This file is a part of the Rhodium site archive. This Aug 2004 static snapshot is hosted by Erowid
as of May 2005 and is not being updated. > > Back to Rhodium Archive Index > >
[] [] [Chemistry Archive]

Reduction of
Phenylacetonitriles to Phenethylamines

HTML by Rhodium

Cobalt Boride (Co2B)

Optimum Reduction of Nitriles Using NaBH4 and CoCl2 in THF/H2O (2:1)1

The following is a representative procedure. A pink solution of CoCl2·6H2O (0.445 g, 1.87 mmol) and benzonitrile (1.92g, 18.6 mmol) in THF (66 ml, distilled) and water (33 ml) was stirred vigorously and cooled intermittently with an ice-water bath while NaBH4 (1.40 g, 37.1 mmol) was added in portions over 8 min. The reaction was exothermic, producing a black precipitate and copious quantities of hydrogen. TLC analysis after 50 min indicated traces of starting material, so additional NaBH4 (0.41 g) was added. After a total time of 2h 28% NH4OH solution (2 ml) was added and the mixture transferred to centrifuge tubes. After centrifugation the supernatant (two liquid phases) was decanted and the sediment washed with more of the same solvent mixture. The combined supernatants were concentrated at reduced pressure to remove the bulk of THF, then the aqueous residue was extracted with 4x40 ml of DCM. The combined DCM layers were dried (MgSO4) and concentrated to afford benzylamine (1.82g, 91%) as a pale oil, identical in every respect with an authentic sample.

Karel: NaBH4/CoCl2 does reduce phenylacetonitriles, but in case of an ortho-substituted substrate (4-bromo-2,5-dimethoxyphenylacetonitrile) my yields were only about 20% (the rest being unreacted nitrile). Using 10 mmol (2.56 g) of the nitrile allowed for the recycling of 2.11g after separation of amine product. Instead using AlH3 in THF2 for the reduction of 4-bromo-2,5-dimethoxyphenylacetonitrile gave a 90% yield of 2C-B.

Sodium - n-Butanol

Reduction of Nitriles and Oximes with Na/n-BuOH3

The usefulness of n-butyl alcohol in the sodium reduction of a number of aliphatic cyanides and oximes has been investigated. In a considerable number of preliminary experiments upon the reduction of n-butyl cyanide it was found that: (1) better results were obtained by adding the sodium to the boiling solution of the cyanide than by adding the cyanide solution to the sodium; (2) if the temperature of the reaction mixture were kept below the boiling point mechanical stirring increased the yield of amine, but it was of no advantage to stir the refluxing solution; (3) the use of more than 7 atoms of sodium per mole of cyanide did not appreciably increase the yield of amine; and (4) all of the sodium dissolves within a convenient time if slightly more than 3 moles of n-butyl alcohol is used per atom of sodium. It is probable that (3) would not be true for all reductions of this type. The yields of amines from a series of methyl alkyl ketoximes varied from 69 to 86% of the theoretical amounts. Heptaldoxime gave 69% of the amine. n-Butyl and n-amyl cyanide gave 86 and 78%, respectively, of the corresponding amines.


A solution of 1 mole of cyanide or oxime in 2300 mL of n-butyl alcohol in a 5-liter two-necked flask fitted with a reflux condenser was heated to boiling and 161 g (7 moles) of sodium added in 10-20 g pieces at short intervals through the large neck of the flask. This required from ten to fifteen minutes. After the reaction had moderated the mixture was refluxed until practically all of the sodium had dissolved. The reaction mixture was allowed to cool, 1.5 liters of water added and the mixture distilled to remove the amine and alcohol. More water was added as necessary. The distillate was made slightly acid with hydrochloric acid and distilled to a volume of 300 ml, 500 ml of water added and distillation continued until no more alcohol came over. The water solution of amine hydrochloride was saturated with sodium hydroxide, the upper oily layer of amine separated and further dried over sodium hydroxide and then over sodium. The amine was now distilled from sodium in a 250ml special Claisen flask. If the amine is properly dried there is very little forerun and no high boiling residue.


  1. J. Am. Chem. Soc. 108, 67-72 (1986)
  2. J. Am. Chem. Soc. 90, 2927-2938 (1968)
  3. J. Am. Chem. Soc. 56, 487 (1934)