Formation and Diffusion of Subsurface Adsorbates at Electrodes

ABSTRACT: We report direct observation of the formation of a subsurface species at metal electrodes in liquid electrolytes and of its migration within...
1 downloads 0 Views 2MB Size
Subscriber access provided by Kaohsiung Medical University

Communication

Formation and Diffusion of Subsurface Adsorbates at Electrodes Björn Rahn, and Olaf M. Magnussen J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b04903 • Publication Date (Web): 07 Jul 2018 Downloaded from http://pubs.acs.org on July 7, 2018

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 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 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.

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 8 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

Journal of the American Chemical Society

Formation and Diffusion of Subsurface Adsorbates at Electrodes Björn Rahn and Olaf M. Magnussen* Institute of Experimental and Applied Physics, Kiel University, Olshausenstr. 40, 24098 Kiel, Germany Supporting Information Placeholder ABSTRACT: We report direct observation of the formation of a subsurface species at metal electrodes in liquid electrolytes and of its migration within the solid’s surface layer, below a chemisorbed electrochemical double layer. Using in situ video-rate scanning tunneling microscopy, we find for adsorbed sulfide on bromide-covered Ag(100) electrodes reversible transitions between adsorption sites on top of the surface and within a vacancy in the first Ag layer. In the latter state, the sulfide surface diffusion can be enhanced by orders of magnitude, which we attribute to vacancy-mediated diffusion underneath the bromide adlayer. The high dynamics within the surface layer, indicated by these observations, may open up alternative pathways in electrocatalytic reactions and growth processes.

The mechanisms of reactions at electrochemical interfaces are traditionally discussed by considering pathways that involve adsorbed species on the electrode surface on in the adjacent electrolyte. However, there is mounting evidence that not only adsorbates on top of the surface but also subsurface species can play a vital role in electrochemical reactions, especially in the context of electrocatalysis for energy storage and conversion. Prime examples for this are the dehydrogenation of formic acid on Bdoped Pd 1, 2 and the electrochemical reduction of carbon dioxide on Cu electrodes.3, 4 For the latter system, recent studies suggest that subsurface oxygen near the Cu surface plays an important role in this process and substantially enhances the reactivity and selectivity towards multicarbon products.5, 6 The presence of subsurface O here was concluded on the basis of spectroscopic data revealing Cu+ species. Direct microscopic observations of subsurface species at this – or in fact any – electrochemical interface are still lacking, however. Many electrochemical phase formation processes involve insertion of a (typically adsorbed) electrolyte species into the electrode bulk, but the precise atomic-scale mechanisms, especially in the initial stages, are often still under debate. For some well-known cases, such as the surface oxidation of Pt(111) in non-specifically adsorbing electrolyte, recent studies by surface-sensitive in situ methods and ab initio theory have helped to unravel the complex potential- and time-dependent place exchange of oxygen with Pt surface atoms (see e.g. ref. 7 and references therein). For most other metals, including Cu and Ag, such detailed data are still lacking and many open questions regarding subsurface species exist. These include the reversibility of place exchange between surface and subsurface sites, the potential range, in which subsurface species can exist, the precise exchange mechanisms, and the influence of the electrochemical environment, specifically chemisorbed anions. In the present work we present in situ high-speed scanning tunneling microscopy (Video-STM) data which show the formation and surface migration of subsurface sulfur adsorbates on bromide-

covered Ag(100) electrodes. This system bears many similarities to the aforementioned case of subsurface oxygen at Cu electrodes: Oxygen and sulfur both exhibit comparable behavior on coinage metal surfaces, involving complex surface reconstructions and an enhancement of surface transport, driven by a tendency to form surface complexes.8-10 In situ structural studies show that sulfur11 and bromide12, 13 adsorb in the fourfold-hollow sites of Ag(100), covering the electrode surface with a close-packed c(2×2) adlayer at sufficiently positive potentials where saturation coverage has been reached. This anionic layer constitutes the inner part of the electrochemical double layer and is a highly suitable model system, because of its well-ordered and particularly simple interface structure. In previous studies, we demonstrated that small atomic and molecular adsorbates can replace individual halides of the c(2×2) halide adlayers on Cu(100) and Ag(100) electrodes and studied the surface diffusion in these systems.14-16 Here we employed the same strategy for studies of sulfur adsorbates on the c(2×2)-Br covered Ag(100) electrode surface. In situ Video-STM sequences (Figure 1), obtained at sulfur coverages of 1 to 2 % of a monolayer (ML), confirm the presence of an Sad species (appearing as higher protrusions) that resides in sites of the c(2×2)-Br lattice, i.e., in the fourfold-hollow sites of the Ag surface (the apparent deviations from a square lattice are image distortions resulting from the high scan rate). The surface mobility of these species is low. According to quantitative surface diffusion studies (to be published), the hopping rate of Sad to neighboring c(2×2) sites is ≤ 1 s-1 at potentials positive of -0.10 V. However, occasionally these Sad species undergo spontaneous changes that we assign to a transition of the sulfur adsorbate into a subsurface site. A characteristic example of this behavior can be seen in Figure 1 (marked by white arrows): Between 100 and 200 ms, the marked Sad completely disappears. Instead, close to its previous position, two neighboring adsorbates of the c(2×2) lattice appear slightly darker (i.e., lower) than the surrounding adlattice. In the subsequently recorded images this feature moves a distance of ≈23 Å (300 ms) and then is converted back to an (immobile) Sad (400 ms). Such exchange events are not always accompanied by large displacements. For example, in the same video sequence an event is observed (marked by black arrows) where the Sad is found in a site next to its original position after reappearing on the surface. Initial and final lattice position in this case are identical to the two darker sites, marking the intermediate state (400 ms).

ACS Paragon Plus Environment

Journal of the American Chemical Society 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

Figure 2. (a,b) In situ Video-STM sequence in 1 mM KBr at -0.13 VSCE, showing (a) the motion of subsurface sulfur in the Ag(100) electrode surface and (b) incorporation of sulfur into the Ag surface layer at steps. (c) Schematic model of the Sss adsorption geometry. Two mechanisms seem feasible for the formation of Sss on the Ag(100) surface: Either the Sad steps down into an existing vacancy or it undergoes place exchange with an Ag atom in the surface layer, resulting in the creation of an Ag adatom. The vacancies or adatoms involved in this process are apparently too mobile to be directly observable in the images. On clean Ag(100) and Cu(100) the diffusion barriers of these species are in the range 0.37-0.59 eV, resulting in hopping frequencies of 108-105 s-1 at room temperature.17, 18 They thus will change their position much faster than the time between successive scan lines in our STM measurements (0.48 ms), provided that the halide adlayer does not reduce the metals surface mobility by several orders of magnitude. Support for the latter was found in Video-STM studies on c(2×2)-Cl covered Cu(100), where Cu adatoms could only be observed, when they were temporarily trapped by specific dimer configurations.19 In addition, a second mechanism of Sss formation was found, occurring at the steps of the Ag substrate, which are typically oriented along the [100] directions. In the presence of Sad, strong local fluctuation are observed along the step edge, manifesting as transiently attached species and kinks, which in some cases separate from the step and become fully formed Sss (Figure 2b). We assign this to an adsorption of Sad at step sites, followed by direct incorporation into the neighboring Ag lattice of the upper terrace. Sss formation was found in the entire studied potential range, down to -0.27 V. To understand its origin compare these observations with results for sulfur on clean Ag(100). The latter has been studied in detail by Thiel and coworkers, using STM under ultrahigh vacuum (UHV) conditions and density functional theory (DFT).9, 10, 20 These studies indicate that at low coverage sulfur energetically prefers adsorption in the fourfold-hollow sites. Occupation of higher-coordinated sites along steps has to be forced by high Sad coverages. Evidence for Sss formation in the low coverage regime was not reported. However, for coverages > 0.25 ML the formation energy for surface vacancies can be lowered, resulting ultimately in the (√17 × √17)R14° reconstructed phase, in which half of the sulfur atoms occupy sites within Ag vacancies. Our findings suggest that the coadsorbed c(2×2)-Br adlayer can cause a similar effect. DFT calculations for Ag(111)

Figure 1. Sequence of in situ Video-STM images obtained on Ag(100) electrodes in 1 mM HClO4 + 1 mM KBr at -0.12 VSCE. Sulfur adsorbates on the surface (bright protrusions) occupy sites of the c(2×2)-Br adlattice and are almost immobile on the subsecond timescale. Two place exchange events of sulfur to subsurface positions and back to sites on the surface are marked by arrows. This characteristic surface feature can adopt both of the symmetrically equivalent positions on the c(2×2) lattice, which are rotated by 90° to each other, demonstrating that its shape is not an imaging artifact caused by the shape of the STM tip. In some cases, changes between the two orientations can even be directly seen in subsequent images of a video sequence (Figure 2a). These observations can only be rationalized by a sulfur species that is located precisely between two neighboring c(2×2) lattice sites, i.e., at the position of an underlying Ag surface atom. We thus attribute this state to a subsurface sulfur species (Sss), which occupies a monovacancy in the Ag(100) surface layer (schematically illustrated in Figure 2c). The two c(2×2) lattice sites on top of the Sss still exhibit visible maxima and are most likely occupied by Brad. Other adsorption geometries which would be compatible with the observed symmetry, e.g. a bridge-bonded Sad on top of the two neighboring Brad, seem highly unlikely. Taking into account that sulfur is a strongly electronegative adsorbate species, the subsurface location could also explain why the two neighboring c(2×2)-Br maxima have a lower apparent height in the STM images. Furthermore, motion of Sss to one of the 4 neighboring sites of the Ag surface lattice would result in a 90° rotation of the observed feature, as seen in Figure 2a.

2

ACS Paragon Plus Environment

Page 2 of 8

Page 3 of 8 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

Journal of the American Chemical Society indeed show that the surface vacancy formation energy can be substantially lowered by electronegative adsorbates.21 Interestingly, indications that bromide coadsorbates may induce exchange of Sad with metal surface atoms were also found in our recent study of Sad diffusion on c(2×2)-Br covered Cu(100) electrodes.16 Although here place exchange was not directly observed and Sad diffusion occurred exclusively via jumps between neighbor c(2×2) sites, an unusual increase in Sad mobility towards more positive potentials was found. This potential-dependence is opposite to that observed on Cl-covered Cu(100), which can only be explained by a different Sad diffusion mechanism in the presence of Brad. On the basis of detailed DFT calculations it was tentatively suggested that in the presence of Brad the diffusion of Sad on top of the surface is less favorable than an exchange diffusion mechnism, where the sulfur atom transiently resides in a subsurface site. In this case, Br coadsorbates would on both metals promote exchange, but with a different effect on the sulfur transport: On Ag(100) it opens up a second transport channel, whereas on Cu(100) it completely switches the transport mechanism. According to our STM observations, the subsurface sulfur species can be surprisingly mobile, despite the presence of the closepacked c(2×2)-Br adlayer on top. However, their motion is more complex than that of the Sad observed in parallel on this surface. The latter’s slow diffusion can be perfectly described by a simple random walk with jumps between neighboring sites, as evidenced by Poisson-like jump distributions. This is not the case for Sss diffusion: While some of the Sss remain at or close to their original location, others perform long jumps of up to 50 Å in just 0.2 s. The latter behavior closely resembles the vacancy-mediated diffusion mechanism observed by UHV-STM for various metal atoms embedded in Cu(100) and Cu(111) surfaces.22-26 Here, multiple encounters with fast moving vacancies in the metal surface layer lead to a sequence of rapid displacement events (“slide puzzle motion”), manifesting as a sudden long-jump in the STM experiments. Because this vacancy-mediated diffusion primarily depends on the presence of highly mobile vacancies, it should also be possible for Sss at electrochemical interfaces. The presence of the halide adlayer apparently does not inhibit vacancy mobility, rather it may even lower the formation energy and thus the density of surface vacancies (see above). Within this scenario two transport mechanisms for sulfur on Brcovered Ag(100) coexist: Predominantly, Sad move via conventional hopping diffusion on the surface, but after transition into the subsurface state vacancy-mediated diffusion is possible. The latter is much faster and therefore can compete with the Sad diffusion on top of the surface. Specifically, sulfur adsorbates undergo exchange approximately 1-4 times per minute and travel on average 18 Å each time they are in the subsurface state. Taking into account that for the surface diffusion of Sad within the c(2×2) lattice the average diffusion length l = 2 ⋅ d c (2 x 2) ⋅ ν ⋅ ∆t = 45 Å

vance for the wide range of reactions that involve oxygen species, e.g. oxidation of adsorbed species, abstraction of oxgen, and oxidation of metal surfaces.

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website. Experimental procedures (PDF) Video of the experiment partly shown in Figure 1 (mpeg)

AUTHOR INFORMATION Corresponding Author * [email protected]

Notes The authors declare no competing financial interests.

ACKNOWLEDGMENT We gratefully acknowledge financial support by the Deutsche Forschungsgemeinschaft through grant MA1618/15.

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

-1

for ∆t = 60 s and ν = 1 s , the transport rates via the two diffusion channels seem comparable. Similar as for desorption-mediated surface diffusion,27, 28 where weakly-bound adsorbates perform brief flights through the adjacent solution phase, the subsurface pathway for interface transport may allow species to bypass 2D barriers, e.g., close-packed coadsorbate adlayers. This may be relevant for understanding the surface dynamics of complex interface reactions, involving multiple surface species. Our observations clearly demonstrate that the findings for Cu and Ag under UHV conditions, i.e., the highly dynamic state of the atomic surface layer at room temperature,17, 18, 22-26 are also valid for electrochemical interfaces, despite or perhaps precisely because of the presence of chemisorbed ions. A particularly interesting open question is whether similar behavior may occur for other chalcogens, especially oxygen. The latter would be of rele-

14. 15. 16. 17. 18. 19. 20. 21. 22. 23.

3

K. Jiang, J. Chang, H. Wang, S. Brimaud, W. Xing, R. J. Behm and W.-B. Cai, ACS Appl. Mat. Interf., 2016, 8, 7133-7138. J. S. Yoo, Z.-J. Zhao, J. K. Nørskov and F. Studt, ACS Catal., 2015, 5, 6579-6586. Y. Hori, Mod. Aspects Electrochem., 2008, 42, 89-189. R. Kortlever, J. Shen, K. J. P. Schouten, F. Calle-Vallejo and M. T. M. Koper, J. Phys. Chem. Lett., 2015, 6, 4073-4082. A. Eilert, F. Cavalca, F. S. Roberts, J. Osterwalder, C. Liu, M. Favaro, E. J. Crumlin, H. Ogasawara, D. Friebel, L. G. M. Pettersson and A. Nilsson, J. Phys. Chem. Lett., 2017, 8, 285-290. D. Gao, I. Zegkinoglou, N. J. Divins, F. Scholten, I. Sinev, P. Grosse and B. Roldan Cuenya, ACS Nano, 2017, 11, 4825-4831. J. Drnec, D. A. Harrington and O. M. Magnussen, Curr. Op. Electrochem., 2017, 4, 69-75. P. A. Thiel, M. Shen, D.-J. Liu and J. W. Evans, J. Vac. Sci. Technol., A, 2010, 28, 1285-1298. D.-J. Liu and P. A. Thiel, The Journal of Chemical Physics, 2018, 148, 124706. S. M. Russell, M. Shen, D.-J. Liu and P. A. Thiel, Surf. Sci, 2011, 605, 520-527. E. Lastraioli, F. Loglio, M. Cavallini, F. C. Simeone, M. Innocenti, F. Carla and M. L. Foresti, Langmuir, 2010, 26, 17679-17685. B. M. Ocko, J. X. Wang and T. Wandlowski, Phys. Rev. Lett, 1997, 79, 1511-1514. T. Wandlowski, J. X. Wang and B. M. Ocko, J. Electroanal. Chem., 2001, 500, 418-434. T. Tansel and O. M. Magnussen, Phys. Rev. Lett., 2006, 96, 026101-1 - 026101-4. Y.-C. Yang and O. M. Magnussen, Phys. Chem. Chem. Phys, 2013, 15, 12480-12487. B. Rahn, R. Wen, L. Deuchler, J. Stremme, A. Franke, E. Pehlke and O. M. Magnussen, Angew. Chem. , Int. Ed, 2018, 57, 6065-6068. U. Kurpick and T. S. Rahman, Surf. Sci., 1999, 427-428, 15-21. D.-J. Liu, Phys. Rev. B, 2010, 81, 035415. Y.-C. Yang, A. Taranovskyy and O. M. Magnussen, Angew. Chem., Int. Ed, 2012, 51, 1966-1969. M. Shen, S. M. Russell, D.-J. Liu and P. A. Thiel, J. Chem. Phys., 2011, 135, 154701-1 - 154701-9. T. E. Jones, T. C. R. Rocha, A. Knop-Gericke, C. Stampfl, R. Schloegl and S. Piccinin, Phys. Chem. Chem. Phys., 2014, 16, 90029014. T. Flores, S. Junghans and M. Wuttig, Surf. Sci., 1997, 371, 1-13. R. van Gastel, E. Somfai, S. B. van Albada, W. van Saarloos and J. W. M. Frenken, Phys. Rev. Lett, 2001, 86, 1562-1565.

ACS Paragon Plus Environment

Journal of the American Chemical Society 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

24. R. van Gastel, E. Somfai, S. B. van Albada, W. van Saarloos and J. W. M. Frenken, Surf. Sci, 2002, 521, 10-25. 25. M. L. Anderson, M. J. D'Amato, P. J. Feibelman and B. S. Swartzentruber, Phys. Rev. Lett., 2003, 90, 126102-1 - 126102-4. 26. R. van Gastel, R. Van Moere, H. J. W. Zandvliet and B. Poelsema, Surf. Sci., 2011, 605, 1956-1961. 27. R. Walder, N. Nelson and D. K. Schwartz, Phys. Rev. Lett., 2011, 107, 156102-1 - 156102-5. 28. O. V. Bychuk and B. O'Shaughnessy, Phys. Rev. Lett, 1995, 74, 17951798.

4

ACS Paragon Plus Environment

Page 4 of 8

Page 5 of 8 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

Journal of the American Chemical Society SYNOPSIS TOC

5 ACS Paragon Plus Environment

Journal of the American Chemical Society 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

Figure 1 78x117mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 6 of 8

Page 7 of 8 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

Journal of the American Chemical Society

Figure 2 78x71mm (300 x 300 DPI)

ACS Paragon Plus Environment

Journal of the American Chemical Society 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 Content figure

ACS Paragon Plus Environment

Page 8 of 8