How Is Green Hydrogen Produced? A PV, Wind and Electrolyzer Case Study

How is green hydrogen produced?

Green hydrogen is made by splitting water into hydrogen and oxygen inside an electrolyzer, using electricity from renewable sources such as photovoltaics (PV) and wind. Because PV and wind only generate power intermittently, the electrolyzer’s output follows the weather — which is why the ratio of solar, wind and electrolyzer capacity, not just the technology itself, determines how much hydrogen a plant can reliably deliver.

How a PV+wind hydrogen plant is actually wired together

In the most common real-world setup, PV and wind feed an electrolyzer directly, hydrogen goes to a demand “behind the fence,” and any leftover power is sold off as residual electricity.

From there the value chain branches: hydrogen typically passes through low-pressure storage, then — if higher pressure is needed — a compressor feeds medium-pressure storage for pipeline injection or trailer export, or high-pressure storage for vehicle refueling. Batteries, direct industrial use and grid off-take sit alongside the electrolyzer as alternative destinations for the electricity itself, not the hydrogen.

Everything else — fuel cells, CHP units, burners, chemical use — can be added on top of this core setup later. It rarely changes the fundamental result, which is why the first and most important step is sizing the PV/wind/electrolyzer combination itself.

Why the “right ratio” depends on the demand profile

The PV/wind/electrolyzer ratio only makes sense once you know what the hydrogen is actually for — continuous chemical demand, shift-based production, or refueling.

An ammonia plant or chemical derivative process typically needs continuous, 24/7/365 supply, usually on-site. Hydrogen for industrial processes such as soldering or surface treatment follows a production schedule — often five days a week, during shifts. Hydrogen for mobility depends on trailer availability and refueling patterns, which themselves vary by use case (retail stations run roughly 6am–9pm, bus depots more like 8pm–2am). Every one of these profiles carries some uncertainty — but almost every customer still expects supply to be effectively 100% reliable.

Setting up the case study: one electrolyzer, four locations, eight ratios

To find the pattern, the case study fixed the electrolyzer at 100 MW and tested eight PV/wind combinations at four very different renewable-resource locations.

The electrolyzer assumptions were kept simple and technology-agnostic: a specific energy consumption of 57 kWh/kg (typical for both PEM and alkaline including balance-of-plant) and a power turndown range of 10–100%. PV and wind capacity were then varied from zero up to 4× the electrolyzer’s rating.

The four sites were chosen for contrast rather than realism (no permitting, land or grid factors were considered): Amarillo, Texas — Panhandle solar and wind widely seen as one of the best hydrogen locations in any industrial country; Larissa, Greece — a stable-solar, low-wind agricultural region already connected to the gas pipeline network; Rajkot, India — near the world’s largest refinery, with excellent year-round solar and monsoon-affected wind; and Fraserburgh, Scotland — sitting on roughly 20% of Europe’s wind resource, but with a weak electrical grid heading south.

PV-only vs. wind-only vs. combined — what the weekly profiles actually look like

Wind-only production swings hardest, PV-only is smooth but capped, and combining both narrows the gaps without eliminating them.

A “PV only” week produces a clean, repeating pattern of daytime peaks with predictable gaps overnight. A “wind only” week can look excellent for days and then collapse for nearly as long — and doubling wind capacity mostly amplifies both the peaks and the troughs rather than smoothing them out. Combining wind and solar shrinks the gaps and lifts the floor, but even then there are periods — a “Dunkelflaute,” when neither wind nor solar produce enough electricity — that no renewable ratio alone can fully solve.

The Amarillo results make the pattern concrete. Against a theoretical ceiling of about 294 tons of hydrogen per week (100 MW running continuously at 57 kWh/kg, roughly 15,300 t/yr):

What changes when the geography changes

The same PV/wind capacity behaves very differently depending on where it’s built — and every location has its own weak season.

Larissa, Greece

Solar is remarkably stable across roughly 35 weeks a year, though winter capacity factor drops to about half of the summer value. Wind alone gives a useful but highly variable baseload of 40–250 t/week. Combining 200 MWp PV with 200 MW wind produces a smooth ~150 t/week at around 50% capacity factor; doubling wind again shifts that to ~200 t/week, but pushing PV and wind further barely moves the needle.

Rajkot, India

Solar is excellent and constant nearly all year; wind is fairly steady but strongly seasonal, peaking in summer and fall. Wind alone is the least predictable option here — some weeks near 90 t/week, others above 280 t/week. Even the best PV+wind combination (Case 7) still runs into monsoon weeks 37–47, where output drops and — if the off-taker is, say, a refinery expecting steady supply — a bridging solution is unavoidable.

Fraserburgh, Scotland

Wind is strong and constant; solar is weak and highly seasonal, as expected this far north. Even a modest 2:1 wind-to-electrolyzer ratio delivers 90% capacity factor in many weeks — but some weeks fall below 50%, and with no local grid, there is no buyer for the surplus power those good weeks produce.

Why doubling renewable capacity rarely doubles hydrogen output

Across every location tested, doubling PV or wind capacity added roughly 12–24% more annual hydrogen — never close to double.

The pattern holds everywhere: once the electrolyzer is already saturated for part of the week, extra PV or wind capacity mostly produces electricity the plant can’t turn into hydrogen. The 400/400 ratio gets every location close to its ceiling — but it also raises the same question every time:

The real constraint isn’t production — it’s reliability

No off-taker accepts “hydrogen if it’s available” — which means storage, not just bigger renewables, is what actually secures a supply agreement.

A pipeline with genuinely unlimited demand can absorb whatever an electrolyzer produces, whenever it produces it. Almost nothing else works that way. Ammonia plants, refineries and refueling networks all expect dependable delivery, so the weak weeks — Dunkelflaute in Texas and Scotland, winter in Greece, monsoon in India — have to be bridged with storage: a public pipeline acting as a buffer, underground or salt-cavern storage, or a battery. A battery can help smooth short gaps, but it adds cost on top of an already renewables-heavy CAPEX, so it rarely solves the problem outright.

The takeaway for sizing a real project

Combine PV and wind at roughly a 1:1 ratio with the electrolyzer for a stable ~50% capacity factor, then deliberately decide whether to chase higher utilization with storage — or accept a smaller, reliable “behind-the-fence” plant.

An even electricity production profile is close to mandatory for any bankable business case. If “make hydrogen” is the answer to “there’s no grid,” the project still needs an answer for what happens to surplus electricity. And if “use the surplus to make more hydrogen” is the answer to curtailment, the project still needs enough full-load hours to make the extra equipment worthwhile. These two unresolved questions, more than any single technology choice, are why a number of green hydrogen projects were paused or cancelled in 2024 and 2025.

Watch a clip from the Webinar

Read next

Article based on a Mission Hydrogen webinar. Technical figures are the speaker’s field estimates, presented for general understanding rather than as design specifications.