Electroactive self-assembled monolayers of unique geometric structures by using rigid norbornylogous bridges

Article abstract of DOI:10.1002/chem.201101588

Herein, we describe the synthesis of straight (S) and L-shaped (L) norbornylogous bridges (NBs) with an anthraquinone moiety at the distal end as the redox-active head group and two thiol feet at the proximal end, by which the molecules assemble on gold s

Full text of DOI:10.1002/chem.201101588

FULL PAPER
DOI: 10.1002/chem.201101588
Electroactive Self-Assembled Monolayers of Unique Geometric Structures
by Using Rigid Norbornylogous Bridges
Nadim Darwish,[a] Paul K. Eggers,[a] Paulo Da Silva,[a] Yi Zhang,[b] Yujin Tong,[b]
Shen Ye,*[b] J. Justin Gooding,*[a] and Michael N. Paddon-Row*[a]
Abstract: Herein, we describe the syn-
thesis of straight (S) and L-shaped (L)
norbornylogous bridges (NBs) with an
anthraquinone moiety at the distal end
as the redox-active head group and two
thiol feet at the proximal end, by which
the molecules assemble on gold surfa-
ces. The NB molecules were shown to
impedance spectroscopy. The surface
selection rules associated with the IR
band intensities allowed the estimation
of the position of the anthraquinone
moiety with respect to the surface and
the tilt of the bridge with respect to the
surface normal, both in pure and dilut-
ed monolayers. It is shown that the S-
and L-NBs hold the plane of the an-
thraquinone moiety close to the surface
normal or the surface tangent, respec-
tively. Neither NB molecule changes its
orientation if spaced by diluents on the
surface. The difference in the structure
of the S- and L-NB SAMs provides a
suitable framework for the investiga-
tion of factors that govern electron
transfer of anthraquinone moieties
across self-assembled monolayers with
limited structural ambiguity, as com-
pared with the commonly used struc-
turally flexible alkanethiol monolayers.
form
self-assembled
monolayers
(SAMs) with a well-behaved surface
redox process. The SAMs were charac-
terized by using in situ IR spectrosco-
py, cyclic voltammetry, scanning tun-
nelling microscopy and electrochemical
Keywords: anthraquinones · bridg-
ing ligands · electrochemistry · IR
spectroscopy · monolayers
Introduction
the electrolyte affects their ET is critical to the understand-
ing of ET in biological systems. Quinone-terminated self-as-
sembled monolayers (SAMs) have also been highlighted for
such applications as electrochemically gated transistors.[8,9]
Because the redox potential of quinones can be altered by
using the pH of the electrolyte, they can be exploited to
obtain a pH-gated transistor.[8]
Despite numerous studies that investigate redox-active
SAMs with a wide range of redox-active groups, only a few
studies have investigated quinone derivatives as the redox-
active terminal group.[2,10–19] Conformationally flexible al-
Quinone derivatives play an important role in biological sys-
tems and are attractive for the preparation of electro-active
devices.[1,2] In particular, quinone derivatives are important
to understand the proton-coupled electron-transfer (ET) re-
actions that are critical steps in photosynthesis, cellular res-
piration and mitochondrial ATP synthesis.[3] The most popu-
lar quinoid compounds are anthraquinone derivatives, which
have a wide variety of uses in such areas as hydrogen perox-
ide production, oxygen reduction and as a molecular carri-
er.[4–6] In biological species, quinone derivatives are typically
membrane-bound and redox reactions occur at the aque-
ous–biomembrane interface.[7] Therefore, a detailed knowl-
edge of their surface chemistry and how their interface with
AHCTUNGTREGkNNNU anethiols that mediate between the quinone moieties and
the gold surface are often used. The flexible nature of the
backbone of an alkanethiol results in ambiguity as to the
actual position of the redox reaction centre. This problem is
important because the position and the environment around
the redox-active group at interfaces can significantly influ-
ence the ET reaction. This problem can be overcome by
using rigid norbornylogous bridges (NBs) with well-defined
organization of the anchoring points to the gold surface and
the positioning of the redox centre relative to the surface.
The NB molecules can hold the quinone moiety at a well-
understood distance and orientation above the surface, de-
fined as the distal end of the diluents in mixed monolayers.
[a] Dr. N. Darwish, Dr. P. K. Eggers, Dr. P. Da Silva,
Prof. Dr. J. J. Gooding, Prof. Dr. M. N. Paddon-Row
School of Chemistry
The University of New South Wales
Sydney, NSW, 2052 (Australia)
Fax : (+612)93855384
E-mail: justin.gooding@unsw.edu.au
m.paddonrow@unsw.edu.au
[b] Dr. Y. Zhang, Dr. Y. Tong, Prof. S. Ye
Catalysis Research Centre
Hokkaido University
Polynorborane bicycloACHTUNRGTNE[NUG 2.2.0]hexane bridges, referred to as
norbornylogous bridges, have the properties required to in-
vestigate ET processes across SAMs with limited structural
ambiguity compared with those previously achieved with
flexible alkanethiols.[20,21] NB is a versatile spacer, which is
particularly useful for several reasons. Firstly, the bridge is
Sapporo 001-0021 (Japan)
Fax : (+81)11-706-9126
E-mail: ye@cat.hokudai.ac.jp
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201101588.
Chem. Eur. J. 2012, 18, 283 – 292
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
283

structurally rigid, which means that the terminal redox
moiety is held in a fixed, well-defined orientation. Secondly,
the nature of the bridge allows systematic variation of the
length and configuration, which allows a study of the de-
pendence on distance and the effect of bridge configuration
on the dynamics of the ET. Finally, the nature of the NBs
allows a wide range of redox-active moieties to be attached
to the termini.[22–28]
In previous studies using an ellipsometry technique,
SAMs formed from tetrathiolated NBs (Scheme 1a) were
shown to tilt by an angle of about 30 to 348 to the surface
1) The molecules are totally rigid, which allows precise
knowledge of the position of the redox-active moiety.
2) The difference in the structures of the S- and L-NBs
allows two different orientations of the anthraquinone
moiety with respect to the gold surface or the surface of
diluents in mixed monolayers. This difference allows in-
vestigation of both the effect of the orientation of the
redox active group on the electron transfer process and
also how the diluent in a mixed SAM affects electron
transfer.
3) The molecules are long enough to be able to position the
anthraquinone redox-active group above the surface of
the diluents in a mixed monolayer. A short bridge neces-
sitates the use of shorter diluents, which is known to
form less-ordered SAMs.[31] To address this issue, the S-
and L-NB molecules were synthesised to be longer than
14 ꢁ.
Scheme 1. a) Tetrathiolated and b) ferrocene-terminated NBs.
normal.[29] It was suggested that both thiol groups bond to
the gold surface and that the NB molecules are more stable
on the gold surface compared with other commonly used al-
kanethiols.[29] Eggers et al. studied a series of NB molecules
with a ferrocene (Fc) moiety fused to one end of the NB
bridge (Scheme 1b).[30] In situ IR spectroscopy was used to
show that the NB molecules change their tilt angle during
the redox reaction (Fc to Fc+) in SAMs formed from ferro-
cene-terminated NBs only.[30] The change in tilt angle was
ascribed to the space requirement needed for the Fc+ to
À
bind to the ClO4 counterions. However, in mixed SAMs
composed of NB molecules diluted with shorter alkane-
thiols, the NB molecules did not change their orientation
during the redox reaction. The fixed orientation was ex-
plained by the ability of the NB molecules to lift the ferro-
cene moiety above the surface defined by the distal end of
the diluents. In such a position, the ferricinium cations are
not sterically congested and, therefore, ion pairing with
4) The molecules possess a trans CH2SH stereochemistry at
the proximal end, at which the molecules bond to the
gold surface. These “feet” lend more rigid anchoring
points to the gold surface.
Figure 1 presents the geometry-optimized structures of
the S- and L-NB molecules performed by using density
functional theory at the B3LYP/6-31G(d) level of theory.[32]
The distance between the thiol groups and the carbonyl
groups for L- and S-NB is 14.2 and 19.9 ꢁ, respectively. The
synthetic procedure leads to the formation of a racemic
modification that is not resolved and is directly used in the
preparation of the SAMs because both enantiomers should
bond in an identical fashion to the gold surface.[29]
À
ClO4 does not necessitate a change in the tilt angle of the
bridge.[30]
This paper has two aims. The first is to describe the syn-
thesis of straight and L-shaped NB compounds (S- and L-
NB, respectively) that are terminated at one end with an-
thraquinone moieties and with two vicinal (CH2SH) groups
in a trans relationship at the proximal end. The second is to
investigate the surface and electrochemical properties of the
SAMs formed from these molecules on gold surfaces by
using in situ IR spectroscopy, scanning tunnelling microsco-
py and electrochemical impedance spectroscopy.
Results and Discussion
Four criteria were taken into consideration when design-
ing the NB compounds described in this paper:
The synthesis of S- and L-NBs followed standard proce-
dures.[33,34] Key steps are depicted in Scheme 2 and detailed
284
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 283 – 292

Electroactive Self-Assembled Monolayers
FULL PAPER
Figure 1. B3LYP/6-31G(d)-optimized structure of a) S-NB and b) L-NB;
distances shown are between the thiol groups and the carbonyl groups.
The OCH3 groups in the backbone of S-NB and the OH groups in the
backbone of the L-NB were omitted for clarity.
Figure 2. ORTEP diagram of the X-ray crystal structure of the L-NB pre-
cursor (9), showing the exo-addition of cyclopentadiene and the trans re-
lationship of the two CH2OTs groups. Ellipsoids were drawn at the 50%
probability level.
were then achieved by reacting precursors 10 and 16 with 6-
7-dimethyl-1,3-diphenyl-naphthoACHTUNTRGNEUNG[2,3-c]furan-5,8-dione
(INF)[38] by using a Diels–Alder reaction followed by dehy-
dration[36] to produce the anthraquinone moiety (Scheme 3).
The reaction of INF with the NB was characterized by X-
Scheme 2. Synthesis of the L- and S-NB precursors. Full schemes and ex-
perimental procedures are presented in the Supporting Information.
schemes and experimental procedures are presented in the
Supporting Information. The norbornadiene dimer under-
went a ruthenium-catalyzed thermal [2+2] cycloaddition of
dimethyl acetylenedicarboxylate (DMAD),[35] followed by
reduction of the esters to give hydroxyl groups protected
with para-methoxybenzyl groups (PMB). The NB was then
extended from the other side by a second cycloaddition of
DMAD to give key compound 4. The L-NB precursor was
achieved by treating 4 with cyclopentadiene to give 5. The
exo-isomer selectivity of the cyclopentadiene addition was
confirmed by the X-ray structure of 9 (Figure 2). The exo-
isomer selectivity agrees with those reported previously for
similar reactions.[34,36,37] The synthesis of the S-NB precursor
was achieved by a similar procedure to the one used for L-
NB. However, instead of treating 4 with cyclopentadiene to
give the L-NB precursor, compound 4 was treated with
quadricyclane to give the S-NB precursor. The final targets
Scheme 3. The anthraquinone moiety was achieved by treating the S- and
L-NB precursors with INF followed by dehydration.
Chem. Eur. J. 2012, 18, 283 – 292
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
285

M. N. Paddon-Row, J. J. Gooding, S. Ye et al.
ray crystallography (see the Supporting Information). Both
the Diels–Alder and the dehydration reaction required
harsher conditions for the L-NB compound than for the S-
NB compound. This difference in behaviour is probably due
to the steric hindrance present in the L-NB, which makes
the Diels–Alder and the dehydration reaction comparatively
more difficult. In contrast to the dehydration of S-NB using
trifluoroacetic acid (TFA) at 408C, the dehydration reaction
for L-NB was achieved by using para-toluenesulfonic acid
(PTSA) in dry benzene under reflux and inert atmospheric
conditions. These conditions were also sufficient to cleave
the methoxy groups at the cyclobutane ring to form hydrox-
yl groups at that position instead.
showed oxidation and reduction waves that are ascribed to
the oxidation and reduction of the anthraquinone moiety.
The area under the Faradaic peaks is similar for both the S-
and L-NB SAMs, which indicates that the molecules in each
SAM have nearly the same surface coverage of (5.20Æ0.5)ꢂ
1013 moleculescmÀ2
;
this
translates
to
around
1.92 nm2 moleculeÀ1. At first, this similarity in surface cover-
age obtained from CV may be surprising, but considering
the dimensions of the molecules (in-plane and out-of-plane,
Figure S2 in the Supporting Information) and the conforma-
tion of the two CH2SH groups on the surface (ascertained
by using in situ IR spectroscopy; see below), the area
needed for each S-NB is 1.60 nm2 moleculeÀ1, which is com-
parable to the value of 1.76 nm2 moleculeÀ1 calculated for L-
NB (see Figure S2 in the Supporting Information). When a
voltage of À750 mV is applied to the electrode, the anthra-
quinone redox moiety is reduced to the anthrahydroqui-
none. The IR spectra at À750 mV (Figure 3b) are subtracted
from an IR spectrum at 0 mV,
Characterisation of pure monolayers: The anthraquinone
moiety undergoes a 2eÀ, 2H+ redox reaction in aqueous
electrolytes (Figure 3a, inset).[39] Cyclic voltammetry (CV)
performed on the pure S- and L-NB SAMs (Figure 3a)
at which the anthraquinone
moiety is in its oxidized form.
This leads to differential spectra
in which the downward bands
(decrease in the IR absorption)
are related to the consumption
of the anthraquinone species
and the upward bands (increase
in the IR absorption) are relat-
ed to the formation of anthra-
hydroquinone species. Accord-
ing to surface selection rules on
highly reflective metal surfa-
ces,[40] bonds that are perfectly
parallel to the surface should
be invisible in the IR spectrum.
Thus, it is only possible to ob-
serve IR bands for bonds that
have a vertical component rela-
tive to the surface.[41–43] Fig-
ure 3b presents the IR spectra
of SAMs formed from either
the S- or the L-NB molecules
(i.e., in the absence of diluents).
The downward
bands at
1662 cmÀ1 are assigned to C=O
stretching in the anthraquinone
moiety. Bands at similar fre-
quencies were previously re-
ported for the C=O stretch in
benzoquinone (BQ) species in
solution (1672 cmÀ1),[44] C=O in
the oxidized form of thiolated
hydroquinone SAMs, that is,
H2QSH
(1647 cmÀ1),[18]
H2QC1SH (1650 cmÀ1)[45] and 2-
Figure 3. a) Cyclic voltammograms of pure S- (c) and L-NB (c) SAMs at 250 mVsÀ1, pH 7.4. b) In situ
(11-mercaptoundecyl)
hydro-
IR spectra of pure S- (c) and L-NB (c) SAMs at À750 mV subtracted from an IR spectrum recorded at
0 mV.
quinone (1660 cmÀ1)[14] SAMs
286
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 283 – 292

Electroactive Self-Assembled Monolayers
FULL PAPER
on gold surfaces. The upward bands at 1601 and 1629 cmÀ1
are both assigned to the aromatic C=C skeletal stretch of
the benzene ring in the anthrahydroquinone form (species
appearing). The wavenumbers are close to those reported
for an aromatic ring C=C stretching of 1626[46] and 1580,
1598 cmÀ1[47] for anthrahydroquinone derivatives in solution
and to the 1580 cmÀ1 upward band corresponding to the C=
C of SAMs formed from 2-(11-mercaptoundecyl) hydroqui-
none on a gold surface by using in situ IR spectroscopy.[14]
The IR spectra for both S- and L-NB SAMs at different sur-
face potentials are presented in Figure S6 (Supporting Infor-
mation).
both molecules. To determine the conformation of the
CH2SH groups, an IR band that corresponds to a functional
group that maintains the same vertical component and,
therefore, has a similar intensity in both SAMs should be
identified. The carbonyl groups on the anthraquinone in the
S- and L-NB SAMs will have a similar vertical component
to the surface regardless of any possible conformation of the
two CH2SH groups. Therefore, the intensity of the IR band
of the carbonyl groups (1662 cmÀ1) can be used as an inter-
nal standard to which the intensities of other bands are com-
pared. Unlike the similar orientation of the carbonyl groups
À
in each SAM, the orientation of the C C and the C=C
The downward bands at 1294, 1380 and 1439 cmÀ1 are all
bonds in the quinone/hydroquinone ring will be different for
each SAM; this orientation depends on the conformation of
the two CH2SH points of attachment to the gold surface.
Table 2 presents the IR band intensities of the (C=C and
À
assigned to the C C stretch of the quinone ring in the an-
thraquinone form. These bands are close to those reported
À
for the ring C C stretching of 1,4-naphthoquinone and 9,10-
[47,48]
anthraquinone in solution (1295, 1328, 1370, 1452 cmÀ1
)
and to the 1333 and 1293 cmÀ1 ring C C stretching of an-
thraquinone-2-carboxylic acid on gold surfaces.[49] The large
upward band between 2870 and 3600 cmÀ1 is assigned to the
À
Table 2. IR band intensities normalized to the carbonyl band at
1662 cmÀ1
.
Wavenumber [cmÀ1
]
L-NB
S-NB
À
O H stretching of water. Because an upward band repre-
1662 (carbonyl)
1.0
1.0
sents an increase in the IR absorption compared with that at
the reference potential (0 mV vs. Ag/AgCl), the increase in
1629 (upward) C=C aromatic anthrahydroquinone
1601 (upward) C=C aromatic anthrahydroquinone
0.24
0.26
0.25
0.63
1.1
0.52
0.48
0.51
1.24
1.97
À
À
1439 (downward) C C anthraquinone
the intensity of the O H stretching at À750 mV can be at-
À
1380 (downward) C C anthraquinone
tributed to an increase in the quantity of water at the inter-
face as the terminal groups switch from the anthraquinone
species (at 0 mV) to the more hydrophilic anthrahydroqui-
none species (at À750 mV). The assignment of the IR bands
for both S- and L-NB SAMs are summarized in Table 1. The
À
1294 (downward) C C anthraquinone
À
C C) of the anthraquinone moiety normalized to the band
intensity of the internal standard (C=O). The normalized
À
band intensities (Table 2), of the C C and the C=C bonds of
Table 1. Assignment of the upward and downward IR bands observed
for the S- and L-NB monolayers.
the quinone/hydroquinone ring in both the oxidized and the
reduced form of the S-NB SAM are larger by an average
factor of around 1.9 than that of the L-NB SAM. The larger
factor in the S-NB SAM is attributed to the greater vertical
Assignment
Wavenumber [cmÀ1
]
C=O
1662
C=C stretching (upward)
1601, 1629
1294, 1380, 1439
2870–3600
À
component of the C C (quinone) and the C=C (hydroqui-
À
C C stretching (downward)
À
O H bending
none) bonds in S-NB compared with L-NB. This indicates
that the L-NB molecules place the anthraquinone moiety in
a position at which it is more parallel to the gold surface.
Based on this, it is possible to deduce the conformation of
the CH2SH groups on the gold surface that best describe the
IR data. Figure 4a,b presents a proposed conformation of
the CH2SH groups in the case of the L-NB and Figure 4c,d
shows the same conformation of the CH2SH groups for the
S-NB. In both the S- and L-NB molecules, the two CH2SH
groups are pointing in opposite directions and are both
bonded to the gold surface. This CH2SH conformation
places the carbonyl groups for both S- and L-NB at the
same angle with respect to the gold surface. Conversely, the
quinone ring of the L-NB molecule (circled in Figure 4a) is
close to parallel with respect to the surface. However, the
quinone ring in the S-NB molecule (circled in Figure 4d) is
situated closer to the surface normal. Other schematic draw-
ings that represent the orientation of the quinone terminal
group are presented in Figure S3 in the Supporting Informa-
tion. The assembly of each NB molecule on the surface can
be sketched as analogous to a runner at the starting gun.
in situ IR spectra for the pure S- and L-NB SAMs showed
the same number of IR bands at the same wavenumbers
(Figure 3b and Table 1). However, the relative band intensi-
ties in each SAM, are different. The intensity ratio of the
À
bands assigned for the C C stretching of the quinone ring
(1439, 1380, 1294 cmÀ1) to the carbonyl band (1662 cmÀ1) is
larger for the S-NB SAM than for the L-NB SAM. The dif-
À
ference in behaviour suggests that the C C bonds of the
quinone ring in the S-NB molecule are more normal to the
surface compared with that of the L-NB.
The observation of bands representing anthraquinone and
anthrahydroquinone moieties indicates that the quinone
moiety in both compounds possesses a vertical component
with respect to the surface. As mentioned in the introduc-
tion, the S- and L-NB molecules have the same backbone
and the same points of adhesion to the gold surface (2ꢂ
CH2SH). Therefore, it is expected that the conformation of
the two CH2SH groups on the surface will be similar for
Chem. Eur. J. 2012, 18, 283 – 292
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
287

M. N. Paddon-Row, J. J. Gooding, S. Ye et al.
Characterisation of diluted monolayers: Figure 6a presents
cyclic voltammograms of a pure SAM formed from L-NB
molecules and a diluted SAM formed from a mixture of L-
Figure 4. The proposed orientation of a,b) L-NB and c,d) S-NB (two dif-
ferent views). Angle q is attributed to the NB backbone tilt in the xz
plane. Angle a is attributed to the NB backbone tilt in the yz plane. The
angle of the carbonyl groups with respect to the gold surface (a) is simi-
lar for both the S- and L-NB. The quinone ring (circled in a) and d)) in
L-NB has a smaller vertical component with respect to the surface com-
pared with that circled for the S-NB. According to these 3D models, a is
ꢀ258 and q is ꢀ308.
The knees of the runner (one lower and one higher) resem-
ble the CH2 groups in a trans relationship and the feet re-
semble the sulfur atoms that are attached to the gold surface
(Figure 5). We propose a two-step mechanism for the assem-
bly of the trans CH2SH feet on the gold surface. First, one
of the sulfur atoms binds to the gold surface. Second, the
other CH2SH strand rotates around the carbon atom of the
CH2SH group until it allows the other sulfur atom to bind to
the gold surface.
Figure 6. a) Cyclic voltammograms of pure L-NB SAM (c) compared
with that of a diluted L-NB/6-mercaptohexan-1-ol SAM (c); scan
rate: 250 mVsÀ1. b) In situ IR spectra of the pure L-NB SAM (c) com-
pared with that of the diluted L-NB/6-mercaptohexan-1-ol SAM (c) at
À750 mV.
NBs and 6-mercaptohexan-1-ol (1:4 molar ratio) in solution.
From the area under the Faradaic peak, the calculated sur-
face coverage for the NBs in the pure SAM was around
5.20ꢂ1013 moleculescmÀ2, whereas that for the diluted
monoACHTNUGRTNEUNG
layer was around 2.60ꢂ1013 moleculescmÀ2. The inten-
sities of the IR bands (Figure 6b) in the diluted SAM are
weaker than those of the pure SAM, as expected with fewer
L-NB molecules present on the surface. The IR band inten-
sities in the diluted SAM preserve the same relative ratio
they had in the pure L-NB SAM. In particular, the intensity
ratio of the band at 1662 cmÀ1 (C=O stretching) to the inten-
sity of the band at 1294 cmÀ1 (C C stretching) is 1.05, which
À
is close to the relative ratio of the intensities of these two
bands in the pure SAM (1.1). This suggests that the L-NB
molecules do not change their orientation on the surface
when they are diluted. The composition of the mixed SAM
was further diluted to have only a quarter of the NB mole-
Figure 5. A schematic drawing of a runner at the starting gun alongside
the structures of S- and L-NB attached to a gold surface. The runner’s
knees resemble the carbon atoms of the CH2SH and his feet resemble
the sulfur atoms.
288
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 283 – 292

Electroactive Self-Assembled Monolayers
FULL PAPER
cules on the surface; again, the IR bands intensities were
scaled proportionally and the IR band intensities diminished
to approximately one quarter of the intensity of the bands
in the pure SAM. Also, the relative band intensities again
preserved the ratio they had in the pure SAM.
The same behaviour was observed for a SAM formed
from S-NB molecules diluted with 9-mercaptononan-1-ol
(Figure 7a). The surface coverage of the NB molecules cal-
plane tilt (xz plane) and out-of-plane tilt (yz) of the NB
molecules shown for the pure SAMs in Figure 4 is retained
for NB molecules in the diluted monolayers. This tilt angle
is close to the tilt angle suggested for SAMs formed only
from alkanethiols.[50,51] Therefore, in mixed NB–alkanethiol
SAMs the L-NB molecule will place the anthraquinone moi-
eties very close to parallel to the distal surface of the di-
luents, whereas the S-NB molecules will place the anthraqui-
none moieties close to the normal of the distal surface of
the diluents.
STM studies and electrochemical impedance spectroscopy
of pure SAMs: STM images (Figure 8) reveal that SAMs
formed from S- and L-NBs have different structures. In the
Figure 7. a) Cyclic voltammograms of a pure S-NB SAM (c) compared
with that of a diluted S-NB/9-mercaptononan-1-ol SAM (c); scan rate:
250 mVsÀ1. b) In situ IR spectra of the pure S-NB SAM (c) compared
with that of the diluted S-NB/9-mercaptononan-1-ol SAM (c) at
À750 mV.
culated from the area under the Faradaic peak for the pure
and diluted SAMs (5.20ꢂ1013 and 3.23ꢂ1013 moleculescmÀ2,
respectively). As in the case of the diluted L-NB SAM, the
band intensities in the diluted S-NB SAM preserved the rel-
ative ratio observed in the pure S-NB SAM (Figure 7b). The
ratio of the intensity of the band at 1662 cmÀ1 (C=O stretch-
Figure 8. STM images of a) an unmodified (bare) Au (111) single gold
crystal surface (570ꢂ570 nm, scale bar 110 nm), b) a pure L-NB SAM
(100ꢂ100 nm, scale bar 20 nm), c) a pure L-NB SAM (20ꢂ20 nm, scale
bar 4.0 nm), d) a pure S-NB SAM (465ꢂ465 nm, scale bar 93 nm), e) a
pure S-NB SAM (40ꢂ40 nm, scale bar 7.9 nm).
ing) to the intensity of the band at 1294 cmÀ1 (C C stretch-
case of L-NB SAMs (Figure 8b and c), a highly-ordered net-
work-like structure is observed. However, in the case of S-
NB SAMs, no network-like structure is observed (Figure 8d
and e).
These images agree with the structure of the SAMs sug-
gested by the in situ IR spectroscopy. In the case of the S-
NB SAM, the anthraquinone head groups, which are orient-
ed close to the surface normal, are spaced out due to the
space required by the backbone of the NBs. However, we
À
ing) is 1.88, which is close to the relative ratio of these two
bands in the pure SAM (1.97). Because the relative intensi-
ties of the IR bands are similar for both pure and diluted
SAMs, this strongly suggests that the orientation of NB mol-
ecules are determined by the conformation of the two
CH2SH groups on the gold surface, and this conformation
will be similar whether the NB molecules are spaced by di-
luents or present as the only component. Therefore, the in-
Chem. Eur. J. 2012, 18, 283 – 292
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
289

M. N. Paddon-Row, J. J. Gooding, S. Ye et al.
speculate that the nearly parallel orientation of the anthra-
quinone moieties in the L-NB SAM allows interaction be-
tween the anthraquinone head groups on the surface despite
the space required for the backbone of the bridges. There-
fore, the ordered arrays of L-NB bridges observed may be a
consequence of interactions between the anthraquinones. A
magnified STM image of the L-NB SAM (5ꢂ5 nm, 25 nm2)
appears to have 11 bright spots (Figure S5 in the Supporting
Information). Each spot correlates to an individual molecule
and, therefore, the area per molecule is 2.3 nm2. As men-
tioned above, the area per L-NB molecule calculated by
using CV is around 1.92 nm2 moleculeÀ1, which is reasonably
close to the value obtained from the STM images.
sulating monoACTHUNGETRNNUlG ayer; Wang et al. suggest that phase angles of
80 to 888 confirm the presence of a highly insulating dielec-
tric material on the electrode surface.[54] Conversely, the
phase angle at low frequency for the S-NB SAM was about
778, which indicates that it is less insulating than the L-NB
SAM. A more quantitative information about the structures
of the monolayers was obtained by fitting the impedance
data to the circuit presented in Figure 10, inset. The resist-
The difference in the structure of S- and L-NB SAMs, ob-
served in the STM images, suggests that S-NB SAMs have
more space between the anthraquinone moieties than the L-
NB SAMs. This hypothesis was explored by using electro-
chemical impedance spectroscopy (EIS), which averages in-
formation over the entire surface area. EIS gives quantita-
tive information on the dielectric properties of the SAMs
and, therefore, the accessibility of the underlying gold sur-
face to ions in solution. The measurements were obtained
over a frequency range of 100 KHz to 1 Hz and at a dc po-
tential that does not overlap with the Faradaic process (0 V
vs. Ag/AgCl). This technique is exceedingly sensitive to any
defects in the SAM. The resulting impedance data can be
represented as phase angle versus frequency plots, referred
to as Bode plots (Figure 9). The phase angle reflects the rel-
ative magnitude of the capacitive impedance and the solu-
tion resistance. Ideally, the Bode phase plot of a Helmholtz
capacitor should give a phase angle of 908 at low frequen-
cies. Deviation of the phase angle from 908 (at low frequen-
cies) would suggest an electrical current leakage through the
dielectric layer of the Helmholtz capacitor.[52–54] At higher
frequencies, the total impedance is dominated by the solu-
tion resistance and the phase angle decreases towards zero.
In this study, the Bode phase plot (Figure 9) demonstrates
that the L-NB SAM behaves as a close-to-ideal capacitive
SAM with a phase angle of about 878, which indicates an in-
Figure 10. Nyquist plots of the impedance of SAMs formed from L-NB
!
*
(
) and S-NB ( ); the symbols represent experimental data and the
lines are the best fits. Inset: The equivalent circuit used to fit the impe-
dance data. The fitting parameters are presented in Table S1 in the Sup-
porting Information.
ance of the L-NB SAM ((8.10Æ0.90) MWcm2) was found to
be about an order of magnitude greater than the resistance
of the S-NB SAM ((1.10Æ0.11) MWcm2). This difference in
the resistance values for the two SAMs is ascribed to more
solvent penetration in the S-NB SAM than in the L-NB
SAM. The difference in the dielectric properties can also be
inferred from the capacitance values of each SAM. The ca-
pacitance of the L-NB SAM is (4.55Æ0.01) mFcmÀ2, where-
as that of the S-NB SAM is (9.84Æ0.08) mFcmÀ2. The capac-
itance of a monolayer is given by C=ere0A/d, in which C is
the capacitance, er is the dielectric constant of the SAM, e0
is the permittivity of free space, A is the surface area and d
is the thickness of the monolayer. Therefore, for the systems
presented herein there are two factors that are important
with regards to the magnitude of the measured capacitance:
the dielectric constant of the SAM, which is directly propor-
tional to the capacitance, and the thickness of the film. The
higher capacitance of the S-NB SAM can be ascribed to a
higher dielectric constant of the S-NB SAM due to solvent
penetration through the backbone of the NB molecules. De-
spite being a thicker film, the solvent penetration in the S-
NB SAM increases the dielectric constant of the SAM. The
L-NB SAM has a lower capacitance due to a greater ability
to block the penetration of the solvent owing to the anthra-
quinone moieties being held close to parallel to the surface.
The difference in the dielectric properties supports the hy-
pothesis that the anthraquinones form a close packed layer
!
*
Figure 9. Bode phase plots of S- ( ) and L-NB ( ) SAMs. The impe-
dance spectra were obtained at 0 V vs. Ag/AgCl and an ac amplitude of
10 mV at pH 7.4.
290
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 283 – 292

Electroactive Self-Assembled Monolayers
FULL PAPER
in dynamic mode with a spectral resolution of 4 cmÀ1. Each IR spectrum
was recorded during a fixed potential with a waiting time of 100 s. Details
of the in situ IR spectroscopy measurements are described else-
where.[57,58] All IR spectra were subtracted from a spectrum recorded at
0 V (vs. Ag/AgCl) in the same solution in which the anthraquinone was
in its oxidized form. All electrochemical experiments were carried out by
using an Ivium CompactStat potentiostat (Netherlands) at RT. A three-
compartment electrochemical cell was used for electrochemical charac-
terization. The reference electrode was a saturated KCl Ag/AgCl elec-
trode and all potentials are referred to the reference. The counter elec-
trode was a platinum gauze. The electrolyte solution used was phosphate
buffer (PBS; 0.1m, pH 7.4) with a controlled ionic strength of 0.6m NaCl
using NaH2PO4, Na2HPO4 and NaCl (Wako Pure Chemicals, Japan). The
electrolyte solution was deaerated with Ar (99.99%, Hokkaido Air
Water, Japan) for at least 15 min before each measurement.
in the L-NB SAM, whereas the anthraquinones in the S-NB
SAM are not closely associating. The EIS agrees with this
hypothesis because the lower penetration of solvent in the
L-NB SAM is consistent with the anthraquinone rings form-
ing a shelter over the hydrophobic backbone of the bridge
and, therefore, preventing the electrolyte from penetrating
through to the electrode surface. Conversely, the anthraqui-
none moiety in the S-NB SAM does not form a complete
shelter and, therefore, cannot completely block the electro-
lyte from penetrating through the backbone of the NB mol-
ecules. The difference in the dielectric properties of both
SAMs further indicates that the L-NB molecules hold the
anthraquinone groups parallel to the gold surface, whereas
the S-NB molecules place the anthraquinone groups close to
the surface normal.
Scanning tunneling microscopy: STM images were acquired by using a
Picoscan system (Molecular Imaging Corp., USA) under ambient condi-
tions. SAMs were formed by incubating Au (111) surfaces in a 1 mm solu-
tion of the NB compounds prepared in a solution of CH2Cl2. The STM
tip was prepared by mechanically cutting a Pt/Ir wire. All images were
acquired in a constant-current mode. The typical imaging conditions
were bias voltages of 0.35 V and a tunnelling current of 25 pA. Images
were manipulated with WxSM software (Nanotec, Spain) and the con-
trast was enhanced by applying a low-pass filter.
Conclusion
The synthesis of straight and L-shaped NB compounds with
an anthraquinone redox-active head group and two thiol
groups is described herein. These compounds were shown to
form self-assembled monolayers on gold surfaces. In situ IR
spectroscopy, STM imaging and the dielectric properties of
the SAMs demonstrated that this class of NB molecules can
precisely locate the redox moiety in a well-defined orienta-
tion with respect to the electrode surface and with respect
to the surface of diluents in a mixed monolayer. The anthra-
quinone rings of SAMs formed from L-NBs were shown to
be located close to parallel to the surface, whereas those
from SAMs prepared from S-NBs were found to be close to
the surface normal. These systems are currently being ex-
ploited to study the influence of the environment surround-
ing the anthraquinone moiety on the redox reaction by
forming SAMs in which the environment around the anthra-
quinone moiety is altered by using diluents of different
length and different distal groups.
Electrochemical impedance spectroscopy: The electrochemical impe-
dance spectroscopy experiments were recorded by using a Solartron 1287
electrochemical interface with a Solartron Impedance/Gain-Phase Analy-
ser. The AC amplitude was 10 mV. Data analysis was carried out by
using the program Z View by Scribner Associates. The measurements
were performed in a three-electrode cell containing an Au (111) working
electrode by forming a meniscus with the electrolyte, a platinum foil
counter electrode and a saturated KCl Ag/AgCl reference electrode.
X-ray structure determination: CCDC-826618 (compound S1 in the Sup-
porting Information) and CCDC-826619 (9) contain the supplementary
crystallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Acknowledgements
N.D. acknowledges ARCNN for a travel fellowship to Hokkaido Univer-
sity, Japan. The Australian Research Council under the Discovery Proj-
ects Funding Scheme (DP0556397) is acknowledged. Dr. Mohan Bhadb-
hade from the Analytical Center, UNSW, is acknowledged for the X-ray
structures presented in this paper. This work was partially supported by a
Grant-in-Aid for Exploratory Research (21655074) and a Grant-in-Aid
for Scientific Research on Innovative Areas “Coordination Program”
(22108501) from MEXT, Japan.
Experimental Section
In situ IR spectroscopy: 6-Mercaptohexan-1-ol and 9-mercaptononan-1-
ol were purchased from Sigma–Aldrich (>98%). The detailed schemes
and experimental procedures for the synthesis of S- and L-NBs are given
in the Supporting Information. SAMs were formed on gold films (thick-
ness ꢀ50 nm) that were chemically deposited on the reflecting surface of
a hemicylindrical Si prism. The procedure of chemically depositing gold
is described elsewhere.[55] Before surface modification with SAMs, the
gold surfaces were electrochemically cleaned by scanning the potential
between 0.0 and +1.5 V (vs. Ag/AgCl) for 10 cycles in 0.5m H2SO4 to
remove possible organic additives adsorbed on the surface during the Au
chemical deposition. SAMs formed from pure NB were formed by incu-
bating the surfaces in a solution of NB (1 mm) in dichloromethane
(CH2Cl2) for 18 h. Mixed SAMs were prepared by incubating the surfaces
in a solution of NB and alkanethiol diluent (molar ratio 1:4; total concen-
tration 1 mm) in CH2Cl2.[56] The surfaces were incubated in the mixed so-
lution for 20 min, after which they were removed and placed in a solution
of alkanethiol (1 mm) in CH2Cl2 for 18 h. The in situ IR spectroscopy
measurements were carried out by using a BioRad FTS-60A/896 spec-
trometer equipped with a MCT detector. The IR spectra were recorded
[1] M. Gꢃmez, F. J. Gonzꢄlez, I. Gonzꢄlez, J. Electroanal. Chem. 2005,
578, 193.
[2] A. D. Abhayawardhana, T. C. Sutherland, J. Phys. Chem. C 2009,
113, 4915.
[3] B. L. Trumpower, J. Biol. Chem. 1990, 265, 11409.
[4] S. M. Golabi, J. B. Raoof, J. Electroanal. Chem. 1996, 416, 75.
[5] A. Sarapuu, K. Vaik, D. J. Schiffrin, K. Tammeveski, J. Electroanal.
Chem. 2003, 541, 23.
[6] K. L. Wong, G. Pawin, K.-Y. Kwon, X. Lin, T. Jiao, U. Solanki,
R. H. J. Fawcett, L. Bartels, S. Stolbov, T. S. Rahman, Science 2007,
315, 1391.
[7] M. Matsson, D. Tolstoy, R. Aasa, L. Hederstedt, Biochemistry 2000,
39, 8617.
[8] P. Petrangolini, A. Alessandrini, L. Berti, P. Facci, J. Am. Chem.
Soc. 2010, 132, 7445.
Chem. Eur. J. 2012, 18, 283 – 292
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
291

M. N. Paddon-Row, J. J. Gooding, S. Ye et al.
[9] X. Li, B. Xu, X. Xiao, X. Yang, L. Zang, N. J. Tao, Faraday Discuss.
2006, 131, 111.
[10] A. T. Hubbard, Chem. Rev. 1988, 88, 633.
[33] H. Oevering, M. N. Paddonrow, M. Heppener, A. M. Oliver, E. Cot-
saris, J. W. Verhoeven, N. S. Hush, J. Am. Chem. Soc. 1987, 109,
3258.
[11] S. A. Trammell, D. A. Lowy, D. S. Seferos, M. Moore, G. C. Bazan,
N. Lebedev, J. Electroanal. Chem. 2007, 606, 33.
[12] S. A. Trammell, M. Moore, T. L. Schull, N. Lebedev, J. Electroanal.
Chem. 2009, 628, 125.
[34] R. N. Warrener, I. G. Pitt, D. N. Butler, Chem. Soc. Chem. Commun.
1983, 1340.
[35] T. Mitsudo, K. Kokuryo, T. Shinsugi, Y. Nakagawa, Y. Watanabe, Y.
Takegami, J. Org. Chem. 1979, 44, 4492.
[13] Y. Sato, M. Fujita, F. Mizutani, K. Uosaki, J. Electroanal. Chem.
1996, 409, 145.
[14] S. Ye, A. Yashiro, Y. Sato, K. Uosaki, J. Chem. Soc. Faraday Trans.
1996, 92, 3813.
[36] N. J. Head, A. M. Oliver, K. Look, N. R. Lokan, G. A. Jones, M. N.
Paddon-Row, Angew. Chem. Int. Ed. 1999, 38, 3219.
[37] K. Kumar, R. J. Tepper, Y. Zeng, M. B. Zimmt, J. Org. Chem. 1995,
60, 4051.
[15] S. A. Trammell, M. Moore, D. Lowy, N. Lebedev, J. Am. Chem. Soc.
2008, 130, 5579.
[16] W. Zhang, S. M. Rosendahl, I. J. Burgess, J. Phys. Chem. C 2010,
114, 2738.
[17] A. G. Larsen, K. V. Gothelf, Langmuir 2005, 21, 1015.
[18] T. Sasaki, I. T. Bae, D. A. Scherson, B. G. Bravo, M. P. Soriaga,
Langmuir 1990, 6, 1234.
[38] B. L. Serpaud, Y. Lepage, Bull. Soc. Chim. Fr. 1977, 539.
[39] S. A. Trammell, N. Lebedev, J. Electroanal. Chem. 2009, 632, 127.
[40] R. Greenler, J. Chem. Phys. 1966, 44, 310.
[41] L. Pranger, A. Goldstein, R. Tannenbaum, Langmuir 2005, 21, 5396.
[42] K. V. G. K. Murty, M. Venkataramanan, T. Pradeep, Langmuir 1998,
14, 5446.
[43] M. Moskovits, J. Chem. Phys. 1982, 77, 4408.
[19] J. J. Hickman, D. Ofer, P. E. Laibinis, G. M. Whitesides, M. S.
Wrighton, Science 1991, 252, 688.
[20] A. J. Black, T. T. Wooster, W. E. Geiger, M. N. Paddon-Row, J. Am.
Chem. Soc. 1993, 115, 7924.
[21] T. T. Wooster, P. R. Gamm, W. E. Geiger, A. M. Oliver, A. J. Black,
D. C. Craig, M. N. Paddon-Row, Langmuir 1996, 12, 6616.
[22] M. N. Paddon-Row, Acc. Chem. Res. 1982, 15, 245.
[23] M. N. Paddon-Row, Acc. Chem. Res. 1994, 27, 18.
[24] V. Balaji, L. Ng, K. D. Jordan, M. N. Paddon-Row, H. K. Patney, J.
Am. Chem. Soc. 1987, 109, 6957.
[44] B. R. Clark, D. H. Evans, J. Electroanal. Chem. 1976, 69, 181.
[45] I. T. Bae, M. Sandifer, Y. W. Lee, D. A. Tryk, C. N. Sukenik, D. A.
Scherson, Anal. Chem. 1995, 67, 4508.
[46] M. Umadevi, V. Ramakrishnan, Spectrochim. Acta Part A: Mol.
Spectrosc. 2003, 59, 393.
[47] S. N. S. Singh, R. S. Singh, Spectrochim. Acta Part A 1968, 24, 1591.
[48] A. Latef, J. C. Bernede, S. Benhida, Thin Solid Films 1991, 195, 289.
[49] S. W. Han, S. W. Joo, T. H. Ha, Y. Kim, K. Kim, J. Phys. Chem. B
2000, 104, 11987.
[50] A. Ulman, J. E. Eilers, N. Tillman, Langmuir 1989, 5, 1147.
[51] K. Sinniah, J. Cheng, S. Terrettaz, J. E. Reutt-Robey, C. J. Miller, J.
Phys. Chem. 1995, 99, 14500.
[25] A. D. Joran, B. A. Leland, G. G. Geller, J. J. Hopfield, P. B. Dervan,
J. Am. Chem. Soc. 1984, 106, 6090.
[26] S. I. Khan, A. M. Oliver, M. N. Paddon-Row, Y. Rubin, J. Am.
Chem. Soc. 1993, 115, 4919.
[27] K. A. Jolliffe, T. D. M. Bell, K. P. Ghiggino, S. J. Langford, M. N.
Paddon-Row, Angew. Chem. 1998, 110, 959; Angew. Chem. Int. Ed.
1998, 37, 915.
[52] E. Boubour, R. B. Lennox, Langmuir 2000, 16, 7464.
[53] E. Boubour, R. B. Lennox, J. Phys. Chem. B 2000, 104, 9004.
[54] W. Wang, S. Zhang, P. Chinwangso, R. C. Advincula, T. R. Lee, J.
Phys. Chem. C 2009, 113, 3717.
[55] H. Miyake, S. Ye, M. Osawa, Electrochem. Commun. 2002, 4, 973.
[56] P. K. Eggers, H. M. Zareie, M. N. Paddon-Row, J. J. Gooding, Lang-
muir 2009, 25, 11090.
[28] K. A. Jolliffe, S. J. Langford, A. M. Oliver, M. J. Shephard, M. N.
Paddon-Row, Chem. Eur. J. 1999, 5, 2518.
[29] W. R. Yang, M. W. Jones, X. Li, P. K. Eggers, N. Tao, J. J. Gooding,
M. N. Paddon-Row, J. Phys. Chem. C 2008, 112, 9072.
[30] P. K. Eggers, P. Da Silva, N. A. Darwish, Y. Zhang, Y. Tong, S. Ye,
M. N. Paddon-Row, J. J. Gooding, Langmuir 2010, 26, 15665.
[31] J. J. Gooding, F. Mearns, W. Yang, J. Liu, Electroanalysis 2003, 15,
81.
[57] S. Ye, Y. Sato, K. Uosaki, Langmuir 1997, 13, 3157.
[58] S. Ye, W. Zhou, M. Abe, T. Nishida, L. Cui, K. Uosaki, M. Osawa,
Y. Sasaki, J. Am. Chem. Soc. 2004, 126, 7434.
[32] W. Koch, M. C. Holthausen, A Chemistꢀs Guide to Density Function-
al Theory, 2nd ed., Wiley-VCH, Weinheim, 2001.
Received: May 24, 2011
Published online: November 23, 2011
292
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 283 – 292