Carbon Steel Corrosion from Produced Water 3

[jskernahan]

Hi all,

I'm replacing an old section of produced water pipe and am trying to figure out if I should internally coat the carbon steel (A106-B). I don't have any info an the existing pipe so can't use that as a benchmark, however I have a water analysis that shows a chloride content of 13,896 mg/L. The system is closed so there is no oxygen ingress, does anyone have any insight as to what kind of corrosion to expect in this line? Is there a good argument for not internally coating the pipe based on the analysis and fact that there's no oxygen ingress? The new pipe will have a corrosion allowance of 1.5 mm and I've attached the water analysis. Any help would be appreciated. Thanks,

James

 

[Chumpes]

hi James
what is the service temperature?
did you investigate Langelier Index ? (

https://corrosion-doctors.org/Cooling-Water-Towers/Index-Lang-calcul.htm)

remark : i think you can say that chloride content is 14000ppm or 1.4% because 13,896 is too precise and stings the eyes.
regards

 

[GabrieleB]

With this content of chlorides and the PH quite high (above 7), you can use the CS withouth any internal lining. The important thing is that you are able to avoid the oxigen ingress. As a general consideration I would have prefer a corrosion allowance of 3mm for a longer design life of the pipe, even though such conditions are not particularly critical for a carbon steel line pipe (ASTM A 106 Gr.B or equivalent are OK).

 

[jskernahan]

Great thanks for the help guys. I'll look into beefing up the CA and will round my numbers for future posts

 

[cavbilly]

I reviewed the analysis you linked to. You state it is a sealed produced water system. Are you certain there is nil oxygen content in the water at the point of origin? I would suggest at least one further test, and that would be ORP. Based on the TDS, species, and the pH, an ORP value for no oxygen should be around 150-200 mV, and oxygenated would be up to 350 mV if saturated.
Another test would be the use of a field dissolved oxygen meter capable of reading at least somewhat below 0.5-1.0 mg/L of water as O[sub]2[/sub].
If oxygen is confirmed or presumptively confirmed low, then by all means proceed.
Any possibility of air ingress by some Venturi effect on the line at any point, and downstream from that point you will be asking for trouble.

 

[Metalero]

Do you have a picture of the internal surface of the removed section? Has it Corrosion?
Why do you think you need coating?

Stiff polygon says corrosive. Langelier or Ryznar index maybe says pitting corrosion, but you need operation temperature.
What about the flow: continous or intermitent?

Do you clea perodically the infernal surface of the pipe?

 

[Chumpes]

@cavbilly
0.5 to 1 mg/L is roughly 0.5 to 1 ppm !! this is quite high for oxygen in water vs. corrosion.
low oxygen is << 50-100 ppb and oxygen probes could usually measure down to few ppb. when deaerated water is required, less than 10ppb oxygen can be usually be met.

 

[cavbilly]

@Chumpes - yes you are correct. Oxygen in that range can and will contribute to corrosion. We are talking about "deep" produced water here, and I suspect
it does not meet the definition of de-oxygenated water. With the chemistry of the original poster's water, there had better not be much oxygen, or there will be problems.
Some instruments are better (lower limits of detectability) than others. Having said that, the quenched luminescence probes are relatively easily calibrated when they
are set up for readings in the ppb range in the first place, although some of those meters are meant for tests on wastewater, where they consistently look at some fraction of saturation above 80%.

For boiler applications, I am well aware of the requirements for exceedingly low oxygen residuals, unless it is oxygenated treatment protocol, where it is around 30 ppb, fully de-aerated, then back injected when intentional.

 

[SJones]

Since this is produced water, there will have been, at some stage, CO[sub]2[/sub]around. Thus, there is a possibility of some CO[sub]2[/sub]corrosion - how much can be estimated by corrosion modelling. It shouldn't be too much given the high level of bicarbonate. However, if the measurements are not made correctly, some of that bicarbonate could actually be carboxylic acids which will change things on the corrosivity front. In addition, produced water systems are prone to bacterial contamination that can give rise to MIC and under deposit corrosion, so maybe get some bug checking done. Internal coating could be a reasonable barrier against the threat of MIC/under deposit/deadleg corrosion if piping constructability considerations allow it.

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.

 

[blacksmith37]

Long ago, cement lined CS was the standard for gathering lines; I don't remember looking at many failures so it likely worked pretty well.

 

[springrise]

Hello James,
based on the analysis this is wastewater not produced water (could be e.g. brine from softener regeneration, reverse osmosis concentrate, etc.). The water has 200 German degrees of hardness!
Given the huge salinity, low pH (note 7,3 IS corrosive for carbon steel) I would suggest using coating or plastic pipe if possible.

 

[cavbilly]

@springrise
I cannot agree more, this is quite corrosive water in its nature, and with this much hardness, trouble is likely.

 

[georgeverghese]

Dissolved O2 meaurements are unreliable at these low values - it is highly likely you've got O2 ingress via all sorts of chemical dosing systems / corrosion inhibitor injection units running off atmospheric tanks further upstream. If you're looking for a fix with a reasonable life, go for lean duplex SS at the least(22Cr).