light dm in the light of cresst-ii

Transcription

1 light dm in the light of cresst-ii Jure Zupan U. of Cincinnati based on T. Schwetz, JZ ; J. Kopp, T. Schwetz, JZ

2 the question CoGeNT, DAMA, CRESST claim signals Is it (can it be) dark matter? three important results at TAUP CRESST-II result surface events in CoGeNT KIMS result Bernabei et al. [DAMA], Aalseth et al. [CoGeNT], ; Angloher et al., [CRESST-II],

3 outline elastic spin independent scattering several comments on experimental results inelastic DM and isospin violating DM parameter goodness of fit test 3

4 spin independent scattering WIMP nucleon cross section ΣSI cm Xenon 100 DAMA CRESST II Limits: 90 Contours: 90, 3Σ CRESST comm. run CDMS low thr KIMS CDMS v km s, v esc 550 km s Kopp, Schwetz, JZ, m Χ GeV 4

5 modulation in CoGeNT? events / kev kg day kev kev fit with no modulation has acceptable goodness-of-fit:! 2 no mod is 9.5, 16.8, 11.7 for 15 d.o.f. for 3 eng. bands! 2 no mod=20 for 15 d.o.f. for kev (17%) 2.8" preference for modulation 5! 2 no mod! 2 cos! 2 8 = 14.5 cosine fit 10.5 GeV, 9.5e-41 cm 2 8 GeV, 7.4e-41 cm 2 = 16.8 / 15 = 7.7 / 13! = kev! 2 = 9.5 / 15 no mod 2! 2 = 5.9 / 13 0 cos -2! 2-4! = = days since Dec Schwetz, JZ, Aalseth et al. [CoGeNT],

6 phase of the modulation Schwetz, talk at PPC11, Cern, June

7 variations of phase peak at day 152 in standard halo model only variations especially pronounced for light DM if experiments compared within the same v min range the phase should be the same 7 Kuhlen et al.,

8 10-39 comparing dama and cogent for esi CoGeNT (and DAMA) to be consistent with the rest one would need L eff drop to zero below measurements q Na should be significantly larger energy calibration of Xenon10 S2 analysis needs to be off CDMS made a major calibration error (in Ge and Si) 90%, 99% CL (2 dof) v 0 = 230 km/s, v esc = 550 km/s DAMA/LIBRA q = 0.3 +/ Na q Na = 0.5 +/- 0.1 Hooper, Collar, Hall, McKinsey, ; Hooper, Kelso, Xenon100 L eff =0 below 4keV see talk by J. Collar " p [cm 2 ] CoGeNT (unmod) CoGeNT (modulation) DAMA CDMS Si 05 Ge low-thr 10 CoGeNT XENON100 low L eff m! [GeV] Schwetz, JZ Q y variation Farina, Pappadopulo, Strumia, Volansky,

9 surface events CoGeNT found that bulk of their signal is due to surface event will there be a signal left once results finalized? what is the effect of this contamination on best fit? makes the disagreement between CoGeNT and DAMA worse due to this uncertainty will disregard CoGeNT 9

10 surface events CoGeNT found that bulk of their signal is due to surface event will there be a signal left once results finalized? J. Collar, talk at TAUP, Sept 2011 what is the effect of this contamination on best fit? makes the disagreement between CoGeNT and DAMA worse due to this uncertainty will disregard CoGeNT 9

11 surface events CoGeNT found that bulk of their signal is due to surface event will there be a signal left once results finalized? what is the effect of this contamination on best fit? makes the disagreement between CoGeNT and DAMA worse due to this uncertainty will disregard CoGeNT 9

12 surface DAMA q ± 10 events CoGeNT found that bulk of their signal is due to surface event will there be a signal left once results finalized? what is the effect of this contamination v on best esc 550 km s fit? WIMP nucleon cross section ΣSI cm CDMS low thr XENON 100 makes Kopp, the Schwetz, disagreement JZ, between CoGeNT and DAMA worse WIMP m ass m Χ GeV Lim its: 90 Countours: 90, 3Σ CoGeNT n o con tam. CoGeNT sm all con tam. CoGeNT large con tam. v km s due to this uncertainty will disregard CoGeNT 9

13 new kims bound CsI detector the bound excludes any signal from I for DAMA KIMS rate( kev) < cnts/kg/day/ kev (90% C.L.) DAMA S m (2-4 kev)=0.0183± cnts/kg/ day/kev! if DAMA is due to DM then scattering on Na if due to DM! points to light DM 10

14 DAMA, new kims bound CsI detector the bound excludes any signal from I for DAMA KIMS rate( kev) < cnts/kg/day/ NR evet rates (PSD analysis) kev (90% 1 year C.L.) data excluding det0, 8, 11 (Total exposure: kg days) DAMA S m (2-4 kev)=0.0183± cnts/kg/ day/kev! if DAMA is due to DM then scattering on Na if due to DM! points to light DM Sun Kee Kim, talk at TAUP

15 new kims bound CsI detector the bound excludes any signal from I for DAMA KIMS rate( kev) < cnts/kg/day/ kev (90% C.L.) DAMA S m (2-4 kev)=0.0183± cnts/kg/ day/kev! if DAMA is due to DM then scattering on Na if due to DM! points to light DM 10

16 sodium quenching factor what is the quenching factor qna? smaller than previous measurements moves the best fit point to higher DM masses dominated by scattering on iodine 11

17 sodium quenching factor J. Collar, talk at TAUP what is the quenching factor qna? smaller than previous measurements moves the best fit point to higher DM masses dominated by scattering on iodine 11

18 sodium quenching factor what is the quenching factor qna? smaller than previous measurements moves the best fit point to higher DM masses dominated by scattering on iodine 11

19 sodium quenching factor what is the quenching factor qna? smaller than previous measurements moves the best fit point to higher DM masses WIMP n ucleon cross section ΣSI cm CDMS low th r DAMA XENON 100 CDMS WIMP m ass m Χ GeV dominated by scattering on iodine KIMS Kopp, Schwetz, JZ, DAMA m odified q Na Na recoils on ly DAMA m odified q Na Na I recoils Lim its: 90 Cou n tou rs: 90,3Σ 11

20 CRESST Angloher et al. [CRESST-II], CaWO 4 crystal, 730 kg days of data 67 events observed in the signal region background from e/# events, $ events, neutron events, Pb recoil signal ~30±9 evnts (24±8 evnts) the significance of extra signal is >4! 12

21 CRESST Angloher et al. [CRESST-II], CaWO 4 crystal, 730 kg days of data 67 events observed in the signal region background from e/# events, $ events, neutron events, Pb recoil signal ~30±9 evnts (24±8 evnts) Angloher et al. [CRESST-II], the significance of extra signal is >4! 12

22 CRESST Angloher et al. [CRESST-II], CaWO 4 crystal, 730 kg days of data 67 events observed in the signal region background from e/# events, $ events, neutron events, Pb recoil signal ~30±9 evnts (24±8 evnts) the significance of extra signal is >4! 12

23 questions elastic spin-independent with standard halo: not a good fit to all experiments could a nongeneric DM model be viable? could nuclear and astrophysical uncertainties make a big difference? 13

24 dark matter variations a number of variations on DM one can consider isospin violating couplings velocity suppressed interactions inelastic scattering endothermic, exothermic scattering through resonances additional momentum dependence light mediators, derivative interactions,... leptophilic interactions spin dependent interactions... Kamionkowski, Kurylov, hep-ph/ ; Giuliani, hep-ph/ ; Cotta et al., ; Kang et al., ; Feng et al., ; Chang et al., ; Frandsen et al., Tucker-Smith, Weiner, hep-ph/ Graham, Harnik, Rajendran, Saraswat, Bai, Fox, Feldstein, Fitzpatrick, Katz, ; Chang, Pierce, Weiner, Kopp, Niro, Schwetz, JZ,

25 other uncertainties astrophysical uncertainties see talk by A. Green vary velocity profiles, v esc, etc integrate them out Fox, Liu, Weiner Fox, Kribs, Tait, channeling Bozorgnia, Gelmini, Gondolo, ; ; nuclear and atomic physics quenching factors, L eff in S1, Q y in Xenon S2 nuclear form factor uncertainties 15

26 tensions a complete check for old CoGeNT and DAMA was done by four groups the bottom line: none of the models can make both signals +other bounds work including uncertainties focus on CRESST-II Schwetz, JZ ; Farina, Pappadopulo, Strumia, Volansky ; Fox, Kopp, Lisanti, Weiner ; Hooper, Kelso ; McCabe could it be due to DM? 16

27 inelastic dark matter CRESST target contains W, is heavy if scattering of DM inelastic scattering on light nuclei no signal ( ) 1 mn E v min = d + δ 2mN E d µ χn W the heaviest, there is a solution, %~90keV maybe tension with Xenon-100 (PG test values of 2%-20%) very dependent on astrophysics details a coincidence problem 17

28 inelastic dark matter Limits: Contours: 90, 3Σ WIMP nucleon cross section ΣSI cm 2 CRESST target contains W, is heavy if scattering of DM inelastic CRESST comm. run scattering on light nuclei CRESST II no signal ( ) 10 1 mn E v min = 40 d + δ 2mN E d µ χn 10 W the heaviest, 41 there is a solution, %~90keV Kopp, Schwetz, JZ, m Χ GeV maybe tension with Xenon-100 (PG test values of 2%-20%) very dependent on astrophysics details a coincidence problem Xenon 100 KIMS idm, 90keV DAMA v km s, v esc 550 km s 17

29 inelastic dark matter CRESST target contains W, is heavy if scattering of DM inelastic scattering on light nuclei no signal ( ) 1 mn E v min = d + δ 2mN E d µ χn W the heaviest, there is a solution, %~90keV maybe tension with Xenon-100 (PG test values of 2%-20%) very dependent on astrophysics details a coincidence problem 17

30 inelastic dark matter CRESST target contains W, is heavy if scattering of DM inelastic scattering on light nuclei no signal CRESST CRESST ( ) 50 com 1 XE100 CDMS mn E v min = LT d + δ 2mN E d µ χn W the heaviest, there is a solution, %~90keV Χ maybe tension 40 with Xenon-100 (PG test values of 2%-20%) 35 CRESST CRESST com very dependent on astrophysics details a coincidence 30 problem kev Kopp, Schwetz, JZ,

31 inelastic dark matter CRESST target contains W, is heavy if scattering of DM inelastic scattering on light nuclei no signal ( ) 1 mn E v min = d + δ 2mN E d µ χn W the heaviest, there is a solution, %~90keV maybe tension with Xenon-100 (PG test values of 2%-20%) very dependent on astrophysics details a coincidence problem 17

32 isospin violating dm DM could couple differently to p than n counter example: higgs mediated scattering dominated by strangeness content if couplings to u, d dominate IVDM possible phenomenologically interesting if couplings such that scattering on Xe suppressed not possible to suppress it completely in detector several isotopes 18

33 IVDM and CRESST define effective atomic number here tan""f n /f p Xe suppressed for f n /f p =-0.7 but then Si, Ge large also W suppressed, Ca and O large in CRESST-II scattering exclusively on Ca an O now 39% scattering on O, 61% on Ca exactly opposite to idm solution where only W 19

34 IVDM and CRESST define effective atomic number A 2 eff / A Si / Ca / O Ge Xe here tan""f n /f p Xe suppressed for f n /f p =-0.7 but then Si, Ge large f n / f p also W suppressed, Ca and O large in CRESST-II scattering exclusively on Ca an O now 39% scattering on O, 61% on Ca exactly opposite to idm solution where only W 10-4 Schwetz, JZ, Na I W 19

35 IVDM and CRESST WIMP nucleon cross section Σeff cm CRESST II DAMA IVDM, f n f p 0.7 Limits: 90 Contours: 90, 3Σ KIMS CRESST comm. run CDMS low thr CDMS Si Xenon 100 CDMS v km s, v esc 550 km s Kopp, Schwetz, JZ, m Χ GeV 20

36 IVDM and CRESST 55 WIMP nucleon cross section Σeff cm CRESST II DAMA Χ IVDM, f n f p 0.7 Limits: 90 Contours: 90, 3Σ KIMS CRESST comm. run CDMS low thr CDMS Si Xenon CDMS v km s, v esc 550 km s Kopp, Schwetz, JZ, CRESST CRESST com XE100 CDMS LT CDMS Si CRESST CRESST com f n f p Kopp, Schwetz, JZ, m Χ GeV 20

37 spectrum pick three points close to minima for the three models idm - reasonable description of data but only 12.2 signal events predicted scattering off W exclusively esi - much better description with 24.8 events scattering from W negligible, mostly off O and Ca IVDM - again smaller signal predicted, O(10) evnts. scattering mostly off O and Ca 21

38 number of events spectrum 8 pick three points 6 close to minima for the three models 4 idm - reasonable description of data but only 12.2 signal events 2predicted scattering off W exclusively esi - much better description with E nr [kev] 24.8 events scattering from W negligible, mostly off O and Ca IVDM - again smaller signal predicted, O(10) evnts. scattering mostly off O and Ca 21 esi m! = 12.5, " = 2.7e-41 m! = 14.2, " = 2.6e-40, f n / f p = -0.7 m! = 35, " = 2.1e-38, # = 95 signal+background IVDM idm

39 spectrum pick three points close to minima for the three models idm - reasonable description of data but only 12.2 signal events predicted scattering off W exclusively esi - much better description with 24.8 events scattering from W negligible, mostly off O and Ca IVDM - again smaller signal predicted, O(10) evnts. scattering mostly off O and Ca 21

40 PG test Maltoni, Schwetz, hep-ph/ parameter goodness of fit test based on! 2 function i=1,2 datasets whose compatibil. are tested! 2 PG gives the penalty due to additional data if follows a! 2 distribution 22

41 PG test 23

42 conclusions CRESST could be a signal of DM, if scattering inelastic or isospin violating J. Zupan Two topics on dark matter Virginia Tech, Nov 14, 2011