ANAIS-112 status: two years results on annual modulation
J. Amaré, S. Cebrián, D. Cintas, I. Coarasa, E. García, M. Martínez, M. A. Oliván, Y. Ortigoza, A. Ortiz de Solórzano, J. Puimedón, A. Salinas, M. L. Sarsa, P. Villar
AANAIS-112 status: two years results on annualmodulation
J. Amar´e , , S. Cebri´an , , D. Cintas , , I. Coarasa , , E. Garc´ıa , ,M. Mart´ınez , , , M.A. Oliv´an , , , Y. Ortigoza , ,A. Ortiz de Sol´orzano , , J. Puimed´on , , A. Salinas , , M.L. Sarsa , and P. Villar , Centro de Astropart´ıculas y F´ısica de Altas Energ´ıas (CAPA), Universidad de Zaragoza,Pedro Cerbuna 12, 50009 Zaragoza, Spain Laboratorio Subterr´aneo de Canfranc, Paseo de los Ayerbe s.n., 22880 Canfranc Estaci´on,Huesca, Spain Fundaci´on ARAID, Av. de Ranillas 1D, 50018 Zaragoza, Spain Fundaci´on CIRCE, 50018, Zaragoza, SpainE-mail: [email protected]
Abstract.
ANAIS (Annual modulation with NaI Scintillators) is a dark matter directdetection experiment located at the Canfranc Underground Laboratory (LSC), in Spain. Thegoal is to confirm or refute in a model independent way the DAMA/LIBRA positive result:an annual modulation in the low-energy detection rate compatible with the expected signalinduced by dark matter particles in the galactic halo. This signal, observed for about 20 years,is in strong tension with the negative results of other very sensitive experiments, but a directcomparison using the same target material, NaI(Tl), was still lacking. ANAIS-112, consisting of112.5 kg of NaI(Tl) scintillators, is taking data at the LSC since August 2017. Here we presentthe preliminary annual modulation analysis corresponding to two years of data (exposure of220.69 kg × y) and the ANAIS-112 projected sensitivity for the scheduled 5 y of operation.
1. Introduction
Large experimental efforts devoted to unraveling the nature of the dark matter (DM) particleshave been carried out, either by direct [1], indirect [2] or accelerator searches [3], which arecomplementary to each other. For about twenty years, the DAMA/LIBRA collaboration hasbeen claiming the observation of an annual modulation in the detection rate, which fulfills all ofthe requirements expected for the contribution of weakly interacting DM particles, distributedin the Milky Way halo. DAMA/LIBRA detector is installed at Gran Sasso UndergroundLaboratory (LNGS), in Italy, and it consists of highly radiopure NaI(Tl) scintillators, having atotal mass of 250 kg [4]. The current statistical significance of the DAMA/LIBRA modulationresult reaches the 12 σ level. However, it has neither been reproduced by any other experiment,nor ruled out in a model independent way [5, 6, 7, 8, 9, 10]. Compatibility among the differentexperimental results in most conventional WIMP-DM scenarios is actually disfavored [11, 12].Other experiments using the same target are crucial to ascertain whether the DAMA/LIBRApositive signal is a signature of the halo dark matter particles or some systematics. There areseveral efforts around the world pursuing this goal [13, 14, 15]. COSINE-100 and ANAIS-112 a r X i v : . [ a s t r o - ph . I M ] O c t xperiments are presently in data taking phase, and both have already published preliminaryanalysis. COSINE-100 experiment is located at the Yangyang Underground Laboratory, in SouthKorea and it consists of 106 kg of NaI(Tl) detectors, immersed in a liquid scintillator, whichallows the identification and subsequent reduction of some radioactive backgrounds. An analysisof the first 59.5 live days using about 61 kg effective mass allowed the COSINE collaborationto exclude the 3 σ DAMA/LIBRA region at 90% C.L. in a model dependent way [10], and theannual modulation analysis of 1.7 years, with a total exposure of 97.7 kg × y, recently publishedis compatible at 1 σ both with DAMA/LIBRA signal and with the absence of modulation [16].ANAIS-112 experiment is taking data at the Canfranc Underground Laboratory in Spain sinceAugust 2017. It consists of 112.5 kg of NaI(Tl)detectors, disposed in a 3x3 array of modules,12.5 kg each. Most relevant features of ANAIS modules, built by Alpha Spectra Inc., are theMylar window built-in to allow low energy calibration with external sources, and outstandingoptical quality, which added to the high-efficiency Hamamatsu photomultipliers (PMTs) enablea light collection at the level of 15 photoelectrons (phe) per keV in all the modules. The ANAIS-112 shielding consists of 10 cm of archaeological lead, 20 cm of low activity lead, an anti-radonbox (kept under overpressure with radon-free nitrogen gas) and 40 cm of a combination of watertanks and polyethylene bricks. An active muon veto made up of 16 plastic scintillators coversthe top and sides of the set-up [17]. ANAIS-112 DAQ hardware and software is robust and it hasbeen fine-tuned in the operation of several prototypes. The signals from the two PMTs coupledto each NaI(Tl)-module are fully processed: signal is divided into a trigger signal, a high-energysignal, and for the low-energy signal the waveform is digitized at 2GS/s with high resolution.Trigger is done by the coincidence (logical AND) of the two PMT trigger signals in a 200 nswindow, while the trigger of each PMT is at phe level. Cd sources are used every two weeks tocalibrate the experiment and correct, if necessary, for possible gain drifts. Energy calibration isdone by combining the information from the
Cd lines and K and Na crystal contaminationsenergy depositions at very low energy, being the latter tagged by coincidences with high-energygammas, and corresponding to 3.2 and 0.87 keV. This procedure allows calibrating ANAIS-112down to the threshold with high accuracy. Triggering below 1 keV with almost 100% efficiencyis guaranteed by the observation of the Na events population. However, analysis threshold isset up at 1 keV because of the PMT-related events rejection, which forces to establish eventsselection criteria whose acceptance efficiencies for bulk scintillation events decrease down toabout 15% at 1 keV [17].
2. Annual modulation analysis and results
Energy and time distribution from single-hit events in the ROI is kept blinded in our analysisprotocol. After the first year of data taking we unblinded 10% of those events, chosen randomlydistributed days along the data taking, for fine tuning the events rejection procedures andcarry-out general background assessment and sensitivity estimates. All of this information waspublished [17, 18, 19]. After 1.5 years, using the designed analysis protocol, the unblinding ofthe ROI and annual modulation analysis allowed the confirmation of the sensitivity projections,while producing some tension with DAMA/LIBRA annual modulation result [20]. This resultcorresponded to an effective exposure of 157.55 kg × y.After 2 years, using the same analysis protocol, we have carried out a new unblinding ofANAIS-112 data and the corresponding annual modulation analysis. While preparing a morecomplete analysis, we present here the result of a model independent analysis searching formodulation in the same regions as DAMA/LIBRA has published ([1-6] keV and [2,6] keV)using an exposure of 220.69 kg × y. Data from all the modules are added together and we use aleast-squared fit, modeling the data as: All the energies used throughout this paper are electron equivalent energies. igure 1.
ANAIS-112 fit results for two years of data in energy regions [1-6] and [2-6] keVin the modulation (blue) and null hypothesis (red). Data are displayed after subtracting theconstant and exponential functions fitted to eq. 1. DAMA/LIBRA results are shown in greenfor comparison. R ( t ) = R + R · exp ( − t/τ ) + S m · cos ( ω · ( t − t )) (1) R and R are free parameters, but τ is fixed to the value obtained for the time evolution of ourbackground model in the corresponding energy range. Period and phase are fixed to the expectedvalues for the standard galactic halo: 1 y and maximum at June, 2, respectively. S m is fixed tozero for the null hypothesis and left unconstrained for the modulation hypothesis. Figure 1 showsthe fit results of detection rate in the ROI to eq 1. The non-modulated components have beensubtracted in the graphical representation. Results are consistent with the null hypothesis, andbest fits for the modulation hypothesis are compatible with the absence of modulation within1 σ . Moreover, those best fits are incompatible with DAMA/LIBRA results at about 2.6 σ .These results, together with the sensitivity projections, are summarized in Figure 2. We quoteour sensitivity to DAMA/LIBRA as the ratio of DAMA modulation result over the standarddeviation on the modulation amplitude derived from ANAIS-112 data ( S DAMAm /σ ( S m )). Resultsfrom two-year data of ANAIS-112 confirm our sensitivity prospects [19], being at present at 2 σ level, and supporting our goal of reaching 3 σ in 5 years of data taking (see Figure 3). Figure 2.
Comparison betweenANAIS-112 results on annual mod-ulation using two years of data andDAMA/LIBRA modulation bestfit. Estimated sensitivity is shownat different confidence levels ascoloured bands: green at 1 σ , yellowat 2 σ , and cyan at 3 σ . igure 3. ANAIS-112 sensitivityto DAMA/LIBRA signal in units of σ C.L. as a function of real time.Cyan bands represent the 68% C.L.DAMA/LIBRA uncertainty in themodulation amplitude. Black dotsand numbers are the experimen-tal sensitivities obtained in the twoANAIS-112 analysis, which corre-spond to 1.5 and 2 years.
3. Conclusions
ANAIS-112 data taking is progressing smoothly since August 2017, accumulating 220.69 kg × yuntil beginning of September 2019. The achieved sensitivity to test DAMA/LIBRA result is at2 σ level. Best fits ( S m = − . ± . − . ± . σ . The confirmation of our sensitivity prospectsimplied by this result, guarantees our ability to test DAMA/LIBRA at 3 σ in less than threeyears from now, bringing a new light into this long standing puzzle. Acknowledgments
This work has been financially supported by the Spanish Ministerio de Econom´ıa yCompetitividad and the European Regional Development Fund (MINECO-FEDER) under grantFPA2017-83133-P, the Consolider-Ingenio 2010 Programme under grants MultiDark CSD2009-00064 and CPAN CSD2007-00042, the LSC Consortium, and the Gobierno de Arag´on and theEuropean Social Fund (Group in Nuclear and Astroparticle Physics and I. Coarasa predoctoralgrant). We thank the support of the Spanish Red Consolider MultiDark FPA2017-90566-REDC.Authors would like to acknowledge the use of Servicio General de Apoyo a la Investigaci´on-SAI,Universidad de Zaragoza and technical support from LSC and GIFNA staff.
References [1] Liu J, Chen X, Ji X, 2017
Nature Phys. Nature Phys. Nature Phys. et al. Eur. Phys. J. C Eur. Phys. J. C Nucl. Phys. At. Energy et al. Phys. Rev. Lett.
Phys. Rev. Lett. et al.
Phys. Rev. Lett. et al.
Phys. Rev. Lett. et al.
Phys. Rev. Lett. et al.
First results from the CRESST-III low-mass dark matter program, arXiv:1904.00498[10] Adhikari G et al.
Nature
Phys. Lett. B et al. JCAP et al.
Eur. Phys. J. C et al. Eur. Phys. J. C et al. J. Phys. Conf. Ser. et al.
Phys. Rev. Lett. et al.
Eur. Phys. J. C et al. Eur. Phys. J. C et al. Eur. Phys. J. C et al. Phys. Rev. Lett.123