- Email: [email protected]
Formation mechanisms of diamond precursors during the nucleation process
ELSEVIER
Diamond and Related Materials 6 (1997) 1047-1050
D|AMOND AND RE TED TER|ALS
Formation mechanisms of diamond precursors during the nucleation process S. Yugo *, K. Semoto, N. Nakamura, T. Kimura, H. Nakai, M. H a s h i m o t o University of Electro-Communications, 1-5-1 Chofugaoka, Clu~l-shi. Tokyo. 182. Japan Received 9 September 1996; accepted 9 December 1996
Abstract The aim of this report is to reconsider our diamond nucleation model in light of the many reports published recently. We found that our model agrees with most of the recent reports; however, it does not agree with some of them with respect to the kinetics of nucleus f o ~ a t i o n . This disagreement stems from the question of whether or not a nuclevs precursor should be treated as an embryonic cluster which is a small critical nucl2us. © 1997 Elsevier Science S.A.
Keyword~: Bias; CVD; Diamond: Nucleation
1. introduction
2. Nucleation model from the bias effect point of view
Many of the reports published so thr have outlined the diamond nucleation mechanism as follows. First, an activated species (high-energy carbon atoms or hydrocarbons) which is supplied from plasma migrates on a substrate and forms a cluster at a stable site. This process is achieved with much difficulty, because the cluster produced tends to evaporate due to the presence of high-density hydrogen plasma and the high-temperature substrate. In order to overcome this problem, the carbon concentration in the raw material is increased or the substrate surface is scratched. Also, substrates with high carbon affinity are selected to stabilize the diamond nucleus. The cluster produced is transtbrmed into the diamond nucleus by plasma energy. As we explained, the diamond nucleation mechanism is generally clear; however, the structure of the precursor and tl~e mechanism ef the transition from an amorphous carbon cluster to a dim~ond nucleus remain unsolved. In the present ~eport, we reconsider our diamond nucleation model by co,.moating it with many recent findings.
We have developed the bias processing method as one of the substrate processing methods for accelerating diamond nucleation and for determining the mechanism of diamond nucleation [I 4 ] . Fig. 1 shows a schematic diagram of our diamond nucleation model that takes the ion irradiation effe~,t into consideration. In the presence of applied bias, ions activated by energy in a plasma sheath undergo both internal diffusion into and surface migration on the substrate. Most activated species evaporate due to high-temperature hydrogen plasma, and some of the activated species induce carbide migration on the substrate surface and form a cluster by mixing with the substrate. The structure of the cluster formed at the beginning of this orocess is not completely clear; however, it is considered to be a hydrogenated spZ-rich amorphous carbon cluster, which can be regarded as a precursor of the diamond nucleus. These clusters are irradiated with low-energy ions, and amorphous components are removed. At the same time, sp 2 bonds are transformed sequentially to sp 3 bonds. After repeated transformation, an sp3-rich cluster grows into a critical nucleus of 1-2 nm size, which is a diamond nucleus. A high concentration of hydrogen in plasma may induce the removal of amorphous carbon and prevent sp 2 bond formation at carbon terminal bonds, judging from hydrogen assisted diamond nucleation. The explanation men-
* Corresponding author. Fax: + 81 424803801; e-mail: [email protected] tPaper presented at the 7th European Conference on Diamo~ad, Diamond-like and Related Materials, Tours, 8-13 September 1996.
0925-9635/97/$17.00 Copyright © 1997 Elsevier Science S.A. All rights reserved I'll S0925-9635(97)00003-4
1048
X
Yugo
ct aL Dhonond and Rehtted lthaterials 6 (1997) 1047-1050
•P l a s m a
diffusion
• p
:~
i
..
"
(~,.
".
men
'~'1
.
.
. ' .' ' .
ion sheath
~
~.
.
g
"
,
j
sputtering H, etching
.
sp2-evaporation sputtering @
H2etchin~-L-~.. ~s~c
,~. •
~
ion bombardment
i T
( ....
~
sorsi
l
.~ ~
~
!transformation ,'from sp2 to sp3
Fig. I. A former model of" the generation of diamond nuclei by ion impinging effects.
tioned above outlines our diamond nucleation model based on the bias effect.
3. Reports on diamond nucleation regarding bias effect
Recently, many reports on diamond nucleation including the bias effect have been published. The following items regarding diamond nucleation have been selected from recent publications. (1) Effective ion energy [5-8]. (2) Acceleration of migration and carbonization reaction [9-13].
(3) Transition from carbon
[14-181.
sp 2
bonds
to
sp 3
bonds
(4) Subplantation effect of ions [1%20]. (5) Stress effect and excess h,:at generation effect [21-23]. (6) Graphite sheet coagulation effect and diamond generation from graphite edge [24-26]. (7) Removal of amorphous sp2 component by hydrogen and prevention of carbon double bond formation [27-30]. (8) Effect of substrate temperature [31 ]. (9) Secondary electron effect [32]. Some of these items agree with our model; however,
S. Yugo et al. / Diamond and Related Materials 6 (1997) 1047-1050
some of them disagree such as the structure of the precursor and the transition mechanism from carbon sp 2 bonds t o sp 3 bonds. We will reconsider our diamond nucleation model by analyzing the items listed above.
4. Discussion
Regarding item ( l ), generally speaking, the bias effect is observed in the bias voltage range of about - 5 0 to - 2 0 0 V. The effective ion energy ranges from a few eV to a few tenths eV, depending on the equipment used. Low-energy ions may exert some chemical effect such as cleaning of the substrate, surface migration and accelerating reaction, whereas high-energy ions may exert some physical effect such as subplantation and selective etching. This bias method is based on the compound effects of ions with various energies. Acceleration of migration by bias application (item (2)), is also observed in our experiments on submicron selective growth [33], and acceleration of carbide formation is also observed by transmission electron microscopy (TEM) [4]. These effects induce speedy cluster formation and accelerate diamond nucleation. Knowledge of the initial cluster structure is necessary for discussion of item (3). We concluded that the initial cluster is mainly composed of hydrogenated amorphous carbon (including graphite, aromatic and diamond structures) by TEM observation [2], and that it is a diamond precursor. Many reports agree with our conclusion: however, verification is difficult. The most critical question is how the diamond precursor is transformed into a diamond nucleus. The initial cluster is an embryonic cluster for the small cluster: theref ~t'e a process which differs from the reaction of a solid cluster should be considered. The ion effect with lower ion energy is effective for the embryonic cluster compared to that for the solid cluster; therefore, embryonic clusters tend to be more active and reach a stable site relatively easily. This process actively promotes dehydrogenation and transformation from carbon sp 2 bonds t o sp 3 bonds. For this reason, the subplantation effect of ions (item (4)) is considered to be relatively small. Also, the stress effect and excess heat generation effect (item (5)) in the cluster are small; they cannot act as a motivating tbrce for diamond nucleation. When the cluster is graphitized, the possibilities of the graphite sheet coagulation effect and diamond generation from the graphite edge (item (6)) cannot be denied; however, from TEM of microsized diamond particles in amorphous carbon, no transition from graphite crystal to diamond was observed. Therefore, random transition from sp 2 bonds t o sp 3 bonds on the cluster surface, rather than the processes explained in item (6)), mainly account for the mechanism of diamond nucleation.
1049
The nucleation acceleration effect of hydrogen (item (7)) is also observed in our experiment using argon gas instead of hydrogen [2]; no contrasting views have been reported thus far. In order to substitute hydrogen terminals with carbon, a relatively high-temperature substrate is necessary (item (8)). This is proof of the hydrogen effect. No mention of item (9) was made in our previous reports. Although this effect cannot be ignored in diamond which has negative electron affinity, it may not be the most important factor for diamond nucleation judging from the short nucleation time and the report of McGinnis et al. [34]. Finally, Fig. 2 shows a schematic diagram of our modified model that indicates the start of nucleation using an amorphous carbon cluster as the host. However, experimental demonstration of this model is very difficult, since the crystal nucleus is very small and is an embryonic cluster. Furthermore, in experiments,
sp 2 rich d u s t e r
ion
bombardment
sp 3
rich cluster(embryonic)
/
SiC 1 (ion n ~ n g )
Diamond ~a-carbon~
iayor
4
Fig. 2. A modified model of the generation of diamond nuclei by ion impinging effects.
1050
S. Yugo et al. / Diamond and Related MateriaL~ 6 (1997) 1047-1050
we observed results of completed processes but not the nucleation process itself, and this also contributes to difficulty in experimental demonstration of this model.
References [ 1] S. Yugo, T. Kanai, T. Kimura and T. Muto, Appl. Phys. Lett., 58, 11 (1991) 1036. [2] S. Yugo, T. Kanai and T. Kimura, Dhonond Relat. Mater., 1 (1992) 388. [3] S. Yugo, T. Kimura and T. Kanai, Diamond Relat. Mater., 2 ( 1993 ) 328. [4] S. Yugo, K. Semoto, K. Floshina, T. Kimura and H. Nakai, Diamond Relat. Mater., 4 (1995) 903. [5] D.R. McKenzie, D. Muller and B.A. Pailthorpe, Ph),s. Rev. Lett., 67(6) ( 1991 ) 733. [6] B.R. Stoner, B.E. Williams, S.D. Wolter, K. Nishiyama and J.T. Glass, J. Maters. Res., 7 (1992) 257. [7] X. Jiang and C.-P. Klages, AppL Phys. Lett., 65 (1993) 1519. [8] J. Gerber, S. Sattel, K. Jung, H. Ehrhardt and J. Robertson, Diamond Relat. Mater., 4 (1995) 559. [9] B.R. Stoner, G.-H. Ma, S.D. Wolter and J.T. Glass, Phys. Rev. B, 45, 19 (1992) 11 067. [10] R. Kohl, C. Wild, N. Herres, P. Koidl, B.R. Stoner and J.T. Glass, Appl. Phys. Lett., 63 (1993) 1203. [11] X. Jiang, K. Schiffmann and C.-P. Klages, Phys. Rev. B, 50, 12 (1994) 8402. [12] S.D. Wolter, B.R. Stoner, J.T. Glass, P.J. Ellis, D.S. Buhaenko, C.E. Jenkins and P. Southworth, Appl. Phys. Left., 62 (1993) 1215. [13] W. Kulisch, B. Sobisch, M. Kuhr and R. Beckmann, Dhmtond. Relat. Ma;er., 4 (1995) 401. [14] W. Moiler, Appl. Phys. Lett., 59. 4 ( 1991 ) 2391. [15] J. Singh and M. Vellaika, J. Appl. Phys., 73, 15 ( 1993} 2831. [16] Y. Lifshitz, G.D. Lempert, S. Rotter. I. Avigal, C. Uzan-Saguy,
R. Kalish, J. Kulik, D. Marton and J.W. Rabalais, Diamond Relat. Mater., 3 (1994) 542. [17] J. Gerber, S. Sattel, K. Jung, H. Ehrhardt and J. Robertson, Diamond Relat. Mater'., 4 (1995) 559. [18] Y. Ma, T. Tsurumi, N. Shinoda and O. Fukunaga, Diamond Relat. Mater., 4 ( 1995 ) 1325. [19] Y. Lifshitz, S.R. Kasi and J.W. Rabalais, Phys. Rev. Lett., 62, I1 (1989) 1290. [20] J. Robertson, J. Gerber, S. Sattel, M. Weiler, K. Jung and H. Ehhardt, Appl. Phys. Lett., 66, 12 (1995) 3287. [21] J. Robertson, Diamond Relat. Mater., 2 (1993) 984. [22] J. Gerber, M. Weiler, O. Sohr, K. Jung and H. Ehrhardt, Diamond Relat. Mater., 3 (1994) 506. [23] P. Deak, A, Gali, G. Sczigel and H. Ehrhardt, Diamond Relat. Mater., 4 (1995) 706. [24] E. Sandre, A. Bonnot and F. Cyrot-Lackman, Diamond Relat. Mater., 3 (1994) 448. [25] J.C. Angus, Z. Li, M. Sunkara, Y. Wang, C. Lee, W.R. Lambrecht and B. SegaU, 2nd lnt. Col!f. AppL Dia. Fihns and Rel. Mat., MYU, Tokyo, 1993 p. 153. [26] J. Robertson, Diamond. Relat. Mater'., 4 (1995) 549. [27] R. McKenzie, D. Muller, B.A. Pailthorpe, Z.H. Wang, E. Kravtchinskaia, D. Segal, P.B. Lukins, P.D. Swift, P.J. Martin G. Amaratunga, P.H. Gaskell and A. Saeed, Diamond Relat. Mater., 1 ( 1991 ) 51. [28] Y. Shigesato, R.E. Boekenhauer and B.W. Seldom Appl. Phys. Lett., 63 (1993) 314. [29] B.W. Seldom R. Csencsits, J. Rankin, E.R. Boekenhauer and Y. Shigesato, J. Appl. Phys., 75 (1994) 5001. [30] W. Piekarczyk and S. Prawer, Diamond Relat. Mater., 2 (1993~ 41. [31] S. Sattel, M. Weiler, J. Gerber, T. Giessen, H. Roth, M. Scheib. K. Jung and H. Ehrhardt, Diamond Relat. Mater., 4 (1995) 33~. [32] M. Schreck, T. Baur and B. Stritzker, Diamond Relat. Mater., 4 (1995) 553. [33] S. Katsumata and S. Yugo, DhI. Films Technol., 3, 4 (1994) 19'9. [34] S.P. McGinnis, M.A. Kelly and S.B. Hagstrom, AppL Phys. LeH., 66, 5 (1995) 3117.
Diamond and Related Materials 6 (1997) 1047-1050
D|AMOND AND RE TED TER|ALS
Formation mechanisms of diamond precursors during the nucleation process S. Yugo *, K. Semoto, N. Nakamura, T. Kimura, H. Nakai, M. H a s h i m o t o University of Electro-Communications, 1-5-1 Chofugaoka, Clu~l-shi. Tokyo. 182. Japan Received 9 September 1996; accepted 9 December 1996
Abstract The aim of this report is to reconsider our diamond nucleation model in light of the many reports published recently. We found that our model agrees with most of the recent reports; however, it does not agree with some of them with respect to the kinetics of nucleus f o ~ a t i o n . This disagreement stems from the question of whether or not a nuclevs precursor should be treated as an embryonic cluster which is a small critical nucl2us. © 1997 Elsevier Science S.A.
Keyword~: Bias; CVD; Diamond: Nucleation
1. introduction
2. Nucleation model from the bias effect point of view
Many of the reports published so thr have outlined the diamond nucleation mechanism as follows. First, an activated species (high-energy carbon atoms or hydrocarbons) which is supplied from plasma migrates on a substrate and forms a cluster at a stable site. This process is achieved with much difficulty, because the cluster produced tends to evaporate due to the presence of high-density hydrogen plasma and the high-temperature substrate. In order to overcome this problem, the carbon concentration in the raw material is increased or the substrate surface is scratched. Also, substrates with high carbon affinity are selected to stabilize the diamond nucleus. The cluster produced is transtbrmed into the diamond nucleus by plasma energy. As we explained, the diamond nucleation mechanism is generally clear; however, the structure of the precursor and tl~e mechanism ef the transition from an amorphous carbon cluster to a dim~ond nucleus remain unsolved. In the present ~eport, we reconsider our diamond nucleation model by co,.moating it with many recent findings.
We have developed the bias processing method as one of the substrate processing methods for accelerating diamond nucleation and for determining the mechanism of diamond nucleation [I 4 ] . Fig. 1 shows a schematic diagram of our diamond nucleation model that takes the ion irradiation effe~,t into consideration. In the presence of applied bias, ions activated by energy in a plasma sheath undergo both internal diffusion into and surface migration on the substrate. Most activated species evaporate due to high-temperature hydrogen plasma, and some of the activated species induce carbide migration on the substrate surface and form a cluster by mixing with the substrate. The structure of the cluster formed at the beginning of this orocess is not completely clear; however, it is considered to be a hydrogenated spZ-rich amorphous carbon cluster, which can be regarded as a precursor of the diamond nucleus. These clusters are irradiated with low-energy ions, and amorphous components are removed. At the same time, sp 2 bonds are transformed sequentially to sp 3 bonds. After repeated transformation, an sp3-rich cluster grows into a critical nucleus of 1-2 nm size, which is a diamond nucleus. A high concentration of hydrogen in plasma may induce the removal of amorphous carbon and prevent sp 2 bond formation at carbon terminal bonds, judging from hydrogen assisted diamond nucleation. The explanation men-
* Corresponding author. Fax: + 81 424803801; e-mail: [email protected] tPaper presented at the 7th European Conference on Diamo~ad, Diamond-like and Related Materials, Tours, 8-13 September 1996.
0925-9635/97/$17.00 Copyright © 1997 Elsevier Science S.A. All rights reserved I'll S0925-9635(97)00003-4
1048
X
Yugo
ct aL Dhonond and Rehtted lthaterials 6 (1997) 1047-1050
•P l a s m a
diffusion
• p
:~
i
..
"
(~,.
".
men
'~'1
.
.
. ' .' ' .
ion sheath
~
~.
.
g
"
,
j
sputtering H, etching
.
sp2-evaporation sputtering @
H2etchin~-L-~.. ~s~c
,~. •
~
ion bombardment
i T
( ....
~
sorsi
l
.~ ~
~
!transformation ,'from sp2 to sp3
Fig. I. A former model of" the generation of diamond nuclei by ion impinging effects.
tioned above outlines our diamond nucleation model based on the bias effect.
3. Reports on diamond nucleation regarding bias effect
Recently, many reports on diamond nucleation including the bias effect have been published. The following items regarding diamond nucleation have been selected from recent publications. (1) Effective ion energy [5-8]. (2) Acceleration of migration and carbonization reaction [9-13].
(3) Transition from carbon
[14-181.
sp 2
bonds
to
sp 3
bonds
(4) Subplantation effect of ions [1%20]. (5) Stress effect and excess h,:at generation effect [21-23]. (6) Graphite sheet coagulation effect and diamond generation from graphite edge [24-26]. (7) Removal of amorphous sp2 component by hydrogen and prevention of carbon double bond formation [27-30]. (8) Effect of substrate temperature [31 ]. (9) Secondary electron effect [32]. Some of these items agree with our model; however,
S. Yugo et al. / Diamond and Related Materials 6 (1997) 1047-1050
some of them disagree such as the structure of the precursor and the transition mechanism from carbon sp 2 bonds t o sp 3 bonds. We will reconsider our diamond nucleation model by analyzing the items listed above.
4. Discussion
Regarding item ( l ), generally speaking, the bias effect is observed in the bias voltage range of about - 5 0 to - 2 0 0 V. The effective ion energy ranges from a few eV to a few tenths eV, depending on the equipment used. Low-energy ions may exert some chemical effect such as cleaning of the substrate, surface migration and accelerating reaction, whereas high-energy ions may exert some physical effect such as subplantation and selective etching. This bias method is based on the compound effects of ions with various energies. Acceleration of migration by bias application (item (2)), is also observed in our experiments on submicron selective growth [33], and acceleration of carbide formation is also observed by transmission electron microscopy (TEM) [4]. These effects induce speedy cluster formation and accelerate diamond nucleation. Knowledge of the initial cluster structure is necessary for discussion of item (3). We concluded that the initial cluster is mainly composed of hydrogenated amorphous carbon (including graphite, aromatic and diamond structures) by TEM observation [2], and that it is a diamond precursor. Many reports agree with our conclusion: however, verification is difficult. The most critical question is how the diamond precursor is transformed into a diamond nucleus. The initial cluster is an embryonic cluster for the small cluster: theref ~t'e a process which differs from the reaction of a solid cluster should be considered. The ion effect with lower ion energy is effective for the embryonic cluster compared to that for the solid cluster; therefore, embryonic clusters tend to be more active and reach a stable site relatively easily. This process actively promotes dehydrogenation and transformation from carbon sp 2 bonds t o sp 3 bonds. For this reason, the subplantation effect of ions (item (4)) is considered to be relatively small. Also, the stress effect and excess heat generation effect (item (5)) in the cluster are small; they cannot act as a motivating tbrce for diamond nucleation. When the cluster is graphitized, the possibilities of the graphite sheet coagulation effect and diamond generation from the graphite edge (item (6)) cannot be denied; however, from TEM of microsized diamond particles in amorphous carbon, no transition from graphite crystal to diamond was observed. Therefore, random transition from sp 2 bonds t o sp 3 bonds on the cluster surface, rather than the processes explained in item (6)), mainly account for the mechanism of diamond nucleation.
1049
The nucleation acceleration effect of hydrogen (item (7)) is also observed in our experiment using argon gas instead of hydrogen [2]; no contrasting views have been reported thus far. In order to substitute hydrogen terminals with carbon, a relatively high-temperature substrate is necessary (item (8)). This is proof of the hydrogen effect. No mention of item (9) was made in our previous reports. Although this effect cannot be ignored in diamond which has negative electron affinity, it may not be the most important factor for diamond nucleation judging from the short nucleation time and the report of McGinnis et al. [34]. Finally, Fig. 2 shows a schematic diagram of our modified model that indicates the start of nucleation using an amorphous carbon cluster as the host. However, experimental demonstration of this model is very difficult, since the crystal nucleus is very small and is an embryonic cluster. Furthermore, in experiments,
sp 2 rich d u s t e r
ion
bombardment
sp 3
rich cluster(embryonic)
/
SiC 1 (ion n ~ n g )
Diamond ~a-carbon~
iayor
4
Fig. 2. A modified model of the generation of diamond nuclei by ion impinging effects.
1050
S. Yugo et al. / Diamond and Related MateriaL~ 6 (1997) 1047-1050
we observed results of completed processes but not the nucleation process itself, and this also contributes to difficulty in experimental demonstration of this model.
References [ 1] S. Yugo, T. Kanai, T. Kimura and T. Muto, Appl. Phys. Lett., 58, 11 (1991) 1036. [2] S. Yugo, T. Kanai and T. Kimura, Dhonond Relat. Mater., 1 (1992) 388. [3] S. Yugo, T. Kimura and T. Kanai, Diamond Relat. Mater., 2 ( 1993 ) 328. [4] S. Yugo, K. Semoto, K. Floshina, T. Kimura and H. Nakai, Diamond Relat. Mater., 4 (1995) 903. [5] D.R. McKenzie, D. Muller and B.A. Pailthorpe, Ph),s. Rev. Lett., 67(6) ( 1991 ) 733. [6] B.R. Stoner, B.E. Williams, S.D. Wolter, K. Nishiyama and J.T. Glass, J. Maters. Res., 7 (1992) 257. [7] X. Jiang and C.-P. Klages, AppL Phys. Lett., 65 (1993) 1519. [8] J. Gerber, S. Sattel, K. Jung, H. Ehrhardt and J. Robertson, Diamond Relat. Mater., 4 (1995) 559. [9] B.R. Stoner, G.-H. Ma, S.D. Wolter and J.T. Glass, Phys. Rev. B, 45, 19 (1992) 11 067. [10] R. Kohl, C. Wild, N. Herres, P. Koidl, B.R. Stoner and J.T. Glass, Appl. Phys. Lett., 63 (1993) 1203. [11] X. Jiang, K. Schiffmann and C.-P. Klages, Phys. Rev. B, 50, 12 (1994) 8402. [12] S.D. Wolter, B.R. Stoner, J.T. Glass, P.J. Ellis, D.S. Buhaenko, C.E. Jenkins and P. Southworth, Appl. Phys. Left., 62 (1993) 1215. [13] W. Kulisch, B. Sobisch, M. Kuhr and R. Beckmann, Dhmtond. Relat. Ma;er., 4 (1995) 401. [14] W. Moiler, Appl. Phys. Lett., 59. 4 ( 1991 ) 2391. [15] J. Singh and M. Vellaika, J. Appl. Phys., 73, 15 ( 1993} 2831. [16] Y. Lifshitz, G.D. Lempert, S. Rotter. I. Avigal, C. Uzan-Saguy,
R. Kalish, J. Kulik, D. Marton and J.W. Rabalais, Diamond Relat. Mater., 3 (1994) 542. [17] J. Gerber, S. Sattel, K. Jung, H. Ehrhardt and J. Robertson, Diamond Relat. Mater'., 4 (1995) 559. [18] Y. Ma, T. Tsurumi, N. Shinoda and O. Fukunaga, Diamond Relat. Mater., 4 ( 1995 ) 1325. [19] Y. Lifshitz, S.R. Kasi and J.W. Rabalais, Phys. Rev. Lett., 62, I1 (1989) 1290. [20] J. Robertson, J. Gerber, S. Sattel, M. Weiler, K. Jung and H. Ehhardt, Appl. Phys. Lett., 66, 12 (1995) 3287. [21] J. Robertson, Diamond Relat. Mater., 2 (1993) 984. [22] J. Gerber, M. Weiler, O. Sohr, K. Jung and H. Ehrhardt, Diamond Relat. Mater., 3 (1994) 506. [23] P. Deak, A, Gali, G. Sczigel and H. Ehrhardt, Diamond Relat. Mater., 4 (1995) 706. [24] E. Sandre, A. Bonnot and F. Cyrot-Lackman, Diamond Relat. Mater., 3 (1994) 448. [25] J.C. Angus, Z. Li, M. Sunkara, Y. Wang, C. Lee, W.R. Lambrecht and B. SegaU, 2nd lnt. Col!f. AppL Dia. Fihns and Rel. Mat., MYU, Tokyo, 1993 p. 153. [26] J. Robertson, Diamond. Relat. Mater'., 4 (1995) 549. [27] R. McKenzie, D. Muller, B.A. Pailthorpe, Z.H. Wang, E. Kravtchinskaia, D. Segal, P.B. Lukins, P.D. Swift, P.J. Martin G. Amaratunga, P.H. Gaskell and A. Saeed, Diamond Relat. Mater., 1 ( 1991 ) 51. [28] Y. Shigesato, R.E. Boekenhauer and B.W. Seldom Appl. Phys. Lett., 63 (1993) 314. [29] B.W. Seldom R. Csencsits, J. Rankin, E.R. Boekenhauer and Y. Shigesato, J. Appl. Phys., 75 (1994) 5001. [30] W. Piekarczyk and S. Prawer, Diamond Relat. Mater., 2 (1993~ 41. [31] S. Sattel, M. Weiler, J. Gerber, T. Giessen, H. Roth, M. Scheib. K. Jung and H. Ehrhardt, Diamond Relat. Mater., 4 (1995) 33~. [32] M. Schreck, T. Baur and B. Stritzker, Diamond Relat. Mater., 4 (1995) 553. [33] S. Katsumata and S. Yugo, DhI. Films Technol., 3, 4 (1994) 19'9. [34] S.P. McGinnis, M.A. Kelly and S.B. Hagstrom, AppL Phys. LeH., 66, 5 (1995) 3117.