DeAngelis et al.
diastereoselectivity. Rh2Piv4 is the optimal catalyst for the
transformations, and it is essential to carry out the reactions at
low temperature (-78 °C) in order to achieve high yield and
diastereoselectivity. A model is set forth to explain the diaste-
reoselectivity of the cycloaddition based on an endo transition
state.
FIGURE 1. Proposed model to explain diastereoselectivity.
Experimental Section
SCHEME 3. “Crossover” Experiment Demonstrates That
the 1,3-Dipolar Cycloaddition Had Taken Place at -78 °C,
and Not While the Reaction Warmed to Room Temperature
Representative Procedures for Dioxolane Formation: 2â,5â-
Di(4-fluorophenyl)-4â-ethoxycarbonyl-4R-ethyl-1,3-dioxolane (4g).
A dry round-bottomed flask was charged with Rh2(piv)4 (2 mg,
0.004 mmol) and the flask was evacuated and filled with nitrogen.
Anhydrous CH2Cl2 (5 mL) and 4-fluorobenzaldehyde (332 mg, 2.68
mmol) were added, and the flask was cooled by a bath of dry ice/
acetone. Ethyl 2-diazobutanoate (95 mg, 0.67 mmol) was dissolved
in 3 mL of anhydrous CH2Cl2 and added to the reaction mixture
via syringe pump at a rate of 1 mL/h. Following addition, the
reaction mixture was allowed to warm to room temperature and
mesitylene (80 mg, 0.67 mmol) was added. 1H NMR analysis was
then performed and the diastereomer ratio was determined to be
>95:5. The solvent was then removed and the residue was
chromatographed on silica gel with 1-5% ethyl acetate/hexanes
as the eluent to give 168 mg (70%) of the title compound as a
clear oil. A similar experiment gave the title compound in 61%
yield. 1H NMR (CDCl3, 400 MHz) δ 7.85 (m, 2H), 7.35 (m, 2H),
7.14 (m, 2H), 7.03 (m, 2H), 6.13 (s, 1H), 4.99 (s, 1H), 3.77-3.88
(m, 1H), 3.52-3.62 (m, 1H), 2.20-2.31 (m, 1H), 1.88-2.01 (m,
1H), 1.06 (t, J ) 7.2 Hz, 3H), 0.94 (t, J ) 7.3 Hz, 3H); 13C NMR
1
(CDCl3, 100 MHz) δ 170.5 (u), 163.9 (u) (d, J(CF) ) 246 Hz),
163.0 (u) (1J(CF) ) 246 Hz), 132.4 (u) (d, 4J(CF) ) 3 Hz), 131.2
(u) (d, 4J(CF) ) 3 Hz), 129.9 (dn) (d, 3J(CF) ) 8 Hz), 128.2 (dn)
(d, 3J(CF) ) 8 Hz), 115.5 (dn) (d, 2J(CF) ) 21 Hz), 115.2 (dn) (d,
2J(CF) ) 22 Hz), 103.5 (dn), 89.0 (u), 87.1 (dn), 61.1 (u), 29.0
(u), 13.7 (dn), 8.8 (dn); IR (neat, cm-1) 2967, 1736, 1607, 1511,
1346, 1225, 1054, 1012, 832; HRMS-ESI m/z [M + Na] calcd for
C20H20O4F2Na 385.1227, found 385.1241.
Carbonyl ylides derived from R-alkyl-R-diazoesters are
unusual because they form and react productively at -78 °C.
Under similar conditions (-78 °C) to those reported in Table
2, ethyl diazoacetate and benzaldehyde give only diethyl maleate
and diethyl fumerate. The sense of diastereoselectivity is also
unusual. Endo transition states via conformer B (Figure 1) have
been proposed for the reactions of carbonyl ylides derived from
methyl R-trimethylsilyl-R-diazoacetate,2g methyl R-trifluorom-
ethyl-R-diazoacetate,2d and methyl R-phenyl-R-diazoacetate.2f
The reason that the reactions of carbonyl ylides derived from
R-alkyl-R-diazoesters apparently arise from conformer A may
be related to steric considerations. An empirical observation
from Table 2 is that the diastereoselectivity is highest when
the alkyl substitutent is small and the ester substitutent is large.
For example, the ethyl ester 4c (R ) Me) is produced with
94:6 dr, whereas the ethyl esters 4f (R ) Et) and 4r(Et) (R )
Bn) are produced in 90:10 and 86:14 dr, respectively. Thus,
the increase in the size of the R-alkyl substitutent (Bn > Et >
Me) is accompanied by a decrease in diastereoselectivity (Me
> Et > Bn). This trend can be countered by increasing the size
of the ester substitutent. Thus, t-Bu ester 4r(t-Bu) (R ) Bn) is
formed with 94:6 dr. Therefore, these observations are consistent
with prior observations in the literature that carbonyl ylides with
relatively large R-alkyl substitutents (Ph,2f TMS,2g CF32d) are
controlled by endo approach via conformer B.
2â,5â-Di(4-nitrophenyl)-4â-ethoxycarbonyl-4R-benzyl-1,3-di-
oxolane (4r). A dry round-bottomed flask was charged with Rh2-
(piv)4 (1 mg, 0.003 mmol) and the flask was evacuated and filled
with nitrogen. Anhydrous CH2Cl2 (5 mL) and 4-nitrobenzaldehyde
(302 mg, 2.00 mmol) were added, and the flask was cooled by a
bath of dry ice/acetone. Ethyl 2-diazohydrocinnamate (102 mg, 0.50
mmol) was dissolved in 3 mL of anhydrous CH2Cl2 and added to
the reaction mixture via syringe pump at a rate of 1 mL/h. Following
addition, the reaction mixture was allowed to warm to room
1
temperature and mesitylene (60 mg, 0.50 mmol) was added. H
NMR analysis was then performed and the diastereomer ratio was
determined to be 84:16. The solvent was then removed and the
residue was chromatographed on silica gel with 5-8% ethyl acetate/
hexanes as the eluent to give 139 mg (59%) of the title compound
as a white solid, mp 55-59 °C. Also isolated was 26 mg (11%) of
4r (minor diastereomer) as a white solid. A similar experiment gave
4r in 54% yield.
Spectoscopic data for 4r: 1H NMR (CDCl3, 400 MHz, δ): 8.37
(d, J ) 9.0 Hz, 2H), 8.28 (d, J ) 9.0 Hz, 2H), 7.97 (d, J ) 9.0 Hz,
2H), 7.62 (d, J ) 8.9 Hz, 2H), 7.32-7.46 (m, 5H), 5.98 (s, 1H),
5.27 (s, 1H), 3.65 (m, 1H), 3.47 (m, 1H), 3.45 (d, J ) 14.7 Hz,
1H), 3.30 (d, J ) 14.7 Hz, 1H), 0.87 (t, J ) 7.3 Hz, 3H); 13C
NMR (CDCl3, 100 MHz) δ 169.5 (u), 148.9 (u), 148.2 (u), 142.6
(u), 141.9 (u), 134.5 (u), 130.9 (dn), 130.8 (dn), 128.6 (dn), 127.5
(dn), 127.4 (dn), 123.7 (dn), 123.6 (dn), 103.2 (dn), 88.2 (u), 84.8
(dn), 61.5 (u), 40.2 (u), 13.7 (dn); IR (neat, cm-1) 2364, 2343,
1736, 1520, 1497, 1439, 1345, 1291, 1200, 1085, 1013, 851, 746,
697. Anal. Calcd for C25H22N2O8: C 62.76, H 4.63, N 5.86, O
26.75. Found: C 62.77, H 4.71, N 5.74, O 26.49.
In summary, general conditions are described for the Rh-
catalyzed generation of carbonyl ylides from R-diazoesters that
possess â-hydrogens. Subsequent cycloaddition with aromatic
aldehydes gives tetrasubstituted dioxolanes with unusually high
1438 J. Org. Chem., Vol. 73, No. 4, 2008