1. Problem and Architectural Premise

Existing rechargeable electrochemical energy-storage devices fall into a small number of architectural classes, each defined by the physical mechanism through which energy is stored and the boundary condition under which the device operates. PEM and SOFC fuel cells store no energy internally; they convert externally supplied hydrogen and externally supplied oxygen, and their operational boundary is open at both reactant ports. Metal-air batteries store energy by oxidizing a metal anode using atmospheric oxygen ingress through a permeable cathode, and their operational boundary is open at the air interface. Lithium-ion and other intercalation batteries store energy through cation insertion into a host lattice, with no covalent bond formation at the active site, and their operational boundary is closed but their storage mechanism is van der Waals or weak ionic. Nickel-metal-hydride cells store hydrogen as a metal hydride at the negative electrode and oxidize at a nickel hydroxide cathode, separated by bulk aqueous potassium hydroxide electrolyte. None of these classes admits a closed-boundary device that stores both reactants of water-splitting as covalent bonds to a single class of active material.

The architectural premise of the carbon-bound hydrogen-oxygen cell is that the two products of water electrolysis, hydrogen and oxygen, can each be held as covalent bonds to carbon, on two spatially separated electrodes within a single sealed enclosure, such that charge is stored as the covalent-bond population imbalance between the two electrodes and discharge is the reversal of that imbalance through proton transport across a polymer membrane. The premise rests on three structural facts: graphitic carbon admits stable C-H bond formation at coverages up to approximately 8 weight percent (the graphane regime); graphitic carbon admits stable C-O bond formation at coverages up to approximately 20 weight percent in carboxyl, hydroxyl, and epoxide functional groups (the graphene-oxide regime); and the activation barriers separating the bound and unbound states are large enough to admit multi-month charge retention while small enough to admit cycling at practical current densities under modest catalysis.

The consequence is a closed-boundary device whose mass is substantially conserved across cycles. No air ingress is required; no fuel is resupplied; no electrolyte is topped off; no metal is consumed. Only electrons cross the cell boundary, through the external circuit. The cell is structurally distinct from every prior class because every prior class either requires open-boundary mass exchange or stores energy through a non-covalent mechanism. The architectural premise is the joint constraint: covalent storage on both reactant species, on a common active-material class, within a sealed enclosure, with reversible water-splitting electrochemistry as the cycling mechanism.

2. The Two-Scaffold Covalent-Storage Primitive

The core architectural primitive is a pair of carbon scaffolds: geometrically separated, electrochemically connected through a proton-conducting polymer membrane, and functionally differentiated as hydrogen-storing and oxygen-storing electrodes. The scaffolds are manufactured from a common class of high-surface-area graphitic carbon (graphene, graphene oxide, reduced graphene oxide, carbon nanotubes, activated carbon, carbon black, hierarchical porous carbons, or composites thereof) with surface area typically in the range of 200 to 2000 square meters per gram and pore-size distributions spanning micropores below 2 nm to mesopores in the 2 to 50 nm range.

Both scaffolds are, in their as-manufactured state, substantially identical in composition. They are differentiated emergently, through an initial conditioning protocol of 5 to 50 charge-discharge cycles, into a hydrogen-storing functional state and an oxygen-storing functional state. The differentiation is determined by polarity: the electrode connected to the negative terminal during charging develops the C-H bond population characteristic of the graphane regime; the electrode connected to the positive terminal develops the C-O bond population characteristic of the graphene-oxide regime. After conditioning, the differentiation is stable across subsequent cycling within the operating envelope.

The covalent-storage primitive is the joint state of the two scaffolds: at any state of charge, the hydrogen-storing electrode carries some fraction of its maximum C-H coverage in the range of 0 to approximately 8 weight percent, and the oxygen-storing electrode carries some fraction of its maximum C-O coverage in the range of 0 to approximately 20 weight percent. The state of charge is the joint coverage state, and the open-circuit voltage between the two electrodes is the Gibbs free-energy difference between the bound and unbound states summed across the two scaffolds, in the range of 1.0 to 1.3 volts at typical operating coverages.

The two-scaffold separation is not a design preference; it is electrochemically required. Reversible operation requires that oxidation occur at one electrode and reduction at the other, without internal short. Co-locating the two functions on a single scaffold would admit internal recombination and defeat charge retention. The primitive is therefore irreducibly two-scaffold, with the proton-conducting membrane as the connecting mass-transport pathway.

3. Reversible Water-Splitting Electrochemistry

The cycling chemistry is reversible water-splitting performed in situ within the polymer-electrolyte membrane and at the carbon-electrode interfaces. During charging, an applied potential drives the cleavage of water sorbed within the membrane: protons are reduced at the hydrogen-storing electrode and form covalent C-H bonds with the carbon scaffold (the graphane-formation half-reaction), while oxygen evolved at the oxygen-storing electrode forms covalent C-O bonds in carboxyl, hydroxyl, and epoxide functional groups (the graphene-oxide-formation half-reaction). The net charging reaction binds water across the two scaffolds as separated covalent populations; no gaseous hydrogen or oxygen accumulates, and no bulk liquid water is produced.

During discharge, closure of the external circuit allows electrons to flow from the hydrogen-storing electrode through the load to the oxygen-storing electrode, while protons traverse the polymer membrane from the hydrogen side to the oxygen side via Grotthuss-mechanism transport. C-H bonds are oxidized to release protons; C-O bonds are reduced to release oxygen; the released species recombine within the membrane to reform water at the membrane-sorbed state. The discharge reaction is the inverse of the charging reaction, and the cell is substantially mass-conserving across the cycle.

The electrochemistry is catalysed at the carbon surfaces by trace transition-metal catalysts (platinum-group metals at low loading, transition-metal oxides, transition-metal carbides, or metal-free heteroatom-doped carbon catalysts) incorporated into the scaffold during fabrication. Catalyst loading is selected to admit practical exchange current densities, typically in the range of 10 to 1000 milliamperes per square centimeter at modest overpotentials, without compromising the stability of the bound states.

Reversibility is bounded by the cycling envelope: cycling within the coverage ranges given above and within the temperature envelope of approximately negative 20 to plus 80 degrees Celsius admits cycle counts in the thousands without significant capacity fade. Excursions outside the envelope drive irreversible side reactions including carbon corrosion, membrane dehydration, and catalyst dissolution.

4. Polymer Electrolyte and Absence of Bulk Water

The proton-conducting electrolyte is a solid polymer membrane selected from the class of perfluorinated sulfonic-acid ionomers (Nafion-class), sulfonated polyaryletheretherketone (sulfonated PEEK), sulfonated polysulfone, sulfonated polyimide, anion-exchange ionomers operated in proton-conduction modes via co-ion exchange, ceramic proton-conductors such as cesium dihydrogen phosphate or barium zirconate, or hybrid polymer-ceramic composites. Membrane thickness is typically in the range of 25 to 200 micrometres, and proton conductivity is typically in the range of 10 to 200 millisiemens per centimeter at operating temperature and humidity.

Water within the cell exists in only two forms. The first is covalently bound water, hydrogen held as C-H on one electrode and oxygen held as C-O on the other, which is not water in the molecular sense but is the dissociated, separated form of what was water before charging. The second is sorbed water, structurally bound within the hydrophilic nanopocket domains of the polymer membrane. Sorbed water occupies the sulfonate cluster regions of perfluorinated sulfonic-acid membranes and the analogous hydrophilic domains of alternative chemistries, in mass fractions typically in the range of 5 to 30 weight percent of the dry membrane mass. There is no bulk liquid water at any state of charge or discharge. The membrane is wet at the molecular scale and dry at the bulk scale.

The absence of bulk liquid water is structurally significant. It eliminates the freezing failure mode of bulk-aqueous architectures, eliminates the corrosion failure mode driven by dissolved-species transport, and eliminates the gravity-orientation dependence of liquid-electrolyte cells. It also distinguishes the architecture sharply from nickel-metal-hydride and from any aqueous-electrolyte battery class.

5. Operating Parameters and Engineering Envelope

The cell operates within a defined engineering envelope. Open-circuit voltage is 1.0 to 1.3 volts, set by the Gibbs free-energy difference between charged and discharged states; cycling voltage windows typically span 0.7 to 1.5 volts to admit useful round-trip efficiency without driving side reactions. Operating temperature is approximately negative 20 to plus 80 degrees Celsius, with the lower bound set by membrane proton conductivity and the upper bound set by membrane dehydration and carbon-oxide stability. Charge and discharge current densities are typically in the range of 1 to 100 milliamperes per square centimeter of geometric electrode area, with peak rates limited by proton transport across the membrane and by interfacial charge transfer at the carbon-electrolyte interface.

Hydrogen storage at the hydrogen-storing electrode is in the range of 1 to 8 weight percent in the charged state, substantially reduced (typically below 0.5 weight percent) in the discharged state. Oxygen storage at the oxygen-storing electrode is in the range of 2 to 20 weight percent in the charged state, substantially reduced (typically below 1 weight percent) in the discharged state. Specific energy at the cell level depends on the joint coverages and on the mass fractions of inactive components but typically falls in the range of 100 to 400 watt-hours per kilogram for engineered cells.

Self-discharge is bounded by the activation barriers separating bound and unbound states. Practical implementations exhibit retained charge fractions above 90 percent over storage intervals of one to several months at ambient temperature in the absence of applied load. The metastability of the charged state is the operational mechanism by which the cell holds its charge; it is not a parasitic effect to be minimized but a load-bearing property of the architecture. Cycle life within the envelope is in the thousands of cycles before significant capacity fade, with end-of-life typically determined by membrane degradation rather than by carbon-scaffold degradation.

6. Alternative Embodiments

The architectural class admits multiple embodiments. The carbon source for the two scaffolds may be selected from graphene, graphene oxide, reduced graphene oxide, single- and multi-walled carbon nanotubes, activated carbon of vegetable or pyrolytic origin, carbon black, hierarchical porous carbons, carbide-derived carbons, or composites and blends. The two scaffolds may be of identical or differing carbon source within a single cell.

Catalysts may be selected from platinum-group metals at loadings below 0.5 milligrams per square centimeter, non-platinum transition-metal-based catalysts (iron, cobalt, nickel, manganese in oxide, sulfide, nitride, or carbide form), metal-organic-framework-derived catalysts, single-atom catalysts on heteroatom-doped carbon supports, or metal-free heteroatom-doped carbon catalysts (nitrogen, sulfur, phosphorus, boron, or co-doped). Catalyst incorporation may be by impregnation, electrodeposition, atomic-layer deposition, or in-situ generation during scaffold synthesis.

The polymer electrolyte may be selected from any of the classes named above, in any thickness consistent with the conductivity envelope, in single-layer or multi-layer composite construction. Form factor admits cylindrical, prismatic, pouch, planar-stack, and bipolar-plate geometries. Fabrication methods admit slurry casting, spray deposition, electrospinning, chemical-vapor deposition, plasma-enhanced deposition, hot-pressing of pre-formed components, and roll-to-roll lamination, in any consistent combination.

Optional dopants may be incorporated into one or both carbon scaffolds, including nitrogen, sulfur, phosphorus, boron, and oxygen-functional-group pre-loading, to tune the binding energies of the C-H and C-O states and to shift the operating voltage window within the disclosed range. Conditioning protocols may vary in cycle count, voltage profile, and current density within the disclosed envelope.

7. Composition with Broader System Architecture

The carbon-bound hydrogen-oxygen cell composes with the broader system architecture of a rechargeable energy-storage installation. At the cell level, multiple cells are combined in series and parallel to achieve required pack voltage and capacity. The sealed-boundary property is preserved at the pack level: pack-level enclosures inherit the mass-conservation property of the constituent cells, and no pack-level provision for air ingress, hydrogen resupply, or electrolyte top-off is required.

At the management-system level, the cell composes with a battery-management system that monitors state of charge through coulomb-counting against the joint coverage state, monitors state of health through impedance spectroscopy at the membrane and interfacial scales, and enforces the cycling envelope through current and voltage limiting. The metastability of the charged state simplifies state-of-charge estimation: open-circuit voltage is a near-monotonic function of the joint coverage state on multi-hour relaxation timescales, admitting accurate state-of-charge readout without active learning or model-based estimation.

At the application level, the architectural class admits stationary energy-storage applications where multi-month self-discharge bounds matter (seasonal storage, backup power, emergency reserves), portable applications where mass conservation and the absence of metal consumption matter (consumer electronics, mobile equipment), and transport applications where the absence of bulk liquid water and the resulting orientation-independence matter. The class composes with photovoltaic, wind, and grid-connected charging sources without architectural modification.

8. Class Membership: Eight Required Elements

Class membership is defined by the joint presence of eight architectural elements. Element one is a sealed enclosure that prevents mass exchange between the cell interior and the external environment across the operational lifetime, with electron transport admitted only through the external circuit terminals. Element two is two carbon-based electrodes constructed from a graphitic-carbon active-material class, geometrically separated within the enclosure. Element three is a proton-conducting electrolyte that admits Grotthuss-mechanism proton transport between the two electrodes without bulk liquid water at the cell scale.

Element four is reversible covalent C-H bond storage at one of the two electrodes, in the graphane-class regime of approximately 1 to 8 weight percent at full charge. Element five is reversible covalent C-O bond storage at the other of the two electrodes, in the graphene-oxide-class regime of approximately 2 to 20 weight percent at full charge in carboxyl, hydroxyl, and epoxide functional groups. Element six is reversible water-splitting electrochemistry, in which charging cleaves membrane-sorbed water across the two electrodes and discharging reverses the reaction with recombination at the membrane-sorbed state.

Element seven is substantial mass conservation across the charge-discharge cycle, with cell mass invariant within instrumental precision across the operational lifetime. Element eight is electron flow exclusively through the external circuit, with no internal short and no internal mass-coupled bypass. The eight elements jointly define the class. Any device exhibiting all eight elements is within the class regardless of specific carbon source, specific catalyst, specific membrane chemistry, specific form factor, specific fabrication method, specific dopant inventory, specific operating envelope within the disclosed ranges, specific conditioning protocol, or specific monitoring subsystem. The class admits implementations developed subsequent to filing.

9. Prior-Art Distinctions

The carbon-bound hydrogen-oxygen cell is structurally distinct from each of the recognized prior classes of electrochemical energy-storage devices. It is distinct from PEM and SOFC fuel cells because fuel cells are open-boundary devices: hydrogen is supplied externally to the anode and oxygen is supplied externally to the cathode; no covalent storage of either reactant occurs at the electrodes; and the device performs no internal energy storage. The carbon-bound cell stores both reactants internally, as covalent bonds, within a sealed enclosure.

It is distinct from metal-air batteries (zinc-air, lithium-air, aluminum-air, iron-air, sodium-air) because metal-air batteries require atmospheric oxygen ingress through a permeable cathode and consume a metal anode through oxidation. The carbon-bound cell admits no atmospheric oxygen ingress and consumes no metal; both electrodes are carbon-based and both retain their carbon scaffold across cycling.

It is distinct from intercalation batteries (lithium-ion, sodium-ion, potassium-ion, magnesium-ion) because intercalation batteries store energy through cation insertion into a host lattice, with no covalent bond formation at the active site; the host-cation interaction is van der Waals, weak ionic, or coordination-class. The carbon-bound cell stores energy through covalent C-H and C-O bond formation, which is structurally and energetically distinct from intercalation.

It is distinct from nickel-metal-hydride cells because nickel-metal-hydride cells use a metal hydride at the negative electrode (typically a rare-earth or transition-metal alloy), a nickel oxyhydroxide cathode, and a bulk aqueous potassium hydroxide electrolyte. The carbon-bound cell uses a carbon-hydrogen covalent storage at the negative electrode, a carbon-oxygen covalent storage at the positive electrode, and a polymer membrane with no bulk liquid electrolyte. It is also distinct from solid-state lithium and sodium batteries, from redox flow batteries, and from supercapacitors, in each case along a structurally identifiable dimension of the eight-element class definition.

10. Disclosure Scope

This article discloses the architectural class of carbon-bound hydrogen-oxygen cells under U.S. provisional application 64/052,368, filed April 29, 2026. The disclosure recites the two-scaffold covalent-storage primitive, the sealed-enclosure mass-exchange constraint, the proton-conducting polymer-electrolyte requirement and admissible chemistries, the reversible C-H and C-O bond formation chemistry, the carbon-source generality and surface-area and porosity envelopes, the optional dopant incorporation, the catalyst diversity and loading ranges, the fabrication-method generality, the initial-conditioning electrode-differentiation protocol, the charged-state metastability and self-discharge envelope, the substantial mass-conservation property, the absence of bulk liquid water at any state of charge, the open-circuit-voltage thermodynamic basis, the operating-parameter envelope, the eight-element class-membership definition, the alternative embodiments, the composition with broader system architecture, and the structural distinctions from PEM/SOFC fuel cells, metal-air batteries, intercalation batteries, nickel-metal-hydride cells, and adjacent classes.

The class admits implementations regardless of the specific selection within each disclosed dimension, regardless of specific subsystem implementations, and regardless of specific monitoring or control approaches. The class admits implementations developed subsequent to filing, including implementations using carbon sources, catalysts, electrolytes, dopants, fabrication methods, or conditioning protocols not specifically named in the disclosure but consistent with the eight-element class definition.

This article is presented as a disclosure document of the architectural class. It is not a product specification, performance guarantee, or commercial offering. Specific implementations within the class are subject to the engineering envelope and operating parameters disclosed herein.