Fe-N/C oxygen reduction catalysis prepared by covalent attachment of 1,10-phenanthroline to a carbon surface

Article abstract of DOI:10.1149/1.3541256

Nonprecious metal catalysts (NPMC) for the oxygen reduction reaction were prepared by surface modification of a carbon black with 5,6-diamino-1,10- phenanthroline through diazonium and benzimidazole coupling chemistry. Both methodologies covalently attach 1,10-phenanthroline to the surface. Oxygen-free cyclic voltammograms (CV) obtained for these modified carbons showed a clearly defined redox peak at ca. 650 mV versus normal hydrogen electrode, which we attribute to phenanthroline-type groups on the carbon surface. Introduction of Fe and subsequent pyrolysis at 700°C under a N H₃ / N₂ atmosphere yielded catalysts that were highly active toward the oxygen reduction reaction. The redox peak at 650 mV was present in the oxygen-free CVs for the most active catalysts, though less intense. Nevertheless, this indicates the presence pyridinic-type active sites in the heat treated catalysts and may be a useful tool in the study of NPMC.

Full text of DOI:10.1149/1.3541256

Journal of The Electrochemical Society, 158 ͑3͒ B331-B336 ͑2011͒
B331
0
013-4651/2011/158͑3͒/B331/6/$28.00 © The Electrochemical Society
Fe–NÕC Oxygen Reduction Catalysis Prepared by Covalent
Attachment of 1,10-Phenanthroline to a Carbon Surface
a
b
a, ,z
Allen D. Pauric, Brian J. MacLean, and E. Bradley Easton *
a
Faculty of Science, University of Ontario Institute of Technology, Oshawa, Ontario, Canada L1H 7K4
b
Department of Chemistry, St. Francis Xavier University, Antigonish, Nova Scotia,
Canada B2G 2W5
Nonprecious metal catalysts ͑NPMC͒ for the oxygen reduction reaction were prepared by surface modification of a carbon black
with 5,6-diamino-1,10-phenanthroline through diazonium and benzimidazole coupling chemistry. Both methodologies covalently
attach 1,10-phenanthroline to the surface. Oxygen-free cyclic voltammograms ͑CV͒ obtained for these modified carbons showed
a clearly defined redox peak at ca. 650 mV versus normal hydrogen electrode, which we attribute to phenanthroline-type groups
on the carbon surface. Introduction of Fe and subsequent pyrolysis at 700°C under a NH /N atmosphere yielded catalysts that
3
2
were highly active toward the oxygen reduction reaction. The redox peak at 650 mV was present in the oxygen-free CVs for the
most active catalysts, though less intense. Nevertheless, this indicates the presence pyridinic-type active sites in the heat treated
catalysts and may be a useful tool in the study of NPMC.
©
2011 The Electrochemical Society. ͓DOI: 10.1149/1.3541256͔ All rights reserved.
Manuscript submitted July 21, 2010; revised manuscript received November 29, 2010. Published January 20, 2011. This was Paper
5
12 presented at the Vancouver, Canada, Meeting of the Society, April 25–30, 2010.
4
Widespread adoption of proton exchange membrane fuel cells
PEMFC͒ is hindered by various issues. Of critical importance is
have observed a wave for some sputtered Fe/N/C catalysts, and
1,2
͑
many others have observed this at times for catalysts prepared by
wet chemical methods. The exact natures of these waves have
never been fully understood.
Our approach to NPMCs has been to employ surface chemical
functionalization to more systematically synthesize the active site in
order to better relate structure and activity. Our initial work investi-
gated catalysts synthesized by attaching nitrogen-containing func-
1
4
the reduction or complete elimination of platinum. The high cost and
limited availability of platinum are roadblocks to the large scale
3
implementation of PEMFCs. Consequently, substantial research has
4
-8
been conducted with nonprecious metal catalysts ͑NPMC͒. They
promote the oxygen reduction reaction ͑ORR͒ at the PEMFC cath-
9
ode. However, oxygen reduction is slow even on platinum, making
the development of a good nonplatinum based catalyst a worthwhile
goal.
tional groups to a carbon black surface via a series of
16
aminosilanes. After coordination of iron, samples were pyrolyzed
to induce catalytic activity. One major advantage of this method is
that chemisorption is more specific than physisorption, thereby al-
lowing the attachment to occur at the end of the basal plane. This
places the nitrogen and iron in locations where active sites can form.
This produced catalysts with good ORR activity, though not as ac-
tive as some other catalysts reported in the literature. Furthermore,
catalysts derived from the physisorption of molecules that already
contain pyridinic functionalities have proven to be highly active
Thus, we have focused on chemically attach-
ing molecules that already contain pyridinic nitrogen functionalities.
Here, we have chemically attached 5,6-diamino-1,10-phenanthroline
aphen͒ to the surface of a carbon black. This was accomplished via
two different methods. The first method employed a diazonium cou-
pling reaction, which has been commonly used to graft functional
One of the most ambiguous factors around NPMCs is that the
exact chemical nature of the active site is still debated. Active cata-
lysts are typically formed by high temperature ͑600–1000°C͒ py-
rolysis in an ammonia atmosphere of a carbon with adsorbed iron on
4
,10-13
the surface. They can also be formed by sputtering.
While the
1
7
exact structure of the active site responsible for the ORR activity is
still debated, the need for surface nitrogen functionalities is univer-
sally accepted. However, there are numerous types of N-containing
species that can be present, including pyrrolic, pyridinic, and
graphene geometries as shown in Fig. 1. The activity of each type of
group is also debated, though catalysts rich in pyridinic nitrogen
1
5,18
toward the ORR.
14
͑
have been linked to highly active ORR catalysts. One of the most
widely accepted models comes from Dodelet and co-workers, who
had postulated that the active site consists of Fe bridging two gra-
1
9,20
1
5
groups onto carbon surfaces.
The second method relies on a
phitic crystallites that contain a pyridinic nitrogen. In all likeli-
hood, most NPMC contain a mixture of active site structures, each
with varying activity. This is not surprising when one considers the
lack of chemical control in their synthesis. Thus, it is quite challeng-
ing to increase the active site density and achieve a greater activity.
The electrochemical identification and quantification of active
sites is also challenging for NPMCs. The quantification of Pt-based
catalytic sites can be performed electrochemically by cyclic voltam-
metry ͑CV͒, where the charge due to H-adsorption ͑determined in
2
1
benzimidazole coupling reaction. These coupling methods allow
for the directed formation of chemical linkages on the edges of
graphitic basal planes, believed to be a preferential location for ac-
tive site formation. Thus, these modified surfaces closely represent
the type of active sites believed to be responsible for catalytic ac-
tivity. The oxygen free voltammetry as well as the ORR activity of
these catalysts is discussed and related to the structure of the active
site.
the absence of O ͒ can be related to the electrochemical surface area
2
Experimental
2
͑
ECSA͒ of Pt ͑210 ␮C = 1 cm of Pt surface͒. This continues to be
a valuable tool for the evaluation and comparison of Pt-based cata-
lysts. Unfortunately, such a tool currently does not exist for NPMCs,
Synthesis of 5,6-diamino-1,10-phenanthroline.— 5,6-diamino-
1,10-phenanthroline ͑aphen͒ was synthesized following a combina-
2
2
23
as they do not always have such obvious features in their O -free
tion of the procedures reported by MacDonnell and Mattay. In
this strategy, the diamine is synthesized from 1,10-phenanthroline-
5,6-dioxime by a palladium-catalyzed reduction using hydrazine hy-
drate as a hydrogen source. Isolation of the dioxime from 1,10-
phenanthroline-5,6-dione required careful control of the reaction pH
with barium carbonate to prevent the formation of unwanted by-
products. Washing of the dioxime was carried out by using a copious
quantity of water, followed by air drying, and then diethyl ether. The
2
CVs. In fact, in many cases, the voltammetry is featureless, with no
discernible waves, making the quantification of catalyst sites impos-
sible by CV. In certain cases, however, a redox wave can clearly be
observed at high potentials ͑ Ͼ 0.6 V͒. For example, Easton et al.
*
Electrochemical Society Active Member.
E-mail: brad.easton@uoit.ca
z
1
13
purity of the compound was confirmed with H and C NMR.
Downloaded on 2015-03-05 to IP 132.203.227.61 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

B332
Journal of The Electrochemical Society, 158 ͑3͒ B331-B336 ͑2011͒
graphene
N
N
nitrile
N
C
N
N
pyrrolic
pyridinic
Figure 3. ͑Color online͒ Schematic representation of the benzimidazole cou-
pling reaction that grafts 5,6-diamino-1,10-phenanthroline onto the surface
of a carbon black to give the product referred to as BP-B.
Figure 1. Depiction of common nitrogen containing species that can be
found of the surface of carbon.
Catalyst synthesis.— Catalysts were synthesized by chemically
modifying a carbon black through either diazonium or benzimida-
zole coupling methods. Details of each method are given below. The
carbon black used in this study was Black Pearls 2000 ͑Cabot͒ and
is hereafter referred to as BP.
of 10 mM iron͑III͒ chloride for 48 h. After filtration and drying at
1
00°C, both samples were ground into a fine powder. Heat-
treatment was performed through pyrolysis of 30 mg of each sample
under either pure flowing nitrogen or a 50/50 mixture of ammonia/
nitrogen for 2 h at 700°C in a tube furnace ͑Barnstead/Thermolyne
21100͒. The final catalysts based upon BP-D and BP-B, which were
heat treated in the absence of ammonia, are referred to BP-D-Fe700
and BP-B-Fe700, respectively. When heat treated in the presence of
ammonia, these catalysts are referred to as BP-D-Fe700A and BP-
B-Fe700A, respectively.
Diazonium coupling.— Aphen was attached using a diazonium
2
4
coupling method similar to that reported by Belmont et al. A sche-
matic diagram of this reaction is shown in Fig. 2. 200 mg of BP was
combined with 25 mg of aphen ͑1 equiv͒ along with 4 equiv of
concentrated sulfuric acid and 2 equiv of sodium nitrite. This con-
verts the amino groups of aphen into diazonium moieties, which
subsequently react with the carbon surface. The mixture was stirred
at room temperature for 30 min, after which the modified carbon
was collected through suction filtration, washed with water, dried
overnight at 100°C, and ground into a fine powder. The product is
hereafter referred to as BP-D.
Electrochemical measurements.— Electrochemical characteriza-
tion was performed through the preparation of electrode inks. Ap-
proximately 10 mg of sample was dispersed in 200 ␮L of 5%
Nafion and 300 ␮L of 50/50 isopropyl alcohol/water to a total vol-
ume of 500 ␮L. After 30 min of sonication, 4 ␮L of ink was depos-
ited on the surface of a glassy carbon disk ͑5 mm diameter͒ of a
rotating ring-disk electrode ͑RRDE͒. This deposition procedure
Benzimidazole coupling.— Aphen was attached using a benzimi-
2
dazole coupling method similar to that reported by Smith and
yielded a catalyst loading of 0.40 mg/cm on the electrode. A plati-
2
1
Pickup. A schematic diagram of this reaction is shown in Fig. 3.
num ring ͑7.5 and 8.5 mm inside and outside diameters, respec-
tively͒ surrounded the glassy carbon disk. The ink was allowed to
air dry for 10 min and placed in a one compartment cell containing
0.5 M H SO ͑aq͒ electrolyte at room temperature. A standard
25 mg of aphen and 100 mg of BP were combined in 25 mL of 4M
HCl and subsequently refluxed for 24 h. The sample was collected
via suction filtration, washed with water, acetone, and methanol,
dried overnight at 100°C, and ground to a fine powder. The product
is hereafter referred to as BP-B.
2
4
calomel electrode was used in conjunction with a platinum wire
counter electrode. Measured potentials were corrected to the revers-
ible hydrogen electrode ͑RHE͒ potential.
Iron coordination and heat-treatment.— Iron was introduced
into both the BP-D and BP-B samples by soaking them in a solution
All electrochemical measurements were made using Pine Instru-
ments AFCBP1 bipotentiostat controlled using AFTERMATH software.
Data were obtained at sweep rates of 100 and 10 mV/s. First, a
steady state CV was obtained under nitrogen at 100 mV/s, followed
by a subsequent scan at 10 mV/s. Afterward, the system was purged
with oxygen, and the scans were repeated at 100 and 10 mV/s.
Furthermore, the electrode rotation rate was varied, changing the
diffusion rate of oxygen to the electrode surface and yielding a
corresponding change in current. The platinum ring was held at a
potential of 1.2 V to oxidize any hydrogen peroxide the samples
produced and thus obtain %H O through the following relationship
2
2
2
00IR
%
H O =
͓1͔
2
2
NID + IR
The term IR is the current obtained from the ring, ID is the current
obtained from the disk, and N is the collection efficiency, which was
determined as 0.1933 for our system through ferricyanide calibra-
tion.
X-ray photoelectron spectroscopy (XPS) measurements.— All
XPS measurements were performed on a Thermo Instruments 310-F
Microlab equipped with a monochromatic Mg K␣ X-ray source
͑h␯ = 1253.6 eV͒. The N1s peaks were fitted, using a mixed
Figure 2. ͑Color online͒ Schematic representation of the diazonium coupling
reaction that grafts 5,6-diamino-1,10-phenanthroline onto the surface of a
carbon black to give the product referred to as BP-D.
Downloaded on 2015-03-05 to IP 132.203.227.61 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 158 ͑3͒ B331-B336 ͑2011͒
B333
Figure 6. ͑Color online͒ Proposed structure of aphen attached to the carbon
surface from a polyaniline-type linkage.
1
7
quinalone ͑pyridinic͒ species. Based on this evidence, we believe
that the process at 0.65 V corresponds to the redox reaction shown
in Fig. 5.
Figure 4. ͑Color online͒ Steady-state CVs obtained for BP, BP-D, and BP-
D-Fe ͑not heat treated͒. Measurements were made at a sweep rate of
The low potential peak at ca. 0.05 V has also been observed for
17
other NPMC catalysts. Furthermore, Smith and Pickup have ob-
1
00 mV/s in N -saturated 0.5 M H SO at room temperature.
2 2 4
served an unexpected low potential peak for anthraquinone ͑AQ͒
2
1
modified carbon surfaces prepared in a similar manner. They have
proposed that this originates from a secondary amine linkage be-
tween the AQ and the carbon surface. Thus, we attribute this wave
to the presence of aphen bound to the carbon surface by secondary
amine “polyaniline-like” linkages, as shown in Fig. 6. This is sup-
ported by the recent work by Birss and Gharaibeh who have pre-
pared catalysts using phenlyene diamine linked to a carbon surface
via polyaniline-type linkages. Figure 7 shows that upon heat-
treatment, the peak at 0.05 V disappears. The polyaniline-type ni-
trogen responsible for this peak either evolved off during pyrolysis
or incorporated itself into the carbon support. Thus, polyaniline-type
nitrogen cannot be responsible for the ORR activity of heat treated
catalysts.
Gaussian–Lorentzian shaped profile with a Shirley background.
Identical parameters were used for all the observed N1s peaks. The
peak broadening was constrained to a maximum of 2 eV.
2
9
Results
Diazonium coupling—electrochemical and XPS character-
ization.— The first step in characterization of the modified carbon
͑
that was not heat treated͒ was to obtain a steady-state CVs under
nitrogen. Figure 4 shows that upon diazonium coupling of aphen,
two peaks appear at ca. 0.05 and 0.650 V versus RHE. The stability
of the observed peaks was confirmed with 3 h of cycling between 0
and 1 V with no observable change in either peak. While the peaks
confirm that aphen is present on the BP surface, their location and
size indicate the type and quantity of aphen functionality. A high
potential redox wave has been previously observed for other
The charge originating from the peak at 0.650 V has been mea-
sured for each catalyst at different stages of development and are
complied in Table I. The size of the wave diminished significantly
upon heat-treatment. However, the peak was substantially larger ͑ca.
4
,14,16,25
NPMCs,
though the absence of this peak does not negate
ORR activity. Because we have not added Fe to our samples, the
peak must originate from the attached aphen moiety. We attribute
the wave at 0.65 V versus RHE to the redox behavior of the phenan-
throline ring and is therefore indicative of a pyridinic type nitrogen
on the BP surface after coupling. This is supported by work of
Cobos-Murcia et al. who reported similar electrochemical signatures
2
6
from the electropolymerization of 5-aminophenanthroline. Similar
waves have been observed for N-enriched carbons for supercapaci-
27,28
tor applications,
which were attributed to the redox behavior of
pyridinic-type species on the carbon surface. In addition, Jaouen et
al. have shown that this species cannot originate from an Fe redox
couple, and they proposed this wave may originate from surface
Carbon surface
Carbon surface
Figure 7. ͑Color online͒ Steady-state CVs obtained for BP700, BP-D700,
and BP-D-Fe700. Measurements were made at a sweep rate of 100 mV/s in
N -saturated 0.5 M H SO at room temperature.
Figure 5. The proposed redox behavior of pyridinic functionalities on the
carbon surface occurring at 0.65 V vs SHE.
2
2
4
Downloaded on 2015-03-05 to IP 132.203.227.61 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

B334
Journal of The Electrochemical Society, 158 ͑3͒ B331-B336 ͑2011͒
Table I. Variation in the onset potential for ORR and the charge
under wave at 0.65 V for catalysts of different compositions.
Charge Under Wave at
0.65 V ͑mC͒
Sample
VORR ͑V vs RHE͒
BP
BP700
BP-D
BP-D-Fe
BP-D700
BP-D-Fe700
BP-B
BP-B-Fe
BP-B700
BP-B-Fe700
0.38
0.34
0.44
0.46
0.64
0.76
0.42
0.40
0.58
0.76
0
0
1.106
1.049
0.026
0.278
3.130
8.242
0.035
0.140
1
0 times͒ for catalysts heat treated in the presence of Fe than those
heat treated without, indicating that Fe acts to help retain nitrogen in
a pyridinic state. The presence of this peak after heat-treatment in-
dicates that more pyridinic nitrogen is present and is likely respon-
sible for a large portion of the activity toward the ORR.
Figure 8 shows the ORR linear sweep voltammagram for cata-
lysts derived from diazonium coupling at various stages of develop-
ment. A key parameter is the onset potential ͑VORR͒, which is de-
fined as the potential at which the ORR current is 5% of the
Figure 9. ͑Color online͒ %H O as a function of disk potential for BP-D-
2
2
Fe700 and BP-D-Fe700A. Measurements were made at a sweep rate of
10 mV/s in O2-saturated 0.5 M H2SO4 at room temperature with a rotation
rate of 900 rpm.
30
diffusion-limited current, and is complied in Table I. Figure 8a
details the change in ORR activity on BP without heat-treatment. BP
on its own has a low VORR of ca. 0.37 V. Initial modification via
diazonium coupling even reduces the limiting current. However,
upon addition of Fe, the VORR increases to ca. 0.46 V versus RHE,
which is in addition to increasing the limiting current. The increase
in activity is likely due to Fe populating some of the active sites.
Figure 8b shows the ORR activity after the diazonium-derived cata-
lysts were heat treated. Heat-treatment significantly increased the
activity of BP-D with an onset potential of ca. 0.64 V, which is on
ous active sites as any oxygen that is incompletely reduced to hy-
drogen peroxide cannot diffuse far before it is further reduced by a
nearby active site.
3
2
Figure 10 compares the N1s XPS spectra obtained for BP-D-Fe-
00 and BP-D-700A. All spectra were fit well with three peaks that
7
are known to originate from pyridinic ͑397–399.5 eV͒, pyrrolic
17
͑
400.2–400.9 eV͒, and graphene ͑401–403 eV͒ type nitrogen. The
relative concentrations for each of the nitrogen species are listed in
Table II. It is clear that not all of the nitrogen has remained in a
pyridinic state after heat-treatment, with much of the nitrogen form-
ing pyrrolic or graphene type functionalities. The total nitrogen in
the samples was ca. 0.6 atom %. However, it is clear that the am-
monia treatment has increased the amount of pyridinic nitrogen
4
par with other results in the literature for metal-free ORR catalysts.
However, as Black Pearls contains trace amounts of iron
1
6
͑
0.02 wt %͒, we cannot completely rule out the impact of Fe,
though we would expect this to be minimal. However, the amounts
of Fe in these catalysts are certainly well below the point of Fe
31
saturation of active sites.
Heat-treatment in the presence of additional Fe dramatically in-
creased the onset potential to 0.76 V, reaching 0.80 V upon the
introduction of ammonia. This high activity is further supported by
the low peroxide production shown in Fig. 9. Only a small propor-
tion of the current is due to an incomplete reduction of oxygen to
H O ͑ca. 2.5% at 0.3 V͒. The low %H O is indicative of numer-
2
2
2
2
Figure 10. ͑Color online͒ XPS N1s narrow scan spectra obtained for ͑a͒
BP-D-Fe700 and ͑b͒ BP-D-Fe700A.
(
a)
(b)
Table II. The relative concentration of nitrogen species present as
determined by XPS.
Figure 8. ͑Color online͒ Comparison of the disk current densities obtained
for BP-D-based catalysts at various stages of development: ͑a͒ prior to heat-
treatment and ͑b͒ after pyrolysis at 700°C. Measurements were made at a
sweep rate of 10 mV/s in O -saturated 0.5 M H SO at room temperature
Sample
Pyridinic ͑%͒
Pyrrolic ͑%͒
Graphene ͑%͒
BP-D-Fe700
BP-D-Fe700A
BP-B-Fe700A
39.91
43.52
49.71
35.15
30.49
31.38
24.94
25.97
18.9
2
2
4
with a rotation rate of 900 rpm.
Downloaded on 2015-03-05 to IP 132.203.227.61 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 158 ͑3͒ B331-B336 ͑2011͒
B335
Figure 11. ͑Color online͒ Steady-state CVs obtained for BP, BP-B, and
BP-B-Fe ͑not heat treated͒. Measurements were made at a sweep rate of
Figure 12. ͑Color online͒ Steady-state CVs obtained for BP700, BP-B700,
and BP-B-Fe700. Measurements were made at a sweep rate of 100 mV/s in
N -saturated 0.5 M H SO at room temperature.
1
00 mV/s in N -saturated 0.5 M H SO at room temperature.
2 2 4
2
2
4
available over heat-treatment under inert atmosphere. The use of
ammonia assists in the maintenance of nitrogen content and the
preferential formation of pyridinic nitrogen. An increase in the pro-
portion of pyridinic nitrogen explains the greater ORR activity ob-
served for the ammonia treated catalyst.
Heat-treatment does lead to a significant increase in the activity,
with a VORR of ca. 0.58 V achieved for BP-B-700, which is slightly
lower than that achieved with BP-D-700. Heat-treatment in the pres-
ence of Fe lead to the expected increase in activity ͑VORR = 0.76 for
BP-B-Fe700͒. Pyrolysis under ammonia further increased the VORR
to ca. 0.82 V. The difference in activity after heat-treatment under
ammonia is more pronounced in Fig. 14 where peroxide production
was reduced significantly when ammonia was introduced.
Benzimidazole coupling—electrochemical and XPS characteriza-
tion.— Figure 11 shows the oxygen-free voltammetry for the non-
heat treated BP-B derived catalysts. These materials show the same
polyaniline-type and pyridinic-type waves that were observed for
those made by diazonium coupling. Comparison of Fig. 4 and 11
reveals that low potential polyaniline-type peak is more intense for
BP-B than for BP-D. This can partially be explained by the higher
loading of aphen attached in BP-B. However, it may also be due the
fact that the reaction conditions used to form BP-B are similar to
those used to form polyaniline. Furthermore, we have observed that
this peak disappears at low aphen loadings and becomes more
Figure 15 shows the N1s XPS spectra obtained for BP-B-
Fe700A. Much like BP-D-Fe700A, this catalyst has a pyridinic ni-
trogen content of 50 atom %. This high abundance of pyridinic
nitrogen is responsible for its high ORR activity and low peroxide
6
,34
production.
As with the BP-B-based samples, the remainder of
the nitrogen present was in a pyrrolic or graphene state ͑see Table II͒
with a total nitrogen content of ca. 0.5 atom %.
Conclusions
33
prominent at higher aphen loadings. Thus, reaction conditions
must be carefully controlled because a mixture of benzimidazole
and polyaniline type linkages can form. In contrast to diazonium
coupling, addition of Fe leads to an increase in the size of both
peaks. Cycling between 0 and 1 V versus RHE for 3 h again yielded
no change in peak shape for any of the samples. Figure 12 demon-
strates that as with diazonium coupling, heat-treatment eliminates
the polyaniline peak and reduces, but does not eliminate the pyri-
dinic peak, and that the larger pyridinic peak is associated with the
sample that contained Fe. It is interesting to note that the charge
under this peak is smaller for BP-B-Fe700 than for BP-D-Fe-700,
despite its substantially a higher initial charge. This may be related
to the thermal stability of the polyaniline/benzimidazole linkages,
though further studies are needed to confirm this.
Regardless of the method of attachment, Fe–N/C catalysts de-
rived from aphen displayed good activity toward ORR, with a quite
low %H O ͑ca. 2%͒. After heat-treatment, these catalysts displayed
2
2
Figure 13 shows the ORR linear sweep voltammograms for cata-
lysts derived from benzimidazole coupling at various stages of de-
velopment. Initial modification via benzimidazole coupling leads to
a slight increase in VORR ͑ca. 0.425 V͒, which was not observed for
the BP-D. One possible reason for this is the higher loading of
aphen on BP-B or the greater number of nitrogen-containing func-
tionality in the BP-B catalysts. The addition of Fe does not lead to a
significant increase in VORR but does leads to a modest increase
limiting current, indicating the beneficial effect of Fe on catalytic
activity.
(a)
(b)
Figure 13. ͑Color online͒ Comparison of the disk current densities obtained
for BP-B-based catalysts at various stages of development: ͑a͒ prior to heat-
treatment and ͑b͒ after pyrolysis at 700°C. Measurements were made at a
sweep rate of 10 mV/s in O -saturated 0.5 M H SO at room temperature
2
2
4
with a rotation rate of 900 rpm.
Downloaded on 2015-03-05 to IP 132.203.227.61 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

B336
Journal of The Electrochemical Society, 158 ͑3͒ B331-B336 ͑2011͒
vided the control necessary to relate the surface functionalities on
the carbon surface to the redox waves present in the oxygen-free
voltammetry. The clear understanding that the high potential redox
peak originates from the proposed active site is the first step toward
the development of an electrochemical tool to quantify the number
of active sites.
Acknowledgments
This work was supported by the Natural Sciences and Engineer-
ing Research Council ͑NSERC͒ of Canada and UOIT. We also thank
Ruiyao Wang ͑Queens University͒ for assistance with XPS measure-
ments.
University of Ontario Institute of Technology assisted in meeting the
publication costs of this article.
References
1
. S. Zhang, X. Z. Yuan, J. N. C. Hin, H. Wang, K. A. Friedrich, and M. Schulze, J.
Power Sources, 194, 588 ͑2009͒.
2
3
4
. M. B. Ji and Z. D. Wei, Energy, 2, 1057 ͑2009͒.
. H. Tsuchiya and O. Kobayashi, Int. J. Hydrogen Energy, 29, 985 ͑2004͒.
. E. B. Easton, A. Bonakdarpour, R. Yang, D. A. Stevens, and J. R. Dahn, J. Elec-
trochem. Soc., 155, B547 ͑2008͒.
5
6
. R. A. Sidik, A. B. Anderson, N. P. Subramanian, S. P. Kumaraguru, and B. N.
Popov, J. Phys. Chem. B, 110, 1787 ͑2006͒.
. C. Medard, M. Lefevre, J. P. Dodelet, F. Jaouen, and G. Lindbergh, Electrochim.
Acta, 51, 3202 ͑2006͒.
Figure 14. ͑Color online͒ %H O as a function of disk potential for BP-B-
2
2
Fe700 and BP-B-Fe700A. Measurements were made at a sweep rate of
0 mV/s in O -saturated 0.5M H SO at room temperature with a rotation
7. C. W. B. Bezerra, L. Zhang, K. Lee, H. Liu, A. L. B. Marques, E. P. Marques, H.
Wang, and J. Zhang, Electrochim. Acta, 53, 4937 ͑2008͒.
1
2
2
4
8
9
. G. Wu, Z. Chen, K. Artyushkova, F. H. Garzon, and P. Zelenay, ECS Trans., 16͑2͒,
59 ͑2008͒.
. H. A. Gasteiger, S. S. Kocha, B. Sompalli, and F. T. Wagner, Appl. Catal., B, 56,
͑2005͒.
rate of 900 rpm.
1
9
1
0. E. B. Easton, A. Bonakdarpour, and J. R. Dahn, Electrochem. Solid-State Lett., 9,
A463 ͑2006͒.
ORR activity that not only surpassed that achieved with our previ-
16
ous work on surface attached with aminosilanes but compares well
with the performance of some of the most active NPMCs reported to
11. R. Yang, A. Bonakdarpour, E. B. Easton, and J. R. Dahn, J. Electrochem. Soc.,
1
54, A275 ͑2007͒.
7,17
this date in the literature.
However, our methodology has pro-
12. E. B. Easton, R. Yang, A. Bonakdarpour, and J. R. Dahn, Electrochem. Solid-State
Lett., 10, B6 ͑2007͒.
1
1
3. G. C.-K. Liu and J. R. Dahn, Appl. Catal., A, 347, 43 ͑2008͒.
4. J. P. Dodelet, in N4-Macrocyclic Metal Complexes, Chap.3, J. Zagal, F. Bedioui,
and J. P. Dodelet, Editors, Springer, New York ͑2006͒.
1
1
5. M. Lefevre, E. Proietti, F. Jaouen, and J. P. Dodelet, Science, 324, 71 ͑2009͒.
6. A. D. Pauric, A. W. Pedersen, T. Andrusiak, and E. B. Easton, J. Electrochem. Soc.,
1
57, B370 ͑2010͒.
1
1
7. F. Jaouen, J. Herranz, M. Lefevre, J. P. Dodelet, U. I. Kramm, I. Herrmann, P.
Bogdanoff, J. Maruyama, T. Nagaoka, A. Garsuch et al., ACS Appl. Mater. Inter-
faces, 1, 1623 ͑2009͒.
8. M. Bron, J. Radnik, M. Fieber-Erdmann, P. Bogdanoff, and S. Fiechter, J. Elec-
troanal. Chem., 535, 113 ͑2002͒.
1
2
2
2
2
2
2
9. S. Baranton and D. Belanger, Electrochim. Acta, 53, 6961 ͑2008͒.
0. C. A. Dyke and J. M. Tour, J. Am. Chem. Soc., 125, 1156 ͑2003͒.
1. R. D. L. Smith and P. G. Pickup, Electrochem. Commun., 11, 10 ͑2009͒.
2. S. Bodige and F. M. MacDonnell, Tetrahedron Lett., 38, 8159 ͑1997͒.
3. A. Kleineweischede and J. Mattay, Eur. J. Org. Chem., 2006, 947.
4. J. A. Belmont, R. M. Amici, and C. P. Galloway, U.S. Pat. 7,294,185 ͑2007͒.
5. A. H. C. Sirk, S. A. Campbell, and V. I. Birss, J. Electrochem. Soc., 155, B592
͑2008͒.
2
6. J. A. Cobos-Murcia, L. Galicia, A. Rojas-Hernandez, M. T. Ramirez-Silva, R.
Alvarez-Bustamante, M. Romero-Romo, G. Rosquete-Pina, and M. Palomar-
Pardave, Polymer, 46, 9053 ͑2005͒.
2
7. D. Hulicova-Jurcakova, M. Kodama, S. Shiraishi, H. Hatori, Z. H. Zhu, and G. Q.
Lu, Adv. Funct. Mater., 19, 1800 ͑2009͒.
2
2
8. S. Shrestha and W. E. Mustain, J. Electrochem. Soc., 157, B1665 ͑2010͒.
9. S. A. Gharaibeh and V. I. Birss, Abstract 510, The Electrochemical Society Meet-
ing Abstracts, Vol. 1001, Vancouver, Canada, April 25–30, 2010.
30. L. Birry, J. H. Zagal, and J. P. Dodelet, Electrochem. Commun., 12, 628 ͑2010͒.
31. F. Jaouen and J. P. Dodelet, Electrochim. Acta, 52, 5975 ͑2007͒.
32. A. Bonakdarpour, M. Lefevre, R. Z. Yang, F. Jaouen, T. Dahn, J. P. Dodelet, and J.
R. Dahn, Electrochem. Solid-State Lett., 11, B105 ͑2008͒.
3
3. E. B. Easton, A. D. Pauric, B. J. MacLean, Abstract 936, The Electrochemical
Society Meeting Abstracts, Vol. 1002, Las Vegas, NV, November 10–15, 2010
Figure 15. ͑Color online͒ XPS N1s narrow scan spectra obtained for BP-B-
Fe700A.
34. M. Lefevre and J. P. Dodelet, Electrochim. Acta, 48, 2749 ͑2003͒.
Downloaded on 2015-03-05 to IP 132.203.227.61 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).