Dipole-induced structure in aromatic-terminated self-assembled monolayers: A study by near edge x-ray absorption fine structure spectroscopy

Article abstract of DOI:10.1063/1.1737303

Near edge x-ray absorption fine structure spectroscopy (NEXAFS) was used to study the structure of self-assembled monolayers presenting aromatic rings at a surface. It was observed that the fluorine substitution at asymmetric positions in the aromatic rings was used to generate a layer of dipoles at the surface of the monolayer. It was observed that the fluorine substituted aromatic rings were more ordered than unsubstituted aromatic rings by a factor of two based on the polarization dependence of the lowest C 1s to π^* transition. It was found that this result was consistent with the influence of dipole-dipole interactions and quadrupolar interactions between the aromatic groups due to the substitution of fluorine atoms.

Full text of DOI:10.1063/1.1737303

Dipole-induced structure in aromatic-terminated self-assembled monolayers—A study
by near edge x-ray absorption fine structure spectroscopy
Yan-Yeung Luk, Nicholas L. Abbott, J. N. Crain, and F. J. Himpsel
Citation: The Journal of Chemical Physics 120, 10792 (2004); doi: 10.1063/1.1737303
View online: http://dx.doi.org/10.1063/1.1737303
View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/120/22?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
The structure of the Au(111)/methylthiolate interface: New insights from near-edge x-ray absorption spectroscopy
and x-ray standing waves
J. Chem. Phys. 130, 124708 (2009); 10.1063/1.3102095
Polarized near-edge x-ray-absorption fine structure spectroscopy of C 60 -functionalized 11-amino-1-undecane
thiol self-assembled monolayer: Molecular orientation and Evidence for C 60 aggregation
J. Chem. Phys. 122, 154703 (2005); 10.1063/1.1880952
Polarized total-reflection x-ray absorption fine structure for self-assembled monolayer of zinc porphyrin at
air–water interface
J. Chem. Phys. 118, 10369 (2003); 10.1063/1.1580095
Combinatorial near-edge x-ray absorption fine structure: Simultaneous determination of molecular orientation
and bond concentration on chemically heterogeneous surfaces
Appl. Phys. Lett. 82, 266 (2003); 10.1063/1.1535271
Orientation and order in microcontact-printed, self-assembled monolayers of alkanethiols on gold investigated
with near edge x-ray absorption fine structure spectroscopy
J. Vac. Sci. Technol. A 15, 2173 (1997); 10.1116/1.580529
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
141.210.2.78 On: Tue, 25 Nov 2014 01:58:33

JOURNAL OF CHEMICAL PHYSICS
VOLUME 120, NUMBER 22
8 JUNE 2004
Dipole-induced structure in aromatic-terminated self-assembled
monolayers: A study by near edge x-ray absorption fine structure
spectroscopy
Yan-Yeung Luka) and Nicholas L. Abbottb)
Department of Chemical and Biological Engineering, University of Wisconsin-Madison,
Madison, Wisconsin 53706
J. N. Crain and F. J. Himpselb),c)
Department of Physics, University of Wisconsin-Madison, Madison, Wisconsin 53706
͑Received 10 November 2003; accepted 11 March 2004͒
The structure of self-assembled monolayers presenting aromatic rings at a surface is studied by near
edge x-ray absorption fine structure spectroscopy ͑NEXAFS͒. Fluorine substitution at asymmetric
positions in the aromatic rings is used to generate a layer of dipoles at the surface of the monolayer.
We find that fluorine substituted aromatic rings are more ordered than unsubstituted aromatic rings
*
by a factor of two based on the polarization dependence of the lowest C 1s to ␲ transition, which
is associated with transitions from phenyl carbons attached to hydrogens. This result is consistent
with the influence of dipole–dipole interactions and quadrupolar interactions between the aromatic
groups due to the substitution of fluorine atoms. The work also serves to illustrate how subtle
variations in the orientation of an end group of a self-assembled monolayer can be determined by
using NEXAFS. © 2004 American Institute of Physics. ͓DOI: 10.1063/1.1737303͔
dimensional hydrogen bond network into the monolayer.9
Abbott and co-workers reported on structural differences be-
tween SAMs formed from semifluorinated alkanethiols as
compared to alkanethiols, and the impact of these structural
difference on the anchoring of liquid crystals.10
I. INTRODUCTION
Self-assembled monolayers ͑SAMs͒ of alkanethiols on
gold substrates form the basis of an important tool for inves-
tigations of surface-driven phenomena in a range of disci-
plines. Examples include control of the orientations of liquid
crystals on SAMs supported on obliquely deposited gold
films,1 control of crystallographic orientations of inorganic
crystals by using various functional groups tethered to
SAMs,2 protein immobilization3 and chemical sensing.4 The
versatility of these experimental systems is largely due to the
well-organized structure of the SAM on a gold substrate.5
Because of the close packing of the alkanethiols on gold
substrates, the functional groups presented by SAMs are
typically oriented.6 In order to tailor the structures and
chemical properties of SAMs, it is important to decipher the
intermolecular interactions that govern the ordering and to
find ways to enhance the order.
Several past examples demonstrate that it is possible to
design the structure and properties of SAMs via specification
of the chemical functionality of the molecules used to form
the SAMs.7 Langer and co-workers introduced bulky groups
that protect acids at the terminus of SAMs to prevent the
close packing of the monolayer via steric hindrance.8 This
surface, upon deprotection, forms a loosely packed mono-
layer that reversibly reorganizes under the influence of an
electric field. Hutchison and co-workers demonstrated the
formation of thermally stable SAMs by introducing a three-
Noncovalent interactions between aromatic groups are
known to impact a wide variety of phenomena in chemistry
and biology.11,12 A classic example is the stabilization of the
DNA duplex due to the stacking of the nucleotide bases. The
stacking of the nucleotide bases is due to interactions of ␲
orbitals between the bases.13 Similar stacking interactions
between aromatics are also believed to contribute substan-
tially to the stabilization of the tertiary structure of
proteins.14 Whereas considerable effort has been dedicated to
the study of aromatic stacking in naturally occurring sys-
tems, there is growing interest in the use of non-natural aro-
matic stacking for the design of materials,15 for use in su-
pramolecular chemistry16 and for biological applications.17
One of the most notable examples of such non-natural aro-
matic stacking arises in systems containing perfluoro-
substituted aromatic rings.18,19 While it is known that ben-
zene in the solid state adopts an edge-to-face structure, it has
been found that
a one-to-one mixture of benzene–
hexafluorobenzene adopts a face-to-face ␲ stacking of alter-
nating molecules of benzene and hexafluorobenezene.18 This
unique arrangement of aromatic molecules in the solid state
reflects molecular interaction arising from the electric quad-
rupole moments of substituted aromatic rings.19
a͒
Current address: Department of Chemistry, Department of Chemical
Engineering and Materials Science, Syracuse University, Syracuse, NY
13244-4100.
In this work, we report on the organization of SAMs, as
influenced by the dipole moments and quadrupole moments
of fluorine-substituted aromatic groups that are tethered to
b͒
Authors to whom correspondence may be addressed.
c͒
Electronic mail: fhimpsel@facstaff.wisc.edu
0021-9606/2004/120(22)/10792/7/$22.00
10792
© 2004 American Institute of Physics
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
141.210.2.78 On: Tue, 25 Nov 2014 01:58:33

J. Chem. Phys., Vol. 120, No. 22, 8 June 2004
Aromatic-terminated self-assembled monolayer
10793
FIG. 1. Self-assembled monolayers formed from alkanethiols 1 and 2. The
arrows represent the dipole moments of the aromatic rings. The orientations
and magnitudes of the dipoles change when three hydrogen atoms are re-
placed by fluorine atoms.
FIG. 2. Synthesis of phenyl-terminated alkanethiols.
the SAMs. The polarization dependence of near edge x-ray
absorption fine structure ͑NEXAFS͒ spectra is used to study
the organization and orientation of the terminal aromatic
groups. The structural order of the monolayers induced by
fluorine-substitution in the aromatic terminal groups are
measured and compared to unsubstituted aromatic terminal
groups. We also present a specific model that describes the
ordering quantitatively. Our work complements previous
NEXAFS by Buck and co-workers, which focused on the
role of the substrate-thiol bond and chain length effects on
the orientation. The nature of the aromatic end group was not
changed by the authors.20
In this paper, we report the use of alkanethiols 1 and 2 to
present 2, 3, 4 fluoro-substituted phenoxy groups or unsub-
stituted phenoxy groups at a surface ͑Fig. 1͒. Substitution of
fluorine atoms in the aromatic rings causes minimal pertur-
bation to steric interactions between the aromatic terminal
groups. However, partial substitution of fluorine atoms at
asymmetric positions of an aromatic ring induces a strong
dipole moment due to the electron withdrawing effects of the
fluorine atoms.11,17 This dipole lies across the fluorophenyl
group and perpendicular to the alkyl chain, whereas the di-
pole moment in alkanethiol 2 lies orthogonal to that of al-
kanethiol 1 ͑Fig. 1͒. Furthermore, the electron withdrawing
effect of the fluoro groups perturbs the electron density of the
p electrons and thus reduces the molecular electric quadru-
pole moments in the aromatic ring of 1 to a great extent. As
a result, when assembled on a gold substrate, SAMs formed
from 1 or 2 present a layer of aromatic rings on the surface
with arrays of dipole moments that are in orthogonal orien-
tations and with substantially different quadrupole moments.
In contrast to the dipole of the unsubstituted SAM, the dipole
of the fluorine-substituted SAM can rotate with angle ␤ and
thus optimize dipole–dipole coupling by forming an antipar-
allel arrangement.
11-Undec-enyloxy-benzene „A… To a solution of 0.257 g
of phenol ͑2.735 mmol, 2 eq.͒ and 66 mg of NaH ͑60%
dispersion͒ in 15 ml DMF stirred for 20 min, 0.3 ml of
11-bromo-1-undecene ͑1.367 mmol, 1 eq.͒ was added. The
solution mixture was stirred for 5 h, and then mixed with 50
ml hexane–ether, 50 ml water. The aqueous phase was then
extracted three times with hexane; the combined organic
phases was dried with MgSO4 , concentrated in vacuo and
purified by flash chromatography ͑1% ethyl acetate–hexane͒
to give 246 mg olefin A ͑0.998 mmol, 73%͒ as a clear oil: 1H
NMR ͑250 MHz, CDCl3) ␦1.26–1.29 ͑br s, 10H͒, 1.53–1.58
͑qui, 2H͒, 1.67–1.75 ͑br s, 2H͒, 2.00–2.04 ͑dd, 2H͒, 3.96–
4.00 ͑t, 2H͒, 4.88–4.96 ͑m, 2H͒, 5.74–5.84 ͑m, 1H͒, 6.81–
6.98 ͑m, 3H͒, 7.2–7.26 ͑m, 2H͒.
1
11-Undecenyloxy-trifluorobenzene „C…: H NMR ͑250
MHz, CDCl3) ␦1.26–1.29 ͑br s, 10H͒, 1.53–1.58 ͑qui, 2H͒,
1.67–1.75 ͑br s, 2H͒, 2.00–2.04 ͑dd, 2H͒, 3.96–4.00 ͑t, 2H͒,
4.88–4.96 ͑m, 2H͒, 5.74–5.84 ͑m, 1H͒, 6.55–6.68 ͑br m,
1H͒, 6.78–6.79 ͑q, 1H͒.
Thioacetic acid S-„11-phenoxy-undecyl… ester „B… A
solution of olefin A ͑0.41 g, 1.664 mmol͒ in dry THF ͑25 ml͒
containing thiolacetic acid ͑0.33 ml, 1.96 mmol͒ and AIBN
͑32.5 mg, 0.193 mmol͒ was irradiated in a photochemical
reactor ͑Rayonet reactor lamp, Southern New England Ultra-
violet Co., model no. RPR-100͒ for 5 h under nitrogen
(ϳ1 atm). Concentration of the reaction mixture in vacuo,
followed by flash chromatography ͑3% ethyl acetate–
hexane͒ gave 483 mg of B as a clear oil ͑1.498 mmol, 90%͒:
1H NMR ͑250 MHz, CDCl3) ␦1.29–1.33 ͑m, 14H͒, 1.55–
1.64 ͑br m, 4H͒, 2.33 ͑s, 3H͒, 2.85–2.90 ͑t, 2H͒, 3.96–4.00
͑t, 2H͒, 6.81–6.98 ͑m, 3H͒, 7.2–7.26 ͑m, 2H͒.
Thioacetic
acid
S-†11-„2,3,4-trifluoro-phenoxy…-
1
undecyl‡ ester „D…: H NMR ͑250 MHz, CDCl3) ␦1.29–
1.33 ͑m, 14H͒, 1.55–1.64 ͑br m, 4H͒, 2.33 ͑s, 3H͒, 2.85–
2.90 ͑t, 2H͒, 3.96–4.00 ͑t, 2H͒, 6.55–6.68 ͑br m, 1H͒, 6.78–
6.79 ͑q, 1H͒.
11-Phenoxy-undecane-1-thiol „1… To a solution of thio-
acetate B ͑53 mg, 0.164 mmol͒ in MeOH ͑15 mL͒ was added
5 drops of HCl ͑12 N͒, and refluxed under nitrogen for 3 h.
The residue product was concentrated in vacuo followed by
purification of the residues by flash chromatography on silica
gel to give 37 mg of the desired thiol 1 ͑0.131 mmol, 80%͒.
1H NMR ͑250 MHz, CDCl3) ␦1.33–1.43 ͑m, 14H͒, 1.50–
1.66 ͑br m, 4H͒, 2.45–2.46 ͑tt, 2H͒, 3.96–4.00 ͑t, 2H͒, 6.81–
6.98 ͑m, 3H͒, 7.2–7.26 ͑m, 2H͒.
II. EXPERIMENT
The compounds 1 and 2 were prepared by a common
3-step route of organic synthesis ͑Fig. 2͒. The syntheses
started with SN2 alkylation of undecenyl bromide with phe-
nol or 2, 3, 4-trifluorophenol to afford the alkylated aromat-
ics. The terminal alkenes were converted to the correspond-
ing thioester by treatment with thiol acetic acid and AIBN
under photolytic conditions.21 Acidic hydrolysis under reflux
afforded the desired aromatic-terminated alkanethiols 1
and 2.21
1
11-„2,3,4-Trifluorophenoxy…-undecane-1-thiol „2…: H
NMR ͑250 MHz, CDCl3) ␦1.33–1.43 ͑m, 14H͒, 1.50–1.66
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
141.210.2.78 On: Tue, 25 Nov 2014 01:58:33

10794
J. Chem. Phys., Vol. 120, No. 22, 8 June 2004
Luk et al.
͑br m, 4H͒, 2.45–2.46 ͑tt, 2H͒, 3.96–4.00 ͑t, 2H͒, 6.55–6.68
͑br m, 1H͒, 6.78–6.79 ͑q, 1H͒.
A. Deposition of gold films and preparation of SAMs
Gold films with thicknesses of ϳ1000 Å were deposited
onto silicon wafers ͑test grade, Silicon Sense Connecting
Technologies, NH͒ mounted on rotating planetaries by using
an electron beam evaporator ͑VES-3000-C manufactured by
Tek-Vac Industries, Brentwood, NY͒. The rotation of the
substrates on the planetaries ensured that the gold was de-
posited without a preferred direction of incidence. A layer of
titanium ͑thickness ϳ100 Å) was used to promote adhesion
between the silicon wafer and the gold film. The rates of
deposition of gold and titanium were ϳ0.2 Å/s. The pres-
sure in the evaporator was less than 5ϫ10Ϫ7 Torr before and
during each deposition. Self-assembled monolayers ͑SAMs͒
of alkanethiols 1 or 2 were formed on gold films by immers-
ing the films in ethanolic solutions containing 2 mM of either
alkanethiol for 6 h.
FIG. 3. Definition of the angles ␣, ␤, and ␹ that characterize the orientation
of the aromatic rings on SAMs. ␣ is the tilt angle of the phenyl ether bond
with respect to the surface normal. ␤ is the angle of rotation of the aromatic
ring around the phenyl ether bond. ␪ is the angle of the electric field vector,
with respect to the normal ͑which is equal to 90° minus the angle of inci-
dence from normal͒. ␹ is the rotation of the phenyl endgroup around the
O-alkane bond. Because the O-alkane bond is nearly perpendicular to the
normal, ␹ has little effect on the orientation of the plane of the phenyl ring
with respect to the surface normal.
B. NEXAFS spectroscopy
Near edge x-ray absorption fine structure ͑NEXAFS͒
spectroscopy allows the direct characterization of the chemi-
cal properties and orientations of molecules bound to
surfaces.22 Electrons in occupied core level states are pro-
moted to empty valence orbitals by absorption of polarized
soft x-rays. Each absorption event produces Auger and sec-
ondary electrons. The number of detected electrons is ap-
proximately proportional to the number of core holes, as
long as the escape depth is small compared to the absorption
length.22 The effective absorption depth changes somewhat
with the polar angle of incidence.23 However, this effect is
identical for the fluorinated and nonfluorinated molecules.
The electrons yield provides a convenient measure of the
absorption coefficient. By utilizing tunable photons from a
synchrotron, the photon energy is scanned across the C 1s
absorption edge.
NEXAFS spectroscopy was performed at the Synchro-
tron Radiation Center ͑SRC͒ on the HERMON beam line
using secondary electron detection. A load lock was used to
introduce the samples into the measurement chamber, which
was maintained at a base pressure of less than 10Ϫ10 Torr.
The photon energy interval from 275 to 325 eV was chosen
to completely span the C 1s absorption edge and all of the
bon atoms sampled in the experiment. The scale given for
the absorption in Fig. 4 refers to the pre-edge background.
III. RESULTS AND DISCUSSION
Figure 3 defines the parameters used to describe the
structure and order of the terminal aromatic rings of the
SAMs. We define ␣ to be the polar angle of the phenyl ether
tether with respect to the normal, and ␤ to describe the ori-
entation of the aromatic ring around the ether bond. The tilt
angle of the alkanethiols is largely determined by the pack-
ing of SAMs on the ͑1,1,1͒ gold.24 Because the gold film was
deposited on rotating planetaries, we do not expect there to
be a macroscopic azimuthal preference of the tilt direction of
the alkanethiols.25
The bond angle of the phenyl ether linkage ͑␣͒ is deter-
mined by the electronic structure of the molecule
(ϳ130°).26 Thus, we expect little variation in the angle of ␣
in the SAMs formed from both 1 and 2, where ␣ approxi-
mately equals to 70°. Because of the trans-conformation of
the methylene units along the alkyl chain, each alkanethiol
occupies an area of about 24 Å2 ͑5.6 Å in diameter͒.27 Since
the longest distance across the facet of a benzene ring is
about 5.0 Å, we assume that the aromatic rings experience
little steric hindrance from neighboring alkanethiol mol-
ecules in the SAM, and hence possess a high degree of ro-
tational freedom ͑rotation about ␤͒.
*
*
and ␴ transitions. The x-rays were Ͼ90% linearly po-
␲
larized. For each sample, scans were taken at angles of the
polarization vector with respect to the sample normal be-
tween 85° ͑near normal incidence͒ and 20° ͑near grazing
incidence͒, as indicated in Fig. 3. The position of the light
spot on the sample was moved every few scans to prevent
extended exposure of the SAMs to photons and radiation
damage. Spectra were normalized to a gold film evaporated
in situ on silicon to remove the transmission function of the
optics. To account for angle-dependent changes of the ab-
sorption path and the collection efficiency, we normalized
the spectra such that they coincide for photon energies far
below and far above the C 1s absorption edge. This proce-
dure effectively normalizes the signal to the number of car-
The positions and intensities of peaks measured in NEX-
AFS provide chemical fingerprints of specific moieties in the
SAM molecules. Their assignment is based on a combination
of the core level shift induced by electronegative neighbors
such as oxygen and fluorine and the energetic position of
*
unoccupied molecular orbitals, such as the ␲ orbitals of the
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
141.210.2.78 On: Tue, 25 Nov 2014 01:58:33

J. Chem. Phys., Vol. 120, No. 22, 8 June 2004
Aromatic-terminated self-assembled monolayer
10795
core-to-valence electronic excitations are sensitive to the
bonding environment, the NEXAFS resonances are governed
by the effects of chemical substitutents, and thus the location
of atoms within the molecule.22,29 Each SAM is character-
ized by six core-to-valence transitions of the C 1s into the
corresponding anti-bonding orbitals. Peak ‘‘a’’ corresponds
to the transition of C 1s of the aromatic carbons tethered
*
with hydrogen to the ␲
anti-bonding orbitals in SAMs
1
prepared from both alkanethiols 1 or 2. Peak ‘‘b’’ corre-
sponds to the transition of C 1s of the aromatic carbons
tethered with fluorine/oxygen ͑SAM prepared from al-
kanethiols 1͒ or with just oxygen ͑SAM prepared from al-
*
kanethiols 2͒ to the ␲ anti-bonding orbitals in both SAMs
1
prepared from alkanethiols 1 or 2.29 Peak ‘‘c’’ corresponds to
*
the transition of C 1s of the aliphatic carbons to the C–H
anti-bonding orbitals ͑Rydberg resonance͒ in both SAMs
prepared from alkanethiols 1 or 2.30 Peaks ‘‘d’’ and ‘‘e’’ cor-
respond to the same transitions as peaks ‘‘a’’ and ‘‘b,’’ re-
spectively, except the final anti-bonding orbital is the higher-
*
lying ␲ . Peak ‘‘f’’ corresponds to the transition of C 1s of
2
*
the aliphatic carbons to the C–C ␴ anti-bonding orbitals in
both SAMs prepared from alkanethiols 1 or 2.31 Substitution
of electronegative atoms of fluorine and oxygen on aromatic
rings induces a partial positive charge on the carbon. which
increases the binding energy of the C 1s level.29 Thus, the
transition of the C 1s of a carbon bonded to a fluorine atom
and the C 1s of a carbon bonded to a oxygen atom to both
FIG. 4. NEXAFS spectra at the C 1s edge as a function of different polar-
ization angles (85° to 25° from the sample normal͒ for SAMs formed from
͑A͒ alkanethiols 1, and ͑B͒ alkanethiols 2. A strong polarization dependence
*
*
␲
and ␲
anti-bonding orbitals are not resolved for
1
2
SAMs prepared from alkanethiols 1. Furthermore, the shift
*
in the transition energy of the C 1s to ␲ anti-bonding or-
*
*
of the C 1s-to-␲ ,␴ transitions is observed ͓peaks ͑a͒–͑f͔͒. The angle is
between the electric field vector E and the sample normal ͑␪ in Fig. 3͒.
bitals ͑peak ‘‘b’’͒ due to fluorine and oxygen substitution
*
also causes an overlap of peak ‘‘b’’ with the C 1s to C–H
transition ͑peak ‘‘c’’͒.
aromatic ring. Figure 4 shows the polarization dependence of
the C 1s NEXAFS absorption spectra of SAMs formed from
alkanethiols 1 and 2. The spectra are normalized below and
above the absorption edge (h␯ϭ280 and 320 eV͒ and the
assignments of the resonances for the SAMs formed from
Inspection of Fig. 4 indicates two characteristic differ-
ences between the transition intensities of SAMs prepared
from alkanethiols 1 and 2 that reflect the substitution of fluo-
rine on the aromatic rings on SAMs. First, because the sub-
stitution of the fluorine atoms eliminates three C–H bonds,
the intensity of the transition from C 1s ͑aromatic carbon
28
alkanethiols 1 and 2 are tabulated in Table I. Because the
TABLE I. Assignment of resonances for the NEXAFS spectra of SAMs formed from alkanethiols 1 and 2.
SAMs formed from alkanethiol 1 ͑fluoro-substituted phenyl͒ SAMs formed from alkanethiol 2 ͑phenyl͒
Peak
a
Assignment
*
Photon
energy ͑eV͒
284.9
Peak
a
Assignment
*
Photon
energy ͑eV͒
284.9
CvC ␲
CvC ␲
1
1
͑neighboring C–H͒
͑neighboring C–H͒
b1
b2
c
CvC ␲
287.3a
287.3
287.0
288.7
ϳ290a
ϳ290a
292.9
¯
b2
c
¯
¯
287.5
*
1
͑neighboring C–F͒
*
*
CvC ␲
CvC ␲
1
1
͑neighboring C–O͒
͑neighboring C–O͒
*
C–H
*
C–H
͑alkyl, phenyl͒
287.0
͑alkyl, phenyl͒
*
*
d
CvC ␲
d
CvC ␲
288.3
2
2
͑neighboring C–H͒
͑neighboring C–H͒
*
e1
e2
f
CvC ␲
¯
e2
f
¯
¯
2
͑neighboring C–F͒
*
*
CvC ␲
CvC ␲
Unresolved
2
2
͑neighboring C–O͒
C–C ␴ ͑alkyl͒
͑neighboring C–O͒
*
*
C–C ␴ ͑alkyl͒
292.9
aThe ␲ C–F/O resonances are masked by the C 1s absorption edge of c, d, or f making an accurate determi-
*
2
nation of the peak position difficult.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
141.210.2.78 On: Tue, 25 Nov 2014 01:58:33

10796
J. Chem. Phys., Vol. 120, No. 22, 8 June 2004
Luk et al.
*
tethering hydrogen͒ to ␲ anti-bonding orbitals ͑peak ‘‘a’’͒
1
on the fluorophenyl rings ͑SAM prepared from 1͒ is signifi-
cantly smaller than that of the phenyl rings ͑SAM prepared
from 2͒. Second, the intensity missing from peak ‘‘a’’ is
transferred to peak ‘‘b’’ when the ring is substituted with
*
fluorine, because the transition from C 1s to ␲ is shifted
1
up in energy by the fluorine-induced C 1s core level shift.
The probability of core-level to anti-bonding electronic
transitions depends on the angle between the polarization
vector of the light and the anti-bonding p-orbitals in the mol-
ecules. For C 1s excitations, the transition probability is a
maximum when the polarization vector is parallel to the or-
bital, and the transition intensity is dipole-forbidden when
the polarization vector is perpendicular to the orbital ͑fol-
lows a cos2 ␪ law͒. Thus, by using the linearly polarized
*
synchrotron x-rays, the orientation of the ␲ orbitals of the
aromatic end groups can be extracted. The excitation of the
core electrons to the anti-bonding orbitals in NEXAFS spec-
troscopy depends strongly on the orientation of the bonding
geometry relative to the polarization of the incident x-ray.
Figure 4 reveals a similar angular dependence of the inten-
sity of absorptions within each SAM prepared from either 1
or 2. For absorptions related to the aliphatic chains of both
SAMs, the intensity of transitions associated with the ␴-bond
*
FIG. 5. Polarization dependence of absorption for the lowest ␲ transition
of the H-bonded C atoms in the phenyl rings ͑peak a in Fig. 4͒. Fluorophe-
nyl ͑full circles͒ exhibits stronger polarization dependence than phenyl
͑open circles͒ indicating better ordering. Solid ͑fluorophenyl͒ and dashed
͑phenyl͒ lines show fits to the data of the form AϩB cos2 ␪.
spectra at each angle, SAM 2 minus SAM 1, as shown for
␪ϭ55 degrees in Fig. 6. Thereby, the contribution from
peaks c is removed to first order. Gaussian fits to the main
*
skeleton ͑C 1s to C–C ␴ , peak ‘‘f’’͒ increases with in-
crease in the grazing angle, ␪, of the incident polarized
*
peaks confirm the ␲ resonances at 284.9 and 287.3 eV. An
1
x-rays. However, the intensity of transitions associated with
additional weak feature at 287.0 eV indicates a small residual
*
the methylene groups ͑C 1s to C–H , peak ‘‘c’’͒ decrease
of the C–H resonance, possibly due to the extra C–H bonds
in the phenyl of molecule 2. The residual C–H peak does not
have any angular dependence, which shows that there is no
significant difference in the orientation of the alkane back-
bone between the two SAMs.. A further complication arises
from the presence of two components of peak b ͑b1 and b2͒
associated with transitions from carbon atoms neighboring
fluorine ͑b1͒ and oxygen ͑b2͒ in the phenyl rings. However,
with increase in the grazing angle ͑Fig. 4͒. Because the
maximum excitation of NEXAFS occurs when the electric
field vector E is parallel to the molecular orbital of
excitation,22 this result is consistent with the ␴ bond ͑C–C͒
being tilted closer to the normal of the surface whereas the
C–H bonds in the methylene groups are oriented towards the
parallel to the surface. This type of angular dependence is a
general characteristic of SAMs formed from alkanethiols on
gold.28
*
both transitions have the same ␲ final state, have initial
1
state chemical shifts that are nearly identical, due to the simi-
While the structures of the two SAMs seem qualitatively
similar, a quantitative comparison of the angular dependence
lar electronegativity of F and O, and should exhibit the ori-
*
of the intensity of the ␲ resonance of the aromatic rings
1
does reveal significant differences between the structures of
the two SAMs. Figure 5 shows the angular dependence of
*
the intensity of the ␲
resonance ͑at 285 eV͒ of SAMs
1
prepared from 1 and 2 at different angles of incidence of the
*
polarized x-rays. We focused on the ␲
resonance for a
1
detailed quantitative analysis because it is the best resolved
resonance in the C 1s absorption edges. After normalizing
*
the intensity of the ␲ resonance at ␪ϭ90 degrees for both
1
SAMs, the SAMs with the fluorine-substituted aromatic
rings clearly show a stronger angular dependence than the
SAMs with unsubstituted rings ͑Fig. 5͒. The modulation of
the intensity of the fluorine-substituted SAMs is almost twice
as large as the unsubstituted SAMs ͑0.51 versus 0.27, see the
full and dashed fit curves͒.
FIG. 6. Analysis of higher transitions using difference spectra. ͑a͒ The
NEXAFS spectrum for fluorinated phenyl rings subtracted from those with-
out fluorine substitution at an angle of ␪ϭ55°. Similar difference spectra at
*
*
other polar angles yield ͑b͒ the polarization dependence for the ␲ transi-
Next, we examine the ␲ resonance at 287.3 eV ͑peak
1
tion of the F/O-terminated C atoms ͑open circles, peak b in Fig. 4͒ in the
phenyl rings and for the C–H bond ͑filled circles, peak c in Fig. 4͒. The
calculated curve for the higher ␲ transition is derived from the fits in
b1͒, which is expected to exhibit the same polarization de-
pendence as peak a. The analysis of peak b1 is complicated
by an overlap with the C–H resonance ͑peak c͒ For a more
quantitative analysis we take the difference of the NEXAFS
*
*
Fig. 5, indicating the same polarization dependence as for the lower ␲
transition.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
141.210.2.78 On: Tue, 25 Nov 2014 01:58:33

J. Chem. Phys., Vol. 120, No. 22, 8 June 2004
Aromatic-terminated self-assembled monolayer
10797
entation dependence of the planar phenyl rings. As a result
peak b can be analyzed without treating the C–O and C–F
components independently.
orientational order of organic surfaces have also been dem-
onstrated to impact the alignment of liquid crystals at the
macroscopic level.1,35 Because an anti-parallel alignment is
energetically preferred for the dipole–dipole interactions, the
small spread in the ␤ angle of the fluorine-substituted aro-
matic ring likely underlies the high order in the measured
The areas of the peaks in the difference spectra enable a
*
quantitative analysis of the angular dependence of the ␲
resonance at 287.3 eV, as shown in Fig. 6͑b͒. The area as a
*
function of angle is described by the relationship
NEXAFS signal of the C 1s to ␲ transition for the fluorine-
substituted monolayers. Furthermore, because the intensity
of the ␲ transition was measured to increase with increase in
the angle of incidence of the x-rays, the planes of the aro-
matic rings containing the ␲ orbitals are tilted perpendicular
rather than parallel to the surface ͑see above͒. These orien-
tational organizations are consistent with a head-to-tail ar-
rangement of dipoles as well as the interactions of the mo-
lecular quadrupolar moments between the fluoro-substituted
aromatic rings.19 For SAMs with aromatic groups that have
no fluorine substitution, the interactions of the dipole mo-
ment of the aromatic groups are not facilitated by a preferred
value of ␤, thus likely resulting in the less ordered assembly
of phenyl rings on the surface. These results, when com-
bined, demonstrate that dipole moment interactions and the
molecular quadrupolar effects of aromatic rings can be used
to guide subtle variations in the structure of SAMs. We dem-
onstrate in this work that subtle changes in the organization
of SAMs due to fluorine-substitution of aromatic groups can
be probed by using NEXAFS.
Areaϭ AϩB cos2 ␪͒,Ϫ CϩD cos2 ␪͒,
͑
͑
where the coefficients A,B are for endgroups of SAM 2, with
only oxygen and C,D of SAM 1, with both oxygen and fluo-
rine. The coefficients are explicitly calculated using ͑1͒ the
*
ratios of B/A and D/C as determined from fits to the ␲
peaks in Fig. 5, ͑2͒ the relative concentrations of carbon
atoms neighboring fluorine ͑b1͒ and oxygen ͑b2͒ in the end-
groups of SAMS 1 and 2 and ͑3͒ a normalization to the
difference intensity at the magic angle ␪ϭ54.7° where the
intensity is independent of the molecular orientation.22 The
calculated curve ͓solid line Fig. 6͑b͔͒ is in excellent agree-
ment with the experimental data, thus corroborating our find-
ing for the 285 eV resonance that the modulation of the
intensity of the fluorine-substituted SAMs is almost twice as
large as the unsubstituted SAMs.32
The above described result leads us to the general con-
clusion that the distribution in the orientation of the aromatic
rings is substantially narrower for the fluorine-substituted
aromatic rings than for unsubstituted rings. If perfect order
existed ͑i.e., all phenyl rings stacked parallel to each other
and oriented perpendicular to the plane of the surface͒, the
IV. CONCLUSION
To conclude, we used NEXAFS to study the orientations
of aromatics groups presented at the outer surfaces of SAMs
supported on films of gold. Asymmetric substitution of aro-
matic end groups with fluorine is used to influence the order-
ing of the SAMs via electrostatic dipole interactions. The
polarization dependence of the NEXAFS signal correspond-
*
␲
intensity in Fig. 5 would change between 0 and 1 with
change in polarization of the incident x-rays. The magnitude
1
*
of the modulation in the ␲
intensity decreases with in-
1
crease in the spread of orientations of the phenyl rings. Be-
cause the electric field vector cannot be perpendicular to all
the phenyls at any angle of incidence, the baseline in Fig. 5
will increase from zero to a finite value when a spread in
orientations of the phenyl groups is present. There are sev-
eral specific models that can be used to describe the orienta-
tional disorder of the phenyl groups, such as disorder in the
twist angle ␤ ͑e.g., oscillations around the symmetric posi-
tions ␤ϭ0° and 90°), disorder in the oxygen–phenyl bond
direction, both in-plane and out-of-plane, symmetry breaking
by a fixed angle ␤ between 0° and 90°, and the coexistence
of several domains with different azimuthal orientations of
the tilted alkyl chains. In this paper, we do not attempt to
employ specific models to describe the orientations of the
phenyl groups. Instead, we emphasize the change in the
overall orientational order of the phenyl rings, defined as the
spread in the orientation of the phenyl rings, averaged over
all domains as dictated by the substitute of fluorine atoms
into the aromatic groups of the SAMs. Explicit calculations
for the symmetric geometries ␤ϭ0° and ␤ϭ90° show that
the ␤ϭ0° is much closer to the experimental situation, i.e.,
the planes of the phenyl rings are almost perpendicular in-
stead of parallel to the surface of the gold film.
*
ing to the C 1s to ␲ transition in the phenyl groups is found
to increase by a factor of 2 with fluorine substitution of the
phenyl groups, thus indicating a higher degree of orienta-
tional order in the presence of the fluorine-substituted aro-
matic groups. This observation is consistent with stabiliza-
tion of the orientational order via the dipole–dipole and the
molecular quadrupolar interactions between the aromatic
rings in the monolayer. Because the substitution of fluorine
atoms does not usually cause additional steric hindrance, we
conclude that weak intermolecular interactions due to fluo-
rine substitution of aromatic groups can be used to control
the details of the structure of the SAMs.
ACKNOWLEDGMENTS
This research was supported by the NSF under Award
No. DMR-0240937 and Materials Research Science and En-
gineering Center on Nanostructured Materials and Interfaces
͑NSF-DMR-0079983͒. The NEXAFS experiments were con-
ducted at the SRC, which is supported by the NSF under
Award No. DMR-0084402.
Strong molecular interactions within SAMs, such as hy-
drogen bonding, have been shown in past studies to confer
additional stability on SAMs,33 as well as to increase the rate
of electron transfer across SAMs.34 Subtle changes in the
1 V. K. Gupta and N. L. Abbott, Science 276, 1533 ͑1997͒.
2 J. Aizenberg, A. J. Black, and G. M. Whitesides, Nature ͑London͒ 398,
495 ͑1999͒.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
141.210.2.78 On: Tue, 25 Nov 2014 01:58:33

10798
J. Chem. Phys., Vol. 120, No. 22, 8 June 2004
Luk et al.
3 Y. Y. Luk, M. L. Tingey, D. J. Hall, B. A. Israel, C. J. Murphy, P. J.
Bertics, and N. L. Abbott, Langmuir 19, 1671 ͑2003͒.
21 C. Palegrosdemange, E. S. Simon, K. L. Prime, and G. M. Whitesides, J.
Am. Chem. Soc. 113, 12 ͑1991͒.
22 J. Stohr, NEXAFS Spectroscopy ͑Springer Verlag, NY, 1996͒.
23 W. Gudat and C. Kunz, Phys. Rev. Lett. 29, 169 ͑1972͒.
24 M. D. Porter, T. B. Bright, D. L. Allara, and C. E. D. Chidsey, J. Am.
Chem. Soc. 109, 3559 ͑1987͒.
25 S. Follonier, W. J. W. Miller, N. L. Abbott, and A. Knoesen, Langmuir 19,
10501 ͑2003͒.
26 J. Barrett and E. Abel, Structure and Bonding ͑Wiley, NY, 2002͒.
27 G. E. Poirier, E. D. Pylant, and J. M. White, J. Chem. Phys. 105, 2089
͑1996͒.
4 R. R. Shah and N. L. Abbott, Science 293, 1296 ͑2001͒.
5 C. D. Bain and G. M. Whitesides, Science 240, 62 ͑1988͒.
6 G. E. Poirier, Chem. Rev. 97, 1117 ͑1997͒.
7 Y. Y. Luk, M. Kato, and M. Mrksich, Langmuir 16, 9604 ͑2000͒.
8 J. Lahann, S. Mitragotri, T. N. Tran, H. Kaido, J. Sundaram, I. S. Choi, S.
Hoffer, G. A. Somorjai, and R. Langer, Science 299, 371 ͑2003͒.
9 R. S. Clegg and J. E. Hutchison, J. Am. Chem. Soc. 121, 5319 ͑1999͒; R.
S. Clegg, S. M. Reed, and J. E. Hutchison, ibid. 120, 2486 ͑1998͒; R. S.
Clegg, S. M. Reed, R. K. Smith, B. L. Barron, J. A. Rear, and J. E.
Hutchison, Langmuir 15, 8876 ͑1999͒; R. Valiokas, M. Ostblom, S. Sved-
hem, S. C. T. Svensson, and B. Liedberg, J. Phys. Chem. B 106, 10401
͑2002͒.
28 R. D. Peters, P. F. Nealey, J. N. Crain, and F. J. Himpsel, Langmuir 18,
1250 ͑2002͒; J. N. Crain, A. Kirakosian, J. L. Lin, Y. D. Gu, R. R. Shah,
N. L. Abbott, and F. J. Himpsel, J. Appl. Phys. 90, 3291 ͑2001͒.
29 A. P. Hitchcock, P. Fischer, A. Gedanken, and M. B. Robin, J. Phys.
Chem. 91, 531 ͑1987͒; J. L. Solomon, R. J. Madix, and J. Stohr, Surf. Sci.
255, 12 ͑1991͒.
10 W. J. Miller, V. K. Gupta, N. L. Abbott, M. W. Tsao, and J. F. Rabolt, Liq.
Cryst. 23, 175 ͑1997͒.
11 M. L. Waters, Curr. Opin. Chem. Biol. 6, 736 ͑2002͒.
12 S. M. Butterfield, P. R. Patel, and M. L. Waters, J. Am. Chem. Soc. 124,
9751 ͑2002͒; C. D. Tatko and M. L. Waters, ibid. 124, 9372 ͑2002͒; M. J.
Rashkin and M. L. Waters, ibid. 124, 1860 ͑2002͒.
30 P. S. Bagus, K. Weiss, A. Schertel, C. Woll, W. Braun, C. Hellwig, and C.
Jung, Chem. Phys. Lett. 248, 129 ͑1996͒.
13 E. T. Kool, Annu. Rev. Biophys. Biomol. Struct. 30, 1 ͑2001͒.
14 N. Kannan and S. Vishveshwara, Protein Eng. 13, 753 ͑2000͒.
15 P. Kirsch and M. Bremer, Angew. Chem., Int. Ed. Engl. 39, 4217 ͑2000͒.
16 G. W. Coates, A. R. Dunn, L. M. Henling, D. A. Dougherty, and R. H.
Grubbs, Angew. Chem., Int. Ed. Engl. 36, 248 ͑1997͒.
31 G. Hahner, M. Kinzler, C. Woll, M. Grunze, M. K. Scheller, and L. S.
Cederbaum, Phys. Rev. Lett. 67, 851 ͑1991͒; D. A. Outka, J. Stohr, J. P.
Rabe, J. D. Swalen, and H. H. Rotermund, ibid. 59, 1321 ͑1987͒.
32
*
Note that the modulation of the ␲ intensity in the difference curves of
17 B. M. O’Neill, J. E. Ratto, K. L. Good, D. C. Tahmassebi, S. A. Helquist,
J. C. Morales, and E. T. Kool, J. Org. Chem. 67, 5869 ͑2002͒.
18 J. H. Williams, Acc. Chem. Res. 26, 593 ͑1993͒.
Fig. 6 is not directly comparable to the raw data of Fig. 5 because it is
calculated from the difference spectra. Separate orientations for the end-
groups of the two SAMs cannot be extracted from the difference spectra.
33 A. K. Boal and V. M. Rotello, Langmuir 16, 9527 ͑2000͒.
19 J. C. Ma and D. A. Dougherty, Chem. Rev. 97, 1303 ͑1997͒.
20 H. T. Rong, S. Frey, Y. J. Yang, M. Zharnikov, M. Buck, M. Wuhn, C.
Woll, and G. Helmchen, Langmuir 17, 1582 ͑2001͒.
34 S. Sek, A. Misicka, and R. Bilewicz, J. Phys. Chem. B 104, 5399 ͑2000͒.
35 J. Stohr, M. G. Samant, J. Luning et al., Science 292, 2299 ͑2001͒.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
141.210.2.78 On: Tue, 25 Nov 2014 01:58:33