Highly Durable Graphene Nanosheet Supported Iron Catalyst for Oxygen Reduction Reaction in PEM Fuel Cells

To prepare graphene nanosheet supported NPMCs using PTAm as a nitrogen precursor and iron acetate (FeAc) as a metal precursor, a desired amount of graphene and PTAm was added to 4M HCl solution, and the solution was subjected to low temperature heat treatment at 80°C for 24 h while stirring. The proposed coupling reaction between the support and PTAm is illustrated in Figure 1a.¹⁷ The mixture was filtered, washed with deionized water and dried overnight in an oven at 60°C. The dried sample was then ground to a fine powder and ultrasonically mixed with 1 wt.% FeAc in 50 mL of ethanol for one hour. The solution was again filtered and dried in an oven. The remaining dried precipitate was pyrolyzed at various temperatures ranging from 700 to 1000°C for one hour under nitrogen gas flow. These catalysts are referred as FPGNP (non-pyrolyzed), FPG700, FPG800, FPG900 and FPG1000 accordingly.

Zoom In Zoom Out Reset image size

The electrocatalytic activity was evaluated in a three-electrode electrochemical cell using a Pine-Instrument's Bipotentiostat AFCBP1, equipped with a speed rotator. All RDE measurements were performed in acidic 0.1 M HClO₄ electrolyte at room temperature, using a Ag/AgCl reference electrode. To prepare the glassy carbon working electrode (0.19635 cm²), 4 mg of catalyst sample was ultrasonically dispersed in 1 mL of ethanol. 30 μL of the ink and 5 μL of 0.5 wt.% Nafion solution were applied to the glassy carbon disk, leading to identical electrode catalyst loadings for each sample during testing. ORR curves were recorded at a scan rate of 10 mV/s with the electrolyte saturated with oxygen gas under various electrode rotation speeds (100, 400, 900 and 1600 rpm). ORR curves were corrected by removing background currents obtained under the same testing conditions in nitrogen saturated electrolyte. Transmission electron microscopy (TEM) was utilized to observe the surface morphology, and X-ray photoelectron spectroscopy (XPS) was conducted to determine surface atomic compositions and configurations.

The morphology of FPG800 observed by TEM is displayed in Figure 1b. Many small graphene nanoplatelets were observed clustered together, possibly resulting from the linkage between PTAm molecules and adjacent graphene sheets, since PTAm contains many nitrogen sites that can be coupled to the edge plane of the graphene and the functional groups.¹⁷ These adjacent graphene nanosheet clusters can serve to host one of the most promising metal-nitrogen active sites structures upon pyrolysis.¹⁸

Half-cell RRDE experiments were conducted for all samples to determine their catalytic activity for the ORR, with polarization curves plotted in Figure 2a at a rotation rate of 900 rpm. From these results, it is clear that a pyrolysis step is crucial, where FPGNP demonstrated negligible ORR activity. FPG800 displayed optimal activity, demonstrating an onset potential of 0.853 V and a half-wave potential of 0.692 V (vs. RHE) and full RRDE data at various rotation rates provided in Figure 2b. At pyrolysis temperatures above or below 800°C reduced onset potentials and current densities were observed. Moreover, the H₂O selectivity of these samples was determined from ring current data, with high selectivity values providing indication of a more efficient 4 electron reduction pathway compared to the inefficient 2 electron reduction forming destructive H₂O₂ species. Figure 3 provides H₂O selectivity of all four pyrolyzed graphene based catalysts with the order of highest to lowest H₂O selectivities being FPG800 > FPG900 > FPG 1000 and FPG700, consistent with ORR activity results. Specifically, FPG800 shows the best H₂O selectivity of 99.9% at 0.4 V vs. RHE, indicating negligible selectivity towards the two electron reduction pathway. The results of the onset potentials, current densities obtained at 0.6V vs. RHE and H₂O selectivities are tabulated in Table 1 for all synthesized catalysts.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

To evaluate the effect of utilizing graphene as the NPMC support, the same type of catalyst was fabricated using Ketjen 600JD carbon black support. This porous carbon has been previously established as a good electrocatalyst support and thus serves as a good performance evaluation benchmark. Catalyst synthesis conditions were identical with the NPMC, with the exception of the carbon support material utilized. Pyrolysis was carried out at the same temperatures as well; however, the activity of only the highest performing material pyrolyzed at 900°C is illustrated in Figure 3a. From the presented curves, it can clearly be seen that the ORR performance of the graphene based catalysts exceeds that of the ketjen black based catalyst for this type of catalyst, which effectively demonstrates the promising potential for the graphene as a catalyst support.

The durability of FPG800 was investigated by accelerated degradation testing (ADT) consisting of 1000 cycles in N2 saturated electrolyte at a scan rate of 50 mV/s between 0 and 1.2 V vs. RHE. Durability data is provided in Figure 4 along with commercial carbon supported platinum (Pt/C) for comparison. Surprisingly, FPG800 showed only a slight decrease of 0.055 V in the half-wave potential while Pt/C showed a much higher potential decrease of 0.125 V. Moreover, the current densities obtained at 0.2 V vs. RHE before and after the cycles for FPG800 showed almost no difference (0.016 mA/cm²) while the difference for Pt/C was 0.055 mA/cm². The resulting current density at 0.2 V vs. RHE for FPG800 following ADT was approximately 10% higher than that of the commercial catalyst. Overall, FPG800 showed superior durability than that of the commercial carbon supported platinum catalyst under the studied conditions. This provides indication of the promising stability of these materials most likely arising due to the high graphitic content.¹⁵

Zoom In Zoom Out Reset image size

XPS was utilized to determine the surface atomic composition and chemistry of the catalyst samples. XPS confirmed the presence of a variety of carbon, and nitrogen species, along with trace amounts of iron as illustrated in Figure 5. A decreasing trend was observed for the surface nitrogen content with increasing pyrolysis temperatures, consistent with previous reports indicating decomposition of nitrogen species at elevated temperatures.¹⁹ High resolution N1s spectra along with a detailed breakdown of the contributing species obtained after peak de-convolution are provided in Figure 6 and 7, respectively. It should be noted that the scale for the intensity obtained for each catalyst has been adjusted for better visibility in Fig. 6. Specifically, the N1s spectra was broken down into three peaks representing pyridinic (398.3 eV), pyrrolic (400.5 ∼ 401 eV), and quaternary (401 ∼ 403 eV) nitrogen functionalities.²⁰ Following pyrolysis at 700°C, the relative amount of pyridinic nitrogen increased to 59.41 at.% of all the nitrogen groups, and dropped slightly to 54.79 at.% for the sample pyrolyzed at 800°C. By further increasing the final pyrolysis temperature, the relative amount of pyridinic nitrogen decreases, as these species are relatively unstable at temperatures exceeding 800°C.²¹ The slight decrease in the relative amount of pyridinic nitrogen from 700 to 800°C, combined with a stable iron surface content observed from 700 to 800°C despite significant changes in the observed ORR activity can possibly indicate specific active site formation, or a transition of the active site from FeN₄/C to FeN₂₊₂/C structures. It has been reported that, while FeN₄/C is the most prominent active site at lower pyrolysis temperatures FeN₂₊₂/C catalytically active sites arise in the range of 700 to 800°C and reach the highest concentration at 800°C.²² It has also been reported previously for catalysts using FeAc as the iron precursor and with low iron content (less than 0.2 wt%), FeN₂₊₂/C active sites dominate the catalytic behavior, while forming better electrical contact with the conductive graphene plane.^22,23 Moreover, as it can be seen in Figure 7 the relative amount of pyrrolic and quaternary nitrogen increases with increased pyrolysis temperature, as these forms of nitrogen are stable at higher temperatures.²¹ Thus, for this particular NPMC, there is no correlation between these two nitrogen groups and the performance. Thus, the catalyst performance reaches the highest at the temperature of 800°C and drops with increasing temperature due to both FeN₂₊₂/C active site and the overall nitrogen content decrease.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

We have successfully synthesized a novel NPMC using PTAm as a nitrogen precursor, FeAc as iron precursor and graphene nanosheets as a catalyst support. This NPMC demonstrates promising electrocatalytic activity and durability for the ORR, rendering graphene nanosheets as a suitable replacement to traditional nanostructured carbon support materials. A pyrolysis procedure was deemed crucial for active site formation, where the NPMC sample heat treated at 800°C was found to display optimal ORR activity and H₂O selectivity, specifically, an onset potential of 0.853 V vs RHE, a half-wave potential of 0.682 V vs RHE, and a H₂O selectivity of ca. 99.9%. Moreover, high stability through ADT was demonstrated most likely due to the high graphitic content of the catalyst support material. Thus, graphene nanosheets are presented as an ideal catalyst support material, with promising ORR activity and stability demonstrated. Future investigations will focus on catalyst optimization by varying iron precursor loading and synthesis conditions, which have previously been deemed important factors influencing the ORR activity.

This work was financially supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) and the University of Waterloo.