indoline 4 and tetrahydroquinoline 5 products. With deacti-
vated substrates, the formation of the superelectrophile B,
containing a m-hydrido bridged carbocation, could be postu-
lated. Then, B leads to the more stable (poly)cationic species E
which is the precursor of the anti-Markovnikov product.
In conclusion, we showed that an anti-Markovnikov addition
on N-allylic derivatives occurred in superacid. Moreover, during
intramolecular processes, such a reaction became completely
regioselective in the presence of strongly deactivated nucleophiles.
The selectivity of the reaction probably stemmed from the ability
to form m-hydrido bridged carbocation containing superelectro-
philic intermediates. Such an unprecedented anti-Markovnikov
addition to a non-conjugated double bond, under acidic condi-
tions, expands the possibilities for application of HF/SbF5 super-
acid chemistry to a range of strategies in organic synthesis.
Financial support from the CNRS, the University of Poitiers,
the Region Poitou-Charentes is gratefully acknowledged. We also
would like to thank Bernard Langlois for helpful discussions.
Fig. 3 Results from low-temperature 13C NMR studies of anilines in
HF/SbF5 (compared to theoretical values in ppm).15
Fig. 4 Reaction of labelled substrates in HF/SbF5.
an arenium–ammonium dication from substrate 6, though, under
these conditions, the concentration of the diprotonated species
should be very low because of a fast exchange with the mono-
protonated one, which makes it difficult to detect by the slow
NMR technique. In contrast, the reaction of the deactivated
substrate 8 in superacid was not accompanied by protonation of
the aromatic ring. Only the diprotonated species 10 was detected.
It should be noted that the observed chemical shifts fitted well with
the theoretical NMR chemical shifts calculated for such postulated
intermediates.15 These observations confirm the formation of a
dicationic intermediate, either of type D from activated substrates
or of type B0 from deactivated substrates (Fig. 1).
Notes and references
1 M. Beller, J. Seayad, A. Tillack and H. Jiao, Angew. Chem., Int. Ed.,
2004, 43, 3368 and references cited therein.
2 V. V. Markovnikov, Ann. Chem. Pharm., 1870, 153, 228.
3 For a recent example of palladium-catalyzed anti-Markovnikov
addition of allylic amine, see: R. J. DeLuca and M. S. Sigman,
J. Am. Chem. Soc., 2011, 133, 11454.
4 To the best of our knowledge the only reported examples of anti-
Markovnikov electrophilic addition on nitrogen containing deri-
vatives are conjugate additions, see for example conjugate additions
in superacid: Y. Zhang, J. Briski, Y. Zhang, R. Rendy and D. A.
Klumpp, Org. Lett., 2005, 7, 2505.
To gain further insight into the mechanism, experiments
with labelled substrates were also performed (Fig. 4). After
reaction with the superacid, substrate 10b produced a mixture of
products 4b and 5b incorporating 0% deuterium, confirming a
D/H exchange that proceeded through protonation/elimination
of the aromatic ring. In contrast, the exclusive formation of the
trideuterated tetrahydroquinoline 50h from the labelled deacti-
vated substrate 10h confirmed that the tetrahydroquinolines
resulting from deactivated systems could not arise via protonation
of the aromatic ring. To precise the mechanism underlying the
reactions of deactivated anilines, we investigated the formation of
B0-type unsymmetrical m-hydrido bridged intermediate. If this
non-classical cation is formed in superacid during the anti-
Markovnikov process, it would be expected to act as a hydride
abstracting agent and to perform, in the presence of perdeuterated
cyclohexane, an ionic hydrogenation process.18 This was con-
firmed by the formation of a g-labelled product 13.19
Taken together, these experimental observations and DFT
modelling suggest that two reaction pathways may operate,
depending on the aromatic ring substitution (Scheme 2). After
initial protonation of the nitrogen atom, (poly)cationic super-
electrophilic species of type B and/or D could be formed, depending
on the aromatic substitution. With activated substrates, the
formation of the dicationic ammonium–arenium dication D
could be favored. The latter acts as the precursor of a mixture of
5 J. C. Jacquesy, in Carbocation Chemistry, ed. G. A. Olah, John Wiley
and Sons, New York, 2004, pp. 359–376 and references cited therein.
6 S. Thibaudeau, A. Martin-Mingot, M.-P. Jouannetaud, O. Karam
and F. Zunino, Chem. Commun., 2007, 3198; F. Liu, A. Martin-
Mingot, M.-P. Jouannetaud, O. Karam and S. Thibaudeau, Org.
Biomol. Chem., 2007, 9, 4789.
7 G. A. Olah and D. A. Klumpp, Superelectrophiles and their
chemistry, John Wiley and Sons, New York, 2008.
8 Charge–charge repulsion in superelectrophiles can have unusual
effects in structure, bonding and charge distributions, please see:
D. A. Klumpp, Chem.-–Eur. J., 2008, 14, 2004.
9 T. S. Sorensen, in Stable Carbocation Chemistry, ed. G. K. S. Prakash
and P. v. R. Schleyer, Wiley, New York, 1997, pp. 75–136; H.-S. Andrei,
N. Solca and O. Dopfer, Angew. Chem., Int. Ed., 2008, 47, 395.
10 In superacid HF/SbF5, increasing the amount of SbF5 increases the
acidity strength of the media. J.-C. Culmann, M. Fauconet, R. Jost and
J. Sommer, New J. Chem., 1999, 23, 863 and references cited therein.
11 These results are in accordance with previously observed results in
superacid: F. Liu, A. Martin-Mingot, M.-P. Jouannetaud,
F. Zunino and S. Thibaudeau, Org. Lett., 2010, 12, 868.
12 C. Hansch, A. Leo and R. W. Taft, Chem. Rev., 1991, 91, 165.
13 Trifluoromethylated products were found to be unstable over silica
that could explain the low obtained yields.
14 G. A. Olah, K. K. Laali, Q. Wang and G. K. S. Prakash, Onium
ions, John Wiley and Sons, New York, 1998, pp. 280–281 for
amides and pp. 101–102 for the ether.
15 Please see ESIz for more details.
16 In this process, Hammond’s postulate allows the comparison of the rela-
tive energies of the intermediates instead of that of the transition struc-
tures. Please see: G. S. Hammond, J. Am. Chem. Soc., 1955, 77, 334.
17 A similar spectrum has been obtained by G. Olah and his group:
G. A. Olah, K. Dunne, D. P. Kelly and Y. K. Mo, J. Am. Chem.
Soc., 1972, 94, 7438.
18 K. Y. Koltunov, G. K. S. Prakash, G. Rasul and G. A. Olah,
J. Org. Chem., 2002, 67, 4330.
19 Product 11 probably comes from cyclohexane reaction (formed
after D/H exchange by s protonation of C6D12): A. Goeppert and
J. Sommer, New J. Chem., 2002, 26, 1335.
Scheme 2 Postulated reaction pathways for cyclic products formation.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 5877–5879 5879