B, N- and P, N-doped graphene as highly active catalysts for oxygen reduction reactions in acidic media

Article abstract of DOI:10.1039/c3ta01648j

Graphene has been highlighted recently as a promising material for energy conversion due to its unique properties deriving from a two-dimensional layered structure of sp²-hybridized carbon. Herein, N-doped graphene (NGr) is developed for its application in oxygen reduction reactions (ORRs) in acidic media, and additional doping of B or P into the NGr is attempted to enhance the ORR performance. The NGr exhibits an onset potential of 0.84 V and a mass activity of 0.45 mA mg^-1 at 0.75 V. However, the B, N- (BNGr) and P, N-doped graphene (PNGr) show onset potentials of 0.86 and 0.87 V, and mass activities of 0.53 and 0.80 mA mg^-1, respectively, which are correspondingly 1.2 and 1.8 times higher than those of the NGr. Moreover, an additional doping of B or P effectively reduces the production of H ₂O₂ in the ORRs, and shows much higher stability than that of Pt/C in acidic media. It is proposed that the improvement in the ORR activity results from the enhanced asymmetry of the spin density or electron transfer on the basal plane of the graphene, and the decrease in the energy gap between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) of the graphene through additional doping of B or P.

Full text of DOI:10.1039/c3ta01648j

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
B, N- and P, N-doped graphene as highly active catalysts
for oxygen reduction reactions in acidic media†
Cite this: J. Mater. Chem. A, 2013, 1,
3694
Chang Hyuck Choi,a Min Wook Chung,b Han Chang Kwon,a Sung Hyeon Parka
and Seong Ihl Woo
ab
*
Graphene has been highlighted recently as a promising material for energy conversion due to its unique
properties deriving from a two-dimensional layered structure of sp2-hybridized carbon. Herein, N-doped
graphene (NGr) is developed for its application in oxygen reduction reactions (ORRs) in acidic media,
and additional doping of B or P into the NGr is attempted to enhance the ORR performance. The NGr
exhibits an onset potential of 0.84 V and a mass activity of 0.45 mA mgꢀ1 at 0.75 V. However, the B, N-
(BNGr) and P, N-doped graphene (PNGr) show onset potentials of 0.86 and 0.87 V, and mass activities of
0.53 and 0.80 mA mgꢀ1, respectively, which are correspondingly 1.2 and 1.8 times higher than those of
the NGr. Moreover, an additional doping of B or P effectively reduces the production of H2O2 in the
ORRs, and shows much higher stability than that of Pt/C in acidic media. It is proposed that the
improvement in the ORR activity results from the enhanced asymmetry of the spin density or electron
transfer on the basal plane of the graphene, and the decrease in the energy gap between the highest
occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) of the
graphene through additional doping of B or P.
Received 1st November 2012
Accepted 18th January 2013
DOI: 10.1039/c3ta01648j
www.rsc.org/MaterialsA
electrochemistry, graphene has also been recognized as a
promising material, and it has been modied by various
1 Introduction
Carbon-based materials, especially N-doped carbon, have been
highlighted as non-precious metal catalysts for oxygen reduc-
tion reactions (ORRs) due to their high methanol–CO tolerance
and good durability compared to commercially available noble
metal catalysts (e.g. Pt or Pd).1 In alkaline media, various
carbon-based catalysts have been developed and they show
comparable ORR activities to that of Pt.2,3 However, despite their
rapid advance, carbon-based catalysts have exhibited much
lower ORR activities than that of Pt in acidic media.4,5 Therefore,
an enhancement of the ORR activity of the carbon-based cata-
lysts is one of the most important topics in the research on
polymer electrolyte membrane fuel cells (PEMFCs) including
direct methanol fuel cells (DMFCs), which are operated in the
acidic environment.
methods to be applied as an ORR catalyst. As a support for
noble metal nanoparticles, graphene provides a high surface
area and a facile electron path, and the catalyst demonstrates
enhanced ORR activity and stability.7 Moreover, graphene has
been used as a catalyst itself via modication by the doping of
heteroatoms and by decoration with polymers and transition
metal complexes.8,9 However, most research regarding graphene
as a catalyst itself has focused on its activity in alkaline media,
whereas there are only a few investigations in which it is applied
as an ORR catalyst in acidic media. This may be due to the
relatively low ORR activity of graphene-based materials
compared to that of other carbon-derived materials (e.g.
graphite and carbon nanotubes (CNTs)).4,5
Tsai et al. synthesized FeCN/graphene catalysts and exhibi-
ted an onset potential of 0.75 V (vs. RHE) in ORRs.10 Similarly,
Choi et al. prepared Fe-complex decorated graphene catalysts
and revealed an onset potential of 0.85 V (vs. RHE) and a mass
activity of 0.053 mA mgꢀ1 at 0.75 V (vs. RHE).11 Compared to
other carbon-based catalysts reported elsewhere, which show an
onset potential of approximately 0.9 V for ORRs in acidic media,
these graphene-based catalysts show much lower ORR perfor-
mance levels.4,5 Recently, Byon et al. reported highly active
graphene-based catalysts prepared from FeCl2, C3N4, and
graphite oxide (GO), demonstrating an onset potential of 0.9 V
(vs. RHE) and a mass activity of 1.5 mA mgꢀ1 at 0.75 V (vs.
RHE).12 This catalyst exhibited the best ORR activity among the
Since its discovery, graphene has enabled signicant
advances in energy conversion and storage technologies due to
its unique properties derived from the two-dimensional
layered structure of sp2-hybridized carbon.6 In the area of
aDepartment of Chemical and Biomolecular Engineering, Daejeon 305-701, Republic of
Korea
bGraduate School of EEWS (WCU), Korea Advanced Institute of Science and
Technology, Daejeon 305-701, Republic of Korea. E-mail: siwoo@kaist.ac.kr; Tel:
+82 42 350 3918
† Electronic supplementary information (ESI) available: Catalyst compositions,
XPS results, EDS mapping images, LSV results before and aer the
acid-leaching step. See DOI: 10.1039/c3ta01648j
3694 | J. Mater. Chem. A, 2013, 1, 3694–3699
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
graphene-derived catalysts that act as catalysts themselves in (0.06 g) and phosphoric acid (0.2 mL) for B and P sources,
acidic media, but even this catalyst did not outperform the respectively, and the precursors were added to the precursor
activities of other carbon-based catalysts. Although Li et al. mixture before evaporation. To verify the roles of B and P in
reported
a CNT–graphene complex catalyst that showed ORRs, B-doped graphene (BGr) and P-doped graphene (PGr)
outstanding ORR performance in acidic media among all were also prepared via the same method of BNGr and PNGr in
carbon-based catalysts,13 this catalyst is restricted to CNT- the absence of DCDA.
derived catalysts, rather than graphene-derived catalysts.
The physical properties of the graphene-derived catalysts
However, graphene is still a highly attractive material for ORRs were examined via X-ray diffraction (XRD), transmission electron
in acidic media because it has many edge sites deriving from its microscopy (TEM), Raman spectroscopy, elemental analysis
unique planar structure, where pyridinic-N sites can be (EA), inductively coupled plasma (ICP), energy dispersive spec-
produced.14 Pyridinic-N in carbon is known to be the most trometry (EDS) and X-ray photoelectron spectroscopy (XPS). The
active site for ORRs among the various N-doping types of XRD patterns were acquired from a D/MAX-2500 (Rigaku) oper-
graphitic-N, pyrrolic-N, and pyridinic-oxide.4,15 Therefore, ated at 40 kV and 300 mA. The step-scan patterns were collected
although graphene-derived materials show low levels of activity, in the 5ꢁ to 60ꢁ range (2-theta ranges) with a step size of 0.01ꢁ and
they still have strong potential as a non-noble metal catalyst for a scan speed of 1ꢁ minꢀ1. The TEM images were taken with a
ORRs in acidic media.
JEM2100-F (Jeol Ltd.) operated at 200 kV. The Raman spectros-
Recently, B- and P-doped carbon without N-doping, and B, N- copy was performed using a LabRAM HR UV/Vis/NIR (Horiba
doped carbon were proposed as promising ORR catalysts in Jobin Yvon) with a laser source of 514 nm. The EA was performed
alkaline media;16–19 however, the activities of these materials in in order to obtain the bulk compositions of the catalysts using a
acidic media have not yet been studied extensively. In our FlashEA 1112. The amounts of B and P in the graphenes were
previous reports, it was noted that additional doping of B or P obtained via an ICP analysis (POLY SCAN 61 E). The EDS data
into N-doped graphite induces a modication of the physical were obtained using a Horiba EDS. The XPS analysis was per-
and chemical properties of N-doped graphite, resulting in a formed using a Sigma Probe (Thermo VG Scientic) equipped
signicant improvement of the ORRs in acidic media.20,21 with a microfocused monochromator X-ray source.
Herein, for the development of highly active graphene-based
The electrochemical properties of the graphene-derived
catalysts for ORRs in acidic media, B or P is additionally doped catalysts were characterized using the CHI700D (CH Instru-
into N-doped graphene (NGr) by using boric acid for B, N-doped ments Inc.) and RRDE-3A (ALS Co.) in a three-electrode beaker
graphene (BNGr) and phosphoric acid for P, N-doped graphene cell equipped with a Pt wire counter electrode (ALS Co., 002233),
(PNGr), and their catalytic ability is discussed.
an Ag/AgCl reference electrode (ALS Co., 012167), and a ring disk
electrode (ALS Co., 011169). All potentials shown in this paper
were converted to the reversible hydrogen electrode (RHE)
scale. The catalysts (10 mg) were dispersed in a Naon ink
2 Experimental
All chemicals were obtained from Aldrich. GO was synthesized solution (1 mL and 1 wt% Naon content), and the ink (5 mL) was
via the method reported by Tour and coworkers.22 Briey, then dropped onto the glassy carbon (3 mm) of the ring disk
graphite (3 g) was dispersed in a concentrated acid mixture of electrode. The inks were dried at room temperature. Cyclic vol-
H2SO4 and H3PO4. Aer the addition of KMnO4 (18 g), the tammetry (CV) was conducted in a 0.1 M HClO4 electrolyte
solution was heated at 50 ꢁC for 12 h. The solution was poured purged with nitrogen or oxygen for more than 1 h, at a scan rate of
onto ice (400 mL), and then H2O2 (5 mL) was added. GO was 40 mV sꢀ1 from 0 to 1.2 V (vs. RHE). Linear sweep voltammetry
obtained aer iterative centrifugation with an HCl solution and (LSV) was performed in a 0.1 M HClO4 electrolyte purged with
deionized (DI) water. rGO was obtained from the chemical nitrogen or oxygen for more than 1 h, with a scan rate of 5 mV sꢀ1
reduction of GO with hydrazine. GO (5 g) was dispersed in DI- from 1.04 to 0.14 V (vs. RHE) and a rotation speed of 900 rpm. The
water aꢁnd was sonicated for 1 h. The reduction was conducted ORR currents were obtained by subtracting the LSV results of an
at 100 C for 24 h with the addition of hydrazine (50 mL) fol- oxygen-purged electrolyte from the results of a nitrogen-purged
lowed by ltration, washing, and drying in a vacuum oven. NGr electrolyte, in order to remove the capacitance of the catalysts.
was prepared by the pyrolysis of a graphene–dicyandiamide The H2O2 formation yield was derived from the current at the Pt
(DCDA) mixture in the presence of small amounts of Co and Fe ring disk electrode. A constant potential (1.2 V, vs. RHE) was
(<1 wt%). In DI-water (100 mL), rGO (0.6 g), CoCl2–6H2O (6 mg), applied to the Pt ring disk electrode during the ORR experiment
FeCl2–4H2O (16 mg) and DCDA (1.2 g) were dissolved and and a collection efficiency of 0.37 was used.5 The current–time
sonicated for 1 h. The mixture powder was gathered by evapo- chronoamperometric results were obtained at 0.6 V (vs. RHE) for
ration of the solution, which was then heat-treated at 900 ꢁC for 12 h in 1 M HClO4 with continuous oxygen bubbling and
3 h under an Ar atmosphere. Aer acid leaching at 80 ꢁC for 8 h commercial Pt/C (20 wt%, E-tek) was used for comparison. All of
in a H2SO4 solution (0.5 M, 100 mL), metal chlorides and DCDA the equations to analyze the results are shown in the ESI.†
were mixed with the powder and the heat-treatment step was
conducted again. This secondary pyrolysis step was conducted
because it has been reported that a secondary pyrolysis step
improves the catalytic activity of carbon materials.4,5 Additional The catalysts were prepared by pyrolysis (900 ꢁC) of the dopant
doping of B (BNDC) or P (PNDC) was performed with boric acid sources on graphene (rGO) derived from graphite oxide (GO).
3 Results and discussion
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 3694–3699 | 3695

View Article Online
Journal of Materials Chemistry A
Paper
Fig. 1 (a) XRD patterns of GO, rGO, and prepared graphene derived catalysts. (b)
Magnified image at the graphite (002) peak is also presented in the figure.
Fig. 3 Raman spectroscopy results of the prepared catalysts: (a) full range and
(b) magnified images at the 2D band region. Calculated ID/IG values and peak
positions of the 2D band are indicated in the figure.
The interlayer spacing of GO calculated from the XRD results
˚
was 8.7 A, which indicates the successful oxidization of graphite
as shown in Fig. 1.22 Aer the hydrazine reduction of GO, the
XRD pattern of rGO showed a broad peak at ꢂ24.7ꢁ, which
exhibited similar patterns to that of graphene reported else-
where.14 However, intense peaks at 26.7ꢁ newly emerged
through the heteroatom-doping steps, which included the
pyrolysis of a precursor mixture. These were assigned to the
XRD patterns of graphite.23 The graphitic structure of the cata-
lysts was determined to have resulted from two different causes.
The rst is the generation of new graphite materials on the
graphene layer and the second is the restacking of graphene
sheets. However, as shown in Fig. 2, the morphologies of NGr,
BNGr, and PNGr were similar to that of rGO, a two-dimensional
sheet structure, and newly generated graphite structures were
not observed aer the pyrolysis step (Fig. S1†).24 Therefore, it
can be concluded that the XRD patterns of graphite for the
heteroatom-doped graphene catalysts resulted from the
restacking of graphene sheets, which induces the restoration of
graphite properties from graphene. In the XRD patterns, the
intensities of the peaks at 26.7ꢁ decreased with the additional
doping of B or P, and it was referred that the degree of the
restacking of graphene sheets is in the order of PNGr < BNGr <
NGr.
resulted from the many defect sites in the graphene layers.25
The 2D-band positions of the prepared catalysts were 2717.3,
2709.6, and 2708.7 cmꢀ1 for the NGr, BNGr, and PNGr,
respectively. These values are between the value of ꢂ2680 cmꢀ1
for single layer graphene and ꢂ2730 cmꢀ1 for bulk graphite.
Therefore, this indicates that several single layers of graphene
were restacked in the prepared graphene-derived catalysts.
The compositions of the prepared catalysts were analyzed by
elemental analysis (EA) and ICP (Table S1†), and the results
disclosed that heteroatoms (N, B, and P) were effectively doped
into graphene via pyrolysis with DCDA, boric acid, or phos-
phoric acid. EDS mapping images of C and the dopants corre-
sponded well to the TEM images (Fig. S2†). This result indicated
that heteroatoms were doped evenly on the graphene surface.
The doping phases of N, B, and P were examined by XPS
analysis, as shown in Fig. 4. In graphene, N is present in the
forms of pyridinic-N (N1, 398.6 eV), graphitic-N (N2, 400.9 eV),
and pyridinic-oxide (N3, 403.0 eV).24,26 Among the N-doping
phases (N1, N2 and N3), N was dominantly doped in graphene
The restacking of graphene sheets aer the additional
doping of B or P was also supported by the Raman spectroscopy
results. As shown in Fig. 3, D- and G-bands emerged in all of the
prepared catalysts, and revealed a high D-band intensity which
Fig. 4 XPS results of the dopant elements (N, B, and P) in the prepared catalysts.
The peaks are deconvoluted; N0 (BN), N1 (pyridinic-N), N2 (graphitic-N), and N3
(pyridinic-oxide) for N-dopant; B1 (BN), B2 (BC3), B3 (partially oxidized carbon-
BN3, BCO2, and BC2O) for B-dopant; and P1 (MexP), P2 (P–O), and P3 (P–C) for P-
dopant. The scale bars in the XPS-N1s results indicate 450 cps.
Fig. 2 TEM images of the prepared catalysts: (a) rGO, (b) NGr, (c) BNGr, and (d)
PNGr.
3696 | J. Mater. Chem. A, 2013, 1, 3694–3699
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
in the form of pyridinic-N (Table S2†), which preferred to be
doped at the edge of graphene. Aer the additional doping of B
or P, the portion of pyridinic-N increased but that of pyridinic-
oxide decreased. This phenomenon was observed in our
previous report.21 In the XPS-N1s of BNGr, a new N-doping
phase, the BN type (N0, 397.9 eV) emerged.27 This bonding was
also observed in the XPS-B1s of BNGr. The XPS-B1s of BNGr was
convoluted with BN (B1, 190.3 eV), BC3 (B2, 191.0 eV), and
partially oxidized B (B3, 192.0 eV; e.g. BN3, BCO2, and BC2O).27–30
The dominant doping phase of B was BC3 (48.9%) and some of
the B was bonded to C, N, and O on the graphene surface
(Table S3†). Moreover, the XPS-P2p results of PNGr disclosed the
doping phases of MexP (P1, 129.4 eV; Me ¼ Co and Fe, x ¼ 1–2),
P–O (P2, 132.9 eV), and P–C (P3, 135.0 eV),31–33 and 77.6% of P
was doped in the form of the P–O phase (Table S3†). The XPS-
O
1s results revealed that the XPS intensity of the surface oxygen
Fig. 6 Electrochemical properties of the prepared graphene-based catalysts for
ORRs in 0.1 M HClO4 with a 900 rpm rotation speed: (a) LSV results, (b) Tafel-plots,
and (c) mass activities at 0.75 V (vs. RHE). (d) ORR result of PNGr was compared
with that of Pt/C.
groups increased through the additional doping of B or P
(Fig. S3†). This was because the additional dopant, B or P,
preferred to bond with oxygen and this oxygen was not elimi-
nated perfectly via the heat-treatment step.
Fig. 5 shows the CV results of the prepared graphene-based
catalysts in 0.1 M HClO4 saturated with N2 and O2. Regardless of
the dopant, all of the prepared catalysts showed a reduction
current under O2 conditions, which indicates the ORR activities
of the graphene-based catalysts. It was proposed that the
reduction peak potential obtained from the CVs in an O2-satu-
rated electrolyte represents the ORR performance of the carbon-
based catalysts.34 The peak potentials of the graphene-based
catalysts were 0.57, 0.61, and 0.64 V (vs. RHE) for NGr, BNGr,
and PNGr, respectively. These results indicate that the addi-
tional doping of B or P into NGr signicantly improved the ORR
activity in acidic media.
TheLSV results also supported that additional dopingof B orP
enhanced the ORR activity of NGr (Fig. 6a). The mass activities of
the prepared catalysts were calculated from Tafel-plots (Fig. 6b).
Onset potentials of the graphene-based catalysts were 0.84, 0.86,
and 0.87 V (vs. RHE) for NGr, BNGr, and PNGr, respectively.
Additional doping of B or P increased the onset potentials.
Especially, P-doping showed an onset potential of nearly 0.9 V,
which was similar tothat ofother carbon-based catalysts reported
by other leading groups.4,5 The half-wave potentials were 0.58,
0.61, and 0.64 V (vs. RHE) for NGr, BNGr, and PNGr, respectively,
indicating an increased ORR activity aer the additional doping
of B or P. The mass activity of NGr obtained at 0.75 V (vs. RHE) was
0.45 mA mgꢀ1, but this value was enhanced signicantly to 0.53
and 0.80 mA mgꢀ1 for BNGr and PNGr, respectively (Fig. 6c).
Although a direct comparison is not valid due to the different
operation conditions, the mass activities of the graphene-derived
catalysts were compared with those of graphite-derived catalysts
reported in our previous work (Fig. S4†). The results show similar
tendencies in terms of the mass activity according to the type of
dopant: N-doping < B, N-doping < P, N-doping. PNGr, having the
highest ORRactivityamongthepreparedcatalysts, showed nearly
1.8-fold higher ORR activity than that of NGr. This activity was
much higher than those of other graphene-derived materials
(0–0.053 mA mgꢀ1),10,11 and was comparable with that of the
catalyst reported by Shao-Horn and coworkers (1.5 mA mgꢀ1),12
which showed the best ORR activity in acidic media among all
graphene-based catalysts. However, its activity is still lower than
that of commercial Pt/C (Fig. 6d).
Fig. 5 CV results of the prepared catalysts in 0.1 M HClO4 saturated with N2
(solid line) and O2 (dotted line), respectively. Oxygen reduction peak potentials in
the CV results are also indicated in the figure.
Fig. 7 (a) H2O2 production yields with calculated number of electrons trans-
ferred at 0.75 V (vs. RHE), and (b) current–time chronoamperometric results
obtained at 0.6 V (vs. RHE) for 12 h.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 3694–3699 | 3697

View Article Online
Journal of Materials Chemistry A
Paper
Additional doping of B or P into the NGr not only advances in the graphene basal plane,41,42 and experimentally it was
the catalytic activity in ORRs but also improves the ORR proved that heteroatom doping increased the electric conduc-
pathway and stability. As shown in Fig. 7a, the presence of B or P tivity of graphene materials.43,44 The other possible reason is the
in the NGr decreased H2O2 production during ORRs, which led decrement of energy gap between HOMO and LUMO of gra-
to a signicant degradation of performance in the fuel cell phene. Aihara reported that the decreasing HOMO–LUMO gap
operations; from 13.4% for NGr to 1.8% and 2.2% of H2O2 derived high chemical activity due to the facilitated formation
production for BNGr and PNGr at 0.75 V (vs. RHE), respectively. of the activated complex for any potential reactions including
Therefore, it was revealed that oxygen was reduced by nearly a ORRs.45 Theoretically, it was calculated that doping of B and P
four-electron pathway and was mainly transferred to water in decreased energy gaps in graphene effectively,19,46 thus, addi-
the BNGr and PNGr. Fig. 7b shows current–time chro- tional doping of B or P in NGr induced the decrement in energy
noamperometric results in 1 M HClO4 with continuous O2 gap of graphene materials and this might be one of the reasons
bubbling at 0.6 V (vs. RHE) for 12 h. Aer 12 h of operation in for improved ORR activity in the acidic environment.
severe acidic media, Pt/C reveals a signicant performance
degradation, indicating the poor stability of Pt/C in the acidic
environment. However, all of the graphene-derived catalysts
4 Conclusions
resulted in high resistibility to a severe acidic environment, and
they still showed stable ORR performance aer 12 h of
operation.
It has been proposed that metal–nitrogen complexes (Me–
Nx) and N-doped carbons are active sites in ORRs.35,36 If Me–Nx
As a catalyst for ORRs in acidic media, graphene was modied
by doping of N and additional dopants, B and P. In the modi-
cation steps, DCDA, boric acid, and phosphoric acid were used
as doping sources for N, B, and P, respectively, and it was
conrmed that the heteroatoms were doped evenly on the gra-
sites were active sites in the prepared graphene-based catalysts,
phene surface. Regardless of the dopant types, the graphene-
the destruction of Me–Nx sites through acid leaching in the
based catalysts revealed high onset potentials, ꢂ0.86 ꢃ 0.01 V
prepared catalysts could cause a decrease in the ORR activity.
(vs. RHE). However, additional doping of B and P in NGr
For example, Li et al. reported that the dissolution of Fe–Nx
signicantly improved catalytic activity in ORRs. The NGr
active sites decreased the onset potential by 0.1 V.13 However, as
showed 0.45 mA mgꢀ1 of mass activity at 0.75 V, but BNGr and
shown in Fig. S5,† a signicant performance degradation was
PNGr indicated 0.53 and 0.80 mA mgꢀ1 of mass activities,
respectively. These values were 1.2 and 1.8 times higher than
that of the NGr, and much higher activities compared with
those of other graphene-based catalysts reported elsewhere.
Moreover, the presence of B and P also increased the catalytic
stability against a severe acidic environment. Additional doping
of B and P in NGr derived electrophysical modications of
graphene, resulted in enhancement of asymmetry of spin
density, facile electron transfer on the graphene basal plane,
and decrement of energy gap between HOMO and LUMO. The
electrophysical modications result in the production of reac-
tive complex in graphene with an increment of conductivity
not observed and this indicated that Me–Nx sites were not active
sites in the case of the prepared graphene-derived catalysts.
This conclusion was supported by other reports that Me–Nx is
not stable and is destroyed at temperatures over 900 ꢁC;37
therefore, it was supposed that Me–Nx could not be generated in
the graphene-derived catalysts that were heat-treated at 900 ꢁC.
Moreover, it was shown that additionally doped B or P did not
generate new active sites in the ORRs. As shown in Fig. S6,† BGr
and PGr, synthesized by the same method for BNGr and PNGr
without DCDA (N-source), indicated much lower ORR activities
and onset potentials compared with those of NGr.
For the prepared graphene-based catalysts, it was suggested
through the basal plane, and it was suggested that these
that the development of ORR activity was resulted from
modications induced improvement of catalytic activity for
improvement of the electrophysical properties of NGr by addi-
graphene materials for ORRs. Therefore, it is expected that this
tional doping of B or P, enhancement of asymmetry of spin
study will provide a new guideline for the catalyst design of the
density, facile electron transfer on the graphene basal plane,
graphene-derived catalysts that have signicantly higher ORR
activities than those of novel graphite- or CNT-derived catalysts.
and decrement of energy gap between highest occupied
molecular orbital (HOMO) and lowest unoccupied molecular
orbital (LUMO). It was reported that asymmetric spin density
provides oxygen reduction ability in the graphene layer.38 The
Acknowledgements
asymmetry of spin density in carbon materials is reinforced via
This work was supported by the National Research Foundation
doping of heteroatoms, and this mechanism has been provided
of Korea (NRF) grant funded by the Korea government (MEST)
in many cases such as N, B, P, and S as dopants.21,38,39 Therefore,
(no. 2009-0092783).
it suggests that additional doping of B or P in the NGr increased
the asymmetry of spin density in graphene and improved the
ORR activity. The second reason is the increment of electron
transfer on the graphene basal plane. Brownson et al. proposed
Notes and references
the low ability of the single layer graphene as an electro-
chemical material due to poor electron density on the graphene
basal plane.40 However, many DFT studies revealed that doping
of heteroatoms in graphene derived signicant electron density
1 Z. W. Chen, D. Higgins, A. P. Yu, L. Zhang and J. J. Zhang,
Energy Environ. Sci., 2011, 4, 3167–3192.
2 K. Gong, F. Du, Z. Xia, M. Durstock and L. Di, Science, 2009,
323, 760–764.
3698 | J. Mater. Chem. A, 2013, 1, 3694–3699
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
3 D. S. Geng, Y. Chen, Y. G. Chen, Y. L. Li, R. Y. Li, X. L. Sun, 24 C. H. Choi, S. H. Park and S. I. Woo, Int. J. Hydrogen Energy,
S. Y. Ye and S. Knights, Energy Environ. Sci., 2011, 4, 760–764. 2012, 37, 4563–4570.
4 G. Liu, X. G. Li, J. W. Lee and B. N. Popov, Catal. Sci. Technol., 25 L. Qu, Y. Liu, J. B. Baek and L. Dai, ACS Nano, 2010, 4, 1321–
2011, 1, 207–217. 1326.
5 G. Wu, K. L. More, C. M. Johnston and P. Zelenay, Science, 26 P. H. Matter, L. Zhang and U. S. Ozkan, J. Catal., 2006, 239,
2011, 332, 443–447. 83–96.
6 D. Chen, L. Tang and J. Li, Chem. Soc. Rev., 2010, 39, 3157– 27 D. Brigg and M. P. Seah, Practical Surface Analysis, John
3180. Wiley and Sons, New York, NY, USA, 1996.
7 Q. Yue, K. Zhang, X. Chen, L. Wang, J. Zhao, J. Liu and J. Jia, 28 M. Koh and T. Nakajima, Carbon, 1998, 36, 913–920.
Chem. Commun., 2010, 46, 3369–3371.
8 Y. Liang, Y. Li, H. Wang, J. Zhou, J. Wang, T. Regier and
H. Dai, Nat. Mater., 2011, 10, 780–786.
29 J. Ozaki, N. Kimura, T. Anahara and A. Oya, Carbon, 2007, 45,
1847–1853.
30 S. Jacques, A. Guette, X. Bourrat, F. Langlais, C. Guimon and
C. Labrugere, Carbon, 1996, 34, 1135–1143.
9 L. F. Lai, J. R. Potts, D. Zhan, L. Wang, C. K. Poh, C. H. Tang,
H. Gong, Z. X. Shen, L. Y. Jianyi and R. S. Ruoff, Energy 31 P. E. R. Blanchard, A. P. Grosvenor, R. G. Cavell and A. Mar,
Environ. Sci., 2012, 5, 7936–7942. Chem. Mater., 2008, 20, 7081–7088.
10 C. W. Tsai, M. H. Tu, C. J. Chen, T. F. Hung, R. S. Liu, 32 F. Claeyssens, G. M. Fuge, N. L. Allan, P. W. May and
W. R. Liu, M. Y. Lo, Y. M. Peng, L. Zhang, J. J. Zhang,
D. S. Shy and X. K. Xing, RSC Adv., 2011, 1, 1349–1357.
11 J. Y. Choi, D. Higgins and Z. W. Chen, J. Electrochem. Soc.,
2012, 159, B87–B90.
M. N. Ashfold, Dalton Trans., 2004, 3085–3092.
33 O. S. Panwar, M. A. Khan, M. Kumar, S. M. Shivaprasad,
B. S. Satyanarayana, P. N. Dixit and R. Bhattacharyya, Jpn.
J. Appl. Phys., 2009, 48.
12 H. R. Byon, J. Suntivich and Y. Shao-Horn, Chem. Mater., 34 H.-C. Kong, X. Yuan, X.-Y. Xia and Z.-F. Ma, Int. J. Hydrogen
2011, 23, 3421–3428. Energy, 2012, 37, 13082–13087.
13 Y. Li, W. Zhou, H. Wang, L. Xie, Y. Liang, F. Wei, J. C. Idrobo, 35 M. Ferrandon, A. J. Kropf, D. J. Myers, K. Artyushkova,
S.J. PennycookandH.Dai,Nat. Nanotechnol., 2012, 7, 394–400.
14 K. R. Lee, K. U. Lee, J. W. Lee, B. T. Ahn and S. I. Woo,
Electrochem. Commun., 2010, 12, 1052–1055.
15 H. Kim, K. Lee, S. I. Woo and Y. Jung, Phys. Chem. Chem.
Phys., 2011, 13, 17505–17510.
U. Kramm, P. Bogdanoff, G. Wu, C. M. Johnston and
P. Zelenay, J. Phys. Chem. C, 2012, 116, 16001–16013.
36 D. von Deak, D. Singh, E. J. Biddinger, J. C. King, B. Bayram,
J. T. Miller and U. S. Ozkan, J. Catal., 2012, 285, 145–151.
37 N. P. Subramanian, X. G. Li, V. Nallathambi, S. P. Kumaraguru,
H. Colon-Mercado, G. Wu, J. W. Lee and B. N. Popov, J. Power
Sources, 2009, 188, 38–44.
16 Z. W. Liu, F. Peng, H. J. Wang, H. Yu, W. X. Zheng and
J. Yang, Angew. Chem., Int. Ed., 2011, 50, 3257–3261.
17 L. Yang, S. Jiang, Y. Zhao, L. Zhu, S. Chen, X. Wang, Q. Wu, 38 L. P. Zhang and Z. H. Xia, J. Phys. Chem. C, 2011, 115, 11170–
J. Ma, Y. Ma and Z. Hu, Angew. Chem., Int. Ed., 2011, 50,
7132–7135.
18 S. Wang, E. Iyyamperumal, A. Roy, Y. Xue, D. Yu and L. Dai,
Angew. Chem., Int. Ed., 2011, 50, 11756–11760.
19 S. Wang, L. Zhang, Z. Xia, A. Roy, D. W. Chang, J. B. Baek and
L. Dai, Angew. Chem., Int. Ed., 2012, 51, 4209–4212.
11176.
39 Z. Yang, Z. Yao, G. Li, G. Fang, H. Nie, Z. Liu, X. Zhou,
X. Chen and S. Huang, ACS Nano, 2012, 6, 205–211.
40 D. A. C. Brownson, L. J. Munro, D. K. Kampouris and
C. E. Banks, RSC Adv., 2011, 1, 978–988.
¨
41 R. Pekoz and Sx. Erkoç, Physica E, 2009, 42, 110–115.
20 C. H. Choi, S. H. Park and S. I. Woo, J. Mater. Chem., 2012, 22, 42 Y. Chen, B. Gao, J. X. Zhao, Q. H. Cai and H. G. Fu, J. Mol.
12107–12115. Model., 2012, 18, 2043–2054.
21 C. H. Choi, S. H. Park and S. I. Woo, ACS Nano, 2012, 6, 7084– 43 Z. S. Wu, W. Ren, L. Xu, F. Li and H. M. Cheng, ACS Nano,
7091. 2011, 5, 5463–5471.
22 D. C. Marcano, D. V. Kosynkin, J. M. Berlin, A. Sinitskii, 44 T. Q. Lin, F. Q. Huang, J. Liang and Y. X. Wang, Energy
Z. Sun, A. Slesarev, L. B. Alemany, W. Lu and J. M. Tour,
Environ. Sci., 2011, 4, 862–865.
ACS Nano, 2010, 4, 4806–4814.
45 J. Aihara, J. Phys. Chem. A, 1999, 103, 7487–7495.
23 C. H. Choi, S. H. Park and S. I. Woo, Appl. Catal., B, 2012, 46 Z. W. Liu, F. Peng, H. J. Wang, H. Yu, W. X. Zheng and
119–120, 123–131.
X. Y. Wei, J. Nat. Gas Chem., 2012, 21, 257–264.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 3694–3699 | 3699