Enantioselective addition of trimethylsilyl cyanide to aldehydes catalysed by bifunctional BINOLAM-AlCl versus monofunctional BINOL-AlCl complexes

Article abstract of DOI:10.1016/j.tet.2004.06.137

A highly enantioselective cyanation of aldehydes takes place by using a bifunctional catalyst derived from 3,3′-bis(diethylaminomethyl) substituted binaphthol (BINOLAM) and dimethylaluminium chloride. The addition is of wide scope and runs best in toluene at temperatures ranging from -20 to -40°C, in the presence of 4 A? MS and triphenylphosphine oxide as additives. The (R)- or (S)-cyanohydrins result when using (S)- or (R)-BINOLAM-AlCl complexes, respectively. The valuable ligand can be recovered by simple extractive work-up and recycled without loss of efficiency (both in terms of chemical and stereochemical yields). This methodology is applied to the Shibasaki synthesis of epothilone A. All the evidence available for the BINOLAM-AlCl enantioselective addition of TMSCN to aldehydes call for the intervention of a hydrocyanation reaction, addition of a catalytic amount of hydrogen cyanide, generated in situ, to an aldehyde, followed by O-silylation. In order to determine the role of the basic amino groups of BINOLAM, comparative studies are carried out with the monofunctional 1,1′-binaphthol-derived complex BINOL-AlCl. Graphical Abstract.

Full text of DOI:10.1016/j.tet.2004.06.137

Tetrahedron 60 (2004) 10487–10496
Enantioselective addition of trimethylsilyl cyanide to aldehydes
catalysed by bifunctional BINOLAM-AlCl versus monofunctional
BINOL-AlCl complexes
a
Jesus Casas, Carmen Najera, Jose M. Sansano and Jose M. Saa
a,
a
b,
*
*
´
´
´
´
´
a
´ ´ ´ ´
Departamento de Quımica Organica and Instituto de Sıntesis Organica (ISO), Universidad de Alicante, Apartado 99, 03080 Alicante, Spain
b
´
Departament de Quımica, Universitat de les Illes Balears, 070122-Palma de Mallorca, Spain
Received 23 April 2004; revised 7 June 2004; accepted 25 June 2004
Available online 24 August 2004
Abstract—A highly enantioselective cyanation of aldehydes takes place by using a bifunctional catalyst derived from 3,30-
bis(diethylaminomethyl) substituted binaphthol (BINOLAM) and dimethylaluminium chloride. The addition is of wide scope and runs
˚
best in toluene at temperatures ranging from K20 to K40 8C, in the presence of 4 A MS and triphenylphosphine oxide as additives. The (R)-
or (S)-cyanohydrins result when using (S)- or (R)-BINOLAM-AlCl complexes, respectively. The valuable ligand can be recovered by simple
extractive work-up and recycled without loss of efficiency (both in terms of chemical and stereochemical yields). This methodology is
applied to the Shibasaki synthesis of epothilone A. All the evidence available for the BINOLAM-AlCl enantioselective addition of TMSCN
to aldehydes call for the intervention of a hydrocyanation reaction, addition of a catalytic amount of hydrogen cyanide, generated in situ, to
an aldehyde, followed by O-silylation. In order to determine the role of the basic amino groups of BINOLAM, comparative studies are
carried out with the monofunctional 1,10-binaphthol-derived complex BINOL-AlCl.
q 2004 Elsevier Ltd. All rights reserved.
1. Introduction
Fortunately though, a new generation of bifunctional
BINOL-derived complexes2a,4 emerged as valuable syn-
thetic tools. These catalysts possess functional groupings to
simultaneously bind a basic substrate and an acidic reactant
in a proper manner so that rate, among other factors, is
favourably affected.2a,4 In particular, Shibasaki et al.
reported a series of bifunctional Lewis Acid-Lewis Base
catalysts (LALB) 1 (Fig. 1) derived from BINOL as the
chiral scaffold where an aluminium atom of a bisaryl-
oxyaluminium chloride moiety functions as the Lewis acid
centre (LA) and a suitable functionality at the 3,30 positions
works as the Lewis base centre (LB).5 Specifically, the
phosphine oxide-derived catalysts 1 (RZH, XZPOAr2)
were found to be very efficient for the asymmetric
cyanosilylation of aldehydes,6 imines7 (Strecker synthesis)
and isoquinolines8 (Reissert-type reaction). As an added-
value goal, a polymer-supported catalyst 1 [RZlinker-
(polymer, XZP(O)Ph2] was eventually developed in order
to recover and recycle the chiral ligand, however with some
loss of enantioselectivity.7
The preparation of non-racemic cyanohydrins is an
important goal for current asymmetric organic synthesis.1
The reason for this is clear-cut: enantiomerically pure
cyanohydrins are important building blocks for the synthesis
of several 1,2-bifunctional products such as a-hydroxy-
carbonyl compounds, b-amino alcohols, a-amino acids and
new materials.1,2 Ten years ago, cyclic peptides and
enzymes were overwhelmingly used in the synthesis of
non-racemic cyanohydrins rather than chiral metal com-
plexes,2d but this order has nowadays been completely
reversed.2a Many metal complexes have been successfully
employed as Lewis acids for the addition of HCN or
TMSCN to aldehydes and ketones,2 as for example,
magnesium, zirconium, titanium, aluminium, yttrium,
lanthanum, samarium, vanadium and gadolinium com-
plexes containing mono- or polydentate ligands.2 Nakai
et al.3 were the first to employ monofunctional BINOL-
metal complexes (titanium in particular) for the enantio-
selective cyanosilylation of aldehydes, but only observed
modest enantioselectivity for aliphatic aldehydes.
Recently, we have developed a bifunctional Lewis Acid-
¨
Bronsted Base (LABB) aluminium complex, namely (S)- or
(R)-BINOLAM-AlCl 3, which retains the C2-symmetric
BINOL scaffold but possesses two diethylaminomethyl
arms, instead of the phosphine oxide ones, presumably
Keywords: Asymmetric catalysis; Cyanohydrins; Binolam; Binol; Alu-
minium complex.
* Corresponding authors. Tel./fax: C34-965903549 (C.N.); tel.: C34-
97117326 (J.M.S.); e-mail addresses: cnajera@ua.es; jmsaa@uib.es
working as Bronsted bases.9–11 BINOLAM-AlCl complexes
¨
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.06.137

10488
J. Casas et al. / Tetrahedron 60 (2004) 10487–10496
2. Results and discussion
The optimisation of the variables of the reaction between
benzaldehyde and commercial TMSCN was started by
examining a number of selected Lewis acids and additives
(Scheme 1 and Table 1). For this purpose, all catalysts were
generated in situ by mixing a weighed amount of the ligand
212 (BINOLAM, 10 mol%, relative to aldehyde) with the
appropriate Lewis acid (10 mol%), at room temperature for
1 h, in the selected solvent. This operation was then
followed by the addition of benzaldehyde and commercial
TMSCN (300 mol%) in one portion, at 0 8C, in the presence
˚
of 4 A molecular sieves (MS) (200 mg/mmol of aldehyde).
1
The reaction was monitored by HNMR spectroscopy and,
at the end, after acidic extractive workup with 2 M
hydrochloric acid (more advantageous than neutral workup
which led to mixtures of partially silylated cyanohydrins),
we isolated the pure cyanohydrin 4a. Its enantiomeric
purity, and those of other cyanohydrins 4 was checked by
HPLC (Chiracel OD-H, Chiralpack AD and AS) on
appropriate derivatives (see Section 3 for details). The
acidic aqueous phase was treated with a buffered solution of
1 M ammonia/1 M ammonium chloride until the pH turned
basic. Upon extraction with ethyl acetate we recovered
ligand 2 (BINOLAM) in almost quantitative yield (O95%)
and purity (O98% according to its optical rotation).
Figure 1.
have been proved to be versatile bifunctional catalysts in the
asymmetric synthesis of cyanohydrins,9 O-methoxy-
carbonyl cyanohydrins10 and O-phosphorylated cyano-
hydrins11 from aldehydes, in excellent chemical yields and
very high enantiomeric ratios using TMSCN, CNCO2Me
and CNPO(OEt)2 as cyanation agents, respectively. In
continuing with our studies on the synthetic applications of
chiral (S)- and (R)-BINOLAM-AlCl complexes, we
describe in this article a comprehensive study of these
catalysts for the enantioselective synthesis of cyanohydrins
upon reaction of aldehydes with TMSCN.9 Also, as a
reference point, this analysis includes the study of the
enantioselective catalysis promoted by the analogous
monofunctional BINOL-AlCl catalyst.
Scheme 1.
Initial experiments with titanium(IV) complexes, generated
with TiCl2(OPri)2 or Ti(OPri)2, gave satisfactory enantio-
meric ratios (er) when working in toluene in the presence of
˚
4 A molecular sieves, even slightly better than the
aluminium(III) complex obtained from dimethylaluminium
chloride (Table 1, entries 1–4). However, unlike the
Table 1. Cyanation of benzaldehyde catalysed by preformed Lewis acid (S)-2 complexes
Run
Lewis acid
Solvent
CH2Cl2
Additives
T (8C)
T (h)
4a (%)a
erb
TiCl2(OPri)2
TiCl2(OPri)2
Ti(Opri)4
AlClMe2
AlClMe2
AlClEt2
0
0
0
0
0
0
0
0
K20
K20
K20
K20
K20
K20
K20
24
24
1
26
3
65
83
99
87
99
99
99
99
99
99
83
99
89
99
99
55/45
88/12
78/22
67/33
88/12
90/10
77/23
75/25
O99/1
O1/99
94/6
˚
4 A MS
4 A MS
4 A MS
4 A MS
4 A MS/Ph3PO
4 A MS/Ph3PO
˚
4 A MS/Ph3PO
1
2
3
4
5
6
7
8
˚
˚
˚
˚
PhCH3
PhCH3
PhCH3
PhCH3
PhCH3
CH2Cl2
PhCH3
PhCH3
PhCH3
PhCH3
PhCH3
PhCH3
PhCH3
PhCH3
˚
8
AlClMe2
AlClMe2
AlClMe2
14
20
6
6
16
9
12
4
3
Ph3PO
˚
9
4 A MS/Ph3PO
˚
c
10
11
12
13
14
15
AlClMe2
4 A MS/Ph3PO
˚
AlClMe2
AlClMe2
AlClMe2
3 A MS/Ph3PO
˚
5 A MS/Ph3PO
˚
91/9
96/4
50/50
55/45
d
4 A MS/Ph3PO
˚
AlMe3
AlCNEt2
4 A MS/Ph3PO
˚
4 A MS/Ph3PO
a
The isolated yields given refer to the cyanohydrins obtained after acidic hydrolysis.
Enantiomeric ratios were determined by chiral HPLC analysis (Chiralcel OD-H).
Reaction performed with (R)-BINOLAM.
b
c
d
A 5 mol% charge of (S)-3 was used as catalyst.

J. Casas et al. / Tetrahedron 60 (2004) 10487–10496
10489
titanium complexes, the activity of the aluminium complex
could be further modulated by manipulation of additives and
temperature. For this reason, we then focussed our study on
the enantioselective addition of TMSCN to aldehydes
catalysed by aluminium(III) complexes. A crucial discovery
to the optimisation process was the favourable effect
observed on addition of triphenylphosphine oxide
(40 mol%)6 to the aluminium complexes,13 which led to
an increase of both reaction rate and er (Table 1, entries 4
and 5). In the course of these optimisations we also
exception to this rule, namely that of furfural, is simply due
to a change in the CIP priority of the substituents. Absolute
configurations were assigned on the basis of literature data
(see Section 3 for individual details). The enantiomeric
ratios reported here were determined by chiral HPLC
(Chiralcel OD-H and Chiralpack AD and AS) analyses of
the corresponding O-acetyl, O-benzoyl, O-TMS, or
O-TBDMS cyanohydrins (see Section 3 for individual
details). Prior control tests demonstrated that no racemisa-
tion occurred during these derivatisations. So, it can be
taken for granted that the enantiomeric ratios given actually
represent the enantiomeric ratios of cyanohydrins
themselves.
˚
recognized the importance of a second additive, namely 4 A
MS.14 Mediocre results (both in terms of rate and er) were
obtained in its absence (Table 1, entry 8), whereas in the
˚
presence of 4 A MS, dried at 120 8C for 4 h, excellent results
were obtained (Table 1, entry 5). Thermogravimetric
As shown below, the cyanation reactions appear to be of
general applicability (Scheme 2 and Table 2). Thus,
aromatic aldehydes led to the corresponding cyanohydrins
with excellent er’s and chemical yield by working at
K20 8C (Table 2, entries 1–11). Somewhat lower er’s were
obtained for the case of m-phenoxybenzaldehyde, possibly
because of the steric demand of the phenoxy group, as
illustrated in Table 2 (entries 8 and 9). The behaviour of
heteroaromatic aldehydes also provided indirect evidence in
favour of the hydrocyanation mechanism. Thus, whilst
furfural gave good results when lowering the temperature
down to K40 8C (96/4 er), nicotinaldehyde behaved as a
special case as it required a large excess of TMSCN for the
reaction to give 99% yield (Table 2, entries 12–14), the
enantioselectivity being always a mediocre 75/25 er. In this
case, the heterocyclic nitrogen atom of the pyridine ring
˚
analysis of this partially dried 4 A MS revealed the presence
of 7.5% of water content. On the other hand, the use of
˚
ultradried 4 A MS (200 8C, 6 h, under high vacuum) led to
an extremely slow reaction. So, a small amount of water
seemed to be required for activity.15 However, water itself
can not be used, as suggested by the fact that operating
˚
without 4 A MS, but in the presence (added at the start in
one portion) of an equivalent amount of water leads to
cyanohydrin 4a in very low yield and enantiomeric ratio,
possibly because water destroys the catalyst. Accordingly
˚
we realised at this point that 4 A MS were possibly acting as
an excellent carrier of a limited amount of water, as recently
demonstrated for a similar reaction.16 So, what is the role of
water? Since water could induce a limited hydrolysis of
TMSCN to HCN, we examined and proved by NMR (1H
and 13C) that adding either water or 4 A MS to an NMR tube
˚
¨
works as a basic site (Bronsted base, BB) to capture the
containing TMSCN produced HCN. It appeared then that,
under these reaction conditions, the enantioselective
addition of trimethylsilyl cyanide to the aldehyde was
actually a hydrocyanation reaction followed by O-silylation.
The use of 3 or 5 A MS also led to excellent er, at K20 8C,
˚
but nevertheless inferior to those obtained with 4 A MS
HCN involved in the hydrocyanation reaction, thereby
slowing down the reaction rate and facilitating the
competition of non-asymmetric routes. Conjugated alde-
hydes were also suitable substrates for this cyanation
reaction, the best er’s being obtained when operating at
K20 and K40 8C, depending on the structure of the
aldehyde (Table 2, entries 15–17). Aliphatic aldehydes, on
the other hand, gave cyanohydrins 4 in very good yields and
er’s when operating at K40 8C in short reaction times
(Table 2, entries 18–22). For the particular case of heptanal,
we checked the influence of (a) using an alkylphosphine
oxide such as tri-n-butylphosphine oxide (40 mol%) instead
of triphenylphosphine oxide, and (b) lowering the tempera-
ture to K60 8C, but no improvement was detected (Table 2,
entries 20 and 21, respectively). The ketones did not react
under these conditions, similarly as reported for the
Shibasaki’s conditions employing complexes 1.6
˚
(Table 1, entries 9, 11 and 12). In further optimising work
we also learned that the highest er was reached by operating
at K20 8C in the presence of triphenylphosphine oxide
˚
(40 mol%) and 4 A MS in toluene as solvent (the mixture
was somewhat heterogeneous in this solvent). Other suitable
solvents such as dichloromethane (Table 1, entries 5 and 7),
THF, chloroform and diethyl ether (not included in Table 1)
led to poorer results. As precursory aluminium species we
found dimethylaluminium chloride to be preferred over
diethylaluminium chloride because the former induced a
faster reaction (Table 1, entries 5 and 6). On the other hand,
trimethylaluminium or, more interestingly, diethylalumi-
nium cyanide led to no asymmetric induction (Table 1,
entries 14 and 15), thereby proving that aluminium cyanide
species were not the key derivatives involved in the
enantioselective cyanation observed. The best catalyst
charge was established as 10% molar (relative to aldehyde),
after realising that a substantial reduction in rate occurred
using a 5% molar charge (Table 1, entry 13).
Scheme 2.
Several operational advantages of our methodology with the
BINOLAM-AlCl catalyst need to be mentioned: (a) the
procedure is operationally simple as all reagents are added
at once (no slow pump addition was needed as required by
Shibasaki’s method)8 and takes place in short reaction times
and easy-to-reach operating temperatures; (b) the chiral
ligand employed is stable and it can be recovered almost
Concerning the stereochemical issues, we noticed the
following systematic trends: the (S)-BINOLAM-AlCl
complex leads to the enantiomerically enriched (R)-
configured cyanohydrins 4, and the opposite was also true:
(R)-BINOLAM-AlCl complex leads to (S)-configured
cyanohydrins 4 (Table 1, entries 2, 6, 10). The only

10490
J. Casas et al. / Tetrahedron 60 (2004) 10487–10496
Table 2. Synthesis of enantiomerically enriched 4 catalysed by either (R)- or (S)-3 complexes
Run
R
3
T (8C)
Time (h)
4
Yield (%)a
Conf.b
(R)
erc
1
2
3
4
5
6
7
8
Ph
Ph
Ph
(S)
(R)
(S)d
(S)
(S)
(R)
(S)
(S)
(S)
(R)
(S)f
(R)
(S)
(S)
(S)
(S)
(R)
(R)
(R)
(R)
(R)
(R)
K20
K20
K20
K20
K20
K20
K20
K20
K40
K20
K20
K20
K20
K40
K20
K20
K40
K40
K40
K40
K60
K40
6
6
6
4a
4a
4a
4b
4c
4d
4e
4f
99
99
99
99
O95
99
99
70
45
55
90
99
99
99
59
99
99
99
99
99
87
99
O99/1
O99/1
O99/1
O99/1
99/1e
(S)
(R)
(R)
(R)
(S)
(R)
(R)
(R)
(S)
(R)
(S)
(S)
(S)
(R)
(R)
(S)
(S)
(S)
(S)
(S)
(S)
4-(MeO)C6H4
4-(Me2N)C6H4
2-ClC6H4
4-ClC6H4
3-(PhO)C6H4
3-(PhO)C6H4
20
22
8
21
48
48
24
39
8
98/2
O99/1
85/15
89/11
97/3
9
4f
10
11
12
13
14
15
16
17
18
19
20
21
22
3,4-(OCH2O)C6H3
6-(MeO)-2-naphthyl
3-Pyridyl
4g
4h
4i
95/5
75/25e
88/12
96/4
2-Furyl
2-Furyl
5
4j
12
21
6
12
4.5
3.5
3.5
10
12
4j
(E)-C5H11CH]CH
(E)-PhCH]CH
(E)-PhCH]CH
PhCH2CH2
4kg
4l
98/2
91/9
4l
O99/1
94/6h
83/17i
65/35i
65/35i
60/40k
4m
4n
4n
4n
4o
CH3(CH2)5
CH3(CH2)5
CH3(CH2)5
c-(C6H11)
j
a
Isolated yields of the cyanohydrins 4 after acidic hydrolysis with 2 M hydrochloric acid, except for 4i, which was obtained after quenching with water.
Assigned by comparison to literature data.
Determined by chiral HPLC analysis (Chiralcel OD-H, Chiralpak AD and AS) on the corresponding O-acetyl derivatives.
Reaction performed with recovered ligand obtained by extractive work-up and recrystallisation.
Six equivalents of TMSCN were required.
Determined by chiral HPLC analysis (Chiralcel OD-H) of the corresponding O-TMS derivative.
This compound was characterised as its methoxycarbonyl derivative.
Determined by chiral HPLC analysis (Chiralcel OD-H) on the corresponding O-TBDMS derivative.
Determined by chiral HPLC analysis (Chiralcel OD-H) on the corresponding O-benzoyl derivative.
Tri-n-butylphosphine oxide (40 mol%) was added instead of triphenylphosphine oxide.
Determined by chiral GC analysis (g-cyclodextrin) of the corresponding O-methoxycarbonyl derivative.
b
c
d
e
f
g
h
i
j
k
quantitatively and reused without loss in efficiency (Table 2,
entry 3); (c) the process can be scaled-up to 2.5 mmol, at
least, thereby affording compound 4a after 12 h in O95%
yield and 98.5/1.5 er.
Scheme 3.
We have made efforts to clarify the non-trivial classification
of the BINOLAM-AlCl catalyst as being bifunctional or
monofunctional. For that purpose, we focussed our attention
on monofunctional complexes (R)- or (S)-BINOL-AlCl 5,
which lack the amino arms at C3 and C30 characteristic of
BINOLAM-AlCl, and comparatively studied their effi-
ciency in the enantioselective addition of TMSCN to
aldehydes. We prepared complex (S)-BINOL-AlCl 5 as
described before for BINOLAM-AlCl, and then we
optimised the conditions for the addition of TMSCN to
benzaldehyde. Once again, we found that both triphenyl-
The absence of a clear-cut basic site in BINOL-AlCl 5 (as
compared to BINOLAM-AlCl 3) suggested that addition of
an external base might erode or even annihilate the
enantioselectivity of the reaction by opening-up competing
non-asymmetric processes. This reasoning was actually
supported by experiment as the addition of a substoichio-
metric amount of triethylamine (20 mol%) to the otherwise
standard reaction conditions led to almost racemic
(er 60/40) product (Table 3, entry 6). This result also
suggested that other basic centres in the substrates or
otherwise, capable of capturing HCN might affect the
chemical yield and/or result in lower enantioselectivities.
The results actually found for a variety of aldehydes can be
defined as capricious. Thus, arylaldehydes and hetero-
arylaldehydes and a,b-unsaturated aldehydes either did not
react or afforded almost racemic cyanohydrins (Table 3,
˚
phosphine oxide and 4 A MS (7.5% water content)
accelerated the reaction and pushed the enantioselectivity
of the process towards excellence. Best results were actually
˚
obtained using (S)-BINOL-AlCl together with 4 A MS
(200 mg/mmol relative to aldehyde) and triphenylphosphine
oxide (40 mol%), in toluene, at K20 8C. In this manner,
after acidic workup (2 M HCl), cyanohydrin 4a was isolated
in high yield and er (Scheme 3, Table 3, entry 1). The total
˚
or partial absence of 4 A MS and triphenylphosphine oxide
did not benefit either the chemical or optical yields (Table 3,
entries 1–3). As shown above for the catalysis by
BINOLAM-AlCl, the key operation behind the overall
addition of TMSCN to aldehydes catalysed by BINOL-AlCl
5, under the above reaction conditions, should also be
considered a hydrocyanation reaction.

J. Casas et al. / Tetrahedron 60 (2004) 10487–10496
10491
Table 3. Synthesis of enantiomerically enriched 4 catalysed by the (S)-BINOL-AlCl complex
Run
R
T (8C)
Time (h)
4
Yield (%)a
Conf.b
erc
1
2
3
4
5
6
7
8
Phd
K20
K20
K20
K20
K20
K20
K20
K20
K20
K20
K20
K20
K20
K20
K40
K20
24
24
24
4
3
9
48
24
48
40
24
24
24
12
18
14
4a
4a
4a
4a
4a
4a
4b
4e
4h
4i
4j
4k
4l
4m
4m
4n
9
10
11
99
99
99
24
—
40
99
—
—
(R)
(R)
(R)
(R)
(R)
(R)
—
—
Phe
90/10
92/8
96/4
95/5
60/40
50/50
—
56/44
97/3j
—
Phf
Ph
Phg
Phh
4-(MeO)C6H4
4-ClC6H4
6-(MeO)-2-Naphthyl
3-Pyridyli
2-Furyl
(E)-C5H11CH]CH
(E)-PhCH]CH
PhCH2CH2
PhCH2CH2
CH3(CH2)5
—
9
(R)
(R)
—
—
—
(R)
(R)
(R)
10
11
12
13
14
15
16
—
—
—
O95
O95
O95
90/10k
94/6k
54/46l
a
Isolated yields of the cyanohydrins 4 after acidic hydrolysis using 3 equiv. of TMSCN.
By comparison to known optical rotations.
Determined by chiral HPLC analysis (Chiralcel OD-H, Chiralpak AD and AS) of the corresponding O-acetyl derivatives.
In the absence of both additives.
˚
In the absence of 4 A MS.
In the absence of triphenylphosphine oxide.
The aluminium source was Et2AlCN.
20 mol% of dry triethylamine was added.
The reaction was quenched with water.
Determined by chiral HPLC analysis (Chiralcel OD-H) of the corresponding O-TMS derivative.
Determined by chiral HPLC analysis (Chiralcel OD-H) of the corresponding O-TBDMS derivative.
Determined by chiral HPLC analysis (Chiralcel OD-H) of the corresponding O-benzoyl derivative.
b
c
d
e
f
g
h
i
j
k
l
entries 7–9 and 11–13). The case of nicotinaldehyde,
however, came as a surprise as the cyanohydrin was
obtained in high yield and er (Table 3, entry 10). This
result is now understood as a consequence of the
intermediate formation of the pyridinium salt, likely the
actual species undergoing reaction. Aliphatic aldehydes, on
the other hand, gave different results according to their
structures (Table 3, entries 14–16), that suggest the likely
influential role of CH-p interactions. Thus, whereas strictly
aliphatic aldehydes yielded almost racemic mixtures
(Table 3, entry 16), those having a b-phenyl substituent
yielded the corresponding cyanohydrin in good chemical
yield and er (Table 3, entries 14–15). As for the processes
catalysed by (S)-BINOLAM-AlCl 3, those catalysed by the
pump).6b,27 In our case, slow addition of 2 equiv. of
TMSCN was not productive (Table 4, entry 3) because a
large amount of unreacted aldehyde 6 was observed in the
crude reaction product. Fortunately, when the reaction was
carried out at K20 8C using an excess of TMSCN
(9 equiv.), added in one portion, the reaction was completed
in 36 h in excellent chemical yield and very good
enantiomeric ratio (Table 4, entry 4). Lowering the
operating temperature to K40 8C did not produce a
substantial improvement in the enantiomeric ratio of
compound 4p (Table 4, entry 5). It is worth noting that
the thiazole ring (pKaZ2.4), is less basic than the pyridine
ring (pKaZ5.2), and so is not able to induce by itself a
racemic side-process. (S)-BINOL-AlCl 5 was also tested as
a catalyst in this transformation but the yield was quite low
and the cyanohydrin 4p was finally isolated as a racemic
mixture. In all of the examples the reaction was quenched by
the addition of trifluoroacetic acid at K20 or K40 8C (see
Table 4), the enantiomeric ratios were determined by chiral
HPLC analyses (Daicel Chiralpak AD) of its O-TBDMS
derivative and the absolute configuration was determined by
comparing the optical rotation with that obtained from a
pure sample.6b
(S)-BINOL-AlCl
cyanohydrins.
5
complex gave (R)-configured
Some of the examples depicted in Tables 2 and 3 are
interesting molecules, for example, 4a is a part of new
cyanogenic glycosides isolated from the leaves and roots of
Phyllagatis rotundifolia,17 4b is used in the synthesis of the
naturally occurring (K)-tembamide and (K)-aegeline,3,18
cyanohydrin 4f19,20 is an intermediate in the industrial
production of pyrethroid insecticides,21 compound 4d is an
intermediate in the synthesis of the anti-thrombotic agent
clopidogrel,22 cyanohydrins 4g and 4i are direct precursors
of biogenic amines and chiral 2-amino-1-(3-pyridinyl)etha-
nol,23 respectively, and product 4k is employed in the
preparation of sphingosines24 and coriolic acid.25 In
addition, a direct application of this methodology was the
elaboration of the key intermediate (S)-4p, used in the
synthesis of epothilone A,26 involving the cyanation of the
aldehyde 6 containing a thiazole moiety (Scheme 4).
Previously, compound (S)-4p was prepared at K40 8C in
48 h by adding 1.5 equiv. of TMSCN very slowly (syringe
The proposed mechanism for this reaction significantly
departs from that reported by Shibasaki et al. in a number of
Scheme 4.

10492
J. Casas et al. / Tetrahedron 60 (2004) 10487–10496
Table 4. Enantioselective synthesis of compound 4p
Run
TMSCN (equiv)
T (8C)
T (h)
4p (%)a
erb
1
2
3
4
3
K20
K20
K20
K20
K40
K20
22
48
24
36
48
24
33
52
0
O98
O98
54
96/4
88/12
—
96/4
96/4
50/50
1.5C1.5c
2d
9
9
6
5
6e
a
Isolated yields given refer to the cyanohydrin obtained after quenching the reaction with trifluoroacetic acid.
Enantiomeric ratios were determined by chiral HPLC analysis (Chiralpak AD) of the O-TBDMS derivative.
The second 1.5 equiv. of TMSCN was added after 1 d.
Slow syringe pump addition over 24 h.
Performed with (R)-BINOL-AlCl 5 complex.
b
c
d
e
relevant issues. According to kinetic studies they showed
carbonyl10 and O-phosphorylated11 cyanohydrins, which,
in absence of triphenylphosphine oxide, exhibited
moderate and strong positive non-linear effects (NLE),28
respectively. The pentacoordinated, highly stable,29 inter-
mediate species 8 (Scheme 5) was proposed from the very
broad band observed at 47 ppm in the 27Al NMR
experiment.30
that the two existing phosphine oxide units should play two
different roles: the phosphine oxide located in the integrated
arm of the ligand activated the trimethylsilyl cyanide while
the added phosphine oxide was required to generate a
pentavalent aluminium complex and prevent oligomerisa-
tion.6,13 The study of the 31P NMR spectra of an equivalent
amount of BINOLAM-AlCl complex and triphenyl-
phosphine oxide showed only one signal at 29.5 ppm (free
triphenylphosphine oxide 30.1 ppm) supporting the
existence of catalytic species 7 (Scheme 5). This reaction
also presented linear effects when the enantiomeric excess
of cyanohydrin 4a was plotted versus the corresponding
enantiomeric excess of the catalytic complex; unlike
other reactions catalysed by chiral BINOLAM-AlCl 3
complexes, such as the asymmetric synthesis of O-methoxy-
The contribution of the chlorine to the catalytic cycle of
BINOLAM-AlCl 3 complex was important; when it was not
present the reaction was not so enantioselective as deduced
from entries 14 and 15 of Table 1. A weak interaction of the
aldehydic proton and the chlorine atom,31 together with a
stronger interaction between the carbonyl oxygen atom and
the aluminium centre, presumably fixed the aldehyde to the
chiral catalytic domain.
Scheme 5.

J. Casas et al. / Tetrahedron 60 (2004) 10487–10496
10493
1
The 13C and H NMR spectra of equimolar mixtures of
never with a typical band of pentacoordinated aluminium
species at 47 ppm being observed. The 31P NMR spectra of
an equivalent amount of BINOL-AlCl complex and
triphenylphosphine oxide revealed six different signals at
41.2, 41.5, 42.1, 43.6, 45.3 and 46.3 ppm (free triphenyl-
phosphine oxide 30.1 ppm) corresponding to, at least, six
different coordination sites.
˚
TMSCN/water of 4 A MS, TMSCN/Ph3PO and TMSCN/
˚
Et3N revealed that water contained in 4 AMS interacted
instantaneously to form hydrogen cyanide and trimethyl-
silanol and no noticeable chemical shifts, apart from the
pure compound signals, were observed for the other two
mixtures, thus indicating that triphenylphosphine and
triethylamine are not so good activating agents of
TMSCN. So, the small amount of HCN, generated by the
water content of molecular sieves would react with a
diethylamino group of the ligand in 9 (Scheme 5). As
described above, if we add an exact amount of water
(equivalent to the amount contained by the MS) the reaction
failed and the er of 4a was also very low working with
extremely anhydrous molecular sieves (high vacuum,
200 8C, 6 h) or with molecular sieves saturated in water
(overnight, air, rt). The diethylaminomethyl group would
In summary, we have shown that the BINOLAM-AlCl
catalytic complex is more efficient than the otherwise
simpler LA catalyst BINOL-AlCl in the hydrocyanation of
aldehydes. The catalyst works as a bifunctional LABB by
ligating the carbonyl group to the chiral aluminium, thereby
becoming a pentacoordinated chloro-aluminium species.
This LABB catalyst has similar efficiency to Shibasaki’s
LALB phosphine oxide catalyst while allowing reactions to
be carried out at higher temperatures and with easier set up.
In addition, we have demonstrated that the chiral ligand (S)-
BINOLAM 3 can be easily and quantitatively recovered at
the end of the reaction and recycled, thus suggesting the
possible application of this methodology in large-scale
processes.
¨
act as a Bronsted base capturing the HCN and, conse-
quently, activating the nucleophile, which is supported by
the fact that the addition of a competing base such as
triethylamine (20 mol%), at K20 8C, lowered the er down
to 60:40 and also by the basic nitrogen of 3-pyridinecarbox-
aldehyde also favouring a racemic pathway in the reaction
with (S)-BINOLAM-AlCl 3 (Table 2, entry 12). A further
indirect proof for the intervention of pentacoordinated
aluminum species in the above reactions resulted from
examination of the (S)-BINOL-AlX catalysed cyanation of
benzaldehyde. In a previous work published by Nakai et al.
the in situ formed dicyanotitanium(IV) complex 11 was
predicted as the catalytically active species.3 By analogy,
since both (S)-BINOL-AlCN 12 and (S)-BINOL-AlCl 5
3. Experimental
3.1. General
All reactions were carried out under argon, including the
transfer of the solid reagents to the reaction vessel.
Anhydrous solvents were freshly distilled under an argon
atmosphere and aldehydes were also distilled prior to use.
Molecular sieves were treated at 120 8C for 4 h. (S)- and
(R)-BINOLAM were prepared according to the literature
protocol.9–12 Melting points were determined with a
Reichert Thermovar hot plate apparatus and are uncor-
rected. IR spectra were recorded on a Nicolet 510 P-FT and
only the structurally most important peaks are listed. NMR
spectra were obtained on a Bruker AC-300 using CDCl3 as
solvent and TMS as internal standard unless otherwise
stated. Optical rotations were measured on a Perkin Elmer
341 polarimeter. HPLC analyses were performed on a
Shimadzu LC-10AD equipped with the corresponding
chiral column (Chiralcel OD-H and Chiralpack AD and
AS) described for each compound, using mixtures of
n-hexane/isopropyl alcohol as mobile phase. Chiral GC
analysis was performed on a HP-5890 using a WCOT
g-cyclodextrin column. Low-resolution electron impact
(EI) mass spectra were obtained at 70 eV on a Shimadzu
QP-5000 and high resolution mass spectra were obtained on
a Finnigan VG Platform. HRMS (EI) were recorded on a
Finnigan MAT 95S. Microanalyses were performed on a
Perkin–Elmer 2400 and a Carlo Erba EA1108. Analytical
TLC was performed on Schleicher and Schuell F1400/LS
silica gel plates and the spots were visualized with UV light
at 254 nm. Flash chromatography employed Merck silica
gel 60 (0.040–0.063 mm).
¨
lack the Bronsted base arm present in the BINOLAM ligand
we assumed that catalysis, if occurring, would take place
through the standard tetracoordinated species. In agreement
with this reasoning we found that, in both cases,
benzaldehyde cyanohydrin was obtained in quantitative
yield and high enantioselection as revealed by a 95/5 or 96/4
er, respectively. Therefore we tentatively conclude that
pentacoordinated aluminum catalysts (LABB catalysts) are
more efficient hydrocyanation catalysts than tetracoordi-
nated ones (LA catalysts) alone.
The absence of a NLE in the reaction catalysed by
complexes 5 and 12 also suggest that monomeric species
control the process. Thus, the BINOL-AlCN complex 12
would participate as a monofunctional catalytic species in
the catalytic cycle through a direct cyanide ion transfer from
the aluminium centre in the tetracoordinated complex
obtained from compound 12 and the ligated aldehyde, re-
generating the active complex after a silylation reaction of
the aluminium alcoholate with TMSCN. In order to
experimentally justify this hypothesis, 27Al NMR
(CD2Cl2, 500 MHz) experiments were recorded. The 27Al
NMR experiments of BINOL-AlCl 5 and BINOL-AlCN 12
alone and with additives afforded very confusing results, but
3.2. General procedure for the enantioselective synthesis
of cyanohydrins 4 mediated by (S)- or (R)-BINOLAM-
AlCl or by (S)-BINOL-AlCl
To a suspension of enantiopure (S)- or (R)-BINOLAM 212

10494
J. Casas et al. / Tetrahedron 60 (2004) 10487–10496
or (S)- or (R)-BINOL (0.025 mmol), triphenylphosphine
˚
oxide (0.1 mmol, 28 mg) and 4 A molecular sieves
3.2.4. (S)-2-(2-Chlorophenyl)-2-hydroxyacetonitrile
(4d). Colorless oil; [a]2D5ZK7.0 (c 0.77, CHCl3, 96% ee
from HPLC) [lit.34 [a]D25ZK2.5, for the (R) enantiomer (c
(previously dried at 120 8C for 4 h) in dry toluene (1 mL),
under an inert atmosphere (argon), was added dimethyl-
aluminium chloride (1 M solution in hexanes, 0.025 mmol,
25 mL). The resulting suspension was then stirred at room
temperature for 1 h. This mixture was cooled at K20 or
K40 8C (see Table 2) and then freshly distilled aldehyde
(0.25 mmol) and TMSCN (0.75 mmol, 100 mL) were added
2.9, CHCl3, 67% ee)]; IR (film) nmax: 3409 and 2253 cmK1
;
1H NMR dH: 3.82 (br. s, 1H), 5.86 (s, 1H), 7.36–7.44 (m,
3H), 7.69–7.73 (m, 1H); 13C NMR dC: 60.8, 117.9, 127.6,
128.3, 130.0, 131.0, 132.6, 132.8; MS (EI) m/z: 143 (MCK
14, 2%), 139 (100), 111 (41), 75 (26), 51 (18); HRMS calcd.
for C8H6NOCl: 167.5961, found: 167.5959. HPLC:
Daicel Chiralcel OD-H, lZ254 nm, n-hexane/2-propanol
99/1, 1.0 mL/min, trZ7.9 and 9.2 min for the
O-acetylcyanohydrin.
1
in one portion. The reaction was monitored by H NMR
spectroscopy and when it was judged complete a 2 M
aqueous solution of hydrochloric acid (2 mL) and ethyl
acetate (2 mL) were added, stirring vigorously the resulting
mixture for 1 h. The emulsion was filtered trough a celite
pad and the organic layer was separated, dried (MgSO4) and
evaporated, affording a residue which was purified by flash
chromatography obtaining pure cyanohydrins 4. These
compounds were derivatised for chiral HPLC analysis by
treatment with 3 equiv. of the base in dry dichloromethane
and 1.2 equiv. of the corresponding chloride (AcCl/
pyridine, TBDMSCl/imidazole/DMAP(cat), BzCl/triethyl-
amine or MeOCOCl/triethylamine) overnight at room
temperature. Yields, enantiomeric ratios and other con-
ditions are given in Tables 1–4. The retention time of the
major enantiomer is given in bold. The physical and
spectroscopic data for compounds 4 follow:
3.2.5. (R)-2-(4-Chlorophenyl)-2-hydroxyacetonitrile
(4e).32Colorless oil; [a]D25ZC31.2 (c 0.8, CHCl3, 98% ee
from HPLC) [lit.32,33 [a]D25ZC27.2 (c 1.48, CHCl3, 66%
ee)]; IR (film) nmax: 3411 and 2252 cmK1; 1H NMR dH: 3.89
(br. s, 1H), 5.54 (s, 1H), 7.41–7.50 (m, 4H); 13C NMR dC:
62.8, 118.4, 127.9, 129.3, 133.6, 135.9; MS (EI) m/z: 163
(MC, 4%), 139 (75), 111 (37), 75 (25), 49 (13); HPLC:
Daicel Chiralcel OD-H, lZ254 nm, n-hexane/2-propanol
99/1, 1.0 mL/min, trZ12.8 and 14.5 min for the
O-acetylcyanohydrin.
3.2.6. (R)-2-Hydroxy-2-(3-phenoxyphenyl)acetonitrile
(4f).35 Colorless oil; [a]D25ZC12.9 (c 1, CHCl3, 78% ee
from HPLC); [lit.33 [a]D25ZC14.0 (c 0.83, CHCl3, 79%
ee)]; IR (film) nmax: 3414 and 2251 cmK1; 1H NMR dH: 2.78
(br. s, 1H), 5.51 (s, 1H), 7.02–7.42 (m, 9H); 13C NMR dC:
63.2, 116.6, 118.5, 119.3, 119.6, 120.9, 123.9, 129.9, 130.6,
137.0, 156.3, 158.1; MS (EI) m/z: 199 (MCK26, 13%), 198
(100), 169 (55), 141 (49), 115 (29), 77 (52); HPLC:
Daicel Chiralcel OD-H, lZ254 nm, n-hexane/2-propanol
90/10, 1.0 mL/min, trZ14.0 and 20.4 min for the
O-acetylcyanohydrin.
3.2.1. (R)-Mandelonitrile (4a).32 Colorless oil; [a]D25ZC
42.2 (c 1, CHCl3, 99% ee from HPLC) [lit.3 [a]D25ZC36.8
(c 2, CHCl3, 85% ee)]; IR (film) nmax: 3222 and 2249 cmK1
;
1H NMR dH: 3.18 (br. s, 1H), 5.53 (s, 1H), 7.42–7.45 (m,
3H), 7.50–7.53 (m, 2H); 13C NMR dC: 63.6, 118.7, 126.6,
129.2, 129.8, 135.0; MS (EI) m/z: 107 (MCK26, 8%), 106
(78), 105 (88), 77 (100), 51 (74); HPLC: Daicel Chiralcel
OD-H, lZ254 nm, n-hexane/2-propanol 99/1, 1.0 mL/min,
trZ11.0 and 14.0 min for the O-acetylcyanohydrin.
3.2.7. (S)-2-Hydroxy-2-(3,4-methylenedioxyphenyl)-
acetonitrile (4g).36 Colorless oil; [a]2D5ZK41.7 (c 1,
CHCl3, 94% ee from HPLC) [lit.34 [a]D25ZC40.8, (R)
enantiomer (c 1.1; CHCl3, 73% ee)]; IR (film) nmax: 3454
and 2239 cmK1; 1H NMR dH: 3.07 (br. s, 1H), 5.43 (s, 1H),
6.00 (s, 2H), 6.81–6.84 (m, 1H), 7.01–7.12 (m, 2H); 13C
NMR dC: 63.4, 101.6, 107.2, 108.6, 118.7, 120.7, 128.8,
143.3, 143.8; MS (EI) m/z: 151 (MCK26, 8%), 150 (84),
149 (100), 121 (31), 91 (10), 63 (26); HPLC: Daicel
Chiralcel OD-H, lZ254 nm, n-hexane/2-propanol 99/1,
1.0 mL/min, trZ16.1 and 18.2 min for the
O-acetylcyanohydrin.
3.2.2. (R)-2-Hydroxy-2-(4-methoxyphenyl)acetonitrile
(4b).34 Colorless oil; [a]D25ZC40.7 (c 1, CHCl3, 99% ee
from HPLC) [lit.34 [a]D25ZC36.3 (c 0.97, CHCl3, 83% ee)];
IR (film) nmax: 3420 and 2249 cmK1; 1H NMR dH: 3.65 (br.
s, 1H), 3.81 (s, 3H), 5.46 (s, 1H), 6.93 (d, JZ8.7 Hz, 2H),
7.42 (d, JZ8.7 Hz, 2H); 13C NMR dC: 55.3, 63.1, 114.4,
119.0, 127.6, 128.2, 160.4; MS (EI) m/z: 161 (MCK2, 1%),
145 (3), 135 (100), 107 (13), 92 (15), 77 (40); HPLC:
Daicel Chiralcel OD-H, lZ254 nm, n-hexane/2-propanol
99/1, 1.0 mL/min, trZ11.2 and 12.6 min for the
O-acetylcyanohydrin.
3.2.3. (R)-2-Hydroxy-2-[4-(N,N-dimethylamino)phenyl]-
acetonitrile (4c). Pale yellow prisms, mp 105–106 8C (from
n-hexane/ethyl acetate); [a]2D5ZC83.3 c 0.6, CHCl3, 98%
ee from HPLC) [lit.34 [a]D25ZC45.1, CHCl3, 53% ee)]. IR
(film) nmax: 3432 and 2244 cmK1; 1H NMR dH: 2.56 (br. s,
1H), 2.99 (s, 6H), 5.43 (s, 1H), 6.75 (d, JZ8.8 Hz, 2H), 7.39
(d, JZ8.8 Hz, 2H); 13C NMR dC: 40.36, 63.31, 112.45,
119.31, 123.05, 127.97, 151.16; MS (EI) m/z: 14 (MCK27,
84%), 148 (100), 132 (14), 77 (22), 51 (13); HRMS calcd.
for C10H12N2O: 176.2194, found: 176.2198. HPLC: Daicel
Chiralcel OD-H, lZ254 nm, n-hexane/2-propanol 99/1,
1.0 mL/min, trZ6.7 and 7.2 min for the O-TMS-
cyanohydrin.
3.2.8. (R)-2-Hydroxy-2-(6-methoxynaphthalen-2-yl)-
acetonitrile (4h).36 Pale yellow needles, mp 113–114 8C
(from n-hexane/ethyl acetate); [a]2D5ZC24.0 (c 0.9, CHCl3)
(90% ee from HPLC); IR (KBr) nmax: 3439 and 2240 cmK1
;
1H NMR dH: 3.91 (s, 3H), 4.81 (br. s, 1H), 5.89 (s, 1H), 7.19
(dd, JZ2.4, 8.9 Hz, 1H), 7.32 (d, JZ2.3 Hz, 1H), 7.61 (m,
1H), 7.87 (m, 2H), 7.98 (s, 1H); 13C NMR (d6-DMSO) dC:
55.6, 63.5, 106.6, 120.2, 120.7, 125.4, 126.3, 128.4, 129.3,
130.4, 133.1, 135.8, 159.3; MS (EI) m/z: 186 (MCK27,
100%), 157 (38), 115 (41); HRMS calcd. for C13H11NO2:
187.0759, found: 187.0739; Microanalyses required for
C13H11NO2: C, 72.0%; H, 5.1%; N, 6.3%; found: C,
73.2%; H, 5.2%; N, 6.5%; HPLC: Daicel Chiralcel OD-H,

J. Casas et al. / Tetrahedron 60 (2004) 10487–10496
10495
lZ254 nm, n-hexane/2-propanol 99/1, 1.0 mL/min, trZ
22.3 and 26.4 min for the O-acetylcyanohydrin.
3.2.14. (S)-2-Hydroxy-n-octanenitrile (4n).6b Colorless
oil; [a]2D5ZK8.0 (c 1.3, CHCl3, 66% ee from HPLC) [lit.6b
[a]2D5ZK13.3 (c 1, CHCl3, 98% ee)]; IR (film) nmax: 3327
and 2247 cmK1; 1H NMR dH: 0.9 (t, JZ7.0 Hz, 3H), 1.22–
1.38 (m, 6H), 1.42–1.54 (m, 2H), 1.82–1.88 (m, 2H), 2.19
(br. s, 1H), 4.47 (t, JZ6.7 Hz, 1H); 13C NMR dC: 13.9, 22.4,
24.4, 28.5, 31.5, 36.2, 61.4, 120.0; MS (EI) m/z: 141 (MC,
6%), 129 (11), 114 (19), 101 (28), 75 (40), 55 (100); HPLC:
Daicel Chiralcel OD-H, lZ254 nm, n-hexane/2-propanol
99.6/0.4, 0.5 mL/min, trZ31.8 and 35.6 min for the
O-benzoylcyanohydrin.
3.2.9. (S)-2-Hydroxy-2-(3-pyridinyl)acetonitrile (4i).37
Pale yellow oil; [a]D25ZK22.1 (c 1CHCl3) (50% ee from
HPLC) [lit.37 [a]D25ZC21.4, (R) enantiomer (c 0.95,
CHCl3) 50% ee)]; IR (film) nmax: 3429 and 2239 cmK1
;
1H NMR dH: 5.53 (br. s, 1H), 5.65 (s, 1H), 7.34 (dd, JZ4.8,
7.9 Hz, 1H), 7.82 (d, JZ7.9 Hz, 1H), 8.61 (d, JZ4.8 Hz,
1H), 8.66 (s, 1H); 13C NMR dC: 61.5, 118.9, 123.7, 132.2,
134.1, 147.6, 150.4; MS (EI) m/z: 108 (MCK26, 4%), 84
(100); HPLC: Daicel Chiralcel OD-H, lZ254 nm,
n-hexane/2-propanol 99/1, 1.0 mL/min, trZ12.1 and
15.3 min. for the O-trimethylsilyl cyanohydrin.
3.2.15. (S)-2-Cyclohexyl-2-hydroxyacetonitrile (4o).39
Pale yellow oil; [a]D25ZK2.3 (c 2.0, CHCl3, 20% ee from
GC) [lit.39 [a]D25ZC8.2, (R) enantiomer (c 0.77, CHCl3,
79% ee)]; IR (film) nmax: 3441 and 2249 cmK1; 1H NMR dH:
1.07–1.35 (m, 6H), 1.68–1.88 (m, 5H), 3.47 (br. s, 1H), 4.26
(d, JZ6.6 Hz, 1H); 13C NMR dC: 25.8, 27.7, 28.1, 45.7,
66.1, 119.4; MS (EI) m/z: 112 (MCK27, 6%), 94 (18), 83
(46), 68 (27), 55 (100); GC: WCOT g-CD column (0.25 nm
diameter, stationary phase FS-Lipodex-E with a film thick
3.2.10. (S)-2-(2-Furyl)-2-hydroxyacetonitrile (4j).38
Colorless oil; [a]2D5ZC45.6 (c 0.8, CHCl3, 92% ee from
HPLC) [lit38 [a]D25ZC50.3 (c 1.6, CHCl3, 99% ee)]; IR
(film) nmax: 3270 and 2258 cmK1; 1H NMR dH: 2.98 (br. s,
1H), 5.54 (s, 1H), 6.43 (dd, JZ3.0, 1.9 Hz, 1H), 6.60 (d, JZ
3.0 Hz, 1H), 7.49 (d, JZ1.9 Hz, 1H); 13C NMR dC: 60.7,
109.9, 110.8, 117.0, 144.2, 148.5; MS (EI) m/z: 106 (MCK
17, 1%), 96 (100); HPLC: Daicel Chiralpak AS, lZ254 nm,
n-hexane/2-propanol 95/5, 1.0 mL/min, trZ21.0 and
28.9 min for the O-acetylcyanohydrin.
of 0.25 mm), TinjectorZ250 8C, TdetectorZ260 8C, Tcolumn
Z
90 8C (3 min) and 180 8C (10 8C/min), PZ120 kPa, trZ
21.2 and 21.5 min for the O-methoxycarbonylcyanohydrin.
3.2.16. Synthesis of (S)-2-hydroxy-3-methyl-4-(2-methyl-
4-thiazolyl)-3-butenenitrile (4p).6b The reaction was
performed as described before starting from the aldehyde
66b adding 9 equiv. of trimethylsilylcyanide in one portion,
keeping the reaction at K20 8C for 38 h. The work up and
purification methods were done according to the literature6b
obtaining pure compound 4p as colorless oil; [a]2D5ZK14.8
(c 0.7, CHCl3, 92% ee from HPLC) [lit.6b [a]D25ZK16.5, (c
3.2.11. (2R,3E)-2-(Methoxycarbonyloxy)-3-nonenenitrile
(O-methoxycarbonyl-4k). Colorless oil; [a]2D5ZK14.4 (c
2.0, CHCl3, 96% ee from HPLC); IR (film) nmax: 2249 and
1763 cmK1; 1H NMR dH: 0.89 (t, JZ6.7 Hz, 3H), 1.25–1.30
(m, 4H), 1.40–1.45 (m, 2H), 2.13 (m, 2H), 3.86 (s, 3H),
5.54–5.61 (m, 1H), 5.66–5.68 (m, 1H), 6.19 (dt, JZ14.6,
7.2 Hz, 1H); 13C NMR dC: 13.8, 22.3, 27.8, 31.1, 31.9, 55.6,
65.1, 115.2, 119.4, 141.5, 154.0; MS (EI) m/z 211 (MC,
2%), 154 (34), 136 (22), 120 (31), 106 (38), 93 (46), 80 (99),
69 (57), 55 (100); HRMS calcd. for C11H17NO3: 211.1208,
found: 211.1212; HPLC: Daicel Chiralcel OD-H, lZ
210 nm, n-hexane/2-propanol 95/5, 0.5 mL/min, trZ9.6
and 10.8 min.
0.7, CHCl3, 99% ee)]; IR (film) nmax: 3354 and 2346 cmK1
;
1H NMR dH: 2.19 (s, 3H), 2.73 (s, 3H), 4.92 (s, 1H), 6.66 (s,
1H), 7.05 (s, 1H); 13C NMR dC: 14.3, 19.2, 67.7, 117.7,
118.5, 121.6, 133.9, 151.6, 165.2; m/z: 193 (MC, 1%), 177
(49), 150 (46), 136 (41), 75 (49); HPLC: Daicel Chiralpak
AD, lZ254 nm, n-hexane/2-propanol 99/1, 0.5 mL/min,
trZ11.2 and 12.3 min, for the O-TBDMS-cyanohydrin.
3.2.12. (2S,3E)-2-Hydroxy-4-phenyl-3-butenenitrile
(4l).39 Colorless oil; [a]D25ZK22.7 (c 1, CHCl3, 99% ee
from HPLC) [lit.33 [a]D25ZC19.2, (R) enantiomer (c 1.9,
CHCl3, 72% ee)]; IR (film) nmax: 3413 and 2249 cmK1; 1H
NMR dH: 3.65 (br. s, 1H), 5.14 (d, JZ5.9 Hz, 1H), 6.25 (dd,
JZ15.8, 5.9 Hz, 1H), 6.91 (d, JZ15.8 Hz, 1H), 7.25–7.40
(m, 5H); 13C NMR dC: 61.8, 118.1, 122.3, 127.0, 128.8,
129.0, 134.7, 135.2; MS (EI) m/z: 142 (MCK17, 1%), 131
(100), 103 (53), 77 (39); HPLC: Daicel Chiralpak AS, lZ
254 nm, n-hexane/2-propanol 99.3/0.7, 1.0 mL/min, trZ
16.5 and 18.8 min for the O-acetylcyanohydrin.
Acknowledgements
This work has been supported by the DGES of the Spanish
´
Ministerio de Ciencia y Tecnologıa (BQU2001-0724-C02),
by Generalitat Valenciana (CTIOIB/2002/320 and
GRUPOS03/134) and by the University of Alicante.
3.2.13. (S)-2-Hydroxy-4-phenylbutanenitrile (4m).6b
Colorless oil; [a]2D5ZC11.3 (c 0.9; CHCl3, 88% ee from
HPLC) [lit.6b [a]D25ZC6.6 (c 0.68, CHCl3, 97% ee)]; IR
References and notes
1. (a) Catalytic Asymmetric Synthesis, 2nd ed.; Ojima, I., Ed.;
Wiley-VCH: Weinheim, 2000. (b) Comprehensive Asym-
metric Catalysis I–III; Jacobsen, E. N.; Pfaltz, A.; Yamamoto,
H., Eds.; Springer: Berlin, 1999. (c) Helmchen, G.; Hoffmann,
R. W.; Mulzer, J.; Schaumann, E.; Methods of Organic
Chemistry, Stereoselective Synthesis. In Houben-Weyl;
Theime: Stuttgart/New York, Vol. E21, 1995.
1
(film) nmax: 3433 and 2248 cmK1; H NMR dH: 2.11–2.19
(m, 2H), 2.83 (t, JZ7.1 Hz, 2H), 3.55 (br. s, 1H), 4.40 (t,
JZ6.7 Hz, 1H), 7.18–7.33 (m, 5H); 13C NMR dC: 30.6,
36.5, 60.3, 119.8, 126.5, 128.4, 128.6, 139.5; MS (EI) m/z:
135 (MCK26, 6%), 134 (63), 105 (33), 91 (100), 78 (45),
65 (17), 51 (22); HPLC: Daicel Chiralcel OD-H, lZ
254 nm, n-hexane/2-propanol 99/1, 1.0 mL/min, trZ8.9 and
12.2 min for the O-TBDMS-cyanohydrin.
2. For reviews about the asymmetric synthesis of cyanohydrins,
see: (a) North, M. Tetrahedron: Asymmetry 2003, 14,

10496
J. Casas et al. / Tetrahedron 60 (2004) 10487–10496
147–176. (b) Gregory, R. J. H. Chem. Rev. 1999, 99,
19. Fadnavis, N. W.; Babu, R. L.; Sheehe, G.; Deshpande, A.
Tetrahedron: Asymmetry 2001, 12, 1695–1699.
20. Tolman, C. A.; Parshall, G. W. J. Chem. Educ. 1999, 76,
178–189.
3649–3682. (c) Effenberger, F. Angew. Chem., Int. Ed. Engl.
1994, 33, 1555–1563. (d) North, M. Synlett 1993, 807–820.
3. Mori, M.; Imma, H.; Nakai, T. Tetrahedron Lett. 1997, 38,
6229–6232.
21. (a) Beckmann, M.; Haack, K.-J. Chem. Unserer Zeit. 2003, 37,
88–97. (b) Breuer, M.; Ditrich, K.; Habicher, T.; Hauer, B.;
Kebeler, M.; Stu¨rmer, R.; Zelinski, T. Angew. Chem., Int. Ed.
2004, 43, 788–824.
4. For reviews about bifunctional catalysis, see: (a) Shibasaki,
M.; Kanai, M.; Funabashi, K. Chem. Commun. 2002,
1189–1199. (b) Matsunaga, S.; Ohshima, T.; Shibasaki, M.
¨
Adv. Synth. Catal. 2002, 344, 3–15. (c) Groger, H. Chem. Eur.
J. 2001, 7, 5246–5251. (d) Rowlands, G. J. Tetrahedron 2001,
57, 1865–1882.
22. van Langen, L. M.; van Rantwijk, F.; Sheldon, R. A. Org.
Process Res. Devel. 2003, 7, 828–831.
23. Duquette, J.; Zhang, M.; Zhu, L.; Reeves, R. S. Org. Process
Res. Devel. 2003, 7, 285–288.
5. (a) Shibasaki, M.; Kanai, M. Chem. Pharm. Bull. 2001, 49,
511–524. (b) Shibasaki, M.; Kanai, M.; Funabashi, K. Chem.
Pharm. Bull. 2001, 49, 511–524.
24. Johnson, D. V.; Felfer, U.; Griengl, H. Tetrahedron 2000, 56,
781–790.
6. (a) Hamashima, Y.; Sawada, D.; Kanai, M.; Shibasaki, M.
J. Am. Chem. Soc. 1999, 121, 2641–2642. (b) Hamashima, Y.;
Sawada, D.; Nogami, H.; Kanai, M.; Shibasaki, M.
Tetrahedron 2001, 57, 805–814.
25. Johnson, D. V.; Griengl, H. Tetrahedron 1997, 53, 617–624.
26. (a) Nicolaou, K. C.; Hepworth, D.; King, N. P.; Finlay,
M. R. V.; Scarpelli, R.; Pereira, M. M. A.; Bollbuck, B.; Bigot,
A.; Werschkun, B.; Winssinger, N. Chem. Eur. J. 2000, 6,
2783–2800. (b) Bode, J. W.; Carreira, E. M. J. Org. Chem.
2001, 66, 6410–6424.
7. Takamura, M.; Hamashima, Y.; Usuda, H.; Kanai, M.;
Shibasaki, M. Angew. Chem., Int. Ed. 2000, 39, 1650–1652.
8. Takamura, M.; Funabashi, K.; Kanai, M.; Shibasaki, M. J. Am.
Chem. Soc. 2000, 122, 6327–6328.
´
9. For a preliminary communication, see: Casas, J.; Najera, C.;
´
Sansano, J. M.; Saa, J. M. Org. Lett. 2002, 4, 2589–2592.
´ ´
10. Casas, J.; Baeza, A.; Sansano, J. M.; Najera, C.; Saa, J. M.
Tetrahedron: Asymmetry 2003, 14, 197–200.
´ ´
11. Baeza, A.; Casas, J.; Najera, C.; Sansano, J. M.; Saa, J. M.
Angew. Chem., Int. Ed. 2003, 42, 3143–3146.
27. (a) Sawada, D.; Shibasaki, M. Angew. Chem., Int. Ed. 2000,
39, 209–213. (b) Sawada, D.; Kanai, M.; Shibasaki, M. J. Am.
Chem. Soc. 2000, 43, 10521–10532.
28. For some recent reviews of NLE see: (a) Kagan, H. B. Adv.
Synth. Catal. 2001, 343, 227–233. (b) Kagan, H. B. Synlett
2001, 888–899. (c) Soai, K.; Shibata, T.; Sato, I. Acc. Chem.
Res. 2000, 33, 382–390. (d) Heller, D.; Drexler, H.-J.; Fischer,
C.; Buschmann, H.; Baumann, W.; Heller, B. Angew. Chem.,
Int. Ed. 2000, 39, 495–499.
12. The diethylamino substituted (S)- and (R)-BINOLAM 2 were
used as phase-transfer catalysts for the enantioselective
alkylation of the aldimine-Schiff bases of alanine esters:
29. Mason, M. R.; Smidth, J. M.; Bott, S. G.; Barron, A. R. J. Am.
Chem. Soc. 1993, 115, 4971–4984.
´
´
´
Casas, J.; Najera, C.; Sansano, J. M.; Gonzalez, J.; Saa, J. M.;
Vega, M. Tetrahedron: Asymmetry 2001, 12, 699–702. And
for the addition of diethylzinc to aldehydes see: Kitajima, H.;
Iti, K.; Katsuki, T. Chem. Lett. 1996, 343–344.
30. (a) Zhong, Z.; Dijkstra, P. J.; Feijen, J. Angew. Chem., Int. Ed.
2002, 41, 4511–4513. (b) Atwood, D. A.; Harvey, M. J. Chem.
Rev. 2001, 101, 37–52.
31. Corey, E. J.; Lee, T. W. Chem. Commun. 2001, 1321–1329.
32. Liang, S.; Bu, X. R. J. Org. Chem. 2002, 67, 2702–2704.
33. Hayashi, M.; Miyamoto, Y.; Inoue, T.; Oguni, N. J. Org.
Chem. 1993, 58, 1515–1522.
13. The influence of Ph3PO to prevent oligomerisation of the
catalyst and to activate trimethylsilyl cyanide has been pointed
¨
out byVogl, E. M.; Groger, H.; Shibasaki, M. Angew. Chem.,
Int. Ed. 1999, 38, 1570–1577.
14. The role of molecular sieves 4 A as H2O donor has been
34. Matthews, B. R.; Jackson, W. R.; Jayatilake, G. S.; Wilshire,
C.; Jacobs, H. A. Aust. J. Chem. 1988, 41, 1697–1709.
35. Brown, R. F. C.; Donohue, A. C.; Jackson, W. R.; McCarthy,
T. D. Tetrahedron 1994, 50, 13739–13752.
˚
demonstrated for the case of binaphthol-derived titanium
complexes: Terada, M.; Matsumoto, Y.; Nakamura, Y.;
Mikami, K. Chem. Commun. 1997, 281–282.
36. Tanaka, K.; Mori, A.; Inoue, S. J. Org. Chem. 1990, 55,
181–185.
¨
15. Vogl, M.; Groger, H.; Shibasaki, M. Angew. Chem., Int. Ed.
Engl. 1999, 38, 1570–1577.
37. Chen, P.; Han, S.; Lin, G.; Huang, H.; Li, Z. Tetrahedron:
Asymmetry 2001, 12, 3273–3279.
16. Shimizu, M.; Ogawa, T.; Nishi, T. Tetrahedron Lett. 2001, 42,
5463–5466.
38. Ziegler, T.; Horsch, B.; Effenberger, F. Synthesis 1990,
575–578.
17. Ling, S.-K.; Tanaka, T.; Kouno, I. J. Nat. Prod. 2002, 65,
131–135.
39. Ooi, T.; Miura, T.; Takaya, K.; Ichikawa, H.; Maruoka, K.
Tetrahedron 2001, 57, 867–873.
18. Cho, B. T.; Kang, S. K.; Shin, S. H. Tetrahedron: Asymmetry
2002, 13, 1209–1217.