Mesoporous β-MnO2 air electrode modified with pd for rechargeability in lithium-air battery

Article abstract of DOI:10.1149/2.090112jes

The electrochemical performance and electrode reactions using ordered mesoporous β-MnO₂ modified with Pd as a cathode catalyst for rechargeable Li-air batteries was reported. Well-ordered mesoporous β-MnO₂ was prepared using mesoporous silica KIT-6 as a template under hydrothermal synthesis of Mn(NO₃)₂H₂O. The obtained mesoporous β-MnO₂ shows narrow pore size distribution of 1 nm. With the dispersion of small amounts of Pd to β-MnO₂, mesoporous β-MnO₂ exhibited a high initial discharge capacity of 817 mAhg_-cat. with high reversible capacity. Charging potential is suppressed at 3.6 V vs. LiLi, which is highly effective for preventing the decomposition of organic electrolyte. The mesoporous β-MnO₂Pd electrode shows good rate capability and cycle stability. Ex-situ and in-situ XRD results suggested that the observed capacity comes primarily from the oxidation of Li to Li₂O ₂ followed by Li₂O after discharge to 2.0 V vs. LiLi. Electron spin resonance measurements suggest that the formation of superoxide anion radicals contributs to the oxidation of Li and the radicals were recovered during charge. Ex-situ FTIR measurement suggested that no electrolyte decomposition was observed and no Li₂CO₃ was formed during discharge when ethylene carbonate (EC)-diethyl carbonate (DEC) (3:7), which is highly stable for Li-air battery, was used as the electrolyte.

Full text of DOI:10.1149/2.090112jes

Journal of The Electrochemical Society, 158 (12) A1483-A1489 (2011)
A1483
0013-4651/2011/158(12)/A1483/7/$28.00 © The Electrochemical Society
Mesoporous β-MnO2 Air Electrode Modified with Pd
for Rechargeability in Lithium-Air Battery
Arjun Kumar Thapa, Yuiko Hidaka, Hidehisa Hagiwara,
Shintaro Ida, and Tatsumi Ishihara∗,z
Department of Applied Chemistry, Faculty of Engineering, Kyushu University, Fukuoka 819-0395, Japan
The electrochemical performance and electrode reactions using ordered mesoporous β-MnO2 modified with Pd as a cathode catalyst
for rechargeable Li-air batteries was reported. Well-ordered mesoporous β-MnO2 was prepared using mesoporous silica KIT-6 as a
template under hydrothermal synthesis of Mn(NO3)2 · H2O. The obtained mesoporous β-MnO2 shows narrow pore size distribution
of 1 nm. With the dispersion of small amounts of Pd to β-MnO2, mesoporous β-MnO2 exhibited a high initial discharge capacity
of 817 mAh/g-cat. with high reversible capacity. Charging potential is suppressed at 3.6 V vs. Li/Li+, which is highly effective for
preventing the decomposition of organic electrolyte. The mesoporous β-MnO2/Pd electrode shows good rate capability and cycle
stability. Ex-situ and in-situ XRD results suggested that the observed capacity comes primarily from the oxidation of Li+ to Li2O2
followed by Li2O after discharge to 2.0 V vs. Li/Li+. Electron spin resonance measurements suggest that the formation of superoxide
anion radicals contributs to the oxidation of Li+ and the radicals were recovered during charge. Ex-situ FTIR measurement suggested
that no electrolyte decomposition was observed and no Li2CO3 was formed during discharge when ethylene carbonate (EC)-diethyl
carbonate (DEC) (3:7), which is highly stable for Li-air battery, was used as the electrolyte.
© 2011 The Electrochemical Society. [DOI: 10.1149/2.090112jes] All rights reserved.
Manuscript submitted July 25, 2011; revised manuscript received September 19, 2011. Published November 14, 2011. This was
Paper 1011 presented at the Las Vegas, Nevada, Meeting of the Society, October 10–15, 2010.
The coming new energy economy must be based on a cheap and
sustainable energy supply. Combined with sustainable resources such
as wind and solar power, chemical energy storage using batteries
can contribute to a potential solution. Currently, battery research and
development is focused on energy storage and conversion with high-
energy high-power density and reliable safety.1–4 Metal-air batteries
have attracted considerable attention because of their extremely large
specific capacity. The reason for such a large specific capacity is that
these cells consist of lithium metal as an anode and an air electrode
for activation of oxygen in air; hence these metal-air batteries have
a simple structure. Among the various metal-air battery systems, the
lithium-air battery is the most attractive because it has the highest
energy density per unit weight. The cell discharge reaction occurs
between Li and oxygen to yield Li2O or Li2O2, with a theoretical dis-
charge voltage of ca. 3.0 V and a theoretical specific energy density up
to 5200 Wh/kg-Li. In practice, the storage of oxygen in the battery is
unnecessary, because air can be directly used. Therefore, the theoret-
ical specific energy (excluding oxygen) is 11.140 kWh/kg-Li, which
is much higher than that of other advanced batteries and methanol
direct fuel cells. Abraham and Jiang reported an Li-air battery using
a nonaqueous electrolyte.5 They suggested that lithium peroxide is
a discharge product based on 2(Li+ + e−) + O2 → Li2O2, which
resulted in a theoretical voltage of 2.96 V. However, because of low
oxygen solubility in a nonaqueous electrolyte, the reported power
density of an Li-air battery using a nonaqueous electrolyte is much
lower than the theoretical value.6,7 Dobley et al. and Kuboki et al. em-
ployed liquid organic solvents as the electrolyte.8,9 When employing
an organic or ionic liquid-based electrolyte solution, the cell reaction
produces insoluble Li2O or Li2O2, which precipitates in the pores
of the porous carbon-based air electrode to block the further intake
of oxygen, abruptly terminating the discharge reaction. Recently, a
lithium-air rechargeable battery employing MnO2 as an air electrode
was reported.10,11 In 2006, Bruce et al. reported that a rechargeable
Li–O2 battery using a carbonate-based electrolyte containing Super
S carbon with electrolytic MnO2 as the catalyst had an initial dis-
charge capacity of 1000 mAh/g (based on the weight of carbon) at
a current rate of 70 mA/g and a capacity retention rate of 60% after
50 cycles.12 Recently, Mizuno et al. also reported that a rechargeable
Li–O2 battery using propylene carbonate (PC) electrolyte containing
Ketjenblack (EC600JD) carbon with MnO2 as the catalyst had an
initial discharge capacity of 820 mAh/g-cat. and a capacity retention
rate of 60% after 100 cycles, which is the longest cycle life reported
for a rechargeable Li–O2 to date.13 However, decrease in the charge
potential is required because of low energy efficiency for charge and
discharge (62% of the reported cell) and also leads to the decompo-
sition of electrolyte resulting in the formation of Li2CO3.14 For this
purpose, further improvement in the catalytic activity of an air elec-
trode, one is larger discharge capacity and another is decreased charge
potential, is required. MnO2 can be used as an alternative low cost
electrocatalyst for oxygen reduction/evolution reactions.15 However,
MnO2 prepared by the traditional ceramic route has limited electro-
catalytic activity because of its large particle size and low specific
surface area.16 In addition, decomposition of electrolytes containing
propylene carbonate (PC) leads to the formation of Li2CO3; this is the
main reason for the high charge potential.13 Previously, we studied
mesoporous α-MnO2 modified with a Pd cathode air electrode for a
rechargeable Li-air battery, which showed higher reversible capacity
as well as higher current density than the standard Li-air battery.17 In
the current study, we synthesized mesoporous β-MnO2 using silica
KIT-6 as a template and applied it as a cathode catalyst for an Li-
air battery. We found that a mixture of Pd and mesoporous β-MnO2
shows high oxidation activity and Li+ undergoes reduction to form
Li2O2 or Li2O, even in a dry atmosphere. We also pointed out negli-
gible decomposition of organic electrolyte by decreasing the charge
potentials and using the stable electrolyte.
Experimental
Sample preparation.— Mesoporous β-MnO2 was prepared us-
ing mesoporous silica KIT-6 as a template. The mesoporous silica
KIT-6 template was prepared in according with the literature.18,19
Mn(NO3)2 · 6H2O (3 g) was dissolved in ethanol (15 ml) and then the
mesoporous silica KIT-6 template was added in a beaker. The mixture
was stirred continuously at room temperature till all of the solution
was adsorbed. The solid product was then heated slowly to 573 K for
3 h and allowed to cool slowly. The product was reheated to 623 K
at the rate of 1 K/min for 5 h. Mesoporous β-MnO2 was treated with
hot 2M NaOH solution to remove the silica KIT template and then
washed with distilled water and ethanol several times. The product
was then kept at 333 K for 12 h to remove the moisture.
∗
Measurement of air electrode performance for Li-air battery.—
The electrochemical characterizations were carried out using a
Electrochemical Society Active Member.
z E-mail: ishihara@cstf.kyushu-u.ac.jp
Downloaded on 2014-11-28 to IP 128.114.34.22 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

A1484
Journal of The Electrochemical Society, 158 (12) A1483-A1489 (2011)
b)
a)
β-MnO2
β-MnO2
1
2
3
4
5
10
20
30
40
50
60
70
80
2θ / o
2θ / o
Figure 1. (a) Low-angle and (b) wide-angle XRD patterns of mesoporous β-MnO2.
Swagelok-type cell. The cathode was formed by casting a mixture
of mesoporous β-MnO2, PdO, and PTFE-coated teflonized acety-
lene black (TAB-2) as a conducting binder (mol ratio of 75:15:10).
The resulting mixture was pressed onto a stainless steel mesh and
the cathode was dried at 433 K for 12 h under vacuum. It is also
noted that PdO in air electrode was reduced to Pd during first charge
because only X-ray diffraction peak assigned to metallic Pd was
observed.
Li-air cell measurements were carried out at room temperature.
The Li-air battery configuration used in this study consists of a
lithium foil (12 mm in diameter) as an anode electrode separated
by two pieces of porous polypropylene film (Celgard 3401). The cell
was gas tight except for the stainless steel mesh window that ex-
posed the porous cathode to O2 atmosphere. The electrolyte used
was 1M lithium bis(trifluoromethanesulfonyl)imide-ethylene carbon-
ate:diethyl carbonate (LiTFSI-EC:DEC, 3:7 by volume) supplied by
Ube Chemical Co. Ltd. Charge–discharge performance was carried
out at a voltage range of 4.0–2.0 V at current densities of 0.025–
5.1 mA cm−2. In this study, for capacity comparison, we normalized
the observed capacity by the total weight of the air electrode, not by
the weight of carbon. A practical lithium-air battery should be tested
under ambient conditions. The moisture content in the air reacts with
the lithium negative electrode and CO2 consumes all the lithium salt
in the electrolyte. To avoid such negative effects, the cell was tested
in a pure dry oxygen atmosphere.
and then sealing the electrodes in a polyvinyl bag in an argon-
filled glove box to prevent any reaction with moisture or CO2 in
the air. Electron spin resonance (ESR) measurement was performed
at room temperature using Bruker, type EMK8 to remove the air
electrode from the cell and to analyze the oxygen radicals by di-
lution with silica glass sand. FT-IR measurement was performed
with transmission method using KBr windows and mercury cad-
mium telluride (MCT) semiconductor detector was used. Commercial
FT-IR spectrometer (Type 610, JASCO co ltd.) was used and the sam-
ples was mounted in the gastight IR cell after charge and discharge
measurement.
Results and Discussion
Structure and morphology of prepared β-MnO2.— Figure 1a illus-
trates the low-angle XRD patterns of mesoporous β-MnO2. A sharp
diffraction intensity peak was observed at 1.02◦, suggesting the for-
mation of a mesoporous structure. The estimated diameter of the
mesopores in β-MnO2 was 0.85 nm, which corresponds to the in-
ner diameter of mesoporous β-MnO2. From the XRD pattern, it was
confirmed that mesoporous β-MnO2 was successfully synthesized by
a hydrothermal synthesis process. Figure 1b shows the wide-angle
XRD pattern of the prepared samples, and it was confirmed that all
diffraction peaks could be assigned to the mesoporous β-MnO2 phase.
Because no diffraction peaks from an impurity phase were observed
in the wide-angle XRD pattern, it can be reasonably concluded that
a single β-MnO2 phase with a mesoporous structure was successfully
prepared. From the XRD database (JCPDS No. # 24-0735), the pre-
pared sample could be assigned to a tetragonal lattice of β-MnO2
[space group P42/mnm], where the lattice parameter values are a
= 4.404 Å, b = 4.404 Å, and c = 2.876 Å.
Characterization.— Powder X-ray diffraction (XRD) analysis
of mesoporous β-MnO2 was carried out with CuKα radiation (λ
= 1.541 Å, Rigaku Rint 2500) at the low-angle scanning range of
0.5–5◦ and a wide-angle range of 10–80◦ (2θ) with a scan speed of
0.02◦/s. In-situ XRD measurement during discharge and charge was
performed in this study. The commercial in-situ XRD measurement
cell (Rigaku) was used in this study. The cell consists of a polyimide
film window with an open hole. An air electrode of 0.2 g and metal Li
were set in the XRD measurement cell. The in-situ cell was put into
a polyethylene bag and oxygen was substituted for the atmosphere. A
current density of 0.02 mA/cm2 was applied, and when the potential
reached the target value, the XRD measurement was performed using
CuKα line (Rigaku Rint-Mini).
Figure 2a shows the scanning electron microscope images of
mesoporous β-MnO2 in which small microparticles of 0.1–0.3 μm
b)
a)
A Keyence VE-7800 was used to obtain a scanning electron
microscope image. A JEOL JEM 2100 operated at 200 kV was
used to obtain a transmission electron microscope image. Spe-
cific surface area (SBET) was determined from nitrogen adsorption–
desorption isotherm using Nippon Bell, BELSORP 18PLUS-FS,
Japan.
10nm
Examination of the discharge electrode involved disassembling
the cell in the glove box, rinsing the cathode electrode three times
with diethyl carbonate for removing the solvent under vacuum,
Figure 2. (a) SEM image of mesoporous β-MnO2 and (b) TEM image of
mesoporous β-MnO2.
Downloaded on 2014-11-28 to IP 128.114.34.22 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 158 (12) A1483-A1489 (2011)
A1485
4.0
4.0
3.5
3.0
2.5
2.0
β-MnO
a)
b)
β-MnO/Pd
2
2
3.5
3.0
2.5
2.0
513 mAh/g
500 600
817 mAh/g
0
100
200
300
400
0
100 200 300 400 500 600 700 800 900
-1
-1
Capacity / mAh g
Capacity / mAh g
Figure 3. Charge–discharge curves of lithium-air battery using mesoporous (a) β-MnO2 and (b) Pd/β-MnO2 air electrode in an O2 atmosphere between 4.0–
2.0 V at a current density of 0.025 mA cm−2
.
diameter were aggregated to form large spherical secondary parti-
cles of ca. 1 μm diameter. The TEM image of mesoporous β-MnO2
shown in Figure 2b indicates many pores with no regular structure.
Therefore, the pore structure seems to be the so-called wormhole-like
pore structure. The estimated size of the pore from the TEM obser-
vation was ca. 1 nm, which agrees closely with that estimated from
the low-angle diffraction peak (0.8 nm). N2 adsorption–desorption
isotherm curves of mesoporous β-MnO2 were also studied. Meso-
porous β-MnO2 showed a typical type IV curve with H1 hysteresis
loop, which is characteristic of a typical mesopore structure. SBET of
mesoporous β-MnO2 was 29.0 m2 g−1. The pore size distribution of
mesoporous β-MnO2 was 4.0–6.1 nm with a pore volume of 0.38 cm3
discharge curve of mesoporous β-MnO2 electrode with added Pd for
the Li-air battery shows that the efficiency of charge and discharge
increased drastically from 65% to 77.8%.
The rate property of Li-air is generally not good; however, if the
air catalyst is highly active, then a reasonably high rate property is
expected. In this study, the charge and discharge rate properties of the
cell were also studied using the mesoporous β-MnO2/Pd air electrode.
The charge–discharge curve of the cell is shown in Figure 4 at various
current densities. By applying a current density of 0.025 mA cm−2
,
the mesoporous β-MnO2/Pd electrode showed a discharge capacity
of 817 mAh g-cat.−1. The charge capacity decreased as the current
density increased; however, it was found that for a current density
as high as 5.1 mA cm−2, the β-MnO2/Pd electrode still retained the
discharge capacity of 93 mAh g-cat.−1, which is 11.3% of its initial
capacity. The mesoporous β-MnO2/Pd electrode showed high capacity
comparable to current lithium-ion batteries such as those of LiMn2O4.
This indicates that the rate property of the current Li-air battery is
reasonably high, which can be attributed to the high activity of the
present air electrode. Here the mesoporous β-MnO2/Pd air electrode
showed a large capacity even at high current density. The porocity of
the air electrode was not high and slso Li foil with small surface area
used for anode. Therefore, one reason for the decreased discharge
capacity at high current density might be the slow dissolution rate
of Li metal into electrolyte and also slow diffusion of Li+ in the air
electrode to active site. In any case, although the Li-air battery is
considered to have a low rate property and shows only large capacity
at a low current density, reasonably high capacity was exhibited at
high current density for the cell using Pd/mesoporous β-MnO2.
Cycle stability at low current density is also studied. Figure 5
shows the capacity as a function of cycle number of the meso-
porous β-MnO2/Pd air cathode electrode at low and high current
densities. The mesoporous β-MnO2/Pd electrode showed an initial
g
−1. The observed pore seems to be space between particles but not
mesopores.
Charge–discharge property of Pd/mesoporous β-MnO2.— The
electrocatalytic activities of β-MnO2/TAB-2 (90:10) and β-
MnO2/Pd/TAB-2 (75/15/10) air electrode in the Li-air battery were
measured. Althoughpropylenecarbonatebasedelectrolytewas widely
used for organic electrolyte of Li-air battery, we used 1M lithium
bis(trifluoromethanesulfonyl)imide-ethylene carbonate:diethyl car-
bonate (LiTFSI-EC:DEC, 3:7 by volume) for electrolyte because of
high chemical stability. However, in order to prevent decomposition,
applied potential should be lower than 4 V. Figure 3 shows the charge–
discharge curve of the Li-air cell. Mesoporous β-MnO2 had an ini-
−1
tial discharge capacity of ca. 513 mAh g-cat. during the first cycle
(Figure 3a). The discharge plateau curve was large and flat at
2.8–2.9 V, which is comparableto reported data.11 However, thecharge
potential plateau curve was observed at 3.7 V, which is slightly higher
than the theoretical potential of 2.95 V vs. Li/Li+. In our previous
study, we found that Pd was highly active for Li+ oxidation and de-
composition of Li2O2 or Li2O.16,17 Loading of Pd on MnO2 shows
high energy efficiency for charge and discharge; hence, loading of Pd
on mesoporous β-MnO2 was further studied. The addition of a small
amount of Pd to the mesoporous β-MnO2 increased the capacity to ca.
817 mAh g-cat.−1, as shown in Figure 3b. With the dispersion of a small
amount of Pd to β-MnO2, the discharge potential plateau curve also
shifted toward the high potential of 2.9 V vs. Li/Li+, which is close to
the theoretical value of 2.95 V vs. Li/Li+ for Li2O2 formation. Increase
in discharge potential is simply explained by the decreased electrode
over potential by loading of the active Pd catalyst. The charge poten-
tial was also suppressed at 3.6 V vs. Li/Li+ and showed a large and
flat plateau curve with high reversible capacity. The above charge–
−1
discharge capacity of 578 mAh g-cat. during the initial −c1ycle. Af-
ter 13 cycles, it retained the capacity of 417 mAh g-cat. at 0.12
mA cm−2. At high curren−t 1densities, it showed an initial discharge
capacity of 208 mAh g-cat. at 1.91 mA cm−2 during the initia−l1cy-
cle. After 13 cycles, it retained the capacity of 134 mAh g-cat. at
1.91 mA cm−2. The capacity retention rates of the β-MnO2/Pd elec-
trode at low and high current densities were 72.2% and 64.9%, respec-
tively, after 13 cycles, which is much higher than the renention rate
reported by other groups.10,11,13 Therefore, the cycle stability of the
cell is reasonably high and the Li-air cell can be used as a rechargeable
battery.
Downloaded on 2014-11-28 to IP 128.114.34.22 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

A1486
Journal of The Electrochemical Society, 158 (12) A1483-A1489 (2011)
150
150
100
50
800
600
400
200
0
Discharge
β-MnO2/Pd electrode
Charge
0.12 mA/cm2
100
50
0
Discharge
Charge
Coulombic efficiency %
1.91 mA/cm2
Current density : 5.1 mA/cm2
0
2
4
6
8
10
12
14
0
0
20
40
60
80
Cycle number
Cycle number
Figure 4. Charge–discharge capacities versus cycle performance of meso-
porous β-MnO2/Pd/TAB (75/15/10) electrode at current densities of 0.12 and
Figure 6. Charge discharge capacities versus cycle numbers of the Li-air cell
using mesoporous β-MnO2/Pd/TAB (75/15/10) air electrode at current density
1.91 mA cm−2
.
of 5.1 mA cm−2
.
Further longer cycle stability was studied at higher current density,
5.1 mA cm−2. Figure 6 shows the capacity as a function of cycle num-
ber of the mesoporous β-MnO2/Pd air cathode electrode at the current
density of 5.1 mA cm−2. The mesoporous β-MnO2/Pd air electrode
showed an initial discharge capacity of 93 mAh g-cat.−1 during the ini-
tial cycle. The charge–discharge measurement showed a decrease in
discharge capacity during the second cycle. However, further charge-
discharge measurement showed a gradual increase in capacity of up to
broad peaks, which are assigned to the peak from Li2O, were also
observed at ca. 34 and 66◦. These results suggest that Li2O2 forms at
the initial highpotential range as well as at the low potential range
of 2.0 V. In contrast, the strong peak of Li2O at 2.0 V suggests
that Li2O is also formed during discharge, in particular, at the low
potential range. It has been reported by several groups that Li2O2 is
the main product in Li-air batteries; however, this study confirmed
that Li2O is also formed during discharge. In addition, as illustrated in
Figure 7, it is evident that no peaks were assigned to Li2CO3, which is
another possible product during discharge of the Li-air battery. In our
previous study, we found that low carbon content in an air electrode
effectively prevents Li2CO3 formation and results in low discharge
potential. Therefore, in this study, the low discharge potential of the
cell using the β-MnO2/Pd air electrode can be explained by the absence
of Li2CO3 formation. It is also noted that Li2O was formed when Pd
was used as the air electrode. This could be explained by the high
activity of Pd for the dissociation of the oxygen–oxygen (O-O) bond.
The discharge products of the air electrode were further studied
with Raman spectroscopy. Raman spectroscopic analysis is reported
to be highly sensitive to Li2O and Li2O2. As shown in Figure 8,
characteristic Raman peaks for Li2O and Li2O2 were observed after
discharge at 250 and 500 cm−1, respectively. Because the peak in-
tensity of Li2O was much stronger than that of Li2O2, the Raman
spectra suggest that the main product in β-MnO2/Pd was Li2O after
−1
95 mAh g-cat. during the 92nd cycles. After the 93rd cycle, capacity
fading was observed owing to increase in internal resistance of the
cell caused by evaporation of the carbonate electrolyte. This is the
first study to report that Li-air battery can be rechargeable even at a
current density as high as 5.1 mA cm−2 for up to ca. 100 cycles.
The ex-situ XRD of the β-MnO2/Pd electrodes before charge and
after discharge to 200 mAh g-cat.−1, 400 mAh g-cat.−1, and 2.0 V
vs. Li/Li+, respectively, were measured in this study, because two
potential plateaus were observed for discharge in the Li-air battery.
Figure 7 shows the ex-situ XRD results of the β-MnO2/Pd electrodes
−1
before charge and after discharge. After discharge to 200 mAh g-cat.
,
the mesoporous β-MnO2/Pd electrode showed a new peak at 52◦,
which corresponds to the stainless steel (SUS) mesh. After discharge
to 400 mAh g-cat.−1, the ex-situ XRD of the β-MnO2/Pd electrode
showed a peak at 23◦, which was assigned to the peak from the
polythene bag. At 35◦ another new peak started to appear that could
be assigned to the formation of Li2O2. After discharge to 2.0 V, the
broad peak at 35◦ was retained. In addition to the peak at 35◦, new
(a) Before charge
β-MnO
Pd
Li2O2
2
(b) Discharge to 200 mAh/g
(c) Discharge to 400 mAh/g
(d) Discharge to 2.0V
4.0
3.5
3.0
2.5
Li2O
(d)
(c)
2.0
1.5
0.13
0.25
1.91
2.55
0.63
5.1
0.025
1.27
(b)
(a)
0
200
400
600
800
1000
Capcity / mAhg-1
10
20
30
40
50
60
70
80
2θ / o
Figure 5. First charge–discharge curves of the cell using mesoporous β-
MnO2/Pd/TAB (75/15/10) air electrode at different current densities. Number
in figure means currdent density for charge and discharge (mA/cm2).
Figure 7. Ex-situ XRD of mesoporous β-MnO2/Pd/TAB (75/15/10) electrode
before discharge and after discharge to 200 and 400 mAh g-cat. at 2.0 V.
−1
Downloaded on 2014-11-28 to IP 128.114.34.22 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 158 (12) A1483-A1489 (2011)
A1487
discharge to 2.0 V. Many studies have reported Li2O2 as the main
discharge product11,12 and small Li2O formation amount after dis-
charge. Furthermore, in this study, we observed Li2O as the main
discharge product. This difference could be explained by the activity
of the air electrode in relation to the dissociation of the O-O bond,
which will be discussed in detail later in this paper. In contrast, no
Li2CO3 peaks were observed at 1090 cm−1 in the Raman spectra after
discharge, which suggests that Li2CO3 was not formed. Therefore,
both Raman spectra and XRD measurements show negligible Li2CO3
formation. This could be related to the low charge potential around 3.7
V and the small amount of carbon binder used for the air electrode.
It has also been reported that significant decomposition of carbon-
ate electrolytes, in particular, propylene carbonate (PC), occurs at a
potential higher than 4.0 V.13,20,21
MnO2
Carbon
after discharge
before discharge
The effect of air instead of O2 was further studied. It was found
that the discharge capacity increased monotonously as the oxygen
partial pressure increased. Therefore, it is reasonable to conclude that
a high oxygen partial pressure is suitable for achieving a high capacity.
However, to date, the application of real air has not been studied in
detail. Compared with the discharge in O2, the potential plateau and
the discharge capacity both decreased slightly in dry air (Fig. 9).
However, decrease in capacity was not large; a fairly large discharge
capacity of ca. 417 mAh g-cat.−1 was observed. In contrast, the charge
potential increased to 3.6 V vs. Li/Li+ when dry air was substituted
for O2. XRD measurement after discharge in dry air also suggests that
no Li2CO3 diffraction peaks was observed.
0
250
500
750 1000 1250 1500 1750 2000
Wavelength (cm-1)
Figure 8. Ex-situ Raman spectroscopy of mesoporous β-MnO2/Pd/TAB
(75/15/10) electrode before discharge and after discharge to 2.0 V.
4.0
3.5
3.0
Charge and discharge mechanism on β-MnO2/Pd.— The charge–
discharge mechanism of the Li-air battery is not yet well understood.
As shown in Figure 3, two potential plateaus were observed at 2.9 and
2.7 V in the discharge curves and at 3.5 and 3.7 V in the charge curves.
This suggests that Li2O2 and Li2O formations were related to charge
and discharge in the studied cell. In-situ XRD measurements were
performed for the Pd/β-MnO2 air electrode in the Li-air cell by using a
homemade cell in substituted atmosphere. Figure 10 shows the in-situ
XRD patterns of the Pd/β-MnO2 air electrode during discharge. Before
discharge, it is obvious that all diffraction peaks can be assigned to Pd
and β-MnO2. With decreasing potential, the intensity of the diffraction
peaks from Pd and β-MnO2 increased slightly. This might be explained
by the slight change in the height of the sample because insoluble
Li2O2 and Li2O products are deposited. At ca. 2.8–2.7 V, diffraction
peaks were observed around 34.8 and 49.8◦ (shoulder peaks of Pd) and
2.5
2.0
1st
5th
3rd
dry air atmosphere
0
100
200
300
400
500
-1
Capacity / mAh g
Figure 9. Charge–discharge curves of a lithium-air battery using mesoporous
β-MnO2 modified with Pd air electrode in dry air atmosphere between 4.0–
2.0 V at a current density of 0.12 mA cm−2
.
Figure 10. In-situ XRD of mesoporous β-
MnO2/Pd/TAB (75/15/10) cathode air electrode for
rechargeable Li-air battery during discharge state in
open-air atmosphere.
Downloaded on 2014-11-28 to IP 128.114.34.22 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

A1488
Journal of The Electrochemical Society, 158 (12) A1483-A1489 (2011)
3000000
g=2.0072
After charge
2000000
1000000
0
g=2.0054
y
t
i
s
n
e
t
Figure 11. Electron spin resonance (ESR) spectra of Pd/β-MnO2
during discharge and charge.
n
I
After Discharge
at 2.8V
g=2.0072
-1000000
3200
3400
3600
3800
G value
these peaks can be attributed to Li2O2. As the potential decreased, at
around 33.8◦, a diffraction peak that can be attributed to Li2O became
significant and the peaks assigned to Li2O2 were weakened. Therefore,
Li2O2 was formed at high potential plateau and followed by Li2O
at low potential. This observation is valid considering the standard
potential of Li2O2 (2.95 V vs. Li/Li+) and Li2O (2.90 V vs. Li/Li+)
for the Li-air battery. Considering the change in weak strength, it is
likely that Li2O formed successively from Li2O2.
ESR measurements were also performed on active oxygen species
in this study. On the basis of Raman spectroscopy results, Bruce
et al. reported that oxygen radical is an active species for oxida-
tion of Li.21 Figure 11 shows the ESR spectra of the air electrode
at 2.8 and 2.0 V for discharge and 4 V for charge. Evidently, two
ESR signals at g = 2.007 were observed at the initial discharge pro-
cess. The ESR signal at g = 2.007 is typical of superoxide anion
radicals (O2−).22,23 After discharge, the −intensity of this ESR peak
decreased drastically, suggesting that O2 contributes to the forma-
tion of Li2O2 or Li2O. Although we did not detect LiO2, Bruce et
al. reported that the initial product in an air electrode is LiO2 from
the Raman spectra.21 Considering the ESR and in-situ XRD results,
the reaction steps for discharge in the present cell seem to be as
follows:
separation of the O-O bond, respectively, are suspected during the dis-
charge process. The similar reaction route was also reported by Laoire
et al. 27,28
During charge, Li2O might largely decompose at 3.7 V to form
Li and oxygen, which might indicate the presence of superoxide an-
ion radicals because the ESR signal was recovered after discharge.
Furthermore, a small amount of Li2O2 was also observed at the fi-
nal stage of discharge (down to 2.0 V) as shown in Figs. 3 and 9.
Based on the linear potential sweep measurement, decomposition
of Li2O2 was observed around 3.5 V and of Li2O at a potential
higher than 3.7 V. Therefore, a small potential plateau at 3.5 V during
charge can be attributed to the decomposition of Li2O2 and a large
plateau at a potential higher than 3.7 could indicate the decompo-
sition of Li2O. However, the charge process requires further detail
study.
Further, carbonate-based electrolytes decomposed with oxygen
radicals through lithiated organic carbonates to form Li2CO3. In this
study, changes in electrolytes before and after the charge and discharge
cycle were measured with FTIR. Figure 12 shows the FTIR spectra
of the pristine Pd/β-MnO2 electrode after first discharge and then
after recharging to 4.0 V in 1M LiTFSI-EC:DEC(3:7) under O2 at-
mosphere. Evidently, IR peaks assigned to carbonate were observed;
however, no significant change in the IR spectra of EC-DEC (3:7)
electrolyte was observed. Mizuno et al. reported that lithium alkyl
carbonate absorption peaks at 1650, 1360, and 825 cm−1 were formed
during charge and discharge.13 However, no peaks were observed at
the frequency shown in Figure 12, suggesting that in this study, no sig-
nificant electrolyte decomposition occurs. McCloskey et al. reported
that electrolyte decomposition is strongly related to the electrolyte
compound used.20 Therefore, the electrolyte of 1M LiTFSI/EC-DEC
(3:7) used in this study seems to be reasonably stable when electro-
chemical windows between 2.0 and 4.0 V are used and when small
amounts of carbon are present in the air electrode. This also agrees
well with the result that negligible Li2CO3 was formed in this study.
One reason for negligible Li2CO3 formation is that the charge po-
tential was limited to 4.0 V and a small amount of carbon was used
as the air electrode. The results from many other groups indicate a
large amount of carbon in the air electrode. However, in our previous
results, it was found that a MnO2 electrode with no carbon shows high
reversibility without forming Li2CO3.16 In any case, in this study, the
organic electrolyte 1M LiTFSI/EC-DEC (3:7) did not decompose sig-
nificantly during charge and discharge in a electrochemical window
from 2 to 4 V vs. Li/Li+. This is important for reversibility of Li-air
battery.
O2 + e−= O−2
O−2 + Li+= LiO2
LiO2 + Li++e−= Li2O2
Li2O2 + 2Li+ + 2e− = 2Li2O
According to this reaction mechanism, the Li ion reacts with
molecular oxygen. In the final step, the O-O bond fractures to form
Li2O. A similar reaction mechanism is also reported for the forma-
tion of H2O2 from O2 and H2.24,25 Pd is a well-known catalyst for
the fracture of O-O bond.26 Therefore, Li2O formation, which has
not been reported by other groups, was observed on the examined
Pd/β-MnO2 composite catalyst in this study. It has been reported
that superoxide anion radicals form on oxygen vacancy21 and MnO2
is easily partially reduced, i.e., Mn3+ is easily formed. Therefore,
it is expected that mesoporous β-MnO2 could be an active site for
the formation of superoxide anion radicals. The bifunctional roles
of MnO2 and Pd for the formation of superoxide anion radicals and
Downloaded on 2014-11-28 to IP 128.114.34.22 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 158 (12) A1483-A1489 (2011)
A1489
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
recharged electrode
discharged electrode
Figure 12. FTIR spectra of Pd/β-MnO2 pristine electrode af-
ter first discharge and then charge in 1M LiTFSI-EC:DEC
(3:7) under O2 atmosphere.
pristine electrode
2000
1800
1600
1400
1200
1000
800
600
Wavenumber (cm-1)
Conclusion
3. J. Chen and F. Cheng, Acc. Chem. Res., 42, 713 (2008).
4. J. M. Tarascon and M. Armand, Nature, 414, 359 (2001).
5. K. M. Abraham and Z. Jiang, J. Electrochem. Soc., 143, 1 (1996).
6. J. Read, J. Electrochem. Soc., 149, A1190 (2002).
7. J. Read, K. Mutolo, M. Ervin, W. Behl, J. Wolfenstine, A. Driedger, and D. Foster,
J. Electrochem. Soc., 150, A1351 (2003).
In summary, this study reveals that a mesoporous β-MnO2/Pd elec-
trode is highly active for a rechargeable Li-air battery and is effective
for increasing discharge and decreasing charge potentials. This con-
figuration results in energy efficiency. Even at reasonably high current
density, high reversible capacity was achieved in both pure O2 and dry
air atmospheres. These results suggest that lithium-air batteries can
be rechargeable even at high current densities. Using the mesoporous
β-MnO2/Pd electrode, ex-situ and in-situ XRD results show the for-
mation of Li2O2 followed by Li2O after discharge to 2.0 V. The ESR
measurements also suggest that superoxide anion radicals contribute
to the oxidation of Li ions and during the charge process, superoxide
anion radicals are recovered. In this study, energy efficiency of up to
80.5% for charge and discharge were achieved. This is much higher
than the reported data to date. This level of energy efficiency can be
achieved by the bifunctional activities of Pd/β-MnO2 for charge and
discharge reactions and by the prevention of EC-DEC (3:7) electrolyte
decomposition by limiting the amount of carbon binder and charge
potential up to 4.0 V. On the other hand, comparing with discharge
capacity by using Pd/α-MnO2 which has molecular size tunnel struc-
ture, much larger discharge capacity was observed on Pd/β-MnO2 in
spite of the similar surface area. Measoporous structure of β-MnO2 is
wormhole-like pore and that of α-MnO2 is straight. Therefore, I think
simple straight mesopore is easily blocked with deposited Li oxide
and so mesopore with complicated structure seems to be effective for
achieving the large capacity in Li-air battery. Consequently, controle
of mesopore structure is important for air electrode of a Li-air battery.
8. A. Dobley, J. DiCarlo, and M. K. Abraham, Proc. 41st Power Sources Conferences,
Philadelphia, PA, P.61 (2004).
9. T. Kuboki, T. Okuyama, T. Ohsaki, and N. Takami, J. Power Sources, 146, 766
(2005).
10. A. Debert, J. Bao, G. Armstrong, and P. G. Bruce, J. Power Sources, 174, 1177
(2007).
11. T. Ogasawara, A. Debart, M. Holfazel, P. Novak, and P. G. Bruce, J. Am. Chem. Soc.,
128, 1390 (2006).
12. A. Debart, A. J. Paterson, J. Bao, and P. G. Bruce, Angew. Chem. Inter. Ed., 47, 4521
(2008).
13. F. Mizuno, S. Nakanishi, Y. Kotani, S. Yokoishi, and H. Iba, Electrochemistry, 78,
403 (2010).
14. G. Q. Zhang, J. P. Zheng, R. Liang, C. Zhang, B. Wang, M. Au, M. Hendruchson,
and E. J. Plichta, J. Electrochem. Soc., 158, A822 (2011).
15. F. Cheng, J. Shen, B. Peng, Y. Pan, Z. Tao, and J. Chen, Nature Chemistry, 3, 79
(2011).
16. A. K. Thapa, K. Saimen, and T. Ishihara, Electrochem. Solid-State Lett., 13, A165
(2010).
17. A. K. Thapa and T. Ishihara, J. Power Sources, 196, 7016 (2011).
18. F. Kleitz, S. H. Choi, and R. Ryoo, Chem. Commun., 17, 2136 (2003).
19. T. W. Kim, F. Klutz, P. Paul, and R. Ryoo, J. Am. Chem. Soc., 127, 7601
(2005).
20. B. D. McCloskey, D. S. Bethune, R. M. Sheldy, G. Girishkumar, and A. C. Luntz, J.
Phys. Chem. Lett., 2, 1161 (2011).
21. S. A. Freunberger, Y. Chen, Z. Peng, J. M. Griffin, L. J. Hardwick, F. Barde, P. Novak,
and P. G. Bruce, J. Am. Chem. Soc., 133, 8040 (2011).
22. J. Y. Wu, S. H. Chien, and B. Z. Wan, Ind. Eng. Chem. Res., 40, 94
(2011).
23. H. Kanai, M. Linthuoto, M. Matsumura, and J. B. Moffat, J. Molecular Catalysis A,
251, 181 2006).
Acknowledgments
24. A. Staykov., T. Kamachi, T. Ishihara, and K. Yoshizawa, J. Phys. Chem. C, 112,
19501 (2008).
This study was financially supported by the Li-EAD project of the
New Energy and Industrial Technology Development Organization
(NEDO) of Japan.
25. M. Piccinini, E. N. Ntainjua, J. K. Edwards, A. F. Carley, J. A. Moulijn, and G. J.
Hutchings, Phys. Chem. Chem. Phys., 12, 2488 (2010).
26. K. Honkala and K. Laasonena, J. Chem. Phys. 115, 2297
(2001).
27. C. O. Laoire, S. Mukerjee, K. M. Abraham, E. J. Plichta, and M. A. Hendrickson, J.
Phys. Chem. C., 113, 20127 (2009).
References
28. C. O. Laoire, S. Mukerjee, K. M. Anraham, E. J. Plichta, and M. A. Hendrickson, J.
Phys. Chem.C., 114, 9178 (2010).
1. M. Winter and R. J. Brodd, Chem. Rev., 104, 4245 (2004).
2. M. Armand and J. M. Tarascon, Nature, 451, 652 (2008).
Downloaded on 2014-11-28 to IP 128.114.34.22 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).