Emissions from Valves and Flanges - Service Classification

[Cokers]

Hi All,

My question is with regards to classification of services for the purposes of limiting emissions from control valves, emergency isolation valves, flanges etc.
Some criteria used are toxicity of fluid, emmissions of Hazardous Atmospheric Pollutants (HAPs), or VOCs as examples.

The question is with regards to rich amine, what would it be classified as?

My opinion is that it would be classified as toxic due to the concentration of H2S regardless of any chemical bond preventing its release.
Since certain materials are toxic at different concentrations, we revert to NFPA and/or European Commission standards - which tend to be more direct.

Co-workers are challenging my justification of toxicity of rich amine by stating that H2S has a chemical bond that prevents emissions from valves etc. I agree to a certain extent but if NFPA and others do not state this, toxicity shall be a function of the concentration alone.

Your expert opinions are appreciated.

 

[sshep]

Hey Cokers,

I am going to pitch in here, but would love to hear from others. Sometime back I created a training module for our company engineers to help make such determinations using data from ASME BPV code (vessels), B31.3 (piping), and NFPA. Because of the way the codes are written, toxic is somewhat qualitatively defined, so you must also include your own engineering standards in making these determinations.

This was my guidence:
If both the BVP and piping codes are considered, toxic includes these three subcategories: Lethal, Category M (piping), and damaging to human tissue. Lethal and Category M are basically the BPV and piping code equivalents. These can generally be associated with fluids which have a NFPA health rating of 4 (like H2S). Damaging to human tissue is fluid which has an NFPA health rating of 3, excluding when it is an asphyxiant only hazard (like N2).

When dealing with mixtures as we often do, a company based standard may apply using NFPA mixing rules. At that point you must consider whether the fluid is liquid or vapor, and also the actual conditions of use- usually to identify situations where the hazard is worse than calculated by strict application of the simple mixing rules.

For the simple mixing rules we use:
Liquid: use mass%.
If sum(NFPA 4) is 7% or more then "Lethal and Category M"
If sum(NFPA 3 and 4) is 25% or more then "Damaging to human tissue"
Gases: use vol%.
If sum(NFPA 4) is 1% or more then "Lethal and Category M"
If sum(NFPA 3 and 4) is 25% or more then "Damaging to human tissue"

But as an engineer, you can make a higher risk determination if based on your knowledge and experience the fluid under conditions of use poses a greater hazard than above.

This being said, your determination will depend on the H2S (NFPA 4 hazard) concentration and the NFPA rating of the amine. There used to be a standard called NFPA 49 that gave a list of components with their health ratings, but this publication is now obsolete, and replaced by the criteria of NFPA 704. Now the toxicity data from the MSDS is usually required to make the determination.

Sorry for the length, but this is my understanding. Again, I would love to hear from others on the subject. You might also get guidence from one of the code forums.

best wishes always,
Sean Shepherd

 

[Chance17]

Will H2S be released if amine solution leaks from the valve packing?
I see the answer as yes.
Most of the H2S will bond to amine.
But there will be a small vapor H2S relaase-thus emissions.
Your co-wokers should recognize the chemical bonding is saturated.

Most rich amine systems have filters to remove solids.
From my personal experience, there is alot of residual H2S vapors inside the filter when the elements are replaced.
Evidence to me that chemical bonding is insufficient to prevent emissions.

 

[dcasto]

I am with your co-workers. I've never seen the rich amine system considered a category M. From the still overhead on, that might be considered a category M if the H2S levels are high enough or the radius of exposure is large enough.

SSHEP sums it up as engineering experience factor. Every H2S exposure issue has been in the sour gas and still overhead systems in my 35 years of experience. The use of administrative controls handle 90% of the issues, those rare 10% require the use of engineering controls or systems. I've installed acid gas compression with 35% to 65% H2S and I've never used or seen used bellows type valve stems or anything beyond double distance pieces with sweet gas purge on compressor packing.

 

[Cokers]

I apologize for not responding earlier and thank you to all for the valued responses.

The direction we went ahead with, is based the concentration of the toxic component in the mixture as stated by sshep. The European Directive gives a more direct approach for the mixing as well and we opted to going with the more conservative of NFPA 704 and the Eurpoean Directive.

It is also true that experience matters. In my experience I have not seen rich amine classified either.

 

[JLSeagull]

I assume that your question pertains to the 100 ppm EPA VOC and LDAR monitoring regulations. The question is valid. A response can be more complex to standardize suitable operating and service issues.

This topic has appeared before in a different manner such including eng-tips

http://www.eng-tips.com/viewthread.cfm?qid=185677&page=7

http://www.eng-tips.com/viewthread.cfm?qid=214341&page=27

http://www.eng-tips.com/viewthread.cfm?qid=155734

and web searches for government stuff such as

http://www.epa.gov/compliance/neic/field/leak.html

http://www.epa.gov/compliance/resources/newsletters/civil/enfalert/emissions.pdf

...
Normally the operating companies define the design criteria. Process engineers define the fluid. Different national regulations apply the leakage rate requirements. Different manufacturers offer different packing selections to address those regulations.

Few service descriptions or material balances provide sufficient normal and upset stream data to clearly define all issues. Thus, some engineers extend the toxic criteria for uniformity. Excluding the utilities, if many of the process applications require enviro-seal packing then some engineers concentrate upon the temperature to select the TFE or graphite based enviro-seal material instead of environmental vs. general packing. Restated, some engineers apply emmision limiting control valve packing in fluids that may not require such limits in a manner that provides consistent trim repair parts etc.