Shape fixing via salt recrystallization: A morphology-controlled approach to convert nanostructured polymer to carbon nanomaterial as a highly active catalyst for oxygen reduction reaction

Article abstract of DOI:10.1021/jacs.5b00292

Herein, we report a "shape fixing via salt recrystallization" method to efficiently synthesize nitrogen-doped carbon material with a large number of active sites exposed to the three-phase zones, for use as an ORR catalyst. Self-assembled polyaniline with a 3D network structure was fixed and fully sealed inside NaCl via recrystallization of NaCl solution. During pyrolysis, the NaCl crystal functions as a fully sealed nanoreactor, which facilitates nitrogen incorporation and graphitization. The gasification in such a closed nanoreactor creates a large number of pores in the resultant samples. The 3D network structure, which is conducive to mass transport and high utilization of active sites, was found to have been accurately transferred to the final N-doped carbon materials, after dissolution of the NaCl. Use of the invented cathode catalyst in a proton exchange membrane fuel cell produces a peak power of 600 mW cm^-2, making this among the best nonprecious metal catalysts for the ORR reported so far. Furthermore, N-doped carbon materials with a nanotube or nanoshell morphology can be realized by the invented method.

Full text of DOI:10.1021/jacs.5b00292

Subscriber access provided by SUNY DOWNSTATE
Article
Shape Fixing via Salt Recrystallization: A Morphology-Controlled
Approach to Convert Nano-structured Polymer to Carbon
Nanomaterial as a High Active Catalyst for Oxygen Reduction Reaction
Wei Ding, Li Li, Kun Xiong, Yao Wang, Wei Li, Yao Nie, Siguo Chen, XueQiang Qi, and Zidong Wei
J. Am. Chem. Soc., Just Accepted Manuscript • Publication Date (Web): 14 Apr 2015
Downloaded from http://pubs.acs.org on April 15, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 9
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Shape Fixing via Salt Recrystallization: A Morphology-
Controlled Approach to Convert Nano-structured Polymer to
Carbon Nanomaterial as a High Active Catalyst for Oxygen
Reduction Reaction
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Wei Ding, Li Li, Kun Xiong, Yao Wang, Wei Li, Yao Nie, Siguo Chen, Xueqiang Qi, Zidong Wei *
Chongqing Key Laboratory of Chemical Process for Clean Energy and Resource Utilization, School of Chemistry and
Chemical Engineering, Chongqing University, Chongqing, 400044; China; zdwei@cqu.edu.cn (Wei)
ABSTRACT: Herein, we report a “shape fixing via salt recrystallization” method to efficiently synthesize nitrogen-doped carbon
material with a large number of active sites exposed to the three-phase zones, for use as an ORR catalyst. Self-assembled polyani-
line with a 3D network structure was fixed and fully sealed inside NaCl via recrystallization of NaCl solution. During pyrolysis, the
NaCl crystal functions as a fully sealed nanoreactor, which facilitates nitrogen incorporation and graphitization. The gasification in
such a closed nanoreactor creates a large number of pores in the resultant samples. The 3D network structure, which is conducive
to mass transport and high utilization of active sites, was found to have been accurately transferred to the final N-doped carbon
materials, after dissolution of the NaCl. Use of the invented cathode catalyst in a proton exchange membrane fuel cell produces a
peak power of 600 mWcm-2, making this among the best non-precious metal catalysts for the ORR reported so far. Furthermore, N-
doped carbon materials with a nanotube or nanoshell morphology can be realized by the invented method.
plate synthesized catalyst catalyzed cathode produced a power
of 420 mW cm−2. Recently, by pyrolyzing well-defined metal–
ligand coordinated metal–organic frameworks (MOFs) or po-
rous organic polymer (POP), NPMCs were produced with
surface areas of up to 903 m2g-1 and an initial performance of
close to the level reached by platinum-based catalysts.6 In
addition to the number of active sites, the exposure of the ac-
tive sites to the interface, where electrons, protons, oxygen and
product water can flow in or exit out, is particularly important.
Otherwise, the active sites cannot be utilized by the ORR.7
Therefore, it would be better if the NPMCs were porous
enough or had an enough space to accommodate the linkage of
various species-transport channels to their active sites: voids
for O2 and water, a polymer chain for protons, and a carbon
framework for electrons. At this point, the traditional direct
pyrolysis of the mixture of nitrogen-containing compounds,
carbon, and transition metal precursors frequently fails in pro-
ducing the necessary porous structure, leading to relatively
poor transport properties and limited active sites that are ac-
INTRODUCTION
Proton exchange membrane fuel cells (PEMFCs) are elec-
trochemical power generators, with potential applications in
vehicle propulsion. To reduce their cost and encourage wide-
spread use, research has focused on replacing the expensive
Pt-based electrocatalyst with a lower-cost alternative for catal-
ysis of the oxygen reduction reaction (ORR) on the cathode
side of the cells.1 Great efforts have been undertaken to pro-
duce non-precious metal catalysts (NPMCs) by pyrolyzing
nitrogen-containing precursors on a carbon support material. 2
However, the power density of these NPMC-based cathodes
have been low compared with Pt-based cathodes, due to fewer
active-sites and poor mass-transport properties.3
One widely used strategy for the introduction of adequate
active-sites is to increase the surface area of the NPMCs by
using hard or soft template methods.4 However, the template
methods frequently fail in increasing the number of ORR ac-
tive-sites because the templates
Scheme 1. Schematic of the “Shape Fixing via Salt Recrystallization” method.
hinder contact between the nitro-
gen-containing precursors and
metals, which are needed to in-
duce or catalyze the transfor-
mation of nitrogen-containing
precursors into materials that can
catalyze the ORR. To increase
precursors-metals contact in tem-
plate methods, an advanced tem-
plate strategy has been developed
by using a low volatility precursor
to form more N-containing active
centers at the precursor’s molten
state.5 The fuel cell with this tem-
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 9
1
2
3
4
5
6
7
8
8
cessible for the ORR.3, Recently, we reported a space-
confinement-induced method to selectively synthesize planar
pyridinic and pyrrolic N -doped graphene. 9 It exhibited a high
yield of planar pyridinic and pyrrolic N-doped graphene and
acceptable ORR catalysis. The density of the exposed sites
activated by planar pyridinic and pyrrolic N, however, is still
not high enough, as the exposed active sites only appear at the
fringe of a large sheet of graphene.
acid (SA) were dissolved in 200 ml deionized water with
magnetic stirring at room temperature for 24 h. The stirring
was then stopped. After 10 min, a 50 mL aqueous solution of
9.294 g ammonium persulfate (APS) was added at an addition
rate of 0.8 ml/min, and the reaction was left for 48 h. The re-
sulting PANI was washed with water, methanol, and ether
several times. Finally, the product was dried under vacuum at
room temperature for 24 h.
9
Herein, we report a “shape fixing via salt recrystallization”
method to efficiently synthesize N-doped carbon material,
with many active sites exposed to the three-phase zones, as a
catalyst for the ORR. The procedure is shown in Scheme 1. A
3D PANI network was fabricated via a self-assembly process.
(Unless otherwise stated, PANI refers to the 3D PANI net-
work.) After adsorption of FeCl3, an 80 °C, super-saturated
NaCl solution was poured into a beaker containing PANI. The
water was then evaporated so that the NaCl recrystallized
around the PANI on the bottom of the beaker. This cycle of
the addition of NaCl solution, water evaporation and NaCl
recrystallization was repeated until the whole PANI was fully
buried so that it appeared to be tightly sealed inside NaCl crys-
tal. Thus, the original shape of the 3D PANI network would be
fixed by NaCl crystal. After the NaCl crystal-sealed PANI was
dehydrated at 120 °C, the temperature was raised to 900 °C at
a heating rate of 6 K min-1, held at 900 °C for 3 hours under
flowing N2, and finally cooled down to room temperature. The
NaCl crystal and iron species were washed off in a hot 0.5 M
H2SO4 solution. The resultant sample is referred to as CPANI-
Fe-NaCl. For comparison, samples obtained from pyrolysis
using NaCl crystal only, and those without either FeCl3 or
NaCl, were termed as CPANI-NaCl and CPANI, respectively.
In this process, the NaCl crystal serves as a fully closed nano-
reactor, which facilitates the N incorporation and graphitiza-
tion. Moreover, when the temperature reaches 800.9 °C, i.e.,
the melting temperature of NaCl, the NaCl molten salt can
facilitate the formation of active-sites. The original 3D struc-
ture of the PANI is well preserved and various pores, from
micro- to meso- to macro-sized, are formed in large quantities
in the high temperature pyrolysis, due to the gasification of
various raw materials in a closed space. Electrochemical eval-
uations showed that the invented catalyst exhibits excellent
conductivity, high ORR activity, and good stability in acidic
electrolytes. In the present work the half-wave potential of the
ORR on an electrode made of the invented catalyst was found
to lag behind the state-of-the-art carbon-supported platinum by
only 58 mV in acid electrolyte. The PEMFC, prepared with
the invented catalyst as the cathode, achieved a maximum
power density of 600 mW cm-2. It is demonstrated that N-
doped carbon materials with 3D network morphology, nano-
tubes and nanoshells, can be realized by the “shape fixing via
salt recrystallization” method. To our knowledge, this is the
first report of such a method to efficiently construct carbon
nanomaterials for catalysis of the ORR with controlled porous
structure, active sits and morphology.
Synthesis of PANI Nanotubes and PANI Nanoshells
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The PANI Nanotubes and PANI Nanoshells were synthe-
sized by a severe chemical oxidation and self-assembly
polymerization, similar to that reported previously 10. Aniline
(3.725 g) and SA (1.379 g for nanotubes, 3.408 g for
nanoshells) were dissolved in deionized water (200 ml). APS
(9.320 g) was also dissolved in deionized water (20 ml) in a
separate vial. The two solutions were vigorously mixed at
room temperature. A brown precipitate appeared immediately,
and the subsequent reaction was carried out at room tempera-
ture without any disturbance. After 7days, the precipitate was
filtered and purified using a Buchner funnel with a water aspi-
rator to wash the product several times with water, methanol,
and ether. Finally, the product was dried in vacuum at room
temperature for 24 h.
Preparation of CPANI-Fe-NaCl by the “Salt-
crystallization” method
Typically, 0.6 g of PANI and 0.6 g of FeCl3.6H2O were sus-
pended in 10 ml deionized water, with magnetic stirring, at
room temperature, for 12 h. After the adsorption of FeCl3, 15
ml supersaturated NaCl solution was added to the suspension,
and magnetic stirring was continued for 2 h. The suspension
was raised to 60 °C to evaporate off water. During the evapo-
ration, an 80°C over-saturated NaCl solution was continually
and slowly added into the suspension, until the NaCl crystal-
lized, encasing the PANI. To minimize the creation of air
pockets during evaporation, the sample was repeatedly trans-
ferred to a vacuum chamber and held at a pressure of -0.1 MPa
at 60°C for 1 hour after adding a small amount of 80°C super-
saturated NaCl solution. The composite was then dried under
vacuum at 60 °C for 24 h. The composite was then heated to
120 °C, after which the temperature was raised to 900°C at a
heating rate of 6 K min-1and held at 900°C for 3 hours under
flowing N2. The NaCl crystal and iron species in the sample
were finally washed off in 0.5 M H2SO4 solution at 80 °C.
Synthesis of CPANI and CPANI-NaCl
The CPANI-NaCl was prepared under conditions similar to
those used for CPANI-Fe-NaCl. For CPANI-NaCl, Fe salt was
not used, and NaCl was washed with room temperature water.
For CPANI, PANI was directly pyrolyzed, without applying
the NaCl crystallization procedure.
RESULTS AND DISCUSSION
The “shape fixing via salt recrystallization” procedure was
monitored by nitrogen adsorption-desorption at 77 K. As
shown in Figure 1, the adsorption hysteresis loop for PANI
suggests a mesoporous structure, with a pore size distribution
centered at 50.0 nm (via the Barrett-Joyner-Halenda method).
After being encased in NaCl crystal, the PANI display a very
small area by Brunauer-Emmett-Teller (BET) theory (0.7 m2 g-
1) compared to the original value (30.7 m2 g-1). The pore vol-
umes of the NaCl-buried PANI approach zero for all pore ra-
EXPERIMENTAL SECTION
Synthesis of PANI 3D Networks
Aniline monomer was distilled under reduced pressure be-
fore use. The PANI 3D network was synthesized by a slow
chemical oxidation and self-assembly polymerization. In the
typical synthesis, 3.728 g of aniline and 0.554 g of salicylic
ACS Paragon Plus Environment

Page 3 of 9
Journal of the American Chemical Society
1
2
dius, indicating that the PANI were fully encapsulated by the
carbon structure and lack of lattice fringe of the CPANI. Con-
NaCl crystal. After pyrolysis at 900°C and removal of the
NaCl crystal in a hot 0.5 M H2SO4 solution, the BET surface
area of the final CPANI-Fe-NaCl increased to 265.7 m2 g-1.
The BET surface increase is mainly due to the formation of
micropores after pyrolysis. As shown in Figure 1a, the pore
volumes for those radii ranging from 100 to 2.5 nm are un-
changed by the transition from PANI to CPANI-Fe-NaCl,
while pore volumes of less than 2.5 nm experience a substan-
tial change. These results indicate that the 3D structure was
well preserved and vast numbers of micropores were formed
in the “shape fixing via salt recrystallization” method.
versely, the CPANI-NaCl shows relatively clear graphitic
fringes, suggesting an improved graphitic structure. This result
is also supported by the Raman spectra (see Figure S5), which
show the relative higher intensity ratio of G-band to D-band
(IG/ID) for the CPANI-NaCl, implying that the graphitization
degree of the CPANI-NaCl is higher than that of the CPANI.
This is attributed to the effect of molten salts that promotes the
degree of graphitization in the heat treatment process, which is
consistent with the previous finding11. The clearest lattice
fringe and the highest IG/ID, corresponding to CPANI-Fe-NaCl
(Figure 3c, 3f, and S5), indicate that the degree of graphitic
structure can be further improved with the introduction of an
Fe element.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. (a) Pore size distribution for PANI, PANI in NaCl, and
CPANI-Fe-NaCl, assessed by BJH. (b) Nitrogen sorption iso-
therms for PANI, PANI in NaCl, and CPANI-Fe-NaCl.
Figure 2a show that the PANI 3D networks were comprised
of PANI nanofibers with a diameter of ∼100 nm. With the
“shape fixing via salt recrystallization” method, as shown in
Figure 2b, the PANI 3D network morphologies were faithfully
transferred to the 3D carbon networks (see also Figure S1a–b).
In contrast, the TEM and SEM images of CPANI synthesized
by traditional pyrolysis reveal a large-scale, spherical or lump-
type morphology (see Figure S2a and Figure S2b). Only one
tiny difference can be observed between the morphology of
the PANI 3D network and that of the carbon 3D network; that
is, there are raised knots on the surface of the PANI but con-
cavities on the CPANI-Fe-NaCl. This further demonstrated
that a carbon nanotube with a diameter of∼200 nm, length
of∼600 nm and thickness of ∼50 nm (Figure S3a–b), as well
as a carbon nanoshell, can be synthesized via the proposed
method; all resultant samples accurately inherit their precur-
sors’ original structures (Figure 2c, 2d, 2e, and 2f; see also
Figure S1c–f). These excellent conversions demonstrate the
high efficiency of the “shape fixing via salt recrystallization”
method in controlling morphology and pore structure. To in-
vestigate the role of Fe during the synthesis, we synthesized
CPANI-Fe by pyrolyzing PANI absorbed FeCl3. Our results
show that Fe ions affect the graphitic structure during the syn-
thesis but they cannot preserve the original sharp of PANI. As
shown in Figure S4a and b, the original nano-fiber morpholo-
gy of PANI transform into a lump-type and graphitic shells-
like morphology. The graphitic shells-like morphology doesn’t
appear in the case of PANI pyrolysis without Fe participation,
i.e., CPANI. The graphitic shells appearing in CPANI-Fe
sample tells that the Fe species catalytically transform PANI
into graphitic carbon. In contrast, the CPANI-NaCl preserves
network morphology after pyrolysis (Figure S4c) indicating
the original sharp of PANI can be preserved by NaCl alone.
Figure 2. TEM images of as-prepared 3D PANI network (a),
PANI nanotubes (c), PANI nanoshell (e), and their corresponding
carbonized products: CPANI-3D-Fe-NaCl (b), CPANI-NT-Fe-
NaCl (d), and CPANI-NS-Fe-NaCl (f).
Figure 3．High-magnification TEM images showing the edges of
CPANI (a, d), CPANI-NaCl (b, e) and CPANI-Fe-NaCl (c, f).
High-resolution
transmission
electron
microscopy
(HRTEM), as shown in Figure 3, demonstrated the amorphous
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 9
1
2
3
4
TGA/DTA at a heating rate of 6 K min-1 in N2 atmosphere
away from the surface. This will be beneficial for the active
was used to gauge how readily the precursors transform to
carbon materials via the “shape fixing via salt recrystalliza-
tion” method. As shown in Figure 4, PANI-NaCl exhibits ex-
cellent thermal stability, with weight loss starting at 443.8 °C.
The PANI, however, shows poor thermal stability. Its decom-
position started at temperatures as low as 199.8 °C. At 900 °C,
only 4.1 % of the PANI residues remained, while 74.3 % and
67.1 % of the PANI-NaCl and PANI-Fe-NaCl residues re-
mained, respectively. In addition, no obvious endothermic or
exothermic processes were found in the PANI. Because of the
uneven degree of the PANI polymerization, its decomposition
(exothermic), dehydrogenation (endothermic) and carboniza-
tion (endothermic) may occur randomly and thus not occur at
a certain temperature range. This result indicates a random
transformation of PANI to carbon through direct pyrolysis.
For PANI-NaCl, an endothermic process with weight loss was
found at 520.7 °C, followed by the occurrence of another en-
dothermic process without weight loss at 577.1 °C. The two
endothermic processes are attributed to the dehydrogenation of
PANI at the lower temperature and the graphitization at the
higher. With the addition of FeCl3, only one exothermic pro-
cess at 337.6 °C exhibited weight loss. One possible explana-
tion is the catalytic effect of iron species during pyrolysis. The
TGA/DTA plot of NaCl shows only one endothermic process
at 800.9 °C corresponding to the melting of NaCl crystal (Fig-
ure S6). With the NaCl crystal melting at 800.9 °C, the graphi-
tization of the pyrolyzed PANI was conducted in NaCl molten
salts. It should be noted that the difference in the specific
weights of NaCl molten salts (1.5 g cm-3) and pyrolyzed PANI
(1.8～2.1 g cm-3) is too small to overcome the high viscosity
of the NaCl molten salt, thereby avoiding phase separation.
Therefore, the graphitization of the pyrolyzed PNAI takes
place inside the NaCl molten salts. As a matter of fact, as the
samples cooled down to room temperature, the final carbon-
based products were still sealed inside the NaCl crystal, just as
shown in Scheme 1. The obvious weight loss after 800.9 °C in
the TGA curve of Figure 4 comes from the vaporization of
NaCl molten salts. As a result, the dehydrogenation, nitrogen
doping and graphitization of PANI, as well as the gasification
of various additives in the present method, are always con-
ducted in a sealed system. This will, at most, prevent weight
loss and produce large quantities of pores in the final product,
due to the trapped gas. As mentioned above, the invented
method is conducive to increased active sites, induced by pla-
nar pyridinic N and pyrrolic N. The substance interactions and
phase transitions may play important roles during the pyrolysis.
The polyaniline is in a doped form. Its backbone carries posi-
tive charge.12 Thus, during the absorption of FeCl3, Cl- ions
will be preferentially adsorbed on the positive-charged PANI,
and then Fe3+ ions as shown in Figure 4. The outside of PANI-
FeCl3 was first covered by NaCl solution and then NaCl solids
with the recrystallization of NaCl. The thick NaCl solid layer
prevents the PANI from decomposition at a relatively lower
temperature. With decomposition temperature increasing, the
dedoping and dehydrogenation of PANI may finish before
PANI structure collapse. Nitrogen atoms with a lone pair elec-
tron expose to the outside with the dedoping and dehydrogena-
tion of PANI. The exposed N reacts with Fe3+ ions, and forms
FeNx/C complex. With the NaCl crystal melting at 800.9 °C,
the excess Fe species are dissolved in molten NaCl and flow
sites because too much Fe species would destroy them at high
temperature.13
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. TGA-DTA plot of PANI, PANI-NaCl, and PANI-Fe-
NaCl; and schematic of substance interactions during the synthe-
sis.
X-ray photoelectron spectroscopy (XPS) analysis was per-
formed to investigate the content and chemical state of the
nitrogen on the surface of the prepared catalysts. The existence
of C-N and C-C (graphite-type) bonds in the matrix confirms
the N-doped carbon structure of the prepared catalysts (see
Figure S7). The relative N/C ratio was only 3.04 % in CPANI
but was 3.52 % and 5.85 % in CPANI-NaCl and CPANI-Fe-
NaCl, respectively; indicating that nitrogen loss during pyroly-
sis at high temperatures was alleviated in the “shape fixing via
salt recrystallization” method. The high resolution N1 s spec-
tra for all catalysts were fitted with five different signals, hav-
ing binding energies of 398.6, 399.2, 400.3, 401.4, and ~403
eV, corresponding to pyridinic N, nitrile N, pyrrolic N, qua-
ternary N, and oxidized pyridinic N, respectively (Figure S8).
Pyridinic N (most likely including Me−N) and pyrrolic N are
generally believed to activate neighboring carbon atoms for
catalysis of the ORR.9, 14 As shown in Figure S8, the content of
ACS Paragon Plus Environment

Page 5 of 9
Journal of the American Chemical Society
1
2
pyridinic and pyrrolic N in the CPANI-NaCl and the CPANI-
supported platinum catalyst (Johnson Matthey, UK, 40 wt %,
Fe-NaCl is higher than that in CPANI, indicating that the
“shape fixing via salt recrystallization” method is more suita-
ble for facilitating more active-site formation. Besides, 11 %
of the N species in CPANI-NaCl are nitrile N, but no nitrile N
is detected in CPANI-Fe-NaCl, indicating that the addition of
iron species accelerates N incorporation into carbon-ring. The
CPANI that have the highest content of quaternary N show the
poorest activity for catalysis of the ORR in acidic medium.
Figure S9 show that Fe presents in the CPANI-Fe-NaCl as
Fe3+ (3.18 atm%), which may exist as a stable complex form
rather than simple inorganic ions because the CPANI-Fe-NaCl
was treated in a hot H2SO4 solution. Na and Cl were not de-
tected in the CPANI-Fe-NaCl after hot H2SO4 solution wash-
ing. It means that NaCl has been washed off by a hot solution.
For CPANI-NaCl, Na and Cl were detected as form of NaCl.
This is because of the residual NaCl in sample after room
temperature water washing.
denoted as JM Pt/C) at a Pt loading of 50 μgPt cm-2. These
results are comparable with those of the best NPMCs reported
to date. 6b, 9, 15, 16 The H2O2 yield was as low as 1.5 % on a
CPANI-Fe-NaCl-catalyzed electrode and 6.8% for a CPANI-
NaCl-catalyzed electrode at the relative positive potential of
0.8 V (vs RHE). The H2O2 yield decreased with a negative-
shifting potential. At 0.5 V (vs RHE), it was only 0.28% on the
CPANI-Fe-NaCl-catalyzed electrode and was 0.85% on the
CPANI-NaCl-catalyzed electrode, which is comparable with
that on the JM Pt/C electrode (0.30%). In any respect, CPANI
exhibited a poor activity for the ORR. The electron transfer
number for the ORR on the CPANI-Fe-NaCl-catalyzed elec-
trode was estimated according to the Koutecky–Levich plot.
As shown in Figure S13, the electron transfer number is 3.9 ±
0.1 on the CPANI-Fe-NaCl-catalyzed electrode in the poten-
tial range of 0.6-0.3 V, indicating a four-electron pathway for
catalysis of the ORR and explaining the low H2O2 yield on the
electrodes made of the prepared samples. The stability of the
CPANI-Fe-NaCl- and CPANI-NaCl-catalyzed electrodes,
together with the JM-Pt/C-catalyzed electrode, was assessed
by cycling the electrodes in a potential range of 0-1.2 V in N2-
saturated 0.5 M H2SO4 (see Figure S14). No noticeable change
in catalytic activity of the ORR was observed on the CPANI-
Fe-NaCl-catalyzed electrode, but the ORR half-wave poten-
tials on the electrodes catalyzed by CPANI-NaCl and by JM
Pt/C were negatively shifted by 27 mV and by 73 mV, respec-
tively, after CV cycling, indicating the excellent stability of
CPANI-Fe-NaCl but relatively poor stability of CPANI-NaCl
and JM-Pt/C. The relatively poor stability of CPANI-NaCl is
attributed to the existence of large numbers of nitrile species in
the CPANI-NaCl, which may accelerate destruction of the
catalyst under harsh conditions.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Electrical conductivity is a prominent concern for materials
used in electrochemical applications. Electrochemical imped-
ance spectroscopy (EIS) at high frequency (see the Supporting
Information for details) revealed that the electrical resistance
of the CAPNI-NaCl and the CPANI-Fe-NaCl are on the same
order as that of commercial Vulcan XC-72R carbon (Figure
S10). In other words, the electrical conductivity of the
CAPNI-NaCl and CPANI-Fe-NaCl is as good as that of com-
mercial Vulcan XC-72R carbon. Conversely, CPANI and
PANI demonstrate very poor electrical conductivity, as illus-
trated in Figure S10. These physical analysis data combined
with TEM results demonstrate that NaCl preserves the original
morphology of precursors and promote the degree of graphiti-
zation; while iron species enhance both of graphitization and
nitrogen incorporation during the synthesis.
The cyclic voltammograms (CVs) of CPANI and CPANI-
NaCl in N2-saturated H2SO4 solution are virtually featureless,
while the CV of CPANI-Fe-NaCl reveals a pair of well-
developed redox peaks at~0.63 V vs the reversible hydrogen
electrode (RHE) (see in Figure S11). The peaks are expected
for a reversible process involving surface Fe species (Fe3+ and
Fe2+). The Fe species are very stable. They cannot be removed
by either acidic solution treatment at 80 °C or by long-term
electrochemical cycling in acidic solution (see Figure S14).
The double layer capacities of the CPANI-Fe-NaCl and the
CPANI-NaCl are similar to each other, but they are twice
larger than that of CPANI. However, active sites of a metal
catalyst can but active sites of a nitrogen-doped carbon cata-
lyst cannot be checked by the CVs. The CVs of the electrode
made of nitrogen-doped carbon catalysts only tell the capacity
of its electric double layer, which indicates the sol-
id/electrolyte interface of the electrode but does nothing with
the activity.
The activity of the catalysts was evaluated using a rotating
ring–disk electrode (RRDE). All of the RRDE experiments
were performed in 0.1 m HClO4. The polarization curves and
the Tafel slops for the catalysts are shown in Figure 5a and
Figure S12, respectively. The CPANI-Fe-NaCl catalyst exhib-
ited the highest activity, and the CPANI-NaCl metal-free cata-
lyst also showed a good activity for the ORR. The ORR half-
wave potentials measured on a CPANI-Fe-NaCl-catalyzed
electrode and a CPANI-NaCl-catalyzed electrode were only
58 mV and 102 mV behind that on a state-of-the-art carbon-
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 9
1
2
containing species, Fe elements exist in FeNx/C complex, of
which FeN4/C and N-FeN2+2/C are highly active for the ORR
in acid medium.17 In contrast, some argued that Fe only pro-
motes key properties of catalyst, such as electronic conductivi-
ty, morphology, and nitrogen-doping level, but it does not
directly participate in active site.18 We made a comparative
investigation between Fe-free and Fe-doping catalyst, that is,
CPANI-NaCl and CPANI-Fe-NaCl. Our results show that
their electronic conductivity, morphology, and nitrogen-
doping level are quite similar to each other but their ORR ac-
tivities are different. As shown in Figure 5a, the electrode
made of the CPANI-NaCl catalysts (Fe-free) shows an excel-
lent ORR activity, indicating the nitrogen-doped carbon with-
out Fe also catalyzes the ORR. This is ascribed to numerous
inner pores formed in the present method. The sites activated
by planar pyridinic and pyrrolic N atoms with sp2 hybrid or-
bitals, which appear at the fringes of a graphene sheet and the
fringes of graphene inner pores, are active for the ORR. The
present method makes such active sites increase obviously, for
instance, 68% for the given graphene sheet as shown in Figure
5c. On the other hand, the electrode made of the CPANI-Fe-
NaCl catalysts (Fe-doping) shows much better ORR activity.
The catalytic role of Fe complex for the ORR is obvious,
which is in accordance with previous studies based on Möss-
bauer spectroscopy.15a, 19 According to Frédéric Jaouen’s re-
search result, the Fe/N/C-catalyst contains two types of FeNx
sites assigned to micropore-hosted sites in edges and gra-
phene-defect-hosted sites in inner plane, while the later are
more stable than the former one during the carbon corrosion.19
The present method might benefit more FeNx sites formation
owing to the existence of numerous inner pores, as shown in
Figure S19. At this stage, we cannot exclude another role of
Fe species in bettering graphitic structure, which is well
known to be conducive for the catalysis of the ORR as report-
ed before.9, 15, 19
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The performance of the single cell with these catalyst-based
cathodes and Pt/C anodes is shown in Figure 5b. An open-
circuit voltage (OCV) of 0.85 V was observed and the current
density reached as high as 2.1 A cm-2 at 0.28 V. A higher
OCV of 0.9 V was observed when the MEA was prepared
using Nafion 117 to substitute Nafion 112 to reduce the cross-
over current (see in Figure S16). A maximum power output of
600 mW cm-2 was achieved with the CPANI-Fe-NaCl-
catalyzed cathode, with 300 mW cm-2 for the metal-free,
CPANI-NaCl-catalyzed cathode. In contrast, the maximum
power output is only 180 mW cm-2 for the CPANI-catalyzed
cathode, which is consistence with previous report about the
single cell performance of a NC-catalyzed cathode obtained
from direct pyrolyzed PANI.16 The CPANI-Fe catalyst pro-
duced without NaCl, also show a poor performance with the
maximum power output of only 284 mW cm-2 (Figure S17). It
tells that NaCl participation is very important in synthesis of
high active nitrogen-doped carbon catalysts.
Figure 5. (a) Steady-state plots of ORR polarization (bottom) and
H2O2 yield (top) for different catalysts in O2-saturated 0.1 m
HClO4. The loading is 50 μgPt cm-2 for Pt/C (40%) and is 0.6 mg
cm-2 for non-Pt catalyst. (b) Polarization curves and correspond-
ing power densities of membrane electrode assemblies fabricated
with the CPANI-Fe-NaCl, CPANI-NaCl and CPANI cathode
catalysts. (c) Schematic of active sites on edges and in pores.
To investigate whether the remained NaCl in CPANI-NaCl
involve the ORR, the ORR activity of CPANI-NaCl was eval-
uated after repeatedly washing with hot water until no Cl- was
detected. As shown in Figure S15, the activities show un-
changed after removal of NaCl, indicating that NaCl do not
involve the catalysis of the ORR. As mentioned above, Fe
species are hardly removed from the CPANI-Fe-NaCl, even
by being treated in a hot acidic solution or by being long-term
cycled in acidic solution (Figure S14a). It suggests that Fe
elements may exist in a very stable complex rather than simple
inorganic ions, otherwise, Fe would be leached off from the
catalysts. According to recent studies relied on Mössbauer
spectroscopy and X-ray absorption spectroscopy about Fe
state after pyrolysis of the mixture of PANI and iron-
In the present work, the loading of nitrogen-doped carbon
catalysts in the catalyst layer is much higher than that of Pt
catalysts, 4.0 mg cm-2 vs 0.1 or 0.3 mg cm-2. Thus the ohmic
voltage loss in the catalyst layer is not negligible owing to the
thick catalyst layer but negligible in the case of Pt/C catalysts
owing to its ultrathin catalyst layer. The ohmic resistance
would affect the I-V slope. As shown in Figure 5b, the I-V
slopes of the three types of catalysts are different from each
ACS Paragon Plus Environment

Page 7 of 9
Journal of the American Chemical Society
1
2
3
4
5
other. The I-V slopes indicate the electric resistance of the
Supporting Information
three types of cathodes, that is, RCPANI-Fe-NaCl ≈RCPANI- NaCl
≪
RCPANI. This is consistence with the EIS result illustrated in
Figure S10.
Experimental details and more characterization and results. This
material is available free of charge via the Internet at
http://pubs.acs.org.
An internal comparison has been made with the commercial
Pt/C catalyst. As shown Figure S17, with the JM Pt/C-
catalyzed cathodes, the maximum power output is 450 mW
cm-2 at Pt loading of 0.1 mg cm-2 and is 820 mW cm-2 at 0.3
mg Pt cm-2. The CPANI-Fe-NaCl with the maximum power
output of 600 mW cm-2 show a comparable performance to
that of the commercial Pt/C. In addition, the single cell per-
formances of the different morphology CPANI-Fe-NaCl cata-
lyzed cathodes are evaluated as shown in Figure S18. The
maximum power output is 530 mW cm-2 with the nanotube
CPANI-Fe-NaCl-catalyzed cathode, and is 490 mW cm-2 with
the nanoshell CPANI-Fe-NaCl-catalyzed cathode. The cell
performances observed in these tests are among the best re-
ported world-wide, to date.6 In contrast to efforts that focus on
improving the surface area, the best catalyst found in the pre-
sent work has a BET surface area of only 265.7 m2 g-1, but
exhibits excellent cell performance despite this. This is at-
tributed to the unique advantages of the “shape fixing via salt
recrystallization” method, by which a high microporosity is
achieved, with a high density of ORR active sites located
along the efficient, mass transport pathway (composed of me-
so- and macropores), leading to a high utilization of active
sites.
6
7
8
9
AUTHOR INFORMATION
Corresponding Author
zdwei@cqu.edu.cn
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Notes
The authors declare no competing financial interests.
ACKNOWLEDGMENT
This work was financially supported by the China National 973
Program (2012CB720300 and 2012CB215500), by the NSFC of
China (Grant Nos. 21436003 and 51272297).
REFERENCES
(1) (a) Gasteiger, H. A.; Markovic, N. M. Science 2009, 324, 48;
(b) Strasser, P.; Koh, S.; Anniyev, T.; Greeley, J.; More, K.; Yu, C.;
Liu, Z.; Kaya, S.; Nordlund, D.; Ogasawara, H.; Toney, M. F.; Nils-
son, A.. Nature Chem. 2010, 2, 454; (c) Wang, C.; Vliet, D.; More, K.
L.; Zaluzec, N. J.; Peng, S.; Sun, S.-H.; Daimon, H.; Wang, G.-F.;
Greeley, J.; Pearson, J.; Paulikas, A. P.; Karapetrov, G.; Strmcnik, D.;
Markovic, N. M.; Stamenkovic, V. R.. Nano Lett. 2010, 11, 919; (d)
Wu, J.; Zhang, J.; Peng, Z.; Yang, S.; Wagner, F. T.; Yang, H.. J. Am.
Chem. Soc. 2010, 132, 4984.
(2) (a) Bezerra, C. W. B.; Zhang, L.; Lee, K.; Liu, H.; Marques, A.
L. B.; Marques, E. P.; Zhang, J.. Electrochim. Acta 2008, 53, 4937;
(b) Lefevre, M.; Proietti, E.; Jaouen, F.; Dodelet, J-P.. Science 2009,
324, 71; (c) Pylypenko, S.; Mukherjee, S.; Olson, T. S.; Atanassov, P..
Electrochim. Acta 2008, 53, 7875.
(3) Jaouen, F.; Proietti, E.; Lefèvre, M.; Chenitz, R.; Dodelet, J.-P.;
Wu, G.; Zelenay, P.. Energy Environ. Sci. 2011, 4, 114.
(4) (a) Yang, S.-B.; Feng, X.-L.; Wang, X.-C.; Müllen, K.. Angew.
Chem. 2011, 50, 5339; (b) Rao, C.-V.; Cabrera, C.-R.; Ishikawa, Y.. J.
Phys. Chem. Lett. 2010, 1, 2622; (c) Wang, X. B.; Liu, Y. Q.; Zhu,
D. B.; Zhang, L.; Ma, H. Z.; Yao, N.; Zhang, B. L.. J. Phys. Chem. B
2002, 106, 2186.
CONCLUSION
In summary, we have demonstrated a “shape fixing via salt
recrystallization” method to efficiently synthesize N-doped
carbon nanomaterials with a high density of active sites, as an
ORR catalyst. In this method, the NaCl crystal functions as a
fully sealed nanoreactor, facilitating the N incorporation and
graphitization. The gas from gasification in such a closed
nanoreactor produces a large quantity of pores in resultant
samples. As for an Fe-free catalyst, this method makes it pos-
sible for the active sites (activated by planar pyridinic and
pyrrolic N atoms with sp2 hybrid orbitals, which only appear at
the outside fringe of a graphene sheet) to appear in large quan-
tities on the edges of numerous pores, as shown in Figure 5c.
As for an Fe-doping catalyst, the present method benefits for-
mation of more FeNx sites in inner pores and graphitic struc-
ture. A favorable catalyst structure leads to the availability of
efficient mass transport pathways and a high utilization of the
active sites. It results in an excellent ORR activity with a half-
wave potential only 58 mV behind that of Pt/C in an acidic
medium. The PEMFC with the CPANI-Fe-NaCl-catalyzed
cathode outputs a peak power of 600 mW cm-2, which is
among the best for non-precious metal catalysts reported so far
for the ORR. Additionally, no other method is comparable to
this invention, in terms of its high yield per batch of N-doped
carbon catalysts. The “shape fixing via salt recrystallization”
method represents a powerful approach for the preparation of
high-performance carbon nanostructures by confined pyrolysis
of nanopolymers and, more importantly, provides insight into
the design of advanced PEMFC cathode catalysts.
(5) Serov, A.; Artyushkova, K.; Atanassov, P.. Adv. Energy Mater.
2014, 4, 1301735.
(6) (a) Proietti, E.; Jaouen, F.; Lefevre, M.; Larouche, N.; Tian, J.;
Herranz, J.; Dodelet, J. P.. Nat. Commun. 2011, 2, 416; (b) Yuan, S.-
W.; Shui, J.-L.; Grabstanowicz, L.; Chen, C.; Commet, S.; Reprogle,
B.; Xu, T.; Yu, L. P.; Liu, D. J.. Angew. Chem. Int. Ed. 2013, 52,
8349; (c) Tian, J.; Morozan, A.; Sougrati, M. T.; Lefèvre, M.;
Chenitz, R.; Dodelet, J.-P.; Jones, D.; Jaouen, F.. Angew. Chem. Int.
Ed. 2013, 52: 6867.
(7) (a) Wei, Z.-D.; Chen, S. G.; Liu, Y; Sun, C. X.; Shao, Z. G.;
Shen, P. K.. J. Phys. Chem. C 2007, 42, 15456; (b) Zhang, H.; Wang,
X.; Zhang, J.; Zhang, J.. In PEM Fuel Cell Electrocatalysts and Cata-
lyst Layers; Zhang, J., Ed.; Springer London: 2008, 889.
(8) (a) Ignaszak, A.; Ye, S.; Gyenge, E.. J. Phys. Chem. C 2008,
113, 298; (b) Antolini, E.. Appl. Catal, B 2009, 88,1.
(9) Ding, W.; Wei, Z.-D.; Chen, S. G.; Qi, X. Q.; Yang, T.; Hu, J. -
S.; Wang, D.; Wan, L. -J.; Alvi, S. F.; Li, L.. Angew. Chem. Int. Ed.
2013, 52, 11755.
(10) Zhang, L.J.; Wan, M. X.. Adv. Funct. Mater. 2003, 13, 815.
(11) Kamali, A. R.; Fray. D. J.. Carbon 2013, 56, 121.
(12) Wang, Y.; Guan, X. N.; Wu, C.-Y.; Chen, M.-T.; Hsieh, H.-
H.; Tran, H. D.; Huang, S.-C.; Kaner, R. B.. Polym. Chem. 2013, 4,
4814.
(13) (a) Nabae, Y.; Moriya, S.; Matsubayashi, K.; Lyth, S. M.;
Malon, M.; Wu, L.; Islam, N. M.; Koshigoe, Y.; Kuroki, S.;
ASSOCIATED CONTENT
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 9
1
2
3
4
5
6
7
8
Kakimoto, M.; Miyata, S.; Ozaki, J.. Carbon, 2010, 48, 2613; (b)Wu,
L.-B.; Nabae, Y.; Moriya, S.; Matsubayashi, K.; Islam, N. M.;
Kuroki, S.; Kakimoto, M.; Ozaki, J. Miyata, S.. Chem. Commun.,
2010, 46, 6377.
(14) (a) Yu, D. S.; Zhang, Q.; Dai, L. M.. J. Am. Chem. Soc. 2010,
132, 15127; (b) Liu, G.; Li, X.G.; P.; Ganesan, Popov, B. N.. Electro-
chim. Acta 2010, 55, 2853; (c) Liu, G.; Li, X.G.; Lee, J. W.; Popov.
B. N. Catal. Sci. Technol. 2011, 1, 207; (d) Liang, Y.; Li, Y.; Wang,
H.; Zhou, J.; Wang, J.; Regier, T.; Dai, H.. Nat. Mater. 2011, 10, 780.
(e) Wang, S.; Zhang, L.; Xia, Z.; Roy, A.; Chang, D.-W.; Baek, J.-B.;
Dai, L.. Angew. Chem. 2012, 51,4209.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(15) Liang, H.-W.; Wei, W.; Wu, Z. -S. Feng, X. l. Müllen, K.. J.
Am. Chem. Soc. 2013, 135, 16002.
(16) Wu, G.; More, K. L.; Johnston, C. M.; Zelenay, P.. Science
2011, 332, 443.
(17) (a) Kramm, U. I; Lefevre, M; Larouche, N; Schmeisser, D;
Dodelet, J.-P.. J. Am. Chem. Soc. 2014, 136, 978; (b) ramm, . .
erranz, J. arouche, N. Arruda, T. M. efevre, M. Jaouen, F.
Bogdanoff, P.; Fiechter, S.; Abs-Wurmbach, I.; Mukerjee, S.; Dodelet,
J.-P.. Phys. Chem. Chem. Phys. 2012, 14, 11673 (c) Zhang, S. M.; Liu,
B.; Chen, S. L.. Phys. Chem. Chem. Phys. 2013, 15, 18482.
(18) (a) Nallathambi, V.; Lee, J. W.; Kumaraguru, S. P.; Wu, G.;
Popov, B. N. J.. Power Sources 2008, 183, 34. (b) Woods, M. P.;
Biddinger, E. J.; Matter, P. H.; Mirkelamoglu, B.; Ozkan, U. S.. Catal.
Lett. 2010, 136, 1.
(19) Goellner, V.; Baldizzone, C.; Schuppert, A.; Sougrati, M. T.;
Mayrhofer, K.; Jaouen, F.. Phys. Chem. Chem. Phys. 2014, 16, 18454.
ACS Paragon Plus Environment

Page 9 of 9
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment