Single Molecule Electrochemistry: Impact of Surface Site Heterogeneity

Nov 4, 2016 - ... https://cdn.mathjax.org/mathjax/contrib/a11y/accessibility-menu.js ..... All of the surface sites are treated theoretically by the r...
0 downloads 0 Views 2MB Size
Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Article

Single Molecule Electrochemistry: Impact of Surface Site Heterogeneity Bo Fu, Colin Van Dyck, Stephanie Zaleski, Richard P. Van Duyne, and Mark A. Ratner J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.6b05252 • Publication Date (Web): 04 Nov 2016 Downloaded from http://pubs.acs.org on November 7, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry C is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Single Molecule Electrochemistry: Impact of Surface Site Heterogeneity AUTHOR NAMES Bo Fu‡§, Colin Van Dyck†§*, Stephanie Zaleski†, Richard P. Van Duyne†¶& and Mark A Ratner† AUTHOR ADDRESS †

Department of Chemistry, Northwestern University, 2145 Sheridan Road, Evanston, Illinois, 60208, United States



Department of Physics and Astronomy, Northwestern University, 2145 Sheridan Road, Evanston, Illinois, 60208, United States

&

Department of Biomedical Engineering, Northwestern University, 2145 Sheridan Road, Evanston, Illinois, 60208, United States



Program in Applied Physics, Northwestern University, 2145 Sheridan Road, Evanston, Illinois, 60208, United States

Phone Number Colin Van Dyck,

847-467-4985 1

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract

Probing the electrochemistry of single molecules is a direct pathway towards a microscopic understanding of a variety of electron transfer processes related to energy science, such as electrocatalysis and solar fuel cells. In this context, Zaleski et al recently studied the single electron transfer reaction of the dye molecule rhodamine-6G (R6G) by electrochemical single molecule surface-enhanced Raman spectroscopy (EC-SMSERS) (J. Phys. Chem. C 2015, 119, 28226-28234). In that work, the reductions of the dye molecule R6G were not only observed in the same potential range as in the ensemble surface CV, but also seen under some less negative potentials. Aiming at understanding and explaining this experiment theoretically, we relate the binding energy of an adsorbed R6G+ on AgNP with its reduction potential and further use periodic density functional theory (DFT) to calculate this adsorption energy at different local surface sites. Several local surface sites, including well-defined crystal facets and defective surfaces, are considered. We find that the calculated adsorption energy distribution of the strongest binding states at each surface site closely matches the potential range of the experimentally observed Faradaic events. Moreover, the underpotential events are explained by the meta-stable adsorption states with less binding strength compared to those corresponding to Faradaic events. Our study reveals the importance of the heterogeneity of surface structures on an AgNP and offers a new perspective on understanding single molecule electrochemical behavior.

2

ACS Paragon Plus Environment

Page 2 of 30

Page 3 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

1. Introduction Electrochemistry is at the heart of various disciplines such as electrocatalysis, solar energy conversion, organic light emitting diodes and molecular electronic devices. Studying electrochemical reactions at the ensemble level has been established for several decades, which typically involves measuring the current induced by trillions of molecules. Considerable experimental and theoretical efforts towards a deeper understanding of nanoscale electrochemistry have started in the past two decades. The first detection of single molecule electrochemical process was achieved by Fan and Bard in 19951 using “redox cycling” method2-3, in which the weak current response from a single chemically reversible redox molecule was magnified by allowing the repeated reduction and oxidation of this molecule within a nanogap between a pair of biased electrodes. Instead of directly detecting Faradaic current of electroactive molecules4, state-of-art high resolution optical spectroscopies, such as single molecule fluorescence and surfaceenhanced Raman spectroscopy (SERS), can be employed for the electrochemical detection of a single molecule if the redox states of the molecule are characterized by different optical responses.5-6 For instance, the single molecule fluorescence spectroscopy has been applied in probing the electrochemical behavior of a dye molecule, cresyl violet, whose fluorescence intensity experience a sharp increase upon oxidation.7 Similarly, by examining Nile Blue with single molecule SERS (SMSERS) during an electrochemical cycle, the reduction and oxidation processes can be captured with the loss and reappearance of the SMSERS signal, respectively. 8-9

3

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

In

parallel

with

the

experimental

effort

towards

Page 4 of 30

investigating

nanoscale

electrochemistry, considerable efforts have been made in studying electrocatalytic processes on nanoparticles with density functional theory (DFT)10.

For instance,

extensive theoretical studies on the electrochemical behavior of reactions such as the hydrogen evolution reaction, which is critical in solar energy conversion processes, are reported in the literature.11-14 The catalytic mechanism is usually attributed to the existence of atomic level binding sites on the nanoparticle surface. However, these theoretical efforts are mainly focused on electrocatalytic processes on diatomic or triatomic molecules that possess very few degrees of freedom.

To the best of our

knowledge, little to no theoretical work exists to describe the electrochemical behavior of relatively large molecules observed in optical spectroscopy correlated electrochemical reactions.15-16

In this context, we focus on recent experimental work by Zaleski et al

17

, where the

authors examine single electron transfer events at the nanoscale using electrochemical single molecule surface-enhanced Raman spectroscopy (EC-SMSERS). Only a brief summary of the experiment is presented here; an extensive description can be found in reference 17. In the experiment, single R6G molecules are adsorbed to an Ag nanoparticle (AgNP) surface; the AgNPs are then covalently attached to an indium tin oxide (ITO) working electrode. The authors use an air-tight glass spectroelectrochemical cell in order to ensure the observation of single electron transfer events. A reduction or oxidation event of R6G is indicated by the disappearance or appearance of the R6G cation SMSERS signal during an electrochemical cycle, respectively. The SMSERS signal of a R6G is monitored as the applied potential is stepped from 0 V to -1.2 V and back to 0 V 4

ACS Paragon Plus Environment

Page 5 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

in 0.1 V increments; SMSERS signal loss events occur due to the reduction of the R6G molecule. The statistical distribution of the SMSERS signal loss potential, or reduction potential, is represented as a histogram as shown in Figure 1 in comparison to the bulk R6G surface cyclic voltammogram (CV). Careful examination of the ensemble surface CV of R6G in Figure 1C shows that the number of molecules that contribute to the cathodic peak experiences a sharp decrease compared to the oxidative peak. The authors attribute this to the fact that the R6G neutral radical species desorbs from the AgNP after reduction, which they also observe in the single molecule electrochemistry experiment R6G in a way that the SMSERS signal return during the backward potential scan is only seen in 2 out of 44 SMSERS signal loss events. The comparison between the SMSERS signal loss potential histogram and the ensemble surface CV shows that the majority of recorded SMSERS signal loss events are within the expected Faradaic potential window, however, some SMSERS signal loss events were found at less negative potentials that are beyond the Faradaic potential range (from - 1.2 V to - 0.7 V). The 3 events at – 0.5 V and - 0.6 V are believed to be the reduction events, as they appear at the tail of the Faradaic events distribution. However, the 4 events captured at much less negative potentials, 0.3 V and - 0.4 V, are more likely caused by some other process, since the reduction events distribution should decay exponentially toward the less negative potentials. These non-Faradaic events are out of the scope of this article, so only the reduction events found at - 0.5 V and - 0.6 V will be discussed in this paper. We refer to the 3 single molecule reduction events at less negative potentials as underpotential events by comparing them with the reduction peak in the ensemble surface CV. The reduction reaction in this underpotential range is less thermodynamically favorable, so in the bulk electrochemistry 5

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

only a very small amount of molecules contribute to the Faradaic current and their contribution cannot be seen in the ensemble surface CV due to the much larger capacitive double layer current. The authors postulate that the observed underpotential reduction events is due to local surface curvature features which are rich on the AgNP but minor on the bulk Ag surface.

Figure 1. (A)Schematic representing the one-electron redox reaction of R6G cation (R6G+) (left) to its neutral radical form (R6G·) (right). (B)Representative SMSERS spectra acquired at 532 nm of R6G+ (red) and R6G· (blue); the loss in R6G+ SMSERS signal signifies the electrochemical conversion of R6G+ to R6G·. The bottom schematic represents the sample used for SMSERS measurements, where the R6G+ molecules are adsorbed to AgNPs on an ITO working electrode. (C) SMSERS signal loss histogram as compared to a bulk surface cyclic voltammogram on an Ag electrode. This figure was adapted with permission from Reference 17. Copyright (2015) American Chemical Society. 6

ACS Paragon Plus Environment

Page 6 of 30

Page 7 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

The primary aim of this work is to develop a theoretical description of the origin of the distribution in the single molecule reduction histogram shown in Figure 1C. As suggested in previous studies17-18, we assume that the reduction potential distribution observed in SMSERS experiment originates from the various binding sites available for the R6G on an AgNP. To support this hypothesis, this work explores the electrochemical reduction of R6G on various Ag surface topologies with DFT calculations. We find a semiquantitative agreement between theory and experiment for the potential window of the experimentally determined Faradaic events. Furthermore, our approach allows for identifying a class of meta-stable adsorption states as the candidates for the experimentally observed underpotential events. Our work provides an ab initio understanding on how the electrochemical reactions at the nanoscale is connected to the nanoscale surface structures and the microscopic configurations.

2. Hypothesis and computational procedure The electrochemical reaction being studied in this work is  6  ads + → ← 6 ads (1)

where (ads) refers to the adsorbed R6G on the AgNP on which the electrochemical reduction occurs. The standard electrochemical potential   of such an electron transfer reaction is defined by the free energy difference of the reaction, i.e.   = −

∆ 

. An exact

free energy of the electrochemical system should include the contributions from all aspects, e.g. ionic vibrations, adsorbate’s translation and rotation, but these are often technically impossible tasks for large adsorbate like R6G. In order to capture the 7

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 30

dominant contributions, we simplify the theoretical framework by ignoring contributions from some aspects based on reasonable assumptions and experimental facts. First of all, the ITO substrate is excluded in our model. In the SMSERS experiment, the AgNP covalently attached to the ITO acts as the working electrode and is held at one of the constant potential steps while the reduction reaction occurs. Although ITO often has variations in the local conductivity19-20, we assume that the AgNPs involved in the observed SMSERS signal loss events are in perfect electrical contact with the substrate. This assumption is corroborated by the fact that there are some AgNP aggregates that undergo no SMSERS signal loss upon potential steps, indicating that they are on a nonconductive portion of the ITO17; all of the observed SMSERS loss events are thus assumed to be on conductive portions of the ITO and the AgNP is in complete electrical contact. Therefore, we assume here that the variation in the reduction potential for different events arises from the single variation in the electron affinity (EA) of the R6G+ adsorbed on the AgNP. Doing so, we evaluate the reduction potential of the R6G+ by its EA and assume that the solvent and entropic contributions to the reduction potential are the same for each event. This assumption is reasonable, because R6G is a relatively large molecule compared to size of the surface atom (Ag) and the disturbance from the solvent and thermal effect doesn’t significantly affect its energetics. The electron affinity of the adsorbed R6G is defined as the following:  =   −   + 1 ,

(2)

where  is the number of electrons in the adsorbed R6G. The total energy of the adsorbed R6G can be decomposed into the energy of the isolated molecule, 8

ACS Paragon Plus Environment

Page 9 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

 !" #6 / % , and the adsorption energy  which is defined as the total energy difference between the separated and combined R6G-AgNP systems. Therefore, the electron affinity associated with the R6G reduction reaction is given by:  = &'( R6G +  R6G on Ag / − &'( R6G +  R6G on Ag / (3) The adsorption energy of the R6G0 here is technically challenging to evaluate, mainly because the R6G0 is not stable on the neutral metallic surface, it exists only under an applied electrochemical potential. However, the experimentally observed desorption of R6G0 implies that only a very weak binding exists between the R6G0 and the AgNP upon reduction, so this experimental fact enables us to drop  R6G on Ag in equation (3) and focus on the major contribution in EA. The equation (3) is then reduced to  =  !" +  R6G on Ag , (4) where the electron affinity of the isolated R6G cation 1'( is a constant. Equation (4) provides a direct link between the experimentally measured reduction potential and the adsorption energy of the R6G cation on the Ag surface. We can therefore postulate that the distribution observed in the experimental single molecule reduction events will also be seen in the adsorption energies of the R6G+. The surface structure of the AgNPs used in the experimental study is much more complicated than a perfect single crystal surface; it includes not only well-defined crystal facets but also defect sites like kinks, steps, and corners. This complexity of the NP surface thus provides a wide spectrum of binding sites, leading to various molecular adsorption states. Several experimental studies have shown that metallic nanoparticles exhibit multiple crystal facets and defective surface structures.21-23 Some theoretical works further discovered that defect sites play an 9

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

important role in electrocatalysis. For instance, step sites contribute most to the overall rate of ammonia synthesis24 and the defective sites on AgNP are found more active than the close-packed [111] crystal facet for various reactions like O2 dissociation25.

Figure 2. We investigated the adsorption energies of the R6G cation on eight different types of surfaces that are illustrated above. Surface structures on the left depict welldefined, defect-free crystal facets: terrace-[111], terrace-[110], terrace-[100] and terrace-[211]. The Ag surface structures on the right-hand side depict defect-rich surface sites constructed from the terrace-[111] site. The total orientations listed for each Ag surface site represent the number of different R6G molecular configurations tested as starting geometries. A variety of binding sites on AgNP have been reported as active sites for electrocatalytic reactions in the literature.25-32 There are atop, bridge and hollow sites on well-defined crystal facets, and adatom, vacancy, step and kink sites on defective surface structures. Among all these available binding sites on an AgNP, we consider here eight specific surface sites, as illustrated in Figure 2: four well-defined, defect-free Ag crystal facets (terrace-[111], terrace-[110], terrace-[100] and terrace-[211]) and four defect-rich terrace-[111] sites (adatom, kink, vacancy or step). This choice of specific sites is 10

ACS Paragon Plus Environment

Page 10 of 30

Page 11 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

motivated by their frequent observation in reported experiments using AgNPs. The atop, bridge and hollow sites are not considered here for the adsorption of R6G+, as these atomic level sites only contribute to the variation of adsorption when the size of the adsorbate is comparable to the crystal lattice constant. All of the surface sites are treated theoretically by the repeated slab model with periodic boundary condition. We believe that the use of a periodic slab model is a robust approximation of the AgNP surface structure for 3 reasons. First, the AgNP employed in the experiment ranges from 40 nm to 60 nm in diameter 17, whereas the size of the R6G molecule is approximately 1 nm33. Second, both theoretical predictions by Plieth 34 and experimental findings from Ivanova and Zamborini 35 have shown that the AgNP exhibits a redox potential that approaches its bulk value by an accuracy of 0.01 eV as its size goes beyond 25 nm in diameter. Finally, a slab model is considered to be more accurate than a cluster model to evaluate adsorption energy.36-37 To model defect-free terrace-[111], terrace-[110], terrace-[100] and terrace-[211] sites, we employ five layers of Ag atoms constructed as face centered cubic (fcc) lattice crystal surfaces with the experimental lattice constant (4.09 Å) and the corresponding Miller indices. The defect-rich Ag surfaces are then constructed from the terrace-[111] site by adding a partially packed layer to create a kink, step, or adatom surfaces or by removing a single Ag atom to create a vacancy. Constructing the surface sites in this way ensures that there are always five Ag layers below the molecule at any position. Indeed, five Ag layers are found to be necessary to accurately describe the adsorption R6G+ on an Ag slab. An additional Ag layer only changes the calculated energies by less than 0.01 eV(see Supporting Information), which is an order of magnitude less than the potential 11

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

steps of 0.1 V used in the EC-SMSERS experiment. In the normal direction, the slabs are separated by a vacuum region with a thickness of 60 Å to exclude inter-slab interaction. A unit cell with a large enough lateral dimension is required to eliminate the interaction between the R6G cation and its neighbors along x and y directions. Considering the computational cost, a large unit cell size of 20 Å × 20 Å × 60 Å is found to be adequate. The total energy of the R6G+ in such a unit cell only differs from that of an ideal isolated R6G+ by less than 0.03 eV (see Supporting Information). Therefore, we employ repeated (7×7) unit cell for terrace-[110] site and repeated (9×8) unit cell for all the other terrace sites and defect sites, which guarantees a large enough unit cell for all the surface sites. The non-flat structure of the R6G molecule introduces multiple degrees of freedom to the adsorption geometry, which can result in a variety of adsorption states characterized by different molecular orientations at the same surface site. The xanthene plane of the R6G+ is perpendicular to the phenyl plane in its gas phase geometry. We define the adsorption geometries of R6G+ on Ag surface as “parallel”, “flipped”, “twisted parallel”, “twisted flipped”, and “skewed”, as illustrated in Figure S6. In the “parallel” geometry, a R6G+ in its gas phase geometry has its xanthene plane in parallel with Ag surface with the ester group on the phenyl ring pointing away from the surface. The “flipped” geometry is then described by the same gas phase R6G+ with its two ethylamine groups on xanthene plane and the substituent group of the phenyl ring pointing toward the Ag surface. Next, twisting the phenyl ring of a R6G+in “parallel” or “flipped” geometry, in which the phenyl plane is no longer perpendicular to the xanthene plane, gives us “twisted parallel” or

“twisted flipped” geometry respectively. Lastly, the “skewed” 12

ACS Paragon Plus Environment

Page 12 of 30

Page 13 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

geometry is described as one ethylamine group and the ester group on the pendant phenyl ring “contacting” the Ag surface. These five configurations are used as starting geometries for terrace, vacancy-[111] and adatom-[111] sites. The stair-like structure on kink-[111] and step-[111] sites allows multiple adsorption geometries for either “flipped” or “parallel” configuration. For instance, the “flipped” R6G+ could make different binding by letting the two ethylamine groups of xanthene plane or the ester group of the phenyl ring rest on the stair-like structure. We consider here only “parallel” and “flipped” configurations at the kink and step sites (see Figure S13 and Figure S14). The twisted molecules at kink, step sites and the “twisted parallel” geometry at vacancy-[111], adatom-[111] sites are not considered due to their expected weak binding energies. Finally, a total of 39 different binding configurations are explored in this work. Self-consistent density functional theory (DFT) calculations using a plane-wave implementation are employed to obtain the stabilized adsorption states of R6G+ on the Ag surface. The PBE (Perdew-Burke-Ernzerhof) generalized gradient approximation38 is used for the exchange-correlation potential. The atomic cores are described by the pseudopotentials generated by projector augmented-wave (PAW) method39-40. The electronic wave functions are represented on a plane wave basis set with a cutoff energy of 400 eV. A single gamma k-point sampling is enough for the convergence of the selfconsistent electron density in all the employed large unit cells. Methfessel-Paxton smearing 41 of order 1 with a value of sigma of 0.001 eV and Pulay mixing42 are applied to aid convergence. The atomic positions of the three bottom Ag layers are fixed at the experimental lattice constant 4.09 Å43 and all the other nuclei are allowed to relax until the residual forces on all these ions are less than 0.025 eV/Å. Van der Waals (vdW) 13

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 30

correction is introduced to energies and forces at each optimization step with the postSCF Tkatchenko-Scheffler scheme44. This correction is essential to describe the interaction between the R6G+ molecule and the Ag surface. It gives reasonable binding energy and adsorption distances for physisorbed molecules45. The calculated interaction energy between R6G+ and Ag surface without the vdW correction is very small (~0.02 eV), which could not explain the experimentally observed strong binding of R6G+ (see Supporting Information). All DFT calculations were performed with the Vienna Ab initio Software Package (VASP).46-48 The charge state description of the R6G cation is accomplished in our electronic structure method by assigning a deficit electron to the system. In order to avoid electrostatic energy divergence for the charged periodic system, a neutralizing uniform countercharge density is added across the entire unit cell.49 This prevents us from calculating the adsorption energy from a direct energy difference between the combined and the separated systems. Therefore, a specific procedure is employed here to calculate the adsorption energy as  = '23 + 4(5 6758 + 96: 6758

(5)

'23 is the direct R6G-Ag interaction (mainly constituted by the image and vdW interaction), which is evaluated as the energy difference between the bound state and an “off” state obtained by shifting the molecule 20 Å away from the Ag surface. Bader analysis50 is applied upon each “off” state, and the charge state of a cation is always seen on the molecule with an accuracy of 0.001 elementary charge. The adsorbed R6G+ not only distorts itself but also modifies the Ag surface structure. 4(5 6758 , the relaxation energy of the molecule, describes the distortion of the adsorbate R6G+ from its gas phase 14

ACS Paragon Plus Environment

Page 15 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

geometry and

96: 6758 , the relaxation energy of the Ag surface, captures the

reconstruction of the Ag slab due to the adsorption. Therefore 4(5 6758 is calculated as the difference between the energies of the isolated R6G+ in its adsorbate geometry and in its gas phase geometry. The energies of both geometries are calculated in a cubic periodic box with a side length of 30 Å (see Supporting Information). Similarly, 96: 6758 is given by the energy difference between the Ag slab modified by the molecule and the Ag slab relaxed with 3 fixed bottom Ag layers; both energies are calculated in the same unit cell as that of the bound state. This specific procedure ensures a well-defined adsorption energy that excludes the additional effects from the neutralizing background charge.

3. Results and Discussion In this section, based on the direct link established in equation (4), we demonstrate one possible origin of the observed variation in R6G reduction potential by examining various binding states of R6G on different Ag surfaces. We first examine the binding motifs with the largest magnitude of adsorption energy on each surface site, i.e., the respective most stable configurations, as summarized in Figure 3. These configurations are the most probable binding motifs and are therefore the most likely to be observed experimentally. The adsorption distance, measured from the top Ag layer to the closest molecular atom along the normal direction on the flat terrace sites and the vacancy-[111] site, mostly ranges in between 1.99 Å and 2.05 Å except that terrace[100] site allows for a much shorter distance of 1.59 Å. When the Ag surface is mainly flat, as is the case in terrace-[111], terrace-[110], terrace-[100] and vacancy-[111] sites, R6G+ preferentially adsorbs in the “parallel 15

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

configuration”. This configuration enables the strongest binding motif of R6G+ on a flat surface, primarily due to the maximized interaction between the xanthene plane of R6G+ and the Ag surface. As shown in Figure 3, the adsorption energies of the most stable binding motifs on these flat surface sites (terrace-[111], terrace-[110], terrace-[100] and vacancy-[111]) are -2.18 eV, -2.11 eV, -1.93 eV and -2.08 eV respectively. With similar molecular relaxation energies (0.27 eV - 0.30 eV) and surface relaxation energies (0.01 eV – 0.05 eV) in these states, the molecule-surface interaction '23 is the dominant contribution that determines the ordering of the aforementioned adsorption energies, its values are -2.49 eV, -2.40 eV, -2.25 eV and -2.40 eV respectively. The close-packed [111] facet with surface atom density of 0.14 atoms/Å2 and inter-layer distance of 2.36 Å stands out with the largest '23 (-2.49 eV). The vacancy-[111] and terrace-[110] sites come after with the same '23 (-2.40 eV) for different reasons. With one less surface atom in the top layer, it is easily understood that the vacancy-[111] site comes with a smaller '23 than that at the terrace-[111] site. Nonetheless, the terrace-[110] site possesses the smallest surface atom packing density (0.08 atoms/Å2) among these flat surfaces, but a much smaller inter-layer distance of 1.45 Å compared to the other facets raises the '23 to the same value (-2.40 eV) as that at vacancy-[111] site. Lastly, terrace-[100] site with a surface atom density of 0.12 atoms/Å2 and inter-layer distance of 2.05 Å only allows a '23 of -2.25 eV, which makes it the least probable site to adsorb among all the surface sites.

16

ACS Paragon Plus Environment

Page 16 of 30

Page 17 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 3. Schematic illustrating the binding motifs which are the most stable binding states on all eight surface sites between -1.93 and -2.37 eV. The molecule-surface distance is measured along normal direction, given by the distance from the lowest atom in molecule to average positions of top layer of silver atoms along normal direction. The numbers below each configuration are  (red), '23 (blue), 4(5 6758 (black) and 96: 6758 (purple). Except for the “parallel” geometries discussed above, “twisted parallel” geometry also leads to a strong binding motif on terrace-[111] site whose adsorption energy is -2.06 eV. This binding motif comes with a similar '23 of -2.54 eV as that in the “parallel” configuration since the xanthene plane is still in parallel with the surface. But the binding 17

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

strength is reduced significantly by a 4(5 6758 of 0.45 eV resulting from the twisted phenyl ring. Exploration of the defect-rich surface sites reveals that the roughness of an Ag surface structure leads to a stronger binding than that on flat surfaces. First, the “parallel” geometry binds onto the terrace-[211] site with an adsorption energy of -2.37 eV, exhibiting the highest stability among all the binding motifs (see Figure 3). This is mainly because terrace-[211] sites offers much more roughness than the other terrace sites and exhibits a stepped surface structure, although it is constructed as a well-defined crystal facet. We find a minor twist on the phenyl ring in this “parallel” configuration at the terrace-[211] site, however, this binding motif shows similar 4(5 6758 (0.28 eV) and 4(5 96: (0.01 eV), and allows for a much larger molecule-surface interaction (-2.66 eV) compared to what is found in the other “parallel” geometries on flat surfaces. Second, another kind of “parallel configuration” with its xanthene plane leaning against the stairlike defect structure is identified as the most stable binding motif at the kink-[111] and step-[111] sites. The kink-[111] site allows for an adsorption energy of -2.32 eV with a '23 of 2.76 eV and relaxation energies of 0.41 eV (4(5 6758 ) and 0.03 eV (96: 6758 ). The step-[111] site ( = -2.28 eV) comes after with a lower '23 (-2.55 eV), less 4(5 6758 (0.24 eV) and the same 96: 6758 (0.03 eV). Lastly, the adatom-[111] site prefers the “flipped” configuration as the most probable R6G+ adsorption state. The adsorption of a flipped R6G+ onto flat surfaces tends to be weak, because the induced distortion is significant (4(5 6758 ~ 0.5 eV, see Figure 4) compared to other stable binding motifs such as “parallel” configurations on flat surfaces and “flipped” configurations on kink-[111] and step-[111] sites. However, the single additional Ag 18

ACS Paragon Plus Environment

Page 18 of 30

Page 19 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

atom at the adatom-[111] site causes -2.60 eV of '23 , making the “flipped” configuration the most stable binding motif on this site, characterized by an adsorption energy of -2.18 eV, similar with the most stable state at terrace-[111] (-2.18 eV).

Figure 4. Flipped R6G+ configurations on terrace-[111], terrace-[100], vacancy-[111], step[111] and kink-[111] sites are presented. The numbers below each configuration are  (red), '23 (blue), 4(5 6758 (black) and 96: 6758 (purple). It is observed here that the R6G distorts more on a flat surface than on step-[111] and kink-[111], since the stair-like structure on the latter conforms to the overall shape of the R6G molecule. It is noteworthy that the flipped R6G+ configuration strongly binds at step-[111] and kink-[111] sites. As illustrated in Figure 4, the stair-like structure at the step-[111] site is a similar structural fit with the flipped R6G molecule, where not only a large '23 (-2.54 eV) but also a small 4(5 6758 (0.24 eV) is observed. This is also observed for the kink[111] site, on which the adsorption energy of the “flipped” geometry is -2.26 eV, with 2.43 eV of '23 and 0.15 eV of 4(5 6758 . 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The eight R6G-Ag binding motifs with the highest stability at each explored Ag surface site (see Figure 3) can be associated with the Faradaic events observed in the ensemble surface CV, because they are the binding states most favored by the eight individual surface site studied. The adsorption energies of these binding motifs, ranging from -1.93 to -2.37 eV, constitute the “Faradaic region” energy window in Figure 5. Referring to equations (2-4) which link the adsorption energy of a R6G+ with its reduction potential, we do not attempt to match the exact theoretical values with the experimental values, but we instead compare the energy window ranges. In fact, the 0.44 eV adsorption energy range of the “Faradaic region” in Figure 5 does match closely to the 0.5 V reduction potential range of Faradaic events observed in the EC-SMSERS, up to the experimental accuracy (0.1 V) (see Figure 1C). This semi-quantitative agreement between the calculated adsorption energy range of the most stable binding motifs on all the surface sites and the observed potential range of Faradaic events supports our proposed hypothesis concerning the existence of various binding sites and the weak binding of the neutral R6G0on AgNP.

20

ACS Paragon Plus Environment

Page 20 of 30

Page 21 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 5. Eads #R6G+ on Ag% versus binding motifs. The 8 crosses on the right side of the blue dashed line represent the 8 most stable configurations for each unique surface type, which fall in a 0.44 eV window. This window corresponds to the potential range of experimental Faradaic events in Figure 1C. In the underpotential region, 11 meta-stable binding states within an adsorption energy range between -1.73 and -1.93 eV are the candidates for underpotential events.

Now that we have determined a class of stable adsorption states that corresponds to the experimental EC-SMSERS Faradaic events, we can start to investigate the nanoscopic origin of the experimentally observed underpotential events. The experimentally determined SMSERS signal loss potential histogram captures some underpotential reduction events within a potential range of approximately 0.2 V (see Figure 1) below the Faradaic events range. These underpotential events are considered here to be associated with the reduction of a variety of meta-stable binding motifs. These meta-stable states are the explored binding motifs that fall within a 0.2 eV adsorption energy range just next to 21

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the theoretical Faradaic region (see Figure 5). They are characterized by a weaker binding strength, compared to the most stable adsorption state at each respective surface site. Accordingly, they are much less probable and not observed in a bulk electrochemical experiment. In spite of being less probable, the adsorbed R6G+ may remain at a metastable state if the energy barrier for switching toward a more stable molecular orientation is significantly high compared to the thermal fluctuation (kBT). The above assumption may be satisfied if the adsorption state transition is achieved by a desorb-switch-adsorb process, because such a transition will have to overcome an energy barrier that amounts to the magnitude of the '23 which is more than 1 eV (see Figure S3 for the potential well of such an interaction for the [110]-(parallel) binding motif ).

We identified 11 R6G-Ag binding configurations, or “meta-stable” states, from the adsorption energy range between -1.73 and -1.93 eV, as depicted in the underpotential region of Figure 5 and illustrated in Figure 6. These meta-stable states can be classified into two group of binding motifs. The first group of candidates for underpotential events are 6 binding motifs on defect-free surface site, they are “flipped”, “twisted flipped” and “twisted parallel” geometries at the terrace-[110] and terrace-[100] sites respectively. These binding motifs are featured with relatively large '23 (between -2.23 and -2.41 eV) and significantly heavy molecular distortion (4(5 6758 between 0.36 eV and 0.52 eV) compared to the most stable binding motifs at defect-free sites. The defect-rich sites then provide another several candidates. They are three “flipped” geometries at step-[111], vacancy-[111] and kink-[111] sites, one tilted “parallel” configuration at the step-[111] site and a “parallel” geometry at the adatom-[111] site. 22

ACS Paragon Plus Environment

Page 22 of 30

Page 23 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 6. Schematics of all of the binding motif candidates for meta-stable states from the adsorption energy window between – 1.73 and -1.93 eV. The number below each configuration is ; The schematic illustration and the adsorption energies for all the explored binding geometries are summarized in Supporting Information. The binding states whose adsorption energies are even less than 1.73 eV in magnitude are not discussed here, because their binding strengths are too weak and possibly are not observed in the experiment. 4. Conclusions and Perspectives

23

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The ability to probe the electrochemical reaction at the single molecule level enables an excellent interplay between theory and experiment. In order to provide an explanation for (1) the broad reduction potential distribution and (2) the underpotential single electron transfer events observed experimentally with SMSERS, this study explored different binding configurations of R6G+ on various defect-free and defect-rich Ag surface sites, which can in turn affect the experimentally observed reduction potential of the molecule. The surface structure on the AgNP is modeled as a planar periodic surface and eight binding sites are considered. Using DFT, we demonstrate that multiple binding sites on an AgNP leads to the reduction potential distribution observed in the EC-SMSERS experiment. A semi-quantitative agreement was obtained between the theoretical adsorption energy window and the experimental SMSERS single loss potential distribution, further supporting our hypothesis. Moreover, we identified a few metastable adsorption states as candidates for the experimentally observed underpotential reduction events. These states are less stable and thus less probable to be observed, but the significant geometry reorganization required to reach a more stable adsorption state may make it possible to capture them experimentally, which provides a rationalization of the observed underpotential events. Significant molecular distortions observed in many of the studied R6G-Ag binding motifs can lead to distinct changes in the vibrational modes of the molecule. This change ought to be reflected in the Raman signature of the adsorbed molecule. The simulated Raman spectra of various R6G-Ag binding motifs may enable another excellent way of connecting theory to experiment, and can further provide means for visualizing how surface defects and binding states on AgNP affect the observed optical response. Our 24

ACS Paragon Plus Environment

Page 24 of 30

Page 25 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

theoretical approach can be extended to an array of complex molecules and surfaces, which is critical when considering other electrochemical systems relevant to electrocatalytic processes on metal nanoparticles, for example.

Furthermore, we

anticipate that both the kinetics and thermodynamics of the electrochemical reaction of R6G will exhibit site-dependent behavior. The kinetics dispersion of the electron transfer reaction at the electrode surface will be studied by applying ab initio DFT to the MarcusHush-Chidsey model51-54 which is a successful application of the Marcus theory on the kinetics of electrode reactions. ASSOCIATED CONTENT

Supporting Information. The determination of employed unit cell sizes; the determination of number of layers included in the slab; the importance of the vdW correction; Bader analysis of the charge on R6G; the schematic illustration of the starting geometries; the schematic illustration of all the explored binding motifs; the table for adsorption energies of all the binding motifs; proposed energetics of the underpotential events. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION

Corresponding Author *email: [email protected]

Author Contributions §

B.F. and C.V.D contributed equally to this work. All authors have given approval to the final version of the manuscript.

Notes The authors declare no competing financial interest.

25

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACKNOWLEDGEMENTS

This work was supported by Air Force Office of Scientific Research MURI (FA955014-1-0003). We gratefully acknowledge the computational resources from the Quest high performance computing facility at Northwestern University and the Extreme Science and Engineering Discovery Environment (XSEDE) program which is supported by National Science Foundation grant number ACI-1053575. S.Z. would like to acknowledge Michael Mattei for helpful discussions.

REFERENCES:

1. Fan, F. R.; Bard, A. J., Electrochemical Detection of Single Molecules. Science 1995, 267, 871-874. 2. Anderson, L. B.; Reilley, C. N., Thin-Layer Electrochemistry: Use of Twin Working Electrodes for the Study of Chemical Kinetics. Journal of Electroanalytical Chemistry (1959) 1965, 10, 538-552. 3. Anderson, L. B.; Reilley, C. N., Thin-Layer Electrochemistry: Steady-State Methods of Studying Rate Processes. Journal of Electroanalytical Chemistry (1959) 1965, 10, 295-305. 4. Lemay, S. G.; Kang, S.; Mathwig, K.; Singh, P. S., Single-Molecule Electrochemistry: Present Status and Outlook. Accounts of Chemical Research 2012, 46, 369-377. 5. Mathwig, K.; Aartsma, T. J.; Canters, G. W.; Lemay, S. G., Nanoscale Methods for SingleMolecule Electrochemistry. Annual review of analytical chemistry 2014, 7, 383-404. 6. Hill, C. M.; Clayton, D. A.; Pan, S., Combined Optical and Electrochemical Methods for Studying Electrochemistry at the Single Molecule and Single Particle Level: Recent Progress and Perspectives. Physical Chemistry Chemical Physics 2013, 15, 20797-20807. 7. Lei, C.; Hu, D.; Ackerman, E. J., Single-Molecule Fluorescence Spectroelectrochemistry of Cresyl Violet. Chem. Commun. 2008, 5490-5492. 8. Cortés, E.; Etchegoin, P. G.; Le Ru, E. C.; Fainstein, A.; Vela, M. E.; Salvarezza, R. C., Monitoring the Electrochemistry of Single Molecules by Surface-Enhanced Raman Spectroscopy. J. Am. Chem. Soc. 2010, 132, 18034-18037. 9. Cortes, E.; Etchegoin, P. G.; Le Ru, E. C.; Fainstein, A.; Vela, M. E.; Salvarezza, R. C., Strong Correlation between Molecular Configurations and Charge-Transfer Processes Probed at the Single-Molecule Level by Surface-Enhanced Raman Scattering. J. Am. Chem. Soc. 2013, 135, 2809-2815. 10. Norskov, J. K.; Bligaard, T.; Rossmeisl, J.; Christensen, C. H., Towards the Computational Design of Solid Catalysts. Nature Chemistry 2009, 1, 37-46.

26

ACS Paragon Plus Environment

Page 26 of 30

Page 27 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

11. Liu, P.; Rodriguez, J. A., Catalysts for Hydrogen Evolution from the [Nife] Hydrogenase to the Ni2p(001) Surface:  The Importance of Ensemble Effect. J. Am. Chem. Soc. 2005, 127, 1487114878. 12. Greeley, J.; Nørskov, J. K.; Kibler, L. A.; El-Aziz, A. M.; Kolb, D. M., Hydrogen Evolution over Bimetallic Systems: Understanding the Trends. ChemPhysChem 2006, 7, 1032-1035. 13. Hinnemann, B.; Moses, P. G.; Bonde, J.; Jørgensen, K. P.; Nielsen, J. H.; Horch, S.; Chorkendorff, I.; Nørskov, J. K., Biomimetic Hydrogen Evolution:  Mos2 Nanoparticles as Catalyst for Hydrogen Evolution. J. Am. Chem. Soc. 2005, 127, 5308-5309. 14. Bonde, J.; Moses, P. G.; Jaramillo, T. F.; Norskov, J. K.; Chorkendorff, I., Hydrogen Evolution on Nano-Particulate Transition Metal Sulfides. Faraday Discuss. 2009, 140, 219-231. 15. Datta, P.; Sardar, D.; Saha, R.; Mondal, T. K.; Sinha, C., Structure, Photophysics, Electrochemistry and Dft Calculations of [Ruh(Co)(Pph3)2(Coumarinyl-Azo-Imidazole)]. Polyhedron 2013, 53, 193-201. 16. Stadelman, B. S.; Kimani, M. M.; Bayse, C. A.; McMillen, C. D.; Brumaghim, J. L., Synthesis, Characterization, Dft Calculations, and Electrochemical Comparison of Novel Iron(Ii) Complexes with Thione and Selone Ligands. Dalton Transactions 2016, 45, 4697-4711. 17. Zaleski, S.; Cardinal, M. F.; Klingsporn, J. M.; Van Duyne, R. P., Observing Single, Heterogeneous, One-Electron Transfer Reactions. The Journal of Physical Chemistry C 2015, 119, 28226-28234. 18. Rowe, G. K.; Carter, M. T.; Richardson, J. N.; Murray, R. W., Consequences of Kinetic Dispersion on the Electrochemistry of an Adsorbed Redox-Active Monolayer. Langmuir 1995, 11, 1797-1806. 19. Hill, C. M.; Pan, S., A Dark-Field Scattering Spectroelectrochemical Technique for Tracking the Electrodeposition of Single Silver Nanoparticles. J. Am. Chem. Soc. 2013, 135, 17250-17253. 20. Liau, Y.-H.; Scherer, N. F.; Rhodes, K., Nanoscale Electrical Conductivity and Surface Spectroscopic Studies of Indium−Tin Oxide. The Journal of Physical Chemistry B 2001, 105, 32823288. 21. Hansen, K. H.; Worren, T.; Stempel, S.; Lægsgaard, E.; Bäumer, M.; Freund, H. J.; Besenbacher, F.; Stensgaard, I., Palladium Nanocrystals on Αl2o3 : Structure and Adhesion Energy. Physical Review Letters 1999, 83, 4120-4123. 22. Hansen, P. L.; Wagner, J. B.; Helveg, S.; Rostrup-Nielsen, J. R.; Clausen, B. S.; Topsøe, H., Atom-Resolved Imaging of Dynamic Shape Changes in Supported Copper Nanocrystals. Science 2002, 295, 2053-2055. 23. Hansen, T. W.; Wagner, J. B.; Hansen, P. L.; Dahl, S.; Topsøe, H.; Jacobsen, C. J. H., Atomic-Resolution in Situ Transmission Electron Microscopy of a Promoter of a Heterogeneous Catalyst. Science 2001, 294, 1508-1510. 24. Honkala, K.; Hellman, A.; Remediakis, I. N.; Logadottir, A.; Carlsson, A.; Dahl, S.; Christensen, C. H.; Nørskov, J. K., Ammonia Synthesis from First-Principles Calculations. Science 2005, 307, 555-558. 25. Nørskov, J. K., et al., Universality in Heterogeneous Catalysis. J. Catal. 2002, 209, 275278. 26. Hoshi, N.; Kato, M.; Hori, Y., Electrochemical Reduction of Co2 on Single Crystal Electrodes of Silver Ag(111), Ag(100) and Ag(110). J. Electroanal. Chem. 1997, 440, 283-286. 27. Eberhardt, D.; Santos, E.; Schmickler, W., Hydrogen Evolution on Silver Single Crystal Electrodes—First Results1. J. Electroanal. Chem. 1999, 461, 76-79.

27

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

28. Shi, C.; Hansen, H. A.; Lausche, A. C.; Norskov, J. K., Trends in Electrochemical Co2 Reduction Activity for Open and Close-Packed Metal Surfaces. Physical Chemistry Chemical Physics 2014, 16, 4720-4727. 29. Pozun, Z. D.; Tran, K.; Shi, A.; Smith, R. H.; Henkelman, G., Why Silver Nanoparticles Are Effective for Olefin/Paraffin Separations. The Journal of Physical Chemistry C 2011, 115, 18111818. 30. Rocca, M., et al., Phase Transition of Dissociatively Adsorbed Oxygen on Ag(001). Physical Review B 2000, 61, 213-227. 31. de Mongeot, F. B.; Cupolillo, A.; Valbusa, U.; Rocca, M., O2 Dissociation on Ag(001): The Role of Kink Sites. Chemical Physics Letters 1997, 270, 345-350. 32. Kokalj, A.; Dal Corso, A.; de Gironcoli, S.; Baroni, S., The Interaction of Ethylene with Perfect and Defective Ag(001) Surfaces. The Journal of Physical Chemistry B 2002, 106, 98399846. 33. Klingsporn, J. M.; Jiang, N.; Pozzi, E. A.; Sonntag, M. D.; Chulhai, D.; Seideman, T.; Jensen, L.; Hersam, M. C.; Duyne, R. P. V., Intramolecular Insight into Adsorbate–Substrate Interactions Via Low-Temperature, Ultrahigh-Vacuum Tip-Enhanced Raman Spectroscopy. J. Am. Chem. Soc. 2014, 136, 3881-3887. 34. Plieth, W. J., Electrochemical Properties of Small Clusters of Metal Atoms and Their Role in the Surface Enhanced Raman Scattering. The Journal of Physical Chemistry 1982, 86, 31663170. 35. Ivanova, O. S.; Zamborini, F. P., Size-Dependent Electrochemical Oxidation of Silver Nanoparticles. J. Am. Chem. Soc. 2010, 132, 70-72. 36. Greeley, J.; Nørskov, J. K.; Mavrikakis, M., Electronic Structure and Catalysis on Metal Surfaces. Annual Review of Physical Chemistry 2002, 53, 319-348. 37. Mavrikakis, M.; Stoltze, P.; Nørskov, J. K., Making Gold Less Noble. Catal. Lett., 64, 101106. 38. Perdew, J. P.; Burke, K.; Ernzerhof, M., Generalized Gradient Approximation Made Simple. Physical Review Letters 1996, 77, 3865-3868. 39. Blöchl, P. E., Projector Augmented-Wave Method. Physical Review B 1994, 50, 1795317979. 40. Kresse, G.; Joubert, D., From Ultrasoft Pseudopotentials to the Projector AugmentedWave Method. Physical Review B 1999, 59, 1758-1775. 41. Methfessel, M.; Paxton, A. T., High-Precision Sampling for Brillouin-Zone Integration in Metals. Physical Review B 1989, 40, 3616-3621. 42. Pulay, P., Convergence Acceleration of Iterative Sequences. The Case of Scf Iteration. Chemical Physics Letters 1980, 73, 393-398. 43. Ashcroft, N. W.; Mermin, D. N., Solid State Physics; Holt, Rinehart and Winston, 1976. 44. Tkatchenko, A.; Scheffler, M., Accurate Molecular Van Der Waals Interactions from Ground-State Electron Density and Free-Atom Reference Data. Physical Review Letters 2009, 102, 073005. 45. Lüder, J.; Sanyal, B.; Eriksson, O.; Puglia, C.; Brena, B., Comparison of Van Der Waals Corrected and Sparse-Matter Density Functionals for the Metal-Free Phthalocyanine/Gold Interface. Physical Review B 2014, 89, 045416. 46. Kresse, G.; Furthmüller, J., Efficiency of Ab-Initio Total Energy Calculations for Metals and Semiconductors Using a Plane-Wave Basis Set. Computational Materials Science 1996, 6, 1550.

28

ACS Paragon Plus Environment

Page 28 of 30

Page 29 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

47. Kresse, G.; Furthmüller, J., Efficient Iterative Schemes for Ab Initio Total-Energy Calculations Using a Plane-Wave Basis Set. Physical Review B 1996, 54, 11169-11186. 48. Kresse, G.; Hafner, J., Ab Initio Molecular-Dynamics Simulation of the Liquid-MetalAmorphous-Semiconductor Transition in Germanium. Physical Review B 1994, 49, 14251-14269. 49. Makov, G.; Payne, M. C., Periodic Boundary Conditions in Ab Initio Calculations. Physical Review B 1995, 51, 4014-4022. 50. Yu, M.; Trinkle, D. R., Accurate and Efficient Algorithm for Bader Charge Integration. The Journal of Chemical Physics 2011, 134, 064111. 51. Chidsey, C. E. D., Free Energy and Temperature Dependence of Electron Transfer at the Metal-Electrolyte Interface. Science 1991, 251, 919. 52. Feldberg, S. W., Implications of Marcus−Hush Theory for Steady-State Heterogeneous Electron Transfer at an Inlaid Disk Electrode. Analytical Chemistry 2010, 82, 5176-5183. 53. Costentin, C.; Robert, M.; Saveant, J.-M., Reorganization Energies and Pre-Exponential Factors in the One-Electron Electrochemical and Homogeneous Oxidation of Phenols Coupled with an Intramolecular Amine-Driven Proton Transfer. Physical Chemistry Chemical Physics 2010, 12, 13061-13069. 54. Laborda, E.; Henstridge, M. C.; Batchelor-McAuley, C.; Compton, R. G., Asymmetric Marcus-Hush Theory for Voltammetry. Chem. Soc. Rev. 2013, 42, 4894-4905.

29

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Table of Contents Artwork

30

ACS Paragon Plus Environment

Page 30 of 30