Biomimetic reduction of O2 in an acid medium on iron phthalocyanines axially coordinated to pyridine anchored on carbon nanotubes

Article abstract of DOI:10.1039/c7ta02381b

An efficient and inexpensive catalyst for the oxygen reduction reaction (ORR) is the key missing component for large-scale development of fuel cells. Bio-inspired tethered electrocatalysts could be the solution to this problematic reaction. Either unsubstituted Fe phthalocyanine (FePc) or Fe hexadecachloro-phthalocyanine (16(Cl)FePc) was anchored to carbon nanotubes (CNTs) via a pyridine axial ligand. The results show that the fifth coordination plays a major role in increasing the catalytic activity of FePc and 16(Cl)FePc for the ORR. The coordination also allows the decoupling of the metal centre from the carbon support, thus changing the geometrical and electronic structure and hindering the production of H₂O₂. The pentacoordinated catalysts were stable in acidic pH according to the rotating disk analysis, but the activity of the hexadecachloro compound was not higher than that of the unsubstituted phthalocyanine. Cl atoms reduced the coupling between O₂ and Fe, mismatching the energy of the frontier orbitals and lowering the activity towards the reduction of O₂.

Full text of DOI:10.1039/c7ta02381b

Journal of
Materials Chemistry A
View Article Online
View Journal
COMMUNICATION
Biomimetic reduction of O2 in an acid medium on
iron phthalocyanines axially coordinated to
pyridine anchored on carbon nanotubes
Cite this: DOI: 10.1039/c7ta02381b
ab
ab
Received 17th March 2017
Accepted 15th May 2017
*
Francisco J. Recio,
Ricardo Venegas,
Jorge Riquelme,cd Karinna Neira,d
Jose F. Marco,e Ingrid Ponce,d Jose H. Zagald and Federico Tasca
d
*
´
´
DOI: 10.1039/c7ta02381b
rsc.li/materials-a
An efficient and inexpensive catalyst for the oxygen reduction reaction a catalyst in the ORR1–4 and other reactions5–8 has been inves-
(ORR) is the key missing component for large-scale development of tigated for several years as reviewed in the references,1 but little
fuel cells. Bio-inspired tethered electrocatalysts could be the solution progress has been made in improving major drawbacks such as
to this problematic reaction. Either unsubstituted Fe phthalocyanine an activity lower than Pt and poor stability in the presence of
(FePc) or Fe hexadecachloro-phthalocyanine (16(Cl)FePc) was fuel cell electrolytes, especially in acidic environments.4,9–12 In
anchored to carbon nanotubes (CNTs) via a pyridine axial ligand. The an attempt to overcome these drawbacks, the following strate-
results show that the fifth coordination plays a major role in increasing gies have been adopted: (i) the synthesis of complexes with
the catalytic activity of FePc and 16(Cl)FePc for the ORR. The coordi- more positive redox potentials;13–15 (ii) the pyrolysis of the MN4
nation also allows the decoupling of the metal centre from the carbon complexes with different carbon supports;16–19 and (iii) the
support, thus changing the geometrical and electronic structure and support of MN4 complexes on different carbon nanostructures
hindering the production of H2O2. The pentacoordinated catalysts with or without coordination of a h ligand.20–26 Already in the
were stable in acidic pH according to the rotating disk analysis, but the 70's, Randin13 and Beck14 had proposed that the activity of MN4
activity of the hexadecachloro compound was not higher than that of catalysts could be related to the M(III)/(II) redox potential, but
the unsubstituted phthalocyanine. Cl atoms reduced the coupling rather few authors developed this concept any further.4 Volcano
between O2 and Fe, mismatching the energy of the frontier orbitals and correlations have then been obtained in which the activity is
lowering the activity towards the reduction of O2.
correlated with the formal redox potential of MN4.15,27–29 The
heat treatment modies the ligand structure around the metal,
thereby making it more electron-withdrawing, and shis the
M(III)/(II) redox potential in the positive direction. This partially
explains the higher activity of these materials,30 which can
compete with Pt-based catalysts.3,31 Unfortunately, the struc-
tural changes are still unpredictable and therefore difficult to
rationalize.30,32,33 On the other hand, electrocatalysts with
increased activity and stability21–24 have been produced through
the synthesis of new penta-coordinated catalysts whose struc-
tures mimic the structures that exist in nature, such as the
active site of the cytochrome c oxidase and haemoglobin.
In this communication, we analyse the catalytic activity of
unsubstituted Fe phthalocyanine (FePc) and of Fe hexadeca-
chloro-phthalocyanine (16(Cl)FePc) for the reduction of O2 in
acid, in the absence and in the presence of a pyridine axial
ligand attached to the external walls of the CNTs. FeN4 moieties
were either absorbed directly onto the CNT by p interactions
(CNT–FeN4) or coordinated to pyridine axial ligands previously
attached to the CNT by diazonium salt modication (CNT–Py–
FeN4). Among the MN4 catalysts, Fe-based MN4 are the most
promising due to their high activity and the fact that they
promote the direct ORR to water, i.e. bypassing the production
Introduction
The oxygen reduction reaction (ORR) is an important process in
fuel cells, but its sluggish kinetics gives rise to a high over-
potential in the cathode, which causes a voltage loss of the cell
under operating conditions. This problem is partially solved by
using very costly Pt-based catalysts in the O2 cathode. A class of
catalysts known as “non-precious metal catalysts” (NPMC), have
been developed to substitute Pt used in the cathode of fuel cells.
Metal-phthalocyanines and metal-porphyrins i.e. moieties con-
taining a metal (e.g. Fe or Co) and four pyrrolic nitrogen atoms
(MN4) belong to this class of catalysts. The role of MN4 as
a
´
´
Facultad de Quımica, Ponticia Universidad Catolica de Chile, Santiago, Chile.
E-mail: javier.recio@uc.cl
b
´
Centro de Nanotecnologıa y Materiales Avanzados, CIEN-UC, Ponticia Universidad
´
Catolica de Chile, Santiago, Chile
cFacultad de Ciencias Naturales, Matematicas
´
y
Medioambiente, Universidad
´
Tecnologica Metropolitana, Santiago, Chile
d
´
´
Facultad de Quımica y Biologıa, Universidad de Santiago de Chile, Santiago, Chile.
E-mail: federico.tasca@usach.cl
e
´
´
Instituto de Quımica Fısica “Rocasolano”, CSIC, Madrid, Spain
This journal is © The Royal Society of Chemistry 2017
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Communication
of H2O2 in alkaline media. 16(Cl)FePc is known for the high Fe(III)/(II) redox couple in 16(Cl)FePc has a higher redox poten-
redox potential of the Fe redox centre which stems from the tial compared to the unsubstituted FePc due to the presence of
electron-withdrawing effect of the Cl atoms present on the electron-withdrawing Cl atoms, which are distributed in all
phthalocyanine ring,1,15,34 and presents higher activity than positions in the phthalocyanine ligand (0.610 V, Fig. 1a). The
unsubstituted FePc.1 While the activity of FePc anchored on basicity of the carbon nanotube can also alter the electron
CNTs via an axial ligand has been reported,20–24 little is known density on the Fe centre by withdrawing electron density from
about the effect of the axial ligand on other MN4 compounds. the metal centre. The shi of Fe(III)/(II) transition to more
When 16(Cl)FePc is coordinated to pyridine, the electron positive values is favourable for ORR catalysis.4,30 Similar effects
cloud is shared between the ligand and the Cl groups, which have been observed in graphitic carbon supports with different
causes a change in the geometry of the catalyst and lowers its Lewis basicity, which also affects the electron density of the Fe
performance.
centre. In addition, a linear relationship is found between the
Lewis basicity of the graphitic support and the turnover
frequency for the ORR on FeNx/C sites, i.e., the higher the
basicity, the higher the turnover frequency.39
Materials and methods
Iron(II)
phthalocyanine
(FePc)
and
iron(II)
In the presence of the pyridine moiety, the redox potential of
1,2,3,4,8,9,10,11,15,16,17,18,22,23,24,25-hexadeca(chloro) the Fe(III)/(II) redox couple in FePc shis around 40 mV towards
phthalocyanine (16(Cl)FePc) were supplied by Porphy Chem more positive potentials as an effect of electron-pulling from the
(Dijon, France). 4-Aminopyridine (Py), N,N-dimethylformamide pyridine linker. Similar results were obtained for the CNT–Py–
(DMF), isopropyl alcohol, NaOH, and H2SO4 were obtained 16(Cl)FePc catalyst (0.650 V, Fig. 1a). The mean values (obtained
from Sigma (St. Louis, USA). Double-walled CNTs were provided from ve repeated experiments) of the redox potentials of the
by Nanocyl (Sambreville, Belgium). We chose working with Fe(III)/(II) redox couples and surface concentration values of
double-walled CNTs instead of multi- or single-walled CNTs CNT–MN4 and CNT–Py–MN4 determined by the integration of
because of their higher purity (>90%).35,36 Similar to that in ref. the redox peak are summarized in Table 1.
22, the functionalization of the CNTs with Py was obtained
XPS analyses conrm the existence of an oxidized state of Fe
using the diazonium reaction. Briey, 5 g of NaNO2 was dis- in CNT–Py–FePc (Fig. 1b). The spectrum that corresponds to
solved in 10 ml of H2O; 7 g of Py was dissolved in 5 ml of 4 M FePc contains a sharp spin–orbit doublet (Fe 2p3/2 ¼ 708.4 eV
HCl; and 0.1 g of CNTs was dispersed in 200 ml of DMF. The and Fe 2p1/2 ¼ 721.0 eV) and broad features at 710.5 eV and
CNT dispersion was added to the NaNO2 and Py solution. CNT– 723.0 eV. Early investigations identied these broad features as
Py was collected by ltering. Next, CNT–Py was modied with shake-up satellites of the Fe2+ in the phthalocyanine.40 Recent
either FePc or 16(Cl)FePc by reuxing in N2 to obtain either investigations have associated the sharp features with the
CNT–Py–FePc or CNT–Py–(16)ClFePc. The ink formulation was presence of Fe2+ and the broad features appearing at higher
obtained by dispersing 10 mg of CNT–Py–FePC or CNT–Py–(16) binding energy as arising from the presence of a Fe3+ spin–orbit
ClFePc in 1 ml of a mixture of 25% volume isopropyl alcohol doublet.22,38 The corresponding binding energies of these broad
and 75% H2O. 10 ml of catalyst ink was loaded onto the surface features are fully compatible with those shown by the Fe 2p3/2
of the electrode and let to dry (nal loading was 0.1 mg cmꢀ2). and Fe 2p1/2 core levels of Fe3+ species. The spectrum recorded
The working electrode was a glassy carbon rotating ring disk from the CNT–Py–FePc sample does not show these sharp
electrode (GCE) of 4 mm diameter with a Pt ring from ALS features but only a spin–orbit doublet characterized by binding
(Tokyo, Japan). Electrochemical experiments were carried out energies (Fe 2p3/2 ¼ 710.7 eV and Fe 2p1/2 ¼ 723.8 eV) typical of
on a Chi Instruments (Austin, USA) electrochemical bi-poten- Fe3+. The disappearance of the sharp features at lower binding
tiostat. In RRDE experiments, the ring potential was set to 0.6 V energies in the Fe 2p spectrum of the CNT–Py–FePc sample and
vs. SCE. XPS data were recorded under a vacuum better than 5 ꢁ the shi of the spectrum to higher binding energies (as
10ꢀ10 mbar using a PHOIBOS-150 electron analyser (SPECS), Al compared with the spectrum recorded from FePc) are compat-
Ka radiation and a constant pass energy of 20 eV.
ible with a decrease in the electron density around the iron
atom in the phthalocyanine structure. This suggests, as
mentioned above, that in the pyridine sample there is
a permanent displacement of the electron population of the
Results and discussion
MN4 compounds are known for being stable in basic pH, but iron atom due to the presence of this ligand. The Fe 2p XPS
only a few authors have reported on the catalytic activity of MN4 spectrum recorded from (16)ClFePc presents intense Fe3+
coordinated to a h ligand in an acidic environment.23,24,37,38 components and clear shoulders at 708.4 eV and 721.0 eV,
In Fig. 1a, we show the schematic structure and the electro- indicating that this sample also contains a Fe2+ contribution.
chemical characterization in N2 saturated 0.1 M H2SO4 solution This component is, in any case, much smaller than the Fe2+
of the CNTs modied with pyridine (CNT–Py) and of the FeN4 contribution observed in FePc. This implies that the electron
used in this work (i.e. FePc and 16(Cl)FePc), in the absence and density around the iron is lower in this sample compared to
in the presence of the pyridine axial link. In all the voltammo- FePc, probably because it is displaced towards the Cl atom,
grams the redox peaks of the Fe(III)/(II) redox couple are high- which is in agreement with the voltammetric results shown in
lighted. In the case of FePc, the Fe(III)/(II) redox peak appears at Fig. 1a. The spectrum corresponding to CNT–Py–(16)ClFePc
potentials close to 0.350 V vs. SCE. As mentioned above, the shows a clear increase of the sharp Fe2+ features in comparison
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2017

View Article Online
Communication
Journal of Materials Chemistry A
with the spectrum of (16)ClFePc. This indicates that the addi-
tion of CNT–Py increases the electron density around the iron
atom, contrarily to the observed in the CNT–Py–FePc sample.
Nevertheless, the intensity of the Fe2+ contribution to the
spectrum of CNT–Py–(16)ClFePc is lower than that to the spec-
trum of FePc, which implies that the CNT–Py–(16)ClFePc
sample is an intermediate situation between FePc and CNT–Py–
FePc.
ORR polarization curves are presented in Fig. 2a. In the
presence of the axial ligand, the overpotential for the ORR is
reduced by 80 mV. The onset potential of the electrocatalytic
waves starts at 0.55 V for electrodes modied with CNT–Py–FePc
and at 0.56 V for the CNT–Py–16(Cl)FePc. The coordination of
both axial positions in FePc (octahedral complex geometry) is
energetically much more favourable than single axial site
coordination (square pyramidal geometry of the complex).24
The total number of electrons determined by Koutecky–
Levich extrapolation (inset, Fig. 2a) for the catalysts is
summarized in Table 1. The pentacoordination favoured the
4eꢀ reduction process, as opposed to CNT–FeN4, which
promoted 2eꢀ. The production of hydrogen peroxide was also
corroborated by RRDE, as shown in Fig. 2b. In the absence of
the axial ligand, more peroxide is formed. The CNT–16(Cl)FePc
catalyst is the most active in terms of peroxide formation,
while in the presence of the pyridine ligand very low amounts
of peroxide are detected. The peroxide formed on electrodes
only modied with CNT–Py is presented as a comparative
reference.
Tafel plots exhibit slopes that vary from ꢀ0.085 V to ꢀ0.104 V
decꢀ1, which could be attributed to a mixture of two parallel
mechanisms with different rate-determining steps. A schematic
mechanism based on previous studies4,24,39,41 is presented in
Scheme 1. The mechanism provides two pathways for the rate-
determining step. The pathways are polarization dependent
and consider that Fe2+ species are thermodynamically favour-
able at high polarization. It must be noted that the reactions
proposed in Scheme 1 are only possible mechanistic steps that
occur aer the rate-determining step.
As observed for FeN4 macrocyclics adsorbed directly on
graphite electrodes,15,27–29 for CNT–MN4 and CNT–Py–MN4, the
onset potential for the ORR also appears close to the Fe(III)/(II)
formal redox potential, so the reaction essentially starts when
Fe(II) active sites are generated on the surface from the reduc-
tion of Fe(III). Fig. 2c shows the relationship between the formal
potential of the catalysts and the catalytic activity expressed as
log I (E ¼ cte) normalized by FeN4 surface concentration (G).
Fig. 1 (a) Electrochemical characterization by cyclic voltammetry and
schematic representation of CNT–FePc; CNT–Py–FePc; CNT–16(Cl)
FePc; CNT–Py–16(Cl)FePc; and CNT–Py. Conditions: N2 or O2 satu-
rated 0.1 M H2SO4 solution and a scan rate of 0.1 V sꢀ1 (b) XPS Fe 2p
spectra of FePc; CNT–Py–FePc; 16(Cl)FePc; and CNT–Py–16(Cl)FePc.
Table 1 Fe(III)/(II) formal potential, Tafel slope, number of electrons transferred during the ORR, and surface concentration values determined by
the integration of the redox peaks of the Fe(III)/(II) redox couples for CNT–FePc; CNT–Py–FePc; CNT–16(Cl)FePc; or CNT–Py–16(Cl)FePc
Tafel slope (V decꢀ1
)
No eꢀ
G (mol cmꢀ2
)
ꢂ
Complex
E(Fe(
(V vs. SCE)
III)/Fe(II))
CNT–FePc
0.352 ꢃ 0.004
0.395 ꢃ 0.003
0.613 ꢃ 0.003
0.648 ꢃ 0.006
ꢀ0.085 ꢃ 0.012
ꢀ0.098 ꢃ 0.003
ꢀ0.104 ꢃ 0.016
ꢀ0.086 ꢃ 0.002
2.56 ꢃ 0.07
3.57 ꢃ 0.02
1.73 ꢃ 0.02
3.45 ꢃ 0.07
1.62 ꢁ 10ꢀ8
3.92 ꢁ 10ꢀ9
3.68 ꢁ 10ꢀ8
8.87 ꢁ 10ꢀ9
CNT–Py–FePc
CNT–16(Cl)FePc
CNT–Py–16(Cl)FePc
This journal is © The Royal Society of Chemistry 2017
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Communication
Fig. 2 (a) Polarization curves recorded at electrodes modified with CNT–FePc; or CNT–Py–FePc; or CNT–16(Cl)FePc; or CNT–Py–16(Cl)FePc.
Conditions: O2 saturated 0.1 M H2SO4 solution, 400, 800, 1200, and 2400 rpm, and a scan rate of 0.005 V sꢀ1. (b) H2O2 oxidation measured at
a Pt ring electrode when the disk was modified with CNT–16(Cl)FePc; or CNT–FePc; CNT–Py–FePc; CNT–Py–16(Cl)FePc; or CNT–Py.
Conditions: O2 saturated 0.1 M H2SO4 solution. A Pt ring polarized at 0.6 V, while the disk was rotating at 1200 rpm and a scan rate of 0.005 V sꢀ1
.
(c) Plots of catalytic activity as log i/G measured at 0.2 V vs. the formal potential of Fe(III)/(II) for CNT–FePc; CNT–Py–FePc; CNT–16(Cl)FePc; or
CNT–Py–16(Cl)FePc. Conditions: O2 saturated 0.1 M H2SO4 solution and 1200 rpm.
followed in the presence of the axial ligand that anchors the
MN4 complex to the CNTs, the increase in activity is more
pronounced for FePc than for 16(Cl)FePc. Thus, the electron
withdrawing groups reduce the gap between the energy of the
frontier orbitals of the metal complex and the O2 molecule.4,43,44
The back-bonding processes could explain the difference
between the behaviour of FePc and 16(Cl)FePc. When O2 binds
to the metal in MN4, its 2p electrons interact with the partially
lled d orbitals of the same. These processes are accompanied
by intermolecular electron transfer, in which O2 accepts charge
density from the partially lled d orbitals of the metal via back-
bonding to the p* antibonding orbital, and donates charge
from a lled p molecular orbital to a half-lled dz orbital of the
metal.2,11,12 The formation of the bond between O2 and Fe
requires that the energies of the predominant d orbitals be the
same as or similar to those of the charge transfer intermediate
species.4,15,30,43 Therefore, the active site in the anchored
complex is harder compared to the adsorbed catalyst. A hard
active site in FePc and 16(Cl)FePc (from hard–so acid base
(HSAB) principle) would promote high activity for the ORR, as
O2 is a hard base.4,30,42,43 In the case of the CNT–Py–16(Cl)FePc
derivative, the metal centre suffers the inductive effects of
chlorine atoms located on the ligand, and the inuence of the
pyridine axial ligand, which promotes a higher oxidation state.10
However, this decrease in the electron density of the metal
Scheme 1 Proposed ORR mechanism for CNT–Py–FeN4.
This conrms once more that regardless of the spatial cong-
uration of the FeN4 complexes, shis of Fe(III/II) to more positive
potentials are benecial to the catalysis of the ORR, as the redox
potential is a reactivity descriptor that predicts the increase of
reactivity4,13,14,28,30 in these catalysts. Although this trend is
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2017

View Article Online
Communication
Journal of Materials Chemistry A
centre, due to the chlorine atoms of the macrocyclic ligand and
pyridine axial ligand, causes changes in the back-bonding
processes, as well as the mismatch of the respective donor–
acceptor orbital energies, namely, less electronic coupling
between the donor and the acceptor and less activity for CNT–
Py–16(Cl)FePc systems compared to CNT–Py–FePc.
7 R. O. Lezna, S. Juanto and J. H. Zagal, J. Electroanal. Chem.,
1998, 452, 221–228.
8 M. J. Aguirre, M. Isaacs, F. Armijo, N. Bocchi and J. H. Zagal,
Electroanalysis, 1998, 10, 571–575.
9 C. Linares, D. Geraldo, M. Paez and J. H. Zagal, J. Solid State
Electrochem., 2003, 7, 626–632.
10 A. A. Gewirth and M. S. Thorum, Inorg. Chem., 2010, 49,
3557–3566.
11 A. A. Tanaka, C. Fierro, D. Scherson and E. B. Yaeger, J. Phys.
Chem., 1987, 91, 3799–3807.
Conclusions
Likewise in nature, the h axial ligand plays a critical role in
´
the increase of the catalytic activity of FePc and 16(Cl)FePc for 12 J. Zagal, M. Paez, A. A. Tanaka, J. R. dos Santos and
the ORR in the trend CNT–MN4 < CNT–Py–MN4. The activity of C. A. Linkous, J. Electroanal. Chem., 1992, 339, 13–30.
FeN4 can be tuned by changing the electron-withdrawing power 13 J.-P. Randin, Electrochim. Acta, 1974, 19, 83–85.
of the ligand substituents on the macrocyclic ligand, which 14 F. Beck, J. Appl. Electrochem., 1977, 7, 239–245.
makes it possible to tune the donor–acceptor electronic 15 J. Masa, K. Ozoemena, W. Schuhmann and J. H. Zagal, J.
coupling (Fe–O2 adduct).1,24 The higher the separation between
Porphyrins Phthalocyanines, 2012, 16, 761.
the energy of the frontier orbitals of the donor (FePc) and the 16 D. Scherson, A. A. Tanaka, S. L. Gupta, D. Tryk, C. Fierro,
acceptor (O2), the lower the reactivity.43 Thus, the electron-
withdrawing groups reduce the gap between the energies of the
R. Holze, E. B. Yeager and R. P. Lattimer, Electrochim. Acta,
1986, 31, 1247–1258.
frontier orbitals. However, when comparing CNT–Py–FePc with 17 M. Lefevre and J. P. Dodelet, ECS Trans., 2012, 45, 35.
CNT–Py–16(Cl)FePc, the activity for CNT–Py–16(Cl)FePc was 18 R. Zhang, Y. Peng, Z. Li, K. Li, J. Ma, Y. Liao, L. Zheng, X. Zuo
lower than the activity of CNT–Py–FePc. The lower increase in
activity for CNT–Py–16(Cl)FePc in the presence of the back 19 I. Kruusenberg, L. Matisen and K. Tammeveski, Int. J.
ligand compared to FePc is attributed to two pulling effects, Electrochem. Sci., 2013, 8, 1057–1066.
namely the presence of the electron-withdrawing –Cl groups on 20 I. Ponce, J. F. Silva, R. Onate, S. Miranda-Rojas, A. Munoz-
and D. Xia, Electrochim. Acta, 2014, 147, 343–351.
˜
˜
´
the phthalocyanine ligand and the axial ligand, which operate
simultaneously on the metal centre. This causes changes in the
Castro, R. Arratia-Perez, F. Mendizabal and J. H. Zagal, J.
Phys. Chem. C, 2011, 115, 23512–23518.
˜
backbonding processes and the mismatching of the respective 21 I. Ponce, J. F. Silva, R. Onate, M. C. Rezende, M. A. Paez,
donor–acceptor orbital energies.
J. H. Zagal, J. Pavez, F. Mendizabal, S. Miranda-Rojas,
˜
´
Finally, the presence of the conjugated axial ligand allows for
the decoupling of the metal centre from the electrode surface,
A. Munoz-Castro and R. Arratia-Perez, J. Phys. Chem. C,
2012, 116, 15329–15341.
thereby changing the geometrical and electronic structure, 22 R. Cao, R. Thapa, H. Kim, X. Xu, M. Gyu Kim, Q. Li, N. Park,
which hinders the production of H2O2 and therefore explains
the higher stability in acidic environments.
M. Liu and J. Cho, Nat. Commun., 2013, 4, 2066.
23 P.-J. Wei, G.-Q. Yu, Y. Naruta and J.-G. Liu, Angew. Chem., Int.
Ed., 2014, 53, 6659–6663.
˜
24 J. Chlistunoff and J.-M. Sansinena, J. Phys. Chem. C, 2014,
Acknowledgements
118, 19139–19149.
¨
F. T. acknowledges the nancial support of Fondecyt 11130167, 25 K.-K. Turk, I. Kruusenberg, J. Mondal, P. Rauwel, J. Kozlova,
and Dicyt 021342TG_DAS, and “Proyectos Basales”. J. H. Z
acknowledges the nancial support of Millenium RC120001,
L. Matisen, V. Sammelselg and K. Tammeveski, J.
Electroanal. Chem., 2015, 756, 69–76.
Fondecyt 1140199 and Project Anillo CT-1412. F. J. R acknowl- 26 K.-K. Urk, K. Kaare, I. Kruusenberg, M. Merisalu, U. Joost,
edges the nancial support of Fondecyt 1161117. I. P.
acknowledges the nancial support of PAI-CONICYT 79150041.
L. Matisen, V. Sammelselg, J. H. Zagal and K. Tammeveski,
J. Electrochem. Soc., 2017, 164, 338–344.
27 J. Zagal, P. Brinda and E. Yeager, J. Electrochem. Soc., 1980,
127, 1506–1517.
Notes and references
˜
28 J. H. Zagal, F. Javier Recio, C. A. Gutierrez, C. Zuniga,
´
1 J. H. Zagal, S. Griveau, J. F. Silva, T. Nyokong and F. Bedioui,
Coord. Chem. Rev., 2010, 254, 2755–2791.
M. A. Paez and C. A. Caro, Electrochem. Commun., 2014, 41,
24–26.
2 A. B. Sorokin, Chem. Rev., 2013, 113, 8152–8191.
3 Y. Liu, X. Yue, K. Li, J. Qiao, D. P. Wilkinson and J. Zhang,
Coord. Chem. Rev., 2016, 315, 153–177.
4 J. H. Zagal and M. T. M. Koper, Angew. Chem., Int. Ed., 2016,
55, 14510–14521.
29 J. Masa and W. Schuhmann, Chem.–Eur. J., 2013, 19, 9644–
9654.
30 J. Zagal, I. Ponce, D. Baez, R. Venegas, J. Pavez, M. Paez and
M. Gulppi, Electrochem. Solid-State Lett., 2012, 15, B90–B92.
31 R. Bashyam and P. Zelenay, Nature, 2006, 443, 63–66.
`
5 M. Mimica, F. Bedioui and J. H. Zagal, Electrochim. Acta, 32 F. Jaouen, A. M. Serventi, M. Lefevre, J.-P. Dodelet and
2002, 48, 323–329. P. Bertrand, J. Phys. Chem. C, 2007, 111, 5971–5976.
6 S. Griveau, M. Gulppi, J. Pavez and J. H. Zagal, 33 S. Gupta, D. Tryk, I. Bae, W. Aldred and E. Yeager, J. Appl.
Electroanalysis, 2003, 15, 779–785.
Electrochem., 1989, 19, 19–27.
This journal is © The Royal Society of Chemistry 2017
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Communication
34 F. Tasca, F. J. Recio, R. Venegas, D. A. Geraldo, M. Sancy and 39 N. Ramaswamy, U. Tylus, Q. Jia and S. Mukerjee, J. Am.
J. H. Zagal, Electrochim. Acta, 2014, 140, 314–319. Chem. Soc., 2013, 135, 15443–15449.
35 A. Morozan, S. Campidelli, A. Filoramo, B. Jousselme and 40 G. V. Ouedraogo, D. Benlian and L. Porte, J. Chem. Phys.,
S. Palacin, Carbon, 2011, 49, 4839–4847. 1980, 73, 642–647.
36 F. Tasca, W. Harreither, R. Ludwig, J. J. Gooding and 41 E. F. Holby and P. Zelenay, Nano Energy, 2016, 29, 54–64.
L. Gorton, Anal. Chem., 2011, 83, 3042–3049. 42 J. Zagal, J. Electrochem. Soc., 1980, 127, 1506.
37 X. Wang, B. Wang, J. Zhong, F. Zhao, N. Han, W. Huang, 43 J. H. Zagal and G. I. Cardenas-Jiron, J. Electroanal. Chem.,
M. Zeng, J. Fan and Y. Li, Nano Res., 2016, 9, 1497–1506. 2000, 489, 96–100.
38 Y. Peng, Z. Li, D. Xia, L. Zheng, Y. Liao, K. Li and X. Zuo, J. 44 G. I. Cardenas-Jiron and J. H. Zagal, J. Electroanal. Chem.,
´
´
´
´
Power Sources, 2015, 291, 20–28.
2001, 497, 55–60.
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2017