Triazatriangulene as Binding Group for Molecular Electronics

Article pubs.acs.org/Langmuir Triazatriangulene as Binding Group for Molecular Electronics Zhongming Wei,†,‡,⊥ Xintai Wang,†,⊥ Anders Borges,† Marco Santella,†,‡ Tao Li,† Jakob Kryger Sørensen,† Marco Vanin,† Wenping Hu,§ Yunqi Liu,§ Jens Ulstrup,∥ Gemma C. Solomon,† Qijin Chi,∥ Thomas Bjørnholm,† Kasper Nørgaard,*,† and Bo W. Laursen*,† † Nano-Science Center & Department of Chemistry, University of Copenhagen, Universitetsparken 5, DK-2100 Copenhagen Ø, Denmark ‡ Sino-Danish Centre for Education and Research (SDC), Niels Jensens Vej 2, DK-8000 Aarhus C, Denmark § Beijing National Laboratory for Molecular Sciences, Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China ∥ Department of Chemistry and NanoDTU, Technical University of Denmark, Building 207, Kemitorvet, 2800 Lyngby, Denmark S Supporting Information * ABSTRACT: The triazatriangulene (TATA) ring system was investigated as a binding group for tunnel junctions of molecular wires on gold surfaces. Self-assembled monolayers (SAMs) of TATA platforms with three different lengths of phenylene wires were fabricated, and their electrical conductance was recorded by both conducting probe-atomic force microscopy (CP-AFM) and scanning tunneling microscopy (STM). Similar measurements were performed for phenylene SAMs with thiol anchoring groups as references. It was found that, despite the presence of a sp3 hybridized carbon atom in the conduction path, the TATA platform displays a contact resistance only slightly larger than the thiols. This surprising finding has not been reported before and was analyzed by theoretical computations of the transmission functions of the TATA anchored molecular wires. The relatively low contact resistance of the TATA platform along with its high stability and directionality make this binding group very attractive for molecular electronic measurements and devices. ■ INTRODUCTION Thiol, amine, isocyanide, C60, and other groups have commonly been used to link the functional molecular electronic element to different kinds of substrates.14,16−18 In the molecular junctions based on SAMs formed from molecular wires with these anchors, the molecular wires are commonly densely packed causing the total current density to contain contributions from both intramolecular (through-bond tunneling) and intermolecular (wire-to-wire tunneling) transport.19,20 To understand the intrinsic molecular properties, the transport pathway should, however, be dominated by through-bond transport.21 An anchoring group that can hinder the intermolecular coupling, and yet provide good electronic contact to the electrode and a structurally well-defined SAM, is highly desirable to obtain true single molecule transport characteristics and still benefit from the multitude of parallel junctions in SAM devices. In addition, a wider space between the molecular adsorbates would enable chemical reactivity of the adsorbates to be monitored. The triazatriangulenium (TATA+) ion is a large cationic πsystem where various side chains can easily be introduced on the three nitrogen atoms (Scheme 1).22,23 Addition of Molecular electronics where single molecules are employed as the active transport component in devices has attracted tremendous attention since the first theoretical prediction of a single molecule rectifier in 1974.1−5 A wide range of different methods have been developed to form suitable size nanogap electrodes and record the charge transport properties of individual or small groups of molecules in the self-assembled monolayers (SAMs).6−12 Binding groups are of paramount importance in SAM-based molecular electronics. Increased electronic coupling between the molecule and the surface can efficiently improve the transport properties, and stronger binding can increase the thermal stability of the device.13 A binding group should have the ideal characteristics of: (i) strong and specific interaction with the metal electrode surface, leading to well-defined and reproducible SAMs; (ii) good electronic coupling between electrode and molecular wire; and (iii) being generally applicable with different types of molecular wires.14,15 In some cases, the binding group has been found to be even more important than the molecular wire itself in determining whether hole-dominated (p-type) or electrondominated (n-type) transport in the molecular junctions prevails.16 © 2014 American Chemical Society Received: October 14, 2014 Published: November 26, 2014 14868 dx.doi.org/10.1021/la504056v | Langmuir 2014, 30, 14868−14876 Langmuir ■ Scheme 1. (a) Molecular Structures of TATA 1−3 and (b) Optimized Geometry of TATA 1 (Alkyl Chains Were Changed into a Methyl Group for Clarity); (c) Molecular Structures of Bis-thiol SS1−3 and Thiol S1−3 Series; and (d) Schematic Structural Comparison between Thiol, Bisthiol, and TATA Anchored Terphenyl Molecular Wires Article EXPERIMENTAL SECTION Materials. SS1−3 and S1−3 (Scheme 1c) were obtained from Sigma-Aldrich or Alfa Aesar. Trioctyl-TATA+BF4− was synthesized by a reported procedure.22 Detailed synthesis procedures and characterization for TATA+BPh4− and TATA 1−3 are given in the Supporting Information. Formation of SAMs. The compounds SS1−3, S1−3 (0.25 mM), and TATA1−3 (0.05 mM) were dissolved in Ar purged ethanol. Freshly prepared gold substrates (single-crystal Au (111) electrodes) were immediately immersed into the solutions. The containers were filled with Ar and sealed. SAMs were grown for 48 h at room temperature in the dark. After removal from the solutions, the sample was rinsed with copious amounts of ethanol and dried under a stream of N2. SAMs of alkyl/phenyl thiol matrix were prepared by immersing Au substrates into alkyl thiol solution for 1 h, and then immersing into phenyl thiol solution for 24 h (pentane and heptanethiol were used as matrix for S2 and S3 thiol SAMs, respectively). Measurements. Electrical properties were obtained with CP-AFM using a Digital Instruments Nanoscope III at room temperature in air. For CP-AFM, a conductive contact mode Pt/Cr coated tip (Veeco model SCM-PIC Silicon probe, NanoWorld) with an additional 30 nm sputter Au was used. The current−voltage (I−V) characteristics of the SAM were measured by combining an AFM, a Keithley 2400 digital source meter, a Keithley 6154 power source, and an input control and output recording program in Labview. The conductive tip in contact with SAM was the top-electrode, and the Au substrate underneath the SAM was connected with a conducting wire as external electrode. The load force was maintained at about 4 nN to achieve a soft contact. All I−V curves were recorded using the same tip and at different places (five cyclic measurements at each place). STM measurements were carried out in the electrochemical, ECSTM mode with independent electrochemical potential control of the working electrode and tip.35,36 A single-crystal Au (111) electrode (SPL, The Netherlands and MaTeck, Germany) was electropolished with H2SO4, and the oxide was removed with HCl. This was followed by annealing, first in a muffle furnace at 860 oC for 8 h, and then in a hydrogen flame followed by quenching in hydrogen saturated water. A PicoSPM EC-STM instrument was used. A Teflon cell was used with a platinum wire as reference and counter electrodes, calibrated against a saturated calomel electrode (SCE). Electrochemically etched tungsten tips (d = 0.25 mm) coated with Apiezon wax served as scanning probe. EC-STM of the TATA SAMs was performed using 0.1 M HClO4 (pH = 1.69). With the tunneling current set point varied from 30−100 pA, the bias voltage from −0.4 to −1.3 V, and sample potential set between −0.3 and 0.5 V vs SCE, I and P gains were adjusted according to image quality. I−V curves were measured by localizing the tip at a position where the tunneling current reaches the set point of 1 nA, with the working potential kept constant at 0.1 V. The feedback loop was switched off and the bias voltage swept from 0.3 to −0.3 V. Tunneling spectra were recorded from 50−200 individual scans, with a scan rate = 0.5 V/s. Electrochemical measurements were recorded using a threecompartment cell Autolab PGSTAT 12 system controlled by the GPES 4.9 software. A freshly prepared reversible hydrogen electrode (RHE) was used as reference electrode, and a Pt wire annealed in a hydrogen flame as counter electrode. The reference electrode was calibrated vs SCE after each experiment. Cyclic voltammetry (CV) was recorded for 10 mM K3[Fe(CN)6]/K4[Fe(CN)6] solution with 1 M KCl as supporting electrolyte and with a scan rate of 50 mV/s. Electrochemical reductive desorption data were obtained by linear sweep voltammetry in basic solution (KOH, pH > 13) with a scan rate of 10 mV/s. The electrolyte solutions were degassed with argon for 15 min to 1 h before all measurements. Purified water (MILLI-Q, 18 MΩ cm) was used throughout, and all glassware was boiled in 15% nitric acid and washed several times with MILLI-Q water before use. Water contact angles (Supporting Information Figure S4) on the SAMs were measured under ambient conditions by depositing a drop (∼1 μL) of pure MILLI-Q water on the SAM surface. The contact angles were carbanions to the central carbon atom in triangulenium ions leads to neutral platform molecules with the carbanion segment oriented normal to the now curved plane of the triangulene ring system.24−26 Herges and Magnussen reported that such TATA platforms bind efficiently to Au surfaces and form well-defined monolayers with the wire segment oriented normal to the substrate.27−34 The large footprint of the TATA platform (>150 Å2) ensures no in-plane contact between the wires in the monolayer. In a series of studies, Herges and Magnussen demonstrated these features of the TATA platforms and demonstrated their advantages for anchoring molecular structural switches (azobenzenes) that require free volume for optimal operation.27−34 These features may also be attractive for anchoring molecular wires for SAM-based molecular tunnel junctions. For this application, the large footprint of the TATA platform would eliminate any wire-to-wire transport, as well as providing invariant electronic contact to the Au substrate/electrode. To evaluate the potential of TATA as a binding/anchoring group for molecular tunnel junctions, we have designed three simple phenylene wire molecules with TATA binding groups (TATA 1−3, Scheme 1a) and compared the electronic transport properties of the SAMs of these molecules with SAMs of similar phenylene wires with thiol binding groups either in one end of the wire (S1−3) or in both ends (bis-thiol, SS1−3) (Scheme 1c). Comparative tunnel junction measurements for this series of phenylene wires with different anchoring groups were recorded to evaluate the performance of the TATA anchoring group (Scheme 1d). Measurements were performed with both conducting probe-atomic force microscopy (CP-AFM) and scanning tunneling microscopy (STM), and supplemented by theoretical transport calculations analyzing the details of tunneling transport through the TATA binding group. Such functional electronic characterization of TATA-based molecules has not been addressed before. 14869 dx.doi.org/10.1021/la504056v | Langmuir 2014, 30, 14868−14876 Langmuir Article determined from optical images a few seconds after application of the drop. Quantum Chemical Calculations. All geometry optimizations were performed using Density Functional Theory (DFT) as implemented in the ASE/GPAW packages.37,38 Here, we provide a basic outline of how the structures were obtained. Additional details of how the geometries were obtained are provided in the Supporting Information. The geometries of the thiols were optimized in a vacuum using the PBE functional.39 The optimized dihedrals between successive rings were approximately 33°. The thiols were chemisorbed (terminal hydrogen atoms removed) to a Au fcc(111) surface consisting of three layers of 4 × 4 Au atoms with a nearest neighbor distance of 2.885 Å (corresponding to a lattice constant of 4.08 Å). The binding geometry of the S−Au interface was assumed to be a fcc hollow site at a vertical distance of 2.04 Å and the corresponding H− Au binding geometry (in the monothiols) a hollow site site at a vertical distance of 2.4 Å. The geometry of the bare TATA platform, with the nitrogen substituents set as methyl groups, was optimized with a 2 × 2 Monckhorst−Pack k-point sampling to within 0.05 eV/Å on a 6 × 6 × 3 Au (111) surface using the Landgreth−Lundqvist vdW-DF2 functional40 with the Cooper exchange functional41 to best model the expected dispersion interaction that controls binding between the TATA-platform and Au. The molecular wires were substituted onto the optimized TATA-platform with a bond length from the corresponding optimized TATAs in vacuum. The coherent transport properties were calculated using both the ASE transport calculator and gDFTB.42−45 A single layer of Au atoms was included in the local currents calculation to investigate the Au−TATA interface. Figure 1. EC-STM images of TATA 1 SAM. Scan area: (a) 50 nm × 50 nm, (b) 30 nm × 30 nm. (Tunneling current set point, 0.03 nA; working potential, −0.1 V; solution pH,: 1.69; voltage bias, −0.9 V.) approximately 1.4 nm, and the mean molecular area (footprint) of a single molecule is around 1.7 nm2. The unit cell obtained for TATA 1 was similar to the findings of Herges and Magnussen.27−29 The white spots in the STM images (Supporting Information Figure S1) for TATA 2 and 3 indicate that the phenyl groups point up from the gold substrate, which accords with expectations that the phenylene wires are perpendicular to the TATA platform due to the sp3 hybridization of the central carbon. The density and quality of the SAMs were characterized further by cyclic voltammetry of an aqueous K3[Fe(CN)6]/ K4[Fe(CN)6] solution recorded with the SAM covered Au substrates as electrodes, Figure 2. As compared to the bare Au ■ RESULTS AND DISCUSSION Experimental Characterization and Properties of the TATA SAMs. The syntheses of phenyl-, biphenyl-, and tertphenyl TATAs (TATA 1−3) were performed by addition of the phenylene lithium anions to the TATA+ cations as outlined in Scheme 2. Contrary to the previously reported Scheme 2. Synthesis of TATA 1−3 Figure 2. Cyclic voltammograms of aqueous K3[Fe(CN)6]/K4[Fe(CN)6] solution recorded with the three TATA 1−3 SAMs on Au as electrodes (scan rate: 100 mV/s). synthesis of TATA platforms (including that of phenylsubstituted TATA 1),27,30,34 TATA+BF4− was first converted into the tetraphenylborate salt (TATA+BPh4−). The solubility of TATA+BPh4− is significantly higher in the low-polarity solvent (THF) needed to handle the lithium carbanions, leading to improved yields and reduced reaction time. The morphology and density of TATA 1−3 SAMs on Au were characterized by EC-STM and cyclic voltammetry (CV). The EC-STM images showed that TATAs form well-ordered SAMs on Au substrates as reported by Herges and Magnussen.27−29 TATA 1 molecules have a triangular shape and form a highly ordered hexagonal lattice structure (Figure 1). The center-to-center distance between two molecules is electrode, the peaks for the Fe(II)/Fe(III) redox signal totally disappear when the bis-thiols SS1−3 SAMs cover the Au electrode (Supporting Information Figure S2), implying that these SAMs block access to the Au surface completely. Weak redox signals for S1−3 with wide peak separation suggest less perfect/dense SAM (Supporting Information Figure S2). Figure 2 shows the cyclic voltammograms for TATA 1−3 SAMs on Au. The Fe(II)/Fe(III) redox signal can still be seen for the TATA SAMs, but with strongly reduced intensity and increasing peak separation with increasing wire length. As compared to the thiol SAMs, this indicates that solvent gets considerably closer to the Au surface, in agreement with the 14870 dx.doi.org/10.1021/la504056v | Langmuir 2014, 30, 14868−14876 Langmuir Article (Supporting Information Figure S8) than their TATA analogues with the same length of phenylene wire. While comparing the same terphenyl wire with thiol (S), bis-thiol (SS), and TATA anchors, the resistance measured by CP-AFM gave the trend: TATA 3 > S3 > SS3. I−V correlations of S2 and S3 were also recorded by ECSTM, Figure S7. To decouple the intermolecular tunneling and make a direct comparison between the conductance of TATA and thiol anchored molecules, the phenylene wires were imbedded in a matrix of SAMs of nonconductive alkanethiols.48 Individual S2 and S3 molecules were easily identified by ECSTM in the mixed SAMs. These EC-STM results (Table 1) showed a trend similar to that of the CP-AFM data with respect to relative conductance. Most of the I−V curves were further symmetric with respect to bias. The exponential increase of resistance with increasing wire length for short molecular wires is consistent with an offresonant tunneling process, which can be described by eq 1:5,51,52 dilute nature of the TATA SAMs, in part imposed by the octyl side chains on the TATA platform.27−29 Vertical Tunnel Junctions and Transport Properties of the TATA SAMs. To evaluate the electronic transport properties of TATA anchored molecular wires, molecular junctions based on the TATA SAMs were characterized by both conducting probe-atomic force microscopy (CP-AFM) in air and EC-STM in solution. CP-AFM46,47 and EC-STM48−50 can efficiently record the charge transport properties of individual or small groups of molecules in the SAMs, respectively. Figure 3 shows typical I−V curves of the molecular R = R 0 exp(βl) (1) R is the junction resistance, R0 is the effective contact resistance, β is the tunneling attenuation factor, and l is the molecular wire length. Here, we defined l only by the length of the phenyl wire (from the top C to bottom C of the oligophenylene, Scheme 1d). We consider the length of the anchoring groups (S or TATA) an intrinsic property of the moiety that is best represented in the contact resistance R0, to allow a reasonable comparison between different binding groups. The fit of resistance versus molecular length in the SAM junctions (Figure 4) was grouped according to the different anchoring groups and measuring methods. The β and R0 values calculated from the fittings are summarized in Table 2. The β value depends on the intrinsic conjugation in the wire backbone and the alignment of the junction Fermi level with the molecular frontier orbitals.51,53 As shown in Figure 4 and Table 2, CP-AFM results of the TATA and SS series (and ECSTM result of the S series) show nearly the same value of β at about 0.6 Å−1. This value is similar to the result (0.61 ± 0.1 Å−1) of Rampi and Whitesides for thiol anchored oligophenylene SAMs in mercury drop junctions.54 EC-STM for the TATAs showed a β value of 0.79 Å−1, which is slightly higher but still in reasonable agreement with the values from the other methods. The similar β values for phenylene wires with thiol and TATA binding groups support that the TATA group indeed may be used as an anchor to disclose the nature of molecular wires. At the same time, the TATA platform anchor can control the orientation intermolecular distance between parallel wires in the junctions and ascertain that wire-to-wire tunneling is blocked. R0 of the TATAs obtained by CP-AFM is higher than that of the SS series. In part this can be ascribed to the very different Figure 3. Typical I−V curves of the molecular junctions based on TATA 1−3 obtained by (a) CP-AFM and (b) EC-STM. junctions based on TATA 1−3. The resistances (each value is the average obtained from at least 50 curves and determined over a small voltage range of ±0.1 V) are summarized in Table 1. The CP-AFM measured resistance for each TATA was between 1 and 2 orders lower than the values obtained from STM. This reflects most likely the larger number of molecules (about 10−60) present in the CP-AFM junction. The bis-thiol (SS1−3) and thiol (S1−3) series analogues were characterized by CP-AFM and EC-STM for comparison. Unfortunately, we could not get high-quality S1 and S2 SAMs (checked by CV, see the Supporting Information). These results are also summarized in Table 1. All of the SS1−3 compounds had lower resistance and more dense histograms Table 1. Summary of the Average Resistances (Measured by CP-AFM and EC-STM) from Molecular Junctions Based on the SAMs resistance (Ω) CP-AFM STM a TATA 1 TATA 2 TATA 3 SS1 SS2 SS3 5.72 × 106 7.05 × 107 1.00 × 108 4.34 × 109 1.13 × 109 7.14 × 1010 1.69 × 106 3.22 × 107 2.82 × 108 S2 S3 4.01 × 108a 3.44 × 108 6.67 × 109a Measured as guests in an alkanethiol host SAM. 14871 dx.doi.org/10.1021/la504056v | Langmuir 2014, 30, 14868−14876 Langmuir Article Figure 4. Semilog plot of resistance versus molecular wire length from experimental and calculated values. Table 2. Summary of the β Values and R0 (Measured by CPAFM and EC-STM) from Molecular Junctions Based on the SAMsa TATA 1−3 STM TATA 1−3 CP-AFM SS1−3 CP-AFM S2−3 STM a β (Å−1) R0 (Ω) 0.79 ± 0.09 0.61 ± 0.03 0.59 ± 0.05 0.65 (9.26 ± 4.85) × 106 (1.10 ± 0.23) × 106 (3.64 ± 1.21) × 105 3.87 × 106 Figure 5. Electrochemical reductive desorption voltammograms for (a) S1−3 and (d) TATA 1−3 (solution pH, 12.9; scan rate, 10 mV/s). most molecules in thiol SAMs desorb at a single potential reflecting a cooperative effect (as the sharpness of the desorption peak is positively related to the pair interaction energy between a molecule and its closest neighbors60). As distinct from in thiol SAMs, desorption of TATA molecules is a gradual process starting when the applied potential has reached −0.95 V vs SCE. This relatively higher value indicates that TATA molecules can tolerate a much wider electrochemical potential window than their thiol-anchored analogues, especially for shorter wires. The absence of length and cooperativity effects in TATA desorption accords with the open structures of the SAM dictated by the large platform footprint of TATA. A more detailed study of the electrochemical desorption of TATA versus thiols was conducted for S2 and TATA 2 SAMs, which have the same biphenyl wire. Time-dependent reductive desorption for these two SAMs is shown in Supporting Information Figure S3. A fixed potential was first applied to the SAM-modified Au electrode for a period of time. The electrode was then cleaned and transferred to the K3[Fe(CN)6]/ K4[Fe(CN)6] solution for CV. The relative current density for the Fe(II)/Fe(III) redox signal is taken as a measure of the fraction of adsorbate molecules remaining on the surface. Almost all molecules had desorbed from the surface for S2 after 1 h at ED = −0.75 V vs SCE (desorption peak in Figure 5) (see Supporting Information Figure S3b). For TATA 2, several different potential values (P2−P6, Supporting Information Figure S3c) were used to check the stability as no clear desorption peak appeared. The CV results showed that TATA molecules still cover the Au surface efficiently even after a potential of −0.8 V vs SCE had been applied for 1 h (Supporting Information Figure S3d). More negative potential and longer time were needed to detach the TATA from Au as compared to thiol. We therefore concluded that TATA had notably higher stability than thiol, which is most commonly Errors from the linear fit. density of wires in the two types of SAM, giving rise to different numbers of parallel molecules in the junctions. The apparent R0 of single TATA can be more precisely calculated as 9.26 × 106 Ω from the EC-STM data. This contact resistance is only slightly larger than for thiols (S2−3, 3.87 × 106 Ω). Tao and associates reported the R0 of bis-thiol and diamine anchoring groups as 5.87 × 104 Ω and 4.04 × 106 Ω, respectively, in junctions where the single molecule was bound to both a substrate and a Au STM tip.55 TATA 1−3 only had one end anchored to substrate, but still showed R0 comparable to those of the diamine and monothiol. The TATA platform thus shows surprisingly good transport properties, which may be a result of both the large π-system and the N atoms bonding to metal substrate. Stability of the SAMs. The strength of the bond between anchor and substrate is very important for the stability of SAMs.27−29 We investigated the stability of TATA 1−3 SAMs, and of S1−3 SAMs for comparison by electrochemical reductive desorption.56−60 Figure 5a shows the potentials required to desorb SAMs of S1−3. It is clearly seen that the stability of thiol-based SAMs depends on the length of the phenylene wires. As the number of phenyl rings increases, the desorption peak Edesorption (ED) is shifted by about 150 mV per added phenyl ring toward negative potentials (ED = −0.59, −0.75, and −0.9 V vs SCE for S1−3). This indicates that higher energy is needed to desorb the thicker SAM. SAMs of the longer molecules are thus more stable and form better SAMs in the series of S1−3, emphasizing the significant role of intermolecular interactions in these dense SAMs.58 The desorption pattern is completely different for TATA 1−3, Figure 5b. Desorption of TATA SAMs is independent of the wire length, and all three curves look quite similar. Moreover, 14872 dx.doi.org/10.1021/la504056v | Langmuir 2014, 30, 14868−14876 Langmuir Article used as anchor in SAMs. This advantage means that TATA is very suitable for SAMs intended for electrochemical operation. Quantum Chemical Calculations. Theoretical modeling of single molecules in junctions can give valuable insight into the nature of their conductive properties. Measurements of the thermoelectric properties of oligo-phenylenes with thiol binding groups predict that transport is mediated by the highest occupied molecular orbital (HOMO), with distinct contributions from the sulfur atoms.16,61 Large spreads in the measured conductance of thiol-based molecules indicate that the transport properties of these molecules depend intricately on the exact geometry at this interface.62 The conductive properties of TATAs have not been investigated previously. We compare here the characteristics of a single TATA with those of its corresponding single monoand bis-thiols. We consider adsorption on a flat Au fcc(111) surface as a model system for the thiols and assume that the sulfur atoms are bound in a hollow-site position with the wire perpendicular to the surface as is the case for the corresponding TATAs. This allows for a direct comparison of the different types of binding groups. We optimized the geometry of the bare TATA platform on the Au(111) electrode as outlined in the Supporting Information and found a well-defined binding geometry on the surface where the characteristic inverted umbrella shape of the TATA is flattened somewhat upon adsorption. We investigated the conductive properties at zero bias in the coherent tunneling regime where the current, I, can be calculated using the Landauer formula in eq 2: I= −2e h Figure 6. Calculated transmission and projected density of states (PDOS) using DFT for (a) S1−3 and (b) TATA 1−3. The PDOS for the different molecules are displaced vertically for clarity. ∞ ∫−∞ [fL (E − μL ) − fR (E − μR )]T(E) dE (2) HOMO peak has distinct sulfur character. Contributions from wire atoms dominate the PDOS around the HOMO peak as the length of the wire increases. The transmission of the shortest TATA reveals a broad split HOMO peak. If we compare with the PDOS, we see that contributions to the density of states with distinct TATA character dominate the density of states in the energy range around this peak. As the length of the wire increases, a peak in the transmission directly associated with the wire shifts to dominate the value of the transmission near the Fermi energy. These calculations therefore indicate that the TATA platform contributes to the HOMO character of the transport in a way similar to thiols. It is well established that the positions of the HOMO and LUMO peaks in the transmission relative to the electrode Fermi energy are poorly described by density functional methods. This is due to the inadequate description of electron correlation effects and the interpretation of the Kohn−Sham eigenvalues as quasiparticle excitation energies.63 Although DFT usually predicts trends in bandgaps correctly, the magnitude of the bandgap is underestimated and the magnitude of the conductance therefore grossly overestimated. Here, we use the transmission to fit the conductance to the molecular length according to eq 1. This has previously been shown to yield reasonable results.64 Figure 7 shows β calculated by fitting the value of ln(h/(2e2T(E)) to the length of the bridging wire as shown in the inset for E = Ef. The attenuation factor of the TATAs is higher than those of the thiols in a wide range around the Fermi energy. This is equivalent to the statement that the transmission near the Fermi energy falls off faster with increasing length for the TATAs than the thiols. The calculated PDOS indicated that the “wire” character of the frontier orbital Here, e is the electronic charge and h is the Planck constant. f L and f R are the temperature and voltage-dependent Fermi functions, and μL and μR the electrochemical potentials of the left and right electrodes, respectively. The transmission, T(E), is calculated by eq 3: T (E) = Tr[ΓLGR ΓR G A ] (3) ΓL and ΓR are half the imaginary parts of the left and right electrode self-energies, respectively. GR and GA are the retarded and advanced Green’s functions of the scattering region. All of the matrixes in eq 3 depend on energy. The transmissions of the monothiols and the TATAs are shown in Figure 6. The transmissions of the bis-thiols exhibit characteristics similar to those of the monothiols and can be found in the Supporting Information, along with further details of the calculations. The calculated current at low bias only depends on the transmission in a narrow range around the Fermi energy. We plot the transmission over a wide energy range to illustrate which peaks contribute to the value of the transmission close to the Fermi energy. The thiol transmission is characterized by a broad band gap that decreases as the length of the wire increases. The density of states of the scattering region can be obtained directly from the retarded Green’s function in eq 3. By projecting the density of states onto the basis functions localized on either anchoring group or wire, we can see which parts of the molecule contribute to the electronic structure at a given energy. These projected density of states (PDOS) are shown in the top panels of Figure 6. For the shortest thiol, the PDOS onto the anchor and wire give equal contributions in the energy range around the HOMO peak. This indicates that the transmission at the 14873 dx.doi.org/10.1021/la504056v | Langmuir 2014, 30, 14868−14876 Langmuir Article Figure 8. Local transmission for TATA 1, calculated using gDFTB. The radii of the arrows indicate the magnitude of the local transmission divided by the total transmission. The color red (blue) indicates a positive (negative) contribution. An arrow is drawn only when the local transmission between a pair of atoms exceeds 0.5% of the total transmission. Left: Local transmission at the Fermi level. Right: Local transmission at E − Ef = 2.0 eV. exact binding geometry on Au and sets TATA apart from other small binding groups such as thiols and amines.66,67 Figure 7. Calculated attenuation factors for TATA 1−3, S1−3, and SS1−3 as a function of energy. The inset shows ln(Resistance) as a function of bridge length and the fit used to calculate the attenuation factor at the Fermi energy. The resistance is calculated as h/(2e2T(E)). ■ CONCLUSIONS We have addressed comprehensively the binding of a series of TATA molecules with variable-length aromatic wire units to gold surfaces using a combination of chemical synthesis, electrochemistry, CP-AFM, and EC-STM, supported by DFT computation of both structural and transport properties of the bound TATA molecules. Neither explicit computation of the transport properties nor the combination of multifarious experimental techniques with the theoretical approach has been addressed before. The results here showed that TATA has great potential as a novel binding/anchoring group for molecular electronics. SAMs of TATA 1−3 were fabricated and characterized in detail. Vertical tunnel junctions based on TATA SAMs were characterized by both CP-AFM and EC-STM, and the results showed identical trends. Analogous bis-thiol (SS1−3) and thiol (S1−3) bound wires were studied as a comparison. Similar β values indicate that TATA is a good anchor for probing the intrinsic molecular wire properties as compared to thiol anchoring groups. Despite the presence of a central saturated carbon atom at the TATA bridge head and noncovalent binding of the TATA unit, the contact resistance R0 was found to be surprisingly low. Theoretically calculated transmission functions of the TATAs and transport properties of TATAbased molecules were analyzed and found to support broadly the experimental data. Projections of the density of states onto the individual atoms further showed that the TATA platform contributed to the HOMO character of the transport in a fashion similar to that for the thiols, leading to the surprising similarity in the contact resistance for these two groups. Notably, the noncovalent binding of the TATA platform to Au is thus comparable to or even better than thiols, and independent of the attached wire segment. In conjunction with the possibility of arranging molecular wires strictly normal to the substrate and yet in a dilute, noninteracting monolayer, these features offer several exciting potential advantages, including suppression of in-plane transport, access for reagents to functional wires, and free volume for electromechanically operated switches. The overarching objective of the study was to explore to which extent the large TATA “footprint” would be a useful novel type strongly bound platform for molecular tunneling junctions with the molecular wire fragments spaced out on the (HOMO) responsible for transport increases with increasing length. We can therefore associate the larger calculated attenuation factor of the TATAs when compared to the thiols with wire contribution to the current that falls of faster with length for TATAs than for thiols. We can explain this effect by the narrower “wire peaks” in the transmission for the TATAs because the wire is less electronically coupled to the electrodes for the TATAs than for the thiols. The β values at the Fermi energy for S1−3, SS1−3, and TATA 1−3 were calculated to be 0.30, 0.29, and 0.35 Å−1, respectively. The corresponding contact resistances were 3.97 × 105, 1.94 × 104, and 1.84 × 107 Ω. These results show the same trend as the experimental data. That is, despite all of the differences between these two binding groups, TATA contributes with similar “gateway states” to those found in thiols. As these latter states are the orbitals that dominate the low bias conductance, the transport properties, including the contact resistance, are not so different from what might have been anticipated from the chemical structures alone. The interface between electrodes and TATA-substituted molecular wires provides an interesting tunneling barrier that has not previously been investigated using a theoretical approach. A local description of the tunneling path of the electron as described elsewhere45 can provide intuitive insight into the nature of the transport. In this description, the total transmission is written as a sum of contributions between the atoms in the scattering region. Conservation of current then implies that the sum of contributions between atoms on opposite sides of any imaginary plane perpendicular to the direction of current is constant. Plotting the contributions as arrows, one can get a sense of the tunneling path through the molecule. Around the Fermi energy, the local currents reveal a regime where the tunneling path is through the central saturated carbon atom of the TATA. This is shown in Figure 8 (left). At other energies, the tunneling path shows direct injection from the TATA to the first unsaturated carbon in the wire, as shown in Figure 8 (right). This illustrates a tunneling pathway that bypasses the central saturated carbon atom, which is usually associated with poor conductive properties.65 Figure 8 also reveals that no single path dominates the interface between electrode and the TATA molecule. This suggests that the transport properties of TATA are relatively insensitive to the 14874 dx.doi.org/10.1021/la504056v | Langmuir 2014, 30, 14868−14876 Langmuir Article transistor of a single organic molecule with access to several redox states. Nature 2003, 425, 698−701. (11) Li, T.; Hauptmann, J. R.; Wei, Z.; Petersen, S.; Bovet, N.; Vosch, T.; Nygård, J.; Hu, W.; Liu, Y.; Bjørnholm, T.; Nørgaard, K.; Laursen, B. W. Solution-processed ultrathin chemically derived graphene films as soft top contacts for solid-state molecular electronic junctions. Adv. Mater. 2012, 24, 1333−1339. (12) Li, T.; Jevric, M.; Hauptmann, J. R.; Hviid, R.; Wei, Z.; Wang, R.; Reeler, N. E. A.; Thyrhaug, E.; Petersen, S.; Meyer, J. A. S.; Bovet, N.; Vosch, T.; Nygård, J.; Qiu, X.; Hu, W.; Liu, Y.; Solomon, G. C.; Kjaergaard, H. G.; Bjørnholm, T.; Nielsen, M. B.; Laursen, B. W.; Nørgaard, K. Ultrathin reduced graphene oxide films as transparent top-contacts for light switchable solid-state molecular junctions. Adv. Mater. 2013, 25, 4164−4170. (13) von Wrochem, F.; Gao, D.; Scholz, F.; Nothofer, H.-G.; Nelles, G.; Wessels, J. M. Efficient electronic coupling and improved stability with dithiocarbamate-based molecular junctions. Nat. Nanotechnol. 2010, 5, 618−624. (14) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Self-assembled monolayers of thiolates on metals as a form of nanotechnology. Chem. Rev. 2005, 105, 1103−1170. (15) Smith, R. K.; Lewis, P. A.; Weiss, P. S. Patterning self-assembled monolayers. Prog. Surf. Sci. 2004, 75, 1−68. (16) Tan, A.; Balachandran, J.; Sadat, S.; Gavini, V.; Dunietz, B. D.; Jang, S.-Y.; Reddy, P. Effect of length and contact chemistry on the electronic structure and thermoelectric properties of molecular junctions. J. Am. Chem. Soc. 2011, 133, 8838−8841. (17) Martin, C. A.; Ding, D.; Sørensen, J. K.; Bjørnholm, T.; van Ruitenbeek, J. M.; van der Zant, H. S. J. Fullerene-based anchoring groups for molecular electronics. J. Am. Chem. Soc. 2008, 130, 13198− 13199. (18) Hong, W.; Manrique, D. Z.; Moreno-García, P.; Gulcur, M.; Mishchenko, A.; Lambert, C. J.; Bryce, M. R.; Wandlowski, T. Single molecular conductance of tolanes: Experimental and theoretical study on the junction evolution dependent on the anchoring group. J. Am. Chem. Soc. 2012, 134, 2292−2304. (19) Song, H.; Lee, H.; Lee, T. Intermolecular chain-to-chain tunneling in metal−alkanethiol−metal junctions. J. Am. Chem. Soc. 2007, 129, 3806−3807. (20) Okabayashi, N.; Paulsson, M.; Ueba, H.; Konda, Y.; Komeda, T. Inelastic tunneling spectroscopy of alkanethiol molecules: Highresolution spectroscopy and theoretical simulations. Phys. Rev. Lett. 2010, 104, 077801. (21) Fracasso, D.; Valkenier, H.; Hummelen, J. C.; Solomon, G. C.; Chiechi, R. C. Evidence for quantum interference in SAMs of arylethynylene thiolates in tunneling junctions with eutectic Ga−In (EGaIn) top-contacts. J. Am. Chem. Soc. 2011, 133, 9556−9563. (22) Laursen, B. W.; Krebs, F. C. Synthesis of a triazatriangulenium salt. Angew. Chem., Int. Ed. 2000, 39, 3432−3434. (23) Laursen, B. W.; Krebs, F. C. Synthesis, structure, and properties of azatriangulenium salts. Chem.Eur. J. 2001, 7, 1773−1783. (24) Lofthagen, M.; Chadha, R.; Siegel, J. S. Synthesis, structures, and dynamics of a macrocyclophane. J. Am. Chem. Soc. 1991, 113, 8785− 8790. (25) Laursen, B. W.; Krebs, F. C.; Nielsen, M. F.; Bechgaard, K.; Christensen, J. B.; Harrit, N. 2,6,10-Tris(dialkylamino)trioxatriangulenium ions. Synthesis, structure, and properties of exceptionally stable carbenium ions. J. Am. Chem. Soc. 1998, 120, 12255−12263. (26) Bosson, J.; Gouin, J.; Lacour, J. Cationic triangulenes and helicenes: synthesis, chemical stability, optical properties and extended applications of these unusual dyes. Chem. Soc. Rev. 2014, 43, 2824− 2840. (27) Baisch, B.; Raffa, D.; Jung, U.; Magnussen, O. M.; Nicolas, C.; Lacour, J.; Kubitschke, J.; Herges, R. Mounting freestanding molecular functions onto surfaces: The platform approach. J. Am. Chem. Soc. 2009, 131, 442−443. (28) Kuhn, S.; Baisch, B.; Jung, U.; Johannsen, T.; Kubitschke, J.; Herges, R.; Magnussen, O. Self-assembly of triazatriangulenium-based surface. This would at the same time block conceivable interfering lateral tunneling pathways and enable monitoring chemical reactivity of the bound TATA-based molecular wires. TATA thus constitutes an exciting new binding group for molecular electronics with both geometric and electronic structural advantages. ■ ASSOCIATED CONTENT S Supporting Information * Experimental details about the synthesis, EC-STM, cyclic voltammetry, and water contact angles of the SAMs, UV−vis absorption spectra, more CP-AFM and EC-STM I−V curves and resistance histograms of the molecular junctions, calculated junction structures for all molecules, local currents throughout the bandgap for TATA, transmission, and PDOS for SS1−3. This material is available free of charge via the Internet at http://pubs.acs.org. ■ AUTHOR INFORMATION Corresponding Authors *E-mail: [email protected]. *E-mail: [email protected]. Author Contributions ⊥ These authors contributed equally. Notes The authors declare no competing financial interest. ■ ACKNOWLEDGMENTS The work was supported by the Danish-Chinese Center for Molecular Nanoelectronics funded by the Danish National Research Foundation, the European Union seventh Framework Programme (FP7/2007-2013) under the grant agreement no. 270369 (“ELFOS”), and from the European Research Council/ ERC Grant agreement no. 258806. We are grateful to Associate Prof. Jingdong Zhang at Technical University of Denmark for help with the EC-STM measurements. ■ REFERENCES (1) Aviram, A.; Ratner, M. A. Molecular rectifiers. Chem. Phys. Lett. 1974, 29, 277−283. (2) Moth-Poulsen, K.; Bjørnholm, T. Molecular electronics with single molecules in solid-state devices. Nat. Nanotechnol. 2009, 4, 551− 556. (3) Akkerman, H. B.; Boer, B. d. Electrical conduction through single molecules and self-assembled monolayers. J. Phys.: Condens. Matter 2008, 20, 013001. (4) Tao, N. J. Electron transport in molecular junctions. Nat. Nanotechnol. 2006, 1, 173−181. (5) Song, H.; Reed, M. A.; Lee, T. Single molecule electronic devices. Adv. Mater. 2011, 23, 1583−1608. (6) Halik, M.; Hirsch, A. The potential of molecular self-assembled monolayers in organic electronic devices. Adv. Mater. 2011, 23, 2689− 2695. (7) Li, T.; Hu, W.; Zhu, D. Nanogap electrodes. Adv. Mater. 2010, 22, 286−300. (8) Osorio, E. A.; Bjørnholm, T.; Lehn, J.-M.; Ruben, M.; van der Zant, H. S. J. Single-molecule transport in three-terminal devices. J. Phys.: Condens. Matter 2008, 20, 374121. (9) Hines, T.; Diez-Perez, I.; Hihath, J.; Liu, H.; Wang, Z.-S.; Zhao, J.; Zhou, G.; Müllen, K.; Tao, N. Transition from tunneling to hopping in single molecular junctions by measuring length and temperature dependence. J. Am. Chem. Soc. 2010, 132, 11658−11664. (10) Kubatkin, S.; Danilov, A.; Hjort, M.; Cornil, J.; Brédas, J.-L.; Stuhr-Hansen, N.; Hedegård, P.; Bjørnholm, T. Single-electron 14875 dx.doi.org/10.1021/la504056v | Langmuir 2014, 30, 14868−14876 Langmuir Article functional adlayers on Au(111) surfaces. Phys. Chem. Chem. Phys. 2010, 12, 4481−4487. (29) Jung, U.; Kuhn, S.; Cornelissen, U.; Tuczek, F.; Strunskus, T.; Zaporojtchenko, V.; Kubitschke, J.; Herges, R.; Magnussen, O. Azobenzene-containing triazatriangulenium adlayers on Au(111): Structural and spectroscopic characterization. Langmuir 2011, 27, 5899−5908. (30) Kubitschke, J.; Näther, C.; Herges, R. Synthesis of functionalized triazatriangulenes for application in photo-switchable self-assembled monolayers. Eur. J. Org. Chem. 2010, 5041−5055. (31) Jung, U.; Schütt, C.; Filinova, O.; Kubitschke, J.; Herges, R.; Magnussen, O. Photoswitching of azobenzene-functionalized molecular platforms on Au surfaces. J. Phys. Chem. C 2012, 116, 25943− 25948. (32) Hauptmann, N.; Scheil, K.; Gopakumar, T. G.; Otte, F. L.; Schütt, C.; Herges, R.; Berndt, R. Surface control of alkyl chain conformations and 2D chiral amplification. J. Am. Chem. Soc. 2013, 135, 8814−8817. (33) Otte, F. L.; Lemke, S.; Schütt, C.; Krekiehn, N. R.; Jung, U.; Magnussen, O. M.; Herges, R. Ordered monolayers of free-standing porphyrins on gold. J. Am. Chem. Soc. 2014, 136, 11248−11251. (34) Tellkamp, T.; Shen, J.; Okamoto, Y.; Herges, R. Diazocines on molecular platforms. Eur. J. Org. Chem. 2014, 2014, 5456−5461. (35) Zhang, J.; Kuznetsov, A. M.; Medvedev, I. G.; Chi, Q.; Albrecht, T.; Jensen, P. S.; Ulstrup, J. Single-molecule electron transfer in electrochemical environments. Chem. Rev. 2008, 108, 2737−2791. (36) Zhang, J.; Welinder, A. C.; Chi, Q.; Ulstrup, J. Electrochemically controlled self-assembled monolayers characterized with molecular and sub-molecular resolution. Phys. Chem. Chem. Phys. 2011, 13, 5526−5545. (37) Bahn, S. R.; Jacobsen, K. W. An object-oriented scripting interface to a legacy electronic structure code. Comput. Sci. Eng. 2002, 4, 56−66. (38) Mortensen, J. J.; Hansen, L. B.; Jacobsen, K. W. Real-space grid implementation of the projector augmented wave method. Phys. Rev. B 2005, 71, 035109. (39) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 1996, 77, 3865−3868. (40) Dion, M.; Rydberg, H.; Schröder, E.; Langreth, D. C.; Lundqvist, B. I. Van der Waals density functional for general geometries. Phys. Rev. Lett. 2004, 92, 246401. (41) Cooper, V. R. Van der Waals density functional: An appropriate exchange functional. Phys. Rev. B 2010, 81, 161104. (42) Pecchia, A.; Carlo, A. D. Atomistic theory of transport in organic and inorganic nanostructures. Rep. Prog. Phys. 2004, 67, 1497−1561. (43) Elstner, M.; Porezag, D.; Jungnickel, G.; Elsner, J.; Haugk, M.; Frauenheim, T.; Suhai, S.; Seifert, G. Self-consistent-charge densityfunctional tight-binding method for simulations of complex materials properties. Phys. Rev. B 1998, 58, 7260. (44) Frauenheim, T.; Seifert, G.; Elsterner, M.; Hajnal, Z.; Jungnickel, G.; Porezag, D.; Suhai, S.; Scholz, R. A self-consistent charge densityfunctional based tight-binding method for predictive materials simulations in physics, chemistry and biology. Phys. Status Solidi B 2000, 217, 41−62. (45) Solomon, G. C.; Herrmann, C.; Hansen, T.; Mujica, V.; Ratner, M. A. Exploring local currents in molecular junctions. Nat. Chem. 2010, 2, 223−228. (46) Mativetsky, J. M.; Pace, G.; Elbing, M.; Rampi, M. A.; Mayor, M.; Samorì, P. Azobenzenes as light-controlled molecular electronic switches in nanoscale metal−molecule−metal junctions. J. Am. Chem. Soc. 2008, 130, 9192−9193. (47) Wei, Z.; Li, T.; Jennum, K.; Santella, M.; Bovet, N.; Hu, W.; Nielsen, M. B.; Bjørnholm, T.; Solomon, G. C.; Laursen, B. W.; Nørgaard, K. Molecular junctions based on SAMs of cruciform oligo(phenylene ethynylene)s. Langmuir 2012, 28, 4016−4023. (48) Bumm, L. A.; Arnold, J. J.; Cygan, M. T.; Dunbar, T. D.; Burgin, T. P.; Jones, L.; Allara, D. L.; Tour, J. M.; Weiss, P. S. Are single molecular wires conducting? Science 1996, 271, 1705−1707. (49) Xu, B.; Tao, N. J. Measurement of single-molecule resistance by repeated formation of molecular junctions. Science 2003, 301, 1221− 1223. (50) Darwish, N.; Díez-Pérez, I.; Da Silva, P.; Tao, N.; Gooding, J. J.; Paddon-Row, M. N. Observation of electrochemically controlled quantum interference in a single anthraquinone-based norbornylogous bridge molecule. Angew. Chem., Int. Ed. 2012, 51, 3203−3206. (51) Choi, S. H.; Frisbie, C. D. Enhanced hopping conductivity in low band gap donor−acceptor molecular wires up to 20 nm in length. J. Am. Chem. Soc. 2010, 132, 16191−16201. (52) Choi, S. H.; Kim, B.; Frisbie, C. D. Electrical resistance of long conjugated molecular wires. Science 2008, 320, 1482−1486. (53) Salomon, A.; Cahen, D.; Lindsay, S.; Tomfohr, J.; Engelkes, V. B.; Frisbie, C. D. Comparison of electronic transport measurements on organic molecules. Adv. Mater. 2003, 15, 1881−1890. (54) Holmlin, R. E.; Haag, R.; Chabinyc, M. L.; Ismagilov, R. F.; Cohen, A. E.; Terfort, A.; Rampi, M. A.; Whitesides, G. M. Electron transport through thin organic films in metal−insulator−metal junctions based on self-assembled monolayers. J. Am. Chem. Soc. 2001, 123, 5075−5085. (55) Chen, F.; Li, X.; Hihath, J.; Huang, Z.; Tao, N. Effect of anchoring groups on single-molecule conductance: Comparative study of thiol-, amine-, and carboxylic-acid-terminated molecules. J. Am. Chem. Soc. 2006, 128, 15874−15881. (56) Kakiuchi, T.; Usui, H.; Hobara, D.; Yamamoto, M. Voltammetric properties of the reductive desorption of alkanethiol self-assembled monolayers from a metal surface. Langmuir 2002, 18, 5231−5238. (57) Sumi, T.; Uosaki, K. Electrochemical oxidative formation and reductive desorption of a self-assembled monolayer of decanethiol on a Au(111) surface in KOH ethanol solution. J. Phys. Chem. B 2004, 108, 6422−6428. (58) Munakata, H.; Oyamatsu, D.; Kuwabata, S. Effects of ωfunctional groups on pH-dependent reductive desorption of alkanethiol self-assembled monolayers. Langmuir 2004, 20, 10123− 10128. (59) Marmisollé, W. A.; Capdevila, D. A.; de la Llave, E.; Williams, F. J.; Murgida, D. H. Self-assembled monolayers of NH2-terminated thiolates: Order, pKa, and specific adsorption. Langmuir 2013, 29, 5351−5359. (60) Aoki, K.; Kakiuchi, T. Probability theory of desorption kinetics of self-assembled alkanethiols stabilized with pair interaction. J. Electroanal. Chem. 1998, 452, 187−192. (61) Malen, J. A.; Doak, P.; Baheti, K.; Tilley, T. D.; Segalman, R. A.; Majumdar, A. Identifying the length dependence of orbital alignment and contact coupling in molecular heterojunctions. Nano Lett. 2009, 9, 1164−1169. (62) Venkataraman, L.; Klare, J. E.; Tam, I. W.; Nuckolls, C.; Hybertsen, M. S.; Steigerwald, M. L. Single-molecule circuits with welldefined molecular conductance. Nano Lett. 2006, 6, 458−462. (63) Peng, G.; Strange, M.; Thygesen, K. S.; Mavrikakis, M. Conductance of conjugated molecular wires: Length dependence, anchoring groups, and band alignment. J. Phys. Chem. C 2009, 113, 20967−20973. (64) Quek, S. Y.; Choi, H. J.; Louie, S. G.; Neaton, J. B. Length dependence of conductance in aromatic single-molecule junctions. Nano Lett. 2009, 9, 3949−3953. (65) Wold, D. J.; Haag, R.; Rampi, M. A.; Frisbie, C. D. Distance dependence of electron tunneling through self-assembled monolayers measured by conducting probe atomic force microscopy: Unsaturated versus saturated molecular junctions. J. Phys. Chem. B 2002, 106, 2813−2816. (66) Ulrich, J.; Esrail, D.; Pontius, W.; Venkataraman, L.; Millar, D.; Doerrer, L. H. Variability of conductance in molecular junctions. J. Phys. Chem. B 2006, 110, 2462−2466. (67) Quek, S. Y.; Kamenetska, M.; Steigerwald, M. L.; Choi, H. J.; Louie, S. G.; Hybertsen, M. S.; Neaton, J. B.; Venkataraman, L. Mechanically controlled binary conductance switching of a singlemolecule junction. Nat. Nanotechnol. 2009, 4, 230−234. 14876 dx.doi.org/10.1021/la504056v | Langmuir 2014, 30, 14868−14876