How to Make a Hydrogen Fuel Cell and How It Works

A hydrogen fuel cell generates electricity through a chemical reaction between hydrogen and oxygen, producing only water and heat as byproducts. The core process is surprisingly simple: hydrogen molecules split into protons and electrons at one electrode, the electrons travel through an external circuit (powering whatever you connect), and everything recombines with oxygen at the other electrode to form water. Building one yourself ranges from a weekend science project with household materials to a sophisticated engineering challenge involving specialized membranes, catalysts, and precision assembly. The gap between those two extremes is where most of the interesting engineering lives.

How a Hydrogen Fuel Cell Generates Electricity

A fuel cell works like a battery that never runs down, as long as you keep feeding it fuel. On one side, called the anode, hydrogen gas flows in. A platinum-based catalyst causes each hydrogen molecule to split into two protons and two electrons. The protons pass through a thin membrane that only allows positively charged particles through. The electrons, blocked by the membrane, are forced to take the long way around through an external circuit. That flow of electrons is your electricity. On the other side, called the cathode, the protons and electrons meet up again, react with oxygen from the air, and produce water.

The membrane at the center is usually a material called Nafion, a fluoropolymer with sulfonic acid groups that create channels for protons to hop through. At different moisture levels, protons move through Nafion by different mechanisms: hopping along the membrane’s surface, jumping between water molecules in a chain-like relay, or simply drifting as part of water molecules carrying a charge.¹ This is why keeping the membrane properly hydrated matters so much, as we will see later.

The theoretical efficiency of this process is a point of genuine debate among researchers. A common claim in textbooks is that fuel cells are not limited by the same thermodynamic ceiling that constrains heat engines. One analysis showed that the maximum fuel cell efficiency actually can be derived from the same principles as that ceiling, suggesting the advantage is overstated.² Other researchers have pushed back, demonstrating that under certain conditions, a fuel cell’s theoretical efficiency can exceed that of a heat engine operating between the same temperatures.³ In practice, a typical PEM fuel cell converts roughly 40 to 60 percent of the hydrogen’s energy into electricity, with the rest becoming heat.

The Parts That Make It Work

Every PEM fuel cell shares the same basic architecture, layered like a sandwich. At the center sits the proton exchange membrane. On each side of the membrane is a thin catalyst layer, usually platinum nanoparticles deposited on carbon. Outside each catalyst layer is a gas diffusion layer (GDL), a porous sheet typically made of carbon paper or carbon cloth that lets reactant gases reach the catalyst while also conducting electrons away. The outermost layers are bipolar plates with machined channels that distribute hydrogen on one side and air or oxygen on the other.

The gas diffusion layer does double duty. Beyond distributing gas evenly across the catalyst, it manages water. Researchers have found that treating the GDL with PTFE (the same material as nonstick cookware coatings) and controlling how that coating is distributed significantly affects performance, particularly under humid conditions where water removal becomes critical.⁴ A uniform PTFE distribution creates better pathways for liquid water to escape without blocking the gas flow.

Bipolar plates are where the flow field design lives. The channels carved into these plates guide hydrogen and oxygen to the reaction sites. Common channel patterns include serpentine (a single winding path), interdigitated (dead-end channels that force gas through the diffusion layer), and parallel (straight channels side by side).⁵ Recent work on biomimetic designs inspired by airfoil profiles has achieved power density improvements of around 6 percent over conventional patterns, with particular gains under high-humidity conditions where water tends to clog channels.⁶

Assembling the Membrane Electrode Assembly

The membrane electrode assembly (MEA) is the heart of any fuel cell, and building one is where the difference between a classroom demo and a functional power source becomes clear. In a professional setting, the catalyst layer is applied to the membrane or GDL by ink coating, spraying, or decal transfer, then the whole sandwich is fused together using a hot press.

The pressing conditions make a real difference. A systematic investigation of MEA fabrication found that temperature was the single most important variable, followed by pressure and the interaction between time and temperature. The study identified that lower temperatures and moderate pressures produced better results, with the best performance coming from pressing at 100°C, about 1,000 psi, for 2 minutes.⁷ Too much heat or pressure can damage the membrane or crush the porous catalyst structure, reducing the area where reactions can happen. Too little and the layers do not bond well, creating high resistance at the interfaces.

For someone building a simple demonstration cell at home or in a classroom, the approach is much more forgiving. Basic setups can use pencil graphite electrodes, a small piece of membrane material, and a container of water with a battery to first electrolyze water into hydrogen and oxygen, then reverse the connections to generate a small voltage. You will not power anything meaningful, but you can observe the fundamental principle at work and measure the voltage with a multimeter. Kits designed for this purpose are commercially available and typically produce milliwatts of power.

From Single Cell to Full Stack

A single fuel cell produces less than one volt, so practical applications require stacking dozens or hundreds of cells in series, like stacking batteries end to end. This is where the engineering gets tricky. The stack needs to be clamped together tightly enough to prevent gas leaks and maintain good electrical contact between layers, but not so tightly that you crush the porous gas diffusion layers and choke off gas flow.

Research on this balancing act has found that as stacks grow larger, the required clamping pressure increases. A 100-cell stack needs about 0.7 megapascals of clamping force for effective sealing, while a 400-cell stack requires around 0.9 megapascals.⁸ The voltage output of individual cells within the stack is not uniform either. Cells near the edges behave differently from those in the middle, and the output follows an uneven pattern along the clamping direction.

Sealing between cells is a persistent challenge. Stacks use various gasket materials including silicone rubber, EPDM, and fluoroelastomers, each with different trade-offs in chemical resistance, long-term compression behavior, and tolerance to the acidic, hot conditions inside the stack.⁹ Manufacturing tolerances, slight misalignment of bipolar plates, and thermal expansion during operation all conspire to create uneven pressure distributions that can lead to leaks or localized performance losses. Bolt torque patterns, gasket material selection, and frame design all play roles in managing this.

The Water Balancing Act

Water management is one of the trickiest aspects of operating a PEM fuel cell. The membrane needs to stay moist to conduct protons effectively, but the cathode constantly produces water as a reaction product. Too much water floods the gas channels and blocks oxygen from reaching the catalyst. Too little dries out the membrane and increases its resistance, choking performance from the other direction.¹⁰

In air-breathing cells (the kind you might build as a project, which draw oxygen passively from surrounding air rather than using a pressurized supply), this problem is even more pronounced. The thermal behavior of the cell determines whether it floods or dries out at a given operating point. Reducing thermal resistance pushes the transition between these two failure modes to higher current densities, broadening the useful operating range. But push too far, and the cell can only flood, never dry out, because it cannot shed heat fast enough to evaporate excess water.¹¹

Real fuel cell systems manage water through a combination of humidity control of incoming gases, careful temperature regulation, and flow field designs that naturally sweep water out of channels. Some designs use gravity, others use pressure differences, and some actively recirculate humidified exhaust gas.

Thermal Management

Roughly half the energy in the hydrogen that enters a fuel cell becomes heat rather than electricity. In a large stack, that heat must be removed continuously and evenly. Uneven temperatures across the stack create hot spots that accelerate degradation and cold spots where water condenses and causes flooding.

Cooling approaches range from passive heat spreaders and dedicated air channels for small stacks to liquid coolant loops for larger systems. Liquid cooling using water or antifreeze is the standard for automotive-scale stacks, where dedicated cooling plates sit between groups of cells.¹² The design of these cooling plates and the coolant flow rate strongly influence temperature uniformity. Higher coolant flow rates reduce temperature differences across the plate, improving consistency across the stack.¹³ More exotic approaches include evaporative cooling and boiling-based phase change systems, though these remain less common in commercial products.

What Wears a Fuel Cell Down

PEM fuel cells degrade through several mechanisms, and the membrane is often the weakest link. Hydrogen peroxide forms as a side product during operation, primarily at the anode, and it decomposes into highly reactive hydroxyl radicals.¹⁴ These radicals attack the membrane’s chemical structure, stripping away the sulfonic acid groups that allow proton conduction. The result is a membrane that conducts protons less effectively and eventually develops physical holes, allowing hydrogen and oxygen to cross over and react directly, which generates more radicals and accelerates the damage.¹⁵

The platinum catalyst is also vulnerable. Carbon monoxide, even in trace amounts in the hydrogen supply, binds strongly to platinum and blocks the sites where hydrogen would normally react.¹⁶ This is called catalyst poisoning, and it is the reason PEM fuel cells are so picky about hydrogen purity. Hydrogen produced by reforming natural gas or methanol always contains some CO and needs to be cleaned up before use, or the cell performance drops quickly. Over time, platinum nanoparticles also migrate, dissolve, and clump together (a process called Ostwald ripening), reducing the available surface area for reactions.

Moving Beyond Platinum

Platinum is expensive and scarce, which is a major reason fuel cells remain costly. A significant research effort aims to replace it with cheaper alternatives. The most promising candidates are iron-nitrogen-carbon catalysts: essentially nitrogen-doped carbon materials with individual iron atoms dispersed throughout the structure. These catalysts have shown adequate activity for the oxygen reduction reaction at the cathode, and researchers have begun testing them under conditions relevant to automotive use.¹⁷ Durability remains the biggest challenge. These catalysts degrade faster than platinum under the acidic, oxidizing conditions inside a fuel cell, though progress has been steady.

PEM Versus Solid Oxide Fuel Cells

PEM fuel cells are not the only type. Solid oxide fuel cells (SOFCs) use a ceramic electrolyte and operate at much higher temperatures, typically 600 to 1,000°C compared to the 60 to 80°C range for PEM cells. SOFCs can run on a wider variety of fuels, including natural gas and biogas, and use cheaper catalysts because the high temperatures speed up reactions without needing platinum. They are better suited for stationary power generation, where their waste heat can be captured for heating or industrial processes.¹⁸

PEM cells win on portability, startup speed, and power density. They can go from cold to operating in seconds, while SOFCs take minutes to hours to reach their working temperature. This makes PEM the clear choice for vehicles and portable electronics. Alkaline fuel cells, phosphoric acid fuel cells, and molten carbonate fuel cells also exist, each with its own operating temperature range and niche applications, but PEM and SOFC dominate current research and commercialization.

Storing and Supplying Hydrogen

Building a fuel cell is only half the puzzle. You also need a reliable hydrogen supply. For vehicles, the most common approach is compressed gas stored in carbon fiber-reinforced tanks at pressures around 350 to 700 bar. These tanks are engineered to survive crashes and extreme temperatures, but they are bulky and expensive.

An alternative being explored is metal hydride storage, where hydrogen is absorbed into a solid metal alloy and released by heating. Researchers have investigated integrating these systems into small fuel cell vehicles, with a focus on how the thermal management of the storage system interacts with vehicle-level performance.¹⁹ The advantage is that metal hydrides store hydrogen at much lower pressures and higher volumetric density, potentially making systems safer and more compact. The EU’s HYDRIDE4MOBILITY project has developed and tested integrated hydrogen storage and refueling systems paired with PEM fuel cells for utility vehicles.²⁰

For a home or classroom project, hydrogen supply is simpler but still requires caution. Small electrolyzer cells can split water into hydrogen and oxygen on demand, and some demonstration kits feed this directly to a miniature fuel cell. At this scale, the amounts of gas are tiny and the pressures negligible, but hydrogen is still flammable and should be handled in a well-ventilated space.

Hydrogen Safety in Practice

Hydrogen is lighter than air, odorless, colorless, and burns with an almost invisible flame. These properties create real safety concerns, particularly in enclosed spaces like vehicle cabins. Research on hydrogen leak scenarios in fuel cell vehicles has found that sensor placement matters enormously. A strategy of placing sensors toward the front and middle of the vehicle, with additional coverage at the rear, enables the fastest detection of leaks from the most likely failure points. Opening windows, even partially, significantly reduces hydrogen concentration inside the cabin and lowers the risk of both explosion and oxygen displacement.²¹

For anyone building or experimenting with fuel cells at home, the safety considerations scale with the amount of hydrogen involved. A small demonstration cell producing a few milliliters of gas per minute poses minimal risk in an open room. A system producing liters per minute for a meaningful power output demands proper ventilation, leak detection, and flame-resistant materials nearby. Hydrogen’s flammability range in air is wide, from about 4 to 75 percent by volume, which means even relatively small leaks can reach ignitable concentrations in poorly ventilated enclosures.

How Green Is a Hydrogen Fuel Cell Vehicle, Really

The environmental case for hydrogen fuel cells depends almost entirely on how the hydrogen is produced. A lifecycle analysis comparing fuel cell and battery electric vehicles in India found that a fuel cell vehicle running on green hydrogen (produced by renewable-powered electrolysis) emits about 105 grams of CO₂ per kilometer, less than a battery electric vehicle charged from the current grid mix at roughly 185 grams per kilometer.²² Even fuel cell vehicles using blue hydrogen (from natural gas with carbon capture) came in lower than battery vehicles on the current grid in that study.

Cold climates complicate the picture. A separate lifecycle assessment found that hydrogen fuel cell vehicles using conventional materials could emit up to 367 grams of CO₂ per kilometer on highways in severe winter conditions, while battery electric vehicles in the same scenario ranged from about 86 to 233 grams per kilometer depending on materials.²³ Lightweight materials like carbon fiber composites brought the fuel cell vehicle figure down to around 244 grams per kilometer. The takeaway is that neither technology has a universal emissions advantage. Grid carbon intensity, climate, driving patterns, vehicle weight, and hydrogen source all shift the calculus. A fuel cell vehicle fueled by gray hydrogen from unabated natural gas reforming, which is still the most common production method globally, loses its climate advantage entirely.