Corrosion Control--Types of Corrosion

Lee H. Erb

Originally published October 1997

Although there may be some differences of opinion on types of corrosion, I look at all corrosion as an electrical (or galvanic) process. The cause of the galvanic process is usually listed as the type. A designer will select material and processes that best suits the design requirements including avoiding or mitigating corrosion.

Electrolytic Corrosion

I prefer to use this term when moisture is present creating an electrolyte to permit electron flow. The moisture may come from condensation after takeoff as the trapped air cools at altitude or after landing as the cool aircraft causes condensation of the local humidity. This is very insidious and is part of the reason for commercial airliners to have the lower skin replaced periodically. It usually happens at the faying (that’s an old ship builder’s term) surfaces. Sometimes you cannot find it until you have taken rivets out and separated the parts (rib from skin, etc).

Salt Water Corrosion

This I consider an extreme form of Electrolytic corrosion but it is one that just about everyone understands. This is what Navy aircraft have to fight the most. In most specifications for a new airplane the Navy will give a minimum thickness of aluminum alloy such as .026 inch. Anything thinner loses a high percentage of its strength by the time corrosion is noticed.

Electrical Power

This type of corrosion is galvanic corrosion that is accelerated by being in the path of an electrical power circuit. Most homebuilt aircraft do not have a lot of accessories that are electric powered. They can stay away from this type of corrosion by having a wire for return circuit. More in the future on electrical bonding.

Intergranular Corrosion

This can be caused by poor processing during manufacture of the alloy (hence the need to be sure it meets a given specification). Usually it is caused by selecting the wrong chemical in attempting a conversion coating on a given alloy. Poor control of heat treatment processing is also a cause.

Over-etching aluminum can cause this type of corrosion.

Exfoliation Corrosion

This type of corrosion is similar to intergranular corrosion except the corrosion follows grain boundaries and "large" chunks fall out. Extrusions can be susceptible where grain boundaries are stretched and/or rough surfaces occur during the extruding process.

If you have a corrosion critical part you may wish to use bar or rod material that has been wrought instead of extruded. In industry extruded stock is used a lot but it usually goes through several inspections before going on an aircraft.

Fretting Corrosion

Fretting corrosion is caused by two surfaces rubbing together at a very small amplitude. I sometimes think that it should not be considered so much a corrosion as a wear. It can be eliminated by placing a very thin sheet of Nylon or Teflon between the surfaces.

Stress Corrosion

The high-strength heat-treatable wrought aluminum alloys in certain tempers are susceptible to stress-corrosion cracking, depending upon the product, section size, direction and magnitude of stress.

Stress corrosion cracking is where the internal stresses (residual stresses) vary across a section so that when they are loaded with additional outside forces, the grain boundaries at the surface start to break. When a part (sheet, extrusion, etc) is quenched and the outside layer cools too quickly, tension stresses are set up on the outside and compression stresses in the middle. This is sometimes taken care of by stretching or shot peening.

Local stresses (assembly stresses) can also be caused by selection of too small diameter of high shear fasteners (such as bolts), shrink or press fits, taper pins, and clevis joints in which tightening of the bolt imposes a sustained bending load on the clevis lugs.

I will never use 7075-T6 because the residual internal stresses can cause cracking without ever being loaded. I will use 7075-T7351.

I am aware of one light airplane that suffered stress corrosion around a bushing in the main spar of a wing. I suspect that particular airplane had been flown many times outside the design envelope.


Improper heat treatment can cause an embrittlement of an otherwise ductile material. Titanium makes a very good light weight spring but Titanium embrittlement can be caused by Cadmium. The cadmium, under pressure and/or heat, will flow (infuse) between the grains of titanium. This weakens the grain boundaries and when the titanium is stressed, a crack will initiate.

Cad plate applied on steel bolts and not subsequently baked can cause the bolts to break from "hydrogen embrittlement." More on embrittlement in the future.


Galling is not a corrosion but with all the talk about corrosion between dissimilar metals, galling problems should be discussed.

Galling is a condition where two parts (made from the same alloy) slide over one another and start ripping the surface between them.

Nuts and bolts are not made of the same material. If they were, they would gall and disassembly would be impossible. Each is made from a different alloy. To me it is interesting that the nut is always made the sacrificial member. If there is any shearing of threads, the nut will shear first. Many times if a nut is unknowingly sheared on assembly, it will fall off and the bolt will remain in place, averting disaster. More on bolts and nuts sometime in the future.


Thanks to Bob Urban and Stan Klein for trying to get me to understand.


1. "Dissimilar Metals," MIL-STD-889
2. "Metallic Materials and Elements for Aerospace Vehicle Structures," Military Handbook, MIL-HNBK-5
3. "Aircraft Corrosion Control," available from EAA (1-800-843-3612, stock number 21-37597, $10.95

EAA Chapter 1000 Home Page
E-Mail: Web Site Director Russ Erb at

Contents of The Leading Edge and these web pages are the viewpoints of the authors. No claim is made and no liability is assumed, expressed or implied as to the technical accuracy or safety of the material presented. The viewpoints expressed are not necessarily those of Chapter 1000 or the Experimental Aircraft Association.
Revised -- 22 April 1998