Ceramic 5 / Ferrite magnets -- possible leeching

[PLukacovic]

We are using cermaic 5 magnets in liquid level sensors that are in heated deionized water, mixed with hydrogen gas. After a failure in the device, we noticed what appears to be rust around the magnets, and now we're trying to figure out if the rust (if that is what it is) came from particles in the water attracted to the magnets, or if the deionized water and/or the hydrogen is degrading the magnets and causing them to leech. We will be conducting experiments of our own, but it's nice to get input on the possibilities before we start.

I did some cursory research and found that ceramic 5 is MO*Fe2O3, with "M" being Barium, Strontium, or a combination of the two. I do not know the exact chemical makeup of the magnets we are using.

Any and all information on ceramic 5 magnets is greatly appreciated.

Thank you

 

[MJR2]

While ceramic magnets are generally considered to resist corrosion heated deionized water mixed with hydrogen gas would be rather agressive. I'd lean towards the magnets 'rusting'. What is the resistivity of your deionized water when you start?

Mike

 

[PLukacovic]

I don't know what the resistivity of the deionized water was when we started the experiment, or even if that data was taken. I do know that this failure happened past the 10,000 hour mark (how far past, I don't know), so the magnets had plenty of exposure to the degradating environment.

I was just presented some data an employee conducted a few yeas ago testing the magnets in deionized water for six months. It looks like the magnets contain Barium, and a little Strontium. Also, after the six months there is a weight loss of 0.32% at room temperature and 0.38% at 120F. A spectral analysis of the material was conducted after the experiment, but I can't find any data relating the an analysis beforehand, which seems to defeat the purpose of conducting such an analysis.
The end composition by mass is 50% BR, 29% Fe, 10% O, 9% Sr, 1% W and 0.2% Ca.

 

[MagMike]

What is the color of the corrosion product? That'll give you some clues to where the corrosion is coming from.

Iron oxide is reddish-brown. Strontium-oxide is white. I'm not sure what color Barium-oxide would be.

 

[PLukacovic]

It's definitely rust-colored. The reddish-brown material can be seen clearly around the stem of the floats on the level sensors (the magnets are housed inside the floats).

I have attached a photo showing the float and rust.

 

[EdStainless]

My guess is that the ferrite is not fully stable. Maybe an excess of one metal or another. You have two options.
1. long term thermal aging may help
2. Coat with paralene

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube