Sciencemadness Discussion Board

The road to vanillin

Serybva - 14-4-2022 at 08:33

Pursuing my exploration of the organic chemistry universe after the synthesis of a few esters I went for vanillin, smells good and can be synthesized from eugenol aka clove oil (which I find totally insane).

First eugenol has to be isomerized to yield isoeugenol.
Then isoeugenol is esterified with acetic acid to give isoeugenyl acetate.
Isoeugenyl acetate can then be oxidized with KMnO4 and hydrolized to vanillin.

So first, I have to get my hands on some isoeugenol, and after a few hours searching the internet on eugenol isomerization I've came up with this:

1 mol eugenol
6 mol KOH (apparently yields are terrible, so the more KOH you put in, the better)
8h reflux @ 150°c in glycerol

The first attempt.

50g eugenol (304.5mmol) were mixed with 102.5g of KOH (1.827mol) and and a glycerol amount that felt right in a three-necked 1l RBF.

It then was refluxed for about 8 hours at 150° and left to cool overnight.

The next day mixture had hardened a lot and was warmed and acidified to ph 1 with 23% HCl, and then extracted 3 times with 50ml of DCM.

The RBF I used for the reaction was badly etched and went directly to the bin.

TLC analysis shows that I have a lot of stuff in my DCM, so DCM was distilled off and the extract left aside in a beaker so I can figure out how to extract the isoeugenol from the tarry solid I had obtained after DCM evaporation.

At that time I had no powerful nor efficient enough vacuum pump allowing me to perform vacuum distillation, so I choosed column chromatography.

While the purification of isoeugenol I made using column chromatography was definitely instructive, it also was a real pain in the ass and quite expensive, consumed about 1.5l of n-pentane (that stuff doesn't come cheap), 500ml of ethyl acetate and 400g of silica gel (sometimes recycled between runs).
All those solvents amounts, time and money (about a hundred euros) spent so I can end up with 1.71g of almost pure isoeugenol which represents a 3.4% yield, meeeeeh.

After more research on how the yield can be improved I stumbled upon a patent that used less KOH (which hopefully won't eat my glassware), another solvent and an argon atmosphere in order to avoid the formation of tarry by products.

I also invested in a two staged vacuum pump able to run continuously to perform vacuum distillation, quicker and much cheaper in the long run.

Second attempt
Here's the protocol for the second attempt:

- 50g eugenol
- 60g KOH
- 100ml ethylene glycol (patent uses pentane-1,5-diol if I recall correctly but ethylene glycol should be ok)
- Argon gas cylinder

Eugenol, KOH and ethylene glycol were mixed in a cheap (I learned my lesson) two necked 1l RBF with a condenser on top, then I flushed the atmosphere out of the apparatus with argon flowing from the second neck of the RBF, then when I was confident that all the air in the apparatus was replaced with argon I put the thermometer in place and turned on the heating mantle for a 6h reflux.

The first 2h everything went fine, reaction mixture was colored but still transparent.

I had to stop the reaction because I had stuff to attend to, of course I was aware that while cooling the apparatus will allow air to flow inside but I though that before oxygen had a chance to reach the mixture it would have been cold enough so it won't form tarry products, lol.

When I came back 1h after stopping the reaction the reaction mixture was all dark/brown, and I knew I had fucked up but decided to carry on anyway and turned the heating mantle on again for 4 more hours of reflux after flushing the air again with argon.

After allowing the reaction mixture to cool under argon stream it was acidified and extracted using toluene, toluene was distilled for recycling and crude isoeugenol fractionally distilled under vacuum (at something like 4mmHg) through a packed column.

After 1h of vacuum distillation nothing did came over, indeed all my isoeugenol was in the column packing, I though I would have more than that but here we are, another ridiculously low yield.
Pump oil smells isoeugenol despite the cold trap, the vacuum is probably too high I won't grease the apparatus joints next time.

TLC shows that I have all the isoeugenol went in the column packing, just a tiny bit in the pump oil and nothing in the crap I was left with after the distillation.

After rinsing the column packing with acetone and evaporating the said acetone, I'm left with 9.15g of nearly pure isoeugenol, which represent a 18.3% yield, way better than the pitiful yield I got from the first attempt.

I think argon atmosphere really help preventing tarry crap formation and greatly improves the yield.

However this is still far from the 70% yield claimed in the patent.

The next time I'll try another method with a ruthenium based catalyst, 156€/10g isn't cheap but potassium hydroxide and 3-necked RBF aren't either.
Moreover the aforementioned method is said to have 98~ish% yields and of course catalyst can be reused.

That's all for now


laserlisa - 15-4-2022 at 02:31

Fun project!
Is it really necessary to protect the phenol with an acetyl group for the double bond cleavage though?

PS. Stirring estragole in methanol with a catalytic amount of PdCl2 overnight gave quantitative yield for me. Would recommend!

Boffis - 16-4-2022 at 06:41

Interesting experiment and nice write up. I have often seen this type of rearrangement referred to wonder how easy they are in real life, so its nice to see someone report actual results. I look forward to your report on the use of a ruthenium catalyst. By the way what is the ruthenium compound?

@laserlisa, do you have a reference or some more involved details of the palladium chloride catalysed rearrangement?

Lionel Spanner - 18-4-2022 at 04:02

I can remember doing this very synthesis many years ago in undergraduate labs, starting from steam-distilled clove oil. The second step (esterification) is not necessary - permanganate will oxidise isoeugenol directly to vanillin.

At one point in the synthesis we used nitrobenzene as a solvent, but I forget which step this was.

[Edit] Please ignore the above. The final step was oxidation with nitrobenzene, no permanganate involved - which could be done directly from isoeugenol.

[Edited on 18-4-2022 by Lionel Spanner]

Serybva - 25-4-2022 at 00:31

Quote: Originally posted by laserlisa  
Fun project!
Is it really necessary to protect the phenol with an acetyl group for the double bond cleavage though?

PS. Stirring estragole in methanol with a catalytic amount of PdCl2 overnight gave quantitative yield for me. Would recommend!

It doesn't absolutely have to but yields are low since the Oh group will be oxidized as well by KMnO4


Quote: Originally posted by Boffis  
Interesting experiment and nice write up. I have often seen this type of rearrangement referred to wonder how easy they are in real life, so its nice to see someone report actual results. I look forward to your report on the use of a ruthenium catalyst. By the way what is the ruthenium compound?

@laserlisa, do you have a reference or some more involved details of the palladium chloride catalysed rearrangement?

Thanks !
This is triruthenium dodecacarbonyl but other ruthenium or osmium catalysts are suitable, here's the patent https://patents.google.com/patent/US4138411A/en

[Edited on 25-4-2022 by Serybva]