Dissolution of CuSO4 Aqueous Electrolyte

Oct 17, 2017 - Investigated by In Situ Soft X‑ray Absorption Spectroscopy. Juan-Jesús Velasco-Vélez,*,†,‡. Katarzyna Skorupska,. †. Elias Fr...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/JPCB

The Electro-Deposition/Dissolution of CuSO4 Aqueous Electrolyte Investigated by In Situ Soft X‑ray Absorption Spectroscopy Juan-Jesús Velasco-Vélez,*,†,‡ Katarzyna Skorupska,† Elias Frei,‡ Yu-Cheng Huang,§,∥ Chung-Li Dong,§ Bing-Jian Su,⊥ Cheng-Jhih Hsu,§ Hung-Yu Chou,§ Jin-Ming Chen,∥ Peter Strasser,# Robert Schlögl,†,‡ Axel Knop-Gericke,‡ and Cheng-Hao Chuang*,§

J. Phys. Chem. B 2018.122:780-787. Downloaded from pubs.acs.org by UNIV OF SUNDERLAND on 10/06/18. For personal use only.



Department of Heterogeneous Reactions, Max Planck Institute for Chemical Energy Conversion, Mülheim an der Ruhr 45470, Germany ‡ Department of Inorganic Chemistry, Fritz-Haber-Institut der Max-Planck-Gesellschaft, Berlin 14195, Germany § Department of Physics, Tamkang University, New Taipei City 25137, Taiwan ∥ National Synchrotron Radiation Research Center, Hsinchu 30076, Taiwan ⊥ Department of Mechanical Engineering, National Central University, Chungli 320, Taiwan # Department of Chemistry, Technical University Berlin, 10623 Berlin, Germany ABSTRACT: The electrodeposition nature of copper on a gold electrode in a 4.8 pH CuSO4 solution was inquired using X-ray absorption spectroscopy, electrochemical quartz crystal microbalance, and thermal desorption spectroscopy techniques. Our results point out that the electrodeposition of copper prompts the formation of stable oxi-hydroxide species with a formal oxidation state Cu+ without the evidence of metallic copper formation (Cu0). Moreover, the subsequent anodic polarization of Cu2Oaq yields the formation of CuO, in the formal oxidation state Cu2+, which is dissolved at higher anodic potential. It was found that the dissolution process needs less charge than that required for the electrodeposition indicating a nonreversible process most likely due to concomitant water splitting and formation of protons during the electrodeposition.



INTRODUCTION

It is well accepted that the overall reaction of copper deposition and dissolution in acidic conditions involves two electron transfers and is described simply as follow:

Among other methods, copper thin films can be deposited by physical vapor deposition (PVD), chemical vapor deposition (CVD), and sputtering but they had been shown to be expensive procedures. Electrodeposition is an attractive method used to prepare a variety of materials allowing precise control of the chemical composition and structure of the electrodeposited material.1 Electrodeposition of copper plays an important role in a multitude of applications; for example replacing aluminum with copper in semiconductor technology2 thanks to its ability to cover imperfections,3 high electrical conductivity (higher than aluminum),4 and because it works as an effective thermal expansion barrier. In addition electrodeposited copper is used extensively as an interconnect via filler material on printed circuit boards (PCbs) and as an electrode in Li ion batteries applications5 or as electrocatalyst6 in chemical energy conversion and storage. Moreover, the anodic deposition of copper species onto a conductive substrate through the formation of an insoluble electroactive film of oxides or oxihydroxide7 is of prime nature in applications such as electrocatalysis or in the fabrication of oxide ceramics and metal oxide semiconductors. © 2017 American Chemical Society

Cu 2 + + e− ⇋ Cu+

(1)

Cu+ + e− ⇋ Cu 0

(2)

where a cupric ion is reduced to form metallic copper and the other way around; a metallic atom is oxidized to produce a cupric ion where the assumption is that the copper is completely oxidized or reduced. Nevertheless, the processes that govern these reactions are not well understood. Galvanostatic measurements suggested that the redox process between Cu2+ and Cu+ is rate controlling, while Cu+ exists in reversible equilibrium with Cu0 at the electrode surface due to the fact that the overpotential, at a constant current density, changes in a similar way as an equivalent circuit composed of a resistance in parallel with a capacitance.8,9 Kinetic studies of Cu deposition and dissolution revealed that the redox process Special Issue: Miquel B. Salmeron Festschrift Received: July 9, 2017 Revised: September 22, 2017 Published: October 17, 2017 780

DOI: 10.1021/acs.jpcb.7b06728 J. Phys. Chem. B 2018, 122, 780−787

Article

The Journal of Physical Chemistry B between Cu2+ and Cu+ is rate controlling with reaction 2 being intrinsically faster than reaction 1 yielding an irreversible process with an exchange current density i0,2 ≫ i0,1 and reaction 2 faster than reaction 1. Therefore, under deposition conditions the current associated with reaction 2 cannot be larger than that associated with reaction 1 in a stationary state. In contrast, Cu+ exists in a reversible equilibrium with Cu0 at the surface at low potential polarization, yielding a posterior diffusion of the adsorbed copper (Cuad) to the lattice (Culattice).10 At higher polarization both reactions are irreversible hindering the dissolution of Cu+ in the electrolyte because the reaction 2 is faster than reaction reaction 1 yielding Cu0. On the other hand, under anodic polarization, when the Cu+ ion concentration reaches a certain level, the Cu+ ions form Cu2+ and Cu0 according to this reaction: Cu 0 → Cu+ + e−

(3)

2Cu+ → Cu 2 + + Cu 0

(4)

is used to separate the liquid phase from the vacuum in the main chamber where the detector (channeltron) is placed. The continuous flow of liquid was assured with a positive displacement micro pump (120SP series from BioChem Fluidics, USA). On the Si3N4 membrane a 20 nm thin film of Au was sputtered and used as working electrode (see below for electrode preparation by sputtering method). The main body of the cell is made of polyether ether ketone (PEEK) which is an electrical insulator and chemically inert. The cell is completed of two extra electrodes used as reference and counter (platinum wire).14 Note that the election of Pt pseudoreference electrode is due to space constrictions in the electrochemal cell. The potential was calibrated to Ag/AgCl following the procedure described by Kasem et al.15 Au-Thin Film Working Electrode Preparation. The Au working electrode was deposited by sputtering with a Cressington 208HR sputter coated machine on a Si3N4 100 nm thick membrane from the company NORCADA. First of all, an adhesion layer of Ti (99.99%, Elektronen-Optik-Service GmbH, Dortmund, Germany), 3 nm nanometer thick, was deposited in a 0.1 mbar Ar atmosphere at a current of 40 mA during 30 s. After that a 20 nm film of Au (99.99%, ElektronenOptik-Service GmbH, Dortmund, Germany) was deposited by sputtering in a 0.1 mbar Ar atmosphere at a current of 40 mA for 140 s. This approach yields the formation of a homogeneous polycrystalline thin film used as working electrode.16 The X-ray transmission through this membrane is estimated to ∼74% of the incoming intensity at the edge of interest (the signal intensity is attenuated to around ∼92% of initial intensity through 100 nm of Si3N4 and ∼80% through 20 nm Au).The accurate surface area of the working electrode is around ∼0.016 cm2. Electrolyte Preparation. The electrolyte was prepared by diluting 0.798 g of CuSO4 (Sigma-Aldrich, anhydrous powder, 99.99%) in 1 L of Milli-Q water (18.2 MΩ) at room temperature (RT), 25 °C. The electrolyte was continuously saturated with pure N2 gas by bubbling, which minimizes the presence of other dissolved gases in the liquid. The electrolyte is acidic with a 4.8 pH. Potentiostat. The potentiometric control was assured with a VersaSTAT4, Princeton Applied Research, which allows different potentiometric and amperometric control as well as impedance spectroscopy among others. Electrochemical Quartz Crystal Microbalance (EQCM). Commercially available calibrated quartz crystals (AT-cut 9 MHz) were used as working electrode (with an effective surface of around ∼0.020 cm2) for the in situ mass measurements and resistance variations. The quartz crystals were coated with a gold layer (200 nm) on Ti film, and they were used as received from the manufacturer. These crystals were placed in a PEEK cell with a special holder with a 90 μL reservoir for the electrolyte. The system was operated with an analyzer QCM922A (SEIKO EG&G) which allows measurements from 9 to 27 MHz with a resolution of 0.01 Hz and 0.01 Ω at 25−27 MHz range using the third overtone of a 9 MHz crystal. The mass change from the initial stage is determined by the resonance frequency variation using the Sauerbrey equation, which yields a mass-frequency sensitivity of 1 ng/ Hz. Using this setup 5 mM CuSO4 electrolyte (4.8 pH) was continuously flowed in the electrochemical cell at a flow rate of 0.8 mL/min avoiding the depletion of the ions or changes in the electrolyte concentration. The electrodes in the cell were the quartz crystal coated with Au (working electrode), platinum

where it is assumed that Cu+ prompts a disproportionation redox reaction 4 yielding the dissolution of copper in form of Cu2+. However, it has been reported some time ago11 that possible oxide intermediates may be formed during the electrodeposition process, where the oxygen is supplied by the water.12 This oxidation state contradicts the generally accepted mechanism introduced by Mattson and Bockris,8 which does not take into account the existence of intermediates as oxides or oxi-hydroxide species. This discrepancy is because Cu+ is not stable in acidic (pH < 3.5) conditions undergoing a rapid equilibrium with Cu which explains why copper is electrodeposited as a metal.13 Accordingly, the formation of copper oxide/hydroxide intermediates cannot be quenched in acidic conditions hindering the total understanding of their electroformation and deposition. Hereby, we use a combination of advanced in situ X-ray absorption spectroscopy (XAS) and quartz crystal microbalance (QCM) under potentiostatic control to reveal the electronic structure and, thereby, the complex reactions that govern the electrodeposition process of copper in acidic media (4.8 pH). To the best of our knowledge, no comprehensive systematic experiments in the characterization of the electronic structure at the Cu L-edges have been performed during electrodeposition/ dissolution of copper. It was found that the electrodeposition and dissolution processes in acidic media involve the formation of nonmetallic copper, most likely in form of oxi-hydroxide.



METHODS Beamline. In situ synchrotron radiation-based experiments were performed at the beamline 20A1 of the National Synchrotron Radiation Research Center (NSRRC) in Hsinchu (Taiwan). This beamline has one horizontal and one vertical focusing mirror, as well as a grating monocromator with four different gratings and a refocusing toroidal mirror. It covers energy spectral range from 60 to 1250 eV (soft X-ray range), with an average resolution of 5000. For the experiments a grating with a groove density of 1200 l/mm was used yielding a resolution of 3000 and 8 × 1010 photons/s flux with a beamspot of (1.5 × 1) mm2. In Situ Electrochemical Cell. The flow liquid cell was operated inside the main chamber of the beamline 20A1 endstation at a background pressure of ∼10−8 mbar while aqueous solutions circulated on the back side of a 100 nm thick Si3N4 membrane (from the company Norcada, Canada), which 781

DOI: 10.1021/acs.jpcb.7b06728 J. Phys. Chem. B 2018, 122, 780−787

Article

The Journal of Physical Chemistry B

polarization there is a large plateau region dominated by ion diffusion.20 Reversed sweep into the anodic polarization yields the formation of an anodic peak corresponding to the oxidation-electrodissolution of the deposited copper into the solution, which is indicated by a decrease in the mass. In addition, the density of the current at higher anodic polarization tends to zero, at higher cathodic polarizations, indicating the complete dissolution of the electrodeposited copper as well.21 Changes in the resistance shows that the electrodeposited material presents higher contact resistance than the pristine sputtered Au electrode. Taking into account that the bulk resistance of metallic copper is lower (16.78 nΩ· m, at 20 °C)22 than that of gold (25.00 nΩ·m),23 the electrodeposited copper should not be presumably in metallic form. The implication of this fact is that the electrodeposited material should contain a complex mix of metallic and/or oxidized copper as the fact that low concentrations of oxygen in the bulk yields high conductivity, like in copper metals,24 indicating a rich oxygen environment. Therefore, the electrodeposited copper oxide is a p-type semiconductor yielding the formation of a Schottky barrier junction at the solid-electrolyte interface affecting the charge transfer due to the formation of a space charge region.25The electrodeposited mass, the resistance and charge transferred to the system start to increase until the potential is reversed and raised up to a potential of −0.18 V vs Ag/AgCl, establishing the existence of two regions for the overall process: the first one ascribed to the electroreduction of Cu2+ to Cu+ and the subsequent formation of cuprous oxide5 and the second one dominated by a oxidation/dissolution process.10 Changes in the current (anodic oxidation) indicate the existence of an oxidized state in the form of Cu+ or Cu2+ at the interface involving bonds with O and/or OH groups yielding, at around −0.18 V vs Ag/AgCl, a drop in the mass resistance and charge. Note that, the onset-potential difference between mass (frequency) and resistance results from the heterogeneous structure, island growing, and noninteracting characteristics.26 The reduction in the resistance at anodic polarization is ascribed to different processes that occur simultaneously like: (i) the dissolution of the electrodeposited copper and the consequent demising of the layer thickness (the total resistance is inversely proportional to the electrodeposited copper oxide thickness). (ii) Due to disproportionation reactions that form metallic copper or other surface oxides with lower resistance than the electrodeposited copper oxide. One interesting effect is revealed by the lack of reversibility in the total charge transferred to the system as the difference between the electrodeposited drove charge (109 mC) and the final transferred charge after the dissolution (54 mC), which indicates the existence of a nonreversible process during the electrodeposition. Roughly, during the cathodic reduction process the electrodeposited copper requires two times the total charge drive due to reoxidation/dissolution processes. Therefore, during the electrodeposition other process, like water splitting and the hydrogen evolution reaction (HER), should occur, thereby requiring more electrons per Cu atom electrodeposited. These processes cannot be compensated by copper precipitation, which yields a strong pH increase. Thus, the anodic peak in the inverse scan is associated with the stripping process of the deposited copper. The charge related to the copper electrodeposition is higher than that corresponding to the electro-oxidation process.27 This behavior can be attributed to the participation of a competing reduction process of H2O or H+ during the electrodeposition to molecular

wire (counter electrode) and a Ag/AgCl reference electrode. This technique allows the monitoring of variations in the redox current, charge transfer, frequency and resistance depending in the applied potential making this technique a powerful tool for the characterization of electrodeposited materials. For easier interpretation the variation in the frequency is directly converted into mass using constant parameters, in dynamic flow, and the Sauerbrey equation. TPD Measurements. TPD analysis was performed in samples electrodeposited in a gold foil (25 μm thick with 99.985% metal purity from the company Alfa Aesar, Germany) from 5 mM CuSO4 electrolyte at a potential of −0.7 V vs Ag/ AgCl (CV plateau region dominated by ions diffusion) during 300 s. The desorption measurement was conducted on a selfconstructed TPD/TDS setup equipped with a IR furnace from Behr (IRF 10) and QMS 200 quadrupol mass spectrometer from Pfeiffer Vacuum. The sample was pretreated for 12 h under HV condition. The desorption was performed at a heating rate of 25 °C/min until 525 °C. As background correction, a blank measurement (reactor without sample) for each m/z ratio is subtracted from the mass spectrometry data.



RESULTS AND DISCUSSION Changes in the resonance frequency, current, and resistance were monitored depending on the applied potential by means of an in situ EQCM, with a quartz/gold resonant electrode17 yielding a direct observation of the mass change, electrochemical reactions, growth kinetics, and interfacial properties (viscosity and density) under liquid environments.18 Figure 1

Figure 1. Current/charge−voltage curve of copper deposition and desorption on gold in 5 mM CuSO4. In situ mass/resistance−voltage curve recorded in EQCM.

shows the changes in the redox current, charge transfer, mass (frequency), and resistance in both potentiostat/galvanostat devices as a function of the applied potential (cyclic voltammogramm, CV). The CV in Figure 1 is related to the electrodeposition of copper in 5 mM CuSO4. Variations in the process current at a given potential were correlated to the variation in the mass, resistance, and charge transfer in the Au working electrode. At −0.28 V vs Ag/AgCl the cathodic current starts to increase indicating a process ascribed to the electroreduction of Cu19 and its electrodeposition as the increase in the mass proves (see Figure 1). At higher cathodic 782

DOI: 10.1021/acs.jpcb.7b06728 J. Phys. Chem. B 2018, 122, 780−787

Article

The Journal of Physical Chemistry B

(Cu2O). On the other hand, it exhibits a strong resonance peak at 930.8 eV when it is in the formal oxidation state Cu2+, in analogy to cupric oxide (CuO). Formally, it is because CuO presents open shell 3d9 electronic configuration in contrast to Cu2O which has closed-shell31 3d10. Figure 2 shows the in situ Cu L2−3 spectra as a function of the applied potential and the measured faradaic current. Each spectrum was recorded in 1 min using a continuous potential variation rate of 0.833 mV/s giving a potential resolution of 0.05 V. It is obvious that the electrodeposition of copper occurs in the oxidation state of Cu+ (tentatively ascribed to cuprite) as the peak at 933.6 eV (L3) in points (b) and (c) in Figure 2 indicates following the same trends in the intensity increase as in the mass (EQCM measurements). After that, the potential is reversed to more anodic polarization yielding a gradual oxidation from Cu+ to Cu2+ with the coexistence of both phases at around 0.1 V as Figure 2 point (d) shows. Higher anodic polarization yields the total oxidation of copper which is ascribed to the increase in the total galvanic current process. Polarizing to a more anodic potential yields the dissolution of copper as the decay in the Cu L2−3 signal intensity points out, see Figure 2e and f, following the same behavior as that described in the in situ EQCM measurements. Nevertheless, even the oxidation of copper at different potentials is clear; it is not obvious that the presence of oxygen is responsible of these oxidation states. Accordingly, temperature program desorption (TPD) measurements were conducted in a gold foil coated with electrodepostied copper at −0.5 V vs Ag/AgCl (CV plateau region dominated by ions diffusion) during 5 min. Figure 3 shows the desorption of H2O/OH (m/z = 18) and O2 (m/z = 32) from room temperature (20 °C) up to 525 °C. The first peak at 100 °C and a shoulder at 40 °C is due to the desorption of physisorbed H2O. Beyond, until 250 °C, the chemisorbed species associated with H2O/OH (m/z = 18) evolve.40,41 The fine structure and the broadened profile indicate that water is the results of a reaction of the copper species upon heating. The fast heating applied allowed for sensitive detection and inhibits readsorption and reaction but induces artifacts from diffusion and reaction limitations concluding existence of the poly condensation of a (oxi)-hydroxide species. That the result of the condensation is in part an oxide becomes apparent by the sharp decomposition peak in the m/e 32 trace indicating the oxide decomposition.

hydrogen, the HER. A portion of the generated hydrogen diffuses away from the electrode interface and hence the charge is missing during the anodic scan. Even the information provided by the in situ EQCM is really valuable, changes in the mass, current, charge and resistance at a given potential can be ascribed to different reasons, such as adsorption/desorption, oxidation/reduction, or deposition/ dissolution. After all, although EQCM is a powerful method, details in the atomistic information related to the electronic structure are systematically missed. Therefore, these measurements were complemented with in situ X-ray absorption spectroscopy in fluorescence mode (XAS-FY) using the abovementioned in situ flow cell in the beamline 20A1 of the NSRRC. The XA spectra yield information on the electronic transitions from core levels to unoccupied final state of a specific element, allowing spectroscopic identification of materials.28 Note that the thin layer of electrodeposited material (in the hundreds of nanometer range) assures that these electrodeposited electrodes are not susceptible to selfadsorption effects.29 In transition metals L2−3 are dominated by the excitation of 2p electrons to unoccupied 3d states (LUMOs orbital) providing information about the electronic structure and coordination environment of these complexes. Cu L-edge spectra can be used to determine unambiguously the oxidation state of different copper oxide species. Depending on the oxidation state, copper presents characteristic spectral shapes which make it easy to determine the oxidation state of Cu as indicated in Table 1. The wide variety of energies ascribed to Table 1. Peaks Assignment of Different Copper Oxides oxidation state Cu0 Cu+ Cu2+

main peak position (eV) 932.7,14 933.7,31 934.5,34 934,38 933.739 932,32 933.7,31 934,33 934.5,34 933.7,35 933.8,36 931.3,37 934,38 933.7,39 933.6This work 934,32 930,14 931.3,33 931,33 931,34 931.0,35 931.3,36 933.8,37 931.3,39 930.8 This work

different copper oxidation states suggests the presence of unresolved coexistence of several species and/or intermediates.30 In particular Cu+ displays an absorption resonant peak at around 933.6 eV due to its comparison with cuprous oxide

Figure 2. Cu L2−3 spectra depending on the applied potential as well as the current in 5 mM CuSO4 with Pt as counter and Ag/AgCl as reference electrodes. Each spectrum was recorder in 1 min using a continuous potential variation rate of 0.833 mV/s yielding a potential resolution of 0.05 V. For a better interpretation some significant spectra were selected at different points of interest: (a) +0.18 V, (b) -0.10 V, (c) -0.50 V, (d) +0.70 V, (e) +0.35 V, (f) +0.88 V, and (g) +0.88 V. 783

DOI: 10.1021/acs.jpcb.7b06728 J. Phys. Chem. B 2018, 122, 780−787

Article

The Journal of Physical Chemistry B

It is evident from the in situ XAS-FY measurements that copper is electrodeposited from CuSO4 in the form of Cu+ as the prominent peak at 933.6 eV (L3) proves. This fact together with the TPD measurement indicates that the electrodeposited copper exists in the form of oxi-hydoxide (Cu2Oaq), which is deposited under cathodic polarization as the EQCM measurements proved. In addition, the formation of a less conductive layer than Au is supported by the EQCM resistance measurements. Accordingly, the electrodeposition of Cu2Oaq involves the reduction of Cu2+ ions to Cu+ species which precipitate as Cu+ and due to the concomitant electrochemical discharge of H2O to H+, generating Cu2Oaq and thus increasing locally the pH at the vicinity of the electrode. Thus, in acidic conditions, copper cations are involved in the two-step reduction mechanism. This process is described as Figure 3. TPD measurements of an electrodeposited Cu electrode onto a Au foil showing the desorption of H2O/OH (m/z = 18) and O2 (m/z = 32) from room temperature (20 °C) up to 520 °C.

Cu 2 + + e− → Cu+

(5)

2Cu+ + H 2O → Cu 2Oaq + 2H+

(6)

+

yielding the formation of stable Cu which is an insoluble solid according to the Pourbaix diagram13 indicating a gradual increasing in the local pH47 which is not compensated by the Cu2O precipitation. This process is shown in Figure 4 involving 2e−, for the reduction of Cu2+ to Cu+ and the reduction of water into H+. On the reverse scan, anodic polarization causes the oxidation/dissolution of Cu2Oaq. The cathodic charge is thereby not fully recovered due to diffusion hydrogen transport away from the interface. Taking into account this fact and the in situ XAS-FY measurements the oxidation/dissolution processes are described as

Usually, the assignment of Cu+ and Cu2+ had been ascribed to cuprous and cupric oxides in the form of Cu2O and CuO respectively. However, TPD indicates that oxidation is induced by OH groups yielding the formation of formally (oxi)hydroxide species.42 The resulting solid is either a mixed-valent oxi-hydroxide surface film on a stoichiometric bulk or possibly a bulk oxi-hydroxide. This intermediate product is still reactive indicating the existence of Cu+ and Cu2+ oxi-hydroxide surface compounds43 or to the coexistence of CuOx-Cu(OH)x hydrous nature.44 In addition to this fact, there is a formation of a Cu2O and mixed Cu2O/Cu(OH)2 layer in alkaline and neutral electrolytes in concordance with the Pourbaix diagram. Under acidic conditions and anodic polarization the oxide species are inherently adsorbates constituting the initial state of an eventual bulk-phase oxidation. During the cathodic cycle, it is more likely the coexistence of Cu−O than Cu−OH groups indicative of the presence of oxygen bounded atoms. These two phases (adsorbed oxygen and oxide) are the precursors of the formation of an oxide multilayer and ultimately a bulk copper-oxide becoming favored the Oads versus the OHads in acidic media, which diminishes the activity of OH−. Thus, the ordered Oads structure is stabilized in acidic media by hydrogen bonding with hydronium ions45 proceeding the transformation of Cu(OH)2 to CuO.46

Cu 2Oaq → Cu 0 + CuOaq

(7)

Cu 2Oaq → CuOaq + e−

(8)

CuOaq + 2H+ → Cu 2 + + H 2O

(9)

At moderate anodic polarization the Cu2Oaq species can be disproportionated to Cu0 (which is not easy to detect owing to the presence of the other phases, especially cuprous oxides due to peak overlapping, see Table 1) and CuOaq reaction 7, indicating why the resistance decreases during the anodic cyclic. The reaction can also directly proceed to CuOaq mediating one e− charge reaction 8. Therefore, soluble Cu2+ ions reduce the

Figure 4. Electrodeposition and dissolution processes derived from the in situ XAS-FY, EQCM, and TPD characterization. 784

DOI: 10.1021/acs.jpcb.7b06728 J. Phys. Chem. B 2018, 122, 780−787

Article

The Journal of Physical Chemistry B amount of electrodeposited mass in the WE as the EQCM indicates. Thus, the oxidation is an irreversible process that forms CuOaq which is dissolved at higher anodic polarization yielding Cu2+ and H2O inducing the continuously dissolution of this material until the electrode is completely depleted of copper due to reaction 9, process that is shown in Figure 4 as well. It is noteworthy that the charge necessary to drive the dissolution is only one electron in contrast with the two electrons necessary for the electrodeposition, which is supported by the charge measurements done with the in situ EQCM. Therefore, the irreversibility is due to concomitant water splitting that is of different effectiveness when the potential is anodic or cathodic cycled as consequence of different electrode oxidation state.

sequences call for the parasitic inclusion of nonequilibrium species of which more types may exist than anticipated from solution chemistry as consequence of the here demonstrated overlay deposition processes with electrocatalytic transformations of the electrolyte species. The effects of aging freshly deposited films, of high electrocatalytic activity and the pronounced effects of thermal post treatment of such materials find their chemical explanation in the complexity of the initial deposition reactions.

CONCLUSIONS Recent advances in spectroscopy aim to provide new insights in the knowledge of the electrochemical processes. Classically, the electrodeposition/dissolution of copper has been described by the interpretation of variations in macroscopic magnitudes as changes in the charge, current, or impedance. However, changes in these magnitudes cannot be related unequivocally to transformations in the electronic structure or chemical composition of these materials. Here, a combination of in situ advanced XAS, EQCM, and TPD proved that the electrodeposition of copper in acidic media involves the formation of Cu+ in the form of a stable solid without the evidence of the formation of any copper metallic species (Cu0), at least within the detection limit due to the superposition of the cuprous oxide peak. Moreover, the dissolution of copper, leads to the condensation of CuO that subsequently dissolved as Cu2+ions as the in situ XAS-FY and EQCM proved. Consequently, these results provide new insights in the understanding of copper electrodeposition and dissolution processes revealing the role of the HER reaction in the formation of copper oxi-hydroxide groups. Furthermore, these experiments contribute significantly to the interpretation of the complex reaction scenario governed by performing the operando valence state determination of the above electrodeposited oxi-hydroxide complexes. Kinetic processes govern this reaction scenario. Both the condensation chemistry of Cu species under varying local pH and the potential-induced deposition-dissolution reactions are combinations of solid−liquid interface processes. Deposition and electrocrystallization reactions are intertwined with poly condensation reactions controlled by the local pH. The present article reveals in exemplary fashion how a projection of the kinetic processes onto a set of experimental conditions can look like. The ambient reaction temperature slows down all diffusion-controlled equilibration processes and tends to preserve gradient-induced inhomogeneity in the material. The result is a high energy material with residual chemical reactivity. We tried to operate within steady state experiments at each data point to minimize transient phenomena as additional complication. We note that changes in overall pH, the temperature and the agitation will quantitatively affect the findings. The qualitative sequence of events as represented in Figure should, however be independent from the kinetic boundary conditions. Future works will elucidate the sensitivity of the present findings against parameter and concentrate there in temperature and overall pH effects. A general conclusion is that electrodeposition of a metal under a defined set of conditions may not lead to a homogeneous film material. The complexity of the reaction

ORCID



AUTHOR INFORMATION

Corresponding Authors

*[email protected]. *[email protected].



Juan-Jesús Velasco-Vélez: 0000-0002-6595-0168 Elias Frei: 0000-0003-3565-1199 Chung-Li Dong: 0000-0002-4289-4677 Peter Strasser: 0000-0002-3884-436X Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was further supported by the Ministry of Education and Science of the Russian Federation (RFMEFI61614X5060007) and the Bundesministerium für Bildung und Forschung (05K14EWA) through the joint Russian− German research project “Synchrotron and Neutron Studies for Energy Storage” (SYNESTESia)”. P.S. acknowledges partial funding by the German Federal Ministry of Education and Science (BMBF) under grants #03SF0523A - “CO2EKAT” and #033RC004E - “eEthylene”. We thank DAAD for financial support in the framework of Taiwanese−German collaboration (project ID 57218279). C.H.C. acknowledges financial support from projects 104-2112-M-032-005-MY2 and 105-2911-I-032501. We thank Dr. Travis Jones for his help during the manuscript preparation.



REFERENCES

(1) Brenner, A. Electrodeposition of Alloys: Principles and Practice; Elsevier, 2013. (2) Andricacos, P. C.; Uzoh, C.; Dukovic, J. O.; Horkans, J.; Deligianni, H. Damascene Copper Electroplating for Chip Interconnections. IBM J. Res. Dev. 1998, 42 (5), 567−574. (3) Dixit, P.; Miao, J. Aspect-Ratio-Dependent Copper Electrodeposition Technique for Very High Aspect-Ratio Through-Hole Plating. J. Electrochem. Soc. 2006, 153 (6), G552−G559. (4) Lu, L.; Shen, Y.; Chen, X.; Qian, L.; Lu, K. Ultrahigh Strength and High Electrical Conductivity in Copper. Science 2004, 304 (5669), 422−426. (5) Morales, J.; Sanchez, L.; Bijani, S.; Martınez, L.; Gabas, M.; Ramos-Barrado, J. R. Electrodeposition of Cu2O: an Excellent Method for Obtaining Films of Controlled Morphology and Good Performance in Li-Ion Batteries. Electrochem. Solid-State Lett. 2005, 8 (3), A159−A162. (6) Hori, Y.; Konishi, H.; Futamura, T.; Murata, A.; Koga, O.; Sakurai, H.; Oguma, K. Deactivation of Copper Electrode” in Electrochemical Reduction of CO2. Electrochim. Acta 2005, 50 (27), 5354−5369. (7) Casella, I. G.; Gatta, M. Anodic Electrodeposition of Copper Oxide/Hydroxide Films by Alkaline Solutions Containing Cuprous Cyanide Ions. J. Electroanal. Chem. 2000, 494 (1), 12−20.

785

DOI: 10.1021/acs.jpcb.7b06728 J. Phys. Chem. B 2018, 122, 780−787

Article

The Journal of Physical Chemistry B (8) Mattsson, E.; Bockris, J. M. Galvanostatic Studies of the Kinetics of Deposition and Dissolution in the Copper+ Copper Sulphate System. Trans. Faraday Soc. 1959, 55, 1586−1601. (9) Reid, J. D.; David, A. P. Impedance Behavior of a Sulfuric AcidCupric Sulfate/Copper Cathode Interface. J. Electrochem. Soc. 1987, 134 (6), 1389−1394. (10) Bockris, J. M.; Enyo, M. Mechanism of Electrodeposition and Dissolution Processes of Copper in Aqueous Solutions. Trans. Faraday Soc. 1962, 58, 1187−1202. (11) Meilor, J. W. A Comprehensive Treatise on Inorganic and Theoretical Chemistry; Longmans: London, 1928; Vol. III. (12) Lu, D. L.; Tanaka, K. I. Single-Crystal Cu2O Formation on Amorphous Carbon Electrode and Effect of Anions on the Crystal Habit of Cu2O Particles. J. Electrochem. Soc. 1996, 143 (7), 2105− 2109. (13) Pourbaix, M. Lectures on Electrochemical Corrosion; Springer Science & Business Media, 1973. (14) Jiang, P.; Chen, J. L.; Borondics, F.; Glans, P. A.; West, M. W.; Chang, C. L.; Salmeron, M.; Guo, J. In Situ Soft X-ray Absorption Spectroscopy Investigation of Electrochemical Corrosion of Copper in Aqueous NaHCO3 Solution. Electrochem. Commun. 2010, 12 (6), 820−822. (15) Kasem, B. K. K.; Jones, S. Platinum as a Reference Electrode in Electrochemical Measurements. Platinum Met. Rev. 2008, 52 (2), 100− 106. (16) Velasco-Velez, J. J.; Wu, C. H.; Pascal, T. A.; Wan, L. F.; Guo, J.; Prendergast, D.; Salmeron, M. The Structure of Interfacial Water on Gold Electrodes Studied by X-Ray Absorption Spectroscopy. Science 2014, 346 (6211), 831−834. (17) Ward, M. D.; Buttry, D. A. In Situ Interfacial Mass Detection with Piezoelectric Transducers. Science 1990, 249 (4972), 1000−1008. (18) Buttry, D. A.; Ward, M. D. Measurement of Interfacial Processes at Electrode Surfaces with the Electrochemical Quartz Crystal Microbalance. Chem. Rev. 1992, 92, 1355. (19) Velasco-Velez, J. J.; Jones, T. E.; Pfeifer, V.; Dong, C.-L.; Chen, Y.-X.; Chen, C.-M.; Chen, H.-Y.; Lu, Y.-R.; Chen, J.-M.; Schlögl, R.; et al. Trends in Reactivity of Electrodeposited 3d Transition Metals on Gold Revealed by Operando Soft X-Ray Absorption Spectroscopy During Water Splitting. J. Phys. D: Appl. Phys. 2017, 50, 024002. (20) Łukomska, A.; Plewka, A.; Łoś, P. Electroreduction of Cupric (II) Ions at the Ultramicroelectrodes from Concentrated Electrolytes− Comparison of Industrial and Laboratory Prepared Aqueous Solutions of Copper (II) Ions in Sulfuric Acid Electrolytes. J. Electroanal. Chem. 2009, 633 (1), 92−98. (21) Velasco-Velez, J. J.; Pfeifer, V.; Hävecker, M.; Weatherup, R. S.; Arrigo, R.; Chuang, C. H.; Stotz, E.; Weinberg, G.; Salmeron, M.; Schlögl.; et al. Photoelectron Spectroscopy at the Graphene−Liquid Interface Reveals the Electronic Structure of an Electrodeposited Cobalt/Graphene Electrocatalyst. Angew. Chem., Int. Ed. 2015, 54 (48), 14554−14558. (22) Giancoli, D.; Electric Currents and Resistance. Physics for Scientists and Engineers with Modern Physics, 4th ed.; Prentice Hall: Upper Saddle River, 2009; Vol. 658. (23) Siegel, J.; Lyutakov, O.; Rybka, V.; Kolská, Z.; Švorčík, V. Properties of Gold Nanostructures Sputtered on Glass. Nanoscale Res. Lett. 2011, 6 (1), 96. (24) Drobny, V. F.; Pulfrey, L. Properties of Reactively-Sputtered Copper Oxide Thin Films. Thin Solid Films 1979, 61 (1), 89−98. (25) Trotochaud, L.; Boettcher, S. W. Precise Oxygen Evolution Catalysts: Status and Opportunities. Scr. Mater. 2014, 74, 25−32. (26) Cuenca, A.; Agrisuelas, J.; Garcìa-Jareño, J. J.; Vicente, F. Oscillatory Charges of the Heterogeneous Reactive Layer Detected with the Motional Resistance During the Galvanostatic Deposition of Copper in Sulfuric Solution. Langmuir 2015, 31, 12664. (27) Gomez, H.; Riveros, G.; Ramirez, D. Chronoamperometric Cu (II) Analysis at Gold Ultramicroelectrodes in Concentrated Sulfuric Acid Solutions. Int. J. Electrochem. Sci. 2017, 12 (2), 985−993. (28) Stöhr, J. NEXAFS Spectroscopy; Springer Science & Business Media, 2013; Vol. 25.

(29) Tröger, L.; Arvanitis, D.; Baberschke, K.; Michaelis, H.; Grimm, U.; Zschech, E. Full Correction of the Self-Absorption in SoftFluorescence Extended X-Ray-Absorption Fine Structure. Phys. Rev. B: Condens. Matter Mater. Phys. 1992, 46 (6), 3283. (30) Greiner, M.; Jones, T. E.; Klyushin, A. Y.; Knop-Gericke, A.; Schlögl, R. Ethylene Epoxidation at the Phase Transition of Copper Oxides. J. Am. Chem. Soc. 2017, 139 (34), 11825. (31) Jiang, P.; Prendergast, D.; Borondics, F.; Porsgaard, S.; Giovanetti, L.; Pach, E.; Newberg, J.; Bluhm, H.; Besenbacher, F.; Salmeron, M. Experimental and Theoretical Investigation of the Electronic Structure of Cu2O and CuO Thin Films on Cu (110) Using X-Ray Photoelectron and Absorption Spectroscopy. J. Chem. Phys. 2013, 138 (2), 024704. (32) Hävecker, M.; Knop-Gericke, A.; Schedel-Niedrig, T.; Schlögl, R. High-Pressure Soft X-Ray Absorption Spectroscopy: a Contribution to Overcoming the “Pressure Gap” in the Study of Heterogeneous Catalytic Processes. Angew. Chem., Int. Ed. 1998, 37 (13−14), 1939− 1942. (33) Velasco-Vélez, J. J.; Pfeifer, V.; Hävecker, M.; Wang, R.; Centeno, A.; Zurutuza, A.; Algara-Siller, G.; Stotz, E.; Skorupska, K.; Teschner, D.; et al. Atmospheric Pressure X-ray Photoelectron Spectroscopy Apparatus: Bridging the Pressure Gap. Rev. Sci. Instrum. 2016, 87 (5), 053121. (34) Sham, T. K.; Coulthard, I.; Lorimer, J. W.; Hiraya, A.; Watanabe, M. Reductive Deposition of Cu on Porous Silicon from Aqueous Solutions: An X-Ray Absorption Study at the Cu L3,2 Edge. Chem. Mater. 1994, 6 (11), 2085−2091. (35) Chen, C. S.; Chen, T. C.; Chen, C. C.; Lai, Y. T.; You, J. H.; Chou, T. M.; Lee, J. F.; Chen, C. H. Effect of Ti3+ on TiO2-Supported Cu Catalysts Used for CO Oxidation. Langmuir 2012, 28 (26), 9996− 10006. (36) Zhang, X.; Zhou, J.; Song, H.; Chen, X.; Fedoseeva, Y. V.; Okotrub, A. V.; Bulusheva, L. G. Butterfly Effect” in CuO/Graphene Composite Nanosheets: a Small Interfacial Adjustment Triggers Big Changes in Electronic Structure and Li-Ion Storage Performance. ACS Appl. Mater. Interfaces 2014, 6 (19), 17236−17244. (37) Nachimuthu, P.; Thevuthasan, S.; Kim, Y. J.; Lea, A. S.; Shutthanandan, V.; Engelhard, M. H.; Gullikson, E. M.; et al. Investigation of Copper (I) Oxide Quantum Dots by Near-Edge XRay Absorption Fine Structure Spectroscopy. Chem. Mater. 2003, 15 (20), 3939−3946. (38) Knop-Gericke, A.; Hävecker, M.; Schedel-Niedrig, T.; Schlögl, R. Characterization of Active Phases of a Copper Catalyst for Methanol Oxidation Under Reaction Conditions: an In Situ X-Ray Absorption Spectroscopy Study in the Soft Energy Range. Top. Catal. 2001, 15 (1), 27−34. (39) Schedel-Niedrig, T.; Hävecker, M.; Knop-Gericke, A.; Schlögl, R. Partial Methanol Oxidation Over Copper: Active Sites Observed by Means of In Situ X-ray Absorption Spectroscopy. Phys. Chem. Chem. Phys. 2000, 2 (15), 3473−3481. (40) Sakka, Y.; Uchikoshi, T.; Ozawa, E. Sintering and Gas Desorption Characteristics of Copper Ultrafine Powders. Mater. Trans., JIM 1990, 31 (9), 802−809. (41) Nygren, M. A.; Pettersson, L. G. H2O Interaction with the Polar Cu2O (100) Surface: a Theoretical Study. J. Phys. Chem. 1996, 100 (5), 1874−1878. (42) Texier, F.; Servant, L.; Bruneel, J. L.; Argoul, F. In Situ Probing of Interfacial Processes in the Electrodeposition of Copper by Confocal Raman Microspectroscopy. J. Electroanal. Chem. 1998, 446 (1), 189−203. (43) Szpyrkowicz, L.; Ricci, F.; Montemor, M. F.; Souto, R. M. Characterization of the Catalytic Films Formed on Stainless Steel Anodes Employed for the Electrochemical Treatment of Cuprocyanide Wastewaters. J. Hazard. Mater. 2005, 119 (1), 145−152. (44) Patil, U. M.; Nam, M. S.; Lee, S. C.; Liu, S.; Kang, S.; Park, B. H.; Jun, S. C. Temperature Influenced Chemical Growth of Hydrous Copper Oxide/Hydroxide Thin Film Electrodes for High Performance Supercapacitors. J. Alloys Compd. 2017, 701, 1009−1018. 786

DOI: 10.1021/acs.jpcb.7b06728 J. Phys. Chem. B 2018, 122, 780−787

Article

The Journal of Physical Chemistry B (45) Chan, H. Y. H.; Takoudis, C. G.; Weaver, M. J. Oxide Film Formation and Oxygen Adsorption on Copper in Aqueous Media as Probed by Surface-Enhanced Raman Spectroscopy. J. Phys. Chem. B 1999, 103 (2), 357−365. (46) Candal, R. J.; Regazzoni, A. E.; Blesa, M. A. Precipitation of Copper (II) Hydrous Oxides and Copper (II) Basic Salts. J. Mater. Chem. 1992, 2 (6), 657−661. (47) Lopez-Salvans, M. Q.; Sagues, F.; Claret, J.; Bassas, J. Fingering Instability in Thin-Layer Electrodeposition: General Trends and Morphological Transitions. J. Electroanal. Chem. 1997, 421 (1−2), 205−212.

787

DOI: 10.1021/acs.jpcb.7b06728 J. Phys. Chem. B 2018, 122, 780−787