Skip to content
Beaconby Winda
FactCorrosion and Damage Mechanisms · General and localized corrosion

What is galvanic corrosion?

·5 sources

Short answer

Galvanic corrosion is the faster corrosion of one metal when it is electrically connected to a more noble metal in a shared electrolyte, and API RP 571 (third edition, 2020) lists it as a damage mechanism in section 3.31. The more active metal, the anode, corrodes faster; the more noble cathode corrodes little or not at all.

By , Editor·Researched and drafted by ·Checked by , Editor·Last checked ·Next review

Two different metals immersed in the same conducting liquid usually sit at different potentials. Connect them, and electrons flow from one to the other: the less resistant metal becomes the anode and corrodes faster, while the more resistant one becomes the cathode and is largely spared 4. The extra loss on the anode is galvanic corrosion, also called bimetallic, dissimilar metal or contact corrosion, and the same effect used on purpose is how a sacrificial anode protects steel 2.

When does it happen?

Three conditions have to meet: two dissimilar metals, an electrolyte, and an electrically conductive path between the metals 3. The UK National Physical Laboratory (NPL) adds two refinements. The electrolyte can be a thin condensed film, damp soil, salt deposits or corrosion products, and the electrical path can run through structural steel, cabling or earthing, so the junction itself need not be wet. A cathodic reaction must also keep running on the noble metal, which in most practical cases is the reduction of dissolved oxygen 2.

How fast the anode is lost then depends on factors that NPL lists as area ratio, temperature, flow rate and the make-up of the electrolyte, among others 2:

  • Area ratio. A large cathode coupled to a small anode concentrates attack on the anode. In still, aerated water the total galvanic attack scales with the cathode's area, which NPL calls the catchment area principle, so steel rivets or bolts in nickel, copper or stainless steel plate are a known bad pairing 2.
  • Conductivity. Seawater conducts about 40,000 microsiemens per centimetre (µS/cm), against 50 to 1,500 µS/cm for supply water, so in seawater attack spreads further and small anodes can corrode severely 2.
  • Temperature. Polarity can reverse: above about 60 °C iron becomes anodic to zinc, and in seawater the aluminium and iron couple reverses at about 95 °C 2.
  • Biofilm and chlorination. In natural seawater a biofilm makes passive alloys very efficient cathodes. Water heated 25 to 30 °C above ambient grows no biofilm, and oxygen reduction becomes 2 to 3 orders of magnitude less efficient; dosing 0.5 to 1.0 milligrams per litre (mg/L) of chlorine also cuts cathode efficiency, by 1 to 2 orders of magnitude 2.

In oil and gas process fluids the effect is usually smaller than in seawater. Laboratory studies reviewed by NPL, mostly in sweet brines and one in a sour brine, found that coupling a passive alloy to an active one no more than doubled the active alloy's corrosion rate at equal areas, and up to tripled it at a 4:1 passive-to-active area ratio 2. Downhole, where a corrosion resistant alloy (CRA) touches carbon steel, the risk shifts to the cathode: hydrogen evolved on the CRA has embrittled duplex stainless steel tubulars and alloy K-500 drill collars 2.

Which oil and gas metal pairs are at risk?

The galvanic series ranks metals by their potential in seawater. The NASA and NPL versions agree on the broad order: graphite, platinum and gold at the noble end, with titanium and passive stainless steels close behind, copper alloys in the middle, then cast iron and carbon steel, and aluminium alloys, zinc and magnesium at the active end 3 2. Stainless steels such as 316 appear twice: high while their protective film holds, far lower once it breaks down 2.

NPL also sorts common seawater alloys into four groups. Group 1 holds noble, passive alloys (super duplex and super austenitic stainless steels, titanium, nickel-chromium-molybdenum alloys); group 2 holds alloys that resist corrosion but suffer crevice attack (22Cr duplex, 316, alloys 400, K-500 and 825); group 3 is mainly the copper alloys; group 4 holds carbon steel, cast iron and aluminium alloys. Alloys within one group can be coupled without significant problems 2.

Beacon's table applies that grouping to pairs found on oil and gas facilities; notes come from NPL unless marked 2 3.

Pair and typical locationMetal that corrodesNPL seawater groupsWhat the sources record
Carbon steel pipe with a 316 stainless steel valve or fittingCarbon steel4 and 2A small anode beside a large cathode makes it worse; NPL advises coating the cathode in preference to the anode
Carbon steel weld metal in carbon steel pipeWeld metal4 and 4A weld bead about 50 mV more negative than the pipe corroded rapidly in partly deaerated seawater (about 0.1 mg/L oxygen)
Super duplex piping with a nickel aluminium bronze pump or valveBronze1 and 3At a 10:1 cathode-to-anode area ratio over 63 days, heavy attack in natural seawater, little or none in chlorinated seawater
Super duplex piping with a group 2 spiral-wound gasket (such as 316)Gasket1 and 2Has caused failures at flanged joints on offshore platforms
CRA tubing in contact with carbon steel casingCarbon steelProcess fluid, not groupedHydrogen evolved on the CRA cathode can embrittle it
Carbon steel structure with zinc or aluminium anodesThe anodes, by designBelow steel in the seriesThis is cathodic protection (NASA series places zinc and aluminium alloys below mild steel)

The potential gap is a guide, not a forecast. NPL records pairs only 50 millivolts (mV) apart that have corroded, and pairs 800 mV apart that have been coupled successfully, because the rate depends on the galvanic current, which area, flow, temperature and chemistry control 2.

How is it found?

Galvanic attack gives itself away by location: corrosion builds up on the anode at, and close to, the joint between the two metals 3. Careful visual examination, ideally before the surface is cleaned, helps tell it apart from other forms of corrosion 4. In a thin film or low-conductivity water the attack stays close to the junction and can be intense, while in seawater it spreads further 2.

It can be quick: NASA's Kennedy Space Center corrosion laboratory found a stainless steel screw corroding aluminium after six months at its atmospheric test site 3.

Isolation flanges, polymer gaskets and insulating sleeves and washers break the circuit 3, but isolation has to be checked rather than assumed. Later modifications can bridge an insulated joint, and earthing, often a safety requirement offshore, can connect both sides of an insulated flange; NPL recommends confirming isolation when equipment enters service 2.

For fixed equipment, API Recommended Practice (RP) 571, third edition (March 2020), covers galvanic corrosion as its own damage mechanism in section 3.31, and API includes that section in the body of knowledge for its piping inspector recertification 1.

Is galvanic corrosion the same as dissimilar metal weld cracking?

No. Galvanic corrosion is electrochemical and needs a wet electrolyte; a weld can suffer it when the weld metal is anodic to the parent metal, as in the seawater case above 2. Dissimilar metal weld cracking is a different problem: welds joining ferritic steels to austenitic alloys, used widely in power plants, fail early from carbon diffusing out of the ferritic steel, local creep strain and mismatched thermal expansion, with no electrolyte involved 5. Galvanic coupling, by contrast, can go on to start pitting on the anode 2.

Common questions

Is galvanic corrosion the same as dissimilar metal weld cracking?
No. Galvanic corrosion is electrochemical attack on the more active metal of a wet couple. Dissimilar metal weld cracking is an early failure of welds between ferritic steels and austenitic alloys, driven by carbon migration, creep strain and mismatched thermal expansion, with no electrolyte needed.
Does a bigger potential difference always mean worse galvanic corrosion?
No. Pairs only 50 mV apart have corroded and pairs 800 mV apart have been coupled without trouble, because area ratio, flow, temperature and how well the cathode reduces oxygen set the actual current.
Why can an insulating flange fail to stop it?
If both sides of the flange are earthed, or a later modification bridges it, the two metals are electrically connected again. Isolation therefore has to be confirmed when the equipment enters service, not assumed from the design.

Beacon explains these concepts in its own words for education. It does not reproduce or replace the standards cited. Engineering decisions need the current edition of the standard and a qualified engineer.

Sources

  1. 1.API 570 Recertification Publications Effectivity Sheet, listing API RP 571, Damage Mechanisms Affecting Fixed Equipment in the Refining Industry, 3rd Edition, March 2020, section 3.31 Galvanic Corrosion · American Petroleum Institute ·
  2. 2.Guides to Good Practice in Corrosion Control No. 5: Bimetallic Corrosion (R. Francis, A. Turnbull, G. Hinds, 2020) · National Physical Laboratory
  3. 3.Forms of Corrosion (galvanic corrosion and galvanic series in sea water) · NASA Kennedy Space Center Corrosion Technology Laboratory ·
  4. 4.Fontana's Eight Forms of Corrosion (after Fontana and Greene, 1967) · AMPP
  5. 5.Additive Manufacturing of Functionally Graded Materials between Ferritic and Austenitic Alloys, PhD dissertation (J. S. Zuback, 2019) · The Pennsylvania State University