Suggested Lab Tests for 316L & Chlorides

[steris]

Hello,

My company has an application where we use welded 316L plate at elevated temperatures (250F-300F) in a steam environment with chlorides present. Historically, we have seen issues with pitting, SCC and generalized chloride corrosion in the heat affected zone of the welds. We are redesigning these welded joints and want empirical data to help choose the best weld configuration. For example, we are experimenting with different filler metals, heat input, shielding gases, etc. and we like to see what combination provides the most corrosion resistant end product.

Sorry if this is a newbie question, but I've never had to deal with this topic before. Does anyone have any suggestions of specific accelerated corrosion tests that would be appropriate for this application?

 

[EdStainless]

Which failure modes do you want to look at?
Do you look at microstructures of the HAZ? Do you use quantitative metallography to measure carbide formation? Your weld procedure should produce no detectable Cr carbides, if it does work on the procedure.
Your shielding should be good enough that there is no more than a light straw tint on the HAZ (and no color anywhere else).
Overalloyed filler can help, but it is a double edged sword as the higher Ni&Mo alloys have even more segregation on solidification. I have seen 2.5% min Mo 316L filler used. You could also use 317L (or 317LMN) filler if you could find it.

For pitting resistance electrochemistry is hard to beat for speed and quantitative results. For 316L we used a couple of different solutions. One being 1000ppm Cl (as NaCl) pH=5 (adjust with HCl) and testing at 70 or 75C. We also did testing in 1% synthetic seawater at pH=4 and 60C. It depended on the application.
You are testing small areas (1 sq cm) and the test is fairly rapid (you can do 4 or 5 in one day), but the equipment is very specialized and requires some training. Gamry makes some of the best.
The results that you get are real numbers so it is easier to compare samples to each other.

Testing for CSCC is harder and for another post.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[SJones]

Be careful what you start, this sounds like it could turn into a PhD project with all the variables that you will have to juggle.

ASTM G61; ASTM G150; NACE Corrosion 2018, Paper 11173 (neglect the H2S discussion); other search results for "four point bend testing of welds"

Steve Jones
Corrosion Management Consultant

http://www.linkedin.com/in/drstevejones

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

 

[EdStainless]

So your corrosion is either a result of start up/shutdown cycles or it is happening in a section of equipment that is cool enough to have condensation, because you will not get pitting or CSCC without an electrolyte.

Some of the best CSCC test work was done years ago by Avesta. They took strip samples (if welded the welds can be across the center or lengthwise down the middle) and put them in 4 point bending (people typically use either the spec min Yield strength or 90% of the actual Yield strength). A very sensitive thermocouple is spot welded to the back center of the sample. The fixture was built using ceramic contact points to allow electrical and thermal isolation. Then they pass electrical current through the sample to heat it. The temp has to be controlled to within a fraction of a degree. Then once everything is stable you begin dripping a mild Cl solution (I think that they used 1,000ppm) onto the center of the sample. Just as the drop dries up you add another. Of course as the test goes on the local Cl levels are increasing, which is common in real applications. Then you record the time to failure. Though failure can be tricky because you really want to know when the sample starts to crack. You can detect that because the load will drop and the sample deflection changes, so they use a small microswitch to detect the movement of the sample.

In pitting test I usually run duplicate samples and if they are close to each other I am satisfied. With SCC I always run tripe samples and then decide if I need to run more. The scatter in SCC testing can be huge.

This is a very involved field. I have multiple volumes on my shelves that cover just pitting and crevice corrosion, multiples on SCC, and even more that just cover corrosion testing.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[steris]

First, thanks for the feedback!

Ed - you raise some good questions. I'm sorry but I don't really know how to answer the failure mode question. In terms of present conditions, we see may visual evidence of corrosion after anywhere from 7-15 years in service depending on field conditions. The corrosion witnessed is typically pitting or cracking. These are observations done by naked eye and reported by end users who are untrained in corrosion. There is no microscopic examination performed by the end users in the field.

For pitting resistance, does that electrochemical experiment have a specific testing standard I would ask a lab to adhere to?

In terms of the various variables, we basically have 4 sample configurations that we want to test. Each sample weldment would be around 10"x10"x2" and would contain representative welds. To be clear, I'm not looking to get exact numerical corrosion rates but rather compare these samples against each other to see which combination is most corrosion resistant. The suggestion of running pitting in duplicate is very helpful!

Are there specific corrosion testing labs in the USA that anyone would recommend?

 

[EdStainless]

Either find a local lab where you can sit down and talk to the corrosion engineer there that does the EC testing. These tests tend to be very customized and initially there will be some trial and error testing.
The other option is to do it yourself.
For EC work there are test methods (ASTM G150) but no standard for the details of the test or standardization of results.
Pitting tests are done using samples the size of a postage stamp and fairly thin.
CSCC work is commonly done on strip samples 1" x 5" x <0.060".
For CSCC you have to be able to control the stress and temperature to have any chance of SCC actually happening and any chance of repeating the test.
CSCC will only happen when there is actual corrosion (even if just microscopic). 316 can be used in concentrated brine at high pH and low temp with zero risk of cracking.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[steris]

Thanks Ed! Do you think salt spray tests or boiling in salt water would be appropriate for pitting or generalized chloride corrosion?

 

[EdStainless]

Immersion tests can work for pitting (control pH, Cl, and temp) but they have very high variability and they are simply go/no-go.
We had alloys where we knew the critical pitting temp in a specific environment should always be above 70C, so we tested every production lot at 60C for a quick (72hr) check to see that things were fine.
Salt spray really justs tests the surface condition. If you want to test that then it works, but cleaning, pickle, passivate, and roughness will all change the results.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed