Carbon 107 (2016) 667e679

Contents lists available at ScienceDirect

Carbon
journal homepage: www.elsevier.com/locate/carbon

Platinum nanoparticles supported on nitrogen and fluorine co-doped

graphite nanofibers as an excellent and durable oxygen reduction
catalyst for polymer electrolyte fuel cells
S. Gouse Peera, A. Arunchander, A.K. Sahu*
CSIR e Central Electrochemical Research Institute, Madras Unit, CSIR Madras Complex, Taramani, Chennai 600 113, India

a r t i c l e i n f o a b s t r a c t

Article history: Nitrogen and fluorine co-doped graphite nanofibers (NFeGNFs) are explored as highly corrosion-
Received 29 February 2016 resistant support materials and investigated for their possible interaction with Pt as an efficient oxy-
Received in revised form gen reduction reaction (ORR) catalyst in an acidic medium. The Pt/NFeGNF catalyst exhibits enhanced
7 June 2016
ORR behavior with greater electrochemical stability, which is exclusively attributed to the simultaneous
Accepted 9 June 2016
presence of N and F heteroatoms, which induces charge delocalization in the GNF matrix based on the
Available online 22 June 2016
electronegativity difference between carbon and heteroatoms. This facilitates a uniform deposition of Pt
nanoparticles over NFeGNFs and interaction between Pt nanoparticles and the GNF support matrix. The
Pt/NFeGNF catalyst was subjected to 10,000 repeated potential cycles between 0 and 1.05 V in acidic
media, resulting in a 30-mV loss in ORR activity, compared to the activity loss of about 200 mV for
commercial Pt/C catalysts. The carbon corrosion and catalyst stability are also determined with more
stringent corrosive test cycles, that is, maintaining the potential of the cell as high as 1.4 V for 180 s and
later decreasing it to 0.6 V for 30 s versus reversible hydrogen electrode (RHE) for 500 cycles, mimicking
the actual conditions of operating fuel cells. Under these harsh corrosive conditions, the Pt/NFeGNF
catalyst displayed excellent stability with only a 90-mV loss in E1/2 in relation to the 270-mV loss in E1/2
for commercial Pt/C catalysts. While evaluating the performance of the Pt/NFeGNF catalyst in a polymer
electrolyte membrane fuel cell (PEMFC), a peak power density of 867 mW cm
2 was achieved with a
minimal Pt loading of 0.1 mg cm
2, which demonstrates the potential application of this catalyst as a
cathode catalyst in fuel cell reactions.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction decrease in catalytic activity and PEMFC performance. Acidic

environment, oxygen atmosphere, and sudden increase in cathodic
Oxygen reduction reaction (ORR) usually occurs at the cathode potential >1.2 V during the start-up/shutdown process greatly in-
of a polymer electrolyte fuel cell (PEFC), which is catalyzed by high crease the carbon corrosion, by which Pt nanoparticles agglomerate
Pt loading to minimize the over potential loss. In order to reduce and detach from the carbon support, thereby decreasing catalytic
the weight, volume, and cost of a PEMFC stack, several studies have activity [2e6].
focused on lowering the Pt amount in PEMFC electrodes [1]. Pt To date, the most popular support material used has been
nanoparticles supported on carbon (Pt/C) represents a state-of-the- Vulcan XC-72 carbon due to its suitable specific surface area and
art ORR catalyst that essentially contributes to the overall efficiency high electronic conductivity. However, due to several shortcomings
of the fuel cell. However, degradation of the Pt/C catalyst, as a in terms of durability of this support material, studies have focused
consequence of carbon corrosion, is primarily responsible for the on developing alternative catalyst supports, such as carbon nano-
tubes [7e9], carbon nanofibers [10e17], mesoporous carbon
[18e22] and graphene alone or in combination with other doped
atoms, to enhance both the catalytic activity and durability
* Corresponding author. [23e26]. Among all carbon supports discussed here, graphite
E-mail addresses: [email protected], [email protected]
nanofibers (GNFs) are promising in terms of their unique fibrous
(A.K. Sahu).

http://dx.doi.org/10.1016/j.carbon.2016.06.021
0008-6223/© 2016 Elsevier Ltd. All rights reserved.
668 S.G. Peera et al. / Carbon 107 (2016) 667e679

structure, high degree of graphitization, and the tendency of 2. Experimental section

interaction of metal nanoparticles with GNF surface during catalyst
synthesis [27]. Prior to the impregnation of Pt nanoparticles, GNFs 2.1. Materials
are usually surface-functionalized by strong oxidizing acids to
introduce surface-bound polar hydroxyl and carboxylic acid groups GNFs (diameter 80e200 nm, length 5e20 mm) were purchased
on the GNF surface [28]. However, the acid oxidation inevitably from Sigma-Aldrich. Chloroplatinic acid, ethylene glycol (EG),
leads to problems such as structural damage, loss in electrical ethanol, and isopropyl alcohol (IPA) were obtained from Acros Or-
conductivity, and carbon corrosion of GNFs due to the highly ganics. Pt/C (40 wt% Pt on Vulcan XC-72 R carbon) was obtained
oxidized surface. from Alfa Aesar (Johnson Matthey Ltd.). All chemicals were used as
One of the best ways to functionalize GNF and to improve the received. Deionized (DI) water (18.4 MU cm) used for the experi-
interaction of metal precursors is by doping with heteroatoms, such ments was produced by a Millipore system.
as nitrogen (N), phosphorous (P), boron (B), sulfur (S), and fluorine
(F), into the GNF matrix [29e31]. Heteroatom doping to the GNF 2.2. N doping and NeF co-doping to GNF and deposition of Pt
matrix induces catalytic active centers and the resulting GNFs are nanoparticles
often considered promising candidates as metal-free ORR catalysts.
The mechanism involves the electronegativity difference between N-doped GNF (N/GNF) and NeF-co-doped GNF (NF/GNF) cata-
the heteroatom and GNF, which creates atomic charge density/spin lysts were synthesized by a process similar to our previous report
density distribution on adjacent carbon atoms serving as ORR [38]. Melamine (C6H3N6) and ammonium fluoride (NH4F) were
active sites [32]. This strong electronegativity difference in the GNF considered as the nitrogen and fluorine source, respectively, for the
matrix also facilitates metalesupport interaction and acts as a preparation of doped catalysts. For the N/GNF catalyst, the mass
tapping state to help prevent Pt detachment. Especially, N doping ratio of GNF and C6H3N6 was optimized to 1:20; for the NeF-co-
effectively prevents the agglomeration of Pt nanoparticles through doped GNF catalyst, the optimized amount of C6H3N6 and NH4F was
the formation of NePt bonds with the support [31,33e35]. More- as reported earlier. The resulting samples were pyrolyzed at
over, fluorination of the GNF induces CeF bonds, which modify the 1000  C for 1 h under a N2 atmosphere.
electronic density of surrounding carbon atoms and also increase Pt nanoparticles were deposited on pristine GNF (Pt/GNF) and
the thermal and electrochemical stability of Pt nanoparticles under heteroatom-doped GNF catalysts (Pt/NeGNF and Pt/NFeGNF) by
fuel cell operating conditions [36,37]. Very recently, we have syn- the EG reduction method. Briefly, GNF, NeGNF, and NFeGNF sup-
thesized N and F co-doped graphene catalyst (NeF/GNF) derived ports were separately suspended in EG solution and ultrasonicated
from GNF as a metal-free catalyst and found enhanced activity to- to form an ink. Chloroplatinic acid dissolved in EG solution was
wards ORR in acidic electrolyte without any loss in its catalytic added dropwise to this prepared ink under vigorous stirring for 2 h.
activity up to 20,000 repeated potential cycles [38]. F doping into The pH was adjusted between 9 and 10 by adding aqueous 0.5 M
N-doped GNFs structurally transforms GNFs into graphene sheets NaOH solution drop wise, and the mixture was heated at 140  C in
with many open-edge active sites suitable for adsorption and an oil bath for 6 h for the complete reduction of Pt precursors. The
subsequent reduction of O2 molecules. The existence of pyridinic N, product was filtered and washed copiously with DI water. The
graphitic N, and semi-ionic CeF bonds in the NeF/GNF catalyst was resulting Pt/GNF, Pt/NeGNF, and Pt/NFeGNF catalysts were dried at
found to enhance the ORR activity and deliver a peak power density 80  C under vacuum for 12 h. Pt loadings of ~20 wt% were achieved
of 165 mW cm
2 in a real H2eO2 fuel cell at 80  C and 2-bar for all the catalysts.
pressure in our earlier work [38].
Due to the promising ORR activity and durability of the NeF/ 2.3. Physico-electrochemical characterizations
GNF catalyst, the present study focuses on further enhancing the
catalytic activity and PEMFC power output by impregnating Pt Powder X-ray diffraction (XRD) patterns for all the Pt-supported
nanoparticles in the NeF/GNF catalyst and using very less Pt catalysts were obtained on a Philips Pan Analytical X-ray diffrac-
loading in the cathode electrode. Pt was impregnated in N/GNFs tometer using CuKa radiation of wavelength 1.54 Å. X-ray photo-
and NF/GNFs, and the catalytic behavior along with the PEMFC electron spectroscopy (XPS, Thermo Fisher Scientific, ESCALAB 250
performance was investigated in detail. Among all the catalysts XPS system) with a monochromated Al Ka source at 15 keV and
developed, the Pt/NFeGNF catalyst exhibits the greatest electro- 150 W was used to identify the presence of heteroatoms and the
chemical active surface area (ECSA), kinetic current, and mass ac- binding energy (BE) of different species in the catalysts with their
tivity, which is exclusively attributed to the synergistic effect of N interaction to Pt nanoparticles. Pt particle size and distribution in
and F co-doped into the GNF surface and their specific interaction the catalysts were analyzed using a transmission electron micro-
with the Pt nanoparticles. The Pt/NFeGNF catalyst was subjected to scope (TEM) (Model: Tecnai 20 G2). For these measurements, the
10,000 repeated potential cycles in acidic media (aqueous 0.5 M samples were suspended in acetone with ultrasonic dispersion for
HClO4 solution), showing only a 30-mV loss in ORR activity, 3 min. Subsequently, a drop of the suspension was deposited on a
compared to the activity loss of about 200 mV in the case of com- carbon-coated copper grid and allowed to dry. Morphological im-
mercial Pt/C catalysts. More stringent corrosive test cycles were ages of Pt/NFeGNF catalysts and mapping of the elements were
also performed maintaining the cell at a higher potential of ~1.4 V also studied under a high-resolution scanning electron microscope
for 180 s and later decreasing it to 0.6 V for 30 s versus reversible (Model: FEI Quanta FEG 200).
hydrogen electrode (RHE) for 500 cycles to investigate carbon Electrochemical tests on the catalysts were performed using a
corrosion and the suitability of the catalyst; the Pt/NFeGNF catalyst rotating-disk electrode (RDE) apparatus connected to a potentiostat
was found to have excellent durability compared to commercial Pt/ (Biologic, model VSP/VMP 3B-20) at 25  C. A glassy carbon (GC) disk
C catalysts. While evaluating the performance of the Pt/NFeGNF with a geometric area of 0.071 cm2 was used as a substrate. A
catalyst in the PEMFC, a peak power density of 867 mW cm
2 at a catalyst (Pt/GNF, Pt/NeGNF, and Pt/NFeGNF) suspension was ob-
load current density of 2008 mA cm
2 was achieved with a minimal tained by adding 2 mg of the catalyst to 500 mL of water followed by
Pt loading of 0.1 mg cm
2 at a cell temperature of 70  C and 2-bar ultrasonication for 30 min. A 3.5-mL aliquot of the above dispersed
absolute pressure. suspension was transferred onto the GC disk to obtain a Pt loading
of 20 mg cm
2. Then, 5 mL of 0.05 wt% Nafion solution was dropped
 S.G. Peera et al. / Carbon 107 (2016) 667e679 669

Pt/NF-GNF further sonicated for 1 h and coated on top of the cathode electrode
C1s O1s until a Pt loading of 0.10 mg cm
2 was obtained. For the anode,
N1s F1s
20 wt% commercial Pt/C was used with 7 wt% Nafion until a Pt
Pt 4f
loading of 0.10 mg cm
2 was achieved. MEAs were obtained by
sandwiching the pretreated Nafion-212 membrane between the
Intensity (a.u.)

cathode and the anode, which were then subjected to hot

C1s compaction under a pressure of 20 kg cm
2 at 130  C for 3 min [39].
N1s O1s Pt/N-GNF
MEAs were coupled with Teflon gas-sealing gaskets and placed in
Pt 4f single-cell test fixtures with a parallel serpentine flow field
machined on graphite plates. The galvanostatic polarization data
were obtained using a fuel cell test station (Biologic: FCT-150 S)
C1s with gaseous H2 (80% RH) and gaseous O2 (80% RH) to the anode
and cathode of the polymer electrolyte fuel cells (PEFCs) at a flow
O1s Pt/GNF rate of 525 and 650 mL min
1, respectively. All the MEAs were
Pt 4f evaluated with an active area of 25 cm2 in PEFCs at a cell temper-
ature of 70  C with an absolute pressure of 2 bar on both the anode
and cathode sides.
0 100 200 300 400 500 600 700 800
Binding Energy (eV) 3. Results and discussion

Fig. 1. XPS survey spectra for Pt/GNF, Pt/NeGNF, and Pt/NFeGNF catalysts.
Fig. 1 shows the XPS survey spectra for the catalysts Pt/GNF, Pt/
NeGNF, and Pt/NFeGNF. All the spectra of catalysts show peaks
onto the catalyst and allowed to dry under ambient condition. Pt corresponding to C, O, and Pt. Besides these peaks, the Pt/NeGNF
wire and saturated calomel electrode (SCE) were used as the and Pt/NFeGNF catalysts show the presence of N and N, F peaks,
counter and reference electrodes, respectively. All potentials are respectively. This confirms the successful doping of N and N, F el-
reported in terms of the RHE scale for convenience unless other- ements into the hexagonal graphitic framework of the carbon lat-
wise stated. Cyclic voltammograms (CVs) were recorded in the tice in the Pt/NeGNF and Pt/NFeGNF catalysts, respectively. The
potential range between 0 and 1.05 V at a scan rate of 50 mV s
1 in atomic percentages of doped nitrogen and fluorine atoms in the
0.5 M HClO4 aqueous solution purged with N2 gas to remove dis- matrix of NeGNF and NFeGNF and the chemical states of the
solved O2. The region for hydrogen adsorption between 0.05 and different species of N and F in the doped catalysts are listed in
0.4 V on the backward potential scan was used to estimate the Table 1.
ECSA. Linear sweep voltammograms (LSVs) for ORR activity mea- The size of Pt nanoparticles in Pt/GNF, Pt/NeGNF, and Pt/
surements were performed using an RDE in aqueous 0.5 M HClO4 NFeGNF catalysts were assessed with the help of XRD analysis. The
solution saturated with O2 at a scan rate of 5 mV s
1 at 1600 ro- XRD patterns for Pt/GNF, Pt/N-GNF, and Pt/NFeGNF catalysts are
tations per minute. To test the stability and durability of the cata- shown in Fig. 2, and the diffraction peaks at 2q of 39.8 , 46.2 , and
lysts, CVs were repeated by potential cycling for 10,000 cycles, and 67.4 are assigned to Pt (1 1 1), Pt (2 1 0), and Pt (2 2 0), respectively.
the ECSA values were recorded for every 2000th cycle. LSVs were All the Pt diffraction peaks are found to be broadened in the case of
also recorded before and after 10,000 cycles to evaluate the ORR the Pt/NeGNF and Pt/NFeGNF catalysts compared to Pt/GNF, which
behavior of the catalysts. For evaluating the stability of the catalyst indicates a reduced Pt crystallite size for Pt/NeGNF and Pt/NFeGNF
support, a more stringent test method was used by maintaining the catalyst. The average Pt crystallite size is calculated using the
catalyst at a higher potential of 1.4 V for 150 s and then decreasing DebyeeScherrer formula considering Pt (1 1 1) peaks, which are
the potential to 0.60 V for 30 s to complete a cycle. This process was about 5.7, 4.5, and 3.1 nm for Pt/GNF, Pt/NeGNF, and Pt/NFeGNF
repeated for 500 cycles, and the CV and LSV data were recorded for catalysts, respectively.
all the catalysts. The experiments were also performed with com- Fig. 3 shows the TEM images for Pt/GNF, Pt/NeGNF, and Pt/
mercial Pt/C (20 wt%) catalysts for comparison. NFeGNF catalysts. In the case of the Pt/GNF catalyst, severe ag-
glomerations of Pt nanoparticles are seen over the GNF surface, and
many regions of the GNF are left without any Pt nanoparticles
2.4. Fabrication of membrane electrode assemblies and fuel cell (Fig. 3a and b). This is due to the hydrophobic and inert nature of
performance evaluation GNF, which lacks sufficient surface functional groups to anchor
metal nanoparticles. By contrast, nanosized Pt particles are wide-
Commercial SGL DC 35 gas diffusion layers (GDLs) were used as spread and uniformly dispersed over the GNF surface in the case of
anode and cathode electrodes for membrane electrode assembly Pt/NeGNF and Pt/NFeGNF catalysts (Fig. 3c and g). This is because
(MEA) fabrication. For the reaction layers, the prepared catalysts GNF undergoes surface functionalization with N doping, which acts
were dispersed in IPA and ultrasonicated for 30 min, followed by as an anchoring site for Pt nanoparticles and facilitates the uniform
the addition of 30 wt% Nafion solution. The resultant slurries were dispersion of metal nanoparticles on the NeGNF surface [40e42].

Table 1
Atomic percentage of nitrogen and fluorine and chemical states of the different species of N and F in the doped catalysts.

Catalyst N (at. %) F (at. %) Pyridinic N binding Graphitic N binding PteN bonding binding Ionic eCeF binding Semi-ionic eCeF binding
energy (eV) energy (eV) energy (eV) energy (eV) energy (eV)

NeGNF 4.5 0 398.28 400.25 e e e

NFeGNF 13 1.1 398.18 400.34 e 685.00 687.67
Pt/NFeGNF 397.95 399.93 397.2 684.75 685.95
670 S.G. Peera et al. / Carbon 107 (2016) 667e679

Pt (1 1 1) Further, high-resolution (HR)-TEM images of the Pt/NFeGNF cata-

Pt/GNF lyst are shown in Fig. 3g, wherein clear lattice fringes of Pt nano-
particles can be seen with the d-spacing of 0.226 nm being
Pt (2 1 0)
Pt (2 2 0) attributed to the (1 1 1) plane of Pt nanoparticles. The Pt nano-
particle size distributions of all the catalysts are represented in the
form of histograms, as shown in Fig. 3hej. Although all the catalysts
Intensity (a.u.)

Pt (1 1 1)
have more or less similar average Pt particle size, the Pt/NeGNF and
Pt/N-GNF Pt/NFeGNF catalysts show comparatively smaller Pt nanoparticles,
Pt (2 1 0) results in large Pt electro chemical surface area (ECSA) available for
Pt (2 2 0)
the ORR. Moreover, N doping into GNF creates the carbon active
sites adjacent to the N atoms based on the charge delocalization
Pt (1 1 1) principles. These carbon active sites are also prone toward O2
adsorption and their subsequent reduction. Introduction of F into
Pt/NF-GNF
Pt (2 1 0) N/GNF opens up the fibrous structure, which is then transformed
Pt (2 2 0) into wrinkled graphene sheets (Fig. 3e) with many folded active
sites, as also observed in our earlier reports and validated by den-
sity functional theory (DFT) study [38].
30 35 40 45 50 55 60 65 70 75 Fig. 4 shows the CVs of Pt-deposited catalysts in 0.5 M HClO4
2 theta (degree) aqueous solution with a scan rate of 50 mV s
1. All the catalysts
show clear H2 adsorption/desorption peaks in the region between
Fig. 2. Comparative XRD pattern of Pt/GNF, Pt/NeGNF, and Pt/NFeGNF catalysts. 0 and 0.25 V and redox peaks associated with PteO/PteOH at a

Fig. 3. TEM images for (aeb) Pt/GNF, (ced) Pt/NeGNF, and (eeg) Pt/NFeGNF catalysts. The particle size distribution curves of (h) Pt/GNF, (i) Pt/NeGNF, and (j) Pt/NFeGNF catalysts.
 S.G. Peera et al. / Carbon 107 (2016) 667e679 671

Fig. 4. Cyclic voltammograms for (a) Pt/C, Pt/GNF, Pt/NeGNF, and Pt/NFeGNF catalysts recorded at 25  C in 0.5 M HClO4 solution at a scan rate of 50 mV s
1; (b) oxide and hydroxide
potential region of the catalysts; (c) hydroxyl surface coverage (WOH ads) of the catalysts calculated based on the eOHads peak obtained from the CV curves.

Fig. 5. Linear sweep voltammograms for (a) Pt/C, Pt/GNF, Pt/NeGNF, and Pt/NFeGNF at 1600 rpm in O2-saturated aqueous 0.5 M HClO4 solution at a scan rate of 5 mV s
1. (b) Mass
activities for Pt/C, Pt/GNF, Pt/NeGNF, and Pt/NFeGNF catalysts.

potential above 0.6 V. An enhanced charge generation is observed the uniform deposition of Pt nanoparticles over the doped GNF
for Pt/NeGNF and Pt/NFeGNF catalysts, which indicates that a large surface. It is worth noting that not only does the ECSA determine
amount of Hþ ions are adsorbed/desorbed on these catalysts the activity of the catalysts but the potential at which PteO for-
compared with Pt/GNF and Pt/C. This enhanced charge is directly mation and its subsequent reduction occur is also responsible for
associated with the volume of Pt nanoparticles exposed for the the intrinsic activity of the catalyst. Generally, the onset potential of
chemical reaction, usually termed as electrochemical active surface PteO species determines the stability of Pt species, and the PteOH
area (ECSA) of the catalyst. Accordingly, the ECSA values for all the reduction potential determines the kinetics of the ORR reaction.
catalysts are calculated from the charge corresponding to the H2 Fig. 4b shows the enlarged portion of the redox peaks of catalysts. It
desorption region in the reverse scan of the voltammograms, which is seen that the onset of PteO species for the Pt/NeGNF and Pt/
are found to be 75, 71, 115, and 124 m [2] g
1 Pt for Pt/C, Pt/GNF, Pt/ NFeGNF catalyst appears at potentials of ~0.85 V, larger than the Pt/
NeGNF, and Pt/NFeGNF catalysts, respectively. Improved ECSA for GNF and Pt/C catalysts. Usually, the oxidation resistance of Pt
the Pt/NeGNF and Pt/NFeGNF catalysts is exclusively attributed to nanoparticles increases with increased onset potential of the PteO
672 S.G. Peera et al. / Carbon 107 (2016) 667e679

Fig. 6. XPS deconvoluted spectra for (a) N1s and (b) F2p NFeGNF catalyst; (c) N1s, (d) F2p, and (e) C1s of Pt/NFeGNF catalyst; (f) combined Pt 4f spectra of Pt/GNF, Pt/N-GNF, and Pt/
NFeGNF catalyst.

species, which makes the Pt nanoparticles less oxophilic [43,44]. which help remove oxide/OH species from the Pt nanoparticles,
However, the PteOH reduction peak of the Pt/NFeGNF catalyst and hence enhances the ORR kinetics [30]. Moreover, well-
shows the highest onset potential of ~0.84 V among all the cata- dispersed Pt nanoparticles and their close interaction with the
lysts, indicating the higher intrinsic ORR activity of the former. support material provide a more facile path for the electron
Further PteO formation and PteOH reduction for the Pt/NFeGNF transfer, thus boosting the ORR kinetics.
catalyst occurs almost at the same potential. This indicates that Fig. 5a shows the LSVs recorded in 0.5 M HClO4 electrolyte at a
oxide formation and reduction occur quickly, refreshing the Pt scan rate of 5 mV s
1. The obtained LSVs of the catalysts illustrate
catalyst surface for faster ORR kinetics [45]. Fig. 4c represents the coherent results with the CV data. The Pt/NFeGNF catalyst shows
adsorption of hydroxyl species (eOH % coverage), which is calcu- the highest ORR onset potentials of 0.98 V among all the catalysts.
lated from the eOHads peak region in the CV backward scan. The OH Further, the Pt/NFeGNF catalyst shows much steeper kinetic curves
% coverage was obtained by normalizing the hydroxyl adsorption in the region 0.8e1.05 V, indicating significant improvement in the
charges to the ECSA of the respective catalysts. It is seen that the Pt/ ORR activity for this catalyst. The observed ORR activity is in the
NFeGNF catalyst shows lower eOH % coverage in relation to the order of Pt/GNF < Pt/C < Pt/NeGNF < Pt/NFeGNF. Further, RDE
others, which confirms weak adsorption/quick desorption of the polarization plots corrected for mass transport and kinetic pa-
eOH species, thus enhancing the ORR activity. Further, the oxo- rameters, such as kinetic current (ik) and mass activity (im), for all
philic nature of the NeGNF surface could provide charged species, the catalysts are measured at 0.9 V, as shown in Fig. 5b. The ik
 S.G. Peera et al. / Carbon 107 (2016) 667e679 673

values for Pt/C, Pt/GNF, Pt/NeGNF, and Pt/NFeGNF are found to be Fig. 6e. The peaks located at 283.6, 284.2, 284.9, 285.8 and 288.7 eV

0.19,
0.24,
0.36, and
0.81 mA cm
2, and the mass activities corresponds to sp3 carbon, sp2 carbon, eCeN, eCeO and eCeF
are 9.5, 12.2, 18.3, and 40.8 A g
1 Pt, respectively. The mass activities bonding respectively. Appearance of eCeN and eCeF in the spectra
of Pt/NeGNF and Pt/NFeGNF catalysts are, respectively, 1.9 and 4.3 confirms the doping of N and F in to the carbon matrix [51,52]. Pt4f
times higher than that of the commercial Pt/C catalyst. The excel- spectra of Pt/GNF, Pt/N-GNF and Pt/NFeGNF catalysts are shown in
lent performance of the Pt/NFeGNF catalyst is attributed to the Fig. 6f. Pt/GNF catalyst possesses two peaks at 71.80 and 75.08 eV
synergistic ORR activity of N and F dopants and their interaction corresponding to the spin-splitting of 4f7/2 and 4f5/2 regions of
with Pt nanoparticles. Further, highly dispersed smaller Pt nano- metallic Pt . The metallic peaks down shifted to lower BE in case of
particles are also responsible for the increase in ORR activity of the Pt/N-GNF and these peaks are further downshifted for Pt/NFeGNF
Pt/NFeGNF catalyst by increasing the ECSA available for the elec- catalyst. This observation clearly indicates specific interaction be-
trochemical reaction. tween Pt nanoparticles and NeF doped carbon supports as a result
The XPS spectra of NF/GNF, Pt/GNF, Pt/NeGNF, and Pt/NFeGNF of electron donation from N and F to the Pt [30,34,53]. This down
catalysts are shown in Fig. 6. To explore the interaction of the shift in BE for Pt/NFeGNF catalyst is also due to the lowering of Pt d-
support (NF/GNF) with the Pt nanoparticles, the XPS spectra of N band center relative to Fermi level which facilitates reduction of O2
and F were compared with the Pt/NFeGNF catalyst (Fig. 6a,b). The N molecules from the Pt surface [54e57]. In the present case, lower
spectra of the Pt/NFeGNF catalyst were deconvoluted into three BE for Pt/N-GNF and Pt/NFeGNF catalysts are ascribed to this
different peaks at the BEs of 398.8, 399.1, and 400.2 eV corre- phenomenon and weaken the eOH adsorption species from Pt
sponding to pyridinic N, pyrrolic N, and graphitic N, respectively surface facilitating faster ORR kinetics [58]. This is also in consistent
(Fig. 6c). Pyridinic N refers to N atoms at the edges of graphene with the % eOH coverage from the CVs (Fig. 4c).
planes which bonded to two carbon atoms, pyrrolic N atoms are Fig. 7 shows the fuel cell polarization and power density curves
linked to five membered heterocyclic rings and bonded to two for the MEAs with Pt/GNF, Pt/C, Pt/NeGNF, and Pt/NFeGNF as
carbon atoms, and in the case of graphitic N, the N is bonded to cathode catalysts at a loading of 0.1 mg cm
2, evaluated under
three carbon atoms and often called substituted N or quaternary N. identical experimental conditions. A significant enhancement in
The enhanced ORR activity of Pt/NFeGNF catalyst in this study is the activation region is observed for the PEMFCs with the Pt/
attributed to presence of all forms of N species discussed above. An NeGNF and Pt/NFeGNF catalysts. As commercial Pt/C is used as an
additional peak is also observed at relatively lower BE of 397.2 eV anode catalyst for all the MEAs, any variation in the PEMFC per-
and is attributed to N atoms bonded to Pt nanoparticles (PteN formance is attributed only to the ORR kinetic behavior of the
bonding) [31,46,47]. Since, pyridinic-N presents at the edges of cathode catalyst. The improved ORR activity and superior perfor-
graphene sheets, the adsorption energy of Pt on graphene is mance of the PEMFCs comprising the Pt/NeGNF and Pt/NFeGNF
enhanced at the edge locations due to dangling bonds and allows catalyst are clearly established in the polarization data. The ob-
for the formation of stable PteN bonds at the edges of graphene tained PEFC performances are in the order Pt/NFeGNF > Pt/
sheets [48]. Holmes et al., reports the interaction of Pt with types of NeGNF > Pt/C > Pt/GNF catalysts with the peak power densities of
N atoms and found that Pt ripening is unfavorable with pyridinic N 867, 775, 561, and 548 mW cm
2 at 70  C and 2-bar pressure. The
type compared to other form of N atoms [40]. Hence, the presence enhanced power density of the Pt/NFeGNF catalyst is attributed to
of pyridinic N and their interaction with Pt in the catalyst Pt/ the improved dispersion of Pt nanoparticles and their interaction
NFeGNF primarily contributes in enhancing the ORR activity in the with the N and F dopants of the support as described earlier.
present study. Fig. 6d shows deconvoluted F spectra which contain To evaluate the long-term stability of all the catalysts, a dura-
two different chemical states of F, one at 684.8 and the other at bility test was performed under potential cycling between 0 and
685.85 eV corresponds to ionic-F and semi-ionic-F bonds with 1.05 V up to 10,000 cycles at a sweep rate of 50 mV s
1, and the CV
carbon respectively. It is also proposed in our earlier study that responses are presented at every 2000 potential cycles as shown in
semi-ionic CeF bond is more active towards the ORR which po- Fig. 8aed. All the catalysts exhibit reduction in the hydrogen
larizes the adjacent carbon atoms to a greater extent especially in
the presence of N atoms [38]. Hence, the existence of semi-ionic
eCeF bonds is responsible for enhanced activity of the Pt/
NFeGNF catalyst. Moreover, presence of F greatly enhances the
durability of Pt catalyst, as F is more stable at the edges than in the
interior region makes the edge positions of carbon supports stable
and are resistant to carbon corrosion [49]. Hence, the combined
existence of pyridinic-N and stable edge occupation of F contributes
to the enhanced durability of Pt nanoparticles for Pt/NFeGNF
catalyst [50]. To understand the changes in chemical states of N and
F spectra in the NFeGNF support before and after Pt decoration,
XPS spectra for pristine NFeGNF sample is shown in Fig. 6a & b
based on our earlier report for fair comparison. The corresponding
BE values are also provided in Table 1. From the data, there is no
significant change in the BE of N spectra is observed between
NFeGNF and Pt/NFeGNF catalyst. However, there is a notable
change in the F spectra is observed. The BE of semi-ionic CeF bond
is shifted to the lower value in case of Pt/NFeGNF catalyst (BE of
NFeGNF is 687.67 eV; BE of Pt/NFeGNF is 685.95 eV). This shift in
BE may attribute to the specific interaction of Pt nanoparticles with
the F atoms in the Pt/NFeGNF catalyst and helps enhancing ORR
Fig. 7. Fuel cell polarization and performance curves for H2eO2 PEFCs comprising Pt/C,
activity and durability of Pt/NFeGNF catalyst. The carbon spectrum Pt/GNF, Pt/N-GNF, and Pt/NFeGNF catalysts as cathode catalysts at 70  C and 2 bar-
of the Pt/NFeGNF catalysts is deconvoluted to five peaks corre- absolute pressure with gaseous H2 (80% RH) and gaseous O2 (80% RH) to the anode and
sponding to different bonding configurations of carbon as shown in cathode of the PEFCs, respectively.
674 S.G. Peera et al. / Carbon 107 (2016) 667e679

Fig. 8. Cyclic voltammograms for (a) Pt/C, (b) Pt/GNF, (c) Pt/NeGNF, and (d) Pt/NFeGNF catalysts by repeated CV cycling test; (e) loss in ECSA values for all the catalysts with respect
to the potential cycling and (f) with respect to each cycle.

adsorption regions, indicating a decrease in ECSA with respect to The half-wave potentials shifted negatively by 200, 180, 130, and
potential cycling [59,60]. The ECSA values are normalized to their 30 mV for the Pt/C, Pt/GNF, Pt/NeGNF, and Pt/NFeGNF catalysts,
corresponding initial value, and the percentage loss of ECSA for all respectively. The calculated rate of voltage loss is 1.8, 1.2, 0.9, and
the catalysts is plotted with the number of potential cycling shown 0.2 mV h
1 for the Pt/GNF, Pt/C, Pt/NeGNF, and Pt/NFeGNF cata-
in Fig. 8e. For the Pt/C and Pt/GNF catalysts, the loss in ECSA after lysts, respectively (Fig. 9e). This remarkable stability of the Pt/
10,000 potential cycles is ~67%, whereas the losses in ECSA are NFeGNF catalyst is solely attributed to the ameliorating effect of N
about 29% and 12% for Pt/NeGNF and Pt/NFeGNF catalysts, and F heteroatoms in the graphene matrix and the interaction of Pt
respectively. Based on these findings, it is evident that the degra- nanoparticles with the graphene supports. Moreover, the presence
dation rate of the Pt/NFeGNF catalyst is greatly reduced compared of CeF bonds also enhances the electrochemical stability of Pt
to that of the commercial Pt/C and Pt/N-GNF catalysts. The average nanoparticles under fuel cell operating conditions. As a result, the
rate of surface area loss is 4.8, 3.9, 3.3, and 1.1 (10
3) m2 g
1 aggregation, dissolution, and detachment of Pt nanoparticles are
cycle
1 for the Pt/GNF, Pt/C, Pt/NeGNF, and Pt/NFeGNF catalysts, prevented.
respectively (Fig. 8f). This indicates that the Pt/NFeGNF catalyst has To further evaluate the catalyst support stability, a more strin-
a significant longer lifetime and is about four times more durable gent test method is used by maintaining the potential of the cata-
than the commercial Pt/C and Pt/GNF catalysts in 0.5 M HClO4 lyst as high as 1.4 V for 150 s and then decreasing the potential to
acidic medium. To understand the O2 reduction behavior, LSVs are 0.60 V for 30 s to complete a cycle [61]. This process was repeated
also conducted after 10,000 potential cycles as shown in Fig. 9aed. for 500 cycles for the Pt/C, Pt/NeGNF, and Pt/NFeGNF catalysts and
 S.G. Peera et al. / Carbon 107 (2016) 667e679 675

Fig. 9. Linear sweep voltammograms of the catalysts before and after the stability tests: (a) Pt/C, (b) Pt/GNF, (c) Pt/NeGNF, and (d) Pt/NFeGNF catalyst at 1600 rpm in O2-saturated
aq. 0.5 M HClO4 solution at a scan rate of 5 mV s
1; (e) rate of ORR activity degradation with respect to time.

their corresponding CV and LSV data were recorded as shown in atmosphere.

Fig. 10aef. This testing protocol mimics the actual conditions of In order to investigate Pt agglomeration and carbon support
operating fuel cells, wherein the sudden increase in the potential corrosion after 500 cycles of the corrosive test, the Pt/C and Pt/
occurs frequently during the start-up and shutdown process of a NFeGNF catalysts were collected form the GC electrode and sub-
PEMFC, which leads to severe carbon corrosion [62e65]. After 500 jected to TEM analysis shown in Fig. 11. After 500 cycles, in the case
cycles under this condition, a significant reduction in the ECSA of the Pt/C catalyst, the original globular carbon support
value is observed for the commercial Pt/C catalyst with a drastic morphology is severely affected, which leads to Pt particle size
negative shift in ORR behavior in terms of onset, half-wave po- growth, particle sintering, and agglomeration. By contrast, in the
tentials, and limiting current regions (Fig. 10a and b). By contrast, case of the Pt/NFeGNF catalyst, Pt particles are well separated and
the Pt/NeGNF and Pt/NFeGNF catalysts showed greatly enhanced no sign of agglomeration is observed, except for albeit to an in-
electrochemical behavior after 500 potential cycles. The losses of crease in their Pt particle size. The stable carbon support is also
ECSA for the Pt/C, Pt/NeGNF, and Pt/NFeGNF catalysts were found clearly seen in this catalyst after the durability test. Generally, the
to be 73%, 47%, and 35%, respectively. The reduction in the ORR half- degradation of Pt-based catalysts occurs by means of i) corrosion of
wave potential of the Pt/C, Pt/NeGNF, and Pt/NFeGNF catalysts the carbon support, ii) Pt nanoparticle dissolution/agglomeration,
were found to be 270, 120, and 90 mV, respectively. The remarkable iii) or Pt nanoparticle detachment from the support. Less graphi-
ORR activity and stability of the Pt/NFeGNF catalyst are attributed tized carbon supports are more prone to carbon corrosion, whereas
to the stability of the N and F-doped graphene support that pre- highly graphitized carbon supports or un-functionalized carbon
vents Pt nanoparticle agglomerations. These results indicate that supports decrease the stability of Pt nanoparticles due to the poor
the Pt/NFeGNF catalyst could be an ideal catalyst in a real fuel cell bonding strength between Pt and the support materials. The
676 S.G. Peera et al. / Carbon 107 (2016) 667e679

Fig. 10. Cyclic voltammograms of (a) Pt/C, (c) Pt/NeGNF, and (e) Pt/NFeGNF recorded after 500 cycles of test studies. CV recorded at a scan rate of 50 mV s
1 in 0.5 M HClO4
electrolyte. LSV curves for (b) Pt/C, (d) Pt/NeGNF, and (f) Pt/NFeGNF catalysts recorded after 500-cycle test. LSVs were recorded at a scan rate of 5 mV s
1 in 0.5 M HClO4 electrolyte.

dissolution and/or agglomeration (Ostwald ripening) of platinum stable NFeGNF support, which strengthen the binding of Pt
nanoparticles occurs either by migration of Pt ions through the nanoparticles with the support, thereby mitigating agglomeration.
ionomer in the catalyst or by Pt dissolution and subsequent However, for the commercial Pt/C catalyst, Pt nanoparticles are
deposition on neighboring particles. Generally, larger Pt nano- detached from the Vulcan carbon support, as the support medium
particles are more stable against the dissolution phenomenon is more prone to carbon corrosion. As a result, the active ECSA
because of their low surface energy, whereas very small-sized Pt available for ORR is reduced in the case of commercial Pt/C. The
nanoparticles tend to dissolute and redeposit on the neighboring Pt average Pt nanoparticle size increases from an initial value of 3e5
nanoparticles and attain a suitable size to minimize the surface to 4e8 nm for the Pt/C catalyst, whereas it increases from 1 to 2 to
energy [66]. From the TEM image of Pt/NFeGNF catalysts (Fig. 11), 4e8 nm in the case of the Pt/NFeGNF catalyst, after 500 cycles of
in addition to the presence of a graphene support layer even after the corrosive test (Fig. 11gei). Thus, it can be concluded that the
500 cycles of the corrosive test, a few larger Pt nanoparticles are NeF/GNF support, which has features similar to graphene, is more
also seen. This is attributed to the very small size of the Pt nano- resistive to carbon corrosion in an acidic medium. The Pt particles
particles of the Pt/NFeGNF catalyst, which are sensitive to disso- incorporated on the surface of these support materials significantly
lution and their redeposition on the adjacent Pt nanoparticles. enhance the electrochemical ORR behavior with great stability.
Although larger nanoparticles are observed, a high number of Pt The excellent activity and stability of the Pt/NFeGNF catalyst is
nanoparticles are still available for the ORR reaction, due to the confirmed by the following factors: The presence of N and F atoms
 S.G. Peera et al. / Carbon 107 (2016) 667e679 677

Fig. 11. TEM images of Pt/C catalysts (aeb) before and (ced) after 500 cycles; (eef) Pt/NFeGNF catalyst after 500 cycles (1.4 V for 150 s and 0.60 V for 30 s). The average particle size
distribution curves of the catalysts (g) Pt/C, before cycling (h), Pt/C, after cycling, and (i) Pt/NFeGNF.

in graphene frameworks not only facilitates the anchoring of metal 4. Conclusions

nanoparticles but also increases its conductivity by raising the
Fermi level toward the conduction band [35,67]. Charge delocal- N and F co-doped GNF is explored as a highly stable support
ization of the NFeco-doped graphene also facilitates the interaction material for the synthesis of Pt/NFeGNF catalysts for PEMFC ap-
of Pt nanoparticles with the support materials, as explained by Gao plications. The good distribution and interactions of Pt particles
et al. [68] It has been reported that the delocalized electrons in the with NF/GNF help increase the ECSA and greatly promote the ORR
p orbital of p sites on graphene overlap with the Pt d orbitals due to activity. The stability and durability of the Pt/NFeGNF catalyst in
p backbonding, in which the electron in the Pt d orbital is shared acidic medium are determined by maintaining the cell voltage at a
with a p* antibonding orbitals of the graphene support, inducing a higher potential of about 1.4 V followed by potential cycling,
strong metalesupport interaction. This p backbonding phenome- mimicking the actual fuel cell environmental conditions. The su-
non is further facilitated at the location of the carbon atoms perior electrochemical behavior and durability of the Pt/NFeGNF
substituted by N. Hence, the p back bonding effect will be much catalyst are solely due to the synergistic effect of the N and F co-
stronger in the case of NF-co-doped graphene, facilitating strong doped GNF surface and its specific interaction with the Pt nano-
metalesupport interactions with Pt nanoparticles and favoring particles. While evaluating the Pt/NFeGNF catalyst in the PEMFC, a
strong catalyst stability. Thus, the Pt/NFeGNF catalyst serves as an peak power density of 867 mW cm
2 is achieved at 70  C with a
excellent ORR catalyst with a potential application as a cathode minimal Pt loading of 0.1 mg cm
2. The excellent power density at
catalyst in fuel cell reactions. low Pt loading demonstrates the potential application of this
catalyst in PEFC technology, which may reduce the weight and cost
of a fuel cell stack.
678 S.G. Peera et al. / Carbon 107 (2016) 667e679

Acknowledgments [21] K.K. Tintula, A.K. Sahu, A. Shahid, S. Pitchumani, P. Sridhar, et al., Directed
synthesis of MC-PEDOT composite catalyst-supports for durable PEFCs,
J. Electrochem. Soc. 158 (2011) B622eB631.
Financial support from CSIR, New Delhi, India, through the [22] K.K. Tintula, A.K. Sahu, A. Shahid, S. Pitchumani, P. Sridhar, et al., Mesoporous
development of advanced materials for next-generation energy carbon and poly(3,4-ethylenedioxythiophene) composite as catalyst support
efficient devices (D-NEED) (PSC 0109) is acknowledged. The au- for polymer electrolyte fuel cells, J. Electrochem. Soc. 157 (2010)
B1679eB1685.
thors thank Dr. V. V. Giridhar, scientist in charge, CECRI-Madras [23] M. Liu, R. Zhang, W. Chen, Graphene-supported nano electrocatalysts for fuel
Unit, and Dr. Vijayamohanan K. Pillai, director, CSIR-CECRI, for cells: synthesis, properties, and applications, Chem. Rev. 114 (2014)
their constant encouragement and support. 5117e5160.
[24] N. Jha, P. Ramesh, E. Bekyarova, X. Tian, F. Wang, et al., Functionalized single-
walled carbon nanotube-based fuel cell benchmarked against US DOE 2017
Appendix A. Supplementary data technical targets, Sci. Rep. 3 (2013) 2257e2264.
[25] Y. Shao, J. Sui, G. Yin, Y. Gao, Nitrogen-doped carbon nanostructures and their
composites as catalytic materials for proton exchange membrane fuel cell,
Supplementary data related to this article can be found at http:// Appl. Catal. B 79 (2008) 89e99.
dx.doi.org/10.1016/j.carbon.2016.06.021. [26] G. Wu, D.Y. Li, C.S. Dai, D.L. Wang, N. Li, Well-dispersed high-loading Pt
nanoparticles supported by shell
core nanostructured carbon for methanol,
Langmuir 24 (2008) 3566e3575.
References [27] D. Long, W. Li, J. Miyawaki, W. Qiao, L. Ling, et al., Meso-channel development
in graphitic carbon nanofibers with various structures, Chem. Mater. 23
[1] Z. Qi, A. Kaufman, Low Pt loading high performance cathodes for PEM fuel (2011) 4141e4148.
cells, J. Power Sources 113 (2003) 37e43. [28] O.H. Suk, K.H. Lim, B. Roh, I. Hwang, H. Kim, Corrosion resistance and sintering
[2] J. Speder, A. Zana, I. Spanos, J.J.K. Kirkensgaard, K. Mortensen, et al., effect of carbon supports in polymer electrolyte membrane fuel cells, Elec-
Comparative degradation study of carbon supported proton exchange mem- trochim. Acta 54 (2009) 6515e6521.
brane fuel cell electrocatalysts e the influence of the platinum to carbon ratio [29] L. Li, L. Hu, J. Li, Z. Wei, Enhanced stability of Pt nanoparticle electrocatalysts
on the degradation rate, J. Power Sources 261 (2014) 14e22. for fuel cells, Nano Res. 8 (2015) 418e440.
[3] D.A. Stevens, J.R. Dahna, Thermal degradation of the support in carbon- [30] Y. Zhou, K. Neyerlin, T.S. Olson, S. Pylypenko, J. Bult, et al., Enhancement of Pt
supported platinum electrocatalysts for PEM fuel cells, Carbon 43 (2005) and Pt-alloy fuel cell catalyst activity and durability via nitrogen-modified
179e188. carbon supports, Energy Environ. Sci. 3 (2010) 1437e1446.
[4] H. Tang, Z. Qi, M. Ramani, J.F. Elter, PEM fuel cell cathode carbon corrosion due [31] D. He, Y. Jiang, H. Lv, M. Pan, S. Mu, Nitrogen-doped reduced graphene oxide
to the formation of air/fuel boundary at the anode, J. Power Sources 158 supports for noble metal catalysts with greatly enhanced activity and stability,
(2006) 1306e1312. Appl. Catal. B 132 (2013) 379e388.
[5] Y. Shao, J. Wang, R. Kou, M. Engelhard, J. Liu, et al., The corrosion of PEM fuel [32] K. Gong, F. Du, Z. Xia, M. Durstock, L. Dai, Nitrogen-doped carbon nanotube
cell catalyst supports and its implications for developing durable catalysts, arrays with high electrocatalytic activity for oxygen reduction, Science 323
Electrochem. Acta 54 (2009) 3109e3114. (2009) 760e764.
[6] Y. Shao, G.P. Yin, Y.Z. Gao, Understanding and approaches for the durability [33] Y. Zhao, L. Yang, S. Chen, X. Wang, Y. Ma, et al., Can boron and nitrogen Co-
issues of Pt-based catalysts for PEM fuel cell, J. Power Sources 171 (2007) doping improve oxygen reduction reaction activity of carbon nanotubes?
558e566. J. Am. Chem. Soc. 135 (2013) 1201e1204.
[7] Y. Lin, X. Cui, X. Ye, Electrocatalytic reactivity for oxygen reduction of [34] S. Pylypenko, A. Queen, T.S. Olson, A. Dameron, K. O’Neill, et al., Tuning
palladium-modified carbon nanotubes synthesized in supercritical fluid, carbon-based fuel cell catalyst support structures via nitrogen functionaliza-
Electrochem Commun. 7 (2005) 267e274. tion. II. Investigation of durability of PteRu nanoparticles supported on highly
[8] W.Z. Li, C.H. Liang, W.J. Zhou, J.S. Qiu, Z.H. Zhou, et al., Preparation and oriented pyrolytic graphite model catalyst supports as a function of nitrogen
characterization of multiwalled carbon nanotube-supported platinum for implantation dose, J. Phys. Chem. C 115 (2011) 13676e13684.
cathode catalysts of direct methanol fuel cells, J. Phys. Chem. B 107 (2003) [35] Y.K. Zhou, T. Holme, J. Berry, T.R. Ohno, D. Ginley, et al., Dopant-induced
6292e6299. electronic structure modification of HOPG surfaces: implications for high ac-
[9] C. Wang, M. Waje, X. Wang, J.M. Tang, R.C. Haddon, Y.S. Yan, Proton exchange tivity fuel cell catalysts, J. Phys. Chem. C 114 (2010) 506e515.
membrane fuel cells with carbon nanotube based electrodes, Nano Lett. 4 [36] J. Sivek, Leenaerts, B. Partoens, F.M. Peeters, First-principles investigation of
(2004) 345e348. bilayer fluorographene, J. Phys. Chem. C 116 (2012), 19240-45.
[10] C.A. Bessel, K. Laubernds, N.M. Rodriguez, R.T.K. Baker, Graphite nanofibers as [37] K.N. Wood, S. Pylypenko, T.S. Olson, A.A. Dameron, K. O’Neill, et al., Effect of
an electrode for fuel cell applications, J. Phys. Chem. B 105 (2001) 1115e1118. halide-modified model carbon supports on catalyst stability, ACS Appl. Mater.
[11] D. Sebasti an, A.G. Ruíz, I. Suelves, R. Moliner, M.J. L azaro, et al., Enhanced Interfaces 4 (2012) 6728e6734.
oxygen reduction activity and durability of Pt catalysts supported on carbon [38] S.G. Peera, A.K. Sahu, A. Arunchander, S.D. Bhat, J. Karthikeyan, et al., Nitrogen
nanofibers, Appl. Catal. B 115 (2012) 269e275. and fluorine co-doped graphite nanofibers as high durable oxygen reduction
[12] D. Sebasti zaro, A. Stassi, et al., Optimizing the
an, I. Suelves, R. Moliner, M.J. La catalyst in acidic media for polymer electrolyte fuel cells, Carbon 93 (2015)
synthesis of carbon nanofiber based electrocatalysts for fuel cells, Appl. Catal. 130e142.
B 132 (2013) 22e27. [39] A.K. Sahu, K.G. Nishanth, G. Selvarani, P. Sridhar, S. Pitchumani, et al., Polymer
[13] D. Sebastian, M.J. Lazaro, R. Moliner, I. Suelves, A.S. Arico, et al., Oxidized electrolyte fuel cells employing electrodes with gas-diffusion layers of mes-
carbon nanofibers supporting PtRu nanoparticles for direct methanol fuel oporous carbon derived from a solegel route, Carbon 47 (2009) 102e108.
cells, Int. J. Hydrogen. Energy 39 (2014) 5414e5423. [40] T. Holme, Y. Zhou, R. Pasquarelli, R. O’Hayre, First principles study of doped
[14] D. Sebasti zaro, The effect of carbon
an, I. Suelves, E. Pastor, R. Moliner, M.J. La carbon supports for enhanced platinum catalysts, Phys. Chem. Chem. Phys. 12
nanofiber properties as support for PtRu nanoparticles on the electrooxidation (2010) 9461e9468.
of alcohols, Appl. Catal. B 132 (2013) 13e21. [41] B.P. Vinayan, R. Nagar, N. Rajalakshmi, S. Ramaprabhu, Novel platinumecobalt
[15] S.G. Peera, A.K. Sahu, S.D. Bhat, S.C. Lee, Nitrogen functionalized graphite alloy nanoparticles dispersed on nitrogen-doped graphene as a cathode
nanofibers/Ir nanoparticles for enhanced oxygen reduction reaction in poly- electrocatalyst for PEMFC applications, Adv. Funct. Mater. 22 (2012)
mer electrolyte fuel cells (PEFCs), RSC Adv. 4 (2014) 11080e11088. 3519e3526.
[16] S.G. Peera, A.K. Sahu, A. Arunchander, N. Krishna, S.D. Bhat, Deoxyribonucleic [42] L. Cong-Sen, L. Xiao-Chen, W. Guo-Chong, L. Ru-Ping, Q. Jian-Ding, Preparation
acid directed metallization of platinum nanoparticles on graphite nanofibers of nitrogen-doped graphene supporting Pt nanoparticles as a catalyst for
as a durable oxygen reduction catalyst for polymer electrolyte fuel cells, oxygen reduction and methanol oxidation, J. Electroanal. Chem. 728 (2014)
J. Power Sources 297 (2015) 379e387. 41e50.
[17] S.G. Peera, K.K. Tintula, A.K. Sahu, S. Shanmugam, P. Sridhar, et al., Catalytic [43] L. Li, S.G. Chen, Z.D. Wei, X.Q. Qi, M.R. Xia, et al., Experimental and DFT study
activity of Pt anchored onto graphite nanofiber-poly (3,4- of thiol-stabilized Pt/CNTs catalysts, Phys. Chem. Chem. Phys. 14 (2012)
ethylenedioxythiophene) composite toward oxygen reduction reaction in 16581e16587.
polymer electrolyte fuel cells, Electrochim. Acta 108 (2013) 95e103. [44] I. Takahashi, S.S. Koch, Examination of the activity and durability of PEMFC
[18] G.S. Chai, I.S. Shin, Y.J. Sung, Synthesis of ordered, uniform, macroporous catalysts in liquid electrolytes, J. Power Sources 195 (2010) 6312e6322.
carbons with mesoporous walls templated by aggregates of polystyrene [45] Z.L. Wang, T.S. Ahmad, M.A. El-Sayed, Steps, ledges and kinks on the surfaces
spheres and silica particles for use as catalyst supports in direct methanol fuel of platinum nanoparticles of different shapes, Surf. Sci. 380 (1997) 302e310.
cells, Adv. Mater. 16 (2004) 2057e2061. [46] P.L. Kuo, W.F. Chen, H.Y. Huang, I.C. Chang, S.A. Dai, Stabilizing effect of
[19] S.H. Joo, S.J. Choi, I. Oh, J. Kwak, Z. Liu, et al., Ordered nanoporous arrays of pseudo-dendritic polyethylenimine on platinum nanoparticles supported on
carbon supporting high dispersions of platinum nanoparticles, Nature 412 carbon, J. Phys. Chem. B 110 (2006) 3071e3077.
(2001) 169e172. [47] A. Adenier, M.M. Chehimi, I. Gallardo, J. Pinson, N. Vila, Electrochemical
[20] J. Cao, Z. Chen, J. Xu, W. Wang, Z. Chen, Mesoporous carbon synthesized from oxidation of aliphatic amines and their attachment to carbon and metal
dual colloidal silica/block copolymer template approach as the support of surfaces, Langmuir 20 (2004) 8243e8253.
platinum nanoparticles for direct methanol fuel cells, Electrochim. Acta 88 [48] K. Kong, Y. Choi, B.H. Ryu, J.O. Lee, H. Chang, Investigation of metal/carbon-
(2013) 184e192. related materials for fuel cell applications by electronic structure
 S.G. Peera et al. / Carbon 107 (2016) 667e679 679

calculations, Mater. Sci. Eng. 26 (2006) 1207e1210. [59] B. Merzougui, S. Swathirajan, Rotating disk electrode investigations of fuel cell
[49] M. Brzhezinskaya, A. Vinogradov, Electronic structure of fluorinated carbon catalyst degradation due to potential cycling in acid electrolyte,
nanotubes, carbon nanotubes, in: Jose Mauricio Marulanda (Ed.), 2010, ISBN J. Electrochem. Soc. 153 (2006) A2220eA2226.
978-953-307-054-4 (InTech). [60] Y. Shao, G. Yin, Y. Gao, Understanding and approaches for the durability issues
[50] J. Cheng, Y. Li, X. Huang, Q. Wang, A. Meib, P.K. Shen, Highly stable electro- of Pt-based catalysts for PEM fuel cell, J. Power Sources 171 (2007) 558e566.
catalysts supported on nitrogen-self-doped three-dimensional graphene-like [61] S. Yuyan, K. Rong, W. Jun, V.V. Vilayanur, K. Ja-Hun, et al., The influence of the
networks with hierarchical porous structures, J. Mater. Chem. A 3 (2015) electrochemical stressing (potential step and potential-static holding) on the
1492e1497. degradation of polymer electrolyte membrane fuel cell electrocatalysts,
[51] R. Lv, Q. Li, A.R. Botello-Mendez, T. Hayashi, B. Wang, et al., Nitrogen-doped J. Power Sources 185 (2008) 280e286.
graphene: beyond single substitution and enhanced molecular sensing, Sci. [62] C.A. Reiser, L. Bregoli, T.W. Patterson, J.S. Yi, J.D.L. Yang, Reverse-current decay
Rep. 2 (2012) 586e594. mechanism for fuel cells, Electrochem. Solid State Lett. 8 (2005) A273eA276.
[52] K.I. Ho, C.H. Huang, J.H. Liao, W. Zhang, L.J. Li, et al., Fluorinated graphene as [63] S.D. Knights, K.M. Colbow, J. St-Pierre, D.P. Wilkinson, Aging mechanisms and
high performance dielectric materials and the applications for graphene lifetime of PEFC and DMFC, J. Power Sources 127 (2004) 127e134.
nanoelectronics, Sci. Rep. 4 (2014) 5893e5900. [64] T.W. Patterson, R.M. Darling, Damage to the cathode catalyst of a PEM fuel cell
[53] J. Zhu, G. He, L. Liang, Q. Wan, P.K. Shen, Direct anchoring of platinum caused by localized fuel starvation, Electrochem. Solid State Lett. 9 (2006)
nanoparticles on nitrogen and phosphorus-dual-doped carbon nanotube ar- A183eA185.
rays for oxygen reduction reaction, Electrochim. Acta 158 (2015) 374e382. [65] K.K. Tintula, A. Jalajakshi, A.K. Sahu, S. Pitchumani, P. Sridhar, et al., Durability
[54] H. Tsunoyama, N. Ichikuni, H. Sakurai, T. Tsukuda, Effect of electronic struc- of Pt/C and Pt/MC-PEDOT catalysts under simulated start-stop cycles in
tures of Au clusters stabilized by poly(N-vinyl-2-pyrrolidone) on aerobic polymer electrolyte fuel cells, Fuel Cells 13 (2013) 158e166.
oxidation catalysis, J. Am. Chem. Soc. 131 (2009) 7086e7093. [66] Y. Shao-Horn, W.C. Sheng, S. Chen, P.J. Ferreira, et al., Instability of supported
[55] B. Hammer, J.K. Nørskov, Electronic factors determining the reactivity of metal platinum nanoparticles in low-temperature fuel cells, Top. Catal. 46 (2007)
surfaces, Surf. Sci. 343 (1995) 211e220. 285e305.
[56] M. Mavrikakis, B. Hammer, J.K. Nørskov, Effect of strain on the reactivity of [67] S. Jiang, Y. Ma, G. Jian, H. Tao, X. Wang, et al., Facile construction of PteCo/CNx
metal surfaces, Phys. Rev. Lett. 81 (1998) 2819e2822. nanotube electrocatalysts and their application to the oxygen reduction re-
[57] B. Hammer, Y. Morikawa, J.K. Nørskov, Co chemisorption at metal surfaces and action, Adv. Mater. 21 (2009) 4953e4956.
overlayers, Phys. Rev. Lett. 76 (1996) 2141e2144. [68] S. Gao, H. Fan, X. Wei, L. Li, Y. Bando, D. Golberg, Nitrogen-doped carbon with
[58] X. Jiangfeng, F. Gengtao, T. Yawen, Z. Yiming, C. Yu, L. Tianhong, One-pot mesopore confinement efficiently enhances the tolerance, sensitivity, and
synthesis of three-dimensional platinum nanochain networks as stable and stability of a Pt catalyst for the oxygen reduction reaction, Part. Part. Syst.
active electrocatalysts for oxygen reduction reactions, J. Mater. Chem. 22 Charact. 30 (2013) 864e872.
(2012) 13585e13590.