# SGS and Carbonix Want to Reinvent Ironmaking. The Pilot Plant Will Tell Us if They Can.
Here’s what this really means: a testing, inspection and certification company and an Indigenous-owned startup are going to find out whether a new direct reduction technology actually works when you run it for extended campaigns — not just in a lab slide deck. SGS and Carbonix signed a definitive agreement in June 2026 to pilot Carbonix’s P180 technology at SGS’s Metallurgical Centre of Excellence in Lakefield, Ontario, and that’s the entire story worth caring about. Everything else is framing.
Let me be blunt about why the framing matters anyway. This isn’t the first time we’ve seen this circus — a promising process concept, a credible testing partner, and a press release full of sustainability language — but the structure here is actually the right one, and that’s rarer than it should be. Carbonix owns the IP. SGS runs the independent validation. That separation matters, because the people who eventually have to finance, build and operate a plant don’t take a founder’s word for anything. They take data from a third party with eight decades of metallurgical piloting behind it.
SGS Lakefield is the real asset in this announcement. Since 1941, that facility has completed more than 22,000 projects, taking processes from early testing through piloting and scale-up into commercial production. That track record doesn’t guarantee P180 works. But it does mean that if the technology stumbles, we’ll know about it — and if it succeeds, the data will be defensible in front of lenders and engineering firms. Mackie, SGS’s VP of Metallurgy and Consulting for North America, put it plainly enough: independent validation is how a new process earns confidence. He’s right. I’ve watched plenty of technologies die not because the chemistry failed but because nobody trusted the numbers behind it.
Now, what P180 actually claims to do — because the ambition here is significant. The process is designed to turn iron ore concentrate into hot briquetted iron, the high-purity feed that electric arc furnaces blend with scrap. It combines three steps that conventionally run at separate facilities — upgrading ore to direct-reduction grade, agglomerating it, and reducing it to metallic iron — into a single integrated flow. Integration sounds like a footnote. It isn’t. Every integration point is a place where processes that work independently can interfere with each other, and where scale-up risk multiplies.
The hydrogen angle deserves a skeptical look. P180 makes its own hydrogen on site via methane pyrolysis — splitting pipeline natural gas into hydrogen and solid carbon rather than carbon dioxide. The hydrogen reduces the ore, and the design keeps part of that carbon in the iron, where steelmakers can use it. On paper, that’s a clever route: it sidesteps the green hydrogen cost problem entirely, since blue and green hydrogen remain painfully expensive and structurally scarce for bulk metallurgical use. Methane pyrolysis has been the “next big thing” in clean hydrogen circles for a decade, and the physics is sound. The engineering — running pyrolysis reactors continuously at industrial throughput, handling solid carbon byproduct, keeping the whole loop stable — is where most attempts have hit the wall. Integrated, sustained operation of P180 has not yet been demonstrated. That sentence, buried near the bottom of the announcement, is the most important one in it. The entire pilot program exists to fix that.
The commercial logic is where Carbonix makes its strongest case, and it doesn’t depend on carbon markets. Ontario added high-purity iron to its critical minerals list in March 2026, citing the province’s shift to electric arc furnace steelmaking. Those furnaces need clean, high-iron feed alongside scrap, and Ontario produces no HBI today. That’s a real supply chain gap, not a manufactured one. Blast furnace capacity is retiring, EAF share is growing, and the domestic supply of DR-grade pellets and HBI is thin on both sides of the border. Whoever solves the clean iron feed problem for North American EAFs owns a durable niche. Paul Pede, Carbonix’s CEO, frames it as targeting that Ontario gap, and the logic holds — provided the technology clears the pilot.
The program structure is what you’d want to see. SGS will upgrade and agglomerate the ore feed, run the reactor campaigns, perform assays and characterisation, and report each milestone against defined acceptance criteria. It concludes with a Pre-FEED study and a scale-up basis of design for a demonstration plant. Translation: by the end of this, there will be a bankable technical document — or a clear statement that the technology isn’t ready. Either outcome has value. Most press releases in this space never get that far; they stall between the lab and the demonstration plant, where capital requirements jump an order of magnitude and good ideas go to die.
A few practical notes for anyone tracking this space. First, pilot success is necessary but nowhere near sufficient. Moving from an extended campaign at Lakefield to a demonstration plant to commercial operation typically takes five to eight years and hundreds of millions in capital, and each stage kills a meaningful share of contenders. Second, watch the acceptance criteria, not the headlines. The interesting questions are metallization rate, carbon content consistency in the HBI, reactor uptime across campaigns, and how the pyrolysis loop handles continuous duty — not the emissions claims. Third, the economics hinge on natural gas prices relative to hydrogen alternatives; P180’s advantage over green hydrogen DR routes is largely a cost story, and gas markets are unforgiving.
There’s also the strategic angle worth noting. SGS is folding this into its critical minerals processing expansion and its sustainability strategy, which is fine — but don’t confuse corporate strategy documents with technical validation. The only thing that matters is what comes out of those reactor campaigns. If P180 hits its targets at Lakefield, Ontario gets a credible path to domestic HBI production and the North American steel supply chain gets a new option. If it doesn’t, this becomes another entry in the long history of ironmaking concepts that worked at pilot scale and died at commercial scale.
The steel industry has heard grand promises about clean ironmaking before. The difference here is that the validation is being run by people whose reputation depends on reporting the failures honestly. That’s the right setup. Now the reactors have to run.
