Formation of an atomically abrupt interface between a polycyclic aromatic molecule and the silicon (001) surface via direct Si-C linkage

Article abstract of DOI:10.1021/jp0265844

The reaction of acenaphthylene with the Si(001) surface has been investigated as a model system for understanding how complex ??-systems interact with a ??-bonded semiconductor surface. Scanning tunneling microscopy (STM) reveals that bonding occurs over a dimer row for more than 90% of the adsorbed molecules. Fourier transform infrared spectroscopy (FTIR), using isotopically labeled derivatives of acenaphthylene, shows that bonding occurs through the 1,2-alkene, leaving the aromatic portion of the molecule mostly unperturbed. The reaction between acenaphthylene and Si(001) represents a potential method for coupling extended ??-electron systems with group IV semiconductor surfaces.

Full text of DOI:10.1021/jp0265844

224
J. Phys. Chem. B 2003, 107, 224-228
Formation of an Atomically Abrupt Interface between a Polycyclic Aromatic Molecule and
the Silicon (001) Surface via Direct Si-C Linkage
Michael P. Schwartz, Robert J. Halter, Robert J. McMahon, and Robert J. Hamers*
Department of Chemistry, UniVersity of WisconsinsMadison, 1101 UniVersity AVenue,
Madison, Wisconsin 53706
ReceiVed: July 22, 2002; In Final Form: October 25, 2002
The reaction of acenaphthylene with the Si(001) surface has been investigated as a model system for
understanding how complex π-systems interact with a π-bonded semiconductor surface. Scanning tunneling
microscopy (STM) reveals that bonding occurs over a dimer row for more than 90% of the adsorbed molecules.
Fourier transform infrared spectroscopy (FTIR), using isotopically labeled derivatives of acenaphthylene,
shows that bonding occurs through the 1,2-alkene, leaving the aromatic portion of the molecule mostly
unperturbed. The reaction between acenaphthylene and Si(001) represents a potential method for coupling
extended π-electron systems with group IV semiconductor surfaces.
I. Introduction
The Si(001) surface consists of pairs of atoms that are
formally linked together with a very weak SidSi double bond.1
Recent studies have shown that many aspects of the chemistry
of the Si(001) surface are analogous to the chemistry of the
π-bonds of organic alkenes and silenes, but with a significant
Figure 1. Previously reported [4+2]-cycloaddition reaction of 9,10-
amount of diradical and zwitterionic character introduced to the
silicon dimer by the unusual geometry of the surface atoms and
the very low strength of the surface π-bond.2,3 Because of the
importance of silicon in microelectronics, there is great interest
in understanding how to control interfaces between silicon and
organic compounds. Organic molecules that contain π-conju-
gated systems are of especially high interest because many of
these molecules are organic semiconductors,4 and may have the
ability to function as active devices.5,6 To successfully integrate
organic materials with silicon-based technology, it is critical to
be able to prepare very well-defined interfaces between silicon
and a variety of organic materials. It is especially important to
control how the interface formation affects the π-conjugation
within the molecule, since loss of π-conjugation would signifi-
cantly alter the electronic properties of the organic material.
Molecules in which the conjugated π-electron system extends
through the entire molecule can directly link to the Si(001)
surface, but the formation of covalent Si-C bonds at the
interface is accompanied by loss of π-conjugation.7 This
problem can be avoided by using an external functional group
such as an isothiocyanate,8 vinyl,9 amine,10 or thiol11 group. If
the π-conjugated electron system is tethered to the surface
through only one bond, however, there are still many degrees
of freedom that are uncontrolled, generally leading to disordered
molecular layers. Recent studies showed that 9,10-phenan-
threnequinone could be linked to the Si(001) surface with two
Si-O-C linkages per molecule, thereby forming a rigid layer
of oriented π-conjugated molecules, as shown in Figure 1.12,13
It is likely that Si-O-C linkages are comparatively insulating
interfaces, in analogy to the well-known insulating properties
of SiO2. Since electron-donating and electron-withdrawing
phenanthrenequinone with a Si dimer, from ref 12.
groups can strongly affect band alignment and conductivity at
the interface, there is interest in designing other types of specific
molecular structures that will yield predictable electronic
properties at organic-inorganic interfaces.14-16
In this paper, we report the selective attachment of a pure
π-conjugated hydrocarbon, acenaphthylene, to the Si(001)
surface. Selective isotopic labeling studies show that acenaph-
thylene links to the surface in such a way as to preserve the
aromaticity. This produces a layer of rigidly bonded molecules
with intact, conjugated π-electron systems, linked to the surface
via a direct Si-C bond.
II. Experimental Section
Experimental measurements were performed using scanning
tunneling microscopy (STM) and Fourier transform infrared
spectroscopy (FTIR). STM and FTIR experiments were per-
formed in separate ultrahigh vacuum (UHV) chambers having
base pressures <10-10 Torr.
All STM and FTIR experiments used “on-axis” silicon
samples that were oriented to (001) ( 0.5°. STM samples were
highly doped (<0.1 ohm cm, As-doped), while FTIR samples
were lightly doped (>5 ohm cm, B-doped). IR spectra reported
here were obtained using nominally unpolarized light.
Wafers polished on both sides were used in all IR spectros-
copy experiments. From these large wafers individual smaller
samples (∼ 1 cm × 2 cm) were cut. The narrow edges (∼ 0.5
mm × 1 cm) were then polished at a 45° angle, making each
Si sample a small prism. For all experiments, samples were
degassed overnight at ∼840 K in UHV and then flash annealed
* Corresponding author. Phone: 608-262-6371. Fax: 608-262-0453.
E-mail: rjhamers@facstaff.wisc.edu.
10.1021/jp0265844 CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/10/2002

Interface between a Polycyclic Aromatic Molecule and Si(001)
J. Phys. Chem. B, Vol. 107, No. 1, 2003 225
Figure 2. Molecular structures for (a) acenaphthylene, (b) acenaph-
thylene-d2, and (c) acenaphthylene-d6.
to ∼1400 K until pressure stabilized at <3 × 10-10 Torr,
producing a clean (2 × 1)-reconstructed surface.17
Infrared absorption spectra for adsorbed molecules were
obtained using a Mattson RS-1 FTIR spectrometer coupled to
a UHV system through two BaF2 windows. The infrared light
was focused into the narrow edge of the trapezoidal samples.
After propagating down the ∼2 cm length of the sample the
light emerged from the narrow edge at the opposite end of the
sample, where it was collected and focused onto a liquid
nitrogen-cooled InSb detector. Spectra were acquired with 4
cm-1 resolution.
Direct observation of adsorbed molecules was achieved with
a home-built UHV scanning tunneling microscope (STM). While
a range of imaging conditions was explored, all images shown
here were obtained using a sample bias of -2.3 V and a
tunneling current of 80 pA.
Figure 2 shows molecular structures and atomic numbering
for (a) acenaphthylene, (b) acenaphthylene-d2, and (c) acenaph-
thylene-d6. The carbon atoms in the 1 and 2 positions form a
CdC bond that is conjugated to the aromatic portion of the
molecule, and will be referred to as the 1,2-alkene. The aromatic
portion of the acenaphthylene molecule is referred to as the
naphthalene ring. Acenaphthylene (Figure 2a) was purchased
from Fisher (99+% purity). Isotopically labeled samples (Figure
2b and 2c) were synthesized as explained in general synthetic
methods (Part III).
In both vacuum chambers used for these studies, acenaph-
thylene was introduced using a leak valve and pressure was
measured using an ionization gauge. Both the location of the
leak valve and other geometric properties of the vacuum systems
make the effective pressure at the sample higher than the
background pressure rise detected by the ionization gauge. All
pressures and doses reported here are based on the uncorrected
background pressure measurement. Therefore, the true exposures
at the sample are expected to be several times higher than the
reported values. Due to the low vapor pressure of acenaphth-
ylene, there was also significant time lag between achieving
maximum dosing pressure and return to base pressure. Because
of this lag in pressure, it is impossible to determine precise
saturation exposure. On the basis of a maximum time lag, we
can estimate the saturation exposure to be less than 1 langmuir
(1 L ) 10-6 Torr s ). Larger exposures of acenaphthylene at
room temperature did not change the spectra from those reported
at lower coverage.
Figure 3. (a) STM image of subsaturation coverage acenaphthylene
on the Si(001) surface. Image was acquired at -2.3 V bias and 8 pA
tunneling current. Each of the species present is centered over a dimer
row. (b) Schematic representation of the position of acenaphthylene
molecules relative to the underlying dimer row.
material.18 Acenaphthene (1.02 g, 6.61 mmol) was dissolved
in 10 mL of Me2SO-d6 and heated slightly to speed up
dissolution. After cooling to room temperature, NaH (60%
dispersion, 52.2 mg, 1.31 mmol) was added and the reaction
heated to 78 °C for 22 h. Water (5 mL) and HCl (1 M, 5 mL)
were added and the acenaphthene precipitated from solution as
a gray solid which was collected by vacuum filtration. Sublima-
tion (0.02 Torr, 50-60 °C bath temp) gave acenaphthene-d4 as
white needles in 100% yield. Deuterium incorporation into the
methylene positions was >98% by 1H NMR and no deuterium
incorporation into the ring was observed via 2H NMR. 1H NMR
δ 7.58 (dd, J ) 8.1, 0.6 Hz, 1H), 7.42 (m, 1H), 7.26 (m, 1H).
2H NMR δ 3.44 (s). 13C NMR δ 145.9, 139.4, 131.6, 127.7,
122.2, 119.2, 29.5 (m). MS (EI) m/z (relative intensity) 158 (M+,
100), 129 (10), 78 (26). mp 92-93.5 °C.
[1,2-2H]-Acenaphthylene (referred to as acenaphthylene-
d2). The bromination/debromination procedure of Anderson et
al.19 was followed to give acenaphthylene-d2 from [1,1,2,2-2H]-
acenaphthene in 37% overall yield after purification by sublima-
tion (0.02 Torr). 1H NMR δ 7.80 (d, J ) 8.1 Hz, 1H), 7.68 (d,
J ) 6.6 Hz, 1H), 7.54 (t, J ) 7.2 Hz, 1H). 2H NMR δ 7.17 (s).
13C NMR δ 139.6, 129.0 (t, J ) 26.1 Hz), 128.2, 128.1, 127.6,
127.2, 124.1. MS (EI) m/z (relative intensity) 154 (M+, 100),
128 (24), 77 (75). HRMS calcd for C12H6D2 154.0752, found
154.0752. mp 87-89 °C.
[3,4,5,6,7,8-2H]-Acenaphthylene (referred to as acenaph-
thylene-d6). [3,4,5,6,7,8-2H]-Acenaphthene was synthesized in
an identical manner as [1,1,2,2-2H]-acenaphthene above, starting
1
from perdeuterioacenaphthene and protio Me2SO. H NMR δ
2
3.40 (s). H NMR δ 7.66 (s, 1D), 7.47 (s, 1D), 7.30 (s, 1D).
MS (EI) m/z 160 (M+, 100), 131 (11). Bromination/debromi-
1
nation then provided [3,4,5,6,7,8-2H]-acenaphthylene: H NMR
III. General Synthetic Methods
δ 7.08 (s). 2H NMR δ 7.93 (s, 1D), 7.80 (s, 1D), 7.65 (s, 1D).
MS (EI) m/z (relative intensity) 158 (M+, 100), 130 (8).
Me2SO was distilled under vacuum, discarding the first 30%
of the distillate. Me2SO-d6 and other compounds from com-
mercial sources were used without further purification. 1H and
13C NMR spectra (Bruker AC-300 spectrometer) and 2H NMR
spectra (Bruker AC-250 spectrometer) were obtained in CDCl3
or CHCl3, respectively, with chemical shifts referenced to Me4-
Si (1H) or CD2Cl2 (2H). Air-sensitive reactions were run under
N2 in oven-dried glassware.
IV. Results
Figure 3a shows an STM image of a Si(001) surface with a
very low coverage of adsorbed acenaphthylene molecules. At
low coverage, STM images provide information about the spatial
location of the adsorbed molecules relative to the underlying
Si(001) crystal lattice, as well as the number of inequivalent
binding sites. Analysis of STM images like that shown in Figure
[1,1,2,2-2H]-Acenaphthene. The deuterated sample was
prepared by base-catalyzed exchange in the protio starting

226 J. Phys. Chem. B, Vol. 107, No. 1, 2003
Schwartz et al.
Si(001) surface held at 110 K. This spectrum is characterized
by peaks at 2954, 2977, 2999, 3025, 3041, 3062, and 3099 cm-1
,
as well as shoulders at 3079 and 3110 cm-1. With the exception
of the small peak at 2977 cm-1 and the shoulder at 3110 cm-1
,
this spectrum is in excellent agreement with a literature spectrum
of acenaphthylene solid.20
A comparison between acenaphthylene/Si(001) and solid
acenaphthylene reveals that attachment results in the appearance
of new peaks in the alkane C-H stretching region that were
not present for the parent molecule. The increase in intensity
of the C-H stretching peaks below 3000 cm-1 suggests that a
significant fraction of carbon atoms undergo a change from sp2
to sp3 hybridization when acenaphthylene bonds to the Si(001)
surface.
To determine which specific carbon atoms form bonds
between the acenaphthylene molecule and the Si(001) surface,
we collected and analyzed FTIR spectra in which selected H
atoms within the acenaphthylene molecule were replaced with
deuterium. Figure 4c shows an FTIR spectrum of a Si(001)
surface exposed to 8 L (4 × 10-8 Torr, 200 s) acenaphthylene-
d2, while Figure 4d shows the corresponding spectrum of a
multilayer film deposited onto a cold sample. In the spectrum
for the monolayer chemisorbed onto Si(001), high-frequency
C-H peaks (typical of sp2 hybridized carbon) are observed at
3036 and 3059 cm-1, with a shoulder at ∼3021 cm-1. Lower-
frequency peaks (typical of sp3 hybridized carbon) are observed
at 2885 and 2938 cm-1, but these peaks have a very low
intensity.
Figure 4. FTIR spectra for (a) Si(001) surface exposed to saturation
coverage acenaphthylene at 300 K, (b) multilayer of acenaphthylene,
(c) Si(001) surface exposed to saturation coverage acenaphthylene-d2
at 300 K, (d) multilayer of acenaphthylene-d2, (e) Si(001) surface
exposed to saturation coverage acenaphthylene-d6 at 300 K and (f)
multilayer of acenaphthylene-d6 at 110 K.
3a indicates that more than 90% of the molecules give rise to
a bright protrusion centered on top of a dimer row, as
schematically represented in Figure 3b. The presence of a
dominant feature in the STM images that represents more than
90% of the adsorbed molecules indicates that there is one
preferred bonding configuration. The remaining 10% of the
molecules appear to be bonded in other alternative configura-
tions and may be associated with defects of the underlying
surface.
To gain further information about the bonding geometry,
FTIR spectroscopy was used to determine how chemical bonds
in acenaphthylene change upon adsorption onto the Si(001)
surface. Because all carbon atoms in acenaphthylene are sp2
hybridized (unsaturated), evaluation of the C-H stretching
region provides clues about how attachment to the Si(001)
surface occurs.
Figure 4 shows FTIR spectra of a Si(001) surface exposed
to (a) saturation coverage of acenaphthylene at 300 K, (b) a
multilayer of acenaphthylene, (c) saturation coverage of acenaph-
thylene-d2 at 300 K, (d) a multilayer of acenaphthylene-d2, (e)
saturation coverage of acenaphthylene-d6 at 300 K, and (f) a
multilayer of acenaphthylene-d6. Multilayer spectra were ac-
quired by lowering the sample temperature to ∼110 K followed
by further exposure to over 10 L acenaphthylene, using the
acenaphthylene monolayer-terminated surface as a background.
Figure 4a shows the FTIR spectrum of a Si(001) surface
exposed to 0.6 L (3 × 10-8 Torr, 20 s) acenaphthylene. Further
exposure did not result in any changes in the FTIR spectrum,
indicating that the Si(001) surface is saturated after exposure
to 0.6 L acenaphthylene. The FTIR spectrum of acenaphthylene
adsorbed on Si(001) is characterized by several peaks between
2800 and 3200 cm-1. The peaks at 2888 and 2944 cm-1 are in
the region typically associated with C-H stretching vibrations
of sp3 hybridized carbon atoms, such as in alkanes. The higher-
lying peaks at 3034 and 3059 cm-1 are in the region typically
associated with sp2 hybridized carbon, such as in alkenes.
For comparison, Figure 4b shows the FTIR spectrum of a
multilayer of acenaphthylene formed by deposition on a
The corresponding spectrum of the multilayer film (Figure
4d) shows several peaks in the C-D and C-H stretching
regions. While we were unable to find previous assignment of
the FTIR peaks for acenaphthylene-d2, a comparison to acenaph-
thylene combined with the fact that substitution of H atoms with
D atoms shifts FTIR peaks to a lower frequency allows us to
make reasonable assignments. A peak at 2323 cm-1 (in the C-D
stretching region) is attributed to the C-D stretch of the 1,2-
alkene since this is the only portion of the molecule that contains
C-D bonds. Several weak-strong peaks are observed in the
C-H stretching region from 2971 to 3110 cm-1. These peaks
are attributed primarily to C-H stretching of the naphthalene
portion of the acenaphthylene-d2 molecule.
While we can conclude that the majority of the observed
absorbance in the C-H stretching region is due to C-H
stretches of the naphthalene portion of the acenaphthylene-d2
molecule, it is also possible that there is some contribution from
combination bands. Styrene-d3, a molecule that is similar to
acenaphthylene-d2 in that it is fully conjugated and contains
only sp2 hypridized carbon atoms, exhibits several combination
bands in the 2800-3100 cm-1 region.21 Therefore, it is possible
that some small peaks in the FTIR spectrum for acenaphthylene-
d2 may also be due to combination bands. However, combination
bands for styrene-d3 were observed to be quite weak,21 and it
is expected that the same would be true for acenaphthylene-d2.
Although there are some differences in relative intensity, the
C-H vibrational frequencies of acenaphthylene-d2 chemisorbed
onto the Si(001) surface (Figure 4c) are nearly identical to those
of the multilayer film (Figure 4d). This similarity suggests that
adsorption onto the Si(001) surface leaves the naphthalene
portion of the molecule relatively unperturbed. Unfortunately,
frequency-dependent changes in detector response, combined
with the low number of D atoms in each molecule, make it
difficult to detect the C-D vibrations of the acenaphthylene-d2
monolayer. Consequently, we also investigated the adsorption
of acenaphthylene-d6, in which all H atoms on the naphthalene

Interface between a Polycyclic Aromatic Molecule and Si(001)
J. Phys. Chem. B, Vol. 107, No. 1, 2003 227
ring are substituted by D, leaving only H atoms that are part of
the 1,2-alkene group.
Figure 4e shows the FTIR spectrum for a Si(001) surface
that has been exposed to 10 L (5 × 10-8 Torr, 200 s)
acenaphthylene-d6. This spectrum is characterized by a strong
C-D stretch at 2275 cm-1 and a very weak but broad feature
extending from 2825 to 3000 cm-1. Notably, no significant
absorbance is observed above 3000 cm-1, even though the
spectrum in this region appears quite flat and with good signal-
to-noise.
Figure 5. Three possible bonding configurations for acenaphthylene
adsorbed on the Si(001) surface. (a) Adsorption through the 1,2-alkene
group to form a four-membered Si2C2 ring. (b) Adsorption through the
naphthalene portion of the ring to form a four-membered Si2C2 ring.
(c) Adsorption through the naphthalene portion of the ring resulting in
a [4+2] product.
For comparison, Figure 4f shows the FTIR spectrum of a
multilayer of acenaphthylene-d6. The multilayer shows peaks
at 2253, 2275, and 2282 cm-1 and a shoulder at ∼2243 cm-1
,
acenaphthylene/Si(001). For the parent acenaphthylene mol-
ecule, the changes in the FTIR spectrum upon adsorption
indicate that some C atoms change from sp2 to sp3 hybridization,
resulting in the appearance of new C-H features at 2888 and
2944 cm-1 (Figure 4a). A comparison of the spectra for
acenaphthylene-d2 (Figure 4c) and acenaphthylene (Figure 4a)
shows that both molecules have similar spectral features in the
3000-3100 cm-1 region. Since acenaphthylene-d2 has only H
atoms on the naphthalene portion of the molecule, the FTIR
peaks at 3036 and 3059 cm-1 can be attributed to C-H
stretching of the naphthalene ring. These FTIR peaks are
identical (within the resolution of the experiment) to the
acenaphthylene FTIR peaks at 3034 and 3059 cm-1. Therefore,
we can conclude that the FTIR peaks at 3034 and 3059 cm-1
for acenaphthylene are due to naphthalene ring C-H absorption.
Also, the presence of intense vibrational features >3000 cm-1
for chemisorbed acenaphthylene-d2 (Figure 4c) along with the
absence of these features for acenaphthylene-d6 (Figure 4e)
strongly suggests that chemisorption of acenaphthylene occurs
primarily through the 1,2-alkene portion of the molecule.
The small peaks at 2885 and 2938 cm-1 (sp3 hybridized C-H
stretching region) for acenaphthylene-d2/Si(001) appear similar
to the 2888 and 2944 cm-1 peaks observed for acenaphthylene/
Si(001). The appearance of low-frequency FTIR peaks for
acenaphthylene-d2/Si(001) suggests that either a small amount
of acenaphthylene-d2 bonds to the surface through the naph-
thalene ring or that the deuteration may be less than 100%
complete. However, a comparison of the intensity of these peaks
shows that the 2885 and 2938 cm-1 peaks of acenaphthylene-
d2/Si(001) (Figure 4c) are much lower in intensity than the 2888
and 2944 cm-1 peaks of acenaphthylene/Si(001) (Figure 4a),
while the C-H peaks at higher frequency have similar intensity.
This suggests that most (but not all) acenaphthylene-d2 mol-
ecules bond through the 1,2-alkene group, with some smaller
fraction (<10% based on STM images) possibly bonding
through the naphthalene ring.
each within the C-D stretching region, and a sharp C-H
vibration at 3094 cm-1 along with several other weak vibrations
between 2800 and 3150 cm-1. Once again, it is possible that
some of these peaks may be due to combination bands.21
Comparing the spectrum of the chemisorbed molecule (Figure
4e) to the multilayer film (Figure 4f) shows that upon interacting
with the surface, the vibration at 3094 cm-1, which is due to
sp2 hybridized C-H stretching of the parent molecule (Figure
4f), disappears and is replaced by a new feature at much lower
frequency (2800-3000 cm-1) (Figure 4e), characteristic of sp3
hybridized carbon. In contrast, the C-D vibrations between
2200 and 2300 cm-1 (Figure 4f) are almost completely
unaffected by bonding to the surface (cf. Figure 4e). This
comparison strongly suggests that bonding of acenaphthylene-
d6 to the Si(001) surface occurs almost exclusively via the 1,2-
alkene group, leaving the naphthalene ring nearly unperturbed.
In summary, STM shows that more than 90% of the
acenaphthylene molecules are adsorbed to Si(001) through a
single bonding configuration. FTIR spectra show that (1)
acenaphthylene reacts with the Si(001) surface in such a way
as to give rise to unsaturated C-C bonds within the molecules,
(2) the CdC bond of the 1,2-alkene group is saturated to form
a C-C bond, and (3) the naphthalene ring portion of acenaph-
thylene is mostly unperturbed.
V. Discussion
Evaluation of the experimental data for attachment of
acenaphthylene to the Si(001) surface reveals several pieces of
information that help to establish the bonding geometry. First,
there is a unique STM feature that accounts for more than 90%
of the adsorbed molecules (Figure 3a). Even though the STM
images cannot definitively determine the bonding configuration,
they do show that there is a predominant species, thereby
establishing that the FTIR spectra can be treated as arising from
molecules in a single configuration instead of a mixture of
configurations.
Analysis of the bonding configuration from FTIR hinges upon
several points. The first is that the C-H stretching vibrations
of sp2 hybridized carbon (as in alkenes) virtually always occur
at frequencies >3000 cm-1, while C-H stretching vibrations
of sp3 hybridized carbon (as in saturated alkanes) are nearly
always <3000 cm-1. Previous studies with a variety of alkenes
and more complex organic molecules show that this trend is
also true for molecules bonded to Si surfaces.7,9,22-25 When
unsaturated molecules such as styrene bond to the Si(001)
surface, the high-lying vibrational modes are shifted to lower
frequency due to the transition from sp2 to sp3 hybridization
and by the donation of electron density from silicon to the
carbon atom.9
There are several possible bonding configurations that could
be formed when acenaphthylene adsorbs to the Si(001) surface.
Three of these are shown in Figure 5. The absence of significant
IR absorbance in the Si-H stretching region (2000-2100 cm-1
)
suggests that the majority of the adsorbed species bond without
dissociation. On the basis of the IR data, we can rule out several
configurations. Since the naphthalene ring remains unperturbed
after adsorption, bonding configurations of the type shown in
Figure 5b and 5c and other similar or more complex schemes
involving the naphthalene portion of the molecule can be ruled
out. Coupling the lack of naphthalene ring interaction with
evidence for the side group changing from an alkene (sp2) to
an alkane (sp3) species shows that the only reasonable bonding
configuration is the one shown in Figure 5a. Therefore, we
conclude that acenaphthylene adsorbs to the Si(001) surface
through the side group in a manner similar to a [2+2]-
Shifts in vibrational frequency upon chemisorption give
important information about the bonding configuration(s) for

228 J. Phys. Chem. B, Vol. 107, No. 1, 2003
Schwartz et al.
(2) Hamers, R. J.; Coulter, S. K.; Ellison, M. D.; Hovis, J. S.; Padowitz,
D. F.; Schwartz, M. P.; Greenlief, C. M.; Russell, J. N., Jr. Acc. Chem.
Res. 2000, 33, 617-624.
(3) Cao, X.; Hamers, R. J. J. Am. Chem. Soc. 2001, 123, 10988-10996.
(4) McGinness, J.; Corry, P.; Proctor, P. Science 1973, 183, 853-855.
(5) Aviram, A.; Ratner, M. Chem. Phys. Lett. 1974, 29, 277-283.
Figure 6. Reductive silylation of acenaphthylene to form 1,2-bis-
(trimethylsilyl)acenaphthene, from ref 30.
(6) Collier, C. P.; Wong, E. W.; Belohradsky, M.; Raymo, F. M.;
Stoddart, J. F.; Kuekes, P. J.; Williams, R. S.; Heath, J. R. Science 1999,
285, 391-394.
(7) Coulter, S. K.; Hovis, J. S.; Ellison, M. D.; Hamers, R. J. J. Vac.
Sci. Technol. 2000, 18, 1965-1970.
cycloaddition reaction and the naphthalene portion of the
molecule remains relatively unperturbed.
(8) Ellison, M. D.; Hamers, R. J. J. Phys. Chem. B 1999, 103, 6243-
Selective chemistry of acenaphthylene has previously been
observed for reactions with several substituted ethylene species.
The 1,2-alkene group of acenaphthylene preferably reacts
through a [2+2] cycloaddition with acrylonitrile,26 tetracyano-
ethylene (TCNE),27 and several more complex substituted
ethylene species (each under irradiation).28 Reactions under
irradiation between TCNE and acenaphthylene resulted in
several products (in addition to the [2+2] product), all involving
the 1,2-alkene portion of the molecule.29 Reductive silylation
of acenaphthylene also proceeded through the 1,2-alkene group
to form 1,2-bis-(trimethylsilyl)acenaphthene, as shown in Figure
6.30 In each case, even under very reactive conditions, reaction
of acenaphthylene occurred through the 1,2-alkene portion of
the molecule, leaving the naphthalene ring unperturbed.
Several Si(001) surface reactions with CdC containing
molecules, in which the CdC portion of the molecule reacts
with a SidSi dimer to form a four-member C2Si2 ring, have
also been observed.2,9,23,31-41 In particular, styrene, a molecule
containing an aromatic ring species with a conjugated vinyl side
group, attaches selectively through the vinyl side group, resulting
in an intact aromatic species tethered to the Si(001) surface.9
Selectivity for styrene originates in the fact that the interaction
between the π electrons of the vinyl group and the end of a
SidSi dimer is attractive at relatively large separations, whereas
interaction with the ring involves an energy barrier. This leads
to selective bonding of styrene to the Si(001) surface through
the vinyl group into a [2+2] adduct via a low-symmetry
pathway.9 While computational work has not been done for this
study, it is likely that a similar explanation can account for the
reaction of acenaphthylene with the Si(001) surface. A reaction
barrier that is larger for attachment through the naphthalene
portion of the molecule than for the 1,2-alkene side group would
result in a kinetically mediated pathway that leads to the product
shown in Figure 5a.
6251.
(9) Schwartz, M. P.; Ellison, M. D.; Coulter, S. K.; Hovis, J. S.;
Hamers, R. J. J. Am. Chem. Soc. 2000, 122, 8529-8538.
(10) Cao, X.; Coulter, S. K.; Ellison, M. D.; Liu, H.; Liu, J.; Hamers,
R. J. J. Phys. Chem. B 2001, 105, 3759-3768.
(11) Coulter, S. K.; Schwartz, M. P.; Hamers, R. J. J. Phys. Chem. B
2001, 105, 3079-3087.
(12) Fang, L.; Liu, J.; Coulter, S. K.; Cao, X.; Schwartz, M. P.; Hamers,
R. J. Surf. Sci. 2002, 514, 362-375.
(13) Hacker, C. A.; Hamers, R. J. J. Phys. Chem. B, submitted for
publication.
(14) Liang, C. X.; Newton, M. D. J. Phys. Chem. 1993, 97, 3199-
3211.
(15) Sachs, S. B.; Dudek, S. P.; Hsung, R. P.; Sita, L. R.; Smalley, J.
F.; Newton, M. D.; Feldberg, S. W.; Chidsey, C. E. D. J. Am. Chem. Soc.
1997, 119, 10563-10564.
(16) Meyer zu Heringdorf, F. J.; Reuter, M. C.; Tromp, R. M. Nature
2001, 412, 517-520.
(17) Hamers, R. J.; Tromp, R. M.; Demuth, J. E. Phys. ReV. B 1986,
34, 5343-5357.
(18) Bailey, R. J.; Card, P. J.; Shechter, H. J. Am. Chem. Soc. 1983,
105, 6096-6103.
(19) Anderson, A. G., Sr.; Anderson, R. G. J. Am. Chem. Soc. 1955,
77, 6610-6611.
(20) Baranovic, G.; Jordanov, B.; Schrader, B. J. Mol. Struct. 1994, 323,
103-115.
(21) Condirston, D. A.; Laposa, J. D. J. Mol. Spectrosc. 1976, 63, 466-
477.
(22) Yoshinobu, J.; Tsuda, H.; Onchi, M.; Nishijima, M. Solid State
Commun. 1986, 60, 801-805.
(23) Hamers, R. J.; Hovis, J. S.; Lee, S.; Liu, H.; Shan, J. J. Phys. Chem.
1997, 101, 1489-1492.
(24) Teplyakov, A. V.; Kong, M. J.; Bent, S. F. J. Am. Chem. Soc. 1997,
119, 11100-11101.
(25) Lopinski, G. P.; Fortier, T. M.; Moffatt, D. J.; Wolkow, R. A. J.
Vac. Sci. Technol., A 1998, 16, 1037-1042.
(26) Plummer, B. F.; Hall, R. A. Chem. Commun. 1970, 44-45.
(27) Haga, N.; Nakajima, H.; Takayanagi, H.; Tokumaru, K. Chem.
Commun. 1997, 1171-1172.
(28) Plummer, B. F.; Songster, M. J. Org. Chem. 1990, 55, 1368-1372.
(29) Haga, N.; Nakajima, H.; Takayanagi, H.; Tokumaru, K. J. Org.
Chem. 1998, 63, 5372-5384.
VI. Conclusion
(30) Laguerre, M.; Felix, G.; Dunogues, J.; Calas, R. J. Org. Chem.
1979, 44, 4275-4278.
(31) Yoshinobu, J.; Tsuda, H.; Onchi, M.; Nishijima, M. J. Chem. Phys.
1987, 87, 7332-7340.
(32) Bozack, M. J.; Taylor, P. A.; Choyke, W. J.; Yates, J. T., Jr. Surf.
Sci. 1986, 177, L933-L937.
(33) Clemen, L.; Wallace, R. M.; Taylor, P. A.; Dresser, M. J.; Choyke,
W. J.; Weinberg, W. H.; Yates, J. T., Jr. Surf. Sci. 1992, 268, 205-216.
(34) Hovis, J. S.; Liu, H.; Hamers, R. J. J. Phys. Chem. B 1998, 102,
6873-6879.
π-Conjugation is responsible for most of the interesting
electronic characteristics of organic molecules. Therefore, simple
schemes that lead to selective attachment of large π-systems to
semiconductor surfaces are highly desirable. We have developed
a simple, selective attachment scheme in which the 1,2-alkene
group of acenaphthylene is used to couple an extended π-system
to the Si(001) surface. The chemistry of acenaphthylene on
Si(001) may be a model system for developing ways to couple
more complex π-systems to semiconductor surfaces.
(35) Konecny, R.; Doren, D. J. Surf. Sci. 1998, 417, 169-188.
(36) Cheng, C. C.; Wallace, R. M.; Taylor, P. A.; Choyke, W. J.; Yates,
J. T., Jr. J. Appl. Phys. 1990, 67, 3693-3699.
(37) Mayne, A. J.; Avery, A. R.; Knall, J.; Jones, T. S.; Briggs, G. A.
D.; Weinberg, W. H. Surf. Sci. 1993, 284, 247-256.
(38) Imamura, Y.; Morikawa, Y.; Yamasaki, T.; Nakatsuji, H. Surf. Sci.
Acknowledgment. The authors acknowledge support from
the National Science Foundation Grant DMR-9901293 and the
U.S. Office of Naval Research. R. J. Hamers and M. P. Schwartz
also thank the S. C. Johnson Co. Distinguished Fellowship
program for support.
1995, 341, L1091-L1095.
(39) Mayne, A. J.; Cataldi, T. R. I.; Knall, J.; Avery, A. R.; Jones, T.
S.; Pinheiro, L.; Hill, H. A. O.; Briggs, G. A. D.; Pethica, J. B.; Weinberg,
W. H. Faraday Discuss. 1992, 94, 199-212.
References and Notes
(40) Liu, H.; Hamers, R. J. J. Am. Chem. Soc. 1997, 119, 7593-7594.
(1) Appelbaum, J. A.; Baraff, G. A.; Hamann, D. R. Phys. ReV. B 1976,
14, 588.
(41) Huang, C.; Widdra, W.; Weinberg, W. H. Surf. Sci. 1994, 315,
L953-L958.