Investigating the Kinetic Mechanisms of the Oxygen Reduction

Article pubs.acs.org/JPCC

Investigating the Kinetic Mechanisms of the Oxygen Reduction Reaction in a Nonaqueous Solvent Nelson A. Galiote,† Dayse C. de Azevedo,‡ Osvaldo N. Oliveira, Jr.,‡ and Fritz Huguenin*,† †

Departamento de Química, Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto, Universidade de São Paulo, 14040-901 Ribeirão Preto (SP), Brazil ‡ Instituto de Física de São Carlos, Universidade de São Paulo, CP 369, 13560-970 São Carlos (SP), Brazil S Supporting Information *

ABSTRACT: The high theoretical energy density of lithium−oxygen batteries brings the promise of higher performance than existing batteries, but several technological problems must be addressed before actual applications are made possible. Among the difficulties to be faced is the slow oxygen reduction reaction (ORR), which requires a suitable choice of catalysts and electrolytic solution. This can only be achieved if the kinetics and mechanism of this reaction are known in detail. In this study, we determined the rate constants for each elementary step of ORR for a platinum electrode in 0.1 mol·L−1 LiClO4/1,2-dimethoxyethane (DME), using a kinetic model in the frequency domain. We found that the energy storage capacity of lithium−air batteries can be increased by converting a large amount of lithium superoxide into lithium peroxide during the electrochemical step in comparison with chemical disproportionation. The mechanisms for ORR were supported by data from an electrochemical quartz crystal microbalance (EQCM): ORR could be distinguished from parasitic reactions induced by solvent degradation, and agglomerates of LixO2 (1 ≤ x ≤ 2) were adsorbed on the electrode. The rate-limiting step for ORR was the electron transfer to the oxygen molecules strongly adsorbed onto platinum sites, particularly as a large amount of reaction product (Li2O2) adsorbed onto the electrode. Even though Pt sheets are likely to be impracticable for real applications due to their low surface area, they were useful in making it possible to determine the kinetics of ORR steps. This can now be employed to devise more involved electrodes, such as those containing dispersed Pt nanoparticles.

■

transfer.14−16 Materials like Pd and Pt appear to exhibit this intermediate adsorption strength.14 Even though platinum is expensive as electrocatalyst in energy sources and batteries, it is still the best ORR catalyst in aqueous media and one of the most efficient in aprotic media.15,17−20 Therefore, studying the kinetics in these electrochemical systems remains relevant. ORR has been studied in aprotic media where the superoxide anion arises.21−24 ORR has been suggested to occur via a oneelectron, two-electron, or four-electron mechanisms in nonaqueous systems, to afford LiO2, Li2O2, and Li2O, respectively, according to the global reactions:25,26

INTRODUCTION Population growth has pushed the energetic demand and the global dependence on oil, with the ensuing rise in emission of pollutant gases. The use of suitable devices to store the excess energy originating from “green sources” or a convenient substitute for combustion engines in transportation is a possible solution.1−3 However, huge challenges must be addressed before these technologies become successful. Among the few alternatives is the lithium−air battery, which has high theoretical specific energy density (11.7 kW·h·kg−1),4−7 with a practical specific energy density close to 3.5 kW·h·kg−1 being reported.8 These batteries have been studied extensively since the report on the first nonaqueous secondary battery.9 Catalysts in the positive electrode can enhance the oxygen reduction reaction (ORR) rate to deliver and store energy efficiently. Several materials may be applied for this purpose, including carbonaceous materials (e.g., graphene sheets), transition-metal oxides (manganese spinel and cobalt oxides), inorganic− organic composites (metal macrocycle derivatives), metal oxides (Fe2O3, Fe3O4, and CuO), noble metals, and metallic alloys (Pt, Pd, and Au).10−13 Electrocatalysts should allow for suitable oxygen adsorption at the electrode surface, which must be strong enough to decrease the activation energy barrier for electro-reduction while being sufficiently weak to allow these anions to bind with lithium ions after the first electron © 2014 American Chemical Society

Li+ + e− + O2 → LiO2 ,

E° = 3.00 V (vs Li/Li+)

2Li+ + 2e− + O2 → Li 2O2 ,

(1)

E° = 2.96 V (vs Li/Li+) (2)

4Li+ + 4e− + O2 → 2Li 2O,

E° = 2.91 V (vs Li/Li+) (3)

In practical systems, additional electroreduction steps occur at the positive electrode, with intermediates emerging and Received: May 30, 2014 Revised: August 27, 2014 Published: August 29, 2014 21995

dx.doi.org/10.1021/jp5053584 | J. Phys. Chem. C 2014, 118, 21995−22002

The Journal of Physical Chemistry C

Article

decomposing, and competition with solvent degradation also exists. The formation of Li2O is kinetically limited, and the major discharge product is Li2O2, even at standard potentials more negative than 2.91 V. In theoretical and experimental studies using density functional theory (DFT) and spectroscopic measurements, respectively, Li2O2 production was reported. According to these studies, one-electron transfer to the O2 molecule initiates ORR, thus leading to the superoxide anion (O2−) being weakly adsorbed on the electrode surface.4,27−30 Solvent molecules can solvate this adsorbed species, which can also react with cations such as Li+ in the electrolytic solution.31 Several organic solvents undergo nucleophilic attack by superoxide ions, yielding discharge products that are different from the expected Li x O 2 species.32−36 Other possible solvent degradation reactions are proton abstraction by the superoxide and autoxidation in the presence of the solution saturated with oxygen at high potentials.37,38 These issues limit the practical potential window, making the choice of a suitable solvent for ORR crucial.32−34,37,38 Several degradation products (as lithium carbonate, acetates, formates, and hydroperoxides) have been detected for ORR in ether solutions,27,39 but a theoretical approach to predict solvent stability has suggested the use of ethers as aprotic solvent for lithium−air batteries.36,37 Since ORR in nonaqueous solvents and lithium salts is not yet fully understood, here we investigated the kinetics of ORR using a LiClO4/1,2-dimethoxyethane (DME) electrolyte solution and a platinum electrode. Platinum sheets are not suitable for batteries due to their surface area, but they are suitable to study ORR mechanisms in detail since effects from other variables, such as size distribution in nanoparticles and contribution from more than one catalyst in alloys and composites, are eliminated. To estimate the kinetic constants of the reaction steps, we conducted measurements in the frequency domain and developed a kinetic model.

+

− k4

Pt−LiO2(ads) + Li + e → Pt−Li 2O2(ads)

k5′

v2 = k 2θO2 exp( −bE 0) exp(bE)

K1K5[O2 ] + 1 + K5

(15)

K5K1[O2 ](1 − θLiO2) K1K5[O2 ] + 1 + K5

⎛1 ⎞ ⎜ ⎟ K1K5[O2 ] + 1 + K5 ⎝ K5 ⎠

(16)

K5(1 − θLiO2)

(17)

where K5 = (k5/k5′ ). Equation 18 gives the temporal variation of θLiO2 which goes to zero in the steady-state, necessary to relate the reaction steps and θLiO2: dθLiO2 dt

= v2 − v3 − v4 = 0

(18)

In the transient state during the perturbation ac potential, Taylor expansion helps to express the step rates as42,43 ⎛ ∂v ⎞ ⎛ ∂v ⎞ 2 ⎟ ΔθLiO2e(jωt ) + ⎜ 2 ⎟ ΔEe(jωt ) v2 = v20 + ⎜⎜ ⎟ ⎝ ∂E ⎠θ ⎝ ∂θLiO2 ⎠ E LiO2 (19)

⎛ ∂v ⎞ 3 ⎟ (jωt ) v3 = v30 + ⎜⎜ ⎟ ΔθLiO2e ∂ θ ⎝ LiO2 ⎠ E

(5)

(6)

(20)

⎛ ∂v ⎞ ⎛ ∂v ⎞ v4 = v40 + ⎜⎜ 4 ⎟⎟ ΔθLiO2e(jωt ) + ⎜ 4 ⎟ ΔEe(jωt ) ⎝ ∂E ⎠θ ∂ θ ⎝ LiO2 ⎠ E LiO2

(7)

(21)

where j = (−1)1/2, ω is the angular frequency, and v0i is the steady-state rate for reaction i. The terms ΔθLiO2 and ΔE are associated with the oscillation amplitude of the coverage degree and with the sinusoidal ac potential. They obey the following relation:

(8)

where the subscript “ads” means species adsorbed onto the platinum sites. The electrochemical steps have been considered out of equilibrium due to overpotential. On the basis of these reaction steps, the corresponding rates are obtained: v1 = k1θPt[O2 ] − k1′θO2

(14)

K5(1 − θLiO2)

θLi 2O2 =

k5

Pt−Li 2O2(ads) ⇄ Pt + Li 2O2

(13)

θO2 =

k3

k 3′

v5 = k5θLi 2O2 − k5′θPt

θPt =

(4)

2Pt−LiO2(ads) ⇄ Pt 2−Li 2O2(ads) + O2

(12)

θPt + θO2 + θLiO2 + θLi 2O2 = 1

K1

Pt−O2(ads) + Li+ + e → Pt−LiO2(ads)

v4 = k4θLiO2 exp( −bE 0) exp(bE)

where b = (αnF/RT), E is the equilibrium potential, and E is the interfacial potential. θPt is the surface coverage degree associated with the platinum free sites, and θO2,θLiO2, and θLi2O2 are the surface coverage degrees for O2, LiO2, and Li2O2, respectively. The term [O2] represents the oxygen molar concentration dissolved in DME. θPt, θO2 and θLi2O2 can be related to the surface coverage degree of the adsorbed intermediate θLiO2 (eqs 14−17) through the mass balance, which makes its substitution in eq 9−13 possible.

■

− k2

(11)

0

KINETIC MODEL The elementary steps are based on the ORR mechanism proposed in the literature.25,31,40,41 The exception is the last chemical equilibrium, which considers partial dissolution of Li2O2 adsorbed onto the platinum electrode in DME: Pt + O2 ⇄ Pt−O2(ads)

v3 = k 3(θLiO2)2 − k 3′θLi 2O2[O2 ]

0 θLiO2 = θLiO + ΔθLiO2 exp(jωt ) 2

(9)

(22)

θ0LiO2is

where the coverage degree of adsorbed LiO2 at the steady-state. The temporal variation of θLiO2 at the transient

(10) 21996



Article

state allows one to associate ΔθLiO2 with ΔE as expressed below: Γ

dθLiO2 dt

= ΓFjωΔθLiO2 exp(jωt ) = v2 − v3 − v4

Equation 25 shows the total impedance (ZT), which depends on the capacitive reactance (Zc) associated with the doublelayer capacitance (Cdl) in parallel with the faradaic impedance

(23)

where Γ is the maximum surface coverage. Hence, eqs 19 and 21 provide the oscillatory faradaic current density (Δif), which is the sum of faradaic contributions Δν2 and Δν4 (Δν2 = ν2 − ν20 and Δν4 = ν4 − ν40) multiplied by the Faraday constant, to determine the impedance values for each angular frequency.

ΔE Zf = Δi f

Zf − 1

and in series with the electrolytic solution resistance (Rs):

⎛ Zf Zc ⎞ ⎟⎟ + R s ZT = ⎜⎜ ⎝ Zf + Zc ⎠

(24)

(− =−

k 2K5K1[O2 ]ebE K1K5[O2 ] + 1 + K5

k 2K5[O2 ]K1ebE K1K5[O2 ] + 1 + K5

+ k4ebE

)⎛⎝− ⎜

+ 2k 3θLiO2 +

where

k 2K5K1[O2 ](1 − θ LiO2)bebE K1K5[O2 ] + 1 + K5 k 3′[O2 ] K1K5[O2 ] + 1 + K5

(25)

⎞ − k4θLiO2bebE⎟ ⎠

+ k4e

bE

+ Iω ΓF

+

k 2K5K1[O2](1 − θLiO2)bebE K1K5[O2 ] + 1 + K5

− k4θLiO2bebE (26)

Equation 26 was used to fit the electrochemical impedance spectroscopy data. The rate constants were scanned; the selected ones were based on the lowest difference between the experimental and theoretical impedance modulus.43 The methodology used to determine the errors for each rate constant is described in the Supporting Information.

electrochemically active area of 0.361 cm2 and an integral sensitivity constant of 0.0815 cm2 s−1 g−1. The resonance frequency shift was measured with a HP-5370B Universal Timer/Counter. Changes in the resonance frequency of the EQCM crystal were transformed into mass changes using the Sauerbrey equation.46

EXPERIMENTAL SECTION All the reagents were purchased from Sigma-Aldrich and used as received: 1,2-dimethoxyethane containing molecular sieves (with water content lower than 0.005%); lithium perchlorate salt grade (99.99% purity); and tetrabutylammonium perchlorate (TBAClO4, purity higher than 99.0%). The electrolyte solutions were prepared prior to use and stored in a glovebox (MBRAUN) with water and oxygen concentration lower than 10 ppm. A platinum sheet was used as the working and counter electrode, and an Ag/Ag+ saturated in electrolyte solution was employed as a quasi-reference electrode. This latter electrode had a potential of 2.94 V versus Li/Li+, so all the potentials referred to hereafter are based on the reference Li/Li+. A LiClO4 (DME) or TBAClO4/DME electrolytic solution (0.1 mol·L−1) was employed in the electrochemical experiments. All the experiments were carried out using an Autolab PGSTAT30 potentiostat/galvanostat. Ac electrochemical impedance spectroscopy was employed between 10 kHz and 10 mHz. Several dc potentials were tried using 5 mV of superimposed ac amplitude. This impedance was measured after pretreatment for at least 2000 s, to achieve a pseudosteady state. Here, these states were defined as a 10% change in current density within half an hour. Ultrapure N2 (99.999% purity) or O2 (99.998% purity) from Linde-AGA were purged into the solution for 20 min before the experiments; the gas flow was maintained during the electrochemical measurements. Oxygen solubility in DME is 9.57 mmol.cm−3.26 Cyclic voltammetry was performed in acid medium (0.5 M H2SO4) to determine the value of Γ (=2.46 × 10−9 mol of sites per cm2), considering an oxidation charge associated with atomic hydrogen of 210 μA per square centimeter of real area for polycrystalline platinum.44,45 The program Maple v13.0 was used to fit the experimental data. For EQCM experiments, the substrates were 6 MHz AT-cut quartz crystals coated with platinum. They have a piezoelectrically and

RESULTS AND DISCUSSION Cyclic voltammetry experiments in an electrochemical cell using electrolytic solution deaerated with nitrogen gas and saturated with oxygen gas indicate that the parasitic reaction in the electrolytic solution in the absence of oxygen for the potential window used in this work can be neglected, as shown in Figure 1. These data show that the reduction process in 0.1

■

■

Figure 1. Cyclic voltammograms using electrolytic solution (− −) deaerated with nitrogen gas and () saturated with oxygen gas at 20 mV s−1. Inset Figure shows the third (red -•-) and fifth (blue line) voltammetric cycles.

M LiClO4/DME saturated with O2 happens at approximately 2.7 V, as expected for ORR in ether/lithium ion based electrolytes.14,47−49 The electrooxidation process takes place at potentials more positive than 3.5 V, confirming the reaction irreversibility in the conditions used, with low cyclability of the electrochemical system being illustrated in the third and fifth voltammetric cyles in the inset in Figure 1. This is a consequence of the insolubility of the reaction products in 21997



Article

DME, which tends to passivate the electrode surface with increasing number of voltammetric cycles. The potentiodynamic profile of the potential changed along the cycles, probably owing to other reaction pathways originating from alterations in the chemical environment due to partial passivation of the platinum surface.35,47,50,51 The potential, charge (q), and mass change (Δm) measured with an electrochemical quartz crystal microbalance (EQCM) during the linear negative potential scan at 20 mV s−1 and in 0.1 M LiClO4/DME saturated with O2 are shown in Figure 2. This

Figure 3. Chronoamperometric curves at 2.34, 2.10, and 1.90 V, using electrolytic solution saturated with oxygen gas.

to minimize the effects of these oscillations, which last about 100 s. Moreover, the impedance data did not change after several consecutive measurements in this frequency range. DME, molecular oxygen, and platinum catalysts participate in these current oscillations, because the latter do not occur during chronoamperometric experiments performed with other electrodes and with aerated or deaerated solvents (for instance, propylene carbonate or glass carbon electrode). Studies about these oscillations will be presented elsewhere. Figure 4 shows the Nyquist and Bode diagrams (in the inset) obtained in oxygen saturated electrolytic solution containing 0.1 M LiClO4/DME, as well as in oxygen-saturated 0.1 M TBAClO4/DME, at (a) 2.34, (b) 2.10, and (c) 1.90 V. The diagrams are superimposed with the ac potential perturbation of 5 mV from 10 kHz to 10 mHz. With extrapolation of the semicircle at 2.34 V, the charge transfer resistance (Rct) is determined as 155 kΩ·cm2 in the presence of oxygen dissolved in 0.1 M LiClO4/DME. Rct decreases as the dc potential becomes less positive, and two semicircles emerge at 2.10 and 1.90 V. The semicircles at high and low frequencies are probably associated with ORR and solvent degradation, respectively. In the Bode diagrams, two distinct processes are identified which display different time constants, at less positive potentials. The Rct values associated with ORR are 20 and 7 kΩ·cm2 at 2.10 and 1.90 V, respectively. Electrochemical impedance spectroscopy in oxygen-saturated electrolytic solution consisting of 0.1 M TBAClO4/DME may confirm ORR and DME degradation reactions in 0.1 M LiClO4/DME, since tetrabutylammonium cations (TBA+) can sustain ORR via one-electron transfer without any solvent degradation26TBA+ ions are poorly solvated and tend to form a stable TBA−O2 complex.39,55,56 According to the hard− soft acid−base theory (HSAB), O2− is a soft base with higher affinity for TBA+ (poorly solvated soft acid) than for the Li+ ion (hard acid).26 Note that a second semicircle (or second time constant) does not appear in the Nyquist diagrams at 2.34, 2.10, or 1.90 V when 0.1 M TBAClO4/DME is the electrolytic solution. Therefore, the second semicircle at low frequencies in 0.1 M LiClO4/DME at 2.10 and 1.90 V (Figures 4b and 4c) refers to solvent degradation by O2− or Li2O2.39,57 Solvent degradation reactions (not included in the reaction mechanism) at 2.10 and 1.90 V modify coverage degrees of reaction products and intermediates of ORR at the steady-state, and this may compromise determination of the rate constants. However, we opted to consider kinetic constants as estimated at these dc potentials, because the inclusion of degradation

Figure 2. Bidimensional projections for E − q (a), E − Δm (b), and q − Δm planes during ORR in 0.1 M LiClO4/DME saturated with O2 at 20 mV s−1.

experiment was aimed at determining the chemical species taking part in the electrochemical reactions. The mass change is small at potentials more positive than 2.10 V, but it increases significantly at more negative potentials owing to solvent degradation. The slope in the mass change as a function of the electroreduction charge is ca. 3.4 × 10−4g C−1 at 2.34 V, which is an intermediate value between those expected for formation of Li2O2 (2.4 × 10−4 g C−1) and LiO2 (4.0 × 10−4 g C−1). It seems that agglomerates of Li2O2 and LiO2 were formed, as predicted using density functional theory (DFT) calculations and Raman spectroscopy.16,52,53 Differently from results in aqueous medium, these EQCM data support a nondissociative adsorption for oxygen molecules, as Li2O is formed during dissociative adsorption and the ratio for mass change/charge expected for Li2O (7.8 × 10−5 g C−1) was not observed. Nondissociative adsorption for oxygen molecules has also been supported by isotope labeling experiments.54 The chronoamperometric curves at 2.34, 2.10, and 1.90 V in Figure 3 are evidence for ORR in oxygen-saturated 0.1 M LiClO4/DME. The current density decreased fast in the beginning, but the steady state was not reached even after a long time because the reaction products partially blocked the platinum surface, as mentioned above. These experiments helped to determine the time that would be necessary to obtain pseudosteady states, which is important when conducting electrochemical impedance spectroscopy at dc potentials. These data also provide important information about how the current oscillates during the electrochemical reactions. These oscillations correspond to ca. 5% of the total current density at 2.34 V, and increase to ca. 10% at 2.10 and 1.90 V. Since the kinetic model above involves the steady state for ORR, the impedance data had to be fitted at high frequencies (from 2 kHz to 3 Hz) 21998



Article

Figure 4. Nyquist diagrams in oxygen saturated electrolytic solution containing (○) 0.1 M LiClO4/DME and (•) 0.1 M TBAClO4/DME at (a) 2.34 V, (b) 2.10 V, and (c) 1.90 V. Bode diagrams are shown in the inset: (blue ○) 0.1 M LiClO4/DME and (red ●) 0.1 M TBAClO4/DME.

Figure 5. (•) Experimental and (○) theoretical impedance data for ORR at (a) 2.34, (b) 2.10, and (c) 1.90 V. () Theoretical impedance data using CPE. Complex diagrams of capacitance are shown in the inset figures: (red ●) Experimental and (blue ○) theoretical impedance data.

reaction in the reaction mechanisms would increase the number of variables (degrees of freedom), thus turning the rate constants unreliable. Moreover, we did not observe solvent degradation for measurements at 2.34 V, although parasitic reactions were reported for open circuit potential in the presence of molecular oxygen.54 Therefore, the coverage degree for the chemical species in the proposed reaction mechanism can be taken as (almost) independent of parasitic reactions. Figure 5a through 5c illustrate the fitting of the experimental impedance data from 2 kHz to 5 Hz at (a) 2.34 V, (b) 2.10 V, and (c) 1.90 V, using eq 26. First, we fitted the impedance data using a double-layer capacitance of 40 μF.cm−2 at 2.34 V and of 20 μF.cm−2 at 2.10 and 1.90 V. Table 1 lists the rate constants and their errors, determined from parabolic fitting around the values of each kinetic constant (see Supporting Information S1, S2, and S3). For the sake of completeness in analyzing the impedance data, we fitted them using a constant-phase element

(CPE) with the kinetic constants of Table 1, since frequency dispersions associated with the nonfaradaic processes are observed and are not included in the kinetic model. The impedance of 1/Q(jω)n replaced the double-layer capacitance, where Q is a constant with dimension Fsn‑1. The insets in Figure 5 show these fittings in the complex capacitance diagram, to allow for better visualization at higher frequencies. Moreover, the beginning of a second semicircle associated with other processes (solvent degradation reactions) arises at low frequencies, but we have not fitted it because it is not part of our kinetic model. The most significant differences (at most four times) between rate constants at 2.34 V and those estimated at less positive potentials may be ascribed to solvent degradation reactions (as mentioned above), in addition to changes in the chemical environment due to formation of insoluble Li2O2 product. 21999



Article

Table 1. Kinetic and Equilibrium Constants as a Function of Potential constants K1 k2, k3, k′3, k4, K5

mol cm−2·s−1 mol cm−2·s−1 cm·s−1 mol cm−2·s−1

2.34 V

2.10 V

1.90 V

(2.30 ± 0.13) × 102 (1.52 ± 0.15) × 10−6 4.33 ± 0.38 2.54 ± 0.02 (2.36 ± 0.10) × 102 1.90 ± 0.02

(2.18 ± 0.01) × 102 (1.42 ± 0.3) × 10−6 4.760 ± 0.024 0.620 ± 0.007 (9.00 ± 0.007) × 102 7.40 ± 0.01

(2.36 ± 0.08) × 102 (1.59 ± 0.5) × 10−6 9.76 ± 0.10 0.70 ± 0.11 (2.45 ± 0.49) × 102 7.37 ± 0.02

Notes

These kinetic constants suggest that electron transfer to the oxygen molecules adsorbed onto the platinum sites (eq 5) is the rate-limiting step. The k2 values are close to those reported in the literature;24,25 the difference may lie in the solvent employed and in kinetic model adopted. K1 values are high because molecular oxygen strongly adsorbs onto platinum sites.14,16,20 The low Li2O2 solubility in organic medium and the K5 values in Table 1 suggest that Li2O2 formed during ORR remained almost totally adsorbed onto the platinum sites, contributing to electrode surface passivation. This conclusion is consistent with the decreased current density during the chronoamperommetric experiments and as a function of the number of voltammetric cycles, already mentioned. High k4 values indicate larger lithium ion affinity for O22− as compared with O2−.26,58 The difference between k3 and k4 suggests that Li2O2 emerges preferentially during the electrochemical step, to enhance the current density, the energy density and the power of lithium−air batteries. It is worth pointing out that though the number of fitting parameters is large, the values inferred from the fitting were consistent with expectation.

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We are grateful to FAPESP (Projects 2012/21629-1, 2011/ 12668-0, and 2011/21545-0), nBioNet (CAPES), and INCTADAPTA/FAPEAM/CNPq (573976/2008-2) for the financial support. We are also grateful to Prof. Luis Gustavo Dias (FFCLRP/USP) for the statistical treatment.

■

■

CONCLUSIONS ORR was found highly irreversible for a platinum electrode in an electrolytic solution of LiClO4 and DME, according to the cyclic voltammetry experiments. The low cyclability of this electrochemical system results from the low solubility of the reaction products in DME. The EQCM data suggested formation of Li2O2 and LiO2 during ORR in an organic medium, whose rate of elementary steps was investigated using a kinetic model in the frequency domain. The kinetic constants are consistent with the literature, pointing to high adsorption energy for oxygen molecules, slow electron transfer to molecules adsorbed onto the platinum sites, and partial blockage of these sites due to formation of reaction products that are insoluble in the organic solvent. Most significantly, the reaction product (Li2O2) may preferentially arise during the electrochemical step, thus increasing the number of electrons transferred to each oxygen molecule adsorbed onto the platinum sites and enhancing the power and energy storage capacity of lithium−air batteries. Because these findings are likely to be independent of the form of Pt, ORR can now be studied in detail for platinum nanoparticles that are more suitable for practical devices.

■

ASSOCIATED CONTENT

S Supporting Information *

Parabolic fittings to determine the error of the rate constants. This material is available free of charge via the Internet at http://pubs.acs.org.

■

REFERENCES

(1) Herranz, J.; Garsuch, A.; Gasteiger, H. A. Using Rotating Ring Disc Electrode Voltammetry to Quantify the Superoxide Radical Stability of Aprotic Li−Air Battery Electrolytes. J. Phys. Chem. C 2012, 116, 19084−19094. (2) Wagner, F. T.; Lakshmanan, B.; Mathias, M. F. Electrochemistry and the Future of the Automobile. J. Phys. Chem. Lett. 2010, 1, 2204− 2219. (3) Alotto, P.; Guarnieri, M.; Moro, F. Redox Flow Batteries for the Storage of Renewable Energy: A Review. Renew. Sust. Energy Rev. 2014, 29, 325−335. (4) McCloskey, B. D.; Bethune, D. S.; Shelby, R. M.; Girishkumar, G.; Luntz, A. C. Solvents’ Critical Role in Nonaqueous Lithium− Oxygen Battery Electrochemistry. J. Phys. Chem. Lett. 2011, 2, 1161− 1166. (5) Electrochemistry: Rechargeable Li-Air Battery. NatureRes. Highlights 2012, 488, 8−8. (6) Scrosati, B. Bottled Lightning: Superbatteries, Electric Cars, and the New Lithium Economy. Nature 2011, 473, 448−449. (7) Bruce, P. G.; Freunberger, S. A.; Hardwick, L. J.; Tarascon, J.-M. Li-O2 and Li-S Batteries with High Energy Storage. Nat. Mater. 2012, 11, 19−29. (8) Kwabi, D. G.; Ortiz-Vitoriano, N.; Freunberger, S. A.; Chen, Y.; Imanishi, N.; Bruce, P. G.; Shao-Horn, Y. Materials Challenges in Rechargeable Lithium-Air Batteries. MRS Bull. 2014, 39, 443−452. (9) Abraham, K. M. A Polymer Electrolyte-Based Rechargeable Lithium/Oxygen Battery. J. Electrochem. Soc. 1996, 143, 1−5. (10) Cheng, F.; Chen, J. Metal-Air Batteries: From Oxygen Reduction Electrochemistry to Cathode Catalysts. Chem. Soc. Rev. 2012, 41, 2172−2192. (11) Ren, X. M.; Zhang, S. S.; Tran, D. T.; Read, J. Oxygen Reduction Reaction Catalyst on Lithium/Air Battery Discharge Performance. J. Mater. Chem. 2011, 21, 10118−10125. (12) Débart, A.; Bao, J.; Armstrong, G.; Bruce, P. G. An O2 Cathode for Rechargeable Lithium Batteries: The Effect of a Catalyst. J. Power Sources 2007, 174, 1177−1182. (13) Shao, Y.; Park, S.; Xiao, J.; Zhang, J.-G.; Wang, Y.; Liu, J. Electrocatalysts for Nonaqueous Lithium−Air Batteries: Status, Challenges, and Perspective. ACS Catal. 2012, 2, 844−857. (14) Lu, Y. C.; Gasteiger, H. A.; Shao-Horn, Y. Catalytic Activity Trends of Oxygen Reduction Reaction for Nonaqueous Li-Air Batteries. J. Am. Chem. Soc. 2011, 133, 19048−19051. (15) Lima, F. H. B.; Zhang, J.; Shao, M. H.; Sasaki, K.; Vukmirovic, M. B.; Ticianelli, E. A.; Adzic, R. R. Catalytic Activity-D-Band Center Correlation for the O2 Reduction Reaction on Platinum in Alkaline Solutions. J. Phys. Chem. C 2007, 111, 404−410.

AUTHOR INFORMATION

Corresponding Author

*(F.H.) E-mail: fritz@ffclrp.usp.br. 22000



Article

(16) Dathar, G. K. P.; Shelton, W. A.; Xu, Y. Trends in the Catalytic Activity of Transition Metals for the Oxygen Reduction Reaction by Lithium. J. Phys. Chem. Lett. 2012, 3, 891−895. (17) Liang, Y.; Li, Y.; Wang, H.; Dai, H. Strongly Coupled Inorganic/ Nanocarbon Hybrid Materials for Advanced Electrocatalysis. J. Am. Chem. Soc. 2013, 135, 2013−2036. (18) Lima, F. H. B.; de Castro, J. F. R.; Santos, L. G. R. A.; Ticianelli, E. A. Electrocatalysis of Oxygen Reduction on Carbon-Supported Pt− Co Nanoparticles with Low Pt Content. J. Power Sources 2009, 190, 293−300. (19) Shao, Y. Y.; Ding, F.; Xiao, J.; Zhang, J.; Xu, W.; Park, S.; Zhang, J. G.; Wang, Y.; Liu, J. Making Li-Air Batteries Rechargeable: Material Challenges. Adv. Funct. Mater. 2013, 23, 987−1004. (20) Lu, Y. C.; Gallant, B. M.; Kwabi, D. G.; Harding, J. R.; Mitchell, R. R.; Whittingham, M. S.; Shao-Horn, Y. Lithium-Oxygen Batteries: Bridging Mechanistic Understanding and Battery Performance. Energy Environ. Sci. 2013, 6, 750−768. (21) Maricle, D. L.; Hodgson, W. G. Reducion of Oxygen to Superoxide Anion in Aprotic Solvents. Anal. Chem. 1965, 37, 1562− 1565. (22) Sawyer, D. T.; Chlericato, G.; Angelis, C. T.; Nanni, E. J.; Tsuchiya, T. Effects of Media and Electrode Materials on the Electrochemical Reduction of Dioxygen. Anal. Chem. 1982, 54, 1720− 1724. (23) Sawyer, D. T.; Valentine, J. S. How Super Is Superoxide? Acc. Chem. Res. 1981, 14, 393−400. (24) Chiericato, G.; Rodrigues, S.; Da Silva, L. A. Comparative Study between Two Methods of Resistance Compensation in the Cyclic Voltammetric Determination of the Standard Electrochemical Rate Constant of Dioxygen. Electrochim. Acta 1991, 36, 2113−2117. (25) Laoire, C. O.; Mukerjee, S.; Abraham, K. M.; Plichta, E. J.; Hendrickson, M. A. Elucidating the Mechanism of Oxygen Reduction for Lithium-Air Battery Applications. J. Phys. Chem. C 2009, 113, 20127−20134. (26) Laoire, C. O.; Mukerjee, S.; Abraham, K. M.; Plichta, E. J.; Hendrickson, M. A. Influence of Nonaqueous Solvents on the Electrochemistry of Oxygen in the Rechargeable Lithium−Air Battery. J. Phys. Chem. C 2010, 114, 9178−9186. (27) Freunberger, S. A.; Chen, Y.; Drewett, N. E.; Hardwick, L. J.; Barde, F.; Bruce, P. G. The Lithium-Oxygen Battery with Ether-Based Electrolytes. Angew. Chem., Int. Ed. 2011, 50, 8609−8613. (28) Ottakam Thotiyl, M. M.; Freunberger, S. A.; Peng, Z.; Bruce, P. G. The Carbon Electrode in Nonaqueous Li-O2 Cells. J. Am. Chem. Soc. 2013, 135, 494−500. (29) Hummelshoj, J. S.; Luntz, A. C.; Norskov, J. K. Theoretical Evidence for Low Kinetic Overpotentials in Li-O2 Electrochemistry. J. Chem. Phys. 2013, 138, 034703. (30) Viswanathan, V.; Thygesen, K. S.; Hummelshoj, J. S.; Norskov, J. K.; Girishkumar, G.; McCloskey, B. D.; Luntz, A. C. Electrical Conductivity in Li2O2 and Its Role in Determining Capacity Limitations in Non-Aqueous Li-O2 Batteries. J. Chem. Phys. 2011, 135, 214704. (31) Lu, Y. C.; Gasteiger, H. A.; Crumlin, E.; McGuire, R.; ShaoHorn, Y. Electrocatalytic Activity Studies of Select Metal Surfaces and Implications in Li-Air Batteries. J. Electrochem. Soc. 2010, 157, A1016− A1025. (32) Bryantsev, V. S.; Giordani, V.; Walker, W.; Blanco, M.; Zecevic, S.; Sasaki, K.; Uddin, J.; Addison, D.; Chase, G. V. Predicting Solvent Stability in Aprotic Electrolyte Li-Air Batteries: Nucleophilic Substitution by the Superoxide Anion Radical (O2(*−)). J. Phys. Chem. A 2011, 115, 12399−12409. (33) Bryantsev, V. S.; Uddin, J.; Giordani, V.; Walker, W.; Addison, D.; Chase, G. V. The Identification of Stable Solvents for Nonaqueous Rechargeable Li-Air Batteries. J. Electrochem. Soc. 2013, 160, A160− A171. (34) Bryantsev, V. S. Predicting the Stability of Aprotic Solvents in Li-Air Batteries: Pk(a) Calculations of Aliphatic C-H Acids in Dimethyl Sulfoxide. Chem. Phys. Lett. 2013, 558, 42−47.

(35) Freunberger, S. A.; Chen, Y.; Peng, Z.; Griffin, J. M.; Hardwick, L. J.; Barde, F.; Novak, P.; Bruce, P. G. Reactions in the Rechargeable Lithium-O2 Battery with Alkyl Carbonate Electrolytes. J. Am. Chem. Soc. 2011, 133, 8040−8047. (36) Sahapatsombut, U.; Cheng, H.; Scott, K. Modelling of Electrolyte Degradation and Cycling Behaviour In a lithium−Air Battery. J. Power Sources 2013, 243, 409−418. (37) Bryantsev, V. S.; Faglioni, F. Predicting Autoxidation Stability of Ether- and Amide-Based Electrolyte Solvents for Li-Air Batteries. J. Phys. Chem. A 2012, 116, 7128−7138. (38) Zhu, D.; Zhang, L.; Song, M.; Wang, X.; Mei, J.; Lau, L. W. M.; Chen, Y. G. Solvent Autoxidation, Electrolyte Decomposition, and Performance Deterioration of the Aprotic Li-O−2 Battery. J. Solid State Electrochem. 2013, 17, 2865−2870. (39) Younesi, R.; Hahlin, M.; Bjorefors, F.; Johansson, P.; Edstrom, K. Li-O−2 Battery Degradation by Lithium Peroxide (Li2O2): A Model Study. Chem. Mater. 2013, 25, 77−84. (40) Girishkumar, G.; McCloskey, B.; Luntz, A. C.; Swanson, S.; Wilcke, W. Lithium−Air Battery: Promise and Challenges. J. Phys. Chem. Lett. 2010, 1, 2193−2203. (41) McCloskey, B. D.; Scheffler, R.; Speidel, A.; Girishkumar, G.; Luntz, A. C. On the Mechanism of Nonaqueous Li−O2electrochemistry on C and Its Kinetic Overpotentials: Some Implications for Li−Air Batteries. J. Phys. Chem. C 2012, 116, 23897−23905. (42) Armstron, Rd; Henderso, M. Impedance Plane Display of a Reaction with an Adsorbed Intermediate. J. Electroanal. Chem. 1972, 39, 81−90. (43) Orazem, M. E.; Tribollet, B. Electrochemical Impedance Spectroscopy; John Wiley & Sons: NJ, 2008; p 533. (44) Santos, V. P. d.; Tremiliosi Filho, G. Correlaçaõ Entre a Estrutura Atômica Superficial E O Processo De Adsorçaõ -Dessorçaõ ́ De Hidrogênio Em Eletrodos Monocristalinos Pt(111), Reversivel Pt(100) E Pt(110). Quim. Nova 2001, 24, 856−863. (45) Gasparotto, L. H. S.; Ciapina, E. G.; Cantane, D. A.; Gomes, J. F.; Lima, F. H. B.; Ticianelli, E. A.; Tremiliosi, G. Altering the Adsorptive and Electronic Properties of Pt through Poly(Vinyl Alcohol) Adsorption. Electrochim. Acta 2013, 104, 358−366. (46) Sauerbrey, G. Verwendung Von Schwingquarzen Zur Wägung Dünner Schichten Und Zur Mikrowägung. Z. Phys. 1959, 155, 206− 222. (47) Meini, S.; Piana, M.; Tsiouvaras, N.; Garsuch, A.; Gasteiger, H. A. The Effect of Water on the Discharge Capacity of a Non-Catalyzed Carbon Cathode for Li-O2 Batteries. Electrochem. Solid. State Lett. 2012, 15, A45−A48. (48) Laoire, C. O.; Mukerjee, S.; Plichta, E. J.; Hendrickson, M. A.; Abraham, K. M. Rechargeable Lithium/Tegdme-Lipf6/O2 Battery. J. Electrochem. Soc. 2011, 158, A302−A308. (49) Gallant, B. M.; Kwabi, D. G.; Mitchell, R. R.; Zhou, J.; Thompson, C. V.; Shao-Horn, Y. Influence of Li2O2 Morphology on Oxygen Reduction and Evolution Kinetics in Li−O2 Batteries. Energy Environ. Sci. 2013, 6, 2518−2528. (50) Aurbach, D.; Daroux, M.; Faguy, P.; Yeager, E. The Electrochemistry of Noble-Metal Electrodes in Aprotic OrganicSolvents Containing Lithium-Salts. J. Electroanal. Chem. 1991, 297, 225−244. (51) Cota, L. G.; de la Mora, P. On the Structure of Lithium Peroxide, Li2O2. Acta Crystallogr., B 2005, 61, 133−136. (52) Lau, K. C.; Assary, R. S.; Redfern, P.; Greeley, J.; Curtiss, L. A. Electronic Structure of Lithium Peroxide Clusters and Relevance to Lithium-Air Batteries. J. Phys. Chem. C 2012, 116, 23890−23896. (53) Yang, J.; Zhai, D.; Wang, H. H.; Lau, K. C.; Schlueter, J. A.; Du, P.; Myers, D. J.; Sun, Y. K.; Curtiss, L. A.; Amine, K. Evidence for Lithium Superoxide-Like Species in the Discharge Product of a Li-O2 Battery. Phys. Chem. Chem. Phys. 2013, 15, 3764−3771. (54) McCloskey, B. D.; Scheffler, R.; Speidel, A.; Bethune, D. S.; Shelby, R. M.; Luntz, A. C. On the Efficacy of Electrocatalysis in Nonaqueous Li−O2 Batteries. J. Am. Chem. Soc. 2011, 133, 18038− 18041. 22001



Article

(55) Wang, H.; Xie, K.; Wang, L.; Han, Y. N-Methyl-2-Pyrrolidone as a Solvent for the Non-Aqueous Electrolyte of Rechargeable Li-Air Batteries. J. Power Sources 2012, 219, 263−271. (56) Nasybulin, E.; Xu, W.; Engelhard, M. H.; Nie, Z. M.; Burton, S. D.; Cosimbescu, L.; Gross, M. E.; Zhang, J. G. Effects of Electrolyte Salts on the Performance of Li-O‑2 Batteries. J. Phys. Chem. C 2013, 117, 2635−2645. (57) Assary, R. S.; Lau, K. C.; Amine, K.; Sun, Y. K.; Curtiss, L. A. Interactions of Dimethoxy Ethane with Li2o2 Clusters and Likely Decomposition Mechanisms for Li-O‑2 Batteries. J. Phys. Chem. C 2013, 117, 8041−8049. (58) Zheng, D.; Wang, Q.; Lee, H. S.; Yang, X. Q.; Qu, D. Catalytic Disproportionation of the Superoxide Intermediate from Electrochemical O2 Reduction in Nonaqueous Electrolytes. Chem.Eur. J. 2013, 19, 8679−83.

22002


Investigating the Kinetic Mechanisms of the Oxygen Reduction

Recommend Documents