Alloy 20 / Acetic Acid - Method of attack? 1

[TiCl4]

In reviewing the possibility of temporarily using an alloy 20 line for 85% acetic acid service, I noticed the corrosion rate was much higher than acceptable - 400 mpy at ambient temperature. This made me curious - what is the mechanism of attack on this alloy? Is it similar across other organic acids? A brief google search revealed very little in this regard.

 

[EdStainless]

This is from the MTI MS-2 Vol. 2: Formic, Acetic and Other Organic Acids.
I have never heard a good explanation as to why higher alloy grades perform worse in these acids, but I know that very clean alloys do better so it must be linked to localized grain boundary attack.

"The use of non-molybdenum-containing stainless steels is rarely reliable in actual practice
and the molybdenum-bearing grades are always preferred for both corrosion resistance
and product purity."

"The best-known of these alloys is alloy 20Cb-3 (UNS N08020). Laboratory tests indicate
a rate of less than 1 mpy (0.025 mm/y) up to at least 50% boiling acetic acid and <5 mpy
(<0.13 mm/y) in 99% acid. However, in laboratory hot-wall tests, alloy 20Cb-3 was more
severely attacked than type 316. Also, in plant tests, the relative merits vs. type 316L are
unpredictable. Field corrosion tests showed rates of 170 mpy (4.3 mm/y) for type 316L vs.
25 mpy (0.63 mm/y) for alloy 20Cb-3 in one distillation column while the performance was
reversed in the next still."

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[TiCl4]

Thanks, Ed. So I understand from the reference you provided that higher alloy stainless grades with moly are better suited to organic acid than, say, 304 grades of stainless. However, there does not seem to be a clear understanding of the mechanism of corrosion, and field results can vary widely based on service conditions.

If implemented, what would your recommended frequency of checks be to gauge initial corrosion rate in a pipe? Every month for the first couple months of service to gauge if fast corrosion is occurring, then set inspection rate based on initial data?

 

[EdStainless]

It is interesting, the highly alloyed SS grades (6%Mo) are no better than 316L in these acids.
You need to think about this, to start with you need to mark locations and take baseline readings either before or very early in service. The thickness variations will be large enough that future readings will rely on these baselines.
Since you could expect corrosion rates of hundreds of MPY setting the test interval of 3-4 months until you get a clear picture is a good idea. Be careful what locations you select as you may be much higher loss rates in some locations (near welds, elbows, outlet of valves).

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

"The use of non-molybdenum-containing stainless steels is rarely reliable in actual practice
and the molybdenum-bearing grades are always preferred for both corrosion resistance
and product purity."

This suggests that pitting is the mode of attack.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[EdStainless]

IM, sort of, the trick is that higher Mo grades aren't much of an improvement over 316.
I have also done some work in Sulfuric where we got the best results with very clean 310.
Cr is your friend if you can prevent ANY carbide formation.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

You know this stuff better than I do, Ed, but there are situations where more Mo does make a difference. Resistance to naphthenic acid corrosion seems to require SS with a minimum of 3% Mo.


"Everyone is entitled to their own opinions, but they are not entitled to their own facts."