I've spent 40 years in fuel system engineering. In the 1980s and 90s, underground storage tank regulations required double-wall tanks to contain leaks. Engineers designed to that standard. Inspectors signed off. Everybody went home satisfied.
Ninety percent of the leaks were in the piping. The code didn't cover pipe.
The major oil companies knew better. They specified double-wall fiberglass pipe. It worked perfectly. It was slow to install, itchy, difficult. Nobody loved it. Then a company called Total Containment came along with a flexible double-wall pipe. Easier to handle, faster to install, everybody wanted it. I warned that it was susceptible to microbial degradation. I was called Chicken Little. TC pipe went into thousands of sites across the country. It failed. The company went out of business. The contamination was already in the ground.
The majors had the right answer the whole time. The industry chased the easier one.
I tell this story because I'm watching the same sequence play out at Oak Orchard.
Brown and Caldwell — a firm I respect — will design the industrial treatment plant to DEC's requirements. That's what engineers do. They design to the code they're given. The problem isn't Brown and Caldwell. The problem is the code.
DEC's current standard requires monitoring for approximately 40 PFAS compounds using EPA Method 1633. Brown and Caldwell will design a biological treatment system — a membrane bioreactor — that the county's own permit filing acknowledges cannot effectively remove PFAS. The treatment plan rests on Micron's unverified claim, made at design workshops, that PFAS will be present in "low concentrations." That's not a permit limit. That's a promise from the company that won't tell anyone what chemicals it uses.
From the county's own engineering report, Page 1-9: "Micron stated during the workshops that PFAS and mercury are anticipated in low concentrations in the waste load coming from their facilities. These compounds will be limited in the discharge from Micron to a level that can be removed in the biological treatment to values below the discharge requirements." Not a measured value. Not a laboratory result. What Micron said at a meeting.
Here's what concerns me about the biological treatment approach beyond the obvious: PFAS are surfactants. They concentrate at air-water interfaces. The aeration bubblers in a biological treatment system generate aerosols. Some apparent "reduction" of PFAS in biological systems may not be treatment at all — it may be phase transfer. Liquid to air. Discharged into the atmosphere over Clay, New York instead of the Oneida River. For long-chain PFAS this is a documented pathway.
For the short-chain compounds that modern semiconductor fabs actually use, the more likely pathway is concentration in the sludge — 14,083 wet tons of it per year from FAB1 alone, with no PFAS testing required before land application.
The PFAS won't be destroyed. They'll go somewhere — air, sludge, or river. Probably all three.
The code doesn't require destruction. B&C will design to the code. I've seen this movie before. The difference is we're standing at the moment before the pipe gets specified. After 60% design completion, it's a change order. Nobody wants a change order on a $2.6 billion plant.
The fiberglass worked. It was harder to install. That's always how it goes.
This isn't a story about faraway communities or statistical risk assessments. The people who will live with the consequences of this design decision fish in these rivers, boat on these lakes, and camp on these shores. Senior county officials have told me directly that their families recreate on the waters that receive the Oneida River's flow. They are stakeholders in the most literal sense — not as administrators managing a permit, but as people whose children swim in this watershed.
That matters because it means the people making this decision understand what's at stake better than anyone. They don't need to be convinced this water is worth protecting. They need a code that actually protects it. Right now they don't have one. Brown and Caldwell will design to what DEC requires. DEC requires monitoring, not destruction. Monitoring without destruction is documentation of a problem, not a solution to one.
The ask isn't to stop the plant. The ask is to require that the plant be designed to destroy PFAS — not just detect them, not just concentrate them in sludge, not just aerosolize them over a residential community — before the design locks in at 60% completion.
After that, it's a change order. And nobody wants a change order on a $2.6 billion plant.