% Ferrite a corrosion factor for 316L Stainless Steel???

[JodiSteel]

I have a vendor that is requesting a deviation from a drawing spec for 316L stainless steel that requires 0% ferrite on a Schaeffler diagram. The deviation has lower nickel which would put us at 5% ferrite. The application is a structural "arm" in a molten bath of 55 %Al - 43.5 %Zn - 1.5 %Si at 1100 F. My concern is accelerated corrosion. Can anybody explain the impact on corrosion properies of increasing % ferrite from 0 to 5?

Thanks in advance!

 

[unclesyd]

With a cursory scan on several papers on Ferrite in SS and corrosion problems there is no mention of any case that would pertain to your enviroment.
My data shows corrosion rates of SS vs Aluminum varying all over the map. I would think the Zinc wouldn't help. I try to dig deeper into the stack.

Evidently the requirement is based on previous performance data where the rate experienced was equated to the magnetic properties. This could lead to ferrite is bad so none is a lot better.

The difference between 0%-5% is really a call of none vs some. That is small difference in measuring ferrite in anything but plate products.

Has the plate been tested?
Was the plate quenched annealed?

 

[metengr]

This post got me thinking about the potential for liquid metal embrittlement of the 316L SS with molten zinc. Some additional information related to this subject – “ the 18-8 stainless steels are generally attacked by molten aluminum, zinc, antimony, bismuth, cadmium and tin”.

The above information was taken directly from ASM Metals Handbook, Desk Edition, page 386. I would not recommend using the 18-8 stainless for this application.

 

[metengr]

Additional reference material - section 18-14 of the 'Handbook of Stainless Steels' by Peckner and Bernstein, and I quote - Austenitic and ferritic stainless steels have limited to poor corrosion resistance to tin, zinc, aluminum and cadmium at their melting points and poor resistance above.

 

[arunmrao]

My experience in making melting pots suggests that Stainless steels are not suited. Though some users continue to insist on using them. Plain gray irons perform better.

Electrodes in salt bath is another example. The high Nickel electrodes perform poorer compared to wrought irons.

 

[unclesyd]

JodiSteel,
Have they used this material in this particular service before?

metengr,

Here is a reference that I have on two tables of general values of corrosion resistance of SS versus
pure metals (liquid white metals). I don’t have any further access to any references such as this one.

Aluminum All types of SS Embrittlement penetration at melting point 1220°F

Zinc All types of SS Poor @ 930°F Embrittlement above about 1372°F

I have another reference that states essentially 1200°F is the starting point of all thing bad that happen to SS in liquid white metals. The kicker in this application is that it is an alloy with "silicon".
Another general statement is the importance of the surface area to liquid volume ratio due to varoius solubilities in the liquid metal.

2:102-124. Jackson, CB, ed., Liquid Metals Handbook and Supplements (US Atomic Energy Commission, US Government Printing Office, Washington, DC, 1954, 1955). ...

WD Manly “Fundamentals of Liquid Metal Corrosion” Corrosion 12 pp 336t (1956)

There is also mention of several reports from Argonne National labs.

 

[PAN]

The penetration of zinc could be started at the melting point (~790 F).

In your application (1100 F), I will not recommend to use austenitic stainless steel.

 

[EdStainless]

I have been the SS/zinc route, it is bad news. We ended up using a high Si Cast iron.

As to ferrite content. In a plate, sheet or bar product the residual ferrite should be low, say less than 0.5%. It doesn't matter what the equalibrium value is as long as it was annealed correctly. Where you have issues is with unannealed welds. They can be near the equalibrium value. The application that I know of that this is a real problem are all either acidic or ultra high purity.

= = = = = = = = = = = = = = = = = = = =
Corrosion never sleeps, but it can be managed.

http://www.trenttube.com/Trent/tech_form.htm

 

[JodiSteel]

Wow, thanks for all the posts!

In response to some of the veins people are taking, the 316L has been used routinely for 10 years in this application. It’s not perfect, you do lose mass over time but you do get several years of service. Basically, this is a hot dip galvalume operation and the 316L is the alloy used for the pot equipment. We are continuously hot dipping low carbon steel strip approximately .019"x42" in cross-section. I know that low carbon steel would be a terrible application for the pot equipment because the strip will dissolve in the bath in a matter of 20 minutes if the line stops unexpectedly!

There are no welds on the arm. The plate is bought as stock which has been annealed to 2000 F with rapid quench. It’s then machined to desired shape.

So I am afraid I am still left with the same question, is going from 0%-5% ferrite going to increase corrosion attack?

 

[metengr]

If you are willing to put up with and factored external corrosion in your design, the 5% volume of ferrite will probably not make a difference. Some of the ferrite will probably revert to sigma phase upon continuous exposure to 1100 deg F, over time, but in your situation it makes no significant difference.

 

[EdStainless]

If it is annealed and quenched it should be less than 0.5% residual ferrite. The equalibruim number (from the SD) may be higher. With Ni prices high today it isn't uncommon to see calculated FNs of 10-14.

= = = = = = = = = = = = = = = = = = = =
Corrosion never sleeps, but it can be managed.

http://www.trenttube.com/Trent/tech_form.htm

 

[JodiSteel]

EdStainless, I need a tech teach on "calculated FN". Is that the Nickel Equivalent?