CO2 corrosion modeling and CO2 solubility in water 1

[Saad M.M.]

Greetings All,

This is my first participation in this great forum.

I am interested in corrosion modeling for oil and gas pipelines and I’ve been reading almost every publication related to this subject including INTETECH’s, M-506, FREECORP’s, De Waard-Milliams-Lotz, … etc. I am a user of ECE, and I have been challenging the limitation in this software.
I have developed good understanding so far. Few things I am still not sure about, and I thought that I can find the answer here.

1- Isn’t CO2 solubility a function of temperature? Why all the models look at the partial pressure only? For example pCO2 of 3 psi could be developed with a total pressure of 100 psia and 3%CO2 and also could be developed at 1000 psia and 0.3%CO2. However, CO2 solubility will vary with water temperature. As the temperature increases, the solubility decreases.
2- The effect of H2S is widely known to decrease the CO2 corrosion. It’s never been highlighted that the H2S solubility in water is higher than that of CO2 by one fold at any given temperature. To me this fact explains the effect of H2S in reducing the CO2 corrosion more than the protective FeS film (not that I disagree with FeS beneficial effects).
3- pH estimation varies to high degree between one model to another. I am confused between terms like “formation water”, “condensed water” and “Water saturated with corrosion products or Fe++”. Which is more critical? Can I assume condensed water as a conservative approach?

I know these questions could be challenging, but thinking out loud could help.

 

[EdStainless]

We look at P and not T with CO2 solubility because it is the larger factor, by a huge margin.
The difference in CO2 solubility at 100F and 400F is insignificant if the pressure is 1000psi.

The estimation of pH is tough. Which approach gives you the lowest estimate? That is the worst case. I worked on a project that was very low H2S and very high CO2, with high T and P. They ended up building a test rig to measure pH because the calculations were so far off.

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

 

[SJones]

1.
Some models address solubility by using a Henry's Law approach. The CO2 partial pressure simply serves as the data input to the algorithm. Cleverer models allow one to use actually measured dissolved CO2 (see NACE Corrosion 2005, Paper 05550)

2.
H2S can actually increase corrosion in some instances. Solubility might play a part, but it's a lot more complex than that. Get more information from starting with NACE Corrosion 2010, Paper 10278 and following through the references.

3.
The water terminology relates to the nature of the water expected in the system. What is 'critical' is using the one that most realistically reflects the situation being modelled. For a given gas phase composition, condensed water is likely to have the highest initial corrosion rate because it will initially lack the ions to raise pH

Take a look at

http://www.ife.no/en/publications/2009/matkor/prediction

and a forthcoming NACE Technical Committee report from TG447, "Selection Of Pipeline Flow And Internal Corrosion Models" if it gets published.

Finally, remember the maxim of oil & gas pipeline corrosion prediction: "how lucky do you feel?"

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.