Introduction

Oxygen reduction reaction (ORR) plays a key role for both metal-air batteries and low-temperature fuel cells1,2,3,4,5,6,7,8. The sluggish electron-transfer kinetics process demands high loading of active Pt catalyst which hinders large scale application of fuel cell because of the limited supply, high cost and finite lifetime of platinum9,10,11. To address these problems, the common method is to reduce Pt usage by alloying Pt with transition metal12,13,14,15,16,17. Abruna and coworkers reported that a wide range of intermetallic compounds exhibit enhanced electrocatalytic activity when compared to pure Pt18,19,20,21,22. Sun and coworkers first reported that structurally ordered PtFe is more electrocatalytic active than PtFe with chemically disordered face centered cubic structure for ORR23,24. However, the long-term stability of alloy catalysts, due to the second metal dissolution, particle growth and corrosion of the carbon support at high potential, remains a major challenging.

The durability of electrocatalysts appears one of the most important issues that has to be addressed before the commercialization of proton exchange membrane fuel cells25,26,27,28,29,30,31,32,33,34. Schuth employed the combination of highly graphitized carbon to reduce carbon corrosion and interconnected pore system in order to encapsulate Pt nanoparticles to overcome the long-term catalyst degradation35. Nørskov36,37, Chorkendorff38,39,40 and Yoo41,42,43,44 teams reported a stable cathode catalysts of Pt alloyed with early transition metals. Markovic and Adzic et al. also demonstrated stable a cathode catalysts of Pt alloyed with a 3d transition metal45,46,47,48,49,50,51,52. Considerable improvements in catalytic performance have been achieved.

In this work, we present a novel approach to develop durable Pt-based intermetallic electrocatalysts towards ORR by N-anchor-metal. In addition to provide a promising electrocatalyst candidate, this work demonstrates a novel design strategy of catalyst by N-anchor-metal, which can be extended to a wide variety of durable alloy catalysts.

The supported N-containing intermetallic N-Pt3Fe1 nanoparticles were synthesized by a simple two-stage approach. At first, supported chemically disordered Pt3Fe1 nanoparticles were prepared via ultrasonic-assisted electroless deposition in a mixed solution of ethylene glycol (EG)/H2O without using surfactant. Subsequently, the supported N-containing intermetallic compound N-Pt3Fe1 nanoparticles were obtained via annealing of the as-prepared supported chemically disordered Pt3Fe1 nanoparticles under NH3 atmosphere at 873 K for 3 hours. To evaluate the N-anchor effect in N-Pt3Fe1/C, the letter were also prepared via annealing of the obtained supported chemically disordered Pt3Fe1 nanoparticles under 95 vol%Ar + 5 vol%H2 atmosphere.

The crystal structure of products was characterized with X-ray techniques. Figure 1 shows the X-ray diffraction (XRD) patterns of as-prepared Pt3Fe1/C, intermetallics Pt3Fe1/C and N-containing intermetallics N-Pt3Fe1/C, respectively. The XRD pattern of the as-prepared Pt3Fe1/C displays the distinct faced centered cubic pattern associated to chemically disordered Pt solid solution structure. After annealed under 95 vol%Ar + 5 vol%H2 atmosphere, the structure was converted from chemically disordered structure (A1 phase) to chemically ordered structure (L12 phase, space group: Pm-3m). XRD patterns of powder obtained via annealing of the as-prepared Pt3Fe1/C under NH3 atmosphere suggest they have a chemically ordered Pt3Fe1 faced centered cubic structure with the Pm-3m space group, similar to powders obtained via annealing of as-prepared Pt3Fe1/C under Ar + H2 atmosphere. However, the diffraction peaks of N-Pt3Fe1/C are slightly shifted to lower angles compared to those of Pt3Fe1/C (Figure S1). The observation is related to the expansion of lattice as a result of nitrogen incorporated into intermetallic Pt3Fe1 structure. The crystal structure from chemically disordered to chemically ordered during annealing is shown in figure 2. To verify the phase transformation during the annealing process, X-ray absorption spectroscopy (XAS) experiments were also performed. The results shown in figure S2 further verify the occurrence of a structural phase transition and the formation of an ordered structure (L12 phase).

Figure 1
figure 1

XRD patterns of as-prepared Pt3Fe1/C, intermetallics Pt3Fe1/C and N-containing intermetallics N-Pt3Fe1/C.

Vertical lines show the peak positions of chemically ordered intermetallics Pt3Fe1(JCPDS card No.89-2050).

Full size image
Figure 2
figure 2

Illustration of phase transition during annealing process.

Full size image

Transmission electron microscopy (TEM) images of the as-prepared nanoparticles of supported chemically disordered and ordered Pt3Fe1 are shown in figure S3. The average size of as-prepared Pt3Fe1 nanoparticles was ca. 2 nm. After annealing under reductive atmosphere, the size of intermetallics Pt3Fe1 nanoparticles increased to ca. 5 nm. TEM images of supported N-containing intermetallic N-Pt3Fe1 nanoparticles are shown in figure 3. The N-containing intermetallic N-Pt3Fe1 nanoparticlse with diameters of ca.5 nm are highly dispersed on XC-72 carbon black.

Figure 3
figure 3

TEM images of supported N-containing intermetallic compound N-Pt3Fe1 nanoparticles obtained at 873 K for 3 hours.

Full size image

Figure 4a shows N 1s X-ray photoelectron spectra (XPS) of N-Pt3Fe1/C while no N 1s signal can be collected in the intermetallic Pt3Fe1/C. The N 1s peak can be deconvoluted into three peaks. The peak located in ca.398.2 eV can be assign to N which interacts with intermetallics Pt3Fe1 and the other two peaks can be assign to N which interact with oxygen species according Fairbrother's works53. Figure 4b and 4c show Pt 4f spectra of intermetallic Pt3Fe1/C and N-containing intermetallic N-Pt3Fe1/C, respectively. Each Pt 4f peak can be deconvoluted in two pairs of doublets. The doublet peaks of labelled 1 and 1′ are generated by photoelectrons emitted from Pt(0) while the other doublet peaks of labelled 2 and 2′ are generated by photoelectrons emitted from Pt(II). The smaller amount of Pt(II) is observed in the N-containing intermetallic N-Pt3Fe1/C. XPS spectra of Fe in the N-Pt3Fe1/C also displays an enhanced intensity of peak at low energy, suggesting a decreased contribution of the higher oxidation state Fe species(figure S4). XPS results indicate also that the introduction of nitrogen enhances the oxidation resistance of the N-Pt3Fe1. We claim it as the N-anchor effect. As to the origin of the enhanced corrosion tolerance, potentiodynamic polarization was employed to evaluate the corrosion behavior of N-Pt3Fe1/C, as shown in figure S5. The corrosion potential of the N-Pt3Fe1/C is higher than that of Pt3Fe1/C and the corrosion current of the N-Pt3Fe1/C is lower than that of Pt3Fe1/C. That means the N-Pt3Fe1/C could show a good durability during ORR process. Furthermore, figure S2 shows the Pt L3 edge XANES spectra. The intermetallic N-Pt3Fe1/C exhibits a decreased Pt L3-edge white line intensity compared to the intermetallic Pt3Fe1/C. The L3 edge XANES spectroscopy at the Pt originates from the electron excitation from core 2s to 5d unoccupied state. The decrease in the white line intensity reflects the decreased number of unoccupied d-states of Pt in the N-Pt3Fe1/C catalyst, implying the high resistance to be oxidized for Pt.

Figure 4
figure 4

XPS spectra(a, b, c, d) of as-prepared Pt3Fe1/C(C 1s), intermetallics Pt3Fe1/C (Pt 4f and C1s) and N-containing intermetallics N-Pt3Fe1/C (Pt4f, N 1s and C1s).

Full size image

Previous works showed that doping nitrogen into a carbon support by nitrogen ion beam or ammonia reaction at 1173 K significantly impedes Pt nanoparticles migration and coarsening54,55,56,57. In this work, to investigate the interaction of NH3 with the support of XC-72 at 873 K, we measured C1s XPS spectra of as-prepared Pt3Fe1/C, intermetallic Pt3Fe1/C and N-containing intermetallic N-Pt3Fe1/C, as shown in figure 4d. It clearly shows that there is no evidence of interaction between nitrogen and carbon.

To further characterize N in N-Pt3Fe1/C, N K-edge XAS spectrum was also measured. Figure 5 compares the calculated theoretical and experimental spectra, which present four significant features marked with vertical dashed lines. Using the “fingerprint” of the N K-edge XAS, we may show that the simulated spectrum of N-Pt3Fe1 with the N atom in a tetrahedral site matches the raw spectrum.

Figure 5
figure 5

Experimental N K-edge XAS spectra of N-Pt3Fe1/C and calculated octahedral-N-Pt3Fe1 and tetrahedral-N-Pt3Fe1.

Full size image

In figure 6a we compare typical CV curves of commercial Pt/C (Johnson Matthey HiSPEC 3000), chemically ordered Pt3Fe1/C and N-Pt3Fe1/C in Ar-saturated 0.1 M HClO4. All the Pt-based electrocatalysts show the region of H-adsorption and H-desorption in the potential range of 0.05 to 0.40 V, the double-layer capacitance region located from 0.40 to ca. 0.60 V and the region of Pt oxidation and Pt-oxide reduction in the range of ca. 0.60–1.20 V. The electrochemical surface area (ECSA) of the electrocatalysts has been calculated by integrating H-desorption charges, a method used to normalize the kinetic current density to evaluate the intrinsic electrocatalytic activity of Pt-based electrocatalysts. The CV curves did not show any anodic currents ascribed to the oxidation/dissolution of Fe, demonstrating that Fe is stabilized by N-anchor. The EDS line scanning and mapping results, as shown in figure S6, point out that the structure of the N-Pt3Fe1/C nanoparticle is intact and no structural transformation after potential cycles from 0.6 to 1.2 V in O2-bubbling perchloric acid solution, further witnessing the good durability of the catalyst.

Figure 6
figure 6

(a) Cyclic Voltammograms and bar plots of ECSA of the commercial Pt/C, Pt3Fe1/C and N-Pt3Fe1/C. (b) plots of kinetic current density (normalized by ECSA) versus cycle number for the commercial Pt/C, Pt3Fe1/C and N-Pt3Fe1/C at 0.9 V.

Full size image

The electrocatalytic activity for ORR was evaluated using rotating disk electrode in O2-saturated 0.1 M HClO4 at room temperature. In order to compare the specific activity for different electrocatalysts according to the Levich-Koutecky equation, the kinetic current was calculated from the polarization curve (Figure S7) by considering the mass-transport correction and normalized with respect to electrochemical active surface area. Both Pt3Fe1/C and N-Pt3Fe1/C show a higher activity towards ORR than that of the commercial Pt/C at the potential of 0.9 V. Their enhanced activities were due to their chemically ordered Pt3Fe1 intermetallic compound structure17,23. Accelerated durability tests (ADT) were performed by cycling the potential between 0.6–1.2 V (vs. NHE) in O2-bubbling 0.1 M HClO4 at a scan rate of 200 mV s−1. After ADT tests, the specific activities of electrocatalysts were recorded as shown in figure 6b. It has been shown that N-Pt3Fe1/C showed the best durability after 5000 cycles. The kinetic current densities at 0.9 V were only decreased ca. 7% for the N-Pt3Fe1/C after 20000 potential cycles, whereas the Pt3Fe1/C shows an activity degradation of 49%. The degradation of specific activity of the intermetallic Pt3Fe1/C is mainly due to the destruction of intermetallics structure deduced by the leaching of Fe during ADT tests. We assigned the enhanced durability of N-Pt3Fe1/C vs. Pt3Fe1/C to the N-anchor effect which could promote corrosion resistance of Pt3Fe1 intermetallics.

This work has highlights a novel strategy to promote the durability of intermetallic Pt3Fe1 electrocatalyst. Supported nitrogen-containing intermetallic N-Pt3Fe1 electrocatalyst was synthesized via a facile two-stage approach. Actually, the N-Pt3Fe1/C electrocatalyst shows stability under ADT tests and only show a 7% specific activity loss after 20000 potential cycles from 0.6 to 1.2 V (vs. NHE) in O2-bubbling perchloric acid solution. The superior durability of the N-Pt3Fe1/C is assigned to the N-anchor effect that could promote corrosion resistance of electrocatalysts. This strategy of durability and activity enhancement towards ORR could be applied to the design of other alloy electrocatalysts for fuel cells.

Methods

Synthesis of as-prepared Pt3Fe1/C

As-prepared Pt3Fe1/C was synthesized by one-pot reduction of Pt and Fe inorganics in EG/H2O mixed solution in an ultrasonic cleaning bath. 0.3 mmol H2PtCl6 6H2O and 0.1 mmol FeSO4 7H2O were dissolved into 200 mL EG/H2O mixed solution (EG: H2O = 1:1 in volume ratio, containing 6 mL concentrated sulfuric acid) in a three-neck flask. 234.1 mg XC-72 was then added to the flask above, yielding Pt loadings of ca.20% in weight. After the mixed solution was vigorous stirred in the ultrasonic cleaning bath for 2 hours at room temperature, the temperature of the mixed solution was increased to 65°C and 5 mol L−1 NaOH solution was added to adjust pH to 11. 40 mmol sodium hypophosphite (NaH2PO2 H2O) was then added into the mixture solution. The process of the reaction is under the protection of N2. After 10 hours reaction, the obtained supported Pt-Fe nanoparticles were then filtered, washed copiously with water and dried at 80°C overnight.

Preparation of intermetallics Pt3Fe1/C

To obtain intermetallics Pt3Fe1/C, as-prepared Pt3Fe1/C was put in a quartz tube and annealed at the temperature of 873 K for 30 min under mixture gas of 95%Ar + 5% H2.

Preparation of N-containing intermetallics N-Pt3Fe1/C

To obtain N-containing intermetallics N-Pt3Fe1/C, as-prepared Pt3Fe1/C was put in a quartz tube and annealed at the temperature of 873 K for 3 hours under NH3 atmosphere.

Characterization

The X-ray diffraction (XRD) patterns of the samples were obtained using a Bruker D8 Advance diffractometer with Cu Kα(λ = 1.5405Å) radiation source (40 kV, 40 mA). The morphology, structure and component of nanoparticles were investigated on FEI Tecnai G2 F30 field-emission transmission electron microscope (TEM) and FEI Titan G2 80–200 Probe Cs-corrector Scanning transmission electron microscope (STEM). The bulk composition of the prepared catalysts was measured using the inductively coupled plasma-atomic emission spectrometry (ICP-AES) on an IRIS Intrepid spectrometer after dissolution of the samples in aqua regia and then dilution using 1 M HCl. X-ray photoelectron spectroscopy (XPS) data were collected on an AXIS-Ultra instrument from Kratos Analytical using monochromatic Al Kα radiation (hν = 1486.6 eV) and low-energy electron flooding for charge compensation. To compensate for the effects of surface charges, the binding energies were calibrated using the C 1s hydrocarbon peak at 284.80 eV. The Shirley method was used to correct the background of all spectra. After making the corrections, the spectra were analyzed using the XPSPEAK 4.1 software package. To deconvolute the spectra, we used a convolution of a mixed Gaussian-Lorentzian function series corrected by an asymmetric component to reflect the many-body effects in metal. Pt L3-edge X-ray absorption fine structure (XAFS) spectra of intermetallics N-Pt3Fe1/C and N-containing intermetallics N-Pt3Fe1/C, as well as Pt L3-edge XAFS spectrum of Pt foil as a reference compound, were measured in fluorescence mode at the Beijing Synchrotron Radiation Facility (BSRF). X-ray absorption spectra were background-subtracted and then normalized to the high energy atomic absorption. To allow comparisons to be made, the edge energy was subtracted and defined as energy zero for the Pt L3-edge XANES spectra.

The electrochemical measurements were performed using a VMP3 multichannel potentiostat/galvanostat (Bio-Logic SAS, France) at a constant temperature of 25°C unless stated otherwise. The glassy carbon rotating disk electrode (GC-RDE) with an area of 0.1257 cm2 was used as the working electrode. A GC film electrode and a normal hydrogen electrode (NHE) were used as the counter electrode and the reference electrode, respectively. All the potentials in this study were given on the reference of NHE. Before using a GC electrode as a substrate for the catalysts, it was polished with 0.05 μm alumina to yield a mirror finish. In order to deposit the electrocatalysts on the working electrode, we prepared the ink as follows: Typically ca. 3 mg of the electrocatalyst was dispersed in isopropyl alcohol (1 mL) together with one drop of 2% Nafion solution in an ultrasonic bath for 20 min. The suspension (10 μL) was pipetted on to the GC substrate and dried in an air oven for 30 min.

To obtain the corrosion behavior of electrocatalysts, the potentiodynamic polarization measurement was conducted in O2-saturated 0.1 M HClO4 solutions. Then the potential curves was obtained plotting the potential as a function of the logarithm of the current density.

Cyclic voltammetry (CV) measurements were carried out in Ar-saturated 0.1 M HClO4 solutions at 50 mV s−1. The electrochemical surface area (ECSA) was estimated by measuring the charge associated with hydrogen desorption (after double layer correction) between 0.05 and 0.40 V by taking the conversion factor to be 210 μC cmPt−2. The ORR measurements were performed in O2-saturated 0.1 M HClO4 solutions using GC-RDE at a rotation of 1600 rpm and a sweep rate of 10 mV s−1. The kinetic currents for ORR on GC-RDE were calculated using the Koutecky-Levich equation (1) from the ORR polarization:

Where j is the experimentally measured current density, jd is the diffusion limiting current density and jk is the kinetic current density.

The ADT tests were performed in O2-bubbling 0.1 M HClO4 solutions by applying cyclic potential sweeps between 0.6 and 1.2 V at a sweep rate of 200 mV s−1 for the given number of cycles.