System for solvent selection 3

[Jonas82]

Is there anyone that has any tip about a system/method for selecting the optimum solvent for an extraction process?

Thanks for your help!

Jonas

 

[25362]

If meant for an industrially-sized plant: cheap, available on short notice, easily recoverable for recycling (e.g., a lower BP than extract and raffinate), chemically and thermally stable, non toxic, non corrosive, selective at ambient temperatures for the specific extraction purpose, soluble (to reduce needed dosage), easily separable at the extraction step by density differences, bio-degradable, low latent heat of vaporization, available commercial knowhow, sometimes it is desirable that the solvent has a low boiling heterogeneous azeotrope with water. Did I forget to mention any other desirable characteristic ?

 

[epoisses]

It's not easy if you start from scratch as I read in an earlier posting of yours.

My 2 cents:

1. Round up all potential candidates, i.e. all solvents that could do the job on lab scale.
2. Eliminate those that do not meet basic boundary conditions in terms of handling (toxicity etc), availability, stability etc (25362's post).
3. Define the process units and utilities required for each of the candidates (extractor, stripper, regenerator, etc). Consider using advanced techniques e.g. divided wall columns to reduce investment.
4. Define dimensions and process conditions of each of those units using simulation software - for each solvent.
5. Then estimate investment and operating cost and select the most promising candidates to be studied in further detail. You will note that you will quickly lose interest in e.g. solvents that cannot be sufficiently well recovered or solvents that require very high energy input.

But before starting from scratch, make an exhaustive list of commercially available processes, list the advantages and disadvantages of each and use this as an indication of which way to go.

 

[BaAn]

For rounding up potential solvent candidates you could contact the Dortmund Data Bank. They provide thermodynamic and transport properties of pure components and mixtures as well as special programs for data correlation, estimation and process synthesis. Part of their software package is a tool which allows selection of solvents for an extraction process based on in-house measurements of solubilities:

http://www.ddbst.de/new/Default.htm

However, the software cannot provide answers to all the other relevant issues pointed out by 25362 and epoisses.