Communications
using 30 mol% of chiral Lewis acid 11a gave 6 in high yield
benzoate products 6 and 7 are more sensitive, which causes a
reduction in the yield and contamination by the resulting
enones. In reactions using substrates 17 and 18, which contain
a bulky mesityl and a benzyl group, respectively, the yields
and ee values were also superior to those of the parent
benzoate substrate (Table 2, compare entries 3and 4 with
entry 1). A dramatic improvement in the selectivity was
observed with substrate 19 compared to that with the parent
benzoate substrate (Table 2, entry 5), which suggests that
steric effects play an important role in the conjugate radical
addition. With substrates 20 and 21, the yields and ee values
were only slightly better than those obtained with the parent
benzoate substrate (Table 2, entries 6 and 7). It is noteworthy
that reactions with the bulky tert-butyl radical consistently
proceeded with higher selectivity than did those with the less
bulky isopropyl radical. This trend parallels that observed
previously in the synthesis of anti-propionate aldol deriva-
tives.[5]
Three cyclic ketones were investigated, which differed in
ring size (5–7), in an effort to broaden the scope of the aldol
products available (Table 3). For these reactions the p-
methoxy benzoate was used because it had previously given
the best yields and the most stable products (Table 2).
Enantioselectivity was poor with the five-membered sub-
strate 34 when catalyst 13 was used (Table 3, entry 1).
Surprisingly, good selectivity and excellent yields were
achieved for both isopropyl and tert-butyl radical addition
using the alternative salen catalyst 11b that had a triflate
counterion (Table 3, entry 2). As discussed earlier, reactions
were efficient with six-membered ring substrate 16. The
enantioselectivity was high for the seven-membered ring 35,
but yields were poor when 30 mol% of catalyst 13 was used
(Table 3, entry 4). Increasing the catalyst loading to 50 mol%
gave the desired product in excellent yield and high selectivity
(Table 3, entry 5). A rationale for this difference in reactivity
with respect to catalyst loading is
and diastereoselectivity (Table 1, entry 5). The ee value for
the product was much higher than that observed with
scandium-based Lewis acids. The reaction with chiral salen
12, which contains a cyclohexanediamine unit, was also facile
and gave enhanced enantioselectivity (Table 1, entry 6).
Increasing the catalyst loading to 100 mol% did not improve
either the chemical yield or the selectivity (Table 1, entry 7).
Counterion exchange was carried out in an effort to increase
the Lewis acidity of the aluminum center. Catalysts 13–15
were prepared by a rapid exchange of counterions between
commercially available 12 and the corresponding silver salts
in dichloromethane at room temperature. The chiral Lewis
acid with a triflate counterion (13) gave the highest enantio-
selectivity (Table 1, entry 8). Further changes to the counter-
ion to give Lewis acids 14 and 15 did not lead to further
improvement (Table 1, entries 9 and 10). The addition of the
bulkier tert-butyl radical gave enhanced enantioselectivity
(Table 1, compare entries 8 and 11). The above experiments
demonstrate that cyclic ketone aldols can be synthesized by
conjugate radical addition in which two stereocenters are
installed in a single step. The aldol products are formed in
excellent yields, and the levels of diastereoselectivity and
enantioselectivity are high.
In an effort to tune the reactivity and selectivity, a series of
substrates were synthesized that contained different acyl
groups, and their effectiveness toward radical addition was
evaluated (Table 2). The optimal chiral salen catalyst 13 was
used, and additions were carried out using both isopropyl and
tert-butyl radicals. Compound 16, which contains an electron-
rich methoxy benzoate group, showed higher selectivity than
the parent benzoate substrate (Table 2, compare entries 1 and
2) for both isopropyl and tert-butyl radical additions. The
yields were much higher because the methoxy benzoate
products 22 and 23 are more resistant to b-elimination; the
not clear. The diastereoselectivity
in all of the above reactions was
Table 2: Effect of acyl groups on the reactivity and selectivity.
excellent (d.r. ꢀ 99:1). It is note-
worthy that isopropyl radical addi-
tion was most enantioselective with
the
seven-membered
ketone
(Table 3, compare entry 4 with
entries 2 and 3). The same trend
was not present for tert-butyl rad-
ical addition, where the six-mem-
bered ring gave slightly higher
Entry
R1(starting material)
R=Isopropyl[a]
R=tert-Butyl[a]
selectivity
(Table 3,
compare
prod.
yield [%][b]
ee [%][c]
prod.
yield [%][b]
ee [%][c]
entries 3and 4). These experiments
indicate that radical conjugate
additions can provide access to
aldols from cyclic ketones of vari-
ous ring size.[11]
The scope of radicals that can
be used was also evaluated
(Table 4). For these experiments
16 was used as the substrate and
30 mol% of 13 was used as the
catalyst. In general, primary, sec-
1
2
3
4
5
6
7
phenyl (5)
6
80
95
85
90
85
85
86
77
84
82
93
94
84
83
7
78
90
70
80
75
80
80
85
95
90
91
97
91
94
4-methoxyphenyl (16)
2,4,6-trimethylphenyl (17)
benzyl (18)
2,4,6-trimethylbenzyl (19)
1-naphthyl (20)
22
24
26
28
30
32
23
25
27
29
31
33
2-naphthyl (21)
[a] For detailed reaction conditions see the Supporting Information. The diastereoselectivity was
determined by NMR spectroscopy. For reactions with 99:1 selectivity, the minor isomer could not be
detected. [b] Yield of isolated products after purification by column chromatography. [c] The ee values
were determined by HPLC on a chiral stationary phase.
9232
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 9231 –9234
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
Take advantage of blue reference links
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
Sibi, Mukund P.
