Cuprous oxide template synthesis of hollow-cubic Cu₂O@Pd_xRu_y nanoparticles for ethanol electrooxidation in alkaline media

Abstract

Surfactant-free and low Pd loading Cu₂O@Pd_xRu_y hollow-cubes were facilely prepared via a galvanic replacement reaction with electrodeposited Cu₂O cubes as sacrificed templates. The morphology and composition of the as-synthesized nanoparticles were characterized, through which the Pd : Ru atomic ratio, metal loading, surface composition and structure of several Cu₂O@Pd_xRu_y nanoparticles were determined. The promotional role of Ru on electrooxidation of ethanol in alkaline media over Cu₂O@Pd_xRu_y was investigated. By comparison, the following electrocatalytic activity order of the catalysts was obtained: Pd/C < Cu₂O@Pd < Cu₂O@Pd₁Ru₃ < Cu₂O@Pd_3.5Ru₁ < Cu₂O@Pd_2.1Ru₁ < Cu₂O@Pd₁Ru_1.6 < Cu₂O@Pd₁Ru₁. Combined with the stability study, the hollow-cubic Cu₂O@Pd₁Ru₁ nanoparticles show superior catalytic activity and stability toward electrooxidation of ethanol in alkaline media compared to the other catalysts studied in this work which is attributed to the appropriate Pd : Ru ratio and the hollow-cubic structure.

---

1. Introduction

As one of the promising alternative clean and renewable energy sources, direct ethanol fuel cells (DEFCs) attract significant attention mainly due to the high theoretical energy density, huge sustainable production source (ethanol can be produced from agricultural products and biomass), easy storage, facile transportation and low toxicity of ethanol.¹ However, the application and commercialization of DEFCs are mainly limited by the high price-performance ratios of anode catalysts which are involved in the electrocatalytic oxidation of ethanol.² It is because of the fact that efficient electrocatalysts for ethanol oxidation reaction (EOR) commonly have high Pt or Pd contents which are expensive and prone to be deactivated by CO poison.³

It is reported that operating DEFCs in alkaline media has several advantages over in acid environment, such as the oxygen reduction reaction kinetics are commonly higher in alkaline medium when using Pt catalysts and the low corrosiveness of high pH environment allows the utilization of non-noble metals as/in electrocatalysts.^4–6 Especially, in alkaline medium, both the electrocatalytic activity and stability of Pd towards electrooxidation of ethanol are remarkably higher than that of Pt.⁴ Since Pd is much cheaper and more abundant than Pt, therefore, Pd-based catalysts have been emerged as promising candidates for the anode materials in alkaline DEFCs.^4–7 Various Pd-based materials have been reported as electrocatalysts for ethanol oxidation under alkaline condition, such as Pd nanoparticles,⁸ Pd nanowire arrays⁹ and a huge diversity of Pd-based nanomaterials (e.g. PdAu, PdCu, PdSn, PdNi, PdAg/graphene oxide),^10,11 improvements in catalytic performance for ethanol oxidation have been achieved as a result of the special structures and/or components of the catalysts. Ruthenium is known as a high-performance co-catalyst of Pt for electrooxidation of alcohols (especially methanol), due to the bi-functional mechanism¹² and electronic effect.¹³ Therefore, efforts have been paid to prepare Ru-containing catalysts to explore the promotion effect of Ru for electrooxidation of alcohols.^14,15 Recently, Pd–Ru composite catalysts toward alcohols electrooxidation were reported by several groups and considerable improvements in catalytic performance were observed.^5,16–29 However, most of the reported PdRu composite catalysts were synthesized via chemical reactions in which the shape of the nanoparticles are difficult to control and the nanoparticles are easy to aggregate. Besides, organic solvents (e.g. ethylene glycol) or surfactants (PVP) are usually employed in the synthesis which may produce nanoparticles with contaminated surfaces. Galvanic replacement reaction (GRR) became an attractive method to prepare nanostructures due to the merits such as facile to control composition, shape and size of the resultant nanostructures, and the surfaces of the nanostructures are clean. What is even more important is that novel nanostructures (usually hollow structures) with low noble metal loading could be obtained by GRR when using non-noble metal nanostructures as sacrificed templates.^30,31

In this work, low Pd loading hollow-cubic Cu₂O@Pd_xRu_y nanoparticles with various Pd : Ru atomic ratio were prepared via GRR with electrodeposited Cu₂O cubes as templates. The as-prepared Cu₂O@Pd₁Ru₁ hollow-cubes (Cu₂O@Pd₁Ru₁ HCs) show superior catalytic activity for electrooxidation of ethanol in alkaline media. This method is simple, easy to handle and, especially, the fabrication procedure is surfactant free, which will facilitate further application of the as-prepared nanoparticles in electrooxidation of ethanol.

2. Experimental

2.1 Materials

Copper(II) trifluoroacetate (Cu(TFA)₂·xH₂O), potassium trifluoroacetate (KTFA, 98%), potassium tetrachloropalladate(II) (K₂PdCl₄, 99.99%), ruthenium(III) chloride (RuCl₃·xH₂O, 99.98%) and Nafion (5 wt%, Sigma-Aldrich) were purchased from Sigma-Aldrich. Pd/C (20 wt%, JM) is purchased from Shanghai Hesen Electric Co., Ltd., China. All chemical reagents were used directly without any purification. All solutions were prepared with Milli-Q ultrapure water (Millipore, ≥18.2 MΩ cm).

2.2 Preparation of Cu₂O nanocubes and Cu₂O@Pd_xRu_y

Cu₂O nanocubes were prepared by a modified method reported by Guo et al.³² The cubic Cu₂O nanoparticles were electrodeposited on Ti sheets (1 cm × 1 cm) using chronoamperometry at −0.06 V (vs. SCE) for 3600 seconds in Cu(TFA)₂ (5 mM) and KTFA (0.1 M) solution. The Ti sheets were polished and cleaned according to the method reported in ref. 33 before the Cu₂O electrodeposition. Cu₂O@Pd_xRu_y nanoparticles were prepared by immersing Cu₂O cubes into 2.6 mL 2 mM K₂PdCl₄ + a mM RuCl₃ (a = 0.5, 1, 2, 3, 4, 6, 8) solution for 9 hours at ∼277 K, respectively. Cu₂O@Ru hollow-cubes (HCs) were prepared by immersing Cu₂O cubes into 2.6 mL 2 mM RuCl₃ solution for 9 hours at ∼277 K. For comparison, a longer time GRR of Cu₂O cubes in 2 mM K₂PdCl₄ or 2 mM RuCl₃ solution were carried out for 12 hours at ∼277 K, respectively. Throughout this paper the GRR time is 9 hours, unless indicated otherwise. According to the ref. 31 and 33, the galvanic replacement reaction on Cu₂O can be expressed as the following equations:

Cu₂O + PdCl₄²⁻ + 2H⁺ → Pd⁰ + 2Cu²⁺ + H₂O + 4Cl⁻ (1)

3Cu₂O + 2Ru³⁺ + 6H⁺ → 2Ru⁰ + 6Cu²⁺ + 3H₂O (2)

2.3 Characterization of the catalysts

The morphologies of the as-prepared nanomaterials were characterized by scanning electron microscopy (SEM) using a JEOL 6701F equipped with an energy dispersive X-ray (EDX) detector system. The compositions of the products were determined by a Rigaku Ultima IV X-ray diffractometer (XRD) with Cu Kα radiation (λ = 1.5406 Å) and a Kratos Axis Ultra DLD X-ray photoelectron spectrometer (XPS) with a pass energy of 40 eV and a monochromated Al-Kα X-ray source (1486.6 eV) at 150 W. The binding energies with ±0.1 eV accuracy were determined with respect to the position of the adventitious C 1s peak at 284.8 eV. The Pd 3d, Ru 3p, Cu 2p and O 1s XPS spectra were deconvoluted via XPS PEAK 4.1 software using a mixed Gaussian–Lorentzian fitting with a Shirely background subtraction over the fitting energy range.³⁵ Cu, Pd and Ru contents in catalysts were determined by a PerkinElmer Optima 2100 DV ICP optical emission spectrometer (ICP-OES). The catalyst loading was determined as follows: firstly, the mass of Cu₂O was determined by m = (Q/F)(M/n), where Q is the quantity of electricity recorded by electrochemical workstation during electrodeposition, F is the Faraday constant (96 485 C mol⁻¹), M is the molar weight, and n is the stoichiometric number of electrons consumed in the electrode reaction; then, Pd (or Ru) loading was evaluated by determining the Pd (or Ru) loss before and after the GRRs. The Pd (or Ru) loss during the GRRs was evaluated by determining the concentration difference of PdCl₄²⁻ (or Ru³⁺) before and after the GRRs by ICP-OES.

2.4 Electrochemical measurements

Electrochemical experiments were performed with an Autolab Electrochemical Workstation (PGSTAT30, Metrohm Autolab, The Netherlands). A conventional three-electrode cell with SCE as reference electrode, a Pt sheet (1 cm × 1 cm) as counter electrode and the modified Ti sheet (1 cm × 1 cm) as working electrode, respectively. Commercial Pd/C catalyst solution was dropped on Ti sheet (1 cm × 1 cm) and dried at room temperature, Pd loading was 30 μg cm⁻². All the working electrodes were coated with 20 μL 0.5% Nafion and dried at room temperature before electrochemical experiments.

3. Results and discussion

The catalysts loading and elemental compositions of the Cu₂O@Pd_xRu_y resultants prepared via GRR were determined by ICP-OES and summarized in Table 1. The catalysts are denoted as Cu₂O@Pd_xRu_y where x and y are the Pd : Ru atomic ratios calculated from ICP-OES results. Interestingly, although the standard redox potential of Ru³⁺/Ru (0.25 V) is much negative than PdCl₄²⁻/Pd (0.59 V),³⁴ the Pd loading is still decreasing with increasing Ru content in GRR precursor. It was also confirmed by SEM investigation that it is easier for GRR of Cu₂O with Ru³⁺ than with PdCl₄²⁻ (details will be discussed later).

Table 1 The catalysts loading and elemental compositions of the Cu₂O@Pd_xRu_y

Catalyst	Catalyst loading on Ti (mg cm^−2)	Pd loading (wt%)	Ru loading (wt%)	Cu loading (wt%)	Pd : Ru : Cu (atomic ratio)	GRR precursor Pd + Ru
Cu2O@Pd	0.261	30	0	62	1 : 0 : 3.5	2 mM + 0 mM
Cu2O@Pd3.5Ru1	0.285	43	12	46	3.5 : 1 : 5.5	2 mM + 0.5 mM
Cu2O@Pd2.1Ru1	0.259	38	17	39	2.1 : 1 : 3.7	2 mM + 0.7 mM
Cu2O@Pd1Ru1	0.272	22	21	51	1 : 1 : 3.9	2 mM + 1 mM
Cu2O@Pd1Ru1.6	0.248	15	23	55	1 : 1.6 : 6.1	2 mM + 2 mM
Cu2O@Pd1Ru3	0.223	13	35	46	1 : 3 : 6.2	2 mM + 3 mM
Cu2O@Ru	0.261	0	28	64	0 : 1 : 3.6	0 mM + 2 mM

---

The morphologies of as-prepared Cu₂O nanocubes and Cu₂O@Pd_xRu_y nanoparticles were characterized by SEM shown in Fig. 1. The Cu₂O nanocubes electrodeposited on Ti sheet are uniform and have well-defined shapes with average edge length ca. 1 μm (Fig. 1A). An appropriate reaction time and precursor concentration of GRR on Cu₂O with Ru³⁺, PdCl₄²⁻, and the mixture solution of Ru³⁺ and PdCl₄²⁻ will produce hollow-cubic Cu₂O@Pd_xRu_y nanostructures. Here, several typical hollow-cubic Cu₂O@Pd_xRu_y nanostructures obtained by a 9 hours GRR are showed in Fig. 1B–D. The wall thicknesses of Cu₂O@Ru (Fig. 1B), Cu₂O@Pd (Fig. 1C) and Cu₂O@Pd₁Ru₁ (Fig. 1D) are 15–20, 32–38 and 12–18 nm, respectively. The Cu₂O@Pd_xRu_y hollow-cubic structures turn collapse with increasing Ru³⁺ contents in GRR precursor, e.g. the Cu₂O@Pd₁Ru_1.6 nanostructures are collapsed exhibited in Fig. 1E. Besides, a longer time (12 h) GRR on Cu₂O with PdCl₄²⁻ or Ru³⁺ will also produce broken structures (Fig. 1F and G), while a shorter time GRR cannot generate hollow-cubic structures or generate hollow-cubes with very thick wall, both of the longer and shorter times GRR resultants exhibit lower catalytic activity toward ethanol electrooxidation in experiments and will not be further discussed below.

Fig. 1 SEM images of (A) Cu₂O, (B) Cu₂O@Ru HCs, (C) Cu₂O@Pd HCs, (D) Cu₂O@Pd₁Ru₁ HCs, (E) Cu₂O@Pd₁Ru_1.6, (F) Cu₂O@Ru (12 h), (G) Cu₂O@Pd (12 h) and (H) Cu₂O@Pd₁Ru₁ HCs after 300 cycles CV between −0.8 to 0.2 V (vs. SCE) at 50 mV s⁻¹ in 1.0 M C₂H₅OH + 1.0 M KOH solution. Inset in (D) are element mappings of an individual Cu₂O@Pd₁Ru₁ HCs.

To confirm the crystal structure of the nanomaterials, XRD patterns of the as-prepared nanoparticles deposited on carbon paper were recorded in Fig. 2. The diffraction peaks at 2θ = 36.46, 42.36, 61.44 and 73.55° are ascribed to the (111), (200), (220), and (311) of cubic Cu₂O (JCPDS 78-2076), respectively. Two weak peaks at 2θ = 43.41 and 50.61° are ascribed to cubic Cu (JCPDS 85-1326) which are impurities co-electrodeposited in Cu₂O. The diffraction peaks corresponding to Cu₂O disappear in XRD pattern of Cu₂O@Ru (Fig. 2B), indicating most of the Cu₂O were replaced by Ru. However, no obvious diffraction features of Ru were observed in XRD patterns of Cu₂O@Ru (Fig. 2B) and Cu₂O@Pd₁Ru₁ NCs (Fig. 2D) which could be ascribed to the amorphous Ru structure in samples. Very weak diffraction peaks at 2θ = 40.18 and 46.75° are assigned to the (111) and (200) of cubic Pd (JCPDS 87-06399) in XRD pattern of Cu₂O@Pd₁Ru₁ NCs. The 2θ position of Pd(111) peaks for Cu₂O@Pd₁Ru₁ NCs are very close to Cu₂O@Pd NCs (Fig. 2C), indicating that Pd and Ru do not form an alloy.

Fig. 2 XRD patterns for (A) Cu₂O, (B) Cu₂O@Ru, (C) Cu₂O@Pd and (D) Cu₂O@Pd₁Ru₁.

XPS spectra were depicted in Fig. 3 to analyze the surface composition of the as-prepared Cu₂O@Pd₁Ru₁ NCs. XPS spectrum of Pd 3d region in Fig. 3A shows an intense doublet centering at 335.2 and 340.5 eV which are assigned to Pd 3d_5/2 and Pd 3d_3/2, respectively. The deconvolution of Pd 3d_5/2 peak reveals the existence of Pd⁰ (335.2 eV) and a small fraction of PdO (336.1 eV) and PdO₂ (337.5 eV).^5,35 Since the Ru 3d peaks overlap with C 1s peak (carbon sources from environment), XPS spectrum of Ru 3p were recorded and showed in Fig. 3B. The Ti 2p_3/2 peak is determined at 458.5 eV with a spin orbital splitting (SOS) of 5.7 eV which are ascribed to TiO₂ (substrate). The deconvolution of Ru 3p_3/2 peak with an SOS of 22.2 eV shows the existence of peaks at around 462.4 (Ru⁰), 463.3 (Ru₂O) and 464.8 eV (RuCl₃), which are assigned to ruthenium atoms in different chemical environments. The Ru 3p_3/2 peak of higher oxidation state (etc. RuO₃) and satellite peaks of Ru 3p_3/2 were not deconvolved, due to the difficulty of the binding energy determination.³⁶ Therefore, the fitting of Ru 3p peaks cannot match the experimental spectrum very well. From the peak intensity, one can see that Ru atoms in Cu₂O@Pd₁Ru₁ NCs are dominated by Ru₂O. According to the XPS results, no significant shift of the peaks of metallic Pd⁰ and Ru⁰ was observed which may indicate Pd and Ru do not form an alloy in Cu₂O@Pd_xRu_y NCs. This result is in line with XRD patterns. The Cu 2p_3/2 core level spectrum in Fig. 3C was used to evaluate the surface Cu oxidation state. Although, it is hard to resolve Cu and Cu₂O by this deconvolution due to the very close binding energy, no Cu diffraction peaks for the XRD pattern of Cu₂O@Pd₁Ru₁ NCs were observed. Herein, the fitting peak at 932.5 eV is assigned to Cu₂O, while the peak at 934.5 eV is ascribed to CuO (generated by GRR) which is confirmed by the satellite peak at 942.5 eV. Fig. 3D shows the XPS spectra of O 1s, which reveals two peaks at 530.2 and 531.5 eV corresponding to Cu oxides (Cu₂O/CuO) and Ti–OH on the Ti substrate, respectively.³⁷

Fig. 3 XPS spectra of Cu₂O@Pd₁Ru₁ in region of (A) Pd 3d, (B) Ru 3p, (C) Cu 2p_3/2 and (D) O 1s.

The catalytic performance of Cu₂O@Pd_xRu_y was studied by electrochemistry. CV curves of the as-prepared Cu₂O@Pd_xRu_y, Cu₂O/Ti and commercial Pd/C investigated in 1.0 M KOH solution from −0.9 to 0.3 V were displayed in Fig. 4. No obvious redox peaks were observed for the CV curves of Cu₂O/Ti and Cu₂O@Ru (Fig. 4B), while the voltammograms of synthesized Cu₂O@Pd_xRu_y have similar features but are various in current densities. Typically, a pair of hydrogen adsorption and desorption peaks could be observed in the potential region from −0.9 to −0.65 V for all catalysts but commercial Pd/C.³⁵ In anodic sweeps, weak peaks in the potential region from −0.6 to −0.4 V which are generally ascribed to the adsorption of hydroxyl group on catalysts surface could be observed for Cu₂O@Pd_xRu_y. It indicates the promoter effect of Ru on hydroxyl group adsorption which will increase the catalysts tolerance to carbon monoxide.^4,38 The anodic peaks in region from −0.2 to 0 V corresponding to the formation of PdO could be observed for the as-prepared catalysts.^35,39 The positive anodic peaks higher than 0 V could be ascribed to the formation of higher valence oxides, RuO₂, because they can only be observed in CVs of high Ru content catalysts i.e. Cu₂O@Pd₁Ru_1.6 and Cu₂O@Pd₁Ru₃. In cathodic scan, the strong peaks situated between −0.35 and −0.46 V are attributed to the reduction of PdO species, while the shoulder at ca. −0.15 V in CVs of Cu₂O@Pd₁Ru_1.6 and Cu₂O@Pd₁Ru₃ are ascribed to the reduction peaks of RuO₂.⁴⁰ The more positive shift of the reduction potential of Pd oxide, the weaker of the chemisorption of oxygen species and reduced oxophilicity which will be facile for the reproduce of fresh active sites.²² The electrochemical active surface area (ECSA) of the catalysts was evaluated according to ref. 39 based on the reduction coulombic charge of PdO with assumption of 0.43 mC cm⁻² for monolayer PdO reduction. The ECSAs are 177, 281, 304, 386, 235, 195 and 41 cm² mg_Pd⁻¹ for Cu₂O@Pd, Cu₂O@Pd_3.5Ru₁, Cu₂O@Pd_2.1Ru₁, Cu₂O@Pd₁Ru₁, Cu₂O@Pd₁Ru_1.6, Cu₂O@Pd₁Ru₃ and Pd/C, respectively. The ECSAs of the catalysts containing Ru are larger than that of Cu₂O@Pd confirmed the syngeneic effect of Pd and Ru, and the ECSA of Cu₂O@Pd₁Ru₁ (386 cm² mg_Pd⁻¹) is larger than that of Pd/Ru reported in ref. 22 which may be ascribed to the special structure of the catalyst, i.e. the hollow cubic structure could facilitate the mass diffusion during electrocatalysis.^41,42 It is also comparable with the PdRu/RGO (reduced graphene oxide) catalyst reported by Kim's group.¹⁸

Fig. 4 Cyclic voltammograms of (A) Cu₂O@Pd_xRu_y and (B) Cu₂O/Ti, Cu₂O@Pd, Cu₂O@Ru, Cu₂O@Pd₁Ru₁ and commercial Pd/C in 1.0 M KOH solution at a scan rate of 50 mV s⁻¹. The current densities of Cu₂O/Ti and Cu₂O@Ru are 10 times enlarged.

The catalytic performance of Cu₂O@Pd_xRu_y HCs towards ethanol oxidation was studied in 1.0 M ethanol + 1.0 M KOH solution by electrochemistry. The CV curves shown in Fig. 5A reveal the Cu₂O@Pd₁Ru₁ catalyst holds the highest forward anodic scan peak current density (j_f, ∼993 mA mg_Pd⁻¹), while the Cu₂O@Pd₁Ru_1.6 possesses the highest backward scan peak current density (j_b, ∼1573 mA mg_Pd⁻¹). The forward to backward current density ratio, j_f/j_b, decreases with the increasing of Ru content in Cu₂O@Pd_xRu_y and touch the bottom while the catalyst is Cu₂O@Pd₁Ru_1.6. It is well known that the forward anodic peak is ascribed to the oxidation of ethanol absorbed on the catalyst surface, while the backward scan peak is corresponding to the removal of the incomplete oxidized carbonaceous products generated during the forward anodic scan.^4,5 Therefore, the increase of reverse scan current density confirms the promotion effect of Ru in Cu₂O@Pd_xRu_y, i.e. the introduction of Ru will facilitate the generation of clean and active surface sites and accordingly will enhance electroactivities and durabilities of Cu₂O@Pd_xRu_y HCs.³³ However, Pd still dominates the electroactivities of the catalysts, because both the forward and backward peak current densities become very low for Cu₂O@Pd₁Ru₃ with ultra-low Pd loading. By evaluating the forward scan peak current density, the catalysts show an electrocatalytic activity of Pd/C (110 mA mg_Pd⁻¹) < Cu₂O@Pd (345 mA mg_Pd⁻¹) < Cu₂O@Pd₁Ru₃ (345 mA mg_Pd⁻¹) < Cu₂O@Pd_3.5Ru₁ (664 mA mg_Pd⁻¹) < Cu₂O@Pd_2.1Ru₁ (772 mA mg_Pd⁻¹) < Cu₂O@Pd₁Ru_1.6 (871 mA mg_Pd⁻¹) < Cu₂O@Pd₁Ru₁ (993 mA mg_Pd⁻¹). The electrocatalytic activity of Cu₂O@Pd₁Ru₁ towards ethanol oxidation is better than the reported results e.g. Pd/Ru,²² Pd_xRu_1−x/C⁵ and PdRu/RGO¹⁸ nanocomposites. These results are consistent with the ECSAs. For comparison, CV curves of Cu₂O/Ti, Cu₂O@Pd, Cu₂O@Ru, Cu₂O@Pd₁Ru₁ and commercial Pd/C, in 1.0 M ethanol + 1.0 M KOH solution, were plotted in Fig. 5B, from which no obvious redox peaks were observed for the CV curves of Cu₂O/Ti and Cu₂O@Ru.

Fig. 5 (A) CV curves of Cu₂O@Pd_xRu_y and Pd/C recorded at the 100th cycle of each sample. (B) CV curves of Cu₂O/Ti, Cu₂O@Pd, Cu₂O@Ru, Cu₂O@Pd₁Ru₁ and commercial Pd/C recorded at the 100th cycle of each sample. (C) CV curves of Cu₂O@Pd₁Ru₁ from first to 300th cycle. (D) Plots of forward and backward scan peak current densities of Cu₂O@Pd₁Ru₁ (blue), Cu₂O@Pd₁Ru₃ (red), Cu₂O@Pd (green) and Pd/C (black) vs. CV cycle numbers (solid: forward; open: backward). (E) Chronoamperometry curves of Cu₂O@Pd_xRu_y and Pd/C. All experiments are measured in 1.0 M C₂H₅OH + 1.0 M KOH solution and at a scan rate of 50 mV s⁻¹ for CV.

To study the stability of the catalysts, the 1^st to 300th cycle of CV curves of the catalysts were investigated. Fig. 5C shows the CV curves of Cu₂O@Pd₁Ru₁ HCs with increasing cycle number and Fig. 5D shows the corresponding plots of peak current densities vs. cycle numbers of Cu₂O@Pd₁Ru₁, Cu₂O@Pd₁Ru₃, Cu₂O@Pd and Pd/C, respectively (the others are not shown). Both the forward and backward peak current densities increase with cycle number before reaching a maximum value (within 110th cycle for all catalysts), this could be attributed to the exist of metal oxides on the surface of the catalysts, and the metal oxides will be reduced or removed during the CV scans. From Fig. 5D, it is found that the forward and backward scan peak current densities of Cu₂O@Pd₁Ru₁ HCs, after 300 cycles, remain 94% and 91% of their maximum values, respectively, indicating high stability of Cu₂O@Pd₁Ru₁ HCs. The correspondence values are 91% and 86% for Cu₂O@Pd₁Ru₃, 46% and 31% for Cu₂O@Pd, and 54% and 46% for Pd/C, respectively. These results show the introduction of Ru in Cu₂O@Pd_xRu_y will improve the stability of the catalysts. This may be due to the fact that the OH_ad species which will facile transform poison species (e.g. CO) on the catalyst surface to CO₂ or other non-poison products could easier generated on the surface of Ru or RuO_x. The morphology of Cu₂O@Pd₁Ru₁ HCs after 300 cycles of CV scans were checked and a small parts of the Cu₂O@Pd₁Ru₁ HCs were founded collapsed down (see SEM in Fig. 1H) after many CV cycles, this will also result in a slight decreasing in peak current density.

To further evaluate the stability and electrocatalytic activity of the catalysts toward ethanol electrooxidation, chronoamperometric experiments recorded at −0.30 V of the as-prepared Cu₂O@Pd_xRu_y nanoparticles and commercial Pd/C were carried out in 1.0 M ethanol + 1.0 M KOH solution (Fig. 5E). One can find that the current density of Cu₂O@Pd₁Ru₁ HCs is higher than that of the other catalysts, further proving the as-prepared Cu₂O@Pd₁Ru₁ catalysts possess superior electrocatalytic activity and electrochemical stability for ethanol electrooxidation in alkaline medium. This result is in agreement with the above CV experiments.

4. Conclusions

In summary, surfactant-free and low Pd loading hollow-cubic Cu₂O@Pd_xRu_y nanoparticles were prepared with Cu₂O cubes as template via a simple GRR and, the catalytic activity and stability of the as-prepared Cu₂O@Pd_xRu_y toward electrooxidation of ethanol in alkaline media were investigated and compared with commercial Pd/C catalyst. By comparison, the hollow-cubic Cu₂O@Pd₁Ru₁ nanoparticles show superior catalytic activity and stability toward electrooxidation of ethanol in alkaline media than the other catalysts studied in this work, owing to the appropriate Pd : Ru ratio and high surface area of the hollow-cubic structures. The electrocatalytic activity order of the catalysts are: Pd/C < Cu₂O@Pd < Cu₂O@Pd₁Ru₃ < Cu₂O@Pd_3.5Ru₁ < Cu₂O@Pd_2.1Ru₁ < Cu₂O@Pd₁Ru_1.6 < Cu₂O@Pd₁Ru₁. These results provide supports for (1) Ru is a promising co-catalyst for high performance electrocatalysts toward ethanol oxidation under alkaline medium; and (2) considering the facile synthetic process, low noble metal loading and surfactant free of the catalysts surface, the GRR of noble metal ions with non-noble metal structures will play an important role in the synthesis of low price and high performance catalysts, especially for multinary/hybrid catalysts.

Acknowledgements

The authors acknowledge the financial support from the National Natural Science Foundation of China (No. 21505019, 21375016 and 21475022), the Natural Science Foundations of Guangdong Province (2015A030310272).

References

1. M. Z. F. Kamarudin, S. K. Kamarudin, M. S. Masdar and W. R. W. Daud, Int. J. Hydrogen Energy, 2013, 38, 9438–9453.
2. B. Braunchweiga, D. Hibbittsb, M. Neurockb and A. Wieckowskia, Catal. Today, 2013, 202, 197–209.
3. H. Mistry, A. S. Varela, S. Kühl, P. Strasser and B. R. Cuenya, Nat. Rev. Mater., 2016, 1, 16009.
4. E. Antonlini and E. R. Gonzalez, J. Power Sources, 2010, 195, 3431–3450.
5. E. A. Monyoncho, S. Ntais, F. Soares, T. K. Woo and E. A. Baranova, J. Power Sources, 2015, 287, 139–149.
6. M. A. F. Akhairi and S. K. Kamarudin, Int. J. Hydrogen Energy, 2016, 41, 4214–4228.
7. H. Wang, C. W. Xu, F. L. Cheng, M. Zhang, S. Y. Wang and S. P. Jiang, Electrochem. Commun., 2008, 10, 1575–1578.
8. N. Tian, Z. Y. Zhou, N. F. Yu, L. Y. Wang and S. G. Sun, J. Am. Chem. Soc., 2010, 132, 7580–7581.
9. Y. Zhang, M. Zhang, Z. Q. Cai, M. Q. Chen and F. L. Cheng, Electrochim. Acta, 2012, 68, 172–177.
10. W. Hong, C. S. Shang, J. Wang and E. K. Wang, Electrochem. Commun., 2014, 48, 65–68.
11. A. C. Chen and C. Ostrom, Chem. Rev., 2015, 115, 11999–12044.
12. M. Watanabe and S. Motoo, J. Electroanal. Chem. Interfacial Electrochem., 1975, 60, 267–273.
13. J. A. R. Van Veen, T. Frelink and W. Visscher, Surf. Sci., 1995, 335, 353–360.
14. A. O. Neto, R. R. Dias, M. M. Tusi, M. Lindardi and E. V. Spinacé, J. Power Sources, 2007, 166, 87–91.
15. M. C. Figueiredo, O. Sorsa, R. M. Arán-Ais, N. Doan, J. M. Feliu and T. Kallio, J. Catal., 2015, 329, 69–77.
16. T. F. Li, Y. Y. Huang, K. Ding, P. Wu, S. C. Abbas, M. A. Ghausi, T. Zhang and Y. B. Wang, Electrochim. Acta, 2016, 191, 940–945.
17. W. L. Li, Y. Y. Huang, D. Tang, T. Zhang and Y. B. Wang, Electrochim. Acta, 2015, 174, 178–184.
18. E. J. Lim, Y. Kim, S. M. Choi, S. Lee, Y. Noh and W. B. Kim, J. Mater. Chem. A, 2015, 3, 5491–5500.
19. P. Y. Bi, W. Hong, C. S. Shang, J. Wang and E. K. Wang, RSC Adv., 2016, 6, 12486–12490.
20. S. Kabir, A. Serov, K. Artyushkova and P. Atanassov, Electrochim. Acta, 2016, 203, 144–153.
21. K. Eid, H. J. Wang, V. Malgras, Z. A. Alothman, Y. Yamauchi and L. Wang, J. Phys. Chem. C, 2015, 119, 19947–19953.
22. K. Zhang, D. Bin, B. B. Yang, C. Q. Wang, F. F. Ren and Y. K. Du, Nanoscale, 2015, 7, 12445–12451.
23. Z. W. Yang, X. Zhao, T. J. Li, W. L. Chen, Q. X. Kang, X. Q. Xu, X. X. Liang, Y. Feng, H. H. Duan and Z. Q. Lei, Catal. Commun., 2015, 65, 34–40.
24. D. S. Wu, M. N. Cao, M. Shen and R. Cao, ChemCatChem, 2014, 6, 1731–1736.
25. L. Ma, H. He, A. Hsu and R. R. Chen, J. Power Sources, 2013, 241, 696–702.
26. K. Hubkowska, M. Łukaszewski, U. Koss and A. Czerwiński, Electrochim. Acta, 2014, 132, 214–222.
27. I. Shimada, Y. Oshima and J. Otomo, J. Chem. Eng. Jpn., 2014, 47, 514–520.
28. D. Dang, H. L. Gao, L. J. Peng, Y. L. Su, S. J. Liao and Y. Wang, Acta Phys.-Chim. Sin., 2011, 27, 2379–2384.
29. J. Bagchi and S. K. Bhattacharya, Transition Met. Chem., 2007, 32, 47–55.
30. X. H. Xia, Y. Wang, A. Ruditskiy and Y. N. Xia, Adv. Mater., 2013, 25, 6313–6333.
31. M. D. Susman, R. Popovitz-Biro, A. Vaskevich and I. Rubinstein, Small, 2015, 11, 3942–3953.
32. X. Y. Guo, W. Q. Lv and X. Y. Li, J. Phys. Chem. C, 2014, 118, 11062–11077.
33. S. H. Ye, X. J. He, L. X. Ding, Z. W. Pan, Y. X. Tong, M. M. Wu and G. R. Li, Chem. Commun., 2014, 50, 12337.
34. W. M. Haynes, D. R. Lide and T. J. Bruno, CRC Handbook of Chemistry and Physics, CRC Press, Boca Raton, FL, 95th edn, 2014.
35. E. A. Monyoncho, S. Ntais, N. Brazeau, J. J. Wu, C. L. Sun and E. A. Baranova, ChemElectroChem, 2016, 3, 218–227.
36. D. J. Morgan, Surf. Interface Anal., 2015, 47, 1072–1079.
37. S. P. Chenakin, G. Melaet, R. Szukiewicz and N. Kruse, J. Catal., 2014, 312, 1–11.
38. V. M. Schmidt and R. Lanniello, J. Phys. Chem., 1996, 100, 17901–17908.
39. Y. Feng, D. Bin, K. Zhang, F. F. Ren, J. Wang and Y. K. Du, RSC Adv., 2016, 6, 19314–19321.
40. B. D. McNicol and R. T. Short, J. Electroanal. Chem. Interfacial Electrochem., 1978, 92, 115–120.
41. Y. Fan, Y. Zhang, H. M. Li, W. M. Shen, J. L. Wang and M. M. Wei, RSC Adv., 2016, 6, 43980–43984.
42. Y. Fan, Y. Zhang, P. F. Liu, Y. Cui, X. R. Ge and Q. S. Jing, Int. J. Hydrogen Energy, 2016, 41, 6773–6780.

---