Influence of Inner- and Outer-Sphere Electron Transfer Mechanisms

ARTICLE pubs.acs.org/JPCC

Influence of Inner- and Outer-Sphere Electron Transfer Mechanisms during Electrocatalysis of Oxygen Reduction in Alkaline Media Nagappan Ramaswamy and Sanjeev Mukerjee* Department of Chemistry and Chemical Biology, Northeastern University Center for Renewable Energy Technology, 360 Huntington Avenue, Northeastern University, Boston, Massachusetts 02115, United States

bS Supporting Information ABSTRACT: Oxygen reduction reaction (ORR) is generally considered to be more facile in alkaline media compared to its acidic counterparts. The fundamental reasoning for this statement has been quite elusive and not understood very well. A pertinent review of the literature in alkaline media on noble and non-noble metal electrocatalysts is presented here along with experimental results to investigate the rationale behind the so-called kinetic facility in alkaline media. Increasing the pH from 0 to 14 has several effects on the electrode
electrolyte interface in terms of the working electrode potential range, the strength of adsorption of the reaction intermediates, and spectator species. Besides these, the reasons for kinetic facility are investigated from the perspective of the changes in the double layer structure and electrochemical reaction mechanisms in transitioning from acidic to alkaline environment. In this context, specifically adsorbed hydroxyl species are found to promote an outer-sphere electron transfer ORR mechanism in alkaline media. A surface independent outer-sphere electron transfer component is proposed to be the reason for the so-called facile kinetics of ORR in alkaline media on a wide range of non-noble metal surfaces. However, this outer-sphere process predominantly leads only to a 2e
peroxide intermediate as the final product. The importance of promoting the electrocatalytic inner-sphere electron transfer mechanism by facilitation of direct adsorption of molecular oxygen and the stabilization of the peroxide intermediate on the active site are emphasized with the usage of chalcogen modified transition metals and pyrolyzed biomimetic metal porphyrins as electrocatalysts.

1. INTRODUCTION Oxygen reduction reaction (ORR) on noble and non-noble metal surfaces remains one of the well investigated electrochemical processes. This interest stems from both a technological and fundamental standpoint. Under acidic conditions, Pt and Pt alloys remain the mainstay as catalyst materials for ORR due to the acid stability criterion; in alkaline electrolyte, a wide range of non-noble metals and their oxides are understood to be stable enough for practical applications. Alkaline fuel cells (AFC), once considered very promising, failed to attract continued research interest primarily due to issues such as carbonate precipitation (and therefore the onerous need for electrolyte scrubbers) and electrolyte leakage.1 Further, the rapid growth of proton exchange membrane (PEM) fuel cells shifted research interest into the acidic counterpart. Despite worldwide research effort in PEMFC, widespread commercialization is strongly predicated on component costs and striking an optimum balance between performance and durability. Appleby2 in 1970 envisaged the improbability of increasing the rate constants for ORR on noble metals due to the compensating changes between the preexponential factors and heat of activation, and apparently this scenario has not changed significantly since then. Further, the poor selectivity of Pt materials in the presence of impurities and fuel crossover aggravates electrocatalysis in PEMFC. Realization of the fact that a wider range of non-Pt based catalyst materials r 2011 American Chemical Society

can be employed at high pH environments and recent research efforts in developing metal cation-free alkaline anion exchange membranes (AAEM) for hydroxide anion transport that does not suffer from carbonate precipitation have spurred research activities in AFCs.3
6 This is despite (i) the 2
3 orders of magnitude lower hydroxide anion conductivity in state of the art alkaline membranes compared to proton transport in acidic membranes7 and (ii) manifestation of poisoning due to carbonate anion exchange in alkaline membranes.8 The claim that AFC performs better than PEMFC remains quite unjustified because such claims have usually been based on comparisons between phosphoric acid fuel cells and AFCs.9 On the contrary, any comparison of performance between PEMFC and liquid electrolyte based AFC technologies roughly exhibits equivalent performances.9,10 However, the applicability of non-Pt materials in alkaline media is completely justified given the ample evidence in the literature employing several non-noble materials as electrocatalysts and the dispensability of the so-called acidstability criterion at high pH environments.1,11,12 Although several ambitious engineering designs were tested to prevent leakage and carbonate precipitation issues while improving AFC Received: May 19, 2011 Revised: July 21, 2011 Published: August 05, 2011 18015

dx.doi.org/10.1021/jp204680p | J. Phys. Chem. C 2011, 115, 18015–18026

The Journal of Physical Chemistry C performances, no revival of the prospects of AFC technology is noticed.1,13,14 Replacement of liquid electrolytes in conventional AFC with metal cation free AAEM that can transfer hydroxide ions (OH
) can revitalize the AFC technology and impart a new momentum to it.3,4 ORR pathway rather than the ORR mechanism has typically been addressed in the literature due to the easy accessibility of the former from rotating ring-disk electrode (RRDE) based studies and the complexity in understanding the latter on the basis of electrochemical and spectroscopic results.15,16 Pt and Pt alloy based catalysts remain well investigated primarily from electrochemical kinetic studies on single crystals and in situ X-ray absorption spectroscopic studies.17,18 ORR pathways are found to be similar in both acid and alkaline media on Pt based materials.15,19 On the basis of the initial propositions by Damjanovic et al.,19
21 the rate determining step on Pt electrodes is widely agreed to be the first electron transfer step to the adsorbed molecular O2 with or without rapid proton transfer.19
21 The analogous kinetic studies involve the elucidation of relationships of dE/d(log i), dE/d(log pO2), and dE/d(pH).19,20 It is also understood that the ORR kinetics are qualitatively different on prereduced, oxide free electrodes and electrodes that contain a thin film of oxide coverage on it.22
24 Two Tafel slope regions in both acid and alkaline media are typically observed. The Tafel slope of
2.3RT/F (
60 mV/dec) obtained in the low current density region follows the Temkin adsorption isotherm due to intermediate oxide coverage arising from ORR reaction intermediates. In the high current density region, the Tafel slope of
2  2.3RT/F (
120 mV/dec) is governed by the Langmuir isotherm because significant oxide coverage ceases to exist at these potentials.20 Tarasevich et al.25 later pointed out that the adsorbed OH species on Pt surfaces that inhibit O2 adsorption arise primarily from water activation whereas the ORR reaction intermediates typically exhibit lower coverage values compared to water activation products. Over all the pH range and entire current density region it is known that the reaction order with respect to molecular O2 is unity.20 Also, from the kinetic studies it is now known that the reaction order of the rate determining step with respect to H3O+, and OH̅ (i.e., dE/d(pH)), is respectively 3/2 and
1/2 in the low current density region.20 In the high current density region, it is observed that the reaction order with respect to H3O+, and OH̅ (dE/d(pH)), is respectively 1 and 0. This unusual “fractional” reaction order with respect protons or hydroxyl ions is the manifestation of the dependence of free energy of activation (ΔG*) on the intermediate oxide coverage (as governed by Temkin adsorption isotherm) and, hence, on the pH of the electrolyte. For a recent review of the kinetics, see Spendelow et al.,6 Gottesfeld, 26 and Adzic et al. 27 Further a combination of electrochemical and spectroscopy studies has shown that, Pt
M alloys (M = Co, Ni, Cr, etc.) show improved kinetics due to a combination of geometric and electronic factors that serve to inhibit OH poisoning of the Pt sites.18,28
32 A major alternative viewpoint to the rate determining step in ORR was proposed by Yeager et al.,33 wherein it was proposed that ORR on Pt surfaces is likely to involve dissociative chemisorption of molecular O2 with the initial adsorption of O2 with or without an electron transfer as the rate determining step. Also, on the basis of hydrogen
deuterium kinetic isotope studies, the rate determining step was not likely to involve proton transfer. In gas phase studies there are several indicators for dissociative chemisorption of molecular O2.15 However, under electrochemical

ARTICLE

conditions where solvation effects and other surface adsorbed species (OHads, electrolyte anions) are present, adsorption of molecular O2 is likely to be weakened, adding evidence to the proposition of the rate determining step put forth by Yeager et al.33 ORR on Pt based catalysts is understood to proceed via “parallel” routes with the 4e
“direct” or “series” pathway as the predominant route and a minor route involving the 2e
pathway to peroxide. This parallel generation of peroxide is mainly related to the oxides, anions, and impurities on the surface that weaken the adsorption of the peroxide intermediate.15,32 The RRDE technique has been extensively used to understand the kinetics of ORR.34 This technique involves the convective transport of dissolved molecular oxygen from the bulk to the electrode surface prior to diffusive transport within the diffusion layer to the catalyst site. This technique allows the detection of stable reaction intermediates generated at the disk during ORR by potential control of the ring electrode. Various kinetic models have been developed to understand the reaction pathways involved during ORR. The first model was developed by Damjanovic et al.35 following which Wroblowa et al.36 and Anastasijevic et al.,37,38 proposed extensive models. Briefly there are two 4e
pathways and one 2e
pathway. 4e
transfer could be either a direct 4e
or a series 4e
pathway. The direct pathway involves the concerted transfer of 4e
to adsorbed molecular oxygen to form H2O without the formation of peroxide intermediates. The series 4e
pathway involves sequential transfer of electrons to adsorbed molecular oxygen to form the adsorbed peroxide species, which without desorbing from the surface is involved in another 2e
transfer to form water. Although these two pathways are rather ideal definitions of the two possible 4e
pathways, it is important to understand the following aspects of this distinction as “direct” and “series”. The direct 4e
transfer requires the concerted transfer of 4e
and 4H+ to the dissociatively chemisorbed molecular oxygen. Evidence for dissociative adsorption of O2 under electrochemical conditions is not available in the literature, making this pathway very less likely. So it is quite possible that the so-called “direct” 4e
pathway proceeds via the peroxide pathway such that the peroxide intermediate does not desorb from the catalyst surface to any appreciable extent. So in this particular case, the ring-disk kinetic studies are incapable of making the distinction between a direct 4e
pathway and a series pathway involving 4e
transfer.25 The 2e
pathway involves the transfer of two electrons to the adsorbed molecular oxygen, forming the peroxide species, and the peroxide intermediate diffuses to the electrolyte bulk without any further reduction. This is the case where peroxide is the final product and the catalyst is incapable of reducing the peroxide intermediate any further. Further an “interactive” pathway was also defined to treat catalyst materials exhibiting heterogeneity in active sites where intermediate species undergo surface diffusion and undergo further reduction at more active sites.37,38 Given the pKa values for the first and second ionization of H2O2 at 25 °C (pK1 = 11.69 and pK2 = ∼20) the predominant peroxide species for pH > 12 is HO2
.25 Further, another important distinction in alkaline media is that in a 4e
“series” pathway, the lower working electrode potential range on an absolute scale causes the HO2
intermediate to desorb from the catalyst surface; however, in the analogous case of acidic media the higher working electrode potential range decreases the possibility of H2O2 desorption from the electrode surface.6 As mentioned by Appleby,39 alkaline electrolyte essentially acts as a homogeneous 18016

dx.doi.org/10.1021/jp204680p |J. Phys. Chem. C 2011, 115, 18015–18026

The Journal of Physical Chemistry C catalyst due to its stabilization effect on the intermediate product HO2
. For electrocatalytic reactions proceeding via inner-sphere electron transfer mechanism, it is typically assumed that either molecular adsorption of reactant species (dissociatively or nondissociatively) or an electron transfer is the first step.40 For neutral, nonpolar species like molecular O2, direct molecular O2 adsorption is likely to be inhibited relative to, for example, the adsorption of superoxide radical anion (O2•
) unless the free energy of adsorption of O2 molecule is very exothermic on a specific catalytic surface. This is especially true under fuel cell conditions where the cathodic reaction typically occurs at potentials well positive of the potential of zero charge (pzc). Multistep, multielectron transfer processes like ORR that involve many adsorbed intermediates undoubtedly classify as an innersphere electron transfer reaction. However, among the many elementary reaction steps involved in ORR there could be a surface-independent outer-sphere electron transfer component in the overall electrocatalytic 4e
inner-sphere electron transfer reaction. In that perspective O2 reduction by one-electron transfer to superoxide (O2•
) is observed at E° =
0.3 ( 0.03 V vs SHE corresponding to ΔG° = 30 ( 2 kJ mol
1 with both O2 and O2•
remaining in the aqueous phase.41,42 Given the pH independence of this redox couple (O2/O2•
), the potential of this reaction does not change as the pH is varied from 0 to 14.43 Due to the occurrence of four proton transfer steps in O2 reduction to H2O/ OH
, its standard reduction potential changes by 0.828 V from 1.229 to 0.401 V vs SHE as the pH value changes from 0 to 14. This causes the overpotential for the first electron transfer step (O2/O2•
) to decrease from 1.53 V at pH = 0 to 0.7 V at pH = 14, indicating a sharp decrease in overpotential at alkaline pH conditions. Markovic et al.43 argued on the basis of a modified Pourbaix diagram approach that the above-mentioned decrease in overpotential is the primary thermodynamic reason for the applicability of a wide range of non-noble materials in alkaline media. Due to the high overpotential required for the O2/O2•
redox couple in acidic media, only certain specific catalyst surfaces such as platinum that offer high free energy of adsorption for O2 can catalyze ORR in acidic media. In alkaline media, a decrease in overpotential for O2/O2•
causes almost all electronically conducting electrode material to be ORR active at alkaline pH.43 Though the decrease in overpotential for the first electron transfer is certainly significant, this argument is primarily of thermodynamic origin. The concept of involving the possibilities of outer-sphere electron transfer during ORR in alkaline media bears importance, and it was pointed out earlier by Bockris1 and Appleby39 that the exchange current density values in alkaline media exhibit near-independence on a large number of electrode materials including silver, gold, manganese oxides, perovskites, and various carbon surfaces. So certain steps in the overall ORR process in alkaline media could proceed via a nonelectrocatalytic pathway.39 In this report we investigate the kinetics of ORR in alkaline media from the perspective of the reaction mechanisms and the double layer structure. Although the fundamental electrochemical aspects of ORR on Pt in acidic media have been thoroughly investigated by various research groups and continue to be a subject of intense study, oxygen reduction on Pt in alkaline media exhibit certain interesting behaviors that have not been discussed previously in detail in the literature. An outer-sphere electron transfer mechanism in alkaline media is found to be responsible for the undesired 2e
peroxide intermediate. The neccessity to

ARTICLE

promote the inner-sphere electrocatalytic process by facilitating direct molecular oxygen adsorption is emphasized with the application of chalcogen modified transition metals, or biomimetic metal macrocycles. A combination of pertinent review of the literature along with experimental results shown here are used to unravel the various possible ORR reaction mechanisms in alkaline media.

2. ELECTROCHEMICAL CHARACTERIZATION All electrochemical measurements were made at room temperature using a RRDE setup from Pine Instruments connected to an Autolab (Ecochemie Inc., model-PGSTAT 30) bipotentiostat. Alkaline (0.1 M NaOH) and acidic (0.1 M HClO4) electrolytes were prepared using sodium hydroxide pellets (semiconductor grade, 99.99%, Sigma-Aldrich) and double-distilled 70% perchloric acid (GFS Chemicals) respectively. 30% Pt/C catalyst from BASF-ETEK (Somerset, NJ) was used as received. Ru/C and Se/Ru/C were synthesized in-house at a metal loading of 20% by weight via aqueous routes.71 Iron(III) meso-tetraphenylporphyrin chloride (FeTPPCl) was procured from Alfa Aesar and used as received. FeTPPCl was mixed with Black Pearl carbon in the mass ratio 1:4 and ball milled for 2 h at 400 rpm followed by pyrolysis at 800 °C for 2 h under argon atmosphere. Catalyst inks were prepared by dispersing 25 mg of the catalyst in 10 mL of 1:1 Millipore H2O:isopropyl alcohol mixture along with 100 μL of 5 wt % Nafion solution as a binder. A 10 μL aliquot of the catalyst ink was dispensed on a 5.61 mm diameter glassy carbon (GC) disk. Gold ring electrode was held at 1.1 V vs RHE in alkaline electrolyte and at 1.3 V vs RHE in acidic electrolyte to detect stable peroxide intermediate. Collection efficiency of the disk-ring electrode was 37.5%. All potentials are referred to reversible hydrogen electrode (RHE) scale prepared from the same solution as the bulk electrolyte unless otherwise stated. More details on the RRDE methodology can be found in our previous publications.32 3. RESULTS AND DISCUSSIONS 3.1. Structural Aspects of the Double-Layer. As mentioned above in traversing a pH range of 14 from acidic to alkaline media, the working electrode potential decreases by about ∼0.83 V on the basis of the Nernst equation. Spendelow et al.6 pointed out that the change in electrode potential is likely to have significant consequences on the energetics of adsorption of reactants, intermediates, and products. Further, this decrease in potential also affects the strength of adsorption of spectator species, contaminants, and anions from the supporting electrolytes.6 Strong poisons in acidic media such as halide anions do not poison the platinum active sites in alkaline media. This is primarily due to the fact that the lower working electrode potential in alkaline media causes the excess surface charge on the electrode to be relatively more negative than that in acidic media. This excess negative charge repels the chloride anions away from the inner-Helmholtz plane (IHP). This is also one reason why typically higher peroxide anion (HO2
) intermediate is detected at the ring in alkaline media compared to the neutral H2O2 intermediate in acidic media.6 Besides the effect of working electrode potential on adsorption strength, it is likely that there is a significant change in the doublelayer structure at the electrode/electrolyte interface as the pH is 18017

dx.doi.org/10.1021/jp204680p |J. Phys. Chem. C 2011, 115, 18015–18026

The Journal of Physical Chemistry C

ARTICLE

only act as solvent but also serve as the source of protons required in ORR. IHP is now populated by specifically adsorbed hydroxyl species (OHads arising from OH
anion adsorption), solvent water dipoles, and chemisorbed O2. Alkali metal ions are typically well solvated and are classically expected to populate the OHP. In the case of a typical electrocatalytic inner-sphere mechanism, electron transfer to O2,ads to form (O2•
)ads followed by proton transfer from water molecules take place. Proton transfer could occur from water molecules coadsorbed on the electrode surface or could also be from the partial solvation shell of adsorbed O2/(O2•
)ads. Increasing alkalinity of the supporting electrolyte (pH > 12) causes the rate of proton transfer from water to decrease concomitant to the decrease in water activity. This is primarily the reason for increased stability of the superoxide radical anion O2•
in strongly alkaline electrolytes.44 The above ideas primarily reflect the electrocatalytic inner-sphere electron transfer mechanism. The outer-sphere electron transfer to form superoxide O2•
is characterized by the following equation:41 O2, aq þ e
f ðO2 •
Þaq

Figure 1. Schematic illustration of the double-layer structure during ORR in acidic (left) and alkaline (right) conditions. Insets (a) and (b) illustrate the inner- and outer-sphere electron transfer processes.

changed from 0 to 14. As shown in Figure 1, in a typical aqueous acidic electrolyte of oxygen saturated 0.1 M HClO4 the primary constituents are hydronium ions (H3O+), perchlorate anions (ClO4
), solvated molecular oxygen, and the solvent water molecules. A brief description of the various constituents of the compact part of the electrochemical double layer and their influence on ORR is given below. Given that the concentration of the acidic/alkaline supporting electrolyte is typically g0.1 M, the diffuse part of the double layer is not considered. Cathodic potentials of oxygen reduction in an operating fuel cell typically occur at potentials well positive of the potential of zero charge (pzc). At these conditions in acidic media, chemisorbed molecular O2 (either dissociatively or nondissociatively adsorbed), specifically adsorbed hydroxyl species (OHads arising from water activation), and solvent water dipoles constitute the IHP. Solvated molecular O2, and ClO4
anions populate the outer-Helmholtz plane (OHP). For potentials positive of pzc, the distance of closest approach of H3O+ ions to the electrode surface is limited to the OHP owing to its net positive charge. Only after the first electron transfer to O2,ads to form the superoxide (O2•
)ads species, protonation of the (O2•
)ads intermediate by the transfer of proton from OHP to the IHP likely occurs. Because protons in acidic media have very high mobility, this step is not rate limiting. To be an efficient electrocatalyst, the ORR reaction intermediates remain adsorbed on the catalyst site until 4e
and 4H+ are transferred followed by desorption of stable H2O molecules as the final product. In the case of alkaline media (0.1 M NaOH) although the double layer structure is not dramatically different, there are some important aspects that need to be taken into account, as shown in Figure 1. At high pH environment water molecules not

E0 ¼
0:33 V vs SHE

ð1Þ

As mentioned above, the overpotential for this reaction in acidic media (pH = 0) is ∼1.53 V, which decreases significantly to ∼0.7 V in alkaline media (pH = 14). This decrease in overpotential implies that strong chemisorption of O2 to the electrode surface is not a prerequisite. Other noncovalent forces such as long-range dipole
dipole interactions or the free energy associated with hydrogen bonding (typically 1.15 V, the primary oxide species of interest to ORR conditions below 1.0 V is typically (OHads)γ
1 (0 < γ < 1, where γ represents the electrosorption valency characterizing the magnitude of charge transfer from the adsorbate to the electrode) and Oads. Ideally, deprotonation (step c) is expected to commence only after complete charge transfer (step b) of all adsorbed OH
species, indicating discrete potential windows for each step. In reality, due to surface heterogeneity there is sufficient overlap of potential regions for steps (b) and (c). In this context, recent study by Jerkiewicz et al.54 show that Pt
Oads formation takes place at potentials 12 it is HO2
.25 ORR process primarily takes place at potentials positive of the potential of zero charge. This gives rise to an electrostatic attractive interaction between the positive charge on the electrode surface and the anionic nature of the peroxide intermediate in alkaline media. Further, the cationic nature of the Fe2+ active site gives an added advantage in electrostatically stabilizing the anionic HO2
species in alkaline media. Such an electrostatic double-layer effect stabilizes the peroxide intermediate on the electrode surface and ensures the complete 4e
conversion of O2 in alkaline media. This electrostatic advantage is clearly absent in acidic media given the neutral charge on H2O2. Figure 5c shows the ring current due to peroxide oxidation measured during ORR in both acidic and alkaline electrolytes; the corresponding peroxide yields are shown in Table 1. Clearly the peroxide yield during ORR on FeTPP/C (pyrolyzed at 800 °C) is higher in acidic media by an order of magnitude compared to that in alkaline media (Table 1). The onset potential for ORR in 0.1 M NaOH is 0.95 V vs RHE whereas the corresponding peroxide oxidation current does not begin until 0.8 V. In 0.1 M HClO4, the onset potential for both ORR and peroxide oxidation is 0.8 V. This is further proof for the instability of peroxide intermediate on the active site in acidic media because the weak binding of the H2O2 intermediate on the Fe2+ active site facilitates its desorption and/or decomposition into the bulk electrolyte. This desorption of H2O2 into the bulk electrolyte due its weak binding on the active site is the primary source of peroxide detected at the ring in acidic media. At very high overpotentials (