Synthesis of a Schiff's base chiral ligand with a trifluoromethyl carbinol moiety

Article abstract of DOI:10.1016/j.tetasy.2006.08.023

A Schiff's base ligand with a trifluoromethyl carbinol moiety was synthesized, and a structure of the active complex for asymmetric induction on the reaction of diethylzinc with aldehydes is proposed.

Full text of DOI:10.1016/j.tetasy.2006.08.023

Tetrahedron: Asymmetry 17 (2006) 2654–2658
Synthesis of a Schiff’s base chiral ligand with a
trifluoromethyl carbinol moiety
Yasser Samir Sokeirik, Masaaki Omote, Kazuyuki Sato,
Itsumaro Kumadaki and Akira Ando*
Faculty of Pharmaceutical Sciences, 45-1, Nagaotoge-cho, Hirakata 573-0101, Japan
Received 11 August 2006; accepted 30 August 2006
Available online 24 October 2006
Abstract—A Schiff’s base ligand with a trifluoromethyl carbinol moiety was synthesized, and a structure of the active complex for
asymmetric induction on the reaction of diethylzinc with aldehydes is proposed.
Ó 2006 Published by Elsevier Ltd.
1. Introduction
2. Results and discussion
Chiral Schiff bases have been used as a unique class of
chiral ligands, and their metal complexes are used for
asymmetric Diels–Alder reactions, cyanosilylation of alde-
hydes, cyclopropanations, and so on.1 On the other hand,
we have reported a new axially dissymmetric ligand with
two trifluoromethyl (CF3) carbinol moieties.2 A CF3 group
exhibits a large steric effect on an adjacent center,3 and the
hydroxy group of a CF3 carbinol moiety could be superior
to other aliphatic hydroxy groups in binding to metals due
to its higher acidity and stability against elimination, race-
mization, or oxidation.2
The synthesis of 1a and its related compounds was accom-
plished according to Scheme 1.
HO
NO2
NO2
NO2
O
H
Br
(a)
(b)
CF3
CF3
2
3
4 (S)
NH2HO
H
(c)
(d)
CF3
1a
Our new project was to synthesize another asymmetric tri-
dentate Schiff’s base ligand 1a, which has one stereogenic
center based on a trifluoromethyl carbinol moiety, and
examine the activity as a chiral ligand for asymmetric
reactions (Fig. 1).
5
Scheme 1. Synthesis of 1a. Reagents and conditions: (a) n-BuLi,
CF3COOC2H5, THF, ꢀ100 °C; (b) Catecholborane, (R)-CBS catalyst
10 mol %, toluene, ꢀ60 °C; (c) H2/Pd–C, MeOH; (d) salicylaldehyde
derivatives, MgSO4, dry EtOH.
H
OH
HO
F3C
The synthesis started with commercially available o-
bromonitrobenzene 2. The CBS reduction of 3 to 4
proceeded smoothly to give up to 97% ee. Amino alcohol
5 obtained by the reduction was recrystallized from toluene
to give enantiomerically pure 5. The absolute configuration
was determined to be (S) by the deamination of 5 to the
alcohol and comparison of the sign already rotation with
that previously reported.4 The condensation of 5 with
salicylaldehyde afforded the desired Schiff bases 1a.
N
1a
Figure 1. New Schiff’s base ligand with a trifluoromethyl carbinol moiety.
*
Corresponding author. Tel.: +81 72 866 3140; fax: +81 72 850 7020;
e-mail: aando@pharm.setsunan.ac.jp
0957-4166/$ - see front matter Ó 2006 Published by Elsevier Ltd.
doi:10.1016/j.tetasy.2006.08.023

Y. S. Sokeirik et al. / Tetrahedron: Asymmetry 17 (2006) 2654–2658
2655
To evaluate the catalytic activity of 1a as a ligand, we
examined the typical addition reaction of diethylzinc to
an aldehyde, as shown in Scheme 2.
Next, we examined the activity of ligand 1a with various
types of aldehydes. The results are shown in Table 2.
Ligand 1a showed moderate to high ee for aromatic alde-
hydes (entries 1–6), while low ee for an aliphatic aldehyde
(entry 7). The ee of this reaction is comparable to that
using our previous axially dissymmetric ligands,2 while this
needs no titanium tetraisopropoxide. The ee is better than
that of SALEN ligand, 10 mol % of which was used to give
77% ee.6
Et2Zn (2.4 eq)
CHO
OH
1a (5 mol%)
H
Solv.
Yield ee
Toluene 97% 86%
Hexane 99% 85%
We then became interested in the mechanism of the reac-
tion catalyzed by ligand 1a. First, we examined whether
chiral amplification would be observed or not, when 1a
with low ee was used. Plotting the ee of the product versus
the ee of 1a showed a non-linear relationship with a mod-
erate enantiomeric amplification (25% ee of the ligand gives
46% ee of the product) (Fig. 2). This suggests that the
ligand might not work in a simple monomeric form but
in a dimeric form, as Noyori’s model of enantiomeric
amplification.7
Scheme 2. Asymmetric addition of diethylzinc to benzaldehyde.
The reaction proceeded in non-polar solvents, but did not
in polar solvents such as THF, diethyl ether, or methylene
chloride. Hexane took a longer reaction time than toluene,
but gave a cleaner reaction and proved much easier to be
evaporated than toluene.
Next, we have introduced a few substituents onto the sali-
cylaldehyde part, and their activities for the addition reac-
tion of diethylzinc were then examined (see Table 1).
100
80
60
40
20
0
Table 1. Effect of substituents on the activity
Ligand
R1
R2
Yield (%)
ee
1a
1b
1c
1d
1e
1f
H
H
t-Bu
H
H
t-Bu
H
87
86
—
—
Racemic
83
40
t-Bu
t-Bu
CH3
H
NR
NR
20
92
78
H
R1
H
3
0
25
50
75
100
OH
R O
CF3
e.e.ligand (%)
N
R2
Figure 2. Examination of chiral amplification.
1a-e R3= H
1f R3 = CH3
Katagiri et al. have synthesized a series of fluorinated ami-
no alcohols and evaluated activities as chiral ligands using
addition reactions of diethylzinc. They proposed that the
ligands might act in an aggregated form.8 To examine
whether this was true for our ligand, we plotted the effect
of the concentration of the ligand on the ee and the yield
of the product (Fig. 3). The concentration of 1a has a slight
effect on the ee of the product, which is different to Kata-
giri’s findings. We think that our catalyst must work as a
well defined dimeric complex.
The substituents ortho to the phenolic hydroxy group led to
a loss of the catalytic activity (1b, 1c, or 1d). No reaction
was observed for 1b or 1c from 0 °C to room temperature,
while in the case of 1d, a slow reaction proceeded at room
temperature leading to a racemic product in a low yield.
Substitution at the para-position of the hydroxy group as
in 1e did not show any decrease in activity (83% ee).
Table 2. Reaction of various aldehydes
Entry
Aldehyde
Solvent
Time (h)
Yield (%a)
eeb (abs. config.c)
1
2
3
4
5
6
7
p-Cl–C6H4CHO
o-Cl–C6H4CHO
o-F–C6H4CHO
p-CH3–C6H4CHO
p-(CH3)2CH–C6H4CHO
p-CH3OCO–C6H4CHO
C6H5CH2CH2CHO
Hexane
Hexane
Hexane
Hexane
Hexane
Hexane–toluene
Hexane
24
36
24
36
36
24
48
91
83
86
96
83
88
76
81 (R)
77 (R)
89 (R)
77 (R)
78 (R)
75 (R)
12 (R)
a Isolated yields.
b Determined by chiral GC.
c Determined by the sign of rotation.5

2656
Y. S. Sokeirik et al. / Tetrahedron: Asymmetry 17 (2006) 2654–2658
form a complex.9 These results are consistent with Katag-
100
80
60
40
20
0
iri’s report, which suggests coordination of the trifluorom-
ethyl carbinol oxygen with zinc. This complex must be the
active complex, since it was the only complex detected by
13C and 19F NMR. The asymmetric induction by 1f seems
to support this speculation.
The effect of
ligand mol%
on the
chemical yield
of the product
The effect of
ligand mol%
on the ee of
the product
Furthermore, the above speculation was supported by the
following observations. The zinc complex from (S)-1a of
25% ee showed two peaks in 19F NMR at 10.4 and
10.7 ppm in a ratio of 1:1.44. The former peak must be
attributed to the homodimeric zinc complex, Zn((S)-1a)2
or Zn((R)-1a)2, and the latter the heterodimeric complex,
Zn((R)-1a)((S)-1a). If the former were much more stable
than the latter, only the former peak would be observed.
If the latter were more stable, the peak ratio would be
1:3. The peak ratio 1:1.44 means that about 60% of the
heterodimer was formed. The mixture of 25% ee contains
37.5% of (R)-1a, and a half of heterodimeric complex,
30% is this isomer. Thus, 30% of (R)-1a is used as the hete-
rodimeric complex and 7.5% as in homodimeric complex,
showing that the heterodimeric complex is more stable
than the homodimeric complex. The complex formed from
racemic 1a showed the latter peak with a trace of the for-
mer. The heterodimeric complex accelerates the reaction
less than the homodimeric complex. We believe this is the
reason why a small enantiomeric amplification was ob-
served when 25% ee of (S)-1a was used.
0
25
50
75
Ligand mol%
Figure 3. Effect of concentration of ligand.
To estimate the structure of our complex, we examined a
high resolution MS/ESI for the complex formed from 1a
and diethylzinc. Interestingly, we were able observe a peak
that might be due to one of two structures illustrated in
Figure 4 with the peak of free 1a.
H
H
O
O
O
O
O
Zn
Zn
N
O
N
N
HO
OH
N
F3C
CF3
F3C
CF3
m/z=653.085 (M+I)
Complex B
m/z=653.085 (M+I)
Complex A
3. Conclusion
Figure 4. Structures of zinc complex of 1a estimated by MS/ESI.
In conclusion, we have synthesized a new Schiff base ligand
with a chiral trifluoromethyl carbinol moiety. The enantio-
meric excess of the reaction of diethylzinc with an aldehyde
in the presence of this ligand seems to be the highest among
the ligands with iminoalcohol moieties. This ligand does
not require titanium tetraisopropoxide, as in the case of BI-
NOL. We were able to estimate the active complex of this
ligand to contain two moles of chiral ligand and one zinc
atom. This would help understanding of the reactions cat-
alyzed by chiral ligands and open the door for many appli-
cations of this type of simple Schiff’s base ligand.
The 13C NMR indicated a big shift for the carbinol carbon
from 67 to 82 ppm indicating that it was strongly affected
by complex formation, while a small shift was observed
for both the phenolic and imino carbons (162 to 168 and
164 to 172 ppm, respectively). These results suggest that
zinc is strongly bound to the carbinol oxygens and is
weakly coordinated to phenolic ones, namely, the complex
must be complex A. The 19F NMR showed a shift from
15.2 to 10.2 ppm for the CF3 fluorine by the complex for-
mation with diethylzinc. To confirm the formation of this
complex, MS/ESI of the complex of 1f derived from o-anis-
aldehyde was also examined. We could identify a similar
complex (see Fig. 5).
4. Experimental
4.1. Synthesis of (S)-1-(2-aminophenyl)-2,2,2-trifluoro-
ethanol 5
H3C
CH3
O
4.1.1. 2,2,2-Trifluoro-1-(2-nitrophenyl)ethanone 3. To a
solution of 2-bromonitrobenzene (5.00 g, 24.8 mmol) in
THF (120 mL) at ꢀ100 °C was added n-BuLi in hexane
(1.58 M, 17.2 mL, 27.2 mmol) in 45 min. The resulting mix-
ture was stirred for an additional 1 h at the same tempera-
ture. To this solution was added ethyl trifluoroacetate
(3.24 mL, 27.2 mmol) in 20 min, and the resulting mixture
kept at this temperature under stirring for an additional
6 h. The mixture was quenched with 2 M HCl (50 mL),
and the whole mixture extracted with Et2O. The Et2O layer
was dried over MgSO4, and evaporated under vacuum to
give an oily black mass, which was purified by column
O
Zn
N
N
O
O
CF
3F3C
M/z= 681.116 (M+1)
Figure 5. Structures of zinc complex of 1f estimated by MS/ESI.
From this result, we can conclude that the two trifluoro-
methyl carbinol oxygens coordinate with a zinc atom to

Y. S. Sokeirik et al. / Tetrahedron: Asymmetry 17 (2006) 2654–2658
2657
chromatography (SiO2, 5% Et2O in hexane) to give 3
(2.82 g, 52%). A pale yellow oil. 1H NMR (CDCl3) d:
8.33 (1H, d, J = 8.3 Hz), 7.91 (1H, t, J = 7.3 Hz), 7.84
(1H, dt, J = 7.3, 1.0 Hz), 7.58–7.55 (1H, d, J = 7.9 Hz).
19F NMR (CDCl3) d (from C6H5CF3): ꢀ12.85 (3F, s). IR
(neat) cmꢀ1: 1748.
7.13–7.04 (4H, m), 7.01–6.95 (3H, m), 6.59 (1H, d,
J = 7.8 Hz), 5.38–5.32 (1H, q, J = 6.5 Hz), 2.40 (1H, s, dis-
appeared with D2O). 13C NMR (CDCl3–toluene d8) d:
76.76 (q, J = 32 Hz), 117.56, 118.70, 119.3, 119.45, 125.13
(q, J = 280 Hz), 127.04, 128.38, 130.33, 132.83, 133.84,
137.51, 147.77, 161.33, 164.68. 19F NMR (CDCl3–toluene
d8) d: ꢀ14.51 (3F, d, J = 7.6 Hz). IR (KBr) cmꢀ1: 3412,
1622. MS m/z: 295 (M+). HRMS Calcd for C15H12F3NO2:
295.0820 (M+). Found: 295.0828.
4.1.2. (S)-2,2,2-Trifluoro-1-(2-nitrophenyl)ethanol 4. To a
stirred suspension of (R)-5,5-diphenyl-2-methyl-3,4-pro-
pano-1,3,2-oxazaborolidine (124 mg) in dry toluene
(20 mL) was added catecholborane (1.01 mL), and the
reaction mixture was stirred for 1 h at room temperature.
The mixture was cooled to ꢀ85 °C, and 3 (1.00 g) in dry
toluene (10 mL) was added in 1 h. The mixture was
warmed up to ꢀ65 °C and kept stirring for 12 h. The reac-
tion mixture was quenched with a cold mixture of 10%
NaOH (20 mL) and 30% H2O2 (10 mL) and the whole mix-
ture was stirred for 3 h at room temperature. The organic
phase was separated, and the aqueous phase extracted with
Et2O. The combined organic phase was dried over MgSO4,
and evaporated under vacuum to give the crude product,
which was purified by column chromatography
4.2.2. (S,E)-2,4-Di-tert-butyl-6-((2-(2,2,2-trifluoro-1-hydr-
oxyethyl)phenylimino)methyl)phenol 1b. The reaction time
was 18 h. The product was purified by flash column chro-
matography (SiO2, 5% Et2O in hexane) to give a yellow
20:5
1
oil. Yield 75%. ½aꢁD ¼ ꢀ162:0 (c 1.41, CHCl3). H NMR
(CDCl3) d: 12.83 (1H, s, disappeared with D2O), 8.44
(1H, s), 7.64 (1H, d, J = 7.8 Hz), 7.41 (1H, d, J =
2.4 Hz), 7.36 (1H, dd, J = 7.8, 1.5 Hz), 7.26 (1H, dd,
J = 7.8, 1.5 Hz), 7.15 (1H, d, J = 2.4 Hz), 6.98 (1H, dd,
J = 7.8, 1.0 Hz), 5.35 (1H, q, J = 6.5 Hz), 2.40 (1H, s, dis-
appeared with D2O), 1.56 (9H, s), 1.14 (9H, S). 19F NMR
(CDCl3–toluene d8) d: ꢀ14.51 (3F, d, J = 7.6 Hz). MS m/z:
407 (M+). HRMS Calcd for C23H28F3NO2: 407.2072 (M+).
Found: 407.2078.
(SiO2, 20% Et2O in hexane) to give 4 (898 mg, 89%) as a
15
yellow oil in ee up to 95% by chiral GC. ½aꢁD ¼ þ154:2
(c 0.98, CHCl3). 1H NMR (CDCl3) d: 8.00 (1H, dd,
J = 7.8, 1.5 Hz), 7.97 (1H, d, J = 7.8 Hz), 7.75–7.70 (1H,
m), 7.61–7.54 (1H, m), 6.25 (1H, q, J = 6.5 Hz), 3.55
(1H, s, disappeared with D2O). 19F NMR (CDCl3) d:
ꢀ14.55 (3F, d, J = 7.5 Hz). IR (neat) cmꢀ1: 3550.
4.2.3.
(S,E)-2-tert-Butyl-6-((2-(2,2,2-trifluoro-1-hydroxy-
ethyl)phenylimino)methyl)phenol 1c. The reaction time
was 12 h. The product was purified by flash column chro-
matography (SiO2, 5% Et2O in hexane) to give a yellow
20:5
1
oil. Yield 82%. ½aꢁD ¼ ꢀ180:2 (c 1.03, CHCl3). H NMR
(CDCl3) d: 13.02 (1H, s, disappeared with D2O), 8.45
(1H, s), 7.62 (1H, d, J = 7.8 Hz), 7.40–7.34 (2H, m), 7.28
(1H, t, J = 7.8 Hz), 7.18 (1H, t, J = 7.8 Hz), 7.00 (1H, d,
J = 7.8 Hz), 6.82 (1H, t, J = 7.8 Hz), 5.56 (1H, q,
J = 6.8 Hz), 2.89 (1H, s, disappeared with D2O), 1.34
(9H, s). 19F NMR (CDCl3–toluene d8) d: ꢀ14.69 (3F, d,
J = 6.3 Hz). MS m/z 351 (M+). HRMS Calcd for
C19H20F3NO2: 351.1446 (M+). Found: 351.1447.
4.1.3. (S)-1-(2-Aminophenyl)-2,2,2-trifluoroethanol 5.
A
solution of 4 (500 mg) in MeOH (30 mL) was shaken under
an atmospheric pressure of H2 in the presence of 10% Pd–C
(50 mg) for 4 h. After the catalyst was filtered off, the fil-
trate was concentrated under vacuum to afford a solid
material, which was recrystallized from toluene to give 5
(371 mg, 85%) as colorless needles. Mp 143–145 °C.
20:5
½aꢁD ¼ þ32:7 (c 1.40, CH3OH). 1H NMR (DMSO) d:
7.19–7.17 (1H, d, J = 7.3 Hz), 7.07–7.04 (1H, t,
J = 7.3 Hz), 6.76–6.59 (3H, m), 5.07 (1H, s, disappeared
with D2O), 4.67 (2H, s, disappeared with D2O). 19F
NMR (DMSO) d: ꢀ14.51 (3F, d, J = 7.6 Hz). IR (KBr)
cmꢀ1: 3412, 3340, 3200. MS m/z: 191 (M+). HRMS Calcd
for C8H8F3NO: 191.0558 (M+). Found: 191.0563.
4.2.4.
(S,E)-2-Methyl-6-((2-(2,2,2-trifluoro-1-hydroxy-
ethyl)phenylimino)methyl)phenol 1d. The reaction time
was 4 h. The product was recrystallized from hexane–
Et2O to give yellow crystals. Yield 81%. Mp 95–97 °C.
20:5
1
1
½aꢁD ¼ ꢀ208:3 (c 1.70, CHCl3). H NMR (CDCl3) d: H
NMR (CDCl3) d: 12.69 (1H, s, disappeared with D2O),
8.52 (1H, s), 7.71 (1H, d, J = 7.3 Hz), 7.45 (1H, t,
J = 7.3 Hz), 7.37 (1H, t, J = 7.3 Hz), 7.31–7.23 (2H, m),
7.08 (1H, d, J = 7.3 Hz), 6.88 (1H, t, J = 7.3 Hz), 5.62
(1H, q, J = 6.5 Hz), 2.87 (1H, s, disappeared with D2O),
2.33 (3H, s). 19F NMR (CDCl3) d: ꢀ14.51 (3F, d,
J = 7.6 Hz). MS m/z: 309 (M+). HRMS Calcd for
C16H14F3NO2: 309.0977 (M+). Found: 309.0967.
4.2. General procedure for preparation of chiral Schiff’s base
ligands 1
Aminoalcohol 5 (1 equiv) was dissolved in dry MeOH, and
anhydrous MgSO4 (excess) and the corresponding salicyl-
aldehyde (1.1 equiv) added. The mixture was stirred for
an appropriate time followed by TLC. After the reaction
was completed, MgSO4 was filtered off, and the solvent
was evaporated under vacuum. The resulting Schiff’s
base was purified as follows.
4.2.5.
(S,E)-4-tert-Butyl-2-((2-(2,2,2-trifluoro-1-hydroxy-
ethyl)phenylimino)methyl)phenol 1e. The reaction time
was 4 h. The product was recrystallized from hexane–
4.2.1. (S,E)-2-((2-(2,2,2-Trifluoro-1-hydroxyethyl)-phenyl-
imino)methyl)phenol 1a. The reaction time was 4 h. The
Et2O to give a yellow solid. Yield 86%. Mp 139–142 °C.
20:5
½aꢁD ¼ ꢀ212:8 (c 1.21, CHCl3). 1H NMR (CDCl3) d:
product was recrystallized from hexane–Et2O to give a yel-
12.34 (1H, s, disappeared with D2O), 8.54 (1H, s), 7.72
(1H, d, J = 7.8 Hz), 7.50–7.44 (2H, m), 7.41–7.35 (2H,
m), 7.09 (1H, dd, J = 7.8, 1.0 Hz), 6.98 (1H, d,
J = 8.7 Hz), 5.62 (1H, q, J = 6.4 Hz), 2.83 (1H, s, dis-
20:5
low solid. Yield 93%. Mp 115–117 °C. ½aꢁD ¼ ꢀ159:5 (c
1
1.10, CHCl3). H NMR (CDCl3) d: 12.58 (1H, s, disap-
peared with D2O), 7.99 (1H, s), 7.64 (1H, d, J = 7.8 Hz),

2658
Y. S. Sokeirik et al. / Tetrahedron: Asymmetry 17 (2006) 2654–2658
appeared with D2O), 1.33 (9H, s). 19F NMR (CDCl3–
References
toluene d8) d: ꢀ14.72 (3F, d, J = 6.2 Hz). IR (KBr) cmꢀ1
:
3415, 1620. MS m/z: 351 (M+). HRMS Calcd for
1. For review, see: Katsuki, T. Chem. Soc. Rev. 2004, 33, 437–
444.
2. Omote, M.; Kominato, A.; Sugawara, M.; Sato, K.; Ando, A.;
Kumadaki, I. Tetrahedron Lett. 1999, 40, 5583–5585.
3. Concerning Steric effect of a CF3 group, see: Kumadaki, I.
Rev. Heteroatom. Chem. 1993, 9, 181.
C19H20F3NO2: 351.1446 (M+). Found: 351.1447.
4.2.6.
amino)phenyl)ethanol 1f.
(S,E)-2,2,2-Trifluoro-1-(2-(2-methoxy-benzylidene-
solution of (100 mg,
A
5
0.52 mmol) and 72 mg (0.52 mmol) of o-anisaldehyde in
dry toluene was refluxed under Dean and Stark condenser
for 4 h with continuous removal of water. The resulting
Schiff’s base was kept in 2 mL dry toluene and used as
stock solution, MS m/z: 309 (M+). HRMS Calcd for
C16H14F3NO2: 309.0977 (M+). Found: 309.0984.
4. Meyers, A. I. Tetrahedron Lett. 1988, 29, 5617–5620.
5. Kriis, K.; Kanger, T.; Muurisepp, A.; Lopp, M. Tetrahedron:
¨ ¨
Asymmetry 2003, 14, 2271–2275; Bauer, T.; Gajewiak, J.
Tetrahedron: Asymmetry 2005, 16, 851–855; Huan, W.; Hu, Q.;
Pu, L. J. Org. Chem. 1999, 64, 7940–7956; Vaestilae, P.;
Pastor, I. M.; Adolfsson, H. J. Org. Chem. 2005, 70, 2921–
2929; Priego, J.; Mancheno, O. G.; Cabrera, S.; Carretero,
˜
J. C. Chem. Commun. 2001, 2026–2027.
6. Cozzi, P. G.; Papa, A.; Umani-Ronchi, A. Tetrahedron Lett.
1996, 37, 4613–4616.
4.3. General procedure for the reaction of Et2Zn with
aldehyde
7. Kitamura, M.; Suga, S.; Niwa, M.; Noyori, R. J. Am. Chem.
Soc. 1995, 117, 4832–4842; Kitamura, M.; Yamakawa, M.;
Oka, H.; Suga, S.; Noyori, R. Chem. Eur. J. 1996, 2, 1171–
1181; Kitamura, M.; Suga, S.; Makoto, N.; Noyori, R.; Zhai,
Z.-X.; Suga, H. J. Phys. Chem. 1994, 98, 12776–12781;
Kitamura, M.; Oka, H.; Suga, S.; Noyori, R. J. Am. Chem.
Soc. 1998, 120, 9800–9809.
8. Katagiri, T.; Fujiwara, Y.; Takahashi, S.; Ozaki, N.; Une-
yama, K. Chem. Commun. 2002, 986–987.
9. This complex was not formed when either half or 1 equiv of
diethylzinc was used. It only formed in the presence of a large
excess of diethylzinc.
To a solution of the Schiff’s base catalyst (0.034 mmol) in
dry solvent (5 mL) was added Et2Zn (1.62 mL of 1 M solu-
tion in hexane, 1.62 mmol). The mixture was stirred for 1 h
at room temperature and then cooled to 0 °C. The appro-
priate aldehyde (0.68 mmol) was added slowly at this tem-
perature. The mixture was kept stirring at this temperature
and followed by GLC. After the reaction was completed,
the mixture was quenched with 2 M HCl. The product
was extracted with an appropriate solvent and purified by
flash column chromatography. The ee was checked by
chiral GC. The results shown in Table 2 were obtained.