remove 0.04 mole/s water from 906 mole/s stream

[ahp]

There is a stream with this composition : (mole/s)

Methane 344.71
Acetylene 3.77
Ethylene 315.31
Ethane 2.23
Propylene 113.80
Propane 119.97
Butune 6.67
Butylene 0.14
Water 0.04

I wonder if there is a way to remove the water ?
On my book, it is written that I should remove the water before send the stream to separation-train.
I can't find the reason for this.
Does anyone know anything ?
Thank you.

 

[TD2K]

Based on that analyses, you have about 45 ppmv of water. The only way I can think of to remove the water further would be molecular sieves which would get you down to 1 ppmv or less. However, your water content is about an order of magnitude less than typical inlet water loadings to mol sieve beds that I'm familiar with.

As to why the water is supposed to be removed, that's impossible to say without more information. If the downstream separation train is a cryogenic chill-down process such as in an ethylene plant, it would be to prevent the water from freezing out and plugging equipment and/or hydrate formation but I'm guessing.

 

[ahp]

TD2K,
You're just right about the water, it is an ethylene plant. If the water would freeze and plug the equipment, what is the minimum amount of water that is allowed ?

ops, nearly forgot that there is also about 300mole/s of hydrogen in this stream. I'm a little confuse in designing separation-train with the presence of H2. Do you mind giving some information on this ?
thanks

 

[TD2K]

Is this a conceptual design or what?

The analyses you have is almost certainly off a PFD or process simulator. With the additional H2 you mention, you have a water content of about 30 ppmv. What stream is this for and what pressure and temperature is it at? As a first pass, you can look at the partial pressure of the water vapor and see what temperature that corresponds to off steam tables (it's rough but gives you an idea).

Ethylene plant chill down trains operate at about 500 psig (3500 kPa, maybe somewhat less, it's been a long time ago) or at least, the one I worked in did. 30 ppmv water at 500 psig is about 0.015 psiA or -50C dewpoint. At that point, you've already will have dehyrated the gas long before this point simply to avoid hydrates which will likely start forming around 40F to 60F I would guess at this pressure.

Are you seeing suspected freezing problems or is someone just saying '100% of the water has to be removed?'.

Our molecular sieve beds were supposed to be removing water to less than a ppmv. I'm not sure what the exact number is, moisture measurements at that level are very difficult to get accurately. We did monitor the outlet water but used it as an indication when breakthrough was starting (water levels start to rise quickly at breakthrough) and we switched the beds at that point rather than letting it reach a specific value.

 

[Samiran]

Dear ahp,

Minimum water requirement should be back calculated. For e.g. if the cryogenic temp for ethylene is -x C then the water dewpoint of the gas at the inlet to the cryogenic part must be below -x C. Otherwise you may get water condensation. Hence, look up in tables to get the dewpoint (ppm of water then can be calculated) at -x and pressure P.

 

[ahp]

TD2K and Samiran,
Thank you for your replies. Those are help me.