Asymmetric cyanation of aldehydes, ketones, aldimines, and ketimines catalyzed by a versatile catalyst generated from cinchona alkaloid, achiral substituted 2,2′-biphenol and tetraisopropyl titanate

Article abstract of DOI:10.1002/chem.200900936

Full investigation of cyanation of aldehydes, ketones, aldimines and ketimines with trimethylsilyl cyanide (TMSCN) or ethyl cyanoformate (CNCOOEt) as the cyanide source has been accomplished by employing an in situ generated catalyst from cinchona alkaloid, tetraisopropyl titanate [Ti(OiPr)₄] and an achiral modified biphenol. With TMSCN as the cyanide source, good to excellent results have been achieved for the Strecker reaction of N-Ts (Ts=p-toluenesulfonyl) aldimines and ketimines (up to >99% yield and >99% ee) as well as for the cyanation of ketones (up to 99% yield and 98% ee). By using CNCOOEt as the alternative cyanide source, cyanation of aldehyde was accomplished and various enantioenriched cyanohydrin carbonates were prepared in up to 99% yield and 96% ee. Noteworthy, CNCOOEt was successfully employed for the first time in the asymmetric Strecker reaction of aldimines and ketimines, affording various a-amino nitriles with excellent yields and ee values (up to >99% yield and >99% ee). The merits of current protocol involved facile availability of ligand components, operational simplicity and mild reaction conditions, which made it convenient to prepare synthetically important chiral cyanohydrins and examino nitriles. Furthermore, control ex-periments and NMR analyses were performed to shed light on the catalyst structure. It is indicated that all the hydroxyl groups in cinchona alkaloid and biphenol complex with Ti^IV, forming the catalyst with the structure of (biphenoxide) Ti(OR*)(Oi'Pr). The absolute configuration adopted by biphenol 4 m in the catalyst was identified as S configuration according to the evidence from control experiments and NMR analyses. Moreover, the roles of the protonic additive ((iPrOH) and the tertiary amine in the cinchona alkaloid were studied in detail, and the real cyanide reagent in the catalytic cycle was found to be hydrogen cyanide (HCN). Finally, two plausible catalytic cycles were proposed to elucidate the reaction mechanisms.

Full text of DOI:10.1002/chem.200900936

DOI: 10.1002/chem.200900936
Asymmetric Cyanation of Aldehydes, Ketones, Aldimines, and Ketimines
Catalyzed by a Versatile Catalyst Generated from Cinchona Alkaloid,
Achiral Substituted 2,2’-Biphenol and Tetraisopropyl Titanate
Jun Wang, Wentao Wang, Wei Li, Xiaolei Hu, Ke Shen, Cheng Tan, Xiaohua Liu, and
Xiaoming Feng*[a]
Abstract: Full investigation of cyana-
tion of aldehydes, ketones, aldimines
and ketimines with trimethylsilyl cya-
nide (TMSCN) or ethyl cyanoformate
(CNCOOEt) as the cyanide source has
been accomplished by employing an in
situ generated catalyst from cinchona
alkaloid, tetraisopropyl titanate [Ti-
carbonates were prepared in up to
99% yield and 96% ee. Noteworthy,
CNCOOEt was successfully employed
for the first time in the asymmetric
Strecker reaction of aldimines and ket-
periments and NMR analyses were per-
formed to shed light on the catalyst
structure. It is indicated that all the hy-
droxyl groups in cinchona alkaloid and
biphenol complex with TiIV, forming
the catalyst with the structure of
AHCTUNGTRENNiGUN mines, affording various a-amino ni-
triles with excellent yields and ee
values (up to >99% yield and >99%
ee). The merits of current protocol in-
volved facile availability of ligand com-
ponents, operational simplicity and
mild reaction conditions, which made it
convenient to prepare synthetically im-
portant chiral cyanohydrins and a-
amino nitriles. Furthermore, control ex-
(biphenoxide)TiACTHNGUTRENNU(G OR*)ACHTUNGTRENUN(NG OiPr). The ab-
solute configuration adopted by biphe-
nol 4m in the catalyst was identified as
S configuration according to the evi-
dence from control experiments and
NMR analyses. Moreover, the roles of
the protonic additive (iPrOH) and the
tertiary amine in the cinchona alkaloid
were studied in detail, and the real cya-
nide reagent in the catalytic cycle was
found to be hydrogen cyanide (HCN).
Finally, two plausible catalytic cycles
were proposed to elucidate the reac-
tion mechanisms.
ACHTUNGTRENNUNG(OiPr)4] and an achiral modified biphe-
nol. With TMSCN as the cyanide
source, good to excellent results have
been achieved for the Strecker reaction
of N-Ts (Ts=p-toluenesulfonyl) ald-
ACHTUNGTRENNUNGimines and ketimines (up to >99%
yield and >99% ee) as well as for the
cyanation of ketones (up to 99% yield
and 98% ee). By using CNCOOEt as
the alternative cyanide source, cyana-
tion of aldehyde was accomplished and
various enantioenriched cyanohydrin
Keywords: asymmetric catalysis
cinchona alkaloid cyanation
cyanohydrins · Strecker reaction
·
·
·
Introduction
ble attention of organic chemists due to the rich chemistry
of cyano group (CN).[3] For example, chiral cyanohydrins
could be elaborated into a number of key intermediates, in-
cluding a-hydroxy acids, a-hydroxy ketones, a-hydroxy alde-
hydes, b-hydroxy amines, and so on.[4–8] Whereas, chiral a-
amino nitriles could be conveniently converted to a variety
of useful chiral a-amino acids and 1,3- diamines.
Background: Since the first synthesis of cyanohydrin by
Winkler in 1832[1] and the first synthesis of a-amino nitrile
using a three-component condensation method by Strecker
in 1850,[2] a-functionalized nitriles have attracted considera-
Before 1986, the efficient catalysts developed for the
preparation of optically active cyanohydrins were merely re-
stricted to enzymes[4a–c,e] (R- or S-oxynitrilase) and cyclic di-
peptides.[4a,c,e] Over the past two decades, great advances
have been achieved in the chemically catalytic asymmetric
cyanation of carbonyl compounds, which provided conven-
ient ways to access various enantioenriched cyanohy-
drins.[4d,f–l,5–7] For the cyanation of aldehydes, the procedures
developed by Belokon’ and North,[4h] Shibasaki,[4i] Choi,[5i]
[a] Dr. J. Wang, W. Wang, W. Li, X. Hu, K. Shen, Dr. C. Tan, Dr. X. Liu,
Prof. Dr. X. Feng
Key Laboratory of Green Chemistry & Technology
Ministry of Education, College of Chemistry
Sichuan University, Chengdu 610064 (P.R. China)
Fax : (+86)28-8541-8249
E-mail: xmfeng@scu.edu.cn
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200900936.
11642
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11642 – 11659

FULL PAPER
Kagan,[5j] Uang,[5l] Saꢁ,[5m] Feng,[5n] Corey,[5o] Pu,[5p] Ishiha-
ra,[5q] and Ohkuma[5v] are the pinnacle of what has been
achieved. However, cyanation of ketones was more chal-
(BINOL) and their derivatives are well known for their ex-
cellent performance in the catalytic asymmetric synthesis.
However, the combined use of these two ligands within one
catalyst system was not attempted until recently. At the very
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
lenging due to the greater steric congestion and lower elec-
trophilicity of ketones. Recently, breakthroughs have been
accomplished in this longstanding problem.[8] The impressive
catalysts comprised the titanium catalysts by Shibasaki[8b,c]
and Feng,[8i] the lanthanide systems by Shibasaki,[8d,g,h,p] the
aluminium catalysts by Snapper and Hoveyda,[8f] as well as
Feng,[8l] the nucleophilic catalyst by Deng,[8e,j,k] the boron
catalyst by Corey,[8q] the thiourea catalyst by Jacobsen,[8r,x]
the amino acid salt catalyst by Feng,[8s] and so on.
beginning, we found that cinchonine/[TiACTHNUTRGENUG(N OiPr)4] could cata-
lyze the cyanation of aldehydes with CNCOOEt and N-Ts
aldimines with TMSCN, although very low ee values were
obtained. To our delight, by combining (R)-BINOL as a
second ligand, the results were significantly improved, sug-
gesting that an improved chiral environment was created
around the catalytic center. Based on these findings, two
catalyst systems have been evolved. Firstly, a four-compo-
nent catalyst system consisted of cinchonine, (S)-6,6’-dibro-
mo-1,1’-bi-2-naphthol, (1R,2S)-(À)-N-methylephedrine, and
As to the asymmetric cyanation of imines (Strecker reac-
tion), the first catalytic version was reported by Liptonꢂs
group in 1996,[10a] after which many efficient catalysts were
successively developed.[9–11] For the cyanation of ald-
[TiACTHUNTRGENNG(U OiPr)4] was discovered to be highly efficient for the
asymmetric cyanocarbonylation of aldehydes.[6i] Secondly, a
relatively simple catalyst generated from cinchonine, achiral
ACHTUNGTRENNUNG
imines,[10] efficient chiral metal complexes included AlIII,[10c,h]
TiIV,[10f,k,n,11j] and ZrIV.[10d] Chiral metal-free catalysts involved
guanidine,[10e] urea and thiourea,[10b,g,t,u,y] N,N’-dioxide,[10i,j,aa]
chiral ammonium salt,[10o] chiral quaternary ammonium
salt,[10p,w] Brønsted acid,[10q] and bisformamide.[10v] For the
similar difficulties encountered in the cyanation of ketones,
Strecker reaction of ketimines was not as fruitful as that of
aldimines. Highly efficient catalysts were rare and could be
counted on oneꢂs fingers: Jacocsenꢂs urea Schiff base,[11a,c]
Shibasakiꢂs glucose derived ligand-Gd complex,[11d–g] Fengꢂs
N,N’-dioxide,[11h,i,k] and Fengꢂs cinchonine/Ti/biphenol com-
plex.[11j]
3,3’-disubstituted biphenol, and [TiACTHNGUTRENU(NG OiPr)4] was developed
for the asymmetric Strecker reaction of N-Ts imines, afford-
ing the desired products in high yields with excellent ee val-
ues.[11j] Especially, a phenomenon of “asymmetric activa-
tion”[12] was observed in the latter case. The inducing ability
of the chiral ligand was significantly magnified by the axially
flexible achiral ligand through coordinative interaction with
the central metal TiIV.
Herein, a full investigation of cinchona alkaloid/[Ti-
AHCTUNGTERG(NNNU OiPr)4]/achiral 3,3’-disubstituted 2,2’-biphenol in the asym-
metric cyanation of aldehydes, ketones, aldimines and keti-
mines was presented.[13] Besides trimethylsilyl cyanide
(TMSCN), an alternative cyanide source, ethyl cyanofor-
mate (CNCOOEt), was also studied in full detail. Excellent
results have been achieved in the cyanation of aldehydes, al-
dimines and ketimines. Noteworthy, to the best of our
knowledge, CNCOOEt has never been successfully em-
ployed in the Strecker reaction before. The catalyst struc-
ture and reaction mechanism were studied in detail by con-
General ligand/catalyst for the catalytic asymmetric cyana-
tion of C=O and C=N bonds: Exploiting ligand/catalyst with
wide applications was challenging and appealing. In the
field of asymmetric cyanation of C=X (X=O, N), such li-
gands/catalysts were listed as follows. The glucose-derived
ligand developed by Shibasaki was efficient for the cyana-
tion of aldehydes, ketones and ketimines by using its AlIII,
TiIV, SmIII or GdIII complex.[4d,i] The AlIII or TiIV complex of
the tripeptide Schiff base ligand developed by Snapper and
Hoveyda could efficiently promote the asymmetric cyana-
tion of ketones[8f] and aldimines.[10f] The AlIII or TiIV com-
plex of C2 symmetric Schiff base was successfully used in the
cyanation of aldehydes and ketones by Belokon’ and
North,[4h] in the cyanation of ketones by our group,[8i,l] and
in the Strecker reaction of aldimines by Jacobsen.[10c] The
BINOL-based bifunctional AlIII catalyst was suitable for the
cyanation of aldehydes, aldimines, and Reissert reaction
(analogous Strecker reaction).[4d,i] The urea and thiourea cat-
alyst developed by Jacobsen was suitable for the asymmetric
cyanation of aldimines,[10g,t,u,y] ketimines,[11a,c] ketones,[8r,x] and
aldehydes[8r] (two examples were reported for aldehydes). In
this article, a general catalyst system accommodated alde-
hydes, ketones, aldimines and ketimines for the asymmetric
cyanation is described.
1
trol experiment, H NMR and 13C NMR analyses. Plausible
catalytic cycles were proposed to explain the reaction
course.
Results and Discussion
Catalytic asymmetric cyanation of aldimines, ketimines,
ketones and aldehydes with TMSCN as the cyanide source
Catalytic asymmetric cyanation of aldimines with TMSCN
as the cyanide source: In our initial studies on the cyanation
of N-Ts benzaldimine 14a with TMSCN, it was found that
while 1a, 1a/[Ti
lytic activity or poor enantioselectivity, the combinatorial
catalyst, cinchonine 1a/[Ti(OiPr)4]/(R)-BINOL 3, provided
ACHTUNGTNERNUG(OiPr)4] or (R)-3/[TiACHTUNGTREN(NUNG OiPr)4] gave low cata-
AHCTUNGTRENNUNG
moderate enantioselectivity (66% ee) with quantitative
yield in toluene at À208C (Table 1, entries 1–4). Surprising-
ly, replacing (R)-BINOL with (S)-BINOL in the catalyst
system led to comparable ee value with the same stereo-
chemistry control, suggesting that the absolute configuration
Development of cinchona alkaloid/Ti/diol catalyst system
for the cyanation reaction and the phenomenon of asymmet-
ric activation: Cinchona alkaloids, 1,1’-bi-2-naphthol
Chem. Eur. J. 2009, 15, 11642 – 11659
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11643

X. Feng et al.
Table 1. Searching for an efficient combinatorial catalyst system.[a]
halogen or alkyl substituents, biphenols 4l–p with planar
and electron-rich aryl group at the 3,3’-position were synthe-
sized. Excitingly, dramatic improvement of enantioselectivi-
ty was generally observed by using these bipols except 4l,
furnishing the desired product 15a with 94–95% ee (Table 1,
entries 17–21). It was suggested that the aryl modified bipols
were suitable to create a more favorable asymmetric cavity
around the metal center.
Entry
Catalyst[b]
ee [%][c]
Entry
Catalyst[b]
ee [%][c]
1[d]
2
1a
1a/Ti
À16
48
12
66
62
71
62
64
4
34
74
70
72
68
71
70
74
95
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
1a/Ti/4n
1a/Ti/4o
1a/Ti/4p
1a/Ti/5a
1a/Ti/5b
1a/Ti/5c
1a/Ti/6
1a/Ti/7
1a/Ti/8
1a/Ti/9
1a/Ti/10
1a/Ti/11
1a/Ti/12
1a/Ti/13
1b/Ti/4m
2a/Ti/4m
2b/Ti/4m
94
95
94
3[d]
4
(R)-3/Ti
1a/Ti/(R)-3
1a/Ti/(S)-3
1a/Ti/4a
1a/Ti/4b
1a/Ti/4c
1a/Ti/4d
1a/Ti/4e
1a/Ti/4 f
1a/Ti/4g
1a/Ti/4h
1a/Ti/4i
1a/Ti/4j
1a/Ti/4k
1a/Ti/4l
1a/Ti/4m
65
5
6
7
8
À11
6
58
55
9
72 (69)[e]
40
10
11
12
13
14
15
16
17
18
36
24
46
12
93
À90
À79
[a] Unless noted otherwise, reactions were performed with benzaldimine
14a (0.1 mmol), TMSCN (0.2 mmol, but 0.12 mmol was used for en-
tries 17–21), catalyst (20 mol%) in toluene (0.5 mL) at À208C for 18 h,
and full conversion of benzaldimine was observed by TLC detection.
[b] The Ti catalyst was L1(cinchona alkaloid)/[Ti
1:1:1 or L1 (L2)/[Ti(OiPr)4] 1:1. When L2=monophenols 6–10 (en-
tries 31–35), L1/[Ti(OiPr)4]/L2 1:1:2. [c] The negative sign means that the
major enantiomer was R configuration. [d] Large amount of the unreact-
ed imine and trace of the product were obtained. [e] L1/[Ti(OiPr)4]/L2
1:1:1.
ACHUTGTNRNE(UNG OiPr)4]/L2ACHTUNGTRNE(NUGN phenol)
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
of the product 15a was mainly governed by cinchonine
rather than chiral BINOL (Table 1, entries 4,5). So, we won-
dered whether the chiral BINOL 3 could be replaced by the
achiral 2,2’-biphenol (bipol) 4a, which was more advanta-
geous for it could be used without asymmetric synthesis or
resolution. The feasibility of this assumption was quickly
confirmed and even slightly higher ee value (71% ee) was
obtained (Table 1, entry 6).
It was anticipated that with the chiral ligand unchanged,
improved enantioselectivity might be achieved by reasona-
ble modification of achiral 2,2’-biphenol. Therefore, an array
of modified bipols 4b–k (Figure 1) with varying electronic
properties and steric hindrances were synthesized and evalu-
ated (Table 1, entries 7–16). Negative effect was observed
when electron-withdrawing group, such as Br, was intro-
duced onto the bipol (Table 1, entries 7,8). Substituents at
ortho position of the hydroxyl groups greatly affected the
outcomes of the enantioselectivity. When R4 was kept con-
stant as tBu, the enantiomeric excess increased along with
the increment of the steric hindrance of R3 (Table 1, en-
tries 9–11). However, the much bulkier 1-adamantyl led to a
slight decrease of ee (Table 1, entry 12). If R3 was kept con-
stant as tBu or 1-adamantyl, variation of substituents at the
5,5’-position of bipol had little influence on enantioselectivi-
ty (Table 1, entries 11–16). Given no significant improve-
ment of ee was achieved by modifying 2,2’-biphenol with
Figure 1. Ligands evaluated in this study.
To find out whether the axis presented in bipol was essen-
tial to achieve good results, we examined some other diols.
As could be seen, inferior results were observed when bi-
phenol 4 was replaced by the structurally similar diols 2,2’-
methylene-bis-(phenol) (5) with a methylene linking two
phenol fragments (Table 1, entries 22–24). Phenols 6–10
which were monomers of the corresponding biphenols 3 and
4 gave poor results either (Table 1, entries 25–29). For in-
stance, when two equiv of 2-phenyl phenol 8 was used in-
stead of 3,3’-diphenyl 2,2’-biphenol (4n), low enantioselec-
tivity was obtained. Moreover, pyrocatechol 11 and hydro-
quinone 12–13 with two hydroxyls attached on one benzene
ring also showed bad results compared with 2,2’-biphenol
(4a; Table 1, entries 30–32). As has been demonstrated, the
axis in the biphenol was very crucial and indispensable to
achieve excellent enantioselectivity.
Other cinchona alkaloids were tried to replace cinchonine,
but none of them exhibited superior result (Table 1, en-
tries 33–35). Noteworthy, when cinchonidine 2a, the diaste-
reomer of 1a, was used, the product was obtained in 90%
ee with inversed configuration, which made it facile to
11644
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11642 – 11659

Asymmetric Cyanation
FULL PAPER
obtain the R enantiomer of a-aminonitrile 15a (Table 1,
entry 34).
mation). It was found that phenol and some other alcohols
such as MeOH gave comparable results with iPrOH. Reac-
tion parameters such as cyanide source,[15] central metal, sol-
vent, and reaction temperature were also investigated, but
no superior result was obtained. The optimal conditions
Interestingly, when the catalyst loading was gradually de-
creased from 30 to 5 mol%, not only the reaction became
sluggish but the enantioselectivity decreased steadily from
97 to 84% ee (Table 2, entries 1–5). Inspired by the previous
reports using protic additives to improve results,[14] iPrOH
(1.2 equiv) was added to the reaction mixture of catalyst
(5 mol%), aldimine 14a, and TMSCN in toluene at À208C.
To our delight, the reaction proceeded smoothly, giving the
product 15a in high yield with high ee (Table 2, entry 6).
Triggered by this, we reasoned that under high catalyst load-
were identified as catalyst (1a/[TiACTHNUTRGNEN(UG OiPr)4]/4m 1:1.2:1.2,
5 mol%), TMSCN (1.2 equiv), iPrOH (1.2 equiv), 0.2m in
toluene at À208C, under which, a series of N-Ts aldimines
were examined and generally high yields and ee values were
obtained (Table 3).
Table 3. Substrate scope for the cyanation of N-Ts aldimines with
TMSCN.[a]
Table 2. Investigation of the catalyst loading with TMSCN as the cyanide
source.[a]
Entry
R
Product
t [h]
Yield [%][b]
ee [%][c]
1
2
3
4
5
6
7
Ph
15a
15b
15c
15d
15e
15 f
15g
2.5
3
3
22
8
18
8
>99
>99
97
>99
>99
>99
>99
97(S)
96
93
97
94
Entry Catalyst [mol%] iPrOH [mol%] t [h] Yield[b] [%] ee [%]
4-FC6H4
4-ClC6H4
4-MeC6H4
3-MeC6H4
2-MeC6H4
4-MeOC6H4
1
2
3
4
30
20
15
10
5
none
none
none
none
none
120
none
60
120
4
4
24
24
54
99
99
99
99
98
97
95
92
90
84
97
46
94
96
95
79
97
5
6[c]
7[d]
8[d]
9[d]
10[d]
5
2.5 99
8
15h
3
>99
97
20
20
20
20
4
4
4
4
35
99
99
99
9
1-naphthyl
2-naphthyl
2-furyl
15i
15j
15k
15l
4
22
3
92
97
93
94
91
96
90
94
10
11
12
180
[a] Unless noted otherwise, reactions were performed with imine
(0.1 mmol), TMSCN (0.12 mmol) with catalyst (1a/[Ti(OiPr)4]/4m 1:1:1,
5–30 mol%) in toluene (0.5 mL) at À208C. [b] Isolated yield. [c] 1a/[Ti-
ACHTUNGTRENNUNG(OiPr)4]/4m 1:1.2:1.2. [d] The produced iPrOH was removed under
2-thienyl
5
AHCTUNGTRENNUNG
13
15m
4
96
91
14[d]
15[d]
16[d]
17[d]
18[e]
19[f]
20[g]
cyclohexyl
iPr
tBu
nC5H11
Ph
Ph
15n
15o
15p
15q
15r
15s
15t
6
6
6
6
4
4
4
>99
96
98
61
>99
94
92
84
84
79
94
94
90
vacuum after catalyst preparation.
ings (20 or 30 mol%), the large amount of iPrOH (ca. 60 or
90 mol%) generated from the catalyst preparation must be
responsible for the excellent results obtained without addi-
tion of iPrOH (Table 2, entries 1,2). To verify this assump-
tion, some control experiments were performed. As expect-
ed, when the released iPrOH was removed after the catalyst
Ph
75
[a] Unless otherwise noted, ArSO2 =p-tosyl (Ts) and reactions were car-
ried out with imine (0.1 mmol), TMSCN (0.12 mmol), catalyst (5 mol%),
iPrOH (0.12 mmol) in toluene (0.5 mL) at À208C. [b] Isolated yield.
[c] Determined by HPLC. [d] 15 mol% of catalyst was used. [e] Ar=4-
ClC6H4. [f] Ar=Ph. [g] Ar=2,4,6-trimethylphenyl.
preparation (1a/[TiACHTUNGTRENNUNG(OiPr)4]/4m 1:1:1, 20 mol%), both the
yield and ee were dramatically diminished (Table 2, entry 7).
Remarkably, the excellent result could be regained by
adding 60 mol% of iPrOH back to the reaction system
(Table 2, entry 8). In addition, presence of excessive iPrOH
(120–180 mol%) had no adverse influence on the result
(Table 2, entries 9,10). Therefore, without the addition of
iPrOH, the inferior results obtained under low catalyst load-
ings would not be ascribed to the reduced amount of active
catalytic species. The real reason might lie in the fact that
the quantity of iPrOH produced in the catalyst preparation
step was far from enough to give excellent result (for the de-
tailed studies on the role of iPrOH, see Section on the In-
vestigation of the role of iPrOH).
Catalytic asymmetric cyanation of ketimines with TMSCN
as the cyanide source: Although 1a/[TiACTHNUTRGEN(UNG OiPr)4]/4m 1:1.2:1.2
was optimized for the reaction of N-Ts aldimines, it could
be efficiently applied to the cyanation of N-Ts ketimines. As
the results shown in Table 4, excellent enantioselectivities
(up to 99% ee) and high yields were attained for a wide
spectrum of ketimines using 5 mol% of catalyst. In some
cases (e.g., substrates bearing electron-donating group such
as methyl or methoxyl, substrates with bulky hindrance, and
aliphatic ketimines), 10 mol% of catalyst was required to
furnish satisfactory reactivity and enantioselectivity.
Most of the substituted acetophenone derived imines
were converted to the products in high yields with excellent
enantioselectivities (Table 4, entries 1–9). Cyclic ketimine
was also tested, giving the product in high yield with good
To further optimize the reaction conditions, the amount
of iPrOH and some other protonic reagents were examined
with 5 mol% of catalyst (for details, see Supporting Infor-
Chem. Eur. J. 2009, 15, 11642 – 11659
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11645

X. Feng et al.
ee with 10 mol% of catalyst (Table 4, entry 10). Heteroaro-
matic ketimines showed high reactivities and enantioselec-
tivities (Table 4, entries 11,12). Both the aryl ethyl and aryl
propyl ketimines gave excellent results, but aryl cyclohexyl
ketone derived imine was unreactive (Table 4, entries 13–
15). Besides, a,b-unsaturated ketimine 16p gave the synthet-
ically important product with high ee and complete regiose-
lectivity (Table 4, entry 16). As to the chalcone derived
imine 16q, the desired 1,2-adduct was obtained in high yield
with moderate ee (Table 4, entry 17). Among the aliphatic
ketimines evaluated, cyclohexyl methyl ketimine afforded
the best result, whereas primary alkyl methyl ketimine 16r
and tertiary alkyl methyl ketimine 16t provided less satisfac-
tory results (Table 4, entries 18–20).
to apply the current catalyst to the asymmetric synthesis of
tertiary a-amino nitriles with two aryl groups on the chiral
center. As shown in Table 5, the reactivity and enantioselec-
tivity drastically depended on the catalyst loading and the
substitution pattern of the aryl group. While low reactivity
and moderate enantioselectivity (26% yield and 76% ee)
were observed for the cyanation of 2-fluorobenzophenone
derived ketimine 18a in the presence of 5 mol% of catalyst,
excellent result (96% yield and 98% ee) could be obtained
by increasing the catalyst loading to 10 mol% (Table 5, en-
tries 1,2). Similar to the trend observed in the asymmetric
reduction of diarylkeACTHNUTRGNEUNG
tones,[16b–d,g] a significant ortho-substitu-
ent effect was observed, namely, ortho-substituted diarylke-
timines were the most suitable substrates and could be
highly enantioselectively converted to the desired products
(Table 5, entries 2–5). It should be mentioned that in the
Strecker reaction of aldimines and aryl alkyl ketimines, the
presence of ortho-substituent in substrate was detrimental to
the enantioselectivities. Surprisingly, no product was detect-
ed when strongly electron-withdrawing group CF3 was pres-
ent (Table 5, entry 6). meta- and para-Substituted benzophe-
none imines gave the desired products in high yields but
poor enantioselectivities (Table 5, entries 7–12). Presumably,
the presence of an ortho-sub-
Chiral diaryl compounds such as diarylmethanols and di-
ACHTUNGTRENNUNGarylmethylamines are important intermediates for the syn-
thesis of pharmaceutically relevant products with antihista-
minic, antiarrhythmic, diuretic, antidepressive, laxative,
local-anesthetic and anticholinergic properties.[16a,b] Al-
though a large number of efficient methods have been es-
tablished for the asymmetric synthesis of sec-diaryl com-
pounds,[16c–j] the reports on the asymmetric synthesis of tert-
diaryl compounds are quite limited.[16k,l] So, it is interesting
stituent in the substrate greatly
benefited the enantiofacial dis-
Table 4. Substrate scope for the cyanation of N-Ts ketimines with TMSCN.[a]
crimination for the cyanide
attack in the transition state.
Catalytic asymmetric cyanosi-
Entry
Ketimine
R1 =Ph, R2 =Me
Product
Catalyst [mol%]
t [h]
Yield [%][b]
ee [%][c]
lylation of ketones and alde-
hydes: Followed the studies of
asymmetric cyanation of ald-
1
2
3
4
5
6
7
8
9
17a
17b
17c
17d
17e
17 f
17g
17h
17i
5
5
5
5
4
4
4
8
>99
90
>99
>99
>99 (91)
99 (95)
>99
>99 (>99)
90
>99(S)
98
R1 =4-FPh, R2 =Me
R1 =4-ClPh, R2 =Me
R1 =4-BrPh, R2 =Me
R1 =4-MePh, R2 =Me
R1 =4-MeOPh, R2 =Me
R1 =3-ClPh, R2 =Me
R1 =2-FPh, R2 =Me
R1 =2-naphthyl, R2 =Me
>99(S)
>99(S)
>99 (94)(S)
>99 (94)(S)
>99
AHCTUNGTERGiNNUN mines and ketimines, 1a/Ti/
4m was attempted in the cya-
nosilylation of ketones. Un-
fortunately, the reaction pro-
ceeded sluggishly and the de-
sired cyanohydrin silyl ether
21a was obtained in 86% yield
and 38% ee after four days
(Table 6, entry 1). To improve
the result, some modified bi-
phenols were screened (see
Supporting Information). Sur-
prisingly, 3,3’-di-a-naphthyl bi-
phenol (4l) provided the best
enantioselectivity (74% ee),
while it gave the poorest result
among the 3,3’-diaryl substitut-
ed biphenols in the cyanation
of benzaldimine 14a. Solvent
screening (see Supporting In-
formation) revealed that n-
hexane was the most suitable,
although the catalyst was un-
dissolved in it. The reactivity
10 (5)
10 (5)
5
10 (5)
5
4 (50)
4 (50)
4
4 (24)
22
90 (84)
>99(S)
10
17j
10 (5)
51 (50)
99 (76)
82 (70)(S)
11
12
13
14
15
R1 =2-furyl, R2 =Me
R1 =2-thienyl, R2 =Me
R1 =Ph, R2 =Et
17k
17l
17m
17n
17o
5
5
22
22
4 (4)
4 (45)
48
97
97
>99 (66)
69 (93)
–
99
>99
>99 (74)
>99 (64)
–
10 (5)
10 (5)
10
R1 =Ph, R2 =nPr
R1 =Ph, R2 =cyclohexyl
16
17
18
17p
17q
17r
5
10
5
17
20
8
97
>99
99
98
71
45
10
10 (5)
10 (5)
4
99
>99 (77)
98 (77)
67 (77)[d]
94 (80)
69 (65)
19
20
R1 =cyclohexyl, R2 =Me
R1 =tBu, R2 =Me
17s
17t
4 (4)
4 (4)
[a] Unless otherwise noted, reactions were carried out with imine (0.1 mmol), TMSCN (0.12 mmol), iPrOH
(0.12 mmol) in toluene (0.5 mL) at À208C. [b] Isolated yield. [c] Determined by HPLC and the absolute con-
figuration was determined by comparison of the optical rotation with literature data.[11h] [d] The reaction was
performed at À458C.
11646
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11642 – 11659

Asymmetric Cyanation
FULL PAPER
Table 5. Substrate scope for the catalytic asymmetric Strecker reaction of
Table 7. Substrate scope for the catalytic asymmetric cyanosilylation of
ketones.[a]
N-Ts diarylketimines.[a]
Entry
Ketone
Product
t [d]
Yield [%][b]
ee [%]
Entry
R
Product Catalyst [mol%] t [h] Yield [%][b] ee [%][c]
1
2
3
4
5
6
7
8
9
PhCOCH3
21a
21b
21c
21d
21e
21 f
21g
21h
21i
1
1
1
1
2
2
1
1
1
2
98
99
98
99
95
83
92
99
96
97
94(S)
92
87(S)
98
90(S)
94(S)
88
70(S)
32
1
2
3
4
2-F
2-F
19a
19a
19b
19c
19c
19d
19e
19 f
19g
19h
19i
5
45
5
26
96
95
90
90
76
98
97
94
97
–
0
2
20
24
28
28
4-F-PhCOCH3
4-Cl-PhCOCH3
4-Br-PhCOCH3
4-Me-PhCOCH3
4-MeO-PhCOCH3
4-NO2-PhCOCH3
3-Cl-PhCOCH3
2-F-PhCOCH3
PhCOC2H5
10
10
10
10
10
10
10
10
10
10
10
2-Cl
2-Me
2-Me
2-CF3
3-Cl
3-Me
4-F
4-Cl
4-Me
4-MeO 19j
45
45
45
45
11
8
6
20
8
5[d]
6
–
7
8
9
10
11
12
>99
>99
>99
>99
>99
>99
10
21j
74
[a] Reaction conditions: ketone (0.2 mmol), TMSCN (0.3 mmol), 1a/Ti/4l
1:1:1, 5 mol%), 2.0m in n-hexane at À208C. [b] Isolated yield.
8
[a] Unless otherwise noted, reactions were carried out with imine
(0.1 mmol), TMSCN (0.12 mmol), iPrOH (0.12 mmol) in toluene
(0.5 mL) at À208C. [b] Isolated yield. [c] Determined by HPLC. [d] The
reaction was performed at À458C.
yield and moderate ee (Table 7, entry 10). Other ketones
such as aliphatic, heterocyclic, and cyclic ketones were also
tested, but only 13–30% ee values were obtained.
Besides, the cyanosilylation of benzaldehyde 22a was also
investigated. After optimizing the conditions (see Support-
ing Information), up to 92% ee was obtained when 1b/Ti/
4m 5:5:2.5 was used as catalyst under the concentration of
0.1m in toluene at À208C. However, separation of pure cya-
nohydrin silyl ether 23a from the reaction mixture was diffi-
cult via flash chromatography as the product was contami-
nated by some impurity with the same Rf value.[17] Due to
this reason, further studies were not proceeded. Despite
this, current catalyst system has shown great potential in the
asymmetric cyanation of aldehyde. As a matter of fact, it
was successfully used in the cyanation of aldehydes with
CNCOOEt as the cyanide source (see Section on Catalytic
asymmetric cyanation of aldehydes).
and enantioselectivity were greatly enhanced, giving the
product in 97% yield and 87% ee within 40 h (Table 6,
entry 2 vs. 3). Increasing the concentration of ketone to
2.0m not only greatly enhanced the reactivity and enantiose-
lectivity of the reaction, but also allowed the catalyst load-
ing and amount of TMSCN to be reduced to 5 mol% and
1.5 equiv, respectively (Table 6, entry 4). Additionally, em-
ploying iPrOH (0.5–1.0 equiv) as additive had no positive
effect on the result.
Table 6. Optimizing conditions for the catalytic asymmetric cyanosilyla-
tion of acetophenone.[a]
Catalytic asymmetric cyanation of aldimines, ketimines, ke-
tones and aldehydes with CNCOOEt as the cyanide source
Entry
Catalyst
t [h]
Yield [%][b]
ee [%]
Undoubtedly, HCN and TMSCN are two of the most com-
monly used cyanide sources for the cyanation. Although re-
markable results have been accomplished, some problems
including high cost, high toxicity, or high volatility are asso-
ciated with them. To overcome these problems, more and
more attention has been focused on the development of al-
ternative cyanide sources which are relatively cheap, less
toxic and easily handling. Till now, much progress has been
achieved for the cyanation of carbonyl compounds (espe-
cially for the aldehydes) using carbonyl cyanide RCOCN
(R=EtO, MeO, Me, Ph),[6] alkyl cyanophosphorylates,[6]
acetone cyanohydrin,[7a–c] and alkali metal cyanides (NaCN/
KCN)[7d–f] as cyanide sources, in which, the stable, less toxic,
relatively cheap, and readily available ethyl cyanoformate
(CNCOOEt) attracted the most attention.[6a–o]
1
2
1a/Ti/4m 1:1:1
1a/Ti/4l 1:1:1
1a/Ti/4l 1:1:1
1a/Ti/4l 1:1:1
96
96
40
20
86
55
97
98
38
74
87
94
3[c]
4[d]
[a] Reaction conditions: acetophenone (0.1 mmol), TMSCN (0.2 mmol),
catalyst (10 mol%), in toluene (0.5m) at À208C. [b] Isolated yield. [c] n-
Hexane was used as solvent. [d] The reaction was run with catalyst
(5 mol%), TMSCN (0.15 mmol), and 2.0m in n-hexane.
As other optimizations gave no superior result, the opti-
mal condition was identified as TMSCN (1.5 equiv), 1a/Ti/4l
1:1:1 (5 mol%), 2.0m in n-hexane at À208C. Under this con-
dition, the substrate scope was examined. para-Substituted
acetophenones with either electron donating or withdrawing
group were suitable substrates (Table 7, entries 1–7). How-
ever, for the meta- and ortho-substituted substrates, less sat-
isfactory results were obtained (Table 7, entries 8,9). Propio-
phenone was transformed to the desired product in good
For the asymmetric Strecker reaction, unfortunately, al-
though some cyanide reagents other than HCN and
TMSCN have once been tried in some reports, disappointing
Chem. Eur. J. 2009, 15, 11642 – 11659
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11647

X. Feng et al.
results were generally observed.[10f,q,r,u] To the best of our
knowledge, the only successful example was reported by
Listꢂs group who developed a thiourea catalyzed asymmetric
cyanation of aldimines with acetyl cyanide (CH3COCN) as
the cyanide reagent, affording highly enantiopure N-acetyl
amino nitriles.[10t,u,y] However, their attempt to involve ket-
Table 8. Substrate scope for catalytic asymmetric cyanation of aldehydes
with CNCOOEt.[a]
Entry
R
Product
t [h]
Yield [%][b]
ee [%][c]
ACHTUNGTRENNUNG
imines as substrates failed.[10y] Therefore, it is still challeng-
1
2
3
4
5
6
7
8
C6H5
4-FC6H4
24a
24b
24c
24d
24e
24 f
24g
24h
24i
24j
24k
24l
24m
24n
5
5
7
7
5
5
7
15
7
7
98
95
90
91
94
97
99
95
99
99
98
99
96
98
94(S)
90
92(S)
91
ing while desirable to realize a highly efficient Strecker reac-
tion with a relatively cheap and convenient cyanide source.
In this part, the asymmetric cyanation of C=O and C=N
with CNCOOEt employing the above developed self-assem-
bled catalyst was described. Considering the problem en-
countered in the cyanosilylation of benzaldehyde, we would
like to begin this part with the study of cyanation of alde-
hyde with CNCOOEt.
4-ClC6H4
4-BrC6H4
4-PhC6H4
4-MeC6H4
4-MeOC6H4
3-PhOC6H4
3-MeOC6H4
3-MeC6H4
2-MeC6H4
2-naphthyl
2-furyl
93
93(S)
90(S)
96
93(S)
92
90
90
93
90
9
10
11
12
13
14
7
15
15
15
Catalytic asymmetric cyanation of aldehydes with
CNCOOEt as the cyanide source: After systematical optimi-
zation (see Supporting Information), we found that the opti-
mal catalyst was the same as that optimized for the cyanosi-
lylation of benzaldehyde. When iPrOH was used as additive,
the reactivity was greatly enhanced with the ee slightly im-
proved. Interestingly, varying the amount of iPrOH (0.25–
3.0 equiv) had no obvious effect on the outcomes.
2-thienyl
15
24o
15
99
84(S)
16
tBu
cyclohexyl
iPr
24p
24q
24r
24s
15
15
15
15
92
99
91
98
95(S)
88(S)
83(S)
72(S)
17[d]
18
19
nC5H11
20
24t
15
85
76(S)
[a] Reaction conditions: 1b (10 mol%), 4m (5 mol%), [TiACHTUNGTRENNUNG(OiPr)4]
(10 mol%), aldehyde (0.1 mmol), CNCOOEt (0.15 mmol), iPrOH
(0.15 mmol) in toluene (0.2 mL) at À208C. [b] Isolated yield. [c] Deter-
mined by HPLC or GC. The absolute configurations were determined by
comparison with literature data (for detail, see Supporting Information).
The substrate scope was investigated under the condition
of 10 mol% of 1b, 5 mol% of 4m, 10 mol% of [TiACTHNURGTNEUNG(OiPr)4],
aldehyde (0.1 mmol), CNCOOEt (1.5 equiv) and iPrOH
(1.5 equiv) in toluene at À208C. Various aromatic aldehydes
were converted to the corresponding products in high yields
and ee values (Table 8, entries 1–11). Especially, 3-phenoxy-
benzaldehyde gave the product in high yield with the best ee
(up to 96%), which was useful for the synthesis of the insec-
ticide fenvalerate Aa.[4c] 2-Naphthaldehyde and heterocyclic
aldehydes also afforded the corresponding products in high
ee values and high yields (Table 8, entries 12–14). When (E)-
cinnamaldehyde was subjected to the reaction, only 1,2-ad-
dition product was afforded in 99% yield with 84% ee
(Table 8, entry 15). Some representative aliphatic aldehydes
were evaluated too. Excitingly, steric bulkier pivalaldehyde
gave the corresponding product with 95% ee and 92% yield
after 15 h (Table 8, entry 16). Cyclohexanecarbaldehyde and
isobutyraldehyde gave 88 and 83% ee, respectively (Table 8,
entries 17,18). Whereas, linear aldehydes such as n-hexanal
and (E)-but-2-enal gave moderate ee values (Table 8, en-
tries 19,20). It is suggested that for the cyanation of aliphatic
aldehydes, bulkier steric hindrance significantly benefited
the enantioselectivity. In addition, acetophenone was un-
reactive under the current conditions. It is still a challenge
to achieve the asymmetric cyanation of acetophenone with
CNCOOEt to date.[18]
[d] 1b (15 mol%), 4m (7.5 mol%) and [TiACHTNUTRGNEUNG(OiPr)4] (15 mol%) were used.
directly employed in the cyanation of ketimines without any
modification.[19] Ketimine 16a could be quantitatively trans-
formed to a-amino nitrile 17a in 99% ee within 42 h
(Table 9, entry 1). Interestingly, unlike the cyanation of alde-
hyde affording the cyanohydrin O-carbonate, no N-ethoxy-
carbonyl protected amino nitrile was observed. This might
be due to the different reaction mechanisms underwent (see
below). Slightly increasing the amount of CNCOOEt and
iPrOH drastically enhanced the reactivity (Table 9, entry 2
vs. 1).
Under the optimized conditions, a series of N-Ts ket-
AHCTUNGTREGiNNNU mines were tested. The results in Table 9 showed, excellent
enantioselectivities (up to 99% ee for most of ketimines)
and high yields were achieved for the aryl methyl ketimines
except ortho-substituted imine 16h (Table 9, entries 2–10).
Heteroaromatic ketimines, aryl ethyl and aryl propyl ket-
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
given in high yield and ee with complete regioselectivity
(Table 9, entry 15). Also worthy of note is that cyanation of
cyclohexyl methyl ketimine exhibited excellent enantioselec-
tivity (>99% ee) (Table 9, entry 16). tert-Butyl methyl ket-
AHCTUNGTREGiNNNU mine gave moderate ee (Table 9, entry 17). Furthermore,
the catalyst system was extended to the challenging Strecker
reaction of unsymmetric diarylketimines. Similar to the case
Catalytic asymmetric cyanation of ketimines with
CNCOOEt as the cyanide source: As there is still no report
dealing with the asymmetric Strecker reaction with
CNCOOEt as the cyanide source, the contribution to this
area was highly desirable. It was found that the catalyst
system optimized for the cyanation of aldehydes could be
with TMSCN as the cyanide source, ortho-substituted diACHTUNGTRENNUNGarACHTUNGTRENNUNGyl-
11648
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11642 – 11659

Asymmetric Cyanation
FULL PAPER
Table 9. Substrate scope for the Strecker reaction of N-Ts ketimines with
Table 10. Optimization of the reaction conditions for the cyanation of
benzaldimine.[a]
CNCOOEt.[a]
Entry L1 L1/4m/[Ti
U
c[b]
[m]
t
Yield[c] ee[d]
Entry
1[d]
2
3
4
5
6
7
8
R1
R2
Product
t [h]
Yield [%][b]
ee [%][c]
[mol%]
[h] [%]
[%]
Ph
Ph
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
17a
17a
17b
17c
17d
17e
17 f
17g
17h
17i
42
22
20
20
20
22
30
30
22
72
22
30
22
22
99
99
99
>99
>99
>99
97
98
99
97
99
99(S)
99(S)
99
99(S)
99(S)
>99(S)
>99(S)
>99
66
99(S)
>99
>99
97
1
2
3
4
5
6
7
8
9
1b 10:5:10
1b 10:10:10
1a 10:10:10
2a 10:10:10
2b 10:10:10
1a 10:10:10
1a 10:10:10
1a 10:10:10
1a 5:5:5
0.2
0.2
0.2
0.2
0.2
0.5
0.5
0.5
0.5
42 93
63 90
63 93
63 89
63 85
30 99
22 99
144 40
46 99
90
93
97
90(R)
86(R)
96
96
95
4-FC6H4
4-ClC6H4
4-BrC6H4
4-MeC6H4
4-MeOC6H4
3-ClC6H4
2-FC6H4
2-naphthyl
2-furyl
9
96
10
11
12
13
14
17k
17l
17m
17n
[a] Unless noted, reactions were carried out with imine (0.1 mmol),
CNCOOEt (0.2 mmol) in toluene at À208C. [b] Concentration. [c] Isolat-
ed yield. [d] Determined by HPLC. Unless noted, (S)-15a was obtained.
2-thienyl
Ph
Ph
95
99
95
nPr
98
15
Me
17p
20
99
96
16
17
18
19
20
21
22
cyclohexyl
tBu
Me
Me
Ph
Ph
Ph
Ph
Ph
17s
17t
19a
19b
19c
19 f
19i
20
20
72
60
72
20
20
99
96
99
99
98
99
99
>99
65
96
96
93
8
Under the conditions applied in Table 10, entry 7, an
array of N-Ts aldimines were evaluated. As shown in
Table 11, most of the aryl aldimines were smoothly convert-
ed to the products in the range of 95–97% ee with high
yields (Table 11, entries 1–7). Condensed aromatic and het-
erocyclic substrates also gave satisfying results (Table 11, en-
tries 8–10). When a,b-unsaturated aldimine 14m was tested,
the desired 1,2-adduct 15m was isolated in 80% yield with
84% ee (Table 11, entry 11). Besides, this catalyst system
was suitable for various N-Ts alkyl imines such as cyclic, a-
disubstituted, and a-trisubstituted substrates, giving good to
excellent ee values (Table 11, entries 12–14).
2-FC6H4
2-ClC6H4
2-MeC6H4
3-MeC6H4
4-MeC6H4
12
[a] Unless noted, reactions were carried out with imine (0.1 mmol),
CNCOOEt (0.2 mmol), iPrOH (0.2 mmol) in toluene (0.2 mL) at À208C.
[b] Isolated yield. [c] Determined by HPLC and the absolute configura-
tions were determined by comparing with literature data (for detail, see
Supporting Information). [d] CNCOOEt (1.5 equiv) and iPrOH
(1.5 equiv) were used.
ACHTUNGTRENNUNGketimines were the most suitable substrates, while meta- or
para-substituted ones only gave poor enantioselectivities
(Table 9, entries 18–22). In general, CNCOOEt showed
lower reactivity than TMSCN, and higher catalyst loading,
higher concentration and longer reaction time were required
to achieve excellent results.
Table 11. Substrate scope for the Strecker reaction of N-Ts aldimines
with CNCOOEt.[a]
Entry
R
Product
t [h]
Yield [%][b]
ee [%][c]
Catalytic asymmetric cyanation of aldimines with
CNCOOEt as the cyanide source: Thereafter, the Strecker
reaction of N-Ts aldimines with CNCOOEt was examined.
Under the similar conditions, aldimime 14a was converted
to 15a in 93% yield and 90% ee within 42 h and no N-
ethoxycarbonyl protected amino nitrile was detected
(Table 10, entry 1). Higher ee value (93%) was obtained by
1
2
3
4
5
6
Ph
15a
15b
15c
15d
15e
15g
22
24
24
24
24
30
99
95
97
99
99
99
96(S)
95
83
97
95
4-FC6H4
4-ClC6H4
4-MeC6H4
3-MeC6H4
4-MeOC6H4
97
7
15h
30
99
97
adjusting the ratio of 1b/[TiACHTNUTRGNENG(U OiPr)4]/4m to 1:1:1, although
8
9
10
1-naphthyl
2-naphthyl
2-thienyl
15i
15j
15l
30
24
30
99
99
99
93
95
86
the reaction became slightly sluggish (Table 10, entry 2).
Among the cinchona alkaloids, cinchonine 1a gave the best
enantioselectivity (up to 97% ee) (Table 10, entries 3–5).
Moreover, both the high concentration and large amount of
iPrOH were necessary to achieve high reactivity (Table 10,
entries 6,7). In the absence of iPrOH, the reaction proceed-
ed extremely slowly (Table 10, entry 8). Decreasing the cata-
lyst loading to 5 mol% led to low reactivity, either
(Table 10, entry 9).
11
15m
48
80
84
12
13
14
cyclohexyl
iPr
tBu
15n
15o
15p
30
24
24
90
61
58
94
86
86
[a] Reaction conditions: imine (0.1 mmol), CNCOOEt (0.2 mmol),
iPrOH (0.2 mmol) and toluene (0.2 mL) at À208C. [b] Isolated yield.
[c] Determined by HPLC.
Chem. Eur. J. 2009, 15, 11642 – 11659
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11649

X. Feng et al.
Reaction mechanism
cinchonine 1a in toluene was a suspension, 1a/[Ti
gave a relatively clearer solution, and 4m/[Ti(OiPr)4] was a
red clear solution (Table 12, entries 1–3). In contrast, when
1a, [Ti(OiPr)4], and 4m were combined to prepare the cata-
ACHTUNGTRENNUNG(OiPr)4]
AHCTUNGTRENNUNG
Insight into the catalyst structure
AHCTUNGTRENNUNG
Control experiments for the catalyst structure studies: The
catalyst structure studies were focused on the complex of
1a/Ti/4m 1:1:1 because it showed the best performance and
widest application among the three catalyst systems de-
scribed above. In this section, some control experiments
were designed to shed some light on the catalyst structure.
Taking cyanation of ketimine 16a with TMSCN as the
model reaction, some modified cinchona alkaloids and bi-
phenols were evaluated to probe the relationship between
the ligand structure and catalytic efficiency. As demonstrat-
ed in Table 12, all the components in the catalyst were very
crucial to ensure the excellent activity and enantioselectivity
(Table 12, entries 1–4). When one or two of the 2-naphthyl
groups at the 3,3’-position of biphenol 4m were removed,
both the yield and ee suffered, which indicated that the
large steric hindrance at the 3,3’-position of 4m was critical
for the construction of the suitable chiral environment for
the asymmetric catalysis (Table 12, entries 5,6). Also, the
role of hydroxyl groups in 4m and 1a was evident. When
one or two hydroxyl groups of 4m were blocked by methyl,
poor results were obtained (Table 12, entries 7,8). Likewise,
if the hydroxyl on C-9 of cinchonine 1a was methylated, ex-
tremely low reactivity and enantioselectivity were observed,
too (Table 12, entry 9).
lyst, a clear orange solution was formed (Table 12, entry 4).
Interestingly, when 4m was partially or fully methylated, the
catalyst solutions were turbid, which had a similar appear-
ance to 1a/[Ti
27 with [Ti(OiPr)4] might hardly occur (Table 12, entries 2,
and 7,8). When the hydroxyl in cinchonine was methylated,
a solution with similar appearance to that of 4m/[Ti(OiPr)4]
ACHTUNGTERNUN(NG OiPr)4], indicating the complexation of 26 or
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
was observed, suggesting 28 might fail to coordinate with
TiIV, and probably the complex formed was the same as that
in 4m/[TiACHTUNGRTNENG(U OiPr)4] (Table 12, entry 3 vs. 9). Thus, both the
difference in catalytic efficacy and the appearance of the
catalyst solutions implied that all hydroxyl groups in cincho-
nine 1a and biphenol 4m were very important, and they
must have participated into the complexation with TiIV in
the catalyst preparation.
1H NMR and 13C NMR analysis for the catalyst structure
studies: To obtain some direct evidences of the catalyst
1
structure, H NMR and 13C NMR analysis were performed.
It should be mentioned that for the NMR analysis, quinidine
1b was used instead of cinchonine 1a, because: 1) 1a and
1b showed comparable results in the catalysis; 2) the me-
thoxyl group (MeO) in 1b was characteristic in NMR analy-
sis, which might supply more useful information.
Besides the catalytic efficiency, the appearance of the dif-
ferent catalyst solutions was also outlined in Table 12. While
When the mixture prepared from quinidine 1b, biphenol
4m, and [TiACTHNUGTRENUNG(OiPr)4] in 1:1:1 ratio was subjected to NMR
1
analysis, to our surprise, very clean and resolvable H NMR
and 13C NMR spectra were obtained, indicating only one
kind of TiIV complex formed in this combination. It was
Table 12. Control experiment for the catalyst structure studies.[a]
speculated that complex 30, (biphenoxide)TiACTHNUTRGENN(GU OR*)ACHTUNGTRENNUNG(OiPr),
was the most possible species generated (Scheme 1). To
1
verify our assumption, assignment of the H and 13C NMR
spectra was desirable. However, it was found that most of
the signals in the downfield were overlapped, which made it
difficult to analyze the aromatic region in both 1H NMR
(7.0–8.0 ppm) and 13C NMR (120–150 ppm) spectra. As a
result, the chemical shifts related to biphenol 4m and some
aromatic protons on quinidine 1b could not be correctly as-
signed. Thus, we mainly focused on assigning the resonances
1
Entry Catalyst[b] Solution
appearance
t [h] Yield [%][c] ee [%][d]
in the upfield of the spectra (0–6.5 ppm for H NMR and 0–
120 ppm for 13C NMR). By using dept-135 and two-dimen-
sional NMR techniques (HMQC and H,H-COSY), partial
1
2
3
4
5
6
7
8
9
1a
1a/Ti
4m/Ti
turbid
turbid
clear
white
white
red
22
22
22
76
65
10
16(S)
27(S)
–
1
assignment of the H and 13C NMR spectra of complex 30
was accomplished with the data summarized in Table 13.
1a/Ti/4m clear
orange
3.5 >99
>99(S)
14(S)
62(S)
26(S)
8(S)
1
For comparative purposes, the H and 13C chemical shifts of
1a/Ti/4a
1a/Ti/25
1a/Ti/26
1a/Ti/27
28/Ti/4m
clear
clear
turbid
turbid
clear
orange 22
orange 22
29
45
68
40
30
quinidine 1b were also given. Significantly, for the 13C NMR
spectrum, the spectral line corresponding to C-8 and C-9 in
1b shifted from 59.8 and 72.1 ppm to 65.8 and 82.2 ppm, re-
spectively, which provided the direct evidence demonstrat-
ing the bond formation between TiIV and the oxygen atom
on C-9 (Figure 3 and Table 13, entries 7,8). Additionally,
complex 30 was almost quantitatively formed since no un-
bound ligand 1b was detected in 1H NMR and 13C NMR
white
white
red
24
22
24
8(S)
[a] Unless noted, reactions were carried out with catalyst (10 mol%),
imine (0.1 mmol), TMSCN (1.2 equiv), iPrOH (1.2 equiv) and toluene
(0.5 mL) at À208C. [b] The Ti catalyst was L1(cinchona alkaloid)/[Ti-
A
ACHTUNGTRENNUNG(phenol) 1:1.2:1.2 or L1(L2)/[TiACHTUNTGERN(NUGN OiPr)4] 1:1. [c] Isolated yield.
11650
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11642 – 11659

Asymmetric Cyanation
FULL PAPER
Table 13. 1H and 13C chemical shifts for 1b and 1b moiety in the complex
30.[a]
Scheme 1), which was agree with the generation of the pro-
posed structure 30.
Entry Carbon
atom
1H chemical shifts in
13C chemical shifts in
CDCl3 [ppm]
Overall, although the high complexity of the catalyst
structure made the complete assignment of the NMR spec-
tra impossible, all data at hand strongly supported the gener-
ation of structure 30 in the catalyst preparation step. ESI-
HRMS characterization of the present catalytic system was
also performed.[20] However, the peak corresponding to the
complex 1b/Ti/4m 1:1:1 was not detected. Instead, the
signal corresponding to quinidine exhibited overwhelming
intensity. Obviously, this contradicted the result from the
NMR analysis that nearly no unbound quinidine existed in
the catalyst solution. Thus, it was supposed that the complex
30 must be unstable under ESI-HRMS conditions.
Interestingly, as shown in Table 13, the upfield shifts for
nearly all the aliphatic protons on the quinuclidine moiety
in quinidine 1b (except the hydrogen atom on the C-8)
were observed upon complexation, indicating that quinucli-
dine moiety might be in the shielding zone of the naphtha-
lene of 4m (Table 13, entries 1–6, 9,10).
CDCl3 [ppm][b]
quinidine 1b moiety in
quinidine 1b moiety in
1b
complex 30
1b
complex 30
1
2
3
4
C-2
C-3
C-4
C-5
2.87, 3.27 2.33, 2.89
49.6
40.1
28.2
26.5
50.2
50.0
38.7
27.4
2.22
1.75
1.50
1.33
ꢀ1.18
0.5, ꢀ1.18
ꢀ25.3
5
C-6
2.74, 2.87 1.08, 2.55
51.0
6
7
8
9
C-7
C-8
C-9
1.16, 2.02 0.84, ꢀ1.18
21.4
59.8
72.1
24.9
65.8
82.2
3.06
5.54
6.01
5.0
8.6
7.49
–
3.61
5.55
4.52
4.21, 4.39
8.42
6.95
–
C-10
C-11
C-2’
C-3’
C-4’
C-5’
C-6’
C-7’
C-8’
C-9’
C-10’
CH3O
140.8
114.4
147.4
118.4
147.7
101.3
157.6
121.5
131.5
126.7
144.2
55.6
137.3
115.0
147.4
120.0
ND[c]
101.6
157.0
121.1
131.5
ND
10
11
12
13
14
15
16
17
18
19
20
7.18
–
6.51
–
7.30
7.95
–
7.30
8.02
–
–
–
ND
55.3
3.94
3.74
The absolute configuration of biphenol 4m in complex 30
[a] The sample was prepared with quinidine 1b (0.1 mmol), [TiACTHNUTRGNEUNG(OiPr)4]
(0.1 mmol), and biphenol 4m (0.1 mmol) in CDCl3 (0.5 mL) at 358C for
0.5 h. Then, it was directly transferred to NMR tube for analysis. For the
full spectra, see Supporting Information. [b] The chemical shifts of the
hydrogen on the corresponding carbon atom. [c] The chemical shift was
not determined.
Control experiments for the studies of the absolute configu-
ration of biphenol 4m: It is well-known that achiral biphe-
nols such as 4m have two low energy chiral conformations
which interconvert rapidly at room temperature. But once it
complexed with a metal such as TiIV, the conformation
would be locked.[12] It thus led to two questions: 1) which
configuration would be preferentially adopted by biphenol
4m upon complexation to form complex 30? 2) whether the
chirality of biphenol 4m could be fully controlled by chiral
quinidine 1b? To probe these problems, 3,3’-di-2-naphthyl-
1,1’-bi-2-naphthol (29) which
spectra (Figures 2 and 3). Moreover, from the spectra,
1.0 equiv of isopropoxyl group (iPrO) bound to TiIV and
3.0 equiv of released iPrOH were observed in the catalyst
solution (the corresponding chemical shifts were shown in
had an analogous structure
with biphenol 4m was synthe-
sized from BINOL. Unlike the
axially flexible achiral biphe-
nol 4m, both the R and S en-
antiomers of 29 were available.
Considering that the 2-naph-
thyl groups at the 3,3’-position
of biphenol 4m played an im-
portant role in constructing an
ideal chiral environment for
the highly efficient asymmetric
catalysis, the distinct spacial
orientations of 3,3’-di-2-naph-
thyl influenced by the axial
chirality of (R)- and (S)-29
must lead to different catalytic
outcomes. As expected, while
1a/Ti/(R)-BINOL and 1a/Ti/
(S)-BINOL gave comparable
results in the cyanation of N-
Ts aldimine 14a (Table 1, en-
tries 4,5), 1a/Ti/(R)-29 and 1a/
Figure 2. 1H NMR spectra were taken in CDCl3. a) 1b, b) 1b/[Ti
1:1:1, d) 1b/[Ti(OiPr)4]/(R)-29 1:1:1, and e) 1b/[Ti(OiPr)4]/(S)-29 1:1:1. For the full spectra, see Supporting In-
formation.
(OiPr)4]/4m 1:1:1, c) 1b/[Ti
E
G
ACHTUNGTRENNUNG
Chem. Eur. J. 2009, 15, 11642 – 11659
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11651

X. Feng et al.
NMR analyses for the studies
of the absolute configuration
of biphenol 4m: According to
Mikami,[12c] the excellent cata-
lytic behavior of the present
catalyst system could arise by
either formation of a single,
highly enantioselective catalyst
or formation of two diastereo-
meric catalysts with huge dif-
ferences in activity and selec-
tivity (Scheme 2). Obviously,
the above control experiments
were unable to tell us which
case exactly existed in the cur-
rent catalyst system. On the
other hand, although both 1H
and 13C NMR spectra of 1b/Ti/
4m 1:1:1 indicated that proba-
Figure 3. 13C NMR spectra were taken in CDCl3. a) 1b, b) 1b/[Ti
1:1:1, d) 1b/[Ti(OiPr)4]/(R)-29 1:1:1, and e) 1b/[Ti(OiPr)4]/(S)-29 1:1:1. For the full spectra, see Supporting In-
formation.
(OiPr)4]/4m 1:1:1, c) 1b/[Ti
N
U
ACHTUNGTRENNUNG
bly there was no isomer exist-
ing in the solution, the possibil-
ity of peak overlapping of the
two diastereomeric isomers
could not be excluded unam-
biguously.
To get more information,
NMR analyses of 1b/Ti/(R)-29
1:1:1, 1b/Ti/(S)-29 1:1.1, and
1b/Ti/(Æ)-29 1:1:1 were con-
ducted. As shown in Figures 2
and 3, both 1b/Ti/(R)-29 1:1:1
and 1b/Ti/(S)-29 1:1:1 gave
similar spectra to 1b/Ti/4m
1:1:1, indicating the complex-
ing behavior of 1b/Ti/(R)-29
Scheme 1. Generation of the complex 30.
and 1b/Ti/(S)-29 were similar
to that of 1b/Ti/4m, and the
analogous complexes 31 and 32
Ti/(S)-29 showed remarkable difference in enantioselectivity
(Table 14, entries 1,2). Similarly, in the Strecker reaction of
N-Ts ketimine 16a, the axial chirality of 29 dramatically af-
fected the result as well. It should be mentioned that (S)-29/
Ti could not catalyze the reaction (Table 14, entry 3). When
1a/Ti/(R)-29 was used, the reaction proceeded slowly and
gave relatively low ee value. In sharp contrast, 1a/Ti/(S)-29
was as efficient as 1a/Ti/4m (Table 14, entries 4–6). Interest-
ingly, 1a/Ti/(Æ)-29 gave excellent result, indicating the reac-
tion process was dominated by 1a/Ti/(S)-29 which was much
more active than 1a/Ti/(R)-29 (Table 14, entry 7). When cin-
chonidine 2a, diastereomer of 1a, was employed as the
chiral inducer, (R)-29 rather than (S)-29 was found to be the
chirally matched ligand. Of note, 2a/Ti/(R)-29 exhibited in-
versed asymmetric induction relative to 1a/Ti/(S)-29
(Table 14, entries 8–10). Based on these observations, it was
reasonable to speculate that for catalyst system 1a/Ti/4m,
biphenol 4m must adopt S configuration in the most active
species.
were supposed to form (Figures 2,3 and Scheme 3). Interest-
ingly, the spectrum of 1b/Ti/(S)-29 was somewhat cleaner
than that of 1b/Ti/(R)-29, implying chirally matched pair of
ligands complexed with TiIV much better (Figures 2,3). Fur-
thermore, although similar, the NMR spectra of diastereo-
meric complexes 31 and 32 were apparently different. Char-
acteristically, the proton on C-9 showed different chemical
shifts (5.516 and 5.653 ppm for 31 and 32, respectively)
(Figure 2). Also, as shown in 13C NMR spectra, almost all
the aliphatic carbons of the alkoxide (-OR*) gave different
chemical shifts (Figure 3). Especially, the spectrum of 1b/Ti/
(Æ)-29 1:1:1 was almost identical to the sum of the spectrum
of 1b/Ti/(R)-29 1:1:1 and 1b/Ti/(S)-29 1:1:1; it was an about
1:1 mixture of 31 and 32 (according to the peak integration
at 5.516 and 5.653 ppm) (Figure 2, spectrum c). No 1b/Ti/
(R)-29/(S)-29 or its analogue was observed.
With these observations in hand, supposing biphenol 4m
would not preferentially adopt R or S configuration during
complexation, 1b/Ti/4m would give a similar spectrum to
11652
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11642 – 11659

Asymmetric Cyanation
FULL PAPER
Table 14. Studies of the absolute configuration adopted by biphenol
ly supported that biphenol 4m must selective-
ly adopt S configuration to form complex 30
(Scheme 2).
4m.[a]
We also investigated the NMR spectra of
1b/Ti/(Æ)-29 1:1:2 and 1b/Ti/(Æ)-29 1:2:2 (for
spectra, see Supporting Information). As ex-
pected, selective generation of 1b/Ti/(S)-29
1:1:1 was observed. Therefore, the chirality
matched 1b/Ti/(S)-29 1:1:1 was much easier
to form than 1b/Ti/(R)-29 1:1:1 under other-
wise identical conditions. It also provided a
Entry
Imine
Catalyst[b]
t [h]
Yield [%][c]
ee [%][d]
1
2
3
4
5
6
7
8
9
14a
14a
16a
16a
16a
16a
16a
16a
16a
16a
1a/Ti/(R)-29
1a/Ti/(S)-29
(S)-29/Ti
1a/Ti/(R)-29
1a/Ti/(S)-29
1a/Ti/4m
1a/Ti/(Æ)-29
2a/Ti/(R)-29
2a/Ti/(S)-29
2a/Ti/4m
4
1
8
3.5
3.5
3.5
3.5
8
98
>99
trace
25
>99
>99
>99
95
62(S)
95(S)
–
71(S)
>99(S)
>99(S)
>99(S)
98(R)
68(R)
94(R)
clue to the question why 1bACHTUNTRGNEUNG(1a)/Ti/(S)-4m
was selectively formed in the catalyst prepa-
ration.
Investigation of the role of iPrOH, the actual
cyanide reagent in the catalytic cycle, and the
role of tertiary amine in the cinchona alka-
loid: In the Strecker reaction of aldimines
and ketimines with either TMSCN or
CNCOOEt as cyanide source, as well as in
the cyanoformation of aldehydes, iPrOH
proved requisite to enhance the catalytic effi-
ciency. So, searching for the exact role of
iPrOH was necessary. Initially, we thought
that the added iPrOH might transform complex 30 to some
more effective species. But it was quickly ruled out by the
NMR analysis, in which no notable change was observed for
8
8
49
87
10
[a] Unless noted, reactions were carried out with catalyst (10 mol%),
imine (0.1 mmol), TMSCN (1.2 equiv), iPrOH (1.2 equiv) and toluene
(0.5 mL) at À208C. [b] The Ti catalyst was L1(cinchona alkaloid)/[Ti-
G
ACHTUNGTRENNUNG(phenol) 1:1.2:1.2 or L2/[TiCAHTUNGTRNE(NUGN OiPr)4] 1:1. [c] Isolated yield.
the H and 13C NMR spectra of complex 30 (prepared on
1
0.2 mmol scale in 0.5 mL CDCl3) after the addition of
iPrOH (1.5 mmol). Then, it was envisaged that iPrOH
would react with TMSCN or CNCOOEt to produce HCN,
which might act the actual nucleophilic cyanide reagent in
the catalytic cycle.
Surprisingly, although it was reported that alcohols involv-
ing primary, secondary, and tertiary alcohols reacted with
TMSCN very fast under neat reaction conditions (generally,
>95% yield in 5 min at 258C),[21] the reaction of secondary
alcohol such as iPrOH with TMSCN hardly occurred in di-
luted conditions. When equimolar amount of TMSCN and
iPrOH (0.2 mmol) were sequentially added and mixed in
0.5 mL CDCl3 at room temperature, only about 3% conver-
sion was detected by NMR even after 16 h (Table 15,
entry 1). By adding 20 mol% of quinidine 1b, the reaction
proceeded swiftly and TMSCN was consumed within 0.5 h
(Table 15, entry 2). Also, the reaction of TMSCN and
iPrOH could be catalyzed by 5 mol% of 1b/[TiACTHNURGTNEUNG(OiPr)4]/4m
Scheme 2. Reaction of 4m, quinidine 1b and [TiACTHNUGTRENUNG(OiPr)4].
1:1:1, and TMSCN could be consumed within 1.5 h
(Table 15, entry 3).
1b/Ti/(Æ)-29. However, as shown in Figures 2 and 3, both
Considering there were two basic nitrogen atoms in quini-
dine 1b, their difference in catalytic ability was then exam-
ined. In the presence of 20 mol% of tertiary amine such as
Et3N, TMSCN was completely converted to HCN within
5 min (Table 15, entry 4). In contrast, when 20 mol% of
quinoline was used, only 10% TMSCN reacted with iPrOH
after 5 min, and about 60% TMSCN was converted to HCN
after 2 h (Table 15, entry 5). Therefore, the tertiary amine of
1
the H and 13C NMR spectra of 1b/Ti/4m and 1b/Ti/(Æ)-29
were completely different, suggesting biphenol 4m must
have preferred adopting R or S configuration upon complex-
1
ing. In addition, the H and 13C NMR spectra of 1b/Ti/4m
exhibited considerable similarity to that of 1b/Ti/(S)-29
rather than 1b/Ti/(R)-29. All of these observations com-
bined with the conclusions from control experiments strong-
Chem. Eur. J. 2009, 15, 11642 – 11659
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11653

X. Feng et al.
of 1b/[Ti
the reaction proceeded very
slowly, and only 38%
ACHTUNGTERN(NUGN OiPr)4]/4m 1:1:1. But
CNCOOEt was converted
after 3 h (Table 15, entry 11).
As shown in the cyanation
of aldehyde and Strecker reac-
tion with CNCOOEt as cya-
nide source, generally 1b/Ti/
4m 2:2:1 was more reactive
than 1b/Ti/4m 1:1:1 while
giving comparable enantiose-
lectivity. Why? According to
Scheme 3. The reaction of 29, quinidine 1b and [TiACTHUNRGTNEUNG(OiPr)4].
the NMR analysis, the spectra
of 1b/Ti/4m 2:2:1 were almost
Table 15. Investigation of the reaction of TMSCN/CNCOOEt with alco-
hols.[a]
identical to the sum of the spectra of 1b/Ti 1:1 and 1b/Ti/
4m 1:1:1. So, the most catalytically active species in 1b/Ti/
4m 2:2:1 was probably the same as that in 1b/Ti/4m 1:1:1.
Then, the reactivity difference between 1b/Ti/4m 2:2:1 and
1b/Ti/4m 1:1:1 might be attributed to the higher catalytic
ability of the former to accelerate the HCN release step,
which might be the rate-determining step in the asymmetric
catalysis, especially for the asymmetric Strecker reaction.
Indeed, when 1b/Ti/4m 2:2:1 was used to catalyze the reac-
tion of CNCOOEt and iPrOH, it was much faster than the
case employing 1b/Ti/4m 1:1:1, and 50% CNCOOEt was
converted after 1.5 h (Table 15, entry 12 vs. 11).
The significant importance of iPrOH, combined with the
above observations that iPrOH would smoothly react with
TMSCN/CNCOOEt in the presence of catalyst involving
1b/Ti/4m, implied that HCN might be the real cyanide re-
agent in the cycle. This assumption was further verified by
some control experiments. As shown in Table 16, either
HCN (prepared from TMSCN and MeOH) or TMSCN/
MeOH (added separately) gave identical result to TMSCN/
iPrOH in the Strecker reaction of aldimines 14a (Table 16,
entries 1–3).
Entry Cyanide
source
R-OH Catalyst
([mol%])
t
Conv.
[%][b]
1
2
3
4
TMSCN
TMSCN
TMSCN
TMSCN
iPrOH none
iPrOH quinidine (20)
iPrOH 1b/Ti/4m (5:5:5) <1.5 h
iPrOH Et3N (20)
16 h
0.5 h
ꢀ3
>99
>99
>99
<5 min
10 min
(2 h)
5
TMSCN
iPrOH
14 (60)
(20)
MeOH none
iPrOH none
MeOH none
iPrOH quinidine (20)
MeOH quinidine (20)
iPrOH 1b/Ti/4m
(10:10:10)
iPrOH 1b/Ti/4m
(10:10:5)
6
TMSCN
10 min
15 h
15 h
0.5 h
0.5 h
3 h
>99
0
0
50
>99
38
7
8
9
CNCOOEt
CNCOOEt
CNCOOEt
CNCOOEt
CNCOOEt
10
11[c]
12[c]
CNCOOEt
1.5 h
55
[a] Reactions were carried out with TMSCN or CNCOOEt (0.2 mmol),
iPrOH or MeOH (0.2 mmol) in CDCl3 (0.5 mL) at room temperature.
[b] Conversion of TMSCN or CNCOOEt; Determined by NMR. [c] 1.0m
[TiACHTUNGTRENNUNG(OiPr)4] in CDCl3 was used to prepare 1b/Ti/4m.
Table 16. Investigation of the real cyanide reagent for the Strecker reac-
tion.[a]
quinuclidine in quinidine 1b should be more efficient than
the quinoline moiety to promote the HCN generation. Ac-
cordingly, the observed high catalytic ability of quinidine 1b
as well as complex 1b/[TiACTHNUTRGNE(UGN OiPr)4]/4m must be primarily con-
tributed by the tertiary amine in quinuclidine. Specially,
MeOH was so reactive that it could react with TMSCN
without any catalyst in CDCl3, and the reaction completed
in less than 10 min (Table 15, entry 6).
Entry Cyanide source and additive ([mmol]) t [h] Yield [%] ee [%]
1
2
3
4
TMSCN (0.12) + iPrOH (0.12)
TMSCN (0.12) + MeOH (0.12)
HCN (0.12)[b]
CNCOOEt (0.2) + iPrOH (0.2)
CNCOOEt (0.2) + MeOH (0.2)
HCN (0.12) + MeOH (0.12)
CNCOOEt (0.2) + MeOH (0.1)
CNCOOEt (0.12) + MeOH (0.12)
1.0
0.5
0.5
60
2.0
1.0
2.0
5.0
99
99
99
93
99
99
99
99
96
96
96
96
84
84
95
92
With respect to CNCOOEt, we found it was inert to
MeOH and iPrOH in CDCl3 at room temperature. No reac-
tion was observed even after 15 h (Table 15, entries 7,8). In
the presence of 20 mol% of quinidine, the reaction of
CNCOOEt with MeOH proceeded swiftly, and CNCOOEt
was completely consumed within 0.5 h (Table 15, entry 9). In
contrast, only about half of the CNCOOEt was consumed
when iPrOH was used under otherwise identical conditions
(Table 15, entry 10). As expected, the reaction of
CNCOOEt and iPrOH could also be catalyzed by 5 mol%
5
6[c]
7
8
[a] Reactions were performed with aldimines 14a (0.1 mmol), catalyst
(1a/[Ti
A
[b] HCN was freshly prepared in situ from equimolar MeOH and
TMSCN in toluene at room temperature. [c] Before the addition of
HCN, to the mixture of catalyst (1a/Ti/4m 1:1:1, 10 mol%) and aldimines
(0.1 mmol) was added MeOH (1.2 equiv, relative to imine), and the mix-
ture was stirred for 0.5 h at À208C.
11654
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11642 – 11659

Asymmetric Cyanation
FULL PAPER
On the other hand, as the asymmetric addition of HCN to
imines was extremely fast, the HCN release step would
become the rate-determining step if it proceeded slowly. Ac-
tually, it was the case in the Strecker reaction with
CNCOOEt as the cyanide source. Due to the low reactivity
of CNCOOEt toward iPrOH even at room temperature
(Table 15, entry 11), it was not surprising that the Strecker
reaction would require extremely long time to reach com-
pletion at À208C (Table 16, entry 4). We speculated that if
the rate of HCN release step could be significantly in-
creased, the reaction time would be greatly shortened. In-
spired by the above finding that MeOH reacted with
CNCOOEt more quickly than iPrOH, MeOH was attempt-
ed as an additive in the Strecker reaction of aldimines 14a.
To our delight, the reactivity was greatly enhanced under
otherwise identical conditions, although slight decrease of ee
was observed (Table 16, entry 5). It was supposed that the
reduced ee might be attributed to the adverse effect of
MeOH to the catalyst system. As anticipated, in the pres-
ence of 1.2 equiv of MeOH (relative to imine), the reaction
sufficient to promote the cyanation of aldehyde with
CNCOOEt, which was remarkably different from the
Strecker reaction in which stoichiometric amount of HCN
was required.
Table 17. Investigation of the real cyanide reagent for cyanation of ben-
zaldehyde with CNCOOEt.[a]
Entry
HCN [mol%][b]
t [h]
Conversion [%][c]
ee [%]
1
2
3
4
5
6
0
5
10
20
40
60
22
22
6
4
4
40
99
99
99
99
99
86
92
90
84
68
62
2
[a] Reactions were performed with benzaldehyde 22a (0.1 mmol), 1a/[Ti-
AHCTUNGTERG(NNNU OiPr)4]/4m 10:10:5 (mol%) (the produced iPrOH was removed after
catalyst preparation), and HCN (0–60 mol%) in toluene (0.2 mL) at
À208C. [b] HCN was freshly prepared in situ from equimolar MeOH and
TMSCN in toluene at room temperature. [c] The conversion of benzalde-
hyde.
of HCN with aldACHTUNGTRENNUNGimines 14a exhibited decreased enantiose-
lectivity (84% ee) (Table 16, entry 6). So, with the dosage of
CNCOOEt maintained (2.0 equiv), decreasing the amount
of MeOH to 1.0 equiv was attempted. Excitingly, both the
high reactivity (2 h) and enantioselectivity (95% ee) were
obtained (Table 16, entry 7). However, when CNCOOEt
and MeOH were synchronously reduced from 2.0 to
1.2 equiv, slightly low reactivity (5 h) and enantioselectivity
(92% ee) were shown (Table 16, entry 8).
Moreover, no obvious influence on the yield and ee value
was observed when 0–3.0 equiv of iPrOH was used as addi-
tive (see Supporting Information). It could be explained by
the fact that the intermediate product cyanohydrin was
much more reactive than iPrOH to react with CNCOOEt.
Thus, once the reaction was initiated, CNCOOEt would
prefer to react with continually generated cyanohydrin. As a
result, the reaction of iPrOH and CNCOOEt was greatly
suppressed, and the HCN would be maintained at a certain
suitable concentration until the reaction finished.
Finally, as demonstrated, the tertiary amine of cinchona
alkaloid in the catalyst complex played a critical role in the
generation of nucleophilic HCN. Moreover, it was expected
to work as Lewis base to activate HCN in the catalytic
cycle. To demonstrate this assumption, some control experi-
ments were performed. As shown in Table 18, when (S)-29/
Ti 10:10 was used as catalyst, the reaction of aldimine 14a
and HCN proceeded very slowly and no ee was observed
(Table 18, entry 1). In contrast, when Et3N was added, the
reaction proceeded very fast (Table 18, entry 2). Further-
more, when quinidine hydrochloride (QD·HCl) or N-Me
quinidinium iodide (QD·MeI) was used instead of quinidine
1b, poor reactivity and ee were observed, which verified the
dual role of the tertiary amine in the cinchona alkaloid
moiety of the catalyst (Table 18, entries 3,4).
As the cyanocarbonylation of aldehydes was concerned,
addition of iPrOH could also enhance the reactivity without
affecting the yield and enantioselectivity. Thus, the real cya-
nide source might also be HCN. As for the fact that the cy-
anation reaction finished in very short time (5 h) while
CNCOOEt reacted with iPrOH slowly under the same reac-
tion conditions, we assumed that HCN release step must not
be the rate-determining step in this case. It was hypothe-
sized that the small amount of HCN produced at the early
stage of the reaction should be recyclable. This assumption
was strongly supported by the following control experi-
ments, in which instead of iPrOH (the in situ produced
iPrOH during the catalyst preparation was also removed
under reduced pressure), 0–60 mol% of HCN was used as
additive. As shown in Table 17, it was found that in the ab-
sence of iPrOH or HCN, the reaction proceeded very slowly
(Table 17, entry 1). When 5–10 mol% of HCN was added,
the reaction was gradually accelerated and high ee was ob-
tained, implying that the small amount of HCN could
smoothly drive the reaction to completion (Table 17, en-
tries 2,3). Although further increasing the amount of the
HCN steadily enhanced the reactivity, the enantioselectivity
suffered (Table 17, entries 4–6). One possible explanation
would be that the reaction would partially proceed through
the uncatalytic pathway in the presence of large amount of
highly reactive HCN (as the control experiment showed,
benzaldehyde could react with HCN without catalyst in tol-
uene at À208C). Above all, catalytic amount of HCN was
Catalytic cycles and transition states: According to the
above control experiments, spectra data and discussions, the
catalyst structure and the real cyanide reagent were identi-
fied. It is reasonable to assume that the reaction proceeds
through a dual activation mechanism.[4i,6f,t] As illustrated in
Scheme 4 and 5, TiIV acts as Lewis acid to activate the elec-
trophile (C=O or C=N) while the tertiary amine in cinchona
Chem. Eur. J. 2009, 15, 11642 – 11659
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11655

X. Feng et al.
Table 18. Investigation of the dual role of cinchona alkaloid moiety in
the catalyst.[a]
Entry
Catalyst [mol%]
t [h]
Yield [%][b]
ee [%]
1
2
3
4
(S)-29/Ti 10:10
8
3
8
8
<22
99
30
0
0
10
0
(S)-29/Ti/Et3N 10:10:10
4m/Ti/QD·HCl 10:10:10
4m/Ti/QD·MeI 10:10:10
10
[a] Reactions were performed with aldimines 14a (0.1 mmol), HCN
(0.12 mmol), catalyst (10 mol%) in toluene (0.5 mL) at À208C. [b] Isolat-
ed yield.
Scheme 5. Proposed catalytic cycle for the cyanation of aldehydes with
CNCOOEt.
alkaloid works as the Lewis base to activate nucleophile
HCN generated in situ from alcohol (iPrOH/MeOH) and
TMSCN/CNCOOEt.
For the Strecker reaction, after the formation of active
species A, enantioselective addition of cyanide ion (CNÀ) to
C=N can easily occur to give intermediate B. Then, a cata-
lytic cycle finishes by intramolecular combination of a
proton to give the product and regenerate the catalyst
(Scheme 4).
more favorable. Moreover, in TS1 the Re face attack is
much more accessible than the Si face because the Si face
was well shielded by the 2-naphthyl in the catalyst. As a
result, the product with S configuration is selectively pro-
duced.
Scheme 4. Proposed catalytic cycle for the Strecker reaction of imines
with TMSCN or CNCOOEt.
Scheme 6. Proposed transition states.
Otherwise, for the cyanation of benzaldehyde with
CNCOOEt, the real cyanide reagent is catalytic amount of
recyclable HCN. As depicted in Scheme 5, likewise, active
species A is formed via dual activation of the substrates.
Conclusion
ꢁ
Then, enantioselective addition of C N to C=O gives inter-
mediate B which readily affords the cyanohydrin by intra-
molecular combination of a proton. Meanwhile, the catalyst
was regenerated. Promoted by the catalyst, the cyanohydrin
would easily react with CNCOOEt to give the desired prod-
uct and reproduce the recyclable HCN. Then, the next cata-
lytic cycle starts (Scheme 5).
The asymmetric cyanation of aldehydes, ketones, aldimines
and ketimines with TMSCN or CNCOOEt as the cyanide
source was studied in detail. Excellent reactivities and enan-
tioselectivities have been achieved for most of these reac-
tions employing a powerful versatile catalyst system generat-
ed from readily available cinchona alkaloid, achiral 3,3’-dis-
Based on the experimental observations, two plausible
transition states TS1 and TS2 are proposed to elucidate the
asymmetric induction (Scheme 6). As the repulsion between
the larger group RL in the substrate and the 2-naphthyl in
the catalyst might occur for TS2, the transition state TS1 is
ubstituted 2,2’-biphenol and [Ti
ACHTUNGTNER(NUGN OiPr)4] under mild reaction
conditions. For instance, 1a/[Ti(OiPr)4]/4m 1:1:1 was general
AHCTUNGTRENNUNG
for the cyanation of N-Ts imines (including aldimines, aryl
alkyl ketimines and unsymmetrical diarylketimines) with
TMSCN as well as for the cyanation of aldimines with
11656
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11642 – 11659

Asymmetric Cyanation
FULL PAPER
[4] For reviews on the asymmetric cyanation of aldehydes and ketones
as well as the application of cyanohydrins, see: a) M. North, Synlett
1993, 807; b) F. Effenberger, Angew. Chem. 1994, 106, 1609; Angew.
Chem. Int. Ed. Engl. 1994, 33, 1555; c) R. J. H. Gregory, Chem. Rev.
1999, 99, 3649; d) M. Shibasaki, M. Kanai, K. Funabashi, Chem.
Commun. 2002, 1989; e) M. North, Tetrahedron: Asymmetry 2003,
14, 147; f) J.-M. Brunel, I. P. Holmes, Angew. Chem. 2004, 116, 2810;
Angew. Chem. Int. Ed. 2004, 43, 2752; g) Synthesis and application
of nonracemic cyanohydrins and a-amino nitriles, Tetrahedron Sym-
posium-in Print, Vol. 60 (Ed.: M. North), 2004, pp. 10383–10568;
h) T. R. J. Achard, L. A. Clutterbuck, M. North, Synlett 2005, 1828;
i) M. Kanai, N. Kato, E. Ichikawa, M. Shibasaki, Synlett 2005, 1491;
j) F. X. Chen, X. M. Feng, Curr. Org. Synth. 2006, 3, 77; k) N. H.
Khan, R. I. Kureshy, S. H. R. Abdi, S. Agrawal, R. V. Jasra, Coord.
Chem. Rev. 2008, 252, 593; l) M. North, D. L. Usanov, C. Young,
Chem. Rev. 2008, 108, 5146; m) J. Gawronski, N. Wascinska, J.
Gajewy, Chem. Rev. 2008, 108, 5227.
[5] For the selected asymmetric cyanation of aldehydes with TMSCN or
HCN as cyanide source, see: a) M. T. Reetz, F. Kunisch, P. Heit-
mann, Tetrahedron Lett. 1986, 27, 4721; b) S. Kobayashi, Y. Tsu-
chiya, T. Mukaiyama, Chem. Lett. 1991, 541; c) H. Nitta, D. Yu, M.
Kudo, A. Mori, S. Inoue, J. Am. Chem. Soc. 1992, 114, 7969; d) M.
Hayashi, Y. Miyamoto, T. Inoue, N. Oguni, J. Org. Chem. 1993, 58,
1515; e) E. J. Corey, Z. Wang, Tetrahedron Lett. 1993, 34, 4001; f) Y.
Hamashima, D. Sawada, M. Kanai, M. Shibasaki, J. Am. Chem. Soc.
1999, 121, 2641; g) Y. N. Belokon’, S. Caveda-Cepas, B. Green, N. S.
Ikonnikov, V. N. Khrustalev, V. S. Larichev, M. A. Moscalenko, M.
North, C. Orizu, V. I. Tararov, M. Tasinazzo, G. I. Timofeeva, L. V.
Yashkina, J. Am. Chem. Soc. 1999, 121, 3968; h) Y. N. Belokon’, M.
North, T. Parsons, Org. Lett. 2000, 2, 1617; i) J. S. You, H. M. Gau,
M. C. K. Choi, Chem. Commun. 2000, 1963; j) I. P. Holmes, H. B.
Kagan, Tetrahedron Lett. 2000, 41, 7453; k) Y. Hamashima, D.
Sawada, H. Nogami, M. Kanai, M. Shibasaki, Tetrahedron 2001, 57,
805; l) C.-W. Chang, C.-T. Yang, C.-D. Hwang, B.-J. Uang, Chem.
Commun. 2002, 54; m) J. Casas, C. Nꢁjera, J. M. Sansano, J. M. Saꢁ,
Org. Lett. 2002, 4, 2589; n) Y. Li, B. He, B. Qin, X. M. Feng, G. L.
Zhang, J. Org. Chem. 2004, 69, 7910; o) D. H. Ryu, E. J. Corey, J.
Am. Chem. Soc. 2004, 126, 8106; p) Y. C. Qin, L. Liu, L. Pu, Org.
Lett. 2005, 7, 2381; q) M. Hatano, T. Ikeno, T. Miyamoto, K. Ishi-
hara, J. Am. Chem. Soc. 2005, 127, 10776; r) Y. H. Wen, X. Huang,
J. L. Huang, Y. Xiong, B. Qin, X. M. Feng, Synlett 2005, 2445;
s) Y. N. Belokon’, V. I. Maleev, M. North, D. L. Usanov, Chem.
Commun. 2006, 4614; t) Y. N. Belokon’, W. Clegg, R. W. Harrington,
C. Young, M. North, Tetrahedron 2007, 63, 5287; u) F. Yang, S. P.
Wei, C. A. Chen, P. H. Xi, L. Yang, J. B. Lan, H. M. Gau, J. S. You,
Chem. Eur. J. 2008, 14, 2223; v) N. Kurono, K. Arai, M. Uemura, T.
Ohkuma, Angew. Chem. 2008, 120, 6745; Angew. Chem. Int. Ed.
2008, 47, 6643.
[6] For the asymmetric cyanation of aldehydes and ketones with
CNCOOEt as cyanide source, see: a) S. K. Tian, L. Deng, J. Am.
Chem. Soc. 2001, 123, 6195; b) J. Tian, N. Yamagiwa, S. Matsunaga,
M. Shibasaki, Angew. Chem. 2002, 114, 3788; Angew. Chem. Int. Ed.
2002, 41, 3636; c) J. Tian, N. Yamagiwa, S. Matsunaga, M. Shibasaki,
Org. Lett. 2003, 5, 3021; d) Y. N. Belokon’, A. J. Blacker, L. A. Clut-
terbuck, M. North, Org. Lett. 2003, 5, 4505; e) Y. N. Belokon’, A. J.
Blacker, P. Carta, L. A. Clutterbuck, M. North, Tetrahedron 2004,
60, 10433; f) N. Yamagiwa, J. Tian, S. Matsunaga, M. Shibasaki, J.
Am. Chem. Soc. 2005, 127, 3413; g) Y. N. Belokon’, E. Ishibashi, H.
Nomura, M. North, Chem. Commun. 2006, 1775; h) Q. H. Li, L.
Chang, X. H. Liu, X. M. Feng, Synlett 2006, 1675; i) S. H. Gou, X. H.
Chen, Y. Xiong, X. M. Feng, J. Org. Chem. 2006, 71, 5732; j) S. H.
Gou, J. Wang, X. H. Liu, W. T. Wang, F. X. Chen, X. M. Feng, Adv.
Synth. Catal. 2007, 349, 343; k) S. H. Gou, X. H. Liu, X. Zhou,
X. M. Feng, Tetrahedron 2007, 63, 7935; l) S. K. Chen, D. Peng, H.
Zhou, L. W. Wang, F. X. Chen, X. M. Feng, Eur. J. Org. Chem. 2007,
639; m) W. T. Wang, S. H. Gou, X. H. Liu, X. M. Feng, Synlett 2007,
2875; n) Y. N. Belokon’, W. Clegg, R. W. Harrington, E. Ishibashi,
H. Nomura, M. North, Tetrahedron 2007, 63, 9724; o) D. Peng, H.
Zhou, X. H. Liu, L. W. Wang, S. K. Chen, X. M. Feng, Synlett 2007,
CNCOOEt; 1a/[Ti
nosilylation of ketones; 1b/[Ti
ally applicable to the cyanation of aldehydes and N-Ts ket-
imines with CNCOOEt. Most significantly, it was for the
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
first time that CNCOOEt was successfully used in the asym-
metric Strecker reaction of aldimines and ketimines, afford-
ing various a-amino nitriles in high yields and ee values.
Compared with the case using TMSCN as the cyanide
source, CNCOOEt showed relatively low reactivity but gave
comparable ee values and yields. Moreover, control experi-
ments and NMR analyses were performed to shed light on
the catalyst structure. It has proved that the sterically de-
manding aromatic substituent at 3,3’-position of biphenol
was crucial to achieve high enantioselectivity and all the hy-
droxyl groups in cinchona alkaloid and biphenol should par-
ticipate in the complexation with TiIV. The absolute configu-
ration adopted by biphenol 4m in the catalyst was deter-
mined as S configuration according to the evidences from
control experiments and NMR studies. The roles of the pro-
tonic additive (iPrOH) and the tertiary amine in the cincho-
na alkaloid were studied in detail, and the real cyanide re-
agent in the catalytic cycle was found to be HCN. Finally,
two plausible catalytic cycles were proposed to elucidate the
reaction mechanisms. Application of the described catalyst
system to other reactions such as cyanation addition to C=C
bond as well as ring opening reaction with cyanide reagents
is in progress.
Experimental Section
Typical procedure for the cyanation of imines with CNCOOEt: Under
Ar atmosphere, [TiACHTUNGTRENNUNG(OiPr)4] was added to a dry tube containing a suspen-
sion of cinchona alkaloid, biphenol and dry toluene. The mixture was
stirred at 358C for 0.5 h. it was diluted with dry toluene to prepare the
catalyst solution in a certain concentration (such as 0.05 or 0.1m). The
catalyst prepared above (0.1m in toluene, 100 uL, 0.01 mmol) was intro-
duced to a dry reaction tube charged with N-Ts imine (0.1 mmol) and tol-
uene (0.1 mL) under Ar atmosphere at À208C. Then, CNCOOEt
(0.2 mmol) and iPrOH (0.2 mmol) were added successively with stirring.
The reaction was monitored by TLC. When the substrate was consumed,
the reaction mixture was subjected to silica gel column chromatography
to afford the desired N-Ts amino nitrile. The ee value was determined by
chiral HPLC.
Acknowledgements
We thank the National Natural Science Foundation of China (Nos.
20732003 and 20602025) for financial support. We also thank Sichuan
University Analytical & Testing Center for NMR analysis and State Key
Laboratory of Biotherapy for HRMS analysis.
[1] F. W. Winkler, Liebigs Ann. Chem. 1832, 4, 246.
[2] A. Strecker, Ann. Chim. Farm. 1850, 75, 27.
[3] a) K. Friedrich, K. Wallenfels in The Chemistry of the Cyano Group
(Ed.: Z. Rappaport), Wiley-Interscience, New York, 1970, p. 67;
b) A. J. Fatiadi in Preparation and synthetic applications of cyano
compounds (Eds.: S. Patai, Z. Rappaport), Wiley-Interscience, New
York, 1983, p. 1057.
Chem. Eur. J. 2009, 15, 11642 – 11659
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11657

X. Feng et al.
2448. For examples of using RCOCN (R=Ph, Me, MeO) and alkyl
cyanophosphorylates as cyanide sources, see: p) A. Baeza, J. Casas,
C. Nꢁjera, J. M. Sansano, J. M. Saꢁ, Angew. Chem. 2003, 115, 3251;
Angew. Chem. Int. Ed. 2003, 42, 3143; q) Y. Abiko, N. Yamagiwa,
M. Sugita, J. Tian, S. Matsunaga, M. Shibasaki, Synlett 2004, 2434;
r) S. Lundgren, E. Wingstrand, M. Penhoat, C. Moberg, J. Am.
Chem. Soc. 2005, 127, 11592; s) A. Baeza, C. Nꢁjera, J. M. Sansano,
J. M. Saꢁ, Tetrahedron: Asymmetry 2005, 16, 2385; t) A. Baeza, J.
Casas, C. Nꢁjera, J. M. Sansano, J. M. Saꢁ, Eur. J. Org. Chem. 2006,
1949.
Chem. 1998, 110, 3369; Angew. Chem. Int. Ed. 1998, 37, 3186;
e) E. J. Corey, M. J. Grogan, Org. Lett. 1999, 1, 157; f) C. A. Krueg-
er, K. W. Kuntz, C. D. Dzierba, W. G. Wirschun, J. D. Gleason, M. L.
Snapper, A. H. Hoveyda, J. Am. Chem. Soc. 1999, 121, 4284;
g) M. S. Sigman, P. Vachal, E. N. Jacobsen, Angew. Chem. 2000, 112,
1336; Angew. Chem. Int. Ed. 2000, 39, 1279; h) M. Takamura, Y. Ha-
mashima, H. Usuda, M. Kanai, M. Shibasaki, Angew. Chem. 2000,
112, 1716; Angew. Chem. Int. Ed. 2000, 39, 1650; i) B. Liu, X. M.
Feng, F. X. Chen, G. L. Zhang, X. Cui, Y. Z. Jiang, Synlett 2001,
1551; j) Z. G. Jiao, X. M. Feng, B. Liu, F. X. Chen, G. L. Zhang,
Y. Z. Jiang, Eur. J. Org. Chem. 2003, 3818; k) W. Mansawat, W.
Bhanthumnavin, T. Vilaivan, Tetrahedron Lett. 2003, 44, 3805;
l) J. M. Keith, E. N. Jacobsen, Org. Lett. 2004, 6, 153; m) S. Naka-
mura, N. Sato, M. Sugimoto, T. Toru, Tetrahedron: Asymmetry 2004,
15, 1513; n) V. Banphavichit, W. Mansawat, W. Bhanthumnavin, T.
Vilaivan, Tetrahedron 2004, 60, 10559; o) J. Huang, E. J. Corey, Org.
Lett. 2004, 6, 5027; p) T. Ooi, Y. Uematsu, K. Maruoka, J. Am.
Chem. Soc. 2006, 128, 2548; q) M. Rueping, E. Sugiono, C. Azap,
Angew. Chem. 2006, 118, 2679; Angew. Chem. Int. Ed. 2006, 45,
2617; r) J. Blacker, L. A. Clutterbuck, M. R. Crampton, C. Grosjean,
M. North, Tetrahedron: Asymmetry 2006, 17, 1449; s) S. Nakamura,
H. Nakashima, H. Sugimoto, N. Shibata, T. Toru, Tetrahedron Lett.
2006, 47, 7599; t) S. C. Pan, J. Zhou, B. List, Angew. Chem. 2007,
119, 3028; Angew. Chem. Int. Ed. 2007, 46, 2971; u) S. C. Pan, B.
List, Org. Lett. 2007, 9, 1149; v) Y. H. Wen, Y. Xiong, L. Chang, J. L.
Huang, X. H. Liu, X. M. Feng, J. Org. Chem. 2007, 72, 7715; w) T.
Ooi, Y. Uematsu, J. Fujimoto, K. Fukumoto, K. Maruoka, Tetrahe-
dron Lett. 2007, 48, 1337; x) M. Negru, D. Schollmeyer, H. Kunz,
Angew. Chem. 2007, 119, 9500; Angew. Chem. Int. Ed. 2007, 46,
9339; y) S. C. Pan, B. List, Chem. Asian J. 2008, 3, 430; z) S. Wꢃnne-
mann, R. Frçhlich, D. Hoppe, Eur. J. Org. Chem. 2008, 684;
aa) Y. H. Wen, B. Gao, Y. Z. Fu, S. X. Dong, X. H. Liu, X. M. Feng,
Chem. Eur. J. 2008, 14, 6789.
[7] For use of acetone cyanohydrin and alkali metal cyanides as cyanide
sources, see: a) T. Ooi, K. Takaya, T. Miura, H. Ichikawa, K. Maruo-
ka, Synlett 2000, 1133; b) T. Ooi, K. Takaya, T. Miura, H. Ichikawa,
K. Maruoka, Tetrahedron 2001, 57, 867; c) A. Watanabe, K. Matsu-
moto, Y. Shimada, T. Katsuki, Tetrahedron Lett. 2004, 45, 6229;
d) C. Girard, H. B. Kagan, Angew. Chem. 1998, 110, 3088; Angew.
Chem. Int. Ed. 1998, 37, 2922; e) Y. N. Belokon’, A. V. Gutnov,
M. A. Moscalenko, L. V. Yashkina, D. E. Lesovoy, N. S. Ikonnikov,
V. S. Larichev, M. North, Chem. Commun. 2002, 244; f) N. H. Khan,
S. Agrawal, R. I. Kureshy, S. H. R. Abdi, V. J. Mayani, R. V. Jasra,
Eur. J. Org. Chem. 2006, 3175.
[8] For the asymmetric cyanation of ketones, see: a) Y. N. Belokon’, B.
Green, N. S. Ikonnikov, M. North, V. I. Tararov, Tetrahedron Lett.
1999, 40, 8147; b) Y. Hamashima, M. Kanai, M. Shibasaki, J. Am.
Chem. Soc. 2000, 122, 7412; c) Y. Hamashima, M. Kanai, M. Shiba-
saki, Tetrahedron Lett. 2001, 42, 691; d) K. Yabu, S. Masumoto, S.
Yamasaki, Y. Hamashima, M. Kanai, W. Du, D. P. Curran, M. Shiba-
saki, J. Am. Chem. Soc. 2001, 123, 9908; e) S. K. Tian, L. Deng, J.
Am. Chem. Soc. 2001, 123, 6195; f) H. Deng, M. P. Isler, M. L. Snap-
per, A. H. Hoveyda, Angew. Chem. 2002, 114, 1051; Angew. Chem.
Int. Ed. 2002, 41, 1009; g) K. Yabu, S. Masumoto, M. Kanai, D. P.
Curran, M. Shibasaki, Tetrahedron Lett. 2002, 43, 2923; h) S. Masu-
moto, M. Suzuki, M. Kanai, M. Shibasaki, Tetrahedron Lett. 2002,
43, 8647; i) F. X. Chen, X. M. Feng, B. Qin, G. L. Zhang, Y. Z. Jiang,
Org. Lett. 2003, 5, 949; j) S. K. Tian, R. Hong, L. Deng, J. Am.
Chem. Soc. 2003, 125, 9900; k) S. K. Tian, Y. G. Chen, J. F. Hang, L.
Tang, P. Mcdaid, L. Deng, Acc. Chem. Res. 2004, 37, 621; l) F. X.
Chen, H. Zhou, X. H. Liu, B. Qin, X. M. Feng, G. L. Zhang, Y. Z.
Jiang, Chem. Eur. J. 2004, 10, 4790; m) Y. C. Shen, X. M. Feng, Y.
Li, G. L. Zhang, Y. Z. Jiang, Eur. J. Org. Chem. 2004, 129; n) B. He,
F. X. Chen, Y. Li, X. M. Feng, G. L. Zhang, Eur. J. Org. Chem. 2004,
4657; o) B. He, F. X. Chen, Y. Li, X. M. Feng, G. L. Zhang, Tetrahe-
dron Lett. 2004, 45, 5465; p) M. Suzuki, N. Kato, M. Kanai, M. Shi-
basaki, Org. Lett. 2005, 7, 2527; q) D. H. Ryu, E. J. Corey, J. Am.
Chem. Soc. 2005, 127, 5384; r) D. E. Fuerst, E. N. Jacobsen, J. Am.
Chem. Soc. 2005, 127, 8964; s) X. H. Liu, B. Qin, X. Zhou, B. He,
X. M. Feng, J. Am. Chem. Soc. 2005, 127, 12224; t) Y. Xiong, X.
Huang, S. H. Gou, J. L. Huang, Y. H. Wen, X. M. Feng, Adv. Synth.
Catal. 2006, 348, 538; u) S. S. Kim, S. H. Lee, J. M. Kwak, Tetrahe-
dron: Asymmetry 2006, 17, 1165; v) S. S. Kim, Pure Appl. Chem.
2006, 78, 977; w) Q. H. Li, X. H. Liu, J. Wang, K. Shen, X. M. Feng,
Tetrahedron Lett. 2006, 47, 4011; x) S. J. Zuend, E. N. Jacobsen, J.
Am. Chem. Soc. 2007, 129, 15872; y) B. Qin, X. H. Liu, J. Shi, K.
Zheng, H. T. Zhao, X. M. Feng, J. Org. Chem. 2007, 72, 2374; z) K.
Shen, X. H. Liu, Q. H. Li, X. M. Feng, Tetrahedron 2008, 64, 147;
aa) A. Alaaeddine, T. Roisnel, C. M. Thomas, J. F. Carpentier, Adv.
Synth. Catal. 2008, 350, 731; ab) M. Hatano, T. Ikeno, T. Matsumura,
S. Torii, K. Ishihara, Adv. Synth. Catal. 2008, 350, 1776.
[11] For the asymmetric Strecker reaction of ketimines, see: a) P. Vachal,
E. N. Jacobsen, Org. Lett. 2000, 2, 867; b) J. J. Byrne, M. Chavarot,
P.-Y. Chavant, Y. Vallꢄe, Tetrahedron Lett. 2000, 41, 873; c) P.
Vachal, E. N. Jacobsen, J. Am. Chem. Soc. 2002, 124, 10012; d) S.
Masumoto, H. Usuda, M. Suzuki, M. Kanai, M. Shibasaki, J. Am.
Chem. Soc. 2003, 125, 5634; e) N. Kato, M. Suzuki, M. Kanai, M.
Shibasaki, Tetrahedron Lett. 2004, 45, 3153; f) N. Kato, M. Suzuki,
M. Kanai, M. Shibasaki, Tetrahedron Lett. 2004, 45, 3147; g) N.
Kato, T. Mita, M. Kanai, B. Therrien, M. Kawano, K. Yamaguchi,
H. Danjo, Y. Sei, A. Sato, S. Furusho, M. Shibasaki, J. Am. Chem.
Soc. 2006, 128, 6768; h) X. Huang, J. L. Huang, Y. H. Wen, X. M.
Feng, Adv. Synth. Catal. 2006, 348, 2579; i) J. L. Huang, X. H. Liu,
Y. H. Wen, B. Qin, X. M. Feng, J. Org. Chem. 2007, 72, 204; j) J.
Wang, X. L. Hu, J. Jiang, S. H. Gou, X. Huang, X. H. Liu, X. M.
Feng, Angew. Chem. 2007, 119, 8620; Angew. Chem. Int. Ed. 2007,
46, 8468; k) Z. R. Hou, J. Wang, X. H. Liu, X. M. Feng, Chem. Eur.
J. 2008, 14, 4484; l) K. Shen, X. H. Liu, Y. F. Cai, L. L. Lin, X. M.
Feng, Chem. Eur. J. 2009, 15, 6008.
[12] For reviews on the asymmetric activation of racemic ligands, see:
a) P. J. Walsh, A. E. Lurain, J. Balsells, Chem. Rev. 2003, 103, 3297;
b) J. W. Faller, A. R. Lavoie, J. Parr, Chem. Rev. 2003, 103, 3345;
c) K. Mikami, M. Yamanaka, Chem. Rev. 2003, 103, 3369; d) K.
Ding, H. Du, Y. Yuan, J. Long, Chem. Eur. J. 2004, 10, 2872.
[13] Part of the data in this article was communicated previously; see
ref. [11j].
[9] For reviews on the asymmetric Strecker reaction, see: a) L. Yet,
Angew. Chem. 2001, 113, 900; Angew. Chem. Int. Ed. 2001, 40, 875;
b) H. Grçger, Chem. Rev. 2003, 103, 2795; c) C. Spino, Angew.
Chem. 2004, 116, 1796; Angew. Chem. Int. Ed. 2004, 43, 1764;
d) S. J. Connon, Angew. Chem. 2008, 120, 1194; Angew. Chem. Int.
Ed. 2008, 47, 1176.
[10] For the asymmetric Strecker reaction of aldimines, see: a) M. S.
Iyer, K. M. Gigstad, N. D. Namdev, M. Lipton, J. Am. Chem. Soc.
1996, 118, 4910; b) M. S. Sigman, E. N. Jacobsen, J. Am. Chem. Soc.
1998, 120, 4901; c) M. S. Sigman, E. N. Jacobsen, J. Am. Chem. Soc.
1998, 120, 5315; d) H. Ishitani, S. Komiyama, S. Kobayashi, Angew.
[14] Isopropanol was added in one portion. Slow addition of iPrOH was
unnecessary to ensure high enantioselectivity. For some discussions
of the role of protic additives, see: ref. 10f, h. For selected examples
using protonic additive to achieve good results, see: ref. [8f,r,s,10f–
h,n,11f–k].
[15] Cyanide agents including tri-n-butyltin cyanide ((nBu)3SnCN), ace-
tone cyanohydrin, and acetyl cyanide (CH3COCN) were examined.
Compared with TMSCN, generally much lower reactivities were ob-
served under otherwise identical reaction conditions.
[16] For reviews, see: a) Comprehensive Asymmetric Catalysis (Eds.:
E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer, Berlin, 1999;
11658
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11642 – 11659

Asymmetric Cyanation
FULL PAPER
b) F. Schmidt, R. T. Stemmler, J. Rudolph, C. Bolm, Chem. Soc. Rev.
2006, 35, 454. For the recent synthesis of chiral sec-diarylmethanols,
see: c) J. Wu, J. Ji, A. S. C. Chan, Proc. Natl. Acad. Sci. USA 2005,
102, 3570; d) A. Kokura, S. Tanaka, T. Ikeno, T. Yamada, Org. Lett.
2006, 8, 3025; e) M. D. Truppo, D. Pollard, P. Devine, Org. Lett.
2007, 9, 335; f) J. Sedelmeier, C. Bolm, J. Org. Chem. 2007, 72, 8859;
g) C.-T. Lee, B. H. Lipshutz, Org. Lett. 2008, 10, 4187. For the recent
synthesis of chiral sec-diarylmethylamines, see: h) H.-F. Duan, Y.-X.
Jia, L.-X. Wang, Q.-L. Zhou, Org. Lett. 2006, 8, 2567; i) R. B. C.
Jagt, P. Y. Toullec, D. Geerdink, J. G. Vries, B. L. Feringa, A. J. Min-
naard, Angew. Chem. 2006, 118, 2855; Angew. Chem. Int. Ed. 2006,
45, 2789; j) Z.-Q. Wang, C.-G. Feng, M.-H. Xu, G.-Q. Lin, J. Am.
Chem. Soc. 2007, 129, 5336. For the synthesis of chiral tert-diarylme-
thanols, see: k) M. A. Fernꢁndez-IbꢁÇez, B. Maciꢁ, A. J. Minnaard,
B. L. Feringa, Org. Lett. 2008, 10, 4041. For the synthesis of chiral
tert-diarylmethylamines, see: l) J. Clayden, J. Dufour, D. M. Graing-
er, M. Helliwell, J. Am. Chem. Soc. 2007, 129, 7488.
Kagan, Tetrahedron Lett. 2002, 43, 5715), in which cyanohydrin ace-
tate could be isolated in pure form.
[18] Although CNCOOEt was successfully used in the cyanation of ali-
phatic ketones, the reaction of CNCOOEt with aromatic ketones
was not mentioned in Dengꢂs report (see: ref. [8e]). In addition, Be-
lokon’ and Northꢂs attempt to employ CNCOOEt as the cyanide
source in the cyanation of ketones also failed (see ref. [6d,e]).
[19] When 1a/Ti/4m 1:1:1 or 1b/Ti/4m 1:1:1 was used instead of 1b/Ti/
4m 2:2:1, although the enantioselectivity was excellent (99% ee),
longer reaction time (78 h) was required to complete the reaction.
On the other hand, lowering the catalyst loading from 10 to 5 mol%
led to a dramatic decrease in reaction rate, and 95% yield and
99% ee were obtained after 5 d.
[20] The spectrum was obtained in positive-ion mode by stepwise dilut-
ing the toluene or CDCl3 solution of 1b/Ti/4m 1:1:1 with acetoni-
trile.
[21] K. Mai, G. Patil, J. Org. Chem. 1986, 51, 3545.
[17] The impurity might be the silylated biphenol. The ee value of the
product was determined by HPLC on Chiralcel OD-H column after
converting 23a to the corresponding cyanohydrin acetate according
to the reported method (S. Norsikian, I. Holmes, F. Lagasse, H. B.
Received: April 8, 2009
Published online: September 22, 2009
Chem. Eur. J. 2009, 15, 11642 – 11659
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11659