Документ взят из кэша поисковой машины. Адрес оригинального документа : http://zebu.uoregon.edu/~uochep/talks/talks03/rad.pdf
Дата изменения: Wed Jul 23 00:50:31 2003
Дата индексирования: Tue Oct 2 09:36:16 2012
Кодировка:

ALCPG July 2003

CCD Radiation Damage Studies
J. Brau, O. Igonkina, N. Sinev, C. Potter (University of Oregon)
July 13, 2003

Outline: · Introduction · Repeating 1999 year study · Irradiation of CCD with electrons · Measurement of CTI and trap/pixel distribution · Conclusions · Plans

O. Igonkina for U.Oregon

CCD Radiation Damage Studies

1

ALCPG July 2003

CCD

CCD gates

v v1 v2 3
~20 µ m

Pixel

The CCD at SLD has shown an outstanding performance, but it also saw the radiation damage when undamped beam run through the detector.
Glass backing p substrate

signal loss

CCD readout

n channel

CCD is proposed as vertex detector for next linear collider. It has excellent space resolution ( 4µm) and is very thin (0.4-0.1%X0 ).

190K

220K

O. Igonkina for U.Oregon

CCD Radiation Damage Studies

2

ALCPG July 2003

Goal

· to predict the radiation damage of a CCD in the environment of linear collider · to verify background conditions (source of radiation) at linear collider by studying SLD VXD3 CCD · to minimize impact of radiation damage on the detector operation

It is very important to understand (quantitatively and qualitatively) the effects from radiation by different type of particles (neutrons, electrons).

· to understand what kind of traps are created : cluster of traps or single traps per pixel nature of trap (VO, V2 O, V2 , etc; acceptors or donors) charge retention time by traps (depends on trap sort) · to measure the charge transfer inefficiency (CTI) due to damage · to find out if can traps be annealed

O. Igonkina for U.Oregon

CCD Radiation Damage Studies

3

ALCPG July 2003

Observables

Analysis of the different CCDs: irradiated with neutrons, irradiated with electrons, CCD from SLD VXD3

· the distribution of number of traps per pixel · the space distribution of damaged pixel (to study background conditions at SLD) · the charge transfer inefficiency · the charge retention time by traps and its temperature dependence · the temperature and speed of damage annealing

See IEEE Trans. Nucl. Sci NS-47 (2000) 1898 for results on radiation damage by 5 · 109 neutrons/cm2 . This talk is concentrated on the work on radiation damage by 1012 electrons/cm2 . Next workshop - analysis of the SLD VXD3

O. Igonkina for U.Oregon

CCD Radiation Damage Studies

4

ALCPG July 2003

CCD Radiation Damage Test-stand

LEDs LFM

T=-73

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Data A/D board AUX Control board PC

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

CCD
Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

o

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Ў

Cryostat

Ў

Ў

Fastbus crate N
2

Two types of LED were used : · "Uniform" light source which flashes the complete CCD surface uniformly. The average signal intensity is about 25-30 electrons/pixel · "Narrow line" source which flashes only few pixels per column. The average signal intensity is 100 times larger.

O. Igonkina for U.Oregon

CCD Radiation Damage Studies

5

ALCPG July 2003

Timing Diagram of the test

Regular measurement scheme
Start of the cycle
LED BC I clk

Signal light pulse

Next cycle start

Signal readout 25 BC
Start of the cycle LED BC I clk

Data processing about 0.8 s

Measurement with sacrificial charge
Injection of traps filling charge Injection of test charge

Variable delay to allow traps emptying

Variable delay to allow traps filling

Data processing

O. Igonkina for U.Oregon

CCD Radiation Damage Studies

6

ALCPG July 2003

Example: CCD Image with Uniform Light

Each pixel is clearly identified

Varying delay between sacrificial light and signal one could have all traps filled or empty

traps

traps are filled - no signal lost
shown in 1999 year studies

O. Igonkina for U.Oregon

CCD Radiation Damage Studies

7

ALCPG July 2003

Trap Fill Time

Ntraps
data 2003, c =15 ms

Changing the delay between signal injection and start of read out from 40 ms to 15 ms showed that the filling time c is of the order of tens ms! which contradicts to the expectation that c is of the order of ns (C. Damerrel, RAL-P-95-008)

data 1999, c =40 ms First comparison 2003 vs 1999 - lo oks like the traps have annealed. but data 2003 with c =40 ms showed that just more time is needed to trap electrons

O. Igonkina for U.Oregon

CCD Radiation Damage Studies

8

ALCPG July 2003

No Annealing at Room Temperature

Comparison of the 1999 and 2003 data before irradiation with electrons (collected under exactly the same conditions) shows same amount of traps in the same pixels. The expectation that traps will anneal during 1 year at room temperature is not confirmed.
data 2003, c =40 ms

data 1999, c =40 ms

Ntraps(2003)/Ntraps(1999)

O. Igonkina for U.Oregon

CCD Radiation Damage Studies

9

ALCPG July 2003

Irradiation with Electrons

During last month we have

irradiated the CCD with about 1012 e- /cm2 of 60 MeV energy

at NLCTA test beam (SLAC) and continued the analysis of the traps :

· using the "flat" uniform light source compare number of traps per pixel before and after irradiation · using the "narrow line" light source compare the charge transfer inefficiency before and after irradiation

O. Igonkina for U.Oregon

CCD Radiation Damage Studies

10

ALCPG July 2003

Traps Before and After Irradiation with Electrons
n traps/ pixel

However, no new pixel with a lot of new traps are observed - electron damage does not cause trap clusters (as expected). Some old clusters of traps were annealed during irradiation of CCD by electrons. Ntrap - Average < - 12.6 year 1999 3552 2003 before irr. 3503 2003 after irr. 2754

1999, after neutron damage 10 2003, b efore electron damage 2003, after electron damage
5

10

4

10

3

10

2

-30

-25

-20

-15

-10

-5

0

5

10

deviation of Ntraps from average

O. Igonkina for U.Oregon

CCD Radiation Damage Studies

11

ALCPG July 2003

Charge Transfer Inefficiency (CTI)
The CTI is measured with the narrow line light. The signal has large intensity, but travels through about 1500 pixels during readout. The amplitudes of the CCD output at 2 different conditions are compared : a with minimum delay between "sacrificial" charge and signal light pulse (almost all traps are filled - no signal loss) b with large ( 1s) delay between "sacrificial" charge and signal light pulse (almost all traps are empty - cause signal loss) Average CTI= 1 - b/a Before irradiation with electrons CTIbef ore = 14.4% ; After irradiation with electrons CTIaf ter = 52.9% Irradiation with 1012 e- /cm2 results in large CTI

The dose is to b e estimated more precise
O. Igonkina for U.Oregon CCD Radiation Damage Studies 12

ALCPG July 2003

Comparison of the Neutron and Electron Damage

SLD The radiation damage by 5 · 109 neutrons /cm2 resulted in clusters of traps. The traps created during irradiation with electrons ( 1012 e- /cm2 ) distributed uniformly over the surface of the CCD but cause large charge transfer inefficiency. That should help us to distinguish between 2 types of the radiation in analysis of SLD VXD3 and verify the background condition at the linear collider.

VXD3

CCD

O. Igonkina for U.Oregon

CCD Radiation Damage Studies

13

ALCPG July 2003

Conclusions

· After 4 years the CCD still had the same traps - no significant annealing at room temperature is observed.

· Traps filling with signal charge takes noticeable time - much larger that the time signal spend in each pixel. We expect that the increased readout speed will lead to decrease of charge transfer inefficiency. One could, however, start to see CTI from shallow traps at very high readout speed (see http://www.hep.lancs.ac.uk/lcfi/talk Aachen.pdf )

· The large CTI is observed after electron irradiation damage. The new traps, however, are distributed uniformly over CCD and do not combine clusters. This suggests that we should be able to distinguish the radiation damage by neutrons and electrons analyzing the SLD VXD3 CCD and to understand damage observed there

O. Igonkina for U.Oregon

CCD Radiation Damage Studies

14

ALCPG July 2003

Plans

· Finish the analysis of the electron radiation damage. We still have spare CCD (not damaged yet) to be irradiated with electron beam at NLCTA (SLAC).

· Understand more accurately the dependence of the charge transfer inefficiency on the measurement parameters (temperature, delay between signal and sacrificial light, readout speed)

· Application of presented technique to SLD VXD3 should show the origin of the its radiation damage and clearfy the background conditions at SLD. This in turn will allow to predict the radiation damage at future linear collider.

· Develop the technique to minimize the impact of the radition on the detector operation.

O. Igonkina for U.Oregon

CCD Radiation Damage Studies

15