I post this back in the Hive days. More here http://www.zoklet.net/bbs/archive/index.php/t-32584.html

"Caffeic acid (3,4-dihydroxycinnamic acid) from green coffee (unroasted) contains about 6% total Caffeic acid and Chlorogenic acid (hydrolysis of chlorogenic acid gives caffeic acid). It is also found in roasted coffee. Isolation from green coffee beans, J. Agr. Food. Chem Vol 8, pg 58 (1960) & roasted coffee, Arch. Pharm. Vol 293, pg 721 (1960). One kilo of ground coffee beans yields approx 50-55 grams of caffeic acid. Both caffeic acid and caffeine are freely soluble in boiling water..."

- WizardX


That just screams MDA from coffee beans does it not? If xxxxxxxx becomes scheduled the method in my last post should still be assessable though this I'd think. Anyways, after extraction I'd assume one could methylenate the caffeic acid like a catechol and subsequently hydrogenate the resulting product for 3,4-MD-phenylpropionic acid. I adapted a couple existing procedures (refs available) to hopefully result in the desired carboxylic acid...


Experiemental:

A solution of 18 g of 3,4-dihydroxycinnamic acid, 12 ml of 50% aqueous sodium hydroxide and 20 ml of DMSO is heated to 98 C and stirred at that temperature for 30 minutes. This solution at 98 C. is then added over a 30 minute period to a refluxing solution of 12 ml of DCM in 30 ml of DMSO. Thereafter, the reaction mixture will be stirred at reflux for 1 hour. Steam is passed into the mixture to achieve steam distillation of the product. The distillate (100 ml) is extracted with 10 ml of DCM, and the extract washed once with 10 ml of water. The DCM solution should then be concentrated in vacuo at 40 C to yield product.


A soultion of 19 g of 3,4-methyenedioxycinnamic acid and 550 ml of 10% sodium hydroxide solution. The mixture is warmed to 90 C. 55 g of nickel-aluminum alloy powder is added in small portions. Addition should be complete within an hour and the mixture should be maintained at 90–95 C for 1 hour further. The mixture is then filtered, and the metallic residue is washed with 50 ml of hot 10% sodium hydroxide solution and two 50 ml portions of hot water in such a manner that the solid is always covered with liquid. The cooled filtrate and washings are added dropwise to 300 ml of concentrated hydrochloric acid at such a rate that the temperature does not exceed 80–85 C. Separation should begin as contents are cooled to room temp.


Notes:

1. I'm not sure if the COOH might interfere in the methylenation. If so, one might prepare 3,4-dihydroxyamphetamine and then methyenate as a last step. Maybee?

2. Cinnamic acid derivitives are notoriously hard to reduce. Our choice of catalyst seems limited to Na/Hg, Raney nickle or platinum dioxide. Possibly possibly possibly Pd/C.

---

Preparation of 3,4-dihydroxystyrene. http://www.freepatentsonline.com/5324804.html

200 g (1.11 mol) of caffeic acid are dissolved in 1.2 l of dimethyl formamide dissolved and the solution is stirred for 3 h at 150° C. After cooling to room temperature, the solution is poured on to 3 kg of ice, saturated with sodium chloride and extracted twice with ethyl acetate. The organic phase is washed with 2% NaHCO3 solution and then with water, dried over Na2SO4 and concentrated by evaporation, giving 120 g (80% ) of a highly viscous liquid which can be recrystallised from toluene. The substance is, however, sufficiently pure and can be further processed without recrystallisation.

Elemental analysis: calcd=C: 70.52, H: 5.97; found=C: 70.25, H: 6.02.

1 H-NMR (CDCl 3 /DMSO): ?=4.99; 5.03; 5.44; 5.51 ppm (dd, 2H); ?=6.47; 6.50; 6.53; 6.56 ppm (m, 1H); ?=6.66-6.74; 6.74; 6.89 (m, 3H); ?=8.53 ppm (s, 1H); ?=8.63 ppm (s, 1H).

WizX, it is now known that there are at least two options when performing the methylenation of even 4-allylcatechol with DCM, the use of CuO with DMSO & DCM is said to provide better yields than are possible with vanillin (less prone to oxidation).

The 3,4-methylenedioxy-a-methylhydrocinnamic acid would be an absolute shoe in for a simple cheap route to MDA (via the Hoffman on the amide). Nothing hard to get, well HBr is not easily sourced here, but nothing overly complex. Transfer hydrogenation of that double bond using Aluminium-based/Al-Ni based catalysts should be easy enough.

Caffeic acid, 3,4-Dihydroxy-cinnamic acid,

I'm not sure if the COOH on the caffeic acid will/might interfere with CuO, DMSO & DCM in the methylenation? The CuO will react with the COOH. If so, one might prepare 3,4-dihydroxyamphetamine and then methyenate as a last step. Maybe?

I concur, however, how do you propose to synthesize the 3,4-methylenedioxy-a-methylhydrocinnamic acid from Caffeic acid, 3,4-Dihydroxy-cinnamic acid?

Caffeic acid, 3,4-Dihydroxy-cinnamic acid,

3,4-(Methylenedioxy)cinnamic acid, http://www.chemblink.com/products/2373-80-0.htm 

---

Can anyone spot the error at http://www.erowid.org/archive/rhodium/chemistry/mda.dalcason.html

From the methyleneated derivative of caffeic acid, my first guess is the catalytic cinnamic Hunsdiecker:

http://pubs.acs.org/doi/abs/10.1021/jo025868h

From the methylenedioxy-halostyrene, a number of routes are available. A Grignard reaction utilizing a methyl halide and a copper (I) salt produces isosafrole.

Basic hydrolysis produces methylenedioxyphenylacetaldehyde; an olefination gives safrole.

Methyleneations of this aldehyde exist, including the famous Wittig and less famous Corey-Chaykovsky reactions based on phosphonium and sulfonium/sulfoxonium ylides respectively, and a reaction using bis(iodozincio)methane described here -- titanium tetrachloride and zinc may sound at first like an attractive reagent mixture but TiCl4 or "tickle" as it is known is volatile highly toxic and hydrolytically unstable to the point of being pyrophoric and I'd frankly rather work with lead chloride, which is ironically a safer catalyst for the zinc olefination.

The "classic" Takai olefination uses iodoform and CrCl2, producing safrole-plus-an-iodine-atom which has no name because "iodosafrole" means something entirely different. The resulting safrole-plus-an-iodine-atom can be reduced quite easily to safrole. This is the most attractive option, considering the ease of generating iodoform (I2 / acetone / NaOH) and the safety of CrCl2 -- far superior to phosphines and titanium halides.

The problem with most methylations is that you have to use methylating reagents to accomplish them -- this is always harder than it should be. The Takai reactions -- chromium (II) couplings are interesting but perhaps hampered by the fact that iodinated reagents are absolutely required -- a rxn that with iodoform gives 87% yields gives only 32% yields with bromoform and nothing with chloroform.

Not what I am talking about - check the attached files - the acid-catalyzed crossed aldol condensation between vanillin and MEK gives the expected a-methylphenylbutenone, which on oxidation with hypochlorite gives (as a methyl ketone) the a-methylcinnamic acid and chloroform. Reduction of this, followed by methylenation of the phenolic diol, would give a-methyl-3,4-methylenedioxyphenylpropionic acid. Form the amide from that (I think the ethyl ester would probably be a better bet - it is going to be a liquid at a much lower temperature, then a Hoffman reaction (hypochlorite again) directly to MDA.

The use of CuO and Methylene Chloride to achieve decent yields in the methylenation, the fact that if you have HI you can reduce the double-bond & remove the methyl ether in one go, plus the ease of access for each step's chemicals, make this a very, very OTC route. Personally I'd try the methylenation straight after forming the ester (post-reduction). The ester will be in the same phase as the DCM, thus removing the need for PTC, while it should not fuck with the CuO. Nice and back to front, there is no need to fuck around with peracids if adding chlorine and NaOH to the reaction mixture will do the job (albeit twice). The other stage in which methylenation 'could' be tried, is straight after forming the hydrocinnamamide - I strongly doubt the amide will interfere with either the methylenation or the CuO. I choose this route due to the simplicity of the various reactions, the ease with which the various components can be sourced (MEK is not precisely hard to find), Copper powder is available for fuck all, DCM is piss easy to find, so is chlorine, NaOH & dry hydrogen chloride gas.

It isn't as "nice" as safrole, no it ain't. Then again, there is no fucking around with "wackers", Pd-salts, Hg-salts and reductive amination.

That Catalytic Hunsdiecker Reaction in DCM using Triethylamine as the catalyst (with N-bromosuccinimide) looks bloody tasty - it would give direct access in decent yields to 3,4-methylenedioxy-2-bromopropane which would shit all over a route to Safrole IMHO. Getting hold of the Tritheylamine and the Succinimide aren't precisely the easiest tasks (they'll either have to be bought or synthesized), but with the MD-2-bromopropane to hand (rapidly) then either ammonia/methylamine to give MDA/MDMA respectively. Nice find.