Composite membranes for flow batteries

Abstract

To provide a flow battery system including composite membranes that achieves significant cost reduction.SOLUTION: A flow battery system comprises: a positive electrolyte flowing through a positive reaction chamber; a negative electrolyte flowing through a negative reaction chamber; and a non-ionomeric membrane comprising a hydrogel physically separating the positive reaction chamber from the negative reaction chamber; wherein the membrane is ionically conductive but resistant to hydraulic crossover and at least one of the positive and negative electrolytes comprises an aqueous solution. The membrane consists of a hydrogel-forming material (e.g., poly (vinyl alcohol)) and an optional porous polymeric material.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to U.S. Patent Application No. 62 / 300,323, filed February 26, 2016.

[0002] This disclosure was developed with funds received under a contract with the U.S. Department of Energy (Contract DAR0000352), which has certain rights in the present invention.

[0003] The present invention relates to the field of flow batteries, and more particularly to systems and devices including composite membranes for use in such batteries, as well as methods of making and using the same. [Background technology]

[0004] A flow battery is a rechargeable battery that uses an electrolyte that moves ("flows") through an electrochemical cell to convert chemical energy from the electrolyte into electricity (and vice versa during charging). The active materials used in flow batteries generally consist of ionizable metal salts or redox-active organic compounds dissolved in a fluid such as water or an organic solvent. Additional salts or acids, such as NaCl or HCl, may also be provided to the fluid to produce a highly conductive electrolyte.

[0005] The electrolytes for flow batteries are typically stored in separate tanks and pumped through the individual cells of the flow battery in a controlled manner, usually according to the applied charge / discharge current. Multiple cells can be used in series or parallel (i.e., "stacked") to produce the desired electrical characteristics of the battery.

[0006] During discharge, each cell consists of a positive (cathode) and negative (anode) electrode and a separator membrane. The electrodes catalyze the desired reaction. The membrane prevents the electrodes from contacting while allowing the conduction of ions necessary to complete the electrical circuit. The separator must also prevent mixing of the circulating positive and negative electrolytes and minimize migration of species created in the electrolytes due to crossover or mixing with other components (e.g., other electrolytes) during charge. Additional mechanical and control structures can be used to create and maintain the desired flow of electrolytes / reactants through the cell or cells.

[0007] True flow batteries have all active chemical species flowing through the battery and stored in an external tank. Reduction-oxidation (redox) flow batteries, such as vanadium or iron / chromium redox batteries, store electrical energy in chemical form and then distribute the stored energy in electrical form through a spontaneous reverse oxidation-reduction reaction. The discharged electrolyte can flow through a reactor cell to which an external voltage source is applied so that the electrical energy can be converted back to chemical energy.

[0008] "Hybrid" flow batteries have at least one active material physically located within the cell / stack, such as zinc metal in a zinc-bromine battery or iron in an all-iron battery, while the other active material is dissolved in an electrolyte. Such hybrid flow batteries still utilize separate positive and negative electrolytes and require a separator or membrane.

[0009] A variety of chemistries have been used in flow batteries, with typical cell voltages ranging from <1.0 to about 2.4 volts, with non-limiting examples being hydrogen-lithium bromate, iron-chromium, all-iron, vanadium-vanadium sulfate (or mixed sulfates and chlorides), sodium-bromine polysulfide, and zinc-bromine.

[0010] In effect, flow batteries resemble fuel cells in that they rely on electron transfer (rather than intercalation or diffusion). Similarly, flow batteries share characteristics with rechargeable conventional batteries, except that the active materials in a flow battery can be easily replenished outside the cell (e.g., in a storage tank through the application of an appropriate charge). As such, flow batteries have advantages over conventional batteries, which have limited discharge capacity based on the active materials contained within the cell and, in the case of secondary batteries, relatively limited cycling capabilities due to the inherent limitations of repeated intercalation. Flow batteries are advantageous compared to fuel cells because they do not require specialized catalysts and can replenish reactants through the application of electrical current, without requiring total cell or reactant replacement. Nevertheless, while flow batteries hold great promise, to date they have mostly found use in larger stationary applications and / or in combination with other power generation methods.

[0011] Examples of flow batteries can be found in U.S. Patent Publication No. 2014 / 0227574; U.S. Provisional Patent Application No. 62 / 239,469, filed October 9, 2015, and International Patent Application No. US15 / 50676, filed September 17, 2015. The entire disclosure of each of these documents is incorporated herein by reference, and the inventors further reserve the right to seek to establish priority to these applications to the maximum extent permitted by applicable law.

[0012] One of the most common membrane materials in use today is a sulfonated tetrafluoroethylene-based fluoropolymer-copolymer sold under the trade name NAFION by EI du Pont de Nemours and Company, based in Wilmington, Delaware. This material has a Teflon backbone with acidic sulfonic acid groups, resulting in cation conductivity, resistance to chemical reactivity and moderate temperatures, and relatively good water and gas permeability. However, the material's unique properties also entail a relatively high cost, and NAFION® has poor conductivity in flow batteries that rely on ions other than proton ionic charge transfer (e.g., potassium, sodium, chloride, etc.).

[0013] Another approach to membranes is disclosed in U.S. Patent Publication No. 2013 / 011504. An ion-exchange membrane for redox flow batteries is described. The membrane is made from a composition including an ion-conducting material, such as an ion-conducting monomer or polymer (e.g., a quaternary ammonium salt), and a water-soluble support. In some examples, a non-effective ionic material is selectively substituted for the ion-conducting polymer. The resulting membrane is described as useful in flow batteries with non-aqueous electrolytes and solvents.

[0014] Porous and microporous membranes have also been considered for use in flow batteries. However, these materials are extremely sensitive to pressure differences, insofar as the porosity allows the liquid / fluid to pass through (as opposed to following the intended flow path of the electrolyte in question). As a result, the use of porous membranes often requires additional components to monitor and regulate the pressure and flow rate of both electrolytes. To the extent that pores are physically blocked, the resulting localized imbalance in porosity presents additional challenges, even if such control components are used.

[0015] In light of the above, cost-effective alternatives to known membrane materials are needed. Furthermore, membranes that do not require detailed monitoring and regulation of pressure differentials within flow batteries, particularly via complex control components, would be welcome. Materials that are more compatible with aproton-rich electrolytes, and more specifically with iron (e.g., Fe) in various charge states, are also needed. 2+ , Fe 3+ ) or vanadium (e.g., V 2+ , V 3+ , V 4+ , V 5+ , etc.) and other multivalent (i.e., other ionic species other than iron and / or vanadium exhibiting several different valencies) and / or cationic species (e.g., K + , Na + , H + There is also a need for materials that exhibit selectivity between . . . . . Summary of the Invention

[0016] In light of the above, membranes for flow batteries have been considered. These membranes consist of a hydrogel-forming material (e.g., poly(vinyl alcohol), hereafter "PVA") and an optional porous polymer material. These membranes are placed between half-cells of a flow battery and allow the selective flow of ions through the membrane without relying on traditional ionomers. These membranes are expected to be particularly useful in the field of iron flow batteries, but their selectivity has utility in a wider range of multivalent flow battery chemistries.

[0017] In certain embodiments, the flow battery includes any combination of the following features: · Positive electrolyte flowing through the positive reaction chamber; · Negative electrolyte flowing through the negative reaction chamber; a non-ionomeric membrane containing a hydrogel that physically separates the positive reaction chamber from the negative reaction chamber; The membrane is ionically conductive but resistant to hydraulic crossover; · At least one of the positive and negative electrolytes comprises an aqueous solution; · At least one of the positive and negative electrolytes contains a multivalent ionic species; The polyvalent species is vanadium (2 + ) ions, vanadium (3 + ) ions, vanadium (4 + ) ions, vanadium (5 + ) ions, cuprous ions, cupric ions, ferric ions, and ferrous ions; · The hydrogel contains poly(vinyl alcohol); · the hydrogel is selected from poly(vinyl alcohol), poly(acrylic acid), poly(ethylene glycol), and combinations thereof; · Hydrogels effectively seal the pores in polymer matrices; · The polymer substrate is microporous; · The polymer substrate consists of polyethylene, polypropylene, and combinations thereof; · The polymer substrate has a void volume of 30% to 80%; · The polymer substrate has a void volume of more than 80%; · The membrane consists of a hydrogel; The pH of the positive and negative electrolytes is less than 3.0, more preferably less than 2.0, and most preferably less than 1.5; and The hydrogel may be at least one of hydrolyzed, crystalline, semi-crystalline, and crosslinked.

[0018] In another aspect, a method of making a membrane for a flow battery can include any combination of the following features: providing a substrate having a plurality of pores that form a void volume; Impregnating the pores with hydrogel: · Processing hydrogels to produce non-ionomeric but ion-conducting membranes; · the hydrogel is selected from poly(vinyl alcohol), poly(acrylic acid), poly(ethylene glycol), and combinations thereof; The processing of the hydrogel includes at least one of thermal crosslinking, chemical crosslinking, photochemical crosslinking, and hydrolysis of the hydrogel; · impregnating the pores includes at least one of film casting, roll-to-roll casting, infiltration, and dip coating; · Impregnating the pores results in a hydrogel layer deposited on at least one side of the substrate with a thickness of 0.1 to 25 micrometers; - at least one of cleaning the pores and removing air from the pores before impregnating them; Submerging the substrate in alcohol; the hydrogel comprises poly(vinyl alcohol) selected to have a purity of at least 99% by weight; and At least 95% by weight of the hydrogel is hydrolyzed.

[0019] In further embodiments, the membrane with tolerance to pH levels approaching 1.0 that is selective for multivalent ionic species comprises any combination of the following features: · Microporous polymer membranes; · Hydrogel impregnated into the membrane; · the hydrogel is selected from poly(vinyl alcohol), poly(acrylic acid), poly(ethylene glycol), and combinations thereof; The polymer film comprises at least one of polyethylene and polypropylene; · The polymer film is virtually bubble-free; The polymer membrane has a porosity of 30% to 80%, more preferably 40% to 60%; The polymer membrane has a thickness of 25-200 micrometers before impregnation with the hydrogel; The polymer membrane has a thickness of less than 200 micrometers after impregnation with the hydrogel.

[0020] In a final aspect, the method of making and operating a flow battery can include any combination of the following features: Providing or forming a composite membrane according to any embodiment disclosed herein; providing first and second electrolytes, each having a multivalent ionic species dissolved therein, on different sides of the composite membrane; and circulating the first and second electrolytes through separate flow paths in response to an electrical or charging load; Multivalent ion species are vanadium (2 + ) ions, vanadium (3 + ) ions, vanadium (4 + ) ions, vanadium (5 + ) ions, cuprous ions, cupric ions, ferric ions, and ferrous ions; the first electrolyte contains ferric ions and the second electrolyte contains ferrous ions; and At least one of the first electrolyte and the second electrolyte has a pH of less than 3.0, more preferably less than 2.0, and most preferably less than 1.5.

[0021] While the above features are identified separately as separate embodiments of the invention, it is to be understood that the present disclosure also contemplates combining features from one disclosed embodiment with any of the other embodiments. To the extent that one embodiment may include a method and another embodiment may include an apparatus, means for adapting the method to the apparatus (and vice versa) are also contemplated and disclosed.

[0022] The accompanying drawings illustrate various systems, apparatus, devices, and related methods, where like reference numerals refer to like parts throughout. [Brief explanation of the drawings]

[0023] [Figure 1] Figure 1 is a schematic diagram of a hybrid flow battery.

[0024] [Figure 2] FIG. 2 is a schematic representation of crosslinks within a hydrogel. DETAILED DESCRIPTION OF THE INVENTION

[0025] Reference will now be made to exemplary embodiments, examples of which are illustrated in the accompanying drawings. It will be understood that other embodiments may be utilized and that structural and functional changes may be made. Furthermore, features of various embodiments may be combined or modified. As such, the following description is presented for purposes of illustration only and is in no way limiting of the various substitutions and modifications that may be made to the illustrated embodiments. In this disclosure, numerous specific details are set forth to provide a thorough understanding of the subject disclosure. It will be understood that aspects of the present disclosure may be embodied in other embodiments, etc., that do not necessarily include all aspects set forth herein.

[0026] As used herein, the words "example" and "exemplary" mean an example or illustration. The words "example" or "exemplary" do not indicate a primary or preferred aspect or embodiment. Unless the context suggests otherwise, the word "or" is intended to be inclusive rather than exclusive. As an example, the phrase "A uses B or C" includes inclusive permutations (e.g., A uses B, A uses C, or A uses both B and C). As another note, unless the context suggests otherwise, the articles "a" and "an" are generally intended to mean "single or plural," with the use of the plural being exemplary rather than required.

[0027] Hydrogel-forming materials, such as poly(vinyl alcohol), poly(acrylic acid), or poly(ethylene glycol), as well as blends of two or more of these materials, have found use in electrochemical cells. For example, U.S. Pat. No. 5,211,827 describes a conventional electrochemical cell using a solid, nonporous composite membrane formed by uniformly dispersing a hydrogel in an inert matrix. Similarly, U.S. Pat. No. 4,614,575 describes the use of nonionic aqueous gels to facilitate electrolyte flow while preventing gas permeation in various electrochemical systems. Generally, the use of nonionic poly(vinyl alcohol) gel film separators in alkaline silver-zinc batteries dates back even further. In all cases, these challenges or the incorporation of soluble PVA membranes in a single aqueous electrolyte with lower-cost, less complex alternatives may have led to the relative lack of further development of PVA separators.

[0028] In conventional batteries, the necessary ionic conduction through the hydrogel occurs when the liquid battery electrolyte penetrates and swells the hydrogel. Cross-linking is necessary to control the degree of swelling and prevent the hydrogel polymer from dissolving in the electrolyte. For fuel cells, ionomers such as polystyrene sulfonic acid or NAFION® are typically incorporated into the separator. Conduction occurs when water is absorbed into the hydrogel, ionizing the acid groups of the ionomer and freeing protons for conduction. However, NAFION® and other ionomers do not allow for the transfer of H + , Na + , and / or K + Compared to multivalent cations (V 2+ , Fe 2+ , Cr 3+ ), causing a significant loss of ionic conductivity, which makes multivalent cation-based flow battery systems highly inefficient.

[0029] Recent literature (reviewed by Maiti in The Journal of Power Sources, Vol. 216, pp. 48-66, 2012) suggests that the modification of PVA by sulfonation or the incorporation of hydrophobic or other moieties may prove useful in PVA-based membranes for direct methanol fuel cells, given PVA's high selectivity for water over alcohols. However, these structures create ionomeric properties in the membrane.

[0030] In contrast to the use of similar membranes for conventional batteries and fuel cells, separators / membranes for flow batteries must provide ionic conductivity and also prevent (or at least minimize) the migration of dissolved ionic reactants from the positive electrolyte to the negative electrolyte and vice versa. That is, reactant migration, particularly via hydraulic osmosis, is a significant challenge unique to flow batteries, which limits the use of conventional battery and fuel cell separators (e.g., low-cost microporous membranes) in flow batteries.

[0031] Movement of reactants can be driven by 1) a naturally occurring concentration gradient (i.e., by diffusion), 2) a naturally occurring potential difference (i.e., by migration), or 3) a pressure difference (total pressure or osmotic pressure) that results in bulk movement of liquid electrolyte through the membrane (hydraulic osmosis). In conventional batteries, only ions of the supporting electrolyte, such as potassium and hydroxyl ions for alkaline batteries, are present in the electrolyte, so reactant movement is not an issue. In fuel cells, unwanted movement of fuel (hydrogen or methanol) molecules can be an issue, but these substances are not ionic, and movement is controlled by various factors. Only in flow batteries is conduction of some ions desirable while conduction of others minimized, and only in flow batteries does hydraulic osmosis of the electrolyte become an issue to be avoided.

[0032] Therefore, we propose to exploit the advantages of a microporous membrane as a support structure in combination with a cross-linker and conductive material such as a hydrogel, particularly poly(vinyl alcohol), as a means of protection against excessive pressure differentials in flow batteries. The resulting structure achieves significant cost savings compared to NAFION® and other ionomer / ion exchange materials, while significantly mitigating the drawbacks of using only microporous membranes in flow batteries.

[0033] Furthermore, the optimal configuration of a composite separator using a hydrogel in a flow battery is unique compared to similar disclosures of hydrogels in electrochemical cells found above and elsewhere. The combination of a hydrogel with a membrane in the context of a flow battery (i.e., hydraulic permeation, diffusion, etc.) entails factors such as the pore size and pore volume of the porous support, the composition and amount of the hydrogel-forming material and its degree of crosslinking, and whether the hydrogel material forms a continuous film on the surface of the support and / or partially or completely fills the pores of the support. Insofar as these factors are not relevant to the inventors' knowledge, or even if they are relevant, they are considered prior art with respect to conventional electrochemical cells, the composites and systems contemplated herein are unique.

[0034] In addition to semi-crystalline, nearly fully hydrolyzed PVA (see below) as a component of the composite membrane of a flow battery, many other options are envisioned. For example, PVA modified with cross-linkable moieties (e.g., acrylates, epoxides) can be used to infiltrate the porous separator material, followed by thermal, chemical, or photochemical cross-linking. Non-ionomeric alternatives to PVA include hydrophilic polymers such as poly(ethylene oxide) or PEO, poly(N-vinylpyrrolidone) or PVP, poly(2-hydroxyethyl methacrylate) or polyHEMA, poly(acrylamide) and derivatives and copolymers, and poly(ethyl-2-oxazoline) and related materials with appropriate cross-links, either physical or chemical, to render them insoluble in water. Bio-derived hydrophilic polymers such as chitosan, gelatin, and hydroxypropyl cellulose may also be considered.

[0035] Blends of one or more of these polymers may also be important. One or more blending components may include polyelectrolytes such as poly(acrylic acid) or its neutralized derivatives such as poly(sodium acrylate), and various copolymers of acrylic acid with, for example, acrylamide in various amounts. Biologically derived polyelectrolytes such as alginate and hyaluronic acid may also be considered.

[0036] Multilayer composite membranes, such as a layer of cross-linked PVA sandwiched between two conventional porous separators, are also envisioned, which would provide better mechanical stability and prevent direct exposure of the relatively soft hydrogel material to the flowing electrolyte.

[0037] PVA embodiments of the present invention should use very pure forms of the material. Ideally, the poly(vinyl acetate) precursors commonly used to generate the final alcohol in situ should be at least 90%, at least 95%, and more preferably at least 99% pure. Failure to use sufficiently pure PVA and / or precursors can result in undesirable impurities that can reduce the performance of the resulting membrane.

[0038] PVA embodiments include depositing a hydrogel material onto a microporous membrane by film casting, roll-to-roll casting, infiltration, dip coating, and other techniques. The PVA must then fill or otherwise effectively occlude the pores of the microporous separator to effectively prevent fluid flow through the pores. The surface coating applied to either / both sides of the membrane should have a depth of approximately 0.1 to 25 micrometers per side. In some embodiments, all pores are filled. In some embodiments, one or both sides of the separator are coated. Ultimately, the application or deposition of the hydrogel must be sufficient to avoid or significantly reduce the aforementioned problems associated with uncoated porous structures.

[0039] A pore is said to be effectively blocked when its effective diameter is reduced by about two orders of magnitude (e.g., a 100 nanometer pore is reduced to an effective size of about 1 nanometer). However, the coating is effectively continuous on its surface, and effective blocking includes complete filling of the pore, substantial coating of the pore's inner diameter along at least a portion of its path through the pore, and / or sufficient filling of the pore on one or both surfaces to effectively (substantially) block the pore. Ultimately, the effectiveness and / or utility of a membrane is determined by its hydraulic permeability and ion selectivity characteristics, and effective blocking of a pore is merely a reflection of these properties.

[0040] Examples of microporous separators suitable for use in the present invention include inert, thin polymer films with thicknesses of 1 to 1,000 micrometers (μm) made from polypropylene and / or polyethylene, covering a wide range of molecular weights. Membranes made from or containing polytetrafluoroethylene can also be used as supports. The pore size of such membranes should be between 0.01 and 1.00 μm, with a porosity or void volume of 30 to 80%, with more preferred upper and lower limits bounded by any two values ​​selected from the following porosity / void volume ranges: 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, and 80%. Membranes can be uniaxial or biaxial in nature, and can be composed of multiple layers of polymeric material. Other notable properties include stability in electrolyte, exposure to active redox species, mechanical properties necessary to support a free-standing membrane, surface properties compatible with hydrogel penetration and coating, and wettability (both for cleaning / bubble removal and for hydrogel penetration). Preferred characteristics are further demonstrated in the following examples, which should be understood as embodying specific embodiments of the invention.

[0041] Surprisingly, the inventors have found that these composite films exhibit improved wettability (compared to the microporous film alone) and unexpectedly low resistivity (i.e., less than 0.3 ohm-cm 2(see below). Furthermore, the inventors believe these composites have a level of ion selectivity for ferrous and ferric ion species relative to other cationic species. Further testing may demonstrate similar utility for vanadium, copper selectivity, and possibly other flow battery and hybrid-type systems. In this context, ion selectivity means that relatively larger multivalent cations may be conducted through the membrane, but at a slower rate than smaller diameter monovalent cations. In this way, composite membranes may differ significantly from NAFION® and other ionomers, where physical partitioning is achieved by incorporating cations into the NAFION® material itself. Without wishing to be bound by any particular theory, the inventors believe that by preventing the passage of larger cations, the composite membranes may be effectively nanoporous in this regard, and may be likened in this regard to nanofiltration membranes, as opposed to the more porous microfiltration and ultrafiltration membranes (which are useful for capturing substances with a molecular weight above 10,000, such as proteins), and the less porous reverse osmosis membranes (which are useful for trace organics and monovalent cations).

[0042] Finally, the composite separator demonstrated sufficient ionic conductivity (i.e., greater than 0.01 S / cm) for appropriate ionic species, particularly compared to current industry-standard ionomer membranes such as Nafion®. Ionic conductivity performance can also be reflected by the comparative performance of membranes with and without the present invention, as reflected by the battery's discharge, cycling, and / or charging performance, as well as its rate capability and total capacity. Furthermore, related measurements, such as resistance measurements (e.g., area-specific resistance), can be useful in verifying the improved ionic conductivity of the present invention compared to conventional membranes.

[0043] Ionic conductivity can be measured by equilibrating the membrane in a given electrolyte. Equilibration involves immersing the membrane and allowing sufficient time for complete wetting of the membrane. To measure conductivity, comparisons (i.e., with or without the membrane present) or multipoint measurements (i.e., above and beyond the membrane on both sides) may be performed. Diffusion coefficients can also be calculated and compared.

[0044] For certain polymer substrates, it may be advantageous to ensure that all undesired materials are first purged from the pores prior to the introduction of the hydrogel material. This can be accomplished by any number of procedures, including flushing the pores with one or more solvents to remove any residual oil or other residual components from the original manufacturing process. Additionally, it may be desirable to pre-wet the membrane to avoid trapping air within the pores, which could lead to undesired reactions. Methanol and other alcohols are particularly useful in both of these applications. In practice, the substrate is "pre-wet" with alcohol and then introduced into the aqueous solution to ensure that air bubbles are expelled and / or the hydrogel / PVA can be readily incorporated into the pores.

[0045] The composite membranes discussed herein may be particularly useful in ferric-ferrous chemistries and other multivalent ion flow battery chemistries (both traditional and hybrid). Additives can be used to enable sustained operation at low pHs of 1.0 or less in mixed sulfate / chloride electrolytes without experiencing unviable levels of hydrogen evolution. These electrolytes are also sufficiently conductive to prevent precipitation of ferric hydroxide. Preferred pH levels for the electrolyte are below 3.0, below 2.0, below 1.5, and below 1.0.

[0046] The composite hydrogel may comprise a mixture of PVA, poly(acrylic acid), and poly(ethylene glycol). The bonds formed within the porous support may be covalent, ionic, or a combination of the two. The hydrogel itself may be partially crosslinked according to various known methods to improve its structural stability, particularly with respect to minimizing swelling and avoiding dissolution of the hydrogel material itself. In some embodiments, a 3 wt. % aqueous solution of highly hydrolyzed PVA (i.e., at least half of the material, more preferably approaching 90%) is coated onto a microporous substrate, followed by one or a series of methanol treatments lasting several hours to stabilize the PVA structure, increase crystallinity, and promote PVA-PVA interactions while minimizing PVA-water interactions.

[0047] Notably, flow batteries present unique challenges and opportunities compared to previously proposed uses of PVA in conventional batteries and fuel cells due to the need for selective ionic conduction and sufficient physical separation of each aqueous mixture inherent in the opposing cells of a flow battery. These considerations are in direct contrast to conventional batteries and fuel cells, where pressure gradients and / or hydraulic crossover are not important. While it is true that fuel cells can experience fuel crossover, fuel cell technology is completely unconcerned with ionic crossover.

[0048] As used herein, resistance to hydraulic penetration may refer to hydraulic penetration normalized to thickness, 9x10 -12 The coating thickness (k w / t m ,cm 2 / cm), and or 10 -14 cm 2The following absolute values ​​are included (for comparison, open and non-enclosed spaces have significantly greater hydraulic permeability values ​​than 1.000000): Another comparison criterion for this resistance is the relative amount for a membrane coated according to embodiments described herein compared to a reduction in permeability of at least 25 times for the coated membrane to be considered resistant. Pressure gradients between half-cells can also indicate undesirable hydraulic crossover. Other measures, including indirect measures such as absolute or relative full or half performance on discharge, cycling, and / or charge, as well as rate capability and resulting cell capacity, can also be strong indicators of improved and acceptable hydraulic permeation resistance.

[0049] The ability of PVA and other hydrogels to crosslink may allow for highly porous substrates (approaching 80% or greater porosity). In some embodiments, completely self-supporting membranes can be formed entirely from hydrogels. Such self-supporting hydrogels include extremely pure PVA that can withstand temperatures of 50-60°C or higher. In other embodiments, sufficient purity of PVA (i.e., greater than 95%) can result in the formation of crystalline structures that allow the hydrolyzed PVA to self-crosslink. Extremely porous substrates can include simple meshes formed from threadlike or woven screen-like materials, expanded materials with regularly or irregularly formed apertures, and / or other planar materials with apertures formed therein.

[0050] Although the composite membranes are expected to be particularly useful in multivalent flow batteries, especially all-iron systems, the composite membranes described herein also exhibit ion selectivity for monovalent cations. + or NH4 + and Cl - The conductivity of the electrolyte can remain comparable even in liquid solutions of 0.01%. Apart from this, lower pH levels, in combination with possible vanadium-based hydrogen inhibitors, further improve the performance of the composite membrane in multivalent systems.

[0051] The composite membrane also shows no apparent degradation when cycled at both room temperature and elevated temperature (50° C.).

[0052] The methods described in the specific examples below encompass other aspects of the invention. While specific devices are disclosed, reasonably equivalent devices and techniques are encompassed by specific embodiments of the methods of the invention. Similarly, the flow battery systems described herein constitute specific embodiments of the invention, particularly insofar as the use of composite membranes simplifies and improves overall system performance and components.

[0053] FIG. 1 illustrates an embodiment of a flow battery system 100 suitable for use in connection with aspects of the present invention. Flow cell 100 includes two half-cells 102 and 104 separated by a separator 106. Half-cells 102 and 104 include electrodes 108 and 110, respectively, and are in contact with an electrolyte such that an anodic reaction occurs on the surface of one electrode and a cathodic reaction occurs on the other electrode. Electrolyte flows through each of half-cells 102 and 104 as oxidation and reduction reactions occur. In FIG. 1, a cathodic reaction occurs at electrode 108 (referred to herein as the positive electrode or cathode) of half-cell 102 during discharge, and an anodic reaction occurs at electrode 110 (referred to herein as the negative electrode or anode) of half-cell 104 during discharge.

[0054] The electrolyte in half-cells 102 and 104 flows through the system to storage tanks 112 and 114, respectively, and fresh / regenerated electrolyte returns from the tanks to the half-cells. In FIG. 1 , the electrolyte in half-cell 102 flows through pipe 116 to holding tank 112, and the electrolyte in tank 112 flows through pipe 118 to half-cell 102. Similarly, the electrolyte in half-cell 104 flows through pipe 120 to holding tank 114, and the electrolyte from tank 114 flows through pipe 122 to half-cell 104. The system can be configured as needed to facilitate or control the flow of electrolyte through the system, and the system can include, for example, any suitable pump or valve system. In the embodiment of FIG. 1 , the system includes pumps 124 and 126 for pumping electrolyte from tanks 112 and 114 to the half-cells, respectively. In some embodiments, the holding tanks can separate the electrolyte that has flowed through each cell from the electrolyte that has not. However, discharged or partially discharged electrolyte can also be mixed.

[0055] The electrodes 108 and 110 can be coupled to supply or receive electrical energy from a load or source. Other monitoring and control electronics included in the load can control the flow of electrolyte through the half-cells 102 and 104. Multiple cells 100 can be electrically coupled (“stacked”) in series to achieve higher voltages or in parallel to achieve higher currents.

[0056] In some embodiments, preferred composite membranes will include a stable carbon backbone formed from PVA. The PVA will be at least semi-crystalline to provide stability in water without the need for covalent crosslinking, as illustrated in the general molecular schematic of FIG. 2. In that illustration, hydrogel material 200 includes a PVA backbone 210 with selective crosslinks 220 formed between each backbone structure 210. The composite will also be non-ionomeric, resulting in the absence of ionic groups such as sulfonate or carboxylate, which may chelate multivalent ions, particularly ferrous and ferric ions. [Example]

[0057] Preparation of PVA and PVA composite separators Solutions of 2 wt % and 3.5 wt % poly(vinyl alcohol) or PVA (Sigma-Aldrich, 99+% hydrolysis, average molecular weight approximately 130 kD) in water were prepared by dissolving the PVA at approximately 85 °C with occasional stirring, followed by cooling to room temperature to obtain a clear, mildly viscous solution. As shown below, microporous polymer separators were coated with the poly(vinyl alcohol) solution using a doctor blade and / or by dip coating. The wet composite films were dried by heating in an air oven at 50 °C for 15 minutes. The resulting composites were then immersed in pure methanol for 16 hours to extract residual water from the composite films, increase the partial crystallinity, and improve the mechanical strength and underwater stability of the poly(vinyl alcohol) through physical cross-linking.

[0058] PVA films were coated using a casting knife film applicator and then polymerized in methanol. The thickness of the PVA coating was controlled by adjusting the blade vertically. After polymerization, the thickness of the composite separator was measured using a micrometer. The properties of the basic microporous support materials are summarized in Table 1. Table 2a shows the properties of the PVA-only membrane.

[0059] Ionic Conductivity Research Electrochemical impedance spectroscopy (EIS) was performed using a Solartron 1280B potentiostat (Ametek, UK) to obtain the ionic conductivity of the coated and uncoated separators. The ionic conductivity of both the uncoated microporous separator and the PVA composite separator was measured in a 1 M HCl electrolyte in an in-house designed and fabricated conductivity cell. The in-house fabricated cell consisted of two CPVC flow fields and a graphite current collector, 1 cm wide and 3.69 cm long (A = 3.69 cm). 2 The cell consisted of a 1-mm separator and a 1-mm gap between each current collector and separator, defined by a Teflon spacer. Current was collected from this cell through two brass plates pressed onto the back of the graphite current collector. Grafoil (Graftech, USA) was used to minimize the contact resistance between the brass plates and the graphite current collector. The electrolyte was pumped vertically through the in-house cell to an external chamber via a peristaltic pump (Cole-Parmer, USA) during all measurements. Conductivity values ​​were measured at room temperature. The resistance due to the membrane was obtained by subtracting the resistance measurement of the membrane sample from the resistance measurement without the membrane sample. This final resistance value was used to calculate the membrane conductivity.

[0060] Hydraulic and ferric ion crossover studies The crossover test was performed using a 6.5 cm diameter separator with conductive carbon felt electrodes pressed against either side of the separator. 2 The study was performed with both coated and uncoated separators in a ferric ion cell. Two separate external reservoirs were used in both studies. Equal flow rates were used during the ferric ion crossover test to ensure there were no pressure differences across the separator that could affect the results. One reservoir held 250 mL of concentrated electrolyte consisting of 1 M FeCl2 and 1 M FeCl3. The other reservoir held 1 M FeCl3, a small amount of concentrated HCl to balance the pH on the concentrated side, and Cl2 on the concentrated side. -The cell consisted of 250 mL of dilute electrolyte containing 2 M NH4Cl to match the volume of the electrolyte. Maintaining a pH balance allowed for ferric and ammonium ion gradients. In this way, crossover measurements were performed with a ferric ion concentration difference (delta C) of 1 mol / L. A potential of 0.5 V was applied to the cell, allowing the flow of ferric ions across the membrane to be detected as an oxidation current. The effective diffusivity of ferric ions, D eff was calculated as [(current*thickness) / (nF*delta C)].

[0061] Correspondingly, the same 6.5 cm 2 Hydraulic crossover tests were performed using this cell. Both reservoirs were filled with 150 mL of DI water and a few drops of HCl to maintain conductivity. The two reservoirs were positioned with a height difference of 78.7 cm (31 inches) between the streams, creating a pressure difference across the separator. The change in liquid volume in each reservoir over time was monitored, and the hydraulic permeability was calculated using Darcy's law. Hydraulic pressure and ferrous ion permeability values ​​were measured at room temperature.

[0062] result This concept was tested with two commercially available microporous separators: Celgard 2325, a three-ply material consisting of two layers of porous polypropylene sandwiching a single layer of porous polyethylene developed primarily for Li-ion batteries, and Daramic 175, a single layer of porous polyethylene also containing dispersed silica particles intended for use in lead-acid batteries. Relevant properties of these materials are listed in Table 1 below. Thickness measurements of composite separators with a thin coating of PVA showed that in both cases, the thickness was consistent with that of the porous support, indicating that the PVA was incorporated into the pores of the support and not forming a film on the surface. When a thicker coating was applied, the thickness of the composite increased. The properties of the composite separators were then compared to those of the uncoated material, and the results are shown in Tables 2–4 below.

[0063] Table 2a shows the properties of pure PVA membranes. While such membranes are possible, they involve certain structural / physical limitations and have proven more difficult to operate compared to composites. The ionic conductivity of the pure PVA film was very high (approximately 54% of the electrolyte conductivity), indicating high proton conductivity through the membrane. The thicker PVA separator maintained its mechanical integrity throughout the duration of the flow test.

[0064] It should be noted from Table 2b that the ionic conductivity of the coated composite Daramic was negligible compared to the uncoated one. This may be due to uneven filling of the larger pores. However, the coated Daramic maintained its ionic conductivity without a significant increase in area resistance. Furthermore, the ionic conductivity of the monolayer Celgard 2400 increased dramatically with coating thickness. This suggests that the PVA film fills the nanoscale pores, allowing enhanced wettability for electrolyte transport. We further speculate that the PVA not only captures the electrolyte and enables ionic conduction, but that the PVA in the support pores of these composites also improves the wettability of the support, allowing for higher levels of conductivity despite occupying a portion of the pore volume. With a thicker coating of PVA on either separator, the conductivity of the composite decreased, but remained acceptable for a flow battery.

[0065] From Table 3, it can be seen that with the Celgard separator, the PVA composite had a significantly lower diffusion rate of ferrous ions, improving ionic conductivity but impeding the desired diffusion of reactant ions. However, this result was not observed with the coated Daramic sample, which had a similar or slightly higher diffusivity for ferrous ions compared to the uncoated sample. This may indicate that the amount of PVA in the Daramic composite was not optimal, which may be related to the fact that both the pore size and pore volume of the Daramic material were larger than those of the Celgard separator.

[0066] Table 4 shows that incorporating PVA onto and into the porous support reduced the hydraulic permeability in both cases. This is a desirable result, even though the magnitude of the reduction in the first series of samples is small. For the second set of samples, where enough PVA was coated onto the support to result in a film of significant thickness, the hydraulic permeability was about an order of magnitude lower. This indicates that by optimizing the amount of PVA in the composite membrane, it should be possible to significantly reduce the hydraulic permeability while maintaining an acceptable level of conductivity. [Table 1] [Table 2] [Table 3] [Table 4] [Table 5] [Table 6]

[0067] The above description and examples identify various non-limiting embodiments of the present invention. Use of the term "flow battery" can also include hybrid flow batteries, if desired. References to positive and negative electrodes should be understood in the context of the discharge reaction, unless otherwise stated or clear from the context.

[0068] Any explicit or implicit range of values ​​set forth herein necessarily includes the disclosure of multiple intervals between the lower and upper limits of the range. As a non-limiting example, a range stated as 0.01 micrometers to 1.0 micrometers encompasses any combination of its upper boundary intervals, including any two selected from 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, and 0.9, even if a specific interval is not explicitly specified. Similarly, any value set forth in the examples shall be deemed to constitute a disclosed lower or upper limit when considered in combination with other values ​​from the same data set. Ratios and other limitations may also be inferred by comparing comparable values ​​disclosed herein.

[0069] All measurements should be read in context and with reference to the most common units and means of measurement as readily understood by one of ordinary skill in the art. Unless stated otherwise, all measurements are made at ambient temperature and pressure, and percentages refer to mass unless otherwise specified. Modifications may occur to those skilled in the art and able to make and use the invention. The disclosed embodiments are intended for illustrative purposes only and are not intended to limit the scope of the invention or the subject matter recited in the claims.

[0070] In one aspect of the present invention, (Section 1) a positive electrolyte flowing through the positive reaction chamber; a negative electrolyte flowing through the negative reaction chamber; a non-ionomeric membrane comprising a hydrogel physically separating the positive reaction chamber from the negative reaction chamber; wherein the membrane is ionically conductive but resistant to hydraulic crossover, and at least one of the positive and negative electrolytes comprises an aqueous solution.

[0071] In one aspect of the present invention, (Section 2) Item 1, the flow battery according to item 1, wherein the hydrogel comprises poly(vinyl alcohol).

[0072] In one aspect of the present invention, (Section 3) The polyvalent species is vanadium (2 + ) ions, vanadium (3 + ) ions, vanadium (4 + ) ions, vanadium (5 + Item 3. The flow battery according to item 1 or 2, comprising at least one of copper ions, copper(II) ...

[0073] In one aspect of the present invention, (Section 4) 4. The flow battery according to any one of items 1 to 3, wherein at least one of the positive and negative electrolyte solutions contains a polyvalent ion species.

[0074] In one aspect of the present invention, (Section 5) Item 5. The flow battery according to any one of items 1 to 4, wherein the hydrogel is selected from poly(vinyl alcohol), poly(acrylic acid), poly(ethylene glycol), and combinations thereof.

[0075] In one aspect of the present invention, (Section 6) Item 6. The flow battery according to any one of Items 1 to 5, wherein the hydrogel effectively seals pores in the polymer substrate.

[0076] In one aspect of the present invention, (Section 7) Item 7. The flow battery according to any one of Items 1 to 6, wherein the polymer substrate is microporous.

[0077] In one aspect of the present invention, (Section 8) Item 8. The flow battery according to any one of Items 1 to 7, wherein the polymer substrate is made of polyethylene, polypropylene, or a combination thereof.

[0078] In one aspect of the present invention, (Section 9) Item 9. The flow battery according to any one of items 1 to 8, wherein the polymer substrate has a void volume of 30% to 80%.

[0079] In one aspect of the present invention, (Section 10) Item 10. The flow battery according to any one of items 1 to 9, wherein the polymer substrate has a void volume of more than 80%.

[0080] In one aspect of the present invention, (Section 11) Item 11. The flow battery according to any one of items 1 to 10, wherein the membrane is made of a hydrogel.

[0081] In one aspect of the present invention, (Section 12) Item 12. The flow battery according to any one of items 1 to 11, wherein the positive and negative electrolyte solutions have a pH of less than 3.0.

[0082] In one aspect of the present invention, (Section 13) Item 13. The flow battery according to any one of items 1 to 12, wherein the hydrogel is at least one of a hydrolyzed hydrogel, a crystalline hydrogel, a semi-crystalline hydrogel, and a crosslinked hydrogel.

[0083] In one aspect of the present invention, (Section 14) 1. A method for producing an ionically conductive, hydraulic crossover resistant composite membrane, comprising: providing a substrate having a plurality of pores forming a void volume; impregnating the pores with a hydrogel; treating the hydrogel to form a non-ionomeric membrane; wherein the hydrogel is selected from poly(vinyl alcohol), poly(acrylic acid), poly(ethylene glycol), and combinations thereof.

[0084] In one aspect of the present invention, (Section 15) Item 15. The method of item 14, wherein treating the hydrogel comprises at least one of thermal crosslinking, chemical crosslinking, photochemical crosslinking, and hydrolysis of the hydrogel.

[0085] In one aspect of the present invention, (Section 16) Item 16. The method according to item 14 or 15, wherein the impregnating the pores includes at least one of film casting, roll-to-roll casting, infiltration, and dip coating.

[0086] In one aspect of the present invention, (Section 17) 17. The method according to any one of paragraphs 14 to 16, wherein the impregnating of the pores results in a layer of hydrogel deposited on at least one surface of the substrate with a thickness of 0.1 to 25 micrometers.

[0087] In one aspect of the present invention, (Section 18) Item 18. The method according to any one of items 14 to 17, further comprising at least one of washing the pores and removing air from the pores before impregnating the pores.

[0088] In one aspect of the present invention, (Section 19) Item 19. The method according to any one of items 14 to 18, further comprising submerging the substrate in alcohol.

[0089] In one aspect of the present invention, (Section 20) Item 20. The method according to any one of items 14 to 19, wherein the hydrogel comprises polyvinyl alcohol selected to have a purity of at least 99% by weight.

[0090] In one aspect of the present invention, (Section 21) Item 21. The method according to any one of items 14 to 20, wherein at least 95% by weight of the hydrogel is hydrolyzed.

[0091] In one aspect of the present invention, (Section 22) Item 22. The method according to any one of items 14 to 21, wherein multiple layers of hydrogel are successively impregnated onto the substrate.

Claims

[Claim 1] a positive electrolyte flowing through the positive reaction chamber; a negative electrolyte flowing through the negative reaction chamber; a non-ionomeric membrane comprising a hydrogel physically separating the positive reaction chamber from the negative reaction chamber; 10. A flow battery system wherein the membrane is ionically conductive but resistant to hydraulic crossover, and at least one of the positive and negative electrolytes comprises an aqueous solution.

Images