Embrittlement of Duplex Stainless Steel (UNS 31803) 1

[sanshu1111]

One of our facility is operating under thermal cycle with temperature raging from 60 Dec C to 260 Dec C. After operating it for appx 5 years, we found some cracks on our piping system (flange and flange to elbow weld joint/ HAZ) which is in DSS UNS 31803 construction.

We send this piping material for the testing and found formation of intermetallic phase (singma/ chi phases) which resulted into brittle failure. Tensile properties of the material were meeting the requirements but elongation was only appx 6% and impact values are appx 2 jouls only.

Many codes and literatures limit use of DSS at appx 300 Deg C, API TR 983 C raise a concern that prolonged exposure at above 260 Deg C, depending on DSS grade may initiate embrittlement. It would have been easier to conclude the same as we are operating very close to 260 Deg C but we do not find any material degradation on the elbow and piping which exhibit very good impact properties/ elongation and no internetallic phase formation. This degradation is noticed only in flange section of the piping.

We have already checked mill test certificates and as per available MTC's, flange material was good with good impact properties and no intermetallic phase formation. Now one possibility is that the data reported in MTC's were falsified and there was some problem with the flange heat treatment during manufacturing. It may be worth highlighting here that all the flanges are from the same heat and have cracks.

Is it also possible that some wrong fabrication process/ heat treatment can provide correct initial properties but can lead to material degradation after some time?

 

[davefitz]

I am not sure of the geometry of the flange, but if there is cycling service with fast temperature changes then it may be recommended that the flange be a forged weld neck flange such that welding zone is removed from the flange face by at least 1 pipe diameter, or 5 wall thicknesses. Fast temperature transients will lead to very high thermal stresses near the root of the flange face, and if the weld zone is in that vicinity then the HAZ will be exposed to severe thermal stresses and a fatigue crack may form.

"Nobody expects the Spanish Inquisition!"

 

[metengr]

Yes, it is very possible there was no solution treatment performed or was incorrect.

 

[EdStainless]

My first bet is that that the original material was not correct.
You can anneal this material very quickly, and get rid of all apparent secondary phases.
Your properties (elong and impact) will be fine. But if the anneal hold times are too short you will leave a lot of remaining segregation in the structure. What this means is that the re-formation of secondary phases will happen much more easily (at lower temp and faster) than in thoroughly annealed material.
There is also the question of weather any of this material ever got any hotter in service than 260C? Even just for a few minutes during start up or shut down? This damage is cumulative and over the years it would add up.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[sanshu1111]

Impact values reported in MTC are in the range of 150 joule, is it possible to get so high values with no or incorrect heat treatment of flanges?

Normal operating temperature for the heating cycle (heating cycle time duration is appx 1/ 4 of total cycle time) is appx 250 - 255 Deg C and there is a possibility that it reaches upto appx 265/ 270 Deg C for short duration. Is the embrittlement phenomenon for DSS happens only above 260 Deg C or it is possible even at a lower temperature depending on exposure time? I notice that ASME Sec VIII Div 2 limit use of UNS 31803 for a maximum temperature of 204 Deg C, what is the reason for the same when Div 1 and B 31.3 limit it to 316 Deg C?

We did not find any material degradation in other piping components, is it possible when we are operating very close to 260 Deg C, a temperature limit for which a flag is raised in API TR 983C?

 

[SJones]

Reported Charpy energy values in inspection documents may bear no resemblance to those of the actual products if the supplier has been allowed free rein.

NACE Corrosion 2012, Paper 2012-1096
NACE Corrosion 2014, Paper 2014-3651
EEMUA 218
NORSOK M-650
ISO 17781
ISO 17782

Steve Jones
Corrosion Management Consultant

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

All answers are personal opinions only and are in no way connected with any employer.

 

[EdStainless]

If they dissolved all of the secondary phases in the original anneal, but still had a lot of segregation they would pass impacts fine.
When we were validating anneals we took samples and aged them at 285C for various times and then tested (electrochemical not impact). We could clearly see that the under-annealed samples began the embrittlement process much sooner and faster than predicted.

There is not temperature where there is no damage, just that it may take a very long time. Depending on chemistry and thermomechanical history you could be picking up some damage at 260C.
But my guess is that it is because the material was under annealed in the first place.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[sanshu1111]

Thanks for all your responses. Any idea about ASME Sec VIII Div 2 limit of 204 Deg C for UNS 31803 whereas limits in Div 1 and B31.3 are 316 Deg C.

With best wishes for Christmas and a Very happy New Year.