1,2-Asymmetric induction in the radical addition of organotin hydrides to (-)-menthyl(E)-2,3-disubstituted propenoates

Article abstract of DOI:10.1016/S0022-328X(97)00637-2

The results obtained in the free radical hydrostannation of (-)-menthyl(E)-2,3-diphenylpropenoate (1) with tri-n-butyl- and triphenyltin hydride, and of (-)-menthyl(E)-2-phenyl-2-butenoate (7) with trimethyltin hydride are reported. The absolute configuration of the new organotin adducts was determined by combining ¹H- and ¹³C-NMR data with chemical correlation. The additions took place in all cases following a syn stereochemistry that led to diastereomeric excesses ranging between 73 and 100%. The observed stereochemistry is explained, taking into account both the allylic strain and the hyperconjugation with β-trialkyltin substituent existing in the intermediate radicals. Full ¹H-, ¹³C- and ¹⁹Sn-NMR data are given.

Full text of DOI:10.1016/S0022-328X(97)00637-2

Journal of Organometallic Chemistry 555 (1998) 151–159
1,2-Asymmetric induction in the radical addition of organotin hydrides
to (−)-menthyl(E)-2,3-disubstituted propenoates1
1,
S.D. Mandolesi, L.C. Koll 2, A.B. Chopa 3, J.C. Podesta´ *
Instituto de In6estigaciones en Qu´ımica Orga´nica, Departamento de Qu´ımica e Ing. Qu´ımica, Uni6ersidad Nacional del Sur, A6enida Alem 1253,
Bah´ıa Blanca 8000, Argentina
Received 4 June 1997; received in revised form 9 September 1997
Abstract
The results obtained in the free radical hydrostannation of (−)-menthyl(E)-2,3-diphenylpropenoate (1) with tri-n-butyl- and
triphenyltin hydride, and of (−)-menthyl(E)-2-phenyl-2-butenoate (7) with trimethyltin hydride are reported. The absolute
1
configuration of the new organotin adducts was determined by combining H- and 13C-NMR data with chemical correlation. The
additions took place in all cases following a syn stereochemistry that led to diastereomeric excesses ranging between 73 and 100%.
The observed stereochemistry is explained, taking into account both the allylic strain and the hyperconjugation with i-trialkyltin
1
substituent existing in the intermediate radicals. Full H-, 13C- and 19Sn-NMR data are given. © 1998 Elsevier Science S.A. All
rights reserved.
Keywords: 1,2-Asymmetric induction; Organotin hydrides; (−)-Menthyl(E)-2,3-disubstituted propenoates
propenoate (1) and of trimethyltin hydride to (−)-
menthyl(E)-2-phenyl-2-butenoate (2). These studies
were carried out with the aim of determining whether
both the olefin substituents and the type and size of the
organic ligands attached to the tin atom have any effect
on the stereoselectivity of these reactions.
1. Introduction
The addition of organotin hydrides to prochiral acti-
vated alkenes [1] has been included among the early
examples of reactions that showed that 1,2-asymmetric
induction was possible in radical reactions ([2]a). We
have recently reported [3] a study on the addition of
trimethyltin hydride to (−)-menthyl(E)-2,3-diphenyl-
propenoate, which confirmed the stereoselectivity of
these free radical reactions and also enabled us to
determine the absolute configuration of the organotin
adducts.
2. Results and discussion
The addition under free radical conditions of tri-n-
butyltin hydride to (−)-menthyl(E)-2,3-diphenyl-
In the present paper, we report the results obtained
in the free radical additions of tri-n-butyl- and
triphenyltin hydrides to (−)-menthyl(E)-2,3-diphenyl-
propenoate (1) leads to
a mixture of the four
diastereoisomers expected, as shown in Scheme 1. The
analysis by 119Sn-NMR of the crude product shows it
to consists of a mixture of four diastereoisomers, two of
which are in higher proportion (41 and 48%). On the
other hand, the addition of triphenyltin hydride to the
same olefin (Scheme 1) leads to a mixture of only two
diastereoisomers (31 and 69%). Similarly, hydrostanna-
tion of (−)-menthyl(E)-2-phenyl-2-butenoate (2)
* Corresponding author. Tel.: +54 91 26420/280345; fax: +54 91
883933; e-mail: jpodesta@criba.edu.ar
1 Dedicated to the memory of F. Gordon Thorpe (Department of
Chemistry, Lancaster University, UK).
2 Member of CONICET, Capital Federal, Argentina.
3 Member of CIC, Provincia de Buenos Aires, Argentina.
0022-328X/98/$19.00 © 1998 Elsevier Science S.A. All rights reserved.
PII S0022-328X(97)00637-2

152
S.D. Mandolesi et al. / Journal of Organometallic Chemistry 555 (1998) 151–159
Scheme 1. Addition of triorganotin hydrides to (−)-menthyl(E)-2,3-diphenylpropenoate (1) and (−)-menthyl(E)-2-phenyl-2-butenoate (2).
3
(Scheme 1) with trimethyltin hydride gives a mixture of
four diastereoisomers (according to the 119Sn-NMR
spectrum: 43.4, 43.4, 5.6 and 7.6%).
angles close to 60°. Similarly, the values of J(Sn–C–
C–Ph) coupling constants for compounds 3, 4 and 6–9
within the range 35.0–52.9 Hz (Table 1, C-2%) indicate
a dihedral angle of about 180° between the trialkylstan-
nyl moiety and the phenyl group attached to C-2.
The 119Sn-NMR analysis of the crude products ob-
tained in the addition of tri-n-butyltin- and triphenyltin
hydrides to the ester 2 also shows these products to
consist of mixtures of four tributyltin adducts (40.3,
40.3, 11.3 and 8.1%) and only two triphenyltin adducts
(76 and 24%), respectively.
1
The H-NMR spectra (Table 2) of compounds 3, 4
and 6–9 show that the J(H, H) coupling constants for
3
the protons attached to C-2 and C-3 lie between 8.2
and 12.5 Hz, indicating that these protons are an-
tiperiplanar. The 3J(Sn–C–C–H) coupling constants
for compounds 3, 4 and 6–9 with values ranging from
25.2 to 52.0 Hz suggest a dihedral angle close to 60°.
Taking into account all these values, it is possible to
attribute a threo configuration (Fig. 1, I) to all these
diastereoisomers, i.e. (2R, 3R)- and (2S, 3S)- for com-
pounds 3, 4, 6 and 7, and (2R, 3S)- and (2S, 3R) for
compounds 8 and 9.
Although separation of all these diastereomers by
column chromatography (silica gel 60) is not entirely
feasible, this method enabled us to separate the
stereoisomers obtained in higher yield (3, 4, 8 and 9,
Scheme 1) from those obtained in lower yield (mixtures
5+5% and 10+10%), and also the diastereoisomers 6
1
and 7. The main H-, 13C- and 119Sn-NMR characteris-
tics of the compounds are summarized in Tables 1 and
2.
On the other hand, the 13C-NMR spectra (Table 1)
show that the values of the 3J(Sn, CꢀO) coupling
constants for compounds 5, 5%, 10 and 10% lie between
66.1 and 81.7 Hz, indicating a dihedral angle of about
180°. The small values of 3J(Sn–C–C–Ph) coupling
constants for these compounds—not observed (5 and
5%), 15.2 (10) and 13.6 Hz (10%)—suggest a dihedral
angle close to 60° between the trialkylstannyl group and
The 13C-NMR chemical shifts (Table 1) were as-
signed through the analysis of the multiplicity of the
signals by means of DEPT experiments and taking into
account the magnitude of J(13C, 119Sn) coupling con-
n
stants. The use of the Karplus-type relationship existing
[4] between the value of the 3J(13C, 119Sn) coupling
constants and the dihedral angle, together with 1H-
NMR data (Table 2), enabled us to deduce the stereo-
chemistry of the adducts. Thus, the 3J(Sn, CꢀO)
coupling constant with values ranging from 11.7 to 16.2
Hz for adducts 3 and 6–9, and not observed in com-
pound 4 (Table 1, C-1) correspond ([4]b) to dihedral
1
the phenyl group attached to C-2. The H-NMR spec-
tra (Table 2) of compounds 5, 5%, 10 and 10% show that
the 3J(H, H) coupling constants for the protons at-
tached to C-2 and C-3 are within the range 12.2–13.5
Hz, indicating a dihedral angle of 180° between these

S.D. Mandolesi et al. / Journal of Organometallic Chemistry 555 (1998) 151–159
153
Table 1
13C- and 119Sn-NMR data of compounds 3–10+10%a
Com-
R
R1 C(1)
C(2)
C(3)
C(1%)
C(2%)
C(3%)
Other signals 119Sn
pound
c
3
4
n-Bu Ph 173.95 (14.6) 54.74 (n.o.)
n-Bu Ph 174.26 (n.o.) 54.92 (n.o.)
n-Bu Ph 174.30 (66.1) 55.14 (n.o.)
n-Bu Ph 174.30 (66.1) 55.54 (n.o.)
37.34 (261.9) 10.27 (314.1) 140.08 (43.2) 144.30 (26.7)
37.48 (261.9) 10.53 (314.1) 140.40 (48.3) 144.25 (26.7)
−
−
−12.0
− 12.7
−113.5
9.0
6.7
d
e
5b
5%b
6
39.08 (n.o.)
39.45 (n.o.)
40.70 (358.4) 139.30
(491.2)
9.28 (311.5)
9.35 (n.o.)
138.97 (n.o.) 144.58 (25.4)
138.97 (n.o.) 144.43 (n.o.)
141.90 (35.0) 139.30 (26.0)
f
g
Ph
Ph 174.00 (11.7) 54.90 (13.5)
Ph 174.60 (16.2) 55.20 (15.2)
Me 174.24 (13.4) 56.88 (10.8)
Me 174.22 (12.6) 57.09 (9.0)
Me 174.66 (81.7) 58.23 (n.o.)
Me 174.58 (80.8) 58.21 (n.o.)
h
i
7
8
Ph
40.80 (368.8) 139.60
(489.3)
23.46 (394.4) −9.79
(319.5)
22.44 (391.3) −9.99
(318.6)
22.32 (398.2) −10.46
(314.2)
142.90 (35.9) 139.95 (24.2)
140.08 (44.0) 16.42 (12.5)
140.18 (52.9) 15.97 (10.0)
140.16 (15.2) 16.29 (n.o.)
139.95 (13.6) 16.17 (n.o.)
−124.4
7.2
Me
Me
Me
Me
j
9
10.3
11.3
7.9
l
10k
10%k
m
21.99 (394.9) −10.74
(312.7)
aIn CDCl3; chemical shifts, l, in ppm with respect to TMS (13C spectra) and Me4Sn (119Sn spectra); nJ(Sn, C) coupling constants in Hz (in
brackets); n.o.=not observed.
bFrom mixtures (5+5%) with either 5 or 5% in excess.
c13.66; 16.08; 20.94; 21.90; 23.06; 25.78; 27.47 (58.5 Hz); 29.05 (17.8 Hz); 31.25; 34.18; 39.34; 74.79; 123.50; 126.41; 127.27; 127.91; 127.98.
d13.93; 15.81; 20.33; 22.32; 23.27; 25.34; 27.78 (57.22 Hz); 29.32 (19.1 Hz); 31.72; 34.58; 41.44; 47.73; 74.87; 123.30; 126.42; 127.15; 127.20; 127.83;
127.91; 127.96.
e13.47; 15.58; 20.65; 21.80; 23.09; 25.31; 27.25 (58.5 Hz); 28.75 (19.1 Hz); 31.18; 34.21; 40.33; 47.03; 74.04; 123.97; 126.53; 127.59; 127.95; 128.39;
128.79.
f13.47; 15.45; 20.81; 25.31; 27.25 (58.5 Hz); 22.87; 28.75 (19.1 Hz); 31.18; 34.21; 40.24; 46.83; 73.85; 124.17; 127.03; 127.79; 128.16; 128.46.
g15.33; 20.33; 21.78; 22.53; 24.68; 30.98; 33.95; 39.61; 46.89; 74.82; 124.17; 126.44; 127.64; 127.76; 127.81; 127.96; 128.01; 128.20; 128.30; 128.43;
128.55; 128.90; 135.87; 136.90; 137.13; 137.36.
h15.78; 20.75; 21.86; 22.98; 25.71; 31.11; 33.97; 39.63; 41.76; 46.46; 75.10; 124.48; 126.49; 128.23; 127.89; 128.01; 128.23; 128.56; 137.31; 139.36.
i16.05; 20.72; 21.83; 23.03; 25.79; 31.15; 34.07; 39.91; 46.60; 74.41; 126.61; 127.77; 128.20.
j15.48; 20.41; 21.95; 22.83; 24.98; 31.24; 34.11; 40.75; 47.14; 74.23; 126.67; 127.86; 128.18.
k
From mixtures (10+10%) with either 10 or 10% in excess.
l17.42; 20.87; 21.86; 22.32; 23.60; 26.18; 31.69; 34.60; 40.81; 47.26; 74.85; 127.74; 128.60; 128.71.
m15.32; 20.60; 21.10; 22.26; 23.40; 26.07; 31.63; 34.55; 40.73; 46.91; 74.53; 127.15; 127.79; 128.02.
ducts, led to the mixtures of the corresponding i-
bromo esters. In the case of adducts 3, 4, 6, and 7, the
resulting i-bromo esters could be separated and were
identified by comparison with authentic samples [3]. On
the other hand, we were not able to separate the
mixture of i-bromo esters resulting from the bromod-
estannylation of adducts 8 and 9. Direct reduction of
the mixtures of i-bromo esters (11+11%) with an ex-
cess of lithium aluminium hydride yielded (R)-(−)-2,3-
diphenylpropan-1-ol (13a) [5] in the case of the bro-
moesters derived from adducts 3 and 6 and (R)-(−)-
protons. These values strongly suggest that compounds
5, 5%, 10 and 10% have the erythro configuration (Fig. 1,
II), i.e. (2S, 3R)- and (2R, 3S)- for compounds 5 and 5%,
and (2R, 3R)- and (2S, 3S)- for compounds 10 and 10%.
The absolute configuration of adducts 3, 4 and 6–9
was established by chemical correlation according to
Scheme 2.
Bromodestannylation of adducts 3, 4 and 6–9, in
carbon tetrachloride and using a ratio adduct/bromine
1:2 in the case of trimethyl- and tri-n-butylstannyl
adducts and a 1:4 ratio for the triphenylstannyl ad-

154
S.D. Mandolesi et al. / Journal of Organometallic Chemistry 555 (1998) 151–159
Table 2
1H-NMR data of compounds 3–10+10%a
Compound
R
R1
Hh 3J(Sn, H)
Hi 2J(Sn, H)
3J(Hh, Hi)
Hk
Other signals
d
e
f
3
4
n-Bu
n-Bu
n-Bu
n-Bu
Ph
Ph
Me
Me
Me
Ph
Ph
Ph
Ph
Ph
Ph
Me
Me
Me
Me
4.29 (35.4)
4.28 (28.8)
4.34 (n.o.)
4.28 (n.o.)
4.42 (25.2)
4.44 (26.7)
3.61 (52.0)
3.46 (40.2)
3.48 (39.0)
3.40 (n.o.)
3.29 (57.7)
3.39 (57.1)
3.42 (55.6)
3.39 (43.7)
3.93 (61.0)
3.73 (69.3)
1.87 (m)
1.93 (m)
2.09 (m)
1.96 (m)
11.5
12.5
13.5
13.4
11.8
8.2
4.71 (m)
4.64 (m)
4.43 (m)
4.45 (m)
4.32 (m)
4.30 (m)
4.65 (m)
4.48 (m)
4.63 (m)
4.66 (m)
5b
5%b
6
g
h
i
7
8
9
j
9.8
k
l
11.0
12.2
12.9
10c
10%c
m
Me
aIn CDCl3; chemical shifts, l, in ppm with respect to TMS; nJ(Sn, H) coupling constants in Hz (in brackets); multiplicity: d=doublet,
m=multiplet (in brackets), n.o.=not observed.
bFrom mixtures (5+5%) with either 5 or 5% in excess.
cFrom mixtures (10+10%) with either 10 or 10% in excess.
d0.87 [d, 3H, 3J(H, H) 7.0 Hz]; 0.89 [d, 3H, 3J(H, H) 7.0 Hz]; 0.94 [d, 3H, 3J(H, H) 6.9 Hz]; 1.28 [d, 3H, 3J(H, H) 7.2 Hz]; 1.38–1.62 (m, 6H);
1.64–1.75 (m, 1H); 2.09–2.20 (m, 1H); 6.85–6.91 (m, 2H); 7.02–7.30 (m, 8H).
e0.80 [d, 3H, 3J(H, H) 7.5 Hz]; 0.85 [d, 3H, 3J(H, H) 7.9 Hz]; 0.89 [d, 3H, 3J(H, H) 7.6 Hz]; 0.92 [d, 3H, 3J(H, H) 6.9 Hz]; 1.21–1.52 (m, 7H);
1.71 (m, 1H); 1.96 (m, 1H); 6.80–6.90 (m, 2H); 7.07–7.17 (m, 8H).
f0.71 [d, 3H, 3J(H, H) 7.0 Hz]; 0.78 [d, 3H, 3J(H, H) 6.7 Hz]; 0.76 [d, 3H, 3J(H, H) 6.7 Hz]; 0.80 [d, 3H, 3J(H, H) 7.0 Hz]; 1.03–1.33 (m, 9H);
1.40–1.50 (m, 1H); 1.52–1.62 (m, 1H); 6.94–7.08 (m, 2H); 7.11–7.57 (m, 8H).
gSignals superimposed with those belonging to isomer 5.
h0.26 [d, 3H, 3J(H, H) 7.5 Hz]; 0.45 [d, 3H, 3J(H, H) 7.5 Hz]; 0.77 [d, 3H, 3J(H, H) 7.4 Hz]; 0.94–1.10 (m, 7H); 1.24 (m, 1H); 1.52 (m, 1H);
6.83–6.95 (m, 2H); 7.00–7.09 (m, 8H); 7.24–7.47 (m, 15H).
i0.51 [d, 3H, 3J(H, H) 6.7 Hz]; 0.75 [d, 3H, 3J(H, H) 7.3 Hz]; 0.79 [d, 3H, 3J(H, H) 6.7 Hz]; 0.84–1.22 (m, 7H); 1.25 (m, 1H); 1.55 (m, 1H);
6.84–6.94 (m, 2H); 6.96–7.14 (m, 8H); 7.18–7.46 (m, 15H).
j0.026 [s, 3J(Sn, H) 51.4 Hz]; 0.74 [d, 3H, 3J(H, H) 6.7 Hz]; 0.81 [d, 3H, 3J(H, H) 7.3 Hz]; 0.86 [d, 3H, 3J(H, H) 6.7 Hz]; 1.00 [d, 3H, 3J(H, H)
7.3 Hz, 3J(Sn, H) 65.0 Hz]; 1.41 (m, 1H); 1.05–1.36 (m, 7H); 7.20–7.31 (m, 5H).
k0.006 [s, 3J(Sn, H) 51.2 Hz]; 0.29 [d, 3H, 3J(H, H) 6.7 Hz]; 0.48 [d, 3H, 3J(H, H) 6.7 Hz]; 0.83 [d, 3H, 3J(H, H) 6.7 Hz]; 0.89 [d, 3H, 3J(H, H)
7.3 Hz]; 1.38 (m, 1H); 1.80 (m, 1H); 1.05–1.27 (m, 7H); 7.06–7.24 (m, 5H).
l−0.26 [s, 3J(Sn, H) 50.0 Hz]; 0.45 [d, 3H, 3J(H, H) 6.7 Hz]; 0.77 [d, 3H, 3J(H, H) 6.8 Hz]; 0.81 [d, 3H, 3J(H, H) 6.7 Hz]; 1.17–1.35 (m); 1.44
(m); 1.79 (m); 7.10–7.37 (m).
m−0.06 [s, 3J(Sn, H) 52.4 Hz], signals of 10 and 10% superimposed.
2-phenylbutan-1-ol (13b) [6] in the case of the bro-
moesters obtained from adduct 8. Similarly, reduction
of mixtures (12+12%) led to (S)-(+)-2,3-diphenyl-
propan-1-ol (14a) [5] in the case of the bromoesters
derived from adducts 4 and 7, and to (S)-(+)-2-
phenylbutan-1-ol (14b) [6] in the case of the bro-
moesters obtained from adduct 9.
Working back from the stereochemistry of the
propanols obtained (13a, 13b, 14a and 14b) it is possi-
ble to make the stereochemical assignments for the
starting adducts. Thus, the absolute configuration of
adducts 3 and 6 is (2R, 3R), of adduct 8 is (2R, 3S), of
adducts 4 and 7 is (2S, 3S), and 9 is (2S, 3R).
acyclic activated olefinic systems takes place with a high
degree of stereoselectivity. The fact that this stereoselec-
tivity is almost the same whether the starting olefin
contains a methyl or a (−)-menthyl ester group, indi-
cates that in these additions, the observed stereoselec-
tivity is independent of the size of the ester group.
The observed threo stereochemistry of the main or
only products obtained in the hydrostannation of
alkyl(E)-2-phenyl-3-methyl (phenyl) propenoates indi-
cates that these additions take place following a prefer-
ential
syn
steric
course.
This
preferential
stereochemistry could be explained [2], taking into ac-
count the fact that acyclic radicals can react with high
stereoselectivity if they adopt preferred conformations.
The six possible intermediate alkyl radicals resulting
These results and those reported previously [1,3]
(Table 3) clearly indicate that the hydrostannation of

S.D. Mandolesi et al. / Journal of Organometallic Chemistry 555 (1998) 151–159
155
from the addition of the tin radicals to the olefins 1 and
2 are shown in Fig. 2. Fig. 2 also shows the expected
products according to the side of the hydrogen transfer
by another molecule of tin hydride in the last step of
the radical chain.
Most of the examples of 1,2-induction follow an
A-strain model [2,7], where a conjugating substituent
on the radical-bearing carbon dictates that the smallest
substituent on the adjacent stereocenter points to the
same direction as the conjugating group. Considering
that this energetically favored arrangement is not
present in the conformations C and F (Fig. 2), and also
that the tin hydride would be approaching the radicals
C and F between the two largest groups in order to
effect the hydrogen transfer, we can consider these
radicals as intermediates in high-energy pathways that
could be discarded.
Although there are many examples ([7]a) of reactions
in which the conjugating substituents attached to the
radical carbon are either phenyl groups (benzylic sys-
tems) or ester groups (heteroallylic systems), we have
not found any reference about radical systems in which
both conjugating groups, phenyl and ester, are attached
to the radical center. The h-phenyl-i-trialkylstannyl
radicals depicted in Fig. 2 belong to the latter class of
radical systems, i.e. they have two conjugating groups
attached to the radical carbon.
Considering that these additions lead to major or
only products originated in a syn pathway, the radicals
A and B will be first considered. Both radicals contain
the hydrogen attached to carbon i on the same side of
a conjugating group: the ester group in radical A and
the phenyl group in radical B. Then, on the basis of
steric effects, namely the hydrogen transfer between the
medium or the larger group and the C-3 hydrogen, and
A-strain effects—A[1,2]-strain for radical A and A[1,3]-
strain for radical B—radical B appears to be disfavored
relative to radical A in the second step of the propaga-
tion chain. However, in radical B, the trialkylstannyl
group occupies an eclipsed position relative to the
half-filled carbon p orbital which, according to Kochi
et al. [8] is the preferred conformation for these stannyl
radicals due to stabilization by hyper- and homocon-
jugative effects. Moreover, in favor of radical B is the
fact that the phenyl group is a better conjugating group
than the ester group [9].
On the other hand, in radical E, the conjugating
phenyl group points to the same direction as the hydro-
gen atom attached to C-3, but there is no eclipsing of
the trialkylstannyl group with the radical orbital, and
although in radical D, this eclipsing exists, the conjugat-
ing group is, in this case, the ester group.
Taking into account the previous discussion, we
strongly believe that the observed stereoselectivity is
due to the fact that the intermediate radical will adopt
preferentially configuration B and, therefore, that the
hydrogen transfer will take place anti with respect to
the trialkylstannyl group and not to the larger R1
(methyl or phenyl) substituent. The difference of
diastereoselectivity observed between the hydrostanna-
tions carried out with triphenyltin hydride (d.e. 100%)
and with trimethyl- and tri-n-butyltin hydrides (d.e. ca.
75%) could be ascribed both to the higher reactivity of
the triphenyltin hydride (better donor) and to its bulki-
ness that would make interconversion of B to D slower.
This conclusion is also supported by previous studies
on the hydrostannation of methyl(E)-2-methyl-3-
phenylpropenoate and methyl(E)-2-methyl-2-butenoate
(methyl tiglate) with trimethyl-, tri-n-butyl-, and
triphenyltin hydride [1]. These studies showed that the
configuration of the main or only products ([4]b) was
the erythro, i.e. the opposite of that of the products
obtained in the hydrostannation of olefins 1 and 2, and
of their methyl ester analogs. The analysis of the inter-
mediate radicals can help again to justify the observed
stereochemistry. Replacing in Fig. 2 the phenyl group
attached to the carbon radical by a methyl group will
show the intermediate h-methyl-i-trialkylstannyl radi-
cals corresponding to these additions.
The reversal of the stereochemistry could be then
explained, considering that, in this case, the only inter-
mediate radical in which the two stabilizing effects,
namely the trialkylstannyl group eclipsed with the half-
filled carbon p orbital and the conjugating group on the
same side as the smaller group (hydrogen) in the neigh-
boring stereocenter, are present, is G (Fig. 3).
The discussed results strongly suggest that in the case
of the hydrostannation of acyclic activated olefinic sys-
tems, the observed stereoselectivity is related to the
existence of preferred conformations in the intermediate
radicals arising from both A-strain effects and the
hyperconjugative interaction existing between the i-tri-
alkylstannyl substituent and the half-filled carbon p
orbital. Our results are in agreement with those recently
reported by Giese et al. [10], in that the degree and the
direction of the 1,2-induction is decided by the sub-
stituents at the radical center. But we feel that this
factor alone is not enough to explain the observed
Fig. 1. Preferred conformations of threo and erythro compounds
3–10% (only one stereoisomer of each is shown).

156
S.D. Mandolesi et al. / Journal of Organometallic Chemistry 555 (1998) 151–159
Scheme 2. Reaction sequences for obtaining the absolute configuration of adducts 3, 4 and 6–9.
reversal of the stereochemistry in the particular case of
the hydrostannation of acyclic activated olefinic sys-
tems.
Further work in order to obtain more information on
the stereochemistry of hydrostannations is in progress.
Argentina) with a Bruker AC 200 instrument (1H and
13C). IR spectra were recorded with a Perkin-Elmer
599B spectrophotometer. Microanalyses were per-
formed at Dortmund University and at INQUIMAE
(University of Buenos Aires, Argentina). Specific rota-
tions were measured with a Polar L-vP, IBZ Messtech-
nik. All the solvents and reagents used were analytical
reagent grade. Triorganotin hydrides were prepared by
reduction of the corresponding chlorides with lithium
aluminium hydride and the starting olefins, 1 and 2,
were prepared as described [3]. Caution: due to its very
high toxicity, trimethyltin hydride must always be han-
dled in a very efficient fume cupboard.
3. Experimental
The NMR spectra were determined partly at Dort-
mund University (Germany) (1H, 13C and 119Sn) and at
Alicante University (Spain) (1H and 13C), using a Bruker
AM 300 instrument, and partly at IQUIOS (Rosario,

S.D. Mandolesi et al. / Journal of Organometallic Chemistry 555 (1998) 151–159
157
Table 3
Organotin hydride additions to (E)-2,3-disubtituted alkyl propenoates
Hydride
Configuration
R2=(−)-Menthyl
R1=Ph
R2=Methylb
R1=Ph
R1=Me
R1=Me
Me3SnH
n-Bu3SnH
Ph3SnH
Threo
Erythro
Threo
Erythro
Threo
Erythro
89
11
89
11
100
—
87
13
80
20
85
15
100
—
75
25
100
—
100
—
85a
15a
100a
—
aFrom the 119Sn-NMR spectra.
bSee [1,3].
All the reactions were carried out following the same
procedure. One experiment is described in detail in each
case in order to illustrate the method used.
(AIBN) as a catalyst (this optimal time of reaction and
the use of an adequate excess of organotin hydride were
indicated by earlier experiments in which the reaction
was monitored by taking samples at intervals and ob-
serving the disappearance of the Sn–H absorption by
IR, and by checking at the end of the reaction that the
1H-NMR spectrum of the reaction mixture no longer
showed the presence of unchanged olefin). The 1H-
NMR spectrum showed that under these conditions, a
quantitative yield (based on starting olefin) of a mixture
of diastereoisomeric adducts 3, 4 and 5+5% was ob-
tained. The relative amount of each diastereoisomer in
the mixture was 41.5 (3), 47.8 (4), 5.1 (5), and 5.5% (5%)
as shown by the integration of the 119Sn-NMR spec-
trum.
3.1. Addition of organotin hydrides to
(−)-menthyl(E)-2,3-diphenylpropenoate (1) and
(−)-menthyl(E)-2-phenyl-2-butenoate (2). Reaction of
(−)-menthyl(E)-2,3-diphenylpropenoate (1) with
tri-n-butyltin hydride. Synthesis of (−)-menthyl(2R,
3R)-, (2S, 3S)-, and (2RS, 3SR)-2,3-diphenyl-3-
(triphenyl-stannyl) propanoates (3, 4, 5, and 5%)
1
The H, 13C and 119Sn-NMR data of the new organ-
otin compounds are included in Tables 1 and 2; other
physical characteristics, reaction conditions, and ele-
mental analyses (C, H) are given in Table 4.
Ester 1 (10 g, 0.0276 mol) was treated for 6 h with
tri-n-butyltin hydride (12.03 g, 0.0414 mol) under
nitrogen at 85 °C and with azobisisobutyronitrile
Column chromatography on silica gel 60 of the crude
mixture, yielded 13.2 g of a mixture of compounds 3
and 4, which were eluted with light petroleum (b.p.
30–65°C) and light petroleum (b.p. 30–65°C)/carbon
Table 4
Some physical properties, reaction conditions and elemental analyses
Compound no. IRa w(CꢀO)
M.p. (°C)b or v2D0
[h]2D5 (concentration)c
Timed (85°C)
Elemental analyses: found (calc.) (%)
C
H
3
4
6
7
8
9
1725e
1729e
1713
1720
1725
1730
1.5290
1.5250
174–175
95–96
99–100
44–45
−25.0° (1.70)
−8.0° (1.26)
−49.0° (8.80)
−15.0°(1.80)
−24.0°(0.70)f
+26.0°(0.84)f
360
360
240
240
120
120
68.22 (67.99)
67.73 (67.99)
72.17 (72.38)
72.59 (72.38)
59.62 (59.37)
59.51 (59.37)
8.75 (8.94)
9.22 (8.94)
6.58 (6.49)
6.23 (6.49)
8.31 (8.23)
8.35 (8.23)
c
aIR spectra as KBr pressed disc except when otherwise stated; w in cm−1
.
bRecrystallized from ethanol. In benzene except when otherwise stated.
f
dIn min; hydride/olefin ratio: 1.5. eFilm. Ethyl ether.

158
S.D. Mandolesi et al. / Journal of Organometallic Chemistry 555 (1998) 151–159
Fig. 2. Intermediate h-phenyl-i-trialkylstannyl radicals (only one enantiomer of each is shown).
tetrachloride 1:1, and 1.54 g of a mixture of compounds
5 and 5%, which eluted in the last fractions of the
chromatography. Compounds 3 and 4 were obtained
pure from the chromatography of the mixture 3+4,
using a fraction collector [eluent: light petroleum (b.p.
30–65°C)/carbon tetrachloride 1:1]. The mixture of
compounds 5+5% could not be separated by this
method, but we were able to obtain mixtures enriched
in each diastereoisomer which enabled to obtain the
NMR characteristics of each diastereoisomer.
The crude products obtained in the hydrostannation
of ester 1 with triphenyltin hydride and of ester 2 with
trimethyltin hydride (see experimental conditions in
Table 4), were purified by column chromatography and
the adducts separated by fractional recrystallization
(ethanol).
3.2. Bromodestannylation reactions. Reaction of
(−)-menthyl(2R, 3S)-2-phenyl-3-(trimethylstannyl)-
butanoate (8) with bromine. Synthesis of the mixture of
(−)-menthyl(2R, 3S)- and (2R, 3R)-2-phenyl-3-
bromobutanoates (in Scheme 2: 11+11% with R=Me)
with LiAlH4. Synthesis of (R)-(−)-2-phenylbutan-1-ol
(13b)
To a solution of 8 (1 g, 0.0022 mol) in carbon
tetrachloride (6 ml) was added dropwise a solution of
bromine in carbon tetrachloride (5.4 ml of a 0.8 M
solution, 0.0044 mol), with stirring, in the dark. After 4
1
h, the H-NMR showed a quantitative yield of a mix-
ture of (−)-menthyl(2R, 3S)- and (2R, 3R)-2-phenyl-3-
bromobutanoates (11+11%, R=Me). The solvent was
distilled off under reduced pressure. The crude product
purified by column chromatography (silica gel 60),
yielded 0.75 g, (89.3%) of the mixture of compounds
11+11% (R=Me), eluted with carbon tetrachloride.
3.3. Reduction of the i-bromo esters. Reaction of the
mixture of (−)-menthyl(2R, 3S)- and (2R,
3R)-2-phenyl-3-bromobutanoates (in Scheme 2:
11+11%, R=Me) with LiAlH4. Synthesis of
(R)-(−)-2-phenylbutan-1-ol (13b)
To a suspension LiAlH4 (0.47 g, 0.0124 mol) in
anhydrous ether (5 ml) was added with stirring a solu-
Fig. 3. Preferred intermediate h-methyl-i-trialkylstannyl radical.

S.D. Mandolesi et al. / Journal of Organometallic Chemistry 555 (1998) 151–159
159
tion of 11+11% (R=Me) (1.20 g, 0.0031 mol) in
References
ether (60 ml). The mixture was heated under reflux
for 6 h. After cooling, the mixture was decomposed
by the addition of HCl (20%, 6 ml). The organic
layer was dried with MgSO4 and the solvent was dis-
tilled off under reduced pressure. The elimination of
the (−)-menthol generated in the reaction under re-
duced pressure with the aid of a cold finger led to
alcohol 13b (0.43 g, 90%), [h]2D0 −24.5° (c, 0.71;
ether) ([6] [h]2D0 −19.45° (neat)).
[1] (a) J.C. Podesta´, A.B. Chopa, A.D. Ayala, J. Organomet. Chem.
212 (1981) 163. (b) J.C. Podesta´, A.B. Chopa, A.D. Ayala, J.
Organomet. Chem. 229 (1982) 223. (c) A.B. Chopa, L.C. Koll,
M.C. Savini, J.C. Podesta´, W.P. Neumann, Organometallics 4
(1985) 1036.
[2] (a) N.A. Porter, B. Giese, D.P. Curran, Acc. Chem. Res. 24
(1991) 296. (b) N.A. Porter, B. Giese, D.P. Curran, Stereochem-
istry of Radical Reactions, VCH, Weinheim, 1995.
[3] J.C. Podesta´, A.B. Chopa, L.C. Koll, S.D. Mandolesi, J.
Organomet. Chem. 434 (1992) 269.
[4] (a) D. Doddrell, I. Burfitt, W. Kitching, et al., J. Am. Chem.
Soc. 96 (1974) 1640. (b) T.N. Mitchell, J.C. Podesta´, A.D. Ayala,
A.B. Chopa, Magn. Reson. Chem. 26 (1988) 497.
[5] R. Buchan, M.B. Watson, J. Chem Soc. C (1969) 2465.
[6] W.A. Bonner, T.W. Greenlee, J. Am. Chem. Soc. 81 (1959) 3336.
[7] (a) W. Smadja, Synlett (1994) 1. (b) D.P. Curran, G. Thoma, J.
Am. Chem. Soc. 114 (1992) 4436.
[8] The magnitude of this interaction can be as large as 5 kcal
mol−1 see: (a) T. Kawamura, J.K. Kochi, J. Am. Chem. Soc. 94
(1972) 648. (b) T. Kawamura, P. Meakin, J.K. Kochi, J. Am.
Chem. Soc. 94 (1972) 8065. (c) A.R. Lyons, M.C.R. Symons, J.
Chem. Soc. Ser. Chem. Commun. (1971) 1068.
[9] F.G. Bordwell, X.-M. Zhang, M.S. Alnajjar, J. Am. Chem. Soc.
114 (1992) 7623.
[10] B. Giese, M. Bulliard, J. Dickhaut, R. Halbach, C. Hassler, U.
Hoffmann, B. Hinzen, M. Senn, Synlett (1995) 116.
Acknowledgements
This work was supported by CONICET (Capital
Federal, Argentina), CIC (Provincia de Buenos Aires,
Argentina), and Universidad Nacional del Sur (Bah´ıa
Blanca, Argentina). The generous help of Prof. M.
Gonza´lez Sierra (IQUIOS, Rosario, Argentina), Prof.
T.N. Mitchell (Dortmund University, Dortmund,
Germany), and Lic. Diego Alonso (Universidad de
Alicante, Spain) in obtaining the NMR spectra is
gratefully acknowledged.
.