Carbon films grown on Pt(1 1 1) as supports for model gold catalysts

Article abstract of DOI:10.1016/j.susc.2006.04.035

Carbon films were grown on a Pt(1 1 1) single crystal by ethylene decomposition at elevated temperatures (1000-1300 K). Depending on the preparation conditions, different carbon structures formed on the metal surface such as flat and curved graphitic layers, carbon particles and carbon nanowires. Although these carbon films exhibited a high density of surface defects, gold interacted only weakly with the carbon surface. CO adsorption on the Au/carbon systems was very similar to that observed for various Au/oxide systems previously studied. This finding strongly indicates that CO adsorption on gold is essentially independent of the nature of support.

Full text of DOI:10.1016/j.susc.2006.04.035

Surface Science 600 (2006) 2688–2695
www.elsevier.com/locate/susc
Carbon films grown on Pt(111) as supports for model gold catalysts
D.E. Starr , E.M. Pazhetnov , A.I. Stadnichenko , A.I. Boronin b,*,
a
b
b
a,
*
S.K. Shaikhutdinov
a
Department of Chemical Physics, Fritz-Haber Institute, Faradayweg 4-6, Berlin 14195, Germany
b
Boreskov Institute of Catalysis, Lavrentieva 5, Novosibirsk 630090, Russia
Received 2 January 2006; accepted for publication 13 April 2006
Available online 22 May 2006
Abstract
Carbon films were grown on a Pt(111) single crystal by ethylene decomposition at elevated temperatures (1000–1300 K). Depending
on the preparation conditions, different carbon structures formed on the metal surface such as flat and curved graphitic layers, carbon
particles and carbon nanowires. Although these carbon films exhibited a high density of surface defects, gold interacted only weakly with
the carbon surface. CO adsorption on the Au/carbon systems was very similar to that observed for various Au/oxide systems previously
studied. This finding strongly indicates that CO adsorption on gold is essentially independent of the nature of support.
ꢀ
2006 Elsevier B.V. All rights reserved.
Keywords: Carbon; Thin films; Nanocarbon; Gold
1
. Introduction
step edges (see recent review [9]). A consequence of this
weak interaction is that structural studies of Au nanoparti-
cles using scanning tunneling microscopy (STM) are very
difficult due to tip induced displacement of the particles.
In order to form and stabilize smaller particles and circum-
vent these difficulties, ion pre-sputtered graphite surfaces
have been suggested as model carbon supports [9–11].
We believe that an alternative approach could be the
growth of carbon films on metal substrates potentially
allowing greater control of the defect structure on the car-
bon surface. It has been well documented in the literature,
that carbon overlayers can be formed on many metal sur-
faces such as Ni and Pt through the decomposition of
hydrocarbons [12–15]. In most cases, the resulting carbon
overlayer is graphitic in nature and exposes the basal
(0001) plane. The formation of graphite islands on
Pt(111) has been shown directly using STM by Land
et al. through the heating of pre-adsorbed ethylene to
The interaction between a gold particle and its support
is believed to play an important role in the catalytic prop-
erties of the particle, particularly for low temperature CO
oxidation (see reviews [1–3]). The supports involvement is
manifested in numerous ways including direct participation
in the reaction or through stabilization of the active gold
species. In order to elucidate reaction mechanisms in these
complex systems surface science studies employ model sys-
tems providing at times an atomic level description of reac-
tion processes. Most of these studies have been performed
by depositing gold onto oxide single crystals and thin films
(
[3–8] and references therein). For carbon supported model
catalysts, highly oriented pyrolitic graphite (HOPG) is typ-
ically used as a substrate. However, the interaction of gold
with the graphite basal surface is weak, resulting in the for-
mation of large metal particles preferentially located at the
800 K [16]. Upon further heating to 1000 K, the graphite
islands coalesced. More recently, however, Pazhetnov
et al. [17] have suggested that carbon films on Pt surfaces
at elevated temperatures (>1200 K) may exhibit graphene
layers with strongly bent surfaces similar to those in fulle-
*
Corresponding authors. Tel.: +44 30 8413 4114; fax: +44 30 8413 4105
(
S.K. Shaikhutdinov).
E-mail addresses: boronin@catalysis.nsk.su (A.I. Boronin), shaikhut-
dinov@fhi-berlin.mpg.de (S.K. Shaikhutdinov).
0
039-6028/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.susc.2006.04.035

D.E. Starr et al. / Surface Science 600 (2006) 2688–2695
2689
renes and nanotubes. Since the interaction of metal atoms
with carbon surfaces is predicted to depend on the curva-
ture of a graphite sheet [18], one expects that the nucleation
and growth of metal particles on these carbon supports
would differ significantly from HOPG, which in turn might
affect their reactivity.
placed ꢀ1 mm from the nozzle of the QMS and heated at a
rate of 5 K/s. All high purity gases were dosed via a cali-
brated directional doser.
STM images were obtained with Pt–Ir tips at typical
tunneling parameters of V = 20–100 mV and I ꢀ 1 nA on
carbon films, and V ꢀ 0.1–3 V and I ꢀ 0.1–0.5 nA on the
Au/carbon samples. The images were essentially indepen-
dent of the bias polarity. The images presented in the paper
were subjected to plane correction and low pass filtering.
XPS spectra were recorded using Al Ka radiation
(hm = 1486.6 eV) and analyzer pass energy 20 eV. The bind-
ing energies were referenced to the Au4f7/2 (84.0 eV) and
Cu2p3/2 (932.7 eV) values.
In this paper, we have studied the structure of the car-
bon films formed on Pt(111) at high temperatures (1000–
1
373 K) as well as the structure of gold particles deposited
on these films using STM and X-ray photoelectron
spectroscopy (XPS). Further, temperature programmed
desorption (TPD) was used to study the adsorption proper-
ties of both the carbon and Au/carbon systems. Our results
do not show any preferential nucleation of the Au particles
along the curved graphite surfaces. In addition, the TPD
results are quite similar to the results for CO adsorption
on oxide supported Au particles, therefore suggesting that
the adsorption of CO on Au is independent of the nature of
the support.
3. Results and discussions
3.1. Surface morphology of the carbon films
Following the growth of the LT films, LEED patterns
exhibited attenuated Pt(111) diffraction spots and a diffuse
ring surrounding these spots as has been previously ob-
served [14,15]. This ring corresponds to the lattice constant
2
. Experimental
˚
The experiments were performed in two UHV chambers
of ꢀ2.45 A as for the (0001) basal plane of graphite. For-
À10
(
base pressure <5 · 10
mbar). One (the ‘‘STM’’ cham-
mation of the ring instead of the individual spots indicates
a high degree of disordering of the graphite single-layer do-
mains, which have various orientations with respect to the
Pt(111) substrate. Some segmentation of this ring was also
observed upon higher temperature annealing, indicating a
slight preferential alignment of the graphite overlayer with
the Pt(111) surface.
ber in Berlin) was equipped with STM, low energy electron
diffraction (LEED), Auger electron spectroscopy (AES)
and a differentially pumped quadrupole mass spectrometer
(
QMS) for temperature programmed desorption (TPD)
measurements. The other (the ‘‘XPS’’ chamber in Novosi-
birsk) was equipped with a VG ESCALAB spectrometer,
LEED and QMS.
STM images of the LT films revealed wide terraces
separated by steps having heights that corresponded to
Pt(111) monoatomic steps underneath the film (Fig. 1a).
However, the terrace surface was atomically rough and
inhomogeneous, on average. These images exhibited do-
mains where the Moire structure with a periodicity of
The Pt(111) crystal was cleaned by numerous cycles of
Ar sputtering and annealing at 1300 K. Traces of surface
À6
carbon were removed by oxidation in 10 mbar O at
2
6
00 K followed by flashing to 1000 K. The resulting surface
had very sharp LEED diffraction spots characteristic of
Pt(111). Also, neither AES nor XPS showed any impurity
signals.
˚
ꢀ22 A was clearly observed (Fig. 1b), resulting from a mis-
match between the Pt(111) and graphite(0001) hexagonal
˚
Following the procedures reported in Ref. [17], the low
temperature (LT) carbon film was formed on Pt(111) by
lattices having unit cell parameters of 2.87 and 2.45 A,
respectively. However, ill-defined structures with high cor-
rugation amplitudes covered most of the surface as shown
in Fig. 1c. Tentatively, we have suggested that the carbon
surface is represented by small graphite domains and there-
fore possesses a high density of domain boundaries (see
also STM results below for the HT film). This is in agree-
ment with the observation of the graphite-induced diffrac-
tion ring rather than spots in LEED and consistent with
other studies of this surface where ethylene was used as a
precursor [16]. An alternative explanation could be the for-
mation of a second, presumably amorphous, carbon layer
since carbon deposition is performed in a very high excess
of ethylene. This would be consistent with the high rate of
carbon gasification observed on the LT film [17]. Note also,
that annealing the surface at 1250 K for 10 min in UHV
did not significantly alter the carbon film morphology by
STM even though some segmentation of the ring pattern
was observed by LEED.
À7
exposure to 10 mbar of ethylene at 300 K followed by
heating to 1000 K with the rate 3 K/s and keeping the
sample at this temperature for 30 min. After heating was
switched off, the sample was cooled in ethylene ambient
down to 800 K before pumping out the ethylene. The high
temperature (HT) carbon film was prepared as follows. The
À7
sample was exposed to 10 mbar of ethylene at 300 K for
3
min, heated up to 1173 K with the rate 3 K/s and kept at
this temperature for 5 min followed by slow (0.05 K/s)
heating to 1373 K where the sample was kept for 5 min be-
fore slowly cooling down (À0.03 K/s) to 800 K and pump-
ing out the ethylene.
In the ‘‘STM’’ chamber, Au was deposited from a Au-
˚
filled Mo crucible, with a deposition rate of ca. 0.1 A/
min. In the ‘‘XPS’’ chamber, Au was deposited from a
Au wire wrapped around a W-filament. For TPD measure-
ments (performed in the ‘‘STM’’ chamber), the sample was

2690
D.E. Starr et al. / Surface Science 600 (2006) 2688–2695
Fig. 2a shows a typical large-scale STM image of the HT
film. In contrast to the LT film, a large density of irregu-
larly shaped particles (see inset) is clearly seen. In addition,
carbon nanowires up to hundreds of nm in length and
5–10 nm in diameter are formed (see also Fig. 3a).
The terraces consist of graphite domains of various rota-
tional orientations with respect to each other as resolved in
Fig. 2b. As a result, numerous domain boundaries are ob-
served. These boundaries may show periodic structures,
somewhat resembling the surface of the carbon nanotubes
on graphite [20]. The graphitic domains show a hexagonal
˚
lattice of protrusions with a ꢀ2.5 A periodicity similar to
the basal plane of HOPG. However, Fig. 2c and d shows
that the surface looks different within the same domain:
some regions show a honeycomb-like structure (1), similar
to HOPG, others – trimer-like (2) structures or ‘‘commas’’
(3). In addition, the star-like feature seen in the middle of
Fig. 2d were frequently observed on this sample and may
be tentatively assigned to positions where the various rota-
tional orientations of the ‘‘trimer’’ structures meet. It is
clear that these effects cannot be assigned to tip asymmetry
effects and seem to be electronic in nature possibly resulting
from the different stacking of the graphite layers, which are
bonded through relatively weak van der Waals forces. For-
mation of numerous small domains indicates that the sur-
face is highly strained which is quite likely due to the
large lattice mismatch (ꢀ12%) between graphite(0001)
and Pt(111). Therefore, it may well be that the various
graphite layers are not in registry, which leads to different
incommensurate structures and hence STM images.
Relatively few domains exhibited the Moire structure
typically observed for thin graphite films on Pt(111) (see
Fig. 2f). Since this superstructure most likely would result
from only a single layer of graphite on Pt(111), we con-
clude that the HT carbon film is relatively thick due to their
lack of prominence, which is consistent with the LEED and
AES results.
Annealing of the HT film in UHV at 1300 K for 40 min
resulted in more frequent observation of nanowires on the
surface (cf. Fig. 3a and b). No preferential location of the
wires along the step edges is seen, in fact many wires cross
multiple steps. The annealed samples also begin to show
patches, which are atomically smooth and free of domain
boundaries and particles (marked by the star in Fig. 3b).
Further annealing causes a significant reduction of the par-
ticles’ density and increases the areas of smooth surface, as
seen in Fig. 3c. Finally, annealing at 1300 K for 2 h has re-
moved carbon from the surface as revealed both by STM
and AES, presumably due to carbon dissolution into the
Pt crystal bulk.
Fig. 1. (a) Large-scale STM image, presented in differentiated contrast, of
low temperature carbon films showing flat terraces separated by mon-
atomic steps of a Pt substrate underneath the film. (b and c) STM images
of the different areas of the film. Inset in (b) shows atomically resolved
Moire structure which arises due to coincidence of hexagonal graph-
ite(0001) and Pt(111) lattices.
Figs. 2a and 3a and c show that carbon nanowires
apparently grow from very large particles, the shape of
which resemble liquid-like droplets as shown in Fig. 3d
and also seen in Fig. 2a. Formation of liquid-like Fe parti-
cles during interaction with amorphous carbon at elevated
temperatures was previously demonstrated by Zaikovskii
et al. using in situ high resolution electron microscopy
LEED patterns of the HT films were different from the
LT films showing only the single diffraction ring and no
Pt spots. AES spectra did not reveal Pt signals as well. This
means that the HT film is thicker than the LT film, and pre-
sumably of at least 10–15 A thickness based on the Auger
electron escape depth of Pt [19].
˚

D.E. Starr et al. / Surface Science 600 (2006) 2688–2695
2691
Fig. 2. STM images of the high temperature carbon film (see the text). Inset in (a) shows the close-up of the numerous particles formed on the surface.
Inset in (b) zooms in on the particle seen in the middle of the image. In addition to honeycomb structure (1) typical for graphite, the ‘‘trimer’’- (2) and
‘
‘comma’’-like (3) features are observed. The images (a, b and e) are presented in differentiated contrast.
[
21]. Note also, that Fujita et al. [22] has recently reported
which is why our observation of carbon nanowires on Pt
is very unusual.
on the observation of carbon nanowires on a Ni(111) sin-
gle crystal surface doped with carbon. This finding may be
rationalized by the high affinity of 3d-metals for carbon. In
fact, the strong carbon–metal interaction lays in the mech-
anism of the formation of filamentous carbon and carbon
nanotubes on Ni, Co and Fe, i.e. metals which readily form
metal carbides (see for instance Ref. [23]). However, plati-
num is known to be very resistant to carbide formation,
Note also that the morphology and AES spectra of the
À6
HT carbon film were not affected by exposure to 10 mbar
of O at 1000 K for 15 min, which implies high resistance of
2
the HT films towards gasification. As expected, no hydro-
gen and CO adsorption above 100 K has been detected
by TPD for all films studied. Also water desorbs from these
films only in a single peak at ꢀ170 K, which implies forma-

2692
D.E. Starr et al. / Surface Science 600 (2006) 2688–2695
Fig. 3. STM images, presented in differentiated contrast, of the high temperature carbon film upon high temperature annealing in vacuum at 1300 K for
0 min (b) and 80 min (c) (see the text). A large particle resembling a liquid-like feature is zoomed in (c). The star marks the atomically smooth surface
free of domain boundaries and particles), which increases in area upon prolonged annealing in vacuum.
4
(
tion of a physisorbed multilayer water film and no water
dissociation. Therefore, the TPD results indicate that the
carbon surface is essentially inert, even though the films
contain a large number of defects as revealed by STM.
3
.2. XPS characterization of the carbon films
As prepared, the carbon films exhibit C 1s binding ener-
gies (BE) of 284.2 and 284.7 eV for the LT and HT films,
respectively, indicating different natures of the carbon
formed. This is consistent with the STM results, presented
above, showing a different morphology of the carbon over-
layers. In order to monitor changes occurring while the film
transforms from the LT film into the HT state, we have
performed the following experiment. The carbon film, pre-
À7
Fig. 4. C 1s region of the XPS spectra of the carbon films after heating of
a Pt(111) crystal up to the specified temperature and subsequent cooling
in 10 mbar of ethylene. All spectra were recorded at 1000 K. The inset
pared at 1000 K, was further heated in 10 mbar of ethyl-
ene at the specified temperature and cooled down to
À7
1
000 K in ethylene ambient, the spectra were then recorded
shows the BE values as a function of the preparation temperature.
at 1000 K in vacuum. Fig. 4 shows a series of C 1s XPS
spectra for different treatment temperatures. It is clearly
seen that the carbon peak gains intensity by a factor of
two, which is consistent with the increasing thickness of
the HT film as implied by AES and LEED (see Section
3.1). In addition, the BE starts to shift to higher energies
by ca. 0.5 eV above 1250 K (see inset in Fig. 4). It is impor-
tant to note that in situ XPS measurements during stepwise

D.E. Starr et al. / Surface Science 600 (2006) 2688–2695
2693
heating of the LT film to 1370 K in ethylene ambient did
not show any changes in the spectra unless the sample is
cooled below 1200 K. This finding strongly suggests that
all changes, observed in Fig. 4, develop while the sample
is cooling.
rather the spectra of carbon nanotubes and fullerenes than
HOPG [17]. Meanwhile, based on XPS results, the LT films
are suggested to be pure graphitic in nature.
3.3. Au on the carbon films
In order to rationalize this effect, we have to recall that
at elevated temperatures, above ꢀ1200 K, carbon may mi-
grate into the Pt bulk [13]. As a result, the near surface re-
gion of the Pt crystal accumulates large quantities of
carbon. During the slow cooling cycle this carbon may seg-
regate back to the surface and therefore interact with sur-
face carbon and ethylene to form a thick carbon film.
However, it must be the presence of ethylene in an ambient
atmosphere during cooling that governs this film formation
since prolonged annealing of the HT film at 1300 K in vac-
uum in fact lead to a gradual disappearance of carbon
structures from the surface. Therefore, it is the complex
interplay between carbon bulk dissolution, surface segrega-
tion and reaction in ethylene ambient that results in a thick
carbon film with various forms of carbons present on the
surface as observed by STM on the HT films. As clearly
seen in Figs. 2 and 3, these carbon structures may exhibit
curved surfaces, which can explain the XPS results that
the C 1s and Auger C-KLL lines for the HT films resemble
Gold, vapor deposited on both the LT and HT films,
was examined by STM at 300 K. Various tunneling param-
eters were used bearing in mind possible tip induced effects
frequently observed for gold/graphite systems [9]. How-
ever, no small particles that could be unambiguously as-
signed to Au, were observed on the surface even when
Au deposition was performed at 120 K. Domain bound-
aries, terrace steps, carbon particles and nanowires, which
might nucleate Au particles, looked very similar to those
prior to gold deposition measured at the same tunneling
conditions. Only a few but very large (above 10 nm in size),
well shaped particles (and therefore assigned to the Au par-
ticles) on both the HT and LT carbon supports have been
observed as shown in Fig. 5a and b. This is in contrast
to our previous results of Au deposited on oxide films
(Al O ,FeO,Fe O ) where small particles, randomly dis-
2
3
3
4
tributed on the support, were observed by STM [6,7].
Therefore, the present STM data indicate that the interac-
Fig. 5. STM images, presented in differentiated contrast, of gold deposited at 120 K and annealed to 500 K on LT (a) and HT (b) carbon films. The images
are recorded at high tunneling bias and low current (3 V, 0.1 nA) in order to avoid tip induced displacement of the particles, however, traces of such
movement are seen in (a). Large hexagonally shaped particles are formed as indicated by the arrows in (b). The other particles in this image are similar to
those observed on the pristine carbon films with irregular shapes and therefore assigned to carbon (see Fig. 2). The TPD spectra for gold on the LT (c) and
HT (d) films are shown for as deposited at 120 K gold and those annealed to 500 K during the first CO TPD run.

2694
D.E. Starr et al. / Surface Science 600 (2006) 2688–2695
could be formed. In the low temperature regime
(
ꢀ1000 K), the thin carbon film is graphitic in nature. At
higher temperatures, thicker carbon films, exposing carbon
nanowires and curved graphitic layers, are observed. We
suggest that the structure of the carbon film on Pt is the re-
sult of the complex interplay of carbon dissolution/segrega-
tion and therefore may depend on other parameters such as
heating/cooling rates as well. The difference in the struc-
tures of the LT and HT film may explain the different reac-
tivity towards oxygen as previously observed [17].
The carbon films exhibited a surface with a high density
of defects, however, gold species vapor deposited on these
films are found to interact only weakly with the carbon film
supports such that only relatively large ꢀ10 nm particles
were observed by STM at room temperature. This is in con-
trast to Au deposited on various oxide films under the same
conditions, where much smaller particles were stable upon
heating to 500 K. Therefore, the STM data indicate a weak
interaction of metallic gold with carbon substrates, inde-
pendent of the nature and defect structure of the carbon.
Our TPD studies showed that the CO adsorption behav-
ior (desorption temperature, adsorption capacity) is very
similar to that observed for various gold/oxide systems
studied in our group. These results fully support the con-
clusion [6,7] that CO adsorption on gold is essentially inde-
pendent on the nature of support used, however, the
support may influence stability of the smallest Au species
towards sintering.
Fig. 6. In situ Au4f spectra obtained for Au deposited on the HT film
upon stepwise heating in vacuum from 300 to 1000 K. Inset shows the
intensity ratio of Au4f and C 1s signals as a function of the annealing
temperature.
tion of gold with the carbon supports is weaker than with
the oxide surfaces, and that this interaction is not affected
by the nature of the carbon support (LT vs HT film).
Fig. 5c and d presents CO TPD spectra for the two Au/
C surfaces. These spectra are very similar to those we have
observed for Au particles on oxide films [6,7]. As deposited
at 120 K Au species adsorb CO much stronger than do an-
nealed Au particles (Tdes ꢀ 235 K vs ꢀ185 K). In addition,
the CO adsorption capacity drops upon annealing. Both
observations indicate the adsorption of CO takes place
only on low-coordinated Au surface atoms, which decrease
in number upon heating and concurrent sintering. In sum-
mary, these results fully support our previous conclusions
that CO adsorption on gold is essentially independent on
the nature of support used, however, the support may influ-
ence the stability of the smallest Au species, which in turn
depends on the Au/support interaction [3].
Acknowledgments
We acknowledge the Russian National Program
‘
(
‘Development of the Scientific Potential of High School’’
grant UR 05.01.206).
XPS measurements were also conducted for ꢀ0.05 ML
References
of gold deposited at 300 K. The Au4f
BE value of
/2
7
8
4.0 eV indicates that the gold species are essentially in a
[
[
[
1] M. Haruta, Chem. Rec. 3 (2003) 75.
metallic state and seem to form relatively large particles
at 300 K, otherwise one would expect higher BE values
as previously observed for very small gold particles [24–
2] G.C. Bond, D.T. Thompson, Catal. Rev. Sci. Eng. 41 (1999) 319.
3] R. Meyer, C. Lemire, S. Shaikhutdinov, H.-J. Freund, Gold Bull. 37
(2004) 72.
2
7]. Stepwise (50 K step) annealing for 30 min at the
[4] A. Kolmakov, D.W. Goodman, Surf. Sci. 490 (2001) L597.
[
5] V.A. Bondzie, S.C. Parker, C.T. Campbell, J. Vac. Sci. Technol. A 17
1999) 1717.
6] C. Lemire, R. Meyer, S. Shaikhutdinov, H.-J. Freund, Surf. Sci. 552
2004) 27.
temperatures specified in Fig. 6 results only in a strong
decrease of the Au4f signal intensity and not of C 1s. Since
gold does not desorb at these temperatures, it seems possi-
ble that the Au migrates into the film through the numer-
ous defects present on the film as has been previously
observed for small metal particles deposited on thin alu-
mina films [28]. An alternative explanation could be the
further sintering of gold at elevated temperatures.
(
[
(
[7] S. Shaikhutdinov, R. Meyer, M. Naschitzki, M. B a¨ umer, H.-J.
Freund, Catal. Lett. 86 (2003) 211.
[
8] E. Wahlstr o¨ m, N. Lopez, R. Schaub, P. Thostrup, A. Rønnau, C.
Africh, E. Laegsgaard, J.K. Nørskov, F. Besenbacher, Phys. Rev.
Lett. 90 (2003) 026101.
[9] H. H o¨ vel, I. Barke, Prog. Surf. Sci. 81 (2006) 53.
[
10] I. Lopez-Salido, D.C. Lim, Y.D. Kim, Surf. Sci. 588 (2005) 6.
[11] Y.J. Zhu, A. Schnieders, J.D. Alexander, T.P. Beebe, Langmuir 18
2002) 5728.
4
. Summary
(
[
[
[
12] K. Baron, D.W. Blakely, G.A. Somorjai, Surf. Sci. 41 (1974) 45.
13] J.C. Hamilton, J.M. Blakely, Surf. Sci. 90 (1980) 199.
14] B. Lang, Surf. Sci. 53 (1975) 317.
Carbon films were grown on a Pt(111) single crystal by
À7
ethylene (>10 mbar) decomposition at 1000–1300 K.
LEED, AES, XPS and STM studies show that, depending
on the preparation conditions, different types of carbon
[15] Zi-pu Hu, D.F. Ogeltree, M.A. van Hove, G.A. Somorjai, Surf. Sci.
180 (1987) 433.

D.E. Starr et al. / Surface Science 600 (2006) 2688–2695
2695
[
[
[
16] T.A. Land, T. Michely, R.J. Behm, J.C. Hemminger, G. Comsa, J.
Chem. Phys. 97 (1992) 6774.
17] E.M. Pazhetnov, S.V. Koshcheev, A.I. Boronin, Kinetika i Kataliz
Engl. Transl. 44 (2003) 1.
[23] M.S. Dresselhaus, G. Dresselhaus, P. Avouris (Eds.), Carbon
Nanotubes: Synthesis, Structure, Properties, and Applications,
Springer, Berlin, 2001.
[24] G.K. Wertheim, S.B. Di Cenzo, S.E. Youngquist, Phys. Rev. Lett. 52
(1983) 2310.
18] M. Menon, A.N. Andriotis, G.E. Froudakis, Chem. Phys. Lett. 320
(
2000) 425.
[25] S. Shukla, S. Seal, NanoStruct. Mater. 11 (1999) 1181.
[26] E. Thune, E. Carpene, K. Sauthoff, M. Selbt, P. Reinke, J. Appl.
Phys. 98 (2005) 34304.
[27] I.R. Videnovic, P. Oelhafen, J. Appl. Phys. 97 (2005) 74308.
[28] M. Heemeier, S. Stempel, Sh.K. Shaikhutdinov, J. Libuda, M.
B a¨ umer, R.J. Oldman, S.D. Jackson, H.-J. Freund, Surf. Sci. 523
(2003) 103.
[
[
[
19] Handbook of Auger Electron Spectroscopy, JEOL, Tokyo, 1982.
20] C. Rettig, M. B o¨ decker, H. H o¨ vel, J. Phys. D 36 (2003) 818.
21] O.P. Krivoruchko, V. Zaikovskii, K.I. Zamaraev, Dokl. Akad. Nauk
3
29 (1993) 744.
[
22] D. Fujita, T. Kumakura, K. Onishi, K. Sagisaka, T. Ohgi, M.
Harada, Surf. Sci. 566–568 (2004) 361.