Non-condensing method for steam recovery from exhaust gas?

[Sheps75]

Cyclonic separators can be used to collect water in after-treatment when hydrocarbons are burnt, but (for one thing) this expends energy in cooling the walls of the cyclone.

Is there a means to, say, use a similar effect of centrifugal force to concentrate the water molecules near the outer wall of a device, but without having to condense them before (as in the cyclone) gravity does it's bit and droplets form.

I had considered that the inner surface of the separator might look more toroidal than conical, but employ some effect similar to osmosis where a section of the device's wall acts as a micro-sieve; letting steam but not (large-amounts of) CO2, etc through.

I'm interested in keeping the steam recovered from the gas flow as hot as possible, and I also don't care how clean it is (acidification, etc isn't a problem).

I'm designing a system for onboard gasification of torrefied wood to be used on a lorry/truck, if you're wondering...

 

[moon161]

For starters, compare the molecular weight of water, (18) with that of O2 (32), N2 (28), C02 (44), simple Hydrocarbon CnH2n+2 (only methane vapor is lighter than water vapor) N0x, etc. Water vapor is lighter than almost everything in flue gas and intake air, so it won't go to the outside of a centrifuge. I expect the only way a centrifuge will pull water out is as droplets of liquid after phase change.

 

[140Airpower]

Also, separating gasses by centrifuging is extremely inefficient when they are cold and gets exponentially worse when hot. But, it's very easy to separate out water droplets by centrifuge. Otherwise, a cyclonic separator is simply collecting condensed water on its outer surface.

 

[ivymike]

I was going to say something like the other two, but then I noticed the molecular sieve question... I don't know a reason why that wouldn't work, offhand, except that Gore-Tex doesn't stand for the temps.

 

[140Airpower]

I should have said, "Otherwise, a cyclonic separator is simply collecting condensed water on its inner surface".

About Gore-Tex, the molecular sieve principle will work with high temperature materials coated with Teflon, but at high temperatures, you will not get liquid water.

Recovering water from exhaust can be done with a condenser. It will be bulky and may require energy to circulate a cooling medium. This looks like the same problem with steam engines. Recovering water is no problem for stationary installations, but probably not a good idea for a car.

 

[rmw]

If there is moisture on the inner surfaces of the separator, it will collect ash from the wood combustion and soon plug up. With experience with lots of wood combustion - albeit none was torrefied per se (some was essentially bone dry however) - the last thing one wanted was moisture of any kind in the back end of the process.

rmw

 

[patprimmer]

I can't see the rate of transfer with a molecular sieve being anywhere near enough to be practical. I am unaware of effective molecular seives that could tolerate prolonged exposure to exhaust without being damaged one way or another.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[Sheps75]

Thanks for all the responses. And "D'oh" on the atomic weight of H20!

Thanks -ivymike- for translating my gibberish to mean 'molecular sieve'.
And thinking of 'Gore-Tex'.

Thanks also -140Airpower-.
Re: - "About Gore-Tex, the molecular sieve principle will work with high temperature materials coated with Teflon, but at high temperatures, you will not get liquid water."
- That's not a problem, as I don't want Li-H20, I want steam; ideally SH steam.

Thanks -rmw-, however any ash is to be filtered out after gasification (so, prior to combustion in the engine). For this reason I shouldn't think the H20 separator stage will be exposed to all but the tiniest amount (and size) of ash particles.

-Pat-, what is it in the exhaust that is most likely to damage the sieve? Thanks.

 

[patprimmer]

Well my thoughts where temperature, pressure and carbon and acid fumes, especially with a backfire.

Molecular seives are very fine by definition and require very large surface areas to process large volumes. It takes very little contamination to clog them so the need regular cleaning. Also all I have seen would need a very well designed back up to give back fire pressure level resistance. Teflon is good to about 400 deg C at which point it starts to break down, eventually producing hydrogen fluoride which is a VERY nasty chemical. ie it eats even glass, but more frightening, human bone marrow.



My only experience is in a material developed for kidney dialysis and the technology was modified and converted to water treatment, clothing , wound dressings, contact lenses and stay fresh food packaging. They where all thermoplastic and depended on mixing non compatible materials at the molecular level of dispersion then dissolving out one component. Different raw materials gave different diffusion rates for different fluids.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[140Airpower]

Sheps75, you can capture water from exhaust gases simply by cooling it, but capturing gaseous water -that is steam- simply and efficiently and cheaply cannot be done mechanically. Centrifugal force to separate heavy molecules in solutions requires tens of thousands to hundreds of thousands of G's and time. To separate light, highly energetic and mobile gas molecules might require millions of G's and/or a chain, perhaps thousands of stages long, where a minute degree of separation is accomplished at each stage, all needing a large plant and a lot of energy.
The other approach, capturing the CO2 does not leave any steam as far as I know.
Are you just trying to save some heat?

 

[Sheps75]

Thanks both.
Maybe there is a material that can act as a Molecular Sieve but can survive slightly higher temperatures. Cooling the inlet air with evap. cooling could be a means to control the exhaust temps, but this would cause atomisation problems (unless we switch to gas fuels, propane, etc = complications).

To reiterate: I want the H20 as steam, not water. Hot as possible.
The principle is to drive a Water-gas reaction that gasifies a secondary (woody biomass) fuel. Around half the SH steam needed would ideally be recovered from exhaust gases following the in-cylinder combustion process; greatly reducing the size of water tank carried, and also reducing the size of an exhaust-mounted heat-exchanger needed to heat to steam.

 

[ivymike]

how about using a closed-circuit exhaust gas cooler system to condense the water, and using exhaust heat upstream of that cooler to reboil the water? That'd reduce the size of the exhaust cooler while still providing fairly hot steam from recovered water.

 

[Sheps75]

Re - closed-circuit exhaust gas cooler: Yes. I had wondered if an approach along those sort of lines was possible, but decided it must go against the laws of thermodynamics (, or something!).

Now I'm not so sure...
So the exhaust temp after the manifold would initially drop as it transferred heat to the water (the water that will be condensed further down the exhaust) and steam was (re-)formed.
Then the exhaust temp would tend to rise again as it 'took in' the excess heat from the condenser circuit.
Finally (for the exhaust gases, prior to exiting the tailpipe) the temps are dropped below boiling point to condense out the water.

The system would need some additional power for pumps, etc and probably an additional air-cooled condenser stage to return enough cold fluid to the E-gas cooler, but it might just work! .... Thanks!!

 

[140Airpower]

Sheps75, No violation of thermodynamics. I thought your objection to the condenser route was for other reasons, like bulk, weight, etc.
The hot exhaust gases include mainly N2, water vapor and CO2. A boiler using exhaust heat would absorb heat sufficient to re-boil water condensed from the exhaust. The boiler itself should condense some water from the exhaust flow and a downstream air-cooled radiator could condense most of the rest of the water. Some would be lost.

 

[Sheps75]

Thanks; although weight and bulk are still a seriously major hurdle, this appears to be a design that could actually be built and integrated with a larger IC-engined vehicle, lorries/artics, buses, etc.

Getting something over two-thirds of the condensible liquids recovered would appear to make the extra weight of the gear worthwhile on a long haul - based on a quick back-of-the-envelope analysis.

 

[ivymike]

one problem with your description

So the exhaust temp after the manifold would initially drop as it transferred heat to the water (the water that will be condensed further down the exhaust) and steam was (re-)formed.
Then the exhaust temp would tend to rise again as it 'took in' the excess heat from the condenser circuit.
Finally (for the exhaust gases, prior to exiting the tailpipe) the temps are dropped below boiling point to condense out the water.

3 separate paths to consider:
Exhaust
The exhaust out of the manifold would cool as it transferred heat to the water in the boiler (correct)
Then the exhaust temp would fall further as it passed through the exhaust gas cooler (condenser circuit), and temps are dropped below the dewpoint to condense out the water
The exhaust exits at this point.

Recovered water
The condensed water is pumped through the boiler and boils to make the steam

Cooling circuit for exh cooler (may be combined with engine jacket water circuit)
The cooling water for the exhaust cooler passes through the exhaust cooler where it is heated by the exhaust (but remains below boiling)
it passes through a radiator where it is cooled back down
it is pumped back to the exhaust cooler

 

[GregLocock]

Sheps75, now you've got hot liquid carcinogenic acidic water what are you going to do with it? Inquiring minds are interested.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[Sheps75]

Thank you -IvyMike- for expanding the my description; making it more legible in the process (I tried to sum-up all three ccts in on paragraph).


GregLocock asks "Now you've got hot liquid carcinogenic acidic water what are you going to do with it? Inquiring minds are interested."
Earlier I said: "[The steam is recovered to] drive a Water-gas reaction that gasifies a secondary (woody biomass) fuel." and "I also don't care how clean it is (acidification, etc isn't a problem)."
...More specifically, this "woody biomass" is wood torrefied at up to 320degC (similar to charcoal but with more of the wood's energy conserved), which is then powdered.

The job of the (now) superheated steam is to drive the above WG process and create a stream of syngas (mainly CO and H2 & some CO2s,CH4s etc). The acidification shouldn't be a problem, unless a catalyst is involved (which it might need to be-?).
The nice thing about this idea is that the 'thrown-away' exhaust heat is recovered by allowing enough heat (primarily, the steam's) to push the reaction along, without the need for oxidisation of the fuel. That said, the system will probably need a conventional fuel whilst reaching temperature (or Torrefied fuel used up, etc); but depending on the chemistry and the heat-flows some additional Hydrocarbon is expected to be consumed in normal running.

NB: This is only a Proof-of-Concept at this stage I might add!... Donor vehicle for which would be an old petrol Landrover. But it's really imagined to best suit bigger vehicles, as mentioned previously.

 

[Sheps75]

Thanks again for all the feedback and advice on this design, and thanks for being so constructive in the tone also.

But now I've got a nagging feeling that, having explained my design more or less in full, the unspoken consensus out there is this is, well, 'out there'! So, as a final ask: Is anyone willing to offer me there most critical perception of this approach? I'd like to hear why it can't work; and if not definitely can't work, why it would make no economic sense, etc. The only aspect I'd rather not discuss in detail at this point is safety issues (not dismissing them, but); unless those are linked to adding a great deal to cost or weight/size of the gear involved.

Much appreciated!...