Pre-heat HX corrosion at NHT installation due to LPG Recovery. 2

[Reformator]

Dear all,

Which material will be good for tube bundles of HX:

Working parameters:

The battery of HX at NHT represents a critical area in terms of service life. Technological parameters in which these exchangers work:
- Tube bundle -gasoline ~45 bar, recovered from several process machinery + hydrogen gas at 45 bar working temperature: product T int = 50-55c, T product output = 120C.
- Shell -hydrophinized gasoline + gases with H2, H2S, NH4Cl and condensate water, injected for the purpose of driving and / or dissolution of any deposits or precipitates.
-
Working parameters: T int = 140-160 C, T output = 110-120 C, P ~ 40 bar.
A simulation of the reaction environment estimated a composition of the reaction effluent:
43.3 mol % H2, 0.24 mol % H2S, 130 ppm NH3 and 10 mol% water, and in the water collected in the dome of the vessel, chlorine ions in proportion of 7.6-18 mg / l, sulphides 652-705 mg/l, ammonia nitrogen 260-288 mg / L and pH= 6.46 - 6.58.


This year due to the start of LPG Recovery, content of H2S (0.4 % - 1ppm) doubled at NHT which affected to pre heat HX.

Current tube Bundle material: 17Cr-12Ni-2.5Mo
Please help me to choose proper material

 

This task is not done by crowdsourcing.
Please hire a HX design engineer.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[EdStainless]

What is wrong with the current unit? Perhaps starting with some good failure analysis would inform the discussion.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[Reformator]

Let me first tell you a little bit of history. 4X4 Heat Exchangers. From Shell side it is gasoline + water+ H2+H2S+NH4CL+CL while at tube side are gases + H2. HX functioned normally before refinery decided to build LPG Recovery, which started to function few month ago and all H2 absorbed by gasoline directly transferred to NHT.

We noticed from analysis that content of H2S increased drastically( 2 times ). So,

43.3 mol % H2, 0.24 mol % H2S, 130 ppm NH3 si 10 mol% water. (Now H2S is 0.48 mol%)


Corroded HX are S1D and S1H. And I must mention that eddy current examination of tube bundle showed us that at the dew point, where water condenses, tube bundle thickness decreased up to 90 %. (Corrosion is uniformly increased from 70 – 90 % along length of HX )

Corrosion Team concluded that at such environment, metallurgy must to be upgraded.

 

[EdStainless]

One of two things are off here, either there is a lot more acid forming than this analysis would indicate or there is something else in the process. There are very few acids that will do this in austenitic SS. If you H2SO4 is high enough the chlorides will be converted to HCl. This mixture (if it is up in the 5% range) will cause general corrosion in 316. If that is really what you have you have limited options for alloys, and they are fairly exotic. If you had more dilution it might help.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[Reformator]

I forget to write that Cl is also exists: 7.6 mg/l

Is it possible that Cl dissolved in water will cause pitting of 306 and at high pressure at 40 bar, H will induce into crystal lattice and ssc will begin?

To be honest I am a little bit confused. If i will choose austenitic material, which is resistant to wet H2S: Cl corrosion will begin. If i will choose Cl resistant material: it will be susceptible to H2S attack. Is there anything in the middle?

 

[Dhurjati Sen]

Dear Reformator,

If you are confused and crowdsourcing is the method, why not explore titanium tubes?

Regards.

DHURJATI SEN
Kolkata, India

 

[EdStainless]

I am not sure that Ti would be a good option with hydrogen in the process, once Ti hydrides it is permanent.
Likewise I might hesitate at using a superferritic or duplex SS because of the risk of hydrogen embrittlement. Though it is much less of a risk since it is reversible and you are hot enough that it is less a factor.
Higher alloy austenitic SS are both more resistant to pitting and cracking.
Such SS grades crack because of the Cl, not the H. But low pH makes the risk of both worse.
I presume that your pH measurement on a flashed sample and at process conditions.
Have you extracted tubes to verify the ECT? If you haven't then you are just guessing.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[Chumpes]

hello
remark : should say chloride ions and not chlorine ions

continuous wash water injection into the effleunt is advisable due to possible corrosion under solid hygroscopic ammonium salts deposits (the effluent gas has 10%mol H2O !!! are you sure this value is consistent ? ?)

going through this very fast, I would advise to investigate the possible use of Ni-Alloy 825 for the tubes of this feed/effluent heat exchanger :
- good resistance to generalized corrosion in wet H2S
- good resistance to chloride SCC
- with continuous wash water injection

regards