Sciencemadness Discussion Board - Salting out ethanol and distillation

Salting out ethanol and distillation

Pages: 1 2

careysub - 13-10-2016 at 15:13

Interesting that the K2CO3 also gave 100%! I'm amazed this isn't more common knowledge.

According to this Caltech thesis the drying of ethanol with K2CO3 stops at 94%. This guy did a lot of work on developing a potential commercial process based on this, so I suspect he is correct.

Attachment: Prentice_lb_1925.pdf (2.7MB)
This file has been downloaded 851 times

Maroboduus - 13-10-2016 at 20:50

Where'd you find that?

careysub - 14-10-2016 at 05:55

Quote: Originally posted by Maroboduus

Where'd you find that?

Right here:
http://thesis.library.caltech.edu/6717/

And I found it with Google typing in "dehydration ethanol potassium carbonate" (though possibly not in that order), which it turns out is nearly the title of the thesis.

[Edited on 14-10-2016 by careysub]

Maroboduus - 14-10-2016 at 10:33

Wow, usually when I try something that simple I get 5,000 hits from flaky chat sites about explosives, drugs, & etc; so I just assumed it was something less straightforward.

Never thought about college theses being searchable online either. Bet there's some interesting stuff in those university sites.

aga - 14-10-2016 at 10:58

Quote: Originally posted by careysub

According to this Caltech thesis the drying of ethanol with K2CO3 stops at 94%.

The 100% deltaH was quoting was probably due to operator error.

[Edited on 14-10-2016 by aga]

Maroboduus - 14-10-2016 at 14:41

Quote: Originally posted by aga

Quote: Originally posted by careysub

According to this Caltech thesis the drying of ethanol with K2CO3 stops at 94%.

The 100% deltaH was quoting was probably due to operator error.

[Edited on 14-10-2016 by aga]

That thesis may be from Caltech, but it's an engineering student in the 1920's who admits in his paper that he's making various simplifying assumptions. (pretending his Carbonate is fully dry when it's up to 3.5% water, assuming there's no carbonate at all dissolved in the EtOh...).

He could well be off by a few percent. Hell, he's an undergrad; they might not even be letting him use the good set of hydrometers.

careysub - 14-10-2016 at 14:52

Fair enough. The SM crew needs to develop some truly definitive results. The gauntlet it thrown down!

What would be the most reliable method?

(BTW, the fact that it was written in 1925 should only be a relevant factor in questioning the result if we were comparing his results to ethanol/water measurements made with some better techniques not then available. This is not the case, the methods used on this thread are no better than what he had.)

Maroboduus - 14-10-2016 at 16:04

Quote: Originally posted by careysub

(BTW, the fact that it was written in 1925 should only be a relevant factor in questioning the result if we were comparing his results to ethanol/water measurements made with some better techniques not then available. This is not the case, the methods used on this thread are no better than what he had.)

Fair enough. But all I've got for this sort of work is a cheap Belle and Tralle hydrometer that's only somewhat more accurate than my highly trained palate, so I'm afraid I can't be much help.

Most of the work I've done in the past is stuff you characterize with melting points, so my analytical gear is a little embarrassing (A Thiele tube full of Wesson oil and a magnifying glass).

I've been out of this game since the early 1990s, and just got inspired to take it up again recently. I'm still digging out random bits of glassware and ancient chemicals from here and there. (why the hell didn't I put all that stuff in one place? Last week I found some 20 year old THF in with a box of Cassette tapes I was afraid to even move it).

careysub - 14-10-2016 at 16:22

I have just ordered a moderately priced Tralle/Proof hydrometer that should be accurate to 1%, which will certainly be able to distinguish between 94% and something close to 100%.

I also have an experiment in progress (I will check tonight) to see if the KMnO4 color test really works to identify absolute ethanol (reportedly no color equals >99.5%). I have ethanol sitting on top of an excess of active 3A sieves which should render it absolute by the time I get home.

aga - 14-10-2016 at 22:51

Quote: Originally posted by careysub

see if the KMnO4 color test really works to identify absolute ethanol

It's more of a quick test to see if there's a >1 % of water in your EtOH rather than an indication of absolute dryness.

Be sure that the KMnO₄ is dry !

Maroboduus - 15-10-2016 at 07:09

If you were here, would you be able to see the egg on my face?

Being finally spurred into action by less insouciant members, I took out my hydrometers and had a look.
The lousy one is a broad range hydrometer (1.0-2.0 ), but among the others is a Tralle hydrometer by Bellwether, which, according to it's insert has a 1% subdivision range and a 2% subdivision tolerance.. Took some alcohol out of the fridge just now so it can warm to room temp for testing it's proof. (It's a beverage, but its just water/ethanol in a somewhat stronger form than commercial vodka.) Will be going under the house later to start the preparation of Potassium Carbonate (the fireplace ash is accessible from down there), as I need some for both this and to treat my solutions of leached vegetable compost and rabbit crap which are part of an investigation on making KNO3 the old fashioned way.

You and Aga have managed to shame me into action, so I suppose I owe you both thanks.
Aga, I've said some harsh things to you on this site in the short time I've been here, and I'm sorry. I can understand your frustration at the lack of action on some of these threads.

careysub - 15-10-2016 at 07:45

Quote: Originally posted by aga

Quote: Originally posted by careysub

see if the KMnO4 color test really works to identify absolute ethanol

It's more of a quick test to see if there's a >1 % of water in your EtOH rather than an indication of absolute dryness.

Be sure that the KMnO₄ is dry !

My initial experience is that I need to prepare some standards of known water content to compare the KMnO₄.

The super-dry ethanol developed some color - though only a tiny amount considering the potency of the MnO4 chromophore. So I do not know how much that differs from the coloring of 0.5% water, 1% water and so forth. I think I will take pix of a row of test tubes with different concentrations for comparison.

I think I will wait for the hydrometer before proceeding as a check before preparing the necessary super-dry ethanol reagent.

I'll post results in a week (assuming the hydrometer gets here in time).

aga - 15-10-2016 at 08:34

Quote: Originally posted by Maroboduus

Aga, I've said some harsh things to you on this site in the short time I've been here, and I'm sorry.

I honestly don't take any offence at anything anyone says online.

Glad to see some action !
(yes, i should be Doing more too)

The KMnO₄ definitely needs to be dry.
The last time i was messing with ethanol drying agents, i used (what was much later discovered to be) ever-so-slightly damp stuff, and this threw my results totally out of whack.

The temptation of an 'instant' indicator was too great.
It would have been better to have used the ethanol refractometer with a 50:50 dilution of the sample.

In 99.99% dry ethanol, there is no colour detectable in the liquid by eye with pot permanganate added.

(pot permanganate added to the liquid, not the eye

)

Add in the tiniest amount of water and the pink/purple is very obvious, so i'd suggest using it as a 'nearly there' indicator and do the actual measurement with an accurately diluted sample.

zwt - 15-10-2016 at 16:38

Quote: Originally posted by Maroboduus

It's a beverage, but its just water/ethanol in a somewhat stronger form than commercial vodka.

A very strong, non-commercial, vodka-like alcoholic beverage, which can be described, but not named?
You sure there's no methanol (or other contaminants) in your <s>moonshine</s> beverage?

Eagerly awaiting the results, in any case, of this, and the rabbit crap.

Maroboduus - 15-10-2016 at 19:26

Made from vodka. I don't MAKE distilled liquor. No point when you can buy it with the tax properly paid for it's manufacture. That's why I can't name it.

I just concentrate it. I do this to make liqueurs out of it. I was making some up to marinate fruitcakes for the holidays, so I had some on hand. I concentrate it so I can dilute it with the flavoring and still have 80-100 proof.

So there's really no accepted name I know of for what I've got. But it was made from a halfway-decent quality vodka by running it through an all glass still with ground joints and a Vigreux column. Discarded the first 5 to 7 percent, and collected a 4 or 5 degree fraction.

I used to make alcohol from mash. Way back when they had what was called an experimental alternative fuel license. For a nominal fee you could make and keep something like 3-5 gallons at a time to keep on your property to run engines. I believe they got rid of it because they figured out many of the licensees were using it as cover to make a little moonshine.

The compost and rabbit crap need some potassium carbonate as well. I'll see if I can find a place to buy some so I won't have to wait for the crude stuff I'm Making up from ashes.
Might pick up some sodium carbonate too for comparison.

[Edited on 16-10-2016 by Maroboduus]

Maroboduus - 17-10-2016 at 17:33

Bought some anhydrous K carbonate today, and Na Carbonate monohydrate.
The 'ethanol' tested out at 180 proof. Actually a little higher than I expected with what I hear around here about Vigreux columns.

Took two 150 ml samples and added potassium carbonate to one and Sodium carbonate to the other. Quantities were calculated to be 10-15% more than what should be required to saturate the water in each sample.

I don't think the higher starting proof of the alcohol for this test will alter the results. Eventually, any sample of ethanol/water being dried in this way will become alcohol floating on a saturated solution of the salt. The action is likely all at the brine/alcohol interface, so I doubt the amount of brine below that interface matters as long as the brine is saturated at endpoint.
Likewise, I don't think any undissolved material in the brine will change the equilibrium either.
If this is disputed I can always re-dilute the alcohol to 100 proof and run another test.

EDIT: after standing 3 hrs with occasional shaking the sodium carbonate is mostly still a free flowing powder, but the potassium carbonate has partly dissolved and there's a little collected fluid on the bottom. Decantation and testing with the hydrometer showed that it is now 189 proof. The material was returned to the flask with the rest of the alcohol/water/Potassium carbonate mixture to sit for an extended period.

This may be all it'll do, and it's closer to 94% than the accuracy of my hydrometer. These solutions will be allowed to sit for a few days to see if anything else happens.

I thought I'd seen claims for the effectiveness of Sodium carbonate as well for this purpose, but so far it doesn't seem to be doing much. maybe it's just used as a conventional crystalline drying agent, in which case I probably need to get it anhydrous for any action. (The K Carbonate was anhydrous, but the Na Carbonate was the monohydrate).

This has got me curious about how well CaO does. Supposedly it gives 200 proof with a treatment and distillation from rigorously dry glassware, but my source for that info is a 19th century formulary, so I don't know.

I mean hell, the same book I got that out of tells you to treat some sheep disease by driving a knitting needle up the sheep's nose.

[Edited on 18-10-2016 by Maroboduus]

Maroboduus - 18-10-2016 at 18:07

The ethanol over Potassium carbonate read the same 94.5% after 24 hours. This is closer to that Caltech thesis than the accuracy of my hydrometer.
The Ethanol over Sodium carbonate rose from 90% to 90.5% over the same period of time. The Sodium carbonate never seemed to dissolve or get sticky or lumpy. This 'change' is also well within the tolerance of my hydrometer.

Looks like that guy was right on. At least I like to think so since while it disproves my suspicions about his accuracy, it at least shows that my hydrometer and correction tables work fairly well.

This certainly is a quick easy way to get almost azeotropic purity in Ethanol. This could probably save a distillation step in making 200 proof, since 94% is probably close enough that it could be dried to 200 by any standard method.

From what the paper said about negligible residue I suppose this treatment alone on ethanol of 100 proof or higher would give the functional equivalent of distilled 190 proof for almost any purpose. And it fully dried in less than 3 hrs with only occasional shaking.

I may set some potash aside for re-use as a drying agent for alcohol. It seems like a fine way to reclaim dilute solutions, even VERY dilute solutions if a quick distillation is used first.

BUT, it's not going on my fruitcake!

Bert - 22-11-2017 at 17:21

I am really tired and somewhat out of my head, perhaps- But how does anhydrous tripotassium phosphate compare to say, 68% to about 95% nitric acid for attraction to water?

And what non water miscible, reasonably cheap and benign liquid likes fuming nitric acid more than the (partially) hydrated phosphate, but would not itself be nitrated or form a Sprengel explosive with strong nitric acid?

Just wondering.

[Edited on 23-11-2017 by Bert]

deltaH - 2-12-2017 at 12:04

I'm glad to see some lateral thinking about the possible uses of salting out.

For the salting out effect to work, you'd need an organic component to bind to the nitric acid and so make it 'saliphobic'. This is nothing else than a nitration (which I assume is what you'd want the anh. nitric acid for anyway). K3PO4 and K2CO3 are out because they would instantly react with the nitric acid, but KNO3, NH4NO3 or Na2SO4(an) may well work. Surely there's a body of literature on this; nitrations in mixtures of nitric acid and nitrate salts?

pots-o-potash - 7-8-2018 at 16:20

Potassium bisulfate or Potassium peroxymonosulfate might be more useful for salting out nitric acid (no chance of reacting to form nitrate and bisulfate) and the oxidizing properties of the latter might also reduce the evolution of NO/NO2 (ie more will be oxidized to (NO3)- in solution, at least while there's excess water present). The availability of Potassium peroxymonosulfate from pool supply stores/warehouses (and online) at technical grade might also make it more appealing. (albeit if you're already producing concentrated nitric acid via distillation of a nitrate salt and sulfuric acid, you'd conveniently have a mix of sulfate and bisulfate waste material left over, or mostly straight bisulfate if you distilled at lower temperatures)

It also might be useful for producing a reasonable nitrating solution in leu of nitric acid + sulfuric acid (ie near-azeotropic nitric acid mixed with solid, dry bisulfate or peroxymonosulfate might work or at least work as well as nitrate salt+sulfuric acid mixtures) It also might boost performance of other nitrating mixtures. (getting off topic of the salting-out subject though, but an interesting topic to experiment with)

I'm also not entirely sure why people have found it so hard to source info on salting out ethanol. At very least, the potassium carbonate example is common as a chem class lab demo (also common to use with methanol, and the isopropanol equivalent obviously has a wide range of salts to use, including chlorides, so would be less useful). Details on the purity resulting and examples of industrial use are more difficult to find, granted, and perhaps more so a few years back with less vast, easily searched patents and research papers online, but the basic chem lab demo examples were out there. (the methanol example is also somewhat marginal as potassium carbonate is slightly soluble in methanol, and sodium carbonate might actually work better in that case, or sodium sesquicarbonate as that's somewhat more soluble)

Additionally, I stumbled on this:

https://onlinelibrary.wiley.com/doi/abs/10.1002/aic.14872 salting out ABE fermentation products:

Quote:

The salting‐out phase equilibria for acetone, 1‐butanol, and ethanol (ABE) from dilute aqueous solutions using potassium carbonate (K2CO3) and dipotassium hydrogen phosphate trihydrate (K2HPO4⋅3H2O) as outstanding salting‐out agents were investigated. Increasing the salt concentration strengthened the salting‐out effects and improved the distribution coefficients of all three solvents (ABE) significantly. Temperature had a slight effect on the phase equilibria. The K2HPO4 solution (69 wt %) showed a stronger salting‐out effect than the K2CO3 solution (56 wt %) on recovering ABE from dilute aqueous solutions. Dilute aqueous solutions containing more solvents increased the recoveries of acetone and 1‐butanol, while the results showed a negligible effect on the solubility of ABE. The solubility of ABE was also correlated well with the molar number of salt per gram of water in the aqueous phase. A new equation demonstrated this satisfactorily. © 2015 American Institute of Chemical Engineers AIChE J, 61: 3470–3478, 2015

The high solubility of the diphosphate makes it more favorable, the same reason potassium salts are more applicable than their sodium counterparts in these examples (carbonates and phosphates ... though sulfates and bisulfates of sodium are actually more soluble and might be more relevant ... same for nitrates obviously and chlorates/perchlorates, but I doubt the latter would be particularly appealing for this purpose ... though perchlorate is quite stable and wouldn't be an oxidation risk, plus NaClO4 is incredibly soluble and would phase-separate rapidly due to the high density of solution, so if it works for salting out ethanol, it might actually be rather appealing).

In any case, potassium di-phosphate appears to be an attractive option and one being investigated for commercial viability.

OTOH, with ABE fermentation specifically, it might be more economical to use a cheaper salt (like calcium chloride) to separate the butanol+acetone and deal with the remaining ethanol+water+salt+etc layer separately. (still if using it purely for fuel production in situations where the ethanol content is attractive or acceptable, extracting it as a single blended fuel stream might be most valubale)

I'm not sure how any organic impurities would fit in to such separation, either (leftover carbohydrates, proteins, lipids, etc, or which of those would be acceptable fuel contaminates. (albeit at least some of the fatty organics might be acceptable in a gasoline substitute and possibly increase the energy density at the expense of some octane ... also likely boost the vapor pressure of the ethanol by reducing polarity of the liquid, same as blending with hydrocarbons)

You'd also have fractional freezing as a possible stage somewhere amongst all that. (you could also perform the fermentation at reduced pressure, allowing acetone and possibly azeotropic ethanol to evaporate or boil off while rehydrating the fermentation mash/liquor as necessar ... it might also delay poisoning of the yeast and improve cost efficiency that way, albeit if yeast also survives in ice, it could be valuable to apply fractional freezing to separate butanol before fermentation reaches its end point)

Hmm, a quick search found this on yeast and freeze/thaw cycles, which seems interesting, though less positive towards yeast surviving http://aem.asm.org/content/63/10/3818.full.pdf

The behavior of Clostridium acetobutylicum (or other Clostridium species) would be more relevant, though.

Also, survival in brine solutions (of various salts of various PH levels) would also be relevant, as continuous salting out might be a viable option. (or at least salting out of the butanol, as the least polar component, whilst evaporation-separating the acetone and possibly ethanol)

60% n-butanol, 30% acetone, 10% ethanol makes an appealing gasoline additive or possible substitute (It'd run lean in non-flex-fuel vehicles if used straight, but might be acceptable ... though blending with various mono and di-cyclic aromatic hydrocarbons, alcohols/phenols, and ethers like anisole would balance it out quite readily and result in a high octane rating, plus further reduce the corrosive/solvent properties of acetone and ethanol beyond what butanol does alone). Blending with isobutanol would also be appealing, be it microbial or synthetic in production. (syngas-derived isobutanol has been used since the 1930s and would work nicely with biomass/char or coal/petroleum/natural gas derived syngas sources ... or various sorts of organic waste, biological or otherwise)

On a side note, the lower BTU/energy density of some oxygenate fuels doesn't always equate to lower MPG or engine-energy-output/unit fuel due to combustion efficiency and other characteristics that impact real-world thermodynamic efficiency of internal combustion engines. (hence both anecdotes and formal research papers on improving MPG with acetone and various other oxygenates in limited quantities and/or a lower than expected drop in MPG at high portion blends)

Edit:

Also worth mentioning is Kleanstrip brand Green Denatured Alcohol. Rather than the ~50% methanol of their typical denatured blend (Crown seems to use even more), the SDS lists <5% methanol, 1.5% ethyl acetate, and 80-90% ethanol with the balance presumably water.

http://www.kleanstrip.com/product/green-denatured-alcohol

http://www.kleanstrip.com/uploads/documents/QKGA75003_SDS-16...

also some previous formulations using MIBK instead of ethyl acetate:
https://www.woodcraft.com/media/W1siZiIsIjIwMTcvMDkvMDYvMTYv...

http://go.rockler.com/tech/RTD20000423AA.pdf

(also broader margins for the chemical content)

[Edited on 8-8-2018 by pots-o-potash]

nimgoldman - 19-10-2018 at 19:40

I tried to reproduce the experiment but I got nowhere near the results published here.

I took 200 mL of distilled mash which is 56-58% ABV:

It therefore contains about 88 mL water. Given the solubility of K₃PO₄ in water, I needed little less than 80g. Here on the left 80 g of salt, on the right 200 mL of alcohol. It looks like a lot of salt, but actually the powder is very fluffy:

After adding the alcohol, the solution got very hot indicating that reaction happened. I mixed the salt thorougly and it created a viscouc, almost gelatinous layer on bottom:

Not all the salt dissolved indicating that the water is saturated now:

I've let it sit for several hours, decanted about 90 mL of the top alcohol layer and immediately measured the concentration.

It was about 82%:

So nowhere nere 99-100% claimed here by the refractometer measurement. Or should it be left sitting for longer time? Or should it be filtered? I have not observed any phase separation in the liquid, just the alcohol and salty water. My salt might not be completely dried - I assumed it is anhydrous from the label. It is 97% pure.

Much higher concentration than this can be achieved in just one run of this distillate though a fractional distillation which is time consuming but still less laborious than the tedious recycling of the salt.

Has anyone else had success with the method?

[Edited on 20-10-2018 by nimgoldman]

Mr. Rogers - 20-10-2018 at 19:22

What's wrong with using dry CuSO4?

nimgoldman - 21-10-2018 at 12:26

Quote: Originally posted by Mr. Rogers

What's wrong with using dry CuSO4?

I don't know but that's not the experiment I have been reproducing.

I have been investigating salting out ethanol with K₃PO₄ which supposedly gave superior results.

There was no confirmation in years so I was tempted to try and possibly debunk the tripotassium phophate method myself.

[Edited on 21-10-2018 by nimgoldman]

deltaH - 27-10-2018 at 07:33

I'd suggest drying your tripotassium phosphate in an oven first like @aga did as your salt may be hydrated if it's a commercial product.

Once dried, guard against exposing it to air. It's terribly hygroscopic.

Poppy - 7-11-2018 at 15:48

What if the activity of the salty mixture is raised as to the H bonds stay busy for a longer mean time in this solution. Lets consider K2CO3 with Ca acetate plus sodium chloride? Not much of the later ones so as to do not shift the hydration function of the carbonate but even so permitting more of the K2CO3 to be added in the separation, giving a shift to higher yields
Imma absolutetly spend 200 bucks to try this in a year or near future if possible

fusso - 7-11-2018 at 16:02

Quote: Originally posted by Poppy

What if the activity of the salty mixture is raised as to the H bonds stay busy for a longer mean time in this solution. Lets consider K2CO3 with Ca acetate plus sodium chloride? Not much of the later ones so as to do not shift the hydration function of the carbonate but even so permitting more of the K2CO3 to be added in the separation, giving a shift to higher yields

Wont Ca acetate react with K2CO3 to ppt CaCO3?

andrewchen5678 - 21-3-2019 at 23:28

Has anyone tried potassium citrate to salt out alcohol? According to wikipedia, it is insoluble in ethanol 95%. https://en.wikipedia.org/wiki/Potassium_citrate. I tried it before and it can also separate out ethanol similar to K2CO3, but I don't know if there are any drawbacks vs K2CO3.

[Edited on 22-3-2019 by andrewchen5678]

fusso - 22-3-2019 at 02:50

Maybe citrate will decompose easier.

Arkada - 21-1-2023 at 02:23

Came in to testify that K3PO4 does indeed work magic. Takes around 1.5g to consume 1g of water. Dries at 150C.
More people should know this

Pages: 1 2