Preparation and enhanced electrocatalytic activity of graphene supported palladium nanoparticles with multi-edges and corners

Abstract

Herein, we reported a facile and green synthesis of palladium nanoparticles with a concave and convex surface formed by multi-edges and corners only involving palladium precursor and ascorbic acid. Ascorbic acid was employed as both reductant and structure-directing agent. The as-prepared palladium nanoparticles were then assembled on reduced graphene oxide. The morphology and chemical composition of the products were confirmed by several characterization techniques. The fast development of fuel cells inspired us to examine the electrocatalytic activity of the products towards methanol, ethanol and formic acid electrooxidation in alkaline medium comparing with that of the commercial one. The characterization and electrochemical results were carefully analyzed and discussed, showing the product can be applied as a well-performed electrocatalyst towards the electrooxidation of methanol, ethanol, and formic acid in alkaline media. The possible electron transfer route was proposed with the assistance of MedeA VASP calculation.

---

Introduction

Noble metal nanomaterials have been attracting considerable attention due to their unique properties and potential applications in many fields.¹ As is known, the properties of noble metal nanomaterials critically depend upon their morphologies. Thus, researchers are still focusing on the preparation of noble metal nanomaterials with novel morphologies. Among various noble metal nanomaterials, palladium (Pd) nanomaterials were one of the most interesting topical materials which have broad applications in catalysis, fuel cells, and batteries.² Diverse morphologies of Pd nanomaterials have been synthesized via different approaches, such as electrodeposition, photochemical method, and wet-chemical route.³ In these methods, wet-chemical approach is one convenient and efficient way to produce Pd nanomaterials with high quality in high yield. Reducing agent is one necessary chemical during the preparation process of producing Pd nanomaterials in the wet-chemical approach. Among the various reducing agents, ascorbic acid (AA), a relatively weak one possessing a green and environmentally friendly nature, has been widely developed in the preparation of Pd nanomaterials with diverse morphologies like nanocube, nanooctahedron, nanorod, nanodentrities and nanobars.⁴ To the best of our knowledge, Pd nanoparticles with multi-edges and corners have never been reported employing AA as both reductant and structure-directing agent.

One of the most promising applications of Pd nanomaterials is fuel cells, the high-efficient and low-emission advanced power sources for portable electronic devices.⁵ It has been revealed that in most widely studied direct alcohol fuel cells, anode electrocatalyst is one of the most important factors that affect the cell performance.⁶ Though for a long time Pt nanomaterials were considered to be the best electrocatalyst in low-temperature fuel cell, the high cost and low abundance have heavily limited their practical applications.⁷ In contrast, Pd-based nanomaterials were much cheaper but with comparable or even better performance in alkaline media, especially towards the electrooxidation of ethanol.⁸ For example, Sawangphruk and co-workers fabricated ultraporous palladium nanocatalyst by controllable electrodeposition, coated it on flexible carbon fiber paper modified with reduced graphene oxide nanosheets and finded it exhibiting higher catalytic activity toward the electrooxidation of ethanol in alkaline media, more excellent poisoning tolerance to carbonaceous species, and higher stability than other electrodes.^8e In another example, Wang et al. synthesized carbon-supported well-dispersed Pd–Ni–P ternary catalyst and investigated the catalytic property for ethanol oxidation reaction in alkaline media, of which the results showed higher electrocatalytic activity can be obtained on the prepared catalyst comparing with the state-of-the-art similar catalysts.^8d Moreover, Pd nanomaterials were also found as well-performed electrocatalysts in direct formic acid fuel cell, not only in acid media, but also in alkaline electrolyte.⁹ Therefore, Pd nanomaterials were now considered as very promising alternatives to the expensive Pt materials and thus received more and more attentions. However, as mentioned above, Pd nanomaterials with novel morphology exhibiting well-performed catalytic property prepared by green approach is still a challenge under investigation, and multi-edges and corners are supposed to endow the nanostructure with larger specific surface area, resulting in obtaining larger current densities.

In this paper, we fabricated palladium nanoparticles with multi-edges and corners (Pd NPs) with rough surface employing ascorbic acid as both reductant and structure-directing agent. The morphology and chemical composition of the prepared nanoparticles were characterized by a series of measurements. Then Pd NPs were assembled onto the surface of reduced graphene oxide (RGO) to obtain Pd/RGO nanomaterial, and the electrochemical behaviors of both Pd nanoparticle and the hybrid nanomaterial were investigated towards methanol, ethanol, and formic acid in alkaline media in comparison with commercial Pd/C catalyst.

Results and discussion

Characterization of nanomaterials

SEM and TEM were employed to characterize the morphology of the nanomaterials. Fig. 1A and B show the SEM images of Pd NPs in different magnifications. As can be seen in Fig. 1A, the as-prepared Pd nanoparticles were in good uniformity and relatively well-dispersed on the substrate. With enlarging the magnification (Fig. 1B), nanoparticles looked like buds ready to burst, leading to the formation of a rough surface with concave and convex details contributed by the multi-edges and corners. Based on the calculation of more than 200 nanoparticles, the average diameter of the prepared Pd NPs was estimated to be 66 nm, and the corresponding particle size distribution histogram was shown in Fig. 1C. Along with SEM characterization, EDX was employed to identify the chemical composition of Pd NPs. As shown in Fig. 1D, two main peaks appearing around 3 keV could be ascribed to Pd element (a much higher peak was originated from silicon wafer, and C was from conductive adhesive), indicating the nanoparticle was composed of metallic Pd.

Fig. 1 SEM images of the as-prepared Pd NPs in low (A) and high (B) magnifications. (C) Particle size distribution histogram of the prepared Pd NPs. (D) EDX image of Pd NPs.

TEM was further employed to examine the surface morphology of the as-prepared nanomaterials. Fig. 2 is the typical TEM (A–C) and HRTEM (D) images of Pd NPs (A, B and D) and Pd/RGO nanomaterial (C). Magnification of Fig. 2B showed clear structure details of the Pd NPs, including the rough surface, and the uneven edges and corners. With the help of HRTEM characterization (Fig. 2D), the interplanar spacing was identified as 0.223 nm, which corresponds to the (111) plane of the face-centered cubic (fcc) Pd. The Pd/RGO hybrid nanomaterial was also examined by TEM. As shown in Fig. 2C, thin RGO nanosheet with wrinkles can be easily observed, and Pd NPs are assembled on its surface without the distortion of Pd NPs, further indicating the successful fabrication of Pd/RGO hybrid nanomaterial.

Fig. 2 (A–C) TEM images of Pd NPs (A and B), Pd/RGO nanomaterials (C). (D) HRTEM image of an individual Pd NP.

XRD was used to characterize the chemical composition and crystal structure of the prepared nanomaterials. Fig. 3A shows the XRD patterns of graphite oxide (GO) (a), RGO (b), Pd NPs (c) and Pd/RGO nanomaterial (d). The diffraction pattern of GO (curve a in Fig. 3A) shows a sharp peak at 2θ value of 9.94°, which corresponds to the (001) plane of GO,¹⁰ while weak peaks appeared at around 28.8° as pictured in curves b and d of Fig. 3A, accompanied with the vanishment of the peak corresponding to (001) plane of GO, suggesting GO has been reduced to RGO. Besides that, the diffraction peaks on curves c and d of Fig. 3A at 2θ values of 40.3°, 46.8°, 68.2°, 82.2° and 86.8° can be indexed to (111), (200), (220), (311) and (222) planes of the fcc phase of palladium according to the JCPDS no. 87-0645, further supporting the successful fabrication of Pd NPs and the assembly on RGO. To obtain further detailed information about Pd NPs, XRD data of Pd NPs were analyzed using Rietveld refinement. As shown in Fig. 3B, the calculated pattern matches well with the experimental data. The refined lattice parameters a and c are both 3.8903 Å. This is consistent with the JCPDS parameters and further indicates the nanoparticles we prepared are composed of Pd element.

Fig. 3 (A) XRD patterns of GO (a), RGO (b), Pd NPs (c) and Pd/RGO nanomaterials (d). (B) Rietveld refined XRD pattern of Pd/RGO nanomaterial.

The surface composition and the oxidation state of the hybrid nanomaterials were examined by XPS. Fig. 4A and B show the deconvoluted high-resolution Pd 3d (A) and C 1s (B) regions of XPS spectrum of the Pd/RGO nanomaterial. In Fig. 4A, peak of binding energy at 335.25 eV could be fitted by two spin–orbit doublets with Pd 3d_5/2 components located at 335.15 eV and 336.12 eV which can be attributed to metallic Pd and Pd(II) phases, respectively.¹¹ Since the depth of XPS investigation is no more than 10 nm, only surface composition can be detected. The data obtained in Fig. 4A indicate that two valence states of Pd⁰ (36.84%) and Pd²⁺ (53.16%) exist in the surface layer. Combining with XRD results, it can be deduced that the Pd NP we prepared is composed of metallic Pd. The appearance of Pd²⁺ in the surface layer might be assigned to the valence change caused by the interaction between ascorbic acid and Pd NPs. The XPS data also prove the influence of ascorbic acid on the synthesis of Pd NPs. Besides that, as shown in Fig. 4B, the peak fitting of C 1s showed the mainly deconvoluted peaks centered at 284.7, 286.2, 287.2 and 288.6 eV are corresponding to the C=C, C–O, C=O, and O–C=O groups, respectively.¹⁰ Comparing with the peak fitting of C 1s for GO (Fig. 4C), the peak densities of C=C peak in hybrid nanomaterial become stronger, while those from the oxygeneous functional groups become weaker, suggesting the efficient reduction of GO to RGO. XPS results further proved that the successful fabrication of the hybrid nanomaterial are composed of Pd and RGO.

Fig. 4 Deconvoluted XPS Pd 3d (A) and C 1s (B and C) regions of Pd/RGO nanomaterial (A and B) and GO (C). (D) Raman spectra of GO and Pd/RGO nanomaterial.

Raman spectroscopy is a helpful approach to characterize graphene. Fig. 4D shows the Raman spectra of GO and Pd/RGO. It is found that the peaks at wavenumbers of 1354 cm⁻¹ and 1597 cm⁻¹ can be assigned to D band, relative to the order or disorder of the graphite edge, and G band associated with the graphitic stacking structure. Based on calculation, I_D/I_G ratio of GO and Pd/RGO nanomaterial can be estimated to be 0.90 and 0.94. The higher intensity ratio of Pd/RGO over GO indicates the reduction extent of RGO is much higher. These results were well consistent with the XRD and XPS data previously discussed.

Electrochemical measurements

Since Pd-based nanomaterials were found to be well-behaved as electrocatalysts towards the electrooxidation of organic molecules such as methanol, ethanol and formic acid, the as-prepared nanomaterials were directly applied onto the surface of glassy carbon electrode (GCE) to investigate the corresponding electrocatalytic activities, while employing commercial 20% Pd/C catalyst as contrast. The electrochemical behaviors of commercial 20% Pd/C catalyst (a), Pd nanoparticle (b) and Pd/RGO nanomaterial (c) were firstly measured in 0.5 M KOH aqueous solution since the subsequent experiments were all carried out in alkaline media, as shown in Fig. 5. It can be seen that the peaks ascribing to the adsorption/desorption of hydrogen appeared between −0.8 V (vs. Ag/AgCl) and −0.6 V (vs. Ag/AgCl), and the reduction peaks assigned to PdO and the oxidation of Pd were shown around −0.3 V (vs. Ag/AgCl) and 0.4 V (vs. Ag/AgCl), respectively.¹² The peak current density of Pd/RGO (74.61 mA mg⁻¹) is about 1.7 times as large as that of Pd NPs (43.37 mA mg⁻¹), and about 5.5 times of that obtained on commercial 20% Pd/C catalyst (13.48 mA mg⁻¹), indicating Pd/RGO nanomaterial exhibits a well-performed catalytic activity in alkaline media, which is higher than that of Pd nanoparticles and commercial Pd/C catalyst.

Fig. 5 CV curves of commercial 20% Pd/C catalyst (a), Pd NPs (b) and Pd/RGO nanomaterials (c) in 0.5 M KOH aqueous solution at the scan rate of 50 mV s⁻¹.

Due to the important usage of nanomaterials in direct methanol fuel cell, our investigation started from the experiments towards methanol electrooxidation in alkaline electrolyte. Fig. 6 shows the electrochemical results and data analyses of the experiments in KOH electrolyte containing methanol. Fig. 6A is the CV curves of commercial 20% Pd/C catalyst (a), Pd nanoparticle (b) and Pd/RGO nanomaterial (c) in 0.5 M KOH aqueous solution containing 0.5 M methanol at the scan rate of 50 mV s⁻¹. All curves were normalized to unit Pd mass of the catalysts. As shown in Fig. 6A, two obvious oxidation peaks can be seen on the three investigated nanomaterials. One appeared at about −0.14 V (vs. Ag/AgCl) on the forward scan can be attributed to the oxidation of methanol,¹³ the other one appeared at −0.28 V (vs. Ag/AgCl) on the reverse scan. As is seen in Fig. 6A, Pd/RGO nanomaterial shows the best catalytic performance comparing with the other two catalysts. We normalized the mass activities of different catalysts towards methanol oxidation in KOH aqueous solution to unit Pd mass or unit total catalyst mass to compare the catalytic activities of the three nanomaterials. As can be seen on the left sides of each column groups (noted as left columns) in Fig. 6B, when current densities were normalized to unit Pd mass, Pd/RGO nanomaterial exhibits a maximum current density of 198.4 mA mg⁻¹, which is 2 times of that on Pd NPs (103.5 mA mg⁻¹) and 2.5 times of that on commercial 20% Pd/C catalyst (80.6 mA mg⁻¹). The enhanced activity of Pd nanoparticle compared to commercial catalyst should be attributed to the novel morphology of the as-prepared Pd nanoparticle, and the enhanced activity of Pd/RGO nanomaterial comparing with Pd nanoparticle should be contributed by the introduction of graphene that greatly improved the conductivity and enlarged the specific surface area of the hybrid nanomaterial. Besides that, the catalytic activities towards methanol electrooxidation normalized to unit total mass (right sides of each column groups, noted as right columns in Fig. 6B) were also compared, a maximum current density of 66.13 mA mg⁻¹ can be obtained on Pd/RGO nanomaterial, which is 4 times as large as that gained from commercial Pd/C catalyst (16.2 mA mg⁻¹), and is comparable with that acquired on pure Pd NPs (103.5 mA mg⁻¹). Such comparison showed the as-prepared Pd NPs exhibit excellent electrocatalytic activity towards methanol, and the introduction of RGO can significantly reduce the catalyst cost while maintaining the major part of the catalytic performance. Moreover, the prepared Pd/RGO nanomaterial also shows enhanced mass activity in comparison with Pd nanoflowers in previous report,¹⁴ also indicating the well-performed catalytic activity of the hybrid. Subsequently, we carried out an investigation on the noble metal cost on different catalysts because of its rare abundance and high cost. As shown in Fig. 6C, when desiring the same catalytic current towards methanol oxidation (500 mA), the required Pd amount is much less for Pd/RGO nanomaterial (2.52 mg) than Pd nanoparticle (4.83 mg) and commercial catalyst (6.2 mg). The lower Pd amount requirement indicates that the fabricated Pd NPs with novel morphology are of excellent catalytic activity via this green preparation, and the introduction of RGO can significantly reduce the catalyst cost when requiring the same catalytic current. Furthermore, the stabilities of the nanomaterials were also investigated. Fig. 6D shows the amperometric i–t curves of commercial 20% Pd/C catalyst (a), Pd NPs (b) and Pd/RGO nanomaterial (c) modified GCEs in 0.5 M KOH aqueous solution containing 0.5 M methanol at applied potential of −0.28 V (vs. Ag/AgCl). After a long period of time (8000 s), better stability and higher current density can be obtained on Pd/RGO nanomaterial comparing with Pd nanoparticle and commercial 20% Pd/C catalyst. The comparison shows the hybrid nanomaterial might be more suitable for application as electrocatalyst towards methanol electrooxidation.

Fig. 6 (A) CV curves of commercial 20% Pd/C catalyst (a), Pd nanoparticle (b) and Pd/RGO nanomaterial (c) modified GCEs in 0.5 M KOH aqueous solution containing 0.5 M methanol at the scan rate of 50 mV s⁻¹. (B) Mass catalytic activity towards methanol oxidation in KOH aqueous solution normalized to unit Pd mass (left columns) or unit total catalyst mass (right columns) of the catalysts, respectively. (C) Pd amount required for obtaining the same catalytic current of 500 mA. (D) Current density–time curves of commercial 20% Pd/C catalyst (a), Pd nanoparticle (b) and Pd/RGO nanomaterial in 0.5 M KOH aqueous solution containing 0.5 M methanol at applied potential of −0.28 V.

With the development of direct methanol fuel cell, methanol was considered to be non-environmentally friendly, resulting in the replacement of fuel became extremely urgent. According to recent reports, ethanol, taking the advantages of nontoxicity, recyclability, easiness to store, transport and detect with large reserve, has been regarded as an excellent alternative to substitute methanol in direct fuel cell as a kind of sustainable energy source. In this study, we further examined the electrocatalytic activity of the as-prepared nanomaterials towards ethanol electrooxidation. CV curves of ethanol electrooxidation in 0.5 M KOH aqueous solution containing V (vs. Ag/AgCl) and −0.28 V (vs. Ag/AgCl) appear on the forward and reverse scans, respectively. Pd nanoparticle (curve b) and commercial Pd/C catalyst (curve a) also showed similar but much lower signals. The electrocatalytic activities of the catalysts were compared according to the oxidation peak in the forward scan. As shown in the left columns in Fig. 7B, when the current densities of the catalysts were normalized to unit Pd mass, the maximum current densities on Pd NPs and commercial Pd/C catalyst are measured to be 218.7 mA mg⁻¹ and 92.31 mA mg⁻¹, whilst the maximum current density obtained on Pd/RGO nanomaterial is 511.2 mA mg⁻¹, which is calculated to be about 2.3 and 5.5 times as large as that on Pd nanoparticle and commercial Pd/C catalyst, respectively. In comparison with the above-discussed methanol electrooxidation, the electrocatalytic activity towards ethanol electrooxidation obtained on Pd/RGO nanomaterial is more significant, suggesting the as-prepared nanomaterial possesses excellent electrocatalytic activity towards ethanol electrooxidation. Furthermore, the catalytic activities of different nanomaterials towards ethanol electrooxidation normalized to total mass of the catalysts were also measured, as shown in the right columns in Fig. 7B. Maximum current density of 170.4 mA mg⁻¹ can be obtained on Pd/RGO nanomaterial, which is 9.2 times of that on commercial Pd/C catalyst (18.46 mA mg⁻¹), and is comparable with that on Pd NPs (218.7 mA mg⁻¹), which is similar to the discussion about methanol oxidation. The calculation not only shows excellent catalytic activity of the prepared nanomaterial towards ethanol oxidation which is much more significant than that towards methanol oxidation, but also reflects the cost saving potential of the hybrid nanomaterial. In the further study, as shown in Fig. 7C, the required Pd amount when obtaining the same catalytic current of 500 mA normalized to unit Pd mass is significantly reduced for Pd/RGO nanomaterial (0.98 mg) than that for Pd nanoparticle (2.29 mg) and commercial Pd/C catalyst (5.42 mg), indicating the indispensable contribution of as-prepared Pd nanoparticle fabricated by this green approach and the introduction of RGO, which shows improved Pd utilization of the hybrid nanomaterial. The stability of the nanomaterials towards ethanol electrooxidation was also examined by amperometric i–t technique. Fig. 7D shows the current density–time curves of commercial 20% Pd/C catalyst (a), Pd nanoparticle (b) and Pd/RGO nanomaterial (c) in 0.5 M KOH aqueous solution containing ethanol at applied potential of −0.28 V (vs. Ag/AgCl). From the curves in Fig. 7D, improved stability and higher current density can be observed on Pd/RGO nanomaterial comparing with Pd nanoparticle and commercial Pd/C catalyst after 8000 s examination. It can be considered that with the addition of graphene during the preparation of the nanocomposite, more active sites of the Pd/RGO catalysts were exposed. Stability results showed that the hybrid nanomaterial can be considered as excellent electrocatalyst towards ethanol oxidation. According to the electrochemical results towards ethanol oxidation, we can conclude that the as-prepared hybrid nanomaterial exhibits excellent electrocatalytic performance for ethanol oxidation in comparison with Pd NPs and commercial Pd/C catalyst. Experiments on methanol and ethanol electrooxidation also revealed that the prominent electrocatalytic properties might probably make the new electrocatalyst to be applicable in direct alcohol fuel cells.

Fig. 7 (A) CV curves of commercial 20% Pd/C catalyst (a), Pd nanoparticle (b), and Pd/RGO nanomaterial (c) modified GCEs in 0.5 M KOH aqueous solution containing 0.5 M ethanol at the scan rate of 50 mV s⁻¹. (B) Mass activity towards ethanol oxidation in KOH aqueous solution normalized to unit Pd mass (left columns) or unit total catalyst mass (right columns) of the catalysts, respectively. (C) Pd amount required for obtaining the same catalytic current of 500 mA. (D) Current density–time curves of commercial 20% Pd/C catalyst (a), Pd nanoparticle (b) and Pd/RGO nanomaterial (c) in 0.5 M KOH aqueous solution containing 0.5 M ethanol at applied potential of −0.28 V.

In other promising applications such as electrooxidizing formic acid in acidic and alkaline media,⁹ the as-prepared Pd/RGO nanomaterial also exhibits competitive performance. Fig. 8A shows the CV curves of commercial 20% Pd/C catalyst (a), Pd nanoparticle (b) and Pd/RGO nanomaterial (c) modified GCEs in 0.5 M KOH solution with 0.5 M formic acid at the scan rate of 50 mV s⁻¹. The current densities of different catalysts were normalized to unit Pd mass. In Fig. 9A, two peaks could be observed at around −0.4 V (vs. Ag/AgCl) in the forward and reverse scans respectively on hybrid nanomaterial (curve c), the intensities of which are much higher than that of the two competitors (curves a and b). The catalytic current densities of different catalysts towards formic acid electrooxidation in KOH were normalized to unit Pd mass and unit total catalyst mass, as shown in Fig. 8B. When the current densities were normalized to unit Pd mass (left columns in Fig. 8B), the peak current density of Pd/RGO hybrid nanomaterial can achieve 1.081 A mg⁻¹, while peak current densities on Pd nanoparticle and commercial Pd/C catalyst are calculated to be 0.70 A mg⁻¹ and 0.45 A mg⁻¹, respectively. The maximum current density for formic acid oxidation on Pd/RGO catalyst is estimated to be 1.5 times and 2.4 times larger than that on Pd nanoparticle and commercial catalyst. When the maximum current densities of different catalysts were normalized to the total catalyst mass, as shown in the right columns in Fig. 8B, the current density on Pd/RGO nanomaterial is obtained as 0.36 A mg⁻¹, which is 4 times as large as that on commercial Pd/C catalyst (0.09 A mg⁻¹), and about 50% of that on Pd NPs (0.7 A mg⁻¹). Based on the calculation of the mass ratio of Pd to RGO, results similar to that towards methanol or ethanol oxidation can also be revealed on this comparison. Furthermore, we also calculated the required Pd amount of different nanomaterials when desiring the same current density of 2 A (Fig. 8C). The lower amount of Pd cost on Pd/RGO nanomaterial (1.86 mg) than Pd nanoparticle (2.86 mg) and commercial Pd/C catalyst (4.44 mg) indicates the hybrid nanomaterial applied as electrocatalyst is much more economic and efficient. After the stability investigation as shown in Fig. 8D, we can deduce that the current density and stability of Pd/RGO nanomaterial are much higher than that of Pd nanoparticle and commercial catalyst after 8000 s test at applied potential of 0.2 V. Electrochemical results towards formic acid oxidation in alkaline media reveal that the hybrid nanomaterial exhibits prominent electrocatalytic performance for formic acid oxidation comparing with Pd nanoparticle and commercial Pd/C catalyst. All of the above electrochemical studies manifest the as-prepared Pd nanomaterial possesses the capability of electrooxidizing different organic molecules. The utilization of multi-fuel might be beneficial to broaden the application to a wider range of the direct fuel cell, resulted in leaving out the multiple procedures such as purification. Therefore, employing multi-fuel directly in direct fuel cells might enable the application to be potential in catalyzing different kinds of organic molecules in waste water which may make the waster water get better use.

Fig. 8 (A) CV curves of commercial 20% Pd/C catalyst (a), Pd nanoparticle (b), and Pd/RGO nanomaterial (c) in 0.5 M KOH solution containing formic acid at the scan rate of 50 mV s⁻¹. (B) Mass activity towards formic acid oxidation in KOH aqueous solution normalized to unit Pd mass (left columns) or unit total catalyst mass (right columns) of the catalysts, respectively. (C) Pd amount required for obtaining the same catalytic current of 2 A. (D) Current density–time curves of commercial 20% Pd/C catalyst (a), Pd nanoparticle (b) and Pd/RGO nanomaterial (c) in 0.5 M KOH aqueous solution containing 0.5 M formic acid at applied potential of 0.2 V.

Fig. 9 Bonding electron distribution in the vertical direction through the diagonal of the 4 × 4 supercell of graphene with one Pd atom located at the B site.

To analyze the mechanism of electron transfer between palladium NPs and graphene, relative theoretical calculation on the interaction between palladium adatom and graphene was carried out. In our calculations, Pd adatom on graphene prefers B-site bonding than T-site or H-site, which is consistent with the literature previously reported.¹⁵ Fig. 9 displays the bonding electron distribution in the vertical direction through the 4 × 4 parallelogram graphene supercell with one Pd atom located at the B site. The bonding electron distribution Δρ(r) shown in Fig. 9 is defined according to the following equation.

Δρ(r) = ρ(r) − [ρ_gra(r) + ρ_ada(r)] (1)

where ρ(r) is the charge density of the adatom-graphene system, while ρ_gra(r) and ρ_ada(r) represent the charge densities of graphene layer and adatom, respectively. Significant covalent bonding can be observed between the palladium adatom and graphene, leading to the larger diffusion barriers. Besides that, the bonding electron density is mainly localized on the carbon atoms next to the Pd adatom. In addition, the mechanism of electron transfer between palladium NPs and graphene is proposed, and the corresponding schematic illustration is shown in Fig. 10. According to the electromagnetic principles, charge distribution on metal surface is strongly depended upon curvature radius. The charge density σ at a point on the surface of radius r can be calculated by the equation:

σ = Q/4πr² (2)

where Q represents charge, and r is the curvature radius. The smaller the curvature radius, the larger the charge density, which revealed the nature of point discharge effect. A great quantity of edges and corners exist on the surface of the as-prepared palladium nanostructure, and carry dozens of positive charges in the light of electromagnetic principles. As is known, oxidation process occurs on the anode part of the direct fuel cells, resulting in the loss of plenty of electrons. These electrons would be firstly captured by the corners of the prepared palladium NPs, and transferred to graphene via the strong covalent interaction between Pd NPs and graphene, and then the external circuit, accomplishing the catalysis on fuel cell anodes, and supplying directional current in the external circuit loop. This electron transfer is not considered to enhance the catalytic performance of Pd NPs with multi-edges and corners towards organic molecules. However, this kind of structure could result in the spontaneous formation of localized electric field, which could impel the fast movement of electrons in the hybrid nanomaterial and increase the electron transfer efficiency. This is hopeful to effectively reduce the internal resistance of the cell.

Fig. 10 Schematic illustration of electron transfer on Pd/RGO nanomaterial.

Experimental

Chemicals

Chemicals like ascorbic acid (AA), palladium chloride and N,N-dimethylformamide (DMF) were all of at least analytical grade and used as received without further purification. Water used throughout the whole experiment was ultrapure water with resistivity of no less than 18.2 MΩ cm.

Preparation of palladium nanoparticles (Pd NPs)

A typical experiment of preparing Pd NPs was performed as follows: briefly, 7 mL of 0.4 mM H₂PdCl₄ aqueous solution was first placed in a 50 mL volume beaker. Freshly prepared ascorbic acid aqueous solution (35.4 mM, 10 mL) was then added into the above solution at the rate of 1 mL min⁻¹ under vigorous stir. The color of the mixed solution changed from light yellow to brown, indicating the accumulation of Pd NPs. After stirring for 3 h at room temperature, the nanoparticles were obtained and collected by centrifugation, washed with water, and suspended in 1 mL water for further characterization.

Preparation of Pd NPs assembled RGO (Pd/RGO) nanomaterials

Graphite oxide (GO) was prepared from graphite powder based on the modified Hummers' method.¹⁶ The as-prepared GO was dispersed in DMF (0.5 mg L⁻¹), ultrasonicated for 30 min, and heated in an oil bath (153 °C) for 6 h, obtaining RGO. For the preparation of Pd/RGO nanomaterial, 1 mL of Pd NPs (0.72 mg L⁻¹) was mixed with 700 μL of RGO (0.5 mg L⁻¹) under ultrasonic treatment for 1 h. Then the product was collected by centrifugation, washed with water and suspended in 1 mL of water for further investigation.

Apparatus

Transmission electron microscopy (TEM) and high-resolution TEM (HRTEM) measurements were performed on a FEI Tecnai G2 S-Twin F20 with an accelerating voltage of 200 kV. The scanning electron microscopy (SEM) images and energy-dispersive X-ray spectroscopy (EDX) data were obtained on a FEI Helios NanoLab 600i FIB equipped with an EDX analyzer. X-ray photoelectron spectroscopy (XPS) analysis was carried out on an ESCALAB-MKII spectrometer (United Kingdom). X-ray diffraction (XRD) patterns were collected on a D/Max 2500 V/PC X-ray diffractometer using Cu (50 kV, 200 mA) radiation. The 2θ scanning angle range was 5–90° at a scanning rate of 8° min⁻¹. The Fourier transformed infrared spectra were identified using PerkinElmer FT-IR spectrometer in the range 500–4000 cm⁻¹. Raman spectra were measured with a Renishaw 2000 Raman spectrometer with a CCD detector and a holographic notch filter (Renishaw Ltd., Gloucestershire, U.K.). Radiation of 514.5 nm from an air-cooled argon ion laser was used for excitation.

Electrochemical measurements

The electrochemical properties were examined on a CHI 660E electrochemical workstation (Chenhua Instruments, Shanghai, China) with a conventional three-electrode cell, in which a modified glassy carbon electrode (GCE, 3 mm in diameter) was used as working electrode, a platinum wire as counter electrode and a Ag/AgCl (saturated KCl) electrode as reference electrode, respectively. For the fabrication of nanomaterial modified GCE, GCE was first carefully polished to mirror, and certain amount of desired solution was casted onto the GCE surface and dried in air. Then, 10 μL of Nafion (0.5 wt%) in methanol was dropped onto the surface of the modified GCE. Electrochemical measurements including cyclic voltammetry (CV) and amperometric i–t technique were all carried out in 0.5 M KOH aqueous solution with or without 0.5 M different organic molecules at the scan rate of 50 mV s⁻¹.

Computational method

The calculations were performed based on the density functional theory with the generalized gradient approximation in the form of Perdew–Burke–Ernzerhof functional implemented using MedeA VASP.¹⁷ We employed one palladium adatom in a 4 × 4 parallelogram graphene supercell with periodic boundary conditions as the model. In this study, adatoms positioned on graphene at the top of a carbon atom is labeled the top (T) site, whilst that at the middle of a carbon–carbon bond is noted as the bridge (B) site, and that at the hexagonal center site is labeled hollow (H) site, respectively.

Conclusions

Palladium nanoparticles with multi-edges and corners were fabricated employing ascorbic acid as both reductant and structure-directing agent. A series of characterizations such as TEM, SEM, XPS and XRD were performed to investigate the morphology and chemical composition of the as-prepared nanoparticles. The Pd NPs were assembled onto the surface of reduced graphene oxide, obtaining the novel electrocatalyst. The as-prepared hybrid nanomaterial was examined as anode electrocatalyst towards methanol, ethanol and formic acid electrooxidation in alkaline media. Comparing with palladium nanoparticle and commercial Pd/C catalyst, the electrocatalytic performance of the as-prepared hybrid nanomaterial was shown to be excellent which should be the contribution of the novel morphology of the nanoparticle and the enhanced conductivity and surface area contributed by reduced graphene oxide. The interaction between Pd NPs and graphene is analyzed by MedeA VASP calculation, and the possible electron transfer on the anode electrocatalyst was proposed. The capability of electrooxidizing different organic molecules might enable the nanomaterials to be employed to catalyze multi-fuel in potential applications including waste water utilization.

Acknowledgements

Financial supports by the National Natural Science Foundation of China (21401012, 21501014), Science and Technology Project of Jilin Province (20140520079JH), and Youth Fund of Changchun University of Science and Technology (XQNJJ-2013-10) is gratefully acknowledged.

Notes and references

1. (a) S. J. Guo, S. Zhang, X. L. Sun and S. H. Sun, J. Am. Chem. Soc., 2011, 133, 15354–15357; (b) L. C. He, Y. Liu, J. Z. Liu, Y. S. Xiong, J. Z. Zheng, Y. L. Liu and Z. Y. Tang, Angew. Chem., Int. Ed., 2013, 52, 3741–3745; (c) W. W. He, H. K. Kim, W. G. Warner, D. Melka, J. H. Callahan and J. J. Yin, J. Am. Chem. Soc., 2014, 136, 750–757; (d) S. Linic, P. Christopher, H. L. Xin and A. Marimuthu, Acc. Chem. Res., 2013, 46, 1890–1899; (e) S. J. Guo and S. H. Sun, J. Am. Chem. Soc., 2012, 134, 2492–2495.
2. (a) T. Vlaar, E. Ruijter, B. U. W. Maes and R. V. A. Orru, Angew. Chem., Int. Ed., 2013, 52, 7084–7097; (b) M. M. Liu, Y. Z. Lu and W. Chen, Adv. Funct. Mater., 2013, 23, 1289–1296; (c) J. J. Xu, Z. L. Wang, D. Xu, L. L. Zhang and X. B. Zhang, Nat. Commun., 2013, 4, 2438.
3. (a) A. Safavi, H. Kazemi and S. H. Kazemi, J. Power Sources, 2014, 256, 354–360; (b) H. H. Zhang, B. Liu, J. Wang, K. Feng, B. Chen, C. H. Tung and L. Z. Wu, Tetrahedron, 2014, 70, 6188–6192; (c) J. N. Zheng, S. S. Li, X. H. Ma, F. Y. Chen, A. J. Wang, J. R. Chen and F. F. Feng, J. Power Sources, 2014, 262, 270–278; (d) Z. B. Shao, W. Zhu, H. Wang, Q. B. Yang, S. L. Yang, X. D. Liu and G. Z. Wang, J. Phys. Chem. C, 2013, 117, 14289–14294.
4. (a) K. L. Huang, Z. T. Liu and C. L. Lee, Electrochim. Acta, 2015, 157, 78–87; (b) W. X. Niu, L. Zhang and G. B. Xu, ACS Nano, 2010, 4, 1987–1996; (c) Y. J. Xiong, H. G. Cai, Y. D. Yin and Y. N. Xia, Chem. Phys. Lett., 2007, 440, 273–278; (d) L. Wang and Y. Yamauchi, Chem.–Asian J., 2010, 5, 2493–2498; (e) B. Lim, H. Kobayashi, P. H. C. Camargo, L. F. Allard, J. Y. Liu and Y. N. Xia, Nano Res., 2010, 3, 180–188.
5. (a) L. Wang, Y. Wang, A. Li, Y. Yang, Q. Tang, H. Cao, T. Qi and C. Li, J. Power Sources, 2014, 257, 138–146; (b) L. Z. Li, M. X. Chen, G. B. Huang, N. Yang, L. Zhang, H. Wang, Y. Liu, W. Wang and J. P. Gao, J. Power Sources, 2014, 263, 13–21; (c) S. Zhang, S. J. Guo, H. Y. Zhu, D. Su and S. H. Sun, J. Am. Chem. Soc., 2012, 134, 5060–5063; (d) S. J. Guo, S. Zhang and S. H. Sun, Angew. Chem., Int. Ed., 2013, 52, 8526–8544; (e) D. R. MacFarlane, N. Tachikawa, M. Forsyth, J. M. Pringle, P. C. Howlett, G. D. Elliott, J. H. Davis, M. Watanabe, P. Simon and C. A. Angell, Energy Environ. Sci., 2014, 7, 232–250.
6. (a) A. L. Wang, H. Xu, J. X. Feng, L. X. Ding, Y. X. Tong and G. R. Li, J. Am. Chem. Soc., 2013, 135, 10703–10709; (b) Z. Y. Shih, C. W. Wang, G. B. Xu and H. T. Chang, J. Mater. Chem. A, 2013, 1, 4773–4778; (c) H. An, L. N. Pan, H. Cui, B. J. Li, D. D. Zhou, J. P. Zhai and Q. Li, Electrochim. Acta, 2013, 102, 79–87; (d) Y. T. Zhang, G. Chang, H. H. Shu, M. Oyama, X. Liu and Y. B. He, J. Power Sources, 2014, 262, 279–285.
7. (a) B. C. H. Steele and A. Heinzel, Nature, 2001, 414, 345–352; (b) F. Bensebaa, A. A. Farah, D. S. Wang, C. Bock, X. M. Du, J. Kung and Y. Le Page, J. Phys. Chem. B, 2005, 109, 15339–15344.
8. (a) R. Awasthi and R. N. Singh, Carbon, 2013, 51, 282–289; (b) C. Z. Zhu, S. J. Guo and S. J. Dong, Chem.–Eur. J., 2013, 19, 1104–1111; (c) J. D. Cai, Y. Y. Huang and Y. L. Guo, Electrochim. Acta, 2013, 99, 22–29; (d) Y. Wang, F. F. Shi, Y. Y. Yang and W. B. Cai, J. Power Sources, 2013, 243, 369–373; (e) M. Sawangphruk, A. Krittayavathananon and N. Chinwipas, J. Mater. Chem. A, 2013, 1, 1030–1034; (f) G. Z. Hu, F. Nitze, X. Jia, T. Sharifi, H. R. Barzegar, E. Gracia-Espino and T. Wagberg, RSC Adv., 2014, 4, 676–682.
9. (a) T. Jin, S. J. Guo, J. L. Zuo and S. H. Sun, Nanoscale, 2013, 6, 160–163; (b) L. Zhang, L. Wan, Y. R. Ma, Y. Chen, Y. M. Zhou, Y. W. Tang and T. H. Lu, Appl. Catal., B, 2013, 138, 229–235; (c) J. F. Chang, L. Feng, C. P. Liu, W. Xing and X. L. Hu, Angew. Chem., Int. Ed., 2014, 53, 122–126; (d) Z. L. Liu, B. Zhao, C. L. Guo, Y. J. Sun, Y. Shi, H. B. Yang and Z. Li, J. Colloid Interface Sci., 2010, 351, 233–238; (e) Z. L. Liu, B. Zhao, C. L. Guo, Y. J. Sun, F. G. Xu, H. B. Yang and Z. Li, J. Phys. Chem. C, 2009, 113, 16766–16771.
10. A. N. Geraldes, D. F. da Silva, E. S. Pino, J. C. M. da Silva, R. F. B. de Souza, P. Hammer, E. V. Spinace, A. O. Neto, M. Linardi and M. C. dos Santos, Electrochim. Acta, 2013, 111, 455–465.
11. J. Y. Park and S. Kim, Int. J. Hydrogen Energy, 2013, 38, 6275–6282.
12. J. B. Xu, T. S. Zhao, Y. S. Li and W. W. Yang, Int. J. Hydrogen Energy, 2010, 35, 9693–9700.
13. Y. M. Zhu, H. Uchida, T. Yajima and M. Watanabe, Langmuir, 2001, 17, 146–154.
14. Z. Yin, H. J. Zheng, D. Ma and X. H. Bao, J. Phys. Chem. C, 2009, 113, 1001–1005.
15. I. Suarez-Martinez, A. Felten, J. J. Pireaux, C. Bittencourt and C. P. Ewels, J. Nanosci. Nanotechnol., 2009, 9, 6171–6175.
16. Y. X. Xu, H. Bai, G. W. Lu, C. Li and G. Q. Shi, J. Am. Chem. Soc., 2008, 130, 5856–5857.
17. (a) X. J. Liu, C. Z. Wang, Y. X. Yao, W. C. Lu, M. Hupalo, M. C. Tringides and K. M. Ho, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 83, 235411; (b) J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.

---