Clean Methanol: Which Route Actually Pencils Out?

The bill before the plant

For one tonne of e-methanol, suppose your hydrogen costs three dollars a kilogram and your carbon dioxide costs one hundred dollars a tonne. With rounded input quantities, those two items alone come to seven hundred and eight dollars. You haven't paid for the methanol plant yet.

Those prices are assumptions. I'm not quoting today's market. The point is to see what the project has to pay for before anyone adds a green premium to the sales forecast.

My answer up front: methanol is a credible product, and lower-carbon production is a real engineering proposition. The investment case depends on a much narrower question. Can your particular supply chain deliver the right product, with the right carbon evidence, at a cost an actual buyer will cover?

We're going to compare the main routes, then work through the input bill that a project presentation can make surprisingly hard to find. I've built a small input-cost screen alongside this episode. It separates purchased hydrogen from onsite electrolysis and leaves the missing costs visible.

Because a familiar molecule can hide a very unfamiliar business.

Start with a real delivery

Take a plant in southern Denmark. European Energy reported first e-methanol production and customer deliveries in twenty twenty-five. Its stated annual capacity is forty-two thousand tonnes. In an August twenty twenty-six update, co-owner Mitsui reported supplies into both shipping and chemical applications.

That is useful evidence. There is a product and there are reported deliveries to customers. It moves the discussion beyond an artist's impression of a future factory.

But capacity, production and profit answer different questions. Forty-two thousand tonnes of annual capacity doesn't tell us how many saleable tonnes actually left the gate, what they cost, or what price the customer paid. Those disclosures don't independently establish sustained utilization or a project margin.

So I wouldn't dismiss the technology because a complete public cost sheet isn't available. I also wouldn't fill that missing sheet with optimistic guesses.

The useful next question is what changed upstream to make this methanol different from the incumbent product.

The molecule stays the same

Conventional methanol can start with natural gas or coal. The process makes synthesis gas, adjusts and cleans it, and uses a catalytic synthesis step to produce methanol. Biomass can also provide the carbon.

For our comparison, keep five lanes on the page: the fossil benchmark; biomass or waste through syngas; biogas or biomethane through reforming; carbon dioxide plus electrolytic hydrogen; and fossil-based capture or hybrid proposals making a lower-carbon claim. Some actual projects combine lanes.

These aren't five universal certification categories. They're a way to expose where the carbon and hydrogen come from, which equipment and contracts you need, and which risks remain.

If the product meets the same specification, you haven't invented a different methanol molecule. You've changed the supply chain behind it. That can change its cost and lifecycle emissions enormously. The label on the tank doesn't tell you how much.

Start with the route that follows directly from the gasification discussion.

Biomass and waste: follow the accepted output

In the biomass or waste route, the upstream job is to turn a variable material into a synthesis feed the downstream plant can use. The equipment between the feed pile and the methanol loop still matters. So do the rejected material, contaminants, utility needs and residual streams.

A cheap feedstock can be useful. You still need a mass balance and a contract that survive the actual feed specification. If the material changes, which part of the plant absorbs that change, and who pays?

For mixed waste, keep the carbon fractions visible. A process can use carbon efficiently without making every part of that carbon renewable. The eventual lifecycle and qualification claim needs its own evidence.

Then follow the saleable output. A gasifier test isn't a demonstration that the entire site can repeatedly supply methanol within the buyer's specification. Ask for the integrated run: actual feed, accepted output, interruptions, utilities and residual handling over a meaningful operating period.

That distinction also appears in China's official pilot instructions, which separate construction and commissioning targets from later high-load stable production. A commissioning event and a sustained operating record are different evidence.

My screening judgment here is conditional: advance a strong integrated evidence package. Redesign a feed or cleanup arrangement that doesn't fit. Wait where the operating proof is missing. Stop where the claimed advantage disappears when the full system is included.

Biogas, capture and the hybrid problem

Biogas or biomethane can supply another route into methanol production. That changes the upstream feed system; it doesn't remove the need to document gas quality, reliable supply and the attributes being sold with it.

I'd ask what else that gas could be sold for. A feedstock with another valuable use doesn't become free because the methanol model would prefer it that way. The relevant cost is the supply arrangement you can actually secure.

Now consider a fossil-based plant adding carbon capture, lower-carbon hydrogen, or a mixture of inputs. There may be a genuine emissions improvement. To evaluate it, show which streams are captured, what remains, what energy the capture system consumes, and where the captured carbon goes.

Carbon used to make a fuel is different from carbon placed into durable storage. When that fuel is used, its carbon can return to the atmosphere. A recycling step needs a lifecycle comparison; the capture step alone doesn't settle the result.

Hybrids deserve the same discipline. If renewable hydrogen improves carbon utilization in a biomass system, show that actual flow. If fossil and renewable material are blended, show the boundary and the accounting. A neat color on a slide isn't a substitute for the underlying quantities.

And this brings us to the route where the input arithmetic is especially revealing.

E-methanol: open the hydrogen invoice

For the carbon-dioxide route, hydrogen reacts with carbon dioxide to make methanol and water. The rounded quantities in the International Renewable Energy Agency's technical reference are about one hundred and ninety kilograms of hydrogen and one point three eight tonnes of carbon dioxide for each tonne of methanol. These are approximate input quantities, not a guarantee of actual plant yield.

Return to our illustration. At three dollars per kilogram, hydrogen contributes five hundred and seventy dollars. At one hundred dollars per tonne, carbon dioxide contributes another one hundred and thirty-eight. Together, seven hundred and eight dollars per tonne of methanol.

That subtotal excludes the rest of the plant and the supply chain. Conversion losses, compression and other utilities, maintenance, equipment, financing, storage and delivery still have to be accounted for. If your carbon dioxide price excludes capture or purification, those costs still need somewhere to go.

What if you make the hydrogen onsite? Use a separate calculation. Assume fifty kilowatt-hours of electricity per kilogram of hydrogen. For our rounded hydrogen requirement, that's nine point five megawatt-hours per tonne of methanol, just for this hydrogen electricity component.

At assumed power prices of thirty, sixty and ninety dollars per megawatt-hour, that component becomes two hundred and eighty-five, five hundred and seventy, and eight hundred and fifty-five dollars. Keep the same carbon dioxide assumption, and the two-input subtotals are four hundred and twenty-three, seven hundred and eight, and nine hundred and ninety-three dollars.

These are sensitivity cases. They're not competing commercial bids. In the onsite case, the electrolyser itself still costs money. And you mustn't add an all-in purchased-hydrogen price on top of the electricity used to make that same hydrogen. Pick a consistent boundary.

A low headline power price also needs an operating profile. Can that power be delivered when needed? What storage or flexibility is required? How does the plant's annual accepted output change? The unit electricity calculation is a useful first check. It doesn't answer those system questions.

The same caution applies to research targets. A government-supported project can be investigating an attractive methanol cost. A target in a research description is evidence of the objective, rather than evidence that a commercial plant already achieves it.

The carbon history travels with the fuel

The carbon dioxide input deserves its own invoice and its own evidence file. Biogenic carbon, fossil process carbon and carbon captured from air can enter similar downstream equipment while supporting different lifecycle and qualification claims.

The electricity also has a history. So do the feedstock, transportation, processing and final use. You need the full boundary relevant to the claim you're making. A good result at the synthesis reactor doesn't erase emissions elsewhere.

This is why I would ask for the lifecycle calculation and the records behind it, rather than accept green, clean or low-carbon as interchangeable descriptions. And I'd ask whether the buyer accepts that particular method, boundary and supporting evidence.

For example, Canada's current Fuel Life Cycle Assessment Model methodology covers stages from feedstock production through fuel use, and its twenty twenty-six update revised the natural-gas-methanol dataset. That is a modeling framework. It isn't an automatic credit award for a new methanol proposal.

Keep three statements separate: we produced methanol; we demonstrated a particular lifecycle result; and a customer or program will pay for that verified result. Each statement needs evidence of its own.

A market signal still needs a contract

Shipping gives methanol a useful application to examine. The European Commission describes its maritime fuel rules as a technology-neutral lifecycle greenhouse-gas-intensity framework. That creates a different decision from a rule telling every ship to buy the same named fuel.

A fuel producer therefore needs to understand which verified attribute has value for a particular buyer, and how much of that value reaches the producer. A customer's possible compliance saving isn't automatically your methanol premium.

I'd also keep a proposed global rule separate from an effective regional rule. For the separate International Maritime Organization process, approval, adoption and entry into force are different legal steps. Check the adopted text and the effective date before assigning revenue to a global rule. Our cost illustration assumes no income from a proposed mandate.

The dedicated policy episode can work through those documents and eligibility details. Here, the commercial test is simpler: who's buying, what exactly are they buying, how is the price determined, and who carries the risk if the fuel's claimed attributes aren't accepted?

The five evidence gates

Before I would advance a project, I'd want five connected evidence files.

First, inputs: feedstock, power, hydrogen or carbon dioxide, with quality, quantity, timing and commercial terms.

Second, integrated operation: actual accepted methanol output, supporting utilities, interruptions and residual handling. Nameplate capacity belongs beside that record, not in place of it.

Third, carbon: a defensible lifecycle boundary and traceable attributes accepted for the intended sale.

Fourth, delivered economics: the complete cost stack on the buyer's delivery basis. Don't compare a plant-gate input subtotal with someone else's delivered finished-fuel price.

Fifth, the buyer: specification, price mechanism, volume, term and allocation of qualification risk. A customer logo doesn't tell you all of that.

I would mark each gate documented, conditional or missing. I wouldn't give an undisclosed project a precise bankability score just because the chart looks better with numbers in it.

The verdict

So which methanol route wins? The evidence here supports a conditional answer. The product and the production chemistry are credible. Real deliveries strengthen that case. A universal cost winner needs site-specific input, operating, carbon and contract evidence that a pathway label cannot supply.

Biomass and waste need integrated feed-to-product proof. Biogas needs a real supply and attribute case. E-methanol needs a hydrogen-and-carbon-dioxide bill that survives the complete plant economics. Fossil capture and hybrids need measured boundaries and an honest account of what remains. Keep the fossil route as a transparent benchmark.

Then choose among all four actions: advance, redesign, wait for evidence, or stop. The evidence can change the action. It doesn't have to change the molecule.

The two questions I'd take away are these. What does an accepted tonne actually cost at the buyer's delivery point? And what verified attribute makes that buyer willing to pay more for it?

The companion input screen lets you change the assumptions without pretending the omitted costs have vanished. The policy follow-up will examine how the qualification rules affect that premium. Its release date isn't assigned.

This is analysis of technology and project economics, not legal or investment advice. If this way of testing a hard-tech claim is useful, subscribe to Bankable.