Bankable
Real Bet? · RB02 · Series-Opener Due-Diligence Report

The colors of hydrogen, ranked: which ones actually pencil?

Hydrogen is one molecule; the "color" is just how you made it. We put every color on one cost-vs-carbon map, add the subsidies, and rank which are real bets — because a single policy line, not the chemistry, quietly decides who gets paid.

Verdict: two real bets (green; honest blue) · one wildcard (white) · the rest is marketing
Coverage: Hydrogen production pathways (grey/brown/blue/turquoise/green/pink/white) Prepared by: Bankable (over a decade of FOAK energy commercialization) Sources: 18 (agency / gov / peer-reviewed lead) Basis: public data only · $/kg-H₂ · lifecycle kg CO₂e/kg

00Executive summary

One molecule, a rainbow of recipes — and only two numbers that matter.

"Hydrogen is the future" is a slogan, not an answer. Which hydrogen? The colors are a marketing vocabulary for the production pathway, and they hide a 3–6× cost spread and a 40× carbon spread. Two numbers settle every argument: dollars per kilogram, and kilograms of CO₂ per kilogram of hydrogen. Map all the colors on those two axes, overlay the subsidies, and the ranking falls out — with one twist that surprises almost everyone.

$1–2 → $3–6 /kg
The cost spread: grey (the incumbent) at $1–2/kg versus green at $3–6/kg (the IEA's wider band runs to $12). Roughly a 3–6× gap across the rainbow.[9][2]
4 kg CO₂e/kg
The gate. The US 45V credit pays nothing above 4 kg CO₂e/kg and the full $3/kg only at ≤0.45. Grey (~10) and brown (~18) are nowhere near it; this one line ranks the bets.[3][4]

Here is the twist. Honest green hydrogen costs ~$4.5/kg and loses to grey on raw price — until the 45V credit, earned only by clearing that carbon line, drags its net cost down to ~$1.50/kg, level with the dirtiest incumbent. The chemistry didn't change; the subsidy did. That is the whole episode: the 4 kg line, not the color, gates the money. And in 2025 a budget bill moved that line's deadline — a rule change that re-scores every project at once.[5]

The portable takeaway Next time someone sells you a "clean hydrogen" project, ask the two numbers — $/kg and kg CO₂/kg — and then ask which side of the 4 kg subsidy line it lands on. Color is branding. Carbon intensity and the credit schedule are the business.

01The colors, fast

Same H₂, different recipe — and the catch in each

Grey — steam-methane reforming of natural gas, no capture. The incumbent: ~$1–2/kg, ~9–12 kg CO₂/kg, and ~95% of the world's ~97 Mt of hydrogen in 2023 was made from unabated fossil fuels.[1][7][9] The bar everything else has to beat — and the thing being greenwashed whenever someone calls hydrogen "clean" by default.

Brown / black — coal gasification. Even dirtier than grey (roughly ~18–20 kg CO₂/kg, route-dependent), and dominant in China's installed base.[7] The incumbent to retire.

Blue — grey plus carbon capture (SMR or ATR + CCS), ~$2–3/kg.[9] The honesty test has two parts, both cited: the real capture rate (proponents need "well over 90%"; real-world SMR plants capture far less) and upstream methane leakage (Howarth & Jacobson modeled 1–4%, ~3.5% midpoint, and found blue's lifecycle footprint "quite high" — in the worst case higher than just burning the gas).[8] Real bet only if it's honest about both; a fossil lifeline otherwise.

Turquoise — methane pyrolysis: CH₄ → solid carbon + 2 H₂, so the carbon comes out as a solid and there's no CO₂ to capture or store.[11] Monolith's Olive Creek 1 (Lincoln, Nebraska) has run at commercial scale since 2020 (~2,500 t/yr H₂ + ~14,000 t/yr carbon black).[10] Base-case economics reach ~$1/kg — but only if the carbon-black byproduct sells, and flooding that market would crash its price.[11] Watch: byproduct economics are the whole ballgame.

Green — electrolysis from renewables. The headline: ~$3–6/kg today (IEA's band stretches to $4–12, falling toward $2–9 by 2030), near-zero operational carbon if the power is genuinely clean.[2][9] The catch is in that "if": on a dirty grid, electrolytic hydrogen can be dirtier than grey — which is exactly why 45V scores lifecycle carbon, not nameplate. And electricity is 55–70% of green's cost, so green's price is really the power price in disguise.[15][3]

Pink (nuclear electrolysis) — same near-zero carbon as green; its edge is a high capacity factor (an electrolyzer runs ~90%+ of the time on nuclear vs ~30–50% on solar/wind), which slashes per-kg capital cost. Qualifies for 45V like green. We ship no standalone $/kg — kept qualitative.[3] Yellow (solar-only electrolysis) folds into green. White / gold (naturally occurring geologic H₂) is the wildcard: the Bourakébougou, Mali reservoir powers a village and was characterized in Nature Scientific Reports as spontaneously recharging, with newer claims in France's Lorraine basin and elsewhere.[12] "Cheap if real" — but at-scale extraction economics are unproven, so we ship no cost number. Huge if true.

02The comparison matrix

Every color on cost, carbon, maturity — and the verdict

ColorHow madeCost $/kgCO₂ kg/kgMaturityWhere it pencilsVerdict
GreySMR, no capture$1–2 [9]9–12 [7]Commercial (95%+ supply) [1]Today's industry, until carbon is pricedIncumbent
Brown/BlackCoal gasification~$1–2~18–20 [7]·MEDCommercial (China base)Nowhere cleanOverhyped
BlueSMR/ATR + CCS$2–3 [9]~1 → >grey [8]Components commercialHigh capture + low leak onlyReal if honest
TurquoiseCH₄ pyrolysis → H₂ + solid C~$1* [11]low if heat clean [11]Commercial demo (2020) [10]When carbon black sellsWatch
GreenRenewable electrolysis$3–6 (–12) [2][9]~0 if clean power [3]Commercial, scalingSubsidy + cheap-power nichesReal Bet
PinkNuclear electrolysispower-driven~0 [3]Rides nuclear build-outHigh-capacity-factor sitesWatch
WhiteGeologic, extractedunproven [12]~0 (natural)Exploration onlyIf/where it's foundWildcard

*Turquoise ~$1/kg holds only if the carbon-black byproduct sells. Point values are illustrative midpoints of the sourced ranges; brown's exact carbon intensity is a medium-confidence peer-LCA figure (lead with grey's sourced 9–12 and read brown as "even dirtier"). Pink and white carry no cost number on purpose — the sourcing gate forbids a $/kg we can't stand behind.

03The cost-vs-carbon map

The money chart — and the subsidy zone that decides who gets paid

45V gate · 4 kg left of here = eligible $0$1$2 $3$4$5$6 Cost ($/kg H₂) 048 121620 Lifecycle carbon intensity (kg CO₂e/kg H₂) Green$4.5 · ~0 (clears full credit) Blue$2.5 · ~1.5 (honest case) Turquoise~$1 · low* Grey$1.5 · ~10 (the incumbent) Brown$1.5 · ~18 Pink White (carbon ~0; cost not shipped — dashed)
Each color plotted on cost ($/kg, vertical) vs lifecycle carbon (kg CO₂/kg, horizontal). The shaded band left of the dashed line is the 45V-eligible zone (≤4 kg); the darker sliver is the full-$3 credit zone (≤0.45 kg). Grey and brown sit far to the right — outside the money. Subsidies act as a vertical shift, pulling eligible colors down the cost axis. [S2][S3][S7][S8][S9]

04The money map

Incentives are tiered by carbon — so policy, not chemistry, reshuffles the board

Every major subsidy is keyed to lifecycle carbon intensity. A color's kg CO₂/kg decides which credit it qualifies for, and the credit decides whether it pencils. Five levers run the table:

LeverWhat it paysWho it favors
US 45V Clean H₂ PTCUp to $3/kg, four-tier by lifecycle CI. Full credit at ≤0.45 kg; zero above 4 kg. Final rules (Jan 2025) lock the "three pillars" — additionality, hourly matching by 2030, deliverability.[3][4]Honest green & pink (clean power)
US 45Q CCS credit$85/t CO₂ for saline storage, $180/t for direct air capture (2024–26 vintages).[6]Blue & any capture play
Canada Clean H₂ ITCA refundable 15–40% investment tax credit on project capital, tiered by the same lifecycle carbon math — full 40% under 0.75 kg, nothing at 4 kg or more. Methane pyrolysis (turquoise) added Dec 2024; capture equipment gets its own 50% CCUS ITC.[16][18]The same 4 kg gate, paid as capex
Canada carbon priceThe industrial price trajectory was redrawn May 2026: from a line pointing at $170/t by 2030 down to ~$115/t ($95 held in 2026; $130 by 2035).[17]Re-prices every capture project
EU Hydrogen BankA fixed €/kg premium over 10 years. First auction (2024) cleared at €0.48/kg (ceiling €4.5); awards ran €0.33–1.88/kg; a later round cleared near €0.60/kg.[13]RFNBO (EU-additional) green
EU CBAMThe carbon border tariff now lists hydrogen as a covered good; definitive charging phases in from 2026, pricing carbon into imported grey/blue.[13]Penalizes dirty imports
ChinaGreen-H₂/ammonia standards + provincial mandates — but the real lever is the electrolyzer cost lead: Chinese alkaline systems run ~¼ the Western price (~$185–200/kW in 2024).[14]Green capex, globally
The rule that re-scores everything In 2025 the "One Big Beautiful Bill Act" shortened the 45V window — facilities must now begin construction before Jan 1, 2028 to qualify.[5] One legislative line just re-priced every hydrogen project in America. And the pen moves in every capital: in May 2026 Ottawa redrew its industrial carbon-price curve from $170/t-by-2030 to ~$115/t, re-pricing every Canadian capture project by one announcement.[17] That is the episode's thesis in a sentence: the subsidy map moves, and the rankings move with it.

05The flip: what 45V does to green

Green loses on raw cost — and wins on net cost, because of one credit

This is the chart the slogan never shows you. On raw cost, honest green (~$4.5/kg) is 2–3× the grey incumbent. Apply the 45V credit it earns by clearing the carbon line, and its net cost collapses to ~$1.50/kg — into the grey band. Nothing about the molecule changed.

Green hydrogen cost — before vs after the $3/kg 45V credit
grey incumbent $1–2/kg Green, before 45V Green, NET after 45V $4.50 $1.50 − $3.00/kg 45V credit (earned at ≤0.45 kg CO₂e/kg) $0$2$4$6

The credit acts as a ~$3/kg vertical shift. Blue makes a smaller move (it must first clear the 4 kg gate at all, then earns a far smaller tier); grey and brown earn nothing, because they never clear the gate. Run your own numbers in the companion model. [3][4][8][9]

The companion workbook (Bankable-RB02-Hydrogen-Colors-TEA.xlsx) makes this editable: change any color's cost or carbon intensity and the 45V credit, the net-cost flip, and the carbon-price breakeven all recompute live. It also answers the carbon-price question directly — with no subsidy at all, green beats grey on total cost once carbon is priced at roughly $300/t; for context the EU ETS has traded ~€60–90/t and 45Q already pays $85/t for stored CO₂.[6]

06The ranking & per-color verdicts

Two real bets, two to watch, one wildcard, two to retire

Green — Real Bet, in its niches and with subsidy + cheap power. It is the headline for a reason: a clear carbon advantage, a credit built for it, and a global capex curve bending down (China). The honest caveat is the "if the power is clean" pillar and the power-price sensitivity. Blue — Real Bet only if it's honest about capture rate and methane leakage; a fossil lifeline otherwise, and the one most likely to be oversold. (We red-team it in a later episode.)

Turquoise — Watch: real if the carbon-black market holds, with a genuine commercial-scale plant already running. Pink — Watch: rides the nuclear build-out; its capacity-factor edge is real but it's hostage to new nuclear. White — Wildcard: enormous if the geology and extraction economics prove out; unproven at scale today, so a watch-item, not a bet.

Grey & brown — the incumbents to beat, overhyped only when sold as "clean." Grey is the honest, cheap, dirty baseline every clean color is measured against; brown is the one to retire first.

The through-line: the color wheel is marketing; the carbon line and the credit schedule are the business. Which color we tear down first is up to you — the poll on the episode orders the deep-dives.


07Sources

18 sources — agency / government / peer-reviewed lead. Original-language titles preserved. Confidence flags carried through from the research dossier.

1
IEA (2024). Global Hydrogen Review 2024 — Executive Summary & Hydrogen Demand.agency
97 Mt H₂ demand in 2023, "almost completely covered by hydrogen produced from unabated fossil fuels." iea.org
2
IEA (2024). Global Hydrogen Review 2024 — Hydrogen Production.agency
Green ~$4–12/kg today → $2–9/kg by 2030; LCOH-by-source. iea.org
3
US DOE. Clean Hydrogen Production Tax Credit (45V) Resources.gov
Four-tier credit scaled to lifecycle CI; up to $3/kg; ≤0.45 kg for full credit. energy.gov
4
US Treasury (Jan 2025). Final Rules for the Clean Hydrogen Production Tax Credit (press release jy2768) + Energy Innovation explainer.gov
Three pillars (additionality, hourly temporal matching by 2030, deliverability); annual-average emissions < 4 kg CO₂e/kg. treasury.gov
5
"One Big Beautiful Bill Act" (2025) — 45V window shortened; Kirkland & Ellis alert (+ Sidley, Fuel Cells Works).legal
Facilities must begin construction before Jan 1, 2028 to qualify. kirkland.com
6
US 45Q — Congressional Research Service IF11455 + EIA Today in Energy.gov
$85/t saline sequestration; $180/t direct air capture (2024–26 vintages). congress.gov
7
Grey/SMR carbon intensity — RSC Green Chemistry (d3gc02410e) & RSC Advances (d5ra09599a) LCA; ScienceDirect S0360319925004835; IEEFA Reality Check (2022).peer-review brown CI = MED
9–12 kg CO₂/kg (LNG route to ~13.9); fossil routes the highest-emitting. pubs.rsc.org
8
Howarth & Jacobson (2021), Energy Science & Engineering; IEEFA, Blue Hydrogen: Not Clean, Not Low-Carbon, Not a Solution; ScienceDirect S030626192500618X.peer-review
Capture "well over 90%" needed (real-world lower); methane leakage 1–4% (~3.5% midpoint); blue's footprint can exceed burning gas. ieefa.org
9
Production cost bands — ScienceDirect S0360319925016234 (TEA) + ECOSENSE + GEP.TEA synthesis
Grey $1–2/kg; blue $2–3(–3.5)/kg; green $3–6/kg (cross-checks IEA, S2). sciencedirect.com
10
Turquoise / Monolith — Olive Creek 1, Lincoln NE: IDTechEx + DigitalRefining + Chemical Engineering.industry
Commercial since 2020; ~2,500 t/yr H₂ + ~14,000 t/yr carbon black. idtechex.com
11
Turquoise economics — Thunder Said Energy + RSC Energy & Environmental Science (d4ee06191h).TEA + peer
~$1/kg base case, carbon-black-price-dependent; no CO₂ separation/storage step. thundersaidenergy.com
12
White/geologic H₂ — Nature Scientific Reports s41598-023-38977-y (2023, Bourakébougou, Mali); CNN explainer.peer-review cost = LOW
Spontaneously recharging reservoir, village power, 1987 well fire; Lorraine (France) 2023. At-scale economics unproven. nature.com
13
European Commission — EU Hydrogen Bank (Innovation Fund IF24 auction) + EU Hydrogen Observatory + PtX Hub.EU
Fixed €/kg RFNBO premium over 10 yr; first auction cleared €0.48/kg (ceiling €4.5); IF24 bids €0.33–1.88/kg; CBAM lists hydrogen. observatory.clean-hydrogen.europa.eu
14
China electrolyzer cost lead — S&P Global Commodity Insights + H2TECH + pv-magazine.industry
Chinese alkaline ≈ ¼ the Western price; ~$185–200/kW (2024). spglobal.com
15
Green LCOH electricity share — ScienceDirect S0306261925012450 + US DOE PEM clean-hydrogen-production-cost record.peer-review + gov
Electricity = 55–70% (typically >60%) of green LCOH. sciencedirect.com
16
Canada Clean Hydrogen Investment Tax Credit — PBO LEG-2324-021-S + CRA Clean Hydrogen ITC pages + BLG clean-economy ITC update (Apr 2026) + IEA policy 17330.gov + PBO
Refundable 15/25/40% by CI tier (<0.75 / 0.75–2 / 2–4 kg CO₂e/kg); no credit ≥4 kg; methane pyrolysis added Dec 16, 2024; enacted s.127.44–49 ITA. pbo-dpb.ca
17
Canada industrial carbon pricing — Canada.ca federal benchmark (updated) + ICAP, "Canada publishes new carbon price trajectory" (May 27, 2026).gov + ICAP
2021 benchmark: $65/t (2023) +$15/yr → $170/t by 2030; revised effective May 15, 2026: $95 held in 2026, ~$115 by 2030, $130 by 2035. icapcarbonaction.com
18
Canada CCUS Investment Tax Credit — IEA policy 13346 + PBO LEG-2324-022-S.IEA + PBO
Refundable 50% on capture equipment (60% DAC), dedicated geological or in-concrete storage. iea.org

Want the editable model and the deep-dives?

This report is the free tier. The numbers here are live in the companion workbook, and each color gets its own teardown next.

Free · this report + the video
The full ranking, the cost-vs-carbon map, and the 45V money map — sourced and public.

Each color gets its own full teardown — whichever wins the episode poll goes first. Cast your vote on the episode. · bankable.show

08Method & disclaimer

Every quantitative claim is traced to a source above. Point values used in the matrix, the scatter, and the companion model are illustrative midpoints of the sourced ranges (e.g. grey $1–2/kg → $1.50), chosen to demonstrate the 45V mechanics — not project-grade cost estimates. Confidence flags are carried through: brown's carbon intensity is medium-confidence; white hydrogen's cost is low-confidence and therefore never shipped as a $/kg; pink is kept qualitative. The 45V credit schedule and the three pillars are as of the January 2025 Treasury final rules, with the 2025 OBBBA construction-deadline change noted. The companion workbook is illustrative and pending qualified-reviewer (P.Eng) sign-off before any external use. Numbers sourced 2026-06-17; the Canada policy layer (Clean Hydrogen ITC, CCUS ITC, the May-2026 carbon-price revision) added 2026-07-02.