Reactor MoC : High Salt Concentration

[TiCl4]

I have a potential small reactor design that will alternate between high pH (10% by weigh NaOH, so 14+ pH) that is later taken to a slightly acidic side with HCl, resulting in a 6-10% salt (NaCl) solution. The reactor temperature will be ambient, so no high temp conditions. I've reviewed some literature on chloride pitting, and have mostly settled on C276 as my prime material of choice. I'm exploring different options, of course, but budgetary estimates for an Alloy 20 reactor were only about 10% less than a C276.

Before settling, I have ordered several corrosion tabs so I can place them in this reaction mass (acidic side), seeing what, if any, corrosion develops over the course of 6-8 weeks.

A couple of questions:
1. Given my reaction is ambient temperature, will heating the material for the corrosion tabs testing be beneficial for accelerating corrosion to see it in a short timeframe? Or should I keep the tab testing near temps the reactor will see?
2. Are the below charts generally correct? Any other materials or other corrosion mechanisms to watch out for?

 

[EdStainless]

Generally a "C" type alloy would be selected for your service.
But what do you mean by 'slightly acidic'?
A pH of 5, or of 3?
Even with highly corrosion resistant alloys the alternating conditions will lead to some corrosion.
Generally a PREn of >40 is used for seawater service.
You could warm you test slightly (what is the hottest day that you expect? 100F) but if you heat too much you may change the corrosion mechanism and be testing the wrong thing.

 

[TiCl4]

Ed,

This is a developing process through R&D at the moment, so I don't have the exact pH. Given the nature of some of the reactants, I would expect a final pH of near 4 to be required as a starting baseline. Would going lower change material selection from C276 to something else?

How do alternating PH conditions lead to corrosion over and above simple exposure to high/low PH environments?

 

[EdStainless]

I am going from memory here, but I want to recall that 686 should stand up better in the alternating pH situation.
You are going to need very solid weld procedures.
And you will need to qualify those procedures with extensive corrosion testing.
Even if I used C22 I would weld with 686.

 

[TiCl4]

I am going from memory here, but I want to recall that 686 should stand up better in the alternating pH situation.
You are going to need very solid weld procedures.
And you will need to qualify those procedures with extensive corrosion testing.
Even if I used C22 I would weld with 686.

Having not spec'd a vessel for corrosive service before, what would typical specifications before for welding for corrosive environments? I assume there would be a limit on heat to limit the HAZ as well as certain cleanliness requirements, but haven't really delved into this area before.

 

[EdStainless]

You need to have raw material specs.
Your weld procedures will have a bunch of extra required variables.
Your procedure qualification should include microstructural evaluation (no detrimental phases) and corrosion testing.
For a corrosion test I would use G48-C at 85C for 72hrs.
There should be no corrosion attack.

 

[TiCl4]

FYI we performed some initial corrosion tests in the lab with a few materials. In order of the pictures below, they are 2205, Alloy 20, and C276. This is 1-week of immersion under flowing conditions.

Also FYI, the pH condition of the material was 0.8, not 4.

It's quite obvious that C276 has superior corrosion resistance to the actual material (contains more than just HCl and NaOH). Given the multiplicity of products expected in this vessel, and given this particular product will only be made a few times a year, this 1 week of testing is the equivalent of approximately 1.5 years of exposure. The remainder of products are non-corrosive to 304 SS, so this material choice is solely to resist attack from this particular product.

I believe the rust blooms on the C276 are residual iron contamination from the cutting process from the mill, not from actual base material corrosion. I scrubbed with a non-iron pad and solvent washed before immersion, but it looks like some iron still got through.

Any thoughts, looking at this?

In order: 2205, Alloy 20, C276

 

[EdStainless]

What do they look like when rinsed off?
There will be little to no price difference going from C276 to C22.
The real concern is the source.
If you get the cheapest generic 276 out there you may be in for some rude surprises.
Some of it is quite bad.
The welding will still be a serious issue.

 

[TiCl4]

I'll bring them out for good in 1 week and rinse/clean. This was a 1-week visual inspection made just due to the change in hue of some of the samples. We measured 4-point thickness on corners and also the weight before the test, so I'll be able to tell any thickness changes from corrosion and total metal loss.

What do you mean by the "cheapest generic 276"? I had planned on spec'ing Hastelloy C-276 per ASTM B906 with a surface finish found in 13.1.1 of that standard. Is that a "generic" C276? I also thought about requiring ERNiCrMo-4 or ERNiCrMo-10 for filler material, and also examining the micro-structure of the as-welded specimens that will be subjected to the G48 test to ensure no mu or P phases are formed during welding.

Lastly, I've seen a recommendation for using pulsed TIG for welding due to better heat dispersion, which helps avoid those phases. Is this something also worth requiring for construction?

 

[EdStainless]

No, you don't want to write your spec like that because it requires purchasing only from Haynes.
There are many other good sources (ATI, Amperam, Special Metals and so on).
If you only deal with one mill you may have trouble getting all of the product forms that you need.
If you have time to test I would try the ERNiCrMo-14 filler.

https://www.specialmetals.com/divisions/welding-products/tradenames/inco-weld/fm686cpt.pdf

And I would be pickier about sourcing the filler than the plate.
Regardless of welding method if it passes your testing you should be good.

 

[TiCl4]

Firstly, thank you for all the help you’ve given thus far. I am sorry for the continued questions, but my company doesn’t have any experience with these types of alloys, so I am having to rely on senior Google and helpful strangers like yourself.

(Edit: Answered my own question. Found out Hastelloy is a trademarked name, which is why you mentioned it would force the manufacturer to buy from Haynes. Will update the spec to state UNS10276 per ASTM B906 standard instead of Hastelloy.)

Is there a good resource to reference that goes into these sorts of details?

 

[EdStainless]

It might be better if you and I talked privately about this.

 

[TiCl4]

I've sent you a PM.