Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02263662 1999-02-17
W O 98/10476 rCTfUS97115588
LITHIUM MANGANESE OXY-FLUORIDES FOR
LI-ION RECHARGEABLE BATTERY ELECTRODES
BACKGR~UND OF THE INVENTION
The present invention relates to lithium manganese oxide
intercalation compounds useful as active electrode materials in
Li-ion rechargeable batteries and, particularly, to oxy-
fluoride complexes of such compounds and their use to improve
the cycling stability and capacity of such batteries.
Litllium ma1.lcJanese oxide intercalatioll compounds,
nominally LiMn204, have been increasingly proven to be effective
and economical materials for the fabrication of secondary,
rechargeable Li-ion electrolytic cells and composite batteries.
Successful batteries of this type are described in U.S. Pat.
Nos. 5,296,31g and 5,~60,904. These batteries exhibit an
a~mirable level of electrical storage capacity and recharge
cycling stability over a wide range of voltages; however, these
properties have not been considered entirely satisfactory to
meet the increasingly stringent requirements of modern
electronic equipment and applications.
Extensive investigations have been undertaken to improve
the noted properties, and such works have resulted in
determinations that variations in the structural parameters of
the LiMn204 spinel, for example, the a-axis lattice dimension of
the compound, contribute significantly to ultimate cell
... . , ~
CA 02263662 1999-02-17
W O 98/1047~ rCTAUS97/15588
performance. Such structural parameters have in turn been found
to depend to a great extent upon the constitution of the
intercalation compound and upon the conditions of its
synthesis. In this respect, it has been generally agreed, for
instance, that an a-axis parameter of less than 8.23 A promotes
desirable recharging stability over extended cycles.
Approaches to achieve this advantageous parameter range
have included close control of synthesis conditions, such as
described by Tarascon in U.S. Pat. No. 5,425,932, to gain the
advantage of smaller a-axis dimensions exhibited by higher Mn
valence levels; and cationic substitutions, such as noted by
Tarascon et al., ~. Electrochem. Soc., Vol. 138, No. 10,
pp. 2~3S9-2864, Oc~:ober 1991, or by re~lacement of a portion of
tLle Mn atoms Wit].l Co, Cr, or Fe, such as suggested in Europeax
Patent 390,185. A number of other investigators have
recommended an increased level of lithium insertion to obtain a
similar efEect from a replacement of Mn, according to the
representative structural formul~, (Li)tet[Mh2-~Li]oct~4~ a~ an
effec~ ive mei:lll,5 of improving cycling stability, bu~ this
pl-aCt:iCe ha5 been found to result in a sacrifice of cell
capacity, as was observed with the earlier Mn replacement
approach.
In contrast to these previously implemented expedients,
the present invention utilizes anionic substitution to provide
a means for achieving concurrent improvements in both cycling
stability and cell capacity and enables the fabrication of
batteries capable of long-lasting and high-powered operation.
CA 02263662 1999-02-17
WO 98tlO476 PCT/U~97tlSS88
SU~A~Y QF THF INVENTION
We have discovered that -he inadequacies of prior
practices may be remedied by anlonic substi~ution of a portion
of the nominal LiMn2O4 oxygen a-oms with flu,orine. Although such
substitutions alone were initially observed to result in
expansion of the a-axis parameter beyond the preferred range,
apparently due to Mn valence reduction, we found, upon further
investigation, that a contemporary increase in Li substitution
for Mn surprisingly achieved a dramatic shift of a-axis
dimension into the optimum range below 8.23 A. Electrolytic
battery cel~s comprising these fluoro-substituted electrode
materials thexeafter exhibited remarkable cell capacity, as
lS well as cyclin~ s~abili~y.
Preparation of these advan~ageous oxy-fluoride spinel
derivatives may most simply follow the usual practice, such as
noted in Tarascon, U.S. 5,425,932, of anneallng at about 800~C
~oi.chiome~ric mixtures of appropriate precursor compounds,
t~f1i.cally 1,i.~CO3, I.iF, ancl MnO2. These derivatives may al50
include precursors for cationic substitutions as earlier-noted
in EP 390,185. The resulting intercalation materials that may
be ef~ectively employed to achieve an improvement in prior
electrolytic cells are therefore represented in the general
formula, Li1~X ~ n2 x yO4 zFz, where M is a metal, such as Co, Cr,
or Fe, and x S 0.4, y < 0.3, and 0.05 S z < 1Ø
Series of battery cell positive electrode compositions
prepared with the oxy-fluoride compounds varying primarily in x
and z formula components, i.e., Li and F, were examined by x-ray
CA 02263662 1999-02-17
WO98/10476 PCT~S~7/15S88
diffraction analysis to determine the resulting a-axis lattice
parameters and were then incorporated into test cells in the
usual-manner, as described in the above-mentioned patents. The
cells were subjected to repeated charge/discharge cycling to
determine the effect of compola~.d constitution on the level of
electrical storage capacity exnibited by the cells, generally
as mAhr/g of electrode compound, as well as on the cycling
stability, i.e., the ability to maintain the initial level of
capacity over extended cycling.
BRIEF DESCRIPT~ N OF THE DRAWING
The present invention will be described wi~h reference to
the accompanying drawing of which:
FIG. 1 is the x-ray diffraction pattern of an invention
compound, LiltxMyMn2 x yO~ zFz~ .~.here x = O.1, y = O, and z = O.l;
FIG. 2 is a graph of a-axis lattice dimensions v. z of
invention compounds, Lil+xMyMn2-x-yoa-zFz~ where x = O.05, y - O,
and z S 0.5;
FIG. 3 is a graphic comparison of capacity and cycling
stability v. number of charginy cycles for ba~tery cells
comprising positive electrode compounds of FIG. 2;
FIG. 4 is a graphic comparison of capacity and cycling
stability v. number of charging cycles for cells comprising
prior ~il+xMn20~ electrode compounds and a compound of the
present invention;
CA 02263662 1999-02-17
WO g8110476 rCT/USg7/15588
FIG. 5 is a graphic comparison of a-axis lattice
dimension v. z of invention compounds, Lil+xMyMn2 x yO~_zFz, where
x < 0.~, y = O, and z < 0.4;
FIG. 6 is a graphic comparison of capacity and cycling
stability v. number of charging cycles for cells comprising
invention compounds, Lil,xMyMn~.~y04 zFz, where x = O, y = O, and
z < 0.4;
FIG. 7 is a graphic comparison of capacity and cycling
stability v. number of charging cycles for cells comprising
invention compounds, Lil+xMyMn~ cyO4 zFz, where x = 0.1, y = O,
a~d z < 0.4;
FIG. 8 is a graphic comparison of capacity and cyclin~
stability v. number of charging cycles for cells comprising
invention compounds, Lil+xMyMn2 x yO4 zFz, where x = O.2, y = O,
and z S 0.4; and
F~G. 9 is a graphic comparison of capacity and cycling
stability v. number of charging cycles for cells comprising
invention compounds, Lil+,cMyMn2 x yO4 zFz~ where x - O, y = 0.2,
and z S 0.1.
DESCRIPTION OF THE INVENTION
Lil+xMn204 intercalation materials employed in prior
practices (according to present formula designation,
Lil+X ~ n2_x_yO4_zFz, where y = O and z = O) were prepared for use
CA 02263662 1999-02-17
W098tlO476 PCT~Sg7tl5588
as performance control samples in the manner described in t~e
aforementioned U.S. 5,425,932, ~Isins stoichiometric mixtures of
the primary precursor compounds, for example, 9.23 parts by
weight of Li2CO3 to 43.46 parts of MnO2 to obtain the nominal
LiMn204. Test cells of these control samples, as well as samples
of the present invention materials to be described later, were
likewise prepared and tested in galvanostatic and
potentiostatic studies, generally as described in that patent
specification. Such test cells comprised lithium foil negative
electrodes as a practical expedient, since experience has
confirmed that performance results achieved in this manner are
objectively comparable to those obtained with Li-ion cell
compositions described in the other above-noted patent
specifications. .~ clltional tests, as indicated below, were
nonetheless conducted with Li-ion compositions comprising the
present materials to obtain further confirmation of this
correlation in results.
Example 1
In a typical preparation of an intercalation material of
the present invention, stoichiometric proportions of the
precursors, MnO2 (EMD-type), Li2CO3, and LiF, were thoroughly
mixed in an agate mortar and pestle in a weight ratio of
60.94:12.82:1, and the mixture was annealed in air in an alumina
crucible in the manner of the control samples to obtain a test
composition of Lil+xMyMn2 x yO4 zFz, where x = 0.1, y = O, and
z = 0.1 ~Lil lMnl 903 gFo ~)- Specifically, the mixture was heated
at a regular rate over a period of about 12 hours to a
temperature of 800~C a~ which it was maintained for about 12
hours. The sample was then cooled to room temperature at a
CA 02263662 1999-02-17
W O 98/1047~ rcT~us~7/l5s88
regular rate over a period of about 24 hours. After a
mix/grinding, the sample was reheated over a period of 5 hours
to 80~~C where it was held for about 12 hours before being
finally cooled to room temperature over a period of about 24
hours. The resulting oxy-fluoride compound was characterized by
CuKa x-ray diffraction (XRD) examination to obtain the graphic
pattern shown in FIG. 1. The clearly-defined peaks of the
pattern confirmed the well-cr~stallized, single-phase product
of the synthesis.
Example 2
A series of oxy-fluoride compounds of the present
:invellt:i.orl was similarly prepared with appropriate combinations
of precursor compounds to yield Li1~xMyMn2 x yO4 zFz, where
x = 0.05, y = 0, and z = 0, 0.05, 0.10, 0.15, 0.20, 0.35, and
0.50. The resulting samples were characterized by XRD and the
respective a-axis lattice parameters were calculated. A plot of
these parameter dimensions as shown in FIG. 2 indicates the
reg3ular i.ncrease which trac]cs and is indicative of the increase
in fluorine substitution.
Portions of the same samples were individually
incorporated with about 10~ conductive carbon and 5~
polyvinylidene fluoride binder and formed as a layer on an
aluminum foil substrate to provide positive test cell
electrodes. Arranged in the usual manner with a lithium foil
electrode and intervening glass fiber separator saturated with
a 1 M electrolyte solution of LiPF6 in a 2:1 mixture of ethylene
carbonate:dimethylcarbonate, the sample electrodes formed test
cells which were subjected to charge/discharge cycling over the
CA 02263662 1999-02-17
WO98/1047G P~ bs~/15588
range o~ 3.4 - 4.5 V at a C/5 rate (full cycle over 5 hours).
The capacity of each cell was traced during a period of up to
35 cycles to provide an indic~tion, as seen in FIG. 3, of the
rate of change of that property, i.e., the c~cling stability of
the cell, with extended recharging. Traces,31-36 reflect the
above-stated increasing levels of fluorine substitution, z,
from 0.05 to 0.5. A comparison of the results depicted in FIG.s
2 and 3 graphically confirms the general tendency toward loss of
both capacity and cycling stability with an increase in a-axis
dimension above the preferred limit of about 8.23 A.
Exam~le 3
~ series o~ unsubstituted intercalation compounds of ~he
prior art varying only in Li, i.e., Lil~xM~Mn2 x_yO4_zFz, where
x = 0.05, 0.075, and 0.1, y - 0, and z = 0, was prepared and
tested in similar manner to provide an indication of the effect
of that variable on the capaclty and cycling stability of
resulting cells. ~s may be seen in FIG. 4 as traces 41-43 of
increasing Li con~ent, ~hat variance alone improves cycling
stability, but significantly reduces cell capacity. The
performance of an additional cell prepared with the oxy-
fluoride (x = 0.1, z = 0.1) compound of Example 1 is also
represented in FIG. 4, at trace 44, and reflects the surprising
e~fect achieved by the present invention. In particular, a
comparison of traces 43 and 44 having like Li content reveals
the outstanding improvement in both capacity and cycling
stability resulting from this combination with fluorine
substitution.
CA 02263662 1999-02-17
W098/104~6 PCT~S97115588
Example 4
Series of oxy-fluoride compounds were prepared varying in
both Li and F, i.e., Ll ~M~Mn c.~0~ 7Fz, where x = O, O.l, and
0.2, y = O, and z = O, 0.05, ~.l, 0.2, and G.4. The variations
of a-axis lattice parameter for each series are shown in FIG. 5
as traces 52-56 of increasing Li and indicate the remarkable
effect of the combination of Li and F content on achieving an
optimum range of this parameter.
Exam~le 5
The series of compounds of Example 4 comprising x = O was
used to pr.epar~.? batteL-y cells which were tested in the manner
described above. The results shown in FIG. 6 as traces 61-65 of
increasing fluorine content indicate the effect on capacity and
cycling stability of a compound favoring F in the Li:F ratio.
ExamPle 6
~,0
The series of compounds of Example 4 comprising x = O.l
was used to prepare battery cells which were tested in the
manner described above. The results shown in FIG. 7 as traces
71-75 of increasing fluorine content indicate the improvement
on capacity and cycling stability of a closer balance of F in
the Li:F ratio.
Example 7
The series of compounds of Example 4 comprising x = 0.2
was used to prepare battery cells which were tested in the
manner described above. The results shown in FIG. 8 as traces
CA 02263662 1999-02-17
WO98/10476 PCT~S~7115588
81-85 of increasing fluorine content indicate the further
effect, particularly on cycling stability of a still closer
balance of F in the Li:F ratio.
Exam~le 9
A series of compounds of the present invention with both
cationic (Cr) and anionic substitutions, Lil+x~Mn2_x_yO4_z~z,
where x = 0, y = 0.2, and z = 0, 0.05, and O.l, was prepared in
the above manner by combining appropriate stoichiometric
amounts of precursors, for example, lO.3:2.31:l.0:0.086 weight
ratio of MnO2, Li2Co3, Cr2O3, and LiF (LiCrO 2Mnl 8~3 gFo 05). The
resulting materials were used to prepare test cells whose
performance improvemellt was comparable to the foregoing
results, as shown at FIG. 9 in traces 92-96 of increasing
fluorine content. Similar results may be obtained with cationic
Co and Fe substitutions.
Exam~le lO
A series of Li-ion battery cells was prepared with the
positive electrode materials of Example 6, and employing
petroleum coke negative electrodes and polyvinylidene copolymer
matrix electrolyte/separator elements, as described in above-
noted ~.S. 5,460,904. Tests of repeated charge cycling showed
cell capacities and cycling stability comparable to those of
Example 6.
It is expected that other embodiments of the present
invention will become apparent to the skilled artisan in light
of the foregoing description, and such variations are intended
to be included within the scope of this invention as recited in
the appended claims.
- 10 -
.... . . . . . . ...
