DAMA Slides/Graphics

Transcription

1 DAMA Slides/Graphics Rick Gaitskell Center for Particle Astrophysics UC Berkeley source at Gaitskell

2 Best fit to Ann Mod data alone Best Fit Minimum DAMA NaI/1-4 DAMA NaI/1-4 (3σ) New CDMS limit Modulation Amplitude Best-fit WIMP model s expected annual modulation does not appear to fit data; lowest point of 3σ contour is much worse. Why? Additional constraint applied during max likelihood analysis: DC WIMP signal implied by AC signal must not exceed observed DC count rate best-fit cross-section is decreased mean over 2-6 kevee (22 66 kev recoil) Best fit to Ann Mod data alone DAMA 2000 paper Figure 2 DAMA 1 DAMA 2 DAMA DAMASummaryGaitskell3.pp June ,000 kg-day 15,000 kg-day 38,000 kg-day Rick Gaitskell

3 DAMA - Cartoon of Lowest Energy Bins 30,000 evts 12,000 evts 6,000 evts ~600 days x 100 kg ~300,000 b/g events in range kev recoil ~50,000 WIMP events (latter expected in sample but are not observed directly) DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

4 Bernabei et al. ROM2F/98/34, INFN/AE-98/20, LNGS Preprint DAMA - Energy Spectra Data re-plotted by Gaitskell, CfPA Rate corresponding to best fit to Ann Mod data alone (σ = cm 2 ) is shown as magenta dashed line - exceeds signal in 2-3 kev bin green: WIMP component assuming (50 GeV,7.2x10-42 cm 2 ) red points: individual spectra red line: average spectrum Very low background Very low background Energy is kevee, not recoil energy: 9% iodine quenching factor Plotted spectra are already efficiency-corrected Possible problems: Signal entirely determined by 2-3 and 3-4 kevee bins Large drop in background in lowest bin; worsened by large inferred WIMP component Strange background shape; would expect flat or rising Large fluctuations in low energy background Consistency of signal among detectors? % is ratio of 2-3 kev bin background to 3-4 kev background rate DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

5 DAMA Sensitivity of Different Energy Bins The shape of DAMA Signal Region is determined by gross sensitivity of experiment, not by any WIMP mass selectivity in analysis Reduction in sensitivity corresponds to pivot point of modulation WIMP spectrum passing through that bin DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

6 Compatibility of CDMS and DAMA CDMS results incompatible with DAMA Figure 2 data (left) at > 99.98% CL Best simultaneous fit to CDMS and DAMA predicts too little annual modulation in DAMA, too many events in CDMS DAMA 2000 paper Figure 2 CDMS data predicted spectrum Estimate full DAMA likelihood function: Two experiments are incompatible at 99.8% CL Ignore multiple scatters: 96.8% CL DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

7 CDMS & Expected WIMP Spectra DAMA NaI/1-4 best-fit WIMP (52 GeV, cm 2 ): CDMS expects 20 WIMP events DAMA NaI/1-4 lowest 3σ contour (40 GeV, cm 2 ): CDMS expects 6 WIMP events Expected WIMPs Detected CDMS NR spectrum after subtraction of most likely neutron background Recoil Energy (kev) WIMP spectra corrected for CDMS efficiency function DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

8 Slides of Location DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

9 Gran Sasso National Laboratory (LNGS) DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

10 Gran Sasso National Laboratory (LNGS) First Director: Zichichi CUORE Large Volume Detector (SN Det) (Opera) (will be removed in 2001) (MiBeta - Fiorini Group) HDMS Heidelberg-Moscow & HDMS DAMA DAQ DAMA NaI ICARUS First Module 0.6kt Sept 99 CRESST DAMA Updated DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

11 DAMA - Experimental Apparatus Very elegant experimental setup - in place >1996 Low Activity NaI scintillator kg NaI crystals, each viewed by 2 PMTs Located at Gran Sasso Underground Lab (4000 mwe) + Photon and Neutron shielding Two modes of Background discrimination Pulse shape Annual modulation: ~2% modulation amplitude PMT NaI NaI NaI NaI PMT DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

12 DAMA NaI Experiment Run 100 kg of ultrapure NaI Now 58,000 kg-days See annual modulation signal DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

13 In depth/technical slides DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

14 DAMA - Energy Histograms from NaI Array Lowest energy bin is most important when setting dark matter limit Earlier paper contained spectra for all 9 detectors separately (Note log plot covers 6 orders of magnitude) Dark Matter sensitivity is determined primarily by behaviour of bottom bins covering 2-3keVee (which is recoil energy, after adjusting for nuclear recoil quenching factor on I nuclei of 0.09) Bernabei et al. ROM2F/98/34, INFN/AE-98/20, LNGS Preprint DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

15 DAMA - Energy Resolution/Trigger Issues Bernabei et al. ROM2F/98/34, INFN/AE-98/20, LNGS Preprint Blue histograms show estimated resolution at 46.5 kev, 13 kev and 2.5 kev 210 Pb 46.5 kev X-ray [kev] = 58.5% E / kev Data re-plotted by Gaitskell, CfPA Spectra incompatible with resolution Energy calibration - requires X% stability Large trigger efficiency correction: trigger stability? Trigger Efficiency 100% >8 kevee 40% in 2-3 kevee bin Drop in spectrum inconsistent with resolution function Energy [kevee] DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

16 Annual Modulation of Rate & Spectrum DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

17 Modulation Animation in NaI 150 GeV WIMP 50 GeV WIMP DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell!

18 Annual Modulation Not distinguish between WIMP signal and Background directly From the amplitude of the modulation, we can calculate the underlying WIMP interaction rate WIMP Signal ±2% Background Dec June Dec June Dec June Dec June DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

19 DAMA Modulation 3 year result ROM2F/2000/01 LNGS Preprint (just released) DAMA 1 5,000 kg-day DAMA 2 15,000 kg-day DAMA ,000 kg-day Rebuilt Shown above is residual modulation in events/kevee/kg/day 2-6 kevee range Total Counts 2-6 kevee WIMPs+Background is ~700 events/day Note that most of modulation is expected in 2-3 kevee bin New data covering 3 year data was released February 2000 LNGS approved (Dec 1999) upgrade to 250 kg NaI array DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

20 DAMA allowed region & SUSY predictions DAMA/1-4 (3σ) ~2.3 event/kgge/day DAMA/1-4 (3σ) Most Likely Fit (52 GeV, cm 2 ) DAMA/1-4 (3σ) Lowest cross-section (40 GeV, cm 2 ) DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

21 DAMA NaI Questions Annual Modulation Experiment It isn t simply a matter of sufficient statistics. if σ~10-5 pb then 100 kg NaI array is sensitive enough However, must also demonstrate that other possible fake sources are not present And, that the control of experiment environment is much better than 1% level of apparent signal E.g. Not only confirm energy calibration, but that stability of width of lowest energy bin is good at <1%) BIN WIDTH/THRESHOLD STABILITY Physical Shape of Raw Background Spectrum Looks odd, especially when compared to other NaI Low Background Experiments DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

22 DAMA NaI Pulse Shape Discrim. Question Pulse Shape Discrimination NOT used in annual modulation identification of 3σ signal Doesn t really help reduce background in lowest energy bin 2-3 kevee (which due to I QF=9% dominates stats.) due to poor signal/noise Saclay and UKDM have both identified additional population of events during background running that are neither electron or nuclear recoils. Anomalous events not been observed in DAMA data, and exclusion limit is quoted accordingly DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

23 DAMA NaI Outlook New data from 2 year run has been published 3 times previous data Have they provided improved comment on systematics? INFN Approved upgrade of DAMA to 250 kg NaI Anomolous Events seen in other NaI Experiments UKDM (Boulby) & Saclay (Modane) NaI experiments Close to an explanation (may be surface contamination) Signal Region Being checked by CDMS DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell

24 Test of Compatibility with DAMA (not using DAMA-0) We have compared CDMS upper limit CL to a region allowed by DAMA signal without including constraint from DAMA's 1996 (NaI-0) upper limit Region including this constraint is made under assumption that DAMA's signal and upper limit are both correct -- but this may not be the case. If CDMS is incompatible with DAMA's signal, either DAMA's signal is wrong or CDMS's limit is wrong. The fact that DAMA has data producing an upper limit should not make their signal appear to be more compatible with other upper limits. The large difference between DAMA's two regions is due to fact that their upper limit is already somewhat incompatible with their signal (although not too significantly); the most likely (M,σ) point for their signal is ruled out by their 90% CL upper limit DAMASummaryGaitskell3.pp June 2000 Rick Gaitskell