Mescaline synthesis questions

[methymouse]

OK, so SWIM is about to get back to his lab, and he's been been strongly considering making some mescaline with all the vanillin and nitromethane he has.  I haven't done all the calculations, but here's the basic game plan, together with a few questions:

(1) Take vanillin and brominate via in situ bromine generation (H2O2/GAA/H2SO4/NaBr), a la uemura (

Post 297806

(uemura: "How to prepare hydroxyvanillin", Methods Discourse)).  The sodium bromide should be able to replace the potassium bromide without any problem (I think).

(2) Place magnesium turnings into methanol with a crystal of iodine; and reflux until all the magnesium has reacted.  This should create at best an 8% solution.  Then much of the methanol is distilled off, so as to create the equivalent of a 5M solution (in terms of the methoxide ions).  Some of the Mg(OMe)2 will crystalize out of solution, but hopefully will disolve back in as the reaction progresses and heat is added.  The workup will then be as on

https://www.thevespiary.org/rhodium/Rhodium/chemistry/mmda.mescaline.html

to make syringaldehyde; just replacing the sodium methoxide solution with a magnesium methoxide solution.

(3) The syringaldehyde is then methylated either via DMS or methyl iodide.  SWIM is leaning towards methyl iodide, since it is much less toxic and also easier to make.  The workup in Vogel with RP/I/MeOH should be fine, though SWIM's also considering generating HI in an iodine/turpentine/methanol solution, and then distilling off the MeI.  Methanol and turpentine seem to be miscible, so this should work fine.  Trimethyl phosphate would be optimal, but SWIM can't seem to find any phosphorus pentoxide needed to make it.  

(4) The 3,4,5 trimethoxy benzaldehyde is then condensed with nitromethane in GAA with an ammonium acetate catalyst in an ultrasonic cleaner, as in

https://www.thevespiary.org/rhodium/Rhodium/chemistry/nitrostyrene.ultrasound.html

(5) Reduce with Urushibara nickel as in

https://www.thevespiary.org/rhodium/Rhodium/chemistry/amph.urushibara.html

Questions:

(1) Comments on step 2 would be much appreciated.  Is MgBr2 soluble in methanol/EtOAc?
(2) Is P2O5 available at all?  SWIM hasn't been able to find a chemical company which sells to individuals who sells it, nor has he been able to find any OTC sources.  Should he bother to keep looking?
(3) Is there any problem with using Urushibara nickel to reduce nitrostyrenes?  I've only seen it used with nitropropenes, though I don't know of a reason why it wouldn't work.

Thanks in advance,
MethyMouse

[Bandil]

1)
I don't know enough about Mg-methoxide. Do you have some references on its use and preparation? I doubt that MagnesiumBromide is soluble in methanol or ethylacetate...

2)
Why bother making trimethylphosphate, when it can be bought cheaply and unsuspiciosly from most chemical companies?

3)
Not that i know of no...

Good luck!

Regards
Bandil

[methymouse]

Perhaps I'm bad at finding good chemical companies which sell to individuals.  I would probably buy it if I could find it.  I guess I'll look more and harder.  TMP is a precursor to certain kinds of chemical weapons, so it's not totally without suspicion.  That's not too worrisome, though.

UTFSE for stuff on mangesium methoxide.  It was discussed here a while ago for methoxylation as an OTC alternative to sodium methoxide.  Rhodium and Osmium both stated that there was no reason they could think of that it wouldn't work, though I don't think anyone has tried it. 

SWIM will do a small trial run on a few grams of bromovanillin before trying to push through a big batch.  If it could be made to work well, then that would be a nice advance.  Everything should be straightforward, as long SWIM can keep the reaction very dry during the methoxylation  

[Rhodium]

The workup in Vogel with RP/I/MeOH should be fine, though SWIM's also considering generating HI in an iodine/turpentine/methanol solution, and then distilling off the MeI.

Don't forget that the HI from I₂/Turpentine is untested in modern times, and that it isn't only HI doing the job of iodination when RP/I₂/MeOH is heated, but also to a significant degree PI₃ formed in situ from the elements.

Then much of the methanol is distilled off, so as to create the equivalent of a 5M solution (in terms of the methoxide ions).  Some of the Mg(OMe)2 will crystalize out of solution, but hopefully will disolve back in as the reaction progresses and heat is added.

Note that the activity (~reactivity) of the methoxide are dependant on the actual concentration of dissolved methoxide ions in the solution, and not the amount present in the flask.

[methymouse]

"Note that the activity (~reactivity) of the methoxide are dependant on the actual concentration of dissolved methoxide ions in the solution, and not the amount present in the flask."

So I suppose the whole reaction would then either take a really long reflux or be fine.  It all hinges on the soluability of magnesium methoxide in hot methanol.  That will call for some experimentation, and I won't speak more on it until next week when SWIM starts his experiments.

Would it be possible to do a melt-phase reaction, as in the melt-phase methylation writeup at Rhodium's site?  I suspect that might lead to oxidation of the magnesium methoxide, but how much of a problem this would be, I have no idea.

Regarding HI, it depends what you mean by "untested."  I was able to make something acidic with this method (and I can't imagine what other acid it could be).  If memory serves, something like a .5 molar solution in water (that was with HI dissolved in turpentine extracted to an aqueous layer).  The reaction seems to require heating, and appears to reverse itself with cooling.  I saw the color change from yellowish back to purple when the turpentine solution was allowed to cool.  But, if I reflux it with methanol, the methanol should react with the HI, kicking the equilibrium in the right direction. Of course the kinetics might be shit with a low-concentration solution and with no PI3...

Bandil:  After several hours of Google-searching, I found three chemical suppliers which sell it, none of which sell to individuals.  I suspect you may be living in a more civilized country than I.

[Rhodium]

I was able to make something acidic with this method

Sure, that I do not doubt, but have you determined how effective the reaction was, as in how many moles of HI was formed from a certain amount of iodine? Also, as 'turpentine' is a really wild mix of natural products, I don't think that you can get a reasonable yield of MeI by just heating a mixture of iodine, methanol and turpentine...

[chilly_willy]

I was also interested in mesc. at one time and found what I thought was the most feasible route.
1.  Iodovanillin here:

Post 303014

(cesium: "5-iodovanillin (Chloramine T + NaI)", Methods Discourse)
2.  Hydroxyvanillin here (from Rhodium):
"2.8g 5-Iodovanillin, 1.6g hydrated CuSO4 and 76mL 4N NaOH were refluxed(105°) for 4.5h with continuous stirring under N2. After cooling to 60-70°, the mixture was filtered under suction and the residue washed with 3x10mL hot water. The alkaline solution was cooled to 10° and acidified to pH 3-4, by the dropwise addition of concentrated HCl. During this addition the mixture was stirred continuously and the tmperature maintained below 25°C. The resulting mixture, which contained a small amount of percipitate, was extracted continuously with ether for 16h. After being dried over anhydrous MgSO4, the ether was removed (60-65°) to leave 1.5g dark gray product. All but 0.10g was dissolved in hot benzene, from which, after concentration to 75mL, 1.15g 5-hydroxyvanillin crystallized, yield 68%, mp 128-129°. Recrystallization from benzene, with charcoaling, gave a chromatographically pure product......"
3. 3,4,5 TMB here (from Rhodium):
"Into a 1-l. round-bottomed flask equipped with a magnetic stirrer and a reflux condenser were placed: 50 g (0.298 mol) 5-hydroxyvanillin, 500 ml acetone, 91.0 g (0.716 mol) dimethylsulfate, 100 g (0.806 mol) finely ground Na2CO3.H2O, 10 ml of 10% KOH in methanol. Of course equimolar amounts of anhydrous Na2CO3 or K2CO3 can be used.
The heterogenous mixture was stirred under vigorous reflux for 24 hours, after which the reflux condenser was replaced with a descending condenser. The solvent was distilled at a bath temp. of ca. 100° until the distillation ceased. To the solid residue was added 400 ml of water and the heterogenous mixture was stirred VIGOROUSLY for 2 hours at room temperature followed by one hour at 0-5° (ice-cooling). The light brown crystals were filtered by suction, washed with 3*150 ml of ice water and air dried to constant weight. Yield 55.1g (94%), m.p. 72.5-74°. After distillation at 0.5mm/130°, there was obtained a 90% yield of 3,4,5-trimethoxybenzaldehyde, m.p. 73.5-75, VPC purity 99.3%."
4. 3,4,5 TMbNS from here:

https://www.thevespiary.org/rhodium/Rhodium/chemistry/mescaline.cathyd.html

5. Mescaline from here:

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

Your only real hitch should be the dimethylsulfate..but it CAN be bought and worked with safely.

[methymouse]

Things like "refluxed(105°) for 4.5h with continuous stirring under N2" and "extracted continuously with ether for 16h" are just the sort of thing SWIM wants to avoid.  Admittedly, SWIM will have to keep the reaction with some metal methoxide very dry, but it doesn't require a soxlet extractor or ether.  And what on earth is wrong with Uemura's bromination method?  It doesn't involve non-OTC NaI or chloramine, and gives perfectly acceptable yields(~90%).  For that matter, his preparation of hydroxy-vanillin gives a higher yield without a continuous solvent extraction.  The Zn/HCl is a good idea!     I hadn't noticed that this would work with nitrostyrenes before. 

As for DMS, I'll have to take your word for it that it can be bought.  It can also be made relatively easily if you have either fuming sulfuric acid or a good vaccuum source.  I imagine that if SWIM could get DMS, he could probably get TMP which is orders of magnitude safer and gives similar yields.

Edit:  I hadn't noticed that the reaction was with chloramine T, not chloramine.  In any case, sodium iodide is still much harder to get and more expensive than sodium bromide.

[chilly_willy]

Try replacing: rufluxing with N2 with "refluxing with helium from party supply store."  And instead of continuous extraction with ether...try "simple solvent extraction with hot boiling toluene" as is suggested at rhodiums site.  I guess when it comes to the sodium bromide vs. sodium iodide it would be a toss up.  I hadnt noticed any particular difficulty in obtaining it or an outrageously high price.  The Chloramine-T route appealed because the reaction can be performed at low temperature followed by an easy workup.  Also, I wouldnt try synthing the DMS.  The sulfuric+methanol method has been written about..but I had trouble finding a post where someone actually completed it successfully.

[methymouse]

Antoncho and Vitus seem to have already solved all MeI problems one could have in

Post 301629

(Antoncho: "Two tried-and-true ways to make MeI !", Chemistry Discourse)     , so I think SWIM won't be doing anything with DMS.

As for NaI vs. NaBr, SWIM can get NaBr at a hardware store in the swimming pool section for something like $5/pound.  He's never seen NaI OTC.  Online, the lowest price he's seen was something like $12/100g.  $/mole, NaBr wins out.

I definitely should have read Rhodium's page on 5-OH-vanillin more carefully.  There are a couple of neat tricks there, as you say--toluene extraction and the bisulfite aduct.

And yes, I know that helium is an inert gas, though running it for several hours is not very cheap--my main concern.

Thanks a lot, everybody!      SWIM can't wait to get back to his lab and start trying some of this stuff out.

[sYnThOmAtIc]

The ultrasonic condensation? Has this been tested by anyone here or truly known to be as proposed?