Silylstannylation of allenes and silylstannylation-cyclization of allenynes. Synthesis of highly functionalized allylstannanes and carbocyclic and heterocyclic compounds

Article abstract of DOI:10.1021/jo049010z

Catalyzed by Pd(0), trialkylsilyltrialkylstannane (R₃Si- SnR′₃) reagents undergo highly selective additions to 1,2-dien-7-ynes and 1,2-dien-8-ynes to give 2-vinylalkylidenecyclopentanes with silicon and tin substituents on the double bonds. Similar additions of distannanes and borostannanes show that the reactions with silylstannanes are superior in terms of ease of handling of the bifunctional reagents and the isolation of the products after the reaction. The chemo- and regioselectivities are controlled by the enhanced reactivity of the allene unit, while the (Z)-geometry of the exocyclic stannylvinylidene is a consequence of the syn-carbometalation and subsequent reductive elimination from Pd with retention of configuration at the vinyl carbon. Synthesis of highly functionalized pyrrolidines and indolizidines and the reluctance of certain kinds of allenynes and silicon-tin reagents to undergo the cyclization illustrate the scope and limitations of the reaction. Based on the isolation of intermediates, a mechanism for the formation of the cyclic compounds is proposed. Model transition states to explain the stereoselectivity in cyclization of substituted allenynes are provided. Further elaboration using the vinyltin and vinylsilane moieties should lead to highly functionalized carbocyclic and heterocyclic compounds. Under similar conditions, addition of silylstannanes to highly functionalized allenes gives E-allylstannanes with high stereoselectivity. Functional groups such as THP- and silyl-ethers, lactones, β- and γ-lactams, α,β-unsaturated esters, olefins, and substituted acetylenes are tolerated under the reaction conditions.

Full text of DOI:10.1021/jo049010z

Silylsta n n yla tion of Allen es a n d Silylsta n n yla tion -Cycliza tion of
Allen yn es. Syn th esis of High ly F u n ction a lized Allylsta n n a n es a n d
Ca r bocyclic a n d Heter ocyclic Com p ou n d s
Ramaiah Kumareswaran, Seunghoon Shin,† Isabelle Gallou, and T. V. RajanBabu*
Department of Chemistry, The Ohio State University, 100 West 18th Avenue, Columbus, Ohio 43210
rajanbabu.1@osu.edu
Received J une 11, 2004
Catalyzed by Pd(0), trialkylsilyltrialkylstannane (R3Si-SnR′3) reagents undergo highly selective
additions to 1,2-dien-7-ynes and 1,2-dien-8-ynes to give 2-vinylalkylidenecyclopentanes with silicon
and tin substituents on the double bonds. Similar additions of distannanes and borostannanes
show that the reactions with silylstannanes are superior in terms of ease of handling of the
bifunctional reagents and the isolation of the products after the reaction. The chemo- and
regioselectivities are controlled by the enhanced reactivity of the allene unit, while the (Z)-geometry
of the exocyclic stannylvinylidene is a consequence of the syn-carbometalation and subsequent
reductive elimination from Pd with retention of configuration at the vinyl carbon. Synthesis of
highly functionalized pyrrolidines and indolizidines and the reluctance of certain kinds of allenynes
and silicon-tin reagents to undergo the cyclization illustrate the scope and limitations of the
reaction. Based on the isolation of intermediates, a mechanism for the formation of the cyclic
compounds is proposed. Model transition states to explain the stereoselectivity in cyclization of
substituted allenynes are provided. Further elaboration using the vinyltin and vinylsilane moieties
should lead to highly functionalized carbocyclic and heterocyclic compounds. Under similar
conditions, addition of silylstannanes to highly functionalized allenes gives E-allylstannanes with
high stereoselectivity. Functional groups such as THP- and silyl-ethers, lactones, â- and γ-lactams,
R,â-unsaturated esters, olefins, and substituted acetylenes are tolerated under the reaction
conditions.
In tr od u ction
barriers for enantiomerization (<60 kJ mol-1) irrespec-
tive of the size of the silicon and tin groups.6 As expected,
the dienes undergo highly selective reactions such as
protodestannylation, Sn-halogen exchange, Stille-cou-
pling, and Diels-Alder reactions (after destannylation).
However, reactions of unsymmetrical diynes (R * H in
Many traditional methods of cyclization are limited by
the fact that often functional groups are depleted in the
ring-forming event.1 Several strategies have been de-
signed to overcome this problem, including invention of
methods that incorporate elements of various bifunctional
reagents (e. g., R3SiH, R3SnH, R3Si-SiR′3, R3Si-SnR′3,
R3Si-BR′2, and R3Sn-BR′2) concurrent with the cycliza-
tion reaction.2 We have been interested in the use of
readily available, air- and moisture-stable trialkylsilyl-
trialkylstannanes3 in the cyclization reactions of diynes4
and allenynes.5 We recently reported the details of silyl-
stannylation cyclization of diynes in which helically chiral
(ZZ)-1,2-bis-alkylidenecyclopentanes are produced (eq 1).
These fascinating molecules have very low activation
(2) For recent review articles, see: (a) Suginome, M.; Ito, Y. Chem.
Rev. 2000, 100, 3221. (b) Smith, N. D.; Mancuso, J .; Lautens, M. Chem.
Rev. 2000, 100, 3257. (c) Itoh, K.; Matsuda, I.; Yamamoto, K. J . Synth.
Org. Chem. J pn. 1999, 57, 912. For recent representative publications
that deal with R3SiH and R3SnH, see: (d) Ojima, I.; Vu, A. T.; Lee,
S.-Y.; McCullah, J . V.; Moralee, A. C.; Fujiwara, M.; Hoang, T. H. J .
Am. Chem. Soc. 2002, 124, 9164 and references therein. (d) Kisanga,
P.; Widenhoefer, R. A. J . Am. Chem. Soc. 2000, 122, 10017. (e) Lautens,
M.; Mancuso, J . Org. Lett. 2000, 2, 671. (f) R3Si-BR2: Onozawa, S.;
Hatanaka, Y.; Tanaka, M. J . Chem. Soc., Chem. Commun. 1997, 1229.
Suginome, M.; Matsuda, T.; Ito, Y. Organometallics 1998, 5233. (g)
R3SnBR2: Onozawa, S.; Hatanaka, Y.; Choi, N.; Tanaka, M. Organo-
metallics 1997, 16, 5389. (h) R3SiSiR3/R3SnSnR3/R3SiSnR3: Obora, Y.;
Tsuji, Y.; Kakehi, T.; Kobayashi, M.; Shinkai, Y.; Ebihara, M.;
Kawamura, T. J . Chem. Soc., Perkin Trans. 1 1995, 599. [SiSn]-
mediated cyclization of enynes: (i) Mori, M.; Hirose, T.; Wakamatsu,
H.; Imakuni, N.; Sato, Y. Organometallics 2001, 20, 1907. (k) Of bis-
allenes: Kang, S.-K.; Baik, T.-G.; Kulak, A. N.; Ha, Y.-H.; Lim, Y.;
Park, J . J . Am. Chem. Soc. 2000, 122, 11529. (l) allene-aldehydes:
Kang, S.-U.; Ha, Y.-H.; Ko, B.-S.; Lim, Y.; J ung, J . Angew. Chem., Int.
Ed. 2002, 41, 343. (m) Pauson-Khand-type reactions: Shibata, T.;
Kadowaki, S.; Hirase, M.; Takagi, K. Synlett 2003, 573. Brummond,
K. M.; Chen, Y.; Sill, P.; You, L. J . Am. Chem. Soc. 2002, 124, 15186.
(n) Stoichiometric Ti(II)-mediated cyclization of allenynes: Urabe, H.;
Takeda, T.; Hideura, D.; Sato, F. J . Am. Chem. Soc. 1997, 119, 11295.
† Current address: Department of Chemistry, Hanyang University,
Seoul 133-791, Korea.
(1) For recent reviews on transition-metal-catalyzed cyclization
reactions, see: (a) Tamao, K.; Kobayashi, K.; Ito, Y. Synlett 1992, 539.
(b) Lautens, M.; Klute, W.; Tam, W. Chem. Rev. 1996, 96, 49. (c)
Negishi, E.-i.; Cope´ret, C.; Ma, S.; Liou, S.-Y.; Liu, F. Chem. Rev. 1996,
96, 365. (d) Ojima, I.; Tzamarioudaki, M.; Li, Z.; Donovan, R. J . Chem.
Rev. 1996, 96, 635. (e) Grubbs, R. H.; Chang, S. Tetrahedron 1998,
54, 4413. (f) Trost, B. M.; Krische, M. J . Synlett 1998, 1. (g) Sato, F.;
Urabe, H.; Okamoto, S. Chem. Rev. 2000, 100, 2835. (h) Brummond,
K. A.; Kent, J . L. Tetrahedron 2000, 56, 3263.
10.1021/jo049010z CCC: $27.50 © 2004 American Chemical Society
Published on Web 09/17/2004
J . Org. Chem. 2004, 69, 7157-7170
7157

Kumareswaran et al.
eq 1) give a mixture of products arising from incorpora-
tion of Si and Sn at either of the two termini (eq 2), and
alternate methods have to be resorted to for controlling
the regiochemistry of the silylstannylation. In an attempt
to circumvent this problem, and to probe the scope of the
cyclization reactions in other related π-unsaturated
systems, we investigated allenynes (eq 3), in which high
chemoselectivity was expected to prevail due to the
increased reactivity of the allene in the Pd-catalyzed
addition reactions.7 Structural features of the allenyne
substrates and the nature of the Si-Sn reagents some-
times prevented the cyclization, even though the initial
addition to the allene (eq 4) proceeded with excellent
regio- and stereoselectivity. Thus, a parallel study to
define the scope, limitations, and functional group com-
patibility of the silylstannylation and distannylation of
these and other highly functionalized allenes (eq 4) was
also undertaken. Simple allenes have been subjected to
the silylstannylation reaction under different conditions
by Mitchell.8 As recorded later in this paper, the regio-
and stereoselectivity and tolerance to various functional
groups appear to be significantly better under our modi-
fied conditions. The details of these investigations are
disclosed in this paper.
F IGURE 1. Allenynes used for silylstannylation studies.
SCHEME 1. Syn th esis of Allen yn es 1-3
SCHEME 2. Syn th esis of Su bstr a te 4
allenynes, shown in Figure 1, was chosen to illustrate
the various approaches for their preparation, functional
group compatibility of the cyclization reaction, and the
potential utility of the products in synthesis. Particularly
appealing are applications for the synthesis of N-hetero-
cyclic compounds such as pyrrolidines, pyrrolizidines, and
indolizidines prepared from allenynes such as 5-7.
The substrates 1-3 were prepared by propargylation
of the allenyl derivatives, which in turn were prepared
by Crabbe reaction9 (Scheme 1).
Scheme 2 describes the synthesis of the substrate 4
starting with the known allenyl alcohol 9.10 Swern
oxidation of 9 proceeds uneventfully to give the aldehyde
10, which can be transformed into the allenyne 4 in
overall 56% yield in two steps. The aldehyde 10 is also
useful in the preparation of other allene derivatives (vide
infra).
Resu lts a n d Discu ssion
A. Silylsta n n yla tion -Cycliza tion of Allen yn es.
Syn th esis of Su bstr a tes for Cycliza tion . A set of
Substrates 5 and 6 were prepared from succinimide
(Schemes 3 and 4). For the preparation of 5, succinimide
was reacted with 3-butynol under Mitsunobu conditions11
(3) (a) Chenard, B. L.; Laganis, E. D.; Davidson, F.; RajanBabu, T.
V. J . Org. Chem. 1985, 50, 3666. See also: (b) Mitchell, T. N.; Killing,
H.; Dicke, R.; Wickenkamp, R. J . Chem. Soc., Chem. Commun. 1985,
354.
(4) Gre´au, S.; Radetich, B.; RajanBabu, T. V. J . Am. Chem. Soc.
2000, 122, 8579.
(5) Shin, S.; RajanBabu, T. V. J . Am. Chem. Soc. 2001, 123, 8416.
(6) Warren, S.; Chow, A.; Fraenkel, G.; RajanBabu, T. V. J . Am.
Chem. Soc. 2003, 125, 15402.
(7) For a review of Pd-catalyzed reactions of allenes, see: Zimmer,
R.; Dinesh, C. U.; Nandanan, E.; Khan, F. A. Chem. Rev. 2000, 100,
3067.
(9) Starting materials: (a) Woo, L. W. L.; Smith, H. J .; Barrell, K.
J .; Nicholls, P. J . J . Chem. Soc., Perkin Trans. 1 1993, 2549. (b)
Djahanbini, D.; Cazes, B.; Gore, J . Tetrahedron 1987, 43, 3441. (c)
Oppolzer, W.; Ruiz-Montes, J . Helv. Chim. Acta 1993, 76, 1266. (d)
Urabe, H.; Takeda, T.; Hideura, D.; Sato, F. J . Am. Chem. Soc. 1997,
119, 11295. (f) Brandsma, L.; Verkruijsse, H. D. Synthesis of Acetylenes,
Allenes and Cumulenes: A Laboratory Manual; Elsevier Scientific:
New York, 1981.
(8) Mitchell, T. N.; Schneider, U. J . Organomet. Chem. 1991, 407,
319.
(10) Chevliakov, M. V.; Montgomery, J . J . Am. Chem. Soc. 1999,
121, 11139.
7158 J . Org. Chem., Vol. 69, No. 21, 2004

Synthesis of Highly Functionalized Allylstannanes
SCHEME 3. Syn th esis of Allen yn e 5
SCHEME 5. Syn th esis of Allen yn e 7
SCHEME 4. Syn th esis of Allen yn e 6
to give the N-(3-butynyl) derivative 11, which was
reduced with NaBH4 in ethanol under conditions de-
scribed by Hart et al.12 to get the 4-ethoxypyrrolidone 12.
Lewis acid-assisted addition of propargyltrimethylsilane
afforded the allenyne substrate 5 in overall 30% yield
starting from succinimide. Likewise, allenyne 6 was
prepared from succinimide in 48% yield by the route
shown in Scheme 4. The propargylation of the acylimin-
ium ion13 derived from 14 with tri-n-butylstannylallene
proceeds in excellent yield in the presence of trimethyl-
silyl triflate to give the allenyne 6.
SCHEME 6. Syn th esis of Su bstr a tes 8a a n d 8b
One of our goals in this project is to develop generally
applicable synthesis of highly functionalized pyrrazo-
lidines, indolizidines, and quinolizidines.14 In an initial
attempt to explore the viability of these methods for the
synthesis of these condensed heterocyclic compounds, we
prepared substrate(s) 7 as shown in Scheme 5. The
synthesis starts from propargyl alcohol 15, which was
converted into allenol 1615 in multigram scale. Mitsunobu
reaction of this alcohol with succinimide gave 65% yield
of 17. Controlled reduction of 17 produced the hydroxy-
lactam in high yield as a mixture of two diastereomers.
Reduction of the imide 17 in ethanol under acidic
conditions to get the ethoxy lactam gave unexpected
olefinic side products in the present example.12 The
alcohol(s) 18 were converted into acetoxylactam(s) 19.
Under optimized conditions, reaction of 19 with allenyl-
tributylstannane16 using TMSOTf as a Lewis acid af-
forded the required allenyne 7 as a chromatographically
inseparable mixture of isomers in 1:1 ratio. However,
reduction of the lactam function in 7 led to the pyrro-
lidines 20 and 21, which were easily separated by
chromatography.
Synthesis of two other highly functionalized allenes
carrying a â-lactam functionality is shown in Scheme 6.
These compounds were prepared from previously known
allenyl derivative 2217 to test the limits of tolerance of a
number of sensitive functional groups to the Pd-catalyzed
reaction conditions and as potential precursors for phar-
maceutically relevant bicyclic â-lactams.18
(11) Mitsunobu, O.; Wada, M.; Sano, T. J . Am. Chem. Soc. 1972,
94, 679.
(12) (a) Burnett, D. A.; Choi, J .-K.; Hart, D. J .; Tsai, Y.-M. J . Am.
Chem. Soc. 1984, 106, 8201. (b) See also: Hubert, J . C.; Wijnberg, J .
B. P. A.; Speckamp, W. N. Tetrahedron 1975, 31, 1437.
(13) For a recent review of acyliminium ion chemistry, see: (a)
Maryanoff, B. E.; Zhang, H.-C.; Cohen, J . H.; Turchi, I. J .; Maryanoff,
C. J . Chem. Rev. 2004, 104, 1431. (b) Hart, D. J . Alkaloids: Chem.
Biol. Perspect. 1988, 6, 227. (c) Hiemstra, H.; Speckamp, W. N. In
Comprehensive Organic Synthesis; Trost, B. M., Fleming, I., Eds.;
Pergamon: Oxford, 1991; Vol. 2, pp 1047-1082.
(14) Recent general reviews on pyrrolizidines, indolizidines, and
quinolizidine alkaloids: (a) Daly, J . W.; Garraffo, H. M.; Spande, T.
F. Alkaloids from amphibian skins. In Alkaloids: Chemical and
Biological Perspectives; Pelletier, S. W., Ed.; Pergamon Press: Oxford,
1998; Vol. 13, pp 1-161. (b) Colegate, S. M.; Dorling, P. R. In Handbook
of Plant and Fungal Toxicants; D’Mello, J . P. F., Ed.; CRC Press: Boca
Raton, 1997; Chapter 1, pp 1-18. (c) Nash, R. J .; Watson, A. A.; Asano,
N. In Alkaloids: Chemical and Biological Perspectives; Pelletier, S.
W., Ed.; Pergamon Press: Oxford, 1996; Vol. 11, pp 345-375. (d) Daly,
J . W. J . Nat. Prod. 1998, 61, 162. (e) Michael, J . P. Indolizidine and
Quinolizidine Alkaloids. Nat. Prod. Rep. 1997, 14, 630; 1998, 15, 571;
1999, 16, 675; 2000, 17, 579.
(15) Kimura. M.; Tanaka, S.; Tamaru, Y. Bull. Chem. Soc. J pn. 1995,
68, 1687.
(16) Boaretto, A.; Marton, D.; Tagliavini, G. J . Organomet. Chem.
1985, 297, 149.
(17) Prasad, J . S.; Liebeskind, L. S. Tetrahedron Lett. 1988, 29, 4253.
J . Org. Chem, Vol. 69, No. 21, 2004 7159

Kumareswaran et al.
TABLE 1. Cycliza tion of Allen yn es Med ia ted by R 3SiSn R′3
a
b
Key: (A) 5 mol % Pd2(dba)3 with 10 mol % P(C6F5)3; (B) 5 mol % Pd(PhCN)2Cl2 with 10 mol % P(C6F5)3. Isolated yields; yields
determined by NMR in parentheses. c Isolated material contained 5% cyclic product 24 (entry 2).
Silylsta n n yla tion -Cycliza tion of Allen yn es. The
results of cyclization of various allenynes (Figure 1)
mediated by silylstannane reagents are shown in Table
1. A typical substrate 1a , in the presence of Ph3Sn-
7160 J . Org. Chem., Vol. 69, No. 21, 2004

Synthesis of Highly Functionalized Allylstannanes
SCHEME 7. Silylsta n n yla tin -Cycliza tion of a
P r ototyp ica l Allen yn e
Table 1 shows the generality of the cyclization reaction.
Reaction of Ph3Sn-SiMe2But with N-tosylallenyne 2
gives the cyclic product 26 in >90% yield at room
temperature, whereas the oxygenated derivative 3 gives
the products 27 and 28 as single diastereomers under
the conditions cited (entries 5 and 6). The less reactive
trimethylsilyltributylstannane takes 5 h at 80 °C to
complete the cyclization of 3 to give the product 28. When
3 is treated with trimethylsilyltributylstannane at room
temperature in the presence of (PhCN)2PdCl2 and (C6F5)3P,
a mixture of an allylstannane and the cyclic product 28
(1:3) is obtained.
Structures of the cyclic products 27 and 28 were
confirmed by IR, 1H, NOE difference, and 13C NMR
spectroscopy, HMQC, COSY, HRMS, and elemental
analysis.19 For the NOE data, while consistent with the
assigned configuration of the THF derivatives 27 and 28,
it should be stressed that in five-membered carbocyclic
and heterocyclic compounds this method is usually not
reliable in assigning vicinal stereochemistry. However,
this configuration is assigned to the vinylstannanes based
on a reasonable model for the cyclization (vide infra,
under Mechanism and Stereoselectivity, Figure 3) and
the fact that only a single diastereomer is produced in
the reaction.
TABLE 2. Effect of P h osp h in e Liga n d s on F or m a tion of
23 (See Sch em e 7)
% conversion
(12 h)a
% conversion
(48 h)
entry
ligand
P(C6F5)3
P(3,5-Me2-C6H3)3
PPh3
1
2
3
4
5
61
39
<10
<10
<10
>90
>90
>60
>60
81
PBu3
PtBu3
a
Conversions based on in situ NMR; no side products were
detected. For conditions see the text.
The adduct 27 is cleanly converted into the corre-
sponding bromo compound in 74% yield by treatment
with NBS in CH2Cl2 at room temperature.19 Reaction
with formic acid in CH2Cl2 leads to protodestannylation
(78%).19
SiMe2But (1.1 equiv), Pd2(dba)3‚CHCl3 (5 mol % in Pd),
and P(C6F5)3 (10 mol %) in 1 mL of C6D6, undergoes an
exceptionally clean reaction (by NMR: >90% conversion
to the product, rest starting material) at room temper-
ature to give the cyclic product 23 in 80% isolated yield
(Scheme 7). The lower preparative yield of the product
is an indication of the instability of this relatively
sensitive material, which was isolated by column chro-
matography on silica gel using hexane containing 5%
Et3N. The structure and configuration of 23 were unam-
biguously established by NMR spectroscopy (1H, 13C,
COSY, NOE difference spectra) and elemental analysis.
The identity of the SnC-H and the geometry of the double
bond is clear from the coupling pattern (δ 6.40; d, 1J Sn-H
) 74 Hz) and the NOE difference spectrum (Figure 1,
Supporting Information). Among the large number of
monophosphine ligands that were screened for the reac-
tion (Table 2) only P(3,5-Me2-C6H3)3 and (C6F5)3P) gave
any appreciable reaction after 12 h at room temperature
(39% and 61%, respectively). The source of Pd, while not
as critically important, showed a definite trend in reac-
tivity when used in conjunction with (C6F5)3P: (PhCN)2-
PdCl2 > [Pd(allyl)Cl]2/AgOTf ∼ Pd2(dba)3‚CHCl3 . PdCl2-
(Ph3P)2 ∼ Pd(PPh3)4.
Syn th esis of P yr r olid in es a n d In d olizid in es. Al-
lenyne 4 would represent an advanced intermediate for
the synthesis of highly functionalized pyrrolidines such
as kainic acids or its congeners.20 When 4 was treated
with Me3Si-SnBu3 (1.1 equiv), a Pd precursor (5 mol %)
and ligand (10 mol %) in C6D6 at 80 °C, an efficient
conversion into a mixture of products, 29-cis and
29-tr a n s, was observed (Table 1, entry 7). Use of Pd-
(PhCN)2Cl2 was less efficient compared to Pd2(dba)3 in
this case. The 1,3-diastereoselctivity (1:1) remains the
same in both instances or when different Si-Sn reagents
were used. Sterically less encumbered Me3SiSnBu3 gave
higher yield (76%) and cleaner conversion compared with
Me2ButSiSnPh3 (50% yield).
Cyclization of allenynes derived from readily available
N-acyliminium ions13 should provide an expeditious entry
into highly functionalized bicyclic N-heterocyclic com-
pounds. Two examples are shown in entries 8 and 9 in
Table 1. The allenyne 5, prepared in three steps from
succinimide and 3-butynol (Scheme 3), gave the product
indolizidine (30) as a single isomer. The relative stereo-
chemistry of the vicinal appendages has been assigned
as trans on the basis of NMR experiments, models for
the cyclization, and the results described in the next
When the reaction was repeated with a less reactive
silylstannane, Bu3SnSiMe3, we were able to isolate an
uncyclized adduct 25 in excellent yield (81% isolated, rest
cyclized product 24) and stereochemical purity (Scheme
7). On prolonged heating (45 °C, 48 h) the intermediate
allylstannane 25 is quantitatively converted into the
cyclic product 24. Presumably, the corresponding allyl-
triphenylstannane is too reactive to accumulate in solu-
tion to any appreciable degree. The acyclic intermediate
(19) See the Supporting Information for details.
(20) (a) Kainic Acid a Tool in Neurobiology; McGeer, E. G., Olney,
J . W., McGeer, P. L., Eds.; Raven Press: New York, 1978. (b)
Hollmann, M.; Heinemann, S. Cloned Glutamate Receptors. Annu. Rev.
Neurosci. 1994, 17, 31. (c) Shinozaki, H. In Excitatory Amino Acid
Receptors. Design of Agonists and Antagonists; Krogsgaard-Larsen, P.,
Hansen, J . J ., Eds.; Ellis Horwood: New York, 1992; p 261. (d) Parsons,
A. F. Recent developments in kainoid amino acid chemistry. Tetrahe-
dron 1996, 52, 4149. (e) Moloney, M. G. Excitatory Amino Acids. Nat.
Prod. Rep. 1998, 205. (f) Moloney, M. G. Excitatory Amino Acids. Nat.
Prod. Rep. 1999, 485 and references therein.
1
25 was fully characterized by H, NOE difference, 13C,
119Sn, and IR spectra and HRMS.
(18) Kant, J .; Walker, D. G. In Organic Chemistry of â-Lactams;
George, G. I., Ed.; VCH: New York, 1993; pp 121-196.
J . Org. Chem, Vol. 69, No. 21, 2004 7161

Kumareswaran et al.
section where in a related system unambiguous relative
stereochemical assignments were made on the basis of
solid-state structure derived from X-ray crystallographic
analysis. Relative juxtaposition of the allene and acety-
lene are reversed in the substrate 6, and an alternate
substitution pattern of the indolizidine (31) results upon
cyclization.
34 and 35 (eq 7). The bulk of the tert-butyldimethyl-
Indolizidines with multiple alkyl substituents, exem-
plified by indolizidine-223A,21 are a class of compounds
that has attracted some recent attention. Our approach
to the synthesis of this class of compounds which relies
upon the allenyne cyclization is shown in eq 5. The
synthesis of the starting allenyne is shown in Scheme 5.
Cyclization of allenyne 7 proceeds stereoselectively with
respect to the newly created stereogenic center (only 6-â).
The diastereomeric ratio reflects nearly the isomeric ratio
of the starting material 7 at C5. The 1:1 ratio of two
isomers 32 and 33 is evident in the 13C NMR spectrum
in which two signals were observed for each carbon atom
even though the two compounds could not be separated
by column chromatography. The confirmation of the
stereochemistry of the products came from the corre-
sponding destannylated products 34 and 35 (vide infra).
A strong cross-peak between the Ph3Sn-CHdCR2 and
the allylic hydrogens (C8) in the 2D NOSY spectrum of
32/33 clearly indicates the double-bond geometry shown.
Incidentally, the free amines 20 or 21 (Scheme 5) derived
from the lactams failed to undergo the cyclization (eq 6).
silyl group, along with the reluctance to the formation
of a primary carbonium ion, should be responsible for this
unusual result. The structures of the two products were
unambiguously established by a combination of 1D and
2D NMR studies (see the Experimental Section for
details) and X-ray crystallographic analysis. The C5-
hydrogens from the two isomers were carefully identified
1
in the H NMR with the help of the 2D-COSY spectrum.
This peak appears as a doublet of triplet (J ) 12, 4.5 Hz)
in 34 and as a doublet of doublet (J ) 8, 8 Hz) in 35. The
ratio of the two products was found to depend on the
strength of the acid used, and the reaction conditions,
suggesting some form of equilibration or selective loss of
one of the indolizidines under the reaction conditions. The
product, when recrystallized from mixture of methylene
chloride and hexane (2:5) as a solvent, gave a well-
behaved solid, still as a mixture of 34 and 35. This
mixture showed a moderately sharp melting point of 56-
58 °C. A randomly picked, suitable crystal was subjected
to X-ray crystallographic analysis. An ORTEP represen-
tation of the structure is shown in Figure 2.19 This solid
was identified as the minor component (35) of the mixture
by NMR methods. This, in combination with the exten-
sive NMR studies, corroborates the structures of the two
indolizidines 34 and 35 and, thus, the stereoselective
nature of the cyclization (i.e., exclusive formation of the
C6-â-vinyl derivative).
Lim ita tion s of th e Rea ction . F or m a tion of Acyclic
Ad d u cts. The cyclization appears to be limited to al-
lenynes with terminal acetylenes. For example, substrate
1b, an internal acetylene, failed to undergo the cycliza-
tion reaction even under forcing conditions (eq 8). While
t-BuMe2SiSnPh3 failed to react, Me3SiSnBu3 gave only
an acyclic adduct 36 under the standard conditions.
Attempts to cyclize this adduct under more forcing
conditions lead to decomposition of the material.
Selective Desta n n yla tion of 32 a n d 33 a n d Rela -
t ive Con figu r a t ion of t h e Cycliza t ion P r od u ct s.
Protodesilylation22 and protodestannylation23 of vinyl-
silanes and vinylstannanes are known to proceed in the
presence of strong acids such as trifluoroacetic acid (TFA)
and hydrofluoric acid in various organic solvents. How-
ever, all our efforts to carry out this transformation on a
mixture of 32 and 33 using various acids such as
camphorsulfonic acid (CSA), TFA, HF, and HI led only
to a selective destannylation to get a mixture of products
For some substrates, the success of the reaction also
depends on the choice of the silylstannane reagent. For
example, substrate 6, which underwent smooth cycliza-
tion with BuMe2SiSnPh3, gave only acyclic product 37
with Me3SiSnBu3 (Table 1, entries 9 and 12, and eq 9).
Under similar conditions, 5 also gave acyclic product 38
(eq 10). Likewise, the â-lactam substrates 8a and 8b gave
(21) Garraffo, H. M.; J ain, P.; Spande, T. F.; Daly, J . W. J . Nat.
Prod. 1997, 60, 2. IND 223A. See also ref 14.
(22) Protodesilylation: (a) Utimoto, K.; Kitai, M.; Nozaki, H.
Tetrahedron Lett. 1975, 2825. (b) Chan, T. H.; Fleming, I. Synthesis
1979, 761.
(23) Protodestannylation: Stork, G.; Mook, R., J r. J . Am. Chem. Soc.
1987, 109, 2829.
7162 J . Org. Chem., Vol. 69, No. 21, 2004

Synthesis of Highly Functionalized Allylstannanes
F IGURE 2. Solid-state structure and ORTEP diagram of indolizidine 35.
SCHEME 8. P ossible Mech a n ism of
Silylsta n n yla tion /Cycliza tion of a n Allen yn e
only acyclic allylstannanes 62a and 62b under the
standard reaction conditions (entry 13, and eq 11).
followed by a reductive elimination of Pd(0) from 41 with
retention of configuration. These processes result in the
exquisite stereoselectivity seen in these reactions.
Inspection of models suggests that in the cyclization
of 3, of the two transition states (leading to the trans-27
and cis-27) the one leading to the former is relatively
strain-free (Figure 3). This model assumes, that the anti-
configuration for the π-allyl-Pd-intermediate would be
more stable resulting in an anti-orientation of the vinyl-
R3Si group and the C3-substituent. Note that the favor-
able TS has a quasi-chairlike appearance while the other
one has “boatlike” features.
Mech a n ism a n d Ster eoselectivity. Details of the
mechanism of silyl-stannylation and the ensuing cycliza-
tion remains unknown. A unified mechanism that ac-
counts for the observed results is shown in Scheme 8.
Coordination of the Pd2+ on the terminal π-bond on the
sterically more accessible face (39) would lead to an anti
π-allyl Pd-complex 40, which upon reductive elimination
with formation of the less congested allylstannane and
regeneration of Pd(0) would give the primary product 25.
A more energetically demanding insertion of the acety-
lene into this π-allyl Pd complex takes place at a higher
temperature leading to the cyclic product 24 via 41.24 In
support of such a mechanism, we have observed that
isolated 25 can be converted into 24 using Pd(0) and
(C6F5)3P at elevated temperatures. Support for the oxida-
tive addition of allylstannanes comes largely from ex-
perimental observations on the allylstannane additions
to acetylenes published by Shirakawa et al.24c,d
(24) (a) A theoretical study of silylstannylation: Hada, M.; Tanaka,
Y.; Ito, M.; Murakami, M.; Amii, H.; Ito, Y.; Nakatsuji, H. J . Am. Chem.
Soc. 1994, 116, 8754. (b) For a metal-catalyzed carbocyclization by
intramolecular reactions of allylsilanes and allylstannanes with alkynes,
see: Ferna´ndez-Rivas, C.; Me´ndez, M.; Echavarren, A. M. J . Am. Chem.
Soc. 2000, 122, 1221. This reaction in which the trialkyltin moiety is
lost is mechanistically different from latter stages of the silylstannyl-
ation/cyclization. For intermolecular allylstannane additions to acety-
lenes catalyzed by Ni and Pd, see: (c) Shirakawa, E.; Yamasaki, K.;
Yoshida, H.; Nakao, Y.; Hiyama, T. J . Am. Chem. Soc. 1999, 121,
10221. (d) Shirakawa, E.; Yoshida, H.; Nakao, Y.; Hiyama, T. Org. Lett.
2000, 2, 2209. (e) A less likely alternate mechanism involving a
palladacyle cannot be ruled out at this time.
Reversibility of C-Sn bond formation is also implied
in Mitchell’s original studies on additions of stannanes
to allenes.8 The carbometalation of the Pd-allyl interme-
diate 40 is expected to be a syn-addition and this will be
J . Org. Chem, Vol. 69, No. 21, 2004 7163

Kumareswaran et al.
and n-Bu3SnH.25 Under these conditions, only hydrostan-
nylation of the allene in 7 resulted and the product was
identified as the enyne 47 (eq 13) after protodestan-
nylation by the same procedure shown in eq 12.
F IGURE 3. Relatively strain-free transition state leading to
trans-27.
C. Silylsta n n yla tion of High ly F u n ction a lized
Allen es Un d er New P r otocols. Mitchell first reported
the reactions of distannanes and silylstannanes with
allenes.3b,8 The addition of ditin reagent is apparently
reversible, and under catalysis of Pd(0), a thermodynamic
mixture of products is obtained. The more useful addition
of silylstannanes under the reaction conditions reported
by Mitchell [(1:1 allene and Me3SiSnMe3 or Me3SiSnBu3,
1 mol % Pd(PPh3)4] leads to a mixture of products, typical
examples of which are shown in eqs 14 and 15. Compared
to these reactions and the corresponding Pt-catalyzed
diboronation26 of allenes, we find that our modified
conditions give more stereoselective reactions. Since
allylstannanes are useful intermediates in their own
right we have expanded the synthesis of functionalized
allylstannanes by this route.
F IGURE 4. Favored transition states in the cyclization of 5
and 6.
The stereoselectivity in the cyclization of 5 and 6 can
also be explained on the basis of the most favorable
conformation of the putative intermediate π-allyl Pd
complexes shown in Figure 4. As in the case of the
cyclization of 3 (Figure 3), the steric interactions in the
anti-π-allyl Pd-complex undergoing the cyclization would
satisfactorily account for the observed stereochemical
result.
In the case of the cyclization of 4 (Table 1, entry 7),
such analysis does not easily explain why the stereose-
lectivity is completely lost.
B. Allen yn e Cycliza tion u sin g Oth er X-Y Re-
agen ts. Com par ison of Silylstan n ylation with Oth er
Sim ila r Rea gen ts. The silyltin reagents are generally
superior to other bis-functionalization reagents in this
type of cyclizations. For comparison, cyclization reactions
of 1a , 3, and 7 with a boron-tin reagent and ditin
reagents were carried out, and the results are shown in
Table 3. The borostannylation is a relatively poor reaction
giving 42 from 1a as the major cyclic product in no more
than 50% yield upon isolation (entry 1). J udged by “in
situ” NMR analysis, the distannylation reaction proceeds
to give a good yield of 43 (entry 2) at 45 °C, and the
corresponding acyclic adduct 44 at room temperature.
Typical results for substrates 3 and 7 are shown in
entries 4 and 5. Isolation of the Sn-Sn and Sn-B
compounds presents significant problems, and upon
chromatography severe losses are encountered. The
distannylated products are best isolated after proto-
destannylation with CSA in moist acetonitrile as shown
in eq 12 for 46.
The 4-(2-tetrahydropyranyloxy)-1,2-butadiene (48) was
prepared as shown in eq 14.27 Syntheses of additional
(25) (a) Lautens, M.; Smith, N. D.; Ostrovsky, D. J . Org. Chem. 1997,
62, 8970. (b) Lautens, M.; Mancuso, J . Org. Lett. 2000, 2, 671.
(26) Ishiyama, T.; Kitano, T.; Miyaura, N. Tetrahedron Lett. 1998,
39, 2357. For a R3SiSnR′3-mediated addition of aryl iodides and vinyl
iodides to allenes, see: Yang, F.-U.; Cheng, C.-H. J . Am. Chem. Soc.
2001, 123, 761.
(27) Preparation of 48: Brandsma, L.; Verkruijsse, H. D. Synthesis
of Acetylenes, Allenes and Cumulenes: A Laboratory Manual; Elsevier
Scientific: New York, 1981.
Recently, Lautens et al. reported cyclization of 1,6-
diynes and 1,6-enynes using catalytic amounts of Pd(0)
7164 J . Org. Chem., Vol. 69, No. 21, 2004

Synthesis of Highly Functionalized Allylstannanes
TABLE 3. Cycliza tion of Allen yn es Med ia ted by Bor on -Tin a n d Tin -Tin Rea gen ts
a
b
Key: (A) 5 mol % Pd2(dba)3 with 10 mol % P(C6F5)3; (B) 5 mol % Pd(PhCN)2Cl2 with 10 mol % P(C6F5)3. Yields determined by NMR
in parentheses. c Not isolated due to instability on column chromatography. Configuration of 45 tentative, assigned by analogy to entries
d
4-6 of Table 1. e Major side product acyclic bis-adduct (45a ). f Isolated as 47 after double-destannylation with CSA, CH3CN, rt, 4 h (see
eq 12).
allenes used in this study are shown in Scheme 9. The
N-allenyl derivative 55 was prepared from the known
propargyl compound 54,28 which is produced in situ by
treatment of the serinol derivative with excess NaH. The
sodium benzylate that is produced in the medium cata-
lyzes the isomerization of the propargyl derivative into
the allenyl derivative 55. Substrates 59 and 60 were also
prepared from serine by established chemistry via inter-
mediate 910 shown in Scheme 9.
Typical examples of silylstannylation of allenes are
shown in Table 4. The allenynes 1a ,b, 3, and 5-7 have
been discussed previously in connection with the silyl-
stannylation-cyclization reaction. These substrates give
addition to the allene without cyclization under mild
conditions. The inevitable byproducts in these reactions,
formed to varying degrees, are the cyclic products. By
careful monitoring of the reaction by NMR spectroscopy,
the reaction can be optimized to get maximum yields of
the expected allylstannanes, which are formed with
excellent regio- and stereoselectivity. As shown in the
case of 1a (Scheme 7), the reaction can also be controlled
by the use of different silylstannanes. In some instances,
different silylstannanes show dramatic difference in
reactivity and monitoring of the reaction is not needed.
For example, substrate 7, which gave an excellent yield
of the indolizidines 32 and 33 with Me2Si(t-Bu)SnPh3 (eq
8), gave only acyclic adducts (51 and 52) under a variety
of reaction conditions with Me3SiSnBu3 and Me3SiSnPh3
respectively. These adducts were conveniently identified
after destannylation with CSA and moist acetonitrile (eq
16).
Note that the silylstannylation of allenes gives clean
(E)-allylstannanes (e.g., 48-53), whereas distannylation
gives a mixture of (E)- and (Z)-products (compare eq 14
and entry 8, Table 4). Other successful substrates include
(28) Hanessian, S.; Ninkovic, S. J . Org. Chem. 1996, 61, 5418.
J . Org. Chem, Vol. 69, No. 21, 2004 7165

Kumareswaran et al.
TABLE 4. Silylsta n n yla tion of F u n ction a lized Allen es
a
b
Key: (A) 5 mol % Pd2(dba)3 with 10 mol % P(C6F5)3; (B) 5 mol % Pd(PhCN)2Cl2 with 10 mol % P(C6F5)3. Isolated yields; yields
determined by NMR in parentheses. c Isolated material contained 8% cyclic product 24 (Scheme 7). Conversion based on NMR. e Identified
d
as 63, the destannylated product (∼40%; see eq 16).
N-alleneoxazolidinone 55 and, surprisingly, 60, which
carries a reactive R,â-unsaturated ester moiety. The
silylstannylation of 60 gives a substrate 61, which is
useful for further annulation reactions. An example of a
7166 J . Org. Chem., Vol. 69, No. 21, 2004

Synthesis of Highly Functionalized Allylstannanes
SCHEME 9. Syn th esis of F u n ction a lized Allen es
(Ta ble 4)
acetylenes, or with less reactive trialkylsilyltrialkylstan-
nane reagents) to undergo the cyclization illustrates
limitations of the reaction. However, in these instances,
the intermediate allylstannanes, resulting from the
initial addition to the allenes, can be isolated. Based on
the isolation of these intermediates, and their subsequent
conversion to the cyclic products under more forcing
conditions, a mechanism for the reaction is proposed.
Model transition states to explain the stereoselectivity
in the cyclization of substituted allenynes are provided.
Further elaboration using the vinyltin and vinylsilane
moieties should lead to highly functionalized carbocyclic
and heterocyclic compounds. Examples of Sn-halogen
exhange and protodestannylation are illustrated. Under
similar conditions, addition of silylstannanes to highly
functionalized allenes give (E)-allylstannanes with high
stereoselectivity. Functional groups such as THP and
silyl ethers, lactones, â- and γ-lactams, R-â-unsaturated
esters, olefins, and substituted acetylenes are tolerated
under the reaction conditions.
Exp er im en ta l Section
novel radical cyclization to form a highly substituted
piperidine is shown in eq 17. Other similar protocols for
cyclization including Lewis acid and transition metal-
mediated processes24b can be envisioned for these highly
functionalized molecules. Nucleophilic reactions of the
E-allylstannanes with electrophiles such as carbonyl
compounds would also expand the utility of these inter-
mediates. Such studies are in progress.
Silylsta n n yla tion -Cycliza tion of Allen yn es. Synthesis
of prototypical substrates 4 and 7 and silylstannane-mediated
cyclization of these compounds will be described here. For the
details of the syntheses of other substrates 1a ,b, 2, 3, 5, 6,
20, 21, and 55, procedures for reactions with silylstannane,
distannane, and borostannane reagents, and characterization
of the products, see the Supporting Information.
The syntheses of the silylstannane3a,6 and borostannane29
reagents have been described before.
Syn th esis of (S)-3-(Bu ta -2,3-d ien yl)-4-eth yn yloxa zoli-
d in -2-on e (4) (Sch em e 2). To a solution of DMSO (0.35 mL,
4.9 mmol) in 4 mL of CH2Cl2 was added (COCl)2 (0.21 mL, 2.2
mmol) dropwise at -60 °C. After 5 min, a solution of alcohol
(S)-3-(buta-2,3-dienyl)-4-(hydroxymethyl)oxazolidin-2-one (9)10
(345 mg, 2.04 mmol) in 4 mL of CH2Cl2 was added dropwise
over 8 min. After the mixture was stirred for 15 min at -60
°C, Et3N (0.57 mL, 4.1 mmol) was added, the mixture was
warmed slowly to -15 °C over 20 min, and the solvent was
evaporated to dryness. The crude mixture of the resulting (R)-
3-(buta-2,3-dienyl)-2-oxooxazolidin-4-carbaxaldehyde 10 was
dissolved in MeOH (5 mL) at 0 °C ,and to this solution were
added Seyferth-Gilbert30 reagent (735 mg, 4.08 mmol) and
K2CO3 (564 mg, 4.08 mmol). The resulting mixture was stirred
at rt for 20 h. The reaction mixture was poured into brine (40
mL), and the aqueous layer was extracted with ether (40 mL
× 4). The combined organic layer was dried (MgSO4) and
evaporated, and the residue was subjected to flash chroma-
tography (EtOAc/Hex ) 1/2) to get 187 mg (56%) of 4 as an
oil: IR (neat, NaCl) 3288, 3065, 2984, 2920, 2118, 1957, 1748,
1479, 1416, 1372, 1217, 1183, 1084, 1024 cm-1; 1H NMR (500
MHz, CDCl3) δ 2.50 (d, J ) 2.1 Hz, H, CCH), 3.71 (dddd, J )
2.0, 3.5, 7.6, 15.2 Hz, H, NCH2), 4.18-4.24 (m, H, NCH2), 4.22-
4.28 (m, H, OCH2), 4.46 (td, J ) 1.2, 8.5 Hz, H, OCH2), 4.59
(ddd, J ) 2.0, 6.2, 8.5 Hz, H, NCH), 4.81-4.88 (m, 2H, Cd
CdCH2), 5.08-5.14 (m, H, CH)CdC); 13C NMR (125 MHz,
CDCl3) δ 41.3, 46.9, 67.4, 75.7, 77.6, 78.8, 85.6, 157.2, 209.7;
HRMS calcd for [M + Na] 186.0525, found 186.0538.
Con clu sion
Trialkylsilyltrialkylstannane (R3Si-SnR′3) reagents
undergo highly selective Pd-catalyzed additions to 1,2-
dien-7-ynes and 1,2-dien-8-ynes to give 2-vinylalkyli-
denecyclopentanes with silicon and tin substituents on
the double bonds. Similar additions of distannanes and
borostannanes show that the reactions with silylstan-
nanes are superior in terms of ease of handling of the
bifunctional reagents and the isolation of the products
after the reaction. The chemo- and regioselectivities are
controlled by the enhanced reactivity of the allene unit
vis-a`-vis the acetylene. The (Z)-geometry of the exocyclic
stannylvinylidene is a consequence of the syn-carbometa-
lation and subsequent reductive elimination from Pd with
retention of configuration at the vinyl carbon. This
method is particularly useful for the synthesis of highly
functionalized pyrrolidines and indolizidines. Reluctance
of certain kinds of allenynes (especially with internal
Syn th esis of Allen yn e 7 (Sch em e 5). (a ) P r ep a r a tion
of 1-(Hep ta -1,2-d ien -4-yl)p yr r olid in e-2, 5-d ion e 17 by a
Mitsu n obu Rea ction . In a 250 mL two-necked, round-
(29) (a) Wong, T.-T.; Busse, P. J .; Niedenzu, K. Inorg. Chem. 1970,
9, 2150. (b) Bradley, E. B.; Herber, R. H.; Busse, P. J .; Niedenzu, K. J .
Organomet. Chem. 1973, 52, 297.
(30) (a) Gilbert, J . C.; Weerasooriya, U. J . Org. Chem. 1982, 47, 1837.
(b) Muller, S.; Liepold, B.; Roth, G. J .: Bestman, H. J . Synlett 1996,
521.
J . Org. Chem, Vol. 69, No. 21, 2004 7167

Kumareswaran et al.
bottomed flask equipped with an addition funnel, stirring bar,
and N2 inlet were placed allenol 16 (44.6 mmol, 5 g), succin-
imide (44.6 mmol, 3.79 g), and PPh3 (44.6 mmol, 11.68 g) in
70 mL of THF at rt. A THF (20 mL) solution of diethyl
azodicarboxylate (44.6 mmol, 7.76 g) was added dropwise at
rt over a period of 40 min, and the resultant dark mixture
was stirred for 24 h. Analysis of the reaction by TLC indicated
a complete consumption of the allenol. The reaction mixture
was concentrated to dryness on a rotary evaporator, and the
residue was triturated with 30% EtOAc in hexanes (3 × 50
mL). The combined washings was concentrated to give a crude
product which upon flash column chromatography (15% EtOAc
in hexanes) afforded 17 (5.59 g, 65%) as a thick pale yellowish
oil: 1H NMR (400 MHz) δ 5.44 (q, J ) 7 Hz, 1H), 4.77 (dt, J
) 7, 3 Hz, 2H), 4.68-4.63 (m, 1H), 1.94-1.88 (m, 1H), 1.84-
ice bath was removed, and the reaction mixture was stirred
overnight. Saturated aq NaHCO3 was added to quench the
reaction and the mixture extracted with CH2Cl2 (3 × 30 mL).
The combined organic part was washed with brine, dried
(MgSO4), and concentrated on a rotary evaporator to get a
crude product which was purified by flash column chromatog-
raphy to obtain 7 (2.3 g, 72%) as a colorless viscous oil.
Analysis of the product by NMR indicated that it is a mixture
of two diastereomers in nearly 1:1 ratio: 1H NMR (500 MHz)
δ 5.27 and 5.24 (2q, J ) 5.5 Hz, 1H), 4.82-4.73 (m, 2H), 4.58-
4.53 and 4.37-4.33 (2m, 1H), 3.82-3.77 and 3.74-3.7 (2m,
1H), 2.58-2.34 (m, 3H), 2.3-2.13 (m, 2H), 1.99-1.93 (m, 2H),
1.73-1.57 (m, 2H), 1.37-1.2 (m, 2H), 0.89 and 0.88 (2t, J )
7.5 Hz, 3H); 13C NMR (125 MHz) δ 208.9 and 208.2, 175.7 and
175.5, 92.1 and 90.1, 80.4 and 80.3, 77.5 and 77.3, 71.2, 57.0
and 55.3, 52.2 and 49.9, 35.8 and 32.8, 30.6 and 30.3, 25.4 and
25.3, 24.5 and 24.4, 20.3 and 19.9, 14.0 and 13.98; HRMS (EI/
CI) calcd for C14H19NONa+ 240.135882, found 240.135242.
1.78 (m, 1H), 1.31-1.21 (m, 2H), 0.89 (t, J ) 7.5 Hz, 3H); 13
C
NMR (100 MHz) δ 208.6, 177.1, 89.3, 77.1, 51.2, 33.9, 28.2,
19.9, 13.7.
Cycliza t ion of 1a (Ta b le 1, E n t r y 1). (Z)-Diet h yl
3-(1-(ter t-bu tyldim eth ylsilyl)vin yl)-4-((tr iph en ylsta n n yl)-
m eth ylen e)cyclop en ta n e-1,1-d ica r boxyla te (23). To a so-
lution of Pd2(dba)3‚CHCl3 (1.3 mg, 0.0025 mmol Pd), P(C6F5)3
(2.7 mg, 0.0050 mmol), and Ph3Sn-SiMe2But (26 mg, 0.055
mmol) in 1 mL of C6D6 at rt was added allenyne 1a (12.5 mg,
0.050 mmol), and the reaction mixture was kept in an NMR
(b) Red u ction of 17 to 1-(Hep ta -1, 2-d ien -4-yl)-5-h y-
d r oxyp yr r olid in -2-on e (18). In a 100 mL two-necked, round-
bottomed flask equipped with a rubber septum, stirring bar,
and N2 inlet was placed 17 (5 mmol, 965 mg) in 30 mL of
methanol at 0 °C. Sodium borohydride (10 mmol, 380 mg) was
added as three approximate portions, and the resultant
mixture was stirred at the same temperature for 40 min.
Water (3 mL) was added to quench the reaction and concen-
trated on a rotary evaporator. The residue was extracted with
CH2Cl2 (3 × 25 mL). The combined organic part was washed
with brine, dried (MgSO4), and concentrated. Purification by
flash column chromatography (60% EtOAc in hexanes) af-
forded a thick colorless oil which was found to be the desired
18 (829 mg, 85%) as a mixture of ∼1:1 diastereomers: 1H NMR
(500 MHz) δ 5.35 and 5.2 (2q, J ) 6.5 Hz, 1H), 5.32 and 5.27
(2t, J ) 4 Hz, 1H), 4.8-4.77 (m, 2H), 4.5-4.7 (m, 1H), 3.99
and 3.79 (2d, J ) 7 Hz, 1H, exchangeable with D2O), 2.64-
2.56 (m, 1H), 2.27-2.2 (m, 2H), 1.92-1.88 (m,1H), 1.77-1.64
(m, 2H), 1.37-1.2 (m, 2H), 0.89 and 0.87 (2t, J ) 6 Hz, 3H);
13C NMR (125 MHz) δ 208.9 and 208.2, 175.5 and 175.3, 92.1
and 90.4, 82.9 and 82.2, 77.3 and 77.3, 51.2 and 50.3, 36.4 and
33.8, 29.4, 29.2 and 28.9, 20.1 and 19.8, 14.1 and 13.9.
(c) Con ver sion of 18 in to 1-(Hep ta -1, 2-d ien -4-yl)-5-
oxop yr r olid in -2-yl Aceta te (19). In a 100 mL single-necked
flask equipped with a stirring bar and a ground glass jointed
one way stopcock connected to N2 was placed 18 (20 mmol,
3.9 g) in 50 mL of CH2Cl2 at rt. Acetic anhydride (25 mmol,
2.36 mL), pyridine (25 mmol, 2.02 mL), and 50 mg of DMAP
were added successively. The reaction flask was evacuated and
refilled with N2. Analysis by TLC after stirring for 3 h
indicated a total disappearance of the starting material while
showing a nonpolar spot. The reaction mixture was diluted
with 50 mL of CH2Cl2, transferred into a separatory funnel,
and washed with water and then brine. The organic part was
dried (MgSO4), concentrated on a rotary evaporator, and dried
(0.1 mmHg at 40-50 °C for 30 min) to get a viscous oil (3.97
g). Analysis of the product by NMR indicated it to be a mixture
of 1:1 diastereomers and also pure enough to proceed for the
next step without further purification: 1H NMR (500 MHz) δ
6.28 (dd, J ) 13 Hz, 1H), 5.12 (qd, J ) 6.5, 1.5 Hz, 1H), 4.81-
4.74 (m, 2H), 4.57-4.49 (m, 1H), 2.58 (dt, J ) 16.5 Hz, 9 Hz,
1H), 2.35-2.2 (m, 2H), 2 and 1.99 (2s, 3H), 1.98-1.93 (m, 1H),
1.67-1.62 (m,1H), 1.54-1.51 (m, 1H), 1.36-1.22 (m, 2H), 0.878
and 0.875 (2t, J ) 7.5 Hz, 3H); 13C NMR (125 MHz) δ 208.9
and 208.1, 176.1 and 176.1, 170.5 and 170.4, 91.1 and 90.2,
84.4 and 82.8, 77.8 and 77.6, 50.7 and 49.7, 35.6 and 33.2, 28.9,
27.3 and 27.1, 21.6 and 21.5, 19.8 and 19.7, 14.1 and 13.9.
1
tube. Progress of the reaction was followed by H NMR. After
2 days at rt, the contents of the NMR tube was evaporated
and the residual oil was purified by flash chromatography
(Hex/Et3N ) 95/5) to get 29 mg (80%) of product as a colorless
oil. 23: 1H NMR (400 MHz, C6D6) δ -0.24 (s, 3H), -0.16 (s,
3H), 0.77 (s, 9H), 0.90 (t, J ) 7.1 Hz, 3H, CH3CH2O), 0.93 (t,
J ) 7.1 Hz, 3H, CH3CH2O), 2.30 (dd, J ) 5.5, 13.0 Hz, H, CH2-
CH), 3.16 (dd, J ) 9.1, 13.0 Hz, H, CH2CH), 3.22 (d, J ) 15.5
Hz, H, CH2CdC), 3.70-3.75 (m, 2H, CH2CH, CH2CdC), 3.90-
4.05 (m, 4H, CH3CH2O), 5.36 (t, J ) 1.6 Hz, H, SiCdCH2),
5.95 (t, J ) 1.8 Hz, SiCdCH2) together 2H, 6.40 (s, J Sn-H ) 74
Hz, 1H), 7.25-7.35 (m, 9H), 7.68-7.85 (m, 6H); Figure 1 in
the Supporting Information shows in detail all the NOE
contacts from which the configurations of the double bonds
were established; 13C NMR (100 MHz, C6D6) δ -6.1, -4.4, 13.4,
13.6, 26.9, 41.1, 46.4, 50.1, 59.0, 60.9, 61.0, 119.6, 126.8, 128.3
(J Sn-C ) 50 Hz), 128.6, 137.0 (J Sn-C ) 38 Hz), 138.9, 151.1,
163.6, 170.5, 171.1. Anal. Calcd for C38H48O4SiSn: C, 63.78;
H, 6.76. Found: C, 64.16, H, 6.74.
Scr een in g for Op tim a l Liga n d for Cycliza tion (Ta ble
2). To a solution of Pd2(dba)3‚CHCl3 (1.3 mg, 0.0025 mmol Pd),
phosphine ligand (0.0050 mmol), and Ph3Sn-SiMe2But (26 mg,
0.055 mmol) in 1 mL of C6D6 at rt was added allenyne 1a (12.5
mg, 0.050 mmol), and the reaction mixture was kept in an
1
NMR tube. Progress of the reaction was followed by H NMR.
The reaction gave a clean conversion into the cyclic product,
and no discernible side product was observed except starting
material. The conversions are based on the amount of product
and starting material left at the end of the prescribed time in
solution as estimated by NMR spectroscopy.
Scr een in g for th e P d P r ecu r sor for th e Cycliza tion
in Sch em e 7. To a solution of Pd precursor (0.0025 mmol Pd),
P(C6F5)3 (2.7 mg, 0.0050 mmol), and Ph3Sn-SiMe2But (26 mg,
0.055 mmol) in 1 mL of C6D6 at rt was added allenyne 1a (12.5
mg, 0.050 mmol), and the reaction mixture was kept in an
1
NMR tube. Progress of the reaction was followed by H NMR.
The reaction gave clean conversion into the cyclic product, and
no discernible side product was observed except starting
material.
Cycliza tion of Su bstr a te 4 (En tr y 7, Ta ble 1). To a
solution of Pd2(dba)3‚CHCl3 (2.6 mg, 0.0050 mmol Pd) and
P(C6F5)3 (5.3 mg, 0.010 mmol) in C6D6 (1 mL) were added Bu3-
Sn-SiMe3 (40.0 mg, 0.110 mmol) and the starting allenyne 4
(16.3 mg, 0.100 mmol). The progress of the reaction was moni-
tored by 1H NMR spectroscopy. After 4 h at 80 °C, the 1H NMR
spectrum indicated a mixture of the acyclic adduct and cyclic
product 29 (1:3), and after 12 h at 80 °C, the conversion was
(d ) P r op a r gyla tion of 19 to 1-(Hep ta -1,2-d ien -4-yl)-5-
(p r op -2-yn yl)p yr r olid in -2-on e (7). In a 250 mL two-necked,
round-bottomed flask equipped with a stirring bar, rubber
septum, and N2 inlet were placed 19 (15 mmol, 3.555 g) and
allenyltributylstannane (45 mmol, 14.81 g) in 60 mL of CH2-
Cl2 at 0 °C. Trimethylsilyl trifluoromethanesulfonate (37.5
mmol, 6.78 mL) was added dropwise through a syringe. The
7168 J . Org. Chem., Vol. 69, No. 21, 2004

Synthesis of Highly Functionalized Allylstannanes
TABLE 5. Silylsta n n yla tion -Cycliza tion of Su bstr a te 4
entry
Pd source
[Si-Sn]
conditions
yielda (%)
cis/trans
1
2
3
4
Pd(PhCN)2Cl2
Pd(PhCN)2Cl2
Pd2(dba)3‚CHCl3
Pd2(dba)3‚CHCl3
Ph3SnSiMe2But
Bu3SnSiMe3
80 °C/3 h
80 °C/3 h
55 °C/1 h
55 °C/1 h
0
59
(50)
76
N/A
1:1
1:1
1:1
Ph3SnSiMe2But
Bu3SnSiMe3
a
Isolated; values estimated by NMR are shown in parentheses.
complete. The ratio of cis-(E,6S,7aR)-tetrahydro-6-(1-(trimeth-
ylsilyl)vin yl)-7-((t r iph en ylst a n n yl)m et h ylen e)pyr r olo-
[1,2-c]oxazol-3(1H)-one (29-cis) and trans-(E,6R,7aR)-tetra-
hydro-6-(1-(trimethylsilyl)vinyl)-7-((triphenylstannyl)methyl-
ene)pyrrolo[1,2-c]oxazol-3(1H)-one (29-tr a n s) was 1:1 as iden-
tified by the 1H NMR spectrum of the crude mixture. The
solvent was removed, and the resulting residue was subjected
to column chromatography (Et3N/Hex ) 5/95). Careful separa-
tion of diastereomers gave 29-cis (23 mg, 44%) and 29-tr a n s
(15 mg, 29%) both as colorless oils. 29-cis: IR (neat, NaCl)
2955, 2923, 1760, 1619, 1461, 1387, 1296, 1250, 1201, 1150,
SnPh3 (5.5 mmol, 2.558 g) were added. The rubber septum was
quickly replaced with a glass stopper, and the dark mixture
was heated at 45 °C for 24 h. Analysis by TLC indicated the
completion of the reaction. The reaction mixture was concen-
trated, and the residue was purified by flash column chroma-
tography (15% EtOAc in hexanes) to afford a sticky solid (2.66
g, 78%) which was found to be the desired product as a mixture
of 32 and 33. Major isomer 32: 1H NMR (500 MHz) δ 7.56-
7.5 (m, 6H), 7.37-7.33 (m, 9H), 5.99 (t, J ) 35 Hz, 1H), 5.87
(t, J ) 1 Hz, 1H), 5.47 (d, J ) 0.5 Hz, 1H), 4.31 (ddd, J ) 16,
7.5, 3.5 Hz, 1H), 3.97-3.92 (m, 1H), 3.69 (d, J ) 7.5 Hz, 1H),
3.09 (dd, J ) 13, 7 Hz, 1H), 2.45-2.4 (m, 2H), 2.3 (dd, J ) 13,
3.5 Hz, 1H), 2.18-2.13 (m, 1H), 1.8-1.74 (m, 1H), 1.6-1.56
(m, 1H), 1.38-1.35 (m, 1H), 1.23-1.18 (m, 2H), 0.87 (t, J )
7.5 Hz, 3H), 0.6 (s, 9H), 0.16 (s, 3H), -0.51 (s, 3H); 13C NMR
(125 MHz) δ 173.6, 157.7, 149.5, 139.9, 139, 137.5, 137.3, 129,
128.9, 125.5, 54.2, 53.5, 50.5, 45.4, 31.6, 27.4, 25.9, 20.4, 17.7,
14.3, -5.2, -5.6; HRMS (EI/CI) calcd for C29H55NOSiSnNa+
604.296929, found 604.29609.
Desta n n yla tion of In d olizid in es 32 a n d 33 to 6-(1-(ter t-
Bu tyld im eth ylsilyl)vin yl)h exa h yd r o-7-m eth ylen e-5-p r o-
p ylin d olizin -3(5H)-on e 34 a n d 35 (Eq 7). In a single-necked
flask equipped with a stirring bar and a ground glass jointed
one-way stopcock attached to N2 was placed indolizidine
mixture 32 and 33 (2 mmol, 1.364 g) in 30 mL of CH3CN and
0.5 mL of water. Camphorsulfonic acid (8 mmol, 1.856 g) was
added at rt in one portion, and the mixture was stirred for 4
h to see a complete consumption of the starting material while
showing a slightly polar spot on TLC analysis. The reaction
mixture was concentrated to dryness, dissolved in 100 mL of
CH2Cl2, and transferred into a separatory funnel. Successive
washings with satd aq NaHCO3 and brine, drying (MgSO4),
and concentration gave a crude product which was purified
by flash column chromatography (20% EtOAc in hexanes) to
afford an inseparable mixture of 34 and 35 (573 mg, 86%) as
a solid: mp 56-58 °C. Indolizidine 34: 1H NMR (500 MHz) δ
5.05 (d, J ) 1.5 Hz, 1H), 4.99 (t, J ) 2 Hz, 2H), 4.91 (br s,
1H), 4.22 (dt, J ) 12, 4.5 Hz, 1H), 3.68-3.63 (m, 1H), 3.15 (br
s, 1H), 2.52 (dd, J ) 13, 4 Hz, 1H), 2.42 (ddd, J ) 12, 6, 4 Hz,
1H), 2.23-2.12 (m, 1H), 1.95 (t, J ) 13 hz, 1H), 1.66-1.58 (m,
1H), 1.5-1.44 (m, 1H), 1.4-1.22 (m, 2H), 1.2-1.12 (m, 2H),
0.89 (t, J ) 7.5 Hz, 3H), 0.84 (s, 9H), 0.06 (s, 3H), 0.05 (s, 3H);
13C NMR (125 MHz) δ 173.6, 145.6, 142.1, 132.3, 112.8, 53.7,
52.1, 50.5, 44.1, 30.0, 28.4, 27, 24.8, 19.3, 17.6, 14.4, -4.9. 35:
1H NMR (500 MHz) δ 5.91 (br s, 1H), 5.84 (br s, 2H), 5.49 (br
s, 1H), 4.38 (dd, J ) 8.5, 6.5 Hz, 1H), 3.6-3.54 (m, 1H), 3 (br
s, 1H), 2.37 (dd, J ) 10.5, 2 Hz, 1H), 2.42 (ddd, J ) 12, 6, 4
Hz, 1H), 2.23-2.2 (m, 1H), 2.15 (t, J ) 12.5 Hz, 1H), 1.66-
1.58 (m, 1H), 1.5-1.44 (m, 1H), 1.4-1.22 (m, 2H), 1.2-1.12
(m, 2H), 0.89 (t, J ) 7.5 Hz, 3H), 0.85 (s, 9H), 0.13 (s, 3H),
0.05 (s, 3H); 13C NMR (125 MHz) δ 173.8, 149.9,143.7, 127.9,
114.6, 55.7, 53.1, 50.5, 38.8, 34.6, 31, 27.4, 25.6, 19.6, 17.4,
14.3, -3.9; HRMS (EI/CI) calcd for C20 H35NOSiNa+ 356.238009,
found 356.23704.
1
1073, 1023 cm-1; H NMR (500 MHz, CDCl3): δ 0.20 (s, 9H),
SnBu3 peaks were omitted, 3.34 (dd, J ) 6.8, 11.6 Hz, H,
NCH2), 3.39 (br d J ) 6.3 Hz, H, ring CHC(Si)dCH2), 3.81 (d,
J ) 11.6 Hz, H, NCH2), 4.26 (dd, J ) 3.8, 8.6 Hz, H, OCH2),
4.39 (ddd, J ) 1.9, 3.8, 8.8 Hz, H, NCH), 4.62 (t, J ) 8.8 Hz,
H, OCH2), 5.57 (t, J ) 1.8 Hz, H, C(Si)dCH2 cis to Si), 5.71 (t,
J ) 1.9 Hz, H, C(Si)dCH2, trans to Si), 6.19 (s, J Sn-H ) 50 Hz,
H, CdCH(Sn); 13C NMR (125 MHz, CDCl3): δ -0.1, 10.4, 14.1,
27.7, 29.5, 52.7, 52.8, 63.7, 69.6, 126.6, 127.6, 152.7, 161.1,
162.5; HRMS calcd for C24H45NO2SiSn [(M + Na)+] 550.2134,
found 550.2160.
29-tr a n s: IR (neat, NaCl) 2956, 1766, 1621, 1464, 1392,
1301, 1249, 1199, 1150, 1085, 1029 cm-1; 1H NMR (500 MHz,
CDCl3) δ 0.15 (s, 9H), SnBu3 peaks were omitted, 2.93 (dd, J
) 6.2, 11.7 Hz, H, NCH2, anti), 3.56 (br t, J ) 7.6 Hz, H, ring
CHC(Si)dCH2), 4.15 (dd, J ) 9.1, 11.7 Hz, H, NCH2, syn), 4.24
(dd br, J ) 1.8, 7.6 Hz, H, NCH), 4.35 (dd, J ) 2.6, 8.7 Hz, H,
OCH2, anti), 4.56 (dd, J ) dd, J ) 7.8, 8.6 Hz, H, OCH2, syn),
5.45 (dd, J ) 1.3, 2.2 Hz, H, C(Si)dCH2 cis to Si), 5.71 (t, J )
1.9 Hz, H, C(Si)dCH2, trans to Si), 6.08 (t, J ) 1.9 Hz, J Sn-H
) 46 Hz, H, CdCH(Sn)); 13C NMR (125 MHz, CDCl3) δ -0.2,
10.5, 14.1, 27.7, 29.5, 51.3, 53.3, 63.3, 68.0, 123.7, 125.4, 153.1,
157.3, 161.9; HRMS calcd for C24H45NO2SiSn [(M + Na)+]
550.2134, found 550.2130.
The structural assignments of 29-cis and 29-tr a n s were
made by NOE experiments (see Figure 1 in the Supporting
Information for details), as well as a full characterization by
1H, 13C, COSY, HMQC, IR, and high-resolution MS.
The more polar isomer was assigned as the 29-cis struc-
ture: based on the strong Hb-Hc NOE contact, R and â OCH2
protons can be assigned as Hc and Hd, respectively. A long-
range NOE contact from He to Hd (â) would be possible because
of the concave geometry and observation of this He-Hd contact
supports 2,4-cis stereochemistry. Likewise, the less polar
component was assigned to 29-tr a n s structure: strong Hc-
Hb NOE contact indicates that R and â OCH2 protons can be
assigned as Hc and Hd, respectively. The absence of He-Hc/Hd
is a good indication that the (Si)CdCH2 substituent is on the
convex side (or pseudoequatorial) of the bicyclic system. The
observation of He-Hb contact also supports these assignments.
The cyclization reaction was carried out with different Pd
sources and Si-Sn reagents and the results are shown in Table
5.
Recrystallization of the mixture from methylene chloride
and hexane gave crystals suitable for X-ray analysis, and the
ORTEP diagram is shown in Figure 2.
Ster eoselective Silylsta n n yla tion of High ly F u n ction -
a lized Allen es. Syn th esis of Allyl Sta n n a n es. Addition of
silylstannanes under the new protocol to allenes 6 and 55 are
described. For other examples, see the Supporting Informa-
tion.
Cycliza t ion of 7 t o Isom er ic (Z)-6-(1-(ter t-Bu t yld i-
m eth ylsilyl)vin yl)h exa h yd r o-7-tr ip h en ylsta n n yl)m eth yl-
en e)-5-p r op ylin d olizin -3(5H)-on e 32 a n d 33 (Ta ble 1,
En tr y 10, Eq 5). In a two-necked, round-bottomed flask
equipped with a reflux condenser, stirring bar, rubber septum,
and N2 inlet were placed Pd2(dba)3‚CHCl3 (0.25 mmol, 259 mg)
and P(C6F5)3 (0.5 mmol, 266 mg). A solution of allenyne 7 (5
mmol, 1.085 g) in 30 mL of benzene followed by t-BuMe2Si-
J . Org. Chem, Vol. 69, No. 21, 2004 7169

Kumareswaran et al.
Ad d ition of Tr im eth ylsilyltr ibu tylsta n n a n e to 6. 5-(2-
P r op yn yl)-1-((2E)-4-t r i-n -b u t ylst a n n yl-3-t r im et h ylsilyl-
bu ten yl)-2-p yr r olid in on e (37) (Ta ble 4, En tr y 4, Eq 9).
To a solution of 1-(2,3-butadienyl)-5-(2-propynyl)-2-pyrrolidi-
none (20 mg, 0.11 mmol) in 1 mL C6D6 were added Bu3SnSiMe3
(36 µL, 0.11 mmol, 1 equiv), Pd2(dba)3 (2 mg, 0.002 mmol, 2
mol %), and P(C6F5)3 (6 mg, 0.011 mmol, 10 mol %). The
reaction was monitored by 1H NMR. After 4 h at room
temperature, the reaction was complete (100% yield by NMR).
The solvent was evaporated and the crude mixture purified
by column chromatography (SiO2: hexanes/diethyl ether 2/1)
to give 46 mg (80% isolated yield) of a colorless oil: 1H NMR
(500 MHz, CDCl3) δ 5.27 (dd, J ) 6.5, 5.5, 1 H, NCH2CH),
4.08 (dd, J ) 15.9, 5.3, 1 H, NCH2), 3.71-3.64 (m overlapping
dd, J ) 16.2, 6.1, 1 + 1 H, NCH2 + H5), 2.52 (m, 1 H, H4), 2.40
(m, 2 H, CH4CH2), 2.35 (m, 1 H, H4), 2.15 (m, 1 H, H3), 1.95
(m, 2 H, H3 + CCH), 1.86 (d, J ) 11.4, J Sn-H ) 32.9, 2 H,
SnCH2CSi), 1.63 (m, 6 H, SnCH2), 1.40 (m, 6 H, SnCH2CH2),
1.00 (m, 15 H, Sn(CH2)2CH2CH3), 0.16 (s, 9 H, SiCH3); NOE
NCH2CH f SiMe3: 2%, NCH2CH f SnCH2CSi: 0%; 13C NMR
(125 MHz, C6D6) δ 173.5, 144.8, 128.6, 79.8, 71.2, 56.2, 39.5,
29.9, 29.7, 29.6, 27.8, 23.7, 13.9, 13.6, 10.6, -1.4; 119Sn (186
MHz, CDCl3) δ -17.8; IR (film) 3312, 1695 cm-1; HRMS calcd
for C26H49NOSiSnNa M•+ ) 562.24977, found 562.24654.
Ta ble 4, En tr y 9. (E)-4-(ter t-Bu tyld im eth ylsila n yloxy-
m eth yl)-3-[2-(ter t-bu tyld im eth ylsila n yl)-4-tr ip h en ylsta n -
n a n ylp r op -2-en yl]oxa zolid in -2-on e (56). To a solution of
Pd(PhCN)2Cl2 (1.9 mg, 0.0050 mmol Pd), P(C6F5)3 (5.3 mg,
0.010 mmol), and Ph3Sn-SiMe2But (51 mg, 0.110 mmol) in 1
mL of C6D6 at rt was added allene 55 (26.9 mg, 0.100 mmol),
and the mixture was kept in an NMR tube at rt. After 8 h at
rt, the 1H NMR spectrum indicated a clean conversion into
the desired product 56. Solvent was evaporated from the
contents of the NMR tube, and the residual oil was purified
by flash chromatography (Hex/Et3N ) 95/5) to get 44 mg (60%)
of product as an off-white solid (mp 144-146 °C). 56: IR (thin
film) 3049, 2953, 2927, 2895, 2856, 2252, 1748, 1605 cm-1; 1H
NMR (500 MHz, CDCl3) δ -0.15 (s, 3H), -0.10 (s, 3H), -0.01
(s, 3H), 0.00 (s, 3H), 0.84 (s, 9H), 0.86 (s, 9H), 2.35 (dd, J Sn-H
) 90 Hz, J ) 1.8, 12.2 Hz, H, CH2SnBu3), 2.46 (d, J Sn-H ) 90
Hz, J ) 12.2 Hz, H, CH2SnBu3), 3.00 (t, J ) 8.0 Hz, H, CH2-
OTBS), 3.44 (dd, J ) 2.2, 10.3 Hz, H, OCH2CH), 3.52 (dd, J )
5.1, 10.5 Hz, H, OCH2CH), 3.82 (dd, J ) 2.9, 10.5 Hz, H, CH2-
OTBS), 3.85 (m, H, OCH2CH), 6.02 (d, J Sn-H ) 38 Hz, J ) 1.8
Hz, H, NCHdCSi), 7.32-7.42 (m, 9H), 7.45-7.62 (m, 6H); see
Figure 1 (Supporting Information) for NOE data; 13C NMR
(125 MHz, CDCl3) δ -6.2, -5.5, -5.1, 16.2, 17.8, 18.4, 26.0,
27.2, 57.0, 61.5, 65.3, 126.6, 126.8, 129.0 (J Sn-C ) 48 Hz), 129.5,
137.5 (J Sn-C ) 34 Hz), 139.0, 157.4.Anal. Calcd for C37H53NO3-
Si2Sn: C, 60.49; H, 7.27; N, 1.91. Found: C, 60.57; H, 7.19;
N, 1.88.
Ack n ow led gm en t. Financial assistance by the U.S.
National Science Foundation (CHE-0079948 and CHE-
0308378) is gratefully acknowledged.
Su p p or tin g In for m a tion Ava ila ble: Full experimental
details for the preparation of staring materials not listed in
the printed version, silylstannylation, distannylation, and
related reactions of these substrates, and spectroscopic and
chromatographic data for characterization of compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
J O049010Z
7170 J . Org. Chem., Vol. 69, No. 21, 2004