The Colors of Hydrogen, Ranked: Which Ones Actually Pencil Out?

Why We Color-Code a Colorless Gas

Hydrogen doesn’t exist lying around in a tank. You have to make it, and making it is where all the carbon and all the cost hide. So the entire rainbow collapses to two numbers: dollars per kilogram (the cost) and kilograms of CO₂ per kilogram of hydrogen (how dirty it was to make). Every color is just a different spot on that map — cost on one axis, carbon on the other — and once you see it that way, the marketing falls apart in your hands.

Let me give you the verdict up front. Most of the rainbow is marketing. Today, only two colors are honest near-term bets; one is a genuine wildcard that could change everything or amount to nothing; and the part nobody tells you is that the subsidies, not the chemistry, quietly decide all of it.

The hydrogen from a coal plant is the exact same molecule as the hydrogen from a solar-powered electrolyzer. The color is just a marketing word for the recipe.

Grey: The Cheap, Filthy Incumbent

Grey is what basically all hydrogen is right now. Take natural gas, blast it with steam, strip the hydrogen off the methane. It’s a hundred-year-old process, it’s cheap, and it’s everywhere: global hydrogen demand hit about 97 million tonnes in 2023, almost all of it from unabated fossil fuels. So when someone tells you we already have a hydrogen economy — we do, and it’s grey.

Grey costs somewhere around $1–2/kg, dirt cheap. And here’s the number that defines this whole thing: making one kilogram of grey hydrogen throws off roughly 9–12 kg of CO₂. The fuel we’re selling as clean emits about ten times its own weight in carbon dioxide. Its dirtier cousin, brown hydrogen — cracked out of coal, big in China — runs higher still, around 18–20. That’s the baseline. Cheap, filthy, and the bar every clean color has to beat.

Blue: Grey With a Catch

Blue is grey with carbon capture bolted onto the back end — same cheap gas, minus the emissions, around $2–3/kg. It’s the one the oil and gas majors love, because it keeps natural gas in the game. But it has two problems the brochures skate past. The first is the capture rate: to call this clean you need to grab well over 90% of the carbon, and a lot of real-world plants capture nowhere near that. The second is sneakier — methane leakage. Natural gas is methane, methane leaks on its way to the plant, and it’s a brutal greenhouse gas. A well-known 2021 study by Howarth and Jacobson ran the leakage at 1–4% and concluded blue’s full lifecycle footprint can be worse than just burning the gas.

Honest blue is a decent bridge. Dishonest blue is a fossil lifeline wearing a clean costume.

Turquoise: The Clever One

Turquoise starts from the same natural gas, but instead of burning the carbon it cracks the methane apart with heat so the hydrogen comes off as gas and the carbon comes out as a solid — literally a black powder, carbon black, the stuff in tires and ink. Because the carbon never becomes CO₂, there’s no gas to capture and bury; you just sweep up the solid. The base case lands near $1/kg of hydrogen — but only if you can sell the carbon black, and if everyone starts cracking methane you’d flood that market and crash the price the whole model depends on. It’s real, though: a company called Monolith has run a commercial-scale plant in Nebraska since 2020, making a couple thousand tonnes of hydrogen and around fourteen thousand tonnes of carbon black a year. Turquoise isn’t science fiction; it’s a “does the byproduct math hold at scale” question.

Green: The Headliner

Green is the one everybody pictures — split water with electricity in an electrolyzer, and if that electricity comes from solar or wind, the carbon is basically zero. The dream is real; the problem is the price tag. Green runs roughly $3–6/kg today, and the IEA has seen it as high as $12. Here’s the single most important fact about it: electricity is 55–70% of the cost. Green hydrogen is basically your electricity bill wearing a lab coat. Which means green isn’t really one number — it’s a bet on cheap, clean power. And one catch people miss: green is only green if the grid is. Plug an electrolyzer into a coal-heavy grid and you can make hydrogen dirtier than grey — which is exactly why the rules got so strict.

Pink and White: The Cousin and the Wildcard

Pink is green’s nuclear cousin — same water-splitting, but the electricity comes from a reactor. The clever part: a reactor runs flat out around the clock, so your expensive electrolyzer works most of the time instead of sitting idle, and that utilization quietly improves the economics. Pink rides whatever happens with nuclear, so it’s a watch-item.

Then there’s white hydrogen — also called gold, or natural, hydrogen. You don’t make this one at all; it’s down there, made by the planet, and you’d drill for it. Sounds like a fairy tale, except a village in Mali called Bourakébougou has literally run on a natural-hydrogen well for years, and researchers found the reservoir even seems to refill itself. If white hydrogen is real at scale, it’s potentially clean and cheap straight out of the ground — which would blow up this entire ranking. But right now it’s almost pure exploration; nobody’s proven you can pull it out economically at scale, so anyone quoting you a price is guessing.

The Money Map: Subsidies Do the Ranking

None of these clean colors beat grey’s $1.50 on cost alone — not yet. So governments pay the difference, and how they pay reshuffles the whole board. The big one is the US clean-hydrogen tax credit, 45V. It pays up to $3/kg — enough to make green flat-out competitive with grey — but it’s tiered by how clean your hydrogen actually is over its full lifecycle. Full credit requires coming in under about 0.45 kg CO₂e/kg; above 4 kg, you get nothing. Hold onto that 4 kg line, because it’s the gate. Grey at ten and brown at eighteen aren’t close. Honest green and pink sail under it. Blue has to fight for it. The carbon number literally decides who gets paid.

And because my map is North America, here’s Canada: the same play through a different door. Instead of paying per kilogram, Ottawa hands back 15–40% of a project’s capital cost, tiered by the exact same carbon math — full 40% under 0.75 kg, nothing at 4 kg or more. Same gate, different currency. They even folded in methane pyrolysis, so turquoise gets paid north of the border too, and carbon-capture gear has its own 50% credit. Europe runs it differently again — the EU Hydrogen Bank auctions a fixed subsidy per kilogram (its first auction cleared around €0.48/kg), and CBAM, the carbon border tariff, now taxes the carbon in imported hydrogen. And China is playing a completely different game: instead of subsidizing the hydrogen, it’s winning the machines — Chinese electrolyzers run about a quarter the price of Western ones.

The chemistry barely moved, but the economics flipped overnight. A politician with a pen can re-rank this entire industry in an afternoon.

That’s not a metaphor. In 2025 a US budget bill moved up 45V’s deadline, so to even qualify you now have to start building before 2028. And this spring, Ottawa redrew its industrial carbon-price curve — from a line pointing at $170/tonne by 2030 down to about $115 — re-pricing every capture project in the country with one announcement. A single rule change re-scores every hydrogen project at once.

The Verdict

Green — a real bet. Genuinely clean, expensive today, but it’s where the money and the policy are actually built, and it works wherever clean power is cheap. Blue — a real bet, conditionally: a legitimate bridge if it’s honest about capture and leakage, a fossil lifeline if it isn’t. Watch the capture rate like a hawk. Turquoise — a watch: clever, genuinely lower-carbon, real hardware in the ground, but it lives or dies on selling that solid carbon. Pink — a watch: quietly sensible, riding the nuclear wave. White — the wildcard: it could rewrite everything or be a mirage, and we won’t know for a few years. Grey and brown — the incumbents: cheap and filthy, the thing every other color is trying to kill.

Here’s the one thing to keep. Next time anyone pitches you hydrogen, ask exactly two questions: what does it cost per kilogram, and how many kilograms of carbon did it take to make one kilogram of the stuff. Cost and carbon. The color is marketing, but those two numbers are physics and money — and they don’t lie.