Changing foam concentrate is a system change, not a procurement decision

Foam changeover is being run by procurement in most organisations that have started it. That is the wrong department, because the concentrate is the one component of a foam system that every other component was tested against.

The date everyone is watching is the wrong one

The marine sector has now passed its threshold: the prohibition on the use of firefighting foams containing PFOS on board ships took effect on 1 January 2026. Behind it, the regulatory machinery keeps moving — the US Coast Guard spent this summer consolidating type-approval criteria for marine foam fire-extinguishing systems into the Code of Federal Regulations, with the comment period closing on 27 July 2026.

Land-side process facilities in the Gulf have no single equivalent cut-off. That absence is being read, wrongly, as time. It is not time; it is only the absence of a date. The direction of travel on fluorinated foam is settled everywhere the question has been asked, and an operator whose jetty receives vessels that have already converted will be asked why the shore system has not.

But the compliance date was never the right engineering trigger anyway. The trigger is the approval basis of the system you already own.

A foam system is a matched set

A fixed foam system is not a tank of concentrate attached to some pipe. It is a matched set — concentrate, proportioner, pipework, discharge device — and, critically, a fire test that qualified that particular combination. Change one member of the set and every other member is holding an assumption that no longer describes reality.

The specific places where the assumption breaks:

Proportioning. Fluorine-free concentrates commonly differ in viscosity from the fluorinated agents they replace. Proportioners sized and calibrated for a thinner concentrate can fall outside their accurate range on a thicker one, and the failure is quiet — the system runs, the foam is simply lean.
Discharge devices. Expansion ratio and operating range have to match the agent. A device that produced a stable blanket with one concentrate is not guaranteed to produce one with another.
Hydraulics. Water supply, tank capacity, pump performance, pipe diameters and valve sizes may all require adjustment once the design application changes.
Application rate and duration. Extinguishing and burnback behaviour differ between agent families. This is the parameter most often carried across unexamined, because it lives in a document nobody reopens.
Residue. Tanks, pipework, pumps and valves must be fully emptied and decontaminated, or replaced. Trace quantities of legacy foam left in a system can be enough for the replacement agent to fail the compliance test it was bought to pass.

Any one of these turns a substitution into a re-design. In practice you rarely get only one.

“Fluorine-free” is not a specification

The most expensive misunderstanding in this whole subject is treating F3 as a category you can procure against. It is not. Approvals and listings attach to tested combinations, not to chemistry families. Two fluorine-free concentrates from two manufacturers are not interchangeable with each other, and neither of them is automatically interchangeable inside a system approved with something else.

Which means a written statement that a facility “has moved to fluorine-free foam” tells a reviewer almost nothing. The questions that carry information are narrower: which agent, verified against which device, at what application rate, evidenced by which test. A changeover that cannot answer those has replaced a documented system with an undocumented one — and has done it while believing it improved its regulatory position.

The schedule decides this, not the chemistry

Decontamination is where these projects actually fail, and they fail on the calendar rather than on the engineering. Legacy stocks need approved disposal routes, and specialist capacity for that disposal is limited and priced accordingly. Cleaning a system properly is not a task that fits into whatever gap remains at the end of a turnaround.

That has an unwelcome implication for planning. The decision to change agent has to be taken far enough ahead of the outage that the disposal route is contracted and the re-verification scope is defined before the plant comes down — not discovered during it.

The pressure is sharper in the Gulf than elsewhere, because the region is specifying an unusual amount of new capacity at exactly this moment. ADNOC Gas confirmed in August a further tranche of awards under its Rich Gas Development programme, covering new gas processing and NGL fractionation capacity. Facilities being specified in 2026 will be in service well beyond any plausible endpoint for fluorinated foam.

So the new-build question and the retrofit question are the same question arriving from opposite directions. On new build, specifying an agent on a phase-down trajectory is buying a retrofit and paying for it twice. On existing assets, the retrofit competes for turnaround windows that are already allocated years out. Both are decided now, or decided badly later.

What this means in practice

Four questions establish where a facility actually stands:

Which agent is our foam system approved with, and does the concentrate currently in the tank match that approval?
If we change agent, what is the re-verification scope — proportioning accuracy, discharge device expansion, hydraulic capacity, application rate — and who signs it off?
What is the decontamination and disposal plan, does the disposal route have capacity in our window, and is it contracted?
On anything currently being specified, is the chosen agent one we expect to remain permitted for the design life of the asset?

A foam changeover that is managed as a procurement action produces a compliant invoice and an unverified system. The two are easy to confuse right up until the day they are not.

Mainstream Petrochemical Instruments FZE provides fire safety engineering for high-hazard industrial facilities. This article is general technical commentary and is not a design recommendation for any specific facility.