Radiation effects on optoelectronic components and systems

1
Radiation effects on optoelectronic components
and systems

• Karl Gill
• CERN, CMS Experiment

2
Outline

• 1 Introduction
• 1.1 Technologies
• 1.2 LHC radiation environments
• 1.3 Review of radiation damage mechanisms
• 2 Radiation damage effects
• 2.1 Components
• 2.2 System implications

3
Optoelectronics

• Photonics - The technology of generating and
  harnessing light and other forms of radiant
  energy whose quantum unit is the photon
• definition from Photonics Magazine
• Applications
• Communication
• Imaging
• Sensing
• Information display

4
LHC Opto-applications

• Widespread use of optoelectronics and
  (fibre-)optics at LHC
• readout and control optical links
• monitoring and calibration
• alignment

5
Technologies

• Many device technologies and materials
• Transmitters - lasers, LEDs, (modulators),.
• Receivers - Photodiodes, CCDs, APD, .
• Passive components - fibres, lenses,.
• Switches - optocouplers
• Materials include Si, GaAs, InGaAs, InGaAsP,
  InP, SiO2

6
Active materials
Emitters
Detectors
Ref 1
7
COTS issues

• COTS in many LHC systems
• Benefit from industrial developments
• cheaper, reliable devices
• However COTS not made for LHC environment
• no guarantees of long-life at LHC
• validation testing of COTS is mandatory

8
Radiation damage overview
Radiation Environment
Interaction of radiation with material
Ionization
SEU
Displacement
Defect creation
Component Effects
Annealing
Effects at system level
9
Environments

• Optoelectronics already employed in variety of
  harsh radiation environments
• e.g. civil nuclear and space applications

Total dose (Gy)
1E8
Space (p,e)
Nuclear (g, also n)
1E6
1E4
1E2
1E0
1E-2
1E-2
1E0
1E2
1E4
1E6
Dose rate (Gy/hr)
Ref 2
10
LHC Radiation environments experiments

• e.g CMS neutrons

(courtesy M. Huhtinen)
11
LHC Radiation environments experiments

• e.g CMS photons

(courtesy M. Huhtinen)
12
Radiation damage mechanisms

• Displacement
• Ionization
• Transient
• also annealing

13
Displacement damage

• e.g. displacement cascade of 30keV Si recoil
• most NIEL in last 5kEV
• final cluster of defects
• 100Å size
• high defect density in crystal lattice

Ref 3
14
defects in band-gap

• can cause several effects
• depends upon
• position in band gap
• type of defect
• donor/acceptor
• single or multiple levels

Ref 4
15
generation-recombination at defects
Emission and capture transitions via defect state
in band-gap
Ec
ET
Ev

• Carrier lifetime
• generally, most damaging defects near centre of
  band-gap

(for defect at mid-gap)
Ref 1
16
Lifetime degradation

• Radiation damage introduces more defects
• K is damage factor - depends on particle type,
  energy
• F is fluence
• lifetime therefore decreases with fluence (or
  dose)

(assume linear)
(Messenger eqn)
or
Ref 5
17
Non-ionizing energy loss

• dependence upon particle type and energy

• Damage factors (K) related to NIEL

Ref 6
18
Ionization damage
Ref 4
19
Ionization damage effects

• charge trapped in oxide or at interface

Ref 7
20
Defects in glasses

• defects (colour centres) created by irradiation
  
• bonds broken by ionisation/displacement

• defects absorb/scatter incident photons

Ref 8
21
Transients

• Single-event upsets (SEU)
• passage of energetic particlecauses ionization
• primary from charged particle or heavy ion
• secondary ionisation from recoiling nucleus
• Variety of effects
• corruption of individual bits
• can kill a component!

22
Transient ionization

• ionizing energy deposition (e.g. for Si, 40MeV
  p)

• large direct ionization peak
• recoils also contribute

• Ionization pulses cause SEU

Ref 9
23
Summary of issues at LHC

• Many types of optical and optoelectronic
  components in LHC systems
• various radiation environments TK----gtcavern
• Spectrum of damage effects from total dose,
  fluence, SEU
• Next step to look at effects in more detail
• concentrate on LHC RD and other relevant sources

24
Outline

• 1 Introduction
• 1.1 Technologies
• 1.2 LHC radiation environments
• 1.3 Review of radiation damage mechanisms
• 2 Radiation damage effects
• 2.1 Components
• 2.2 System implications

25
Component sensitivity
Danger!!
Beware
Probably OK!
26
Transmitters

• LEDs
• Edge-emitting lasers
• Surface emitting lasers
• (modulators)

Increasing rad-hardness
27
Basic LED structure

• P-n diode
• Light from spontaneous emission (recombination)

Ref 10
28
pigtailed LED

• Multi-mode fibre pigtail
• butt-coupled for optimum efficiency

Ref 10
29
LED characteristic

• LED Light-current and current-voltage
  before/after irradiation

Ref 11
30
LED damage 1

• damage vs fluence

• Significant damage at low particle fluences

Ref 12
31
LED damage 2

• damage vs fluence - e.g. different LED types

• Biasing increases resistance
• enhances annealing

Ref 12
32
LED damage 3

• damage vs fluence (ATLAS SCT)

Ref 13
33
Transmitters

• LEDs
• Edge-emitting lasers
• Surface emitting lasers

Increasing rad-hardness
34
Edge-emitting laser

• stripe geometry
• cleaved ends form Fabry-Peror optical cavity

Ref 14
35
DCPBH-MQW

• e.g. double-channel-planar-buried-heterostructure
  type

Ref 15
36
Laser characteristics

• 1310nm DCPBH-MQW before/after pion irradiation

Main effects Ithr increase Eff decrease Rs
same Vthr increase

• Ep330MeV, Fp2x1014p/cm2

Ref 16,17
37
Damage picture

• non-radiative recombination at defects
• competes with laser recombination

38
Different vendors

• Ithr and Eff changes vs neutron fluence

• similar effects in all 1310nm InGaAsP lasers

Ref 17
39
Different particles

• DIthr vs fluence of different particles

Ref 16

• Damage correlated to NIEL? probably.

40
Annealing (temperature)

• damage anneals (faster at higher temperature)

Ref 18

• Note tracker operating at -10C

41
Annealing (current)

• damage anneals faster at higher forward bias

Ref 18

• recombination enhanced annealing (?)

42
Reliability

• irradiated device lifetime gt 10 years??
• Ageing test at 80C

• No additional degradation in irradiated lasers
• acc. Factor 400 relative to -10C operation
• lifetime gtgt10years

Ref 19
43
Other EEL parameters

• includes
• wavelength
• facet reflectivity
• beam profile
• series resistance
• turn-on time
• NOT affected up to 100kGy or 1015n/cm2 (1MeV)

44
Transmitters

• LEDs
• Edge-emitting lasers
• Surface emitting lasers

Increasing rad-hardness
45
VCSEL structure

• surface emitting laser diode

• Very small active volume

Ref 15
46
VCSEL damage effects 1
Ref 20

• Similar damage effects as in edge-emitters
• smaller absolute changes - smaller device volume

47
VCSEL 2

• Damage vs fluence
• (6MeV n)
• Siemens devices

• 25 annealing at room T after irradiation

Ref 20
48
VCSEL 3

• ATLAS measurements

Ref 21

• Before after 2.9x1015n/cm2 (1MeV n) !!

49
VCSEL 4

• lifetime (reliability) of irradiated devices

Ref 22

• Equivalent to 3700 LHC-years at 20mA !

50
Transmitter damage summary
Displacement damage
Defects in band-gap
Non-radiative recombination
carrier lifetime degradation
Non-radiative
radiative
tNR
tR

• LED
• EEL
• VCSEL

g
Increasing rad-hardness

• Competition between radiative and non-radiative
  transitions

51
Components review

• Transmitters (lasers, LEDs, - total dose,
  fluence)
• Receivers (Photodiodes, APD, CCD)
• Passive Components (Fibres, lenses)
• Other components (Optocouplers)

52
Recievers

• total fluence effects
• p-i-n photodiode (InGaAs and Si)
• CCDs
• APDs see notes
• SEU effects in receivers

53
Optical detectors

• Many types of material used
• cover different wavelength spectra
• look at GaInAs and Si

Ref 1
54
p-i-n photodiode

• Basic structure

Ref 1
55
p-i-n operation

• Operation principles

Ref 1
56
InGaAs p-i-n characteristics

• Output current vs incident power
• InGaAs p-i-n -5V
• before/after 2x1014p/cm2

• Increase in Ileak
• decrease in Iphoto

Ref 17
57
p-i-n damage picture
trapping recombination
generation
leakage current
signal loss
58
Different vendors - leakage

• leakage current (InGaAs, 6MeV neutrons)

Ref 16

• similar damage over many (similar) devices

59
Different vendors - response

• Photocurrent (InGaAs, 6MeV neutrons)

• Significant differences in damage
• depends mainly if front or back-illuminated

Ref 16
60
Front-illum. vs back-illum.

• electron hole pairs created at InGaAs/InP
  interface in back-illuminated diodes

• holes must travel through InGaAs in back-illum.
  p-i-n
• holes travel less distance in front-illuminated
  p-i-n.

Defects acceptor type (good hole traps)
61
Different particles (leakage)

• leakage current (InGaAs, different particles, 20C)

Ref 17

• higher energy p, p more damaging than n

62
Different particles (response)

• different particles

Ref 17

• higher energy p, p more damaging than n

63
InGaAs p-i-n annealing

• After pion irradiation (room T, -5V)

• Leakage anneals a little
• No annealing of response

Ref 17
64
InGaAs p-i-n reliability

• irradiated device lifetime gt 10 years??
• Ageing test at 80C

• No additional degradation in irradiated p-i-ns
• lifetime gtgt10years

Ref 19
65
ATLAS Si p-i-n damage

• 35 loss of response
• Ileak 60nA (20C)
• rise time still lt 2ns

Ref 21
66
ATLAS Si p-i-n reliability

• ATLAS SCT Si p-i-n ageing

Ref 23

• No degradation, lifetime 2720years ! (90CL)

67
Recievers

• focus on total fluence effects
• InGaAs p-i-n photodiode
• Si p-i-n photodiode
• CCDs
• APDs
• then look at SEU in control link receivers

68
CCDs

• Basic structure and operation

Ref 24
69
CCD leakage

• leakage current increase

Ref 25
70
CCD leakage spikes

• variations in leakage density

Ref 25

• linked to small size of pixels

71
CCD RTS

• Random telegraph unstable switching of leakage
  current

Ref 25
72
CCD CTI

• charge transfer inefficiency

Ref 25
73
detector bulk-damage summary
trapping recombination
generation
leakage current
signal loss
74
APDs

• Basic structure

Ref 1
75
APD damage (gain)

• Effect of irradiation on gain (CMS ECAL)

F 2x1013 (1MeV equivalent) n/cm2
Ref 26
76
APD damage (quant. eff.)

• Damage to quantum efficiency (CMS ECAL)

F 2x1013 (1MeV equivalent) n/cm2
Ref 26
77
Recievers

• focus on total fluence effects
• InGaAs p-i-n photodiode
• Si p-i-n photodiode
• CCDs
• APDs
• then look at SEU in control link receivers

78
PD SEU

• photodiodes sensitive to SEU

Ref 27

• strong dependence upon particle type and angle

79
PD SEU bit-errors

• photodiodes sensitive to SEU

Ref 27

• Can change 0 to a 1 if signal above threshold
  at the time of decision

80
PD BER test

• test setup

Ref 28

• Photodiode and receiver chip irradiated

81
PD BER 1

• BER with 59MeV protons in InGaAs p-i-n (D80mm)
• 90 angle 1-100mW optical power

• large BER up to high power
• long ionizing track in active layer of p-i-n
• direct ionization effect

Ref 28
82
PD BER 2

• BER with 59MeV protons (cont.)
• smaller angles

• lower BER at lower angles
• shorter ionizing track in active volume of p-i-n
• nuclear recoil effect

Ref 28
83
PD BER 3

• energy deposition (e.g. for Si with 40MeV p)

• large direct ionization peak
• recoils contribute individual events with large
  energy deposition

Ref 9
84
PD BER 4

• compare BER for 59MeV p and 62MeV n0

• Neutrons give nuclear recoils
• same collision X-section as for protons
• sBERNerrors/F
• sBER(n) sBER(p)
• confirms nuclear recoil effect for p

Ref 28
85
Components review

• Transmitters (lasers, LEDs, - total dose,
  fluence)
• Receivers (Photodiodes, APD, CCD)
• Passive Components (Fibres, lenses)
• Other components (Optocouplers)

86
Defects in glasses

• defects (colour centres) created by irradiation
  
• bonds broken by ionisation/displacement

Ref 8

• defects absorb/scatter incident photons

87
Colour centres

• e.g. irradiated lenses
• collimated beam damage
• (different Ce concentration affects darkening)

courtesy D. Doyle (ESTEC) and A.Gusarov (SCK-CEN)
88
Fibre types

• MM
• short data links
• good coupling to VCSELs, LEDs
• ATLAS SCT, Larg
• CMS ECAL
• SM
• telecoms
• CMS Tracker

Ref 2,10
89
Effect of dopants/impurities in fibres

• Avoid phosphorus!
• (Note also strong wavelength dependence)

Ref 29
90
Radiation hardening

• some fibres become more resistant after high doses

• Defects passivated by mobile oxygen atoms

Ref 30
91
Optical bleaching

• damage dependence on light power in fibre

• Modern telecom fibres less sensitive

Ref 31
92
Fibre attenuation vs dose

• Gamma damage (CMS-TK COTS single-mode fibres)

• 1310nm

Ref 32
93
Fibre attenuation vs fluence

• Neutron damage (CMS TK)

• damage actually most likely due to gamma
  background

Ref 32
94
Fibre annealing

• damage recovers after irradiation (e.g. gamma)

• Damage therefore has dose-rate dependence

Ref 32
95
Integrated components

• lenses
• Ball-lenses often found in fibre-coupled packages
• Glass covers
• on TO-packages

Ref 10
96
Lens darkening
lens
LD
PD
fibre

• Output efficiency decreases if lenses or covers
  darkened
• (also loss of response in some packaged
  photodiodes)

Ref 33
97
Summary of damage in glass

• Main concern is attenuation
• many factors affect damage in glass
• impurities
• wavelength of light
• production process
• dose rate
• previous irradiation history
• temperature
• light power level
• Should test samples under application conditions

98
Components review

• Transmitters (lasers, LEDs, - total dose,
  fluence)
• Receivers (Photodiodes, APD, CCD)
• Passive Components (Fibres, lenses)
• Other components (Optocouplers)

99
Optocouplers

• total dose / fluence
• SEU
• data from COTS used in space applications
• Johnston et al., NSREC, RADECS

100
structures

• various types
• e.g. LED phototransistor

sandwich
lateral
Ref 34
101
P damage

• LED output degradation
• Photoresponsedecrease
• Note
• low fluence!

Ref 34
102
Gain photoresponse

• Photoresponse more important than decrease in
  transistor gain

Ref 34
103
Optocoupler SEU

• another type
• SEU from protons measured
• vs angle
• vs energy

Ref 9
104
SEU pulses

• Lid (LED) removed
• detector is most sensitive element
• 64MeV protons
• many pulses almost saturate

Ref 9
105
X-sect vs angle

• Strong anglular dependence
• pronounced at lower energies
• direct ionization responsible

Ref 9
106
SEU X-sect

• For protons incident from all directions

Ref 9
107
Component sensitivity
Danger!!
Beware
Probably OK!
108
Component effects summary
109
System Issues

• Depends on system!
• use CMS Tracker optical link as example
• system overview
• COTS validation procedure

110
e.g. Optical link technology
E.g. CMS Tracker optical links lasers
single-mode fibre array connectors
photodiodes

• Transmitter - 1310nm InGaAsP EEL
• Fibres and connectors - SM Ge-doped fibre
• Receivers - InGaAs p-i-n
• plus electronics

Ref 35,36
111
CMS Tracker readout and control links
Analogue Readout 50000 links _at_ 40MS/s
FED
Detector Hybrid
Tx Hybrid
96
Rx Hybrid
processing
MUX

A
buffering
APV
4
DAQ
21
D
amplifiers
12
12

C
pipelines
1281 MUX
PLL Delay
Timing
DCU
TTCRx


TTC
Digital Control 2000 links _at_40MHz
FEC
Control

64
4
TTCRx
CCU
CCU
8
processing
buffering
CCU
CCU
Back-End
Front-End
Ref 37
112
System specs

• Analogue readout links

• Last 2 columns filled in for each device type
  after testing

Ref 35
113
LHC Radiation environments Trackers

• (charged hadrons)

(courtesy M. Huhtinen)
114
LHC Radiation environments Trackers

• (neutrons)

(courtesy M. Huhtinen)
115
COTS

• Recall COTS not made for LHC environment
• for applications in radiation environments
• NO guarantees of
• rad-hardness
• reliability
• require validation of COTS
• develop test procedures relevant to application

116
E.g. COTS lasers for CMS Tracker

• 1-way InGaAsP EEL on Si-submount with lid

Ref 35
117
Validation procedures

• e.g. Lasers for analogue links

Highlighted Market survey validation tests
(in-system) lab tests
g irradiation
p irradiation
n irradiation
annealing
ageing
(in-system) lab tests
118
Radiation test system

• Test setup for in-situ measurements

• Similar for p-i-n and fibre studies

119
Validation (component rad-damage)

• L-I characteristic before/after irradiation

Irradiation 100kGy 60Co g 1015n/cm2 (0.8MeV)

• Increase in laser threshold, decrease in
  efficiency

120
Lab testing

• In-system test-bed

• measure threshold, gain, noise, linearity,
  rise-time (bandwidth)

Ref 38
121
Validation (in-system)

• Transfer characteristics

• gain decrease
• Increase in d.c. bias-point

122
Validation (in-system)

• Noise normalized to peak-signal

• Decrease in signal/noise
• gain loss
• more noise at higher currents

123
Validation (in-system)

• Linearity
• not much change

124
Validation (in-system)

• rise times

Dt
90
10
Dt 3.0ns before and after irradiation (limited
by receiver bandwidth)
125
validation summary

• laser validation for CMS TK analogue optical link
• validated radiation hardness of components
• validated system performance with irradiated
  lasers
• potentially sensitive system parameters include
• dynamic range
• signal/noise
• linearity
• bandwidth/settling time

126
System implications

• Allow compensation for damage effects
• threshold increases
• programmable d.c. offset bias
• efficiency loss
• (and variation in optical coupling at connectors)
• variable gain at transmitter
• therefore optimize dynamic range

127
Summary 1 - recall radiation damage overview
Radiation Environment
Interaction of radiation with material
Ionization
SEU
Displacement
Defect creation
Component Effects
Annealing
Effects at system level
128
Summary 2 - recall component effects
129
Conclusions - important issues to consider
What is the radiation environment?
What are the damage effects in components?
What are the implications at the system level?
What are the relevant validation
procedures (attention COTS!)
130
References - 1
1 Semiconductor Technology, S.M. Sze, published
by Wiley (1985) 2 RADECS (and NSREC) Technical
training courses. 3 Cluster damage for
example (a) J.C. Moreno-Marin et al., Nucl.
Instr. and Meth. B48, 404-407 (1990) (b) V.A.J.
Van Lint et al., IEEE Trans. Nucl. Sci., 19, 181
(1972) 4 J.E. Gover and J.R. Srour, Sandia Labs
Report SAND85-0776 (1985) 5 G.C. Messenger,
IEEE Trans. Nucl. Sci., 39, No. 3, 468-473
(1992) 6 Non-ionizing energy loss (a) A. Van
Ginnekin Fermilab Note FN-522 (1989) (b) G.P.
Summers et al., IEEE Trans. Nucl. Sci., 34, 1134
(1987) 7 Ionizing Radiation Damage Effects in
MOS Devices and Circuits, T. Ma and P.
Dressendorfer, published by Wiley
Interscience. 8 J. Robertson, Phil. Mag. B, 52,
No. 3, 371-377 (1985) 9 A.H. Johnston et al.,
IEEE Trans. Nucl. Sci. 46, No. 6, 1335-1341
(1999) 10 Optical Communication Systems (2nd
Edition), J. Gowar, Published by Prentice Hall,
1993. 11 B.H. Rose and C.E. Barne, J. App.
Phys., 53(3), 1772-1780 (1982). 12 A.H.
Johnston et al., IEEE Trans. Nucl. Sci. 46, No.
6, 1781-1789 (1999). 13 J. Beringer et al.,
ATLAS Internal Note INDET-NO-183 (1997) 14
Fundamentals of Optical Fiber Communications, W.
Van Etten and J. Van der Plaats, Published by
Prentice Hall, 1991. 15 Seminconductor Lasers
(2nd Edition), G.P. Agrawal and N.K. Dutta,
published by Van Nostrand Reinhold, 1993. 16 K.
Gill et al., Proceedings of 4th Workshop on
Electronics for LHC Experiments, London,
1998. 17 K. Gill et al., Proceedings of SPIE
3440, 89-99 (1998) 18 K. Gill et al., submitted
to Conference on Photonics for Space, SPIE Annual
Meeting 2000.
131
References - 2
19 K. Gill et al., Proceedings of 1999 RADECS
conference, Fontevraud, 1999. 20 K. Gill et
al., Internal Report. 21 D.G. Charlton et al.,
Nucl. Instr. And Meth. A443, 430-446 (2000) 22
ATLAS SCT VCSEL annealing and ageing
tests http//www.atlas.uni-wuppertal.de/optolink/
wuppertal/annealing.html 23 ATLAS SCT
photodiode ageing tests http//www.ep.ph.bham.ac.
uk/user/mahout/irradn/sctmeeting/jun99/ 24 V.
Radeka, Nucl. Instr. and Meth. A226209-218
(1984) 25 G.R. Hopkinson, IEEE Trans. Nucl.
Sci. 46, No. 6, 1790-1796 (1999). 26 K. Deiters
et al., Nucl. Instr. and Meth. A442 193-197
(2000) 27 P. Marshall et al., IEEE Trans. Nucl.
Sci., 43, No. 2, 645-653 (1996) 28 F. Faccio et
al, submitted to RADECS Workshop, 2000 29 E.J.
Friebele et al., App. Opt. 19, No. 17, 2910-2916
(1980) 30 D. Griscom, J. App. Phys., 77 (10)
5008-5013 (1995) 31 B.D. Evans, Proc. SPIE
1174, 20-26 (1989) 32 J. Troska, Proc. SPIE
3440, 112-119 (1998) 33 P. Marshall et al.,
IEEE Trans. Nucl. Sci., 39, No. 6, 1982-1989
(1992) 34 A.H. Johnston et al., IEEE Trans.
Nucl. Sci. 43, No. 6, 3167-3173 (1996). 35 CMS
Tracker Optical Links www site
http//cern.web.cern.ch/CERN/Divisions/ECP/CME/Op
ticalLinks/ 36 F. Vasey et al, IEEE Trans.
Nucl. Sci., 45, No. 3, 331-337, (1998) 37 CMS
Tracker Technical Design Report., CERN LHCC 98-6,
(1998) 38 F. Jensen et al, CMS Technical Note,
CMS Note 99-074 (1999)