This research has been funded by the European Unions Horizon 2020 - PowerPoint PPT Presentation

This research has been funded by the European Union’s Horizon 2020 Research and Innovation Programme under the Marie Sklodowska-Curie Grant Agreement No 841261 (DarkSphere) 1

NEWS-G NEWS-G collaboration Searching for Canada, France, Greece, UK, USA low mass Dark Matter Using an innovative gaseous detector the Spherical Proportional Counter 6th collaboration meeting LPSC, Grenoble June 2019

The sensor Anode ● Metallic ● Semiconducting Wire ● Metallic core ● Insulating surface Supporting tip ● Insulator ● ● ○ Supporting Rod ● Metallic ○ ● Simple design ● Resistive ● Single readout ● coating ● 3 I.Giomataris et al ,JINST,2008, P09007

Polyethylene 30 cm Lead 15cm Copper 8 cm ● NOSV Copper vessel (Ø 60 cm) ● Equipped with a 6.3 mm Ø sensor Sedine 60 cm Ø SPC ● Chemically cleaned several times for Radon deposit removal

NEWS-G collaboration, Astropart. Phys. 97, 54 (2018), doi: 10.1016/j.astropartphys.2017.10.009 NEWS-G (2017) DAMIC Crest-II (2015) CDMS-lite Gas Mixture: Ne+0.7%CH 4 at 3.1 bar (280 g) Exposure: 9.6 kg*days (34.1 live-days x 0.28 kg)

Snoglobe at LSM ● NEWS-G is preparing to install a new detector at SNOLAB ● H-rich mixtures ● Expected to be sensitive to WIMP masses ~100 MeV ● Detector already operating at LSM for a commissioning run Preliminary Ne+ 6%CH 4 NEWS-G Snolab CH 4

Snoglobe at LSM ● NEWS-G is preparing to install a new detector at SNOLAB ● H-rich mixtures ● Expected to be sensitive to WIMP masses ~100 MeV ● Detector already operating at LSM for a commissioning run Preliminary

Electric field strength in large volume SPCs

10 Low E-field region

11

Giganon, A. et al, 2017. “A Multiball Read-out for the Spherical Proportional Counter.”, JINST ● Decoupling the E-field needed for gain vs the E-field to drift charge

Giganon, A. et al, 2017. “A Multiball Read-out for the Spherical Proportional Counter.”, JINST A C H I N O S S i n g l e

Giganon, A. et al, 2017. “A Multiball Read-out for the Results with the prototypes Spherical Proportional Counter.”, JINST He:Ar:CH 4 (80:11:9) 640 mbar HV 1 = 2015 V HV 2 = -200 V 2 mm Ø anodes

Resistive layer materials tested: ● Araldite/Cu, Araldite/Graphite ● Commercial resistive paste ● DLC (Diamond Like Carbon) 3D design Measurement of the 5.9 keV 55 Fe X-ray line 5.9 keV Escape peak Implemented modules using 3D printing ● He:Ar:CH4 (56:37:7) ● 455 mbar ● HV1 = 1100 V, HV2 = -100 V 2 mm Ø anodes ●

Purifiers: ● Getter ● Oxysorb 600 mbar He+10% CH 4 without contaminant filtering 600 mbar He+10% CH 4 with contaminant filtering 5.9 keV 5.9 keV 55 Fe 55 Fe Using 1.49 keV Oxisorb Al Fluorescence 9310 ADU 4620 ADU 1.49 keV 8.9% (σ) 22.4% (σ) Al Fluorescence Contaminants: Oxygen, Water, electronegative gases...

● SAES MicroTorr Purifier (MC700 902-F) then used ● Improved filtering efficiency in large sphere – attachment problem ‘solved’ ● Incorporated into recirculation system with RGA Charge Loss and Oxygen Concentration over Time while gas passes circulated through MicroTorr Purifier

Q. Arnaud et al. (NEWS-G Collaboration), Phys. Rev. D 99, 102003 (2019) Common DAQ for timing Parallel photo-detector analysis between two to tag laser events channels Tunable transmission to control the mean number of electrons A powerful UV laser ● 213 nm laser used to extract primary electrons from wall of SPC capable of extracting 100s ● Photo detector in parallel tags events and monitors laser power of electrons ● Laser intensity can be tuned to extract 1 to 100 photo electrons

Q. Arnaud et al. (NEWS-G Collaboration), Phys. Rev. D 99, 102003 (2019) ● N photo-electrons are extracted from the surface of the sphere: Poisson Fit results ● Each photo-electron creates θ = 0.09 ±0.02 S avalanche pairs: <G> = 30.26 ± 0.21 ADU Nth convolution of Polya χ 2 /ndf = 0.97 ● Sum the contributions of all N photo-electrons ● The overall response is convolved with a Gaussian to model baseline noise Laser in pulsed mode fixed to a low intensity

Q. Arnaud et al. (NEWS-G Collaboration), Phys. Rev. D 99, 102003 (2019) Fit of 270 eV and 2.82 lines with ● Ar37 produced by flat background irradiating Ca power with a K-Shell: 2.82 keV high flux of fast neutrons W = 27.6 eV/pair* F = 0.19 ● Together with laser calibrations, can find W (mean Ionization energy) L-Shell: 270 eV with 1% precision for target W = 27.6 eV/pair gas, and set upper limits on F = 0.26 F (Fano factor) Detector response modeled: ● Primary ionisation (COM-Poisson) *The W-value at 2.82 keV was D. Durnford et al, Phys. Rev. D 98, 103013 (2018) calculated directly from <G> and ● Avalanche (Polya) fixed for this fit

Laser events The laser can be used to monitor the detector response during physics runs 37 Ar 2.82 keV peak Long-term fluctuations in gain can be caused by temperature changes, O 2 contamination, sensor damage... 37 Ar 2.82 keV corrected Laser monitoring data could even be used to correct for long-term fluctuations Q. Arnaud et al. (NEWS-G Collaboration), Phys. Rev. D 99, 102003 (2019)

● 4N Aurubis copper (99.99% pure) ○ Spun into two hemispheres ● Copper has no long-lived isotopes 63 Cu(n, ⍺ ) 60 Co from fast neutrons – mostly cosmic ● muon spallation ● Contaminants : U and Th decay chain traces ○ Measured for NEWS-G ~10 μBq/kg 210 Pb out of equilibrium - 28.5 mBq/kg ○ 5.5 MeV 6.0 MeV 5.3 MeV 7.7 MeV

● Using PNNL expertise in electroforming Cu The setup during electroplating at LSM ● The inner surface of the detector was electroplated to stop Bremsstrahlung X-rays from 210 Pb and 210 Bi β-decays in copper ● 0.5 mm pure copper plated on inner surface at LSM: expected background from 210 Pb and 210 Bi under 1 keV reduced from 4.58 dru<1 keV to 1.96 dru

Plated surface Hemisphere after plating ~0.036 mm/day ~1.3 cm/year ● Good surface quality achieved ● Hemispheres electron-beam welded together ● Detector already operating at LSM ● Copper was deposited at a rate of ~36 μm/day ○ Result is promising for possibly a whole detector electroformed underground

Preliminary

● Known that these filters emanate radon: ○ Observed ~600 mBq alpha particles, up from ~1 mBq without filter ~40 days before 222 Rn acceptably low, but ○ depositing daughters (e.g. 210 Pb) ● Alpha particles not directly a great problem Daughter nuclei are a problem ● Tests with an Carboxen 100 active carbon filter to capture Rn ○ Side effect, removal of some CH 4 from mixtures ○ Looking into optimum temperature of filter ● Recirculation system in development

Comparison between a CH4 and Ne/CH4 run

Gas quality (2) Improvements Gas filtering Vacuum conditions 1.E-4 mbar→1.E-5 mbar→1.E-6 mbar Rise time vs Amplitude 2D Amplitude 1D Histo Histo Leak Rate ≈ 1.4E-6 mbar*L/s Wall event e He/CH4 (90/10) ⇒ Not a dramatic effect - Attachment 600 mbar e Volume - HV1=1820 V event HV2=+225 V Gas quality Ball Φ2 mm No OXISORB Contaminants ~ppm ↓ Oxisorb ~100 ppb 1.49 ⇒ Big improvement 5.9 keV keV ( 55 Fe) ( 27 Al) He/CH4 (90/10) 600 mbar Increased drift velocity less HV1=1840 V attachment Conversion e - HV2=+300 V Ball Φ2 mm OXISORB

Best Estimate of 210 Po & 210 Pb from two measurements of 210 Po 2 measurements of 210 Po in our copper. Implied: 210 Po ~ 80 mBq/kg ● 210 Pb ~ 60 mBq/kg ● 10 7 decays of 210 Pb and 210 Bi in Cu Walls in 2 bar Ne+10% CH 4 Simulation shows this gives 4.58 dru < 1 keV Reduced to 1.96 dru if 500 μm pure copper plated onto surface XIA UltraLo-1800 See: XMASS collaboration arXiv:1707.06413 https://www.xia.com/ultralo-theory.html

Advantage of spherical geometry - A C~Electronic noise ↑ Electronic noise ↑ Threshold Capacities for a 1 m 3 detector in different geometries Parallel Plate Spherical ≈ 3500 pF ≈ 1.5 pF r A Cylindrical ≈ 115 pF 36

● Copper benefits from ‘ electrowinning ’ – has high reduction potential ● Reduces on surface more readily than most contaminants - 238U, 232Th ect ● Refined ultra-pure copper surface builds up

Preparation Procedure ● Cleaned with detergent ● Sanded ● Cleaned again ● Surface chemically etched with 3% H 2 0 2 , 1% H 2 SO 4 in deionised (DI) water ○ Shown to be effective etchant while less aggressive than some alternatives ● Electrolyte of H 2 SO 4 , H 2 O and CuSO 4 ○ Pump and filter to move electrolyte and remove particulates 38

Electroplating Copper ● Some ions reduce more readily than others – reduction potentials ● Copper benefits from ‘ electrowinning ’ – high reduction potential +0.34 V ● Reduction potential of: ○ Uranium: -1.80 V ○ Thorium: -1.90 V ○ Lead: -0.44 V ○ All lower than copper → refined during electroplating ● Using PNNL expertise, already electroformed Cu for Majorana 39 Experiment