Best way to cool down hot gases with particles 2

[nbog]

I have hot gas stream (H20, N2) from reactor. T=~1000 C. It contains fine particles (product) that go to filter. I have to cool it down but avoid condensation. I am thinking about quenching with water and air but that will just add more water to the system. I need advice on which way to go:
- I still have to calculate where it would be from steam/water eqilibria line if I add quench water but I just have a feeling that I will need exchanger. What type of exchanger is best suited for this type of application?
Thanks,
Nick

 

[MortenA]

Have you thought about cooling in a fluid bed (cooling with cold air)?

NIRO is one of the well know fluid bed manufactors. Look at this brochure that cointains (among other things) info on fluid bed cooling

http://www.niroinc.com/html/chemical/cpdfs/BNA 500-GB Fluid Bed Proc.pdf

Best regards

Morten

 

[speco]

Nbog,

I have used air-cooled exchangers for similar applications, such as flue gas containing some fly-ash loading. Exchangers could used either finned or bare tubes. Vertical tubes work well, allowing particuates to pass through without accumulating in the tubes. Cover plates are a good idea for cleaning. This usually involves large diameter (around 1.5") tubes to minimize pressure drop.

It's another way to go.

Regards,

Speco

http://www.stoneprocess.com

 

[nbog]

Morten, my particles are ultrafine (microns), so I am not sure how fluidized bed will work. Sory I didn't provide this info in my post.

 

[Montemayor]

Nick:

Your query sends everyone scurrying every which way and coming up with dead ends, guesses, and up blind alleys simply because you don’t furnish ALL the basic data and scope of what you are doing. Let me cite what is confusing or giving problems:

1. 1,000 oC is not a common, expected reactor outlet temperature. This hot a temperature requires a lot of special know-how and background as to the application, fluids involved, pressures, and the process itself;
2. You say you have to cool down the exiting gas stream, but avoid its condensation. How can you consider that you are going to condense steam (and much less N2) at 1,000 oC? Without knowing the actual composition and pressure of the binary we can’t tell what would be a safe temperature to quench down to without causing condensation. But you can. And yet you don’t state those conditions. It’s hard to imagine that you don’t have a super-superheated vapor that can tolerate a lot of additional cooling without condensing the steam portion. But we can’t do anything but ponder.
3. You want to quench with water or air, but that will add more water to the system. You never stated that you can’t dilute the stream. If so, why don’t you give us that scope of work? Also, how can you consider adding air? Doesn’t that also dilute (or contaminate) the stream? Since we don’t know what you are doing, it’s next to impossible to suggest or recommend anything.
4. You say you want to consider an exchanger (at 1,000 oC??). Forget about a TEMA design. You’re talking about what everyone does in these type of circumstances: use a coil-type of apparatus. The expansion compensations are astronomical and a pipe coil (or tube) arrangement is about the only thing that will withstand the movements. The material of construction is totally another hard issue. And the fouling (and plugging) factors are going to be a nightmare.

One of the scary issues on this Forum is that we don’t know you personally or your qualifications. We don’t know if you really understand and dominate the issues at hand and whether you can apply any recommendation or suggestion safely. When we don’t even get all the basic data and scope of work, this makes it doubly dubious as to whether or not any hint or suggestion might hurt somebody – physically or economically. For any professional engineer this is a very important issue. It seems that every time we get this type of query that is deficient in basic data and scope of work, the storyline starts to change as the thread moves along. The “rest of the story” doesn’t come out until later, when everyone realizes that a lot of time, effort, and brain cells have been wasted or spent erroneously on the wrong subject.

I recommend you to tell us all of the application, all the basic data, and what your scope of work is. Perhaps we can go from there. This is not only the safest, most efficient route to take, it is also the most common-sense way to approach any problem.

 

[nbog]

Morten, thanks for your critics. I appologize.
Here is the process (pilot plant) I am looking at:
Metal salt solution is dispersed in a chamber (operating slightly below atm pressure - use fan to get the gas out). Heat is provided by torch. Salts are converted to oxides. Particles + steam + torch gases are output from the chamber. I have to collect particles.

I am looking at water flow rates 1-5 lit/min. Torch power is 50-150 kW. My calculations for 1.0 l/min of water show that to get to 1000 C at this conditions (to convert salts to oxides), I have to provide around 72 kW at the torch.
Now I have to do something with exit stream before sending it to filters.

 

[Montemayor]

nbog:

Thank you for the great cooperation and the data. Now everyone has the same, accurate description to sink our teeth into and come up with credible and constructive suggestions / recommendations.

 

[nbog]

Yeah, forgot to add that you can assume torch gases (N2 for simplicity) to flow at 200 l/min std.

2 ways came to my mind:
1. Add quench water to bring the T down and eventually, if not good enough, add some air to dilute the stream to keep partial pressure of water above the saturation line
2. Go with exchanger which requires far more capital investment and operating costs than quenching, but I just wanted someone else look at it in case I missed something.

 

[rzrbk]

No new cooling options, just some thoughts.

The choice between the above two options seems an economic one, with the choice being one or more exchangers for indirect cooling vs. a bigger fan to handle the added steam/air. Assuming contamination, and materials issues could be handled in either option, I would lean toward an exchanger, which would allow a more gradual cooling versus the basically instantaneous change from 1000 degC to say 150 degC.

If low-pressure steam is available as a coolant, this would solve many of the fouling/expansion/etc issues with using a liquid, and should require a much smaller exchanger than using air. This would also automatically limit the quench temperature to above that where water would condense.

A second exchanger after the filters would allow the steam to be condensed out, greatly reducing the size of your fan.