Jump to content



Featured Articles

Check out the latest featured articles.

File Library

Check out the latest downloads available in the File Library.

New Article

Product Viscosity vs. Shear

Featured File

Vertical Tank Selection

New Blog Entry

Low Flow in Pipes- posted in Ankur's blog

Base Oil Hydrotreater Reactor Effluent Hhps Quench Line Failures

h2s polythionic acid hydrotreater base oil lubricants

This topic has been archived. This means that you cannot reply to this topic.
6 replies to this topic
Share this topic:
| More

#1 fseipel

fseipel

    Gold Member

  • Members
  • 81 posts

Posted 14 June 2014 - 09:06 PM

We've had some corrosion in a 3" Sch. 80 Alloy 800 pipe downstream of a base oil hydrotreater train.  Corrosion (cracking) causes pinholes in the high pressure (~1300 psig) line.  This line comes off top of the HHPS.  Combined Hydrogen/H2S gas stream enters at ~450 F and after quench is ~250-300 F.  Problems (leaks) occur downstream of the water injection quill.  This has been a recurring problem every year or so.  I have obtained API 932-B 'Design, Materials, Fabrication, Operation, and Inspection Guidelines for Corrosion Control in Hydroprocessing Reactor Effluent Air Cooler (REAC) Systems' which seems applicable but am looking for follow-on information & your expert advice.

 

1.) Are there any other recommended references which would be useful?  I'm working in a waste oil re-refinery so my resources are not as vast as for someone at a crude oil refinery.

 

2.) We currently just use softened municipal water pretreated w/carbon to remove chlorine, calcium, and magnesium.  With test strips, I'm seeing < 0.2 ppm chlorine.  Hardness < 2 ppm.  So that seems okay.

 

3.) What, if any, deoxygenation should be conducted?  Standard calls for 15 ppbw Oxygen in the quench water; I expect I have much more; I don't have a DO meter but I'm not de-oxygenating at all presently.  Recommendations?  I'm not a microbiologist but I'm guessing the tap water contains 2-6 ppm DO, much higher than the 15 ppbw level called for in this standard.  I'm getting a deaerator soon for boiler, should I just take boiler feedwater, cool it & use it for quench or are any other additives problematic?  Or is O2 more a concern during lay-up?  Currently we don't do soda ash rinse before shutdowns.

 

4.) Failures in past have been to heat affected zone adjacent to welds, and to welds themselves.  Any recommendations on a testing company familiar with this type of corrosion/service to inspect the pipe & make recommendations?

 

5.) Should post weld heat treating be used?  Most references say no, but seems to me it would take the carbide back into solution -- failures have been in heat affected zone near welds or in welds themselves.  Line used to be SS.  Old SS line (from before when I started) has rust in welds; appears to me whoever made it used a carbon steel brush... I don't think welds themselves failed since switching to Inconel but heat affected area adjacent to them did, presumably due to carbide precipitation?

 

I'm going to look at quench water flow rate, remove nozzle & inspect spray pattern, and do some other investigation work.

 

6.) I'm not sure if Inconel 800 is best alloy choice per the standard since standard notes concerns with PTA SCC.  Any advice here?

 

7.) Any advice on injection quill spray angle or pipe size?

 

8.) Any inspection plan recommendations?  We were thinking of doing shear wave UT quarterly & during outages.

 



#2 Bobby Strain

Bobby Strain

    Gold Member

  • Members
  • 3,528 posts

Posted 14 June 2014 - 10:13 PM

Sounds like you should consult an experienced metallurgist.

 

Bobby



#3 fseipel

fseipel

    Gold Member

  • Members
  • 81 posts

Posted 15 June 2014 - 08:53 AM

Bobby: Yes, we intend to do that, but I had some questions beyond the scope of the metallurgy (process questions).  Another related question, is whether high-pressure clamps such as PLIDCO sleaves, would be appropriate for temporary repairs of this line?  If so, should the clamps that can withstand a complete failure (circumferential rupture) be specified?  I wasn't clear on how quickly a crack might propagate along the weld. I'm working on assembling technical references so any recent citations to quench design articles or standards are of of interest.



#4 TS1979

TS1979

    Gold Member

  • Members
  • 164 posts

Posted 15 June 2014 - 12:24 PM

Just a few thoughts:

 

1. If the problem happens downstream of the quench water injection point, the problem most likely caused by the water injection. When you add water to sour exhaust gas stream, acid will be formed. H2S may be too weak an acid to cause serious problem for Inconel 800. The problem may be due to chlorine from the base oil hydrotreating process instead of softened water. Is that possible to change the quench water injection location or add ammonia or soda in the quench water to increase the PH value of the water?

2. The DO in the quench water doesn't seem to be the cause of the problem, especially when most of the exhaust gas is H2S and H2.

3. Low temperature provides the condition for water existence and acid formation (~250 - 300F).



#5 fseipel

fseipel

    Gold Member

  • Members
  • 81 posts

Posted 15 June 2014 - 04:44 PM

TS1979: Thanks much for your response.  I am going to check the pH and chlorine concentration in the rinse water -- I have read guidelines for how much water we should be injecting, once I know ammonium bisulfide concentration in the water I should be better equipped to know whether I need to increase quench rate or adjust pH.  There were deposits in the 3" pipe which I removed, I do not yet know what they are -- corrosion byproducts or salts from quench -- I suspect the latter.  What testing do you recommend for those crystals?

 

I realize from the limited reading I've done so far that acid can form.  Failures have been at or near welds; probably some sort of stress corrosion cracking.  Is C-276 a better choice here?  I was asking about Oxygen in case it was considered polythionic acid.  I've also learned from some further reading I need to consider soda ash wash when plant is shutdown.  Is that the case EVEN if aren't breaking flanges or exposing it to air?  The inconel is not exhibiting significant general corrosion.

 

Adding ammonia or soda is certainly possible; I have 25% sodium hydroxide on site in a large storage tank.

 

I don't fully understand, what is meant by 'change the quench water injection location'?  Does this mean moving it downstream of the gas cooler?    I suppose that would reduce corrosion since it would be cooler.  But then, the cooler would be subjected to possible deposition of bisulfide, correct?

 

The crankcase re-refinery is SMALL; there is no air cooler after reactor, just the hot high pressure separator, followed by the quench injection, and then a water-cooled tube heat exchanger.  Hydrotreater processes about 30 GPM of liquid feed.



#6 TS1979

TS1979

    Gold Member

  • Members
  • 164 posts

Posted 17 June 2014 - 10:15 PM

I am not expert on this topic but just a few thoughts to shed some light to your problem. Certainly you can do element analysis for the deposit. I think that your thought is correct - the deposits are coming from quench water. Inside the piping, the section your injection point will become an evaporator - salts from the water will be left on the piping wall. Downstream of the injection point, the flow is most likely two-phase flow - vapor with water droplets. Therefore, there will be not only corrosion but also erosion. With erosion and corrosion, the failure point often appears at the elbow location.

 

By the way, what is the purpose of the quench water if you have a water-cooled heat exchanger?



#7 fseipel

fseipel

    Gold Member

  • Members
  • 81 posts

Posted 18 June 2014 - 06:13 PM

TS1979: The quench is to eliminate salts. As the reactor effluent (hydrogen gas containing ammonia and hydrogen sulfide) cools, ammonium bisulfide salt forms.  These deposits would foul the cooler if they are not dissolved by the quench.  So the quench is a must, regardless of whether the gas cooler is air-cooled or water-cooled.  At least, it's a must if there is much H2S and NH3 in the reactor effluent gases.  I once designed a hydrotreater WITHOUT a quench, but it was for a terminal alkene; a highly refined hydrocarbon with essentially NO sulfur and NO Nitrogen.  So in that case, I didn't need the quench because no salts formed in the effluent gases when cooled. I am new to hydrotreaters that include this quench but it is routine for hydroprocessing where sulfur must be removed since H2S is a byproduct, and typically, so is NH3.  I understand completely evaporation occurs and erosion, we are going to pull the injection quill out & inspect the spray pattern, I also tried to thermal image it unsuccessfully (pipe is uninsulated), but the emissivity of stainless prevented obtaining good results -- I wanted to see how temperature of pipe skin changed circumferentially & axially to see if I had a spray distribution problem while running.  I understand evaporative cooling occurs.  There must be a lot of ammonia too as pH goes UP above 7 in effluent water.  I'm not seeing signs of erosion.  Failure is in heat effected zone, cracks, I believe this is due to some sort of stress cracking.  I'm still looking for more technical references to quench design & we're going to check velocities against standards.






Similar Topics