Electrocatalytic reduction of dioxygen at a modified glassy carbon electrode based on Nafion-dispersed single-walled carbon nanotubes and cobalt-porphyrin with palladium nanoparticles in acidic media

Article abstract of DOI:10.1016/j.electacta.2011.02.090

Cathodic dioxygen (O₂) reduction was performed at a modified glassy carbon electrode (GCE) by single-walled carbon nanotubes (SWCNT)/Nafion (NF) film with cobalt (II) tetra (2-amino-phenyl) porphyrin (CoTAPP) and palladium (Pd) nanoparticles incorporated and employed as doping agents. Both the electrochemical behavior of SWCNT with a P(CoTAPP)-Pd nanoparticle matrix and the electrocatalytic reduction of O₂ were investigated using transmission electron microscopy (TEM), cyclic voltammetry (CV) and rotating ring-disk electrode (RRDE) techniques in 0.1 mol l ^-1 H₂SO₄ aqueous solutions. The electrocatalytic reduction of O₂ at the SWCNT/NF/P(CoTAPP)-Pd composite film established a pathway of four-electron transfer reductions into H₂O. Hydrodynamic voltammetry revealed that the modified electrode was catalyzed effectively by the four-electron transferred reduction of dioxygen into H₂O with minimal generation of H₂O₂. The SWCNT/NF/P(CoTAPP)-Pd composite film showed a highly efficient electrocatalytic performance. P(CoTAPP)-Pd was an effective mediator for the reduction of dioxygen and was responsible for the enhanced catalytic activity.

Full text of DOI:10.1016/j.electacta.2011.02.090

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Electrochimica Acta 56 (2011) 4924–4929
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Electrocatalytic reduction of dioxygen at a modified glassy carbon electrode
based on Nafion®-dispersed single-walled carbon nanotubes and
cobalt–porphyrin with palladium nanoparticles in acidic media
Mohammad Shamsuddin Ahmed, Haesang Jeong, Jung-Min You, Seungwon Jeon∗
Department of Chemistry and Institute of Basic Science, Chonnam National University, Gwangju 500-757, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Cathodic dioxygen (O2) reduction was performed at a modified glassy carbon electrode (GCE) by single-
walled carbon nanotubes (SWCNT)/Nafion® (NF) film with cobalt (II) tetra (2-amino-phenyl) porphyrin
(CoTAPP) and palladium (Pd) nanoparticles incorporated and employed as doping agents. Both the
electrochemical behavior of SWCNT with a P(CoTAPP)–Pd nanoparticle matrix and the electrocatalytic
reduction of O2 were investigated using transmission electron microscopy (TEM), cyclic voltammetry
(CV) and rotating ring-disk electrode (RRDE) techniques in 0.1 mol l−1 H2SO4 aqueous solutions. The
electrocatalytic reduction of O2 at the SWCNT/NF/P(CoTAPP)–Pd composite film established a pathway
of four-electron transfer reductions into H2O. Hydrodynamic voltammetry revealed that the modified
electrode was catalyzed effectively by the four-electron transferred reduction of dioxygen into H2O with
minimal generation of H2O2. The SWCNT/NF/P(CoTAPP)–Pd composite film showed a highly efficient
electrocatalytic performance. P(CoTAPP)–Pd was an effective mediator for the reduction of dioxygen and
was responsible for the enhanced catalytic activity.
Received 2 December 2010
Received in revised form 16 February 2011
Accepted 20 February 2011
Available online 1 March 2011
Keywords:
Dioxygen reduction
Electrocatalytic
Cobalt (II) porphyrin
Hydrodynamic voltammetry
Palladium nanoparticles
© 2011 Elsevier Ltd. All rights reserved.
1. Introduction
structure with a highly hydrophobic surface [20], a property that
allows CNT interaction with various compounds through ␲–␲
Over the last few decades, significant effort has been directed
towards the development of low-cost electrocatalysts for oxygen
reduction reactions (ORR). Finding a low-cost electrocatalyst for
ORR is important for practical applications in clean energy gen-
eration systems such as fuel cells and metal-air batteries [1,2].
The determination of O2 is also an important analytical target in
biological areas [3]. Oxygen can be measured using electrochem-
ical methods in order to gain the advantages of high sensitivity
and selectivity. Several studies have examined the electrocatalytic
reduction of O2 at metalloporphyrin-modified electrodes [4–10].
Cobalt [11–13], copper [14] and iron [14,15] porphyrins have been
used as homogeneous catalysts. In the presence of an effective cat-
alyst in aqueous solution, dioxygen can be reduced to H2O2 or
water via two or four-electron transfer. Some complexes applied
as homogeneous catalysts react slowly with O2; most of them
are insoluble in aqueous media. Surface modification of electrodes
with carbon nanotubes (CNTs) has lead to recent developments
in the field of electrocatalysis. Among such developments, detec-
tion of bio-organic and inorganic compounds at the CNT matrices
has been widely reported [16–19]. CNTs exhibit a ␲-conjugative
electronic and hydrophobic interactions [21–23]. CNTs are also
used for the preparation of sandwiched film-modified electrodes
for electrocatalytic study [24,25]. Similarly, electropolymerization
is a simple but powerful method used to target selective mod-
ification of different types of electrodes with desired matrices.
Numerous conjugated polymers (porphyrin) have been electro-
chemically synthesized for fabrication of chemical and biochemical
sensor devices [4–10]. The conjugated polymers used for sensor
devices exhibit interesting enhancements in the electrocatalytic
activity towards oxidation or reduction [20]. Oxygen is reduced
through a complicated series of reactions involving the forma-
tion of a H2O2 intermediate [26]. Analysis of H2O2 formation can
provide additional information regarding the effects on the reac-
tion mechanism. RRDE is used to determine kinetic parameters
for O2 reduction and to detect intermediate formation. It is gen-
erally recognized that NF has a microscopically phase-separated
structure that consists of hydrophilic ionic clusters, a hydropho-
bic backbone and an intermediate region [27–30]. Impregnation
of NF into gas-diffusion electrodes in proton exchange membrane
fuel cells (PEMFCs) should affect the mass-transport, and the elec-
trode kinetics of the fuel cell reactions [31]. Many research groups
have reported that the presence of an NF film does enhance the
kinetic current for O2 reduction at NF-coated Pt electrodes [32–34].
To develop a low cost ORR electrocatalyst, Pd nanoparticles were
∗
Corresponding author. Tel.: +82 62 530 0064; fax: +82 62 530 3389.
E-mail address: swjeon3380@naver.com (S. Jeon).
0013-4686/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2011.02.090

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
M.S. Ahmed et al. / Electrochimica Acta 56 (2011) 4924–4929
4925
preferred over platinum and gold due to the lower price. Palladium
has been reported to present considerable catalytic activity for ORR
in acidic electrolyte solution and preferentially proceeds through a
four-electron pathway, which is important for oxygen sensing [35].
Few research groups have used only Pd nanoparticles in ORR.
In the present work, we have explored the P(CoTAPP)–Pd archi-
tectures whereby the molecules oriented on the surface of the
electrodes by electrochemistry were used in the electroreduction
of O2. The SWCNT/NF-coated GCE was covered with P(CoTAPP)–Pd
using electropolymerization in a 0.1 mol l−1 tetrabutylammonium
perchlorate (TBAP)/acetonitrile (AN) solution and further chemi-
cally processed. In order to provide a stronger understanding of the
potential and currents of SWCNT/NF/P(CoTAPP)–Pd composite film
in the reduction of dioxygen in 0.1 mol l−1 H2SO4 aqueous solutions,
the electrochemical behavior, stability and electrocatalytic activity
towards ORR of the newly synthesized composite electrode were
electrochemically analyzed using CV and RRDE techniques.
Fig. 1. Electropolymerization of CoTAPP in 0.1 mol l−1 TBAP/AN solution at a scan
rate of 100 mV/s.
electrodes for 1 min to desorb chloride ions [31]. The ring collection
efficiency (N = 0.18) was determined using a solution of ferrocene.
Following cleaning, the dispersed solution was prepared by
dispersion of 1.0 mg SWCNT and 0.5 ml NF in 0.5 ml AN to gen-
erate a black solution. For a homogeneous suspension, the solution
was sonicated with ultrasonic agitation for 30 min. To prepare the
SWCNT/NF/GCE-modified electrode, 4× 2 ␮l of the black solution
was cast onto the GCE surface and the solvent evaporated at RT.
After coating, electropolymerization of CoTAPP (2.0 mg CoTAPP in
4.0 ml 0.1 M TBAP/AN solution) was performed between sweeping
potentials +1.5 V to −1.5 V at 0.1 V/s for 15 cycles (Fig. 1). The par-
tially modified electrode was soaked in 0.25 wt% PdCl2 and 0.5 wt%
NaBH4 aqueous solution for 20 min and 60 min, respectively, to
prepare the Pd nanoparticles (Pd0). Finally, electrode was washed
several times with distilled water. Electrochemical measurements
using the SWCNT/NF/P(CoTAPP)–Pd modified electrode were con-
ducted in Ar and dioxygen-saturated 0.1 M H2SO4 solution for the
electrocatalytic reduction of O2.
2. Experimental
2.1. Instruments
Voltammetric measurements were accomplished with a three-
electrode potentiostat [CHI 700C electrochemical workstation
(USA)] in a grounded Faraday cage. A platinum-wire electrode sep-
arated from the analyte compartment by a medium porosity glass
frit was used as an auxiliary electrode. A calibrated Ag/AgCl elec-
trode [3 M NaCl solution] supplied by Bioanalytical Systems Inc.
(BAS) was used as a reference electrode, with a potential of approx-
imately 45 mV relative to a saturated calomel electrode (SCE). A
GCE (3 mm in diameter) was employed as a working electrode
after modification with a composite film. An EG&GPARC Model 636
RRDE system and a CHI 700C electrochemical workstation bipo-
tentiostat were used for hydrodynamic voltammetry experiments.
A rotating GC disk (4.3 mm in diameter)-platinum ring electrode
was used as a working electrode. Electrochemical impedance spec-
troscopy (EIS) was performed with a Versa State 3, manufactured
by Metek, USA. Transmission electron microscopy (TEM) images
were taken by a TECNAI model FI-20 (FEI, Netherland) in ethanol
after 20 min ultrasonication. The dioxygen concentration in the
dioxygen-saturated solution was 1.2 mM (similar to Ref. [13]). All
potentials were recorded with respect to the Ag/AgCl electrode at
room temperature (RT).
3. Results and discussion
3.1. TEM studies of SWCNT/NF/P(CoTAPP)–Pd film
TEM is a powerful instrument that was used to observe
the surface of the nanoparticles. The morphology of the
SWCNT/NF/P(CoTAPP)–Pd composite film was investigated by
TEM. Fig. 2 displays the TEM micrographs of the composite film.
Fig. 2a demonstrates that Pd nanoparticles were set successfully
onto an electropolymerized SWCNT/NF film; large Pd nanoparticles
were found on the surface of SWCNT/NF/P(CoTAPP). Approxi-
mately 5–10 nm nanoparticles were cast on the modified electrode
(Fig. 2b).
2.2. Chemicals
CoTAPP was synthesized as described in the literature [36]. The
SWCNT (1.2–1.5 nm in diameter produced by the arc method) was
purchased from Aldrich (Korea) and purified with 6 M HCl solu-
tion. NF (5 wt% in lower aliphatic alcohols and water), PdCl2 and
NaBH4 were also purchased from Aldrich. All other reagents used
were of analytical grade. High purity argon was used for deaera-
tion. All experiments were carried out at RT. Doubly distilled water
with resistibility over 18 Mꢀ cm in a quartz apparatus was used to
prepare all aqueous electrolyte solutions. The 0.1 M H2SO4 (Fischer
Scientific) solution was used as the supporting electrolyte.
3.2. EIS measurement
EIS is an important experiment that provides information about
impedance of the electrode. It is well known that the Nyquist plot
has two regions: the first is a semicircular portion and the second
is the linear portion. The semicircular portion at higher frequen-
cies corresponds to the electron-transfer limited process which
controls the electron transfer kinetics of the redox probe at the
electrode interface. The linear portion at lower frequencies corre-
sponds to the diffusion process. As shown in Fig. 3, the Nyquist
plots display the impedance spectroscopy of (i) bare GCE, (ii)
SWCNT/NF/PCoTAPP–Pd/GCE and (iii) SWCNT/NF/PCoTAPP/GCE in
0.1 mol l−1 KCl containing 5.0 mM K3Fe(CN)6/K4Fe(CN)6 (1:1) by
applied frequency from 105 Hz to 10−2 Hz. The bare GCE showed
a large semicircle up to 2.8 kꢀ, indicating that the bare GCE was
2.3. Preparation of working electrodes
An RRDE containing a GC disk and Pt ring sealed in a polytetraflu-
oroethylene holder was polished with 0.05 ␮m alumina suspension
on a polishing cloth (BAS, USA) and cleaned with ultrasonication in
distilled water. After the surface was polished to a mirrored surface,
it was electropolished in 10% HCl solution at 1 mA cm−2 for 30 min
to remove any Pt contamination. The solution was replaced with
0.1 mol l−1 H2SO4 and hydrogen was evolved at both ring and disk

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
4926
M.S. Ahmed et al. / Electrochimica Acta 56 (2011) 4924–4929
Fig. 2. TEM images of SWCNT/NF/P(CoTAPP)–Pd.
an obstacle and made the electron transfer at the interface more
difficult when compared to other modified GCEs. A very small semi-
circle portion for both (ii) SWCNT/NF/PCoTAPP–Pd/GCE and (iii)
SWCNT/NF/PCoTAPP/GCE, and they showed small resistances of
150 ꢀ and 223 ꢀ, respectively. The results indicated that both mod-
ified GCE electrodes were highly conductive and made the electron
transfer at the interface easier than with a bare GCE.
follows [13,38]:
ꢀ
ꢁ
1/2
˛n˛F
RT
ip = 0.4958nFAC
1.15RT
v1/2D1/2
(1)
(2)
ꢁEp
=
˛n˛F
where n is the number of transferred electrons, A is the surface
area of the GCE (surface area of 0.070686 cm2 was employed) and
D and C are the oxygen diffusion coefficient and the bulk concen-
tration, respectively. The value of the oxygen diffusion coefficient
was 1.9E−5 cm2/s, ꢂ is the scan rate in V/s, ˛ is the transfer coef-
ficient, n˛ is the apparent number of electrons transferred in the
rate-determining step and ꢁEp is the peak potential change when
the scan rate increases 10-fold. Other symbols mentioned used
their standard values. The number of transferred electrons was cal-
culated as 3.4 for the reduction of O2 using Eq. (1). No anodic peak
was present, indicating that the catalytic reduction reaction process
was irreversible in the experiment.
3.3. Electrocatalytic O2 reduction on
SWCNT/NF/P(CoTAPP)–Pd/GCE
The SWCNT/NF/P(CoTAPP)–Pd/GCE electrode was employed for
the reduction of molecular oxygen in this study via CV. The
peak potential of the dioxygen reduction wave appeared for
the SWCNT/NF/P(CoTAPP)–Pd-coated electrodes. Typical CVs were
taken in Ar and oxygen saturated 0.1 mol l−1 H2SO4 solutions,
and all had potentials ranging from 0.6 V to −0.1 V at scan rate
of 25 mV/s, shown in Fig. 4. As expected, there was a negligible
peak potential for Ar saturated CV. A significant and clear peak
potential appeared at 0.21 V for the O2 saturated acid solution and
had a current response of 41 ␮A (much higher than Refs. [11,13]).
The reduction reaction started from 0.5 V, while the peak currents
increased linearly (correlation coefficient 0.997) with the square
root of the scan rates (Fig. 4 inset). The study suggested that the
electrocatalytic processes were controlled by molecular oxygen dif-
fusion in a modified electrode [37]. In a typical irreversible reaction,
the relationship between the peak current and scan rate was as
3.4. Hydrodynamic voltammetry
To analyze experimental results in this study, we used a sim-
plified model [39] shown in Scheme 1. Path 1 shows how O2 is
reduced directly to H2O through a four-electron transfer. Path 2
is the sequential reaction path where O2 is first reduced to H2O2
through a two-electron transfer followed by a two-electron reduc-
tion to H2O (Path 3), or by the release of the formed H2O2 into bulk
solution (Path 4).
Fig. 4. CVs of SWCNT/NF/PCoTAPP–Pd electrode in Ar- and O2-saturated 0.1 mol l−1
H2SO4 solution at 25 mV/s; inset: plot of peak current versus square root of scan
rate.
Fig. 3. Nyquist plots of (i) bare GCE, (ii) SWCNT/NF/PCoTAPP–Pd and (iii)
SWCNT/NF/PCoTAPP.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
M.S. Ahmed et al. / Electrochimica Acta 56 (2011) 4924–4929
4927
1
2
3
O2 (b)
O2 (a)
H2O2 (a)
4
H2O
H2O2 (b)
Scheme 1. A model for ORR (a) adsorbed on electrode surface and (b) in bulk solu-
tion.
The RRDE experiments of hydrodynamic voltammetry
were useful in confirming the electrocatalytic reduction
pathway of O2. The electrocatalytic O2 reduction activity deter-
mined by the RRDE method using SWCNT/NF/P(CoTAPP) and
SWCNT/NF/P(CoTAPP)–Pd-modified electrodes were investigated
in 0.1 mol l−1 H2SO4 solution under O2 saturated conditions at
a rotation speed of 25–2500 rpm. The disk potential employed
was from +0.5 V to −0.5 V at 20 mV/s, and the ring potential was
constant at 1.2 V (Fig. 5). In Fig. 5, the RRDE voltammograms were
obtained from the SWCNT/NF/P(CoTAPP)-modified electrodes;
the O2 reduction started at approximately 0.3 V and the current
response was significantly high. The Koutecky–Levich equation
[40] was used for the calculation of the n value as follows:
Fig. 6. Koutecky–Levich plots for ORR catalyzed by SWCNT/NF/P(CoTAPP) at −0.1,
−0.2 and −0.3 V applied potentials.
ilim = 0.62nFAD2/3Cω1/2v−1/6
(3)
where ilim is the limiting current (␮A) on the plateau, ω is
angular velocity (rad/s), v is the kinematic viscosity (cm2/s), the
kinematic viscosity was 0.01 cm2/s and n is the electron num-
ber. The Koutecky–Levich plots for the RRDE voltammograms
of SWCNT/NF/P(CoTAPP) obtained at various rpm are shown in
Fig. 6. The solid lines were obtained from the plateau current.
The dependence of transferred electrons on the cathodic poten-
tial was calculated from Eq. (3). According to the n value of the
reaction, two electrons were involved, which indicated that the pri-
mary product was H2O2. Nevertheless, the RRDE voltammograms
(in +0.6 V to −0.2 V at 20 mV/s, ring potential of 1.2 V) exhibited by
the SWCNT/NF/P(CoTAPP)–Pd-modified electrode (Fig. 7) showed
that O2 reduction started from a more positive potential (approx-
imately 0.5 V) and the current response was higher than without
Pd nanoparticle modified electrodes (comparable to Ref. [11]). The
Fig. 7. RRDE voltammograms recorded from SWCNT/NF/P(CoTAPP)–Pd at 25, 100,
400, 900, 1600 and 2500 rpm; ring potential remained constant at 1.2 V; scan rate
of 20 mV/s.
initial potentials were used in order to determine the potential at
which the electrocatalytic activity began, which defined the poten-
tials and the O2 reduction wave departures from the baseline in
both cases. The Koutecky–Levich plots are shown in Fig. 8 (data
obtained from Fig. 7). The n values were also calculated by Eq. (3).
The two dashed lines were calculated theoretically for two and four
electrons in both Figs. 6 and 8. According to the n value, four elec-
trons were involved, which indicated the main product to be H2O.
Fig. 5. RRDE voltammograms obtained from SWCNT/NF/P(CoTAPP) at 25, 100, 400,
900, 1600 and 2500 rpm; ring potential was constant at 1.2 V; scan rate of 20 mV/s.
Fig. 8. Koutecky–Levich plots for ORR catalyzed by SWCNT/NF/P(CoTAPP)–Pd at 0.0,
−0.1 and −0.2 V applied potentials.