NovaSolver›PEM Electrolyzer Hydrogen Production Simulator Back
Energy Engineering
PEM Electrolyzer Hydrogen Production Simulator
A real-time simulator for proton-exchange-membrane (PEM) water electrolyzer stacks. Adjust the cell voltage, current density, number of cells and operating temperature to see hydrogen production rate, HHV efficiency, specific energy consumption and total overpotential update at once. Useful for concept design of green hydrogen plants and for studying the efficiency-versus-current-density trade-off.
Parameters
Cell voltage V_cell
V
Operating voltage per cell. 1.481 V is the thermoneutral voltage; typical values are 1.7-2.0 V.
Current density j
A/cm²
Current per unit electrode area. Higher means more compact but raises overpotential.
Cell area A
cm²
Cells per stack N
Number of cells stacked in series.
Operating temperature T
°C
PEM upper limit is around 80 °C. Higher T lowers the reversible voltage.
Faradaic efficiency η_F
%
Fraction of current actually producing H₂ (after gas-crossover losses).
Results
—
Stack power (kW)
—
H₂ output (kg/hr)
—
H₂ output (Nm³/hr)
—
Specific energy (kWh/kg H₂)
—
HHV efficiency (%)
—
Overpotential (V)
—
PEM cell cross-section — electrolysis animation
At the anode, water splits into O₂ and H⁺; the protons cross the PEM membrane to the cathode, where they recombine with electrons coming back through the external circuit to form H₂. Bubble density scales with current density.
Polarization curve — cell voltage vs current density
I = cell current (= j·A), N = number of cells, η_F = Faradaic efficiency, F = 96485 C/mol. The factor 2 in the denominator is the two electrons required per H₂ molecule. HHV efficiency is simply the ratio of the thermoneutral voltage V_tn = 1.481 V to the actual cell voltage, scaled by η_F.
Temperature-corrected reversible voltage V_rev (1.229 V at 25 °C, Nernst approximation) and total overpotential η_op (activation + ohmic + concentration losses). T is in kelvin.
PEM Electrolyzer Efficiency
🙋
You hear about green hydrogen everywhere — but what is a PEM electrolyzer actually doing? Is it different from the standard electrolysis you do in a chemistry class?
🎓
Loosely: PEM means Proton Exchange Membrane electrolysis. At the anode water splits into H₂O → O₂ + 4H⁺ + 4e⁻, and only the protons (H⁺) cross a thin polymer film (about 50 μm of Nafion). The electrons travel through the external circuit, come back at the cathode, and 2H⁺ + 2e⁻ → H₂ produces hydrogen. Compared with alkaline electrolysis there is no KOH solution, so the unit is compact, and the membrane lets you push much higher current density (PEM ~2 A/cm² vs alkaline ~0.4 A/cm²). It also responds in seconds to changes in power, so it pairs especially well with intermittent solar and wind power. That is why it is treated as the front-runner for green-hydrogen plants.
🙋
With the defaults (1.8 V, 2 A/cm²) the tool gives an HHV efficiency of 80%. Is that good? And what is the theoretical maximum?
🎓
The theoretical ceiling is at the thermoneutral voltage V_tn = 1.481 V, which corresponds to exactly 100% HHV. Below that voltage the reaction is endothermic and would need external heat to proceed, so production essentially stops. So industrial operation always sits above ~1.6 V. 80% at 1.8 V/cell is typical of today's commercial PEM stacks and is a good number. Alkaline electrolysis comes in around 65-70%, and SOEC (solid oxide) can reach 80-90% but needs 700 °C+ heat. Reaching 80% at room-temperature start-up like PEM does is industrially very valuable.
🙋
So if I drop the cell voltage to 1.6 V, efficiency jumps over 90%, right? Why don't operators just do that?
🎓
That is exactly where plant design earns its keep. Try V_cell = 1.6 V on the slider — HHV efficiency does climb past 90%, but in reality the achievable current density drops to 0.3-0.5 A/cm². To produce the same hydrogen flow you need roughly four times the electrode area, so CAPEX explodes. Conversely 2.0 V/cell lets you push 3 A/cm² with very small electrodes but drops efficiency to 74%, raising OPEX. Cheap-power regions (Australian or Middle-Eastern PV) push high current density and small footprint; expensive-power regions push low current density and large area. 1.7-1.9 V/cell is the pragmatic answer most plants settle on.
🙋
If the overpotential is 0.6 V and all of that becomes heat… that is a lot of waste heat, isn't it?
🎓
Sharp. With the defaults: 0.611 V × 1000 A × 100 cells = 61.1 kW of heat dumped (about 34% of the 180 kW input). That is why every commercial stack has a cooling loop circulating ultrapure water to keep the stack near 70 °C. Interestingly, if that heat is recovered into a district-heating loop, the apparent 80% stack efficiency can be lifted to a 90% "system" efficiency. In Japan the Fukushima Hydrogen Energy Research Field (FH2R, ~10 MW) is one of the projects exploring this sector-coupling approach — using heat, oxygen and hydrogen all at once.
🙋
Last one: Faradaic efficiency 98% — where do the remaining 2% go?
🎓
The main culprit is gas crossover. A small fraction of the H₂ produced at the cathode back-diffuses through the PEM, reaches the anode, recombines with O₂ and reverts to water. Current flowed, but no hydrogen came out. Thinner membranes give better proton conductivity but more crossover, so vendors mix and match thinner Nafion NR212 (50 μm) for high efficiency and thicker N117 (175 μm) for low crossover. At low part-load — typical for renewables tracking — crossover gets relatively worse and η_F can fall to ~95%. We use 98% as a sensible default, but for real plant data use the part-load value.
Frequently Asked Questions
Faraday's law gives the molar hydrogen rate per cell as n_H2 = I/(2F) (F = 96485 C/mol; the factor 2 is the two electrons needed per H2 molecule). All N cells of a stack are in series, so the same current I flows through every cell and the stack output becomes n_H2 = N·I/(2F)·η_F, where η_F is the Faradaic efficiency (the share of current that actually produces hydrogen; 97-99% is typical for PEM). This tool computes I = j·A from current density j and cell area A and reports the result in both kg/hr and Nm³/hr.
HHV (higher heating value) is the combustion heat of hydrogen when the product water is liquid: 39.4 kWh/kg. LHV (lower heating value) leaves the product water as vapour: 33.3 kWh/kg. Electrolyzer efficiency is usually quoted on an HHV basis as η_HHV = 39.4 / (input power / hydrogen mass) × 100, expressing "what fraction of the input power became chemical energy in hydrogen". Commercial PEM stacks reach 70-80% HHV, which corresponds to about 60-67% on an LHV basis. Mixing the two gives an 18% difference, so always check which basis a datasheet uses.
Yes. η_HHV = V_tn / V_cell · η_F with V_tn = 1.481 V, so efficiency scales inversely with cell voltage. For V_cell = 2.0 V the efficiency is about 74%; for V_cell = 1.6 V it climbs above 91%. The catch is that lower voltage also implies lower current density, so the electrode area must grow to keep the same hydrogen output. "High efficiency" therefore means a large, expensive plant, while "high current density" means a cheap but lossy plant. Real plants settle around 1.7-2.0 V/cell as a CAPEX-vs-OPEX optimum.
Overpotential is the gap between the actual cell voltage and the reversible voltage V_rev (1.229 V at standard conditions). It is the extra voltage needed to actually drive the reaction. It splits into (1) activation overpotential — the kinetic barrier on the electrode catalyst, dominated by the anode OER; (2) ohmic overpotential — IR losses in membrane and electrodes; and (3) concentration overpotential — gas blanketing at high current density. A typical PEM stack runs at 0.4-0.8 V of total overpotential. All of it becomes heat, so overpotential × current × number of cells equals the waste heat that the cooling loop must remove. This tool computes V_cell − V_rev.
Real-World Applications
Green-hydrogen plants powered by renewables: PEM electrolyzers are the core hardware of Power-to-Gas projects that store surplus solar or wind power as hydrogen. Fukushima Hydrogen Energy Research Field (FH2R, 10 MW class) and REFHYNE at the Shell Rhineland refinery (10 MW class) are commercial-scale examples already in operation. PEM's second-scale response to load swings — vs the tens-of-minutes response of alkaline electrolysis — is the key reason it pairs so well with intermittent renewables.
Hydrogen refuelling stations for fuel-cell vehicles (FCVs): PEM units of 30-300 Nm³/hr (60-600 kg/day) are deployed as on-site producers at hydrogen stations. The tool's default case (1000 A, 100 cells, 180 kW) outputs about 41 Nm³/hr ≈ 80 kg/day, enough to refuel roughly 16 FCVs per day at 5 kg per vehicle. Distributed production of this kind is critical for the rollout of Toyota Mirai-class cars and fuel-cell buses.
Hydrogen feedstock for steel and chemical plants: The industry is shifting away from steam methane reforming (SMR) toward green hydrogen for ammonia and methanol synthesis, refinery hydrotreating and direct-reduced iron. Sweden's HYBRIT pilot replaces coke with hydrogen in steelmaking, cutting CO₂ by ~95%. These industries need hydrogen on the scale of hundreds of thousands of tonnes per year, and PEM is being scaled to meet them.
Grid-service operation: Electrolyzers count as controllable demand. When the grid has surplus power the stack ramps up; when it is short the stack throttles down. In Denmark and Germany electrolyzers have begun bidding into Frequency Containment Reserve (FCR) markets, earning grid-service revenue on top of hydrogen sales. This "multi-use" operating mode is increasingly part of the business case.
Common Misconceptions and Pitfalls
The biggest trap is treating "hydrogen → only water out the tailpipe" as automatically zero-emission. What matters is the power source. Run an 80%-efficient electrolyzer on average Japanese grid power (~0.45 kg-CO₂/kWh) and you emit about 27 kg-CO₂ per kg of hydrogen — worse than steam-methane-reformed "grey" hydrogen at ~10 kg-CO₂/kg-H₂. Some industry voices call this "brown" hydrogen. To label hydrogen "green", essentially 100% renewable power is required. The EU's RED III directive demands three properties — additionality, temporal correlation, geographical correlation — without which the electricity does not count as renewable for hydrogen.
Next, the gap between catalogue efficiency and full-plant efficiency. Even with an 80% HHV stack, the AC-DC rectifier (~95%), circulation pumps, cooling, water purification (5-10% auxiliary load), and hydrogen compression (3-4 kWh/kg from 30 bar to 700 bar) all stack up. The system efficiency typically falls to 60-65%. This tool computes only the stack — for full-system evaluation, add the balance-of-plant losses. Recovering waste heat into district heating, on the other hand, can lift the overall figure back to 75-80%.
Finally, the concern that PEM relies on scarce, expensive iridium. World annual iridium production is only ~9 t, and at today's loading (~2 mg/cm²) building 100 GW of PEM would consume several years of supply. Stack vendors are working hard to cut the loading to about 0.2 mg/cm². Nel, Plug Power, Siemens Energy and others are aligned on tens-of-GW manufacturing capacity by the 2030s. Catalyst cost is not modelled in this tool — at very large plant sizes, always cross-check catalyst availability separately.
How to Use
Set cell voltage (0.5–2.5 V per cell) and current density (0.2–2.0 A/cm²) using sliders or numeric inputs.
Define active cell area (100–500 cm²) and number of cells in stack (10–200 cells).
Read real-time outputs: stack power, H₂ production rate, specific energy consumption, and HHV efficiency percentage.
Worked Example
A 60-cell PEM stack with 250 cm² per cell, 1.8 V per cell, and 1.0 A/cm² current density operates at 27 kW total power. Hydrogen output reaches 9.2 kg/hr (103 Nm³/hr). Specific energy consumption is 48.5 kWh/kg H₂, yielding 70% HHV efficiency. Overpotential across the stack measures 0.35 V due to activation and ohmic losses.
Practical Notes
Operating point selection: Higher current density (1.5–2.0 A/cm²) increases H₂ output but raises overpotential and reduces efficiency; use 0.8–1.2 A/cm² for grid-linked renewable hydrogen plants.
Cell voltage range: Industrial PEM stacks typically run 1.6–2.1 V per cell; below 1.6 V indicates insufficient driving force; above 2.1 V signals excessive losses.
Scaling calculations: Multiply stack power by duty cycle and annual hours to estimate yearly energy demand and H₂ volume for pipeline or storage sizing.