VOCs abasement in petrochemical plant

[Knowhow_]

We have a project in a petrochemical plant to treat the VOCs from the sewage tank. There are hydrocarbons, including benzene，toluene, and xylene, and H2S in the stream. The inlet concentration of non-methane hydrocarbon is about 6,000 mg/m3, and the air flow rate is 1,700 m3/h. The current technological process is wet scrubber & plasma & activated carbon. However, the effect is not good, and we can't meet the emission standard of VOCs or non-methane hydrocarbon. And it's very strange that sometimes the outlet concentration of non-methane hydrocarbon is even higher than its inlet concentration.
Do anyone of you have experience in similar cases? Could you give me some advice?
Many thanks.

 

[bimr]

What type of activated sludge system do you have?

It would appear that it would be more attrative to develop a systems approach. Use of an activated sludge process such as the pure oxygen activated sludge system and catalytic oxidation for the exhaust stream.

 

[moltenmetal]

Wet scrubber & "plasma" and activated carbon. What do you mean by "plasma"?

The only way your non-methane hydrocarbons exiting an activated carbon bed are higher than they are at the inlet is if you are de-sorbing hydrocarbons from a fully saturated carbon bed. The only right way to use activated carbon beds is to have two in series with sampling between them. As long as the sampling frequency is high enough, when you see hydrocarbons at the sample point rise to something close to the inlet concentration, you have fully saturated the first bed and need to switch in a new bed.

Activated carbon, like any adsorbent, works best if it is fed a stream which is not saturated in water. If you feed a water-saturated stream to carbon (likely for any stream leaving a wet scrubber), you will end up with condensation in the carbon bed which will block some of the pores and cause you to lose sorption capacity.

If you have a steady 6g/m3 of hydrocarbons x 1700 m/hr, that's 10.2 kg/hr going to the carbon bed. That's a fair bit of material that you're wasting- and you'd probably need on the order of 50 kg/hr of fresh carbon to adsorb it. A sorbent bed system capable of on-site regeneration would allow you to recover some of that product and would reduce your disposal costs, in return for the cost of some energy. A thermal oxidizer would allow you to eliminate the disposal cost but would cost you even more energy as the feed isn't rich enough in hydrocarbons to totally fuel it.

 

[bimr]

Generally agree with moltenmetal's comments. I would go further and state that activated carbon is a poor choice for this application because of the low absorption capacity of carbon.

You really need to look into the use of the oxygen based activated sludge system that uses mechanical mixers instead of air for mixing. The cost of treating the air stream will be tremendous.

 

[Knowhow_]

Thanks for all of your replies.

bimr, I'm a little confused about your question. We just treat the gas collected from the sewage storage tank, and I don't think that there is any treatment of the sewage. Do you mean that we should add the water treatment with pure oxygen activated sludge system? Sound like control from the I also considered to use RCO, but I don't know much about the parameters, could you give me some advice?

moltenmetal, your detailed answer impressed me. "Plasma" here is non-thermal plasma, which uses electricity to produce ozone and some other strong oxidants to destroy VOCs. Do you think that non-thermal plasma is useful for VOCs abasement? I agree with you that the increase in non-methane hydrocarbon after the activated carbon bed could be caused by desorption. However, the temperature is just about 6 [sup]o[/sup]C, would the desorption happen under the condition? And we have set a demister before the activated carbon bed to decrease the moisture of the stream. 6 g/m[sup]3[/sup] of non-methane hydrocarbons is the inlet concentration of the whole system, and activated carbon is set at the end for repolishing, it's supposed that the concentration is much lower than that if the upstream treatment is effective. If we just use thermal oxidizer to treat the air stream, could 6 g/m[sup]3[/sup] of non-methane hydrocarbons enough for totally fueling it?

 

[moltenmetal]

Non thermal plasma does nothing in damp air in my opinion. I'm unaware of anyone feeding damp, contaminated air through what amounts to an ozone generator as a means to treat it, but perhaps someone is trying this. From my experience, if you want to generate ozone, you need scrupulously dry air or it won't form in the first place. Generating hydroxyl radicals directly in damp air is possible but I would imagine the yield would be terrible, i.e. it would take a huge amount of electricity to do meaningful destruction. If there is significant hydrocarbon vapour at the discharge of this non-thermal plasma unit, you've already proven that it isn't working. Note that ozone is a very, very expensive oxidant, and you're generating large quantities of VOCs- you'd need to make a huge mass of ozone to destroy them appreciably.

The colder the air is, the higher the sorption capacity will be, but if you feed air to a saturated carbon bed, it will desorb compounds from the carbon into the air until a new equilibrium is reached, i.e. the adsorption process is reversible.

If your air is at 6 C, your bioreactor is also running pretty cold...those poor cold organisms are probably not eating much, therefore letting the vapour leave pretty rich in "food", i.e. organic molecules that are volatile and which they might otherwise eat if they were happier. This is constantly a worry with aerobic biodegradation of volatile species in water- they are likely to operate as air strippers, shifting the problem from the water to the exiting air stream.

6000 mg/m3 doesn't sound like enough to fuel a thermal oxidizer even with good heat recovery, but you could check that with a heat of combustion calc and by talking to thermal oxidizer manufacturers.

 

[bimr]

Sorry, assumed that the "sewage" tank was a treatment tank. The terminology "sewage" tank is a bit unusual as are the comments about organics. Most industrial facilities keep industrial wastewater and domestic sewage separate, not combined.

Is it possible to take this vent to the aeration based at the wastewater treatment plant and sparge it into the basin?

 

[Knowhow_]

I'm afraid that there is no bioreactor in our system. But I also considered treating the gas with biofilter or biotrickling. I'm worried that the inlet concentration of non-methane hydrocarbon is too high for bioreactors to treat, and the efficiency could be very low. And I'm wondering that if there is any bacteria that could degrade non-methane hydrocarbon, such as benzene, effectively. Thanks for moltenmetal's suggestion.

Sorry, I didn't notice the terminology. bimr, you inspired me, I've never think about that method. Do you mean that using the air stream as the aeration in the basin? And combined with moltenmetal's idea, I think that there should be some bacterial, such as activated sludge in the basin, or the volatile compounds would volatilize to the air again when sparged into the basin.

 

[bimr]

Regarding: Do you mean that using the air stream as the aeration in the basin?

Probably the least expensive method unless you have a long piping run.