I found this posted on another board, the author was reporting on the demethylation of vanillin, but looking at the patent it is based on it has a
much wider application. The original poster comments that it might be applicable to iso-eugenol, which I have my doubts about.
Here is what what was said to have worked for vanillin.
"A solution of 212g (1.6 M) AlCl3 (anhydrous) and 63.28g (0.8 M) Thiourea in 1L of DCM was made and let stir for 30 mins at room temperature,
protected from atmospheric moisture. To this a solution of 60.86g (0.4 M) of Vanillin in 300 ml DCM was added. The solution was brought to reflux left
with good stirring for 5 hours. During which time the mixture had turned turned slightly darker. The reaction was cooled to room temperature and made
acidic with 15% HCL. The HCL was added slowly and in small amounts. The resulting solution had a yellowish clear upper layer, a red middle layer, and
small brown oilly bottom layer.
The aqueous layer as removed. The red DCM layer was washed with 2 x 300 ml of H20, and the H20 added to the removed aqueous layer. The brown layer was
washed with dissolved in 300 ml ethyl acetate and filtered. The aqueous layer was extracted with 3 x 300 ml ethyl acetate. All the ethyl acetate was
combined and the solvent was removed over low heat. A large yellow precipitate with a melting point of 157C and weighting 49.6g (81.4% yield) was
left.
Discussion:
The real trick here is that the bulk of di-hydroxy product is found in the aqueous layer, NOT the DCM layer. The DCM contains a small amount of the
di-hydroxy and all the unreacted vanillin. The water washed of the DCM is probably un-necessary, and may even hurt your purity of your product. It was
believed the brown layer may contain some precipitated product, hence the extraction, it too could probably be eliminated. "
If it is ok, I will post a link to the original thread. From my knowledge it looks like a legitimant procedure, but it was found on a somewhat non
scientific message board.
Green Impstoichiometric_steve - 21-12-2008 at 00:06
If it is a patent, then it may be just as well plain BS. You never know what you get with patents.greenimp - 21-12-2008 at 07:30
well i read the patent, and its not BS. Hard Acid/Soft Base theory is big in this field of work. It sounds like the thiourea is acting like a weak
nucleophile, just like a RSH, and attacking the CH3 group.
The reaction basically is AlCl3 -> electrons on oxygen of ether group -> AlCl3(-)-RO(+)CH3 -> AlCl3(-)-RO(+) + H4N2-C-S-CH3?????-> add HCL
-> AlCl3 + ROH
There is very good papers in the literature describing thiol groups used in demethylation reactions. Dodecane thiol being one of the more practical
reagents. However, I can not find one paper describing thiourea used in this manner. It makes sense considering thioureas resonance structures, but
I don't know why someone who patents this idea wouldn't also publish the idea.
Green Imp
edited to add the carbon in thiourea
[Edited on 21-12-2008 by greenimp]
[Edited on 21-12-2008 by greenimp]stoichiometric_steve - 21-12-2008 at 11:19
where is there a thiol in thiourea?greenimp - 21-12-2008 at 12:53
I didn't mean to imply there was one. Just that it was performing the same role as a thiol in the reaction. Thiourea has two resonance forms as far
as i know, thiourea and isothiourea. The later has a positive charge on the sulfur atom, and a negative on one of the nitrogen atom.
What I don't understand is what happens to the thiourea during the course of the reaction. What I posted above is my best guess.
Green ImpPanache - 21-12-2008 at 17:07
i have all those reagents and some time over summer, i might give it a whirl. If it works it will be the first time a patent i have tried worked, lol.greenimp - 21-12-2008 at 17:50
One step ahead of you. I should have some results soon. I can say this much the
reagent pair forms as described.
V16greenimp - 22-12-2008 at 01:04
Just finished a run of this. I can say that I believe it works as stated in the patent. Too tired to write a report up now, but I am
re-crystallizing the product right now, and will do a melting point in the morning. If it looks good I'll write a full report.
Green ImpNerro - 22-12-2008 at 03:30
Wouldn't the aldehydde be activated by the AlCl3? It would seem like an entirely possible route to let the sulphur attack on the resulting carbocation
to form some freakish adduct...Nicodem - 22-12-2008 at 06:27
Quote:
Originally posted by Nerro
Wouldn't the aldehydde be activated by the AlCl3? It would seem like an entirely possible route to let the sulphur attack on the resulting carbocation
to form some freakish adduct...
The nucleophilic addition of thiourea on the aldehyde is reversible (in other words irrelevant for the reaction outcome). On the other hand, the
nucleophilic attack on the methoxy group is not reversible.greenimp - 22-12-2008 at 14:14
Nicoderm is right, the attack on the methyl group drives the reaction forward.
Anyway, here is my attempt.
53g (0.4M) AlCl3 (anhydrous) was placed in a 2L round bottom flask that had been rinsed with acetone and heat dried to remove all moisture. To this
250 ml of dry DCM was added and stirring was begun with a magnetic stir bar. The solution turned dark red. To this 23g of Thiourea was added in
small amounts. After its addition was complete, a reflux condenser with a drying tube was attached to the 2L flask. As the Thiourea went into
solution it came to boil and over 30 mins changed color to a brown with green overtones. The solution was allowed to cool to room temperature and a
solution of 16g vanillin in 50ml DCM was prepared. Over the course of 25 mins the vanillin solution was slowly added to the flask at about 2 ml/min.
As the vanillin was added 2 layers initially formed in the flask, a thick brown bottom layer and white/brown upper. After about 15 mls addition the
solution came to boil, and a small yellow precipitate began to form on the wall of the flask. By 30 mls the brown layer had broken up, but a
yellow-orange precipitate had formed to the extent that magnetic stirring was no longer possible. Once the entire vanillin solution had been added,
the flask was maintained at reflux with external heating for 3.5 hours, at the end of which very significant amounts of yellow/orange precipitate had
collected in the bottom and sides of the flask.
After cooling in a cold water bath 250 ml of 5% HCL was added 10mls at a time with very good manual stirring. Great care was taken to ensure the
solution did not over heat the DCM. The addition of the HCL caused a large single clump of deep orange precipitate to form. The flask was allowed to
stand for 30mins with good stirring. During which time the orange precipitate broke up, turned yellow then brown, then grey. It was removed by vacuum
filtration. After drying the filtrate was grey green, weight 9.34 grams, for a yield of 67% of theoretical. I did a crude melting point test on a
small sample, using a digital thermometer on hot plate, and it begun to melt at 153, and fully melted at 157. Protocatechualdehyde decomposes at
152-154.
An attempt was made to clean up the product. 4 grams of filtrate was re suspended in 250 ml H20. It went into solution easily, creating a brown
solution. This was extracted with 3 x 125 ml ethyl acetate. The ethyl acetate combined and mostly removed by distillation at atmospheric pressure.
The resulting residue was let dry in a crystallizing dish overnight. What was left was a red/black gummy solid.
I should also say, that my ethyl acetate comes from distilling non-acetone nail polish remover. Shipping costs have gotten so high that it is cheaper
to get it this way now. The rest of the reagents are lab grade.
I think my crystallization failed. Most protocols call for recrystalization from ether, but I didn't have any, and the original poster of this
document crystallized theirs from ethyl acetate. I saved half the product, so I will try a different method, maybe just from a small volume of
boiling water?
Finally, the patent states the reagent pair forms a "olive-drab" color, which i really didn't get. To test this i combined 2.5g AlCl3 with 1g
thiourea in a small 50ml beaker. Upon stirring the solids with a glass rod it turned to a oily olive drab liquid.
So what do you guys think?
Green ImpKlute - 22-12-2008 at 15:03
Very nice work!
I have had good results by recrystallizing aroamtic aldehydes with ethyl acetate and petroleum ether. I am sure it is a cheap and avaible alternative
for ether in your case, worth a try.
Impatient to hear more! Keep up the good work!greenimp - 22-12-2008 at 15:20
Thanks. I think the ethyl acetate should work too, but something went wrong. It may have been distilling it at atmospheric temperature. From the
melting point test, this molecule doesn't like the heat very much.
Also, I washed the filter cake with some acetone, and presto, green color all gone. Now just a white-brown crystal. Which is close to what I have
seen it described as.
Now to see if the yield can get better. The patent uses a final step where the reaction is swapped into 1,1,2,2-tetrachloroethylene and heated at 70
C for a further hour. This got their yields up from 65-70% to over 90%!! Tetrachloroethylene is not something I have though. Also, I will have to get
some more AlCl3 first. Does anyone have a write up for the procedure for making it from Al in xylene with dry HCL gas? I looked but can only find
people talking about it in passing.
But all I can say is this is a no BS patent. Works as stated.
Green ImpQuantum_Dom - 24-12-2008 at 23:52
Sciencemadness is by far the best chemistry forum that I have ever consulted so far.
Thank you very much for this important thread !appetsbud - 26-12-2008 at 06:26
I found this whilst browsing through an A-level textbook from 1969, AlCL3 synthesis is mentioned briefly as a side note to the synthesis of methyl
phenyl ketone.
"Aluminium may be prepared by passage of dry chlorine over heated, dry aluminium but it is probably more satisfactory to purchase it fresh; once the
bottle has been opened it is apt to deteriorate unless very well sealed up."
i just thought i would mention it, as i was reminded of this thread when i came across it. It was found in Practical Chemistry for Advanced Level, by
A.E.Somerfield497 - 28-12-2008 at 02:13
Quote:
Tetrachloroethylene is not something I have though
Tetrachloroethylene has to be one of the most OTC chlorinated solvents around. They sell it fairly pure (and cheap!) by the gallon at any auto parts
store such a Nappa, used as a brake cleaner.497 - 31-12-2008 at 18:19
Quote:
The patent uses a final step where the reaction is swapped into 1,1,2,2-tetrachloroethylene and heated at 70 C for a further hour. This got their
yields up from 65-70% to over 90%!!
After looking at the patent, I doubt using TCE would help you much with vanillin. I think the only reason they use it is because the substrate is a
big polycyclic thing that won't dissolve well in other solvents (according to them).
I guess it's worth a try since TCE is so easy to get. I'd like it if someone proved me wrong...greenimp - 31-12-2008 at 21:16
I think its more than that. The specifically say the reagent pair is insoluble in the tetrachloroethane (TCE). Plus they dissolve their substrate
DCM to begin with. My theory is the the TCE soluablizes the AlCL3-O-R, allowing some of the unreacted substrate to be accessed.
I picked up some TCE today, so I will do a small scale run tomorrow or the next day and let you all know how it goes.
Green Imp497 - 31-12-2008 at 22:27
Hmm, you're right, I read over it in more detail and that seems to be what happens. But why don't they do it (heat in TCE) for the other examples?greenimp - 31-12-2008 at 22:39
Most of the other examples are liquids i believe. No need to use the DCM.
Green ImpFormula409 - 23-1-2009 at 02:11
Someone should see if this works with AlI3. It's much easier to generate in my opinion.
Formula409.stoichiometric_steve - 24-1-2009 at 03:21
Quote:
Originally posted by Formula409
Someone should see if this works with AlI3.
It most likely does. The difference is that you might get Methyl Iodide instead of Methyl Chloride which is a different hazard altogether.peach - 26-1-2009 at 18:28
Thiourea is a solid if you get it pure, and the AlCl3 is obviously solid as well. The substrate will likely provide some dissolving power, but I
expect the kenetics will be superior if the mix is dissolved to a homogeneous liquid phase.
Anhydrous AlCl3 is a lot easier than AlI3. There's a writeup in Vogel I think. HCl is far superior to Cl2 temperature wise.
[Edited on 26-1-2009 by peach]
[Edited on 26-1-2009 by peach]
[Edited on 26-1-2009 by peach]Formula409 - 27-1-2009 at 00:01
Quote:
Originally posted by peach
Anhydrous AlCl3 is a lot easier than AlI3. There's a writeup in Vogel I think. HCl is far superior to Cl2 temperature wise.
Oh really? AlI3 is just a matter of refluxing the elements in a solvent such as hexane, benzene, etc. Refluxing in DCM would even (theoretically)
enable the AlI3 to be used in the demethylation without any further work up.
Formula409.peach - 27-1-2009 at 07:21
Yes, really.
If you have a regular supply of iodine that's great, but HCl is far easier for most.
Anhydrous AlCl3 is as difficult as bubbling the gas through DCM loaded with Al powder and, as with AlI3, it's then ready for use in situ.Siddy - 30-1-2009 at 23:01
This is a good find, the clean up seems a little complex due to the solubility of the di-phenol product.
If one were to be isolating the product to perform a methylenation reaction on the Catechol, could one not simply add hydroxide while the crude
product is still in the DCM/AlCl3 mix - or would the AlCl3 interfere?497 - 31-1-2009 at 18:11
Uhh you'd probably end up with a gelatinous mess of Al(OH)3...greenimp - 31-1-2009 at 18:14
The hydroxy-vanillin is not soluble in the 5% aq. layer, nor the DCM layer (the vanillin is though)
Basically you get a solid, filter it out, and re-crystallize it. Not much easier than that.
The tetra-chloroehtane step boosts yields considerably.
I can say this much the
reagent pair forms as described.