NACE applicability criterion for liquid streams

[DieguitoI]

Hello!

I have a doubt regarding on how to apply Annex C.2 of NACE ISO 15156-2 2009 Standard. For a gas free liquid system, the H2S partial pressure for assess the applicability of the Nace Standard, i.e. pH2S>0.05 psia, is calculated at a pressure corresponding to the bubble pressure at the operating temperature (Pb) and for the H2S molar fraction of the gas phase in equilibrium with the liquid at the bubble point (xH2S). Thus: pH2S=pB*xH2S. In my application I have a bi-phase liquid stream HC+H2O. What bubble pressure I have to consider the one related to the HC phase or the one of H2O phase? Since SSC corrosion can happens only in presence of an aqueous phase, to me it seems more correctly to consider the equilibrium bubble pressure of the H2O phase alone, but Annex C of Nace standard states:
“For a liquid-full pipeline downstream of gas separation units, a good approximation for bubble-point pressure is the total pressure of the last gas separator”
That is it seems to consider the bubble pressure of the HC phase.

Regards

 

[Chumpes]

Hello
The problem is that in full liquid system you don't have a vapor phase but this does not mean that the activity of H2S, H2 or other gases (expressed as an equivalent partial pressure) is nil.
In case of a separator drum just upstream the full liquid stream to consider, the bubble point pressure in the liquid stream is (with excellent approximation) the total pressure in the separator. Because liquid HC are at equilibrium with vapor HC and liquid water is at equilibrium with water vapor (ppH2O) in the upstream separator, then the bubble point pressure of the liquid HC and the bubble point pressure of the liquid water are equal to the total pressure in the separator. Also in the separator, gas H2S is at equilibrium with dissolved H2S in the water phase and in the liquid HC phase.

Imagine that the full liquid HC+H2O+dissolved H2S stream downstream of a separator is a little bit heated (+0.1°C) at constant pressure, or little depressurized (-0.1bar) at constant temperature, then you will have formation of a vapor phase (fist bubble) which composition is almost exactly the same as the vapor composition (HC, H2O, H2S) in the upstream separator, because the equilibrium with the gas phase is achieved in the upstream separator. If the increase in temperature or the decrease in pressure of the full liquid Stream become significant, then the approximation is no more perfectly valid.

When checking sour service limits of different streams, you better have to consider the properties and equilibriums of the HC phase because thermodynamic models used for HC equilibriums are usually poor with electrolytic calculations that you could also have performed for the water phase, and also because H2S does not dissolves much in water (the presence of a water phase will not modify significantly partial pressure H2S due to relativly poor dissolution ability of H2S in water). Detailed calculation of sufides (H2S, HS- and S2-) cocentrations in water requires electrolytic models (influence of pH)..

Regards

 

[SJones]

Surely, it's a bubble point pressure of the system, rather than a bubble point pressure of either the hydrocarbon, or the water, alone. The value should be available from good process simulation software such as OLI Studio Stream Analyser, or UniSim. Remember that it is also necessary to derive the H2S content of the gas at its point of liberation. This content will differ if the gas is liberated solely from hydrocarbon, or solely from water (see attached partitioning discussion). Given the imprecise nature of working with the thresholds of ISO 15156, a number of users would argue that the highest computed bubble point pressure should be used for the sake of conservatism.

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.

 

[DieguitoI]

Dear Sirs,
Thanks for your valuable replies.