Article pubs.acs.org/cm
Electrodeposition of Crystalline Co3O4A Catalyst for the Oxygen Evolution Reaction Jakub A. Koza,‡ Zhen He,‡ Andrew S. Miller, and Jay A. Switzer* Department of Chemistry and Graduate Center for Materials Research, Missouri University of Science and Technology, Rolla, Missouri 65409-1170, United States S Supporting Information *
ABSTRACT: Crystalline films of Co3O4 are deposited by electrochemically oxidizing a tartrate complex of Co2+ in an aqueous, alkaline solution at elevated temperatures. The crystallinity and stability of the films are a strong function of the deposition temperature. Films deposited at temperatures from 50 to 90 °C are amorphous, but films deposited from refluxing solution at 103 °C are crystalline. The crystalline films adhere strongly to the substrate, whereas the amorphous films peel off of the substrate when dried due to drying stresses. The crystalline films deposit with the normal spinel structure, with a lattice parameter of 0.8097 nm and crystallite size of 26 nm. The catalytic activity of Co3O4 for the oxygen evolution reaction (OER) of the crystalline and amorphous films is compared by Tafel analysis in alkaline solution at pH 14. The crystalline Co3O4 film has a Tafel slope of 49 mV/decade and an exchange current density of 2.0 × 10−10 A cm−2, whereas an amorphous film deposited at 50 °C has a Tafel slope of 36 mV/decade and an exchange current density of 5.4 × 10−12 A cm−2. Because the films deposited from refluxing electrolyte deposit directly as crystalline films, it is possible to deposit them epitaxially on single-crystal Au(100). This opens up the possibility to study the catalytic activity of different Co3O4 planes exposed to the electrolyte. KEYWORDS: water electrolysis, oxygen evolution catalyst, cobalt oxide, Co3O4, electrodeposition deposit Co3O4 films have led to the formation of amorphous and/or hydrated layers.11,12 In our work, crystalline Co3O4 films are deposited galvanostatically from a solution of 5 mM Co(II) and 6 mM L-tartrate in 2 M NaOH at temperatures ranging from 50 to 103 °C (reflux). The method does not require heat treatment of the deposit, because the material deposited at 103 °C is crystalline. Moreover, the films can be grown epitaxially on a single-crystal Au(100) substrate. This opens up the possibility to study the catalytic activity toward the OER of different Co3O4 crystal planes exposed to the electrolyte.
1. INTRODUCTION The kinetic bottleneck of electrochemical and photoelectrochemical (PEC) water splitting is the oxygen evolution reaction (OER).1,2 OER catalysts include transition metal oxides such as RuO2, IrO2, PtO2, MnO2, and Co3O4.3 Although the most active of these catalysts are the expensive noble metal oxides RuO2, IrO2, and PtO2, there is interest in developing OER catalysts based on earth-abundant metals such as Co. Co3O4 is slightly less active than the noble metal oxides for water oxidation in alkaline solution.3 Kanan and Nocera have also shown that an amorphous cobalt-phosphate (CoPi) OER catalyst can be electrodeposited, which is stable in neutral solution.4 Here, we report that crystalline (Fd3m space group) spinel Co3O4 films can be produced by electrochemical oxidation of Co(II) complexed with tartrate (tart) at elevated temperatures. Co3O4 is reported to be a p-type semiconductor with a bandgap of about 2.4 eV.5−7 The bangap is larger than that of Si (1.1 eV), which makes it a promising candidate for silicon-based PEC water splitting applications. That is, it will catalyze the oxygen evolution reaction, but it will not significantly attenuate light absorption by Si. Co3O4 is also a good candidate for a PEC cell using other n-type semiconductors as photoanodes, such as TiO2, Fe2O3 or WO3. In contrast to Si, these oxides are stable in alkaline solution. Co3O4 is generally produced by thermal methods such as thermal decomposition, hydrothermal, or spray pyrolysis.3,8−10 To the best of the authors’ knowledge, attempts to electro© 2012 American Chemical Society
2. EXPERIMENTAL SECTION Deposition Solution. The electrolyte used for deposition of Co3O4 has the following composition: 5 mM Co2+ and 6 mM tartrate (tart) in 2 M NaOH. In order to prepare the electrolyte a Co2+-(tart) stock solution was prepared by dissolving an appropriate amount of Co(NO3)2.7H2O and L-tartaric acid in deionized (DI) water to form a 0.1 M Co2+ and 0.12 M tartrate solution. Finally, a proper volume of the stock solution was added to a NaOH solution to form the nominal composition. The tartaric acid was used to complex Co2+ ions to be soluble at pH 14, where according to the Pourbaix diagram Co3O4 is thermodynamically stable. Received: April 24, 2012 Revised: July 9, 2012 Published: August 24, 2012 3567
dx.doi.org/10.1021/cm3012205 | Chem. Mater. 2012, 24, 3567−3573
Chemistry of Materials
Article
Electrochemical Methods. All electrodeposition experiments were performed in a three-neck flask with an electrolyte volume of 200 mL in a standard three-electrode setup using a Princeton Applied Research EG&G 273A potentiostat. The films were deposited at temperatures from 50 °C to reflux. The temperature was maintained with a heating mantle and monitored by a thermometer. A condenser was attached to one of the necks of the flask to prevent electrolyte losses due to evaporation. The electrolyte prior to depositing was purged with Ar for about 60 min to prevent atmospheric oxidation of Co2+. An Ar atmosphere above the electrolyte was maintained during all experiments. A stainless steel type 430 (SS), Au rotating disk, and single-crystal Au(100) were used as the working electrodes. XRD, morphology studies, linear sweep voltammograms, Tafel plots, and electrochemical stability studies were done on stainless steel substrates. Cyclic voltammograms used to study the surface redox chemistry were done on a Au rotating disk electrode. Epitaxy studies were done on a Au(100) substrate. Current densities are reported with respect to the geometric area of the working electrodes. An ultrahigh purity pyrolytic graphite rod served as the counter electrode, and a Ag/AgCl/KClsat. reference electrode was used. The SS and Au substrates were mechanically polished, sonicated in acetone and rinsed with DI water prior to experiments. The Au(100) substrate was also electropolished prior to depositing according to the procedure described elsewhere.13 All films were deposited anodically by passing a constant charge of 0.9 C cm−2. The catalytic properties of the deposits were investigated in 1 M KOH solution at room temperature in a three-electrode cell using a Brinkmann PGSTAT 30 Autolab potentiostat. The deposit served as the working electrode and a Pt mesh was the counter electrode. The OER experiments were performed on fresh layers rinsed with DI water immediately after deposition. The overpotentials were calculated with respect to the thermodynamic potential of oxygen evolution at pH 14 (0.401 V vs NHE.) and corrected for the IR drop in the solution. The electrolyte resistance was determined with electrochemical impedance spectroscopy. The double layer capacitance of the films was measured to compare the roughness factors of the films by determining the scan rate dependence of the current in cyclic voltammograms that were run over a small potential range (±20 mV) in an acetonitrile solution with 0.1 M tetrabutylammonium hexafluorophosphate as the electrolyte. Characterization. The structure of the films was determined using a high-resolution Philips X-Pert MRD X-ray diffractometer (XRD) with a CuKα1 radiation source (λ = 1.54056 Å). The morphology of the films was studied by scanning electron microscopy (SEM, Hitachi S4700 FESEM) and optical microscopy. The thickness of the deposited films was determined with a profilometer (Sloan DEKTAK IIA).
Figure 1. Electrochemistry of the Co(II)(tart) electrolyte. (a) Linear sweep voltammograms measured at different electrolyte temperatures at a scan rate of 50 mV s−1 on a SS substrate. Potentials at each temperature were corrected for the temperature dependence of the potential of the reference electrode. (b) Thickness vs charge density dependencies obtained in a refluxing electrolyte (103 °C) at current densities of 0.1 (black) and 0.25 mA cm−2 (red). The dashed red line is a linear fit to the 0.25 mA cm−2 data.
the deposition mechanism proposed by Casella for hydrated cobalt oxide compounds.11,12 From the temperature dependence on the LSVs it is clear that by increasing the electrolyte temperature, the current density of the oxidation peak increases and the onset of the oxidation peak shifts toward lower potentials. This leads to a decrease in the potential at which films can be grown at a given current density. Figure 1b shows the dependence of the deposit thickness on the charge density passed during galvanostatic depositions from a refluxing electrolyte at 0.1 and 0.25 mA cm−2 current densities. A linearity obtained at 0.25 mA cm−2 can be seen, whereas the dependence obtained at 0.1 mA cm−2 shows a negative deviation from linearity. This suggests that at 0.1 mA cm−2 another parallel reaction consumes a significant portion of charge. The most probable explanation is the oxidation of the tartrate ion, because this reaction was observed in the cyclic voltammograms measured in the electrolyte without Co ions (see Figure S1 in the Supporting Information). An estimation of the current efficiency at 0.25 mA cm−2 based on the proposed deposition mechanism, the film thickness, and assuming the density of the deposit equal to that of bulk material (6.056 g cm−3), gave a value of roughly 100%,
3. RESULTS AND DISCUSSION The electrochemistry of the Co(II)tartrate electrolyte was studied by linear sweep voltammetry (LSV) and galvanostatic depositions. Figure 1a shows the LSVs at a stainless steel (SS) working electrode obtained at different temperatures. Potentials at each temperature were corrected for the temperature dependence of the potential of the reference electrode. As the potential increases, a wave in the LSV is observed, which corresponds to the Co(II)/Co(III) redox couple (eq 1) followed by a steep increase of the current due to the OER. The Co3O4 deposition is believed to proceed via an electrochemical-chemical (EC) route 2Co2 +(tart) ⇆ 2Co3 + + 2(tart) + 2e−
(1)
2Co3 + + Co2 +(tart) + 8OH− → Co3O4 + 4H 2O + (tart)
(2)
The deposition scheme is analogous to that proposed by Kothari et al. for Fe3O4 deposition14 and is in agreement with 3568
dx.doi.org/10.1021/cm3012205 | Chem. Mater. 2012, 24, 3567−3573
Chemistry of Materials
Article
regardless of the deposition time. In the case of the deposition at 0.1 mA cm−2 the efficiency decays from 100% at the beginning to about 60% after 2.5 h (0.9 Ccm−2) of the process. Figure 2 shows optical micrographs of the films deposited at 0.25 mA cm−2 and different temperatures. The layers deposited
Figure 2. Morphology of the films electrodeposited on stainless steel. (a−d) Optical images of the film deposited at 50 °C showing the peeling of the film from the substrate upon drying. (a) As-deposited film with a water film on it. The film was dried in air at room temperature; the time passed between (b) and (d) was 2.5 min. (e, f) Optical microscope images of the films deposited at (e) 75 and (f) 103 °C. Same scale bar for e and f. All films were deposited on a SS substrate at 0.25 mA cm−2 by passing a charge density of 0.9 C cm−2. Figure 3. XRD analysis of the deposited films. (a) XRD patterns of the films deposited at different conditions as indicated. Dependence of the Co3O4 crystallite size on (b) the electrolyte temperature at 0.25 mA cm−2 and on (c) the current density at 103 °C. All films were deposited on a stainless steel substrate by passing a charge density of 0.9 C cm−2.
at low temperatures adhere very poorly. Films peel off from the substrate upon drying, as shown for the film deposited at 50 °C in Figure 2a−d, which represent a time sequence. We attribute this peeling to drying stresses for the amorphous films. Hence, for studies of catalytic activity, the films deposited at 50 °C were maintained in a wet state. The adhesion of the film deposited at 75 °C (Figure 2e) improved significantly, but the film has a cracked-mud morphology. A similar observation was made by Kanan and Nocera for electrodeposited CoPi films.4 Spataru et al. also reported that electrodeposited, amorphous/ hydrated Co3O4 films displayed a very nonuniform morphology. The films appeared discontinuous with isolated particles and a preferential deposition at the edges.15 However, the film deposited from a refluxing electrolyte in our work is very adherent, with a smooth and crack-free surface (Figure 2f). The structure of the deposit was determined by X-ray diffraction (XRD). Figure 3a shows XRD patterns of the layers deposited at 0.25 mA cm−2 and temperatures of 50 (black curve) and 103 °C (refluxing electrolyte − blue curve). The XRD pattern of the film deposited at 50 °C displays the SS substrate peaks and very broad and low intensity peaks, which can be assigned as Co3O4. This indicates an amorphous nature of the film. The pattern is similar to the one published by Nakaoka et al. for an electrodeposited Co3O4 film.16 They observed only a single broad and low intensity (311) peak. Similar results were also published on Co3O4 electrodeposition by other researchers. It was reported that as-deposited Co3O4 films were amorphous and/or hydrated11,12,15,17 and required heat treatment to crystallize.17−19 However, the film deposited in our work from a refluxing electrolyte (103 °C) is crystalline, with well-resolved peaks in the XRD pattern corresponding to Co3O4 (a = 0.8097 nm). The crystallinity of the electrodeposited Co3O4 was a strong function of the deposition temperature and current density. The crystallite size was estimated from the line broadening of the (311) peak using the Scherrer equation. As shown in Figure
3b, the crystallite size increases as the temperature is increased, with a dramatic increase in size at reflux temperatures. The crystallite size increases from a value of about 2 nm at 50 °C to about 26 nm at 103 °C. Hence, even though the film deposited at reflux temperatures is crystalline according to X-ray diffraction, the crystallite size is still in the nanometer range. For a series of films grown at reflux temperature, the crystallite size decreases as the current density is increased (see Figure 3c). These results are consistent with the EC mechanism proposed in eq 1 and 2. Although the rate of the electrochemical reaction (eq 1) is constant at a constant current density, the rate of the chemical reaction (eq 2) would be faster at higher temperatures. Hence, at higher temperatures and lower currents, there would be lower supersaturation of Co(III) at the electrode surface, leading to more crystallographically ordered films. The electrodeposition from a refluxing electrolyte can be considered to be an electro-hydrothermal process. 20−24 Yoshimura and co-workers introduced the electro-hydrothermal method for BaTiO3 synthesis in 1997.20 The deposition of BaTiO3 employs an anodic dissolution of the working electrode, which serves as a source of one of the metal ions. In our case, Co3O4 is deposited from solution precursors, which gives a greater degree of freedom regarding the choice of the substrate. Preliminary investigations have shown, for example, that Co3O4 can be deposited onto p+-Si (see Figure S2 in the Supporting Information). The catalytic properties of Co3O4 films deposited at 50 and 103 °C for the OER were studied in 1 M KOH at room temperature by Tafel analysis. The films were tested using a 3569
dx.doi.org/10.1021/cm3012205 | Chem. Mater. 2012, 24, 3567−3573
Chemistry of Materials
Article
lower than that reported for CoPi (60 mV/decade). The films operating in neutral solutions may have advantages over the films operating in base with respect to silicon-based PEC applications. However, there are several reports in which Si was stabilized to work in alkaline solutions, such as an electrodeposited Tl2O3 film on n-Si28 or a tunnel oxide layer of TiO2 on n-Si produced by atomic layer deposition.29The Tafel plots in Figure 4 are based on the geometric area of the stainless steel substrate. To accurately compare the exchange current densities of electrocatalysts, it is necessary to determine the electrochemically active area. We attempted to measure the electrochemically active area by several methods. We measured linear sweep voltammograms with a rotating disk electrode using reversible, outer-sphere redox couples in order to compare the mass-transport-limited currents of the redox couples on the Co3O4 electrodes to those on a planar Au electrode. Attempts with 1,4-benzoquinone in aqueous solution and ferrocene in acetonitrile were unsuccessful, because the Co3O4 electrode was electrochemically inert for these redox couples. Another method is to measure the double layer capacitance of the electrode.30−32 This measurement is not possible in aqueous solution with Co3O4 because of a large pseudocapitance of the material due to intercalation of cations into the oxide.30−32 To eliminate pseudocapaciatance due to cation intercalation, we measured the double layer capacitance in an acetonitrile solution with 0.1 M tetrabutylammonium hexafluorophosphate as the electrolyte. The CVs used to measure the double layer capacitance are shown in Figure 5a for the amorphous film deposited at 50 °C and in Figure 5b for the crystalline film deposited at 103 °C. Scan rates of 50, 100, 150, 200, 250, 300, and 400 mV s−1 were used for each film. The specific double layer capacitances determined from the plot of current density versus scan rate in Figure 5c are 13 μF cm−2 for the crystalline film deposited at 103 °C and 12 μF cm−2 for the amorphous film deposited at 50 °C. Because the double layer capacitance of smooth (roughness factor = 1) Co3O4 is not known, it is not possible to determine the true electrochemically active area of the two films. However, because the double layer capacitances of the two films are very similar, the roughness factors of the films should also be similar. Hence, the ratio of the exchange current densities of 2.0 × 10−10 and 5.4 × 10−12 A cm−2 for the crystalline and amorphous films of 37 is approximately correct. This method should provide a means to accurately measure the roughness factor of a Co3O4 electrode if a reference sample with a known roughness factor could be obtained. The stability of the Co3O4 films on stainless steel substrates was studied in 1 M KOH at 10 and 100 mA cm−2 for 48 h (Figure 6). The overpotential vs time curves at an applied current density of 10 mA cm−2 do not show significant changes over time for either film. The curves measured at 100 mA cm−2 show an increase in overpotential, with the overpotential of the crystalline film increasing more than that of the film deposited at 50 °C. We note, however, that the amorphous film deposited at 50 °C needed to be maintained in the wet state to prevent it from peeling from the stainless steel substrate (see Figure 2a− d). The stability of the crystalline film electrodeposited in this study was similar to films deposited via chemical spray pyrolysis that were previously reported by Singh et al. during long time anodic polarization at 25 mA cm−2.33,34 XRD studies of the electrodeposited crystalline film after the stability study did not show any change of the crystal structure. SEM investigations also did not show any significant change in the film morphology. This suggests that the films prepared by our
stainless steel substrate in an unstirred solution. Note that the Tafel plot for the amorphous film deposited at 50 °C had to be run on a film that was not allowed to dry, to prevent it from peeling off of the stainless steel substrate. Figure 4 shows the
Figure 4. Steady-state Tafel plots measured in 1 M KOH for films deposited at 50 °C (black line and open squares) and 103 °C (red line and open circles), with a fit to the linear portion of the curves (dashed lines). Literature data of Co3O4 (solid triangles) and CoPi (asterisk) are plotted for comparison.4,8 Co3O4 was deposited at 0.25 mA cm−2 on a stainless steel substrate by passing a charge density of 0.9 C cm−2. The overpotential was calculated with respect to the thermodynamic potential at pH 14 of 0.401 V vs NHE. The data were corrected for the IR drop in the solution.
steady state polarization curve measured on a film deposited at 0.25 mA cm−2 from a refluxing electrolyte (red line and open circles) and an amorphous film deposited at 50 °C (black line and open squares). The data was corrected for the IR drop in the solution, but was not corrected for the IR drop in the film. The current densities were higher for the amorphous film at low overpotentials and current densities, but the crystalline film was more active at current densities higher than about 40 mA cm−2. We attribute this crossover in apparent activity at 40 mA cm−2 to the higher resistance of the amorphous film. The crystalline Co3O4 film has a Tafel slope of 49 mV/decade and an exchange current density of 2.0 × 10−10 A cm−2, whereas the amorphous film deposited at 50 °C has a Tafel slope of 36 mV/ decade and an exchange current density of 5.4 × 10−12 A cm−2. The Tafel slopes provide information on the reaction mechanism, and the exchange current densities measure the catalytic activity of the films at zero overpotential. The Tafel slopes of 49 and 36 mV/decade observed for the crystalline and amorphous films are in the 40 mV/decade range that is consistent with a mechanism involving a pre-equilibrium consisting of a one-electron electrochemical step and a possible chemical step followed by a one-electron electrochemical rate determining step.25−27 The electrodeposited crystalline film is comparable in activity to that of Co3O4 deposited by spray pyrolysis on Ni (Tafel slope of 51 mV/decade, wine triangles in Figure 4).8 In addition, the data for a CoPi film in neutral solution is shown in Figure 4. The catalytic activity of Co3O4 is higher than that of CoPi. At an overpotential of 0.41 V the current density is 1 mA cm−2 on CoPi and is approximately 17 mA cm−2 on the crystalline Co3O4 film. Also, the Tafel slope obtained on the crystalline Co3O4 film (49 mV/decade) is 3570
dx.doi.org/10.1021/cm3012205 | Chem. Mater. 2012, 24, 3567−3573
Chemistry of Materials
Article
Figure 6. Plot of the overpotential as a function of time at current densities of 10 mA cm−2 (black) and 100 mA cm−2 (red) measured in 1 M KOH for films deposited at 50 °C (dashed lines) and 103 °C (full lines). Co3O4 deposited at 0.25 mA cm−2 on a stainless steel substrate by passing a charge density of 0.9 C cm−2. The overpotential was calculated with respect to the thermodynamic potential at pH 14 of 0.401 V vs NHE. The data were corrected for the IR drop in the solution.
Figure 7c were run with rotation rates of 0 and 250 rpm. The linear dependence of the current density on the scan rate shown in the insets of panels a and b in Figure 7 is consistent with a surface confined redox couple. This is also verified by the independence of the CVs on the rotation rate in Figure 7c. Notice, however, that the current densities observed are on the order of tens of mA/cm2, much higher than expected for a simple surface confined redox process (typically about 0.1 mA/ cm2). These results suggest that either the redox reaction extends into the bulk of the film or that the surface roughness of the films is high. The crystalline and amorphous films both show two anodic peaks (A1 and A2) and three cathodic peaks (C1, C2, and C3) in panels a and b in Figure 7. The anodic peak A1 is more clearly defined in the amorphous film than in the crystalline film, and the anodic peaks A1 and A2 are both higher in current density for the amorphous film. These results suggest that there are more active sites on the amorphous film, because the electrochemically active area of the two films is very similar (based on the double layer capacitance measurements). Other workers12,15,31,35 have assigned the A1 and A2 peaks to the oxidation of Co(II) to Co(III) and the oxidation of Co(III) to Co(IV), respectively. The assumed reactions for the oxidations occurring at peaks A1 and A2 are shown in eqs 3 and 4. The oxygen evolution reaction occurs at potentials slightly positive of peak A2 for both films, suggesting that Co(IV) is the active catalyst for the OER. The 40 mV/decade Tafel slope observed for both films is consistent with the oxidation of CoOOH to CoO2 being the rate-determining step.
Figure 5. CVs of the Co3O4 films deposited on stainless steel substrates at (a) 50 °C and (b) 103 °C, measured in 0.1 M tetrabutylammonium hexafluorophosphate in acetonitrile at scan rates from 50 to 400 mVs−1. (c) Plots of the current density at 0 V vs the scan rate used to determine the specific capacitance of the films.
method may be suitable for use in PEC devices by fulfilling the long-term stability requirement. The catalytic activity of the Co3O4 films has been attributed to the surface redox activity of the Co(II) and Co(III) in the films.8,12,15,16,27,31,35 This surface reactivity can be probed by running cyclic voltammograms (CVs) of the films. Figure 7 shows CVs in 1 M KOH of an amorphous film deposited at 50 °C (Figure 7a) and a crystalline film deposited at 103 °C (Figure 7b). The CVs were run on an Au rotating disk electrode. The CVs in panels a and b in Figure 7 were run in a quiescent solution at a series of scan rates, and the CVs in
Co3O4 + H 2O + OH− ⇆ 3CoOOH + e− −
CoOOH + OH ⇆ CoO2 + H 2O + e
−
(3) (4)
Because Co3O4 deposits directly at reflux temperatures as a crystalline film without any subsequent heat treatment, it is possible to electrodeposit epitaxial films of Co3O4 onto singlecrystal Au. Figure 8a shows the XRD pattern of the Co3O4 film deposited on a Au(100) single-crystal substrate. The film grows 3571
dx.doi.org/10.1021/cm3012205 | Chem. Mater. 2012, 24, 3567−3573
Chemistry of Materials
Article
Figure 8. XRD evidence of the epitaxial growth of Co3O4 on a Au(100) substrate. (a) X-ray diffraction pattern of the film. (b) (311) pole figure. The arrows in b show the substrate orientation.
study the catalytic activity toward the OER of different planes exposed to the electrolyte. Both the in- and out-of-plane orientations can be controlled by carefully choosing the substrate and deposition parameters as shown for other electrodeposited oxide films.36−38 It is known that different atomic planes have not only different atomic densities but also differ with respect to their electronic structures.39 Xie et al. have shown that the {110} planes in Co3O4 have the highest catalytic activity toward CO oxidation at low temperature due to its highest density of Co3+ ions.40 Hu et al. have also shown that the {112} planes in Co3O4 are more reactive in the catalytic combustion of methane than the {001} and {011} planes.9 It is reasonable to assume that different planes of Co3O4 will differ with respect to the OER catalytic activity. This should improve the fundamental understanding of the processes involved in the OER at Co3O4.
Figure 7. CVs of the Co3O4 films deposited on a Au rotating disk electrode at (a) 50 °C and (b) 103 °C measured in 1 M KOH quiescent solution at different scan rates. The insets in a and b show the linear dependence of the A1 and A2 peak current densities on the scan rate. (c) Comparison of the CVs measured for films deposited at 50 and 103 °C in quiescent solution (open points) and at 250 rpm rotation rate (solid lines).
4. CONCLUSIONS We have shown that films of spinel Co3O4 can be deposited by electrochemical oxidation of Co(II)tartrate in alkaline solution at elevated temperatures (reflux). The as-deposited films are fully crystalline and do not require any further heat treatment. The films deposited from refluxing electrolyte grow with the normal spinel structure, with a lattice parameter of 0.8097 nm and crystallite size of 26 nm. The electrolyte temperature is the most critical parameter controlling the crystallinity and stability of the film. A dramatic increase of the crystallite size was observed at reflux temperatures. The crystalline films deposited at 103 °C are very smooth, crack-free, and adherent, whereas amorphous films deposited at lower temperatures peel off of the substrate after drying. The crystalline Co3O4 film has a Tafel slope of 49 mV/decade and an exchange current density of 2.0 × 10−10 A cm−2, whereas an amorphous film deposited at
with a preferred {111} out-of-plane orientation. Only a very small intensity peak of the (311) reflection can be seen. A (311) pole figure was measured (Figure 8b), which shows distinct peaks revealing that the film is epitaxial. The pole figure was compared to the stereographic projections (see Figure S3 in the Supporting Information) and the epitaxial relationship was found to be Co3O4(111)[112]̅ ||Au(100)⟨001⟩. The low intensity peak in the center of the pole figure corresponds to the (311) plane observed in the XRD pattern (Figure 8a). The epitaxial growth of Co3O4 films opens up the possibility to 3572
dx.doi.org/10.1021/cm3012205 | Chem. Mater. 2012, 24, 3567−3573
Chemistry of Materials
Article
50 °C has a Tafel slope of 36 mV/decade and an exchange current density of 5.4 × 10−12 A cm−2. The Tafel slopes are consistent with a mechanism involving a pre-equilibrium consisting of a one-electron electrochemical step (Co(II) → Co(III)) and a possible chemical step followed by a oneelectron electrochemical rate determining step (Co(III) → (Co(IV)). Both the crystalline and amorphous electrodeposited Co3O4 are more active catalysts than CoPi for the OER. We have also shown that the films can be electrodeposited epitaxially on single-crystal Au(100). The film grows with a preferred {111} out-of-plane orientation on Au(100). The epitaxial relationship is Co3O4(111)[112̅]||Au(100)⟨001⟩. The epitaxial growth of Co3O4 opens up the possibility to study the catalytic activity for the OER of different crystal planes exposed to the electrolyte. In future work, we plan to electrodeposit the crystalline Co3O4 OER catalyst onto n-type semiconductor photoanodes such as n-Si, TiO2, Fe2O3, and WO3 for photoelectrochemical water splitting.
■
(16) Nakaoka, K.; Nakayama, M.; Ogura, K. J. Electrochem. Soc. 2002, 149, C159. (17) Wu, J. B.; Lin, Y.; Xia, X. H.; Xu, J. Y.; Shi, Q. Y. Electrochim. Acta 2011, 56, 7163. (18) Yuan, Y. F.; Xia, X. H.; Wu, J. B.; Huang, X. H.; Pei, Y. B.; Yang, J. L.; Guo, S. Y. Electrochem. Commun. 2011, 13, 1123. (19) Xia, X. H.; Tu, J. P.; Zhang, J.; Xiang, J. Y.; Wang, X. L.; Zhao, X. B. ACS Appl. Mater. Interfaces 2009, 2, 186. (20) Yoo, S. E.; Yoshimura, M.; So̅miya, S. J. Mater. Sci. Lett. 1989, 8, 530. (21) Yoshimura, M.; Yoo, S.-E.; Hayashi, M.; Ishizawa, N. Jpn. J. Appl. Phys. 1989, 28, L2007. (22) Yoshimura, M.; Suchanek, W. Solid State Ionics 1997, 98, 197. (23) Suchanek, W. L.; Watanabe, T.; Sakurai, B.; Kumagai, N.; Yoshimura, M. Rev. Sci. Instrum. 1999, 70, 2432. (24) Song, S.-W.; Han, K.-S.; Yoshimura, M. J. Am. Ceram. Soc. 2000, 83, 2839. (25) Castro, E. B.; Gervasi, C. A. Int. J. Hydrogen Energy 2000, 25, 1163. (26) Gerken, J. B.; McAlpin, J. G.; Chen, J. Y. C.; Rigsby, M. L.; Casey, W. H.; Britt, R. D.; Stahl, S. S. J. Am. Chem. Soc. 2011, 133, 14431. (27) Castro, E. B.; Gervasi, C. A.; Vilche, J. R. J. Appl. Electrochem. 1998, 28, 835. (28) Switzer, J. A. J. Electrochem. Soc. 1986, 133, 722. (29) Chen, Y. W.; Prange, J. D.; Dünen, S.; Park, Y.; Gunji, M.; Chidsey, C. E. D.; McIntyre, P. C. Nat. Mater. 2011, 10, 539. (30) Trasatti, S.; Petrii, O. A. Pure Appl. Chem. 1991, 63, 711. (31) Boggio, R.; Carugati, A.; Trasatti, S. J. Appl. Electrochem. 1987, 17, 828. (32) Minguzzi, A.; Fan, F.-R. F.; Vertova, A.; Rondinini, S.; Bard, A. J. Chem. Sci 2012, 3, 217. (33) Singh, R. N.; Hamdani, M.; Koenig, J. F.; Poillerat, G.; Gautier, J. L.; Chartier, P. J. Appl. Electrochem. 1990, 20, 442. (34) Singh, R. N.; Koenig, J. F.; Poillerat, G.; Chartier, P. J. Electrochem. Soc. 1990, 137, 1408. (35) Casella, I. G. J. Electroanal. Chem. 2002, 520, 119. (36) Switzer, J. A.; Shumsky, M. G.; Bohannan, E. W. Science 1999, 284, 293. (37) Liu, R.; Vertegel, A. A.; Bohannan, E. W.; Sorenson, T. A.; Switzer, J. A. Chem. Mater. 2001, 13, 508. (38) Sorenson, T. A.; Morton, S. A.; Waddill, G. D.; Switzer, J. A. J. Am. Chem. Soc. 2002, 124, 7604. (39) Xu, X.-L.; Chen, Z.-H.; Li, Y.; Chen, W.-K.; Li, J.-Q. Surf. Sci. 2009, 603, 653. (40) Xie, X.; Li, Y.; Liu, Z.-Q.; Haruta, M.; Shen, W. Nature 2009, 458, 746.
ASSOCIATED CONTENT
S Supporting Information *
XRD pattern and Raman spectrum of the film deposited on p+Si, stereographic projections, LSVs of the substrates and films deposited on SS and Ti, and Tafel plots of the crystalline Co3O4 film with and without IR correction. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Author Contributions ‡
These authors contributed equally
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by the U.S. Department Of Energy, Office of Basic Sciences, under Grant DE-FG02-08ER46518.
■
REFERENCES
■
(1) Walter, M. G.; Warren, E. L.; McKone, J. R.; Boettcher, S. W.; Mi, Q.; Santori, E. A.; Lewis, N. S. Chem. Rev. 2010, 110, 6446. (2) Zhong, D. K.; Gamelin, D. R. J. Am. Chem. Soc. 2010, 132, 4202. (3) Trasatti, S. Electrochim. Acta 1984, 29, 1503. (4) Kanan, M. W.; Nocera, D. G. Science 2008, 321, 1072. (5) Petitto, S. C.; Langell, M. A. J. Vac. Sci. Technol. A 2004, 22, 1690. (6) Long, M.; Cai, W.; Cai, J.; Zhou, B.; Chai, X.; Wu, Y. J. Phys. Chem. B 2006, 110, 20211. (7) Kandalkar, S. G.; Gunjakar, J. L.; Lokhande, C. D.; Joo, O.-S. J. Alloys Compd. 2009, 478, 594. (8) Singh, S. P.; Samuel, S.; Tiwari, S. K.; Singh, R. N. Int. J. Hydrogen Energ. 1996, 21, 171. (9) Hu, L.; Peng, Q.; Li, Y. J. Am. Chem. Soc. 2008, 130, 16136. (10) Wendt, H.; Hofmann, H.; Plzak, V. Mater. Chem. Phys. 1989, 22, 27. (11) Innocenzo, G, C. J. Electroanal. Chem. 2002, 520, 119. (12) Casella, I. G.; Di Fonzo, D. A. Electrochim. Acta 2011, 56, 7536. (13) Boonsalee, S.; Gudavarthy, R. V.; Bohannan, E. W.; Switzer, J. A. Chem. Mater. 2008, 20, 5737. (14) Kothari, H. M.; Kulp, E. A.; Limmer, S. J.; Poizot, P.; Bohannan, E. W.; Switzer, J. A. J. Mater. Res. 2006, 21, 293. (15) Spataru, N.; Terashima, C.; Tokuhiro, K.; Sutanto, I.; Tryk, D. A.; Park, S.-M.; Fujishima, A. J. Electrochem. Soc. 2003, 150, E337.
NOTE ADDED AFTER ASAP PUBLICATION This article was published ASAP on September 4, 2012, with errors in eqs 1, 3, and 4, and a minor text error in the Experimental Section. The corrected version was published ASAP on September 7, 2012.
3573
dx.doi.org/10.1021/cm3012205 | Chem. Mater. 2012, 24, 3567−3573
