Sciencemadness Discussion Board

Best non-polar solvent for caffeine

Cavatica - 22-1-2015 at 03:14

Hi there! First post.

I'm planning on doing a caffeine extraction, but am uncertain as to what to use for NPS. As a premise, I would like to avoid the chlorinated alternatives, and make the process as green as possible. I have access to d-limonene, and would like to know if anyone here has any experience using this as the NPS in caffeine extraction. Thanks in advance!

Jylliana - 22-1-2015 at 04:12

The first solvent that came to mind was dichloromethane, but that has chlorine.
So... the next best thing is diethyl ether.

EDIT: I thought I had already done an extraction with diethyl ether, but I mixed up two names, so I didnt extract it with this.

It could be used, though.

[Edited on 22-1-2015 by Jylliana]

Cavatica - 22-1-2015 at 04:17

Yes, I would like to stay miles away from ether as well. :) I have nowhere near the experience or confidence to be fooling around with something as volatile and flammable!

gdflp - 22-1-2015 at 04:17

I completely agree with Jylliana, diethyl ether is the best solvent for this reaction without using chlorinated solvents. However, if you don't have access to diethyl ether another possible solvent is ethyl acetate, though yields will be lower due to it's slight solubility in water.

NitreRat - 22-1-2015 at 05:15

Ethyl acetate would be the go-to solvent for this extraction. It's cheap, volatile and non-toxic (within reason). It can even be easily synthesized from from vinegar and vodka. The only drawbacks I can think of are its flammability and strong smell. With non-chlorinated non-polar solvents there's always a trade-off between flammability and volatility. Whilst limonene would be safer to work with, it's much more difficult to strip the solvent once you've extracted your product and the higher temperatures could damage your product.

If you have the equipment for vacuum distillation I would recommend using limonene.
If you have a basic distillation set-up I would recommend using ethyl acetate. Just remember, when stripping the solvent, to heat the distillation flask with a hot water bath (avoid flames) and cool the receiving flask with ice.
If you have zero glassware you could wait for a hot day and do the extraction with lighter fuel and a plate ghetto style.

kmno4 - 22-1-2015 at 05:32

Quote: Originally posted by gdflp  
However, if you don't have access to diethyl ether another possible solvent is ethyl acetate, though yields will be lower due to it's slight solubility in water.

Solubilities of Et2O and EtAc in water are of the same order and differ only by ~1g/ml (as far I remember)

Cavatica - 22-1-2015 at 06:40

But d-limonene - is it feasible?

Bert - 22-1-2015 at 07:40

Supercritical CO2.

http://www.sweetmarias.com/decaf.co-2method.html

http://youtu.be/TvmM6qa2a90

http://en.m.wikipedia.org/wiki/Supercritical_carbon_dioxide

ethyl acetate...

http://reviews.homedepot.com/answers/1999/product/202994064/...


[Edited on 22-1-2015 by Bert]

Mailinmypocket - 22-1-2015 at 08:17

Quote: Originally posted by Cavatica  
But d-limonene - is it feasible?


Maybe? Try it. It certainly is not the best non polar solvent for caffeine, by definition of the thread title.

Why not use a tried and tested solvent? D-limonene has a high boiling point compared to other solvents such as chloroform, diethyl ether, dichloromethane... Letting it evaporate would likely take forever at rt. Save your d-limonene and make some chloroform instead with sodium hypochlorite and acetone (OTC) or using other well known methods found in books or this forum.

[Edited on 22-1-2015 by Mailinmypocket]

Cavatica - 22-1-2015 at 08:38

Thanks, Mailinmypocket. Being aware of the issue with the boiling point of d-lim, I might just migrate the caffeine back from the non-polar into an acidic phase (as caffeine hydrochloride), then basify with NaOH and evap enough of the solution to achieve crystallization by cooling. That, or sublimation, but I've never tried that. I would really like to try with d-l. I've even heard sunflower oil as a potential solvent. Would use DCM or chloroform instead, but, well. I'm aiming at low toxicity.

Nicodem - 22-1-2015 at 09:27

You can forget about using limonene. Did you check the logPow of caffeine? It is -0.07, so there is not a way it would usefully partition to limonene.

Dichloromethane, ethyl acetate, chloroform, 2-butanone (MEK) and 2-methyltetrahydrofuran (MeTHF) are about all there is useful among the more common solvents. 1-Butanol or 2-butanol might also work, but I'm unsure of that, as caffeine does not solvate well with alcohols.
Quote: Originally posted by Cavatica  
Thanks, Mailinmypocket. Being aware of the issue with the boiling point of d-lim, I might just migrate the caffeine back from the non-polar into an acidic phase (as caffeine hydrochloride), then basify with NaOH and evap enough of the solution to achieve crystallization by cooling.

It is quite obvious that you read none of the many threads on this topic that we already have here. You also did not check the pKa value of caffeine. Don't worry, we see this all the time on this forum, as it is a typical demonstration of how beginners prefer to make assumptions rather than checking the literature. I suggest you to start by first reading the existing topics and then read also all the experimentals for caffeine extraction (we even have one published on the forum, with pictures and all).
Quote:
I've even heard sunflower oil as a potential solvent.

Seriously? Where?
Quote:
Would use DCM or chloroform instead, but, well. I'm aiming at low toxicity.

Caffeine is also toxic, even more so than dichloromethane and chloroform. In contrast to the last two, death by caffeine poisoning is terribly painful and agonizing.

Chemosynthesis - 22-1-2015 at 10:14

Caffeine certainly can be toxic, but I wouldn't characterize caffeine as more toxic than dichloromethane or chloroform without specifying acute doses (or dose/time if chronic), method of administration (dermal or inhalational may be more relevant than oral), and animal model used.

I have used ethyl acetate as an exercise in decaffeination. There are obviously many options as stated: http://www.scientificamerican.com/article/how-is-caffeine-re...

Cavatica - 22-1-2015 at 10:20

Painfully aware of my novicity, and I should naturally be reading up on all the available info. I am, however gradually. That being said, making assumptions in the sandbox is often a good way to establish heuristics, and to trigger feedback of a more general nature so as to cast light on the pressing issues. Just starting out here! Very grateful for the feedback.

Cavatica - 22-1-2015 at 10:25

Also, Nicodem, the sunflower oil reference: Hossain & Chong 2011 (PDF)

ChemistryGhost - 22-1-2015 at 15:16

2-methyl-2-butanol can extract caffeine. Though 2-methyl-2-butanol is less toxic than ethanol and most solvents, caffeine itself can be somewhat toxic. I have prolintane floating around for myself, but that's another story.

Cavatica - 23-1-2015 at 04:25

Thanks, ChemistryGhost. In what sense would you consider ethanol truly toxic?

tandpasta - 24-1-2015 at 07:06

Well, everything is toxic in large enough doses. Since ethanol is what common folk call alcohol, it poses no threat of toxicity at all to healthy adults. The caffeine will kill the person long before any residual ethanol will pose any risk.

But maybe ChemistryGhost can explain why he considers 2-methyl-2-butanol less toxic than ethanol. As the oral (rat) LD50 is 1000 mg/kg for 2-methyl-2-butanol and 7060 mg/kg for ethanol.

Seems to me like ethanol is 7 times less toxic.

Mailinmypocket - 24-1-2015 at 07:20

Is the toxicity of the solvent concerning because the caffeine will be consumed? Or are you just trying to avoid using certain solvents for environmental reasons?

ChemistryGhost - 30-1-2015 at 11:50

Quote: Originally posted by tandpasta  
Well, everything is toxic in large enough doses. Since ethanol is what common folk call alcohol, it poses no threat of toxicity at all to healthy adults. The caffeine will kill the person long before any residual ethanol will pose any risk.

But maybe ChemistryGhost can explain why he considers 2-methyl-2-butanol less toxic than ethanol. As the oral (rat) LD50 is 1000 mg/kg for 2-methyl-2-butanol and 7060 mg/kg for ethanol.

Seems to me like ethanol is 7 times less toxic.
You are looking only at quantities, not the degree of damage that the chemical causes. Ethanol produces acetaldehyde, a toxic metabolite. After it does some damage, it then produces acetic acid, a non-toxic metabolite. 2-methyl-2-butanol produces non-toxic metabolites 2-Methyl-2,3-butanediol and 2-methyl-2-butyl gluconoride. Rodent studies show it to be 20 times stronger than regular ethanol.
2-methyl-2-butanol is less damaging. :cool:

NitreRat - 30-1-2015 at 14:25

Quote: Originally posted by ChemistryGhost  

Ethanol produces acetaldehyde, a toxic metabolite. After it does some damage, it then produces acetic acid, a non-toxic metabolite. 2-methyl-2-butanol produces non-toxic metabolites 2-Methyl-2,3-butanediol and 2-methyl-2-butyl gluconoride. Rodent studies show it to be 20 times stronger than regular ethanol.
2-methyl-2-butanol is less damaging. :cool:


To my knowledge, there are several alcohols which have lower toxicity profiles and/or more potent effects than ethanol. 2-Propanol and 4-Hydroxybutanoic acid are some examples that come to mind. Is it simply historical, cultural and infrastructural reasons that keep ethanol as most societies social-lubricant of choice? I enjoy a nice glass of wine (...or two) as much as the next person, but the number of decent human beings that have had their lives destroyed, directly or indirectly, by ethyl alcohol does make you question the current state of affairs. If I were a more suspicious man, I might say the major alcohol companies probably benefit from keeping things the way they currently are.

Ultimately this bleeds into the ancient (but ever-topical) debate on dealing with recreational drug use and so probably belongs on different forum, but it sure would be nice if anhydrous ethanol wasn't so expensive.

Chemosynthesis - 30-1-2015 at 15:26

Quote: Originally posted by ChemistryGhost  
You are looking only at quantities, not the degree of damage that the chemical causes.

You are probably best phrasing this in the light of "irreversible toxicity" as that is typically what someone in the field might call it. Valid perspective, but a distinction that is interestingly not always made.
Quote: Originally posted by NitreRat  
To my knowledge, there are several alcohols which have lower toxicity profiles and/or more potent effects than ethanol. 2-Propanol and 4-Hydroxybutanoic acid are some examples that come to mind. Is it simply historical, cultural and infrastructural reasons that keep ethanol as most societies social-lubricant of choice?
My lay opinion is that this is likely the case. Just to add another substance to the list, tert-amyl alcohol has been used as well, with higher potency than ethanol. Measuring relative alcoholic beverage content given a higher acute toxicity and thus a lower therapeutic/recreational(?) window may be slightly problematic for regulatory or industrial purposes, but certainly possible.
Quote:
I enjoy a nice glass of wine (...or two) as much as the next person, but the number of decent human beings that have had their lives destroyed, directly or indirectly, by ethyl alcohol does make you question the current state of affairs. If I were a more suspicious man, I might say the major alcohol companies probably benefit from keeping things the way they currently are.
/rquote] You would probably need to quantify different aspects of destruction, as any psychoactive alternatives would still have addictive, acute toxicity, sociological, and indirect deleterious effects (ex. drunk driving).
I also see a potential argument for the hypothetical diminishment of some chronic, irreversible toxicity as being a reasonable argument at this point.

CaptainPike - 3-2-2015 at 08:15

Check out the YouTube https://www.youtube.com/watch?v=Xzh-6ZDitQ8 video where Dr Rob Stockman does a caffeine extraction on coffee using ethyl acetate. The details of the extraction are interspersed by an handwaving professors ranting which I suppose may help some folks the first few times. But Dr. Stockman puts forth the nuts and bolts:

1) cool the coffee down – caffeine is less soluble in the cold water.
2) protonate the caffeine with the addition of a little sodium hydroxide.
3) using a separatory funnel, extract with additions of ethyl acetate.
4) they dry on a rotovap, but you could distill off most of the EtOAc then dry on a watch glass.

Personally, I like drinking the coffee and letting my body do the extraction ;)

Regarding the ethyl acetate… I have used Klean-Strip® MEK substitute, and they claim it's 100% ethyl acetate. I can imagine having concerns about using this to "strip", any organic material which might, in part, be consumed by a human. You could distill this, being mindful of the still head temperature, to at least assuage one's concern about manufacturing metals, debris, etc. making its way into the final product.


The credit goes to have2know, for this great source of inexpensive solvent/organic reagent available to the home experimenter. At least if his home is in America.

One more thing… I believe the separation is quite good using EtOAc – you can literally, "cap 'n' shake" (and vent), as long as you're willing to wait a little while for the layers to separate. I just mean that you won't wind up with a cloudy emulsion that lasts forever.

[Edited on 3-2-2015 by CaptainPike]

Caffeine Extraction

nah89 - 1-6-2015 at 14:26

Could I go about extracting caffeine from Coffee/Tea by going about a normal A/B extraction? My thought process using sodium carbonate for the basification, then pouring in toluene (I have no access to DCM), separating off the water and then adding HCl to form a salt. Since the salt would be soluble in water I would then separate off the water layer and then evaporate or should I just stop before adding the acid and evap off? I also read somewhere that I could do a recrystallization with hot acetate. Would that work?

Sorry, I'm rather new to home chemistry and I'm doing this for the sake of doing it, not consuming it. Thanks! I looked around on old threads but they all involve using DCD or chloroform.

j_sum1 - 1-6-2015 at 16:08

Well, chloroform is relatively easy to make. But I think that there are many suitable solvents. A quick tour of some youtube videos will help. My current favourite on caffeine is this one:
https://www.youtube.com/watch?v=fkhO3461T0E

Incidentally, the same youtuber also has a video on the haloform reaction which can produce chloroform. Although I haven't watched that one or some of the more recent videos.

subsecret - 1-6-2015 at 17:04

I actually extracted my caffeine into acidified water, then deprotonated the caffeine in the separatory funnel so none would be lost.