Sciencemadness Discussion Board

Caffeine extraction low yield?

Booze - 28-3-2017 at 17:14

So I got bored today, and so I went to a hardware store and distilled paint stripper and got relatively pure DCM. I think there might be a small methanol contamination, but that is fine by me. I used this DCM to extract caffeine from coffee.

First I mixed 10 grams of ground coffee (I gave the beans a very light roast to keep most of the caffeine) with 5 grams of sodium carbonate. The sodium carbonate would make a lot of the things in the coffee soluble in water. I dissolved these in 60 mL of water.

I put a round bottom flask filled with snow and placed it on the beaker to limit water loss. I boiled it for 15 minutes. After that, I filtered it do get rid of the coffee grounds. It was painfully slow. I poured in 100 mL of hot water through the filter.

Then I added 15 mL of the DCM I made (I don't have a seperatory funnel, so I just mixed these in a beaker and added them to my addition funnel). I did this 3 times. The caffeine should be much more soluble in DCM than water.

Now I took my 45 mL of DCM and I added some sodium chloride solution to get rid of the water. I did the same process where I mixed the solutions and seperated them with my addition funnel.

I boiled the DCM away, and was left with basically no caffeine. With the little that was there, I scraped it out and put it on my scale, and the scale wasn't good enough to measure it, and it can go to .01 grams. What gives with the low yield?

CharlieA - 28-3-2017 at 18:02

From 10 g of coffee, you can expect ca. 135 mg caffeine. Assuming 100% recovery, you should have been able to weigh (I know, "mass") this amount with your scale. But the assumption of 100% recovery is optimistic. Maybe you could try again with more coffee (say 100g or more).

Booze - 28-3-2017 at 18:36

Quote: Originally posted by CharlieA  
From 10 g of coffee, you can expect ca. 135 mg caffeine. Assuming 100% recovery, you should have been able to weigh (I know, "mass") this amount with your scale. But the assumption of 100% recovery is optimistic. Maybe you could try again with more coffee (say 100g or more).

Okay. I wasn't sure the yield would be so low, but whatever.

Also why was this thread moved? This is organic chemistry, not beginning.

Texium - 28-3-2017 at 18:41

Quote: Originally posted by Booze  
Also why was this thread moved? This is organic chemistry, not beginning.
It's an unreferenced beginner level experiment. Therefore it goes in Beginnings.

NEMO-Chemistry - 30-3-2017 at 17:52

Caffeine pills are not as much fun, but they do give a higher yield. here in the UK real coffee beans are fairly pricey, well considering the rest of the world :D.

When I did this experiment, I used the grounds from a coffee shop (a fairly large amount. I made coffee first in batches using a soxhlet extractor, I dont remember how many times I filled my large soxhlet up with grounds.

But I do remember the disappointment with the yield!! I then used the energy pills, it was much better yield wise, but TBH no where near as much fun.

Booze - 31-3-2017 at 08:39

Quote: Originally posted by NEMO-Chemistry  
Caffeine pills are not as much fun, but they do give a higher yield. here in the UK real coffee beans are fairly pricey, well considering the rest of the world :D.

When I did this experiment, I used the grounds from a coffee shop (a fairly large amount. I made coffee first in batches using a soxhlet extractor, I dont remember how many times I filled my large soxhlet up with grounds.

But I do remember the disappointment with the yield!! I then used the energy pills, it was much better yield wise, but TBH no where near as much fun.

That's too bad. I bought my beans from Costa Rica (they were green) and I roasted them until they were soft enough to crunch, but still had their caffeine.

Hexavalent - 31-3-2017 at 12:36

Quote: Originally posted by Booze  

After that, I filtered it do get rid of the coffee grounds. It was painfully slow. I poured in 100 mL of hot water through the filter.


By how much did the aqueous solution cool before you added the DCM? The solubility of caffeine in water drops dramatically with decreasing temperature, from 666 mg/mL at boiling to 16 mg/mL at room temperature (Sigma Aldrich).
Quote: Originally posted by Booze  

Then I added 15 mL of the DCM I made (I don't have a seperatory funnel, so I just mixed these in a beaker and added them to my addition funnel). I did this 3 times. The caffeine should be much more soluble in DCM than water.


How intimately did you mix the DCM and aqueous phases? Extractions generally require a good mixing of the two phases, hence why shaking in a separating funnel is preferable. Did you get any emulsion formation? I'm told that DCM has a tendency to do this in the presence of base and will clearly affect your recovery of caffeine.

Don't forget that any methanol present in your DCM will migrate to the aqueous phase, possibly increasing the solubility of caffeine there. What boiling fraction of your paint stripper did you collect as DCM?
Quote: Originally posted by Booze  

I boiled the DCM away, and was left with basically no caffeine. With the little that was there, I scraped it out and put it on my scale, and the scale wasn't good enough to measure it, and it can go to .01 grams. What gives with the low yield?


A better idea would have been to weigh the empty vessel prior to removing the solvent, then re-weighing afterwards (and hence finding the mass of solid, by difference).

What balance are you using? I've had a few cheap Chinese ones in the past that claimed to have milligram (0.001 g) resolution but couldn't detect anything weighing less than around 150 mg. Have you used this scale before with such small masses?


[Edited on 31-3-2017 by Hexavalent]

Booze - 31-3-2017 at 13:00

Quote: Originally posted by Hexavalent  
Quote: Originally posted by Booze  

After that, I filtered it do get rid of the coffee grounds. It was painfully slow. I poured in 100 mL of hot water through the filter.


By how much did the aqueous solution cool before you added the DCM? The solubility of caffeine in water drops dramatically with decreasing temperature, from 666 mg/mL at boiling to 16 mg/mL at room temperature (Sigma Aldrich).
Quote: Originally posted by Booze  

Then I added 15 mL of the DCM I made (I don't have a seperatory funnel, so I just mixed these in a beaker and added them to my addition funnel). I did this 3 times. The caffeine should be much more soluble in DCM than water.


How intimately did you mix the DCM and aqueous phases? Extractions generally require a good mixing of the two phases, hence why shaking in a separating funnel is preferable. Did you get any emulsion formation? I'm told that DCM has a tendency to do this in the presence of base and will clearly affect your recovery of caffeine.

Don't forget that any methanol present in your DCM will migrate to the aqueous phase, possibly increasing the solubility of caffeine there. What boiling fraction of your paint stripper did you collect as DCM?
Quote: Originally posted by Booze  

I boiled the DCM away, and was left with basically no caffeine. With the little that was there, I scraped it out and put it on my scale, and the scale wasn't good enough to measure it, and it can go to .01 grams. What gives with the low yield?


A better idea would have been to weigh the empty vessel prior to removing the solvent, then re-weighing afterwards (and hence finding the mass of solid, by difference).

What balance are you using? I've had a few cheap Chinese ones in the past that claimed to have milligram (0.001 g) resolution but couldn't detect anything weighing less than around 150 mg. Have you used this scale before with such small masses?


[Edited on 31-3-2017 by Hexavalent]


I let it cool to room temperature. I didn't notice any presipitation occuring, so I went on with the DCM extraction, and I did mix it pretty good. My scale is a jewlery one, and I have been using it to weigh silver beads I made.

Hexavalent - 31-3-2017 at 13:13

Quote: Originally posted by Booze  

I let it cool to room temperature. I didn't notice any presipitation occuring, so I went on with the DCM extraction, and I did mix it pretty good.


If you had 160 mL of water present, then that should be enough to dissolve all of your theoretical mass of caffeine at room temperature, so disregard my earlier point about solubility.

Again, what boiling point fraction of DCM did you collect? Did you notice any emulsion formation when attempting to separate layers?

Booze - 2-4-2017 at 08:21

Quote: Originally posted by Hexavalent  
Quote: Originally posted by Booze  

I let it cool to room temperature. I didn't notice any presipitation occuring, so I went on with the DCM extraction, and I did mix it pretty good.


If you had 160 mL of water present, then that should be enough to dissolve all of your theoretical mass of caffeine at room temperature, so disregard my earlier point about solubility.

Again, what boiling point fraction of DCM did you collect? Did you notice any emulsion formation when attempting to separate layers?

I did a super amateur job and just distilled the paint thinner until it was a goop. I knew it was DCM because it boiled at a very low teperature and it says so on the label. It did also say methanol, but that was like 7% and I'm not too worried about it. There was also probably some other stuff, but I threw away the box.

LD5050 - 2-4-2017 at 10:04

Next time after you distill the DCM from the paint stripper throw it in a separating funnel and do a few water washings. The methanol is soluble in H2O and will remove all of it from the DCM. After dry the DCM with a drying agent and you will be left with some pretty pure DCM. VERY easy.

Hexavalent - 2-4-2017 at 10:28

Quote: Originally posted by LD5050  
After dry the DCM with a drying agent and you will be left with some pretty pure DCM. VERY easy.


Best choice would be calcium chloride, which forms adducts with the lower alcohols and thus will remove any remaining methanol from the DCM.

LD5050 - 2-4-2017 at 20:59

[/rquote]

Best choice would be calcium chloride, which forms adducts with the lower alcohols and thus will remove any remaining methanol from the DCM. [/rquote]

Good call I didn't even think of that, I usually use MgSO4 but I will be now using calcium chloride.