Toluene/Water Solution 1

[Elpmek]

Hi all,

I'm looking at concentration a water wet toluene solution by distillation. I have about 1500L of toluene which I need to get down to about 100L. I understand the solubility of water in toluene to only be about 0.03% w/w (my toluene will be saturated with water) but I'm looking to do a constant volume distillation so would like a Txy curve to work with? Any idea how I might get this for my system?

 

[georgeverghese]

Try searching for this in DECHEMA, Chemistry Data Series, Volume I subsection for binary aqueous VLE data. Presume you are aware this binary forms a 2phase min boiling azeotrope at 84egC, 1atm at 56mole% water.

 

[MortenA]

What is you water content now? For Touluene its not like e.g. alkohol. The two liquids will seperate into destinct phases that should be possible to seperate (im stating something a bit basic here i know but i cant just gauge how deep your initial knowledge on the subject is). Would it be possible for you to seperate the two phases by gravety and then use destallation for the rest? DECHEMA is not free but if its out there you will find it there. Most common process simulator will have these data (VLE) build in.

 

[Elpmek]

I will have already done a phase seperation from water but there is some residual water left in the toluene because it is very slightly soluble. I've used ASPEN to try and get a VLE but the curve I get is all over the place, definietly not correct.

 

[moltenmetal]

Generally you don't do this kind of a purification by distillation. A solid phase dessicant/sorbent will be more efficient than trying to distill water from toluene.

 

[Elpmek]

I'm not looking to seperate the water and toluene, I don't care how much water is in the final solution, I'm simply looking to concentrate the solution from 1500L to about 100L. I need the VLE data to set the temperature for the distillation and probably model a constant volume distillation since the reactor we have available for this is small than 1000L.

 

[moltenmetal]

Simple: ignore the water. At that concentration, the water will not matter at all- it will "steam distill" with the toluene, adding an insignificant amount of enthalpy to the overall distillation.

 

[Latexman]

I'm confused. You have 1500 L of 99.97% toluene / 0.03% water solution? And, you want 100 L of the same solution? That's what it sounds like you want to me. If not, please explain.

Good luck,
Latexman

Technically, the glass is always full - 1/2 air and 1/2 water.

 

[Elpmek]

Yes, I want to concentrate my toluene solution down to 100L, but I don't care what the composition of the final solution is. I have a molecule in the solution so I just want to reduce the volume of the solution, the amount of water in the final solution doesn't matter for any downstream processing.

Having thought about it I guess it's obvious that any water that is present will come off at the toluene/water azeotrope temperature of ca. 86°C.. Once all the water is gone the dry toluene will start coming off at 111°C or whatever it's BP is??

How would I go about finding the water/toluene azeotrope boiling point at reduced pressures? Say 0.2bar?

 

[ornerynorsk]

By definition, one cannot further concentrate a (for practical purposes) a pure solvent. You are only going to end up evaporating 1400L. Or, if you fully recover the distillate, you will still have 1499-1/2 liters.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[Elpmek]

I have a molecule in the toluene solution so I am trying make it a more concentrated solution in terms of g/L.

 

[Latexman]

What are the characteristics of this "other molecule" (OM)? Compared to toluene does OM have a low vapor pressure, high vapor pressure? Is it soluble? Density higher or lower than toluene? You have to give us something to go on.

Good luck,
Latexman

Technically, the glass is always full - 1/2 air and 1/2 water.

 

[Elpmek]

The OM is a large steroid molecule which is very soluble in toluene, it doesn't crytallise out until we reach about 20L of toluene in this process so we don't need to be concerned with it. I only mentioned it to explain that I'm not trying to concentrate the tolunene but I'm trying to concentrate the OM solution.

 

[Latexman]

To me "large steroid molecule" = low vapor pressure compared to toluene. Do you just boil off the toluene/water or do you have a rectification column? Do you recycle the toluene for latter use?

Good luck,
Latexman

Technically, the glass is always full - 1/2 air and 1/2 water.

 

[Elpmek]

Yeah I mean the OM is a solid with a very very low vapour pressure, negligable compared to toluene. We are just boiling it off and not recyling anything.

 

[StoneCold]

Elpmek

So you want to feed strip your solution to concentrate the large steroid molecule? That's not very complicated, if you are doing it at atmospheric pressure it probably boils around 110C. Fill the vessel about 80% full, heat it up. Continue to fill the vessel as toluene is boiled off. You may want to do it under vacuum as organic molecules tend to break at higher temperatures. The operation is the same, you just distill at a lower temperature. Is there more to the question that we have not covered?

Regards
StoneCold

 

[Elpmek]

But it doesn't boil at 110°C because of the water/toluene azeotrope at 85°C. All I want to know is where I can find out what the azeotrope boiling point is at reduced pressures?

 

[Latexman]

For a rough estimate, on a vapor pressure curve for toluene, draw a point at 85 C and 1 atm. Now, draw a curve through that point that is parallel to or equidistant from the toluene vapor pressure curve.

Good luck,
Latexman

Technically, the glass is always full - 1/2 air and 1/2 water.

 

[StoneCold]

Elpmek
You know for a fact that you form the azeotrope in your system using a single stage flash to concentrate the steroid? Or are you using a multi-stage distillation setup and you are getting the azeotrope for a few minutes? Either way you still just distill the toluene/water overhead. No issues.

Regards
StoneCold

 

[georgeverghese]

Elpmek, From memory, there is a procedure in the subsection "Prediction of Azeotropic Data" in Perry Chem Engg Handbook, 6th edition that you may find useful for hetero azeotropes such as TOL - water. I've checked out this procedure myself for a few cases and it seems to work well.
Sadly, this entire subsection has been expunged from the 7th edition.