Structure of Thiol Self-Assembled Monolayers Commensurate with the GaAs (001) Surface.

Subscriber access provided by Service des bibliothèques | Université de Sherbrooke
Langmuir is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Article
Structure of Thiol Self-Assembled Monolayers
Commensurate with the GaAs (001) Surface
Oleksandr Voznyy, and Jan J. Dubowski
Langmuir, 2008, 24 (23), 13299-13305 • Publication Date (Web): 30 October 2008
Downloaded from http://pubs.acs.org on November 29, 2008
More About This Article
Additional resources and features associated with this article are available within the HTML version:
• Supporting Information
• Access to high resolution figures
• Links to articles and content related to this article
• Copyright permission to reproduce figures and/or text from this article

Structure of Thiol Self-Assembled Monolayers Commensurate with the
GaAs (001) Surface
Oleksandr Voznyy and Jan J. Dubowski*
Department of Electrical and Computer Engineering, Centre of Excellence for Information Engineering
(CEGI), UniVersite´ de Sherbrooke, Sherbrooke, Que´bec J1K 2R1, Canada
ReceiVed April 4, 2008. ReVised Manuscript ReceiVed September 15, 2008
Observed properties of thiol self-assembled monolayers (SAMs) on GaAs (001) surfaces can be explained by the
presence of surface reconstructions, but their exact form is generally unknown. We propose a new approach to
modeling the SAM-surface interface based on using alkanethiol dense packing structures as a starting point and
adjusting the surface reconstruction to accommodate them. Obtained in such a way, model SAMs adsorb along the
trenches in the [110] direction and exhibit a 19° tilt and ( 45° twist angles, in agreement with available experimental
data. The molecules of the SAM bind to both Ga and As, and cover only 50% of the available surface sites. The
requirements for the SAM formation process to achieve the proposed structures are discussed.
1. Introduction
Self-assembled monolayers (SAMs) of organic molecules on
solid substrates are of high technological and fundamental
interests.1,2 Particularly on semiconductor surfaces, their potential
applications are in bio-3,4 and chemical sensing,2 molecular
electronics,5,6 passivation,7,8 nanolithography,9 and precursors
for growth of other compounds.10
Recently, the procedure for the growth of highly ordered and
dense alkanethiol SAMsonGaAs (001)was reported.11Molecules
in such SAMs are tilted ∼14° from the surface normal,11-15
compared to previously reported tilts of∼57°,16-18 but the factors
affecting the quality of themonolayers are not yetwell understood.
The structure with 14° tilt suggests denser SAMs than those on
Au (111) with the tilt of ∼30°.11,13,19 However, the spacing
between thiols in crystalline phase is incommensurate with the
underlying square GaAs lattice, and the exact structure of the
thiol-substrate interface remains unknown. In previous works
it was suggested that maximum density molecular packing may
be important in the film assembly process, and some disruption
of the ideal GaAs lattice should arise upon SAM formation, at
the same time leaving a significant fraction of surface atoms
unbound.13,14 It should be noted, however, that the morphology
of semiconductor surfaces is not represented by the areas of
significantly large atomically flat planes, such as on metal or
nanocrystalline oxide surfaces, despite the fact that the bulk
substrate is monocrystalline. Even ideal GaAs surfaces prepared
using molecular beam epitaxy exhibit reconstructions with the
roughness of 1-2 atomic layers and the surface unit cell size of
1-2 nm.20 The wet etching procedure used for thiol deposition
is expected to disrupt the surface even more. Formation of the
amorphous As overlayer and microroughness are the examples
of the possible drastic changes after etching.12,21 Thus, the
complexity of the thiol-GaAs interface is expected. However,
it remains unclear whether thiol-thiol interactions play a
significant role in the formation of the final surface morphology,
or SAMs simply adapt to the available surface reconstruction at
any given point.
The bonding chemistry of thiolates to GaAs surface has been
of great debate in literature (for a comprehensive review of the
problemsee, e.g., ref 14).X-rayphotoelectron spectroscopy (XPS)
of the samples prepared using thiol deposition from liquid can
resolve the presence of a component inAs 3d spectra, presumably
related toAs-Sbinding.11-16,22However, this component cannot
be unambiguously assigned to As-S, since it strongly overlaps
with As0 component and with As-H, which is usually disre-
garded, although temperature programmed desorption (TPD) and
theoretical studies on Ga-rich surfaces suggest that hydrogen
may stay on the surface.23-25 The spectra ofGa 3d region exhibit
no significant differences between freshly etched and thiolated
samples, which is usually interpreted as the absence of Ga-S
bonding. High-resolution XPS studies14,22 show the presence of
the Ga component shifted less than 0.5 eV from the bulk GaAs
peak and assigned toGa2O3 or surfaceGa.However, experiments
* Corresponding author. E-mail: jan.j.dubowski@usherbrooke.ca.
(1) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M.
Chem. ReV. 2005, 105, 1103.
(2) Ulman, A. Chem. ReV. 1996, 96, 1533.
(3) Ding, X.; Moumanis, K.; Dubowski, J. J.; Frost, E. H.; Escher, E. Appl.
Phys. A: Mater. Sci. Process. 2006, 83, 357.
(4) Tanaka, M.; Sackmann, E. Phys. Stat. Sol. A 2006, 203, 3452.
(5) Cai, L. T.; Cabassi, M. A.; Yoon, H.; Cabarcos, O. M.; McGuiness, C. L.;
Flatt, A. K.; Allara, D. L.; Tour, J. M.; Mayer, T. S. Nano Lett. 2005, 5, 2365.
(6) Lodha, S.; Carpenter, P.; Janes, D. B. J. Appl. Phys. 2006, 99.
(7) Lunt, S. R.; Ryba, G. N.; Santangelo, P. G.; Lewis, N. S. J. Appl. Phys.
1991, 70, 7449.
(8) Moumanis, K.; Ding, X.; Dubowski, J. J.; Frost, E. H. J. Appl. Phys. 2006,
100, 034702.
(9) Amro, N. A.; Xu, S.; Liu, G. Y. Langmuir 2000, 16, 3006.
(10) Nishimura, K.; Nagao, Y.; Sakai, K. J. Cryst. Growth 1993, 134, 293.
(11) McGuiness, C. L.; Shaporenko, A.; Mars, C. K.; Uppili, S.; Zharnikov,
M.; Allara, D. L. J. Am. Chem. Soc. 2006, 128, 5231.
(12) Adlkofer, K.; Tanaka, M. Langmuir 2001, 17, 4267.
(13) McGuiness, C. L.; Blasini, D.; Masejewski, J. P.; Uppili, S.; Cabarcos,
O. M.; Smilgies, D.; Allara, D. L. ACS Nano 2007, 1, 30.
(14) McGuiness, C. L.; Shaporenko, A.; Zharnikov,M.;Walker, A. V.; Allara,
D. L. J. Phys. Chem. C 2007, 111, 4226.
(15) Nesher, G.; Vilan, A.; Cohen, H.; Cahen, D.; Amy, F.; Chan, C.; Hwang,
J. H.; Kahn, A. J. Phys. Chem. B 2006, 110, 14363.
(16) Jun, Y.; Zhu, X. Y.; Hsu, J. W. P. Langmuir 2006, 22, 3627.
(17) Sheen, C. W.; Shi, J. X.; Martensson, J.; Parikh, A. N.; Allara, D. L.
J. Am. Chem. Soc. 1992, 114, 1514.
(18) Zhang, Q.; Huang, H. Z.; He, H. X.; Chen, H. F.; Shao, H. B.; Liu, Z. F.
Surf. Sci. 1999, 440, 142.
(19) Schreiber, F. Prog. Surf. Sci. 2000, 65, 151.
(20) Schmidt, W. G. Appl. Phys. A: Mater. Sci. Process. 2002, 75, 89.
(21) Aspnes, D. E.; Studna, A. A. Appl. Phys. Lett. 1985, 46, 1071.
(22) Shaporenko,A.;Adlkofer,K.; Johansson, L. S.O.; Tanaka,M.; Zharnikov,
M. Langmuir 2003, 19, 4992.
(23) Yi, S. I.; Chung, C. H.; Weinberg, W. H. J. Vac. Sci. Technol. A 1997,
15, 1168.
(24) Donev, S.; Brack, N.; Paris, N. J.; Pigram, P. J.; Singh, N. K.; Usher, B. F.
Langmuir 2005, 21, 1866.
(25) Voznyy, O.; Dubowski, J. J. J. Phys. Chem. C 2008.
13299Langmuir 2008, 24, 13299-13305
10.1021/la8010635 CCC: $40.75  2008 American Chemical Society
Published on Web 10/30/2008

on adsorption of thiols and hydrogen sulfide on Ga-containing
surfaces23,24 and our previous simulations25 suggest that thiols
preferentially bind toGa, even if bothGa andAs sites are available
on the surface. Analysis of charges on atoms in our ab initio
calculations25,26 suggests that Ga-S related peaks should be
between the above-mentioned surfaceGaandbulkGacomponents
and thus be practically undetectablewithXPS. The clear presence
of a Ga-S signal in recent time-of-flight secondary ion mass
spectrometry (ToF-SIMS) results14 supports our conclusion that
the nature of thiol-GaAs bonding cannot be unambiguously
determined based solely on XPS.
Another important question is whether thiol SAMs can be
used for the passivation of GaAs surface, i.e., improvement and
stabilization of electronic properties of GaAs by unpinning the
Fermi level (removal of the band bending) and reduction of the
surface recombination velocity. Passivation is achieved by
suppression of surface states in the bandgap (arising mainly due
to surface oxides) and by protecting the surface from further
oxidation. Experimental reports, however, do not provide an
unambiguous picture concerning the role of thiols in passivation
of the GaAs surface. Significant improvement of photolumi-
nescence intensitywas reported for intrinsic andn-type samples.7,8
Also, XPS data indicate reasonable protection from oxida-
tion,13,14,22 andRaman spectroscopy shows the reduction of band
bending and improvement of long-term stability.27 However,
more recent Raman spectroscopy data on samples covered with
presumably higher-quality SAMs suggest that the removal of
band bending is not achieved on n-type samples but is noticeable
on p-type samples.14
In view of the recently revived interest in thiol SAMs growth
on GaAs and the limited amount of experimental techniques
suitable for analysis of thismaterial system, theoreticalmodeling
can provide information crucial for understanding the nature of
the thiol SAM interface with GaAs and other semiconductors.
In this work, we propose a new approach for searching the exact
SAM-surface interface for cases when surface reconstruction
is unknown. The approach is based on using thiol dense packing
as a starting point and fitting of the surface reconstruction to
achieve commensurability with the SAM. Obtained in such a
way, thiol SAM structures on GaAs(001) are in agreement with
available experimental data13 and provide new insight into the
bonding chemistry and passivation properties of these SAMs.
The requirements for the SAM formation process to achieve the
proposed structures are discussed.
2. Model
Experimental infrared reflection spectroscopy (IRS) data show
a high degree of crystallinity of alkanethiol SAMs on GaAs
(001), comparable to that of bulk alkanethiols.11,13,15,17 Fitting
of simulated IRpeak intensities to experimental data also provides
the tilt (deviation from surface normal) and twist (rotation about
the chain axis) angles of the molecules in the SAM.11,13,17 Since
the surface structure of thewet-etchedGaAs substrate is unknown,
the crystalline structure of thiols remains the only known starting
point to proceed with modeling of the SAM. This is in contrast
to conventional modeling approaches, where surface structure
is assumed to be known and a SAM is adjusted on it, e.g., as
used for alkanethiols on Au (111).28,29 The importance of
thiol-thiol interactions was realized previously and the as-
sumption of SAM crystallinity was used in the search for thiol
SAMs structure onGaAs (001).13,14 However, it had not resulted
in a successful model, since GaAs surface was still assumed to
be atomically flat, and no accurate quantitative characterization
of SAMs (distances and angles between thiols) was used.
On the basis of the weak dependence of CH3 IRS peak
intensities on the change from an odd to an even amount of
carbons in thiol chains, it was suggested that the SAM should
contain two differently oriented chains per unit cell, resulting in
a herringbone packing with a 90° setting angle between the
C-C-C planes.13 This type of structure was resolved for c(4
× 2) SAMs of thiols on gold,30,31 Langmuir monolayers,32 and
bulk alkanes,33,34 on the basis of the presence of splitting in CH2
scissor deformationmode or grazing incidenceX-ray diffraction
(GIXRD) data. However, such a splitting can be resolved only
with low temperature measurements that were not performed for
thiols on GaAs.
To verify the experimental suggestion about chain orientations,
we investigated both monoclinic and orthorhombic structures,
with one and two chain orientations per unit cell, respectively.
To describe the thiol-thiol interactions, we used theDREIDING
force field35 as implemented in the Accelrys Discovery Studio
package. Empirical molecular mechanics (MM) is an accurate
and reliable theoretical tool for the description of interactions
between small organicmolecules, as opposed to density functional
theory (DFT) calculations, which generally fail to account for
van derWaals attractions. Figure 1 shows the twodensest possible
packing structures of alkanethiols obtained in our simulations.
Our observations of thiol packing coincide with previously
suggested ideas of interlocking, reduction of empty volume, and
steric limits.29 Our structural parameters (shown in Figure 1) are
within 1% error from the most recent experiments.33,34
Unlike previous similar studies of thiols on gold that failed
to resolve the exact SAM structure,28,29 we usedMM to describe
only the interactions between thiols and not thosewith the surface.
As has been previously found in our ab initio calculations, thiols
bind to GaAs surface sites via the formation of the single highly
directed covalent bonds.25,26 This is simpler than binding to a
Au surface,where single, double, and triple coordination of sulfur
is possible. Such a strictly defined bonding geometry on GaAs
makes the matching of the SAM to the surface more straight-
forward and allows one to replace the combination of DFT and
MMsimulations, needed to describe correctly both thiol-surface
and thiol-thiol interactions, with a simple geometrical fitting of
two crystalline structures. Both DFT and MM simulations were
still used in our work to gain deeper understanding of the
adsorption process and to facilitate the search of new possible
geometries.
3. Results
3.1. Discrete Tilts. Both structures in Figure 1 have an area
per molecule of 18.4 Å2,33,34 the projection of which on the
substrate increases as 1/cosθwith tilt. In the case of an atomically
flat surface, which is an assumption made for IRS fitting, the
area per GaAs surface atom is 16 Å2. For the reported 14°
tilt11-13,15 this suggests a 100% coverage and a lattice mismatch
of at least 9%. For 50% coverage of surface sites, the available
(26) Voznyy, O.; Dubowski, J. J. J. Phys. Chem. B 2006, 110, 23619.
(27) Dorsten, J. F.;Maslar, J. E.; Bohn, P.W.Appl. Phys. Lett. 1995, 66, 1755.
(28) Li, T. W.; Chao, I.; Tao, Y. T. J. Phys. Chem. B 1998, 102, 2935.
(29) Ulman, A.; Eilers, J. E.; Tillman, N. Langmuir 1989, 5, 1147.
(30) Laibinis, P. E.;Whitesides, G.M.; Allara, D. L.; Tao, Y. T.; Parikh, A. N.;
Nuzzo, R. G. J. Am. Chem. Soc. 1991, 113, 7152.
(31) Nuzzo, R. G.; Korenic, E. M.; Dubois, L. H. J. Chem. Phys. 1990, 93,
767.
(32) Kaganer, V. M.; Mohwald, H.; Dutta, P. ReV. Mod. Phys. 1999, 71, 779.
(33) Kitaigorodskii, A. I. Organic Chemistry Crystallography; Consultants
Bureau: New York, 1961.
(34) Takahashi, Y.; Kumano, T. J. Polym. Sci. B 2004, 42, 3836.
(35) Mayo, S. L.; Olafson, B. D.; Goddard, W. A. J. Phys. Chem. 1990, 94,
8897.
13300 Langmuir, Vol. 24, No. 23, 2008 Voznyy and Dubowski

area per thiol increases to 32 Å2. Preserving the densest packing
of thiols and taking into account the cos θ dependence, this can
be accommodated by a 55° tilt, close to previously reported
experimental values.16-18
For the dense packing structures shown in Figure 1, the flat
surface condition and the same orientation of S-surface bonds
can be achieved only at discrete values of tilts, the minimal of
which is 32° (see Supporting Information). This discreteness of
tilt values is responsible for the known phase transitions in
Langmuir films.32,33 For other tilt angles, the sulfur-substrate
bond (and, as a result, the CH3 endgroup) orientation alternates
from molecule to molecule and can significantly diminish the
odd-even dependence of CH3 IR intensities, as observed
experimentally,13 even formonoclinic (singly twisted) structure.
Tilting a densely packed slab of thiols to reproduce experi-
mental 43° twist and 14° tilt values11,13 results in a relatively
rough interface and randomly oriented headgroups (see Sup-
porting Information). Such conditions could be matched, e.g., to
an amorphous As layer covering the GaAs surface. The presence
of elemental As is always observed after wet etching of GaAs
and even after SAM formation.12,14,15,22 Stable thiol monolayers
with the molecules standing almost upright were also reported
on explicitly created thick As overlayers.12 The choice by the
SAM to tilt by 14°, which increases the mismatch of area per
thiol and area per surface site evenmore, compared to that already
present at 0° tilt, can be thought of as the way to enhance the
disorder in the headgroup orientation to better match the
disordered surface.
However, for a 45° twist, which is close to the experimentally
estimated 43°, the monolayer shows a long-range order in
headgroup orientations, contradicting the amorphous surface
hypothesis. Adsorption of a SAM on the amorphous overlayer
would alsomake possible tilting in any direction, thus, producing
multiple peaks inGIXRDmaps for any azimuth, e.g., as observed
for Langmuir monolayers on liquids.32 However, experimental
GIXRD results of thiols on GaAs13 reveal peaks only along
distinct azimuthal directions,which suggests an ordered substrate
and, likely, a different interface than in the case of an explicitly
created As overlayer.12
3.2. Structure ofAlkanethiol SAMonGaAs (001) Surface.
Two equivalent interplane distances in GIXRD data suggest that
thiols tilt in either nearest-neighbor (NN)or next-nearest-neighbor
(NNN) thiol direction32 and pack, likely, in the orthorhombic
structure (see Figure 1). The azimuthal orientation of the
equivalent peaks and GIXRD-derived NN thiol distances13
indicate that the 5 Å NN distance lies along the 0° azimuth, i.e.,
along GaAs [110] (or [-110]) direction. The angle of 115°
between equivalent GIXRD peaks, that is greater than 111° in
ideally packed structures, suggests a tilt along the NNN thiol
direction, i.e., along the 7.37 Å diagonal.
Our efforts to fit either of the two structures shown in Figure
1 on a square GaAs lattice (i.e., flat surface) were unsuccessful
for any combination of SAMparameters (tilt value and direction,
SAM orientation relative to the substrate) due to significant
differences in NN distances and in the symmetries of the SAM
and the substrate. Moreover, MM simulations attempting to put
thiols one after another on all available sites of a flat surface
resulted in the formation of gauche defects and the impossibility
of 100% coverage due to steric repulsion between themolecules.
SinceGIXRDdata shows preferential alignment of thiols along
the [110] (or [-110]) direction, it was suggested that adsorption
of thiols starts at step edges or etch-pit edges,13 which are known
to appear following the wet etching procedure.21 Those step
edges expose the less reactive (111) Ga plane and run along
[110]. Trenches exposing Ga or As atoms from a second (and
even third) surface atomic layer naturally exist on all GaAs (001)
reconstructions and run along [110] for As-rich reconstructions
andalong [-110] forGa-richones.20Ourpreviouswork suggested
that thiols preferentially bind toGa sites that have empty dangling
bonds.25 Those dangling bonds formweak bondswith the thiols’
sulfur lone pairs, increasing the physisorbed thiol dwell time and
its chances to chemisorb. In the case when ammonia is added
to the thiol solution,11,13,14 thiolsmay dissociate into SR- ions,36
which are also expected to preferentially adsorb on Ga empty
dangling bonds rather than on As filled ones.
Introduction of the step edges indeed allowed achieving
geometrical commensurability of the SAM with the surface, but
only along the step edge and not with the rest of the surface.
Thus, further fittingwas performedby adding thiols on the surface
one after another and performing MM geometry optimization,
mimicking the real adsorption process of thiols. Figure 2 shows
that a thiol adsorbed on a Ga site in a trench slightly tilts in the
direction of the hollow between As dimers. This allows one to
accommodate the sp3 hybridization ofGa and the p3 hybridization
of S and, at the same time, to reduce the steric repulsion of the
first CH2 unit from the surface, as was described previously.26
The resulting ∼45° twist coincides with suggestions from IRS
data fitting.11 Having one thiol adsorbed, the steric repulsion
forbids the adsorption of new molecules on the closest Ga sites
at 4 Å distance. The closest next adsorption site would be a Ga
(36) Tajc, S. G.; Tolbert, B. S.; Basavappa, R.; Miller, B. L. J. Am. Chem. Soc.
2004, 126, 10508.
Figure 1. Monoclinic (a) and orthorhombic (b) packing structures of
bulk alkanes with interplane and NN distances indicated. Carbon chains
are shown as sticks with the topmost carbon as spheres, and hydrogen
atoms are omitted for clarity. van der Waals volumes of the molecules
are shown in background. Planes potentially responsible for GIXRD
peaks are shown as thick gray lines.
Structure of Thiol SAMs on GaAs Surfaces Langmuir, Vol. 24, No. 23, 2008 13301

site on the same side of the trench, which is 8 Å away, or a Ga
site on the other side of the trench, along the ∼5.65 Å diagonal
of the GaAs lattice. Thiols adsorbed along both sides of the
trench form a zigzag structure, reminiscent of the ideally packed
structure tilted in the NNN direction, and in agreement with the
above-mentioned expectations from GIXRD data. Further
adsorption sites are the As sites in the ridge of As dimers which
are located above the forbidden Ga atoms (see Figure 2). Those
sites coincide with the pockets in the zigzag structure of thiols
and provide the maximal van der Waals interaction with the
already adsorbedmolecules. It should be noted that to grow such
a SAM without defects there should be no adjacent ridges of As
dimers, which are observed, for example, in As-rich �2(2 × 4)
reconstruction.20 Figure 3 shows the resulting structure, which
requires a set of trench-ridge pairs rather than one step-edge.
This reconstruction differs from the known stable reconstructions
of freeGaAs (001) surfaces.20Weanticipate that the liquid etchant
together with adsorbed thiols is responsible for the creation of
such a structure.
Our model SAM tilts by 19° along the trench direction and
is less densely packed than the ideal crystalline structures shown
in Figure 1. The available spacing of 8Åalong the trench between
the surface sites participating in bonding with thiols (Figure 2)
is 2.5% larger than the corresponding distance of 7.8 Å in the
bulk thiol slab tilted in the NNN direction by 19°, a tilt providing
the flat-surface condition along the trench direction. In the
perpendicular direction, theGaAs lattice is 6% larger than required
by the ideally packed structure (Figure 3ab). Such a mismatch
reduces the extent of thiols interlocking; however, it does not
induce compressive strain neither in the SAM nor in the surface
structure. Reduced density SAMs provide less protection against
oxygen penetration to the surface and also allow thiols to pivot
about their binding sites, which may be responsible for the
broadening of diffraction peaks observed experimentally.13
In the proposed SAMmodel, thiols cover only 50%of exposed
surface atoms and have an area of 21.33 Å2 per thiol, which is
larger than 16 Å2 per thiol available on a flat surface at 100%
coverage. For ideally packed thiols, 21.33 Å2 would correspond
to a 30° tilt.A structurewith awider trench, exposing an additional
As row in the middle of the trench (as observed in As-rich �2(2
× 4) reconstruction20), can also accommodate a SAM (see
Supporting Information), but its density would be even smaller.
Another interesting feature of our model is that adsorption on
Ga sites located on the opposite sides of As ridges requires
opposite orientation of thiols (see the first vs third and third vs
fourth columns in Figure 3a and Figure 3b), which breaks the
periodicity observed for either of the thiol structures shown in
Figure 1, but can be nevertheless easily accommodated due to
the slightly smaller density of the SAM on surface. As a result,
thiols adsorbed on As ridges (columns 2 and 5) can be oriented
(twisted) perpendicular to either the left or right neighbor (Figure
3c), which may also be responsible for blurring of the GIXRD
signal. Thiol chains adsorbed on top of As ridges are forced to
orient the C-C-C plane near the surface along the ridge, to
accommodate the idealAs-Sdirection (Figure 3a). Only at some
distance from the surface can the longer chain thiols rotate to
accommodate the twist observed in the densest packing structures
(Figure 3c).
The SAM structure proposed in Figure 3 is in qualitative
agreementwith available experimental data. Despite the tilt angle
of 14-15° estimated from IRS fitting and near-edge X-ray
absorption fine structure (NEXAFS) data,11,13 the area per thiol
obtained from GIXRD data and corrected for cos θ dependence
suggests a tilt of ∼20° or a nonmaximum packing density of
Figure 2.Adsorption of thiols on a step edge exposing the Ga-rich (111)
plane.Dangling bonds, i.e., sites available for thiol adsorption, are shown
with dotted lines.
Figure 3. Proposed structure of an alkanethiol SAMand the GaAs (001)
surface reconstruction needed to accommodate it (a) compared to the
ideal orthorhombic structure tilted by the same angle (b). Columns 1
and 6 have different thiol twists in the ideal structure and our model.
Note also the shift between the 3rd and 4th columns in our model. Top
view along the chains (c) shows the packing density and relative
orientation ofmolecular chains. Thiols adsorbed onAs sites are indicated
with lighter gray.
13302 Langmuir, Vol. 24, No. 23, 2008 Voznyy and Dubowski