Fe(VI) catalyzed manganese redox chemistry: Permanganate and super-iron alkaline batteries

Article abstract of DOI:10.1021/jp012178t

In principle, alkali permanganates represent a substantial cathodic charge source for electrochemical storage, but high rate charge transfer has been inefficient. This study presents a novel Fe(VI) species (super-iron) and manganese redox chemistry synerg

Full text of DOI:10.1021/jp012178t

J. Phys. Chem. B 2001, 105, 11933-11936
11933
Fe(VI) Catalyzed Manganese Redox Chemistry: Permanganate and Super-Iron Alkaline
Batteries
Stuart Licht,* Susanta Ghosh, Vera Naschitz, Nadezhda Halperin, and Leonid Halperin
Department of Chemistry and Institute of Catalysis, Technion Israel Institute Of Technology,
Haifa, 32000, Israel
ReceiVed: June 7, 2001; In Final Form: September 19, 2001
In principle, alkali permanganates represent a substantial cathodic charge source for electrochemical storage,
but high rate charge transfer has been inefficient. This study presents a novel Fe(VI) species (super-iron) and
manganese redox chemistry synergism. Also presented is the high discharge energies from cathodes which
utilize this phenomenon in a conventional cylindrical battery configuration. Batteries formed with a cathode
combining BaFeO4 and KMnO4, and using a conventional zinc alkaline anode in an AAA cylindrical cell
configuration, exhibit an unusually high discharge capacity of 2.2 Wh. The discharge efficiency of solid
alkaline permanganate or manganate cathodes is probed, and improved charge transfer of Mn(VII) redox
chemistry in the presence of Fe(VI) is demonstrated. In a reciprocal manner, enhancement of Fe(VI) charge
transfer are demonstrated upon inclusion of manganese salts, and also that added cesium further improves
the combined Fe(V)/Mn(VII) cathode.
1. Introduction
2. Experimental Section
Solution phase alkaline permanganate provides a high power
and capacity cathodic discharge.1,2 Yet for solid alkaline
permanganate cathodes, effective reductive charge transfer has
remained a challenge. Whereas variations of solid manganese
dioxide salts have been of wide interest as commercial cathodes,
this is not the case for the higher valence, higher intrinsic
capacity, manganese salts such as the alkali or alkali earth
permanganates or manganates.3-5
We have previously detailed the synthesis of K2FeO4 and
BaFeO4 salts and their use as a cathode.5-8 A cathode composite
is formed by mixing an Fe(VI) salt (either K2FeO4, BaFeO4),
and/or manganese salt, with graphite and electrolyte. CsMnO4
is synthesized by precipitation from KMnO4 with Cs2(SO4),11
and other materials are commercial analytical grade.
In the experiments, components are removed from standard
commercial AAA alkaline cells (a cylindrical cell configuration
with diameter 10.4 mm, and a 42 mm cathode current collector
case height), and the outer MnO2 mix, replaced with a pressed
super-iron and/or (per)manganate salt mix; followed by reinser-
tion of the separator, Zn anode mix, gasket, and anode collector
and resealing of the cell. The cathode composites contain various
cathode salts with added graphite as a matrix to support the
cathode reduction. Cells were discharged as indicated at a 2.8
Ω, a 75 Ω, or a 275 Ω, constant resistance load, or in one case
a constant power of 0.7 W. Cell potential time variation was
measured via PC/LabView data acquisition, and cumulative
discharge, as ampere hours, determined by subsequent integra-
tion.
We recently introduced batteries with several energy and
environmental advantages based on Fe(VI) species6-9 whose
chemistry had been relatively unexplored,10 and in accord with6
FeO42- + 5/2H2O + 3e- f 1/2Fe2O3 + 5OH-
(1)
The term “ferrate” has been variously applied to both Fe(II)
and Fe(III) compounds. Due to their highly oxidized iron basis,
multiple electron transfer, and high intrinsic energy, we refer
to cathodes containing Fe(VI) compounds as “super-iron”
cathodes. A low level (2-5 wt %) of permanganate or
manganate salts when added to Fe(VI) salts, is sufficient to
renew an Fe(VI) salt when blocked by a small presence of a
passivating Fe(III) oxide, which we generalized for either
permanganate or manganate with7
3. Results and Discussion
This study probes cathodic charge transfer through the
preparation and use of sealed cylindrical alkaline cells. Although
not isolating the cathode redox couple as effectively as in a
galvanostatic, or potentiostatic, three-electrode configuration,
this configuration has several advantages: the same type of
anode was used throughout the study (as removed from the
commercial Zn alkaline cell) and exhibits facile (low polariza-
tion, efficient) charge transfer. Hence, the charge-transfer
attributes of the cathode are demonstrated, and which limit the
observed discharge characteristics of these cells. Advantages
of this two-electrode cell are its reproducibility and sealing
(preventing competing oxygen reduction effects), that the cell
requires a minimum of electrolyte, and that it provides a direct
2MnO4- + Fe2O3 + 2OH- f 2MnO2 + 2FeO42- + H2O
(2)
3MnO42- + H2O + Fe2O3 f 3MnO2 +2OH- + 2FeO4
2-
(3)
This study presents a novel Fe(VI) species (super-iron) and
manganese redox chemistry intercatalysis, and the high dis-
charge energies from cathodes which utilize this phenomenon
in a conventional alkaline cylindrical battery configuration.
* Corresponding author. E-mail: chrlicht@techunix.technion.ac.il.
10.1021/jp012178t CCC: $20.00 © 2001 American Chemical Society
Published on Web 11/02/2001

11934 J. Phys. Chem. B, Vol. 105, No. 48, 2001
Letters
Figure 1. Cell potential and energy capacity of alkaline super-iron AAA cells. Cells contain various indicated weight fractions of Fe(VI), Mn(VII),
and Mn(VI) salts, in the cathode mix, and use either a KOH or CsOH electrolyte. The “pure” cathode mix includes 3.5 g of KMnO4, 4.9 g of
CsMnO4, or 4.1 g of BaMnO4 in the 9 wt % graphite mix. The combined mass of a BaFeO4/KMnO4 cell cathode increases smoothly with the
increasing fraction of BaFeO4, from 3.5 g of KMnO4 in the 0% BaFeO4 cathode to 4.3 g in a pure BaFeO4 cathode. The sodium permanganate mix
also includes solid NaOH to avoid an over wet mix, as well as a higher graphite content (2.1 g of NaMnO4‚H2O and NaOH in a 9:1 weight ratio)
mixed with 32 wt % graphite. KMnO4 or BaMnO4 cathodes contain 9 wt % 18 m KOH electrolyte; while CsMnO4 contains 9 wt % 18m CsOH,
and the NaMnO4‚H2O cathode uses no added electrolyte (water is inherent in the salt). Top and middle sections: cells discharged at low constant
load rate (75 or 275 Ω). Lowest section: cells discharged at high constant load rate using a constant load of 2.8 Ω or a constant power of 0.7 W.
electrochemical comparison to a well-characterized (widely
distributed, commercial AAA alkaline) existing cell which
provides ∼0.3 Wh at high (0.7 W) discharge and ∼1.4 Wh at
low discharge rate.6
Composite alkaline BaFeO4 cathodes exhibit high energy
capacities.6,7,9 This “super-iron” battery energy is seen in the
95% BaFeO4 curve in the top of Figure 1, exhibiting a capacity
of 1.8 Wh to a 0.8V discharge. Included are this study’s novel
even higher discharge storage capacity, which may be derived
in alkaline cells containing both Fe(VI) and Mn(VII) salts. The
cells are prepared with identical zinc anodes and separators, as
removed from commercial AAA alkaline cells. Under these
conditions, and as seen in the figure’s middle, a cathode
comprised of only KMnO4, exhibits less than half of the capacity
of the BaFeO4 cathode. As previously noted,7 Fe(VI) discharge
is blocked by a small presence of a passivating Fe(III) oxide.
The 75 Ω energy capacity of pure BaFeO4 is 1.75 Wh with a
discharge profile highly similar to that of the 95% BaFeO4/5%
KMnO4 composite cathode detailed in the top section of Figure
1. Cells containing high fractions of both BaFeO4 and KMnO4
discharge to a higher capacity than cathodes comprised of either
salt alone. Hence, a cathode with either 50:50 or 75:25 relative
weight percent of BaFeO4 to KMnO4 yields respective discharge
capacities of 1.83 and 1.95 Wh. Finally as seen in the figure,
utilization of a CsOH, rather than KOH electrolyte further
enhanced the discharge energy to 2.1-2.2 Wh. The resultant

Letters
J. Phys. Chem. B, Vol. 105, No. 48, 2001 11935
cathode to 340 or 290 mAh/g, respectively, in 30%/70% or 3%/
97% KMnO4 /BaFeO4 composite cathodes.
In the storage cell, low solubility, or insolubility is preferred
to minimize parasitic cathode/anode interactions. As we have
previously summarized, the lighter alkali permanganate salts
have a high aqueous solubility, e.g., 4 m (m ≡ molal) for
LiMnO4, and 0.5 m for KMnO4, whereas the respective
solubility of 0.07 and 0.01 m for RbMnO4 and CsMnO4 is very
low. Whereas the solubility of the alkali earth magnesium,
calcium, strontium and barium permanganates, is very high (e.g..
9, 8, and 2 m respectively for Ca(MnO4)2, Sr(MnO4)2 and Ba-
(MnO4)2), the solubility of barium manganate, BaMnO4, is very
low.7 BaFeO4 is insoluble in water. As with other Fe(VI) salts,
the permanganate salts also generally exhibit a rapid decrease
in solubility with increases in hydroxide concentration.6,7
Figure 1, middle, summarizes the measured discharge of
several pure permanganate or manganate cathode alkaline AAA
cells. Despite the lower intrinsic Mn(VI f IV) capacity of the
barium manganate salt, this salt’s cathode approaches 1.0 Wh,
yielding a higher discharge capacity than the sodium or
potassium permanganate cathode cells. As is evident in the
figure, the measured discharge capacity is higher, despite the
lower intrinsic 4e- capacities, for the heavier alkali cation
permanganates compared to the lighter alkali permanganates.
The measured capacity of sodium, potassium, and cesium
permanganate cathodes is ∼0.45, 0.8, and 1.3 Wh. Note, the
sodium permanganate discharge required a higher fraction (32
wt %) of graphite to generate a discharge.
Figure 2. Measured and theoretical charge capacities of alkaline AAA
cells containing various fractions of Fe(VI) and Mn(VI) salts. The K2-
FeO4/KMnO4 composite cathodes contain a fixed 3.5 g of K2FeO4 and/
or KMnO4. Cathodes also contain 9 wt % graphite and 18 m KOH.
Note, all measured capacities above 1.25 Ah approach intrinsic
limitations of the ∼1.4 maximum Ah packed zinc anode.
volumetric capacity of 600 Wh/cc is high compared a maximum
400 Wh/cc for a high performance MnO2 cathode alkaline AAA
cell.
The permanganate to manganate reduction, or direct reduction
to MnO2, in aqueous alkaline media are
The cathode reduction is supported by a conductive matrix
provided through inclusion of graphite in the cathode mix. As
we have previously demonstrated, BaFeO4 with as low as
3-10% added carbon is capable of sustaining high current
densities with low polarization losses.6,7 In the presence of
BaFeO4, the enhanced Mn(VII) charge transfer indicated in
Figure 2, is attributed in part to the improved conductive matrix
that this Fe(VI) salt provides. Figure 3, probes the experimental
4e- (for permanganate) or 3e- (for manganate) efficiency,
determined by comparison of the measured cumulative discharge
ampere hours, as a fraction of the intrinsic charge determined
from the mass of the salt. In Figure 3, utilization of higher
weight fractions (cathodes employing 32 wt %, rather than 9
wt %) graphite greatly improves the percent storage capacity.
Hence, in each case reductive charge transfer appears to be
limited by an insufficient conductive matrix.
MnO4- + 1e- f MnO4
E ) 0.56 V vs SHE (4)
2-
MnO4 + 2H2O +3e- f MnO2 + 4OH-
-
E ) 0.58 V vs SHE (5)
The MnO2 product can undergo a further 1e- reduction, as
utilized in the conventional commercial alkaline cell:
2MnO2 + H2O +2e- f Mn2O3 + 2OH-
E ) 0.35 V vs SHE (6)
As seen in Figure 2, inclusion of even small amounts of the
Fe(VI) salts, such as 5% BaFeO4 or K2FeO4, dramatically
enhances charge transfer, yielding higher measured cathodic
capacities for KMnO4. At 75 Ω load in the AAA cell
configuration, KMnO4 or K2FeO4 alone discharge to 160 mA/
g, whereas a 40% KMnO4/60% K2FeO4 composite cathode
discharges to 300 mAh/g. In principle, permanganate can
undergo an eq 5 and 6 total of a 4e- alkaline cathodic reduction.
KMnO4 has a large theoretical discharge capacity, but exhibits
inefficient charge transfer, measured as a low experimental
capacity in the figure, which also includes the theoretical
(intrinsic) storage capacity of cells containing a variety of
relative compositions of BaFeO4 and KMnO4. The intrinsic
capacities are calculated from the mass of KMnO4 and BaFeO4
in the cell, assuming a 4 faradays mole-1 Mn(VI f III), and 3
F mol-1 Fe(VI f III), reduction, and subsequently converted
to ampere hours. Experimentally, Fe(VI) salts can attain the
complete 3e- faradaic capacity of eq 1, although K2FeO4 is less
charge efficient at higher current densities than BaFeO4.6-9 As
evident in the figure, a wide BaFeO4/KMnO4 composition range,
including cells varying from 20% to 80% BaFeO4 (and 20% to
80% KMnO4 ), each exhibits a high discharge capacity. Charge
capacity increases from 160 mAh/g with the pure KMnO4
In Figure 3, it is observed that the CsMnO4 cathode, with
inclusion of 32 wt % graphite, approaches 100% of the
theoretical 4 e- capacity. Three voltage plateaus are visible in
this CsMnO4 discharge, and are evidently attributable to
permanganate, manganate and MnO2 reduction steps occurring
at ∼1.67, 1.3, and 0.9 V respectively during 0-25, 25-75, and
75-95 of the percent storage capacity. Also in Figure 3 as
expected for the overall 3e- reduction in the barium manganate
discharge, when compared to a 4e- permanganate reduction the
lowest potential discharge step is accentuated (elongated), due
to the larger relative contribution of the final Mn(IV f III)
reduction described in eq 6.
The CsMnO4 cathode facilitates more efficient charge transfer,
compared to KMnO4, and has the additional previously dis-
cussed benefit of lower solubility than the potassium salt.
However, the intrinsic capacity of CsMnO4 is lower due to its
heavier formula weight. Instead in the super-iron alkaline cell,
cesium benefits may be incorporated, without the need to include
the heavier cesium permanganate salt, by direct replacement of
the KOH electrolyte with a CsOH electrolyte. In Figure 1, top,

11936 J. Phys. Chem. B, Vol. 105, No. 48, 2001
Letters
Figure 3. Salt cation and conductance (wt % graphite variation) effects on the discharge storage capacity of permanganate and manganate alkaline
AAA cells during discharge at low constant load rate (75 Ω). The percent storage capacity is determined by the measured cumulative ampere hours,
compared to the theoretical capacity. The 9 wt % graphite cells are detailed in the Figure 1 legend. In the 32 wt
% graphite cells, the mass of KMnO4 is 2.3 and that of CsMnO4 is 2.7 g. The potassium and barium cathodes contain 9 wt % 18 m KOH electrolyte,
and the cesium cathode contains 9 wt % 18 m CsOH.
this adds more efficient reduction toward the end of the
discharge, and further enhances total energy capacity.
an AAA cylindrical cell configuration, exhibit an unusually high
discharge capacity of 2.2 Wh. The discharge efficiency of solid
alkaline permanganate or manganate cathodes is probed, and
improved charge transfer of Mn(VII) redox chemistry occurs
in the presence of Fe(VI) is demonstrated. In a reciprocal
manner, Fe(VI) electrochemistry is enhanced upon inclusion of
manganese salts, and added cesium further improves the
combined Fe(V)/Mn(VII) cathode.
Figure 1, bottom, demonstrates that the additional perman-
ganate enhanced super-iron cell energy capacities realized at
low discharge rate in Figure 1, top, do not substantially impair
the high discharge rate performance of the cell. Under rapid
discharge conditions the capacity of the alkaline MnO2 cell falls
to a small fraction of its energy capacity in the low rate domain.
We had shown that the super-iron alkaline cell capacity
advantage over conventional alkaline cells is particularly large
in the high discharge rate domain, amounting to a severalfold
increase in capacity.6 In the lower section of Figure 1 these
high capacities are shown for both a constant load (2.8 Ω)
discharge and for a constant power (0.7 W) discharge, for the
cell containing a cathode composed primarily of BaFeO4. As
seen in the left-most curve, under these 2.8 Ω discharge
conditions, the KMnO4 leads to not only a lower capacity but
also due to the higher polarization, the lower cell potential
generates a lower power of only ∼0.4 W during the discharge.
Significantly, it is also seen that a cathode comprised of equal
KMnO4 and BaFeO4 weight fractions, retains the majority of
the capacity advantage, approaching 0.8 Wh, when also
discharged at the 2.8 Ω constant load (again, ∼0.6-0.7 W).
Acknowledgment. We thank B. Wang, and D. Rozen who
participated in the analyses.
References and Notes
(1) Licht, S.; Peramunage, D.; Myung, N. J. Phys. Chem. B 1998, 102,
6780.
(2) Licht, S. Electrochem. Commun. 1999, 1, 33.
(3) Rong, Z. et al. Science 1997, 276, 926.
(4) Thackeray, M. Prog. Solid State Chem. 1997, 25, 1.
(5) Chabre, Y.; Pannetier, J. Prog. Solid State Chem. 1995, 23, 1.
(6) Licht, S.; Wang, B.; Ghosh, S. Science 1999, 285, 1039.
(7) Licht, S.; et al. Electrochem. Commun. 1999, 1, 522; Electrochem.
Commun. 1999, 1, 527; Electrochem. Commun. 2000, 2, 535; Electrochem.
Commun. 2001, 3, 340.
(8) Licht, S.; et al. Electrochem. Solid State Lett. 2000, 3, 209;
Electrochim. Solid State Lett. 2001, 4, A209209; J. Electrochem. Soc. 2001,
148, A1072.
(9) Licht, S.; et al. J. Power Sources 2001, 99, 7; J. Power Sources
2001, 101/102, 167.
4. Conclusions
(10) Johnson; M. D.; Read, J. F. Inorg. Chem. 1996, 35, 6795.
(11) Komissarova, P. D.; Krestovnikova, S. A. Metody Poluch. Khim.
Reakt. Prep. 1967, 16, 72.
Cathodes formed with a cathode combining BaFeO4 and
KMnO4, in a cell using a conventional zinc alkaline anode in