Crevice corrosion in 316L threaded fitting

[isakleivsson]

Hi,
We har experiencing crevice corrosion on some threaded fittings, so looking for any guidance in troubleshooting and solving the problem. New here, not sure whether to add a bunch of photos along with this or not.

The equipment is 316L thin wall (2-3mm) piping that has 316L threaded check-valves connected on a 1.5" BSPT nipple, operated at 10barG. The threads are sealed with a combination of Teflon tape and Loctite 577. The piping is the distribution of a cleaning system that delivers sea water, detergent and desinfectant (see specs below) to various nozzles on some equipment that is placed on deck of fish carrying ships. This is in use a few hours a day intermittently, and detergent and ozone are for short periods in these cycles. The pitting is significant, but not yet through the thicker wall of the check valve. We became aware of this as some piping was dissasembled due to improper weld procedure that also caused pitting on the boundary of internal "sugaring" and caused a leak. It has been in service for about a year, but we do not know the number of washing cycles. On average probably used once a week.

The crevice corrosion is occuring on a number (not all) of the check valves, about 1/2 thread from the tip of the nipple (probably at the end of the starting thread chamfer). It is not at the bottom (with respect to gravity), the nipples are not welded on in a specific rotation with the respect to the threads. None of the nipples have any pitting on this or other parts of their thread. The orientation of the check valve is horizontal, and there is inner geometry that catches some liquid regardless of draining procedure (height difference in male/female threads, check valve seat etc). The piping is acid passivated, but the check valve does not appear to be (at least not in the same way, it has shiny threads). There is a few threaded adapters with pitting on the 60deg cone hose connections that are after the check valve.

Detergent (others use similar ones)
- pH 13 NaOH mixture. Safety sheet says <1% NaOH in the concentrate. mixed 2-3% in seawater to be 11 pH before it enters our piping.

Desinfection
- Mix of Hydrogen Peroxide (20-25%), Acetic acid (10-15%) and peracetic acid (~5%). Mixed 1% in seawater before it enters our piping.

The system runs a spray cycle (seawater), a detergent cycle, a rinse cycle (seawater), desinfection cycle and a final rinse cycle (seawater). Some likely deviate from procedure and do not use the final seawater rinse. The seawater is filter and UV-treated, but according to some accounts a bacteria test comes back positive if the final seawater rinse is performed.

The questions I have:
- Is the seawater rinse cycle critical? Seems from other forum posts seawater (especially stagnant) has no problem corroding 316L in crevices like these.
- Is it possible to reason to or to test whether the crevice corrosion is happening during the washing cycle or in the stagnant, after/between cycle?
- Will orienting this in a less horizontal way that drains more of the water drastically change the outcome, or is the amount of water/liquid that is going to be "stuck" in the threads going to do this regardless?
- Cathodic protection. Will installing a sacrificial cathode in the piping system have any effect on this? It will only be possible to add one for the bulk of the piping, so once it is drained wont each remnant liquid be unaffected? There are zinc anodes in other parts of our equipment, which are filled with seawater during use, but are not in any way in a liquid connection with this pipe system.
- Would passivating the check valve have a significant effect.
- Would a different joint type be less prone? I.e. if the check valve is flanged with gasket.

Any other tips/input is appreciated. Seems a "sealing" the threads option is possible, but from other forum posts also not guaranteed in any way.

 

[TugboatEng]

Chlorides are trouble with 316 SS. If you're draining the pipe the remaining water with it's low chloride concentration will become super concentrated as the water evaporates and attack the SS. Consider blowing hot air through the piping after flushing for rapid drying.

Threaded connection will always be susceptible to crevice corrosion. Sealants can't solve that as there will be crevice corrosion at the toe of the sealant. BUT, you can reduce this effect by using ONLY the 577. You need to trust in your sealant and not come up with your own concoctions. Also PTFE tape is one of the worst for causing crevice corrosion. I don't have specific experience about the suitability of 577 for acids but I do know that that it works very well in seawater allocations.

It's good to think of anodic protection as being line of sight, while this may not be exactly true, anodes don't protect well around corners. Piping has too many corners.

 

[Dhurjati Sen]

1. As sea-water rinse cycle given in procedure, no point in deviating from it.
2. There is a basic difference between crevice and pitting corrosion. As per you, since it is crevice corrosion, so removing the crevice(s) is primary. Once there is a crevice, corrosion would occur, maybe at different rates during different cycles of operation, but this is not so relevant.
3. Refer point no. 2, removal of crevice(s) is primary.
4, 5 & 6. Painting is more appropriate.

 

[Compositepro]

Sea water is not sterile, so why would you rinse with it? If you are using UV to sterilize the rinse water, your testing is showing that not to be effective.

 

[EdStainless]

Are the valves made from castings or wrought material?
If castings they are likely not annealed properly.
You really should go to an all welded system.
And at that point also go to a better alloy.
the most cost effective option is likely 2205

 

[MintJulep]

As noted by @TugboatEng, 316 isn't a good choice where there are chlorides. Apparently the combination of chlorides + acid increases the susceptibility to crevice corrosion.

The alternating exposures to acid and basic solutions might be mucking up the passive layer by selectively leaching out Cr and Mo, increasing Fe in what's left behind. This is apparently temperature dependent.

 

[EdStainless]

The fact that this line is not used consciously makes things worse.
As it it sits some water will evaporate and this will concentrate the impurities.