Asymmetric synthesis of enantio-enriched acyclic α-amino alkylstannanes and rearrangement behavior of carbanions thereof

Article abstract of DOI:10.1016/S0040-4039(02)02811-3

An efficient stereocontrolled synthesis of enantio-enriched N-Boc-N-allyl-α-amino alkylstannanes and N,N-diallylic α-amino alkylstannanes starting from enantio-enriched α-hydroxy alkylstannane has been developed. The aza-Wittig rearrangement of enantio-de

Full text of DOI:10.1016/S0040-4039(02)02811-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 1239–1242
Asymmetric synthesis of enantio-enriched acyclic a-amino
alkylstannanes and rearrangement behavior of carbanions thereof
Takahiro Tomoyasu,† Katsuhiko Tomooka* and Takeshi Nakai
Department of Applied Chemistry, Graduate School of Science and Engineering, Tokyo Institute of Technology, Meguro-ku,
Tokyo 152-8552, Japan
Received 31 October 2002; revised 2 December 2002; accepted 6 December 2002
Abstract—An efficient stereocontrolled synthesis of enantio-enriched N-Boc-N-allyl-a-amino alkylstannanes and N,N-diallylic
a-amino alkylstannanes starting from enantio-enriched a-hydroxy alkylstannane has been developed. The aza-Wittig rearrange-
ment of enantio-defined N,N-diallyl-a-amino alkyllithium, generated by tin–lithium exchange, is shown to proceed predominantly
with inversion of configuration at the Li-bearing carbon terminus. © 2003 Elsevier Science Ltd. All rights reserved.
The Wittig rearrangement of allyl ether systems
(Scheme 1, Y=O), particularly its asymmetric version
using enantio-enriched substrates, has enjoyed wide-
spread application in stereocontrolled synthesis.1 In
contrast, despite its potential utility, the aza-Wittig
rearrangement (Scheme 1, Y=NR%%) is far less exploited
in asymmetric synthesis because of the considerable
difficulty encountered in the generation of enantio-
enriched a-amino carbanions.2 In their pioneering
work, Gawley and co-workers have reported the aza-
Wittig rearrangement of (R)-N-allyl-2-lithiopyrrolidine,
generated by a tin-lithium transmetalative protocol,3
which affords the aza-Wittig product in moderate enan-
tiomeric purity.4 However, no aza-Wittig rearrange-
ment of acyclic a-amino alkylstannanes, initiated by
tin–lithium exchange, has been reported yet except for
the examples of primary a-amino methylstannane sys-
pounds, i.e. a-amino alkylstannanes, had so far been
tedious,7 at first we examined improvement of its syn-
thetic route from easily available a-hydroxy alkylstan-
nane (S)-18 via mesylation followed by the SN2 reaction
with amino reagents (Scheme 3).9,10
As the result, the required N-allyl aminostannane (R)-2
(>95% ee) was obtained in 65% overall yield from
a-hydroxy phenylpropylstannane (S)-1 (>95% ee) with-
out loss of stereochemical integrity.11,12 The key feature
of this method is the high degree of inversion of
tems.5 We now report
a new efficient synthetic
approach to acyclic enantio-enriched N,N-disubstituted
a-amino alkylstannanes A and our preliminary results
of the aza-Wittig rearrangement thereof, which involves
an enantio-defined a-amino alkyllithium B generated by
tin–lithium exchange (Scheme 2).
Scheme 1.
As a substrate for this rearrangement, initially we chose
N-allylic a-amino alkylstannanes which contained a
Boc group on nitrogen to stabilize the configuration of
the resulting a-amino carbanion due to its coordination
effect.6 While the preparation of this class of com-
* Corresponding author. Fax: +813-5734-3931; e-mail: ktomooka@
o.cc.titech.ac.jp
† Present address: Otsuka Pharmaceutical Factory, Inc., Muya-cho,
Naruto, Tokushima 772-8601, Japan
Scheme 2.
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)02811-3

1240
T. Tomoyasu et al. / Tetrahedron Letters 44 (2003) 1239–1242
Scheme 3.
configuration in the SN2 process. Similar reactions
using N-Boc-benzylamine and phthalimide provide the
corresponding stannane (R)-3 and (R)-4 as an enan-
tiopure form, respectively.11,13 Next, we examined the
transmetalative rearrangement of the thus-obtained
(R)-2. However, disappointingly, reaction of (R)-2 with
n-BuLi provided neither [1,2]- nor [2,3]-Wittig rear-
rangement product, and, instead, deuterated product 5
was obtained by quenching with D2O at that tempera-
ture (Scheme 4).14
one. The requisite stannane (R)-7 was prepared from
phthalimide derived (R)-4 via reaction with hydrazine
hydrate followed by bis-allylation.15 Reaction of N,N-
diallyl aminostannane (R)-7 with n-BuLi was found to
afford the desired aza-Wittig rearrangement product 8
in 50% yield (Scheme 5).16,17 The absolute stereochem-
istry and enantiopurity of 8 was determined to be 48%
ee (R) by the HPLC analysis of its (S)-methylbenzyl
urea derivative 9.18
Since tin–lithium exchange proceeds with retention of
configuration,19 it appears that the rearrangement pro-
ceeds predominantly with inversion at the Li-bearing
carbon terminus. While a complete explanation of the
loss of enantiospecificity is not possible at present, it
might be taken into account that the a-amino alkyl-
lithium intermediate might be configurationally labile
This result suggests that the Boc-protected a-amino
alkyllithium thus generated is still not reactive enough
to undergo the aza-Wittig rearrangement, probably due
to the remarkable dipole stabilization of the a-amino
carbanion. These observations prompted us to redesign
the rearrangement system for a non-Boc substituted
Scheme 4.
Scheme 5.

T. Tomoyasu et al. / Tetrahedron Letters 44 (2003) 1239–1242
1241
Scheme 6.
and/or aza-Wittig rearrangements would proceed with
low stereospecificity. Furthermore, a similar reaction of
(R)-10 (E/Z mixture) was found to afford a mixture of
[1,2]-rearrangement product 11 and [2,3]-rearrangement
product 12, which means that this class of rearrange-
ment proceeds with low periselectivity (Scheme 6).20
Chem. Soc. 1971, 93, 4027–4031; (b) Quintard, J.-P.;
Elissondo, B.; Jousseaume, B. Synthesis 1984, 495–498;
(c) Elissondo, B.; Verlhac, J.-B.; Quintard, J.-P.; Pereyre,
M. J. Organomet. Chem. 1988, 339, 267–275; (d)
Ahlbrecht, H.; Baumann, V. Synthesis 1994, 770–772; (e)
Pearson, W. H.; Stevens, R. P. Synthesis 1994, 904–906;
(f) Katritzky, A. R.; Chang, H.-X.; Wu, J. Synthesis
1994, 907–908; (g) Burchat, A. F.; Chong, J. M.; Nielsen,
N. J. Org. Chem. 1996, 61, 7627–7630.
In summary, we have developed a convenient method
for the preparation of enantio-enriched N-Boc-N-allyl-
a-amino alkylstannanes and N,N-diallylic a-amino
alkylstannanes from enantio-enriched a-hydroxy alkyl-
stannane. Furthermore, we have demonstrated that the
first example of aza-Wittig rearrangement of acyclic
enantio-enriched N,N-diallylic a-amino alkylstannanes
via tin-lithium exchange proceeds predominantly with
inversion of configuration at the lithium-bearing carbon
terminus. Further work is in progress to elucidate the
mechanism of this rearrangement and to enhance the
synthetic potential thereof.
8. (a) Marshall, J. A.; Gung, W. Y. Tetrahedron Lett. 1988,
29, 1657–1660; (b) Chan, C. M.; Chong, J. M. J. Org.
Chem. 1988, 53, 5584–5586; (c) Tomooka, K.; Igarashi,
T.; Nakai, T. Tetrahedron Lett. 1994, 35, 1913–1916.
9. Chong’s group has reported that enantio-enriched a-
phthalimido alkylstannane can be prepared by the Mit-
sunobu reaction of enantio-enriched a-hydroxy alkyl-
stannane and phthalimide, see: Ref. 3a.
10. The SN2 reaction with alkoxides has been reported:
Tomooka, K.; Igarashi, T.; Nakai, T. Tetrahedron Lett.
1993, 34, 8139–8142.
11. The enantiopurity of (R)-2,3 was determined by chiral
HPLC analysis (Chiralcel OD-H, 0.46×25 cm).
Acknowledgements
12. All the compounds were characterized by 1H and 13C
NMR analysis. Data for selected products are as follows.
1
This research was supported by a Grant-in-Aid for
Scientific Research (B) from the Ministry of Education,
Culture, Sports, Science and Technology, Japan.
(R)-2: H NMR (CDCl3) l 0.80–0.95 (m, 15H), 1.24–1.58
(m, 12H), 1.43 (s, 9H), 2.00–2.15 (m, 2H), 2.50–2.65 (m,
2H), 2.92–3.00 (m, 1H), 3.54 (dd, J=7.4, 15.0 Hz, 1H),
4.03 (dd, J=5.8, 15.0 Hz, 1H), 5.05–5.13 (m, 2H), 5.70–
5.85 (m, 1H), 7.16–7.31 (m, 5H). 13C NMR (CDCl3) l
10.6, 13.7, 27.6, 28.5, 29.2, 34.6, 34.9, 49.0, 53.8, 79.0,
116.6, 125.7, 128.3, 128.3, 135.0, 142.3, 155.4. [h]2D5 +10.5°
References
1
1. Reviews: (a) Marshall, J. A. In Comprehensive Organic
Synthesis; Trost, B. M.; Fleming, I., Eds.; Pergamon:
New York, 1991; Vol. 3, pp. 975–1014; (b) Nakai, T.;
Mikami, K. Org. React. 1994, 46, 105–209; (c) Nakai, T.;
Tomooka, K. Pure Appl. Chem. 1997, 69, 595–600.
2. Review: Vogel, C. Synthesis 1997, 497–505.
3. A tin–lithium transmetalative method provides easy
access to configurationally stable a-amino alkyllithiums:
(a) Chong, J. M.; Park. S. B. J. Org. Chem. 1992, 57,
2220–2222; (b) Burchat, A. F.; Chong, J. M.; Park. S. B.
Tetrahedron Lett. 1993, 34, 51–54; (c) Gawley, R. E.;
Zhang, Q. J. Org. Chem. 1995, 60, 5763–5769.
(c 1.0, CHCl3). (R)-3: H NMR (CDCl3) l 0.80–0.95 (m,
15H), 1.19–1.52 (m, 12H), 1.46 (s, 9H), 2.00–2.11 (m,
2H), 2.45–2.55 (m, 2H), 2.81–2.87 (m, 1H), 4.07 (d,
J=15.0 Hz, 1H), 4.67 (d, J=15.0 Hz, 1H), 7.06–7.37 (m,
10H). 13C NMR (CDCl3) l 10.6, 13.7, 27.5, 28.5, 29.1,
31.6, 34.4, 48.8, 53.3, 79.3, 125.7, 127.2, 127.8, 128.2,
128.3, 128.7, 138.8, 142.2, 155.5. [h]2D5 +17.6° (c 0.25,
1
CHCl3). (R)-4: H NMR (CDCl3) l 0.81–1.53 (m, 27H),
1.95–2.09 (m, 1H), 2.28–2.43 (m, 1H), 2.59 (t, J=7.8 Hz,
2H), 4.04 (dd, J=5.6, 10.2 Hz, 1H), 7.05–7.25 (m, 5H),
7.65–7.69 (m, 2H), 7.75–7.80 (m, 2H). 13C NMR (CDCl3)
l 10.3, 13.6, 27.3, 28.9, 34.6, 34.9, 37.4, 122.8, 125.6,
128.2, 132.0, 133.6, 141.4, 169.1. [h]2D5 +76.0° (c 1.0,
4. Gawley, R. E.; Zhang, Q.; Campagna, S. J. Am. Chem.
1
Soc. 1995, 117, 11817–11818.
CHCl3). (R)-7: H NMR (CDCl3) l 0.81–0.91 (m, 15H),
1.24–1.52 (m, 12H), 1.72–1.86 (m, 1H), 2.02–2.19 (m,
1H), 2.47–2.60 (m, 1H), 2.63 (dd, J=14.0, 7.3 Hz, 2H),
2.81–2.95 (m, 1H), 3.09 (dd, J=10,2, 5.4 Hz, 1H), 3.29
(dd, J=14.0, 4.8 Hz, 2H), 5.05–5.19 (m, 4H), 5.81 (dddd,
J=17.0, 10.2, 7.3, 4.8 Hz, 2H), 7.16–7.30 (m, 5H). 13C
NMR (CDCl3) l 10.6, 13.6, 27.6, 29.4, 34.8, 35.2, 57.0,
57.5, 116.4, 125.5, 128.2, 128.5, 137.6, 142.9. [h]2D5 −60.0°
(c 0.25, CHCl3). (R)-8: 1H NMR (CDCl3) l 1.69–1.77 (m,
2H), 2.17–2.31 (m, 2H), 2.63–2.69 (m, 3H), 3.23 (dd,
J=5.9, 1.3 Hz, 2H), 5.04–5.18 (m, 4H), 5.74–5.93 (m,
5. (a) Peterson, D. J.; Ward, J. F. J. Organomet. Chem.
1974, 66, 209–217; (b) Coldham, I. J. Chem. Soc., Perkin
Trans. 1 1993, 1275–1276.
6. The first synthesis of N-allyl-N-Boc-a-amino alkylstan-
nane was reported by Chong’s group. However, they
have not commented on the aza-Wittig rearrangement
thereof, see: Ref. 3a.
7. A wide variety of methods have been reported for the
preparation of acyclic a-amino alkylstannanes but all give
primary or racemic products. (a) Peterson, D. J. J. Am.

1242
T. Tomoyasu et al. / Tetrahedron Letters 44 (2003) 1239–1242
2H), 7.17–7.30 (m, 5H). 13C NMR (CDCl3) l 32.0, 35.6,
38.3, 49.6, 55.5, 115.7, 117.4, 125.7, 128.3, 128.3, 135.4,
137.2, 142.5. (R)-10 (major diastereomer): 1H NMR
(CDCl3) l 0.81–0.91 (m, 15H), 1.24–1.73 (m, 18H), 1.75–
1.90 (m, 1H), 2.09–2.22 (m, 1H), 2.49–2.65 (m, 3H),
2.81–2.95 (m, 1H), 3.10–3.31 (m, 3H), 5.40–5.63 (m, 4H),
7.16–7.35 (m, 5H). 13C NMR (CDCl3) l 10.7, 13.7, 17.8,
27.6, 29.4, 34.8, 35.0, 50.8, 56.6, 125.5, 127.4, 128.2,
128.6, 130.2, 143.0. 11 (major diastereomer): 1H NMR
(CDCl3) l 1.55–1.69 (m, 8H), 2.13–2.17 (m, 2H), 2.54–
2.61 (m, 3H), 3.07–3.20 (m, 2H), 5.34–5.54 (m, 4H),
7.07–7.23 (m, 5H). 13C NMR (CDCl3) l 13.1, 17.8, 31.3,
32.1, 35.7, 49.1, 56.5, 125.7, 125.9, 126.3, 127.0, 128.3,
128.3, 129.8, 142.6. 12 (major diastereomer): 1H NMR
(CDCl3) l 1.03–1.08 (m, 3H), 1.62–1.90 (m, 5H), 2.41–
2.52 (m, 2H), 2.55–2.85 (m, 2H), 3.17–3.30 (m, 2H),
5.07–5.13 (m, 2H), 5.51–5.84 (m, 3H), 7.18–7.34 (m, 5H).
13C NMR (CDCl3) l 15.5, 17.7, 32.6, 32.8, 40.3, 49.6,
18. Authentic samples of (S)-8 and (S,S)-9 were prepared as
depicted below. The starting stannane (S)-14 (>95% ee)
was prepared by our reported method, see: Tomooka, K.;
Shimizu, H.; Inoue, T.; Shibata, H; Nakai, T. Chem. Lett.
1999, 1, 759–760.
60.5, 114.8, 125.6, 126.9, 128.3, 128.3, 130.0, 141.9, 142.8.
1
13. The enantiopurity of (R)-4 was determined by H NMR
analysis after conversion (hydrazine hydrate/MTPACl) to
a MTPA amide of (R)-6.
14. Treatment of (R)-2 with n-BuLi at −78“0°C led to
complex mixtures of products.
15. Unfortunately, a previously developed synthetic method
from (S)-1 was not applicable to the preparation of
diallyl derivatives.
16. The intramolecular coordination between the lithium and
the olefinic p-bond may play a very important role for
configurational stability of a-amino alkyllithiums, see:
Komine, N.; Tomooka, K.; Nakai, T. Heterocycles 2000,
52, 1071–1074.
17. In contrast, a similar reaction of N-allyl-N-Me-a-amino
alkylstannane 13 gave no rearrangement product. Thus,
the reasonable coordination ability of the olefinic p-bond
in the transition state C appears to be well suited for
aza-Wittig rearrangement.
19. Retention of stereochemistry in tin–lithium exchange
reactions is well-known for the preparation of stereo-
chemically defined a-hetero alkyllithiums, see: Pearson,
W. H.; Lindbeck, A. C.; Kampf, J. W. J. Am. Chem. Soc.
1993, 115, 2622–2636 and references cited therein.
20. Rearrangement products 11 and 12 were obtained as a
complex mixture of several diastereomers. Thus, the
enantiopurity of the products has not yet been deter-
mined.