α-Pinene-type chiral Schiff bases as tridentate ligands in asymmetric addition reactions

Article abstract of DOI:10.1016/j.tetasy.2011.03.013

A group of tridentate Schiff bases derived from (+)-α-pinene were synthesized. The steric effects in the transition state, the importance of π-π stacking interactions as well as the electronic effects of aryl aldehydes according to Hammett constant values in the enantioselective addition of Et₂Zn to aldehydes with the use of Schiff bases as chiral ligands are described. Also, a variety of aldehydes were cyanated using a catalyst prepared in situ from titanium tetraisopropoxide and chiral Schiff bases. The influence of a conjugated double-bond in the cyanation substrates on enantioselectivity was observed. The chemical structures of the chiral Schiff base-titanium alkoxide complexes are discussed based on their ¹H and ¹³C NMR spectra. 3D models of the Zn₂-complex catalyst and Ti-complex catalyst containing α-pinane-type Schiff bases based on X-ray diffraction experiments are postulated. The models presented were consistent with the reported chirality of the addition product and observed ee.

Full text of DOI:10.1016/j.tetasy.2011.03.013

Tetrahedron: Asymmetry 22 (2011) 648–657
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
a-Pinene-type chiral Schiff bases as tridentate ligands in asymmetric
addition reactions
Magdalena Jaworska a, , Ewelina Błocka a, Anna Kozakiewicz b, Mirosław Wełniak a
⇑
a
´
Department of Organic Chemistry, Nicolaus Copernicus University, Gagarina 7, 87-100 Torun, Poland
b Department of Crystallochemistry and Biocrystallography, Nicolaus Copernicus University, Gagarina 7, 87-100 Torun, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
A group of tridentate Schiff bases derived from (+)-a-pinene were synthesized. The steric effects in the
transition state, the importance of p–p stacking interactions as well as the electronic effects of aryl alde-
hydes according to Hammett constant values in the enantioselective addition of Et2Zn to aldehydes with
the use of Schiff bases as chiral ligands are described. Also, a variety of aldehydes were cyanated using a
catalyst prepared in situ from titanium tetraisopropoxide and chiral Schiff bases. The influence of a con-
jugated double-bond in the cyanation substrates on enantioselectivity was observed. The chemical struc-
tures of the chiral Schiff base–titanium alkoxide complexes are discussed based on their 1H and 13C NMR
Received 3 February 2011
Accepted 3 March 2011
Available online 4 May 2011
spectra. 3D models of the Zn2-complex catalyst and Ti-complex catalyst containing a-pinane-type Schiff
bases based on X-ray diffraction experiments are postulated. The models presented were consistent with
the reported chirality of the addition product and observed ee.
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
The main aim of our work was the preparation of chiral Schiff
bases derived from (1S,2S,3R,5S)-2-amino-3-hydroxymethylp-
a
-Pinene is an important source of chirality because both of its
inane3 and their application as tridentate ligands for asymmetric
synthesis. Diethylzinc addition to aldehydes and the cyanation of
aldehydes were chosen as representative examples of enantiose-
lective reactions according to the usefulness of secondary alcohols
and cyanohydrins as chiral building block described broadly in the
literature.12
enantiomers are naturally available. Using classical organic synthe-
sis methods,
a-pinene can be easily converted into many useful
derivatives such as alcohols, ketones, amines, aminoalcohols and
so on. The pinane-type compounds are broadly used in asymmetric
synthesis as chiral auxiliaries,1 chiral ligands,2 and building blocks.3
Some of them are commercially available and easily accessible, for
example, Alpine-BoraneÒ (B-3-pinanyl-9-borabicyclo[3.3.1]non-
ane), effective in the stereoselective reduction of certain acetylenic
ketones4; (+)-pinanediol, used to prepare chiral double allylation
reagents5; and (R)-Alpine-BoramineÒ, as a polarity-reversal catalyst
for the kinetic resolution of racemic esters or ketones.6 Pinene-type
aminoalcohols are also precursors of oxazaborolidines—the effec-
tive organocatalysts for the asymmetric reduction of ketones with
2. Results and discussion
2.1. Synthesis and structural investigations on
Schiff bases
a-pinene-type
The synthesis of aminoalcohol 3, shown of Scheme 1, was per-
formed according to the procedure described in the literature.3
The regio- and stereoselectivity of CSI cycloaddition was previously
described by Sasaki et al.13 so that b-lactam 1 should have ee >99%
which was confirmed by comparison with the specific rotations in
the literature.3
BH3.7,8 Recently, cis- -pinene9 were
c-aminoalcohols derived from a
employed in the asymmetric diethylzinc addition to aldehydes with
moderate enantioselectivities. Moreover, salicylidene Schiff bases
containing a cis-c-aminoalcohol pinane frame-work were used
with limited success as precatalysts in the catalytic oxidation of sul-
fides into chiral sulfoxides.10 However, other tridentate Schiff base
catalysts have also been applied in asymmetric nucleophilic addi-
tion reactions, although only a few reports with high enantioselec-
tivities are known.11
In the reaction of 1 with ethanol containing HClgas, b-aminoester
2 was obtained after distillation under reduced pressure (1 Torr).
The crystallization of crude oily product 3 in hexane at ꢀ20 °C
obtained after LiAlH4 reduction of 2 allowed us to avoid problem-
atic aminoalcohol 3 purification by using hydrochloride derivative.3
The reaction of aminoalcohol 3 with selected salicylaldehydes
afforded six chiral Schiff bases 4a–f (Scheme 2). Two different
syntheses were applied.10,11b,14
⇑
Corresponding author. Tel.: +48 56 611 4521; fax: +48 56 654 2477.
E-mail address: magdajaworska@vp.pl (M. Jaworska).
0957-4166/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2011.03.013

M. Jaworska et al. / Tetrahedron: Asymmetry 22 (2011) 648–657
649
H
NH2
NH2
COOEt
N
O
i
iii
ii
OH
1
3
2
(+)-α-pinene
e.e.=97%,
[α]D21 = +50.7, neat
Scheme 1. Synthesis of (1S,2S,3R,5S)-2-amino-3-hydroxymethylpinane 3. (i) CSI, KOH, Na2SO3, rt, 48 h, 76–80%; (ii) 10% HClgas in EtOH, reflux, 12 h, 70%; (iii) LiAlH4, THF,
reflux, 4 h, crystallization from hexane at ꢀ20 °C, 80%.
R3
R4
NH2
R2
N
Procedure A or B
OH
R1
HO
OH
3
4a: R1=R2=R3=R4= H (57%)
4b: R1= Me, R2=R3=R4= H (46%)
4c: R1= t-Bu, R2=R4= H, R3= Me (88%)
4d: R1= t-Bu, R2=R4= H, R3= t-Bu (60%)
4e: R1=R2=R4= H, R3= Br (77%)
4f: R1= H, R2= N,N-diethyloamino, R3=R4= H (64%)
Scheme 2. Synthesis of Schiff bases 4a–f. Procedure A: aldehyde, p-TsOH, anhydrous MgSO4, toluene, reflux, 24 h. Procedure B: aldehyde, anhydrous MgSO4 or Na2SO4,
MeOH, rt, 24 h.
Table 1
To assess the configuration of the stereogenic centers, the con-
formation of the molecules and the intramolecular interactions,
Crystal data and structure refinement for 1 and 4d
1
4d
the crystal structure was determined for b-lactam 1 and Schiff base
4d. Some details of the diffraction experiments and structural
refinement are summarized in Table 1.
The molecular structures of 1 and 4d with the atom numbering
scheme is presented on Figures 1 and 2, respectively.
Formula sum
Formula weight
Color
Crystal system
Space group
C
11H17NO
C26.50H44.75N1O3.50
433.38
Yellow
179.26
Colorless
Orthorhombic
P212121
Orthorhombic
P21212
The valence geometry of the compound investigated was typi-
cal for the lactam ring. The four-membered lactam ring C2–N3–
C4–C5 was planar, as characterized by the torsion angles varied
from ꢀ2.28(13)° to 2.22(13)°. This ring and the pinane gem-di-
methyl group C10–C8–C11 were positioned on the opposite sides
of the central C1–C2–C5–C6 plane (Fig. 1).
The asymmetric unit of 4d structure is presented on Figure 2.
The conformation of the O1–C11–C3–C2–N1–C12–C13–C14–O2
was trans–cis–trans–trans–cis–cis, as characterized by the consecu-
tive O1–C11–C3–C2, C11–C3–C2–N1, C3–C2–N1–C12, C2–N1–
C12–C13, N1–C12–C13–C14 and C12–C13–C14–O2 torsion angles
being 168.0(3)°, 25.0(4)°, ꢀ158.3(3)°, 178.6(3)°, ꢀ0.1(5)° and
3.2(5)°, respectively. The H2O and CH3OH solvent molecules were
found in the structure and take part in intra- and intermolecular
H-bonds. The crystal structure revealed the disorder of the tert-bu-
tyl group bound to the phenolic C17 atom with the major popula-
tion of 65%.
Unit cell dimensions
a (Å)
b (Å)
6.6530(10)
10.295(2)
14.968(3)
1025.2(3)
4
10.3853(5)
18.5593(8)
14.1549(6)
2728.3(2)
4
c (Å)
V (Å3)
Z
Absorption coefficient
0.074
0.068
(mmꢀ1
)
F(0 0 0)
h range for data collection
(°)
Reflections collected
Independent reflections
Transmission max/min
Goodness-of-fit on F2
Final R indices [I > 2r(I)]
392
2.40–28.42
955
2.19–24.99
6186
16,239
2295 [R(int) = 0.0238]
0.9887/0.9682
1.072
R1 = 0.0405,
wR2 = 0.1030
R1 = 0.0457,
wR2 = 0.1064
0.4(18)
4804 [R(int) = 0.0663]
0.9913/0.9583
0.933
R1 = 0.0675,
wR2 = 0.1677
R1 = 0.1259,
wR2 = 0.1915
2(3)
R indices (all data)
Absolute structure
parameter
For 1 and 4d, the absolute structure could not be reliably deter-
mined with the Flack method.15 The correct configuration was
determined by comparison to the known chirality on C1 and C5 of
2
2
2
1=2
R1
¼
RjjF0j ꢀ jFcjj=
RjjF0j; wR2 ¼ ½
R
wðF20 ꢀ F2c Þ =
R
ðwjF0j Þ ꢁ
.
the (+)-a-pinene substrate, and is (1S,2S,5R,7S) and (1S,2S,3R,5S)
for 1 and 4d, respectively.
the enantioselectivity (Table 2; entries 1–3). Therefore, the influ-
ence of the temperature and the time of reaction were investigated
using Schiff base 4d (Table 2; entries 6–8). The best results for the
chiral ligands investigated were observed when both the catalyst
formation and addition to benzaldehyde were performed at
ꢀ70 °C. Increasing the reaction time from 24 to 48 h did not cause
an increase in reaction yield. As a result, other experiments were
performed for 24 h.
2.2. Enantioselective addition of diethylzinc to aldehydes
In the first stage, Schiff bases were tested as chiral ligands 4a–f
in the enantioselective addition of diethylzinc to benzaldehyde
(Scheme 3). The results of studied reactions are summarized in
Table 2.
Decreasing the temperature of the active catalyst formation
process with ligand 4a did not have a remarkable influence on
When using 4a, 4b and 4d–f as chiral ligands, the addition prod-
uct (S)-1-phenyl-1-propanol was obtained (Table 2; entries 1–4

650
M. Jaworska et al. / Tetrahedron: Asymmetry 22 (2011) 648–657
Table 2
The asymmetric addition of diethylzinc to benzaldehyde with Schiff bases 4a–f as
chiral ligands
Entry
Ligand
Time (h)
Yieldd (%)
eee (%)
Config.f
1
2
3
4
5
6
7
8
9
10
4a
4a
4a
4b
4c
4d
4d
4d
4e
4f
24a
24b
24c
24b
24b
24a
24b
48b
24b
24b
46
11
30
54
54
76
55
57
46
16
22
21
21
14
6
(S)
(S)
(S)
(S)
(R)
(S)
(S)
(S)
(S)
(S)
4
36
36
29
12
a
Reaction was carried out in the temperature range of 0 °C to +20 °C.
Reaction was carried in the temperature range of ꢀ70 °C to +20 °C.
Reaction was carried in the temperature range of ꢀ20 °C to +20 °C.
Isolated products.
Determined by HPLC using OD-H column.
The configuration was determined by the measurement of the specific rotation
b
c
d
e
f
and comparison with the literature values.16
Figure 1. Structure of 1. The thermal ellipsoids are plotted at 30% probability level.
Figure 2. Structure of 4d. The thermal ellipsoids are plotted at 30% probability
level.
O
OH
*
Figure 3. Model of the active complex Zn–Zn containing ligand 4d with the
1.20% mol ligand 4a-f
substrate orientation obtained with molecular mechanics calculations.17
H
Et2Zn, toluene
2. NH4Claq
presence of an intramolecular H-bond O2–H2Aꢂ ꢂ ꢂN1 indicated
the ability of 4d molecule to coordinate bidentately to Zn without
major conformational changes. Analysis of CSD entries18 for similar
imine ligands did not show any significant factors restricting the
conformational freedom of the O1–C11–C3–C2–N1 fragment,
Scheme 3. Asymmetric addition of diethylzinc to benzaldehyde.
and 6–10) in contrast to the use of 4c (Table 2; entry 5). The change
of configuration might be due to the different steric hindrance
caused by the tert-butyl at C15 and the methyl at C17 (Fig. 2) in
the phenolic ring of 4c. The best ee = 36% was obtained when
ligand 4d, which has two bulky tert-butyl substituents in the
phenolic ring at C15 and C17 (Fig. 2), was used (Table 2; entry
7). Despite the significant distance from the reaction center, the
increase of enantioselectivity was influenced by the substituent
at C17 (Fig. 2) in the salicylidene moiety of ligands 4d and 4e.
Smaller substituents at C17, for example, methyl 4c caused a
decrease in enantioselectivity (Table 2; entry 5).
which suggests the possibility of tridentate coordination g
3–NO2,
to Zn or Ti(IV) in the active complex with a formation of the six-
membered chelate rings.
Modeling studies revealed that the bulk of the ligand pinane
ring enforces the binding of benzaldehyde with its Si face exposed
to Et2Zn as well as for the Re face, the distance between the pinane
moiety of the ligand and the aromatic ring of the substrate is too
close (1.6 Å). Therefore, addition to the Re face would require relax-
ation with possible conformational changes. The importance of the
Based upon the crystal structure of 4d a model of the active
complex (Fig. 3) was built. The model obtained was optimized with
molecular mechanics methods with an ARGUSLAB program using a
UFF force field.17
p
(the
–
p
interactions between the ligand and substrate in the Si face
distance being 3.9 Å) was also observed. Consequently,
pꢂ ꢂ ꢂp
the model showed that the addition to the Si face should be
preferred.
The Schiff bases reported herein have three heteroatoms, which
could participate in the coordination to the metal center. The
To demonstrate the influence of electronic and steric effects of
the substrate for the asymmetric ethylation reaction, different aryl

M. Jaworska et al. / Tetrahedron: Asymmetry 22 (2011) 648–657
651
aldehydes were used as substrates under the optimized conditions.
The results are summarized in Table 3.
Significantly, lower ee values than those obtained for benzalde-
hyde occurred for o-chloro-, o-bromo-, m-methoxy and o-methoxy-
benzaldehyde (Table 3; entries 1 and 3–5). The presence of these
substituents probably caused an increase in the steric hindrance
near the carbonyl carbon atom which would hinder the addition of
the ethyl group. Some electronic effects for other substrates were
also observed. The substrates with electron-withdrawing groups
at the para-position (Table 3; entries 7 and 8) of the aryl aldehydes
afforded a higher ee than those with electron-donating groups
(Table 3; entries 2 and 3). Furthermore, the product of Et2Zn addition
to m-chlorobenzaldehyde had a higher ee (54%) (Table 3; entry 6)
than for benzaldehyde (Table 2; entry 7). A correlation between
the enantioselectivity and the electronic Hammett constant is
shown in Figure 4.
The Hammett plot of the enantioselectivity versus the rpara val-
ues for ligand 4d showed a linear correlation (R2 = 0.9837), which
was consistent with the results reported by Zhang et al.19 It is ex-
pected that the presence of the electron-withdrawing groups (po-
Figure 4. Plot of the enantiomeric excess of the addition of Et2Zn to para-
sitive Hammett constants r) leads to an increase in Lewis acidity of
substituted aryl aldehydes versus Hammett constant
r for ligand 4d.
the carbonyl carbon atom in the aryl aldehydes, while the presence
of electron-donating groups in the aryl aldehydes (negative
Hammett constants r) diminishes it. The origin of the remarkable
substituents such as a tert-butyl group at C15 in the phenolic ring
(Fig. 2) led to an increase in enantioselectivity (Table 4; entries 3
and 4). Contrary to this, the presence of bulky tert-butyl groups
at C17 led to decrease in enantioselectivity (ligand 4d). The Schiff
base 4c with 3-tert-butyl-5-methylsalicylidene moiety was the
best chiral ligand and gave the product (R)-2-hydroxy-2-phenyl-
acetonitrile with an ee = 60% (Table 4; entry 3). Moreover, the
investigation of the electronic effects of substituents on ligands
(4e and 4f) showed that electron-donating and electron-withdraw-
ing groups on C16 and C17 (Fig. 2) in the salicylidene moiety did
not cause an increase in the ee (Table 4; entries 5 and 6), which
is in contrast to the results reported by Li.21
electronic effect described here on the enantioselectivity remains
unclear. The mechanism of the diethylzinc addition to aldehydes
using aminoalcohols as chiral ligands is coupled with the forma-
tion of dinuclear zinc complexes.20 Probably, aldehyde binding to
zinc complexes with Schiff base ligands, with the imine bond con-
jugated with the aromatic ring involved in the salicylidene moiety,
led to the formation of a transition state that is stabilized by the
electron-withdrawing substituents on the para-position. Similar
results for Schiff base ligands have also been reported in our previ-
ous work.14
In conclusion, the enantioselectivity of the tested reaction in-
creased if more reactive substrates were used. Some deviations
of the observed electronic effects might be attributed to the change
in inherent activity of the catalysts shown in the proposed model
of the transition state (Fig. 3).
Encouraged by the results obtained for benzaldehyde as the
substrate and in the presence of 20 mol % 4c as a chiral ligand,
the cyanation reactions of a variety of aldehydes were studied.
The results are summarized in Table 5.
Aromatic and some aliphatic aldehydes gave similar ee values
(54–62%) (Table 5; entries 1–3 and 5). Considerably lower ee val-
ues were obtained for cyclohexanecarboaldehyde (45%) and furfu-
ral (42%) (Table 5; entries 4 and 8). It is possible that the oxygen
atom in the heteroaromatic furfural was a crucial element in the
transition state formation with the titanium complex and caused
the decrease of ee On the other hand, the steric interactions of
the cyclohexane ring of an aliphatic cyclohexanecarboaldehyde
also caused a decrease in the ee. Much better results were obtained
when methacrolein and (E)-cinnamaldehyde were tested (Table 5;
entries 6 and 7). High enantioselectivity (ee >99%) in the case of
2.3. Enantioselective cyanation of aldehydes
In the next stage, the cyanation of aldehydes was examined
with benzaldehyde in the presence of 20 mol % of the chiral Schiff
bases 4a–f (Scheme 4). The catalyst–titanium (IV) complex was
formed in situ when titanium (IV) isopropoxide was mixed with
a pinane-type ligand at room temperature. The results of the tested
ligands are summarized in Table 4.
The results indicated that the enantioselectivity was signifi-
cantly influenced by the structure of the ligands. More bulky
Table 3
Addition of diethylzinc to aldehydes with Schiff base 4d as a chiral ligand
Entry
Ligand
Aldehyde
Time (h)
Yielda (%)
eeb (%)
Config.c
1
2
3
4
5
6
7
8
9
4d
4d
4d
4d
4d
4d
4d
4d
4d
4d
o-Methoxybenzaldehyde
m-Methoxybenzaldehyde
p-Methoxybenzaldehyde
o-Chlorobenzaldehyde
o-Bromobenzaldehyde
m-Chlorobenzaldehyde
p-Chlorobenzaldehyde
p-Bromobenzaldehyde
p-Dimethylaminobenzaldehyde
Cyclohexanecarboaldehyde
72
72
72
72
72
72
72
72
72
72
74
90
45
66
51
62
60
70
65
54
4
8
6
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
—
4
10
54
82
90
0
10
20
(S)
a
b
c
Isolated products.
Determined by HPLC using an OD-H column or by GC using a b-Dex capillary column.
The configuration was determined by measurement of the specific rotation and comparison with the literature values.16

652
M. Jaworska et al. / Tetrahedron: Asymmetry 22 (2011) 648–657
Table 6
CN
*
O
1. 20% mol Schiff base - Ti(O-i-Pr)4
TMSCN
Enantioselective addition of TMSCN to (E)-cinnamaldehyde catalyzed by Schiff base
4c–titanium tetraisopropoxide complex
OH
H
2. 1M HCl
Entry Ligand Aldehyde
Mol % of
ligand 4c [%]
Yielda eeb Config.c
[%]
Scheme 4. Enantioselective trimethylsilylcyanation of benzaldehyde.
1
2
3
4
4c
4c
4c
4c
(E)-Cinnamaldehyde 10
76
43
10
99
99
98
98
(R)
(R)
(R)
(R)
5
1
0.5
67d
Table 4
Enantioselective addition of TMSCN to benzaldehyde catalyzed by Schiff base–
titanium (IV) isopropoxide complexes
a
b
c
Isolated products.
Determined by GC using b-Dex capillary column.
Entry
Ligand
Yielda [%]
eeb[%]
Config.c
The configuration was determined by the measurement of the specific rotation
and comparison of the literature values.11c
1
2
3
4
5
6
4a
4b
4c
4d
4e
4f
79
45
73
47
99
89
22
9
60
50
0
(R)
(R)
(R)
(R)
(R)
(R)
d
The reaction time was 72 h.
Using tridentate Schiff base ligands, Hayashi identified the cata-
lytic active titanium complex as the 1:1 (ligand/titanium) adduct
marked as LꢃTi(O-i-Pr)2. They also noticed that the bis(Schiff
base)compounds marked as Lꢃ2Ti were not active in the catalytic
cyanation of aldehydes. The structure of the titanium complexes
with our pinane-type Schiff base ligands using NMR spectrometry
and a molecular mechanics method17 has been studied. The aldim-
ine region with well separated resonances in 1H and 13C NMR spec-
tra was diagnostic for our investigations and revealed explicit
differences between the chemical shifts of imine bonds in the li-
gand and in their titanium complexes.
15
a
b
c
Isolated products—cyanohydrins.
Determined by HPLC using an OD-H column of their acetates.
The configuration was determined by the measurement of the specific rotation
and comparison with the literature values.11b
Table 5
Enantioselective addition of TMSCN to aldehydes catalyzed by a Schiff base 4c–
titanium tetraisopropoxide complex
Entry Ligand Aldehyde
Yielda [%] eeb [%] Config.c
When equal molar amounts of chiral ligand 4c (with bulky sub-
stituents at C15 in the phenolic ring) and titanium tetraisopropox-
ide were combined, only two types of titanium complexes: LꢃTi(O-
i-Pr)2 and Lꢃ2Ti were formed, which is in contrast to the results de-
scribed in the literature.11d In Figure 5b, in the aldimine region,
four resonances were observed. According to DEPT 135 and HMBC
spectra resonances at 165.10 and 163.23 ppm as imine C–H car-
bons as well as 160.00 and 159.31 ppm as quaternary carbons
bonded to phenolic –OH were identified. The presence of two var-
ious resonances of the imine carbons proved the occurrence of two
different titanium complexes. In comparison with the pure ligand,
the 13C{H} NMR spectrum in the investigated mixture unbonded li-
gand 4c was not present. When ligand 4c and titanium tetraiso-
propoxide in a 2:1 ratio were combined, the formation of only
one Lꢃ2Ti complex was observed. In the 13C{H} NMR spectrum
(Fig. 5c) in the aldimine region, only one resonance of an imine
C–H carbon at 165.10 ppm and one resonance at 160.90 ppm of a
quaternary carbon bonded with a phenolic –OH were detected.
These resonances corresponded with those presented on the
13C{H} NMR spectrum of ligand/titanium tetraisopropoxide 1:1 ra-
tio mixture so that in the 1:1 mixture two kinds of titanium com-
plexes could have occurred. Therefore, the 1H NMR spectra of the
ligand and mixtures of ligand and titanium isopropoxide in two ra-
tios of 1:1 and 2:1 (Fig. 6) were also recorded.
1
2
3
4
5
6
7
8
4c
4c
4c
4c
4c
4c
4c
4c
p-Methoxybenzaldehyde
m-Methoxybenzaldehyde
m-Chlorobenzaldehyde
Cyclohexanecarboaldehyde 75
Hexanal
(E)-Cinnamaldehyde
Methacrolein
Furfural
89
53
94
62
62
58
45
54
>99
80
42
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
45
83
82
40
a
Isolated products.
b
Determined by HPLC using an OD-H column or by GC using a b-Dex capillary
column.
c
The configuration was determined by the measurement of the specific rotation
and comparison with the literature values.11c
(E)-cinnamaldehyde was certainly influenced by electronic effects
of the double bond conjugated with the aromatic ring. The elec-
tronic effects could be related to an increase of electrophilicity of
the carbonyl carbon atom in the substrate so that the enantioselec-
tivity could be higher as well the reaction speed. Aliphatic methac-
rolein without an aromatic ring conjugated with the double bond
system also gave satisfactory enantioselectivity (ee = 80%), thus it
should be stressed that the presence of double bond in the sub-
strate near the reaction center had a remarkable influence on the
reaction ee.
According to the high enantioselectivity of the (E)-cinnamalde-
hyde cyanation reaction, the effect of the amount of ligand 4c on
the enantioselectivity and yield was also studied and the results
are summarized in Table 6.
In fact, catalyst (Schiff base 4c–titanium complex) loading did
not significantly influence the enantioselectivity or yield. Even if
the amount of catalyst at 1 or 0.5 mol % level was used, the enanti-
oselectivity was still at the same level when using 5–20 mol %.
However, when the amount of catalyst was 0.5 mol %, the reaction
time needed to be increased from 48 to 72 h in order to obtain a
better yield.
The 1H NMR spectrum contained in the aldimine region two
singlets at 8.05 and 8.00 ppm with an integration ratio of 1:2. In
the mixture of the 1:1 catalytic active species LꢃTi(O-i-Pr)2 was
predominantly formed. However, an octahedral coordination
geometry of Lꢃ2Ti species still could not be determined. Flores-
Lopéz et al.11d mentioned that tridentate Schiff bases (Lꢃ) gave five
possible geometrical isomers of Lꢃ2Ti; two pairs of C2-symmetric
diastereoisomers and one isomer with both steric inequivalent li-
gands. According to our results, the Lꢃ2Ti species was one of the
C2-symmetric diastereoisomers, because in the 1H and 13C{H}
NMR spectra, only one aldimine C–H resonance (Figs. 5c and 6c)
was observed.
2.4. Structural studies on the catalytic active titanium catalyst
Based on the crystal structure of 4d, the model of the active Ti-
complex with ligand 4c (Fig. 7) was built. The obtained model was
optimized with a molecular mechanics method in ARGUSLAB program
using UFF force field.17
Hayashi et al.11b and Flores-Lopéz11d have studied the titanium
complexes with Schiff bases using NMR and mass spectrometry.

M. Jaworska et al. / Tetrahedron: Asymmetry 22 (2011) 648–657
653
Figure 5. 13C{H} NMR spectra of (a) ligand 4c; (b) mixture of ligand 4c and titanium tetraisopropoxide 1:1; (c) mixture of ligand 4c and titanium tetraisopropoxide 2:1;
diagnostic areas of the hydroxy-aldimine resonances are enlarged.
This model suggests the preferred addition to the Re face of the
aldehyde. The bulk of both the tert-butyl group of the phenolic ring
enantioselectivity of the diethylzinc additions. The described inter-
actions were consistent with presented theoretical 3D models ob-
tained using molecular mechanics methods based on the crystal
structure of 4d. The asymmetric cyanation of a variety of aldehydes
with a catalyst prepared from Ti(O-i-Pr)4 and chiral Schiff bases
possessing different groups in the salicylidene moiety provides
an efficient method for the synthesis of optically active cyano-
and the
a-pinane moiety cause steric hindrance thus enhancing
CNꢀ addition to the Re side of benzaldehyde. This conclusion is
consistent with the observed configuration and ee of the obtained
1-phenylpropanol.
The NMR measurements and molecular mechanics calculations
were compared with results from the catalytic asymmetric cyana-
tion of aldehydes. Ligand 4c gave medium yield (73%) as well as
ee = 60%. The medium yield could be explained by the presence
of inactive Lꢃ2Ti species. The formation of this complex was
significantly connected with the increasing percentage of
unreacted titanium tetraisopropoxide, which slowly catalyzes the
cyanide addition reaction but gives a racemic product.11c When
ligand 4c was tested with other substrates, similar results
(yields = 40–94%; ee = 42–62%) were obtained. In contrast, using
(E)-cinnamaldehyde and methacrolein as substrates, an increase
in the enantioselectivity of the cyanation product was observed.
hydrins, especially when using
a,b-unsaturated aldehydes as
substrates. The enantioselectivity of the cyanide addition to (E)-
cinnamaldehyde was up to 99% even if only 0.5–1 mol % of chiral
titanium-Schiff base catalyst was examined. The 1H and 13C NMR
spectroscopic analysis of the Schiff base-titanium complexes re-
vealed the presence of two different titanium complexes, Lꢃ2Ti
and LꢃTi(O-i-Pr)2. The theoretical calculations of catalytic active
LꢃTi(O-i-Pr)2 species provided an explanation for the substituent-
dependent enantioselectivity and can be applied to the design of
other Schiff base precatalysts. For the tridentate Schiff base-
titanium catalysts investigated herein, the position of the achiral
substituents in the salicylidene moiety as well as the electronic
effects from double bonds in the substrates had a significant
influence on the enantioselectivity.
3. Conclusion
The Schiff bases derived from (1S,2S,3R,5S)-2-amino-3-hydrox-
ymethylpinane may serve as chiral ligands and precatalysts in
enantioselective syntheses including cyanide or Et2Zn addition to
aldehydes. Herein, we emphasize the importance of: (1) steric
interactions between ligand and substrate; (2) electronic effects
4. Experimental
4.1. Instrumental
of substituents in the substrates; and (3)
the aromatic rings of the substrate and ligand in controlling the
p
–
p
stacking between
Melting points were determined with a Büchi apparatus in open
capillaries and are uncorrected. Optical rotation was measured

654
M. Jaworska et al. / Tetrahedron: Asymmetry 22 (2011) 648–657
Avance III 700 MHz, respectively, in CDCl3 at ambient temperature.
Chemical shifts are reported in parts per million (d scale), coupling
constants (J values) are listed in Hertz and tetramethylsilane was
used as the internal standard (d = 0 ppm). Infrared spectra are re-
ported in reciprocal centimeters (cmꢀ1) and are measured either
as a neat liquid or as a nujol mull or in HCB. GC was performed
on a Perkin–Elmer AutoSystem XL chromatograph using b-Dex
325 capillary column (30 m, 0.25 mm), or on Shimadzu GC-14A
using Zebron ZB-5 capillary column. HPLC analyses were per-
formed on a Shimadzu LC-10AT chromatograph using Chiralcel
OD-H column (250 ꢄ 4.6 mm). Colorless crystals of 1 and yellow
crystals of 4d have been obtained from the diisopropyl ether and
methanol-water solution, respectively. The X-ray data were col-
lected at 292 K with an Oxford Sapphire CCD diffractometer using
MoKa radiation k = 0.71073 Å and x-2h method. The numerical
absorption correction was applied with CrysAlis171 package of
programs, Oxford Diffraction, 2000.22 Structures were solved by di-
rect methods and refined with the full-matrix least-squares meth-
od on F2 with the use of SHELX-97 program package.23 The hydrogen
atoms have been located from the difference electron density maps
and constrained during refinement. The structural data have been
deposited with Cambridge Crystallographic Data Centre, the CCDC
numbers 812136 and 812137 for 1 and 4d, respectively.
4.2. Materials
TLC was performed on silica gel PolygramÒ Sil G/UV254
(0.2 mm). Regular column chromatography was carried out using
Silica Gel 60 (0.06–0.2 mm). All solvents were purchased from
POCh Gliwice, Poland. Toluene was distilled from sodium prior to
use. Diethylzinc, TMSCN, aldehydes, (+)-
¼ þ50:7, neat}, CSI, LiAlH4 were purchased from Sigma–Al-
drich or Fluka. 3,5-Di-tert-butyl-2-hydroxybenzaldehyde and 3-
tert-butyl-5-methyl-2-hydroxybenzaldehyde were obtained
a-pinene {ee = 97%,
½ ꢁ
a 2D1
according to the literature procedures24 and were confirmed to
be pure by the melting point (mp 58–59 °C, (lit.25 mp 58–60 °C)
and mp 69–70 °C (lit.24 mp 69–71 °C), respectively. 3-Methylsali-
cylaldehyde was obtained according to the literature26 and it was
confirmed to be pure by GC analysis. (1S,2S,3R,5S)-2-Amino-3-
hydroxymethylpinane was prepared according to the literature
Figure 6. 1H NMR spectra of (a) ligand 4c; (b) mixture of ligand 4c and titanium
tetraisopropoxide 1:1; (c) mixture of ligand 4c and titanium tetraisopropoxide 2:1.
procedures3 [the reaction route (+)-
a-pinene?b-lactam?b-
aminoester?b-aminoalcohol] and was confirmed to be pure by
the specific rotation measurement {½a D20
¼ ꢀ11:9 (c 1.2, MeOH),
ꢁ
lit.3
½
a 2D0
ꢁ
¼ ꢀ11:0 (c 2.0, MeOH)} after recrystallization from hex-
ane at ꢀ20 °C.
4.3. General procedures for the synthesis and purification of
Schiff bases 4a–f
4.3.1. Procedure A
A 100-ml round-bottom flask equipped with a magnetic stirrer
bar and reflux condenser was filled with 5 mmol (1.0 g) of amino-
alcohol, 60 ml of anhydrous toluene, 5 mmol of the appropriate
aldehyde, 20 mg of p-toluenesulfonic acid and 6.0 g of anhydrous
MgSO4. The reaction mixture was stirred for 24 h at reflux and
then, after cooling to room temperature, it was filtered onto a Cel-
ite pad of 1 cm thickness in a fritted glass funnel. If necessary, an
organic solution was dried with anhydrous MgSO4. After filtration,
the solvent was removed and the crude product was purified using
the purification procedure described below.
Figure 7. Model of the active Ti-complex formed with ligand 4c with the
orientation of the substrate as obtained with molecular mechanics calculations.17
4.3.2. Procedure B
with PolAAr 3000 automatic polarimeter in a 10 cm cell at 589 nm.
Elemental analyses were performed at Elementary Analysensys-
teme GmbH VarioMACRO CHNanalyzer. 1H and 13C NMR spectra
in solution were recorded on Bruker Avance III 400 MHz or Bruker
A 100-ml round-bottom flask equipped with a magnetic stirrer
bar was charged with 5.5 mmol (1.0 g) of aminoalcohol, 40 ml of
methanol and 6.0 g of anhydrous MgSO4 or Na2SO4. To the stirred
mixture,
a methanol solution of the appropriate aldehyde

M. Jaworska et al. / Tetrahedron: Asymmetry 22 (2011) 648–657
655
(5.5 mmol in 10–15 ml of MeOH) was added dropwise. The reac-
tion mixture was stirred for 24 h at room temperature and then
was filtered onto a Celite pad of 1 cm thickness in a fritted glass
funnel. If necessary, an organic solution was dried over anhydrous
MgSO4. After filtration, the solvent was removed and the crude
product was purified using the purification procedure described
below.
J = 7.8 Hz, J = 10.8 Hz, 1H, CH2OH), 6.78 (t, J = 7.2 Hz, 1H, CH-aryl
(C-5 in phenol ring)), 7.28 (dd, J = 1.7 Hz, J = 7.6 Hz, 1H, CH-aryl
(C-6 in phenol ring)), 7.21 (ddd, J = 0.8 Hz, J = 1.8 Hz, J = 7.3 Hz,
1H, CH-aryl (C-4 in phenol ring)), 8.29 (s, 1H, CH–imine), 14.44
(br s, 1H, OH). 13C NMR (176.1 MHz, CDCl3): d 15.62 (CH3), 23.68
(CH3), 28.26 (CH3), 28.61 (CH3), 29.55 (CH2), 31.31 (CH2), 38.89
(C), 40.17 (CH), 41.82 (CH), 53.80 (CH), 65.10 (C), 66.51 (CH2,
CH2OH), 117.47 (CH), 117.94 (C), 126.57 (C), 129.45 (CH), 133.35
(CH), 161.17 (C), 161.49 (CH, imine). IR (nujol mull): 3250.1,
2359.0, 1625.2, 1270.4, 1246.6, 1165.4, 1070.5, 1033.6, 844.8,
747.6. Anal. Calcd for C19H27NO2: C, 75.71; H, 9.03; N, 4.65; O,
10.62. Found: C, 75.49; H, 8.71; N, 4.80.
4.3.3. Purification procedures
The crude product was dissolved in 20 ml of methanol and
added directly to a water solution of sodium metabisulfite in a
crystallizer. The mixture was placed in a fume hood and evapo-
rated slowly to dryness. The solid obtained was triturated in a por-
celain mortar. The powder was placed on a fritted glass funnel and
washed with chloroform using a filter pump. The chloroform solu-
tion was dried over anhydrous Na2SO4, filtered and the solvent was
evaporated under reduced pressure.
4.6. 2-tert-Butyl-6-((E)-((1S,2S,3R,5S)-3-(hydroxymethyl)-2,6,6-
trimethylbicyclo[3.1.1]-heptan-2-ylimino)methyl)-4-
methylphenol 4c
The resulting product was recrystallized using the appropriate
solvents (4b and 4d) or purified using column chromatography
and then recrystallized (4a). In the case of 4c and 4e, after evapo-
ration of solvents, 10 ml of diethyl ether were added and then solu-
tion was dried under vacuum for 1.5 h.
2-Hydroxy-3-tert-butyl-5-methylbenzaldehyde was used as a
substrate using procedure B. Yellow-orange solid (88%). Purified
by recrystallization from methanol. Mp 67–70 °C, ½a D20
¼ ꢀ4:9 (c
ꢁ
0.41, MeOH). 1H NMR (400 MHz, CDCl3): d 1.20 (s, 3H, CH3 (bond
to C-6)), 1.28 (d, J = 10.3 Hz, 1H, H-7), 1.33 (s, 3H, CH3 (bond to
C-6)), 1.46 (s, 9H, CH3–t-Bu), 1.47 (s, 3H, CH3 (C-2)), 1.62 (ddd,
J = 2.1 Hz, J = 7.5 Hz, J = 13.6 Hz, 1H, H-4), 2.00 (t, J = 5.6 Hz, 1H,
H-1), 2.02–2.05 (m, 1H, H-5), 2.07 (br s, 1H, CH2OH), 2.14–2.19
(m, 1H, H-4), 2.24 (dddd, J = 2.4 Hz, J = 3.9 Hz, J = 9.7 Hz,
J = 11.7 Hz, 1H, H-7), 2.30 (s, 3H, CH3), 2.39–2.48 (m, 1H, H-3),
3.68 (dd, J = 7.0 Hz, J = 11.3 Hz, 1H, CH2OH), 3.84 (dd, J = 7.4 Hz,
J = 11.0 Hz, 1H, CH2OH), 6.94 (d, J = 1.4 Hz, 1H, CH-aryl (C-6 in phe-
nol ring)), 7.14 (d, J = 2.1 Hz, 1H, CH-aryl (C-4 in phenol ring)), 8.30
(s, 1H, CH–imine), 14.21 (br s, 1H, OH). 13C NMR (176.1 MHz,
CDCl3): d 20.69 (CH3), 23.70 (CH3), 28.30 (CH3), 28.67 (CH3),
29.44 (3 ꢄ CH3, t-Bu), 29.54 (CH2), 31.36 (CH2), 34.79 (C), 38.88
(C), 40.23 (CH), 42.05 (CH), 53.75 (CH), 65.14 (C), 66.55 (CH2,
CH2OH), 118.64 (C), 126.21 (C), 129.93 (CH), 130.36 (CH), 137.36
(C), 158.65 (C), 162.03 (CH, imine). IR (HCB mull): 2923.5,
2362.3, 1609.3, 1457.7, 1374.9. Anal. Calcd for C23H35NO2: C,
77.27; H, 9.87; N, 3.92; O, 8.95. Found: C, 77.32; H, 9.95; N, 4.04.
4.4. 2-((E)-((1S,2S,3R,5S)-3-(Hydroxymethyl)-2,6,6-
trimethylbicyclo[3.1.1]heptan-2-ylimino)-methyl)phenol 4a
2-Hydroxybenzaldehyde was used as a substrate using proce-
dure A. Yellow solid (57%). Purified by column chromatography
(hexane/ethyl acetate—3:1, v/v) and then by recrystallization from
hexane. Mp 89–92 °C,
½
a 2D0
ꢁ
¼ þ70:8 (c 0.048, MeOH) {lit.10
22
½
aꢁ
¼ þ2:8 (c 2.8, CHCl3)}. 1H NMR (700 MHz, CDCl3): d 1.21 (s,
580
3H, CH3 (bond to C-6)), 1.26 (d, J = 9.8 Hz, 1H, H-7), 1.34 (s, 3H,
CH3 (bond to C-6)), 1.51 (s, 3H, CH3 (C-2)), 1.55 (ddd, J = 2.2 Hz,
J = 7.6 Hz, J = 13.6 Hz, 1H, H-4), 1.58 (br s, 1H, CH2OH), 2.02 (t,
J = 5.6 Hz, 1H, H-1), 2.03-2.06 (m, 1H, H-5), 2.19 (dddd, J = 2.4 Hz,
J = 3.9 Hz, J = 9.5 Hz, J = 13.9 Hz, 1H, H-4), 2.29 (dddd, J = 2.3 Hz,
J = 3.9 Hz, J = 9.5 Hz, J = 13.4 Hz, 1H, H-7), 2.44-2.48 (m, 1H, H-3),
3.67 (dd, J = 6.35 Hz, J = 11.0 Hz, 1H, CH2OH), 3.82 (dd, J = 7.8 Hz,
J = 11.0 Hz, 1H, CH2OH), 6.86 (td, J = 0.8 Hz, J = 7.5 Hz, 1H, CH-aryl
(C-5 in phenol ring)), 6.96 (d, J = 8.3 Hz, 1H, CH-aryl (C-3 in phenol
ring)), 7.28 (dd, J = 1.63 Hz, J = 7.5 Hz, 1H, CH-aryl (C-6 in phenol
ring)), 7.33 (ddd, J = 1.67 Hz, J = 7.0 Hz, J = 8.8 Hz, 1H, CH-aryl (C-
4 in phenol ring)), 8.32 (s, 1H, CH–imine), 14.50 (br s, 1H, OH).
13C NMR (176.1 MHz, CDCl3): d 23.68 (CH3), 28.26 (CH3), 28.62
(CH3), 29.52 (CH2), 31.32 (CH2), 38.88 (C), 40.14 (CH), 41.73 (CH),
53.70 (CH), 65.15 (C), 66.42 (CH2, CH2OH), 117.73 (CH), 117.90
(CH), 118.60 (C), 131.80 (CH), 132.68 (CH), 161.28 (CH, imine),
163.29 (C). IR (nujol mull): 2362.9, 1627.9, 1507.8, 1280.1,
1068.1, 754.6. Anal. Calcd for C18H25NO2: C, 75.22; H, 8.77; N,
4.87; O, 11.13. Found: C, 74.94; H, 8.39; N, 4.75.
4.7. 2,4-Di-tert-butyl-6-((E)-((1S,2S,3R,5S)-3-(hydroxymethyl)-
2,6,6-trimethylbicyclo-[3.1.1]heptan-2-ylimino)methyl)phenol
4d
3,5-Di-tert-butyl-2-hydroxybenzaldehyde was used as a sub-
strate using procedure B. Yellow solid (60%). Purified by recrystal-
lization from methanol. Mp 137–139 °C, (lit.10 mp 129–130 °C),
22
580
½
a 2D0
ꢁ
¼ þ37:5 (c 0.044, MeOH) {lit.11a
½
aꢁ
¼ ꢀ6:7 (c 2.2, CHCl3)}.
1H NMR (700 MHz, CDCl3): d 1.21 (s, 3H, CH3 (bond to C-6)), 1.29
(d, J = 10.4 Hz, 1H, H-7), 1.34 (s, 3H, CH3 (bond to C-6)), 1.35 (s,
9H, 3 ꢄ CH3, t-Bu), 1.48 (s, 9H, 3 ꢄ CH3, t-Bu), 1.49 (s, 3H, CH3 (C-
2)), 1.62 (ddd, J = 1.8 Hz, J = 7.6 Hz, J = 13.7 Hz, 1H, H-4), 1.97 (br
s, 1H, CH2OH), 2.02 (t, J = 6.1 Hz, 1H, H-1), 2.03–2.06 (m, 1H, H-
5), 2.16–2.22 (m, 1H, H-4), 2.24–2.29 (m, 1H, H-7), 2.42–2.49 (m,
1H, H-3), 3.69 (dd, J = 6.4 Hz, J = 10.5 Hz, 1H, CH2OH), 3.87 (dd,
J = 7.9 Hz, J = 11.0 Hz, 1H, CH2OH), 7.12 (d, J = 2.0 Hz, 1H, CH-aryl
(C-6 in phenol ring)), 7.41 (d, J = 2.2 Hz, 1H, CH-aryl (C-4 in phenol
ring)), 8.34 (s, 1H, CH–imine), 14.34 (br s, 1H, OH). 13C NMR
(176.1 MHz, CDCl3): d 23.69 (CH3), 28.30 (CH3), 28.79 (CH3),
29.48 (3 ꢄ CH3, t-Bu), 29.74 (CH2), 31.32 (CH2), 31.53 (3 ꢄ CH3, t-
Bu), 34.15 (C), 35.08 (C), 38.88 (C), 40.19 (CH), 42.00 (CH), 53.74
(CH), 65.09 (C), 66.57 (CH2, CH2OH), 118.03 (C), 126.24 (CH),
126.97 (CH), 137.04 (C), 139.64 (C), 158.91 (C), 162.40 (CH, imine).
IR (nujol mull): 3219.8, 2359.5, 1623.8, 1251.9, 1163.3, 1028.8,
877.4, 722.4. Anal. Calcd for C26H41NO2: C, 78.15; H, 10.34; N,
3.51; O, 8.01. Found: C, 78.36; H, 10.50; N, 3.61.
4.5. 2-((E)-((1S,2S,3R,5S)-3-(Hydroxymethyl)-2,6,6-
trimethylbicyclo[3.1.1]heptan-2-ylimino)methyl)-6-
methylphenol 4b
2-Hydroxy-3-methylbenzaldehyde was used as a substrate
using procedure B. Yellow solid (46%). Purified by recrystallization
from hexane. Mp 82–83 °C, ½a D20
ꢁ
¼ þ63:0 (c 0.268, MeOH). 1H NMR
(400 MHz, CDCl3): d 1.20 (s, 3H, CH3 (bond to C-6)), 1.27 (d,
J = 10.4 Hz, 1H, H-7), 1.34 (s, 3H, CH3 (bond to C-6)), 1.50 (s, 3H,
CH3 (bond to C-2)), 1.55 (ddd, J = 2.0 Hz, J = 7.7 Hz, J = 13.8 Hz,
1H, H-4), 2.00–2.06 (m, J = 5.6 Hz, 2H, H-1, H-5), 2.05 (br s, 1H,
CH2OH), 2.18 (dddd, J = 2.4 Hz, J = 3.5 Hz, J = 9.9 Hz, J = 13.5 Hz,
1H, H-4), 2.24–2.30 (m, 1H, H-7), 2.31 (s, 3H, CH3), 2.42–2.50 (m,
1H, H-3), 3.68 (dd, J = 6.34 Hz, J = 10.9 Hz, 1H, CH2OH), 3.82 (dd,

656
M. Jaworska et al. / Tetrahedron: Asymmetry 22 (2011) 648–657
4.8. 4-Bromo-2-((E)-((1S,2S,3R,5S)-3-(hydroxymethyl)-2,6,6-
trimethylbicyclo[3.1.1]-heptan-2-ylimino)methyl)phenol 4e
evaporated to obtain enantiomerically enriched 1-phenyl-1-propa-
nol. The crude product was purified by column chromatography
using appropriate eluents (hexane/ethyl acetate). The ee was
determined with HPLC using Chiralcel OD-H column or GC analysis
using b-Dex capillary column. Yields were determined via GC
analysis using ZB-5 capillary column or by isolation of the product
(Tables 2 and 3).
5-Bromo-2-hydroxybenzaldehyde was used as a substrate using
procedure A. Yellow solid (77%). Purified by recrystallization from
diethyl ether. Mp 59–62 °C, ½a D20
ꢁ
¼ þ51:5 (c 0.466, MeOH). 1H NMR
(700 MHz, CDCl3): d 1.21 (s, 3H, CH3 (bond to C-6)), 1.32 (d,
J = 10.7 Hz, 1H, H-7), 1.35 (s, 3H, CH3 (bond to C-6)), 1.49 (m, 1H,
H-4), 1.52 (s, 3H, CH3 (C-2)), 2.03 (t, J = 5.7 Hz, 1H, H-1), 2.04–
2.07 (m, 1H, H-5), 2.14–2.19 (m, 1H, H-4), 2.20 (br s, 1H, CH2OH),
2.29–2.33 (m, 1H, H-7), 2.45–2.50 (m, 1H, H-3), 3.66 (dd,
J = 6.3 Hz, J = 10.5 Hz, 1H, CH2OH), 3.74 (dd, J = 8.5 Hz, J = 11.5 Hz,
1H, CH2OH), 6.88 (d, J = 9.3 Hz, 1H, CH-aryl (C-3 in phenol ring)),
7.40 (m, 2H, CH-aryl (C-6 and C-4 in phenol ring)), 8.19 (s, 1H,
CH–imine), 14.58 (br s, 1H, OH). 13C NMR (176.1 MHz, CDCl3): d
23.64 (CH3), 28.22 (CH3), 28.70 (CH3), 29.50 (CH2), 31.15 (CH2),
38.88 (C), 40.08 (CH), 41.61 (CH), 53.55 (CH), 65.49 (C), 66.27
(CH2, CH2OH), 108.72 (C), 119.67 (C), 120.19 (CH), 133.89 (CH),
135.49 (CH), 160.15 (CH, imine), 163.07 (C). IR (nujol mull):
1627.0, 1280.0, 1154.3, 1074.0, 820.2, 722.4. Anal. Calcd for
4.11. Typical procedure for the asymmetric trimethylsilyl-
cyanation of aldehydes and measuring the ee’s of products
Schiff base 4c (71.5 mg, 0.20 mmol) and dry CH2C12 (2.5 ml)
were placed in a flame-dried Schlenk tube. To this solution at room
temperature Ti(O-i-Pr)4 (0.05 ml, 0.18 mmol) was added and the
resulting solution was stirred for 1 h under an N2 atmosphere.
Then the mixture was cooled to ꢀ20 °C and TMSCN (0.29 ml,
2.3 mmol) and then freshly distilled benzaldehyde (106 mg,
1.0 mmol) was added to the solution, and the whole was stirred
for 48 h at ꢀ20 °C. After this, the mixture was poured into a mix-
ture of 1 M HCl (30 ml) and ethyl acetate (150 ml) and stirred vig-
orously for 12 h at rt. The mixture was then extracted with ethyl
acetate (3 ꢄ 30 ml), and then the combined extracts were washed
with saturated NaHCO3 (4 ꢄ 50 ml) and brine (2 ꢄ 50 ml), dried
over anhydrous Na2SO4, and then evaporated. The residue was
purified by column chromatography on silica gel (eluent/hexane/
ethyl acetate = 5:1 (v/v)) to give the (R)-riched cyanohydrin, for
C18H24BrNO2: C, 59.02; H, 6.60; Br, 21.81; N, 3.82; O, 8.74. Found:
C, 59.15; H, 6.72; N, 3.60.
4.9. 5-(Diethylamino)-2-((E)-((1S,2S,3R,5S)-3-(hydroxymethyl)-
2,6,6-trimethylbicyclo-[3.1.1]heptan-2-ylimino)methyl)phenol
4f
example,
2-hydroxy-2-phenylacetonitrile
(97.09 mg,
73%),
a 2D0
ꢁ
¼ þ25:8 (c 2.32, CHC13) {lit.11b
½
a 2D4
ꢁ
¼ þ36:8 (c 2.00, CHCl3)
4-Diethylamino-2-hydroxybenzaldehyde was used as a sub-
strate using procedure B. Yellow solid (64%). Purified by recrystal-
lization from diethyl ether/hexane solution (1:1, v/v) (this
compound was purified using only recrystallization without using
½
for the (R)-enantiomer in 85% ee}, 1H NMR (CDCl3): d 7.38–7.60
(m, 5H), 5.54 (s, 1H), 2.92 (br s, 1H).
After measuring the specific rotation, the pure cyanohydrin
(93.09 mg) was converted directly into the corresponding acetate
by reaction with acetyl chloride (0.09 ml) and pyridine (0.04 ml)
in CH2Cl2 (4 ml) at room temperature under a nitrogen atmosphere
for 1 h. The reaction was quenched by water (5 ml) and extracted
with CH2Cl2 (3 ꢄ 10 ml). The organic extracts were washed with
brine (10 ml) and dried over anhydrous sodium sulfate. Evapora-
tion of solvents gave the corresponding acetate. The enantiomeric
excess of the cyanohydrin acetate was determined by HPLC meth-
od on a Chiralcel OD-H column, using hexane/2-propanol = 99:1,
flow rate = 0.7 ml/min, t(R) = 14.60 min, t(S) = 16.30 min.
the purification procedure described in the literature11a). Mp 151–
22
580
153 °C, ½a 2D0
ꢁ
¼ þ92:1 (c 0.038, MeOH) {lit.10
½
aꢁ
¼ þ13:1 (c 0.37,
CHCl3)}. 1H NMR (700 MHz, CDCl3): d 1.17 (s, 3H, CH3 (bond to C-
6)), 1.21 (t, J = 7.0 Hz, 6H, 2 ꢄ CH3 in –NEt2), 1.26 (d, J = 10.5 Hz, 1H,
H-7), 1.34 (s, 3H, CH3 (bond to C-6)), 1.51 (s, 3H, CH3 (C-2)), 1.52
(ddd, J = 2.0 Hz, J = 7.5 Hz, J = 14.0 Hz, 1H, H-4), 1.99-2.03 (m, 1H,
H-5), 2.07 (t, J = 5.5 Hz, 1H, H-1), 2.08–2.13 (m, 1H, H-4), 2.30–
2.34 (m, 1H, H-7), 2.40–2.45 (m, 1H, H-3), 2.53 (br s, 1H, CH2OH),
3.39 (q, J = 7.0 Hz, 4H, 2 ꢄ CH2 in –NEt2), 3.69 (dd, J = 5.6 Hz,
J = 11.0 Hz, 1H, CH2OH), 3.79 (dd, J = 9.1 Hz, J = 11.5 Hz, 1H,
CH2OH), 5.99 (d, J = 2.1 Hz, 1H, CH-aryl (C-5 in phenol ring)), 6.10
(dd, J = 2.1 Hz, J = 8.9 Hz, 1H, CH-aryl (C-3 in phenol ring)), 6.93
(d, J = 9.1 Hz, J = 7.5 Hz, 1H, CH-aryl (C-6 in phenol ring)), 7.76 (s,
1H, CH–imine), 14.60 (br s, 1H, OH). 13C NMR (176.1 MHz, CDCl3):
d 12.87 (2 ꢄ CH3 in –NEt2), 23.63 (CH3), 28.19 (CH3), 29.06 (CH3),
29.26 (CH2), 30.50 (CH2), 38.65 (C), 39.94 (CH), 41.23 (CH), 44.56
(2 ꢄ CH2), 53.13 (CH), 65.12 (C), 65.72 (CH2, CH2OH), 99.67 (CH),
102.87 (CH), 108.12 (C), 134.14 (CH), 153.31 (C), 157.36 (CH,
imine), 173.78 (C). IR (nujol mull): 3267.9, 2365.8, 1610.2,
1508.3, 1340.2, 1237.3, 1140.1, 1077.6, 1008.3, 827.9. Anal. Calcd
for C22H34N2O2: C, 73.70; H, 9.56; N, 7.81; O, 8.93. Found: C,
73.50; H, 9.48; N, 7.65.
4.12. Reaction of titanium (IV) isopropoxide with 1 equiv of
ligand 4c (for NMR measurements)
Under a nitrogen atmosphere, to the orange ligand 4c (0.0715 g,
0.2 mmol) solution in 3.5 ml of CDCl3 at 20 °C, 1 equiv of Ti(O-i-Pr)4
(0.0568 g, 0.06 ml, 0.2 mmol) was added. After stirring for 1 h the
dark red solution was transferred into 5 mm NMR tube and the
NMR spectrum was recorded on a 700 MHz Bruker Avance III spec-
trometer. The 13C{1H}NMR spectrum showed the presence of a C–H
resonance at 164.13 ppm assigned to C6 (see Fig. 5). This resonance
corresponds to a catalytic active complex LꢃTi(O-i-Pr)2.
4.10. Typical procedure for asymmetric addition of Et2Zn to
aldehydes
Acknowledgments
This work was supported in part by Marshal of Kujawsko-
Pomorskie Voivodeship—‘Scholarship for Ph.D. students 2008/
2009-ZPORR’ No. SPS.IV-3040-UE/265/2009 and by the Nicolaus
Copernicus University Grant UMK CH-526.
Diethylzinc (2 mmol, 2 ml, 1 M solution in hexanes) was added
to a solution of chiral Schiff base 4d (79.92 mg, 0.20 mmol) in dry
toluene (5 ml) at ꢀ70 °C under an atmosphere of N2. After 1 h, the
aldehyde (1 mmol) was added and the reaction mixture was
warmed to room temperature. The mixture was stirred for the
appropriate time (Scheme 3) and was quenched with a saturated
solution of ammonium chloride (5 ml). After 10 min of stirring,
the mixture was extracted with ethyl acetate (3 ꢄ 10 ml), dried
over anhydrous MgSO4 and gravity filtered. The solvent was
References
1. Irako, N.; Hamada, Y.; Shiori, T. Tetrahedron 1995, 51, 12731–12744.
2. Markowicz, S. W.; Pokrzeptowicz, K.; Karolak-Wojciechowska, J.; Czylkowski,
R.; Omelan´ czuk, J.; Sobczak, A. Tetrahedron: Asymmetry 2002, 13, 1981–1991.

M. Jaworska et al. / Tetrahedron: Asymmetry 22 (2011) 648–657
657
3. Szakonyi, Z.; Martinek, T.; Hetényi, A.; Fülöp, F. Tetrahedron: Asymmetry 2000,
11, 4571–4579.
1996, 50, 234–242; (d) Tanaka, K.; Mori, A.; Inoue, S. J. Org. Chem. 1990, 55,
181–185.
4. Midland, M. M.; Graham, R. S. Org. Synth. 1990, Coll. Vol. 7, 402.
5. Peng, F.; Hall, D. G. J. Am. Chem. Soc. 2007, 129, 3070–3071.
6. Dang, H.-S.; Diart, V.; Roberts, B. P. J. Chem. Soc., Perkin Trans. 1 1994, 1033–
1041.
13. Sasaki, T.; Eguchi, S.; Yamada, H. J. Org. Chem. 1973, 38, 679–686.
14. Jaworska, M.; Ła˛czkowski, K. Z.; Wełniak, M.; Welke, M.; Wojtczak, A. Appl.
Catal., A 2009, 357, 150–158.
15. Flack, H. D. Acta Crystallogr., Sect. A 1983, 39, 876–881.
16. Bauer, T.; Gajewiak, J. Tetrahedron: Asymmetry 2005, 16, 851–855.
17. ARGUSLAB 4.0.1, M. A. Thompson, Planaria Software LLC, Seattle, WA, http://
www.arguslab.com.
´
7. Ła˛czkowski, K. Z.; Pakulski, M. M.; Krzeminski, M. P.; Jaisankar, P.; Zaidlewicz,
M. Tetrahedron: Asymmetry 2008, 19, 788–795.
8. Masui, M.; Shioiri, T. Tetrahedron 1995, 30, 8363–8370.
9. Szakonyi, Z.; Balázs, A.; Martinek, T. A.; Fülöp, F. Tetrahedron: Asymmetry 2006,
17, 199–204.
10. Koneva, E. A.; Volcho, K. P.; Korchagina, D. V.; Komarova, N. I.; Kochnev, A. I.;
Salakhutdinov, N. F.; Tolstikov, A. G. Russ. Chem. Bull., Int. Ed. 2008, 57,
108–117.
11. (a) Tanaka, T.; Yasuda, Y.; Hayashi, M. J. Org. Chem. 2006, 71, 7091–7093; (b)
Hayashi, M.; Miyamoto, Y.; Inoue, T.; Oguni, N. J. Org. Chem. 1993, 58, 1515–
1522; (c) Sokeirik, Y. S.; Omote, M.; Sato, K.; Kumadaki, I.; Ando, A. Tetrahedron:
Asymmetry 2006, 17, 2654–2658; (d) Flores-Lopéz, L. Z.; Parra-Hake, M.;
Somanathan, R.; Walsh, P. J. Organometallics 2000, 19, 2153–2160.
12. (a) Bolshan, Y.; Chen, C.; Chilenski, J. R.; Gosselin, F.; Mathre, D. J.; O’Shea, P. D.;
Roy, A.; Tillyer, R. D. Org. Lett. 2004, 6, 111–114; (b) Ziegler, T.; Hörsch, B.;
Effenberger, F. Synthesis 1990, 575–578; (c) Kanerva, L. T. Acta Chem. Scand.
18. Allen, F. H. Acta Crystallogr., Sect. B 2002, 58, 380–388.
19. Zhang, H.; Xue, F.; Mak, T. C. W.; Chan, K. S. J. Org. Chem. 1996, 61, 8002–8003.
20. Kitamura, M.; Okada, S.; Suga, S.; Noyori, R. J. Am. Chem. Soc. 1989, 111, 4028–
4036.
21. Li, Y.; He, B.; Qin, B.; Feng, X.; Zhang, G. J. Org. Chem. 2004, 69, 7910–7913.
22. CrysAlis CCD171 and RED171 Package of Programs, Oxford Diffraction, 2000.
23. Sheldrick, G. M. SHELXS97 and SHELXL97, University of Goettingen, Germany,
1997.
24. Larrow, J. F.; Jacobsen, E. N.; Gao, Y.; Hong, Y.; Nie, X.; Zepp, C. M. J. Org. Chem.
1994, 59, 1939–1942.
25. Claus, P.; Schilling, P.; Gratzl, J. S.; Kratzl, K. Monatsch. Chem. 1972, 103, 1178–
1193.
26. Huber, W. Helv. Chim. Acta 1953, 36, 1109–1115.