Fermi National Accelerator Laboratory, October 2006
The measurement of the SiPM photon detection efficiency at ITC-irst
Nicoleta Dinu
INFN – Trento, Italy
Nicoleta Dinu 2Fermi National Accelerator Laboratory, October 2006
• Photon Detection Efficiency (PDE) of the SiPM
• Experimental methods used for the PDE measurement
• PDE of the first SiPM prototypes developed at ITC-irst
• Summary and outlook
Outline
Nicoleta Dinu 3Fermi National Accelerator Laboratory, October 2006
Photon detection efficiency of the SiPM
geomtriggeringSiPM PQE εη ××=
1. QE – the quantum efficiency• probability that a photon generate an e/h pair in the active region of the device (e.g. n+/p
junction of a pixel) - wavelength dependent
a) transmission efficiency through ARC b) internal quantum efficiency(ARC = stack of dielectric layers with appropriate thickness and refraction index)
sourcelightbyemitedphotonsofnrrecordedpulsesoutputofnr
. .
=ηTraditional PDE:
PDE of the SiPM:
p+
πpn+
Anti-ReflectiveCoating (ARC)
SiPM pixel
p+
πpn+
ARC
SiPM pixel
Nicoleta Dinu 4Fermi National Accelerator Laboratory, October 2006
Total area includes dead regions given by:• quenching resistors• trenches• metal layers
Active area: • ∼ 15- 30% of total area
depending of the layout design
(design not in scale)
-VbiasBack contactp+ silicon wafer
Front contact
ppnn++ppππ nn++
Active area
Trenches with Al
Rquenching
hhνν
Photon detection efficiency of the SiPM (cont)2. Ptriggering – the triggering probability (Pt = Pe + Ph - Pe*Ph)
• probability that a carrier (e or h) traversing the high field region triggers an avalanche• Pe & Ph are linked to the impact ionization rates of the electrons and holes
• electrons have higher ionization rates than holes• both electrons and holes ionization rates increase with the electric field (e.g. overvoltage)
3. εgeom – the geometrical efficiency (active area / total area)
SiPM
Nicoleta Dinu 5Fermi National Accelerator Laboratory, October 2006
The experimental set-up
Halogen white lamp(OSRAM)
Neutral filters(filter transmittance 10%)
Monocromator(Jobin Yvon HR250)
Photodiode (UDT 221)
SiPM + preamplifier
Optometer
Stage with 3D micrometric movement (50µm precision)
Nicoleta Dinu 6Fermi National Accelerator Laboratory, October 2006
The methods for the PDE measurement
surfaceSiPMtheonphotonsincident
SiPM thebyrecordedphotonsSiPM N
N
=η
surface SiPMthe onphotonsincidentN
reference calibrated detector(photodiode)
hcλΦN mmW
mmsphinc⋅= )
2/(2//. .
SiPMthecorded by photons reN
DC methodqG
IIN
SiPM
darklight
mms phrec ⋅−
=2//..
Pulses counting method
counts/s) (dark - counts/s) (lightNmms phrec
2//..=
1
2
(for low optical power densities)
Nicoleta Dinu 7Fermi National Accelerator Laboratory, October 2006
1.E+05
1.E+06
1.E+07
1.E+08
1.E+09
1.E+10
1.E+11
1.E+12
300 350 400 450 500 550 600 650 700 750
wavelength (nm)
Nr.
of in
cide
nt p
hoto
ns/ s
/ mm
^2
Set-up calibration
1.E+05
1.E+06
1.E+07
1.E+08
1.E+09
1.E+10
1.E+11
1.E+12
300 350 400 450 500 550 600 650 700 750
wavelength (nm)
Nr.
of in
cide
nt p
hoto
ns/ s
/ mm
^2
0 filter1st filter2nd filter3rd filter1+2 filters calculated1+2+3 filters calculated1+2 filters measured
hcλΦN mmW
mmsphinc⋅= )/(
2
2//. .
)/(
)(
)()/(
2
2 1WA
A
mmmmW
phot
phot
phot RI
A⋅=Φ
no filters
1 filter2 filters3 filters
Photodiode sensibility:• ~ 4 x 107 photons/s/mm2
Calibration method:• light beam without any filter• each filter separately• each filter factor is calculated at all
wavelengths• if 2 or 3 filters are inserted
simultaneously, the optical power density is calculated based on each filter factor determined previously
Nicoleta Dinu 8Fermi National Accelerator Laboratory, October 2006
1.5V overvoltage: y = 0.0652x + 8772.82.0V overvoltage: y = 0.0844x + 168112.5V overvoltage: y = 0.1109x + 106523.0V overvoltage: y = 0.1261x + 179103.5V overvoltage: y = 0.1461x + 244344.0V overvoltage: y = 0.1674x + 30282
1.0E+04
1.0E+05
1.0E+06
1.0E+07
1.0E+08
1.E+05 1.E+06 1.E+07 1.E+08 1.E+09
incident photons /s/mm^2
reco
rded
pho
tons
/s/m
m^2
2.0V overvoltage: y = 0.0824x + 5054.32.5V overvoltage: y = 0.1004x + 6517.63.0V overvoltage: y = 0.1222x + 4809.33.5V overvoltage: y = 0.1424x + 103324.0V overvoltage: y = 0.1618x + 11199
1.E+04
1.E+05
1.E+06
1.E+07
1.E+08
1.E+05 1.E+06 1.E+07 1.E+08 1.E+09
incident photons /s/mm^2
reco
rded
pho
tons
/s/m
m^2
PDE @ 550nm – DC & pulses counting methods (1) DC method
device with εgeom ~ 22%
device with εgeom ~ 22%
Pulses counting method
qGII
NSiPM
darklightmms phrec ⋅
−=2//..
counts/s) (dark - counts/s) (lightNmms phrec
2//..=
Nicoleta Dinu 9Fermi National Accelerator Laboratory, October 2006
PDE @ 550nm – DC & pulses counting methods (2)
Very good agreement in between the DC and counting pulses methodsPDE increases linearly with the overvoltage at least up to 5V overvoltage
0
5
10
15
20
25
0 1 2 3 4 5
Overvoltage (V)
PDE
@ 5
50nm
(%)
Pulses counting methodDC methodLinear (Pulses counting method)Linear (DC method)
Nicoleta Dinu 10Fermi National Accelerator Laboratory, October 2006
0
2
4
6
8
10
12
14
16
18
300 400 500 600 700 800 900
wavelength (nm)
PDE
%
1.5V overvoltage2.0V overvoltage2.5V overvoltage3.0V overvoltage3.5V overvoltage
0
2
4
6
8
10
12
14
16
18
300 400 500 600 700 800 900
wavelengths (nm)
PDE
%
1.5V overvoltage2.0V overvoltage2.5V overvoltage3.0V overvoltage3.5V overvoltage4.0V overvoltage
Photon detection efficiency – DC method device with εgeom ~ 16%
device with εgeom ~ 22%
Maximum PDE in the range 500 ÷ 600 nm
• ~ 16% @ 4V overvoltage for a SiPM of εgeom ~ 22%
For low λ• the PDE is reduced by the Ptriggering
(only holes trigger the avalanche)For high λ
• the PDE is reduced by the QE(QE was optimized for low λ)
Nicoleta Dinu 11Fermi National Accelerator Laboratory, October 2006
Quantum efficiency
30
40
50
60
70
80
90
100
110
300 400 500 600 700 800 900
wavelengths (nm)
QE
%
0V-1V-2V-3V-4V-5V
diode with A = 1mm2
30
40
50
60
70
80
90
100
110
300 400 500 600 700 800 900
wavelengths (nm)
QE
%
0V-1V-2V-3V-4V -5V
diode with A = 0.5625mm2
Diode:• Test structure with the same (n+/p
junction + ARC) as each SiPM pixel• Works as a photodiode at low
reversed bias (0V, 1V or 2V)• Allows the measurement of the QE
(transmission through ARC & internal quantum efficiency)
The impact ionization effect already visible at 3-4V
QE > 95% in the blue region(optimized for λ~ 420nm)
Nicoleta Dinu 12Fermi National Accelerator Laboratory, October 2006
1.5V overvoltage2.0V overvoltage2.5V overvoltage3.0V overvoltage3.5V overvoltage
0
10
20
30
40
50
60
70
80
90
100
300 400 500 600 700 800
wavelengths (nm)
Trig
gerin
g pr
obab
ility
%
0102030405060708090
100
300 400 500 600 700 800
wavelengths (nm)
Trig
gerin
g pr
obab
ility
%
1.5V overvoltage2.0V overvoltage2.5V overvoltage3.0V overvoltage3.5V overvoltage4.0V overvoltage
Triggering probability
device with εgeom ~ 22%
device with εgeom ~ 16% Calculated from:
geomtriggeringSiPM PQE εη ××=
Only holes trigger the avalanche
Both holes and electrons trigger the avalanche
Only electrons trigger the avalanche
Nicoleta Dinu 13Fermi National Accelerator Laboratory, October 2006
Light absorption
13
14
15
16
17
18
19
20
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2depth (um)
Dop
ing
conc
.
0E+00
1E+05
2E+05
3E+05
4E+05
5E+05
6E+05
7E+05
E fie
ld (V
/cm
)
DopingField
pn+
h e
At low wavelengths only the holes cross the high field region & trigger the avalanche⇒ triggering probability @ low λ (e.g. 385, 390, 395nm) = hole triggering probability
At high wavelengths only the electrons cross the high field region & trigger the avalanche⇒ triggering probability @ high λ (e.g. 700nm) = electron triggering probability
Simulated doping profile and electric field of the SiPMAttenuation of the light intensity in silicon (Beer-Lambert law)
Nicoleta Dinu 14Fermi National Accelerator Laboratory, October 2006
0.00
0.20
0.40
0.60
0.80
1.00
1.20
0 1 2 3 4 5
Overvoltage (V)
trigg
erin
g pr
obab
ility
hole probability (390nm)hole probability (395nm)electron probability (730nm)Lineare (hole probability (390nm))Lineare (hole probability (395nm))Lineare (electron probability (730nm))
Electron & hole triggering probability
Pholes
Pelectrons
Pe & Ph increase linearly with the overvoltage up to 4VThe slight difference of 0V point could arises from the unavoidable statistical variation of the Vbreakdownacross the structure
Ref. data: W. Oldham & all,• “Triggering phenomena in avalanche
diodes”, IEEE Trans. on Electron Devices Vol. ED-19, No.9, Sept. 1972
Nicoleta Dinu 15Fermi National Accelerator Laboratory, October 2006
Photon detection efficiency of the SiPM devices developed at ITC-irst
Two experimental methods:• DC & pulses counting• very good agreement in between the two methods (λ=550nm)
Photon detection efficiency:• Depends of three factors:
• geometrical efficiency: ~15-30% (e.g. function of the layout design)• quantum efficiency: > 95% in the blue region (optimized for 420nm)• triggering probability: Pe > Ph
• Maximum in the range 500-600 nm :• ~ 16% @ 4V overvoltage for a device of εgeom= 22%• for low λ it is reduced by the Ptriggering (only holes trigger the avalanche)• for high λ it is reduced by the QE (optimized for blue region)
• Increases linearly with the overvoltage (at least up to 4V overvoltage)
Summary
Top Related