Download Mandic_RD50_CERN_10

Annealing effects in n+p strip detectors irradiated with
high neutron fluences
Igor Mandić1, Vladimir Cindro1, Andrej Gorišek1,Gregor Kramberger1,
Marko Milovanović1, Marko Mikuž1,2, Marko Zavrtanik1
1Jožef
Stefan Institute, Ljubljana, Slovenia
2 Faculty of Mathematics and Physics, University of Ljubljana, Slovenia
I. Mandić, 17th RD50 Workshop, CERN, 17 November – 19 November 2010
1
• final results of annealing studies with hamamatsu detecotrs
• paper accepted for publication in NIM
Detectors:
• p-type, FZ, 320 µm thick, 75 µm strip pitch, 1x1 cm2 , produced by Hamamatsu
1) ATLAS07-PSSSSD_Series I, W45, BZ3-P15:
2) ATLAS07-PSSSSD_Series I, W19, BZ3-P18:
3) ATLAS07-PSSSSD_Series I, W22, BZ3-P3:
4) ATLAS07-PSSSSD_Series I, W16, BZ3-P21:
Φ = 2∙1014 n/cm2
Φ = 5∙1014 n/cm2
Φ = 1∙1015 n/cm2
Φ = 5∙1015 n/cm2
• detectors irradiated with neutrons in reactor in Ljubljana
2
Setup:
• SCTA128VG chip
• VME module SEQSI (for clock, commands...)
• Tektronix digital scope for data acquisition
• 90Sr source, photomultiplier, scintillator,
power supplies, coincidence circuit ......
• Most probable value (MPV) from fit of Landau + Gaus to
distribution of measured signal cluster heights
 scale defined with signals from not irradiated detector
• measure collected charge and leakage current,
after annealing steps at 60 C
• measurements done after: 80, 240, 560, 1200 2480 and 5040 minutes at 60 C.
I. Mandić, 17th RD50 Workshop, CERN, 17 November – 19 November 2010
1
Annealing of MPV
Low fluences, low voltages:
 standard behavior: beneficial annealing followed by reverse annealing
High voltages:
 rise of MPV with reverse annealing can be observed
I. Mandić, 17th RD50 Workshop, CERN, 17 November – 19 November 2010
2
Vfd from the kink: ~600 , ~700, ~800, ~900
Vfd from the kink: ~900, ~900, ~1000, ~1200
Time at 60 C
80
240
560
1200
2480
5040
Vfd (2e14)
490
580
710
860
1000
1100
Vfd (5e14)
980
1200
1500
1900
2300
2500
Vfd calculated with following parameters:
Stable: gc = 0.017 cm-1, short term: ga = 0.018 cm-1, τa = 19 min, (60°C) (V. Cindro et al.)
Long term: gY = 0.053 cm-1, τY = 1100 min (60°C, G. Kramberger et al.)
Vfd estimated from the plot higher then expected from Vfd: hint of multiplication
5
MPV
High fluences, high voltages:
MPV drops due to short term annealing:
• Neff drops  smaller peak electric field  less multiplication
MPV rises due to long term annealing:
• Neff rises  larger peak electric field  more multiplication
It seems that breakdown voltage is lower at 5e14 and 1e15 than at 2e14 and 5e15
 breakdown voltage decreases with reverse annealing
6
Leakage current (guard ring not bonded)
• Increase of leakage current with annealing  multiplication
7
Compare with calculation (annealing parameters from M. Moll et al. NIMA 426(1999) p. 87)
• high voltage: current higher then expected even if full depletion assumed
(some disagreament at 2e14  maybe error on fluence )
8
Noise
Noise increases when multiplication large
9
Signal/Noise
Signal/Noise ratio doesn’t change significantly
10
Discussion about noise:
Noise can be written as:
ENC2= ENCp2+ENCs2,
ENCs : serial noise for SCT128 and this detector capacitance ~ 800 e
ENCI2 : shot noise proportional to current I
If no multiplication: ENC ~ 900 el  dominated by ENCs, shot noise negligible
If multiplication:
 I 2  F  M 2  I0
M – multiplication factor, F = 2 if only electrons get multiplied (our case)
Shot noise for SCT128 chip (triangular shaping):
ENCI
2
2
2
2
  p   I   p  F  M  I
3e
3e
M can be estimated as:
τp = 20 ns
peaking time
e = 1.6·10-19 As elementary charge
I Measured
M
I Calculated
11
Icalculated from depleted volume (larger M)
Icalculated from whole detector volume
Total noise increase due to shot noise
if there is no multiplication (F = M = 1) small:
12
Noise increase larger than caluclated:
-underestimated M ?
- dissagreement in the last point before breakdown – microdischarges?
13
Conclusions
• at high fluences and high bias voltages multiplication effects influence annealing
behavior:
 multiplication increases with annealing time larger than
~ 500 minutes @ 60°C (~170 days @ 20°C)
 increase of space charge concentration  higher electric fields  more multiplication
 collected charge increases with reverse annealing
 leakage current increases with reverse annealing
 multiplication decreases with beneficial annealing
 collected charge decreases with reverse annealing
• noise increases because of multiplication  shot noise increase
• signal/noise ration doesn’t change significantly
14