Nickel core–palladium shell nanoparticles grown on nitrogen-doped graphene with enhanced electrocatalytic performance for ethanol oxidation

Abstract

Herein, we report a facile two-step strategy for green synthesis of nickel core–palladium shell nanoclusters on nitrogen-doped graphene (Ni@Pd/NG) without any surfactant and additional reducing agent. During the synthesis, nitrogen-doped graphene acted as both the active substance and support by taking advantage of its moderate reducing and highly dispersing capacities. Characterization indicated that a uniform dispersion of Ni@Pd nanoparticles on nitrogen-doped reduced graphene oxide had a 2.8 nm average particle size. Unexpectedly, the as-prepared Ni@Pd/NG hybrid exhibited much greater activity and stability than that of Pd/graphene and commercial Pd/C electrocatalyst at the same Pd loadings. Possible mechanisms for the enhanced electrocatalytic performance of nitrogen-doped reduced graphene oxide after combining with Ni@Pd are proposed. The present study provides an efficient strategy to synthesize highly efficient electrocatalysts.

---

1. Introduction

Among all types of fuel cells, direct ethanol fuel cells (DEFCs) based on liquid fuels is a promising biomass-derived power source and is being widely investigated due to its unique properties, including higher theoretical mass energy density, lower toxicity and operating temperature, and easier handling and transportation than other types of fuel cells.^1–3 Moreover, ethanol can be easily obtained on a large scale from chemical industry and fermentation of agricultural products or biomass.⁴ In developing DEFCs technology, highly active electro-catalysts are important and necessary to completely oxidize ethanol. Therefore, it has become a hot research topic to design and develop efficient ethanol electro-oxidation anode catalysts for DEFCs, especially in alkaline media.⁵ Nowadays, the electrocatalysts of anode catalyst in DEFCs have predominantly relied on Pt based catalysts, which have been extensively investigated.^6–9 Moreover, the high cost remains a choking point for the general application of fuel cells for a long time, especially due to the high cost of the noble metal supported electrocatalyst, which is one of the most critical units in fuel cell systems. It is necessary to develop new catalyst materials or improve the efficiency of existing catalysts and supports for alcohol oxidation. Thus, there is a strong motivation to increase catalyst utilization via their dispersion as small particles on a support material. Recently, Pd-based nano-electrocatalysts are emerging as an excellent substitute to Pt-based catalysts and have been proven to be a promising candidate for direct ethanol fuel cells because of higher abundance, lower cost, higher electrocatalytic activity and greater resistance to intermediate products for ethanol electro-oxidation in alkaline medium.¹⁰ However, there are also a number of challenges and obstacles for Pd based nano-electrocatalysts such as low utilization efficiency of Pd, which hinders the practical application in commercialization of portable fuel cell technology. It is still necessary to develop low-cost, effective catalysts for DEFCs. Most noble metal NPs catalytic reactions occur only on the NPs surface and a large fraction of atoms in the core are catalytically inactive. Therefore, to make a large percentage of noble metal atoms available for catalysis and to reduce their consumption, the inner noble metal atoms could be replaced by other non-noble metals.^11,12 A variety of core–shell nanoparticles have also been reported as catalysts in DMFCs. Density functional theory studies suggested that the enhanced catalytic activity for the core–shell nanoparticle originates from a combination of an increased availability of CO-free Pt surface sites on the Ru@Pt nanoparticles and a hydrogen-mediated low-temperature CO oxidation process that is clearly distinct from the traditional bifunctional CO oxidation mechanism.¹³ Currently, more research focus is on the core/shell structure with a noble metal shell and base transition metal (Ni, Co, Cu, Fe) core.^14–16 Consequently, a noble metal at the outer surface and base metal second atomic layer instead of a noble metal as electrocatalyst promotes changes in the catalytic activity, selectivity, and stability due to synergistic effects.¹⁷ In addition to the active metal regulation, another alternative effective approach to enhance electrocatalytic activity is to seek and develop novel catalyst supports.^18–20 Graphene, as an atomic-layer-thick two dimensional material, displays intriguing potential benefit as a support material for DEFCs due to its many unique chemical and physical properties such as superior electrical conductivity, high surface-to-volume ratio, ultrathin thickness, structural flexibility and chemical stability.^21–23 Recently, heteroatom doped graphene materials have received much attention as supports for electrocatalysts, paving the way for the growth of catalytically active metals with controlled morphology and dispersion on the graphene support surface.^24,25 To further tailor the catalytic support properties of graphene, the nitrogen doping is important and perhaps the most frequently chosen method that can enhance the graphene conductivity and induce n-type semi-conductor behavior, because the nitrogen atom is of comparable atomic size and contains five valence electrons available to form strong valence bonds with carbon atoms.^26,27 More importantly, the nitrogen-doped graphene (NG) with more functional groups for property design would be provided by incorporating different types of nitrogen into the graphene carbon network.^28,29 Recent studies show that nitrogen doping of catalyst supporting materials can provide small particle sizes and a high dispersion of the catalyst nanoparticles, strong bonding between the support and catalyst, electronic structure modification of the catalysts and an increase in the supports electronic conductivity. In addition, studies also show that nitrogen doped graphene can act as good electrocatalysts even in the absence of precious metals.³⁰ Inspired by our previous studies, it is of great interest to develop highly active Ni_core–Pd_shell supported on N-doped graphene catalysts for the ethanol oxidation reaction. Because metal nano-particles interact strongly with the nitrogen-doped graphene surface, it can be highly dispersed and show good stability with the graphene surface and the interaction between Ni core and Pd shell, which make Ni@Pd/NG establish a fairly conductive network to facile charge-transfer and mass-transfer processes. Therefore, the combination of Ni@Pd nanoparticles (NPs) and NG may open up a new avenue for designing next generation DEFCs catalysts.

In this study, Ni@Pd core/shell nanoparticles supported on nitrogen-doped graphene (Ni@Pd/NG) and its electrocatalytic activity and stability for alcohol oxidation are presented. The preparation of Ni@Pd core/shell nanostructure could not only decrease Pd consumption, but also take advantage of the interaction between the Ni core and Pd shell such as the ligand effect downshifting the D-band energy center, which is favourable for promoting the electrochemical activity and stability of the Ni@Pd/NG nanostructure.³¹ At the same time, N-doped graphene not only served as a support, but it also acted as a secondary catalyst that could increase the active sites accessibility. The immobilization of Ni@Pd NPs embedded on the N doped graphene to fabricate the Ni@Pd/NG nanocatalyst would be promising for the purpose of preventing Ni@Pd aggregation. We applied the Ni@Pd/NG to catalyze alcohol oxidation, wherein the possible origin of the large surface area and excellent chemical stability of Ni@Pd/NG was discussed.

2. Experimental details

2.1. Materials

Natural flake graphite was obtained from Qingdao Guyu graphite Co., Ltd. with a 150 nm particle size. Nickel acetate tetrahydrate (Ni(ac)₂·4H₂O), ethylene glycol, ammonia and palladium chloride (PdCl₂) were purchased from Sinopharm Chemical Reagent Co., Ltd., China and used as received without any further purification. Distilled water was also used throughout the experiments.

2.2. Synthesis of GO and NRGO

Graphite oxide (GO) was prepared from purified natural graphite using a modified Hummers' method.³² A mixture of 3.0 g of graphite powder, 150 mL of H₂SO₄ and 6.0 g of NaNO₃ was prepared and immersed into an ice bath. Then, KMnO₄ (8.0 g) was added gradually with stirring and the temperature of the mixture was maintained below 10 °C. After the reaction mixture temperature stabilized, the reaction media was heated to 80 °C and stirred for 2 h until it obtained a pasty brownish appearance and then was diluted with distilled water. Successively, the mixture was stirred for 30 min and 25 mL of 30 wt% H₂O₂ solution was slowly added to the mixture to reduce the residual KMnO₄, after which the color of the mixture changed to brilliant yellow. The solid product was separated by centrifugation and washed repeatedly with 5% HCl solution to remove metal ions followed by 1.5 L of distilled water to remove the acid. For further purification, the resulting solid was re-dispersed in distilled water and dialyzed for one week to remove any residual salts and acids. Finally, the solid was separated by sintered discs and freeze-dried.

To synthesize NRGO material, 80 mg of GO were dispersed into 60 mL water and magnetically stirred for 3 h. GO was exfoliated by sonication for 2 h. Subsequently, a certain amount of urea was also added and the solution was ultrasonicated for another 0.5 h. Finally, the mixture was transferred into a 50 mL Teflon lined stainless steel autoclave and was heated at 180 °C for 12 h. The resulting product was filtered, washed with distilled water several times and freeze-dried.

2.3. Preparation of Ni/NG composite

For synthesizing the Ni/NG compounds, 25 mg of NRGO was dispersed in 100 mL of ethylene glycol (EG) by sonication for 60 min. Then, the solution was added with 15 mg of nickel acetate under vigorously stirring and then subjected to a microwave (MW) oven (900 W) for intermittent microwave heating with a 20 s on and 10 s off procedure for 15 cycles in a microwave oven at 180 °C. The solution pH was adjusted to 10 by adding NaOH (2.0 M). After cooling to room temperature, the product was washed with distilled water 5–6 times to remove the excess EG and subsequent separation by sintered discs and dried in vacuum at 80 °C for 24 h. Finally, the nickel impregnated nitrogen-doped graphene (Ni/NG) was obtained. The Ni contents were analyzed by inductively coupled plasma spectroscopy (ICP, Optima2000DV, USA) analysis, which showed 26.5 wt% of Ni in the Ni/NG compound.

2.4. Preparation of Ni@Pd/NG electrocatalyst

Ni@Pd/NG nanosized electrocatalysts were synthesized using a replacement method. Typically, 20 mL of 0.093 mol L⁻¹ H₂PdCl₄ and 1.65 g Ni/NG were dispersed in 50 mL of distilled water in a beaker. The resulting solution was uniformly dispersed by sonication for 10 min and then vigorously stirred for 24 h at room temperature. This caused spontaneous Ni replacement by Pd as follows:

2Ni/NG + PdCl₄²⁻ → Pd(Ni)/NG + 2Ni²⁺ + 6Cl⁻

These reactions are thermodynamically favorable because the standard potentials of the Ni²⁺/Ni couples are −0.257 V vs. SHE, they are lower than the standard potential of the Pd(II) Cl₄²⁻/Pd couple (+0.951 V vs. SHE). The black solid was separated by sintered discs, washed with deionized water several times and finally dried in a vacuum oven at 60 °C. For comparison, Pd nanoparticles supported on nitrogen-doped reduced graphene oxide (Pd/NG) as an electrocatalyst was also obtained directly by reducing the H₂PdCl₄ in a graphene suspension using ethylene glycol microwave reduction. The theoretical Pd content in both Ni@Pd/NG and Pd/NG were targeted at 15 wt%. ICP analysis obtained the actual Pd contents of 13.8 wt% for Ni@Pd/NG and 13.5 wt% for Pd/NG.

2.5. Catalyst electrode preparation

For electrode preparation, 25.1 mg of Ni@Pd/NG were dispersed in 1 mL ethanol and 1 mL 0.5 wt% Nafion suspension (DuPont, USA) under ultrasonic agitation to form the electrocatalyst ink. The electrocatalyst ink (40 μL) was then deposited onto the glassy carbon rod surface and dried at room temperature overnight. The total Pd loadings were controlled at 0.02 mg cm⁻². All chemicals were of analytical grade and used as received.

2.6. Characterization of the supports and the electrocatalysts

The particle sizes and shapes of the as-prepared samples were examined by a transmission electron microscope (TEM, JEOL-JEM-2010, Japan) operating at 200 kV. The as-synthesized products were washed with absolute ethyl alcohol to remove impurities and then were dispersed in absolute ethyl alcohol. 2–3 drops of the dispersion were dripped onto a carbon-coated TEM grid. The crystalline phase and phase purity of the Pd nanoparticles and Ni@Pd particles were analyzed by X-ray diffraction (XRD) using a D8 Advance X-ray diffractometer (Bruker AXS company, Germany) equipped with Cu-K_α radiation (λ) 1.5406 (Å), employing a 0.02° s⁻¹ scanning rate in the 2θ range from 10° to 80°. X-ray photoelectron spectroscopy (XPS) was carried out on a PHI-5702 and the C 1s line at 291.4 eV was used as the binding energy reference. The graphitization degree of N doped-graphene was determined by a Laser Micro-Raman Spectrometer (Renishaw inVia, Renishaw plc, UK). All electrochemical measurements were performed in a three-electrode cell on an IM6e potentiostat (Zahner-Electrik, Germany) at 30 °C controlled by a water-bath thermostat. A platinum foil (1.0 cm²) and Hg/HgO (1.0 mol dm⁻³ KOH) were used as counter and reference electrodes.

3. Results and discussion

The schematic for forming Ni@Pd/NG nanocomposite is shown in Fig. 1. Once graphite oxide was synthesized, it was dispersed thoroughly in a urea solution by sonication and was heated at 180 °C for 12 h. Then, it was subjected to microwave (MW) heating in a microwave oven after nickel nitrate was added. Finally, the Ni/NG was added to the H₂PdCl₄ solution and continuously stirred for 14 h at room temperature and well dispersed Ni@Pd nanoparticles with a core/shell structure on N-doped graphene were obtained.

Fig. 1 Schematic of Ni@Pd nanocatalysts synthesis on N-doped graphene.

Fig. 2 shows typical XRD patterns of the as-prepared of GO (a), NRGO (b), Pd/NG (c), Ni@Pd/NG (d) composites. All peaks could be indexed with face centered cubic (fcc) structures. The original graphite oxide sample showed a strong diffraction peak centered at 2θ = 9.8°, corresponding to the (001) reflection of graphene oxide, far smaller than that of pristine graphite³³ (Fig. 2a). After surface doping, the N-doped graphene revealed negligible C (001) peaks compared to GO (Fig. 2b). The weakening of the crystalline C (001) peak suggests that urea corrosion and defects resulted from nitrogen doping made the interlayer spacing larger. In addition, as observed from the XRD pattern of Fig. 2c, the diffraction peaks at 2θ = 40.1°, 46.5° and 68.2° could be indexed to the characteristic (111), (200) and (220) cubic crystalline structured Pd, respectively. Moreover, as it is displayed in Fig. 2d, a positive shift of the Pd peaks occurred obviously on the Ni@Pd/NG compared with the Pd/NG, indicating the Ni atoms entered into the Pd crystals, which caused a very small change of the Pd crystal lattice distance. Wu et al.³⁴ reported that the variation of crystal lattice parameters of noble metal could obviously improve its catalytic activities. This can also be attributed to the decorating effect of the Pd atomic shell layer on the alloyed nanoparticles. The Pd particle size was calculated from all the Pd crystal plane parameters and averaged 2.8 nm for Ni@Pd/NG, which was very close to the TEM results.

Fig. 2 Shows typical XRD patterns of the as-prepared GO (a), NRGO (b), Pd/NG (c), Ni@Pd/NG (d) composites.

Raman spectroscopy is a powerful and widely used tool for identifying graphene-based materials and detecting the doping effect of GO, such as defect structures and disordered crystal structures of carbon, and carbon-heteroatoms from carbon–carbon bonds.³⁵ Fig. 3 compares the Raman spectra of GO, NRGO and Ni@Pd/NG. All the samples displayed one intense D band at 1347.27 cm⁻¹ due to the expected crystallite structure of graphite³⁶ and one relatively weak G band due to the optically allowed E_2g mode. A blue shift of the G band is usually a result of the effects of disorder and the presence of isolated short double bond segments.³⁷ It is noteworthy that the NGO (1583.03 cm⁻¹, Fig. 3c) and Ni@Pd/NG composites (1578.18 cm⁻¹, Fig. 3a) exhibited blue shifts compared with that of GO (Fig. 3b). This revealed that the partial N heteroatoms may be introduced onto these structural sites to form substitutional N species during nitridation and can increase the numbers of structural defects on the graphene surface.³⁸ These indicate the electronic interaction between NRGO and Ni@Pd for electron transportation between NRGO and Ni@Pd. In addition, the intensity ratio of the D and G bands (I_D/I_G) is an indication of the number of structural defects and a quantitative measure of edge plane exposure, which is most commonly being used as the doping level and the chemical modification of the graphitic carbon sample.³⁹ It was found that I_D/I_G values of Ni@Pd/NG (1.13) and NGO (1.10) were higher than that of GO (1.02), indicating that successful nitrogen doping into RGO that can induce a higher concentration of structure defects.⁴⁰ Such a higher ratio of D peak and G peak in Ni@Pd/NG is attributable to the decrease in the average size of the sp² domains upon the formation of nanoparticles on the graphene and the structural defects and edge plane exposure caused by nitrogen atom incorporation into the graphene layers.⁴¹

Fig. 3 Raman spectra of (a) (Ni@Pd/NG), (b) (GO), and (c) (NRGO).

The morphology and element distribution of Ni@Pd nanoparticles supported on NG were investigated by TEM and EDS and the results are shown in Fig. 4. Compared with GO (Fig. 4a), the NRGO (Fig. 4b) appeared to obviously aggregate with increased numbers of layers. It was proven experimentally that the defects created by nitrogen doping act as the anchoring sites for the uniform dispersion of small metal nanoparticles homogeneously on the surface even without any additional protective reagents or surfactants in the system (Fig. 4c). The HRTEM image describing the crystalline nature of the Ni@Pd nanoparticles is shown in Fig. 4d. The single crystalline Ni@Pd particles are confirmed, the lattice planes with an interlayer distance of 0.203 nm in the core are indexed to Ni (111) crystal planes, the outer layer with the lattice space of 0.224 nm corresponds to Pd (111) crystal planes. The elemental analysis by EDS proves that the Ni@Pd/graphene is composed of C, Ni and Pd (Fig. 4e, the Cu signal comes from the sample holder). The particle size distribution of Ni@Pd/NG particles derived from the TEM image in a narrow diameter range from 1.5 to 4.5 nm and the average diameter is about 2.8 nm (Fig. 4f). Clearly, nickel nanoparticles interact strongly with the N-doped graphene surface playing a key role in keeping a similar and highly dispersed Ni@Pd particle size on the support. It is known that N-doped graphene tends to interact with metal species. Density functional theory calculation was used to show that extra Ni–C bonds were formed at the interface and more electrons transferred from the interfacial C–C bonds to the Ni–C bonds.⁴² It can be imagined that the Ni particles were supported on N doped graphene first and then the Pd particle coating on the Ni core. At the same time Pd particles were fixed on the graphene because the bottom of Pd also contacts with the graphene. Possibly, palladium and nickel nanoparticles had a strong interaction with the N doped graphene surface, which restrained the aggregation and the growth of the nanoparticles to form larger particles, resulting in the uniform distribution of these nanoparticles.

Fig. 4 TEM images of GO (a), NRGO (b), Ni@Pd/NG (c), HRTEM images of Ni@Pd/NG (d), EDS spectrum of Ni@Pd/NG (e) and size distribution of Ni@Pd/NG metal particles derived from TEM images (f).

XPS were further performed to analyze the chemical compositions, oxidation states and nitrogen bonding configurations of the Ni@Pd/NG catalysts. The high resolution XPS spectra of C 1s (Fig. 5a) showed three peaks (C–C, 283.9 eV; C–N, 285.5 eV; C=O, 287.9 eV). The peak centered at 283.9 eV can be assigned to the adventitious hydrocarbon from the defect-containing sp² hybridized carbon atoms present in graphitic domains, and the C–N bond (285.5 eV) suggests that the N was successfully doped in the graphene nanosheets.⁴³ Moreover, the signal of the high resolution N 1s peak (Fig. 5b) could be deconvoluted into three characteristic peaks, corresponding to the presence of pyridinic N, pyrrolic N and quaternary N in the Ni@Pd/NG.⁴⁴ As in previous reports, the pyridinic N (398.3 eV) contributed to nitrogen atoms at the edge of graphene planes that are bonded to two carbon atoms and donate one p electron to the aromatic π system, the pyridinic edge sites can also provide a stronger interaction between catalyst nanoparticles and carbon support and this reduces the agglomeration of a core–shell nanophase. Pyrrolic N (399.4 eV) refers to the N atom contributing two p-electrons to the p system due to the contribution of pyridine and pyrrole functionalities; and quaternary N (400.8 eV) was derived from the graphene layers by replacing a carbon atom within a graphene plane and contributed two p electrons to the π system.^45,46 The detailed Pd_3d spectra of catalysts (Fig. 5c) displayed a doublet consisting of a high-energy band (Pd 3d_3/2) and a low-energy band (Pd 3d_5/2) at 339.98 and 334.65 eV for Ni@Pd/NG, such binding energy values are in accordance with those reported for the Pd lattice and a Ni@Pd core/shell was formed. The peaks of Ni 2p_3/2 and Ni 2p_1/2, which were located at around 855.5 and 873.4 eV (Fig. 5d), respectively, are assigned to nickel in two surface chemical states on the N doped-graphene. Remarkably, there are some extra peaks labeled as satellite peaks, which are present around the expected Ni 2p_3/2 and Ni 2p_1/2 signals in the Ni 2p region.⁴⁷

Fig. 5 XPS spectra of (a) C 1s, (b) N 1s, (c) Pd 3d and (d) Ni 2p of the Ni@Pd/NG sample.

Cyclic voltammetry (CV) a highly useful technique that is frequently used to investigate the heterogeneous electron transfer ability of an electrode surface. Fig. 6a shows the cyclic voltammograms (CVs) of alcohol electro-oxidation on the abovementioned prepared electrodes in 1.0 M KOH + 1.0 M ethanol solution at 303 K at a 50 mV s⁻¹ scan rate. Two featured strong anodic current peaks were observed in the CV curves of the different catalysts (Fig. 6a). The oxidation peak in the forward scan was reminiscent of the amount of ethanol in the electro-oxidation process at the Pd-based electrocatalysts and was consistent with literature reports.^48,49

Fig. 6 (a) Cyclic voltammograms for ethanol oxidation on Ni@Pd/NG, Pd/NG, Pd/G and Pd/C electrodes and (b) cyclic voltammograms of Ni@Pd/NG, Pd/NG, Pd/G and Pd/C electrodes in 1.0 mol dm⁻³ KOH solution at 303 K and a 50 mV s⁻¹ scan rate.

The reverse scan peak was primarily associated with the removal of carbonaceous species that are not completely oxidized in the forward scan.⁵⁰ More explicitly, the mass activity of the ethanol oxidation in 1 mol dm⁻³ ethanol solution on Ni@Pd/NG was 3650 mA mg_Pd⁻¹, which is higher than that of 2435 mA mg_Pd⁻¹ on Pd/NG and also higher than that of 1528 mA mg_Pd⁻¹ on a Pd/G electrode. The Ni@Pd/NG electrode showed 4.5 times the peak current density as that on Pd/C electrocatalyst at the same Pd loadings under similar electrochemical reaction conditions. The activities of the Ni@Pd/NG electrocatalyst were most likely a result of the particular structure of bimetallic nanodendrites and the high surface area of N-doped graphene as a support being beneficial for the uniform dispersion of the Ni@Pd nanoparticles to increase their mass transfer capability, which displays vast commercial competition. When the Ni was added to form the Ni–Pd alloy, according to Zhang's report,⁵¹ the Ni would transform to Ni(OH)₂ in alkaline media at the reaction potential and consequently increase the OH_ads coverage on the Pd surface, which would ultimately accelerate the reaction rate through the following equation:

The abovementioned results were further verified by comparing the electrochemical active surface area (EASA), which is shown in Fig. 6b in a KOH solution without ethanol. As shown in Fig. 6(b), the peaks between −0.9 and −0.5 V originated from the hydrogen adsorption–desorption.^52,53 On the reverse sweep, the defined peak near −0.25 V was characteristic of the reduction of Pd oxide formed during the positive-going sweep to Pd. No other obvious characteristic peaks appeared in the CV curves in KOH solution with or without alcohol, which is consistent with literature reports.⁵⁴

The EASA of the abovementioned catalysts were studied by CV tests from −0.70 to 0.20 V in 1.0 mol L⁻¹ KOH at a 50 mV s⁻¹ scan rate and were calculated based on the PdO reduction peak adapting, and the assumptions of 212 μC cm⁻² of the electrode surface⁵⁵ were 158.7 m² g⁻¹, 103.6 m² g⁻¹, 85.8 m² g⁻¹ and 59.4 m² g⁻¹ for the Ni@Pd/NG, Pd/NG, Pd/G and Pd/C, respectively. The EASA on Ni@Pd/NG was 2.7 times higher than that on Pd/C. The largest EASAs of Ni@Pd/NG catalyst was mainly due to replacement of Pd active sites on the surface by an N-doped active center and also due to its high specific surface area shell structure. Furthermore, theoretical study has shown that nitrogen-doped reduced graphene oxide leads to a higher positive charge on a carbon atom adjacent to the nitrogen atoms and a positive shift of Fermi energy at the apex of the Brillouin zone of graphene readily attracts electrons from the anode to facilitate the ethanol oxidation process.⁵⁶ Moreover, the Ni@Pd/NG had about a 40 mV negative onset potential compared to Pd/NG. The negative shift indicated that ethanol was more easily oxidized by Ni@Pd/NG, which was evidence that the core–shell structure could tremendously improve the kinetics of the ethanol electro-oxidation.⁵⁷ The results further indicated that the Ni@Pd significantly improved the activity and the output when used in fuel cells.

The stabilities of the Ni@Pd/NG and Pd/G electrodes for ethanol oxidation are shown in Fig. 7. The dotted lines are the cycling difference from the 1st cycle to the 3000th cycle. It is clear that the EASA of the Pd/G electrodes was decreased by 17.6% from 85.8 m² g⁻¹ to 70.7 m² g⁻¹ for ethanol oxidation after the 3000th cycle. However, the EASA of the Ni@Pd/NG was reduced by 8.9% from 158.7 m² g⁻¹ to 144.6 m² g⁻¹. It is evident that the Ni@Pd/NG electrocatalyst had a higher stability than the Pd/G. The promoted electrochemical stability may be due to the stronger interaction force between Ni@Pd and N doped-graphene than the force between Pd and graphene.

Fig. 7 (A) The cyclic voltammograms of the Ni@Pd/NG electrocatalysts and (B) the cyclic voltammograms on Pd/G in 1.0 mol dm⁻³ KOH solution at 303 K and a 50 mV s⁻¹ scan rate. The dotted line shows the difference from the 1st cycle to the 3000th cycle.

The improved mass transfer property of the Ni@Pd/NG electrocatalyst was observed by performing various scan rates for ethanol oxidation (Fig. 8). The deflection from the linear line of the data at lower scan rates in the relationship of peak current density and square root of scan rate on electrode indicated the improved mass transfer. The straight line appeared over the 49 mV s⁻¹ scan rate of Ni@Pd/NG compared with Pd/G (20 mV s⁻¹), indicating N doped-graphene improved mass transfer; however, it showed a straight line at any scan rate with the Pd/C electrode, indicating concentration polarization.

Fig. 8 Plots of the peak current density against the square root of the scan rate for Ni@Pd/NG, Pd/G and Pd/C electrodes in 1.0 mol dm⁻³ ethanol/1.0 mol dm⁻³ KOH solution at 303 K.

The chronoamperometric curves on the three electrodes at a fixed potential of −0.18 V are shown in Fig. 9. It can be observed that the initial current densities decreased rapidly for all the three catalysts due to the formation of reactive intermediates, such as CO_ads, CH₃COOH_ads and CH₃CHO_ads, during the electrochemical reaction process,⁵⁸ but the Ni@Pd/NG showed the highest initial and highest steady-state current density than the Pd/NG and PD/C, indicating that the Ni@Pd/NG hybrid possessed a good catalytic durability and excellent catalytic activity.

Fig. 9 Chronoamperometric curves for Ni@Pd/NG, Pd/NG and Pd/C at −0.18 V (vs. SCE) in 1.0 M KOH + 1.0 M ethanol solution.

The CV data and chronoamperometric measurements suggested that the Ni@Pd/NG exhibits an enhanced catalytic activity and good stability for ethanol electro-oxidation. The superior electrocatalytic performance of the proposed catalyst (Ni@Pd/NG) could be attributed to the following factors. First, it is understood that the catalyst surface structure strongly affected the catalytic activity. The core/shell structure catalysts have different intermetallic surface structures and a downshift in the D-band energy center of the noble metal, which will certainly affect the catalytic activity. Researchers refer to the electric “ligand effect” between the two metal layers of the core/shell catalyst, which could promote desorption of toxic intermediate products such as CO from the noble metal and consequently improve the anodic catalyst activity and stability.^59,60 Second, the N doped on the graphene sheets could not only help forming small, highly concentrated and uniformly dispersed Ni@Pd nanoparticles, but also the doped N acted as a secondary catalyst increasing more catalytic activity centers and the rapid removal of intermediate poisoning species. In addition, the Ni@Pd/NG possessed a larger EASA compared with the other electrocatalysts because of the increased catalytic activity center of N-doped graphene, a narrower size distribution, more uniform distribution and more perfect crystal structure than those of other as-prepared composites.

4. Conclusions

In summary, Ni@Pd/NG nanocomposites have been successfully fabricated through an environmentally friendly two-step strategy and the catalytic properties toward the ethanol oxidation are investigated. As a new catalyst, the Ni@Pd/NG exhibits better kinetics, large specific surface areas, higher tolerance and electrochemical stability than the other catalysts for ethanol electro-oxidation in DEFCs, which was correlated with the core shell architecture, nitrogen doping of graphene, and the synergistic effects between Ni and Pd. Importantly, the Ni cores of the Ni@Pd NPs reduced the consumption of Pd, N-doped graphene had a large specific surface area and more active sites. This remarkable improvement of the electrocatalytic performance may provide new insights into the other graphene-based metallic systems and prepare good electrocatalyst materials in alcohol oxidation for fuel cells.

Acknowledgements

This study was the financially supported by the Natural Science Foundation of Jiangsu (BK20140531), Zhenjiang Industry Supporting Plan (GY2014040), the China Postdoctoral Science Foundation (2015M570410) and Research Foundation for Talented Scholars of Jiangsu University (14JDG187); Dr Suci Meng thanks the National Natural Science Foundation of China (21103073).

References

1. J. Wu and H. Yang, Acc. Chem. Res., 2013, 46, 1848.
2. M. K. Debe, Nature, 2012, 486, 43.
3. Z. X. Yan, M. M. Zhang, J. M. Xie, J. J. Zhu and P. K. Shen, Appl. Catal., B, 2015, 165, 636.
4. D. P. He, K. Cheng, H. G. Li, T. Peng, F. Xu, S. C. Mu and M. Pan, Langmuir, 2012, 28, 3979.
5. Z. X. Yan, M. Cai and P. K. Shen, J. Mater. Chem., 2012, 22, 2133.
6. S. H. Xu, Z. F. Li, Y. F. Ji, S. W. Wang, X. Y. Yin and Y. X. Wang, Int. J. Hydrogen Energy, 2014, 39, 20171.
7. M. M. Zhang, J. M. Xie, Q. Sun, Z. X. Yan, M. Chen and J. J. Jing, Int. J. Hydrogen Energy, 2013, 38, 16402.
8. M. M. Zhang, J. M. Xie, Q. Sun, Z. X. Yan, M. Chen, J. J. Jing and A. M. S. Hossain, Electrochim. Acta, 2013, 111, 855.
9. Z. X. Yan, M. M. Zhang, J. M. Xie and P. K. Shen, J. Power Sources, 2013, 243, 336.
10. S. Y. Wang, D. S. Yu and L. M. Dai, J. Am. Chem. Soc., 2011, 133, 5182.
11. K. Tedsree, C. W. A. Chan, S. Jones, Q. Cuan, W. K. Li, X. Q. Gong and S. C. E. Tsang, Science, 2011, 332, 224.
12. H. Yin, H. Tang, D. Wang, Y. Gao and Z. Tang, ACS Nano, 2012, 6, 8288.
13. A. Alayoglu, A. U. Nilekar, M. Mavrikakis and B. Eichhorn, Nat. Mater., 2008, 7, 333.
14. M. M. Zhang, Y. Li, Z. X. Yan, J. J. Jing, J. M. Xie and M. Chen, Electrochim. Acta, 2015, 158, 81.
15. S. H. Zhou, B. Varughese, B. Eichhorn, G. Jackson and K. McIlwrath, Angew. Chem., Int. Ed., 2005, 44, 4539.
16. M. M. Zhang, Z. X. Yan, Q. Sun, J. M. Xie and J. J. Jing, New J. Chem., 2012, 36, 2533.
17. O. Metin, S. F. Ho, C. Alp, H. Can, M. N. Mankin, M. S. Gultekin, M. Chi and S. Sun, Nano Res., 2013, 6, 10.
18. Z. X. Yan, J. M. Xie, P. K. Shen, M. M. Zhang, Y. Zhang and M. Chen, Electrochim. Acta, 2013, 108, 644.
19. Z. X. Yan, W. Wei, J. M. Xie, S. C. Meng, X. M. Lü and J. J. Zhu, J. Power Sources, 2013, 222, 218.
20. Y. H. Lu, M. Zhou, C. Zhang and Y. P. Feng, J. Phys. Chem. C, 2009, 113, 20156.
21. X. M. Chen, G. H. Wu, J. M. Chen, X. Chen, Z. X. Xie and X. R. Wang, J. Am. Chem. Soc., 2011, 133, 3693.
22. S. Li, H. Yang, Z. Dong, S. Guo, J. Zhao, G. Gou, R. Ren, J. Huang, J. Jin and J. Ma, Catal. Sci. Technol., 2013, 3, 2303.
23. D. He, K. Cheng, H. Li, T. Peng, F. Xu, S. Mu and M. Pan, Langmuir, 2012, 28, 3979.
24. D. P. He, Y. L. Jiang, H. F. Lv, M. Pan and S. C. Mu, Appl. Catal., B, 2013, 132, 379.
25. L. S. Zhang, X. Q. Liang, W. G. Song and Z. Y. Wu, Phys. Chem. Chem. Phys., 2010, 12, 12055.
26. H. L. Guo, P. Su, X. Kang and S. K. Ning, J. Mater. Chem. A, 2013, 1, 2248.
27. L. Sun, L. Wang, C. Tian, T. Tan, Y. Xie, K. Shi, M. Li and H. Fu, RSC Adv., 2012, 2, 4498.
28. H. M. Jeong, J. W. Lee, W. H. Shin, Y. J. Choi, H. J. Shin, J. K. Kang and J. W. Choi, Nano Lett., 2011, 11, 2472.
29. Z. S. Wu, A. Winter, L. Chen, Y. Sun, A. Turchanin, X. L. Feng and K. Müllen, Adv. Mater., 2012, 24, 5130.
30. S. Y. Yang, K. H. Chang and Y. L. Huang, Electrochem. Commun., 2012, 14, 39.
31. A. Tegou, S. Papadimitriou, I. Mintsouli, S. Armyanov, E. Valova, G. Kokkinidis and S. Sotiropoulos, Catal. Today, 2011, 170, 126.
32. M. M. Zhang, Z. X. Yan, Y. Li, J. J. Jing and J. M. Xie, Electrochim. Acta, 2015, 161, 48.
33. B. Zhao, P. Liu, Y. Jiang and D. Pan, J. Power Sources, 2012, 198, 423.
34. A. Wu, S. W. Ong, H. C. Kang and E. S. Tok, J. Phys. Chem. C, 2010, 114, 21252.
35. H. B. Wang, T. Maiyalagan and X. Wang, ACS Catal., 2012, 2, 781.
36. Z. H. Sheng, L. Shao, J. J. Chen, W. J. Bao, F. B. Wang and X. H. Xia, ACS Nano, 2011, 5, 4350.
37. E. F. Antunes, A. O. Lobo, E. J. Corat, V. J. Trava Airoldi, A. A. Martin and C. Veríssimo, Carbon, 2006, 44, 2202.
38. O. S. Kwon, S. J. Park, J. Y. Hong, A. R. Han, J. S. Lee, J. H. Oh and J. Jang, ACS Nano, 2012, 6, 1486.
39. Y. J. Zhang, K. Fugane, T. Mori, L. Niu and J. H. Ye, J. Mater. Chem., 2012, 22, 6575.
40. W. X. Wang, S. H. Liang, T. Yu, D. H. Li, Y. B. Li and X. F. Han, J. Appl. Phys., 2011, 109, C501.
41. D. S. Geng, S. L. Yang, Y. Zhang, J. L. Yang, J. Liu, R. Y. Li, T. K. Sham, X. L. Sun, S. Y. Ye and S. Knights, Appl. Surf. Sci., 2011, 257, 9193.
42. M. T. Zhao, W. Xiao, H. J. Zhang and K. Cho, Phys. Chem. Chem. Phys., 2011, 13, 11657.
43. H. J. Yin, H. J. Tang, D. Wang, Y. Gao and Z. Y. Tang, ACS Nano, 2012, 6, 8288.
44. E. Yoo, J. Nakamura and H. S. Zhou, Energy Environ. Sci., 2012, 5, 6928.
45. L. Sun, L. Wang, C. Tian, T. Tan, Y. Xie, K. Shi, M. Li and H. Fu, RSC Adv., 2012, 2, 4498.
46. S. Maldonado, S. Morin and K. J. Stevenson, Carbon, 2006, 44, 1429.
47. H. Chen, L. F. Hu, M. Chen, Y. Yan and L. M. Wu, Adv. Funct. Mater., 2014, 24, 934.
48. S. Guo, S. Dong and E. Wang, Adv. Mater., 2010, 22, 1269.
49. Q. Tan, C. Du, Y. Sun, G. Yin and Y. Gao, J. Mater. Chem. A, 2014, 2, 1429.
50. Z. Bai, P. Xu, S. Chao, H. Yan, Q. Cui, L. Niu, L. Yang and J. Qiao, Catal. Sci. Technol., 2013, 3, 2843.
51. Z. Y. Zhang, L. Xin, K. Sun and W. Z. Li, Int. J. Hydrogen Energy, 2011, 36, 12686.
52. X. Chen, Z. Cai, X. Chen and M. Oyama, J. Mater. Chem. A, 2014, 2, 315.
53. V. Bambagioni, C. Bianchini, Y. Chen, J. Filippi, P. Fornasiero, M. Innocenti, A. Lavacchi, A. Marchionni, W. Oberhauser and F. Vizza, ChemSusChem, 2012, 5, 1266.
54. H. Yang, H. Wang, H. Li, S. Ji, M. W. Davids and R. Wang, J. Power Sources, 2014, 260, 12.
55. S. Y. Wang, D. S. Yu, L. M. Dai, D. W. Chang and J. B. Baek, ACS Nano, 2011, 5, 6202.
56. S. U. Lee, R. V. Belosludov, H. Mizuseki and Y. Kawazoe, Small, 2009, 5, 1769.
57. H. T. Zheng, Y. Li, S. Chen and P. K. Shen, J. Power Sources, 2006, 163, 371.
58. B. He, G. R. Li, Z. L. Wang, Y. N. Ou and Y. X. Tong, J. Phys. Chem. C, 2010, 114, 19175.
59. A. C. Garcia, V. A. Paganin and E. A. Ticianelli, Electrochim. Acta, 2008, 53, 4309.
60. S. Papadimitriou, A. Tegou, E. Pavlidou, G. Kokkinidis and S. Sotiropoulos, Electrochim. Acta, 2007, 52, 6254.

---