I have a solution of water and oil mixed to as much as possible, since they won't actually mix. The supply is coming down a pipe at a relatively fast rate and I need to separate it so that the water goes down a pipe and the oil goes through another pipe.

How would this be done, without waiting days for it to occur?

Diagram of what I want.

  • $\begingroup$ If you mix oil and water really well, you get something that looks like mayonnaise. It could take years to separate. $\endgroup$ – LDC3 Apr 12 '14 at 16:10
  • $\begingroup$ Maybe if you add a highly polar salt in large quantities it will help break the emulsion? Dump a bunch of sodium chloride in the mixture under gentle stirring and see if it separates faster. You could also possibly selectively filter out one of the liquids by forcing the mixture through a highly polar or non-polar membrane with tiny pore size, but I don't think it would work perfectly in a single stage. Mere heating also promotes the separation of emulsions, I believe. $\endgroup$ – Nicolau Saker Neto Apr 12 '14 at 16:31
  • 1
    $\begingroup$ Dairy farming: cream separator. $\endgroup$ – Uncle Al Apr 12 '14 at 22:17
  • 1
    $\begingroup$ Graphene oxide $\endgroup$ – julien Apr 11 '16 at 17:59
  • $\begingroup$ I'll just point out that for any industrial type application there are bounds to be laws about the amount of contamination in the waste water that would be acceptable. You'd probably be required to do some sort of sampling to insure that the process stays in control too. $\endgroup$ – MaxW Apr 12 '16 at 2:20

Oil-water separation is a big money industrial process, and there are a number of methods for successfully separating the two. So the answer to your question is best approached in two parts: a theoretical answer, and a practical answer.

Theoretically, yes. A number of systems exist and the best are described below. There are off-the-shelf applications, as well as major capital infrastructure for major water treatment plants. Whether these solutions are for you will depend on your application, as well as budget. Hence, the practical answer may be different if you are looking for a cheap, backyard-employable device to solve the problem created by Granny filling the radiator through the sump oil inlet again.

Wikipedia has a very nice summary of water treatment, and includes a lot of information about the API (American Petroleum Institute) oil-water separator, which has been in effective use for decades. It is a very effective device, albeit costly, and can treat upwards of 50000 litres of water per hour. You can read about it in detail here, but I don't doubt it is what you are after.

Something more in the lines of what you describe is the Vertical Gravity separator. It is a pressurized system which promotes separation through convection mixing as the liquid is pushed though a complex path. The differences in liquid densities encourages separation of the two components. VGS systems can treat up to 3000L per hour. A document describing the operation of this system can be found on a link to a pdf document on this page. There are some problems associated with this type of device, in that the 'filter' mesh can get blocked if the material is very dirty.

Of course, any treated water is not 100% pure, and may need additional treatment depending on your requirements.

Overview of VGS for oil-water separation

There are a number of interesting web resources where you can read more about oil-water separation methods. Pielkenrood have an excellent page of various methods, past and present. This blog also has a number of useful pages, although it is a commercial site promoting their own preferred devices.

  • 1
    $\begingroup$ Well then, it turns out there was significantly more depth to this question than I had first assumed. Thanks for this answer. $\endgroup$ – Nicolau Saker Neto Apr 12 '14 at 22:15

Membrane technology can certainly separate immiscible solvents from water. One commercial (lab) solution is from Zaiput, but other technologies are available. Another option could be centrifugal separation. If you look up the type BXP by Rousselet Robatel it will tell you what's possible and it mentions oil-water mixtures. They do a small lab one which scales to their process ones. Again, other suppliers are available.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.