Brass corrosion and finish compatibility

A brass part rarely fails because the alloy was wrong. It fails at the interface between the alloy and whatever is supposed to protect it, and the first sign is usually a white bloom around a hole or a blister the size of a pinhead on a face that was perfect at dispatch.

Brass corrosion is a zinc problem before it is a coating problem. The zinc is the cheaper half of the alloy and the half that leaves first, either as a surface film that turns the part brown, or as selective leaching that turns it red and porous and takes the section with it. Which of those happens depends on three variables: zinc content, environment, and whether the surface was prepared for the coating that went on it.

What corrodes on a brass part first?

The zinc, in one of two ways. In clean dry air it forms a 5–20 nm oxide film within hours and a sulphide film within weeks, which is tarnish: cosmetic, self-limiting, and about 0.1 µm thick after a year indoors. In warm chloride or ammonia-bearing water it dezincifies: the zinc dissolves out of the alloy and leaves a copper sponge that looks red, feels rough, and has almost no strength.

Dezincification needs more than about 15% zinc to proceed, which is why every sheet brass in normal use is susceptible to some degree. It also needs an electrolyte and time. A 63/37 bracket in a dry switchgear cabinet will outlast the building. The same bracket in a cooling-tower enclosure, a dairy wash-down area or a coastal pump house can lose 0.1–0.5 mm of section a year, and the loss is not visible until the part is picked up and crumbles at the edge.

The third mechanism is galvanic. Brass sits well above steel, zinc and aluminium in the galvanic series, so when a brass part is bolted to a galvanised or mild steel frame and water bridges the joint, the steel becomes the anode and corrodes at a rate set by the cathode area. That is the mechanism behind the rust ring around a brass badge on a painted panel, and it is fixed with an isolating washer and a drain, not with a thicker coating.

Brass corrosion and finish compatibility by environment

The table below is a selection guide, not a specification. The fourth column is what our plant can actually do, which is narrower than what a finishing house with plating and lacquering lines can offer.

Two rows are usually argued. The marine row is a material change disguised as a finish question: no 80 µm film rescues a 63/37 bracket that sits in salt spray, and the honest answer is a dezincification-resistant brass or a 316L part. The electrical row is a masking question, and it is the one that fails a factory acceptance test because a coated earth stud reads open.

Why powder coating brass is not the steel process

Brass conducts, so it takes the electrostatic charge and the powder wraps the part well. That is where the similarity with steel ends.

The first difference is the oxide. Brass carries a zinc-rich oxide that forms again within hours of cleaning, and a powder film applied over it adheres to the oxide rather than to the metal. The bond fails at the interface, not at the topcoat, which is why a cross-hatch test can pass on a coupon and a production part can still chip to bare metal at a bracket edge.

The second is the cleaner. Brass is amphoteric: strong alkali attacks it as readily as strong acid, and a highly caustic degreaser that is perfectly safe on CRCA will etch a brass surface grey in a few minutes. The alkaline stage has to be mild, typically pH 9–10.5, and the dwell has to be long enough to lift the drawing lubricant without attacking the alloy.

The third is the rinse water. Brass leaves copper ions behind, and copper ions are the poison that stops a zinc-phosphating bath working. A line that runs brass before steel in the same tank will coat the steel parts poorly for the rest of the shift, and the defect appears on a completely different job.

The pretreatment sequence that works on brass

A brass part that is going to be powder coated needs five things to happen in order, and skipping any one of them moves the failure to a later date rather than removing it.

1. Degrease in a mild alkaline cleaner at 50–60 °C, with the dwell set for the oil actually on the part. Stamping lubricant and protective oil need different times.
2. Rinse in clean water, and keep the rinse moving; a still tank just redistributes the cleaner.
3. Deoxidise in a proprietary acid descaler that removes the zinc oxide without attacking the copper. This is the stage that decides adhesion.
4. Passivate with a benzotriazole-type anti-tarnish or a chromate-free conversion coating, which holds the surface until the powder goes on.
5. Rinse in deionised water and dry before the part cools, so no mineral film is left in the pits.

The seven-tank pretreatment line at Ganesh is the same line that runs steel, with the chemistry and the dwell set per substrate. The stages before the booth are described in degreasing brass parts before coating, and the copper version of the same problem — where the oxide is thicker and the rinse water discolours faster — is in pretreatment for copper parts.

What the curing oven does to the part

Powder cures at 180–200 °C for 10–20 minutes of metal temperature, depending on the chemistry. That is below the 250–325 °C stress-relief range for brass, so a half-hard part does not soften in the oven and a formed spring contact keeps most of its set.

Two things do change. Any residual stress from bending or drawing is still there, and if the service environment carries ammonia, a stressed brass part can crack by stress corrosion within months. The remedy is a stress-relief anneal before coating, not a thicker film. The second change is dimensional: brass expands about 19–20 µm/m·K, roughly 60% more than steel, so a 1,200 mm part grows 0.4 mm from 20 °C to 200 °C and comes back 0.4 mm on cooling. A coated part that is clamped cold on a fixture and released hot will keep a set.

Powder, wet paint or bare: the decision rule

Opaque protection with the best abrasion resistance: powder at 70–90 µm. Thin film where the brass colour has to read through, or where the part cannot take 200 °C: wet paint at 25–40 µm. Visible metal with no film at all: bare, with the tarnish accepted as a maintenance item.

The numbers behind that rule are film thickness and cure temperature. Powder gives roughly twice the film of a two-pack paint, which is why it survives a trolley knock that would mark a painted panel. Paint gives a thinner, smoother film and can be formulated clear, so a brushed brass face still looks like brass; the trade is lower abrasion resistance and a longer cure schedule at 60–80 °C for the two-pack systems we run. A clear powder over brass is possible and is the hardest of the three, but clear coats yellow under ultraviolet light, and a yellowed clear coat on brass reads as a dirty part rather than a protected one.

Edges, holes and masked faces

The laser-cut edge of a brass part is not the same alloy as the face. Zinc boils at 907 °C, below the melting range of the alloy, so the cut front loses zinc and leaves a copper-enriched layer a few micrometres thick. That layer is cathodic to the parent metal, which concentrates attack at the cut edge in a chloride environment — the opposite of what a deburred steel edge does.

Practical consequences: deburr and, where the service is aggressive, specify edge coverage rather than a face DFT. Powder wraps an edge poorly at 90°, and a sharp laser edge holds about 40–50% of the face film thickness. A 0.5 mm edge break gives the film something to cling to and removes the enriched layer at the same time.

Masked faces need the same discipline as on a steel cabinet. Earth studs, threads, gasket lands and any surface that has to stay conductive are taped before the booth, and the tape is removed before the oven or immediately after, not the next morning. A cured powder film on a thread is a re-tap, and re-tapping a brass thread removes the coating and the corrosion protection with it.

What can be proved in-house, and what cannot

Ganesh checks coating thickness with a DFT gauge on several points per face, and adhesion by cross-hatch on a coupon or a non-visible area. Those two checks catch the common failures: a thin edge, a contaminated surface, an under-cured film.

Salt spray is a different matter. ISO 9227 sets out the apparatus, reagents and procedure for neutral salt spray, acetic acid salt spray and copper-accelerated acetic acid salt spray, and it deliberately does not set the exposure period or interpret the result — those belong to the product specification. A 480-hour NSS claim is a test on a stated specimen against a stated acceptance criterion, and it is run by a laboratory, not inferred from a thickness reading. Where a drawing calls for it, the test is a third-party activity and the report is part of the PPAP pack.

What to put on the brass finish schedule

• Alloy and temper, because a DZR grade is a different purchase from C27200.
• Service environment, in words: indoor dry, coastal, wash-down, immersed.
• Finish system, stage by stage: cleaner type, deoxidise, passivation, powder type, DFT range.
• Edges, with the deburr callout and whether edge coverage is required.
• Masked areas, listed by feature rather than by note.
• Electrical requirements, with the bond resistance to be achieved after coating.
• Salt spray or adhesion test, with the standard, the hours and the acceptance criterion.
• Colour and gloss, with a sample or a RAL number, because brass under a translucent coat changes hue with film thickness.

The RFQ checklist carries the finish fields in the same order, and the materials hub links the equivalent schedules for stainless, aluminium and CRCA. Where the environment is the open question, request a quote with the service described in words and we will propose the finish system rather than guess at one.

Where this route stops

We powder coat and wet paint, over a seven-tank pretreatment line with a 16 ft × 16 ft blast chamber and an 18 ft × 14 ft powder booth, as described on the paint shop page. We do not electroplate, so nickel and chrome on brass is a bought-in service and the part leaves our plant unplated. We do not lacquer, we do not hot-dip galvanise, and we do not apply a patinated or antiqued finish.

The other limit is alloy supply. A dezincification-resistant brass may be the correct answer and still be a six-week mill order in the thickness the drawing needs, which is why the material decision belongs at the design stage rather than at the quotation. Where the schedule cannot wait, the coating has to carry the whole duty, and that is a conversation to have before the drawing is released.