Efficient
metal-free
N-doped
mesoporous
carbon
catalysts for ORR by a template-free approach
Guillermo A. Ferrero,a Antonio B. Fuertes,a Marta Sevilla,a* Maria-Magdalena
Titirici *
a
Instituto Nacional del Carbón (CSIC), P.O. Box 73, Oviedo 33080, Spain
b School
of Engineering and Materials Science, Queen Mary University of London
Mile End Road, E1 4NS, London, UK
*
Corresponding
Author:
[email protected]
(M.
Sevilla),
[email protected] (M-M. Titirici)
Abstract
A simple, cost-effective and scalable approach for the synthesis of N-doped
mesoporous carbons that are effective as ORR catalysts is reported. The
synthesis procedure involves two steps: a) production of mesoporous carbons
using a template-free approach based on the carbonization of citrate salts of zinc
and calcium, and b) N-doping by heat treatment in the presence of melamine.
The resulting N-doped carbon possess a high specific surface area, a porosity
made up exclusively of mesopores and a large amount of nitrogen functionalities
(~8-9 wt %). When used as an electrocatalyst for the oxygen reduction reaction
(ORR), the metal-free carbon materials predominantly catalyze the 4 e- process,
with an onset potential of 0.9 V (vs. RHE) and a superior kinetic current density
(~17 mA cm-2) to that of Pt/C at ~0.6 V under basic conditions. In addition, the
1
developed catalysts show a higher stability than commercial Pt/C and excellent
electrocatalytic selectivity against methanol crossover.
Keywords: Carbon, Nitrogen-doping, mesoporosity, oxygen reduction reaction,
catalyst.
2
1. Introduction
Metal-free
carbon
materials
have
gained
increasing
importance
in
electrocatalysis for the cathodic oxygen reduction reaction (ORR) in fuel cells.
This is due to their high efficiency to catalyze the reaction in alkaline medium
along with a better stability, lower cost and broader availability compared to the
widespread used platinum [1-3]. It is well-known that the incorporation into the
carbon framework of certain types of heteroatoms gives rise to a notable
enhancement in the ORR performance in relation to un-doped carbons, which
exhibit extremely low currents and the undesired two-electron pathway towards
harmful H2O2 [2-5]. Although it has been proved that a variety of heteroatoms
such as phosphorus [6], boron [7], and sulfur [8] can, to a certain extent, catalyze
the ORR reaction, carbon materials doped with nitrogen functional groups have
emerged as the best metal-free catalyst for the ORR process [4, 9-11]. Indeed,
N-doped carbons exhibit an excellent long-term durability and an improved ORR
activity as a result of their high electronic conductivity and the higher
electronegativity of N atoms which creates positive charged sites that favor the
adsorption of oxygen molecules and their subsequent reduction [3, 9, 12, 13].
However, it is important to note that the ORR activity of N-doped carbon materials
not only depends on the presence of abundant N-groups, but also on the easy
access of the reactants to these catalytic sites [14-18]. In this respect, carbon
materials with a porosity made up of accessible mesopores are preferred
because the transport of reactants and products towards/from the active sites is
faster than in microporous carbons [12, 19]. A common approach for the
preparation of N-doped mesoporous carbon materials with well-controlled
properties is based on nanocasting techniques by using either a hard or soft
3
template [20-25]. However, these synthetic methodologies are complex,
expensive and time-consuming as they require the fabrication of sacrificial
templates (inorganic templates or structure-directing agents) besides the use of
hazardous substances for template removal [20-27], which limits their
competitiveness against Pt-based catalysts. Therefore, for ensuring the
widespread commercialization of fuel cells, the preparation of N-doped
mesoporous carbons with controlled chemical and structural properties by a
facile, cost-effective and sustainable method is an important challenge to be
overcome. Accordingly, here we present a simple strategy for the fabrication of
N-doped mesoporous carbons with good ORR activity. The procedure comprises
two steps: a) the production of a mesoporous carbon using direct carbonization
of citrate salts of calcium or zinc, and b) the incorporation of nitrogen functional
groups by thermal treatment in the presence of melamine. The resulting N-doped
carbons possess a large specific surface area in the 1190-1350 m2 g-1 range, a
porosity made up almost exclusively of mesopores with size of around 11 nm and
a large number of nitrogen functionalities (~ 8-9 wt %). The synthesized materials
were tested as ORR catalysts in basic media and the results obtained show that
they exhibit an excellent electrocatalytic performance and a better catalytic
durability than commercial platinum-based catalysts.
2.
Experimental Section
2.1.
Synthesis of Materials
In a typical synthesis procedure, 3 g of citrate salt (i.e. (C6H5O7)2Zn3 · 2H2O, or
(C6H5O7)2Ca3 · 4H2O) purchased from Aldrich were heat-treated in a stainless
steel reactor under nitrogen up to 800 ºC at a heating rate of 3 ºC min -1 and held
at this temperature for 1 hour. The resulting black solid was then washed with
4
diluted HCl (10 %). Finally, the carbon particles were collected by filtration,
washed with abundant distilled water and dried at 120 ºC for several hours.
Afterwards, the mesoporous carbon was mixed with melamine, using a ratio
melamine/carbon ratio of 4. Finally, the mixture was heat treated under nitrogen
up to 800 ºC (heating rate of 3 ºC min-1) for 1 hour. The as-obtained N-doped
mesoporous samples are denoted as CX, where X indicate the type of citrate salt
(X= Zn or Ca).
2.2.
Physical characterization
Scanning electron microscopy (SEM) images were obtained on a Quanta
FEG650 (FEI) instrument. The nitrogen sorption isotherms of the carbon samples
were measured at -196 °C using a Micromeritics ASAP 2020 sorptometer. The
apparent surface area was calculated using the BET method. An appropriate
relative pressure range was selected to ensure that a positive line intersect of
multipoint BET fitting (C > 0) would be obtained and Vads(1 − p/po) would increase
with p/po [28]. The total pore volume (Vp) was determined from the amount of
nitrogen adsorbed at a relative pressure (p/po) of 0.95. The micropore volume
(Vmicro) was estimated using the αs-plot method. The reference adsorption data
used for the αs analysis of the carbon samples correspond to a non-graphitized
carbon black sample [29]. The primary mesopore volume (Vmeso) was determined
by difference between the pore volume (Vp) and the micropore volume (Vmi). The
pore size distribution (PSD) was calculated by means of the Kruk-Jaroniec-Sayari
method applied to the adsorption branch [30]. X-ray diffraction (XRD) patterns
were obtained on a Siemens D5000 instrument operating at 40 kV and 20 mA,
using CuKα radiation. X-ray photoelectron spectroscopy (XPS) was performed
on a Specs spectrometer, using Mg Kα (1253.6 eV) radiation from a double anode
5
at 150 W. Elemental analysis (C, N and O) of the samples was carried out on a
LECO CHN-932 microanalyzer.
2.3.
Electrochemical measurements
Electrochemical measurements were conducted using an AUTOLAB PGSTAT
101 and a Multi AUTOLAB M101 (CH Instruments). The catalyst inks were
prepared by ultrasonically dispersing 1.5 mg of CX sample in a solution
containing 100 μL Nafion (5 wt %) solution and 900 μL deionized water. For
comparison, the Pt/C catalyst (20 wt % Pt on graphitized carbon, Sigma-Aldrich)
ink was prepared in the same way, using the same amount of catalyst (i.e. 1.5
mg). The above prepared catalyst inks were deposited onto a polished glassy
carbon electrode (α-Al2O3 slurry, 50 nm) and dried under room temperature. A
conventional three-electrode cell was employed incorporating Ag/AgCl (3 M KCl)
as the reference electrode, a Pt wire as the counter electrode and the catalyst
film coated rotating ring-disk electrode (RRDE) or rotating disk electrode (RDE)
as the working electrode. The electrolyte was 0.5 M H2SO4 solution or 0.1 M KOH
solution. All experiments were carried out at 20 ºC. Before testing, an O 2/N2 flow
was bubbled through the electrolyte in the cell for 30 min to saturate it with O2/N2.
The measured potentials vs. Ag/AgCl (3 M KCl) were converted to the reversible
hydrogen electrode (RHE) scale according to the Nernst equation:
ERHE = EAg/AgCl + 0.059 pH + EoAg/AgCl
(1)
where EAg/AgCl is the experimentally measured potential vs. Ag/AgCl reference
and EoAg/AgCl = 0.21 V at 20 ºC. The values of potential provided along the text
are referenced against RHE unless otherwise stated.
Cyclic voltammetry (CV) was performed from 0 to 1.2 V vs. RHE in 0.1 M KOH
and 0.5 M H2SO4, with a sweep rate of 50 mV s-1.
6
Rotating disk electrode (RDE) linear sweep voltammetry (LSV) measurements
were conducted from 1.2 to 0 V vs. RHE in 0.1 M KOH and 0.5 M H2SO4 at a
scan rate of 10 mV s-1 under disk rotation rates of 400, 800, 1200, 1600, 2000
and 2400 rpm. A working electrode of 3.0 mm diameter GC rotating disk electrode
was used.
The apparent number of electrons transferred for ORR on N-CC catalyst was
determined by the Koutecky-Levich equation given by:
1 1 1
1
1
= + =
+
⁄
1
2
J JL JK B ω
JK
1
B = 0.62 n F C0 (D0 )2⁄3 v 6
where 𝐽 is the measured current density, 𝐽𝐾 is the kinetic current density, 𝐽𝐿 is the
diffusion-limited current density, ω is the electrode rotation rate, F is the Faraday
constant (96485 C mol-1), C0 is the bulk concentration of O2 (1.1 x10-3 mol L-1 for
0.5 M H2SO4 solution and 1.2 x10-3 mol L-1 0.1 M KOH solution), D0 is the diffusion
coefficient of O2 (1.4 x 10-5 cm2 s-1 for 0.5 M H2SO4 solution and 1.9 x 10-5 cm2 s1
for 0.1 M KOH solution) and  is the kinetic viscosity of the electrolyte (0.01 cm2
s-1 for both 0.5 M H2SO4 solution and 0.1 M KOH solution) [31-34]. The onset
potential is defined as the potential at which the current density is -0.1 mA cm-2.
For the rotating ring-disk electrode (RRDE) tests, the disc potential was scanned
at 10 mV s-1, while the ring potential was held at 1.5 V vs. RHE in order to oxidize
any H2O2 produced [35, 36]. The working electrode was a 5 mm GC disk
electrode and a Pt ring electrode (375 µm gap). The H2O2 collection efficiency at
the ring (N=0.249) was provided by the manufacturer. The following equations
were used to calculate n (the apparent number of electrons transferred during
ORR) and % H2O2 (the percentage of H2O2 released during ORR) [34]:
7
𝑛=
4 𝐼𝐷
𝐼𝐷 + (𝐼𝑅 /𝑁)
% 𝐻2 𝑂2 = 100
2 𝐼𝑅 /𝑁
𝐼𝐷 + (𝐼𝑅 /𝑁)
where 𝐼𝐷 is the Faradaic current at the disk, 𝐼𝑅 is the Faradaic current at the ring,
N is the H2O2 collection coefficient at the ring.
3.
Results and discussion
3.1.
Structural characteristics of the N-doped mesoporous carbons
The mesoporous carbons were synthesized by direct heat-treatment of citrate
salts of calcium or zinc at 800ºC followed by acid washing. The mechanism of
formation of these carbons has already been thoroughly discussed by us
elsewhere [37]. Briefly, at temperatures below 320ºC the dehydration of the salts
occurs. When the temperature increases (∼ 350-500 ºC), two simultaneous
processes takes place: the pyrolysis of the organic moiety and the formation of
inorganic species (i.e. CaCO3 or ZnCO3). At higher temperatures, the pyrolysis
of the carbonaceous material is completed and the carbonates decompose into
the corresponding nano-oxides (i.e. CaO or ZnO) which remain embedded within
the carbonaceous framework. The removal of these inorganic nanoparticles -that
act as endotemplates- is accomplished by acid washing, which leads to a carbon
with numerous cavities in the mesopore range. In order to obtain nitrogen-doped
mesoporous carbons, the corresponding mesoporous carbons were mixed with
melamine and heat-treated up to 800 ºC as described in the experimental section.
During this process, at temperatures < 500 ºC, the polymerization of the
melamine takes place giving rise a carbon nitride allotrope (g-C3N4).
Subsequently, at higher temperatures, this compound decomposes releasing
8
numerous nitrogen-containing species (e.g. C2N2+, C3N2+, C3N3+, etc) that react
with the mesoporous carbon, promoting thereby the incorporation of numerous
nitrogen functional groups in the carbon framework [38]. The simplicity of this
methodology lies in the use of commercial products which are readily available,
such as metal citrates. In spite of its simplicity, however, the textural properties
of the resultant carbons (BET surface area, pore volume and narrow pore size
distribution) are comparable to those of hard- and soft-templated mesoporous
carbons, as will be shown.
The morphology of the N-doped mesoporous carbons was investigated by
means of scanning electron microscopy (SEM). As can be seen in Figures 1a-b,
the carbon samples are made up of particles that possess an irregular
morphology with size of around 2 µm. The porosity of the N-doped mesoporous
carbons was investigated by means of nitrogen physisorption analysis and the
results are displayed in Figures 1c-d and Table 1. The N-doped carbon samples
possess a mesoporous structure, as can be inferred from the nitrogen sorption
isotherms and the pore size distributions (PSDs) displayed in Figures 1c-1d. More
specifically, the PSDs reveal that CCa has a porosity made up of large
mesopores centered at 10.8 nm (Figure 1c), while CZn possesses two pore
systems centered at 3.2 nm and 10.5 nm (see Figure 1d). It is envisaged that the
mesoporous structure will allow the reactants to reach the catalytic sites with
minor diffusion limitations [12, 19]. Moreover, these samples exhibit high specific
surface areas (1190 m2 g-1 for CZn and 1350 m2 g-1 for CCa) and large pore
volumes (1.2 cm3 g-1 for CZn and 1.79 cm3 g-1 for CCa). Additionally, these
carbons exhibit an amorphous-like structure with low-degree of graphitization as
9
can be inferred from the XRD patterns in Figure S1 which show two broad bands
at around 2θ=24º and 44º.
Table 1. Physico-chemical properties of the mesoporous carbon.
Sample Code
SBET
(m2 g-1)
Vp
(cm3 g-1) a
Vmicro
(cm3 g-1) b
Vmeso
(cm3 g-1) c
C
H
CZn
1190
1.20
0.10
1.10
82.2
0.42
8.5
8.8
1.4
CCa
1350
1.79
0.06
1.73
79.4
0.44
9.2
10.9
0.9
Chemical Composition (wt %)
N
O
Electrical
Conductivity
(S cm-1)
Pore volume at p/po ~ 0.95. b Micropore volume determined by using the s-plot method
applied to the N2 adsorption branch. c Mesopore volume obtained by the difference
between pore volume (Vp) and micropore volume (Vmicro).
a
10
a)
b)
1 µm
c)
1 µm
d)
CCa
CZn
10.5 nm
CCa
CZn
3.0
3
dV/dlog (D), cm g
1000
800
600
400
200
0
0.0
4.0
3.5
-1
1200
3
Adsorbed volume, (cm STP/g)
1400
2.5
2.0
3.2 nm
1.5
10.8 nm
1.0
0.5
0.2
0.4
0.6
0.8
Relative pressure (p/po)
1.0
0.0
1
10
100
Pore size (D), nm
Figure 1. SEM images of (a) CCa and (b) CZn, and (c) nitrogen sorption
isotherms and (d) pore size distributions of the N-doped mesoporous carbons.
As determined by elemental analysis, the N-doped mesoporous carbons
have high nitrogen contents of 8.5 wt % and 9.2 wt % for the CZn and CCa
samples respectively (see Table 1). Further investigation of the nature of Ngroups present in the carbon samples was carried out by means of XPS analysis
(see Figures 2a-b). In the case of the CZn sample, the deconvoluted XPS N1s
spectra exhibits two main peaks at 398.36 eV and 400.4 eV that can be assigned
to pyridinic-N (57.4 %) and pyrrolic-N (33.8%) respectively, and two minor peaks
11
at 401.8 eV and 403.3 eV, which can be attributed to quaternary-N (3.9 %) and
pyridine-N-oxides (4.8 %). Similarly, two main contributions were found in CCa
sample, which were assigned to pyridinic-N (58.8 %) and pyrrolic-N (28.5%), but
only one minor contribution at 401.28 eV, attributable to quaternary-N (12.7%)
[39, 40]. Additionally, the XPS analysis discards the presence of any impurity or
metal trace in the carbon framework (see Figure S2). Furthermore, both nitrogendoped mesoporous carbon possess good electronic conductivities in the 0.9-1.4
S cm-1 range (Table 1), an important feature for ensuring a fast and smooth
electron transfer pathway [41].
b)
CZn
a)
Pyridinic
B.E.~398.4
CCa
Experimental data
Fitted envelope
Pyridinic
B.E.~398.3
Experimental data
Fitted envelope
Quaternary
B.E.~401.8
N-Oxide
B.E.~403.3
396
398
400
402
Binding Energy (eV)
404
Pyrrolic
B.E.~399.9
Intensity (a.u.)
Intensity (a.u.)
Pyrrolic
B.E.~400.4
Quaternary
B.E.~401.3
396
398
400
402
Binding Energy (eV)
404
c)
12
Figure 2. High-resolution XPS N 1s core level spectra of (a) CCa and (b) CZn,
and (c) scheme of the different nitrogen functionalities identified in the
mesoporous carbons.
3.2.
ORR performance of the N-doped mesoporous carbons
The catalytic activity of the mesoporous N-doped carbon materials was examined
by cyclic voltammetry and rotating disk/ring-disk electrode measurements
(RDE/RRDE) in a three-electrode electrochemical cell using both acid and basic
electrolyte media. For comparison, commercial 20% Pt with the same mass
loading was also tested. In a first step, cyclic voltammetry experiments were
performed under N2 and O2 atmospheres in 0.1 M KOH electrolyte using a scan
rate of 50 mV s-1. As can be seen in Figures 3a-3b, featureless voltammetric
curves are observed in the N2-saturated solution for both materials. However, a
well-defined cathodic peak current at approximately 0.757 V is observed when
the electrolyte is saturated with O2, which is indicative of enhanced
electrocatalytic activity. This cathodic peak is more pronounced in the case of
CCa, which hints at a better electrocatalytic performance in comparison to CZn.
In order to confirm this result, linear sweep voltammograms (LSV) were
measured with the RDE electrode in O2-saturated KOH electrolyte at 1600 rpm.
Figure 4a evidences that both nitrogen-doped mesoporous carbons experience
significant changes in the ORR polarization curves between 1.08 and 0.68 V,
CCa being the material that shows the most positive onset potential, at ~0.9 V,
value which is only 30 mV more negative than that of commercial Pt/C catalyst.
By contrast, the CZn sample offers a relatively worse electrocatalytic behavior,
with an onset potential 20 mV more negative than that of CCa. Furthermore, its
half-wave potential (E1/2) is 20 mV lower than that of CCa and the diffusion limiting
13
current
is 4.1 mA cm-2, smaller than that of CCa (4.9 mA cm-2), although
comparable to that of Pt/C. It is worth noting that the value of E1/2 of CCa (0.75
V) is only 30 mV more negative than that of the Pt/C used as reference, 70 mV
compared to the best Pt/C reported so far [42] and 130 mV compared to the best
metal free N-doped carbon catalyst [43, 44]. In order to further compare the
electrocatalytic performance, the kinetics of the ORR reaction were investigated
under different RDE rotation rates at a scanning rate of 10 mV s-1. As can be seen
in Figure S3, for potentials > 0.83 V, the current densities are independent of the
electrode rotation rate indicating kinetic control, whereas for potentials < 0.58 V,
plateau current densities appear indicating diffusion control. The value of these
limiting current densities increases with rotation rate due to the shortened
diffusion layer [34]. Regardless of the rotation rate, the limiting current density is
always larger in the case of CCa than in the case of CZn, confirming the higher
ORR activity of CCa. Analysis of the Koutecky-Levich plots (Figure 4b) at different
potentials reveals that the CCa sample exhibits a number of electrons transferred
> 3.6 over the whole range of potentials, suggesting a preferential 4 electrons
pathway towards OH-, whereas the CZn carbon exhibits a slightly lower number
of electrons transferred, in the 3.3-3.8 range. The somewhat lower selectivity of
CZn might be attributed to the presence of a larger amount of pyrrolic-N in CZn.
Thus, several studies suggest that this kind of nitrogen species may contribute to
the activity of the two-electron process [21, 25, 45]. The kinetic current density
(JK), which is another parameter normally used to characterize the activity of ORR
catalysts, was determined by means of the RDE measurements and calculated
on the basis of the Koutecky-Levich analysis [23]. The Koutecky-Levich plot of J1
vs ω-1/2 at a potential of 0.58 V on the CCa and CZn electrode exhibits good
14
linearity in 0.1 M KOH (see Figure 4c). The calculated JK value for the CCa
sample is 16.9 mA cm-2, which is almost twice the value obtained for the CZn
carbon (9.5 mA cm-2) and superior to that of commercial Pt/C catalyst (8.9 mA
cm-2) (see Figure 4d). Furthermore, this value is superior to that of previous
reports on nitrogen-doped carbons [20, 23, 46-48].
b)
CCa
CZn
1.0
1.0
0.5
0.5
-2
0.0
-0.5
-1.0
-1.5
O2
-2.0
N2
-2.5
Current density (mA cm )
-2
Current density (mA cm )
a)
0.0
0.3
0.6
0.9
Potential (V vs RHE)
1.2
0.0
-0.5
-1.0
-1.5
O2
-2.0
N2
-2.5
0.0
0.3
0.6
0.9
Potential (V vs RHE)
1.2
Figure 3. Cyclic voltammograms at a scan rate of 50 mV s-1 for the (a) CCa and
(b) CZn mesoporous carbons in N2- and O2-saturated 0.1 M KOH electrolytes.
15
-
Number of electrons (e )
b)
0
-2
Current density (mA cm )
a)
-1
-2
-3
CCa
CZn
Pt/C
-4
-5
0.0
0.3
0.6
0.9
Potential (V vs RHE)
3.5
3.0
2.5
0.6
0.4
0.2
Potential (V vs RHE)
0.8
d)
0.2
-2
-1
0.3
16
CCa
CZn
0.1
0.0
jK (mA cm )
2
0.4
-1
0.5
J (mA cm )
CCa
CZn
2.0
0.0
1.2
c)
4.0
0.02
0.03
0.04
-1/2
-1/2
w (rpm )
12
8
4
0.05
0
CCa
CZn
Pt/C
Figure 4. (a) RDE polarization curves at 1600 rpm for CCa and CZn samples
compared to 20 wt% Pt/C at different rotation rates in O 2-saturated electrolyte,
(b) number of electrons calculated from Koutecky–Levich analysis, (c) Koutecky–
Levich plot at 0.58 V and (d) electrochemical activity given as the kinetic-limiting
current density (JK) at 0.58 V. Electrolyte: 0.1 M KOH.
To confirm the reaction pathway, a rotating ring-disk electrode (RRDE)
system was used, which makes possible to determine the amount of HO2produced during the oxygen reduction reaction. The experiments were conducted
at a scan rate of 10 mV s-1 and a rotating speed of 1600 rpm. As can be observed
in Figure 5a, the CCa and CZn samples display relatively good catalytic activity
16
and selectivity in KOH aqueous solution, with ring currents an order of magnitude
lower than those recorded on the disk. The yield of peroxide formed in the
presence of both N-doped carbons is < 15 % regardless of the potential (see
Figure 5b), confirming that the reaction proceeds mainly via the four-electron
pathway. For comparison purposes, Pt/C was also tested, showing a low
peroxide yield of ~ 5%, as expected. This result is relevant from an operation
point of view, as the H2O2 produced on the two-electron process causes corrosion
problems [49]. The high performance in KOH electrolyte of the N-doped
mesoporous carbons is further strengthened by a better stability than that of
commercial Pt/C, as evidenced by the chronoamperometric experiments shown
in Figure 6a (10 % higher current retention after 10000 s for both N-doped
carbons), and an excellent electrocatalytic selectivity against methanol
electrooxidation, as revealed by the chronoamperometric experiments in Figure
6b. Thus, a sharp 40 % current loss is registered for commercial Pt/C after
methanol injection, in contrast to a slight current variation in the case of N-doped
mesoporous carbons.
b)
CCa
CZn
Pt/C
30
25
20
15
10
5
0
0.0
20
CCa
CZn
Pt/C
15
H2O2 yield (%)
Disk current (mA)
Ring current (A)
a)
-0.4
10
5
-0.8
-1.2
0.0
0.3
0.6
0.9
Potential (V vs RHE)
1.2
0
0.0
0.2
0.4
0.6
Potential (V vs RHE)
17
0.8
Figure 5. (a) RRDE curves performed at a rotating speed of 1600 rpm and a scan
rate of 10 mV s-1 while the Pt ring electrode was polarized at 1.5 V, and (b)
peroxide yield of CCa, CZn and Pt/C catalysts in O2-saturated 0.1 M KOH
solution.
b)
100
100
80
60
CCa
CZn
Pt/C
40
20
0
2500 5000 7500 10000
time (s)
Relative current (%)
Relative current (%)
a)
75
50
25
0
CH3OH
0
CCa
CZn
Pt/C
200 400 600 800 1000
time (s)
Figure 6. Chronoamperometric responses at a constant rotation speed of 1600
rpm in O2-saturated 0.1 M KOH solution (a) over 10000 s at 0.58 V and (b) over
ca. 1000 s in the presence of methanol (the arrow symbolizes the addition of
methanol).
The effective ORR activity offered by these N-doped carbons may be
attributed to: i) their high porosity made up of large mesopores which reduces
mass-transport limitations [12, 50] and ii) the high density of catalytically active
sites for O2 chemisorption and reduction coupled to a relatively good electronic
conductivity that reduces kinetic limitations [12, 16, 41, 51-54]. Furthermore,
these results clearly show that an appropriate tailoring of the chemical and
structural properties of carbon materials that allow competing with Pt-based
catalysts can be done through cost-effective and environmentally sound
processes. On the other hand, the superior performance of CCa catalyst over
18
CZn correlates well with its higher surface area, higher nitrogen content and,
importantly, higher proportion of pyridinic- and quaternary-N [1, 21, 55-57], which
have been shown to be the active sites for the ORR process, and lower amount
of pyrrolic-N, which seems to contribute to the activity of the two-electron process
as already commented [21, 45, 58]. All these characteristics allow to compensate
for the somewhat lower electronic conductivity of this material (see Table 1).
The electrocatalytic activity of the synthesized materials was also
investigated in acid electrolyte (0.5 M H2SO4). Cyclic voltammetry experiments
were performed in N2- and O2-saturated electrolyte and the results are shown in
Figures 7a-7b. As is usual for N-doped carbon materials [59], the N-doped
mesoporous carbons here developed possess less activity in acid medium, with
a non-visible peak in the cathodic reduction scan. This behavior is further
confirmed by the LSVs at different rotating speeds in Figures S4a-4b, where no
diffusion limiting currents are observable and the reaction is controlled by the
electron transfer kinetics for potentials down to ~ 0.49 V. As can be seen, the Ndoped mesoporous carbons have an onset potential (0.76 V for CCa and 0.67 V
for CZn) which is 50-140 mV more negative than that of commercial Pt/C. A
comparison of both carbons reveals that, as already pointed out in basic media,
CCa offers a better catalytic performance than CZn, with higher onset and halfwave potentials (0.37 V for CCa and 0.33 V for CZn). In spite of the lower activity
recorded in acid medium, their electrocatalytic activity is comparable or higher, to
other metal-free carbon catalyts [15, 48, 60-63].
With regards to the ORR
mechanism operating in acid medium, RRDE measurements show similar results
to those obtained in KOH, i.e. reduction of oxygen takes place predominantly
through the efficient four electron pathway, with a peroxide yield lower than 15 %
19
(Figure 7c). RRDE results reveal also the superior selectivity offered by CCa over
CZn with a lower H2O2 yield over the whole range of potentials. These results
prove the better catalytic activity of the CCa catalyst over CZn in both basic and
acid electrolytes.
a)
b)
CCa
CZn
0.0
-0.5
O2
-1.0
N2
-1.5
0.6
0.9
1.2
15
CCa
CZn
Pt/C
-0.5
O2
N2
-1.0
0.0
d)
0.3
0.6
0.9
1.2
Potential (V vs RHE)
4.0
3.5
3.0
2.5
2.0
0.0
0.2
0.4 0.6
Potential (V vs RHE)
10
5
0
0.0
0.0
0
-2
20
0.5
Current density (mA cm )
0.3
Potential (V vs RHE)
c)
% H2O2
-2
0.5
-2.0
0.0
Current density (mA cm )
1.0
Number of electrons
-2
Current density (mA cm )
1.5
0.2
0.4
0.6
Potential (V vs RHE)
-1
-2
Pt/C
Pt/C (after 3500 cycles)
CCa (first cycle)
CCa (after 3500) cycles
CZn (first cycle)
CZn (after 3500) cycles
-3
0.8
0.0
0.3
0.6
0.9
1.2
Potential (V vs RHE)
Figure 7. Cyclic voltammograms at a scan rate of 50 mV s-1 for the (a) CCa and
(b) CZn mesoporous carbons in N2- and O2-saturated H2SO4 electrolytes (c)
comparison of the peroxide yield of CCa and CZn with that of 20 wt% Pt/C in O2saturated 0.5 M H2SO4 (inset: number of electrons transfer in the case of CCa),
and (d) RDE LSVs of CCa, CZn and Pt/C at 1600 rpm in O2-saturated 0.5 M
H2SO4 before and after 3500 potential cycles.
20
The durability of the N-doped carbon catalysts was assessed by using the
testing method proposed by the US Department of Energy's, which consists in an
accelerated durability test protocol that entails the cycling of the catalysts
between 0.6 and 1.0 V (vs NHE) at 50 mV s−1, under N2 atmosphere, in 0.5 M
H2SO4 [64, 65]. Figure 7d reveals the higher stability of the N-doped mesoporous
carbons in comparison to commercial Pt/C, similarly to the basic medium. Thus,
after 3500 cycles, there is a slight decrease of activity in the case of CCa and
CZn carbons, whereas commercial Pt/C suffers a diminution of 70 mV on the
onset potential and a 30 % loss on the diffusion-limited current density.
Interestingly, after the test, the performance of CCa is comparable to that of Pt/C
(slightly more negative onset and E1/2 potentials, but higher limiting current).
Hence, the initially less active catalyst, can match Pt/C activity after long-term
operation.
4.
Conclusions
In summary, a facile template-free synthesis approach towards nitrogen-doped
mesoporous carbons useful as metal-free ORR electrocatalysts is herein
presented. The synthesis procedure consists on the direct carbonization of citrate
salts followed by an additional heat-treatment step in the presence of melamine.
We have investigated systematically the pore structure, chemical composition
and electrocatalytic activity of the mesoporous carbons derived from two different
citrate salts (i.e. zinc and calcium). The carbon derived from calcium citrate
exhibited an onset potential of 0.9 V (vs. RHE) and a kinetic current density of ca.
17 mA cm-2 in KOH electrolyte, both values being superior to those of the carbon
derived from zinc citrate. Moreover, its catalytic activity is comparable to that of
commercial Pt/C with a higher long-term stability and selectivity against methanol
21
electrooxidation. In acid electrolyte, the onset potential is only 50 mV inferior to
that of Pt/C and 90 mV superior to that of the carbon derived from zinc citrate.
The good electrocatalytic performance offered by the calcium citrate-derived Ndoped mesoporous carbons is the result of an optimum mesopore transport
system which reduces mass-transport limitations and a high density of catalytic
sites (i.e. pyridinic and quaternary-N) which efficiently catalyze the four-electron
process. The lower amount of pyrrolic-N in comparison to the zinc citrate-derived
carbons seems to be essential for reducing hydrogen peroxide generation.
Acknowledgments
This research work was supported by the Spanish Ministerio de Economía y
Competitividad, MINECO (MAT2012-31651), Fondo Europeo de Desarrollo
Regional (FEDER) and FICYT Regional Project (GRUPIN14-102). G. A. F.
thanks the MINECO for his predoctoral contract and M. S. thanks the Spanish
Ministerio de Ciencia e Innovación for her Ramón y Cajal contract.
References
[1]
Gong K, Du F, Xia Z, Durstock M, Dai L. Nitrogen-Doped Carbon Nanotube
Arrays with High Electrocatalytic Activity for Oxygen Reduction. Science.
2009;323(5915):760-4.
[2]
Dai L, Xue Y, Qu L, Choi H-J, Baek J-B. Metal-Free Catalysts for Oxygen
Reduction Reaction. Chem Rev. 2015;115(11):4823-92.
22
[3]
Wang D-W, Su D. Heterogeneous nanocarbon materials for oxygen
reduction reaction. Energy Environ Sci. 2014;7(2):576-91.
[4]
Trogadas P, Fuller TF, Strasser P. Carbon as catalyst and support for
electrochemical energy conversion. Carbon. 2014;75(0):5-42.
[5]
Lee K, Zhang L, Zhang J. PEM Fuel Cell Electrocatalysts and Catalyst
Layers. Springer London: 2008.
[6]
Wu J, Jin C, Yang Z, Tian J, Yang R. Synthesis of phosphorus-doped
carbon hollow spheres as efficient metal-free electrocatalysts for oxygen
reduction. Carbon. 2015;82(0):562-71.
[7]
Jo G, Shanmugam S. Single-step synthetic approach for boron-doped
carbons as a non-precious catalyst for oxygen reduction in alkaline medium.
Electrochem Commun. 2012;25(1):101-4.
[8]
Yang Z, Yao Z, Li G, Fang G, Nie H, Liu Z, et al. Sulfur-doped graphene
as an efficient metal-free cathode catalyst for oxygen reduction. ACS Nano.
2012;6(1):205-11.
[9]
Shao Y, Sui J, Yin G, Gao Y. Nitrogen-doped carbon nanostructures and
their composites as catalytic materials for proton exchange membrane fuel cell.
Applied Catalysis B: Environmental. 2008;79(1):89-99.
[10] Wu K-H, Wang D-W, Su D-S, Gentle IR. A Discussion on the Activity Origin
in Metal-Free Nitrogen-Doped Carbons For Oxygen Reduction Reaction and their
Mechanisms. ChemSusChem. 2015;8(17):2772-88.
[11] Masa J, Xia W, Muhler M, Schuhmann W. On the Role of Metals in
Nitrogen-Doped Carbon Electrocatalysts for Oxygen Reduction. Angew Chem Int
Ed. 2015;54(35):10102-20.
[12] Daems N, Sheng X, Vankelecom IFJ, Pescarmona PP. Metal-free doped
carbon materials as electrocatalysts for the oxygen reduction reaction. J Mater
Chem A. 2014;2(12):4085-110.
[13] Strelko VV, Kuts VS, Thrower PA. On the mechanism of possible influence
of heteroatoms of nitrogen, boron and phosphorus in a carbon matrix on the
catalytic activity of carbons in electron transfer reactions. Carbon.
2000;38(10):1499-503.
[14] Chen P, Wang L-K, Wang G, Gao M-R, Ge J, Yuan W-J, et al. Nitrogendoped nanoporous carbon nanosheets derived from plant biomass: an efficient
catalyst for oxygen reduction reaction. Energy Environ Sci. 2014;7(12):4095-103.
[15] Mo Z, Liao S, Zheng Y, Fu Z. Preparation of nitrogen-doped carbon
nanotube arrays and their catalysis towards cathodic oxygen reduction in acidic
and alkaline media. Carbon. 2012;50(7):2620-7.
[16] Yin J, Qiu Y, Yu J. Porous nitrogen-doped carbon nanofibers as highly
efficient metal-free electrocatalyst for oxygen reduction reaction. J Electroanal
Chem. 2013;702(0):56-9.
[17] Anastasopoulos A, Davies J, Hannah L, Hayden B, Lee C, Milhano C, et
al. The Particle Size Dependence of the Oxygen Reduction Reaction for Carbon‐
Supported Platinum and Palladium. ChemSusChem. 2013;6(10):1973-82.
[18] Brun N, Wohlgemuth SA, Osiceanu P, Titirici MM. Original design of
nitrogen-doped carbon aerogels from sustainable precursors: application as
metal-free oxygen reduction catalysts. Green Chemistry. 2013;15(9):2514-24.
[19] Jaouen F, Lefèvre M, Dodelet J-P, Cai M. Heat-Treated Fe/N/C Catalysts
for O2 Electroreduction: Are Active Sites Hosted in Micropores? J Phys Chem B.
2006;110(11):5553-8.
23
[20] Chen Y, Ma R, Zhou Z, Liu G, Zhou Y, Liu Q, et al. An In Situ SourceTemplate-Interface Reaction Route to 3D Nitrogen-Doped Hierarchical Porous
Carbon as Oxygen Reduction Electrocatalyst. Advanced Materials Interfaces.
2015;2(11):n/a-n/a.
[21] Hung C-T, Yu N, Chen C-T, Wu P-H, Han X, Kao Y-S, et al. Highly
nitrogen-doped mesoscopic carbons as efficient metal-free electrocatalysts for
oxygen reduction reactions. J Mater Chem A. 2014;2(47):20030-7.
[22] Liang J, Zheng Y, Chen J, Liu J, Hulicova-Jurcakova D, Jaroniec M, et al.
Facile Oxygen Reduction on a Three-Dimensionally Ordered Macroporous
Graphitic C3N4/Carbon Composite Electrocatalyst. Angew Chem Int Ed.
2012;51(16):3892-6.
[23] Liu R, Wu D, Feng X, Müllen K. Nitrogen-Doped Ordered Mesoporous
Graphitic Arrays with High Electrocatalytic Activity for Oxygen Reduction. Angew
Chem Int Ed. 2010;49(14):2565-9.
[24] Liu Y-L, Shi C-X, Xu X-Y, Sun P-C, Chen T-H. Nitrogen-doped
hierarchically porous carbon spheres as efficient metal-free electrocatalysts for
an oxygen reduction reaction. J Power Sources. 2015;283:389-96.
[25] Fellinger T-P, Hasche F, Strasser P, Antonietti M. Mesoporous NitrogenDoped Carbon for the Electrocatalytic Synthesis of Hydrogen Peroxide. J Am
Chem Soc. 2012;134(9):4072-5.
[26] Liu Y, Goebl J, Yin Y. Templated synthesis of nanostructured materials.
Chem Soc Rev. 2013;42(7):2610-53.
[27] Bayatsarmadi B, Zheng Y, Jaroniec M, Qiao SZ. Soft-Templating
Synthesis of N-Doped Mesoporous Carbon Nanospheres for Enhanced Oxygen
Reduction Reaction. Chemistry – An Asian Journal. 2015;10(7):1546-53.
[28] ISO 9277:2010. Determination of the specific surface area of solids by gas
adsorption - BET method. Second Edition of ISO 9277, ISO. Geneva; 2012.
[29] Kruk M, Jaroniec M, Gadkaree KP. Nitrogen Adsorption Studies of Novel
Synthetic Active Carbons. J Colloid Interface Sci. 1997;192(1):250-6.
[30] Kruk M, Jaroniec M, Sayari A. Application of Large Pore MCM-41
Molecular Sieves To Improve Pore Size Analysis Using Nitrogen Adsorption
Measurements. Langmuir. 1997;13(23):6267-73.
[31] Wu Z-Y, Xu X-X, Hu B-C, Liang H-W, Lin Y, Chen L-F, et al. Iron Carbide
Nanoparticles Encapsulated in Mesoporous Fe-N-Doped Carbon Nanofibers for
Efficient Electrocatalysis. Angew Chem Int Ed. 2015:n/a-n/a.
[32] Lide DR. CRC handbook of chemistry and physics: CRC press; 2004.
[33] Gasteiger HA, Ross PN. Oxygen Reduction on Platinum Low-Index
Single-Crystal Surfaces in Alkaline Solution: Rotating Ring DiskPt(hkl) Studies.
J Phys Chem. 1996;100(16):6715-21.
[34] Xing W, Yin G, Zhang J. Rotating Electrode Methods and Oxygen
Reduction Electrocatalysts: Elsevier; 2014.
[35] Wang D-W, Li F, Chen Z-G, Lu GQ, Cheng H-M. Synthesis and
Electrochemical Property of Boron-Doped Mesoporous Carbon in
Supercapacitor. Chem Mater. 2008;20(22):7195-200.
[36] Akinwolemiwa B, Peng C, Chen GZ. Redox Electrolytes in
Supercapacitors. J Electrochem Soc. 2015;162(5):A5054-A9.
[37] Ferrero GA, Sevilla M, Fuertes AB. Mesoporous carbons synthesized by
direct carbonization of citrate salts for use as high-performance capacitors.
Carbon. 2015;88(0):239-51.
24
[38] Fuertes AB, Ferrero GA, Sevilla M. One-pot synthesis of microporous
carbons highly enriched in nitrogen and their electrochemical performance. J
Mater Chem A. 2014;2(35):14439-48.
[39] Pels JR, Kapteijn F, Moulijn JA, Zhu Q, Thomas KM. Evolution of nitrogen
functionalities in carbonaceous materials during pyrolysis. Carbon.
1995;33(11):1641-53.
[40] Schmiers H, Friebel J, Streubel P, Hesse R, Köpsel R. Change of chemical
bonding of nitrogen of polymeric N-heterocyclic compounds during pyrolysis.
Carbon. 1999;37(12):1965-78.
[41] Qiu K, Guo ZX. Hierarchically porous graphene sheets and graphitic
carbon nitride intercalated composites for enhanced oxygen reduction reaction.
J Mater Chem A. 2014;2(9):3209-15.
[42] Loukrakpam R, Luo J, He T, Chen Y, Xu Z, Njoki PN, et al.
Nanoengineered PtCo and PtNi Catalysts for Oxygen Reduction Reaction: An
Assessment of the Structural and Electrocatalytic Properties. J Phys Chem C.
2011;115(5):1682-94.
[43] Liang H-W, Zhuang X, Brüller S, Feng X, Müllen K. Hierarchically porous
carbons with optimized nitrogen doping as highly active electrocatalysts for
oxygen reduction. Nat Commun. 2014;5.
[44] Jin H, Huang H, He Y, Feng X, Wang S, Dai L, et al. Graphene Quantum
Dots Supported by Graphene Nanoribbons with Ultrahigh Electrocatalytic
Performance for Oxygen Reduction. J Am Chem Soc. 2015;137(24):7588-91.
[45] Lai L, Potts JR, Zhan D, Wang L, Poh CK, Tang C, et al. Exploration of the
active center structure of nitrogen-doped graphene-based catalysts for oxygen
reduction reaction. Energy Environ Sci. 2012;5(7):7936-42.
[46] Liu X, Zhu H, Yang X. One-step synthesis of dopamine-derived
micro/mesoporous nitrogen-doped carbon materials for highly efficient oxygenreduction catalysts. J Power Sources. 2014;262:414-20.
[47] Xu J, Dong G, Jin C, Huang M, Guan L. Sulfur and Nitrogen Co-Doped,
Few-Layered Graphene Oxide as a Highly Efficient Electrocatalyst for the
Oxygen-Reduction Reaction. ChemSusChem. 2013;6(3):493-9.
[48] Parvez K, Yang S, Hernandez Y, Winter A, Turchanin A, Feng X, et al.
Nitrogen-Doped Graphene and Its Iron-Based Composite As Efficient
Electrocatalysts for Oxygen Reduction Reaction. ACS Nano. 2012;6(11):954150.
[49] Zhang J. PEM fuel cell electrocatalysts and catalyst layers: fundamentals
and applications: Springer Science & Business Media; 2008.
[50] Wang H, Bo X, Luhana C, Guo L. Nitrogen doped large mesoporous
carbon for oxygen reduction electrocatalyst using DNA as carbon and nitrogen
precursor. Electrochem Commun. 2012;21(0):5-8.
[51] Geng D, Chen Y, Chen Y, Li Y, Li R, Sun X, et al. High oxygen-reduction
activity and durability of nitrogen-doped graphene. Energy Environ Sci.
2011;4(3):760-4.
[52] Sheng X, Daems N, Geboes B, Kurttepeli M, Bals S, Breugelmans T, et
al. N-doped ordered mesoporous carbons prepared by a two-step nanocasting
strategy as highly active and selective electrocatalysts for the reduction of O2 to
H2O2. Applied Catalysis B: Environmental. 2015;176–177(0):212-24.
[53] Zhang L, Su Z, Jiang F, Yang L, Qian J, Zhou Y, et al. Highly graphitized
nitrogen-doped porous carbon nanopolyhedra derived from ZIF-8 nanocrystals
25
as efficient electrocatalysts for oxygen reduction reactions. Nanoscale.
2014;6(12):6590-602.
[54] Tao G, Zhang L, Chen L, Cui X, Hua Z, Wang M, et al. N-doped
hierarchically macro/mesoporous carbon with excellent electrocatalytic activity
and durability for oxygen reduction reaction. Carbon. 2015;86(0):108-17.
[55] Niwa H, Horiba K, Harada Y, Oshima M, Ikeda T, Terakura K, et al. X-ray
absorption analysis of nitrogen contribution to oxygen reduction reaction in
carbon alloy cathode catalysts for polymer electrolyte fuel cells. J Power Sources.
2009;187(1):93-7.
[56] Xu B, Zheng D, Jia M, Cao G, Yang Y. Nitrogen-doped porous carbon
simply prepared by pyrolyzing a nitrogen-containing organic salt for
supercapacitors. Electrochim Acta. 2013;98(0):176-82.
[57] Wang D-W, Li F, Yin L-C, Lu X, Chen Z-G, Gentle IR, et al. Nitrogen-Doped
Carbon Monolith for Alkaline Supercapacitors and Understanding NitrogenInduced Redox Transitions. Chemistry – A European Journal. 2012;18(17):534551.
[58] Fellinger T-P, Hasché F, Strasser P, Antonietti M. Mesoporous NitrogenDoped Carbon for the Electrocatalytic Synthesis of Hydrogen Peroxide. J Am
Chem Soc. 2012;134(9):4072-5.
[59] Chen Z, Higgins D, Yu A, Zhang L, Zhang J. A review on non-precious
metal electrocatalysts for PEM fuel cells. Energy Environ Sci. 2011;4(9):3167-92.
[60] Elumeeva K, Fechler N, Fellinger TP, Antonietti M. Metal-free ionic liquidderived electrocatalyst for high-performance oxygen reduction in acidic and
alkaline electrolytes. Materials Horizons. 2014;1(6):588-94.
[61] Wei W, Liang H, Parvez K, Zhuang X, Feng X, Müllen K. Nitrogen-Doped
Carbon Nanosheets with Size-Defined Mesopores as Highly Efficient Metal-Free
Catalyst for the Oxygen Reduction Reaction. Angew Chem Int Ed.
2014;53(6):1570-4.
[62] Perazzolo V, Durante C, Pilot R, Paduano A, Zheng J, Rizzi GA, et al.
Nitrogen and sulfur doped mesoporous carbon as metal-free electrocatalysts for
the in situ production of hydrogen peroxide. Carbon. 2015;95:949-63.
[63] Wohlgemuth S-A, White RJ, Willinger M-G, Titirici M-M, Antonietti M. A
one-pot hydrothermal synthesis of sulfur and nitrogen doped carbon aerogels
with enhanced electrocatalytic activity in the oxygen reduction reaction. Green
Chemistry. 2012;14(5):1515-23.
[64] Wu G, More KL, Johnston CM, Zelenay P. High-Performance
Electrocatalysts for Oxygen Reduction Derived from Polyaniline, Iron, and Cobalt.
Science. 2011;332(6028):443-7.
[65] Li Y, Zhou W, Wang H, Xie L, Liang Y, Wei F, et al. An oxygen reduction
electrocatalyst based on carbon nanotube-graphene complexes. Nat Nano.
2012;7(6):394-400.
26