Stainless Steel Weld Corrosion Problem

Brine being basicaly a salt solution will cause an electrolytic action between the different metals disolving one in favour of the other. The reaction will be speeded up by an increase in temperature or an increase in salt in the solution. You could drain down rince out and reweld then cover the affected weld area with a good two part epoxy paint or the blue acid resistant paint used to coat containers containing things like sulphuric acid. A physical barrier method is the best failing this line the tank with good quality plastic sheeting. a Glass lining is the ultimate. In a nut shell the salt is disolving the welds out. If you can cover them all will be well. All the best. PS your spelling is not that bad and if you click the ABC button it will activatre an automatic spell check this highlights the errors and gives the correct spelling or at least lets you know something is up.

I had something similar a while back on a mobile repair job. I can't remember what they had in their tanks but all I had was E309 (TIG process). It seemed to work, maybe because of the higher Cr and Ni content or because of the welding process?

I think you had better switch over to SS 316L for tanks instead of 304, with correspondingly suitable elctrodes.

What you are experiencing is known as inter-granular corrosion. In a nutshell, chromium in the grains prevents corrosion by producing a thin oxide layer on the surface. Heat treating of stainless is a tricky process, certain temperature ranges are avoided as the chromium will precipitate out of the grain as a carbide. This robs the grain of its normal chromium content, rendering it less resistant to corrosion.

Try a forced water quench immediately after welding, this will decrease the time the steel is in the critical temperature range and fewer chromium carbides will precipitate out.

merkelerk, Agree, this sounds like inter-granular corrosion. Sensitization - chromium carbide precipitation - occurs in the range roughly 480 to 870 deg C if memory serves. I guess using 304L would help, but not practical to rebuild the tanks. Would smaller and shorter weld runs be sufficient to reduce the time in this temperature range? Passivation is always a good idea but would it have any effect in this case?

An earlier post mentioned that TIG welding produced no inter-granular corrosion. This is likely because of the quenching effect of the surrounding steel. As in induction hardenening, an actual water or oil quench is often not required. The base metal, having a much greater mass, acts as a heat sink to the very small amount of metal heated by the TIG torch and virtually quenches the weld as soon as the torch moves on.

Sensitization or sensitizing is the correct term and the temperature range (according to my books) is 480 - 760 C.

Stainless that has been sensitized can be re-heated to 1037 - 1093 Celsius and held at that temp. for a time (enough time for the carbides to dissolve and diffuse back into the ferrite/austenite grains) and then immediately quenched in water. Keep in mind that prolonged exposure at high temperatures will cause grain growth that will adversely affect toughness and strength.

Passivation would not likely help in this case (and may have been done by the supplier), I would have to check a little deeper, but I think that the chromium carbides would be too stable to be reacted with nitric acid.

If stick welding is a must, then as you mentioned, short and prefferably quenched welds would help.

Just a suggestion - is it possible to cover the welds with some high-performance coating, epoxy or something? After griblasting etc. There's plenty on market, but obviously it depends on tank details.

Cheers.....Codey

Thanks for all the great info. I have tried a belazona epoxy type coating on 2 of these tanks. Although I haven't been able to get back into the tanks to inspect these 2 have started to leak as well. I am assuming that the coating has failed. We have always had good results with belazona products and I really thought this would be the answer. I have done some research on the reaction tank materials and found the agitator as well as some of the other materials were made from carpenter 20 (Hastelloy?) This probably explains why allot of the original tanks and stacks were made from Cyprus wood. I also found that the piping we use to pump the brine from the reaction tank (316L sch. 40) is also corroding away in the weld seams. The mill is not going to spring for the cost of Hastelloy tanks and piping because the ROI time is long. I am looking into going back to carbon steel materials and add allot of corrosion allowance. It seems this was the best answer back in the 50's when the system was designed. The weld quenching idea is interesting and I may try this as it would be easy enough to perform. I am still not sure what type electrode would be best to weld with before I quench .

pipewelder

Please read my earlier post before you pursue the heavy carbon steel approach. Fixing your post weld treatment to restore the corrosion resistance will be far more successful than converting over to carbon steel. Converting to carbon steel is an environmental safety disaster in the making.

Regarding the other pipe corroding near its seams, well of course, it s going to fail somewhere and sometime in this application, and the seam area is the highest concentration of material differences. Schedule replacement before 50% of wall is diminished.

milo

Thanks Milo and everyone else too, I will try the water quench and the pickle and passivation. On the side I own a small piping fabrication shop where we do alot of 304 and 316 TIG welding specialty pipe fabrication. I always passivate the welds in my shop for as much appearance as anything. Do you think going to a E-316L or 312 type of electrode would be any help or just stay with the 308L for the re-weld?

thanks, pipewelder

A few years back there was a coating called Flakeline from Ceilcote co. If it's still going it might work, or possibly they can offer something that does.

Cheers..........Codey

A lot of good comments and thoughts on this one!

A good 3 part coating system with primer and 2 part epoxy after repair and aggressive cleaning of the welded area would be cheapest way to rectify.

Always look to the possibility of contamination not removed after the weld.

Great question and Great responses.

If you were to look closely, my experience tells me that the corrosion is not the weld itself, but in fact starts at the Heat Affected Zone immediately adjacent to the weld.

This heat affected zone is 'different' from both the weld zone and the base metal by the fact that it contains surface defects and contaminations, such as welding oxide, heat tint, microslag inclusions and scale formed during fabrication. Not to mention the formation of chromium depleted layer.

Left untreated, these will reduce the corrosion resistance of the stainless. A galvanic cell will form, and by my reading of the chemicals in use, the stainless will certainly be vigorously attacked.

In order to give the weldment the best possible chances for this environment, a post weld treatment by acid pickling will remove the welding oxides,surface exposed slag, and the chromium depleted layer. after /during pickling, application of mechanical means such as brushing and hi-pressure jets will help restore the surface integrity.

From the facts that you've given, you've got $2000 welds and 15 cent post weld treatments.

Please look at ASTM A 380 and all will be made clear to you.

http://engineers.ihs.com/document/abstract/FODZHBAAAAAAAAAA

By the way, you did a great job of assuring that we had sufficent facts to understand your problem!

Hope you find this posting helpful.

milo

Milo:

I think you are probably right and if one were to examine the microstructure of the fracture, it would likely be intergranular (follow the grain boundries) as opposed to transgranular (ductile grain shear) due to carbide precipitation in the adjacent heat affected zone. That's the primary reason for specifying L (low carbon) modified stainless steels for welded structures.

One caution on acid pickling. Have to be careful to avoid hydrogen embrittlement due to absorption of nascent hydrogen from the pickling solution. That would only compound the problem and almost guarantee brittle fracture. Inhibitors can be added to the pickling bath or subsequent bakeout can be employed.

Good morning Bluestone.

If they were going to immerse the weldment in a tank for a long time that might be a concern, but I was speaking of commercial gels and pastes. The amount of time is insufficient to diffuse hydrogen into the material from a surface treatment applied as I described. Improper electrodes on the other hand could indeed load up the weld metal with Hydrogen, a very different story, as you pointed out.

I guess that your comments are instructive that it does need to be stated when one posts that they are describing a commercial pickling practice with proper attention paid to bath, inhibitors, etc., applied using best commercial means, rather than just dip and hold in raw acid.

The real lesson is that rather than just blame the material, we should always look at our PROCESSES. These are usually culpable as the facts in this case seem to suggest.

milo

Your welding rod is incompatible with the chlorides in the brine. Most stainless steels are not resistant to chlorides. HCl and Sod. Hypochlorite are examples of chloride materials that will eat your welds quickly. Is it possible to line your tanks with a high temp fiberglass or switch to fiberglass altogether? There are chemicals that you may be able to add to the solution that can help with the corrosion. Contact your local water treatment company. One would be Superior Water Services in Atlanta.

In 1984, I had faced a problem like that where the corrosion propagated in the weld itself in a stainless steel piping system. I discovered that the pipe fitters and welders used a brush for cleaning which was used before in carbon steel weldment. The brush was contaminated with fine carbon steel particles, in turn causes a dissimilar materials inside the weldment, and causes corrosion for the weldment.

The corrosion is caused by the heat from welding. The chrome in stainless got burnt off.

Stainless steel's corrosion resistant is based on the chrome at surface of the metal. Under normal condition that layer is self healing. Heat from welding removed the chrome from the surface. Depends on how much the material was heated, thickness of the chromless layer vary. Only way to fix it is to remove that layer and expose the metal. Try sand/grind down the welded surface. Sand blasting may be better.

I had frame cut 304SS plates which need to machine 1/32" off to prevent it from corroding.

Pineapple

BRINE KILLED seam welded with a E-308, SMAW .But the plates remained intact.

Took 2 days to respond. Studied some serious research papers in between.

All data/theory point to one course of corrective action:

Weld without any flux . Means ,MIG/MAG with 304L filler wire. Problems will end .

Before attempting this remedy you will probably like to ask Consultants like Hendrix:

info@hghouston.com

In due time please tell us at CR4 ---of improvements.

You could clean with sand blast then electro plate with nickel. Then for belt and braces go with the two part epoxy paint. Polish with a product called MER this resists salt very well, see you local auto parts store.

Wrong filler metal; no coating will help; heat treat will not cure problem; 304 is austinitic microstructure; 308 weld is dendritic; problems arise in weld due to delta ferrite content, chrome carbide precipitation, among other things; 316 base material and 316l weld is much more resistant to chloride attack; 23yrs as metallurgist doing failure analysis have seen this before. If you cannot change base metal on existing tank you must use correct filler metal in weld to resist chloride attack.

erssk aws-cwi

So good! An Expert's Recap.

I am no metallurgist-but agree with you 100%.

MIG/MAG+NO FLUX+304 FILLER WIRE

310;310cb;310mo;330 filler metal will typically deposit a fully austinitric microstructure; the 304 filler is being eaten away remember

erssk aws-cwi

Thanks erssk, I will try the 310 filler metal. I already have a procedure and qualified weldors for this type of welding electrode and keep a suppply of rods on hand. We currently weld the tank with 308 electrodes, not 304. I will also passivate after welding.

pipewelder

Free iron is a big problem with stainless, and aluminum (just set an aluminum soda can on a 316SS flat surface); underdiluted Clorox used for algae abatement in clear water rinse tanks can initiate corrosion also, and I've heard plain old table salt sprinkled on stainless supposedly can cause problems. As you may have guessed I am not a chemist nor a metallurgist. Never had one problem with weld corrosion except the above exceptions. My application was photographic processing, color film and paper, also black and white. No problems with 304 or 316 stainless. Not sure if regular cleansing and wiping down/airing out with your tanks is being done to allow the surface layer to re-establish itself, or if that's even an issue.

Next time build your tanks out of 316L marine grade ss for holding brine.

The only long term solution alternative I have experience with is employing a high quality reinforced PVC membrane like RenoSys or Delifol.

sjm