Galvanic Corrosion in Mixed-Metal Joints: How to Specify Fastener Material, Isolation and Verification When No Numeric Threshold Exists

Specify the joint, not a corrosion-proof fastener

A galvanic cell needs dissimilar electrochemical potentials, metallic contact and an electrolyte. Remove or control any one of the three and the cell stops ¹. That is why the specification question is not "which fastener resists corrosion" but "what does this joint expose, and which member becomes the anode" ²⁰. In practice the less noble metal becomes the anode and corrodes. So the fastener should sit on the noble side of the pair or close enough to the joined metal that the difference is negligible ¹². The practical framing is to specify the fastener as part of the joint system: both base metals, the finish on each component, the expected electrolyte exposure. And whether any isolation can be maintained through installation and service life ²⁰. When you draft the RFQ, list the base metals and their finishes, the service electrolyte, and the maintenance access to the joint. A supplier quoting against that list can tell you which member is anodic; a supplier quoting against a fastener grade alone cannot.

Why you cannot write a millivolt threshold into the RFQ

Independent publishers agree on the direction of risk: the further apart two metals sit on the galvanic series, the faster the anodic member is attacked, and close pairing or isolation is the mitigation ^{12, 13}. One source lists pairings with large potential differences — aluminium with copper, carbon steel with stainless, aluminium with stainless, zinc with copper, magnesium with almost anything — as ones to avoid or isolate heavily ¹⁴. None of them gives numeric potentials or an allowable potential difference, so a millivolt limit in your RFQ would be a number you invented. Ask the supplier instead to state the galvanic-series position of each joined metal and of the proposed fastener, to name the anodic member for your electrolyte. And to justify the pairing or the isolation on that basis. That is a clause the evidence supports; a threshold is not. The same caution applies to area ratio. A small anode against a large cathode concentrates the corrosion current on the small anodic area. But the actual exposed areas, coating continuity, drainage and crevices decide the real risk rather than a material-name rule ¹. A small stainless fastener in a large bare carbon-steel member is not automatically the worst ratio; a small coating defect beside a larger passive stainless surface can be the more concentrated anode ¹.

Match the fastener to the joined metal, or isolate the joint

The consistent guidance is to use a fastener of the same material as the joined component, or one very close to it on the galvanic scale: zinc-plated steel for carbon steel, a compatible stainless grade for stainless, aluminium for aluminium ¹⁵. That is one supplier's stated position, and it comes with its own caveat — corrosion compatibility is only one part of the choice. And clamp load, thread performance, vibration resistance and service temperature still govern whether the joint performs ¹⁵. Aluminium is where the trade-off bites. Aluminium fasteners reduce the mismatch but may not meet strength, wear or installation requirements, so stainless is frequently specified instead, with isolating washers or sleeves and joint sealing where exposure is expected ²⁰. If you go that route, note that the aluminium is the anode and the risk scales with the electrolyte present ¹⁷. Where matching is not possible, breaking electrical continuity can be more effective than changing the fastener material alone. Non-metallic washers, sleeves, spacers and bonded sealing washers isolate the head and shank and seal the interface against electrolyte ingress ¹⁰. Treat this as a joint-stack decision: the isolation has to survive the same service life as the fastener, which is the subject of the next check.

Qualify the isolation hardware before it enters the RFQ

Isolation works by removing the metallic contact and sealing the electrolyte path, but the components doing that job are non-metallic and sit under preload. They must withstand compression, temperature, UV exposure and chemical contact; if the bushing cracks or the washer cold-flows, metal-to-metal contact returns ¹⁰. Stack-up changes also affect clamp load and torque response, so the tightening specification has to be revisited once isolation is added ¹⁰. There is a further restriction to check on structural work. Dielectric washers and bolt sleeves can fully isolate the bolt, but the dielectric material must be evaluated for loading compatibility, and for high-strength connections shims and washers are not allowed under existing standards ². If your joint is a high-strength bolted connection, confirm with the supplier and your connection designer whether isolation hardware is permissible at all before you write it into the RFQ. Ask the supplier to state the compression, temperature and chemical conditions the washer or sleeve is qualified for, and to confirm the effect on clamp load and torque. No dielectric strength, temperature rating or torque rating for isolation hardware appears in this source set, so those values have to come from the supplier's own data.

Set the service environment, because it decides the verdict

The same material pair can be acceptable or unacceptable depending on where it lives. A theoretically risky combination might perform adequately in a dry, controlled environment, while even a compatible pair can degrade in saltwater ¹⁴. Steel into aluminium shows little to no galvanic effect in dry indoor service and significant aluminium degradation at the fastener interface in marine or outdoor exposure ¹⁷. Humidity, temperature and salt exposure all shift the corrosion rate, which is why the environment belongs in the RFQ as a stated condition rather than an assumption ²⁴. For marine, coastal or high-chloride service, corrosion-resistant stainless grades are the indicated starting point ²⁵. Write the electrolyte, chloride level, humidity and temperature range into the RFQ and require the supplier to confirm the pairing against that environment. A quote that does not restate the environment is a quote you cannot compare with another.

What to put in the verification package

Each check below rests on a named standard with its own duration or acceptance criterion, so the supplier can quote against it and you can accept or reject against it. The last item is the exception: no source in this set gives an isolation continuity or resistance test with acceptance criteria, so you have to define it and agree it with the supplier.

• Sustained-load hydrogen embrittlement test per ISO 15330: 200 hours minimum at about 75% of proof load, no failures ⁹
• Alternative embrittlement evaluation per ASTM F519: 200 hours at 75% of notched fracture strength ³
• Process control for plated or coated fasteners per ASTM F1940, with baking validated by sustained-load or process qualification testing ^{3, 6}
• Torque-tension test per ISO 16047, reporting the friction coefficient — a typical lubricated target is 0.12–0.18 ⁹
• Cyclic corrosion test per VDA 621-415 or GMW 14872 rather than continuous salt spray alone ⁹
• Coating adhesion by cross-cut per ISO 2409, sampled from each production batch, with no coating removal at cut edges ⁹
• Isolation continuity or resistance test: define the method and acceptance criterion yourself and agree it with the supplier before award

When the documentation premium is worth paying

Verification burden has a price. One source reports that hot-dip galvanizing to ASTM F2329 carries a 10–25% cost premium over ASTM A153. And attributes it to mandatory adhesion testing on every batch lot, higher coating thickness minimums for small fasteners and heavier requirements for high-strength bolts ¹¹. That source's own recommendation is to specify F2329 only where critical structural applications, high-strength fasteners or rigorous quality documentation justify the premium ¹¹. For your RFQ, the question is whether the joint needs per-batch adhesion evidence or whether the simpler standard gives adequate assurance. General construction hardware in non-critical applications is described as adequately served by A153 with a lighter inspection burden ¹¹. If your joint is a high-strength or structural connection, the premium buys documentation that matches the consequence of failure; if it is an indoor enclosure, it buys evidence the joint does not need.

Where the sources disagree or fall silent

Two of these rows are genuine disagreements you should resolve with the supplier rather than assume. The rest are silences, and each one becomes a supplier-verified item in the RFQ instead of a figure you supply yourself.

FAQ

Related pages

• Surface finishes product page
• Threaded Fasteners vs Snap-Fits, Rivets and Adhesives: What the Evidence Can Settle Before You Specify
• Installing a Bolt to a Specified Torque: Procedure, Tooling and Verification
• Thread Pitch and Engagement Length for Steel, Aluminium and Plastic Joints: What the Evidence Settles
• Cold Heading vs Machining for Fasteners: How to Choose a Route on Strength Class, Coating and Qualification Evidence