Transcript Radiation Damage in Silicon Detectors

EPFL – LPHE, Lausanne, January 29, 2007
Radiation Tolerant Sensors
for Solid State Tracking Detectors
- CERN-RD50 project –
http://www.cern.ch/rd50
Michael Moll
CERN - Geneva - Switzerland
RD50
Outline
 Introduction: LHC and LHC experiment
 Motivation to develop radiation harder detectors
 Introduction to the RD50 collaboration
 Part I: Radiation Damage in Silicon Detectors (A very brief review)


Microscopic defects (changes in bulk material)
Macroscopic damage (changes in detector properties)
 Part II: RD50 - Approaches to obtain radiation hard sensors


Material Engineering
Device Engineering
 Summary and preliminary conclusion
Michael Moll – Lausanne, 29. January 2007
-2-
RD50
LHC - Large Hadron Collider
Start : 2007
• Installation in
existing LEP tunnel
• 27 Km ring
p
p
• 1232 dipoles B=8.3T
•  4000 MCHF
(machine+experiments)
• pp s = 14 TeV
Ldesign = 1034 cm-2 s-1
• Heavy ions
(e.g. Pb-Pb at
s ~ 1000 TeV)
LHC experiments located at 4 interaction points
Michael Moll – Lausanne, 29. January 2007
-3-
RD50
LHC Experiments
+ LHCf
CMS
Michael Moll – Lausanne, 29. January 2007
-4-
RD50
LHC Experiments
LHCf
CMS
Michael Moll – Lausanne, 29. January 2007
-5-
RD50 LHC example: CMS inner tracker
Inner Tracker
CMS
Outer Barrel
Inner Barrel
(TOB)
End Cap
(TIB)
(TEC)
Inner Disks
2.4 m
(TID)
Total weight
12500 t
Diameter
15m
Length
21.6m
Magnetic field
4T
CMS – “Currently the Most Silicon”
 Micro Strip:
Pixel
Pixel Detector
 ~ 214 m2 of silicon strip sensors, 11.4 million strips
 Pixel:




Inner 3 layers: silicon pixels (~ 1m2)
66 million pixels (100x150mm)
Precision: σ(rφ) ~ σ(z) ~ 15mm
Most challenging operating environments (LHC)
Michael Moll – Lausanne, 29. January 2007
-6-
RD50
Status December 2006
 LHC Silicon Trackers close to or under commissioning
 CMS Tracker (12/2006)
(foreseen: June 2007 into the pit)
 ATLAS Silicon Tracker (08/2006)
August 2006 – installed in ATLAS
CMS Tracker Outer Barrel
Michael Moll – Lausanne, 29. January 2007
-7-
RD50
Motivation for R&D on
Radiation Tolerant Detectors: Super - LHC
• LHC upgrade
SUPER - LHC (5 years, 2500 fb-1)
LHC (2007), L = 1034cm-2s-1
Ministrip (?)
Pixel (?)
16
10
f(r=4cm) ~ 3·1015cm-2
500 fb-1
5
Super-LHC (2015 ?), L = 1035cm-2s-1
5 years
f(r=4cm) ~ 1.6·1016cm-2
2500 fb-1
• LHC (Replacement of components)
e.g. - LHCb Velo detectors (~2010)
- ATLAS Pixel B-layer (~2012)
Macropixel (?)
5
total fluence eq
eq [cm-2]
10 years
1015
5
neutrons eq
pions eq
1014
5
ATLAS SCT - barrel
(microstrip detectors)
ATLAS Pixel
13
10
0
10
other charged
hadrons eq
20
30
40
50
60
[M.Moll, simplified, scaled from ATLAS TDR]
r [cm]
• Linear collider experiments (generic R&D)
Deep understanding of radiation damage will be fruitful for linear collider experiments where
high doses of e, g will play a significant role.
Michael Moll – Lausanne, 29. January 2007
-8-
RD50
The CERN RD50 Collaboration
http://www.cern.ch/rd50
RD50: Development of Radiation Hard Semiconductor Devices for High Luminosity Colliders
 Collaboration formed in November 2001
 Experiment approved as RD50 by CERN in June 2002
 Main objective:
Development of ultra-radiation hard semiconductor detectors for the luminosity
upgrade of the LHC to 1035 cm-2s-1 (“Super-LHC”).
Challenges: - Radiation hardness up to 1016 cm-2 required
- Fast signal collection (Going from 25ns to 10 ns bunch crossing ?)
- Low mass (reducing multiple scattering close to interaction point)
- Cost effectiveness (big surfaces have to be covered with detectors!)
 Presently 264 members from 52 institutes
Belarus (Minsk), Belgium (Louvain), Canada (Montreal), Czech Republic (Prague (3x)),
Finland (Helsinki, Lappeenranta), Germany (Berlin, Dortmund, Erfurt, Freiburg, Hamburg, Karlsruhe), Israel
(Tel Aviv), Italy (Bari, Bologna, Florence, Padova, Perugia, Pisa, Trento, Turin), Lithuania (Vilnius),
The Netherlands (Amsterdam), Norway (Oslo (2x)), Poland (Warsaw (2x)), Romania (Bucharest (2x)), Russia
(Moscow), St.Petersburg), Slovenia (Ljubljana), Spain (Barcelona, Valencia), Switzerland (CERN, PSI),
Ukraine (Kiev), United Kingdom (Diamond, Exeter, Glasgow, Lancaster, Liverpool, Sheffield),
USA (Fermilab, Purdue University, Rochester University, SCIPP Santa Cruz, Syracuse University, BNL,
University of New Mexico)
Michael Moll – Lausanne, 29. January 2007
-9-
RD50
Outline
 Motivation to develop radiation harder detectors
 Introduction to the RD50 collaboration
 Part I: Radiation Damage in Silicon Detectors
(A very brief review)
 Microscopic defects
 Macroscopic damage
(changes in bulk material)
(changes in detector properties)
 Part II: RD50 - Approaches to obtain radiation hard sensors


Material Engineering
Device Engineering
 Summary and preliminary conclusion
Michael Moll – Lausanne, 29. January 2007
-10-
RD50
Radiation Damage – Microscopic Effects
 Spatial distribution of vacancies created by a 50 keV Si-ion in silicon.
(typical recoil energy for 1 MeV neutrons)
M.Huhtinen 2001
van Lint 1980
I
V
I
V
particle
SiS
EK>25 eV
V
Vacancy
+
I Interstitial
point defects
(V-O, C-O, .. )
EK > 5 keV point defects and clusters of defects
Michael Moll – Lausanne, 29. January 2007
-11-
RD50
particle
Radiation Damage – Microscopic Effects
SiS
EK>25 eV
V
Vacancy
+
I Interstitial
point defects
(V-O, C-O, .. )
EK > 5 keV point defects and clusters of defects
60Co-gammas
Electrons
Neutrons (elastic scattering)
Compton Electrons
Ee > 255 keV for displacement  En > 185 eV for displacement
with max. Eg 1 MeV E > 8 MeV for cluster
 En > 35 keV for cluster
e
(no cluster production)
Only point defects
point defects & clusters
10 MeV protons
Mainly clusters
24 GeV/c protons
1 MeV neutrons
Simulation:
Initial distribution of
vacancies in (1mm)3
after 1014 particles/cm2
[Mika Huhtinen NIMA 491(2002) 194]
Michael Moll – Lausanne, 29. January 2007
-12-
RD50
Primary Damage and secondary defect formation
 Two basic defects
I - Silicon Interstitial
V - Vacancy
 Primary defect generation
I, I2 higher order I (?)
 I -CLUSTER (?)
V, V2, higher order V (?)
 V -CLUSTER (?)
I
V
Damage?!
 Secondary defect generation
I
V
Main impurities in silicon: Carbon (Cs)
Oxygen (Oi)
I+Cs  Ci  Ci+Cs  CiCS
Ci+Oi  CiOi
Ci+Ps  CiPS
V+V  V2
V+Oi  VOi 
V+Ps  VPs
V+V2 
V+VOi 
V3
V2Oi
I+V2  V
I+VOi 
Oi .......................
Damage?! (“V2O-model”)
Michael Moll – Lausanne, 29. January 2007
-13-
Example of defect spectroscopy
RD50
- neutron irradiated -
Deep Level Transient Spectroscopy
0.8
DLTS-signal (b1) [pF]
0.6
0.4
0.2
? + VV(-/0) + ?
Ci(-/0)
Introduction Rates
Nt/eq:
CiCs(-/0)
E(35K)
E(40K)
E(45K)
VOi(-/0)
Ci :
1.55 cm-1
VV(--/-)
0
-0.2
-0.4
-0.6
H(220K)
60 min
170 days
50
Ci(+/0)
100
CiCs :
0.40 cm-1
CiOi :
1.10 cm-1
CiOi(+/0)
150
200
Temperature [ K ]
 Introduction rates of main defects
 1 cm-1
 Introduction rate of negative space charge  0.05 cm-1
250
example : eq = 11014 cm-2
defects  11014 cm-3
space charge  51012cm-3
Michael Moll – Lausanne, 29. January 2007
-14-
RD50 Impact of Defects on Detector properties
Shockley-Read-Hall statistics
(standard theory)
charged defects
 Neff , Vdep
e.g. donors in upper
and acceptors in
lower half of band
gap
Trapping (e and h)
generation
 CCE
 leakage current
shallow defects do not
Levels close to
contribute at room
midgap
temperature due to fast
most effective
Inter-center charge
transfer model
(inside clusters only)
enhanced generation
 leakage current
 space charge
detrapping
Impact on detector properties can be calculated if all defect parameters are known:
n,p : cross sections
E : ionization energy
Nt : concentration
Michael Moll – Lausanne, 29. January 2007
-15-
RD50 Reverse biased abrupt p+-n junction
Poisson’s equation
q
d2
 2 f x   0  Neff
dx
 0
Electrical
charge density
Electrical
field strength
Positive space charge, Neff =[P]
(ionized Phosphorus atoms)
depleted
zone
neutral bulk
(no electric field)
+VB<Vdep
+VB>Vdep
particle
(mip)
Full charge collection only for VB>Vdep !
depletion voltage
Electron
potential energy
Vdep 
q0
 0
 N eff  d 2
effective space charge density
Michael Moll – Lausanne, 29. January 2007
-16-
Macroscopic Effects – I. Depletion Voltage
RD50
Change of Depletion Voltage Vdep (Neff)
…. with time (annealing):
1000
500
102
 600 V
type inversion
100
50
10
5
101
1014cm-2
1
10-1
100
"p-type"
n-type
[M.Moll: Data: R. Wunstorf, PhD thesis 1992, Uni Hamburg]
10
0
10
1
10
2
eq [ 10 cm ]
12
10
3
10
 Neff [1011cm-3]
10
5000
3
| Neff | [ 1011 cm-3 ]
Udep [V] (d = 300mm)
…. with particle fluence:
10-1
-2
• “Type inversion”: Neff changes from positive to
negative (Space Charge Sign Inversion)
before inversion
p+
n+
p+
n+
after inversion
8
6
NY
NA
4
NC
gC eq
2
NC0
[M.Moll, PhD thesis 1999, Uni Hamburg]
0
1
10
100
1000 10000
annealing time at 60oC [min]
• Short term: “Beneficial annealing”
• Long term: “Reverse annealing”
- time constant depends on temperature:
~ 500 years (-10°C)
~ 500 days ( 20°C)
~ 21 hours ( 60°C)
- Consequence: Detectors must be cooled
even when the experiment is not running!
Michael Moll – Lausanne, 29. January 2007
-17-
RD50
Radiation Damage – II. Leakage Current
Change of Leakage Current (after hadron irradiation)
…. with time (annealing):
…. with particle fluence:
-1
6
10-2
10-3
-4
10
10-5
10-6 11
10

n-type FZ - 7 to 25 Kcm
n-type FZ - 7 Kcm
n-type FZ - 4 Kcm
n-type FZ - 3 Kcm
p-type EPI - 2 and 4 Kcm
80 min 60C
1012
eq [cm ]
-2
1014
 is constant over several orders of fluence
and independent of impurity concentration in Si
 can be used for fluence measurement
80 min 60C
5
4
4
3
3
2
2
.
0
1
17
-3
oxygen enriched silicon [O] = 2 10 cm
parameterisation for standard silicon
1
[M.Moll PhD Thesis]
10
100
1000
o
10000
annealing time at 60 C [minutes]
[M.Moll PhD Thesis]
Leakage current
per unit volume
and particle fluence
6
5
1
1015
Damage parameter  (slope in figure)
I
α
V   eq

1013
n-type FZ - 780 cm
n-type FZ - 410 cm
n-type FZ - 130 cm
n-type FZ - 110 cm
n-type CZ - 140 cm
p-type EPI - 380 cm
(t) [10-17 A/cm]
I / V [A/cm3]
10


Leakage current decreasing in time
(depending on temperature)
Strong temperature dependence
 E

I  exp  g

2
k
T
B


Consequence:
Cool detectors during operation!
Example: I(-10°C) ~1/16 I(20°C)
Michael Moll – Lausanne, 29. January 2007
-18-
RD50
Radiation Damage – III. CCE (Trapping)
Deterioration of Charge Collection Efficiency (CCE) by trapping
Trapping is characterized by an effective trapping time eff for electrons and holes:
where
Inverse trapping time 1/ [ns-1]
Increase of inverse trapping time (1/) with fluence
0.5
24 GeV/c proton irradiation
0.4
data for electrons
data for holes
0.3
0.2
0.1
[M.Moll; Data: O.Krasel, PhD thesis 2004, Uni Dortmund]
0
0
2.1014 4.1014 6.1014 8.1014
1015
particle fluence - eq [cm-2]
1
 eff e,h
 N defects
….. and change with time (annealing):
Inverse trapping time 1/ [ns-1]


1

Qe,h (t )  Q0 e,h exp 
t
  eff e,h 


0.25
24 GeV/c proton irradiation
eq = 4.5.1014 cm-2
0.2
0.15
data for holes
data for electrons
0.1
[M.Moll; Data: O.Krasel, PhD thesis 2004, Uni Dortmund]
5 101
5 102
5 103
annealing time at 60oC [min]
Michael Moll – Lausanne, 29. January 2007
-19-
RD50
Summary: Radiation Damage in Silicon Sensors
Two general types of radiation damage to the detector materials:
 Bulk (Crystal) damage due to Non Ionizing Energy Loss (NIEL)
- displacement damage, built up of crystal defects –
Influenced
by impurities
in Si – Defect
Engineering
is possible!
I.
Change of effective doping concentration (higher depletion voltage,
under- depletion)
II. Increase of leakage current (increase of shot noise, thermal runaway)
Same for
III. Increase of charge carrier trapping (loss of charge)
all tested
Silicon
materials!  Surface damage due to Ionizing Energy Loss (IEL)
- accumulation of positive in the oxide (SiO2) and the Si/SiO2 interface –
affects: interstrip capacitance (noise factor), breakdown behavior, …
Impact on detector performance and Charge Collection Efficiency
(depending on detector type and geometry and readout electronics!)
Signal/noise ratio is the quantity to watch
 Sensors can fail from radiation damage !
Can be
optimized!
Michael Moll – Lausanne, 29. January 2007
-20-
RD50
Outline
 Motivation to develop radiation harder detectors
 Introduction to the RD50 collaboration
 Part I: Radiation Damage in Silicon Detectors (A very brief review)


Microscopic defects
Macroscopic damage
(changes in bulk material)
(changes in detector properties)
 Part II: RD50 - Approaches to obtain radiation hard sensors


Material Engineering
Device Engineering
 Summary and preliminary conclusion
Michael Moll – Lausanne, 29. January 2007
-21-
RD50
Approaches of RD50 to develop
radiation harder tracking detectors
 Defect Engineering of Silicon
 Understanding radiation damage
Scientific strategies:
• Macroscopic effects and Microscopic defects
• Simulation of defect properties and defect kinetics
• Irradiation with different particles at different energies
 Oxygen rich silicon
• DOFZ, Cz, MCZ, EPI
I.
Material engineering
II.
Device engineering




Oxygen dimer enriched silicon
Hydrogen enriched silicon
Pre-irradiated silicon
Influence of processing technology
 New Materials
III. Variation of detector
operational conditions
CERN-RD39
“Cryogenic Tracking Detectors”
 Silicon Carbide (SiC), Gallium Nitride (GaN)
 Diamond: CERN RD42 Collaboration
 Device Engineering (New Detector Designs)





p-type silicon detectors (n-in-p)
Thin detectors
3D and Semi 3D detectors
Cost effective detectors
Simulation of highly irradiated detectors
Michael Moll – Lausanne, 29. January 2007
-22-
RD50
Defect Engineering of Silicon
 Influence the defect kinetics by incorporation of impurities or defects
 Best example: Oxygen
Initial idea: Incorporate Oxygen to getter radiation-induced vacancies
 prevent formation of Di-vacancy (V2) related deep acceptor levels
Observation: Higher oxygen content  less negative space charge
(less charged acceptors)
V2 in
clusters
 One possible mechanism: V2O is a deep acceptor
O
VO (not harmful at room temperature)
VO
V2O (negative space charge)
Ec
VO
V
V2O(?)
EV
Michael Moll – Lausanne, 29. January 2007
-23-
Spectacular Improvement of
g-irradiation tolerance
RD50
Depletion Voltage
250
CA: <111> STFZ
CB: <111> DOFZ 24 h
CC: <111>DOFZ 48 h
CD: <111> DOFZ 72 h
CE: <100> STFZ
CF: <100> DOFZ 24h
CG: <100> DOFZ 48h
CH: <100> DOFZ 72h
30
20
100
10
50
0
0
1500
1000
Irev [nA]
150
40
Neff [1011 cm-3]
Vdep [V]
200
Leakage Current
CA: <111> STFZ
CB: <111> DOFZ 24h
CC: <111> DOFZ 48h
CD: <111> DOFZ 78h
CE: <100> STFZ
CF: <100> DOFZ 24h
CG: <100> DOFZ 48h
CH: <100> DOFZ 72h
500
(a)
200
400
600
800
(b)
0
1000
0
0
(b)
200
400
600
800
1000
dose [Mrad]
dose [Mrad]
 No type inversion for oxygen enriched silicon!
 Slight increase of positive space charge
(due to Thermal Donor generation?)
 Leakage increase not linear and depending on
oxygen concentration
[E.Fretwurst et al. 1st RD50 Workshop]
See also:
- Z.Li et al. [NIMA461(2001)126]
- Z.Li et al. [1st RD50 Workshop]
Michael Moll – Lausanne, 29. January 2007
-24-
of microscopic defects
RD50 Characterization
- g and proton irradiated silicon detectors 
2003: Major breakthrough on g-irradiated samples
 For the first time macroscopic changes of the depletion voltage and leakage current
can be explained by electrical properties of measured defects ! [APL, 82, 2169, March 2003]

since 2004: Big steps in understanding the improved radiation tolerance of
oxygen enriched and epitaxial silicon after proton irradiation
[I.Pintilie, RESMDD, Oct.2004]
Levels responsible for depletion voltage
changes after proton irradiation:
Almost independent of oxygen content:
 Donor removal
“Cluster damage”  negative charge
Influenced by initial oxygen content:
 I–defect: deep acceptor level at EC-0.54eV
(good candidate for the V2O defect)
 negative charge
Influenced by initial oxygen dimer content (?):
 BD-defect: bistable shallow thermal donor
(formed via oxygen dimers O2i)
 positive charge
BD-defect
I-defect
Michael Moll – Lausanne, 29. January 2007
-25-
Oxygen enriched silicon – DOFZ
RD50
- proton irradiation -
• DOFZ (Diffusion Oxygenated Float Zone Silicon)
 1982 First oxygen diffusion tests on FZ [Brotherton et al. J.Appl.Phys.,Vol.53, No.8.,5720]
 1995 First tests on detector grade silicon [Z.Li et al. IEEE TNS Vol.42,No.4,219]
 1999 Introduced to the HEP community by RD48 (ROSE)
Standard
300
4
Oxygenated
2
0
0
400
200
100
1
2
3
4
24 GeV/c proton [10 cm ]
14
-2
[RD48-NIMA 465(2001) 60]
5
10
9
8
7
6
5
4
3
2
1
0
700
1 5 K  cm < 1 1 1 > - stan d ard
600
1 5 K  c m < 1 1 1 > - o x y g en ate d
500
400
300
200
Vdep [V] (300 mm)
500
6
http://cern.ch/rd48
|Neff| [1012cm-3]
8
600
Vdep [V] (300 mm)
|Neff| [1012cm-3]
10
Carbonated
RD48
Later systematic tests reveal strong
variations with no clear dependence
on oxygen content
First tests in 1999 show clear
advantage of oxygenation
Carbon-enriched (P503)
Standard (P51)
O-diffusion 24 hours (P52)
O-diffusion 48 hours (P54)
O-diffusion 72 hours (P56)
ROSE
100
[M .M o ll - N IM A 5 1 1 (2 0 0 3 ) 9 7 ]
0
2
4
6
 2 4 G e V /c p ro to n [1 0
8
14
10
-2
cm ]
However, only non-oxygenated
diodes show a “bad” behavior.
Michael Moll – Lausanne, 29. January 2007
-26-
RD50
Silicon Growth Processes
 Floating Zone Silicon (FZ)
Poly silicon
 Czochralski Silicon (CZ)
 The growth method
used by the IC industry.
 Difficult to produce
very high resistivity
RF Heating coil
Single crystal silicon
Float Zone Growth
 Basically all silicon detectors made
out of high resistivety FZ silicon
Czochralski Growth
 Epitaxial Silicon (EPI)
 Chemical-Vapor Deposition (CVD) of Si
 up to 150 mm thick layers produced
 growth rate about 1mm/min
Michael Moll – Lausanne, 29. January 2007
-27-
RD50
Oxygen concentration in FZ, CZ and EPI
Epitaxial silicon
DOFZ and CZ silicon
EPI
layer
5
1018
5
Cz as grown
1017
1017
5
5
16
10
5
0
DOFZ 72h/1150oC
DOFZ 48h/1150oC
DOFZ 24h/1150oC
50
100
[M.Moll]
150
depth [mm]
200
250
1016
5
 CZ: high Oi (oxygen) and O2i (oxygen dimer)
concentration (homogeneous)
 CZ: formation of Thermal Donors possible !
75 mu
50 mu
5
5
1018
CZ substrate
25 mu
Data: G.Lindstroem et al.
O-concentration [1/cm3]
O-concentration [cm-3]
 DOFZ: inhomogeneous oxygen distribution
 DOFZ: oxygen content increasing with time
at high temperature
1018
5
1017
5
1016
5
0
SIMS 25 mm
SIMS 50 mm
SIMS 75 mm
simulation 25 mm
simulation 50 mm
simulation 75mm
[G.Lindström et al.,10th European Symposium on
Semiconductor Detectors, 12-16 June 2005]
10 20 30 40 50 60 70 80 90 100
Depth [mm]
 EPI: Oi and O2i (?) diffusion from substrate
into epi-layer during production
 EPI: in-homogeneous oxygen distribution
Michael Moll – Lausanne, 29. January 2007
-28-
RD50
standard
for
particle
detectors
Silicon Materials under Investigation by RD50
Material
Symbol
Standard FZ (n- and p-type)
Diffusion oxygenated FZ (n- and p-type)
used for
LHC
Pixel
detectors
“new”
material
Magnetic Czochralski Si, Okmetic, Finland
(n- and p-type)
Czochralski Si, Sumitomo, Japan (n-type)
Epitaxial layers on Cz-substrates, ITME, Poland
(n- and p-type, 25, 50, 75, 150 mm thick)
Diffusion oxygenated Epitaxial layers on CZ
 (cm) [Oi] (cm-3)
FZ
1–710 3
< 51016
DOFZ
1–710 3
~ 1–21017
MCz
~ 110 3
~ 51017
Cz
~ 110 3
~ 8-91017
EPI
50 – 400
< 11017
EPI–DO
50 – 100
~ 71017
 DOFZ silicon
- Enriched with oxygen on wafer level, inhomogeneous distribution of oxygen
 CZ/MCZ silicon
- high Oi (oxygen) and O2i (oxygen dimer) concentration (homogeneous)
- formation of shallow Thermal Donors possible
 Epi silicon
- high Oi , O2i content due to out-diffusion from the CZ substrate (inhomogeneous)
- thin layers: high doping possible (low starting resistivity)
- as EPI, however additional Oi diffused reaching homogeneous Oi content
 Epi-Do silicon
Michael Moll – Lausanne, 29. January 2007
-29-
RD50
Standard FZ, DOFZ, Cz and MCz Silicon
800
• type inversion at ~ 21013 p/cm2
• strong Neff increase at high fluence
 Oxygenated FZ (DOFZ)
• type inversion at ~ 21013 p/cm2
• reduced Neff increase at high fluence
 CZ silicon and MCZ silicon
600
12
FZ
<111>
DOFZ <111> (72 h 11500C)
MCZ <100>
CZ <100> (TD killed)
10
8
400
6
4
200
|Neff| [1012 cm-3]
 Standard FZ silicon
Vdep (300mm) [V]
24 GeV/c proton irradiation
2
0
0
2
4
6
8
10
0
proton fluence [1014 cm-2]
 no type inversion in the overall fluence range (verified by TCT measurements)
(verified for CZ silicon by TCT measurements, preliminary result for MCZ silicon)
 donor generation overcompensates acceptor generation in high fluence range
 Common to all materials (after hadron irradiation):
 reverse current increase
 increase of trapping (electrons and holes) within ~ 20%
Michael Moll – Lausanne, 29. January 2007
-30-
EPI Devices – Irradiation experiments
RD50
 Epitaxial silicon
G.Lindström et al.,10th European Symposium on Semiconductor Detectors, 12-16 June 2005
G.Kramberger et al., Hamburg RD50 Workshop, August 2006
 Layer thickness: 25, 50, 75 mm (resistivity: ~ 50 cm); 150 mm (resistivity: ~ 400 cm)
 Oxygen: [O]  91016cm-3; Oxygen dimers (detected via IO2-defect formation)
12000
105V
(25mm)
Neff(t0) [cm-3]
2.1014
25 mm, 80 oC
50 mm, 80 oC
75 mm, 80 oC
230V
(50mm)
1014
0
0
320V
(75mm)
2.1015
4.1015 6.1015
eq [cm-2]
8.1015
150 mm - neutron irradiated
75 mm - proton irradiated
75 mm - neutron irradiated
50 mm - neutron irradiated
50 mm - proton irradiated
10000
Signal [e]
23 GeV protons
1016
 Only little change in depletion voltage
 No type inversion up to ~ 1016 p/cm2 and ~ 1016 n/cm2
high electric field will stay at front electrode!
reverse annealing will decreases depletion voltage!
 Explanation: introduction of shallow donors is bigger
than generation of deep acceptors
8000
6000
4000
2000
[Data: G.Kramberger et al., Hamburg RD50 Workshop, August 2006]
0
0
[M.Moll]
20 40 60 80 100
eq [1014 cm-2]
 CCE (Sr90 source, 25ns shaping):
 6400 e (150 mm; 2x1015 n/cm-2)
 3300 e (75mm; 8x1015 n/cm-2)
 2300 e (50mm; 8x1015 n/cm-2)
Michael Moll – Lausanne, 29. January 2007
-31-
RD50
Advantage of non-inverting material
p-in-n detectors (schematic figures!)
Fully depleted detector
(non – irradiated):
Michael Moll – Lausanne, 29. January 2007
-32-
RD50
Advantage of non-inverting material
p-in-n detectors (schematic figures!)
Be careful, this is a very schematic
explanation, reality is more complex !
Fully depleted detector
(non – irradiated):
heavy irradiation
inverted
non inverted
inverted to “p-type”, under-depleted:
non-inverted, under-depleted:
• Charge spread – degraded resolution
•Limited loss in CCE
• Charge loss – reduced CCE
•Less degradation with under-depletion
Michael Moll – Lausanne, 29. January 2007
-33-
RD50
50 mm thick silicon detectors:
- Epitaxial silicon (50cm on CZ substrate, ITME & CiS)
- Thin FZ silicon (4Kcm, MPI Munich, wafer bonding technique)
Vdep [V]
200
Ta=80oC
ta=8 min
150
100
50
0
0
EPI (ITME), 50mm
FZ (MPI), 50mm
1.4
1.2
1.0
0.8
0.6
0.4
0.2
150
Ta=80oC
EPI (ITME), 9.6.1014 p/cm2
100
Vfd [V]
250
|Neff| [1014 cm-3]

Epitaxial silicon - Annealing
50
FZ (MPI), 1.7.1015 p/cm2
[E.Fretwurst et al., Hamburg]
20
40
60
80 100
proton fluence [1014 cm-2]
0 0
10
101
102
103
104
105
annealing time [min]
[E.Fretwurst et al.,RESMDD - October 2004]


Thin FZ silicon: Type inverted, increase of depletion voltage with time
Epitaxial silicon: No type inversion, decrease of depletion voltage with time
 No need for low temperature during maintenance of SLHC detectors!
Michael Moll – Lausanne, 29. January 2007
-34-
RD50
Property
Eg [eV]
Ebreakdown [V/cm]
me [cm2/Vs]
mh [cm2/Vs]
vsat [cm/s]
Z
r
e-h energy [eV]
Density [g/cm3]
Displacem. [eV]
New Materials: Epitaxial SiC
“A material between Silicon and Diamond”
Diamond
5.5
107
1800
1200
2.2·107
6
5.7
13
3.515
43
R&D on diamond detectors:
RD42 – Collaboration
http://cern.ch/rd42/
GaN
3.39
4·106
1000
30
31/7
9.6
8.9
6.15
15
4H SiC
3.3
2.2·106
800
115
2·107
14/6
9.7
7.6-8.4
3.22
25
Si
1.12
3·105
1450
450
0.8·107
14
11.9
3.6
2.33
13-20
 Wide bandgap (3.3eV)
 lower leakage current
than silicon
 Signal:
Diamond 36 e/mm
SiC
51 e/mm
Si
89 e/mm
 more charge than
diamond
 Higher displacement
threshold than silicon
 radiation harder than
silicon (?)
Michael Moll – Lausanne, 29. January 2007
-35-
RD50 SiC: CCE after neutron irradiation
 CCE before irradiation




material produced by CREE
55 mm thick layer
neutron irradiated samples
tested with b particles
 Conclusion:
 SiC is less radiation tolerant than
expected
 Consequence:
-
 CCE after irradiation (example)
Collected Charge ( e )
 100 % with  particles and MIPS
3000
@ 950 V
2000
1000
Before irradiation
0
0.1
1E14
1E15
2
1E16
Fluence ((1MeV) n/cm )
[F.Moscatelli, Bologna, December 2006]
 RD50 will stop working on this topic
Michael Moll – Lausanne, 29. January 2007
-36-
RD50
Outline
 Motivation to develop radiation harder detectors
 Introduction to the RD50 collaboration
 Part I: Radiation Damage in Silicon Detectors (A very brief review)


Microscopic defects
Macroscopic damage
(changes in bulk material)
(changes in detector properties)
 Part II: RD50 - Approaches to obtain radiation hard sensors


Material Engineering
Device Engineering
 Summary and preliminary conclusion
Michael Moll – Lausanne, 29. January 2007
-37-
RD50
Device engineering
p-in-n versus n-in-p detectors
p-type silicon after high fluences:
n-type silicon after high fluences:
p+on-n
n+on-p
p-on-n silicon, under-depleted:
n-on-p silicon, under-depleted:
• Charge spread – degraded resolution
•Limited loss in CCE
• Charge loss – reduced CCE
•Less degradation with under-depletion
•Collect electrons (fast)
Be careful, this is a very schematic explanation,
reality is more complex !
Michael Moll – Lausanne, 29. January 2007
-38-
RD50
n-in-p microstrip detectors
n-in-p: - no type inversion, high electric field stays on structured side
- collection of electrons
n-in-p microstrip detectors (280mm) on p-type FZ silicon time [days at 20oC]
500
1000
1500
2000
Detectors read-out with 40MHz
20 0
25
24 GeV/c p irradiation
20
CCE ~ 6500 e (30%)
after 7.5 1015 p cm-2 at
900V
15
10
5
[Data: G.Casse et al., NIMA535(2004) 362]
0
0
[M.Moll]
2
4
6
8
10
fluence [1015cm-2]
CCE (103 electrons)
CCE (103 electrons)


18
16
14
12
10
8
6
4
2
0
2500
800 V
1.1 x 1015cm-2
500 V
3.5 x 1015cm-2 (500 V)
7.5 x 1015cm-2 (700 V)
[Data: G.Casse et al., to be published in NIMA]
M.Moll
0
100
200
300
400
o
time at 80 C[min]
500
 no reverse annealing visible in the CCE measurement !
e.g. for 7.5  1015 p/cm2 increase of Vdep from
Vdep~ 2800V to Vdep > 12000V is expected !
Michael Moll – Lausanne, 29. January 2007
-39-
Introduced by: S.I. Parker et
al., NIMA 395 (1997) 328
3D detector - concepts
 “3D” electrodes: - narrow columns along detector thickness,
- diameter: 10mm, distance: 50 - 100mm
 Lateral depletion: - lower depletion voltage needed
- thicker detectors possible
- fast signal
- radiation hard
n-columns
p-columns
PLANAR
3D
p+
p+
n
50 mm
- - -++
+
+ +
p+
+
300 mm
RD50
- ++
+ +
+
wafer surface
n-type substrate
Michael Moll – Lausanne, 29. January 2007
-40-
RD50
3D detector - concepts
 “3D” electrodes: - narrow columns along detector thickness,
- diameter: 10mm, distance: 50 - 100mm
n-columns
 Lateral depletion: - lower depletion voltage needed
- thicker detectors possible
- fast signal
- radiation hard
p-columns
wafer surface
n-type substrate
 Simplified 3D architecture
 n+ columns in p-type substrate, p+ backplane
 operation similar to standard 3D detector
 Simulations performed
 Fabrication:
 IRST(Italy), CNM Barcelona
metal strip
hole
[C. Piemonte et al.,
NIM A541 (2005) 441]
hole
 Simplified process
 hole etching and doping only done once
 no wafer bonding technology needed
Hole depth 120-150mm
Hole diameter ~10mm
C.Piemonte et al., STD06, September 2006
 First CCE tests under way
Michael Moll – Lausanne, 29. January 2007
-41-
RD50
Comparison of measured collected charge on different
radiation-hard materials and devices
 In the following: Comparison of collected charge as published in literature
 Be careful: Values obtained partly under different conditions




irradiation
temperature of measurement
electronics used (shaping time, noise)
type of device – strip detectors or pad detectors
 This comparison gives only an indication of which material/technology could be used,
to be more specific, the exact application should be looked at!
 Remember: The obtained signal has still to be compared to the noise
 Acknowledgements:
 Recent data collections:
Mara Bruzzi (Hiroschima conference 2006)
Cinzia Da Via (Vertex conference 2006)
Michael Moll – Lausanne, 29. January 2007
-42-
RD50
Comparison of measured collected charge on different
radiation-hard materials and devices
4000
signal [electrons]
SiC, n-type, 55 mm, (RT, 2.5ms) [Moscatelli et al. 2006]
3000
sample:
irradiation:
measurement:
analysis:
4H-SiC layer, 55mm, pad detector
24 GeV/c protons
Sr-90 source, 2.5 ms shaping, room temperature
mean values presented
2000
1000
0
0
20
40
60
80
14
-2
eq [10 cm ]
100
120
[M.Moll 2007]
Michael Moll – Lausanne, 29. January 2007
-43-
RD50
Comparison of measured collected charge on different
radiation-hard materials and devices
signal [electrons]
6000
pCVD-Diamond, 500mm, (RT, ms) [RD42 2002]
SiC, n-type, 55 mm, (RT, 2.5ms) [Moscatelli et al. 2006]
5000
sample:
irradiation:
measurement:
analysis:
4000
polycrystal, 500mm thick, strip
24 GeV/c protons
testbeam, ms shaping
most probable values
3000
2000
1000
0
0
20
40
60
80
14
-2
eq [10 cm ]
100
120
[M.Moll 2007]
Michael Moll – Lausanne, 29. January 2007
-44-
signal [electrons]
RD50
Comparison of measured collected charge on different
radiation-hard materials and devices
7000
6000
5000
4000
3000
2000
1000
0
0
pCVD-Diamond, 500mm, (RT, ms) [RD42 2005]
pCVD-Diamond, 500mm, (RT, ms) [RD42 2002]
SiC, n-type, 55 mm, (RT, 2.5ms) [Moscatelli et al. 2006]
sample:
irradiation:
measurement:
analysis:
20
polycrystal, 500mm thick, strip
24 GeV/c protons
testbeam, ms shaping
most probable values
40
60
80
14
-2
eq [10 cm ]
100
120
[M.Moll 2007]
Michael Moll – Lausanne, 29. January 2007
-45-
RD50
Comparison of measured collected charge on different
radiation-hard materials and devices
signal [electrons]
10000
pCVD-Diamond, 500mm, (RT, ms) [RD42 2006]
pCVD-Diamond, 500mm, (RT, ms) [RD42 2005]
pCVD-Diamond, 500mm, (RT, ms) [RD42 2002]
SiC, n-type, 55 mm, (RT, 2.5ms) [Moscatelli et al. 2006]
8000
sample:
irradiation:
measurement:
analysis:
6000
polycrystal, 500mm thick, strip
24 GeV/c protons
testbeam, ms shaping
most probable values
Diamond quality increasing
[2000-2006]
4000
2000
0
0
20
40
60
80
14
-2
eq [10 cm ]
100
120
[M.Moll 2007]
Michael Moll – Lausanne, 29. January 2007
-46-
RD50
Comparison of measured collected charge on different
radiation-hard materials and devices
signal [electrons]
10000
pCVD-Diamond, 500mm, (RT, ms) [RD42 2002-2006] (scaled)
SiC, n-type, 55 mm, (RT, 2.5ms) [Moscatelli et al. 2006]
8000
sample:
irradiation:
measurement:
analysis:
6000
polycrystal, 500mm thick, strip
24 GeV/c protons
testbeam, ms shaping
most probable values
4000
2000
0
0
20
40
60
80
14
-2
eq [10 cm ]
100
120
[M.Moll 2007]
Michael Moll – Lausanne, 29. January 2007
-47-
RD50
Comparison of measured collected charge on different
radiation-hard materials and devices
signal [electrons]
12000
10000
pCVD-Diamond, 500mm, (RT, ms), strip, [RD42 2002-2006] (scaled)
n-epi Si, 150 mm, (-30oC, 25ns), pad [Kramberger 2006]
n-epi Si, 75 mm, (-30oC, 25ns), pad [Kramberger 2006]
SiC, n-type, 55 mm, (RT, 2.5ms), pad [Moscatelli et al. 2006]
8000
6000
4000
2000
0
0
20
40
60
80
14
-2
eq [10 cm ]
100
120
[M.Moll 2007]
Michael Moll – Lausanne, 29. January 2007
-48-
RD50
Comparison of measured collected charge on different
radiation-hard materials and devices
signal [electrons]
25000
p-FZ Si, 280 mm, (-30oC, 25ns), strip [Casse 2004]
p-MCZ Si, 300 mm, (-30oC, ms), pad [Bruzzi 2006]
n-epi Si, 150 mm, (-30oC, 25ns), pad [Kramberger 2006]
n-epi Si, 75 mm, (-30oC, 25ns), pad [Kramberger 2006]
pCVD-Diamond, 500mm, (RT, ms), strip, [RD42 2002-2006] (scaled)
SiC, n-type, 55 mm, (RT, 2.5ms), pad [Moscatelli et al. 2006]
20000
15000
10000
5000
0
0
20
40
60
80
14
-2
eq [10 cm ]
100
120
[M.Moll 2007]
Michael Moll – Lausanne, 29. January 2007
-49-
RD50
Comparison of measured collected charge on different
radiation-hard materials and devices
signal [electrons]
25000
3D FZ Si, 235 mm, (laser injection, scaled!), pad [Da Via 2006]
p-FZ Si, 280 mm, (-30oC, 25ns), strip [Casse 2004]
p-MCZ Si, 300 mm, (-30oC, ms), pad [Bruzzi 2006]
n-epi Si, 150 mm, (-30oC, 25ns), pad [Kramberger 2006]
n-epi Si, 75 mm, (-30oC, 25ns), pad [Kramberger 2006]
pCVD-Diamond, 500mm, (RT, ms), strip, [RD42 2002-2006] (scaled)
SiC, n-type, 55 mm, (RT, 2.5ms), pad [Moscatelli et al. 2006]
20000
15000
10000
5000
0
0
20
40
60
80
14
-2
eq [10 cm ]
100
120
[M.Moll 2007]
Michael Moll – Lausanne, 29. January 2007
-50-
RD50
Comparison of measured collected charge on different
radiation-hard materials and devices
signal [electrons]
25000
3D FZ Si, 235 mm, (laser injection, scaled!), pad [Da Via 2006]
p-FZ Si, 280 mm, (-30oC, 25ns), strip [Casse 2004]
p-MCZ Si, 300 mm, (-30oC, ms), pad [Bruzzi 2006]
n-epi Si, 150 mm, (-30oC, 25ns), pad [Kramberger 2006]
n-epi Si, 75 mm, (-30oC, 25ns), pad [Kramberger 2006]
20000
sCVD-Diamond, 770mm, (RT, ms), [RD42 2006] (preliminary data, scaled)
pCVD-Diamond, 500mm, (RT, ms), strip, [RD42 2002-2006] (scaled)
15000
SiC, n-type, 55 mm, (RT, 2.5ms), pad [Moscatelli et al. 2006]
10000
5000
0
0
20
40
60
80
14
-2
eq [10 cm ]
100
120
[M.Moll 2007]
Michael Moll – Lausanne, 29. January 2007
-51-
RD50
Signal Charge / Threshold
 Do not forget: The signal has still to be compared to the noise (the threshold)
Michael Moll – Lausanne, 29. January 2007
-52-
RD50
Summary – Radiation Damage
 Radiation Damage in Silicon Detectors
 Change of Depletion Voltage (type inversion, reverse annealing, …)
(can be influenced by defect engineering!)
 Increase of Leakage Current (same for all silicon materials)
 Increase of Charge Trapping (same for all silicon materials)
Signal to Noise ratio is quantity to watch (material + geometry + electronics)
 Microscopic defects
 Good understanding of damage after g-irradiation (point defects)
 Damage after hadron damage still to be better understood (cluster defects)
 CERN-RD50 collaboration working on:
 Material Engineering (Silicon: DOFZ, CZ, EPI, other impurities,. ) (Diamond)
 Device Engineering (3D and thin detectors, n-in-p, n-in-n, …)
 To obtain ultra radiation hard sensors a combination of material and
device engineering approaches depending on radiation environment,
application and available readout electronics will be best solution
Michael Moll – Lausanne, 29. January 2007
-53-
RD50

Summary – Detectors for SLHC
At fluences up to 1015cm-2 (Outer layers of SLHC detector) the change of the depletion
voltage and the large area to be covered by detectors are major problems.
 CZ silicon detectors could be a cost-effective radiation hard solution
no type inversion (to be confirmed), use cost effective p-in-n technology
 oxygenated p-type silicon microstrip detectors show very encouraging results:
CCE  6500 e; eq= 41015 cm-2, 300mm

At the fluence of 1016cm-2 (Innermost layers of SLHC detector) the active thickness of
any silicon material is significantly reduced due to trapping.
The two most promising options besides regular replacement of sensors are:
Thin/EPI detectors : drawback: radiation hard electronics for low signals needed
(e.g. 2300e at eq 8x1015cm-2, 50mm EPI)
3D detectors

: looks very promising,
drawback: technology has to be optimized
SiC and GaN have been characterized and abandoned by RD50.
Further information: http://cern.ch/rd50/
Michael Moll – Lausanne, 29. January 2007
-54-
RD50
Comparison of measured collected charge on different
radiation-hard materials and devices
Line to guide the eye for planar devices
25000
p FZ Si 280mm; 25ns; -30°C [1]
p-MCz Si 300mm;0.2-2.5ms; -30°C [2]
n EPI Si 75mm; 25ns; -30°C [3]
n EPI Si 150mm; 25ns; -30°C [3]
sCVD Diam 770mm; 25ns; +20°C [4]
pCVD Diam 300mm; 25ns; +20°C [4]
n EPI SiC 55mm; 2.5ms; +20°C [5]
3D FZ Si 235mm [6] 160V
25000
20000
15000
10000
5000
0
pixels
strips
15000
# electrons
# electrons
20000
14
10
[1] G. Casse et al. NIM A (2004)
[2] M. Bruzzi et al. STD06, September 2006
[3] G. Kramberger, RD50 Work. Prague 06
[4] W: Adam et al. NIM A (2006)
[5] F. Moscatelli RD50 Work.CERN 2005
[6] C. Da Vià, "Hiroshima" STD06
(charge induced by laser)
10000
5000 1015
10
-2
1 MeV n fluence [cm ]
16
M. Bruzzi, Presented at STD6 Hiroshima Conference,
Carmel, CA, September 2006
0
14
15 in partial depletion, collected
16
 Thick (300mm) p-type planar
detectors can operate
charge
10
10
10
15
-2
higher than 12000e up to 2x10 cm .
-2
Fluence
[cm
]
 Most charge at highest fluences collected with 3D detectors
 Silicon comparable or even better than diamond in terms of collected charge
(BUT: higher leakage current – cooling needed!)
Michael Moll – Lausanne, 29. January 2007
-55-
RD50
Comparison of measured collected charge on different
radiation-hard materials and devices
Michael Moll – Lausanne, 29. January 2007
-56-