2
G. Villa et al. / Tetrahedron Letters xxx (2014) xxx–xxx
(
1)
coming from the olefin either in acetal 5 or in ketone 6
O
Me CuLi
O
2
1
9
(Scheme 3). We conducted a rapid screening of hydroborating
agents to ascertain whether the steric demand around the boron
center influences the diastereoselectivity. Following this approach,
ref. 1b (ent-1)
ref. 8 (rac-1)
ent- or rac-1
ent- or rac-3
we selected the BH
catecholborane for its promising preliminary results, and disi-
amylborane (Sia BH) for its high steric demand. The results are
ÁSMe
2
complex for its low steric demand,
3
1
7
(
2)
2
O
O
summarized in Table 1.
Me2CuLi
ref. 9
5 steps
ref. 9
O
Acetal 5 was always deprotected to the ketone 1/epi-1 during its
organoborane-catechol mediated reduction presumably due to the
acidic properties of catechol. Reduction of the ketone was always
observed during the hydroboration of 6. As the screened hydrobo-
rating agents are known to react faster with terminal olefins than
ketones, this observation suggests that the hydroboration is
unusually slow due to a large steric demand at C-5. When the
hydroborating agent was used in excess, both the ketone and the
exo-methylene group reacted. For substrate 6, the crude interme-
diate alcohol 8 obtained by the hydroboration/radical-mediated
protonolysis was directly oxidized to the trimethyldecalone 1
before analysis and yield determination. The hydroboration with
O
O
rac-4
rac-2 (WMK)
rac-1
Scheme 1. Previous synthesis of ent-1 and rac-1.
methylene in compound 5,9 synthesized in three steps from
2
0
Wieland–Miescher ketone (2), in which the two quaternary car-
bons are formed (Scheme 2). We have previously examined the
direct transformation of the exocyclic methylene in 5 and 6 to
the target 1, but the stereochemical control was not good enough.
The diastereoselectivity in the hydrogenation using several
BH
3
ÁSMe
2
of both substrates 5 and 6 was performed at 0 °C, leading
9
16
to moderate diastereoselectivities (dr 3–5.5:1) and low to moder-
ated yields. Hydroborations with catecholborane require heating
and afford the desired trimethyldecalone in only 44% yield from
conditions and catalysts never exceeded 3:1 for 1/epi-1. All the
preliminary studies were conducted in the racemic series.
Recently, Renaud and co-workers17 reported a mild radical
6
5
as a 5:1 ratio of diastereomers. Starting from dioxolanyl acetal
, only degradation of the substrate was observed. Hydroborations
procedure for the transformation of organoboranes to alkanes (rad-
ical-mediated protonolysis), in which 4-tert-butylcatechol (TBC), a
well-established radical inhibitor and antioxidant, acts as a source
2
with in situ generated Sia BH were slow and had to be performed
at room temperature, but gave the highest diastereomeric ratios
and yields from both 5 and 6 (Table 1, entries 2 and 5). The best
1
8
of hydrogen atoms. An efficient chain reaction is observed due to
the exceptional reactivity of phenoxyl radicals toward alkylbor-
anes. The reaction has been applied to a wide range of organoboron
derivatives such as B-alkylcatecholboranes, trialkylboranes,
pinacolboronates, and alkylboronic acids. We thought that this
methodology, involving a one-pot chain reduction of alkylboron
compounds with catechols, might allow us to take advantage of
the improved stereocontrol provided by the hydroboration and
O
O
O
1) Borane
) TBC, air,
O
+
2
8
0 °C
5
1
epi-1
avoid the multi-step sequence reported for the transformation of
DMP
9
2
to 1 (via 5).
O
Thus, the challenge was to set the all-cis configuration of the
methyl groups in ketone 1 by direct reduction of an organoborane
1) Borane
) TBC, air,
OH
2
8
0 °C
6
8
O
O
O
O
3
steps
10% HCl
ref. 9
DMP = Dess-Martin periodinane
TBC = 4-tert-butylcatechol
O
ref. 9
O
Scheme 3. One-pot hydroboration/radical protonolysis. Reaction conditions: see
Table 1.
2
5
6
O
H2, Pd-C
ref. 9
O
+
O
5
Table 1
Hydroboration and radical mediated alkylborane protonolysisa of exo-methylene
compounds 5 and 6 to trimethyldecalone 1: racemic series
7
(minor)
epi-7 (major)
Entry
Alkene
Borane
Yield (%)b
1/epi-1
1
2
5
5
6
6
6
BH
Sia
BH
3
2
3
ÁSMe
2
2
20c
80
58
44
70
3:1d
9:1
5.5:1
5:1
H2 (30 atm), cat.
O
O
BH
6
+
e
3
ÁSMe
solvent
ref. 9
e
4
Catecholborane
e
5
Sia
2
BH
6.5:1
1
epi-1
a
All reactions performed as two-step one-pot reaction following the general
procedure reported in the experimental section (footnote 21). Diastereoselectivities
as observed in the 1H NMR reaction mixture, based on characteristic chemical shifts
ofba-carbonyl protons.
cat.
solvent
CH2Cl2
C6H6
1/epi-1
1.5:1
2:1
Pd-C (70%)
Isolated yield of the mixture of diastereomers.
(Ph3)3RhCl (1%)
c
Unoptimized yield.
PtO2 (10%)
CH2Cl2
3:1 (95% yield)
d
Diastereoselectivity lower than the one reported in Ref. 9. This point was not
further investigated due to the low yield of the reaction.
e
Yield and diastereomeric ratio were determined after oxidation of initially
Scheme 2. Approach to trimethyldecalone 1 via catalytic hydrogenation according
to Ref. 9 (racemic series).
formed alcohol using Dess–Martin periodinane (DMP) to 1/epi-1.