Multiscale charge injection and transport properties in self-assembled monolayers of biphenyl thiols with varying torsion angles

Article abstract of DOI:10.1002/chem.201201858

This article describes the molecular structure-function relationship for a series of biphenylthiol derivatives with varying torsional degree of freedom in their molecular backbone when self-assembled on gold electrodes. These biphenylthiol molecules chemisorbed on Au exhibit different tilt angles with respect to the surface normal and different packing densities. The charge transport through the biphenylthiol self-assembled monolayers (SAMs) showed a characteristic decay trend with the effective monolayer thickness. Based on parallel pathways model the tunneling decay factor β was estimated to be 0.27 A^-1. The hole mobility of poly(3-hexylthiophene)-based thin-film transistors incorporating a biphenylthiol SAM coating the Au source and drain electrodes revealed a dependence on the injection barrier with the highest occupied molecular orbital (HOMO) level of the semiconductor. The possible role of the resistivity of the SAMs on transistor electrodes on the threshold voltage shift is discussed. The control over the chemical structure, electronic properties, and packing order of the SAMs provides a versatile platform to regulate the charge injection in organic electronic devices. Controlled charge injection! Biphenylthiols with different substituents between phenyl groups in the bridge position, once chemisorbed onto gold source-drain electrodes, can be used to vary the charge injection and electrical performances of organic thin-film transistors. By varying the torsion angle between the phenyl rings, one can introduce changes in packing densities and tilt angles of the molecules self-assembled on the electrode that play a direct role in influencing the charge transport through the molecular monolayer (see figure). Copyright

Full text of DOI:10.1002/chem.201201858

FULL PAPER
DOI: 10.1002/chem.201201858
Multiscale Charge Injection and Transport Properties in Self-Assembled
Monolayers of Biphenyl Thiols with Varying Torsion Angles
Appan Merari Masillamani,[a] Nfflria Crivillers,[a] Emanuele Orgiu,[a] Jürgen Rotzler,[b]
David Bossert,[b] Ramakrishnappa Thippeswamy,[d] Michael Zharnikov,*[d]
Marcel Mayor,*[b, c] and Paolo Samorì*[a]
Abstract: This article describes the mo-
lecular structure–function relationship
for a series of biphenylthiol derivatives
with varying torsional degree of free-
dom in their molecular backbone when
self-assembled on gold electrodes.
These biphenylthiol molecules chemi-
sorbed on Au exhibit different tilt
angles with respect to the surface
normal and different packing densities.
The charge transport through the bi-
phenylthiol self-assembled monolayers
(SAMs) showed a characteristic decay
trend with the effective monolayer
thickness. Based on parallel pathways
model the tunneling decay factor b was
estimated to be 0.27 ꢀꢀ1. The hole mo-
bility of poly(3-hexylthiophene)-based
thin-film transistors incorporating a bi-
phenylthiol SAM coating the Au
source and drain electrodes revealed
a dependence on the injection barrier
with the highest occupied molecular or-
bital (HOMO) level of the semicon-
ductor. The possible role of the resis-
tivity of the SAMs on transistor elec-
trodes on the threshold voltage shift is
discussed. The control over the chemi-
cal structure, electronic properties, and
packing order of the SAMs provides
a versatile platform to regulate the
charge injection in organic electronic
devices.
Keywords: biphenylthiols · charge
tunneling · hole injection barrier ·
molecular electronics · self-assem-
bly
Introduction
form highly ordered self-assembled monolayers (SAMs).[3]
These junctions can feature distinct functionalities including
electrical-current rectification,[2c,4] photo-responsive current
switching,[2b,5] wettability, and work function modification.[6]
These tunable properties, which depend on the choice of the
molecule and metal electrode, have been shown to make
SAMs ideal components for optimizing the charge injection
at metal–organic interfaces for application in organic thin-
film transistors (OTFTs). A thorough knowledge on the
transport properties through the SAMs is therefore impor-
tant for the design and further fabrication of the organic
electronic devices incorporating these monolayers. The
charge transport through molecules sandwiched between
two metallic electrodes has been probed by using a variety
of different techniques, such as conducting-probe atomic
force microscopy (CP-AFM),[2b,8] Hg-droplet-based electro-
des,[9] mechanically controlled break junctions,[10] scanning
tunneling microscopy (STM),[2a,11] nanopores,[12] crossed-wire
tunneling junctions,[13] junctions formed by surface-diffusion-
mediated deposition (SDMD),[14] and non-intrusive confor-
mal junctions on the SAM with an eutectic alloy of liquid-
metallic gallium–indium (GaInE) tip as the probe elec-
trode.[15] Each of the above-mentioned techniques has its
own pros and cons, with a variable range of set-up complexi-
ties, junction areas, geometry of contacts, and conditions
necessary for experimentation. The GaInE-based junction is
particularly interesting, since it exploits the peculiar proper-
ty of liquid-metallic gallium–indium to behave like a non-
Newtonian liquid at room temperature. Importantly, it can
be molded into a conical tip that retains its shape under am-
The prediction by Aviram and Ratner that organic mole-
cules could exhibit an electrical-current rectification proper-
ty[1] triggered remarkable research efforts, which have been
blooming with the emergence of molecular electronics. The
nanoscale electronic properties in molecule-based junctions
can be controlled through a proper molecular design.[2]
particular approach for molecular-junction preparation is
the chemisorption of sulfur-exposing molecules on metallic
surfaces like Au, Ag and Pt, which provides a facile route to
A
[a] A. M. Masillamani, Dr. N. Crivillers, Dr. E. Orgiu, Prof. P. Samorꢁ
ISIS & icFRC, Universitꢂ de Strasbourg & CNRS
8 allꢂe Gaspard Monge, 67000 Strasbourg (France)
E-mail: samori@unistra.fr
[b] J. Rotzler, D. Bossert, Prof. M. Mayor
Department of Chemistry, University of Basel
St. Johannsring 19, 4056 Basel (Switzerland)
E-mail: marcel.mayor@unibas.ch
[c] Prof. M. Mayor
Karlsruhe Institute of Technology (KIT)
Institute for Nanotechnology
P.O. Box 3640, 76021 Karlsruhe (Germany)
[d] Dr. R. Thippeswamy, Prof. M. Zharnikov
Angewandte Physikalische Chemie, Universitꢃt Heidelberg
Im Neuenheimer Feld 253, 69120 Heidelberg (Germany)
E-mail: Michael.Zharnikov@urz.uni-heidelberg.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201201858.
Chem. Eur. J. 2012, 00, 0 – 0
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

bient conditions. Significantly, GaInE features an intrinsic
property to form a thin oxide layer (ca. 1–2 nm) when ex-
posed to air; importantly, such a self-bonding oxide (Ga2O3)
sheath enables the bulk of the underlying GaInE to hold its
shape.[15] The ease of forming reliable and reproducible coni-
cal-shaped GaInE tips has been exploited to study the
charge transport of delicate mcm2-sized single-molecule-
layer junctions.[2c,15,16]
In contrast, other types of counter electrodes, like those
based on a Hg drop, needed a protecting alkanethiol layer
coating the counter electrode to improve the yield[17] by mit-
igating formation of junction short-circuits when placing the
drop in contact with the SAM on the metallic substrate.[9b,18]
Similarly, a thin conducting film of poly(3,4-ethylenedi-
oxythiophene)–poly(styrenesulfonate) (PEDOT:PSS) poly-
mer[5c,19] or graphene[20] has been employed to protect the
SAM prior to deposition of top Au electrodes in large-area
molecular junctions.
between the molecular conductance with the square of the
cosine of the torsion angle.[25]
The different mono-thiolated biphenyl derivatives de-
signed, synthesized and exploited in the present study are 1)
9H-fluorene-2-thiol, denoted here as 5mBPT due to the
fluorene group bridging the two individual phenyl rings; 2)
1,2-bis(9,10-dihydrophenanthren-2-yl)disulfane denoted as
6mBPT; 3) (1,1’-biphenyl)-4-thiol denoted as BPT; and 4)
2,2’-dimethyl(1,1’-biphenyl)-4-thiol denoted as tBPT (i.e.,
twisted BPT).
Experimental Section
Thiol solution preparation: 1 mm solutions in CHCl3 were prepared and
stored in an inert N2 filled atmosphere of a glove box (Jacomex).
SAM fabrication procedure: For X-ray spectroscopy and charge-trans-
port measurements with a GaInE setup, the SAMs were prepared on
Traditionally n-alkanethiols have been widely studie-
d,[8a,9b,11,12,16c,19,21] and successfully employed to change the
wettability and work function properties of gold contacts,
but their scope as functional components for molecular elec-
tronic devices is limited. The better conducting aromatic
compounds offer a far more versatile platform to modulate
their electronic properties by altering the chemical structure
of the molecule[16a,22] or by influence of external stimuli like
thermal energy or light.[2b,16b,18b]
Here we have designed and synthesized a new library of
monothiolated biphenyl molecules featuring different tor-
sion angles between the two phenyl subunits along the mo-
lecular backbone. Although there have been studies on the
single-molecular conductance of biphenyl units thiolated in
both a and w positions,[2a,23] the charge transport through bi-
phenyl monothiol SAMs with varying torsion in the long
molecular axis have not yet been explored on a macroscopic
commercial goldACTHUNGETRNNU(G 111) evaporated on mica substrates (Georg Albert
PVD, Germany) with an Au thickness of 300 nm. The substrates were im-
mersed and incubated in the corresponding 1 mm biphenylthiol solution
under an N2 atmosphere for 24 h to obtain the desired monolayers. For
work function and contact-angle experiments, Au with a thickness of
50 nm was evaporated onto glass slides with an adhesion Cr layer of
10 nm at a base pressure of about 10ꢀ6 mbar in a Plassys MEB 300 ther-
mal evaporator and then immediately immersed in the respective thiol
solution for 24 h under inert conditions inside a glovebox. For all the ex-
periments, once removed from the thiol solution, the substrates were
rinsed with copious amounts of CHCl3 in ambient conditions. All the sub-
strates were treated with UV/Ozone prior to immersion in the respective
thiol solutions for SAM formation.
X-ray spectroscopy measurements: The structural characterization of the
SAMs was performed by means of synchrotron-based high-resolution X-
ray photoelectron spectroscopy (HRXPS) and angle-resolved near-edge
X-ray absorption fine structure (NEXAFS) spectroscopy. The HRXPS
and NEXAFS spectroscopy experiments were performed at the bending
magnet HE-SGM beamline at the synchrotron storage ring BESSY II in
Berlin (Germany). The measurements were carried out under room tem-
perature and ultra-high vacuum (UHV) conditions with a base pressure
of at least 1.5ꢅ10ꢀ9 mbar or higher. The time for spectral acquisition was
carefully selected such that no noticeable sample damage by the primary
X-rays was observed during the measurements.[26] The spectra were re-
corded in normal emission geometry. C 1s, S 2p, and Au 4f HRXPS spec-
tra were acquired at photon energies (PEs) of 350 and 580 eV. The O 1s
spectral range was monitored and the binding energy (BE) scale was ref-
erenced to the position of the Au 4f7/2 emission line of an Au substrate
with a SAM of dodecanethiol at an energy of 83.95 eV, which is given by
the latest ISO standard.[27] The energy resolution was approximately
0.3 eV at a PE of 350 eV. The resulting HRXPS spectra was fitted by
symmetric Voigt functions of either a Shirley-type or linear background;
the fits were performed self-consistently (i.e., the same peak parameters
were used in identical regions of the spectrum), which resulted in the
spectrum accuracy for the BE at FWHM values in the range of 0.05 eV.
In the case of the S 2p3/2,1/2 doublets, we used a pair of peaks with the
same FWHM, a branching ratio of 2 (2p3/2/2p1/2), and a spin-orbit splitting
verified by the fit of about 1.18 eV.[28]
scale. On the one hand, once chemisorbed on AuACTHNUTRGNEUG(N 111), we
investigated the charge transport through the monothiol bi-
phenyl SAMs embedded in a two-terminal junction based
on a top GaInE counter electrode. On the other hand, we
functionalized the source and drain electrodes of OTFTs
with monothiol biphenyl SAMs and exploited them to tune
the charge injection at the metal–organic interface of the de-
vices.
The torsion angle between individual aromatic rings in
doubly thiolated biphenyl derivatives has been shown, in
a single-molecule break junction based on STM, to govern
the electrical conductivity across the single-molecule
scale.[2a,23] Both theoretical[22c] and experimental re-
sults[2a,23,24] provided evidence for a correlation between the
cosine square of the torsion angle and the single-molecule
conductance. The increasing torsion angle is accompanied
by a decrease in the extent of the geometrical overlap of pz
orbitals; as a result the corresponding single-molecule con-
ductance decreases, since the electron-transfer rate changes
as a function of the square of pz orbital overlap. Moreover if
the contribution due to tunneling transport through the s or-
bitals is neglected, the theory predicts a linear relationship
The NEXAFS spectra were acquired at C K-edge in partial electron
yield mode with a retarding voltage of ꢀ150 V and linearly polarized
light with a polarization factor of about 91%. To gather details of the ori-
entational order of the SAMs the angle of the incident light was varied
from 90 to 208 in steps of 10–208, which corresponds to progression of
the E vector from the surface normal to the plane close to the sample
surface.[29] The energy resolution of the spectra was about 0.3 eV. The
raw NEXAFS spectrum of each sample was normalized by dividing by
the incident photon flux for a clean gold reference sample[30] and the
&
2
&
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Molecular Electronics
FULL PAPER
energy scale was gauged to the most pronounced p* resonance of highly
deposited on the SiO2 layer, serving as source-drain electrodes, were
used to fabricate the thin-film transistor. The channel length (L) and
width (W) for the TFTs that were characterized amounted to 10 and
10000 mm, respectively. To complete the OTFT device a poly(3-hexylthio-
phene) (P3HT) solution (120 mL of 2 mgmLꢀ1 in toluene) was spin-
coated on the substrates. After spin-coating of P3HT, the solvent was al-
lowed to evaporate for about 48 h prior to electrical characterization.
This characterization was performed by means of a dual channel source
meter (Keithley 2636 A SMU) in the dark. Triaxial cables were employed
to minimize signal losses. The complete fabrication and electrical charac-
terization were performed inside a N2-filled glovebox (Jacomex) with O2
levels <10 ppm, with the exception of the rinsing step after SAM forma-
tion to remove physisorbed material on the electrodes, which was per-
formed in ambient conditions.
oriented pyrolytic graphite at 285.38 eV.[31]
Charge transport in SAMs: To estimate the rate of charge transport
through the SAMs on Au a two terminal setup composed of a liquid-me-
tallic alloy GaInE was employed as the top-probe electrode. An airtight
digital syringe held an adequate reservoir of GaInE, which could be ma-
nipulated to dispense the liquid metal in a controlled fashion, and was
then molded to a conical shaped tip (see Figure S5 Supporting Informa-
tion). The junction, comprising the SAM sandwiched between Au and
GaInE, could be formed very accurately by means of a servo-controlled
xyz stage through careful approach of the GaInE tip. This was accom-
plished by raising the stage supporting the functionalized substrate to
gently contact the apex of the hanging GaInE tip (monitored in real-time
by using a CCD camera). To ensure a greater control during approach in
the z-axis direction (vertical) and higher versatility of movement in the
x–y axes direction (lateral), we kept the position of the syringe holding
the GaInE tip fixed, thereby only moving the stage supporting the
sample. The necessary potentials across the junctions were provided by
an external supply unit (Keithley 2635 SMU, Keithley instruments Inc)
interfaced through a triaxial cable to minimize signal losses. In particular,
a DC bias was applied to the GaInE tip, while holding the Au electrode
in electrical ground. Current versus potential (I–V) traces were recorded
by applying potential sweep across the junction from ꢀ0.5!+0.5 V (uni-
directional trace), starting from ꢀ0.5 V proceeding to +0.5 V and ending
at ꢀ0.5 V for the case of a bidirectional trace (see Supporting Informa-
tion) in 20 mV steps. The measured I values were translated into a corre-
sponding current density (J) by normalizing to the junction contact area.
The area in contact with the GaInE tip was estimated from the image
captured by using a USB camera and calibrating that to a sample of
known thickness. The procedure for contact area estimation was per-
formed with the following steps. A cross-sectional side view image of
a substrate with standard thickness (IPMS Fraunhofer) was captured and
the corresponding number in pixels was estimated for this thickness.
From this the distance value for an individual pixel was extracted, as
a way to calibrate our optical system. Images were captured for each
junction formed and then the total
Morphological characterization: The morphology of the semiconductor
film on the Au electrodes of the OTFT device configuration was mapped
by atomic force microscopy (AFM) from Digital Instruments Dimension
3100 (Bruker Instruments Inc, USA) operating in the intermittent con-
tact mode.
Ionization energy of P3HT: The ionization energy of a drop-casted
P3HT film was recorded by ambient ultraviolet photoelectron spectrosco-
py by using a Riken AC2 spectrophotometer. An estimate of the ioniza-
tion energy was attained from relation of square root of the photoelec-
tron yield for the incident PE.
Results
Synthesis: Several different strategies can be envisaged to
introduce a thiol moiety on aromatic systems. The maybe
most straightforward ones are displayed in Scheme 1. Nucle-
ophilic aromatic substitution of aryl halides with a strong
number of pixels was estimated in
the region of tip contact by using Im-
ageJ 1.42q (an open source software).
The number of pixels was then re-
translated into distance in mm, corre-
sponding to the diameter for the
junctions when the GaInE tip estab-
lished contact with the SAM. This di-
ameter was then used to estimate the
contact area by using a standard for-
mula for the area of a circle. It has
previously been shown that the actual
area of the GaInE tip in contact with
Scheme 1. Retrosynthetic strategies to introduce the sulfur moiety to a biphenyl core.
an ITO surface is roughly 25% of the
actual area,[4] this is probably due to
the formation of micromenisci[32] at the sides of the liquid-metallic GaInE
tip apex. Hence to arrive at an accurate consensus of the current density
for a given applied bias in the sweep voltage range, we measured a large
number of I–V traces, in “n” junctions.
nucleophilic thiol is a valuable method if the resulting thio-
phenol has to be obtained in a protected form (method A).
A second strategy can be the introduction of the thiol by re-
acting a lithiated aryl with sulfur, forming the sulfide, fol-
lowed by trapping with a protection group or protonation
(method B). Method C can be an alternative strategy if the
syntheses of aryl halide precursors are troublesome, for ex-
ample, because of low selectivity. The precursor for the final
Newman–Kwart rearrangement is a phenol for which a varie-
ty of different synthetic routes are known. Furthermore the
polarity of the phenol often improves the purification and
isolation procedures. A major disadvantage of method C is
the increased number of synthetic transformations. The
monomethoxy biphenyl can also be converted to the thiol
Electrode wettability tests: The aqueous contact angle at the SAM-ambi-
ent interface was measured by means of a Krꢆss DSA 100 goniometer.
The contact angle and the derived work of adhesion provided qualitative
information on the surface wettability of electrodes incorporated in
OTFTs.
Work function measurements: The work function of the modified Au
substrates was measured by means of macroscopic Kelvin probe (Kelvin
Probe Technology Ltd). The contact potential difference method, using
a commercial Au reference tip was employed to record the work function
of the Au and Au–SAM surface.
OTFTs fabrication and electrical characterization: 1.5ꢅ1.5 cm2 Si/SiO2
substrates (IPMS Fraunhofer) composed of thermally grown SiO2 layer
on n++ Si, serving as the gate electrode, and pre-patterned Au electrodes
ꢀ
by C S hetero-cross-coupling protocols (method D).
Chem. Eur. J. 2012, 00, 0 – 0
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

M. Zharnikov, M. Mayor, P. Samorꢁ et al.
(1,1’-Biphenyl)-4-thiol (BPT) was synthesized according
to synthetic method B. The commercially available 4-bromo-
biphenyl was treated with tert-butyl lithium at ꢀ788C for
1 h. Then ground elemental sulfur was added to the lithiated
biphenyl in small portions. The final target compound BPT
was obtained after quenching of the sulfide with hydro-
chloric acid and purification by basic extraction followed by
sublimation in 52% (Scheme 2 top).
methylene protons, due to the increased acidity of this
doubly stabilized benzylic position. To prevent side product
formation, commercially available 2-bromo-9H-fluorene was
treated with sodium thiomethoxide in 1,3-dimethylimidazoli-
dinone (DMI) at 1208C. Nucleophilic aromatic substitution
yields the methyl-protected thiophenol, which was depro-
tected in situ by the excess of sodium thiomethoxide. After
quenching with hydrochloric acid, basic extraction afforded
target compound 5mBPTin
high purity (83%).
The
1,2-bis(9,10-dihydro-
phenanthren-2-yl)disulfane
(6mBPT) was synthesized by
use of a Newman–Kwart rear-
rangement of 9,10-dihydrophen-
anthren-2-ol.[61] Attempts to
monobrominate 9,10-dihydro-
phenanthrene at the 4-position
were not successful. Even
though the formation of the de-
sired starting material was ob-
1
served by H NMR spectrosco-
py, it was impossible to isolate
2-bromo-9,10-dihydrophenan-
threne from the crude, due to
the lack of polarity of all
formed compounds. To avoid
the difficulties in regioselectivi-
ty and isolation, in a first step
Scheme 2. Synthesis of 4-thiobiphenyls with varying torsion angles (BPT, tBPT, and 5mBPT). a) 1.) tBuLi,
THF, ꢀ788C, 1 h, then RT, 2.) S8, RT, 3 h, 3.) aq. HCl, RT, 30 min; b) sodium 2-methylpropane-2-thiolate,
DMF, reflux 2 h; c) o-tolylboronic acid, [PdACTHUNRTGNEUN(G PPh3)4], Cs2CO3, toluene/EtOH 2:1, 958C, 24 h; d) BBr3 (1m in di-
chloromethane), 08C, 2.5 h, then MeOH; e) sodium thiomethoxide, DMI, 1208C, 18 h, then aq. HCl.
2,2’-Dimethyl(1,1’-biphenyl)-4-thiol (tBPT) was synthe-
sized in a three-step sequence starting from 1-bromo-4-
fluoro-2-methylbenzene (Scheme 2, middle). The thiol
moiety was introduced at the first step of the synthesis, since
nucleophilic substitution of 2,2’-dimethyl-4-fluorobiphenyl
with sodium tert-butylthiolate was not successful. Fluoroben-
zene was transformed to (4-bromo-3-methylphenyl)-tert-bu-
tylsulfane by a nucleophilic aromatic substitution using
9,10-dihydrophenanthrene was acetylated in the 2-position
by using a Friedel–Crafts acylation protocol[5c] in 37% yield.
A sequence of Baeyer–Villiger oxidation followed by ester
cleavage afforded 9,10-dihydrophenanthren-2-ol in 77%
over both steps (Scheme 3). The O-thiocarbamate necessary
for the envisaged Newman–Kwart rearrangement was then
synthesized by reacting the dihydrophenthrene-2-ol with di-
methylcarbamoylchloride and sodium hydride as a base. The
thermally activated Newman–Kwart rearrangement, in
which an O-thiocarbamate rearranges to a S-thiocarbamate,
was carried out at 2608C in diphenyl ether. After successful
rearrangement the solvent was removed by column chroma-
sodium tert-butylthiolate as
a nucleophile. Subsequent
Suzuki cross-coupling with o-tolylboronic acid by using tet-
rakis(triphenylphosphine)palladium as a catalyst and cesium
carbonate as a base in a solvent mixture of toluene/EtOH
2:1 afforded the tert-butyl-pro-
tected biphenylthiol. Deprotec-
tion by boron tribromide yield-
ed after quenching with metha-
nol the desired target com-
pound tBPT in 23% yield over
three steps. Biphenylthiol tBPT
was further purified by Kugel-
rohr distillation.
For the synthesis of the fluo-
rene
derivative
(5mBPT)
method A was used (Scheme 2
bottom). In contrast to the syn-
thesis of BPT, the presence of
a strong base readily leads to
Scheme 3. Synthesis of 1,2-bis(9,10-dihydrophenanthren-2-yl)disulfane (6mBPT). a) Acetyl chloride, AlCl3,
dichloromethane, 08C, 1 h, then RT, 2 h; b) 1.) m-CPBA, dichloromethane, 08C, 3 h, then RT, 95 h, 2.) conc.
aq. HCl, MeOH, RT, 18 h; c) dimethylcarbamoyl chloride, NaH, DMF, 808C, 2 h, then RT, 15 h; d) Ph2O,
deprotonation of the bridging 2608C, 2 h; e) KOH, MeOH, 808C, 3 h.
&
4
&
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Molecular Electronics
FULL PAPER
tography and the obtained S-thiocarbamate cleaved in
strong basic medium to yield the free thiol 6mBPT in 98%
yield. The free thiol was not stable towards formation of di-
sulfides even when stored at ꢀ188C under argon atmos-
phere.
All desired compounds were further purified by vacuum
sublimation (solids) or Kugelrohr distillation prior to the
physical chemical investigations. The target compounds
were characterized by NMR spectroscopy, mass spectrome-
try, GC-MS, and elemental analysis. The purity of tBPT
after storage over two months under argon atmosphere at
08C was confirmed by GC-MS.
ferently (but strongly) bound SAM constituents, while the
latter to the unbound or weakly bound molecules that are
either caught in the hydrocarbon matrix of the SAM or
physisorbed at the SAM/ambience interface;[35] the content
of all these species is however quite low. Further, the effec-
tive thicknesses of the SAMs were calculated according to
standard procedure by using the intensity ratios of the C 1s
and the Au 4f signals and the standard attenuation lengths
for the Au 4f and C 1s emissions at the given kinetic ener-
gies.[34,36] In addition to the above thickness values, the pack-
ing density in the target SAMs were estimated on the basis
of the S 2p/Au 4f intensity ratio, using a SAM with a known
packing density, namely, dodecanethiol on AuACHTNUTRGNE(UNG 111) in the
Structural characterization of the SAMs: The structural
analysis on the surface of the AuACTHNUGRTNEUNG(111) with the different bi-
phenylthiol SAMs chemisorbed on its surface was per-
formed by means of HRXPS complemented by NEXAFS
spectroscopy to cast light onto the electronic structure and
tilt angle of the adsorbed molecular layers.
given case, as a reference.[37] The derived values of the effec-
tive thickness and packing density of the target SAMs are
displayed in Table 1. A smaller value of the thickness of the
SAM in comparison to the estimated length of the BPT
molecule (ca. 12 ꢀ) in the gas phase suggests that the SAM
constituents have some degree of inclination. The packing
density of the target SAMs exhibits the following trend
BPT>6mBPT>5mBPT>tBPT.
HRXPS on biphenylthiol SAMs: The C 1s and S 2p HRXPS
spectra of all the SAMs used in this study are shown in
Figure 1. The spectra were gathered at wave energies of 350
and 580 eV for the C 1 s and at 350 eV for the S 2p emis-
Table 1. Effective SAM thickness in ꢀ along with packing density of the
molecules in for the different BPT SAMs in comparison to the reference
of dodecanethiol (C12) SAM.
Effective
thickness [ꢀ]
I
(S 2p)/I
E
)
Packing density
ꢅ1014 [molcmꢀ2
]
C12
15
4.63
3.67
3.68
4.25
2.91
5mBPT
6mBPT
BPT
11.68
9.08
11.45
8.13
0.06867
0.068842
0.079533
0.054446
tBPT
NEXAFS measurements: Figure 2 shows the C K-edge
NEXAFS spectra of the target SAMs. The spectra in the
left-hand panel were acquired at the so-called magic X-ray
incidence angle (558), which leads to spectra that are inde-
pendent of the molecular orientation, representative of the
electronic structure of the unoccupied molecular orbitals.[29]
The curves in the right-hand panel are the differences be-
Figure 1. C 1s (left and middle panels) and S 2p (right panel) HRXPS
spectra of the target SAMs acquired at photon energies of 350 eV (left
and right panels) and 580 eV (middle panel). The gray lines are guide for
the eye and are placed at a BE of 284.2 eV.
sion. In the C 1s spectra a single emission at 284.2--284.3 eV
is observed, which is attributed to the aromatic backbone of
the SAM constituents. A characteristic emission in a similar
binding energy (BE) range of 284.2–284.3 eV has previously
been reported for biphenyl-substituted dithiols (BPDTs).[33]
The S 2p spectra of the SAMs are dominated by the S
2p3/2,1/2 doublet at 162.0 eV (S 2p3/2), corresponding to a thio-
late-type bonding of the SAM constituents to the sub-
strate.[33,34] This doublet is the only feature in the spectra of
6mBPT and 5mBPT SAMs, but is accompanied by an addi-
tional weak doublet at a BE of about 161.0 eV (S 2p3/2) in
the spectra of BPT and tBPT monolayers and by a further
weak doublet at a BE of 163.4–163.5 eV in the spectrum of
tBPT SAMs. The former doublet can be assigned to the dif-
Figure 2. C K-edge NEXAFS spectra of the target SAMs acquired at an
incidence angle of 558 (left panel) and the difference of the spectra ac-
quired at X-ray incidence angles of 908 and 208 (right panel). Horizontal
dashed lines in the right panel correspond to zero.
Chem. Eur. J. 2012, 00, 0 – 0
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

M. Zharnikov, M. Mayor, P. Samorꢁ et al.
tween the spectra acquired at X-ray incidence angles of 90
and 208; such curves are representative of the molecular ori-
entation.[29] The spectra are dominated by the intense p1*
resonance of the phenyl rings at about 285.1 eV accompa-
nied by the respective p2* resonance at 288.8 eV, the R*/C-
S* resonance at about 287.8 eV, and several broad s* reso-
nances at higher photon energies (the assignments were per-
formed in accordance with literature data).[28,30,33,34,38] Note
that the pattern of the absorption resonances for the SAMs
used in this study is similar to that of benzene, as can be ex-
pected.[36,39] In the case of the BPT SAM, this pattern can
be attributed to the localization of the excited molecular or-
bitals at a particular aromatic ring.[39] In accordance with the
literature data,[33] the aliphatic bridging group in 5m and
6mBPT does not change this pattern significantly, which
means that the conjugation between the aromatic rings is
not affected. Note that this correlates with the spacing be-
tween the rings in these compounds, which exceeds a critical
value of 1.477 ꢀ.[40] On the basis of the intensity of the p1*
resonance, the quality of the target SAMs follows the trend
BPT>5mBPT>6mBPT>tBPT, in agreement with the
HRXPS results (see the previous section). In addition, the
p1* resonance for the tBPT SAMs is somewhat broader
than for the other monolayers of this study, which is a further
evidence for the higher degree of heterogeneity in this par-
ticular film.
In addition to the above analysis of the 558 spectra, infor-
mation on the molecular orientation in the target SAMs
could be obtained on the basis of the entire set of the
NEXAFS data. In particular, the average tilt angle of the ar-
omatic chains in these films can be estimated by a quantita-
tive analysis of the angular dependence of the NEXAFS res-
onance intensities.[29] For this analysis, the p1* resonance was
selected as the most intense and distinct resonance in the
absorption spectra. In addition, the intensities of this reso-
nance could be derived directly from the NEXAFS spectra,
without a sometimes ambiguous fitting procedure. Once
these intensities I are known, the average tilt angle a of the
p1* orbitals with respect to the surface normal can be de-
rived from a standard expression for a vector-type orbital
[Eq. (1)],[29] in which A is a constant, a is the average tilt
angle of the molecular orbital, P is the polarization factor
and q is the incidence angle of the incoming X-rays.
This expression allows estimation of the tilt angle of the
molecular axis as soon as some assumptions about the value
of the twist angle are made. In the case of the SAMs of this
study it is reasonable to assume a herringbone arrangement,
which is typical of aromatic bulk systems and of aromatic
SAMs.[42] This means two different spatial orientations of
the aromatic moieties occurring with oppositely signed (i.e.,
reverse) twist angles #1 =ꢀ#2 and the similar tilt angles f1 =
f2. The different signs of the twist angle are not a problem,
since cos# is an even function. The most reasonable assump-
tion for the absolute value of the twist angle is 328, which is
the value observed in the biphenyl bulk materials[42a] and
well applicable to aromatic SAMs.[41,43] Table 2 shows the es-
Table 2. Tilt angle and parameters estimated from the NEXAFS data.
qt [8][a]
a [8][b]
q [8][c]
f [8][d]
5mBPT
6mBPT
BPT
0[44]
20[45]
69.0
66.9
71.5
62.4
32.0
32.0
32.0
32.0
25.0
27.6
22.0
33.0
0[42a]
tBPT
79.7[33]
[a] Solid-state torsion angle. [b] Average tilt angle of p1* molecular orbi-
tal. [c] Assumed twist angle. [d] Average tilt angle all in degrees.
timated average tilt angles of p1* orbitals (a), the tilt angles
of molecules (f), and the assumed twist angle (#). Note that
we assume a coplanar arrangement of individual rings
within the biphenyl backbone in the BPT SAMs, which is
a reasonable assumption considering that this arrangement
is typical for densely-packed romatic compounds in which
the dihedral rotation is lifted due to the intermolecular in-
teraction.[42a] The backbone of 5mBPT SAMs has a planar
conformation,[44] while the torsion angle for the backbone of
the 6mBPT SAM is about 16.88 (see Table 3 and refer-
ence [45]). The torsion angle for the backbone of the tBPT
SAM is close to 808 (see Table 3 and reference [33]), so that
the assumed twist angle can only be considered as an aver-
age value over the both rings. The accuracy of the derived
tilt angles is about ꢂ38, which is partly related to the uncer-
tainty of the twist angles. The derived tilt angles represent
average values over macroscopic area of the SAM and thus
can be taken as a fingerprint of the orientational order.
Indeed, a totally disordered molecular film exhibits no
linear dichroism and can be associated with an average tilt
angle of a vector orbital of 54.78, since IACTHNUTRGENU(GN a,q) does not
Iða; qÞ ¼
ꢂ
ꢀ
ꢁ
ꢃ
depend on q at a=54.78 [see Eq. (1)]. A deviation from this
value, as seen for the SAMs studied here, means that a cer-
tain degree of orientational order exists, with a higher
degree for a higher deviation. The highest orientational
order is observed for the BPT and 5mBPT SAMs, in ac-
cordance with the planar conformation of 5mBPT[44] and
the assumed planar conformation of BPT;[40,42a] such a con-
formation should favor high orientational order and dense
molecular packing. Even slight deviation from the planar
conformation, as is the case for 6mBPT,[45] results in
a lower packing density (see previous section) and some de-
terioration of the orientational order (see Table 2). Finally,
1
1
2
1
2
A P ꢁ 1 þ ð3 cos2 q ꢀ 1Þð3 cos2 a ꢀ 1Þ þ ð1 ꢀ PÞ sin2 a
3
ð1Þ
The tilt angle a of the p1* orbitals is directly related to
the tilt angle f of the molecular axis with respect to the sur-
face normal and to the average twist angle # of the aromatic
rings with respect to the plane spanned by the surface
normal and the molecular axis [Eq. (2)].[41]
cos a ¼ cos # sin f
ð2Þ
&
6
&
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Molecular Electronics
FULL PAPER
Table 3. Molecular structure and charge transport properties of biphenylthiol SAMs.
[c]
[d]
[e]
Molecular
structure[a]
Torsion
njn
Jobs
Jsc
Yield
[%][f]
(mlogjJjꢂs)
[Acmꢀ2
mJ ꢅ10ꢀ3
[g]
[h]
angle [8][b]
G
]
ACHTUNGTRENNUNG ]
[Acmꢀ2
5mBPT
1.1
24
8696
6
99.93
(ꢀ2.81ꢂ0.62)
1.55
6mBPT
BPT
16.8
36.4
30
20
3292
3206
18
3
99.45
99.90
(ꢀ2.30ꢂ0.35)
(ꢀ2.71ꢂ0.10)
5.01
1.94
tBPT
79.7
34
6487
6
99.91
(ꢀ1.38ꢂ0.58)
41.69
[a] Molecular structure of different BPT derivatives used. [b] Torsion angle between the planes of the phenyl rings from X-ray structure adapted from
reference [23]. [c] Number of junctions formed. [d] Total number of J observations for each SAM. [e] Number of I–V traces that resulted in a short (i.e.,
junctions in which I reached 10ꢀ3 A). [f] Yield gives an estimate of work junctions, that is, the ratio of the difference between the total and shorting J(V)
traces to the total J observations. [g] Mean of log|J| from the log-normal distribution of J. [h] Mean value of J.
a distinct nonplanar conformation adopted by tBPT causes
even lower packing density (see previous section) and even
less orientational order (see Table 2). Thus, the results from
the HRXPS and NEXAFS seem to corroborate the fact that
the quality and packing density of the SAMs is the highest
for the BPT SAM, slightly worse for the 5mBPT SAMs,
distinctly worse for the 6mBPT SAMs, and the worst for
the tBPT SAMs, in full correlation to the torsion angle in
the biphenyl backbone. The slightly better quality of the
BPT SAMs as compared to the 5mBPT monolayers is pre-
sumably related to the presence of the side bridge in the last
case, which can be associated with additional steric de-
mands.
actual range of J distribution across each of the SAMs. The
oxide skin covering the GaInE bulk was reported to have an
electrical resistivity of 0.4 MWcm, corresponding to a J of
5.5 Acmꢀ2 at 0.2 V,[4] which is much lower than the resistivi-
ty of any SAM in this study. The above current density is
nearly three orders of magnitude greater than the analogous
value for the tBPT SAM (J=2.1ꢅ10ꢀ2 Acmꢀ2 at 0.2 V),
which is the most conductive sample in this study. This
would imply that the corresponding resistance of the least
resistive tBPT SAM is about three orders of magnitude
greater than the Ga2O3 layer. Hence, the resistivity contribu-
tion of the oxide skin to the overall J measured for the junc-
tions can be reasonably neglected, since the total resistance
is the direct sum of the two individual resistive components
of the oxide in series with that of the SAM. Figure 3a shows
the variation of the mean of log|J| with the applied bias
across the junctions for all SAMs of this study. The charge
transport through these monolayers clearly shows a trend
that would be expected in the case of non-resonant tunnel-
ing.[47] The distribution of J through the SAM follows a log-
normal tendency this is attributed to the exponential de-
pendence of charge tunneling over the distance d.[9b,16c,48]
Figure 3b displays the log|J| distributions for the 5mBPT,
6mBPT, BPT and tBPT SAMs. The log-normal histograms
for each of the monolayers were populated by taking the
log|J| values at a bias voltage of ꢀ0.4 V. From the peak of
the Gaussian fitting of the histograms, it was possible to esti-
mate the mlog|J| value, which offers an estimate of highest
likelihood occurrence of J for a particular SAM. All histo-
gram plotting and corresponding data analysis were done
using the ggplot2[49] package of R.[50] Although in a chemisor-
bed SAM the molecules are believed to be arranged in well-
ordered (i.e., similar to a 2D crystal pattern) manner, it is
likely that there can be structural defects which may arise
from several factors, including the presence of physisorbed
adlayers, trace impurities on the electrode, and defects in-
duced by the substrate surface profile among others. As
these features influence the charge transport across the
SAMs, it is imperative to acquire a large number data to
Electrical characterization: charge transport and OTFTs
Charge transport through biphenylthiol SAMs: To gain in-
sight into the charge transport across the target SAMs sup-
ported on Au, we performed I–V measurements for two-ter-
minal junctions by using a GaInE tip as the top counter elec-
trode. We describe the junctions formed following the
E
ꢀ
ꢀ
naming convention Au SAM/Ga2O3 j jGaIn in which “ ”
indicates a covalently bonded contact, “/” denotes the physi-
cally formed contact between the surface of the SAM and
the Ga2O3 layer, and “j j” denotes the interface between
Ga2O3 and GaInE bulk.
Tunneling is known to be the governing charge-transport
mechanism through p-conjugated molecules shorter than
about 3 nm.[46] The current density (J) due to tunneling typi-
cally follows an exponential decay with molecular length ac-
cording to the simplified approximation of Simmons model,
J=J0 eꢀbd in which d is the molecular length (in ꢀ; in this
present work d is considered to be the SAM effective thick-
ness) and b (in ꢀꢀ1) is the tunneling decay constant. To
gather an adequate amount statistical data for the range of
charge tunneling through the target SAMs, we formed a min-
imum of at least 20 junctions for each of these monolayers
and analyzed the resulting data without selecting or elimi-
nating, any of the recorded data points to arrive at the
Chem. Eur. J. 2012, 00, 0 – 0
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

M. Zharnikov, M. Mayor, P. Samorꢁ et al.
the more narrow distribution of the log|J| in SAMs of BPT
and 6mBPT, in which the molecules exhibit more order and
a greater packing density on the electrode surface. The
thickness of the SAMs depends on the tilt angle with respect
to the surface normal, that is, the greater the tilt angle the
lower the effective thickness that corresponds to the separa-
tion between electrodes. From the plot of log|J| versus d
(i.e., effective monolayer thickness) the parameters J0 and
b can be extracted on the basis of the y-intercept and slope
of the linear fit, respectively. We observed that the SAM
with a greater tilt angle, that is, tBPT–SAM, was more con-
ductive than that with a lower tilt angle, that is, 5mBPT–
SAM, by a factor of about 27 in the linear scale (ca. 2 in
log|J| magnitude). In oligo-aromatic architectures the
charge-tunneling contributions probably result from
through-bond and through-space tunneling.[51] Owing to the
minimal difference in the molecular length for the biphe-
nylthiol derivatives, it was not possible to unequivocally re-
solve the influence of tunneling due to the through-bond
molecular length. To understand if the current density ex-
hibits dependence on the SAM medium, we plotted the
log|J| with respect to the effective monolayer thickness
(Figure 4). We observed a b of (0.27ꢂ0.08) ꢀꢀ1 from data
extracted at a bias of ꢀ0.4 V (Figure 4); this is in agreement
Figure 3. a) Characteristics of mean log current density as a function of
junction bias for the Au–5mBPT/Ga2O3 j jGaInE, Au–6mBPT/Ga2O3 j j
GaInE, Au–BPT/Ga2O3 j jGaInE and Au–tBPT/Ga2O3 j jGaInE junctions
indicating forward and reverse traces. The arrow indicates the bias value
from which the logjJj values were extracted to populate the histogram.
b) logjJj distribution profiles at a bias of ꢀ0.4 V for the corresponding
biphenylthiol SAMs 5mBPT, 6mBPT, BPT, and tBPT. The total
number of recorded J observations (Jobs) are indicated in each histogram.
Figure 4. Plot of logjJj vs effective monolayer thickness for BPT SAMs
at a bias of ꢀ0.4 V showing the respective tunneling decay factor estimat-
ed from the slope.
with the b extracted from a series of phenylene-oligomer-
based SAMs measured in a large-area molecular junc-
tion.[47b] On closer inspection of the decay of current density
with respect to the effective thickness for the biphenylthiol
monolayers, it is clear that there is characteristic attenuation
with increase in the distance of separation. It is interesting
to note that the log|J| dependence with cosine square torsion
angle for our biphenylthiol SAMs exhibits an opposite trend
observed in similar molecules on a single-molecule scale
(see the Supporting Information).[23] We attribute this to the
likely weaker cohesive interaction between the SAM and
top GaInE electrode,[4] which is considered to be of van der
Waals nature, being in contrast to the much stronger cova-
obtain a meaningful estimate of J for each of the target
monolayers. In total more than 20000 J values were collect-
ed with a minimum of 3206 I–V traces recorded for each of
the SAMs on at least two samples per SAM. The corre-
sponding J values were used to populate the histograms.
The statistical data for the recorded J values are collected in
Table 3 along with the molecular chemical structure.
The higher degree of disorder and lower packing density
of the tBPT SAMs influences the J showing clearly several
distinct multiple peaks (with broader distribution range) in
the log|J| histogram (see Figure 3b). This is in contrast with
&
8
&
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Molecular Electronics
FULL PAPER
lent electronic coupling between the biphenyl dithiol mole-
cules and gold electrodes studied by Vonlanthen et al.[23]
surface wetting properties of the Au–SAM were estimated
over macroscopic areas from static-water contact-angle (qca)
measurements giving values spanning between about 70 and
798 (see Table 4), highlighting a higher hydrophobic nature
of the coated electrodes with respect to the bare gold. The
qca for Au-BPT SAM (ca. 78.98) was in excellent agreement
for a similar type of metal–SAM surface reported by us ear-
lier.[7a] Note that these values match closely to the contact
angle reported in the literature for non-substituted biphe-
Electrical characterization of OTFTs with biphenylthiol–
SAM functionalized electrodes: To obtain information on
the charge-injection properties of OTFT electrodes modified
with biphenylthiol-based SAMs, we fabricated and charac-
terized transistors in bottom-gate bottom-contact device ge-
ometry by using P3HT as active semiconducting layer. The
channel length (L) and width (W) for the OTFTs under
study amounted to 10 and 10000 mm, respectively. Given
that the performance of OTFTs was sensitive enough to
detect the subtle differences in charge-transport across
SAMs chemisorbed on the source and drain electrodes,[6a]
we sought to determine the influence of the varying torsion-
angle in the molecules composing the SAM.
The route to modify the electrode work function (FM) by
means of SAMs and thereby to adjust the energy levels in
the system to either inject or extract charge carriers (elec-
trons or holes) more effectively in organic electronic devices
have been vastly exploited.[7a,52] In particular, the chemisorp-
tion of alkanethiols and perfluoroalkanethiols has been used
to lower or raise the work function of Au electrodes, respec-
tively, to ultimately improve the injection of electron or
holes.[52a] The shift in the work function is generally attribut-
ed to the total of effective dipole moment formed between
nylthiol SAMs on AuACTHNUTRGNEUNG
(111).[55]
The output (drain current (ID) vs. source-drain voltage
(VDS)) and transfer (ID vs. gate voltage (VGS)) characteristics
were measured for the devices incorporating electrodes
modified with the BPT family of SAMs and compared with
devices with non-functionalized electrodes. Figure 5 shows
the OTFT device configuration and the ID vs. VDS plots ob-
tained for the device with the BPT-SAM-modified electro-
des. The recorded ID vs. VGS curves for the TFTs with the
electrodes modified with the different SAMs are shown in
Figure 6. The other key device performance indicators like
ꢀ
the Au S bond along with the dipole moment due to the
chemical composition of the molecule in the SAM,[53] the
tail groups contributing notably to the overall interfacial
dipole.[54] Among different possibilities, the work function of
the metal can be measured by means of the contact poten-
tial difference (CPD) technique of a macroscopic Kelvin
probe. We extracted the work function of Au electrodes
coated with biphenyl-based SAMs from the CPD by using
an Au reference tip (Fꢃ5.1 eV). Table 4 gives the respec-
tive data along with the work function of a bare Au elec-
trode. The work function for the BPT SAM was measured
to be about (4.81ꢂ0.01) eV, which is in the range of the re-
ported value in literature.[7a] The work function difference
between the more disordered Au–tBPT SAM with respect
to the ordered Au–BPT SAM was approximately 210 meV,
note that as the molecules are intrinsically quite non-polar,
the effective shift in the work function due to the variability
in packing densities and tilt angle of the assembled mono-
layer is fairly evident from the recorded values of FM. The
Figure 5. a) OTFT device geometry with biasing configurations. b)
Output characteristics for TFT based on Au–BPT SAM modified electro-
des (W=10000 mm and L=10 mm) with respective VGS denoted.
Table 4. Contact angle, work function of electrodes, and other key device performance indicators.
Electrode
surface
N
N
Fbh+
[meV][a]
(mh+,satꢂs)ꢅ10ꢀ3
A
ACHTUNGTRENNUNG
[b]
N
[cm2 Vꢀ1 sꢀ1
]
bare Au
5mBPT
6mBPT
BPT
(18.6ꢂ4.2)
(71.6ꢂ3.4)
(69.3ꢂ0.6)
(78.9ꢂ1.9)
(68.0ꢂ2.2)
(4.88ꢂ0.06)
(4.98ꢂ0.02)
(4.93ꢂ0.02)
(4.81ꢂ0.02)
(5.02ꢂ0.03)
25
121
72
43
164
N
G
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
tBPT
ACHTUNGTRENNUNG
[a] Hole injection barrier with respect to the IE level of P3HT. [b] Average hole mobility calculated in the saturation regime along with the standard de-
viation for at least 8 devices. The surface of the clean bare Au is hydrophilic,[59] the qca of which was measured within 60 s after performing UV/Ozone
treatment.
Chem. Eur. J. 2012, 00, 0 – 0
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
9
&
ÞÞ
These are not the final page numbers!

M. Zharnikov, M. Mayor, P. Samorꢁ et al.
boiling point (110.68C), since it was reported that higher
boiling point solvents tend to promote formation of poly-
meric micro crystalline domains in the active channel, which
can enhance charge transport.[56] The devices incorporating
the BPT SAM exhibited the highest field-effect mobility
with mh+ of 9.0ꢅ10ꢀ3 cm2 Vꢀ1 sꢀ1. This result can be attribut-
ed to the combination of low energetic injection barrier and
better electrode wettability by the nonpolar P3HT (the
P3HT film has a qca of ca. 1068 on the surface of SiO2)[57]
over the other, more hydrophobic BPT-based monolayer
(see Table 4). The interfacial packing of the P3HT is likely
to be better in more hydrophobic electrodes. In contrast, the
devices with Au–tBPT electrodes exhibited the lowest field-
effect mobility (mh+ of 6.4ꢅ10ꢀ3 cm2 Vꢀ1 sꢀ1) in comparison
with all other SAMs. This lower mobility for devices that
contain Au–tBPT can be correlated with the relatively large
hole injection barrier of 164 meV, which is greater by
a factor of about 4 than the devices with highest mobility in-
jection barrier for Au–BPT electrodes. The carrier mobility
in organic semiconductors has been shown to be influenced
by the injection barrier with the charge-injecting elec-
trode.[37] Hence, the lower the injection barrier, the higher is
the current-injection efficiency. It has been shown for
OTFTs that the contact resistance plays a big role in the
lower source-drain bias,[58] while at sufficiently high source-
drain potentials the resistance offered by the contacts is lim-
ited.[58b] Since we estimate the field-effect hole mobility at
a VDS of ꢀ60 V, the effect of the contact resistance has been
reasonably omitted. Note that the standard deviation of mo-
bility was within 15% indicating good consistency for all the
tested devices.
Figure 6. a) Transfer curves for the OTFTs based on electrodes with dif-
ferent BPT SAMs showing variation of square root of ID vs VGS (left
axis) and semilog ID vs VGS (right axis) when VDS =ꢀ60 V with lines indi-
cating 5mBPT (red), 6mBPT (brown), BPT (green), tBPT (blue) and
bare Au (black). b) Electrical resistivity dependence on effective thick-
ness of the biphenylthiol SAMs estimated for a bias of ꢀ0.4 V with
a linear least-squares fitting.
turn on voltage (Von) and the threshold voltage (VT) were
estimated from the transfer plots. The Von is the voltage
value for which a change in slope sign occurs in the log
scale on the transfer curves and was extracted from the ID–
Discussion
VGS in the linear regime. VT is the intercept with the gate
In order to address the charge transport through the SAMs,
the molecular structure, monolayer structural defects and
packing density are among the pertinent factors that should
be considered. The HRXPS data showed that the well-or-
dered BPT SAM on Au had a packing density about 1.5
times greater than the tBPT monolayer, which features the
lowest orientational order. The molecular packing densities
are probably heavily influenced by the intramolecular steric
hindrance in tBPT SAM, which indeed features the highest
torsion angle. Closer inspection of the histograms obtained
from the charge-transport measurements revealed a narrow-
er log|J| distribution for well-ordered BPT, 6mand 5mBPT
SAMs if compared to the wider distribution for disordered
tBPT SAM. This suggests that the measured J values are
sensitive to the structural characteristics of the biphenyl
SAMs chemisorbed on Au. We believe that the charge
transport through our biphenylthiol SAMs is due to a combi-
nation of pathways tunneling through the molecule along
with transport through interchain hops between adjacent
molecules. While the length of the pathway through the mo-
lecular chain is the length of the molecule in its long axis, in
the case of the parallel pathways model proposed by Slowin-
pffiffiffiffiffiffiffi
voltage axis from the plot of
jIDj versus VGS (Figure 6)
and provides the value of gate potential at which most of
the traps levels are filled up. The hole injection barrier
(Fbh+) is defined as the energy difference between ioniza-
tion energy (IE) and the work function of the functionalized
Au electrode. Note that the IE for the P3HT film used in
our devices is about (4.86ꢂ0.02) eV as measured by ambient
photoelectron spectroscopy (see the Supporting Informa-
tion). The evaluated injection barrier can give a good indica-
tion of the energetic mismatch at the metal–SAM/semicon-
ductor interface. The field-effect mobility was extracted
pffiffiffiffiffiffiffi
from the saturation regime by estimating the slope of jIDj
versus VGS and then using Equation (3) in which Ci is the ca-
pacitance of the gate dielectric layer (230 nm of SiO2) and
has a value of 15 nFcmꢀ2.
ꢄ
ꢄ
ꢄ
ꢄ
ꢄ
!
pffiffiffiffiffiffiffi
2
2L
@
jI j
D
msat
¼
ð3Þ
WCi @VGS
VDS
For the P3HT solution, we used toluene, which has a high
&
10
&
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Molecular Electronics
FULL PAPER
ski et al.[51] the pathway due to through-space hops decreases
with the effective thickness of the SAM. Least-squares fit-
ting of the current density with the effective monolayer
thickness yielded b of (0.27ꢂ0.08) ꢀꢀ1.
The electrical resistivity (1) of the biphenylthiol SAMs
can be estimated from the current density (estimated from
the charge transport measurements with GaInE probe) de-
pendence with the applied electric field (E). As expected
the 1 shows a monotonic increase with increasing effective
thickness (Figure 6b). Note that despite the polycrystalline
nature of the Au electrodes in the OTFT and the monocrys-
talline character of the Au used for the measurements with
a GaInE-based junction (see Supporting Information for
AFM images), a fairly good correlation between the electri-
cal resistivity and the relative threshold voltage shift for the
transistor comprising of biphenylthiol-coated electrodes was
found. The VT extracted from the transfer curve in Figure 6a
was measured under biasing conditions in which VGS was
swept from +20!ꢀ60 V, while holding VDS fixed at ꢀ60 V.
It was observed that the VT was about +2.2 V for transistors
containing Au–tBPT SAM electrodes and the corresponding
electrical resistivity 1ꢃ0.12 GWcm also being the lowest;
on the other hand, the devices with Au–5mBPT electrodes
had a more negative shift of VT of ꢀ9.2 V and the highest 1
ꢃ2.2 GWcm (see Figure 6b). This difference in the VT for
devices based on Au–5mBPT electrodes of nearly 18 V in
the upper bounds when compared Au–tBPT can be attribut-
ed directly to the resistivity increase by a similar one-to-one
correlation (i.e., the 1 increases by factor of about 18 for
electrodes containing 5mBPT with respect to tBPT SAMs).
For P3HT transistors the majority charge carrier are holes
(i.e., increase in ID tends to predominantly occur when VGS
is negative). Note that while the potential sweep for the
gate electrode starts from positive voltage and proceeds to
the negative (Figure 6a) the TFTs exhibited VT shifts to-
wards more negative voltages in the order of tBPT,
6mBPT, BP, and 5mBPT, being in good agreement with
the concomitant trend of increasing resistivity through the
biphenylthiol SAMs. The OTFT charge-carrier mobility un-
ambiguously reveals, as expected, that the hole injection
barrier from the electrodes plays an important role in the
transport through the semiconducting film (Figure 7). It is
interesting to note that although the injection barrier for
bare Au and Au–BPT electrodes were in the same range,
confined within error bars (see Table 4), for the device with
just the bare gold the mobility was about 0.0078 cm2 Vꢀ1 sꢀ1,
which in comparison to devices with Au–BPT electrodes
were lower by a factor of about 1.2.
Figure 7. Hole carrier mobility in saturation mode and respective work
function for OTFTs based on electrodes with and without biphenylthiol
SAMs. The dashed line indicates the ionization energy level of P3HT,
which is taken as the reference to estimate the hole injection/extraction
barrier to the source-drain electrode.
mobility than the Au–BPT which has a more hydrophobic
surface. This suggests that, as in various cases previously re-
ported in literature, the mobility of our TFTs is dictated by
1) the injection barrier of electrode with semiconductor, 2)
electrode surface wettability of the semiconductor film and
also 3) the packing density of the molecules in the SAM.
The well-ordered BPT SAM on electrodes exhibited a mobi-
lity which was about 1.4 times greater than that of the mobi-
lity for devices based on more disordered Au–tBPT SAM,
indeed reflecting the variable packing density of the mole-
cules adsorbed on the Au electrodes and the energetic mis-
match at the metal–SAM/semiconductor interface. Overall,
the transistor responses show that our devices are an effec-
tive means to monitor the energy-level adjustment and the
corresponding electronic coupling at the interface and the
threshold voltage-shift scale with the tunneling resistivity
through the biphenyl SAMs on the Au electrodes.
Conclusion
In summary, the charge transport through a family of biphe-
nylthiol-based SAMs with varying torsion angles chemi-
sorbed on Au surfaces exhibits a good degree of correlation
with the packing density and orientational order in the
monolayers. The last parameters were mostly governed by
molecular conformation, which was specifically adjusted by
the site bridging or substitution of/at the individual phenyl
rings. The efficiency of the charge tunneling through the
SAMs shows a characteristic decay with the increasing sepa-
ration between the metallic contacts (i.e, increase of the ef-
fective SAM thickness). The charge-transport experiments
revealed a through-space b of (0.27ꢂ0.08) ꢀꢀ1. For transis-
tors the shift in the threshold voltage scaled as a function of
resistivity through the biphenyl SAMs on the electrodes. A
good agreement between the transistor-electrode/SAM ener-
getic-injection-barrier dependence on the field-effect mobili-
ty was found. These results show that a proper design of
thiolated molecules chemisorbed on Au electrodes is needed
The top-surface morphology of the P3HT film on the
modified electrodes exhibited an isotropic grain-like pattern
(see the respective AFM images in the Supporting Informa-
tion) featuring a fairly similar root-mean-square roughness
(RRMS). The increase in mobility for devices with BPT-
coated electrodes can be attributed to the better arrange-
ment of the semiconductor at the interface with the more
hydrophobic surface over the bare hydrophilic Au electrode
surface, which having a similar injection barrier has a lower
Chem. Eur. J. 2012, 00, 0 – 0
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
11
&
ÞÞ
These are not the final page numbers!

M. Zharnikov, M. Mayor, P. Samorꢁ et al.
[15] R. C. Chiechi, E. A. Weiss, M. D. Dickey, G. M. Whitesides, Angew.
Chem. 2008, 120, 148–150; Angew. Chem. Int. Ed. 2008, 47, 142–
144.
for realizing more efficient organic electronic devices
through improved control over the electrode–semiconductor
interface.
[16] a) D. Fracasso, H. Valkenier, J. C. Hummelen, G. C. Solomon, R. C.
Chiechi, J. Am. Chem. Soc. 2011, 133, 9556–9563; b) G. D. Lilly,
A. C. Whalley, S. Grunder, C. Valente, M. T. Frederick, J. F. Stod-
dart, E. A. Weiss, J. Mater. Chem. 2011, 21, 11492–11497; c) M. M.
Thuo, W. F. Reus, C. A. Nijhuis, J. R. Barber, C. Kim, M. D. Schulz,
G. M. Whitesides, J. Am. Chem. Soc. 2011, 133, 2962–2975.
[17] Yield is referred to as the ratio of the difference between the total
number of J(V) traces to the traces that ended up with a short (in-
strument reached compliance limit) with the total J observations.
[18] a) E. Tran, M. Duati, V. Ferri, K. Mꢆllen, M. Zharnikov, G. M.
Whitesides, M. A. Rampi, Adv. Mater. 2006, 18, 1323–1328; b) V.
Ferri, M. Elbing, G. Pace, M. D. Dickey, M. Zharnikov, P. Samorꢁ,
M. Mayor, M. A. Rampi, Angew. Chem. 2008, 120, 3455–3457;
Angew. Chem. Int. Ed. 2008, 47, 3407–3409.
Acknowledgements
This work was supported by the ERC project SUPRAFUNCTION (GA-
257305), the International Center for Frontier Research in Chemistry
(icFRC, Strasbourg), and the German Research Foundation (DFG)
through the grant ZH 63/14-1. N.C. thanks the Marie-Curie IEF-OPTSU-
FET (PIEF-GA-2009-235967), while E.O. and A. M.M. acknowledge the
ITN-SUPERIOR (PITN-GA-2009-238177) for their fellowships. Re-
search in Basel was supported by the Swiss National Science Foundation
(SNF) and the Swiss Nanoscience Institute (SNI). R.T. and M.Z. thank
A. Nefedov and C. Wçll (KIT) for the technical cooperation at BESSY
II as well as BESSY II staff for the assistance during the synchrotron-
based experiments.
[19] H. B. Akkerman, P. W. M. Blom, D. M. de Leeuw, B. de Boer,
Nature 2006, 441, 69–72.
[20] G. Wang, Y. Kim, M. Choe, T.-W. Kim, T. Lee, Adv. Mater. 2011, 23,
755–760.
[21] a) J. C. Love, L. A. Estroff, J. K. Kriebel, R. G. Nuzzo, G. M. White-
sides, Chem. Rev. 2005, 105, 1103–1170; b) Y. Qi, I. Ratera, J. Y.
Park, P. D. Ashby, S. Y. Quek, J. B. Neaton, M. Salmeron, Langmuir
2008, 24, 2219–2223.
[1] A. Aviram, M. A. Ratner, Chem. Phys. Lett. 1974, 29, 277–283.
[2] a) L. Venkataraman, J. E. Klare, C. Nuckolls, M. S. Hybertsen, M. L.
Steigerwald, Nature 2006, 442, 904–907; b) J. M. Mativetsky, G.
Pace, M. Elbing, M. A. Rampi, M. Mayor, P. Samorꢁ, J. Am. Chem.
Soc. 2008, 130, 9192–9193; c) C. A. Nijhuis, W. F. Reus, A. C. Siegel,
G. M. Whitesides, J. Am. Chem. Soc. 2011, 133, 15397–15411.
[3] a) C. D. Bain, G. M. Whitesides, Science 1988, 240, 62; b) W. B. Cald-
well, D. J. Campbell, K. Chen, B. R. Herr, C. A. Mirkin, A. Malik,
M. K. Durbin, P. Dutta, K. G. Huang, J. Am. Chem. Soc. 1995, 117,
6071–6082; c) G. Pace, V. Ferri, C. Grave, M. Elbing, C. von Hꢃ-
nisch, M. Zharnikov, M. Mayor, M. A. Rampi, P. Samorꢁ, Proc. Natl.
Acad. Sci. USA 2007, 104, 9937–9942.
[22] a) D. Li, B. I. Swanson, J. M. Robinson, M. A. Hoffbauer, J. Am.
Chem. Soc. 1993, 115, 6975–6980; b) F. Anariba, J. K. Steach, R. L.
McCreery, J. Phys. Chem. B 2005, 109, 11163–11172; c) W. Haiss, C.
Wang, R. Jitchati, I. Grace, S. Martꢈn, A. S. Batsanov, S. J. Higgins,
M. R Bryce, C. J. Lambert, P. S. Jensen, R. J. Nichols, J. Phys. Con-
dens. Matter 2008, 20, 374119; d) A. B. Ricks, G. C. Solomon, M. T.
Colvin, A. M. Scott, K. Chen, M. A. Ratner, M. R. Wasielewski, J.
Am. Chem. Soc. 2010, 132, 15427–15434; e) G. Bruno, F. Babudri,
A. Operamolla, G. V. Bianco, M. Losurdo, M. M. Giangregorio,
O. H. Omar, F. Mavelli, G. M. Farinola, P. Capezzuto, F. Naso, Lang-
muir 2010, 26, 8430–8440; f) A. Operamolla, O. H. Omar, F. Babu-
dri, G. M. Farinola, F. Naso, J. Org. Chem. 2007, 72, 10272–10275;
g) S. Casalini, F. Leonardi, C. A. Bortolotti, A. Operamolla, O.
Hassan, L. Paltrinieri, C. Albonetti, G. M. Farinola, F. Biscarini, J.
Mater. Chem. 2012, 22, 12155–12163.
[23] a) A. Mishchenko, D. Vonlanthen, V. Meded, M. Bꢆrkle, C. Li, I. V.
Pobelov, A. Bagrets, J. K. Viljas, F. Pauly, F. Evers, M. Mayor, T.
Wandlowski, Nano Lett. 2010, 10, 156–163; b) D. Vonlanthen, A.
Mishchenko, M. Elbing, M. Neuburger, T. Wandlowski, M. Mayor,
Angew. Chem. 2009, 121, 9048–9052; Angew. Chem. Int. Ed. 2009,
48, 8886–8890.
[24] A. Mishchenko, L. A. Zotti, D. Vonlanthen, M. Burkle, F. Pauly,
J. C. Cuevas, M. Mayor, T. Wandlowski, J. Am. Chem. Soc. 2011,
133, 184–187.
[4] C. A. Nijhuis, W. F. Reus, G. M. Whitesides, J. Am. Chem. Soc. 2009,
131, 17814–17827.
[5] a) N. Crivillers, E. Orgiu, F. Reinders, M. Mayor, P. Samorꢁ, Adv.
Mater. 2011, 23, 1447–1452; b) S. Karpe, M. Ocafrain, K. Smaali, S.
Lenfant, D. Vuillaume, P. Blanchard, J. Roncali, Chem. Commun.
2010, 46, 3657–3659; c) A. J. Kronemeijer, H. B. Akkerman, T. Ku-
dernac, B. J. van Wees, B. L. Feringa, P. W. M. Blom, B. de Boer,
Adv. Mater. 2008, 20, 1467–1473.
[6] a) P. Stoliar, R. Kshirsagar, M. Massi, P. Annibale, C. Albonetti,
D. M. De Leeuw, F. Biscarini, J. Am. Chem. Soc. 2007, 129, 6477–
6484; b) X. Cheng, Y. Y. Noh, J. Wang, M. Tello, J. Frisch, R. P.
Blum, A. Vollmer, J. P. Rabe, N. Koch, H. Sirringhaus, Adv. Funct.
Mater. 2009, 19, 2407–2415.
[7] a) E. Orgiu, N. Crivillers, J. Rotzler, M. Mayor, P. Samorꢁ, J. Mater.
Chem. 2010, 20, 10798; b) M. E. Gemayel, M. Treier, C. Musumeci,
C. Li, K. Mꢆllen, P. Samorꢁ, J. Am. Chem. Soc. 2012, 134, 2429–
2433.
[8] a) V. B. Engelkes, J. M. Beebe, C. D. Frisbie, J. Am. Chem. Soc.
2004, 126, 14287–14296; b) V. B. Engelkes, J. M. Beebe, C. D. Fris-
bie, J. Phys. Chem. B 2005, 109, 16801–16810.
[9] a) Y. Selzer, A. Salomon, D. Cahen, J. Phys. Chem. B 2002, 106,
10432–10439; b) E. A. Weiss, R. C. Chiechi, G. K. Kaufman, J. K.
Kriebel, Z. Li, M. Duati, M. A. Rampi, G. M. Whitesides, J. Am.
Chem. Soc. 2007, 129, 4336–4349.
[10] a) M. T. Gonzꢇlez, S. M. Wu, R. Huber, S. J. van der Molen, C. Scho-
nenberger, M. Calame, Nano Lett. 2006, 6, 2238–2242; b) M. A.
Reed, C. Zhou, C. J. Muller, T. P. Burgin, J. M. Tour, Science 1997,
278, 252–254; c) J. Reichert, R. Ochs, D. Beckmann, H. B. Weber,
M. Mayor, H. von Lohneysen, Phys. Rev. Lett. 2002, 88, 176804.
[11] X. D. Cui, A. Primak, X. Zarate, J. Tomfohr, O. F. Sankey, A. L.
Moore, T. A. Moore, D. Gust, G. Harris, S. M. Lindsay, Science 2001,
294, 571–574.
[25] S. Woitellier, J. Launay, C. Joachim, Chem. Phys. 1989, 131, 481–
488.
[26] a) B. Jꢃger, H. Schꢆrmann, H. U. Mꢆller, H. J. Himmel, M. Neu-
mann, M. Grunze, C. Wçll, Z. Phys. Chem. 1997, 202, 263–272;
b) M. Wirde, U. Gelius, T. Dunbar, D. L. Allara, Nucl. Instrum.
Methods Phys. Res. Sect. B 1997, 131, 245–251; c) M. Zharnikov, M.
Grunze, J. Vac. Sci. Technol. B 2002, 20, 1793–1807.
[27] X-ray photoeletron spectrometers—calibration of the energy scales,
ISO 15472 Surface chemical analysis 2001.
[28] W. Azzam, B. I. Wehner, R. A. Fischer, A. Terfort, C. Wçll, Lang-
muir 2002, 18, 7766–7769.
[29] J. Stçhr, NEXAFS spectroscopy, Springer, Berlin, New York, 1992.
[30] S. Frey, V. Stadler, K. Heister, W. Eck, M. Zharnikov, M. Grunze, B.
Zeysing, A. Terfort, Langmuir 2001, 17, 2408–2415.
[31] P. E. Batson, Phys. Rev. B 1993, 48, 2608–2610.
[32] R. S. Timsit, Appl. Phys. Lett. 1982, 40, 379–381.
[12] W. Wang, T. Lee, M. A. Reed, Phys. Rev. B 2003, 68, 035416.
[13] J. G. Kushmerick, D. B. Holt, J. C. Yang, J. Naciri, M. H. Moore, R.
Shashidhar, Phys. Rev. Lett. 2002, 89, 086802.
[33] A. Shaporenko, M. Elbing, A. Blaszczyk, C. von Hꢃnisch, M.
Mayor, M. Zharnikov, J. Phys. Chem. B 2006, 110, 4307–4317.
[34] M. Zharnikov, M. Grunze, J. Phys. Condens. Matter 2001, 13, 11333.
[14] A. P. Bonifas, R. L. McCreery, Nat. Nanotechnol. 2010, 5, 612–617.
&
12
&
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Molecular Electronics
FULL PAPER
[35] M. Zharnikov, J. Electron Spectrosc. Relat. Phenom. 2010, 178, 380–
393.
[51] K. Slowinski, R. V. Chamberlain, C. J. Miller, M. Majda, J. Am.
Chem. Soc. 1997, 119, 11910–11919.
[36] J. A. Horsley, J. Stohr, A. P. Hitchcock, D. C. Newbury, A. L. John-
son, F. Sette, J. Chem. Phys. 1985, 83, 6099–6107.
[37] Y. Shen, M. W. Klein, D. B. Jacobs, J. Campbell Scott, G. G. Mallia-
ras, Phys. Rev. Lett. 2001, 86, 3867–3870.
[38] C. Fuxen, W. Azzam, R. Arnold, G. Witte, A. Terfort, C. Wçll,
Langmuir 2001, 17, 3689–3695.
[39] T. Yokoyama, K. Seki, I. Morisada, K. Edamatsu, T. Ohta, Phys. Scr.
1990, 41, 189.
[52] a) B. de Boer, A. Hadipour, M. M. Mandoc, T. vanWoudenbergh,
P. W. M. Blom, Adv. Mater. 2005, 17, 621–625; b) G. Heimel, L. Ro-
maner, E. Zojer, J. L. Bredas, Acc. Chem. Res. 2008, 41, 721–729;
c) S. A. DiBenedetto, A. Facchetti, M. A. Ratner, T. J. Marks, Adv.
Mater. 2009, 21, 1407–1433; d) D. Boudinet, M. Benwadih, Y. Qi, S.
Altazin, J. M. Verilhac, M. Kroger, C. Serbutoviez, R. Gwoziecki, R.
Coppard, G. Le Blevennec, Org. Electron. 2010, 11, 227–237; e) K.
Singh, T. Nelson, J. Belot, T. Young, N. R. Dhumal, T. Kowalewski,
R. McCullough, P. Nachimuthu, S. Thevuthasan, L. M. Porter, ACS
Appl. Mater. Interfaces 2011, 3, 2973–2978.
[40] A. Hargreaves, S. H. Rizvi, Acta Crystallogr. 1962, 15, 365–373.
[41] N. Ballav, B. Schꢆpbach, O. Dethloff, P. Feulner, A. Terfort, M. Zha-
mikov, J. Am. Chem. Soc. 2007, 129, 15416–15417.
[53] G. Heimel, L. Romaner, E. Zojer, J.-L. Brꢂdas, Nano Lett. 2007, 7,
932–940.
[42] a) J. Trotter, Acta Crystallogr. 1961, 14, 1135–1140; b) A. Bashir, D.
Kꢃfer, J. Mꢆller, C. Wçll, A. Terfort, G. Witte, Angew. Chem. 2008,
120, 5328–5331; Angew. Chem. Int. Ed. 2008, 47, 5250–5252.
[43] a) S. C. Chang, I. Chao, Y. T. Tao, J Am Chem. Soc 1994, 116, 6792–
6805; b) A. A. Dhirani, R. W. Zehner, R. P. Hsung, P. Guyot-Sion-
nest, L. R. Sita, J. Am. Chem. Soc. 1996, 118, 3319–3320.
[44] G. M. Brown, M. H. Bortner, Acta Crystallogr. 1954, 7, 139.
[45] R. Cosmo, T. W. Hambley, S. Sternhell, J. Org. Chem. 1987, 52,
3119–3123.
[54] a) I. Campbell, S. Rubin, T. Zawodzinski, J. Kress, R. Martin, D.
Smith, N. Barashkov, J. Ferraris, Phys. Rev. B 1996, 54, R14321–
R14324; b) N. V. Venkataraman, S. Zꢆrcher, A. Rossi, S. Lee, N.
Naujoks, N. D. Spencer, J. Phys. Chem. C 2009, 113, 5620–5628.
[55] J. F. Kang, A. Ulman, S. Liao, R. Jordan, G. H. Yang, G. Y. Liu,
Langmuir 2001, 17, 95–106.
[56] J. F. Chang, B. Sun, D. W. Breiby, M. M. Nielsen, T. I. Sçlling, M.
Giles, I. McCulloch, H. Sirringhaus, Chem. Mater. 2004, 16, 4772–
4776.
[46] a) S. Ho Choi, B. Kim, C. D. Frisbie, Science 2008, 320, 1482–1486;
b) Q. Lu, K. Liu, H. Zhang, Z. Du, X. Wang, F. Wang, ACS Nano
2009, 3, 3861–3868.
[57] L. Robinson, J. Isaksson, N. D. Robinson, M. Berggren, Surfactant
Sci. Ser. 2006, 600, L148–L152.
[47] a) T. Ishida, W. Mizutani, Y. Aya, H. Ogiso, S. Sasaki, H. Tokumoto,
J. Phys. Chem. B 2002, 106, 5886–5892; b) A. Kronemeijer, E. Huis-
man, H. Akkerman, A. Goossens, I. Katsouras, P. van Hal, T. Geuns,
S. van der Molen, P. Blom, D. de Leeuw, Appl. Phys. Lett. 2010, 97,
173302.
[58] a) G. Horowitz, R. Hajlaoui, R. Bourguiga, M. Hajlaoui, Synth. Met.
1999, 101, 401–404; b) V. Podzorov, S. E. Sysoev, E. Loginova, V. M.
Pudalov, M. E. Gershenson, Appl. Phys. Lett. 2003, 83, 3504–3506.
[59] C. G. Worley, R. W. Linton, J. Vac. Sci. Technol. A 1995, 13, 2281–
2284.
[48] C. A. Nijhuis, W. F. Reus, G. M. Whitesides, J. Am. Chem. Soc. 2010,
132, 18386–18401.
[60] S. Liao, R. Harvey, J. Pataki, EP0363128 (A2), 1990.
[49] H. Wickham, ggplot2: elegant graphics for data analysis, Springer,
New York 2009.
[50] R Development Core Team, R: A language and environment for
statistical computing, Springer, Vienna, Austria 2011.
Received: May 28, 2012
Published online: && &&, 0000
Chem. Eur. J. 2012, 00, 0 – 0
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
13
&
ÞÞ
These are not the final page numbers!

M. Zharnikov, M. Mayor, P. Samorꢁ et al.
Controlled charge injection! Biphe-
nylthiols with different substituents
between phenyl groups in the bridge
position, once chemisorbed onto gold
source-drain electrodes, can be used to
vary the charge injection and electrical
performances of organic thin-film tran-
sistors. By varying the torsion angle
between the phenyl rings one can
introduce changes in packing densities
and tilt angles of the molecules self-
assembled on the electrode that play
a direct role influencing the charge
transport through the molecular mono-
layer (see figure).
Molecular Electronics
A. M. Masillamani, N. Crivillers,
E. Orgiu, J. Rotzler, D. Bossert,
R. Thippeswamy, M. Zharnikov,*
M. Mayor,* P. Samorꢀ* ...... &&&& — &&&&
Multiscale Charge Injection and Trans-
port Properties in Self-Assembled
Monolayers of Biphenyl Thiols with
Varying Torsion Angles
&
14
&
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!