New stable catalysts of the Pauson-Khand annelation

Article abstract of DOI:10.1016/S0040-4039(00)02204-8

A range of phosphine- and phosphite-substituted carbonylcobalt(0) complexes have been synthesised and screened as catalysts of the Pauson-Khand annelation. The readily prepared stable solid, heptacarbonyl(triphenylphosphine)dicobalt(0) 1, performed well (

Full text of DOI:10.1016/S0040-4039(00)02204-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 1183–1185
New stable catalysts of the Pauson–Khand annelation
a
a,
a
b
Alex C. Comely, Susan E. Gibson, * Andrea Stevenazzi and Neil J. Hales
a
Department of Chemistry, King’s College London, Strand, London WC2R 2LS, UK
b
AstraZeneca UK Ltd, Mereside, Alderley Park, Macclesfield SK10 4TG, UK
Received 5 October 2000; accepted 29 November 2000
Abstract—A range of phosphine- and phosphite-substituted carbonylcobalt(0) complexes have been synthesised and screened as
catalysts of the Pauson–Khand annelation. The readily prepared stable solid, heptacarbonyl(triphenylphosphine)dicobalt(0) 1,
performed well (78% yield) in a benchmark annelation reaction. © 2001 Elsevier Science Ltd. All rights reserved.
Some of the most significant contributions to organic
synthesis during the past years have been made by
transition metal-mediated reactions. A prime example is
plexes as catalyst precursors. Livinghouse makes use of
the hexacarbonyldicobalt(0) complex of 2-methyl-3-
butyn-2-ol which is reduced in situ to an active source
6
the Pauson–Khand annelation (PKA), first described in
of a carbonyldicobalt(0) with triethylsilane. Krafft uses
1
1
971, which involves the synthesis of cyclopentenone
either the hexacarbonyldicobalt(0) complex of the sub-
strate as the catalyst precursor, or a complex of a polar
enyne, the cyclopentenone product from which is read-
ily removed from the reaction mixture by acid/base
derivatives via the carbonylcobalt(0)-mediated cyclisa-
tion of an alkyne, an alkene and carbon monoxide.
During its development and application, the PKA has
mainly been applied using a stoichiometric amount of
the transition metal complex, usually by prior forma-
tion of a stable (alkyne)hexacarbonyldicobalt(0) com-
7
wash or silica plug filtration.
As a result of our interest in the use of carbonyl-
cobalt(0) complexes immobilised on ‘polymer-bound
plex. Although the catalytic Pauson–Khand annelation
2
8
(
CPKA) was reported as early as 1973, it was confined
triphenylphosphine’ in the CPKA, we initiated an ‘off-
to the strained reactive alkenes norbornene and norbor-
nadiene and required an excess of the alkyne compo-
nent. The 1990s saw a surge of interest in the CPKA
and the current state-of-the-art can be attributed to
Livinghouse who in 1996 reported a photochemically
polymer’ study of phosphorus derivatives of carbonyl-
cobalt(0) species. Despite (a) the significant use of
phosphine and phosphite substituted alkyne complexes
in the stoichiometric PKA, where co-ordination leads
to a reduction in the rate and overall efficiency of the
9
driven process requiring only mild temperatures (50–
reaction, and (b) the observation that the use of
3
5
5°C) and one atmosphere of carbon monoxide. A
triphenylphosphite as an additive in the Co (CO)
2 8
more recent report from the same group relates that
careful control of temperature to within a narrow win-
mediated CPKA leads to an improvement in reaction
10
efficiency, the performance of preformed complexes of
phosphines and phosphites in the CPKA has not been
assessed. We report here that such complexes do
catalyse the CPKA and that some of the systems stud-
ied provide an attractive practical alternative to the
catalysts currently available.
dow (60–70°C) dispenses with the need for photolytic
4
promotion. Rigorous purification of Co (CO)8 is
2
essential for the success of these systems, however, and
Livinghouse recommends recrystallisation or sublima-
tion immediately prior to use or opening a fresh com-
mercial sample in a glove-box. Such cumbersome
techniques and purification steps can be obviated,
reports Krafft, by prior base washing of the glassware
A representative range of readily available phosphines
and phosphites was selected for conversion into car-
bonylcobalt(0) complexes and testing as PKA catalyst
precursors. Synthesis of the mono- and bis-phosphine
complexes 1–9 was straightforward and generally
achieved by reacting one or two equivalents respectively
of the appropriate phosphine or phosphite with
Co (CO) . Work-up gave the required complexes in
5
and introduction of cyclohexylamine. Persistent prob-
lems associated with the very labile Co (CO)8 have
2
spurred the development of stable cobalt–alkyne com-
*
Corresponding author (n e´ e Thomas); e-mail: susan.gibson@
2
8
kcl.ac.uk
satisfactory yield (58–96%).
0
040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)02204-8

1
184
A. C. Comely et al. / Tetrahedron Letters 42 (2001) 1183–1185
The mono- and bis-phosphine complexes 1–9 were
tested on one of the benchmark PKA reactions, i.e.
the conversion of enyne 10 to cyclopentenone 11. The
conditions selected (5 mol% of catalyst, 70°C, 24 h,
THF, 1.05 atm. CO) were those found to be the most
successful using ‘polymer-bound triphenylphosphine’.
Under these conditions, small amounts of a byproduct
derivatives of carbonylcobalt(0) complexes can indeed
act as catalysts of the CPKA. The bisphosphine/bis-
phosphite complexes give poorer conversions to 11
than their monophosphine/monophosphite analogues,
and this is tentatively attributed to their poorer solu-
bility. The monophosphite complexes, 1, 3, 5 and 7,
however, all give very good conversions of 10 to 11.
Work-up of the product mixture obtained from the
reaction performed with complex 1 [column chro-
8
1
2 are sometimes produced. After the reaction, the
THF was removed from the product mixture and the
1
resulting solid examined by H NMR spectroscopy.
matography (SiO ; hexane–ethyl acetate, 6:4 to 1:1
2
The results of the screening are given in Table 1.
These demonstrate that phosphine and phosphite
gradient elution] gave a crystalline sample of cyclopen-
tenone 11 in 78% yield.
a
b
Table 1. Catalysts and CPKA conversions (11:12:10)
catalyst
5 mol%)
(
TsN
TsN
O
+
12
o
7
0 C, 24 h, THF
1.05 atm CO
10
11
PꢀCo complex
R
CO OC CO
Co Co CO
OC CO CO
CO OC CO
Co Co P
R
R
R
R
P
P
R
R
R
OC CO CO
R
PR3
1
1
12
1
2
84:6:10
61:8:31
P
3
1
1
13
3
4
91:6:3
2:0:98
P
OMe
3
1
1
12
5
6
P
78:0:22
14:0:86
3
1
1
12
P O
7
8
84:3:13
58:0:42
3
1
4
Cobalt complex
not isolable
9
P
0:0:100
SO Na
3
3
a
References for synthesis of catalysts provided.
b
1
1
1:12:10 as determined by H NMR spectroscopy.

A. C. Comely et al. / Tetrahedron Letters 42 (2001) 1183–1185
1185
In conclusion we have demonstrated that a range of
phosphorus substituted carbonylcobalt(0) complexes
catalyse the PKA. The isolated yield of 11 obtained using
J. Chem. Soc., Chem. Commun. 1971, 36.
2. Khand, I. U.; Knox, G. R.; Pauson, P. L.; Watts, W.
E.; Foreman, M. I. J. Chem. Soc., Perkin Trans. 1
1973, 977.
1
(78% using 5 mol% 1, 70°C, 24 h, THF, 1.05 atm. CO)
is comparable with the yields obtained by Livinghouse
and by Krafft for the same cyclisation under simi-
3
4
5
6
. Pagenkopf, B. L.; Livinghouse, T. J. Am. Chem. Soc.
1996, 118, 2285.
lar conditions (86% using 7.5 mol% of rigorously
. Belanger, D. B.; O’Mahony, D. J. R.; Livinghouse, T.
Tetrahedron Lett. 1998, 39, 7637.
. Krafft, M. E.; Bonaga, L. V. R.; Hirosawa, C. Tetra-
hedron Lett. 1999, 40, 9171.
4
purified Co (CO) , 60°C, 12 h, DME, 1 atm. CO; 89%
2
8
using 10 mol% Co (CO)8 and 20 mol% cyclohexyl-
2
amine in base washed glassware, 70°C, 14 h, DME, 1
5
atm. CO ). Complexes 1, 3, 5 and 7 are readily available
. Belanger, D. B.; Livinghouse, T. Tetrahedron Lett.
and stable complexes. For example, complex 1 is a stable
1998, 39, 7641.
solid, easily prepared by stirring triphenylphosphine with
11
7. Krafft, M. E.; Hirosawa, C.; Bonaga, L. V. R. Tetra-
hedron Lett. 1999, 40, 9177.
. Comely, A. C.; Gibson (n e´ e Thomas), S. E.; Hales, N.
J. Chem. Commun. 2000, 305.
. Billington, D. C.; Helps, I. M.; Pauson, P. L.; Thom-
son, W.; Willison, D. J. Organomet. Chem. 1988, 354,
Co (CO) at room temperature for 1 h (65% yield). It
2
8
is considerably more stable than Co (CO) , which not
2
8
8
only makes it easier to handle and measure, but also
renders its reactions much more reliable.
9
2
33.
Acknowledgements
1
0. Jeong, N.; Hwang, S. H.; Lee, Y.; Chung, Y. K. J.
Am. Chem. Soc. 1994, 116, 3159.
The authors gratefully acknowledge AstraZeneca UK
Ltd for studentships (for A.C.C. and A.S.).
11. Szabo, P.; Fekete, L.; Bor, G.; Nagy-Magos, Z.;
Marko, L. J. Organomet. Chem. 1968, 12, 245.
1
1
2. Manning, A. R. J. Chem. Soc. (A) 1968, 1135.
3. Farrar, D. H.; Hao, J.; Poe, A. J.; Stromnova, T. A.
Organometallics 1997, 16, 2827.
References
14. Bartik, T.; Bartik, B.; Hanson, B. E.; Whitmire, K. H.;
1. Khand, I. U.; Knox, G. R.; Pauson, P. L.; Watts, W. E.
Guo, I. Inorg. Chem. 1993, 32, 5833.
.
.