- Email: [email protected]
Issues in the ion implantation of Si for GaAs applications
Implantation
Feature
Issues in the ion implantation of Si for GaAs applications by Sonu Daryanani, Jim Jillsonand Ronald Eddy The ion implantation of Si is one of the key enabling processes involved in the fabrication of GaAs electronic and optical devices. In this article we discuss some of the issues involved from the ion implanter equipment perspective. A comprehensive study of the mass separation issues between the various isotopes of Si for various dose and energy levels is presented. This data illustrates the tradeoffs that exist between throughput and beam purity in the selection between the isotopes. The possible contaminants that could be introduced during the implantation are discussed, as well as techniques that could be used to reduce them. Data on multiple charged implants for deeper profiles is also presented, as well as equipment designs that have been made to improve the beam currents for these implants.
ilicon is the m o s t c o m m o n l y used n type dopant in GaAsbased devices. As device geometries scale d o w n to m e e t the requirements o f higher speeds, greater packing density, as well as lower power d i s s i p a t i o n , precise control o f the doping profile in all three dimensions of the substrate b e c o m e increasingly important. The ion implantation technique has the potential for providing precise three-dimensional dose control with high wafer throughputs and low contamination levels. In this paper we present some of the unique characteristics and requirements that are inv o l v e d in the ion i m p l a n t a t i o n o f silicon. T h e data p r e s e n t e d here is based on Varian's serial process, medium current implanter - the E H P 500 -- which has several features allowing production-worthy implantation for state-of-the-art GaAs devices.
S
Selection of the isotope for implantation Si has three isotopes, with masses of 28, 29, and 30. The most abundant isotope in natural occurrence is at mass 28 and exceeds the other two isotopes in abundance by over a factor of 20, while the ratio in the natural occurrence o f the mass 29 and 30 isotopes is approximately 4:3.
26
The EHP-5000 implanter used in production of GaAs devices.
T h r o u g h p u t considerations thus dictate the preference in the use o f the mass 28 isotope. Ion implantation of silicon at this mass however has the associated risk o f co-implantation of Nz + and CO +, both of which are also at mass 28. Contamination from these two species can be particularly harmful for low dose c h a n n e l i m p l a n t s where precision in the dose is required for threshold control. The interfering
III-Vs Review • Vol.10 No.2 1997 0961-1290/97/517.00©1997, Elsevier Science Ltd
species cause an under-dose both from the F a r a d a y m e a s u r e m e n t s y s t e m (which counts any positive ions traversing it as implanted dose), as well as from the potential of carbon counterdoping the GaAs. The risks involved in this contamination have made many GaAs device manufacturers opt for the use of 29Si+ instead. Figure 1 shows SIMS profiles o f nitrogen in two samples, one o f
Implantation
Feature
Figure 1. Nitrogen SIMS profile for samples ion implanted with 28Si+ and 29Si + .
Figure 2. Nitrogen and boron SIMS profiles for a 2aSi+ implant using boron nitride and alumina insulators.
Sl ; ~ U keY,b 1 3 ~
I
~
ol
o2
Figure 3. Nitrogen and boron SIMS profiles for a 29Si+ implant using boron nitride and alumina insulators.
Figure 4. Aluminum SIMS profiles for a 28Si+ implant using boron nitride and alumina insulators.
which was implanted with 288i+, and the other with the 298i + isotope. In this example the source used alumina insulators for the filament and repeller electrodes, and SiF4 was used as the source gas. The same beam current of 25 btAwas used for both implants. The sample implanted with 288i + clearly shows a nitrogen implant peak, well above the background level found in the sample implanted with29Si +. This indicates that most of the nitrogen has originated in the source region of the implanter, and has been extracted
and implanted at the same potential as the 2Ssi + beam.
28
Boron nitride versus alumina insulators The standard filament insulators on the EHP-500 implanter are made from boron nitride (BN). This material can cause some c o n t a m i n a t i o n for G a A s implants. Specifically, the boron in the BN could be etched off with the SiF 4 gas, and cause BF + to be produced.
IIl-Vs Review • Vol.10 No.2 1997 0961-1290/97/$17.00 ©1997, Elsevier Science Ltd
(I°B~9F)+ has the same mass as 29Si+, and could be a cause of dose shift and contamination. A solution to this problem is the use of alumina insulators. Figure 2 shows nitrogen and boron SIMS profiles obtained for a 2Ssi + implant using BN and alumina insulators. This implant was done at a beam e n e r g y o f 50 keV a n d a d o s e o f 6 x 1013cm -z. These plots show that the integrated nitrogen density is lower in the case of the alumina insulators by a factor of 0.53. The boron levels are h o w e v e r u n c h a n g e d , and no i m planted boron is detected in the profiles. Figure 3 shows the nitrogen and b o r o n SIMS profiles obtained for a 29Si + implant using BN and alumina insulators. This implant was done at a beam energy of 40 keV and a dose of 6 x 1013cm -2. These plots show that the integrated nitrogen density is lower in the case of the alumina insulators, although both are still below the measurement limits on the SIMS. No implanted boron is seen in either profile and there is very little difference in the profiles, which is probably all on the surface. The conclusion from these SIMS plots is that while the alumina insulators do allow reduction in the level of nitrogen that is implanted, especially in the 28Si+ case, the boron concentration is virtually unchanged, with no implanted b o r o n detected in either 28 .+ 29 .+ the $1 or the Sl Implants. This shows that at the dose and energy levels used for these implants, the BN insulators contribute insignificantly to any contamination during a 298i+ implant. The alumina insulators are, however, useful during the implantation of 288i+, in the reduction of the n i t r o g e n c o n c e n t r a t i o n t h a t is achieved. T h e i m p l a n t e d n i t r o g e n seems to originate both from the BN insulators as well as from residual nitrogen in the source chamber. Extensive p u r g i n g o f the source with an inert gas like argon could help in this regard. A n o t h e r potential source of nitrogen is the source t u r b o p u m p in which nitrogen is commonly used as the purge gas, and it is suggested that argon be used for this as well, in the application of 288i+ implantation.
Implantation
Feature
(a)
(b)
I ~INE
o.91
0.0
EXT-U ARC ~ FIL-U
60.0 2 33 1?% 43.8 7,0E-6
MAg-I ~
2 iI
' ' I . . . . .
~
.~rE . . aT~t~'~ . . .
/>
06 • 5 0.4 0.3 0,2 • i
~
BEAMLN 2.~-7 I
. . . . . . .
6~p~0 M~62460T 662~OAMPS I~MU6~O 70310
I
'
'
~
I
?220
ID 3
-
-
~
z 60 50 40
~.e 8 00 e ~.~
OFF
S
00,2
66 4
2
I
3
. ~
Mass separation
I
Ir "l
EXT-U ~ I : FZL-U
252 82 146 ~-Z 43.6 SOLRCE 9 F E 6
5
"--~
/
i
roTE 27m~-gs 4
]
I~I
'I. ]
- /
II
~ ...,
i
42,0
~OMmt~=
-~?
'
•
I'1
29Si+ beam
(=)
I
O~ e ~: e.@
C=F BEAtILN2.6E-? Z~
6
~
l
&4.8
SPECTRUM
29
ID
1 00
4
PEAKS
AM~ M~G-Z BEAM-I 29.~
43.4
46.~.
current of 80 #mA, and (b) at 29Si + beam current of 46 nA using the low
over the complete range of useable b e a m currents. Figure 5 shows spectra that were o b t a i n e d while r u n n i n g the implanter at two different 298i + b e a m currents and with different low b e a m reducer settings. T h e low b e a m reducer is an a p e r t u r e at the exit o f t h e source which can be selectively closed d o w n , hence p e r m i t t i n g smaller b e a m c u r r e n t s to b e used for i m p l a n t a t i o n w h i l e still r u n n i n g r e a s o n a b l y h i g h arc currents for stable plasma production in the source. F i g u r e 5(a) shows t h e s p e c t r u m o b t a i n e d for a 298i + b e a m current o f 80 HA with the reducer n o t used, while Figure 5(b) shows the s p e c t r u m at a 29Si + beam current o f 46.4 n A with the reducer in the n a r r o w position. These plots clearly show the excellent s e p a r a t i o n that is a c h i e v e d b e t w e e n the 298i + peak and the other
A critical issue in the use o f 298i + implants is the ability o f the implanter to clearly separate out the mass 29 peak from the m o r e a b u n d a n t mass 28 peak
I
J
=
"
, n a t i V E STRTOS
II ~'I
i'
-aO 10
OFF
AMU ~IO~-I EAM-I
Figure 5. Beam spectrum showing isotope separation. (a) at beam reducer.
A concern with the use o f a l u m i n a insulators is the p o s s i b i l i t y o f alumin u m o r its c o m p o u n d s b e i n g i m planted. F i g u r e 4 shows the A1 SIMS profiles o b t a i n e d following a 288i + implant at an energy o f 50 keVand a dose o f 6 x 1013cm -2 for b o t h BN and alumina insulators. T h e s e plots s h o w that the A1 is all on the surface and the integrated dose levels are b o t h very similar and are a r o u n d 4 x 101°cm -2 T h i s allows us to conclude that the alumina insulators d o n o t c o n t r i b u t e any significant A1 c o n t a m i n a t i o n for the dose and energy levels that were used.
{~R}
Si isotopes over a range o f beam currents. T h i s allows low dose i m p l a n t s to be a c h i e v e d in a v e r y r e p e a t a b l e and u n i f o r m fashion. Stable and reli able o p e r a t i o n at low dose rates allows c o n t r o l o f the i m p l a n t - i n d u c e d damage and i m p r o v e m e n t in d o n o r activation [1]. B e a m c u r r e n t s o f a few h u n d r e d HA are easily a c h i e v a b l e with 298i +, which gives the user the o p t i o n o f run n i n g this b e a m even for s o m e o f the h i g h dose i m p l a n t s , w h e r e 288i + has conventionally been used.
Multiple-charged implants M u l t i p l e - c h a r g e d i m p l a n t s are o f t e n u s e d to o b t a i n d e e p e r c h a n n e l and
Cb) 1;,10 I{
1ttl
"""
~
19,~
nR
Figure6. Integrated double- and sing/e-charged beam profiles using the in situ energy purity check with (a) the source magnet in the normal position. and (b) with the source magnet polarity reversed.
30
III-Vs Review• Vol.lO No.2 1997 0961-1290/97/$17.00©1997, ElsevierScience Ltd
ImplankrtionFeature
net polarity reversed. The 29Si++ beam is again measured at 52.33 PmA, but the single-charged contamination level has dropped dramatically to 0.07%. This implants
shows that double-charged can be run very reliably on
the EHP-500
implanter,
without
risk
of contamination.
Summary In conclusion, it has been shown that the mass 29 isotope of Si is preferable for reasons of contamination.
The im-
plantatio; at mass 29 requires good mass separation with the more abundant peak at mass 28 over the entire range of beam current operation. Magnetk Quadrupole
use of alumina insulators
The
is shown to
reduce nitrogen contamination in the case of 28Sif implantation. Doublecharged source
implants magnet
with the reversed polarity
have been
shown to improve the energy purity, allowing reliable and repeatable operation.
Acknowledgments A charge exchange reaction that occwa at two specilk reglons in the beamline physically separates two. low intenalty, single%b&rged beams (A and 6) im the main double-charged
The authors thank Mr Marcus Monell of Varian, IIS for his help with the ion implantation of the wafers, and Evans
besm 0.
East, NJ for their excellent SIMS charwell implants.
The
EHP-500
arc current supply on the new EHP-500 source allows achievement of much higher arc currents.‘“Si++ beam currents in excess of 70 PA have been obtained with this source.
has a
number of characteristics that allow production-worthy multiple-charged implant performance: 1.
An in-situ check of the energy purity in the implanter allows an exact determination of the level of the single-charge contamination when running double-charged beams. It has been verified by SIMS that the energy purity check mechanism on the EHP-500 is very accurate and ;he sensitivity sometimes exceeds that obtained by SIMS.
2.
An electrostatic beam filter placed just beyond the analyzer magnet deflects any half energy (singlecharged) beams produced by dissociation in the analyzer magnet away from the path of the double-charged beam.
3.
Use of a larger arc supply: a 15 amp
4.
Use of a reversing switch in the source magnet polarity allows for better fragmentation in the source plasma for multiple-charge performance [2]. This is illustrated in Figure 6. Figure 6(a) shows the integrated double-charged and the contaminant single-charged beam when running a 29Si++ beam at 5 A of source arc current with the source magnet polarity in the normal position. The 29Si+f beam is measured at 50.89 PA, and the level of the single-charge beam at 0.54%. Figure 6(b) shows a plot of these beams with the same source settings, but with the source mag-
acterization.
References [l] T. E. Haynes and 0. W. Holland, Comparative study of implantation induced damage in GaAs and Ge:temperature and flux dependence, Applied PhysicsLetrers, 50 (4) (1991) 452-454. [2] D. R. Swenson, A. Renau, S. R. Walther and M. E. Mack Enhanced Bernas Source for thevarian EHP-500 Medium Current Ion Implanter, 11th InternationalConferenceon Ion Implanterl&hnology,A&in, Zxr, June 1996. Contact: Sonu Daryanani Varian Ion Implant Systems, 35 Dory Rd., Gloucester, MA 01930, USA Tel/fax: +1508-282 2985/283 5391. E-mail: [email protected]
Ill-Vs Review l Vol.10 No.2 1997 0961-1290/97/$17.00 01997, Elsevier Science Ltd
31
Feature
Issues in the ion implantation of Si for GaAs applications by Sonu Daryanani, Jim Jillsonand Ronald Eddy The ion implantation of Si is one of the key enabling processes involved in the fabrication of GaAs electronic and optical devices. In this article we discuss some of the issues involved from the ion implanter equipment perspective. A comprehensive study of the mass separation issues between the various isotopes of Si for various dose and energy levels is presented. This data illustrates the tradeoffs that exist between throughput and beam purity in the selection between the isotopes. The possible contaminants that could be introduced during the implantation are discussed, as well as techniques that could be used to reduce them. Data on multiple charged implants for deeper profiles is also presented, as well as equipment designs that have been made to improve the beam currents for these implants.
ilicon is the m o s t c o m m o n l y used n type dopant in GaAsbased devices. As device geometries scale d o w n to m e e t the requirements o f higher speeds, greater packing density, as well as lower power d i s s i p a t i o n , precise control o f the doping profile in all three dimensions of the substrate b e c o m e increasingly important. The ion implantation technique has the potential for providing precise three-dimensional dose control with high wafer throughputs and low contamination levels. In this paper we present some of the unique characteristics and requirements that are inv o l v e d in the ion i m p l a n t a t i o n o f silicon. T h e data p r e s e n t e d here is based on Varian's serial process, medium current implanter - the E H P 500 -- which has several features allowing production-worthy implantation for state-of-the-art GaAs devices.
S
Selection of the isotope for implantation Si has three isotopes, with masses of 28, 29, and 30. The most abundant isotope in natural occurrence is at mass 28 and exceeds the other two isotopes in abundance by over a factor of 20, while the ratio in the natural occurrence o f the mass 29 and 30 isotopes is approximately 4:3.
26
The EHP-5000 implanter used in production of GaAs devices.
T h r o u g h p u t considerations thus dictate the preference in the use o f the mass 28 isotope. Ion implantation of silicon at this mass however has the associated risk o f co-implantation of Nz + and CO +, both of which are also at mass 28. Contamination from these two species can be particularly harmful for low dose c h a n n e l i m p l a n t s where precision in the dose is required for threshold control. The interfering
III-Vs Review • Vol.10 No.2 1997 0961-1290/97/517.00©1997, Elsevier Science Ltd
species cause an under-dose both from the F a r a d a y m e a s u r e m e n t s y s t e m (which counts any positive ions traversing it as implanted dose), as well as from the potential of carbon counterdoping the GaAs. The risks involved in this contamination have made many GaAs device manufacturers opt for the use of 29Si+ instead. Figure 1 shows SIMS profiles o f nitrogen in two samples, one o f
Implantation
Feature
Figure 1. Nitrogen SIMS profile for samples ion implanted with 28Si+ and 29Si + .
Figure 2. Nitrogen and boron SIMS profiles for a 2aSi+ implant using boron nitride and alumina insulators.
Sl ; ~ U keY,b 1 3 ~
I
~
ol
o2
Figure 3. Nitrogen and boron SIMS profiles for a 29Si+ implant using boron nitride and alumina insulators.
Figure 4. Aluminum SIMS profiles for a 28Si+ implant using boron nitride and alumina insulators.
which was implanted with 288i+, and the other with the 298i + isotope. In this example the source used alumina insulators for the filament and repeller electrodes, and SiF4 was used as the source gas. The same beam current of 25 btAwas used for both implants. The sample implanted with 288i + clearly shows a nitrogen implant peak, well above the background level found in the sample implanted with29Si +. This indicates that most of the nitrogen has originated in the source region of the implanter, and has been extracted
and implanted at the same potential as the 2Ssi + beam.
28
Boron nitride versus alumina insulators The standard filament insulators on the EHP-500 implanter are made from boron nitride (BN). This material can cause some c o n t a m i n a t i o n for G a A s implants. Specifically, the boron in the BN could be etched off with the SiF 4 gas, and cause BF + to be produced.
IIl-Vs Review • Vol.10 No.2 1997 0961-1290/97/$17.00 ©1997, Elsevier Science Ltd
(I°B~9F)+ has the same mass as 29Si+, and could be a cause of dose shift and contamination. A solution to this problem is the use of alumina insulators. Figure 2 shows nitrogen and boron SIMS profiles obtained for a 2Ssi + implant using BN and alumina insulators. This implant was done at a beam e n e r g y o f 50 keV a n d a d o s e o f 6 x 1013cm -z. These plots show that the integrated nitrogen density is lower in the case of the alumina insulators by a factor of 0.53. The boron levels are h o w e v e r u n c h a n g e d , and no i m planted boron is detected in the profiles. Figure 3 shows the nitrogen and b o r o n SIMS profiles obtained for a 29Si + implant using BN and alumina insulators. This implant was done at a beam energy of 40 keV and a dose of 6 x 1013cm -2. These plots show that the integrated nitrogen density is lower in the case of the alumina insulators, although both are still below the measurement limits on the SIMS. No implanted boron is seen in either profile and there is very little difference in the profiles, which is probably all on the surface. The conclusion from these SIMS plots is that while the alumina insulators do allow reduction in the level of nitrogen that is implanted, especially in the 28Si+ case, the boron concentration is virtually unchanged, with no implanted b o r o n detected in either 28 .+ 29 .+ the $1 or the Sl Implants. This shows that at the dose and energy levels used for these implants, the BN insulators contribute insignificantly to any contamination during a 298i+ implant. The alumina insulators are, however, useful during the implantation of 288i+, in the reduction of the n i t r o g e n c o n c e n t r a t i o n t h a t is achieved. T h e i m p l a n t e d n i t r o g e n seems to originate both from the BN insulators as well as from residual nitrogen in the source chamber. Extensive p u r g i n g o f the source with an inert gas like argon could help in this regard. A n o t h e r potential source of nitrogen is the source t u r b o p u m p in which nitrogen is commonly used as the purge gas, and it is suggested that argon be used for this as well, in the application of 288i+ implantation.
Implantation
Feature
(a)
(b)
I ~INE
o.91
0.0
EXT-U ARC ~ FIL-U
60.0 2 33 1?% 43.8 7,0E-6
MAg-I ~
2 iI
' ' I . . . . .
~
.~rE . . aT~t~'~ . . .
/>
06 • 5 0.4 0.3 0,2 • i
~
BEAMLN 2.~-7 I
. . . . . . .
6~p~0 M~62460T 662~OAMPS I~MU6~O 70310
I
'
'
~
I
?220
ID 3
-
-
~
z 60 50 40
~.e 8 00 e ~.~
OFF
S
00,2
66 4
2
I
3
. ~
Mass separation
I
Ir "l
EXT-U ~ I : FZL-U
252 82 146 ~-Z 43.6 SOLRCE 9 F E 6
5
"--~
/
i
roTE 27m~-gs 4
]
I~I
'I. ]
- /
II
~ ...,
i
42,0
~OMmt~=
-~?
'
•
I'1
29Si+ beam
(=)
I
O~ e ~: e.@
C=F BEAtILN2.6E-? Z~
6
~
l
&4.8
SPECTRUM
29
ID
1 00
4
PEAKS
AM~ M~G-Z BEAM-I 29.~
43.4
46.~.
current of 80 #mA, and (b) at 29Si + beam current of 46 nA using the low
over the complete range of useable b e a m currents. Figure 5 shows spectra that were o b t a i n e d while r u n n i n g the implanter at two different 298i + b e a m currents and with different low b e a m reducer settings. T h e low b e a m reducer is an a p e r t u r e at the exit o f t h e source which can be selectively closed d o w n , hence p e r m i t t i n g smaller b e a m c u r r e n t s to b e used for i m p l a n t a t i o n w h i l e still r u n n i n g r e a s o n a b l y h i g h arc currents for stable plasma production in the source. F i g u r e 5(a) shows t h e s p e c t r u m o b t a i n e d for a 298i + b e a m current o f 80 HA with the reducer n o t used, while Figure 5(b) shows the s p e c t r u m at a 29Si + beam current o f 46.4 n A with the reducer in the n a r r o w position. These plots clearly show the excellent s e p a r a t i o n that is a c h i e v e d b e t w e e n the 298i + peak and the other
A critical issue in the use o f 298i + implants is the ability o f the implanter to clearly separate out the mass 29 peak from the m o r e a b u n d a n t mass 28 peak
I
J
=
"
, n a t i V E STRTOS
II ~'I
i'
-aO 10
OFF
AMU ~IO~-I EAM-I
Figure 5. Beam spectrum showing isotope separation. (a) at beam reducer.
A concern with the use o f a l u m i n a insulators is the p o s s i b i l i t y o f alumin u m o r its c o m p o u n d s b e i n g i m planted. F i g u r e 4 shows the A1 SIMS profiles o b t a i n e d following a 288i + implant at an energy o f 50 keVand a dose o f 6 x 1013cm -2 for b o t h BN and alumina insulators. T h e s e plots s h o w that the A1 is all on the surface and the integrated dose levels are b o t h very similar and are a r o u n d 4 x 101°cm -2 T h i s allows us to conclude that the alumina insulators d o n o t c o n t r i b u t e any significant A1 c o n t a m i n a t i o n for the dose and energy levels that were used.
{~R}
Si isotopes over a range o f beam currents. T h i s allows low dose i m p l a n t s to be a c h i e v e d in a v e r y r e p e a t a b l e and u n i f o r m fashion. Stable and reli able o p e r a t i o n at low dose rates allows c o n t r o l o f the i m p l a n t - i n d u c e d damage and i m p r o v e m e n t in d o n o r activation [1]. B e a m c u r r e n t s o f a few h u n d r e d HA are easily a c h i e v a b l e with 298i +, which gives the user the o p t i o n o f run n i n g this b e a m even for s o m e o f the h i g h dose i m p l a n t s , w h e r e 288i + has conventionally been used.
Multiple-charged implants M u l t i p l e - c h a r g e d i m p l a n t s are o f t e n u s e d to o b t a i n d e e p e r c h a n n e l and
Cb) 1;,10 I{
1ttl
"""
~
19,~
nR
Figure6. Integrated double- and sing/e-charged beam profiles using the in situ energy purity check with (a) the source magnet in the normal position. and (b) with the source magnet polarity reversed.
30
III-Vs Review• Vol.lO No.2 1997 0961-1290/97/$17.00©1997, ElsevierScience Ltd
ImplankrtionFeature
net polarity reversed. The 29Si++ beam is again measured at 52.33 PmA, but the single-charged contamination level has dropped dramatically to 0.07%. This implants
shows that double-charged can be run very reliably on
the EHP-500
implanter,
without
risk
of contamination.
Summary In conclusion, it has been shown that the mass 29 isotope of Si is preferable for reasons of contamination.
The im-
plantatio; at mass 29 requires good mass separation with the more abundant peak at mass 28 over the entire range of beam current operation. Magnetk Quadrupole
use of alumina insulators
The
is shown to
reduce nitrogen contamination in the case of 28Sif implantation. Doublecharged source
implants magnet
with the reversed polarity
have been
shown to improve the energy purity, allowing reliable and repeatable operation.
Acknowledgments A charge exchange reaction that occwa at two specilk reglons in the beamline physically separates two. low intenalty, single%b&rged beams (A and 6) im the main double-charged
The authors thank Mr Marcus Monell of Varian, IIS for his help with the ion implantation of the wafers, and Evans
besm 0.
East, NJ for their excellent SIMS charwell implants.
The
EHP-500
arc current supply on the new EHP-500 source allows achievement of much higher arc currents.‘“Si++ beam currents in excess of 70 PA have been obtained with this source.
has a
number of characteristics that allow production-worthy multiple-charged implant performance: 1.
An in-situ check of the energy purity in the implanter allows an exact determination of the level of the single-charge contamination when running double-charged beams. It has been verified by SIMS that the energy purity check mechanism on the EHP-500 is very accurate and ;he sensitivity sometimes exceeds that obtained by SIMS.
2.
An electrostatic beam filter placed just beyond the analyzer magnet deflects any half energy (singlecharged) beams produced by dissociation in the analyzer magnet away from the path of the double-charged beam.
3.
Use of a larger arc supply: a 15 amp
4.
Use of a reversing switch in the source magnet polarity allows for better fragmentation in the source plasma for multiple-charge performance [2]. This is illustrated in Figure 6. Figure 6(a) shows the integrated double-charged and the contaminant single-charged beam when running a 29Si++ beam at 5 A of source arc current with the source magnet polarity in the normal position. The 29Si+f beam is measured at 50.89 PA, and the level of the single-charge beam at 0.54%. Figure 6(b) shows a plot of these beams with the same source settings, but with the source mag-
acterization.
References [l] T. E. Haynes and 0. W. Holland, Comparative study of implantation induced damage in GaAs and Ge:temperature and flux dependence, Applied PhysicsLetrers, 50 (4) (1991) 452-454. [2] D. R. Swenson, A. Renau, S. R. Walther and M. E. Mack Enhanced Bernas Source for thevarian EHP-500 Medium Current Ion Implanter, 11th InternationalConferenceon Ion Implanterl&hnology,A&in, Zxr, June 1996. Contact: Sonu Daryanani Varian Ion Implant Systems, 35 Dory Rd., Gloucester, MA 01930, USA Tel/fax: +1508-282 2985/283 5391. E-mail: [email protected]
Ill-Vs Review l Vol.10 No.2 1997 0961-1290/97/$17.00 01997, Elsevier Science Ltd
31