
==== Front
J Am Chem Soc
J Am Chem Soc
ja
jacsat
Journal of the American Chemical Society
0002-7863
1520-5126
American Chemical Society

39250739
10.1021/jacs.4c07327
Article
Counterion Loss from Charged Surface-Bound Complexes Drives the Formation of Loosely Packed Monolayers
Trang Christina D. M.
https://orcid.org/0000-0001-8840-5093
Mora Perez Carlos
https://orcid.org/0000-0003-0579-7138
Ran Jingyi
https://orcid.org/0000-0002-5140-7500
Prezhdo Oleg V. *
https://orcid.org/0000-0001-7339-8812
Inkpen Michael S. *
Department of Chemistry, University of Southern California, Los Angeles, California 90089, United States
* Email: prezhdo@usc.edu.
* Email: inkpen@usc.edu.
09 09 2024
18 09 2024
146 37 2562525639
29 05 2024
27 08 2024
26 08 2024
© 2024 The Authors. Published by American Chemical Society
2024
The Authors
https://creativecommons.org/licenses/by-nc-nd/4.0/ Permits non-commercial access and re-use, provided that author attribution and integrity are maintained; but does not permit creation of adaptations or other derivative works (https://creativecommons.org/licenses/by-nc-nd/4.0/).

The functionality of multicomponent self-assembled monolayers (SAMs) can be severely diminished by the segregation of like components into nanoscale domains, a process that maximizes favorable short-range intermolecular interactions. Here, we explore the use of a modular family of sulfur-functionalized metal bis(terpyridine) complexes ([M(tpy-R)2]2+(PF6–)2) to prepare mixed SAMs, considering that the comparable structure, dimensions, and ionic composition of these species should render them interchangeable within the adsorbed surface layer. While surface voltammetry experiments show that these SAMs do exhibit compositions representative of their assembly solutions, they also suggest, in line with previous reports, that adjacent complexes in the monolayer are separated by a gap of ∼ 1 nm. Remarkably, X-ray photoelectron spectroscopy studies reveal no F 1s peak features that would confirm the proliferation of PF6– counterions on the surface. We propose that the loosely packed structure of these SAMs results from the loss or exchange of PF6– counterions, which introduces significant repulsive Coulomb interactions between the adsorbed 2+ charged complexes. The hypothesis is supported by an electrostatic model which indicates that these complexes should form close-packed SAMs if mobile counterions are present. First-principles calculations demonstrate that complex–counterion binding interactions are weakened by charge transfer to the gold substrate, suggesting that this may play an important role in the formation of such low-coverage SAMs. Together, this study raises important questions regarding the assembly, organization, and composition of charged SAMs and highlights new opportunities in the design of multicomponent monolayer assemblies with free volume, for example, to facilitate surface-based reactions or support molecular switches.

National Institutes of Health 10.13039/100000002 S10 RR25432 American Chemical Society Petroleum Research Fund 10.13039/100006770 62751-DNI5 University of Southern California 10.13039/100006034 NA Division of Chemistry 10.13039/100000165 2154367 Division of Chemistry 10.13039/100000165 0840366 Division of Biological Infrastructure 10.13039/100000153 0821671 document-id-old-9ja4c07327
document-id-new-14ja4c07327
ccc-price
==== Body
pmcIntroduction

New strategies to prepare functional self-assembled monolayers (SAMs) comprising different, perfectly mixed components in well-defined proportions are highly sought after in molecular nanoscience. These could help control the lateral spacing of functional species on a surface to enable, for example, the tuning of interactions between localized magnetic dopants for quantum information science,1−3 the modulation of intermolecular steric interactions to minimize their influence on surface-based reactions,4,5 or the provision of free volume to facilitate photoswitching operations.6,7 However, such SAMs are challenging to realize due to the distinct thermodynamic (molecule-molecule and molecule-surface interactions) and kinetic (solution diffusion and surface desorption/adsorption rates) properties associated with each component, which impact both SAM formation and temporal stability. As a result, the coassembly of dissimilar molecules at solid–liquid interfaces has been found to result in component segregation into domains,8−10 or SAMs with compositions that diverge from those of their assembly solutions.11 While it has been shown that dissimilar functional groups may be mixed at the molecular level in SAMs by covalently attaching them to the same surface binding group, such as a cyclic disulfide12 or spiroalkanedithiol,13 this approach may be limited to the incorporation of a small number (∼2) of different functional groups into a SAM at one time.

A distinct, emerging strategy that targets the formation of perfectly mixed SAMs with tunable free volume exploits families of “platform” molecules with comparable surface adsorption geometries and external structure.14 Varying an appended functional group that resides within the platform footprint, or substitution of internal atoms, may exert only a minimal influence on the assembly process. Direct interaction of the platform with the substrate, or proximal connection of surface anchor groups (rather than through an extended tether), ensures that components cannot bind to the surface within another’s footprint and limits lateral movements relative to their anchor point. In principle, the platform approach facilitates a highly modular strategy to form mixed SAMs comprising unlimited numbers of different functional groups (each appended to a separate platform) from their solution mixtures. Key examples of platform-type SAM components include bis- and tris-chelating alkanethiols,15−17 triazatrianguleniums (TATAs),14 carboranes,18,19 porphyrins,20 or electrografted calixarenes,21 though a broad array of relevant systems have been investigated.22,23

While platform components with tunable redox properties have been less frequently explored, these provide additional opportunities to evaluate and modify the properties of the SAMs they are used to build. For example, surface voltammetry24 or electrochemical scanning tunneling microscopy20 may be used to probe SAM composition and structure, and changes in component charge state could influence the function of an appended catalyst.25−27 A relevant system was introduced by Gang et al., who found that well-mixed (though likely air-sensitive28,29) monolayers can be formed from [M(typSH)(tpy)]2+ complexes (M = Co, Zn; typSH = 2,2′:6′,2″-terpyridine-4′-thiol).1 Their approach could, in principle, be extended to utilize analogous complexes with different metals that impart distinct redox behavior(s), magnetic characteristics,30 or other properties of interest such as catalytic activity.31 It is important to note that the use of preassembled complexes with directly connected surface-binding groups (platform components) is distinct from the majority of previous work on polypyridyl SAMs, prepared through stepwise in situ complex formation32 or using preformed complexes comprising a distal binding group (Supporting Information (SI), Figure S5). Interestingly, studies of SAMs formed from [M(typSH)(tpy)]2+ and other metal polypyridyl complexes (SI, Table S4) suggest that they may assemble differently from neutral platform molecules due to their ionic composition, exhibiting low surface coverages that indicate the presence of significant spacings between adjacent complexes in the monolayer (loose packing). Though this phenomenon has been attributed to electrostatic repulsion between the surface-bound charged complexes,33,34 the potential role of counterions in compensating those charges has not been explicitly considered. We reason that the sum of Coulomb interactions between all positively charged complexes and negatively charged counterions in such two-dimensional assemblies might be expected to confer a net favorable cohesive energy that results in a close-packed lattice,35 as is well-known, for example, for the ions of Ca2+ and F– in fluorite (CaF2, a prototypical three-dimensional ionic solid).36 Further studies of charged SAMs constructed from metal polypyridyl complexes are needed to help assess the influence of electrostatic interactions on their properties, as well as the potential utility of these modular redox-active building blocks as platform components in mixed assemblies.

Accordingly, here we explore the properties of monolayers prepared using a recently introduced28 family of air-stable metal bis(terpyridine) complexes, [M(tpySS)2](PF6)2 (MSS; M = Fe, Co, Zn, Ru; tpySS = 4′-(methyldisulfide)-2,2′:6′,2″-terpyridine) and [Ru(tpySS)(tpy)](PF6)2 (RuSS-1; tpy = 2,2′:6′,2″-terpyridine) (Figure 1a). These comprise either two or one methyldisulfide group(s) for binding to gold, respectively. Surface voltammograms obtained for mixed MSS/RuSS-1 SAMs confirm these do indeed exhibit compositions comparable to the solutions used to prepare them (a prerequisite for perfect mixing), as well as saturation surface coverages several times lower than the theoretical close-packed limit (Figure 1b). Remarkably, high-resolution X-ray photoelectron studies of these SAMs reveal featureless F 1s spectra, providing no evidence for the proliferation of PF6– counterions on the surface. To help interpret these findings, we develop a general electrostatic model capable of probing trends in the cohesive energy of two-dimensional (2D) ionic lattices with different structures. This model shows that the low surface coverages observed may only be easily rationalized if counterions are lost from complexes upon surface binding, or fixed in a specific unfavorable geometry, whereby repulsive complex-complex Coulomb interactions dominate the Madelung field. First-principles calculations based on density functional theory (DFT) further indicate that counterion loss is facilitated by charge transfer to the gold surface, which serves to weaken complex–counterion binding energies. This implicates such processes may play an important role in the formation of MSS/RuSS-1 SAMs. More broadly, our findings suggest that long-range electrostatic forces within charged monolayers could be generally exploited as part of an alternative, potentially complementary, approach to the platform strategy for controlling SAM structure. As is apparent from this work, such interactions could be used to organize components by their charge state, and/or influence whether these pack closely (with attractive interactions) or further apart (with repulsive interactions) on a surface. Consequently, they should also serve to mitigate the influence of short-range intermolecular van der Waals, π-stacking, or hydrophobic interactions that otherwise direct the structure of neutral (noncharged) SAMs.

Figure 1 (a) Molecular structure of the [M(tpySS)2](PF6)2 (MSS, where M = Fe, Co, Zn, Ru, tpySS = 4′-(methyldisulfide)-2,2′:6′,2′′-terpyridine, R = −SSMe) and [Ru(tpySS)(tpy)](PF6)2 (RuSS-1, where tpy = 2,2′:6′,2′′-terpyridine, R = −H) complexes studied here. (b) Single- and multicomponent self-assembled monolayers (SAMs) of MSS/RuSS-1 are reproducibly formed on gold surfaces following their exposure to 1 mM acetonitrile solutions for ≥18 h. Analysis of surface voltammograms reveal these SAMs exhibit low saturation surface coverages, consistent with an intercomponent spacing of ∼1 nm (assuming a homogeneous hexagonal close-packed structure, possible intercomplex surface-blocking adsorbates are omitted for clarity). (c) We propose that repulsive intermolecular electrostatic interactions serve to laterally space complexes in MSS/RuSS-1 SAMs following the loss of mobile PF6– counterions.

Results and Discussion

Synthesis and Electrochemical Characterization

In a recent study,28 we showed that the tpySS ligand can be utilized for the preparation of a new family of sulfur-functionalized redox-active MSS (M = Fe, Co, Zn) complexes. These compounds were found to be air-stable in solution for ≥7 days, in contrast to thiol-functionalized analogues,28,29 and capable of forming single-component SAMs on gold surfaces. Here we report that analogous RuSS-1 and RuSS complexes may also be synthesized from tpySS and RuCl2(DMSO)(tpy) following the method of Ziessel et al.,37 using reaction temperatures that do not exceed 60 °C. These compounds, comprising d6 Ru2+ second-row transition metal ions, are expected to exhibit a greater stability than their first-row analogues and so provide an important point of comparison.38−40CoSS and ZnSS complexes, for example, have d7 Co2+ or d10 Zn2+ centers which are known to be particularly kinetically labile,41,42 facilitating the dissociation and exchange of ligands between complexes in solution at room temperature. We recognize that such ligand dissociation processes have the potential to influence the composition of SAMs formed from CoSS and ZnSS solutions if free ligand effectively competes with coordination of the intact complexes to the surface, complicating the interpretation of monolayer properties. However, as detailed below, we find no evidence for the incorporation of free ligand in MSS SAMs by X-ray photoelectron spectroscopy (XPS), and the comparable surface coverages of all MSS and RuSS-1 SAMs serves to further support the assertion that these monolayers comprise intact complexes.

In Figure 2a, we plot overlaid solution cyclic voltammograms for MSS (M = Fe, Co, Zn) and RuSS-1. The voltammogram for RuSS is shown in the SI, Figure S2c for clarity. For ease of comparison, solution and surface voltammograms for MSS (M = Fe, Co, Zn) are reproduced here from an earlier report.28 Critically, we maintain the potential window above approximately –1.4 V vs the FcH/[FcH]+ redox couple in voltametric studies of these complexes to avoid irreversible reduction of the disulfide group.28 Within this window, which also lies positive of any anticipated gold-sulfur reductive desorption processes,43 we observe reversible redox events for all complexes except ZnSS, which is redox silent over this potential range. These features are assigned to the respective M2+/3+ couple through comparison to the well-studied solution voltammograms of the parent [M(tpy)2]2+ complexes (SI, Figure S2a and Table S1). In the SI, Figure S2b–c, we overlay additional solution voltammograms for RuSS-1 and RuSS in which we extend the electrochemical window to greater reduction potentials. As previously observed for MSS with other metals,28 these voltammograms comprise additional features that we associate with the formation of thiolate and thiyl radical species following disulfide reduction (selected electrochemical data is provided in the SI, Table S2). For RuSS-1, a complex with only one disulfide group, the integrated charge ratio for the disulfide reduction and Ru2+/3+ peak features is 1:1, confirming that disulfide reduction in these complexes is a 1 e– process.28

Figure 2 (a) Overlaid solution cyclic voltammograms (MeCN–0.1 M nBu4NPF6) for MSS where M = Fe (purple dashed), Co (orange solid), and Zn (green dotted),28 and for RuSS-1 (red short dashed). Redox features observed for FeSS, CoSS, and RuSS-1 are attributed to the M2+/3+ couples. The potential window is maintained above approximately –1.4 V vs the FcH/[FcH]+ redox couple to avoid irreversible reduction of the disulfide group. (b) Overlaid representative surface cyclic voltammograms (CH2Cl2–0.1 M nBu4NPF6) for FeSS, CoSS,28 and RuSS-1 SAMs on gold. For each SAM a single redox feature is observed close to the potential of the corresponding M2+/3+ couple measured in solution, indicating that the complex is attached to the surface. (c) Representative surface cyclic voltammogram (CH2Cl2–0.1 M nBu4NPF6) for a binary FeSS-CoSS SAM formed from a solution containing FeSS and CoSS in a 1:1 molar ratio, showing distinct redox features with approximately equal peak areas. (d) A plot showing that the surface coverage of CoSS (ΓCoSS) in mixed CoSS-ZnSS SAMs increases in proportion to the mole fraction of CoSS in the CoSS-ZnSS solution used to prepare the SAM. Data in panels (a) and (b) are reproduced in part from ref (28). with permission from the Royal Society of Chemistry.

The voltammograms shown in Figure 2a and SI, Figure S2c indicate that SAMs of FeSS, CoSS, and now also RuSS and RuSS-1, may be characterized using surface voltammetry through analysis of their distinct M2+/3+ redox features. To confirm this, we first study single-component SAMs formed by exposing mechanically polished gold disc electrodes to 1 mM solutions of each complex in MeCN for ≥ 18 h. In Figure 2b and SI, Figure S3a, we plot overlaid representative surface cyclic voltammograms for these modified electrodes measured in CH2Cl2–0.1 M nBu4NPF6 (equilibrium voltammograms are presented unless otherwise stated). As previously noted for CoSS and FeSS,28 reversible redox features indicative of adsorbed RuSS and RuSS-1 are also clearly observed at potentials close to the potential of the corresponding M2+/3+ couple measured in solution (ipa/ipc is typically close to 1 for redox-active SAMs, ip ∝ Vs; selected data are provided in the SI, Tables S1 and S3). As also observed for CoSS and FeSS SAMs,28 the peak width and intensity of Ru2+/3+ features for a RuSS-1 SAM do not significantly vary upon repeated potential cycling (SI, Figure S3b). This further highlights the broad stability of these systems with respect to electrochemical characterization and shows that surface-adsorbed MSS/RuSS-1 complexes remain capable of associating and dissociating with solution-based counterions upon the application of an external redox potential that changes their charge state.

Having established that MSS and RuSS-1 complexes can form single-component monolayers on gold surfaces, we next investigated their capacity to form mixed SAMs. In Figure 2c, we plot a surface cyclic voltammogram for a binary FeSS-CoSS SAM prepared from a solution comprising each complex in a 1:1 molar ratio. This voltammogram shows two well-separated features of approximately equal peak areas that we assign to the distinct Co2+/3+ and Fe2+/3+ redox processes expected for each component in the SAM. The total coverages calculated from summing the integrated charge of each feature in these multicomponent SAMs is ∼40 pmol/cm2 (using the average real surface area of the electrodes), in good agreement with typical coverages of pure MSS and RuSS-1 SAMs (SI, Table S3). We also formed binary SAMs using solutions containing redox-active CoSS and redox-inactive ZnSS in different molar ratios (1:0, 3:1, 1:1, 1:3). Following Gang et al.,1 but now using voltammetry, here we evaluate the use of adsorbed ZnSS complexes as a matrix diluent where, for example, they may play a role analogous to alkanethiols coadsorbed with ferrocene-terminated alkanethiols.44,45 For each CoSS-ZnSS SAM, we calculate the surface coverage of CoSS (ΓCoSS) using the integrated charge of the Co2+/3+ redox feature (representative voltammograms are shown in the SI, Figure S4). We plot, in Figure 2d, ΓCoSS as a function of its mole fraction in each SAM preparation solution. Each data point represents the average surface coverage obtained from a set of three SAM-modified electrodes prepared at the same time (errors bars = 1 s.d.), and we measure two independent electrode sets (6 total electrode measurements) for each solution composition. A linear fit to all data points (dotted line, y-axis intercept fixed to 0) shows a clear positive correlation. Such a relationship is consistent with the reported properties of other platform-based SAMs containing two components in different molar ratios.1,21 While some scatter in our data set is evident, most data points fall close to the expected variation in Γ(CoSS) when accounting for the ±1 s.d. error in the average real surface area of the electrodes (gray shaded region around the linear fit). Outliers are attributed to >1 s.d. variations in the mechanically polished electrode real surface area, or adventitious impurities that may influence the surface adsorption and packing density of SAM components.

As noted above, the saturation surface coverages we observe are consistent with reported values for other SAMs comprising charged polypyridyl complexes but much lower than the coverages calculated for monolayers packed at their steric limit (SI, Table S4 and Figure S5). Using a hexagonally close-packed structure comparable to that observed for analogous SAMs in scanning tunneling microscope imaging experiments,34 and approximating each complex as a circle of radius 0.6 nm (assuming free rotation of the complex on the surface), we estimate the maximum surface coverage (Γ) for a MSS SAM to be ∼133 pmol/cm2 (SI, Figure S6a). We obtain Γ ∼ 40 pmol/cm2 only for an adapted model that uses an intercomplex centroid-centroid distance of ∼2.2 nm (intercomplex separation of ∼1 nm; Figure 1b and SI, Figure S6b). Remarkably, only 27% of the total surface area is occupied in the adapted model compared to 91% in the close-packed system. To confirm the low-coverages we obtain for MSS SAMs are not due to recurring electrode contamination or other experimental artifact, we also prepared SAMs comprising 6-(ferrocenyl)hexanethiol (FcSH-1). In the SI, Figure S8a we plot overlaid representative cyclic voltammograms for CoSS and FcSH-1 SAMs on gold disc electrodes, showing that CoSS SAMs exhibit a significantly smaller Co2+/3+ redox feature compared to the corresponding Fe2+/3+ redox feature for FcSH-1 SAMs (ratio of integrated charge = ∼6.8; SI, Table S3). The value we obtain for ΓFcSH-1 = ∼270 pmol/cm2 (based on the real electrode surface area), is in good agreement with reported values for FcSH-1 SAMs formed on flat, template-stripped, gold substrates (ΓFcSH-1 = ∼300 pmol/cm2, calculated using the geometric electrode area).46 It should be noted that the low-coverages obtained for these MSS SAMs may not be attributed to use of the -SSMe linker group, as SAMs formed from analogous complexes comprising –SH anchors exhibit similar coverages.1,28

In support of the assertion that such charged SAMs loosely pack due to unfavorable electrostatic interactions between charged components,33,34 we note that the full width half-maximum (fwhm) for surface voltametric peaks from all MSS and RuSS-1 SAMs, ΔEfwhm, is between 170–240 mV (SI, Table S3). An ΔEfwhm > 90 mV (the ideal width for noninteracting redox sites) indicates significant lateral repulsive interactions within the adsorbed layer.34 The redox potentials of MSS/RuSS-1 complexes in SAMs are larger than that measured in solution by 50–140 mV, indicating they are more difficult to oxidize as a result of smaller average intercomplex distances and unfavorable lateral electrostatic forces (SI, Tables S1 and S3). To help further quantify these repulsive interactions, we apply a method introduced by Laviron to calculate the interaction energy (W) for CoSS SAMs by using the ΔEfwhm of the surface voltammetric peak (see SI for further discussion).47 This approach, which assumes a random distribution of adsorbed species in the monolayer, provides WCoSS = −5.4 kJ mol–1. This value is comparable to that obtained from studies of an Os(II) bis(terpyridine) complex bound to a Pt electrode via a pendant pyridyl group, using a more complex model that explicitly takes into account the geometric arrangement of components on the surface (W = −6.5 kJ mol–1, voltammetry performed in CH2Cl2–0.1 M nBu4NPF6).34 Drawing inspiration from such earlier studies which also found differences between interaction energies measured in different solvents,34 we reasoned it should be possible to increase the surface coverage of such charged SAMs by forming them in solvent environments that better screen electrostatic interactions. However, preliminary studies show that CoSS SAMs prepared and measured in MeCN–0.1 M nBu4NPF6 (a medium with higher ionic atmosphere and an excess of solution-based PF6– ions) exhibit comparable surface coverages to those prepared in pure MeCN and measured in CH2Cl2–0.1 M nBu4NPF6 (SI, Table S3; comparable redox potentials are also observed in each case).

We further considered that if these MSS SAMs are loosely packed due to repulsive Coulomb interactions between components, they should comprise exposed regions of gold surface available for the binding of uncharged species. To test this hypothesis, we first subjected single-component CoSS and FeSS SAMs formed over 60 s or ≥18 h to repeated immersions (3 × 10 s) in a 0.1 mM solution of FcSH-1 in MeCN. After each immersion, we characterized the modified electrode by surface voltammetry (see SI, Figure S8b–e for all stepwise voltammograms). Remarkably, we observe only a small additional redox feature that we attribute to either the binding of FcSH-1 at SAM defect sites or by displacement of adsorbed species. The magnitude of this redox feature is comparable to that observed for a surface saturated hexanethiol (C6SH-1) SAM exposed to FcSH-1 (SI, Figure S9a). In contrast, we observe the growth of a much larger redox feature for an unmodified gold electrode subjected to the same immersion conditions, corresponding to the rapid adsorption of FcSH-1 at uncoordinated surface sites (SI, Figure S9b). We also probed the properties of a CoSS SAM that was repeatedly immersed into a 0.1 mM solution of C6SH-1 in MeCN. Interestingly, surface voltammograms did not show any significant reduction in electrode double layer capacitance that would indicate substantial absorption of an insulating alkane layer (SI, Figure S8d). Together, these studies show that while the coverage of MSS appears low based on integration of the M2+/3+ redox feature, vacant surface sites are either not present or not easily accessible to solution alkanethiols (discussed further below).

X-ray Photoelectron Spectroscopy

We further characterized these SAMs using XPS to probe all surface-adsorbed species, not only the electroactive component(s) evaluated in surface voltammetry experiments. In Figure 3 we plot representative overlaid high-resolution N 1s, S 2p, Co 2p, and F 1s spectra for CoSS powder on carbon tape (left) and single-component CoSS SAMs on Au/silicon substrates (middle). In each case we observe a single N 1s peak at >399.5 eV that we attribute to coordinated terpyridine ligands, given that these binding energies are ∼1 eV higher than the dominant peak of a tpySS SAM control which is assigned to unbound pyridyl groups (SI, Figure S10f).48,49 Importantly, we also find the dominant Co 2p features in these spectra have binding energies that are consistent with either Co2+ or Co3+ (>779.5 eV).50 While these features exclude the presence of significant metallic Co in CoSS SAMs, the low surface concentration of cobalt and the resulting low signal-to-noise ratio of the Co 2p spectra makes the precise oxidation state of cobalt ions difficult to assign unambiguously. Analogous N 1s and M 2p/3d XPS features are identified for other MSS/RuSS-1 SAM samples, as shown in the SI, Figure S11 and Figure 3 (right). We acquired spectra for RuSS-1 rather than RuSS SAMs to facilitate comparisons between monolayers formed from complexes comprising 1 or 2 surface binding groups, and because RuSS-1 could be isolated in higher purity than RuSS. Selected data from all spectra is summarized in the SITables S5–S7. These XPS studies, taken together with the single prominent M2+/3+ redox events observed for each SAM in surface voltammetry studies (Figure 2b), provide strong evidence for the presence of intact complexes and the absence of uncoordinated tpySS ligands in MSS/RuSS-1 monolayers.

Figure 3 High-resolution X-ray photoelectron spectra for the N 1s, S 2p, Co 2p or Ru 3d, and F 1s regions of CoSS powder (left), SAM (middle), and RuSS-1 SAM (right) samples. Signals for N, S, and Co/Ru are clearly observed in each case (a–i), and a strong F 1s peak associated with PF6– counterions is obtained for the CoSS powder sample (j). However, the F 1s region is typically featureless for MSS/RuSS-1 SAM samples (k–l), indicating the absence of PF6– counterions in the adsorbed surface layer. These findings are consistent with XPS studies of other MSS SAMs (SI, Figure S11). Selected peak fitting parameters and assignments for all spectra are provided in the SI, Table S5-S7. In (i) we mark the Ru 3d5/2 spectral peak with an arrow, noting that Ru 3d and C 1s features are found in a similar spectral region.

We further assign the F 1s peak measured for the CoSS powder sample to PF6– counterions and note that this appears at a significantly higher intensity than the peaks observed for Co 2p, S 2p, or N 1s. This is consistent with fluorine’s relatively high atomic sensitivity factor (Co > F > S > N)50 and abundance in this complex (the chemical formula of CoSS is C38H35CoF12N9P2S4). In stark contrast, for CoSS SAMs, and most other MSS/RuSS-1 SAM samples, we observe a featureless F 1s region that strongly indicates PF6– counterions are absent in the adsorbed surface layer (Figure 3 and SI, Figure S11). While we do measure a low intensity F 1s signal for the FeSS SAM sample presented, this is only found in combination with an otherwise uncharacteristic S 2p doublet at 163.86 eV that we assign to a small proportion of physisorbed molecules which have not been removed by solvent washing (discussed further below; see SI, Figure S11c,g).51,52 We reason that the absence of a F 1s signal for these SAMs cannot readily be attributed to the in situ reduction of M2+ to M0 during XPS measurements (allowing counterions to diffuse away), or other instrumental artifact, as the binding energies of M 2p/3d features are characteristic of those metals in higher oxidation states, and the F 1s spectra obtained for FeSS SAMs and CoSS powder demonstrate that fluorine peaks for PF6– counterions can be clearly resolved for complexes that are not directly bound to the gold surface.

The S 2p regions for MSS/RuSS-1 SAMs exhibit more complex spectra than for the other elements discussed above, comprising multiple overlapping doublets that are consistent with the presence of different sulfur environments. To aid in the interpretation of these spectra, we first characterized a series of relevant organic SAM controls. In the SI, Figure S10b, we present a S 2p spectrum for a SAM formed from 1-decanethiol (C10SH-1) on template-stripped gold (AuTS), which shows the expected single doublet at ∼162 eV attributed to bound S.53,54 The corresponding spectra for a 1,10-decanedithiol (C10SH) SAM shows this feature in addition to the anticipated doublet at 163.6 eV associated with the unbound -SH groups of the standing up phase.55 We note that the larger intensity of the bound S component (60%) relative to the unbound S peak feature (40%) indicates that this C10SH SAM comprises a mixture of standing up and lying down phases, as expected from earlier studies.56−58 Importantly, we also studied SAMs formed from 1,10-decanedimethyldisulfide (C10SS, structure shown in the SI, Figure S10a), a molecule comprising the same binding groups as for MSS/RuSS-1 but with an alkane backbone. The S 2p spectrum for C10SS SAMs is complex, comprising the same two doublets observed for C10SH that would be expected if this SAM also comprises a mixture of standing up and lying down phases (161.9 and 163.1 eV with a ∼5:3 intensity ratio), in addition to two new doublets centered around 161 and 168 eV (SI, Figure S10c). As these new features are not observed for long-chain alkanethiol/alkanedithiol SAMs, they appear to be characteristic of SAMs formed from molecules functionalized with the methyldisulfide group. To corroborate this observation, in the SI, Figure S10d we plot the S 2p spectrum for a tpySS SAM formed on an Au/silicon substrate. This spectrum is comparable to that obtained for SAMs of C10SS except that it is missing the doublet associated with unbound sulfur as tpySS binds to gold through a single sulfur functionality. Possible assignments of these different peak features are discussed below.

Consistent with our XPS studies of C10SS and tpySS SAMs, we also observe doublets at ∼161 and ∼168 eV in the S 2p spectra of SAMs formed from methyldisulfide-functionalized CoSS, FeSS, and ZnSS. A feature associated with bound S is also clearly observed at ∼162 eV for MSS/RuSS-1 SAMs, but we do not typically resolve a doublet at ∼163 eV associated with pendant (undercoordinated) organosulfur groups. This suggests that both sulfur groups of MSS complexes are bound to gold on these atomically rough surfaces (root mean squared roughness ∼4.5 nm),59 or that pendant disulfide moieties have been oxidized (in line with the discussion below). While it was not possible to utilize our AuTS substrates (root mean squared roughness ∼0.6 nm)59 for XPS studies of MSS/RuSS-1 SAMs due to the incompatibility of the adhesive with MeCN, future studies using atomically flat substrates could help expose any possible influences of surface roughness on the properties of these monolayers. Though we obtained a better fit for the S 2p spectrum of RuSS-1 SAMs without explicitly including a doublet at ∼161 eV, we recognize that this feature could be present but difficult to resolve given the low signal-to-noise ratio of these measurements and the associated challenges with accurately fitting laboratory-based XPS data. A reduced intensity ∼ 161 eV feature could also be reflective of the 1:1 ratio of M2+ to -SSMe in RuSS-1 SAMs compared to the 1:2 ratio in MSS SAMs. For completeness, we note that in Figure 3d we fit the S 2p region of the CoSS powder sample with two doublets at 163.4 (the major component) and 164.2 eV (minor) to maintain peak widths ∼1 eV consistent with fits of other samples (peak width = ∼1.5 eV when only a single doublet is used). We hypothesize that the apparently distinct sulfur environments present here may reflect differential charging of the insulating powder sample.60,61

S 2p features around 161 and 168 eV have been observed in XPS studies of other gold-sulfur SAMs, although these have not yet been unambiguously assigned. For example, the 161 eV doublet was seen in spectra of SAMs prepared from solutions containing 4-pyridylthiol/bis(4-pyridyl)disulfide,62 and short chain alkanethiols/disulfides such as ethanethiol53 or MeSSMe.63 This feature was also observed in studies of SAMs formed on substrates immersed for short times in octanethiol64 or dithiolreitol65 solutions (1 and 20 min, respectively). The 161 eV doublet has often been associated with the adsorption of atomic sulfur species, given that S 2p doublets are also observed around this binding energy after exposure of gold surfaces to Na2S.66 It was initially proposed that such sulfide adsorbates formed from alkanethiols/disulfides through processes involving scission of the C-S bond.62,67 However, it was later recognized that these more likely originate from the competitive adsorption of adventitious sulfide impurities (discussed further below), particularly for analytes that form monolayers of reduced stability (e.g., those with short rather than long alkyl chains).53,68 The 161 eV doublet has also been assigned to alkanethiol species bound to gold in an sp rather than sp3 configuration,64,69 coordination geometries that have been suggested to occur in low coverage monolayers during the early stages of assembly.

S 2p doublets at ∼168 eV are typically observed for gold-sulfur SAMs exposed to laboratory environments for extended periods (e.g., 12 h,70 6 d71 exposure) or irradiated with UV light.72,73 These peaks are assigned to the presence of sulfonates (–SO3–)73−75 or sulfinates (–SO2–),76,77 following seminal studies that applied laser-induced desorption mass spectroscopy to characterize the adsorbed components of these assemblies. XPS features at 168 eV have also been reported for freshly prepared biphenyldithiol SAMs,78 and multilayer alkanedithiol SAMs reported to comprise interlayer, readily oxidized, disulfide groups.79 Notably, features attributable to oxidized sulfur are observed even for SAMs stored in the dark, thought to originate from reactions of the surface-bound species with adventitious ozone (O3).71,80 It has been suggested that different laboratories may observe different rates of sulfur oxidation for air-exposed samples, given that their local O3 concentrations are expected to vary.80 Less ordered SAMs, such as those prepared from short chain alkanes, have been found to oxidize more rapidly with UV irradiation or over time in air due to the easier penetration of O2/O3 into the adsorbed layer.72,80

Taking these previous studies into consideration, the X-ray photoelectron spectra obtained for saturation coverage MSS/RuSS-1 SAMs formed in air are consistent with SAM structures that result from the initial adsorption of complexes to their electrostatically spaced, loosely packed limit, which provide vacant, intercomplex gold surface sites that facilitate the subsequent binding of -SMe (from dissociation of -SSMe groups) or adventitious sulfides, disulfides, or thiols. Critical in this context, Porter et al. have convincingly demonstrated that the presence of adventitious disulfide and thiol impurities in a sample of ethyl phenyl sulfide were responsible for S 2p XPS and reductive desorption features consistent with those expected for a chemisorbed gold-sulfur SAM.53 Ethyl phenyl sulfide, after rigorous purification by gas chromatography, was shown in that same study to be incapable of forming a chemisorbed SAM on gold. In the present study, we reason that the binding of such adventitious surface-bound redox-inactive species block the adsorption of significant quantities of FeSH-1/C6SH-1 in the postassembly immersion studies detailed above. Interestingly, we also note that the in situ formation of iron bis(terpyridine) SAMs through reactions between surface-bound free terpyridine ligands and an iron(II) tetrafluoroborate hexahydrate salt have also resulted in SAMs with low saturation surface coverages (40–85 pmol/cm2).81 This is consistent with the hypothesis that the coverages of such SAMs are limited by intercomplex electrostatic repulsive forces rather than the specific mechanism of SAM formation.

Our XPS studies show that such MSS/RuSS-1 SAMs are more readily oxidized upon exposure to the ambient laboratory atmosphere than the more ordered SAMs formed from long-chain alkanethiols/alkanedithiols.72,80 Oxidized sulfur species also rapidly form in C10SS and tpySS SAMs as coassembly of each component of the dissociated disulfide also increase the degree of disorder in these monolayers relative to those formed from C10SH-1 or C10SH. As noted above, any surface adsorbed -SMe resulting from dissociated -SSMe may also be expected to oxidize rapidly given that the small methyl group cannot readily impede access of O2/O3 to the surface.72,80 In this context, we recognize that if –SO2– or –SO3– groups are present in the SAM they may themselves serve as counterions for the 2+ metal complexes, resulting in the loss of PF6– counterions through anion exchange and subsequently influencing SAM surface coverages. However, comparable low surface coverages are observed even for CoSS and [Co(typSH)2](PF6)2 (CoSH) SAMs formed and studied by surface voltammetry in the absence of air,28 and for other polypyridyl SAMs with nitrogen-based surface binding groups (SI, Table S4), which do not comprise –SO2– or –SO3– to perform this function. It is therefore apparent that the oxidation of sulfur groups, or the presence or absence of dissociated surface-bound –SMe groups, is not the primary driver of the assembly process that results in the formation of loosely packed polypyridyl SAMs. The comparable surface coverages of SAMs exposed to air or handled exclusively under inert conditions, as determined by surface voltammetry,28 also indicate that the air-oxidation of sulfur species in MSS/RuSS-1 SAMs has minimal impact on the stability of adsorbed complexes under these measurement conditions. Future studies of these SAMs, for example using contact angle goniometry,11,15,16 surface infrared spectroscopy,15,16 or quantitative XPS,82 could provide valuable additional insights into their composition and surface structure.

Electrostatic Model

To better understand the influence of Coulomb interactions on the organization of MSS SAMs, we develop an electrostatic model to qualitatively explore trends in the cohesive (lattice) energies of charged monolayers with different structures. Here, each monolayer is represented as a 2D array of point charges, whereby the sum of attractive and repulsive interactions between all ions in the lattice can be calculated. Our approach enables us to extract the Madelung constant (α) for each 2D system using a shell expansion method,83 where α has a value that is unique to a particular lattice geometry and is independent of the distance between nearest neighbor ions (d).84,85 If α is known, the electrostatic energy per formula unit (EES) for that lattice can be calculated to a first approximation using eq 1 (where Zi = charge on ion i, e = electronic charge, ε0 = vacuum permittivity; this expression excludes short-range repulsive interactions).36 While the use of α to evaluate the properties of 3D ionic crystals such as sodium chloride (αNaCl = 1.7476) or fluorite (αCaF = 2.5194) is well established,84 for example, to calculate lattice energies for ionic solids,36,85 its utility has been only rarely considered in the context of charged surface adsorbates.86 For the ionic lattices studied here (having identical Zi, Zj, and d), we use α to directly compare their electrostatic stability (EES ∝ α). In a related approach, we also use our model to evaluate how EES for different 2D arrays varies with changing intercomplex distance. If EES decreases (becomes more negative) with decreasing distance, it indicates the close-packed array of complexes is more stable than the isolated complex (and vice versa). Further background and model calculations detailing our methodology can be found in the SI. We note that more complex electrostatic models of ions close to an interface have been developed,87 for example, to understand the interfacial properties of ionic liquids, which could be exploited to provide additional insights for charged SAMs in subsequent work.1

We first define, in Figure 4a, the relative bond distances (d) and angles (θ, θ′, ϕ) between the A2+ ([M(tpy)2]2+) and X– (PF6–) components of the AX2 complexes in our simulated MSS monolayers. Here the z-axis is oriented parallel to the surface normal, and the surface lies in the x–y plane. In Figure 4b, we present a schematic showing an illustrative arrangement of these components for an isolated AX2 complex adsorbed on a metal surface (θ = 60°, θ′ = 120°, ϕ = 90°). Here the metal-surface and metal-counterion distances (0.8 and 0.6 nm, respectively), were fixed at values close to those found in computational models and single-crystal X-ray structure data for MSS (see below and the SI, Figure S1). To calculate the individual Coulomb interaction between A2+ and X– ions (I) we use the vacuum permittivity (ε0) unless otherwise stated (EI-I = q1q2e2/4πε0d). While this simplification is not expected to impact qualitative trends, it is expected to overestimate EES as in the real system EI-I are screened by the solvent and molecular dielectric (ε0εr ≠ 1) which will vary for individual ion–ion interactions as a function of complex-complex distance and the A2+-X– geometry (changing θ, θ′, ϕ). Due to the polarizability of the gold substrate, modeled as an ideal metal, we also incorporate image charges (IC) as equal and opposite charges of the A2+ (2– image charge) and X– (1+ image charge) components located at positions corresponding to their mirror reflection in the surface plane. We calculate the energies of ion-image charge interactions as half the Coulomb energy between two point charges (EI-IC = q1q2e2/8πε0d), and assume image charge-image charge interactions are perfectly screened (ε0εr = ∞, EIC-IC = 0) within the metal.87,88 The interactions between the complex and its induced image charges serve to slightly increase the electrostatic stabilization of the adsorbed complex compared to the isolated (gas phase) species (−3.500 vs – 3.535 e2/[4πε0dA-X] for the geometry shown in Figure 4b), which can be considered a result of the effective solvation of the complex by the mobile electrons in the metal. Notably, we find that image charges influence the absolute EES for an adsorbed, isolated AX2 complex with different θ/θ′. However, as discussed further below, these energy changes are small relative to the changes associated with the formation of close-packed 2D arrays (SI, Figures S13 and S14), making it energetically favorable for X– ions in proximity to other AX2 complexes to move to adopt lower energy lattice configurations (Figure S15).

Figure 4 (a) Definitions of the bond angles and lengths for AX2 complexes used in our electrostatic model. (b) An illustrative arrangement of charged components in a surface-bound complex (top) and their corresponding image charges in the metal substrate (bottom). Here, θ = 60°, θ′ = 120°, ϕ = 90 (complex viewed along the y-axis). (c) We generate different 2D hexagonal close-packed arrays of AX2 complexes using a custom computer algorithm, modeling each component as a point charge (sphere diameters not to scale). These arrays are expanded with successively larger hexagonal shells (dotted gray), such that for shell = 1, 2, 3, ... we add 1, 6, 12, ... AX2 complexes, respectively. The complex geometry and distance between each complex (dA-A) can be varied. For illustrative purposes, we show here a 3-shell array (dA-A = 2.4 nm) comprising complexes with the geometry shown in (b) but ϕ = 60° (image charges excluded for clarity). (d) We calculate the Madelung constant (α) for arrays of increasing size until this converges. At convergence, α can be considered representative of the bulk value (the infinite array). Here we present the result for the close-packed analogue of the array shown in (c) (θ = 60°, θ′ = 120°, ϕ = 60°, dA-A = 1.2 nm). Increasing α (Δα = αCP – αisol. = +ve) with increasing array size indicates the 2D lattice is electrostatically stabilized relative to the isolated complex. (e) Left: Schematic adsorbed complexes of the same geometry as in (c) and (d) with 2 (AX2), 1 (AX1+), or 0 (A2+) counterions. Right: Overlaid plots of electrostatic energy (per adsorbed complex, normalized to the energy of the isolated adsorbed complex) as a function of dA-A for AX2, AX1+, and A2+ arrays. Increasing energy with decreasing dA-A indicates the close-packed lattice is energetically more favorable than the isolated complex (and vice versa). Lines mark dA-A estimated from voltammetric analysis (gold dashed), and the close-packed limit (black dotted).

We apply our algorithm to generate 2D hexagonal close-packed AX2 arrays with specific complex geometries and intercomplex distances (dA-A). The size of each array is determined by adding a discrete number of hexagonal AX2 shells around a central AX2 unit (Figure 4c, gray dotted lines indicate shells 2 and 3). This expansion process allows us to systematically calculate α for arrays of increasing size until the value converges. At convergence, α may be considered representative of the bulk value (an infinite array). In Figure 4d we plot α against the number of AX2 units in each array size for arrays where θ = 60°, θ′ = 120°, ϕ = 60°, dA-A = 1.2 nm (a close-packed analogue of the array shown in Figure 4c). We obtain αCP = 1.9741 for the 2D array compared to αisol. = 1.7673 for the isolated adsorbed complex. The increase in α with increasing array size (Δα = αCP – αisol. = +0.2068) indicates that the AX2 complexes are electrostatically stabilized through formation of the 2D close-packed lattice (ΔEES = – 95.77 kJ mol–1). As noted above, we also calculate how EES for this AX2 array varies with changing dA-A, normalizing this to the energy of the isolated adsorbed complex for ease of comparison. This data, plotted in Figure 4e (right), shows that the normalized energy increases with decreasing dA-A, again indicating that the close-packed array is energetically more favorable than the isolated complex. These results firmly challenge the notion that ionic MSS (AX2) complexes should exhibit repulsive electrostatic interactions in a close-packed SAM if mobile counterions are present.

While the AX2 geometries used in Figure 4 serve as useful illustrative examples, the degree of electrostatic stabilization, as well as the energy-dA-A relationship, is strongly dependent on both the position and number of counterions in the lattice. For example, in the SI, Figure S14 we propose a more complex close-packed 2D AX2 lattice geometry, featuring a hexagonal unit cell comprising 4 × AX2 units and θ = θ′ = 90°, that yields a significantly larger EES. In contrast, model AX2 lattices based on the DFT optimized geometry for an isolated CoSS complex (with θ = θ′ ∼ 50°, see below) predicts that these X– positions, if fixed, will destabilize the close-packed lattice (SI, Figure S15a–c). Critically, however, we know from surface electrochemical studies that we have a surface coverage corresponding to dA-A = 2.2 nm for a hexagonal lattice (SI, Figure S6). If we model the same, electrostatically destabilized, lattice geometry used in SI, Figure S15a–c with dA-A = 2.2 nm, we find that there is a large electrostatic driving force for the X– ions to move to positions with θ = θ′ ∼ 90° to form a lattice that is more energetically stable than the isolated complex (SI, Figure S15d). Further electrostatic energy gains are subsequently available by decreasing dA-A = 2.2 nm toward dA-A = 1.2 nm (the close-packed limit), forming the lattice geometry identified in SI, Figure S14. Accordingly, if mobile counterions are present in MSS/RuSS-1 SAMs, we cannot easily rationalize why these would be loosely packed due to unfavorable electrostatic interactions.

Motivated by the XPS studies described above, we further utilize our algorithm to model the potential influence of counterion loss in MSS SAMs. Starting with the same AX2 lattice geometry used in Figure 4e (θ = 60°, θ′ = 120°, ϕ = 60°), we perform additional normalized energy-dA-A analyses now after removing 1 (AX1+) or 2 (A2+) of the X– components (schematic structures shown in Figure 4e, left). Both close-packed (dA-A = 1.2 nm) AX1+ and A2+ lattices are electrostatically destabilized relative to their isolated adsorbed complexes. This finding supports the assertion that counterion loss from adsorbed MSS complexes will favor low-coverage SAMs, in agreement with both the surface voltammetric and XPS analysis presented above. We further note that for AX1+ and A2+ our model predicts changes in EES at dA-A > 2.2 nm (yellow dashed line in Figure 4e), the intercomplex distance calculated from the experimentally determined surface coverages of MSS SAMs. This correlation further helps strengthen associations between the low coverage of MSS SAMs and repulsive electrostatic interactions between charged AX1+ or A2+ components in these adsorbed layers. Coulombic forces driving the separation of SAM components may be partially offset by factors that energetically favor close-packed layers, such as forming the maximum number of Au-S bonds.

Computational Studies

Our ab initio calculations further support the experimental results, demonstrating that complex–counterion binding is weakened by charge transfer to the gold substrate. The calculations provide a detailed atomistic description of the interactions between the surface, the [Co(tpySH)2]2+ = CoSH2+ complex (applied here as a simplified model of CoSS2+) and the PF6– counterions, including electrostatic, covalent and van der Waals interactions. Detailed explanations of the computational methodologies are included in the SI. First, we examined several possible PF6– counterion positions: “top” – both PF6– ions are placed above CoSH2+ toward vacuum Figure S17a; “diagonal” – one PF6– ion is placed above CoSH2+ toward vacuum, and the other one is placed near the Au surface, Figure S17b; “bottom” – both PF6– ions are placed below CoSH2+ near the Au surface, Figure S17c. In all cases, the initial placement of the molecule was set as a bidentate bridging interaction between the S atom and the Au surface, in an fcc chemisorption site, shown to be a favorable position in previous studies for thiols on Au(111).89−92 Upon geometry optimization, the structures exhibited a tridentate-type interaction between the S atom and the Au surface. The calculated energies of each configuration revealed that the “top” orientation was the lowest energy structure. The “diagonal” orientation is 12.35 kJ/mol higher in total energy than the “top” configuration; this energy difference is about 5 times greater than the thermal Boltzmann energy at room temperature (∼5 × kBT). The “bottom” configuration is 60.32 kJ/mol higher in total energy than the “top” configuration; this energy difference is about 15 times greater than the thermal Boltzmann energy at room temperature (∼15 × kBT). The significant differences in the energies of the various configurations of the PF6– counterions highlight a lower probability of the “diagonal” and “bottom” structures in experimental samples, at least for isolated complexes.

The binding energy between CoSH2+ on Au and the PF6– counterions is explored to understand the conditions in which counterions can leave CoSH2+. Given the lower energetically favorable configuration of “top,” it was used when calculating the binding energy between CoSH2+ on the Au surface and the PF6– counterions. Figure 5 shows that under neutral conditions of the system, the binding between the PF6– counterions and the CoSH2+ + Au complex is favorable by –244 kJ/mol. Furthermore, when the counterions are displaced in the out-of-plane direction to the Au surface (PF6 ions move away from the Au surface), the binding becomes weaker but remains favorable over long distances. However, when the system is negatively charged (total charge = –1e) the binding between the PF6– counterions and the CoSH2+ + Au complex is unstable by +191 kJ/mol. When the counterions are displaced, the binding becomes even more unstable. Thus, the calculations show that the counterions can unbind and leave CoSH2+, when the system becomes sufficiently negatively charged. The instability arises because negatively charged Au repulses PF6– anions, Table S13.

Figure 5 Comparison of systems with (a) two, (b) one, and (c) zero PF6– counterions. Detailed data are provided in the SI, Tables S12–S14 and Figures S17–S18. (d) Energy of binding between the Co complex on Au and the PF6– counterions. The lowest energy structure is shown in (a). The insert shows a representative structure with the counterions near the Co core (left), and away from the complex (right). The binding is a favorable under neutral conditions –244 kJ/mol (orange). However, under negatively charged conditions (single electron added to simulation) the binding is unfavorable +191 kJ/mol (blue). As the counterions are displaced in the z-direction perpendicular to Au surface, the binding in the neutral system remains favorable over long distances, while the binding in the charged system is always unfavorable. Zero displacement corresponds to the lowest energy structure (a).

To examine the effect of counterion loss on system stability, we calculate the charge and energetics of the systems with 1 and 0 counterions, Table S13. For a given number of counterions, the system is more stable if the removed counterion is charged, i.e., PF6– is removed and the total charge of the remaining system becomes positive. Adding a negative charge to the system with both counterions present is energetically unfavorable. Considering the energy per atom values reported in Table S13, we find that the total stability increases as we remove PF6– anions. I.e., the energy of the system with both counterions present, −405 (kJ/mol)/atom, is less favorable than the energies of the systems with one or two PF6– anions removed, −521 (kJ/mol)/atom, total charge = +1 and −636 (kJ/mol)/atom, total charge = +2, respectively. This is consistent with the experimental results reported above and can be rationalized by the favorable charge redistribution between CoSH2+ and Au.

As a further explanation for the lack of the PF6– signal in the experimental samples, we calculated the binding energy of PF6– ions of the Au slab model, Figure S18. When the simulation cell (Au + PF6– ions) is charge-neutral (the total charge of the simulation cell = 0), the Au slab is positively charged (+1.8e), while the PF6 ions are negatively charged (−0.9e), resulting in a favorable binding between the ions and the surface (−675 kJ/mol), Table S14. However, when the total charge of the simulation system is negative (−2e), such as when charged PF6– counterions approach neutral Au, Au accepts some of the charge of the counterions and becomes negatively charged itself (−0.15e). Negatively charged Au repulses PF6– anions, resulting in an unstable binding energy (+67,346 kJ/mol). These results demonstrates that when PF6– anions leave CoSH2+, they will not be able to bind to the Au surface, further supporting the experimental lack of PF6– on the sample surfaces.

Additional Discussion

We gain further insights into the role of counterions on MSS SAM structure by considering the organization of these ionic complexes in the solid-state. While adjacent [M(tpySS)2]2+ in either a SAM or bulk material will certainly experience repulsive electrostatic interactions, single-crystal X-ray diffraction studies show these charged species readily pack closely in the presence of counterions (SI, Figure S1). Here, we again attribute this to the long-range Coulomb interactions between all [M(tpySS)2]2+ and PF6– ions in the infinite solid, which confer an overall favorable cohesive energy to this crystal lattice,35 as is well-known for the ions of Ca2+ and F– in fluorite (CaF2) and found above using our electrostatic model for 2D AX2 arrays (Figure 4 and SI, Figure S14). This observation further supports the view that the low coverages of SAMs formed from charged polypyridyl complexes cannot readily be attributed to unfavorable lateral electrostatic interactions unless, for example, the ion composition at the surface is substantially different from that of the crystalline solid. The X-ray crystal structure shown in SI, Figure S1 also reveals that PF6– ions are small enough to occupy free volume within the radius of the complex and are therefore not expected to appreciably impact the surface coverage of MSS SAMs due to their steric bulk. As suggested by our XPS studies and evaluated using the 2D electrostatic model described above, the properties of these MSS/RuSS-1 SAMs are easily rationalized if discreet mobile counterions are absent.

We stress that this is not the first time that counterion loss from adsorbed charged monolayers has been observed, indicating that this phenomenon is not specific to the complexes studied here and may be more widespread than previously recognized. For example, XPS and scanning tunneling microscopy (STM) studies of SAMs comprising TATA-based carbocation components have shown these can lose BF4– ions upon surface adsorption.6,93,94 In these reports it was noted that the loss of counterions implied a significant charge transfer to the substrate,6 and that ion dissociation may be favored by the electric field generated at an STM tip.94 Supporting computational experiments revealed that while a TATA+-BF4– ion pair is stabilized by adsorption on a gold surface, the cohesive energy of the system is reduced by addition of an electron (as we also show above).94 In contrast to SAMs formed from polypyridyl complexes, the potential influence of electrostatic interactions on surface coverage is likely more difficult to discern in these monolayers given that the intercomponent spacing of TATA components is larger, and also influenced by their peripheral long-chain alkyl substituents.14 In a different study, PF6– counterions were not detected by XPS in SAMs formed from a disulfide-functionalized [Ru(bimpyH2)2](PF6)2 complex (bimpyH2 = bis(benzimidazole-2yl)pyridine).95 Here, it was suggested that the benzimidazolyl N–H groups integrated into the complex were deprotonated in situ to maintain charge neutrality, rather than balancing this charge through reduction of the underlying substrate. Charge neutrality could, in principle, be similarly achieved in MSS complexes if both disulfide groups were cleaved to generate the corresponding bis(thione) species,29 or oxidized to a charged sulfinate or sulfonate (see XPS discussion above). However, given that this is not possible for RuSS-1 complexes which comprise only one disulfide group, and noting the reported poor air-stability of the iron bis(thione) complex,29 we do not consider this a viable explanation for charge balance in MSS/RuSS-1 SAMs when PF6– counterions are absent.

Though much of our discussion has focused on net unfavorable electrostatic interactions that increase the lateral spacing of surface-adsorbed components and destabilize a SAM, the same arguments suggest that net favorable electrostatic interactions should result in close-packed SAMs of increased stability. This has been demonstrated for a 3-mercaptopropionic acid SAM studied in aqueous alkaline solution.96 Here, the authors proposed that the formation of a surface-based two-dimensional ionic lattice, comprising deprotonated RCOO– and X+ (X = alkali metals or tetraalkylammonium salts) ionic species, confers strong lateral attractive interactions that make the SAM more difficult to remove from a gold surface by reductive desorption. The structure and properties of SAMs prepared using thiol-terminated dialkyl viologens is also reportedly influenced by favorable lateral electrostatic interactions with intercalating counterions, where the identity of the counterion is also thought to play a role.97−99 Here, for example, postassembly exchange of Cl– for ClO4– was found to result in a more compact, ordered monolayer that could impede permeation of solvent and electrolyte.

Finally, in seeking to rationalize the low coverage of MSS/RuSS-1 SAMs, we note that the electrostatic and DFT models discussed above implicate either: (1) the loss of PF6– counterions through charge transfer to the gold substrate; or (2) the persistence of energetically unfavorable counterion geometries in the assemblies of adsorbed complexes. However, the mechanism(s) and/or driving forces associated with either of these outcomes remains unclear. The electrochemical characterization of MSS/RuSS-1 complexes in solution suggest that only CoSS exhibits a redox potential low enough to transfer charge to the substrate through oxidation of M2+ to M3+, indicating these complexes in general do not provide an energetically accessible source of electrons and implicating the presence of a distinct and independent source of electrons common to all compounds studied. In these ambient temperature, solution-based experiments, we cannot rule out the possibility that impurities, present in either the solvent, sample, or on the gold surface, may serve as reducing agents which help to facilitate the loss of PF6– ions from MSS during SAM formation. Notably, the surface charge associated with a saturation coverage MSS2+ SAM on a 1 cm2 substrate is ∼7.7 × 10–6 C, which could be accommodated through reaction with just ∼80 picomoles of an adventitious reductant that may not easily be identified. Alternatively, we also recognize that ion-exchange between mobile PF6– ions and negatively charged ions with restricted mobility, perhaps coadsorbed on the gold surface, could result in ion geometries that favor the loose packing of complexes in a SAM.

Conclusions

In this work we have extended the SAM platform concept to air-stable, redox-active Fe(II), Co(II), and Ru(II), and redox-inactive Zn(II), bis(terpyridine) complexes comprising directly connected methyldisulfide surface binding groups. While single-component and binary mixed SAMs with tailored compositions are readily formed, the low saturation surface coverages of these SAMs, large ΔEfwhm, and shift in E1/2 upon surface binding also indicate that their mixing behavior is facilitated by repulsive electrostatic interactions that separate individual components in these monolayers by ∼1 nm, rather than due to comparable short-range forces between distinct components with comparable structures and dimensions. These observations reinforce the results of earlier investigations on SAMs constructed from related polypyridyl-based complexes (SI, Table S4),28 suggesting that the previously reported low coverages for these systems are not due to labile and/or peripheral molecule-surface binding groups, or the decomposition of complexes in solution. However, these studies also highlight important questions related to why repulsive interactions dominate in such 2D assemblies in the presence of charge-balancing counterions, particularly given the close packing of their constituent complexes in 3D as evidenced by single-crystal X-ray diffraction.

Remarkably, XPS studies of MSS/RuSS-1 SAMs provide no evidence for the presence of PF6– counterions, suggesting that their loose packing results from unscreened, lateral repulsive electrostatic interactions between adjacent 2+ charged metal bis(terpyridine) complexes on the surface. Here, using an electrostatic model to evaluate extended 2D ionic assemblies, in combination with first-principles calculations of isolated adsorbed complexes, we show that the stoichiometry and position of counterions indeed plays a critical role in the stability of close-packed charged MSS/RuSS-1 SAMs. We show that the relative energies of different counterion geometries are strongly influenced by electrostatic interactions not only between the ions in the complex, but also between ions and image charges in the conducting substrate. These models reveal feasible ionic arrangements that result in strongly net favorable electrostatic interactions between all charged species at the close-packed limit. As such, they firmly dispel the prevailing notion that such 2+ charged components should repel each other in a SAM if counterions are present, and support the hypothesis that the low coverage of MSS/RuSS-1 SAMs results from the loss of mobile PF6– counterions, facilitated either by charge transfer to the gold substrate (as suggested by our DFT calculations), or through exchange with surface immobilized ions that lock the 2D lattice into geometries which do not favor close packing. Further studies of these and analogous systems are needed to probe the mechanism of potential counterion loss, for example, through systematic studies which vary the complex-surface distance, ion identity, or surrounding solvent medium. We recognize that an electrode surface may only serve to help compensate the charge of proximal complexes, suggesting that charged components will better retain counterions, gain electrostatic stability, and form close-packed SAMs when attached at an interface via an extended tether.

Together, this work highlights the importance and potential utility of electrostatic forces, thus far widely neglected, for controlling the spacing, coverage, and stability of charged molecules adsorbed on surfaces. New design strategies which take these into account could produce monolayers in which component-component interactions are dominated by long-range Coulombic forces, rather than the short-range (van der Waals, π-stacking, hydrogen-bonding) interactions of common critical importance to the nanoscale arrangement of most charge-neutral, close-packed SAMs. Such forces may also play a critical role in future platform-inspired approaches, which we hope will help to provide new mechanisms for manipulating the composition and properties of mixed, functional group appended monolayers that enable new synthetic, mechanical, or electrical processes on molecular length scales.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacs.4c07327.Additional experimental details and discussion, synthetic, crystallographic, electrochemical, and XPS data, electrostatic model and computational calculations, as well as 1H and 13C{1H} NMR spectra for all new compounds (PDF)

Supplementary Material

ja4c07327_si_001.pdf

The authors declare no competing financial interest.

Acknowledgments

This work was supported by University of Southern California (USC) startup funds and the ACS Petroleum Research Fund (62751-DNI5). We thank Nils Rotthowe for useful discussions, Thomas Czyszczon-Burton for technical support with XPS measurements, and Latha Venkataraman and Jahan Dawlaty for helpful insights regarding image charges. 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 National Science Foundation (DBI-0821671, CHE-0840366) and National Institute of Health (S10 RR25432) supported the acquisition of the NMR spectrometers used in our work. O.V.P. acknowledges support of the US National Science Foundation (CHE-2154367). Data in panels (a) and (b) of Figure 2 are reproduced in part from ref (28) with permission from the Royal Society of Chemistry.
==== Refs
References

Gang T. ; Yilmaz M. D. ; Ataç D. ; Bose S. K. ; Strambini E. ; Velders A. H. ; De Jong M. P. ; Huskens J. ; Van Der Wiel W. G. Tunable Doping of a Metal with Molecular Spins. Nat. Nanotechnol. 2012, 7 (4 ), 232–236. 10.1038/nnano.2012.1.22306840
Atxabal A. ; Ribeiro M. ; Parui S. ; Urreta L. ; Sagasta E. ; Sun X. ; Llopis R. ; Casanova F. ; Hueso L. E. Spin Doping Using Transition Metal Phthalocyanine Molecules. Nat. Commun. 2016, 7 , 13751 10.1038/ncomms13751.27941810
Graham M. J. ; Zadrozny J. M. ; Fataftah M. S. ; Freedman D. E. Forging Solid-State Qubit Design Principles in a Molecular Furnace. Chem. Mater. 2017, 29 (5 ), 1885–1897. 10.1021/acs.chemmater.6b05433.
Inkpen M. S. ; Leroux Y. R. ; Hapiot P. ; Campos L. M. ; Venkataraman L. Reversible On-Surface Wiring of Resistive Circuits. Chem. Sci. 2017, 8 (6 ), 4340–4346. 10.1039/C7SC00599G.28660061
Hutt D. A. ; Leggett G. J. Functionalization of Hydroxyl and Carboxylic Acid Terminated Self-Assembled Monolayers. Langmuir 1997, 13 (10 ), 2740–2748. 10.1021/la970069t.
Jung U. ; Kuhn S. ; Cornelissen U. ; Tuczek F. ; Strunskus T. ; Zaporojtchenko V. ; Kubitschke J. ; Herges R. ; Magnussen O. Azobenzene-Containing Triazatriangulenium Adlayers on Au(111): Structural and Spectroscopic Characterization. Langmuir 2011, 27 (10 ), 5899–5908. 10.1021/la104654p.21506548
Akiyama H. ; Tamada K. ; Nagasawa J. ; Abe K. ; Tamaki T. Photoreactivity in Self-Assembled Monolayers Formed from Asymmetric Disulfides Having Para-Substituted Azobenzenes. J. Phys. Chem. B 2003, 107 (1 ), 130–135. 10.1021/jp026103g.
Salaita K. ; Amarnath A. ; Maspoch D. ; Higgins T. B. ; Mirkin C. A. Spontaneous “Phase Separation” of Patterned Binary Alkanethiol Mixtures. J. Am. Chem. Soc. 2005, 127 (32 ), 11283–11287. 10.1021/ja042393y.16089456
Stranick S. J. ; Parikh A. N. ; Tao Y.-T. ; Allara D. L. ; Weiss P. S. Phase Separation of Mixed-Composition Self-Assembled Monolayers into Nanometer Scale Molecular Domains. J. Phys. Chem. A 1994, 98 (31 ), 7636–7646. 10.1021/j100082a040.
Imabayashi S.-i. ; Hobara D. ; Kakiuchi T. ; Knoll W. Selective Replacement of Adsorbed Alkanethiols in Phase-Separated Binary Self-Assembled Monolayers by Electrochemical Partial Desorption. Langmuir 1997, 13 , 4502–4504. 10.1021/la970447u.
Bain C. D. ; Whitesides G. M. Molecular-Level Control over Surface Order in Self-Assembled Monolayer Films of Thiols on Gold. Science 1988, 240 (4848 ), 62–63. 10.1126/science.240.4848.62.17748822
Bruening M. ; Cohen R. ; Guillemoles J. F. ; Moav T. ; Libman J. ; Shanzer A. ; Cahen D. Simultaneous Control of Surface Potential and Wetting of Solids with Chemisorbed Multifunctional Ligands. J. Am. Chem. Soc. 1997, 119 (24 ), 5720–5728. 10.1021/ja964434z.
Shon Y.-S. ; Lee S. ; Perry S. S. ; Lee T. R. The Adsorption of Unsymmetrical Spiroalkanedithiols onto Gold Affords Multi-Component Interfaces That Are Homogeneously Mixed at the Molecular Level. J. Am. Chem. Soc. 2000, 122 (7 ), 1278–1281. 10.1021/ja991987b.
Baisch B. ; Raffa D. ; Jung U. ; Magnussen O. M. ; Nicolas C. ; Lacour J. ; Kubitschke J. ; Herges R. Mounting Freestanding Molecular Functions onto Surfaces: The Platform Approach. J. Am. Chem. Soc. 2009, 131 (2 ), 442–443. 10.1021/ja807923f.19113847
Park J.-S. ; Smith A. C. ; Lee T. R. Loosely Packed Self-Assembled Monolayers on Gold Generated from 2-Alkyl-2-Methylpropane-1,3-Dithiols. Langmuir 2004, 20 (14 ), 5829–5836. 10.1021/la036424z.16459597
Park J.-S. ; Vo A. N. ; Barriet D. ; Shon Y.-S. ; Lee T. R. Systematic Control of the Packing Density of Self-Assembled Monolayers Using Bidentate and Tridentate Chelating Alkanethiols. Langmuir 2005, 21 (7 ), 2902–2911. 10.1021/la0475573.15779965
Srisombat L.-o. ; Zhang S. ; Lee T. R. Thermal Stability of Mono-, Bis-, and Tris-Chelating Alkanethiol Films Assembled on Gold Nanoparticles and Evaporated “Flat” Gold. Langmuir 2010, 26 (1 ), 41–46. 10.1021/la902082j.19791779
White K. E. ; Avery E. M. ; Cummings E. ; Hong Z. ; Langecker J. ; Vetushka A. ; Dušek M. ; Macháček J. ; Višnák J. ; Endres J. ; Bastl Z. ; Mete E. ; Alexandrova A. N. ; Baše T. ; Weiss P. S. Competing Intermolecular and Molecule–Surface Interactions: Dipole–Dipole-Driven Patterns in Mixed Carborane Self-Assembled Monolayers. Chem. Mater. 2024, 36 (4 ), 2085–2095. 10.1021/acs.chemmater.3c03210.
Ito M. ; Wei T. X. ; Chen P. L. ; Akiyama H. ; Matsumoto M. ; Tamada K. ; Yamamoto Y. A Novel Method for Creation of Free Volume in a One-Component Self-Assembled Monolayer. Dramatic Size Effect of Para-Carborane. J. Mater. Chem. 2005, 15 (4 ), 478–483. 10.1039/b411121d.
Tao N. J. Probing Potential-Tuned Resonant Tunneling through Redox Molecules with Scanning Tunneling Microscopy. Phys. Rev. Lett. 1996, 76 (21 ), 4066–4069. 10.1103/PhysRevLett.76.4066.10061183
Santos L. ; Mattiuzzi A. ; Jabin I. ; Vandencasteele N. ; Reniers F. ; Reinaud O. ; Hapiot P. ; Lhenry S. ; Leroux Y. ; Lagrost C. One-Pot Electrografting of Mixed Monolayers with Controlled Composition. J. Phys. Chem. C 2014, 118 (29 ), 15919–15928. 10.1021/jp5052003.
Valášek M. ; Mayor M. Spatial and Lateral Control of Functionality by Rigid Molecular Platforms. Chem. - Eur. J. 2017, 23 (55 ), 13538–13548. 10.1002/chem.201703349.28766790
Chinwangso P. ; Jamison A. C. ; Lee T. R. Multidentate Adsorbates for Self-Assembled Monolayer Films. Acc. Chem. Res. 2011, 44 (7 ), 511–519. 10.1021/ar200020s.21612198
Eckermann A. L. ; Feld D. J. ; Shaw J. A. ; Meade T. J. Electrochemistry of Redox-Active Self-Assembled Monolayers. Coord. Chem. Rev. 2010, 254 (15 ), 1769–1802. 10.1016/j.ccr.2009.12.023.20563297
Wei J. ; Diaconescu P. L. Redox-Switchable Ring-Opening Polymerization with Ferrocene Derivatives. Acc. Chem. Res. 2019, 52 (2 ), 415–424. 10.1021/acs.accounts.8b00523.30707548
Lorkovic I. M. ; Duff R. R. ; Wrighton M. S. Use of the Redox-Active Ligand 1,1′-Bis(Diphenylphosphino)Cobaltocene To Reversibly Alter the Rate of the Rhodium(I)-Catalyzed Reduction and Isomerization of Ketones and Alkenes. J. Am. Chem. Soc. 1995, 117 (12 ), 3617–3618. 10.1021/ja00117a033.
Gregson C. K. A. ; Gibson V. C. ; Long N. J. ; Marshall E. L. ; Oxford P. J. ; White A. J. P. Redox Control within Single-Site Polymerization Catalysts. J. Am. Chem. Soc. 2006, 128 (23 ), 7410–7411. 10.1021/ja061398n.16756273
Trang C. D. M. ; Saal T. ; Inkpen M. S. Methyldisulfide Groups Enable the Direct Connection of Air-Stable Metal Bis(Terpyridine) Complexes to Gold Surfaces. Dalton Trans 2023, 52 , 7836–7842. 10.1039/D3DT00955F.37218422
Van Der Geer E. P. L. ; Van Koten G. ; Klein Gebbink R. J. M. ; Hessen B. A [4Fe–4S] Cluster Dimer Bridged by Bis(2,2′:6′,2″-Terpyridine-4′-Thiolato)Iron(II). Inorg. Chem. 2008, 47 (7 ), 2849–2857. 10.1021/ic702062q.18330985
Storrier G. D. ; Colbran S. B. ; Craig D. C. Bis[4′-(4-Anilino)-2,2′:6′,2″- Terpyridine]Transition-Metal Complexes: Electrochemically Active Monomers with a Range of Magnetic and Optical Properties for Assembly of Metallo Oligomers and Macromolecules. J. Chem. Soc., Dalton Trans. 1997, 3011–3028. 10.1039/a702778h.
Winter A. ; Schubert U. S. Metal-Terpyridine Complexes in Catalytic Application – A Spotlight on the Last Decade. ChemCatChem 2020, 12 (11 ), 2890–2941. 10.1002/cctc.201902290.
Nishihara H. ; Kanaizuka K. ; Nishimori Y. ; Yamanoi Y. Construction of Redox- and Photo-Functional Molecular Systems on Electrode Surface for Application to Molecular Devices. Coord. Chem. Rev. 2007, 251 (21–24 ), 2674–2687. 10.1016/j.ccr.2007.04.002.
Campagnoli E. ; Hjelm J. ; Milios C. J. ; Sjodin M. ; Pikramenou Z. ; Forster R. J. Adsorption Dynamics and Interfacial Properties of Thiol-Based Cobalt Terpyridine Monolayers. Electrochim. Acta 2007, 52 (24 ), 6692–6699. 10.1016/j.electacta.2007.04.096.
Figgemeier E. ; Merz L. ; Hermann B. A. ; Zimmermann Y. C. ; Housecroft C. E. ; Gu H.-J. ; Constable E. C. Self-Assembled Monolayers of Ruthenium and Osmium Bis-Terpyridine Complexes Insights of the Structure and Interaction Energies by Combining Scanning Tunneling Microscopy and Electrochemistry. J. Phys. Chem. B 2003, 107 (5 ), 1157–1162. 10.1021/jp026522d.
Izgorodina E. I. ; Bernard U. L. ; Dean P. M. ; Pringle J. M. ; MacFarlane D. R. The Madelung Constant of Organic Salts. Cryst. Growth Des. 2009, 9 (11 ), 4834–4839. 10.1021/cg900656z.
West A. R. Solid State Chemistry and Its Applications, 2nd ed.; John Wiley & Sons Ltd: United Kingdom, 2014.
Ziessel R. ; Grosshenny V. ; Hissler M. ; Stroh C. Cis-[Ru(2,2′:6′,2′′-Terpyridine)(DMSO)Cl2]: Useful Precursor for the Synthesis of Heteroleptic Terpyridine Complexes under Mild Conditions. Inorg. Chem. 2004, 43 (14 ), 4262–4271. 10.1021/ic049822d.15236539
Knaak T. ; González C. ; Dappe Y. J. ; Harzmann G. D. ; Brandl T. ; Mayor M. ; Berndt R. ; Gruber M. Fragmentation and Distortion of Terpyridine-Based Spin-Crossover Complexes on Au(111). J. Phys. Chem. C 2019, 123 (7 ), 4178–4185. 10.1021/acs.jpcc.8b11242.
Benavides P. A. ; Matias T. A. ; Araki K. Unexpected Lability of the [Ru III (Phtpy)Cl 3] Complex. Dalton Trans. 2017, 46 (44 ), 15567–15572. 10.1039/C7DT03658B.29091091
Bark T. ; von Zelewsky A. ; Rappoport D. ; Neuburger M. ; Schaffner S. ; Lacour J. ; Jodry J. Synthesis and Stereochemical Properties of Chiral Square Complexes of Iron(II). Chem. – Eur. J. 2004, 10 (19 ), 4839–4845. 10.1002/chem.200400399.15372687
Bark T. ; Düggeli M. ; Stoeckli-Evans H. ; Von Zelewsky A. Designed Molecules for Self-Assembly: The Controlled Formation of Two Chiral Self-Assembled Polynuclear Species with Predetermined Configuration. Angew. Chem., Int. Ed. 2001, 40 (15 ), 2848–2851. 10.1002/1521-3773(20010803)40:15<2848::AID-ANIE2848>3.0.CO;2-S.
Constable E. C. ; Housecroft C. E. ; Kulke T. ; Lazzarini C. ; Schofield E. R. ; Zimmermann Y. Redistribution of Terpy Ligands—Approaches to New Dynamic Combinatorial Libraries. J. Chem. Soc., Dalton Trans. 2001, 19 , 2864–2871. 10.1039/b104865c.
Widrig C. A. ; Chung C. ; Porter M. D. The Electrochemical Desorption of N-Alkanethiol Monolayers from Polycrystalline Au and Ag Electrodes. J. Electroanal. Chem. Interfacial Electrochem. 1991, 310 (1–2 ), 335–359. 10.1016/0022-0728(91)85271-P.
Creager S. E. ; Rowe G. K. Redox Properties of Ferrocenylalkane Thiols Coadsorbed with Linear N-Alkanethiols on Polycrystalline Bulk Gold Electrodes. Anal. Chim. Acta 1991, 246 (1 ), 233–239. 10.1016/S0003-2670(00)80680-7.
Chidsey C. E. D. Free Energy and Temperature Dependence of Electron Transfer at the Metal-Electrolyte Interface. Science 1991, 251 (4996 ), 919–922. 10.1126/science.251.4996.919.17847385
Nerngchamnong N. ; Thompson D. ; Cao L. ; Yuan L. ; Jiang L. ; Roemer M. ; Nijhuis C. A. Nonideal Electrochemical Behavior of Ferrocenyl–Alkanethiolate SAMs Maps the Microenvironment of the Redox Unit. J. Phys. Chem. C 2015, 119 (38 ), 21978–21991. 10.1021/acs.jpcc.5b05137.
Laviron E. Surface Linear Potential Sweep Voltammetry: Equation of the Peaks for a Reversible Reaction When Interactions between the Adsorbed Molecules Are Taken into Account. J. Electroanal. Chem. Interfacial Electrochem. 1974, 52 , 395–402. 10.1016/S0022-0728(74)80449-3.
Traulsen C. H. H. ; Darlatt E. ; Richter S. ; Poppenberg J. ; Hoof S. ; Unger W. E. S. ; Schalley C. A. Intermixed Terpyridine-Functionalized Monolayers on Gold: Nonlinear Relationship between Terpyridyl Density and Metal Ion Coordination Properties. Langmuir 2012, 28 (29 ), 10755–10763. 10.1021/la301644r.22741945
Zubavichus Y. ; Zharnikov M. ; Yang Y. ; Fuchs O. ; Umbach E. ; Heske C. ; Ulman A. ; Grunze M. X-Ray Photoelectron Spectroscopy and Near-Edge X-Ray Absorption Fine Structure Study of Water Adsorption on Pyridine-Terminated Thiolate Self-Assembled Monolayers. Langmuir 2004, 20 (25 ), 11022–11029. 10.1021/la047980b.15568854
Moulder J. F. ; Stickle W. F. ; E Sobol P. ; Bomben K. D. Handbook of X-Ray Photoelectron Spectroscopy; Chastain J. , Ed.; Perkin-Elmer Corporation, 1992.
Jia J. ; Kara A. ; Pasquali L. ; Bendounan A. ; Sirotti F. ; Esaulov V. A. On Sulfur Core Level Binding Energies in Thiol Self-Assembly and Alternative Adsorption Sites: An Experimental and Theoretical Study. J. Chem. Phys. 2015, 143 (10 ), 104702 10.1063/1.4929350.26374051
Castner D. G. ; Hinds K. ; Grainger D. W. X-Ray Photoelectron Spectroscopy Sulfur 2p Study of Organic Thiol and Bisulfide Binding Interactions with Gold Surfaces. Langmuir 1996, 12 (21 ), 5083–5086. 10.1021/la960465w.
Zhong C.-J. ; Brush R. C. ; Anderegg J. ; Porter M. D. Organosulfur Monolayers at Gold Surfaces: Reexamination of the Case for Sulfide Adsorption and Implications to the Formation of Monolayers from Thiols and Disulfides. Langmuir 1999, 15 , 518–525. 10.1021/la980901+.
Bain C. D. ; Biebuyck H. A. ; Whitesides G. M. Comparison of Self-Assembled Monolayers on Gold: Coadsorption of Thiols and Disulfides. Langmuir 1989, 5 (3 ), 723–727. 10.1021/la00087a027.
Rieley H. ; Kendall G. K. ; Zemicael F. W. ; Smith T. L. ; Yang S. X-Ray Studies of Self-Assembled Monolayers on Coinage Metals. 1. Alignment and Photooxidation in 1,8-Octanedithiol and 1-Octanethiol on Au. Langmuir 1998, 14 (18 ), 5147–5153. 10.1021/la971183e.
Yu J.-J. ; Ngunjiri J. N. ; Kelley A. T. ; Garno J. C. Nanografting versus Solution Self-Assembly of α,ω-Alkanedithiols on Au(111) Investigated by AFM. Langmuir 2008, 24 (20 ), 11661–11668. 10.1021/la802235c.18823084
Bain C. D. ; Troughton E. B. ; Tao Y. T. ; Evall J. ; Whitesides G. M. ; Nuzzo R. G. Formation of Monolayer Films by the Spontaneous Assembly of Organic Thiols from Solution onto Gold. J. Am. Chem. Soc. 1989, 111 (1 ), 321–335. 10.1021/ja00183a049.
Akkerman H. B. ; Kronemeijer A. J. ; van Hal P. A. ; de Leeuw D. M. ; Blom P. W. M. ; de Boer B. Self-Assembled-Monolayer Formation of Long Alkanedithiols in Molecular Junctions. Small 2008, 4 (1 ), 100–104. 10.1002/smll.200700623.18098243
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.17696377
Greczynski G. ; Hultman L. X-Ray Photoelectron Spectroscopy: Towards Reliable Binding Energy Referencing. Prog. Mater. Sci. 2020, 107 , 100591 10.1016/j.pmatsci.2019.100591.
Cólon Santana J. A. Quantitative Core Level Photoelectron Spectroscopy: A Primer; IOP concise physics; Morgan & Claypool Publishers: San Rafael, CA, 2015.
Lamp B. D. ; Hobara D. ; Porter M. D. ; Niki K. ; Cotton T. M. Correlation of the Structural Decomposition and Performance of Pyridinethiolate Surface Modifiers at Gold Electrodes for the Facilitation of Cytochrome c Heterogeneous Electron-Transfer Reactions. Langmuir 1997, 13 (4 ), 736–741. 10.1021/la960637p.
Cometto F. P. ; Macagno V. A. ; Paredes-Olivera P. ; Patrito E. M. ; Ascolani H. ; Zampieri G. Decomposition of Methylthiolate Monolayers on Au(111) Prepared from Dimethyl Disulfide in Solution Phase. J Phys Chem C 2010, 114 (22 ), 10183–10194. 10.1021/jp912060e.
Ishida T. ; Hara M. ; Kojima I. ; Tsuneda S. ; Nishida N. ; Sasabe H. ; Knoll W. High Resolution X-Ray Photoelectron Spectroscopy Measurements of Octadecanethiol Self-Assembled Monolayers on Au(111). Langmuir 1998, 14 (8 ), 2092–2096. 10.1021/la971104z.
Kumar S. ; Soni S. ; Danowski W. ; Van Beek C. L. F. ; Feringa B. L. ; Rudolf P. ; Chiechi R. C. Correlating the Influence of Disulfides in Monolayers across Photoelectron Spectroscopy Wettability and Tunneling Charge-Transport. J. Am. Chem. Soc. 2020, 142 (35 ), 15075–15083. 10.1021/jacs.0c06508.32786759
Buckley A. N. ; Hamilton I. C. ; Woods R. An Investigation of the Sulphur(−II)/Sulphur(0) System on Gold Electrodes. J. Electroanal. Chem. Interfacial Electrochem. 1987, 216 (1 ), 213–227. 10.1016/0022-0728(87)80208-5.
Zhong C.-J. ; Porter M. D. Evidence for Carbon-Sulfur Bond Cleavage in Spontaneously Adsorbed Organosulfide-Based Monolayers at Gold. J. Am. Chem. Soc. 1994, 116 (25 ), 11616–11617. 10.1021/ja00104a071.
Yoshimoto S. ; Yoshida M. ; Kobayashi S. I. ; Nozute S. ; Miyawaki T. ; Hashimoto Y. ; Taniguchi I. Electrochemical Study on Competitive Adsorption of Pyridinethiol with Sulfide onto Au(111) Surfaces. J. Electroanal. Chem. 1999, 473 (1 ), 85–92. 10.1016/S0022-0728(99)00239-9.
Leavitt A. J. ; Thomas P. B. Jr. Chemical Reactivity Studies of Hydrogen Sulfide on Au(111). Surf. Sci. 1994, 314 , 23–33. 10.1016/0039-6028(94)90210-0.
Willey T. M. ; Vance A. L. ; Van Buuren T. ; Bostedt C. ; Terminello L. J. ; Fadley C. S. Rapid Degradation of Alkanethiol-Based Self-Assembled Monolayers on Gold in Ambient Laboratory Conditions. Surf. Sci. 2005, 576 (1–3 ), 188–196. 10.1016/j.susc.2004.12.022.
Lee M. T. ; Hsueh C. C. ; Freund M. S. ; Ferguson G. S. Air Oxidation of Self-Assembled Monolayers on Polycrystalline Gold: The Role of the Gold Substrate. Langmuir 1998, 14 (22 ), 6419–6423. 10.1021/la980724c.
Hutt D. A. ; Leggett G. J. Influence of Adsorbate Ordering on Rates of UV Photooxidation of Self-Assembled Monolayers. J. Phys. Chem. A 1996, 100 (16 ), 6657–6662. 10.1021/jp952734h.
Huang J. ; Hemminger J. C. Photooxidation of Thiols in Self-Assembled Monolayers on Gold. J. Am. Chem. Soc. 1993, 115 , 3342–3343. 10.1021/ja00061a048.
Tarlov M. J. ; Newman J. G. Static Secondary Ion Mass Spectrometry of Self-Assembled Alkanethiol Monolayers on Gold. Langmuir 1992, 8 (5 ), 1398–1405. 10.1021/la00041a026.
Li Y. ; Huang J. ; Mclver R. T. ; Hemminger J. C. Characterization of Thiol Self-Assembled Films by Laser Desorption Fourier Transform Mass Spectrometry. J. Am. Chem. Soc. 1992, 114 (7 ), 2428–2432. 10.1021/ja00033a018.
Garrell R. L. ; Chadwick J. E. ; Myles D. C. ; Severance D. L. ; McDonald N. A. Adsorption of Sulfur Containing Molecules on Gold: The Effect of Oxidation on Monolayer Formation and Stability Characterized by Experiments and Theory. J. Am. Chem. Soc. 1995, 117 (46 ), 11563–11571. 10.1021/ja00151a022.
Chadwick J. E. ; Myles D. C. ; Garrell R. L. Self-Assembly of Sulfinate Monolayers on Gold: New Membrane Mimetics. J. Am. Chem. Soc. 1993, 115 , 10364–10365. 10.1021/ja00075a064.
Azzam W. ; Wehner B. I. ; Fischer R. A. ; Terfort A. ; Wöll C. Bonding and Orientation in Self-Assembled Monolayer of Oligophenyldithiols on Au Substrates. Langmuir 2002, 18 (21 ), 7766–7769. 10.1021/la020426m.
Kohli P. ; Taylor K. K. ; Harris J. J. ; Blanchard G. J. Assembly of Covalently-Coupled Disulfide Multilayers on Gold. J. Am. Chem. Soc. 1998, 120 (46 ), 11962–11968. 10.1021/ja981987w.
Schoenfisch M. H. ; Pemberton J. E. Air Stability of Alkanethiol Self-Assembled Monolayers on Silver and Gold Surfaces. J. Am. Chem. Soc. 1998, 120 (18 ), 4502–5413. 10.1021/ja974301t.
Maeda H. ; Sakamoto R. ; Nishihara H. Rapid Electron Transport Phenomenon in the Bis(Terpyridine) Metal Complex Wire: Marcus Theory and Electrochemical Impedance Spectroscopy Study. J. Phys. Chem. Lett. 2015, 6 (19 ), 3821–3826. 10.1021/acs.jpclett.5b01725.26722877
Kim H. ; Colavita P. E. ; Paoprasert P. ; Gopalan P. ; Kuech T. F. ; Hamers R. J. Grafting of Molecular Layers to Oxidized Gallium Nitride Surfaces via Phosphonic Acid Linkages. Surf. Sci. 2008, 602 (14 ), 2382–2388. 10.1016/j.susc.2008.05.002.
Baker A. D. ; Baker M. D. Rapid Calculation of Individual Ion Madelung Constants and Their Convergence to Bulk Values. Am. J. Phys. 2010, 78 (1 ), 102–105. 10.1119/1.3243281.
Johnson Q. C. ; Templeton D. H. Madelung Constants for Several Structures. J. Chem. Phys. 1961, 34 (6 ), 2004–2007. 10.1063/1.1731810.
Waddington T. C. Lattice Energies and Their Significance in Inorganic Chemistry. Adv. Inorg. Chem. Radiochem. 1959, 1 (C ), 157–221. 10.1016/S0065-2792(08)60254-X.
Foulston R. ; Gangopadhyay S. ; Chiutu C. ; Moriarty P. ; Jones R. G. Mono- and Multi-Layer Adsorption of an Ionic Liquid on Au(110). Phys. Chem. Chem. Phys. 2012, 14 (17 ), 6054–6066. 10.1039/c2cp23901a.22441396
Kaiser V. ; Comtet J. ; Niguès A. ; Siria A. ; Coasne B. ; Bocquet L. Electrostatic Interactions between Ions near Thomas–Fermi Substrates and the Surface Energy of Ionic Crystals at Imperfect Metals. Faraday Discuss. 2017, 199 , 129–158. 10.1039/C6FD00256K.28436506
Griffiths D. J. Introduction to Electrodynamics, 4th ed., international ed.; Always learning; Pearson: Boston, 2013.
Baker T. A. ; Friend C. M. ; Kaxiras E. Atomic Oxygen Adsorption on Au(111) Surfaces with Defects. J. Phys. Chem. C 2009, 113 (8 ), 3232–3238. 10.1021/jp806952z.
Jianming C. ; Naijuan W. ; Shangxue Q. ; Kean F. ; Zei M. S. Chemisorption of Oxygen on Au(111) Surface. Chin. Phys. Lett. 1989, 6 (2 ), 92–95. 10.1088/0256-307X/6/2/012.
Grönbeck H. ; Curioni A. ; Andreoni W. Thiols and Disulfides on the Au(111) Surface: The Headgroup-Gold Interaction. J. Am. Chem. Soc. 2000, 122 (16 ), 3839–3842. 10.1021/ja993622x.
Santiago-Rodríguez Y. ; Herron J. A. ; Curet-Arana M. C. ; Mavrikakis M. Atomic and Molecular Adsorption on Au(111). Surf. Sci. 2014, 627 , 57–69. 10.1016/j.susc.2014.04.012.
Snegir S. ; Dappe Y. J. ; Sysoiev D. ; Huhn T. ; Scheer E. Nonuniform STM Contrast of Self-Assembled Tri-n-Octyl-Triazatriangulenium Tetrafluoroborate on HOPG. ACS Omega 2023, 8 (41 ), 38766–38772. 10.1021/acsomega.3c06454.37867726
Snegir S. ; Dappe Y. J. ; Sysoiev D. ; Pluchery O. ; Huhn T. ; Scheer E. Where Do the Counterions Go? Tip-Induced Dissociation of Self-Assembled Triazatriangulenium-Based Molecules on Au(111). Phys. Chem. Chem. Phys. 2021, 23 (16 ), 9930–9937. 10.1039/D1CP00221J.33861285
Haga M. A. ; Hong H. G. ; Shiozawa Y. ; Kawata Y. ; Monjushiro H. ; Fukuo T. ; Arakawa R. Synthesis and Proton-Coupled Electron-Transfer Reaction of Self-Assembled Monolayers of a Ruthenium(II) Complex Containing Tridentate 2,6-Bis(Benzimidazol-2-Yl)Pyridine on a Gold Surface: Comparison of Acid/Base Chemistry with Bulk Solution Chemistry. Inorg. Chem. 2000, 39 (20 ), 4566–4573. 10.1021/ic990934s.
Kitagawa Y. ; Hobara D. ; Yamamoto M. ; Kakiuchi T. Counterion Binding Induces Attractive Interactions between Negatively-Charged Self-Assembled Monolayer of 3-Mercaptopropionic Acid on Au(111) in Reductive Desorption. J. Solid State Electrochem. 2008, 12 (4 ), 461–469. 10.1007/s10008-007-0471-5.
Han B. ; Li Z. ; Wandlowski T. ; Błaszczyk A. ; Mayor M. Potential-Induced Redox Switching in Viologen Self-Assembled Monolayers: An ATR-SEIRAS Approach. J. Phys. Chem. C 2007, 111 (37 ), 13855–13863. 10.1021/jp073208g.
De Long H. C. ; Buttry D. A. Ionic Interactions Play a Major Role in Determining the Electrochemical Behavior of Self-Assembling Viologen Monolayers. Langmuir 1990, 6 (7 ), 1319–1322. 10.1021/la00097a022.
Li Z. ; Han B. ; Meszaros G. ; Pobelov I. ; Wandlowski T. ; Błaszczyk A. ; Mayor M. Two-Dimensional Assembly and Local Redox-Activity of Molecular Hybrid Structures in an Electrochemical Environment. Faraday Discuss. 2006, 131 , 121–143. 10.1039/B506623A.16512368
