Ni(II)-catalyzed enantioselective Nazarov cyclizations

Article abstract of DOI:10.1039/b806870d

Nazarov cyclizations are catalyzed by a dicationic Ni(ii) complex containing the chiral tridentate phosphine Pigiphos; the catalyst exerts a high degree of torquoselectivity and affords the products in up to 88% ee. The Royal Society of Chemistry.

Full text of DOI:10.1039/b806870d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Ni(II)-catalyzed enantioselective Nazarov cyclizationsw
Irene Walz and Antonio Togni*
Received (in Cambridge, UK) 23rd April 2008, Accepted 3rd June 2008
First published as an Advance Article on the web 17th July 2008
DOI: 10.1039/b806870d
Nazarov cyclizations are catalyzed by a dicationic Ni(II) com-
plex containing the chiral tridentate phosphine Pigiphos; the
catalyst exerts a high degree of torquoselectivity and affords the
products in up to 88% ee.
The Nazarov reaction involves the cyclization of divinyl ketones to
cyclopentenones under the influence of strong acids (Scheme 1).1
Its potential for the synthesis of cyclopentane building blocks is
therefore quite vast.2 Hence, it is hardly surprising that since the
work of Denmark and Jones3 who demonstrated that this reaction
can proceed with high chemoselectivity, the number of studies
concerning Nazarov cyclizations has grown significantly.4
Scheme 2 Pigiphos ligands 2 and their Ni(II) complexes 3.
Furthermore, the Nazarov cyclization is one of the few
electrocyclic transformations that can be promoted in an
enantioselective fashion by chiral catalysts. Problems such as
the torquoselectivity of the cyclization step, as well as the
regioselectivity of the deprotonation–reprotonation sequence,
have attracted considerable interest and led to a number of
specific solutions.5 A remarkable example consists of catalyti-
cally controlling the ring-closing step with a chiral binol
phosphate, thereby generating highly enantiomerically en-
riched cyclopentenones.6 However, this process is limited to
highly reactive dihydropyrane substrates. Despite recent pro-
gress, the genuine control of torquoselectivity of the cycliza-
tion step in truly catalytic Nazarov reactions remains very
rare.7 Several early and late transition-metal salts and com-
plexes bearing a variety of ligands have been shown to be
selective.8 For example, we recently reported that dicationic
V(IV) complexes bearing salen-type ligands are highly active
catalysts for the cyclization of divinyl ketones bearing an ester
group adjacent to the carbonyl unit (1, Scheme 1).8a
a,b-unsaturated nitriles in hydroamination10 and hydrophosphi-
nation11 reactions, we decided to embark on a study addressing
their activity and selectivity in Nazarov cyclizations of substrates
of type 1. These compounds are tetra substituted divinyl ketones
and at the same time unsaturated b-keto esters. They have been
prepared by standard synthetic procedures and were partly
reported before.8a We have chosen these substrates for the
present study because the cyclopentenones generated upon the
corresponding Nazarov cyclizations contain two new contiguous
stereogenic centers, but are formed as single diastereoisomers.
Since the stereogenic center at position 4 of the cyclopentenone
ring is installed in the course of the electrocyclic step, the
enantioselectivity of the reaction corresponds to the degree of
torquoselectivity exerted by the catalyst. The dicationic Ni(II)–tri-
phosphine complexes 3 can be easily prepared in situ by mixing
Pigiphos ligands 2 with [Ni(H2O)6][ClO4]2 in dry THF.
We first examined the influence of various solvents as well as
temperature on the enantioselectivity of the cyclization of divinyl
ketone 1a catalyzed by complex 3a. The results shown in Table 1
clearly demonstrate that dichloromethane is the best solvent in
view of achieving good selectivities. We were pleased to find that
the Nazarov product 4a could be obtained with 86% enantio-
selectivity when working at room temperature though with a
relatively high catalyst loading of 10 mol% and a 1 : 2 nickel :
ligand ratio. However, a temperature increase reduced the
selectivity of the cyclization reaction (compare entry 1 with 2
and 3 in Table 1). THF was found to be as similarly suited a
solvent as CH2Cl2 for the cyclization of dialkenyl ketone 1a
(82% ee). However, only 31% of the product could be isolated
after a reaction time of ten days. Similar considerations apply to
toluene (44% yield after ten days and 76% ee). The strongly
coordinating solvent acetonitrile hampered the reaction comple-
tely and no conversion could be observed.
Given the demonstrated Lewis acidic properties of dicationic
9
Ni(II)–Pigiphosz complexes 3 (Scheme 2) for the activation of
Scheme 1 General formulation of the Nazarov cyclization and sub-
strate type used in this study.
Department of Chemistry and Applied Biosciences, Swiss Federal
Institute of Technology, ETH Zurich, CH-8093 Zurich, Switzerland.
E-mail: atogni@ethz.ch; Fax: +41 44 632 1310;
Tel: +41 44 632 2236
w Electronic supplementary information (ESI) available: Spectro-
scopic data for selected new compounds. CCDC reference number
682373 for compound 5. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/b806870d
Further experiments were carried out with the two modified
Pigiphos ligands 2b–c under the above best conditions. The
3,5-dimethylphenyl (2b) and 2-naphthyl substituted (2c)
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 4315–4317 | 4315

View Article Online
Table 1 Influence of solvents and reaction temperature on the
selectivity
reaction. A very pronounced drop in both reactivity and
selectivity resulted in going from ethyl and propyl to benzyl
and 1-naphthyl ester, respectively. In the latter case no reac-
tion could be observed even after exposing 1h to the catalyst
for one month.
Thus, a size match between the ester group and the aromatic
substituent R3 is essential in order to maintain reactivity and
to obtain a high enantioselectivity. Since both the ester group
and R3 are connected to the same alkene unit, this might
indicate that they are very influential in determining the ideal
substrate conformation needed for the electrocyclic step.
The absolute configuration of the newly formed stereogenic
centers in product 4a was determined by X-ray crystallo-
graphic analysis of a sample obtained after transesterification
to the corresponding (1R)-menthyl ester 5. Compound 4a itself
is accessible in enantiomerically pure form by preparative
chiral HPLC, however, we did not succeed in obtaining
crystals suitable for an X-ray study. As shown in Fig. 1 the
absolute configuration of both centers C4 and C5 of the
cyclopentenone ring is R. This means that for the catalyst
containing (R)-(S)-Pigiphos the conrotatory ring closure of 1a
must have taken place in an anti clockwise manner.
Entry
Solvent
T (1C)
Yield (%)
ee (%)
1
2
3
4
5
6
CH2Cl2
CH2Cl2
Cl(CH2)2Cl
THF
toluene
MeCN
r.t.
40
60
r.t.
r.t.
r.t.
84
79
91
31
86
85
60
82
76
n.a.
44
no reaction
TMP = 2,4,6-trimethoxyphenyl.
Pigiphos afforded the cyclopentenone 4a with only 43% and
62% enantioselectivity, respectively.
Next we examined the influence of the three different
substituents R1, R2 and R3 on the dialkenyl ketone scaffold
1 on the selectivity of the cyclization (Table 2). No significant
effect appears to derive from the nature of R1, both in steric
and electronic terms, at least as illustrated by 4a (R1 = Me)
and 4b (R1 = Ph). Both these products have been obtained at
the same level of enantioselectivity (87% ee). The only differ-
ence between the two substrates 1a and 1b is that the latter
requires a longer reaction time. Similar considerations appear
to apply to R3 when comparing the formation of the tri-
methoxyphenyl (TMP) derivative 4b with that of 4d (83% ee),
bearing a 4-methoxyphenyl (PMP) group instead, which re-
quired a significantly extended reaction time.
The main drawback of the Ni(Pigiphos) catalyst is its low
activity. Even at the 10 mol% level the Nazarov cyclization
takes several days to go to completion. The fact that strongly
coordinating solvents such as acetonitrile completely shut
down the reaction indicates that the thermodynamic stability
as well as the kinetic lability of [Ni(Pigiphos)L]2+ complexes
are key to obtain active catalysts. Given that both compounds
1 and 4 are likely to coordinate to the Ni(Pigiphos) fragment
via the oxygen atom of the ketone unit, one can reasonably
expect a possible product inhibition of the catalytic reaction.
Indeed, the major Ni(Pigiphos) species that can be observed by
31P-NMR spectroscopy during the whole course of the cata-
lytic reaction involving substrate 1a corresponds to the adduct
obtained when mixing equivalent amounts of complex 3a and
However, a significant drop in both enantioselectivity and
yield was found for the combination R1 = Me and R3
=
PMP. The Nazarov product 4c could be isolated in only 32%
yield and 71% ee. The ester group, as illustrated by the series
1a–1e–1g–1h having otherwise the same R1 and R3 substitu-
ents, displays a more drastic influence on the outcome of the
Table 2 Ni-catalyzed Nazarov cyclizations of various dialkenyl
ketonesay
Compound
R1
R2
R3
Yield (%)
ee (%)
4a
4b
4c
4d
4e
4f
4g
4h
a
Me
Ph
Me
Ph
Me
Ph
Me
Me
Et
Et
Et
Et
Pr
Pr
Bn
Npd
TMPb
TMPb
PMPc
PMPc
TMPb
TMPb
TMPb
TMPb
84
85
32
96
80
82
86
87
71
83
82
88
45
n.a.
58
no reaction
Reaction times for full conversion are 6–8 d for substrates having R3 =
TMP and 9–15 d for R3 = PMP. b TMP = 2,4,6-trimethoxyphenyl.
Fig. 1 Molecular structure and absolute configuration of menthyl
ester 5.
c
d
PMP = 4-methoxyphenyl. Np = 1-naphthyl.
ꢀc
This journal is The Royal Society of Chemistry 2008
4316 | Chem. Commun., 2008, 4315–4317

View Article Online
therefore currently pursuing more active modifications of this
nickel catalyst in order to extend the scope of its applications.
The Swiss National Science Foundation supported this
work (PhD grant to I. W.). We thank Pietro Butti and
Francesco Camponovo for the X-ray crystallographic mea-
surement of compound 5.
Scheme 3 Ni-catalyzed tandem Nazarov cyclization–Michael addition.
Notes and references
product 4a. Thus, it appears that the Nazarov product forms a
relatively stable complex with the catalyst and that the con-
centration of the truly catalytically active species must be low.
Despite this observation and because of the ability of the
dicationic Ni(Pigiphos) unit to activate a,b-unsaturated ni-
triles towards nucleophilic attack,10,11 we envisaged a further
functionalization of products 4 as nucleophiles in a Michael-
type addition to a cyanoolefin. The concept of this tandem
process is shown in Scheme 3.
z Pigiphos: bis{(R)-1-[(S)-2-(diphenylphosphanyl)ferrocenyl]ethyl}-
cyclohexylphosphane.
y General procedure for the nickel(II)-catalyzed Nazarov cyclization
(and Michael addition): [Ni(H2O)6][ClO4]2 (5 mmol, 0.1 equiv.) and
Pigiphos 2 (0.01 mmol, 0.2 equiv.) were dissolved in dry THF (1 mL)
under argon. After stirring for 16 h at room temperature, the solvent
was evaporated to dryness and the residue was dried under vacuum for
3 h. Afterwards the Nazarov substrate 1 (0.05 mmol, 1 equiv.) and
CH2Cl2 (1 mL) were added. After the cyclization completed, the
solvent was evaporated and the residue was purified by column
chromatography. For the Michael addition, the solvent was evapo-
rated after the Nazarov cyclization completed, as monitored by TLC.
Acrylonitrile (1 mL) and DBU (5 mmol, 0.1 equiv.) were then added
and the mixture was stirred for 3 h, after which time the reaction was
complete. After evaporation of the solvent the residue was purified by
column chromatography. The enantioselectivities were determined by
chiral HPLC (Dr Maisch, ReproSil Chiral DP). All compounds of
type 4 bearing a TMP substituent show hindered rotation about the
C(4)–Cipso bond at r.t., as shown by NMR spectroscopy (see ESIw).
z Crystal data for compound 5: C32H40O6, M = 520.64, orthorhom-
bic, space group P212121 (no. 19), a = 10.0389(5), b = 15.9508(7), c =
18.4174(8) A, V = 2949.2(2) A3, Dc = 1.173 g cmꢁ3, T = 200(2) K, Z
= 4, 120 948 reflections measured, 14 329 unique (Rint = 0.0493), R1
= 0.0440 [I 4 2s(I)], wR2 = 0.1089. Some disorder present in the
molecule has not been modelled. CCDC reference number 682373. For
crystallographic data in CIF or other electronic format see DOI:
10.1039/b806870d
The completion of the Nazarov reaction of substrates 3 may
be conveniently monitored by TLC. This was followed by the
evaporation of the solvent, addition of an excess of acrylonitrile,
both as a solvent and reagent, as well as a co-catalytic amount of
the base DBU (10 mol%).12 We were pleased to find that under
these conditions a smooth and complete conversion of the
Nazarov cyclization product 4 to the new, highly substituted
cyclopentenones 6 took place at room temperature within ca. 3 h.
Specifically, this procedure has been applied to the three selected
dialkenyl ketones 3a, 3c and 3f and in all cases the addition
product was isolated in good to almost quantitative yield (see
Table 3). The new compounds 6a, 6c and 6f display the same
level of enantioenrichment as their corresponding precursors 4.
Although we did not determine the absolute configuration of the
new products 6, it is reasonable to assume that the generation of
the new quaternary stereogenic center occurs with inversion of
configuration at position 5 of the cyclopentenone ring.
1 I. N. Nazarov and I. I. Zaretskaya, Izv. Akad. Nauk Kaz. SSR, Ser.
Khim., 1941, 211–224.
2 A. J. Frontier and C. Collison, Tetrahedron, 2005, 61, 7577–7606.
3 S. E. Denmark and T. K. Jones, J. Am. Chem. Soc., 1982, 104,
2642–2645.
The results obtained with our nickel catalyst are very encoura-
ging because they concern Nazarov substrates that have been so
far investigated only with stoichiometric amounts of a chiral
Lewis acid.5a Furthermore, the Ni(Pigiphos) system is also able
to catalyze the addition of the Nazarov products to acrylonitrile,
thus representing a bonus with respect to other catalysts. We are
4 For reviews, see: (a) H. Pellissier, Tetrahedron, 2005, 61,
6479–6517; (b) M. A. Tius, Eur. J. Org. Chem., 2005, 2193–2206.
5 (a) V. K. Aggarwal and A. J. Beffield, Org. Lett., 2003, 5,
5075–5078; (b) G. X. Liang and D. Trauner, J. Am. Chem. Soc.,
2004, 126, 9544–9545.
6 M. Rueping, W. Ieawsuwan, A. P. Antonchick and B. J.
Nachtsheim, Angew. Chem., Int. Ed., 2007, 46, 2097–2100.
7 (a) A. Cavalli, M. Masetti, M. Recanatini, C. Prandi, A. Guarna
and E. G. Occhiato, Chem.–Eur. J., 2006, 12, 2836–2845; (b) C.
Prandi, A. Ferrali, A. Guarna, P. Venturello and E. G. Occhiato,
J. Org. Chem., 2004, 69, 7705–7709; (c) D. A. Smith and C. W.
Ulmer II, J. Org. Chem., 1993, 58, 4118–4121; (d) S. E. Denmark,
M. A. Wallace and C. B. Walker, Jr, J. Org. Chem., 1990, 55,
5543–5545.
Table 3 Ni-catalyzed tandem Nazarov cyclization–Michael addition
8 (a) I. Walz, A. Bertogg and A. Togni, Eur. J. Org. Chem., 2007,
2650–2658; (b) M. Janka, W. He, A. J. Frontier and R. Eisenberg,
J. Am. Chem. Soc., 2004, 126, 6864–6865; (c) W. He, X. F. Sun and
A. J. Frontier, J. Am. Chem. Soc., 2003, 125, 14278–14279; (d) L.
Zhang and S. Wang, J. Am. Chem. Soc., 2006, 128, 1442–1443;
(e) L. Bartali, P. Larini, A. Guarna and E. G. Occhiato, Synthesis,
2007, 1733–1737; (f) J. Nie, H.-W. Zhu, H.-F. Cui, M.-Q. Hus and
J.-A. Ma, Org. Lett., 2007, 9, 3053–3056.
9 P. Barbaro and A. Togni, Organometallics, 1995, 14, 3570–3573.
10 L. Fadini and A. Togni, Chem. Commun., 2003, 30–31.
11 A. D. Sadow and A. Togni, J. Am. Chem. Soc., 2005, 127,
17012–17024.
Yield
(%)
ee
(%)
Compound
R1
R2
Et
R3
6a
Me
TMPa
78
63
97
87
(86)c
72
6c
6f
Me
Ph
Et
Pr
PMPb
TMPa
(71)c
88
(88)c
12 This type of procedure has been used before in our laboratory for
similar Ni(Pigiphos)-catalyzed additions of simple, achiral 1,3-dike-
tones to unsaturated nitriles: Y. Liu and A. Togni, to be published. See
a
b
2,4,6-trimethoxyphenyl. PMP
TMP
Value in brackets is for the Nazarov cyclization.
=
=
4-methoxyphenyl.
c
also: Y. Liu, PhD Thesis Nr. 17006, ETH Zurich, 2007.
¨
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 4315–4317 | 4317