Cu(110) Surface in Hydrochloric Acid Solution: Potential Dependent

Subscriber access provided by UZH Hauptbibliothek / Zentralbibliothek Zuerich

Article

The Cu(110) Surface in Hydrochloric Acid Solution: Potential Dependent Chloride adsorption and Surface Restructuring Claudio Goletti J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/jp5073445 • Publication Date (Web): 22 Dec 2014 Downloaded from http://pubs.acs.org on December 28, 2014

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.

The Journal of Physical Chemistry

This document is confidential and is proprietary to the American Chemical Society and its authors. Do not copy or disclose without written permission. If you have received this item in error, notify the sender and delete all copies.

The Cu(110) Surface in Hydrochloric Acid Solution: Potential Dependent Chloride adsorption and Surface Restructuring

Journal: Manuscript ID: Manuscript Type: Date Submitted by the Author: Complete List of Authors:

The Journal of Physical Chemistry jp-2014-073445.R2 Article 22-Dec-2014 Goletti, Claudio; Università di Roma, dipartimento di Fisica Bussetti, Gianlorenzo; Politecnico di Milano, Department of Physics Violante, Adriano; Paul-Drude-Institut fur Festkorperelektronik, Bonanni, Beatrice; Universita' di Roma Tor Vergata, Dipartimento di Fisica Di Giovannantonio, Marco; Istituto di Struttura della Materia - CNR, Serrano, Giulia; Università di Roma, dipartimento di Fisica Breuer, Stephan; Institut für Physikalische und Theoretische Chemie, Gentz, Knud; Univeristy of Bonn, Deoartment of Physical and Theoretical Chemistry Wandelt, Klaus; Institut für Physikalische und Theoretische Chemie,

ACS Paragon Plus Environment

Page 1 of 36

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 Cu(110) Surface in Hydrochloric Acid Solution: Potential Dependent Chloride Adsorption and Surface Restructuring

C. Goletti1,*, G. Bussetti1,#, A. Violante1,@, B. Bonanni1, M. Di Giovannantonio1, G. Serrano1, S. Breuer1,2, K. Gentz2 and K. Wandelt2 1

Dipartimento di Fisica and CNISM, Università di Roma Tor Vergata, Via della ricerca scientifica 1, 00133 Roma, Italy 2

Institut für Physikalische und Theoretische Chemie, Wegeler Str. 12, 53115 Bonn, Germany Abstract

Using Cyclic Voltammetry (CV), Reflectance Anisotropy Spectroscopy (RAS) and in situ Electrochemical Scanning Tunneling Microscopy (EC-STM) we have studied the structure and structural transitions at a Cu(110) electrode surface in 10 m M HCl solution as a function of the applied electrode potential. While at potentials lower than -550

mV vs. Ag/AgCl the in situ STM images reveal the adsorbate-free,

unreconstructed structure of the Cu(110) surface, at increasing potential ≥ -550 mV chloride adsorption, as indicated by CV, leads first to the formation of grooves followed by the growth of added stripes. Both are aligned in the [001] direction as shown by EC-STM - and supported by ex situ Low Energy Electron Diffraction (LEED) - and are the result of a severe but fully reversible restructuring of the surface. This faceting is accompanied by an optical anisotropy peak in RAS centered at about 500 nm (2.48 eV), having a maximum when the linearly polarized light is aligned along the [1-10] direction, i.e. perpendicular to the stripes detected with in situ STM under the same conditions. By cycling the electrode potential through a full cyclic voltammogram and monitoring simultaneously the RAS signal we obtain a hysteresis-like curve which 1



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

supports a two-step kinetics of the restructuring process in agreement with the CV- and STM-data. The investigations demonstrate the power of combined RAS and in situ STM measurements to shine light on potential-driven processes at metal-electrolyte interfaces.

Keywords: Metal/electrolyte interface; chloride adsorption; surface restructuring; in situ Electrochemical Scanning Tunneling microscopy; in situ Reflectance Anisotropy Spectroscopy.

*

Corresponding author

#

Present address: Dipartimento di Fisica, Politecnico di Milano, Piazza Leonardo da Vinci 32, I-20133 Milano (Italy)

@

Present address: Paul-Drude-Institut für Festkörperelektronik, Hausvogteiplatz 5-7 D-10117, Berlin (Germany)

2


Page 2 of 36

Page 3 of 36

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


Introduction. Important technologies are based on processes at solid/liquid interfaces1 like electro-catalysis2 , grafting3 and biofunctionalization of surfaces4 , as well as electro-deposition5 and electro–etching6 of metals down to the nanometer scale. In biosensors the signal is produced by the interaction of the analyte with the sensitive biological layer deposited in liquid onto a solid surface7. The on-chip circuitry in modern electronic devices is nowadays created by a sequence of copper deposition and etching processes, the so-called Damascene process8 , reaching lateral dimensions of about 30 nm. In order to improve and optimize the activity of an electro-catalyst or the sensitivity of a biosensor, or to push electroplating/electro-etching to even smaller dimensions, a deeper mechanistic understanding of these processes on the atomic scale is required. Consequently, there is a growing interest and motivation to study solid/liquid interfaces with the same precision that has been achieved with the so-called “surface science approach” under ultrahigh vacuum conditions. Investigations of the properties and processes at solid/liquid interfaces however pose a new methodological challenge. Unlike in UHV, electron-, ion-, and atom-beam based methods are not applicable or at least of very limited use, because the particles do not penetrate either of the two phases. Therefore, only a much more limited set of experimental methods is available to study a solid surface immersed in a liquid. These are proximity probe-based or photon-based techniques, like Scanning Tunneling Microscopy (STM) 9 , Atomic Force Microscopy (AFM)10 , X-ray and Infrared Absorption Spectroscopy (XAS, IRRAS)11 , Electroreflectance (ER)12 , Ellipsometry13 , Sum-Frequency Spectroscopy (SFS)

14

and – more recently –

Reflectance Anisotropy Spectroscopy (RAS) 15,16. While each of these techniques may provide some specific information about a given interface, we are still far away from the detailed microscopic and spectroscopy information that is obtainable under UHV conditions. Therefore it is sometimes unavoidable, in order to complement a data set obtained in situ, i.e. from the solid surface in contact with the liquid phase, that the sample is emersed from the liquid and transferred into a UHV apparatus in order to take advantage of the plethora of UHV-based methods. The transfer can be done either through air (by applying some suitable safety measures, e.g. by capping the sample), or air-free (by using a dedicated “transfer-system”, in which the sample is transferred through one or more buffer3



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

chambers, which can be valved off and selectively pumped down, into the final UHV analysis chamber17). This strategy, of course, always requires an independent verification whether or not -after the transfer into UHV- the status of the surface is still representative for the one before emersion from the liquid. This result can be achieved by comparing the results obtained with one and the same technique before emersion and after transfer into UHV. Yet, another approach has been to mimic the solid/liquid interface by adsorbing layers of suitably chosen atomic or molecular species together with solvent molecules under UHV conditions18. Undoubtedly, the most realistic and trustworthy approach is to study a solid/liquid interface directly in situ, and if possible with several complementary techniques under identical conditions. To this end, in this work we demonstrate the combined use of RAS and in situ STM for the investigation of a Cu(110) surface in contact with an electrolyte. Due to the technological relevance of copper in electrochemical (EC) processes, as mentioned above, many experimental (both in UHV and in electrochemical environment) and theoretical investigations have been focussed on the low index surfaces of copper, mainly on Cu(100) and Cu(111) and less on Cu(110), in particular to explore the interaction with halogens relevant in corrosion and etching processes19 . The (110) plane has been less extensively studied, owing to its more open geometry it is more reactive than the Cu(100) and Cu(111) surface, respectively. The first important step towards an understanding of the interaction between a Cu(110) surface and chloride ions in hydrochloric acid solution has been made with Electrochemical STM (EC-STM) by Wan et al. 20 and Li et al. 21 , who showed massive reconstruction of the surface. Later there was also an increased interest to explore any changes in the electronic properties of the surface during the interaction with ions in solution. In particular, based on RAS measurements Barritt et al.

22

discussed the existence and modification of a

characteristic electronic surface state of the Cu(110) surface in hydrochloric acid solution. This surface state was well known from measurements in UHV23 . The reactivity of copper surfaces towards chlorine and the concomitant surface restructuring has also been studied extensively in vacuum including STM measurements24,25 . As a consequence of all these separate STM and RAS measurements the picture emerged, that no substantial differences seem to exist between a Cu(110) sample in contact with aqueous HCl solution or being exposed to chlorine under vacuum conditions. 4


Page 4 of 36

Page 5 of 36

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 this paper, we present the combined application of RAS and in situ EC-STM in conjunction with Cyclic Voltammetry (CV) to study the Cu(110) surface in contact with a 10 mM HCl solution in one and the same EC cell. Our EC-STM images of the immersed Cu(110) surface show that – at particular values of the electrode potential – trenches are forming on the surface running along the [001] direction, the result of a beginning faceting process. Upon increasing the electrode potential, further adsorption of chloride ions leads to the evolution of elevated stripes growing also in [001] direction. Under these conditions RAS detects an optical anisotropy peak centred around a wavelength of 500 nm (2.48 eV), having its maximum when the polarized light is aligned perpendicular to the stripes imaged by EC-STM. Finally, from the intensity value at peak maximum as a function of the electrode potential during a complete cyclic voltammogram, we obtain information about the kinetics of the chloride adsorption/desorption induced surface reconstruction process.

2. Experimental. All CV, RAS and STM measurements were carried out in the same electrochemical cell of a dedicated, home-made EC-STM. This EC-STM was designed and built at the Institute of Physical and Theoretical Chemistry of the University of Bonn, Germany, and has been described in detail elsewhere9.In short, the instrument combines an STM with an electrochemical cell of a volume of 2.5 ml. This enables a direct comparison of in situ STM and electrochemical measurements, namely cyclic voltammetry, in the same cell. Furthermore, by means of a specifically coupled design of the bipotentiostat with the electronics of the STM it is possible to record tunneling currents in three different modes, namely i) at constant electrode potential and constant bias voltage (potentiostatic imaging mode), ii) during changing electrode potential, i.e. during the scan of a cyclic voltammogram, and constant bias (potentiodynamic imaging mode), which is possible thanks to the very low drift sensitivity of the whole instrument and a high quality of the tip scanner, and iii) at a fixed surface position with constant electrode potential but varying bias voltage (local spectroscopic mode). All functions of the STM are fully software controlled including safety measures against tip-sample crashes. The tunnelling tips were electrochemically etched from a 0.25 mm diameter tungsten wire and insulated with a hot-melt glue9 . STM and electrochemical cell are placed in a small 5



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

aluminium chamber in order to screen off electromagnetic radiation and noise. The chamber is permanently purged with purified argon9,26 in order to avoid oxygen contamination.STM images were analyzed with the WSxM software27 . The RAS measurements in the UV-VIS range (1.5 – 4.5 eV) were performed using a conventional twopolarizer apparatus28-30 placed above the Cu(110) sample after acquisition of the STM images. The incidence of the incoming light beam was nearly normal (less than 4° off normal) and light reflected from the immersed sample surface was detected. To this end, in order to allow the optical alignment the scanner/preamplifier stage of the STM had to be removed; in other words the STM and RAS data are taken from the same sample under identical conditions, but not at the same time. The consequent temporary exposure of the liquid surface to atmosphere did not produce any oxygen contamination at the metal surface as deliberately demonstrated by XPS measurements. Cyclic voltammetry, characterizing the adsorption and desorption processes at the surface, has been used to compare and ensure the status of the sample during the in situ STM and the consecutive RAS experiments. While the Cu(110) sample constitutes the working electrode in the electrochemical cell, a Pt wire serving as counter electrode and a saturated Ag/AgCl reference electrode (all potentials of the working electrode in this paper are referred to this electrode) were also placed inside the cell. The Cu(110) single crystal, purchased from Mateck, Jülich, Germany, was mechanically polished and then etched for 20 seconds in 50% orthophosphoric acid solution at an electrode potential of +2.5 V. After this procedure, the sample was rinsed with HCl solution and then mounted inside the EC cell filled with 10 mM hydrochloric acid solution. For all solutions, high purity water (Milli-Q purification system, resistivity> 18 MΩ cm) and reagent grade, or better, chemicals were used. At first the quality of the copper surface was checked by STM, showing the cleanliness and structural order of the surface and, thus, the suitability of this sample preparation procedure. During this stage the sample was always held at an electrode potential value near -605 mV with respect to the reference electrode in order to prevent Cl- ion adsorption on the surface (see below). Moreover, we had also the possibility to immerse the sample into an electrochemical cell in a separate preparation chamber of a UHV system, filled with high purity argon, in order to measure CV cycles. After emersion of the sample at a chosen potential, removal of 6


Page 6 of 36

Page 7 of 36

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 EC cell and pumping down the preparation chamber to 10-9 Torr the sample could be transferred without contact to air into the main UHV (p < 10-10Torr) analysis chamber where complementary X-ray Photoelectron Spectroscopy (XPS) and Low Energy Electron Diffraction (LEED) measurements could be performed. The RAS signal is defined as follows:

(1)

where Rα (Rβ)is the intensity of the reflected light from the sample when the light beam is linearly polarized along direction α (β), equal to the square modulus of the Fresnel reflectivity coefficient for the same polarization30 . RAS measures the anisotropy of the sample reflectance between directions α and β. Since the EC-STM measurements of the sample in the EC-STM cell were taken first, the orientation of the surface axes with respect to the sample holder and the EC-cell (fixed within the aluminium housing of the EC-STM) were known, as also the α and β axis orientation of the RAS spectrometer. In the present experiment, the sample has been oriented such that the electric field directions α and β are aligned along the [1-10] and the [001] crystallographic axes of the Cu sample, respectively. In general, ∆R/R contains the anisotropy of the dielectric function of the surface (or of the adsorbed layer) convoluted with the bulk dielectric properties31, that are well known from the literature32,33. Within a classical three layer model34 - where water(transparent in the used photon energy range35), the surface (or the adsorbed layer) and copper bulk are described by their optical functions - the unknown anisotropic dielectric function can be extracted31 . Since the absorption of copper is not negligible in the considered energy range32,33 , the spectra need to be carefully analysed to obtain the optical transitions of the surface (layer). In this paper we will use RAS spectra as a signature of the physical state at the solid/liquid interface, without entering into greater details of the excitation process itself. We just mention that a RAS peak coming from an adsorbate

7



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 36

layer is proportional to the density of molecules or atoms adsorbed onto the surface, thereby allowing a quantitative evaluation of the coverage36,37 . RAS is a powerful technique30 to investigate the optical properties of reconstructed surfaces in UHV38, confined structures39 and intrinsic or extrinsic surface states40. In particular, the RAS spectrum of the adsorbate-free Cu(110) surface, collected in an EC cell, has recently been compared with the well-known Cu(110) anisotropic spectrum acquired in UHV22 .The intrinsic clean Cu(110) electronic properties (i.e. surface states present at the solid-liquid interface) can in principle affect the optical properties induced by the adsorption of Cl- ions onto the copper surface, if their characteristic photon energies fall in the same energy range with a consequent overlap of the corresponding line shapes. In order to separate the optical anisotropy of the layer adsorbed at the solid/liquid interface, we have therefore defined the ∆RAS signal as follows:

(2) where V represents a generic electrode potential, while V0 =-605 mV is the potential applied to the copper electrode in order to have an adsorbate-free surface (see below). It has been demonstrated that the ∆RAS can be used to enhance small but meaningful anisotropic signals in RAS spectra41 . Peaks in a ∆RAS spectrum may therefore in the present case be taken as a direct sign of the modification induced by changes of the surfaces resulting from the Cl- adsorption.

8


Page 9 of 36

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


Fig.1. Cyclic voltammogram (CV) of the Cu(110)/HCl 10 mM system, with a scan rate dV/dt=2 mV/s. The CV is characterized by a reversible current peak pair which is attributed to the chloride adsorption (positive direction: A, B) and chloride desorption process (negative direction:C, D). The numbers indicate the acquisition starting point of different STM images (see text below ). Each STM picture (see Fig.3 and Fig. 4) is completely recorded between two consecutive numbers.

3. Results and discussion. 3.1 Cyclic voltammetry. Fig.1 displays a CV curve acquired in the EC cell with the Cu(110) sample being immersed in a 10 mM HCl solution, after the etching procedure reported above. The arrows indicate the direction of the CV scan (scan rate dV/dt=2 mV/sec). We have limited the scan interval from about -600 mV to -300 mV to avoid both the anodic dissolution of the metal and the cathodic hydrogen evolution. In the potential scan along the positive (negative) direction two peaks A,B (C,D) are detected. Peaks B and C are separated by 30 mV. The same potential difference exists between peaks A and D. The significant increase 9



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

of the anodic current with increasing potential, in particular beyond peak B, is consistent with the very slow scan rate chosen here (see below). In electrochemistry, the anodic (cathodic) current peak of a CV cycle is attributed to adsorption (desorption) of anions present in solution. In our case, the presence of two peaks in both scan directions at similar electrode potentials, suggests that either two different anions are present in solution, or that the same anion is involved in two different electrochemically induced processes at the sample surface, e.g. two different adsorption (desorption) states or one adsorption (desorption) process followed by a surface reconstruction (deconstruction). Our combined EC-STM and RAS results support the second option (see below). The cyclic voltammogram reported in Fig. 1 considerably deviates from previously published ones which showed only one peak in either scan direction related to the adsorption and desorption of Cl- ions, at most followed by a tail following the peak in positive (negative) direction20,21. In these publications the concentration of the solution and, in particular, the scan rates were quite different from our conditions, namely 1 mM HClO4 + 0.1 mM HCl/50 mVs-1 20 and 100 mM HCl/150 mVs-1 21, respectively, as opposed to 10 mMHCl/2 mVs-1 in our case. We interpret the occurrence of the second distinct peak B and the following anodic current by a second slow process which is fully detected only if the scan rate is slow, i.e. if the system is kept at the high potential for much longer time. While the Cl-concentration enters the kinetics of this process linearly, the potential enters exponentially (Buttler-Volmer equation). As a consequence of our much slower scan rate the system lingers much longer at the high potential (above peak A) leading to a much higher turnover during this second process. Conversely, it is the "product" of this slow process at high potentials which decays during the scan in negative direction giving rise to the pronounced peak C. Our in situ STM and RAS data are fully consistent with this CV behaviour as discussed in the following sections.

3.2 Scanning Tunneling Microscopy. Fig.2 shows the Cu(110) surface at constant electrode potential of 605 mV. At this potential no ions are adsorbed on the copper surface as suggested by the CV in Fig.1. The structure shown in Fig.2 agrees with results already published from experiments with Cu(110) in liquid and in UHV21,22,24,42-44 and is in accordance with that expected for the clean Cu(110) surface: indeed the 10


Page 10 of 36

Page 11 of 36

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


parameters a and b of the unit cell, shown in the figure, are equal to 2.54±0.04 Ǻ and 3.66±0.04 Ǻ respectively, in good agreement with the values reported in the literature for the bare Cu(110) surface20,21,22, 42-44

. Consequently, we refer to this surface as the clean or reference Cu(110) surface, in close analogy with

the clean Cu(110) surface prepared in UHV23 . At such negative value of the electrode potential (-605 mV) the metal surface does not undergo significant modification or contamination in the solution. This is consistent with the CV reported in Fig.1, where the adsorption peaks of chloride are located at a potential higher than -500 mV. Some noisy, white spots are also visible in this image, with atomic dimensions and heights of the order of about 1 Å. These spots are apparently floating on the surface, as in consecutive images they do not appear at the same site but have obviously moved. The presence of atomic-size particles floating on the surface, also evident in a larger area image (Fig. 3, panels 1-2), is very likely due to some [CuCl2]- species formed with highly reactive low-coordinated Cu atoms at defect sites, or being a residue from the mass transport inherent in the restructuring/faceting process of the Cu(110) surface discussed below19. Thus, STM images have then been acquired while the electrode potential has been varied through a full CV scan (see Fig. 1). The increasing numbers along the curve in Fig.1 indicate the starting point, i.e. the respective electrode potential, of the corresponding STM image, shown in Fig. 3 and Fig. 4. During acquisition of a single image (e.g., the nth image), the variation between potential values corresponding to two following numbers (n and n+1) in the CV spectrum has been continuously monitored by STM. This means that in this mode (potentiodynamic mode) each STM scan line, (except for Fig.2 taken as a whole at the constant electrode potential of -605 mV, see below), refers to a specific potential value. As a consequence, the scanned area reported in each STM image of Figs. 3 and 4 directly shows the effect of the potential variation in a range of about 50 mV (the difference in electrode potential at which two consecutive images are recorded).

11



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

Fig.2. EC-STM

image (4x4 nm2 ) of the Cu(110) surface taken at E = -605 mV. The arrows indicate the copper unit cell and the main

crystallographic directions. No evidence of adsorbed species is detected at this electrode potential, although some objects of atomic size (white spots) seem to float on the surface (see text for details). The picture has been acquired with Vbias=33 mV (filled states) and a tunneling current of It=8 nA. Crystallographic directions in the surface plane are shown.

STM images acquired during a positive scan are reported in Fig. 3. In panel 1 the image of the clean Cu(110) surface exhibits a terrace having a width > 70 nm. In the scanned area, regions with a spotty appearance are seen on the surface, which at higher magnification reveal a structure similar to the white spots of Fig. 2. Such regions have always been detected on the bare copper substrate in HCl solution, with the individual spots being mobile as mentioned before. When, in positive direction, the onset of the first CV peak (A in Fig.1) is reached, corresponding to the bottom of panel 2 and top of panel 3 in Fig. 3, a restructuring process of the surface is visible. It starts from the step edges (bottom left corner of panel 2) and then evolves into the copper terraces with the formation of long channels (up to several tens of nanometers), aligned along the [001] crystallographic direction (top of panel 3). In contrast to the rather “frizzy” step edges these erosion channels exhibit sharp boundaries, thus

12


Page 12 of 36

Page 13 of 36

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


implying some stabilizing process due to strong chemical interaction between copper atoms and ions in solution19. A similar observation has been reported for Cu(110) surfaces exposed to molecular Cl2 in UHV25.

Fig.3. EC-STM images (81x81 nm2 ) recorded during the positive sweep of the CV cycle, monitoring the adsorption process of Cl- on Cu(110), as a function of the electrode potential. The lateral arrows indicate the variation of the potential value during the STM scan. The numbers of the panels correspond to the numbers reported along the CV cycle in Fig. 1. The pictures have been acquired with Vbias=30 mV (filled states) and a tunneling current of It=1.4 nA.

A recent STM investigation in UHV of Cu(110) surfaces exposed to Cl2 has clearly assessed that a faceting process happens at the Cu(110) plane at high chloride exposures, thus producing (210) planes25. We adopt this finding and conclude that also in our experiment in solution faceting of the Cu(110) surface occurs at potentials > -510 mV (see below). When the potential is changed further to higher values (-470/-480mV, close to the onset of the second anodic current peak B in the CV, see Fig.1) we observe also the development of extended directional stripes well aligned along the [001] direction (panel 3). These stripes – whose width is in the range 2-3 nm - always start after or, at least, in coincidence with the copper restructuring and channel formation, and extend onto copper 13



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 36

terraces, preferentially from initial nucleation sites close to step and channel edges. Most of them are positioned adjacent to a channel. At higher potentials (panels 4, 5, 6) passing the second anodic current peak (B) in Fig.1, the surface appears more and more uniform, the number of stripes and the overall surface roughness increase. Now the stripes create a dense adlayer, where the distance between two adjacent rows is around 2nm. This restructuring process of channel and subsequent stripe formation is obviously induced by the chloride adsorption. The faceting process has now apparently slowed down and at the end of the positive scan (-300 mV, panel 6) the stripes are long up to several tens of nanometers and have a width in the range of 2 to 3 nm. Some stripes have overgrown the lower ones. From height profiles perpendicular to the stripes and channels, an inclination angle of 18°±3° is obtained, consistent with (210) planes (see an example in the Supporting Information, figure S1). The conclusive demonstration that chlorine is on the copper surface comes from LEED and XPS data. After the Cu(110) sample was in contact with the 10 mM HCl solution at about -300 mV for a few seconds and then transferred air-free (see Experimental section), a clear signal from the Cl2p level is measured (spectra not reported). LEED images taken under the same experimental conditions (see Supporting Information, figure S2) show that the 1x1 pattern measured for the clean Cu(110) surface is modified by additional stripes in the [110] direction, in agreement with results of Stickney et al.

45

. These XPS and LEED

observations are consistent with the presence of a very thin layer of chlorine, with some degree of order and with stripes in the [001] direction, but no long range order in the direction of the stripes.

14


Page 15 of 36

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


Fig.4. EC-STM images (81x81 nm2 )recorded during the negative sweep of the CV cycle, monitoring the desorption process of Cl- from Cu(110) and the disappearance of the stripes and erosion channels, as a function of the electrode potential. The lateral arrows indicate the variation of the potential value during the STM scan. The numbers of the panels correspond to the numbers reported along the CV cycle in Fig. 1. The pictures have been acquired with Vbias=30 mV (filled states) and a tunneling current of It=1.4 nA.

Fig. 4 reports a series of STM images acquired in the reverse CV scan, i.e. along the negative direction. During the negative scan, the system undergoes the reverse process (Fig. 4). The starting image ( panel 7) shows again the stripes. The surface morphology remains basically unchanged in panels 8, 9 and 10 -while the first desorption peak C is reached (~ -420/-450 mV), demonstrating the stability of the stripe structure in this potential regime. In panel 11, an abrupt variation is clearly visible at around -500 mV (onset of the desorption peak D in Fig. 1). At this potential value, the stripes disappear and wide copper (110) terraces become visible instead (up to 4 terraces), with “fuzzy” borders showing again mobility at the surface. Some channels are clearly still observable implying a different kinetics of the healing process of the surface when going to negative potentials compared to the positive sweep. Panel 12 proves that for potential values

15



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

between -550 and -600 mV the original clean copper surface is completely regenerated. The occurrence of reversible processes as suggested by the CV in Fig.1 is, thus, confirmed by the STM analysis at the nanoscale regime. Mobile spots on the clean (110) copper planes, earlier assigned to [CuCl2]- species, are now visible again, suggesting that the complex mass transport due to the whole restructuring process has very likely a longer time scale than our acquisition time for a single image. In order to check the evolution of the channels and stripes during their formation and removal, we have directly switched the electrode potential between two specific values, namely from -550 mV to -500 mV and vice-versa. By applying these selected potential values, in one case we stop before the first anodic current peak of the CV (peak A in Fig.1), while in the other case we cross the second cathodic current peak D. The experimental result is reported in Fig.5.

16


Page 16 of 36

Page 17 of 36

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


Fig.5. EC-STM image (81x81 nm2) acquired before and after the abrupt change (dashed line) of the electrode potential from -550 mV to -500 mV (A) and vice-versa (B). The pictures have been acquired with Vbias=153 mV (filled states) and a tunneling current of It=1.5 nA. Significant structural details have been highlighted (see text): a stripe (S) and three channels (C1,C2,C3) due to chloride adsorption.

In panel A we observe that the restructuring process due to the change of the electrode potential (dashed line) is instantaneous. The stripe S which can be noted in the lower part of panel A, below the dashed line appears between just two line scans. Also the reverse process (panel B) happens suddenly, only two channels (C2, C3) are still visible after the stripes have disappeared. Their persistence may have two reasons, either they are "pinned" by some foreign atoms, a phenomenon well known from etch-structures, and/or broader channels are more stable than narrower ones. The latter seems to be supported by the channels C1 and C2 in panel A and B, respectively. In particular the broad part of channel C2 in panel A survives in panel B, while its narrower extension disappears.

17



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

Fig. 6 ∆RAS spectrum (for definition of ∆RAS see text) expressing the difference of two RAS spectra, measured at two different electrode potentials: -605 mV (clean Cu(110) surface) and -300 mV (at the completion of the adsorption process of Cl- ions on the Cu surface). The line shape highlights the optical anisotropic contribution of the Cl- induced modification of the Cu(110) surface

3.3 Reflectance Anisotropy spectroscopy. The evolution of surface morphology as a function of the applied potential described in the previous section, suggests that significant concomitant changes of the electronic properties of the liquid-solid interface (due to the adsorption/desorption of ions on the Cu(110) surfaces, and to the charge transfer between the adsorbed species and the substrate) occur that should perturb the electronic states at the metal/liquid interface. Moreover, the formation of stripes and channels on the surface, oriented along the [001] direction, implies that an important structural anisotropy of the sample exists. As a consequence also the optical properties of the bare substrate must be affected by the adsorption/desorption of the chloride ions, and such modification should be detectable by RAS. Indeed a structure centred at 2.5 eV appears in the ∆RAS spectrum when the electrode potential is set at -300 mV, as shown in Fig.6. From the definition of ∆RAS (see Eq. 2), and from the potential values between which the difference has been evaluated, namely -600 mV and -300 mV, this structure is clearly due to the adsorption of Cl- on the copper surface during the positive scan, whose final effect is the almost complete coverage of the surface by stripes, as shown in the STM images (panel 6, Fig. 3). 18


Page 18 of 36

Page 19 of 36

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


Fig.7. Evolution of the ∆RAS signal as a function of the electrode potential (dV/dt=2mV/s) , acquired at fixed photon energy (2.5 eV). The chosen photon energy corresponds to the maximum of the curve in Fig. 6. The arrows indicate the scan direction (positive and negative direction). The numbers represent the potential values at which STM picture have been acquired (Fig. 3 and 4). Adsorption and desorption processes show different time evolutions. As a consequence, the ∆RAS signal acquired at fixed photon energy goes through a “hysteresis cycle”.

According to Eqs. 1 and 2, a positive sign in the ∆RAS spectrum means that an increase of reflected light along the 110 direction or a decrease of light along the [001] direction occurs when the Cl- induced stripes are formed. The peak intensity (about 2.5x10-3) is consistent with a surface contribution31,38,40 , and the large line width (FWHM~1 eV) suggests that the sample is not homogeneous, in the sense that the stripes are neither equidistant nor equally broad. All these facts are in qualitative agreement with the surface morphology shown in EC-STM images (Fig.3, panel 6). In a previous RAS investigation of the Cu(110)/electrolyte interface22 , it has been demonstrated that surface states exist on the clean copper metal surface immersed in solution, at a photon energy slightly different with respect to that observed under UHV conditions (2.2 eV vs 2.1 eV). The different photon energy (with respect to the clean surface) of our ∆RAS peak centered at 2.5 eV, and the fact that ∆RAS gives the difference of anisotropy at a certain electrode potential (-300 mV) with respect to that of the clean surface, clearly call for 19



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

an assignment of the ∆RAS signal in Fig. 6 to the adsorption of Cl- ions on the copper surface, and, more specific, to the chloride induced formation of the highly anisotropic stripe structure seen in the STM images along the [001] direction of the (110) surface. This conclusion is definitely strengthened by the correlation of the emergence and growth of the 2.5 eV ∆RAS peak with the adsorption of chloride indicated by the double adsorption peak and the high current in positive scan direction of the CV in Fig.1, and then by the XPS results showing that Cl is adsorbed on the metal surface after immersion in HCl solution. Up to date, no calculations exist in literature for the electronic and optical properties of the Cl-covered Cu(110). This means we cannot definitively attribute the measured optical anisotropy to Cl-induced surface states, that is to intrinsic electronic properties of the directional stripes, or to some structural arrangement of atoms within the stripes. In addition, the ∆RAS peak occurs in the spectral range where bulk optical transitions are present32,46 : therefore, also the explanation in terms of surface-modified bulk states47-49 cannot be discarded. However, even in the absence of a detailed understanding, we conclude that the ∆RAS spectrum in fig. 6 is the signature of the chloride-induced stripes on the Cu(110) surface. We have studied the optical anisotropy signal of the system by varying the electrode potential in the same range of the CV cycle (Fig.1) and as in Figs. 3 and 4 in order to correlate its evolution with the structural changes monitored by STM. Fig. 7 reports the dependence of the ∆RAS signal on the electrode potential, measured at fixed photon energy (namely 2.5 eV, corresponding to the peak maximum position in Fig.6). For a useful comparison with previous figures, the numbers along the optical cycle displayed in Fig.7 refer to the same corresponding positions along the CV cycle in Fig.1 and to the related STM images in Figs. 3 (panels 1-6) and 4 (panels 7-12). The signal starts to rise in positive scan direction in accordance with the onset of the first adsorption peak A in the CV (Fig.1), and the slope changes once the second peak B is reached at about -420 mV. In the reverse direction, i.e. the negative scan, the ∆RAS intensity keeps basically constant until -480 mV and then drops rather sharply. The optical anisotropy signal follows the evolution of the morphology changes as displayed by the STM images in Figs.3, 4 and 5. Also this optical tracking indicates that the process is slower during the formation of the chloride induced stripes, than in the opposite, i.e. negative, direction where also a more abrupt change in the morphology is recorded. It is interesting to notice that during the positive scan the flank of the ∆RAS plot has always an appreciable non20


Page 20 of 36

Page 21 of 36

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


zero slope at about -300 mV, indicating that the underlying adsorbate induced restructuring process is still proceeding at this electrode potential.

Fig.8. Evolution of the ∆RAS signal acquired at fixed photon energy ( 2.5 eV, see Fig. 6) as a function of time. At the beginning the electrode potential is set at -500 mV, then it is suddenly changed to -550 mV, and, after few seconds, it is put back to – 500 mV, and so forth. The red dashed lines highlight the slope corresponding to the potential change from -500 mV (Cl- adsorption) to -550 mV (Cl- desorption).

We have also mimicked the experimental condition of Fig. 5 and monitored by ∆RAS the dynamics of the adsorption/desorption process. Fig. 8 reports the variations of the ∆RAS signal at 2.5 eV when the electrode potential is repeatedly switched abruptly (within a few milliseconds) from -500 mV to -550 mV and viceversa. Like the disappearance of the Cl- induced stripes in STM images (Fig.5), the ∆RAS signal suffers a fast reduction (the corresponding slope is marked with a dashed line). On the contrary, the chloride induced surface restructuring follows a slower process with a more complex time dependence: an initial steep slope is followed by a slower further rise. This asymmetry determines a sort of “hysteresis" cycle in the ∆RAS data when they are plotted as in Fig.7. Summarizing so far, all three techniques suggest a two-step process in the chloride adsorption induced restructuring of the Cu(110) surface. i) The cyclic voltammogram in Fig.1 shows two successive current peaks followed by a relatively high current level at positive potentials. ii) The series of STM images in Fig.3 shows a rather spontaneous formation of grooves and stripes of monolayer thickness when crossing the first 21



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

CV adsorption peak A followed by a further growth of the stripes in width, length and, in particular, height when passing the second “adsorption” peak B and beyond. iii) Finally, also the variation of the ∆RAS intensity as a function of the electrode potential in Fig.7, namely the obvious change in slope during the positive scan, indicates a change in kinetics, roughly after 2/3 of the maximum ∆RAS intensity (at -300 mV) has been reached near -420 mV. All three observations as a function of increasing electrode potential support the following scenario: at potentials below -550 mV chloride anions start to react with individual, particularly low coordinated Cu surface atoms, forming mobile complexes, e.g. [CuCl2]-, which appear as mobile bright spots in the STM images. At -550 mV this reaction is enhanced leading first to the formation of grooves, i.e. the consumption of surface copper atoms and thus the missing copper rows (dark in the STM images), followed by the formation of added rows and stripes (bright) of predominantly monoatomic height. The latter first grow in number and length until higher (brighter) layers start to appear. In particular, the thickening seems to be a slower process which continues to go on up even at -300 mV as suggested by the relatively high current in the CV cycle above peak B as well as by the non-zero slope of the ∆RAS signal in positive direction at -300 mV. This ongoing process means that the reconstruction process continues above the second anodic current peak. It is also in line with the large “desorption” peak C in the subsequent negative scan, showing a faster desorption/deconstruction of the Cu surface (consistently with the ∆RAS cycle).

4. Conclusions We have presented an investigation of the structure and structural transitions at a Cu(110) surface in 10 mM HCl solution as a function of the potential applied to the electrode, by using Cyclic Voltammetry, Reflectance Anisotropy Spectroscopy and in situ Electrochemical Scanning Tunneling Microscopy. The evolution of the interface after adsorption/desorption of Cl- ions on/from Cu(110), imaged by STM, has been correlated to the development of a characteristic optical anisotropy signal at 2.5 eV, while ex situ XPS and LEED have confirmed the presence of a chloride layer at a Cu(110) surface prepared in similar conditions. From the resulting optical cycle (measured by cycling the electrode potential, as in CV), we have obtained a 22


Page 22 of 36

Page 23 of 36

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


signature of the adsorption/desorption processes. In particular the data indicate that the adsorption process is characterized by the evolution of characteristic grooves and stripes along the [001] direction, then followed by a reconstruction of the Cu surface, while the desorption of ions from the surface (ending after restoring the clean copper surface) is a slower process. The combined application of EC-STM and RAS paves the way for the possibility to obtain accurate information on the kinetic of adsorption/desorption processes in liquid environment.

5. Acknowledgements Some of the authors (G.B., A.V., B.B.) acknowledge the financial support from “ProgettoVigoni”.

Supporting Information Available Figure S1. EC-STM image of an isolated stripe due to chloride adsorption on Cu(110) (electrode potential 500 mV). Height profile across the stripe shown in the EC-STM image (to evaluate the tilt of the facets due to adsorption). Figure S2. LEED patterns of the Cu(110) surface: clean, and after adsorption of chloride (at -300 mV of electrode potential) in the electrochemical cell introduced in the preparation chamber. This material is available free of charge via the Internet at http://pubs.acs.org.

References. [1] Wandelt, K.; Thurgate, S. (editors), Solid Liquid Interfaces: Macroscopic Phenomena – Microscopic Understanding, Springer, Berlin, 2003. [2] Krischer, K.; Savinova, E. R., Fundamentals of Electrocatalysis, in Handbook of Heterogeneous Catalysis, Wiley (2008), 1873–1905.

23



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 24 of 36

[3] Maschmeyer, T.; Rey, F.; Sankar G.; Thomas J. M. Heterogeneous Catalysts Obtained by Grafting Metallocene Complexes onto Mesoporous Silica, Nature 1995, 378, 159 – 162. [4] Kang, S. M.; Park, S.; Kim, D.; Park, S. Y.; Ruoff, R. S.; Lee, H. Simultaneous Reduction and Surface Functionalization of Graphene Oxide by Mussel-Inspired Chemistry, Adv. Funct. Mater. 2011, 21, 108–112. [5] Mohanty, U. S. Electrodeposition: a Versatile and Inexpensive Tool for the Synthesis of Nanoparticles, Nanorods, Nanowires, and Nanoclusters of Metals, J. Appl. Electrochem. 2011, 41, 257-270. [6] Magnussen, O.M. Electrochemistry as a Nanoscience, Electrochem. Soc. Interface 2006, 15 (3), 23-27. [7] Ming-Pei, L.; Cheng-Yun, H.; Wen-Tsan, L.; Yuh-Shyong, Y. Probing the Sensitivity of Nanowirebased Biosensors Using Liquid-gating, Nanotechnology 2010, 21, 425505. [8] Andricacos, C.; Uzoh, C.; Dukovic, J.O;. Horkans, J.; Deligianni, H. Damascene Copper Electroplating for Chip Interconnections, IBM J. Res. and Develop., 1998, 42, 567-564. [9] Wilms, M. ; Kruft, M. ; Bermes G.; Wandelt K. A New and Sophisticated Electrochemical Scanning Tunneling Microscope Design for the Investigation of Potentiodynamic Processes, Rev. Sci. Inst. 1999, 70, 3641. [10] Hansma, H. G.; Vesenka, J.; Siegerist, C.; Kelderman, G.; Morrett, H.; Sinsheimer, R. L.; Elings, V.; Bustamante, C.; Hansma, P. K. Reproducible Imaging and Dissection of Plasmid DNA Under Liquid with the Atomic Force Microscope, Science 1992, 256, 1180-1184. [11] Pradier, C.M.; Chabal, Y. J. (editors), Biointerface Characterization by Advanced IR Spectroscopy, Elsevier, 2011. [12] Nozik, A.J. (editor), Photoeffects at Semiconductor Electro-lyte Interfaces, American Chemical Society Symposium Series, Washington, D. C., 1981, vol. 146. [13] Höök, F.; Vörös, J.; Rodahl, M.; Kurrat, R.; Böni, P.; Ramsden, J.J.; Textor, M .; Spencer, N.D. ; Tengvall, P.; Gold, J. et al. Comparative Study of Protein Adsorption on Titanium Oxide Surfaces Using in Situ

Ellipsometry,

Optical

Waveguide

Lightmode

Spectroscopy,

and

Quartz

Crystal

Microbalance/Dissipation, Colloids Surf. B, 2002, 24, 155–170; [14] Bain, C. D. Sum-frequency Vibrational Spectroscopy of the Solid/Liquid Interface, J. Chem. Soc., Faraday Trans., 1995, 91, 1281-1296. [15] Sheridan, B.; Martin, D. S.; Power, J. R.; Barrett, S. D.; Smith, C. I.; Lucas, C. A.; Nichols, R. J.;Weightman, P. Reflection Anisotropy Spectroscopy: A New Probe for the Solid-Liquid Interface, Phys. Rev. Lett., 2000, 85, 4618–4621; 24


Page 25 of 36

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


[16] Mazine, V.;Borensztein, Y., Monitoring the Transitions of the Charge-Induced Reconstruction of Au(110) by Reflectance Anisotropy Spectroscopy, Phys. Rev. Lett., 2002, 88, 147403-147406. [17] Richarz, F.; Wohlmann, B.; Hoffschulz, H.; Vogel, U.; Wandelt, K. Surface and Electrochemical Characterization of Electrodeposited PtRu Alloys,

Surf.Sci., 1995, 335, 361-371.

[18] Pirug, G.; Bonzel, H.P.; UHV Simulation of the Electrochemical Double Layer: Adsorption of HClO4/H2O on Au(111), Surf. Sci., 1998, 405, 87-103. [19] Magnussen, O. Ordered Anion Adlayers on Metal Electrode Surfaces, Chem. Rev., 2002, 102, 679-725. [20] Wan, Li-Jun; Itaya, K. In Situ Scanning Tunneling Microscopy of Cu(110): Atomic Structures of Halide Adlayers and Anodic Dissolution, J. Electroanal. Chem., 1989, 473 ,10-18. [21] Li, W.H.; Wang, Y.; Ye, J.H.;Li, S.F.Y. In Situ STM Study of Chloride Adsorption on Cu(110) Electrode in Hydrochloric Acid Aqueous Solution, J. Phys. Chem. B, 2001, 105, 1829-1833. [22] Barritt, E. E.; Smith, C. I.; Martin, D. S.;Gentz, K.; Wandelt, K.;Weightman P. Evidence for the Observation of Surface States at the Cu(110)/Electrolyte Interface, Europhys. Lett., 2010, 92, 57005 [23] Hofmann, Ph.; Rose, K. C. ; Fernandez, V.; Bradshaw, A. M.; Richter, W. Study of Surface States on Cu(110) Using Optical Reflectance Anisotropy, Phys. Rev. Lett., 1995, 75, 2039-2042 [24] Carley, A.F.; Davies, P.R.;Harikumar, K.R.; Jones, R.V. A Low Energy Pathway to CuCl2 at Cu(110) Surfaces, Phys.Chem.Chem.Phys., 2009, 11, 10899-10907 [25] Andryushechkin, B.V.;Cherkez, V. V.; Pavlova, T.V.;Zhidomirov, G.M.; Eltsov, K.N. Structural Transformations of Cu(110) Surface Induced by Adsorption of Molecular Chlorine, Surf. Sci., 2012, 608, 135-145. [26] Wilms, M. "Potentiodynamische Rastertunnelmikroskopie an Fest/Flüssig Grenzflächen - Apparative Entwicklung und Untersuchungen zur Sulfat-Adsorption auf Cu(111)", PhD-thesis, University of Bonn, Germany, 1999. [27] Horcas, I.; Fernandez, R.; Gomez-Rodriguez, J. M.; Colchero, J.; Gómez-Herrero J.; Baro, A. M. WSXM: a Software for Scanning Probe Microscopy and a Tool for Nanotechnology, Rev. Sci. Instrum., 2007, 78(1), 013705. [28] Aspnes, D. E.;Studna, A. A. Anisotropies in the Above—Band-Gap Optical Spectra of Cubic Semiconductors, Phys. Rev. Lett., 1985, 54, 1956-1959 [29] Goletti, C.; Bussetti, G.; Chiaradia, P.; Sassella, A. ; Borghesi, A. The Application of Reflectance Anisotropy Spectroscopy to Organics Deposition, Org. Electron., 2004, 5, 73-81 [30] Weightman, P.; Martin, D. S.; Cole, R. J.; Farrell, T. Reflection Anisotropy Spectroscopy, Rep. Prog. Phys., 2005, 68, 1251-1341 [31] Goletti, C.; Arciprete, F.; Almaviva, S.; Chiaradia, P.; Esser, N.; Richter, W. Analysis of InAs(001) Surfaces by Reflectance Anisotropy Spectroscopy , Phys. Rev. B, 2001, 64, 193301-1(4)

25



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

[32] Palik,E.D. (editor), “Handbook of Optical Constants of Solids”, vol. 1, Academic Press, San Diego, 1985. [33] Ehrenreich, H.; Philipp, H. R. Optical Properties of Ag and Cu, Phys. Rev., 1962, 128, 1622-1629. [34] McIntyre, J.D.E.; Aspnes, D.E. Differential Reflection Spectroscopy of Very Thin Films, Surf. Sci., 1971, 24, 417-434. [35] Mobley,C.D., Optical properties of water, in Handbook of Optics, Second Edition, M. Bass editor, McGraw-Hill, 1994. [36] Goletti, C. ; Chiaradia, P. ; Moretti, L.; Yian, W. ; Chiarotti, G. Activated Chemisorption of Oxygen on Si (111)-2× 1, Surf. Sci., 1996, 356, 68-74. [37] Goletti, C.; Bussetti, G.; Chiaradia, P.; Sassella, A.; Borghesi, A. Highly Sensitive Optical Monitoring of Molecular Film Growth by Organic Molecular Beam Deposition, Appl. Phys. Lett., 2003, 83, 4146-4148. [38] Arciprete, F.; Goletti, C.; Placidi, E.; Hogan, C.; Chiaradia, P.; Fanfoni, M.; Patella, F.; Balzarotti, A. Surface States at the GaAs(001)2×4 Surface, Phys. Rev. B, 2004, 69, 081308 (4p) [39] Steimetz, E.; Zettler, J.T.; Richter, W.; Westwood,

D.I.; Woolf, D.A.; Sobiesierski, Z. Optical

Monitoring of the Development of InAs Quantum Dots on GaAs(001) by Reflectance Anisotropy Spectroscopy , J. Vac. Sci. Technol. B, 1996, 14, 3058-3064. [40] Fazi, L.; Hogan, C.; Persichetti, L.; Goletti, C.; Palummo, M.;Sgarlata, A.;Balzarotti, A. Intermixing and Buried Interfacial Structure in Strained Ge/Si(105) Facets, Phys. Rev. B, 2013, 88, 195312 (6p). [41] Bussetti, G.;Cirilli, S.; Violante, A.;Chiaradia, P.; Goletti, C.; Tortora, L.;Paolesse, R.; Martinelli, E.;D’Amico, A.; Di Natale, C. et al.. Optical Anisotropy Readout in Solid-State Porphyrins for the Detection of Volatile Compounds, Appl. Phys. Lett., 2009, 95, 091906 (3p). [42] LaGraff, J.R.; Gewirth, A.A. In-situ Observation of Oxygen Adlayer Formation on Cu(110) Electrode Surfaces , Surf. Sci. Lett. , 1995, 326, L461-L466. [43] Vogt, M.R.; Lachenwitzer, A.; Magnussen, O.M.; Behm, R.J. In-situ STM Study of the Initial Stages of Corrosion of Cu(100) Electrodes in Sulfuric and Hydrochloric Acid Solution , Surf. Sci., 1998, 399, 49-69. [44] Chua, F.M.; Kuk, Y.; Silverman, P.J. Oxygen Chemisorption on Cu(110): an Atomic View by Scanning Tunneling Microscopy , Phys. Rev. Lett., 1989, 63, 386-389. [45] Stickney, J.L.; Ehlerz, C.B.; Gregory, B.W. Adsorption of Gaseous and Aqueous Hydrochloric Acid on the Low-index Planes of Copper , Langmuir, 1988, 4, 1368-1373. [46] Stahrenberg, K.; Herrmann, Th.; Wilmers, K.; Esser, N.; Richter, W.; Lee, M. J. G., Optical Properties of Copper and Silver in the Energy Range 2.5–9.0 eV, Phys. Rev. B, 2001, 64, 115111. [47] Schmidt, W. G.; Esser, N.; Frisch, A. M.; Vogt, P.; Bernholc, J. ; Bechstedt, F.; Zorn, M.; Hannappel, Th.; Visbeck, S.; Willig, F.et al. W. Understanding reflectance anisotropy: Surface-state signatures and bulkrelated features in the optical spectrum of InP(001)(2×4), Phys. Rev. B, 2000, 61, R16 335-16338. [48] Mendoza, B. S.; Del Sole, R. ; Shkrebtii, A. I. Theory of optical reflectance anisotropy of the natural Si(110) surface, Phys. Rev. B, 1998, 57, R12 709-12712. 26


Page 26 of 36

Page 27 of 36

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


[49] Hahn, P. H.; Schmidt, W. G.; Bechstedt, F. Bulk Excitonic Effects in Surface Optical Spectra, Phys. Rev. Lett., 2002, 67, 016402-1(4).

27



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 Image

28


Page 28 of 36

Page 29 of 36

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


4



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

Page 30 of 36

a b

[110] [001]


Page 31 of 36

-600

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

-550


1

-450

4

[110] [001]

2

-400

5

-500

-350

3

-450

6

-300 ACS Paragon Plus Environment


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

Page 32 of 36

-300

-450

7

-350

10

-500

8

11

-400

-550

9

[110]

12

[001]

-450

-600 ACS Paragon Plus Environment

Page 33 of 36

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


A

B C1

-550 mV

C1

-500 mV

S -550 mV

-500 mV C2

C3

C2


C3


1x10-3

DRAS

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


Page 34 of 36

Page 35 of 36

9

1 0

2 ,5

8 7

1 1

2 ,0

(x 1 0

-3

)

6

D R A S

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


1 ,5 5

1 ,0 4 1 2

0 ,5 0 ,0

3 1

-6 0 0

2

-5 0 0

-4 0 0


E le c tr o d e p o te n tia l (m V )

-3 0 0


3 ,0 -5 0 0 m V

-5 5 0 m V

@

2 .5 e V

2 ,5

)

2 ,0

(x 1 0

-3

1 ,5 1 ,0

∆R A S

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

Page 36 of 36

0 ,5 0 ,0 -0 ,5

-5 5 0 m V 0

1 8 0

3 6 0

-5 0 0 m V 5 4 0

7 2 0

9 0 0


tim e ( s e c )

1 0 8 0

1 2 6 0

1 4 4 0

1 6 2 0

Cu(110) Surface in Hydrochloric Acid Solution: Potential Dependent

Recommend Documents