Stainless steel is not inherently corrosion resistant. It is corrosion resistant because of a chromium oxide film a few atoms thick that reforms whenever it is damaged — provided there is chromium available and oxygen to react with. Contaminate the surface with ordinary carbon steel and you have created sites where that film cannot form.
The ferroxyl test finds those sites. It is quick, it is visual, and on nuclear and pharmaceutical work it is very often contractual.
How iron gets onto stainless in the first place
Almost always through contact with carbon steel during fabrication or handling. The usual routes:
- Shared tooling. A grinding disc, wire brush or file previously used on carbon steel embeds particles into the stainless surface. This is the biggest single source and the easiest to prevent.
- Shared workspace. Grinding sparks from carbon steel work landing on stainless nearby. Sparks are hot particles of steel and they weld themselves to whatever they hit.
- Lifting and handling. Carbon steel slings, chains, hooks, and dragging a component across a steel bench or trestle.
- Storage. Stainless stacked directly on carbon steel racking, or resting on the bed of a steel trailer.
- Fasteners and temporary attachments. Carbon steel clamps, tack plates and temporary supports welded on and removed.
None of these look like contamination. A component can pass a visual inspection, look clean and bright, and still carry embedded iron across the whole surface.
Why it matters more than it looks
Embedded iron does two things, and the second is the serious one.
First, it rusts. Free iron on a stainless surface corrodes exactly as carbon steel does, producing visible rust staining that a client will reasonably reject on appearance alone — particularly on architectural or hygienic work.
Second, and more importantly, it initiates pitting. A rusting iron particle sets up a local cell on the surface. The corrosion products are acidic and depleted in oxygen, and beneath them the passive film cannot reform. What starts as a surface stain becomes a pit into the parent material, and pitting is the failure mechanism that puts holes in stainless pipework.
In a chloride-bearing environment this accelerates sharply, and pits are the classic initiation site for stress corrosion cracking. On nuclear systems, where a through-wall leak is a containment issue rather than a maintenance job, the tolerance for surface iron is effectively zero.
How the test works
The ferroxyl test is a chemical spot test that turns free iron blue.
A solution containing potassium ferricyanide and nitric acid is applied to the cleaned surface, usually on filter paper laid over the area or applied directly. Where free iron is present, ferrous ions go into solution and react with the ferricyanide to form ferric ferrocyanide — Prussian blue. The result is unambiguous: a blue spot or streak marks contamination, and the pattern often tells you the mechanism. A row of dots is spark spatter. A broad smear is a grinding disc. A line across the component is a sling.
The nitric acid in the solution is there to break through the passive film enough to let any embedded iron react, without significantly attacking the stainless itself.
The critical point about interpretation: the test detects free iron, not chemistry. It says nothing about the alloy composition of the component. A perfectly good 316L spool contaminated by a grinding disc will show blue; a mis-supplied carbon steel component would show blue everywhere. Distinguishing those requires positive material identification, which is a different test.
Doing it properly
Several things separate a meaningful ferroxyl test from a coloured smear.
Surface preparation. The surface must be clean and free of oil, grease and residues before testing, or the solution cannot make contact. Degreasing is part of the test, not preparation for it.
Solution age. Ferroxyl solution has a limited working life and degrades with light and time. An old solution gives a weak or absent reaction — a false pass, which is the worst outcome a test can produce. Solutions are made up fresh, dated, and stored dark.
Control samples. The right way to prove the solution is working is a known-contaminated control alongside the component. If the control does not go blue, the test is not running, and no result from that batch means anything.
Dwell time. The reaction is not instant. Reading the surface too early misses light contamination.
Post-test removal. The solution and its reaction products must be removed completely after testing. Leaving ferricyanide residue on a stainless surface is itself a contamination problem, and rinsing to a defined standard is part of the procedure.
What happens after a positive result
A blue indication is not a rejection. It is a decontamination instruction.
Light surface contamination usually comes off with mechanical cleaning using stainless-only or non-metallic media, followed by a re-test. More significant or embedded contamination goes to pickling and passivation — an acid treatment that removes the iron and the damaged surface layer, followed by conditions that let the passive film reform fully.
Either way the component is re-tested afterwards, because the whole point is evidence that the surface is now clean, not evidence that somebody cleaned it.
Worth noting that passivity testing and ferroxyl testing answer different questions and are often specified together. Ferroxyl asks "is there foreign iron on this surface". Passivity asks "has the protective film properly reformed". A component can pass one and fail the other.
Where it is specified, and why there
Ferroxyl testing shows up in three environments, and understanding why explains how strictly it will be enforced.
Nuclear. Free iron on stainless in a system carrying or containing radioactive material is a through-wall leak waiting to happen, and a through-wall leak is a containment event rather than a maintenance job. Specifications here tend to be absolute — no detectable free iron — with testing at multiple production stages rather than only at the end.
Pharmaceutical and food processing. The concern is product contamination as much as component life. A pit in a hygienic system is a place bacteria live that cleaning cannot reach, and the surface finish requirements that go with these industries are undermined entirely by embedded iron.
Chloride service. Offshore, marine, and anything handling seawater or chlorinated process fluid. Chlorides attack the passive film directly, so a surface that starts compromised has very little margin. This is where the gap between "it passed visual inspection" and "it will last twenty years" is widest.
Outside those, ferroxyl testing is often specified on architectural stainless for purely cosmetic reasons — rust staining on a visible surface is a rejection whatever the structural argument — and that is a legitimate use of the test even though the consequence is appearance rather than integrity.
Prevention beats testing
Every fabricator who has paid for a pickling and passivation cycle on a finished vessel reaches the same conclusion: segregation is cheaper.
Dedicated stainless-only tooling, colour-coded and stored separately. Physical separation of stainless and carbon steel work areas, or at minimum screening against spark travel. Non-metallic or stainless slings and lifting gear. Plastic or timber packing under stored components. And ferroxyl testing at goods-in and at key production stages rather than only at final inspection, so contamination is found when it costs an hour rather than after the component is finished.
That last point is the practical one. A positive result on a raw plate costs a clean. The same result on an assembled, welded and polished vessel costs a great deal more.
Training and testing
We carry out ferroxyl testing on site and in our Cumbria facility, and — because most contamination is caused by handling rather than found by inspectors — we also deliver ferroxyl awareness training through Responsive Training Services for fabrication and handling teams.
Fabricators who put their own people through it generally stop failing the test, which is a better outcome for everyone than being very good at detecting a problem after it has happened.
If free iron contamination is in your specification and you are not sure whether your controls are adequate, that is worth a conversation before the first component is rejected rather than after.
TECHNICAL REVIEW — DELETE FROM THE LINE ABOVE, DOWN
Scheduled for 2026-09-01. Not to go live until a Level 3 or the RPA has read it.
Check: 251 impressions, 0 clicks, position 7.8. Ties to the Responsive Training Services ferroxyl course. Chemistry to be confirmed by whoever runs the test — the potassium ferricyanide / Prussian blue mechanism as described, and the solution shelf life and control-sample practice.
Image to shoot: Ferroxyl reaction on a stainless surface with clear blue indications, ideally showing a recognisable pattern (spark spatter or a sling mark).
Internal links already in the text: /ferroxyl-testing/, /passivity-testing/, /responsive-training-services/, /positive-material-identification-pmi/ — confirm they read naturally, do not add more.
