Sciencemadness Discussion Board

System to seperate Ethylene Glycol & PG from water - most efficient method?

RogueRose - 9-8-2017 at 14:47

Ok, so the "gunk" that needs to be separated is a mix of coolant from car/truck cooling systems (used, not new stuff). So there should be at least about 50% water mixed mainly with EG and maybe some PG (and possibly some glycerine will be in the mix in newer "green" mixes).

I'm going to guess that distillation is going to be the only way I can separate the glycols and possibly glycerol but. The thing is that while the mix is 50/50 water/glycol, there is also times when a gallon or two is used to flush the whole system to get all the gunk out and purge the whole system, so this is contaminated with anywhere from 1-2% to 10% glycol in the final flushing water - so this could be kept separate if that would make separation easier.

I was wondering if reverse osmosis could be used in this process, especially the "final flush" that has maybe 1-10% glycol. If a very large amount of the water could be removed, then the remaining could be distilled taking a lot less time and energy.

I was also wondering if the RO could work with the 50/50 mix and if it does work to some degree, what percentage might I be able to expect to get the glycols up to?


The thing is that there will be a need for distilled water (or ultra purified - RO would be adequate I think) so that is actually a benefit as I can use as much as is pulled from this. As for the Glycols (and possible glycerol), that would be more difficult to separate as PG and EG have VERY close BP's at 371 (PG) and 378 (EG) but the freezing point of the three is pretty far apart and I was wondering if freezing under vacuum might help separate these and then filter? The other option I guess would be fractional distillation for the EG/PG but I'm wondering if vacuum would increase the difference between BP's and make the separation easier?

Dr.Bob - 12-9-2017 at 17:43

AFAIK, distillation is the only good answer. You will have a hard time getting "pure" water from EG/PG mixtures, so don't count on that to provide distilled water for lab use. I don't think you will be able to easily separate the EG and PG, they are too similar, and freezing is tough to get to work well, might help enrich them.

I would recommend trying to recover the EG/PG by removing most water by distillation(and using it as clean water for making antifreeze mixtures), then using the EG/PG mixture as cheap antifreeze. It will be very hard to make pure EG or PG from the mix, but it will make great antifreeze in any rough mixture. I had access to some years ago for free, and wish I had saved more of it as antifreeze is expensive now, and I hate wasting money on it. If you want to find the approx. concentration of water to EG, you can use density (weigh a vol. flask of the mixture) or refractive index to determine it. Or use a cheap auto tester.

Sulaiman - 12-9-2017 at 18:17

Smaller glycols like ethylene and propylene are highly polar,
so I guess that fairly strong azeotropes are formed with water,
... try to find azeotrope data before starting.