Home›Insights›Due-diligence standard

Four questions separate a real supercritical process from a promotional one. This is the test we apply before we will engage with any supercritical or hydrothermal technology, including our own. It is written so that a bioresources manager, a strategy lead or a consulting engineer can apply it without specialist supercritical experience.

Why this exists

Treating sewage sludge and other wet organic wastes above the critical point of water is a real and promising route. It avoids the drying penalty that makes conventional thermal treatment uneconomic on a feed that is 75 to 85 percent water, it operates without combustion air and so without a conventional flue-gas train, and it opens a path to taking contaminated solids out of the land-application route entirely.

It is also a young category with very little reliable public material, a small number of genuine practitioners, and a growing number of promotional packages whose numbers do not close. Utilities and their advisers are being asked to form a view on processes they have no in-house basis to assess, in a market where the marketing has run well ahead of the engineering. The four questions below are what we use to tell the difference.

Question one: carbon cannot be created

The most common failure in this category is a product-gas yield larger than the carbon in the feed can supply. It is easy to miss, because the headline mass balance appears to close.

Here is the arithmetic you can do unaided. Take the dry matter in the feed. Take volatile solids as a fraction of it. Take carbon as roughly half of volatile solids by mass. That is the carbon available. Now take the claimed product gas and multiply by its carbon fraction, which for a methane and hydrogen mixture is high. If carbon out exceeds carbon in, the process as specified does not exist. Water contributes hydrogen and oxygen. It does not contribute carbon.

We have reviewed packages in which carbon out exceeds carbon in by a factor between 1.6 and 2.5, while the total tonnage balance is presented as conserving one hundred percent. The tonnage closes because the stated ranges are wide enough to absorb the discrepancy. The elements do not close at all. That is why the question specifies elemental, and not total.

Question two: a carbon-dioxide-free product gas is not available

Some packages claim a product gas containing no carbon dioxide, usually justified by the absence of combustion. The absence of combustion is real, and it does explain the absence of thermal NOx and of a conventional flue gas. It does not remove the carbon dioxide.

Carbon removal in supercritical water proceeds by reforming and water-gas shift. Carbon plus water gives carbon dioxide and hydrogen; carbon monoxide plus water gives carbon dioxide and hydrogen. The oxygen that forms that carbon dioxide comes from the process water and from the feed organics, which are themselves roughly half oxygen by mass in the carbohydrate fraction. That oxygen has to go somewhere, and at these conditions its only sink is carbon dioxide or carbonate. High carbon-to-gas conversion and zero carbon dioxide are mutually exclusive. Reported supercritical water gasification product gas typically runs 20 to 50 percent carbon dioxide.

A more sophisticated version of the claim concedes that carbon dioxide forms but asserts it stays dissolved in the aqueous phase at pressure and is separated before the gas is discharged. That version deserves testing rather than dismissal, and it fails on two counts. First, capacity. Carbon dioxide solubility in water saturates in the region of 55 to 60 grams per kilogram at 250 bar, which on a typical sludge slurry flow is a small fraction of the carbon dioxide the oxygen balance produces. The rest reports to the gas whatever the flowsheet says. Second, and decisively, Henry’s law. Whatever does dissolve comes back out of solution the moment the pressure is let down. You cannot hold the carbon dioxide in the water and deliver a usable gas at low pressure. These are not competing engineering opinions. They are the same physical constraint read twice.

Two secondary tells travel with this claim. Carbonate trapping requires calcium, magnesium or sodium cations in quantities sludge ash does not supply. And a carbon-dioxide-saturated aqueous phase is acidic, in the region of pH 3 to 4, which sits badly beside any claim that the water leaving the process is potable.

Why the question names the detector. A flame ionisation detector with a methaniser is a trace-level technique. Using it to characterise a stream that should contain percent-level carbon dioxide is choosing an instrument that cannot see the thing in dispute. Ask for a thermal conductivity detector or an equivalent percent-level method, and ask to see the trace rather than a tabulated composition.

Question three: reforming is endothermic, and start-up is not steady state

Supercritical water gasification is net endothermic. Heat is also needed to raise the feed water to supercritical conditions, and the high-pressure pumping duty is real. A well-designed continuous plant recovers much of the sensible heat through a feed-effluent recuperator, and a high recuperator effectiveness is what makes the energy balance work. That is an engineering achievement, not a free lunch, and the number that matters is the recuperator’s fouled performance, not its clean performance.

Two claims should stop a buyer. The first is zero external energy for start-up: a cold start to supercritical conditions requires external energy by definition. The second is a recycle of product gas to supply process heat in a strictly non-oxidative system: heat cannot be extracted from a fuel gas without oxidising it, and oxidising it is combustion, which the same package has usually just disclaimed. Asking for the two balances separately surfaces both immediately.

Question four: provenance and auditability

Ask who owns and who licenses the core intellectual property, in what jurisdiction it sits, who the ultimate beneficial owner is, and where a working installation can be visited and independently audited.

Three answers should end the process. That the technical claims cannot be verified without a non-disclosure agreement, when what is being asked for is a mass balance and a chromatogram rather than a design. That the vendor must be the sole channel to any reference site, so independent verification is not available. And an installed base that turns out, on inspection, to be a different process at different conditions on a different feed, presented as evidence for the process being sold. A high-temperature dry-feed gasifier is not evidence for a supercritical wet-feed reactor, whatever the corporate continuity between them.

Where intellectual property, ownership or reference sites sit in a jurisdiction that creates sanctions, financing, insurance or reputational exposure, that is a commercial disqualifier independent of the engineering, and it should be settled before any technical effort is spent.

Two more worth asking

Calorific value has a basis. Where a gas calorific value looks high for the stated composition, check whether the figure is the whole-gas value or the value of a sub-stream. Quoting the calorific value of the hydrocarbon fraction as though it were the whole gas inflates the number by roughly 15 percent, and the hydrogen and carbon dioxide fractions are exactly what pull the real number down. Ask for higher heating value and Wobbe index on the whole gas as produced, with the composition it was calculated from printed alongside.

Materials are the historical failure mode. The first generation of commercial supercritical water plants was closed by materials problems, not by thermodynamics. Three mechanisms did the damage: salt precipitation out of supercritical water and the plugging that follows, chloride stress-corrosion cracking, and fluoride attack, the last directly relevant where the feed carries fluorinated compounds. Nickel-chromium-molybdenum alloys are the right family, but the family name is not an answer. A sound proposal contains a zoned materials-of-construction schedule, a salt-management strategy with a nameplate salt duty, a fluoride capture strategy, and a position on upstream chemical dosing, because chloride-bearing coagulants feed the worst failure mode in the reactor. What should concern you is a single alloy named for the whole plant, a claimed absence of corrosion, or a design life quoted in decades with no corrosion allowance, materials test data or inspection regime behind it.

The checklist

The last line costs nothing and finds more than any other. Where a package contains several documents, put their yields side by side. A real process gives one answer. Promotional packages routinely give three, and the three disagree with each other and with the carbon available in the feed.

Download the one-page checklist (PDF) The four questions and the table above, on a single page, for taking into a meeting. No sign-up.

What this standard is not

It is not a claim that supercritical treatment does not work. It does, and it is the most promising route available for wet contaminated solids that no other thermal process handles economically.

It is not a substitute for technology due diligence before financial close. It is the screen that decides whether that exercise is worth commissioning.

And it is not a checklist that any early-stage technology passes completely. Genuine developers will have open items, and the honest ones will tell you which. What separates a real process from an unreal one is not the absence of gaps. It is whether the developer can tell you precisely where the gaps are, what will close them, and when.

Applying the standard to our own system

A standard the author will not sit under is not a standard. Redrock’s supercritical fluid gasification module, at its current stage of development, answers as follows.

Elemental balance. Closed on the process design basis, on a stated feed characterisation, with carbon accounted through the gas, aqueous and residue phases. Available on request.

Carbon dioxide. Our product gas contains carbon dioxide, in the range the chemistry requires. We make no zero-carbon-dioxide claim, and no buyer should accept one from anybody.

Energy. The design is continuous and recuperative, and the feed-effluent recuperator is specified on guaranteed fouled effectiveness rather than clean. The process requires external energy at start-up. The steady-state balance depends on recuperator performance, which is why it is a guaranteed duty in our enquiry documents rather than an assumption.

Materials. Zoned materials schedule, vertical salt-managed reactor with a nameplate salt duty, lime-based fluoride capture, and a hard design rule prohibiting ferric coagulant upstream because chloride feeds the reactor’s worst failure mode. Hot-zone alloy selection is deliberately held open pending site-specific chloride and fluoride characterisation. We would rather hold it open than name an alloy we cannot yet justify.

PFAS. This is the claim we are asked about most, and the one we will not overstate. What an installed unit does on day one is end the PFAS land-exposure pathway for that works: the contaminated solids leave the soil route entirely, treated on site, with nothing exported. That is real, immediate and warrantable. Destruction is a different claim. The bankable destruction route is oxidative, and our chosen reductive route carries PFAS destruction as a monitored co-benefit that we are mass-balancing across gas, water and residue on the pilot. Until that data exists we quote no destruction efficiency, and we do not describe the mineral residue or the process water as clean until both have been characterised for residual PFAS and fluoride. Sites that need a destruction warranty today should be routed to an oxidative process, ours or anyone else’s.

Gas yields. The yields in our published material are literature ranges for municipal sewage sludge, labelled as such. They are not site values and they are not warranted performance. Site values come from biomethane potential testing and bench gasification on the actual sludge, and until those exist we will not present a number as ours.

Provenance. The intellectual property is developed and held by Redrock. There is no foreign licensor. Our operating reference does not yet exist, which is the honest answer: the first unit is the pilot, and it is being built in part to prove the salt, chloride and fluoride behaviour that closed the previous generation of plants. We will not claim a reference we do not have, and we would encourage any buyer to apply question four to us as rigorously as to anyone else.

Two of those answers are open items. That is what an early-stage technology looks like when it is described accurately.

Use this standard on us, or on anyone else. The checklist is a one-page PDF, free and ungated, made to be forwarded. If you are assessing a supercritical or hydrothermal proposal and want a second opinion on a mass balance, contact info@redrockbioenergy.com.

This page is a due-diligence aid and a technical overview, not a design, procurement or investment recommendation. Process performance and economics are site and feedstock specific. Confirm every figure against bench and pilot data on your own material before relying on it.