Author Topic: Eugenol-->Safrole directly? (Read 12035 times)

El_Zorro · « **on:** March 29, 2002, 09:48:00 AM »

Alright. I'd like to start a discussion about going from eugenol directly to safrole. In the past, all routs to safrole from eugenol have involved demethylating the methoxy group, then methylating the two phenols. Why?, I asked. Eugenol already has everything safrole has, the only difference is, the 4 position phenol isn't connected to the 3 postion methoxy. So why go through several steps to get to safrole, when you could just hitch up the methoxy and the phenol in one step?

In the methylation step from catechol-->benzodioxole, a base is used to deprotonate the phenols in order to react the with the methylene halide.

Now, can't the phenol AND the methyl part of the methoxy group be deprotonated at the same time, and with perhaps a sodium halide, or even an elemental halide to react and close the bridge? I've been reading that in order to deprotonate a methyl group if you use a base with a higher pK_a than the C-H bond. Now, what I don't know is, what is the pK_a of the C-H bods in the 3 position methoxy group of eugenol.(or vanillin) Comments? What bases could be used to deprotonate both the methyl group and the phenol simultaneously? Comments?

I sell crack for the CIA

terbium · « **Reply #1 on:** March 29, 2002, 06:35:00 PM »

Your superbase would be more likely to split the ether than abstract a proton. So, you would still get the dihydroxypropenylbenzene.

Aurelius · « **Reply #2 on:** March 30, 2002, 01:59:00 AM »

So what's the problem with getting dihydroxypropenylbenzene? just methylenate the stupid thing and there's isosafrole.

PrimoPyro · « **Reply #3 on:** March 30, 2002, 06:05:00 AM »

Do you know of any suppliers other than Aldrich for that compound?

Im not asking for any, my question is as simple as it is stated: Do you know of any suppliers for that compound other than Aldrich (or other huge manufacturers)?

I don't. I've looked because I too was very very very interested in that compound as a source.

PrimoPyro

Vivent Longtemps La Ruche!

Chromic · « **Reply #4 on:** March 30, 2002, 08:42:00 AM »

Last I checked eugenol is an allylbenzene (ie a 2-propenylbenzene) like safrole, not a propenylbenzene (ie a 1-propenylbenzene) like isosafrole. Like terb said, O-demethylation will occur before a hydrogen will leave from the methyl group, strong lewis acids/bases will cleave the ether bond... bridging it usually done with base-catalyzed methylenation with dichloromethane...

Why try and reinvent the wheel when there's so many refs showing this works as is?

PrimoPyro · « **Reply #5 on:** March 30, 2002, 08:52:00 AM »

But that phrase halts progress. It is anti-scientific.

Vivent Longtemps La Ruche!

Chromic · « **Reply #6 on:** March 30, 2002, 09:03:00 AM »

Although I haven't looked at it, the reaction mechanism of bridging dichloromethane to two phenoxides ortho to each other is likely S_N2-type chemistry.

You have the phenoxide attacking dichloromethane, and one of the chlorine groups leave (I bet this would work even better with dibromomethane or diiodomethane as they're better leaving groups)... then you have an intermolecular williamson ether-type synthesis going on where the other phenoxide attacks the chloromethyl group that was just formed. This works because chlorine can turn into chloride (Cl-) which makes an excellent leaving group. Hydrogen would have to turn into hydride (H-) to make a good leaving group. Hydrides don't form easily, from what I know, unless you have a very electropositive species (like alkali metals) hydrogen is always going to be floating around as a proton (H+). The problem with El_Zorro's approach is that it's not possible to turn the hydrogen on the methyl into a good leaving group.

Or should I say... I don't think that's possible. I always stand to be corrected.

Btw, point taken about furthering discussion/research...

Rhodium · « **Reply #7 on:** March 30, 2002, 09:23:00 AM »

PP: Source for what compound? Eugenol is all around us for just a few dollars a liter as clove oil.

PrimoPyro · « **Reply #8 on:** March 30, 2002, 09:44:00 AM »

A source for the compound stated in the post right before mine.

Vivent Longtemps La Ruche!

terbium · « **Reply #9 on:** March 30, 2002, 10:20:00 AM »

A source for the compound stated in the post right before mine.
The source for dihydroxypropenylbenzene is the O-demethylation of eugenol.

PrimoPyro · « **Reply #10 on:** March 30, 2002, 10:24:00 AM »

I know that terbium.

Perhaps I am not being clear enough. By "source" I do not imply origin, I imply "retailors of" said compound, that is why I specified Aldrich as an example, and not eugenol as an example.

My post was in its entirety a reply to the one preceeding it, which obviously implied purchasing said chemical, not manufacturing it.

PrimoPyro

Vivent Longtemps La Ruche!

Rhodium · « **Reply #11 on:** March 30, 2002, 10:29:00 AM »

It is very unstable to oxidation, and thus not available from many places, except the microsoft/AOL companies of chemsuppliers.

terbium · « **Reply #12 on:** March 30, 2002, 10:30:00 AM »

Last I checked eugenol is an allylbenzene (ie a 2-propenylbenzene) like safrole, not a propenylbenzene (ie a 1-propenylbenzene) like isosafrole.
Yes, I was using the term propenylbenzene in an attempt to be ambiguous about the position of the double bond.

I was attempting to be ambiguous about the position of the double bond so that I could ignore whether or not a certain O-demethylation process might also isomerize the double bond.

El_Zorro · « **Reply #13 on:** March 30, 2002, 11:59:00 AM »

I realize that there are well-documented synths that involve demethylating the eugenol then methylating it, but they use a two-step synth, and I was looking for a one step, in order to minimize time and chems.

Alright, if the superbase would cleave the ether group and form dihydroxyallylbenzene, and the methylation is catylized by a base, could not the entire synth be performed in one step? Hmmmm..... perhaps not, because polymerization during the methylation step is minimized by dripping the dihydroxy into the DCM. Damn. Even if it did work I have no doubt that it would result in low yeilds.

Second question. What is the pK_a of the methyl section of the methoxy group in eugenol? Without knowing we are all just guessing as to what would happen.

I sell crack for the CIA

terbium · « **Reply #14 on:** March 30, 2002, 12:11:00 PM »

What is the pKa of the methyl section of the methoxy group in eugenol?
This question does not make sense.

Without knowing we are all just guessing as to what would happen.
You are guessing. Needlessly now, I might add, since you have been told what would happen and had pointed out to you at least two major conceptual errors in your proposed process.

Chromic · « **Reply #15 on:** March 30, 2002, 08:07:00 PM »

It's got to be two steps unless there's another way to look at it... but maybe it could be done in one pot since at the end you'll likely vac distill it to remove the impurities that would be carried through.

If you use pyridine.hcl in say DMSO to demethylate, then basify with excess NaOH to form pyridine freebase and the phenoxide, you should be able to do the methylenation in the same pot (well, not really, it would mean pouring the rxn mix into an addition funnel and dripping it into DMSO with DCM, but you know what I mean..). I don't think the pyridine would interfere with the rxn. I'm sure Rhodium will swear and curse at me for making this recommendation (purify dem intermediates... grumble, grumble...) but, hey if you want to save time and have succeeded with the two pot synthesis of safrole from eugenol, why not try it out? We need folks like you to do the research!

Rhodium · « **Reply #16 on:** March 30, 2002, 08:14:00 PM »

My instinct would say the yields/purity would be higher using a two-step rxn. Demethylate with Pyr*HCl in the microwave, and then methylenate using the two-phase PTC reaction on my page.

El_Zorro · « **Reply #17 on:** March 30, 2002, 08:59:00 PM »

Yeah, maybe you're right. This all started when I saw that picture in

Post 238753

(Rhodium: "MDMA from only aliphatic precursors", Serious Chemistry), which basically showed closing the methylenedioxy ring on a eugenol derivative with a 75% yeild. I really haven't been able to get it out of my mind since. Does that SynGen program give references to the reactions it uses?

I sell crack for the CIA

foxy2 · « **Reply #18 on:** March 30, 2002, 10:27:00 PM »

I know I've seen many similar reations with a higher yeild than 75%

Those who give up essential liberties for temporary safety deserve neither liberty nor safety

El_Zorro · « **Reply #19 on:** March 30, 2002, 10:44:00 PM »

Look up

I sell crack for the CIA

Osmium · « **Reply #20 on:** April 02, 2002, 03:31:00 AM »

If you have to use shitty reagents like pyridine hydrochloride to demethylate cheap stuff like eugenol then better forget about this whole scheme right now.
Using difficult to aquire reagents like this one in the first step of a multi-step synth is ridiculous. Consider it a scientific curiosity, but not a working and worthwhile procedure.

I say it again: use aq. H2SO4 to demethylate. Anything else is too expensive, smelly, suspicious, costly and complicated.

I'm not fat just horizontally disproportionate.

Rhodium · « **Reply #21 on:** April 02, 2002, 03:57:00 AM »

Os: I have said it so many times before - H2SO4 should polymerize the ass of eugenol, unless you have at least a 1800's journal article from Uzbekistan to back it up...

Osmium · « **Reply #22 on:** April 02, 2002, 04:34:00 AM »

I don't know what exactly will happen to eugenol, but even if it produce only a 30% yield this is the way to go, since eugenol can be had for less than €20/liter.

I don't have any refs right now, but I have seen several ones over the years. Strong aq. acids will demethylate phenolethers. Look what Mr. Shulgin says:

The amphetamine metabolite, 4-hydroxyamphetamine hydrochloride (4-HA), was prepared by heating 5.0 g 4-MA in 20 mL concentrated HCl at 15 lbs/in. After recrystal-lization from aqueous EtOH, the product weighed 3.8 g and had a mp of 171-172 °C.

http://www.erowid.org/library/books_online/pihkal/pihkal097.shtml

I've seen several other reactions where phenolethers were refluxed with H2SO4, and at least partial demethylation took place. para MeO will hydrolyse the fastest, ortho-MeO the slowest. But they all will eventually come off.

If eugenol won't work satisfactorily then use something else. Vanillyl-methyl-ketone for example.

I'm not fat just horizontally disproportionate.

Aurelius · « **Reply #23 on:** April 02, 2002, 05:12:00 PM »

Rhodium? you've been to Uzbekistan? how interesting

Osmium, have you tried the rxn? if you supply details for how you would run the rxn, aurelius will do the work in the lab. (has more time in lab now)

And just so everybody knows, THIO COMPOUNDS SUCK!!!!!!!!

Rhodium · « **Reply #24 on:** April 02, 2002, 06:12:00 PM »

Great Aurelius! If you actually try the reaction we can settle it once and for all if the reaction is indeed workable.

I wonder if it would be best to isolate the reaction product as the diacetoxy ester, as the diphenol is so oxidation sensitive, or if the best way would be to try to isolate all non-polymeric phenols and either methylenate them directly, or actually isolate the dihydroxyallylbenzene?

We have the physical properties of all the compounds in the database and at my site.

Thio compounds does not suck, but they sure smell. No, I haven't been in Uzbekistan, but all the most revolutionary syntheses we find are always written in an unreadable language...

Chromic · « **Reply #25 on:** April 02, 2002, 07:33:00 PM »

Would isoeugenol be less prone to polymerization? (for steric and stability reasons?) Just a thought. The ketone idea is nice too. I think a reflux in 20% HCl (that's the azeotrope, right?) for a couple days might be an idea as well.

Antoncho · « **Reply #26 on:** April 02, 2002, 09:55:00 PM »

Rhodi: i am absolutely confident that all scientific journals ever published in Uzbekistan are in Russian

It's especially funny since i am currently waiting for a very interesting article from an Uzbekian journal

So this discussion kinda caught my eye...

As for eugenol - guys, H2SO4 will add across the double bond bee4 it demethylates anything, correct?

Now, to dedmethylate vanillin w/it is indeeed a possibility

/to bee continued later/

Aurelius · « **Reply #27 on:** April 03, 2002, 12:54:00 AM »

should reflux be open-air? what about distillation? the diacetoxy compound? do you mean using Acetyl chloride to give the compound or are you talking about a different compound?

Rhodium · « **Reply #28 on:** April 03, 2002, 10:37:00 AM »

The reflux should most preferably be done under inert atmopsphere (as dihydroxyallylbenzene is sensitive to oxidation). A tightly packed CaCl2 moisture protection tube should keep most of the air out of the reaction flask though.

I was referring to protecting the dihydroxyallylbenzene as the diacetoxyallylbenzene (acetylation of it all with acetyl chloride or acetic anhydride) while isolating/purifying it, it might help against decomposition. Some data on the dihydroxyallylbenzene can be found in

https://www.thevespiary.org/rhodium/Rhodium/chemistry/allylpyrocatechol.html

For suggestions on how to perform the H2SO4 demethylation, I think Osmium is more qualified to write an outline on how it should be done.

What is the best distillation vacuum you have available in your lab?

Antoncho · « **Reply #29 on:** April 03, 2002, 10:48:00 AM »

Elder bees, please excuse my ramblings, you of course know all this - and this topic has been discused zillion times bee4 -i just want to somewhat summarize the known facts for the ones still unfamiliar. Comments, please!

There are basically three types of demethylating agents (all acidic, alkaline is also possible, but i'll leave that out):

- mineral acids (HBr, HCl, H2SO4)
- metal halides (AlBr3, FeCl3, etc)
- hydroclorides of organic bases (pyridinium*HCl, anilinium*HCl)

So.... 1st, eugenol unaturated side-chain is profoundly rich in electrons, thus, any Lewis or Broensted acid, xept the mildest ones like organic amine chlorhydrates, will do things to it.

I suggested use of aniline HCl earlier - which has been mentioned as a good demethylating agent for vanillin in one of old Russian textbooks. BTW, remember, those were the pre-microwave days!

OTOH, the side-chain of MVK (methyl vanillyl ketone) is very electron-poor (acidic) - so acids won't hurt it, but there's a major BUT: any Lewis-acidic metal chloride will serve as a Friedel-Crafts catalyst for it, providing an excellent "process for producing polymer" as they say it at USPatent database

That's actually the reason for low yields in FC-acetonylation (P2P via chloroacetone).

I don't know if mineral acids can serve in the same manner (Rh? Os?), but i do know ketones react with strong acids.

Vanillin. This one, once demethylated, will of course bee more of a problem to convert to the desired product. But the odds of successful demethylation are much higher, it seems to me.

1st, it can bee demethylated w/H2SO4. There is a patent.... uh, lemme see....

I am awfully sorry to end my post so abruptly, but i just found something that definitely deserves a thread of it's own

sorry.

Antoncho

foxy2 · « **Reply #30 on:** April 03, 2002, 11:04:00 AM »

Here it is!!!
However the yeild is only 50% and it takes quite allot of LiCl. Hmm I wonder if you could recover the LiCl?

Patent EP1136481

A solution of 354 ml (2.30 mmol) of eugenol and 292 g (6.89 mol) of lithium chloride in 3.7 1 of N,N-dimethylformamide was refluxed for a total of 44 hours (h), and, after 4 h, 18 h and then 7 h, a further 292 g (6.89 mol) of lithium chloride were added each time. After cooling, 2 1 of toluene were added and the resultant precipitate was filtered off with suction and extracted with toluene. The organic extracts were combined and concentrated on a rotary evaporator. After flash chromatography (ether/pentane, 1:1, Rf = 0.37) on silica gel, 173 g (50%) of 4-allylcatechol were obtained.

Those who give up essential liberties for temporary safety deserve neither liberty nor safety

foxy2 · « **Reply #31 on:** April 03, 2002, 11:45:00 AM »

Lignosulfonates. XXVI. Demethylation of lignin during pulping.
Hayashi, Akira; Namura, Yoshifumi.
Nippon Mokuzai Gakkaishi (1967), 13(1), 24-7. (In English)
Abstract
Guaiacol, ethylated guaiacol, vanillyl alc., eugenol, ethylated eugenol, vanillin, ethylated vanillin, vanillic acid, and anisole as model compds. were cooked with 5% NaOH, 5% NaHSO3, and 5% Na2SO3 + 1% NaHCO3, and H2O, and the MeOH formed in the reaction mixt. was detd. to study the demethylation during the cooking process. All guaiacyl compds. examd. were demethylated during the sulfite and the alkali cooking process. Even ethylated compds. having no free phenolic OH groups and anisole were also demethylated. Thus, demethylation is not a special reaction of lignin, but a general hydrolytic reaction of phenolic ethers that proceeds very slowly.

Those who give up essential liberties for temporary safety deserve neither liberty nor safety

Rhodium · « **Reply #32 on:** April 03, 2002, 12:28:00 PM »

Foxy: The latter reference has already been retrieved, and it gives <5% yield when checked by GC, I don't want to think of the isolable yield...

Antoncho · « **Reply #33 on:** April 03, 2002, 12:52:00 PM »

I adore thee. This LiCl/DMF procedure is awesome! Just what we needed.

Antoncho

foxy2 · « **Reply #34 on:** April 03, 2002, 12:54:00 PM »

Rhodium
Do you mean the LiCl ref or the lignin?

Those who give up essential liberties for temporary safety deserve neither liberty nor safety

Rhodium · « **Reply #35 on:** April 03, 2002, 01:15:00 PM »

The latter - demethylation by base on lignin-like compounds.

Regarding the LiCl thing - there MUST be another usable salt...

moo · « **Reply #36 on:** April 03, 2002, 02:35:00 PM »

For some lithium salts are available, but not DMF. Any suggestions for a replacement? Of course, plain formamide could be prepared.

Rhodium · « **Reply #37 on:** April 03, 2002, 02:44:00 PM »

DMSO may be a suitable substitute.

Aurelius · « **Reply #38 on:** April 03, 2002, 06:11:00 PM »

Sorry, the strength of the vacuum pump is not known off-hand. aurelius will find out asap. how strong does it really need to be?

slappy · « **Reply #39 on:** April 03, 2002, 10:43:00 PM »

Can't you just use a slight excess of BBr₃ in DCM? I think that you can also use Lithium Triethylborohydride to demethylate anisoleic methoxy groups.

Can you use Triethylamine Hydrochloride? Pyridine is readily available and cheap (aniline too, though more expensive), just take it up in DCM, and gas with HCl, and the respective *HCl salt will fall out, then wash, dry, and you're good to go.

Antoncho · « **Reply #40 on:** April 03, 2002, 11:29:00 PM »

That was funny.

Antoncho

News: