Why Stainless Steel Rusts Around Welds, and How to Prevent It

You specified stainless steel because it resists rust. So when brown staining appears around the welds a few weeks after installation, it looks as if you were sold the wrong metal. In most cases, you were not.

Why stainless steel resists rust in the first place

Stainless steel does not resist corrosion because it is a completely different metal from ordinary carbon steel. Underneath, it is still mostly iron, and iron rusts. What makes it stainless is chromium (at least 10.5% of the alloy) and what that chromium does at the surface.

When chromium meets oxygen, it forms an extremely thin, tightly bound film of chromium oxide across the surface. This is the passive layer. It is only a few atoms thick and invisible, and it separates the iron underneath from moisture and air. If the film is scratched, it forms again almost immediately, as long as chromium and oxygen are both available. That ability to repair itself is the reason stainless steel works. The protection is not a coating that can wear through: it is a property of the metal, and it keeps renewing itself.

Every problem at a weld has the same basic cause: something has damaged that passive layer in the place where the heat went in. The rest of the part is unaffected and stays bright.

The four things that make a weld rust

Brown staining beside a weld usually comes from one of four causes, and often from several of them together. That is why the cure is rarely a single action.

The four causes are:

• Heat tint: the straw, blue or brown colour next to the weld. When the surface is heated in air, the passive film thickens and its chemistry changes. The chromium in that thicker oxide is drawn from the metal just beneath it, which leaves a chromium-poor layer that protects far less well than the original film.
• Sensitisation: in grades with more carbon, the metal beside the weld passes through a band of roughly 450 to 850°C. That is the range commonly given for austenitic (chromium-nickel) stainless steels such as 304 and 316. In that range, carbon and chromium combine into chromium carbides at the grain boundaries. Those carbides hold chromium that should be dissolved in the metal, so the boundaries do not have enough chromium left to stay passive.
• Iron contamination: free iron transferred onto the stainless steel from carbon-steel tools, grinding wheels, wire brushes, or even from storage on a mild-steel rack. That embedded iron rusts, and the rust runs and stains the stainless steel around it, even though the stainless steel itself is sound.
• An unshielded weld root: the back of the joint oxidises heavily during welding because no inert gas protected it. In the worst cases the root turns dark, rough and crusty, which welders call "sugaring". A root in that state has lost chromium and corrodes easily.

Only one of these causes, sensitisation, depends on how much carbon the alloy contains. The other three depend on how the weld was made, protected and cleaned, which is why choosing the grade alone does not solve the problem.

What the grade can fix, and what it cannot

The grade does real work against two of the four causes. Molybdenum in 316 and 316L improves resistance to chloride pitting. So if a weld zone has lost some of its protection, it copes better in salt air or washdown than the same zone would in 304. The low-carbon "L" grades, 316L and 304L, deal directly with sensitisation. With less carbon available, far fewer chromium carbides can form in the 450 to 850°C band, so the grain boundaries keep their chromium and their corrosion resistance.

Where the metal is welded and then used in a corrosive environment, a low-carbon grade is the sensible default. Stabilised grades, 321 and 347, reach the same result by another route. They contain titanium or niobium, which combine with the carbon first and form their own stable carbides, so the chromium stays in the metal. They are common in higher-temperature service, where sensitisation is a continuing risk.

The "L" grade and the shielding gas fix different problems. Neither one covers for the other, and a rusting weld usually means one of them was skipped.

This is the limit of what the grade can do. 316L does nothing about heat tint, iron contamination or an unshielded root, because those come from the welding process, not from the chemistry of the alloy. You can specify the best grade for welding and still get a rusting joint if the tint was never cleaned off or the root was left unprotected. A better grade gives you more margin against corrosion, but it does not replace a properly made weld.

When you write the specification for welded stainless steel in any chloride or washdown environment, include both of these:

• a low-carbon grade (316L or 304L)
• a statement that post-weld pickling and passivation are required.

A common mistake is to name the grade and leave out the cleaning after welding.

The role of shielding gas and back-purging

Hot and molten stainless steel takes up oxygen from the surrounding air very readily, and all the oxygen it takes up thickens and weakens the surface oxide. The job of the shielding gas is to keep air away from the metal while it is hot enough to react.

On the face of the weld, the side the welder can see, an inert gas such as argon flows from the torch and covers the weld pool. When this is done well, the front of the joint stays bright, with no more than a faint straw tint. Most fabricators protect the front of a weld well, because it is directly in front of them.

The root is the side you cannot see, and it is usually where the trouble starts. On pipe, tube and closed vessels, the inside of the joint is heated as much as the outside, but no gas from the torch reaches it. Unless the inside is filled with inert gas, which is called a back-purge, the root oxidises, becomes heavily tinted and, in the worst case, sugars. That is why a pipe or tank can look perfect on the outside and corrode from the inside. The surface that was left unprotected during welding is the same surface that the fluid touches.

Back-purging means replacing the air inside the pipe or vessel with argon before and during welding, so that the root forms under the same protection as the face. It takes setting up, gas and time, particularly on large or complex assemblies. That is exactly why it is sometimes skipped, and why the hidden surfaces are the ones that fail.

Pickling, passivation and mechanical cleaning: what each one does

After welding, the surface next to the weld is no longer in the condition stainless steel needs to protect itself, and post-weld treatment returns it to that condition. The three methods do different things. If they are confused, the work can look finished when it is not.

The differences to remember are:

• Pickling uses an acid, typically a mixture of nitric and hydrofluoric acid applied as a paste or in a bath. It chemically removes the heat tint and, most importantly, the thin chromium-poor layer of metal beneath it. Pickling removes material, not just colour: it takes off the damaged surface and exposes fresh metal of the correct composition underneath. This is the step that deals with heat tint and with a sensitised or oxidised surface.
• Passivation uses a milder acid, commonly nitric or citric. It does not remove metal. Instead, it cleans free iron off the surface and helps a complete, chromium-rich passive film to form again. Passivation restores the protective layer, but it does not remove heat tint.
• Mechanical cleaning means abrasive blasting with clean media suitable for stainless steel, or grinding or brushing with tools used only on stainless steel. It physically removes tint and oxide. It is a valid alternative to pickling where acids are impractical, but the media and tools must be free of contamination, or they will press fresh iron into the surface. Mechanical cleaning alone does not fully rebuild the passive film, so it is normally followed by passivation.

So the practical order for a weld where corrosion matters is this. First remove the tint and the chromium-poor layer, by pickling or mechanical cleaning, and then passivate to restore the film. TH Steel carries out pickling and passivation in-house, so this finishing step is done in the same workshop, under the same control, as the welding.

Where weld rust matters most

Not every faint tint is a problem, and treating every weld as critical wastes money. The right level of finishing depends on what the part does and where it is used.

Weld condition matters most in these cases:

• Hygienic work: food, beverage, pharmaceutical and process equipment. Here a rough, tinted or sugared internal surface is a corrosion site, and it is also a place where product can collect and bacteria can grow. Internal welds on tanks and pipework need clean roots and full post-weld treatment.
• Marine and coastal work: anywhere chlorides reach the metal. A chromium-poor weld zone that would last indefinitely indoors will pit within weeks in salt air. The passive layer must be fully restored, and the grade may also need to change to 316L or duplex 2205 for the conditions.
• Visible architectural work: handrails, balustrades, façade elements and feature metalwork. Here staining is a defect the client can see, whatever its structural effect. A uniform finish and clean, treated welds are part of the specification, whether or not the environment is aggressive.

By contrast, a light straw tint on a bracket in a dry, indoor location where nobody sees it is usually only cosmetic, and it carries little risk of corrosion. Knowing which case you are in stops a specification from under-protecting a marine tank, or from paying for full finishing on a hidden support. If you are unsure, tell the fabricator what the part does and ask them to advise how much finishing it needs.