(direct) Dark Matter Searches and XMASS experiment DM search: status, prospects and difficulties for large scale experiments XMASS experiment S. Moriyama Kamioka Observatory Institute for Cosmic Ray Resea March 3 rd 2007, KEKTC7
(direct) Dark Matter Searches andXMASS experiment
DM search: status, prospects and difficulties for large scale
experimentsXMASS experiment
S. MoriyamaKamioka ObservatoryInstitute for Cosmic Ray ResearchMarch 3rd 2007, KEKTC7
A lot of evidences on the existence of DM1. Orbital velocities of galaxies in clusters2. Rotational speed of galaxies3. Gravitational lensing of background objects by
galaxy clusters4. Temperature distributions of hot gas in galaxies
and clusters of galaxies5. CMB power spectrum analysis in WMAP6. The Bullet cluster7. Large scale map for dark matter8. …
Dark Matter in the Universe A lot of evidences on the existence of DM
WMAP Cosmic Microwave Backgroundpower spectrum decomposed matter, cold dark matter, and dark energy NASA
The Bullet cluster: NASARed: ordinary matter from X-ray Blue: mass from gravitational lensing
All of them are astrophysical observations. “Direct detection” is an urgent issue now.
WIMP Dark Matter
Particles thermally generated at high temperature decouple when the expansion rate ~ the interaction rate.
Weakly interacting, heavy, neutral, and stable particles can be CDM.
(WIMPs != SUSY here)E.W. Kolb and M.S. Turner, The Early Universe
X=m/Temperature (time )
Com
ovin
g nu
mbe
r de
nsity
Nequillibrium
Increasing<Av>
Dark Matter in the Milky Way
Also, the Milky Way (our galaxy) has a large amount of Dark Matter.
R.P.Olling and M.R.Merrifield MNRAS 311, 369- (2000)
Buldge
Steller disk
Dark Halo
Motion of WIMPs around us
Earth
Sun
Annual, sidereal modulation
DM density ~0.3GeV/cc100GeV WIMPs 1 WIMP / 7cm cubic, =105/cm2/sec
The solar system is moving at ~230km/sec.The earth is moving at ~30km/sec.The earth is rotating.
WIMPs are randomlymoving with Maxwelldistribution <v>~270km/sec
“Wind” of Dark Matter
(direct) Detection method Since they are neutral and stable, what we
can expect is only a collision with ordinary matter.
Electron recoil does not give enough energy but nuclear recoil gives ~100keV if mDM~O(100GeV).
Dark Matterparticles
Energy deposit
What is the best target?Recoil spectrum (spin independent)
Ge and Xe, and Ar give similar rates. Annual modulation should be seen.
Si
Ge
XeSi
XeGe
Red: differential, Blue: integrated
Target Xe
Assuming quenchingfactor of 0.2
June
Dec.
R.J.Gaitskell, Ann. Rev. Part. Sci., 54 (2004) 315.
WIMP mass 100GeV
Diff.~6%
Candidate: Neutrarino in the Supersymmetric Model
SUSY solves the hierarchy problem, stability of boson mass, and it can predict the gauge coupling unification. The prime paradigm beyond the standard model.
Even if SUSY particles are discovered in LHC, direct detection of DM should be done INDEPENDENTLY.
SM particles SUSY particles
Quark
Lepton
Higgs
Gauge particles
uu dd cc ss tt bb
ee ee
gg WW ZZ
uu dd cc ss tt bb~ ~ ~ ~ ~ ~
ee ee ~ ~ ~ ~ ~ ~
HH++ HH-- HH1100 HH22
00~ ~ ~ ~
gg WW ZZ00~ ~ ~ ~
AA00HH++ HH-- HH1100 HH22
00
Neutralino
(1pb = 10-36cm2 )
Expected cross section to nucleon
Large space of SI for proton.
SNOWMASS benchmark, many of them: 10-
8-10-10pb
General MSSM
10-6
10-12
SI f
or p
roto
n (p
b)
100 400 700 M(GeV)
Y.G.Kim et al. 2002
10-8
10-10
Mission in the real world
Once we make detector, we have 10~100Hz background by ambient gamma rays and internal radioactive contaminations, and neutrons.
What we need to realize is “1count/100day detector”.
108-9 reduction needed
DM Searches: past
NaI(Tl) at Gran SassoExposure 107 ton dayAnnual modulation over 7 years
6.3sigma C.L. annual modulation2-6keV data:0.02000+/-0.0032 cpd/kg/keVT0=140+/-22 days (exp. 152.5th day)T=1.00+/-0.01 year
No modulation: 7x10-4 (2/d.o.f.=71/37)
DAMA: NaI(Tl) scintillator
LIBRA experiment will reject/confirm the result.
astro-ph/0307403
DM searches: recent, CDMS (Ge)
Blind analysis done. World best limit obtained: ~O(kg)
4x250g Ge2x100g Si
BG from, e: EQ~EP
Signal : EQ~0.3EP
Ionaization EQ
Bolometric EP
Rise time of Ep, timing diff. btw EP & EQ
1st run Oct.03-Jan.04 2nd run Feb.04-Aug.04
Recoil energy (keV) Recoil energy (keV)
BG band
Signal band~E
Q/E
P
CDMS to SuperCDMS:25kg seems to be straight forward
DM searches: XENON (2 phase LXe)
XENON10: 10kg target first results soon! XENON100: 100kg target 2007-2009 construction
M. Yamashita IDM2006
Liquid xenon scintillator with charge coll.
Rare gas liquid: scale up, purification easy
DM searches: WARP/ArDM (Liquid Ar)
P.Benettiastro-ph/0701286
S2/S1 + difference of scintillation waveformfor e and nuclear recoil e n
Integrated waveforms
1s
L
og(S
2/S
1)
n-lik
e
e-lik
e
Pulse shepe parameter e-like n-like
CDMSII
WARPNeed to remove 39Ar (0.8Bq/kg)100 liter detector soon?
Future: larger scale Exp.
General MSSM
10-6
10-8
10-10
10-12
SI f
or p
roto
n (p
b)
100 400 700 M(GeV)
Y.G.Kim et al. 2002
B. SadouletKEKTC6
~4events/100ton/year10-12 pb is really
challenging!!
~4events/100kg/yearXe recoil E >25keV
+5-10year 10-9 ~ -10pb
CDMSII (current)Sorry for approximate line
2007 CDMS 10-7pbDAMA
2003 DAMA 10-6pb
10-12pb detection requires Super-K @ keV
Super-K
KamLAND (>0.8MeV)
Heidelberg Moscow
Kamioka Ge
ZEPLIN before PSD cut
DM signal for LXe100GeV 10-6pbQF=0.2
DAMANaI
Eve
nts/
kg/k
eV/d
ay
BG requiredFor 10-12pb
CDMSII
Atm, solar neutrino NC nuclear recoil appear
Difficulties appear ~10-9-10-10pb Scale-up is always accompanied with new bac
kground sources which are caused by:1. Low E threshold2. Surface effect3. Impurities in the targets4. Insufficient rej. eff.5. Neutron background
Small signals Surface & rays 39Ar, 85Kr Large # of BG >108
muon, parts inside
neutron
?
Target (Ge, LAr, LXe)Neutron BG will bedominated <10-9pb~Full target schemei.e. Ge, XENON, WARPsingle neutron eventscannot be discriminated
Neutron background
neutrone,
Strategy for larger target, high sensitivity
XMASS!
Neutron shielding by large amount of target itself.
Currently, “gamma rejection” is extensively studied. CDMS, XENON, and WARP have good discriminations with multiple information.
However, <10-9pb, neutron will be most serious BG. It cannot be discriminated. N. Spooner, DM2004
MC: outside Xenon
Blue : γ trackingPink : whole liquid xenonDeep pink : fiducial volume
U-chain gamma rays
XMASS (liquid Xe sci.) at Kamioka Self shielding with large target
for gamma rays
80cm dia.800 kg
All volume20cm wall cut30cm wall cut (10ton FV)
Large self-shield effect
External ray from U/Th-chain
BG can be efficiently reduced in < 500keV low energy region
1MeV0 2MeV 3MeV
250cm dia.23 ton
BG
nor
ma
lize
d b
y m
ass
Self shield for fast neutrons (5MeV)
Assumption: irreducible BG is only single neutron scattering
1/100 reduction is possible in 30cm thickness.
Red: single recoil
250cm dia.23 ton
Fiducialvolume
Single recoil Vertex distributionEvis>5keV
0 60 120Distance from the center (cm)
FV
# of
eve
nts
(cm
-3)n
100kg PrototypeFV 3kg
800kg detectorFV 100kg
23 ton detector, FV 10t
~ 30cm ~ 80cm ~ 2.5m
R&D
Dark matter search
Multipurpose detector
(solar neutrino, …)
“Full” photosensitive, large volume detector
Realize ~FULL COVERAGE
Neutron BGnot serious
Neutron shield30cm outer layer >1/100
Vertex reconstructionLarge photoelectronYield can be obtained~5 p.e./keV
Accurate vertexReconstruction basedon light pattern possible
Liq. Xe (31cm)3
MgF2 window
54 2-inch low BG PMTs
16% photo-coverage
HamamatsuR8778
Confirmation with a prototype detector done.
Demonstration with a prototype detector I
DATA
MC
A B C
+++Reconstruction works well
Data well reproduced by MC
30cm cube
Gamma ray injection from collimators
Confirmed Self shielding
>2 orders of magnitude agreementsGamma rays
Z= +15Z= -15
137Cs: 662keVDATAMC
PMTsaturation
-15 +15cm -15 +15cm
~1/200
~1/10
Reconstructed Z
Arb
itrar
y U
nit
10-1
10-2
10-3
10-4
10-5
10-1
10-2
10-3
10-4
10-5
60Co: 1.17&1.33MeV DATA MC
Reconstructed Z
Demonstration with a prototype detector II
Detail design of the 800kg detector
GEANT4 base simulation done
6cm RMS for 5keV @ boundary of FV Good performance
Background of the 800kg detector
Background origins
Neutrons: enough by
water shield
PMT rays: low BG PMT
Radioactive impurities:
distillation
PMT
n
RI in LXe
Background from PMTs (238U )
1.8 mBq-238U/PMT, most clean PMT ever made (achieved) c.f. usual “low” activity PMT ~O(1Bq/PMT)
BG <100keV 5cm self shield: ~10-3 /day/kg/keV 10cm self shield: ~10-4 /day/kg/keV 20cm self shield: ~10-5 /day/kg/keV (2events/100kg/5keV/year)
Fiducial cutRed: 100kgN
orm
aliz
ed b
y vo
lum
e(d
ay-1kg
-1ke
V-1)
Radioactive contaminationMeasured with the prototype detector
U, Th, Kr near to the goal. Within reach.
Internal origin of backgroundTarget values to achieve our goal
238U: < 1x10-14 g/g
232Th: < 2x10-14 g/g
85Kr: < 1ppt from reactors
(9±6) x10-14 g/g Further reduction by filter< 23 x10-14 g/g Upper limit, use filter3.3±1.1 ppt Distillation system (original)
Sensitivity of the 800kg detector
x 100 sensitive than CDMSII
XMASS FV 0.5ton year (100kg, 5yr)3 discovery
W/O any pulse shape info.
Plots exept for XMASShttp://dmtools.berkeley.eduGaitskell & Mandic
106
104
102
1
10-2
10-4
103
Near future ~5years
Ge, Ar and Xe go down to 10-9pb
Aim to detect DM ~2010
10-6
10-8
10-10
10-12Cro
ss s
ectio
n to
nuc
leon
(pb
)
+10yr, sensitivity of DM exp.
<10-10pb ~10ton volume Thick outer
shield (~30cm) will reduce neutron BG for further >100!
10-6
10-8
10-10
10-12Cro
ss s
ectio
n to
nuc
leon
(pb
)
Summary Direct detection of DM, WIMPs, is an urgent issue. Solid state & Noble gas liquid are the world trend fo
r large scale experiments. Search with <10-9pb, neutron will be most serious b
ackground, and it should be shielded by the target volume itself. XMASS
R&D efforts for XMASS done. XMASS at Kamioka is going to improve factor 100 i
n the sensitivity (~10-9pb). Serious competition in the world. Need to start now!