Nitrogen Tank-Blanketing For Vocs Abatement Vent Header Collector

[EngDaniel]

I am currently working on an emissions abatement project that will involve extracting vapor streams of Volatile Organic Compounds from various tanks and sending them to a Regenerative Thermal Oxidizer (RTO) for incineration before discharge into the atmosphere. The tanks/vessels will be inerted with nitrogen and the excess waste vapor fumes will vent into the pipe headers and sucked to RTO by fans.

My concept design incorporates the use of standard Nitrogen Padding & Depadding Valves, Flame/Detonation Arresters, Oxygen Analyzers / LEL Detectors, Pressure-Vacuum Reliefs, Explosion Reliefs etc. I would welcome any advice on anything else I should consider in the project.

How best can I predict the maximum volumetric flows and the Nitrogen and VOC components concentrations on the main header prior to the RTO.

 

[EmmanuelTop]

Look at API 2000. There you can see how to calculate the maximum in-breathing and out-breathing flows for storage tanks. The header compositions will be the result of these flows.


Dejan IVANOVIC
Process Engineer, MSChE

 

[EngDaniel]

Thanks a lot Dejan for your advice. Well appreciated.
My understanding is that at the end of the day, the API guidance will not help me determine the concentration of Nitrogen and of the VOCs in the waste gas stream. Since with inerting, the waste gas stream is now devoid of oxygen and only remaining with Nitrogen saturated with volatiles from the coal tar product. My desire is for as little Oxygen to be present in the gas stream for safety and I do not wish to compromise that but the RTO vendor is concerned with the Nitrogen concentration and the lack of at least 15% Oxygen vol.

 

[Latexman]

The API guidance may not, but with some basic ChE'ing (vapor pressures, equilibrium, mass balance) a conservative basis for the RTO can be made. Air (or O2) will have to be added for the oxidation.

Good luck,
Latexman

Technically, the glass is always full - 1/2 air and 1/2 water.

 

[EngDaniel]

Thanks Latexman,
I am glad you understand my predicament! Now the challenge is that Coal-Tar is a complex mixture of short & long hydrocarbon chain compounds and the issue of Vapor Pressures is as complex as the mixture itself just like for bitumen or crude oil. That said, how best do we approach the "basic ChE'ing (vapor pressures, equilibrium, mass balance)" to come up with a good conservative estimate.

Yes, the RTO unit chosen is designed to internally handle the dilution of adding an air stream (for oxidation O2) before the combustion chamber. Even with that the RTO vendor has flagged that above a certain concentration of N2 in the waste gas stream the oxidation of VOCs will be compromised.

It remains my wish to deliver an inert gas stream up to the RTO battery limits and have it achieve my intended destruction efficiency!

 

[Latexman]

I have spec'd TO's for 3-4 chemical plants over about 36 years. My experience is with chemical plants where I know which chemical(s) is in which tank. In raw material storage tanks, it's almost always a pure component, which makes it easy. And, in mix tanks, reactors, and other equipment, I know the concentrations of the pure components through recipes and mass balances. I don't have experience characterizing HCs. Some one else will have to chime in on that. I would assume they would be characterized as a mixture of pure HCs, but what do I know, there may be a better way.

Good luck,
Latexman

Technically, the glass is always full - 1/2 air and 1/2 water.

 

[StoneCold]

EngDaniel

If you want to use a thermal oxidizer on your vent stream you might want to consider using natural gas as your blanketing gas. This pushes you above the UEL instead of using nitrogen to push you below the LEL. So you still get explosion protection. However the stream is easily burned in the thermal oxidizer without the problem of excessive nitrogen. This is more practical for outside storage tanks than reactors inside a building, but it sounds like a good option for you here. If you use either nitrogen or natural gas you are going to need a combustion analyzer on the inlet of the thermal oxidizer to measure the gas composition. Then you can add oxygen or natural gas to the vapor stream to get into the combustible range for the thermal oxidizer. These units also need a detonation arrestor to separate them from the vent line header.

Regards
StoneCold

 

[georgeverghese]

Presume you have a combustion air fan at the RTO? What purity is the N2 pad gas?
Depending on how these tanks are filled with coal tar, you may run the risk of blocking up the fine heat dissipation mesh on these expensive detonation arrestors. When that happens, the only protection you've got against flashback from the TO is to enable some positive flow of air into the waste gas burner.
Some attention required to the sample gas preconditioning kit in the LEL analyser - low points to be drained out or designed out - this preconditioning unit is the weak link in the LEL protection loop.

 

[EngDaniel]

Thanks Latexman. Nevertheless you have clarified the problem.

 

[EngDaniel]

StoneCold
Thank you for your advice.

Well using Natural gas is not an option for me both on viability and cost. The natural gas equally displaces O2 in the waste gas stream like N2 does. So I assume that as far as the RTO vendor that boils to the same thing - less than optimum O2.
So far I had only considered having an O2 Analyzer to monitor for flammability/ingress of O2.

I suppose the Combustion Analyzer you mention would have to be after addition of an air stream. Well my RTO of choice will have this addition of air done internally in the RTO. I dont think it would be a good idea to have double dilution points.

Yes I have incorporated Deflagration/Detonation Arresters.

 

[EngDaniel]

Thanks a lot georgeverghese for the pointers.
I will check on the N2 purity.
At this stage I have only considered having Detonation/Deflagration Arresters on the vent header piping and not at tank level. I will relook at that. We have considered steam tracing and insulation to discourage condensation.

Depending on unit cost of arresters I may consider having a Duty-Stand by arrangement to allow for regular maintenance.

Yes gas sampling and conditioning will be a real challenge. I am currently looking at an inline O2 probe thats on the market. That may mean I will not need a sampling system.

Thank you.

 

[Compositepro]

I just noticed that your handle says (Electical). Emissions control is a very complex subject and there are very many options that depend on amount of flow, frequency of flow, predictability of flow. Choices made now can be very costly for years to come. You really should have an experienced Chemical Engineer involved. Vendors will simply try to sell you their products.

I see you mention fans along with N2 inerting. These do not go well together. N2 piping should be positive pressure to avoid sucking air in. Blanketed tanks should have conservation vents so there is no flow out of the tank until a certain pressure is exceeded. Condensers or carbon absorbers on the vents are other options. I got involved in an oxidizer project after the equipment had exploded a few months after installation (first project on a new job). A surge of solvent fumes got to the oxidizer before valves had time to react. The equipment had cost over one million dollars and cost over $200,000 per year to operate. Eventually we were able to schedule operations so the equipment did not have to run 24/7 and got operating cost down to $40,000 per year. That was 30 years ago. An oxidizer for tank vents does not sound like the best option. All cars have charcoal canisters on the tank vents, for example.

 

[EngDaniel]

Compositepro,
Thanks for your valid observations and advice (& the frightening past experience with Oxiders). I am specifically a C&I Engineer with a strong chemical industry background and working along side a team of versatile Chem Engineers who take the lead in the design. However where I am not convinced or need alternative clarity I seek independent opinion as thus.

We are looking at nitrogen blanketing the vessels at a slight positive pressure (say about 1.5kpa)and allowing the vessels to vent excess vapors into the vent header collection system through a PCV at a slightly higher pressure of say 2kpa. The fans (with VSDs)at the RTO will draw and maintain flow in the headers thus sweeping out all fumes to the RTO. At the tank we intend on having the nitrogen blanketing valve, a PVV and the discharge PCV as well as emergency lifting hatches.

I will discuss your observations with my team.

Thanks

 

[georgeverghese]

The following should be enabled

max sampling time for the LEL analyser = t1

time taken for the waste gas to travel from the LEL sampling take off (from the waste gas header) to the RTO shutdown valves - add on for the time taken for these valves to close also = t2

t1 should be < t2

Velocity of the waste gas mix through the header to be made slow enough to make this happen.

Would guess that in this application, loss of N2 padding would be an appropriate safety trip( i.e PSLL on the tank) , which should be a lot more reliable than these LEL analysers ?

 

[EngDaniel]

Thank you georgeverghese for your advice. Indeed I shall have a closer look on the time delays and their effect. However as I have noted earlier I am also looking at employing an in-line, Mettler Toledo, oxygen probe that not require sampling and conditioning system. My thinking is that since the system is meant to operate with little or no oxygen then monitoring of O2 is primary and more appropriate than LEL Analysis (looking at it from a different angle the two appear to be one and the same thing right?)

As you have suggested, I shall definitely incorporate Loss of N2 trip.

 

[georgeverghese]

Sorry, my previous advice should be corrected

2.t1 should be < t2

Agreed, the O2 analyser may be better than an LEL analyser in this application. I'd be wary of not having a preconditioning unit for this O2 analyser in the coal tar waste gas service-pls check again and provide all necessary info to the vendor. Also check if there are other chemical species in this waste gas that may interfere with the primary sensing function ( water vapor, H2S, CO2, aromatics etc)

 

[georgeverghese]

Presume you have a startup RTO bypass stack?

 

[EngDaniel]

Thanks x 10^E6 georgeverghese
Sure I will interrogate the vendor till I am satisfied.But the technology looks like its up for this challenge.You would think the probe is an RTD temperature probe... no gas sample drawn from the process. I will consider duplex sensor option due to high probability of fouling. Well the alternative option of gas sampling and conditioning system and a normal O2 Analyser is still on the table. Yes the system will have fast acting valve by-pass system.

Much appreciated!