The XENONnT Neutron Veto Detector This Talk! > 2500 SLPM GXe) - - PowerPoint PPT Presentation

Shingo Kazama (Nagoya University, KMI)

• n behalf of the XENON collaboration

@TAUP2019, September 11th 2019

The XENONnT Neutron Veto Detector

2

XENONnT Experiment

New TPC LXe Purification Radon Distillation Neutron Veto

~6t Time Projection Chamber To achieve fast cleaning of the large LXe volume (5L/min LXe, > 2500 SLPM GXe) To online remove the

the detector To tag and measure in situ neutron-induced background GXe purification (120 SLPM)

10 times higher sensitivity compared to XENON1T with 20 t-year exposure, reaching spin-independent WIMP-nucleon cross-section of O(1)☓10-48cm2

WIMP mass [GeV/c2]

WIMP-nucleon σSI [cm2]

• r

• j

• n

• d

• v

This Talk!

3

Neutron Background @ XENONnT

• Neutron events become the one of main BGs once

Rn is removed as planned (~1μBG/kg)

• PRELIMINARY estimate:

1.3±0.3 neutron events / yr in [4-50] keVr in 4 t FV without neutron veto For 20 t-year exposure, ~ 6.5 events!

• For the best dark matter discovery potential,

aim at neutron tagging efficiency > 80%

• Helps modeling neutron background
• Can check efficiency with 241AmBe (and

neutron generator)

Active neutron veto

See poster by D.Ramirez on NR BG@ XENONnT

preliminary

4

Neutron Veto (nVeto) System @ XENONnT

• XENON1T water tank (700 t) already works as a

Cherenkov detector for the muon veto

• Gd-loaded water Cherenkov detector technology

from EGADS/SK-Gd experiment in Kamioka

• Build an inner neutron veto detector with higher PMT

coverage & light collection efficiency with highly reflective foil.

• 120 8” PMTs (Hamamatsu R5912-100WA-D30)

will be installed to detect Cherenkov photons

• digitized with CAEN v1730 (500MHz)
• Highly reflective foil will be used to increase light

collection efficiency and to separate optical BGs from the outside of the water tank

• Octagonal support structure made of SUS304L

(in total ~ 700 kg) * Top part will also be covered by reflectors!

5

n

Single-scatter n-capture

157Gd (15.65%): 254 kb, 7.9 MeV γ-rays 155Gd (14.80%): 61 kb, 8.5 MeV γ-rays n+157Gd→158Gd*→158Gd+γ’s n+155Gd→156Gd*→156Gd+γ’s thermalization Cryostat Diffuse reflector

γ-ray

Delay time between single scatter and n-capture (0.2% Gd concentration) ~ 20μs Decays via gamma-cascade

nVeto Working Principle

PMTs

7.6×105 difference compared with 1H(333 mb)

Coverage < 10%

~2m

6

n

Single-scatter

Cryostat Diffuse reflector electron Cherenkov photons

nVeto Working Principle

PMTs thermalization

n-capture γ-ray

Coverage < 10% Cryostat is also covered with the reflector

~2m

nVeto Working Principle: Summary & Requirements

• High light collection efficiency
• Large PMTs with high quantum efficiency and low radioactivity
• Highly reflective foil
• High transparency of Gd-loaded water

Requirements to nVeto@XENONnT

7

PMTs

Coverage < 10%

8

nVeto System @ XENONnT

• Gd is insoluble in water, but Gd sulfate can be

dissolved when octahydrated: Gd2(S04)3・8H20 arXiv:1908.11532

• 0.2% Gd-concentration by weight
• > 90% of neutrons can be captured by Gd
• Others are captured by H (2.2 MeV)
• 3.4 t of Gd-sulfate octahydrate is necessary for

700t of the water tank

• Large PMTs with high quantum efficiency and low

radioactivity

• 120 8” PMTs: R5912-100WA-D30 (QE~33%@380nm)
• ~1Bq/PMT of 232Th, 238U, 40K
• High reflective foil
• ePTFE reflector with ~99% reflectivity (details later)
• High transparency of Gd-loaded water
• Careful checks of all the detector components used

inside the water tank (soak-test in Gd-water)

• Similar water purification system as in EGADS

9

SK-Gd / EGDAS Techniques

Gd-loaded water Cherenkov detector: developed by SK-Gd / EGADS to detect the supernova relic ν

• After full Gd-loading (0.2%), the water transparency is within typical SK values
• Water filtration system can maintain good water quality (will use similar system for nT)
• No Gd-losses are found after more than two years operation
• Gd sulfate quickly dissolves and is homogeneously distributed throughout the detector
• Key point: injecting Gd-sulfate while keeping good water transparency

Lessons learned from the EGADS experiment (arXiv:1908.11532)

• Gd sulfate is essentially transparent to Cherenkov light.

arXiv:1908.11532

200 t water tank

10

Simulation Results: γ-ray emission model

γ-ray spectrum after nCapture

• γ-ray emission model developed by ANNRI-Gd

collaboration already implemented in our MC

• Three different cases are considered in MC
• 0% (pure water), 0.02%, 0.2%(nT target)
• With 10-fold coincidence requirement and 0.2%

concentration, we can achieve > 80% tagging efficiency

• Even without Gd (pure water case), can achieve > 60%

tagging efficiency with 10-fold coincidence

preliminary preliminary

• Three different cases are considered in MC
• 0% (pure water), 0.02%, 0.2%(nT target)
• Even without Gd (pure water case), can achieve

> 60% tagging efficiency with 10-fold coincidence requirement

1000 2000 3000 4000 5000 6000 7000 8000 Energy [keV] 1 10

10

10

10

10 Counts [1/10 keV]

Data GLG4sim Our model

PTEP 023D01

γ-ray emission model

• γ-ray emission model developed by ANNRI-Gd

collaboration already implemented in our MC preliminary preliminary

11

preliminary

Coincidence window 0.2% 0.02% 0%

Fraction of the events vetoed

100μs 0.98 0.72 0.48 150μs 0.99 0.85 0.62 300μs ~1 0.98 0.84 500μs ~1 ~1 0.94

Simulation Results: Timing Information

• For 0.2% case, it may be possible to decrease the

coincidence window down to 150 μs

• For the pure water case, need to increase the

coincidence window up to 500 μs because it takes longer for neutrons to be thermalized and then captured by hydrogen.

Δt (TPC, nVeto)

12

Simulation Results: Reflectivity Dependence

• Photo-coverage of the PMTs is < 10%, thus reflectivity

plays an important role for achieving higher tagging efficiency because of many reflections.

• ePTFE is selected because of its high reflectivity: ~99%
• Also it is radio-pure: U(~20ppt), Th(~30ppt) with ICP-MS
• No degradation of reflectivity has been found after soak-

test in Gd-water (0.2% concentration)

preliminary

Many Reflections!

Before soak-test After 2-month soak-test

ePTFE reflector (1.5mm thickness)

13

See poster by D.Ramirez on NR BG@ XENONnT NR BG rate without nVeto with nVeto

• Target: 0.2% Gd concentration, and
• 10-fold coincidence,150μs window (TPC, nVeto),
• 4t FV, e-lifetime of 1ms, drift field of 200 V/cm
• With nVeto, neutron BG can be reduced from 1.3

to 0.17±0.05 events/year

• For 20 t-year target exposure, the # of neutron BG

events is estimated to be ~ 1event

Simulation Results: NR BG Rate with/without nVeto

preliminary preliminary preliminary

14

• We may discard many TPC events if nVeto has too

much BG rate

• Main contribution is from radioactive impurities in

PMTs

• measured with HP-Ge detector@LNGS
• For 10-fold coincidence case, BG rate is expected to

be ~100Hz.

• With 150μs coincidence window, the loss of TPC

events is ~ 1.5 %.

Simulation Results: Fake BG Rate in nVeto

PMT Component 40K [Bq/PMT] 238U [Bq/PMT] 232Th [Bq/PMT] Window 0.6 < 0.6 0.4 Body 1.1 0.9 1.3

nVeto PMTs radioactivity

* 120 PMTs will be used

15

Calibration

• ▫

▫ γ an ▫ γ n)

• ▫

▫

nVETO: neutron tagging efficiency TPC: nuclear recoil response calibration

• Use AmBe for the calibration of the nVeto
• Average neutron energy 4.6 MeV
• In 57.5 %, a coincident γ-ray with an energy of 4.438MeV
• TPC provides trigger or delay (γ + n)
• Calibration of the TPC
• nVeto provides trigger
• Measure multiple-scattering / single-scattering neutrons

16

Summary & Outlook

• nVeto is essential for the dark matter discovery
• XENONnT nVeto will be based on Gd-loaded Water Cherenkov

technology established by SK-Gd / EGADS

• Neutron tagging efficiency >= 85 % is possible
• requires high photo-coverage, highly reflective foils and

high transparent Gd-load water

• XENONnT nVeto design is almost finalized
• We have tested 125 8” R5912 PMTs at LNGS
• None of them showed a bad behavior.
• Installation of the nVeto will be the last step of XENONnT

construction, and it will start by the end of this year.

• Then, we will start XENONnT commissioning with pure water

in the nVeto.

• In parallel, we are working to get the permission for Gd-

sulphate at LNGS. The goal is to install the Gd-Water purification plant at the beginning of 2020, then we will start the dark matter run with the fully efficient nVeto.

Back Up

18

Water Purification System

Figure 7: Schematic view of the band-pass system and the fast recirculation system (inside dashed line). These systems were built in cooperation with the South Coast Water company.

arXiv:1908.11532

20

Geometry