Low crossover, high durability graphene reinforced proton exchange membrane

Abstract

The present disclosure provides methods for making a composite ionomer-graphene membrane and systems containing such membranes. The method includes providing a graphene layer; subjecting the graphene layer to UV-Ozone treatment to generate a plurality of defects distributed across a surface of the graphene layer, wherein plurality of defects corresponds to a defect density of about 1x1011 cm-2 to about 5x1012 cm-2 and wherein the UV-ozone treatment comprises generating ultraviolet light in an oxygen-containing atmosphere, and wherein molecular oxygen in the oxygen-containing atmosphere absorbs the ultraviolet light to generate oxygen radicals and / or molecular ozone that reacts with the graphene layer to generate the defects; and laminating the treated graphene layer to an ionomer membrane to generate a composite ionomer-graphene membrane.

Description

LOW CROSSOVER, HIGH DURABILITY GRAPHENE REINFORCED PROTON EXCHANGE MEMBRANECROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of and priority to U.S. Provisional Application No. 63 / 626,247, filed on January 29, 2024, and titled “LOW CROSSOVER, HIGH DURABILITY GRAPHENE REINFORCED PROTON EXCHANGE MEMBRANE,” the content of which is herein incorporated by reference in its entirety for all purposes.BACKGROUND

[0002] Hydrogen fuel cells, particularly those based on proton exchange membrane fuel cells (PEMFCs), represent an opportunity for society to transition away from diesel and fossil fuels for heavy machinery and trucking. However, a number of technical challenges prevent current adoption of the technology, namely with the durability of the proton exchange membrane (PEM). A major issue that impacts durability is the hydrogen crossover phenomenon, when molecular hydrogen (H2) fuel diffuses through the membrane without first dissociating into protons, causing mixed potentials and the formation of peroxides that chemically degrade the membrane. As such, several strategies have been explored to extend the lifetime of the membranes, such as the addition of a recombination layer, and the use of 2D material barrier layers. The addition of hygroscopic oxides has been explored with moderate success in reducing crossover, but greatly impacts the proton conductivity due to the proton insulating nature of these materials.

[0003] The 2D material approach stands out as particularly promising-the atomically thin nature of the film suggests that conductivity penalties may be minimal to none, while the intrinsic pore size of the lattice allows for highly selective transport. Moehring et. al. demonstrated the role of defects in crossover mitigation through tailored CVD growth but show increases in proton resistances of over 30 mOhm-cm.

[0004] Reduction of the conductivity penalty, or ideally elimination, while retaining high selectivity is paramount to enabling adoption of the 2D material strategy. Despite the advances in 2D material barrier layers, there are advancements that are still needed. The present disclosure satisfies this need and offers other advantages as well.BRIEF SUMMARY

[0005] Fuel cells are electrochemical devices that convert the chemical energy of hydrogen and oxygen into electricity in an environmentally friendly manner. Some hydrogen fuel cells, particularly those based on proton exchange membrane fuel cells (PEMFCs), may be limited by the hydrogen crossover phenomenon, wherein molecular hydrogen fuel diffuses through the membrane without first dissociating into protons.

[0006] This disclosure provides the use of UV-Ozone induced defects in graphene to eliminate the conductivity penalty commonly associated with adding traditional crossover mitigation strategies in an ionomer membrane. The disclosed methods demonstrate an increase in hydrogen / proton selectivity, a decrease in EE crossover, without impact to current output. Furthermore, disclosed membranes degrade slower than state of the art membranes (e.g., Gore Select membranes) and show no loss in performance after a lOOh accelerated stress test (AST).

[0007] As such, in one embodiment, the present disclosure provides a method for making a composite ionomer-graphene membrane, comprising: providing a graphene layer; subjecting the graphene layer to UV-Ozone treatment to generate a plurality of defects distributed across a surface of the graphene layer, wherein plurality of defects corresponds to a defect density of 1x1011cm’2to 5x1012cm’2and wherein the UV-Ozone treatment comprises generating ultraviolet light in an oxygen-containing atmosphere, and wherein molecular oxygen in the oxygen-containing atmosphere absorbs the ultraviolet light to generate oxygen radicals and / or molecular ozone that reacts with the graphene layer to generate the defects; and laminating the treated graphene layer to an ionomer membrane to generate a composite ionomer-graphene membrane.

[0008] In another embodiment, the present disclosure provides a system comprising: a graphene layer, wherein the graphene layer includes a plurality of defects distributed across a surface of the graphene layer, wherein the graphene layer exhibits a defect density of about IxlO11cm’2to about 5x1012cm’2; and an ionomer membrane, wherein the graphene layer is in contact with and supported by the ionomer membrane.

[0009] These and other objects, aspects and embodiments will become more apparent with the detailed description and figures that follow.BRIEF DESCRIPTION OF THE DRAWINGS

[0010] FIG. l(a)-(c) illustrate fabrication process for ozonated graphene PEMs. (a) Commercially obtained uniform graphene on copper was treated with UV-ozone at 100 °C for a variable amount of time. As exposure time increases, a larger number of effusive pores are created, in contrast to selective pores, (b) Post-exposure, the graphene / copper stack is hot pressed to a GORE Select PEM (8 pm thickness) and the copper is etched away in 1 M ammonium persulfate (APS). The graphene is then encapsulated by another layer of the GORE Select PEM to create a membrane sandwich, (c) The stack is then hot pressed with gas diffusion electrodes (GDE) to form the final test package.

[0011] FIG. 2 shows a SEM image of graphene transfer onto membrane via hot press method.

[0012] FIG. 3 illustrates a cross-sectional view of a membrane electrode assembly (MEA) showing gas diffusion electrodes (GDE) containing 0.3 mg / cm2platinum catalyst on Vulcan and the two sandwiched membranes with a graphene interface.

[0013] FIG. 4(a)-(d) illustrate ozonated graphene characterization. FIG. 4(a) D / G ratios and associated defect densities as a function of exposure time. FIG. 4(b) Representative Raman peaks of each ozonation time studied. FIG. 4(c-f) Raman maps (10 pm x 10 pm, step size 0.25 pm) of D / G ratios of graphene surfaces to illustrate homogeneity of ozonation. FIG. 4(g) Histogram illustrating D / G ratios of ozonated and pristine samples. FIG. 4(h) Pristine graphene HA-ADF STEM image showing pristine periodic graphene lattice. FIG. 4(i) 2 min 45 sec ozone exposed graphene on a SiN grid showing the early formation of point defects, selective pores. The range of pore sizes expands with ozonation time, as new selective pores form while older pores expand into effusive pores.

[0014] FIG. 5 illustrates 5-minute exposed graphene TEM image. Image! analysis suggests pores with an average of 8.65 nm diameter, with 8.2% of the area covered in effusive pores.

[0015] FIG. 6 illustrates 15-minute exposed graphene TEM image. Image! analysis suggests pores with an average of 9.97 nm diameter, with 16.1% of the area covered in effusive pores.

[0016] FIG. 7 illustrates pore diameter by ozone exposure time. As ozone exposure increases, pore size increases, as does variability due to the nucleation and expansion of existing defects.

[0017] FIG. 8(a)-(d) illustrate XPS data of graphene Cis by ozonation time. FIG. 8(a) Only C-C and C-0 bonding is seen from oxygen binding to vacancies and defects in pristine graphene. FIG. 8(b) The 5 minute ozonated sample shows the growth of a C=O peak at 288 eV and a decrease in C-C bonding. FIG. 8(c) The 10 minute ozonated sample shows an increase in both C-0 and C=Obonding and a decrease in C-C signal, suggesting the loss of carbon in the ozonation process and the formation of large-scale pores. FIG. 8(d) The 15-minute sample shows a further decrease in C- C bonding but also a decrease in C=O bonding, suggesting that existing pores expanded, leaving fewer locations for C=O bonding to occur.

[0018] FIG. 9(a)-(d) illustrate fuel cell test data. FIG. 9(a) Polarization curves at 80 °C, 150 kPagauge (250 kPaabs) of 5 cm2MEA packages tested with different ozonated graphene layers. Air and H2 were supplied at flowrates of 0.56 1pm and 0.14 1pm respectively. FIG. 9(b) Current Interrupt (iR) measured cell resistance by ozonation time. FIG. 9(c) Fuel cell power density curves. FIG. 9(d) Crossover obtained by linear sweep voltammetry (LSV). H2 and N2 were both supplied at 0.125 1pm to the anode and cathode respectively. The GORE Select sandwich is referred to as 8 pm|8 pm to indicate thickness of the stack - all membranes tested were nominally 16 pm.

[0019] FIG. 10 illustrates open circuit voltage (OCV) vs irradiation time of embodiment of the disclosure.

[0020] FIG. 1 l(a)-(b) illustrate (a) cell potential versus current density and (b) current density versus cell potential.

[0021] FIG. 12(a)-(b) illustrate (a) cell potential versus current density or (b) current density versus cell potential.

[0022] FIG. 13 illustrates power density at 100% RH, 80 °C by voltage. The 10- and 15-minute ozonated sample show comparable and superior performance to the baseline. The 15-min sample shows a 22% increase in power output at 0.7V, a significant increase in efficiency.

[0023] FIG. 14(a)-(d) illustrate a multifaceted view of fuel cell Membrane characterization, (a) Ohmic conductivity, calculated as the inverse of cell resistance, to crossover ratio (mS / cm) / (mA / cm2). FIG. 14(b) Overpotential at 10 mA / cm2from OCV. FIG. 14(c) Proton and molecular hydrogen flux as a function of irradiation time, (d) Flux based selectivity calculation, an alternative lens to the conduct! vity / crossover ratio in FIG. 14(a).

[0024] FIG. 15 (a)-(f) illustrate accelerated stress testing (AST) for the 5-minute, 10-minute and 15-minute ozonated graphene and the baseline. FIG.15(a) 100 h OCV degradation data held at 90 °C, 30% RH (DOE MEA Chemical Stability Test). FIG.15(b) Degradation rate extracted from linear region of the curve (past 10 hours). FIG.15(c) Polarization curve data before and after the AST. FIG.15(d) Cell resistance before and after stress testing. FIG.15(e) Crossover measurements of ozonated samples before and after AST. FIG.15(f) Comparison of this work (10-minute and 15-minute samples chosen to represent conditions without performance decay) to other similar works in performance, showing decay rate against current density.

[0025] FIG. 16 (a)-(f) illustrate the structure of ozonated graphene nanopores. The nanopores are irregular-shaped, and their sizes are referenced by the largest dimension, defined as the maximum distance between nanopore atoms FIG.16 (a) 0 nm nanopore FIG.16 (b) 2 nm nanopore FIG.16 (c) 4 nm nanopore FIG.16 (d) 6 nm nanopore. FIG.16 (e) 8 nm nanopore FIG.16 (f) 10 nm nanopore.

[0026] FIG. 17 (a)-(d) illustrate proton and hydrogen flux for ozonated graphene nanopores of size 0 nm, 2 nm, 4 nm, 6 nm, 8 nm and 10 nm. FIG.17(a) Proton flux (as hydronium) and hydrogen flux are computed from the slope of the cumulative flux. FIG.17(b) Flux-based selectivity of water transport over hydrogen transport (left axis) and selectivity of hydronium over hydrogen (right axis) in ozonated graphene pores with increasing pore size. FIG.17(c) Water flux as a function of pore size, showing that water transport is throttled at smaller pore sizes relative to baseline. FIG.17(d) Water permeation plotted against selectivity shows the 3 distinct regimes observed in fuel cell testing.

[0027] FIG. 18 (a)-(c) illustrate cumulative flux of FIG.18(a) water and FIG.18(b) hydronium ion, and (c) hydrogen molecules w.r.t simulated time for systems with ozonated nanopore of size 10 nm, 8 nm, 6 nm, 4 nm, 2 nm and 0 nm. Cumulative flux is computed from counting the total number of molecules translocated across the ozonated graphene pore.

[0028] FIG. 19 (a)-(c) illustrate Flux rate of FIG.19(a) Hydronium ion, FIG.19(b) water, and FIG.19(c) Hydrogen. The flux rate was computed from the slope of the cumulative flux plot shown in FIG. 18.

[0029] FIG. 20 (a)-(b) illustrate a comparison of 16 pm Ozonated Graphene / GORE Select membranes against 25 pm NR-211 membranes. Polarization curves obtained at 250 kPaabs, 80 °C, 100% RH. FIG. 20 (a) Polarization performance is improved relative to NR-211 due to the thinner nature of the ozonated graphene membranes, enabling faster proton transport. FIG. 20 (b) Crossover in the ozonated membranes with GORE Select is much lower than the NR-211, making these membranes much more selective.

[0030] FIG. 21(a)-(b) illustrate tensile testing results. FIG. 21(a) Representative Stress-Strain curves for all sample types tested. FIG. 21(b) Average modulus and maximum stress of all samples tested. All samples resulted in consistent moduli and maximum stress values, noting a lack of statistically significant deviations.

[0031] FIG. 22 illustrates a thermogravimetric analysis (TGA). All samples show similar degradation profiles and thermal stability. Differences in curves obtained are attributed to measurement noise. Key metrics such as the TD5 (temperature at which 5 wt% mass loss occurs) and weight fractions at 300 °C were consistent between all samples, providing a coefficient of variation (CV = standard variation / average x 100%) of 4.7% and 1.3%, respectively. Since these values fall within the 2-5% range, the samples can be considered statistically consistent.DETAILED DESCRIPTION

[0032] The present disclosure provides methods and systems of 2D material barrier membranes to sieve between EE and EEO , which methods are highly scalable. The membranes are highly durable fuel cell membranes that represent a significant upgrade over current state of the art membranes. Being atomically thin, the barrier materials provide high separation capacity and excellent selectivity.

[0033] In certain embodiments, the present disclosure provides methods and systems for precise control of nucleated defects in graphene lattices, which allow for ease of transport for protons while minimizing hydrogen transport. Advantageously, the use of the UV-ozone exposure approaches creates pores or defects in graphene, which is capable of high selectivity between kinetic diameters of EEC) (0.269 nm) and EE (0.290 nm), and which affords highly precise control of pore formation.

[0034] In certain aspects, the disclosure provides scalable methods for nucleating selective defects into CVD graphene lattices by UV-Ozone exposure to minimize the proton conductivity detriment while maintaining lower hydrogen permeance at proton exchange membrane fuel cells (PEMFC) operational conditions. As described herein, the methods and systems provide a reduction or complete elimination of the conductivity penalty by introducing pores such as effusive pores into the graphene lattice with a longer exposure time, which leads to the development of a membrane that increases current output while reducing the hydrogen permeation by an upshift of the fuel cell polarization curve, leading to an increase of selectivity.

[0035] In certain embodiments, the present disclosure provides a method to generate a plurality of defects distributed across a surface of a graphene layer. Once such a graphene layer is treated, the method includes laminating the treated graphene layer to an ionomer membrane to generate a composite ionomer-graphene membrane.

[0036] In one embodiment, the present disclosure provides a method to generate a composite ionomer-graphene membrane, which method comprises:providing a graphene layer; subjecting the graphene layer to UV-Ozone treatment to generate a plurality of defects distributed across a surface of the graphene layer, wherein the plurality of defects corresponds to a defect density of about IxlO11cm’2to about 5xl012cm’2and wherein the UV-Ozone treatment comprises generating ultraviolet light in an oxygen-containing atmosphere, and wherein molecular oxygen in the oxygen-containing atmosphere absorbs the ultraviolet light to generate oxygen radicals and / or molecular ozone that reacts with the graphene layer to generate the defects; and laminating the treated graphene layer to an ionomer membrane to generate a composite ionomer-graphene membrane.

[0037] A “composite ionomer graphene membrane” refers to a membrane made by combining an ionomer (a polymer with ionic exchange capabilities) with graphene, a highly conductive carbon material, to facilitate efficient proton transport such as within a fuel cell, enhancing its overall performance. The ionomer provides the primary proton conduction pathway, while the added graphene barrier sheet enhances conductivity, mechanical strength, and allows for ease of transport of protons while minimizing hydrogen transport. Graphene has excellent mechanical strength, is optically transparent, has low toxicity and has high chemical and thermal stability, which provides an edge over other barrier materials.

[0038] Utilization of a UV-O3 source can generate defects in graphene through C-C bond scission (i.e., etching) using oxygen radicals and / or molecular ozone. In certain aspects, the plurality of defects includes selective defects, effusive defects or a combination thereof. As used herein, the term “selective defects” refer to a specific type of structural imperfection or pores within the graphene lattice where a targeted modification or disruption occurs at a particular location, often along a line or at a specific point. The term “effusive defects” refers to a more widespread, distribution of defects across the graphene, wherein defects or pores are not localized to specific areas.

[0039] The methods provide a defect density of about IxlO11cm’2to about 5xl012cm’2and wherein the UV-Ozone treatment comprises generating ultraviolet light in an oxygen-containing atmosphere. It is believed that the molecular oxygen in the oxygen-containing atmosphere absorbs ultraviolet light to generate oxygen radicals and / or molecular ozone that thereafter reacts with the graphene layer to generate the defects. The defect density is between about 100 billion pores or defects to about 5 trillion pores per square centimeter (defects / cm2). For example, the defect density is about 100 billion pores, 125 billion pores, 150 billion pores, 175 billion pores, 200 billion pores, 225 billion pores, 250 billion pores, 275 billion pores, 300 billion pores, 325 billionpores, 350 billion pores, 375 billion pores, 400 billion pores, 425 billion pores, 450 billion pores, 475 billion pores, 500 billion pores, 550 billion pores, 600 billion pores, 650 billion pores, 700 billion pores, 750 billion pores, 800 billion pores, 850 billion pores, 900 billion pores, 950 billion pores, 1 trillion, 2 trillion, 3 trillion, 4 trillion, and / or 5 trillion per square centimeter (defects / cm2). In certain aspects, the defect density is about IxlO11cm’2to about 5xl0ncm’2(about 100 billion pores to about 500 billion pores).

[0040] FIG. 1 (a)-(c) illustrates a fabrication process for ozonated graphene polymer electrolyte membrane PEMs. In certain aspects, a commercially obtained full area coverage graphene on copper is treated with UV-Ozone at about 100 °C for a variable amount of time. As exposure time increases, a larger number of effusive pores are created. The UV-ozone treatment increases pore density up to about 500 billion / cm2. FIG. 1(a) shows the method comprising UV-ozone treatment increases pore density, with pore expansion and nucleation by increasing the length of exposure times to treatment.

[0041] Although CVD graphene is available commercially, a skilled person in the art knows how to synthesize 2D graphene, using CVD processes. A two-dimensional graphene layer, for example, as synthesized using CVD processes is suitable for use in the methods and systems of the present disclosure. The graphene-based sheet or layer may include a single monolayer of graphene. Graphene suitable for use in the methods and systems of the present disclosure may also be synthesized other ways, for example by exfoliation, solution, or epitaxy methods.

[0042] In certain aspects, the graphene layer comprises an atomically thick two-dimensional (2D) layer comprising sp2-hybridized carbon.

[0043] After exposure to the treatment, as shown in FIG. 1(b), the graphene / copper stack is hot pressed to a membrane such as a polymer electrolyte membrane (PEM). As a non-limiting example, GORE Select PEM (e.g., 8 pm thickness) is suitable for use. Next, the copper is etched away in IM ammonium persulfate (APS). The graphene is then encapsulated by another layer of GORE Select PEM to create a membrane sandwich, having the 2D graphene sandwiched between the two membranes. Thereafter, the membrane sandwich is hot pressed with gas diffusion electrodes (GDE, FIG. 1(c)) to form a MEA package.

[0044] In certain aspects, the graphene layer is subjected to the UV-Ozone treatment for a duration of time from about 1 minute to about 20 minutes, such as 1-10 min, 1-15 min, or about 1- 20 minutes, or about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 minutes. In certain instances, the time can be less than 1 minute and longer than 20 minutes.

[0045] In certain aspects, the ozone exposures are conducted at a temperature of about 75°C to about 125°C (e.g., 100°C) to enable sufficient pore formation. The exposure time to UV-ozone is varied to allow for the nucleation of ozone to the surface of the graphene. In certain instances, the temperature can be varied with time exposure. The exposure can begin with high temperature and cool ramped down, or lower temperature and increased in a ramped-up procedure. All such temperature variations are within the scope of the disclosure. Additional exposure times allow for more pore nucleation and pore expansion, which decreases the selectivity of the graphene, but allows for higher proton transport. The use of full area coverage graphene allows for simple control of selectivity and proton flux, making this approach highly scalable to industry.

[0046] In certain aspects, UV-Ozone treatment comprises generating oxygen radicals and / or molecular ozone at a temperature of from about 75°C to about 125°C, such as 80°C, 85°C, 90°C, 95°C, 100°C, 105°C, 110°C, 115°C, or 120°C.

[0047] In certain aspects, providing the graphene layer includes providing the graphene layer on a substrate and wherein the method further comprises removing the graphene layer from the substrate after the UV-Ozone treatment.

[0048] As shown in FIG. 2, the claimed methods lead to full area transfer onto the membrane, as verified by SEM imaging, showing a clear contrast between the coated and uncoated edge of the membrane. In this illustrative example, the choice of commercial PEM was GORE Select.Although this particular selection of membrane is non-limiting, it is noted that this PEM is the current automotive state of the art. Those of skill in the art will know of other PEMs that are also suitable for the present methods. This membrane is ultrathin (8 pm), but as a result, has higher crossover than for example, Nafion®, but has the benefit of higher current output.Advantageously, the use of ozonated graphene generates a low crossover and high current output MEA.

[0049] Electrodes may be attached to one or both sides of the proton conducting ionomergraphene membrane. The electrodes can be made of a metal such as platinum or palladium though any electrode material can be used. Alternatively, electrodes, such as Pt, can be deposited on one or both sides of the membrane by evaporation techniques. In certain aspects, the membrane of the disclosure includes a pair of electrodes, one on either side of the membrane and in direct electrical contact with the membrane in order to allow the passage of current through the membrane. In another aspect, the membrane includes an electrode on the “output” side of the membrane i.e., the side of the membrane from which protons are emitted when in use.

[0050] A second electrode can be in direct contact with the other “input” side of the membrane, or the electrode can be in contact with a solution which is itself in contact with the “input” side of the membrane. In this case the electrode is in electrical contact with the membrane but is not directly bound to the membrane. The solution provides a source of protons. In another aspect, the electrodes do not form part of the membrane structure at all but are each in contact with solutions or gases on the two respective sides of the membrane. Electrical contact is formed due to both of the solutions or gases being in contact with the respective sides of the membrane.

[0051] In certain aspects, the ionomer membrane is a first ionomer membrane, and wherein the method further comprises encapsulating, surrounding or sandwiching the porous graphene between the first ionomer membrane and a second ionomer membrane to generate a composite ionomergraphene membrane sandwich.

[0052] In certain aspects, the ionomer membrane comprises one of more of a sulfonated tetrafluoroethylene polymer, a tetrafluoroethylene polymer, an expanded polytetrafluoroethylene polymer, or a composite of tetrafluoroethylene polymer and a sulfonated tetrafluoroethylene polymer.

[0053] Suitable commercially available ion-conducting membranes used in PEMFCs and are generally formed from perfluorinated sulphonic acid (PFSA) ionomers and the membranes formed from these ionomers are sold under the tradenames GORE-SELECT® Membrane, Nafion® (E.I. DuPont de Nemours and Co.), Aciplex® (Asahi Kasei), Aquivion® (Solvay) and Flemion® (Asahi Glass KK).

[0054] In certain aspects, the ionomer membrane has a thickness from about 5 pm to about 120 pm, or from about 7 pm to about 50 pm, or from about 8 pm to about 30 pm.

[0055] In certain aspects, the ionomer membrane comprises or corresponds to a proton- or ionexchange membrane.

[0056] In certain aspects, the process further includes the step of adding an electrode to the membrane. In a further aspect, the process includes adding a pair of electrodes, one to either side of the membrane. The electrodes can be added mechanically or by evaporation of a layer of metal onto the membrane.

[0057] In certain aspects, the method further comprises positioning the composite ionomergraphene membrane between a first gas diffusion electrode and a second gas diffusion electrode.

[0058] In certain aspects, the method further comprises contacting the first gas diffusion electrode with a fuel containing gas and contacting with the second gas diffusion electrode togenerate a voltage and / or current between the first gas diffusion electrode and the second gas diffusion electrode.

[0059] In certain aspects, the method further comprises contacting the first gas diffusion electrode with a source gas and applying a voltage between the first gas diffusion electrode and the second gas diffusion electrode to generate a current.

[0060] A membrane electrode assembly (MEA) includes a polymer electrolyte membrane of the present invention and two electrodes respectively acting as a cathode and an anode, which are attached to both sides of the polymer electrolyte membrane.

[0061] As a non-limiting example, FIG. 3 shows a cross-sectional view illustrating a MEA package 300 including a first gas diffusion electrode GDE 302 containing 0.3 mg / cm2platinum catalyst on Vulcan, and a similar second GDE 307 on the other side of the sandwiched membrane. As disclosed, there is a first membrane 318 of the sandwich membrane, a second membrane 325 of the sandwich and the graphene interface 336 having a defect density of about 100 billion pores or defects to about 500 billion pores or defects / cm2. The sandwich membrane is situated between the first electrode (e.g., cathode) 302 and the second electrode (e.g., anode) 307. The membrane is configured to conduct protons while providing an electronic insulator and a barrier between fuel cell reactants (e.g., oxygen and hydrogen gas). The graphene layer comprising a plurality of defects aids in the barrier function. The ionomer polymer may be based on any of the polymers already described such as a sulfonated tetrafluoroethylene polymer, a tetrafluoroethylene polymer, an expanded polytetrafluoroethylene polymer, or a composite of tetrafluoroethylene polymer and a sulfonated tetrafluoroethylene polymer.

[0062] In certain aspects, the ionomer is provided on both sides of the 2-D graphene material, and the ionomer may be the same or different. The ionomer may encapsulate, surround or sandwich the 2-D graphene material. It is also possible for one or both of the ionomer membrane layers to be associated with an optional substrate and / or to be provided in the form of a single structure combining the functions of the ionomer and substrate in a single entity. In some aspects, the ionomer such as Gore Select, is sufficient to provide support for the 2-D graphene and consequently the combination of the ionomer and 2-D graphene is employed in that form in a fuel cell.

[0063] Advantageously, the incorporation of pores shows a decrease in hydrogen crossover while exhibiting an increase in current outputs. For example, the incorporation of graphene barrier of this disclosure is able to increase selectivity by 18%-20% or even more in comparison to current state of the art GORE Select membranes. Further, durability is improved by up to 40% after theincorporation of graphene, and performance after durability testing increases and does not degrade at all.

[0064] Turning now to FIG. 4(a)-(i), the ozonated graphene characterization is shown. FIG. 4(a) illustrates the D / G ratio, which represents the defectivity of the graphene lattice; as the ratio increases, the defect density also increases, as calculated by an analysis of the Raman D&G peaks:(2).

[0065] In the second equation, TID is the defect density of the graphene and 2 is the Raman excitation wavelength (e.g., 532 nm). As the exposure time is increased from 0 minutes / pristine graphene to 15 minutes, the D / G peak ratio (hashed bars / dotted bars) increases from 0.185 to 1.667, corresponding to a magnitude increase in defect density. FIG. 4(b) illustrates a typical Raman spectrum for the ozonated graphene, showing the growth of the D peak and the D’ peak relative to the G peak. Furthermore, the 2D peak is seen to lose intensity as well, further indicating increased nucleation of defects. The ozone exposure is seen to be highly uniform as seen from the Raman maps in FIG. 4(c)-(h), with D / G histograms showing a slight left skew in ozonated samples. ADF-STEM imaging illustrates the etching of the pristine graphene lattice (FIG. 4(h)) with a 2 min 45 sec exposure on a TEM grid (FIG. 4(i)), showing the formation of point defects - the birth of effusive pores. This illustrates the trend in exposure time against pore diameter - the range of pore sizes is also seen to expand with ozonation time. This effect is attributed to new pores nucleating and forming while previously formed pores expand. As seen in FIG. 5 and FIG. 6, the increase in ozonation time to suspended graphene on a TEM grid results in a higher effusive pore density, from 8% to 16% pore area coverage, and an average pore diameter increase from 8.6 nm to 9.9 nm in a 5-minute and 15-minute exposed sample (FIG. 7). This analysis skews towards the effusive regime, as these effusive pores cannot be imaged at the same scale as selective pores. As such, a similar trend is expected in the pore sizes and densities on the polymer membrane, although etch rates on the TEM grid are likely higher.

[0066] The chemical environment of the graphene was also studied by X-ray photoelectron spectroscopy (XPS) (FIG. 8 (a)-(d)). As ozonation time increases, a corresponding loss of C-C bonding signal at 248.8 eV is observed, and an increase in the C=O signal at 288 eV, as describedin previous studies of oxidized graphene. As ozonation time increases to 15 minutes, the C=O signal decreases, suggesting that the primary mechanism occurring is pore expansion rather than new pore formation, corroborating the data from the Raman and TEM analysis.

[0067] FIG. 9(a)-(d) shows fuel cell test data. For example, FIG. 9(a) are polarization curves at 80°C, 150 kPaabs of 5 cm2MEA packages tested with different ozonated graphene layers. Air and H2 were supplied at flowrates of 0.56 1pm and 0.14 1pm, respectively. FIG. 9(b) shows current interrupt (iR) measured cell resistance by ozonation time. FIG. 9(c) shows fuel cell power density curves and FIG. 9(d) illustrates crossover obtained by linear sweep voltammetry (LSV). H2 and N2 were both supplied at 0.125 1pm to the anode and cathode respectively. The GORE Select sandwich is referred to as 8 pm |8 pm in the figures to indicate thickness of the stack. All membranes tested were nominally 16 pm.

[0068] A skilled person will recognize that the disclosure encompasses thinner and thicker ionomer membranes, for example, from about 1 pm to about 150 pm. The use of 8 pm is exemplary, but other membrane sizes are suitable (e.g., 1 pm, 2 pm, 3 pm, 4 pm, 5 pm, 6 pm, 7 pm, 8 pm 9 pm, 10 pm, etc.), with the graphene layer sandwiched between them, so that the overall thickness is optimum for high ionic conductivity with low crossover. In certain aspects, there is a combination of large pores sizes and small pores sizes, and this balance can be used in various membrane thicknesses and is especially useful for thinner membranes.

[0069] These assembled MEA packages with the ozonated graphene were tested at beginning of life with polarization curves and crossover measurements. In the fuel cell typical operating range of 0.6V-0.7V, it is clear from FIG. 9(a) that pristine graphene grown by CVD causes significant performance degradation, showing a mere 520 mA / cm2at 0.6V as compared to the GORE Select sandwich, which outputted 1048 mA / cm2. Pristine graphene shows a 50.4% drop in current output and a 37% increase in cell resistance and wattage (FIG. 9(b), (c), but a 44% mitigation in crossover (FIG. 5(d)). This drastic drop in current density makes pristine graphene alone impractical for practical fuel cell applications and is attributed to the graphene film making fuel access to the catalyst heterogeneous, showing a large initial activation loss.

[0070] Ozonated graphene, however, shows an immediate uptick in current density, with the 5, 10 and 15-minute ozonated graphene samples outputting averages of 847 mA / cm2, 1014 mA / cm2and 1047 mA / cm2respectively. Importantly, all 3 samples show a reduction in the cell resistance, from pristine graphene’s 81.4 mOhm-cm2(a 37% increase) to the 15-minute ozonated sample’s 62.9 mOhm-cm2(a 5.4% increase). This implies a real conductance increase of graphene from 45.0 S / cm2to 270 S / cm2, an increase in conductance of 6x. The crossover was observed to be lessmitigated as ozonation time increased (3.81 mA / cm2, 3.99 mA / cm2, and 4.41 mA / cm2for the 5, 10 and 15-minute samples respectively), suggesting a sacrifice in ohmic selectivity for permeation, a tradeoff well established in size selective membrane research. Advantageously, all treated graphene samples illustrated a decrease in crossover compared to the baseline value of 5.2 mA / cm2. This reduction can be attributed to angstrom scale selectivity, as well as a blocking of H2 diffusion in pristine and selective regions.

[0071] Importantly, the 15-minute samples showed an increase in current in the 0.9V-0.6V range, attributed to the shifting of the polarization curve upwards due to crossover mitigation in combination with a lack of impact to ohmic resistance. If operating the fuel cell at 0.7V, the 15- minute sample shows a 19% increase on top of the baseline. It is important to note that the wattage curves (FIG. 9(c)) for the ozonated samples exceed the baseline in the low current regimemass transport losses dominate earlier and earlier as ozonation time increases, thus dropping the maximal wattage achievable in the high current regime by 50 mW / cm2, or 6% of the baseline maximum. However, for many fuel cell applications, such high-power density is not needed, and cells are most often run between 0.6 V-0.7 V. The OCV trend further supports the assertion that ozonated graphene can reshow an increase in OCV compared to the baseline (FIG. 10). Similarly, at 70% RH and 30% RH, the 10 and 15-minute samples exceed the baseline current output, while retaining lower crossover (FIG. 1 l(a)-(b), FIG. 12(a)-(b)). FIG. 11 illustrate (a) cell potential versus current density or (b) current density versus cell potential. FIG. 12 illustrate (a) cell potential versus current density or (b) current density versus cell potential. Regardless, as shown in FIG 13, the ozonated graphene samples increase the wattage at 0.6 V as compared to the pristine graphene by 27.3%, making this a facile method to achieve higher current densities when needed at still lower crossover.

[0072] Advantageously, the ozonated graphene samples increase the maximum wattage as compared to the pristine graphene by 27.3%, making this a facile method to achieve higher current densities when needed at still lower crossover.

[0073] The disclosure also provides a system comprising a membrane made by the foregoing methods.

[0074] In one embodiment, the present disclosure provides a system comprising: a graphene layer, wherein the graphene layer includes a plurality of defects distributed across a surface of the graphene layer, wherein the graphene layer exhibits a defect density of about IxlO11cm'2to about 5xl012cm'2; andan ionomer membrane, wherein the graphene layer is in contact with and supported by the ionomer membrane.

[0075] In certain aspects, commercially obtained graphene can be exposed to UV-ozone treatment. The use of a UV-ozone tool instead of gaseous ozone provides a more controllable method of pore development and expansion, by exposure times on the order of minutes rather than seconds, and the use of temperature as a modulation knob.

[0076] In certain aspects, the graphene layer comprises an atomically thick two-dimensional layer comprising sp2-hybridized carbon.

[0077] In certain aspects, the plurality of defects includes both selective defects and effusive defects.

[0078] In certain aspects, the ionomer membrane comprises a member selected from the group consisting of a sulfonated tetrafluoroethylene polymer, a tetrafluoroethylene polymer, an expanded polytetrafluoroethylene polymer, or a composite of tetrafluoroethylene polymer and a sulfonated tetrafluoroethylene polymer.

[0079] In certain aspects, the ionomer membrane has a thickness from about 1 pm to about 150 pm, from about 5 pm to about 120 pm, or from about 7 pm to about 50 pm, or from about 8 pm to about 30 pm.

[0080] In certain aspects, the ionomer membrane is a proton- or ion-exchange membrane.

[0081] In certain aspects, the ionomer membrane comprises a bottom ionomer membrane and wherein the system further comprises a top ionomer membrane in contact with the graphene layer on a side of the graphene layer opposite to the bottom ionomer membrane.

[0082] In certain aspects, the system comprises a composite structure including the bottom ionomer membrane, the top ionomer membrane, and the graphene layer between the bottom ionomer membrane and the top ionomer membrane.

[0083] In certain aspects, the system further comprises a first gas diffusion electrode and a second gas diffusion electrode, wherein the graphene layer and the ionomer membrane are positioned between the first gas diffusion electrode and the second gas diffusion electrode.

[0084] In certain aspects, the system corresponds to a fuel cell, wherein a fuel containing gas is positioned in contact with the first gas diffusion electrode, and wherein an oxidizing agent containing gas is positioned in contact with the second gas diffusion electrode.

[0085] In certain aspects, the system corresponds to an electrolysis cell, wherein a source gas is positioned in contact with the first gas diffusion electrode for reducing the source gas at the first gas diffusion electrode upon application of a voltage between the first gas diffusion electrode and the second gas diffusion electrode.

[0086] A person of ordinary skill in the art will recognize and appreciate other applications of this technology, which extend outside hydrogen fuel cells. For example, other applications include other non-hydrogen fuel cells such as direct methanol fuel cells, membrane-based separations, membrane electrolysis such as CO2 or proton exchange membrane electrolysis (PEMWE), electrochemical membrane reactors, and highly durable battery separators.

[0087] The disclosure shows that the introduction of ozonated graphene into PEM membranes aids in both performance and durability, providing a compelling alternative to traditional approaches to extend fuel cell life. These methods are scalable, as UV-ozone exposure is facile and inexpensive, while large-area graphene is readily available. The results show that defect- engineered 2D materials can prove to be synergistically integrated into fuel cell stacks without significant infrastructural change, while enhancing all-around performance. The performance at 0.6 V-0.7 V of ozonated graphene and the durability, in particular, make the methods and systems well suited to automotive and heavy machinery as an improvement to the current state-of-the-art.EXAMPLES

[0088] The following are additional details regarding the methods of UV-ozonation as well as graphene characterization and data collection.UV-Ozonation Process

[0089] Commercially obtained CVD graphene on copper (Grolltex, Carlsbad, CA) was placed in a benchtop UV cleaner (Novascan, PSD Thermal Series UV-Ozone Cleaner). The UV light is generated from a low-pressure mercury lamp at 184.9 and 253.7 nm wavelengths about 13 cm from the hot plate where samples are placed. The 184.9 nm light generates oxygen radicals which recombine with molecular oxygen to form ozone. Additionally, molecular ozone is hit by the 253.7 nm light which generates additional oxygen radicals to attack the surface of the samples. The hot plate was allowed to come to 100 °C to enhance the pore formation / oxidative etching process. After the plate reached the desired temperature, the graphene / Cu samples were placed on the hot plate and exposed to the UV-Ozone process for a determined amount of time.Graphene Characterization

[0090] Graphene was characterized on both copper and after transfer to a SiCh substrate after ozonation by Raman spectroscopy (Renishaw inVia micro-Raman system). When on copper, the graphene was characterized by a 432 nm excitation wavelength, while on SiCh after transfer, an excitation wavelength of 532 nm was used. XPS was conducted with a VersaProbe4 XPS tool (Physical Electronics, Chanhassen, MN, USA). Peak labeling was conducted in accordance with previous studies of oxidized graphene. SEM imaging was conducted with an FEI Quanta scanning electron microscope.

[0091] The transfer process for graphene onto SiCh for further characterization involved spin coating 500 nm of PMMA onto the copper / graphene surface after ozonation. The stack was floated on the surface of a beaker with 1 M ammonium persulfate (APS) to etch away the copper. The floating graphene / PMMA stack was fished out using an SiCh wafer, then dried for 24 hours. The PMMA was then reflowed after a bake of 5 minutes at 150 °C. After the drying process to improve the adhesion and the reflow bake, the stack was immersed in acetone to dissolve the PMMA. After 48 hours, the stack was fished out of the acetone and rinsed with IPA and dried.HA-ADF STEM Characterization

[0092] Commercial CVD-grown graphene on Cu foil (Grolltex, Carlsbad, CA, USA) was spin- coated with an 8 wt% solution of poly(methyl methacrylate) in anisole (Sigma-Aldrich, St. Louis, MO, USA). Once the PMMA layer had hardened the copper was etched away using a IM Ammonium Persulfate (Sigma-Aldrich, St. Louis, MO, USA) solution. The graphene, supported by the PMMA layer, was rinsed in deionized water and transferred onto silicon nitride TEM grids (Norcada, Edmonton, AB, Canada). Once dry, the TEM grids were immersed in acetone to dissolve the PMMA layer, leaving behind only monolayer graphene on the TEM grids. Next, the samples were heated to in a tube furnace at 350°C and subjected to 200 seem of 3% EE / 97% Ar to remove residual PMMA particles and remove adsorbed solvents. Finally, the TEM samples were stored under vacuum overnight to guarantee cleanliness before ozone treatment and HA-ADF STEM imaging.

[0093] HA-ADF STEM imaging was carried out using a probe-corrected JEOL neoARM (JEOL, Akishima, Japan) using 80 kV accelerating voltage and a 27 milliradian convergence semiangle. The electron probe was corrected using ASCOR (CEOS GmbH, Heidelberg, Germany) to achieve a coherent flat-phase envelope greater than the convergence angle of 27 milliradians. Cumulative electron dose for atomic-resolution graphene images was on the order of 105- 106e’ / A2, depending on magnification and probe dwell time.Membrane Sandwich Fabrication

[0094] GORE Select membranes were hot pressed to graphene on copper is hot pressed at 500 lbf / cm2of force on a Carver hydraulic press at 120 °C for 3 minutes. The stack was floated on IM APS to allow the copper to etch away. This graphene / membrane stack was fished out and checked for electrical conductivity of the surface with a multimeter to ensure full area coverage of the graphene. Upon this confirmation, the stack was then assembled into an MEA package. The gas diffusion electrodes (GDE) (Fuel Cell Store, College Station, TX) contained a Pt catalyst at a loading of 0.3 mg / cm2on Vulcan. A stack was pressed into an MEA package at 120 °C, 3 min at a force of 20 lbf / cm2containing one GDE, a GORE Select membrane, the coated membrane, and another GDE. Cross sections of MEA’ s were obtained by immersing the MEA into liquid nitrogen for 15 minutes and then cutting with a razor.Fuel Cell Testing

[0095] Fuel cell testing was conducted on a Scribner 850 commercial test stand. A break in process was conducted on the fuel cell packages before any testing was performed: the voltage was cycled at 100% RH (80 °C) between 0.9 V and 0.6 V 10 times, at 32% RH between 0.9 V and 0.6 V 10 times and at 40 °C between 0.9 V and 0.6 V 5 times. Polarization curves were obtained at 30% RH, 70% RH and 100% RH at 80 °C at flowrates of 0.14 lpm / 0.56 1pm H2 / Air. These curves were obtained with galvanostatic control. Crossover analysis was conducted at the previous conditions but at flowrates of 0.125 lpm / 0.125 1pm H2 / N2. Upon the switch to different gases, the system was allowed to equilibrate until a stable reduction in open circuit voltage (OCV) was seen and held for 15 minutes. Upon this stability, a potential of 0.4 V was applied and held to eliminate any hydrogen that had previously built up before running a linear sweep voltammetry measurement.

[0096] Accelerated stress testing (AST) was conducted by the DOE standard for membrane chemical degradation through a lOOh OCV hold at 90 °C, 30% RH. The flowrates were held at 0.07 lpm / 0.16 1pm H2 / Air. Polarization curves and crossover measurements were obtained after the AST by the method described earlier, after allowing the fuel cell system to come to equilibrium at 80 °C / 100% RH.Mechanical & TGA Testing

[0097] Uniaxial tensile testing was carried out using a Shimadzu Autograph AGS- X universal testing machine equipped with a 1 kN load cell. Samples were punched (width = 18.3 mm, length = 24 mm) from films approximately 16 pm in thickness. Uniaxial extension was carried out at arate of 10 mm / min until fracture. Samples were loaded with minimal slack and subjected to pretest loading conditions at the same strain rate (10 mm / min) until a positive force (0.05 N) was detected at which point data collection commenced. All samples were tested under ambient conditions.

[0098] Thermogravimetric analysis (TGA Q500, TA Instruments) was performed to evaluate thermal stability of each film. Films were cut into ~4 mm x 4 mm pieces, and 2-3 layers were stacked to achieve an average total mass of ~4 mg per sample. The samples were placed in a platinum pan, and the analysis was conducted under a nitrogen flow of 10 mL / min. The temperature method followed an initial ramp from room temperature to 30 °C at a rate of 10 °C / min, followed by an isothermal hold for 10 minutes. Subsequently, the temperature was increased at a rate of 10 °C / min until reaching 820 °C.Modeling Method

[0099] Molecular dynamics simulations were performed on the system using LAMMPS software package. For the ozonated graphene pore membrane, the Lennard -Jones parameters for carbon and oxygen atoms were occ =0.34 nm, ecc =0.36 kJ mol’1, and sOo =0.29 nm, eoo =0.59 kJ mol’1, respectively. The intermolecular interactions of the atoms in the membrane are modeled according to the optimized potentials for liquid simulations, all-atom (OPLS-AA) force field. The partial atomic charges for oxygen (c / o=-0.4e) and carbon (qc=+0.2e) were set according to the original OPLS-AA parametrization. The SPC water model was employed for water molecules, while a non-polarizable force field and hydroxide ions. All O-H bonds and H-O-H angles in water, hydronium, and hydroxide ions were constrained to remain rigid. For hydrogen, two-site Lennard-Jones model was utilized.

[0100] The carbon atoms at the boundary of the ozonated graphene pore membrane were fixed in place to maintain the membrane's structure. Additionally, the displacement of all atoms in the membrane along the z-di recti on was set to zero to avoid deformation of the pore and to prevent fluctuations, which could result from the migration of charged atoms in the nanopore under the influence of the electric field. The hydroxide ions were fixed in place to prevent water molecule transport via hydroxide ion movement under the applied electric field.

[0101] The Lennard-Jones interactions were computed using the 12 / 6 potential, and electrostatic interactions were calculated using the Coulomb potential, both with a 12 A cutoff distance. For interactions beyond the cutoff distance, the PPPM method was employed. The system was treated as periodic in all directions. Simulations were conducted in the NVT ensemble at T= 300 K with a 1 fs integration timestep using the Velocity-Verlet algorithm. The system was equilibrated for 5 nsprior to the production run. During the production run, an electric field magnitude of 1 V / Lz was applied in the z-direction, normal to the nanopore membrane, resulting in a voltage difference of 1 V. The flux of water and hydrogen molecules was determined by counting the total number of molecules that translocated through the nanopore. The flux rate was obtained by calculating the slope of cumulative flux across the membrane as a function of simulation time.Example 1

[0102] The membrane properties themselves can be viewed in the lens of two measures of selectivity. The contrasting views are presented in FIG. 14(a)-(d). The simplistic view of the conductivity / crossover ratio, presented in FIG. 14(a) suggests that in the Ohmic regime, the selectivity suffers as a function of ozonation, and that the highest selectivity membrane is atomically pristine graphene. This statement would be accurate in a purely diffusive system, which is not true of a fuel cell, as opposed to pump cells. This finding agrees with previous work in membrane literature. However, this is a necessarily incomplete look at fuel cell membrane performance, which ignores the contribution of the activation region and 2D material impact to it. FIG. 14(b) shows the effect pristine graphene has on the activation overpotential, showing a nearly triple overpotential compared to the baseline. This overpotential is likely due to heterogeneous accessibility to the catalyst.

[0103] The incorporation of a pristine graphene layer within the proton exchange membrane significantly impacts the oxygen reduction reaction (ORR) catalyst kinetics and overall fuel cell performance, as evidenced by increased activation overpotential in FIG. 14(b) and distinct changes in polarization behavior, despite maintaining fully humidified gas feeds. This phenomenon likely arises from disruption of water-mediated proton transport networks. While protons can traverse the graphene barrier through the Grotthus mechanism, the water impermeable nature of pristine graphene prevents the formation of continuous water channels through the membrane, inhibiting dynamic water redistribution. This effect has been previously observed by Moehring et. al. in thinner reinforced membranes below 10 um thickness, as the membrane is not thick enough to facilitate proton redistribution across the area of the membrane before the cathode,

[0104] This disruption of continuous water networks manifests as heterogeneous proton activity across catalyst sites, where some active sites maintain optimal proton access through local water networks (via defects) while others experience limited proton availability and disrupted water- mediated transport. The resulting spatial and temporal variations in local proton concentration affect multiple aspects of cell performance: increased activation overpotential reflecting modified reaction kinetics, enhanced ohmic resistance due to suboptimal membrane hydration distribution,and earlier onset of mass transport limitations from non-uniform water accumulation. These effects are mitigated through ozonation of the graphene layer, which creates water transport pathways through the barrier. This modification restores uniform proton activity across catalyst sites by reestablishing water-mediated proton transport networks, resulting in improved kinetics, reduced ohmic resistance, and enhanced mass transport characteristics by allowing the dynamic redistribution of protons. The systematic improvement across all regions of the polarization curve with ozonated graphene demonstrates that maintaining homogeneous proton activity through efficient interfacial water networks is crucial for optimal fuel cell performance, even under fully humidified conditions. By UV-ozonation, this effect on the catalyst region is significantly mitigated by opening more defects (grain boundaries and vacancies) that facilitate fuel accessibility and homogeneous proton transport. The 10-minute and 15-minute samples eliminated this parasitic effect.

[0105] Accordingly, it becomes prudent to view fuel cell selectivity not only in terms of ohmic selectivity, but primarily in terms of flux ratios calculated by employing the Faraday relation:

[0106] When viewed in terms of fluxes, as seen in FIG. 14(c) the rate of hydrogen permeation increase is much lower than that of proton flux increase, suggesting that ozonated graphene retains some selectivity. When the ratio of fluxes is computed, as in FIG. 14(d), pristine graphene impedes the overall selectivity of the membrane by nearly 13%. However, once defects are introduced to the lattice, the overall selectivity of the membrane is seen to increase past the baseline, peaking at the 10-minute ozonation sample, which shows a 30% increase in selectivity from the baseline. Selectivity in nanoporous graphene membranes can be described as the contribution of both selective pores and large, effusive pores - the data suggests that the pristine and 5-minute sample have many selective pores, but do not facilitate enough HsO+or proton permeance to result in a peak in selectivity. On the other hand, the 15-minute sample loses selectivity due to the larger concentration of effusive pores, compromising the balance between effusive and selective pores. Because ionomers use a chemical transport mechanism rather than a standard size exclusionary permeation model, the notion of an “upper bound” governed solely by kinetic diameters is an inaccurate model of transport to sieve between protons and molecular hydrogen. However, the incorporation of an atomic sieve in conjunction with an ionomer pushes the bounds of both individually, suggesting that this strategy could be effective in enhancing selectivity of other applications, such as electrolyzers.Example 2 Accelerated Stress Testing (AST) of Graphene Membranes

[0107] Another benefit of the introduction of nanoporous graphene is found in the durability data, conducted with the DOE standard accelerated stress test (90 °C, 30% RH), seen in FIG. 15. The baseline sandwich of GORE Select membranes exhibited a degradation rate of 0.38 mV / hr, in line with literature data on PFSA based materials. All ozonated samples were chosen to benchmark against the GORE baseline as they show higher selectivity and performance, implying highest expected durability. As seen in FIG. 15(a)-(b), the 5 -minute and 10-minute ozonated graphene sample showed a degradation rate of 0.21 and 0.26 mV / hr, merely 55% and 61% of the baseline, respectively, and significantly improved compared to CeO? based solutions as seen in FIG. 15(f) and the Table below (Comparison of Ozonated Graphene Durability Results to Similar Works. Ozonated graphene shows superior performance at operational voltages before and after AST.)9. Kutagulla, S.; Le, N. H.; Terry, I.; Bohn, C.; Stacy, B. J.; Favela, C. S.; Slack, J. J. Baker, A. M.; Kim, H.; Shin, S.; Korgel, B. A.; Akinwande, D. Comparative Studies of Atomically Thin Proton Conductive Films to Reduce Crossover in Hydrogen Fuel Cells ACS Appl Mater Interfaces 2023.21. Robeson, L. M. The Upper Bound Revisited. J Memb Sci2008, 320(1-2), 390-400.22. Baker, A. M.; Mukundan, R.; Spemjak, D.; Judge, E. J.; Advani, S. G.; Prasad, A. K.;Borup, R. L. Cerium Migration during PEM Fuel Cell Accelerated Stress Testing. J Electrochem Soc 2016,163(9), F1023- F1031.23. Vinothkannan, M.; Hariprasad, R.; Ramakrishnan, S.; Kim, A. R.; Yoo, D. J. Potential Bifunctional Filler (CeO? -ACNTs) for Nafion Matrix toward Extended Electrochemical Power Density and Durability in Proton-Exchange Membrane FuelCells Operating at Reduced Relative Humidity. 2019.24. Baker, A. M.; Wang, L.; Johnson, W. B.; Prasad, A. K.; Advani, S. G. Nafion Membranes Reinforced with Ceria-Coated Multiwall Carbon Nanotubes for Improved Mechanical and Chemical Durability in Polymer Electrolyte Membrane Fuel Cells. Journal of Physical Chemistry C 2014,118(46), 26796-26802.

[0108] The reduction in current density and power output is minimal (below 10%), in contrast to the hygroscopic oxides, which show proton conductivity losses of over 15-20%. The 15-minute ozonated sample showed a degradation rate of 0.34 mV / hr, still an 11% reduction from the baseline. However, the current density is seen to increase even further relative to the 10-minute sample, showing a current of 1450 mA / cm2after the AST.

[0109] Furthermore, as seen in FIG. 15(c), the performance of the polarization curve before and after the AST is striking; the 10-minute ozonated graphene shows an increase in current output compared to the beginning of life measurement, and the 15-minute sample exhibits a 45% increase in current output. This is attributed to membrane thinning introduced through hydrogen exposure at elevated temperatures in conjunction with chemical degradation of the PFSA, and is corroborated in similar experiments. In effect, the hydrogen barrier graphene layer is intact while the PFSA degrades, allowing for higher current outputs but still relatively reduced crossover. Furthermore, as seen in FIG. 15(e), the crossover for the 10-minute sample only increases by a mere 0.2 mA / cm2, ending at 4.2 mA / cm2, 19% below the beginning of life measurement for the baseline membrane. Due to the larger pores and lower selectivity in the 15-minute sample, the crossover is only 3.8% reduced in the 15-minute sample relative to baseline. In contrast, the baseline also decreased in performance after the AST, with a decrease in current output of 19.8%.Analyzing Pore Dimension on the Hydronium (Proton), Water and Hydrogen Transport

[0110] Ozonated graphene pore membranes with C-O-C nanopore termination of various sizes, as shown in FIG. 16, were generated using VMD and BOVIA Materials Studio. The nanopores are symmetrically shaped, and their sizes are referenced by the largest dimension, defined as the maximum distance between nanopore atoms. The simulation box contains two reservoirs, each filled with 19,854 water molecules, 20 hydronium ions, and 20 hydroxide ions, generated using Packmol. The reservoir at -Z contains 40 H2 molecules, while the reservoir at +Z contains no H2 molecules creating a concentration gradient at t=0 that acts as a driving force for hydrogen crossover. These reservoirs are separated by the ozonated graphene pore membrane, which is oriented with its normal aligned along the z-axis. The simulation domain has dimensions of 12 nm in the x-direction, 12.2 nm in the j’-direction, and 8.4 nm in the z-direction.Example 3 Proton and Hydrogen Flux[OHl] Molecular dynamics simulations were performed, as outlined in the Methods section, to investigate the effect of ozonated graphene pore size on the hydronium ion, water, and hydrogen transport. The production run was carried out for 40 ns, with the initial 10 ns of data discarded to reduce fluctuations hydrogen concentrations in both reservoirs, thereby representing the steadystate operation of the system. The remaining 30 ns of data were used for analysis. An electric field of 1 V / Lz was applied in the direction normal to the nanopore membrane. The simulation results suggest that larger pores facilitate higher proton transport (FIG. 17(a)), higher water transport (FIG. 17(c)), and higher hydrogen transport (FIG. 17(a)), in alignment with conventional expectations. Based on FIG. 17(b), the selectivity of 2 nm, 4 nm and 6 nm nanopores are 7.1, 2.3, and 1.5 x the 10 nm pore’s selectivity, showing a rapid decrease in membrane selectivity with increasing defect size.

[0112] However, continuous water channels facilitate large scale dynamic proton redistribution, and the interruption of these forces proton transport at only hotspots near defects. As seen in FIG. 17(c), and FIGs. 18(a)-(c) and FIG. 19(a)-(c), the water transport through the nanopores increases significantly with defects, such as those caused by ozonation, allowing for the redistribution of protons to happen more seamlessly. By opening up water channels with larger defects, net gains can still be made in fuel cell performance through more effective redistribution and diffusion of protons throughout the membrane, and removes the negative effects seen in the activation region of the catalyst. As such, there are 3 distinct regimes that can be observed in graphene / PFSA membranes. The first is comprised of purely selective pores, which allow for extremely high ohmic selectivity between protons and hydrogen, but impede permeation of protons, leading to reduced catalyst effectiveness. The second is a mixed regime, where selectivity is sacrificed for permeation, resulting in more balanced areal redistribution of protons, allowing for restoration of activation losses while resulting in higher selectivity still. The third regime is a purely effusive system which behaves no different than the baseline. UV ozonation allows for controlled choice in which regime to operate in- the second mixed regime in this work’s case- allowing for tailored solutions that still push the upper bound of selectivity.Example 4

[0113] Hydrogen fuel cells have the potential to revolutionize the heavy-duty transportation sector, but their widespread adoption has been impeded by the limitations of current proton exchange membranes (PEMs). Hydrogen crossover remains a critical challenge, leading to reduced efficiency and accelerated membrane degradation. While previous attempts to incorporate 2Dmaterials into PEMs have shown promise in mitigating crossover, they have unilaterally resulted in compromised proton conductivity and overall performance, caused by mitigation of water permeation through the membrane. [1,2,3]

[0114] This disclosure presents a novel and scalable approach to extend both PEM lifetime and performance using UV-ozone treated monolayer graphene as a selective barrier layer without any degradation in performance.

[0115] This disclosure utilizes a UV-ozone treatment approach allowing for precise control over the size and density of angstrom and nanometer scale defects in the graphene lattice. By tuning the ozone exposure time, we create a defect distribution that maximizes hydrogen / proton selectivity while minimizing the impact on proton conductivity by facilitating the transport of water. Our results demonstrate a 27% increase in selectivity over state-of-the-art Gore Select membranes, a 24% decrease in hydrogen crossover, and up to a 19% enhancement in current density at 0.7 V. Moreover, our ozonated graphene PEMs exhibit an impressive 39% improvement in durability, displaying no performance loss after a 100-hour accelerated stress test.

[0116] The significance of our approach lies in its simplicity and scalability to integrate size selective graphene with traditional PFSA membranes. The UV-ozone treatment is a straightforward post-processing step that can be readily applied to commercially available graphene materials and seamlessly integrated into existing PEM manufacturing processes. This makes our technology a practical solution with immense potential for real-world implementation. This work sets the stage for further advancements in PEM design and opens new avenues for the development of high-performance, ultra-durable fuel cell systems.

[0117] [1] S. Kutagulla et al., ACS Appl Mater Interfaces 2023, 15, 51, 59358-59369; [2] M. Komma et al., ACS Appl. Mater. Interfaces 2024, 16, 18, 23220-23232; [3] N.K. Moehring et al., Nanoscale, 2024, 16, 6973.ReferencesCunanan, C.; Tran, M. K.; Lee, Y.; Kwok, S.; Leung, V.; Fowler, M. A Review of Heavy-Duty Vehicle Powertrain Technologies: Diesel Engine Vehicles, Battery Electric Vehicles, and Hydrogen Fuel Cell Electric Vehicles. Clean Technologies 2021, Vol. 3, Pages 474-489 2021, 3 (2), 474-489.Yue, M.; Lambert, H.; Pahon, E.; Roche, R.; Jemei, S.; Hissel, D. Hydrogen Energy Systems: A Critical Review of Technologies, Applications, Trends and Challenges. Renewable and Sustainable Energy Reviews 2021, 146.Schoemaker, M.; Misz, U.; Beckhaus, P.; Heinzel, A. Evaluation of Hydrogen Crossover through Fuel Cell Membranes. Fuel Cells 2014, 14 (3), 412-415.Kocha, S. S.; Yang, J. D.; Yi, J. S. Characterization of Gas Crossover and Its Implications in PEM Fuel Cells. 2006.Scalable Implementation of Recombination Catalyst Layers to Mitigate Gas Crossover in PEM Water Electrolyzers. 2022.Lin, X.; Zhu, J.; Seow, Y.; Xu, Z. J.; Mbeki, M.; Ernst, M. F.; Komherr, M. Membrane Interlayer with Pt Recombination Particles for Reduction of the Anodic Hydrogen Content in PEM Water Electrolysis You May Also like A Brief Introduction of Electrode Fabrication for Proton Exchange Membrane Water Electrolyzers.Zhu, L. Y.; Li, Y. C.; Liu, J.; He, J.; Wang, L. Y.; Lei, J. Du. Recent Developments in High- Performance Nafion Membranes for Hydrogen Fuel Cells Applications. Pet Sci 2022, 19 (3), 1371-1381.Moehring, N. K.; Chaturvedi, P.; Cheng, P.; Ko, W .; Li, A.-P.; Boutilier, M. S. H.; Kidambi, P. R. Kinetic Control of Angstrom-Scale Porosity in 2D Lattices for Direct Scalable Synthesis of Atomically Thin Proton Exchange Membranes. ACS Nano 2022, 16, 28.Kutagulla, S.; Le, N. H.; Terry, I.; Bohn, C.; Stacy, B. J.; Favela, C. S.; Slack, J. J.; Baker, A. M.; Kim, H.; Shin, S.; Korgel, B. A.; Akinwande, D. Comparative Studies of Atomically Thin Proton Conductive Films to Reduce Crossover in Hydrogen Fuel Cells. ACS Appl Mater Interfaces 2023.Imamura, G.; Saiki, K. Interlayer Interaction in the UV Irradiated Defect Formation of Graphene. Journal of Physical Chemistry C 2014, 118 (22), 11842-11848.Zhao, J.; He, G.; Huang, S.; Villalobos, L. F.; Dakhchoune, M.; Bassas, H.; Agrawal, K. V. Etching Gas-Sieving Nanopores in Single-Layer Graphene with an Angstrom Precision for High- Performance Gas Mixture Separation. Sci Adv 2019, 5 (1).Takamuku, S. Proton Exchange Membrane Involving Compatible Acid-Base Interaction Towards High Performance and Mechanical Property. ECS Meeting Abstracts 2016, MA2016-02 (38), 2556.Koenig, S. P.; Wang, L.; Pellegrino, J.; Bunch, J. S. Selective Molecular Sieving through Porous Graphene. NATURE NANOTECHNOLOGY 2012, 7.Celebi, K.; Buchheim, J.; Wyss, R. M.; Droudian, A.; Gasser, P.; Shorubalko, I.; Kye, J. Il; Lee, C.; Park, H. G. Ultimate Permeation across Atomically Thin Porous Graphene. Science (1979) 2014, 344 (6181), 289-292.Eckmann, A.; Felten, A.; Mishchenko, A.; Britnell, L.; Krupke, R.; Novoselov, K. S.; Casiraghi, C. Probing the Nature of Defects in Graphene by Raman Spectroscopy. Nano Lett 2012, 12 (8), 3925-3930.Freeman, B. D. Basis of Permeability / Selectivity Tradeoff Relations in Polymeric Gas Separation Membranes. Macromolecules 1999, 32 (2), 375-380.(17) Haxhiu, A.; Chan, R.; Kanerva, S.; Kyyra, J. A System Level Approach to Estimate Maximum Load Steps That Can Be Applied on a Fuel Cell Powered Marine DC System. Energy Reports 2021, 7, 888-895.Zavorotnaya, U. M.; Ponomarev, 1. 1.; Volkova, Y. A.; Sinitsyn, V. V. Development of High- Performance Hydrogen- Air Fuel Cell with Flourine-Free Sulfonated Co-Polynaphthoyleneimide Membrane. Membranes 2023, Vol. 13, Page 485 2023, 13 (5), 485.Park, H. B.; Kamcev, J.; Robeson, L. M.; Elimelech, M.; Freeman, B. D. Maximizing the Right Stuff: The Trade-off between Membrane Permeability and Selectivity. Science (1979) 2017, 356 (6343), 1138-1148.Moehring, N. K.; Bashith, A.; Basha, M.; Chaturvedi, P.; Knight, T.; Fan, X.; Pintauro, P. N.; Boutilier, M. S. H.; Karan, K.; Kidambi, P. R. Overcoming the Conductance versus Crossover Trade-off in State-of-the-Art Proton Exchange Fuel -Cell Membranes by Incorporating Atomically Thin Chemical Vapor Deposition Graphene. Nano Lett 2025.Robeson, L. M. The Upper Bound Revisited. JMemb Sci 2008, 320 (1-2), 390-400.Baker, A. M.; Mukundan, R.; Spemjak, D.; Judge, E. J.; Advani, S. G.; Prasad, A. K.; Borup, R. L. Cerium Migration during PEM Fuel Cell Accelerated Stress Testing. J Electrochem Soc 2016, 763 (9), F1023-F1031.Vinothkannan, M.; Hariprasad, R.; Ramakrishnan, S.; Kim, A. R.; Yoo, D. J. Potential Bifunctional Filler (CeO 2 -ACNTs) for Nafion Matrix toward Extended Electrochemical Power Density and Durability in Proton-Exchange Membrane Fuel Cells Operating at Reduced Relative Humidity. 2019.Baker, A. M.; Wang, L.; Johnson, W. B.; Prasad, A. K.; Advani, S. G. Nafion Membranes Reinforced with Ceria-Coated Multiwall Carbon Nanotubes for Improved Mechanical and Chemical Durability in Polymer Electrolyte Membrane Fuel Cells. Journal of Physical Chemistry C 2014, 118 (46), 26796-26802.Ketpang, K.; Lee, K.; Shanmugam, S. Facile Synthesis of Porous Metal Oxide Nanotubes and Modified Nafion Composite Membranes for Polymer Electrolyte Fuel Cells Operated under Low Relative Humidity. ACS Appl Mater Interfaces 2014, 6 (19), 16734-16744.Vinothkannan, M.; Ramakrishnan, S.; Kim, A. R.; Lee, H. K.; Yoo, D. J. Ceria Stabilized by Titanium Carbide as a Sustainable Filler in the Nafion Matrix Improves the Mechanical Integrity, Electrochemical Durability, and Hydrogen Impermeability of Proton-Exchange Membrane Fuel Cells: Effects of the Filler Content. ACS Appl Mater Interfaces 2020, 72 (5), 5704-5716. https: / / doi.org / 10.1021 / ACSAMI.9B18059 / ASSET / IMAGES / LARGE / AM9B18059_0009.JPEG.Bukola, S.; Liang, Y.; Korzeniewski, C.; Harris, J.; Creager, S. Selective Proton / Deuteron Transport through Nafion|Graphene|Nafion Sandwich Structures at High Current Density. 2018.Humphrey, W .; Dalke, A.; Schulten, K. VMD: Visual Molecular Dynamics. J Mol Graph 1996, 74 (1), 33-38.Martinez, L.; Andrade, R.; Birgin, E. G.; Martinez, J. M. PACKMOL: A Package for Building Initial Configurations for Molecular Dynamics Simulations. J Comput Chem 2009, 30 (13), 2157-2164.Johra, F. T.; Lee, J. W.; Jung, W. G. Facile and Safe Graphene Preparation on Solution Based Platform. Journal of Industrial and Engineering Chemistry 2014, 20 (5), 2883-2887.Kumar, N.; Towers, D.; Myers, S.; Galvin, C.; Kireev, D.; Ellington, A. D.; Akinwande, D. Graphene Field Effect Biosensor for Concurrent and Specific Detection of SARS-CoV-2 and Influenza. medRxiv 2023, 2022.Kireev, D.; Sarik, D.; Wu, T.; Xie, X.; Wolfrum, B.; Offenhausser, A. High Throughput Transfer Technique: Save Your Graphene. Carbon N Y 2016, 107, 319-324.Jorgensen, W. L.; Maxwell, D. S.; Tirado-Rives, J. Development and Testing of the OPLS AllAtom Force Field on Conformational Energetics and Properties of Organic Liquids. J Am Chem Soc 1996, 118 (45), 11225-11236.Grootenhuis, P. D. J.; Kollman, P. A. Molecular Mechanics and Dynamics Studies of Crown Ether-Cation Interactions: Free Energy Calculations on the Cation Selectivity of Dibenzo-18- Crown-6 and Dibenzo- 30-Crown-10. J Am Chem Soc 1989, 111 (6), 2152-2158.Fang, A.; Kroenlein, K.; Riccardi, D.; Smolyanitsky, A. Highly Mechanosensitive Ion Channels from Graphene-Embedded Crown Ethers. Nature Materials 2018 18:1 2018, 18 (1), 76-81.Bonthuis, D. J.; Mamatkulov, S. I.; Netz, R. R. Optimization of Classical Nonpolarizable Force Fields for OH- and H3O+. Journal of Chemical Physics 2016, 144 (10).Weiner, S. J.; Kollman, P. A.; Nguyen, D. T.; Case, D. A. An All Atom Force Field for Simulations of Proteins and Nucleic Acids. J Comput Chem 1986, 7 (2), 230-252.Yang, Q.; Zhong, C. Molecular Simulation of Adsorption and Diffusion of Hydrogen in Metal- Organic Frameworks. Journal of Physical Chemistry B 2005, 109 (24), 11862-11864.Barraco, M.; Neyertz, S.; Benes, N. E.; Brown, D. Comparison of Eight Classical Lennard- Jones- Based H2 Molecular Models in the Gas Phase at Temperatures and Pressures Relevant to Hydrogen On-Board Storage Tanks. Journal of Physical Chemistry A 2023, 127 (30), 6335-6346.

[0118] In this application, unless specifically stated otherwise, the use of the singular includes the plural, and the separate use of “or” and “and” includes the other, i.e., “and / or.” Furthermore, use of the terms “including” or “having,” as well as other forms such as “includes,” “included,” “has,” or “had,” are intended to have the same effect as “comprising” and thus should not be understood as limiting.

[0119] Any range described herein will be understood to include the endpoints and all values between the endpoints. Whenever “substantially,” “approximately,” “essentially,” “near,” “about,” or similar language is used in combination with a specific value, variations up to and including 10% of that value are intended, unless explicitly stated otherwise.

[0120] Many alternatives, modifications, and variations are enabled by the present disclosure. While specific examples have been shown and described in detail to illustrate the application of the principles of the present invention, it will be understood that the invention may be embodied otherwise without departing from such principles. For example, disclosed features may becombined, rearranged, omitted, etc. to produce additional embodiments, while certain disclosed features may sometimes be used to advantage without a corresponding use of other features.Accordingly, Applicant intends to embrace all such alternative, modifications, equivalents, and variations that are within the spirit and scope of the present invention. All the publications are hereby incorporated by reference.

Claims

WHAT IS CLAIMED IS:1 A method to generate a composite ionomer-graphene membrane, the method comprising: providing a graphene layer; subjecting the graphene layer to UV-Ozone treatment to generate a plurality of defects distributed across a surface of the graphene layer, wherein plurality of defects corresponds to a defect density of about IxlO11cm’2to about 5xl012cm’2and wherein the UV-Ozone treatment comprises generating ultraviolet light in an oxygen-containing atmosphere, and wherein molecular oxygen in the oxygen-containing atmosphere absorbs the ultraviolet light to generate oxygen radicals and / or molecular ozone that reacts with the graphene layer to generate the defects; and laminating the treated graphene layer to an ionomer membrane to generate a composite ionomer-graphene membrane.

2. The method of claim 1, wherein the graphene layer comprises an atomically thick two-dimensional layer comprising sp2-hybridized carbon.

3. The method of claim 1, wherein the plurality of defects includes both selective defects and effusive defects.

4. The method of claim 1, wherein the graphene layer is subjected to the UV-Ozone treatment for a duration of time from 1 minute to 20 minutes.

5. The method of claim 1, wherein UV-Ozone treatment comprises generating oxygen radicals and / or molecular ozone at a temperature of from about 75 °C to about 125 °C.

6. The method of claim 1, wherein providing the graphene layer comprises providing the graphene layer on a substrate and wherein the method further comprises removing the graphene layer from the substrate after the UV-Ozone treatment.

7. The method of claim 1, wherein the ionomer membrane is a first ionomer membrane, and wherein the method further comprises encapsulating the porous graphene between the first ionomer membrane and a second ionomer membrane to generate the composite ionomergraphene membrane.

8. The method of claim 1, wherein the ionomer membrane comprises a sulfonated tetrafluoroethylene polymer, a tetrafluoroethylene polymer, an expanded polytetrafluoroethylenepolymer, or a composite of tetrafluoroethylene polymer and a sulfonated tetrafluoroethylene polymer.

9. The method of claim 1, wherein the ionomer membrane has a thickness from about 5 pm to about 120 pm.

10. The method of claim 1, wherein the ionomer membrane has a thickness from about7 pm to about 50 pm.

11. The method of claim 1, wherein the ionomer membrane has a thickness from about8 pm to about 30 pm.

12. The method of claim 1, wherein the ionomer membrane has a thickness from about 8 pm to about 30 pm.

13. The method of claim 1, wherein the ionomer membrane comprises or corresponds to a proton- or ion-exchange membrane.

14. The method of claim 1, further comprising positioning the composite ionomergraphene membrane between a first gas diffusion electrode and a second gas diffusion electrode.

15. The method of claim 14, further comprising contacting the first gas diffusion electrode with a fuel containing gas and contacting with the second gas diffusion electrode to generate a voltage and / or current between the first gas diffusion electrode and the second gas diffusion electrode.

16. The method of claim 14, further comprising contacting the first gas diffusion electrode with a source gas and applying a voltage between the first gas diffusion electrode and the second gas diffusion electrode to generate a current.

17. The method of claim 1, wherein the plurality of defects corresponds to a defect density of about 5x1011cm’2.

18. A system comprising: a graphene layer, wherein the graphene layer includes a plurality of defects distributed across a surface of the graphene layer, wherein the graphene layer exhibits a defect density of about IxlO11cm’2to about 5xl012cm’2; and an ionomer membrane, wherein the graphene layer is in contact with and supported by the ionomer membrane.

19. The system of claim 18, wherein the graphene layer comprises an atomically thick two-dimensional layer comprising sp2-hybridized carbon.

20. The system of claim 18, wherein the plurality of defects includes both selective defects and effusive defects.

21. The system of claim 18, wherein the ionomer membrane comprises a member selected from the group consisting of a sulfonated tetrafluoroethylene polymer, a tetrafluoroethylene polymer, an expanded polytetrafluoroethylene polymer, or a composite of tetrafluoroethylene polymer and a sulfonated tetrafluoroethylene polymer.

22. The system of claim 18, wherein the ionomer membrane has a thickness from about 5 pm to about 120 pm.

23. The system of claim 18, wherein the ionomer membrane has a thickness from about7 pm to about 50 pm.

24. The system of claim 18, wherein the ionomer membrane has a thickness from about8 pm to about 30 pm.

25. The system of claim 18, wherein the ionomer membrane is a proton- or ionexchange membrane.

26. The system of claim 18, wherein the ionomer membrane comprises a bottom ionomer membrane and wherein the system further comprises a top ionomer membrane in contact with the graphene layer on a side of the graphene layer opposite to the bottom ionomer membrane.

27. The system of claim 26, comprising a composite structure including the bottom ionomer membrane, the top ionomer membrane, and the graphene layer between the bottom ionomer membrane and the top ionomer membrane.

28. The system of claim 18, further comprising a first gas diffusion electrode and a second gas diffusion electrode, wherein the graphene layer and the ionomer membrane are positioned between the first gas diffusion electrode and the second gas diffusion electrode.

29. The system of claim 28, comprising or corresponding to a fuel cell, wherein a fuel containing gas is positioned in contact with the first gas diffusion electrode, and wherein an oxidizing agent containing gas is positioned in contact with the second gas diffusion electrode.

30. The system of claim 28, comprising or corresponding to an electrolysis cell, wherein a source gas is positioned in contact with the first gas diffusion electrode for reducing the source gas at the first gas diffusion electrode upon application of a voltage between the first gas diffusion electrode and the second gas diffusion electrode.

31. The system of claim 18, wherein the plurality of defects corresponds to a defect density of about 5x1011cm’2.

Images