Membranes for enhancing rates of water dissociation and water formation

Abstract

Membranes for electrochemical technologies that operate at non-extreme acidic or basic conditions are described herein. The membrane can be a bipolar membrane having an anion-exchange layer, a cation-exchange layer, and a polymer layer containing a catalyst juxtaposed between the two layers. The catalyst can improve the rate for water dissociation and water formation at low applied bias thereby decreasing the overpotential and resistance for overall ion transport, especially when unequal pH values are used between the catholyte and anolyte.

Claims

1. A bipolar membrane (BPM) (100) for enhancing a rate of water dissociation or formation, said BPM (100) comprising a cation-exchange membrane (CEM) layer (102), an anion-exchange membrane (AEM) layer (104), and a catalytic polymer layer (106) interfacing and juxtaposed between the CEM layer (102) and the AEM layer (104), wherein the catalytic polymer layer (106) comprises a polymer backbone functionalized with catalytic buffer groups, the catalytic buffer groups comprising proton-transfer functional groups, proton-transfer coordination compounds, or combinations thereof, wherein the proton-transfer catalytic functional groups have a pk.sub.a in the range of about 6 to 8 and the proton-transfer catalytic coordination compounds have a pK.sub.f in the range of about 4 to 10, wherein the proton-transfer functional groups comprise amines, pyrazines pyrimidines, pyridazines, imidazoles, piperazines, morpholines, boronates, phosphonates, nucleobases, or amino acids, and wherein the proton-transfer coordination compounds comprise ligated cationic zinc, copper, iron, or lead.

2. The BPM (100) of claim 1, wherein the pK.sub.f is in the range of about 6 to 8.

3. The BPM (100) of claim 1, wherein a width of the catalytic polymer layer (106) ranges from about 1 nm to about 10 m.

4. The BPM (100) of claim 1, wherein the polymer backbone comprises two or more catalytic buffer groups.

5. A bipolar membrane (BPM) (100) for enhancing a rate of water dissociation or formation, said BPM (100) comprising a cation-exchange membrane (CEM) layer (102), an anion-exchange membrane (AEM) layer (104), and a catalytic polymer layer (106) interfacing and juxtaposed between the CEM layer (102) and the AEM layer (104), wherein the catalytic polymer layer (106) comprises a polymer backbone functionalized with catalytic buffer groups, the catalytic buffer groups comprising proton-transfer functional groups, proton-transfer coordination compounds, or combinations thereof, wherein the proton-transfer catalytic functional groups have a pK.sub.a in the range of 6 to 8 and the proton-transfer catalytic coordination compounds have a pK.sub.f in the range of 6 to 8.

6. The BPM (100) of claim 5, wherein the polymer backbone comprises two or more catalytic buffer groups.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING(S)

(1) The patent application or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

(2) The features and advantages of the present invention will become apparent from a consideration of the following detailed description presented in connection with the accompanying drawings in which:

(3) FIG. 1 shows non-limiting embodiments of a three-layer bipolar membrane (BPM) of the present invention. The BPM includes an acidic cation-exchange membrane (CEM), a catalyst-containing polymer layer, and an alkaline anion-exchange membrane (AEM). Exploded views are included to show the catalysts of the polymer layers.

(4) FIG. 2A is a schematic of three interfacial reactions that are critical to rapid hydrogen oxidation and water electrolysis (Reaction 1): Hydrogen Oxidation and Evolution Reaction (HOR/HER) (left), Oxygen Reduction and Evolution Reaction (ORR/OER) (right), and Water Dissociation and Formation Reaction (WDR/WFR) (center). The rate of current flow can be increased through the use of proton-transfer catalysts for WDR/WFR. The overall reaction is shown. The opposite of this reaction is water electrolysis.

(5) FIG. 2B is a schematic illustrating CO.sub.2 or N.sub.2 reduction via a membrane-electrode assembly.

(6) FIG. 3A shows a proposed mechanism for enhanced water dissociation. The mechanism makes an analogy to a semiconductor and trap states (Shockley-Read-Hall model for fast generation/recombination of charged species in a semiconductor).

(7) FIG. 3B shows an example reaction barrier diagram for enhanced water dissociation.

(8) FIG. 4A are prior art examples of catalyzed water dissociation: 1: Fumasep FBM (0.5 M Na.sub.2SO.sub.4); 2: FBM (pH 0 and pH 7); 3: FBM (pH 0 and pH 14); 4: SPEEK:QPPO 2-D Interface (0.5 M Na.sub.2SO.sub.4); 5: SPEEK:QPPO 3-D Interface (0.5 M Na.sub.2SO.sub.4).

(9) FIG. 4B shows a theoretical current density versus potential curves illustrating the effect of electric field and proton-transfer catalysis for various open-circuit potentials (E.sub.OC).

(10) FIGS. 5A-5B show non-limiting examples of potential current densities for water dissociation and water formation, calculated assuming a small region for water dissociation from parametric microkinetic simulations as a function of pH and functional group pK.sub.a at CEM|AEM, over the pH and pK.sub.a range of [0, 14] (FIG. 5A) and [5.3, 8.7] (FIG. 5B). The current densities are largest when pH=pK.sub.a=7.

(11) FIG. 6 are non-limiting examples of polymer backbones and Brnsted-Lowry conjugate acid functional groups that can be covalently bonded and positioned at CEM|AEM in BPMs.

(12) FIG. 7 depicts a four-electrode, six-compartment electrodialysis setup used to most accurately quantify the Faradaic yield for water dissociation from a middle three-layer BPM by measuring pH or ion concentration changes in the reference electrode compartments over time as current is passed.

(13) FIG. 8 shows a plot relating water dissociation to applied potential for various membranes. The x-values are largest at y-values close to zero (meaning low applied voltage bias) for the PPO materials, which are those made with phosphonate groups on them, shown as saturated phosphate in FIG. 7. The first two membranes are commercial Fumatech FBM and a made-in-house for a material's system that was recently reported to be highly active (sprinkled on graphene oxide as the middle region). Those do not work as well at low applied bias voltage.

(14) FIG. 9 shows a membrane-electrode assembly arrangement indicating the processes of H.sub.2O, H.sup.+, and OH.sup. transport and water dissociation that result in ionic current in a BPM.

(15) FIG. 10 shows a plot which illustrates the effects of the thickness of the middle polymer layer in a membrane-electrode assembly (MEA). This MEA has phosphonated PPO polymer (70% functionalized) as middle layer, shown as saturated phosphate in FIG. 7. The thinner layers seem to have approximately the same EFE effect as the case with no middle layer, but the currents are about 50 larger than without a catalyst. In this example, the pH was rather extreme across the membrane (thought to have a difference of 0 and 14 but could be as small as 0 and 11), but the conclusions are consistent with those shown in FIG. 8.

(16) FIG. 11A shows a plot of simulated exchange current density for overall water dissociation using two buffers, each at 0.5 M and at the indicated pK.sub.a values, and with the same results when these are pK.sub.f value, and pH over 10 nm thickness. Data for one buffer at 1 M concentration is shown as a white dashed diagonal line and is the same as that shown in FIG. 5A.

(17) FIG. 11B shows a plot of simulated exchange current density for overall water dissociation using two buffers, each at 0.5 M and at the indicated pK.sub.a values, and pH 7 over 10 nm thickness. Data for one buffer at 1 M concentration is shown as a black dashed diagonal line.

(18) FIG. 11C shows a plot of simulated exchange current density for overall water dissociation using 1 M buffer at the indicated pK.sub.a value, and with the same results when these are pK.sub.f values, and pH value over 10 nm thickness. Data for the condition of no buffer is shown as a black horizontal line at the bottom.

DETAILED DESCRIPTION OF THE INVENTION

(19) Following is a list of elements corresponding to a particular element referred to herein: 100 bipolar membrane (BPM) 102 cation-exchange membrane (CEM) 104 anion-exchange membrane (AEM) 106 catalyst-containing polymer layer

(20) As used herein, the term proton-transfer catalysis may refer to the catalysis of protonation, deprotonation, hydroxylation, or dihydroxylation (i.e. hydroxide transfer).

(21) It has generally been assumed that EFE water dissociation is strictly required in order to observe greatly enhanced rates of water dissociation, and that applied potential biases of >0.8 V are needed to dissociate water in BPMs, as shown in FIG. 4A, curve 1. However, the previous exceptions suggest that EFE water dissociation is not unique in its ability to enhance the rate of water dissociation. In one study, significant rates of water dissociation were observed at applied potential biases of <0.8V for an electrospun sulfonated poly(ether ether ketone) (SPEEK) and quaternized poly(phenylene oxide) (QPPO) BPM separating electrolytes at the same neutral pH values where the onset potential for rapid water dissociation was 0.7 V and the behavior was characteristic of an EFE process where the current increased exponentially with bias (curve 5). A BPM from WSI Technologies Inc. to separate electrolytes at the same neutral pH demonstrated that the onset potential for rapid water dissociation was 0.6 V and decreased with temperature, indicating a thermally activated process also indicative of an EFE process. Using Fumasep FBM (FumaTecH) with a pH difference of 7 across the BPM, rapid water dissociation occurred at 0.45 V and was characteristic of the CR model (curve 2).

(22) In FIG. 4B, hypothetical current density versus potential curves are shown for conditions where the pH difference across the BPM is 0, 7, and 14 units, as evidenced by the respective Nernstian E.sub.oc values of 0 V, 413 mV=759.2 mV, and 828 mV=1459.2 mV, where (In 10)RT/F=59.2 mV. This plot also depicts how increasing the local electric field, by decreasing the width of the space-charge region, changes the apparent onset potential of an exponential increase in the rate of water dissociation via an EFE process (horizontal shift curves). These plots also depict the effect of increased catalytic activity for water dissociation as proposed in the CR model, which upon introduction of small overpotential biases immediately enhances the rate of water dissociation (vertical shift in curves). Therefore, without wishing to limit the present invention to a particular theory or mechanism, by decreasing the width of the space-charge region and/or introducing a large concentration of catalytic proton-transfer functional groups and/or a large CEM|AEM interfacial area, it is believed that a current density for water dissociation consistent with state-of-the-art reversible fuel cells can be attained at record low overpotentials and for any pH difference.

(23) Referring now to FIG. 1, the present invention features catalytic polymer compositions, which may be used in bipolar membranes (BPM) (100) for enhancing the rate of water dissociation. In some embodiments, the BPM (100) may comprise a cation-exchange membrane (CEM) layer (102), an anion-exchange membrane (AEM) layer (104), and a catalytic polymer layer (106) interfacing and juxtaposed between the CEM layer (102) and the AEM layer (104). The catalytic polymer layer (106) comprises the catalytic polymer composition. In some embodiments, the catalytic polymer composition can have a polymer backbone having proton-transfer functional groups, proton-transfer coordination compounds, or combinations thereof. In some embodiments, the catalytic polymer may be synthesized by ring-opening polymerization, substitution reactions, bromination and subsequent substitution with a functional group, SN coupling with amines and sulfonyl halides (e.g. sulfonyl fluoride and sulfonyl chloride), CN coupling with amines and carboxylic acids or carboxylic acid derivatives (e.g. acyl halides, acyl azides, acylimidazoles, anhydrides, and active esters), click chemistry (e.g. azide and alkyne cycloaddition), cross-coupling such as palladium cross coupling (e.g. Suzuki. Miyaura, Suzuki-Miyaura, Hiyama, and Negishi reactions), electrophilic substitutions (e.g. Friedel-Crafts acylation reactions), nucleophilic substitutions (e.g. Williamson ether synthesis), free radical polymerization, metathesis, reversible addition-fragmentation chain-transfer (RAFT) polymerization, atom-transfer radical polymerization (ATRP), living cationic polymerization, or living anionic polymerization.

(24) In alternative embodiments, the proton-transfer functional groups or proton-transfer coordination compounds may be used to catalyze transfer of protons to or from water without being covalently bound to a polymer backbone in a BPM. As one non-limiting example, the proton-transfer functional groups or proton-transfer coordination compounds may be sandwiched between the CEM and AEM layers of a BPM as small molecules, unbound cations, or as functionalized particles or nanoparticles. In another embodiment, a surface of either the AEM layer or the CEM layer may be functionalized with the proton-transfer functional groups or proton-transfer coordination compounds. In another embodiment, the proton-transfer functional groups or proton-transfer coordination compounds may be bound to a polymer backbone and used as either a single membrane without CEM or AEM layers or in combination with only one of the CEM or AEM layers.

(25) In some embodiments, the catalytic polymer compositions or catalyst-containing polymer layers of the present invention may also be used for CO.sub.2 capture via CO.sub.2 hydration whereby one of the following reactions is made to take place: either Reaction 11 after water dissociation as described above, or Reaction 12 after Reaction 6, the latter of which is the reported mechanism for carbonic anhydrase:

(26) $\begin{array}{cc}{\mathrm{OH}}^{-}\left(\mathrm{aq}\right)+{\mathrm{CO}}_2\stackrel{k_8}{\underset{k_{-8}}{\underset{}{.fwdarw.}}}{\mathrm{HCO}}_3^{-}\left(\mathrm{aq}\right)& \left(11\right)\\ {\left[M\left(\mathrm{OH}\right)L_n\right]}^{\left(m-1\right)+}\left(\mathrm{aq}\right)+{\mathrm{CO}}_2+H_{2}O\stackrel{k_6}{\underset{k_{-6}}{\underset{}{.fwdarw.}}}{\left[M\left(H_{2}O\right)L_n\right]}^{m+}\left(\mathrm{aq}\right)+{\mathrm{HCO}}_3^{-}\left(\mathrm{aq}\right)& \left(12\right)\end{array}$
In these cases, OH.sup. is ultimately transferred to a CO.sub.2 substrate via the intermediate catalyst, and not into water. When Reaction 11 is operative, the rate of the overall CO.sub.2 hydration reaction to form HCO.sub.3.sup.(aq) is rate-limited by water dissociation, meaning that the optimal catalyst for water dissociation described above is the same here. When Reaction 12 is operative, the optimal catalyst has a pK that is halfway between the pK.sub.a of H.sup.+(aq), which is 0, and the pK.sub.a of H.sub.2CO.sub.3 with or without CO.sub.2 hydration, which is 3.5-6.5. If thermodynamically favorable bimolecular reactions are diffusion-limited, then the mechanism that proceeds via Reaction 12 can occur at a rate that is 1-10,000 times faster than the optimal rate when the mechanism proceeds via Reaction 11. In some embodiment, the catalytic polymer compositions of the present invention may catalyze either hydration or dehydration of CO.sub.2.

(27) In some embodiments, the BPM may consist of polymers that present non-extreme acidic or basic conditions when hydrated. In other embodiments, the BPM may be immersed in aqueous electrolyte solution where infiltration with salt species results in non-extreme acidic or basic local conditions. In either case, this allows for a BPM that presents non-extreme acidic or basic local conditions at the catholyte or anolyte, which may be more compatible with the desired electrochemical reactions in terms of their activity, selectivity, and overall stability.

(28) A non-limiting embodiment of the catalyst-containing polymer layer is depicted in FIG. 1, which show the hydrated pathways in a near-neutral-pH material. This material may comprise a copolymer of cationic quaternary ammonium groups and anionic phosphonate groups with a pK.sub.a7 to catalyze water dissociation and to buffer protons in order to maintain desired near-neutral pH conditions while rapidly conducting H.sup.+ and OH.sup..

(29) According to other embodiments, the present invention provides a method of fabricating a bipolar membrane (BPM) (100) for enhancing the rate of water dissociation. In one embodiment, the method may comprise synthesizing a catalytic polymer comprising a polymer backbone with proton-transfer functional groups, proton-transfer coordination compounds, or combinations thereof, adding a layer of said catalytic polymer (106) onto a surface of a cation-exchange membrane (CEM) layer (102), adding a layer of said catalytic polymer (106) onto a surface of an anion-exchange membrane (AEM) layer (104), stacking the CEM layer (102) and the AEM layer (104) such that the surfaces with the catalytic polymer layers (106) are interfacing each other, and heat pressing the stack of the CEM layer (102) and the AEM layer (104) to form a single structure having the unified catalytic polymer layer (106) disposed in between the CEM layer (102) and the AEM layer (104), thereby producing the BPM (100).

(30) In some embodiments, the catalytic polymer layer (106) may be added to the surface of the CEM layer (102) or the AEM layer (104) by drop-coating, blade-coating, or spin-coating the catalytic polymer directly onto the CEM layer (102) or the AEM layer (104). In some other embodiments, the catalytic polymer layer (106) may be added to the surface of the CEM layer (102) or the AEM layer (104) in a layer-by-layer (LbL) assembly to allow for control of polymer interlayer thickness. In one embodiment, the LbL assembly may comprise polymerization of the catalytic polymer directly on the surface of the CEM layer (102) or the AEM layer (104). In another embodiment, the LbL assembly may comprise dip-coating or spray-coating the catalytic polymer onto the CEM layer (102) or the AEM layer (104).

(31) In some embodiments, the polymer backbone may comprise a vinyl copolymer, an acrylate copolymer, poly(benzimidazole), poly(phenylene oxide), polynorbornene, polysulfone, polystyrene, or polyethylene, and related polymers. However, any suitable polymer may be used as the polymer backbone without deviating from the scope of the present invention. In other embodiments, the catalytic proton-transfer functional groups may comprise Brnsted-Lowry conjugate acid/base functional groups. Examples of Brnsted-Lowry conjugate acid/base functional groups include, but are not limited to, amines, pyridines, pyrazines, pyrimidines, pyridazines, imidazoles, piperazines, morpholines, boronates, phosphonates, carboxylates, nucleobases such as guanine, cytosine, adenine, thymine, uracil, amino acids, and related compounds. In one example, pyrimidines have low and high pK.sub.as (1 and 10). Guanine may be used to incorporate a functional group with several pK.sub.as (3, 9, and 12). It is to be understood that other Brnsted-Lowry conjugate acid/base functional groups may be used in accordance with the present invention.

(32) In one embodiment, Lewis acidic cations in the coordination compounds may comprise Zn, Cu, Fe, or Pb. For example, the Lewis acidic cations may be Zn(II), Cu(II), Fe(II), or Pb(II), and may be chelated to enhance stability and localization. Other Lewis acidic cations, formally chelated or not, may be used without deviating from the scope of the present invention.

(33) In some embodiments, the proton-transfer functional groups have a pK.sub.a in the range of about 4 to 10. In other embodiments, the proton-transfer functional groups have a pK.sub.a in the range of about 5 to 9. In preferred embodiments, the proton-transfer functional groups have a pK.sub.a in the range of about 6 to 8. Most preferably, the proton-transfer functional groups have a pK.sub.a of about 7 or close to 7.

(34) In some embodiments, the proton-transfer coordination compounds have a pK.sub.f value for binding OH.sup. in the range of about 4 to 10. In other embodiments, the proton-transfer coordination compounds have a pK.sub.f in the range of about 5 to 9. In preferred embodiments, the proton-transfer coordination compounds have a pK.sub.f in the range of about 6 to 8. Most preferably, the proton-transfer coordination compounds have a pK.sub.f of about 7 or close to 7.

(35) In preferred embodiments, the pH of the catalytic polymer is about 7 or close to 7. In some embodiments, the pH of the catalytic polymer ranges from about 4 to 10. In other embodiments, the pH of the catalytic polymer ranges from about 5 to 9. In preferred embodiments, the pH of the catalytic polymer ranges from about 6 to 8. Without wishing to limit the present invention, the current density of the catalytic polymer is largest when the pH and pK.sub.a or pK.sub.f is about 7 or close to 7, as shown in FIGS. 5A-5B. In some embodiments, the current density for water dissociation may be attained for any pH difference. In some desirable embodiments, the pH on one side of the membrane is slightly acidic and strong base is used on the other side of the membrane, partially because the oxygen evolution reaction (OER) is usually most effective at inexpensive catalysts around a pH of 14. As a non-limiting example, the pH may be 14 on one side of the membrane and 7 for the catholyte on the other side of the membrane such that the local conditions are benign. As another non-limiting example, pH values of 9 and 14 may be used together because 9 is slightly more acidic than the second pK.sub.a of H.sub.2CO.sub.3, which means that little CO.sub.3.sup.2 (carbonate) forms. This is important because carbonate is not very soluble in the presence of most salts or cationic impurities and thus will precipitate out as a solid and clog the membrane, slowing its ability to transport species.

(36) Again, without wishing to limit the present invention to a particular theory or mechanism, it is believed that for any solution/solvent, the pK.sub.a of the proton-transfer functional group catalysts, the pK.sub.f of the proton-transfer coordination compound catalysts, and the pH of the solvated membrane are preferably at or near the midpoint of the pK.sub.a of the protonated solvent molecules and the pK.sub.a of the deprotonated solvent molecules. At this pH, pK.sub.a, and/or pK.sub.f the current density of the catalytic polymer is largest.

(37) In some embodiments, the width of the catalytic polymer layer (106) may range from about 1 nm to about 10 m. For instance, the width may be about 1-10 nm, 10 nm-100 nm, 100 nm-1 m, or 1-10 m. In yet other embodiments, the width of the catalytic polymer layer (106) may be greater than 10 m, for example, 20 m.

(38) In other embodiments, a concentration of the proton-transfer functional groups and/or proton-transfer coordination compounds range from about 0.01 M to about 10 M. For example, the concentration of the proton-transfer catalytic functional groups may be about 0.01-0.1 M, 0.1-1 M, 1-5 M, or 5-10 M. In yet other embodiments, the concentration of the catalytic proton-transfer functional groups may be greater than 10 M, for example, 20 M.

(39) Electrolyzers and fuel cells operate at most at 1-10 A/cm.sup.2. Thus, a membrane must be able to pass that current density as a rate of ion transport. The data reported herein show that using optimal proton-transfer catalytic functional groups at 1 M concentration and over the thickness of 10 nm, microscopically reversible rates as current densities of >0.1 A/cm.sup.2 are possible at effectively zero bias and thus zero wasted voltage across the membrane. Surface roughening, increasing the concentration of catalytic groups, and increasing the thickness of the catalytic polymer layer will increase the rate by these same scaling factors, unless the reaction is limited by another process such as mass transport of water into or out of the membrane, which can be overcome by using thinner catalytic polymer layers. To achieve 1-10 A/cm.sup.2, small biases will be required but can be less than biases required for these same current densities at state-of-the-art electrocatalysts for the O.sub.2 evolution reaction and the O.sub.2 reduction reaction. Therefore, in some other embodiments, an applied potential bias to dissociate water at 1-10 A/cm.sup.2 is less than 0.5 V. As non-limiting examples, the applied potential bias required to dissociate water may be less than 0.3 V or less than 0.1 V. As additional non-limiting examples, a sufficient rate for water dissociation may be 50-10.sup.5 M/s, which as a current density over 10 nm thickness equals 0.005-10 A/cm.sup.2. In some embodiments, a large current density for water dissociation and formation (>5 mA/cm.sup.2) may be attained for any pH difference

(40) In conjunction with any of the aforementioned embodiments, the present invention may feature an electrochemical device comprising a bipolar membrane (100). The BPM of the electrochemical device may be according to any of the BPMs described herein. Examples of the electrochemical device include, but are not limited to, fuel cells, electrolyzers, CO.sub.2 capture devices, and solar fuels devices.

(41) In other embodiments, the present invention further provides a method for enhancing the rate of water dissociation in an electrochemical process. The method may comprise providing a bipolar membrane (BPM) (100) according to any of the BPMs described herein, and utilizing the BPM (100) in the electrochemical process. Without wishing to be bound to a particular theory or mechanism, the BPM (100) can increase the water dissociation rate and lower the applied potential bias required for rapid rates of water dissociation. A rapid rate of water dissociation is considered to be any rate that is at least two orders-of-magnitude faster than the uncatalyzed rate of 10.sup.3 M/s, which as a current density over 10 nm thickness equals 1 A/cm.sup.2.

(42) Referring to FIG. 5A, proton-transfer Brnsted-Lowry functional groups with pK.sub.a7 can be most effective at catalyzing water dissociation. Therefore, protonated/deprotonated amines, pyridines, pyrazines, pyrimidines, pyridazines, imidazoles, piperazines, morpholines, boronates, phosphonates, carboxylates, or nucleobases or amino acids, and related compounds can serve as proton-transfer catalysts for water dissociation (AH/A.sup. in Reactions 4 and 5). For example, deprotonated phosphonic acid groups can serve as optimal proton-transfer catalyst for water dissociation given that their more basic protic group has a pK.sub.a7. However, in order for this protonation state to best serve the role of AH/A.sup. and result in rapid rates of water dissociation, approximately half of the groups should be singly deprotonated and the other half of the groups should be doubly deprotonated to a state that would exist at pH7.

(43) To achieve this, quaternary ammonium groups or other cationic groups may be combined in a 3:1 to 3:2 molar ratio with phosphonic acid groups and drive their ion pairing through phosphonic acid deprotonation and dialysis of the counterions, such as H.sup.+ and Cl.sup., as shown in FIG. 1. These materials can be fabricated by two methods: mixing two different polymers or ionomers, or synthesizing copolymers containing both functional groups at various molar ratios. These molar ratios may exceed 3:1 because not all functional groups may ion pair to the desired ratio. Electrochemical measurements may be conducted using three-layer BPMs with the phosphonate-containing polymers sandwiched between a commercial CEM, like Nafion 111, Nafion 112, Nafion 117, or related, and a commercial AEM, like Neosepta AHA, Sustanion X37, Sustanion XA9, or related, as described below.

(44) FIG. 8 shows that (Basic PPO-Pi and Acidic PPO-Pi), where +XA9 means that an anion-exchange ionomer with quaternary ammoniums were mixed in so that the pH was more basicwhich is a decent condition for these samplesand +Nafion means that a solution of cation-exchange material with acidic sulfonic acids were mixed in so that the pH was more acidicwhich is the worst condition for these samples. PPONH2 means a polymer with an ammonium/amine functional group whose pK.sub.a is not as close to 7 as the PPO-Pi ones. This worked rather wellthe x-intercept is the current density of interest that correlates to the modeled data current densities. Additionally, the data in FIG. 8 only uses the PPO-Pi with varied thickness. In both of these figures an AEM and CEM was used to sandwich this layer. Also, PPO-Pi means phosphonated PPO polymer whose structure was shown in FIG. 6. There are several phosphonated groups shown in that figure and this Pi one is the saturated phosphate one.

(45) While Brnsted-Lowry conjugate bases like phosphonates are effective at stabilizing H.sup.+ during the process of water dissociation, water dissociation can also be catalyzed through stabilization of OH.sup. using Lewis acidic cations ([M(H.sub.2O)L.sub.n].sup.m+/[M(OH)L.sub.n].sup.(m1)+ in Reactions 6 and 7). For example, analogous to pK.sub.a values for Brnsted-Lowry acids, Lewis acidic hydrated Zn(II) ions have a pK.sub.f value of 5 for binding OH.sup. and forming Zn(II)OH species, which stabilize an OH.sup. intermediate during the reaction of carbon dioxide and water to form bicarbonate and H. As such, Zn(II) and other ions with pK.sub.f7 (e.g. Cu(II) with pK.sub.f=6.5 and Pb(II) with pK.sub.f=6.4) may be incorporated into polymers through complexation with functional groups such as imidazoles and polypyridines that are covalently bound to various polymer backbones. For example, vinyl and acrylate copolymers can be synthesized, functionalized polyethylene and polynorbornene polymers can be synthesized by ring-opening metathesis polymerization, and functionalized poly(phenylene oxide) can be synthesized by Friedel-Crafts acylation of poly(phenylene oxide).

(46) Commercial BPMs utilize a variety of interfacial polymer layers at CEM|AEM to catalyze water dissociation, such as cationic microbeads in polystyrene-co-polyvinylbenzoyl chloride, polysulfone containing tertiary and quaternary amines, a polyacrylic acid/polyvinylpyridine salt complex, and hyperbranched polyesters containing amines or hydroxyl groups. Referring to FIG. 6, combinations of specific Brnsted-Lowry functional groups, coordination compounds, and polymer backbones may be used at CEM|AEM. These polymers, copolymers, and composite polymers at CEM|AEM may be incorporated by casting them onto OEMs and AEMs and then heat pressing them to form a single structure.

(47) In a non-limiting embodiment, the CEM may be Nafion 112. In another non-limiting embodiment, the AEM may be Neosepta AHA or Sustanion X37/XA9. There are several techniques that can be employed to functionalize CEM|AEM with the catalytic polymers and form a BPM. Examples of which include, but are not limited to, drop-coating, blade-coating, or spin-coating the catalytic polymers directly on to the CEM, followed by heat pressing an AEM on top with a variable-temperature hydraulic membrane press. Processing conditions, such as deposition rate, pressure, time, and temperature, may be varied to obtain materials with substantial mechanical robustness and adhesion so that electrochemical measurements can be conducted. This can also be achieved by covalently linking thin films via chemical modification of their interfaces or by synthesizing the polymers from one monolithic polymer backbone structure.

(48) In some embodiments, it is desirable to use methods that afford precise control over catalyst polymer layer thickness. Thus, a layer-by-layer (LbL) assembly of polymers onto OEMs may allow for finer control over polymer thickness. In one embodiment, the LbL technique may include dip coating. In another embodiment, the LbL technique may comprise spraying to isolate one side of the membrane and increase the throughput of generating LbL membranes. In other embodiment, gas-phase chemistry may be utilized to polymerize thin films on membrane surfaces. As an example, OEMs are placed inside a vacuum chamber containing a crucible with an oxidation catalyst (i.e. FeCl.sub.3) that is heated so that the catalyst sublimates. Then, a stream of monomer is introduced through the chamber and upon mixing with the catalyst, polymerization occurs on the substrate CEM surface. These techniques can afford conformal coating of polymer films with precisely controlled thicknesses.

Example

(49) The following is a non-limiting example of the present invention. It is to be understood that said example is not intended to limit the present invention in any way. Equivalents or substitutes are within the scope of the present invention.

(50) Microkinetic Models.

(51) In order to determine functional group concentrations and pK.sub.a values that are desired at CEM|AEM for effective proton-transfer catalysis of the water dissociation reaction, 0-D and 1-D kinetic models may be implemented using the COMSOL Reaction Engineering module and the COMSOL finite-element solver to approximate solutions to the multiple coupled partial differential equations or 0-D analytical solutions to the Henderson-Hasselbalch equation and kinetic rate laws with knowledge of the rate constants for the proton-transfer reactions. Unique to the 1-D models, an electric field can be imposed to simulate EFE water dissociation. More realistic kinetic rate constants and CEM|AEM properties may be gleaned from the atomistic and coarse-grained simulations to couple the two modeling regimes.

(52) The inventors have performed preliminary parametric microkinetic simulations to determine the effects of pH and pK.sub.a of catalytic proton-transfer groups on ionic current density, and where the results are the same when instead of pK.sub.a values, pK.sub.f values for coordination compounds are used. The simulations were carried out assuming 1 M catalyst groups over a 10 nm thick region, diffusion-limited second-order proton-transfer reactions, and the literature value for the zeroth-order reaction for water dissociation of 10.sup.3 M/s. It was found that for any pK.sub.a value, the maximum rate of water dissociation occurs at pH 7 and that the exchange current density is largest when the pK.sub.a of the catalytic groups is 7, resulting in an exchange current density of 10-100 mA cm.sup.2 (FIG. 5A-5B). Exchange current densities represent the rate of water dissociation and water formation at equilibrium and therefore, at zero applied bias overpotential. This exchange current density for water dissociation by a CR mechanism is more than an order-of-magnitude larger than any values observed experimentally under modest polarization biases of <0.5 V. Also notable is that the rate of water dissociation is affected little by reasonable variations in reorganization energy and electronic coupling, which means that a simplified Eyring/Arrhenius-type expression for the thermally activated H.sup.+ transfer processes predicts similar behavior to that observed from a Marcus-type model.

(53) Of great practical importance is that the rate of water dissociation increases linearly with concentration of catalyst, width of the catalyst-containing region, and CEM|AEM surface area, suggesting that 0.1 A/cm.sup.2 current densities relevant to practical reversible fuel cells will be possible at little-to-no overpotential and separating electrolytes at any pH value, therefore, enabling proton-coupled electron-transfer catalysts that are stable at any pH value to be used. For example, considering a concentration of proton-transfer catalysts with pK.sub.a=7 increased from 1 M to 10 M and over a thickness of 100 nm instead of 10 nm, exchange current densities may be increase by a factor of 100 to >1 A/cm.sup.2 at pH=7.

(54) These enhanced rates of water dissociation are purely due to the CR model. While this seems at odds with the EFE mechanism that based on prior literature is postulated to dominate experimentally observed water dissociation behavior, a simple analysis suggests a purely CR mechanism can result in rapid rates of water dissociation at very low overpotentials. Rapid rates of water dissociation require a large density/concentration of proton-transfer catalysts with pK.sub.a=7, which is the midpoint between the pK.sub.a of H.sup.+(aq) and H.sub.2O, and operation at pH 7 so that there is an equal concentration of each form of the proton-transfer catalyst, and thus equal capture cross section, for the AH and A.sup. proton-transfer functional group catalysts in Reactions 4 and 5 or [M(H.sub.2O)L.sub.n].sup.m+ and [M(OH)L.sub.n].sup.(m1)+ proton-transfer coordination compound catalysts in Reactions 6 and 7. Results from this analysis are consistent with the data in FIGS. 5A-5B that were obtained using the microkinetic models, thus further validating them.

(55) Macroscale Poisson-Nernst-Planck Simulations.

(56) Information regarding the identity and optimal properties of functional groups for water dissociation gleaned from the microkinetic simulations may be incorporated into a macroscale BPM model. This model can include coupling of the microkinetic modeling framework directly into a 1-D and 2-D Poisson-Nernst-Planck model to compute the behavior of BPMs under different acid, base, and salt electrolyte conditions.

(57) Preliminary results obtained from the Poisson-Nernst-Planck model, that did not include microkinetic processes, indicate that differences in in the concentration of H.sup.+ and H.sup. across CEM|AEM are indefinitely stable when the contacting phase for the BPM is deionized water. Significant ion crossover was shown to occur when the contacting phases were modeled to be aqueous electrolytes, and the ion crossover rates were found to be related to the concentrations of H.sup.+ and OH.sup., as predicted by simple Donnan theory. To the inventors' knowledge, this is the first BPM model that takes into consideration the effects of ions other than H.sup.+ and OH.sup.. Results from the model also indicate that an ion-concentration-dependent electrostatic potential barrier exists at the BPM/solution interfaces, and a large potential barrier at CEM|AEM with a magnitude that is independent of the concentration of ions in the liquid electrolyte but dependent on the rate of ion crossover. Also, by incorporating properties of commercial BPMs, a space-charge region that is 2 nm wide is observed at CEM|AEM. The general Poisson-Nernst-Planck model can also output the spatial distribution of protonation and deprotonation reactions.

(58) Experimental Electrochemical Measurements.

(59) Fuel cells are generally evaluated as membrane-electrode assemblies (MEAs) with gaseous humidified H.sub.2 and O.sub.2 feeds. One benefit of MEAs is that deionized water can be used in the feed stream resulting in zero ion crossover at open circuit in the absence of an ionic current. Without confounding ion crossover processes, MEAs enables accurate quantification of extremely low current densities from uncatalyzed water dissociation so that the effects of catalyst layers can be accurately compared. However, one drawback of MEAs driving H.sub.2 and O.sub.2 redox chemistry is that they do not easily allow selective measurement of solely the electrochemical performance of the BPM, and thus the rate of water dissociation, because current-voltage behavior is convoluted by overpotentials from the redox reactions. This can almost entirely be overcome by driving fast isoenergetic redox reactions at the electrodes. Therefore, by flowing humidified forming gas (5% H.sub.2 in N.sub.2) or pure hydrogen gas past large-area platinized Pt electrodes in both the anode and cathode chambers, they function as reversible hydrogen electrodes (RHEs) with well-defined potentials, zero difference in free energy across the membrane, and low overpotentials for H.sub.2 oxidation and H.sup.+ reduction redox reactions. This can allow for the performance of the BPM and rate of water dissociation to be accurately measured.

(60) Four-electrode measurements using BPMs wetted by near-neutral pH electrolyte solutions have several advantages. For example, through use of a four-electrode, six-compartment cell as shown in FIG. 7, electrochemical measurements afford facile determination of the Faradaic yield and rate of water dissociation in comparison to unintentional ion crossover. This is performed by passing a known amount of charge through the system and comparing that value to the magnitude of the change in pH and/or ion concentration as measured by ICP-MS or ICP-OES. Preliminary data obtained for a phosphonate containing polymer and copolymer as part of a three-layer BPM indicated a >35% Faradaic yield for water dissociation at 100 mV overpotential bias when wetted by near-neutral pH electrolyte solutions. To the inventors' knowledge, this is the largest Faradaic yield for water dissociation at the lowest overpotential ever reported for a BPM in contact with near-neutral pH electrolyte solutions.

(61) Four-electrode measurements also enable the inner Galvani potential difference across the BPM to be measured directly using two nominally identical reference electrodes. The inner Galvani potential difference is the total electric potential drop across the membrane, which in combination with measurements of the thickness of the proton-transfer catalyst layer, can be used to accurately determine the strength of the electric field, which is an important parameter in the EFE model and all proposed numerical models. Thus, two-electrode MEA electrochemical measurements of BPMs in contact with humidified forming gas or pure hydrogen gas may be performed to quantify small rates of water dissociation, as well as four-electrode six-compartment electrochemical measurements (FIG. 8) of BPMs wetted by near-neutral pH electrolyte solutions to quantify Faradaic yields for water dissociation and electric potential drops.

(62) As used herein, the term about refers to plus or minus 10% of the referenced number.

(63) Although there has been shown and described the preferred embodiment of the present invention, it will be readily apparent to those skilled in the art that modifications may be made thereto which do not exceed the scope of the appended claims. Therefore, the scope of the invention is only to be limited by the following claims. In some embodiments, the figures presented in this patent application are drawn to scale, including the angles, ratios of dimensions, etc. In some embodiments, the figures are representative only and the claims are not limited by the dimensions of the figures. In some embodiments, descriptions of the inventions described herein using the phrase comprising includes embodiments that could be described as consisting essentially of or consisting of, and as such the written description requirement for claiming one or more embodiments of the present invention using the phrase consisting essentially of or consisting of is met.

(64) The reference numbers recited in the below claims are solely for ease of examination of this patent application, and are exemplary, and are not intended in any way to limit the scope of the claims to the particular features having the corresponding reference numbers in the drawings.