Liquid metal assisted cracking (T' Al...!)

[Welderpaul]

Al... Come across Liquid Metal Assisted Cracking (LMAC) mentioned in a spec for an upcoming job. Not heard of it before, so found a bit of simple text on it...

If steel is subject to tension, or shear stress, and if various hot liquid metals (zinc, lead) are in contact with it, and if they can enter into the surface of the steel, the liquid metal can infiltrate into the steel at a molecular level. Having molten zinc atoms between grains of the steel reduces the tensile strength to zero at that location.
If the stresses in the steel are high, a crack can form. If this crack reduces the stress, the minute crack can stop at birth, but if the crack results in an increase in stress at the tip of the crack, the size of the crack will extend. In effect, the zinc or lead can be a knife so sharp that it is one atom wide at its edge. Unlike a knife through butter, the liquid metal "knife" does not need pressure to cut: what it does need is tension in the steel; and an engine to maintain the tension in the steel by continuing to pull the faces of the cut apart, allowing the liquid metal to penetrate. When the tension disappears, the crack stops. While sufficient tension remains, the knife slices in. A crack 3/4 of the depth of a section can form in a second.


In the spec i have, it seems to be linked with galvanising. From what i understand, for it to happen, it would be inherent stresses within the metal from fabrication/rolling/shearing etc that crack when subjected to galvanising rather than cracking in service. Is this correct?

 

[gordon stephenson]

It does not just happen with liquid metal, I had to overhaul a gas driven expander (bit like a huge turbo charger running at 27,000rpm) Casing needed a 1" AF Allen key to undo the bolts,Around 80% of the Allen bolt heads sheared straight off from the shank with very little effort,They just snapped like carrots. Our engineering boffins said the process gasses in the atmosphere had made the allen bolts under great tension glass hard and brittle.

 

[hotrodder]

In the spec i have, it seems to be linked with galvanising. From what i understand, for it to happen, it would be inherent stresses within the metal from fabrication/rolling/shearing etc that crack when subjected to galvanising rather than cracking in service. Is this correct?

Don't much about LME (liquid metal embrittlement) but the above is how i understand it although I thought it was also only a bigger deal/real issue with higher strength steels which are more prone to hydrogen cracking/embrittlement. I've read of LME being linked with the brazing of higher strength steels too, specifically brazing 4130 although i don't trust the source (it was book that although generally good also contained other questionable info)

This pdf covers the 'autopsy' of a galvanised sign bridge where they concluded that overlength diagonal braces had been forced into position which resulted in enough stress to cause cracking during galvanising. The zinc then filled these cracks hiding them... http://www.metassoc.com/pdf/MAI_Minutes-6_04.pdf

 

[TechnicAl]

If its a tank that will hold the molten zinc for galvanising you need to use special low Silicon steel. Also the welds need to be low Silicon but not too low or you will get porosity. There are some stick electrodes specifically designed for welding low silicon steel. Losilmat from Esab and Metrodes Nilsil are two there may be others. They weld like s**t but its a means to an end.

Alternatively if the fabrication is normalised after welding it will virtually eliminate the problem. The grains will recrystalise and smaller grains will increase the path of least resistance.

What happens (simplified) is the molten zinc reacts with the silicon in the steel and basically eats its way through. Lead doesnt do it, but when refining Lead they add Zinc to remove any Silver that may be present (there usually is a small amount).

If you are Galvanising you can take advantage of this reaction by using a high silicon steel so that the Zinc reacts and you get a thicker layer. It only happens whilst the zinc is molten. This is why you can see welds after galvanising, the weld often has higher silicon that the plate.

What are you fabbing?

 

[Welderpaul]

Thanks fellas, that is a good link, thanks HR.
We are fabbing a large gantry walkway that houses a small building in the middle. lots of cantilever members, pinned joints and nearly all from circular section.
It was mentioned in the structural engineering spec and i had never come across it before.
Is it a common occurance?

 

[TechnicAl]

No. As I said it really only occurs when the zinc is molten so galvanising baths and lead refining kettles.
I cant imagine why its mentioned for fabbing a gantry. Sounds like someone has seen something and over-reacted.

 

[Welderpaul]

Thanks fellas, learn something new everyday eh