar Remnants (white dwarfs, neutron
stars, black holes)? partly• Their progenitors overproduce infrared radiation.• Their progenitors overproduce element abundances (C, N,
He)• Enormous mass budget.• Requires extreme properties to make them.• NONE of the expected signatures of a stellar remnant
population is found.
• AT MOST 20% OF THE HALO CAN BE MADE OF STELLAR REMNANTS
[Fields, Freese, and Graff (ApJ 2000, New Astron. 1998); Graff,
KF, Walker and Pinsonneault (ApJ Lett. 1999)]
Candidate MACHO microlensing event in M87 (giant elliptical galaxy in VIRGO
cluster, 14 Mpc away)
QuickTime™ and aTIFF (LZW) decompressor
are needed to see this picture.
Baltz, Gondolo, and Shara (ApJ 2004)
Consistent withMACHO data in Milky Way (to LMC)
Good news: cosmologists don't need to "invent" new
particle:• Weakly Interacting
Massive Particles (WIMPS). e.g.,neutralinos
• Axions
ma~10-(3-6) eV
arises in Peccei-Quinn
solution to strong-CP
problem
v
cmhχ
sec/103 3272 ×≈Ω
AXION BOUNDS from ADMXRF cavity experiment (1998)
QuickTime™ and aTIFF (LZW) decompressor
are needed to see this picture.
Overall status of axion bounds
QuickTime™ and aTIFF (LZW) decompressor
are needed to see this picture.
Inflation with the QCD axion
QuickTime™ and aTIFF (LZW) decompressor
are needed to see this picture.
V (a) = V0[1
Chain Inflate:Tunnel from higher to lower minimum in stages, with a fraction of an efold at each stage
Freese, Liu, and Spolyar (2005)
WIMPsWIMPsRelic Density: Ωχh2≈( 3x10-26 cm3/sec / χχsm)
vh
13272 scm103⋅≈Ω
v Annihilation cross section χχsm of Weak Interaction strength gives right answer
Prospects for detection:
Detection
direct
indirect
Neutrinosfrom sun/earth
anomalouscosmic rays
WIMP candidate motivated by SUSY:Lightest Neutralino, LSP in MSSM
Supersymmetry
• Particle theory designed to keep particle masses at the right values
• Every particle we know has a partner: photon photino quark squark electron selectron• The lightest supersymmetric partner is a
dark matter candidate.
Lightest Super SymmetricParticle: neutralino
• Most popular dark matter candidate.• Mass 1Gev-10TeV
(canonical value 100GeV)• Majorana particles: they are their own antiparticles and
thus annihilate with themselves• Annihilation rate in the early universe determines the
density today.• The annihilation rate comes purely from particle physics
and automatically gives the right answer for the relic density!
Dark Matter Annihilation
• Annihilation mediated by weak interaction.
• Thus for the standard neutralino (WIMPS):
• On going searches: LHC, CDMS XENON, GLAST, ICECUBE
€
Ωχh2 =
3×10−27 cm3 /sec
<σv>ann
SUSY dark matter
QuickTime™ and aTIFF (LZW) decompressor
are needed to see this picture.
Coannihilation included (Binetruy, Girardi, and Salati 1984);Griest and Seckel 1991)
Detecting Dark Matter Particles
• Accelerators• Direct Detection• Indirect Detection (Neutrinos)
• Sun (Silk,Olive,Srednicki ‘85)• Earth (Freese ‘86; Krauss, Srednicki, Wilczek ‘86)
• Indirect Detection (Gamma Rays, positrons)• Milky Way Halo (Ellis, KF et al ‘87)• Galactic Center (Gondolo and Silk 2000)• Anomalous signals seen in HEAT (e+), HESS,
CANGAROO, WMAP, EGRET, etc.
Detection of WIMP dark matter
A WIMP in the Galaxy travels through our detectors. It hits a nucleus, and depositsa tiny amount of energy. The nucleus recoils, and we detectthis energy deposit.
Expected Rate: less than one count/kg/day!
Three claims of WIMP dark matter detection: how can we
be sure?
• 1) The DAMA annual modulation• 2) The HEAT positron excess• 3) Gamma-rays from Galactic Center
HAS DARK MATTER BEEN
DISCOVERED?
DAMA annual modulationDrukier, Freese, and Spergel (PRD 1986); Freese, Frieman, and Gould (PRD 1988)
Data do show a 6 modulation
WIMP interpretation is controversial
Bernabei et al 2003
Event rate
€
dR
dE=
NT
MT
×dσ
dE× nv f (v, t)d3v∫
(number of events)/(kg of detector)/(keV of recoil energy)
€
=ρ 0F
2(q)
2mμ 2
f (v, t)
vd3v
v> ME / 2μ 2∫
Spin-independent
Spin-dependent
€
0 =A2μ 2
μ p2
σ p
€
0 =4μ 2
πSp Gp + Sn Gn
2
DAMA: spin-independent?
Spin-independent cross section with canonical Maxwellian halo is excluded by CDMS-II (2004)
EDELWEISS
DAMA
CDMS-IIBaltz&GondoloBottino et al
Positron excess
• HEAT balloon found anomaly in cosmic ray positron flux
• Explanation 1: dark matter annihilation
• Explanation 2: we do not understand cosmic ray propagation
• Upcoming: PAMELA data release, June 2008
Baltz, Edsjo, Freese, Gondolo 2001
Neutralinos at Galactic Center?
• Compact source: 0.01 - 1 pc• J~10 or 103
• Compatible with mass from stellar motions
Ghez et al 2003 Genzel et al 2003
mSUGRA studyHall, Baltz, Gondolo 2004
WMAP microwave emission interpreted as dark matter
annihilation in inner galaxy?
QuickTime™ and aTIFF (LZW) decompressor
are needed to see this picture.
Finkbeiner 2005
Consistent with 100 GeV WIMPs.
Gamma-ray line
• Characteristic of WIMP annihilation• Need good energy resolution
GLAST Simulation
€
χχ → γγ• GLAST may do it below ~80 GeV
Bergstrom, Ullio and Buckley 1998
Three Claims of WIMP detection
• The DAMA annual modulation
• The HEAT positron excess
• Gamma-rays from Galactic Center
Upcoming Data
• LHC (find SUSY)
• Indirect Detection due to annihilation:• GLAST first light May 27, 2008 (gamma rays)• PAMELA (positrons)• ICECUBE (neutrinos)
• Direct Detection: CDMS, XENON, WARP, CRESST, ZEPLIN, COUPP, KIMS …
Streams of WIMPs
• For example, leading tidal stream of Sagittarius dwarf galaxy may pass through Solar SystemMajewski et al 2003, Newberg et al 2003
• Dark matter density in stream ~ Freese, Gondolo, Newberg 2003• New annual modulation of rate and
endpoint energy; difficult to mimic with lab effects
Freese, Gondolo, Newberg, Lewis 2003
€
0.01−0.01+0.20 ρ local
Sagittarius stream
Plot for 20% Sgr stream density (to make effect visible); χp=2.7x10-42cm2
Rate modulation
Endpoint energy modulation
Sagittarius stream
• Increases countrate in detectors up to cutoff in energy spectrum
• Cutoff location moves in time• Sticks out like a sore thumb in
directional detectors• Changes date of peak in annual
modulation• Smoking gun for WIMP detection
Dark Stars: Dark Matter Annihilation in the
First Stars leads to a new phase of stellar evolution.
Katherine Freese
arXiv:0802.1724K. Freese, D. Spolyar, and A. Aguirre
Phys. Rev. Lett. 98, 010001 (2008) D. Spolyar , K .Freese, and P. Gondolo
Also with P. Bodenheimer and N. Yoshida
Our Results
• Dark Matter (DM) in proto-stellar haloes can dramatically alter the formation of the first stars
• The LSP (lightest supersymmetric particle) provides a heat source that prevents the protostar from further collapse, leading to a new stellar phase:
• The first stars in the universe are giant (> 1 A.U.) H/He stars powered by dark matter annihilation rather than by fusion
Basic Picture
• The first stars form in a DM rich environment• As the gas cools and collapses to form the
first stars, the cloud pulls DM in as it collapses.
• DM annihilates and annihilates more rapidly as its densities increase
• At a high enough DM density, the DM heating overwhelms any cooling mechanisms which stops the cloud from continuing to cool and collapse.
Basic Picture Continued
• Thus a gas cloud forms which is supported by DM annihilation
• More DM and gas accretes onto the initial core which potentially leads to a very massive gas cloud supported by DM annihilation.
• If it were fusion, we would call it a star.• Hence the name Dark Star
• The First Stars- standard picture • Dark Matter
• The LSP (lightest SUSY particle) • Density Profile
• DM annihilation: a heat source that overwhelms cooling in Pop III star formation
• Outcome: A new stellar phase• Observable consequences
The First Stars
• Important for:• End of Dark Ages.• Reionize the universe.• Provide enriched gas for later stellar
generations.• May be precursors to black holes which power
quasars.
First Stars: Standard Picture
• Formation Basics:– At z = 10-50
– Form inside DM haloes of (105-106) M
– Baryons initially only 15%
The dominant cooling Mechanism is H2
Not a very good coolant
(Hollenbach and McKee ‘79)
Thermal evolution of a primordial gas
adiabatic contraction
H2 formationline cooling (NLTE)
loitering(~LTE)
3-bodyreactio
n
Heatreleas
e
opaque tomolecular
line
collisioninducedemissionT [K]
104
103
102
number density
opaqueto cont.
anddissociation
adiabatic phaseMust be cool to collapse!
Scales
• Jeans Mass ~ 1000 M
at
• Central Core Mass (requires cooling)
accretion
Final stellar Mass??
in standard picture
€
n ≈104cm−3
Dark Matter + Pop III Stars
• Dark Matter annihilation heats the collapsing gas cloud preventing further collapse, which halts the march toward the main sequence.- A “Dark Star” may result forming
(a new Stellar phase)
Dark Matter Annihilation
• Annihilation mediated by weak interaction.
• Thus for the standard neutralino (WIMPS):
• On going searches: LHC, CDMS XENON, GLAST, ICECUBE
€
Ωχh2 =
3×10−27 cm3 /sec
<σv>ann
Dark Matter
Our Canonical Case:
Minimal supergravity (SUGRA)– Mass 50GeV-2TeV– can be an order of magnitude bigger
Nonthermal relics– can be much larger!
€
<v>ann =3×10−26cm3 /sec
€
Mχ =100GeV
€
<v>ann
€
<v>ann
Dark Matter
• We consider a range:– Mass: 1GeV-10TeV– a range of Cross sections
• Results apply to other candidates– Sterile – K-K particles
Dark Matter Density Profile
• Annihilation rate proportional to square of dark matter density
• We need to know how much dark matter is inside the star: what is the DM profile?
Hierarchical Structure Formation
Smallest objects form first
Pop III stars (105 M)
Merge galaxies
Merge clusters etc.
Numerical Simulations
• NFW Profile (Navarro,Frank,white ‘96)
€
ρ(r)= ρο
rrs
(1+ rrs
)2
€
ρο=
€
ρ(rs)= 1/4 ρο
€
rs= “Scale Radius”
“Central Density”
Other Variables
• We can exchange
• radius at which
€
ρο, rs → Mvir, Cvir
€
Rvir
€
ρDM = 200 × The DM density of the universe at
the time of formation.€
Cvir =Rvir
rs
€
Mvir =2004π3
Rvir3 ρcrit(z)
Dark Matter Density Profile
• Adiabatic contraction (a prescription):– As baryons fall into core, DM particles
respond to potential well.
• Profile
that we find: €
r M(r) = constant
€
ρχ(r)=r−1.9 Outside Core
€
ρχ(n)=5 GeV (n /cm−3)0.8
(using prescription from Blumenthal, Faber, Flores, and Primack ‘86)
DM profile and Gas
Z=20 Cvir=2 M=7x105 M
Blue: Original NFW Profile
Gas ProfileEnvelope
Black: 1016 cm-3
Green: 1010 cm-3
Red: 1013 cm-3
Gas densities:
ABN 2002
How accurate is Blumenthal method for DM density profile?
• Work with Jerry Sellwood (in prep):• There exist three adiabatic invariants. • In spherical case, one is zero. Blumenthal
method conserves only angular momentum; takes into account only circular orbits. With Sellwood, also include radial orbits and invariant. Find that our naivest results were only high by 25%! Adiabatic conraction works where it really shouldn’t.
Dark Matter Heating
Heating rate:
Fraction of annihilation energy
deposited in the gas:
Previous work noted that at
annihilation products simply escape
(Ripamonti,Mapelli,Ferrara 07)
€
Qann =nχ2 <σv>× mχ
€
=ρχ
2 <σv>mχ
€
ΓDMHeating = fQ Qann
€
fQ :1/3 electrons
1/3 photons
1/3 neutrinos€
n≤104cm−3
Crucial Transition
• At sufficiently high densities, most of the annihilation energy is trapped inside the core and heats it up
• When:
• The DM heating dominates over all cooling mechanisms, impeding the further collapse of the core
€
mχ ≈1 GeV
€
mχ ≈100 GeV
€
mχ ≈10 TeV €
n ≈109 /cm3
€
n ≈1013 /cm3
€
n ≈1015−16 /cm3
DM Heating dominates over cooling when the red lines cross the blue/green lines (standard evolutionary tracks from
simulations). Then heating impedes further collapse.
€
<v>ann =3×10−26cm3 /sec
€
ΓDM ∝<σv>mχ
New Stellar Phase
• “Dark Star” supported by DM annihilation rather than fusion
• They are giant diffuse stars that fill Earth’s orbit
• THE POWER OF DARKNESS: DM is only 2% of the mass of the star but provides the heat source
• Dark stars are made of DM but are not dark: they do shine, although they’re cooler than standard
early stars.
€
mχ ≈1 GeV
€
core radius 17 a.u.
Mass 11 M
€
mχ ≈100 GeV
€
core radius 960 a.u.
Mass 0.6 M
Luminosity 140 solar
Key Question: Lifetime of Dark Stars
• How long does it take the DM in the core to annihilate away?
• For example for our canonical case:
• v.s. dynamical time of <103yr: the core may fill in with DM again s.t. annihilation heating
continues for a longer time Are there still some dark stars around today?
€
Te ≈mχ
ρχ <σv>
€
Te ≈600 million years for n≈1013cm−3
Possible effects
• Reionization: Delayed due to later formation of Pop III stars? Can study with upcoming measurements of 21 cm line.
• Solve big early black hole problem.– Massive quasars at high z
Observables
• Dark stars are giant objects with core radii > 1 a.u.– Find them with JWST: NASA’s 4 billion dollar sequel to HST plans to see the first
stars and should be able to differentiate standard fusion driven ones from dark stars, which will be cooler
annihilation products in AMANDA or ICECUBE? Fluxes too low, angular resolution inadequate in current detectors. Someday.
• Can neutralinos be discovered via dark stars or can we learn more about their properties?
DARK STARs (in conclusion)
• The first Stars live in a DM rich environment.
• DM annihilation heating in Pop III protostars can delay/block their production.
• A new stellar phase DARK STARSDriven by DM annihilating and not by
fusion!
What happens next?
• Outer material accretes onto core– Accretion shock
• Once T~106 K:– Deuterium burning, pp chain, Helmholz
contraction, CNO cycle.
• Star reaches main sequence– Pop III star formation is delayed.
Next stage
• Even once/if first star reaches main sequence and has fusion in core, DM annihilation can be very important.
• Can be dominant heat source• Can determine the mass of the stars
(Freese, Spolyar, Aguirre Feb. 08;Iocco Feb. 08)
Basic Idea
• Dark star phase has ended• Next, fusion powered star forms: on main sequence• New star captures DM• Capture rate extremely high, which leads to a very large
luminosity from the DM. • How big?
All first stars
1 M Fixes mass of the first stars or limits DM scattering cross section.
Capture Rate (Particle Physics)
• Scattering – Consider only SD scattering for first stars, which are made
only of H and He. SI scattering is generally subdominant.
• and
• DM luminosity LDM generally independent of DM mass and annihilation rate!
mχ= 100 GeV <v>ann=3x10-26 cm3/s
Limits from Super-K
Capture Rate (Astrophysics)
• Typical Mass of first stars– Up to 103 M (Jeans Mass)
• First stars DM host halo
• DM Velocity – Much slower than Milky Way since typical host halo is much smaller
• DM density
M ~ (10 to 250) M
MDM~106 M
Simulations: 108 GeV/cm3
Adiabatic Contraction: 1018 Gev/cm3(Blumenthal, Faber, Flores, Primack 1987)
(Abel, Bryan, Norman 2002)
Capture Rate per Unit Volume
• nχ (number density of DM) cm-3
• n (number density of H) cm-3
• V(r) escape velocity at a point r
• velocity of the DM
We can neglect the term in the brackets because the DM velocity
is much less than the escape velocity for the first stars, which
makes B big.
If the star moves relative to the DM halo, the term in the brackets changes. Luckily, we can still neglect the term.
• c scattering cross section
Press, Spergel 85 & Gould 88
Capture Rate in the First Stars
Capture Rate s-1
1 Assume constant DM density
2 Conservatively fix v(r) to the escape from surface of star (vesc).
3 Integrate nH giving the number of H in the star, whichproduces the second term in parentheses on the RHS below.
H fraction (fH) DM density (ρχ)
Proton mass (Mp ) DM mass (mχ)
DM Luminosity (LDM)
• Fraction of Energy deposited (f)- we assume a third goes into neutrinos so we take f = (2/3).
• Equilibrium between the annihilation and Capture is very short
Independent of the mass of particle!
Comparison of Luminosity
Black line- LEdd
Green line-LDM
If LDM exceeds Ledd, stellar mass is fixed!
Blue line-L(fitted)
Red () zero metallicity stars on MS (Heger &Woosley)
10M
50M
100M
250M
70M
LDM versus L
• We compare the DM luminosity against fusion luminosity of zero metallicity stars half way through H burning (on the main sequence) for various masses. (Heger,Woosley, in prep)– H burning represents the largest fraction of a star’s life.
• DM luminosity wins for a sufficiently high DM density.
Similar and Simultaneous Work
• Within a few days of each other, we and Fabio Iocco posted the same basic idea: – Both groups found that the DM luminosity can be larger than
fusion for the first stars.
• We went one step further:– We found that when the DM luminosity exceeds the
Eddington luminosity, we can uniquely fix the mass of the first stars.
Eddington Luminosity
•Luminosity of a star at which the radiation pressure overwhelms the gravitational force.
C speed of light
p opacity
G Newton's ConstantM Stellar Mass
•Assume opacity (mean free path)is dominated by Thompson scattering since the surface of first stars is hot.
Max Stellar Mass
• Once LDM exceeds LEdd, star cannot accrete any more matter and mass is determined.• We can solve for the max stellar mass by setting LEdd=LDM.
• We derived the above by fixing v=10 km/s • And also noting that (Vesc)2 (M)0.55,
which follows since R (M)0.45
_ We find that the M and c are inversely related for a fixed
DM density.
Conclusions too
• Even if the Dark Star phase is short.• DM Capture can still alter the first stars.
– DM luminosity larger than fusion Luminosity
– May uniquely determine Mass of first stars
• Conversely- the first stars may offer the best bounds or opportunity to measure the the scattering cross section of DM.
Max Mass
10-38 cm2 10-40 cm2 10-43 cm2
Lines correspond to fixed scattering cross section. We show the relationship between the DM density and mass of the first stars.
Max Mass Example
Lines correspond to fixed cross section. We show the relationshipbetween the DM density and mass of the first stars.
M=10M
ρχ=1014 Gev/cm3
10-38 cm2 10-40 cm2 10-43 cm2
First Stars: Limits on SD ScatteringLimits with Stellar Mass Larger than 1M
5x1012
(GeV/cm3)
5x1015
(GeV/cm3)
1018
(GeV/cm3)Xenon SI
Zeplin SI
Super K
(SD limits examined in Savage, Freese, Gondolo 2005)
First Stars also limit SI Scattering
Super K
Limits with Stellar Mass Larger than 1M
5x1012
(GeV/cm3)
5x1015
(GeV/cm3)
1018
(GeV/cm3)Xenon SI
CDMSII SI
Zeplin SI
Present bounds from DMTools