Advanced Energy Materials - 2023 - Schmid - Rechargeable Oxide Ion Batteries Based On Mixed Conducting Oxide Electrodes

Research Article
www.advenergymat.de
Rechargeable Oxide Ion Batteries Based on Mixed

Conducting Oxide Electrodes
Alexander Schmid,* Martin Krammer, and Jürgen Fleig
electrode surface, rather than only at the

Mixed conducting oxides are frequently studied as electrode materials for solid triple phase boundary, and thus facilitates
oxide fuel or electrolysis cells. However, in principle their ability to change higher current densities of the respective
stoichiometry depending on the oxygen chemical potential also enables charge electrodes in fuel or electrolysis cells.
However, their mixed conducting
storage typical for battery electrodes: Formally neutral oxygen can be incorpo-
nature also gives rise to their ability to
rated by annihilating oxygen vacancies and creating electron holes, similar to change the oxygen (non)stoichiometry
the incorporation of lithium in lithium ion battery electrodes. Electrodes with depending on the oxygen chemical poten-
different reducibility should thus allow operation of an oxide ion battery. This is tial, which may be driven by either oxygen
exemplified for perovskite-type La0.6Sr0.4FeO3−δ (LSF), La0.5Sr0.5Cr0.2Mn0.8O3−δ partial pressure changes or an applied
voltage. In electrochemical impedance
(LSCrMn), and La0.9Sr0.1CrO3−δ (LSCr) electrodes with blocked oxygen sur-
measurements, this variable oxygen (non)
face exchange. Cells with thin film electrodes are prepared by pulsed laser stoichiometry manifests itself as the chem-
deposition on yttria stabilized zirconia electrolytes. Charge/discharge cycles ical capacitance, which describes the dif-
and impedance measurements are performed on half cells between 350 and ferential change in oxygen concentration
500 °C and related to the materials defect chemistry. Electrode capacities of cO per change in oxygen chemical poten-
up to 350 mA h cm−3 and <0.25 % capacity loss over 19 cycles are measured tial μO.[46] Under ideally dilute conditions
and in the absence of electron trap states,
at potentials between 0.2 and −1.1 V versus O2 at 1 bar. Full cells with LSF
the chemical capacitance of an oxide with
cathodes and LSCrMn anodes are successfully operated at 350 to 400 °C with variable oxygen content is given by
electrode related energy densities up to 70 mW h cm−3, coulomb efficiencies
−1
>99 % and good cycling performance. This novel type of rechargeable battery ∂c O 4 e 2  1 4 
Cchem = 4e 2
=  + 
may open further application fields such as elevated temperatures (>200 °C), ∂ µO kT   VO  [ e′ ] +  h •  
where common cation-based cells are not applicable so far. (1)
where [VO ], [h•], and [e′ ] are the concentra-
tions of oxygen vacancies, electron holes,
1. Introduction and electrons; and e, k, and T denote the elementary charge,
Boltzmann's constant, and temperature. By means of Equa-
Mixed ionic electronic conducting (MIEC) perovskite-type oxides tion (1), the chemical capacitance of (especially thin film) mixed
are studied extensively as electrode materials in solid oxide conductors can be used to study their defect chemistry.[23,47,48]
fuel cells (SOFCs) or solid oxide electrolysis cells (SOECs) and Conceptually, the chemical potential dependent stoichiom-
as oxygen permeation membranes.[1–10] Among others, (La,Sr) etry also gives rise to battery-like behavior, provided that oxygen
MnO3−δ (LSM),[11–14] (La,Sr)(Co,Fe)O3−δ (LSCF),[15–39] and (Ba,Sr) exchange with the atmosphere is inhibited. Electrical energy is
(Co,Fe)O3−δ (BSCF)[40–45] are under investigation. Particularly then stored chemically in the MIEC electrode by changing the
for LSCF and BSCF, the ability to conduct both ions and elec- oxygen content, i.e., oxidation/reduction of the electrode mate-
trons allows incorporation or evolution of oxygen at the entire rial, and the oxide, thus, acts as an oxide ion insertion electrode.
This is very similar to lithium ion battery (LIB) electrodes,
which store electrical energy by changing their lithium content.
A. Schmid, M. Krammer, J. Fleig Essentially, voltage driven oxygen stoichiometry changes cor-
Institute of Chemical Technologies and Analytics
TU Wien respond to charging/discharging of the chemical capacitance,
Getreidemarkt 9, Vienna 1060, Austria where self discharging via oxygen exchange with the atmos-
E-mail: [email protected] phere is prevented. Figure 1 illustrates this similarity between
SOFC and LIB electrodes.
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/aenm.202203789. The half cell potential of such an oxide ion insertion elec-
© 2023 The Authors. Advanced Energy Materials published by Wiley-VCH trode should be determined by the materials reducibility, i.e.,
GmbH. This is an open access article under the terms of the Creative the oxygen chemical potential of the transition from electronic
Commons Attribution License, which permits use, distribution and repro- to ionic defects as majority defects, for typical acceptor doped
duction in any medium, provided the original work is properly cited. oxides from holes to oxygen vacancies. At this chemical poten-
DOI: 10.1002/aenm.202203789 tial, the chemical capacitance is maximal[48] and a plateau in the
Adv. Energy Mater. 2023, 13, 2203789 2203789 (1 of 12) © 2023 The Authors. Advanced Energy Materials published by Wiley-VCH GmbH
16146840, 2023, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/aenm.202203789 by Cochrane Germany, Wiley Online Library on [28/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advenergymat.de
Figure 1. Similarity between solid oxide fuel/electrolysis cell and lithium

ion battery electrodes: Both electrodes can store formally neutral spe- Figure 2. Ragone plot, comparing the potential power end energy density
cies (O or Li), shown here in dark blue. Additionally, SOFC electrodes of oxygen ion batteries with other electrochemical technologies. Data for
can exchange neutral species with the atmosphere, indicated by the light Li-ion batteries are taken from Ref. [51], data for fuel cells, supercapaci-
dashed arrows. If the latter path is blocked, SOFC electrodes behave tors and capacitors are Ref. [52], data for high temperature Na-S batteries
analogous to LIB electrodes. from Ref. [53], and data for flow batteries from Ref. [54]. Data for the
oxygen ion battery corresponds to the predicted performance of an opti-
charge–voltage curve can be expected. Materials with different mized cell discussed below.
half cell potentials may then be combined to create an oxide ion
battery, in which oxygen, in the form of oxide ions and electrons, they do not depend on critical elements such as cobalt or
is transported from one electrode to the other and vice versa, lithium. Potential applications of oxygen ion batteries include
similar to Li+ and e− in a Li ion battery. However, even though, supply/demand balancing systems for regenerative power
conceptually, this is straightforward, investigations employing sources or other stationary applications and operation in high
oxygen incorporation for charge storage are restricted to nano- temperature environments. Figure 2 displays a Ragone plot
scaled low temperature systems with aqueous electrolytes.[49,50] comparing the expected performance of oxygen ion batteries
Systematic investigation of such oxygen insertion electrodes in with other electrochemical energy storage technologies.
solid oxide cells and realization of all solid state oxygen ion bat-
teries are not available, to the best of our knowledge.
In this study, we demonstrate that solid state oxygen ion bat- 2. La0.6Sr0.4FeO3−δ Half Cell Measurements
teries can indeed be fabricated and operated at 350 to 500 °C
with their performance being close to that expected from defect Half cell measurements were performed on mixed conducting
chemical considerations. Yttria stabilized zirconia (YSZ) single La0.6Sr0.4FeO3−δ (LSF) thin film electrodes (300 to 600 nm
crystal electrolytes with MIEC thin film electrodes are used as thick), covered by a ZrO2 layer blocking the oxygen exchange.
model cells. Charge/voltage characteristics of La0.6Sr0.4FeO3−δ These working electrodes as well as counter and reference
(LSF), La0.5Sr0.5Cr0.2Mn0.8O3−δ (LSCrMn), and La0.9Sr0.1CrO3−δ electrodes were prepared on an yttria stabilized zirconia (YSZ)
(LSCr) electrodes are determined in half cell measurements. single crystal. Figure 3a shows the reduction/oxidation curves
Measurements reveal volumetric charge densities of up to of an LSF half cell, measured for several cycles at 350 °C.
350 mA h cm−3 at potentials between 0.2 and −1.1 V. Solid During reduction, oxygen is pumped out of the electrode, thus
state oxygen ion batteries utilizing an LSF cathode and an increasing the oxygen vacancy concentration at the expense of
LSCrMn anode were fabricated, with volumetric capacities up electron holes. Insertion/removal of oxygen via reduction/oxi-
to 120 mA h cm−3 (normalized to electrode volume) at cell volt- dation thus ideally corresponds to
ages in the range of 0.6 V.
These oxygen ion batteries may offer certain advantages over 5La 0.6Sr0.4 FeO3 + 2e cc− 5La 0.6Sr0.4 FeO2.8 + O2YSZ
−
(2)
other battery technologies, such as Li-ion, sodium sulphur or
ZEBRA batteries. Consisting only of solid, non-flammable where e cc− denotes electrons in the current collector and O2−
YSZ is
oxides, they present much less of a security risk in the case of an oxide ion in YSZ. In terms of defect chemistry we may write
device failure. Furthermore, they can be fabricated mostly from
abundant elements (e.g., Fe, Ca, Ti, Cr, Mn,), and in particular, OOx + 2e cc− + 2h • VO•• + O2YSZ
−
(3)
Figure 3. a) Charge–voltage curves of an LSF half cell measured at 350 °C in 0.25 mbar O2 over 19 reduction/oxidation cycles in galvanostatic mode
with a current of 58 µA cm−2 (i.e., 5 C) between −0.01 to −1.07 V versus 1 bar O2; all curves are virtually identical. During reduction oxygen is pumped
out of the electrode. The initial cycle is not shown. b) Electrode capacity extracted from data in panel (a). c) Corresponding coulomb efficiency.
which means that by reduction and oxidation, we switch layer at the electrode surface, allowing small amounts of oxygen
between hole (h•) and vacancy ( VO••) dominated LSF. The stoichi- to leak from the surrounding atmosphere into the electrode.
ometry in Equation (2) assumes that we switch from virtually Similar measurements were performed at different temper-
complete hole compensation of the Sr doping to almost com- atures and with different currents (i.e., charge rates) and the
plete oxygen vacancy compensation. According to the Brouwer corresponding capacities and coulomb efficiencies are listed in
diagram of LSF (see below), this can be expected when oper- Table 1. For lower currents (29 µA cm−2 at 350 °C), we reach
ating in a range between 0 and 0.4 V versus O2 and should even higher capacities of ≈94 % of the theoretical capacity
result in a volumetric capacity of 295 mA h cm−3. according to Equation (2). At higher currents and especially
The volumetric charge of our electrodes was calculated by at low temperatures the overpotential losses in the thick elec-
integrating the measured current and normalizing to the total trolyte dominate the voltage and thus also strongly limit the
thin film electrode volume. Between 0 V and the sharp voltage
drop from −0.3 to −0.9 V at 58 µA cm−2, we get 253 mA h cm−3 Table 1. Electrode capacity and coulomb efficiency of an LSF half cell
electrode at different temperatures and at different charge rates. Mean
and thus 86 % of the theoretical value. The experimental value values and standard deviations over 19 cycles are listed.
corresponds to a gravimetric capacity of ≈40 mA h g−1. For com-
parison, this is ≈29 % of the volumetric and 37 % of the gravi- Temperature Current Cut-off potential Capacity Coulomb
metric capacity of the common lithium ion battery electrode [°C] [µA cm−2] efficiency [%]
material LixCoO2 with 670 and 140 mA h g−1 and x between 0.5
Lower [V] Upper [V] [mA h cm−3]
and 1, respectively.[55]
Reduction and oxidation cycles are distinctly separated by 350 29 -1.06 0.04 276.7 ± 0.1 99.5 ± 0.1
≈200 mV due to transport resistances and a corresponding 350 58 -1.07 -0.01 253.4 ± 0.3 99.6 ± 0.1
overpotential of 100 mV in each direction. From the ionic 350 115 -0.87 -0.06 151.3 ± 0.8 99.3 ± 0.7
transport resistance of the YSZ electrolyte (determined by 400 58 -0.88 -0.07 260.9 ± 0.1 97.1 ± 0.1
impedance spectroscopy, see below) we can estimate the YSZ
400 115 -0.90 -0.07 257.0 ± 0.1 98.5 ± 0.1
contribution to the overpotential to be ≈80 mV. Thus, it consti-
tutes the major part of the total overpotential. The remaining 400 288 -0.92 -0.07 230.4 ± 0.3 99.2 ± 0.1
20 mV correspond to an intermediate frequency feature in the 400 570 -0.92 -0.07 151.3 ± 0.6 99.3 ± 0.4
impedance spectra, likely caused by an interfacial resistance 400 1152 -1.11 0.18 179.2 ± 0.8 99.5 ± 0.5
between electrolyte and electrode or contributions from the
450 288 -0.90 -0.10 228.2 ± 0.3 97.8 ± 0.1
current collector.
Figure 3b displays the capacities resulting for the reduction 450 570 -0.93 -0.10 191.6 ± 1.1 98.6 ± 0.2
and oxidation steps extracted from Figure 3a. The capacity is 450 1152 -1.10 -0.10 80.5 ± 0.8 98.5 ± 1.5
virtually constant, decreasing by <0.25 % over 19 cycles. The 450 2880 -1.15 0.38 142.5 ± 1.0 100.0 ± 0.2
corresponding coulomb efficiencies of ≈99.6 % are shown in 500 288 -0.69 -0.09 235.7 ± 0.2 92.9 ± 0.2
Figure 3c. The very small loss of charge in each cycle is most
500 576 -0.77 -0.08 230.1 ± 0.8 96.2 ± 0.3
probably a consequence of the yet non-ideal oxygen blocking
Figure 4. a) Charge–voltage curves of an LSF half cell measured at 350 °C in 0.25 mbar O2 with currents between 29 and 115 µA cm−2 (i.e., 2.5 to 10 C).
b) Charge–voltage curves of an LSF half cell measured between 400 and 500 °C in 0.25 mbar O2 with a current of 288 µA cm−2.
accessible capacity range.Exemplary, charge–voltage curves semicircle at intermediate frequencies is likely caused by an
measured for different C-rates are shown in Figure 4a. Not interface impedance (i.e., charge transfer between electrolyte
surprisingly, the voltage gap between reduction and oxidation and electrode), but may also include contributions from the cur-
increases with increasing current due to the internal resist- rent collector.[20] At low frequencies, the onset of an additional
ances. Moreover, the accessible charge decreases with higher feature is visible, which is more or less capacitive. However,
C-rates. Figure 4b displays charge–voltage curves at different due to very long relaxation times at these low temperatures,
temperatures. Curves at 400 and 450 °C differ only in the only very few points could be measured within reasonable time.
overpotential (visible as the difference between charge and Thus, only a rough estimation of the corresponding capacitance
discharge curve). At 500 °C, also the capacity of the reduction is possible. The approximate value of this capacitance is rather
curve appears to be higher than at lower temperatures, and fur- high (several mF cm−2) clearly suggesting a chemical origin, i.e.
thermore, the oxidation curve does not reach its starting point this is the chemical capacitance of the MIEC electrode.
at zero volumetric charge. However, this is due to significant At higher temperatures, the YSZ high frequency semicircle is
parasitic oxygen exchange with the atmosphere leading to a loss largely lost (due to the high relaxation frequency), but a larger
of charge during the measurement. Therefore, more charge is part of the low frequency feature becomes accessible, which
required to reach the cut-off potential during charging and less develops into a semicircle. Especially at very negative potentials,
charge can be retrieved during oxidation. almost the entire low frequency semicircle is visible in the given
frequency range, see Figure 5b. The diameter of this semicircle
is several orders of magnitude larger than the electrolyte resist-
3. Impedance Measurements of La0.6Sr0.4FeO3−δ ance and corresponds to the differential resistance of not yet fully
known parasitic current paths. (Please note that the small electro-
Half Cells
lyte resistance largely determines the overpotential in the charge–
Additionally, impedance spectra were recorded in a three elec- discharge curves.) However, the capacitance of the low frequency
trode arrangement at different temperatures and potentials. arc can still be interpreted in terms of a chemical capacitance of
Figure 5a shows impedance spectra of an LSF half cell meas- the electrode. This was extracted by fitting the low frequency arc
ured at 350 °C. The high frequency semicircle is independent to an equivalent circuit consisting of an R||CPE element with a
of potential and its resistance corresponds to the ionic trans- series resistor; the CPE with impedance ZCPE = T−1(iω)−P was
port resistance of the YSZ electrolyte.[56] A small depressed used to model an imperfect capacitor with C = (R1 −PT)1/P. Typical
Figure 5. Impedance spectra of an LSF electrode (550 nm) half cell sample at a) 350 °C and b) 500 °C and varying potentials versus 1 bar O2, measured
in 0.25 mbar oxygen versus a reference electrode. Markers indicate measured data points, lines represent fits to an equivalent circuit consisting of an
R||CPE element in series with a resistor.
P values are between 0.9 and 0.95. The chemical capacitances potential of a mixed conducting electrode with high electronic
were normalized to the entire electrode film volume including conductivity and slow or blocked oxygen exchange at the sur-
the volume atop the current collector grid, since the inhibition kT p∗
face is given by µO = µO0 + ln O2 and can be varied by the
of the surface exchange causes the entire film to contribute to 2 1bar 4 eU
∗
the capacitive response. Values between 1940 mF cm−2 at open potential versus 1 bar O2, U, according to pO2 = 1bar exp .
kT
circuit and 23 mF cm−2 at a potential of −0.8 V versus 1 bar O2 ∗
The effective oxygen partial pressure pO2 in the electrode is,
result, which agrees excellently with chemical capacitance meas- thus, directly related to the electrode potential and fort given
urements on similar LSF thin films without oxygen blocking zir- thermodynamic data can be used to calculate [VO•• ] as a function
conia layer,[48] confirming our assignment of the low frequency of the potential. As a reference point, we define the oxidized
capacitance to the electrode chemical capacitance. state (Q = 0) of the electrode at a potential of 0 V, i.e. pO∗ 2 = 1 bar
([VO•• ] = 1.5 × 1017 cm −3 at 350 °C, according to data in Refs. [28, 48]).
Thus, we obtain
4. Reconstruction of the Charge/Voltage
Characteristics from the Chemical Capacitance (
Q (U ) = 2e  VO••  (U ) −  VO••  ( 0V ) ) (4)
∗
In the following, we show how the charge–voltage curve of LSF This is illustrated in Figure 6b. Hence, from the p (i.e., O2
can be directly predicted and understood from its defect model. potential) dependency of the oxygen vacancy concentration,
The defect chemistry of LSF has been investigated extensively we directly get the theoretical charge–voltage curve (with inter-
by a multitude of techniques, including coulometric titration, changed axes) in Figure 6b.
thermogravimetry, statistical thermodynamics calculations, Figure 7a shows the calculated curve from Figure 6b in the
carrier gas titration, electronic conductivity measurements conventional charge/voltage plot, together with measured data
and chemical capacitance measurements.[22,23,28,29,32,48,57–60] at 350 °C. The observed charge/voltage characteristics fit the
Generally, LSF defect thermodynamics is described by defect defect chemical predictions very well. The calculated curve
models including (doubly charged) oxygen vacancies ( VO••), is largely in between the measured reduction and oxidation
electron holes (h•) and, at low oxygen chemical potentials, curves, since it does not include the overpotential in the electro-
electrons (e′). The corresponding Brouwer diagram (defect lyte. Figure 7b displays similar data at 500 °C and here we also
concentrations versus pO2 ) is shown in Figure 6a. A refined demonstrate that the charge/voltage curve can be reconstructed
model may also consider singly charged oxygen vacancies at from the determined chemical capacitance values (measured by
low oxygen partial pressures. However, in our case, we can stay impedance spectroscopy). This chemical capacitance is nothing
with the given model since LSF is only operated in the hole but the differential capacitance dQ/dU often discussed for bat-
conducting regime. Generally, the change in electrode charge teries, and thus we can use
ΔQ is given by the sum of the changes of electronic charge car-
rier concentrations times their respective charges, which is pro- dQ (υ )
U U
portional to the change in vacancy concentration of LSF, i.e., Q (U ) = ∫Cchem (υ )dυ = ∫ dυ (5)
0 0
dυ
∆Q = e∆[e′ ] − e∆[h • ] = 2e∆[VO ] . From the Brouwer diagram, the
oxygen vacancy concentration is known as a function of oxygen to reconstruct the charge–voltage curve (υ is the electrode poten-
partial pressure. More generally, the effective oxygen chemical tial versus 1 bar). The agreement with measured charge–voltage
Figure 6. a) Brouwer diagram of LSF at 350 °C calculated from defect data in Ref. [48]. b) Equivalence of charge/voltage characteristics and Brouwer
diagram. The curve shows the oxygen vacancy concentration from panel (a) on a linear scale. The upper and right axes indicate the corresponding
voltage and charge, respectively, i.e., a rotated charge–voltage curve. Exemplary, the charge at −400 mV is shown (dashed lines), relative to our chosen
reference state at 0 V (dotted lines).
Figure 7. Charge/voltage characteristics of an LSF half cell measured at a) 350 °C/29 µA cm−2 and b) 500 °C/288 µA cm−2, together with curves cal-
culated from the defect model (Brouwer diagram). For 500 °C reconstructed data were also determined from chemical capacitance measurements by
integration, see Equation (5).
curves and the curve from the Brouwer diagram is excellent. or again
Altogether, we can conclude that LSF is a highly promising
cathode material for solid state oxygen ion batteries. OOx + 2 e cc− + 2h • VO•• + O2YSZ
−
(7)
in defect notation.
5. La0.5Sr0.5Cr0.2Mn0.8O3−δ Half Cell Measurements The reduction/oxidation characteristics at 400 °C are
shown in Figure 8. Notably, the electrode capacity for
As an anode in an oxygen ion battery, a material is required 16.9 µA cm−2 is very high (338 mA h cm−3), even surpassing
which is much harder to reduce than LSF. Chromium and that of LSF. This value is again reasonably close to the theo-
manganese are known to be much less reducible than iron and retical capacity according to Equation (6) (385 mA h cm−3),
thus (La,Sr)(Cr,Mn)O3−δ (LSCrMn) is a potential anode material. which is more than for LSF due to the increased Sr content
Here, La0.5Sr0.5Cr0.2Mn0.8O3−δ thin film electrodes were investi- (50 % Sr on the A site). Even higher capacities are achieved
gated; the corresponding redox reaction nominally reads at lower currents, see Table 2. The capacity of 392 mA h cm−3
(1404 C cm−3) at 8.5 µA cm−2 for example, corresponds to
4La 0.5Sr0.5Cr0.2 Mn 0.8 O3 + 2e cc− 4La 0.5Sr0.5Cr0.2 Mn 0.8 O2.75 + O2YSZ
−
(6) 7.28 mmol cm−3 of oxygen vacancies transferred between
Figure 8. a) Charge/voltage characteristics of an LSCrMn electrode measured in a half cell configuration at 400 °C in 0.25 mbar O2 using a current
of 16.9 µA cm−2 (0.8 C). Nineteen cycles between −0.12 and −1.12 V versus 1 bar O2 are shown. b) Extracted capacities. c) Corresponding coulomb
efficiency.
Table 2. Electrode capacity and coulomb efficiency (per cycle) of an 6. Full Cells Measurements
LSCrMn half cell at different temperatures and using different currents.
Mean values and standard deviations over 19 cycles are listed.
Full cells of solid state oxide ion batteries were fabricated by
combining two electrodes with different oxygen insertion (i.e.,
Temperature Current Cut-off potential Capacity Coulomb
half cell) potentials. Charge can then be reversibly stored and
[°C] [µA cm−2] [mA h cm−3] efficiency [%]
retrieved by pumping oxide ions between the two electrodes.
Lower [V] Upper [V] Ideally, and for matching electrode volumes, full charging and
350 8.5 −1.11 −0.11 296.3 ± 16.9 97.0 ± 0.9 discharging corresponds to oxygen insertion/removal according
to the nominal cell reaction
350 16.9 −1.11 −0.11 263.2 ± 4.7 99.7 ± 0.4
350 33.9 −1.11 −0.11 179.2 ± 5.0 99.9 ± 0.8 

5La 0.6Sr0.4 FeO2.8 + 4La 0.5Sr0.5Cr0.2 Mn 0.8 O3 ← 
charging
→
discharging (8)
400 8.5 −1.12 −0.12 391.7 ± 2.0 97.5 ± 0.1 5La 0.6Sr0.4 FeO3 + 4La 0.5Sr0.5Cr0.2 Mn 0.8 O2.75
400 16.9 −1.12 −0.12 337.8 ± 3.9 98.2 ± 0.2
Non-idealities in real cells certainly modify the exact stoichio-
400 33.9 −1.12 −0.12 323.6 ± 3.2 99.6 ± 0.7 metric factors in this equation.
400 84.7 −1.12 −0.12 251.6 ± 1.9 99.9 ± 0.2 Before we discuss full cell characteristics, one peculiarity of
400 169.5 −1.12 −0.12 138.1 ± 1.5 99.5 ± 1.6
these cells has to be addressed. After deposition, the oxygen
(non)stoichiometry is frozen in at a value corresponding to
450 16.9 −1.13 −0.13 368.9 ± 0.8 97.2 ± 0.2
equilibrium with the atmosphere during deposition. For all
450 33.9 −1.13 −0.13 357.2 ± 0.9 98.7 ± 0.2 materials investigated in this study this means that most
450 84.7 −1.13 −0.13 325.9 ± 1.3 99.7 ± 0.4 oxygen vacancies are filled after preparation, despite their dif-
ferent reducibility. However, battery operation requires that
450 169.5 −1.13 −0.13 277.8 ± 4.0 100.1 ± 0.3
one electrode has a substantial amount of oxygen vacancies,
500 16.9 −1.14 −0.14 345.5 ± 4.4 92.3 ± 1.3 whereas the other is essentially completely filled with oxygen.
500 33.9 −1.14 −0.14 350.3 ± 1.9 97.8 ± 0.1 One way to counter this problem is an initial conditioning
(i.e., reduction) step of one electrode via a third, auxiliary, elec-
500 84.7 −1.14 −0.14 330.7 ± 0.2 99.0 ± 0.1
trode which is open to the atmosphere. After this initial con-
500 169.5 −1.14 −0.14 309.5 ± 0.5 99.5 ± 0.2 ditioning step, the cell can be cycled repeatedly by pumping
oxygen from one electrode to the other and back. In our cells,
the role of this auxiliary electrode is played by the reference
the MIEC electrode and the electrolyte. Nominally, this is electrode. Figure 9 illustrates the initial conditioning step
equal to 0.254 oxygen vacancies per formula unit of LSCrMn (Figure 9a), and the repeated charging (Figure 9b) and dis-
(δ = 0.254) and thus we conclude that the stoichiometry charging (Figure 9c) cycles.
changes in LSCrMn are indeed close to those indicated The addition of an auxiliary electrode also adds an
by Equation (6). Except close to fully oxidized, the reduc- advantageous feature to oxygen ion batteries: It allows to
tion and oxidation cycles are separated by an overpotential replenish lost oxygen (or, more precisely, oxygen vacancies) via
which almost matches to the voltage drop due to the elec- the open auxiliary electrode. Hence, capacity losses due to para-
trolyte resistance. Hence, the electrode itself again causes sitic oxygen exchange with the atmosphere can be regenerated, in
no major transport resistance. The half-cell potential is at theory infinitely. This is in contrast to, e.g., lithium ion batteries,
≈−0.65 V versus 1 bar O2. For 19 cycles the capacity even where Li losses due to parasitic reactions are often permanent
slightly increases. Exact reasons behind this 0.25 % increase and mostly cause irreversible degradation of the cell capacity.
per cycle are not known yet. Possibly some defect chemical Figure 10 displays charge/discharge characteristics of a full
or crystallinity changes may play a role. The coulomb effi- cell with LSF cathode and LSCrMn anode. The cell capacity
ciency is at ≈98.5 %, most probably due to some parasitic is almost 100 mA h cm−3, normalized to the volume of both
oxygen exchange reaction. For comparison, we also tested an electrodes. A slight drift of the cell capacity is visible, see
La0.9Sr0.1CrO3−δ thin film electrode. This lead to an even more Figure 10b, most likely due to parasitic oxygen exchange with
negative electrode potential (≈−1.1 V) but a much smaller the atmosphere, which also corresponds to coulomb efficien-
capacitance. Details are given in the Supporting Information. cies slightly below unity, see Figure 10c. It is not yet fully
Similar measurements at different temperatures and with understood why the coulomb efficiency is significantly lower
different currents showed the same trends as observed for LSF in the first cycle. After equilibrating the cell with the atmos-
electrodes, see Table 2. At lower temperatures and higher cur- phere the same behaviour is observed reproducibly, indicating
rents, a significant amount of the voltage drops at the electro- that it is not due to irreversible side reactions. A cell voltage
lyte resistance. Hence, the actual voltage window across which in the range of 0.6 V is found, which agrees with the different
the electrode is cycled becomes reduced. Consequently, the half cell potentials of LSF and LSCrMn. The volumetric energy
accessible electrode capacity is lower at higher currents and density of the cell, calculated from the charge–voltage curve
lower temperatures. The maximum value of 392 mA h cm−3 by integration is 56 mW h cm−3 (normalized again to the elec-
(65 mA h g−1) is found at a rate of 0.34 C at 400 °C. Altogether, trode volume).
LSCrMn turns out to be a promising anode material for oxygen Similar measurements with varying temperatures and cur-
ion batteries. rents revealed even an increase of the cell capacity with either
Figure 9. a) Initial conditioning: Oxygen is pumped out of the LSF cathode into the atmosphere, creating oxygen vacancies at the expense of electron
holes. b) Charging: Oxide ions are pumped from the LSCrMn anode to the LSF cathode, filling the vacancies and forming holes. c) Discharging: Oxide
ions flow from the LSF cathode to the LSCrMn anode.
higher temperatures or lower currents. Up to 119 mA h cm−3 achievable with improved electrodes, e.g., highly doped (La,Sr)
(20 mA h g−1) and 69 mW h cm−3 (11 mW h g−1) were measured CrO3 anodes and optimized cathodes. For the electrolyte we
at 400 °C using a current of 8.5 µA cm−2, which corresponds to assume a thin film of Sm:CeO2, which has a higher conduc-
a rate of 1.3 C. Again, the overpotential can be mostly attributed tivity than YSZ and has also reduced interfacial resistances at
to the thick electrolyte. Exemplary charge/voltage curves for cathodes.[61–63] To estimate the internal resistance of such a cell
different current densities and temperatures are shown in we take the ionic conductivities of Sm:CeO2 and La0.6Sr0.4FeO3−δ
Figure 11, and Table 3 lists the cell capacity, volumetric energy from Refs. [61] and [16], respectively. Since the conductivity
density and coulomb efficiency under varying conditions. of La0.6Sr0.4FeO3−δ depends on oxygen vacancy concentra-
tion, i.e., the electrode charge state, we use the conductivity
at δ = 0.043 which corresponds to an electrode where 78.5%
7. Predicted Performance of an Optimized Thin of vacancies are filled. This is the logarithmic mean between
completely empty (δ = 0.2) and 99 % filled (δ = 0.002). For
Film Cell
example, the assumed ionic conductivity of both electrodes is
Finally, we extrapolate these data to predict the performance of 2.8 × 10−4 S cm−1 at 350 °C. Additionally, we include an extra
an optimized solid state oxide ion battery. We assume anode 25 % of the cell volume to account for thin film electrolyte, elec-
and cathode materials with a volumetric charge density of trical contacts and packaging.
350 mA h cm−3 each and for both an ionic conductivity sim- For a cell with an electrolyte being much thinner than the
ilar to that of LSF. An open circuit cell voltage of 1 V is likely electrodes, this results in volumetric energy densities of up to
Figure 10. a) Charge–voltage curves of a full cell with La0.6Sr0.4FeO3−δ cathode and La0.5Sr0.5Cr0.2Mn0.8O3−δ anode, measured at 350 °C in 0.25 mbar
O2 with a current of 8.5 µA cm−2 (1.6 C) between 0 and 1 V. 9 cycles are shown; the initial charging step is excluded. Capacities are normalized to the
combined volume of both electrodes. b) Electrode capacity extracted from data in panel (a). c) Corresponding coulomb efficiency.
Figure 11. a) Charge–voltage curves of a full cell with La0.6Sr0.4FeO3−δ cathode and La0.5Sr0.5Cr0.2Mn0.8O3−δ anode measured with different currents at
400 °C. b) Charge–voltage curves of a full cell measured with a current of 8.5 µA cm−2 at different temperatures.
140 mW h cm−3, which corresponds to gravimetric energy den- on electrode thickness and cycling rate is also visualized in
sities of up to 30 mW h g−1. For comparison, this is ≈30 % of Figure S2 (Supporting Information). Ideal geometries of prac-
the volumetric energy density of current lithium ion batteries tical devices may depend on the specific applications. Multi-
at 450 mW h cm−3.[51] Depending on cycling rate and operating layer stacks similar to multilayer ceramics capacitors are fea-
temperature, the internal cell resistance limits the electrode sible, with the individual electrode thickness determining the
thickness, but this can be coped with by stacking multiple thin achievable power density.
film cells. Figure 12a shows the predicted volumetric energy
and power densities for a single cell or a stack, both with 10 µm
thick electrodes and a 200 nm electrolyte layer in the tempera- 8. Conclusion
ture range of 200–350 °C. At higher temperatures, electrode
thicknesses up to 50 µm should be possible, see Figure 12b. MIEC oxides can act as electrochemical oxygen storage elec-
For example, cycle rates up to 10 C at 200 °C would corre- trodes if oxygen exchange with the atmosphere is inhibited. This
spond to power densities up to 700 mW cm−3 (145 mW g−1). At was shown for different MIEC thin film electrodes, grown on
350 °C and above up to 12 W cm−3 (2.5 W g−1) are predicted. YSZ electrolyte substrates, covered by an oxygen blocking zir-
Operation much below 200 °C is hardly feasible with common conia layer and characterized at 350 to 500 °C. La0.6Sr0.4FeO3−δ
electrode/electrolyte materials due to the high resistances of (LSF) was investigated as a model cathode and the charge/voltage
ion transport. At room temperature, for example, power den- characteristics were in very good agreement with those pre-
sities below 1 mW cm−3 are expected. However, these slug- dicted by chemical capacitance measurements and defect chem-
gish kinetics should also largely prevent self discharge after ical models. Capacities of up to 277 mA h cm−3 (45 mA h g−1)
cooling down. The dependence of energy and power density were found at a half cell potential of −0.3 V vs. 1 bar O2.
Table 3. Cell capacity, volumetric energy density and coulomb efficiency of an LSF/LSCrMn full cell extracted from charge/voltage measurements at
different temperatures and using different currents. Mean values and standard deviations over 9 cycles are listed.
Temperature [°C] Current [µA cm−2] Cut-off voltage Capacity [mA h cm−3] Energy density [mW h cm−3] Coulomb efficiency [%]
Lower [V] Upper [V]
350 3.4 0 1.0 117.3 ± 0.2 68.0 ± 0.1 98.8 ± 0.1
350 8.5 0 1.0 98.3 ± 0.7 55.3 ± 0.4 99.2 ± 0.6
350 16.9 0 1.0 88.9 ± 0.6 45.1 ± 0.4 99.6 ± 1.0
350 33.9 0 1.0 70.2 ± 1.0 29.7 ± 0.6 99.1 ± 2.7
400 8.5 0 1.0 118.9 ± 0.9 69.3 ± 0.5 99.9 ± 0.1
400 16.9 0 1.0 113.9 ± 0.5 65.1 ± 0.3 99.9 ± 0.2
400 33.9 0 1.0 102.2 ± 0.6 56.8 ± 0.4 99.9 ± 0.3
400 84.7 0 1.0 84.1 ± 0.7 42.1 ± 0.4 99.6 ± 1.1
400 84.7 0 1.25 102.9 ± 1.1 56.5 ± 0.7 99.7 ± 0.5
400 169.5 0 1.0 68.3 ± 0.9 27.9 ± 0.6 98.9 ± 2.8
400 339.0 0 1.2 60.5 ± 0.2 18.2 ± 0.4 98.1 ± 3.3
Figure 12. Predicted energy densities (solid lines) and power densities (dashed lines) of optimized oxygen ion batteries with thin film electrolytes.
a) 10 µm electrodes and 200 nm electrolyte. b) 50 µm electrodes and 1 µm electrolyte.
La0.5Sr0.5Cr0.2Mn0.8O3−δ (LSCrMn) anodes exhibited even higher PLD parameters for the different materials are summarized in
capacities of up to 390 mA h cm−3 (65 mA h g−1) at a half cell Table 4 and a sketch of the resulting geometry is shown in Figure 13a.
Figure 13c,d shows a microscope image of the sample surface and a
potential of −0.65 V. Both electrodes showed very good cycling
SEM image of the sample cross section, respectively.
performance with coulomb efficiencies above 99 %. Ablation targets were produced from powders by pressing and
A full oxide ion battery was fabricated with LSF cathode sintering in air at 1200 °C for 12 h. Pure zirconia powder was purchased
and LSCrMn anode. DC measurements on this cell showed from Alfa Aeser (99.7 %), perovskite powder Pechini synthesis: Metal
capacities as high as 120 mA h cm−3 (normalized to the elec- precursors (SrCO3, La2O3, Fe, Cr(NO3)3, MnCO3, all Sigma–Aldrich
trode volume) at a cell voltage of 0.6 V at 350–400 °C. Again >99.995%) were dissolved in nitric acid and citric acid was added in a 1:1
ratio with respect to total cation amount. The solution was then heated
excellent cycling performance was found with <1 % of the
and evaporated until ignition and self combustion took place, and the
charge being lost per cycle. An additional auxiliary electrode is obtained powder was calcined in air at 850 °C for 8 h. Pe purity of the
required during the initial conditioning cycle, but may also be sintered targets was confirmed by X-ray diffraction.
used to regenerate lost charge. Altogether, these results dem- Electrochemical Measurements: Working and counter electrodes were
onstrate that MIEC based oxide ion batteries are a viable novel contacted by clamping the sample between two platinum meshes, the
electrochemical energy storage device operating at elevated reference electrode was contacted with a fine Pt wire loop around the
temperatures of typically 200–500 °C. Further optimization by sample circumference. The sample holder was then placed in a sealed
tube of fused silica inside a tube furnace. Figure 13b shows a sketch of
increasing the cell voltage using other electrode materials is the sample mounted inside the sample holder.
highly realistic and promises power and energy densities of up Samples were allowed to equilibrate with the measurement
to 12 W cm−3 (at 350 °C) and 140 mW h cm−3, respectively. atmosphere prior to electrochemical measurements by connecting all
three electrode leads and keeping the sample at elevated temperature
until no current was detectable between the electrodes. This was
done in order to establish a defined oxidation state of all electrodes
9. Experimental Section independent of thermal history and oxygen partial pressure during
Sample Preparation: Samples were made from yttria stabilized electrode deposition.
zirconia (YSZ) single crystals (5 × 5 × 1 mm3), with the (100) direction DC current voltage measurements were performed with a Keithly
perpendicular to the square sides and a notch (300 × 300 µm2) carved 2600 source meter in galvanostatic mode in a 4-wire configuration,
around the circumference. A reference electrode was prepared by with currents from 5 to 500 µA and at temperatures from 350 to 500 °C
brushing Pt paste into this notch and sintering in air at 1200 °C for 2 h. in 0.25 mbar O2. Half cells as well as full cells were typically cycled
Platinum current collector grids of 100 nm thickness and 15/35 µm
strip/mesh width were prepared on both large sides of the electrolyte Table 4. Pulsed laser deposition parameters for oxide thin films.
by magnetron sputtering and subsequent photo-lithography and
ion beam etching. A 5 nm thick Ti layer was deposited beneath the Pt LSF LSCr LSCrMn LSC (porous) ZrO2
for better adhesion. Mixed conducting perovskite thin film electrodes Laser pules 43 35 31 25 38 [nm−1]
(300 to 600 nm thickness) were then grown on top of the current collector
grids by pulsed laser deposition (PLD). Finally, a dense layer of undoped Laser fluence 1.1 1.1 1.1 1.4 1.1 [J cm2]
zirconia (1 µm thick) was deposited on top of the electrodes by PLD to Pulse frequency 10 10 10 5 10 [Hz]
isolate them from the outer atmosphere. Two types of cells were produced:
Oxygen pressure 0.04 0.015 0.015 0.4 0.015 [mbar]
1) Half cell samples consisting of a dense working electrode with Substrate 600 700 700 450 650 [°C]
oxygen barrier layer (as described above) and a catalytically highly temperature
active counter electrode of nano-porous PLD grown La0.6Sr0.4CoO3−δ.
Target to substrate 6 6 6 6 6 [cm]
2) Full cells with dense working and counter electrodes both covered by
distance
a zirconia oxygen barrier layer.
Data Availability Statement

The data that support the findings of this study are openly available
in Oxygen Ion Batteries at 10.5281/zenodo.7298433, reference
number 7298433.
Keywords
chemical capacitance, electrochemical energy storage, intercalation
batteries, mixed ionic electronic conducting oxides, solid state batteries
Received: November 8, 2022

Revised: December 21, 2022
Published online: January 25, 2023
[1] X. M. Ge, S. H. Chan, Q. L. Liu, Q. Sun, Adv. Energy Mater. 2012, 2,

1156.
[2] A. Jun, J. Kim, J. Shin, G. Kim, ChemElectroChem 2016, 3, 511.
[3] M. A. Laguna-Bercero, J. Power Sources 2012, 203, 4.
[4] N. Mahato, A. Banerjee, A. Gupta, S. Omar, K. Balani, Prog. Mater.
Sci. 2015, 72, 141.
Figure 13. a) Sketch of a full cell sample consisting of working electrode [5] I. Z. Rahman, M. A. Raza, M. A. Rahman, Adv. Mater. Res. 2012, 445,
(1), reference electrode (2), YSZ substrate (3), counter electrode (4), plat- 497.
inum current collector(5) and zirconia capping layer (6). b) Measurement [6] C. Sun, R. Hui, J. Roller, J. Solid State Electrochem. 2010, 14, 1125.
setup consisting of: Sealed outer tube (1), spring loaded middle tube (2),
[7] C. Sun, U. Stimming, J. Power Sources 2007, 171, 247.
thermocouple (3), gas inlet (4), inner tube (5), impedance analyzer (6),
[8] B. Timurkutluk, C. Timurkutluk, M. D. Mat, Y. Kaplan, Renewable
spacer (7), Pt contacting mesh (8), sample (9). c) Microscope image of
Sustainable Energy Rev. 2016, 56, 1101.
an LSF half cell, top view. d) Secondary electron image of an LSF half cell,
cross section. [9] C. Xia, M. Liu, Adv. Mater. 2002, 14, 521.
[10] J. M. Ralph, C. Rossignol, R. Kumar, J. Electrochem. Soc. 2003, 150,
A1518.
multiple times between a lower and upper cut-off voltage depending [11] R. De Souza, Solid State Ionics 1998, 106, 175.
on material, temperature and current. The voltage at the working [12] X. J. Chen, K. A. Khor, S. H. Chan, Solid State Ionics 2004, 167, 379.
electrode was measured with respect to the reference electrode, i.e., with [13] J. Fleig, H. R. Kim, J. Jamnik, J. Maier, Fuel Cells 2008, 8, 330.
respect to oxygen at the atmospheric partial pressure of 0.25 mbar. The [14] Y. A. Mastrikov, R. Merkle, E. Heifets, E. A. Kotomin, J. Maier,
potential against 1 bar oxygen was then calculated by Nernst's equation. J. Phys. Chem. C 2010, 114, 3017.
Additionally, the cell voltage between working and counter electrode was [15] B. Steele, Solid State Ionics 1998, 106, 255.
measured for full cell measurements. [16] J. Bahteeva, I. Leonidov, M. Patrakeev, E. Mitberg, V. Kozhevnikov,
Impedance spectra were recorded with a Novocontrol Alpha-A K. Poeppelmeier, J. Solid State Electrochem. 2004, 8, 578.
frequency analyser with a PotGal interface at 350 to 500 °C in 0.25 mbar
[17] F. S. Baumann, J. Fleig, G. Cristiani, B. Stuhlhofer,
O2 over a frequency range of 106–10−2 Hz with an AC amplitude of 10 mV.
H.-U. Habermeier, J. Maier, J. Electrochem. Soc. 2007, 154, B931.
A three wire configuration was used, i.e., currents were measured/
[18] F. S. Baumann, J. Maier, J. Fleig, Solid State Ionics 2008, 179, 1198.
applied between working and counter electrode, voltages between
working and reference electrode. Additionally, DC bias voltages between [19] F. S. Baumann, J. Fleig, M. Konuma, U. Starke, H.-U. Habermeier,
200 and −1000 mV were applied to the working electrode. J. Maier, J. Electrochem. Soc. 2005, 152, A2074.
[20] F. S. Baumann, J. Fleig, H. U. Habermeier, J. Maier, Solid State
Ionics 2006, 177, 1071.
[21] E. V. Bongio, H. Black, F. C. Raszewski, D. Edwards, C. J. Mcconville,
Supporting Information V. R. W. Amarakoon, Fuel Cells 2005, 14, 193.
[22] E. Bucher, W. Sitte, J. Electroceram. 2004, 13, 779.
Supporting Information is available from the Wiley Online Library or
[23] T. Kawada, J. Suzuki, M. Sase, A. Kaimai, K. Yashiro, Y. Nigara,
from the author.
J. Mizusaki, K. Kawamura, H. Yugami, J. Electrochem. Soc. 2002,
149, E252.
[24] M. C. Kim, S. Park, H. Haneda, J. Tanaka, S. Shirasaki, Solid State
Acknowledgements Ionics 1990, 40-41, 239.
[25] S. Kogler, A. Nenning, G. M. Rupp, A. K. Opitz, J. Fleig, J. Electro-
This project has received funding from the European Union's Horizon chem. Soc. 2015, 162, F317.
2020 research and innovation programme under grant agreement No [26] M. Kubicek, T. M. Huber, A. Welzl, A. Penn, G. M. Rupp, J. Bernardi,
824072 (Harvestore). M. Stöger-Pollach, H. Hutter, J. Fleig, Solid State Ionics 2014, 256, 38.
[27] J. Mirzababaei, S. Chuang, Catalysts 2014, 4, 146.
[28] M. Kuhn, S. Hashimoto, K. Sato, K. Yashiro, J. Mizusaki, Solid State
Conflict of Interest Ionics 2011, 195, 7.
[29] J. Mizusaki, T. Sasamoto, W. R. Cannon, H. K. Bowen, J. Am. Ceram.
The authors declare no conflict of interest. Soc. 1983, 66, 247.
[30] A. Nenning, A. K. Opitz, C. Rameshan, R. Rameshan, R. Blume, [47] W. C. Chueh, S. M. Haile, Phys. Chem. Chem. Phys. 2009, 11, 8144.
M. Hävecker, A. Knop-Gericke, G. Rupprechter, B. Klötzer, J. Fleig, [48] A. Schmid, G. M. Rupp, J. Fleig, Phys. Chem. Chem. Phys. 2018, 20,
J. Phys. Chem. C 2016, 120, 1461. 12016.
[31] A. K. Opitz, A. Nenning, C. Rameshan, R. Rameshan, R. Blume, [49] J. T. Mefford, W. G. Hardin, S. Dai, K. P. Johnston, K. J. Stevenson,
M. Hävecker, A. Knop-Gericke, G. Rupprechter, J. Fleig, B. Klötzer, Nat. Mater. 2014, 13, 726.
Angew. Chem., Int. Ed. 2015, 54, 2628. [50] M. Hibino, T. Kimura, Y. Suga, T. Kudo, N. Mizuno, Sci. Rep. 2012,
[32] M. V. Patrakeev, I. A. Leonidov, V. L. Kozhevnikov, 2, 601.
K. R. Poeppelmeier, J. Solid State Chem. 2005, 178, 921. [51] N. Muralidharan, E. C. Self, M. Dixit, Z. Du, R. Essehli,
[33] G. M. Rupp, H. Téllez, J. Druce, A. Limbeck, T. Ishihara, J. Kilner, R. Amin, J. Nanda, I. Belharouak, Adv. Energy Mater. 2022, 12,
J. Fleig, J. Mater. Chem. A 2015, 3, 22759. 2103050.
[34] G. M. Rupp, A. Schmid, A. Nenning, J. Fleig, J. Electrochem. Soc. [52] V. Aravindan, J. Gnanaraj, Y.-S. Lee, S. Madhavi, Chem. Rev. 2014,
2016, 163, F564. 114, 11619.
[35] M. Sase, J. Suzuki, K. Yashiro, T. Otake, A. Kaimai, T. Kawada, [53] H. D. Yoo, E. Markevich, G. Salitra, D. Sharon, D. Aurbach, Mater.
J. Mizusaki, H. Yugami, Solid State Ionics 2006, 177, 1961. Today 2014, 17, 110.
[36] J. E. ten Elshof, J. Electrochem. Soc. 1997, 144, 1060. [54] M. V. Holland-Cunz, F. Cording, J. Friedl, U. Stimming, Front. Energy
[37] N. Tsvetkov, Q. Lu, L. Sun, E. J. Crumlin, B. Yildiz, Nat. Mater. 2016, 2018, 12, 198.
15, 1010. [55] B. Xu, D. Qian, Z. Wang, Y. S. Meng, Materials Science and
[38] H. Wang, K. J. Yakal-Kremski, T. Yeh, G. M. Rupp, A. Limbeck, Engineering: R: Reports 2012, 73, 51.
U. Fleig, S. A. Barnett, J. Electrochem. Soc. 2016, 163, 581. [56] C. Ahamer, A. K. Opitz, G. M. Rupp, J. Fleig, J. Electrochem. Soc.
[39] J. Yoo, A. Verma, S. Wang, A. J. Jacobson, J. Electrochem. Soc. 2005, 2017, 164, F790.
152, A497. [57] J. Mizusaki, M. Yoshihiro, S. Yamauchi, K. Fueki, J. Solid State
[40] A. Yan, M. Cheng, Y. Dong, W. Yang, V. Maragou, S. Song, Chem. 1985, 58, 257.
P. Tsiakaras, Appl. Catal., B 2006, 66, 64. [58] J. Mizusaki, M. Yoshihiro, S. Yamauchi, K. Fueki, J. Solid State
[41] J. F. Vente, S. McIntosh, W. G. Haije, H. J. M. Bouwmeester, J. Solid Chem. 1987, 67, 1.
State Electrochem. 2006, 10, 581. [59] A. M. Ritzmann, A. B. Mu noz-García, M. Pavone, J. A. Keith,
[42] S. Gangopadhayay, T. Inerbaev, A. E. Masunov, D. Altilio, E. A. Carter, Chem. Mater. 2013, 25, 3011.
N. Orlovskaya, ACS Appl. Mater. Interfaces 2009, 1, 1512. [60] M. Søgaard, P. Vang Hendriksen, M. Mogensen, J. Solid State
[43] S. Lee, Y. Lim, E. A. Lee, H. J. Hwang, J. W. Moon, J. Power Sources Chem. 2007, 180, 1489.
2006, 157, 848. [61] M. G. Bellino, D. G. Lamas, N. E. Walsöe de Reca, Adv. Funct.
[44] C. Niedrig, S. Taufall, M. Burriel, W. Menesklou, S. F. Wagner, Mater. 2006, 16, 107.
S. Baumann, E. Ivers-Tiffée, Solid State Ionics 2011, 197, 25. [62] R. J. Milcarek, K. Wang, R. L. Falkenstein-Smith, J. Ahn, Int. J.
[45] F. S. Baumann, J. Fleig, H. U. Habermeier, J. Maier, Solid State Hydrogen Energy 2016, 41, 9500.
Ionics 2006, 177, 3187. [63] D. Chen, G. Yang, Z. Shao, F. Ciucci, Electrochem. Commun. 2013,
[46] J. Jamnik, J. Maier, S. Pejovnik, Electrochim. Acta 1999, 44, 4139. 35, 131.

Advanced Energy Materials - 2023 - Schmid - Rechargeable Oxide Ion Batteries Based On Mixed Conducting Oxide Electrodes

Uploaded by

Copyright:

Available Formats

Advanced Energy Materials - 2023 - Schmid - Rechargeable Oxide Ion Batteries Based On Mixed Conducting Oxide Electrodes

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Advanced Energy Materials - 2023 - Schmid - Rechargeable Oxide Ion Batteries Based On Mixed Conducting Oxide Electrodes

Uploaded by

Copyright:

Available Formats

Research Article

Rechargeable Oxide Ion Batteries Based on Mixed

electrode surface, rather than only at the

Figure 1. Similarity between solid oxide fuel/electrolysis cell and lithium

350 33.9 −1.11 −0.11 179.2 ± 5.0 99.9 ± 0.8 

Data Availability Statement

Received: November 8, 2022

[1] X. M. Ge, S. H. Chan, Q. L. Liu, Q. Sun, Adv. Energy Mater. 2012, 2,

You might also like