on layer 00eddata.fnal.gov/lasso/summerstudents/papers/2011/somaschini.pdf · italian summer...
TRANSCRIPT
Aging Studies On Layer 00
Italian Summer Internship Report
Sept,23th 2011
Luca Somaschini Università degli Studi di Milano
Silicon Detectors
DEPLETION VOLTAGE:
The external voltage
(BIAS) required to extend the Electric
field across the whole bulk (i.e. to collect all the charges produced by ionizing particles)
9/23/11 Luca Somaschini 2
Radiation Damages
BULK DAMAGES: Besides
ionization, Hadrons can
DISPLACE lattice atoms;
9/23/11
Interstitial and vacancy defect creation.
exiting particle
incident particle
interstitial vacancy
Frenkel pair
OVERALL EFFECT: changes in the depletion voltage
Luca Somaschini 3
Evolution of Dep. Volt.
9/23/11 Luca Somaschini 4
Signal Bias Scan
Procedure to measure the silicon sensor depletion voltage Fit => Depletion voltage: 95% of charge is collected
9/23/11
Bias voltage (V)0 20 40 60 80 100 120 140 160 180 200
Clu
ster
Cha
rge
(AD
C c
ount
s)
0
5
10
15
20
25
30
35
40
effic
ienc
y
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Fit Parameters 0.735± p0 = 19.647
0.004± p1 = 0.061 1.229± p2 = 48.898 2.686± p3 = 91.203
Extremal Charges 0.19±Charge Max = 28.01 0.19±Charge Min = 10.70
Ladder: f843_0
Luca Somaschini 5
Radiation Dose
9/23/11
What is the relation between Absorbed Dose and Delivered
Luminosity?
If |z| < 50 cm
Measured using 916 dosimeters placed in the detector
* Tesarek, R. J.; D'Auria, S.; Hocker, A.; Kordas, K.; McGimpsey, S.; Worm, S. NIM A 514 p. 188
Luca Somaschini 6
Layer 00
9/23/11
We will restrict our analysis to Layer 00: • It’s the closest layer to the beampipe • It recieves highest radiation doses • The radiation dose received by L00 is sensitive to changes in beam position;
• Only good runs data will be considere; • Good Run List: goodrun_qcd_si.list
Luca Somaschini 7
Layer 00
9/23/11
Y-Z PLANE
2 Barrels of three segment each 94 cm long
72 Ladders Wides: Hamamatsu - Narrows: SGS
Thompson - Narrows Oxygenated: Micron
Luca Somaschini 8
Layer 00
9/23/11
X-Y PLANE
• 6 Wedges • 12 Ladders
• 6 Wides • 6 Narrows
Luca Somaschini 9
Approximation 0
9/23/11
• All the ladders are in the same position
• The beam is in the center of the beam-pipe
1.62 cm
1.62 cmLUMINOSITY ó DOSE (i.e. no differences between a DepVolt vs
Luminosity plot and a DepVoltage vs Dose )
Luca Somaschini 10
9/23/11
Approximation 0
Luminosity (pb-1)0 2000 4000 6000 8000 10000 12000
Dep
Vol
tage
(V)
50
100
150
200
250
f843_0 / ndf 2r 17 / 25
Prob 0.8817p0 3.425± 26.86 p1 0.0006835± 0.01816
/ ndf 2r 17 / 25Prob 0.8817p0 3.425± 26.86 p1 0.0006835± 0.01816
f843_0
Good Fits: low chi-square per
degree of freedom
Luca Somaschini 11
9/23/11
Approximation 0
Luminosity (pb-1)0 2000 4000 6000 8000 10000 12000
Dep
Vol
tage
(V)
50
100
150
200
250
300
f823_1 / ndf 2r 135.7 / 19
Prob 1.229e-19p0 4.893± 41.39 p1 0.0008957± 0.0181
/ ndf 2r 135.7 / 19Prob 1.229e-19p0 4.893± 41.39 p1 0.0008957± 0.0181
f823_1
Bad Fits: very high chi-
square per degree of freedom
Luca Somaschini 12
9/23/11
ChiSquareEntries 72Mean 3.052RMS 1.83
0 2 4 6 8 100
5
10
15
20
25
ChiSquareEntries 72Mean 3.052RMS 1.83
ChiSquare
Radiation Dose
We will work only on ladders with chi-squre per deg. of
freedom <3;
Selection of: 42 of 72
Luca Somaschini 13
Luminosity (pb-1)0 2000 4000 6000 8000 10000
Dep
Vol
tage
(V)
50
100
150
200
250
All Together
Expected Corrections
9/23/11
Two different behavior before inversion point
Luca Somaschini 14
Approximation 1
9/23/11
• Wide and Narrow Ladders have different radii
• The beam is in the center of the beam-pipe
1.35 cm1.62 cm
Conversion of Scan Luminosity to dose
Luca Somaschini 15
Dose (Krad)0 1000 2000 3000 4000 5000 6000 7000 8000
Dep
Vol
tage
(V)
50
100
150
200
250
All Together
Expected Corrections
9/23/11
Still two different behavior before inversion point
Luca Somaschini 16
Narrow Ladders
9/23/11
0 2000 4000 6000 8000 10000
50
100
150
200
250
300
350
400
450
500
NARROWs
Luca Somaschini 17
Wide Ladders
9/23/11
Dose (Krad)0 1000 2000 3000 4000 5000 6000 7000 8000
Dep
Vol
tage
(V)
50
100
150
200
250
WIDEs
The observed split is due to Wides
behaving differently to Narrows
Luca Somaschini 18
Approximation 2
9/23/11
• Use the real position of the ladders
• Consider the real position of the beam
E!ec
tive
dist
ance
Changes in the beam-ladders distance means changes in the absorbed
dose Luca Somaschini 19
Beam Position
9/23/11
The DepVoltage depends on the absorbed dose.
The absorbed dose depends on the beam position between two scans.
Luca Somaschini 20
Beam Position
9/23/11
Store Number1000 2000 3000 4000 5000 6000 7000 8000 9000
Y Be
am P
ositi
on (c
m)
0
0.1
0.2
0.3
0.4
0.5
0.6
Y Beam Position
Movement is about 0.5cm
Luca Somaschini 21
Beam Position
9/23/11
Store Number1000 2000 3000 4000 5000 6000 7000 8000 9000
Y Be
am P
ositi
on (c
m)
0
0.1
0.2
0.3
0.4
0.5
0.6
Y Beam Position
Simple Correction: divide into three
region with similar beam position
Luca Somaschini 22
9/23/11
Store Number0 2000 4000 6000 8000
Y Be
am P
ositio
n
0
0.1
0.2
0.3
0.4
0.5
0.6
Y Beam Position
Store0 1000 2000 3000 4000 5000 6000 7000 8000 9000
)-1
Lum
inos
ity (
nb
0
2000
4000
6000
8000
10000
12000
Luminosity Per Store
Store0 1000 2000 3000 4000 5000 6000 7000 8000 9000
)-1
Tota
l Lum
inos
ity (p
b
0
2000
4000
6000
8000
10000
12000
LuminosityLuminosity
Large beam offset had negligible contribution
Luca Somaschini 23
9/23/11
Plot @ a fixed dose of 6000 krad;
Phi Dependence
The Phi dependence is smaller after the correction. Note the difference of Chi-Square per deg. of freedom
/ ndf 2r 197 / 11p0 0.9± 141.4
)° (q-150 -100 -50 0 50 100 150
Dep
Vol
tage
(V)
0
20
40
60
80
100
120
140
160
180
/ ndf 2r 197 / 11p0 0.9± 141.4
/ ndf 2r 197 / 11p0 0.9± 141.4
Per Wedge Gruping
/ ndf 2r 49.41 / 11p0 0.9± 143.4
()q-150 -100 -50 0 50 100 150
Dep
Vol
tage
(V)
0
20
40
60
80
100
120
140
160
/ ndf 2r 49.41 / 11p0 0.9± 143.4
/ ndf 2r 49.41 / 11p0 0.9± 143.4
Per Wedge Grouping
Luca Somaschini 24
Dose (Krad)0 1000 2000 3000 4000 5000 6000 7000 8000
Dep
Volta
ge (V
)
40
60
80
100
120
140
160
180
200
220
All Together
Results
9/23/11
Reduction of point spread at low luminosity; no effect at high luminosities.
Luca Somaschini 25
Results
9/23/11
Dose (Krad)0 1000 2000 3000 4000 5000 6000 7000 8000
Dep
Vol
tage
(V)
40
60
80
100
120
140
160
180
200
220
All Together
Dose (Krad)0 1000 2000 3000 4000 5000 6000 7000 8000
Dep
Vol
tage
(V)
40
60
80
100
120
140
160
180
200
220
All Together
Non Corrected – Corrected
Luca Somaschini 26
Results
9/23/11
The increase of integrated luminosity has a larger effect compared to changes of beam
position
Luca Somaschini 27
Manufacturer
9/23/11
An important difference appears in the behaviour of ladders of different
manufacturer
Luca Somaschini 28
Dose (Krad)0 2000 4000 6000 8000 10000 12000
Dep
Volta
ge (V
)
0
50
100
150
200
250
300
SGS Thompson- Hamamatsu - MicronSGS Thompson- Hamamatsu - Micron
Manufacturer
9/23/11
Hamamatsu (Wide) SGS Thompson (Narrow)
Micron (Narrow Oxygenated)
Beam Position Corrected
Luca Somaschini 29
Manufacturer
9/23/11
Oxygenated ladders: • Inversion point at higer dose
• Smaller slope • Live ≈ twice as long as Hamamatsu
SLOPES (10-2 V/KRad):
Hamamatsu: 2.49 ± 0.01 SGS – Thompson: 1.78 ± 0.01
Micron (Oxygenated): 1.366 ± 0.006
Oxygenateds have a longer life!! Luca Somaschini 30
Conclusions
9/23/11
• BeamLine – Detector distance, for aging effects, is relevant only at low doses
• At higher doses the trend is driven by luminosity
• X-Y Slope and beam width are not relevant • Great differences come from manufacturers • Oxygenated ladders introduce great
improvements
Luca Somaschini 31
Further Investigations
9/23/11
• Influence of dose derived from beam losses and quenches
• Influence of temperature changes
Luca Somaschini 32
Thanks to
Mentor: Ankush Mitra
Institute of Physics - Academia Sinica
Supervisor: Benedetto Di Ruzza
FNAL
Thanks to
Fermilab Staff for the opportunity of
this summer
Back Up
9/23/11 Luca Somaschini 35
Silicon Detectors Particles traveling through a silicon
sensor can produce free charges.
P-N Junction creates an electric field
depleted from free charges.
Charges can be collected thanks to the
electric field. 9/23/11
Cathode
Luca Somaschini 36
Silicon Detectors Without bias voltage only charges in dep. region are collected.
An Extenal Potential (BIAS) is applied to
extend this region into the bulk.
The higher voltage is the wider electric field
is. 9/23/11 Luca Somaschini 37
Radiation Damages
HAMBURG MODEL: complex THE OVERALL EFFECT: changes in the depletion voltage
REMEMBER THAT: Radiation affects all
devices (detectors, chips …)
*Jeffery Wyss – Exposure to radiation damage: concepts, quantities, environments - http://agenda.infn.it/conferenceTimeTable.py?confId=3245
9/23/11 Luca Somaschini 38
Signal Bias Scan
WHAT: Procedure to measure the silicon sensor depletion voltage;
WHEN: at CDF is performed about every 6
months;
HOW: measure the collected chage at different bias voltages
9/23/11 Luca Somaschini 39
Signal Bias Scan
This gives one couple of values: DEPLETION VOLTAGE vs DELIVERED LUMINOSITY
For: • Each Ladder • Each Bias scan
9/23/11 Luca Somaschini 40
Bias Scan Each Landau X Gauss Gives a point at a
different voltage;
9/23/11
Bias voltage (V)0 20 40 60 80 100 120 140 160 180 200
Clu
ster
Cha
rge
(AD
C c
ount
s)
0
5
10
15
20
25
30
35
40
effic
ienc
y
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Fit Parameters 0.735± p0 = 19.647
0.004± p1 = 0.061 1.229± p2 = 48.898 2.686± p3 = 91.203
Extremal Charges 0.19±Charge Max = 28.01 0.19±Charge Min = 10.70
Ladder: f843_0
Luca Somaschini 41
Dose Mapping
Distributio of Dose obtained
with 916 dosimeters
http://cdf-radmon.fnal.gov/
public/plots
9/23/11 Luca Somaschini 42
Dosimeters Mapping
9/23/11
Dispalcement of the 916
dosimeters
Luca Somaschini 43
Expected Corrections
9/23/11
Some error bars don’t cross the fitted line.
Is that systematics or is it caused by
our approximation?
Luca Somaschini 44
Expected Corrections
9/23/11 Luminosity (pb-1)
0 2000 4000 6000 8000 10000 12000
Dep
Vol
tage
(V)
50
100
150
200
250
f843_0 / ndf 2r 17 / 25
Prob 0.8817p0 3.425± 26.86 p1 0.0006835± 0.01816
/ ndf 2r 17 / 25Prob 0.8817p0 3.425± 26.86 p1 0.0006835± 0.01816
f843_0
Luca Somaschini 45
Luminosity (pb-1)0 2000 4000 6000 8000 10000
Dep
Vol
tage
(V)
50
100
150
200
250
All Together
Expected Corrections
9/23/11
Each ladder has a different behavior
Luca Somaschini 46
Expected Corrections
9/23/11
Luminosity (pb-1)0 2000 4000 6000 8000 10000 12000
Dep
Vol
tage
(V)
50
100
150
200
250
f843_0 / ndf 2r 17 / 25
Prob 0.8817p0 3.425± 26.86 p1 0.0006835± 0.01816
/ ndf 2r 17 / 25Prob 0.8817p0 3.425± 26.86 p1 0.0006835± 0.01816
f843_0
Different Ladders fits have different slopes
Luca Somaschini 47
Expected Corrections
9/23/11
Luminosity (pb-1)0 2000 4000 6000 8000 10000 12000
Dep
Vol
tage
(V)
50
100
150
200
250
f843_0 / ndf 2r 17 / 25
Prob 0.8817p0 3.425± 26.86 p1 0.0006835± 0.01816
/ ndf 2r 17 / 25Prob 0.8817p0 3.425± 26.86 p1 0.0006835± 0.01816
f843_0
Different Ladders fits could overlap
after the correction for the beam
position
Luca Somaschini 48
Expected Corrections
9/23/11
Let’s check by starting to take into account some details, if we can, to
improve our fits.
Luca Somaschini 49
Beam Position
9/23/11
X Beam Position (cm)-0.4 -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05
Y Be
am P
ositi
on (c
m)
0
0.1
0.2
0.3
0.4
0.5
0.6
XY Beam PositionBeam Position
for all RunII stores
Luca Somaschini 50
Beam Position
9/23/11
Store Number1000 2000 3000 4000 5000 6000 7000 8000 9000
X Be
am P
ositi
on (c
m)
-0.4
-0.35
-0.3
-0.25
-0.2
-0.15
-0.1
-0.05
X Beam Position
Movement is about 0.1 cm
Luca Somaschini 51
()q-150 -100 -50 0 50 100 150
Dep
Volta
ge (V
)
0
20
40
60
80
100
120
140
160
Per Wedge GroupingPer Wedge Grouping
9/23/11
Plot @ a fixed dose of 6000 krad;
With Correction Without Correction
Phi Dependence
Luca Somaschini 52
Z (cm)-50 -40 -30 -20 -10 0 10 20 30 40 50
Dep
Volta
ge (V
)
0
20
40
60
80
100
120
140
160
Per Barrel Grouping - AllPer Barrel Grouping - All
9/23/11
Plot @ a fixed dose of 6000 krad;
With Correction Without Correction
Z Dependence
The z dependence is present but is small
Luca Somaschini 53
Z (cm)-50 -40 -30 -20 -10 0 10 20 30 40 50
Dep
Vol
tage
(V)
0
20
40
60
80
100
120
140
160
Per Barrel Grouping - AllPer Barrel Grouping - All
9/23/11
Plot @ a fixed dose of 6000 krad;
With Correction Without Correction
Z Dependence
Luca Somaschini 54
Z (cm)-50 -40 -30 -20 -10 0 10 20 30 40 50
Dep
Volta
ge (V
)
0
20
40
60
80
100
120
140
160
Per Barrel Grouping - AllPer Barrel Grouping - All
9/23/11
Plot @ a fixed dose of 6000 krad;
Z Dependence
Asymmetry: maybe influence of z beam
position
Luca Somaschini 55
Store Number1000 2000 3000 4000 5000 6000 7000 8000 9000
Z Be
am P
ositi
on (c
m)
-15
-10
-5
0
5
10
15
Z Beam Position
Beam Position
9/23/11
It would require an extra investigstion
Luca Somaschini 56
Other Corrections
9/23/11
Store Number0 1000 2000 3000 4000 5000
X Be
am W
itdh
(cm)
0
0.0005
0.001
0.0015
0.002
0.0025
0.003
0.0035
0.004
X Beam WidthX Beam WidthX BEAM WIDTH
IRRELEVANT: 2 order of magnitude smaller than
position effects
Luca Somaschini 57
Other Corrections
9/23/11
Store Number0 1000 2000 3000 4000 5000
Y Be
am P
ositio
n (c
m)
0
0.0005
0.001
0.0015
0.002
0.0025
0.003
0.0035
0.004Y Beam WidthY Beam WidthY BEAM WIDTH
IRRELEVANT: 2 order of magnitude smaller than
position effects
Luca Somaschini 58
Other Corrections
9/23/11
Store Number0 2000 4000 6000 8000
X Be
am S
lope (
mili-r
ad)
0.0002
0.0003
0.0004
0.0005
0.0006
0.0007
0.0008
0.0009
0.001
X Beam SlopeX Beam SlopeX BEAM SLOPE
Luca Somaschini 59
Other Corrections
9/23/11
Store Number0 2000 4000 6000 8000
Y Be
am S
lope
(mili-
rad)
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
-310×
Y Beam SlopeY Beam SlopeY BEAM SLOPE
Luca Somaschini 60
Slope to Distance
9/23/11
R = tan(slope) * 94 cm ≈ 0.4 – 1 micron
IRRELEVANT: 3 order of magnitude smaller than
position effects
R
Luca Somaschini 61
9/23/11
CDF Overview
Luca Somaschini 62
9/23/11
CDF Overview
Luca Somaschini 63
9/23/11
CDF Silicon Overview
Luca Somaschini 64
9/23/11
Dose vs Radius
Evolution of equivalent dose with
distance from the beam
Luca Somaschini 65
9/23/11
Dose vs Radius
Official CDFII plot.
Dep Voltage vs Luminosity.
NO
CORRECTIONS
Luca Somaschini 66