Date post: | 18-Jan-2018 |
Category: |
Documents |
Upload: | audrey-watson |
View: | 219 times |
Download: | 0 times |
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 1
The Search for Dark Matter in the form of WIMPs:CDMS (Cryogenic Dark Matter Search)
Extending the search for dark matter WIMPsbeyond CDMS-II with SuperCDMS
SLAC Experimental SeminarNovember 17, 2004
Blas Cabrera
CDMS Collaboration
At Stanford: Paul Brink, Laura Baudis (now at U Florida), Jodi Cooley,Clarence Chang, Walter Ogburn, Matt Pyle, Daniel Soto, Betty Young (SCU),
Pat Castle, Astrid Tomada and Larry Novak
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 2
CDMS Collaboration
UC Berkeley, Stanford, LBNL, UC Santa Barbara,Case Western Reserve U, FNAL, Santa Clara U, NIST, U Colorado Denver, Brown U, U Minnesota,U Florida, Princeton
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 3
Dark Matter WIMPs• The Science
– Scientific case compelling for dark matter WIMPs both from particle physics side and astrophysics side
• Galaxy formed in dark matter gravitational potential– Our solar system rotates through swarm of WIMPs 109/cm2/s– These would only interact with nuclei not electrons ~ (mN/me)2
– Nearly all backgrounds interact with e’s produced by gammas• CDMS II detectors and program
– Tower 1 (4 Ge and 2 Si detectors) at SUF (neutron limited)– Same Tower 1 at Soudan - PRL 93, 211301 (2004) best by x4– Towers 1& 2 (2004) and Towers 1-5 (2005) - x20 (n limited)
• SuperCDMS detectors and program– Development Project - 5 kg Ge new detectors run at Soudan– Phase A (25 kg Ge) & Phase B (150 kg Ge) run at SNOLab– Can run 1000 kg Ge at SNOLab before neutron limited
• Conclusions: to succeed need SLAC technical and management experience
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 4
timeThe BIG questions
?What is the
Origin of the Universe?
What role did Quantum Gravity play in the birth of the Universe?
The fabric of Space & Time: Was Einstein right?
How did Black Holes form in the early Universe?; Are Gamma Ray Bursts related to Black Holes?
What is Dark Matter? Does Dark Energy really exist?
How did Galaxies form? Which came first, Black Holes or Galaxies?
How does our star work? Is Life in our galaxy unique?
What is the ultimate fate of the Universe?
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 5
Expanding universe - simulations and data
QuickTime™ and aYUV420 codec decompressor
are needed to see this picture.
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 6
Concordance Model of Cosmology
• Supernovae + Cosmic Microwave Background + Large Scale Structure
• Great overall success!
• However raises even more questions about origin of dark energy
Dark Energy density
Matter density
€
WMAP + flat : Ωm = 0.27 ± 0.04 ΩΛ = 0.73 ± 0.04
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 7
Composition of the Cosmos
WIMPs
WMAP best fit
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 8
Strong motivation• Cosmology
– Theory:– Observation:
• Nuclear Physics: baryons - p, n, e– Theory of formation of H, D, He, Li– Observations:
• Astronomy: luminous matter– Stars & gas
• Baryonic dark matter is necessary: MACHOs at most 20% of halo mass
• Non-baryonic dark matter dominates universe hot cold
axionsneutrinosmonopolesWIMPs
X√
√√
Supersymmetry LSP
~1√?X√ √
best DMcandidates
ρ ρcrit = Ωm; Ωm + ΩΛ =1; ΩΛ ≠ 0 still ugly!
€
0.25 ≤ Ωm < 0.30
€
lum ≈ 0.005 0.01 0.1 1
lumDM, B
B DM, ≠B
m Λ
But favored bysupernova data
€
0.03 < ΩB < 0.05
€
χ =N10* ˜ B + N20
* ˜ W 3 + N30* ˜ H 1
0 + N40* ˜ H 2
0
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 9
The Signal and Backgrounds
χ0
v/c 710-4
NucleusRecoils
Er 10’s KeVphonons
Signal (WIMPs)
Er
v/c 0.3
ElectronRecoils
Background (gammas)
Er
ionization
Surface electrons from beta decay can mimic nuclear recoils
Neutrons also interact with nuclei, but mean free path a few cms
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 10
Direct Detection of Neutralinos
• [e. g., Lewin & Smith; and Jungman, Kamionkowski & Griest]
• The observed differential rate of events is given by
• For a Maxwell distribution of incident velocities
dRdQ
=σ0 ρ0
π v0 m χ mρ2 F
2 Q( )T Q( )η Q( )
T Q( ) =exπ −vmin v0( )2
[ ] wηeρe vmin = Qm N mρ
η Q( ) is the detector efficiency as a function of Q
€
R in evts/kg - d, typically σ 0, scalar >> σ 0, spin; ρ 0 WIMP (χ ) at earth, ~ 0.3 GeV/cm3
v0 velocity of sun around galaxy, ~ 220 km/s
mχ , mN mass of neutralino & nucleus, mr =mχ mN
mχ + mN
; recoil energy Q = mr2 v2
mN
1− cosθ *( )
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 11
• To compare the coherent rates of different materials
• We define the fundamental WIMP-nucleon cross section
• Two target materials such as Si and Ge very powerful
mχ =40 GeV
σ0,σχalaρ=5×10−42 χm2neutronbackground
σ0 scalar =4mχ
2 mN4
π mχ + mN( )2
fn
mn
⎛ ⎝ ⎜ ⎞
⎠ ⎟2
, where fn ≅ fp is the WIMP - nucleon coupling
σ0 W n
mrχ n2 =
4π
fn2 =
σ 0 scalar
A2 mrχ N
2 , which gives σ 0 Wn in terms of σ 0 scalarwith A ≅ mN mn
Spin independent scalar cross section
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 12
Nuclear form factor suppression
• For spin-independent or coherent interactions the form factor F2(Q) suppression is shown below and results in suppressed rates for heavy nuclei
50 keV true nuclear recoil threshold is equivalent to about 5 keVee recoil
F Q( ) =3
j1 Qρn / ηχ( )Qρn /ηχ
exπ −Qσ/ηχ( )2 /2[ ], wηeρe ρn ≈ 0.89A1 3 + 0.3( ) fm and σ~1 fm
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 13
Summary and bold future visionLimit at SUF 2002(during CDMS II)
Development Project 5 kg of Ge 2008
SuperCDMS Phase C 1000 kg of Ge
World-best limit today
SuperCDMS Phase A 25 kg of Ge 2011
CDMS II goal 2006
SuperCDMS Phase B 150 kg of Ge 2014
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 14
CDMS ZIP Detectors (Ge/Si)
Qinner
Q outer
A
B
D
C
Rbias
Ibias
SQUID array Phonon D
Rfeedback
Vqbias
Phonon sensors (4) (TES)
Ionization Electrodes (2)x-y-z imaging:
from timing, sharing
WIMPs: σ(Ge) >> σ(Si)
Neutrons: σ(Ge) ~ σ(Si)
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 15
380 mm Al fins
60 mm wide
~25% QP collection eff.
o TES’s patterned on the surface measure the full recoil energy of the interaction
o Phonon pulse shape allows for rejection of surface recoils (with suppressed charge)
o 4 phonon channels allow for event position reconstruction
ZIP detector phonon sensor technology
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 16
The ZIP Detector Signal• Charge & Phonon signals occur on a similar timescale• Phonon pulse time of arrival allows for event position
reconstruction• 20 keV event in a Si & Ge ZIP
Si ZIP Ge ZIP
(EXCELLENT S/N FOR 20 KeV TRUE RECOIL ENERGY)
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 17
Am241 : 14, 18, 20, 26, 60 kev
Cd109 + Al foil : 22 kev
ZIP Phonon Position Sensitivity
Delay Plot
A D
CB
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 18
First operate at SUF (17 mwe)• At SUF
– 17 mwe– 0.5 n/d/kg
• At Soudan– 2090 mwe– 0.8 n/y/kg
• At SNOLab– 6060 mwe– 1 n/y/ton
Log 1
0(Muo
n Fl
ux) (
m-2s-1
)
Depth (meters water equivalent)
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 19
ZIP 1 (Ge)ZIP 2 (Ge)ZIP 3 (Ge)ZIP 4 (Si)ZIP 5 (Ge)ZIP 6 (Si)
SQUID cards
FET cards
4 K
0.6 K0.06 K0.02 K
The shallow Stanford Underground Facility (SUF)
• Tower 1 operated at SUF during calendar 2002 at a depth of 17 mwe sufficient to eliminate the hadronic component of cosmic rays
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 20
Recoil [keV] vs Charge [keVee]
Yellow252Cf
Blueµ coin
Redµ anti
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 21
Tower 1 (4 Ge and 2 Si detectors) at SUF
• SUF Run 21– Calendar 2002– 52.6 kg-d of Ge after
cuts• Saw 19 nuclear recoils
– clean separation of gamma & beta events
– See 10 keV and 67 keV lines for energy calibration
• All consistent with neutrons– Consistent ratio of Ge
singles to Ge multiples
– Consistent ratio of Ge events to Si events
– Only gain as sqrt(MT)• TO DO BETTER NEED TO GO
DEEPER
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 22
Now operate at Soudan (2090 mwe)
• At SUF– 17 mwe– 0.5 n/d/kg
• At Soudan– 2090 mwe– 0.8 n/y/kg
• At SNOLab– 6060 mwe– 1 n/y/ton
Log 1
0(Muo
n Fl
ux) (
m-2s-1
)
Depth (meters water equivalent)
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 23
Outside of the Experiment
plasticscintillators
outerpolyethylene
lead
ancientlead inner
polyethylene
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 24
Run 118 (1T) & Run 119 (2T) in Soudan
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 25
Phonon energy (prg) in keVPhonon energy (prg) in keV
Excellent agreement between data and Monte CarloExcellent agreement between data and Monte Carlo
Energy calibration of Ge ZIP with 133Ba source
Ionization energy in keVIonization energy in keV
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 26
Nuclear recoil calibration of Ge & Si ZIPs with 252Cf
Nuclear recoils in Ge ZIPNuclear recoils in Ge ZIP Nuclear recoils in Si ZIPNuclear recoils in Si ZIP
Excellent agreement between data and Monte CarloExcellent agreement between data and Monte Carlo
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 27
252Cf Neutron & Gamma calibration data
• Upper red dashed line are +/- 2 σ gamma band
• Lower red dashed line are +/- 2 σ nuclear recoil band
• Separate high statistics calibrations with 133Ba gamma source
• Determined with calibration data as was the analysis threshold energy
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 28
133Ba gamma & 252Cf neutron calibrations• Use phonon
risetime and charge to phonon delay for discrimination of surface electrons “betas”
• Cuts and analysis thresholds determined entirely from calibration data with WIMP search data blinded until after the cuts and thresholds were set.
gammas
neutrons
ejectrons
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 29
Example of setting cut with calibration
Calibration - Gaussiandistribution 1000 evts
Data - samedistribution100 evts
Datax10
Calx10
Cut atlast event
Probability0.1 of eventpast cut
On averagearea beyondlast event = 1
On averagearea beyondcut = 0.1
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 30
• In 92 days between October 11, 2003 and January 11, 2004, we collected 52.6 live days - a net exposure of 22 kg-d after cuts
• Below data are shown before (left) and after (right) timing cuts
WIMP search data with Ge detectors
(yellow points are from neutron calibration)
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 31
CDMS-II Reach with five Towers• We have begun to
explore MSSM cross section range
• DAMA largely ruled out for spin independent scalar interactions (see Gelmini & Gondolo hep-ph/0405278)
• Light mass region suggested by Bottino (hep-ph/0307303) largely ruled out
• Another factor of 3-5 improvement at Soudan past CDMS-II (neutrons)
• THEN MUST GO DEEPER - exploring SNOLab in Canada
EGRET gammas asDM annihilation
astro-ph/0408272
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 32
Cryocooler Improvements
• Sumitomo (SHI) RDK415D cryocooler head & compressor
• 1.5 W at 4.2 K• 45 W at 50 K• Cost $50k• First use in dewar
(reduce boiloff)• Second use in Estem
(colder 4 K stage)
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 33
Now installing Towers 3, 4 and 5 in Soudan
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 34
Completed fabrication & testing of T3-5
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 35
Propose to operate at SNOLab (6060 mwe)
• At SUF– 17 mwe– 0.5 n/d/kg
• At Soudan– 2090 mwe– 0.8 n/y/kg
• At SNOLab– 6060 mwe– 1 n/y/ton
Log 1
0(Muo
n Fl
ux) (
m-2s-1
)
Depth (meters water equivalent)
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 36
SuperCDMS Roadmap to SNOLAB
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 37
Possible SLAC roles• At Stanford we know how to fabricate detectors for 25 kg experiment, but for 150 kg and 1000 kg we need to mass produce the fabrication/mounting
• SLAC could take central responsibility for fabrication with a combination of in house facilities (radon suppression) and commercial production.
• Other major areas include electronics (custom boards), DAQ (with digitizers for 10k channels readout at a 50 Hz for calibrations), computing (first pass data reduction, data analysis and Monte Carlo simulations).
• With Spokesperson for SuperCDMS at Stanford, SLAC could & should take the lead role in project management
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 38
Detector development (Paul)• Existing ZIPs3” dia x 1 cm thick
• Thicker ZIPs3” dia x 1” thick(base detector)
• Explore larger ZIPs to 4” dia and up to 4 cm thick
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 39
Detector development plans
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 40
05
1015
02
46
810-1
-0.5
0
0.5
1
X Position [mm]Z Position [mm]
Interleaved Ionization electrodes concept
• Alternative method to identify near-surface events– Phonon sensors on both sides are virtual ground reference.– Bias rails at +3 V connected to one Qamp– Bias rails at -3 V connected to other Qamp– Signals coincident in both Qamps correspond to events drifted out of the bulk.
– Events only seen by one Qamp are < 1.0 mm of the surface.
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 41
Interleaved Ionization electrode design• Design details
– To maintain ~60 pF of capacitance requires keeping bias and ground rails ~ 1 mm apart.
– Phonon sensors ‘contained’ within the (200 mm wide)ground rails.
– First mask-layout recently completed:
Ground ring around side to define the ‘Qouter’ volume containing all surfaces
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 42
New Read-out schemes• Two-stage SQUIDs for reading out new phonon sensors
– Allows lower Rn, more TESs, better phonon sensor surface area coverage.• Will improve effectiveness of present phonon risetime cut even further.
– Allows move to Al-Mn TESs to overcome W Tc variability• Resitivity of Al-Mn < W, hence risk /design constraint of electro-thermal oscillation if change-over to two-stage SQUIDs not implemented.
– Commensurate with NIST-style time-domain multiplexing.• ZIP detector phonon pulses are probably sufficiently slow to utilize this scheme effectively to reduce the readout wiring to room temperature that would otherwise be required.
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 43
• Two-stage SQUID configuration – Ionization detector transformer-coupled to first-stage SQUID
– Eliminate potential microphonic read-out issues associated with FET readout
– Eliminate IR photon leakage
– Eliminate heated FET load on 4 K
– Transformer ~ 12 mm x 6 mmchip, fabricated at NIST.
– Critically damped circuit, ~1 MHz sampling required.
– Simulations give 0.4 keVee FWHM
Ionization read-out using SQUIDs
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 44
How to build a 1000 kg experiment in stages
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 45
Schematic of new ‘SNObox’
x3
x3
Exploring cryocooler system with little or no cryogen servicing
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 46
Summary and bold future visionLimit at SUF 2002(during CDMS II)
Development Project 5 kg of Ge 2008
SuperCDMS Phase C 1000 kg of Ge
World-best limit today
SuperCDMS Phase A 25 kg of Ge 2011
CDMS II goal 2006
SuperCDMS Phase B 150 kg of Ge 2014
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 47
Conclusion• Excellent scientific reach for 1000 kg Ge experiment at SNOLAB with three phases
• SuperCDMS Development Project continues to run Towers 1-5, develops 0.6 kg ZIP detectors, and operates one SNOLAB Tower at Soudan in 2008
• Start SNOLAB installation in 2007 so that detectors can be operated starting in 2009
• SLAC could & should take leadership role in SuperCDMS
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 48
Identify and Reduce 210Pb, 14C & 40K
• Use Van de Graaff to attract positive ion radon daughters 222Rn -> 218Po -> 214Pb -> 214Bi -> 214Po -> 210Pb
• Run VdG for 2 hrs• Wipe surface & count
QuickTime™ and aTIFF (Uncompressed) decompressorare needed to see this picture.
QuickTime™ and aTIFF (Uncompressed) decompressorare needed to see this picture.
3.8d 3.1m 27m 20m .16ms
250kV
ground
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 49
Quantum Universe• Question 6:
– WHAT IS DARK MATTER?– HOW CAN WE MAKE IT IN
THE LABORATORY?
• “…We need to study dark matter directly by detecting relic dark matter particles in an underground detector and by creating dark matter particles at accelerators, where we can measure their properties and understand how they fit into the cosmic picture.”
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 50
Quantum Universe - CDMS!
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 51
Quntum Universe - direct detection
• “… The particle nature of dark matter can be verified by finding the rare events they would produce in a sensitive underground dark matter detector such as CDMS. …”
• “However, to understand the true nature of dark matter particles, particle physics experiments must produce them at accelerators and study their quantum properties. Physicists need to discover how they fit into a coherent picture of the universe. Suppose experimenters detect WIMPs streaming through an underground detector. What are they? Are they the lightest supersymmetric particle? The lightest particle moving in extra dimensions? Or are they something else?”
• Did not go the final step of saying that if supersymmetry discovered at an accelerator then we must see if it is the dark matter with direct detection experiments.
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 52
Earlier MSSM
Baltz & Gondolo PRD67 065503 (2003)Kim,Nihei,Roszkowski, hep-ph/0208069
Baltz & Gondolo hep-ph/0102147
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 53
mSUGRA and relax GUTs
Baer et al, hep-ph/0305191Chattopadhyay et. al, hep-ph/0407039Ellis et al, hep-ph/0306219
Bottino, et al hep-ph/0307303
Blas Cabrera - Stanford UniversitySLAC Experimental Seminar Page 54
mSUGRA and Split Supersymmetry
Baltz & Gondolo hep-ph/0407039A. Pierce, hep-ph/0406144 &G. F. Giudice and A. Romaninohep-ph/0406088