Chlorides in closed loop methanol water coolant

[gonefishn]

We have a large chiller loop (~40k gal)that distributes chilled (-40F) methanol water (60%/40%) to a large number of heat exchangers and condensers. We recently had a 316L ss condenser welded tube failure at the welded seams. Attack appears to have occured from the methanol water side. The first failure occured after 10+ years of service. The second failure occured in about 4 months. Testing revealed a level of 300-400 ppm chlorides. We do not have a good history of chloride levels in the methanol water but theorize it is likely primarily contamination from the process side (the higher pressure side) during the first failure. The high chlorides then possibly caused chloride attack resulting in the second failure. The questions are:
1. Should this level of chlorides (300-400 ppm) at this temperature (-40F)cause this expedited corrosion rate?
2. Does anyone have experience removing chlorides from low temperature methanol water? Return temp is about -30F. System is must run so any filtering process must be done at -30F and can't significantly impact the chiller system performance, i.e, supply temp.

 

[Latexman]

The temperature of interest is on and in the crevices between the tubes and tubesheets. The highest temperature on the non-coolant side is probably where the biggest issues will be. I'd guess this is above the 30 F return temperature, right?

Good luck,
Latexman

 

[gonefishn]

Thanks for your response, Latexman. Our mechanical integrity guy's inspection determined that the second failure occured at the welded tube seams from the methanol side (the cold side)presumeably from chloride attack based on presence of 350 ppm chlorides now in the methanol. Removal of the chlorides from the methanol is my primary concern at this point. Information gleaned so far is that removel with a resin bed at the subzero temps is difficult and I may have to warm up a small slipstream to do so. Will be slow due to need to not impact the operating chiller system performance. Any further comments appreciated.

 

[SJones]

Is the damage mechanism pitting or cracking? Anything about the weld or parent material quality to give an indication as to why attack should be concentrated in the weld? At -30 deg F, even with 300 ppm chloride, corrosion risk should be low.

Steve Jones
Materials & Corrosion Engineer

http://www.linkedin.com/pub/8/83b/b04

 

[StoneCold]

Just a heat exchanger comment. We have a system that runs methanol and some water at -40C. We have had heat exchanger failures. We have attributed them to stress due to low temperatures on the shell side and 100C+ temperatures on the tube side. Some of the failures were C-276 and others were tantalum (very thin tubes. 0.015 wall) We have added expansion joints to the shells of all new exchangers and some repaired exchangers and we have not had any more problems. But we don't have 350ppm chlorides. Can you get a temporary tank and put material in it untill you are down to the minimum liquid level for the system. Treat the material in the tank and then shoot it back. If it was at my facility I guess I would distill off the methanol and dispose of the water and salts. But we have a continuous waste water stream to put it in and lots of vessels where we could do a single stage flash.

Good Luck

StoneCold

 

[gonefishn]

Thanks for the input StoneCold and SJones. Because the original HX lasted many years, and the second only months,we believe that the high chlorides that developed from leakage after the first failure is likely the cause of the second failure. I am told the failure appears to be pitting corrosion in the heat affected zone presumeably from breach of the passivation layer by the chlorides. I have been advised that a strong base anion exchange (SBA) resin is likely the best filter option. But performance at the low temp and in methanol seems to be a bit of an unknownm even to the big boys in that business.

Anyone out there with experience running methanol through a SBA resin?