Effect of Surface finish on Corrosion Resistance of SS316L 2

[bssem]

We are OEM suppliers of instrumentation. There has been a debate on the effect of surface finish on the corrosion resistance of a given material (SS 316L) in given corrosive environment (acids or high chlorides).
Can someone pl. guide me if any such studies have been done to prove that - an as cast part is less resistant to corrosion as comparted to a cast part followed by finish machining. In other words does Surface finish effect Corrosion Resistance? If so by how much?
cheers,
bssem

 

[TVP]

The answer is complicated by the method used to obtain the surface finish. In general, smoother is better, however, if machining and polishing operations expose sulfide inclusions to the surface, resistance to pitting in chloride environments is reduced. mcguire has posted a number of times about this phenomenon.

 

[bssem]

Thanks TVP,
Could you post the other mcguire thread numbers so I can catch-up with you?

 

[EdStainless]

That is why a combination of mechanical and chemical cleaning is used. After the mechanical finishing you have a number of worries; these include smeared metal on the surface forming crevices, freshly exposed inclusions, and imbeded material in the surface.
The mechanical prep should be followed with a passivation, at least. I would prefer a pickle after myself.

Of course, I assume that the castings are well annealed to start with. If they have any measurable magnetism that will be an issue also.

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

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

 

[mcguire]

bssem
There is some data published by S.R.Collins of Swagelok who shows that an a TIG weld, which approximates a casting, has a critical pitting temperature (ASTM G150) of about 10C. If it is electropolished, it increases to about 20C. If it is mechanically polished, it decreases to 0C or less. The test is invalid below about 5C.
I have seen unpublished data which shows that a bright anneal finish is about as good as electropolished, which in turn is close to annealed and pickled. When you mechanically polish you introduce residual stress and expose de-chromed regions, both of which lower pitting resistance.

 

[moltenmetal]

mcguire: do you recommending welding 316/L with different filler metal to increase the critical pitting temperature, regardless of the surface condition? If so, which one?

 

[EdStainless]

If you must use as-welded material, using over-alloyed fillers is a good practice. It will not help the corrosion resistance of the HAZ in the base metal, that still depends on your weld/shielding practices.
There are standard over-alloyed fillers. For 316L you can use a rich 316L composition or use the 317L.
There was a paper at NACE this year, 04291, that describes surface finish vs corrosion resistance for AL-6XN.

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

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

 

[mcguire]

EdStainless is right. You can improve the weld's corrosion resistance but the base metal which just reaches melting temerature is still the weak link. If you cannot anneal the weld, then good pickling is the next best thing.

 

[moltenmetal]

mcguire: I thought your "the weld is basically the same as cast" comment was related to the fact that the 316/L filler metal is designed to give you a little more ferrite to avoid problems with cracking (i.e. like in the cast equivalents), and hence is a less corrosion-resistant material. The metal in the parent metal pool and the HAZ is just recrystallized native 316/L, which if you do a good job of pre-weld cleanliness, shielding and post-weld pickling/passivation, should be as resistant as the parent pipe or fitting. Is that correct, or does the "as welded" condition represent residual stresses which make the parent metal and HAZ less resistant regardless what you do?

Thanks again for your help, and pardon my ignorance, but this is something you probably know off the top of your head whereas it'd take me hours of digging up references and reading to get the same info...

 

[EdStainless]

The problem with the welds and casting is microsegregation. As the material solidifies hte first to freeze is a slightly different composition from the last. Notice the dendritic structure of unannealed welds and castings. These variations in composition have different corrosion resistances. Some of htem will be below the bulk, average' alloy in pitting resistance and failure will start there.
Keeping the welds small, clean and pickling afterwards will get you as close as possible in corrosion resistance.

The ferrite issue is part of it also. Yes, you like welding materials with slight reatined delta ferrite levels. Welding 904L or AL-6XN is difficult because they have zero FN. But just as 4-7% ferrite helps prevent weld cracking the ferrite is also the weak link in resistance to corrosion in acidic environments. In a good anneal the ferrite is gone (<0.5%) and the chemistry is homogenized.

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

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

 

[mcguire]

A casting or a weld is as-solidified metal. As such it is not at thermodynamic equilibrium and contains phases and inhomogeneous regions which would not exist after a long anneal or some mechanical working and a long anneal.
These regions are sometimes lower in elements which supply the corrosion resistance. These are chromium, molybdenum, and nitrogen. There are two schools of thought on which regions are the most detrimental, delta ferrite or the zones around inclusions. That doesn't matter to your question. The parent metal which has melted and resolidified has reduced corrosion resistance from this process of remelting. Pickling removes the troublesome regions from the surface which takes away the easy nucleation sites for pitting. Annealing negates them by making them more homogeneous. Residual stress is a factor only if they are locally high as they are from abrasion of welds. I have lots of references on these points if you need them.

 

[unclesyd]

Surface finish influence on corrosion is very dependent on the specific corrodent and enviroment it's exposed to. Real world conditions seldom meet laboratory testing expectations unless the specific corrodent under actual conditions are used for testing. We never found pitting resistance a real measure or indication of actual service exposure corrosion rates.

Case in point 316/316L exposure to molten Organic Acid (100% Adipic) at 190? C corroded at a rate of <.001 IPY if left in the as welded, as received state. Any grinding or attempt to polish or clean the surface of the plate, pipes, or welds caused a local increase of several fold in local rates. We verified this both by laboratory tests and field tests of different coupons. We have another organic acid that has essentially the opposite effect. It will take a ground surface and polish it in a very short time. In service coupon testing also shows this effect. A ground sample of 316L corrodes at .015 IPY to .020 IPY initially then goes to nil. The corrosion rate of either a mechanically or chemically polished coupon exposed to same material has a corrosion rate of nil.

We have to the case of 304ELC (.015C) exposed to HNO3 + Organics @ 120?C where again we have preferential attack on any areas that are ground. We have in our spec that all welds be left as is, no grinding or cleaning what so ever. If the weld metal (SMAW) is left as is will delay the onset of corrosion to same. Even cursory grinding will cause an immediate attack. If you find the onset os weld decay it can be mitigated by making a fusion pass of the affect area. This service has a pronounced tendency to cause end grain attack, like on the internal projection of a nozzle end or a fastener (bar), or plate. This can be mitigated by making a fusion or weld pass on the exposed end. SMAW works best.

We ran hundreds of Heat Flux tests using unpolished coupons or polished coupons, to better emulate heat exchanger tubing, and very seldom saw any effect on the corrosion rates. The only effect noted but not measured was the bubble formation on the polished surface.

If I were bessem I would leave cast parts and welds alone unless it is absolutely necessary for functionality of the instrument.

 

[mcguire]

Good information from unclesyd,as he usually has. My experience is confined to chlorides and atmospheric corrosion, which are the most common corrodants for stainless unless you in the chemical/petroleum industry. It sure seems certain, based on his experience, that you must prepare for the right threat,because leaving a weld un-treated would be a very bad thing for marine or atmospheric applicaions.

 

[bssem]

Thank-you ALL for your invaluable contributions
cheers,
bssem