Asymmetric Au-core Pd-shell nanoparticles supported on reduced graphene oxide for enhanced electrocatalytic activity

Abstract

Reduced graphene oxide (rGO)-supported asymmetric Au-core Pd-shell bimetallic nanoparticles (AuPd/rGO), i.e. Pd²⁺ is reduced only on the exposed Au(0) nanoparticle (GN) surface, were prepared using a simple two-step synthetic approach. First, negatively charged ∼15 nm GNs were prepared and attached to the amine-functionalized positive surface of a glass slide. Second, the substrate of the adsorbed Au-core nanoparticles was added to the Pd precursor solutions (0.5, 5, and 10 mM of PdCl₂, respectively), resulting in the deposition of a thin asymmetric Pd layer on the surface of the GNs via a tailored galvanic replacement reaction (GRR). This led to more sophisticated structures such as asymmetric core–shell nanoparticles, while avoiding monometallic nanoparticle formation. The compositional/structural features were characterized by high-resolution transmission electron microscopy, scanning TEM, UV-Vis spectroscopy, X-ray diffraction, and X-ray photoelectron spectroscopy. The catalytic activity of AuPd/rGO was investigated by rotating disk electrode voltammetry in 0.1 M NaOH. In particular, AuPd/rGO with a Au : Pd wt% ratio of 0.61 : 0.39 showed superior oxygen reduction reaction (ORR) activity along with satisfactory stability under alkaline conditions.

---

Introduction

Fuel cells are generally viewed as one of the most promising energy conversion systems in the next generation of sustainability.^1–3 On the other hand, the high cost of fuel cells is one of the major challenges hindering their commercialization because expensive Pt or Pt-based catalysts are currently used in these electrochemical devices to catalyze the cathodic oxygen reduction reaction (ORR). In addition, the CO poisoning effect on the carbon-supported Pt catalysts during fuel oxidation and the limited supply of Pt as well as the requirement of a high Pt loading have impeded the widespread applications of fuel cells.^4–6 Pt-free nanoparticles, however, have attracted considerable attention owing to their good electrocatalytic activity with low-cost and stability.⁷ Being a more economic and plentiful metal than Pt, Pd has been intensively investigated over the last decade due to its electrocatalytic activity for the ORR, because it has similar properties to Pt, such as being in the same group in the periodic table, having the same fcc crystal structure, and a similar atomic size.⁸ Several studies have recently reported the cathodic ORR activities on Pd nanocatalysts. For example, the ORR activity of Pd nanorods prepared by Xiao et al. was 10 times higher than that of conventional nanoparticles under acidic conditions.⁹ Shao et al. reported that the ORR activities of Pd catalysts are strongly dependent on their structure and the electrolyte used.¹⁰ More recently, Tammeveski et al. reported that in alkaline media, the specific activity of carbon-supported Pd nanocubes, 7, 10, and 30 nm in size, is more than two times higher than that of spherical Pd nanoparticles.¹¹ Although metallic Pd is an alternative electrocatalyst that is less expensive than Pt, the current performance with Pd alone is inadequate. Therefore, extensive research efforts have been devoted towards the development of bimetallic Pd-based nanomaterials for the ORR at the cathode of fuel cells.

The surface/structure of Au-based nanomaterials was also reported to influence their electrocatalytic properties toward the ORR.^12–14 Catalysts based on pure Au, however, still show lower activity, particularly for the ORR compared to Pt-based ones because of the weak chemisorption properties of Au with the filled d-band.^15–17 Considering the great similarity between Pt and Pd, a class of Au catalysts coated or alloyed with Pd instead of Pt has been investigated. Indeed, bimetallic nanoparticles with unique core–shell structure are expected to show enhanced catalytic activity, because the complex electron interaction between the two electron-rich elements and the lattice strain produced in these core–shell particles could modify the surface electronic properties of the nanoparticles.^18–30 For example, the core–shell Au@Pd nanoparticles via their core–shell Au@Ag/Pd templates display superior activity and durability in catalyzing the ORR, due mainly to the larger lattice tensile effect in the Pd shell induced by the Au core and Ag removal.³⁰ In addition, different methods for the synthesis of Au–Pd core–shell nanoparticles have been reported, such as one-pot coreduction,^26,31,32 two-step seed-mediated growth method,³³ electrolysis of the bulk,³⁴ sonochemical,³⁵ chemical reduction,^36–38 and a galvanic replacement reaction (GRR).³⁹ These synthesis methods can be divided into (1) successive, in which one starts with a seed nanoparticle of one composition and grows a shell around it, and (2) simultaneous, in which the ions of both metals are reduced simultaneously.^40,41 In Au–Pd systems, a variety of syntheses that use either a simultaneous or a successive approach have been reported. In general, the successive approach enables greater control over the shell thickness and the morphology of the nanoparticle product. In one example, Lee's group synthesized core–shell nanoparticles of Au–Pd successfully by adding a PdCl₂ solution to Au nanoparticles (GNs) preformed via GRR-like fabrication.³⁹ The core–shell nanoparticles of Au–Pd were generated by a successive approach and the surface plasmon resonance (SPR) peaks of the GNs were extinguished after the addition PdCl₂. On the other hand, bimetallic Au–Pd core–shell nanoparticles have two main disadvantages in electrocatalytic reactions. First, the capping agents on the surface of the nanoparticles may block mass transport and electron transfer, which seriously impair their electroactivity. Second, monometallic Pd (and/or Au), which is inevitably produced by the synthesis process and lowers the yield of the final core–shell product, has a much lower intrinsic ORR catalytic activity than that of bimetallic Au–Pd due to its unsuitable d-band center, which may weaken the adsorption of oxygen on the catalyst surface, resulting in insufficient performance in catalyzing oxygen reduction.^42,43 Indeed, the turnover rate for the reaction between dihydrogen and dioxygen to form water is considerably higher over AuPd than over the pure Pd catalyst.

This study examined the ORR activity of bimetallic AuPd nanoparticles supported on reduced graphene oxide (rGO) under 0.1 M NaOH. AuPd nanoparticles, which are densely anchored to the rGO nanosheets, were synthesized asymmetrically with controlled Pd distributions via a facile GRR-like fabrication method without the use of surfactants or reducing agents. The morphological and electrochemical properties of the rGO assembled with AuPd nanoparticles were characterized. The results showed that AuPd/rGO with a Au : Pd wt% ratio of 0.61 : 0.39 exhibited superior ORR activity and satisfactory stability.

Experimental

Reagents

All chemicals and solvents used for the synthesis of core–shell nanoparticles were purchased and used as received. Gold(III) chloride hydrate (HAuCl₄·H₂O), (3-aminopropyl)trimethoxy silane (APTMS, C₆H₁₇NO₃Si), 5.0 wt% Nafion, and sodium citrate tribasic dehydrate (Na₃C₆H₅O₇) were purchased from Aldrich. The RBS detergent solution (35 concentrate) was obtained from Fluka. Palladium(II) chloride (PdCl₂) and graphite flake were obtained from Alfa Aesar. Commercial carbon-supported Pd and Pt catalysts (Pd-20/C and Pt-20/C, respectively, with a 20 wt% metal loading on Vulcan XC-72) were purchased from E-TEK. The solutions used in this study were prepared freshly using Milli-Q water (resistivity ≥ 18 MΩ cm). All glassware was cleaned using an aqua regia solution (3 : 1 HCl–HNO₃), followed by a treatment in a base bath (saturated KOH in isopropyl alcohol), and rinsing with Milli-Q water prior to use.

Synthesis of nanoparticles

GNs with a diameter of ∼15 nm were synthesized using a similar method to that described previously.⁴⁴ First, 20 mL of a 1.0 mM HAuCl₄ aqueous solution was heated to boiling, and 2 mL of 38.8 mM citrate solution was added with vigorous stirring. The color of the solution changed gradually to a light yellow, purple, and then a deep red color. When it became a deep red color, stirring was stopped, and the solution was kept in the dark prior to use.

Synthesis of graphene oxide

Graphene oxide (GO) was synthesized using a slight modification of the Hummers method.⁴⁵ Briefly, 5 g of graphite (purity degree ≥ 99.5%) was mixed with 115 mL of H₂SO₄ and 2.5 g NaNO₃ in an ice bath using the three-neck 250 mL flask for one and a half hours. Subsequently, 15 g of KMnO₄ was added slowly to the mixture within 1 h to ensure that the temperature of the mixture did not exceed 5 °C, and was kept stirring for 1 h. The mixture was heated to 35 °C and kept stirring for 2 h. Water (230 mL) was then added and external heating was introduced to maintain the reaction temperature at 98 °C for 15 min. Finally, the oxidation reaction was terminated by the addition of 700 mL water and 50 mL of a 30% H₂O₂ solution. The resulting product in a suspension was washed thoroughly with distilled water and a 5% HCl solution. The GOs were dried completely in a vacuum for 12 h and kept in a refrigerator prior to use.

Preparation of asymmetric core–shell nanoparticles

A glass slide (25 mm × 12 mm, Marienfeld, Germany) was cleaned by sonication in a 15% RBS detergent solution at 90 °C for 5 min, washed repeatedly with deionized water, and then immersed in a 1 : 1 (v/v) mixture solution of methanol and HCl for 30 min.^46,47 After rinsing with deionized water, the glass substrate was dried in an oven at 100 °C for 6 h. To functionalize the surfaces of the glass substrate with an amine, the cleaned glass substrate was immersed in a 1% ethanol solution of APTMS for 30 min.⁴⁶ The amine coating of the glass surfaces was completed by washing with ethanol, sonicating for 3 min, and drying in an oven for 2.5 h. The core GNs were adsorbed onto the amine-functionalized surfaces of the glass substrates by immersing the substrate in a 5 mL solution of ∼15 nm GNs for 12 h. The unbound GNs were removed by washing three times with water and then rinsing with ethanol. The GN-adsorbed substrate needs to always be kept in solution during the washing and rinsing process. The substrate of the adsorbed core GNs was added to a PdCl₂ solution at various concentrations (0.5, 5, and 10 mM, respectively). The solution was stirred for 24 h at room temperature, resulting in the deposition of a Pd thin layer on the surface of the GNs. After washing and rinsing with water and ethanol, the glass substrate, which was covered with the core–shell nanoparticles, was placed in 5 mL of water and sonicated for 30 s using an ultrasonic cleaner. Upon sonication, the asymmetric core–shell nanoparticles desorbed preferentially from the substrates into the ethanol, and the solution was kept in the dark prior to use.

Synthesis of rGO–AuPd composites

Initially, a 100 mL suspension of the prepared GOs (0.25 mg mL⁻¹) was ultrasonicated in three-necked 250 mL flask for 30 min. Subsequently, 5 mL of 35 wt% hydrazine was mixed with 33.3 mL of 30% ammonia water and heated to the reaction temperature at 95 °C for 3 h. The synthesized rGOs were cooled slowly to room temperature and washed thoroughly with distilled water. The rGOs dispersed in water were mixed with a solution of AuPd nanoparticles (2.0 mg mL⁻¹) by stirring for 5 h. The resulting rGO-supported AuPd nanoparticles (AuPd/rGO) were filtered and washed thoroughly with distilled water. The AuPd/rGO was separated from the washing water by centrifugation at 4000 rpm for 20 min. This washing step was repeated at least three times. The thoroughly washed AuPd/rGO that settled at the bottom of the centrifuge cell was then dried completely overnight under a vacuum environment (Fig. S1†). For comparison, a symmetric AuPd/rGO was also prepared as follows. The solution of as-prepared GNs was mixed with a rGO support dispersed in water (2.0 mg mL⁻¹) by stirring for 2 h. A dispersed solution of Au/rGO was added to a PdCl₂ solution at 5 mM concentration, which is the optimal concentration for ORR performance in asymmetric AuPd/rGO. The rest of purification steps are the same as the procedure for asymmetric AuPd/rGO catalysts.

Electrocatalyst characterizations

Field-emission scanning electron microscopy (FE-SEM, Hitachi S-4300), which was equipped with an energy dispersive X-ray spectroscopy (EDS) system, and high-resolution transmission electron microscopy (HR-TEM, FEI Tecnai G2 F20 S-TWIN) were used to verify the surface morphology and composition of the prepared catalysts. The elemental mapping images were taken using a Hitachi HF3300 scanning transmission electron microscope (STEM) with an acceleration voltage of 300 kV. For HR-TEM and STEM analysis, a few drops of the sample suspension were placed on a carbon-coated copper TEM grid and dried in air before the observations. The nanoparticles formed were characterized by UV-Vis spectroscopy (SCINCO UV-Vis Spectrophotometer Mega 2100). Powder XRD was carried out on a Rigaku D/max-2500 with Cu Kα (λ = 1.54 Å). X-ray photoelectron spectroscopy (XPS, ESCALAB 250Xi, Thermo Scientific) was performed using Al Kα radiation in a twin anode at 14 kV × 16 Ma, which was calibrated internally by the binding energy (BE) of C (1s) at 285 eV. The Au and Pd contents in the AuPd catalysts were also analyzed by inductively coupled plasma-mass spectrometer (ICP-MS, Thermo Scientific, ICAP Q).

Electrochemical measurements

Prior to electrode surface modification, the GC rotating disk working electrode (RDE; 0.071 cm²) was wet-polished on an Alpha A polishing cloth (Mark V Lab) using 0.3 and 0.05 μm aluminum slurries, successively. The electrode was rinsed with distilled water and finally sonicated in distilled water for 30 s to remove any residual aluminum slurry from the electrode surface. A 2 mg portion of each catalyst was dispersed in 1 mL of water by 15 min ultrasonication of the catalyst suspension. A 10 μL sample of the catalyst suspension was applied to the GC electrode three times and dried in air for 4 h at room temperature. Subsequently, 10 μL of a Nafion solution (0.05 wt%) was applied over the dried electrode and dried at ambient temperature. Commercial Pd-20/C and Pt-20/C electrodes have also been prepared in a similar manner. Cyclic voltammetry (CV) was conducted using an electrochemical analyzer (601, CH Instruments, Inc.). The measured currents were converted to the current densities by normalization to the geometric surface area (GSA) of each modified electrode, which was determined by chronocoulometry (CC) in an aqueous 10 mM K₃Fe(CN)₆ solution containing 0.1 M KCl as the supporting electrolyte. The activity of the prepared catalysts for the ORR was characterized by RDE voltammetry. All electrochemical RDE experiments were carried out using a RDE-2 of a BASi electrochemical analyzer (Bioanalytical Systems Inc., West Lafayette, IN). The RDE experiments were run in a 0.1 M NaOH aqueous solution saturated with either oxygen or nitrogen (Dong-Sung Gas Co.) at a scan rate of 10 mV s⁻¹ over the potential range of 0.1 and −0.75 V vs. a saturated calomel reference electrode (SCE) with a Pt wire counter electrode. The ORR currents were recorded at electrode rotating speeds ranging from 400 to 3600 rpm.

Results and discussion

Fig. 1 outlines the entire procedure to obtain the asymmetric AuPd/rGO core-sell nanoparticles. The core GNs with mean diameter of ∼15 nm were chosen for this study. An asymmetric layer of the Pd shell was coated onto the GNs using a GRR-like method. The extent of replacement between the GNs and Pd²⁺ ions could be controlled by varying the amount of PdCl₂. To obtain both high catalytic activity and stability in ORR, solutions with various concentrations of PdCl₂ were added to a reaction vessel containing the glass substrate with the adsorbed core GNs. A series of as-prepared asymmetric AuPd/rGO catalysts are denoted as AuPd (Au : Pd wt% ratio)/rGO based on their composition. On the other hand, the standard reduction potential of Pd(II) (Pd²⁺/Pd, 0.951 V vs. NHE) was lower than that of the Au(III) (AuCl₄⁻/Au, +1.002 V vs. NHE),⁴⁸ and a GRR between Pd²⁺ and GNs was not anticipated to be favorable thermodynamically. One possible reason for the observed GRR-like Pd deposition associated with Au etching could be explained by the adsorption of free Pd²⁺ on the GN surface.^38,39 GNs have negative surface charges that can apply attractive forces to positive charges of Pd²⁺ ions, inducing an increase in the local Pd²⁺ concentration and facilitating Pd²⁺ reduction and Au oxidation to AuCl₄⁻ by shifting the equilibrium to the right side in eqn (1).

3Pd²⁺(aq) + 2Au(s) + 8Cl⁻(aq) ↔ 3Pd(s) + 2AuCl₄⁻(aq) (1)

Fig. 1 Schematic illustration of the synthetic process for asymmetric core–shell nanoparticles using APTMS-coated glass slide: (1) surface functionalization of a glass substrate with an amine using APTMS, (2) adsorption of ∼15 nm GNs onto the glass substrate, (3) spontaneous deposition of Pd on GNs, and (4) desorption of the core–shell AuPd nanoparticles from the glass substrate into water by sonication. The Pd shells are recognizable by the different contrast in the HR-TEM images because of the difference in the Au and Pd atomic numbers.

In general, the growth of the second metal on the surface of the core particles is quite complex and can be affected by a number of chemical/physical parameters, e.g., reaction kinetics, the correlation of the surface and interface energies, the reducing and stabilizing agents, and even the difference in electronegativity between the two elements.^49–53 Indeed, this synthetic approach provides some advantages over the other chemical methods of core–shell nanoparticles formation, e.g., any unreacted PdCl₂ and large isolated Pd particles from GRR-like fabrication and self-nucleation as a side reaction are removed readily by washing process (see Fig. S2†). In addition, it does not require the further removal of templates and greatly simplifies the synthetic procedures.

To obtain high catalytic activity and stability in the ORR, various concentrations of PdCl₂ solutions were added to a reaction vessel containing the adsorbed core GNs on the glass substrate. The core–shell AuPd nanoparticles with the PdCl₂ precursor solution can be observed when the Pd²⁺ ion concentration is rather high. The presence of Pd was confirmed by ICP-MS analysis, where a Au : Pd wt% ratio of 0.61 : 0.39 was observed under the optimal precursor concentration of PdCl₂ (5 mM). The ICP-MS results were compared with the XPS results to reveal any differences in composition between the surface layers and the bulk material, and thus confirm the core–shell morphology. In each case, XPS overestimated the expected surface metal compared to the ICP-MS results. Indeed, the XPS-derived elemental analyses revealed surface enrichment of the shell metal (Table S1†).

The morphology of the resulting products was investigated by high-resolution transmission electron microscopy (HR-TEM). Fig. 2 presents HR-TEM images of the AuPd/rGO, in which the lattice fringes indicated by the pairs of lines show a d-spacing of 0.23 nm, corresponding to the (111) lattice spacing of Au and Pd. The right panels of Fig. 2 show images obtained by HAADF-STEM and EDS elemental mapping of the Au_0.61Pd_0.39 catalyst. Pd(0) indicated by arrow (a) in Fig. 2 tended to form on the exposed GNs surface rather than on the basal region (arrow b) between the GNs and glass substrate. Although it was difficult to identify the Au surface remaining from the desorption of AuPd nanoparticles after the GRR-like process, Pd deposited on the surface of the GNs can be detected in AuPd by EDS mapping. This was also confirmed by UV-Vis absorption spectroscopy. As shown in Fig. S3,† strong absorption at 520 nm was observed for the GNs colloidal solution due to the SPR of the nanoparticles prepared with citrate reductase. On the other hand, the SPR signal of symmetric AuPd nanoparticles had disappeared entirely when the Pd precursor concentration was 10 mM of PdCl₂. This shows that the Pd layer coats the Au-core completely. A weak absorption peak by the asymmetric AuPd catalysts was observed at 520 nm at the same PdCl₂ concentration due to the exposed Au-core surface. This is further evidence of the asymmetric core–shell nanostructures.

Fig. 2 HR-TEM image of Au_0.61Pd_0.39 nanoparticles synthesized by a spontaneous reduction reaction. Representative HAADF-STEM image (right top panel) of Au_0.61Pd_0.39 nanoparticles and its corresponding Au (red) and Pd (green) EDX elemental mapping images. The arrows indicate the (a) occupied and (b) unoccupied areas of the cores, highlighting the asymmetrical features of the prepared AuPd nanoparticles.

Fig. 3 presents the XPS profiles of each AuPd/rGO catalyst with varying Au/Pd atomic distributions using different Pd precursor concentrations in the presence of the Au/rGO catalyst. The low-energy band (Au 4f_7/2) peak appeared at a binding energy of 84.0 eV and a high-energy band (Au 4f_5/2) peak appeared at a binding energy of 87.7 eV, which indicates the formation of metallic Au.⁵⁴ This suggests that Au(0) is dominant for the series of AuPd/rGO catalysts. Indeed, the spectral intensities at Au 4f decreased with increasing concentration of the precursor Pd. This shows that the GNs are covered with Pd layers. Fig. 3b presents a typical Pd 3d XPS spectrum of the AuPd/rGO catalysts, showing that the composites consist mainly of Pd(0) with a small amount of Pd²⁺ at 337.2 and 342.7 eV, which may be caused by surface oxidation during the synthesis process and/or XPS sample preparation.

Fig. 3 XPS profiles of (A) Au 4f and (B) Pd 3d for a series of AuPd/rGO catalysts: (a) Au_0.72Pd_0.28/rGO; (b) Au_0.61Pd_0.39/rGO; and (c) Au_0.58Pd_0.42/rGO.

XRD was performed for further structural and composition evaluation of the catalysts. For all samples, the broad peak centered at approximately 23.5° and 24.3° 2θ in Fig. 4 corresponds to the amorphous nature of the Vulcan XC-72 carbon and rGO, respectively. The XRD peaks located at 38.2°, 44.4°, 64.6°, and 77.6° 2θ were assigned to the (111), (200), (220), and (311) planes of Au, respectively, which are in accordance with those reported for metallic Au.⁵⁵ The weak diffraction peaks were assigned to Pd-20/C because of the lower contents of Pd particles in the carbon support. The XRD patterns of AuPd/rGO series showed that the fcc structure of the nanoparticles has distinguished XRD peaks, particularly in the (311) planes. The XRD peaks of the (111), (200), and (220) planes overlap with each other, making it difficult to distinguish them. The XRD peak intensity of the (111) plane in the AuPd/rGO composites gradually became stronger with increasing Pd content to a certain level and decreased with further increases in Pd content: Au_0.61Pd_0.39/rGO > Au_0.72Pd_0.28/rGO ≈ Au_0.58Pd_0.42/rGO. Remarkably, there were no other peaks for compounds containing Pd, suggesting that the Pd ions are reduced completely to crystalline Pd nanoparticles. Compared to Pd-20/C, no XRD peaks assigned to Pd were evident in those of the AuPd/rGO catalysts, indicating the absence of large isolated Pd particles.

Fig. 4 XRD patterns of a series of AuPd/rGO catalysts: (a) rGO; (b) Pd-20/C; (c) Au_0.58Pd_0.42/rGO; (d) Au_0.61Pd_0.39/rGO; and (e) Au_0.72Pd_0.28/rGO.

Before evaluating the electrocatalytic behavior of the catalysts, the Au/rGO and AuPd/rGO-modified GC electrodes were characterized electrochemically by CV in a 0.1 M HClO₄ solution (Fig. 5 and S5†). The shape of the CV curves changed from the three distinctive regions, including a double layer capacitive region between 0.2 and 0.05 V, a hydrogen adsorption/desorption (H_ads/des) between 0.05 and −0.28 V, and a metal oxide formation/reduction of surface Pd (or Au) oxide layer between 1.4 and 0.2 V. In particular, Δj of the double layer between the hydrogen and oxide regions increased from Au/rGO to Au_0.61Pd_0.39/rGO and further decreased with increasing Pd content of Au_0.58Pd_0.42/rGO. This corresponds well to the XRD results and indicates that the change in Δj in the double layers shows a rather porous Pd layer on the Au/rGO frame core, leading to an enlarged surface area. The H_ads/des characteristics occurring at a more negative potential range than 0.05 V were not observed in Au/rGO. Hydrogen adsorption does not occur on gold electrodes.^56,57 In addition, a series of AuPd/rGO catalysts revealed well-defined voltammetry peaks associated with the H_ads/des characteristics that increased with increasing Pd : Au ratio, supporting the successful deposition of a Pd shell on the GNs core. The observed H_ads/des indicates the formation of a thin and porous Pd layer with fine nanostructures compared to bulk Pd. These peaks were not observed when the crystalline Pd size was increased significantly.⁵⁸

Fig. 5 (A) CV curves recorded in a 0.1 M HClO₄ solution deaerated with N₂ purging gas at a scan rate of 50 mV s⁻¹. (B) Plots of the Pd/Au wt% ratio (△) estimated by ICP-MS and integrated reduction charges of surface Pd oxide (□) and Au oxide (○) as a function of the precursor Pd concentration. The charge density was obtained by normalization to the GSA. The electrodes used in the electrocatalytic oxygen reduction reactions: (a) Au/rGO; (b) Au_0.72Pd_0.28/rGO; (c) Au_0.61Pd_0.39/rGO; and (d) Au_0.58Pd_0.42/rGO.

A voltammogram of Pd-20/C is characterized by a single reduction peak at approximately 0.34 V, which was attributed to the reduction of Pd oxide (Fig. S4†). A series of AuPd/rGO catalysts in Fig. 5 show anodic peaks corresponding to the formation of a surface oxide and cathodic peaks associated with the reduction of the oxide. In the cathodic scan direction, the peaks at ca. 0.30 and ca. 0.84 V correspond to the reduction of Pd and Au oxides, respectively. The cathodic peak at 0.84 V is not normally observed in symmetric Au-core Pd-shell nanoparticles due to the blocking effect in the Pd shell induced by the Au core (Fig. S5†).³⁹ On the other hand, a well-defined Au oxide reduction peak appeared in the AuPd/rGO series, supporting the successful synthesis of an asymmetric Pd shell on the Au core. In other words, the Au reduction peaks of the AuPd/rGO catalysts clearly show regions of unreacted gold remaining from the desorption of the core–shell nanoparticles. Unlike the CV peaks for Au_0.58Pd_0.42/rGO, a distinctive reduction peak at ca. 0.54 V was obtained for Au_0.61Pd_0.39/rGO, which indicates the existence of separate surface phases showing distinctly different electrochemical properties.⁵⁹ Careful comparisons of the CV waves related to a metal oxide formation/reduction suggest that Au and Pd are alloyed well in the Au_0.61Pd_0.39/rGO sample compared to other AuPd/rGO compositions.⁵⁵ The improved catalytic activity toward the ORR is believed to be due to the surface alloy effects between the Au and Pd phases in the bimetallic AuPd/rGO catalysts, which will be discussed in the next section. Moreover, the incorporation of Pd to Au can increase the onset potential of surface oxide formation during fuel cell operation, because Au is the most electronegative metal, which can have an electron-withdrawing effect on the neighboring metal atoms in the Au-based nanoparticles, reduce the total quantity of metal dissolution in the AuPd alloy catalysts, and stabilize the AuPd catalyst. Fig. 5B shows the charge amounts for surface Au or Pd oxide reduction and different Pd/Au wt% ratios in a series of AuPd/rGO catalysts depending on the Pd precursor concentration. The integration of the AuO current densities of the AuPd/rGO series decreased in the following order: Au/rGO > Au_0.72Pd_0.28/rGO > Au_0.61Pd_0.39/rGO > Au_0.58Pd_0.42/rGO. The charges corresponding to the Pd real surface area showed a large increase from 0.5 to 5 mM and a rather small increase from 5 to 10 mM of PdCl₂, even if the Pd/Au ratio increased gradually. This suggests that the most efficient porosity of the Pd layer, which is possibly related to the catalyst activity, is achieved on the Au core at 5 mM PdCl₂ among the precursor concentrations studied.

The catalytic performance of Au/rGO, and a series of asymmetric AuPd/rGO catalysts toward the ORR was investigated by RDE voltammetry under alkaline conditions. Fig. 6A shows the normalized RDE voltammetry curves recorded in the O₂-saturated 0.1 M NaOH solution at 1600 rpm with each catalyst-loaded GC electrode. For comparison, the ORR activity of the commercial Pd-20/C and Pt-20/C catalyst-loaded GC electrodes was also studied. The ORR began at more positive potentials for Au_0.61Pd_0.39/rGO, followed by Pt-20/C. The ORR onset potentials were more positive in the following order: Au_0.61Pd_0.39/rGO > Pt-20/C > Au_0.72Pd_0.28/rGO > Au_0.58Pd_0.42/rGO > Pd-20/C > Au/rGO > rGO. The trend of the half-wave potentials (E_1/2) was similar to that of the ORR onset potentials except for Pd-20/C. The E_1/2 value of Au_0.61Pd_0.39/rGO (−0.158 V vs. SCE) was more positive than that of the other AuPd/rGO catalysts. The most positive onset potential and E_1/2 values of Au_0.61Pd_0.39/rGO indicate that the efficient porosity for the Pd layer formed on GNs with an enhanced electrochemically active surface area facilitates the ORR more favorably. Therefore, Au_0.61Pd_0.39/rGO exhibited a higher ORR current density in the mixed kinetic-diffusion controlled region (>−0.01 V) than Pd-20/C and Pt-20/C, supporting its superior ORR activity. Moreover, all aspects investigated support the high ORR activity of asymmetric Au_0.61Pd_0.39/rGO, even surpassing that of symmetric AuPd/rGO counterpart (Fig. S6†).

Fig. 6 (A) RDE polarization curves for oxygen reduction in an O₂-saturated 0.1 M NaOH solution at a scan rate of 10 mV s⁻¹ (rotation speed = 1600 rpm). (B) K–L plots for oxygen reduction obtained from the RDE data presented in (A). The electrodes used in the electrocatalytic oxygen reduction reactions: (a) rGO; (b) Au/rGO; (c) Au_0.72Pd_0.28/rGO; (d) Au_0.61Pd_0.39/rGO; (e) Au_0.58Pd_0.42/rGO; (f) Pd-20/C; and (g) Pt-20/C.

The number of electrons transferred (n) during the course of the ORR was calculated for a series of AuPd/rGO catalysts, and commercial Pd-20/C and Pt-20/C catalysts, from the RDE voltammetry results obtained at various electrode rotation speeds using the Koutecky–Levich (K–L) equation:⁶⁰

where j is the measured limiting current density, j_k is the kinetic current density, j_d is the diffusion-limited current density, n is the number of electrons transferred in the ORR, F is Faraday's constant, C_O2 (1.2 × 10⁻⁶ mol cm⁻³) is the saturated concentration of oxygen, D_O2 (1.9 × 10⁻⁵ cm² s⁻¹) is the diffusion coefficient of oxygen, v is the kinematic viscosity of the solution (1.0 × 10⁻² cm² s⁻¹), and ω is the electrode rotation rate. The parameters for the calculation were obtained from the literature.⁶¹

Fig. 6B presents the K–L plots (j_d⁻¹ vs. ω^−1/2) of the ORR using the RDE current densities at −0.6 V (vs. SCE). The plots exhibit good linearity, indicating that the ORR kinetics is first order with respect to the reactant concentration. The n values for Au_0.72Pd_0.28/rGO, Au_0.61Pd_0.39/rGO, Au_0.58Pd_0.42/rGO, Pd-20/C, and Pt-20/C, which were calculated from the slopes of the K–L plots, were 3.89, 3.98, 3.87, 3.82, and 3.84, respectively. This suggests that oxygen is reduced via a direct four-electron transfer pathway at a series of AuPd/rGO and commercial Pd-20/C and Pt-20/C catalysts in an alkaline solution, whereas the AuPd/rGO series showed a relatively higher n value.

The direct ORR pathway via 4-electron transfer is more desirable rather than a 2-electron transfer pathway producing hydrogen peroxide, which might have adverse effects on the stability of the catalysts. Indeed, the 4-electron transfer ORR pathway, i.e., producing OH⁻ as the main ORR product, is dominant in all three catalysts. All aspects investigated (e.g., the ORR onset and half-wave potentials, limiting current density levels, RDE curve sharpness, and n values) support the high ORR activity of AuPd/rGO (particularly with an optimized Au : Pd ratio = 0.61 : 0.39), which are even comparable to that of commercial Pd-20/C and Pt-20/C. In addition to the effects of the ORR activities of AuPd/rGO, the electrocatalytic stability is also one of the important factors to evaluate the performance of a catalyst. Indeed, similar RDE voltammograms, in terms of the ORR onset potential, curve shape, and limiting current, were obtained with the Au_0.61Pd_0.39/rGO and Pt-20/C-modified GC electrodes for at least 100 repeated RDE runs in an O₂ saturated 0.1 M NaOH solution at a rotation rate of 1600 rpm, as shown in Fig. 7. A closer inspection showed that the limiting current densities decreased by 4.83 and 6.56%, and the E_1/2 values were shifted negatively by 0.01 and 0.013 V for Au_0.43Pd_0.57/rGO and Pt-20/C, respectively. Overall, both the intrinsic structure with an enhanced surface area and high stability of the catalyst were achieved by effectively constructing an asymmetric nanostructure of AuPd/rGO.

Fig. 7 RDE polarization curves for the oxygen reduction obtained repetitively for 100 runs in an O₂-saturated 0.1 M NaOH solution using a (A) Au_0.61Pd_0.39/rGO and (B) Pt-20/C-modified GC electrodes at a scan rate of 10 mV s⁻¹ (rotation speed = 1600 rpm).

Conclusions

A series of rGO-supported asymmetric Au-core Pd-shell nanoparticles as a function of the Pd precursor concentration were prepared using a straightforward synthetic strategy. The nanoparticles produced exhibit unique structural features, where a core GNs (∼15 nm) was decorated spontaneously by a porous Pd layer with part of the core surfaces left unoccupied. The Pd metal precursor reacted spontaneously with GNs formed through a GRR-like mechanism, resulting in a AuPd/rGO series. Inspired by the attractive structure of the AuPd/rGO series, the RDE voltammetry results verified the high catalytic activity of the AuPd/rGO catalysts toward the ORR. Indeed, Au_0.61Pd_0.39/rGO, despite having a lower Pd content than Au_0.58Pd_0.42/rGO, exhibited the best ORR activity among the catalysts: positive ORR onset and E_1/2 potentials, higher n value, greater limiting current density, and steeper RDE curve slope in the mixed kinetic-diffusion controlled region. This developed facile synthesis with a unique structure is expected to be quite valuable for routinely producing new multimetallic core–shell structures with various sizes, morphologies, compositions, and shell thicknesses.

Acknowledgements

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2014R1A1A2054826).

References

1. M. Shao, Q. Chang, J.-P. Dodelet and R. Chenitz, Chem. Rev., 2016, 33, 1285–1288.
2. Y. Nie, L. Li and Z. Wei, Chem. Soc. Rev., 2015, 44, 2168–2201.
3. Y.-J. Wang, N. Zhao, B. Fang, H. Li, X. T. Bi and H. Wang, Chem. Rev., 2015, 115, 3433–3467.
4. J. J. Baschuk and X. Li, Int. J. Energy Res., 2001, 25, 695–713.
5. C. M. Pedersen, M. Escudero-Escribano, A. Velázquez-Palenzuela, L. H. Christensen, I. Chorkendorff and I. E. L. Stephens, Electrochim. Acta, 2015, 179, 647–657.
6. A. Hassan, V. A. Paganin and E. A. Ticianelli, Appl. Catal., B, 2015, 165, 611–619.
7. D.-S. Yang, C. Kim, M. Y. Song, H.-Y. Park, J. C. Kim, J.-J. Lee, M. J. Ju and J.-S. Yu, J. Phys. Chem. C, 2014, 118, 16694–16702.
8. E. Antolini, Energy Environ. Sci., 2009, 2, 915–931.
9. L. Xiao, L. Zhuang, Y. Liu, J. Lu and H. D. Abruna, J. Am. Chem. Soc., 2008, 131, 602–608.
10. M. Shao, J. Odell, M. Humbert, T. Yu and Y. Xia, J. Phys. Chem. C, 2013, 117, 4172–4180.
11. M. Lüsi, H. Erikson, A. Sarapuu, K. Tammeveski, J. Solla-Gullón and J. M. Feliu, Electrochem. Commun., 2016, 64, 9–13.
12. J. Hernández, J. Solla-Gullón, E. Herrero, A. Aldaz and J. M. Feliu, J. Phys. Chem. C, 2007, 111, 14078–14083.
13. X.-R. Li, X.-L. Li, M.-C. Xu, J.-J. Xu and H.-Y. Chen, J. Mater. Chem. A, 2014, 2, 1697–1703.
14. D. Chen, J. Li, P. Cui, H. Liu and J. Yang, J. Mater. Chem. A, 2016, 4, 3813–3821.
15. B. Hammer and J. K. Norskov, Nature, 1995, 376, 238–240.
16. G. C. Bond, Catal. Today, 2002, 72, 5–9.
17. R. J. Davis, Science, 2003, 301, 926–927.
18. J. X. Wang, H. Inada, L. Wu, Y. Zhu, Y. Choi, P. Liu, W.-P. Zhou and R. R. Adzic, J. Am. Chem. Soc., 2009, 131, 17298–17302.
19. P. Strasser, S. Koh, T. Anniyev, J. Greeley, K. More, C. Yu, Z. Liu, S. Kaya, D. Nordlund, H. Ogasawara, M. F. Toney and A. Nilsson, Nat. Chem., 2010, 2, 454–460.
20. T. Balcha, J. R. Strobl, C. Fowler, P. Dash and R. W. J. Scott, ACS Catal., 2011, 1, 425–436.
21. J. Yang, X. Chen, F. Ye, C. Wang, Y. Zheng and J. Yang, J. Mater. Chem., 2011, 21, 9088–9094.
22. K. Tedsree, T. Li, S. Jones, C. W. A. Chan, K. M. K. Yu, P. A. J. Bagot, E. A. Marquis, G. D. W. Smith and S. C. E. Tsang, Nat. Nanotechnol., 2011, 6, 302–307.
23. W. Annan, P. Qing and L. Yadong, Chem. Mater., 2011, 23, 3217–3222.
24. Q. Tan, C. Du, G. Yin, P. Zuo, X. Cheng and M. Chen, J. Catal., 2012, 295, 217–222.
25. J. H. Yang, J. Yang and J. Y. Ying, ACS Nano, 2012, 6, 9373–9382.
26. G. H. Jeong, D. Choi, M. Kang, J. Shin, J.-G. Kang and S.-W. Kim, RSC Adv., 2013, 3, 8864–8870.
27. C.-H. Kuo, L. K. Lamontagne, C. N. Brodsky, L.-Y. Chou, J. Zhuang, B. T. Sneed, M. K. Sheehan and C.-K. Tsung, ChemSuSChem, 2013, 6, 1993–2000.
28. X. Yuan, G. Sun, H. Asakura, T. Tanaka, X. Chen, Y. Yuan, G. Laurenczy, Y. Kou, P. Dyson and N. Yan, Chem.–Eur. J., 2013, 19, 1227–1234.
29. M. Ren, Y. Zhou, F. Tao, Z. Zou, D. L. Akins and H. Yang, J. Phys. Chem. C, 2014, 118, 12669–12675.
30. D. Chen, C. Li, H. Liu, F. Ye and J. Yang, Sci. Rep., 2015, 5, 11949.
31. L. Kuai, X. Yu, S. Wang, Y. Sang and B. Geng, Langmuir, 2012, 28, 7168–7173.
32. G. Fu, Z. Liu, C. Yu, J. Lin, Y. Tang and T. Lu, Nano Res., 2014, 7, 1205–1214.
33. Y. Ding, F. Fan, Z. Tian and Z. L. Wang, J. Am. Chem. Soc., 2010, 132, 12480–12486.
34. M. T. Reetz, W. Helbig and S. A. Quaiser, Chem. Mater., 1995, 7, 2227–2228.
35. H. Remita, A. Etcheberry and J. Belloni, J. Phys. Chem. B, 2003, 107, 31–36.
36. V. Amendola and M. Meneghetti, Phys. Chem. Chem. Phys., 2009, 11, 3805–3821.
37. N. Mottaghi, M. Ranjbar, H. Farrokhpour, M. Khoshouei, A. Khoshouei, P. Kameli, H. Salamati, M. Tabrizchi and M. Jalilian-Nosrati, Appl. Surf. Sci., 2014, 292, 892–897.
38. Z. Sheykhifard, M. Ranjbar, H. Farrokhpour and H. Salamati, J. Phys. Chem. C, 2015, 119, 9534–9542.
39. J. H. Shim, J. Kim, C. Lee and Y. Lee, Chem. Mater., 2011, 23, 4694–4700.
40. M. Mandal, S. Kundu, S. K. Ghosh and T. Pal, J. Photochem. Photobiol., A, 2004, 167, 17–22.
41. J. Xu, A. R. Wilson, A. R. Rathmell, J. Howe, M. Chi and B. J. Wiley, ACS Nano, 2011, 5, 6119–6127.
42. A. Ruban, B. Hammer, P. Stoltze, H. L. Skriver and J. K. Nørskov, J. Mol. Catal. A: Chem., 1997, 115, 421–429.
43. T. Bligaard and J. K. Nørskov, Electrochim. Acta, 2007, 52, 5512–5516.
44. A. D. McFarland, C. L. Haynes, C. A. Mirkin, R. P. Van Duyne and H. A. Godwin, J. Chem. Educ., 2004, 81, 544A–544B.
45. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
46. R. G. Freeman, K. C. Grabar, K. J. Allison, R. M. Bright, J. A. Davis, A. P. Guthrie, M. B. Hommer, M. A. Jackson, P. C. Smith, D. G. Walter and M. J. Natan, Science, 1995, 267, 1629–1632.
47. N. Nath and A. A. Chilkoti, Anal. Chem., 2002, 74, 504–509.
48. A. J. Bard, R. Parsons and J. Jordan, Standard Potentials in Aqueous Solutions; Marcel Dekker, New York, 1985.
49. Z. Peng and H. Yang, Nano Today, 2009, 4, 143–164.
50. F.-R. Fan, D.-Y. Liu, Y.-F. Wu, S. Duan, Z.-X. Xie, Z.-Y. Jiang and Z.-Q. Tian, J. Am. Chem. Soc., 2008, 130, 6949–6951.
51. B. Lim, H. Kobayashi, T. Yu, J. Wang, M. J. Kim, Z.-Y. Li, M. Rycenga and Y. Xia, J. Am. Chem. Soc., 2010, 132, 2506–2507.
52. J. Li, Y. Zheng, J. Zeng and Y. Xia, Chem.–Eur. J., 2012, 18, 8150–8156.
53. J. Zeng, C. Zhu, J. Tao, M. Jin, H. Zhang, Z.-Y. Li, Y. Zhu and Y. Xia, Angew. Chem., Int. Ed., 2012, 51, 2354–2358.
54. T. F. Jaramillo, S.-H. Baeck, B. R. Cuenya and E. W. McFarland, J. Am. Chem. Soc., 2003, 125, 7148–7149.
55. A. Cha, J. H. Shim, A. Jo, S. C. Lee, Y. Lee and C. Lee, Electroanalysis, 2014, 26, 723–731.
56. J. H. Shim, J. Kim, C. Lee and Y. Lee, J. Phys. Chem. C, 2011, 115, 305–309.
57. F. Hernandez and H. Baltruschat, Langmuir, 2006, 22, 4877–4884.
58. J. J. Salvador-Pascual, S. Citalán-Cigarroa and O. Solorza-Feria, J. Power Sources, 2007, 172, 229–234.
59. M. Łukaszewski and A. Czerwiński, J. Solid State Electrochem., 2008, 12, 1589–1598.
60. J. Koutecky and V. G. Levich, Zh. Fiz. Khim., 1958, 32, 1565–1575.
61. S. Wang, D. Yu and L. Dai, J. Am. Chem. Soc., 2011, 133, 5182–5185.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra19155j

---