Enantioselective [1,2]-Stevens rearrangement of quaternary ammonium salts. A mechanistic evaluation

Article abstract of DOI:10.1039/b716488b

Using a supramolecular asymmetric ion pairing strategy, an enantioselective [1,2]-Stevens is feasible on substrates devoid of stereogenic quaternary nitrogen atoms. The Royal Society of Chemistry.

Full text of DOI:10.1039/b716488b

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Enantioselective [1,2]-Stevens rearrangement of quaternary ammonium
salts. A mechanistic evaluationw
ˆ
´
´
´
Maria-Helena Gonc¸ alves-Farbos, Laurent Vial and Jerome Lacour*
Received (in Cambridge, UK) 26th October 2007, Accepted 6th December 2007
First published as an Advance Article on the web 21st December 2007
DOI: 10.1039/b716488b
Using a supramolecular asymmetric ion pairing strategy, an
enantioselective [1,2]-Stevens is feasible on substrates devoid of
stereogenic quaternary nitrogen atoms.
genic center is complete. With this established precedent, it
occurred to us that any other twisted biaryl azepinium entity
of analogous structure and geometry could react similarly and
selectively in [1,2]-Stevens rearrangements—and ‘‘simple’’,
readily available from indole derivatives, diphenylazepinium
cations of type 3 (Table 1) in particular.
The [1,2]-Stevens rearrangement, which is the spontaneous
transformation of ammonium ylides into tertiary amines, has
been strongly studied for its interesting mechanism and for its
synthetic utility.1 Despite the many studies, asymmetric ver-
sions of this reaction still remain a challenge probably reflect-
ing the radical character of the process.2 In fact, strict
enantioselective [1,2]-Stevens rearrangement of quaternary
ammonium ions has not been reported,3 and enantiospecific
transformations occur usually with some loss of enantiomeric
purity.4 These latter reactions that use substrates containing
stereogenic quaternary nitrogen atoms involve intramolecular
transfer of chirality (ToC) from the N-atom to the adjacent
C-position. The enantioselectivity of the ToC depends upon
the ionization conditions as shown recently by West and
Tayama (ee 54 to 99%).5 In view of these precedents, it was
debatable whether an enantioselective [1,2]-Stevens could ever
be developed for compounds devoid of stereogenic quaternary
nitrogen atoms. Herein, on designed substrates and using a
novel supramolecular asymmetric ion pairing strategy,6,7 we
report a ‘‘proof of concept’’ mechanistic study that sees an
enantioselective [1,2]-Stevens rearrangement occurring despite
its diradical mechanism [eqn (1)].
The caveat with compounds 3 was however their configura-
tional lability at ambient and low temperatures. In fact, studies
performed on analogous diphenylazepines or diphenylazepi-
nium salts have revealed that the 7-membered dibenzo-
[c,e]azepinium ring presents an axial chirality with a low
kinetic barrier of enantiomerization (DGz E12–14 kcal
molꢀ1).11,12 Diphenylazepinium cations of type 3 thus exist
as 1 : 1 mixtures of freely interconverting P and M atropi-
somers in solution. As such, these tropos derivatives were not
used in stereoselective [1,2]-Stevens rearrangements.13
Previously, hexacoordinated phosphorus anion BINPHAT
5 (D and L enantiomers, Fig. 1)14 has been shown to be a
general NMR chiral solvating, resolving and asymmetry-in-
ducing reagent for chiral organic cationic species. When
associated with configurationally labile cations, supramole-
cular diastereoselective interactions can occur and one diaster-
eomeric ion pair can become predominant over the other.6 The
occurrence of such behavior, called the Pfeiffer effect,15 is the
Table
1 Enantioselective [1,2]-Stevens rearrangement of the
BINPHAT salts of cations 3a to 3ea
ð1Þ
If enantioselective [1,2]-Stevens rearrangements of quaternary
ammonium ions have been elusive, many useful diastereo-
selective reactions have been developed.8 Previously, Mislow
and Zavada have shown that [1,2]-Stevens rearrangements of
configurationally stable biarylazepinium cations occur readily
upon ylide formation.9,10 The reactions proceed with high
selectivity as, for instance, cations of type 1 [eqn (1)] react
with strong bases to produce dihydrohelicenes of type 2 as
single diastereomers (99% yield).9a The ToC from the biaryl
axis of the diarylazepinium precursor to the new sp3 stereo-
c
Z
Salt
Yield (%) eeb (%) [a]D
ded (%) ToCe (%)
H
H
[3a][D-5] 90
[3a][L-5] 88
33
33
27
20
49
55
(þ)
(ꢀ)
(þ)
(þ)
(þ)
(þ)
33
33
30
20
50
60
100
100
90
100
98
OMe [3b][D-5] 52
OBn [3c][D-5] 50
F
Cl
[3d][D-5] 50
[3e][D-5] 48
92
a
Treatment with 6 (1.5 equiv., CH2Cl2, ꢀ80 1C, 4 h); average of at
b
least two runs. Enantiomeric purity of 4a–4e determined by CSP-
c
HPLC. Sign of the optical rotation of 4a–4e. Diastereoselectivity
d
Department of Organic Chemistry, University of Geneva, Quai E.
Ansermet 30, CH-1211 Geneva 4, Switzerland. E-mail:
jerome.lacour@chiorg.unige.ch
w Electronic supplementary information (ESI) available: Experimental
details. See DOI: 10.1039/b716488b
of the ion pairing determined by 1H and 19F NMR at 193 K, precision
e
ꢁ2–3%. Transfer of Chirality defined as the ratio of the enantio-
selectivity [ee] over the ionic stereoinduction [de], precision ꢁ2–3%.
Chem. Commun., 2008, 829–831 | 829
ꢂc
This journal is The Royal Society of Chemistry 2008

View Article Online
Fig. 1 BINPHAT anion 5 (D and L enantiomers shown).
Fig. 2 Ion pairing of anion 5 and cation 3d. Asymmetric induction as
evidenced by 19F NMR (CD2Cl2, 352 MHz): (a) [3d][Br], 233 K; (b)
[3d][D-5], 233 K, de 50% and (c) [3d][D-5], 193 K, de 52%.
result of the formation of one thermodynamically more stable
ion pair in solution.16 The association of this enantiopure
anion 5 with cations 3 was then interesting for the develop-
ment an asymmetric [1,2]-Stevens rearrangement, any imbal-
ance in the population of the diastereomeric salts, e.g. [P-3][D-
5] over [M-3][D-5], resulting possibly in the preferential for-
mation of one enantiomer of rearranged product 4 over the
other (Table 1).
eomeric ion pair, a series of variable temperature NMR
experiments were performed on the bromide and BINPHAT
salts under essentially the reaction conditions. First, in the 1H
NMR, clear AB patterns appeared at ꢀ40 1C in CD2Cl2 for
the benzylic protons of the bromide salts demonstrating,
without ambiguity, that the exchange between the P and M
conformers of 3 was slow on the NMR-timescale at that (and
lower) temperature. Then, salts [3a][D-5] to [3e][D-5] were
studied at ꢀ40 and ꢀ80 1C and, in all cases, anion D-5 acted
as an NMR chiral solvating agent. NMR signals were clearly
split into two sets, one for each of the atropisomers of 3. In 1H
NMR, the differentiation was better followed in the d 5.8–6.5
ppm region corresponding to some of the aromatic protons.
The singlet signals of the MeO substituent of 3b or of the
fluorine atom of 3d in 19F NMR (Fig. 2) were also effective
probes to follow. More importantly, these experiments re-
vealed an asymmetric induction from 5 onto the cations as one
of the diastereoisomeric ion pairs, [P-3][D-5] or [M-3][D-5], is
clearly favored in solution. Integration of the split signals gave
ratios from 1.5 : 1 to 4.0 : 1 (ꢁ2–3%) corresponding to
diastereomeric excesses of 20% to 60% (ꢁ2–3%), respec-
tively.22 The results are reported in Table 1. Better selectivities
were obtained for the cations bearing electron-withdrawing
halogen atoms.
This hypothesis was assessed by preparing a series of
compounds 3 with various substituents at the 5-position on
the indolinium ring. Salts [3a][Br] to [3e][Br] (Table 1, Z ¼ H,
OMe, OBn, F, Cl), were prepared in one step by condensation
of 2,20-bis(bromomethyl)biphenyl and the respective commer-
cially available or simply prepared indolines (K2CO3, CH3CN,
80 1C, 71–90%). The ion pairing with anion 5 was realized by
mixing solutions of the bromides salts with that of [Me2NH2]-
[D-5] (or its enantiomer, 1.2 equiv.) in CH2Cl2–acetone. Chro-
matography of the mixtures on basic alumina (eluent CH2Cl2)
afforded the desired pairs [3a][D-5] to [3e][D-5] as the only
eluted salts (70–98%).17 Salt [3a][L-5] was prepared similarly
(87%).
For the Stevens rearrangement, care was taken to select a
neutral base so as not to upset the possible asymmetric ion
pairing situation by addition of another salt, and Schwesinger
base P4-t-Bu 6 in particular.18,19 Treatment of the bromide
salts of ammonium cations 3a to 3e with 6 (1.5 equiv., CH2Cl2,
ꢀ80 1C, 4 h) resulted, as desired, in the formation of the
racemic rearrangement products 4a to 4e. The yield was
excellent for 4a (92%) and moderate with the substituted
derivatives 4b to 4e (46–53%, see ESIw).20 With these results
and conditions in hand, salt [3a][D-5] and its enantiomer
[3a][L-5] were treated with 6 to yield non-racemic amines
(þ)-4a and (ꢀ)-4a, respectively. Chiral Stationary Phase
(CSP)-HPLC revealed an enantiomeric purity of 33% for both
(Table 1). The occurrence of an enantioselective reaction was
confirmed by the reactions of salts [3b][D-5] to [3e][D-5] which
afforded amines (þ)-4b to (þ)-4e. The yields were in complete
accordance (ꢁ2%) with that of the bromide salts. Whereas
electron-donating ether substituents led to poor ee values
(20% and 27% for OBn and OMe), the presence of electron-
withdrawing halogen atoms (F and Cl) improved the enantio-
selectivity to 49% and 55%, respectively; values high enough
to be clearly considered as ‘‘proof of concept’’.21 The sub-
stituent effect, somewhat surprising at first glance, could be
rationalized in the course of the following study.
Significantly, the values for the diastereoselectivity within
the asymmetric ion pairs are essentially identical to that of the
enantiomeric purity of the corresponding tertiary amines 4.
Direct comparison between the two sets of data points to the
existence of an essentially linear correlation.23 This indicates
that the ToC from the preferred atropisomers of 3 to the non-
racemic amines 4 occurs with excellent stereoselectivity
(from 90 to 100%, Table 1).
Interestingly, as we have mentioned earlier, achieving such a
high selectivity was not obvious considering the probable
mechanism (Scheme 1). The [1,2]-Stevens rearrangement of 3
should involve principally two successive intermediates. The
first is zwitterionic ylide 7 generated by deprotonation and the
second is radical pair 8 produced by homolytic fragmentation
(Scheme 1). Both compounds 7 and 8 are neutral and hence
afford the possibility for 5 to diffuse out of the reaction pocket.
Loss of enantiomeric purity can then simply occur by (a)
rotation around the biaryl axis of 7 or 8, or (b) by rotation
(1801) of the Caryl–CHN bond of 8. Although it is premature
at this stage to speculate, the high selectivity for the ToC
To establish that the enantioselectivity of the reactions was,
as imagined, the result of the predominance of one diaster-
ꢂc
This journal is The Royal Society of Chemistry 2008
830 | Chem. Commun., 2008, 829–831

View Article Online
Asymmetry, 2003, 14, 917; (h) S. Hanessian and M. Mauduit,
Angew. Chem., Int. Ed., 2001, 40, 3810; (i) R. Pedrosa, C. Andres
and M. Delgado, Synlett, 2000, 893; (j) J. P. Liou and C. Y. Cheng,
Tetrahedron Lett., 2000, 41, 915; (k) L. S. Beall and A. Padwa,
Tetrahedron Lett., 1998, 39, 4159, and references therein.
9 (a) I. G. Stara, I. Stary, M. Tichy, J. Zavada and V. Hanus, J. Am.
Chem. Soc., 1994, 116, 5084; (b) H. Joshua, R. Gans and K.
Mislow, J. Am. Chem. Soc., 1968, 90, 4884.
Scheme 1 Possible intermediates 7 and 8 in the rearrangement of 3 to
4. Loss of chirality may occur by rotation of (a) the biaryl C–C bond
or (b) of the Caryl–CHN bond; configurations are assumed.24
10 V. K. Aggarwal, J. N. Harvey and R. Robiette, Angew. Chem., Int.
Ed., 2005, 44, 5468.
11 (a) S. Superchi, R. Bisaccia, D. Casarini, A. Laurita and C. Rosini,
J. Am. Chem. Soc., 2006, 128, 6893; (b) T. Ooi, Y. Uematsu, M.
Kameda and K. Maruoka, Angew. Chem., Int. Ed., 2002, 41, 1551;
(c) P. Rashidi-Ranjbar, S. Taghvaei-Ganjali, S. L. Wang, F. L.
Liao and A. Heydari, J. Chem. Soc., Perkin Trans. 2, 2001, 1255;
(d) L. A. Saudan, G. Bernardinelli and E. P. Kundig, Synlett, 2000,
483; (e) M. Tichy, M. Budesinsky, J. Gunterova, J. Zavada, J.
Podlaha and I. Cisarova, Tetrahedron, 1999, 55, 7893; (f) B.
Kiupel, C. Niederalt, M. Nieger, S. Grimme and F. Vogtle, Angew.
Chem., Int. Ed., 1998, 37, 3031; (g) I. O. Sutherland and M. V. J.
Ramsay, Tetrahedron, 1965, 21, 3401.
12 Considering first-order kinetics, this correspond to half-lives in
the range of the 10ꢀ4 s and minutes at 25 1C and ꢀ80 1C,
respectively.
13 K. Mikami, K. Aikawa, Y. Yusa, J. J. Jodry and M. Yamanaka,
Synlett, 2002, 1561.
probably indicates that all the steps leading to 4 are extremely
fast on the reaction time-scale.2
In conclusion, this paper reports that a strict enantioselec-
tive [1,2]-Stevens rearrangement of quaternary ammonium
ions is feasible, using enantiopure anionic counterions as
asymmetric auxiliaries in particular. The methodology consti-
tutes an interesting example of double transmission of chir-
ality: (i) a supramolecular transfer of the helical chirality of
anion 5 to the axial chirality of cation 3 and then (ii) its very
effective translation (90 to 100%) during the [1,2]-Stevens
rearrangement to the centered chirality of amines 4.
We are grateful for financial support of this work by the
Swiss National Science Foundation, the State Secretariat for
Education and Science for financial support.
14 BINPHAT: bis(tetrachlorobenzenediolato)mono([1,10]binaphth-
alenyl-2,20-diolato) phosphate(V) anion: J. Lacour, A. Londez,
C. Goujon-Ginglinger, V. Buß and G. Bernardinelli, Org. Lett.,
2000, 2, 4185.
15 (a) R. M. Yeh and K. N. Raymond, Inorg. Chem., 2006, 45, 1130;
(b) C. Bonnot, J.-C. Chambron and E. Espinosa, J. Am. Chem.
Soc., 2004, 126, 11412; (c) M. M. Green, C. Khatri and N. C.
Peterson, J. Am. Chem. Soc., 1993, 115, 4941.
References
1 (a) J. A. Vanecko, H. Wan and F. G. West, Tetrahedron, 2006, 62,
1043; (b) I. E. Marko, in The Stevens and Related Rearrangements,
ed. B. M. Trost and I. Fleming, Wiley, Oxford, 1991, p. 913.
2 (a) Y. Maeda and Y. Sato, J. Chem. Soc., Perkin Trans. 1, 1997,
1491; (b) W. D. Ollis, M. Rey and I. O. Sutherland, J. Chem. Soc.,
Perkin Trans. 1, 1983, 1009; (c) K. Chantrapromma, W. D. Ollis
and I. O. Sutherland, J. Chem. Soc., Perkin Trans. 1, 1983, 1049;
(d) W. D. Ollis, M. Rey, I. O. Sutherland and G. L. Closs, J. Chem.
Soc., Chem. Commun., 1975, 543.
16 (a) O. Winkelmann, D. Linder, J. Lacour, C. Nather and U.
Luning, Eur. J. Org. Chem., 2007, 3687; (b) B. Laleu, G. Bernar-
dinelli, R. Chauvin and J. Lacour, J. Org. Chem., 2006, 71, 7412;
(c) S. A. Vignon, J. Wong, H.-R. Tseng and J. F. Stoddart, Org.
Lett., 2004, 6, 1095; (d) E. Martınez-Viviente, P. S. Pregosin, L.
Vial, C. Herse and J. Lacour, Chem.–Eur. J., 2004, 10, 2912; (e) S.
Hiraoka, K. Harano, T. Tanaka, M. Shiro and M. Shionoya,
Angew. Chem., Int. Ed., 2003, 42, 5182; (f) L. Vial and J. Lacour,
Org. Lett., 2002, 4, 3939.
3 (a) For tertiary amines and in the context of Lewis acid-mediated
enantioselective [2,3]-Stevens rearrangements, competing [1,2]-
rearrangements occur with decent enantioselectivity: J. Blid,
O. Panknin, P. Tuzina and P. Somfai, J. Org. Chem., 2007, 72,
1294; (b) J. Blid, O. Panknin and P. Somfai, J. Am. Chem. Soc.,
2005, 127, 9352.
4 (a) For early examples, see: J. H. Brewster and R. S. Jones, J. Org.
Chem., 1969, 34, 354; (b) R. K. Hill and T. H. Chan, J. Am. Chem.
Soc., 1966, 88, 866. No conclusion on the extent of the ToC can
however be gathered from this early literature.
5 (a) E. Tayama, S. Nanbara and T. Nakai, Chem. Lett., 2006, 35,
478; (b) K. W. Glaeske and F. G. West, Org. Lett., 1999, 1, 31.
6 (a) J. Lacour and D. Linder, Science, 2007, 317, 462; (b) J. Lacour
and R. Frantz, Org. Biomol. Chem., 2005, 3, 15; (c) J. Lacour and
V. Hebbe-Viton, Chem. Soc. Rev., 2003, 32, 373.
7 (a) G. L. Hamilton, E. J. Kang, M. Mba and F. D. Toste, Science,
2007, 317, 496; (b) M. Rueping and A. P. Antonchick, Angew.
Chem., Int. Ed., 2007, 46, 4562; (c) S. Mayer and B. List, Angew.
Chem., Int. Ed., 2006, 45, 4193; (d) R. Gausepohl, P. Buskens, J.
Kleinen, A. Bruckmann, C. W. Lehmann, J. Klankermayer and W.
Leitner, Angew. Chem., Int. Ed., 2006, 45, 3689.
17 (a) V. Desvergnes-Breuil, V. Hebbe, C. Dietrich-Buchecker, J.-P.
Sauvage and J. Lacour, Inorg. Chem., 2003, 42, 255; (b) D.
Monchaud, J. J. Jodry, D. Pomeranc, V. Heitz, J.-C. Chambron,
J.-P. Sauvage and J. Lacour, Angew. Chem., Int. Ed., 2002, 41,
2317.
18 R. Schwesinger and H. Schlemper, Angew. Chem., Int. Ed. Engl.,
1987, 26, 1167.
19 L. Vial, M.-H. Goncalves, P.-Y. Morgantini, J. Weber, G. Ber-
nardinelli and J. Lacour, Synlett, 2004, 1565.
20 While the exact reasons for the lower yields cannot be advanced at
this stage, it is clear that competitive pathways may occur under
these strongly basic conditions—and Hofmann elimination in
particular.
21 The global ee values may appear to be low but they are in line
with that of the enantiospecific studies (ref. 5) if one takes into
account that recrystallizations are necessary at the stage of the
formation of the stereogenic quaternary ammonium ions to
remove minor diastereomeric salts. Without the purification prior
to the Stevens rearrangement, lower levels of enantioselectivity are
achieved.
22 At 193 K, the precision of the measurement is slightly limited by
some line-broadening. At 233 K (ꢀ40 1C), the NMR resolution
was sharper and the diastereomeric excesses measured are in
complete agreement with that observed at ꢀ80 1C.
8 (a) F. Couty, F. Durrat, G. Evano and J. Marrot, Eur. J. Org.
Chem., 2006, 4214; (b) J. A. Vanecko and F. G. West, Org. Lett.,
2005, 7, 2949; (c) M. Valpuesta, A. Diaz, R. Suau and G. Torres,
Eur. J. Org. Chem., 2004, 4313; (d) D. Muroni, A. Saba and N.
Culeddu, Tetrahedron: Asymmetry, 2004, 15, 2609; (e) M. Harada,
T. Nakai and K. Tomooka, Synlett, 2004, 365; (f) S. Hanessian, S.
Parthasarathy and M. Mauduit, J. Med. Chem., 2003, 46, 34; (g)
K. W. Glaeske, B. N. Naidu and F. G. West, Tetrahedron:
23 Linear regression of de ¼ f(ee) affords the following trendline
(y ¼ 0.9109x þ 1.6395, R2 ¼ 0.9879).
24 By analogy with the study of Zavada and coworkers, the (P)-3 and
(M)-3 conformations ought to lead to the formation of (R)-4 and
(S)-4, respectively.
ꢂc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 829–831 | 831