Waste-to-Hydrogen Pencils at $5.30/kg. The Co-Product Decides Whether It Pencils at All.

NREL put the minimum selling price of hydrogen from mixed plastic waste gasification at $3.41/kg in a 2023 techno-economic study. Pyrolysis routes that bolt catalytic steam reforming onto the back end land higher still — published analyses put them at $5.30 to $5.50/kg, depending on how hard you push methane conversion. Methane pyrolysis, the turquoise route running on pipeline gas, clears around $1.80/kg in a molten-salt reactor. So if you came to waste to hydrogen expecting a cost advantage, there isn't one. You start two to three dollars a kilo behind the cheapest low-carbon competitor, and the spreadsheet has to make that gap up somewhere it usually doesn't admit.

I structure capital for waste-to-energy and alternative-fuel projects, and waste-to-hydrogen pro formas have started landing on my desk at roughly one a quarter. They share a tell. The hydrogen revenue line is doing far too much work. Strip it back and most of these aren't hydrogen projects at all — they're carbon or pyrolysis-oil projects with a hydrogen tail, and whether they finance depends on a co-product the deck barely mentions.

The price stack, and why the H2 line loses

Pyrolysis hydrogen is a two-stage problem priced as one. Stage one cracks the polymer in an oxygen-free reactor at 400–700°C; stage two reforms the resulting gas and oil vapours at 700–900°C and runs a water-gas shift to push the hydrogen fraction up, per the 2025 Royal Society of Chemistry review of hydrogen generation from waste plastics. Both stages are endothermic. Both want clean, dry, well-sorted feedstock. Neither is free, and the energy bill is why the number comes in where it does.

Read that table as a financier would. The waste route sits at the bottom. Turquoise hydrogen needs only about 37.4 kJ per mole of H2 at standard conditions because it splits methane into hydrogen and solid carbon and nothing else, according to the Chemical Engineering overview of methane pyrolysis. Waste plastic is not methane. It's a chlorinated, pigmented, multi-polymer mess that has to be shredded, dried, sorted, and held at temperature, and every one of those steps is opex the turquoise plant never sees. The hydrogen yield is real — flash Joule heating of HDPE reaches roughly 47 moles of H2 per kilogram in the RSC review, and catalytic pyrolysis hits 50-plus mmol per gram — but yield was never the binding constraint. Cost was. The waste conversion technology works. The unit economics, on the hydrogen line alone, do not.

That's the part the decks gloss. They benchmark waste-to-hydrogen against green hydrogen from electrolysis at $5–6/kg and declare victory. But green hydrogen is the wrong comparator for a project that has to clear a final investment decision in 2027. The comparator is whatever the offtaker can buy delivered, and increasingly that's blue or turquoise at half the price.

What actually carries the project

Here's where the framing has to change. A plastic pyrolysis plant doesn't produce hydrogen; it produces a slate. Pyrolysis oil — a naphtha-substitute liquid — is usually the largest revenue stream by mass. Solid carbon or char is the second. Hydrogen and light gas are a slice, and in a lot of configurations that gas is burned on-site to heat the reactor rather than sold at all.

The economics of the turquoise route make this explicit. The Chemical Engineering analysis notes that solid carbon priced at $600–900/ton fully offsets the cost of the hydrogen, and even $200/ton carbon gets methane pyrolysis to parity with SMR once offset credits are counted. The hydrogen, in other words, is close to free if the carbon sells. Waste pyrolysis has the same structure with messier products: the project pencils when the oil and char find buyers, and the hydrogen is the thing you book once the slate has already paid for the plant.

And there's a line below the product slate that matters more than any of them. The waste tip fee is the only line that's actually contractual. A plastic pyrolysis plant gets paid a gate fee — often $40 to $90 a tonne in mature markets — to take feedstock that would otherwise go to landfill or export. That fee is contracted with a municipality or a waste hauler, it's senior to commodity price risk, and it's the cash flow a lender will actually size debt against. The hydrogen offtake, the oil offtake, the carbon offtake — those are all merchant or quasi-merchant until proven otherwise. If you want to understand how the pyrolysis gas itself gets cleaned into something sellable before any of this revenue exists, the mechanics of turning pyrolysis gas into a usable syngas stream are a useful companion read.

So the underwriting order inverts. Tip fee first, because it's contractual. Oil and carbon second, because they're the bulk of merchant revenue. Hydrogen last. A waste-to-hydrogen project that can't survive that ordering isn't a waste-to-hydrogen project. It's a hydrogen bet wearing a circular-economy costume, and I've watched two of those fall apart in diligence in the last eighteen months.

The offtake question nobody underwrites early enough

Assume the cost works. You still have to sell the hydrogen, and pyrolysis-derived hydrogen has a quality problem that the pro forma rarely prices.

Raw pyrolysis gas carries carbon monoxide, carbon dioxide, sulphur compounds, tar, and — if any PVC slipped into the feed — halogenated compounds. Fuel-cell and mobility buyers need hydrogen at ISO 14687 grade: a 99.97% minimum fuel index with hard species limits on exactly those contaminants. Getting there means pressure-swing adsorption and polishing trains that cost capital and lose 10–15% of the hydrogen as tail gas. If your offtaker is a refinery hydrotreater or an ammonia plant, the spec is looser and the cleanup cheaper. If it's a hydrogen mobility network, it's not. The pro forma I most distrust is the one that assumes mobility-grade pricing on industrial-grade cleanup capex.

Equity stays in the room as long as the offtake stays in writing. A hydrogen project with a merchant tail and a 20-year debt tenor is asking lenders to underwrite a commodity that has no liquid forward curve.

This is the structural weakness. Hydrogen has no deep traded market, no FOB/CIF spread you can hedge, no settled index. A waste-to-energy plant selling power at least has a grid and, often, a PPA. A waste-to-hydrogen plant selling merchant H2 has a local pipeline customer or it has a stranded asset. On a 2024 PPA renegotiation I worked — a 150 MW waste-to-energy facility — the counterparty reopened the contract mid-term to demand biogenic-share recertification. That was a power deal with a real market underneath it, and it still got reopened. Now picture the same counterparty leverage on a hydrogen offtake with no market underneath it at all.

I'll own a related mistake, because it's the same failure mode. In 2021 I underwrote a deal assuming the offtake's credit support would hold through a sovereign downgrade. It didn't. The whole structure had to be re-tranched in Q3 2022 at a 240 bps spread I hadn't modelled. The lesson transfers directly: offtake credit quality is not a footnote, and for waste-to-hydrogen — where the offtake market barely exists — it's the first thing I now stress, not the last.

Where it doesn't pencil

Not every waste-to-hydrogen claim deserves the same scepticism, so here's where the case is genuinely weak and where it isn't.

It doesn't pencil below scale. Pyrolysis and reforming both carry fixed thermal and gas-cleanup loads, and a plant taking under roughly 30,000 tonnes a year of plastic spreads those costs over too little output. It doesn't pencil on contaminated feedstock — PVC introduces chlorine that attacks reformer catalysts and corrodes downstream metal, and mixed municipal solid waste with high moisture forces a drying bill that the turquoise competitor, again, never pays. It doesn't pencil where there's no industrial hydrogen customer within pipeline distance, because trucking compressed hydrogen erases the margin fast.

Where it does have a case: co-located at a refinery or chemical complex that already consumes hydrogen, where the gate fee is contracted, where the pyrolysis oil has a committed buyer, and where the hydrogen displaces grey hydrogen the site was buying anyway. In that configuration the project is really a waste-conversion plant that happens to short-circuit a hydrogen purchase, and the renewable energy from waste it generates is a genuine emissions reduction rather than a marketing line. That's a financeable structure. It's also a narrow one.

The carbon-credit angle deserves its own warning. Most of the waste-to-hydrogen decks I see lean on avoided-emissions or carbon-removal revenue to close the gap, and that revenue is routinely overstated by 30–50% at the point a registry actually reviews the methodology. I ran a carbon-credit issuance audit for an Asian waste facility in 2023, and a methodology mismatch killed 38% of the expected credits at registry review — credits the financial model had already spent. A waste-to-hydrogen pro forma that needs carbon revenue to clear its hurdle rate should be stress-tested at half the assumed credit volume before anyone signs anything.

What I'd actually underwrite

Can pyrolysis deliver clean hydrogen from waste? Technically, yes — the yields are real and the gas can be polished to ISO 14687. That was never the question. The question is whether it delivers hydrogen anyone can finance, and the answer is: only when the hydrogen is the last thing you're counting on.

Underwrite the gate fee, because it's contractual. Underwrite the oil and the carbon, because they're the bulk of the revenue and they decide whether the plant survives a soft hydrogen year. Treat the hydrogen as a margin enhancer for a buyer who already needs it on-site, not as the product that justifies the build. A project built the other way round — hydrogen first, slate second, tip fee as an afterthought — is the one where the spreadsheet turns out wrong by year three. The chemistry is fine. It's the capital structure that fails, and it fails for reasons that have nothing to do with the reactor.

Disclosure: the author advises on waste-to-energy capital structuring and alternative-fuel trade. This analysis reflects diligence patterns observed across financed and declined projects, not a specific live transaction. For RWE's view of the underlying conversion hardware, see its overview of waste-to-energy pyrolysis systems; for the broader frame, Renewable Waste Energy's circular economy work sits alongside this column.

Sources & Notes

1. Minimum selling price of $3.41/kg for hydrogen from mixed plastic waste gasification: NREL / Green Chemistry techno-economic study, 2023 — docs.nrel.gov/docs/fy23osti/83095.pdf.
2. Pyrolysis process temperatures (400–700°C cracking, 700–900°C reforming), hydrogen yields (~47 mol H2/kg HDPE, 50+ mmol/g catalytic), and gasification-versus-pyrolysis trade-offs: "State-of-the-art and perspectives of hydrogen generation from waste plastics," Royal Society of Chemistry, Chemical Society Reviews, 2025 — pubs.rsc.org.
3. Turquoise hydrogen cost range ($1.80–4.00/kg), reaction energy (37.4 kJ/mol H2), and solid-carbon offset values ($200–900/ton): "Hydrogen Production via Methane Pyrolysis: An Overview of Turquoise H2," Chemical Engineering, 2023 — chemengonline.com.
4. Hydrogen fuel quality specification (99.97% minimum fuel index, species impurity limits): ISO 14687:2019, Hydrogen fuel quality — Product specification — iso.org/standard/69539.html.
5. Plastic pyrolysis plus steam reforming cost range ($5.30–5.50/kg) and three-stage process configuration: techno-economic literature on staged pyrolysis / catalytic reforming / water-gas shift of waste plastic — PMC9986875. Gate fee range and registry-methodology haircut: RWE project experience, 2022–2024.