NASA Technical Reports Server (NTRS) 20050156605: Single Particle Damage Events in Candidate Star Camera Sensors

0.29 MB·English

by  NASA Technical Reports Server (NTRS)

Single particle damage events in
candidate star camera sensors
Pauli Marshall', Cheryl Marshall2, Elizabeth Polidan3,
Augustyn Wacyznski3 and Scott Johnson3
'NASA GSFC Consultant,
2NASA GSFC
3Global Science and Technology, NASA-GSFC
Phone: [phone]
[email protected]
Outline
Solar and trapped protons and shielding
- Energy range and mechanisms
Proton interactions in Si
- -
Ionization (Total Ionizing Dose TID)
- Displacement damage (trapping and increased leakage)
- -
Single event transient effects important but not covered here
Displacement damage effects in detectors
-
Dark current dependence on temperature
-
c'Universal" damage constant for "bulk" dark current
- Damage distributions, hot pixels & random telegraph noise
- Electric field enhancement of leakage currents
Hot pixel mechanisms, introduction rates, and annealing
- Onsrbit HST measurements
- Laboratory measurements (WFC3 CCDs)
- Implications for star camera applications: CCDs, APS, CIDs, hybrids
1
1 10 100 1000
PROTON ENERGY (MeV)
Nonionizing energy loss correlates with displacement damage as LET does
with ionizing dose (rads(Si))
C.J. Dale et a/., "A Comparison of Monte Carlo and Analytic Treatments
of Displacement Damage in Microvolumes, IEEE MS, Vd. 41. 1994.
I Vaca nc y -l nt ers t it ial
able Defect
- @ E 3 0 =%m..mI
Displacement Damage U<ITING
STABLE
pARTlcLE DEFECT
0 /fa*'\ ;
0 0 0 0,a 0 0
0 4'0 0 0 0 0 0
f
0 0 0 0 0 0 0 0
o
INCIDENT Interstitial
PARTICLE
I-$ Vacancy
-I
e
Dopant or Impurity Atom
-Stable divacancy complex defects may also form and are important for
dark current generation
*
Resented by Paul W We 281h Anrud #AS Gudara 6 Cow Ws-. Eb-eckmdgc. CdOraJo.F rsrUn, E9 Mo5 4
2
Proton Spectra Behind Thick Shields
Trapped proton environment decrea-se s exponentially
with energy and is populated out to 400 MeV
Lower energy protons are stopped more easily, but all
protons lose energy when transported through shield
-
Behind 1 inch AI, the average proton energy is >80 MeV
-
>Relatively few protons below 30 MeV
Need to anticipate the damage mechanisms for
higher energy protons (> 50 MeV)
+
I I
A PROTON ENERGY
6-10 MeV > 20 MeV
Log N
b
I I
RECOIL ENERGY
II I
11--22 kkeeVV 1122--2200 kkeeVV
3
Electrical Effects of
Proton-Induced Defects
-1.0@2S!
Conduction Band
\ 4 : \
+ r. I ~~~
'
ValenceBand
Generation Trapping
Recombination Compensation
Dark Current and Tern peratu r e
mamimmu
- E E E l n
+Leakage W n gT emperamel
fEad = 0.63 eV)
Jd is bulk leakage or dark
current: not FET leakage
Kdariks damage factor
8
Eadi s activation energy g 3
Ed = 0.63 e V
-! I
k is Boltman's constant 01 I
k = 8.62 E45 e V/T 2 0 0 2 2 0 240 260 280 300
Temperature (degrees Kehn'n)
Tis Temperature in Kelvin
Srour and Lo find that proton induced leakage from a variety of studies exhibit an
activation energy of from 0.62 to 0.64 eV, and 0.63 works well for both pre and post radiation
-
[l]" Universal Damage Factor for Radiation-Induced Dark Current in Si Devices,'
J.R. Srourand D.H. Lo, IEEETrans. Nucl. Sci., NS-47. No. 6. Dec. 2000. pp. 2451-59
W e dbY W d the tahAnru4 AAS Gud.rra 6 Omw m W .,- cdonfo.F abruTy $9- 8
4
Temporal Dark Current Fluctuations
in a Pixel
I
1
I I
EEV ccws, 10 MeV protons
9
I
A DARK CURRENT CWCd) Nonunifonnity
4 B 12
I I 1 I I I I Can Be Significant
a 12Mev PROTONS Concern
000u(pERIuENT
-
C A L ~ T n N Measurements at
293' Kelvin
High Energy
Tails Due to
statistics of
Nuclear Reactions
DAMAGE ENERGY (WV)
CID Pixel volume is 1300 cubic microns: or -1/8" of reference volume after
roughly 40 krad(Si) damage for highest damage level shown, so consider -300
nA/cm2 at mom temperature and -258" K to reach 10 nNcdf or the average J,
*Expect more broadening of distribution due to higher proton energies
P. W. Marshall, et al., ?roton-Induced Displacement Damage DistribOtions and Extremes in
Silicon Microvolumes," lEEE Trans. On Nucl. Sci, NS-37. No. 6, 1990.
bj'w UasM d LheZUh ARud AAS G WL Co nbd m.Be cLemdge, CdSdQ. FmbQZOOS 10
5
.-
@
-
HST Hot Pixels on Orbit Annealing
4l e n n m o c m m i i
WFC, hot pix > 0.04 e/s .
i " . I . " I ' . '
Accumulation of
0
hot pixels
mandates monthly
warm-ups to -0C
Similar annealing
0
results observed in
ng
3 HST instruments
0.2
0 1st laboratory s
measurements
made for WFC3
I
Months
11
WFC3 Hot Pixel Experiment (1)
mExI0 E H E . B i l
A Hubble Space Telescope Wide Field Camera 3 E2V CCD was irradiated
0
while operating at -83 "C and the dark current studied as a function of
temperature while the CCD was warmed up to a +30"C.
Populations of
0
hot pixels (defined
0 immediately After Inadation
as pixels having A[+1PA RqeercAiefldnES&Ynega rs
Upper limit due to 55 Fe contamination
more than a given
105
dark current -m
:'/
threshold) were E
I-d '04 -
twrahcektheedrt oai hdoetn ptiifxye l
I-0
encountered at any -
103 F
fluence is a new
hot pixel or an Threshold = 144 e-/h
8
Fluence (of 63 MeV protonslcrn2)
ReMled by Pad Mastd d lheZ67h Arm& AASGuida'xe 6 CDnW cadaacc. Becktnidge. Cdora& Febmq5-9 2005 12
6
WFC3 Hot Pixel Experiment (2) @
~ O O € C B ! 0 H H 6 B i l
Hot pixel populations
tracked during warm-up
and cool-down. Annealing
process undenrray by
40C and continues
through wanner
temperatures with no
sharp activation energy.
*The peak annealing rate
at 0°C is an artifact
caused by the saturation
of hot pixels.
WFC3 Hot Pixel Experiment (3)
-@@El0 m e H B B I l
Mean dark
current slightly
105
elevated, but still
less than 2 eJh
Shape of tails 104
ears annealed
changing
Annealed hot +54 103
6
pixels become
warm pixels 102 2 months annealed
10'
10 0
1 10 100 1000 10000
Dark Rate (e-/h)
7
0 Depending on the application either CTE (for CCDs) or
dark current can be the driving radiation effect.
Hot pixels may be caused by small (Coulomb) events in
0
high field regions and by rare nuclear reactions.
Observed in CCDs, CIDs, APS and Si hybrid devices
0
If your tracker is cooled (even modestly) then you will
0
underestimatet he number of hot pixels by following usual
proton test procedures at room temperature.
.This can overwhelm the capacity of the tracker
software.
8