Centrifuge? would it work???

Distillation methods and improvements.

Moderator: Site Moderator

seravitae
Swill Maker
Posts: 225
Joined: Mon Aug 18, 2008 10:31 pm

Re: Centrifuge? would it work???

Post by seravitae »

ressurecting an old thread i know, but just had a late night thought on this. maybe one day I will try it out.

I remember seeing a company produce a cheap centripetal oil-water seperator for the oil spills we had many moons ago. The technique was to pump water at high speed at an angle down a long tube (full of water) such that it created a longitudal vortex. Depending on whether the immisicble phase (oil) was heavier or lighter in density, caused a 'channel' of enriched phase at the center or outer surface of the tube. They had a small outlet pipe at the other end, in the center of the tube, for one phase to pass through, and an outlet pipe at the tube wall for the other phase. Given enough force they were able to seperate the immiscible phases past 98%.

Now because they are immisicble, the surface tension and solubility properties made this type of seperation effective with their equipment.

But, centrifuges are designed to work on different relative densities, which doesnt explicity reference a solubility component.

Where am I going with this? Well, it is *possible* [i would even go as far as to say 'likely'] that if you took a tube of ethanol:water and placed it in a centrifuge, that the water (possessing a higher relative density) would accumulate at the bottom. HOWEVER! It is unlikely it would do this extremely efficiently, to give a pure ethanol and pure water phase, but I believe you would still get a dynamic gradient of ethanol across the tube. To what extent, who knows [this is what science is for!]

If anyone was that interested, i'd suggest building the 'liquid phase vortex tube' as i mentioned above. Take some distilled wash, vortex it through the tube, and collect the inner and outer 'phases' into seperate containers, and test their ethanol content with respect to the initial input mix.

My predictions is that you would get one phase slightly higher in ethanol content and the other slightly lower, even though both phases are miscible. If this predicion is true, then you could almost create the 'reflux' equivilant by taking both one, or both streams, and recirculating them back into the vortex tube. If you recirculated just the 'weak' phase, you'd weaken it to produce more enriched phase (higher recovery factor). If you recirculated just the 'enriched' phase, you'd increase the ethanol concentration (higher purity). If you recirculated a mixture of both, you could probably achieve both circumstances, leading to i guess, a reflux/fractionation equivilant.
seravitae
Swill Maker
Posts: 225
Joined: Mon Aug 18, 2008 10:31 pm

Re: Centrifuge? would it work???

Post by seravitae »

http://www.youtube.com/watch?v=-W8_GpMz ... ator%20oil" onclick="window.open(this.href);return false;" rel="nofollow

for the movie of the device i was talking about. it would be interesting to see what happens if you put ethanol/water through this and just check the concentration of the two outlet phases. i dont think this method would be energy efficient, but it might work.
User avatar
Husker
retired
Posts: 5031
Joined: Thu Aug 17, 2006 1:04 pm

Re: Centrifuge? would it work???

Post by Husker »

I believe (but do NOT have real world experience) that the immiscible vs miscible properties will cause this 'solution' to fall flat on its face. Since ethanol/water (which is almost ALL of what is in the ferment mash), are fully miscible, they form a solution which I believe has no normal property of separation. Even with a simple distillation, we are not separating. We are simply causing the solution to increase in strength, due to the power difference between ethanol and water to phase change from liquid to gas. Ethanol requires less power, water more. So when you put X joules of power into a given mix, it does 'more good' converting the ethanol than water (for a given ABV) from liquid to gas phase, thus the result is a vapor with a higher percentage of ethanol than what was in the liquid mix.

Now, if a centrifuge could spin with enough force that it cause the heavier ethanol molecules to be spun OUT of the mix, then sounds like it 'might' work. However, I am not sure how high of a gravitational pull would be required for that, or if it is even possible. Like I said, no real world experience for myself here.

If you do find these answers, then by all means fill the rest of us in. But, even if this 'IS' possible, like you said, it may not be any cheaper, energy wise.

H.
Hillbilly Rebel: Unless you are one of the people on this site who are legalling distilling, keep a low profile, don't tell, don't sell.
maritime
Novice
Posts: 98
Joined: Wed May 19, 2010 5:29 pm
Location: mid atlantic

Re: Centrifuge? would it work???

Post by maritime »

i think a Wiped Film Evaporator would work. WFE for short
you control the wall temp
there is a rotor with wiper blades spinning inside
you inject the wash into the vessel. this will sling the wash against the heated walls. by controlling the wall temp, you could control what is being evaporated.
the wash keeps geting wiped off the walls and falls to the vessel bottom. it is then pumped back up to the top of the rotor, over and over again.
you could set the wall temp to remove foreshot and heads. then raise the temp till you get into the hearts. when your return slows, raise the temps to get the tails.
i have seen WFE's work. you make a batch of X, you need to remove Y from the formula-after the reaction, run it through the WFE until your residual Y is low enough. normally X with no Y is what you are after. in this case, Y is what we are after.
i think it would work.
it should just leave you with a stripped wash.
7 year column distiller
makes pretty good barn vodka
airhill
Swill Maker
Posts: 287
Joined: Fri Jun 05, 2009 7:42 pm
Location: Fourecks

Re: Centrifuge? would it work???

Post by airhill »

There again you could replace your bubbler plates with spinning disks and drop the reflux into the near center of the disk below. :)
Dnderhead
Angel's Share
Angel's Share
Posts: 13666
Joined: Sun Dec 23, 2007 8:07 pm
Location: up north

Re: Centrifuge? would it work???

Post by Dnderhead »

to complicated for me a 100,000$ still to make 20$ hooch.
airhill
Swill Maker
Posts: 287
Joined: Fri Jun 05, 2009 7:42 pm
Location: Fourecks

Re: Centrifuge? would it work???

Post by airhill »

I was not suggesting you used wipers only spinning disks. Micro grooves on the disks (ie do a scruffy facing job on the lathe) cheap dc small motor on the top, Imho you could get better theoretical plates than packing as you are using mechanical as well as thermodynamic energy, and less height.
There again I only theorise and especially when I have had a few :lol:
Hog Creek
Novice
Posts: 4
Joined: Thu Aug 25, 2011 12:58 pm

Re: Centrifuge? would it work???

Post by Hog Creek »

Haven't they used some kind of spinning cone column to remove the alcohol in some of the alcohol free wines?
airhill
Swill Maker
Posts: 287
Joined: Fri Jun 05, 2009 7:42 pm
Location: Fourecks

Re: Centrifuge? would it work???

Post by airhill »

I know there is one heated spinning cone evaporator but I did not realise they used it to remove alcohol.
Our advantage with disks is that you have the boiler producing vapour so the disks would be relying on boiler energy and not have to be heated.You would have both phase change from the boiler energy and a very large thin film surface area because of the centrifugal force. I am not certain of this but I think you could reduce the height between plates.
Only a theory :)
seravitae
Swill Maker
Posts: 225
Joined: Mon Aug 18, 2008 10:31 pm

Re: Centrifuge? would it work???

Post by seravitae »

Since ethanol/water (which is almost ALL of what is in the ferment mash), are fully miscible, they form a solution which I believe has no normal property of separation. Even with a simple distillation, we are not separating. We are simply causing the solution to increase in strength, due to the power difference between ethanol and water to phase change from liquid to gas.
Simple distillation theory is about seperating compounds based on volatility, whether you achieve full seperation or not depends on the chemical system. The effect of partial seperation is as you say to create a solution with an increased strength, but these are not seperate circumstances. it is entirely possible under the correct distillation conditions to achieve anhydrous ethanol of a concentration high enough to be deemed as 'pure'. Most people just don't have the equipment or desire to do so.


Ethanol requires less power, water more. So when you put X joules of power into a given mix, it does 'more good' converting the ethanol than water (for a given ABV) from liquid to gas phase, thus the result is a vapor with a higher percentage of ethanol than what was in the liquid mix
Not strictly true - your hypothesis appears to suggest that distillation of binary mixtures relies solely on a thermodynamic principle, which it doesn't. chemical and physical-chemical interactions also play a HUGE role in determining phase transition energy constants, azeotropes, etc. Distillation (and boiling for that matter) are simply a product of the vapour pressure of the chemical system reaching equilibrium (and further) with the pressure in the vessel (usually atmospheric)

Now, if a centrifuge could spin with enough force that it cause the heavier ethanol molecules to be spun OUT of the mix, then sounds like it 'might' work. However, I am not sure how high of a gravitational pull would be required for that, or if it is even possible. Like I said, no real world experience for myself here.
ethanol is lighter than water. the water is more dense and would be spun to the bottom.
also please note I did specifically maintain that it would be unlikely that you can spin two distinct phases, and only suggested that you could achieve a concentration gradient within a single phase.
additionally i should point out in the laboratory many of times i have created biphasic systems from miscible solvents. often when growing crystals for x-ray crystallography this technique is used. If very careful you can create a layer of ethanol on top of water (i have done this many times) which persists for some days due to surface tension. So theoretically you could spin out two layers, but as I said, its near impossible. Especially when mixed.


regarding comments about using a WFE, WFE is just thin film evaporation, whilst interesting in its own right, has nothing to do with the theory or general topic of the thread. i dont mind talking about it though, but it's apples vs oranges regarding centrifuge
seravitae
Swill Maker
Posts: 225
Joined: Mon Aug 18, 2008 10:31 pm

Re: Centrifuge? would it work???

Post by seravitae »

also i am extremely busy at the moment but if I get a chance soon I will go into my work lab and create an 'extremist' example, by mixing two fully miscible solvents with the largest difference in density I can find - probably carbon tetrachloride and some hydrocarbon. Equilibrium of density of miscible solvents is quite often met with an interesting optical pattern (shimmering), so given a significantly large difference in density I should be able to mix the solutions fully, centrifuge, and then stir the solution and note for any shimmering. If this works, then it should definately hold true for ethanol:water mixtures, although a similarity in densities would make seperation less efficient and potentially difficult to see any shimmering.
airhill
Swill Maker
Posts: 287
Joined: Fri Jun 05, 2009 7:42 pm
Location: Fourecks

Re: Centrifuge? would it work???

Post by airhill »

"regarding comments about using a WFE, WFE is just thin film evaporation, whilst interesting in its own right, has nothing to do with the theory or general topic of the thread. i dont mind talking about it though, but it's apples vs oranges regarding centrifuge"

Think you'll find on page one (a long time ago :) ) we dived away from the topic heading ,with the approval of its originator, to include any mechanical or mechanically assisted distilling process :lol:

Since you have access to lab equipment actually testing the original question will be very interesting. :)
seravitae
Swill Maker
Posts: 225
Joined: Mon Aug 18, 2008 10:31 pm

Re: Centrifuge? would it work???

Post by seravitae »

Oh, ok, well, I didn't see that, my bad.

WFE is pretty neat and I actually set out to build one some time ago, however didn't really get around to it. I've always stumbled upon making proper seals (vapour proof, we are dealing with flammable vapours here). TFE itself would be quite easy to build and play with, though I've more sooner toyed with the idea of plate exchanger-flash drum for stripping. Most good home brew stores sell 'plate chillers' for chiling wort from a kettle, which have huge surface areas. If you plumbed one end of the plates with a recirculating flow of water or water/glycol at say, 90C, potentially you could push the wash through the other side of the plates to heat it, and then pass the midly superheated mix into a flash drum (either TFE or industrial standard continous distillation style) to achieve a single distillation, which for most, accounts for a significant reduction in volume for subsequent batch distillation.

I believe the whole setup - small heating fluid resivour, hot pump, wash pump, plate exchanger, flash drum and condensor would easily occupy the footprint of a shoebox on its side, and be able to strip an infinitely large wash (though at what rate is another matter)
seravitae
Swill Maker
Posts: 225
Joined: Mon Aug 18, 2008 10:31 pm

Re: Centrifuge? would it work???

Post by seravitae »

Update: In true efficiency, I didn't even need to run the experiment as preliminary research has indicated that it has already been modeled and proved.

Yoav Tsori, Ludwik Leibler, Phase-separation of miscible liquids in a centrifuge, Comptes Rendus Physique, Volume 8, Issues 7-8, September-October 2007, Pages 955-960, ISSN 1631-0705, 10.1016/j.crhy.2007.09.017.
(http://www.sciencedirect.com/science/ar ... 0507002435" onclick="window.open(this.href);return false;" rel="nofollow)


I am unable to post the article for legal reasons but I will post an excerpt of the abstract and the conclusion:
We show that a liquid mixture in the thermodynamically stable homogeneous phase can undergo a phase-separation transition
when rotated at sufficiently high frequency ω. This phase-transition is different from the usual case where two liquids are im-
miscible or where the slow sedimentation process of one component (e.g. a polymer) is accelerated due to centrifugation. For a
binary mixture, the main coupling is due to a term ∝ρ(ωr)2 , where ρ is the difference between the two liquid densities and
r the distance from the rotation axis. Below the critical temperature there is a critical rotation frequency ωc , below which smooth
density gradients occur.
When ω > ωc , we find a sharp interface between the low density liquid close to the center of the centrifuge
and a high density liquid far from the center.
Centrifugation of mixtures, whether of simple liquids or polymers, is shown here to be an effective means to
induce a phase-separation transition. As is quite intuitive, composition variations occur at any rotation frequency,
because of the force which appears due to the different densities of the mixture’s components [12,13,16,18]. The
simple mean-field free-energy model we use reveals a surprising phenomenon: for temperature lower than the crit-
ical temperature Tc , there exists a critical rotation frequency ωc . If the rotation is fast enough, ω > ωc , a demixing
transition occurs where the high density components is pulled toward the outer radius of the centrifuge, lower-density
components are pulled in the other direction, and a sharp interface appears. At these high rotation frequencies, the
centripetal force is large enough to cause large composition differences, which destabilize the homogeneous mixture
in favor of a separated one
.
I have highlited the segments that validate my initial assumptions, mainly that at low speeds a miscible mixture would exist but have a concentration gradient, and at high speeds it is possible to actually make them phase-seperate (though i suggested this would be near impossible practically)
airhill
Swill Maker
Posts: 287
Joined: Fri Jun 05, 2009 7:42 pm
Location: Fourecks

Re: Centrifuge? would it work???

Post by airhill »

Wasn't having a go at you, was just pointing it out. :)
With the centrifuge did they try an ethanol/water mix and if so at what speed did it separate? Mechanically I am having difficulty working out how you could introduce liquid and extract liquid from a high speed centrifuge.
You got any links to diagrams of a flash drum that would work for what you suggested most of the ones I know about are just for separating water mists from steam.
seravitae
Swill Maker
Posts: 225
Joined: Mon Aug 18, 2008 10:31 pm

Re: Centrifuge? would it work???

Post by seravitae »

I know, all good :)

That paper was a phys. chem, so they did the numbers which is applicable to any system. They did not actually test with different compositions, probably because it's not practical industrially. However if i get time, i will see if i can plug in some figures for ethanol:water and see if i can discover the critical rotation speed (given centripital force).

"Continous centrifuges" exist, where you constantly feed in material from a static tube into the center of the rotor and collect 'fractions' of products just like you would in a fractionating still. Either you shape the rotor conically so that the 'fractions' seperate vertically, and collect via small holes (porous rotor), stacked circular splash trays, etc, or you can collect horizontally, by having slight downwards g-force and having tubes coming out of the bottom of the rotor displaced sequentially from the center to the radius. for steady state both systems require a fully flooded rotor as far as i can tell, so you ideally would pump fluid into the rotor faster than it exits, but have the first 'splash tray' recirculate back to your feedstock

regarding a flash drum, the wiki image suffices:

Image

seperation of water mists/droplets from steam, or oil droplets from air, is called 'coalescing', in principal it uses the same 'flash drum' arrangement.

however you can see in the wiki image that you dont necessarily need the distributor tube. simply injecting hot ethanol/water vapour into the side of the drum will cause the hot ethanol rich fraction to vapourise upwards and the ethanol lean fraction to condense below.

instead of using a huge boiler you just use any heating method and constantly pump wash in, and have it vapourise into the flash drum. a plate exchanger would be a safe and fast way to get the wash stream up to boil, though plate exchangers sometimes dont like it when you vapourise the fluid inside
User avatar
guittarmaster
Site Donor
Site Donor
Posts: 407
Joined: Sat Jul 24, 2010 6:51 pm
Location: Misouri

Re: Centrifuge? would it work???

Post by guittarmaster »

Centrifuge separates molecules bases upon mass. I've used them several times in molecular biology where we spin a matrix of cells or cell material. Eg, spinning whole blood to seperate red blood cells and plasma, like when you donate plasma for sugar money.... :shifty:

You can also spin the centrifuge faster until the forces tear apart the lipid by layer and allowing for extraction of proteins and DNA or other cell goo,

you COULD spin you mash to clean it up instead of waiting for it to clear.... if you wait for run your mashes... I don't. But, this is the limit that a centrifuge can help us. +2 points through for thinking outside of box :thumbup:

If you were to spin the boiler like suggested you all you would do is make a more complicated still, This is because molecules do not boil from your mash based upon molecular size, it's based upon bonding energy. The short version is there two kinds of molecular bonds, Van der Waals forces (London dispersion forces) and hydrogen bonds.

hydrogen bonds are what gives water it unusually high boiling point, heat capacity and other physical properties. It is *generally* seen in molecules what have an -OH group; like alcohol, water, and aldehydes.

Van der Waals forces are very week in comparison to hydrogen bonds and is not only the bonding scene of oils, alkanes, alkenes, alkynes, lipids (hydrophobic end), but also liquid CO2, and other non polar molecules.

From the parent site:
Acetone 56.5C (134F) C3H6O, molar weight: 58.08 g mol
Methanol (wood alcohol) 64C (147F) CH4O MW: 32.04 g mol
Ethyl acetate 77.1C (171F) C4H8O2 MW: 88.105 g/mol
Ethanol 78C (172F) C2H6O MW: 46.07 g mol−1
2-Propanol (rubbing alcohol) 82C (180F) C3H8O MW: 60.1 g mol−1
1-Propanol 97C (207F) C3H8O MW: 60.1 g mol−1
Water 100C (212F) H2O MW: 18.01528(33) g/mol
Butanol 116C (241F) C4H10O MW: 74.12 g mol−1
Amyl alcohol 137.8C (280F) C5H12O MW: 88.15 g mol−1
Furfural 161C (322F) C5H4O2 MW: 96.08 g mol−1


As you can see, after adding the molar weights of the compounds there is no correlation with the mass of the molecule and boiling temperature. Ethyl Acetate and Amyl Alcohol have almost the same weight but a 60°C temperature difference of boiling points. Water weights only 18g/mol but boils off with molecules 3 times it's mass...

sorry for the long chemistry lecture, but i wanted to show why, not say "nope doesn't work like that"

GM
War is Peace
Freedom is Slavery
Ignorance is Strength
airhill
Swill Maker
Posts: 287
Joined: Fri Jun 05, 2009 7:42 pm
Location: Fourecks

Re: Centrifuge? would it work???

Post by airhill »

Thanks for that seravitae, it was the terminology threw me as I normally call them VL separators or driers. For that application a vortex separator might be good.
Although the centrifuge would probably have to operate at speeds which would make me a bit nervy :) it would be interesting to know.
GM as far as I understand it there is no heat involved in the centrifuge (aside from mechanically generated heat) so the separation would be according to mass and bonds are being broken by mechanical force. I am assuming the bonds within molecules are stronger than the bonds between molecules.
seravitae
Swill Maker
Posts: 225
Joined: Mon Aug 18, 2008 10:31 pm

Re: Centrifuge? would it work???

Post by seravitae »

guitarmaster:

thanks for the info, im sure the others will find it useful :)
just pointing out that as airhill said, we're not talking about a spinning boiler, we're talking about heatless seperation of miscible solvents via centripetal force alone. :)

airhill: yes, intermolecular bonds are much weaker than intramolecular bonds. if these exist they will be broken, but more bonds will be formed between similar molecules once the phase seperates.
I dont know yet what amount of centripital force would be required for decent seperation (by gradient or sharp phase boundary) but you could potentially limit the 'hazards' by using a continous microcentrifuge and a slower flow rate.

The other thing I just thought of that is awesome - using a centrifuge ie mechanical seperation - doesnt have the importance of raoult's law! - the seperation can go to 100% purity, ie, NO azeotrope as you're not boiling anything!
Bagasso
Distiller
Posts: 1344
Joined: Sat Jun 20, 2009 2:09 pm

Re: Centrifuge? would it work???

Post by Bagasso »

seravitae wrote:The other thing I just thought of that is awesome - using a centrifuge ie mechanical seperation - doesnt have the importance of raoult's law! - the seperation can go to 100% purity, ie, NO azeotrope as you're not boiling anything!
Don't know how it would compare to cleaning oil with a centrifuge but those guys talk about tens to hundreds of liters per minute. There's only 2 liters in a 20l batch at 10% you could be done in less time than it takes to set up a still. Plus you wouldn't have to wait for the wash to clear. I hope it works.
seravitae
Swill Maker
Posts: 225
Joined: Mon Aug 18, 2008 10:31 pm

Re: Centrifuge? would it work???

Post by seravitae »

I imagine that you would not achieve the same flow rates as them, as their seperation is much much more pronounced due to relative densities and also immisicibility. You also raise a good point, i know that in some breweries clarification is actually done using continuous centrifugation, yeast is much heavier than ethanol or water, and is immiscible, so it seperates out. There's no filters to clog and if you optimize carefully you only lose a very small percentage of wash to achieve great clarification. So at the very least it could be fun to play with a small centrifuge style system for clarification.
User avatar
guittarmaster
Site Donor
Site Donor
Posts: 407
Joined: Sat Jul 24, 2010 6:51 pm
Location: Misouri

Re: Centrifuge? would it work???

Post by guittarmaster »

I'm not entirely sure if i understand.. no heat? You'd have to have *some* heat to keep the vapor from re-condensing on the sides, unless your thinking only a centrifuge like the long pipe, and no vapor component.... After all, if you spin a liquid it isn't going to just vaporize on it's own, in-fact you would affectively be raising the boiling point by increasing the pressure needed to for a bubble of vapor.

I'm not going to try to discourage anyone because inspiration takes many forms... however i deeply suspect that centrifuges are far from being practical for the home distiller...

just promise me that you folks will do all your math before shopping for pneumatic motors... :crazy: .. make sure it will work on paper first before spending that time and money on something...


My 2cents is that if your goal is 100% ethanol, than preforming a vacuum filtration through a Semipermeable Membrane or Molecular Seive would be safer, cheaper, and easier to achieve a 100% ethanol sample. Molecular sieves are made for this purpose and for doing it safely, have a proven track record, and are widely used in the industry

http://www.artofdrink.com/ingredients/s ... -of-vodka/
The above link has this to say about them...
Is it possible to get 100% ethanol?

Yes, there are a couple of ways, the sexiest being "molecular sieve" technology. Basically, you start of with a good distillation process and distill until you hit the magic number of 94% ethanol. Then you use the molecular sieve technology to further rectify the ethanol to 100%. You take the 100% ethanol and mix it with the purest water available or pick some place in the world that has the image of pure water and mix it with the ethanol. Voila, instant vodka and a method that could make the marketing geeks go nutty!

There are a couple of other methods including using lime (calcium hydroxide) to dehydrate the ethanol, but this method requires adding a relatively impure mineral to the ethanol, resulting in a lot of possible contamination by trace organics and other chemicals. Doesn't sound sexy. The other way is solvent extraction. Still not cool sounding, well maybe cool, but not healthy sounding.

Over all I'm not a big fan of straight vodka. Vodka has its place, usually in a Kangaroo cocktail or fruity "martini" or when companies distribute flavoured vodkas. That is truly where vodka shines, as a carrier for flavours. On its own, vodka is just ethanol with some extra stuff, but when combined with flavours you can make a great drink. The marketing groups are where the mystique and creativity come from. Sometimes I wonder if a vodka maker discusses the process with the marketing department, before the manufacturing department.
This site, http://www.3angstrom.com/activated_alumina.php has more information about them and their use and a link to order them as well as links to their MSDS

Good luck with it... I'm so gonna look into those Molecular Sieves some more..
War is Peace
Freedom is Slavery
Ignorance is Strength
Bagasso
Distiller
Posts: 1344
Joined: Sat Jun 20, 2009 2:09 pm

Re: Centrifuge? would it work???

Post by Bagasso »

guittarmaster wrote:just promise me that you folks will do all your math before shopping for pneumatic motors... :crazy: .. make sure it will work on paper first before spending that time and money on something...
Actually it has been done in a lab. That is what seravitae linked to in this post: http://homedistiller.org/forum/posting. ... #pr6928743

The question is, what does it take to make it happen in water/ethanol mixture? He offered to do the math. I'm sure no one is buying anything until rough numbers, at least, say it can be done.
airhill
Swill Maker
Posts: 287
Joined: Fri Jun 05, 2009 7:42 pm
Location: Fourecks

Re: Centrifuge? would it work???

Post by airhill »

I think there is a little confusion here because the discussion was both thin film evaporation and an actual centrifuge. The thin film was using centrifugal forces to spread the wash or using a pump to inject heated wash into a vl seperator and would require a condenser. The centrifuge does not create vapour and relies on the forces created by spinning to create layers of substances based on their molecular weight.
Imho thin film is doable, building a high speed centrifuge considering the forces involved would be very difficult (unless you can get one off Ebay) :lol:
seravitae
Swill Maker
Posts: 225
Joined: Mon Aug 18, 2008 10:31 pm

Re: Centrifuge? would it work???

Post by seravitae »

I'm not entirely sure if i understand.. no heat? You'd have to have *some* heat to keep the vapor from re-condensing on the sides, unless your thinking only a centrifuge like the long pipe, and no vapor component.... After all, if you spin a liquid it isn't going to just vaporize on it's own, in-fact you would affectively be raising the boiling point by increasing the pressure needed to for a bubble of vapor.
You can have heated centrifuges if required, but my conversation thus far relates only to a standard centrifuge, we are not talking about distillation inside a centrifuge. The vapour component is ignored.
I'm not going to try to discourage anyone because inspiration takes many forms... however i deeply suspect that centrifuges are far from being practical for the home distiller...
Agreed, though the topic subject is 'Centrifuge? would it work???' :) This is of course, the research and theory subforum. It is merely a thought experiment, and is unlikely to proceed any further then that.
just promise me that you folks will do all your math before shopping for pneumatic motors... .. make sure it will work on paper first before spending that time and money on something...
All I have been doing so far is math/phys. chem. - I am far too busy with actual important stuff to undertake this. Maybe one day I will have a play. I have discovered enough information to say it is 'plausible but not practical'. That is enough for me :)
My 2cents is that if your goal is 100% ethanol, than preforming a vacuum filtration through a Semipermeable Membrane or Molecular Seive would be safer, cheaper, and easier to achieve a 100% ethanol sample. Molecular sieves are made for this purpose and for doing it safely, have a proven track record, and are widely used in the industry
Actually there is an easier way. Once you have passed the pinch point of azeotropic ethanol, distillation to 100% is easy. If you have distilled your ethanol to the pinch point, you can simply add something like dry magnesium sulfate to remove a percent or two of water, and then now that you have bypassed the azeotrope point, you can reflux distill to 100%. We do this in the lab all the time, our ethanol comes as 99%+, but often due to hygroscopicity it gets a little bit wet, but we can easily distill out the water so long as it hasn't absorbed enough water to go below the pinch point.
Also rather than vacuum filtration through a semipermeable membrane, i think you mean 'vapor pervaporation' as this is what industry does with semipermeable membranes with regard to ethanol and other solvents. Regarding your link, which suggests using an impure mineral to the ethanol, it doesnt matter really how pure the substance is, as long as it's deemed food grade safe - as long as there are no volatile organics (which there won't be as most salts like that are calcined at high temperatures) then filtration of the bulk dessicant followed by redistillation would remove any trace of the inorganics.
Solvent extraction is not good for toxicity but also because even nonpolar solvents dissolve water, albeit small amounts, but sometimes enough to make it an issue. When I do extractions in the lab, after performing water or brine washes on an organic solvent, it is always customary that I seperate the organic solvent and add MgSO4 to remove traces of water before solvent removal.

Also molecular sieves I don't recommend as they are fairly basic (alkaline) and can catalyse certain reactions. If you take some acetone and place some molecular sieves in there to dry it, after a few months you will find there is a large component of mesitylene oxide and related aldol components. I recommend only using 'neutral' salts as dessicants.

I think there is a little confusion here because the discussion was both thin film evaporation and an actual centrifuge. The thin film was using centrifugal forces to spread the wash or using a pump to inject heated wash into a vl seperator and would require a condenser. The centrifuge does not create vapour and relies on the forces created by spinning to create layers of substances based on their molecular weight.
Imho thin film is doable, building a high speed centrifuge considering the forces involved would be very difficult (unless you can get one off Ebay)
Thin film evaporation does have a spreader, but that spreader doesnt use centripital forces as a means to do so. It's actually a displacing and shearing force. The wipers spin in a circular fashion only because of the way they were designed - you can have a linear thin film wiper that slides back and forth, it will still spread the material into a thin film, but obviously still has no centripetal force.

Also just for accuracy a centrifuge will seperate substances based on relative density, not molecular weight. Molecular weight determines the average mass of an individual molecule, but it does not tell you how many of those molecules will pack together in a fixed volume, and thus it cannot tell you what the density would be. Iodine has a Mw=126.90g*mol^-1 and d=4.933 g*cm^-3, whereas carbon tetrachloride MW=153.82 g*mol^-1 and d=1.58 g*cm^-3 [liquid state]. That is, carbon tetrachloride has a higher molecular weight and lower density than iodine, which has a lower molecular weight and a higher density.

Also I am talking out my arse because I havent been able to find the numbers, but I think it's doable at home, but whether its worth the effort or not is debatable. I have seen people modify dremel multi-tools for use as microcentrifuges with eppendorf tubes (for biochemistry@home projects) and those dremels can easily spin at 15,000 rpm, much more than a standard benchtop centrifuge.
Bagasso
Distiller
Posts: 1344
Joined: Sat Jun 20, 2009 2:09 pm

Re: Centrifuge? would it work???

Post by Bagasso »

Here is a link to the paper:

http://www.bgu.ac.il/~tsori/my/25cr_phy ... 5_2007.pdf

While I get the gist of it I wouldn't know where I could plug in numbers even if a knew what numbers to look for but two things caught my eye:
If ω is larger than the plotted ωc, demixing occurs, otherwise the mixture displays only weak composition gradients. The numerical value of ωc is quite low, and certainly achievable in conventional laboratory equipment – ωc = 60 rad/s is equivalent to 572 rpm.
and
Centrifugation of mixtures, whether of simple liquids or polymers, is shown here to be an effective means to
induce a phase-separation transition.
A while back I had looked around the net for a counter-top centrifuge used to separate milk and cream thinking it would work to clear washes but couldn't see spending $200 on it. According to the description it goes up to 12,000 rpm. The first quote says they were using less than 600 rpm.

Now I know that this is not the amount needed for an ethanol/water mixture but even 10x that amount is doable.

The second quote says its effective. I could only imagine that it at least compares somewhat to normal distillation if not better.

The last thing that is going through my mind is that even if you can't split the water and ethanol maybe you could centrifuge out whatever it is that makes tails funky. So crystal clear washes and tails removal from the centrifuge and then off to the still to get rid of that pesky water.

And of course if anyone is interested in testing here is the best idea I have seen for a continous centrifuge on the cheap. A centrifuge made from a centrifuge juicer which gets 3600 rpm:
http://www.youtube.com/watch?v=rS26K9xt3Zg
airhill
Swill Maker
Posts: 287
Joined: Fri Jun 05, 2009 7:42 pm
Location: Fourecks

Re: Centrifuge? would it work???

Post by airhill »

I take your point about density :)
I had thought wipers only made spinning disk evaporators more efficient.
If someone has a centrifuge wouldn't the simplest be to dilute some neutral and give it a wizz to see what happens?
Thanks for the paper Bagasso.
Post Reply