Author Topic: Reduction of cathinone salts to PPA derivatives  (Read 29793 times)

0 Members and 1 Guest are viewing this topic.


  • Guest
Reduction of cathinone salts to PPA derivatives
« on: October 18, 2004, 11:10:00 PM »
Cathinone and its derivatives are potentially very useful as amphetamine, phenethylamine or butanamine precursors, but their inherent instability in the freebase form - especially in the presence of additional base such as hydroxide - leads to the inevitable (and often rapid) dimerisation to nasty pyrazines. It is therefore desirable to keep the aminoketones as salts when working with them.

Cathinones can be made by the infamous Delepine reaction on alpha-bromoketones or by an SN2 reaction between virtually any substituted amine and an alpha-bromoketone. There is also the novel method posted by Rhodium (

Post 492480

(Rhodium: "beta-Ketophenethylamines (2C-CATs)", Novel Discourse)
), as well as the introduction of derivatives of amino acids to substituted benzenes followed by hydrolysis (e.g.

Post 475692

(Rhodium: "A two-step method for chiral cathinones", Novel Discourse)

Although I have only tried the reduction method below on one substrate, it is very likely to work equally well on others (e.g. the reduction of methylone to methylenedioxyephedrine, which can be further reduced to MDMA if desired). I will try this on methylone when I get round to resynthesising it.

I mentioned last year that I had tried the method (

Post 475219

(Kinetic: "Benzene -> methaephetamine", Novel Discourse)
), but that didn't work well, and I instead favoured a procedure using the freebase in methanol. Now that I have got it to work, this method gives a better yield and is fast and efficient. But the best thing about it is the solvent used: water.


Reduction of cathinones to phenylpropanolamines ("c1cc(ccc1)C(C(CC)NC)(=O)>>c1cc(ccc1)C(C(CC)NC)O")


10.69g (50 mmol) 2-Methylamino-1-phenyl-1-butanone HCl (MW 213.70)
1.32g (35mmol) NaBH4
75ml deionised water

A solution of the aminoketone salt1 in 40 ml water was cooled to 0°C by immersion of the flask in an ice-water bath, and a solution of NaBH4 in 10 ml water was added over 10 minutes. Immediately on addition of the borohydride a white precipitate began to form. This became heavier as addition proceeded, and became so thick by the end that a further 25 ml water was required to allow continued stirring. After addition the reaction was stirred for a further 20 minutes with cooling, then the residual borohydride was destroyed by the very cautious addition of glacial acetic acid,2 dropwise with the flask still in the ice bath. Enough acid was added to dissolve all the precipitate giving a clear, colourless solution. This was made strongly basic with approximately 50 ml 10% NaOH solution, causing a voluminous white precipitate to fall out of solution. 100 ml ether was added, the aqueous layer was saturated with NaCl and the ether layer was separated off.3 The aqueous layer was further extracted with 2x50 ml ether, then the combined extracts were washed with 100 ml brine and dried over a little MgSO4. Removal of the solvent gave the title product as bright white, beautiful sparkling needles melting at around 83-85°C.4 When melted, the needles reformed as a granular, waxy solid.

Yield: 8.8g (49 mmol, 98%)


1. This must be recrystallised (alcohol-ether is great for this) to ensure no residual traces of acid remain, which will seriously interfere with the sodium borohydride. The same principle applies to the water - it is essential that deionised water is used. I suspect that not using this in a previous attempt was the cause of my earlier problems.

2. This is the most difficult part of the reaction: if the acid isn't added very slowly and with vigorous stirring, the solution will foam heavily and work its way out of the flask via the nearest available exit.

3. The first ether extract does not dissolve all the precipitate, but the ether can be separated off with the precipitate suspended in it. Combination of the ether extracts gives a colourless, homogenous solution. DCM can be used - and less is required, but there is a risk of severe emulsions.

4. The melting point is the same as that obtained after purification of the freebase by vacuum distillation using the earlier reduction method. The freebase can be recrystallised from ether or ethyl acetate, or made into its hydrochloride derivative and recrystallised from an appropriate solvent combination.


  • Guest
« Reply #1 on: October 19, 2004, 08:17:00 PM »
This post casts some doubt on whether the delepine works on alpha-halo-propiophenones:

Post 529515

(psychokitty: "I hope this helps . . .", Novel Discourse)


  • Guest
Delepine reaction
« Reply #2 on: October 19, 2004, 09:44:00 PM »
Sorry starlight, I should have said some cathinones can be made by the Delepine reaction; i.e. those which would give phenethylamines upon removal of the carbonyl function. With a longer side-chain, I agree there is some doubt over the validity of the reaction, but I am yet to be convinced the reaction will not work at all. That said, there are several articles in the literature for making alpha-aminoacetophenones from alpha-haloacetophenones and hexamine, and I am yet to find a reference for the same reaction with alpha-halopropiophenones.

Below is one of the more recent methods for making alpha-aminoacetophenone salts from the corresponding haloketone; the yield is a nice 81%. This means it should be possible to go from e.g. 1,4-dimethoxybenzene to 2-amino-1-(2,5-dimethoxyphenyl)ethanol (beta-hydroxy 2C-H) in an overall 71% yield, via Friedel-Crafts acylation in 90% yield; alpha-bromination in quantitative yield; Delepine reaction in 81% yield; and reduction of the carbonyl to the alcohol in 98% yield. Hopefully, that is. ;)

Taken from J. Med. Chem., 25, 1045-1050 (1982):

Alpha-amino-p-phenylacetophenone Oxime.

A solution of alpha-bromo-p-phenylacetophenone (25.0 g, 90.9 mmol) and hexamethylenetetramine (13.2 g, 94.0 mmol) in CHCl3 (200 mL) was heated with stirring at 48 °C for 4 h. The resulting mixture was cooled to room temperature, and the quaternary salt was collected by filtration: yield 36.2 g. This salt was stirred in a mixture of EtOH (135 mL) and concentrated HCl (67 mL) for 19 h. The hydrochloride of the product was collected by filtration and washed with small portions of H2O: yield 18.2 g. [73.5 mmol, 81%]


  • Guest
methylone to mdma
« Reply #3 on: October 29, 2004, 04:43:00 AM »
Rhodium has in the past stated that the red P / iodine reduction of pseudoepherine-MDMA would tear the molecule apart, and Nicodem (I think that's who it was) has indicated that applying the Clemmonson Hg/Zn reaction (which takes a ketone and gives an alkane back) does not work for aminoketones.  Therefore, by elimination, if one were so inclined as to synth mdma from methylone, the Wolf-Kishner hydrazine/KOH reaction (which, like the Clemmonson reaction, gives alkanes from ketones) would therefore be the most logical reaction scheme, that I know of anyway.  Also, that route has the advantage of being only one step, whereas the synthesis of pseudoephedrine-MDMA requires that the methylone first be reduced with sodium borohydride or some other reducing agent.