A highly enantioselective Strecker reaction catalyzed by titanium-N-salicyl-β-aminoalcohol complexes

Article abstract of DOI:10.1016/j.tet.2004.07.097

N-salicyl-β-amino alcohols 1 were synthesized and evaluated as ligands for catalytic asymmetric Strecker reactions. N-Benzhydrylaldimines derived from aromatic and aliphatic aldehydes reacted with TMSCN in the presence of 10 mol% of Ti-1 complex to give t

Full text of DOI:10.1016/j.tet.2004.07.097

Tetrahedron 60 (2004) 10559–10568
A highly enantioselective Strecker reaction catalyzed by
titanium-N-salicyl-b-aminoalcohol complexes
Vorawit Banphavichit, Woraluk Mansawat, Worawan Bhanthumnavin
*
and Tirayut Vilaivan
Organic Synthesis Research Unit, Department of Chemistry, Faculty of Science, Chulalongkorn University, Phayathai Road, Patumwan,
Bangkok 10330, Thailand
Received 23 April 2004; revised 17 June 2004; accepted 28 July 2004
Available online 15 September 2004
Abstract—N-salicyl-b-amino alcohols 1 were synthesized and evaluated as ligands for catalytic asymmetric Strecker reactions.
N-Benzhydrylaldimines derived from aromatic and aliphatic aldehydes reacted with TMSCN in the presence of 10 mol% of Ti-1 complex to
give the Strecker products in excellent yields and in up to O98% ee. The presence of a protic additive is essential to ensure good conversion
and reaction rate. The reaction conditions are simple and the stereochemical outcome is predictable from the configuration of the ligands,
both enantiomers of which are readily synthesized.
q 2004 Elsevier Ltd. All rights reserved.
1. Introduction
few stereogenic centers as possible, provided that
good reactivity and enantioselectivity can still be
retained.
The Strecker synthesis is one of the most atom-economical
methods for the synthesis of a-amino acids. The classical
Strecker reaction has been known since 1850, but it was not
until 1963 that the first asymmetric Strecker synthesis was
reported.1 The majority of asymmetric Strecker syntheses
developed since then involve the use of chiral auxiliaries,
generally attached to the imine moiety.2 Catalytic versions
of asymmetric Strecker reactions have a relatively recent
history.3 In 1996, Lipton reported the first enantioselective
Strecker reaction catalyzed by a cyclic dipeptide bearing a
guanidine side-chain.4 In less than 10 years, a number of
highly effective catalysts have subsequently emerged. These
catalysts can be entirely organic molecules such as Corey’s
cyclic guanidine5 and Jacobsen’s urea-Schiff base.6 More
recently, a number of promising chiral Lewis acid catalysts
based on metal complexes of ligands such as BINOL7 Schiff
base,8 bisoxazoline9 and a carbohydrate derivative10 have
been developed.
Scheme 1.
Although many of these catalysts provide excellent
enantioselectivities for a wide range of substrates, the
most efficient ones are still fairly large and complex
molecules with multiple stereogenic centers. From a
practical point of view, it would be highly desirable
to develop a small molecule catalyst possessing as
The tridentate N-salicyl-b-aminoalcohol 1 and its parent
Schiff base 2 should form chelating complexes with metals
such as titanium or aluminium. These complexes should
Keywords: Asymmetric; Catalyst; Imine; Hydrocyanation.
* Corresponding author. Tel.: C2-218-7627; fax: C2-218-7598;
e-mail: vtirayut@chula.ac.th
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.07.097

10560
V. Banphavichit et al. / Tetrahedron 60 (2004) 10559–10568
behave as Lewis acids whilst providing a rigid asymmetric
environment and, therefore, be potentially useful for
catalyzing various types of asymmetric reactions. For
instance, a heterobimetallic Li–Al complex of 1 was
reported to catalyze the Michael addition of dialkyl
malonates to cyclic enones.11 Titanium complexes of 2
are highly efficient catalysts for enantioselective synthesis
of cyanohydrins.12 In view of the mechanistic similarity
between hydrocyanation of aldehydes and imines, we
envisaged that such catalysts would also perform well in
the enantioselective Strecker reaction. The results of our
preliminary investigation indicated that chiral N-salicyl-b-
adducts is usually performed by chiral HPLC, this method
was not available to us at the time of this investigation. We
considered 1H NMR spectroscopy as a potential alternative
technique since it should require only a short analysis time,
consume less solvent, require no expensive accessories such
as chiral columns, and most importantly, the sample can be
analyzed without any pre-treatment apart from removal of
the reaction solvents. The only remaining problem was to
find a suitable chiral media to distinguish the signals of each
enantiomer. After trying several different chiral shift or
chiral solvating agents, optically active camphorsulfonic
acids (CSAs) previously used by Hassan et al. proved to be
most effective in resolving the signals of the racemic
N-benzyl a-aminonitrile (4a).15 The splitting of the CaH
signal by up to 0.14 ppm was observed on a 200 MHz NMR
spectrometer upon addition of 1–2 equiv of (S)-CSA to a
0.1 M solution of 4a in CDCl3. The CaH and Ph2CH peaks
of N-benzhydryl substituted a-aminonitrile (4b) also split in
the presence of CSA although the resolution was to a lesser
magnitude than that of the N-benzyl analogues (up to
0.03 ppm for CaH signal and 0.05 ppm for Ph2CH signal).
Nonetheless, a baseline resolution was obtained on a
400 MHz NMR spectrometer. Only a single CaH peak
was detected when enantiomerically pure (S)-4b was treated
with CSA. The ‘satellite’ peaks can be easily detected in
enantiomerically enriched samples containing as little as 1%
of the other enantiomer (98% ee) prepared by addition of
racemic 4b to enantiomerically pure 4b (Fig. 1). No
racemization was observed after the CSA-treated sample
was left at room temperature for several days. As expected,
the position of the enantiomeric CaH peaks reversed if the
aminoalcohols
were effective catalysts for asymmetric Strecker reactions
of N-benzyl substituted imines with TMSCN
1
in combination with Ti(OiPr)4
3
(Scheme 1).13,14 The reported enantioselectivities were,
however, rather moderate (maximum 86% ee). We have
now discovered that the N-substituent of the imine
substrate exerts a very significant effect on the degree of
enantioselective induction. In this paper we disclose the
results of our investigation on enantioselective
Strecker reactions employing this versatile class of ligands
in detail.
2. Results and discussion
2.1. Development of an NMR method for determination
of enantiomeric purity
Although analysis of the enantiomeric purity of Strecker
1
Figure 1. Partial H NMR spectra of aminonitrile 4b (400 MHz, CDCl3) in the presence of (S)-CSA showing the Ph2CH (dR: 5.48; dS: 5.42 ppm) and CaH
signals (dR: 5.26; dS: 5.29 ppm): racemic (top); 98% ee (middle); and enantiomerically pure (S)-4b (bottom). The humps marked by ‘x’ denote satellites due to
1H–13C couplings (1J1H–13Cw140 Hz).

V. Banphavichit et al. / Tetrahedron 60 (2004) 10559–10568
Table 1. Structure and yield of the ligandsa 1a–1k
10561
R0 (b)
R00 (a)
Entry
Ligand
Amino alcohols
R (N)
Yield (%)
1
2
3
4
5
6
7
8
(S)-1a
(R)-1a
(S)-1b
(S)-1c
(S)-1d
(S)-1e
(S)-1f
(S)-1g
(S)-phenylalaninol
(R)-phenylalaninol
(S)-alaninol
H
H
H
H
H
H
H
H
PhCH2
PhCH2
Me
H
H
H
H
H
H
H
H
74
72
51
75
74
73
50
71
(S)-valinol
iPr
(S)-tert-leucinol
(S)-phenylglycinol
(S)-leucinol
(S)-cyclohexyl-ala-
ninol
(S,S)-isoleucinol
(R)-1-aminopropan-
2-ol
tBu
Ph
iBu
cHexCH2
9
10
(S)-1h
(R)-1i
H
H
secBu
H
H
CH3
74
49
11
12
(S)-1j
(S)-1k
(S)-prolinol
(S)-phenylalaninol
–(CH2)3–
PhCH2
H
H
50
59
H
a
XZH for 1a–1j (starting from salicylaldehyde); XZtBu for 1k (starting from 3,5-di-tertbutylsalicylaldehyde).
opposite enantiomer of CSA was added. Importantly, CSA
is very general in inducing a chiral environment for
many other racemic N-benzyl and N-benzhydryl substituted
a-aminonitriles (4) carrying different aromatic and aliphatic
a-substituents. Therefore, we conclude that 1H NMR is
indeed a very reliable method for the determination of
enantiomeric purity of a variety of Strecker adducts.
ligand was obtained as a stable white crystalline solid after
routine work-up and column chromatography. It can be kept
for years at room temperature without detectable degra-
dation. Many other related ligands can be similarly prepared
from the corresponding b-aminoalcohols in 50–75%
yield (Table 1).
2.3. Evaluation of the ligands for catalytic asymmetric
Strecker reaction and optimization of the reaction
conditions
2.2. Synthesis of the ligands
The N-salicyl b-aminoalcohol ligand of type 1 is known in
the literature. It has been synthesized by several methods
including reduction of the salicylimines derived from the
corresponding a-amino esters16 or aminoalcohols.17 Since a
number of chiral b-aminoalcohols are readily available, we
chose to synthesize 1 from the corresponding Schiff bases.
Ligand (S)-1a was synthesized in 74% yield from
commercially available (S)-phenylalaninol and salicylalde-
hyde via the Schiff base (S)-2a by NaBH4 reduction. The
Several 1:1 metal complexes of (S)-1a and its parent Schiff
base (S)-2a were prepared in situ and tested as catalysts for
enantioselective Strecker reactions. The reaction between
N-benzylidenebenzylamine (3: RZPh; R0ZPhCH2) and
TMSCN was used as a model. The reaction was performed
in toluene at 0 8C in the presence of 10 mol% of the catalyst.
Excess TMSCN (2 equiv) was essential to assure a complete
reaction. While many complexes of both ligands provided
Table 2. Effect of solvent and temperature
Entry
1
Solvent
Catalyst (mol %)
10
t (8C)
Time (h)
6
Conversion (%)
97
ee (%)
72
HexanesCPhMe
(1:1)
Et2O
0
2
3
4
5
6
7
8
9
10
10
10
10
10
10
10
1
0
0
0
0
0
K20
25
0
0
0
6
6
6
6
6
12
6
8
8
8
98
90
98
96
98
36
97
77
52
48
72
0
79
8
62
24
46
74
75
THF
CH2Cl2
MeOH
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
10
11
12
5
15
20
96
O99
O99
0
8

10562
V. Banphavichit et al. / Tetrahedron 60 (2004) 10559–10568
Table 3. Effect of ligand structure
Entry
Ligand
R ligand (position)
Time (h)
Conversion (%)
ee (%)
1
2
3
4
5
6
7
8
(S)-1a
(S)-1b
(S)-1c
(S)-1d
(S)-1e
(S)-1f
(S)-1g
(S)-1h
(R)-1i
(S)-1j
(S)-1k
PhCH2 (b)
Me (b)
6
6
6
6
6
6
6
6
8
8
8
98
97
99
98
98
97
98
98
81
77
84
79 (S)
20 (S)
82 (S)
86 (S)
66 (S)
48 (S)
77 (S)
87 (S)
6 (R)
iPr (b)
tBu (b)
Ph (b)
iBu (b)
cHexCH2 (b)
secBu (b)
Me (a)
9
10
11
(CH2)3 (N, b)
PhCH2 (b)
8 (R)
16 (S)
the desired Strecker product (4a) in good to excellent yields,
only the titanium complex of the reduced Schiff base (S)-1a
gave a satisfactory outcome in terms of both conversion and
enantioselectivity. Interestingly, the presence of TMS
groups was not observed in the Strecker adduct, presumably
due to the cleavage of the rather labile N–Si bond by trace of
moisture during work-up. The reaction parameters includ-
ing solvent, temperature and catalyst loading were next
optimized using this screening system (Table 2). It was
evident that toluene is the best solvent for the reaction. Polar
solvents resulted in lower degrees of enantioselectivity. The
effect of temperature is also quite interesting, the best ee was
obtained when the reactions were carried out at 0 8C. Not
unexpectedly, higher temperatures resulted in poorer
selectivity. At temperatures below 0 8C, the reaction was
very slow, and interestingly the enantioselectivity dropped
substantially. Decreasing the catalyst loading resulted in
lower enantioselectivity while the conversion was not much
affected. On the other hand, no improvement in selectivity
was found at O10% catalyst loading, therefore, the optimal
catalyst loading was 10%.
2.4. Effects of ligand structure
In order to investigate the effect of the ligand structure on
the enantioselectivities, the reactions of imine 3 (RZPh;
R0ZPhCH2) with TMSCN in the presence of other catalysts
were repeated under the best condition obtained for ligand
(S)-1a. A good correlation, at least in a qualitative sense,
between the size of the alkyl side-chain and the degree of
enantioselectivity was observed. Therefore, only ligands
bearing a relatively sterically hindered substituent at the
b-position such as 1a (RZPhCH2), 1c (RZiPr), 1d
(RZtBu), and 1h (RZsecBu) provided synthetically useful
selectivities (Table 3: entries 1, 3, 4 and 8). Addition of a
bulky tert-butyl substituent on the salicyl moiety resulted in
Table 4. Enantioselective Strecker reactions of aromatic benzyl and benzhydrylimines (3) catalyzed by Ti(OiPr)4-(S)-1a complexesa,b
R0
Entry
R (substrate)
Conversion (%)
ee (%)
1
2
3
4
5
6
7
8
Ph
PhCH2
PhCH2
PhCH2
PhCH2
PhCH2
PhCH2
Ph2CH
Ph2CH
Ph2CH
Ph2CH
Ph2CH
Ph2CH
Ph2CH
Ph2CH
Ph2CH
Ph2CH
Ph2CH
Ph2CH
Ph2CH
Ph2CH
98
96
97
93
79
72
67
45
64
51
98
95
96
91
94
98
4-ClC6H4
4-MeC6H4
4-MeOC6H4
3-NO2C6H4
2-MeOC6H4
Ph
4-ClC6H4
4-MeC6H4
4-MeOC6H4
3-MeOC6H4
3-NO2C6H4
2-MeC6H4
2-ClC6H4
2-BrC6H4
2-MeOC6H4
1-Naphthyl
2-Naphthyl
2-Furyl
O99
O99
O99
O99
O99
O99
O99
O99
O99
O99
O99
O99
98
9
10
11
12
13
14
15
16
17
18
19
20
97
97c
O98c
90
98
96
98
98
97
O99
O99
2-Thienyl
a
Reaction time: entries 1–6Z9 h; entries 7–20Z48 h.
For reaction conditions see Table 3.
b
c
Baseline separation of the enantiomeric CaH and Ph2CH signals was not achieved. The reported ee values were estimated by comparison of the optical
rotation with known reference compounds (Ref. 8b).

V. Banphavichit et al. / Tetrahedron 60 (2004) 10559–10568
10563
Table 5. Enantioselective Strecker reactions of aliphatic benzhydrylimines catalyzed by Ti(OiPr)4-1 complexesa,b
Entry
R (substrate)
Ligand
conversion (%)
ee (%)
1
2
3
4
5
6
tBu
(S)-1a
(S)-1d
(S)-1h
(S)-1a
(S)-1d
(S)-1h
O99
O99
O99
O99
O99
O99
23
47
51
61
88
91
tBu
tBu
PhCHZCH
PhCHZCH
PhCHZCH
a
Reaction time: 24 h.
For reaction conditions see Table 3.
b
a substantially decreased enantioselectivity (entry 11). In all
aforementioned cases, the absolute configuration of the
Strecker product 4a was determined to be S based on the
chemical shift of the CaH signal in the presence of (S)-CSA
and by comparison of the optical rotation with a known
standard.15 Moving the substituent to the a-position also
resulted in a poor enantioselectivity (compare entries 2 and
9). Ligand (S)-1j bearing an N-alkyl substituent which is
part of a pyrrolidine ring gave, apart from a very low
selectivity, the product of opposite absolute configuration
(entry 10).
values of O90% were routinely observed for various imine
substrates derived from aromatic aldehydes and benz-
hydrylamine (Table 4, entries 8–20). Substituents with
various steric and electronic natures appeared to be well-
tolerated for virtually all imines derived from substituted
benzaldehyde studied. Equally good results were obtained
for non-benzaldehyde derived aromatic imines including
1-naphthyl, 2-naphthyl and heteroaromatic substrates.
Aliphatic substrates such as those derived from pivalalde-
hyde and cinnamaldehyde gave much poorer ee’s at similar
conversion. Nevertheless, the use of ligands bearing bulkier
substituents such as 1d and 1h gave significantly improved
results (Table 5). The S-configuration of the Strecker
products was assumed in all cases (except for entries
19–20 in Table 4, vide infra) by analogy to 4a and 4b.
2.5. Effects of substrate structure
The substrate generality of enantioselective Strecker reac-
tions employing this class of ligand was investigated.
Various substituted aldimines were subjected to cyanide
addition under the best conditions obtained thus far using
the Ti complex of (S)-1a. It was found that optically active
a-aminonitriles 4 were obtained in excellent yields and with
enantioselectivity ranging from poor to fairly good (45–
79% ee) for N-benzyl substituted aldimine substrates (3:
RZPh; R0ZPhCH2) (Table 4, entries 1–6). However, a
dramatic breakthrough was achieved when N-benzhydryl
substituted aldimines (3: R0ZPh2CH) were used as
substrates. The Strecker reaction of N-benzhydrylimine (3:
RZPh; R0ZPh2CH) in the presence of Ti(OiPr)4-(S)-1a
required somewhat longer reaction time to reach O99%
conversion, but provided the corresponding optically active
a-aminonitrile in 98% ee under otherwise identical
conditions (Table 4, entry 7). The bulkier ligands 1c, 1d
and 1h also gave the same conversion (O99%) and ee
range for this substrate (1c: 94%; 1d and 1h: 97%). Ee
2.6. The role of protic additives
Upon scaling up the reaction from 0.1 mmol to 0.5 or
1 mmol, the rate of the reaction was found to be
significantly slower. In many cases the reaction did not
proceed to completion at all although the ee values were
practically the same. Previous works in this area have
demonstrated the importance of proton sources in similar
reactions.7a,8b The proton source can generate HCN from
TMSCN, which is believed to be the actual cyanating
species.8b Furthermore, since the catalyst contains free
hydroxylic functions which should be susceptible to
silylation by TMSCN under the reaction conditions, a
proton source should increase the catalyst turnover by
preventing silylation of, or by regenerating, the hydroxylic
function of the catalyst.18 We were delighted to find that
addition of a proton source such as water or 2-propanol
Figure 2. The rate of cyanation of imine 3 (RZPh; R0ZPh2CH) at 0 8C the presence of 10 mol% Ti(OiPr)4-(S)-1a complex and 0, 0.2, 1.0 and 10 equiv of
2-propanol (left) and in the presence of 10 equiv 2-propanol and with or without Ti(OiPr)4 and/or (S)-1a (10 mol%) (right).

10564
V. Banphavichit et al. / Tetrahedron 60 (2004) 10559–10568
completely restored the catalytic activity. Kinetic analysis
by 1H NMR spectroscopy of the Ti-(S)-1a catalyzed
cyanation of imine 3 (RZPh; R0ZPh2CH) conducted in
the presence of different amounts of 2-propanol additive
suggested that the reaction rate was significantly enhanced
compared to a reaction without the additive (Fig. 2, left).
Most interestingly, the reaction rate reaches a maximum
when an equivalent amount of the additive is used and under
these conditions. The reaction went to completion within 2 h
at 0 8C. A large excess of 2-propanol (10 equiv) resulted not
only in a significantly slower reaction rate, but also provided
almost racemic product. The reactions in the presence of
1.0 equiv iPrOH but without the Ti-(S)-1a catalyst or either
of its components were very slow (Fig. 2, right). In all cases
the conversion was !20% after 2 h at 0 8C. This suggested
that the background hydrocyanation is almost insignificant
under the experimental conditions, at least for this particular
i
substrate. Consequently, the slow addition of PrOH or
performing the reaction at very low temperatures may not be
essential to ensure good enantioselectivities. This would
greatly simplify the synthetic procedure. Indeed, we have
found that the additive can be added at the beginning of the
reaction without any adverse effects on the yield and
enantioselectivity.19 The role of the protic additive is
probably less important for small scale reactions where the
total exclusion of all adventitious moisture was difficult
which explains why the reaction proceeded to completion
without the need for the protic additive.
Scheme 2.
To ascertain that the protic additive is generally beneficial
for large scale reactions with other substrates, a few more
representative reactions were attempted at 1 mmol scale
with 1.0 equiv of 2-propanol added at the beginning and
10 mol% catalyst prepared from ligand (S)-1a (25 mg of the
1
ligand/mmol of substrate). H NMR analysis of the crude
reaction mixture revealed that all reactions were completed
within 4 h at 0 8C, and that the enantioselectivities were
comparable to the small scale reactions. Interestingly, after
attempted purification by passing through silica gel or an
alumina column, significant racemization was observed
for products 4 bearing electron donating substituents (e.g.,
2-furyl, 2-thienyl, 2-and 4-methoxyphenyls) so that the final
products of only 50–75% ee were obtained. We have
confirmed that the racemization was catalyzed by weak
acids including silica gel and neutral alumina. Fortunately,
addition of a small amount of triethylamine to the eluting
solvent largely suppressed the racemization. In all cases, the
isolated yield of the crystalline products were O80% and
the enantioselectivities were good to excellent (Scheme 2).
The absolute configuration of all products derived from the
ligand (S)-1a was confirmed to be S by comparison of the
optical rotations with literature values.8b It should be noted
that, for RZ2-thienyl and 2-furyl, although the sense of
asymmetric induction is the same, the absolute con-
figuration must be designated as R according to Cahn–
Ingold–Prelog sequence rules. The stereoselectivity of the
enantioselective Strecker reaction induced by the Ti-1
catalyst is, therefore, fully controlled by the absolute
configuration of the ligand 1 (Scheme 3).
Scheme 3.
one stereogenic center in the ligand moiety means that the
other enantiomer is much more easily obtained than those
with multiple stereogenic centers. To further demonstrate
the application of this novel catalyst system, the two
enantiomers of the optically active a-aminonitrile precursor
of 1-naphthylglycine were prepared. The reaction of the
imine 3 (RZ1-naphthyl; R0ZPh2CH) with TMSCN and
1.0 equiv PrOH in the presence of 10 mol% Ti(OiPr)4-(S)-
i
1a on a 1 mmol scale afforded the corresponding (S)-a-
aminonitrile in 91% yield and O98% ee. The opposite
enantiomer was similarly obtained in 87% yield and O98%
ee by employing (R)-1a derived from commercially
available (R)-phenylalaninol as the ligand.
3. Conclusions
It is obvious that aminonitriles with opposite configuration
should be obtained in a highly predictable manner simply by
switching to the enantiomeric ligand. The presence of only
In summary, this work has demonstrated that Ti complexes
of N-salicyl-b-aminoalcohols 1 are highly effective catalysts

V. Banphavichit et al. / Tetrahedron 60 (2004) 10559–10568
10565
for enantioselective Strecker reactions of aldimines. It has
been clearly shown that the configuration as well as the
bulkiness of the b-substituent exerts a direct influence on
both the absolute stereochemical outcome and enantio-
selectivity of the Strecker product. Furthermore, we have
illustrated that excellent yields accompanied by extremely
high degrees of enantioselectivity can be obtained when a
substrate with a more sterically demanding N-substituent
was employed. The superiority of this class of optically
active ligands can be emphasized due to their low molecular
weights and the fact that they possess only one stereogenic
center, the starting precursor of which is readily available
with any desirable absolute stereochemistry. In addition, it
has been shown that the reaction conditions are extremely
simple. It is expected that these catalyst systems will offer a
very practical access to optically active a-aminonitriles and
a-amino acids. Further understanding of the mechanistic
details of the catalysis and efforts to improve the catalysts to
accommodate aliphatic substrates are currently in progress.
100 MHz); d 16.1, 49.8, 53.9, 65.6, 116.5, 119.1, 121.7,
128.4, 128.9, 158.0; Anal. Calcd for C10H15NO2: C, 66.27;
H, 8.34; N, 7.73. Found: C, 66.11; H, 8.50; N, 7.75%.
4.1.3. N-(20-Hydroxyphenyl)methyl-(S)-2-amino-3-
methyl-butanol [(S)-1c]. White crystalline solid (0.157 g,
1
75%). Mp 52–54 8C; [a]26DZC15.0 (c 1.0, CHCl3); H
NMR (CDCl3, 400 MHz) d 1.01 and 1.04 (6H, 2!d, JZ
6.8 Hz, 2!CH3), 1.99 (1H, m, CH(CH3)2), 2.54 (1H, m,
CHNH), 3.69 (1H, ABX, JABZ11.1 Hz, JBXZ6.0 Hz,
CHaHbOH), 3.87 (1H, ABX, JABZ11.1 Hz, JAXZ3.8 Hz,
CHaHbOH), 4.05 (2H, s, CH2NH), 6.82 (1H, apparent t, JZ
7.4 Hz, Ar), 6.88 (1H, d, JZ8.0 Hz, Ar), 7.03 (1H, d, JZ
7.6 Hz, Ar), 7.21 (1H, apparent t, JZ7.7 Hz, Ar); 13C NMR
(CDCl3, 100 MHz) d 18.3, 19.2, 28.7, 51.0, 61.2, 64.0,
116.5, 119.1, 123.1, 128.2, 128.8, 158.1; Anal. Calcd for
C12H19NO2: C, 68.87; H, 9.15; N, 6.69; Found: C, 68.65; H,
9.39; N, 6.73%.
4.1.4. N-(20-Hydroxyphenyl)methyl-(S)-2-amino-3,3-
dimethyl-butanol [(S)-1d]. White crystalline solid
(0.165 g, 74%). Mp 58K60 8C; [a]2D5ZC5.0 (c 1.0,
CHCl3); H NMR (CDCl3, 400 MHz) d 1.04 (9H, s, 3!
CH3), 2.41 (1H, ABX, JAXZ3.6 Hz, JBXZ5.8 Hz,
4. Experimental
1
4.1. General procedure for the preparation of ligands 1
CHC(CH3)3), 3.74 (1H, ABX, JABZ10.9 Hz, JBX
5.8 Hz, CHaHbOH), 4.00 (1H, ABX, JABZ10.9 Hz, JAX
Z
Z
A mixture of ethanol (2 mL), an appropriate amino alcohol
(1 mmol), and aldehyde (1 mmol) was stirred at room
temperature until the starting materials were totally
consumed (monitored by TLC). To the yellow mixture
was added NaBH4 (37.8 mg, 1 mmol) with stirring to give a
colorless solution. The reaction was then quenched with dil.
HCl. After neutralization (sat. NaHCO3), the reaction
mixture was concentrated under reduced pressure and the
residue was extracted with ethyl acetate and washed with
water. The organic layer was dried over anhydrous Na2SO4
and the solvent removed. The products were purified by
flash column chromatography (hexanes/ethyl acetate).
3.6, CHaHbOH), 4.02 and 4.10 (2H, AB, JZ13.3 Hz,
CH2NH), 6.82 (1H, apparent t, JZ7.4 Hz, Ar), 6.89 (1H, d,
JZ8.0 Hz, Ar), 7.04 (1H, d, JZ7.2 Hz, Ar), 7.21 (1H,
apparent t, JZ7.8 Hz, Ar); 13C NMR (CDCl3, 100 MHz) d
27.5, 34.2, 53.1, 61.2, 67.8, 116.4, 119.3, 123.7, 128.5,
128.8, 157.9; Anal. Calcd for C13H21NO2: C, 69.92; H,
9.48; N, 6.27; Found: C, 69.95; H, 9.52; N, 6.27%.
4.1.5. N-(20-Hydroxyphenyl)methyl-(S)-2-amino-2-
phenyl-ethanol [(S)-1e]. White crystalline solid (0.201 g,
1
73%). Mp 119–121 8C; [a]2D0ZC64.0 (c 1.0, CHCl3); H
4.1.1. N-(20-Hydroxyphenyl)methyl-(S)-2-amino-3-
phenyl-propanol [(S)-1a]. White crystalline solid
(0.190 g, 74%). Mp 133–134 8C; [a]2D3ZK23.0 (c 1.0,
NMR (CDCl3, 400 MHz) d 3.74 (1H, AB, JZ13.6 Hz,
CHaHbNH, 3.76–3.87 (3H, m, CH2OH and CHNH), 3.97
(1H, AB, JZ13.6 Hz, CHaHbNH), 5.05 (br s, NH and OH),
6.79 (1H, apparent t, JZ7.2 Hz, Ar), 6.87 (1H, d, JZ
8.0 Hz, Ar), 6.92 (1H, d, JZ6.4 Hz, Ar), 7.19 (1H, apparent
t, JZ8.0 Hz, Ar); 13C NMR (CDCl3, 100 MHz) d 50.1, 63.9,
66.5, 116.4, 119.4, 122.8, 127.5, 128.2, 128.6, 128.9, 129.0,
138.7, 157.8; Anal. Calcd for C15H17NO2: C, 74.05; H,
7.04; N, 5.76. Found: C, 74.07; H, 7.03; N, 5.79%.
1
CHCl3); H NMR (CDCl3, 400 MHz) d 2.89 (1H, ABX,
JABZ13.6 Hz, JBXZ7.6 Hz, CHaHbPh), 2.91 (ABX, JAB
Z
13.6 Hz, JAXZ6.4 Hz, CHaHbPh), 3.02 (1H, m, CHNH),
3.58 (1H, ABX, JABZ11.0 Hz, JBXZ5.0 Hz, CHaHbOH),
3.78 (1H, ABX, JABZ11.0 Hz, JAXZ3.8 Hz, CHaHbOH),
4.05 (2H, s, CH2NH), 6.82 (1H, apparent t, JZ7.2 Hz, Ar),
6.87 (1H, d, JZ8.0 Hz, Ar), 7.01 (1H, d, JZ7.2 Hz, Ar),
7.19–7.38 (6H, m, Ar); 13C NMR (CDCl3, 100 MHz) d 37.3,
50.3, 59.7, 62.6, 116.6, 119.2, 122.7, 126.7, 128.3, 128.7,
128.9, 129.2, 138.0, 158.0; Anal. Calcd for C16H19NO2: C,
74.68; H, 7.44; N, 5.44. Found: C, 74.66; H, 7.41; N, 5.43%.
4.1.6. N-(20-Hydroxyphenyl)methyl-(S)-2-amino-4-
methyl-pentanol [(S)-1f]. Light yellow crystalline solid
(0.114 g, 50%). Mp 87–88 8C; [a]2D2ZC15.0 (c 1.0,
1
CHCl3); H NMR (CDCl3, 400 MHz) d 0.95 (6H, t, JZ
6.2 Hz, 2!CH3), 1.36 and 1.44 (2H, 2!m, CH2-
CH(CH3)2), 1.74 (1H, m, CH(CH3)2), 2.81 (1H, m,
CHNH), 3.56 (1H, ABX, JABZ11.1 Hz, JBXZ5.6 Hz,
CHaHbOH), 3.84 (1H, ABX, JABZ11.1 Hz, JAXZ3.4 Hz,
CHaHbOH), 4.04 (2H, s, CH2NH), 4.90 (br s, NH and OH),
6.79 (1H, apparent t, JZ6.4 Hz, Ar); 6.87 (1H, d, JZ
8.0 Hz, Ar) 7.02 (1H, d, JZ6.8 Hz, Ar), 7.19 (1H, apparent
t, JZ7.8 Hz, Ar); 13C NMR (CDCl3, 100 MHz) d 22.7, 22.8,
25.0, 40.2, 49.8, 56.2, 63.2, 116.5, 119.1, 122.8, 128.3,
128.7, 158.1; Anal. Calcd for C13H21NO2: C, 69.92; H,
9.48; N, 6.27. Found: C, 69.74; H, 9.55; N, 6.04%
4.1.2. N-(20-Hydroxyphenyl)methyl-(S)-2-amino-propa-
nol [(S)-1b]. Viscous yellowish oil (0.092 g, 51%).
[a]2D5ZC66.0 (c 1.5, CHCl3); 1H NMR (CDCl3,
400 MHz); d 1.17 (3H, d, JZ6.4 Hz, CH3), 2.92 (1H, m,
CHCH3), 3.54 (1H, ABX, JABZ10.9 Hz, JBXZ6.6 Hz,
CHaHbOH), 3.75 (1H, ABX, JABZ10.9 Hz, JAXZ3.6 Hz,
CHaHbOH), 4.00 and 4.10 (2H, AB, JZ13.8 Hz, CH2NH),
4.20 (br s, NH and OH) 6.81 (1H, apparent t, JZ7.4 Hz, Ar),
6.86 (1H, d, JZ8.0 Hz, Ar), 7.03 (1H, d, JZ7.2 Hz, Ar),
7.19 (1H, apparent t, JZ7.6 Hz, Ar); 13C NMR (CDCl3,

10566
V. Banphavichit et al. / Tetrahedron 60 (2004) 10559–10568
4.1.7. N-(20-Hydroxyphenyl)methyl-(S)-2-amino-3-cyclo-
hexyl-propanol [(S)-1g]. Light yellow crystalline solid
(0.200 g, 71%). Mp 82–84 8C; [a]2D5ZC12.0 (c 1.0,
(S)-2-amino-3-phenyl-propanol [(S)-1k]. Viscous yellow-
ish oil (0.219 g, 59%). [a]2D2ZK33.0 (c 2.6, CHCl3); H
1
NMR (CDCl3, 400 MHz); d 1.32 and 1.45 (2!9H, 2!s,
C(CH3)3)), 2.86 (1H, ABX, JABZ13.6 Hz, JBXZ7.6 Hz,
CHaHbPh), 2.97 (1H, ABX, JABZ13.4 Hz, JAXZ6.4 Hz,
1
CHCl3); H NMR (CDCl3, 400 MHz); d 0.85–1.78 (13H,
m, cHex ring protons and cHexCH2), 2.83 (1H, m, CHNH),
3.55 (1H, ABX, JABZ11.1 Hz, JBXZ5.2 Hz, CCHaHbOH),
3.82 (1H, ABX, JABZ11.1 Hz, JAXZ3.4 Hz, CHaHbOH),
4.03 (2H, s, CH2NH), 5.02 (br s, NH and OH), 6.81 (1H,
apparent t, JZ7.2 Hz, Ar), 6.87 (1H, d, JZ8.0 Hz, Ar), 7.02
(1H, d, JZ7.2 Hz, Ar), 7.19 (1H, apprent t, JZ7.8 Hz, Ar);
13C NMR (CDCl3, 100 MHz) d 26.2, 26.5, 33.6, 34.5, 38.6,
49.7, 55.5, 63.2, 116.5, 119.1, 122.8, 128.4, 128.9, 158.1;
Anal. Calcd for C16H25NO2: C, 72.96; H, 9.57; N, 5.32.
Found: C, 73.16; H, 9.49; N, 5.46%.
CHaHbPh), 3.03 (1H, m, CHNH), 3.58 (1H, ABX, JAB
Z
11.2, JBXZ4.8 Hz, CHaHbOH), 3.76 (1H, ABX, JABZ11.2,
JAXZ3.6 Hz, CHaHbOH), 3.99–4.08 (2H, AB, JZ13.6 Hz,
CH2NH), 6.90 (1H, s, Ar), 7.22–7.37 (6H, m, Ar); 13C NMR
(CDCl3, 100 MHz) d 29.8, 31.8, 34.3, 35.0, 37.4, 51.1, 59.9,
62.7, 122.3, 123.2, 123.3, 126.7, 128.8, 129.4, 136.3, 138.2,
140.9, 154.4; Anal. Calcd for C24H35NO2: C, 78.00; H,
9.55; N, 3.79. Found: C, 78.16; H, 9.71; N, 3.61%.
4.1.12. N-(20-Hydroxyphenyl)methyl-(R)-2-amino-3-
phenyl-propanol [(R)-1a]. White crystalline solid
(0.185 g, 72%). Mp 133–134 8C; [a]2D2ZC24.0 (c 1.0,
4.1.8. N-(20-Hydroxyphenyl)methyl-(S)-2-amino-(S)-3-
methyl-pentanol [(S)-1h]. Yellowish oil (0.166 g, 74%).
[a]2D2ZC54.0 (c 2.0, CHCl3); 1H NMR (CDCl3, 400 MHz);
d 0.95 (6H, m, CH3CH2 and CH3CH), 1.22 and 1.50 (2H,
2!m, CH3CH2), 1.75 (1H, m, CH3CH), 2.66 (1H, m,
CHNH), 3.63 (1H, ABX, JABZ11.2 Hz, JBXZ6.8 Hz,
CHaHbOH), 3.83 (1H, ABX, JABZ11.2 Hz, JAXZ3.6 Hz,
CHaHbOH), 3.98–4.09 (2H, AB, JZ13.6 Hz, CH2NH), 4.85
(br s, NH and OH), 6.81 (1H, apparent t, JZ7.4 Hz, Ar),
6.87 (1H, d, JZ8.0 Hz, Ar), 7.03 (1H, d, JZ7.2 Hz, Ar),
7.20 (1H, apparent t, JZ7.8 Hz, Ar); 13C NMR (CDCl3,
100 MHz) d 11.9, 14.8, 26.2, 35.2, 50.7, 61.0, 62.7, 116.5,
119.2, 123.0, 128.4, 128.9, 158.0; Anal. Calcd for
C13H21NO2: C, 69.92; H, 9.48; N, 6.27. Found: C, 69.85;
H, 9.42; N, 6.20%.
1
CHCl3); H NMR (CDCl3, 400 MHz) d 2.85 (1H, ABX,
JABZ13.6 Hz, JBXZ7.6 Hz, CHaHbPh), 2.94 (1H, ABX,
JABZ13.6 Hz, JAXZ6.4 Hz, CHaHbPh), 3.03 (1H, br,
CHNH), 3.57 (1H, ABX, JABZ11.2 Hz, JBXZ4.8 Hz,
CHaHbOH), 3.76 (1H, ABX, JABZ11.2 Hz, JAXZ3.2 Hz,
CHaHbOH), 4.04 (2H, s, CH2NH), 6.80 (1H, apparent t, JZ
7.2 Hz, Ar), 6.85 (1H, d, JZ8.0 Hz, Ar), 7.00 (1H, d, JZ
7.2 Hz, Ar), 7.19–7.38 (6H, m, Ar); 13C NMR (CDCl3,
100 MHz) d 37.1, 50.1, 59.7, 62.5, 116.6, 119.3, 122.6,
126.7, 128.4, 128.8, 129.0, 129.2, 137.9, 157.9; Anal. Calcd
for C16H19NO2: C, 74.68; H, 7.44; N, 5.44. Found: C, 74.43;
H, 7.29; N, 5.41%.
4.2. General procedure for Ti-catalyzed addition of
TMSCN to imines (small scale)
4.1.9. N-(20-Hydroxyphenyl)methyl-(R)-1-amino-propa-
nol [(R)-1i]. Light yellow crystalline solid (0.089 g, 49%).
1
Mp 87–88 8C; [a]D25.9ZK24.0 (c 1.0, CHCl3); H NMR
Ligand 1 (0.02 mmol) was dissolved in anhydrous toluene
(0.3 mL). Ti(OiPr)4 (6.0 mL, 0.02 mmol) was added to the
reaction and stirred for 10 min at ambient temperature to
give a clear yellow solution. The selected imine (0.2 mmol)
was added and then cooled to 0 8C (ice–salt bath) for
15 min. TMSCN (50 mL, 0.4 mmol) was then added using
syringe. The progress of the reaction and enantioselectivity
was monitored by 1H NMR analysis.
(CDCl3, 400 MHz); d 1.13 (3H, d, JZ6.4 Hz, CH3CH), 2.50
(1H, ABX, JABZ12.0 Hz, JBXZ8.4 Hz, CHaHbCH), 2.62
(1H, ABX, JABZ12.0 Hz, JAXZ3.2 Hz, CHaHbCH), 3.91
(1H, m, CHOH), 3.90 and 3.99 (2H, AB, JZ14.0 Hz,
CH2NH), 5.03 (br s, NH and OH), 6.71 (1H, apparent t, JZ
7.4 Hz, Ar), 6.75 (1H, d, JZ8.2 Hz, Ar), 6.91 (1H, d, JZ
7.2 Hz, Ar), 7.09 (1H, apparent t, JZ7.6 Hz, Ar); 13C NMR
(CDCl3, 100 MHz) d 21.5, 52.5, 55.6, 66.5, 116.4, 119.1,
122.4, 128.5, 128.8, 158.1; Anal. Calcd for C10H15NO2: C,
66.27; H, 8.34; N, 7.73. Found: C, 66.14; H, 8.30; N, 7.67%.
4.3. General procedure for Ti-catalyzed addition of
TMSCN to imines (large scale)
4.1.10. N-(20-Hydroxyphenyl)methyl-(S)-2-hydroxy-
methyl-pyrrolidine [(S)-1j]. Viscous yellowish oil
Ligand (1a) (25 mg, 0.1 mmol) placed in a screw-cap vial
was dissolved in anhydrous toluene (1.5 mL). Ti(OiPr)4
(29.8 mL, 0.1 mmol) was added to the reaction and stirred
for 10 min at ambient temperature to give a clear yellow
solution. 2-Propanol (76.5 mL, 1.0 mmol) was added and
left for another 10 min. The selected imine (1.0 mmol) was
then added and the reaction cooled to 0 8C. Finally, TMSCN
(250 mL, 2.0 mmol) was quickly added in one portion. After
8 h, the crude product was purified by passing through a
plug of neutral alumina eluting with ethyl acetate–hexanes
plus 0.1% Et3N to yield the corresponding a-aminonitriles.
(0.103 g, 50%). [a]D24ZK75.0 (c 1.8, CHCl3); H NMR
1
(CDCl3, 400 MHz); d 1.76–2.08 (4H, m, pyrrolidine
(CH2)2), 2.37 (1H, m, pyrrolidine N–CHaHb), 2.79 (1H,
m, pyrrolidine N–CHaHb), 3.10 (1H, m, N–CHCH2OH),
3.57 (1H, AB, JZ14.0 Hz, ArCHaHbN), 3.69 (1H, ABX,
JABZ11.2 Hz, JBXZ4.8 Hz, CHaHbOH), 3.75 (1H, ABX,
JABZ11.2 Hz, JAXZ3.8 Hz, CHaHbOH), 4.33 (1H, AB,
JZ14.0 Hz, ArCHaHbN), 6.18 (br s, OH), 6.80 (1H,
apparent t, JZ7.2 Hz, Ar), 6.84 (1H, d, JZ8.0 Hz, Ar),
7.00 (1H, d, JZ7.6 Hz, Ar), 7.18 (1H, apparent t, JZ
7.6 Hz, Ar); 13C NMR (CDCl3, 100 MHz) d 23.2, 27.5,
54.5, 58.3, 64.0, 65.5, 116.0, 119.1, 122.9, 128.1, 128.6,
157.6; Anal. Calcd for C12H17NO2: C, 69.54; H, 8.27; N,
6.76. Found: C, 69.14; H, 8.29; N, 6.71%.
4.3.1. (S)-Diphenylmethylamino-phenylacetonitrile.
White solid (0.253 g, 85%), 96% ee: [a]2D2ZK63.0 (c 1.0,
1
CHCl3) {lit.8b [a]D24 (97% ee, c 5.0, CHCl3)ZK64.2}; H
NMR (CDCl3, 400 MHz) d 2.18 (1H, d, JZ12.0 Hz,
NHCH), 4.63 (1H, d, JZ12.0 Hz, CHCN), 5.28 (1H, s,
CHPh2), 7.23–7.55 (11H, m, Ar), 7.60 (4H, m, Ar); 13C
4.1.11. N-(30,50-Di-tert-butyl-20-hydroxyphenyl)methyl-

V. Banphavichit et al. / Tetrahedron 60 (2004) 10559–10568
10567
NMR (CDCl3, 100 MHz) d 52.4, 65.6, 118.8, 127.2, 127.3,
127.5, 127.8, 128.0, 128.8, 129.0, 129.1, 129.2, 135.0,
141.2, 142.8.
127.4, 127.9, 128.0, 128.9, 129.1, 140.9, 142.4. 143.7,
147.3.
4.3.8. (R)-Diphenylmethylamino-thiophen-2-ylaceto-
nitrile. White solid (0.239 g, 83%), 98% ee: [a]2D4Z
K76.0 (c 1.0, CHCl3); 1H NMR (CDCl3, 400 MHz) d
2.22 (1H, d, JZ12.0 Hz, NHCH), 4.62 (1H, d, JZ12.0 Hz,
CHCN), 5.08 (1H, s, CHPh2), 6.84 (1H, dd, JZ5.2, 3.6 Hz,
CH-thiophene), 7.05–7.28 (8H, m, CH-thiophene and Ar),
7.30 (2H, d, JZ7.2 Hz, Ar), 7.40 (2H, d, JZ7.2 Hz, Ar);
13C NMR (CDCl3, 100 MHz) d 48.3, 65.5, 118.2, 126.2,
126.8, 127.0, 127.2, 127.4, 127.9, 128.2, 128.9, 129.2,
138.4, 140.9, 142.6.
4.3.2. (S)-Diphenylmethylamino-2-bromophenylaceto-
nitrile. White solid (0.280 g, 80%), O98% ee: [a]2D2Z
K121.0 (c1.0, CHCl3) {lit.8b [a]D24 (O99% ee, c 5.0,
CHCl3)ZK122}; 1H NMR (CDCl3, 400 MHz) d 4.92 (1H,
s, CHCN), 5.24 (1H, s, CHPh2), 7.22–7.45 (8H, m, Ar), 7.50
(2H, d, JZ7.2 Hz, Ar), 7.58–7.68 (4H, m, Ar); 13C NMR
(CDCl3, 100 MHz) d 52.6, 65.7, 118.3, 123.4, 127.2, 127.8,
127.9, 128.0, 128.3, 128.8, 128.9, 129.4, 130.8, 133.8,
134.5, 140.7, 142.6.
4.3.3. (S)-Diphenylmethylamino-2-chlorophenylaceto-
nitrile. White solid (0.256 g, 84%), 97% ee: [a]2D2Z
K118.0 (c1.0, CHCl3) {lit.8b [a]D24 (O99% ee, c 3.5,
4.3.9. (S)-Diphenylmethylamino-1-naphthylacetonitrile.
White solid (0.316 g, 91%), O98% ee: [a]2D3ZK186.0
(c1.0, CHCl3) {lit.8b [a]D22 (O99% ee, c 2.0, CHCl3)Z
K182.2}; 1H NMR (CDCl3, 400 MHz) d 2.24 (1H, br,
NHCH), 5.20 (1H, s, CHCN), 5.41 (1H, s, CHPh2), 7.21–
7.60 (11H, m, Ar), 7.66 (2H, d, JZ7.2 Hz, Ar), 7.82–7.91
(2H, m, Ar), 7.95 (2H, d, JZ7.2 Hz, Ar); 13C NMR (CDCl3,
100 MHz) d 50.5, 66.1, 118.9, 123.1, 125.3, 125.9, 126.3,
126.9, 127.0, 127.7, 128.1, 128.3, 128.7, 128.9, 129.0,
130.2, 130.4, 130.6, 134.0, 141.1, 142.6.
1
CHCl3)ZK122} H NMR (CDCl3, 400 MHz) d 4.93 (1H,
s, CHCN), 5.26 (1H, s, CHPh2), 7.22–7.42 (8H, m, Ar), 7.28
(2H, d, JZ7.2 Hz, Ar), 7.60 (4H, m, Ar); 13C NMR (CDCl3,
100 MHz) d 50.3, 65.8, 118.2, 127.2, 127.6, 127.7, 127.8,
128.0, 128.8, 128.9, 129.3, 130.4, 130.6, 132.8, 133.5,
140.7, 142.6.
4.3.4. (S)-Diphenylmethylamino-2-methylphenylaceto-
nitrile. White solid (0.245 g, 84%), 98% ee: [a]2D4Z
K161.0 (c 1.0, CHCl3); H NMR (CDCl3, 400 MHz) d
2.03 (1H, br, NHCH), 2.33 (3H, s, CH3), 4.62 (1H, s,
CHCN), 5.32 (1H, s, CHPh2), 7.22–7.36 (7H, m, Ar), 7.40
(2H, m, Ar), 7.48 (2H, d, JZ7.2 Hz, Ar), 7.60 (3H, m, Ar);
13C NMR (CDCl3, 100 MHz) d 19.0, 50.4, 65.9, 118.9,
126.8, 127.1, 127.6, 127.7, 128.0, 128.2, 128.9, 129.0,
129.3, 131.2, 133.3, 136.6, 141.1, 142.9.
4.3.10. (R)-Diphenylmethylamino-1-naphthylaceto-
nitrile [from (R)-1a ligand]. White solid (0.303 g, 87%),
O98% ee: [a]2D3ZC185.0 (c 1.0, CHCl3); 1H NMR
(CDCl3, 400 MHz) d 2.20 (1H, br, NHCH), 5.20 (1H, s,
CHCN), 5.40 (1H, s, CHPh2), 7.20–7.60 (11H, m, Ar), 7.64
(2H, d, JZ7.2 Hz, Ar), 7.82–7.90 (2H, m, Ar), 7.95 (2H, d,
JZ7.2 Hz, Ar); 13C NMR (CDCl3, 100 MHz) d 50.5, 66.1,
118.9, 123.2, 125.3, 125.9, 126.3, 126.9, 127.0, 127.7,
128.1, 128.3, 128.8, 128.9, 129.0, 130.2, 130.4, 130.7,
134.0, 141.1, 142.7.
1
4.3.5. (S)-Diphenylmethylamino-2-methoxyphenyl-
acetonitrile. White solid (0.253 g, 84%), 83% ee:
[a]2D4ZK57.0 (c 1.0, CHCl3); 1H NMR (CDCl3,
400 MHz) d 3.92 (3H, s, OCH3), 4.71 (1H, s, CHCN),
5.23 (1H, s, CHPh2), 7.03 (2H, m, Ar), 7.28–7.43 (8H, m,
Ar), 7.48 (2H, d, JZ7.6 Hz, Ar), 7.56 (2H, d, JZ7.6 Hz,
Ar); 13C NMR (CDCl3, 100 MHz) d 48.6, 55.7, 65.4, 111.5,
119.2, 121.1, 123.5, 127.4, 127.6, 127.7, 127.8, 128.8,
128.9, 129.0, 130.7, 141.6, 143.0, 157.1.
Acknowledgements
This work was partly supported by Chulalongkorn
University’s Radchadapisek Endowment Fund (T.V. and
W.B.), the Ministry of University Affairs (W.M.), the Royal
Golden Jubilee PhD Program (V.B.) and the Thailand Toray
Science Foundation (T.V.).
4.3.6. (S)-Diphenylmethylamino-4-methoxyphenyl-
acetonitrile. White solid (0.256 g, 85%), 91% ee;
[a]2D4ZK38.0 (c 1.0, CHCl3) {lit.8b [a]2D2 (94% ee, c 0.54,
CHCl3)ZK27.7}; 1H NMR (CDCl3, 400 MHz) d 1.94 (1H,
d, JZ12.0 Hz, NHCH), 3.66 (3H, s, OCH3), 4.39 (1H, d,
JZ12.0 Hz, CHCN), 5.06 (1H, s, CHPh2), 6.78 (2H, d, JZ
8.8 Hz, Ar) 7.02–7.22 (6H, m, Ar), 7.29 (4H, m, Ar), 7.40
(2H, d, JZ7.6 Hz, Ar); 13C NMR (CDCl3, 100 MHz) d
51.9, 55.4, 65.6, 114.4, 119.1, 127.1, 127.2, 127.5, 127.7,
128.0, 128.6, 128.8, 129.1, 141.3, 142.9, 160.1.
References and notes
1. Harada, K. Nature 1963, 200, 1201.
2. Review: Duthaler, R. O. Tetrahedron 1994, 50, 1539–1650.
3. Kobayashi, S. Chem. Rev. 1999, 99, 1069–1094. Reviews: (a) .
(b) Yet, L. Angew. Chem., Int. Ed. 2001, 40, 875–877. (c)
Groger, H. Chem. Rev. 2003, 103, 2795–2828.
4.3.7. (R)-Diphenylmethylamino-furan-2-ylacetonitrile.
White solid (0.214 g, 82%), 91% ee: [a]2D1ZK25.0 (c 1.0,
CHCl3); 1H NMR (CDCl3, 400 MHz) d 2.37 (1H, br,
NHCH), 4.65 (1H, s, CHCN), 5.20 (1H, s, CHPh2), 6.42
(1H, dd, JZ3.2, 2.0 Hz, CH-furan), 6.49 (1H, d, JZ3.2 Hz,
CH-furan), 7.24–7.42 (6H, m, Ar), 7.48 (3H, d, JZ7.6 Hz,
CH-furan and Ar), 7.55 (2H, d, JZ7.2 Hz, Ar); 13C NMR
(CDCl3, 100 MHz) d 46.2, 65.2, 109.0, 110.7, 117.1, 127.3,
4. Iyer, M. S.; Gigstad, K. M.; Namdev, N. D.; Lipton, M. J. Am.
Chem. Soc. 1996, 118, 4910–4911.
5. Corey, E. J.; Grogan, M. J. Org. Lett. 1999, 1, 157–160.
6. (a) Vachal, P.; Jacobsen, E. N. J. Am. Chem. Soc. 2002, 124,
10012–10014. (b) Sigman, M. S.; Vachal, P.; Jacobsen,
E. N. Angew. Chem., Int. Ed. 2000, 39, 1279–1281. (c)
Sigman, M. S.; Jacobsen, E. N. J. Am. Chem. Soc. 1998, 120,
4901–4902.

10568
V. Banphavichit et al. / Tetrahedron 60 (2004) 10559–10568
7. (a) Takamura, M.; Hamashima, Y.; Usada, H.; Kanai, M.;
Shibasaki, M. Angew. Chem., Int. Ed. 2000, 39, 1650–1652.
(b) Mori, M.; Imma, H.; Nakai, T. Tetrahedron Lett. 1997, 38,
6229–6232. (c) Ishitani, H.; Komiyama, S.; Kobayashi, S.
Angew. Chem., Int. Ed. 1998, 37, 3186–3188. (d) Chavarot,
M.; Byrne, J. J.; Chavant, P. Y.; Vallee, Y. Tetrahedron:
Asymmetry 2001, 12, 1147–1150.
Soc., Chem. Commun. 1991, 1752–1753. (b) Hayashi, M.;
Miyamoto, Y.; Inoue, T.; Oguni, N. J. Org. Chem. 1993, 58,
1515–1522. (c) Hayashi, M.; Inoue, T.; Miyamoto, Y.; Oguni,
N. Tetrahedron 1994, 50, 4385–4398.
13. Mansawat, W.; Bhanthumnavin, W.; Vilaivan, T. Tetrahedron
Lett. 2003, 44, 3805–3808.
14. The structurally related chiral Ti-Schiff bases (2) (RZPhCH2,
iPr) complexes did catalyze the same reaction to provide the
Strecker product in good yield. However, the enantio-
selectivity was modest (!10% ee).
8. (a) Sigman, M. S.; Jacobsen, E. N. J. Am. Chem. Soc. 1998,
120, 5315–5316. (b) Krueger, C. A.; Kuntz, K. W.; Dzierba,
C. D.; Wirschun, W. G.; Gleason, J. D.; Snapper, M. L.;
Hoveyda, A. H. J. Am. Chem. Soc. 1999, 121, 4284–4285.
(c) Porter, J. R.; Wirschun, W. G.; Kuntz, K. W.; Snapper,
M. L.; Hoveyda, A. H. J. Am. Chem. Soc. 2000, 122,
2657–2658. (d) Josephsohn, N. S.; Kuntz, K. W.; Snapper,
M. L.; Hoveyda, A. H. J. Am. Chem. Soc. 2001, 123,
11594–11599.
15. Hassan, N. A.; Bayer, E.; Jochims, J. C. J. Chem. Soc., Perkin
Trans. 1 1998, 3747–3757.
16. Narasimhan, S.; Balakumar, R. Synth. Commun. 2000, 30,
4387–4396.
17. Suwa, T.; Sugiyama, E.; Shibata, I.; Baba, A. Synthesis 2000,
789–800.
9. Keith, J. M.; Jacobsen, E. N. Org. Lett. 2004, 6, 153–155.
10. Masumoto, S.; Usuda, H.; Suzuki, M.; Kanai, M.; Shibasaki,
M. J. Am. Chem. Soc. 2003, 125, 5634–5635.
18. The presence of both hydroxyl groups in the ligand 1 is
necessary for high activity, since analogues of ligand (1)
lacking either of the hydroxyl groups did not provide the same
level of efficiency both in terms of yield and enantioselectivity.
19. Reactions performed on a larger scale or at lower loading of
catalyst do require a slow addition of 2-propanol to ensure
good enantioselectivity.
11. (a) Narasimhan, S.; Velmathi, S.; Balakumar, R.;
Radhakrishnan, V. Tetrahedron Lett. 2001, 42, 719–721.
(b) Velmathi, S.; Swarnalakshmi, S.; Narasimhan, S.
Tetrahedron: Asymmetry 2003, 14, 113–117.
12. (a) Hayashi, M.; Miyamoto, Y.; Inoue, T.; Oguni, N. J. Chem.