Ethanol Azeotropes - Attn: EtOH cannabis extractors!


So, I assume that you have a mostly purged methanol hash oil. Just add enough ethanol that you can see it’s a viscous solution and purge again. The remaining methanol will be gone very quickly and once the ethanol is purging well the methanol is gone.

I can give more specific advice if that’s my good enough, but it’s the sort of thing that’s not too hard and if I was to give very specific advice I’d have to think about what techniques are being used etc. The boiling point of methanol is lower than ethanol and they don’t form an azeotrope. Methanol is formed naturally in many alcohol preparations and this is the head they really don’t want. So when you have a touch of methanol left and then solvate it all again in ethanol, the methanol isn’t trapped in crystals or goop or anything. So as you heat it, pretty much the very first thing that comes off is all the methanol. Your ethanol is still solvating the material and at elevated temperature / reduced pressure so there isnt the issue of a wax trapping fluids. In the end you will be left with ethanol contamination determined by your operating procedure for an ethanol purge.

Alcohols and terpenes form azeotropes - not as a class but as each individual pair or not… but I urge you to use enough ethanol, but not to go crazy and loose more



So like mix it to a 5:1?



@Lewis_Bronsted is just explaining the physical concept of how viscosity and dissolved solids of a solution affects evaporation of any solvent. As the mixture becomes more viscous or solid, the volatiles have a harder and harder time physically escaping the mixture… so evaporation slows way down.

Therefore, if you want to remove a lower boiling point solvent from a mixture completely, it is best to mix in some higher boiling point solvent to loosen up the fluid, allowing all traces of the lower b.p. solvent to escape freely.

You may be familiar with people using this method on BHO, for example. Instead of purging the BHO into a hard slab over several days in a vacuum oven, and then dissolving the slab in ethanol to Winterize, many choose to simply add ethanol to the gooey butane:resin mixture. This serves the purpose of both loosening the mixture, allowing butane to more rapidly evaporate, and also allowing the resin to dissolve more rapidly in the ethanol. It’s two sides of the same coin.

The same procedure can be done when methanol or some other lower-boiling (and possibly toxic), non-azeotroping solvent needs to be completely removed from your resin.

Don’t let this confuse you, but it could also work with 2 solvents that DO form an azeotrope, as long as that azeotrope is lower-boiling and thereby easy to completely evaporate from the mixture… but this will always take longer to completely remove the toxic/undesirable solvent, since it effectively gives you a single b.p. evaporating solvent (azeotrope) composed of both solvents, instead of letting one totally evaporate while the other mostly stays behind to keep the mixture loose.



“What he said!”




@Photon_noir I’m wrapping my head around this before presenting it to the owners. If our heptane denatured 200 proof has 4.5% heptane, which if im reading correctly can azeotrope to around 15% water, isnt that a composition of roughly only 1 percent water overall? Does that really make a huge difference? I’m planning to buy a LOT of molecular seives to try to dry our denatured ethanol and heptane (liq/liq use).

I’m also curious of how terps affect proof/introduce additional water azeotropes as well. Up until last night I didn’t think inmiscible liquids formed azeotropes. Thank you for waking me up to this.

Appreciate any thoughts you have.


1 Like


I’m wondering, what’s wrong with removing most of the heptane from denatured alcohol? Seems like nobody has given much attention to that question, unless there’s a different thread about it that I haven’t seen?

Fractionating columns are not exactly high tech or expensive to build from tri-clamp fittings. Nor would you need a large column that can distill many gallons per hour, either the binary or ternary azeotrope has a lot of heptane after all. With a decent reflux ratio and stainless scrubbers as packing in a 4-5’ long by 3" dia column, collecting maybe 1/2 to 3/4 gal/hr, I see no reason that wouldn’t pull out near azeotropic 60% heptane leaving very little behind. I’d add just slightly more water than is needed to satisfy the ternary azeotrope, but it would work without water.

Alternatively, maybe force it to separate into two phases and decant the upper phase? The problem here is then you’ll have a bunch of water ethanol mixture that needs to be reconcentrated, so it wouldn’t avoid the need for a column, and the whole volume of ethanol would need to run thru it rather than just the heptane rich fraction, so I see no advantage to this method.

It may be possible to use something other than lots of water to cause phase separation, allowing one to use a normal recovery still to reclaim the ethanol after decanting the heptane. The only nonpolar extractant possibility I know of is silicone oil. Even heavy mineral oil is a bit too soluble to alcohol. From a manufacturer:
“Silicone fluid is highly soluble in hydrocarbon solvents such as toluene, xylene, ligroin, and mineral spirits as well as in chlorinated hydrocarbons. However, it is insoluble in ethanol , methanol, and water.”
It could be easily reused too. No doubt the cleaned up ethanol would need to be run thru a normal pot still, or rotovap to eliminate however much oil that did dissolve into it.
More commonly mentioned in literature is polar solvents (they’re working with much higher % heptane mixtures) and I found papers that show how a nontoxic deep eutectic solvent choline chloride+propylene glycol/glycerin and potentially a little water can work wonderfully for separating ethanol and heptane. Bulk livestock feed grade choline chloride is quite inexpensive too. It won’t let me post links yet, but if you’re interested you can piece these together. researchgate . net/publication/255771075_Deep_eutectic_solvents_as_extraction_media_for_azeotropic_mixtures

researchgate . net/publication/325221289_Effect_of_water_addition_on_extraction_ability_of_eutectic_solvent_choline_chloride_12-propanediol_for_separation_of_hexaneheptaneethanol_systems/amp

The liquid/liquid extraction style technique would also be even more useful to break the ternary azeotrope that’s collected from a fractionating still, in order to recover the heptane and little bit of ethanol as usable solvents. Much smaller volumes of deep eutectic solvent needed that way. Fractionation and then separate the azeotrope is probably how I’d go about it.

If one could find grades of denatured ethanol with methanol or some other less problematic solvents thatd be great but they don’t seem nearly as common or readily available. Nonetheless, buying heptane denatured stuff and fractionating off the heptane looks potentially much easier than fermenting and producing many gallons of azeotropic ethanol. I do have some experience with that… Its not a bad option either. If you do it, my recommendation is find some bulk wine in IBC totes for as cheap as possible and build stills (as many as you need) with beer kegs and 3" tri-clamp spools, packed with scrubbers. Don’t expect over 1 gal/hr of near 95% etoh from a still that size. They don’t have to cost too much, but you better insulate them and have an efficient fuel burning heat source (with a heat tranfer fluid hopefully) or cheap electricity. Its helpful to initially run the wine thru a pot still cranked up as fast as it will go, to “strip” the ethanol out as 40-60% before feeding it into the fractionating still. I’m assuming many of you are reasonably near wine country… :wink:
If you’re lucky enough to find super cheap or free wine like I did, it won’t cost much more to produce the 'shine than denatured alcohol costs. Unless you value your time much. Having the equipment to produce drinkable spirits has its own set of risks and benefits though. I know of a guy down the road who got all the way thru the process of pressing/fermenting apples, building the still, producing the alcohol and then said “screw it, this stuff tastes too good to ruin with a bunch terpenes from trim”
Be aware it can be difficult to avoid some flavor from home made ethanol getting into the extract, whether that’s a problem depends on many factors.

I haven’t used it, but methanol is also intriguing in my opinion. Particularly if you’re already set up to perform extractions below -60 or your post processing steps can handle some extra hydrophilic contaminates.

P.S. I’ve been lurking and sometimes posting in forums on a wide range of subjects since around 2006, but this forum really stands head and shoulders above the rest! I’m really loving it and learning lots, so BIG kudos/congrats/all the rest to everyone who keeps this thing running and the folks contributing their experience and providing real usable high quality information.

In regards to measuring the water content of heptane denatured ethanol. Seems like it would be feasible to use a hydrometer (although it would need to be able to read a bit lower SG than a normal 0-200 proof hydrometer) to get the SG of the fresh denatured 200 proof. Then take a sample, add a measured amount of water, record the new specifc gravity, maybe repeat that again just to be sure. At least until the phases separate.
I’d try it myself but I don’t have the right hydrometer. Maybe measuring the SG at freezer temperature could cancel out the lower density of the heptane to allow the normal hydrometer to work. Hmmm…

Edit 2:
Cool, that works. 95% EtOH 5% n-heptane measures exactly 190 proof (aka 0.8156-0.8160 SG) on my hydrometer, at -32°C. It was syphoned out of a freshly opened barrel and put directly in the freezer two days ago so there shouldn’t be much water.
So that’s a start. I’ll try adding some water to small amount and remeasure once its cooled off.



No,, 15% water means 15% water.

The problem isn’t even the amount of water, though, because the max total water (at least in the azeotrope) is only about 6%.

The problem is the fact that essentially ANY water in the mixture causes the whole azeotrope (of which there is a continuum of ratios for all 3 components) to separate into 2 distinct layers… each of which have different percentages of all 3 compounds: upper layer has about 2/3rds heptane, very little water, and 1/3rd ethanol… the lower layer has up to 15% water, mostly ethanol, and about 5% heptane!
So if you have 6% water in the mixture (which it naturally pulls from air and plant matter), then that 6% water still causes ALL of the liquid to separate into 2 distinct layers which have the ratios mentioned above! Even if you only have 3% water, it still ALL separates into 2 distinct layers which have slightly different ethanol:heptane ratios, and only half the usual rations of water (i.e. top layer is 0.1% water instead of 0.2%, and bottom layer is 7.5% water instead of 15%)!
These layers obviously have very different extraction properties!

Yes, terpenes and other solutes may very well form azeotropes with one or more of each of the 3 solvent components, which would wreak all sorts of havoc on any calculations… but I can verify that the bilayer system still exists after solvent recovery!

@back2thefuture The point of denaturing is to make azeotropes that are very difficult to separate. Sure, one could probably make secondary azeotropes and fractionally distill the ethanol:heptane apart, but something is going to stay azeotroped with something else.



So agitation makes this less of an issue?



Maybe, but not really… the layers separate pretty rapidly. Agitation (when cold, so I can see the difference between the phases) appears to make a heterogeneous matrix of the bottom (more dense) phase with little bubbles of the top (less dense) phase in it, which rapidly climb to the top.



Front load washing machine it is!

thanks @Photon_noir! I hope someday I will be able to help as many people as you do.

1 Like


An old washer, perhaps… the new HE washers stop and start and generally don’t agitate enough to keep it homogenized. Honestly, the only time the azeotrope is truly homogeneous is when it is in vapor phase!

1 Like


Rising film heptane extraction and filtration?

Ps: i think the edibles kicked in

1 Like


Filtering a rising film would be a feat, but interesting idea… vapor phase extraction can be a thing, much like supercritical fluid extraction!



Maybe falling film vapor phase extraction? Inline centrifuge to help with gravity. Hot to cold with media doing the work?

Ps: a 9-5 does sound easier



Nevermind @Poizzin already did this. If only I wasnt afraid of lp gas.



Right I understand the point of denaturants. Just because they’re intended to be inseparable by people trying to bootleg doesn’t mean they’re too inseparable for our purposes. We can all agree that the ethanol heptane azeotrope is lower boiling than ethanol, yet has 10 fold the concentration of heptane as the starting denatured alcohol right? So what is stopping us from fractionating off that azeotrope with a good column until the majority of the heptane is gone?

The boiling point of the ternary azeotrope with water is 10° lower than ethanol. That doesn’t seem like an especially demanding job for a reflux column. Sure getting out most of the heptane is a far cry from making it into drinkable alcohol, so the denaturant isn’t necessarily defeated, just reduced to a reasonable level, say 0.5% or less.

I could be wrong about this, but your assertions thst the azeotropes somehow form separate layers of azeotropic concentrations upon simply mixing water into 5% heptane in EtOH don’t quite add up. As I understand it, azeotropes are a vapor phase phenomenon only. The ternary solubility of ethanol water and alkanes including heptane is in this wonderful compilation of solubility data along with many other solvent combinations. The space needs to be taken out for the link to work.
nist. gov/sites/default/files/documents/srd/jpcrd566.pdf

While I may be interepreting the charts incorrectly, I think they prove that significant amounts (more than you’re stating) of water and heptane can dissolve in liquid ethanol without separation.
I’m really not trying to be an ass here and show up and argue with you in my first couple posts. We’re both trying to clarify what’s going on here for all of our benefit.

This chart is showing molar ratios. For those who arent familiar with these ternary charts, the area above the curve is homogeneous.
The molar volumes of the liquids involed are: ethanol 58ml
heptane 147.5ml
water 18ml
So a liter of denat ethanol with 5% water added will be 15.56 mol ethanol, 0.322 mol heptane, 2.78 mol water or 83.38% ethanol, 1.725% heptane, 14.9% water. I interpret that chart to say that the mixture will be homogenous until at least 30 mole % water is present when the concentration of heptane is so low.

Don’t get me wrong, I’m not disputing that buildup of water could cause problems in an extraction stuation, and problems may well start to occur sooner when the heptane is in the mix. The phase separation could occur at lower water concentrations in cryo temperatures.

Mainly I’m taking issue with the assumption that the ternary azotrope will somehow form on its own in liquid phase, and that the heptane denatured ethanol will absorb more water from the air than normal 200 proof ethanol just because during a distillation a vapor phase azeotrope exists. I just don’t think that’s azeotropes work. Totally open to being proven wrong on this.

The differece in refractive index between heptane and ethanol is about as much as the difference between water and diethyl ether. So if two phases are forming it would certainly be visible in the solvent after distilling it off the extract…

That compilation of ternary solubilities in the link above also has data on methanol+water+alkane and isopropanol+water+alkane systems, as well as many other alcohols, and alkenes, arenes, etc… Seems like it could come in handy around here :slight_smile:



The cold trap condensate on my ffe smells like straight heptane.

1 Like



First of all you da bomb. Secondly, a few questions. Is the growth of the azeotrope limited by the initial amount of n-heptane present? If I am reading the wikipedia table correctly then why does it seem like n-heptane is in much greater concentrations than the other constituents(85%??)?

1 Like


Thank you @Photon_noir. I have 710Spirits n-heptane denatured EtOH specced for an operation launching in May so this is hyper-relevant and was at first rather alarming. Having solved the solvent source way back, nearly at the beginning of the design-phase I first freaked due to recollection of the exorbitant cost of non-denatured EtOH but calmed when I recalled the >99% solvent recovery we’ve designed in. Now however, I wonder how the < -40C extraction might pull the 5% water by freezing: is breaking the 95:5 azeotrope so easy? And since the azeotropic composition is optimal for exclusion of the FATs it seems such loss of water would introduce significant variation in the primary extract quality/composition. This then may require post-solvent recovery titration with water to reestablish the azeotrope, a not-insignificant process. Thoughts?



I looks to me that while EtOH:water are fully miscible as are EtOH:heptane any extension into heptane from the EtOH:Water composition line leads into the 2-phase domain.