New Ideas in Conventional Detectors - T. Shutt SLAC - Rencontres du Vietnam
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
New Ideas in Conventional
Detectors
T. Shutt
SLAC
16th Rencontres du Vietnam
Quy Nhon, Vietnam
Jan 7, 2020
Conventional, in this talk
• Dark matter acts like a particle
• Detectors
- Liquid noble TPCs
- Next generation cryogenic. (Current: J. Gascon talk)
• Not in this talk
- DM acts like a wave - axions etc., talks by H. Liu, K.J. Bae
on Thursday
- CCDs (DAMIC) + other semiconductors, bubble
chambers
T. Shutt - 16th Rencontres du Vietnam, 1/7/2020 2
The LZ Dark Matter Experiment
• LXe TPC - 7
tons
• Nested veto:
LXe skin + Gd-
loaded
scintillator
T. Shutt - 16th Rencontres du Vietnam, 1/7/2020 3
The LZ Dark Matter Experiment
LXe filling this year, first physics in 2021
T. Shutt - 16th Rencontres du Vietnam, 1/7/2020 4
DAMA
Current focus:/ neutrino
LIBRA: other
floor
PDG 2018
No annual modulation signal in other experiments: 5
T. Shutt - 16th Rencontres du Vietnam, 1/7/2020
DAMA
New / LIBRA:
Idea: Also Look Left other
PDG 2018
Symmetric DM?
SIMPS?
No annual modulation signal in other experiments: 6
T. Shutt - 16th Rencontres du Vietnam, 1/7/2020
How to look left
< 1 kg 100
pp solar
104
10 tons
nb
atm, ds
• Noenergy
Low annual modulation
threshold, light signal
target in other experim
XENON100
[ 0.19 keV 16 GeV ]
2
Enr−max MN
mχ =et500
E. Aprile MeV ×Rev. Lett. 118,
al, Phys. × 101101 (2017)
XMASS
• Small detector
K. Abe et al, Phys. Rev. D 97, 102006 (2018)
T. Shutt - 16th Rencontres du Vietnam, 1/7/2020 7
Low energy “Nuclear” Recoils
• Lindhard - adiabatic overlap of electron shells
4 33. Passage of particles through matter
• Most energy goes to heat
(N = N ρZ/A). The former is used throughout this chapter, sin
e A
• Bigger effect at low energy
(dE/dx, X , etc.) vary smoothly with composition when there is
0
(PDB, 2016)
Mass stopping power [MeV cm2/g]
µ+ on Cu
100 µ−
Bethe Rad
Anderson-
Ziegler
Lindhard-
Radiative
Scharff
effects Eµc
10 reach 1%
Minimum
ionization
Nuclear
losses
1
0.001 0.01 0.1 1 10 100 1000
βγ
Electron velocity in
0.1 target
1 atoms10 100 1 10 100
T. Shutt - 16th Rencontres du Vietnam, 1/7/2020 8
Fig. 1. The radiation from the Migdal e↵ect is typica
New Idea: different channel
3-4 orders of magnitude more likely to occur than BRE
Although only a very small fraction (0.001⇠0.1%) of N
accompanies Migdal radiations, the larger energy and E
nature make them easier to be detected than the p
NRs.
• Migdal: initial nuclear kick excites electron shells. X-ray
χ
(Ibe, et al., arXiv:1707.07258)
~~~
Bremsstrahlung
~
~~~
-
~
e
Small probability, depends on shells
~~~~
~~
e
~~~
Ionization electron
~
e
- Boosted energy in ER channel χ e
- Also “Bremsstrahlung” (Kouvaris, Pradler, arXiv:
+
+
1607.01789) e e
Auger electron
(LUX, arXiv:1811.11241) arXiv:1907.12771
NR
FIG. 1. Illustration of the ER signal production from BRE
(green) and Migdal processes (pink) after elastic scatter
between DM ( ) and a xenon nucleus.
Migdal
The data used in previous analyses [8] consists of t
science runs with a livetime of 32.1 days (SR0) and 24
days (SR1), respectively. The two runs were taken un
Brehm slightly di↵erent detector conditions. To maximize
1.2acquired
amount of data keVer under stable detector conditio
threshold
we decided to use SR1 only. The same event selecti
fiducial mass, and background models as described in
are used for the SR1 data, which we refer to as the
S2 data in later text. The exposure of the S1-S2 data
about 320 tonne-days. The interpretation of such S1
analysis is based on the corrected S1 (cS1) signal a
T. Shutt - 16th Rencontres du Vietnam, 1/7/2020 9
Fig. 1. The radiation from the Migdal e↵ect is typica
New Idea: different channel
3-4 orders of magnitude more likely to occur than BRE
Although only a very small fraction (0.001⇠0.1%) of N
accompanies Migdal radiations, the larger energy and E
nature make them easier to be detected than the p
NRs.
• Migdal: initial nuclear kick excites electron shells. X-ray
χ
(Ibe, et al., arXiv:1707.07258)
~~~
Bremsstrahlung
~
~~~
-
~
e
Small probability, depends on shells
~~~~
~~
e
~~~
Ionization electron
~
e
- Boosted energy in ER channel χ e
- Also “Bremsstrahlung” (Kouvaris, Pradler, arXiv:
+
+
1607.01789) e e
Auger electron
arXiv:1907.12771
xpected DM sig-
FIG. 1. Illustration of the ER signal production from BRE
ects using NEST (green) and Migdal processes (pink) after elastic scatter
our indicates a between DM ( ) and a xenon nucleus.
V/c2 assuming a
other two con-
ed teal contour Big
The data used in previous analyses [8] consists of t
science runs with a livetime of 32.1 days (SR0) and 24
y scalar media-
ue contour is for penalty
days (SR1), respectively. The two runs were taken un
slightly di↵erent detector conditions. To maximize
mediator (11.5
tes the expected in rate
amount of data acquired under stable detector conditio
we decided to use SR1 only. The same event selecti
or a given signal fiducial mass, and background models as described in
upper limit. The are used for the SR1 data, which we refer to as the
S2 data in later text. The exposure of the S1-S2 data
ed in the region
about 320 tonne-days. The interpretation of such S1
95 live days and analysis is based on the corrected S1 (cS1) signal a
T. Shutt
dius < 18- 16th Rencontres du Vietnam, 1/7/2020
cm are 10Newish Idea: S2 only in LXe/Ar TPCs
• # electrons, photons comparable
• Light collection ~10%,
e- collection ~100%
• Substantially reduce E threshold
• Time Projection Chamber
(XENON1T - 1907.11485)
12
100 12
Yield NESTCharge
NEST ChargeYield
Yield NEST Charge Yield
eld NESTLight
NEST LightYield
Yield NEST Light Yield
harge Yield 10
80
LUXCH3T
LUX DD Charge
ChargeYield
Yield 10 LUX DD Charge Yield
ght Yield LUXCH3T
LUX DD Light
LightYield
Yield LUX DD Light Yield
Yield (quanta/keV)
harge Yield LLNL
LUX Charge
Xe127 YieldYield
Charge LLNL Charge Yield
harge Yield 8 PIXeY Ar37 Charge Yield 8
60
6 6
40
4 4
20
2 2
(LUX - 1907.06272)
00 0
4 5 00 11 22 33 44 55 0 1 2 3 4 5
keV) Nuclear
Electronrecoil
recoilenergy
energy(keV)
(keV) Nuclear recoil energy (keV)
T. Shutt - 16th Rencontres du Vietnam, 1/7/2020 11S2 only electron background
(Sorensen, Kamdin, arXiv:1711.07025)
“e-train” in LUX
LBECA prototype - sealed
• Several sources of electrons chamber with high
chemical purity
• LBECA: dedicated S2 only experiment
- Electron reduction methods under study
- 100 kg LXe detector proposed
LZ grid under test
• LZ: major program to reduce emission from
grids
- electron signal ~4 times larger than LUX
T. Shutt - 16th Rencontres du Vietnam, 1/7/2020 12Very new idea: doping
• Dissolve H2 (or He) in LXe TPC
• Proton is best low mass target H
- 3 e- Enr~100 eV: mx~100 MeV
•Lindhard nullified by mH, mXe Xe
imbalance
What
• H2 would
and D2: oddsignals
n and p look like?
- or He, if ok with PMTs
• H2 dissolved in LXe
• HydroX - new experiment,
• DM-H2 interactions:
• DM scatters on H2 (proton)
currently LZ-based
• Proton transfers energy to Xe (mHSub-cooled liquid Xe
Outer Cryo
Xe
Variable control valve
HydroX
Line heater
Cold head w/ heater x6
• LZ fiducial becomes shield To/From Main Circula0on Loop
To/From 222Rn removal
Outer Cryostat Vessel
Xenon Tower
• Modify purification to cope
Dis0lla0on Column
with H2 x6
• Need to check
- H2 solubility
Figure 11: The top panel shows a schematic of the existing LZ xenon handling system
xenon gas returns from the circulation loop and is liquified in the “Xenon Tower” before
through transfer lines into the bottom of the detector. Liquid overflows a weir near the top
detector and returns to the Xenon Tower via another transfer line, before evaporating in
- charge yield heat exchanger in the Tower and heading out to the circulation loop as a gas. In the bottom
Preliminary Projections
we have added an intermediate step between the detector and the Tower to remove H2 fr
liquid via distillation. The distillation column could potentially replace the right most vesse
-
Xenon Tower (the reservoir). An alternative scheme (not shown) would use sparging to rem
effect of H2
tivity
H2 by bubbling warm, pure xenon through the liquid. The extracted H2 gas is then re-inject
the liquid xenon returning to the detector.
on S2 phase. Support is requested for approximately 0.4(0.2) total FTE of Penn State engineers
O’Meara and Matthew Weiss in year 1(2-3) and 0.5 FTE of graduate student in Years 1 a
build and operate this system.
Fermilab cryoengineer Kendziora will assist in the design of these systems with 0.25
years 1 and 2. Kendziora has relevant expertise from the design and operation of the F
s: distillation column that separated underground argon from impurities as well as the design
cryogenics for the DarkSide-50 and DarkSide-20k projects in Italy.
Z environment 22
raction of 2.6%, 2.2 kg
ER-like (no discrimination)
-day exposure
T. Shutt - 16th Rencontres du Vietnam, 1/7/2020 14Calibrating Nuclear Recoils
A test chamber at Fermilab)
B. Lenardo, April APIS, 2019
efficient, Hxp ∝ x/p
• Lindhard
2 concentration on -S1
successful atgain,
& S2 yield, S2 highand Xe response
oltage
energies
onse to proton recoils
- But recent lower yields in Si, Xe.
Generic?
t of a low energy neutron source (UCSB/NU)
eVIN)
•
24 keVMust calibrate
n, iron below
for γ attenuation andthreshold
n E-
conduit gives 1-2.5 keV n
• Increasingly active area 124Sb-9Be
- Beams with variety of reactions
- Photo-nuclear sources + gamma
shield Sc
~2 keV Fe
- Notch filters 24 keV
- In-situ calibration ideal
• Need keV neutrons to calibrate H! Photo-nuclear source + filters
11
T. Shutt - 16th Rencontres du Vietnam, 1/7/2020 (HydroX, unpublished) 15has demonstratedthea 3.5ratioseVofbaseline
di↵erent resolution
signal channels, ( ) on while a enabling
its energy on shorter timescales after propa
2
45.6 cm collection area low
extremely [43].energyThere are no obstacles
threshold. materialTransition
boundaries.Edge This
Sensor (TES).excimer
triplet Pyle et prop
al.
Superfluid 4He
out that historically the timescale of energy
expected in further refinement, we expect the thresholds ballisticfrom thanks to the
absorber to low
TESdensity
has been of severely
phononme
in coming years to advance into the sub-eV regime, in the with superfluid mediumtimescales,
the TES response (assumingleadinga to
tem
while retaining areas. Such II. sensors
DETECTOR are LAYOUT
capable ofSuperfluid 4He
• Field ionization to measure ejected “Channeltron” with
field ionization tip
He atom. (Osterman, et. al, TAUP 2019)
• Response is complicated: recent
EFT approach (Esposito, TAUP 2019)
• Beyond this: collective modes
in He approaches mDM ~ keV
(Schutz, Zurek, arXiiv:1604.08206)
T. Shutt - 16th Rencontres du Vietnam, 1/7/2020 17Cryogenic Wild West
• Energy measured to ~meV DIAMOND DETECTORS FOR DIRECT DETECTION OF …
PhysRev D 99, 123005 (2019)
- Superconductors+TES (Hochberg,
Zhao, Zurek arxiV:1504.07237)
- Magnetic bubble chambers (Bunting
et al., arXiv:1701.06566)
- Optical phonons (Knapen et al.,
arXiv:1712.06598)
DIAMOND DETECTORS FOR DIRECT DETECTION OF … PHYS. REV. D 99, 123005 (2019)
- Diamond (Kurinsky, et al., PRD FIG. 9. Nuclear recoil projected reach at 95% C.L. for He (blue), dia
2019) 1 eV (left panel) and 10 meV (right panel). The dashed lines are for a gra
year exposure. The yellow region indicates the neutrino floor for He
-
(computed according to the formalism in Ref. [79]). Also shown in gray
Absorption in superconductor νCLEUS [94], CRESST-III [36,96], CDMSLite [97], Darkside-50 [98
(Hochberg et al., arXiv:1604.06800) intrinsically limited in dynamic range. To account for this, T
- 3D Dirac materials (Hochberg et we assume 3 orders of magnitude in dynamic range, similar
to what has been seen in detectors with O(eV) thresholds
de
to
al., arXiv:1708.08929) [34]. This means that the upper integration limit is set to
10 σ , where the threshold σ is assumed to be five times
3
t t
co
di
the resolution.
• For electron recoils: m ~ keV DM
We show the 95% C.L. projected reach, corresponding to
three signal events, for calorimetric diamond detectors with
fo
th
hi
the thresholds discussed in the previous section in Fig. 9,
FIG. 9. Nuclear recoil projected reach at 95% C.L. for He (blue),compared to the Sileading
diamond (green), low-mass
(red), and Xe (cyan)NRfor limits
energy from the of co
thresholds
1 eV (left1/7/2020
T. Shutt - 16th Rencontres du Vietnam, panel) and 10 meV (right panel). The dashed lines are forνcleus (sapphire,
a gram-day exposure, Ref. [94]),
while the CRESST-III
dot-dashed (CaWO
lines are for 4 , 18 cr
a kilogram-Backgrounds
• Backgrounds below ~keV are Hochberg et al., arXiv:1512.04533
1/
kg/
terra incognita keV/
• Little self shielding
Coherent mass in sub-GeV/c direct dark matter searchesday
at ~kgbackground
photon scattering 2
- Rates better than 1/kg/keV/day
Fermi
Alan E. Robinson
National Accelerator Laboratory, Batavia, Illinois, USA 60510 ⇤
only with enormous effort (Dated: March 17, 2017)
Proposed dark matter detectors with eV-scale sensitivities will detect a large background of atomic
• Robinson: coherent Compton
(nuclear) recoils from coherent photon scattering of MeV-scale photons. This background climbs
steeply below ⇠10 eV, far exceeding the declining rate of low-energy Compton recoils. The upcoming
generation of dark matter detectors will not be limited by this background, but further development
scattering rises at low energy of eV-scale and sub-eV detectors will require strategies, including the use of low nuclear mass target
materials, to maximize dark matter sensitivity while minimizing the coherent photon scattering
background.
• Non particle 'backgrounds”
ro-ph.IM] 17 Mar 2017
PACS numbers: 13.60.Fz, 95.35.+d
Figure 2. Di↵erential rate in units of dR⌫ /d log10 ED for the solar neutrino coherent nuclear scatteri
Robinson, arVix:1610.07656
- PICO particles
background on various target nuclei as well as the expected radiogenic background from cosmogen
3
2 H spallation of the absorber during fabrication and from U/Th/K contamination of the SuperCDM
Interest in sub-GeV/c mass thermal relic dark mat-
Differential cross section dσ/dEr (b/eV)
SNOLAB cryostat. 2
ter models has inspired ideas for direct detection experi- 100 0.1 GeV/c Dark Matter on Si
- CRESST crackophonics ~ca.
Ge Coherent
ments with eV-scale and sub-eV thresholds [1]. For such -2 Si Coherent
10 He Coherent
light dark matter, the recoil energy di↵erential scattering
2000
Ge Compton
rate is restricted to energies below the detectionmatter direction detection.
thresh- 10-4
However, there are two reasons why radiogenic backgrounds a
Si Compton
of secondary importance for light mass dark matter detection. First, the low energy cohere
olds of direct detection experiments motivated by weak- May be subject
neutrino scattering background from pp neutrinos is much larger than the background produc
-
scale and supersymmetric models [2, 3]. 10-6 to structure
WillPenetrating
these MeV-scale
proliferate ata background
photons are lowbytheatmospheric
for
radiogenic backgrounds (comptons, 10-8
Pb decay products, H) have characteristic ener
effects
neutrinos within the high mass dark matter region of interest. Secondly, all
210 3
all of these experiments with di↵erent mechanisms at
energy? scales which are much larger than the light mass dark matter region of interest (Summary
• As we approach neutrino floor for WIMPs, new
frontier has opened at low mass
• Wide variety of new ideas building on existing
technologies
• Exciting time for hunting dark matter
T. Shutt - 16th Rencontres du Vietnam, 1/7/2020 20You can also read
