By a primary amine I mean a compound where the hydroxyl group of the corresponding alcohol is replaced with an amine group.


4 Answers 4


In the conversion of primary alcohols to primary amines

$$\ce{R-CH2OH -> R-CH2NH2}$$

direct alkylation of ammonia normally is the last thing you want to do in the lab.

Under conditions where $\ce{OH}$ is a good leaving group, i.e. in acidic medium, the nucleophilicity of ammonia is reduced due to protonation. Moreover, there's little chance to prevent further alkylation of the initial product.

More promising strategies are:

  • Gabriel synthesis: Hydroxy-halogen-exchange, followed by reaction of the primary halide with phthalimide and subsequent hydrazinolysis

  • Mitsunobu reaction ($\ce{Ph3P, DEAD}$) of the primary alcohol using hydrazoic acid $\ce{HN3}$ as a nucleophile, followed by reduction of the alkyl azide

  • An (almost-one-pot) combination of the Appel and the Staudinger reaction

    1. Heat the primary alcohol ($\ce{R-CH2OH}$) with $\ce{Ph3P}$ and bromotrichloromethane $\ce{CBrCl3}$ in benzene, then cool down to RT (room temperature) to yield $\ce{R-CH2Hal}$

    2. Add $\ce{NaN3}$,tetrabutylammonium bromide and N,N-dimethylformamide and reflux again to obtain the azide $\ce{R-CH2N3}$, then cool down to RT, pour on water, extract with benzene and dry the organic layer

    3. Add triethylphosphite $\ce{(EtO)3P}$ to the organic layer, stir at RT to yield the iminophosphorane $\ce{R-CH2N=P(OEt)3}$

    4. Hydrolyze the iminophosphorane by refluxing with hydrochloric acid, then neutralize to release the amine

  • $\begingroup$ Mitsunobu was my first though. Appel is nice, too. Interestingly, I'd never heard of the Gabriel Synthesis; nice. $\endgroup$ Apr 22, 2016 at 19:26
  • $\begingroup$ In place of using hydrazoic acid, DPPA might be a slightly less dangerous alternative (albeit more expensive and less atom economic). $\endgroup$ Nov 6, 2017 at 0:01

For aliphatic compounds you can use alkylation

$$\ce{R-OH + NH3 -> R-NH2 + H2O}$$

For aromatic compounds like phenol

$$ \begin{matrix} \ce{C6H6O &\quad +& Zn& \ ->& C_6 H_6\\ Phenol& & (dust\ form)& \ & Benzene&} \end{matrix} $$

Zinc dust reduces the OH group in phenol

$$\begin{matrix} \ce{C6H6& + &HNO3& ->[H2SO4]& C6H5NO2 + &H_2 O&\\ Benzene& & (high\ concentration)&& Nitrobenzene& water& } \end{matrix}$$ Nitration

$$\begin{matrix} \ce{C6H5NO2& +& Sn& +& \qquad HCl_{(aq)} ->& C6H5NH2&\\ Nitrobenzene& & Tin& &Hydrochloric & Aniline&\\ &&&&acid} \end{matrix}$$ $ $
Tin and hydrochloric acid act as reducing agents

  • 12
    $\begingroup$ I think you are oversimplifying and skipping a few steps and issues by just saying "alkylation". For a good answer I'd like to see the details and complications. Sure it is possible, but it isn't easy and that's what usually matters in chemistry. $\endgroup$
    – matt_black
    Apr 2, 2014 at 7:30
  • 5
    $\begingroup$ (-1) That "alkylation" doesn't work. $\endgroup$ Apr 22, 2016 at 18:11

In addition to the method provided by t3st you could also try the following for $$\begin{matrix} \ce{Ph-OH& +& Zn& ->& C6H6 \\ Phenol& & Zinc& & Benzene&} \end{matrix}$$ (Just like the first step of t3st 's answer.

Then you could do Electrophilic halogenation of benzene using $\small\ce{AlCl3\ or\ FeBr3}$ as catalyst. $$\begin{matrix} \ce{C6H6& ->[\ce{Cl2/AlCl3}]& Ph-Cl\\ Benzene& & Aryl\ halide&} \end{matrix}$$

Finally you could react aryl-halide with soda-amide($\small\ce{NaNH2}$) to substitute $\ce{Cl}$ with $\ce{NH2}$[citation needed] $$\begin{matrix} \ce{Ph-Cl& ->[\ce{NaNH2}]& Ph-NH2&\\ Aryl halide& & Aniline&} \end{matrix}$$

  • 1
    $\begingroup$ Low yields all the way with this route $\endgroup$
    – Waylander
    Feb 7, 2018 at 10:31

For PhOH to PhNH2 Pd catalysis offers a better route than Zn reduction etc. Formation of the triflate by reaction with triflic anhydride then reaction under Pd catalysis with a source of N (I've used benzophenone imine or acetamide but there are others) then finally aqueous hydrolysis gets to the aniline in fairly respectable yield.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.