Radical hydrogen abstraction-cyclization with a 2-bromovinylsilyl group as a bifunctional tether

Article abstract of DOI:10.1016/j.tet.2005.06.021

In order to develop an efficient method for the regio- and stereoselective introduction of a carbon substituent, a radical hydrogen abstraction-cyclization using 1- and 2-bromovinylsilyl groups as bifunctional tethers was studied. Although, the selective introduction of a carbon substituent at the position β to the hydroxyl was unsuccessful, a carbon substituent was introduced with the 2-bromovinylsilyl ethers via a hydrogen abstraction-cyclization. These reactions were analyzed based on the bond dissociation energies obtained by theoretical calculations.

Full text of DOI:10.1016/j.tet.2005.06.021

Tetrahedron 61 (2005) 7865–7873
Radical hydrogen abstraction–cyclization with a
2-bromovinylsilyl group as a bifunctional tether
*
Makoto Sukeda, Akira Matsuda and Satoshi Shuto
Graduate School of Pharmaceutical Sciences, Hokkaido University, Kita-12, Nishi-6, Kita-ku, Sapporo 060-0812, Japan
Received 11 May 2005; revised 7 June 2005; accepted 8 June 2005
Available online 1 July 2005
Abstract—In order to develop an efficient method for the regio- and stereoselective introduction of a carbon substituent, a radical hydrogen
abstraction–cyclization using 1- and 2-bromovinylsilyl groups as bifunctional tethers was studied. Although, the selective introduction of a
carbon substituent at the position b to the hydroxyl was unsuccessful, a carbon substituent was introduced with the 2-bromovinylsilyl ethers
via a hydrogen abstraction–cyclization. These reactions were analyzed based on the bond dissociation energies obtained by theoretical
calculations.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
ducing a 1-hydroxyethyl or 2-hydroxyethyl group at the
position to the hydroxyl in halohydrins or
b
Intramolecular radical cyclizations employing the tempo-
rary silicon connection approach are very useful for the
introduction of a carbon substituent.1 We previously,
developed a regio- and stereoselective method for intro-
a-phenylselenoalkanols by using an intramolecular radical
cyclization reaction with a dimethyl- or diphenylvinylsilyl
group as the radical acceptor (Scheme 1).2 Thus, the
selective introduction of both the 1-hydroxyethyl and the
Scheme 1.
Keywords: Radical reaction; Hydrogen abstraction; Silicon tether; Cyclization.
*
Corresponding author. Tel.: C81 11 706 3229; fax: C81 11 706 4980; e-mail: shu@pharm.hokudai.ac.jp
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.06.021

7866
M. Sukeda et al. / Tetrahedron 61 (2005) 7865–7873
Scheme 2.
2-hydroxyethyl groups can be achieved, depending on the
concentration of Bu3SnH in the reaction system, via the
5-exo-cyclization intermediate G or the 6-endo-cyclization
intermediate H, respectively, after oxidative cleavage by
treatment of the cyclization products under Tamao
oxidation conditions. We also have developed an efficient
method for the stereoselective introduction of a vinyl and an
ethynyl group via a radical atom-transfer reaction followed
by treatment with fluoride ion,3 and have applied them to the
synthesis of various branched nucleosides and C-glyco-
sides.2,4
Scheme 4.
Although these radical reactions are actually effective for
the regio- and stereoselective introduction of a carbon
substituent, the substrates are limited to halohydrins or
a-phenylselenoalkanols, and the preparation of these
substrates can be troublesome.2b,5 Therefore, we decided
to seek another approach to produce radical C via a
hydrogen abstraction by vinylsilyl radicals, employing
either 1-bromovinylsilylether X1 or 2-bromovinylsilylether
X2 (Scheme 2). Radical hydrogen abstractions have been
studied extensively and many useful reactions have been
reported.6 Brown et al.6b and Robertson et al.6d reported
efficient hydrogen abstraction–cyclizations, namely hydro-
gen abstraction by vinyl radicals and subsequent addition of
the transferred radical to the vinyl group (Schemes 3 and 4).
These results, obtained via effective hydrogen abstraction
by vinyl radicals, suggested that Y1 or Y2 generated from
X1 or X2, respectively, might abstract the hydrogen at the
position b to the silyloxy group to form radical C as shown
in Scheme 2. Compared with the previous procedures in
Scheme 1, this process would make it much easier to
introduce a carbon substituent at the position b to a hydroxyl
group.
We report here, radical hydrogen abstraction–cyclizations
employing either the 1- or 2-bromovinylsilyl group. Studies
on the difference between a- and b-silylvinyl radicals in
their reactivity investigated by theoretical calculations are
also described.
2. Results and discussion
2.1. Working hypothesis
Our strategy, using 1-bromovinyldimethylsilyl (1-BVDS)
ethers X1 or 2-bromovinyldimethylsilyl (2-BVDS) ethers
X2 as the reaction substrates, is shown in Scheme 2.
Treatment of the substrate X1 with a radical initiator would
produce the a-silylvinyl radical Y1. We expected that the
1,5-hydrogen transfer (HT) reaction of Y1 would preferen-
tially proceed to give the desired radical C. We also planned
to investigate the reaction with 2-BVDS ethers X2. If the
1,6-HT reaction of the b-vinylsilyl radical Y2 generated
from the substrate X2 occurred, the radical C would be
formed. Radical C would then cyclize via a 5-exo-mode to
produce the radical D, which corresponds to the key
intermediate in the reaction pathways shown in Scheme 1.
2.2. Synthesis of 1- and 2-bromovinylsilylating reagents
Initially, we prepared 1-bromovinyldimethylsilyl chloride 3
and 2-bromovinyldimethylsilyl bromide 6 as reagents to
introduce the 1- or 2-bromovinylsilyl group into alcohols.
The chlorosilane 3 was synthesized according to the method
reported by Tamao et al. (Scheme 5).7 The 2-bromovinyl-
silylating reagent 6 was prepared from 1,1,3,3-tetramethyl-
1,3-diethynyldisiloxane (4). Hydrostannylation of 4 and the
subsequent stannane–bromine exchange gave 5, which was
Scheme 3.

M. Sukeda et al. / Tetrahedron 61 (2005) 7865–7873
7867
Table 1. Radical hydrogen abstraction–cyclization reaction with the
2-BVDS ether 11
Run Conditions
Yield (%)
12C13 (12:13) 14
1
2
3
4
Bu3SnH, AIBN, PhH, reflux (80 8C)
20 (1.2:1)
33 (2.2:1)
45 (3.3:1)
39 (3.3:1)
64
51
33
33
Bu3SnH, AIBN, PhMe, reflux (111 8C)
Bu3SnH, AIBN, octane, reflux (125 8C)
(TMS)3SiH, AIBN, octane, reflux (125 8C)
Scheme 5.
treated with SiBr4, HMPA and a catalytic amount of water,
followed by distillation to provide the desired bromide 6
(Scheme 6).
using a syringe pump, to a solution of 11 in refluxing
benzene, and the resulting products, without purification,
were treated under Tamao oxidation conditions. At this
stage, the mixture was partially purified using Edelson’s
procedure,10 where the 2-indanol (14) was separated out.
The remaining mixture was acetylated to give a mixture of
12 via 1,5-HT and 13 via 1,6-HT in 20% yield (12/13Z
1.2:1, run 1). Similar reactions at higher temperature in
refluxing toluene (run 2) or octane (run 3) increased the
yield of the branched 2-indanols 12 and 13 (33%, run 2;
45%, run 3). When (TMS)3SiH was used as a reductant in
refluxing octane, compounds 12 and 13 were likewise
obtained (run 4) as in the case of Bu3SnH. We examined
various radical reaction conditions; however, the yield of
the desired compound 13, which had carbon substituents at
the position b to the hydroxyl, was not further improved.
Scheme 6.
2.3. Radical hydrogen abstraction–cyclization with
2-indanol derivatives
We first, tried an approach based on 1,5-HT reactions with
2-indanol derivatives 7, which was prepared by treating
2-indanol with 3 in THF containing pyridine. The substrate
7 was subjected to various reductive radical reaction
conditions with Bu3SnH (0.01 M or lower). However,
none of the desired products, derived from C–C bond
formation at the position b to the hydroxyl, like 9 or 10
(Scheme 7, Fig. 1), was obtained. In these reactions, the
simply reduced product 88 was predominantly formed,
implying that the 1,5-HT step had failed. This result was
consistent with Curran’s report.9
2.4. Radical hydrogen abstraction–cyclization with t-
butylcyclohexanol derivatives
Curran and co-workers also reported that the 1,5-HT step9
failed in the 1-BVDS ether of the mannose derivative 15 as
it did in our case with the substrate 7. However, they
achieved a 1,5-HT reaction with the o-bromophenylsilyl
ethers of the mannose derivative 16 and the trans-t-
butylcyclohexanol 1711 (Fig. 2).
Scheme 7.
Figure 1. Potential products in the reaction of 7 and 11.
Figure 2. Substrates for the radical hydrogen abstraction reaction.
We next prepared the 2-BVDS ethers 11 by treating
2-indanol with 6 in THF containing pyridine, and examined
whether the hydrogen abstraction–cyclization through 1,6-
hydrogen transfer occurred by using 11 as the substrate
(Scheme 8, Table 1). A solution of 1.2 equiv of Bu3SnH and
0.6 equiv of AIBN in benzene was added slowly over 4 h,
In order to make clear the differences in the reactivity of
2-BVDS, 2-BVDS and o-bromophenylsilyl ethers, we
examined the deuterium-labeling experiments using the 1-
and 2-BVDS ethers of trans-t-butylcyclohexanol as sub-
strates under two kinds of reaction conditions with Bu3SnD.
Scheme 8.

7868
M. Sukeda et al. / Tetrahedron 61 (2005) 7865–7873
Scheme 9.
Table 2. Radical hydrogen abstraction–cyclization reaction with 1- or 2-BVDS ethers of trans-t-butylcyclohexanol
Run
Substrates
Conditions
Yield% (D ratio)a
20
21-trans
21-cis
1
2
3
4
18
18
19
19
Standard
Dropwise
Standard
Dropwise
96
(Ha, !1%; Hb, !1%)
74
(Ha, !1%; Hb, !1%)
85
(Ha, 28%; Hb, !1%)
22
(Ha, 3%; Hb, !1%)
—
—
—
—
5
1
(Ha, 25%; Hb, 75%)
50
(Ha, 13%; Hb, 28%)
(Ha, 75%; Hb, 25%)
16
(Ha, 22%; Hb, 10%)
a The ratio of deuterium incorporation was determined as % by 2H NMR experiment.
In one, the standard conditions were those used in the
previous studies by Curran et al.,11 that is, a mixture of the
substrate, Bu3SnD (1.5 equiv) and AIBN (0.3 equiv), was
stirred in refluxing benzene. In the other, the dropwise
conditions of run 3 in Table 1 were followed, except for
using Bu3SnD instead of Bu3SnH (Scheme 9, Table 2).
increased (run 4) compared with the standard conditions.
However, none of the products with a carbon substituent at
the position b to the hydroxyl, such as 13, were detected in
the reaction.13
Thus, it was clear that the b-silylvinyl radical derived from
the 2-BVDS ethers was able to abstract the hydrogen, while
the reactivity seemed to be lower than that of b-silylphenyl
radicals derived from the o-bromophenylsilyl ethers.11
Furthermore, the a-silylvinyl radical derived from
1-BVDS ethers was almost incapable of abstracting the
hydrogen atom at any of the positions in this system.
We first, examined the 1-BVDS ether 18 of trans-t-
butylcyclohexanol. A mixture of 18, Bu3SnD and AIBN
in benzene, was heated under reflux for 4 h (standard
conditions) and the product was purified after Tamao
oxidation. We recovered trans-t-butylcyclohexanol (20) in
2
96% yield, which, by H NMR experiments, was hardly
deuterated (run 1). We also recovered trans-t-butylcyclo-
hexanol (20) under dropwise conditions in 74% yield and it
was also hardly deuterated (run 2). Thus, we could not
isolate the branched t-butylcyclohexanols like 21-trans or
21-cis12 in either of the reaction conditions. These results
indicate that the hydrogen abstraction did not proceed
in the reaction with the 1-BVDS ether of trans-t-
butylcyclohexanol.
2.5. Computational chemistry
Although, the 1,5-hydrogen transfer (HT) is the most
favorable pathway in intramolecular hydrogen transfer
reactions, the 1,5-HT reaction with 1-bromovinylsilyl ethers
(7, 18) did not occur. In hydrogen abstractions, it is essential
that the initially generated radical can be significantly
unstable (reactive) to abstract a hydrogen atom. We
wondered if the silicon atom in the tether moiety might
have stabilized the vinyl radical at the position a, making
the radical rather inactive. Experimental and computational
studies have demonstrated that the a- or b-substituent on
carbon-center radicals often affects the stability of the
radicals.14 The effect of a silicon atom at the position a or b
to aliphatic radicals has been investigated by ab initio
calculations to show that the silicon effectively stabilizes
these radicals.15
We next, examined the 2-BVDS ether 19 of trans-t-
butylcyclohexanol. The substrate 19 was treated under the
standard conditions, and the resulting products, without
purification, were treated under Tamao oxidation con-
ditions. At this stage, the mixture was purified by column
chromatography to give the partly deuterated trans-t-
butylcyclohexanol (20) in 85% yield (run 3). These results
indicate that the hydrogen abstraction occurred to form the
radical at the position a to the hydroxyl, at least to some
extent, which was reduced by Bu3SnD prior to the
cyclization. The remaining mixture was benzoylated to
give the branched trans-t-butylcyclohexanol 21-trans and
its cis isomer 21-cis in low yield. Under the dropwise
conditions, the slightly deuterated t-butylcyclohexanols
were obtained in 22% yield, where the yield of the branched
t-butylcyclohexanols (21-trans, 21-cis) was dramatically
Therefore, we decided to investigate theoretically the a- or
b-silicon effects on non-aliphatic, that is, vinyl and phenyl
radicals. Bond dissociation energies (BDE) of various vinyl
and phenyl radicals were calculated and were compared
with that of the secondary C–H in propane. BDEs are
relevant to the stability of the radical generated by
dissociation, which is useful for understanding hydrogen

M. Sukeda et al. / Tetrahedron 61 (2005) 7865–7873
7869
Table 3. Theoretical bond dissociation energies (BDEs) and relative to the secondary C–H bond dissociation energy in propane (DBDEs)
Speciesa
ZPE (a.u.)b
E0 (a.u.)c
BDE (kJ/mol)
DBDE (kJ/mol)d
CH3CH(–H)CH3 (22)
—
—
409.1
444.8
—
409.6
—
458.7
—
442.5
—
462.7
—
451.4
—
0.0
35.7
—
0.5
—
49.6
—
33.4
—
53.6
—
42.3
—
H2C]C(–H)CMe2OMe (23)
H2C]C%CMe2OMe (230)
H2C]C(–H)SiMe2OMe (24)
H2C]C%SiMe2OMe (240)
(H–)HC]CHCMe2OMe (25)
HC%]CHCMe2OMe (250)
(H–)HC]CHSiMe2OMe (26)
HC%]CHSiMe2OMe (260)
(o-H–)C6H4CMe2OMe (27)
%C6H4CMe2OMe (270)
0.166034
0.152144
0.157112
0.143105
0.166034
0.152449
0.157112
0.143684
0.213960
0.200754
0.204631
0.191771
K310.880777
K310.213389
K562.511049
K561.699963
K310.880777
K310.208106
K562.511049
K561.687426
K464.492189
K463.817990
K715.963861
K715.293962
(o-H–)C6H4SiMe2OMe (28)
%C6H4SiMe2OMe (280)
a The dissociation H atom is shown in parentheses.
b Zero-point energy scaled by 0.9806.
c Energy including ZPE at B3LYP/6-31G(d) level.
d DBDEZBDE (23–28)KBDE (22).
abstraction and free radical reactions.16 Geometry optimiz-
ation and single-point energy calculations were performed
using the density functional theory with the B3LYP
functional17 and 6-31G(d) basis set. The zero-point energies
(ZPE) were calculated using the scale factor 0.9806 reported
by Scott et al.18 The values of BDE were determined by the
isodesmic reaction reported to by Marshall et al.19a and
DiLabio et al.19b to eliminate systematic errors. These
results are summarized in Table 3.
1-bromovinylsilyl ethers (7, 18) did not occur might be
the following. Malacria et al. reported the efficient radical
hydrogen abstraction by a-silylvinyl radical in the radical
cascade reaction (Scheme 10).20 In their substrates, the
three-dimensional positioning of the radical and the
abstracted hydrogen atom is strictly limited to proceed
effective abstraction. However, our substrates (7, 18),
having a 1-bromovinylsilyl group on the hydroxyl, are
conformationally flexible and therefore, have no steric
demand for promoting the intramolecular hydrogen abstrac-
tion. Accordingly, the radical might be reduced by the
reductant in preference to the hydrogen abstraction.
B3LYP calculations showed that the BDE of the vinyl C–H
in allyl ether 23 is ca. 35.7 kJ/mol larger than that of the
secondary C–H in propane (22), while the BDE of the vinyl
C–H in vinylsilyl ether 24 is almost the same as that of the
secondary C–H in propane (22). Therefore, it is likely that
the a-silylvinyl radicals 230 are unable to abstract the
hydrogen atom from a secondary C–H in alkanes. On the
other hand, the BDE of vinylsilyl ether 26 for generating
the b-silylvinyl radical 260 is ca. 30 kJ/mol larger than that
of the secondary C–H in propane (22). Although, the BDE
of 26 is smaller than that of the corresponding carbon
congener 25, the b-silylvinyl radical 260 might be reactive
enough to abstract a hydrogen atom from the secondary
C–H in alkanes. The BDE of the C–H (ortho) in the phenyl
group in 28 is much larger (DBDEZ42.3 kJ/mol) even if
silicon substituents are attached to the phenyl group.
Therefore, this highly reactive property of o-silylphenyl
radical would make the hydrogen abstraction possible even
without the dropwise conditions.11
3. Conclusion
We investigated the radical hydrogen abstraction–cycliza-
tion with 1- or 2-bromovinylsilyl tether as a bifunctional
tether. Although, the selective introduction of a carbon
substituent at the position b to the hydroxyl was
unsuccessful, a carbon substituent was introduced with the
2-bromovinylsilyl ethers in moderate yield, via a hydrogen
abstraction–cyclization. We also investigated theoretically
the cause of the failure of the reaction with 1-bromovinyl-
silyl ethers by comparing the BDE of the vinyl C–H of the
vinylsilyl ether with the BDE of the secondary C–H in
propane, which suggested that the silicon-substituted vinyl
radical, generated from 1-bromovinylsilyl ethers, is too
stable to abstract the hydrogen.
Another reason that the hydrogen abstraction with
4. Experimental
4.1. General procedure
Melting points are uncorrected. NMR spectra were recorded
at 270 and 400 MHz (1H), at 61 MHz (2H), and at 100 MHz
(13C), and are reported in ppm downfield from Me4Si. Mass
spectra were obtained by electron ionization (EI) or the fast
atom bombardment (FAB) method. Thin-layer chromato-
graphy was performed on Merck coated plate 60F254. Silica
gel chromatography was performed with Merck silica gel
5715 or 9385 (neutral). Reactions were carried out under an
argon atmosphere.
Scheme 10.

7870
M. Sukeda et al. / Tetrahedron 61 (2005) 7865–7873
4.1.1. 2-Bromovinyldimethylbromosilane (6). A mixture
of 1,1,3,3-tetramethyl-1,3-diethynyldisiloxane (4, 10 g,
55 mmol), Bu3SnH (32.5 mL, 121 mmol), AIBN (903 mg,
5.5 mmol) was stirred at 80 8C for 2 h. After cooling the
solution to K40 8C, NBS (21.5 g, 121 mmol) and CH2Cl2
(60 mL) were added, and the mixture was stirred at the same
temperature for 2 h and then at 0 8C for 2 h. The resulting
mixture was evaporated, and the insoluble material was
filtered off. The filtrate was purified by distillation (bp 53–
115 8C/3 mmHg) to give crude 5 including some impurities
(10.6 g, ca. 40%). The mixture of crude 5 (10.6 g), SiBr4
(1.37 mL, 11 mmol), HMPA (332 mL, 3.0 wt%), and H2O
(29 mL, 0.25 wt%) was stirred at 60 8C for 18 h. Distillation
of the resulting mixture gave 6 (bp 55–65 8C/13 mmHg,
5.0 g, 19% from 4) as a pale brown liquid: 1H NMR
(270 MHz, CDCl3) d 6.84 (d, 1H, JZ15.5 Hz), 6.63 (d, 1H,
JZ15.5 Hz), 0.67 (s, 6H); HRMS (EI) calcd for C4H789Br2Si
241.8750, found 241.8762 (MC).
purified by column chromatography (neutral SiO2, 0–2%
AcOEt in hexane) to give 11 (435 mg, 98%) as a colorless
1
liquid: H NMR (270 MHz, CDCl3) d 7.21–7.13 (m, 4H),
6.67 (d, 1H, JZ15.8 Hz), 6.55 (d, 1H, JZ15.8 Hz), 4.66 (m,
1H), 3.13 (dd, 2H, JZ6.6, 15.8 Hz), 2.88 (dd, 2H, JZ6.6,
15.8 Hz), 0.24 (s, 6H); 13C NMR (100 MHz, CDCl3) d
140.54, 136.99, 126.41, 124.48, 119.59, 73.97, 42.44,
K1.04; LRMS (EI) m/z 296 (79Br, MC), 298 (81Br, MC).
Anal. Calcd for C13H17BrOSi: C, 52.53; H, 5.76. Found: C,
52.33; H, 5.71.
4.1.5. General procedure for radical hydrogen abstrac-
tion–cyclization reaction with indane derivatives 7 and
11. To a refluxing solution of 7 or 11 (59 mg, 0.20 mmol) in
a solvent (40 mL) was added dropwise a mixture of Bu3SnH
or (TMS)3SiH (0.26–0.30 mmol) and AIBN (0.04–
0.06 mmol) in a solvent [5 mL, when octane was used as
a solvent (Table 1, runs 3 and 4), 10% benzene in octane was
used to dissolve AIBN] over 4 h, and then the resulting
mixture was evaporated. A mixture of the residue, KF
(0.4 mmol), KHCO3 (0.2 mmol), 30% aqueous H2O2
(227 mL, 2.0 mmol) in THF/MeOH (2:1, 3 mL) was stirred
at room temperature for 15–20 h. Saturated aqueous
Na2S2O3 was added, and the resulting mixture was filtered
through Celite. The filtrate was dried (Na2SO4) and
evaporated. A mixture of the residue, CsF (122 mg,
0.8 mmol), CsOH (0.4 mmol), and silica gel (60 mg) in
CH2Cl2 (5 mL) was stirred at room temperature for 30 min
and was applied to a short (2 cm) plug of silica gel. Elution
with 50% hexane in AcOEt afforded the mixture, by which
tin compounds were removed to some extent. 2-Indanol (14)
was separated from the mixture by column chromatography
(SiO2, 5–50% AcOEt in hexane). A mixture of the resulting
mixture, Ac2O (0.2–0.4 mmol), and DMAP (0.2–0.4 mmol)
in MeCN (1 mL) was stirred at room temperature for
15–25 h. To the mixture was added MeOH (0.5 mL), and
evaporated. The residue was partitioned between AcOEt
and H2O, and the organic layer was washed with brine, dried
(Na2SO4), and evaporated. The residue was purified by
column chromatography (SiO2, 8% AcOEt in hexane) to
give a mixture of 12 and 13. The ratio of 12 and 13 was
4.1.2. (G)-2-(1-Bromovinyldimethylsiloxy)indane (7). A
mixture of 2-indanol (14) (403 mg, 3.0 mmol), 1-bromo-
vinyldimethylchlorosilane (3, 540 mL, 3.6 mmol), and
pyridine (315 mL, 3.9 mmol) in THF (30 mL) was stirred
at room temperature for 1 h, and then evaporated. The
residue was partitioned between AcOEt and aqueous HCl
(1 M), and the organic layer was washed with H2O and
brine, dried (Na2SO4), and evaporated. The residue was
purified by column chromatography (neutral SiO2, 0–2%
AcOEt in hexane) to give 7 (851 mg, 95%) as a colorless
1
liquid: H NMR (270 MHz, CDCl3) d 7.21–7.12 (m, 4H),
6.37 (d, 1H, JZ2.0 Hz), 6.33 (d, 1H, JZ2.0 Hz), 4.74 (m,
1H), 3.16 (dd, 2H, JZ6.6, 15.9 Hz), 2.94 (dd, 2H, JZ5.3,
15.9 Hz), 0.33 (s, 6H); 13C NMR (100 MHz, CDCl3) d
140.45, 135.47, 130.73, 126.33, 126.19, 124.68, 124.28,
74.16, 73.03, 42.59, 42.24, K2.27; HRMS (EI) calcd for
C13H BrOSi 296.0232, found 296.0246 (MC).
79
17
4.1.3. (G)-2-Vinyldimethylsiloxyindane (8). A mixture of
2-indanol (14) (81 mg, 0.6 mmol), vinyldimethylchloro-
silane (99 mL, 0.72 mmol), and pyridine (61 mL,
0.75 mmol) in THF (6 mL) was stirred at room temperature
for 1.5 h, and then evaporated. The residue was partitioned
between AcOEt and aqueous HCl (1 M), and the organic
layer was washed with H2O and brine, dried (Na2SO4), and
evaporated. The residue was purified by column chroma-
tography (neutral SiO2, 0–3% AcOEt in hexane) to give 8
1
determined by H NMR analysis. Compounds 12 and 13
were separable by preparative TLC (17% AcOEt in hexane).
2-(2-Acetoxyethyl)-2-acetoxyindane (12): 1H NMR
(400 MHz, CDCl3) d 7.17 (m, 4H), 4.16 (t, 2H, JZ
6.6 Hz), 3.43 (d, 2H, JZ16.8 Hz), 3.21 (d, 2H, JZ16.8 Hz),
2.46 (t, 2H, JZ6.6 Hz), 2.03 (s, 3H), 1.99 (s, 3H); 13C NMR
(100 MHz, CDCl3) d 170.43, 170.23, 139.66, 126.47,
124.14, 89.12, 60.67, 44.57, 35.11, 22.02, 20.94; HRMS
(EI) calcd for C15H18NaO4 285.1096, found 285.1103
(MNaC). 1-(2-Acetoxyethyl)-2-acetoxyindane (13): 1H
NMR (400 MHz, CDCl3) d 7.24–7.20 (m, 4H), 5.57 (m,
1H), 4.22 (t, 2H, JZ6.6 Hz), 3.36 (dt, 1H, JZ6.0, 8.3 Hz),
3.21 (dd, 1H, JZ5.8, 16.6 Hz), 2.99 (dd, 1H, JZ3.2,
16.6 Hz), 2.13–2.03 (m, 2H), 2.06 (s, 3H), 2.03 (s, 3H);
13C NMR (100 MHz, CDCl3) d 170.62, 170.41, 142.92,
139.63, 126.85, 126.54, 124.49, 123.30, 76.16, 62.77,
44.66, 38.43, 26.63, 21.08, 20.94; HRMS (FAB,
positive) calcd for C15H19O4 263.1297, found
263.1283 (MHC).
1
(105 mg, 80%) as a colorless liquid: H NMR (270 MHz,
CDCl3) d 7.19–7.11 (m, 4H), 6.18 (dd, 1H, JZ14.5,
19.8 Hz), 6.02 (dd, 1H, JZ4.6, 14.5 Hz), 5.80 (dd, 1H, JZ
4.6, 19.8 Hz), 4.66 (m, 1H), 3.12 (dd, 2H, JZ6.6, 15.2 Hz),
2.89 (dd, 2H, JZ5.9, 15.2 Hz), 0.22 (s, 6H); 13C NMR
(100 MHz, CDCl3) d 140.63, 137.46, 132.79, 126.10,
124.25, 73.61, 42.29, K1.58; HRMS (EI) calcd for
C13H18OSi 218.1127, found 218.1123 (MC).
4.1.4. (G)-2-(2-Bromovinyldimethylsiloxy)indane (11).
A mixture of 2-indanol (14) (201 mg, 1.5 mmol), 2-bromo-
vinyldimethylbromosilane (6, 282 mL, 1.8 mmol), and
pyridine (158 mL, 1.95 mmol) in THF (15 mL) was stirred
at room temperature for 2 h, and then evaporated. The
residue was partitioned between AcOEt and aqueous HCl
(1 M), and the organic layer was washed with H2O and
brine, dried (Na2SO4), and evaporated. The residue was
4.1.6. trans-1-(1-Bromovinyldimethylsilyl)-4-t-butyl-
cyclohexanol (18). Compound 18 (620 mg, 97%) was

M. Sukeda et al. / Tetrahedron 61 (2005) 7865–7873
7871
obtained as a colorless liquid from trans-t-butylcyclo-
hexanol (313 mg, 2.0 mmol) as described above for the
synthesis of 7, after purification by column chromatography
(neutral SiO2, 0–3% AcOEt in hexane): 1H NMR
(270 MHz, CDCl3) d 6.34 (d, 1H, JZ2.0 Hz), 6.33 (d, 1H,
JZ2.0 Hz), 3.58 (m, 1H), 1.94 (m, 2H), 1.84 (m, 2H), 1.29
(m, 2H), 1.09–0.93 (m, 3H), 0.84 (s, 9H), 0.30 (s, 6H); 13C
NMR (100 MHz, CDCl3) d 136.09, 130.36, 72.79, 46.95,
36.02, 32.20, 27.57, 25.68, K2.04; LRMS (EI) m/z 318
(79Br, MC), 320 (81Br, MC). Anal. Calcd for C14H27BrOSi:
C, 52.65; H, 8.52. Found: C, 52.43; H, 8.34.
positions as a white solid: 2H NMR (61 MHz, CHCl3) d 3.50
(br s, 0.005 H), 1.22 (br s, 0.001 H).
2
2
By the dropwise conditions. After the treatment of 18
(64 mg, 0.2 mmol) according to the above general pro-
cedure (dropwise conditions), the resulting residue was
purified by column chromatography (SiO2, 5–15% AcOEt
in hexane) to give 20 (23 mg, 74%) slightly deuterated at 1
and 2 position as a white solid: 2H NMR (61 MHz, CHCl3)
2
2
d 3.50 (br s, 0.009 H), 2.00 (br s, 0.001 H), 1.21 (br s,
0.002 H).
2
4.1.7. trans-1-(2-Bromovinyldimethylsilyl)-4-t-butyl-
cyclohexanol (19). Compound 19 (464 mg, 73%) was
obtained as a colorless liquid from trans-t-butylcyclo-
hexanol (313 mg, 2.0 mmol) as described above for the
synthesis of 11, after purification by column chromato-
4.2.2. Synthesis of the non-deuterated branched t-butyl-
cyclohexanols (21-trans and 21-cis) as reference com-
pounds. After the treatment of 19 (64 mg, 0.2 mmol)
according to the above general procedure (dropwise
conditions) with Bu3SnH instead of Bu3SnD, trans-t-
butylcyclohexanol 20 was separated from the mixture by
column chromatography (SiO2, 5–50% AcOEt in hexane).
A solution of the resulting mixture, BzCl (46 mL,
0.4 mmol), and pyridine (32 mL, 0.4 mmol) in MeCN
(1 mL) was stirred at room temperature for 20 h and then
evaporated. The residue was partitioned between AcOEt
and H2O, and the organic layer was washed with brine, dried
(Na2SO4), and evaporated. The residue was purified by flash
column chromatography (SiO2, 5–12% AcOEt in hexane)
to give 21-trans (25 mg, 42% from 19) as a white solid and
21-cis (9 mg, 14% from 19) as a white solid.
1
graphy (neutral SiO2, 0–4% AcOEt in hexane): H NMR
(400 MHz, CDCl3) d 6.65 (d, 1H, JZ15.4 Hz), 6.53 (d, 1H,
JZ15.4 Hz), 3.50 (m, 1H), 1.87 (m, 2H), 1.75 (m, 2H), 1.27
(m, 2H), 1.06–0.90 (m, 3H), 0.84 (s, 9H), 0.20 (s, 6H); 13C
NMR (100 MHz, CDCl3) d 137.28, 118.95, 72.27, 46.89,
36.11, 32.18, 27.55, 25.66, K1.03; LRMS (EI) m/z 318
(79Br, MC), 320 (81Br, MC). Anal. Calcd for C14H27BrOSi:
C, 52.65; H, 8.52. Found: C, 52.60; H, 8.44.
4.2. General procedure for radical hydrogen abstrac-
tion–cyclization reaction with 1- or 2-bromovinylsilyl
ethers of trans-t-butylcyclohexanol 18 and 19
trans-1-(2-Benzoyloxy)-4-t-butylcyclohexanol (21-trans):
mp 122–123 8C (white needles from hexane/AcOEt); H
1
By the standard conditions. A mixture of 18 (64 mg,
0.2 mmol) or 19 (64 mg, 0.2 mmol), Bu3SnD (65 mL,
0.24 mmol), and AIBN (3.3 mg, 0.02 mmol) was heated
under reflux for 4 h. The resulting mixture was evaporated,
and a mixture of the residue, KF (23 mg, 0.4 mmol),
KHCO3 (20 mg, 0.2 mmol), and aqueous H2O2 (35%,
194 mL, 2.0 mmol) in THF/MeOH (2:1, 3 mL) was stirred
at room temperature for 15–20 h. To the mixture was added
saturated aqueous Na2S2O3 (1 mL), and the resulting
mixture was filtered through Celite. The filtrate was dried
(Na2SO4) and evaporated.
NMR (400 MHz, CDCl3) d 8.04–7.42 (m, 5H), 4.52 (t, 2H,
JZ6.8 Hz), 2.03 (t, 2H, JZ6.8 Hz), 1.91 (m, 2H), 1.74 (m,
2H), 1.62 (br s, 1H), 1.42 (m, 2H), 1.11 (m, 3H), 0.86 (s,
9H); 13C NMR (100 MHz, CDCl3) d 166.19, 132.58,
129.96, 129.15, 128.05, 71.57, 61.44, 47.40, 39.16, 34.70,
32.18, 27.57, 24.45; LRMS (EI) m/z 304 (MC). Anal. Calcd
for C19H28O3: C, 74.96; H, 9.27. Found: C, 74.90; H, 9.27.
cis-1-(2-Benzoyloxy)-4-t-butylcyclohexanol (21-cis): mp
93–94 8C (white plates from hexane/AcOEt); 1H NMR
(400 MHz, CDCl3) d 8.03–7.42 (m, 5H), 4.52 (t, 2H, JZ
6.6 Hz), 1.92 (t, 2H, JZ6.6 Hz), 1.80 (m, 2H), 1.62 (m, 2H),
1.51 (br s, 1H), 1.44–1.28 (m, 4H), 0.97 (m, 1H), 0.87 (s,
9H); 13C NMR (100 MHz, CDCl3) d 166.18, 132.56,
129.99, 129.18, 128.05, 69.88, 61.31, 47.64, 42.12, 37.69,
32.33, 27.48, 22.30; LRMS (EI) m/z 304 (MC). Anal. Calcd
for C19H28O3: C, 74.96; H, 9.27. Found: C, 74.82; H, 9.31.
By the dropwise conditions. To a refluxing solution of 18 or
19 (64 mg, 0.2 mmol) in octane (40 mL) was added
dropwise a mixture of Bu3SnH (81 mL, 0.30 mmol) or
Bu3SnD (81 mL, 0.30 mmol) and AIBN (10 mg, 0.06 mmol)
in octane/benzene (1:9 10 mL) over 4 h, and the resulting
mixture was stirred for further 1 h, then evaporated.
A mixture of the residue, KF (23 mg, 0.4 mmol), KHCO3
(20 mg, 0.2 mmol), 35% aqueous H2O2 (194 mL, 2.0 mmol)
in THF/MeOH (2:1, 3 mL) was stirred at room temperature
for 15–20 h. Saturated aqueous Na2S2O3 (1 mL) was added,
and the resulting mixture was filtered through Celite. The
filtrate was dried (Na2SO4) and evaporated.
4.2.3. D-label experiment with trans-1-(2-bromovinyl-
dimethylsilyl)-4-t-butylcyclohexanol (19). By the standard
conditions. After the treatment of 19 (64 mg, 0.2 mmol)
according to the above general procedure (normal con-
ditions), trans-t-butylcyclohexanol 20 (26.5 mg, 85%)
partly deuterated at the a or b position of the benzoyloxy
group was separated from the mixture by column
chromatography (SiO2, 5–50% AcOEt in hexane). A mix-
ture of the resulting mixture, BzCl (12 mL, 0.1 mmol), and
pyridine (8 mL, 0.1 mmol) in MeCN (0.5 mL) was stirred at
room temperature for 20 h. The mixture was evaporated,
and the residue was partitioned between AcOEt and H2O,
washed with brine, dried (Na2SO4), and evaporated. The
residue was purified by flash column chromatography (SiO2,
4.2.1. D-label experiment with trans-1-(1-bromovinyl-
dimethylsilyl)-4-t-butylcyclohexanol (18). Slightly
deuterated trans-4-t-butylcyclohexanol (20). By the
standard conditions. After the treatment of 18 (64 mg,
0.2 mmol) according to the above general procedure
(standard conditions), the resulting residue was purified by
column chromatography (SiO2, 5–15% AcOEt in hexane) to
give 20 (30 mg, 96%) slightly deuterated at the 1 and 2

7872
M. Sukeda et al. / Tetrahedron 61 (2005) 7865–7873
4984–4985. (c) Bols, M.; Skrydstrup, T. Chem. Rev. 1995, 95,
1253–1277. (d) Fensterbank, L.; Malacria, M.; Sieburth,
S. McN. Synthesis 1997, 813–854. (e) Gauthier, D. R., Jr.;
Zandi, K. S.; Shea, K. J. Tetrahedron 1998, 54, 2289–2338.
2. (a) Shuto, S.; Kanazaki, M.; Ichikawa, S.; Matsuda, A. J. Org.
Chem. 1997, 62, 5676–5677. (b) Shuto, S.; Kanazaki, M.;
Ichikawa, S.; Minakawa, N.; Matsuda, A. J. Org. Chem. 1998,
63, 746–754. (c) Shuto, S.; Sugimoto, I.; Abe, H.; Matsuda, A.
J. Am. Chem. Soc. 2000, 122, 1343–1351. (d) Kanazaki, M.;
Ueno, Y.; Shuto, S.; Matsuda, A. J. Am. Chem. Soc. 2000, 122,
2422–2432.
5–12% AcOEt in hexane) to give 21-trans (5 mg, 5% from
19) as a white solid and 21-cis (!1 mg, !1% from 19) as a
white solid.
2
Partly deuterated trans-t-butylcyclohexanol (20): H NMR
(61 MHz, CHCl3) d 3.50 (br s, 0.28 2H), 2.01 (br s, 0.01 2H),
1.21 (br s, 0.001 H).
2
Deuterated trans-1-(2-benzoyloxy)-4-t-butylcyclohexanol
(21-trans): 1H NMR (270 MHz, CDCl3) d 8.04–7.41
(m, 5H), 4.52 (d, 0.5H, JZ6.6 Hz), 4.51 (t, 0.75H, JZ
6.6 Hz), 2.02 (d, 1.5H, JZ6.6 Hz), 2.01 (t, 0.25H, JZ
6.6 Hz), 1.91 (m, 2H), 1.73 (m, 2H), 1.60 (br s, 1H), 1.42 (m,
2H), 1.13 (m, 3H), 0.86 (s, 9H).
3. Sugimoto, I.; Shuto, S.; Matsuda, A. J. Org. Chem. 1999, 64,
7153–7157.
4. (a) Ueno, Y.; Nagasawa, Y.; Sugimoto, I.; Kojima, N.;
Kanazaki, M.; Shuto, S.; Matsuda, A. J. Org. Chem. 1998, 63,
1660–1667. (b) Shuto, S.; Yahiro, Y.; Ichikawa, S.; Matsuda,
A. J. Org. Chem. 2000, 65, 5547–5557. (c) Sukeda, M.; Shuto,
S.; Sugimoto, I.; Ichikawa, S.; Matsuda, A. J. Org. Chem.
2000, 65, 8988–8996. (d) Kodama, T.; Shuto, S.; Nomura, M.;
Matsuda, A. Chem. Eur. J. 2001, 7, 2332–2340. (e) Sukeda,
M.; Ichikawa, S.; Matsuda, A.; Shuto, S. Angew. Chem., Int.
Ed. 2002, 41, 4748–4750. (f) Sukeda, M.; Ichikawa, S.;
Matsuda, A.; Shuto, S. J. Org. Chem. 2003, 68, 3465–3475.
5. Giese, B.; Dussy, A.; Elie, C.; Erdmann, P.; Schwitter, U.
Synthesis 1994, 1310–1312.
Deuterated cis-1-(2-benzoyloxy)-4-t-butylcyclohexanol
(21-cis): 1H NMR (400 MHz, CDCl3) d 8.04–7.41 (m,
5H), 4.52 (d, 1.5H, JZ6.6 Hz), 4.51 (t, 0.25H, JZ6.6 Hz),
1.92 (d, 0.75H, JZ6.6 Hz), 1.90 (t, 0.5H, JZ6.6 Hz), 1.80
(m, 2H), 1.57 (m, 3H), 1.44–1.28 (m, 4H), 0.97 (m, 1H),
0.87 (s, 9H).
4.2.4. D-label experiment with trans-1-(2-bromovinyl-
dimethylsilyl)-4-t-butylcyclohexanol (19). By the drop-
wise conditions. After the treatment of 19 (64 mg,
0.2 mmol) according to the above general procedure
(dropwise conditions), trans-t-butylcyclohexanol 20
(7.0 mg, 22%) was separated from the mixture by column
chromatography (SiO2, 5–50% AcOEt in hexane). A mix-
ture of the resulting mixture, BzCl (62 mL, 0.54 mmol), and
pyridine (43 mL, 0.54 mmol) in MeCN (1 mL) was stirred at
room temperature for 15 h. The mixture was evaporated,
and the residue was partitioned between AcOEt and H2O,
and the organic layer was washed with brine, dried
(Na2SO4), and evaporated. The residue was purified by
flash column chromatography (SiO2, 5–12% AcOEt in
hexane) to give 21-trans (30 mg, 50% from 19) as a white
solid and 21-cis (10 mg, 16% from 19) as a white solid.
6. (a) Curran, D. P.; Kim, D.; Liu, H. T.; Shen, W. J. Am. Chem.
Soc. 1988, 110, 5900–5902. (b) Brown, C. D. S.; Simpkins,
N. S.; Clinch, K. Tetrahedron Lett. 1993, 34, 131–132.
(c) Bosch, E.; Bachi, M. D. J. Org. Chem. 1993, 58,
5581–5582. (d) Robertson, J.; Peplow, M. A.; Pillai, J.
Tetrahedron Lett. 1996, 37, 5825–5828. (e) Bertrand, M. P.;
Crich, D.; Nouguier, R.; Samy, R.; Stien, D. J. Org. Chem.
1996, 61, 3588–3589. (f) Kittaka, A.; Asakura, T.; Kuze, T.;
Tanaka, H.; Yamada, N.; Nakamura, K. T.; Miyasaka, T.
J. Org. Chem. 1999, 64, 7081–7093. (g) Chatgilialoglu, C.;
Gimisis, T.; Spada, G. P. Chem. Eur. J. 1999, 5, 2866–2876.
7. Tamao, K.; Maeda, K.; Yamaguchi, T.; Ito, Y. J. Am. Chem.
Soc. 1989, 111, 4984–4985.
2
8. The vinylsilyl ether 8 was prepared in advance as a reference
from 2-indanol (14) and vinyldimethylchlorosilane. The
details are described in Section 4.
Partly deuterated trans-t-butylcyclohexanol (20): H NMR
(61 MHz, CHCl3) d 4.52 (br s, 0.28 2H), 2.02 (br s, 0.13 2H).
9. Curran et al. tried to abstract the anomeric hydrogen atom of
mannose derivatives by a-silylvinyl radical; however, the
hydrogen abstraction was not observed: Yamazaki, N.;
Eichenberger, E.; Curran, D. P. Tetrahedron Lett. 1994, 35,
6623–6626.
Partly deuterated trans-1-(2-benzoyloxy)-4-t-butylcyclo-
hexanol (21-trans): H NMR (61 MHz, CHCl3) d 4.53 (br
2
2
s, 0.10 H), 1.93 (br s, 0.22 H).
2
Partly deuterated cis-1-(2-benzoyloxy)-4-t-butylcyclo-
hexanol (21-cis): H NMR (61 MHz, CHCl3) d 3.50 (br s,
2
10. Edelson, B. S.; Stoltz, B. M.; Corey, E. J. Tetrahedon Lett.
1999, 40, 6729–6730.
2
0.03 H).
11. Curran, D. P.; Somayajula, K. V.; Yu, H. Tetrahedron Lett.
1992, 33, 2295–2298.
4.3. Computations
12. The non-deuterated branched t-butylcyclohexanols 21-cis and
21-trans were prepared in advance as references from the
2-bromovinylsilyl ether 19 and Bu3SnH instead of Bu3SnD as
a reductant with dropwise conditions. The details are described
in the Section 4. The term cis and trans denote the relationship
between the tert-butyl group and the hydroxy groups.
13. The reason why 1,6-HT did not occur in the reaction with
t-butylcyclohexanol derivative 19 but occurred in the reaction
with indane derivatives 11 might be due to differences of the
conformation and/or the BDE between the two substrates. The
substrate 11 has a benzylic hydrogen, the BDE(C–H), of which
is considerably lower than that of usual secondary C–H, and
therefore, the 1,6-HT might have occurred.
Geometry optimizations and single-point energy calcu-
lations were performed using B3LYP/6-31G*. The ZPE was
calculated with scale factor 0.9806.
References and notes
1. (a) Nishiyama, H.; Kitajima, K.; Matsumoto, M.; Itoh, K.
J. Org. Chem. 1984, 49, 2298–2230. (b) Tamao, K.; Maeda,
K.; Yamaguchi, T.; Ito, Y. J. Am. Chem. Soc. 1989, 111,

M. Sukeda et al. / Tetrahedron 61 (2005) 7865–7873
7873
14. (a) Pasto, D. J.; Krasnansky, R.; Zercher, C. J. J. Org. Chem.
1987, 52, 3062. (b) Curran, D. P.; Kim, D.; Liu, H. T.; Shen,
W. J. Am. Chem. Soc. 1988, 110, 5900–5902.
(b) Becke, A. D. J. Phys. Chem. 1993, 98, 5648–5652.
(c) Becke, A. D. Phys. Rev. A 1998, 38, 3098–3100. (d) Becke,
A. D. Phys. Rev. B 1998, 37, 785–789.
15. (a) Coolidge, M. B.; Borden, W. T. J. Am. Chem. Soc. 1988,
110, 2298–2299. (b) Ibrahim, M. R.; Jorgensen, W. L. J. Am.
Chem. Soc. 1989, 111, 819–824.
18. Scott, A. P.; Radom, L. J. Phys. Chem. 1996, 100,
16502–16513.
19. (a) Marshall, P.; Schwartz, M. J. Phys. Chem. A 1997, 101,
2906–2909. (b) DiLabio, G. A.; Pratt, D. A.; LoFaro, A. D.;
Wright, J. S. J. Phys. Chem. A 1999, 103, 1653–1661.
20. (a) Devin, P.; Fensterbank, L.; Malacria, M. J. Org. Chem.
1998, 63, 6764–6765. (b) Bogen, S.; Gulea, M.; Fensterbank,
L.; Malacria, M. J. Org. Chem. 1999, 64, 4920–4925.
16. Feray, L.; Kuznetsov, N.; Renaud, P. In 3.6 Hydrogen Atom
Abstraction; Renaud, P., Sibi, M. P., Eds.; Radical in Organic
Synthesis: Application; Wiley-VCH: Weinheim, 2001; Vol. 2;
p 246.
17. (a) Becke, A. D. J. Phys. Chem. 1993, 98, 1372–1377.