positron beam studies of defects in...

Positron Beam Studies of Defects in SemiconductorsPositron Beam Studies of Defects in Semiconductors

R. Krause-Rehberg, F. Börner, F. Redmann, J. Gebauer, S. Eichler

Universität Halle, FB Physik

� The development of the technique

� Recent progress: lateral and depth resolution enhancement

� Example: The Rp/2 effect in Si

� Future developments

Martin-Luther-Universität Halle

Martin-Luther-Universität

Halle-Wittenberg

Monoenergetic positrons by moderation


� positron annihilation was very successful in defect identification in last decades

� semiconductor technology: thin layers (epitaxy, ion implantation)

� broad energy distribution due to β+ decay

� some surfaces: negative workfunction ⇒ moderation (rather unefficient)

Energy distribution after β+ decayEffect of moderation

Conventional positron beam technique


� positron beam can be made from monoenergetic positrons

� often: magnetically guided for simplicity

� disadvantage: no simple lifetime measurements and bad lateral resolution

� defect studies by Doppler-broadening spectroscopy

� characterization of defects only by line-shape parameters or positron diffusion length

Information from Doppler-broadening spectroscopy


� S parameter does not contain all information

� W parameter determined by annihilation with core electrons ⇒ chemical sensitivity

� R parameter was invented in 1978

S. Mantl, W. Triftshäuser, Phys. Rev. B 17 (1978) 1645

� this parameter was revitalized by the S-W-plotL. Liszkay et al., Appl. Phys. Lett. 64 (1994) 1380

506 508 510 512 514 516 518 5200,0

0,2

0,4

0,6

0,8

1,0

defect-free

defect-riche+ annihilationin GaAs

AS

norm

aliz

ed in

tens

ity

γ energy (keV)

514 515

AW

b d b

b d b

S S S SR

W W W W

− −= =− −

S. Eichler, PhD Thesis, 1997

Ion implantation in Si

0 .92

0 .96

0 .96 0 .98 1 .00

1 .00

1 .04

M e an p o s itro n d ep th (µ m )

P o sitro n e n erg y (ke V )

d e fec t

su r face I

su r face II

b u lk

B :S i 5 0 , 1 5 0 , 3 0 0 k e V

1 ·1 0 cm1 4 -2

re fe re n ce

2 ·1 0 c m1 6 -2

0 1 0 2 0 3 0 4 0

0 0 .59 1 .92 3 .83 6 .24

S pa

ram

eter

W parameter

S. Eichler, PhD Thesis, 1997

.××

4 0 0 6 0 0 8 0 0 1 0 00 1 2 00

1 .00

1 .02

1 .04

1 .06

1 .08

1 .10

S i:S i 5×1 0 cm1 4 -2

S i:S i 1×1 0 cm1 5 -2

annea ling tem pera tu re (K )

norm

aliz

edla

yer

Spa

ram

eter

Complex defect reactions show up in S-W-plot


� defects of self-implantation of Cz-Si (300 keV) show a complex annealing behavior

� during annealing: defect reactions take place

� oxygen-vacancy complexes are formed and grow further

� although more detailed information by S-W-plot: positron lifetime absolutely essential

for determination of open-volume and trapping fraction

Ion implantation in Cz-Si

� coincident detection of second annihilation γ quantum:strong reduction of background

� high-momentum region accessible

� comparison with theoretically calculated spectra:

information on chemical environment of annihilation site

Doppler-coincidence spectroscopy


(Gebauer, unpublished)

E1+E2= 2 m0 c2

=1022 keV

500 505 510 515 520 525

10-5

10-4

10-3

10-2

10-1

100

Rel

ativ

e in

tens

ity

γ-ray energy [keV]

with coincidence

without

K.G. Lynn et al., Phys. Rev. Lett. 38 (1977) 241

M. Alatalo et al., Rev. B 51 (1995) 4176.

� Doppler spectra show only minor changes

� difference or ratio plots: characteristic elemental features

� a single atom in direct neighborhood of a vacancy leaves its mark (VGa-TeAs in GaAs:Te)

� chemical senistivity of DOBS is a real advantage over lifetime spectroscopy

Doppler-coincidence spectroscopy in GaAs


J. Gebauer et al., Phys. Rev. B 60 (1999) 1464

0 10 20 30

0,5

1,0

1,5

Electron momentum pL (10

-3 m

0c)

Te Ga As Si

Rat

io t

o d

efec

t-fr

ee G

aAs

0,8

1,0

10 20 30

0,8

1,0

VGa

-SiGa

Vacancy in GaAs:Te

Experiment

VGa

-SiGa

VGa

-TeAs

VGa

VAs

Theory

Electron momentum pL (10-3 m

0c)

Rat

io t

o d

efec

t-fr

ee G

aAs

0 10 20 3010-6

1x10-5

1x10-4

10-3

10-2

Elemental Si Ga As Te

ρ (p

) [1

03 (m

0c)-1]

Electron momentum pL (10-3 m

0c)


Positron lifetime spectroscopy using monoenergetic positrons

� positron lifetime measurement more difficult

� a system of chopper and bunchers: short

pulses of monoenergetic positrons

� problems with sidepeaks are solved

� another system: Tsukuba

� more systems under construction

Schödelbauer et al., NIM 34 (1988) 258Kögel et al.,

Mat. Sci. Forum 175 (1995) 107

n-type Si

Munich system

� Semiconductor devices nm-sized ⇒ Positron microscopes and microprobes required

� However: positron diffusion length is fundamental limit for lateral resolution

� no sense to improve resolution below 500 nm

Improvement of lateral resolution


H. Greif et al., Appl. Phys. Lett., 71 (1997) 2115

Bonn Positron Microprobe Munich Positron Microscope

W. Triftshäuser et al., NIM B 130 (1997) 265

� for 3D-imaging of defects: depth resolution must also be improved

� positron implantation profile rather broad compared to L+ at energies > 10 keV

� Makhov function roughly describes this profile

Improvement of depth resolution


−=

− m

m

m

z

z

z

mzEzP

00

1

),(

� idea: stepwise removal of surface and

measurement with small e+ energy

� chemical etching:

N.B. Chilton and M. Fujinami, AIP Conf. Proc. 303 (1994) 25H. Kauppinen, C. Corbel, K. Skog, K. Saarinen, T. Laine, P. Hautojärvi, P. Desgardin and E. Ntsoenzok, Phys. Rev. B 55 (1997) 9598F. Börner, S. Eichler, A. Polity, R. Krause-Rehberg, R. Hammer and M. Jurisch, Appl. Surf. Sci. 149 (1999) 151P.J. Simpson, M. Spooner, H. Xia and A.P. Knights, J. Appl. Phys. 85 (1999) 1765

P.G. Coleman and A.P. Knights, Appl. Surf. Sci. 149 (1999) 82

� another idea: surface removal by ion sputtering

� test structure: a-Si/SiO2/c-Si

� sputter conditions:

Improvement of depth resolution by ion sputtering


4 8 5 n m

6 0 0 n m

a -S i

c -S i

S iO 2

A r rad ia tio n

Θ

� Ar+ of 2.5 keV at 60° angle

� 2×10-6 Torr (3 ×10-4 Pa)

� about 3 nm / min

0 2 4 6 8 1 0 1 2 0 .9 0

0 .9 5

1 .0 0

1 .0 5

1 .1 0

S p u tte r s tep

S surf

/S re

f

0 .8

1 .0

1 .2

1 .4

1 .6

1 .8

2 .0

W su

rf /W

ref

c -S i su bstra te S iO 2 (60 0 n m ) a-S i (4 85 nm ) S i o x id e

0 .9 2

0 .9 4

0 .9 6

0 .9 8

1 .0 0

1 .0 2

1 .0 4

S/S

bulk

at 2

.5 k

eV

0 .9 2

0 .9 4

0 .9 6

0 .9 8

1 .0 0

1 .0 2

1 .0 4

S/S

bulk

0 .0 0 .2 0 .4 0 .6 0 .8 1 .0 1 .2

1 .0

1 .2

1 .4

1 .6

1 .8

D ep th ( µ m )

W/W

bulk

at 2

.5 k

eV

1 .0

1 .2

1 .4

1 .6

1 .8

W/W

bulk

c -S i s u b s tra te S iO 2 (6 0 0 n m ) a -S i (4 8 5 n m )

S i o x id e

Surface parameters

Please visit our Poster: „Improved depth resolution of slow positron method- detailed determination of defect profiles“ presented by F. Börner


Defects in high-energy self-implanted Si The Rp/2 effect

� after high-energy (3.5 MeV) self-implantation of Si (5 ×1015 cm-2) and RTA annealing

(900°C, 30s): two new gettering zones appear at Rp and Rp/2 (Rp = projected range of Si+)

� visible by SIMS profiling after intentional Cu contamination

0 1 2 3 41015

1016

1017

Cu

conc

entr

atio

n (c

m-3)

Depth (µm)

RpRp/2

SIMS

� at Rp: gettering by interstitial-type

dislocation loops (formed by excess

interstitials during RTA)

� no defects visible by TEM at Rp/2

� What type are these defects?

TEM image by P. Werner, MPI Halle

Interstitial type

[1,2]

Vacancy type

[3,4]

[1] R. A. Brown, et al., J. Appl. Phys. 84 (1998) 2459[2] J. Xu, et al., Appl. Phys. Lett. 74 (1999) 997[3] R. Kögler, et al., Appl. Phys. Lett. 75 (1999) 1279[4] A. Peeva, et al., NIM B 161 (2000) 1090


Investigation of the Rp/2 effect

by conventional VEPAS

� the defect layers are expected in a

depth of 1.7 µm and 2.8 µm corresponding

to E+= 18 and 25 keV

� implantation profile too broad to

discriminate between the two zones

� simulation of S(E) curve gives the same

result for assumed blue and yellow

defect profile (solid line in upper panel)

� furthermore: small effect only

� no conclusions about origin of Rp/2 effect

possible

0 1 2 3 4 5 6 0 .0 0

0 .0 1

0 .0 2

0 .9 4

0 .9 6

0 .9 8

1 .0 0

1 7

1 10 ×

1 10 ×

1 8

P(z

,E)

D e p th ( µ m )

P o sitron e n e rg y (k e V )

40 35 30 25 20 15 10 0

S/S

bulk

defe

ct d

ensi

ty (

cm -3

)

re fe ren ce im p lanted andan nea led + C u

18 keV25 keV


Getter centers after high-energy self-implantation in Si

0,95

1,00

1,05

1,10

1,15

W/W

ref a

t 7.5

keV

0,995

1,000

1,005

1,010

1,015

S/S

ref a

t 10

keV

0 1 2 3 4 5 6

1016

1017

PositronAnnihilation

SIMS

Cu

dens

ity (

cm-3)

Depth (µm)

TEM

surface RpRp/2

0 1 2 3 4 5 6

1.000

1.005

1.010

1.015

Rp/2 R

p

Depth (µm)

S/S re

f

0.95 1.00 1.05 1.10

surface

running parameter:depth

Rp/2R

p

bulk

W/W

Please visit our Poster: „Impurity gettering in high-energy ion-implanted siliconat Rp/2 by vacancy-type defects“

� the depth-enhanced VEPAS show clearly open-volume

defects at Rp/2 and Rp

� they must be different (see S-W-plot)

� �normal� behavior of W parameter at Rp but high value at

Rp/2: Cu decorates the vacancy-type defect

Doppler-coincidence and lifetime spectroscopy


0 2 4 6 8

0

200

400

600

-60

-40

-20

0

20

0 8 16 24 32

Cu

γ energy (difference to 511 keV)

RTA-annealed

RTA-annealed +Cu contaminated

pL (10-3 m

0c)

as-implanted

Dif

fere

nce

to b

ulk

Si (

a.u.

)

0 1 2 3 4200

250

300

350

400

450

RpR

p/2

τbulk

τ def (

ps)

Depth (µm)

60

80

100

after annealingCu contaminated

Inte

nsity

(%

)

� Doppler-coincidence shows the existence of Cu

at the annihilation site

� lifetime spectroscopy needed for size of open

volume

� at Rp/2: τd=450 ps

� at Rp: τd=320 ps

� lifetimes independent of

copper contamination

� intensities behave

different

� Conclusions

� Rp/2: small vacancy clusters are getter centres

� Rp: positrons are trapped by dislocation loops; trap density

decreases during Cu diffusion; no copper close to the positron

trap


Positron microprobe at LLNL

� Positrons generated by LINAC; twofold re-moderation and bunching system

� E+ = 1�50 keV

� spot about 1 µm

� 107 e+/s expected


Research reactorFRM-II, Munich

Photos: 4-July-2000 (source: internet)

experimental hall

reactor vessel moderator tank

construction site

� positron source by a nuclear

reaction: 113Cd(n,γ)114Cd

� three γ rays → pair production

� continuous positron beam

� ≈1010 slow e+/s expected


Positron source at an Free-Electron Laser

� electron beam at FEL is bunched (length: ≈ 1 ps repetition time: ≈ 100ns)

� beam energy: > 50 MeV power: > 50 kW

� two such systems under construction in Germany: DESY (Hamburg) & ELBE (Rossendorf)

p sn s ≈

� in September 1999: Workshop at DESY (40 participants / 11 European countries):

� Agreement: �EPOS � European Positron Source for Applied Research�

� Conceptual Report now available (done by 20 scientists / 8 countries)

p o ssib lee so u rce+

p ossib lee so urce+

Please visit our Poster: „EPOS - A European Positron Source for Applied Research “


Future developments

� positron beam technique now widely accepted as unique tool for defect

detection

� important: combination of Doppler broadening with positron lifetime

spectroscopy

� powerful: comparison Doppler-coincidence spectroscopy ⇔ theory

� new: - 3D-imaging of defects

- direct observation of defect kinetics in real time

� needed: - focused high-brightness positron beams operated as

user-dedicated facilities- more efficient moderators

This presentation can be found as pdf-file at our Webpage:

http://www.ep3.uni-halle.de/positrons

positron beam studies of defects in...

Documents