| Full Text |
Description
STOICHIOMETRIC LITHIUM COBALT OXIDE AND METHOD
FOR PREPARATION OF THE SAME
Technical Field
[1 ] The present invention relates to a lithium cobalt oxide having a stoichiometric
composition, which can be used as a cathode active material for lithium rechargeable batteries.
Background Art
[2] A report in 1980 that LiCoO 2is useful for a cathode active material of lithium
rechargeable batteries was followed by a lot of research, so LiCoO2 was adopted by commercial enterprises as a cathode active material for lithium rechargeable batteries. But, the high cost of LiCoO contributes significantly (about 25%) to the cost of the battery product. High competition presses producers of rechargeable lithium batteries to lower the cost.
[3] The high cost of LiCoO2 is caused by two reasons: First, the high raw material cost
of cobalt, and second the high cost of establishing reliable quality management and ensuring perfect process control during large scale production.
[4] Especially, the quality management and process control aim to achieve highly re-
producible products having optimized properties, where the performance of every batch fluctuates very little from those optimum properties. High reproducibility and little fluctuations of the performance of LiCoO2 are absolutely essential in current highly-automated high volume lithium battery production lines.
[5] A major problem is that LiCoO2 is a sensitive material. Small changes of
production process parameters cause large fluctuations of the performance of the cathode product. So quality management and process control require much effort and high costs.
[6] LiCoO2 is a stoichiometric phase. Under normal conditions (for example 800°C in
air) no reliable indication for any I ,i:Co non sloichiometry has been reported in the literature.
[7] Only stoichiometric LiCoO2 with a l:Ca ratio very near to 1:1 has properties
which are suitable for the cathode active mnlerial of the commercial lithium batteries. If the Li content is higher than 1:1, LiCoO2 will coexist with a secondary phase which contains the excess lithium and largely consists of Li2 CO3 . Li2CO3 impurities in the commercial LiCoO2 cathode active material are highly undesirable. Such samples are
known to show poor storage properties at elevated temperature and voltage. One typical test to measure the storage properties is storage of fully charged batteries at
90°C for 5 hours.
f 8) If the cathode contains Li2CO3 impurities, this may result in strong swelling
(increase of thickness) of polymer cells. Even the much stronger metal cases of prismatic cells may bulge.
[9] If the Li2CO3ntent is lower than 1:1, then the cathode contains divalent cobalt, i.e.
LiCoO2 coexists with cobalt oxides. Lithium-deficient LiCoO2 shows poor cycling stability at a high voltage (>4.3 V), especially at an elevated temperature It is speculated that the higher catalytic activity of divalent cobalt present in the cobalt oxide phase supports the undesired oxidation of an electrolyte on the surface of LiCoO . Alternatively, divalent cobalt might, especially at a high voltage, dissolve in the electrolyte, and undergo precipitin ion at the anode side, thereby damaging a solid electrolyte interphase (SEI) layer on the anode.
[ 10] Only in a lab, it is easy to prepare stoichiometric LiCoO2 practically free of Li2CO3
or CoOx impurities by simple heating of LiCoO 2. The high cycling stability of such cathodes (in coin cells) has been demonstrated in the literature. It is speculated that the good cycling stability is attributed to two effects: (1) At small scale (lab size samples) the excess lithium (Li2CO3 ) easily evaporates during sintering, and (2) Heating repairs any damage to the surface of LiCoO2 , which was caused by air exposure, probably by a reductive attack by hydrocarbons.
[11] A similar re-heating of LiCoO 2is not effective to solve the problems associated
with the high temperature properties and cycling stability which may occur in the large scale production. First, large scale-produced LiCoO2 has not a damaged surface. After production the product is usually filled into air tight packaging, so any damage caused by air exposure is practically absent. Second, on a large scale, excess lithium does not evaporate practically. Very small amounts of Li2CO3 can be decomposed because volatile phases exist with very small thermodynamic equilibrium partial pressure. At small partial pressure gas transport is very slow, so that only tiny amounts of Li2CO3 can be decomposed. If we deal with large samples then the gas transport is not sufficient to decompose significant amounts of Li2CO3 .
[12] The situation is different if Li2CO3 decomposes in the presence of a lithium
acceptor (such as cobalt oxide). In this case the thermodynamic equilibrium partial pressure is high and the gas transport kinetics is fast enough to decompose Li2CO3 .
[13] More generally, it is very difficult or even impossible to prepare LiCoO2 with the
exact desired Li:Co ratio at large scale. If an excess of cobalt is used, then a cobalt oxide impurity remains. Unfortunately, small impurities of CoOx are practically impossible to be detected by standard quality control methods, but they are very important for the performance of the cathode. If an excess of lithium is used, lithium impurities remain due to the low evaporation at large scale. Even if the premixed (Li
CO3 and Co-oxide) powder would exactly have the desired Li:Co ratio, any inho-mogeneity in the mixed powder would after sintering creates a powder with regions being lithium-rich and other regions being lithium-deficient. Additionally some Li2CO3 can melt before fully reacting with the Co-oxide, and the molten Li2CO3 would tend to
separate downwards. This will conecting II:Co gradient with Li-deficient sample at the top and Li-excess at the bottom of the shteing vessel. As a result, very small amounts of impurity phases (Li2CO3 or Co oxide) ate present.
[14] Much previous art to improve properties of LiCoO2 has been disclosed. Examples
of such efforts are surface coating of LiCoO2 , doping of LiCoO2 with other metal cations and the preparation of non-stoichiometric LiCoO2 at a very high temperature. Each effort created some satisfactory results, but the results are not enough for a mass production process, and make another problem of the costs of additional processes. Disclosure of Invention Technical Problem
[ 15] The invention discloses that a property fluctuation of LiCoO2 in a mass-production
process is primarily caused by the difference of lithium chemical potential of LiCoO2 , and robust LiCoO2 less sensitive to process parameters can be prepared by co-firing LiCoO2 and a solid state lithium buffer material to adjust a stoichiometric composition of lithium and cobalt to a desired range.
[ 16] Conventional LiCoO2 that is mass-produced has a problem in that little deviation of
the composition from the desired stoichiomelry results in significant fluctuation of a lithium chemical potential. If the lithium chemical potential is fixed within a given range, it is possible to achieve excellent high-voltage cycling properties and storage properties of LiCoO2 even under high temperature conditions on the mass production scale.
[17] Therefore, the present invention has been made in view of the above problems, and
it is an object of the present invention to provide a powder comprising LiCoO2 having a stoichiometric composition via heat treatment of a lithium cobalt oxide and a lithium buffer material to make equilibrium of the lithium chemical potential therebetween, a method for preparing the same, the above-mentioned lithium buffer material, an electrode comprising the above-mentioned LiCoO2 -containing powder as an active material, and a rechargeable battery comprising the same electrode.
Technical Solution
[ 18] In accordance with an aspect of the present invention, the above and other objects
can be accomplished by the provision of a powder comprising LiCoO2 having a stoichiometric composition via heat treatment of a lithium cobalt oxide and a lithium buffer material to make equilibrium of a lithium chemical potential therebetween.
[ 19] In accordance with another aspect of the present invention, there is provided a
LiCoO2-containing powder having a constant lithium chemical potential, wherein LiCoO2coexists with a lithium buffer material under equilibrium conditions of the lithium chemical potential therebetween, and the lithium chemical potential of powder is higher than the equilibrium lithium chemical potential between LiCoO2 and a cobalt oxide and is lower than the equilibrium lithium chemical potential between LiCoO2 and Li2CO3 .
[20] In accordance with a further aspect of the present invention, there is provided a
lithium buffer material which acts as a Li acceptor or Li donor to thereby remove Li-excess or supplement Li-deficiency, which coexists with a stoichiometric lithium metal oxide.
[21] In accordance with a still further aspect of the present invention, there is provided a
method for preparing a LiCoO2 -containing powder, comprising a step of providing a homogeneous mixture of LiCoO2 and a lithium buffer material; and a step of heating the resulting mixture to make an equilibrium of a lithium chemical potential.
[22J In accordance with yet another aspect of the present invention, there is provided an
electrode comprising the above-mentioned LiCoO2 -containing powder as an active material, and a rechargeable baltery comprising the same electrode. Brief Description of the Drawings
[23] The above and other objects, leatures and other advantages of the present invention
will be more clearly understood from the following detailed description taken in conjunction with the accompanying drawings, in which:
[24] FIG. 1 is a thermodynamic schematic diagram of lithium chemical potential of
LiCoO2 ;
[25] FIG. 2 is a thermodynamic schematic diagram of lithium chemical potential of Li
buffer (LiMO2);
[26] FIG. 3 is a thermodynamic schematic diagram which shows an equilibrium state of
lithium chemical potential between Li-exccss LiCoO2 and Li buffer (LiMO2);
[27] FIG. 4 is a thermodynamic schematic diagram which shows an equilibrium state of
lithium chemical potential between Li-deficient LiCoO2 and Li buffer (LiMO );
[28] FIG. 5 is a graph showing cycling stability of LiCoO2 at a different Li:Co ratio in
Reference Example 1;
[29] FIG. 6 is a crystallographic map of solid state lithium buffer Li(Mn1/3 Ni1/3 Co1/3 )O2
in Reference Example 4;
[30] FIG. 7 is a graph showing cycling stability of solid state lithium buffer Li(Mn1/3 Ni1/3
Co )O2in Reference Example 5;
[31 j FIG. 8 is an FESEM (Field Emission Scanning Electron Microscope) image which
shows (a) LiCoO2 precursor powder, and (b) TR01 sample obtained by co-firing of 90% LiCoO2 and 10% Li(Mn1/3 Ni1/3 Co )O2 in Example 1-
[32] FIG. 9 is a graph showing cycling stability of a coin cell in Experimental Example
2, comprising TR01 sample as a cathode active material;
[33] FIG. 10 is a graph showing cycling stability of a coin cell in Comparative Example
1, comprising inexpensive LiCoO2 as a cathode active material;
[34] FIG. 11 is a graph showing cycling stability of a polymer cell in Experimental
Example 3, comprising a cathode active material of TR01 (23°C, 45°C, 1C rate (discharge)-0.6C rate (charge), 3.0 V to 4.2 V, 3.0 V to 4.3 V, 3.0 V to 4.35 V, 3.0 V to 4.4 V, 400 cycling);
[35] FIG. 12 is an FESEM image which shows a precursor sample (Precursor 1) and the
resulting electrode active material (4 kg-Final) in Example 2; and
[36] FIG. 13 a graph showing cycling stability of electrode active material in Example
2, wherein the core of LiCoO2 is fully covered by the shell of solid state lithium buffer.
Mode for the Invention
[37] Hereinafter, the present invention will be described in more detail.
[38]
[39]
[401 Generally, the chemical potential can be defined as the changes in a characteristic
thermodynamic state functions (internal energy, enthalpy, Gibbs free energy, and Helmholtz free energy) with respect to the change in the number of moles of a particular constituent, depending on the experimental conditions. The chemical potential can be expressed by the following equation, under the conditions of constant temperature and constant pressure:
[41] (Formula Removed)
[42] wherein µ is a chemical potential, G is Gibbs free energy, N is the number of
molecules, T is a temperature and p is pressure.
[43] Therefore, the chemical potesium IN regarded as the energy state of each material in
the specific system. If two different muterium with different chemical potential coexist in the same system, the reaction Inkes place in order to lower the sum of potentials, and the two materials equilibrate to the same potential.
[44] In the present invention, "i" is lithium, "j" is oxygen, but by a special ther-
modynamic operation called Lagrange transformation "j" is replaced by p(j) which is the oxygen partial pressure. Other "j" (Mn, Co, Ni.) are "frozen", definition of which will be made hereinafter.
[45] The thermodynamic equilibrium state of two or more different stoichiometric
compounds requires the same chemical potential therebetween. If LiCoO2 coexists with impurity materials (Li2CO3 or CoO ), it can be considered that there is the state of
chemical potential equilibrium between LiCoO2 and individual impurities. As shown in FIG. 1, Li2CO3 has a higher lithium chemical potential than LiCoO2 and CoO has a
lower chemical potential than LiCoO2 . As a result, if the composition of LiCoO2
deviates from the stoichiometry (1:1), each equilibrium potential profile exhibits a
stepwise gradient, not a continuous form.
[46] Generally, mass-produced LiCoO2 is somewhat lithium-excess or lithium-deficient,
as discussed hereinbefore. The lithium chemical potential of such a product has always a higher value equilibrated with Li2CO3 , or a lower value equilibrated with CoO x.
Therefore, the mass-produced LiCoO2 is difficult to have a proper value between the upper and the lower chemical potentials (the potential of Li:Co =1:1 composition).
[47] LiCoO2 free of bulk impurity phases (Li2CO3 or CoO x) still has surface defects.
This is because defects diffuse out of the crystallites and accumulate at the surface. As a result, the surface is lithium-deficient, then the lithium chemical potential is low. Alternatively, the surface can be lithium-rich, then the lithium chemical potential is high.
[48] The present invention provides a LiCoO2 -containing powder having a constant
lithium chemical potential, via the heat treatment of LiCoO2 and a material functioning as a Li acceptor and/or a Li donor to bring about equilibrium of the lithium chemical potential therebetween, whereby the lithium chemical potential of LiCoO2 is higher than the equilibrium lithium chemical potential between LiCoO2 and a cobalt oxide and is lower than the equilibrium lithium chemical potential between LiCoO2 and Li2CO3 .
[49] That is, the phrase "preferred range of a lithium chemical potential of LiCoO2 " as
used herein refers to a chemical potential which is higher than the equilibrium potential between LiCoO2 and a cobalt oxide and is lower than the equilibrium potential between LiCoO2 and Li2CO3 .
[50] If LiCoO2 having a fixed chemical potential in the preferred range is used as a
cathode active material of lithium rechargeable batteries, it is advantageously possible to achieve excellent cycling stability at a high voltage.
[51] The lithium chemical potential cannot be easily measured. It is not directly related
to the open circuit voltage (OCV) of an electrical cell at ambient temperature. The
OCV is the lithium potential in a "frozen" cobalt-oxygen lattice framework. As used
herein, the term "frozen" refers to a temperature which is low enough to prevent a ther
modynamic equilibration within a limited time. The entropy of a crystalline substance
is zero at the absolute zero of temperature (0 K), thereby exhibiting completely
different thermodynamic behavior.
[52] [Reaction Scheme 1]
[53]
LiCoO2 - x(Li+ + e-)↔ Li1_xCoO2
[54] Contrary, the lithium chemical potential at room temperature is dominated by the
transition metal composition and the lithium stoichiometry, and furthermore it is related to the conditions during population
[55] As discussed before, the perlomance properties of commercial LiCoO2 as the
electrode active material depend very sensitively on the exact Li:Co ratio. Upon slight deviation from the exact Li:Co ratio, the strong change in properties of LiCoO is caused by the step-like change of lithium chemical potential. Accordingly surface properties, which dominate the storage and high-voltage cycling properties, change stepwise as well.
[56] Obviously, it would be preferable to eliminate the step-like change of the lithium
potential, and to fix the potential within a preferred region. Then a small deviation of lithium stoichiometry away from the optimum stoichiometric value would only cause a small change of the lithium chemical potential, as a result surface properties would be only scatter slightly from the optimum and generally, a more robust cathode material less sensitive to changes of the composition is achieved. Such a robust cathode material can then be prepared at high quality and low cost with less requirements regarding perfect process control and quality management as will be disclosed in the following.
[57]
[58]
[59] FIG. 2 shows an illustrative example explaining some basic thermodynamic
properties of a solid state lithium buffer such as LiMn1/3Ni1/3 CoO2 .
[60] Being a buffer means that it can act as a lithium acceptor and/or a lithium donor,
and that the lithium chemical potential varies little with lithium stoichiometry, and that the absolute value of a slope of a chemical potential versus lithium stoichiometry is much smaller than that of lithium metal oxide which would be removed or supplemented with Li.
[61] If the Li:M ratio in LiMO2 as the solid state lithium buffer is 1:1, basically Mn is
tetravalent, Co is trivalent and Ni is divalent. If the sample is lithium-rich, Li1+x M 1-x O2 ,
a portion of Ni has a trivalent state. If the sample is lithium-deficient, Li1-x M1+x O2 , a
portion of Mn changes from 4 to 3 valent state, or a portion of Co changes from 3 to 2 valent state. The changes of valence state occur within the same crystallographic layer structure and enable achievement of a wide stoichiometric range. Because of the wide stoichiometric range, the changes of lithium chemical potential within a limited region
are not significant.
162] A mixture of LiCoO2 with a solid state lithium buffer such as LiMO (for example,
M=Mn1/3Ni1/3Co1/3) does not react (i.e-it does not equilibrate) at ambient temperature. Therefore, all LiCoO2 particles in the mixture maintain a low or high lithium chemical potential, and the buffer particles have a potential determined by the lithium stoi-chiometry of the buffer.
[63] During heating of the mixture a reaction takes place as follows. At a medium
temperature, possibly above 200°C, the lithium becomes very mobile. This is not sufficient for the equilibration of the lithium chemical potential because the buffer reaction involves a change of transition metal valence state. This is also accompanied by an uptake or release of gas phase oxygen. At a higher temperature, possibly above 400°C, the oxygen becomes mobile, but the transition metal cations are still frozen. Now the lithium chemical potential and the oxygen potential equilibrate. At that temperature, the buffer can consume lithium by decomposition of Li2CO3 impurities, or it can release lithium to lithiate the cobalt oxide impurities. Finally, the lithium chemical potential of LiCoO2 equilibrates at the buffer potential.
[64] At much higher temperatures (>>10000C), the transition metal cations become
mobile, and can therefore react with LiCoO2 10 form a new material.
[65] The present invention disclomenl ,LiCoO2 which is co-fired with a lithium buffer
serving as the Li acceptor or Li douor, al a temperature above 400°C, i.e. at a temperature which is high enougli In achieve an equilibration of lithium and oxygen. Otherwise the temperature is low enough (below 1000°C) so that the transition metal has not fully equilibrated. As a result, stoichiometric LiCoO2 , free of Li2CO3 or cobalt oxide impurities and coexisting with the lithium buffer, is achieved. The lithium chemical potential of LiCoO2 is fixed at the lithium buffer potential. The buffer is chosen so that the lithium potential of LiCoO2 is fixed within a preferred region.
[66] FIG. 3 is a schematic diagram which shows a thermodynamic equilibrium state of a
mixture of LiCoO2 (having a small lithium excess) with a lithium buffer LiMn1/3Ni1/3 Co2
O2 . Initially, the lithium potential is different in LiCoO2 and buffer. During the equi
libration process, the buffer consumes lithium (by decomposition of the Li2CO3
impurities) until stoichiometric LiCoO2 is achieved, and the lithium chemical potential
is equilibrated and fixed within the preferred region.
[67] FIG. 4 is a schematic diagram which shows a thermodynamic equilibrium state of a
mixture of lithium-deficient LiCoO2 with a lithium buffer LiMn1/3 Ni1/3 Co1/3O2 . Initially,
the lithium potential is different in LiCoO2 and buffer. During equilibration, the buffer delivers lithium (to lithiate the cobalt oxide) until stoichiometric LiCoO2 is achieved, and the lithium chemical potential is equilibrated and fixed within the preferred region. [68]
[69]
[70] Lithium buffer materials may serve as lithium acceptors and/or lithium donors.
Lithium donating properties are required to lithiate CoOx impurities to form LiCoO2 .
Lithium donators are known in the art. Non-limiting examples of the lithium donors are lithium-containing oxides such as Li MnO . Otherwise lithium accepting properties are required to decompose excess Li2CO3 impurities. Mild lithium acceptors, strong enough to decompose Li2CO3 but not strong enough to delithiate LiCoO2 , are also
known in the art. Non-limiting examples of the lithium acceptors may include TiO (reacting to form Li TiO ), ZrO (→Li2 ZrO3 ), Al2 O3 (→LiA1O2 ), MnO2 (→Li2 MnO3 ),
LiMn 0 (→Li2 MnO3 ), etc. These compounds can be generalized as oxides which are able to form double oxides with Li O2.
[71] Other examples of lithium-accepting compounds donate anions, preferably fluorine
or phosphate ions, which trap excess lithium thereby forming stable lithium salts. Non-limiting examples of such compounds are MgF (→2LiF + MgO), Li3 AIF6 , A1PO3 (→A12 O 3+ Li PO4 ) and transition metal-based phosphates (such as Co (PO ) and
LiCoPO4 ) etc. Such lithium acceptors may be effective to decompose Li2CO3
impurities, but they cannot lithiate CoOx impurities. Additionally, these compounds are
electrochemically inert, that means they do not contribute to the reversible capacity.
Only small amounts of inert compounds, typically below 1% by weight, should be
added, otherwise the specific reversible capacity of the final cathode will be too low.
[72] The solid state lithium buffer of the present invention preferably has a lithium-
accepting ability and a lithium-donating ability at the same time. Preferably, the solid state lithium buffer of the present invention also has a high reversible capacity. Preferred examples of the lithium buffers according to the present invention may be lithium transition metal oxides of Formula Liz MO2 (0.95
0
[73] Specifically, for example, the lilhium transition metal composite oxides comprising
nickel, manganese and cobalt, such as LiMn1/3Ni1/3 Co1/302 and LiNi0.55 Mn0.3Co0.1502,
are preferred because they act as lithium donators as well as lithium acceptors, and ad
ditionally, because they can be added in larger quantities (>1% by weight, typically
10% by weight or more) without causing deterioration of the reversible capacity of the
final cathode material.
[74] In the present invention, an amount of the solid state lithium buffer is preferably
less than 30% by weight of the total weight of the powder. That is, the preferred content of LiCoO2 in the electrode active material is more than 70% by weight.
2
[75] The minimum amount of the lithium buffer is at least 1 mol%, more typically at
least 10 mol%. For example, if LiCoO2 has 0.25 mol% impurity of Li2CO3 , then adding about 1 mol% of LiMO , followed by co-sintering yields impurity-free LiCoO2
coexisting with lithium-rich Li1+xM1+xO2(approximate composition Li1.2 M0.8O2
[76]Alternatively, a proper amount of lithium-deficient Li1-xM1+xay be added as the
solid state lithium buffer. Alternatively, adding 10 mol% of LiMO2to LiCoO2 with
0.25 mol% impurity of Li2CO3would result in LiCoO2 coexisting with slightly lithium-rich Li1+xM1-xO2(approximate composition Li1.025M0.975O2).
[77] It is recommended to add a sufficient amount of the solid state lithium buffer (or to
adjust the initial lithium stoichiometry) to achieve a preferred lithium stoichiometry of the buffer after co-sintering. If the lithium content of the lithium buffer in the final product is too low or too high, then generally the electrochemical properties of the buffer (for example, reversible capacity) are less.
[78]For example, if LiMO2with M=Mn1/3Ni1/3Co1/3orM=Ni0.55Mn0.3Co0.15 is used as the
lithium buffer material, the lithium stoichiometry of the buffer after the heat treatment should be within a desired range otherwise the electrochemical properties deteriorate. Slightly lithium-rich Li1+x M1-xO2 (x
slightly lithium-deficient Li1-xM1+xO2 (x
[79] It is preferred to ensure that the lithium chemical potential of the solid state lithium
buffer matches the lithium chemical potential of LiCoO2 and is in the preferred range which is higher than the equilibrium potential between LiCoO2 and a cobalt oxide and is lower than the equilibrium potential between LiCoO2 and Li2CO3 . As an example: The preferred Li:M ratio for LiMO with a high content of nickel (>80%) is 1:1. However, at this composition the lithium chemical potential is too high. Otherwise, the lithium chemical potential of lithium manganese spinel is lower than the equilibrium potential between LiCoO2 and the cobalt oxide corresponding to a lower limit of the above-mentioned optimal range. Therefore, spinel is a too strong lithium acceptor which will decompose LiCoO2 .
[80] The electrode active material in the present invention is not limited to a specific
form as long as LiCoO2 contacts with the lithium buffer (material serving as the
lithium acceptor and/or the lithium donor). In the simplest case, the electrode active
material is in the powder form. Typically, the LiCoO2 powder and the solid lithium
buffer powder are mixed, followed by heat treatment (co-firing).
[81] In one preferred embodiment, the heat-treated mixture is a co-fired mixture of an
oxide powder (a) of LiCoO2 and a lithium transition metal oxide powder (b) of Formula Liz MO2 (0.95
Ni (x/(l-x-y)) is in a range of 0.4 to 1.1). Herein, the oxide powder (a) is monolithic particles having D50 of more than 10 D, and the oxide powder (b) is agglomerated
particles having D50 of less than 10 D.
[82] Generally, where a particle size of the electrode active material is larger, this may
lead to a decrease in a surface area for read ion with an electrolyte inside a battery cell, thereby causing significant deterioration of high-voltage storage properties and rate properties and consequently decreasing a particle diameter of the active material. On the other hand, the electrode active meterial with a large particle diameter exhibits relatively high structural stability Including high-temperature properties and decrease of side reactions including electholyte decomposition, as compared to the active material having a small particle diameter.
[83] However, it was surprisingly confirmed that the co-fired mixture according to the
present invention maintains a desired level of excellent high-voltage storage properties, even though the oxide powder (a) has a large particle diameter of more than 10 D. As described before, this is because LiCoO2 having a stoichiometric composition or a mixture having a proper lithium chemical potential can be obtained, due to buffering effects of the Li MO powder (b) as the lithium buffer material, on the LiCoO2 powder (a), during heat treatment.
[84] Alternatively, the solid state lithium buffer can be added in powder form, dispersed
in a solution, or it can be dissolved in the solution.
[85] Instead of a simple mixing of powders, more complex preparation routes might
allow not only to improve cycling and storage properties, but also achieve improvement of the safety. Generally, it is known that LiCoO2 has poor safety and Mn-containing LiMO has better safety. Therefore, if a portion of the buffer covers the LiCoO2 surface, the safety could be improved.
[86] For example, fine particles of LiMO may be coated on the surface of the LiCoO2
particles. A typical method would involve spray-coating or dry-coating of small, preferably monolithic LiMO particles (1-3 0) onto larger LiCoO2 particles (5-20 D). During sintering, not only the Li2CO3 and CoO impurities are consumed, and the lithium chemical potential is fixed within the preferred region, but also the small LiMO particles become strongly attached to the LiCoO2 surface and effectively cover a large fraction of the LiCoO2 surface.
[87] Alternatively, a layer of transition metal hydroxide or transition metal carbonate
can be coated onto the LiCoO2 by precipitation, using the LiCoO2 particles as seeds. After eventually adding further lithium and sintering, not only the impurities are consumed, and the lithium chemical potential is fixed within the preferred region, but additionally a dense thick layer of electrochemically active LiMO effectively covers the LiCoO2 surface.
[88]
[89]
[90] After addition of the lithium buffer to LiCoO2 , a heat treatment follows. In some
cases, additional additives may be added before the heat treatment. The additives may be sources of additional lithium (such as Li2CO3 and LiOHH2 O), or the additives may extract lithium and supply fluorine (such as MgF and Li A1F ), or the additives may be materials suitable to modify the surface of the particles (for example, sintering agents)
[91] The heat treatment typically is made in air, or alternatively it can be carried out in
controlled, oxygen-containing gases with a poor oxygen content or in mixed gas of oxygen and nitrogen.
[92] The heat treatment is carried out at a proper temperature. A suitable temperature
range for the heat treatment is 400 to 1100°C, more preferably 500 to 950°C. A temperature of less than 400°C might be too low. At such a low temperature, the equilibration of lithium potential between the lithium buffer and the LiCoO2 may require an unreasonable long time. If the temperature exceeds 500°C, the equilibration of lithium potential between LiCoO2 and lithium buffer and also the necessary equilibration of the oxygen potential usually occur at reasonable kinetics. If the heat treatment is carried out at a very high temperature exceeding 1 100°C, this may undesirably result in significantly increased process costs. That is, the heat treatment at the very high temperature suffers from high cosin for lnstullation of high temperature equipment, large consumption of energy, and a need for additional processing steps such as grinding and sieving of sintered pales, Additionally, at such temperatures not only the lithium and oxygen potentials equilibrate, but also the transition metal diffusion becomes significant, thus resulting in a doped LiCoO2 .
[93] If the lithium buffer is LiMn1/3Ni1/3Co2O for example, dominantly cobalt from the
LiCoO2 would diffuse into the LiMO2 , and less Mn-Ni from the LiMO2 would diffuse
into the LiCoO2 , resulting in a phase mixture of two LiCo1-x(Mn-Ni)xO2phases, one with x>0.333 and the other with x=0.
[94]
[95] Preparation of electrode and rechargeable batteries>
[96] An electrode comprising the LiCoO2 material according to the present invention as
an electrode active material can be prepared by a conventional method known in the art. For example, the electrode in the present invention may use an electrical conducting material for providing electric conductivity, and a binder for adhesion between the electrode material and a current collector, in addition to such an active material.
[97] A paste is prepared by mixing the above-prepared electrode active material, 1 to 30
wt% of the electrical conducting material and 1 to 10 wt% of the binder in a dispersion solvent, followed by stirring. A laminated electrode structure is prepared by applying
the resulting electrode paste to a metal sheet current collector, and pressing and drying the resulting structure.
[98] A general example of the electrical conducting material is carbon black. The
products sold in market may include various acetylene black series (available from Chevron Chemical Company and Gulf Oil Company), Ketjen Black EC series (available from Armak Company), Vulcan XC-72 (available from Cabot Company) and Super P (available from MMM Company).
[99] Representative examples of the binders may include polytetrafluorethylene
(PTFE), poly vinylidene fluoride (PVdF) or a copolymer thereof, and cellulose.
[ 100]Representative examples of the dispersion solvents may include isopropyl alcohol,
N-methylpyrolidone (NMP) and acetone.
[1011 The metal sheet for the current collector should be a highly electrical conductive
metal to which the paste can be easily attached. Further the metal sheet should be non-reactive in the range of a working voltage of cells. If that condition is satisfied, any metal sheet can be used. Representative examples of the metal sheets may be mesh or foil of aluminum or stainless steel.
[ 102] The present invention provides a rechargeable battery comprising the electrode of
the present invention. The rechargeable battery of the present invention can be prepared by a conventional method known in the art, which is not particularly limited. For example, the battery can be fabricated by interposing a separator between the cathode and the anode and introducing a non-aqueous electrolyte into the resulting electrode assembly. The electrode, separator, electrolyte, and optionally additives known in the art can be used.
[ 103] A porous separator can be used as a separator upon fabrication of the battery.
Specific examples of the porous separator may include, but are not limited to, polypropylene series, polyethylene series, and polyolefin series.
[ 104] The non-aqueous electrolyte for the rechargeable battery of the present invention
contains a cyclic carbonate and/or a linear carbonate. Examples of the cyclic carbonate may include ethylene carbonate (EC), propylene carbonate (PC), and gamma buty-rolactone (GBL). Examples of the linear carbonate may include diethyl carbonate (DEC), dimethyl carbonate (DMC), elhylniethyl carbonate (EMC), and methylpropyl carbonate (MPC). The non-aqueons electeolite for the rechargeable battery of the present invention contains a lithium net In onjunction with the carbonate compound. Specific examples of the lithium sell may include LiCIO4 , LiCF3 SO3 , LiPF6 , LiBF4 ,
'LiAsF6 and LiN(CF3 SO2 )2.
[105]
[106] EXAMPLES
[107] Now, the present invention will be described in more detail with reference to the
following examples. These examples are provided only for illustrating the present invention and should not be construed as limiting the scope and spirit of the present invention.
[108]
[109] [Reference Example 1] Preparation and characterization of LiCoO2 with Li-excess
or Li-deficiency
[110] This reference example is intended to demonstrate that an electrochemical property
of LiCoO2, particularly the cycling stability at a high voltage, sensitively depends on the Li:Co ratio.
[ 111] Commercially available LiCoO2 was purchased from a low-cost supplier and used
as a precursor material for this experiment. Three LiCoO2 samples (Li', LiO and Li+) were prepared from the precursor. Samples Li" and Li+ were in an amount of about 1 kg, and Sample LiO was approx. LOO g.
[112] The lithium-deficient sample Li" was prepared by immersing 1 kg of LiCoO2 into
water, resulting in a total of 2L slurry. 7.6 g of concentrated Li SO was added with stirring the slurry. Three different reactions took place after addition of acid: (1) decomposition of Li2CO3 impurities, (2) dissolution of divalent cobalt, and (3) ion exchange (ion leaching) of Li ions with H+ ions on the surface region of LiCoO2 particles. Initially the pH dropped to about 2, but slowly increased to about 6 after 30 minutes. The powder was recovered by filtration. After drying the powder at 180°C, the above procedure was repeated.
[113] 1CP analysis of the thus-filtered solution showed that total approx. 2.5 mol%
lithium and 0.6 mol% cobalt were extracted from LiCoO2 . In this manner, a lithium
cobalt oxide (LiCoO2 ) with an approximate composition of Li2CO3O was obtained. The lithium-deficient Li0.98 CoO2 was heated at 750°C for 10 hours.
[114] The approx. stoichiometric sample, LiO, was prepared by heating the precursor
LiCoO2 at 850°C for 10 hours. A small amount (100 g) of a sample was used to allow for evaporation of eventual excess lithium impurities.
[115] The lithium-rich sample Li+ was prepared i'rom 1 kg of inexpensive LiCoO2 by
adding 1.5 mol% ball-milled LiOH-H O per 1 mol cobalt followed by a heat treatment at 750°C for 10 hours in air.
[116] X-ray diffraction (XRD) analysis showed that all of 3 samples basically had the
same XRD pattern. Particularly, the lattice constants were identical therebetween. The pH titration of the samples Li", LiO and Li+ revealed that Li" and LiO were basically free of Li2CO3 impurities, whereas Li+ contained about 1% by weight of Li2CO3 . Sample Li
would not be suitable for commercial batteries because the Li2CO3 impurities would cause un-acceptable amounts of gas (for example causing swelling of polymer cells) during charging the battery.
(117] The samples were subjected to electrochemical tests by coin cells at 3.0-4.2,
3.0-4.4 or 3.0-4.5 V and at room temperature (25°C) or elevated temperature (50°C). A typical schedule involved 32 cycles: a charge rate was C/5. During Cycles 1-5, a discharge rate performance was measured ((710, C/5, C/2, 1C and 2C). Cycles 6-30 were carried out at a C/5 discharge rate to investigate the cycling stability. Cycle 31 was carried out at a C/10 discharge rate to investigate the remaining capacity, and Cycle 32 was carried out at a C/1 discharge rate to measure the capacity loss (impedance built-up) at a high-late discharge.
[118] All samples showed an excollent cycling stability at 4.2 V, but exhibited a strong
capacity fading at 4.5 V, especially for the Li-deficient sample. Significant impedance built-up was observed in the lithium-deficient and lithium-rich samples. Details are shown in FIG. 5, and Table 1 below summarizes the results.
[119]
(Table Removed)
(120]
[121] Discussion: The data show that the cycling stability of LiCoO2 at a high voltage
dramatically changes even with a slight changes of a Li:Co ratio. The high-voltage cycling stability (and storage properties at high temperatures) is dominated by surface properties. The surface chemical properties depend on the chemical potential. Because the lithium chemical potential changes stepwise, the high-voltage cycling stability also changes stepwise. If the lithium chemical potential is fixed within a preferred region (according to the present invention), the high-voltage cycling stability can be improved.
[122]
[123] [Reference Example 2] Properties of lithium buffer LiMn1/3Ni1/3Co1/3O 2 as Li
acceptor
[124] The reference example is intended to confirm that LiMn1/3 Ni1/3Co1/3O2 is a lithium
acceptor.
[125] This is proven by mixing LiMn1/3Ni1/3Co1/3O2 with a small amount of Li2CO3 (total
Li:M=1.025:l), followed by a cooking step at 900°C. X-ray diffraction (XRD) analysis showed the following results.
11261 (1) All Li2CO3 was consumed. This fact was also confirmed by pH titration. That i
s, after immersing the sample into water, the remaining Li CO impurities were dissolved in water and detected by pH titration.
1127] (2) The lattice constants (ahex, chex, and the unit cell volume) of the final sample
(2.8602 A, 14.2302633 A, and 33.60586 A)were smaller than those of the initial LiMn 1/3Ni1/3Co1/3O2(2-8620 A' 14-23688 A and 33.66297 A).These results confirm that lithium originating from the Li2CO3 has been introduced into the crystal lattice structure, resulting in Li1+xM1-xO2 (see also Reference Example 4 for the relationship between stoichiometry and lattice constant).
[ 128] Discussion: If LiMn1/3Ni1/3Co1/3O2 is added to LiCoO2 with small amounts of Li2 CO
impurities, followed by co-sintcring, then the Li2 CO3impurities are consumed, the LiMOn buffer is lithiated to give Li1+xM1-xO2 and the lithium chemical potential of LiCoO2 is fixed at the same value as Li1+x M1-x O2 which is below the high value for
LiCoO2 coexisting with Li 2CO3.
[129]
[130] [Reference Example 31 Properties of lithium buffer LiMn1/3Ni1/3 Co1/3 O as Li
donor
[131] This example is intended to demonstrale that LiMn1/3Ni1/3Co1/3O2 is not a overly
strong lithium acceptor and can also act as a lithium donor.
[132] LiCoO and MOOH (M= LiMn1/3Ni1/3Co1/3O ) were mixed in a 5:3 ratio. The resulting
mixture was pressed into the pellent After ("-sintering of the pellets at 800°C for 1
day, X-ray diffraction (XRD) analysis was carried out on the sintered materials and
Rietveld refinement was made. The refinement yielded the following conclusions:
[133] 1) CO3 O4 , LiCoO2 and Li-M-O2 coexist.
[134] 2) The lattice constants and the unit cell volume of the final Li-M-0 were slightly
larger than those of LiMn1/3Ni1/3Co1/3O2 . That is, the final Li-M-O exhibited a small
lithium deficiency. Using the data of Reference Example 4 allows to estimate the
composition as approx. Li 1-xM1+xO2 with x=0.025 (Li:M=0.95).
[135] Applying basic thermodynamic considerations confirms that the lithium chemical
potential of LiMn1/3Ni1/3Co1/3O2 is above the low value of LiCoO2 coexisting with
cobalt oxide impurities. Conclusion: LiMn1/3Ni1/3Co1/3O2 acts as the lithium donor to
lithiate Co 3O4 impurities.
[136]
[137] [Reference Example 4] Relationship between the stoichiometry and the crystal
lattice of lithium buffer LiMn1/3Ni1/3Co1/3O2
[ 138] This reference example is intended to investigate the stoichiometry-lattice relation
of the lithium buffer LiMn1/3Ni1/3Co1/3O2 .
[1391 Commercial LiMn1/3Ni1/3Co1/3O2 with an approximate Li:M ratio of 1:1 was used as
a precursor. Two lithium-rich samples Li1+xM1-xO2 with target Li:M = 1 025-1 and
1.075:1 were prepared by adding Li2CO3to the precursor material and reacting the resulting mixture at 900°C for 24 hours in air. Two lithium-deficient samples Li1-xM1+x
O2 with target Li:M = 0.975:1 and 0.925:1 were prepared by adding a mixed hydroxide
(MOOH, M= Mn1/3Ni1/3Co1/3 ) to the precursor material and reacting the resulting
mixture at 900°C for 24 hours in air.
[140] X-ray diffraction (XRD) analysis showed a gradual and smooth change of a crystal
lattice constant as a function of lithium stoichiometry. Data are given in Table 2 below.
The same data are also shown in FIG. 6.
[141]
[1421
(Table Removed)
[143 ] The above results of Reference Examples 2-4 confirm those of the schematic FIG.
2 to FIG. 4.
[144] Conclusion: LiMn1/3Ni1/3Co1/3O2 is a suitable solid state lithium buffer. It is able to
donate as well as to accept lithium. The buffer potential matches LiCoO2 potential and is within the preferred region. It has a wide non-stoichiometric range.
[145]
[146 ] [Reference Example 5] Electrochemical properties of lithium buffer LiMn1/3Ni1/3
[1471 This example discloses further properties of the solid state lithium buffer LiMn1/3Ni1/3Co1/3O2 .
[148 ] Electrochemical properties of the samples of Reference Example 4 were tested.
Coin cells (with a Li metal anode) were prepared and were tested at 25°C and 60°C.
The charging voltage was 4.4, 4.5 and 4.6 V. Table 3 below summarizes the obtained
results. FIG. 7 shows some of the results.
[149]
(Table Removed)
[150]
[154 ]Within a relatively broad preferred region (about 0.975:1 to 1.025:1) excellent
cycling stability was achieved. The sample with high lithium-deficiency (0.925:1) showed some deterioration of rale performance. Samples with a low or high Li:M ratio (0.925:1 or 1.075:1) showed some deterioration of cycling stability.
[ 152] Discussion: The relatively broad preferred region, and the smooth change of elec-
trochemical properties are caused by the gradual change of lithium chemical potential. Further, other properties including surface chemistry (by pH titration) were checked. A similar slight and continuous variation of properties depending on the Li:M ratio was observed.
[153]
[154] [Example 1] Preparation of LiCoO2 with fixed lithium chemical potential
[155] 3.6 kg of inexpensive LiCoO2 (received from a low-cost producer) and 400 g of
commercial LiMn1/3Ni1/3Co1/3O2 were mixed by slow ball rolling, followed by a co-
sintering at 900°C for 10 hours in air. The samples was filled into vials (no sieving or
grinding was required) shortly after cooling down and was stored and further processed
in a dry room.
[156] FIG. 8 shows FESEM micrographs of the LiCoO2 precursors and the final sample
(Sample name: TR01). The morphology of TR01 was just the same as a mixture of the precursors. Particularly, LiCoO2 and LiMn1/3Ni1/3Co1/3O2 were still separated particles and were not sintered into agglomerates. A temperature of 900°C is high enough for a fast equilibration of lithium and oxygen chemical potentials. Therefore, the lithium potential of LiCoO2 is fixed at a preferred value, which is determined by the lithium chemical potential of the lithium buffer. The value is above the low value for LiCoO2 coexisting with the cobalt-oxide, and below the upper value for LiCoO2 coexisting with Li2CO3 . Furthermore, the LiCoO2 was basically free of the cobalt oxide or Li2CO3
impurities. The absence of Li2CO3 impurities was confirmed by pH titration.
[157]
[158] [Experimental Example 1] Effects of co-firing
[1 59] In order to confirm effects of heat-treatment in the present invention, the elec-
trochemical properties of TR01 prepared in Example 1 were compared with a sample which is a mixture of 90 wt% of heated LiCoO2 and 10 wt% of LiMn1/3Ni1/3Co1/3O2
[1 60] The sample was prepared in the same manner as in Sample TR01, except that two
materials were not heat treated (co-fired). Table 4 below summarizes the results.
1161]
[162]
(Table Removed)
[163] The results of Table 4 showed that a simple mixing of LiCoO2 with LiMn1/3Ni1/3Co1/3O2
without heat-treatment is not sufficient to achieve a high cycling stability at a high
voltage.
[164] Conclusion: A heat treatment is required to achieve equilibration of the lithium
chemical potential. [165]
[166] [Experimental Example 2] Coin-cell test
[167] A cathode was prepared using Sample TR01 prepared in Example 1, as a cathode
active material. Coin cells (with a Li metal anode) were prepared and were tested at 4.4
V and 4.5 V and at 25°C and 50°C. FIG. 9 and Table 5 below summarize the obtained
results.
[168]
[169]
(Table Removed)
[ 170] From the results given in Table 5 and FIG. 9, it can be seen that Sample TR01 of
Example 1 (having the lithium chemical potential fixed in a preferred region) has improved cycling properties at an elevated voltage.
[171]
1172] [Comparative Example I] Coin cell test using low-price LiCoO2
[ 173] Except using low-price LiCoO2 as a cathode active material, a coin cell was
prepared in the same manner as in Experimental Example 2, and tested under the same conditions. LiCoO2 used in this Example was the same product which was used as the precursor in Example 1. The low-price LiCoO2 powder was heated to 900°C and refreshed to have the same heat treatment history as Sample TR01 of Example 1. However, the above comparative sample was not subjected to the treatment to achieve the equilibration of the lithium chemical potential by the action of the lithium buffer.
[174] As shown in FIG. 10, the comparative sample exhibited excellent stability through
the heat treatment, which was, however, lower than that of Sample TR01 (see FIGS. 9 and 10).
[ 175] In addition, it can be seen that the comparative sample exhibited a significant
decrease of the capacity at 4.5 V, 50°C as well as 4.5V, 25°C (see C/10 data) and much more impedance built-up (see voltage depression of C/l).
[176]
[177] [Experimental Example 3] Polymer-cell lest (cycling properties)
[.178] Commercial size polymer cells (3.6 mm thickness) were prepared at pilot plant
scale. The polymer cells contained a cathode composed of 95 wt% TR01 (Example 1), 2.5 wt% PVDF binder and 2.5 wl% conductive additive (Super P), double side coated on 15 micrometer aluminum foil. The anode active material was commercial MCMB (Mesocarbon microbead). A standard commercial electrolyte (not containing overcharge additives) was used. The anode loading was chosen to achieve balanced cells at 4.4 V cell voltage (anode caparlly eathode capacity charged to 4.45 V versus Li/Li+).
[179] The polymer cells were cycled lor 400 cycles (charge rate 0.6 C, discharge rate 1
C, 1 C = 800 mA). Each 100th cycle was a "capacity check" cycle during which a slower charge/discharge rate (0.2 C) was applied. The cells were cycled at 3.0-4.2 V, 3.0-4.3 V, 4.0-3.5 V or 3.0-4.35 V. The cycling temperature was 23°C or 45°C. Frequently the increase of thickness during extended cycling was checked. Furthermore, the evolution of resistance was checked by impedance measurement at 1000 Hz.
[180] FIG. 11 shows the measurement results of cycling stability at 4.2, 4.3, 4.35 and 4.4
V at 23°C and 45°C. Even at the high cell voltage of 4.4 V and at the elevated temperature of 45°C a very high cycling stability without significant impedance built-up was observed.
[181] More importantly, at 23°C a similar rate of capacity loss was obtained for all
voltages, and additionally, the capacity loss at C/l and C/5 rates evolved similar patterns. Also at 45°C, a similar behavior of capacity loss was observed for all voltages. Therefore, it can be concluded that the increase of cell voltage did not cause degradation of the cathode.
1182] Table 6 below summarizes the obtained results.
[183]
M84|
(Table Removed)
[185] Conclusion: The obtained results clearly confirm that a modified LiCoO2 , with a
lithium chemical potential fixed within a preferred region, has an improved excellent stability at high voltage (at least up-to 4.4 V versus Li/Li+) even at elevated temperature (45°C).
[186]
[187] [Experimental Example 4] Polymer-cell test (storage properties)
[188] Polymer cells, prepared in the same manner as in Experimental Example 3, were
charged to 4.2, 4.3 or 4.35 V. After charging, the cells were placed in a temperature chamber and the temperature was increased to 90°C over 1 hour. Cells were kept at 90°C for 4 hours, and then the temperature was decreased to room temperature over 1 hour. During the temperature profile, the cell thickness was automatically monitored. Before and after the test, the cell capacity was measured at C/l and C/5 rates.
[189] No significant increase of thickness was observed at any of the investigated charge
voltages. Also, the recovery ratio did not dccicase with an increased storage voltage. It can be concluded that the increase of cell voltage did not cause degradation of the cathode.
[ 190] Table 7 below summarizes the lesully.
[191]
[192]
(Table Removed)
[193] Conclusion: The obtained results clearly confirm that modified LiCoO2, with a
lithium chemical potential fixed at a preferred region, has improved, excellent storage properties at a high voltage (at least up-to 4.4 V versus Li/Li+).
[194]
[ 1951 [Comparative Example 2] Polymer cell lest using low-price LiCoO2
[196] Except using low-price LiCoO2 as a calhode active material, a polymer cell was
prepared in the same manner as in Experimental Example 3, and tested under the same conditions as Experimental Examples 3 and 4. However, the cell always showed much inferior stability at >4.3 V and always showed strong swelling during a 90°C storage test.
[1 97 ] Cells with standard commercial LiCoO2 exhibited smooth cycling at 4.2-4.25 V,
but at 4.3-4.35 V an increased rate of capacity loss was observed simultaneously with a stronger built-up of capacity difference (=impedance built-up). This behavior was caused by the lack of cycling stability of LiCoO2 at voltages > 4.3 V versus Li/Li+.
[198]
[199] [Example 2] Preparation of LiCoO2 with core-shell structure
[200] (1) Experimental Examples 2, 3 and 4 demonstrate that modified LiCoO2 with a
lithium chemical potential fixed in a preferred region, obtained by co-sintering with a solid state lithium buffer, allows to obtain excellent storage properties and high-voltage cycling properties. This actual example modifies this approach. The LiCoO2 (having a lithium chemical potential fixed in a preferred region) is present in the core of particles, covered by a shell of the solid state lithium buffer. This concept is useful to further improve safety properties.
[201 ] (2) A layer of M(OH)2 (M=Mn Ni ) was precipitated onto LiCoO2 particles
acting as seeds during the precipitation process. During precipitation, a flow of an aqueous MSO solution (2M) and a flow of an aqueous NaOH solution (4M) were added to a reactor (5L) containing 5 kg of LiCoO2in the form of a water-based slurry (Volume = 2.7 L). The pH was adjusted to be within a preferred region, and the temperature was about 85°C. The total time of precipitation was 165 min. A total of 0.06 mol transition metal was precipitated per 1 mol of LiCoO2 After the precipitation, the slurry was filtered and washed, and the resulting powder cake was aged overnight
in 10 L of an aqueous 0.5M LiOH solution, followed by washing and drying at 180°C.
The thus-obtained powder (name: Precursor 1) was used as a precursor for the
preparation of a final cathode material.
|202] (3) To find optimum preparation condilions, a series of small scale samples was
prepared and electrochemically tested at 4.4, 4.5 and 4.6 V and at 25°C and 60°C, respectively. Samples had varying Li2CO3ntents and were prepared by adding small amounts of Li2CO3 to Precursor I, followed by heat treatment at 900°C for 5 hours. In
some cases, a small amount of fluorine (I ,i AlF , 0.2 mol% Al per 1 mol of Co) was
also added prior to the heat treatment.
[203] (4) Finally, a large sample (volume sl/,e: -I- kg, name: 4 kg-Final) was prepared by
adding 48 g of Li2CO3 and 20.5 g of a 2:1 mixture of Li A1F and Li2CO3 to 4 kg of
Precursor 1, followed by heat treatment at 900°C for 6h. Coin cells were prepared
using the thus-obtained samples and were tested at 4.4, 4.5 and 4.6 V at 25°C as well as
at 60°C.
[204] FIG. 12 shows an FESEM micrograph of the thus-prepared cathode active material
with a core-shell structure. Apparently, a shell of solid state lithium buffer fully covers the LiCoO2 core. FIG. 13 shows test results of the cycling stability. High stability during cycling at high voltage and elevated temperature has been demonstrated. The cycling stability was much improved, as compared with the results of state of the art LiCoO2 .
Industrial Applicability
[205] According to the present invention, LiCoO2 having a stoichiometric composition
can be prepared by co-fixing with a material acting as a Li acceptor and/or a Li donor, thus fixing a lithium chemical potential within the preferred range. As a result, it is possible to prepare a LiCoO2 electrode active material which has improved high-temperature storage properties and high-voltage cycling properties, and is robust to composition fluctuation in the production process.
[206] Therefore, the present invention can spend less time and labor to quality control
and process management in the mass-production of the electrode active material, and the production costs of LiCoO2 can be cut down.
[207] Although the preferred embodiments of the present invention have been disclosed
for illustrative purposes, those skilled in the art will appreciate that various modifications, additions and substitutions are possible, without departing from the scope and spirit of the invention as disclosed in the accompanying claims.
Claims
[1] A LiCoO -containing powder comprising LiCoO having a stoichiometric
composition via heat treatment of a lithium cobalt oxide and a lithium buffer material to make equilibrium of a lithium chemical potential therebetween.
[2] The powder according to claim 1, wherein the lithium buffer material is a
material that withdraws Li from Li2 CO3 , a Li-excess form of LiCoO2 , to result in
decomposition of Li2 CO3 , or supplies Li to a cobalt oxide, a Li-deficient form, to
result in production of LiCoO2.
[3] The powder according to claim 1, wherein the lithium buffer material is a
material of Formula Liz MO2 (0.95
a ratio of Mn to Ni (x/(1-x-y)) is in the range of 0.4 to 1.1).
[4] The powder according to claim 1, wherein the lithium buffer material is LiMn1/3
Ni1/3 Co1/3 O2.
[5] The powder according to claim 1, wherein the heat treatment temperature is
lower than the temperature at which the reaction between LiCoO2 and the lithium buffer material takes place to form a new compound.
[6] The powder according to claim 1, wherein the heat treatment temperature is in
the range of 400 to 1100°C.
[7] The powder according to claim I, wherein the content of LiCoO is at least more
than 40 wt%.
[8] The powder according to claim I, wherein LiCoO2 is mixed and contacts with
the lithium buffer material
[9] The powder according to claim I, wherein the lithium buffer material has a core-
shell structure that covers the surface of LiCoO 2.
[10] The powder according to claim 1, which is prepared on the scale of at least 1 kg
batch.
[11] The powder according to claim 1, which is prepared on the scale of at least 20 kg
batch.
[12] The powder according to claim 1, which is used as a cathode active material for a
rechargeable battery.
[1 3] A LiCoO2 -containing powder having a constant lithium chemical potential,
wherein LiCoO2 coexists with a lithium buffer material under equilibrium conditions of lithium chemical potential therebetween, and the lithium chemical potential of powder is higher than the equilibrium lithium chemical potential between LiCoO2 and a cobalt oxide and is lower than the equilibrium lithium chemical potential between LiCoO2 and Li2 CO3 .
[14] A lithium buffer material which acts as a Li acceptor or Li donor to remove Li-
excess or supplement Li-deficiency, each of which coexists with a stoichiometric
lithium metal oxide.
[15] The buffer material according to claim 14, which withdraws Li from Li2 CO3 , a
Li-excess form of LiCoO2 , to result in decomposition of Li2 CO3 , or supplies Li to
a cobalt oxide, a Li-deficient form, to result in production of LiCoO2 .
[16] The buffer material according to claim 14, wherein the buffer material is a
material of Formula Li2 MO2 (0.95
a ratio of Mn to Ni (x/(l-x-y)) is in a range of 0.4 to 1.1).
[17] A method for preparing a LiCoO2 -containing powder, comprising a step of
providing a homogeneous mixture of LiCoO2 and a lithium buffer material; and a
step of heating the mixture to achieve equilibration of a lithium chemical
potential.
[18] The method according to claim 17, wherein the homogeneous mixture has a
core-shell structure that the lithium buffer material covers the surface of LiCoO2 .
[19] The method according to claim 17, wherein the heat treatment temperature is
lower than the temperature at which the reaction between LiCoO2 and the lithium
buffer material takes place to form a new compound.
[20] The method according to claim 17, wherein the heat treatment temperature is in
the range of 400 to 1100°C.
[ 21 ] An electrode comprising the powder of any one of claims 1 to 13, as an active
material.
122] A rechargeable battery comprising the electrode of claim 21.
|