Copper corrosion and finish compatibility

Copper corrosion almost never destroys a sheet metal part. At roughly 1–2 µm a year in urban air, a 1 mm panel would take centuries to perforate, and the green patina that alarms a first-time buyer is a protective film rather than a failure.

The failures that do happen are at the edges of the material decision: a joint that goes high-resistance because the oxide grew under the bolt, a steel panel that rusts in a ring around a copper fixing, a white wall stained green by copper runoff, or a powder coat that lifts because the copper oxidised in the oven before the film set. Each has a cause and a number, and each is cheaper to design out than to explain later.

What actually happens to bare copper?

It oxidises in stages, and the stage determines whether the surface is still usable as a conductor or as a finish.

Within minutes the surface forms a 2–5 nm cuprous oxide film, which is enough to raise contact resistance but not enough to see. Within days the film thickens and darkens to brown. Over months in sulphur-bearing air it goes grey-brown, and over years outdoors it develops the green basic copper carbonate or sulphate that most people recognise as patina. The rate depends on sulphur dioxide, chloride and time of wetness: about 1–2 µm a year in a city, higher within a few hundred metres of the sea or beside a chimney.

Two consequences follow. The first is that copper is a poor choice for a part that must stay bright without a coating. The second is that copper is an excellent choice for a part that must last, because the film that forms is coherent and self-limiting. The mistake is to specify copper for appearance and then treat the tarnish as a defect.

Copper corrosion and finish compatibility by environment

The table is a selection guide for sheet parts. The fourth column is what our line can do, which stops short of plating and lacquering.

The row that surprises buyers is the seventh. Copper is not attacked when it is bolted to steel; the steel is. A copper washer on a galvanised frame looks like a detail and behaves like a small, efficient cathode, and the corrosion appears as a ring of rust a season later.

Why powder coating copper is different from coating steel

Copper conducts, so electrostatic deposition works and the powder wraps the part well. Everything else about the process has to change.

The oxide grows during cure. Powder cures at 180–200 °C for 10–20 minutes of metal temperature. At that temperature copper oxidises quickly — tens of nanometres in minutes in air — and the oxide forms between the passivation and the metal, or under the film if the passivation is thin. The result is a film that passes a tape test on the day and lifts at a scribe a month later.

The cleaner must not etch. Copper is attacked by strong acids and by ammoniacal cleaners, and a bright dip that is right for steel will pit a copper surface. A mild alkaline degrease followed by a controlled acid deoxidise is the sequence that leaves a uniform surface.

The rinse water carries copper. Copper ions in a shared phosphating bath poison it, in the same way brass does, and the damage shows up on the next steel batch. Copper and brass belong on their own sequence.

The coating hides the material. Copper under an opaque powder is a conductor in a plastic shell, which is correct for a shield or a housing and pointless for a decorative panel. If the appearance matters, the finish is the product and should be chosen first.

The pretreatment detail is covered in pretreatment for copper parts, and the degrease stage in degreasing copper parts before coating.

Wet paint, clear coat or bare

A two-pack paint at 25–40 µm cures cooler than powder, at 60–80 °C, and that temperature matters on copper: less oxide grows during cure, so the adhesion window is wider and a clear or tinted film holds its colour better. The trade is abrasion resistance, which is lower than a 70–90 µm powder film.

A clear coat over polished copper is the finish most buyers picture and the one that ages worst. Clear coats yellow under ultraviolet light, and a yellowed film over copper reads as a dirty part. It also locks in whatever oxide was present at coating, so the polish has to happen immediately before the passivation and the part cannot sit overnight.

Bare copper is the right answer in three cases: an electrical contact that must not be insulated, a part whose patina is the design intent, and a part that will be assembled into a sealed enclosure where appearance does not matter. In all three, say so on the drawing, because the default in a fabrication shop is to coat everything that arrives.

Galvanising, plating and the finishes copper cannot take

Hot-dip galvanising is not an option. The bath runs at 445–465 °C, which is close to the annealing range of copper, and the zinc diffuses into the surface to form brittle copper-zinc intermetallics. The part also distorts, because a 1.5 mm sheet at that temperature will not hold flat. A zinc coating on copper is a galvanic cell where the zinc is consumed first, so even a zinc-rich paint is a short-lived protection on a conductive substrate.

Electroplating is the finish copper is genuinely good at — nickel, tin and silver all bond well to a clean copper surface, which is why busbar is often tinned or silver-plated at the joint. None of that happens at Ganesh. Plated copper is a bought-in service with its own lead time, and the plating specification belongs on the drawing with the thickness and the test method.

Anodising, of course, does not apply: copper has no stable anodic oxide of the kind aluminium forms.

The joint is where copper corrosion gets expensive

A bolted copper lap joint that is cleaned and tightened measures in the single-digit micro-ohms. That number is the whole point of the joint, and it is also the number an oxide film destroys.

Cuprous oxide is a semiconductor, and a 50–100 nm film is enough to raise the resistance of a clamped joint by an order of magnitude once the clamp load relaxes. The failure is self-reinforcing: higher resistance means more I²R heating, more heating means faster oxidation, and the joint that measured 8 µΩ at commissioning measures 60 µΩ a year later with a warm spot you can feel.

Three controls prevent it. Clean and abrade the joint faces immediately before assembly, not the day before. Apply a contact grease or a joint compound that excludes oxygen. And specify the torque, because a joint at half the specified torque has half the contact area. For a 50 × 10 mm bar the difference between 20 N·m and the specified 45 N·m on an M10 bolt is the difference between a joint that holds and one that heats.

Copper that is powder coated needs the same discipline at the masked faces. A coated joint face is an insulated joint face, and an insulated joint in a 1,000 A circuit is a fire.

Where copper run-off goes

Copper does not stay where it corrodes. The dissolved metal runs off the surface with rain and deposits as a green stain wherever the water lands, typically 0.5–2 g per square metre per year in a European or Indian urban atmosphere and more at the coast.

That is why a copper flashing on a white rendered wall leaves a green streak three metres long within two years, and why the stain is so hard to remove: the copper has reacted with the render, not simply dried on it. The design answer is a drip edge and a drainage path that ends over a surface that will not be seen, or a coated part that does not release copper in the first place.

The same runoff attacks what it lands on. Water running off copper onto galvanised steel or aluminium carries copper ions that plate out on the less noble metal and set up a galvanic cell on the surface itself, which pits the zinc or the aluminium under a green film. Where a copper part sits above a coated steel panel, the drainage is an engineering detail with a number attached: keep the water off the panel or coat the copper.

Inspection: what proves a copper finish

Thickness and adhesion are checked the same way as on steel: a DFT gauge on several points per face, and a cross-hatch on a coupon or a hidden area. On copper the cross-hatch has a second purpose, because the failure mode is cohesive — the film comes away with oxide attached — and a clean cut with no lifting is the evidence that the pretreatment did its job.

Salt spray is external. ISO 9227 sets the apparatus and procedure for neutral, acetic acid and copper-accelerated salt spray and deliberately leaves the exposure period and the pass mark to the product specification, so a 480-hour claim is a test on a stated specimen, not an inference from a thickness reading. On copper, salt spray produces green staining at scribes and edges long before it produces a perforation, and the acceptance criterion should say whether staining counts as a failure.

What to put on the copper finish schedule

• Grade and temper, because C11000 and C12200 do not tarnish at the same rate in a brazed joint.
• Environment, in words: indoor, coastal, wash-down, ammonia-bearing.
• Finish system, stage by stage, including the passivation and the cure temperature.
• Masked faces, listed by feature: contact pads, earth points, joint faces.
• Edge treatment, since a laser-cut copper edge is thin and takes little film.
• Electrical requirement, with the bond or contact resistance to be achieved after coating.
• Test method and pass mark, with the standard number and the hours.
• Appearance standard, as a sample or a colour reference, not the word "natural".

The RFQ checklist carries the same fields, and the materials hub has the brass and stainless equivalents. If the finish is the unresolved item, request a quote with the environment and the conductive faces marked, and the pretreatment route can be fixed at quotation instead of at first article.

Where this route stops

We cut and form copper, and we powder coat or wet paint it over a seven-tank pretreatment line with a 16 ft × 16 ft blast chamber, an 18 ft × 14 ft powder booth and a curing oven, as set out on the paint shop page. We do not electroplate, we do not lacquer, we do not galvanise, and we do not supply a chemically patinated or antiqued finish.

The honest limit is the one that matters commercially: if the part's whole reason for being copper is appearance, the finish is a specialist job and we are the wrong supplier for it. If the reason is conductance, shielding or thermal duty, the coating is protection and masking is the critical operation — and that is work we do every week.

Whatever the answer, the decision belongs on the drawing. A copper part specified as "copper, finish as standard" gets whatever the shop did last, and the corrosion that follows is not a material problem at all.