characterization of cf 4 primary scintillation andrey morozov
TRANSCRIPT
Characterization of CFCharacterization of CF44 primary primary
scintillationscintillation
Andrey Morozov
What, why and how?What, why and how?
Light emission and photon yield
Effect of applied electric field
Effect of He admixture
Quenching effects (gas aging)
Spectral studies Photon flux measurements
Time resolved measurements
CF4 primary scintillation:
Spectral studiesSpectral studies
Spectral studies: SetupSpectral studies: Setup
MonochromatorPMT
Am-241α-source
Al reflectorR=0.80 ± 0.08
Excitation volume
Quartz window
Grid (0 – 6 kV)
Second order filter
Photon counting
CF4
CFCF44 primary scintillation: Raw spectra primary scintillation: Raw spectra
!Spectra have to be
corrected for:
1) Response of the monochromator and PMT
2) Geometrical factor of the detection
3) Gas aging
Monochromator+PMT response measurementsMonochromator+PMT response measurements
MonochromatorPMT
ES: Entrance slitIF: Optional interference filtersL: LensNF: Neutral filter(s)D: Diaphragm
Tungstenstrip lamp(1900 K)450 – 850 nm
Tungstenhalogen lamp400 – 850 nm
Deuterium lamp200 – 400 nm
ES IF L
NF
D
Spectra corrected for the instrumental responseSpectra corrected for the instrumental response
Spectra from different pressuresare not to scale!
200 300 400 500 600 700 8000.0
0.1
0.2
0.3
0.3
0.4
0.5
0.6
0.7
0.8
0.8
0.9
1.0
Inte
nsity
, a.u
.
Wavelength, nm
0.5 bar 1 bar 2 bar 3 bar 4 bar 5 bar
Need directphoton fluxmeasurements
Absolute photon flux measurementsAbsolute photon flux measurements
Absolute photon flux measurements: setupAbsolute photon flux measurements: setup
PMT
Extension tubes:PMT-to-source distanceis adjustablefrom 80 to 300 mm
Broad band filter (UV or visible)
Light source is ahalf-sphere, radius:~3 mm at 5 bar~13 mm at 1 bar~26 mm at 0.5 bar
Optional interference filter
Extensiontubes
At the limitgiven by the cellgeometry!
Point-light-source approximation:
If valid, no numerical simulations (unknown error bars!) are needed.
Photon fluxes:FOBSERVED ∞ NSOURCE / R2
then FOBSERVED∙R2 should be constant!
Flux vs. PMT-to-source distanceFlux vs. PMT-to-source distance
RLight sourceDetector
Flux vs. PMT-to-source distanceFlux vs. PMT-to-source distance
PMT can “see” the whole light emitting volume!
VISIBLE componentUV component
80 120 160 200 240 2800.00
0.02
0.04
0.06
0.08
0.10
0.12
Fig.2
5 bar4 bar
3 bar
2 bar
1 bar
Cou
nt_r
ate
x R2 , 1
09 m
m2 /s
R, mm
80 120 160 200 240 2800.0
0.5
1.0
1.5
2.0
2.5
3.0
5 bar
2 bar
3 bar
4 bar
1 bar
Cou
nt_r
ate*
R2 , 1
07 m
m2 /s
R, mm
Flux vs. PMT-to-source distanceFlux vs. PMT-to-source distanceMissing double-event photoelectrons are taken into account!
VISIBLE componentUV component
Point source approximation is definitely valid for R > 140 mm for all pressures! No scattering!
80 120 160 200 240 2800.0
0.5
1.0
1.5
2.0
2.5
3.0
5 bar
2 bar
3 bar
4 bar
1 bar
Cou
nt_r
ate*
R2 , 1
07 m
m2 /s
R, mm80 120 160 200 240 280
0.00
0.02
0.04
0.06
0.08
0.10
0.12
Fig.2
5 bar4 bar
3 bar
2 bar
1 bar
21%
16%
8%
10%Cou
nt_r
ate
x R2 , 1
09 m
m2 /s
R, mm
30%
6%
12%
Fluxes for the UV and VIS componentsFluxes for the UV and VIS components
Two lines of fluxmeasurements:
Using broad-band filters andinterference filters.
Photon detection probability vs. wavelength?
XP2020QR1387
PMT
Halogen or deuterium lamp
Neutral filters
Optical beam-line (2 m long) with diaphragms
Interference filter
Photon detection probabilityPhoton detection probability
Interference filters:
200 300 400 500 600 700 8000.0
0.5
1.0
Relative detection probabilitiesRelative detection probabilitiesUV component
PMT: Photonis XP2020Q
Visible component
PMT: Hamamatsu R1387
Max uncertainties (main contributors: the filters and lamps):
25% 35%
CFCF44 emission spectra emission spectra
(all corrections are applied)(all corrections are applied)
Flux measurements have been used to scale the spectra from different pressures
200 300 400 500 600 700 800
0.0
0.2
0.4
0.6
0.8
1.0
Pho
ton
flux
, a.u
.
Wavelength, nm
1 bar 2 bar 3 bar 4 bar 5 bar
Fig.1
Photon yieldPhoton yield
Photon yield calculationPhoton yield calculation
Y = Number_of_photons / (α-Rate ∙ Average_Energy)
From photon flux
measurements
Number of α-particles entering the gas per second multiplied by the average deposited energy
Photon yield: Number of photons (in 4π) produced per MeV of energy deposited in the gas.
Photon yield (Photon yield (UVUV component, integrated) component, integrated)
Uncertainties
Broadband:
16% sum in quadrature
(40% sum)
IF-300 nm:
21% sum in quadrature
(55% sum)0 1 2 3 4 50
500
1000
1500
2000
2500
3000
Fig.3
Broadband (UG11) IF (300nm)
Pho
ton
yiel
d, p
h/M
eV
Pressure, bar
Photon yield (Photon yield (visiblevisible component, integrated) component, integrated)
Uncertainties
Broadband:
19% sum in quadrature
(50% sum)
IF-300 nm:
5 → 1bar:
26% → 72% sum in quadrature
(67% → 120% sum)
0 1 2 3 4 50
1000
2000
3000
4000
5000
6000
7000
Fig.4
Broadband (UV-block) IF (610nm)
Pho
ton
yiel
d, p
h/M
eV
Pressure, bar
Photon yield: cross-checkPhoton yield: cross-check
P
bar
Yield ratios
UV / vis
Spectral ratios
UV / vis
1 2.8 ± 1.0 ~2.5 ± 0.8
2 0.70 ± 0.25 0.76 ± 0.15
3 0.34 ± 0.12 0.50 ± 0.10
4 0.22 ± 0.08 0.35 ± 0.07
5 0.17 ± 0.06 0.23 ± 0.05
The ratios of yields in the UV and visible are comparedwith the flux ratios from the calibrated emission spectra
Overlap! 200 300 400 500 600 700 800
0.0
0.5
1.0
Time spectraTime spectra
Time spectra: setupTime spectra: setup
PMT
Optical filter
Silicon detector(PIPS Canberra)
0 – 300 V
Photon: “Stop”
α-particle: “Start”
Alpha source (Ortec)
Electronics:Start and stop → fast preamps → CF triggers → Time-to-amplitude converter → Multi-channel analyzer
Time spectra: UV componentTime spectra: UV component
Two distinctive decay structures (both ? exponential)
Indications of a very weak non-exponential tail
Weak pressure dependence
Time spectra: Visible componentTime spectra: Visible component
Strongly non-exponential decay
Very likely there is a strong contribution from recombination Clear pressure dependence
Effect of the electric fieldEffect of the electric field
Electric field: SpectraElectric field: Spectra
No effect for low pressures!
For medium-high pressures:Very strong reduction of the visible component and an increase of the UVcomponent with the field!
At 5 bar:Strong reshaping of the UVcomponent with the field!
Electric field: UV componentElectric field: UV component
1 2 3 4 50.0
0.2
0.4
0.6
0.8
1.0
Inte
grat
ed p
hoto
n flu
x, a
.u.
Pressure, bar
220-500 nmrange 0 V/cm 370 V/cm 560 V/cm 1100 V/cm 2200 V/cm
Electric field: Visible componentElectric field: Visible component
1 2 3 4 50.0
0.2
0.4
0.6
0.8
1.0
Inte
grat
ed p
hoto
n flu
x, a
.u.
Pressure, bar
500-800 nm range 0 V/cm 75 V/cm 280 V/cm 560 V/cm 1900 V/cm
Electric field: Electric field: Time spectra of the UV componentTime spectra of the UV component
Very small difference!
E-field dependence: Visible componentE-field dependence: Visible component
0 20 40 60 80 100 120 1401
10
100
1000 3 bar 0 kV/cm 0.5 kV/cm 3 kV/cm
Cou
nts
Time, ns
Very different behavior at high and low pressures!
0 20 40 60 80 100 120 140
10
100
1000 1 bar 0 kV/cm 1 kV/cm
Cou
nts
Time, ns
3 bar 1 bar
E-field dependence: Why UV and visibleE-field dependence: Why UV and visible behave differently? behave differently?
CF3+
CF4+
Recombination UV emission
CF3*Direct excitation
QuenchingVisible emission
Electric field suppressesrecombination UV emissionbenefit from thatwhilevisible emissionshould be reduced
Effect of helium: spectra and flux measurementsEffect of helium: spectra and flux measurements
Visible component(pressures are in bar)
3 CF4 + 2 He: same as 3 CF4
2 CF4 + 3 He: same as 2 CF4
1 CF4 + 4 He: same as 1 CF4
UV component(pressures are in bar)
3 CF4 + 2 He: same as 3 CF4
2 CF4 + 3 He: 20% higher than 2 CF4
1 CF4 + 4 He: 40% higher than 1 CF4
Absolute flux measurements:
Gas aging studies: Time spectraGas aging studies: Time spectra
UV component Visible component
Both components show significant aging!
NN22 quenching quenching
UV component
No effect!
Visible component
Strong effect!
Visible component is strongly affected by nitrogen admixtures!Bad news, since CF4 purifiers (effective) do not remove nitrogen.
Thank you!