1. Listening for the Dark
Harry Nelson
UCSB

2. Plan Massive Dark Matter Direct Detection Xenon CDMS Future 9/15/05
Penn State Colloquium

3. `We Declare a New Order’ (Joel Primack)
Metals (us)
0.01%
Visible Baryons
0.5%
Dark Baryons
5%
Cold Dark Matter
(WIMPs?)
25%
Cosmological Constant 
70%
`We Declare a New Order’ (Joel Primack)
9/15/05
Penn State Colloquium

4. Point out r, v, they tell you about central force and hence mass distribution. Use astrojax to demonstrate: 2 points: v up, F up; can study by v(r).
v; = v/c r
9/15/05
Penn State Colloquium

5. Identfy axes… for the eight planets inner planet… just grabbed velocity data, nothing special. Equation assumes all mass is concentrated at the center.
9/15/05
Penn State Colloquium

6. The power law indicates all mass in Solar System is indeed all concentrated at center. *Level* indicates value of the mass at the center.
b = ± 0.009
a  M = (1.85 ± 0.06) 1030 kg
9/15/05
Penn State Colloquium

7. Stars w/r Galaxy Center: v/c =  constant  0.7 10-3
Repeat for other systems… moons of Saturn, all mass at center by power law, mass of saturn much less by level… Galaxy power law wrong, showing mass not at center. Use Gauss’s law to fix up, assume spherical symmetry. Level gives local mass *density*, about 2/3 p/cm^3, ½ luminous, ½ dark.
Msun = (1.85 ± 0.06) 1030 kg
Planets about sun
MSaturn = (5.5 ± 0.2) 1026 kg
Moons about Saturn
9/15/05
Penn State Colloquium

8. Galactic Dark Matter R is the distance from center v
v is the speed (tangential)
R is the distance from center
9/15/05
Penn State Colloquium

9. What about our home galaxy (Milky Way)? v R 
220 km/s
Honma and Sofue, 1996
:v0=180 km/s
Without the Dark Halo
(Binney & Tremaine)
9/15/05
Penn State Colloquium

10. Milky Way: mainly a dark cloud
Bulge
Disk
Sun: moves in plane of disk
v/c =  0.7 10-3
Maxwellian/Gaussian (simple)
v/c =  0.7 10-3
Particles in `halo’: 3-d
 = mc2  n  1/3 GeV/cm3
(1/2 of total mass density)
Use Bowling Ball. Our galaxy like a fried egg, rotating like a compact disk inside, at 0.7/1000 speed of light. Dark matter at rest on average; we plow through. Mass density is product of mass/particle * number density… raise mass/particle, lower needed number density. DM must move so as not to collapse at center… simplest is Maxwellian, *helps* experimentation.
9/15/05
Penn State Colloquium

11. Design a Particle and an Experiment0
v/c =  0.7 10-3
v/c =  0.7 10-3
We use Germanium, A=73, mc2=67.6 GeV;
others: Si, S, I, Xe, W
Neutral: cool particles neutral –
, n, , K0, Z0, H0…
Chi^2… lightest superpartner… although not the only possibility. Weak scale is about the mass of the W and Z bosons, some new physics keeps the Higgs light GeV is `Massive’
ER½ mGec2 2
 ½ 68 GeV  ½  10-6
 20 keV
 x-ray energy ! Easy!
Massive: Mc2100 GeV hinted
at by accelerator data
`Weak Scale’
9/15/05
Penn State Colloquium

12. Arguments for  characteristic of weak interaction
Particle physics… chose Mc2  100 GeV, `Weak Scale’
Big Bang… independently implies weak cross section as well… Coincidence(s)… or Clues ???
*Big Bang*… weak cross section is just what needed to describe observed ratio of luminous to dark matter, if luminous and dark were once in thermal equilibrium.
9/15/05
Penn State Colloquium

13. What is the weak interaction cross section?
Donald H. Perkins, 1987
Note that this is cross section per proton for inverse beta decay. Note the 100 light years!
9/15/05
Penn State Colloquium

14. Rate governed by scattering cross section, 
Flux… get from density and velocity. Cross section converts the rest into an observable rate..
9/15/05
Penn State Colloquium

15. Coherence, density of states enormous bonus!
Scattering off a proton….
…. Hopeless! 0
Indistinguishable
A = # neutrons + # protons
Back to rate… traditional to quote cross section per *nucleon* (either proton or neutron). Really use a bag of nucleons, so must add up effects of all… add amplitudes *then* square. Also win due to density of states (for very massive wimps). Huge bonus!! Now doable.
Density of States:
9/15/05
Penn State Colloquium

16. Rate of Main Background
Move on to background… as example, consider radioactivity of person at 2 meters… turns out hard to beat! DAMA lives with large background, but uses a huge target mass (100 kg, now upgrading to 250 kg) to boost signal, and then looks for temporal modulation at an expected frequency. CDMS, other limit, tries to distinguish electron recoil background from nuclear recoil background… very small mass. (1-10 kg).
Rate about 103 / (kg-day) !!!
Shield… but that radioactive too
Strategies: DAMA… huge target mass (100 kg),
look for astrophysical modulation
CDMS… small target mass (few kg)
distinguish electron from nucl. recoil
9/15/05
Penn State Colloquium

17. 0 Signal Shape Er MWIMP=100 GeV   10-42 cm2/nucleon A2 Nucleus
Silicon, Sulphur
Germanium
Iodine, Xenon
Nucleus
Recoils Er
Slope: Maxwell-Boltzmann
WIMPs in Galaxy
Diffraction 0
9/15/05
Penn State Colloquium

18. Catalog of Direct Detection Experiments
Indirect Detection:
Super-Kamiokande,
Amanda, Ice-Cube,
HEAT, GLAST,
Egret…
Look for astrophysical
neutrinos, gammas
From WIMP
annihilation
Mine, Name of experiment, technique… mostly particle/nuclear physics detectors, mass, status… mostly non-US world effort.. Note DAMA, NaI (they have evidence for a signal!), CDMS (us)… we operated at Stanford, not deep, for many years; now deep at Soudan.
9/15/05
Penn State Colloquium

19. Direct Detection: Signal and Main Background
v/c  0.3
Electron
Recoils
Background
Sparse Energy Deposition Er
Nucleus
Recoils Er
v/c  710-4
dense energy deposition
efficiency low
distinct energy scale
0 (calibrate: neutron)
Differences the
Basis of Particle ID
9/15/05
Penn State Colloquium

20. ZEPLIN II, III, MAX, XMAS, XENON
IGEX,
DRIFTI, II
CDMS, EDELWEISS
phonons H
ionization Q
scintillation L
ZEPLIN II, III, MAX, XMAS, XENON
NAIAD, ZEPLIN I, DAMA
CRESST I,
PICASSO,
COUPP
CRESST II,
ROSEBUD
Tim Sumner
9/15/05
Penn State Colloquium

21. E. Aprile et al., arXiv:astro-ph/0503621 Tim Sumner
Xenon – nuclear recoils give 1/7 scintillation/energy, compared to electron recoils (``quenching’’)…. Sets recoil energy scale
E. Aprile et al., arXiv:astro-ph/
– Bernabei et al.
– Arneodo et al.
– Akimov et al.
– Aprile et al.
– LIP work
Lindhard
Hitachi
Tim Sumner
9/15/05
Penn State Colloquium

22. Zeplin 1 – Scintillation Alone
Nucl. Recoil
Zeplin 1 – Scintillation Alone
Electron Recoil
125 keV
30 keV
30 keV
Nuclear Recoils –
Faster Risetime…
3.2 kg Xe
Risetime (ns)
9/15/05
Penn State Colloquium

23. Zeplin-I Limit- Background Weakens
300 kg-days
CDMS: 100 kg-days
10-42 cm2
9/15/05
Penn State Colloquium

24. 2-Phase (Liquid/Gas) Noble … Ar or Xe
Nuclear Recoil S1 S2 S2
Electron Recoil S2 S1 S1
9/15/05
Penn State Colloquium

25. Preliminary Preliminary 9/15/05 Penn State Colloquium
Gamma + neutron source
Gamma source
Zeplin-II
Zeplin-II
Preliminary
Preliminary
XENON
9/15/05
Penn State Colloquium
Aprile et al. 2005

26. However, unforeseen backgrounds are the rule as sensitivity increases
Analysis….
Appears scalable
However, unforeseen backgrounds are the rule as sensitivity increases
Promising… ZEPLIN-II and Xenon-10 soon deployed in deep sites
9/15/05
Penn State Colloquium

27. CDMS: Adapt Traditional Ionization Detector0
ER10’s keV
Holes e-
Electrode
Implants E 
1 cm
One of our detectors (now 30 in operation)… also use Silicon as a control. Key is to compare rate of expected signal and rate of expected background.
Germanium (or Silicon)
7.6 cm
¼ kg (1/10 kg)
What rate? (in, say, 1kg)
Backgrounds?
….…. Gamma rays, neutrons, surface beta-decay
9/15/05
Penn State Colloquium

28. Our Hunting Ground Weakly Interacting Experiment Massive
Particle
Experiment
CDMS (shallow)
DAMA (old!)
1/10
9/10
Theory
SUSY,
various constraints
including Big Bang
Review the scales… cross section is *per nucleon* on vertical. Note WIMP acronym. Theoretical predictions are colored areas, the result of adaptation of various SUSY descriptions… until SUSY seen (if seen), lots of free parameters… lower cutoff from LEP. Experiments… DAMA is 3 sigma region… now 6 sigma. CDMS is limit, 9/10 chance below, prior to move to deep site. Nature of curve: most sensitivity when mass of target is about that of projectile (use Oddballs). Sensitivity loss at low mass from bad kinematics (demonstrate with small bouncy ball colliding with oddball). Sensitivity loss at high mass due to reduced number density needed to obtain given mass density.
Gaitskell/Mandic
9/15/05
Penn State Colloquium

29. DAMA – Exploit Annual Modulation
Signal: higher rate in June,
lower in December
Background: constant in time
DAMA exploits the motion of earth relative to sun… sun hurtles along with galactic disk at about 7/1000 speed of light, but earth is simultaneously swining around (consider running with astrojax!). More energy available in June, more events above threshold. Background should be constant, but… is it really?
9/15/05
Penn State Colloquium

30. CDMS Collaboration (Mar. 2002)
In the deep Soudan mine; 3 ignore DOE safety regs (MINOS under construction, hard hats required, not steel toed boots though).
9/15/05
Penn State Colloquium

31. CDMS Collaboration
Brown University
M.J. Attisha, R.J. Gaitskell, J-P. F. Thompson
Case Western Reserve University
D.S. Akerib, P. Brusov, C. Bailey, M.R. Dragowsky, D.D.Driscoll, S.Kamat, A.G. Manalaysay, T.A. Perera, R.W.Schnee, G.Wang
University of Colorado at Denver
M. E. Huber
Fermi National Accelerator Laboratory
D.A. Bauer, R. Choate, M.B. Crisler, R. Dixon, M. Haldeman, D. Holmgren, B. Johnson, W.Johnson, M. Kozlovsky, D. Kubik, L. Kula, B. Lambin, B. Merkel, S. Morrison, S. Orr, E.Ramberg, R.L. Schmitt, J. Williams, J. Yoo
Lawrence Berkeley National Laboratory
J.H Emes, R. McDonald, R.R. Ross, A. Smith
Santa Clara University
B.A. Young
Stanford University
P.L. Brink, B. Cabrera, J.P. Castle,
C.L. Chang, J. Cooley, M. Kurylowicz,
L. Novak, R. W. Ogburn, M. Pyle, T. Saab,
A. Tomada
University of California, Berkeley
J. Alvaro-Dean, M.S. Armel, M. Daal, J. Fillipini, A. Lu, V. Mandic, P.Meunier, N. Mirabolfathi, M.C.Perillo Isaac, W. Rau, B. Sadoulet, D.N.Seitz, B. Serfass, G. Smith, A. Spadafora, K. Sundqvist
University of California, Santa Barbara
R. Bunker, S. Burke, D.O. Caldwell, D. Callahan, R.Ferril, D. Hale, S. Kyre, R. Mahapatra, J.May, H. Nelson, R. Nelson, J. Sander, C.Savage, S.Yellin
University of Florida
L. Baudis, S. Leclerq
University of Minnesota
J. Beaty, P. Cushman, L. Duong, A. Reisetter
Cryogenic
Dark
Matter
Search
Bold faced names… those who have grown sufficiently unproductive.
9/15/05
Penn State Colloquium

32. Nuclear Recoil bad at making Ionization
Holes e-
Nuclear recoil goes at a velocity small compared to the typical electron velocity… electron wavefunction adiabatically distorts, about 1/3 the ionization. Electron recoil (Compton scatter) at a velocity exceeding typical electron velocity in Ge orbital wavefunction, sudden approximation, good ionization efficiency.
Germanium
more ionization!
Need a second, `fair’ measure of deposited energy… sound!  0
Both deposit, say, 20 keV
9/15/05
Penn State Colloquium

33. Our Detectors
`Phonon sensor (4)’ (TES)
Array of Transition Edge Sensors
Ionization Electrodes (2)
x-y-z imaging:
from timing, sharing
Transition edge sensor to detect phonons (phono-cathode), read out with SQUIDS. Segmentation of 4… allows x-y imaging, parallel to face. Other side has ionization electrodes, segmented into edge and inner; allows rejection of edge events. Z from timing… `ZIP’… 50 mK.
Z-coordinate, Ionization, Phonons
ZIP
Operate at Kelvin
9/15/05
Penn State Colloquium

34. Transition-Edge Sensor (TES)
The Phonon Sensor Al
quasiparticle trap
Al Collector W
Transition-Edge Sensor (TES)
Ge or Si
quasiparticle
diffusion
phonons
Cooper Pair
Phonon impinge on thick aluminum, bust cooper pairs, quasiparticles diffuse to edge, where they heat a tiny piece of W kept on edge between normal and superconducting. Resistance goes up, read off current change.
superconducting
normal
T (mK)
Tc ~ 80mK
RTES () 4 3 2 1
~ 10mK
R. Schnee
9/15/05
Penn State Colloquium

35. Pulses
20 KeV
Phonon (4)
Charge (2)
Pulse readout…2 charge electrodes, 4 phonon. Phonon pulses slower, use intercomparison of pulseheights to image. Lots of information in timing.
300
300
Time (s)
Time (s) 30
-30
-30 30
-30 30
9/15/05
Penn State Colloquium

36. Separation of the types of recoils
(electron recoils)
Neutrons cause nuclear recoils too!
Another background…
Hold up gamma source, vertical scale is ionization/phonons, vertical just phonons. Then hold up neutron source, making nuclear recoils… see the difference. But realize, neutrons are irreducible background! Go deep underground to evade!
(nuclear recoils)
Sound
9/15/05
Penn State Colloquium

37. SUF Run 21 Germanium – Low Threshold
1.3 keV x-ray, L-shell Capture
Shallow Site
Neutron Background
10.4 keV Ga x-rays from 71Ge K-shell Capture
R. Bunker
9/15/05
Penn State Colloquium

38. Background Neutrons from Cosmic Ray Muons
Muons from cosmic rays bust up nucleus, eject neutrons which go a long way, hard to shield with local shielding… go deep.
Limited our earlier results…moved to a deep mine
9/15/05
Penn State Colloquium

39. Go Deep Underground to Evade Muons
Vertical… muons/m^2/year, two horizontals, depth in feet for rock, meters water equivalent. Soudan shallow among deep sites… DUSEL… 6500 mwe… San Jacinto attractive.
9/15/05
Penn State Colloquium

40. Deep Facility Can take a canoe to Canada from Soudan. Soudan Mine
Hosts: State of Minn., U Minn.,
Fermilab
690 meters underground
2090 meters water equivalent
9/15/05
Penn State Colloquium

41. Down deep in the Soudan mine
General view of our cavern… electronics above clean room.
HVAC
Mechanical
RF-shielded
Clean room
Shield
Fridge
Front-end
Electronics
Mezzanine
Detector Prep
DAQ/Electronics
Clean Benches
Icebox
Pumps,
Cryogenics
Soudan II
MINOS
connecting tunnel
9/15/05
Penn State Colloquium

42. Outside In
lead
plastic
scintillators
Inside clean room… surrounded by scintillators to detect and allow removal of any particles that might have induced background… WIMPs totally silent. Cross section, photo during assembly.
outer
polyethylene
ancient
lead
inner
polyethylene
9/15/05
Penn State Colloquium

43. A Cold Heart
Results from first…
ZIP 1 (Ge)
ZIP 2 (Ge)
ZIP 3 (Ge)
ZIP 4 (Si)
ZIP 5 (Ge)
ZIP 6 (Si)
4 K
0.6 K
0.06 K
0.04 K
SQUID cards
FET cards
14C
The very heart… lots of copper cans at various temperatures.
poorer resolution
Two `towers’
6 detectors…
10 cm
4 Germanium (0.25 kg each), 2 Silicon (0.1 kg each)
9/15/05
Penn State Colloquium

44.  Calibration (133Barium) (e recoils)
Phonons
Ionization
Hold up gamma source… Barium peak in ionization, both data and MC… phonons have less resolution.
L. Baudis
Energy, KeV
9/15/05
Penn State Colloquium

45. Better Source, Calibration
Ionization
Phonons
9/15/05
Penn State Colloquium

46. n Calib. (252Californium) (nuclear recoils)
Hold up neutron source, phonon energy. Germanium on left...
Reconstructed recoil energy, KeV
S. Kamat
9/15/05
Penn State Colloquium

47. Soudan Data First Run Second Run Two runs! 92 calendar days
53 live days, kg Germanium
Tower 1
Second Run
Two runs!
140 calendar days
74 live days, kg Germanium 0.6 kg Silicon
Double `Exposure’
9/15/05
Penn State Colloquium

48. Reject Multiple Interactions
ZIP 1 (Ge)
ZIP 2 (Ge)
ZIP 3 (Ge)
ZIP 4 (Si)
ZIP 5 (Ge)
ZIP 6 (Si)
4 K
0.6 K
0.06 K
0.04 K
SQUID cards
FET cards
14C
n, 
Backgrounds often do 0
WIMPs don’t scatter twice
Ignore multiple interactions.
9/15/05
Penn State Colloquium

49. First Run
After timing cuts, which reject external electrons
Prior to timing cuts
Nuclear Recoils induced by a neutron source
10.4 keV Gallium line
External e
(address with timing cuts)
B. Cabrera
9/15/05
Penn State Colloquium

50. External e : surface events, ionization missed
 decay contamination,
escaping compton recoils
Ionization not collected
Electrode
Implants E  z
`ZIP’ : `reconstruct’ z with start time, risetime
Germanium
9/15/05
Penn State Colloquium

51. 0 External e  decay contamination, escaping compton recoils
Edge Induces
Phonon Energy Downshift
Phonon Sensors 0
Eventually energy low, pathlength long, `phonons go ballistic’
Distinct Sensor Sharing
Lower Energy
Phonons Have
Longer Pathlength,
Diffuse More Promptly
Primary Phonons
`High Energy,’
`Rayleigh Scatter’
so they diffuse
+ energy downshift
9/15/05
Penn State Colloquium

52. Timing rejects surface/external e
Nuclear Recoils e
External e
9/15/05
Penn State Colloquium

53. WIMP search data with Ge detectors
Z2, Z3, Z5
After timing cuts, which reject external electrons
Prior to timing cuts
Nuclear Recoils induced by a neutron source
10.4 keV Gallium line
B. Cabrera
(yellow points are from neutron calibration)
9/15/05
Penn State Colloquium

54. All the features on one plot
Z2, Z3, Z5
10.4 keV Gallium line
Surface
Rejection:
Omitted
Applied 
Energy
Estimate
Corrected
(non-blind)
10 keV
threshold
(1 20 keV)
0.7  0.35 external e,
0.07 recoils from neutrons expected
2/3 event expected from Barium calib
9/15/05
Penn State Colloquium

55. Second Run – twice the exposure
After timing cuts, which reject most electron recoils
Prior to timing cuts
1.5
1.0
0.5
0.0
10.4 keV Gallium line
1.5
1.0
0.5
0.0
Z2/Z3/Z5/Z9/Z11
Ionization Yield
Ionization Yield
1 candidate (barely)
1 near-miss
Z2/Z3/Z5/Z9/Z11
Recoil Energy (keV)
Recoil Energy (keV)
ESTIMATE: 0.37 0.20 (sys.)  0.15 (stat.) surface electron recoils,
0.05 recoils from neutrons expected
9/15/05
Penn State Colloquium

56. Limits
DAMA
9/15/05
Penn State Colloquium

57. PRELIMINARY The Near Future Improvements 30 detectors in 5 towers of 6
Installed 3 additional towers November 2004
CDMS- Soudan
CDMS -projected
Edelweiss
ZEPLIN-1
PRELIMINARY
Improvements
Cryogenics, backgrounds, DAQ
Currently commissioning
30 detectors in 5 towers of 6
4.75 kg of Ge, 1.1 kg of Si to run through 2006
Improve sensitivity x10
Last fall we installed 18 additional detectors in 3 additional towers, for a total of 4 kilograms of germanium. Running these 5 towers of detectors for 9 months should improve our sensitivity by another order of magnitude. (0:15)
9/15/05
Penn State Colloquium

58. Sensitivity Expectations: Distant Future
9/15/05
Penn State Colloquium

59. Projected Sensitivies
9/15/05
Penn State Colloquium

60. CDMS Collaboration(March 2002)
9/15/05
Penn State Colloquium

61. DAMA – 100 kg of NaI Iodine, A=127 Eobs(KeVee)0.09 Erecoil (KeV)
Sodium, A=23
Eobs(KeVee)0.25 Erecoil (KeV)
Erecoil  Light
NaI
PMT
Copper
Lead
Poly
9/15/05
Penn State Colloquium

62. DAMA Background and Signal
Energy Spectrum
Bkgd  1 cpd/kg/keV
2-6 KeV
8-24 KeV Na(23) KeV I(127)
0.0195 0.019
cpd/kg/keV
through 2003 … 6.3 
Bernabei et al., astro-ph/
through 2000 … 4
9/15/05
Penn State Colloquium

63. Similar Exposure… Stanford Site
Nuclear Recoils
Surface electrons
Z1 () or Z5 (+)
Neutrons!!
Soudan rock filters the muons that make them
(but not WIMPS)
9/15/05
Penn State Colloquium

64. Limits DAMA 1-4 58,000 kg-days CDMS Stanford Edelweiss Non-blind blind
(10’s of
kg-days)
Various Models
9/15/05
Penn State Colloquium