Results from the CUORE experiment

1. Results from the CUORE experiment Antonio Branca– DFA UniversitàdegliStudi di Padova & INFN Sezione di Padova On behalf of the CUORE Collaboration EPS-HEP 2019 – July 10-17, Ghent, Belgium

2. Outline • Bolometric technique for neutrino-less double beta decay search; • Expected signature from ; • Why TeO2? • CUORE: the first bolometric experiment at the ton scale: • The challenge of a rare event search; • Experimental setup; • From detector construction to data taking: • Assembly/installation; • Cooldown and commissioning ; • Detector optimization and performance; • Data-taking; • Physics results: • Neutrino-less DBD analysis; • Background budget; • Two neutrino DBD analysis; • Summary A. Branca

3. The expected signature • Building a high sensitivity experiment: • : select candidates with high natural isotopic abundance or enriched; • : high detection efficiency; • : high detector mass; • : good detector stability over a long period; • : extremely high energy resolution; • : extremely low background environment; Ideal energy spectrum of the two electrons in the final state A. Branca

4. Bolometric technique A (A) Copper frame:10 mK heat sink (B) PTFE holders:weak thermal coupling D Radiation: energy deposit C (D) Si joule heater:reference pulses B E (C) TeO2 crystal:energy absorber (E) NTD Ge thermistor: resistive thermometer Low temperature needed: Readout A. Branca

5. Bolometric technique: 130Te • Bolometric detectors: detector also the source of : • high efficiency; • excellent energy resolution; • large masses are achievable; • 130Te: a good candidate source for : • high natural isotopic abundance; • Q-value (2528 keV) above most of the natural radioactivity; • nuclear matrix elements and phase space on average; A. Branca

6. A rare event search Searching for a rare event (): ……………………… Extremely important to reduce as much as possible backgrounds: • natural radioactivity from outside the detector: • cosmic ray muons induced background; • neutron and gamma fluxes; • natural radioactivity from the detector itself: • long-lived nuclei (40K, 238U, 232Th); • anthropogenic radioactive isotopes (60Co, 137Cs, 134Cs); • cosmogenic radioactive isotopes (60Co); • mechanical vibration noise: • cryogenic system and seismic noise; CUORE installed @ LNGS underground laboratories (~3600 m.w.e.) REDUCTION • Strict protocols for crystal production @ SICCAS; • Cleaning techniques developed @ LNL for copper parts near crystals; • Strict protocol for assembling and installation; REDUCTION Suspension/damping systems and new noise cancellation tools REDUCTION A. Branca

7. Suspension system and external shielding Y-Beam • Abatement of vibrations: mechanical decoupling from the outside environment: • detector suspension independent from that of cryogenic and calibration systems: • detector hung by the Y-Beam through cables made of stainless steel tie bars, Kevlar ropes and copper bars (damping the horizontal oscillations); • 3 minus-K springs connect the Y-Beam to the Main Support Plate, MSP (attenuating the noise of ~35 dB); • elastomers at the structure basis (seismic isolators); • Reduction of radioactive background (from outside): • outer neutron shield: polyethylene + boric acid; • outer gamma shield: lead shield; Minus-K MSP External lead shield (~70 t) H3BO3 panels Polyethylene Elastomers A. Branca

8. Cryogenic system and shielding • Challenging task: cool down ~15 tons @ T < 4 K and ~1.5 tons @ T = 10 mK in a few weeks in a low radioactive environment. • Cryogen-free (dry) cryostat: high duty cycle: • Fast Cooling System (4He gas): T down to ~40 K; • 5 Pulse Tubes cryocoolers: T down to ~4 K; • (Custom) Dilution Refrigerator: T operations 10 mK; • Nominal cooling power: 3 μW @ 10 mK; • Reduction of radioactive background (from detector): • material screening and accurate selection to ensure radiopurity (mainly pure copper, other material in small amount, limited amount of Multi Layer Insulator); • cleaning of copper surfaces facing crystals; • lead shielding (Roman and modern Pb); A. Branca

9. Cryostat commissioning and tower assembly • Cryostat commissioning completed in March 2016: • stable base T = 6.3 mK over 70 days (no detector, full load); • full detector read-out chain (electronics, DAQ) test, temperature stability with Mini-Tower (8 crystal tower); • successful deployment of the calibration sources at base temperature; • Strict protocol adopted for each step of towers assembling/installation (developed thanks to CUORE-0): • assembling:in N2 atmosphere and within glove boxes to avoid radioactive recontamination; • installation: protected area inside clean room flushed with radon free air; protective bags flushed with N2 for overnight and emergency storage; A. Branca

10. Detector installation • Design specifics: • detector arrangement: 19 towers with 13 floors of 4 crystals each; • crystals: 988 crystals, 5 cm3, 750 g each; • total TeO2 mass of 742 kg; • total 130Te mass of 206 kg (all natural abundance); • Reduction/control of radioactive background: • minimization of material/surface facing the crystals; • closely packed crystal array with high granularity; • GOALS: • low background of ; • energy resolution: in the Region Of Interest (ROI); • projected sensitivity: (5 years, 90% C.L.); Installation of all 19 towers completed by August 2016 A. Branca

11. Cooldown and temperature selection • Temperature scans around base temperature to optimize detector resolution and NTDs resistances at design values (): • March 2017: identified best working temperature of 15 mK; • September 2017: with calibration sources deployed, confirmed trend of better resolution on physics events at lower T. Set 11 mKas new working temperature; After towers installation, the cryostat was closed between September-November 2016: • cooldown started on Dec 5, 2016: lasted ~26 days (without counting technical stops for system debugging); • on Jan 26, 2017 reached a stable base temperature of T = 7 mK; cryogenics debugging electronics noise studies A. Branca

12. Detector optimization • Developed tool to minimize vibrational noise from PTs: • use of linear drives (LD) to control PTs’ rotating valves; • PTs phase scan to find the configuration of minimum noise across the whole detector; • NTD resistances fixed by selecting the working temperature, while correct bias currents have been optimized with dedicated measurements: • for each bolometer set bias currents such to avoid distorted signal shapes and maximizing SNR; Reference pulses for different amplitudes @ 10 mK temperature: before and after LD switch on CUORE Preliminary A. Branca

13. First data taking period and detector performance • First phase of detector optimization ended in April 2017 to start science runs: • Dataset 1: 3 weeks of data bracketed by 2 calibration periods (May - June 2017). TeO2 exposure ; • still room for performance improvements; • Second phase of detector optimization restarted in July 2017: • careful investigation/upgrades to the electronics grounding in the CUORE Faraday cage; • Introduced PTs phase scans to refine the abatement of induced noise; • optimization of the operating temperature and detector working points; • software and analysis upgrades; • Resumed data-taking in August 2017: • Dataset 2: same procedure as first dataset (August – September 2017). TeO2 exposure ; Dataset 1 Dataset 2 Improved resolution between the two datasets A. Branca

14. Status of the CUORE experiment • Stop-A:warm-up to 100 K in order to fix a small leak (gate valves connecting DCS to cryostat) • Stop-B+C: • warm-up to 100 K; • solved small leaks and partial obstruction of condensation lines • fixed two PTs; Stop C Stop B Stop A • Introduced the Optimum-Trigger algorithm: allows to reduce the trigger thresholds; • Introduced an algorithm to reduce the correlated noise between bolometers; A. Branca

15. From raw pulses to energy spectrum • Amplitude evaluation: Optimum Filter technique to maximize SNR; • Thermal gain stabilization: use reference pulses at constant energy (pulsers) to correct amplitude from small variations due to temperature drifts; • Calibration: apply calibration function, exploiting 6 known peaks from 232Th in calibration data. • Calibration data are also exploited for: • model detector response to energy deposition; • energy scale and resolution evaluation; • thermal gain stabilization; A. Branca

16. Event selection • Noisy periods of data (about 1% of total exposure discarded); • Basic pulse quality selection (remove noise spikes, pile-up…); • Consider only events with multiplicity equal to 1 (anti-coincidence selection): • Radioactive contaminations likely produce signals in more then one crystal in a given time window; • Pulse shape analysis (PSA): • Define a distance in the 6-dimensional space of the pulse shape parameters; • Calculate a threshold maximizing the experimental sensitivity; • Discard events above the threshold; REMOVED ACCEPTED A. Branca

17. fit results • A 3 components simultaneous UELM (Unbinned Extended Maximum Likelihood) fit in the energy region 2465-2575 keV is performed: • Posited peak at the 0v DBD Q-value: fixed peak position, floating rate common to all channels and both datasets; • Peak for 60Co sum gamma line (2505 keV): floating peak position, floating rate common to all channels and both dataset; • Flat background for multi scatter Compton events from 208Tl line and surface events: common to all channels and dataset dependent; • The peaks are modeled through the detector response function; • The line width for each peak is scaled by the resolution scaling factor obtained from calibration data; Signal decay rate best-fit: Background index best-fit (datasets average): A. Branca

18. Combined limits with previous experiments Combined 90% limits on the half-life and the effective Majorana neutrino mass: CUORICINO: exposure ; CUORE-0: exposure ; CUORE: exposure • Half-life limits: • 130Te: 1.5 × 1025 yr from this analysis • 76Ge: 8.0 × 1025yr from PRL 120, 132503 (2018) • 136Xe: 1.1 × 1026 yr from Phys. Rev. Lett. 117, 082503 (2016) • 100Mo: 1.1 × 1024 yr from Phys. Rev. D 89, 111101 (2014) • 82Se: 2.4 × 1024 yr from Phys. Rev. Lett 120, 232502 (2018) • Nuclear Matrix elements from: • JHEP02 (2013) 025 • Nucl. Phys. A 818, 139 (2009) • Phys. Rev. C 87, 045501 (2013) • Phys. Rev. C 87, 064302 (2014) • Phys. Rev. C 91, 034304 (2015) • Phys. Rev. C 91, 024613 (2015) • Phys. Rev. C 91, 024309 (2015) • Phys. Rev. C 91, 024316 (2015) • Phys. Rev. Lett. 105, 252503 (2010) • Phys. Rev. Lett. 111, 142501 (2013) A. Branca

19. Background: CUORE vs CUORE-0 • Excess of the 210Po peak: • under investigation; • not influencing the ROI counts ( < 1%); • In comparison to CUORE-0: • strong reduction in region: material selection/screening/cleaning; • consistent in almost all the region: same cleaning and assembly procedure; A. Branca

20. Background reconstruction • Geant4 Monte-Carlo simulation: • Different background sources (60 components); • Surface/bulk contaminations of copper/lead/crystals; • Cosmic muons; • Background reconstructed with Bayesian fit on data exploiting JAGS: • Same data used for analysis; • Split in events crystal multiplicity: multiplicity 1 (M1), multiplicity 2 (M2) and multiplicity 2 sum (): • Each one is sensitive to different background contributions; • Geometrical split: outer/inner crystal layers: • Outer layer sensitive to cryostat contaminants; • Inner layer sensitive to tower contaminants; • Prior assigned to the normalization factor of each component: • From material screening, CUORE-0 background, • LNGS cosmic flux; M1 M2 inner outer A. Branca

21. Background reconstruction Model able to reconstruct main features of the CUORE background A. Branca

22. half-life Measured 130Te half life from CUORE: • Dominant systematic due to contamination location and uniformity; • Contribution from signal efficiency and energy thresholds are about one order of magnitude lower; Consistent with previous measurements: (CUORE-0) (NEMO-3) A. Branca

23. half-life In CUORE events are of the the counts in the 1-2 MeV range because of reduced background In CUORE-0 events are of the counts in the 1-2 MeV range A. Branca

24. Summary • With the first data (exposure of ): • set the strongest limit on the neutrino-less double-beta decay half-life of 130Te; • most precise measurement of the two neutrino double-beta decay half-life of 130Te; • Accumulated an exposure of more than to date: • exploring the possibility to lower the trigger thresholds; • physics results will be released shortly; • Gained a lot of experience from the operation of the first ton-scale bolometric array: • developed new methods for noise reduction; • refined methods to set the detector working conditions; • studying and testing new strategies to improve the detector resolution; • Future: • will take data for 5 years; • upgrades to perform particle identification (CUPID); A. Branca