peratures with extended reaction time (Method B, entry 2)7
provided the desired aldol adduct in improved regioselection,
but the diastereoselectivity was diminished and conversion
was low.8,9 The improved regioselection indicated that the
thermodynamic enolization conditions resulted in equilibra-
tion to the desired R-enolate, while the diminished diaster-
eoselectivity implied that significant quantities of the un-
desired R-(Z) enolate were also formed under these condi-
tions. However, use of the aldehyde reaction partner as the
limiting reagent resulted in formation of the desired aldol
adduct with good regio- and diastereoselectivity (entry 3).
We thus conclude that whereas enolization of ketone 2 under
thermodynamic conditions (Method B) results in the forma-
tion of a mixture of enolates, the desired R-(E) enolate is
the more reactive.10
Scheme 2. Regio- and Stereochemical Issues
To address the issue of aldehyde facial selectivity, two
model aldehydes were prepared (Scheme 3). Aldehyde 5 was
Scheme 3. Syntheses of Model Aldehydes
enolate will lead to the undesired R,â-syn adducts; (c)
aldehyde facial selectiVity-anti-Felkin selectivity will lead
to the desired â,γ-syn adducts, whereas Felkin selectivity
will lead to the undesired â,γ-anti adducts.
Ketone 2 was selected as an appropriate model.4 To
address the issues of enolization selectivity, the aldol
additions of boron enolate derived from ketone 2 with
dihydrocinnamaldehyde was investigated (Table 1). Dicy-
Table 1. Optimization of Enolization Selectivity
available using the auxiliary-controlled addition of a Sn(II)-
glycolate enolate, which provided adduct 6 with moderate
diastereoselection.11 Aldehyde 7 was prepared from com-
mercially available 2,3-O-isopropylidene-D-erythronolactone.
When the boron enolate derived from ketone 2 was added
to these aldehydes under the optimized conditions (Table 1,
entry 2), aldol diastereoselectivity was found to be strongly
dependent on the choice of protecting groups (Scheme 4).
In the case of aldehyde 5, the undesired Felkin adduct 8b
was obtained as the major product in 79:21 diastereoselection
enoliztion
methoda
equiv
RCHO
conv
[%]
ratio
3:4
anti:syn
(3)
entry
1
2
3
A
B
B
1.5
1.5
0.5
100b
60b
93c
36:64
90:10
>97:3
90:10
80:20
92:8
a Method A: 1.5 equiv of Cy2BCl, 1.8 equiv of NEt3, Et2O, -78 °C, 1
h. Method B: 1.1 equiv of Cy2BCl, 2.0 equiv of EtNMe2, pentane, 0 °C to
room temperature, 15 h. b Based on ketone. c Isolated yield based on
aldehyde.
(7) Galobardes, M.; Gasco´n, M.; Mena, M.; Romea, P.; Urp´ı, F.;
Vilarassa, J. Org. Lett. 2000, 2, 2599.
(8) Adduct 4 was formed as a single diastereomer, consistent with the
formation of the (Z) enolate, which is expected for R′-siloxy ketones: Murga,
J.; Falomir, E.; Carda, M.; Gonzalez, F.; Marco, J. A. Org. Lett. 2001, 3,
901.
(9) See Supporting Information for experiments that support all stereo-
chemical assignments.
(10) Evans, D. A.; Nelson, J. V.; Vogel, E.; Taber, T. R. J. Am. Chem.
Soc. 1981, 103, 3099. Similarly, it has been noted that (E) crotylboronates
react faster with aldehydes than the corresponding (Z) isomers: Roush, W.
R.; Adam, M. A.; Walts, A. E.; Harris, D. J. J. Am. Chem. Soc. 1986, 108,
3422.
(11) Evans, D. A.; Gage, J. R.; Leighton, J. L.; Kim, A. S. J. Org. Chem.
1992, 1961.
clohexylchloroborane was employed, as this reagent gener-
ally leads to high selectivity for the desired (E) enolate.5
Enolization at low temperatures (Method A, entry 1)6 led to
low levels of regioselectivity. Enolization at elevated tem-
(4) See Supporting Information for details of the synthesis of 2.
(5) Goodman, J. M.; Paterson, I. Tetrahedron Lett. 1992, 33, 7233.
(6) Marco, J. A.; Carda, M.; Falomir, E.; Palomo, C.; Oiarbide, M.; Ortiz,
J. A.; Linden, A. Tetrahedron Lett. 1999, 40, 1065.
3332
Org. Lett., Vol. 7, No. 15, 2005