BACKGROUND OF THE INVENTION The present invention relates to a high voltage, high capacity rechargeable electrochemical battery cell which comprises a positive electrode, a negative electrode, and an interposed separator with an electrolyte comprising, during operation of the cell, a pair of different mobile cation species which individually participate in redox activity at the respective electrodes. More particularly, the invention relates to the preparation and use of a rechargeable battery cell comprising a polyvalent electrode material which participates predominantly, during cycling of the cell, in a redox reaction with the first of a pair of cation species, of which one is polyvalent, present in the cell electrolyte while the second of the cation species reacts predominantly at the opposite electrode of the cell. These contemporary redox reactions enable multiple electron per ion transfer during cell operation with a resulting significant increase in cell capacity without loss of high voltage output.

The present market for compact, light-weight rechargeable batteries is served in great measure by lithium intercalation batteries, particularly Li-ion cells, which, by virtue of the light weight of the lithium electrode and electrolyte component materials, provide a significant level of specific capacity, i. e., the amount of energy per unit of cell weight that can be stored and transferred from a cell. The high reactivity of

lithium yields an additional benefit in providing an exceptionally low electrical potential in an incorporating negative cell electrode, which may comprise lithium metal or alloy, or a lithium-intercalating material. As a further advantage, a wide variety of metal oxide, sulfide, or fluoride materials are available which react with lithium at high electrical potential, thereby enabling their use as positive electrode components in resulting high-voltage battery cells.

The advantageous composite effect of the light weight and high voltage operation of Li-ion cells on the resultant specific energy density of these rechargeable batteries is marred, however, by the limitation that the mobile lithium cation upon which cell operation depends is monovalent and therefore capable of accounting for the operative transfer of only a single electron per available Li+ ion.

Considering the dependence of cell capacity upon the valence of the charge transfer ion, an alternative means of increasing the capacity of an electrochemical cell would logically appear to involve the use of polyvalent reactive components. Such an approach has been considered, as in US Patent 5,601,949; however, the substitution of polyvalent cations for the monovalent lithium in an attempt to achieve higher capacity intercalation battery cells has met with little actual success. The failure of such cells appears to be attributable to a number of causes, not the least of which is the significantly greater size of the polyvalent ion which prevents effective intercalation into negative electrode compositions, such as the graphite or other carbons materials proposed in that patent specification.

An additional deterrent to the effective operation of a polyvalent ion cell is the passivation layer of reaction materials, referred to as a solid/electrolyte interface (SEI), typically of reduction byproducts, e. g., electrolyte cation oxides, fluorides, carbonates, and the like, which form at the surface of the negative cell electrode during the first cycle charging period. While Li+ ions of a common Li-ion intercalation cell are able to diffuse through the SEI layer in order to contact and be reduced at the negative electrode, polyvalent cations cannot diffuse in this manner and are significantly deterred from participating in the essential redox reaction at the negative electrode. Although some reduction of the polyvalent cation may transpire, the reaction occurs at the invariably higher potential of the passivation layer reaction products, thus decreasing the potential difference between the electrodes with a resulting decrease in the operating cell voltage.

Therefore, the practical utilization of polyvalent electrochemical cell components in order to increase cell capacity requires the implementation of a mechanism other than the simple transmission of a species of mobile polyvalent cation between cell electrodes. The present invention provides such a novel and effective mechanism to enable the capacity-improving use of polyvalent cell components.

SUMMARY OF THE INVENTION A rechargeable electrochemical cell prepared according to the present invention comprises a positive electrode member, a

negative electrode member, and an interposed separator member which is ion-transmissive and electron-insulative. Also interposed and contained between the electrode members is an electrolyte comprising a polyvalent cation species, the electrolyte preferably being a non-aqueous solution of a solute providing polyvalent cations such as, e. g., Y3+, La3+, Mg2+, Ca2+, Ba2+, or Sr. The positive electrode member comprises an active material, such as a transition metal oxide, sulfide, fluoride, or carbon fluoride, which can take up and release the polyvalent cation in a reversible oxidation reaction of intercalation, alloying, adsorption, or the like during operation of the cell.

The negative electrode member comprises an active material which provides a source of a second, highly reactive, negative-acting cation species, preferably of an alkali metal, such as Li+, Na+, K+, Rb+, or Cs+, capable of being reversibly released into and taken up from the electrolyte solvent during operation of the cell. Such a negative electrode active material may be the alkali metal, an alloy of the alkali metal, or a carbonaceous material, e. g., coke, hard carbon, or graphite, capable of intercalating the alkali metal cation.

One embodiment of a cell of the present invention comprises a positive electrode member of V205, a negative electrode member of LiSi, and an electrolyte of 0.5 M Y (C104) 3 in a 2: 1 mixture of ethylene carbonate (EC) and dimethyl carbonate (DMC) saturating a borosilicate glass fiber separator membrane.

During the initial discharge of the cell, Y3+ ions from the electrolyte move to the reversible reaction at the positive electrode while Li+ ions from the negative electrode are released into the EC: DMC solvent of the electrolyte. Due primarily to the physical proximity to the positive electrode of the relatively high concentration of Y3+ ions and the higher overall potential

of intercalation, these reactions predominate at their respective electrodes.

Upon recharging of the cell, the reactions tend toward reversal in the usual manner, i. e., with deintercalation or other release of the Y3+ ions from the positive electrode and movement of both cation species toward reduction at the negative electrode. However, due to the rapid formation of passivation products at the surface of the negative electrode, only the Li+ ions are able to diffuse through the SEI layer in order to reach the LiSi negative electrode material where they are reduced at a potential of about that of the theoretical-3.0 V vs SHE.

Despite ever greater applied recharge voltage, the passivation layer at the negative electrode prevents the reduction of the Y3+ ions, which remain in electrolyte solution, thus maintaining the low relative potential of the negative electrode and the resulting high operating voltage of the cell.

The procedures for fabricating laminated polymeric electrolytic cell electrode members which have been widely used in practice, such as described in US Patent 5,460,904, serve well in the preparation of electrode members of cells of the present invention. In this manner, positive electrode members may be readily prepared by dispersing 28 parts of an active material capable of intercalating polyvalent cations, e. g., any of various vanadium oxides and cobalt oxides, preferably in nano-material form, with 6 parts conductive carbon in a matrix composition comprising an organic solution, e. g. in 28 parts acetone, of about 15 parts binder polymer, such as a poly (vinylidene fluoride-co-hexafluoropropylene), and 23 parts of a primary plasticizer for the polymer, e. g., dibutyl phthalate.

The composition is cast as a layer which is air-dried to a membrane at room temperature prior to being cut to desired size for cell fabrication. The membrane specimen may then be laminated to an electrically conductive current collector member and thereafter to counter-electrode and separator members. The laminated assemblage is usually then extracted of incorporated plasticizer with a polymer-inert solvent, such as diethyl ether, prior to the addition of electrolyte solution. Although commercial cells will preferably be fabricated as fully laminated electrode-separator assemblies, experimental laboratory models are more readily assembled for testing in Swagelok test cells which in essence closely resemble the physical pressure style battery cell such as is typified by the familiar"button"battery. This latter style battery structure may be used as well to embody the present invention.

BRIEF DESCRIPTION OF THE DRAWING The present invention will be described with reference to the accompanying drawing of which: FIG. 1 is a diagrammatic representation in cross section of a laminated battery cell embodying the present invention; FIG. 2 is a graph tracing characteristic recycling voltage and specific capacity in a single monovalent cation cell of the prior art;

FIG. 3 is a graph tracing characteristic recycling voltage and specific capacity in a single polyvalent cation cell; FIG. 4 is a graph tracing characteristic recycling voltage and specific capacity in one embodiment of a dual cation cell of the present invention; and FIG. 5 is a graph tracing characteristic recycling voltage and specific capacity in another embodiment of a dual cation cell of the present invention.

DESCRIPTION OF THE INVENTION As seen in FIG. 1, a battery cell structure 10 useful in the present invention comprises, preferably in the form of a laminated assembly of members such as described in the above- mentioned US Patent 5,460,904, a positive electrode member 13, a negative electrode member 17, and an interposed separator member 15 containing cell electrolyte. Current collector members 11,19 associated with the respective positive and negative electrode members provide electrical circuit connections for the cell, such as at extending terminal tabs 12,16. For laboratory test purposes, it is useful to provide an intermediate electrode, such as a silver wire 14, within separator member 15 in order to establish a quasi-reference electrical potential for the respective positive and negative half-cells.

Typically, positive electrode 13 comprises a vinylidene copolymer matrix membrane containing a dispersion of nano-sized

active material, such as a transition metal oxide, e. g., V205, MnO2, or Co304, capable of intercalating or adsorbing polyvalent electrolyte cations, for instance those of the alkaline earth group. Negative counter-electrode 17 comprises a similar copolymer matrix dispersion of a nano-material compound, or simply a metal foil, capable of reversibly plating, alloying, intercalating, or otherwise reacting with, and thus providing a source of, monovalent cations, such as of Li, Na, or other alkali. Separator 15 may likewise be a polymeric membrane, as described in the referenced specification, or it may comprise a widely used microporous membrane or simply a glass fiber mat, any of which is capable of absorbing the non-aqueous electrolyte, e. g., about a 0.5 to 2 M solution of a polyvalent cation compound in a solvent mixture of cyclic and acyclic carbonates. Such an electrolyte may additionally comprise a small amount of a monovalent alkali salt which can benefit the reaction kinetics of the negative electrode and enable fabrication of the cell in the discharged state, as well.

The datum reference established by optional Ag electrode 14 provides a convenient means for determining individually the electrolytic activity of selected composition constituents at the respective electrodes. In this manner, effective electrode and electrolyte combinations may be identified. For example, implementation of such a reference electrode was instrumental in confirming the electrolytic cell mechanism wherein a polyvalent cation species, e. g., Y3+, is denied access to a passivated alkali metal negative electrode and is thus unable to plate or reduce at that electrode in order to effect cell charging, despite applied voltages greatly in excess of that theoretically required. In like manner, it was determined that the highly positive reduction reaction of the Y3+ cation occurring in the

half cell at the accessible surface of a passivated negative electrode in a single cation cell was responsible for the essentially ineffective resultant voltage level exhibited by one of the cells exemplified below.

In fabricating working battery cells, selected cell compositions and components were conveniently assembled in standard Swagelok test cell apparatus in which positive and negative electrode members with intervening electrolyte- saturated separator member are compressed between opposing current collector block members to achieve the essential intermember contiguity. After assembly, each test cell was arranged in circuit with a MacPile automatic cycling control/data-recording system operating in the galvanostatic mode at a preselected cycling rate of about 7.6 mA per g of active material to obtain a characteristic signature voltage/capacity profile of the test cell.

In the light of the foregoing discussion, the following examples will provide the skilled artisan with further guidance toward selection of useful combinations of components and compositions for effective practice of the present invention.

Example I A lithium intercalation test cell was fabricated as a comparative example of the operating voltage level and capacity achieved in a single monovalent cation battery cell typical of the prior art. A positive electrode was cast as a layer of a composition comprising 28 parts by weight of nano-sized V205, 6 parts of conductive carbon black (MMM super P), 15 parts of

poly (vinylidene fluoride-co-hexafluoropropylene) (Elf Atochem, Kynar 2801), and 23 parts of dibutyl phthalate plasticizer in 28 parts of acetone. The layer was dried at 22°C for about 0.5 hr to form a self-supporting membrane, and disks of 1 cm2 were cut from the membrane to provide electrode members comprising about 4 to 10 mg of active material, i. e., V20, 5-In order to the cell for introduction of electrolyte in the manner of prior art laminated cell structures, the plasticizer was extracted from the electrode disk member with diethyl ether.

A negative electrode member of LiSi was likewise prepared from a cast layer of a composition similar to that of the positive electrode, but for the substitution of Si for the V205.

A segment of the dried, extracted layer was overlaid upon a segment of lithium foil and an electrode member disk was cut from the composite material. The LiSi alloy having a surface area of greater than about 0.5 m2/g spontaneously formed in situ at the electrode disk member over a short period of time.

The electrode members were assembled under substantially anhydrous conditions (-80°C dewpoint) in a Swagelok test cell with an intervening disk of borosilicate glass fiber mat saturated with a 1 M electrolyte solution of LiC104 in a 2: 1 mixture of ethylene carbonate (EC) and dimethyl carbonate (DMC).

The cell was then cycled in circuit with the automated test controller/recorder for a number of periods during which the Li+ electrolyte cation reactions of intercalation at the positive electrode during discharge and reduction at the negative electrode during recharge were repeated in the usual manner. The recorded data, comprising the two-electrode output voltage of the cell and indicating a typical specific capacity of about 125

mAh/g, were plotted to yield the characteristic trace depicted in FIG. 2.

Example II A second comparative example of a battery cell comprising a single polyvalent cation was prepared in the manner of Example I utilizing the positive electrode member of Example I and a negative electrode member comprising a nano-sized YSi2 powder active material. The electrolyte was a 0.5 M solution of Y (Cl04) 3 in the 2: 1 mixture of EC: DMC.

Examination, with the aid of quasi-reference electrode 14, of the Y3+ cation half-cell reactions at the respective positive and negative electrodes revealed definite intercalation or adsorption at the positive electrode with reduction or alloying at the negative electrode, thus confirming reversible polyvalent Y3+ cation activity within the cell. Unfortunately, however, the voltage of the negative electrode material reaction of the single polyvalent cation is too high with respect to the positive electrode reaction to establish the significant voltage difference between electrodes which is necessary to provide a useful resultant working voltage level in the battery cell. This shortcoming is apparent in the cycling voltage trace depicted in FIG. 3 which shows a desirable substantial increase in cell capacity, but little useful voltage output.

Example III

A battery cell embodying the present invention, i. e., comprising dual cations including at least one which is polyvalent, was prepared generally in the manner of the foregoing examples, comprising in the respective positive and negative electrodes materials capable of intercalating or adsorbing a polyvalent cation, such as that of yttrium, lanthanum, or an alkaline earth metal, during the discharge cycle segment and of reducing, plating, or alloying with the smaller and more reactive second cation, typically of a monovalent alkali, during the charging cycle segment. In combination with such electrode materials, the electrolyte provides the polyvalent cation and is capable of readily receiving into the electrolyte solution the second cation species.

Specifically, the positive electrode member of this dual cation cell comprised the V205 nano-material of Example I and the negative electrode member comprised the LiSi of that example.

Thus, although the active electrode materials of the cell may serve equally as well in the cell structure of the present invention as in those of the prior art, a surprisingly effective distinction is made in the electrolyte cation employed.

According to this invention, the cation of the electrolyte is selected to be the polyvalent cation of the dual cation combination while the complementary cation is typically the monovalent cation component of the negative electrode composition. In the present example, the electrolyte is a 0.5 M solution of Y (C104) 3. A trace of the cycling voltage of the cell is depicted in FIG. 4 and shows the remarkable specific capacity reaching to about 250 mAh/g while the resultant working voltage is stable and in the range of about 3 to 3.5 V.

The theorized mode of operation of the dual cation cells of the present invention appears to follow the generalized process in which, during cell discharge, the polyvalent Y3+ cations in the electrolyte solution are taken up at the positive electrode while Li+ cations enter the solution from the negative electrode, and, during cell recharge, the Y3+ polyvalent cation species reenters the electrolyte solution while the Li+ cations are reduced and plated or alloyed at the negative electrode to maintain a stable low voltage cell datum at about-3 V vs SHE.

This mode of operation in which the reactions of one cation species predominate at one electrode while the reactions of the complementary cations species predominate at the other electrode has received supporting evidence in a series of energy dispersive spectroscopy (EDS) assays of specimen cell electrodes at various stages of reversible cycling. From these EDS test results, the respective electrode predominance of cation reactions is confirmed in the fact that the Y/V ratios at the positive electrode are consistent with the degree of cell discharge and recharge.

Example IV Another embodiment of the dual cation cell of the invention was prepared with a negative electrode member of metallic lithium on a nickel support, a positive electrode member comprising a Co304 active material, and a 0.5 M electrolyte solution of Y (Cl04) 3. This cell provided results similar to the cell of Example III.

Example V Yet another embodiment of the present invention was prepared with a negative electrode member of metallic sodium on a stainless steel support, a positive electrode member comprising a V205 active material as in Example III, and a 0.33 M electrolyte solution of Y (CF3S03) 3. The trace of cycling characteristics of the cell as depicted in FIG. 5 confirms the high voltage stability and improved capacity of this cell of dual cation configuration. A similar cell limited to a monovalent configuration, i. e., utilizing an electrolyte comprising NaCl04 in propylene carbonate, provided only about 90 mAh/g of active capacity.

It is anticipated that other embodiments and variations of the present invention will become readily apparent to the skilled artisan in the light of the foregoing description and examples, and such embodiments and variations are intended to likewise be included within the scope of the invention as set out in the appended claims.