9372
J. Zhao et al. / Tetrahedron Letters 44 (2003) 9371–9373
Table 1. Regioselective bromination of arylamines via
boron amidesa
Scheme 2.
was the highest yield for this conversion to date
(Scheme 1). No 2-bromoaniline or dibromoaniline was
isolated. Lithium amide, boron amide and N-bro-
moaniline (1c) are the presumable intermediates pro-
duced at the different stages. Employing trimethyl
borate as the lithium arylamide trapping agent excluded
the need for using trimethyltin chloride and potassium
fluoride.9 Interestingly, we found that omitting the
addition of B(OMe)3 led to a complex mixture, from
which 1b was isolated in poor yield (41%).
A tentative free radical pathway for the formation of 1b
from the unstable intermediate 1c is proposed in
Scheme 2. Homolytic cleavage of 1c furnishes anilinyl
radical 1d and a bromine radical. The bromine radical
attacks another molecule of 1c to afford iminocyclo-
hexadiene 1e with concomitant regeneration of a
bromine radical. Finally, upon expedient rearrange-
ment, 1e was converted to the desired product, i.e.
4-bromoaniline (1b). Obviously, careful mechanistic
investigations will be needed in order to acquire further
insights into the nature of the current arylamine brom-
ination reaction.
except for the observed much lower yields for 7b (27%
and 24%, respectively). Our preliminary experimental
results suggest that aromatic bromination is probably
much faster than other alternative reaction pathways.
Further investigation along this line is underway in our
laboratory.
In conclusion, we have successfully developed a new
efficient regioselective mono bromination methodology
for arylamines. Our one-pot, three-stage approach
(lithiation, boron amide formation and bromination)
proved to be highly useful in mildly brominating a
variety of arylamines in up to 94% yields.
To examine the generality of our regioselective bromi-
nation technology, another eight substrates were scruti-
nized (Table 1). The following observations have been
made. (i) N-Methylaniline 2a reacted as cleanly as the
parent 1a and gave 2b in comparable yield (94%).
Bromination of aniline analogs with an electron-with-
drawing group at nitrogen (entries 3 and 4) proceeded
more sluggishly. For instance, by the one-pot treat-
ment, 4-bromoacetanilide (3b) was isolated in only 61%
yield along with a small quantity of 3a recovered. (ii)
ortho-Substituted anilines seemed to give higher yields
of the desired bromination products than para-substi-
tuted counterparts, presumably due to steric effects (see
entries 5–7).9 (iii) Our technology can be applied to
brominating arylamines other than anilines (entries 8
and 9). For 1-naphthylamine (8a), no appreciable
improvement was made as compared with that reported
previously,9 which implicates that the intrinsic proper-
ties of the substrates play a significant role here.
Acknowledgements
We thank the following agencies for financial support:
Chinese Academy of Sciences (‘Hundreds of Talent’
Program); Science
& Technology Commission of
Shanghai Municipality (‘Venus’ Program); National
Natural Science Foundation of China. We are indebted
to Professor Chaozhong Li of SIOC for enlightening
discussions. Special thanks go to all reviewers and
editors for their pertinent suggestions.
References
It is noteworthy that the bromoaniline derivative 7b
was found to be the only product (85%) when 7a12 was
subjected to our standard bromination conditions. No
cyclization, dimerization or proton abstraction prod-
ucts could be isolated. Even in the presence of TFA or
CuBr/CuBr2 (introduced at −78°C 15 min after the
addition of bromine), similar results were obtained
1. Onaka, M.; Izumi, Y. Chem. Lett. 1984, 2007–2008.
2. Majetich, G.; Hicks, R.; Reister, S. J. Org. Chem. 1997,
62, 4321–4326.
3. (a) Berthelot, J.; Guette, C.; Essayegh, M.; Desbene, P.
L.; Basselier, J. J. Synth. Commun. 1986, 16, 1641–1645;
(b) Berthelot, J.; Guette, C.; Desbene, P.-L.; Basselier,
13