Galvanic corrosion

[Bubulus]

I hope to pose what is probably a basic chemical interaction question.

We have attempted to bond an iron oxide magnet to 1010 steel using a 3M aluminum-pigmented one part epoxy. The epoxy contains 30-60% by volume aluminum pigmentation. However, after spending some time in the field, these magnets fail - the magnet has very cleanly de-bonded. All of the epoxy remains stuck to the steel.

There are certainly a number of things that could cause this problem, but I am specifically interested in finding out if the aluminum pigmentation in the epoxy could be corroding the iron oxide. If the aluminum is sucking electrons (or the oxygen atoms) from the iron, the free iron would then rust and the bond could fail. Since this would occur on such an atomic level, it would probably not be discernable to the naked eye. ...right? I'm also assuming that there is an electrolyte in the environment, possibly humid or salty air.

Lastly, I cannot guess how the aluminum pigmentation was manufactured (ie, how much exposure to oxygen it already has). I know that aluminum is very reactive and will oxidize upon exposure to air.

The long and the short of it is, could there be a spontaneous chemical reaction between Fe2O3 and Al?

 

[ChasBean1]

Bubulus, I like your philosophy that the Al(3+, if I remember right) can certainly react with the Fe(2-, again, if I remember right) to exchange electrons to eventually disable the magnet. I'm not a chemical engineer, but I did stay at a Holiday Inn Express last night... I'm a mechanical engineer with a naval nuclear history as an Engineering Lab Technician.

My best recommendation would be to research, find, and try a different bonding agent. Best of luck, -Chas

 

[Bubulus]

Thanks for your reply, Chas. We wound up with a list of 8 or 9 different potential causes for failure (including corrosion) but we weren't able to replicate the problem through any of our testing. The time scale is such that we won't know for a couple years whether or not the changes we made were effective solutions. Wish I could post a more detailed follow up But again, thanks for your time.

Erik

 

[Bubulus]

Thanks for your reply, Chas. We wound up with a list of 8 or 9 different potential causes for failure (including corrosion) but we weren't able to replicate the problem through any of our testing. The time scale is such that we won't know for a couple years whether or not the changes we made were effective solutions. Sorry I can't post a more detailed follow up But again, thanks for your time.

Erik

 

[phoward]

I like Erik's reply and honesty...

Dissimilar metals do interact with each other to cause problems. But I think you are seeking a metal-to-metal solution when you could be seeking a different starting point. Are you limited to a one-part epoxy? Can you use a mechanical attachment? Is magnet failure the separation from the 1010 or demagnetizing or both?

Those are just to start the thinking process, because my actual suggestion is to go to the adhesive manufacturer you are already using and ask them. They have seen the problem before, most likely, and could already have the solution, simple one, at hand.