Particle Models for DM (SUSY Wimps)
Total Page:16
File Type:pdf, Size:1020Kb
Particle Models for DM (SUSY WIMPs) !"#$%&'()*+%&)+)*+',- David G. Cerdeño TAUP 2013, Monterey 10 – 09 – 2013 DM is a necessary and abundant component of the Universe Good candidates for Dark Matter have to fulfil the following conditions • Neutral • Stable on cosmological scales • Reproduce the correct relic abundance • Not excluded by direct/indirect searches • No conflicts with BBN or stellar evolution • In agreement with LHC bounds Many candidates in Particle Physics • Axions • Weakly Interacting Massive Particles (WIMPs) • Asymmetric DM • SuperWIMPs and Decaying DM • WIMPzillas • SIMPs, CHAMPs, SIDMs, ETCs... ... they have very different properties Dark matter can be searched for in different ways Accelerator (DM production) LHC (ILC) Searches Direct Detection Super (DM-nuclei scattering) WIMPs DAMA/LIBRA Axion-like Light DM CDMS, SuperCDMS particles WIMP XENON KIMS COUPP PICASSO R DDM ZEPLIN Super CoGeNT Heavy DM CRESST SIMPLE ZEPLIN ANAIS Indirect Detection XMASS (DM annihilation) ... PAMELA ANTARES Fermi IceCube MAGIC CTA AMS ... probing different aspects of the DM interactions with ordinary matter Accelerator (DM production) LHC (ILC) Searches collider searches Direct Detection (DM-nuclei scattering) q Constraints in one sector DAMA/LIBRA might affect observations in CDMS, SuperCDMS the other two. XENON KIMS COUPP “Redundant” detection can be used to extract DM PICASSO ( ) ZEPLIN q properties. CoGeNT CRESST direct searches COMPLEMENTARITY SIMPLE of DM searches ZEPLIN indirect searches via antiprotons ANAIS Indirect Detection XMASS (DM annihilation) ... PAMELA ANTARES Fermi IceCube MAGIC CTA AMS ... Possible hints of light WIMPs (mX~10 GeV) • DAMA/LIBRA (NaI) Annual modulation signal (cumulative exposure 427,000 kg day) DAMA/LIBRA Coll. ‘10 • CoGeNT (Ge) Irreducible background that can be compatible with 7-10 GeV WIMPs ... with annual modulation Collar et al. ‘10- ‘13 • CRESST II (CaWO4) (730 kg day) Excess over the known background 4 Angloher et al. 1109.0702 served in Detector 3 of Tower 5. These detectors were 1039 103 near the middle of their respective tower stacks. Fig. 2 • CDMS (Si) (140.2 kg days) 2 illustrates the distribution of events in and near the sig- pb cm nal region of the WIMP-search data set before (top) and DAMA/LIBRA (NaI) 3 events Agnese et al. 1304.4279 40 4 after (bottom) application of the phonon timing criterion. 10 10 CoGeNT (Ge) XENON10 section Fig. 3 shows an alternate view of these events, expressed section (Xe) in “normalized” versions of yield and timing that are CRESST (CaWO4) 41 5 cross cross 10 10 transformed so that the WIMP acceptance regions of all CDMS II detectors coincide. (Si) Reconstruction of the compatible After unblinding, extensive checks of the three candi- nucleon regions in the WIMP Spin- nucleon 42 6 10 10 date events revealedindependent no data qualitycross section or analysis vs mass issues that would invalidate them as WIMP candidates. The XENON100 (Xe) CDMS II WIMP signal-to-noise on the ionization channel for the three WIMP (Ge) events (ordered in increasing recoil energy) was measured 1043 107 to be 6.7σ, 4.9σ, and 5.1σ. A study on possible leakage 5 6 7 8 9 10 15 20 30 40 50 into the signal band due to 206Pb recoils from 210Po de- WIMP Mass GeV c2 cays found the expected leakage to be negligible with an upper limit of < 0.08 events at the 90% confidence FIG. 4. Experimental upper limits (90% confidence level) for level. The energy distribution of the 206Pb background the WIMP-nucleon spin-independent cross section as a func- was constructed using events in which a coincident α par- tion of WIMP mass. We show the limit obtained from the ticle was detected in a detector adjacent to one of the 8 exposure analyzed in this work alone (blue dotted line), and Si detectors used in this analysis. combined with the CDMS II Si data set reported in [23, 28] (blue solid line). Also shown are limits from the CDMS This result constrains the available parameter space II Ge standard [17] and low-threshold [29] analysis (dark of WIMP dark matter models. We compute upper lim- and light dashed red), EDELWEISS low-threshold [30] (long- its on the WIMP-nucleon scattering cross section using dashed orange), XENON10 S2-only [31] (dash-dotted green), Yellin’s optimum interval method [25]. We assume a and XENON100 [32] (long-dash-dotted green). The filled re- WIMP mass density of 0.3 GeV/c2/cm3, a most probable gions identify possible signal regions associated with data WIMP velocity with respect to the galaxy of 220 km/s, from CoGeNT [33] (dashed yellow,90%C.L.),DAMA/LIBRA a mean circular velocity of Earth with respect to the [10, 34] (dotted tan,99.7%C.L.),andCRESST[12,35](dash- galactic center of 232 km/s, a galactic escape velocity of dotted pink, 95.45% C.L.) experiments. 68% and 90% C.L. 544 km/s [26], and the Helm form factor [27]. Fig. 4 contours for a possible signal from these data are shown in light blue. The blue dot shows the maximum likelihood point shows the derived upper limits on the spin-independent 2 41 2 at (8.6 GeV/c ,1.9 10− cm ). WIMP-nucleon scattering cross section at the 90% con- × fidence level (C.L.) from this analysis and a selection of other recent results. The present data set an upper limit 41 2 2 of 2.4 10− cm for a WIMP of mass 10 GeV/c .We rates were treated as nuisance parameters and the WIMP are completing× the calibration of the nuclear recoil energy mass and cross section were the parameters of interest. scale using the Si-neutron elastic scattering resonant fea- We profiled over probability distribution functions of the ture in the 252Cf exposures. This study indicates that our rate for each of our known backgrounds. The highest like- reconstructed energy may be 10% lower than the true re- lihood was found for a WIMP mass of 8.6 GeV/c2 and 41 2 coil energy, which would weaken the upper limit slightly. a WIMP-nucleon cross section of 1.9 10− cm .The Below 20 GeV/c2 the change is well approximated by goodness-of-fit test of this WIMP+background× hypoth- shifting the limits parallel to the mass axis by 7%. In esis results in a p-value of 68%, while the background- addition, neutron calibration multiple scattering∼ effects only hypothesis fits the data with a p-value of 4.5%. improve the response to WIMPs by shifting the upper A profile likelihood ratio test finds that the data favor limit down parallel to the cross-section axis by 5%. the WIMP+background hypothesis over our background- ∼ A model of our known backgrounds, including both only hypothesis with a p-value of 0.19%. Though this energy and expected rate distributions, was constructed result favors a WIMP interpretation over the known- for each detector and experimental run for each of the background-only hypothesis, we do not believe this result three backgrounds considered: surface electron recoils, rises to the level of a discovery. neutron backgrounds, and 206Pb recoils. Simulations of Fig. 4 shows the resulting best-fit region from this our background model yield a 5.4% probability of a sta- analysis (68% and 90% confidence level contours) on tistical fluctuation producing three or more events in our the WIMP-nucleon cross-section vs. WIMP mass plane. signal region. The 90% C.L. exclusion regions from CDMS II’s Ge This model of our known backgrounds was used to in- and Si analyses and EDELWEISS low-threshold analy- vestigate the data in the context of a WIMP+background sis cover part of this best-fit region, but the results are hypothesis. We performed a profile likelihood analysis, overall statistically compatible. There is much stronger including the event energies, in which the background tension with the upper limits from the XENON10 and Non-observation in other experiments set upper bounds on the cross section XENON10, XENON100 (Xe), CDMS-II (Ge), Edelweiss (Ge), COUPP (CF3I) have not observed any DM signal, which constrains the scattering cross section These bounds are in tension with the other observations XENON10 (Xe) CoGeNT (Ge) CDMS (Si) DAMA/LIBRA (NaI) CRESST (CaWO4) Ge) - 2009 CDMS II ( Xe) - 2012 XENON100 ( Assumptions: Next generation experiments • “standard” WIMP with only SI interactions (SuperCDMS, XENON1T) will explore further regions of this • Elastic scattering parameter space • Coupling to protons = coupling to neutrons • “Standard Halo Model” 2 2 2 -3 section is dσ/dER =ˆσAmA/(2v µA), with 10 2 µA p n 2 σˆA = 4 [fpZFA(ER)+fn(A Z)FA(ER)] , (2) M∗ − where fp,n are the couplings to protons and neutrons, p,n 10-4 normalized by the choice of mass scale M∗,andFA (ER) are the proton and neutron form factors for nucleus A. p n p FA(ER)andFA(ER) are not identical. FA(ER) is (pb) n what has typically been measured, but FA(ER) may also Z N be probed, for example, through neutrino and electron ! -5 parity-violating scattering off nuclei [14]. However, since 10 CDMS-Si (SUF) the isospin violation from this effect is small compared to CDMS-Ge (SUF) 2 CDMS-Ge (Soudan) the potentially large effects of varying fn/fp, we will set XENON100 (2011) XENON10 (2011) 2 both2 form factors equal to FA(ER).-3 With this approxi- DAMA (Savage et al.) section is dσ/dER =ˆσAmA/(2v µ ), with 10 CoGeNT (2010) mation,A the event rate simplifies to R = σAIA,where 2 2 2 10-6 µA p n µA 2 2 7 8 9 10 11 12 13 σˆ = [f ZF (E )+f (A Zσ)FA =(E )][f,pZ +(2)fn(A Z)] 2 2 (3) -3 A 4 p A R n A sectionR 4 is dσ/dER =ˆσAmA/(2v µ ), with 10 M∗ − M∗ − A m (GeV) vmax X m -1 µ2 3 A 2 10 where fp,n are the couplings to protonsIA = andNT n neutrons,X dEAR dp vf(v) 2 FA(ERn) , (4)2 σˆAp,n!= 4 ![fvpminZFA(ER)+-42fvµn(AA Z)FA(ER)] , (2) M∗ 10 − normalized by the choice of mass scale M∗,andFA (ER) 2 and σA is the zero-momentum-transfer SI cross section are the proton and neutron form factors for nucleus A.