Hydrophilic Membranes for Electrochemical Devices and Method for Preparing Same

This invention relates to articles such as membranes having an extremely thin hydrophilic polymeric shell about surfaces of the article while substantially retaining the geometric configuration of the article, the use of hydrophilic microporous membranes in separation processes, and the method of preparing such articles.

Many polymeric materials are hydrophobic. When such materials are formed into films, beads, membranes or the like, their hydrophobic nature prevents or inhibits "wetting" by water.

When used to describe a surface, the term "hydrophobic" means that water on that surface has a contact angle of greater than ninety degrees. By contrast, the term "hydrophilic" applies to those polymeric surfaces which have a contact angle of less than ninety degrees.

While hydrophobic materials are well known in the art and easily prepared, their usefulness in many processes and products is severely restricted by their hydrophobicity. There have been numerous prior attempts to render a hydrophobic material hydrophilic in order to be useful in processes where water is present and must "wet" the surface of the material.

Several efforts have concentrated in rendering hydrophilic a porous hydrophobic polymeric membrane. Despite the low cost of preparation of such hydrophobic materials in the form of porous membranes, such membranes are not useful as membranes in aqueous systems because capillary forces at the pores of such hydrophobic materials prevent the wetting of the pores by water, aqueous solutions, or other high surface tension organic solutions.

Treatment of the surfaces of hydrophobic materials, such as porous membranes, made from polyolefins has been attempted using surfactant coatings such as the silicone glycol copolymer disclosed in U.S. Pat. No. 3,853,601 (Taskier) or the nonionic alkylphenoxy poly(ethyleneoxy)ethanol surfactant disclosed in U.S. Pat. No. 4,501,793 (Sarada), or a copolymer coating having hydrophilic monomeric units and hydrophobic monomeric units such as an ethylene-vinyl alcohol copolymer disclosed in European Patent Office Publication No. 0 023 459 (Nitadori et al.). Unfortunately, such surfactant treatments to the surfaces of hydrophobic materials may not be permanent due to the washing away of such surface coatings by water or a variety of organic solvents including those used to form the coating on the supporting hydrophobic article. Also, surfactants are commonly known to denature enzymes. See, for example, Molecular Cell Biology, J. Darnell et al. Fds., Scientific American Books, 232, (1986).

Another approach taken in the art is the adsorption of a hydrophilic polymer on a hydrophobic substrate, as disclosed in U.S. Pat. No. 4,794,002 and corresponding European Patent Office Publication 0 221 046 (Henis et al.). A modifying polymer may be adsorbed onto the surfaces of a polysulfone or a polyethersulfone from an aqueous solution of the modifying polymer. But the modifying polymer can be removed with detergent solutions and the like.

Relatively permanent hydrophilic coatings on hydrophobic microporous films have been attempted by further treatment of chemical cross-linking of or ionizing radiation directed against the coating. U.S. Pat. No. 4,346,142 (Lazear) discloses an ionizing radiation process. U.S. Pat. No. 4,776,959 (Kasai et al.) discloses thermally curing a water insoluble vinyl alcohol-vinyl acetate copolymer onto a porous membrane. U.S. Pat. No. 4,753,725 (Linder et al.) discloses semipermeable composite membranes made by reacting PVA/PVA-copolymer films with a monomeric organic compound containing at least two functional groups, a linear or branched polyfunctional oligomer or polymer, and a compound containing cross-linking and ionizable groups. Japanese Publ. No. JP62-14903 (Ohtani et al.) describes using a solution containing a compound having ester side chains and a crosslinking agent to thermally crosslink the ester side chains to hydroxyl or carboxyl reactive sites on the hydrophobic polymer.

Others have attempted to apply hydrophilic poly(vinyl alcohol) directly to the hydrophobic polymer membrane. Japanese Publ. No. JP62-277106 (Ikehara et al.) describes the ionic cross-linking of a poly(vinyl alcohol) on a microporous polymer substrate from a water-soluble poly(vinyl alcohol) polymer containing an inorganic alkaline compound. While poly(vinyl alcohol) has excellent hydrophilicity, processing difficulties are encountered when one attempts to coat hydrophilic poly(vinyl alcohol) directly onto the hydrophobic membrane from a polar or aqueous solution.

Another has attempted to form hollow fiber microporous membranes with poly(vinyl alcohol) chemically bonded to the surfaces of the hollow fiber membrane. U.S. Pat. No. 4,885,086 (Miura) discloses that a hollow fiber membrane is irradiated with ionizing radiations and then reacted with vinyl acetate and hydrolyzed.

The attempts described in the art to provide a hydrophilic poly(vinyl alcohol) coating are based on using atactic poly(vinyl alcohol), which has a low crystallinity content. It is believed that coatings based on atactic poly(vinyl alcohol) are more soluble in a range of solvents and aqueous fluids and consequently the coatings are more readily washed away, particularly when contacted with solvents miscible with the solvents used to bring the hydrophilic material in contact with the hydrophobic membrane.

It is possible to produce poly(vinyl alcohol) which is not atactic. Preparation and the properties of syndiotactic and isotactic poly(vinyl alcohol) have been described in Harris et al., Journal of Polymer Science: Part A-1, Vol. 4, 665-677 (1966), describing the preparation of syndiotactic poly(vinyl alcohol) from poly(vinyl trifluoroacetate) and isotactic poly(vinyl alcohol) from poly(vinyl tert-butyl ether). Further, the production of poly(vinyl trifluoroacetate) as a precursor for syndiotactic poly(vinyl alcohol) has been described in Haas et al., Journal of Polymer Science, Vol. 22, pgs. 291-302 (1956).

Prior uses of such tactic poly(vinyl alcohol) materials have included the preparation of ophthalmic articles, such as contact lenses and coatings for such articles, from non-crosslinked poly(vinyl alcohol) copolymers hydrated to have controlled hydrogel properties and high strength. Co-assigned, related U.S. Pat. Nos. 4,528,325; 4,618,649; 4,693,939; 4,694,037; 4,780,514; 4,840,992; and 4,921,908 (Ofstead) disclose these copolymers and shaped articles, with U.S. Pat. No. 4,693,939 disclosing these copolymers as coatings on articles.

Non-crosslinked crystallized poly(vinyl alcohol) coatings have been disclosed for use with a variety of medical devices. European Patent Publication 0 370 657 (Ofstead) discloses a poly(vinyl alcohol) coating on medical devices (such as catheter guidewires), which is prepared by coating atactic poly(vinyl alcohol) on the device and then annealing the coating to crystallize the poly(vinyl alcohol) to provide a slippery surface.

However, the art of preparing crystallized poly(vinyl alcohol) hydrogel coatings has failed to recognize that in many instances it is desirable to retain the particular geometric configuration of the article being coated. Crystallized poly(vinyl alcohol) which is capable of becoming a hydrogel in the presence of water can disrupt a complex geometric configuration of a supporting structure, such as by blocking the pores of a microporous membrane, if the coating applied to the supporting structure is not carefully controlled.

Electroplating Devices

Electroplating devices require residue barriers to contain particulates generated during the electroplating process and other debris in order to maintain the quality of the electrodeposit of metal on a substrate. A residue barrier must not interfere with the transfer of metallic ions or otherwise provide any resistance to the plating current flow.

Electroplating solutions are generally aqueous solutions. A residue barrier must be hydrophilic and porous in order for the metallic ions in an agitated solution to pass through the barrier while the debris is blocked from passage. Hydrophobic microporous membranes would be excellent residue barriers if only such membranes were hydrophilic.

Electrochemical Cell Separators

Microporous membrane materials are frequently utilized as separator materials for electrochemical cells wherein they provide a physical barrier between the cell electrodes, keeping plates of opposite polarity from coming into direct contact with each other. In addition to having dimensional stability sufficient to maintain a physical barrier, the membrane material must also be non-conductive. The membrane material also must be porous so that cell electrolyte can pass through the separator to provide an internal conducting path between electrodes. Notwithstanding the need for porosity, the membrane material must minimize penetration through the separator of particulate matter either arising from flaking or colloidal dispersion of electrode materials or arising from dendrite formation during charging. The membrane material must also be chemically inert to the environment established by the cell.

Preferably, the separator is in the form of a microporous membrane having a high void volume which permits substantially unimpeded transport of electrolyte through the separator while exhibiting good dendristatic properties.

The present invention describes a supporting structure having a complex geometric configuration and an extremely thin hydrophilic polymer shell which imparts hydrophilicity to the structure. The present invention also describes a method of providing such hydrophilicity while substantially retaining the geometric configuration of the structure.

The present invention also provides a hydrophilic membrane for electroplating residue barriers which traps debris but does not substantially impair electroplating current. The residue barrier of the present invention, either surrounding an anode or positioned between an anode and an cathode of an electroplating device, does not appreciably interrupt the current flow or metallic ion transport vital to an electroplating process while blocking residue and debris harmful to the electroplating process.

The present invention also provides a hydrophilic microporous electrochemical cell separator material as a residue barrier for particulates and debris yet permits substantially unimpeded transport of electrolyte through the separator.

The invention overcomes the deficiencies in the prior art by providing a supporting structure having a complex geometric configuration enveloped at at least a portion of its surfaces by an extremely thin, self-interlocking shell of tactic, hydrophilic homopolymer or copolymer of poly(vinyl alcohol) while substantially retaining the complex geometric configuration of the supporting structure. The tactic poly(vinyl alcohol) shell may be either syndiotactic or isotactic.

As used herein, "complex geometric configuration" refers to the multiplicity and types of surfaces of a supporting structure when observed on a micron scale. The extremely thin shell of poly(vinyl alcohol) on a supporting structure envelops the multiplicity of such surfaces without altering the type of such surfaces. Thus, the poly(vinyl alcohol) shell imparts hydrophilicity to a supporting structure while substantially retaining the complex geometric configuration of the supporting structure.

The multiplicity of surfaces of a supporting structure are enveloped by the poly(vinyl alcohol) shell. "Envelop" means the shell entirely surrounds each of the multiple surfaces and imparts hydrophilicity thereto. Integrity is imparted by the formation of crystalline crosslinks within the shell, i.e., the formation of tie molecules connecting two or more crystallites. Thus, the shell is self-interlocking mechanically about the surfaces of the supporting structure without substantial covalent, ionic, or van der Waals interaction with such surfaces.

The type of surfaces that a supporting structure may have may be expressed in terms of Euclidean geometry, fractal geometry, or a combination of both.

A fractal is an object or process that cannot be represented by Euclidean geometry. With the complexity of natural shapes and surfaces being so jagged that they have more than two dimensions, fractal geometry has become useful to analyze shapes so commonly found in nature. (Van Nostrand's Scientific Encyclopedia Seventh Edition, Van Nostrand Reinhold 1989, p. 1221.)

Euclidean surfaces may be planar, curved, or any other topography which may exhibit a Euclidean geometric configuration.

Fractal surfaces may be porous, tentacular, jagged, uneven, undulating, irregular, asymmetrical, or of any other topography which may exhibit a non-Euclidean geometric configuration.

For example, a porous membrane or bead may appear to have surfaces which are planar or spherical, respectively, i.e., in a Euclidean geometric configuration. But at a micron scale, the membrane and bead have a complex geometric configuration, because a precise examination of the multiplicity of surfaces shows a fractal, three dimensional terrain which defies Euclidean characterization. The pores of the membrane or bead are uneven, irregular, and unpatterned in all of the three dimensions Euclidean geometry measures. The fractal surfaces surrounding such pores generate a complex geometric configuration for the supporting fractal structure. The poly(vinyl alcohol) shell of the present invention envelops such fractal surfaces defining such pores but does not cover or fill such pores or otherwise convert the fractal configuration of the surfaces to a Euclidean configuration.

In another example, a non-woven web may appear to be flat and have a Euclidean geometric configuration. But at a micron scale, surfaces of the web are an unpatterned layering of strands which give the non-woven web a complex geometric configuration, even if the individual strands comprising the web have a Euclidean geometric configuration. The poly(vinyl alcohol) shell of the present invention envelops the strands of the web while substantially retaining the complexity of the surfaces and the fractal configuration of the non woven web.

The hydrophilic, polymeric shell enveloping the supporting structure is "extremely thin", on a scale of monolayers of polymer. "Extremely thin" means that the shell's monolayer dimension is such that it does not substantially clog, smooth, block, or swell in manner to appreciably alter a supporting structure's complex geometry. Unlike a hydrogel coating, which, upon exposure to water would swell and significantly alter a geometric configuration of a supporting structure, the self-interlocking shell of poly(vinyl alcohol) of the articles of the present invention does not appreciably swell, substantially retaining the complex geometry of the article.

A supporting structure such as a membrane may have a Bubble Point Pore Size (c.f. ASTM F-316) of about 0.01 to 20 .mu.m. The present invention finds that an extremely thin shell of hydrophilic polymer having less than about an average of 100 Angstroms thickness forming a shell on fractal surfaces of the membrane reduces the effective pore size less than about 30 percent and desirably less than about 15 percent. The complex geometric configuration of the supporting structure is substantially retained.

Thus, the present invention allows the supporting structure to acquire a hydrophilic surface without altering its physical configuration.

The supporting structure has at least one surface in a complex geometric configuration which the poly(vinyl alcohol) shell may envelop. Nonlimiting examples of a supporting structure include films, porous membranes, beads, woven and non-woven webs, spun threads, hollow porous fibers, and porous fibers. Nonlimiting examples of the composition of the supporting structure may be polymeric, ceramic, cellulosic, glassy, metallic, or carbonaceous.

A polymeric structure may be made from any useful and formable polymeric material which does not dissolve substantially in the presence of solvents used with precursors to make the shell. Nonlimiting examples include without limitation, polyolefins (e.g., polyethylene and polypropylene), polyhalo olefins (e.g., polytetrafluoroethylene and polyvinylidene fluoride), nylon, polyesters (e.g., polyethylene terephthalate), polysulfones, polyethersulfones, poly(2,6-dimethyl,4-phenylene oxide) and derivatives thereof, polyamides, polyimides, polyetherimides, or polymeric materials previously unavailable for forming hydrophilic polymeric structures.

The tactic, hydrophilic shell of a homopolymer or copolymer of poly(vinyl alcohol) is formed in-situ about complex, envelopable surfaces of the supporting structure by hydrolysis (e.g., alcoholysis or ammonolysis) of a tactic hydrophobic polymeric poly(vinyl alcohol) precursor with a hydrolysis reagent.

"Hydrolysis" means the cleaving of an ester or ether group in the presence of a hydrolysis reagent to form an alcohol group.

The hydrophobic polymeric poly(vinyl alcohol) precursor can be any tactic poly(vinyl alcohol) precursor which forms a tactic homopolymer or copolymer of poly(vinyl alcohol), including without limitation, homopolymers of vinyltrifluoroacetate and copolymers of vinyltrifluoroacetate monomers and monomers having a vinylic group therein, homopolymers of vinyl tert-butyl ether monomers, and copolymers of vinyl tert-butyl ether monomers and monomers having a vinylic group therein. For purposes of describing this invention, references to poly(vinyl alcohol) shall include both a homopolymer of poly(vinyl alcohol) and a copolymer of vinyl alcohol and another co-monomer.

The invention also overcomes problems confronted in the prior art by providing a method for generating an extremely thin shell of tactic poly(vinyl alcohol) about envelopable surface(s) of a supporting structure. The method employs applying a tactic, polymeric poly(vinyl alcohol) precursor to surfaces of the supporting structure, and then causing, in-situ, a hydrolysis reaction to form a tactic, hydrophilic poly(vinyl alcohol) shell enveloping such surfaces while retaining the complex geometric configuration of the supporting structure.

The hydrolysis reagent may be a reagent which causes the formation of tactic poly(vinyl alcohol), whether such reaction occurs in liquid or gaseous phase. Preferably, the hydrolysis reagent is a basic reagent having a pH greater than about 7.0. Suitable reagents include, but are not limited to, dissolved or anhydrous ammonia, sodium hydroxide, sodium carbonate, and potassium hydroxide.

Many desirable articles having a hydrophilic polymeric shell thereon may be made and used in accordance with the present invention. The article may take the form of porous membranes where fractal surfaces define pores and interstices in and through the membrane. The shell of poly(vinyl alcohol) does not substantially alter the complex geometric configuration of the membrane. Such membranes may be used to separate particles in a flowing medium.

Many desirable electroplating residue barriers may be made and used in accordance with the present invention. Particularly, a hydrophilic, microporous membrane of the present invention may be used as an anode bag surrounding the anode of an electroplating device. Alternatively, a hydrophilic, porous membrane of the present invention may be used as an electroplating diaphragm separating the cathode and anode compartments with positive pressure electrolyte solution flow through the compartments. Preferably, the diaphragm membrane may be microporous to restrict the movement of micron-sized particles regardless of electrolyte solution flow characteristics.

The present invention further provides a hydrophilic microporous membrane material suitable for use as a separator for an electrochemical cell. The hydrophilic membrane material is substantially non-conductive, chemically inert to the electrochemical cell environment, and provides a residue barrier while having sufficient porosity to allow the cell's electrolyte to pass substantially unimpeded through the membrane. An electrochemical cell separator residue barrier, positioned between an anode and a cathode of an electrochemical cell minimizes movement of electrode debris originating from flaking, colloidal dispersion, or dendrite formation.

The present invention provides a hydrophilic polymeric self-interlocking shell about surfaces of a supporting structure while substantially retaining the complex geometric configuration of the structure, and to permit that hydrophilized article to be used in aqueous systems or with organic solvents without adversely affecting the hydrophilic polymeric shell.

The present invention also provides a method for forming an extremely thin shell of tactic, hydrophilic poly(vinyl alcohol) about a supporting structure, such as a microporous membrane, through the use of a tactic poly(vinyl alcohol) precursor capable of being converted, in-situ, on at least a portion of the complex surfaces of the supporting structure to tactic poly(vinyl alcohol) while substantially retaining the complex geometric configuration of the structure.

A feature of the invention is that the hydrophilic polymeric shell may be prepared using readily available materials reacted at minimally elevated temperatures and pressures.

It is another feature of the invention that the extremely thin hydrophilic polymeric shell of tactic hydrophilic poly(vinyl alcohol) envelops all outer surfaces of the supporting structure and any available interior surfaces without blocking or clogging such pores or interstices or otherwise substantially altering the complex geometric configuration of the supporting structure.

It is another feature of the invention to provide a tactic hydrophilic poly(vinyl alcohol) shell on a supporting structure which has hydroxyl reactive sites available for further reaction.

It is an advantage of the invention that articles produced according to the present invention have a surface shell which is permanent in the presence of aqueous systems or organic solvents, including those employed during use of a hydrophilic article.

It is another advantage of the invention that the tactic, hydrophilic poly(vinyl alcohol) shell provides increased mechanical strength to the polymeric structure, thereby enhancing the stability and sturdiness of an otherwise delicate film, membrane, web, or other structure while substantially retaining the physical configuration of that structure.

For a greater appreciation of embodiments of the invention, a detailed description follows with reference to the drawing.

FIG. 1a is a scanning electron photomicrograph of the outer surfaces of a supporting structure made according to U.S. Pat. No. 4,539,256 without a tactic, hydrophilic poly(vinyl alcohol) shell thereon.

FIG. 1b is a scanning electron photomicrograph of the outer surfaces of a supporting structure with a tactic, hydrophilic poly(vinyl alcohol) shell thereon made according to the present invention.

FIG. 2a is a scanning electron photomicrograph of the outer and cross-sectional surfaces of a polymeric structure made according to U.S. Pat. No. 4,539,256 without a tactic, hydrophilic poly(vinyl alcohol) shell thereon.

FIG. 2b is a scanning electron photomicrograph of the outer and cross-sectional surfaces of a polymeric structure with a tactic, hydrophilic poly(vinyl alcohol) shell thereon made according to the present invention.

FIG. 3a is an illustration of a membrane microstructure which is enveloped by a poly(vinyl alcohol) shell.

FIG. 3b is an exploded view of another illustration of the enveloped polymeric structure of FIG. 3a.

FIG. 4 illustrates a typical electroplating device having an anode bag and a diaphragm of the present invention.

FIG. 5 illustrates a typical electrochemical cell having at least one cathode, at least one anode, electrolyte and at least one separator of the present invention.

The supporting structure may be composed of any individual or combination of compositions of polymeric, ceramic, cellulosic, glassy, metallic, or carbonaceous materials. These materials may be either hydrophobic or hydrophilic in nature.

The supporting structure has a complex geometric configuration at a micron scale and may be formed according to known techniques into membranes, films, woven and non-woven webs, beads, spun threads, porous fibers, porous hollow fibers, or any other three dimensional configuration having a topography which permits the poly(vinyl alcohol) shell to envelop surface(s) of the structure in a self-interlocking fashion.

Non-limiting examples of the types of surfaces which can be enveloped include reticulated porous microstructures and tentacular outer surfaces of a structure.

Desirably, surfaces of any of these supporting structures provide a greater surface area per unit mass than that apparent from the gross Euclidean dimensions of the supporting structure. Many uses of articles are dependent on providing a large surface area per unit mass. The hydrophilic polymeric shell of the present invention imparts hydrophilicity without reducing substantially the surface area of the supporting structure.

Polymeric structures are preferred supporting structures. Of polymeric structures, porous membranes are preferred. More preferably, these porous membranes are microporous.

The effective pore sizes of the structure may be at least several times the mean free path of flowing molecules, e.g., from about a nanometer to about several micrometers. The porous membrane has a reticulated surface structure throughout its mass, which provides surface(s) for enveloping the complex geometric configuration of the membrane with a tactic, hydrophilic, poly(vinyl alcohol) shell.

The polymeric structure may be made from any polymeric material which may be formed into a desired complex geometric configuration.

Non-limiting examples of the polymeric materials used to make polymeric structures are: polysulfones, polyethersulfones, poly(2,6-dimethyl-4-phenylene oxide) and derivatives thereof, polyamides, polyimides, polyetherimides, polyolefins, polyhalo olefins (especially polytetrafluoroethylene), polyesters, nylon, and the like.

Non-limiting examples of suitable polyolefins include (regardless of molecular weight) polyethylene, polypropylene, poly-3-methyl-1-butene, poly-4-methyl-1-pentene, copolymers of ethylene, propylene, 3-methyl-1-butene, or 4-methyl-1-pentene with each other or with minor amounts of other olefins, e.g., copolymers of ethylene and propylene, copolymers of a major amount of 3-methyl-1-butene, and a minor amount of a straight chain n-alkene having from 2 to 18 carbon atoms such as 1-octene, 1-hexadecene, and octadecene or other relatively long chain alkenes, as well as copolymers of 3-methyl-1-pentene, and any of the same alkenes mentioned previously in connection with 3-methyl-1-butene.

A polyolefinic material may also include small amounts of other materials which may be copolymerized or blended therewith, but which do not substantially adversely affect the characteristics of the polyolefinic material.

The material comprising the polymeric structure should have a weight average molecular weight greater than about 1000, and preferably greater than about 50,000, a melt index less than about 1200 grams/10 minutes and preferably less than about 10 grams/10 minutes as measured according to ASTM D 1238-82.

When the polymeric structure takes the form of a porous or microporous membrane or other porous configuration, the polymeric structure should have a porosity of from about 15 percent to about 99 percent, and preferably from about 30 percent to about 95 percent. The porosity measurements are made according to ASTM D-792.

When the polymeric structure takes the form of a membrane or other porous configuration, the structure should have an effective pore size in micrometers, measured according to ASTM F-316, of from about 0.01 .mu.m to about 20 .mu.m, and preferably from about 0.1 .mu.m to about 1.2 .mu.m.

The tactic, hydrophilic poly(vinyl alcohol) is prepared by the reaction of a tactic, polymeric poly(vinyl alcohol) precursor with a hydrolysis reagent. The tacticity of the poly(vinyl alcohol) ranges from about 50 percent tactic triads to about 80 percent tactic triads using Fluorine NMR spectroscopy methods. Pritchard et al., "Fluorine NMR Spectra of Poly(vinyl Trifluoroacetate)" J. Poly. Sci. 4, 707-712 (1966), incorporated by reference herein, discloses calculation of triad tacticities for poly(vinyl alcohol) prepared by various methods.

The extremely thin shell of poly(vinyl alcohol) enveloping surface(s) of a supporting structure is described in terms of monolayers of poly(vinyl alcohol) on complex surfaces of a supporting structure. A "monolayer" is the thickness of the smallest dimension of a crystalline unit cell of poly(vinyl alcohol), about 2.53 Angstroms. The poly(vinyl alcohol) shell may comprise greater than an average of 10 monolayers to impart hydrophilicity to the complex geometric and often hydrophobic surfaces of a supporting structure.

The "extremely thin" self-interlocking shell of poly(vinyl alcohol) imparting hydrophilicity does not appreciably swell upon exposure to water to substantially alter the complex geometric configuration of a supporting structure.

Reference to an "average of" a number of monolayers compensates for the fact that these extremely thin shells are not of exact uniform thickness throughout the entire complex geometric configuration of the supporting structure.

If the supporting structure is porous and it is desired not to block or clog such pores of a nominal 2 micron pore size, the poly(vinyl alcohol) shell may comprise from about an average of about 10 to about 4,000 monolayers. Desirably, the poly(vinyl alcohol) shell may comprise from about an average of 10 to about an average of 400 monolayers. It is presently preferred that the poly(vinyl alcohol) may comprise from about an average of 10 to about an average of 40 monolayers.

Based on the dimensions of pore size of a porous supporting structure and the monolayers of poly(vinyl alcohol) enveloping surfaces of that supporting structure, it is desirable to have a shell of tactic poly(vinyl alcohol) occupy less than 30 percent of the pore size existing in the supporting structure prior to forming such poly(vinyl alcohol) shell. Preferably, the tactic poly(vinyl alcohol) occupies less than 15 percent of the original pore size.

The tactic, hydrophilic poly(vinyl alcohol) shell is relatively insoluble in water or highly polar organic solvents, or nonpolar organic solvents. Such organic solvents include without limitation, dimethylsulfoxide, glycerol, ethylene glycol, and other solvents having a solubility parameter differential from poly(vinyl alcohol) of greater than about 0.4 and desirably greater than about 0.6. The solubility parameter .delta. (H) for poly(vinyl alcohol) is about 12.6. A tactic poly(vinyl alcohol) shell of the present invention resists some washings by solvents having solubility parameters of less than about 12.2 or greater than about 13.0. Moreover, the tactic poly(vinyl alcohol) shell resists repeated washings by solvents having solubility parameters of less than about 12.0 or greater than about 13.2. Solubility parameters for solvents may be found in the Handbook of Chemistry and Physics, 60th Edition, Chemical Rubber Company. The tactic, hydrophilic poly(vinyl alcohol) shell of the present invention resists washout by any of the above-named solvents.

Desirably, the initial solubility of the hydrophilic poly(vinyl alcohol) shell when exposed to the above-named solvents is less than about 1 part per 100 parts of solvent at room temperatures and pressures with no measurable solubilization thereafter. Such relative insolubility of the poly(vinyl alcohol) shell in water and polar and nonpolar organic solvents provides continuing hydrophilicity of the article during usage in the presence of such solvents.

The poly(vinyl alcohol) precursor may be a tactic homopolymer of vinyltrifluoroacetate, a tactic copolymer of vinyltrifluoroacetate monomer and monomer(s) having a vinylic group therein, a tactic homopolymer of vinyl tert-butyl ether, or a tactic copolymer of vinyl tert-butyl ether monomer and monomer(s) having a vinylic group therein.

The weight average molecular weight of the homopolymer or copolymers of poly(vinyltrifluoroacetate) range from about 50,000 to about 2,000,000 and desirably range from about 500,000 to about 1,000,000. Desirably, the syndiotactic homopolymer or syndiotactic copolymer of poly(vinyltrifluoroacetate) is unbranched.

The weight average molecular weight of the homopolymer or copolymers of poly(vinyl tert-butyl ether) range from about 25,000 to about 60,000 and desirably from about 35,000 to about 45,000.

Non-limiting examples of monomers having a vinylic group therein, useful for copolymerization to form the precursor, include vinyl esters having up to six carbon atoms, vinyl ethers having up to eight carbon atoms, and disubstituted ethylenes (such as esters or anhydrides of lower alkyl (C.sub.1 -C.sub.4) substituted or unsubstituted dicarboxylic acids having up to eight carbon atoms). Of these possible monomers, maleic anhydride and vinyl acetate are preferred.

The presently preferred precursor is syndiotactic poly(vinyl trifluoroacetate) homopolymer.

The precursor typically is hydrophobic and is applied in relatively dilute solution. The solvent may be any liquid that wets the surfaces of the supporting structure and solubilizes the precursor.

A solvent which allows spontaneous wetting of the precursor solution on all available surfaces of a hydrophobic supporting structure is preferred. The term "all available surfaces" includes without limitation, the reticulated pores and interstices of a porous hydrophobic article or the tentacular surface of a film, bead or web. Spontaneous wetting provides rapid, even envelopment of all of the available internal and external surfaces of the article, and an application of at least an average of 10 monolayers thickness of the precursor on such surfaces for further processing. It is possible for a porous supporting structure to have some pores having radii smaller than the hydrodynamic radius of the precursor. The surfaces along such smaller pores may not be available for application of the precursor, because the precursor molecule is too big to enter the pore.

The concentration of the precursor in solution determines the ability of the precursor to cover all available surfaces of the supporting structure while substantially retaining the complex geometric configuration of the supporting structure. The concentration of the precursor in the solvent may range from about 0.5 percent (w/v) to about 15 percent (w/v). Desirably, the concentration ranges from about 2 percent (w/v) to about 10 percent (w/v). Preferably, the concentration ranges from about 3 percent (w/v) to about 8 percent (w/v).

The solvent is desirably organic and has a significant vapor pressure at a temperature of less than about 38.degree. C.

When the supporting structure is hydrophobic, the solvent may be any liquid which solubilizes the precursor and wets the supporting structures' surfaces. Non-limiting examples include: ketones, esters, ethers, nitriles, or amides having aliphatic, allcyclic, or aromatic groups. Of these solvents, acetone, ethyl acetate, cyclohexanone, tetrahydrofuran, pyridine, acetophenone, and acetonitrile are desired. Of these solvents, acetone is preferred due to its availability, cost, and handling.

The hydrophobic polymeric poly(vinyl alcohol) precursor applied to the surfaces of the supporting structure is converted, in-situ, to tactic poly(vinyl alcohol) by a hydrolysis reagent which is capable of converting the pendant trifluoroacetate groups of the precursor into hydroxyl groups. The hydrolysis reagent may be applied in either a liquid or a gaseous state. The hydrolysis reagent may be acidic or basic, but desirably it is basic. Thus, a desirable hydrolysis reagent has a pH of greater than about 7.0 and desirably from about 8 to about 10.

Non-limiting examples of a hydrolysis reagent include sodium hydroxide in methanol, sodium carbonate in a methanol/water solution, ammonium hydroxide in methanol, potassium hydroxide in a methanol/water mixture, and aqueous or vaporous ammonia. Of these reagents, ammonia is preferred in the vapor phase or in a methanol/water mixture.

When vaporous ammonia is used, it is presently preferred to hydrate the surfaces of the supporting structure and the shell of tactic poly(vinyl alcohol) with water or moisture vapor to stabilize hydrophilicity of the shell. Otherwise, it is possible for the extremely thin shell of poly(vinyl alcohol) to conformationally rearrange, causing loss of some or a substantial portion of hydrophilicity, if such article is not placed into use in aqueous-based solvents within weeks after manufacture of the hydrophilic article.

The amount of contact between the hydrolysis reagent and the precursor should be sufficient in duration and in concentration to permit complete conversion of the tactic precursor to tactic poly(vinyl alcohol). Desirably, the polymeric supporting structure having the precursor applied to its surfaces is immersed in a solution containing the hydrolysis reagent having a pH of greater than 7.0.

The manufacture of an article having a hydrophilic polymeric shell varies according to its composition and its ultimate shape.

The supporting structure may be formed from commercially available materials depending on form and composition desired by those skilled in the art.

Raw materials suitable as base materials for supporting structures are commercially available. For example, polymeric supporting structures may be prepared from commercially available resins using a variety of extrusion, membrane preparation, or film-forming techniques well known in the art. A preferred method of membrane preparation is disclosed in U.S. Pat. No. 4,539,256, the disclosure of which is incorporated by reference herein.

Membranes of polysulfone are commercially available from Schleider and Schuell of Keene, N.H. Polyolefinic microporous membranes are commercially available from Hoeschst-Celanese of Charlotte, N.C. and references to the methods of manufacture of such polyolefinic microporous membranes may be found in U.S. Pat. Nos. 4,501,793 and 3,853,601, both of which are incorporated by reference herein.

For the poly(vinyl alcohol) precursor, a poly(vinyl trifluoroacetate) homopolymer may be made according to U.S. Pat. No. 2,436,144, incorporated by reference herein. A poly(vinyl trifluoroacetate) copolymer may be made according to U.S. Pat. No. 2,436,144 or according to co-assigned U.S. Pat. Nos. 4,528,325, and 4,618,649, both of which are incorporated by reference herein.

Vinyltrifluoroacetate and comonomers for synthesis of poly(vinyltrifluoroacetate) are commercially available from Polysciences of Warrington, Pa. and Aldrich Chemical of Milwaukee, Wis.

A solution of tactic, poly(vinyl alcohol) precursor in a solvent which wets the supporting structure is then applied onto all available surfaces of the supporting structure, saturating the complex surfaces. Upon evaporation of the solvent, a self-interlocking shell is formed which substantially retains the complex geometric configuration of the structure.

Depending upon the configuration of the supporting structure and its composition, the method of application of precursor solution may involve wiping, dipping, rolling, knifing or extruding steps as the case permits. The solvent may be removed by drying the polymeric shell for such times and at such temperatures and pressures to fully dry the precursor. Processing conditions may be controlled as necessary to permit drying of the precursor on surfaces without covering or clogging available porous surfaces of the supporting structure. The application of precursor may occur batch-wise or continuously according to the manufacturing processing conditions preferred.

For example, to prepare an unclogged porous membrane, the in-situ conversion of tactic precursor to tactic, hydrophilic poly(vinyl alcohol) occurs by hydrolysis at less than about 38.degree. C. using a hydrolysis reagent in either liquid or vaporous phase. A closed reaction vessel for vaporous reaction is preferred. A dipping tank for liquid reaction is preferred.

When ammonia vapor is used, after a closed vessel is employed, the membrane is dipped or sprayed with water or moisture vapor to lock in hydrophilicity of the poly(vinyl alcohol) shell.

While it is preferable to provide hydrophilic surfaces for the supporting structure, it can be desirable in certain articles to be able to reverse hydrophilicity into hydrophobicity after a process step or other intermediate activity. Hydrophilic shells of poly(vinyl alcohol) of the present invention can be conformationally rearranged into hydrophobic surfaces by heating the hydrophilic supporting structure above the glass transition temperature (Tg) of poly(vinyl alcohol), about 80 degrees C., for a limited period of time. Without being limited to a particular theory, it is believed that the increased mobility of poly(vinyl alcohol), above its Tg, allows a conformational rearrangement of hydroxyl groups at outer surfaces of the shell to point in towards the bulk of the shell. The driving force for this rearrangement is a natural desire to minimize interfacial energy (i.e., between the shell surface and air). This results in the outermost few Angstroms of the shell surface being defined by the hydrocarbon backbone of the poly(vinyl alcohol). Since "wetting" and hydrophilicity is defined by the outermost few Angstroms of a surface, the presence of the hydrocarbon backbone in place of hydroxyl groups results in the surface no longer being hydrophilic. Thus, the shell has the same chemistry but is hydrophobic.

Hydrophobic poly(vinyl alcohol) can be converted to hydrophilic poly(vinyl alcohol) by wetting surfaces of the shell-covered supporting structure with a polar, water-miscible solvent, such as methanol or acetone, followed by solvent exchanging water into pores of the supporting structure and drying. In such re-hydrophilization, contact of a poly(vinyl alcohol) shell with such polar, water miscible solvent re-orients hydroxyl groups out from such surfaces. Water or moisture vapor plasticizes poly(vinyl alcohol), swells such shell, and lowers the glass transition temperature thereof in the presence of such a polar environment.

Thus, according to methods of the present invention, one can control hydrophilization of supporting structures, throughout all surfaces, only at outer surfaces, or only in pores or interstices. One can create regio-specific hydrophilic surfaces for a supporting structure according to need. Non-limiting examples of such regio-specific surfaces can be patterned hydrophilicity throughout specific portions of a porous membrane, "facade" hydrophilicity of a porous membrane, or a "sandwich" hydrophilicity having hydrophobic inner pores and interstices.

Alternatively, regio-specific hydrophilization can be achieved by introducing a hydrolysis reagent only to designated portions of surfaces of a supporting structure covered with polymeric poly(vinyl alcohol) precursor. Also, such regio-specific hydrophilization can be achieved by applying polymeric poly(vinyl alcohol) precursor to designated portions of surfaces of the supporting structure, followed by hydrolysis of such portions with hydrolysis reagent.

However, some uses of the article may prefer skinned, covered or clogged pores, to convert a fractal geometric configuration of the article to a Euclidean geometric configuration. In such circumstances, processing conditions or solutions may be adjusted as desired. Three parameters may be adjusted. Choice of solvent influences rate of coverage and evaporation. Precursor concentration determines solution viscosity, rate of pore penetration, and shell thickness. Pore sizes of surfaces of the article also determine rate of pore penetration.

FIG. 1 illustrates the comparison between a hydrophobic microporous membrane (prepared according to Example 23 of U.S. Pat. No. 4,539,256) and a microporous membrane having with tactic, hydrophilic poly(vinyl alcohol) shell (prepared according to the present invention). The outer surfaces of the article in scanning electron photomicrograph 1a are hydrophobic and untreated. The outer surfaces shown in scanning electron photomicrograph 1b are hydrophilic due to the tactic, hydrophilic poly(vinyl alcohol) shell about its surfaces. The treated membrane shown in scanning electron photomicrograph 1b retains its fractal geometric configuration because its complex surfaces are substantially as open and unclogged as the unprocessed membrane in photomicrograph 1a. Thus, the pores and interstices of supporting structure are not filled or occluded by the in-situ generated poly(vinyl alcohol) shell. The fractal configuration of the supporting structure is not converted to a Euclidean configuration. Thus, the membrane retains its structural advantages while adding hydrophilic surfaces.

FIG. 2 illustrates that a microporous membrane having a tactic, hydrophilic poly(vinyl alcohol) shell enveloping both outer and inner porous surfaces of the membrane does not clog or occlude any pores or interstices. Scanning electron photomicrograph 2a of the same hydrophobic membrane as in FIG. 1a and scanning electron photomicrograph 2b of the same hydrophilic membrane as in FIG. 1b both show about 10 microns of cross-section of the membrane at the bottom of the scanning electron photomicrographs, with the remainder being a perspective view of the outer surface. No significant difference can be seen between these two photos.

While not being limited to any particular theory, it is believed that the shell of tactic poly(vinyl alcohol) envelops available surfaces of the supporting structure by forming tie molecules among crystallite molecules. FIG. 3a illustrates a membrane microstructure which is enveloped by a poly(vinyl alcohol) shell. The exploded view of FIG. 3b provides another illustration of the enveloped polymeric structure. Tie molecules of poly(vinyl alcohol), such as those described in Basset, D. C., Principles of Polymer Morphology, Cambridge Univ. Press, 1981, between crystallites of poly(vinyl alcohol), provide the self-interlocking strength of the shell. The complex geometric configuration of the underlying polymeric structure is substantially retained after envelopment of from about an average of 10 to about an average of 4000 monolayers of poly(vinyl alcohol) and desirably from about an average of 10 to about an average of 400 monolayers of poly(vinyl alcohol).

The hydrophilic supporting structures of the present invention can be utilized in several applications involving aqueous fluids or hydrophilic organic solvents. The chemical inertness and complex geometric configuration of many hydrophobic materials or structurally weak hydrophilic materials would make them ideally suited for hydrophilic processes if the supporting structure had hydrophilic surfaces of a self-interlocking shell.

Having a tactic, hydrophilic poly(vinyl alcohol) shell enveloping surfaces of such hydrophobic supporting structures enables such structures to be used in aqueous systems or in hydrophilic organic solvents in which the untreated supporting structure above would be inadequate, incompatible, or ineffective, notwithstanding its complex geometric configuration desired for such mechanical processes. The relative insolubility of the tactic, hydrophilic poly(vinyl alcohol) shell in a large number of organic solvents and water enables the article to be used in those circumstances where the solvent must wet the article in order for the article to perform its intended purpose. A useful measure of "wetting" capability is pore wetting surface energy.

"Pore wetting surface energy" means the surface energy of the supporting structure required for spontaneous wetting of a pore through the wicking of water into the pore via capillary forces. Spontaneous wetting of the pore occurs when the surface energy of the internal surface of the pores is high enough for water to have less than a 90.degree. contact angle with the surface. Analytically, according to Wu, S., Polymer Interface and Adhesion, Marcel Dekker, New York, 1982, p. 244, spontaneous wetting occurs when the capillary force, .DELTA.P.sub.c, in the following equation is positive:

where .sigma..sub.L is liquid surface tension (72.8 dynes/cm for water), .THETA. is the contact angle (<90.degree.), and r is the pore radius.

The magnitude of the positive capillary force correlates to the rate of spontaneous wetting. The variation in pore size and complex geometric configuration also assists in controlling the rate of migration.

The tactic, hydrophilic poly(vinyl alcohol) shell enveloping the supporting structure also provides the advantage of increasing the mechanical strength of a supporting structure. By enveloping internal and external surfaces of a membrane while substantially retaining the complex geometric configuration of the membrane (e.g., a microporous membrane having pore sizes of from about 0.01 .mu.m to about 1.2 .mu.m,) the tactic poly(vinyl alcohol) shell increases the tensile strength and percent elongation properties of the supporting structure. While not limited by any particular theory, it is believed that the enhanced tensile strength is achieved by the covering of acute geometric interstices of the fibrillar structure which would otherwise be probable stress concentration points for failure initiation in the complex geometric configuration. While acute geometric intensities may be lessened, the overall complex geometric configuration is substantially retained.

The presence of a tactic, hydrophilic poly(vinyl alcohol) shell about a supporting structure provides highly reactive hydroxyl sites for further chemical, physical, and biological uses.

The present invention has broad utility in that articles having durable hydrophilic polymeric shells can be prepared from a wide variety of supporting structure materials which can comprise any of several compositions and take any of several forms. The present invention's ability to provide articles having a non-crosslinked self-interlocking shell comprising hydroxyl functionality that displays minimal solubility in water and aqueously soluble organic solvents provides advantages not previously found in the art.

Polyolefinic membranes, especially microporous polyolefinic membranes, could have considerable usage in conjunction with electroplating devices if the membrane were hydrophilic. A microporous polyolefinic membrane has a complex geometric configuration because of a multiplicity of surfaces defining pores and interstices. The tactic, hydrophilic poly(vinyl alcohol) shell on all available surfaces of a microporous polyolefinic membrane converts the hydrophobicity of the polyolefinic membrane to hydrophilicity without substantially altering the complex geometric configuration of the membrane.

A residue barrier in an electroplating device may take the form of an anode bag to trap debris. Debris may be additives used in forming the anode and, in extremely acidic baths, bubbles of gaseous oxygen. When used as an anode bag, the membrane should be microporous.

A residue barrier may also be a diaphragm separating an anode from a cathode in the device to trap particulates and debris from a cathodized substrate being electroplated in the agitated and heated electrolytic solution. When used as a diaphragm, if there is positive pressure of electrolyte solution flow, the diaphragm may be porous. Otherwise, the diaphragm should be microporous.

A residue barrier can also be used in an electrochemical apparatus for the recovery of magnetic metals, such as nickel and iron, from synthetic diamond bearing materials dispersed in an electrolyte. In this use, it is particularly important to protect a cathode with a hydrophilic microporous membrane bag to prevent diamonds and unwanted debris being deposited with metal being recovered.

A residue barrier can also be used in an electrodeposition apparatus for plating batches of particles that either float in the electrolyte or sink, such as an electrode enclosure for use in barrel plating of small metal parts or surface conductive non-metallic parts.

Whether an anode bag, cathode bag, or a diaphragm, the residue barrier provides a physical barrier based on the complex geometric configuration of a membrane while relying on the hydrophilicity of the poly(vinyl alcohol) of the shell to maintain the electrochemical process of metallic ion transport to the substrate.

In FIG. 4, a typical electroplating device 10 is shown. The device 110 is a tank 112 in which a cathode 114, an anode 116, a heater 118, and electrolyte 120 reside. Between cathode 114 and anode 116 is a diaphragm 122 of the present invention. Surrounding anode 116 is an anode bag 124 of the present invention. Preferably, the anode bag 124 also has art overlay of coarse non-woven material (not shown) to further protect the structural integrity of the bag 124 in a turbulent tank 112. Preferably, the overlay is a non-woven polypropylene.

The cathode has the substrate 126 to be electroplated positioned adjacent the diaphragm 122 and the anode 116. The cathode is often rotated on shaft 128 by turning head motor 130 to vigorously agitate the electrolyte 120 in the tank 112 and improve uniformity of deposit thickness.

The anode 116 is typically a titanium basket holding chips of the metal to be electroplated. Often, the metal is nickel or copper. The chips are typically formulated with additives useful to improve anode chip erosion and to facilitate entry of the metal into solution.

An anode bag was previously often made of cotton cloth or polypropylene fabric surrounding a napped or flannel-like interior. Anode bags are provided to contain any additives and debris from entering the electrolytic solution. Because the additives are often insoluble, these residue particulates could mar the surfaces of the electroplated substrate 126 if not contained.

An anode bag previously had pore sizes of about 10 .mu.m or greater to attempt to contain anode additive residues. But the fabric did not block particulate residues smaller than 10 .mu.m in size and did not restrict electrolyte solution agitation near the anode. Moreover, the prior anode bags did unfortunately create about a 8-10% reduction in electroplating current flow. Efforts to use a tighter weave anode bag further inhibited electroplating current flow because the fabric became an electrical insulator.

An anode bag 124 of the present invention may have pore size ten times smaller than the conventional fabric without substantially impairing electroplating current flow. Thus, the complex geometric configuration of the membrane blocks particulates and restricts electrolyte solution agitation near the anode without disrupting the electroplating current flow of metallic ions to the cathode substrate 126 to be electroplated.

An anode bag 124 of the present invention is especially useful for acid copper electroplating which requires grain refiner additives to harden, level, and brighten the deposit. These additives added to the electrolyte 120 deteriorate or are consumed by contact with copper anodes and copper particulates in the electrolyte 120, even without agitation. Because the anode bag 124 of the present invention restricts the flow of electrolyte solution near a copper anode, additives to the electrolyte 120 to improve the substrate 126 are inhibited from deterioration by reaction with copper at the anode. The valuable additives may be better controlled in the tank 112.

The anode bag 124 also contains gaseous oxygen formed at or near the anode when the electrolyte is a nickel-phosphorus solution. The bubbles of oxygen are impeded from blending into the electrolyte 120 in the remainder of tank 112 and rise to the surface of the electrolyte 120 within anode bag 124.

The anode bag 124 of the present invention reduces the need for constant filtering of the electrolyte 120 to remove particulate residues because the anode bag 124 contains such residues more efficiently than currently performed.

However, filtering within a tank 112 remains desirable. Diaphragm 122 of the present invention serves to skim undesired particulate residues from the electrolyte 120 without substantially disrupting electroplating current flow. Diaphragm 122 is stretched over a supporting frame in tank 112 adjacent cathode 114. Floating debris is skimmed from the cathode compartment as the electrolyte 120 overflows the diaphragm 122 into the anode compartment.

In electrolyte containing nickel, cotton cloth has been conventionally used as a diaphragm. But the deficiencies of cotton are the same as for its use as an anode bag. Further, cotton is not durable in an acid copper or an acid nickel-phosphorus electrolyte system where the pH of the electrolyte 120 may be lower than 1.

A diaphragm 122 of the present invention shows no measurable reduction in electroplating current flow. In an acid copper plating electrolyte 120, a poly(vinyl alcohol) shell polyolefin membrane diaphragm 122 is more durable than cotton cloth.

While not being limited to any particular theory, it is believed that the electroplating current flow rate is not disrupted because the pore wetting surface energy of the membrane as diaphragm 122 or anode bag 124 is greater than the surface tension of the electrolyte 120.

Thus, flow of metallic ions is unimpeded while particulate residues and other debris are physically restrained at pores of the membrane. The hydrophilicity of the shell of the membrane retains current flow while the complex geometric configuration of the membrane blocks residue.

Polyolefin-based hydrophilic microporous membrane materials of the present invention are particularly desirable as separator materials for aqueous based electrochemical cells because they exhibit most of the performance properties that are desired for good electrochemical cell separation. More specifically, polyolefin-based hydrophilic membranes of the present invention are substantially non-conductive, chemically inert to the environment of the electrochemical cell, provide good barrier properties for electrochemical cell uses, and are hydrophilic so that they are spontaneously and rapidly wet by hydrophilic electrolyte compositions.

The polyolefin-based hydrophilic microporous membrane materials of the present invention also demonstrate excellent dimensional stability, showing less than about 5% change, and preferably less than about 3% change over a temperature range of from about ambient to about 120.degree. C. The polyolefin-based hydrophilic microporous membrane materials of the present invention exhibit minimal resistance to electrolyte flow through the membrane due to their high porosity level. This low resistivity is due, at least in part, to the fact that very high electrolyte levels are present in the separator at all times when in the electrochemical cell. The high porosity coupled with the hydrophilic properties of the poly(vinyl alcohol) shell on available surfaces contribute to the high electrolyte levels present in the separator. The hydrophilic microporous membrane materials for an electrochemical cell separator should have a porosity of at least 15%, desirably have a porosity of at least 30%, preferably have a porosity of at least 40%, and most preferably have a porosity of at least 50%. Porosity is measured by the test method described in ASTM D792-66, the disclosure of which is incorporated by reference.

The polyolefin-based hydrophilic microporous membrane materials of the present invention offer a very attractive balance between porosity level and barrier properties desirable for electrochemical cell separators. Hydrophilic membranes having maximum pore sizes from about 0.05 to about 1.2 .mu.m, and more preferably from about 0.09 to about 0.2 .mu.m in diameter provide excellent barrier properties. Separators having such pore sizes minimize passage of particulate matter and other debris through the separator. It is important that the separator material exhibit good barrier and dendristatic properties so that particles or dendrites cannot bridge between opposing electrodes which can cause an electrical short and eventual failure of the electrochemical cell. Polyolefin-based hydrophilic microporous membrane materials of the present invention may possess a machine direction tensile strength of at least about 100 kg/cm.sup.2, more preferably a tensile strength of at least about 200 kg/cm.sup.2, and most preferably a tensile strength of at least about 350 kg/cm.sup.2.

In FIG. 5, a typical electrochemical cell 210, such as a battery is shown. The device 210 is a container 212 in which at least one cathode 214, at least one anode 216, and electrolyte 220 reside. Between each cathode 214 and each anode 216 is a separator 222 of the present invention.

An electrochemical cell separator 222 of the present invention restricts the flow of particulate matter, dendrites, or other debris between cathodes 214 and anodes 216 and preserves the electrochemical efficiency of the energy storage and energy dispensing processes during the life of the cell 210. Such debris is less likely to interfere with the electrochemical activity at cathodes 214 and anodes 216.

While not being limited to any particular theory, it is believed that the electrochemical current flow rate is not significantly disrupted because the surface energy of the membrane as electrochemical cell separator 222 is greater than the surface tension of the electrolyte 220. Flow of ions is unimpeded while particulate matter and other debris are physically restrained at pores of the membrane.

The invention is not limited to the embodiments described here or by the examples which follow.

In the examples to follow, certain tests were conducted and are described below:

Gurley Value--This value is a measurement of time in seconds to pass 50 cm.sup.3 of air through a porous film according to ASTM D-726, Method A.

Bubble Point Pore Size--This is a measurement of the maximum effective pore size, in microns, according to ASTM F-316. This value is also referred to as "pore size" in the Examples.

Porosity--This is a measure of the void volume of the porous article, and is derived from the measurement of specific gravity of the article, according to ASTM D-792. The porosity is defined as: ##EQU1## Tensile Strength--Values measured according to ASTM D 638-80 using an Instron model 1122 tensile tester under the following conditions:

Jaw Gap: 5.08 cm

Jaw Speed: 50.8 cm/min

Sample Size: 2.54 cm wide strip

MD and TD values for the tensile strength refer to measurements made in the "machine direction" and the "transverse direction" respectively.

Water Permeability--Water permeability was determined by placing a 74 mm diameter piece of the membrane in a test cell, which used an o-ring to seal the membrane to a sintered stainless steel back-up plate. The cell was equipped with a 350 ml water reservoir and was pressured with compressed nitrogen. The water flow rate was calculated by measuring the volume of water passed through the sample in a given time, with a 10 psi (or 68,947 N/m.sup.2) head pressure. At least three measurements were averaged for each permeability value reported.

Preparation of the poly(vinyl alcohol) precursor, syndiotactic poly(vinyl trifluoroacetate), was performed in a one gallon glass bowl jacketed pressure reactor having a stainless steel lid fitted with a metal turbine agitator blade on a sealed shaft, two mixing vanes, a thermowell and at least two valved openings. The system was purged with a sweep of dried argon to remove moisture and oxygen before adding reactants or solvent. Materials were weighed and transferred in closed vessels under inert gas and anhydrous conditions. Charges were made through rubber septa covering the opened valves in the reactor lid using proper techniques to prevent uptake of atmospheric moisture and oxygen. Into the reactor were placed, in order, 3025 g of Freon 113, 17.5 ml of a premix containing 2.5 g of trifluoroacetic anhydride in 25 ml of Freon 113, 355 g vinyl trifluoroacetate monomer, 14 ml of a second premix containing 2.5 g of bis(4-t-butylcyclohexyl) peroxydicarbonate (commercially available as "Percadox" 16N from Akzo Chemie America, Noury Chemicals of Chicago, Ill.) in 25 ml of Freon 113. The reactor temperature was raised to 45.degree. C. and maintained at that temperature for about 18 hours with an agitator speed at about 1000 rpm. A slight exotherm was observed during the reaction with a maximum system pressure of about 10-12 psig (0.7-0.8 kg/cm.sup.2). The polymerized, syndiotactic poly(vinyl trifluoroacetate) (PVTFA) was isolated by filtration and dried at 40.degree. C. under vacuum overnight.

A microporous polyethylene (PE) membrane, made by thermally induced phase separation as disclosed in Example 23 of U.S. Pat. No. 4,539,256 (Shipman) the disclosure of which is incorporated herein by reference, having a maximum pore size of 0.5 micron, a porosity of 81.5 percent and a thickness of 0.074 mm, was saturation treated with a 4 percent (w/v) acetone solution of PVTFA using an extrusion die. The membrane was dried slowly for 1.6 minutes in a two zone air floatation oven with the two zones set at temperatures of 27.degree. C. and 38.degree. C. respectively, resulting in a 22.2 weight percent add on of the PVTFA shell formed on the external and internal pore surfaces. No substantial blocking of the pores occurred, nor was a PVTFA skin formed on the covered side as evidenced from scanning electron microscopy (SEM) analysis. The complex geometric configuration of the membrane was substantially retained. Bubble point measurements showed a reduction in the maximum pore size to 0.44 micron.

A piece of this dry membrane was placed in an ammonia-saturated glass vessel for 2 minutes in order to convert, in-situ, the PVTFA shell to a poly(vinyl alcohol) (PVA) shell. The ammonia atmosphere was generated by placing a concentrated ammonium hydroxide solution in the bottom of the vessel. A 68 weight percent reduction in the weight of the shell resulted from the hydrolysis reaction.

Fourier Transform Infrared, (FT-IR), spectroscopy (at 4 cm.sup.-1 resolution, 64 scans, between a range of 4000 cm.sup.-1 and 400 cm.sup.-1, through the membrane) confirmed that the 68 weight percent loss in shell weight, which occurred during this basic hydrolysis reaction step was due to the quantitative loss of the trifluoroacetate group from the PVTFA. This amount of weight loss corresponded exactly to the amount of weight loss expected for 100% conversion from PVTFA to PVA. Upon removing the membrane from the ammonia atmosphere, it displayed spontaneous and nearly instantaneous wetting with water. The complex geometric configuration of the membrane was substantially retained throughout the hydrolysis treatment as was evidenced by a pore size loss of less than eight percent.

The ability of the hydrophilic membrane to resist wash-out of the PVA shell by common organic solvents was demonstrated by soaking pieces of the membrane in large amounts of acetone, isopropyl alcohol, and 1,1,1-trichloroethane. After 45 minutes of soaking in each of these solvents, the re-dried membranes retained their hydrophilicity, as shown by their spontaneous and nearly instantaneous wetting with water. The ability of the hydrophilic membrane to resist wash-out of the PVA shell by water was demonstrated by passing 2000 ml of deionized water through a 36 cm.sup.2 piece of this membrane. After drying, the hydrophilicity of the membrane remained unchanged (i.e., it was spontaneously and nearly instantly wetted with water).

The porous properties of the starting PE microporous membrane, the PVTFA covered membrane, and the final PVA shell hydrophilic membrane are reported in Table 1 below.

A hydrophilic membrane was prepared according to the procedure of Example 1 except that the syndiotactic PVTFA solution was applied to the membrane using a #8 wire-wound bar to spread the PVTFA solution on the membrane. The sample was allowed to dry at room temperature in a ventilation hood to produce a PVTFA treated membrane without causing pore blockage or PVTFA skin formation as shown by SEM examination. The PVTFA treated membrane retained the complex geometric configuration of the starting membrane. The sample was reacted with vaporous ammonia as in Example 1 to yield a hydrophilic membrane, as shown by its spontaneous and nearly instantaneous wetting with water.

Table 2 below provides data on mechanical properties of unprocessed hydrophobic microporous membranes and the hydrophilic microporous membranes of the present invention, both before and after in-situ conversion to the corresponding PVA shell membrane. The base PE membrane was the same as that used in Example 1. The PVTFA treated membranes were prepared according to the procedure detailed in Example 2, using various concentrations of syndiotactic PVTFA solutions in acetone as noted in Table 2 below. Hydrolysis of the PVTFA treated membranes was performed in an ammonia atmosphere as in Example 1. The PVTFA add-on was determined by weight difference after a piece of the PVTFA treated membrane had been extracted with acetone to a constant weight. The weight percent of PVA was calculated from the weight percent PVTFA above, assuming 100 percent conversion to PVA. Tensile measurements were performed on 2.54 cm wide strips of membrane as described above. Tensile strength is defined as the Newtons/m.sup.2 at break normalized to the cross-sectional area.

A microporous PE membrane, prepared according to Example 23 of U.S. Pat. No. 4,539,256 (Shipman), having 0.26 micron pore size and a 77 percent porosity was treated with a 4 percent (w/v) solution of syndiotactic PVTFA (prepared according to Example 1) in cyclohexanone. The web was passed through an immersion trough containing the tactic PVTFA solution, which was heated to 46.degree. C. to decrease its viscosity, then passed through a rubber nip station to squeeze off excess solution, and dried in an air floatation oven at a temperature of 40.5.degree. C. to produce a PVTFA treated membrane. Control of the PVTFA add-on was more difficult using this method, and the membrane had a tendency to stretch as it passed through the nip roll station. After hydrolysis with ammonia vapor, as in Example 1, the membrane was hydrophilic as shown by its spontaneous and nearly instantaneous wetting with water.

A microporous polypropylene (PP) membrane, made according to the procedure of Example 9 of U.S. Pat. No. 4,726,989 (Mrozinski), the disclosure of which is incorporated herein by reference, was treated with a 4 percent (w/v) acetone solution of syndiotactic PVTFA following the procedure of Example 2. Upon hydrolysis with ammonia vapor as in Example 2, the membrane was hydrophilic, as demonstrated by its spontaneous and nearly instantaneous wetting with water. The porous properties of the starting PP membrane and the hydrophilic membrane of the present invention are shown in Table 3. Pore size loss was less than 11 percent, demonstrating a substantial retention of the physical structure of the membrane while imparting hydrophilicity to the membrane surfaces.

A 0.023 mm thick PP microporous membrane, prepared according to Example 9 of U.S. Pat. No. 4,726,989, having a 0.2 micron pore size, a 66.7 percent porosity, and a Gurley value of 25.6 sec, was treated with PVTFA according to the procedure of Example 2, using a 2 percent (w/v) solution of syndiotactic PVTFA in acetone followed by in-situ NH.sub.3 hydrolysis. The treatment/hydrolysis operation was repeated three times to prepare the hydrophilic membrane of the present invention. The resulting hydrophilic membrane was instantaneously wet with water. The ability of the PVA shell to resist wash-out by water was demonstrated by placing the membrane in boiling water for 5 hours, drying the membrane, and demonstrating that the membrane was still spontaneously and nearly instantaneously wet with water, even though there was a 1.8 percent reduction in the membrane's weight during the exposure to boiling water.

The syndiotactic PVTFA treated membrane of Example 1 was placed in a stream of anhydrous NH.sub.3 for 2 seconds. The NH.sub.3 stream was directed against the membrane so as to force the ammonia through the pores of the membrane. After this dry ammonolysis treatment, the hydrophilicity of the membrane was comparable to the hydrophilicity of the membrane of Example 1, as shown by spontaneous and nearly instantaneous wetting with water. FT-IR showed that 100 percent conversion of the PVTFA to PVA was accomplished. This demonstrated that the 2 minute ammonolysis time of Example 1 was only required in order to allow the NH.sub.3 vapor to diffuse into the pores, and that by forcing the NH.sub.3 into the pores, the true ease of conversion to PVA is appreciated.

The ability to convert syndiotactic PVTFA treated membranes into PVA shell membranes using a variety of hydrolysis reagents and conditions was demonstrated by dipping the PVTFA treated membrane of Example 1 into solutions of various bases as well as a HCl solution. The results are shown in Table 4.

In order to effect the conversion of PVTFA to PVA throughout the membrane, the solution must be able to wet the PVTFA treated membrane, or the base must be volatile in order to deliver the hydrolysis reagent to the internal pore surfaces (c.f. Examples 10E and 10F). The 30 minute immersion time was probably excessive, but was chosen to ensure complete hydrolysis. All cases that resulted in a hydrophilic membrane (Examples 10B, 10C and 10D) showed 100 percent conversion of the PVTFA to PVA under FT-IR analysis performed according to Example 1.

The PE microporous base membrane of Example 1 was treated, as in Example 2, with 4 percent (w/v) syndiotactic PVTFA solutions in various solvents as noted in the Table 5 below. This example demonstrates that a variety of solvents other than the preferred acetone, including: esters, cyclic ethers, aliphatic and aromatic ketones, nitriles, and amides, can be used to prepare PVTFA treated membranes. Also, shown by comparison, are solvents which could not be made to work, due to the substantial insolubility of PVTFA in these solvents. Copolymers of PVTFA, described in other examples within this disclosure, are not limited to the solvents listed below.

Microporous PE membrane samples, prepared according to Example 23 of U.S. Pat. No. 4,539,256, having a pore size of 0.548 .mu., a thickness of 0.056 mm, a porosity of 88 percent, and a Gurley value of 5.4 sec, were treated, as in Example 2, with 4 percent (w/v) acetone solution of syndiotactic PVTFA copolymers having either vinyl acetate or maleic anhydride as the comonomer. The copolymers were prepared by free radical polymerization of the appropriate ratio of vinyl trifluoroacetate and the corresponding vinyl comonomer (i.e., vinyl acetate or maleic anhydride, respectively) according to Examples 1 and 4, respectively, of U.S. Pat. No. 4,618,649 (Ofstead), which is incorporated herein by reference. In-situ hydrolysis of these treated membranes produced PVA shell membranes which were hydrophilic as evidenced by spontaneous and nearly instantaneous wetting with water. In order to show that the crystallinity of the hydrophilic PVA shell was not excessively disrupted by the incorporation of less than about 5 percent of comonomer, each PVA shell membrane sample was subjected to a water extraction to determine PVA loss. Initial PVTFA and PVA add-ons were determined according to the procedures of Example 3. One liter of water was passed through a disc of each PVA shell membrane having a surface area of 36.3 cm.sup.2 and the resulting PVA weight loss calculated by weight differential of the membrane sample. The weight loss results tabulated in Table 6 show that less than 1 weight percent of the membrane weight was lost due to the water wash step. The hydrophilicity of the washed and dried samples were comparable to the hydrophilicity of the unwashed membranes, demonstrating that the presence of less than about 5% comonomer had not significantly disrupted the crystallinity of the PVA.

A microporous polysulfone membrane having a surfactant coating to render it hydrophilic and having a rated 0.45 micron pore size (obtained from Schleicher & Schuell) was rinsed in isopropyl alcohol to remove the surfactant coating. The then hydrophobic polysulfone membrane was saturation-treated with a 4 percent (w/v) solution of syndiotactic PVTFA in acetophenone (a poor solvent for polysulfone) following the procedure of Example 2. The resulting PVTFA treated membrane was hydrolyzed in an ammonia atmosphere according to the procedure of Example 2 to produce a hydrophilic PVA shell membrane which was spontaneously and nearly instantaneously wet with water.

A microporous polyvinylidene fluoride, PVDF, membrane made according to Example 22 of U.S. Pat. No. 4,539,256, having a 0.21 micron pore size, a 72 sec Gurley value and a 58.3 percent porosity, was saturation-treated with a 4 percent (w/v) solution of syndiotactic PVTFA by forcing the solution through the membrane by applying a partial vacuum to the opposite side of the membrane. The resulting PVTFA treated membrane was hydrolyzed in an ammonia atmosphere according to the procedure of Example 2 to produce a hydrophilic membrane that was spontaneously and nearly instantaneously wet with water.

Polyethylene microporous membranes, prepared according to Example 23 of U.S. Pat. No. 4,539,256, having a range of porosities and pore sizes, were treated with a 4 percent (w/v) solution of syndiotactic PVTFA in acetone according to the procedure of Example 1. Upon hydrolysis with the ammonia vapor according to the procedure of Example 2, hydrophilic PVA shell membranes which were spontaneously and nearly instantaneously wet with water were produced. The porosity, pore size and Gurley values of both the starting membranes, numbers 13A, 13C, 13E, and 13G, and the PVA shell membranes, numbers 13B, 13D, 13F, and 13H, are tabulated in Table 7, along with comments concerning the amount of surface pore blockage that occurred. The extensive pore blockage noted with the 0.101 .mu.m pore size membrane is due to the fact that the solution evaporates more rapidly than it can penetrate into the small pores which results in the formation of a pore-blocking skin at the surface of the membrane. Solvents having lower vapor pressures and/or lower viscosity solutions should lessen the occurrence of this type of pore-blocking skin formation.

The procedure of Example 1 was used to prepare a hydrophilic PVA shell membrane from a microporous PE membrane having a pore size of 0.259 microns and a 77 percent porosity, prepared according to Example 23 of U.S. Pat. No. 4,539,256, except that a 4.7 percent (w/v) solution of syndiotactic PVTFA in cyclohexanone was used, and the two zones of the oven were set to 38.degree. C. and 106.degree. C. respectively. The higher viscosity of the cyclohexanone PVTFA solutions relative to the viscosity of the acetone PVTFA solutions coupled with the relatively high oven temperatures used to dry the treated membranes resulted in the formation of an integral pore-blocking skin on the membrane surface. The presence of the skin was demonstrated by an effectively infinite Gurley air permeability value and by SEM analysis.

Samples of hydrophilic microporous membranes were prepared according to Example 1 and were subjected to extractions by highly polar organic solvents to demonstrate the ability of the PVA shell to resist wash-out. The initial PVA add-on was 9.1 weight percent of the untreated hydrophobic microporous membrane. The samples were weighed and then soaked for 1.5 hours in the indicated solvents, followed by four rinses of water to remove the solvent. The samples were dried, reweighed and a percentage weight loss for the hydrophilic membrane calculated by weight differential.

Samples exposed to each of the above solvents remained hydrophilic to varying degrees. Dimethyl formamide (DMF) caused the greatest percentage weight loss of the PVA shell, perhaps because the DMF dissolves the PVA crystallites (c.f. FIG. 3). Thus, while the hydrophilic polymeric structure produced according to the present invention is resistant to washout by DMF at least after continuous exposure for up to 1.5 hours, care should be taken to select a polymeric structure which does not also degrade or dissolve during exposure to the highly polar solvent.

An inherently hydrophilic microporous Nylon 6,6 membrane, rated with a 0.45 micron pore size, obtained from Schleicher & Schuell of Keene, N.H., was saturation-covered with a 4 percent w/v acetone solution of syndiotactic PVTFA following the procedure of Example 2. The resulting PVTFA envelopment of the internal and external surfaces of the membrane did not block the pores, but due to the hydrophobicity of PVTFA, the Nylon 6,6 membrane was rendered hydrophobic. Upon reaction with the ammonia vapor of ammonium hydroxide, as in Example 2, the PVA shell membrane again became hydrophilic as demonstrated by spontaneous and nearly instantaneous wetting with water. Characterization data, before and after this treatment are tabulated in Table 9, below. This example demonstrated the use of the present treatment to provide hydroxyl functional groups to the surface of a hydrophilic membrane without significantly blocking the pores or reducing the hydrophilicity.

A piece of Gore-Tex.TM. poly(tetrafluoroethylene) membrane, manufactured by W. L. Gore and Associates, Inc. of Elkton, Md. was saturated with a 5 percent w/v acetone solution of syndiotactic PVTFA prepared according to Example 1 using a #14 wire-wound bar to spread the solution. This sample was allowed to dry at room temperature in a ventilation hood to produce a PVTFA shell on the external and internal surfaces of the membrane, without causing pore blockage or PVTFA skin formation, as shown by SEM examination. The sample was reacted with vaporous ammonia as in Example 1 to yield a highly hydrophilic membrane, as shown by being spontaneously and nearly instantaneously wetted with water.

A piece of a calendered spunbonded PE web, commercially available under the trademark "Tyvek T-984", from E.I. DuPont of Wilmington, Del., having an average Gurley air flow of 3.1 sec per 50 cm.sup.3 of air, was saturation-covered with a 4 percent w/v acetone solution of syndiotactic PVTFA according to the procedure of Example 2. After drying and reacting in-situ with ammonia vapor, the PVA shell web was hydrophilic as judged by being spontaneous and nearly instant wettability with water. The web was still through-porous, since water would pass through the web after the hydrophilization and exhibited a Gurley value of 9.8 sec per 50 cc of air.

A polypropylene melt-blown web, was made according to the procedure described in Wente, Van A., "Superfine Thermoplastic Fibers" in Engineering Chemistry, Vol. 48, p. 1342 et. seq. (1956), or in Report No. 4364 of the Naval Research Laboratories, published May 25, 1954, entitled "Manufacture of Superfine Organic Fibers", by Wente, V. A.; Boone, C. D.; and Fluharty, E. L., the disclosures of which are incorporated by reference herein. It was covered with a 6 percent w/v acetone solution of syndiotactic PVTFA according to the procedure of Example 2. After drying and in-situ reaction with ammonia vapor, this PVA shell melt-blown web was hydrophilic as demonstrated by spontaneous and nearly instant wetting with water. The resistance of this hydrophilic treatment to washout was demonstrated by 16 repeated soak/squeeze/dry cycles with pure water, resulting in a melt-blown web that was still as hydrophilic as it was initially, as shown by spontaneous and nearly instant wettability with water.

A polypropylene woven fabric, obtained from the Arthur Kahn Co., of N.Y., N.Y., which was hydrophobic (i.e., a drop of water did not penetrate the fabric when it was placed on the fabric gently) was covered with a 4 percent w/v solution of syndiotactic PVTFA in acetone using the method of Example 2. (The weave of the starting fabric was coarse enough, however, to allow water to penetrate if any pressure was applied to the drop.) This resulted in a shell of PVTFA enveloping the surface of the fabric's fibers. Upon reaction of the PVTFA covered fabric with the ammonia vapor of ammonium hydroxide, as in Example 2, the fabric having a PVA shell about its surfaces became hydrophilic as demonstrated by spontaneous and nearly instant wetting with water.

In order to show the availability of the hydroxyl functional groups of the hydrophilic shell towards chemical derivatization, the hydrophilic membrane from Example 1 was reacted with an acid chloride. Enough sebacyl chloride was added to a glass vessel to cover a piece of the vacuum dried membrane placed in the vessel. These were allowed to react for 1/2 hour at room temperature. The sample was rinsed in 1,1,1-trichloroethane to remove excess acid chloride. Infrared spectroscopy of the reacted membrane showed a new carbonyl absorption at 1737.sup.-1 cm and a decrease in the hydroxyl absorption at 3300.sup.-1 cm, which indicated that esterification of the hydroxyl group had occurred.

A microporous polypropylene (PP) membrane, made by thermally induced phase separation as disclosed in U.S. Pat. No. 4,726,989 (Mrozinski), Example 9, having a Bubble Point maximum pore size of 0.65 .mu.m, an average Gurley of 6.4 sec per 50 cc of air and a thickness of 0.82 mm was extrusion saturated with a 4.5 percent (w/v) solution of syndiotactic PVTFA according to procedure of Example 1 except that the membrane was dried for about 45 seconds. The resulting treated membrane had a PVTFA add-on of 25.6 percent. A PVA shell membrane was prepared by hydrolyzing the PVTFA treated membrane in a stream of anhydrous ammonia according to the procedure of Example 9, followed by hydration with deionized water and drying at room temperature for about four (4) minutes. The PVA shell membrane had a Bubble Point pore size of 0.575 .mu.m and was hydrophilic as demonstrated by it being spontaneously and nearly instantly wetted with water.

The filtration performance of the PVA shell membrane, the untreated PP microporous membrane and a commercially available microporous membrane, namely a 0.22 .mu.m Durapore.TM. polyvinylidene difluoride microporous membrane (available from Millipore Corp, Bedford, Mass.) were compared by measuring the turbidity of the filtrate obtained when each membrane was challenged with a submicron sized suspension. A Hach Ratio Turbidimeter (Model 18900), available from Hach Instruments (Fort Collins, Colo.) was used to determine filtrate turbidity. The challenge suspension was prepared by adding six drops of a Fastek 0.22 .mu.m sized latex sphere suspension (formerly available from Eastman Kodak) to 1600 ml of ultrapure water which had a turbidity of 0.08 Nephelometric Turbidity Units (NTU) to produce a suspension having a turbidity of 117 NTU. A 47 mm diameter disk of the test membrane was placed on the support plate of a Gelman Magnet Filter Holder, the top of the filter holder installed and the filter holder placed on a vacuum filtration flask. A laboratory vacuum of approximately 56 cm Hg was applied to the filter flask and the average time required to collect 100 ml of filtrate for each membrane filter and the turbidity of each filtrate sample as measured on the Hach Turbidimeter are reported in Table 10.

The data in Table 10 shows that the microporous membrane filter based on the PVA shell membrane of the present invention has significantly better particle retention properties than the untreated membrane as is evidenced by the lower turbidity of the filtrate obtained using the PVA shell membrane. Reasonably close filtration rates between the PVA shell membrane and the Durapore membrane implies that the two membranes have porosities which are quite similar but the lower turbidity of the filtrate obtained with the PVA shell membrane suggests that it is likely that the PVA shell membrane has either a smaller pore size or a higher tortuosity as compared to the Durapore membrane and consequently it can provide superior filtration performance.

Microporous membrane filters known to provide absolute control over bacterial contaminants above a critical size can be used to "cold pasteurize" or sterilize thermally sensitive aqueous fluids. Several techniques are used to validate the retentive efficiency, compatibility, and life expectancy of filters with an absolute pore-size rating above 0.02 .mu.m. While a rigorous validation of filter efficiency requires the use of several techniques, an indication of filter efficiency can be provided by challenging the filter with 0.22 .mu.m latex particle and comparing the concentration of spheres up- and downstream of the filter by means of turbidimetric analysis (see Goldsmith et. al., Pharmaceutical Manufacturing, Nov. 1985. pp 31-37). The data in Table 10 suggests that the PVA shell membranes of the present invention have the potential of realizing an "absolute" rating for control of particles larger than 0.22 .mu.m and thus, might be suitable for sterilization of aqueous fluids.

A standard acid copper plating tank having a capacity of 189 liters was filled with an aqueous electrolyte solution containing 210 g/L of copper sulfate, 60 g/L of sulfuric acid, 75 ppm of chloride ion, 0.4% (v/v) of "CuFlex" 327 liquid grain refining additive commercially available from McGean-Rohco, Inc. of Cleveland, Ohio, and 0.125% (v/v) "CuFlex" 326 liquid brightener additive also commercially available from McGean-Rohco, Inc. Temperature in the tank was maintained at 24.degree. C. The cathode and anode were 15.24 cm squares of exposed copper with the reverse sides masked. The anode was covered by anode bags of the compositions shown in Table 11. The cathode and anode were spaced 10 cm. apart. Four volts DC was applied and the amperage was recorded as an indication of the efficiency of the anode bag to pass current. All of the anode bags were soaked in the electrolyte for 18 hours before the test.

An anode bag was constructed from PVA shell PE membrane prepared according to Example 1 and having a Bubble Point Pore Size of 0.463 .mu.m, a porosity of 73.6%, and a thickness of 0.06 mm was fitted over a titanium anode basket containing copper chips. A "Poly Nap" style 7020 anode bag was installed over the membrane anode bag for protection, even though the "Poly Nap" fabric anode bag reduced current flow. The anode assembly was installed in the tank and aqueous electrolyte solution used in Example 23 maintained at the same temperature. The tank also contained a filter with a polypropylene filter cartridge having a retention rating of 1 .mu.m and a nominal length of 25 cm. Platings from this tank were very smooth and shiny, indicating the membrane anode bag maintained a clean solution. The filter cartridge remained unexpectedly clean, and the electroplating current flow rate was at or near the maximum, notwithstanding the presence of the "Poly Nap" fabric anode bag. Deterioration of the "CuFlex" 326 and 327 additives was significantly reduced; the replenishment rate for the additives was halved. After completion of the experiment, the anode bag assembly was removed. A significant quantity of sludge and residues were retained in the assembly, more than when the "Poly Nap" fabric anode bag had been used alone.

A nickel electroplating tank similar to that illustrated in FIG. 4 was assembled, filled with an aqueous electrolyte solution containing 450 g/L of nickel sulfamate and 30 g/L of boric acid. Three different tests were run using the same components except for variation of the composition of the diaphragm separating the cathode and anode compartments.

In the first test, no diaphragm was used. In the second test, a 100% cotton denim cloth was used. In the third test, a PVA shell PE membrane was used. In each of the tests, the conductivity of the diaphragm was measured after the diaphragms were saturated in the electrolyte solution in the tank.

The following conditions were measured: the tank was maintained at 49.degree. C. The aqueous electrolyte solution had a pH of 3.8 and a specific gravity of 1.35. The cathode having a 49 cm diameter rotated at 5 rpm. The diaphragm having a 60 cm diameter covered aperture separating the cathode and anode compartments.

The application of from 1 to 7 volts D.C. generated from about 10 to 80 amps for each test. There was no measurable difference among the conductivity of the PVA shell PE membrane, the cotton cloth, or no diaphragm at all.

With no diaphragm present, any debris present flowed freely during the agitation of the electrolyte between the cathode and the anode compartments. However, the PVA shell PE membrane had smaller pore sizes than the cotton cloth, efficiently reducing the amount of circulating debris without any measurable drop in electroplating current flow.

Samples of a hydrophilic membrane prepared according to Example 1 were subjected to 14 day exposures to 36% H.sub.2 SO.sub.4 and 30% KOH solutions to determine if the PVA shell would be subject to degradation under conditions typically encountered in electrochemical cells having an aqueous-based electrolyte. Results of the exposure study are reported in Table 12.

After exposure to both the acid and alkaline soaks, the hydrophilicity of the membranes was comparable to that of the original membrane. No significant weight loss or dimensional shrinkage of the samples was noted.

A sample of the hydrophilic membrane of Example 1 was subjected to a 14 day exposure to a 36% H.sub.2 SO.sub.4 solution maintained at 49.5.degree. C. to determine if the PVA shell would be subject to thermal or oxidative degradation under elevated temperature conditions which could be encountered in electrochemical cells having a hydrophilic electrolyte. The acid solution was placed in a modified Dewar condenser which was used to condense cyclopentane (B.P. 49.5.degree. C.) to maintain a constant elevated temperature. Results of the exposure study are reported in Table 13.

Examination of the data in Table 13 suggests that there is no significant degradation of the hydrophilic membrane under the elevated temperature conditions used in Example 26.

A sample of the hydrophilic membrane of Example 1 (basis weight 0.022 gm/m.sup.2) was thermally point bonded to a 10 gm/m.sup.2 basis weight polyethylene microfiber web (made using Dow 6806 Polyethylene resin, available from Dow Chemical, Midland Mich., in a procedure similar to that described in Wente, Van E., "Superfine Thermoplastic Fibers" in Industrial Engineering Chemistry, Vol. 48, pages 1342 et seq (1956) or in Report No. 4364 of the Naval Research Laboratories, published May 25, 1954 entitled "Manufacture of Superfine Organic Fibers" by Wente, Van E., Boone, C. D., and Fluharty,, E. L.) in a regularly spaced dot pattern. The point bond areas accounted for approximately 15% of the total surface area of the hydrophilic semipermeable membrane and reduced the porosity of the membrane by 15%. The electrolyte resistivity and compressibility of the resulting laminate was compared to the resistivity and compressibility of a glass microfiber and a silica filled polyethylene battery separator material. Results of this comparison are reported in Table 14.

As shown in Table 14, the electrolyte resistivity and percent compression of the electrochemical cell separators based on the PVA shell membranes of the present invention are significantly superior to those demonstrated by the silica filled PE materials currently being used as electrochemical cell separators.

Neither the embodiments of the invention nor the examples described above limit the scope of this invention.