Copper-Catalyzed Asymmetric Alcoholysis of Azlactones
A R T I C L E S
Figure 4. Kinetic resolution of rac-1g: experimental values (dots); simulated lines as zeroth-order reaction (krel ) 10.2) and as first-order reaction (krel
) 10.2 and 22).
donating substituent (p-MeO) on R3 (1l) decreased the reactivity
(entries 24-27). Comparable rate was observed with higher (5
mol %) catalyst loadings (entry 27, cf. entry 5; 2 mol %). It is
noteworthy that 93% ee of unreacted substrates was obtained
with 1e and 1h, although krel was only 5.6 and 5.9, which clearly
indicates the higher efficiency of zeroth-order kinetic resolution.
We assume the rate law of the reaction as eq 7; thus, the
first-order plot for alcohol should be fitted.
reaction (krel ) 18) as first-order reaction. Catalytic reaction
often shows zeroth-order dependence in the substrate, for
instance, Ru-catalyzed oxidation,12a Rh-catalyzed hydrogenation,12b
Pd-catalyzed Heck reaction,12d alkali-metal-catalyzed amination
of olefins,12e and needless to say, most cases of enzymatic
reactions. When these reactions are applied for kinetic resolution,
it will be easier to achieve high ee than first-order reactions.
Experimental Section
d[1]
dt
-
) k[1]0[ROH]1
(7)
Typical Procedure for Kinetic Resolution: Alcoholysis of 1b
(Table 1, Entries 3 and 5). In a dried 20 mL Schlenk flask was placed
Cu(OAc)2 (1.9 mg, 0.010 mmol), and then the solid was heated at 80
°C under vacuum for 3 min and the flask filled with Ar. To this were
added (S)-DTBM-SEGPHOS (14.2 mg, 0.012 mmol) and toluene (1.0
mL); then, the mixture was stirred at 45 °C for 1 h. The mixture was
cooled to room temperature, and toluene (1.0 mL) solution of 1b (123.6
mg, 0.500 mmol) and 2-methoxyethanol (80 µL, 1.0 mmol) was added.
The mixture was stirred at 20 °C for 17 h (entry 3) or 52 h (entry 5)
in an incubator. The product 2b and the substrate 1b were isolated by
silica gel column chromatography (elution: hexane/Et2O 10/1, 1/1, and
then 1/2). Entry 3: 1b (86.6 mg, 69.7%, 37.5% ee) and 2b (49.3 mg,
30.3%, 81.5% ee). Entry 5: 1b (52.0 mg, 42.1%, 93.2% ee) and 2b
(89.7 mg, 55.3%, 73.7% ee). 1b (93.2% ee) HPLC (CHIRALCEL OD
(DAICEL), hexane/2-propanol ) 200/1, 0.50 mL/min, TR ) 24.1 min
(minor), TR ) 26.6 min (major)), [R]29D -110.9° (c 4.53, benzene).1H
NMR (400 MHz, CDCl3): δ 1.26 (t, J ) 7.3 Hz, 3H) 1.77 (s, 3H),
4.20-4.30 (m, 2H), 7.50 (dd, J ) 7.3, 8.2 Hz, 2H), 7.60 (t, J ) 7.3
Hz, 1H), 8.03 (d, J ) 8.2 Hz, 2H). 13C NMR (100 MHz, CDCl3): δ
13.9, 20.4, 63.0, 72.9, 125.3, 128.2 (2C), 128.8 (2C), 133.3, 163.1,
165.9, 175.1. Anal. Calcd for C13H13NO4: C, 63.15; H, 5.30; N, 5.67.
Found: C, 63.18; H, 5.33; N 5.69. 2b (81.5% ee) HPLC (CHIRALPAK
AD-H (DAICEL), hexane/2-propanol ) 4/1, 1.0 mL/min, TR ) 8.9
min (major), TR ) 11.5 min (minor)), [R]26D + 9.0° (c 1.00, benzene).
1H NMR (400 MHz, CDCl3): δ 1.26 (t, J ) 7.3 Hz, 3H), 1.87 (s, 3H),
3.29 (s, 3H), 3.54-3.57 (m, 2H), 4.21-4.30 (m, 2H), 4.35 (t, J ) 4.5
Hz, 2H), 7.43 (dd, J ) 7.6, 7.8 Hz, 2H), 7.51 (t, J ) 7.6 Hz, 1H), 7.55
(brs, 1H), 7.80 (d, J ) 7.8 Hz, 2H). 13C NMR (100 MHz, CDCl3): δ
14.0, 21.1, 59.0, 62.9, 63.2, 65.4, 70.1, 127.2 (2C), 128.7 (2C), 132.0,
133.5, 166.0, 168.8, 168.9. Anal. Calcd for C16H21NO6: C, 59.43; H,
6.55; N, 4.33. Found: C, 59.30; H, 6.66; N, 4.29.
The reaction of 1b with 0.6 equiv of alcohol for this plot was
carried out by a sampling experiment from the same batch. The
result showed good fit with first-order plot over 90% conversion
based on alcohol (Figure S-3 in Supporting Information), which
suggests eq 7 is appropriate for this reaction. The same plots
were made for the data in Table 1 and showed reasonable fit,
even though the reactions were carried out with different
batches.22 Presumably, the chiral recognition of the two enan-
tiomers of the azalactone occurs in the reversible formation of
an azlactone-chiral Cu-catalyst complex and the attack of the
alcohol on this chiral Lewis acid-substrate complex is the rate-
determining step. On the other hand, a preliminary mechanistic
study for the reaction with spectroscopic and kinetics experi-
ments suggest the possibility of a somewhat more complicated
mechanism than our expectation: the coordination mode may
be different between the fast-reacting substrate and the slow-
reacting substrate (O-O chelate and O-N chelate). Further
investigation for a detailed mechanism is in progress.
Conclusion
The difference of resolution efficiency between zeroth-, first-,
and second-order reaction in kinetic resolution was discussed.
Higher efficiency of zeroth-order kinetic resolution was graphi-
cally visualized and demonstrated by a newly developed copper-
catalyzed alcoholysis reaction of azlactones. Typically, a reaction
which has only low selectivity factor (krel ) 5.6, Table 1, entry
11) as zeroth-order reaction is equivalent to a highly selective
Acknowledgment. We thank Takasago International Corpo-
ration for providing DTBM-SEGPHOS. This work was sup-
ported by a Grant-in-Aid for Scientific Research on Priority
Areas (NO. 16033204, Reaction Control of Dynamic
(22) First-order plot (ln([ROH]/[ROH]0) ) -kt) for 1b (Table 1, entries 2-5),
1f (entries 12-15), and 1g (entries 16-19) were well fitted below 50%
convn (25% convn based on alcohol), 1l (entries 24-27) was moderately
fitted, but 1c (entries 6-9) was not fitted (see Supporting Information Figure
S-4).
9
J. AM. CHEM. SOC. VOL. 128, NO. 13, 2006 4485