Reactivity of γ-benzyloxyallyltins with cyclohexylidene glyceraldehydes

Article abstract of DOI:10.1016/S0022-328X(00)00928-1

Benzyloxyallyltributyltins were obtained in 50-80% yield by S_N2′ reaction of alkyl-cyanocuprates with 3,3-dibenzyloxy-1-tributylstannylprop-1-ene in the presence of boron trifluoride. They reacted with cyclohexylidene glyceraldehyde in the presence of different Lewis acids and the obtained diastereomeric adducts were unambiguously identified after an ozonolysis/deprotection sequence by comparison with authentic aldopentoses. The mechanisms are briefly discussed as well as the relationship of the configuration of the reagents to the selectivity of the allylstannation reaction.

Full text of DOI:10.1016/S0022-328X(00)00928-1

Journal of Organometallic Chemistry 624 (2001) 383–387
www.elsevier.nl/locate/jorganchem
Communication
Reactivity of g-benzyloxyallyltins with cyclohexylidene
glyceraldehydes
Florian Fliegel, Isabelle Beaudet, Jean-Paul Quintard *
Laboratoire de Synthe`se Organique, UMR 6513 du CNRS, Faculte´ des Sciences et des Techniques de Nantes, 2 rue de la Houssinie`re,
BP 92208, F-44322 Nantes Cedex 3, France
Received 7 November 2000; accepted 1 December 2000
Abstract
Benzyloxyallyltributyltins were obtained in 50–80% yield by SN2% reaction of alkyl-cyanocuprates with 3,3-dibenzyloxy-1-tri-
butylstannylprop-1-ene in the presence of boron trifluoride. They reacted with cyclohexylidene glyceraldehyde in the presence of
different Lewis acids and the obtained diastereomeric adducts were unambiguously identified after an ozonolysis/deprotection
sequence by comparison with authentic aldopentoses. The mechanisms are briefly discussed as well as the relationship of the
configuration of the reagents to the selectivity of the allylstannation reaction. © 2001 Elsevier Science B.V. All rights reserved.
Keywords: b-Tributylstannylacrolein dibenzyloxyacetal; Benzyloxyallyltributyltins; Cyclohexylidene glyceraldehyde; Allylstannation; Aldopentoses
1. Introduction
When Lewis acids are able to give transmetallation of
the SnꢀC bond, g-alkoxyallylmetals (stabilized by
chelation of the metal on the oxygen) obtained from
a-alkoxyallyltins react with aldehydes through a cyclic
transition state. This is the actual trend when InCl3 is
used as Lewis acid [8]. The use of SnCl4 is expected to
give similar results on the basis of studies on allyltins
bearing a remote alkoxy functionality [9] on the condi-
tion that allylstannation occurs before destruction of
the reagents [10].
The use of TiCl4 may involve both of these processes
depending on the order of addition of the reagents [11].
Nonetheless in the oxygenated series, the nature of the
functional group is of importance because examples of
destruction of the reagents have been mentioned [10] as
well as expected allyltitanation [12].
In the present paper, we wish to report recent results
obtained in the g-alkoxyallyltin series using a Lewis
acid unable to give the transmetallation reaction. The
early studies in these series achieved with non function-
alized crotyltins have shown a high syn selectivity what-
ever the geometry of the crotyltin [13] and a similar
trend was observed for g-alkoxyallyltins [1,2].
The addition of a- and g-alkoxyallyltins to aldehydes
has been extensively studied in recent years [1,2]. While
a-alkoxycrotyltins have been shown to react with alde-
hydes upon heating, this type of reactivity has been
scarcely used because of the lack of reactivity of the
Z-isomer due to the higher energy of the cyclic chair-
like transition state when compared to the E-isomer
[3–5]. Attempts of activation using Lewis acids can
lead to different possibilities depending on the nature of
the Lewis acid.
In the case of a Lewis acid unable to transmetallate
the SnꢀCallyl bond (for instance BF3), the addition
occurs through an opened transition state where the
reacting species are often g-alkoxyallyltins whatever the
position of the alkoxy group on the allyl unit since
a-alkoxyallyltins isomerize very easily, in a stereospe-
cific fashion, into g-alkoxyallyltins under these experi-
mental conditions [6,7].
* Corresponding author. Tel.: +33-2-5112-5408; fax: +33-2-5112-
5412.
E-mail address: quintard@chimie.univ-nantes.fr (J.-P. Quintard).
0022-328X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(00)00928-1

384
F. Fliegel et al. / Journal of Organometallic Chemistry 624 (2001) 383–387
siloxyallyltributyltins were prepared from b-tributyl-
The syn selectivity was initially explained by an an-
stannylacrolein acetals [21] according to Scheme 1.
Allyltins 3a and 3b were obtained easily as clean Z
isomers according to our previously described proce-
dure [20] while 5a and 5b were obtained via conjugate
addition of higher order lithium alkylcyanocuprates to
b-tributylstannylacrolein 4.
tiperiplanar transition state [13]. The possibility of a
synclinal transition state was shown in an intramolecu-
lar allylstannation [14]. Subsequently the balance be-
tween antiperiplanar and synclinal transition states has
been shown to be very subtle and dependent on sec-
ondary effects such as an inside alkoxy effect [15,16],
the geometry of the double bond [17] or steric hin-
drance of the a substituent or of the b substituent when
a- or b-substituted g-alkoxyallyltins are involved
[18,19].
Due to the easy preparation of differently a-substi-
tuted g-ethoxyallyltins by reaction of RCu(CN)MgX
with b-tributylstannylacrolein acetals [20], we have been
able to test this parameter and to obtain reversed
selectivities as a function of the size of the a-substituent
for reactions on benzaldehyde or 2-furaldehyde [18].
The g-benzyloxyallyltins seem the more interesting
reagents because they could give syn or anti adducts
with aldehydes depending on the size of the R group.
With g-siloxy derivatives, syn selectivity is obtained
even in the presence of a-bulky groups (for instance
R=SnBu3) [25]. Indeed, 3b has been shown to react
with benzaldehyde to give the anti adduct (98% yield,
syn/anti=2/98).
In order to evaluate the validity of the above assump-
tions and to prepare precursors of aldopentoses, 3a and
3b were reacted with (R)-cyclohexylidene glyceralde-
hyde (which has been recently used for the synthesis of
2-C branched 2-deoxypentofuranoses [26]) in the pres-
ence of boron trifluoride etherate (Scheme 2).
With BF3·Et2O (3eq) as Lewis acid (at −78°C in
CH2Cl2), 3a affords a mixture 39/37/24 of 7, 8, 9 in
48% yield while 3b affords a 49/41/6/4 mixture of 7, 8,
9 and 10 in 98% yield (these values were obtained on
the basis of HPLC analysis assuming that the four
diastereomers have similar m values for the benzyloxy
group).
The syn/anti ratios (93/7 for R=Me and 3/97 for
R=t-Bu) were explained by an antiperiplanar transi-
tion state in the first case (R=Me) and by a synclinal
transition state in the second case (R=t-Bu).
On the basis of this preliminary study it seems possi-
ble to obtain aldopentoses from glyceraldehyde and
appropriately substituted g-oxygenated allyltins (the
alkoxy groups being easily removable groups like BnO
and R3SiO).
The identification of the diastereomers 7–10 has been
made after separation by liquid chromatography and
further transformation into aldopentoses according to
the Scheme 3.
Accordingly, g-benzyloxyallyltributyltins and g-
Scheme 1.
Scheme 2.

F. Fliegel et al. / Journal of Organometallic Chemistry 624 (2001) 383–387
385
Each diastereomer (7–10) was transformed according
to this scheme and identified by comparison with au-
thentic samples of aldopentoses (comparison of the
1H-NMR spectra). The following correlation can be
established:
while MgBr2 (reaction at −20°C, in CH2Cl2, 15 h)
induces a preference for the anti diastereomer 7+8/9+
10=24/76 for R=t-Bu (30% yield). It is worth noting
that side products were obtained in both cases.
The second test which has been conducted was a
comparative study between (R) and (S) cyclohexylidene
glyceraldehyde (R)-6 or (S)-6 which were allowed to
react with racemic 3b under conditions in which start-
ing materials were recovered (3b/6=1/1, BF3·Et2O=
1.2 eq, CH2Cl2, −78°C, 2 h) (see Table 1). After flash
chromatography, the recovered organotin exhibits an
uncorrected ‘[h]D’ value of −27.9 (C6H6, c=1.4) start-
ing from (R)-6 and +32.8 starting from (S)-6. On the
basis of the known [h]D values for pure a-n-alkyl-g-
alkoxyallyltins (between +70 and +120 for the (S)
enantiomers [6,20]), taking into account the similarity
of polarisabilities of the bonds around the asymmetric
centre, enrichment of the (S)-allyltin seems likely in the
first case (‘[h]D’= −27.9) and of the (R)-allyltin in the
second case (‘[h]D’= +32.8) since priority of the sub-
stituents around the chiral centre is modified (n-BuBg-
alkoxyvinylBt-Bu).
(89%)
Diastereoisomer 7ꢀꢀꢀꢀꢀꢁarabinose (syn-anti )
(63%)
Diastereoisomer 8ꢀꢀꢀꢀꢀꢁxylose (syn-syn)
(70%)
Diastereoisomer 9ꢀꢀꢀꢀꢀꢁribose (anti-anti )
(75%)
Diastereoisomer 10ꢀꢀꢀꢀꢁlyxose (anti-syn)
It is worth noting that the correlation sequence must
be done from purified diastereomers since significant
difference in reactivity can be observed between two
diastereomers (for instance 7 and 8).
Obviously the size of the R group has not the ex-
pected effect on the diastereoselection since syn selectiv-
ity appears to be the major addition mode in both cases
with a higher preference for R=t-Bu (7+8/9+10=
90/10) when compared with R=Me (77/23 for the
similar ratio)!
This result is in disagreament with the trends previ-
ously observed with benzaldehydes [18]. In order to
obtain more information about the reaction mecha-
nisms, two reactions have been conducted in the pres-
ence of Lewis acids which can be doubly coordinated
(cf. Table 1). With ZnCl2, 7 remains the major product
For a-t-butyl-g-benzyloxyallyltributyltin, the (R)
enantiomer is more reactive than the (S) enantiomer
with (R)-cyclohexylidene glyceraldehyde; this result is
corroborated by a reverse trend with (S)-cyclohexyli-
dene glyceraldehyde.
Scheme 3.
Table 1
Reaction of g-benzyloxyallyltin 3b on cyclohexylidene glyceraldehydes (R)-6 and (S)-6
a
d
Aldehyde
Experimental conditions b
Adducts
Remaining allyltin
Overall yield
7
8
9
10
[h]D
(R)-6 (3eq)
(R)-6 (1.5eq)
(R)-6 (1.5eq)
(R)-6 (1eq)
(S)-6 (1eq)
BF3·Et2O (3eq); −78°C
MgBr2 (1.5eq); −20°C
ZnCl2 (2eq); −25°C to rt
BF3·Et2O; (1.2eq)c
98
30
24
36
45
49
6
66
72
41
18
15
24
6
10
4
66
1
4
n.d.
18
n.d.
0 e
−27.9 f
f
BF3·Et2O; (1.2eq) c
65 (ent-7)
28 (ent-8)
0 e (ent-9)
7 (ent-10)
+32.8
a (R)-6 and (S)-6 were obtained according to litterature [32].
b Experimental conditions: in a Schlenk reactor containing 6 in dry degased CH2Cl2 (10 ml) were added Lewis acid (xeq) at the
above-mentioned temperature before addition of 3b (0.4 mmol, 1eq) in CH2Cl2 (2 ml). The obtained mixture is allowed to react at the desired
temperature until complete disappearance of 6 (TLC monitoring). After hydrolysis (NaHCO3) and usual treatments, compounds 7–10 were
purified by flash chromatography on silica gel using hexane/ether=85/15 [33].
c Same procedure as (b) but the mixture is allowed to react only for 2 h at −78°C in order to have an incomplete reaction.
d HPLC separation of 7–10 was performed on hypersil (5 m, 25×0.45 cm, eluent=ether/hexane=15/85, 1 ml min−1). The retention times were
the following ones: 7 (10.6 min), 8 (9.8 min), 9 (13.3 min), 10 (7.4 min).
e Due to its longer retention time, a small amount of 9 (ca. 3%) might be omitted.
f These values were determined on crude remaining allyltin and therefore cannot be considered as specific optical rotation because remaining
allytin is polluted with non-chiral organotins.

386
F. Fliegel et al. / Journal of Organometallic Chemistry 624 (2001) 383–387
Scheme 4.
2. Comments
Acknowledgements
Though not complete enough to allow a detailed
discussion, these results offer some interesting insight
on stereochemical trends of this allylstannation. The
stereochemical results appear to be consistent with
Marshall’s reports concerning the addition of enan-
tioenriched g-alkoxyallyltins to chiral a-alkoxy alde-
hydes [2] with matched and mismatched effects [27].
In the present case, among the possible transition
states, compound 7 is believed to come from an an-
tiperiplanar transition state AP1 (reaction of (R)-3b on
(R)-6) while compound 8 and 9 might be due mainly to
the addition of (S)-3b to (R)-6 through an antiperipla-
nar transition state AP2 or a synclinal transition state
(which is less disfavoured when R=Me instead of
R=t-Bu because of the difference in steric hindrance)
(Scheme 4).
Similar effects can also account for formation of
ent-7 to ent-10 when (S)-6 reacts with 3b in the presence
of boron trifluoride.
However, these explanations are unable to rationalize
the observed stereochemistry when bidentate Lewis
acids like ZnCl2 or MgBr2 are used. In these cases, the
situation appears to be much more complicated and
probably involves chelations both on the a-alkoxy and
on the b-alkoxy groups of the aldehyde [28].
The authors are grateful to Chemetall g.m.b.h., Con-
tinentale Parker, Schering and Rhodia-Silicones Com-
panies for the gift of organometallic starting materials,
to AGISMED and to the ‘Conseil Ge´ne´ral de Loire
Atlantique’ for financial support (F.F.).
References
[1] Y. Yamamoto, N. Asao, Chem. Rev. 93 (1993) 2207.
[2] J.A. Marshall, Chem. Rev. 96 (1996) 31.
[3] Y. Yamamoto, N. Shida, Advances in Detailed Reaction Mecha-
nisms, vol. 3, JAI Press, Greenwich, CT, 1994, pp. 1–44.
[4] (a) A.J. Pratt, E.J. Thomas, J. Chem. Soc. Perkin Trans I (1989)
1521. (b) V.J. Jephcote, A.J. Pratt, E.J. Thomas, J. Chem. Soc.
Perkin Trans I (1989) 1529.
[5] J.-P. Quintard, G. Dumartin, B. Elissondo, A. Rahm, M.
Pereyre, Tetrahedron 45 (1989) 1017.
[6] J.A. Marshall, G.S. Welmaker, B.W. Gung, J. Am. Chem. Soc
113 (1991) 647.
[7] Even scarcely observed, examples of reaction of a-alkoxyallyltins
with aldehydes without previous rearrangement into g-alkoxyal-
lyltins have been reported. Cf. for instance: J.A. Marshall, W.Y.
Gung, Tetrahedron 45 (1989) 1043 or J.A. Marshall, D.V.
Yashunsky, J. Org. Chem. 56 (1991) 5493.
[8] (a) J.A. Marshall, K.W. Hinkle, J. Org. Chem. 61 (1996) 105. (b)
J.A. Marshall, A.W. Garofalo, J. Org. Chem. 61 (1996) 8732.
[9] E.J. Thomas, Chem. Commun. (1997) 411.
[10] J.A. Marshall, K.W. Hinkle, J. Org. Chem. 60 (1995) 1920.
[11] G.E. Keck, D.E. Abbott, E.P. Boden, E.J. Enholm, Tetrahedron
Lett. 25 (1984) 3927.
[12] H. Paulsen, C. Graeve, R. Fro¨hlich, D. Hoppe, Synthesis (1996)
145.
3. Conclusion
[13] (a) Y. Yamamoto, H. Yatagai, Y. Naruta, K. Maruyama, J.
Am. Chem. Soc. 102 (1980) 7107. (b) Y. Yamamoto, H. Yatagai,
Y. Ishihara, N. Maeda, K. Maruyama, Tetrahedron 40 (1984)
2239.
[14] (a) S.E. Denmark, E.J. Weber, J. Am. Chem. Soc. 106 (1984)
7970. (b) S.E. Denmark, E.J. Weber, T.M. Wilson, T.M. Will-
son, Tetrahedron 45 (1989) 1053.
[15] B.W. Gung, A.J. Peat, B.M. Snook, D.T. Smith, Tetrahedron
Lett. 32 (1991) 453.
[16] B. Muller, J.-P. Fe´re´zou, A. Pancrazi, J.-Y. Lallemand, Bull.
Soc. Chim. Fr. 134 (1997) 13.
[17] G.E. Keck, S.M. Dougherty, K.A. Savin, J. Am. Chem. Soc. 117
(1995) 6210.
[18] S. Watrelot-Bourdeau, J.L. Parrain, J.-P. Quintard, J. Org.
Chem. 62 (1997) 8261.
[19] Y. Nishigaichi, N. Ishida, M. Nishida, A. Takuwa, Tetrahedron
Lett. 37 (1996) 3701.
[20] S. Watrelot, J.-L. Parrain, J.-P. Quintard, J. Org. Chem. 59
(1994) 7959.
In spite of their limited application for the prepara-
tion of the aldopentoses, the above results demonstrate
that the diastereoselectivity of allylstannanes additions
observed with aromatic aldehydes cannot be applied to
glyceraldehyde derivatives.
Furthermore these preliminary results suggest that
significant improvements can be expected in terms of
selectivity using g-alkoxyallyltins with appropriate
configuration. For this purpose the previously described
enantioselective synthesis of g-alkoxyallyltins [20] or
g-aminoallyltins [29] through SN2% opening of b-trib-
utylstannylacrolein derivatives can bring tools of inter-
est to prepare differently protected sugars or azasugars,
the allylstannation being possible on a-alkoxy or a-
amino aldehydes [2] as well as aldimines [9,30,31].

F. Fliegel et al. / Journal of Organometallic Chemistry 624 (2001) 383–387
387
[21] The strategy used for the synthesis of (E)-3,3-diethoxy-1-tributyl-
stannylprop-1-ene which has been previously described in our
group [22,23] was applied successfully to the synthesis of diben-
zyl acetal 2. b-Tributylstannylacrolein acetals are easily hy-
drolyzed on wet silica gel to afford b-tributylstannylacrolein
which has been shown to react easily with higher order stannyl-
cyanocuprates [24].
[22] (a) I. Beaudet, J.-L. Parrain, J.-P. Quintard, Tetrahedron Lett.
32 (1991) 6333. (b) I. Marek, A. Alexakis, J.F. Normant, Tetra-
hedron Lett. 32 (1991) 6337.
[23] F. Suzenet, J.-L. Parrain, J.-P. Quintard, Eur. J. Org. Chem.
(1999) 2957.
[24] D. Madec, J.-P. Fe´re´zou, Tetrahedron Lett. 38 (1997) 6657.
[25] D. Madec, J.-P. Fe´re´zou, Tetrahedron Lett. 38 (1997) 6661.
[26] A. Chattopadhyay, B. Dhotare, S. Hassarajani, J. Org. Chem. 64
(1999) 6874.
[27] S. Masamune, W. Choy, J.S. Petersen, L.R. Sita, Angew. Chem.
Int. Ed. Engl. 24 (1985) 1.
[28] J. Jurczack, S. Pikul, T. Bauer, Tetrahedron 42 (1986) 447.
[29] F. Suzenet, E. Blart, M. Destouches, J.-P. Quintard, J.
Organomet. Chem. 567 (1998) 21.
[30] D.-K. Wang, L.-X. Dai, X.-L. Hou, Tetrahedron Lett. 36 (1995)
8649.
[31] H. Nakamura, H. Iwama, Y. Yamamoto, Chem. Commun.
(1996) 1459.
57); 29.3 (3C, 2JSnC=13); 30.6 (3C, 3JSnC=26); 33.8 (1C,
2
2JSnC=13); 39.9 (1C, 1JSnC=297/312); 73.4; 109.5 (1C, JSnC
=
3
38); 127.3–128.3 (5C); 138.1; 141.1 (1C, JSnC=48); 119Sn-NMR
l (ppm): −29.0. Found: C, 63.23; H, 9.58. C26H46OSn (492.26)
requires C, 63.30; H, 9.40.
5a: see [34].
5b: E isomer: 1H-NMR l (ppm): 0.1 (s, 6H); 0.7–1.0 (m, 24H);
2
0.9 (s, 9H); 1.15–1.6 (m, 12H); 1.86 (d, 1H, 3J=12.6, JSnH
=
60.1); 5.09 (dd, 1H, 3J=12.6, 3J=11.6, 2JSnH=24.1); 6.08 (d,
1H, 3J=11.6, 2JSnH=20.6); 13C-NMR l (ppm): −5.1 (2C);
10.5 (3C, 2JSnH=280/301); 13.7 (3C); 18.2; 25.7 (3C); 27.6 (3C,
2JSnH=60); 29.3 (3C, 2JSnH=18); 30.7 (3C, 2JSnH=20); 33.8;
42.5; 112.9 (1C, 2JSnH=36); 137.5.
7: 1H-NMR l (ppm): 1.05 (s, 9H); 1.3–1.65 (m, 10H); 2.5 (d,
OH, 3J=6.3); 3.57 (ddd, 1H, 3J=3.9, 3J=6.3, 3J=6.4); 3.88
(dd, 1H, 3J=3.9, 3J=8.4); 3.95 (dd, 1H, 3J=6.3, 2J= −8.3);
4.01 (dd, 1H, 3J=6.2, 2J= −8.3); 4.11 (ddd, 1H, 3J=6.2,
2
3J=6.3, 3J=6.4; 4.38 and 4.60 (syst. AB, 2H, J= −10.8); 5.43
(dd, 1H, 3J=8.4, 3J=15.8); 5.73 (d, 1H, 3J=15.8); 7.25–7.40
(m, 5H); 13C-NMR l (ppm): 23.7–36.3 (5C); 29.4 (C(C
6
H3)3);
(CH3)3); 66.0; 69.8 (Bn); 75.0; 75.3; 79.6; 109.4; 121.5;
127.5–128.2 (5C); 138.6; 147.1.
33.1 (C
6
8: 1H-NMR l (ppm): 1.05 (s, 9H); 1.2–1.8 (m, 10H); 2.60 (d,
OH, 3J=5.3); 3.53 (ddd, 1H, 3J=4.3, 3J=5.3, 3J=5.6); 3.76
(dd, 1H, 3J=7.6, 2J= −7.8); 3.78 (dd, 1H, 3J=5.6, 3J=8.7);
3.90 (dd, 1H, 3J=6.3, 2J= −7.8); 4.17 (ddd, 1H, 3J=4.3,
3J=6.3, 3J=7.6); 4.34 and 4.60 (syst. AB, 2H, 2J= −11.9);
5.32 (dd, 1H, 3J=8.7, 3J=15.9); 5.77 (d, 1H, 3J=15.9); 7.25–
[32] (a) J.A.J.M. Vekemans, J. Boerekamp, E.F. Godefroi, Recl.
Trav. Chim. Pays-Bas 104 (1985) 266. (b) D.J. Jackson, Synth.
Comm. 18 (1988) 337. (c) M. Daumas, Y. Vo-Quang, L. Vo-
Quang, F. Le Goffic, Synthesis (1989) 64. (d) C.R. Schmidt,
D.A. Bradley, Synthesis (1992) 587.
7.40 (m, 5H); 13C-NMR
(C(CH3)3); 33.2 (C(CH3)3); 65.8; 69.9 (Bn); 73.0; 75.0; 81.5;
l (ppm): 23.8–36.0 (5C); 29.4
6
6
[33] Meaningful NMR data (in CDCl3) for new compounds: 2:
1H-NMR l (ppm): 0.8–1.6 (m, 27H); 4.57 and 4.69 (syst. AB,
4H, 2J= −11.9); 5.08 (dd, 1H, 3J=4.3, 4J=1.1, 4JSnH=6.7);
6.08 (dd, 1H, 3J=19.3, 3J=4.3, 3JSnH=63); 6.49 (dd, 1H,
3J=19.3, 4J=1.1, 2JSnH=70); 7.2–7.4 (m, 10H); 13C-NMR l
(ppm): 9.3 (3C, 1JSnC=330/346); 13.5 (3C); 27.1 (3C, 3JSnC=53/
56); 28.9 (3C, 2JSnC=21); 67 (2C); 101.7 (1C, 3JSnC=66/69);
127.4 (2C); 127.6 (4C); 128.2 (4C); 134 (1C, 1JSnC=356); 138.1
109.6; 121.1; 127.5–128.3 (5C); 138.3; 148.2.
9: 1H-NMR l (ppm): 1.05 (s, 9H); 1.3–1.65 (m, 10H); 2.38 (s,
OH); 3.85 (dd, 1H, 3J=3.7, 3J=6.3); 3.91 (dd, 1H, 3J=3.7,
3J=8.4); 3.92 (dd, 1H, 3J=6.6, 2J= −8.2); 3.96 (dd, 1H,
3J=6.1, 2J= −8.2); 4.04 (ddd, 1H, 3J=6.1, 3J=6.3, 3J=6.6);
4.37 and 4.59 (syst. AB, 2H, 2J= −11.9); 5.42 (dd, 1H, 3J=8.4,
3
3J=15.8); 5.78 (d, 1H, J=15.8); 7.25–7.40 (m, 5H); 13C-NMR
(2C); 143.8 (1C, 2JSnC=6); 119Sn-NMR
l (ppm): −46.9.
l (ppm): 23.8–36.3 (5C); 29.5 (C(C6 H3)3); 33.3 (C6 (CH3)3); 65.9;
69.8 (Bn); 74.1; 75.0; 80.9; 109.2; 122.9; 127.5–128.3 (5C); 138.3;
148.2.
Found: C, 64.29; H, 8.11. C29H44O2Sn (542.24) requires C,
64.10; H, 8.16.
10: 1H-NMR l (ppm): 1.05 (s, 9H); 1.2–1.7 (m, 10H); 2.30 (d,
OH, 3J=5.2); 3.53 (ddd, 1H, 3J=5.2, 3J=5.2, 3J=6.4); 3.69
(dd, 1H, 3J=6.4, 3J=8.1); 3.77 (dd, 1H, 3J=7.3, 2J= −8.1);
3.98 (dd, 1H, 3J=6.4, 2J= −8.1); 4.22 (ddd, 1H, 3J=5.2,
3J=6.4, 3J=7.3); 4.34 and 4.58 (syst. AB, 2H, 2J= −11.9);
5.36 (dd, 1H, 3J=8.1, 3J=15.9); 5.73 (d, 1H, 3J=15.9); 7.20–
3a: 1H-NMR l (ppm): 0.7–1.7 (m, 30H); 2.52 (ddq, 1H, 3J=7.6,
3J=10.8, 4J=0.9, 2JSnH=60); 4.45 (dd, 1H, 3J=10.8, 3J=6.0,
3JSnH=20); 4.7 and 4.77 (syst. AB, 2H, 2J= −12.7); 5.85 (dd,
3
4
4
1H, J=6.0, J=0.9, JSnH=20); 7.2–7.4 (m, 5H); 13C-NMR l
(ppm): 8.7 (3C, 1JSnC=298/285); 13.7 (3C); 16.6; 18.7; 27.5 (3C,
3JSnC=54); 29.3 (3C, 2JSnC=20); 73.5; 114.6; 127.4–128.3 (5C);
138.0; 139.4; 119Sn-NMR l (ppm): −15.7.
7.40 (m, 5H); 13C-NMR
(C(CH3)3); 33.3 (C(CH3)3); 66.0; 69.8 (Bn); 73.8; 75.7; 81.0;
l (ppm): 22.6–36.1 (5C); 29.5
3b: 1H-NMR l (ppm): 0.7–1.7 (m, 36H); 2.65 (dd, 1H, 3J=12.4,
6
6
3
4J=0.6, 2JSnH=64); 4.52 (dd, 1H, 3J=12.4, 3J=6.2, JSnH
=
109.5; 121.5; 127.5–128.3 (5C); 138.2; 148.1.
7–10: Found: C, 74.17; H, 9.24. C23H34O4 (374.25) requires C,
73.76; H, 9.15.
22/27); 4.71 and 4.79 (syst. AB, 2H, 2J= −12.5); 5.92 (dd, 1H,
3J=6.2, 4J=0.6, 4JSnH=19); 7.2–7.4 (m, 5H); 13C-NMR l
3
(ppm): 10.7 (3C, 1JSnC=283/295); 13.7 (3C); 27.6 (3C, JSnC
=
[34] J.A. Marshall, G.S. Welmaker, J. Org. Chem. 57 (1992) 7158.
.