- Email: [email protected]
Electrocatalytical study of carbon supported Pt, Ru and bimetallic Pt–Ru nanoparticles for oxygen reduction reaction in alkaline media
Accepted Manuscript Title: Electrocatalytical study of carbon supported Pt, Ru and bimetallic Pt-Ru nanoparticles for oxygen reduction reaction in alkaline media Author: M.G. Hosseini P. Zardari PII: DOI: Reference:
S0169-4332(15)00742-4 http://dx.doi.org/doi:10.1016/j.apsusc.2015.03.146 APSUSC 30023
To appear in:
APSUSC
Received date: Revised date: Accepted date:
17-7-2014 23-3-2015 23-3-2015
Please cite this article as: M.G. Hosseini, P. Zardari, Electrocatalytical study of carbon supported Pt, Ru and bimetallic Pt-Ru nanoparticles for oxygen reduction reaction in alkaline media, Applied Surface Science (2015), http://dx.doi.org/10.1016/j.apsusc.2015.03.146 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Highlights Binary catalyst Pt.Ru/C is evaluated towards ORR. Pt.Ru/C nanoparticles revealed best ORR catalytical activity. The 120 mV/dec Tafel slop indicated that the first electron transfer is the rds.
Ac ce p
te
d
M
an
us
cr
ip t
The active number sites of Pt.Ru/C catalyst were 3 times higher than Pt/C.
1 Page 1 of 38
Electrocatalytical study of carbon supported Pt, Ru and bimetallic Pt-Ru nanoparticles for oxygen reduction reaction in alkaline media
ip t
M.G. Hosseini* a, P. Zardaria a) Department of Physical Chemistry, Electrochemistry Research Laboratory, University of
cr
Tabriz, Tabriz, Iran
us
Abstract
Carbon supported Pt, Ru and bimetallic Pt-Ru nanoparticles (Pt/C, Ru/C and Pt.Ru/C) have
an
been prepared by the chemical reduction method. Particle morphology, composition and structure of nanoparticles have been investigated by scanning electron microscopy (SEM) and
M
energy dispersive X-ray (EDX) analysis. SEM results showed a uniform dispersion of nanoparticles with rough and porous structure into carbon supports with the average particle
d
size of 30-64 nm. EDX analysis demonstrated the presence of both Pt and Ru nanoparticles in
te
each gas diffusion electrode. The Pt/C, Ru/C and Pt.Ru/C composites were used as
Ac ce p
electrocatalyst for oxygen reduction reaction (ORR) in alkaline media. The ORR activities of cathodes were characterized using cyclic voltammetry (CV), polarization technique, AC impedance spectroscopy (EIS) and chronoamperometry. CV and polarization curves showed significantly higher activity on Pt.Ru/C electrocatalyst than observed on Pt/C and Ru/C catalysts, which can be related to synergistic effect, which is playing a critical role in ORR activity. The Tafel slope values of 120 mV/dec showed that the first electron transfer is the rate determining step. The EIS results of cathodes under different polarization potentials indicated two different behaviours which depend on the applied dc potentials and reveals
*
Corresponding Author. Tel.: +984113393138; fax: +984113340191.
E-mail address: [email protected] .
2 Page 2 of 38
different electrochemical processes occurring on the electrodes. Keywords: ORR, ODC, Pt electrocatalyst, Pt-Ru electrocatalyst, EIS, CV
1. Introduction
ip t
Nowadays, the chlor-alkali industry is one of the largest chemical processes worldwide. Its two main components - chlorine and caustic soda - are indispensable commodities that are
cr
used for a wide range of applications. Nearly 55 percent of all specialty chemical products
us
manufactured require one of the chlor-alkali products as a precursor [1, 2]. Chlorine and caustic soda, obtained by the electrolysis of sodium chloride solution by three methods
an
included the mercury, diaphragm and ion exchange membrane processes [3]. The brine electrolysis system has been changed from the mercury and the diaphragm process to the ion
M
exchange membrane process in the most advanced countries because of the attempting to reduce the manufacturing costs of chlorine production [4-7]. In the ion exchange membrane
d
process, the total cell reaction for the electrolysis of brine into chlorine and caustic soda
te
proceeds through the following half-cell reactions and the decomposition voltage of the total
Ac ce p
cell reaction is about -2.2 V [8].
Anode reaction: 2Cl- → Cl2 + 2e-
E=+1.358 V/NHE
Cathode reaction: 2H2O + 2e- → 2OH- + H2
E=-0.828 V/NHE
(1) (2)
The electrolysis of brine through the ion exchange membrane process and by hydrogen evolution reaction (HER) cathodes is still one of the most energy-intensive industrial operations and it is necessary to reduce energy consumption. The application of an oxygen depolarized cathode (ODC) by using a gas diffusion electrode (GDE) instead of HER cathode can solve this problem. By replacing the ODC instead of HER cathodes, the half cell reactions can be written as follows: Anode reaction: 2Cl- → Cl2 + 2e-
E=+1.358 V/NHE
(3) 3 Page 3 of 38
Cathode reaction: 2H2O + O2 +4e- → 4OH-
E=+0.401 V/NHE
(4)
Nevertheless, by replacing the HER cathodes in a membrane cell with an ODC, the cell voltage and energy consumption can be reduced by as much as 30-40% at 0.4 A/cm2 [9, 10].
brine in the chlor-alkali cells that have to meet several criteria such as: a) Long term stability in sodium hydroxide media at 80-90 oC.
cr
b) High catalytic activity for ORR.
ip t
Oxygen depolarized cathodes consist of a gas diffusion electrode (GDE) for electrolysis of
us
c) Suitable hydrophobic/hydrophilic pore structure for transport of oxygen and electrolyte to the reaction sites [11].
an
The GDE consists of a reaction layer, gas diffusion layer and current distributor. The reaction layer structure is made of carbon black, Polytetrafluoroethylene (PTFE), and oxygen
M
reduction active catalyst. The gas diffusion layer is made of carbon black and PTFE and facilitates the reactant permeation from the electrolyte to the catalytic sites. The Ni, Ag mesh
d
or highly pyrolytic graphite is used as a current distributor in GDE [12, 13]. In the last
te
decades, Pt catalyst has attracted a great deal of studying due to their promising applications,
Ac ce p
particularly as the electrode catalyst for oxygen reduction in GDEs and considerable progress has been achieved [14-16].
Despite the fact that the Pt catalyst has excellent electrocatalytic activities for oxygen reduction, some obstruction still remain for the commercialization of ODC cells, such as the high cost of Pt. Consequently, the major issues facing ODC cells are the cost down of electrode and finding a new selective oxygen reduction catalyst on ODC electrode. So, many research groups have attempted to find new low cost electrocatalysts that selectively catalyse oxygen reduction reactions [16]. Recent studies [17-22] have examined the use of bimetallic Pt alloys as oxygen reduction catalysts. The goal of these studies is to improve the catalytic activity of platinum, by alloying it with transition metals like Pd, Ru, Ni, for the ORR at cathode sites of GDEs. The alloying of transition metals exhibits significantly higher 4 Page 4 of 38
electrocatalytic activities towards the ORR than platinum alone [23]. This enhancement effect has been explained by models such as the “bifunctional mechanism” [24] or by the “electronic effect” [25], which indicates a promotional effect of the alloyed metal on Pt. Particularly, Pt-Ru has been investigated as a binary system, and has shown the appropriate
ip t
catalytic activities in many different electrochemical reactions [26]. Several studies, carried out on the polycrystalline (PC) Ru, monocrystalline Ru, Pt submonolayers on Ru, and
cr
Pt.Ru/C composite electrodes, have shown their considerable activity for ORR in acidic and
us
alkaline media [27]. The Volcano plot of ORR by different Pt group metals shows that the oxygen is chemisorbed on Ru sites more intensively than Pt sites and the further desorption of
an
oxygen on Ru sites is more difficult than Pt. On the other hand, the molar ratio of Pt:Ru in electrocatalyst layer is very important. If the bimetallic catalyst is prepared by the high molar
M
ratio of Ru, the oxygen will be adsorbed more strongly on the alloy and the future desorption
alloy will be reduced [28].
d
of oxygen molecules will be more difficult, so the electrocatalytical activity of bimetallic
te
The main aim of this work is to accomplish a method to prepare bimetallic Pt.Ru/C
Ac ce p
nanoparticle electrocatalyst which has the optimum amount of Ru sits according to Volcano plots. On the other hand, the amount of Pt will be chosen as minimum as possible due to economic problems of Pt. In order to investigate the electrochemical activity of binary Pt-Ru alloys in GDEs, the Pt, Ru and Pt-Ru bimetallic catalysts were supported on carbon black, used in the catalyst layer of GDEs, and characterized by various electrochemical techniques to study their ORR activity.
2. Material and methods 2.1. Chemicals used Sodium borohydride (NaBH4, 96%), sodium hydroxide (NaOH, 99%) and 2-propanol ((CH3)2CHOH, 99.99%) were received from Merck. Ruthenium trichloride anhydrous
5 Page 5 of 38
(RuCl3·3H2O, 99%) and chloroplatinic acid hydrate (H2PtCl6.xH2O, 99.99%), were purchased from Cole-Parmer and Sigma-Aldrich, respectively. Polytetrafluoroethylene (PTFE, 99%), Vulcan carbon (XC-72R), nafion solution (5wt %) and
TP-060T, respectively. All chemicals used were of analytical grade. 2.2. Synthesis of catalyst
ip t
Teflon treated carbon paper were obtained from Sigma-Aldrich, Cobat, Dupont, Toray EC-
cr
Appropriate amount of Vulcan carbon (XC-72R) was ultrasonically dispersed in the mixture
us
of 2-propanol and de-ionized water for 2 h. Metal precursors (H2PtCl6.xH2O + RuCl3.3H2O) were added to the resulted Vulcan carbon ink. The molar ratio of Pt to Ru was achieved 1:1
an
and the total metal content in the elctrocatalyst was 40 wt% vs. carbon support. The 1 M NaOH solution was added to adjust the pH of ink to 7-8. The mixture was heated up to 80 oC
M
and 10 mL NaBH4 solution was added drop-wise to the suspension under vigorous stirring condition at 80 oC. The solid particles were separated from the solution by filtration, then
d
washed and dried at 120 oC for 6 h. For comparison, 40 wt% Pt/C and 40 wt% Ru/C were
te
prepared by the same method too.
Ac ce p
2.3. Preparation of gas diffusion layer Vulcan carbon (XC-72R) and PTFE solution was ultrasonically dispersed into the mixture of 2-propanol, water and glycerol for 1 h. The PTFE content to Vulcan carbon (XC-72R) was 60 wt%. Then resulted suspension was painted onto the carbon paper (geometric exposed area of 1 cm2). The painted layer was dried at 80 oC and sintered at 350 oC for 30 min. The loading of the gas diffusion layer was obtained equal to 1 mg.cm-2. 2.4. Preparation of gas diffusion electrode The catalyst inks were prepared by mixing the appropriate amount of catalysts with 2propanole, de-ionized water and 5wt% nafion solution. The loading of the nafion and catalysts on the GDLs were about 1 mg.cm-2. After the painting of catalyst inks on the GDLs, the final electrodes were dried at 120 oC for 1 h. The surface morphology of the GDEs was studied 6 Page 6 of 38
with a scanning electron microscope (Philips, Model XL30) and energy dispersive X-ray (EDX) analysis. 2.5. Electrochemical measurements The GDEs were mounted in a Teflon holder containing a high pyrolytic graphite disk as a
ip t
current collector which had arranged for oxygen feed from the back of the electrode. The electrochemical measurements were carried out using three-electrode compartment (Fig. 1).
cr
An Ag/AgCl electrode was used as the reference electrode and platinum as the counter
us
electrode.
Electrochemical experiments were done using the Potentiostat/Galvanostat (EG&G model
an
Parstat 2263) equipped with Frequency Response Analyzer (FRA) controlled by a PC through PowerSuite software in the 0.1 M NaOH solution, at 25 ◦C. Cyclic voltammetry was applying
M
a scan rate of 50 mV.s-1 between -0.6 and +1.1 V vs. Ag/AgCl in the O2 saturated NaOH solution. EIS measurements were performed over a frequency range of 100 KHz to 0.01 Hz,
d
using ±5 mV amplitude of a sinusoidal voltage as perturbing signal. The samples were
te
potentiodynamically polarized to the specified potentials. Experimental data were fitted and
Ac ce p
analysed by ZView (II) software. Tafel polarization measurements were done in the potential range of 1 to -0.2 V vs. Ag/AgCl by applying a scan rate of 1 mV.s-1 and the Tafel slopes and exchange current densities were obtained. Chronoamperometric determinations were achieved by holding the potential of the electrodes at +1.2 V vs. Ag/AgCl for 10 s and then stepping the potential to a value of +0.4 V for 4 h. 3. Results and discussion
3.1. SEM and EDX analyses The SEM graphs and EDX analysis of GEDs are shown in Fig. 2. The surface of GDEs revealed rough and porous structures with high and quite uniform dispersion of Pt and Ru nanoparticles onto/into Vulcan carbon supports with average diameter about 57, 64 and 43 nm for Pt/C, Ru/C and Pt.Ru/C catalysts, respectively. The EDX analysis showed presence of 7 Page 7 of 38
35.47 and 24.38 wt.% Pt in Pt/C and Pt.Ru/C nanoparticles, respectively. The obtained weigh percent of Ru in Ru/C and Pt.Ru/C electrocatalyst was around 43.21 and 23.82 which is in good agreement with nominal values of catalysts (Table 1). 3.2. Cyclic Voltammetry
ip t
Fig. 3 shows the cyclic voltammograms recorded for the Pt/C, Ru/C and Pt.Ru/C GDEs in aqueous 0.1 M NaOH saturated with O2 (dash line) or N2 (solid line), at a scan rate of 50
cr
mV.s−1. All the CV curves recorded under N2-atmosphere, show two distinguish peaks (I, II).
us
While, the CV curves in the O2-saturated electrolyte, show three characteristic regions, namely, the hydrogen region, the double layer charging and the surface oxide
an
formation/reduction. In other words, an additional reductive current is visible, that is absent in the case of saturation with N2. On the anodic scan of O2-saturated electrolyte, the peaks at
M
about -0.3 to 0 V (peak I) are assigned to the oxidation of hydrogen absorbed on the metallic active sites. Peak (I) is clearly intense in the case of Pt/C and Pt.Ru/C catalysts because of the
d
more catalytical active particles in these samples (greater hydrogen adsorption).
It is
te
generally believed that it is very difficult to separate the regions of hydrogen
Ac ce p
adsorption/adsorption and double layer regions for Ru particles. So, the peak (I) isn’t so identifiable in the Ru/C catalyst CV. The double layer region of catalysts is observed in the potential region of 0.0 to 0.7 V. On the anodic sweep and above 0.7 V potential, the formation of oxide film is done on the surface of metallic electrocatalyst (peak II). The onset potential of peak (III) for Pt.Ru/C, Pt/C and Ru/C catalysts are observed about 0.75, 0.68 and 0.1V, respectively. As discussed in the introduction part, by replacing the ODC instead of HER cathode, the activation energy of water reduction process will be reduced and the water reduction peak will be appear in the more positive potential. By the comparison of CV plots for N2 and O2 saturated electrolyte, it can be conclude that the water reduction process started in the more positive potential in the present of O2 molecules. For example, the water reduction potential of Pt/C GDE in the O2 saturated electrolyte is started 1 V more positive than N2 8 Page 8 of 38
saturated one. From the application standpoint, for an efficient operation of an ODC, the current intensity of the ORR induced for the same O2 concentration for all catalysts (comparative study for the saturated O2 solution) has to be as high as possible because high current reflects a fast reaction [29]. Thus, onset potential, current intensity and sharpness of
ip t
the peak (III) in Fig. 3 are the main criteria used in this study to identify the best catalyst for the ORR. The advantage of this method is the simplicity and the accuracy at the condition that
cr
the experiments are well done in a comparative manner, which the case of the present study.
us
The onset potentials of ORR for used catalysts are listed in Table 2. The onset potential of ORR with Pt/C and Pt.Ru/C are located in the same region, while the onset potential of ORR
an
with Ru/C is shifted by 600 mV towards more negative potentials. It can be concluded that the ORR with Pt.Ru/C catalyst is energetically more favourable [30]. Comparison of cyclic
M
voltamograms of Pt/C, Ru/C and Pt.Ru/C electrodes in O2 saturated solutions show that for Pt.Ru/C catalyst, ORR current intensity increase within the ORR region. It can be concluded
d
that the second metal addition has influenced the catalytic activity of electrocatalyst toward
te
ORR. This matter can be related to synergistic effect, which is playing a critical role in ORR
Ac ce p
activity. Despite the hydrogen peaks, the surface areas of catalysts (EASA) were determined from the charge of the monolayer chemisorbed oxygen on the surface, because their adsorption/desorption charge could have some contribution from hydrogen absorption [31]. The EASA values and active site number (NA) of catalysts are also listed in Table 2. The EASA value of the Pt.Ru/C catalyst was approximately 2.75 and 3.6 times higher than that of Pt/C and Ru/C catalysts. So, with increasing of available active surface area, catalytical activity of electrocatalyst towards ORR has been raised. As mentioned in the introduction part, the Ru and Pt alloying in the optimum amount of Volcano plots can increase the ORR activity of catalyst. The CV results lead us to conclude that the bimetallic nanoelectrocatalyst acts more powerful than single Pt/C as ORR catalyst although the Pt loading amount in Pt.Ru/C is half of Pt/C one. This matter can be related to the synergistic effect, which is 9 Page 9 of 38
playing a critical role in ORR. The synergistic effect of the simultaneous presence of Pt and Ru in the electocatalyst can be explained by the spill-over effect of hydrogen and oxygen from Pt surface to Ru and the resultant freedom for Pt active sites [32]. The electroactive surface area of synthesized Pt/C electrodes according to hydrogen adsorption/desorption peak
(10 wt.%, E-TEK) commercial catalyst (32.4 m2.g-1) [33].
cr
3.3. Polarization measurement
ip t
was also estimated equal to153 m2.g-1 which is in a good agreement value of commercial Pt/C
us
Figure 4 shows polarization curves for three electrodes in a conventional three-electrode cell. Polarization curves were produced at a scan rate of 1 mV.s-1 between 1 and -0.2 V vs.
extracted by using the Tafel equation [34]:
M
= (2.303RT/αnF) log(i/io)
an
Ag/AgCl in the O2 saturated NaOH solution. The kinetic parameters of the ORR were
(5)
Where = (E−E0) is the overpotential, R is the gas constant, T is the absolute temperature, α
d
is the transfer coefficient, i0 is the exchange current density, i is the current density, n is the
te
number of electrons, and F is the Faraday’s constant.
Ac ce p
The current density of Pt.Ru/C cathode is higher than Pt/C and Ru/C electrodes according to Fig 4. The kinetic parameters of electrodes are reported in Table 2. The results showed that the ORR current density in Pt.Ru/C GDE is higher than in Pt/C and Ru/C GDEs and the Pt.Ru/C cathode shows lower Tafel slope, which indicated that the Pt.Ru/C electrode has a much better performance in ORR. It can be as a result of weak O-O bond on Pt-modified Ru nanoparticles [35], the synergy effect between Pt and Ru nanoparticles and uniform dispersion of Ru and Pt on the carbon support. The lower Tafel slope and higher current density of Pt.Ru/C cathode can be attributed to the electrocatalytical activity enhancement of Pt.Ru/C electrocatalyst in relative to Pt/C and Ru/C electrocatalysts. The Tafel slopes of ≥120 mV per decade were accounted for all GDEs.
10 Page 10 of 38
There are two different ORR pathways in the alkaline media, i.e. inner-sphere mechanism which occurs in the inner Helmholtz plane (IHP) and the outer-sphere mechanism in outer Helmholtz plane (OHP). In alkaline media, water molecules act as solvent and also proton sources required in the ORR. The loci of species that chemisorbed on the electrode surface
ip t
like hydroxyl species, solvent water dipoles and chemisorbed O2 constitute the IHP. These species populating the IHP covalently interact with the electrode surface. On the other hand,
cr
the OHP consists of solvated species interacting with the electrode via long range electrostatic
us
forces. For example, alkali metal ions are typically solvated and populate OHP. In the case of inner-sphere electron transfer mechanism of ORR, the strong chemisorbed O2 molecules (with
an
or without bond breaking) reduced on oxide-free metal active sites by 4e−/4H+ transfer. The outer-sphere electron transfer mechanism occurs only at the oxide-covered metal sites. The
M
solvated molecular O2 clusters interact with adsorbed hydroxyl species and promote the 2e-
oxide-covered metal sites [36].
d
reaction pathway. Thus, peroxide intermediate (HO2-) is produced as the final product at the
te
The Tafel slope value of ≥120 mV/dec shows that the first electron transfer is the rate
Ac ce p
determining step, according to the inner-sphere electron transfer mechanism. This mechanism is written as follows:
M + O2 + e- → [MOO]-
(6)
[MOO-] + H2O → MOOH + OH-
(7)
MOOH + e- → [MOOH]-
(8)
[MOOH]- + H2O + e- → MOH + 2OH-
(9)
MOH + e- → M + OH-
(10)
Tafel slope higher than 120 mV/dec is often observed in the reactions through some adsorbed layers at the surface [37]. 3.4. Electrochemical impedance spectroscopy
11 Page 11 of 38
An EIS study was carried out at different dc potentials in order to study the mechanism of ORR in O2 saturated alkaline solutions on prepared GDEs. The Nyquist plots of GDEs under different polarization potentials are shown in Fig. 5. The impedance diagrams show two different behaviours which depend on the applied dc
different electrochemical processes occurring on the electrode.
ip t
potentials. In the other words, the shape of plots changes at different potentials, suggesting
cr
The Nyquist plots of O2 reduction on GDEs at the E≥0.7 show two loops, while the
us
impedance spectra acquired at the potential lower than 0.7 V indicate one loop in the high frequency region associated with the time constant of a charge transfer process and semi-
an
infinite diffusive manner related to Warburg component in the low frequency region. This semi-infinite diffusive character is related to the adsorption of reactants and intermediate
M
products. The processes which could be involved on the electrode surface that would produce these changes include [38]:
d
(1) Diffusion of O2 through the gas phase in the pores (of porous carbon supported catalyst)
te
and the electrolyte to the reaction site.
Ac ce p
(2) Adsorption or heterogeneous surface reaction of the oxygen, together with oxygen diffusion.
(3) Charge transfer.
(4) Diffusion of reduction products into the bulk electrolyte ORR is a multi-electron reaction that includes a number of elementary steps involving different reaction intermediates. Many reaction mechanisms have been proposed to describe ORR in aqueous electrolytes. Damjanovic model describes the ORR as a multi-electron reaction which O2 molecules in the vicinity of the electrode are irreversibly reduced directly to H2O through 4-electron transfer or to H2O2 through 2-electron transfer. The H2O2 formed can be reduced to H2O through 2-eltectron transfer or diffuse into the bulk solution. The H2O2 intermediate can be observed when there is no splitting of the O-O bond before H2O2 species 12 Page 12 of 38
is formed. Thermodynamically speaking, the reduction potential of O2 into H2O2 is lower than the potential required forming H2O molecule following 4 electron pathway (E0=0.69 and 1.23 V vs. NHE, respectively). On the other hand, if the produced H2O2 is further transformed into H2O, the overall reaction to reduce O2 into H2O passing through H2O2 will require more
ip t
energy (E0 =1.76 V vs. NHE) [30].
The ORR mechanism on transition metals has also been investigated by theoretical
cr
calculation based on the electronic structure [39, 40]. According to these studies, two main
us
reaction mechanisms are suggested: The dissociative and associative mechanisms. The dissociative mechanism is proposed for a low current density range (more positive potentials),
an
while the associative mechanisms happens in high current density range (more negative
Dissociative Mechanism (E ≥ 0.7 V):
M
potentials) [28]. The brief explanation of these mechanisms is discussed here.
In this mechanism, no H2O2 is produced. On a metal surface, O2 adsorption breaks the O-O
d
bond and forms adsorbed atomic O, which further gains two electrons in the two consecutive
te
steps, forming hydroxide ions. Since there is no adsorbed O2 on the catalyst surface, H2O2
Ac ce p
cannot be formed. This mechanism can be considered a detailed form of the direct 4-electron pathway and can be written as follows: 1/2O2 + M (metal active sites) → Oºads
(11)
Oºads + e- + H2O → OHads + OH-aq
(12)
OHads + e- → OH-ads → OH-aq
(13)
Associative Mechanism (E ˂ 0.7 V): Since adsorbed O2 is present, the O-O bond may not be broken in the following steps, resulting in the formation of H2O2. The H2O2 could either be further reduced to H2O or be a final product. Therefore, the mechanism can be written as follows: O2,ads + 2H2O + 2e- → H2O2 + 2OH-
(14)
H2O2 + 2e- → 2OH-
(15) 13 Page 13 of 38
The further reduction of H2O2(ads) to hydroxide ions occurs only once the enough overpotential has been reached and before the formed H2O2 diffuse into the bulk solution as mentioned in cyclic voltammetry studies [31, 28]. The two loop manner of GDEs in Nyquist plots may be related to two basic steps. On the
ip t
other hand, for E ≥ 0.7 V, two time constants are detected during the impedance measurements (Fig. 5). The first time constant at high frequencies is associated with the
us
further reduction of OHads to produce OH- based Eq. 13.
cr
charge transfer reaction according to Eq. 12, while the second may be associated with the
On the other hand, for E ˂ 0.7 V, the first time constant is related to H2O2 intermediate
an
formation according to Eq. 14 and further semi-infinitive diffusive manner in the low frequency region can be explained by adsorption and diffusion of this intermediate into the
M
bulk solution. In order to obtain quantitative information from impedance spectra in Fig. 5, two electric circuits were employed (Fig. 6). The electric circuit in Fig. 6a was used to
d
simulate the impedance response of those spectra with OCP and E < 0.7 V, while the electric
te
circuit in Fig. 6b was used for those with E ≥ 0.7 V.
Ac ce p
Using the equivalent circuits shown in Fig. 6, a constant phase element (CPE) is suggested instead of pure capacitance (C), due to the non-homogeneous surface of the electrodes. The impedance of CPE is defined as [Yo(jω)n]-1, where Yo has (Ω-1.sn.cm-2) dimension, while the exponent n denotes the correction factor pertaining to the roughness of the electrode and has values that range from 0 to 1. A pure capacitance yields n=1, a pure resistance yields n=0, while n=0.5 represents the ideal Warburg impedance. The true capacitance values can be calculated using the following equation [41]: C=[Yo × R (1-n)] 1/n
(14)
According to equivalent circuits shown in Fig. 6, Rs is associated with the resistance of the solution, connectors, leads and wires. R1 is the charge transfer resistance of the reduction process from O2 to H2O2, R2 is the resistance of adsorbed species with H2O2 as the main 14 Page 14 of 38
intermediate or O2 adsorption into the GDE pores in the figure 6.a. The R1 and R2 circuit components in figure 6b are the charge transfer resistance of the reduction process of Oºads to OHads and the reduction process of OHads to OH-, respectively. Parameters calculated from the equivalent circuits (Fig. 6) of O2 reduction on the GDEs are listed in Table 3. According to
ip t
table 3, the Rs values change as a function of the potential, indicating that the contribution of the apparatus (connectors, leads and wires) from the total impedance of system shouldn’t be
cr
dismissed. So, the Rs values have a contribution function of both electrolyte and the apparatus
us
resistance, i.e., connectors, leads and wires.
The R1 and R2 also show a dependence on the applied dc potentials (Fig. 7). By increasing the
an
positive potential, the R1 values decrease for all of the GDEs. In the potential region of lower than 0.7 V, the adsorption of free O2 molecules happens on the metal catalyst and then O2,ads
M
reduces to OHads (Eq. 8). In the applied dc potential of E ≥ 0.7 V, the adsorption of Oºads free radicals happens. By increasing the positive applied potentials, the adsorbed amount of O2,ads
d
and Oºads increases and the further reduction process occurs more easily in the catalyst layer.
te
So the R1 values will be decreased.
Ac ce p
The R2 values for E ˂ 0.7 V are related to the adsorption of molecules like H2O2,ads into the GDE pores. When the potential is increased to more positive potentials, the adsorption of H2O2,ads species by the oxygen atom orientation to the metal catalysts will be increased and R2 values became higher. In the potential region of E ≥ 0.7 V, the OH,ads species which are reduced to OH-aq, increase and charge transfer happens more easily. So the R2 values will be decreased. Comparison of R values for GDEs shows that the Pt.Ru/C electrocatalyst has the lowest resistance in the whole range of applied dc potentials. The applied dc potentials in the range of higher or lower than 0.7 V, charge transfer reactions happen according to Eq. 12 and 14. Figure 7 shows that the R1 values of Pt.Ru/C are lowest among the GDEs. So, it can be concluded that the charges transfer more easily and rapidly in this ODC, either dissociative mechanism or associative mechanism happens, respectively. 15 Page 15 of 38
This phenomena happens in the situation that the amount of Pt nanoparticles used in the Pt.Ru/C catalysts is half of Pt/C catalyst one. For the applied dc potentials lower than 0.7 V, the R2 values are related to H2O2 or other intermediate adsorption on the active layer of GDEs. The Pt.Ru/C cathode shows the lower R2
ip t
values and it means that intermediates like H2O2 molecules desorbs more easily from this GEDs so it may be prevented from the active site blocking by these intermediates. For
cr
example, in the applied dc potential equal to 0.4 V, the R2 value of Pt/C, Ru/C and Pt.Ru/C
us
GDEs is 55.8, 62 and 17 Ω.cm2, respectively (Table 3). The R2 value of Pt/C GDE is 3.2 times greater than Pt.Ru/C one. The R2 values in higher potential than 0.7 V relate to charge
an
transfer reaction according to Eq. 13. With the potential of 0.9 V, the R2 values of Pt/C, Ru/C and Pt.Ru/C are 36, 42 and 25 Ω.cm2, respectively, and the R2 value of Pt.Ru/C cathode is 1.4
M
and 1.7 times lower than Pt/C and Ru/C GDEs. So, it can be concluded that the second charge transfer reaction happens more rapidly onto the bimetallic nano catalyst.
d
The dependence of true capacitance values of the applied potentials is shown in Fig. 7, too.
te
The plots show a gradual decrease of capacitances by increasing the potentials. The adsorbed
Ac ce p
species enhancement by increasing the more positive potentials causes to decrease of C1 and C2 values. The adsorption of species like O2 molecules increases by applying the more positive potential. So, according to C=ɛɛ0A/d [41], the values of d become greater and causes to C1 and C2 decreases. The true capacitance of Pt.Ru/C GDE also shows the highest amount among the other cathodes. The decrease of C1 and C2 values of Pt/C and Ru/C ODCs in the comparison of Pt.Ru/C is due to the reduction in the electrochemical active surface area of cathodes as mentioned before in the cyclic voltammetry results. As can be concluded according to the EIS and other technique results, the second metal addition has influenced the ORR rate. Some factors that may explain this electrocatalytic performance include: (a) the ratio of low coordinated surface atoms, which increases inversely with particle size; (b) the
16 Page 16 of 38
electronic state of small metal particles; (c) the strong metal-support interaction and (d) the synergistic effects of Pt and Ru catalyst species. 3.5. Chronoamperometery The durability and oxygen diffusion coefficients of GDEs were examined using
ip t
chronoamperometry technique. Chronoamperograms were obtained by holding the potential of the electrodes at +1.2 V for 10 s and then holding it at +0.4 V relative to the Ag/AgCl
cr
electrode for 4 h with oxygen flowing along the electrolyte. Fig. 8a shows the typical current
us
density-time (i-t) plots with the current normalized to the initial current (i/i0) to indicate the fractional decay with time. The current decay is rapid for the Pt/C and Ru/C electrodes,
an
reaching to 15% and 5% of the initial current in 1.5 and 1 h, respectively. In contrast, the Pt.Ru/C electrode needs 3.5 h to reach 20% of initial current. It can be seen clearly that the
M
current density on the Pt.Ru/C electrode is relatively higher than on the Pt/C and Ru/C GDEs, which suggest that paring the Pt and Ru decrease the opportunity of active electrocatalyst
d
particle dissolution from GDEs.
Ac ce p
[42]:
te
By plotting I vs. t−1/2, the linear dependence relationship was obtained for different electrodes
(16)
Where I is the limited current, A the surface area of the electrode, D the diffusion coefficient, C the concentration of oxygen, n the number of electrons in the overall reaction of ORR, F the Faraday’s constant, t the time, and is equal to 3.14. Fig. 8b shows the plot of current density vs. t-1/2 for GDEs in alkaline media. Cottrell parameters are listed in Table 2, also. The results confirmed the higher Cottrell slope and D values for Pt.Ru/C electrode. So, the Pt.Ru/C cathode has more permeability and activity towards oxygen reduction reaction. Conclusion In this investigation, Pt/C, Ru/C and Pt.Ru/C bimetallic electrocatalysts were prepared by the chemical reduction process. The SEM results showed that all the cathodes consist of rough 17 Page 17 of 38
and porous structures and Pt, Ru nanoclusters were deposited quite uniform onto/into Vulcan carbon supports with the average particle size of about 30-45 nm. The ORR activity of cathodes was evaluated in 0.1 M O2 saturated NaOH media. Comparison of cyclic voltamograms of Pt/C, Ru/C and Pt.Ru/C electrodes in O2 saturated solutions indicated that
ip t
Pt.Ru/C catalyst has higher current density and more positive ORR onset potential in the compare of Pt/C and Ru/C electrodes. It can be concluded that the Ru addition has influenced
cr
the catalytic activity of electrocatalyst toward ORR. This matter can be related to synergistic
us
effect, which is playing a critical role in ORR activity. The Pt.Ru/C cathodes showed lower Tafel slope and high current density. An EIS study was carried out at different dc potentials in
an
order to study the mechanism of ORR in O2 saturated alkaline solutions on prepared GDEs. The Nyquist plots of GDEs under different polarization potentials showed two different
M
behaviors, suggesting different associative and dissociative electrochemical processes
te
Acknowledgements
d
occurring at the electrodes.
Ac ce p
The authors would like to acknowledge the financial support of the Office in Charge of Research of Iranian Nanotechnology Society and the financial support of the Office of Vice Chancellor in charge of research of University of Tabriz. Reference:
1. G. G. Botte, Electrochemical Manufacturing in the Chemical Industry, Electrochem. Soc. Interface. 23 (2014) 49-55. 2. Y.Kiros, M. Pirjamali, M. Bursell, Oxygen reduction electrodes for electrolysis chlor-alkali cells, Electrochim. Acta. 51 (2006) 3346-3350. 3. Thomas F. O'Brien, Tilak V. Bommaraju, Fumio Hine, History of the Chlor-Alkali Industry, in Handbook of chlor-alkali technology, vol. 1, Springer, New York, NY, USA, 2004. 4. I, Moussallem, S. Pinnow, N. Wagner, T. Turek, Devolepment of high-performance silverbased gas-diffusion electrodes for chlor-alkali electrolysis with oxygen depolarized cathodes, Chem. Eng. Process. 52 (2012) 125-131. 18 Page 18 of 38
5. T. Mirzazadeh, F. Mohammadi, M. Soltanieh, E. Joudaki, Optimization of caustic current efficiency in a zero-gap advanced chlor-alkali cell with application of genetic algorithm assisted by artificial neural networks, Chem. Eng. J. 140 (2008) 157–164. 6. F. Mohammadi, S.N. Ashrafizadeh, A. Sattari, Aqueous HCl electrolysis utilizing an oxygen reducing cathode, Chem. Eng. J. 155 (2009) 757-762.
ip t
7. E. Joudaki, F. Farzami, V. Mahdavi,S. J. Hashemi, Performance Evaluation of OxygenDepolarized Cathode with PtPd/C Electrocatalyst Layer in Advanced Chlor-Alkali Cell, Chem. Eng. Technol. 33 ( 2010) 1525-1530
cr
8. S. Siracusano, T. Denaro, V. Antonucci, A. S. Aricò, C. Urgeghe, F. Federico, Degradation of oxygen-depolarized Ag-based gas diffusion electrodes for chlor-alkali cells, J. Appl. Electrochem. 38 (2008) 1637-1646.
us
9. Ludwig Lipp, Shimshon Gottesfeld, Jerzy Chlistunoff, Peroxide formation in a zero-gap chlor-alkali cell with an oxygen-depolarized Cathode, J. Appl. Electrochem. 35 (2005) 1015– 1024.
an
10. J. Chlistunoff, Advanced Chlor-Alkali Technology: Final Technical Report, Los Alamos National Laboratory, New Mexico, United States 2004.
M
11. M. Sugiyama, K. Saikai, A. Sakata, H. Aikawa, N. Furuya, Accelerated degradation testing of gas diffusion electrodes for the chlor-alkali process , J. Appl. Electrochem. 33 (2003) 929–932.
te
d
12. S. Chabi, M. Kheirmand, Electrocatalysis of oxygen reduction reaction on Nafion/platinum/gas diffusion layer electrode for PEM fuel cell, Appl. Surf. Sci. 257 (2011) 10408-10413.
Ac ce p
13. M. Hezarjaribi, M. Jahanshahi, A. Rahimpour, M. Yaldagard, Gas diffusion electrode based on electrospun Pani/CNF nanofibers hybrid for proton exchange membrane fuel cells (PEMFC) applications, Appl. Surf. Sci. 295 (2014) 144– 149. 14. N. Ramaswamy, S. Mukerjee, Fundamental mechanistic understanding of electrocatalysis of oxygen reduction on Pt and non-Pt surfaces: acid versus alkaline media, Adv. Chem. Phys. (2012) 1. 15. T. R. Ralph and M. P. Hogarth, Catalysis for Low Temperature Fuel Cells, Platinum Metals Rev, 46 (2002) 3-14. 16. Sanjeev Mukerjee, Supramaniam Srinivasan, Enhanced electrocatalysis of oxygen reduction on platinum alloys in proton exchange membrane fuel cells, J. Electroanal. Chem. 357 (1993) 201-224. 17. U.A. Paulus, A. Wokaun, G.G. Scherer, T.J. Schmidt, V. Stamenkovic, N.M. Markovic, P.N. Ross, Oxygen reduction on high surface area Pt-based alloy catalysts in comparison to well defined smooth bulk alloy electrodes, Electrochim. Acta. 47 (2002) 3787-/3798. 18. V. Stamenkovic´, T.J. Schmidt, P.N. Ross, N.M. Markovic, Surface segregation effects in electrocatalysis: kinetics of oxygen reduction reaction on polycrystalline Pt3Ni alloy surfaces, J. Electroanal. Chem. 554-555 (2003) 191-199. 19 Page 19 of 38
19. U. A. Paulus, A. Wokaun, and G. G. Scherer, T. J. Schmidt, V. Stamenkovic, V. Radmilovic, N. M. Markovic, and P. N. Ross, Oxygen Reduction on Carbon-Supported Pt-Ni and Pt-Co Alloy Catalysts, J. Phys. Chem. B. 106 (2002) 4181-4191.
ip t
20. V. Stamenkovic´, T. J. Schmidt, P. N. Ross, and N. M. Markovic, Surface Composition Effects in Electrocatalysis: Kinetics of Oxygen Reduction on Well-Defined Pt3Ni and Pt3Co Alloy Surfaces, J. Phys. Chem. B. 106 (2002) 11970-11979.
cr
21. J. Zhang, Y. Mo, M. B. Vukmirovic, R. Klie, K. Sasaki, and R. R. Adzic, Platinum monolayer electrocatalysts for O2 reduction: Pt monolayer on Pd(111) and on carbonsupported Pd nanoparticles, J. Phys. Chem. B. 2004, 108, 10955-10964.
us
22. C. Jeyabharathi, P. Venkateshkumar, J. Mathiyarasua, K.L.N. Phani, Platinum–tin bimetallic nanoparticles for methanol tolerant oxygen-reduction activity, Electrochim. Acta. 54 (2008) 448-454.
an
23. A. Nozad Golikand, E. Lohrasbi, M. Ghannadi Maragheh, M. Asgari, Enhancement of electrocatalytic O2 reduction on carbon nanotube-supported Pt alloys nanoparticles in gas diffusion electrodes, J Appl Electrochem. (2009) 39:1443-1449.
d
M
24. T. Frelink, W. Visscher, A. P. Cox, J. A. R. Van veen, Ellipsometry and dems study of the electrooxidation of methanol at Pt and Ru- and Sn- promoted Pt, Electrochim. Acta. 40 (1995) 1537-1543.
te
25. D. Chu, S. Gilman, Methanol Electro‐oxidation on Unsupported Pt‐Ru Alloys at Different
Ac ce p
Temperatures, J. Electrochem. Soc. 143 (1996) 1685-1690. 26. A.L. Ocampo, M. Miranda-Hern´andez, J. Morgado, J.A. Montoya, P.J. Sebastian, Characterization and evaluation of Pt-Ru catalyst supported on multi-walled carbon nanotubes by electrochemical impedance, J. Power Sources, 160 (2006) 915-924. 27. E. Lust,1, E. Härk, J. Nerut, K. Vaarmets, Pt and Pt–Ru catalysts for polymer electrolyte fuel cells deposited onto carbide derived carbon supports, Electrochim. Acta. 101 (2013) 130141. 28. C. Song, J. Zhang, Electrocatalytic Oxygen Reduction Reaction in PEM Fuel Cell Electrocatalysts and Catalyst Layers, Springer, Verlag London Limited. (2008) chapter 2, pp: 89-134. 29. A.J. Bard, L.R. Faulkner, Electrochemical Methods: Fundamentals and Applications, second ed., Wiley, NY, 2001. 30. Malika Ammam, E. Bradley Easton, Oxygen reduction activity of binary PtMn/C, ternary PtMnX/C (X= Fe, Co, Ni, Cu,Mo and, Sn) and quaternary PtMnCuX/C (X= Fe, Co, Ni, and Sn) and PtMnMoX/C (X= Fe, Co, Ni, Cu and Sn) alloy catalysts, J. Power Sources, 236 (2013) 311-320. 20 Page 20 of 38
31. Rosa Rego, A.M. Ferraria, A.M. Botelho do Rego, M. Cristina Oliveira, Development of PdP nano electrocatalysts for oxygen reduction reaction, Electrochim. Acta. 87 (2013) 73-81 32. C. Wang, B. Peng, H. N. Xie, H.X. Zhang, F. F. Shi, W. B. Cai, Facile Fabrication of Pt, Pd and Pt−Pd Alloy Films on Si with Tunable Infrared Internal Reflection Absorption and Synergetic Electrocatalysis, J. Phys. Chem. C. 113 (2009) 13841-13846.
ip t
33. K. Kakaei, Electrochemical Characteristics and Performance of PlatinumNanoparticles Supported by Vulcan/Polyaniline for OxygenReduction in PEMFC, Fuel cells. 12 (2012) 939945
cr
34. N. R. Elezovic, B. M. Babic, LJ. M. Vracar, N.V. Krstajic, Oxygan reduction at platinum nanoparticles supported on carbon cryogel in alkalin solution, J. Serb. Chem. Soc. 72 (2007) 699-708.
an
us
35. T. Lopes, E. Antolini, E.R. Gonzalez, Carbon supported Pt–Pd alloy as an ethanol tolerant oxygen reduction electrocatalyst for direct ethanol fuel cells, Int. J. Hydrogen Energy. 33 (2008) 5563 -5570.
M
36. N. Ramaswamy, S. Mukerjee, FundamentalMechanistic Understanding of Electrocatalysis of Oxygen Reduction on Pt and Non-Pt Surfaces: Acid versus AlkalineMedia, Adv. Phys. Chem. 2012 1-12. 37. N.A. Anastasijevic, Z.M. Dimitrijevic, R.R. ADzic, Oxygen reduction on a ruthenium electrode in alkalin electrolytes, .J Electroanal. Chem. 199 (1986) 351-364.
te
d
38. E. H. Yu, S. Cheng , Bruce E. Logan, K. Scott, Electrochemical reduction of oxygen with iron phthalocyanine in neutral media, J Appl Electrochem, 39 (2009) 705-711.
Ac ce p
39. V. P. Zhdanov, B. Kasemo, Kinetics of electrochemical O2 reduction on Pt, Electrochem. Commun. 8 (2006) 1132–1136. 40. J. K. Norskov, J. Rossmeisl, A. Logadottir, L. Lindqvist, Origin of the Overpotential for Oxygen Reduction at a Fuel-Cell Cathode, J. Phys. Chem. B. 108 (2004) 17886-17892. 41. M. Kissi, M. Bouklah, B. Hammouti, M. Benkaddour, Establishment of equivalent circuits from electrochemical impedance spectroscopy study of corrosion inhibition of steel by pyrazine in sulphuric acidic solution, Appl. Surf. Sci. 252 (2006) 4190-4197. 42. J. Ribeiro, A. R. de Andrade, Investigation of the electrical properties, charging process, and passivation of RuO2–Ta2O5 oxide films, J. Electroanal. Chem. 592 (2006) 153-162.
21 Page 21 of 38
ip t
Table. 1. Surface composition of GDEs by EDX analysis.
cr
Table caption
recorded in 0.1 M O2 saturated NaOH solution.
us
Table. 2. The electrochemical active surface area, Tafel and Cottrell parameters of GDEs,
an
Table. 3. Impedance parameter values of GDEs at different dc potentials; recorded in 0.1 M
Ac ce p
te
d
M
O2 saturated NaOH solution.
22 Page 22 of 38
ip t
Figure caption
cr
Fig. 1. Schematic illustration of the three-electrode compartment set up for electrochemical
us
tests.
Fig. 2. SEM surface morphologies of Pt/C (a), Ru/C (b), Pt.Ru/C (c) and EDX analysis of
an
Pt/C (d), Ru/C (e), Pt.Ru/C (f) GDEs.
Fig. 3. Cyclic voltammograms of Pt/C (a), Ru/C (b), Pt.Ru/C (c) GDEs in 0.1 M N2 saturated
M
(solid line) and O2 saturated (dash line) NaOH solution at a scan rate of 50 mV.s-1. Fig. 4. Comparison of polarization curves for the ORR onto the GDEs in 0.1M O2 saturated
d
NaOH solution and 1mV.s-1 sweep rate.
te
Fig. 5. Impedance diagrams in the complex plane of Pt/C (a), Ru/C (b), Pt.Ru/C cathodes as a
Ac ce p
function of applied potentials, recorded in 0.1 M O2 saturated NaOH media. Fig. 6. The electrical equivalent circuits (EEC) used for the fitting of the experimental results presented in Fig. 5.
Fig. 7. Dependence of the R1, C1 (a) and R2, C2 (b) values of the GDE as a function of different applied potentials; recorded in 0.1 M O2 saturated NaOH media. Fig. 8. Normalized current verse time (a) and current density verse t−1/2 plots (b) of the GDEs in 0.1 M NaOH solution with constant potential +0.4 V (vs. Ag/AgCl) during 4 h.
23 Page 23 of 38
C 64.53 56.79 51.8
W (%) Pt 35.47 24.38
Ru 43.21 23.82
Ac ce p
te
d
M
an
us
cr
catalyst Pt/C Ru/C Pt.Ru/C
ip t
Table. 1. Surface composition of GDEs by EDX analysis.
24 Page 24 of 38
NA (×1016 )
Pt/C Ru/C Pt.Ru/C
7.2 5.5 19.8
0.68 0.1 0.75
91 78 253
Exchange current density (A.cm-2) 1.34×10-3 1.28×10-3 2.05×10-3
Tafel slope (mV/dec)
Cottrell slope (A.s-1/2)
CD1/2 (mol.cm-2.s1/2 )
136.4 150.6 122.3
1.39 × 10-6 3.80 × 10-7 2.6 × 10-6
ip t
E onset (V)
0.304 0.083 0.569
us
EASA (cm2)
Ac ce p
te
d
M
an
catalyst
cr
Table. 2. The electrochemical active surface area, Tafel and Cottrell parameters of GDEs, recorded in 0.1 M O2 saturated NaOH solution.
25 Page 25 of 38
Table. 3. Impedance parameter values of GDEs at different dc potentials; recorded in 0.1 M O2 saturated NaOH solution. Rw (Ω.cm2) 1917 341.6 117.2 1610 267.8 116 357.5 139.4 110.3 -
Yw (Ω-1.sn.cm-2) 13.53 9.17 11.3 30.95 14.60 9.49 36.79 39.47 20.62 -
ip t
0.97 0.98 0.96 0.91 0.82 0.91 0.95 0.96 0.83 0.85 0.87 0.95 0.98 0.81 0.89
R2 (Ω.cm2) 86.30 55.80 40.20 40.80 36.00 92.00 62.00 49.00 47.00 42.00 16.00 17.00 20.00 28.00 25.00
cr
ndl
us
0.66 0.61 0.65 0.67 0.69 0.78 0.85 0.86 0.75 0.79 0.75 0.73 0.75 0.75 0.73
Y0dl (Ω-1.sn.cm-2) 0.13 0.12 0.11 0.12 0.13 0.73 0.21 0.17 0.51 0.39 0.50 0.41 0.44 0.54 0.46
an
R1 (Ω.cm2) 5.8 5.6 5.5 5.4 5.2 5.4 5.3 5.2 5.1 5.1 5.0 4.9 4.9 4.8 4.7
np
M
Pt.Ru/C
Y0f (Ω-1.sn.cm-2) 11 19 12 10 8.6 72 31 62 4 3.1 3.2 4.6 3.9 4.0 4.6
nw 0.79 0.69 0.64 0.93 0.69 0.65 0.82 0.62 0.61 -
d
Ru/C
Rs (Ω.cm2) 7.2 7.2 7.3 7.3 7.3 6.8 7.1 6.9 6.9 7.1 11.3 10.9 10.8 10.7 11.0
te
Pt/C
E (V) OCP 0.4 0.6 0.7 0.9 OCP 0.4 0.6 0.7 0.9 OCP 0.4 0.6 0.7 0.9
Ac ce p
sample
26 Page 26 of 38
ip t
Ac ce p
te
d
M
an
us
cr
Fig. 1.
27 Page 27 of 38
28
Page 28 of 38
d
te
Ac ce p us
an
M
cr
ip t
29
Page 29 of 38
d
te
Ac ce p us
an
M
cr
ip t
ip t cr us an M d te Ac ce p
Fig. 2.
30 Page 30 of 38
31
Page 31 of 38
d
te
Ac ce p us
an
M
cr
ip t
ip t cr us an
Ac ce p
te
d
M
Fig. 3.
32 Page 32 of 38
ip t cr us an
Ac ce p
te
d
M
Fig. 4.
33 Page 33 of 38
ip t cr
70
(a)
an
40
5
30
10
15
20 OCP
M
-Z'' (Ω.cm2)
50
us
14 12 10 8 6 4 2 0
60
20 10
0.4V 0.6V 0.7V
d
0.9V
0
10
20
30
40
50
60
70
2
Z' (Ω.cm )
Ac ce p
te
0
30
(b)
10
25
5
-Z'' (Ω.cm2)
20
0
15
5
10
15 OCP
10
0.4V 0.6V
5
0.7V 0.9V
0 5
10
15
20
25
30
2
Z' (Ω.cm )
34 Page 34 of 38
ip t
30
15
20
5
0
15
cr
10
us
25
2
-Z'' (Ω.cm )
(c)
15
an
10
0
15
Ac ce p
te
10
d
5
20
25
OCP 0.4V 0.6V
M
10
20
0.7V 0.9V
25
30
35
40
2
Z' (Ω.cm )
Fig. 5.
35 Page 35 of 38
ip t cr us
Ac ce p
te
d
M
an
Fig. 6.
36 Page 36 of 38
12
35 Pt.Ru/C
Pt/C
Ru/C
Pt.Ru/C
25
15
8
ip t
R1 (Ω.cm2)
10
Ru/C
C1 (mF/cm2)
(a)
Pt/C
5
cr
6
0.2
0.4
0.6
an
E/V (Ag/AgCl)
160
M
(b)
0.8
0.5
Pt/C
Ru/C
Pt.Ru/C
Pt/C
Ru/C
Pt.Ru/C
te
70
-0.1
40
-0.3
10
0
0.2
0.3
0.1
d
100
Ac ce p
R2 (Ω.cm2)
130
-15
C2 (mF/cm2)
4 OCP 0
us
-5
-0.5 0.4 0.6 E/V (Ag/AgCl)
0.8
Fig. 7.
37 Page 37 of 38
1 (a)
Pt/C Ru/C Pt.Ru/C
ip t
0.50
cr
0.25 -0.5
0-1 0
0.5
1
1.5
2
2.5
3
3.5
4
an
Time (h)
30
M
(b) 25
d
20
15
te
J (mA/cm2)
us
Normalized current
0.75 0.5
Pt.Ru/C
10
Pt/C
Ac ce p
Ru/C
5
0
0
0.01
0.02
0.03 t
-1/2
0.04 (s
0.05
0.06
0.07
-1/2
)
Fig. 8.
38 Page 38 of 38
S0169-4332(15)00742-4 http://dx.doi.org/doi:10.1016/j.apsusc.2015.03.146 APSUSC 30023
To appear in:
APSUSC
Received date: Revised date: Accepted date:
17-7-2014 23-3-2015 23-3-2015
Please cite this article as: M.G. Hosseini, P. Zardari, Electrocatalytical study of carbon supported Pt, Ru and bimetallic Pt-Ru nanoparticles for oxygen reduction reaction in alkaline media, Applied Surface Science (2015), http://dx.doi.org/10.1016/j.apsusc.2015.03.146 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Highlights Binary catalyst Pt.Ru/C is evaluated towards ORR. Pt.Ru/C nanoparticles revealed best ORR catalytical activity. The 120 mV/dec Tafel slop indicated that the first electron transfer is the rds.
Ac ce p
te
d
M
an
us
cr
ip t
The active number sites of Pt.Ru/C catalyst were 3 times higher than Pt/C.
1 Page 1 of 38
Electrocatalytical study of carbon supported Pt, Ru and bimetallic Pt-Ru nanoparticles for oxygen reduction reaction in alkaline media
ip t
M.G. Hosseini* a, P. Zardaria a) Department of Physical Chemistry, Electrochemistry Research Laboratory, University of
cr
Tabriz, Tabriz, Iran
us
Abstract
Carbon supported Pt, Ru and bimetallic Pt-Ru nanoparticles (Pt/C, Ru/C and Pt.Ru/C) have
an
been prepared by the chemical reduction method. Particle morphology, composition and structure of nanoparticles have been investigated by scanning electron microscopy (SEM) and
M
energy dispersive X-ray (EDX) analysis. SEM results showed a uniform dispersion of nanoparticles with rough and porous structure into carbon supports with the average particle
d
size of 30-64 nm. EDX analysis demonstrated the presence of both Pt and Ru nanoparticles in
te
each gas diffusion electrode. The Pt/C, Ru/C and Pt.Ru/C composites were used as
Ac ce p
electrocatalyst for oxygen reduction reaction (ORR) in alkaline media. The ORR activities of cathodes were characterized using cyclic voltammetry (CV), polarization technique, AC impedance spectroscopy (EIS) and chronoamperometry. CV and polarization curves showed significantly higher activity on Pt.Ru/C electrocatalyst than observed on Pt/C and Ru/C catalysts, which can be related to synergistic effect, which is playing a critical role in ORR activity. The Tafel slope values of 120 mV/dec showed that the first electron transfer is the rate determining step. The EIS results of cathodes under different polarization potentials indicated two different behaviours which depend on the applied dc potentials and reveals
*
Corresponding Author. Tel.: +984113393138; fax: +984113340191.
E-mail address: [email protected] .
2 Page 2 of 38
different electrochemical processes occurring on the electrodes. Keywords: ORR, ODC, Pt electrocatalyst, Pt-Ru electrocatalyst, EIS, CV
1. Introduction
ip t
Nowadays, the chlor-alkali industry is one of the largest chemical processes worldwide. Its two main components - chlorine and caustic soda - are indispensable commodities that are
cr
used for a wide range of applications. Nearly 55 percent of all specialty chemical products
us
manufactured require one of the chlor-alkali products as a precursor [1, 2]. Chlorine and caustic soda, obtained by the electrolysis of sodium chloride solution by three methods
an
included the mercury, diaphragm and ion exchange membrane processes [3]. The brine electrolysis system has been changed from the mercury and the diaphragm process to the ion
M
exchange membrane process in the most advanced countries because of the attempting to reduce the manufacturing costs of chlorine production [4-7]. In the ion exchange membrane
d
process, the total cell reaction for the electrolysis of brine into chlorine and caustic soda
te
proceeds through the following half-cell reactions and the decomposition voltage of the total
Ac ce p
cell reaction is about -2.2 V [8].
Anode reaction: 2Cl- → Cl2 + 2e-
E=+1.358 V/NHE
Cathode reaction: 2H2O + 2e- → 2OH- + H2
E=-0.828 V/NHE
(1) (2)
The electrolysis of brine through the ion exchange membrane process and by hydrogen evolution reaction (HER) cathodes is still one of the most energy-intensive industrial operations and it is necessary to reduce energy consumption. The application of an oxygen depolarized cathode (ODC) by using a gas diffusion electrode (GDE) instead of HER cathode can solve this problem. By replacing the ODC instead of HER cathodes, the half cell reactions can be written as follows: Anode reaction: 2Cl- → Cl2 + 2e-
E=+1.358 V/NHE
(3) 3 Page 3 of 38
Cathode reaction: 2H2O + O2 +4e- → 4OH-
E=+0.401 V/NHE
(4)
Nevertheless, by replacing the HER cathodes in a membrane cell with an ODC, the cell voltage and energy consumption can be reduced by as much as 30-40% at 0.4 A/cm2 [9, 10].
brine in the chlor-alkali cells that have to meet several criteria such as: a) Long term stability in sodium hydroxide media at 80-90 oC.
cr
b) High catalytic activity for ORR.
ip t
Oxygen depolarized cathodes consist of a gas diffusion electrode (GDE) for electrolysis of
us
c) Suitable hydrophobic/hydrophilic pore structure for transport of oxygen and electrolyte to the reaction sites [11].
an
The GDE consists of a reaction layer, gas diffusion layer and current distributor. The reaction layer structure is made of carbon black, Polytetrafluoroethylene (PTFE), and oxygen
M
reduction active catalyst. The gas diffusion layer is made of carbon black and PTFE and facilitates the reactant permeation from the electrolyte to the catalytic sites. The Ni, Ag mesh
d
or highly pyrolytic graphite is used as a current distributor in GDE [12, 13]. In the last
te
decades, Pt catalyst has attracted a great deal of studying due to their promising applications,
Ac ce p
particularly as the electrode catalyst for oxygen reduction in GDEs and considerable progress has been achieved [14-16].
Despite the fact that the Pt catalyst has excellent electrocatalytic activities for oxygen reduction, some obstruction still remain for the commercialization of ODC cells, such as the high cost of Pt. Consequently, the major issues facing ODC cells are the cost down of electrode and finding a new selective oxygen reduction catalyst on ODC electrode. So, many research groups have attempted to find new low cost electrocatalysts that selectively catalyse oxygen reduction reactions [16]. Recent studies [17-22] have examined the use of bimetallic Pt alloys as oxygen reduction catalysts. The goal of these studies is to improve the catalytic activity of platinum, by alloying it with transition metals like Pd, Ru, Ni, for the ORR at cathode sites of GDEs. The alloying of transition metals exhibits significantly higher 4 Page 4 of 38
electrocatalytic activities towards the ORR than platinum alone [23]. This enhancement effect has been explained by models such as the “bifunctional mechanism” [24] or by the “electronic effect” [25], which indicates a promotional effect of the alloyed metal on Pt. Particularly, Pt-Ru has been investigated as a binary system, and has shown the appropriate
ip t
catalytic activities in many different electrochemical reactions [26]. Several studies, carried out on the polycrystalline (PC) Ru, monocrystalline Ru, Pt submonolayers on Ru, and
cr
Pt.Ru/C composite electrodes, have shown their considerable activity for ORR in acidic and
us
alkaline media [27]. The Volcano plot of ORR by different Pt group metals shows that the oxygen is chemisorbed on Ru sites more intensively than Pt sites and the further desorption of
an
oxygen on Ru sites is more difficult than Pt. On the other hand, the molar ratio of Pt:Ru in electrocatalyst layer is very important. If the bimetallic catalyst is prepared by the high molar
M
ratio of Ru, the oxygen will be adsorbed more strongly on the alloy and the future desorption
alloy will be reduced [28].
d
of oxygen molecules will be more difficult, so the electrocatalytical activity of bimetallic
te
The main aim of this work is to accomplish a method to prepare bimetallic Pt.Ru/C
Ac ce p
nanoparticle electrocatalyst which has the optimum amount of Ru sits according to Volcano plots. On the other hand, the amount of Pt will be chosen as minimum as possible due to economic problems of Pt. In order to investigate the electrochemical activity of binary Pt-Ru alloys in GDEs, the Pt, Ru and Pt-Ru bimetallic catalysts were supported on carbon black, used in the catalyst layer of GDEs, and characterized by various electrochemical techniques to study their ORR activity.
2. Material and methods 2.1. Chemicals used Sodium borohydride (NaBH4, 96%), sodium hydroxide (NaOH, 99%) and 2-propanol ((CH3)2CHOH, 99.99%) were received from Merck. Ruthenium trichloride anhydrous
5 Page 5 of 38
(RuCl3·3H2O, 99%) and chloroplatinic acid hydrate (H2PtCl6.xH2O, 99.99%), were purchased from Cole-Parmer and Sigma-Aldrich, respectively. Polytetrafluoroethylene (PTFE, 99%), Vulcan carbon (XC-72R), nafion solution (5wt %) and
TP-060T, respectively. All chemicals used were of analytical grade. 2.2. Synthesis of catalyst
ip t
Teflon treated carbon paper were obtained from Sigma-Aldrich, Cobat, Dupont, Toray EC-
cr
Appropriate amount of Vulcan carbon (XC-72R) was ultrasonically dispersed in the mixture
us
of 2-propanol and de-ionized water for 2 h. Metal precursors (H2PtCl6.xH2O + RuCl3.3H2O) were added to the resulted Vulcan carbon ink. The molar ratio of Pt to Ru was achieved 1:1
an
and the total metal content in the elctrocatalyst was 40 wt% vs. carbon support. The 1 M NaOH solution was added to adjust the pH of ink to 7-8. The mixture was heated up to 80 oC
M
and 10 mL NaBH4 solution was added drop-wise to the suspension under vigorous stirring condition at 80 oC. The solid particles were separated from the solution by filtration, then
d
washed and dried at 120 oC for 6 h. For comparison, 40 wt% Pt/C and 40 wt% Ru/C were
te
prepared by the same method too.
Ac ce p
2.3. Preparation of gas diffusion layer Vulcan carbon (XC-72R) and PTFE solution was ultrasonically dispersed into the mixture of 2-propanol, water and glycerol for 1 h. The PTFE content to Vulcan carbon (XC-72R) was 60 wt%. Then resulted suspension was painted onto the carbon paper (geometric exposed area of 1 cm2). The painted layer was dried at 80 oC and sintered at 350 oC for 30 min. The loading of the gas diffusion layer was obtained equal to 1 mg.cm-2. 2.4. Preparation of gas diffusion electrode The catalyst inks were prepared by mixing the appropriate amount of catalysts with 2propanole, de-ionized water and 5wt% nafion solution. The loading of the nafion and catalysts on the GDLs were about 1 mg.cm-2. After the painting of catalyst inks on the GDLs, the final electrodes were dried at 120 oC for 1 h. The surface morphology of the GDEs was studied 6 Page 6 of 38
with a scanning electron microscope (Philips, Model XL30) and energy dispersive X-ray (EDX) analysis. 2.5. Electrochemical measurements The GDEs were mounted in a Teflon holder containing a high pyrolytic graphite disk as a
ip t
current collector which had arranged for oxygen feed from the back of the electrode. The electrochemical measurements were carried out using three-electrode compartment (Fig. 1).
cr
An Ag/AgCl electrode was used as the reference electrode and platinum as the counter
us
electrode.
Electrochemical experiments were done using the Potentiostat/Galvanostat (EG&G model
an
Parstat 2263) equipped with Frequency Response Analyzer (FRA) controlled by a PC through PowerSuite software in the 0.1 M NaOH solution, at 25 ◦C. Cyclic voltammetry was applying
M
a scan rate of 50 mV.s-1 between -0.6 and +1.1 V vs. Ag/AgCl in the O2 saturated NaOH solution. EIS measurements were performed over a frequency range of 100 KHz to 0.01 Hz,
d
using ±5 mV amplitude of a sinusoidal voltage as perturbing signal. The samples were
te
potentiodynamically polarized to the specified potentials. Experimental data were fitted and
Ac ce p
analysed by ZView (II) software. Tafel polarization measurements were done in the potential range of 1 to -0.2 V vs. Ag/AgCl by applying a scan rate of 1 mV.s-1 and the Tafel slopes and exchange current densities were obtained. Chronoamperometric determinations were achieved by holding the potential of the electrodes at +1.2 V vs. Ag/AgCl for 10 s and then stepping the potential to a value of +0.4 V for 4 h. 3. Results and discussion
3.1. SEM and EDX analyses The SEM graphs and EDX analysis of GEDs are shown in Fig. 2. The surface of GDEs revealed rough and porous structures with high and quite uniform dispersion of Pt and Ru nanoparticles onto/into Vulcan carbon supports with average diameter about 57, 64 and 43 nm for Pt/C, Ru/C and Pt.Ru/C catalysts, respectively. The EDX analysis showed presence of 7 Page 7 of 38
35.47 and 24.38 wt.% Pt in Pt/C and Pt.Ru/C nanoparticles, respectively. The obtained weigh percent of Ru in Ru/C and Pt.Ru/C electrocatalyst was around 43.21 and 23.82 which is in good agreement with nominal values of catalysts (Table 1). 3.2. Cyclic Voltammetry
ip t
Fig. 3 shows the cyclic voltammograms recorded for the Pt/C, Ru/C and Pt.Ru/C GDEs in aqueous 0.1 M NaOH saturated with O2 (dash line) or N2 (solid line), at a scan rate of 50
cr
mV.s−1. All the CV curves recorded under N2-atmosphere, show two distinguish peaks (I, II).
us
While, the CV curves in the O2-saturated electrolyte, show three characteristic regions, namely, the hydrogen region, the double layer charging and the surface oxide
an
formation/reduction. In other words, an additional reductive current is visible, that is absent in the case of saturation with N2. On the anodic scan of O2-saturated electrolyte, the peaks at
M
about -0.3 to 0 V (peak I) are assigned to the oxidation of hydrogen absorbed on the metallic active sites. Peak (I) is clearly intense in the case of Pt/C and Pt.Ru/C catalysts because of the
d
more catalytical active particles in these samples (greater hydrogen adsorption).
It is
te
generally believed that it is very difficult to separate the regions of hydrogen
Ac ce p
adsorption/adsorption and double layer regions for Ru particles. So, the peak (I) isn’t so identifiable in the Ru/C catalyst CV. The double layer region of catalysts is observed in the potential region of 0.0 to 0.7 V. On the anodic sweep and above 0.7 V potential, the formation of oxide film is done on the surface of metallic electrocatalyst (peak II). The onset potential of peak (III) for Pt.Ru/C, Pt/C and Ru/C catalysts are observed about 0.75, 0.68 and 0.1V, respectively. As discussed in the introduction part, by replacing the ODC instead of HER cathode, the activation energy of water reduction process will be reduced and the water reduction peak will be appear in the more positive potential. By the comparison of CV plots for N2 and O2 saturated electrolyte, it can be conclude that the water reduction process started in the more positive potential in the present of O2 molecules. For example, the water reduction potential of Pt/C GDE in the O2 saturated electrolyte is started 1 V more positive than N2 8 Page 8 of 38
saturated one. From the application standpoint, for an efficient operation of an ODC, the current intensity of the ORR induced for the same O2 concentration for all catalysts (comparative study for the saturated O2 solution) has to be as high as possible because high current reflects a fast reaction [29]. Thus, onset potential, current intensity and sharpness of
ip t
the peak (III) in Fig. 3 are the main criteria used in this study to identify the best catalyst for the ORR. The advantage of this method is the simplicity and the accuracy at the condition that
cr
the experiments are well done in a comparative manner, which the case of the present study.
us
The onset potentials of ORR for used catalysts are listed in Table 2. The onset potential of ORR with Pt/C and Pt.Ru/C are located in the same region, while the onset potential of ORR
an
with Ru/C is shifted by 600 mV towards more negative potentials. It can be concluded that the ORR with Pt.Ru/C catalyst is energetically more favourable [30]. Comparison of cyclic
M
voltamograms of Pt/C, Ru/C and Pt.Ru/C electrodes in O2 saturated solutions show that for Pt.Ru/C catalyst, ORR current intensity increase within the ORR region. It can be concluded
d
that the second metal addition has influenced the catalytic activity of electrocatalyst toward
te
ORR. This matter can be related to synergistic effect, which is playing a critical role in ORR
Ac ce p
activity. Despite the hydrogen peaks, the surface areas of catalysts (EASA) were determined from the charge of the monolayer chemisorbed oxygen on the surface, because their adsorption/desorption charge could have some contribution from hydrogen absorption [31]. The EASA values and active site number (NA) of catalysts are also listed in Table 2. The EASA value of the Pt.Ru/C catalyst was approximately 2.75 and 3.6 times higher than that of Pt/C and Ru/C catalysts. So, with increasing of available active surface area, catalytical activity of electrocatalyst towards ORR has been raised. As mentioned in the introduction part, the Ru and Pt alloying in the optimum amount of Volcano plots can increase the ORR activity of catalyst. The CV results lead us to conclude that the bimetallic nanoelectrocatalyst acts more powerful than single Pt/C as ORR catalyst although the Pt loading amount in Pt.Ru/C is half of Pt/C one. This matter can be related to the synergistic effect, which is 9 Page 9 of 38
playing a critical role in ORR. The synergistic effect of the simultaneous presence of Pt and Ru in the electocatalyst can be explained by the spill-over effect of hydrogen and oxygen from Pt surface to Ru and the resultant freedom for Pt active sites [32]. The electroactive surface area of synthesized Pt/C electrodes according to hydrogen adsorption/desorption peak
(10 wt.%, E-TEK) commercial catalyst (32.4 m2.g-1) [33].
cr
3.3. Polarization measurement
ip t
was also estimated equal to153 m2.g-1 which is in a good agreement value of commercial Pt/C
us
Figure 4 shows polarization curves for three electrodes in a conventional three-electrode cell. Polarization curves were produced at a scan rate of 1 mV.s-1 between 1 and -0.2 V vs.
extracted by using the Tafel equation [34]:
M
= (2.303RT/αnF) log(i/io)
an
Ag/AgCl in the O2 saturated NaOH solution. The kinetic parameters of the ORR were
(5)
Where = (E−E0) is the overpotential, R is the gas constant, T is the absolute temperature, α
d
is the transfer coefficient, i0 is the exchange current density, i is the current density, n is the
te
number of electrons, and F is the Faraday’s constant.
Ac ce p
The current density of Pt.Ru/C cathode is higher than Pt/C and Ru/C electrodes according to Fig 4. The kinetic parameters of electrodes are reported in Table 2. The results showed that the ORR current density in Pt.Ru/C GDE is higher than in Pt/C and Ru/C GDEs and the Pt.Ru/C cathode shows lower Tafel slope, which indicated that the Pt.Ru/C electrode has a much better performance in ORR. It can be as a result of weak O-O bond on Pt-modified Ru nanoparticles [35], the synergy effect between Pt and Ru nanoparticles and uniform dispersion of Ru and Pt on the carbon support. The lower Tafel slope and higher current density of Pt.Ru/C cathode can be attributed to the electrocatalytical activity enhancement of Pt.Ru/C electrocatalyst in relative to Pt/C and Ru/C electrocatalysts. The Tafel slopes of ≥120 mV per decade were accounted for all GDEs.
10 Page 10 of 38
There are two different ORR pathways in the alkaline media, i.e. inner-sphere mechanism which occurs in the inner Helmholtz plane (IHP) and the outer-sphere mechanism in outer Helmholtz plane (OHP). In alkaline media, water molecules act as solvent and also proton sources required in the ORR. The loci of species that chemisorbed on the electrode surface
ip t
like hydroxyl species, solvent water dipoles and chemisorbed O2 constitute the IHP. These species populating the IHP covalently interact with the electrode surface. On the other hand,
cr
the OHP consists of solvated species interacting with the electrode via long range electrostatic
us
forces. For example, alkali metal ions are typically solvated and populate OHP. In the case of inner-sphere electron transfer mechanism of ORR, the strong chemisorbed O2 molecules (with
an
or without bond breaking) reduced on oxide-free metal active sites by 4e−/4H+ transfer. The outer-sphere electron transfer mechanism occurs only at the oxide-covered metal sites. The
M
solvated molecular O2 clusters interact with adsorbed hydroxyl species and promote the 2e-
oxide-covered metal sites [36].
d
reaction pathway. Thus, peroxide intermediate (HO2-) is produced as the final product at the
te
The Tafel slope value of ≥120 mV/dec shows that the first electron transfer is the rate
Ac ce p
determining step, according to the inner-sphere electron transfer mechanism. This mechanism is written as follows:
M + O2 + e- → [MOO]-
(6)
[MOO-] + H2O → MOOH + OH-
(7)
MOOH + e- → [MOOH]-
(8)
[MOOH]- + H2O + e- → MOH + 2OH-
(9)
MOH + e- → M + OH-
(10)
Tafel slope higher than 120 mV/dec is often observed in the reactions through some adsorbed layers at the surface [37]. 3.4. Electrochemical impedance spectroscopy
11 Page 11 of 38
An EIS study was carried out at different dc potentials in order to study the mechanism of ORR in O2 saturated alkaline solutions on prepared GDEs. The Nyquist plots of GDEs under different polarization potentials are shown in Fig. 5. The impedance diagrams show two different behaviours which depend on the applied dc
different electrochemical processes occurring on the electrode.
ip t
potentials. In the other words, the shape of plots changes at different potentials, suggesting
cr
The Nyquist plots of O2 reduction on GDEs at the E≥0.7 show two loops, while the
us
impedance spectra acquired at the potential lower than 0.7 V indicate one loop in the high frequency region associated with the time constant of a charge transfer process and semi-
an
infinite diffusive manner related to Warburg component in the low frequency region. This semi-infinite diffusive character is related to the adsorption of reactants and intermediate
M
products. The processes which could be involved on the electrode surface that would produce these changes include [38]:
d
(1) Diffusion of O2 through the gas phase in the pores (of porous carbon supported catalyst)
te
and the electrolyte to the reaction site.
Ac ce p
(2) Adsorption or heterogeneous surface reaction of the oxygen, together with oxygen diffusion.
(3) Charge transfer.
(4) Diffusion of reduction products into the bulk electrolyte ORR is a multi-electron reaction that includes a number of elementary steps involving different reaction intermediates. Many reaction mechanisms have been proposed to describe ORR in aqueous electrolytes. Damjanovic model describes the ORR as a multi-electron reaction which O2 molecules in the vicinity of the electrode are irreversibly reduced directly to H2O through 4-electron transfer or to H2O2 through 2-electron transfer. The H2O2 formed can be reduced to H2O through 2-eltectron transfer or diffuse into the bulk solution. The H2O2 intermediate can be observed when there is no splitting of the O-O bond before H2O2 species 12 Page 12 of 38
is formed. Thermodynamically speaking, the reduction potential of O2 into H2O2 is lower than the potential required forming H2O molecule following 4 electron pathway (E0=0.69 and 1.23 V vs. NHE, respectively). On the other hand, if the produced H2O2 is further transformed into H2O, the overall reaction to reduce O2 into H2O passing through H2O2 will require more
ip t
energy (E0 =1.76 V vs. NHE) [30].
The ORR mechanism on transition metals has also been investigated by theoretical
cr
calculation based on the electronic structure [39, 40]. According to these studies, two main
us
reaction mechanisms are suggested: The dissociative and associative mechanisms. The dissociative mechanism is proposed for a low current density range (more positive potentials),
an
while the associative mechanisms happens in high current density range (more negative
Dissociative Mechanism (E ≥ 0.7 V):
M
potentials) [28]. The brief explanation of these mechanisms is discussed here.
In this mechanism, no H2O2 is produced. On a metal surface, O2 adsorption breaks the O-O
d
bond and forms adsorbed atomic O, which further gains two electrons in the two consecutive
te
steps, forming hydroxide ions. Since there is no adsorbed O2 on the catalyst surface, H2O2
Ac ce p
cannot be formed. This mechanism can be considered a detailed form of the direct 4-electron pathway and can be written as follows: 1/2O2 + M (metal active sites) → Oºads
(11)
Oºads + e- + H2O → OHads + OH-aq
(12)
OHads + e- → OH-ads → OH-aq
(13)
Associative Mechanism (E ˂ 0.7 V): Since adsorbed O2 is present, the O-O bond may not be broken in the following steps, resulting in the formation of H2O2. The H2O2 could either be further reduced to H2O or be a final product. Therefore, the mechanism can be written as follows: O2,ads + 2H2O + 2e- → H2O2 + 2OH-
(14)
H2O2 + 2e- → 2OH-
(15) 13 Page 13 of 38
The further reduction of H2O2(ads) to hydroxide ions occurs only once the enough overpotential has been reached and before the formed H2O2 diffuse into the bulk solution as mentioned in cyclic voltammetry studies [31, 28]. The two loop manner of GDEs in Nyquist plots may be related to two basic steps. On the
ip t
other hand, for E ≥ 0.7 V, two time constants are detected during the impedance measurements (Fig. 5). The first time constant at high frequencies is associated with the
us
further reduction of OHads to produce OH- based Eq. 13.
cr
charge transfer reaction according to Eq. 12, while the second may be associated with the
On the other hand, for E ˂ 0.7 V, the first time constant is related to H2O2 intermediate
an
formation according to Eq. 14 and further semi-infinitive diffusive manner in the low frequency region can be explained by adsorption and diffusion of this intermediate into the
M
bulk solution. In order to obtain quantitative information from impedance spectra in Fig. 5, two electric circuits were employed (Fig. 6). The electric circuit in Fig. 6a was used to
d
simulate the impedance response of those spectra with OCP and E < 0.7 V, while the electric
te
circuit in Fig. 6b was used for those with E ≥ 0.7 V.
Ac ce p
Using the equivalent circuits shown in Fig. 6, a constant phase element (CPE) is suggested instead of pure capacitance (C), due to the non-homogeneous surface of the electrodes. The impedance of CPE is defined as [Yo(jω)n]-1, where Yo has (Ω-1.sn.cm-2) dimension, while the exponent n denotes the correction factor pertaining to the roughness of the electrode and has values that range from 0 to 1. A pure capacitance yields n=1, a pure resistance yields n=0, while n=0.5 represents the ideal Warburg impedance. The true capacitance values can be calculated using the following equation [41]: C=[Yo × R (1-n)] 1/n
(14)
According to equivalent circuits shown in Fig. 6, Rs is associated with the resistance of the solution, connectors, leads and wires. R1 is the charge transfer resistance of the reduction process from O2 to H2O2, R2 is the resistance of adsorbed species with H2O2 as the main 14 Page 14 of 38
intermediate or O2 adsorption into the GDE pores in the figure 6.a. The R1 and R2 circuit components in figure 6b are the charge transfer resistance of the reduction process of Oºads to OHads and the reduction process of OHads to OH-, respectively. Parameters calculated from the equivalent circuits (Fig. 6) of O2 reduction on the GDEs are listed in Table 3. According to
ip t
table 3, the Rs values change as a function of the potential, indicating that the contribution of the apparatus (connectors, leads and wires) from the total impedance of system shouldn’t be
cr
dismissed. So, the Rs values have a contribution function of both electrolyte and the apparatus
us
resistance, i.e., connectors, leads and wires.
The R1 and R2 also show a dependence on the applied dc potentials (Fig. 7). By increasing the
an
positive potential, the R1 values decrease for all of the GDEs. In the potential region of lower than 0.7 V, the adsorption of free O2 molecules happens on the metal catalyst and then O2,ads
M
reduces to OHads (Eq. 8). In the applied dc potential of E ≥ 0.7 V, the adsorption of Oºads free radicals happens. By increasing the positive applied potentials, the adsorbed amount of O2,ads
d
and Oºads increases and the further reduction process occurs more easily in the catalyst layer.
te
So the R1 values will be decreased.
Ac ce p
The R2 values for E ˂ 0.7 V are related to the adsorption of molecules like H2O2,ads into the GDE pores. When the potential is increased to more positive potentials, the adsorption of H2O2,ads species by the oxygen atom orientation to the metal catalysts will be increased and R2 values became higher. In the potential region of E ≥ 0.7 V, the OH,ads species which are reduced to OH-aq, increase and charge transfer happens more easily. So the R2 values will be decreased. Comparison of R values for GDEs shows that the Pt.Ru/C electrocatalyst has the lowest resistance in the whole range of applied dc potentials. The applied dc potentials in the range of higher or lower than 0.7 V, charge transfer reactions happen according to Eq. 12 and 14. Figure 7 shows that the R1 values of Pt.Ru/C are lowest among the GDEs. So, it can be concluded that the charges transfer more easily and rapidly in this ODC, either dissociative mechanism or associative mechanism happens, respectively. 15 Page 15 of 38
This phenomena happens in the situation that the amount of Pt nanoparticles used in the Pt.Ru/C catalysts is half of Pt/C catalyst one. For the applied dc potentials lower than 0.7 V, the R2 values are related to H2O2 or other intermediate adsorption on the active layer of GDEs. The Pt.Ru/C cathode shows the lower R2
ip t
values and it means that intermediates like H2O2 molecules desorbs more easily from this GEDs so it may be prevented from the active site blocking by these intermediates. For
cr
example, in the applied dc potential equal to 0.4 V, the R2 value of Pt/C, Ru/C and Pt.Ru/C
us
GDEs is 55.8, 62 and 17 Ω.cm2, respectively (Table 3). The R2 value of Pt/C GDE is 3.2 times greater than Pt.Ru/C one. The R2 values in higher potential than 0.7 V relate to charge
an
transfer reaction according to Eq. 13. With the potential of 0.9 V, the R2 values of Pt/C, Ru/C and Pt.Ru/C are 36, 42 and 25 Ω.cm2, respectively, and the R2 value of Pt.Ru/C cathode is 1.4
M
and 1.7 times lower than Pt/C and Ru/C GDEs. So, it can be concluded that the second charge transfer reaction happens more rapidly onto the bimetallic nano catalyst.
d
The dependence of true capacitance values of the applied potentials is shown in Fig. 7, too.
te
The plots show a gradual decrease of capacitances by increasing the potentials. The adsorbed
Ac ce p
species enhancement by increasing the more positive potentials causes to decrease of C1 and C2 values. The adsorption of species like O2 molecules increases by applying the more positive potential. So, according to C=ɛɛ0A/d [41], the values of d become greater and causes to C1 and C2 decreases. The true capacitance of Pt.Ru/C GDE also shows the highest amount among the other cathodes. The decrease of C1 and C2 values of Pt/C and Ru/C ODCs in the comparison of Pt.Ru/C is due to the reduction in the electrochemical active surface area of cathodes as mentioned before in the cyclic voltammetry results. As can be concluded according to the EIS and other technique results, the second metal addition has influenced the ORR rate. Some factors that may explain this electrocatalytic performance include: (a) the ratio of low coordinated surface atoms, which increases inversely with particle size; (b) the
16 Page 16 of 38
electronic state of small metal particles; (c) the strong metal-support interaction and (d) the synergistic effects of Pt and Ru catalyst species. 3.5. Chronoamperometery The durability and oxygen diffusion coefficients of GDEs were examined using
ip t
chronoamperometry technique. Chronoamperograms were obtained by holding the potential of the electrodes at +1.2 V for 10 s and then holding it at +0.4 V relative to the Ag/AgCl
cr
electrode for 4 h with oxygen flowing along the electrolyte. Fig. 8a shows the typical current
us
density-time (i-t) plots with the current normalized to the initial current (i/i0) to indicate the fractional decay with time. The current decay is rapid for the Pt/C and Ru/C electrodes,
an
reaching to 15% and 5% of the initial current in 1.5 and 1 h, respectively. In contrast, the Pt.Ru/C electrode needs 3.5 h to reach 20% of initial current. It can be seen clearly that the
M
current density on the Pt.Ru/C electrode is relatively higher than on the Pt/C and Ru/C GDEs, which suggest that paring the Pt and Ru decrease the opportunity of active electrocatalyst
d
particle dissolution from GDEs.
Ac ce p
[42]:
te
By plotting I vs. t−1/2, the linear dependence relationship was obtained for different electrodes
(16)
Where I is the limited current, A the surface area of the electrode, D the diffusion coefficient, C the concentration of oxygen, n the number of electrons in the overall reaction of ORR, F the Faraday’s constant, t the time, and is equal to 3.14. Fig. 8b shows the plot of current density vs. t-1/2 for GDEs in alkaline media. Cottrell parameters are listed in Table 2, also. The results confirmed the higher Cottrell slope and D values for Pt.Ru/C electrode. So, the Pt.Ru/C cathode has more permeability and activity towards oxygen reduction reaction. Conclusion In this investigation, Pt/C, Ru/C and Pt.Ru/C bimetallic electrocatalysts were prepared by the chemical reduction process. The SEM results showed that all the cathodes consist of rough 17 Page 17 of 38
and porous structures and Pt, Ru nanoclusters were deposited quite uniform onto/into Vulcan carbon supports with the average particle size of about 30-45 nm. The ORR activity of cathodes was evaluated in 0.1 M O2 saturated NaOH media. Comparison of cyclic voltamograms of Pt/C, Ru/C and Pt.Ru/C electrodes in O2 saturated solutions indicated that
ip t
Pt.Ru/C catalyst has higher current density and more positive ORR onset potential in the compare of Pt/C and Ru/C electrodes. It can be concluded that the Ru addition has influenced
cr
the catalytic activity of electrocatalyst toward ORR. This matter can be related to synergistic
us
effect, which is playing a critical role in ORR activity. The Pt.Ru/C cathodes showed lower Tafel slope and high current density. An EIS study was carried out at different dc potentials in
an
order to study the mechanism of ORR in O2 saturated alkaline solutions on prepared GDEs. The Nyquist plots of GDEs under different polarization potentials showed two different
M
behaviors, suggesting different associative and dissociative electrochemical processes
te
Acknowledgements
d
occurring at the electrodes.
Ac ce p
The authors would like to acknowledge the financial support of the Office in Charge of Research of Iranian Nanotechnology Society and the financial support of the Office of Vice Chancellor in charge of research of University of Tabriz. Reference:
1. G. G. Botte, Electrochemical Manufacturing in the Chemical Industry, Electrochem. Soc. Interface. 23 (2014) 49-55. 2. Y.Kiros, M. Pirjamali, M. Bursell, Oxygen reduction electrodes for electrolysis chlor-alkali cells, Electrochim. Acta. 51 (2006) 3346-3350. 3. Thomas F. O'Brien, Tilak V. Bommaraju, Fumio Hine, History of the Chlor-Alkali Industry, in Handbook of chlor-alkali technology, vol. 1, Springer, New York, NY, USA, 2004. 4. I, Moussallem, S. Pinnow, N. Wagner, T. Turek, Devolepment of high-performance silverbased gas-diffusion electrodes for chlor-alkali electrolysis with oxygen depolarized cathodes, Chem. Eng. Process. 52 (2012) 125-131. 18 Page 18 of 38
5. T. Mirzazadeh, F. Mohammadi, M. Soltanieh, E. Joudaki, Optimization of caustic current efficiency in a zero-gap advanced chlor-alkali cell with application of genetic algorithm assisted by artificial neural networks, Chem. Eng. J. 140 (2008) 157–164. 6. F. Mohammadi, S.N. Ashrafizadeh, A. Sattari, Aqueous HCl electrolysis utilizing an oxygen reducing cathode, Chem. Eng. J. 155 (2009) 757-762.
ip t
7. E. Joudaki, F. Farzami, V. Mahdavi,S. J. Hashemi, Performance Evaluation of OxygenDepolarized Cathode with PtPd/C Electrocatalyst Layer in Advanced Chlor-Alkali Cell, Chem. Eng. Technol. 33 ( 2010) 1525-1530
cr
8. S. Siracusano, T. Denaro, V. Antonucci, A. S. Aricò, C. Urgeghe, F. Federico, Degradation of oxygen-depolarized Ag-based gas diffusion electrodes for chlor-alkali cells, J. Appl. Electrochem. 38 (2008) 1637-1646.
us
9. Ludwig Lipp, Shimshon Gottesfeld, Jerzy Chlistunoff, Peroxide formation in a zero-gap chlor-alkali cell with an oxygen-depolarized Cathode, J. Appl. Electrochem. 35 (2005) 1015– 1024.
an
10. J. Chlistunoff, Advanced Chlor-Alkali Technology: Final Technical Report, Los Alamos National Laboratory, New Mexico, United States 2004.
M
11. M. Sugiyama, K. Saikai, A. Sakata, H. Aikawa, N. Furuya, Accelerated degradation testing of gas diffusion electrodes for the chlor-alkali process , J. Appl. Electrochem. 33 (2003) 929–932.
te
d
12. S. Chabi, M. Kheirmand, Electrocatalysis of oxygen reduction reaction on Nafion/platinum/gas diffusion layer electrode for PEM fuel cell, Appl. Surf. Sci. 257 (2011) 10408-10413.
Ac ce p
13. M. Hezarjaribi, M. Jahanshahi, A. Rahimpour, M. Yaldagard, Gas diffusion electrode based on electrospun Pani/CNF nanofibers hybrid for proton exchange membrane fuel cells (PEMFC) applications, Appl. Surf. Sci. 295 (2014) 144– 149. 14. N. Ramaswamy, S. Mukerjee, Fundamental mechanistic understanding of electrocatalysis of oxygen reduction on Pt and non-Pt surfaces: acid versus alkaline media, Adv. Chem. Phys. (2012) 1. 15. T. R. Ralph and M. P. Hogarth, Catalysis for Low Temperature Fuel Cells, Platinum Metals Rev, 46 (2002) 3-14. 16. Sanjeev Mukerjee, Supramaniam Srinivasan, Enhanced electrocatalysis of oxygen reduction on platinum alloys in proton exchange membrane fuel cells, J. Electroanal. Chem. 357 (1993) 201-224. 17. U.A. Paulus, A. Wokaun, G.G. Scherer, T.J. Schmidt, V. Stamenkovic, N.M. Markovic, P.N. Ross, Oxygen reduction on high surface area Pt-based alloy catalysts in comparison to well defined smooth bulk alloy electrodes, Electrochim. Acta. 47 (2002) 3787-/3798. 18. V. Stamenkovic´, T.J. Schmidt, P.N. Ross, N.M. Markovic, Surface segregation effects in electrocatalysis: kinetics of oxygen reduction reaction on polycrystalline Pt3Ni alloy surfaces, J. Electroanal. Chem. 554-555 (2003) 191-199. 19 Page 19 of 38
19. U. A. Paulus, A. Wokaun, and G. G. Scherer, T. J. Schmidt, V. Stamenkovic, V. Radmilovic, N. M. Markovic, and P. N. Ross, Oxygen Reduction on Carbon-Supported Pt-Ni and Pt-Co Alloy Catalysts, J. Phys. Chem. B. 106 (2002) 4181-4191.
ip t
20. V. Stamenkovic´, T. J. Schmidt, P. N. Ross, and N. M. Markovic, Surface Composition Effects in Electrocatalysis: Kinetics of Oxygen Reduction on Well-Defined Pt3Ni and Pt3Co Alloy Surfaces, J. Phys. Chem. B. 106 (2002) 11970-11979.
cr
21. J. Zhang, Y. Mo, M. B. Vukmirovic, R. Klie, K. Sasaki, and R. R. Adzic, Platinum monolayer electrocatalysts for O2 reduction: Pt monolayer on Pd(111) and on carbonsupported Pd nanoparticles, J. Phys. Chem. B. 2004, 108, 10955-10964.
us
22. C. Jeyabharathi, P. Venkateshkumar, J. Mathiyarasua, K.L.N. Phani, Platinum–tin bimetallic nanoparticles for methanol tolerant oxygen-reduction activity, Electrochim. Acta. 54 (2008) 448-454.
an
23. A. Nozad Golikand, E. Lohrasbi, M. Ghannadi Maragheh, M. Asgari, Enhancement of electrocatalytic O2 reduction on carbon nanotube-supported Pt alloys nanoparticles in gas diffusion electrodes, J Appl Electrochem. (2009) 39:1443-1449.
d
M
24. T. Frelink, W. Visscher, A. P. Cox, J. A. R. Van veen, Ellipsometry and dems study of the electrooxidation of methanol at Pt and Ru- and Sn- promoted Pt, Electrochim. Acta. 40 (1995) 1537-1543.
te
25. D. Chu, S. Gilman, Methanol Electro‐oxidation on Unsupported Pt‐Ru Alloys at Different
Ac ce p
Temperatures, J. Electrochem. Soc. 143 (1996) 1685-1690. 26. A.L. Ocampo, M. Miranda-Hern´andez, J. Morgado, J.A. Montoya, P.J. Sebastian, Characterization and evaluation of Pt-Ru catalyst supported on multi-walled carbon nanotubes by electrochemical impedance, J. Power Sources, 160 (2006) 915-924. 27. E. Lust,1, E. Härk, J. Nerut, K. Vaarmets, Pt and Pt–Ru catalysts for polymer electrolyte fuel cells deposited onto carbide derived carbon supports, Electrochim. Acta. 101 (2013) 130141. 28. C. Song, J. Zhang, Electrocatalytic Oxygen Reduction Reaction in PEM Fuel Cell Electrocatalysts and Catalyst Layers, Springer, Verlag London Limited. (2008) chapter 2, pp: 89-134. 29. A.J. Bard, L.R. Faulkner, Electrochemical Methods: Fundamentals and Applications, second ed., Wiley, NY, 2001. 30. Malika Ammam, E. Bradley Easton, Oxygen reduction activity of binary PtMn/C, ternary PtMnX/C (X= Fe, Co, Ni, Cu,Mo and, Sn) and quaternary PtMnCuX/C (X= Fe, Co, Ni, and Sn) and PtMnMoX/C (X= Fe, Co, Ni, Cu and Sn) alloy catalysts, J. Power Sources, 236 (2013) 311-320. 20 Page 20 of 38
31. Rosa Rego, A.M. Ferraria, A.M. Botelho do Rego, M. Cristina Oliveira, Development of PdP nano electrocatalysts for oxygen reduction reaction, Electrochim. Acta. 87 (2013) 73-81 32. C. Wang, B. Peng, H. N. Xie, H.X. Zhang, F. F. Shi, W. B. Cai, Facile Fabrication of Pt, Pd and Pt−Pd Alloy Films on Si with Tunable Infrared Internal Reflection Absorption and Synergetic Electrocatalysis, J. Phys. Chem. C. 113 (2009) 13841-13846.
ip t
33. K. Kakaei, Electrochemical Characteristics and Performance of PlatinumNanoparticles Supported by Vulcan/Polyaniline for OxygenReduction in PEMFC, Fuel cells. 12 (2012) 939945
cr
34. N. R. Elezovic, B. M. Babic, LJ. M. Vracar, N.V. Krstajic, Oxygan reduction at platinum nanoparticles supported on carbon cryogel in alkalin solution, J. Serb. Chem. Soc. 72 (2007) 699-708.
an
us
35. T. Lopes, E. Antolini, E.R. Gonzalez, Carbon supported Pt–Pd alloy as an ethanol tolerant oxygen reduction electrocatalyst for direct ethanol fuel cells, Int. J. Hydrogen Energy. 33 (2008) 5563 -5570.
M
36. N. Ramaswamy, S. Mukerjee, FundamentalMechanistic Understanding of Electrocatalysis of Oxygen Reduction on Pt and Non-Pt Surfaces: Acid versus AlkalineMedia, Adv. Phys. Chem. 2012 1-12. 37. N.A. Anastasijevic, Z.M. Dimitrijevic, R.R. ADzic, Oxygen reduction on a ruthenium electrode in alkalin electrolytes, .J Electroanal. Chem. 199 (1986) 351-364.
te
d
38. E. H. Yu, S. Cheng , Bruce E. Logan, K. Scott, Electrochemical reduction of oxygen with iron phthalocyanine in neutral media, J Appl Electrochem, 39 (2009) 705-711.
Ac ce p
39. V. P. Zhdanov, B. Kasemo, Kinetics of electrochemical O2 reduction on Pt, Electrochem. Commun. 8 (2006) 1132–1136. 40. J. K. Norskov, J. Rossmeisl, A. Logadottir, L. Lindqvist, Origin of the Overpotential for Oxygen Reduction at a Fuel-Cell Cathode, J. Phys. Chem. B. 108 (2004) 17886-17892. 41. M. Kissi, M. Bouklah, B. Hammouti, M. Benkaddour, Establishment of equivalent circuits from electrochemical impedance spectroscopy study of corrosion inhibition of steel by pyrazine in sulphuric acidic solution, Appl. Surf. Sci. 252 (2006) 4190-4197. 42. J. Ribeiro, A. R. de Andrade, Investigation of the electrical properties, charging process, and passivation of RuO2–Ta2O5 oxide films, J. Electroanal. Chem. 592 (2006) 153-162.
21 Page 21 of 38
ip t
Table. 1. Surface composition of GDEs by EDX analysis.
cr
Table caption
recorded in 0.1 M O2 saturated NaOH solution.
us
Table. 2. The electrochemical active surface area, Tafel and Cottrell parameters of GDEs,
an
Table. 3. Impedance parameter values of GDEs at different dc potentials; recorded in 0.1 M
Ac ce p
te
d
M
O2 saturated NaOH solution.
22 Page 22 of 38
ip t
Figure caption
cr
Fig. 1. Schematic illustration of the three-electrode compartment set up for electrochemical
us
tests.
Fig. 2. SEM surface morphologies of Pt/C (a), Ru/C (b), Pt.Ru/C (c) and EDX analysis of
an
Pt/C (d), Ru/C (e), Pt.Ru/C (f) GDEs.
Fig. 3. Cyclic voltammograms of Pt/C (a), Ru/C (b), Pt.Ru/C (c) GDEs in 0.1 M N2 saturated
M
(solid line) and O2 saturated (dash line) NaOH solution at a scan rate of 50 mV.s-1. Fig. 4. Comparison of polarization curves for the ORR onto the GDEs in 0.1M O2 saturated
d
NaOH solution and 1mV.s-1 sweep rate.
te
Fig. 5. Impedance diagrams in the complex plane of Pt/C (a), Ru/C (b), Pt.Ru/C cathodes as a
Ac ce p
function of applied potentials, recorded in 0.1 M O2 saturated NaOH media. Fig. 6. The electrical equivalent circuits (EEC) used for the fitting of the experimental results presented in Fig. 5.
Fig. 7. Dependence of the R1, C1 (a) and R2, C2 (b) values of the GDE as a function of different applied potentials; recorded in 0.1 M O2 saturated NaOH media. Fig. 8. Normalized current verse time (a) and current density verse t−1/2 plots (b) of the GDEs in 0.1 M NaOH solution with constant potential +0.4 V (vs. Ag/AgCl) during 4 h.
23 Page 23 of 38
C 64.53 56.79 51.8
W (%) Pt 35.47 24.38
Ru 43.21 23.82
Ac ce p
te
d
M
an
us
cr
catalyst Pt/C Ru/C Pt.Ru/C
ip t
Table. 1. Surface composition of GDEs by EDX analysis.
24 Page 24 of 38
NA (×1016 )
Pt/C Ru/C Pt.Ru/C
7.2 5.5 19.8
0.68 0.1 0.75
91 78 253
Exchange current density (A.cm-2) 1.34×10-3 1.28×10-3 2.05×10-3
Tafel slope (mV/dec)
Cottrell slope (A.s-1/2)
CD1/2 (mol.cm-2.s1/2 )
136.4 150.6 122.3
1.39 × 10-6 3.80 × 10-7 2.6 × 10-6
ip t
E onset (V)
0.304 0.083 0.569
us
EASA (cm2)
Ac ce p
te
d
M
an
catalyst
cr
Table. 2. The electrochemical active surface area, Tafel and Cottrell parameters of GDEs, recorded in 0.1 M O2 saturated NaOH solution.
25 Page 25 of 38
Table. 3. Impedance parameter values of GDEs at different dc potentials; recorded in 0.1 M O2 saturated NaOH solution. Rw (Ω.cm2) 1917 341.6 117.2 1610 267.8 116 357.5 139.4 110.3 -
Yw (Ω-1.sn.cm-2) 13.53 9.17 11.3 30.95 14.60 9.49 36.79 39.47 20.62 -
ip t
0.97 0.98 0.96 0.91 0.82 0.91 0.95 0.96 0.83 0.85 0.87 0.95 0.98 0.81 0.89
R2 (Ω.cm2) 86.30 55.80 40.20 40.80 36.00 92.00 62.00 49.00 47.00 42.00 16.00 17.00 20.00 28.00 25.00
cr
ndl
us
0.66 0.61 0.65 0.67 0.69 0.78 0.85 0.86 0.75 0.79 0.75 0.73 0.75 0.75 0.73
Y0dl (Ω-1.sn.cm-2) 0.13 0.12 0.11 0.12 0.13 0.73 0.21 0.17 0.51 0.39 0.50 0.41 0.44 0.54 0.46
an
R1 (Ω.cm2) 5.8 5.6 5.5 5.4 5.2 5.4 5.3 5.2 5.1 5.1 5.0 4.9 4.9 4.8 4.7
np
M
Pt.Ru/C
Y0f (Ω-1.sn.cm-2) 11 19 12 10 8.6 72 31 62 4 3.1 3.2 4.6 3.9 4.0 4.6
nw 0.79 0.69 0.64 0.93 0.69 0.65 0.82 0.62 0.61 -
d
Ru/C
Rs (Ω.cm2) 7.2 7.2 7.3 7.3 7.3 6.8 7.1 6.9 6.9 7.1 11.3 10.9 10.8 10.7 11.0
te
Pt/C
E (V) OCP 0.4 0.6 0.7 0.9 OCP 0.4 0.6 0.7 0.9 OCP 0.4 0.6 0.7 0.9
Ac ce p
sample
26 Page 26 of 38
ip t
Ac ce p
te
d
M
an
us
cr
Fig. 1.
27 Page 27 of 38
28
Page 28 of 38
d
te
Ac ce p us
an
M
cr
ip t
29
Page 29 of 38
d
te
Ac ce p us
an
M
cr
ip t
ip t cr us an M d te Ac ce p
Fig. 2.
30 Page 30 of 38
31
Page 31 of 38
d
te
Ac ce p us
an
M
cr
ip t
ip t cr us an
Ac ce p
te
d
M
Fig. 3.
32 Page 32 of 38
ip t cr us an
Ac ce p
te
d
M
Fig. 4.
33 Page 33 of 38
ip t cr
70
(a)
an
40
5
30
10
15
20 OCP
M
-Z'' (Ω.cm2)
50
us
14 12 10 8 6 4 2 0
60
20 10
0.4V 0.6V 0.7V
d
0.9V
0
10
20
30
40
50
60
70
2
Z' (Ω.cm )
Ac ce p
te
0
30
(b)
10
25
5
-Z'' (Ω.cm2)
20
0
15
5
10
15 OCP
10
0.4V 0.6V
5
0.7V 0.9V
0 5
10
15
20
25
30
2
Z' (Ω.cm )
34 Page 34 of 38
ip t
30
15
20
5
0
15
cr
10
us
25
2
-Z'' (Ω.cm )
(c)
15
an
10
0
15
Ac ce p
te
10
d
5
20
25
OCP 0.4V 0.6V
M
10
20
0.7V 0.9V
25
30
35
40
2
Z' (Ω.cm )
Fig. 5.
35 Page 35 of 38
ip t cr us
Ac ce p
te
d
M
an
Fig. 6.
36 Page 36 of 38
12
35 Pt.Ru/C
Pt/C
Ru/C
Pt.Ru/C
25
15
8
ip t
R1 (Ω.cm2)
10
Ru/C
C1 (mF/cm2)
(a)
Pt/C
5
cr
6
0.2
0.4
0.6
an
E/V (Ag/AgCl)
160
M
(b)
0.8
0.5
Pt/C
Ru/C
Pt.Ru/C
Pt/C
Ru/C
Pt.Ru/C
te
70
-0.1
40
-0.3
10
0
0.2
0.3
0.1
d
100
Ac ce p
R2 (Ω.cm2)
130
-15
C2 (mF/cm2)
4 OCP 0
us
-5
-0.5 0.4 0.6 E/V (Ag/AgCl)
0.8
Fig. 7.
37 Page 37 of 38
1 (a)
Pt/C Ru/C Pt.Ru/C
ip t
0.50
cr
0.25 -0.5
0-1 0
0.5
1
1.5
2
2.5
3
3.5
4
an
Time (h)
30
M
(b) 25
d
20
15
te
J (mA/cm2)
us
Normalized current
0.75 0.5
Pt.Ru/C
10
Pt/C
Ac ce p
Ru/C
5
0
0
0.01
0.02
0.03 t
-1/2
0.04 (s
0.05
0.06
0.07
-1/2
)
Fig. 8.
38 Page 38 of 38