Venting toxic vapours

[Tom_H]

Hi,

We have an older atmospheric tank currently only equipped with a lute pot for relief. It contains a flammable (flash point <20 C/68 F) and quite toxic liquid. It is nitrogen blanketed. It has been considered to equip the tank with a pressure/vacuum valve also sized for fire case. The question is whether to keep the lute pot and also how to make sure venting is to a “safe location”. This would apply to the lute pot today also of course. The plant does not have a flare system. API 2000 is used as guidance. It is possible to do a quick dispersion study I guess. Any good experience to share of this and the fate of the lute pot?

 

[don1980]

I don't consider a lute pot ("liquid seal pot") to be a reliable way to maintain a pressure pad on a tank. It's not easy to ensure that the pot's liquid level is maintained. Also, there's the risk of the fluid freezing or becoming plugged with algae or debris.

The only way to assess the hazard caused by the vent stream is to do a dispersion analysis. The results of that analysis will guide you in choosing a safe location/elevation for this release. Routing tank vents to a flare header is only feasible and safe when there's no risk of any significant pressure inside the flare header. This can be done if the flare system is exclusively for storage tank vents. But even then, the tanks need an emergency vent device routed directly to the atmosphere. It's almost certain that the plant flare isn't a safe destination for this tank vent.

 

[Tom_H]

Thanks for the reply. Yes, I agree, liquid seals are definitely not the preferred solution. There is no flare on site, so the alternative is a vent stack (i.e. open vent with some elevation). It could possibly serve many tanks. But it is of course a costly exercise unless truly needed. Dispersion study seems to be the way to go.

 

[pennpiper]

Tom_H,
If I light a match around a Propane leak I would go BOOM!
If I inhale Chloroform I would go to sleep for a while!

It might help people here if you were to give us better information.
What do you mean when you say:
(a) flammable liquid?
(b) quite toxic liquid?





Sometimes its possible to do all the right things and still get bad results

 

[georgeverghese]

Both dispersion and flame radiation studies should be done for toxic and flammable vapors for the remote vent. Check local environmental emissions guidelines for permissible short and long term operator exposure concentrations at ground level and any other nearby elevated platforms that may be manned. For carcinogens and other chemicals with HSE concerns, incineration of tank filling operation displacement vapors may be the only solution (thermal oxidisers or incinerators).

 

[Tom_H]

pennpiper,

Thank you for your reply. The reason I did not specify the chemical composition is that it is a partly proprietary compound and I did not think it would have any meningful impact on the question to know the exact chemical composition for the purposes of general relief advise. I thought that the problem with having a flammable and toxic substance in a tank would be fairly commonplace in the industry and that there would be best practises (API 2000 does not delve in to detail on this) for how to handle relief of such substances when there is no flare system on site. It would also be problematic with connecting to a flare system (if it existed) as has been mentioned, due to the fact that the tank is atmospheric.

But as I mentioned, the compound has a flash point of just below 20 deg C and is toxic to inhale and can also enter the body via skin. I do not believe it is a carcinogen.

I was also curious what the advise in here would be on whether to keep the liquid seal after installing the PV, or to remove the liquid seal. It could be argued that the liquid seal provides some extra protection, akin to installing two redundant PV:s for example.

But as has been stated in the thread, it seems the best approach is to perform a relief dispersion analysis and take it from there.

 

[MJCronin]

Good discussion of LP tank venting in this paper:

http://ichemeuat.icheme.org/communi...evention/Hazards Archive/XIX/XIX-Paper-63.pdf

Unless your maintenance department sees a future need for the lute (seal)pot during possible relief valve work, I would remove it.

Because of the need to maintain an accurate liquid level in the lute pot, they have never been a good solution to this problem. Seal pots also are emissions sources.

https://books.google.com/books?id=N...UQ6AEwAXoECAkQAQ#v=onepage&q=lute pot&f=false

You may also want to consider flame arrestors in your system ........ Protego (and others) would be happy to sell you what you want ...

https://www.icheme.org/communities/...Prevention/Hazards Archive/XX/XX-Paper-39.pdf

Keep us in the loop ......

Please respect us and tell us of your final solution to this venting problem ....





MJCronin
Sr. Process Engineer

 

[Tom_H]

Thanks for the replies. We ended up performing dispersion simulations showing that there was negligible probability of a toxic gas cloud impacting persons. Or indeed even reaching ground level. In other case, we would have considered an elevated vent as first alternative option before other choices.