Gaitskell Towards One Tonne WIMP Direct Detectors: Have we got what it takes? (CryoArray) Rick Gaitskell Department of Physics & Astronomy University College London source at

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Direct Detection: History & Future Oroville (88) [m = ?? GeV - if significantly better limit obtained at different mass] 90% CL Limit on Cross section for 60 GeV WIMP (scalar coupling) CDMS SUF (T) CDMS Soudan (T) 7 kg Ge+Si Cryodet GENINO (T) 100 kg Ge Diode GENIUS (T) 100 kg Ge Diode CryoArray (T) tonne Cryodet ~1 event kg -1 day -1 ~1 event kg -1 yr -1 ~1 event 100 kg -1 yr -1 LHC Not meant to be a complete list - see Different Colours Indicate Different Technologies NO W rjg Ge NaI Cryodet (T) Target Signal Liq Xe [m=20 GeV] Homestake (87) H-M (94) H’berg-Moscow (98), IGEX (00) DAMA (96) UKDMC (96) [m=100 GeV] DAMA (98) DAMA (00) CDMS SUF (99) CDMS SUF (00) Edelweiss (98) Edelweiss (01) Gaitskell (astroph )

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Background Rejection / WIMP Signal Identification Nuclear Recoil (WIMPs & neutrons) vs   Cryogenic Phonon + Ionisation or Photons (CDMS / Edelweiss / CRESST)  NaI Scintil. Photon Pulse Shape (DAMA ‘96 / Boulby DM ‘97-’02)  Liquid Xe Scintil Photon Pulse Shape (ZEPLIN I Boulby DM ‘01-) Photon + Ionisation (ZEPLIN II+III Boulby DM ‘02- )  Droplets Superheated Bubbles (Picasso/Simple ‘01-) Exploit WIMP “Wind” for signature  Annual Modulation NaI (DAMA 100 kg ‘97-00, 250 kg ‘02-)  Direction Axis of Recoil Gas TPC (DRIFT Boulby DM ‘01-)

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell CDMS I: SUF Site Stanford Underground Facility 17 mwe of rock hadronic component down by >1000 muon flux down by ~5 Low-Background Environment 25 cm polyethylene reduces muon-induced neutron flux from rock and lead by factor > cm Pb reduces photon flux by factor >1000 radiopure cold volume (10 kg) additional internal (ancient) lead shielding Active Scintillator Muon Veto muon veto >99.9% efficient reject ~22 “internal” neutrons/ day produced by muons within shield SHALLOW SITE SUFFERS FROM “EXTERNAL NEUTRONS” MeV from muons in rock polyethylene outer moderator detectorsinner Pb shield dilution refrigerator Icebox outer Pb shield scintillator veto rjg

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell CDMS: Ge BLIP Ionization & Phonon Detectors Four 165 g Ge detectors, for total mass of 0.66 kg during 1999 Run Calorimetric measurement of total energy ENERGY Resolution  = Ionisation 220 eV, Phonons 250 eV Inner Ionization Electrode Outer Ionization Electrode Passive Ge shielding (NTD Phonon Sensors on underside) Tower WiringWiring heat sinkingheat sinking holds cold FETs for amplifiersholds cold FETs for amplifiers BLIP

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell CDMS Nuclear Recoil Discrimination - Event by Event Nuclear recoils arise from  WIMPs  Neutrons Electron Recoils arise from  photons  electrons  alphas (Typical Background) Ionization yield  ionization/recoil energy strongly dependent on type of recoil Recoil energy  Phonons give full recoil energy Neutrons (external source) Gammas (external source) Trigger Threshold NOT A SIMULATION! 1334 gamma events, 616 neutron events 1 / 2 year’s background >> 1 / 2 year’s signal !! rjg

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell 1999 Run Ge BLIP Data Set Combined all single- scatters NR candidates Entire 96 live days operation Ge BLIPs = 10.6 kg-days  Gamma and electron bands well separated from NR band NR candidates are truly NR’s  See a total of 13 events > 10 keV  ~ 1.2 events/kg/day NR Band asymmetric (-3 ,+1.28  )= 90% efficient Expect 40 events for scalar WIMP  given by the DAMA Ann Mod Signal rjg Gammas Surface Betas Nuclear Recoils

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell WIMP SUSY CDMS I + II Status CDMS I  Current limit based on 10.6 kg-days exposure (PRL 84 (2000) 5699) 3x 160 g Ge detectors, with inner fiducial volume 46%  At TAUP will announce updated analysis based on larger fiducial vol Volume 46% -> 65% (See TAUP talk for new limit, and discussion of extra analysis) CDMS II  Infrastructure/buildings installed at Soudan mine  New Tower, total 1 kg detectors being commissioned at Stanford Underground Facility 6 dets: 2 50 g Ge g Si ZIPs  Tower will be installed in Soudan - January 2002 Test systems Jan-Mar’02 Low background operation Apr-Oct’02, target 44 kg-days Ge exposure which will push sensitivity to 0.07 events/kg/day in Ge (Er>10 keV)  Subsequent Towers installed Tower Operate Jan’03 -> (followed by more towers) Tower 1 to 7 (Full complement 7 kg) Operate Jun’04 -> B4 B3 B5 B6 Z1 Ge Z2 Si Z3 Ge Z4 Si Z5 Ge Z6 Si

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Current Experimental & Theoretical Regions DAMA Ann Mod CDMS I Edelweiss DAMA(96) H-M / IGEX Ellis et al. Gondolo… (g-2) Mandic et al UKDMC From limitplot rjg Updated Bednyakov et al.

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell 1 per minute in 4 m 2 shield CDMS II: Site Depths & Muon Flux Depth (mwe) Muon Flux (m -2 s -1 ) Stanford Underground Site SUF->Soudan Muon Flux Falls by ~ MC predict residual punch-through neutron signal events/keV/kg/day 500 Hz muons in 4 m 2 shield

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Already demonstrated discrimination to < 10 event / kg / year  >99.9% rejection of photons >10 keV (~0.5 events/keV/kg/day)  >99% rejection of surface-electrons >15 keV (~0.05 events/keV/kg/day) Identical Icebox: 7 Towers each with three Ge & three Si ZIP detectors  Total mass of Ge = 7 X 3 X 0.25 kg > 5 kg  Total mass of Si = 7 X 3 X 0.10 kg > 2 kg CDMS II Detector Soudan

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Si ZIP Ionization & Phonon Detectors ZIP: At end of fabrication steps involving µm photolithography at Stanford Nanofabrication Facility Advanced athermal phonon detection technology  Superconducting thin films of W/Al 4 K 0.6 K 0.06 K 0.02 K

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell CDMS Ge & Si Fast Phonon & New Electrode Detectors Al/W Grid 60% Area Coverage mm Squares 888 X 1 µm tungsten TES in parallel Aluminum Collector Fins 8 Traps

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell ZIP Detector: Detail of film patterns VIDEO NOT INCLUDED

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell collimator X y Px = (C + D - A - B)/(A+B+C+D) Delay plot xyZIP: Position Sensitivity

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell ZIP Detector: Operational Advantages VIDEO NOT INCLUDED

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Detection Mechanisms Thermal Channel is dumping ground ultimately for all energy

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell CryoArray: A 3 rd Generation Experiment Based on extrapolation of CDMS technology/strategy Basic parameters/goals  1000kg x 2 (live) years  >100 WIMPs at cm 2  <100 background events (prior to subtraction, eg, multiple scattering) Reduce backgrounds  Internal ( ,  ) and external (n) Increase mass and manufacturability of detectors/cryo package  Maintain Performance Possible sites  Soudan (CDMS II) among shallower site  National Underground Facility Depth Shared resources (vetos, assembly, materials screening, fabrication???) 10 4 – 10 5 increase over present limits 10 2 – 10 3 increase over expts under construction

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell “CryoArray”: One Tonne Dark Matter Detector New philosophy in deployment of cryogenic detectors  Use athermal phonon technology + ionisation to allow significant simplification of detector production and assembly  kg in kg units Challenge: Mass production/deployment  Detectors produced commercially by Si chip fab facility Simple bilayer metal circuits (superconductors W & Al)  Appropriate volumes of SQUID electronics now available Warm - Fermilab surface mount format Cold - SQUID arrays available in quantity/consistency Challenge: Modest improvements in existing performance….  Event by event discrimination: Noise (<1 keV) well matched to WIMP recoil CDMS I dets. showed 96 live days ‘free’ of systematics -> target 1 year free  Background reduction benefits from virtual elimination of passive material in detector space (self veto)

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell CDMS I Design Considerations for operation with (BLIP) thermal detectors  Low Thermal Load  Low Microphonics  (High impedance circuits)  Low IR leakage into 10 mK space (detector can’t take heat load)

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell CryoArray Design Considerations for (ZIP) non-thermal detectors  Non-thermal phonons  Most mounting hardware eliminated - low impedance so microphonics is not an issue  FET -> 4K (GaAs) or SQUID Q amp  Spatial Ch Multiplexing reduces electronics channels

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Nuclear Recoil Discrimination - Event by Event Nuclear recoils arise from  WIMPs  Neutrons Electron Recoils arise from  photons  electrons  alphas (Typical Background) Ionization yield  ionization/recoil energy strongly dependent on type of recoil Recoil energy  Phonons give full recoil energy Neutrons (external source) Gammas (external source) Phonon Trigger Threshold NOT A SIMULATION! 1334 gamma events, 616 neutron events

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell CryoArray (Sensitivity <1 per 100 kg-yr,  ~ cm -2 ) Scale up to 1 tonne detector with target (90%CL) <1 evt per 100 kg-yr Reduce  /  backgrounds by factor 20 vs CDMSII   > cts/keVee -1 kg -1 day -1 (This compares to keVee -1 kg -1 day keV for HMDS)   > cts/keVee -1 kg -1 day -1 (Challenge to survey surfaces to this sensitivity) Improve  /  rejection by factor 1-few!   99.5% -> 99.95% (1 in 2000) CDMS I 1999 in-situ calibrations already showed 99.96% (17k event calibs)   95% -> 99.5% (1 in 200) Ge BLIP with aSi contact, (E>25 keVr ) >95% (E>40 keVr) >99.5%, Si ZIP using phonon rise times (E=10-20 keVr ) >98%, (E>20 keVr) >99.5% Without Discrimination:  needs ~10 4 reduction in background from present (HM) levels rjg Revise rej

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Background Projections - CryoArray (1 tonne) dru = 1 event keV -1 kg -1 day -1 Energy Range keV x20 -1 x3 -1 x rjg H-M entry ALREADY HAVE 99.96% IN CDMS I

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Nuclear Recoil Discrimination - Event by Event Nuclear recoils arise from  WIMPs  Neutrons Electron Recoils arise from  photons  electrons  alphas (Typical Background) Ionization yield  ionization/recoil energy strongly dependent on type of recoil Recoil energy  Phonons give full recoil energy Neutrons (external source) Gammas (external source) Phonon Trigger Threshold 1334 gamma events, 616 neutron events

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Background Projections - CryoArray (1 tonne) dru = 1 event keV -1 kg -1 day -1 Energy Range keV x20 -1

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Neutron Background in CryoArray Expect dominant component from muon interactions in rock  Veto in cavity difficult – neutrons from 2 – 3 meters in  Polyethylene shield transparent above 50 MeV Need factor x20 improvement vs CDMS II  Increase depth to => 4000 mwe  Instrument the rock with 2.5 m streamer tubes.  Preliminary simulations indicate >75% of emergent HE neutrons from hadron cascades are >50-cm transverse size.  Augment with ‘umbrella’ veto  Increase shield density inside veto Soudan

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Problem: Performance -> Production Mass Recent challenge has been to demonstrate novel techniques  Low backgrounds &/or Discrimination Now it is time to solve some of problems of scale up …  Detector production issues (+larger sizes of collaboration) … and background reduction  What screening techniques are required? (Low energies / Monte Carlo??) Eliminating gamma/beta contamination in energy range keVee is new field.  Discrimination: Improvements can be shared between improved backgrounds and improved rejection performance Good experience of running technology underground  True comparison of technologies can only come once data from extended operation underground is available - highlights systematics

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell 2nd Kind of Limit on Threshold of Discrimination x1x1 x2x2 x2x2 x1x1 Lateral spread of lines simulates noise Background (e.g. Gamma or Surface Betas) Signal (Nuclear Recoil) In this region x 2 channel for signal events are buried in noise: Example where 2nd limit on discrimination may be relevant: CRESST x1(phonon) x2(photon) CaWO 4 Gamma: x1=50 keVee x2~350 eVdet [based on current #’s - see CRESST II talk] Neutron: x1=50 keVee&r x2~50 eVdet (close to noise threshold) Assumes 0.7% energy of electron recoil signal detected in photons for gamma event and QF (established for O nuclei recoils relative to gammas) is 14% Effective threshold raised due to event-by-event discrimination confused by noise In 2nd case - will single event in x 1 only be background free? 1st Kind 2nd Kind For 1st case - (see CDMS talks on surface electron confusion with nuclear recoils)

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Influence of Coherence & Form Factor 73 Ge 40 Ar,~ 32 S 131 Xe, 127 I Rate kg -1 day -1  = cm 2, m=300 GeV) Recoil Energy Threshold [keVr] Xe=10 keVee (20%QF) I=4.5 keVee (9%QF) Ge=15 keVee (30%QF) (q=0) Rate~A 2 Solid line is integrated rate above threshold (dotted is differential rate)

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Collaboration & Funding Issues Strong interest among CDMS Collaborators Factor of 2 – 3 expansion in scale  Not costed or scheduled in any detail (requires technology/production studies)  Likely to require international collaboration Include cryo detector groups in Europe Funding – CDMS model  NSF  DoE Labs (LBNL, FNAL)  DoE University Programs  ++ International Component Timescale  CDMS II detector production and test facilities begin to ramp down end of  Reasonable to ramp up development and use these facilities + new facilities while taking data with CDMS II Experiment up and running. Therefore would seek detector development funding ~2002-4, project funding ~2005.  Other factors include CDMS II technical and scientific developments, as well as results from the rest of the community.

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Direct Detection: History & Future Oroville (88) [m = ?? GeV - if significantly better limit obtained at different mass] 90% CL Limit on Cross section for 60 GeV WIMP (scalar coupling) CDMS SUF (T) CDMS Soudan (T) 7 kg Ge+Si Cryodet GENINO (T) 100 kg Ge Diode GENIUS (T) 100 kg Ge Diode CryoArray (T) tonne Cryodet ~1 event kg -1 day -1 ~1 event kg -1 yr -1 ~1 event 100 kg -1 yr -1 LHC Not meant to be a complete list - see Different Colours Indicate Different Technologies NO W rjg Ge NaI Cryodet (T) Target Signal Liq Xe [m=20 GeV] Homestake (87) H-M (94) H’berg-Moscow (98), IGEX (00) DAMA (96) UKDMC (96) [m=100 GeV] DAMA (98) DAMA (00) CDMS SUF (99) CDMS SUF (00) Edelweiss (98) Edelweiss (01) Gaitskell (astroph )

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Current SUSY Theory ~ 1 event/100 kg/yr CDMS (Feb 2000) ~1 event/kg/d CryoArray 1 tonne event by event discrim. ~1 event/kg/yr

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell Projected Experimental & Theoretical Regions CDMS II Soudan / Edelweiss II GENIUS TF 2 keV Ellis et al. Gondolo… (g-2) Mandic et al UKDMC NaIAD From limitplot rjg Updated IGEX CryoArray GENIUS GENIUS TF 11 keV GENINO

One Tonne Dark Matter Snowmass July 2001 Rick Gaitskell One Tonne: Conclusion Discriminating Detectors  Allows sharing of improvement budget (target 10 4 over current sensitivity) Good discrimination uses Ge/Si (expensive) target very efficiently  Will allow us to study WIMP physics (~100 events) at  ~ cm2, with background ~100 events characterised 5  (Even better for higher cross- sections) CryoArray (based on CDMS technology) see Gaitskell/astro-ph  Discrimination:  good enough,  within factor few Based on extensive underground running of Ge/Si detectors (1996->) so stats reflect real world scenario  Backgrounds: x20 reductions vs CDMS II, x3 vs best current Heidelberg- Moscow Modest b/g reduction should be possible through materials selection/simplification of structures (little mounting material)  Scaling up to 1000’s detectors CDMSII 42 dets (250 g Ge) / CUORE -> 1000 dets Fabrication of dets in commercial environment is the challenge 10 year development : means opportunity to capitalise on technology