Role of Molecular Dipoles in Charge Transport across Large Area

Article Cite This: J. Phys. Chem. C XXXX, XXX, XXX-XXX

pubs.acs.org/JPCC

Role of Molecular Dipoles in Charge Transport across Large Area Molecular Junctions Delineated Using Isomorphic Self-Assembled Monolayers Jiahao Chen,†,§ Symon Gathiaka,‡ Zhengjia Wang,† and Martin Thuo*,†,§ †

Department of Materials Science and Engineering, Iowa State University, Hoover Hall 2220, Ames, Iowa 50011, United States School of Pharmacy and Pharmaceutical Science, University of California, San Diego, 9500 Gilman Drive, MC 0657, La Jolla, California 92093-0657, United States § Micro-electronic Research Center, Iowa State University, 133 Applied Sciences Complex I, 1925 Scholl Road, Ames, Iowa 50011, United States ‡

S Supporting Information *

ABSTRACT: Delineating the role of dipoles in large area junctions that are based on self-assembled monolayers (SAMs) is challenging due to molecular tilt, surface defects, and interchain coupling among other features. To mitigate SAM-based effects in study of dipoles, we investigated tunneling rates across carboranesisostructural molecules that orient along the surface normal on Au (but bear different dipole moments) without changing the thickness, packing density, or morphology of the SAM. Using the Au-SAM//Ga2O3-EGaIn junction (where “//” = physisorption, “−” = chemisorption, and EGaIn is eutectic gallium−indium), we observe that molecules with dipole moments oriented along the surface normal (with dipole moment, p = 4.1D for both M9 and 1O2) gave lower currents than when the dipole is orthogonal (p = 1.1 D, M1) at ±0.5 V applied bias. Similarly, from transition voltage spectroscopy, the transition voltages, VT (volt), are significantly different. (0.5, 0.43, and 0.4 V for M1, M9, and 1O2, respectively). We infer that the magnitude and direction of a dipole moments significantly affect the rate of charge transport across large area junctions with Δ log|J| ≅ 0.4 per Debye. This difference is largely due to effect of the dipole moment on the molecule-electrode coupling strength, Γ, hence effect of dipoles is likely to manifest in the contact resistance, Jo, although in conformational flexible molecules field-induced effects are expected.

■

INTRODUCTION Advances in processing capabilities in electronic devices through top-down approaches is nearing a critical size-limit, that needs to be overcome if the growth trajectory (Moore’s law) is to be maintained.1 Molecular electronics has emerged as a potential bottom-up alternative, even though the original promise of wave function engineering through molecular synthesis for tunable single molecule electronic components is still an ongoing endeavor.2 Understanding charge transport through unimolecular systems bears an inherent fabrication challenge but the so-called large area junctions have emerged as one of the more reliable platforms in terms of yield of working device, data reproducibility, and stability.3−6 Fundamental relations between molecular properties and charge transport behavior, however, need to be delineated for functional unimolecular devices to be realized. Among these, decoupling stereostructural effects (e.g., barrier width, conformations, phase-evolution) from electronic (e.g., dipoles, band broadening, interfacial coupling, barrier height) molecular properties is vital. Adoption of self-assembled monolayers (SAMs) in large area junctions, viewed as a series of resistors (Figure 1a), for © XXXX American Chemical Society

example, fails to decouple molecular dipoles from steric orientation of the molecules making up the SAM. The uncoupled effect of molecular dipole on charge transport remains unresolved in part due to challenges inherent in pinning a vector quantity on a dynamic (rotating/vibrating) and/or canted SAM system(s). In single molecule junctions, it has been shown that direction of embedded dipoles significantly affects rates of charge transport.7−10 As expected, current across junctions with parallel dipole moments, relative to electron flow, are much larger than across analogous junctions with antiparallel dipole moments (Figure 1b).7 These observations can be attributed to molecular dipoles enhancing polarity at the electrodes through associated mirror charges and/or inductive effects. In SAM-based studies, however, such direct correlations have been a challenge. Thuo et al.11 and later Yoon et al.,12,13 for example, reported that introduction of amides and/or aromatic moieties with associated polarity Received: August 1, 2017 Revised: September 19, 2017 Published: October 13, 2017 A

DOI: 10.1021/acs.jpcc.7b07634 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

Figure 1. (a) Schematic illustration of a molecular junction with EGaIn as top electrode. The junction is analogous to equivalent circuit with a SAM resistance and contact resistance. (b) Schematics demonstrating the effect of dipole on charge transport behavior, raising the question whether the tunneling rates are dependent on direction and/or magnitude of the dipoles. Structure and dipole of carboranethiol molecules and SAMs formed with these molecules. (c) Calculated electrostatic potential (ESP) of M1 (i), M9 (ii), and1O2 (iii), respectively. (d) Estimated dipole moment and direction of M1, M9, and 1O2 molecules. (e) Dipole direction of M1 SAM, M9 SAM, and 1O2 SAM. For brevity and clarity, all hydrogen atoms are ignored.

(dipole) changes in n-alkanethiol-based SAMs does not significantly affect tunneling rates. Kovalchuk et al.14 also reported that molecules with dipoles oriented in different directions gave comparable tunneling rates in conflict to the single molecule junction studies. Discrepancy between SAMbased junctions and the single molecule junctions calls for further evaluation for clarity. From simple Coulombic interactions, however, dipole moments should be influenced by an applied field and vice versa. Transition voltage, VT, a transition point from direct tunneling to Fowler−Nordheim tunneling due to barrier tilting (Figure S1),14−16 however, can be used to delineate any changes due to field-driven dipoledependent band alignments. As such, studies on role of dipoles in charge transport necessitates determination of VT. Besides VT, an experimental caveat in the above studies is that SAM structural and electronic properties were not decoupled, hence the inferences can be limited by lack of a detailed understanding of the nature of the SAM structure.17 We desired to decouple stereoelectronic (dipole moment) and steric (SAM tilt angle and packing density) effects on the rate of charge transport across SAMs and, hence, demonstrate the singular (isolated) effect of the dipole moment on charge transport behavior. We address this challenge by (i) designing SAMs that do not tilt upon forming a SAM, that is, molecules that orient along the surface normal, (ii) fabricating SAMs from isomeric molecules (structural isomers) with differences in both magnitude and direction of their dipole moments due to atomic arrangement but forming isomorphic SAMs (similar packing density, binding strength, thickness, or top interface structure). We chose molecules whose direction and magnitude of dipole can be tuned without significantly altering the stereostructural properties of the molecule, (iii) Addressing changes in barrier height by using a class of isostructural/isomorphic molecules where the elemental composition, dipole moment, or localization of frontier orbitals is not significantly perturbed from one structure to the other.

Carboranethiols, which are primarily boron-based molecules with a few carbons atoms interspersed between the boron atoms, are the candidates for this study. We choose three commercially available molecules, meta-1-carboranethiol (M1), meta-9-carboranethiol (M9), and 1,2-o-carboranethiol (1O2) (Figure 1c-e). Monolayers derived from M1 and M9 on Au are well studied,18−21 giving well-ordered SAMs with similar coverage, thickness, and surface properties (i.e., they are Isomorphic).21 Weiss and co-workers have shown that when these carboranethiols are attached to a surface (Au or Ag), the work function of the surface can be tuned, which is believed to be caused by the molecular dipole moment.18−20 1O2 forms SAMs with similar coverage and thickness as M1 and M9, albeit with a different bottom interfaces where the molecule is anchored via two thiols.20 The dipole moments between M9 and1O2 SAMs are antiparallel, while M1 is oriented orthogonal to the surface normal upon forming SAMs (Figure 1e).21

■

RESULTS Carboranethiol SAMs and Interface. The electrostatic potential (ESP) maps of molecules in free space, and the corresponding minimized structural energies (Figure 1c), were calculated using the Gaussian software package22 (see Supporting Information for details). The magnitude of the associated molecular dipoles is given alongside the molecule while the direction is illustrated by an arrow (Figure 1d). The calculated values are comparable to those in literature.20 The dipoles slightly change upon chemisorption onto the substrate with M1 aligning along the substrate surface (orthogonal to surface normal) while in M9 and 1O2 the dipoles align with the surface normal, albeit in opposite directions.20 The reported dipoles of the carboranes adsorbed on Au were reported to be 1.1 D, 4.1 D, and 4.1 D for M1, M9, and 1O2, respectively.20 The three carboranes have also been shown to structurally orient along the surface normal while retaining the same packing structure and density.20 In this article, we deploy these B


Article


Figure 2. Surface characterizations of substrates and the SAM. (a) Atomic force microscopies of bare template-stripped gold, AuTS and (b) AuTS surface bearing a SAM of carboranethiol, showing a difference in RMS rougnness (RRMS), power spectrum density (PSD) and bearing volume (BV). (c) Wide-angle X-ray diffraction pattern on bare AuTS and one with a SAM of carboranethiol showing that the metal is (111) oriented and that the faceting does not change upon deposition of the SAM.

Figure 3. Tunneling current across carboranethiolate junctions. (a) The tunneling data across all three molecules is summarized into heat maps. (b) Lag plot for tunneling rates across M1 to show the effect of outliers. (c) Calculated mean value for tunneling rates across the three molecules.

support inferences drawn from M1 and M9 if deployed in a comparative manner and not as an equal. Template-stripped gold (AuTS, roughness root-mean-square RRMS = 0.38 ± 0.04 nm) was used as substrate for device fabrication since; (i) carboranethiols oriented along the surface normal when bonded to Au,20,21 (ii) has lower surface roughness compared to analogous, and commonly used, AgTS (Figure 2a), (iii) AuTS surface is uniformly faceted along the (111) direction.23 Upon depostion of carboranethiol SAMs, the

isostructural molecules to form isomorphic SAMs in order to delineate the effect of molecular dipole on charge transport across EGaIn-based large area junctions, Au-SAM//Ga2O3EGaIn junction (where “//” = physisorption, “−” = chemisorption, and EGaIn is eutectic gallium−indium). The M1 and M9 have the same structure at the molecule//Ga2O3 interface but the 1O2//Ga2O3 is slightly different and is included solely to probe the role of dipole directionality. Despite the slight differences, we infer that 1O2 can be used to C


Article


Figure 4. Tunnelling current density across carboranethiolate SAMs and the effect of dipole. (a) Current density at −0.5 and +0.5 V for all three molecules. The dipole moment of the adsorbates are derived from ref 20. (b) Schematic of charge reorganization at the molecule/substrate interface and the resulting HOMO/LUMO shrinkage. (c) The energy level diagram (before contact without pining and band bending) at 0 bias and ±1 V, respectively. (d) The rectification ratio for all three molecules was with respect to the dipole (moment and direction). (e) The TVS for all three molecules at both positive (solid) and negative (open) bias.

(random noncorrelated data points). The outliers shown in the lag plots were removed from consideration (SI Figure S4), and the main portion of the data shows a strong autocorrelation (linear narrow distribution), as shown in Figure 3b, suggesting a nonrandom series of data. Gaussian mean values and standard deviations were also calculated before and after removing the outliers giving statistically indistinguishable values. These results confirm that previously used approach of fitting a Gaussian over tunneling data11,24−28 does not skew with outliers (SI Figure S5), and can be reliably deployed in analyzing charge tunneling characteristics as described in the LAJA method.11 Figure 3c summarizes the charge transport data obtained from the three carboranethiols. The average current densities obtained from M9 and 1O2 at ±0.5 V are comparable while M1 has a significantly higher current density. We observe that the distribution of the data correlates well with the charge density at the vertex of the molecule (Figure 1) with low-electron density at the vertex leading to significant broadening (1O2) in

grain boundaries become less distinguishable by AFM (Figure 2b) while the texture is slightly (but statistically significant) smoother than AuTS (RRMS 0.3 ± 0.03 nm, PSD = 0.077 ± 0.022 nm, p < 0.002), indicating a slight modification in the substrates morphology due to presence of the cage molecules. X-ray analysis confirms that our surfaces are dominated by the (111) orientation with or without a SAM (Figure 2c). Charge Transport Behavior Across the Junctions. All charge transport data were collected using a previously reported method.13,24,25 The raw data were then analyzed by LAJA26 and data summarized into heat-maps (Figure 3a), Gaussian-fits, and 3D histograms (see SI, Figure S2 and S3). We observed significant differences between charge transport properties across these molecules despite being of same barrier width.20 Carborane M9 showed narrowest distribution in log|J| with the distribution widths following the order; M9 < M1 ≪ 1O2. Peak broadening and significant outliers in the data obtained from M1 and 1O2 molecules (Figure 3a) necessitated investigation of autocorrelation through lag plots, to help identify the outliers D


Article

The Journal of Physical Chemistry C the distribution of J (A/cm2). Junctions derived from 1O2 were also observed to create “sticky” contacts which could be due to a stronger contact analogous to so-called “hard” contacts.26 Strong physisorption, however, does not necessary imply stronger electrical coupling since the less “sticky” M9 gave comparable average current densities.

delocalized combined orbitals between the metal and the molecules are thus disentangled to some degree, hence, decreasing interfacial coupling strength (Γ). The resulting broadened bandgap leads to narrowing of HOMO and LUMO states (Figure 4b),16,33 leading to a more “resistive” path across the molecule due to a decrease in energetically accessible/ coupled conduction states. The energy gain, ΔE, per molecule to disentangle the delocalized orbitals is proportional to the potential drop, ΔΦ, (eq 2),32 which is proportional to the effective molecular dipole along the surface normal (eq 3)31,34−36

■

DISCUSSION Effect of Molecular Dipoles on Tunneling Rates. Figure 4a compares the magnitude of the dipole moments of the three molecules with current density at +0.5 and −0.5 V. The molecule with a lower magnitude dipole moment (1.1 D), M1, gave the highest average current density while the higher dipole, M9 and 1O2 (4.1 D for both) gave statistically indistinguishable average current densities that were lower than observed with M1. Dipole moments, however, are characterized by both magnitude and direction. Considering direction of the dipole moments, higher current is observed for the dipole oriented orthogonal to the surface normal. With flat substrates and well-ordered SAMs, the surface normal aligns with the direction of charge injection in carboranethiolate SAMs. Molecule, M1, has a dipole moment orthogonal to the surface normal while M9 and 1O2 have dipole moments oriented along the direction of electron flow although in an antiparallel direction (Figurse 1e and 4a). The directions of the dipole moments relative to the flow of electrons are given as inserts above each value of obtained current density in Figure 4a. From these data, and assuming no significant field effects, we can deduce that the effect of magnitude of dipole moments on the rate of charge transport in these SAMs is Δ log|J| ≅ 0.4 per Debye, a value that is critical in comparative studies of structurally varying molecules. To fully quantify the effect of the dipole, however, the dot product of the two vector quantitates, I·p̂ (where I = current and p̂ = effective dipole moment along the surface normal), should be considered. In this case therefore, M1 has an effective charge vector oriented at angle relative to the surface normal, while M9 and 1O2 are aligned with direction of electron flow, hence can be assumed to have an additive or resistive effect on the injection of charge into the molecules. A parameter that can capture the effect of dipole on tunneling rate is, therefore, the coupling between the electrode and the molecule. Considering molecule−electrode coupling, the rate of charge tunneling, expressed in current density, can be expressed as eq 129,30 2q J (V ) ∼ ΓΓ 1 2 (1) πℏ where Γ is the coupling strength, q is the electronic charge, and ℏ is the Planck constant. Here, we focus on the coupling at the bottom interface since the chemisorbed termini is expected to be in the strong coupling regime.31 If the molecules have significant polarizability and cooperativity, an intramolecular charge redistribution with the metal substrate will occur.32 As shown (Figure 1c), the electrostatic potential differences within the molecule is significantly larger for M9 and 1O2, and thus in SAMs a pseudo-2D layer of oriented dipoles forms on the substrate (Figures 1e and 4b left). This dipole layer (emanating from the assembled molecules) is metastable and, hence, tends to either partially transfer charge from/to the substrate or align upon applying a bias to mitigate charge build-up within the SAM (Figure 4b right). A partial electrostatic potential is then formed and locked at the molecule/metal interface. As a result, the

ΔE =

q·ΔΦ 2

ΔΦ =

∑∫ i

0

(2) π

Ni(θ ) ·Pi(θ ) ·cos θ dθ εr ·ε0

(3)

where P is the dipole moment, N is the density of the dipole, θ is the angle between dipole orientation and the surface normal, and εr·ε0 is the molecule permittivity. For M1, ΔE ∼ 0 because the dipole direction is orthogonal to the surface normal (cos θ ≈ 0), while for M9 and 1O2, energy change is significantly larger. From this expression, therefore, we predict and empirically show that tunneling rates for M9 and 1O2 are significantly lower than M1. We also demonstrate that the difference in tunneling rates is not just dependent on the magnitude of the dipole moment but also on its orientation relative to the surface normal. Kovalchuk et al.14 further supports our inference that SAMs with larger effective molecular dipoles promote disentanglement of mixed orbitals, hence reduced tunneling rates. From the higher current density and the theoretical work that M1 does not significantly affect the work function (see supporting informmation for details), we can infer that a molecule with a dipole moment orthogonal to the surface normal is akin to one without a dipole moment since there is no effect on the SAM-electrode coupling (i.e., ΔΦ = 0). For more detailed understanding of the molecule/electrode coupling, Fermi level pinning/depinning theory34 is used to explain the phenomenon that M9 significantly changes the electronic properties while M1 has minimal electronic perturbations due to the interface dipole coupling effects. Figure 4c shows a schematic of the energy level diagram for the prepared molecular junctions (without pinning), based on molecular energy levels derived from simulation and UPS studies.18,37 According to Liu et al.,34 there should be a potential drop at the Au/M9 interface due to the dipole but no analogous drop at Au/M1 interface due to negligible effective dipole moment, we believe there is Fermi level pinning only at Au/M9 interface but not at Au/M1 interface. As a result, M9 junctions have larger barrier height than M1 due to dipoleinduced energy band shift.34 The Fermi level depinning within M1 SAMs has also been demonstrated in carboranethiol SAMs modified organic semiconductor device,18 where M1-modified device shows significantly improved performance than M9. This is because M1 serves as a depinning insulating layer that reduces the injection barrier (hence the contact resistance).34 To further support inferences drawn so far, transition voltage spectroscopy (TVS) was performed. A transition voltage (VT) was observed for all three molecules (Figure 4e). Compared to previously reported molecules with aromatic groups, the transition voltage for carboranes are diffuse (broad).14,16 This E



■

Article

CONCLUSIONS In this work, we sought to find the effect of molecular dipoles of self-assembled monolayers on charge transport across large area molecular tunneling junction. Using carboranethiolatebased molecular tunneling junction, we are able to make conclusions as follows: 1. Carboranethiols-Oriented Perpendicular to the Au Substrate (along the Surface Normal) Are Ideal for the Study of Effect of Molecular Dipoles on Charge Transport Behavior. The cage structures in the two (M1 and M9) carboranes used in this study are structurally isomorphic (structure, empirical formula, barrier width, pseudoaromaticity, SAM structures and packing density) but distinct in magnitude and orientation of their molecular dipoles which was verified through literature and calculated electrostatic potential maps. To further support our inferences, a structurally slightly different analog of M9 carborane, 1O2 was also investigated realizing that although the SAMs are isomorphic, the latter has two anchoring units that may lead to differences in metal−molecule orbital mixing, hence slight differences in stereoelectronic properties of the surface-bound molecules. SAMs derived from carboranes on Au align along the surface normal (no canting) with comparable packing density and effective thickness and are therefore ideal for delineating the role of surface dipoles on charge transport since structural variabilities are not a concern. 2. The Orientation of the Dipole Significantly Affect the Tunneling Rates Across the Large Area Tunneling Junctions. Dipole moments oriented orthogonal to the surface normal do not affect charge transport behavior across these molecules as much as those oriented along the surface normal. When the dipole moments is oriented orthogonal to the surface normal, resistivity comparable to tridecanethiol (C13) are observed even though the apparent height of M1(or M9 for that matter) carboranethiolate SAMs (9.8 ± 0.3 Å) is comparable to that of C10 SAMs.21 When the dipoles are oriented along the surface normal (M9), however, the current densities are comparable to those of heptadecanethiol (C17). This shows that when two similar (isomorphic) SAMs have differences in magnitude and direction of the dipoles, then a difference in current density equivalent to 4 CC bonds is observed. By isolating stereostructure SAM effects from the dipole, we deduce the effect of dipole moments in charge transport by tunnelling to be Δ log|J| ≅ 0.4 per Debye, a value that is critical in interpreting effect of dipoles in charge tunneling through large-area junctions especially in SAMs where the cant angle and packing density are a concern. This work also demonstrates an angular dependence to the effect of dipoles on charge tunneling, a fact that has been largely ignored in tunneling studies. 3. The Dipole Affects the Coupling between Molecules and the Electrodes of Large Area Tunneling Junctions Due to the Fermi Level Pinning at the Interface. In terms of electrode-SAM coupling, the dipole moment is likely influential during the charge injection step and, as such, effect of dipoles can be expressed as a dot product (two vectors oriented at an angle) where the charge potential and dipole moment affect injection of the charge into the molecular orbitals, hence the degree of interference (decay) between the charge (moving wave) and the molecule (standing wave). We can therefore infer that the dipoles affect the potential drop across the metal−molecule interface, hence,

broadening of the VT is possibly due to the pseudoaromatic cages having empty nonbonding boron p-orbitals, which result in a large number of field tunable states hence a “diffuse” fielddriven tilting of the nonfrontier energy level bands. Lack of isolation from the bottom electrode also complicates the band structure of these molecules on binding to the substrate (discussed under Fermi-level pinning) The transition voltage (VT) when the top-electrode bears a positive bias was calculated from TVS (SI Figure S6 and Table S1) to be VT (M1) = 0.49 ± 0.02 V, VT (M9) = 0.43 ± 0.01 V, and VT 1O2 = 0.40 ± 0.02 V. Considering the broadened transition, we infer that the VT for M9 and 1O2 are not statistically significantly different but M1 is significantly different (ΔVT ≅ 0.07). Since by definition VT is associated with position of molecular levels, this result indicates that M1 has more accessible band(s) than M9, a result that concurs with reported theoretical work37 and agrees well with our inference on Fermi level pinning associated with the relative angle of the dipole moment relative to the surface normal. Interestingly, the F−N plots when the top-electrode bears a positive or negative bias are very different (Figure 4e) and hence the transtion voltage at forward bias and reverse bias are significantly different (see SI Figure S7 and Table S1). We believe this variance is due to the asymmetry of the junctions which is amplified by coupling across the two electrode− molecule interfaces. The ratio of the transtion voltage for forward and reverse bias remains constant across the molecules (Table S1), indicating that the asymmetry is an inherent property of the junction, even though different molecules give different transition voltages. We, therefore, infer that this asymmetry captures the differences in molecule−electrode coupling across the chemisorbed (bottom) and physisorbed (bottom) electrodes. From the above studies, we inferred that a change from direct tunneling to Fowler−Nordheim tunneling regime at 0.4−0.49 V should result in an increase in the rate of tunneling across these barriers. Based on Schottky−Mott rule, the difference in work function of M1 and M9 lead to different degree of band bending, which can be associated with recitification, especially with sweeps larger than the VT due in part to mixed transport mechanism. In these carboranethiol junctions, despite inherent junction asymmetric due to different electrode materials (Figure 4c) recitifcation was observed for all molecules when we expanded the sweep width from ±0.5 to ±1 V. With increased voltage, we observed slight current rectification for all molecules in the order: M1 > M9 > 1O2 (Figure 4d, and SI Tables S2). We observe that the standard deviations in rectification ratios (RR) at 0.5 V follow the general trend σRR‑M1 = 4.8 > σRR‑M9 = 3.3 > σRR‑1O2 = 2.1, with the same trend at 1 V (σRR‑M1 = 14 > σRR‑M9 = 9 > σRR‑1O2 = 6), as shown in SI Figure S8. We infer that the variance in the distributions may be due to differences in degree of barrier tilting above the VT and, potentially, the perturbations of this process by the molecular dipole. More importantly, the trend in RR is similar to the VT (SI Figure S8). The change of work-function through M1/M9 monolayer gives Schottky barrier-induced rectification in those junctions. We infer that differences in barrier height between M1 and M9, as deduced from VT, mostly arise from the Schottky-type behavior rather than structural differences in the molecules (size, packing density, surface-bond strength). This is likely affecting the coupling between the electrode and the molecule, hence M1 with a larger transition voltage shows higher tunneling rate. F


Article


m T//Ga2O3/EGaIn Junctions. J. Am. Chem. Soc. 2014, 136 (1), 16− 19. (13) Yoon, H. J.; Shapiro, N. D.; Park, K. M.; Thuo, M. M.; Soh, S.; Whitesides, G. M. The Rate of Charge Tunneling through SelfAssembled Monolayers Is Insensitive to Many Functional Group Substitutions. Angew. Chem., Int. Ed. 2012, 51 (19), 4658−4661. (14) Kovalchuk, A.; Abu-Husein, T.; Fracasso, D.; Egger, D. A.; Zojer, E.; Zharnikov, M.; Terfort, A.; Chiechi, R. C. Transition Voltages Respond to Synthetic Reorientation of Embedded Dipoles in Self-Assembled Monolayers. Chem. Sci. 2016, 7 (1), 781−787. (15) Beebe, J. M.; Kim, B.; Gadzuk, J.; Frisbie, C. D.; Kushmerick, J. G. Transition From Direct Tunneling to Field Emission in MetalMolecule-Metal Junctions. Phys. Rev. Lett. 2006, 97 (2), 026801. (16) Jia, C.; Guo, X. Molecule−Electrode Interfaces in Molecular Electronic Devices. Chem. Soc. Rev. 2013, 42 (13), 5642−5660. (17) Chen, J.; Liu, J.; Tevis, I.; Andino, R. S.; Miller, C. M.; Ziegler, L. D.; Chen, X.; Thuo, M. M. Spectroscopic Evidence for the Origin of Odd-Even Effects in Self-Assembled Monolayers and Effects of Substrate Roughness. Phys. Chem. Chem. Phys. 2017, 19, 6989−6995. (18) Kim, J.; Rim, Y. S.; Liu, Y.; Serino, A. C.; Thomas, J. C.; Chen, H.; Yang, Y.; Weiss, P. S. Interface Control in Organic Electronics using Mixed Monolayers of Carboranethiol Isomers. Nano Lett. 2014, 14 (5), 2946−2951. (19) Kim, M.; Hohman, J. N.; Cao, Y.; Houk, K. N.; Ma, H.; Jen, A. K.-Y.; Weiss, P. S. Creating Favorable Geometries for Directing Organic Photoreactions in Alkanethiolate Monolayers. Science 2011, 331 (6022), 1312−1315. (20) Hohman, J. N.; Claridge, S. A.; Kim, M.; Weiss, P. S. Cage Molecules for Self-Assembly. Mater. Sci. Eng., R 2010, 70 (3), 188− 208. (21) Hohman, J. N.; Zhang, P.; Morin, E. I.; Han, P.; Kim, M.; Kurland, A. R.; McClanahan, P. D.; Balema, V. P.; Weiss, P. S. SelfAssembly of Carboranethiol Isomers on Au {111}: Intermolecular Interactions Determined by Molecular Dipole Orientations. ACS Nano 2009, 3 (3), 527−536. (22) Frisch, M.; Trucks, G.; Schlegel, H.; Scuseria, G.; Robb, M.; Cheeseman, J.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. Gaussian 09, revision D. 01; Gaussian, Inc., Wallingford, CT, 2009. (23) Weiss, E. A.; Kaufman, G. K.; Kriebel, J. K.; Li, Z.; Schalek, R.; Whitesides, G. M. Si/SiO2-Templated Formation of Ultraflat Metal Surfaces on Glass, Polymer, and Solder Supports: Their Use as Substrates for Self-Assembled Monolayers. Langmuir 2007, 23 (19), 9686−9694. (24) Thuo, M. M.; Reus, W. F.; Simeone, F. C.; Kim, C.; Schulz, M. D.; Yoon, H. J.; Whitesides, G. M. Replacing− CH2CH2−with− CONH−Does Not Significantly Change Rates of Charge Transport through AgTS-SAM//Ga2O3/EGaIn Junctions. J. Am. Chem. Soc. 2012, 134 (26), 10876−10884. (25) Thuo, M. M.; Reus, W. F.; Nijhuis, C. A.; Barber, J. R.; Kim, C.; Schulz, M. D.; Whitesides, G. M. Odd− Even Effects in Charge Transport across Self-Assembled Monolayers. J. Am. Chem. Soc. 2011, 133 (9), 2962−2975. (26) Sporrer, J.; Chen, J.; Wang, Z.; Thuo, M. M. Revealing the Nature of Molecule−Electrode Contact in Tunneling Junctions Using Raw Data Heat Maps. J. Phys. Chem. Lett. 2015, 6 (24), 4952−4958. (27) Jiang, L.; Sangeeth, C. S.; Nijhuis, C. A. The Origin of the Odd− Even Effect in the Tunneling Rates across EGaIn Junctions with SelfAssembled Monolayers (SAMs) of n-Alkanethiolates. J. Am. Chem. Soc. 2015, 137 (33), 10659−10667. (28) Yuan, L.; Jiang, L.; Thompson, D.; Nijhuis, C. A. On the Remarkable Role of Surface Topography of the Bottom Electrodes in Blocking Leakage Currents in Molecular Diodes. J. Am. Chem. Soc. 2014, 136 (18), 6554−6557. (29) 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 (8), 618−624. (30) Grave, C.; Risko, C.; Shaporenko, A.; Wang, Y.; Nuckolls, C.; Ratner, M. A.; Rampi, M. A.; Zharnikov, M. Charge Transport through

effect of dipoles mainly manifests as differences in the coupling strength, Γ.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcc.7b07634. Detailed introduction of background, experimental section, and additional results and discussion (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: +1-515-294-8581. ORCID

Jiahao Chen: 0000-0001-7563-8023 Martin Thuo: 0000-0003-3448-8027 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This material is based upon work supported by the Air Force Office of Scientific Research under award number FA2386-161-4113 and by Iowa State University through startup funds. J.C. acknowledge support from the Catron solar energy center at Iowa State University through a Catron Fellowship.

■

REFERENCES

(1) Whitesides, G. M.; Love, J. C. The Art of Building Small. Sci. Am. 2007, 17, 12−21. (2) Ratner, M. A Brief History of Molecular Electronics. Nat. Nanotechnol. 2013, 8 (6), 378−381. (3) Chiechi, R. C.; Weiss, E. A.; Dickey, M. D.; Whitesides, G. M. Eutectic Gallium−Indium (EGaIn): A Moldable Liquid Metal for Electrical Characterization of Self-Assembled Monolayers. Angew. Chem. 2008, 120 (1), 148−150. (4) Thompson, D.; Nijhuis, C. A. Even the Odd Numbers Help: Failure Modes of SAM-Based Tunnel Junctions Probed via Odd-Even Effects Revealed in Synchrotrons and Supercomputers. Acc. Chem. Res. 2016, 49 (10), 2061−2069. (5) Akkerman, H. B.; Blom, P. W.; De Leeuw, D. M.; De Boer, B. Towards Molecular Electronics with Large-area Molecular Junctions. Nature 2006, 441 (7089), 69−72. (6) Akkerman, H. B.; Kronemeijer, A. J.; Harkema, J.; van Hal, P. A.; Smits, E. C.; de Leeuw, D. M.; Blom, P. W. Stability of Large-Area Molecular Junctions. Org. Electron. 2010, 11 (1), 146−149. (7) Uji, H.; Morita, T.; Kimura, S. Molecular Direction Dependence of Single-Molecule Conductance of a Helical Peptide in Molecular Junction. Phys. Chem. Chem. Phys. 2013, 15 (3), 757−760. (8) Mafé, S.; Manzanares, J. A.; Reiss, H. Nanoscale Switch based on Interacting Molecular Dipoles: Cooperativity Can Improve the Device Characteristics. J. Appl. Phys. 2011, 109 (4), 044302−044302−6. (9) Hiremath, R. K.; Rabinal, M. H. K.; Mulimani, B. G. Dipole Tuning of Charge Transport in Molecular Junctions. Phys. Chem. Chem. Phys. 2010, 12 (11), 2564−2568. (10) Díez-Pérez, I.; Hihath, J.; Lee, Y.; Yu, L.; Adamska, L.; Kozhushner, M. A.; Oleynik, I. I.; Tao, N. Rectification and Stability of a Single Molecular Diode with Controlled Orientation. Nat. Chem. 2009, 1 (8), 635−641. (11) Reus, W. F.; Nijhuis, C. A.; Barber, J. R.; Thuo, M. M.; Tricard, S.; Whitesides, G. M. Statistical Tools for Analyzing Measurements of Charge Transport. J. Phys. Chem. C 2012, 116 (11), 6714−6733. (12) Yoon, H. J.; Bowers, C. M.; Baghbanzadeh, M.; Whitesides, G. M. The Rate of Charge Tunneling Is Insensitive to Polar Terminal Groups in Self-Assembled Monolayers in AgTSS (CH2) n M (CH2) G


Article

The Journal of Physical Chemistry C Oligoarylene Self-Assembled Monolayers: Interplay of Molecular Organization, Metal−Molecule Interactions, and Electronic Structure. Adv. Funct. Mater. 2007, 17 (18), 3816−3828. (31) Karthäuser, S. Control of Molecule-Based Transport for Future Molecular Devices. J. Phys.: Condens. Matter 2011, 23 (1), 013001. (32) Cahen, D.; Naaman, R.; Vager, Z. The Cooperative Molecular Field Effect. Adv. Funct. Mater. 2005, 15 (10), 1571−1578. (33) Moth-Poulsen, K.; Bjørnholm, T. Molecular Electronics with Single Molecules in Solid-State Devices. Nat. Nanotechnol. 2009, 4 (9), 551−556. (34) Liu, Z.; Kobayashi, M.; Paul, B. C.; Bao, Z.; Nishi, Y. Contact Engineering for Organic Semiconductor Devices via Fermi Level Depinning at the Metal-Organic Interface. Phys. Rev. B: Condens. Matter Mater. Phys. 2010, 82 (3), 035311. (35) Iwamoto, M.; Mizutani, Y.; Sugimura, A. Calculation of the Dielectric Constant of Monolayer Films on a Material Surface. Phys. Rev. B: Condens. Matter Mater. Phys. 1996, 54 (11), 8186. (36) Vilan, A.; Ghabboun, J.; Cahen, D. Molecule− Metal Polarization at Rectifying GaAs Interfaces. J. Phys. Chem. B 2003, 107 (26), 6360−6376. (37) Mete, E.; Yılmaz, A.; Danışman, M. F. A van der Waals Density Functional Investigation of Carboranethiol Self-Assembled Monolayers on Au (111). Phys. Chem. Chem. Phys. 2016, 18 (18), 12920− 12927.

H


Role of Molecular Dipoles in Charge Transport across Large Area

Recommend Documents