APC: P. Agnes, D. Franco, Q. Riffard, A. Tonazzo LPNHE

38 Pages·2016·44.65 MB·English

DarkSide
APC: P. Agnes, D. Franco, Q. Riffard, A. Tonazzo
LPNHE: O. Dadoun, S. De Cecco, C. Giganti, A. Navrer-Agasson
Journée Matiére Sombre - APC 01/01/2016
The DarkSide project
The Veto’s
Dual phase Liquid Argon TPC to search for Dark Matter
DarkSide-50 (2013-20A1c8tiv)e neutron veto: !
-  20 ton boron-loaded scintillator!
-  50% PC + 50% TMB!
Nested detectors: 50 kg Liquid Argon TPC
-  2 m radius sphere!
Boron-loaded Liquid S- c11i0n Ltoilwla Btaockrg rVouendt PoM T s !
Water Cherenkov Veto
Active muon veto (passive neutron veto): !
First use of undergrou-  1n0d00 atorn gulotran p u(rde weaptelre! ted of 39Ar)
-  10 m height, 11 m diameter!
-  80 upwards oriented PMTs!
Demonstrate pulse shape discrimination (>107)
DarkSide-20k (2020-2026)
Rejection efficiencies:!
-  >99.5% against radiogenic neutron!
30 ton (20 ton FV) dep-l  e>9t5e%d c oLsmiqogueniidc nAeurtrgonos!n TPC
Designed to be background-free at 100 t yr exposure
D. Franco - APC 19
Similar system of vetoes as DS-50, use of SiPM
ARGO (2025-2035)
300 ton (200 ton FV)
Search for WIMPs up to the neutrino floor
Solar neutrino physics
2
DarkSide Collaboration
Large international collaboration (43 institutes, 277
signatures for the DS-20k proposal)
Entire Liquid Argon community for DM searches to join this
project → no competition with Liquid Argon for DS-20k
Part of the collaboration works on external calibration
experiments, using neutron beams to study the response of
LAr to Nuclear Recoils
SCENE in the US (2014)
ARIS experiment at IPNO (2016)
Future experiments including directionality (ReD, ARIS-2)
3
IN2P3 DarkSide groups
DS-50 analyses
DS 1ton prototype
3 papers in preparation
(G4DS Monte Carlo, NR calibration
f90 (PSD) model, experiments and
DS-50 background budget) directionality
ARIS (done)
APC+LPNHE
ARIS-2
Developers of the
ReD
DS-20k optimization Monte Carlo
simulation
(do we need vetoes to
be background-free? yes!
DS-20k sensitivity ARIA
and discovery potential, (depletion factor needed
solar neutrinos, …) for DS-20K and ARGO)
French groups in DS have leading roles in physics analyses
Leaders of the DS MC simulation
External calibration with neutrons beam
4
Timescale
16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
DS-50
DS-20k
ARGO
DarkSide-50 (37 kg FM) will take data at LNGS until 2018
DarkSide-20k (20 ton FM) will start data taking at LNGS in
2021
Expect approval from INFN and NSF in April 2017
A 1-ton scale prototype will be built starting next year to
test the technologies needed for DS-20k
ARGO will follow DS-20k (200 ton FM)
5
What is DarkSide-50?
a two phase liquid argon (LAr) detector,
within a neutron veto,
within a muon veto,
under a mountain
… and all because of backgrounds
March 17, 2015 DarkSide 50 p. 10
© Stephen Pordes, FNA6L
The Veto’s
DarkSide-50
Experiment installed at LNGS
Active neutron veto: !
-  20 ton boron-loaded scintillator!
Double phase TP-  C50% w PCit +h 5 05%0 T MkBg! of LAr
-  2 m radius sphere!
-  110 Low Background PMTs!
Vetoes system: Liquid Scintillator and
Water Cherenkov
Active muon veto (passive neutron veto): !
-  1000 ton ultra pure water!
2 data releases published
-  10 m height, 11 m diameter!
-  80 upwards oriented PMTs!
Atmospheric Ar campaign (52 days)
Phys.Lett.B.743(456)
Rejection efficiencies:!
-  >99.5% against radiogenic neutron!
Underground Ar campaign (71 days)
-  >95% cosmogenic neutrons!
Phys. Rev. D 93, 081101 (2016)
Background identification
Background reduction
Pulse Shape Discrimination
Depleted UndergroD. Furannco d- AP CA rgon 19
S1/S2 discrimination
Low background materials
Measure neutron flux in borate
Active Shields
scintillator
Demonstrate the potential of the
technology for multi ton
background-free detector
7
DarkSide-50 detector design
Liquid Argon TPC
36 cm x 18 cm radius
50 kg (36.9 fiducial)
38 PMTs
DS50 LAr TPC
Uniform electric field (200 V/cm)
gaseous Ar
1 cm gas pocket, extraction field
(above grid)
anode
(2.8 kV/cm)
Reflectors and TPB coating
extraction grid
Liquid Scintillator Veto (LSV) liquid Ar
~
(~50 kg active volume)

4
0

c
m
30 tons, 2m radius
LS (1:1 TMB+PC)
cathode
110 PMTs
TPC assembled in radon-free clean room and
Water Cherenkov (WCD)
just before insertion into the cryostat
March 17, 2015 DarkSide 50 p. 19
1 kton water, 5.5m radius
80 PMTs
8
Ionization and excitation
WIMPs scatter on LAr → primary photons (S1) + electrons
drifting to the GAr region where they are accelerated and
emit light through electroluminescence (S2)
S1/S2 ratio allow to distinguish electron recoils from
Dual Phase Liquid Argon TPC
nuclear recoils with a rejection power of ~102-103
Recoil
Ionization Excitation
Electrons Ar+ Ar*
WIMP Ar + Ar *
WIMP 2 2
S2
Ar**
Singlet Triplet
Recombination
S1
Ar*
Singlet
Ar *
2
Triplet 99
Davide Franco - APC
S2/S1 Ratio
4
date. Within each exposure segment, the field magnitude electron activity following large-S2 events [9, 42], and
is considered to be constant and uniform. Boundaries in results in 99.0% e�ciency when applied to 3H calibra-
date are September 11, 2014; January 1, 2015; April 1, tion data for WS2014–16. The S2 threshold is set to
2015; October 1, 2015; May 2, 2016. Boundaries in drift 200phd (raw uncorrectSed2pulse area) to avoid events Energy deposition! Heat!
time are 40, 105, 170, 235, 300 µs. Periodic 3H calibra- for which the x ,y position uncertainty is high.
S2 S2
{ }
tions provide each of the 16 exposure segments with a Events for which S2 > 104 phd, S1 > 50phd, log (S2) <
S1/S2 10
unique calibration set from which to construct a unique median 5� or log (S2) >median + 3�
NRS�1 NR 10 ER ER
Excitation! Ionization!
individual response model. These 16 respSons2e m/odSels 1 (boRundaarites eividoent in Fig. 1) are considered far from
take the form of parameter variations of the Noble El- the region of intere(setxacnidtaatrieoing n<or<e d .ionisation)
ement Simulation Technique (NEST) model[33], which A fiducial volume in drift time is defined as 40-300 µs
Phys.Rev.D93,072009
captures both the LXe microphysics of signal production (date-independent). Each of the four date bins has Ar*!
and the detector physics of signal collection. Fits are a uniquely defined radial fiducial selection boundary,
Ar+! Electrons
performed by comparing the measured ER band (median 3.0cm radially inward from the measured PTFE sur-
!
S2
Ar*
and 10-90 percentile width in the S1,log (S2) plane face position for that date bin in observed S2 coordi-
{ 10 } 2!
as in Fig. 1) with tSha2t predicted by the response model. nEatnese, rxgy,y d,ez p.osThiteiowanll! position, a functHioneoaf t!
ER S2 S2 S2 NR
{ }
Specific to each exposure segment, two model parame- � ,z , is measured with 210Pb sub-chain events that Ar+
S2 S2
{ } 2! S2
ters are varied during these fits: the electric field mag- originate on the PTFE surface. The fiducial mass is de- !
Singlet Triplet
nitude, and the recombination fluctuation parameter F termined by scaling the 250kg of active LXe by the ac- ! !
r
(see[3S11, 33, 34, 40]). FieldS-i2nd>ep>enSd1ent parameters that ceEptxancceitfSara1tcitoionno! f 83mKr eventsIothnrSoiu2zg~haStthi1eofindu! cial- Ar**
describe the detector as a whole (e.g. light collection ef- selection criteria. The time-averaged fiducial masses for !
ER band
ficiency in gas(,exS2cirtaestoiolunti o<n<, g io, nanisdagtio),n)are allowed the date bins are 105.4, 107.2, 99.2, and 98.4 kg, in
1 2 (excitation ~ ionisation) S1
to vary while constrained to be equal for all exposure chronological order. A 3% systematic uncertainty across NR! band Recombination
Phys.Rev.D93,072009 !
segments within a given date bin. In each exposure seg- all dateAs irs*e! stimated through comparison with accep-
ment, the measured ER band median di↵ers from the tance fractions of 3H calibration data, of similarly uni-
model band median by less than 1% for all S1. The formEdRist rRibuetiojneicn ttriuoe nrec ofial pcoAstitorio+rn! .: 1E0l2e -c 1tr0o3ns
Exploit different fractions of
16 electric field magnitudes found through these fits are !
S2
BAenr*etti et al. (ICARUS) 1993;
consistent with the values earlier obtained from the elec- 4
energy going to excitation (S1) 2! 9.8 keVee
trostatic field models. This last point deserves emphasis, Benetti7 .5et8.7 al. (WARP) 2006 Discrimination between
3.8 6.3
baecnaudse itohentiwzoatetcihonniqu (esSf2or)e stimating electric field 5.2 Ar+ NR (WIMP-induced) and ER (background)
4
magnitude are completely independent: the electrostatic 3.6 2! S2
2.9 !
fiAeldllmoowdel tisob adseidsotninthge uobisserhve dEeRlec tfrronodmrif tNpaRth sSinPagolloe Atg EneRsTriplet PhD Thesis Defence, 30 Sept ‘16 7
3.4 Title Text
alone, while the NEST fits are based on the S1 and S2 ] 1.7 ! !
)
d
amwpiliStthu1d eas adloinse.crimination power of ph 3.2 NARr**
(

2 !
1N0eu2t-r1on0c3a librations with the DD source were per- S
[ 3 ER band
0
formed in each date bin. For each individual exposure 1
g
(excitation ~ ionisation) o S1 33 keVnr
segment, the best-fit parameters from the corresponding l 2.8 Recombination
Technique available for dual NR! band 27
!
ER calibration are applied to the NEST NR model. The
2.6
21
repsuhltaingsNeRsm noodeblslseho wliqexuceilldenst adgreeetmeencttwoitrh cali- LUX Col., Phys. Rev. Lett. 116,
brations, such that the NR band medians of correspond- 2.4 15
161301 (2016)
(Xe or Ar)
EinRg m Rodeelsjeancdtciaolibnra tifoanscdti↵oerrb:y 1le0ss2t h-a n120.6%3 for 3 9
2.2
all S1. As in[9], the overall energy scale in the response 0 10 20 30 40 50
Benetti et al. (ICARUS) 1993; S1 (phd)
models is fixed by fitting the NEST NR model to a sepa-
Benetti et al. (WARP) 2006
rate in situ energy calibration using tagged neutron mul- Discrimination between 10
FIG. 1. WS2014–16 data passing all selection criteria. Fidu-
tiple scatters[11, 12]. As before, we conservatively as-
NR (WIMP-induced) and ER (background)
cial events within 1 cm of the radial fiducial volume boundary
sume NR light yield to be zero below 1.1keV, the lowest
are indicated as unfilled circles to convey their low WIMP-
energy at which NR light yield was measured in[11]. The
signal probability relative to background models (in particu-
Paolo1 A6gEnResand 16 NR models are then used withPinhDa pThroefislies DefleaTitle Texrnt ctehe, 32006 PSebpwt a‘1ll6background). Exposure-weighted average 7
likelihood ratio (PLR) method[41] to search for evidence
ER and NR bands are indicated in blue and red, respectively
of dark-matter scattering events. It can be seen from the (mean, 10%, and 90% contours indicated). Of the 16 models
light-dashed curves in Fig. 1, representing extrema of the used, the scale of model variation is indicated by showing the
16 ER and NR models, that the scale of model variation extrema boundaries (the upper edge of the highest-S2 model
and the lower edge of the lowest-S2 model) as fainter dashed
is small and diminishes towards the energy threshold.
lines for both ER and NR. Gray curves indicate a data selec-
Events consisting of a single scatter within the active
tion boundary applied before application of the profile likeli-
LXe are selected according to several criteria: a single S2
hood ratio method. Green curves indicate mean (exposure-
preceded by a single S1, an S1 threshold of 2 PMT coinci-
weighted) energy contours in the ER interpretation (top la-
dence, and an upper threshold for the summed pulse area bels) and NR interpretation (lower labels), with extrema mod-
outside S1 and S2 within the trigger window. This last els dashed.
selection removes triggers during high single-extracted-