A study of [Co2(alkyne)(binap)(CO)4] complexes (BINAP = (1,1′-Binaphthalene)-2,2-diylbis(diphenylphosphine))

Article abstract of DOI:10.1002/chem.200401210

Understanding the interaction of chiral ligands, alkynes, and alkenes with cobaltcarbonyl sources is critical to learning more about the mechanism of the catalytic, asymmetric Pauson-Khand reaction. We have successfully characterized complexes of the type

Full text of DOI:10.1002/chem.200401210

A Study of [Co2(alkyne)(binap)(CO)4] Complexes (BINAP=(1,1’-
Binaphthalene)-2,2’-diylbis(diphenylphosphine))
Susan E. Gibson,*[a] Karina A. C. Kaufmann,[a] Jennifer A. Loch,[a]
Jonathan W. Steed,[b] and Andrew J. P. White[a]
Abstract: Understanding the interac-
tion of chiral ligands, alkynes, and al-
kenes with cobaltcarbonyl sources is
critical to learning more about the
mechanism of the catalytic, asymmetric
Pauson–Khand reaction. We have suc-
cessfully characterized complexes of
the type [Co2(alkyne)(binap)(CO)4]
(BINAP=(1,1’-binaphthalene)-2,2’-
diylbis(diphenylphosphine)) and shown
occurs under Pauson–Khand reaction
conditions when
CCO2Me. Attempts to isolate [Co2-
(alkyne)(binap)(CO)x] complexes with
coordinated alkenes led to the forma-
tion of cobaltacyclopentadiene species.
alkyne=HCꢀ
that
diastereomer
interconversion
Keywords: alkynes
·
asymmetric
catalysis · cobalt · cyclopentenone ·
reaction mechanisms
Introduction
have only been reported since the late 1990s. The first
report of a catalytic, asymmetric PKR came from Buch-
wald[3] in 1996. By using 5–20 mol% of complex 1, which is
generated in situ from (S,S)-[Ti-
Methods of simultaneously introducing multiple functional
groups and asymmetry are highly valued in organic synthe-
sis. Ideally, the agent inducing the enantioselectivity is used
in a catalytic quantity. The Pauson–Khand reaction (PKR),
that is, the cyclocarbonylation of an alkyne and an alkene to
form a cyclopentenone, is able to fulfil each of these goals.[1]
Given the molecular complexity that the PKR generates, it
has proven to be an invaluable tool in many total syntheses.
These syntheses range from prostaglandins with a cyclopen-
tenone core to more complex natural products in which the
cyclopentenone ring is further manipulated during the
course of the synthesis.[1a,d,e]
(ebthi)(Me)2] (EBTHI=ethyl-
ene-1,2-bis(h5-4,5,6,7-tetrahy-
dro-1-indenyl) under a CO at-
mosphere, he could obtain
enantiomeric excesses (eeꢀs) as
high as 96% with intramolecu-
lar PKR substrates (Scheme 1).
Since its initial discovery more than 30 years ago,[2] when
stoichiometric amounts of cobalt were employed for cyclo-
pentenone formation, there has been significant progress to-
wards making the PKR either catalytic or asymmetric. Sys-
tems that are both catalytic and enantioselective, however,
Scheme 1. The catalytic, asymmetric PKR employing chiral complex 1.[3]
The bulky EBTHI ligand was, however, incompatible with
some more highly substituted enynes. Buchwald[4] later went
on to study the catalytic, asymmetric PKR of nitrogen-con-
taining enynes, which proved to show large variations in ee
depending on the substituent attached to the nitrogen atom.
The C2 symmetry found in the EBTHI ligand has contin-
ued to play a role in subsequent reports of catalytic, asym-
metric PKRs by Buchwald and others. In fact, to date, reac-
tions giving the highest selectivities for catalytic, enantiose-
lective PKRs all employ axially chiral ligands. Hiroi[5] was
the first to use cobalt in these types of PKRs. He found the
[a] Prof. S. E. Gibson, K. A. C. Kaufmann, Dr. J. A. Loch,
Dr. A. J. P. White
Department of Chemistry, Imperial College London
South Kensington, London SW7 2 AY (UK)
Fax : (+44)207-594-5804
E-mail: s.gibson@imperial.ac.uk
[b] Dr. J. W. Steed
Department of Chemistry, University of Durham
South Road, Durham DH1 3 LE (UK)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
2566
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/chem.200401210
Chem. Eur. J. 2005, 11, 2566 – 2576

FULL PAPER
Scheme 3. Our report of improved conditions for the asymmetric, cobalt-
catalyzed PKR.[8]
Initial reports of rhodium and iridium complexes being
used in asymmetric, catalytic PKRs appeared in 2000.
Jeong[9] found that by using [{RhCl(CO)2}2] in combination
with (S)-BINAP and AgOTf, good enantioselectivities could
be obtained (Scheme 4). Even higher selectivities were ach-
axially chiral ligand (S)-BINAP (2; BINAP=(1,1’-binaph-
thalene)-2,2’-diylbis(diphenylphosphine)) to be superior to
other diphosphines and obtained high enantiomeric excesses
with some standard intramolecular PKR substrates at
20 mol% [Co2(CO)8] loading (Scheme 2). Buchwald[6] has
Scheme 4. The enantioselective, catalytic PKR with rhodium.[9]
ieved by Shibata[10] by using [{Ir(cod)Cl}2] and (S)-tolBINAP
(tolBINAP=2,2’-bis(di-p-tolylphosphine)-1,1’-binaphthyl).
In this same account, Shibata reports the first, and to date
only, example of a catalytic, asymmetric, intermolecular
PKR. Although the cyclocarbonylation of 1-phenyl-1-pro-
pyne and norbornene gave only a 32% yield, the enantiose-
Scheme 2. Hiroiꢀs[5] report of the cobalt-catalyzed, enantioselective PKR.
also reported conditions for the asymmetric, [Co2(CO)8]-cat-
alyzed PKR using the binaphthyl-based phosphite 3. He
could only obtain good enantioselectivities with two intra-
molecular PKR substrates, however. Both Hiroi and Buch-
wald suggest that the bidentate phosphine ligands bridge the
two cobalt centers. A recent report by Consiglio[7] extends
the trend of axially chiral ligands in the PKR. The use of
catalytic amounts of [Co2(CO)8] or CoII salts with a reducing
agent and ligand 4 generates intramolecular PKR products
with good enantiomeric excesses. Our own report[8] in which
Scheme 5. The only example of an intermolecular, catalytic, asymmetric
PKR.[10]
lectivity was high (93%; (Scheme 5). Both Jeong and Shiba-
ta tested a number of other chiral diphosphines, but found
the binaphthyl ligands to be the best. Chung[11] studied the
effectiveness of catalytic [{Rh(cod)Cl}2] and (S)-BINAP in a
water/dioxane solvent system. Use of a surfactant, sodium
dodecylsulfate, leads to acceleration of the PKR.
Recently, methods for the catalytic, enantioselective PKR
that depend on aldehydes as a CO source rather than toxic
CO gas have been developed. By using [{Rh(cod)Cl}2] and
(S)-tolBINAP, Shibata[12] found moderate to good enantiose-
lectivity with standard intramolecular PKR substrates
(Scheme 6). Interestingly, he was able to achieve better re-
we used [Co4(CO)12], an easier to handle and more reliable
substitute for [Co2(CO)8], has recently shown that catalytic,
enantioselective PKRs can be carried out with lower load-
ings of catalyst and ligand than those reported by Hiroi to
give good to excellent yields and enantiomeric excesses
(Scheme 3). Again, high selectivities were only achieved
with C2-symmetric diphosphines, ligands 2, 5, and 6.
Scheme 6. The enantioselective, rhodium-catalyzed PKR with cinnamal-
dehyde as the CO source.[12]
Chem. Eur. J. 2005, 11, 2566 – 2576
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2567

S. E. Gibson et al.
sults with cinnamaldehyde as the CO source under solvent-
free conditions than with CO gas and solvent. Finally,
Jeong[13] has recently reported that the catalytic, asymmetric
PKR can be used for the desymmetrization of meso-dien-
ynes (Scheme 7). Rhodium- and iridium-based catalysts
Scheme 7. Desymmetrization of meso-dienynes by asymmetric Pauson–
Figure 1. The molecular structure of 7.
Khand catalysis.[13]
were used in combination with BINAP and BINAP deriva-
tives under 1 atm of CO or with cinnamaldehyde as the CO
source. Different catalyst systems give the best enantioselec-
tivities and diastereoselectivities for different dienyne sub-
strates.
Given the importance of developing new PKR catalysts
and the success of axially chiral ligands in this chemistry, we
initiated an investigation into how C2-symmetric diphos-
phines direct and sustain the catalytic, enantioselective
PKR. We chose to start by examining the coordination of
BINAP to different cobalt carbonyl sources, cobalt carbon-
yls being employed in a vast majority of PKRs. We have
gone on to synthesize a number of BINAP–cobalt–carbonyl
complexes with coordinated alkynes, alkenes, and enynes to
probe subsequent steps of the PKR mechanism. Some of the
results described here have been presented in a preliminary
communication.[8]
43.1 ppm. Application of these same reaction conditions to
[Co4(CO)12] and (ꢁ)-BINAP gave a complex also exhibiting
this same 31P NMR peak at 43.1 ppm. Bands in the CO
region of the IR spectra of the products formed on mixing
either [Co2(CO)8] or [Co4(CO)12] with (ꢁ)-BINAP were
identical.
Complex 7, along with the structures that will be dis-
cussed later in this paper, is the only example of a molecular
structure of a Co–BINAP complex with a chelating BINAP
ligand. The one literature example of a Co–BINAP complex
with a bridging BINAP will be discussed later in this section.
The asymmetry induced by the BINAP ligand on the struc-
ture of 7 is evident upon comparison with other analogous
[Co2(diphosphine)(CO)6] structures (see Table S1 in the
Supporting Information): the difference in lengths between
[14]
ꢂ
the two Co P bonds is greater than in [Co2(dppe)(CO)6]
and [Co2(dppp)(CO)6] (DPPE=1,2-bis(diphenylphosphino)-
ethane,
DPPP=1,3-bis(diphenylphosphino)propane).[15]
There is also distortion in the binding of the two bridging
CO ligands. The difference in lengths between the longest
Results and Discussion
ꢂ
Co CObridging bond length and the shortest is 0.53 ꢂ, com-
Axially chiral diphosphines coordinate to both [Co2(CO)8]
and [Co4(CO)12] to form enantioselective PKR catalysts in
situ,[5–8] and so it was of interest to us to establish the nature
of their interactions. Experimentation with various reaction
conditions eventually allowed us to isolate the complex
pared with 0.29 and 0.16 ꢂ for the two previously men-
tioned [Co2(diphosphine)(CO)6] examples, respectively. The
trans influence of BINAP phosphorus atom, P(1), can also
ꢂ
be seen in the fact that the Co(1) C(3) bond is longer than
ꢂ
ꢂ
those of Co(1) C(1) or Co(1) C(2). Given that there are
few examples of crystallographically characterized transi-
tion-metal–BINAP complexes and that excellent PKR activ-
ity has been observed with rhodium complexes as described
in the introduction, it is interesting to note that as a result
of his interest in BINAP oxidation by dioxygen, Poꢃ[16] has
formed
on
mixing
[Co2(CO)8]
and
(ꢁ)-BINAP
(Scheme 8).[8] Complex 7 is very air sensitive and must be
carefully handled under an inert atmosphere. A small
number of dark red crystals of 7 were obtained and X-ray
analysis revealed that BINAP is bound to just one of the
two cobalt atoms (Figure 1). Phosphorus NMR analysis of
recently reported
a
structurally characterized [Rh-
crystals of complex
7
revealed
a
single resonance at
(binap)(CO)(Cl)] complex, in which the BINAP occupies cis
coordination sites in the square-planar complex.
We next wished to probe whether 7 plays a role in a typi-
cal catalytic, asymmetric PKR.[8] The enantioselective, cata-
lytic PKR shown in Scheme 9 was repeated, but just prior to
adding the intramolecular PKR substrate, a sample of the
solution formed from mixing [Co4(CO)12] and (S)-BINAP
was removed. Removal of solvent from this sample and
analysis by 31P NMR spectroscopy in [D8]THF gave a spec-
trum containing a peak at 43.1 ppm (alongside peaks corre-
Scheme 8. Synthesis of [Co2(binap)(CO)6].
2568
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2566 – 2576

Pauson–Khand Reaction
FULL PAPER
Scheme 9. A typical Pauson–Khand reaction (top) and the same reaction conducted with sampling of the precatalyst solution (bottom).
Table 1. Conversion of alkynes 8a–f into cobalt–BINAP complexes 9a–f.
sponding to (S)-BINAP and (S)-BINAP monoxide), identi-
cal to the peak obtained from crystals of 7. Addition of the
enyne substrate to the remaining precatalyst solution and
running the reaction with our standard PKR conditions gave
a yield of 45% and an ee of 88% (compared to 55% yield
and 88% ee when no sampling is conducted). This strongly
suggests that 7 is a PKR precatalyst.
Having established the precatalytic nature of 7, we started
to investigate how 7 interacts with alkynes to begin to un-
derstand the early part of the PKR catalytic cycle. After
Entry
Alkyne
t
Isomer
ratio
Yield
[%][a]
[h]
1
2
3
4
5
6
8a
8b
8c
8d
8e
8 f
0.5
2
1
2
2
-
-
52
30
2:1
1:0
2:1
2:1
50[b]
29
66
45
2
[a] Some samples were contaminated with a trace of BINAP. [b] Sum-
mation of yield of separated diastereoisomers (34% + 16%).
In addition, complexes 9c and 9e were characterized by
X-ray crystallography.
The two diastereomers formed on mixing alkyne 8c and
complex 7 can be separated by column chromatography. Iso-
lated reaction yields and inverse gated 31P NMR spectrosco-
py both give a 9cmaj/9cmin ratio of 2:1. These isomers can be
crystallized individually and both 9cmaj and 9cmin show that
the chelating mode of coordination of the BINAP is pre-
served. If the chirality of the BINAP is arbitrarily selected
to be R, the different configurations of the tetrahedral cores
of 9cmaj and 9cmin can be readily observed (Figure 2).
Figure 2 also shows schematically how the orientation of the
alkyne differs between the two isomers.
combining [Co2(CO)8] and (ꢁ)-BINAP as shown in
Scheme 8, alkynes 8a–f were added and the reaction was
stirred for a further 0.5–2 h at 408C (Scheme 10). Workup of
the product mixtures gave air-sensitive brown powders that
1
were characterized by IR, H, 13C, and 31P NMR spectrosco-
py, mass spectrometry, and elemental analysis, and identified
as the BINAP-containing alkyne complexes 9a–f (Table 1).
Alkyne complex 9e serves as a model of a typical enyne
substrate interacting with the dicobalt species (Figure 3).
One would expect a similar complex to form with the analo-
gous enyne, 8 f, but crystallographic studies of 9 f were
plagued by disorder problems (see Figures S5, S6, and S7 in
the Supporting Information). Although isomer separation
was not possible by column chromatography, an isomer
Scheme 10. Synthesis of [Co2(alkyne)(binap)(CO)4].
Chem. Eur. J. 2005, 11, 2566 – 2576
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2569

S. E. Gibson et al.
Scheme 11. Laschatꢀs[17] report of the inactivity of [Co2(3,3-dimethyl-
butyne)(CO)6] in the stoichiometric PKR (top). Our example of 9d in
the stoichiometric PKR (bottom).
Figure 2. The molecular structures of 9cmaj (left) and 9cmin (right).
chiral, bidentate phosphorus–sulfur ligand to [Co2(3,3-dime-
thylbutyne)(CO)6] occurs with a diastereomeric coordina-
tion ratio of 20:1. This is the highest obtained for the com-
plexation of a chiral phosphine to a dicobaltcarbonyl com-
plex containing a monosubstituted alkyne. Nicholas[20] also
reports that, depending on the chiral alkyne used, essentially
complete selectivity can be achieved when complexing tri-
phenylphosphine to [Co2(chiral alkyne)(CO)6].
Nevertheless, the BINAP system does of course lead to
PKR products with high enantioselectivities in many cases.
Although a ratio of only 2:1 was obtained with complexes
9c, and most importantly 9e and 9 f, the enantiomeric ratio
of the cyclopentenone produced from the enyne is 94:6
(88% ee) as shown in Scheme 9. This discrepancy in ratio
could be explained by the fact that the syntheses of 9 take
place at 408C rather than the PKR temperature of 758C.
We therefore became interested in how 9cmaj and 9cmin
would behave at 758C under a CO atmosphere. These iso-
mers were selected for study as they were readily separated
by column chromatography.
Samples of either 9cmaj or 9cmin were heated in DME at
758C for 1 h (under either a CO or nitrogen atmosphere),
cooled, the DME was removed in vacuo, and the 31P NMR
spectrum measured in CDCl3 to reveal mixtures containing
both 9cmaj and 9cmin in all four experiments. The time re-
quired for diastereomer interconversion to lead to the equi-
librium isomer ratio of 2:1 was measured at 758C in DME
in the NMR spectrometer. We have found that more time is
required for 9cmaj to reach this 2:1 ratio under a nitrogen at-
mosphere than under an atmosphere of CO (Figure 4). This
same trend is observed at 608C though, as expected, isomer-
ization rates are slower. Variable-temperature 31P NMR ex-
periments reveal that peaks corresponding to 9cmin first start
to emerge at 508C when an NMR sample of 9cmaj in DME
is prepared under CO and slowly heated in the spectrometer
from room temperature. When the same sample is heated
under nitrogen, 9cmin does not appear until 658C.
Figure 3. The molecular structure of 9e.
ratio of 2:1 was again observed for both 9e and 9 f by
31P NMR spectroscopy.
Complexes 9a–f, in which the BINAP ligand acts as a che-
late, are different from the bridging structures predicted by
Hiroi[5] and Buchwald[6] to be intermediates in their catalyt-
ic, asymmetric PKRs. Their predictions were consistent with
Laschatꢀs[17] isolation of the bridging structure 10 from the
reaction between [Co2(3,3-dimethylbutyne)(CO)6] and (R)-
BINAP.[18] Bridging structure 10 proved to be inert, howev-
er, to typical PKR conditions (Scheme 11).[17] In contrast, re-
action of chelating complex 9d from this study with norbor-
nene gave the expected PKR product. To probe whether the
order of addition of BINAP and alkyne dictated whether
the chelated or bridged BINAP structure formed, 9c was
synthesized following Laschatꢀs protocol of alkyne addition
before BINAP addition. The resulting product gave a
31
ꢂ
P NMR spectrum with Co P peaks identical to those origi-
nally obtained for 9c. No evidence of any bridged product
was detected.
It is interesting to compare the low diastereoselectivity
exhibited by the BINAP system (apart from with phenylace-
tylene) with the high selectivities seen with some reactions
pertaining to the stoichiometric, asymmetric PKR. For ex-
ample, Verdaguer and Riera[19] report that coordination of a
The fact that diastereomer interconversion does not pro-
hibit high enantiomeric excesses in the intramolecular
BINAP–enyne system is in contrast to many literature stud-
ies of intermolecular, stoichiometric systems in which
2570
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2566 – 2576

Pauson–Khand Reaction
FULL PAPER
ꢂ
ample, heating a 1:1 isomer mixture of [Co2(chiral-P S)(2-
ꢂ
methyl-3-butyn-2-ol)(CO)4] with
a
different chiral P S
ligand gave a final ratio of 4.5:1 after 17 h at 708C.[24c,d]
Since heating 9cmaj under CO allows for faster isomeriza-
tion than heating under a nitrogen atmosphere, we propose
that the mechanism of isomerization begins with an associa-
tive addition of CO (Scheme 12). This can then be followed
by dissociation of alkyne or phosphorus. The alkyne can
rotate and recoordinate to generate the other diastereomer
or the other isomer can be generated through a bridging
BINAP species. We have never obtained any evidence of a
bridging BINAP complex, but it could be present in solution
in amounts too small to be detected.
Having examined the behavior of [Co2(alkyne)-
(binap)(CO)4], we next wished to probe how alkenes react
with this species during the next phase of the catalytic cycle.
We hoped to isolate a complex containing bound BINAP,
alkyne, and alkene ligands to model a potential PKR inter-
mediate. Though not crystallographically characterized,
Itami and Yoshida[25] have reported extensive NMR spectro-
scopic data for a ruthenacyclopentene PKR intermediate.
Their approach of attaching a pyridylsilyl group to the
olefin allowed them to characterize this complex, which has
a five-membered ring derived from the metal, an alkene,
and an alkyne. It was less important to us at this point to
use alkynes and alkenes that are typical PKR substrates,
since it was likely that these would turnover to give cyclo-
pentenone product. Regardless of whether we added di-
methyl maleate, styrene, or cyclopentene as the alkene com-
ponent, the same deep red solid was isolated in low yield.
X-ray crystallography confirmed that alkene had not coordi-
nated and revealed that cobaltacyclopentadiene 12 had
formed (Figure 5). Not adding alkene and increasing the
equivalents of alkyne or the reaction temperature for pro-
longed periods of time did not improve the yield of 12
(Scheme 13).
Figure 4. Inverse gated 31P NMR isomerization experiments with 9cmaj
(
^
~
&
c, CO/758C; b, N2/758C; a, CO/608C; + d, N2/608C).
[Co2(phosphine)(alkyne)(CO)x] complexes have had to be
used at relatively low temperatures to avoid interconversion
and to enable high enantiomeric excesses to be obtained in
the PKR. Reports by Pauson,[21] Brunner,[21,22] and Kerr[23]
using the chiral phosphine glyphos (11) in the asymmetric
PKR with external alkenes state
that diastereomer interconver-
sion of the [Co2(glyphos)-
(alkyne)(CO)5] complex takes
place at elevated temperatures,
thereby lowering the ee. Activa-
tion of the asymmetric dicobalt
species possessing either chiral
phosphines or chiral alkynes in the range of 08C to room
temperature is achieved using an N-oxide, such as N-methyl-
morpholine N-oxide or trimethylamine N-oxide, and thus
limits epimerization and allows for increased enantiomeric
excesses.[19,23,24] Other literature reports note how the dia-
stereomeric ratios of dicobaltcarbonyl–alkyne complexes
containing bidentate phosphorus–sulfur ligands can be en-
riched through heating. For instance, heating a 1:1 isomer
Non-BINAP-containing cobaltacyclopentadiene com-
plexes have been isolated as intermediates of cyclotrimeriza-
tion reactions. Knox and Spicer[26] report that the yields of
some cobaltacyclopentadienes can be increased by isolating
[Co2(alkyne)(CO)6] and then adding one additional equiva-
ꢂ
mixture of [Co2(chiral-P S)(3,3-dimethylbutyne)(CO)4] in
toluene at 908C for 2 h, followed by heating under CO at
808C for 66 h, lead to a final ratio of 2:1.[19] In another ex-
Scheme 12. Two possible isomerization pathways for complexes of type 9.
Chem. Eur. J. 2005, 11, 2566 – 2576
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2571

S. E. Gibson et al.
lation of small amounts of cobaltacyclopentadiene com-
plexes.
Experimental Section
Syntheses were carried out under an atmosphere of nitrogen in oven-
dried glassware using standard Schlenk techniques. THF and DME were
distilled from sodium/benzophenone and dichloromethane was distilled
from calcium hydride. Enyne 8 f was prepared according to literature
procedures;[28] the synthesis of 8e was based on these same procedures.
All other reagents are commercially available and were used as received.
NMR spectra were recorded on Bruker AM500 or Avance 360 instru-
ments. Some 31P NMR spectra were collected in the inverse gated mode
to enable peak integration. Melting points were recorded in open capilla-
ries on a Sanyo Gallenkamp melting-point apparatus and are uncorrect-
ed. IR spectra were recorded on a Perkin–Elmer Spectrum RX FT-IR
spectrometer. Mass spectra were recorded on a Micromass AutoSpec-Q
instrument. Elemental analyses were performed by the London Metro-
politan University microanalytical service. HPLC analysis was performed
by using a Unicam Crystal 200 pump, a Unicam Spectra 100 UV/Vis de-
tector (set at 210 nm), and a CHIRALPAK AS column.
Figure 5. The molecular structures of 12 (left) and 13 (right).
Scheme 13. Synthesis of cobaltacyclopentadiene 12.
Synthesis of 7: In a Schlenk flask, anhydrous THF (10 mL) was added to
[Co2(CO)8] (50 mg, 0.15 mmol). After adding (ꢁ)-BINAP (93 mg,
0.15 mmol), the solution was stirred for 30 min at room temperature and
30 min at 408C in the dark. After cooling, nitrogen-saturated hexane
(20 mL) was added and red solid precipitated. The filtrate was removed
by cannula, additional hexane (10 mL) was added, and the filtrate was re-
moved again. The red solid was dried in vacuo. IR (CHCl3): n˜ =2053 (s),
1985 (s), 1890 (w), 1796 cmꢂ1 (m; CO); 31P NMR (162 MHz, [D8]THF):
d=43.1, 25.4, ꢂ13.8 ppm (BINAP monoxide). On one occasion, a small
sample of 7 suitable for a single-crystal structure determination was ob-
tained by slow diffusion of petroleum ether into a solution of 7 in CH2Cl2
at ꢂ208C. IR (CHCl3): n˜ =2053 (s), 1986 (s), 1890 (w), 1850 (sh),
1795 cmꢂ1 (m; CO); 31P NMR (146 MHz, [D8]THF): d=43.1 ppm. All at-
tempts to characterize by other methods were unsuccessful due to the in-
stability of 7.
lent of alkyne in the presence of Me3NO at room tempera-
ture. We, too, tried adding an N-oxide, N-methylmorpholine
N-oxide, because we thought this might generate a vacant
coordination site to which the alkene could bind. As the
structure of complex 13 shows, methyl maleate did indeed
bind to the metal, but the five-membered cobaltacyclopenta-
diene ring derived from two equivalents of alkyne was still
formed (Scheme 14, Figure 5). The core structure of 13 is
very similar to that of 12 and other cobaltacyclopentadiene
complexes.[26,27] Analogues of 12 and 13 with alkynes more
electron-rich than 8c could not be synthesized. Presumably
this is a function of the propensity of electron-poor alkynes
to cyclize.
PKR sampling experiment as shown in Scheme 9: [Co4(CO)12] (16.1 mg,
0.028 mmol) and (S)-BINAP (35.0 mg, 0.056 mmol) were dissolved in
CO-saturated DME (3 mL) in a 10 mL round-bottomed flask fitted with
a reflux condenser and stirred at room temperature under a CO atmos-
phere (1.05 atm) for 15 min. A sample (1 mL) of this solution was re-
moved by syringe and injected into a different round-bottomed flask
where it was dried in vacuo. 31P NMR (146 MHz, [D8]THF): d=43.1,
25.5, ꢂ13.8 (BINAP monoxide), ꢂ14.3 ppm (BINAP). A solution of the
enyne (125 mg, 0.5 mmol) in CO-saturated DME (2 mL) was added to
the original precatalyst solution and heated to 758C for 5 h under CO.
The resulting brown mixture was cooled, filtered through a short pad of
Celite, and concentrated in vacuo. The brown residue was redissolved in
1
CDCl3, and the extent of the reaction was calculated from H NMR spec-
Scheme 14. Synthesis of cobaltacyclopentadiene 13.
troscopy (82%). Purification of the residue by flash column chromatogra-
phy (hexane/ethyl acetate, 7:3, Rf =0.12) gave the known cyclopentenone
product[29,30] as a white solid (62 mg, 45% yield, 88% ee). IR (CHCl3):
ꢂ1
ꢂ
n˜ =1715 (s, C=O), 1652 (m, C=C), 1600 (w, C C(Ar)), 1351, 1162 cm (s,
NSO2); 1H NMR (360 MHz, CDCl3): d=2.06 (dd, 3J(H,H)=4 Hz, 2J-
(H,H)=18 Hz, 1H; CHHCO), 2.44 (s, 3H; CH3), 2.56–2.65 (m, 2H;
CHHCO, NCHHC=), 3.16 (m, 1H; CHCH2CO), 4.00–4.05 (m, 2H;
NCHHC=, NCHHCH), 4.34 (d, 2J(H,H)=17 Hz, 1H; NCHHCH), 5.99
(s, 1H; C=CH), 7.35 (d, 3J(H,H)=8 Hz, 2H; m-ArH), 7.73 ppm (d, 3J-
(H,H)=8 Hz, 2H; o-ArH); 13C NMR (75 MHz, CDCl3): d=21.6 (CH3),
39.8 (CH2CO), 43.9 (CHCH2CO), 47.6 (NCH2CH), 52.4 (NCH2C=),
126.2 (C=CH), 127.4 (Cortho), 130.0 (Cmeta), 133.3 (Cpara), 144.2 (Cipso),
178.8 (C=CH), 207.4 ppm (C=O).
Conclusion
Complexes of the type [Co2(alkyne)(binap)(CO)4] have
been synthesized and shown to isomerize under Pauson–
Khand reaction conditions. The ease of isomerization sug-
gests that the selectivity of initial alkyne coordination is not
critical in determining the enantioselectivity of PKR prod-
ucts. The influence of CO on the interconversion of [Co2-
(binap)(HCꢀCCO2Me)(CO)4 has prompted us to propose
two possible isomerization pathways. Attempts to synthesize
alkene-coordinated PKR intermediates have led to the iso-
Representative procedure for the synthesis of complexes of type 9 and
complex 12
Synthesis of 9c: In a Schlenk flask, anhydrous THF (15 mL) was added
to [Co2(CO)8] (77 mg, 0.225 mmol). After (ꢁ)-BINAP (140 mg,
2572
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2566 – 2576

Pauson–Khand Reaction
FULL PAPER
0.225 mmol) was added, the solution was stirred in the dark for 30 min at
room temperature and 30 min at 408C. Methyl propiolate (28 mL,
0.338 mmol) was added by syringe and stirring continued at 408C in the
dark for 1 h. The flask was allowed to cool and neutral alumina (Gra-
de II) was added before removing the solvent in vacuo. The brown solid
was charged onto a chromatography column of silica and eluted under ni-
trogen first with hexane/diethyl ether (9:1) to remove unreacted
[Co2(CO)8] and then with hexane/diethyl ether (7:3; 9cmin Rf =0.24, 9cmaj
Rf =0.18) to collect the two diastereomers of 9c as dark brown solids
(9cmin: 33 mg, 16%, 9cmaj: 72 mg, 34%).
6.5–7.9 ppm (m, 32H; BINAP ArH); 13C NMR (125 MHz, CDCl3): d=
72.3 (HCꢀ), 75.6 (d, 2J(C,P)=14.0 Hz, HCꢀ), 125.5–138.1 (BINAP ArC),
203.2, 211.1 (CꢀO); 31P NMR (202 MHz, CDCl3): d=53.6, 44.9 ppm
(CoPPh2Ar); MS (FAB): m/z (%): 877 (2) [MꢂH]+, 849 (1)
[MꢂHꢂCO]+, 822 (6) [Mꢂ2CO]+, 794 (34) [Mꢂ3CO]+, 766 (88)
[Mꢂ4CO]+, 681 (100) [BINAP+Co]+, 437 (86) [BINAPꢂPPh2]+; ele-
mental analysis calcd (%) for C50H34Co2O4P2 (878.63): C 68.35, H 3.90;
found: C 68.43, H 4.09.
Synthesis of 9d: The same procedure as in the synthesis of 9c was fol-
lowed except 2 h of heating was required after the addition of phenylace-
tylene (35 mL, 0.338 mmol) as the alkyne. The brown solid was charged
onto a chromatography column of silica and eluted under nitrogen first
with hexane/diethyl ether (9:1) to remove unreacted dicobalt octacarbon-
yl and then with hexane/diethyl ether (8:2; Rf =0.38) to collect the title
Data for 9cmin: Analytically pure 9cmin suitable for a single-crystal struc-
ture determination was obtained by slow diffusion of pentane into a so-
lution of 9cmin in CH2Cl2 at ꢂ208C. M.p. 214–2168C (decomp); IR
(CHCl3): n˜ =2053 (s), 1996 (s), 1947 (m), 1683 cmꢂ1 (m; CO); 1H NMR
(500 MHz, CDCl3): d=3.19 (s, 3H; CH3), 5.66 (d, 3J(P,H)=7.4 Hz, 1H;
HCꢀ), 6.5–7.8 ppm (m, 32H; BINAP ArH); 13C NMR (125 MHz,
CDCl3): d=51.6 (CH3), 70.7 (d, 2J(C,P)=14.6 Hz, ꢀCC), 80.9 (HCꢀ),
120.3–137.4 (BINAP ArC), 172.8 (CCO2), 201.4, 210.8 ppm (CꢀO);
31P NMR (202 MHz, CDCl3): d=52.0, 47.2 (CoPPh2Ar); MS (FAB): m/z
(%): 880 (4) [Mꢂ2CO]+, 852 (31) [Mꢂ3CO]+, 824 (90) [Mꢂ4CO]+, 681
(100) [BINAP+Co]+, 437 (80) [BINAPꢂPPh2]+; elemental analysis calcd
(%) for C52H36Co2O6P2 (936.67): C 66.68, H 3.87; found: C 66.51, H 3.75.
compound as
a dark brown solid (62 mg, 29%). M.p. 197–2008C
(decomp); IR (CHCl3): n˜ =2040 (s), 1982 (s), 1932 cmꢂ1 (m, CO);
1H NMR (360 MHz, CDCl3): d=5.77 (dd, 3J(P,H)=6.2, 3.5 Hz, 1H;
HCꢀ), 6.3–8.1 ppm (m, 37H; ArH, BINAP ArH); 13C NMR (125 MHz,
CDCl3): d=72.5 (HCꢀ), 87.8 (d, 2J(C,P)=20.6 Hz, ꢀCC), 120.3–141.6
(ArC, BINAP ArC), 201.7, 213.7 ppm (CꢀO); 31P NMR (202 MHz,
CDCl3): d=53.5, 36.7 ppm (CoPPh2Ar); MS (FAB): m/z (%): 870 (1)
[Mꢂ3CO]+, 842 (3) [Mꢂ4CO]+, 681 (6) [BINAP+Co]+, 655 (22)
[BINAP(O)2]+, 453 (7) [BINAP(O)ꢂPPh2]+, 437 (7) [BINAPꢂPPh2]+;
elemental analysis calcd (%) for C56H38Co2O4P2 (954.73): C 70.45, H 4.01;
found: C 70.29, H 4.14.
Data for 9cmaj: Analytically pure 9cmaj suitable for a single-crystal struc-
ture determination was obtained by slow diffusion of light petroleum
ether into a solution of 9cmaj in CH2Cl2 at ꢂ208C. M.p. 247–2508C
(decomp); IR (CHCl3): n˜ =2054 (s), 1999 (s), 1956 (m), 1687 cmꢂ1 (m,
Synthesis of sulfonamide precursor to 8e (N-(propane)-p-toluene sul-
fonamide):[28a] p-Toluene sulfinic acid sodium salt hydrate (7.12 g, 40 mmol)
was dissolved in water (75 mL) at 58C and n-propylamine (1.6 mL,
20 mmol) and glacial acetic acid (1.1 mL, 20 mmol) were added. The re-
action was stirred and sodium hypochlorite (40 mL, 30 mmol) was added
dropwise over 1 h. The flask was then allowed to reach room tempera-
ture and stirred for 16 h. The reaction was acidified to pH 6 with concen-
trated hydrochloric acid. The precipitate was filtered, washed with water
(3ꢄ30 mL), and dried in vacuo to give the desired product as a white
crystalline solid (4.10 g, 96%). M.p. 54–558C; IR (CHCl3): n˜ =3376, 1496
(m, NH), 1328, 1160 cmꢂ1 (s, NSO2); 1H NMR (500 MHz, CDCl3): d=
0.50 (t, 3J(H,H)=7 Hz, 3H; CH3), 1.47 (tq, 3J(H,H)=7, 7 Hz, 2H;
NCH2CH2CH3), 2.42 (s, 3H; CH3), 2.88 (t, 3J(H,H)=7 Hz, 2H;
NCH2CH2CH3), 4.77 (s, 1H; NH), 7.30 (d, 3J(H,H)=8 Hz, 2H; ArH),
CO) cmꢂ1 1H NMR (500 MHz, CDCl3): d=3.43 (s, 3H; CH3), 5.72 (dd,
;
3J(P,H)=5.5, 2.5 Hz, 1H; HCꢀ), 6.4–8.0 ppm (m, 32H; BINAP ArH);
13C NMR (125 MHz, CDCl3): d=51.7 (CH3), 73.4 (d, 2J(C,P)=22.1 Hz,
ꢀCC), 73.9 (HCꢀ), 120.3–138.2 ppm (BINAP ArC), 172.0 (CCO2), 200.3,
211.3 ppm (CꢀO); 31P NMR (202 MHz, CDCl3): d=51.9, 40.7 ppm (CoP-
Ph2Ar); MS (FAB): m/z (%): 880 (2) [Mꢂ2CO]+, 852 (12) [Mꢂ3CO]+,
824 (62) [Mꢂ4CO]+, 681 (100) [BINAP+Co]+, 437 (94) [BI-
NAPꢂPPh2]+; elemental analysis calcd (%) for C52H36Co2O6P2 (936.67):
C 66.68, H 3.87; found: C 66.66, H 3.79.
Synthesis of 9a: The same procedure as in the synthesis of 9c was fol-
lowed except only 0.5 h of heating was required after the addition of di-
ethyl acetylenedicarboxylate (58 mL, 0.338 mmol) as the alkyne. The
brown solid was charged onto a chromatography column of silica and
eluted under nitrogen first with hexane/ethyl acetate (19:1) to remove
unreacted [Co2(CO)8] and then with hexane/ethyl acetate (8:2; Rf =0.26)
to collect 9a as a dark brown solid (119 mg, 52%). M.p. 175–1808C
(decomp); IR (CHCl3): n˜ =2063 (s), 2010 (s), 1967 (m), 1687 cmꢂ1 (m,
CO); 1H NMR (500 MHz, CDCl3): d=1.08 (t, 3J(H,H)=7.1 Hz, 3H;
CH3), 1.19 (t, 3J(H,H)=7.1 Hz, 3H; CH3), 4.04–4.22 (m, 4H; CH2), 6.3–
8.0 ppm (m, 32H; BINAP ArH); 13C NMR (125 MHz, CDCl3): d=14.1
(CH3), 60.6, 60.7 (CH2), 71.7 (ꢀCC), 78.2 (d, 2J(C,P)=20.0 Hz, ꢀCC),
120.3–139.1 (BINAP ArC), 171.2, 173.1 (CCO2), 198.9, 211.4 ppm (CꢀO);
31P NMR (202 MHz, CDCl3): d=48.2, 41.6 ppm (CoPPh2Ar); MS
(FAB): m/z (%): 977 (9) [MꢂOEt]+, 966 (2) [Mꢂ2CO]+, 938 (28)
[Mꢂ3CO]+, 910 (100) [Mꢂ4CO]+, 837 (27) [Mꢂ4COꢂCO2Et]+,
681 (60) [BINAP+Co]+, 437 (54) [BINAPꢂPPh2]+; elemental analysis
calcd (%) for C56H42Co2O8P2 (1022.76): C 65.77, H 4.14; found: C 65.59,
H 4.19.
3
7.75 ppm (d, J(H,H)=8 Hz, 2H; ArH); 13C NMR (125 MHz, CDCl3): d=
11.1 (CH3), 21.5 (CH3), 22.9 (NCH2CH2CH3), 44.9 (NCH2CH2CH3), 127.1
(Cortho), 129.6 (Cmeta), 137.0 (Cpara), 143.2 ppm (Cipso); MS (FAB): m/z
(%): 214 (100) [M+H]+, 155 (23) [CH3C6H4SO2]+, 91 (19) [CH3C6H4]+.
Synthesis of 8e:[28b] N-(Propane)-p-toluenesulfonamide (1.77 g, 8 mmol)
was dissolved in anhydrous DMF (15 mL) at room temperature. Sodium
hydride (0.43 g, 11 mmol) was added with care and the mixture was stir-
red for 30 minutes. Propargyl bromide (1.5 mL, 14 mmol) was added and
the reaction was stirred for 1 h. It was then quenched with water (50 mL)
and the product was extracted with diethyl ether (4ꢄ30 mL). The ethere-
al extracts were dried over magnesium sulfate and then concentrated in
vacuo. The crude mixture was then loaded onto a chromatography
column of silica and eluted with hexane/diethyl ether (5:1; Rf =0.16) to
collect 8e, which was then ground in a mortar and dried in vacuo (1.39 g,
67%). M.p. 47–498C; IR (CHCl3): n˜ =3308 (m, CꢀCH), 1348, 1159 ppm
(s, NSO2); 1H NMR (500 MHz, CDCl3): d=0.91 (t, 3J(H,H)=7 Hz, 3H;
CH3), 1.58 (tq, 3J(H,H)=7, 7 Hz, 2H; NCH2CH2CH3), 2.00 (t, 4J(H,H)=
2 Hz, 1H; NCH2CꢀCH), 2.40 (s, 3H; CH3), 3.14 (t, 3J(H,H)=7 Hz, 2H;
NCH2CH2CH3), 4.11 (d, 4J(H,H)=2 Hz, 2H; NCH2CꢀCH), 7.27 (d, 3J-
(H,H)=8 Hz, 2H; ArH), 7.71 ppm (d, 3J(H,H)=8 Hz, 2H; ArH);
13C NMR (125 MHz, CDCl3): d=11.0 (CH3), 20.8 (NCH2CH2CH3), 21.4
(CH3), 36.1 (NCH2CꢀCH), 47.9 (NCH2CH2CH3), 73.5 (NCH2CꢀCH),
76.6 (NCH2CꢀCH), 127.6 (Cortho), 129.4 (Cmeta), 136.0 (Cpara), 143.3 ppm
(Cipso); MS (FAB): m/z (%): 252 (100) [M+H]+, 155 (18)
[CH3C6H4SO2]+, 91 (24) [CH3C6H4]+; elemental analysis calcd (%) for
C13H17NO2S (251.35): C 62.07, H 6.82, N 5.57; found: C 62.25, H 6.98, N
5.54.
Synthesis of 9b: The same procedure as in the synthesis of 9c was fol-
lowed except for the following modifications: After allowing the reaction
between the [Co2(CO)8] and (ꢁ)-BINAP to cool and closing the Schlenk
tap to nitrogen, acetylene gas was gently bubbled into the reaction so-
lution through a syringe needle for 90 s. The acetylene needle was then
removed, a balloon filled with nitrogen was added, and heating continued
in the dark at 408C for 2 h. The flask was allowed to cool and neutral
alumina (Grade II) was added before removing the solvent in vacuo. The
brown solid was charged onto a chromatography column of silica and
eluted under nitrogen first with hexane/diethyl ether (19:1) to remove
unreacted [Co2(CO)8] and then with hexane/diethyl ether (8:2; Rf =0.40)
to collect 9b as a dark brown solid (59 mg, 30%). M.p. 176–1808C
(decomp); IR (CHCl3): n˜ =2042 (s), 1978 (s), 1923 cmꢂ1 (m, CO);
1H NMR (500 MHz, CDCl3): d=5.22 (d, 3J(P,H)=8.1 Hz, 2H; HCꢀ),
Synthesis of 9e: The same procedure as in the synthesis of 9c was fol-
lowed except 2 h of heating was required after the addition of 8e (57 mg,
Chem. Eur. J. 2005, 11, 2566 – 2576
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2573

S. E. Gibson et al.
0.226 mmol) as the alkyne. The brown solid was charged onto a chroma-
tography column of silica and eluted under nitrogen first with hexane/
ethyl acetate (19:1) to remove unreacted [Co2(CO)8] and then with
hexane/ethyl acetate (7:3; Rf =0.74) to collect 9e as an air-sensitive
brown solid (159 mg, 66%). Analytically pure 9e suitable for a single-
crystal structure determination was obtained by slow diffusion of pentane
into a solution of 9e in CH2Cl2 at ꢂ208C. M.p. 176–1788C; IR (CHCl3):
n˜ =2044 (m), 1980 (s), 1940 cmꢂ1 (w, CO); 1H NMR (500 MHz, CDCl3):
too broad to interpret; 13C NMR (125 MHz, CDCl3, major diastereomer,
minor diastereomer resonances difficult to assign with confidence): d=
11.1 (CH3), 20.2 (NCH2CH2CH3), 21.4 (CH3), 49.4 (NCH2CH2CH3), 60.3
(NCH2CꢀCH), 81.6 (NCH2CꢀCH), 83.8 (NCH2CꢀCH), 125.6–142.4
(ArC, BINAP ArC), 201.1, 203.2, 211.9 ppm (CꢀO); 31P NMR (202 MHz,
CDCl3): d=57.9, 51.6, 43.8, 39.9 (CoPPh2Ar), ꢂ14.9 ppm (BINAP); MS
(FAB): m/z (%): 1019 (18) [Mꢂ3CO]+, 991 (54) [Mꢂ4CO]+, 681 (16)
[BINAP+Co]+, 437 (24) [BINAPꢂPPh2]+; elemental analysis calcd (%)
for C61H49Co2NO6P2S (1103.92): C 66.37, H 4.47, N 1.27; found: C 66.32,
H 4.41, N 1.17.
gated 31P NMR spectra were taken every 15 min until either a 9cmaj/9cmin
ratio of 2:1 was reached, 7 h elapsed, or decomposition occurred. Rela-
tive amounts of the two isomers were calculated by integrating the peaks
at 47.2 ppm (9cmin) and 40.7 ppm (9cmaj).
Type C (variable-temperature 31P NMR, RT!758C): Isomer 9cmaj (25 mg,
0.027 mmol) was dissolved in DME (0.7 mL) under a nitrogen or CO at-
mosphere and transferred to a screw-cap NMR tube (also kept under a
nitrogen or CO atmosphere) by means of a syringe. 31P NMR spectra
were taken every 58C from room temperature to 758C to determine
when peaks corresponding to 9cmin are first observed.
Synthesis of 12: The same procedure as in the synthesis of 9c was fol-
lowed except 40 h of heating was required after the addition of methyl
propiolate (60 mL, 0.675 mmol). The dark red solid was charged onto a
chromatography column of silica and eluted under nitrogen first with
hexane/diethyl ether (9:1) to remove unreacted [Co2(CO)8] and then
with hexane/diethyl ether (7:3; Rf =0.15) to collect 12 as a dark red solid
(16 mg, 7%). Analytically pure 12 suitable for a single-crystal structure
determination was obtained by slow diffusion of petroleum ether into a
solution of 12 in CH2Cl2 at ꢂ208C. M.p. 244–2488C (decomp); IR
(CHCl3): n˜ =2020 (s), 1971 (s), 1676 cmꢂ1 (m, CO); 1H NMR (500 MHz,
CDCl3): d=2.19 (s, 3H; CH3), 3.21 (s, 3H; CH3), 5.22 (s, 1H;
CHCHCCO2) 5.87 (d, 4J(P,H)=8.5 Hz, 1H; CHCHCCO2), 6.4–7.9,
8.76 ppm (m, 32H; BINAP ArH); 13C NMR (125 MHz, CDCl3): d=49.3,
51.0 (CH3), 113.0, 116.4 (CHCHCCO2), 120.3–154.0 (BINAP ArC,
CCO2), 175.1, 178.0 (CCO2), 202.4, 207.0 ppm (CꢀO); 31P NMR
(202 MHz, CDCl3): d=49.2, 39.5 ppm (CoPPh2Ar); MS (FAB): m/z (%):
908 (1) [Mꢂ3CO]+, 681 (5) [BINAP+Co]+, 655 (15) [BINAP(O)2]+, 639
(6) [BINAP(O)]+, 437 (70) [BINAPꢂPPh2]+; elemental analysis calcd
(%) for C55H40Co2O7P2 (992.73): C 66.54, H 4.06; found: C 66.44, H 4.03.
Synthesis of 9 f: The same procedure as in the synthesis of 9c was fol-
lowed except 2 h of heating was required after the addition of 8 f (56 mg,
0.226 mmol) as the alkyne. The brown solid was charged onto a chroma-
tography column of silica and eluted under nitrogen first with hexane/
ethyl acetate (19:1) to remove unreacted [Co2(CO)8] and then with
hexane: ethyl acetate (7:3; Rf =0.73) to collect 9 f as an air-sensitive
brown solid (109 mg, 45%). M.p. 143–1448C; IR (CHCl3): n˜ =2043, 1980,
1939 cmꢂ1 (s, CO); 1H NMR (500 MHz, CDCl3): too broad to interpret;
13C NMR (125 MHz, CDCl3, major diastereomer, minor diastereomer
resonances difficult to assign with confidence): d=21.3 (CH3), 49.8
(NCH2CH=CH2), 60.2 (NCH2CꢀCH), 82.0 (NCH2CꢀCH), 83.9 (NCH2Cꢀ
CH), 118.5 (NCH2CH=CH2), 125.4–138.4 (ArCortho,meta,para, BINAP ArC,
NCH2CH=CH2), 142.4 (Cipso), 201.0, 202.9, 211.6 ppm (CꢀO); 31P NMR
(202 MHz, CDCl3): d=58.9, 53.2, 44.0, 39.2 (CoPPh2Ar), ꢂ14.6 ppm
(BINAP); MS (FAB): m/z (%): 1017 (11) [Mꢂ3CO]+, 989 (40)
[Mꢂ4CO]+, 681 (100) [BINAP+Co]+, 437 (65) [BINAPꢂPPh2]+.
Synthesis of 13: In a Schlenk flask, anhydrous dichloromethane (15 mL)
was added to [Co2(CO)8] (77 mg, 0.225 mmol). After (ꢁ)-BINAP
(140 mg, 0.225 mmol) was added, the solution was stirred in the dark for
30 min at room temperature. After adding a condenser to the Schlenk
flask, the mixture was stirred at 408C in the dark for an additional
30 min. Methyl propiolate (28 mL, 0.338 mmol) was added by means of a
syringe and stirring continued at 408C in the dark for 1 h. The flask was
allowed to cool and dimethyl maleate (42 mL, 0.338 mmol) and N-methyl-
morpholine N-oxide (158 mg, 1.35 mmol) were added. The reaction stir-
red for 20 h at room temperature in the dark. Neutral alumina (Grade II)
was added before removing the solvent in vacuo. The dark purple-red
solid was charged onto a chromatography column of silica and eluted
under nitrogen first with hexane/ethyl acetate (9:1) to remove unreacted
[Co2(CO)8] and then with hexane/ethyl acetate (6:4; Rf =0.25) to collect
13 as a dark purple-red solid (25 mg, 10%). Analytically pure 13 suitable
for a single-crystal structure determination was obtained by slow diffu-
sion of petroleum ether into a solution of 13 in CH2Cl2 at ꢂ208C.
M.p. 250–2558C (decomp); IR (CHCl3): n˜ =2023 (s), 1978 (m), 1725
(m), 1681 cmꢂ1 (m, CO); 1H NMR (500 MHz, CDCl3, broad peaks):
d=2.17 (s, 3H; CHCCO2CH3), 3.21 (s, 3H; CHCCO2CH3), 3.56 (s, 3H;
maleate CH3), 3.68 (s, 3H; maleate CH3), 5.23 (s, 1H; CHCHCCO2)
5.74 (s, 1H; CHCHCCO2), 6.3–7.7, 8.65 ppm (m, 34H; BINAP
ArH, O2CCHCHCO2); 13C NMR (125 MHz, CDCl3): d=44.9, 49.5
(O2CCHCHCO2), 49.3, 51.2, 51.4, 51.7 (CH3), 111.1, 117.7
(CHCHCCO2), 120.3–165.6 (BINAP ArC, CCO2) 173.1, 173.2, 173.8,
176.7 (CCO2), 199.3, 207.1 ppm (CꢀO); 31P NMR (202 MHz, CDCl3): d=
47.0, 38.2 ppm (CoPPh2Ar); MS (FAB): m/z (%): 1108 (1) [M]+, 936 (5)
[MꢂCOꢂ(MeO2CCH)2]+, 908 (76) [Mꢂ2COꢂ(MeO2CCH)2]+, 681 (14)
[BINAP+Co]+, 655 (28) [BINAP(O)2]+, 437 (72) [BINAPꢂPPh2]+; ele-
mental analysis calcd (%) for C60H48Co2O10P2 (1108.85): C 64.99, H 4.36;
found: C 64.91, H 4.47.
Complex 9d in a stoichiometric PKR: Complex 9d (382 mg, 0.4 mmol)
dissolved in CO-saturated DME was syringed into a 25 mL round-bot-
tomed flask containing norbornene (377 mg, 4.0 mmol) and fitted with a
reflux condenser under CO (1.05 atm). The reaction system was stirred at
758C for 5 h under CO. The resulting brown solution was cooled and pu-
rified twice by flash column chromatography (hexane/diethyl ether, 49:1,
Rf =0.08) to give the known cyclopentenone product[30, 31] as a white solid
(17.1 mg, 25% yield). An inverse gated 31P NMR spectrum of a sample
of the 9d used is this experiment showed that 76% of the material was
actually 9d; the remainder of the sample consisted of BINAP,
BINAP(O), and BINAP(O)2. The reported yield is calculated based on
1
this integration. IR (CHCl3): n˜ =1697 (s, C=O) cmꢂ1; H NMR (360 MHz,
CDCl3): d=1.00 (d, 2J(H,H)=11 Hz, 1H; CHCHHCH), 1.12 (d, 2J-
(H,H)=11 Hz, 1H; CHCHHCH), 1.26–1.43, 1.58–1.72 (m, 4H;
CHCH2CH2CH), 2.28 (brd, 3J(H,H)=3 Hz, 1H; CH2CHCHCH=), 2.37
(d, 3J(H,H)=5 Hz, 1H; CHCO), 2.50 (brd, 3J(H,H)=3 Hz, 1H;
CH2CHCHCO), 2.71 (m, 1H; CHCH=), 7.30–7.40 (m, 3H; m-ArH, p-
ArH), 7.64 (d, 3J(H,H)=3 Hz, 1H; CH=), 7.70 ppm (m, 2H; o-ArH);
13C NMR (90 MHz, CDCl3): d=28.4, 29.1, 31.3 (CHCH2CH2CH,
CHCH2CH), 38.3, 39.4, 47.7, 54.9 (CHCHCH2CHCH), 127.0, 128.4
(Cortho, Cmeta, Cpara), 131.5 (Cipso), 146.1 (PhC=), 160.2 (CH=), 209.0
(C=O).
Isomerization experiments with 9c
Type A (reaction in DME, 31P NMR in CDCl3): Isomer 9cmaj or 9cmin
(25 mg, 0.027 mmol) was dissolved in DME (3 mL) in a round-bottomed
flask fitted with a reflux condenser under a nitrogen or CO atmosphere.
The solution was stirred at 758C for 1 h. The solution was cooled, the
DME removed in vacuo, and the 31P NMR spectrum of the residue mea-
sured in CDCl3.
Type B (31P NMR every 15 min at 758C or 608C): Isomer 9cmaj (25 mg,
0.027 mmol) was dissolved in DME (0.7 mL) under a nitrogen or CO at-
mosphere and transferred to a screw-cap NMR tube (also kept under a
nitrogen or CO atmosphere) by means of a syringe. With the NMR spec-
trometer set at 758C or 608C, the NMR tube was inserted and inverse
X-ray crystallography: Table 2 provides a summary of the crystallograph-
ic data for compounds 7, 9cmaj, 9cmin, 9e, 9 f, 12, and 13. Data were col-
lected by using Nonius Kappa CCD (7), Oxford Diffraction Xcalibur 3
(9cmaj and 13), Bruker P4 (9cmin, 9e and 12), and Oxford Diffraction PX
Ultra (9 f) diffractometers, and the structures were refined based on F2
by using the SHELXTL and SHELX-97 program systems.[32] CCDC
212506 (7), 248509 (9cmaj), 248510 (9cmin), 255848 (9e), 255847 (9 f),
255849 (12), and 255850 (13) contain the supplementary crystallographic
2574
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2566 – 2576

Pauson–Khand Reaction
FULL PAPER
Table 2. Crystal data, data collection, and refinement parameters for compounds 7, 9cmaj, 9cmin, 9e, 9 f, 12, and 13.[a]
7
9cmaj
9cmin
9e
formula
solvent
Mr
color, habit
crystal size [mmꢂ1
T [K]
C50H32Co2O6P2
1.5CH2Cl2
1035.95
red blocks
0.15ꢄ0.10ꢄ0.10
120
C52H36Co2O6P2
CH2Cl2
1021.53
orange thin plates
0.20ꢄ0.16ꢄ0.03
150
C52H36Co2O6P2
CH2Cl2
1021.53
orange-red plates
0.30ꢄ0.28ꢄ0.05
293
C61H49Co2NO6P2S
CH2Cl2
1188.80
deep red blocky needles
0.67ꢄ0.12ꢄ0.10
203
]
crystal system
space group
a [ꢂ]
b [ꢂ]
c [ꢂ]
monoclinic
P21/n (no. 14)
14.4575(3)
19.3414(4)
16.2073(4)
90
94.5540(10)
90
4517.71(17)
4
triclinic
P1 (no. 2)
triclinic
P1 (no. 2)
monoclinic
P21/c (no. 14)
12.506(4)
31.226(8)
14.528(4)
90
104.28(2)
90
5498(3)
4
¯
¯
10.0063(10)
13.1274(11)
18.7862(14)
73.232(7)
86.586(8)
74.004(8)
2270.7(3)
2
12.9966(15)
13.229(3)
14.5918(18)
90.514(13)
96.565(11)
110.359(18)
2333.5(6)
2
a [8]
b [8]
g [8]
V [ꢂ3]
Z
1calcd [gcmꢂ3
]
1.523
1.494
1.454
1.436
radiation
MoKa
1.034
MoKa
0.971
CuKa
7.684
MoKa
0.850
m [mmꢂ1
]
2q max [8]
relfns measured
50
7931
71
18162
120
6881
47
8076
relfns observed [jFo j>4s(jFoj)]
4786
587
0.058/0.138
14112
586
0.155/0.225
4181
566
0.074/0.172
3728
639
0.078/0.129
variables
[b]
R1/wR2
9 f
12
13
formula
solvent
Mr
C61H47Co2NO6P2S
C55H40Co2O7P2
–
992.67
orange-red plates
0.28ꢄ0.18ꢄ0.05
203
C60H48Co2O10P2
1.5CH2Cl2
1236.17
dark orange blocks
0.29ꢄ0.28ꢄ0.20
173
CH2Cl2
1186.78
red plates
0.12ꢄ0.11ꢄ0.05
173
color, habit
crystal size [mmꢂ1
T [K]
]
crystal system
space group
a [ꢂ]
b [ꢂ]
c [ꢂ]
monoclinic
P21/c (no. 14)
12.6939(12)
30.385(2)
14.4188(15)
90
101.482(9)
90
5450.1(8)
4
triclinic
P1 (no. 2)
triclinic
P1 (no. 2)
¯
¯
12.818(3)
12.998(4)
15.427(4)
85.936(18)
75.981(16)
65.772(19)
2272.8(11)
2
12.9158(5)
13.1071(5)
18.6890(7)
77.726(3)
77.296(3)
60.903(2)
2676.08(18)
2
a [8]
b [8]
g [8]
V [ꢂ3]
Z
1calcd [gcmꢂ3
]
1.446
1.451
1.534
radiation
CuKa
7.017
MoKa
0.856
MoKa
0.892
m [mmꢂ1
]
2q max [8]
relfns measured
142
10336
45
5899
65
14037
relfns observed [jFo j>4s(jFoj)]
8772
721
0.137, 0.338
3203
547
0.074, 0.128
13236
725
0.076, 0.196
variables
[b]
R1/wR2
[a] Details in common: graphite-monochromated radiation, refinement based on F2. [b] R1 =ꢀjjFo jꢂjFc jj/ꢀjFo j ; wR2 ={ꢀ[w(F2oꢂF2c)2]/ꢀ[w(Fo2)2]}1/2
;
wꢂ1 =s2(F2o)+(aP)2 +bP.
data for this paper. These data can be obtained free of charge from the
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
[1] a) J. Blanco-Urgoiti, L. AÇorbe, L. Pꢅrez-Serrano, G. Domꢆnguez, J.
Pꢅrez-Castells, Chem. Soc. Rev. 2004, 33, 32–42; b) S. E. Gibson, A.
Stevenazzi, Angew. Chem. 2003, 115, 1844–1854; Angew. Chem. Int.
Ed. 2003, 42, 1800–1810; c) B. E. Hanson, Comments Inorg. Chem.
2002, 23, 289–318; d) K. M. Brummond, J. L. Kent, Tetrahedron
2000, 56, 3263–3283; e) A. J. Fletcher, S. D. R. Christie, J. Chem.
Soc. Perkin Trans. 1 2000, 1657–1668.
[2] a) I. U. Khand, G. R. Knox, P. L. Pauson, W. E. Watts, J. Chem. Soc.
D 1971, 36–36; b) I. U. Khand, G. R. Knox, P. L. Pauson, W. E.
Watts, M. I. Foreman, J. Chem. Soc. Perkin Trans. 1 1973, 977–981.
Acknowledgements
We thank the EPSRC and Medivir (UK) for studentship support (to
K.A.C.K.), AstraZeneca and the Royal Society (Rosalind Franklin
Award) for consumables support (to J.A.L.), and Peter Haycock and Jon
Cobb for NMR experiments.
Chem. Eur. J. 2005, 11, 2566 – 2576
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2575

S. E. Gibson et al.
[3] a) F. A. Hicks, S. L. Buchwald, J. Am. Chem. Soc. 1996, 118, 11688–
11689; b) F. A. Hicks, S. L. Buchwald, J. Am. Chem. Soc. 1999, 121,
7026–7033.
[4] S. J. Sturla, S. L. Buchwald, J. Org. Chem. 1999, 64, 5547–5550.
[5] K. Hiroi, T. Watanabe, R. Kawagishi, I. Abe, Tetrahedron Lett. 2000,
41, 891–895; b) K. Hiroi, T. Watanabe, R. Kawagishi, I. Abe, Tetra-
hedron: Asymmetry 2000, 11, 797–808.
[6] S. J. Sturla, S. L. Buchwald, J. Org. Chem. 2002, 67, 3398–3403.
[7] T. M. Schmid, G. Consiglio, Tetrahedron: Asymmetry 2004, 15,
2205–2208.
[8] S. E. Gibson, S. E. Lewis, J. A. Loch, J. W. Steed, M. J. Tozer, Orga-
nometallics 2003, 22, 5382–5384.
[21] P. Bladon, P. L. Pauson, H. Brunner, R. Eder, J. Organomet. Chem.
1988, 355, 449–454.
[22] H. Brunner, A. Niedernhuber, Tetrahedron: Asymmetry 1990, 1,
711–714.
[23] A. M. Hay, W. J. Kerr, G. G. Kirk, D. Middlemiss, Organometallics
1995, 14, 4986–4988.
[24] a) H.-J. Park, B. Y. Lee, Y. K. Kang, Y. K. Chung, Organometallics
1995, 14, 3104–3107; b) J. Castro, A. Moyano, M. A. Pericꢇs, A.
Riera, A. Alvarez-Larena, J. F. Piniella, J. Am. Chem. Soc. 2000,
122, 7944–7952; c) X. Verdaguer, A. Moyano, M. A. Pericꢇs, A.
Riera, M. A. Maestro, J. Mahꢆa, J. Am. Chem. Soc. 2000, 122,
10242–10243; d) X. Verdaguer, A. Lledꢈ, C. Lꢈpez-Mosquera,
M. A. Maestro, M. A. Pericꢇs, A. Riera, J. Org. Chem. 2004, 69,
8053–8061.
[9] N. Jeong, B. K. Sung, Y. K. Choi, J. Am. Chem. Soc. 2000, 122,
6771–6772.
[10] T. Shibata, K. Takagi, J. Am. Chem. Soc. 2000, 122, 9852–9853.
[11] W. H. Suh, M. Choi, S. I. Lee, Y. K. Chung, Synthesis 2003, 2169–
2172.
[12] a) T. Shibata, N. Toshida, K. Takagi, Org. Lett. 2002, 4, 1619–1621;
b) T. Shibata, N. Toshida, K. Takagi, J. Org. Chem. 2002, 67, 7446–
7450.
[13] N. Jeong, D. H. Kim, J. H. Choi, Chem. Commun. 2004, 1134–1135.
[14] P. Braunstein, I. Pruskil, G. Predieri, A. Tiripicchio, J. Organomet.
Chem. 1983, 247, 227–237.
[15] L. Shi-Jie, Y. Fan, F. Hong-Xiang, S. Jian-Qiu, Jiegou Huaxue 1993,
12, 405–408.
[16] K. A. Bunten, D. H. Farrar, A. J. Poꢃ, A. Lough, Organometallics
2002, 21, 3344–3350.
[17] V. Derdau, S. Laschat, I. Dix, P. G. Jones, Organometallics 1999, 18,
3859–3864.
[18] Synthesis of 10 using Laschatꢀs protocol (addition of alkyne before
BINAP) was easily reproduced by us. Significantly, subjecting 3,3-di-
methylbutyne to our protocol (addition of BINAP before alkyne)
also generated the bridging complex 10, revealing that the reactivity
of 3,3-dimethylbutyne is significantly different to the alkynes exam-
ined in the study reported here.
[25] K. Itami, K. Mitsudo, K. Fujita, Y. Ohashi, J. Yoshida, J. Am. Chem.
Soc. 2004, 126, 11058–11066.
[26] R. J. Baxter, G. R. Knox, P. L. Pauson, M. D. Spicer, Organometal-
lics 1999, 18, 197–205.
[27] Selected examples of structurally characterized cobaltacyclopenta-
diene complexes: a) M. A. Bennett, P. B. Donaldson, Inorg. Chem.
1978, 17, 1995–2000; b) G. Predieri, A. Tiripicchio, M. Tiripicchio
Camellini, M. Costa, E. Sappa, J. Organomet. Chem. 1992, 423, 129–
139; c) I. Moldes, T. Papworth, J. Ros, A. Alvarez-Larena, J. F. Pi-
niella, J. Organomet. Chem. 1995, 489, C65-C67; d) T. Henkel, A.
Klauck, K. Seppelt, J. Organomet. Chem. 1995, 501, 1–6.
[28] a) F. E. Scully, Jr., K. Bowdring, J. Org. Chem. 1981, 46, 5077–5081;
b) W. Oppolzer, A. Pimm, B. Stammen, W. E. Hume, Helv. Chim.
Acta 1997, 80, 623–639.
[29] a) B. L. Pagenkopf, T. Livinghouse, J. Am. Chem. Soc. 1996, 118,
2285–2286; b) J. W. Kim, Y. K. Chung, Synthesis 1998, 142–144.
[30] S. E. Gibson, C. Johnstone, A. Stevenazzi, Tetrahedron 2002, 58,
4937–4942.
[31] T. Sugihara, M. Yamaguchi, J. Am. Chem. Soc. 1998, 120, 10782–
10783.
[32] SHELXTL PC version 5.1, Bruker AXS, Madison, WI, 1997;
SHELX-97, G. Sheldrick, University of Gçttingen, Gçttingen (Ger-
many), 1998.
[19] X. Verdaguer, M. A. Pericꢇs, A. Riera, M. A. Maestro, J. Mahꢆa, Or-
ganometallics 2003, 22, 1868–1877.
[20] D. H. Bradley, M. A. Khan, K. M. Nicholas, Organometallics 1989, 8,
554–556.
Received: November 26, 2004
Published online: February 28, 2005
2576
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2566 – 2576