2.1. Cyclic Voltammetry
Typical cyclic voltammograms (CVs) of the model Pt(111) electrode in 0.1 M HClO4 and 0.05 M H2SO4 solutions are shown in Figure 1. Cyclic voltammetry in clean HClO4 electrolytes is normally used as a sensitive probe to evaluate the Pt(111) surface status and quality.35 Generally, three distinct regions are distinguishable in the CVs (Figure 1): 1) the hydrogen adsorption/desorption region, between 0.05 V and ∼0.35 V (RHE), 2) the double-layer region, between ∼0.35 V and approximately 0.55 V (RHE) and 3) adsorption/desorption of *OH (where * denotes adsorbed species), between approximately 0.55 V and 0.8 V (RHE). Particularly, the height and sharpness of the “butterfly” peaks at approximately 0.8 V (RHE) are commonly used to evaluate the surface quality of Pt(111) crystals as well as to ensure the cleanliness of the system.35 The Pt(111) electrode in contact with 0.05 M H2SO4 solution reveals three adsorption regions as well (Figure 1). However, the CVs obtained in 0.05 M H2SO4 and 0.1 M HClO4 are only similar in the potential region between 0.05 V and ∼0.35 V (RHE). The region between approximately 0.35 V and 0.55 V (RHE) is assigned to the specific adsorption and desorption of (bi)sulfates with sharp peaks at ∼0.52 V, corresponding to disorder/order phase transitions in the (bi)sulfate adlayer.36 An ordered ()R19.1° superstructure dominates in the region between ∼0.52 V and ∼0.8 V consisting of (bi)sulfates and water molecules.36 At potentials more positive than ∼0.8 V (RHE) further rearrangements in the adsorbate layer occur, most likely triggered by initial *OH-adsorption. The CVs shown in Figure 1 are close to the state-of-the-art CVs known for these systems revealing high-quality Pt(111) surface. Notably, OH-adsorption is suppressed in the H2SO4 electrolyte compared to HClO4 solution in the corresponding electrode potential regions.
Figure 1. CVs characterizing Pt(111) model electrodes in 0.1 M HClO4 (••••) and in 0.05 M H2SO4 (—), dE/dt=50 mV s−1.
Download figure to PowerPoint
Figure 2 A compares two CVs for the Pt(111) and the Cu–Pt(111) NSA obtained in 0.1 M HClO4. The latter CV is characteristic for the Cu NSA as reported previously.23, 28, 29, 42 It should be noted that both hydrogen and hydroxyl adsorption regions are shifted towards more negative and more positive potentials, respectively, revealing the fact that the surface of Cu–Pt(111) NSA binds these adsorbates more weakly than the unmodified Pt(111) electrode. While integration of the CVs between 0.5 V and 1.0 V, performed after correction for the double-layer charging, gives similar OH coverages for both unmodified and modified electrodes, the maximal coverage for the adsorbed hydrogen in the case of NSA is only ∼1/3 ML. This is two times smaller than that estimated for the original Pt(111) surface.
Figure 2. CVs of Cu–Pt(111) NSA (—) in A) 0.1 M HClO4 and B) 0.05 M H2SO4. Dotted lines represent corresponding CVs for the unmodified Pt(111) single crystals for better comparison, dE/dt=50 mV s−1.
Download figure to PowerPoint
Cyclic voltammetry experiments disclose a more complex adsorption dynamics if the electrolyte is replaced with 0.05 M H2SO4. Figure 2 B shows CVs of the Pt(111) and Cu–Pt(111) NSA electrodes obtained in 0.05 M H2SO4.
It should be noted that the hydrogen adsorption region remains almost unchanged for the NSA in both perchloric and sulfuric acids, suggesting that this process remains practically unaffected by (bi)sulfates. In contrast, the butterfly peaks at approximately 0.5 V (RHE) completely disappeared. In fact, the shape of the voltammograms for the NSA, in comparison to the corresponding CV for Pt(111), suggests that specific adsorption of (bi)sulfates is shifted towards more positive potentials (as indicated in Figure 2 B). While the onset potentials for the adsorption are approximately the same (∼0.38 V) for both Pt(111) and Cu–Pt(111) NSA electrodes, comparable voltammetric currents and integrated anodic charges for (bi)sulfate electrosorption in the case of NSA can be only reached at approximately 1.0 V (RHE). It should be emphasized, that while we attribute the anodic peaks between ∼0.8 V and 1.0 V mainly to the adsorption of (bi)sulfates, some co-adsorption of *OH cannot be excluded and requires further detailed investigations to further clarify this.
Figure 3 A shows the results of voltammetric experiments made in 0.1 M HClO4 using the Pt(111) and Cu–Pt(111) SA, where ∼2/3 ML of Cu atoms are preferentially located in the first atomic layer of Pt. The voltammogram characterizing the surface alloy is only shown in the potential region of its stability. Surface Cu atoms are oxidized at more positive potentials. The CV of the SA shown in Figure 3 A reveals stronger adsorbate–substrate interactions in the electrode potential region between ∼0.2 V and ∼0.4 V as compared to the unmodified Pt(111) electrode.23 Notably, integration of the cyclic voltammogram for the Cu–Pt(111) surface alloy gives approximately the same charge as for the Pt(111) surface for the same region ∼160 μC cm−2. However, the presence of Cu atoms in the first atomic layer obviously leads to a noticeable redistribution of the adsorbate binding energies (Figure 3 A), in agreement with ref. 23.
Figure 3. CVs of the Cu–Pt(111) surface alloys (—) in A) 0.1 M HClO4 and B) 0.05 M H2SO4. Dotted lines show corresponding CVs for the unmodified Pt(111) surfaces for comparison, dE/dt=50 mV s−1.
Download figure to PowerPoint
Figure 3 B shows the results of CV experiments performed in 0.05 M H2SO4 using Pt(111) and Cu–Pt(111) SA electrodes. Comparing the shape of CVs, which are recorded in sulfuric (Figure 3 B) and perchloric acid (Figure 3 A) electrolytes, distinct changes can be observed. The peaks in the potential region between ∼0.2 V and ∼0.4 V observed in 0.1 M HClO4, has been considerably shifted (ca. 50 mV) towards more negative potentials in sulfate-containing electrolyte. Notably, integration of the CVs characterizing the Cu–Pt(111) SA in 0.05 M H2SO4 gives much smaller adsorption/desorption charges compared to those estimated for SA electrodes in contact with 0.1 M HClO4 (∼100 μC cm−2 vs. ∼160 μC cm−2, respectively). Obviously, voltammetric experiments reveal a significant effect of (bi)sulfate anions on the SA-electrode/electrolyte interface status. However, simple voltammetric measurements described above give only the integral current that is likely associated with simultaneous adsorption/desorption of hydrogen atoms and (bi)sulfate anions. In order to separate contributions originating from different adsorbates, electrochemical impedance spectroscopy (EIS) experiments have been also performed. The results of these EIS experiments are discussed in Section 2.2.
CVs recorded in both 0.1 M HClO4 and 0.05 M H2SO4 solutions for the Cu pseudomorphic overlayer deposited on the Pt(111) electrode are shown in Figure 4 A. In contrast to previous systems, the resulting CVs obtained in the potential region of the monolayer stability are featureless. Furthermore, the voltammograms are similar in both perchloric and sulfuric acid electrolytes. It has been previously shown in ref. 37 that specific adsorption/desorption of (bi)sulfates on well-ordered Cu(111) surfaces take place at more negative electrode potentials. However, scanning the electrode potential to a more cathodic region (data not shown) does not reveal any peaks or features which could be identified as related to sulfate adsorption/desorption in 0.05 M H2SO4. To illustrate a difference in the voltammetric behavior of the Cu overlayer and Cu(111) electrodes, Figure 4 B compares voltammograms for the Cu(111) electrode (taken from ref. 37) and Cu overlayer on Pt(111) in 5 mM H2SO4 (our data). Specific adsorption and desorption of (bi)sulfate anions at the Cu(111) surface manifest themselves by a pair of peaks in cathodic (at −0.9 V) and anodic (−0.55 V) directions (note that the potentials are reported in Figure 4 B versus the Hg/HgSO4 reference electrode to enable direct comparison with literature data). The former peak is assigned to (bi)sulfate desorption while the latter peak is due to (bi)sulfate adsorption. Surprisingly, the CV of the Cu pseudomorphic overlayer on Pt(111) does not reveal any characteristic features indicating possible adsorption of (bi)sulfates at the Cu overlayer. Experiments using electrochemical quartz crystal microbalance (see Section 2.3) further support the hypothesis that specific adsorption of (bi)sulfates is unlikely at the surface of the Cu overlayer.
Figure 4. A) CVs of the Cu pseudomorphic overlayer on Pt(111) in 0.1 M HClO4 (—) and 0.05 M H2SO4 (•–•–) in comparison with the voltammograms for the unmodified Pt(111) in the same electrolytes (- - - - and •••• lines), dE/dt=50 mV s−1. B) Voltammograms for the Cu overlayer on Pt(111) (—, this work) and Cu(111) single crystal in 5 mM H2SO4 (- - - - and ••••, as reported in ref. 37, dE/dt=10 mV s−1. Note that the potentials are given versus a Hg/HgSO4 reference electrode to compare with literature data.).
Download figure to PowerPoint
2.2. Electrochemical Impedance Spectroscopy Measurements
To investigate adsorption processes at Cu–Pt(111) surface alloy electrodes in more detail, EIS experiments have been performed in 0.05 M H2SO4. The advantage of EIS in characterization of the electrochemical interfaces is based on its ability to separate responses originating from adsorption of different species, which occur simultaneously.
A general physical model shown in Figure 5 A has been elucidated earlier.38, 39 It accurately describes the EIS response of Pt electrodes in contact with H2SO4 electrolytes.48 In Figure 5 A, Rs is the uncompensated resistance, Zdl is the impedance of the electric double layer, Rct and R′ct are the charge transfer resistances for the adsorption of hydrogen and (bi)sulfate species, respectively, Ca and C′a, are the adsorption capacitances for hydrogen and (bi)sulfate species, respectively. Using EIS, in these systems it is indeed possible to identify and separate the response associated with sulfate adsorption from other adsorption processes.38, 39, 48 The charge associated with (bi)sulfate adsorption can be derived from experimental C′a(E) curves available as a result of the EIS data analysis. Figure 5 B shows an example of EIS spectra taken at 0.2 V for Cu–Pt(111) surface alloy electrodes in 0.05 M H2SO4 and as a Nyquist plot. Additionally, the inset in Figure 5 B shows the dependence of the phase shift as a function of the probing frequency. Perfect fitting of all experimental impedance spectra to the model shown in Figure 5 A has been obtained. This additionally proves that adsorption/desorption of hydrogen is accompanied by specific adsorption and desorption of (bi)sulfates. Attempts to use more simplified models resulted in considerable increase in the root-mean-square deviations between the data and fitted curves as well as in estimated errors of individual parameters of the model. To simplify further discussion, only the data attributed to (bi)sulfate adsorption/desorption is discussed later (further detailed information about modeling, fitting, and subsequent data analysis related to impedance spectroscopy of the interface between Pt electrodes and H2SO4 electrolytes can be found in ref. 48).
Figure 5. A) The equivalent electric circuit describing the electrochemical interface between the Cu–Pt(111) surface alloy and 0.05 M H2SO4 electrolyte. Rs is the uncompensated resistance, Zdl is the impedance of the electric double layer, Rct and R′ct are the charge transfer resistances for the adsorption of hydrogen and (bi)sulfate species, respectively. Ca and C′a, are the adsorption capacitances for hydrogen and (bi)sulfate species, respectively; B) A typical example of the experimental impedance spectra (points) with fitting (solid line) at 0.2 V; C) the dependence of (bi)sulfate adsorption capacitance (C(bi)sulfate) on the electrode potential.
Download figure to PowerPoint
Figure 5 C shows the dependence of the adsorption capacitance for (bi)sulfates on the electrode potential. Integration of this curve gives the charge of approximately 18 μC cm−2. Assuming that one electron is transferred through the interface per one adsorbed anion, the total (bi)sulfate coverage is ∼0.08 ML. This is an approximately three times smaller coverage than that found for unmodified Pt. Taking into account that the total voltammetric charge that has been estimated earlier for SA in sulfuric acid is ∼100 μC cm−2, and assuming the rest of the charge is associated with the adsorption/desorption of hydrogen, the H coverage at the Cu–Pt(111) SA is ∼1/3 ML.
Summarizing the results of the EIS measurements, one can conclude that even though the surface of the SA binds hydrogen stronger than Pt(111) in 0.1 M HClO4, the presence of (bi)sulfate anions at the interface suppresses H adsorption in sulfuric acid. Presumably, this is due to an interplay between H and (bi)sulfate binding energies and stable surface structures at different electrode potentials.
2.3. Electrochemical Gravimetric Measurements
In order to further clarify issues related to the drastic differences in the CVs for Cu(111) and a Cu overlayer on Pt(111) in H2SO4, electrogravimetric measurements were performed using a quartz crystal microbalance. However, it should be noted that polycrystalline Pt was used in these measurements due to experimental complications in obtaining Pt(111) thin films at the surface of quartz resonators. Figure 6 A shows the electrode mass change in 0.1 M HClO4 before and after addition of Cu2+ ions at the electrode potential required to form the Cu overlayer (0.33 V).
Figure 6. Results of electrochemical gravimetric measurements. A) massogram characteristic for Cu UPD (at 0.33 V) on polycrystalline Pt electrode in 0.1 M HClO4 (the arrow indicates the time when 1 mL of 0.2 M Cu2+ in 0.1 M HClO4 has been added. B) Massogram illustrating adsorption of (bi)sulfates on an unmodified Pt surface and in absence of their specific adsorption on a Cu overlayer on Pt-electrode at 0.4 V vs. RHE; the arrow indicates time when 1 mL of 5.0 M H2SO4 has been added.
Download figure to PowerPoint
Increase in the electrode mass after addition of Cu2+ ions is associated with the formation of the copper overlayer. The resultant Δm values are in good agreement with the theoretically expected (∼160 ng cm−2) mass change, indicating no significant co-adsorption of heavy anions, such as perchlorates or chlorides. Figure 6 B compares two massograms for unmodified Pt and the Cu overlayer electrodes in 0.1 M HClO4 at the most positive potentials where the overlayer is still stable against anodic oxidation. From Figure 6 B it is clear that addition of (bi)sulfates into the electrolyte does not lead to a significant mass change of the Cu-overlayer electrode. In contrast, (bi)sulfates adsorb at the surface of the unmodified platinum electrode under the same conditions. Therefore, electrogravimetric experiments additionally support the voltammetric data presented in Figure 4, suggesting that no significant specific adsorption of (bi)sulfates takes place at the surface of the Cu overlayer. The origin of the difference in Cu-overlayer properties requires further investigations.