View
33
Download
0
Category
Tags:
Preview:
DESCRIPTION
Annealing of the vacancy-oxygen and divacancy-oxygen complexes in silicon. M. Mikelsen 1 , E. V. Monakhov 1 , J. H Bleka 1 , J. S. Christensen 1 , B. S. Avset 2 , J. Härkönen 3 , B. G. Svensson 1 1 Department of Physics, Physical Electronics, University of Oslo, P.O. Box 1048 Blindern, - PowerPoint PPT Presentation
Citation preview
1
Annealing of the vacancy-oxygen and divacancy-oxygen
complexes in silicon M. Mikelsen1, E. V. Monakhov1, J. H Bleka1, J. S. Christensen1, B. S.
Avset2, J. Härkönen3, B. G. Svensson1
1Department of Physics, Physical Electronics, University of Oslo, P.O. Box 1048 Blindern, N-0316 Oslo, Norway
2SINTEF, Electronics and Cybernetics. P.O. Box 124 Blindern, N-0314 Oslo, Norway(Present address: Statens Strålevern, Grini næringspark 13, POB 55, 1332 Østerås)
3Helsinki Institute of Physics, Helsinki University, Helsinki, Finland
Special thanks to:The Norwegian Research Counsel (NFR) for Financial Support
2madmike@fys.uio.no
Background
During irradiation we get: V + Oi VO
V + V V2
(V2 is also produced directly)
For higher doses: V + VO V2O
Another channel for V2O formation:
V2 + Oi V2O
Basis for I-level identification
Basis for X-level identification
3madmike@fys.uio.no
Background
100 120 140 160 180 200 2200
1
2
3
4
5
6
0.36 eV
15 min at 225OC
14 min at 285OC
35 min at 285OC
112 min at 285OC
2N
ddC
/C (
101
1 c
m-3
)
Temperature (K)
V2(=/-)X(=/-)
X(-/0)
V2(-/0)
M. Mikelsen, E. V. Monakhov, G. Alfieri, B. S. Avset, and B. G. Svensson Phys. Rev. B 72, 195207 (2005)
X was identified as V2O
V2 migrates:
V2 + Oi V2O
4madmike@fys.uio.no
ExperimentSurvey of the samples used in the study:
SampleDoping (P/cm3)
Carbon (cm-3)
Oxygen (cm-3)
MCZ-Si 5.5×1012 ≤ 1016 (5–10)×1017
DOFZ-Si 5.0×1012 (2–4)×1016 (2–3)×1017
Dose: 2 - 5 ×1012 cm-2
6 MeV electrons
DOFZ = Diffusion-Oxygenated Float-Zone Si MCZ = Magnetic Czochralski
Experimental method: DLTS Pre-annealing to insure a V2 V2O transformation. Isothermal annealing in the range 275-355 oC.
5madmike@fys.uio.no
Results: DOFZ - Si
80 100 120 140 160 180 200 2200
10
20
30
40
50 DOFZ-Si
15 min at 225oC
15 min 300oC
10 min 335oC
30 min 335oC
70 min 335oC
2
Ndd
C/C
(1
01
0 c
m-3
)
Temperature (K)
VO(-/0)
6madmike@fys.uio.no
Results: DOFZ - Si
100 120 140 160 180 200 2200
2
4
6
8
10
12
B
CD
V2(-/0)V2(=/-)V2O(=/-) V2O(-/0)
DOFZ-Si
15 min at 225oC
15 min 300oC
10 min 335oC
30 min 335oC
70 min 335oC
2N
ddC
/C (
1010
cm
-3)
Temperature (K)
A
Other levels observed: A at ~157 K (*)
B at ~200 K Similar electrical
parameters as V2O(-/0).
C at ~175 K
D at ~157 K
(*) To be discussed in: ”On the formation of the L-center in silicon during heat treatment in the temperature range 205 to 285°C”M. Mikelsen, E. V. Monakhov, B. S. Avset, B. G. Svensson To be published in “Physica sicripta”
7madmike@fys.uio.no
Results: DOFZ - Si
100 120 140 160 180 200 2200
2
4
6
8
10
12
~0.27 eV~0.31 eV
0.47 eV
0.43eV0.23(7) eV0.23(2) eV 0.47 eV
2Ndd
C/C
(10
10 c
m-3
)
Temperature (K)
0.36 eV
The activation enthalpies for peak C and D were obtained by using a computer program that fits the spectrum to a sum of single peaks.
8madmike@fys.uio.no
Results: MCZ Si
100 120 140 160 180 200 220 240 260 2800
2
4
6
8
10
12
14
16
MCZ-Si
15 min at 300oC
50 min 325oC
125 min 325oC
250 min 325oC
2N
ddC
/C (
1010
cm
-3)
Temperature (K)
MCZ-Si behaves similar to DOFZ But the area between
the two V2O peaks looks a little bit different. (No D-peak)
9madmike@fys.uio.no
Results: V2O annealing
0 20 40 60 80 100 120 140
0.1
1
MCZ: 325oC
VO(-/0) V2O(=/-) V2O(-/0) Fit to V2O(=/-)
[VO
] / [V
O]t=
0 a
nd
[V
2] /
[V2]
t=0
Annealing time (min)
Example: The annealing of V2O(=/-) exhibits first order kinetics. For both MCZ and
DOFZ. For all annealing
temperatures in the range 275 - 355 oC.
The kinetics for V2O(-/0) is affected by overlapping level.
10madmike@fys.uio.no
Results: V2O annealing
0.00155 0.00160 0.00165 0.00170 0.00175 0.00180 0.00185
1E-5
1E-4
1E-3
V2O(=/-) in MCZ-Si V2O(=/-) in DOFZ-Si
Ann
eal
ing
rat
e (s
-1)
Ea=2.0 eV
C0= 2 x1013
1/T (K-1)
11madmike@fys.uio.no
The V2O annealing mechanism It is improbable that V2O anneals by migration and trapping. Since Oi is the main impurity, one would expect V2O2 to form.
V2O2 is believed to be electrically active with two acceptor centers similar to V2 and V2O.
V2O does not anneal by interacting with migrating oxygen dimers. The diffusivity of O2 is to low by a factor 104.
12madmike@fys.uio.no
The V2O annealing mechanism A prefactor, c0, in the 1012 – 1013 range is expected for dissociation. The experimental value is in this range taking in regard the
uncertainty. If dissociation were to occur we would expect an increase in VO;
V2O V + VO
…. So we must investigate the VO annealing, to check
if it really is dissociation.
13madmike@fys.uio.no
Results: VO annealing (DOFZ)
0 50 100 150 200
0
2
4
6
DATA Model
[VO
](1
01
1 c
m-3
)
Annealing time (min)
350 oC
After the initial increase, [VO] decreases with 1st. order kinetics.
14madmike@fys.uio.no
A model for defect annealing
The reactions V2O VO +V (dissociation) V + Oi VO (migration of V) VO + Oi VO2 (migration of VO)
VO V + Oi (dissociation)
VO +V V2O (migration of V)
The diff-equations d[VO]/dt = - Cdiss_V2O(T) [V2O] + 4 R DV(T) [V] [VO]
d[VO]/dt = Cdiss_V2O(T) [V2O] + 4 R DV(T) [V] [Oi] – 4 R DVO(T) [VO] [Oi] - Cdiss_VO(T) [VO] – 4 R DV(T) [V] [VO]
d[V]/dt = Cdiss_V2O(T) [V2O] + Cdiss_VO(T) [VO] – 4 R DV(T) [V] [Oi]
- 4 R DV(T) [V] [VO]
15madmike@fys.uio.no
Results: VO annealing (DOFZ)
0 50 100 150 200
0
2
4
6
DATA Model
[VO
](1
01
1 c
m-3
)
Annealing time (min)
350 oC The model is
highly sensitive to the [Oi] concentration. We used
[Oi]= 2.5 x 1017.
16madmike@fys.uio.no
Results
80 100 120 140 160 180 200 220 240
0
10
20
30
40
50
2N
ddC
/C (
10
10 c
m-3
)
Temperature (K)
DOFZ, 350 oC pre annealed 15 min 30 min 45 min 60 min 75 min 90 min 105 min 120 min 150 min 180 min
VOH
Hydrogen may interact with VO.
17madmike@fys.uio.no
Summary The V2O defect anneals with an activation
energy, Ea, of ~2.0 eV and a prefactor ~2 x 1013 s-1.
Dissociation is the likely annealing mechanism for V2O.
A model that includes VO and V2O dissociation and VO migration and trapping describes the observed annealing kinetics well.
Recommended