2
494
J. Kaldun et al. / Tetrahedron Letters 57 (2016) 2492–2495
Table 1
exceeded 0.33 in the reaction mixture under these conditions. By
keeping this ratio, the amount of 8 can be reduced to 2.5 mol %
Optimization of the reaction conditions for oxazaborolidine 8a
(
addition time of 1a: 120 min), giving (R)-2a in high 98% yield
and excellent 98% ee (entry 6). Increasing the catalyst loading to
0 mol % (addition time of 1a: 30 min) or dilution of the reaction
from 0.25 M to 0.13 M exerted no effect on the enantioselectivity
entries 7 and 8). Finally, variation of the temperature revealed,
again in good agreement with other CBS reductions, that higher
or lower temperatures cause a loss in enantioselectivity and/or
yield (entries 9 and 10).
A short study on the scope of our new catalyst 8 is given in
Table 2, entries 1–7. Under optimized conditions (see Table 1, entry
4), the aryl alkyl ketones 1a–d were smoothly reduced in the pres-
ence of 8 delivering the secondary alcohols (R)-2a–d in high 92–
99% yield and high to excellent 93–98% ee (entries 1–4). 4-Acetyl-
pyridine (1e), which is one of the more problematic aryl methyl
ketones due to the strongly Lewis basic and deactivating nitrogen
atom in the heteroaromatic ring, was also reduced in high 94% ee
O
OH
*
8
(cat.)
+
3 2
BH •SMe
Ph
Me
tolue ne
Ph
Me
1
1a
(R)-2a
(
Entry 8 (mol
Temp
(°C)
Addition time of 1a
(min)
Yield
(%)
ee
(%)
7g
b
c
%
)
1
2
3
4
5
6
7
8
9
1
5
5
5
5
20
20
20
20
20
20
20
20
40
10
0
0
93
99
80
95
95
98
89
81
94
15
91
82
97
97
97
98
97
97
96
92
d
30
60
90
120
30
60
60
60
5
2.5
10
e
5
5
5
0
a
A solution of 1a (500 lmol) in toluene (1 mL) was continuously added over the
indicated time to a solution of catalyst 8, prepared in situ, and BH
in toluene (1 mL).
(
98% yield, entry 5). This result is clearly superior to that reached
3
ꢀSMe
2
(1.0 equiv)
with the well-known CBS catalyst (52% ee under similar condi-
b
20,21
Isolated yield.
Determined by HPLC on chiral phase.
Reaction in THF as the solvent.
Reaction at half concentration (2 + 2 mL toluene).
tions).
As examples for dialkyl ketones, pinacolone (1f) and
c
benzylacetone (1g) were chosen (entries 6 and 7). While the for-
mer one afforded (R)-2f in 97% yield and superb 98% ee, the reduc-
tion of 1g delivered (R)-2g with just 65% ee. The mediocre
enantioselection, presumably caused by the lower steric differenti-
ation between the two aliphatic substituents at the ketone, seems
d
e
The performance of 8 and 9 as chiral catalysts in the enantios-
elective ketone reduction with borane was studied next. The reac-
2
2
to be typical for Oxazaborolidine-type catalysis.
tion conditions were first optimized using catalyst
acetophenone (1a) as the model substrate (Table 1). The oxaz-
aborolidine 8 (5 mol %) was prepared prior to use (vide supra) as
8
and
The borane reduction of 1a in the presence of the tricyclic B-
alkyl catalyst 9 provided the alcohol (R)-2a with good 89% ee
(entry 8). Since 8 and 9 only differ in the atom next to the boron
atom (oxygen vs methylene), the lower stereocontrol of 9 must
be related to it. Either steric or electronic reasons (8: boric ester
vs 9: boronic ester) may account for this. In all reactions, the R-con-
figured alcohols were obtained as the major enantiomers.
a solution in toluene (1 mL) and the achiral reductand BH
3
ꢀSMe
2
(
1.0 equiv) was added. If the ketone 1a (1.0 M in toluene) was
introduced at 20 °C in one batch, the R-configured alcohol (R)-2a
was obtained in 93% yield and acceptable 91% ee (entry 1). Chang-
ing the solvent to THF lowered the asymmetric induction signifi-
cantly (82% ee, entry 2). Continuous addition of 1a over a period
of 30–90 min permitted enhanced levels of stereocontrol (97% ee,
entries 3–5). Thus, as for other CBS reductions, a too large excess
of the ketone in the reaction mixture has to be avoided, in order
to efficiently suppress the non-stereoselective background reac-
Finally, we decided to get some information about the activities
of 8 and 9 relative to the standard CBS catalyst (R)-3 (Scheme 5).
The reduction of acetophenone (1a) under our optimized condi-
tions (see Table 2), in which (R)-3 provided (S)-2a in 91% yield
2
3
and excellent 98% ee, was chosen as the model reaction. Since
catalysts 8 and 9 provided, compared to (R)-3, the enantio-comple-
mentary products with good to high stereocontrol, their relative
reaction rates can be judged from the sense of the asymmetric
induction and the level of the enantiomeric excess in the product,
by using a competition experiment with equimolar amounts of 8/
1
9
tion. Considering yield and reaction time, addition of 1a within
0 min provided the best result. The ratio ketone/catalyst never
6
2
4
Table 2
(R)-3 and 9/(R)-3, respectively. The high 85% ee (S) in the reaction
with 9/(R)-3 reveal that the standard CBS catalyst (R)-3 accelerates
the borane reduction of 1a by far more efficiently than the tricyclic
B-alkyl oxazaborolidine 9. This observation is in contrast to our ini-
tial assumption that the reduced N–B double bond character,
caused by the bowl-like shape of 9, might increase the Lewis acid-
ity/basicity of the boron and nitrogen atom and, thus, enhance the
activity of the catalyst. It might be possible that the coordination of
both reactants to the catalyst and the following internal hydride
transfer is indeed, as anticipated, faster for 9, but this acceleration
is overcompensated by other rate-reducing effects. For example, a
higher dimerization tendency of 9, which would reduce the
amount of active catalyst, or a slower release of the product due
to tighter binding might be a consequence of the increased Lewis
Reduction of prochiral ketones in the presence of 8 and 9a
8
or 9
O
OH
(
5 mol%)
+
BH
3
•SMe
2
R
R'
toluene, 20 °C
R
*
R'
1
(R)-2
0
Yield (%)b
ee (%)c
Entry
Cat.
Ketone
R
R
1d
2
3
8
8
8
8
8
8
8
9
1a
1b
1c
1d
1e
1f
Ph
Ph
4-O
a
4-Py
tBu
CH
Ph
Me
Et
Me
95
99
96
92
98
97
81
96
97
96
f
2
N-Ph
-Indanone
98
4
5
93
94
e
Me
Me
Me
Me
f
6
7
8
98
1g
1a
2
Bn
65
89
a
A solution of the ketone 1 (500
over a period of 60 min to a solution of catalyst 8 or 9 (5 mol %), prepared in situ,
and BH (1.0 equiv) in toluene (1 mL).
lmol) in toluene (1 mL) was continuously added
cat.
(R)-3: 91%, 98% ee (S)
9/(R)-3 (1:1): 99%, 85% ee (S)
8/(R)-3 (1:1): 95%, 22% ee (S)
3
ꢀSMe
2
O
OH
(
5 mol% in sum)
b
Isolated yield.
Determined by HPLC on chiral phase.
See Table 1, entry 4.
c
d
e
f
Ph
Me toluene, 20 °C
Ph
*
Me
1a
2a
2
.0 equiv of BH
3
ꢀSMe
2
were used.
Determined by GC on chiral phase.
Scheme 5. Single catalyst and competition experiments with (R)-3, 8 and 9.