Metal-molecule-metal junctions in langmuir-blodgett films using a new linker: Trimethylsilane

Article abstract of DOI:10.1002/chem.201001181

Herein trimethylsilane (TMS) is demonstrated to be an efficient binding group suitable for construction of metal-molecule-metal (M-mol-M′) junctions, in which one of the metal contacts is an atomically flat gold surface and the other a scanning tunnelling microscopy (STM) tip. The molecular component of the M-mol-M′ devices is an oligomeric phenylene ethynylene (OPE) derivative Me₃Si C≡C{C₆H₄C≡C} ₂C₆H₄NH₂, featuring both Me ₃SiC≡C and NH₂ metal contacting groups. This compound can be assembled into Langmuir-Blodgett (LB) films on Au-substrates by surface binding through the amine groups. Alternatively, low coverage (sub-monolayer) films are formed by adsorption from solution. In the case of condensed monolayers top electrical contacts are formed to STM tips through the TMS end group. In low coverage films, single molecular bridges can be formed between the gold surface and a gold STM tip. The similarity in the I-V response of a one-layer LB film and the single molecule conductance experiments reveals several points of critical importance to the design of molecular components for use in the construction of M-mol-M′ junctions. Firstly, the presence of neighbouring π systems does not have a significant effect on the conductance of the M-mol-M′ junction. Secondly, in the STM configuration, intermolecular electron hopping does not significantly enhance the junction transport characteristics. Thirdly, the symmetric behaviour of the I-V curves obtained, despite the different metal-molecule contacts, indicates that the molecule is simply an amphiphilic electron-donating wire and not a molecular diode with strong rectifying characteristics. Finally, the conductance values obtained from the amine/TMS-contacted OPE described here are of the same order of magnitude as thiol anchored OPEs, making them attractive alternatives to the more conventionally used thiol-contacting chemistry for OPE molecular wires. Junction box: 4-[4-{4-(Trimethylsilylethynyl)phenylethynyl}phenylethynyl]aniline has been assembled into well-packed monolayer films by using the Langmuir-Blodgett technique. Trimethylsilane (TMS) is demonstrated to be an efficient binding group suitable for construction of metal-molecule-metal junctions, making this group an attractive alternative to the more conventionally used thiol-contacting chemistry for molecular wires (see picture). Copyright

Full text of DOI:10.1002/chem.201001181

DOI: 10.1002/chem.201001181
Metal–Molecule–Metal Junctions in Langmuir–Blodgett Films Using a New
Linker: Trimethylsilane
Gorka Pera,[a, b] Santiago Martꢀn,[a, b] Luz M. Ballesteros,[a, b] Adam J. Hope,[c]
Paul J. Low,[c] Richard J. Nichols,[d] and Pilar Cea*[a, b]
Abstract:
Herein
trimethylsilane
monolayers top electrical contacts are
formed to STM tips through the TMS
end group. In low coverage films,
single molecular bridges can be formed
between the gold surface and a gold
STM tip. The similarity in the I–V re-
sponse of a one-layer LB film and the
single molecule conductance experi-
ments reveals several points of critical
importance to the design of molecular
components for use in the construction
of M–mol–M’ junctions. Firstly, the
presence of neighbouring p systems
does not have a significant effect on
the conductance of the M–mol–M’
junction. Secondly, in the STM configu-
ration, intermolecular electron hopping
does not significantly enhance the junc-
tion transport characteristics. Thirdly,
the symmetric behaviour of the I–V
curves obtained, despite the different
metal–molecule contacts, indicates that
the molecule is simply an amphiphilic
electron-donating wire and not a mo-
lecular diode with strong rectifying
characteristics. Finally, the conductance
values obtained from the amine/TMS-
contacted OPE described here are of
the same order of magnitude as thiol
anchored OPEs, making them attrac-
tive alternatives to the more conven-
tionally used thiol-contacting chemistry
for OPE molecular wires.
(TMS) is demonstrated to be an effi-
cient binding group suitable for con-
struction of metal–molecule–metal (M–
mol–M’) junctions, in which one of the
metal contacts is an atomically flat
gold surface and the other a scanning
tunnelling microscopy (STM) tip. The
molecular component of the M–mol–
M’ devices is an oligomeric phenylene
ethynylene (OPE) derivative Me3Si
ꢀ
C C
ꢀ
U
featuring
ꢀ
both Me3SiC C and NH2 metal con-
tacting groups. This compound can be
assembled into Langmuir–Blodgett
ꢁ
(LB) films on Au substrates by surface
binding through the amine groups. Al-
ternatively, low coverage (sub-mono-
layer) films are formed by adsorption
from solution. In the case of condensed
Keywords: conducting materials
·
Langmuir–Blodgett films · molecu-
lar wires · thin layers
Introduction
[a] G. Pera, Dr. S. Martꢀn, L. M. Ballesteros, Dr. P. Cea
Departamento de Quꢀmica Orgꢁnica y Quꢀmica Fꢀsica
Facultad de Ciencias
The development of ever smaller sized electronic devices
has turned interest toward organic molecules as potential
circuit components, such as conductors, rectifiers, transistors
and logic gates.[1–3] To understand their electrical behaviour
direct current-voltage (I–V) measurements are a require-
ment for molecules arranged into metal-molecule-metal
junctions. There are now a number of methods capable of
achieving this feat and they include mechanical break junc-
tions,[4,5] nanopores,[6,7] cross-wire junctions,[8] scanning tun-
nelling microscopy (STM)[9–13] or conducting atomic force
microscopy (c-AFM).[14–16] Molecular conductance values
obtained from such devices, depend not only on the inherent
molecular features, but also on other important parameters,
such as the metal–molecule contact. Often, small variations
in the nature or characteristics of the metal–molecule con-
Universidad de Zaragoza, 50009 (Spain)
E-mail: pilarcea@unizar.es
[b] G. Pera, Dr. S. Martꢀn, L. M. Ballesteros, Dr. P. Cea
Instituto de Nanociencia de Aragꢂn (INA)
edificio I+D+I, C/Mariano Esquilor s/n Campus Rio Ebro
50018, Zaragoza (Spain)
[c] A. J. Hope, Prof. P. J. Low
Department of Chemistry
University of Durham
Durham DH1 3 LE (UK)
[d] Prof. R. J. Nichols
Department of Chemistry
University of Liverpool
Crown Street, Liverpool, L69 7ZD (UK)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201001181.
13398
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13398 – 13405

FULL PAPER
tact can have a pronounced effect on charge transport in the
junction.[17,18]
LB technique in previous studies.[27–31] Removing the alkyl
tail allows a direct contact between the conjugated skeleton
of the molecule through functional groups to the gold sub-
strate and the STM tip, thereby improving the electrical
transport through these molecular assemblies.[13,32–36]
The Langmuir–Blodgett (LB) technique is an effective
tool for transferring well-ordered molecular films at the air–
water interface onto a solid support, permitting a degree of
control over the molecule–surface interaction or contact, as
it is applicable for forming both chemisorbed and physisor-
bed films.[19] Therefore, the use of the LB metal–organic as-
sembly technologies provides diverse opportunities for the
exploration of different metal–molecule contacts that, so
far, have been limited to a few interfaces formed by the tra-
ditional self-assembly (SA) method. The thiol moiety has
been used extensively to attach molecules to gold surfaces
using such SA methods. How-
Results and Discussion
Characterisation of Langmuir and Langmuir–Blodgett films:
Figure 1a shows the representative surface pressure (p–A)
and surface potential (DV–A) isotherms of TMS-OPE-NH2
on a water subphase. The p–A isotherm is characterised by
ever, the shortcomings of the
ꢁ
Au S attachment have become
apparent in recent years.[20] For
ꢁ
example, the Au S bond is
mobile at room temperatures
giving rise to device instabilities
such as stochastic switch-
ing.[21–23] Measurements at the
single molecule level have dem-
ꢁ
onstrated that Au S contacts
have multiple contact values,
Figure 1. a) p–A (solid line) and DV–A (dashed line) isotherms of TMS-OPE-NH2 on water; b) p–A isotherm
(left) and tilt angle, q , formed by the normal to the surface and the transition dipole moment of the molecule
(right) versus the area per molecule.
which depend on the nature of
the coordination bond between
the sulphur and the gold con-
tact; moreover, these contacts
have shown unexpectedly high resistance.[24] To overcome
these problems it is clear that alternative contact chemistries
should be intensively investigated and LB technologies pro-
vide a powerful platform for achieving this.
a zero surface pressure in the 1.3–0.90 nm2 moleculeꢁ1 range,
featuring a lift-off at approximately 0.90 nm2 moleculeꢁ1 fol-
lowed by an increase of the surface pressure upon compres-
sion. Changes in the slope reveal a progressive orientation
and/or reorientation of the molecules at the interface during
compression. DV–A isotherms are well-known to anticipate
the phase changes, showing deviations a few ꢄ2 before they
are detected in the p–A isotherms.[37] The sudden decrease
of DV values at approximately 0.47 nm2 upon compression is
especially noteworthy and is consistent with a collapse of
the monolayer, in which the dipole moments are randomly
distributed in a three-dimensional arrangement of TMS-
OPE-NH2 molecules. Such a collapse can also be observed
at slightly lower areas per molecule in the p–A isotherm.
Molecular organisation in Langmuir films was investigat-
ed in situ by UV/Vis reflection spectroscopy through the re-
flection of unpolarised light under normal incidence. Nor-
malised reflection spectra (DRn =DR·Area per molecule, in
which DR is the reflection) recorded at different values of
the surface pressure are illustrated in Figure 2. As can be
observed, the intensity of the band at l=310 nm, character-
istic of the phenylene-ethynylene moiety,[38] decreases con-
siderably upon compression. This phenomenon is indicative
of a gradual decrease of the tilt angle formed by the normal
to the surface and the dipole transition moment of the mole-
cules revealing a progressive orientation of the molecules at
the interface upon compression.[39] In addition, the band has
a marked blue-shift of approximately 33 nm with respect to
To the best of our knowledge, trimethylsilane (TMS) has
not yet been employed in metal–molecule–metal junctions
studies, although oligothiophene derivatives[25] and n-alka-
nes[26] bearing the TMS group have been shown to form
long-range and highly-stable self-assembled monolayers
onto reconstructed AuACTHNUGRTENUNG(111). Herein we examine the role of
this unexplored linker by measuring the conductivity of
both single molecules (SMC) and assemblies of molecules
positioned onto gold substrates by the LB technique. Thus,
the I–V response of an asymmetric oligomeric phenylene
ethynylene (OPE) derivative (TMS-OPE-NH2), functional-
ised at one of the terminal position with a TMS group
ꢀ
linked to the OPE core through a C C triple bond, and fur-
ther functionalised with an amine binding group, is present-
ed here. TMS-OPE-NH2 lacks the supporting, but insulating,
alkyl chain (“tail”), which has been included in the molecu-
lar skeleton of OPE derivatives assembled into films by the
Chem. Eur. J. 2010, 16, 13398 – 13405
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13399

P. Cea et al.
(Figure 3). At a surface pressure of 5 mNmꢁ1 the monolayer
features some domains, which coalesce as the surface pres-
sure increases. This phenomenon is accompanied by a cer-
tain increase in the brightness of the images indicative of a
gradual tilt of the molecules as confirmed by the reflection
experiments (Figures 1b and 2). At 15 mNmꢁ1 a much more
homogeneous film can be observed, although some less well
covered regions can still be seen in the BAM images. At
20 mNmꢁ1 a highly homogeneous monolayer is formed.
Langmuir monolayers were transferred onto solid sub-
strates, initially immersed in the subphase by the vertical
dipping method at a surface pressure of 20 mNmꢁ1, to form
one-layer LB films. The transfer ratio calculated by the
trough software was 1. This deposition rate was also as-
sessed using a quartz crystal microbalance (QCM). Thus, the
frequency change (Df) for a QCM quartz resonator before
and after the deposition process was determined. Using the
Sauerbrey equation [Eq. (1)]:[42]
Figure 2. Left: Normalised reflection spectra, DRn, of TMS-OPE-NH2
spread onto water recorded at the indicated surface pressures (solid
lines). Right: The black dotted line shows the UV/Vis absorption spec-
trum of TMS-OPE-NH2 1ꢅ10ꢁ5 m in chloroform solution and the dashed
red line shows the UV/Vis spectrum of a one-layer LB film.
2f02Dm
the absorption band in solution (Figure 2, dotted line), at-
tributable to the formation of H-aggregates (side by side
alignment of transition dipole moments of dyes).[40] There is
a small red shift of 4 nm of the maximum wavelength of the
normalised reflection spectra upon compression, which
could be attributed to a less polar environment when the
surface pressure increases. In keeping with this suggestion, a
bathochromic shift of 2 nm in this characteristic absorption
band of TMS-OPE-NH2 is observed in CCl4 solution (non-
polar) compared to the CHCl3 solution (polar). The angle,
q, defined as the angle formed by the normal to the surface
and the transition dipole moment of the molecule (inset of
Figure 1b) can be calculated from the normalised reflection
spectra.[39] The variation of this angle upon compression is
shown in Figure 1b. Notably, the spectra recorded in the gas
phase of the isotherm (1.3–0.9 nm2) show a large dispersion
in the angle values, probably as a result of the presence of
non-uniform domains in the monolayer underneath the
fibre-optic detector, which produce significant fluctuations
in the signal.[41] However, as soon as the isotherm takes off,
all the spectra follow a logical evolution.
Df ¼ ꢁ
ð1Þ
A11q=2m1q=2
for which f0 is the fundamental resonant frequency of
5 MHz, Dm(g) is the mass change, A is the electrode area,
1q is the density of the quartz (2.65 gcmꢁ3), and mq is the
shear module (2.95·1011 dyncmꢁ2), and TMS-OPE-NH2 mo-
lecular weight (389 gmolꢁ1), the surface coverage (G) is
3.6·10ꢁ10 molcmꢁ2, which is in excellent agreement with the
estimated value for the saturated surface coverage,
3.5·10ꢁ10 molcmꢁ2, determined from the molecular area of
TMS-OPE-NH2 at the air–water interface at a surface pres-
sure of 20 mNmꢁ1.
XPS experiments revealed that under these transference
process conditions, the amine group of the OPE derivative
bonds directly to the gold substrate giving the following
schematic structure: Au-NH2-OPE-TMS. Figure 4 shows the
XPS scans of the N1s region in LB films, as well as in the
powder. The XPS spectrum of a solid sample of NH2-OPE-
TMS powder shows a peak at 400.9 eV, corresponding to a
free amine group (Figure 4a, top). In contrast, a peak corre-
sponding to N atoms in the LB film, with a binding energy
of 399.2 eV, is observed in the XPS spectrum of the LB
Brewster angle microscopy (BAM) investigations were
made during the compression of the Langmuir film and
gave further insight into the formation of the monolayer
Figure 3. The Brewster angle microscopy (BAM) images of the Langmuir films, at the indicated surface pressures, for TMS-OPE-NH2 on water. The
field of view along the x axis for the BAM images is 3000 mm.
13400
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13398 – 13405

Metal–Molecule–Metal Junctions
FULL PAPER
and section analysis profile of a one-layer LB film are
shown in Figure 5, revealing a homogeneous surface (the
film roughness, calculated in terms of the root mean squared
(RMS), is quite low, ꢂ0.05 nm), in which the mica is wholly
covered by the monolayer (lower surface pressures of trans-
ference show the presence of holes or incompletely covered
surfaces).
A UV/Vis spectrum of a one-layer LB film of TMS-OPE-
NH2 on quartz was also obtained (dashed red line in
Figure 2). The spectrum is similar in profile to the reflection
spectra obtained at the air–water interface upon compres-
sion, with a maximum absorption feature at l=310 nm. This
similarity in both spectra indicates that the 2D H-aggregates
formed at the air–water interface are preserved in LB films.
Electrochemical electron transfer currents at electrodes
under controlled potential provide an indirect measure of
defect densities in thin films[45] and can be conveniently
studied by cyclic voltammetry for the film coated electrodes.
Cyclic voltammograms (CVs) obtained from aqueous solu-
tions containing 1 mm K3[Fe(CN)6] and 0.1m KCl for a bare
gold and for a gold working electrode modified by a one-
layer LB film deposited at 20 mNmꢁ1 are shown in Figure 6.
The electrochemical response of a bare gold electrode ex-
hibits a clear voltammetric wave for ferricyanide. The ab-
sence of these peaks for the electrode modified by the LB
film indicates a large passivation of the electrode and a lack
(or low density) of holes or defects in the monolayer. This
indicates that TMS-OPE-NH2 yields high-quality films.
STM for I–V measurements: The electrical properties of a
one-layer LB film deposited onto gold substrates initially
immersed in the subphase at 20 mNmꢁ1 were investigated
using a STM. Current-voltage (I–V) curves were recorded
and averaged from multiple (525) scans at different loca-
tions on the substrate and using different samples to ensure
the reproducibility and reliability of the measurements. Nev-
ertheless, the vertical position of the STM tip with respect
to the monolayer is a determining factor in the recorded
electrical properties of the assembled molecules. If the STM
tip is not in contact with the monolayer, a gap exists be-
tween tip and monolayer. Alternatively, at close enough tip
proximities the STM tip will penetrate the monolayer. In
such circumstances it is difficult to know how far the tip
penetrates and thus what portions of the molecules contrib-
Figure 4. a) XPS spectra of N1s region for the powder (top) and for a
one-layer LB film deposited at 20 mNmꢁ1 onto
a gold substrate
(bottom), b) XPS spectra of the Au4f region for bare Au substrate and a
Au substrate covered by one monolayer of NH2-OPE-TMS.
films (Figure 4a, bottom). These observations are consistent
with the amine group being adsorbed onto the gold surface
in the LB films.[43]
The lack of alkyl chains in this OPE derivative, commonly
used to stabilise LB films by promoting strong van der
Waals interactions between
neighbouring molecules, and
the presence of the trimethylsi-
lane group opens the question
of whether the LB films of
TMS-OPE-NH2 prepared here
are well packed, forming an
uniform and homogeneous
monolayer, or not. The mor-
phology of the films transferred
onto freshly cleaved mica sub-
Figure 5. AFM image (left) and section analysis profile (right) of a one-layer LB film transferred at 20 mNmꢁ1
strates can be evaluated by
AFM imaging.[44] The image onto freshly cleaved mica.
Chem. Eur. J. 2010, 16, 13398 – 13405
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13401

P. Cea et al.
point where metal-tip contact occurs is the conductance
quantum G0 (G0 =2e2 hꢂ77.4 mS), provide the basis for an
estimation of the gap separation (tip-substrate distance) at a
given current according to Equation (2).
lnðG0 ꢄ Ut=I0Þ
ð2Þ
s ¼
d lnðIÞ=ds
I–V curves for a one-layer TMS-OPE-NH2 LB film using
several set-point parameters (Ut =0.6 V and I0 =0.3, 0.5 and
1.0 nA), which give different initial tip-substrate distances
(1.57, 1.50 and 1.42 nm) according to Equation (2), are
shown in Figure 7a. In agreement with the thickness of the
monolayer, 1.49ꢃ0.04 nm, at 0.5 nA and 0.6 V (s=1.50 nm)
the tip is positioned just above the monolayer. Meanwhile,
for 1.0 nA (s=1.41 nm) the tip penetrates inside the mono-
layer (conductance and rectification increase) and for
0.3 nA (s=1.57 nm) there is a notable gap between the tip
and the LB film (junction conductance decreases).
Figure 6. Cyclic voltammogram (CV) of a one-layer LB film of TMS-
OPE-NH2 deposited on a gold electrode at 20 mNmꢁ1(dotted line). CVs
were recorded by immersing the LB film in aqueous solutions of 1 mm
K3[Fe(CN)6] and 0.1m KCl using a scan rate of 0.05 Vsꢁ1 at 208C. A Agj
AgCljsaturated KCl reference electrode was employed and the counter
electrode was a Pt sheet. The solid line shows the response of a bare gold
electrode in the same solution and conditions.
ute to the current. Therefore, it
is necessary to know, before re-
cording the I–V curves, both
the thickness of the monolayer
and the tip-substrate distance
(s) in order to position the
STM tip in a defined manner
above the monolayer and so
avoid either penetration of the
STM tip into the film or the ex-
istence of a substantial gap be-
tween the STM tip and the
monolayer. The thickness of the
monolayer (1.49ꢃ0.04) nm,
was determined by using the at-
tenuation of the Au4f signal in
the XPS spectra (Figure 4b) as
explained in the Experimental
Section. This thickness gives an
average tilt angle for the mole-
Figure 7. a) I–V curves of a one-layer LB film of TMS-OPE-NH2 transferred onto AuACTHNUTRGNEUNG
(111) at 20 mNmꢁ1 using
several set-point parameters: 0.3 nA (s=1.57 nm) (black dashed line); 0.5 nA (s=1.50 nm) (red solid line);
and 1.0 nA (s=1.42 nm) (black dotted line). An I–V curve constructed from single-molecule conductance
values obtained by using the I(s) method is also shown (blue circles). The error bars represent the standard de-
viation. Ut =0.6 V. b) I–V curve of a one-layer LB film of TMS-OPE-NH2 at 0.5 nA (dashed line) and fitting
according to the Simmons equation, F=0.68 eV, a=0.42 (red solid line).
cules in the LB film of 488 from the surface normal, taking
in account the molecule length (2.23 nm) estimated by mo-
lecular modelling. This value is in good agreement with the
tilt angle obtained from reflection spectroscopy at the air–
water interface (468) at a surface pressure of 20 mNmꢁ1
(Figure 1b). The tip-substrate distance (s) was measured
making use of a careful calibration where the set-point pa-
rameters (I0 ꢀ“set-point current” and Ut ꢀ“tip bias”) can be
converted to an absolute gap separation (s), as follows. I(s)
scans (exponential dependence of current (I) on distance
(s)), which display a monotonic exponential decrease of the
tunnelling current (no wire formation) as the tip is retracted
were recorded at regular intervals during the measurements.
These monotonic exponential decay curves were then plot-
ted as ln(I) versus s. Averaging the slope of the correspond-
ing dln(I)/ds curves, (dln(I)/ds was typically in the order of
7.59ꢃ0.88 nmꢁ1), and assuming that the conductance at the
In addition to results from the conductance studies for the
monolayer covered surfaces, Figure 7a also shows an I–V
curve constructed from single molecule conductance (SMC)
values for TMS-OPE-NH2 obtained by using the I(s)
method at 10 different bias voltage values. The I(s) method
developed by Haiss et al. has been widely used to determine
the single-molecule conductance of several com-
pounds[10,24,46] and description of it can be found in the liter-
ature.[9,24,47] The SMC-curve coincides with the I–V curve ob-
tained for the LB film covered surface recorded at 0.5 nA
setpoint current. These results indicate that with these pa-
rameters the STM tip is located directly above the LB film
and electronically coupled to a single molecule. Under such
conditions the STM tip probes the conductance of a single
TMS-OPE-NH2 molecule embedded in the LB film by using
the trimethylsilane group to form contact with the gold
STM tip. There is similarity between both I–V curves, de-
13402
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13398 – 13405

Metal–Molecule–Metal Junctions
FULL PAPER
spite the different molecular surroundings in the two cases.
In the LB film case the molecule is packed together with
neighbouring TMS-OPE-NH2 molecules, whereas no such
neighbours exist for the SMC determinations. This indicates
that for the LB film intermolecular electron hopping does
not significantly enhance the junction transport characteris-
tics and that the presence of neighbouring p systems does
not greatly influence the STM determined molecular con-
ductance. From this observation it is concluded that the exis-
tence of intermolecular interactions and 2D aggregates (H-
aggregates) in the LB films (as demonstrated by UV/Vis
spectroscopy) does not have a significant impact on the
junction electrical characteristics.
electron. To fit the I–V data in Figure 7b, F and a are the
fit parameters. A good agreement between the data and the
model is obtained with F=0.68 eV, and a=0.42. In spite of
having molecular “asymmetry”, this effective barrier height
is slightly higher than that obtained by Lu et al.[13] for a sym-
metric OPE with amine groups in both ends (Fꢂ0.60 eV).
These collected electrical measurements indicate that the
TMS can be used as an effective anchoring group in metal–
molecule–metal junctions without significant impairment of
transport through the molecular wire, when bench marked
against thiol end-groups. This shows that TMS is an efficient
linker for electron transport through molecular junctions.
The I–V response for the SMC-curve and for the I–V
curve obtained at 0.5 nA is close to ohmic between bias vol-
tages of ꢁ0.6 to +0.6 V giving a molecule conductance
value of approximately 1.2ꢅ10ꢁ5 G0. This value is the same
order of magnitude as the conductance value obtained for
other OPE derivatives using thiols as anchoring groups
(1.8ꢅ10ꢁ5 G0).[48,49] Therefore, it is concluded that both the
amino and trimethylsilane groups are alternative linkers
that provide electronic coupling between the electrodes and
the molecule which is essentially as efficient as that through
thiols. However, outside the ꢁ0.6 to +0.6 voltage range the
response deviates from linearity showing a sigmoidal behav-
iour over the full voltage region. In addition, the SMC-curve
and the I–V curve at 0.5 nA set-point current are nearly
symmetrical despite the asymmetry of the molecule. This
implies that its behaviour is like that of a molecular wire for
which the molecule is simply an amphiphilic electron-donat-
ing wire, as has been previously reported for similar OPE
derivatives,[35,50–52] and does not behave like a molecular
diode with strong rectifying characteristic produced by the
asymmetric molecular junction.
Conclusion
An antisymmetric OPE derivative, with an amine group on
one end and a trimethylsilane group linked to the aromatic
ring through a triple bond at the other, has been synthesised
and assembled into well-packed monolayer films by means
of the Langmuir–Blodgett technique. Langmuir films were
prepared at the air–water interface and characterised by p–
A and DV–A isotherms, which demonstrate that this mole-
cule can form true monolayers at the air–water interface.
Analysis of UV/Vis reflection spectra revealed the forma-
tion of two dimensional H-aggregates and a gradual transi-
tion of the molecules to a more vertical position upon com-
pression. These monomolecular films were transferred un-
disturbed onto solid substrates with a transfer ratio close to
1 with a Au-NH2-OPE-TMS structure as demonstrated by
XPS. Cyclic voltammograms showed the absence (or low
density) of holes or defects in the monolayers of this mole-
cule despite of the absence of alkyl chains commonly used
to stabilise LB films. Electrical characteristics of the LB
films on gold substrates were determined from I–V curves,
with a gold STM tip positioned sufficiently above the mono-
layer as determined from calibration of the tip–substrate dis-
tance and determination of the thickness of the LB film. I–
V curves recorded at different set-point parameters were
compared with those constructed from single-molecule con-
ductance values ob-
The I–V curves characteristics indicate that the mecha-
nism of transport through these metal-molecule-metal junc-
tions is non-resonant tunnelling. A widely applied tunnelling
model that can be used for comparison with the experimen-
tal I–V data is the Simmons model.[53] In this model, the cur-
rent I is given by Equation (3),
ꢄꢀ
ꢁ
ꢂ
ꢀ
ꢁ
ꢃ
ꢀ
ꢁ
ꢂ
ꢀ
ꢁ
ꢃꢅ
ð3Þ
1=2
1=2
1=2
1=2
tained by using the
I(s) technique. The
Ae
4p2ꢀhs2
eV
2
2ð2mÞ
ꢀh
eV
2
eV
2
2ð2mÞ
ꢀh
eV
2
I ¼
F ꢁ
exp ꢁ
a F ꢁ
s ꢁ F þ
exp ꢁ
a F þ
s
coincidence
be-
tween both curves
for which V is the applied potential, A is the contact area of
the molecule with the tip (0.48 nm2 in concordance with the
isotherm shown in the Figure 2 at the surface pressure of
20 mNmꢁ1and since the STM tip is electronically coupled to
a single molecule in the LB films according to Figure 7a), s
is the width of the tunnelling barrier which was assumed to
be the through-bond distance between the end groups in
OPE molecular wire as calculated with a molecular model-
ling program (2.23 nm), F is the effective barrier height of
the tunnelling junction (relative to the Fermi level of the
Au), a is related to the effective mass of the tunnelling elec-
tron and m and e represent the mass and the charge of an
shows that the conductance of a single TMS-OPE-NH2 mol-
ecule embedded in the LB film is probed. Importantly, it is
concluded that TMS is an alternative anchoring group,
which provides effective electronic coupling at metal–mole-
cule contacts. Finally, analysis of the I–V curve showed a
non-resonant tunnelling mechanism with a symmetrical re-
sponse indicative of non-rectifying molecular wire-like char-
acteristics.
Chem. Eur. J. 2010, 16, 13398 – 13405
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13403

P. Cea et al.
Acknowledgements
Experimental Section
G.P., L.M.B., S.M. and P.C. are grateful for financial assistance from the
Ministerio de Ciencia e Innovaciꢂn from Spain and fondos FEDER in
the framework of project CTQ2009–13024. P.J.L holds an EPSRC Lead-
ership Fellowship. R.J.N. thanks EPSRC for funding. G.P. gratefully ac-
knowledges the award of a FPU fellowship from MEC. S. M acknowledg-
es his Juan de la Cierva position from Ministerio de Ciencia e Innovaciꢂn
(Spain) and L.M.B acknowledges her grant from Banco Santander. We
are also grateful to Dr. Jordi Dꢀaz from the University of Barcelona and
Dr. Guillermo Antorrena from Instituto de Nanociencia de Aragꢂn for
their help with the AFM and XPS measurements, respectively.
Film fabrication: 4-[4-{4-(Trimethylsilylethynyl)phenylethynyl}phenyl-
ACHTUNGTRENNUNG
A
100 mm2 housed in a constant temperature (20ꢃ18C) clean room was
used to prepare the films. The surface pressure (p) of the monolayers
was measured by using a Wilhelmy paper plate pressure sensor. Ultra-
pure Milli-pore Milli-Q water (resistivity 18.2 MWcm) was used as sub-
phase. The spreading solutions (1ꢅ10ꢁ5 m) in TMS-OPE-NH2 were pre-
pared in chloroform (HPLC grade, 99.9% purchased from Sigma). To
construct the Langmuir films, the solution was spread using a Hamilton
micro-syringe held very close to the aqueous surface, allowing the surface
pressure to return to a value as close as possible to zero between each
addition. The spreading solvent was allowed to completely evaporate
over a period of at least 15 min before compression of the Langmuir film
at a constant sweeping speed of 0.02 nm2 moleculeꢁ1 minꢁ1. Each com-
pression isotherm was recorded at least three times to ensure the repro-
ducibility of the results so obtained. Surface potential measurements
were carried out using a Kelvin Probe provided by Nanofilm Technologie
GmbH, Gçttingen, Germany. During monolayer compression, p–A and
DV–A isotherms were recorded simultaneously. Meanwhile, a commercial
UV/Vis reflection spectrophotometer, with a light source FiberLight
DTM 6/50 and an absolute wavelength accuracy <0.3 nm and a resolu-
tion (Raylight-criterion)> 3 nm, was used to obtain the reflection spectra
of the Langmuir films upon compression.[39] A commercial mini-Brewster
Angle Microscope (mini-BAM), from Nanofilm Technologie, was em-
ployed for the direct visualisation of the monolayers at the air/water in-
terface.
[1] D. Nilsson, S. Watcharinyaron, M. Eng, L. Li, E. Moons, L. Johan-
son, M. Zharnikov, A. Shaporenko, B. Albinsson, J. Martensson,
Langmuir 2007, 23, 6170–6181.
[2] A. M. Rawlett, T. J. Hopson, I. Amlani, R. Zhang, J. Tresek, L. A.
Nagahara, R. K. Tsui, H. Goronkin, Nanotechnology 2003, 14, 377–
384.
[3] U. H. F. Bunz, Adv. Polym. Sci. 2005, 177, 1–52.
[4] M. A. Reed, C. Zhou, C. J. Muller, T. P. Burgin, J. M. Tour, Science
1997, 278, 252–254.
[5] A. Salomon, D. Cahen, S. Lindsay, J. Tomfohr, C. B. Engelkes, D. D.
Frisbie, Adv. Mater. 2003, 15, 1881–1890.
[6] J. Chen, M. A. Reed, A. M. Rawlett, J. M. Tour, Science 1999, 286,
1550–1552.
[7] E. A. Osorio, K. O’Neill, M. Wegewijs, N. Stuhr-Hansen, J. Paaske,
T. Bjornholm, H. S. J. van der Zant, Nano Lett. 2007, 7, 3336–3342.
[8] B. Kim, J. M. Beebe, C. Olivier, S. Rigaut, D. Touchard, J. G. Kush-
merick, X. Y. Zhu, C. D. Frisbie, J. Phys. Chem. C 2007, 111, 7521–
7526.
[9] W. Haiss, H. van Zalinge, S. J. Higgings, D. Bethell, H. Hçbenreich,
D. J. Schiffrin, R. J. Nichols, J. Am. Chem. Soc. 2003, 125, 15294–
15295.
[10] G. Sedghi, K. Sawada, L. J. Esdaile, M. Hoffmann, H. L. Anderson,
D. Bethell, W. Haiss, S. J. Higgins, R. J. Nichols, J. Am. Chem. Soc.
2008, 130, 8582–8583.
[11] A. M. Moore, B. A. Mantooth, Z. J. Donhauser, F. Maya, D. W.
Price, Y. Yao, J. M. Tour, P. S. Weiss, Nano Lett. 2005, 5, 2292–2297.
[12] K. Liu, X. Wang, F. Wang, ACS Nano 2008, 2, 2315–2323.
[13] Q. Lu, K. Liu, H. Zhang, Z. Du, X. Wang, F. Wang, ACS Nano
2009, 3, 3861–3868.
The films were deposited at a constant surface pressure by the vertical
dipping method with a dipping speed of 0.6 cmminꢁ1. The solid substrates
used to support the LB films were quartz, mica, and gold. UV/Vis spectra
were acquired on a Varian Cary 50 spectrophotometer. Atomic Force Mi-
croscopy (AFM) experiments were performed by means of a Multimode
extended microscope with Nanoscope IIIA electronics from Digital In-
struments, using the tapping mode. The data were collected with a silicon
cantilever provided by Nanoworld, with a force constant of 42 mN and
operating at a resonant frequency of 285 kHz. The images were collected
with a scan rate of 1 Hz, an amplitude set point lower than 1 V, and in
ambient air conditions.
Cyclic voltammetry (CV) experiments were completed by using an elec-
trochemical cell containing three electrodes. The working electrode con-
sisted of either a gold electrode or a gold substrate with the deposited
LB film, the counter electrode was a platinum sheet, and the reference
electrode was Ag jAgCljsaturated KCl.
[14] X. D. Cui, A. Primak, X. Zarate, J. Tornfohr, O. F. Sankey, A. L.
Moore, A. T. Moore, D. Gust, G. Harris, S. M. Lindsay, Science 2001,
294, 571–574.
[15] B. Kim, J. M. Beebe, Y. Jun, X. Y. Zhu, C. D. Frisbie, J. Am. Chem.
Soc. 2006, 128, 4970–4971.
The XPS spectra of LB films on Au substrates were recorded using a
Kratos AXIS ultra DLD spectrometer with a monochromatic AlKa X-ray
source (1486.6 eV) employing a pass energy of 20 eV at photoelectron
take-off angles of 908 with respect to the sample plane. Surface charging
[16] J. M. Beebe, B. Kim, C. D. Frisbie, J. G. Kushmerick, ACS Nano
2008, 2, 827–832.
[17] J. M. Beebe, V. B. Engeelkes, L. L. Miller, C. D. Frisbie, J. Am.
Chem. Soc. 2002, 124, 11268–11269.
[18] V. B. Engelkes, J. M. Beebe, C. D. Frisbie, J. Am. Chem. Soc. 2004,
126, 14287–14296.
[19] S. H. Gyepi-Garbrah, R. Silerova, Phys. Chem. Chem. Phys. 2002, 4,
3436–3442.
[20] F. Chen, X. Li, J. Hihath, Z. Huang, N. J. J. Tao, J. Am. Chem. Soc.
2006, 128, 15874–15881.
[21] G. K. Ramachandran, T. J. Hopson, A. M. Rawlett, L. A. Nagahara,
A. Primak, S. M. Lindsay, Science 2003, 300, 1413–1416.
[22] Z. K. Keane, J. W. Ciszek, J. M. Tour, D. Natelson, Nano Lett. 2006,
6, 1518–1521.
[23] S. Yasuda, S. Yoshida, J. Sasaki, Y. Okutsu, T. Nakamura, A. Tanina-
ka, O. Takeuchi, H. Shigekawa, J. Am. Chem. Soc. 2006, 128, 7746–
7747.
[24] W. Haiss, S. Martin, E. Leary, H. van Zalinge, S. J. Higgins, L. Bouff-
ier, R. J. Nichols, J. Phys. Chem. C 2009, 113, 5823–5833.
[25] C. Reese, M. E. Roberts, S. R. Parkin, Z. Bao, Adv. Mater. 2009, 21,
3678–3681.
was compensated by referencing the adventitious AuACHTUNGTNRE(UNG 4f7/2) peak at
84.0 eV. The thickness of LB films on the gold substrates was estimated
using the attenuation of the Au4f signal from the substrate according to:
ILB film ¼ Isubstrate expðꢁd=l sin qÞ, where d is the film thickness, ILB
and
and Au4f
7/2
film
Isubstrate are the average of the intensities of the Au4f
5/2
peaks attenuated by the LB film and from clean gold, respectively, q is
the photoelectron take-off angle, and l=4.2ꢃ0.1 nm[54] is the effective
attenuation length of the photoelectron.
An Agilent STM running Picoscan 4.19 Software was used for all meas-
urements of molecular conductance. The tip potential is referred to as Ut.
STM tips were freshly prepared for each experiment by etching of a
0.25 mm Au wire (99.99%) in a mixture of HCl (50%) and ethanol
(50%) at +2.4 V. Gold films employed as substrates were purchased
from Arrandee, Schroeer, Germany. These were flame-annealed at ap-
proximately 800–10008C with a Bunsen burner immediately prior to use
to prepare atomically flat AuACHTNUGRTNEUNG
(111) terraces.[55]
13404
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13398 – 13405

Metal–Molecule–Metal Junctions
FULL PAPER
[26] N. Katsonis, A. Marchenko, S. Taillemite, D. Fichou, G. Chouraqui,
C. Aubert, M. M. , Chem. Eur. J. 2003, 9, 2574–2581.
[27] G. Pera, A. Villares, M. C. Lꢂpez, P. Cea, D. P. Lydon, P. J. Low,
Chem. Mater. 2007, 19, 857–864.
[28] A. Villares, D. P. Lydon, L. Porrꢆs, A. Beeby, P. J. Low, P. Cea, F. M.
Royo, J. Phys. Chem. B 2007, 111, 7201–7209.
[29] Z. X. Tang, R. K. Hicks, R. J. Magyar, S. Tretiak, Y. Gao, H. L.
Wang, Langmuir 2006, 22, 8813–8820.
[30] A. Villares, S. Martin, I. Giner, J. Diaz, D. P. Lydon, P. Low, P. Cea,
Soft Matter 2008, 4, 1508–1514.
[41] A. Gil, I. Arꢀstegui, A. Suꢁrez, I. Sꢁndez, D. Mçbius, Langmuir
2002, 18, 8527–8534.
[42] G. Sauerbrey, Z. Phys. 1959, 155, 206–222.
[43] L. Kankate, A. Turchanin, A. Gçlzhꢇuser, Langmuir 2009, 25,
10435–10438.
[44] M. Haro, B. Giner, C. Lafuente, M. C. Lꢂpez, F. M. Royo, P. Cea,
Langmuir 2005, 21, 2796–2803.
[45] M. D. Porter, T. B. Bright, D. L. Allara, C. E. D. Chidsey, J. Am.
Chem. Soc. 1987, 109, 3559–3568.
[46] C. S. Wang, A. S. Batsanov, M. R. Bryce, S. Martin, R. J. Nichols,
S. J. Higgings, V. M. Garcia-Suarez, C. J. Lambert, J. Am. Chem.
Soc. 2009, 131, 15647–15654.
[31] A. Villares, D. P. Lydon, B. J. Robinson, G. Ashwell, F. M. Royo,
P. J. Low, P. Cea, Surf. Sci. 2008, 602, 3683–3687.
[32] Z. J. Donhauser, B. A. Mantooth, K. F. Kelly, L. A. Bumm, J. D.
Monnell, J. J. Stapleton, D. W. Price Jr, A. M. Rawlett, D. L. Allara,
J. M. Tour, P. S. Weiss, Science 2001, 292, 2303–2307.
[33] R. Huber, M. T. Gonzalez, S. Wu, M. Langer, S. Grunder, V. Ho-
rhoiu, M. Mayor, M. R. Bryce, C. S. Wang, R. Jitchati, C. Schonen-
berger, M. Calame, J. Am. Chem. Soc. 2008, 130, 1080–1084.
[34] X. Y. Xiao, L. A. Nagahara, A. M. Rawlett, N. J. J. Tao, J. Am.
Chem. Soc. 2005, 127, 9235–9240.
[47] W. Haiss, R. J. Nichols, H. van Zalinge, S. J. Higgins, D. Bethel, D. J.
Schiffrin, Phys. Chem. Chem. Phys. 2004, 6, 4330–4337.
[48] W. Haiss, C. S. Wang, I. Grace, A. S. Batsanov, D. J. Schiffrin, S. J.
Higgins, M. R. Bryce, C. J. Lambert, R. J. Nichols, Nat. Mater. 2006,
5, 995–1002.
[49] N. Weibel, A. Alfred Błaszczyk, C. von Hꢇnisch, M. Mayor, I. Pobe-
lov, T. Wandlowski, F. Chen, N. Tao, Eur. J. Org. Chem. 2008, 136–
149.
[35] K. Liu, G. Li, X. Wang, F. Wang, J. Phys. Chem. C 2008, 112, 4342–
4349.
[50] G. J. Ashwell, B. Urasinska, W. D. Tyrrell, Phys. Chem. Chem. Phys.
2006, 8, 3314–3319.
[36] A. Villares, G. Pera, S. Martin, R. J. Nichols, D. P. Lydon, L. Apple-
garth, A. Beeby, P. J. Low, P. Cea, Chem. Mater. 2010, 22, 2041–
2049.
[37] O. N. Oliveira, Jr., C. Bonardi, Langmuir 1997, 13, 5920–5924.
[38] A. Beeby, K. Findlay, P. J. Low, T. B. Marder, J. Am. Chem. Soc.
2002, 124, 8280–8284.
[51] A. Villares, D. P. Lydon, P. J. Low, B. J. Robinson, G. J. Ashwell,
F. M. Royo, P. Cea, Chem. Mater. 2008, 20, 258–264.
[52] J. Reichert, R. Ochs, D. Beckmann, H. B. Weber, M. Mayor, H. von
Lohneysen, Phys. Rev. Lett. 2002, 88, 176804.
[53] J. G. Simmons, J. Appl. Phys. 1963, 34, 1793–1803.
[54] C. D. Bain, G. M. Whitesides, J. Phys. Chem. 1989, 93, 1670–1673.
[55] W. Haiss, D. Lackey, J. K. Sass, K. H. Besocke, J. Chem. Phys. 1991,
95, 2193–2196.
[39] P. Cea, S. Martꢀn, A. Villares, D. Mçbius, M. C. Lꢂpez, J. Phys.
Chem. B 2006, 110, 963–970.
[40] X. Song, J. Perlstein, D. G. Whitten, J. Am. Chem. Soc. 1997, 119,
9144–9159.
Received: May 3, 2010
Published online: October 7, 2010
Chem. Eur. J. 2010, 16, 13398 – 13405
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13405