Caustic Scrubber Design 1

[OilBoiler]

Hey everyone,

Right now, I need to design a caustic scrubber for our fuel gas. I already have the design flowrates for the gas, temperature, pressure, and H2S concentration. From that I calculated the amount of caustic needed, since the molar ratio of H2S to NaOH is 1:1 and then converted it to 25% caustic.

Anyway, I'm good until that point, but I have a question about how to do the rest! We are going to use a vessel that is already in place and out of service.

I'm guessing what I need to do is figure out whether this vessel is appropriate or not and probably need to design the packing. What kind of parameters do I need to calculate to give to the packing vendors? Gas velocity? Do I need to go ahead and calculate NTUs and all that good stuff? I also have to spec out demisters. How can I approach that?

And somebody mentioned water wash for a caustic scrubber, is there a reason why we would want water?

I just wanted to get some feedback and if you have designed these, based on your experience if you wouldn't mind giving me some pointers.

Like always, your feedback is very much appreciated!

A newbie.

 

[mbeychok]

OilBoiler:

This is not a direct answer to your basic question. I just want to offer some comments:

(1) Using caustic to remove H2S from fuel gas essentially means that the process is semi-batch ... since you cannot regenerate the spent caustic by removing the H2S and therefore the caustic must be replaced when spent. Thus, you should consider having two caustic scrubbers so that one caustic scrubber (with fresh caustic) is in use while the other scrubber (with spent caustic) is being emptied and recharged with fresh caustic.

(2) A water scrubber following the caustic scrubber is a good idea to remove any caustic carryover with the scrubbed gas. Demisters should not be counted on as 100 percent insurance against carryover during normal operations as well as during upset conditions.

(3) More importantly, why are you using caustic in a non-regenerable semi-batchh process rather than using a regenerable process such as a monoethanol amine or diethanol amine system? Caustic is quite expensive and the spent caustic constitutes a quite difficult waste disposal problem.




Milton Beychok
(Contact me at www.air-dispersion.com)

 

[AndreChE]

OilBoiler

The caustic scrubber is used to absorb the sour gases like H2S and CO2.

To design this, you need to consider the system as an absorption column. For that you need to calculate how many equilibrium stages you have. Do not forget you have , at least, one chemical reaction (H2S + 2NaOH --> Na2S + 2H2O) if your stream doesn't have CO2.

Your system, for sure, will not have 100% pure caustic. In our scrubber we use 50%. Water is added to the top of the scrubber in order not have corrosion and to dilute the caustic. You need to have caustic recirculations, otherwise your absorption will not be effective. Normally, you have two recirculations in the bottom section of the column.

You can have a continuous process without problems but don't forget the effluent: Na2S. Can your water treatment deal with such amount? I just finished to define a Process Design Basis for a Wet Air Oxidation to oxidize all sulphides to sulphates.

Another issue is pH (another reason for water). Don't forget to take into account the neutralisation of the effluent. Before this, we have pH = 12-13 and at this level, all the biological treatment of your waste water... use CO2 or H2SO4. We had H2SO4 but the pH was very difficult to control. With CO2, the reaction is very smooth.

Demisters, you should think on that significantly. We have two.

Another topic: Amines, why not? It is also often used...

Important paper you should read:

"Optimize caustic scrubbing systems", Hydrocarbon Processing, May 1978, Marcello Piccioti

 

[roker]

hi,

we use caustic for scrubbing HCl from a gas stream, the lower part of the column is a caustic redervoir since the operation is semi-batch. the caustic is circulated at the lower packing while water is circulated at the upper packing both packings in the same column, when the caustic is at low concentration the spent caustic is emptied to such a level that caustic circ can still be operated and a new batch of fresh 20% caustic is filled to the column.
but for H2S treatment with caustic you have some ractions in parallel also NaOH+H2S <--> NaHS+H20 IN EQUILIBRIUM,
by the way NaHS is used for floatation of Cu in mines.
so it is recommended to use amine for H2S treatment.

regards,
roker

 

[sshep]

Hello OilBoiler,

This should be a fun job for a newbie, but is actually quite simple and I am sure you will learn alot. As you have a vessel already your diameter and height is known. You appear to have calculated your theoretical caustic consumption based on a blowdown concentration of 25%. You are ready to go:

Gas Load: The massflow is determined by your fuel gas flow. The packing vendor will need this and the density (plus temp, pressure, vapor viscosity, and a few other basic data). Before contacting a vendor the superficial gas velocity should be calculated from the vessel diameter. If this is a reasonable velocity then the vessel may work, if not then the vessel is too small and you shouldn't waste any effort evaluating it further.

Liquid Load: The vendor will need this. It seems that you are envisioning a single circulation loop. Multiple loops running at different caustic concentrations are used to conserve caustic. Set the circulation rate based on a reasonable irrigation (gal/sq-ft) of the tower diameter. The incremental operating and capital cost of a generous flow will be minimal so don't choose something too small. The packing vendor will also design the top distributor.

NOTE: the use of a water wash is to protect downstream equipment from salts. This is especially critical in an ethylene plant (i.e. you may have a centrifugal compressor downstream), but may not be important in your case and will add extra cost to your design.

Estimating the H2S removal: This is a more difficult subject where you can use the methods described in the cited paper. A process simulation will help. See also my post in thread798-110363. You know how much height you have to work with and what your top H2S spec is (you didn't give it). You have already set the concetration you plan to use although the concentration to the top of the tower could be a bit higher as a result of judicious use of the make-up caustic.

Miscellanious: The location of make-up, blowdown, and sizing of pipes, pumps, control valves, etc is probably straightforward. I see you making enough posts to believe that this will not be a problem for you. If you need guidence on reasonable superficial velocities and irrigation rates this forum can help, or take a look through Kister, or Perry to see what is typical.

Environmental: The disposal of spent caustic is not a trivial concern. Simple nuetralization will cause the caustic to liberate H2S. We are currently using wet air oxidation to destroy the spent caustic in one of our units, but this requires significant extra investment. You can pay for deepwell injection which is not nice or cheap either. I am sure there are other options, but in any case this may be a more difficult issue than any faced in your scrubber design.

Best wishes,
sshep

 

[bewdley]

Hi OilBoiler,
I have spent some time in H2S/NaOH contactors, H2S coming from desulphuration. The problem is not the H2S/NaOH, but the probably huge amount of impurities due to the non-H2S compunds entrained. Therefore, the design of the column shall be foul-proof as much as possible. Consider that contact time is an important parameter, and is affected by temperature - you indeed need a tempearture control.
The efficiency of the system in terms of H2S removal, if correctly designed, is good - we were constantly below detection as residual H2S, and even the alarm threshold at 4 ppm was never reached at the stack.
Consider that, if the ratio is correct, you can obtain a sellable product from this reaction (NaHS, used in tanning industry); the disposal problem, in this case, concerns the sludge due to the impurities above. This is in solid form, and we used to send it to a dump. Difficult to estimate the volume, not knowing the amount of impurities.

 

[siretb]

Yes, stoechiometry is 1 to 1, but, as already said your gas will probably contain something else.
if it's CO2, although the absorption of CO2 into caustic is a slow reaction, it will grab some. in a 1996 paper by Al Wohoush ,where the system studied contained 1% H2S and 10%CO2 about 5-7% of the CO2 was scrubbed, consuming caustic.
The other issue is the VLE and H2S - HS- S-- equilibrium. You need fairly high pH if you do not want a residual partial pressure that is significant. For these reasons, if you are still decided to proceed with a caustic scrubber, you'd need
1) to run a simulation (HYSYS ASPEN ....) , forcing some CO2 absorbtion if you have CO2 and chech that the pH profile if OK, and that the equilibrium partila pressure of H2S is OK. You get feed rates and bleed rates.
2) select a packing and tower diameter. A reasonable choice, is to operate at a F loading (F=U * sqrt(rhog) ) u in m/s rhog in kg/m3) of around 2. you'll find good packings.
3) ask the vendor for the HTU. It will be probably larger than 0.5 m for this system. You get the bed heigth.

As someone quoted, fouling, scaling, ... is a MAJOR issue with this system.
My opinion is that caustic scrubbing of H2S should be used only where only a very small quantity of H2S is to be removed (high loading --> MEA or other amines); yet another optiion is to scrub with bleach.

 

[121202]

Oilboiler,

If you want a packing or internals supplier to provide you with a solution, which is highly recommended, you need to produce a detailed process datasheet. The datasheet\specification should have the following details as a minimum:-

- Stream descriptions and compositions
- Stream flowrates, operating pressures and temperatures
- Stream physical properties
- Vessel design pressure and temperature conditions
- A vessel sketch showing all of the key dimensions and approximate locations of all nozzles.
- A simplified process flow diagram
- A reasonably detailed process description
- Performance requirements and process guarantees (i.e. H2S specification to be met and the maximum permissible carryover)

The alternative is to get a design house involved to develop all the specifications and drawings for you.