APPLIED PHYSICS LETTERS
VOLUME 74, NUMBER 18
3 MAY 1999
Epitaxial aluminum carbide formation in 6H–SiC by high-dose Al1
implantation
J. Stoemenosa)
Aristotle University of Thessaloniki, Physics Department, 54006 Thessaloniki, Greece
B. Pécz
Research Institute for Technical Physics and Materials Science, H-1525 Budapest, Hungary
V. Heera
Research Center Rossendorf, D-01314 Dresden, Germany
~Received 14 December 1998; accepted for publication 2 March 1999!
Aluminum carbide precipitates are formed after Al ion implantation with dose 331017 cm22 at
500 °C into single crystalline 6H–SiC. The aluminum carbide (Al4C3) precipitates are in epitaxial
relation
with
6H–SiC
matrix,
having
the
following
orientation
relation,
@0001#6H–SiC//@0001#Al4C3, and @112̄0#6H–SiC//@112̄0#Al4C3, as transmission electron
microscopy reveals. The aluminum carbide appears around the maximum of the Al depth
distribution. Silicon precipitates were also detected in the same zone. © 1999 American Institute
of Physics. @S0003-6951~99!00518-5#
Silicon carbide ~SiC! has outstanding properties for electronic and mechanical applications under extreme
conditions.1,2 Very often the material properties of SiC are
modified by impurity atoms. In order to understand and control these modifications basic information about the chemical
and structural changes induced by the impurities in the SiC
lattice is necessary. Particular effects like precipitation and
compound formation can be expected if the impurity concentration exceeds the solubility limit.
One of the most important impurity elements in SiC is
Al. It is a very attractive p-type dopant because of its lower
ionization energy compared to other acceptors.3 Relatively
high Al concentrations up to 0.1% are required in order to
achieve low resistivity p-type SiC. Still higher Al concentrations are necessary for metallic-like conduction4 or ohmic
contact formation.5 Al metallization of SiC including Al
compound formation is a further problem under
investigation.6 Composite materials frequently contain Al or
an Al compound and SiC components which can react during
thermal treatment.7,8 Therefore the study of the behavior of
high concentrations of Al atoms in SiC is of particular interest. One way to perform such studies under definite conditions is to use ion implantation of Al impurities into single
crystalline SiC ~Refs. 3, 4, and references therein!. However,
most of them are devoted to p-type doping of SiC. Only a
few experiments were performed using ion doses high
enough to produce Al peak concentrations exceeding 1%. In
order to reduce radiation defects in SiC, implantations are
often performed at elevated temperature.9
In this letter we report on compound formation in 6H–
SiC after high dose Al implantation performed at elevated
temperature. The results have been obtained during an experiment initially designed for the investigation of ion beam
induced crystallization and surface erosion of SiC.10 Doses
of 131017 Al1 cm22 and 331017 Al1 cm22 with an energy
of 350 keV have been implanted into ~0001! oriented 6H–
SiC single crystalline substrates, partly through amorphized
surface regions, at a temperature of 500 °C. According to
TRIM calculations maximum Al concentrations of about 5%
and 15%, respectively, can be expected in a depth of 380 nm.
An extended, highly defected zone was formed by the
Al1 implantation. In Fig. 1~a!, a cross section micrograph of
the layer implanted with 331017 cm22 Al1 ions is shown in
FIG. 1. Cross-sectional transmission electron microscopy ~XTEM! micrographs and SADP ~insets! of the implanted region. ~a! Precipitates having a
laminar shape are denoted by the letter P. Diffraction spots from 6H–SiC
and Si are shown in the inset on the right-hand side of the micrograph. Extra
spot, which was indexed as the 101̄1 reflection of Al3C4 is shown in the
inset on the left-hand side of the micrograph. ~b! XTEM micrograph from an
Al3C4 precipitate. The moiré pattern denoted by the letter K is attributed to
the reflections ~000.12! Al3C4 //~0006!6H–SiC. Small stacking faults in the
6H–SiC matrix are denoted by arrows.
a!
Electronic mail: [email protected]
0003-6951/99/74(18)/2602/3/$15.00
2602
© 1999 American Institute of Physics
Appl. Phys. Lett., Vol. 74, No. 18, 3 May 1999
the depth region between 300 and 600 nm. Within this damaged zone a narrow band of large crystalline precipitates,
denoted by the letter P, were formed, around the mean projected ion range of 380 nm. Selected area diffraction pattern
~SADP! taken from the precipitated area, reveals the presence of extra spots, denoted by an arrow, which can be indexed as the 111 diffraction spot of silicon shown in the inset
on the right-hand side of Fig. 1~a!. The 111 silicon spot
reveals that the silicon precipitates have the preferential orientation of @111#Si //@0006#SiC. Extra spots in the diffraction
pattern shown in the inset on the left-hand side of Fig. 1~a!
were indexed as the 101̄1 reflection of the compound Al4C3.
The presence of Al4C3 was also confirmed by the moiré
pattern shown in Fig. 1~b!. The moiré fringes, denoted by the
letters K, are formed by interference of the imaging electron
beams originating from the two superimposed mismatched
lattices. These moiré patterns are perpendicular to the c axis
of the 6H–SiC matrix. The spacing D of the moiré pattern is
1.19 nm. This is very close to the theoretical one; 1.18 nm,
taking into account that the d lattice spacing for Al4C3 and
0006
000.12
50.207 nm and d SiC
50.251 nm,
6H–SiC are d Al
4C3
respectively.11
The precipitates were further studied by planar view
TEM ~PVTEM!. A typical SADP from the precipitated zone
when the electron beam is exactly parallel to the 6H–SiC c
axis is shown in Fig. 2~a!. The characteristic 112̄0 and 101̄0
type 6H–SiC diffraction spots of the ~0001! zone are evident. In addition, similar Al4C3 diffraction spots, which
are in perfect epitaxial orientation with the 6H–SiC
matrix, are also observed. The spots are attributed to the
Al4C3 precipitates having the orientation relation with
the 6H–Si matrix, @0001#6H–SiC//@0001#Al4C3, and
@112̄0#6H–SiC//@112̄0#Al4C3. No Al4C3 precipitates with
other orientations were observed. Most of the Si precipitates
are also in epitaxial relation with the 6H–SiC matrix having
the orientation relationship @0001#6H–SiC//@111#Si and
@112̄0#6H–SiC//@220#Si. However, about 30% of the Si precipitates are randomly oriented, as diffraction rings reveal in
Fig. 2~a!.
The dark field ~DF! micrograph taken from the common
112̄0 reflection of 6H–SiC and Al4C3 is shown in Fig. 2~b!.
The Al4C3 precipitates are not distinguished directly by DF
observation because the 112̄0 reflections of 6H–SiC and
Al4C3 are very close, as shown in Fig. 2~a!, the misfit for the
common (112̄0) planes is 7.7%. In this case the presence of
Al4C3 precipitates is evident from the moiré pattern produced
by the common 112̄0 reflections, as shown in the inset in
Fig. 2~b!.
In contrast to the results of the implantation with the
higher dose, no precipitates were found after implantation
with 131017 cm22 Al1 ions. One can speculate that either a
critical Al concentration above 5% or a critical number of
displacements above 20 dpa is necessary to stimulate the
precipitation. Probably, this precipitation is a thermodynamically driven phase separation like in spinodal decomposition.
The permanent atomic displacements during the ion implantation facilitate this process which otherwise could not start
because of the low diffusivities of the atoms in SiC. It is
known that the formation of Al–C bonds is energetically
Stoemenos, Pécz, and Heera
2603
FIG. 2. PVTEM from the precipitated zone. ~a! SADP with the electron
beam exactly parallel to the c axis of the 6H–SiC. Diffraction spots attributed to the Al4C3 precipitates have the orientation relation
@112̄0#Al4C3 //@112̄0#6H–SiC revealing perfect epitaxy with the 6H–Si matrix. Part of the Si precipitates exhibit the orientation relation
@112̄0#6H–SiC//@220#Si. The 111, 220, and 311 diffraction rings, denoted
by the letters F 1 , F 2 , and F 3 , also belong to the Si precipitates, which are
randomly oriented. Extra spots from higher Lawe zone, which belong to
Al4C3 and 6H–SiC are also observed, denoted by the letters E 1 and E 2 ,
respectively. This is a size effect due to the high density of stacking faults.
~b! PVTEM dark field micrograph from the common 6H–SiC and Al4C3
112̄0 reflection. Moiré pattern due to the 112̄0 6H–SiC//Al4C3 reflections is
shown in the inset.
favored in the ternary system Al/C/Si.8 The energetic preference of Al–C bonds gives also the explanation why doping
of SiC with Al is so successful. Almost all of the Al atoms
introduced into the SiC lattice occupy Si sites where they are
electrically active.12
The formation of aluminum carbide during high dose
Al1 implantation in 3C–SiC at 550 °C and in 6H–SiC at
800 °C has been reported by other authors.7,13,14 In these
cases the precipitates exhibit some texture and are elongated
along a specific orientation but no epitaxial relation with the
matrix was observed. No Si precipitates were reported previously in the zone implanted with Al. According to Du
et al.13 the Si loss could be caused by outdiffusion. Implantation at very high temperature does not facilitate the formation of Al3C4. For example no Al3C4 is formed after Al1
implantation in 6H–SiC with a dose of 231016 cm22 at
1400 °C, subsequently annealed at 1800 °C for 5 s.15
Transition metal carbides have a high electrical conduc-
2604
Appl. Phys. Lett., Vol. 74, No. 18, 3 May 1999
tivity and thermal stability.16 Therefore, epitaxial layers of
Al3C4 in SiC could be interesting for contact formation and
wiring. However, for this purpose the formation of continuous layers is necessary which could be impeded by the Si
precipitation. Further studies are in progress in order to investigate the phase formation after Al implantation into SiC.
One of the authors, B. Pecz, acknowledges the financial
support of the OTKA T 030447 Project.
J. B. Casady and R. W. Johnson, Solid-State Electron. 39, 1409 ~1996!.
W. E. Lee and W. M. Rainforth, Ceramic Microstructures—Property
Control by Processing ~Chapman and Hall, London, 1994!, p. 515.
3
T. Troffer, M. Schadt, T. Frank, H. Itoh, G. Pensl, J. Heindl, H. P. Strunk,
and M. Maier, Phys. Status Solidi A 162, 277 ~1997!.
4
H. Wirth, D. Panknin, E. Niemann, and W. Skorupa, Appl. Phys. Lett. 74,
979 ~1999!.
5
L. Spiess, O. Nennewitz, H. Weishart, J. Lindner, W. Skorupa, H. Ro1
2
Stoemenos, Pécz, and Heera
manus, F. Erler, and J. Pezoldt, Diamond Relat. Mater. 6, 1414 ~1997!.
J. Crofton, L. M. Porter, and J. R. Williams, Phys. Status Solidi B 202,
581 ~1997!.
7
Z. Yang, H. Du, M. Libera, and I. Singer, J. Mater. Res. 10, 1441 ~1995!.
8
T. P. D. Rajan, R. M. Pillai, and B. C. Bai, J. Mater. Sci. 33, 3491 ~1998!.
9
V. Heera and W. Skorupa, Mater. Res. Soc. Symp. Proc. 438, 241 ~1997!.
10
V. Heera, J. Stoemenos, R. Kögler, M. Voelskow, and W. Skorupa, J.
Appl. Phys. 85, 1378 ~1999!.
11
P. B. Hirsch, A. Howie, R. B. Nicholson, and D. W. Pashley, Electron
Microscopy of Thin Crystals ~Butterworths, London, 1965!.
12
V. Heera, J. Pezoldt, X. J. Ning, and P. Pirouz, Inst. Phys. Conf. Ser. 142,
509 ~1996!.
13
H. Du, Z. Yang, M. Libera, D. C. Jacobson, Y. C. Wang, and R. F. Davis,
J. Am. Ceram. Soc. 76, 330 ~1993!.
14
Z. Yang, H. Du, M. Libera, and I. Singer, Mater. Res. Soc. Symp. Proc.
354, 281 ~1995!.
15
A. V. Suvorov, O. L. Lebedev, A. A. Suvorova, J. Van Landuyt, and L. O.
Usov, Nucl. Instrum. Methods Phys. Res. B 127/128, 347 ~1997!.
16
T. Y. Kosolapova, Carbides-Properties, Production and Applications
~Plenum, New York, 1971!.
6