Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2024 Oct 4;146(41):28516–28526. doi: 10.1021/jacs.4c11241

α,ω-Alkanedibromides Form Low Conductance Chemisorbed Junctions with Silver Electrodes

Thomas M Czyszczon-Burton †, Enrique Montes ‡, Jazmine Prana †, Sawyer Lazar †, Nils Rotthowe †, Sully F Chen †, Héctor Vázquez ‡,*, Michael S Inkpen †,*
PMCID: PMC11487573  PMID: 39364997

Abstract

graphic file with name ja4c11241_0006.jpg

Chemical groups capable of connecting molecules physically and electrically between electrodes are of critical importance in molecular-scale electronics, influencing junction conductance, variability, and function. While the development of such linkage chemistries has focused on interactions at gold, the distinct reactivity and electronic structure of other electrode metals provides underexplored opportunities to characterize and exploit new binding motifs. In this work we show that α,ω-alkanedibromides spontaneously form well-defined junctions using silver, but not gold, electrodes through application of the glovebox-based scanning tunneling microscope-based break junction method. We systematically evaluate, through a series of additional studies, whether these molecular components form physisorbed or chemisorbed contact geometries, and if they undergo secondary chemical reactions at the silver surface. Critically, we find that the same junctions form when using different halide, or trimethylstannyl, terminal groups, suggestive of an electronically transparent silver–carbon(sp3) contact chemistry. However, the experimental conductance of the junctions we measure with silver electrodes is ∼30× lower than that observed for such junctions comprising gold–carbon(sp3) contacts, which does not align with predictions based on first-principles calculations. We further exclude the possibility that the proposed silver alkyl species undergo α- or β-hydride elimination reactions that result in a distinct contact chemistry through conductance measurements of control molecules that cannot undergo such processes. Applying insights provided from prior temperature-programmed desorption studies and a robust series of atomistic simulations, we ultimately propose that in these experiments we measure alkoxide-terminated junctions formed from the reaction of the chemisorbed alkyl with oxygen that is coadsorbed on the silver surface. This work, in demonstrating that high conductance contact chemistries established using model gold electrodes may not be readily transferred to other metals, underscores the need to directly characterize the interfacial electronic properties and reactivity of electrode metals of wider technological relevance.

Introduction

Inspired by the established chemistry of self-assembled monolayers,1 among the first reported single-molecule junctions comprised air-stable gold electrodes bridged by thiol-functionalized organic compounds.2,3 A wide variety of linker groups have since been identified to be compatible with gold,4 and a subset of these have been found to form junctions with other metal electrodes. For example, room temperature scanning tunneling microscope-based break junction (STM-BJ) studies with silver electrodes have utilized aurophilic thiol (−SH),5 nitrile (−CN),6,7 amine (−NH2),5−10 pyridyl (−C5H4N),11,12 alkyne (−CCH),13−15 and carboxylic acid (−COOH)16−18 linkers. The chemical properties of such linker groups, in combination with those of the metal (projected density of states, work function), are known to impact single-molecule and large-area junction properties in several important ways–exerting a strong influence over the conducting orbital level alignment and electronic coupling,4,10,19−23 junction mechanical stability,24,25 or even imparting conductance switching functionality.26−28 Beyond such STM-BJ studies with gold and silver electrodes a wider array of metals are capable of forming junctions comprising multiple-atom, single-atom, or molecular bridges, as was first demonstrated through mechanically controlled break junction studies in vacuum.29,30

Interestingly, while different electrode metals exhibit distinct surface structures, electronic properties, and intrinsic reactivities, the ability of a chemical group to function as a linker for some metals but not others has not yet been comprehensively evaluated. For example, we are only aware of one study by Jo et al., who report that methoxy (−OMe) groups, which do not bind to gold, are capable of forming junctions with silver leads.31 The discovery and further development of non-gold contact chemistries formed in solution measurements has clear potential to drive significant advances in molecular electronics and surface assembly, for example, by establishing new metal-molecule bonding interactions with useful mechanical or electronic character, or enabling the selective self-assembly of molecules on mixed metal substrates. More broadly, the identification of appropriate contact chemistries to form molecular junctions with less diffusive electrode metals than gold may prove critical to establishing the utility of prospective hybrid molecule-semiconductor integrated circuitry.32 Studies of single-molecule junctions that expose interfacial chemical transformations33 could also provide important insights into reaction discovery and development in on-surface chemistry,34 as well as studies of heterogeneous catalytic chemical processes.35

Electrode linker groups are commonly incorporated into organic molecules via synthetic routes that start from halide-functionalized precursors (particularly for −Br and −I). Such precursors are commercially ubiquitous due to their utility in a wide range of chemical transformations, and the distinct rates of reactivity of carbon–halogen bonds which enable modular, multistep reaction strategies. These factors also contribute to growing interest in the direct use of halides as electrode linkers. This approach could not only significantly reduce the number of synthetic steps required to prepare molecular junction components, but also enable the characterization of systems in which the incorporation of conventional contact groups is challenging. With gold electrodes, it has been reported that junctions can be formed using thienyl-,36 alkyl-37 and aryl-iodides,38−41 as well as halide-capped clusters.42 Depending on the experimental conditions, such junctions are proposed to comprise either physisorbed gold-halogen (coordinating through the halogen lone pairs) or chemisorbed gold–carbon (Au–C) contacts (following oxidative addition of aryl iodides), although their exact contact chemistry has not always been determined unambiguously.

The use of organohalides to directly form chemisorbed Au–C(sp3) and Au–C(sp2) contacts is an attractive proposition, avoiding established precursors with toxic -SnMe3 groups,23,43 laborious preinstallation of Au–C bonds,23 or the in situ reduction of (potentially explosive) diazonium salts.44 For completeness, we also note that Au–C(sp) contacts may be formed through the in situ deprotonation of terminal alkynes.45,46 Au–C(sp3) linkages are among the most electronically transparent contacts for single-molecule junctions, due to Au–C “gateway” states that can be well coupled to both the electrode and the molecular backbone (e.g., in benzyl–gold contacts).23,43 While aryl iodides are typically considered among the most reactive of the organohalides, exhibiting faster rates of oxidative addition than aryl bromides, aryl chlorides, or alkyl halides,47−49 their use to form gold-aryl linked junctions in situ at room temperature has so far been demonstrated only through the application of reducing potentials in an electrochemical environment,39 or exposure to atomically rough surfaces in polar solvents that generate a large interfacial electric field.40,50 Despite the increased reactivity of non-gold metals toward activation of carbon–halogen bonds relative to gold,51,52 the application of these electrode materials to form molecular junctions from organohalides has not yet been reported. Indeed, the only non-gold metal–carbon (M–C) contacts probed in single-molecule junctions are Ag–C(sp) linkages formed from terminal alkynes,13−15 or M–C(sp2) contacts formed in situ from N-heterocyclic carbene precursors (M = Ag, Cu),53 leaving their properties broadly underexplored.

In this work we show, through application of the glovebox STM-BJ method, that simple α,ω-alkanedibromides, CnBr (n = 4–12, where “n” represents the number of backbone methylene groups), form well-defined single-molecule junctions with silver electrodes under an inert nitrogen atmosphere (Figure 1, top). Remarkably, these junctions form spontaneously from nonpolar solutions at room temperature, even at low tip–substrate bias (10 mV). To establish the nature of the interfacial contact, we performed a robust series of control experiments and first-principles transmission calculations based on density functional theory (DFT). Specifically, these studies systematically evaluate whether these CnBr junctions comprise chemisorbed or physisorbed contact geometries or undergo secondary chemical reactions at the silver surface. Critically, measurements of C8X (X = −I, −Cl, or −SnMe3) produced near-identical conductance features to those of C8Br, indicating that each of these compounds are precursors that react in situ to produce junctions of similar geometry. However, the observed peak features occur at an unusually low conductance: ∼30× lower than Au–C(sp3) linked junctions of the same alkane length,23 and 2.5× lower than for analogous Ag–S linked junctions which comprise two additional S atoms in the molecular backbone (Figure 3a). Transmission calculations further show that alkane junctions linked by Ag–C(sp3) contacts should be slightly more—not significantly less—conducting than their Au–C(sp3) linked analogues (Figure 3b), together establishing that we do not measure chemisorbed Ag–C(sp3) linked alkane junctions in these experiments. Additional control experiments rule out other plausible secondary processes such as α- or β-hydride elimination reactions from the proposed silver alkyl species. Drawing insights from prior surface science studies,54−57 we propose instead that while Ag–C(sp3) linked junctions do initially form from haloalkane precursors, these chemisorbed alkyl groups subsequently react with surface-adsorbed oxygen (noting that O2 is present in ppm quantities) to form silver-alkoxides that we trap and measure in molecular junctions (Figure 1, bottom). This hypothesis is supported here by transmission calculations which show Ag–O-alkyl terminated junctions exhibit conductance values consistent with the features we observe experimentally.

Figure 1.

Figure 1

Open in a new tab

A schematic showing a single-molecule junction formed using silver electrodes from solutions of α,ω-dibromoalkanes (CnBr) under an inert nitrogen atmosphere inside a glovebox. Bottom: CnBr precursors can feasibly interact with the electrode surface in different ways, forming alkyl halide- (H), alkyl- (C), or alkoxide-contacted (O) junctions following loss of the halide and reaction with surface-adsorbed oxygen.

Figure 3.

Figure 3

Open in a new tab

(a) Overlaid 1D histograms from conductance measurements of C10SMe, C10Br (reproduced from Figure 2a for convenience) and C10SH with silver electrodes. The conductance of C10Br junctions is intermediate between those with physisorbed and chemisorbed sulfur linkers, where C10SH junctions exhibit a higher conductance than C10Br despite comprising two additional S atoms in the proposed molecular backbone for junctions of the CC-type. Histograms were constructed from 6,000 to 10,000 traces. The corresponding 2D conductance-displacement histograms for C10SMe and C10SH are shown in the SI, Figure S7. (b) DFT-based transmission spectra of Ag-C10 and Au-C10 (decane-based) junctions, plotted relative to the Fermi level (EF). Inset: scattering state of the Ag-C10 junction at E = −0.6 eV. (c) Overlaid 1D histograms for junctions formed from iodide- (C8I, purple), bromide- (C8Br, red; reproduced from Figure 2a for convenience), chloride- (C8Cl, green) and trimethylstannyl-terminated (C8SnMe3, pink) 1,8-octane wires (6,800–10,000 traces). C8I was measured using a solution concentration of 0.01 mM to facilitate the clean breaking of junctions. C8SnMe3 was measured using a solution concentration of ∼8 mM at 50 mV (comparable conditions to previous studies23), after 40 h of solution–substrate contact to maximize peak intensity. While the conductance peak in the presented data for C8Cl appears more intense than that for the other α,ω-dihalooctanes, peak intensities were found to vary for all analytes from experiment to experiment. Corresponding 2D histograms for C8X are shown in the SI, Figure S8a-d. (d) Calculated transmission spectra of C4Sn(Me2) (dark green), C4Br (HC, dark red), and C4Br (HH, light red) junctions. (e) Schematic structures for the different junction geometries calculated in (d).

Results and Discussion

Conductance Measurements of α,ω-Dibromoalkanes with Silver Electrodes

We perform conductance measurements with silver electrodes using a STM-BJ setup housed inside an inert atmosphere glovebox that has been described previously (see SI for more details).58,59 To remove the majority of native oxide from oxyphilic silver surfaces, we polish evaporation slug substrates using silicon carbide sandpaper and mechanically cut tip wire inside the glovebox. These processes produce new metallic silver surfaces that have never been exposed to air. Immediately after removing surface oxide, STM-BJ measurements are performed by applying a voltage bias (V) across the tip and substrate while measuring the current (I), converted to conductance G (= I/V), as the electrodes are pushed in and out of contact. In each conductance-displacement measurement, the initially formed large area junction thins to a single-atom contact with increasing tip–substrate displacement, before breaking to form a nanoscale gap between undercoordinated silver atoms. We repeat these measurements thousands of times whereby the resulting traces are compiled, without data selection, into one- and two-dimensional (1D and 2D) conductance histograms.

During experiments under nitrogen or in an anhydrous solvent environment under nitrogen, we observe steps in individual conductance-displacement traces close to 1 G0 (= 2e2/h = 7.75 × 10–5 S), characteristic of the formation of silver atomic point contacts, which combine to form well-defined histogram features (SI, Figure S1). We find the conductance steps for silver atom contacts are typically shorter than the corresponding steps observed using gold electrodes, in agreement with previous work10 and consistent with reports that atomic chains form more readily with gold than silver.60 While STM-BJ studies using silver electrodes exposed to air have reported an additional conductance feature close to ∼0.1 G0 that is attributed to the formation of Ag–O–Ag junctions,10,61 this feature is not typically pronounced in the glovebox-based studies presented here. As evidenced below, additional conductance features are observed below 1 G0 in the presence of molecules in solution that can bridge the nanogap-separated silver electrodes. Importantly, we find our inert atmosphere glovebox STM-BJ approach can facilitate conductance measurements of molecular junctions formed using silver electrodes at ambient temperature and pressure over extended periods (hours or days) without apparent reoxidation of the electrode surface.

In Figure 2a we plot overlaid conductance histograms obtained from measurements of 0.1 mM solutions of CnBr in tetradecane (TD). In each case we identify a prominent feature (black dotted arrow, Figure 2a), with a shoulder or peak of lower intensity at ∼2× the primary peak conductance. We attribute these features to 1 and 2 molecule junctions formed from CnBr, respectively (SI, Figure S2).62−64 In Figure 2b we plot representative traces for CnBr measurements (n = 6–12), showing that the step length corresponding to these junctions increases with the number of bridge carbon atoms (see SI, Figure S3 for 2D histograms). While the intensity of conductance peaks in 1D histograms is convoluted with the probability of junction formation, these changes in step length also broadly account for the higher intensity features observed in measurements of CnBr junctions comprising longer alkane backbones (Figure 2a). The strong correlation between step and molecular length further supports the assignment of these features to molecular junctions formed from CnBr. In contrast to an earlier report,37 our attempts to form molecular junctions with C8X solutions (X = Br, I) using gold electrodes provide no clear conductance peak features (see SI, Figure S4 and associated discussion). This indicates that CnX do not readily form molecular junctions with gold and implicates a distinct haloalkane-Ag interfacial chemistry.

Figure 2.

Figure 2

Open in a new tab

(a) Overlaid 1D histograms obtained from scanning tunneling microscope-based break junction (STM-BJ) measurements of 0.1 mM CnBr solutions in tetradecane at 750 mV bias (n = 4–12). Histograms are constructed from 6,000 to 10,000 consecutively measured traces without data selection. Arrows indicate assigned high (black dotted) and low (blue solid) conductance peaks for C8Br. The histogram for C4Br is truncated for clarity. Corresponding 2D conductance-displacement histograms are provided in the SI, Figure S3. (b) Sample conductance-displacement traces for selected junctions (n = 6–12), offset along the displacement axis for clarity. (c) A plot of conductance versus number of repeat methylene units (n) for high (black triangles) and low (blue squares) conductance peaks. Exponential fits to each data series using G = Gcexp(−βn) gives β(high) = 0.98/n, β(low) = 1.00/n, Gc(high) = 1.6 × 10–1 G0, and Gc(low) = 1.2 × 10–2 G0.

When probing shorter CnBr junctions (n = 4–10), we also observe an additional, lower conductance peak feature of reduced intensity (blue solid arrow, Figure 2a). This peak typically arises from a step that occurs after breaking of the high conductance junctions, as illustrated by the example traces shown in SI, Figure S5 and in 2D histograms which show regions of reduced counts for displacements prior to the onset of the low conductance feature (SI, Figure S3, S8h, and S11b). Accordingly, we suggest that this feature results from (i) a geometrical rearrangement of the electrode (e.g., pulling out a silver atom from the surface) and/or the molecule during junction displacement that changes the molecule-electrode coupling or frontier orbital energy alignment; or (ii) the measurement of a different junction comprising a distinct contact chemistry (see below). The origin of this component appears distinct from that of a similar feature observed for covalent Au–C junctions, which was in that case attributed to a periodic electrode deformation that resulted in intermittent jumps to a low conductance geometry with junction elongation.23 Related conductance features have also been observed in some studies of alkanedithiols measured using gold electrodes, attributed to the formation of Au(SR)2 interfacial contacts.65

In Figure 2c we plot conductance versus molecular length for each series of peaks, using the most probable values of junction conductance obtained from Gaussian fits to histogram peaks (SI, Table S1). The conductance of both series exhibits an exponential length dependence, indicative of through-molecule tunneling transport. Tunneling decay constants (β ∼ 1/n), obtained through fits to each series using G = Gcexp(−βn) (where Gc is the contact conductance), agree with previously observed values obtained for junctions with alkane backbones measured on gold or silver.10,24 However, the nature of the molecule-electrode contact formed from CnBr in the primary series remains unclear. Based on earlier studies of halide-terminated molecules on gold (discussed above), instead of a single dominant peak we might expect to observe three peak features associated with junction geometries comprising two chemisorbed Ag–C(sp3) linkages (CC), two physisorbed halogen-bonded Ag–Br interactions (HH), or one of each type of contact (HC; see Figure 1 and Figure 3b,e).

Evaluating the Formation of Chemisorbed Ag–C(sp3) Contacts

We first consider the possibility that the dominant peaks from CnBr measurements correspond to CC-type junctions. If this was the case these junctions would exhibit a surprisingly low conductance, ∼30× smaller than reported values for the corresponding gold-alkane-gold junctions formed from trialkylstannyl precursors (SI, Table S1 and Figure S6).23 To further illustrate this point we present, in Figure 3a, overlaid conductance histograms from measurements of 1,10-bis(methylthio)decane (C10SMe), C10Br, and 1,10-decanedithiol (C10SH). This comparison shows that the conductance of C10Br junctions is intermediate between junctions comprising physisorbed and chemisorbed sulfur linkers. It is particularly notable that the conductance of C10SH junctions, comprising two additional sulfur atoms compared to the proposed CC-type junctions formed from C10Br, is 2.5× higher. With gold electrodes the opposite trend is found, as junctions formed from C8SH are ∼22× lower in conductance than the Au–C(sp3) linked gold-octane-gold analogue.23,66 Remarkably, conductance histograms of C4Br and C6Br comprise no substantial secondary features that align with the primary conductance peaks obtained from studies of C8Br and C12Br (Figure 2a). If these junctions had a CC geometry, this would appear to indicate that the rate of formation of alkane dimers via C–C homocoupling reactions on silver electrodes is significantly slower than with gold electrodes, as dimers are commonly observed in studies of Au–C linked junctions.23,39,40 This contrasts with expectations based on the results of temperature-programmed desorption (TPD) experiments, where silver is reportedly unique in mediating alkyl homocoupling reactions at temperatures >190 K relative to other decomposition pathways.67,68 Accordingly, the silver-alkyl motif is expected to exhibit a high reactivity at room temperature, as further illustrated through unsuccessful attempts to isolate related molecular complexes.69 In contrast, analogous gold-alkyl complexes are readily prepared70 and have been successfully utilized as controls with preinstalled Au–C bonds to help assess whether Au–C interfacial contacts have formed in molecular junctions.23

We performed DFT-based transmission calculations of Au–C(sp3) and Ag–C(sp3) contacted junctions in an attempt to rationalize the surprisingly low conductance of junctions formed from CnBr. In Figure 3b, we compare the transmission spectra of C10 (decane-based) junctions comprising direct, covalent C(sp3) bonds to Ag or Au electrodes, calculated at the DFT level (with energy plotted relative to the Fermi level (EF = 0)). Remarkably, both spectra are qualitatively similar, exhibiting a broad peak centered at −0.6 eV that derives from electronic states localized at the Au–C and Ag–C bonds. In Figure 3b-inset, we show the real-space representation of the current-carrying eigenchannel of the Ag-C10 junction at this energy, which reveals the electrode and backbone states are well coupled through the Ag–C(sp3) bond. In the SI, Figure S12, we present additional transmission calculations of Ag-alkyl junctions with different backbone lengths. These spectra all exhibit qualitatively similar features, comprising a broad transmission resonance close to the Fermi level (EF) which results in highly conducting junctions (see Table S3 for selected data). Importantly, our analogue benchmark simulations with gold electrodes are consistent with the previous findings of Cheng et al.23 Across the different backbone lengths, our calculations yield transmission values at EF that are ∼2× higher for Ag-alkyl junctions than for Au-alkyl junctions.

While it is well-known that DFT calculations suffer from inherent deficiencies in the calculation of molecular resonance energies, we stress these are unlikely to alter the relative calculated conductance of the Ag- and Au-alkyl junctions studied here. Notably, corrections to these deficiencies based on the DFT+Σ methodology have been devised for weakly interacting junctions,71 yet such adjustments may not be appropriate for junctions linked through strongly interacting chemisorbed Ag–C(sp3) or Au–C(sp3) contacts. Assuming a rigid translation of the DFT spectra toward lower energies (Figure 3b and SI, Figure S12), large shifts in the Ag–C resonance states (>3 eV) would be needed to achieve the relative conductance values measured here for junctions formed from CnBr. While smaller resonance shifts of 1.4–2 eV have been reported for weakly bonded molecular junctions,72−74 level corrections might be expected to be even smaller in covalently bonded interfaces.75,76 Accordingly, the required value of >3 eV for Ag-alkyl covalent bonds appears too high to be reasonable. Furthermore, we cannot physically rationalize why such resonance shifts would be so much larger for Ag-alkyl junctions than for analogue Au-alkyl systems, to reconcile the ∼30× lower measured conductance of Ag-alkyl junctions compared to the same junctions formed with gold electrodes. Accordingly, DFT transmission calculations show that the conductance of alkyl junctions with Ag–C(sp3) linkages should be comparable to, and even slightly higher than those with Au–C(sp3) bonds, and strongly suggest that the CnBr junctions measured experimentally do not comprise chemisorbed Ag–C(sp3) bonds (i.e., they do not have CC contact geometries).

Considering the Prescence of Silver-Halogen Physisorbed Contacts

We next performed a series of control studies comprising alkanes with different terminal groups to establish whether the junctions formed from CnBr have HC or HH contact geometries. Schematic structures of these junction geometries are shown in Figure 3e. We plot, in Figure 3c, overlaid conductance histograms for measurements of C8X (X = I, Br, Cl, SnMe3; C8SnMe3 was measured as a mixture in a ∼ 1:0.3 molar ratio with 1-trimethylstannyloctane, see SI, Figures S21–23) which all show a near identical high conductance feature at 6–7 × 10–5 G0. This finding strongly suggests that the junctions we measure in each case result from the in situ cleavage of these terminal groups to form junctions of identical composition and geometry, particularly given that the -SnMe3 group is not expected to form a physisorbed contact. To rigorously evaluate this proposal, we calculate the conductance of model junction geometries comprising terminal groups that interact with silver electrodes through one (HC), or two (HH), physisorbed Ag–Br or chemisorbed Ag–Sn contacts (CnSn(Me2), feasibly formed through cleavage of a Sn-Me bond).

In Figure 3d, we plot overlaid transmission functions for such junction geometries with C4 (butane-based) backbones. Additional transmission spectra for analogous HC and HH junctions comprising different alkyl halides are shown in the SI, Figures S13 and S14. The representative spectra shown in Figure 3d exhibit transmission peaks at energies characteristic of either the H or C silver-molecule contact. Spectra of HC junctions show a broad peak at −0.7 eV, the signature of the Ag–C resonance which dominates the zero-bias conductance, as well as a peak around 2–3 eV which corresponds to the Ag-halide bond (SI, Figure S15). Spectra of HH junctions exhibit two peaks in the unoccupied range of the spectrum (>2 eV) corresponding to both Ag-halide resonances. In contrast, CnSn(Me2) junctions comprising Ag–Sn(Me2) contacts show two peaks below EF, both derived from the Ag–Sn contact, the broader of which dominates the zero bias conductance. Notably, the calculated conductance of HC junctions is >1 order of magnitude, and the conductance of CnSn(Me2) junctions is almost 2 orders of magnitude, higher than HH junctions (SI, Figure S16 and Table S6). Furthermore, simulations of HC and HH junctions show that junction conductance changes with the identity of the halogen atom. Compared to bromide-terminated HC junctions, the calculated conductance is ∼15% smaller and ∼40% higher for chloride and iodide contacts, respectively (SI, Table S4). Such percentage differences are approximately twice as large in HH junctions (SI, Table S5). These calculations corroborate expectations that the conductance of these junctions should exhibit a significant dependence on the identity of the terminal group, contrasting with our experimental results which show the conductance of junctions formed from C8X exhibit the same conductance. This further supports the hypothesis that these halide and trimethylstannyl groups are lost through reaction at the silver surface and are not incorporated into the junctions we measure experimentally.

We highlight additional features of the experiments shown in Figure 3c and related control studies that provide important insights. In contrast to studies using CnBr, we found C8I was more readily measured as a 0.01 mM solution, rather than at a 0.1 mM concentration where it was difficult to repeatedly break tip–substrate point contacts. It also proved challenging to routinely observe conductance peaks in measurements of C8Cl at 0.1–1 mM concentrations. These observations are consistent with the expectation that organoiodides (organochlorides) should interact most strongly (weakly),77 and react at a faster (slower) rate,52 with metal electrodes, as discussed above. We note that the peak feature at 1 G0 exhibits a diminished intensity in measurements comprising halide precursors, relative to measurements of C8SnMe3 (Figure 3c), C10SH, C10SMe (Figure 3a), or clean silver (SI, Figure S1), suggesting the presence of halide inhibits the formation of stable silver atomic points contacts. While a small peak feature for C8SnMe3 at 6–7 × 10–5 G0 was observed immediately in conductance measurements after addition of a ∼ 8 mM solution to the silver substrate, this feature was found to increase in intensity after extended solution-substrate contact. Accordingly, the conductance histogram for C8SnMe3 provided in Figure 3c was obtained ∼40 h after introduction of solution to help maximize the histogram peak intensity. A similar observation was reported in studies of 1,4-bis(trimethylstannyl)benzene on gold electrodes,23 in which conductance steps appeared 2–3 h after addition of a 10 mM solution to the substrate (suggesting slow conversion of the Sn–C bond to a M–C interfacial contact). However, in a repeat of this experiment using silver electrodes we have observed peak features of much higher intensity after only ∼4 h of solution-substrate contact, which suggest that the precise tip geometry, and/or the substrate roughness and cleanliness, play an important role in the probability of junction formation from this reactive precursor (SI, Figure S8g,h). For completeness, we also performed conductance measurements of C8SnMe3 with gold electrodes (SI, Figure S8e). Here we obtained the same high conductance peak feature as previously reported,23 directly illustrating the stark differences between the conductance of the resulting junctions formed with gold and silver electrodes. Conductance measurements of C8X (X = Br, Cl) and C6Br at lower applied bias (100 or 10 mV rather than 750 mV) provide histograms with similar conductance features to those obtained at higher bias, showing that large electric fields are not a prerequisite for junction formation (SI, Figure S9). Conductance measurements of C10Br using a template-stripped silver substrate78 gave identical conductance features as those observed using mechanically polished silver substrates (SI, Figure S10), indicating that the formation of these junctions is not impacted by surface roughness or adventitious contamination during oxide removal.

Excluding the Prescence of Distinct Contacts Formed through Hydride Elimination

We perform an additional series of conductance measurements to confirm that the junctions formed from CnBr do not result from other plausible secondary processes such as α- or β-hydride eliminations. While these chemical reactions have not previously been observed for silver-alkyl species in TPD studies,57 they have been observed to occur on surfaces functionalized with different metal–alkyl groups79 and in discrete metal complexes.80,81 We recognize that such reactions could result in the formation of carbenic, or π-bound vinylic interfacial contacts, respectively (Figure 4a). To verify that these contacts do not form we perform a series of control measurements using the compounds shown in Figure 4b, in which we have replaced α- and/or β-hydrogens relative to the Br–C (and therefore the projected Ag–C) bond with different groups that have higher bond dissociation energies. This approach follows established strategies in molecular inorganic chemistry that have been shown to stabilize otherwise reactive ligands.81 In Figure 4c, we present overlaid conductance histograms for C10Br and C10Br-Me, a molecular analogue in which α-hydrogens have been replaced by methyl groups. Both measurements provide a peak at the same conductance, indicating the formation of CnBr junctions does not involve α-hydride elimination processes resulting in carbenic contacts. We attribute the low intensity of the conductance feature in measurements of C10Br-Me (and 1,8-dibromoperfluorooctane (C8Br–F), Figure 4f) to the bulky groups proximal to the Br–C bond that may sterically inhibit junction formation or stability. To help maximize the histogram peak intensity, conductance measurements of C10Br-Me were performed after leaving the solution in contact with the substrate overnight.

Figure 4.

Figure 4

Open in a new tab

(a) Reaction scheme showing possible α- and β-hydride elimination processes to form junctions with carbenic or vinylic contacts. (b) Control molecules with substituents at α- (C10Br-Me), β- (Tol2Br), or both α- and β-positions (C8Br-F) that cannot readily undergo hydride elimination. (c) Overlaid 1D histograms of C10Br-Me (4,000 traces) and C10Br showing that each forms junctions of similar conductance. This shows the formation of these junctions do not involve α-hydride elimination processes resulting in carbenic contacts. (d) Overlaid 1D histograms for bitolyl-based junctions formed from Tol2Br and Tol2SMe (peak separation indicated by black solid arrow). The junctions formed from bromide-terminated molecules exhibit a conductance that is higher by a similar ratio to alkane-based junctions formed from C10Br and C10SMe (Figure 2a; alkane: ×9.6, tolyl: ×3.5). These comparisons indicate we form junctions from Tol2Br that comprise a similar molecule–electrode contact chemistry as junctions formed from C10Br, and that this process does not involve β-hydride elimination reactions that result in vinylic contacts. We note that Tol2Br and Tol2SMe were measured in 1,2,4-trichlorobenzene solutions to facilitate their dissolution, and Tol2Br was measured at a concentration of 1 μM to facilitate the clean breaking of junctions. (e) Proposed schematic structures for the junctions formed in (d), illustrating their similarities. X = O for the alkoxide-terminated junctions proposed below based on prior temperature-programmed desorption studies and our transmission calculations. (f) Overlaid 1D histograms showing that the conductance of C8Br-F junctions is only a factor of ∼2 lower than that of C8Br junctions. This suggests that these junctions exhibit similar contact chemistries, with conductance only influenced by backbone substituents that tune the energetic alignment of conducting orbitals with the silver EF, further corroborating the studies using C10Br-Me and Tol2Br. Data for CnBr reproduced from Figure 2a for convenience.

In Figure 4d, we present overlaid conductance histograms for bitolyl-based junctions formed from 4,4′-bis((methylthio)methyl)biphenyl (Tol2SMe) and 4,4′-bis(bromomethyl)biphenyl (Tol2Br), a molecule with no β-hydrogens. We use the conductance of C10SMe (Figure 3a) and Tol2SMe junctions as a point of reference to evaluate whether the junctions formed from C10Br (Figure 2a) and Tol2Br exhibit a markedly distinct or comparable contact chemistry (Figure 4e). We find that the peak conductance of C10SMe and Tol2SMe junctions is lower by a similar factor relative to the peak conductance of junctions formed from their Br-terminated analogues (by ×9.6 and ×3.5, respectively), implicating a similar contact chemistry. Notably, if the absence of β-hydrogens could stabilize a distinct, highly conducting C(sp3)-Ag contact chemistry we would expect the conductance of junctions formed from Tol2Br to be orders of magnitude higher than the conductance of Tol2SMe junctions, given that the Ag–C bond would be both strongly coupled and well aligned with the backbone π-system (as shown for analogous junctions with gold electrodes).43 Considering that Tol2Br cannot undergo β-hydride elimination reactions to form vinylic contacts, these results indicate that such processes also do not occur in CnBr junctions. As such, we conclude that this decomposition pathway is not responsible for the low contact conductance observed. In further support of this deduction, our attempts to directly measure alkene-terminated alkanes in junctions with silver electrodes have so far provided only inconsistent conductance features distinct to those shown here. Finally, in Figure 4f, we show that the conductance of junctions formed from C8Br-F, a molecule with no α- or β-hydrogens, is a factor of ∼2 lower than that of C8Br junctions, providing additional supporting evidence that junctions formed from CnBr do not comprise significantly different carbenic or vinylic contacts. We propose that the slightly lower conductance of C8Br-F junctions may instead be attributed to the electron-withdrawing fluorine backbone substituents that are expected to impact level alignment between the conducting molecular orbital(s) and silver electrode EF.82

Evidence in Support of Silver Alkoxide Contacts

The studies above show that the conductance features of junctions formed from CnBr are not easily attributed to formation of the CC, HC, or HH geometries that might be expected based on previous molecular conductance studies, nor to secondary processes involving α-, or β-hydride elimination from a chemisorbed silver alkyl species. To identify alternative plausible contact chemistries for these junctions, we turn to earlier reports that have applied TPD and high-resolution electron energy loss spectroscopy (HREELS) to probe chemical reactions of alkyl halides on silver surfaces. Alkyl iodides have been most extensively studied, and are found to chemisorb on clean Ag(110) and Ag(111) surfaces at temperatures starting around 110 and <150 K, respectively.55−57 The alkyl fragments (e.g., CnH(2n+1)) subsequently desorb as homocoupled alkanes (C2nH2(2n+1)) at temperatures above 190 K without evidence of dehydrogenation, and adsorbed iodide is found to desorb above 700 K.56,68 These reports further support our assertion that large electric fields are not required to drive the chemisorption of CnX in STM-BJ studies at room temperature (SI, Figure S9), and suggest they occur spontaneously across the silver surface. Remarkably, when both ethyl iodide and molecular oxygen are adsorbed on a Ag(110) surface (studied in ratios between 2:1 and 1:0), additional products including diethyl ether and ethoxide are formed.54,55 While the desorption peak for butane was found to vary between 218 and 238 K depending on surface coverage, features attributed to the reaction products of surface ethoxide (acetaldehyde, ethanol) were centered around 270–280 K; this highlights the increased thermal stability of chemisorbed Ag–O–alkyl relative to Ag–alkyl species. It was further shown that alkyls and oxygen compete for available silver surface sites, the presence of adsorbed oxygen does not inhibit the cleavage of C–I bonds (at least for an oxygen monolayer coverage of ∼30%), and that the formation of homocoupled alkyl products decreases as the coverage of oxygen increases from 0 to 33%.55

These observations are particularly relevant to experiments presented here, as the O2 content of the nitrogen atmosphere present during our glovebox STM-BJ measurements can increase from <1 ppm to ∼5–20 ppm during a typical experiment (monitored using an oxygen analyzer fuel cell). Even for an excellent glovebox atmosphere comprising 0.1 ppm of O2 at ∼1 atm, O2 will be exposed to our polycrystalline silver surface at a rate of ∼80 Langmuir/s. This corresponds to a dose of >90,000 L if conductance data for the clean silver surface prior to adding analyte solution is measured in just 20 min (see the SI for more details). While the sticking coefficient of O2 on silver varies for different crystal faces (e.g., 10–3 for Ag(110) at 300 K), the doses of O2 applied in glovebox STM-BJ studies are far greater than those typically introduced under ultrahigh vacuum to achieve submonolayer or saturation coverages.83−85 We note that after molecular junction components are added in solution on top of a substrate the relative ratio of oxygen and reactive chemical groups at the interface will vary, being influenced by their relative solution concentrations and diffusion rates.

Together, these TPD studies confirm that while the carbon–halogen bond of alkyl halides can undergo rapid scission at room temperature to form silver-alkyl species, these groups are inherently unstable and can subsequently react with adsorbed surface oxygen to form alkoxides. We hypothesize that it is these alkoxides (CnO) that we trap and measure in molecular junctions (Figure 1). The formation of such junctions, comprising an interfacial Ag–O–C(sp3) contact that is expected to be less conducting than Ag–C(sp3), would explain the low junction conductance we observe experimentally in measurements of CnX. However, we note that the provision of indisputable experimental evidence to confirm this hypothesis poses a significant challenge. Complexes containing preinstalled Ag–C(sp3) bonds are difficult to isolate (noted above), rendering their utilization as controls here impractical. Excluding carboxylate-contacted junctions which are expected to bind through a distinct coordination geometry,16−18 we know of no reported methods to form and probe chemisorbed Ag–O linked junctions and only one recent study of Au–O linked junctions–formed through conductance measurements of deprotonated aryl–OH groups.86 To date, our attempts to form Ag–O junctions directly from the less acidic alkyl–OH precursors have proven unsuccessful.

In lieu of experimental measurements, we evaluate this hypothesis using computational models. We prepare junctions with each end of the alkyl backbone bonded to an oxygen atom on a silver electrode. Our simulations show that oxygen atoms adsorb on the hollow site on top of silver trimers, in agreement with previous findings,87 and that Ag–O–C contacts are more favorable than Ag–C linkages by ∼1 eV (SI, Table S8). In Figure 5a, we plot the transmission function for a C10O junction with two such Ag–O–C(sp3) contacts, calculated at the DFT level. The occupied resonance at −1.5 eV derives from the Ag–O gateway state and dominates the zero-bias conductance as it extends across EF. We show, in Figure 5b, the real-space representation of the current-carrying eigenchannel state calculated at EF, confirming that the oxygen 2p states are strongly involved in transport through these junctions. Notably, and in remarkable agreement with the trend observed experimentally (Figure 3a), the calculated conductance for this junction geometry lies between the calculated values for thioether- and thiolate-linked junctions on silver electrodes: 4.5 × 10–6 G0 for C10SMe, 1.7 × 10–5 G0 for C10O, and 9.5 × 10–5 G0 for C10S (SI, Figure S17). Transport calculations of CnO junctions comprising different alkane lengths (n = 4–10; SI, Figure S18) further illustrate that the resonance associated with the Ag–O bond is further from EF and narrower than that associated with Ag–S and Ag–C(sp3) bonds, yet remains capable of enhancing transmission at EF relative to junctions comprising Ag–S(Me) contacts. This makes these oxygen-bonded alkyl junctions less conductive than analogous junctions with chemisorbed Ag–S linkers and significantly less transmissive than with Ag–C(sp3) contacts. While the calculated conductance of CnO junctions agrees well with the measured results, falling within a factor of 2 (SI, Table S7), DFT values generally overestimate measured conductance.71 Accordingly, rather than focusing on the exact quantitative values, we propose the trends in calculated conductance strongly support the existence of an Ag–O–C(sp3) contact chemistry in the junctions formed from CnBr and measured experimentally using the STM-BJ method.

Figure 5.

Figure 5

Open in a new tab

(a) Transmission spectrum of a C10O junction, comprising a decane-based backbone with Ag–O–C(sp3) contacts to both electrodes. (b) Transmission eigenchannel calculated at the EF. Oxygen atoms adsorb on the hollow site of Ag trimer structures.

Conclusion

Here we have demonstrated that alkyl halides uniquely and spontaneously form chemisorbed molecular junctions with silver electrodes, leveraging the ability of intrinsically reactive non-gold metal surfaces to activate otherwise inert functional molecular groups. Critically, however, we show that these junctions do not exhibit the high conductance expected for Ag–C(sp3) linkages based on measurements of analogous precursors on gold, or from first-principles calculations based on DFT. We hypothesize that the most likely contact incorporates an insulating oxygen component through a multistep process that involves the reaction of an intermediate silver-alkyl species with surface adsorbed oxygen, resulting in the formation of Ag–O–C(sp3) linkages. These findings expose fundamental challenges in transferring core concepts and methodologies in molecular electronics, established using model systems comprising gold electrodes, between different electrode materials with distinct properties.

More broadly, this work highlights the application of glovebox STM-BJ methods to probe interfacial chemical reactions and charge transport processes through molecular conductance measurements with air-sensitive electrodes. We anticipate that the established enhanced reactivity of simple, halide-functionalized molecules at silver and other non-gold electrodes67,88 will facilitate additional studies focused on the formation and evaluation of different M–C contact chemistries that build upon the approaches used here. In pursuing such experiments, we emphasize the value of exploring and drawing inspiration from the wider surface science literature, which we recognize contains results and discussion of substantial relevance to the field of molecular electronics, but which have not yet been well-integrated into this field. Given the oxophilicity of many metals, we also propose that the development of new, robust approaches to form molecular junctions comprising metal–oxygen contacts will prove invaluable in evaluating the interfacial chemistry of chemisorbed species in junctions formed from non-gold electrodes.

Acknowledgments

This work was primarily supported by funding from the University of Southern California (USC), the National Science Foundation (NSF CAREER Award to M.S.I., CHE-2239614), and the Czech Science Foundation (GAČR) under project 23-05891S. N.R. was supported by a Marie Skłodowska Curie Global Fellowship (HOPELEC: 898657) within the Horizon 2020 Programme. Computational resources were supplied by the project “e-Infrastruktura CZ” (e-INFRA LM2018140) provided within the program Projects of Large Research, Development and Innovations Infrastructures. Instrumentation in the USC Chemistry Instrument Facility was acquired with support from the USC Research and Innovation Instrumentation Award Program. Additionally, funds provided by the NSF (DBI-0821671, CHE-0840366) and National Institute of Health (S10 RR25432) supported the acquisition of the NMR spectrometers used in our work.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacs.4c11241.

  • Additional experimental details and synthetic methods, as well as conductance, computational, and spectroscopic data. (PDF)

Author Contributions

§ T.M.C.-B. and E.M. contributed equally to this work.

The authors declare no competing financial interest.

Supplementary Material

ja4c11241_si_001.pdf (7MB, pdf)

References

  1. Ulman A. Formation and Structure of Self-Assembled Monolayers. Chem. Rev. 1996, 96 (4), 1533–1554. 10.1021/cr9502357. [DOI] [PubMed] [Google Scholar]
  2. Zhou C.; Muller C.; Burgin T.; Tour J.; Reed M. Conductance of a Molecular Junction. Science 1997, 278 (5336), 252–254. 10.1126/science.278.5336.252. [DOI] [Google Scholar]
  3. Xu B.; Tao N. J. Measurement of Single-Molecule Resistance by Repeated Formation of Molecular Junctions. Science 2003, 301 (5637), 1221–1223. 10.1126/science.1087481. [DOI] [PubMed] [Google Scholar]
  4. Su T. A.; Neupane M.; Steigerwald M. L.; Venkataraman L.; Nuckolls C. Chemical Principles of Single-Molecule Electronics. Nat. Rev. Mater. 2016, 1 (3), 16002. 10.1038/natrevmats.2016.2. [DOI] [Google Scholar]
  5. Li H.; Su T.; Camarasa-Gómez M.; Hernangómez-Pérez D.; Henn S.; Pokorný V.; Caniglia C. D.; Inkpen M.; Korytár R.; Steigerwald M.; Nuckolls C.; Evers F.; Venkataraman L. Silver Makes Better Electrical Contacts to Thiol-Terminated Silanes than Gold. Angew. Chem., Int. Ed. 2017, 56 (45), 14145–14148. 10.1002/anie.201708524. [DOI] [PubMed] [Google Scholar]
  6. Li D. F.; Mao J. C.; Chen D. L.; Chen F.; Ze-Wen H.; Zhou X. Y.; Wang Y. H.; Zhou X. S.; Niu Z. J.; Maisonhaute E. Single-Molecule Conductance with Nitrile and Amino Contacts with Ag or Cu Electrodes. Electrochim. Acta 2015, 174, 340–344. 10.1016/j.electacta.2015.06.018. [DOI] [Google Scholar]
  7. Kim H. S.; Kim T. Dependence of the Conductance Change on the Molecular Orbitals in Ag and Au Electrodes. J. Korean Phys. Soc. 2016, 68 (2), 279–282. 10.3938/jkps.68.279. [DOI] [Google Scholar]
  8. Kim T. Conductance of a Single Molecule Junction Formed with Ni, Au, and Ag Electrodes. J. Korean Chem. Soc. 2014, 58 (6), 513–516. 10.5012/jkcs.2014.58.6.513. [DOI] [Google Scholar]
  9. Kim T. Substituent Effect on the Conductance of Single-Molecule Junctions Formed with Silver Electrodes. J. Korean Phys. Soc. 2015, 67 (12), 2077–2080. 10.3938/jkps.67.2077. [DOI] [Google Scholar]
  10. Kim T.; Vázquez H.; Hybertsen M. S.; Venkataraman L. Conductance of Molecular Junctions Formed with Silver Electrodes. Nano Lett. 2013, 13 (7), 3358–3364. 10.1021/nl401654s. [DOI] [PubMed] [Google Scholar]
  11. Adak O.; Korytár R.; Joe A. Y.; Evers F.; Venkataraman L. Impact of Electrode Density of States on Transport through Pyridine-Linked Single Molecule Junctions. Nano Lett. 2015, 15 (6), 3716–3722. 10.1021/acs.nanolett.5b01195. [DOI] [PubMed] [Google Scholar]
  12. Zhou X. Y.; Wang Y. H.; Qi H. M.; Zheng J. F.; Niu Z. J.; Zhou X. S. Single-Molecule Conductance of Dipyridines Binding to Ag Electrodes Measured by Electrochemical Scanning Tunneling Microscopy Break Junction. Nanoscale Res. Lett. 2014, 9 (1), 1–6. 10.1186/1556-276X-9-77. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Song K.; Lin J.; Song X.; Yang B.; Zhu J.; Zang Y.; Zhu D. Formation of Covalent Metal-Carbon Contacts Assisted by Ag + for Single Molecule Junctions. Chem. Commun. 2023, 59 (41), 6207–6210. 10.1039/D3CC01113E. [DOI] [PubMed] [Google Scholar]
  14. Li S.; Yu H.; Chen X.; Gewirth A. A.; Moore J. S.; Schroeder C. M. Covalent Ag-C Bonding Contacts from Unprotected Terminal Acetylenes for Molecular Junctions. Nano Lett. 2020, 20 (7), 5490–5495. 10.1021/acs.nanolett.0c02015. [DOI] [PubMed] [Google Scholar]
  15. Song K.; Lin J.; Song X.; Zhang M.; Gu Q.; Zang Y.; Zhu D. In Situ Creation of Organometallic Molecular Junctions via Terminal Alkynes. J. Phys. Chem. C 2023, 127 (18), 8850. 10.1021/acs.jpcc.3c01337. [DOI] [Google Scholar]
  16. Zhou X.-S.; Liang J.-H.; Chen Z.-B.; Mao B.-W. An Electrochemical Jump-to-Contact STM-Break Junction Approach to Construct Single Molecular Junctions with Different Metallic Electrodes. Electrochem. Commun. 2011, 13 (5), 407–410. 10.1016/j.elecom.2011.02.005. [DOI] [Google Scholar]
  17. Wang Y. H.; Hong Z. W.; Sun Y. Y.; Li D. F.; Han D.; Zheng J. F.; Niu Z. J.; Zhou X. S. Tunneling Decay Constant of Alkanedicarboxylic Acids: Different Dependence on the Metal Electrodes between Air and Electrochemistry. J. Phys. Chem. C 2014, 118 (32), 18756–18761. 10.1021/jp505374v. [DOI] [Google Scholar]
  18. Peng Z.-L.; Chen Z.-B.; Zhou X.-Y.; Sun Y.-Y.; Liang J.-H.; Niu Z.-J.; Zhou X.-S.; Mao B.-W. Single Molecule Conductance of Carboxylic Acids Contacting Ag and Cu Electrodes. J. Phys. Chem. C 2012, 116 (41), 21699–21705. 10.1021/jp3069046. [DOI] [Google Scholar]
  19. Engelkes V. B.; Beebe J. M.; Frisbie C. D. Length-Dependent Transport in Molecular Junctions Based on SAMs of Alkanethiols and Alkanedithiols: Effect of Metal Work Function and Applied Bias on Tunneling Efficiency and Contact Resistance. J. Am. Chem. Soc. 2004, 126 (43), 14287–14296. 10.1021/ja046274u. [DOI] [PubMed] [Google Scholar]
  20. Kim B.; Choi S. H.; Zhu X. Y.; Frisbie C. D. Molecular Tunnel Junctions Based on π-Conjugated Oligoacene Thiols and Dithiols between Ag, Au, and Pt Contacts: Effect of Surface Linking Group and Metal Work Function. J. Am. Chem. Soc. 2011, 133 (49), 19864–19877. 10.1021/ja207751w. [DOI] [PubMed] [Google Scholar]
  21. Wang J. G.; Prodan E.; Car R.; Selloni A. Band Alignment in Molecular Devices: Influence of Anchoring Group and Metal Work Function. Phys. Rev. B - Condens. Matter Mater. Phys. 2008, 77 (24), 1–8. 10.1103/PhysRevB.77.245443. [DOI] [Google Scholar]
  22. Adak O.; Korytár R.; Joe A. Y.; Evers F.; Venkataraman L. Impact of Electrode Density of States on Transport through Pyridine-Linked Single Molecule Junctions. Nano Lett. 2015, 15 (6), 3716–3722. 10.1021/acs.nanolett.5b01195. [DOI] [PubMed] [Google Scholar]
  23. Cheng Z. L.; Skouta R.; Vazquez H.; Widawsky J. R.; Schneebeli S.; Chen W.; Hybertsen M. S.; Breslow R.; Venkataraman L. In Situ Formation of Highly Conducting Covalent Au-C Contacts for Single-Molecule Junctions. Nat. Nanotechnol 2011, 6 (6), 353–357. 10.1038/nnano.2011.66. [DOI] [PubMed] [Google Scholar]
  24. Park Y. S.; Whalley A. C.; Kamenetska M.; Steigerwald M. L.; Hybertsen M. S.; Nuckolls C.; Venkataraman L. Contact Chemistry and Single-Molecule Conductance: A Comparison of Phosphines, Methyl Sulfides, and Amines. J. Am. Chem. Soc. 2007, 129 (51), 15768–15769. 10.1021/ja0773857. [DOI] [PubMed] [Google Scholar]
  25. Frei M.; Aradhya S. V.; Hybertsen M. S.; Venkataraman L. Linker Dependent Bond Rupture Force Measurements in Single-Molecule Junctions. J. Am. Chem. Soc. 2012, 134 (9), 4003–4006. 10.1021/ja211590d. [DOI] [PubMed] [Google Scholar]
  26. Chen X.; Roemer M.; Yuan L.; Du W.; Thompson D.; del Barco E.; Nijhuis C. A. Molecular Diodes with Rectification Ratios Exceeding 105 Driven by Electrostatic Interactions. Nat. Nanotechnol. 2017, 12, 797–803. 10.1038/nnano.2017.110. [DOI] [PubMed] [Google Scholar]
  27. 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 Single-Molecule Junction. Nat. Nanotechnol. 2009, 4 (4), 230–234. 10.1038/nnano.2009.10. [DOI] [PubMed] [Google Scholar]
  28. Zhou Q.; Song K.; Zhang G.; Song X.; Lin J.; Zang Y.; Zhang D.; Zhu D. Tetrathiafulvalenes as Anchors for Building Highly Conductive and Mechanically Tunable Molecular Junctions. Nat. Commun. 2022, 13 (1), 1–8. 10.1038/s41467-022-29483-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Smit R. H. M.; Noat Y.; Untiedt C.; Lang N. D.; van Hemert M. C.; van Ruitenbeek J. M. Measurement of the Conductance of a Hydrogen Molecule. Nature 2002, 419 (6910), 906–909. 10.1038/nature01103. [DOI] [PubMed] [Google Scholar]
  30. Agraït N.; Yeyati A. L.; van Ruitenbeek J. M. Quantum Properties of Atomic-Sized Conductors. Phys. Rep. 2003, 377 (2–3), 81–279. 10.1016/S0370-1573(02)00633-6. [DOI] [Google Scholar]
  31. Jo H. Y.; Yoo P. S.; Kim T. Oxygen-Silver Junction Formation for Single Molecule Conductance. J. Korean Chem. Soc. 2015, 59 (1), 18–21. 10.5012/jkcs.2015.59.1.18. [DOI] [Google Scholar]
  32. Lortscher E. Wiring Molecules into Circuits. Nat. Nanotechnol 2013, 8 (6), 381–384. 10.1038/nnano.2013.105. [DOI] [PubMed] [Google Scholar]
  33. Stone I.; Starr R. L.; Zang Y.; Nuckolls C.; Steigerwald M. L.; Lambert T. H.; Roy X.; Venkataraman L. A Single-Molecule Blueprint for Synthesis. Nat. Rev. Chem. 2021, 5, 695–710. 10.1038/s41570-021-00316-y. [DOI] [PubMed] [Google Scholar]
  34. Lindner R.; Kühnle A. On-Surface Reactions. ChemPhysChem 2015, 16 (8), 1582–1592. 10.1002/cphc.201500161. [DOI] [PubMed] [Google Scholar]
  35. Hannagan R. T.; Giannakakis G.; Flytzani-Stephanopoulos M.; Sykes E. C. H. Single-Atom Alloy Catalysis. Chem. Rev. 2020, 120 (21), 12044–12088. 10.1021/acs.chemrev.0c00078. [DOI] [PubMed] [Google Scholar]
  36. Xiang L.; Hines T.; Palma J. L.; Lu X.; Mujica V.; Ratner M. A.; Zhou G.; Tao N. Non-Exponential Length Dependence of Conductance in Iodide-Terminated Oligothiophene Single-Molecule Tunneling Junctions. J. Am. Chem. Soc. 2016, 138, 679–687. 10.1021/jacs.5b11605. [DOI] [PubMed] [Google Scholar]
  37. Peng L. L.; Huang B.; Zou Q.; Hong Z. W.; Zheng J. F.; Shao Y.; Niu Z. J.; Zhou X. S.; Xie H. J.; Chen W. Low Tunneling Decay of Iodine-Terminated Alkane Single-Molecule Junctions. Nanoscale Res. Lett. 2018, 13, 121. 10.1186/s11671-018-2528-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Komoto Y.; Fujii S.; Hara K.; Kiguchi M. Single Molecular Bridging of Au Nanogap Using Aryl Halide Molecules. J. Phys. Chem. C 2013, 117 (46), 24277–24282. 10.1021/jp404858x. [DOI] [Google Scholar]
  39. Starr R. L.; Fu T.; Doud E. A.; Stone I.; Roy X.; Venkataraman L. Gold-Carbon Contacts from Oxidative Addition of Aryl Iodides. J. Am. Chem. Soc. 2020, 142 (15), 7128–7133. 10.1021/jacs.0c01466. [DOI] [PubMed] [Google Scholar]
  40. Stone I. B.; Starr R. L.; Hoffmann N.; Wang X.; Evans A. M.; Nuckolls C.; Lambert T. H.; Steigerwald M. L.; Berkelbach T. C.; Roy X.; Venkataraman L. Interfacial Electric Fields Catalyze Ullmann Coupling Reactions on Gold Surfaces. Chem. Sci. 2022, 13, 10798–10805. 10.1039/D2SC03780G. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Li Y.; Buerkle M.; Li G.; Rostamian A.; Wang H.; Wang Z.; Bowler D. R.; Miyazaki T.; Xiang L.; Asai Y.; Zhou G.; Tao N. Gate Controlling of Quantum Interference and Direct Observation of Anti-Resonances in Single Molecule Charge Transport. Nat. Mater. 2019, 18 (4), 357–363. 10.1038/s41563-018-0280-5. [DOI] [PubMed] [Google Scholar]
  42. Wei C.; Ye J.; Su Y.; Zheng J.; Xiao S.; Chen J.; Yuan S.; Zhang C.; Bai J. Halide Anchors for Single-Cluster Electronics. CCS Chem. 2022, 5, 1–9. 10.31635/ccschem.022.202202180. [DOI] [Google Scholar]
  43. Chen W.; Widawsky J. R.; Vázquez H.; Schneebeli S. T.; Hybertsen M. S.; Breslow R.; Venkataraman L. Highly Conducting π-Conjugated Molecular Junctions Covalently Bonded to Gold Electrodes. J. Am. Chem. Soc. 2011, 133 (43), 17160–17163. 10.1021/ja208020j. [DOI] [PubMed] [Google Scholar]
  44. Hines T.; Díez-Pérez I.; Nakamura H.; Shimazaki T.; Asai Y.; Tao N. Controlling Formation of Single-Molecule Junctions by Electrochemical Reduction of Diazonium Terminal Groups. J. Am. Chem. Soc. 2013, 135 (9), 3319–3322. 10.1021/ja3106434. [DOI] [PubMed] [Google Scholar]
  45. Olavarria-Contreras I. J.; Perrin M. L.; Chen Z.; Klyatskaya S.; Ruben M.; van der Zant H. S. J. C-Au Covalently Bonded Molecular Junctions Using Nonprotected Alkynyl Anchoring Groups. J. Am. Chem. Soc. 2016, 138 (27), 8465–8469. 10.1021/jacs.6b03383. [DOI] [PubMed] [Google Scholar]
  46. Millar D.; Venkataraman L.; Doerrer L. H. Efficacy of Au-Au Contacts for Scanning Tunneling Microscopy Molecular Conductance Measurements. J. Phys. Chem. C 2007, 111 (47), 17635–17639. 10.1021/jp0756101. [DOI] [Google Scholar]
  47. Miyaura N.; Suzuki A. Palladium-Catalyzed Cross-Coupling Reactions of Organoboron Compounds. Chem. Rev. 1995, 95 (7), 2457–2483. 10.1021/cr00039a007. [DOI] [Google Scholar]
  48. Sambiagio C.; Marsden S. P.; Blacker A. J.; McGowan P. C. Copper Catalysed Ullmann Type Chemistry: From Mechanistic Aspects to Modern Development. Chem. Soc. Rev. 2014, 43 (10), 3525–3550. 10.1039/C3CS60289C. [DOI] [PubMed] [Google Scholar]
  49. Ariafard A.; Lin Z. Understanding the Relative Easiness of Oxidative Addition of Aryl and Alkyl Halides to Palladium(0). Organometallics 2006, 25 (16), 4030–4033. 10.1021/om060236x. [DOI] [Google Scholar]
  50. Sorenson S. A.; Patrow J. G.; Dawlaty J. M. Solvation Reaction Field at the Interface Measured by Vibrational Sum Frequency Generation Spectroscopy. J. Am. Chem. Soc. 2017, 139 (6), 2369–2378. 10.1021/jacs.6b11940. [DOI] [PubMed] [Google Scholar]
  51. Livendahl M.; Espinet P.; Echavarren A. M. Final Analysis: Is Gold a Catalyst in Cross-Coupling Reactions in the Absence of Palladium?. Platin. Met. Rev. 2011, 55 (3), 212–214. 10.1595/147106711X579128. [DOI] [Google Scholar]
  52. Björk J.; Hanke F.; Stafström S. Mechanisms of Halogen-Based Covalent Self-Assembly on Metal Surfaces. J. Am. Chem. Soc. 2013, 135 (15), 5768–5775. 10.1021/ja400304b. [DOI] [PubMed] [Google Scholar]
  53. Doud E. A.; Inkpen M. S.; Lovat G.; Montes E.; Paley D. W.; Steigerwald M. L.; Vázquez H.; Venkataraman L.; Roy X. In Situ Formation of N-Heterocyclic Carbene-Bound Single-Molecule Junctions. J. Am. Chem. Soc. 2018, 140 (28), 8944–8949. 10.1021/jacs.8b05184. [DOI] [PubMed] [Google Scholar]
  54. Jones G. S.; Barteau M. A.; Vohs J. M. Mechanism of Diethyl Ether Formation on Ag(110) and Its Dependence on Coadsorbed Oxygen Species. J. Phys. Chem. B 1999, 103 (7), 1144–1151. 10.1021/jp983997p. [DOI] [Google Scholar]
  55. Jones G. S.; Barteau M. A.; Vohs J. M. The Formation of Diethyl Ether via the Reaction of Iodoethane with Atomic Oxygen on the Ag(110) Surface. Surf. Sci. 1999, 420 (1), 65–80. 10.1016/S0039-6028(98)00821-8. [DOI] [Google Scholar]
  56. Zhou X.-L.; Solymosi F.; Blass P. M.; Cannon K. C.; White J. M. Interactions of Methyl Halides (Cl, Br and I) with Ag(111). Surf. Sci. 1989, 219, 294–316. 10.1016/0039-6028(89)90214-8. [DOI] [Google Scholar]
  57. Zhou X.-L.; White J. M. Thermal Decomposition of C2H5I on Ag(111). Catal. Lett. 1989, 2, 375–384. 10.1007/BF00768180. [DOI] [Google Scholar]
  58. Venkataraman L.; Klare J. E.; Tam I. W.; Nuckolls C.; Hybertsen M. S.; Steigerwald M. L. Single-Molecule Circuits with Well-Defined Molecular Conductance. Nano Lett. 2006, 6 (3), 458–462. 10.1021/nl052373+. [DOI] [PubMed] [Google Scholar]
  59. Miao Z.; Quainoo T.; Czyszczon-Burton T. M.; Rotthowe N.; Parr J. M.; Liu Z.; Inkpen M. S. Charge Transport across Dynamic Covalent Chemical Bridges. Nano Lett. 2022, 22 (20), 8331–8338. 10.1021/acs.nanolett.2c03288. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Smit R. H. M.; Untiedt C.; Yanson A. I.; Van Ruitenbeek J. M. Common Origin for Surface Reconstruction and the Formation of Chains of Metal Atoms. Phys. Rev. Lett. 2001, 87 (26), 266102. 10.1103/PhysRevLett.87.266102. [DOI] [PubMed] [Google Scholar]
  61. Kim T. Silver Electrodes for Reversible Oxygen Sensor Applications. J. Korean Phys. Soc. 2015, 67 (5), 823–826. 10.3938/jkps.67.823. [DOI] [Google Scholar]
  62. Reuter M. G.; Solomon G. C.; Hansen T.; Seideman T.; Ratner M. A. Understanding and Controlling Crosstalk between Parallel Molecular Wires. J. Phys. Chem. Lett. 2011, 2 (14), 1667–1671. 10.1021/jz200658h. [DOI] [Google Scholar]
  63. Obersteiner V.; Egger D. A.; Zojer E. Impact of Anchoring Groups on Ballistic Transport: Single Molecule vs Monolayer Junctions. J. Phys. Chem. C 2015, 119 (36), 21198–21208. 10.1021/acs.jpcc.5b06110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Capozzi B.; Low J. Z.; Xia J.; Liu Z.-F.; Neaton J. B.; Campos L. M.; Venkataraman L. Mapping the Transmission Functions of Single-Molecule Junctions. Nano Lett. 2016, 16 (6), 3949–3954. 10.1021/acs.nanolett.6b01592. [DOI] [PubMed] [Google Scholar]
  65. Leary E.; Zotti L. A.; Miguel D.; Márquez I. R.; Palomino-Ruiz L.; Cuerva J. M.; Rubio-Bollinger G.; González M. T.; Agrait N. The Role of Oligomeric Gold-Thiolate Units in Single-Molecule Junctions of Thiol-Anchored Molecules. J. Phys. Chem. C 2018, 122 (6), 3211–3218. 10.1021/acs.jpcc.7b11104. [DOI] [Google Scholar]
  66. Inkpen M. S.; Liu Z.; Li H.; Campos L. M.; Neaton J. B.; Venkataraman L. Non-Chemisorbed Gold-Sulfur Binding Prevails in Self-Assembled Monolayers. Nat. Chem. 2019, 11, 351–358. 10.1038/s41557-019-0216-y. [DOI] [PubMed] [Google Scholar]
  67. Zaera F. The Surface Chemistry of Hydrocarbon Fragments on Transition Metals: Towards Understanding Catalytic Processes. Mol. Phys. 2002, 100 (19), 3065–3073. 10.1080/00268970210130254. [DOI] [Google Scholar]
  68. Zhou X. L.; White J. M. Reactions of Iodoethane, Iodopropane, 2-Iodopropane, Chloroiodomethane Adsorbed on Silver(111). J. Phys. Chem. 1991, 95 (14), 5575–5580. 10.1021/j100167a039. [DOI] [Google Scholar]
  69. Whitesides G. M.; Bergbreiter D. E.; Kendall P. E. Preparation and Thermal Decomposition of N-Alkyl(Tri-n-Butylphosphine)Silver(I) Reagents. J. Am. Chem. Soc. 1974, 96 (9), 2806–2813. 10.1021/ja00816a024. [DOI] [Google Scholar]
  70. Porter K. A.; Schier A.; Schmidbaur H.. α,ω-Bis[(Triphenylphosphine)Gold(I)] Hydrocarbons. In Perspectives in Organometallic Chemistry; The Royal Society of Chemistry, Cambridge, UK, 2003; pp 74–85. [Google Scholar]
  71. Quek S. Y.; Khoo K. H. Predictive DFT-Based Approaches to Charge and Spin Transport in Single-Molecule Junctions and Two-Dimensional Materials: Successes and Challenges. Acc. Chem. Res. 2014, 47 (11), 3250–3257. 10.1021/ar4002526. [DOI] [PubMed] [Google Scholar]
  72. Zotti L. A.; Bürkle M.; Pauly F.; Lee W.; Kim K.; Jeong W.; Asai Y.; Reddy P.; Cuevas J. C. Heat Dissipation and Its Relation to Thermopower in Single-Molecule Junctions. New J. Phys. 2014, 16 (1), 015004. 10.1088/1367-2630/16/1/015004. [DOI] [Google Scholar]
  73. Quek S. Y.; Venkataraman L.; Choi H. J.; Louie S. G.; Hybertsen M. S.; Neaton J. B. Amine-Gold Linked Single-Molecule Circuits: Experiment and Theory. Nano Lett. 2007, 7 (11), 3477–3482. 10.1021/nl072058i. [DOI] [PubMed] [Google Scholar]
  74. Markussen T.; Jin C.; Thygesen K. S. Quantitatively Accurate Calculations of Conductance and Thermopower of Molecular Junctions. Phys. Status Solidi B 2013, 250 (11), 2394–2402. 10.1002/pssb.201349217. [DOI] [Google Scholar]
  75. Refaely-Abramson S.; Liu Z.-F.; Bruneval F.; Neaton J. B. First-Principles Approach to the Conductance of Covalently Bound Molecular Junctions. J. Phys. Chem. C 2019, 123 (11), 6379–6387. 10.1021/acs.jpcc.8b12124. [DOI] [Google Scholar]
  76. Montes E.; Vázquez H. Calculation of Energy Level Alignment and Interface Electronic Structure in Molecular Junctions beyond DFT. J. Phys. Chem. C 2021, 125 (46), 25825–25831. 10.1021/acs.jpcc.1c07407. [DOI] [Google Scholar]
  77. Fatemi V.; Kamenetska M.; Neaton J. B.; Venkataraman L. Environmental Control of Single-Molecule Junction Transport. Nano Lett. 2011, 11 (5), 1988–1992. 10.1021/nl200324e. [DOI] [PubMed] [Google Scholar]
  78. 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. 10.1021/la701919r. [DOI] [PubMed] [Google Scholar]
  79. Zaera F. The Surface Chemistry of Hydrocarbons on Transition Metal Surfaces: A Critical Review. Isr. J. Chem. 1998, 38 (4), 293–311. 10.1002/ijch.199800035. [DOI] [Google Scholar]
  80. Schrock R. R. Alkylidene Complexes of Niobium and Tantalum. Acc. Chem. Res. 1979, 12 (3), 98–104. 10.1021/ar50135a004. [DOI] [Google Scholar]
  81. Mowat W.; Shortland A.; Yagupsky G.; Hill N. J.; Yagupsky M.; Wilkinson G. Elimination Stabilized Alkyls. Part I. Chromium, Molybdenum, Tungsten, and Vanadium. J. Chem. Soc., Dalton Trans. 1972, 533–542. 10.1039/dt9720000533. [DOI] [Google Scholar]
  82. Venkataraman L.; Park Y. S.; Whalley A. C.; Nuckolls C.; Hybertsen M. S.; Steigerwald M. L. Electronics and Chemistry: Varying Single-Molecule Junction Conductance Using Chemical Substituents. Nano Lett. 2007, 7 (2), 502–506. 10.1021/nl062923j. [DOI] [PubMed] [Google Scholar]
  83. Backx C.; de Groot C. P. M.; Biloen P. Adsorption of Oxygen on Ag(110) Studied by High Resolution ELS and TPD. Surf. Sci. 1981, 104, 300–317. 10.1016/0039-6028(81)90137-0. [DOI] [Google Scholar]
  84. Wachs I. E.; Madix R. J. The Oxidation of Methanol on a Silver (110) Catalyst. Surf. Sci. 1978, 76 (2), 531–558. 10.1016/0039-6028(78)90113-9. [DOI] [Google Scholar]
  85. Engelhardt H. A.; Menzel D. Adsorption of Oxygen on Silver Single Crystal Surfaces. Surf. Sci. 1976, 57 (2), 591–618. 10.1016/0039-6028(76)90350-2. [DOI] [Google Scholar]
  86. Lawson B.; Skipper H. E.; Kamenetska M. Phenol Is a pH-Activated Linker to Gold: A Single Molecule Conductance Study. Nanoscale 2024, 16 (4), 2022–2029. 10.1039/D3NR05257E. [DOI] [PubMed] [Google Scholar]
  87. Li W.-X.; Stampfl C.; Scheffler M. Oxygen Adsorption on Ag(111): A Density-Functional Theory Investigation. Phys. Rev. B 2002, 65 (7), 075407. 10.1103/PhysRevB.65.075407. [DOI] [Google Scholar]
  88. Zaera F. Preparation and Reactivity of Alkyl Groups Adsorbed on Metal Surfaces. Acc. Chem. Res. 1992, 25 (6), 260–265. 10.1021/ar00018a003. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ja4c11241_si_001.pdf (7MB, pdf)

Articles from Journal of the American Chemical Society are provided here courtesy of American Chemical Society

RESOURCES