Description of Related Art Lithium batteries are familiar to almost everyone. Smaller lithium batteries can be found in the form of watch batteries. Larger primary lithium batteries are also relatively common. The larger batteries range in size from camera batteries to multi-kilogram units capable of delivering hundreds of Watt-hours (Wh) of energy.

Less familiar are lithium secondary, or rechargeable, batteries. Lithium batteries represent a small but growing portion of the rechargeable battery market. Although lithium secondary batteries have the theoretical potential to store over 800 Wh of energy per kilogram, even the best current systems have capacities only slightly over 100 Wh/kg, which falls short of the 150 Wh/kg desired for vehicle applications.

Present lithium secondary batteries generally fall into two categories, those employing a liquid electrolyte and those employing a solid polymer electrolyte. Liquid electrolytes exhibit the highest level of conductivity, however, in order to keep the anode and the cathode from coming into contact, they have to be used in conjunction with a porous supporting separator. The inclusion of a separator limits the achievable energy density of the system. Liquid electrolytes also require the system to contain a liquid capable of migrating within the enclosure of the cell.

Solid polymer electrolytes can be made very thin. However, even with a thinner electrolyte, battery performance is hindered because the ionic conductivity of the polymers currently in use is much lower than that of common liquid electrolytes.

Due to safety considerations and the flammability of lithium metal batteries, lithium ion batteries (prepared without lithium metal) have been developed for the commercial market. Capacities in excess of 1,200 mA/hr have been reported for some cell types containing lithium cobalt oxide cathodes (an intercalation compound)

combined with lithium intercalated graphite anodes, but their energy densities approached only 100 Wh/kg.

The chemistry for these ion-type batteries, with intercalation compounds used at both the anode and the cathode, are illustrated in Figure 1. Lithium ions are inserted or deinserted within the structure of the intercalation material depending on the direction of flow of lithium cations and electrons. Since metallic lithium is not formed during charging when intercalation materials are used at the anode, the accumulation of dendritic deposits is precluded.

The use of an intercalated carbon anode reduces the maximum energy storage density somewhat, due to an increase in the equivalent weight of the stored lithium from 6.94 g ex'for pure lithium to 79.0 g ex'for LiC6, the fully filled intercalated compound.

Part of the lost energy is recovered with the elimination of the metal (aluminum or stainless steel) support used in batteries having lithium metal anodes.

The major limitation on lithium ion battery power and stability appears to be the poor conductivity of the solid polymer electrolyte and irregularities in the interface between the solid polymer electrolyte and the solid electrodes. A lithium battery design utilizing a conducting solid polymer electrolyte membrane that assists the mobility of lithium ions flowing between the battery electrodes with liquid-like conductivity levels would be very beneficial.

Polyethylene oxide (PEO) has been extensively studied as a polymer electrolyte due to its solvation properties, however, pure PEO has low room temperature conductivity (< 10-6). PEO is a linear polyether that is highly crystalline (70-85%). The melting point of the crystalline phase (Tm) is 65° C and the glass transition temperature of the amorphous phase (Tg) is about 60° C. Above 60° C the conductivity level of PEO with lithium salts rises to the range of 10 to 10 Scm''at 100° C. PEO bears ether oxygen donor atoms which form polymer-salt coordination complexes with alkali metal and transition metal ions.

Efforts to improve the conductivity by reducing the crystallinity of PEO through cross linking (as described in U. S. Patent No. 5,009,970) or copolymerization with a less symmetrical co-monomer have shown some success, but with conductivity levels in the 10-5 S cm~'range, the results are still inadequate.

Other efforts have produced gels formed between polyacrylonitrile (PAN) and plasticizing agents such as propylene carbonate (PC) and ethylene carbonate (EC) as taught in U. S. Patent No. 5,589,295. Plasticizing agents are used to lower the glass transition temperature and the crystallinity of the polymer which theoretically increases the conductivity of the resulting matrix. The PAN gels are conductive at room temperature (>10-3 Scmi'), however, they have several limitations. They are too soft, have poor mechanical properties, and they are unstable when used in conjunction with a lithium anode. In addition, the shelf life of a PAN gel battery is less than one year.

PEO complex has also been modified by adding super acids, or their salts, and zeolites or alumina (as described in U. S. Patent No. 5,360,686), however, these modifications do not achieve a desirable level of lithium ion conductivity. Similarly, silica sol-gels and silica-PEG combinations have good conductivity, but they have poor mechanical properties and they require support from a supplemental membrane. These limitations are similar to those encountered with liquid electrolytes.

Propylene carbonate, ethylene carbonate, as well as other ether compounds, have been used as plasticizing agents and oxygen carriers in liquid and gel electrolyte lithium batteries. These solvents have structures of linear and cyclic ethers and esters. Linear and cyclic ethers and esters have high electrolytic conductivity at room temperature (~10- 3 Scm~') and high dielectric constants. However, in order to be used in a battery, they require a supporting matrix to prevent the anode and cathode from coming in direct contact and shorting the cell. The presence of this supporting matrix reduces the actual cross section area of the electrolyte and makes the conductive path more tortuous, lowering the overall conductivity level of the electrolyte. The support matrix also imposes limits on the thickness of the cell that can be produced. The latter is a significant consideration when attempting to maximize the energy storage density for applications such as electric vehicles. Size limitations, combined with the fact that a liquid electrolyte battery requires a mobile, potentially corrosive component, makes the use of these materials impractical.

It is clear from the published reports such as U. S. Patent No. 5,300,376, that liquid electrolytes have superior conductivity levels over polymer electrolytes. However, liquid electrolyte batteries suffer from several problems which seriously reduce their useful

lives. Polymer electrolyte batteries have an acceptable lifetime with poor power densities due to the poor conductivity levels of the electrolytes. A solid polymer electrolyte with the conductivity of a liquid electrolyte would provide a battery with a desirable power density and long life.

Sulfonic acids have been used to enhance lithium conductivity in polymer electrolytes in the past. When sulfonic acid functionalities were grafted onto the ends of polyethylene oxides (PEO), the conductivity increased, but not to the desired level of 10-3 S cari'. Cross linking PEO with agents containing perfluorosulfonate functionalities produced a similar result. Both unsatisfactory results are due to the fact that the polymers remained mostly crystalline even after the addition of the sulfonate functionalities. A polymer with sulfonate functionalities and a less crystalline structure would provide a highly conductive solid electrolyte.

An ideal polymer electrolyte for use in lithium batteries will have high lithium conductivity, such as that observed with liquid electrolytes, while retaining good physical properties (such as strength, flexibility and dimensional stability) and be resistant to degradation.

The design of a conventional lithium ion solid polymer electrolyte (SPE) battery involves a solid redox system in which the lithium ion is reversibly oxidized or reduced at the electrode surfaces which are separated by a solid polymer film. At the same time, the ionic conductivity of the polymer allows the mobile lithium ions to traverse between the electrode surfaces. Therefore, the solid polymer acts as an electronically insulating and ionically conducting separator. The anode acts as a source of lithium ions during discharge, and the cathodic electrode acts as a lithium ion sink. Upon recharging, the direction of the lithium ion flow is reversed; therefore, the lithium ion redox couple potentials dominate during both the discharge and recharge events. The lithium ion is preferred among alkali metal ions for battery designs on the basis of its redox potential (in excess of three volts) and its high mobility, having a room temperature conductivity as high as 10-3 Scm~'. A solid solution polymer electrolyte that is adapted to optimize the conductivity of lithium ions in a lithium ion battery would be highly desirable.

Summarv of the Invention The present invention provides an ionically conductive separator, an electrolytic cell using a separator and a method for making an ionically conductive separator. The ionically conducting separator has a perfluorinated polymer matrix having pendant ion exchange groups forming ionically conductive regions extending through the perfluorinated polymer matrix. A plasticizer is disposed within the perfluorinated polymer matrix and alkali metal ions, such as lithium, are associated with the ionically conductive regions of the polymer. The separator is preferably substantially gas impermeable.

The separator may comprise a dimensionally stable porous matrix having the perfluorinated polymer matrix formed therein. The polymer matrix can be a blend of at least two polymers, wherein one of the polymers is sulfonated. One of the polymers can be a reinforcing component such as PTFE.

The separator can be selected from perfluorosulfonic acid, sulfonated polystyrene, sulfonated polytrifluorostyrene, sulfonated polyparaphenylene, sulfonated styrene- butadiene copolymers and mixtures thereof. Likewise, examples of suitable plasticizers include, cyclic ethers, linear ethers, small molecule ethers, short chain polyethers, sulfonate salts, and mixtures thereof. Additionally, the plasticizer may be selected from propylene carbonate, diethylene glycol monobutyl ether, perfluorobutylmethylether, perfluorooctylsulfonate salt, and y-butyrolactone. Preferably, the separator comprises single phase mixture having at least 20 wt% plasticizer. In the single phase mixture, the plasticizer and the polymer are in a solid state solution at room temperature.

The invention also provides an electrolytic cell comprising a cathode capable of accepting lithium, an anode containing lithium in electrical communication with the cathode, a separator having a single phase mixture of an ionically conducting polymer, an alkali metal ion, and a plasticizer, wherein the separator is positioned between the anode an the cathode. The lithium anode can have the formula LiXC where x equals 0.16 and the lithium cathode can have the formula Li Mo2, where y equals 0.16.

The present invention also provides a method for making a separator comprising, providing an ionically conducting polymer in lithium salt form, removing water from the

polymer, and dissolving a plasticizer into the polymer to produce a single phase separator. The polymer is preferably sulfonated and fluorinated.

Additionally, the present invention provides an alternative method for making a separator comprising, providing a solution of an ionically conducting, ion exchange polymer, replacing water present in the polymer solution with an organic solvent, drying the polymer solution, mixing the dry polymer solution with a plasticizer, and forming a membrane by solvent casting the polymer/plasticizer mixture on an inert surface.

Preferably, an alkyl lithium compound, such as butyl lithium, is added to the polymer/plasticizer mixture eliminate residual water and convert the polymer to a lithium salt form.

Brief Description of the Drawings So that the above recited features and advantages of the present invention can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to the embodiments thereof which are illustrated in the appended drawings. It is to be noted, however, that the appended drawings illustrate only typical embodiments of this invention and are therefore not to be considered limiting of its scope, for the invention may admit to other equally effective embodiments.

Figure 1 is a schematic view of a typical lithium ion battery also referred to as a "rocking chair"battery.

Figure 2 is a schematic diagram of the chemical structure of some typical classes of perfluorosulfonic acid (PFSA) polymers, as used in separators described herein, such as NAFION 115 (n; 6.5 and m = 1).

Figure 3 is a schematic diagram illustrating the key features of a perfluorosulfonic acid (PFSA) membrane, i. e., C-a PTFE-like backbone phase separated from, I-an ionic region that closely resembles an aqueous electrolyte by E-an interphase or side chain region.

Detailed Description of the Invention The present invention provides a solid solution polymer electrolyte (SSPE) with high conductivity levels and structural integrity making it especially suitable for use in lithium primary or secondary batteries.

The present invention provides a solid solution polymer electrolyte (SSPE) that is produced by dissolving organic compounds having solvating properties for lithium salts and normally liquid at near ambient temperatures, such as small molecule ethers or short chain polyethers in a compatible organic polymer such as perfluorosulfonic acid (PFSA) or related polymers. Some suitable organic polymers include perfluorosulfonic acids, sulfonated polystyrene, polytrifluorostyrene, sulfonated polyparaphenylene, and sulfonated styrene-butadiene copolymers. One such organic polymer is commercially available from duPont de Nemours, Inc. under the name NAFION'. The chemical structure of NAFIONs is shown in Figure 2.

Figure 3 is a schematic diagram of the morphology of typical PFSA materials in the lithium salt form. As shown, there are three different regions within the structure.

The backbone region C is largely crystalline, and known from x-ray diffraction studies to have a structure nearly identical to pure polytetrafluoroethylene (PTFE). An ether side chain region E exhibits behavior that is dominated by ether side chains. An ionic region I is dominated by sulfonate end groups. Regions E and I are amorphous.

The PFSA polymers of Figure 3 readily absorb ethers and similar organic molecules, virtually exclusively into the regions labeled E and I. The addition of ether or other certain organics to the polymer has the effect of expanding the volume of the polymers by increasing the volume of the amorphous regions and smoothing out the differences between the ion rich regions (I) and the ether side chain regions (E). The expansion of the amorphous regions makes a large fraction of the polymer behave like a liquid by reducing the glass transition temperature of the polymer. Lithium ions exhibit increased mobility above the glass transition temperature of the polymer. This provides a SSPE with liquid-like mobilities of lithium while the whole of the polymer remains knitted together into a solid by the crystalline regions C.

PFSA polymers are generally synthesized by the copolymerization of a derivatized, or active, co-monomer with tetrafluoroethylene, (TFE). After synthesis, the resulting polymer is both hydrophobic and electrochemically inert and is converted into an active ionomer by a base hydrolysis process. The result of this step is an ionomer in its salt form. This can be converted to the proton form by ion-exchange with a strong acid or to other salt forms by ion exchange with an aqueous solution of the appropriate

compound. In this electrolyte, the sulfonate groups (R-S03-) act as the stationary counter charge for mobile Li+ cations, which are generally monovalent.

One embodiment of the present invention provides an organic ion carrier PFSA mixture that is a single phase blend enabling a liquid-like lithium conductivity level, while maintaining the structural rigidity of the PFSA membrane. The organic ion carrier can be any ether capable of dissolving in the PFSA matrix, such as perfluorobutylmethyl ether (C4F90CH3). This partially fluorinated ether acts to swell PFSA by entering its structure, and in the process softens the crystalline structure of the polymer slightly, thus improving its ability to solvate lithium ions.

The resulting SSPE has several advantages over the gel membranes described above. First, the matrix of the SSPE is fully fluorinated thus making it immune to reductive attack by reduced lithium. Second, the side chains on the polymer contain ether functionalities to aid in coordinating lithium ions. Coordination is used here in the sense of a classical coordination compound where lone pairs of electrons on the oxygen atoms in the chain"donate"charge to the lithium ion to neutralize its positive charge by spreading it over a larger volume. Third, the polymer also contains anionic sulfonate functionalities which serve as fixed anion counter charges for mobile lithium ions. The presence of these fixed anion counter charges is important because it reduces, if not eliminates the need to add lithium salts. The charge density produced by the fixed anion counter charges present in PFSA materials is comparable to that produced by the type of lithium salt additions typically used (about 0.8 M), in polyether systems. The fixed anion counter charges also prevent the formation of mobile ion pairs. This is useful because mobile ion pairs are larger and less susceptible to the electric field gradient present in the battery, thus reducing conductivity.

In another embodiment, other organic ion carrier compounds can be dissolved in the PFSA material to increase the concentration of sulfonate anions available to serve as counter charges for mobile lithium ions. These include fluorinated sulfonate compounds such as triflic acid (CF3SO3H) and its salts, perfluorooctylsulfonate (C8F, 7SO3) salts, a full range of other partially and fully fluorinated sulfonate compounds and partially or fully fluorinated phosphonate compounds with chain lengths from one carbon to ten carbons. Also useful in this role are a variety of more complex fluorosulfonate,

fluorodisulfonate and fluorophosphonate compounds, such as bis (trifluoromethylsulfonyl) methane ((CF3SO2) 2CH2), (HO3S (CF2) 2) 2Os (HO) 2P (O) OCH2 (CF2) 4CH20P (O) (OH) 2, C4F9SO2N (HO) SO2CF2CF2) 2, and related compounds. These compounds are preferably mostly or fully fluorinated and therefore resistant to degradation by reduced lithium. Several methods can be used for the production of SSPE membranes. The first of these routes uses PFSA films as a starting material. These films are available as sheets as thin as 25 um. The proton form of the PFSA film is converted to the lithium ion form by ion exchange using lithium hydroxide or a mixture of lithium hydroxide and lithium nitrate. If the film is obtained in the protonated form, a single 60 minute immersion treatment should be sufficient to complete the exchange of lithium ions for protons. If the film is initially in the sodium form, an initial conversion to the proton form using a strong acid is required.

The lithium ion form of the PFSA film is then rigorously dried using both heat and vacuum to remove as much water as possible. The presence of water will not inhibit the uptake of ethers, or other compounds by PFSA materials, however, it can react with lithium intercalated in the carbon anode to reduce the stored charge and produce hydrogen gas. The drying step is followed by dissolving an organic ion carrier compound (generally an ether) into the PFSA film. The organic ion carrier can be either a pure liquid or dissolved in a volatile solvent, the latter is the preferred method of addition for materials which are not liquids at room temperature, such as perfluorooctylsulfonate.

Organic ion carrier compounds will readily swell PFSA polymers, but in some cases, when the swelling kinetics are poor, it is necessary to dissolve the desired organic ion carrier in a volatile solvent prior to adding it to the membrane. The volatile solvent must be low boiling and have an ability to rapidly dissolve in the membrane and swell the membrane. Volatile solvents suitable for dissolving the organic ion carrier include methyl alcohol, ethyl alcohol, and most preferably tetrahydrofuran. The low boiling solvent is subsequently removed by evaporation in a vacuum oven. The swelling procedure is then repeated until the membrane is saturated with the desired organic ion carrier.

Another method of membrane preparation involves recasting PFSA from solution.

Normal solutions of PFSA polymers contain up to 50% water, which must be removed

prior to recasting. Water is removed by solvent transfer to a solvent that is compatible with the organic carrier compound. After carrying out the solvent transfer, the dry PFSA solution and the organic ion carrier solution are combined and blended. To ensure that the solution is completely dry, any residual water that remains can be removed by treating the solution by adding a small amount of butyl lithium solution. Treatment with the butyl lithium also ensures that the PFSA polymer is exchanged into the lithium form. An SSPE is then produced by solvent casting onto a smooth inert surface and allowing it to dry.

For some PFSA materials it is possible to make anhydrous solutions directly.

This is accomplished following the procedure described by Martin and Moore (U. S.

Patent No. 4,731,263) using one of the solvents that they list, N, N'-dimethylformamide (DMF), dimethylsulfoxide (DMSO), N-methyl pyrrolidone, tetramethyl urea, triphenylphosphate, dimethylacetomide, sulfolane, and butyrolactone, all of which are aprotic and boil at temperatures above 120°C, or another solvent having similar properties. Although dissolution is described as occurring at room temperature, both heat and agitation are reported to accelerate the process.

In the application of the present invention, lower equivalent weight ionomer membranes are preferred because they have a higher density of fixed counter ions permitting a higher density of charge carriers and enhancing their mobility. Lowering the equivalent weight of an ionomer generally leads to an improvement in ionic conductivity.

But the gain in ionic conductivity provided by low equivalent weight ionomers is often offset by the loss of certain physical properties, such as tear strength and puncture resistance.

In another variation, the SSPE is comprised of multiple layers, each with a composition different from that of the layers adjacent to it. A polymer resistant to reductive attack by lithium is used to fabricate the face of the SSPE in direct contact with the anode of the battery. This same polymer may also be used for the face in contact with the cathode. The core of the SSPE is fabricated from another, less resistant polymer. The layers are bonded and the liquid component is dissolved in the solid to produce a composite membrane. This approach permits a backbone capable of forming a very conductive solid solution using a polymer that is normally susceptible to degradation by reduced lithium to be used by protecting the polymer from contact with reduced lithium

by a thin (1,000 to 50,000 thick) film of polymers resistant to attack by reduced lithium.

Composite membranes fabricated by the processes of the invention may comprise a reinforcing component that supplies structural integrity to the membrane while the low molecular weight ionomer may confer to the membrane the desired ionic conductivity.

The desired SSPE exhibits good physical strength, high ionic conductivity, shape stability, resistance to oxidation, and be durability. A dimensionally stable composite membrane may include organic polymers or non-polymeric substrates, such as quartz wool, sodium free glass and woven alumina fibers. Such a reinforced ionically conducting membrane may be formed by blending at least two thermoplastic polymers, forming a membrane, and derivatizing one of the components of the blend. For example, a composite membrane having controlled ion conductivity and structural integrity may be fabricated by melting a precursor to an ion conducting polymer, such as the sulfonyl fluoride intermediate form of a PFSA with an inert polymer providing structural strength, such as PTFE. The two polymers are thoroughly blended, and fabricated into a membrane by methods such as extrusion or casting. The ionically conducting membrane may then be obtained by hydrolyzing the precursor with a strong base in the case of sulfonyl fluoride precursors to PFSA membranes.

For example, a composite membrane comprising an ion conducting component and a reinforcing component may be formed by combining a solution of a polymer, such as NAFION, which is ion conducting, with a suspension of a polymer, such as PTFE, which will provide the structural reinforcing component in the membrane. The mixture may be processed to the perfluorosulfonate ionomer from the proton form or to the tetraalkylammonium form which is a pseudo-thermoplastic, and not ionically conductive.

The mixture is then precipitated to form a gelatinous to solid mass. The gelatinous to solid mass is fabricated into a sheet and sintered. The sintered sheet is then fabricated into a composite membrane and if the perfluorosulfonate component had been converted to an alkyl ammonium form, that component should be converted back to an ion conducting form through the use of a strong mineral acid, such as nitric acid followed by neutralization with a lithium containing base, such as lithium hydroxide.

The reinforcing component must be porous and is preferably substantially inert.

The pores or voids in the reinforcing component may be filled with a polymer, such as a non ion-conducting thermoplastic precursor of a polymer that is ion conducting. After filling the pores of the matrix with the precursor, the precursor may be processed and transformed into the ion conducting polymer. The process allows for the fabrication of a reinforced ionically conducting membrane with controlled ion exchange properties, which may be more difficult to obtain by directly filling the pores of the matrix with the non-thermoplastic ion conducting form of the polymer.

For example, a polymer, such as the sulfonyl fluoride precursor form of a PFSA polymer, may be impregnated into the void portion of a porous, but substantially inert substrate sheet or other fabricated shape. After impregnation, the added polymer may be converted into its active ion exchange form.

The following examples show the function of this invention and some of its preferred embodiments.

Example 1 A NAFION'105 perfluorosulfonic acid membrane was converted from the proton form to the lithium form by immersion overnight in an aqueous solution containing an excess of lithium hydroxide. The lithium membrane was dried overnight in a vacuum at 110 °C and protected from moisture until needed. To produce the solid solution, a piece of the membrane was weighed and placed in a container of propylene carbonate (C403H6, also known as 1,2-propanediol cyclic carbonate) and allowed to remain there for several days. At the end of that time, the membrane was again weighed and found to have gained 66.0 % in weight, to yield a solid solution containing 39.7 % liquid ether (propylene carbonate).

Example 2 A NAFIONO 105 perfluorosulfonic acid membrane was converted from the proton form to the lithium form by immersion overnight in an aqueous solution containing an excess of lithium hydroxide. The lithium membrane was dried overnight in a vacuum at 110 °C and protected from moisture until needed. To produce the solid solution, a piece

of the membrane was weighed and placed in a container of diethylene glycol monobutyl ether (C803H, 5) and allowed to remain there for several days. At the end of that time, the membrane was again weighed and found to have gained 110.1 % in weight, to yield a solid solution containing 52.4 % liquid ether.

Example 3 The conductivity of the film produced in Example 1 was determined using the van der Pauw four-probe method at a temperature of 22°C. This is a method originally developed for determining the conductivity of thin sheets. Since it is a four-probe method, the conductivity values obtained are free of contact resistance. This solid solution film was found to have a conductivity of 1.92 mS cm-'.

Example 4 The conductivity of the film produced in Example 2 was determined using the van der Pauw four-probe method at a temperature of 22°C. This is a method originally developed for determining the conductivity of thin sheets. Since it is a four-probe method, the conductivity values obtained are free of contact resistance. This solid solution film was found to have a conductivity of 1.78 mS crri'.

Example 5 As a control the conductivity of a NAFION'105 membrane film was determined using the van der Pauw four-probe method at a temperature of 22°C. This film was found to have a conductivity of 0.87 mS cm-'.

While the foregoing is directed to the preferred embodiment of the present invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims which follow.