Author Topic: Synthesis of 4-Fluoroamphetamine  (Read 9904 times)

0 Members and 1 Guest are viewing this topic.


  • Guest
kick ass synthesis of fluoro-amphetamine
« Reply #40 on: January 29, 2003, 03:16:00 PM »
Synthesis of N-(beta-haloalkyl)-N-alpha-dimethyl-arylethylamines and their conversion into aziridium salts, E. Zara-Kaczian, Gy Deak, J. Hasko-Breuer and A. Neszmelyi., Acta Chimica Academiae Scientiarum Hungaricae, Tomus 79 (4), pp. 433-447 (1973)

Chimimanie's voice:
This article is one of those from foreign countries that is a true pearl for us Clan'Chimist; I like those unknown journals whose contributors chose to do so many great reactions and combine them in their article, really the less subside the chemists have the greatest their work!

This article describe two quick ways to this compound, one of those is based on the 'Oh so speaked but not so tried' reaction of chloroacetone on a grignard reagent (made from the cheap p-bromo-fluoro-benzene) and the second one start from the cheap fluorobenzene and react it with allyl chloride to make the fluoro-phenyl-chloropropane derivative. I think these two methods (especially the second) are kicking ass, they make p-fluoro-amphetamine in 3 steps from unwatched and cheap precursors in not too bad yield.

Anyway 'Felicitation Mr Zara-Kaczian!'

This is an excerpt of the text:

For the grignard reaction with chloroacetone:

Reproduction of the procedure in the litterature gave the yield specified (40%), and its improvement was attempted by varying the reaction conditions.(...) It was found, however, that under certain conditions - when the reaction was carried out at 0°C in 5h - a product other than the expected one formed. It proved to be an alcohol, probably 1-chloro-2-(p-fluorophenyl)-2-propanol, that is, the addition process became the predominating reaction.


By Friedel-Crafts reaction, about the chloropropane derivative:

Another reaction path was also considered for the preparation of metamphetamine derivatives halogenated in the nucleus. The Friedel-Crafts reaction of fluoro- or chlorobenzene (Chimimanie's voice: not chlorobenzene for our purpose!) in the presence of iron (III) chloride gives a 1-aryl-2-chloropropane[6] carrying a halogen atom in the benzene ring. This product yield with methylamine the corresponding halogenated metamphetamine; when methylaminoethanol is used, the N-beta-hydroxyethyl derivative is obtained; the latter reaction has not been achived previously. (Chimimanie's voice: they give a ref with methylamine in ethanol solution in a bomb, same ref I give below, the hydroyethyl is made with a bad yield (8.7%) with methylaminoethanol as solvent, this is better to swap with azide and reduct)


When fluorobenzene was used in the reaction, the product obtained in 48% yield proved to be a mixture of ortho and para isomers (72% para, 28% ortho).



A mixture of p-fluorobenzyl methyl ketone (41.1 g; 0.26 mol), N-methylformamide (95.5g; 1.62 mole) and 100% formic acid (12.5 g; 0.27 mole) was refluxed in a rbf, equipped with a reflux condenser, for 12 hrs. After cooling, the solution was diluted with water (140 ml) and extracted with ether (3x140 ml). The ether was evaporated from the combined extract and the residue refluxed with conc. HCl (200 ml) for 12 hrs. After cooling, the mixture was shaken with ether (3x140 ml), the aqueous phase was made alkaline (pH 8) with 30% NaOH (about 120 ml) while cooling, then saturated with Na2CO3. An oil separated which was extracted with ether (5x140 ml) and dried over anhydrous Na2SO4. The ether was evaporated and the residue was distilled in vacuum to give a colourless liquid (26.15g; 58%) bp 86°C/4.5 torr; n20=1.4887; lit bp 87-89°C/10 torr n20=1.4922

1-o-Fluoro and 1-p-fluoro-2-chloropropane

Anhydrous iron(III) chloride (16.2g; 0.1 mole) was added to fluorobenzene (192.2g; 2.0 mole), the mixture was cooled to -21°C and allyl chloride (38.25g; 0.5 mole) was added to it dropwise during about 2hrs, with continuous cooling and stirring. Stirring was aplied for 3 hours more at -16°C, then the reaction mixture was poured into a mixture of ice (500 g) and conc. HCl (50 ml); the organic phase was washed with 1:1 HCl (3x70ml), with saturated NaHCO3 solution (2x70ml) and finally with saturated NaCl solution (2x100ml) until free from acid, then dried over anydrous Na2SO4. After evaporation of the fluorobenzene, the residue was fractionated in vacuum; the product collected at 14 torr pressure and 90-94°C temperature (41.7g; 48.4%) was a mixture of the o- and p-isomers.
The isomeric mixture was separated by means of a preparative gas chromatograph of type Carlo-Erba Fractovap Mod P , on a column packed with 6% Apieson L/Chromosorb 6 (6m in lenght, 25mm diam); H2 carrier gas and 150°C column temperature applied.  :-[

A blunt of ortho and para isomers is sufficiant for me... 8)

The chlorine atom can be swapped with sodium azide and then reducted to the amine by various methods, UTFSE.

p-fluoro-phenylacetone From [5]

A grignard reagent was prepared from 35g (0.2 mole) of p-fluorobromobenzene and 4.6g (0.19 mole) of magnesium. To this was added 18.5 g (0.2 mole) of chl oroacetone in 50ml of ether as rapidly as the refluxing of the reaction mixture would allow. The ether was removed by heating the reaction flask in an oil-bath and at 100°C the residue foamed with formation of a gel from which ether was removed slowly by heating at 135-140°C for forty-five minutes. The flask was cooled, ice and dilute acid were added, the heavy oil which separated was removed with ether and the ether solution dried and fractionated. There was obtained 11.2g (37%) of practically pure ketone distilling at 106-107°C (18mm)

A variation of this procedure in which the reaction mixture was not heated above 100°C gave a product that could not be satisfactorily fractionated. Substantially pure ketone was finally obtained by conversion to the sodium bisulfite addition compound followed by regeneration with dilute sodium carbonate. It distilled at 108°C (18 mm) as a light yellow oil with n20d 1.4965, d20 4 1.107, Md calcd 40.07, obsd 40.17. After a few days of standing, crystals of p-fluorobenzoic acid were deposited. Carbon and hydrogen analyses gave low results even with material kept in a sealed ampoule. A solid derivative was therefore prepared. The dinitrophenylhydrazone did not form but the semicarbazone mp 200.5-201.5° separated readily. Analysis by the Jamieson method of titrating with potassium iodate gave only fair results.

alpha-Methyl-p-fluorophenetylamine ibid

A mixture of 8.7g (0.057 mole) of p-fluorobenzyl methyl ketone and 16g (0.36 mole) of formamide was heated in a 100ml rbf attached to an air condenser for twelve hours at a temperature where vigorous bubbling occured. The amide was hydrolyzed by refluxing for eleven hours, and the amine layer separated. Distillation gave 3.6g (41% yield) of pure amine, bp 95-96°c (17 mm) The amine rapidly took up CO2 from the air forming a solid salt. The hydrochloride salt melt at 156-157°C

[5]: Suter and Weston, JACS 63, 602, 1941
[6]: Patrick, F M, JACS 68, 1009 (1946)


  • Guest
Fluorination Articles
« Reply #41 on: June 29, 2004, 01:50:00 PM »

There is no way to get a regioselective reaction

The authors of these articles seem to disagree with that statement  ;)

JOC 34 1387-91 (1968)

JOC 34 2835-9 (1969)

JOC 35 723-7 (1970)