Dicobalt hexacarbonyl-3-butyn-1-yl cobaloxime: Synthesis X-ray crystal structure and reactions of the first combined cobaloxime-dicobalt hexacarbonyl alkyne complex

Article abstract of DOI:10.1016/S0022-328X(03)00385-1

Reaction of 1-butynyl cobaloxime (1), synthesised by displacement of the corresponding tosylate with cobaloximato nucleophile, with dicobalt octacarbonyl gives the first trimetallic cobaloxime-dicobalt hexacarbonyl-alkyne complex (2). This trimetallic species is characterised by X-ray crystallography and undergoes the Pauson-Khand reaction to give cobaloxime-substituted cyclopentenones. We believe these are the first examples of Pauson-Khand reactions of alkynes which contain a carbon-to-metal sigma bond.

Full text of DOI:10.1016/S0022-328X(03)00385-1

Journal of Organometallic Chemistry 677 (2003) 125ꢀ132
/
www.elsevier.com/locate/jorganchem
Dicobalt hexacarbonyl-3-butyn-1-yl cobaloxime: synthesis X-ray
crystal structure and reactions of the first combined cobaloximeꢀ
/
dicobalt hexacarbonyl alkyne complex
Michael P. Coogan *, Lisa S. Stanton, Thilo Walther 1
Department of Chemistry, Cardiff University, Park Place, PO Box 912, Cardiff, Wales, CF10 3TB, UK
Received 2 April 2003; received in revised form 3 May 2003; accepted 5 May 2003
Abstract
Reaction of 1-butynyl cobaloxime (1), synthesised by displacement of the corresponding tosylate with cobaloximato nucleophile,
with dicobalt octacarbonyl gives the first trimetallic cobaloximeꢀdicobalt hexacarbonyl-alkyne complex (2). This trimetallic species
/
is characterised by X-ray crystallography and undergoes the Pauson-Khand reaction to give cobaloxime-substituted cyclopente-
nones. We believe these are the first examples of Pauson-Khand reactions of alkynes which contain a carbon-to-metal sigma bond.
# 2003 Elsevier Science B.V. All rights reserved.
Keywords: Cobaloxime; Cobalt carbonyl; Trimetallic complex; Pauson-Khand; Multinuclear complex
1. Introduction
cursors [7b] and their ability to stabilise carbocations as
the cobaloximato p cation (Scheme 2) [7c]. Very
recently, new interest in cobaloxime chemistry has
been generated by the ability of both B12 and cobalox-
imes to dehalogenate priority pollutants such as tetra-
chloroethene [7d].
Although multinuclear systems are of great current
interest [8], no system has yet been reported in which
these two very different classes of organocobalt species
are combined in the same molecule and this paper
reports the first studies of such compounds. It is possible
Metal alkyne and alkynyl [1a] complexes including
dicobalt hexacarbonylꢀalkyne [1b] complexes are of
/
interest due to both their many potentially useful
properties and applications [1a,2] and their applications
in synthetic chemistry, notably the Pauson-Khand
reaction [3] and the Nicholas reaction [4] in which a
Lewis acid removes an oxy-group from the propargyl
position of a dicobalt hexacarbonyl alkyne complex to
give a cobalt-stabilised cation, which can then be
trapped with nucleophiles (Scheme 1). Often in synthesis
this reaction is followed by oxidative demetallation of
the alkyne to give overall propargylic substitution of
oxy-alkynes. Also of interest are alkyl cobaloximes,
that combined cobaloximeꢀdicobalt hexacarbonyl sys-
/
tems could provide useful routes to cyclic organometal-
lics via the Pauson-Khand reaction, or that cations
stabilised by both cobalt-containing units could be
isolated or applied as synthetic intermediates. We were
particularly intrigued as to whether these functional
groups could be connected via a sigma-bonded frame-
work with the alkynyl functionality, and hence the
dicobalt hexacarbonyl complex, pendant from the
cobaloxime cobalt, as opposed to through the axial
pyridine ligand in the cobaloxime unit. The various
activating effects of the cobaloxime group and the
dicobalt hexacarbonyl would thus be acting upon the
same portion of a methylene chain, and this may either
lead to useful reactivity or instability of the complexes.
which have been used as model systems for the carbonÃ
/
cobalt bond of coenzyme vitamin B12 [5], as catalysts in
living radical polymerisation [6] and in synthetic chem-
istry [7] due to both their properties as radical pre-
* Corresponding author. Tel.: ꢁ
874030.
E-mail address: cooganmp@cf.ac.uk (M.P. Coogan).
/
44-2920-874066; fax: ꢁ44-2920-
/
1
Current address: Technische Universitat Clausthal, Institut fur
¨
¨
Organische Chemie, Leibnizstrasse 6, 38678 Clausthal-Zellerfeld,
Germany.
0022-328X/03/$ - see front matter # 2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0022-328X(03)00385-1

126
M.P. Coogan et al. / Journal of Organometallic Chemistry 677 (2003) 125ꢀ132
/
Scheme 1.
Although single crystals suitable for X-ray diffraction
have not been obtained, NMR and MS indicate that this
is the correct structure as the mass of 410 and the proton
and carbon NMR indicates reduction of the alkyne to
an alkene. Additionally, DEPT confirms the presence of
a methyl group indicating reduction of the propargyl
unit. It is not obvious how this material is formed and it
differs from the literature precedent in which propargyl
chlorides are reported to give mixtures of the propargyl
and allenyl products [10]. The reduction of the propar-
gyl unit to give alkenyl products is explicable by the use
of sodium borohydride in the synthesis of the cobalox-
imato nucleophiles, and it seemed possible that excess
borohydride reduces one of the expected products A or
B which, along with reduction, rearranges to give a 2-
cobaltopropene system. Although the standard condi-
tions [9] for alkylation of cobaloximes which were
followed in this work include the addition of a second
portion of borohydride after the cobaloximato-nucleo-
phile has been treated with alkyl halide, and this step is
an obvious culprit for over-reduction, omission of this
second addition, in fact, led to higher yields of 4. In
order to probe the mechanism of formation of this
unexpected product, a reaction was performed between
the cobaloximato nucleophile and propargyl bromide
under standard conditions but with the addition of
excess acetone 15 min after the addition of propargyl
bromide, in order to destroy excess borohydride, quench
Scheme 2.
Two possible approaches to these combined systems
are available: the displacement of a leaving group from a
preformed dicobalt hexacarbonyl complex, and the
treatment of an alkynyl cobaloxime with dicobalt
octacarbonyl. As it is known that treatment of dicobalt
hexacarbonyl complexes with organometallic nucleo-
philes can displace cobalt from the complex [8c], the
latter approach was chosen.
2. Results and discussion
2.1. Synthesis and characterisation
Treatment of propargyl bromide with the cobalox-
imato-nucleophile (3) prepared under standard condi-
tions [9] gave not the expected propargyl cobaloxime A,
nor the allene B which might be expected from an SN2?
mechanism, but a product which we have tentatively
assigned as 2-propenylcobaloxime (4) (Scheme 3).
Scheme 3.

M.P. Coogan et al. / Journal of Organometallic Chemistry 677 (2003) 125ꢀ
/132
127
any reductive rearrangements, and allow the isolation of
intermediates A or B. Two products which were
tentatively assigned as A and B were indeed observed
in the crude 1H-NMR of the products of these reactions.
In addition to the signals assigned to pyridine and
DMGH units, the characteristic signals which allowed
assignment as A are a doublet of integration 2H at d 3.9
ppm and a triplet of integration 1H at d 2.2 ppm with a
mutual coupling of 2.5 Hz, typical of a propargyl
system, similarly B was assigned on the basis of a
doublet of integration 2H at d 5.4 ppm and a triplet of
integration 1H at d 4.8 ppm with a mutual coupling of
5.9 Hz, typical of a mono-substituted allene. Although
this crude product was a reasonably clean mixture of
these two species, each of them appeared to be unstable
even in the absence of reductant, and attempts to isolate
either of these putative intermediates failed. Although
mechanistic studies are ongoing, a tentative proposal for
the formation of 4 is illustrated in Scheme 3. In this,
both A and B are converted a common intermediate
cobaloxime p cation C by protonation of the p system
by the medium, which in each case involves migration of
cobalt, mirroring the mechanisms precedented in
Scheme 2 [7c,9]. This intermediate C is then reduced
by borohydride with addition of hydride at the least
substituted carbon to give 4.
Fig. 1. ORTEP view of But-3-yn-1-ylcobaloxime 1.
significant broadening of the signals attributed to the
protons attached to these carbons is observed (see Fig.
1).
It might have been expected that the stability of a
cobaloxime p cation [7,9] and of the alkynyl anion
would lead to a lengthening of the methylene to
quaternary CÃ
/
C bond (Scheme 6), but in fact this
˚
˚
˚
length is 1.477 A compared to 1.470 A 12 to 1.475 A [13]
in some of the few examples of crystallographically
characterised hex-1-ynes (in which the cobalt of 1 is
replaced by a methylene unit) indicating that this
putative resonance form 6 makes little if no contribu-
tion. But-3-yn-1-yl cobaloxime (1) is stable to moderate
heat (decomposition occurs under prolonged heating
greater than 80 8C), air and water, but is slowly
degraded in bright light. Treatment of 1 with dicobalt
octacarbonyl in toluene or DCM (degassed) leads to the
immediate formation of a deep red solution and the
evolution of a gas (assumed to be carbon monoxide)
(Scheme 5). NMR of the product obtained upon
pumping down these solutions indicate that the dicobalt
hexacarbonyl alkyne complex has formed as the alkyne
proton (d 1.87 ppm CDCl3) disappears and is replaced
by a broad signal at d 5.88 ppm (CDCl3) characteristic
of the complex. The reaction is essentially quantitative;
however, the crude product is always contaminated by a
small amount of black insoluble material assumed to be
metallic cobalt which can be removed by filtration of a
toluene solution, followed by precipitation of the
product as very dark red to black crystals (68%) with
Given the lack of success with propargyl bromide,
attention was turned to the homopropargyl system.
Treatment of but-3-yn-1-ol tosylate (5) [11] with the
sodium cobaloximato nucleophile prepared under stan-
dard conditions [9] gave the required but-3-yn-1-ylco-
baloxime (1) in a reasonable yield (Scheme 4). The air
and water stability of these organometallics is high-
lighted by the fact that crystals suitable for X-ray
diffraction were grown from a methanolꢀwater mixture
/
under ambient atmosphere. To the best of our knowl-
edge, this is the first crystallographically characterised
alkyne-containing cobaloxime, although such species
have been invoked previously. The NMR of this and
subsequent cobaloximes show signals corresponding to
the methylene unit directly attached to cobalt in the 1H-
NMR in the typical methylene envelope (in this case
dHꢂ
1.56); however, the interpretation of 13C-NMR is
/
complicated by broadening of the signal corresponding
to the carbon directly attached to the cobalt atom. In
many cases, this signal is difficult to observe and
appears as such a weak broad signal that it can only
be confidently assigned by the use of Hꢀ13
1
/
C correla-
tion. This broadening is presumably a result of coupling
60ꢀ
/
80% petroleum ether (see Fig. 2).
to the highly quadrupolar 59Co nucleus; however, no
Scheme 4.
Scheme 5.

128
M.P. Coogan et al. / Journal of Organometallic Chemistry 677 (2003) 125ꢀ132
/
atom-transfer oxidant has been used to promote the
Pauson-Khand reaction, and attempts to promote the
Pauson-Khand reactions of 1 with NMO lead to the
formation of a mixture of demetallated products,
demonstrating the limit to which the cobaloxime unit
may be exposed to oxidants. The Pauson-Khand may
also be thermally promoted; however, heating alkyl
Scheme 6.
Again, this tricobalt organometallic species is reason-
ably stable to air and water, seeming to be oxidised
slowly in solution but, again, it could be crystallised
from aqueous alcoholic solution providing that oxygen
cobaloximes homolyses the cobaltÃ
/
carbon bond and an
attempt to run the reaction at 100 8C in toluene led to
decomposition to an intractable black tar. Cobaloxime
homolysis is an effective method for initialising alkene
was excluded. Crystallisation from toluene/60ꢀ
petroleum ether gave single crystals of dicobalt
hexacarbonylꢀbut-3-yn-1-yl cobaloxime which were
suitable for X-ray diffraction studies. Again, all bond
lengths (CÃC, CÃCo, CoÃCo) are consistent with the
small number [14] of examples of acyclic dicobalt
hexacarbonylꢀbut-1-ynyl complexes which have been
/
80%
polymerisation, and it is likely that the alkyneꢀalkene
/
mixture which represents the Pauson-Khand reaction is
highly sensitive to the alkyl-cobaloxime/alkyl radical
couple generated from these species at high temperature.
The yields represent the optimum result of a range of
temperature/solvent combinations which were investi-
gated, and the balance between driving the Pauson-
Khand reaction to completion and formation of by-
products was best at reflux in dichloromethane.
Although the yields are thus lower than idea, they
represent an unpromoted Pauson-Khand reaction of an
/
/
/
/
/
crystallographically characterised indicating that there is
no significant contribution from a p cobaloxime cation.
2.2. Reactions
alkyneꢀdicobalthexacarbonyl complex which is rela-
tively stable and thus may be expected to react
sluggishly.
/
Having established that the two forms of cobaltÃ
/
carbon species are compatible and can exist in the
same molecule (indeed show remarkable stability), it
was of interest to see if the reactivity of the dicobalt
hexacarbonyl unit was retained in the presence of
cobaloxime substitution. The substrate in hand was
considered suitable for a Pauson-Khand [3] reaction and
indeed this proceeded cleanly with strained bicyclic
alkenes in moderate yields (Scheme 7). These yields
compare disfavourably with the literature equivalents
for metal-free systems, but in many cases these reactions
are not strictly comparable as NMO [3c] or a similar
The complex tricyclic products of the Pauson-Khand
reactions were studied by 1D and 2D NMR and all the
peaks unambiguously assigned by application of
gCOSY, HSQC, HMBC and NOSY spectroscopy.
Generally, the assignment of spectral peaks of cobalox-
imes may be complicated by difficulties in observing the
carbon (and occasionally its attached hydrogens) di-
rectly attached to cobalt due to quadrupolar broadening
from 59Co.
Fig. 2. ORTEP view of dicobalt hexacarbonyl but-3-yn-1-ylcobaloxime 2.

M.P. Coogan et al. / Journal of Organometallic Chemistry 677 (2003) 125ꢀ
/132
129
Scheme 7.
3. Conclusion
In conclusion, the first combined cobaloximeꢀ
Table 1
Crystal data and experimental details for 1, orange block, stable under
ambient conditions collected on a Nonius KappaCCD solved using
direct methods [15]
/
dico-
balt hexacarbonyl alkyne complex has been prepared,
and it has been shown that these remarkably stable
organotrimetallics retain the synthetically useful chem-
istry associated with the Pauson-Khand reaction to give
organometallic products. Studies are ongoing into
reducing the cobaloxime to alkyne connectivity and
into other sigma-bonded organometallic species suitable
for use in the Pauson-Khand reaction.
Formula
C17H24CoN5O4
421
293
Molecular weight
Temperature (K)
Crystal system
Space group
Monoclinic
P2(1)/C
0.71073
˚
l (A)
Unit cell dimensions
˚
a (A)
˚
9.104(2)
14.491(3)
14.516(3)
90
b (A)
˚
c (A)
a (8)
b (8)
4. Experimental
90.60(3)
90
g (8)
3
V (A )
˚
1908.9(7)
4
0.932
4.1. Crystal data
Z
Absorption coefficient (mmꢃ1
Crystal size (mm3)
u Limits (8)
)
Crystallographic data for compounds 1 and 2 have
been given in Tables 1 and 2, respectively.
0.2ꢄ
/
0.2ꢄ/0.2
3.08ꢀ/27.52
Reflections collected
25112
4367
Unique observed reflections [Foꢀ4sFo]
/
Rint
Goodness-of-fit on F2
0.0980
1.110
4.1.1. General
NMR spectra were recorded on a Bruker DPX 400
MHz at 399.9 MHz (proton) and 100.0 MHz (carbon)
and on an INOVA-500 MHz at 499.9 MHz (proton)
and 125.7 MHz (carbon). Unless otherwise stated NMR
spectra were recorded in CDCl3. IR spectra were
Final R indices [I ꢀ2s(I)]
/
R1
wR2(F)2
0.0428
0.1052
All data
R1
wR2(F)2
0.0580
0.1118
recorded on a PerkinꢀElmer 1600 FT-IR: mass spectra
/
were recorded on a VG Fisons Platform II or at the
EPSRC national mass spectrometry service in Swansea
(HRMS). Unless otherwise specified, all chemicals and
reagents were commercial materials used without
further purification. Solvents for reactions involving
dicobalt octacarbonyl were dried with molecular sieves
loximes 7 and 8 are less stable than their alkynyl
precursor 1.
4.1.2. But-3-yn-1 p-toluenesulphonate (5) [11]
and degassed by three freezeꢀ
/
pumpꢀ
/
thaw cycles prior
To DCM (260 ml) at 0 8C was added triethylamine
(11.2 ml, 80 mmol), then 3-butyn-1-ol (3.9 ml, 50 mmol).
p-TosCl (10.7 g, 55 mmol) was added as a solid in
portions during a period of 5 min. The reaction was
stirred at ambient temperature for 60 h and then
quenched with ice water (30 ml). The organic layer
was separated and treated with aqueous hydrochloric
to use. Exposure of cobaloximes to light was minimised
where possible. A satisfactory analysis of 4 was never
obtained. Although 4 could be crystallised from metha-
nol to give material which was pure by NMR, this
homogeneity was rapidly lost and unidentified decom-
position products appeared. Similarly, homoallyl coba-

130
M.P. Coogan et al. / Journal of Organometallic Chemistry 677 (2003) 125ꢀ132
/
Table 2
one portion and the mixture was stirred for one more
hour. The reaction mixture was adsorbed onto silica (2
g). This brown powder was put on top of a plug of fresh
silica (20 g). This was rinsed with petroleum ether (200
ml) and with ethyl acetate (500 ml) to give an orange
solution. Evaporation and crystallisation from methanol
gave an orange powder (0.2 g, 12%) (Rf 0.16 ether) (4).
Crystal data and experimental details for 2, dark red needle, stable
under ambient conditions, collected on Nonius KappaCCD solved
using direct methods [15]
Formula
C23H24Co3N5O10
707
Molecular weight
Temperature (K)
Crystal system
Space group
150
triclinic
¯
P1
m.p. 180 8C (decomp.) dHꢂ
Hz, 2H, pyrÃH2,6) 7.67 (m, 1H, pyrÃ
pyrÃH3,5), 4.55 (s, 1H, HHCÄC), 4.21 (s, 1H, HHCÄ
C), 2.04 (s, 12H, 4ꢄ DMGH CH3), 1.55 (s, 3H, CH3)
ppm; dCꢂ150.3, 150.1 (both NÄC, DMGH, pyrÃ
C2,6), 138.0 (pyrÃC4), 126.3 (very weak, CÃCo(CH3)),
125.3 (pyrÃC3,5), 115.5 (H2CC), 29.6 (CH3), 13.6
(DMGH CH3) ppm. DEPT confirms assignments as
/
8.57 (dd, Jꢂ/1.4 and 6.4
/
˚
l (A)
0.71073
/
/H4), 7.28 (m, 2H,
Unit cell dimensions
˚
/
/
/
a (A)
8.9732(2)
11.2696(3)
15.4905
/
˚
b (A)
˚
c (A)
/
/
/
a (8)
b (8)
69.5420(10)
76.8790(10)
84.090(2)
1428.87(7)
2
/
/
/
g (8)
3
V (A )
˚
CH, CH3; m/z (ES) 410 [Mꢁ
/
Hꢁ],nmax(Nujol)ꢂ
3407
/
Z
(broad, weak) 2865, 1558, 1237 cmꢃ1
.
Absorption coefficient (mmꢃ1
Crystal size (mm3)
u Limits (8)
)
1.782
0.2ꢄ
/
0.1ꢄ0.05
/
2.98 26.00
23589
Reflections collected
Unique observed reflections [Foꢀ
Rint
Goodness-of-fit on F2
/4sFo]
5562
0.0844
1.104
4.1.4. But-3-yn-1-yl cobaloxime (1)
Cobalt(II) chloride hexahydrate (0.96 g, 4.03 mmol),
dimethylglyoxime (0.92 g, 7.92 mmol), aqueous NaOH
(50%, 0.585 g, 14.63 mmol) and pyridine (0.34 ml, 4.2
mmol) were each added successively in one portion to
MeOH (22 ml). The mixture was deoxygenated for 10
min by bubbling N2 and then cooled to 0 8C. Sodium
borohydride (0.215 g, 5.68 mmol) was added in two
aliquots within 1 min and the mixture stirred for 10 min.
5 (0.94 g, 4.2 mmol) was added and the mixture was
stirred for 3 h at 0 8C. The reaction mixture was
adsorbed onto silica (3.5 g). This brown powder was
put on top of a plug of fresh silica (35 g). This was
washed with petroleum ether (100 ml), petroleum ether/
ethyl acetate (1:1) (500 ml) and with ethyl acetate (500
ml) to give an orange solution. Evaporation of the ethyl
acetate portion followed by crystallisation from metha-
nol (40 ml MeOH per 1 g raw product) gave an orange
Final R indices [I ꢀ2s(I)]
R1
wR2(F)2
/
0.0539
0.1412
All data
R1
wR2(F)2
0.0672
0.1544
acid (1 M, 30 ml), sodium bicarbonate (30 ml) and with
brine (30 ml), then dried over magnesium sulphate and
evaporated to give the 3-butyn-1-o-tosylate (4) as a pale
yellow liquid (11.2 g, 87.5%) of sufficient purity to be
used in further steps. dHꢂ
H2,6), 7.3 (d, Jꢂ8.0 Hz, 2H, ArÃ
Hz, 2H, CH2CH2O), 2.49 (dt, Jꢂ
CH2CH2O), 2.38 (s, 3H, CH3), 1.91 (t, Jꢂ
/
7.73 (d, Jꢂ
H3,5), 4.03 (t, Jꢂ
7.0 and 2.7 Hz,
2.7 Hz, 1H,
/
8.0 Hz, 2H, ArÃ
/
/
/
/
7.0
/
/
HCC) ppm [11].
powder (Rf 0.45 EtOAc) (0.918 g, 54%) (1). m.p. 225ꢀ
230 8C (decomp), dHꢂ8.49 (dd, Jꢂ1.3 and 6.1 Hz,
2H, pyrÃH2,6), 7.67 (m, 1H, pyrÃH4), 7.26 (m, 2H,
pyrÃH3,5), 2.06 (s, 12H, 4ꢄ CH3 DMGH), 1.87 (t, Jꢂ
2.5 Hz, 1H, HCÅC), 1.71 (dt, Jꢂ2.5 and 8.7 Hz, 2H,
HCCCH2), 1.56 (t, Jꢂ8.7 Hz, 2H, CH2Co) ppm; dCꢂ
150.3 (MeCÄN, ligand), 150.1 (pyrÃC2,6), 138.0 (pyrÃ
C4), 125.7 (pyrÃC3,5), 85.3 (HCC), 67.3 (HCÅC), 27.5
(weak, broad, CH2Co), 18.9 (propargy CH2), 12.5 (CH3,
DMGH) ppm; assignments CH, CH2, CH3 confirmed
by DEPT other than d 27.5 ppm (weak, broad, CH2Co),
which was not observed by DEPT, assignment of this
/
/
/
/
/
4.1.3. Prop-1-enyl-2-cobaloxime (4)
/
/
/
All work was done under dry nitrogen and with
degassed solvent. Cobalt(II) chloride hexahydrate (0.96
g, 4.03 mmol), dimethylglyoxime (0.92 g, 7.92 mmol),
aqueous NaOH (50%, 0.585 g, 14.63 mmol) and
pyridine (0.34 ml, 4.2 mmol) were added successively
each in one portion to MeOH (22 ml). The mixture was
deoxygenated for 10 min by bubbling N2 and then
cooled to 0 8C. Sodium borohydride (0.215 g, 5.68
mmol) was added in two aliquots within 1 min and the
mixture stirred for 10 min. Progargyl bromide (80%
solution in toluene, 0.45 ml, 4.2 mmol) was added via
syringe in one portion and the mixture was stirred for 2
h while it attained ambient temperature. Additional
sodium borohydride (0.067 g, 1.77 mmol) was added in
/
/
/
/
/
/
/
/
/
peak was confirmed by 13Cꢀ1
methylene signal at dH 1.56 ppm,nmax(Nujol)ꢂ
2086 (weak), 1558, 1227 cmꢃ1; m/z (ES) 422 [Mꢁ
/
H correlation to the
3236,
Hꢁ];
/
/
C17H24N5O4Co requires C, 48.46; H, 5.74; N, 16.62;
Found: C, 48.33; H, 5.72; N, 16.55%.

M.P. Coogan et al. / Journal of Organometallic Chemistry 677 (2003) 125ꢀ
/132
131
4.1.5. But-3-yn-1-yl cobaloxime dicobalt hexacarbonyl
(2)
by typical and distinct d value 210. gCOSY, HSQC and
HMBC from these signals allowed assignment of 6-17,
1
H12,13, end/exo differentiated by Hendo, NOESY to
To a solution of 1 (125 mg, 0.3 mmol) in dry, degassed
toluene (10 ml) was added dicobalt octacarbonyl (105
mg, 0.3 mmol) as a solid in one portion. The mixture
was stirred under nitrogen for 3 h during which time
cobalt carbonyl dissolved, evolution of carbon mon-
oxide ceased and the solution turned a deep red colour.
H15 and H13exo NOESY to H14. H17a and H17b
distinguished by stronger correlation in gCOSY between
H17a and H12,13endo, attributed to W-like relationship
(see Fig. 3).
Evaporation followed by crystallisation (toluene/60ꢀ
80% petroleum ether) gave red-black needles (144 mg,
68%) (2) m.p. ꢀ300 8C. dHꢂ8.53 (d, Jꢂ4.9 Hz, 2H,
pyrÃH2,6), 7.66 (m, 1H, pyrÃH4), 7.26 (m, 2H, pyrÃ
H3,5), 5.88 (br s, 1H, H4), 2.4 (t, Jꢂ7.9 Hz, 2H, H2),
2.07 (s, 12H, 4ꢄ DMGH CH3), 1.59 (t, Jꢂ7.9 Hz, 2H,
H1) ppm; dC (CD3CN) 198.5 (very broad, weak CO),
149.1, 148.7 (DMGH CÄN, PyÃC2,6), 137.2 (PyÃC3,5),
134.0 (PyÃC4), 32.5 (weak, CH2Co), 10.2 (CH2CH2Co)
ppm; m/z (FAB) 708 [MꢁH], 629 [MꢁHꢃpyr], 573
[MꢁHꢃpyrꢃ2CO], 545 [MꢁHꢃpyrꢃ3CO];
nmax(Nujol)ꢂ3410 (broad weak), 2855, 2081, 2031,
/
4.1.7. Tricyclo[4.3.12,5]-8-en-8-ethylcobaloxim-7-
decanone (8)
/
/
/
1 (0.5 g, 1.19 mmol) was dissolved in DCM to which
dicobalt octacarbonyl (0.44 g, 1.29 mmol) was added in
one portion and the solution was stirred for 30 min
during which time evolution of carbon monoxide ceased
and a deep red colour developed. Norbornadiene (0.6
ml, 5.54 mmol) was then added in one portion and the
reaction mixture was heated to reflux for 3 h. The
reaction mixture was adsorbed onto 2.5 g of silica.
Column chromatography (35 g of silica, petrol ether/
ethyl acetate (1:3), then ethyl acetate) gave a brown
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
1980 1558 1227 cmꢃ1; C23H24N5O10Co3 requires C,
39.06; H, 3.42; N, 9.90; Found: C, 39.33; H, 3.37; N,
9.71%.
powder (Rf 0.33 EtOAc) (8) (278 mg, 43%). dHꢂ
/
8.55
(d, Jꢂ5.0 Hz, 2H, H1), 7.68 (t, Jꢂ7.6 Hz, 1H, H2),
/
/
7.31 (m, 2H, H3), 7.03 (s, 1H, H16), 6.16 (m, 1H, H13),
6.08 (m, 1H, H12), 2.80 (s, 1H, H11), 2.52 (s, 2H,
4.1.6. Tricyclo[4.3.12,5]-8-en-8-ethylcobaloxim-7-
decanone
H14,15), 2.07 (s, 13H, 12ꢄ
H7), 1.54 (m, 2H, H6), 1.23 (d, Jꢂ
1.09 (d, Jꢂ9.2 1H, 17a) ppm; dCꢂ
/
H5ꢁ
/
H10), 1.68 (m, 2H,
9.2 Hz, 1H, H17b),
209.3 (C9), 157.9
/
7 (0.5 g, 1.19 mmol) was dissolved in DCM to which
dicobalt octacarbonyl (0.44 g, 1.29 mmol) was added in
one portion and the solution was stirred for 30 min
during which time evolution of carbon monoxide ceased
and a deep red colour developed. Norbornylene (0.51
mg, 5.43 mmol) was added in one portion and the
reaction mixture was heated to reflux for 3 h. The
reaction mixture was adsorbed onto 2.5 g of silica.
Column chromatography (35 g of silica, petrol ether/
ethyl acetate (1:3), then ethyl acetate) gave a brown
/
/
(C16), 150.8 (C8), 149.8 (C4), 149.4 (C1), 138.3 (C3),
137.4 (C13), 136.9 (C12), 125.1 (C2), 52.3 (C10), 47.4
(C15), 43.4 (C11), 42.8 (C14), 41.2 (C17), 26.5 (broad,
weak C6), 25.0 (C7), 12.0 (C5) ppm; nmax(Nujol)ꢂ
(broad weak), 2855, 1694, 1599, 1558 cmꢃ1; m/z (ES)
542 [Mꢁ
Hꢁ]. Crystallisation from MeOH/H2O gave a
brown powder. m.p. 119ꢀ121 8C. C25H34CoN5O5 (M×
/
3467
/
/
/
H2O) requires C, 55.25; H, 6.31; N, 12.89; Found: C,
54.95; H, 6.31; N, 12.72%.
powder (Rf 0.3 EtOAc) (7) (245 mg, 38%). dHꢂ
/8.58 (d,
Assignments arrived at from 1Hꢀ1H correlation
/
Jꢂ5.0 Hz, 2H, H1), 7.73 (t, Jꢂ7.5 Hz, 1H, H3), 7.32
/
/
working from uniquely identifiable 1H H16 and distinct
pair H12,13 with pairs H12,13 H11,14 H 10,15 differ-
entiated by NOESY and gCOSY correlations between
H16 H15 and H13. H17a and H17b distinguished by
stronger correlation in gCOSY between H17b and H15,
H10, attributed to W-like relationship (see Fig. 4).
(m, 2H, H2), 7.04 (s, 1H, H16), 2.45 (s, 1H, H15), 2.30
(s, 1H, H11), 2.14 (s, 12H, H5), 2.09 (m, 2H, H14, H10),
1.73 (m, 2H, H7), 1.61 (m, 4H, 2ꢄ
H13exo), 1.23 (m, 2H, H12endo, H13endo), 0.96 (d,
Jꢂ9.3 Hz, 1H, H17a), 0.88 (d, Jꢂ9.3 Hz, 1H, H17b)
ppm; dCꢂ210.6 (C9), 157.9 (C16), 149.8 (C1), 149.4,
/
H6, H12exo,
/
/
/
149.3 (C4, C8), 137.4 (C3), 152.1 (C2), 53.6 (C10), 48.0
(C15), 38.8 (C11), 37.9 (C14), 31.0 (C17), 29.0 (C12),
28.4 (C13), 27.4 (broad, weak C6), 24.7 (C7), 12.0 (C5)
ppm; m/z (ES)ꢂ
1689, 1558, 1232 cmꢃ1. Crystallisation from MeOH/
H2O gave brown powder. C25H38CoN5O6 (M×2H2O)
/
544 [Mꢁ
/
Hꢁ];nmax(Nujol)ꢂ
/
2855,
/
requires C, 53.28; H, 6.79; N, 12.43; Found: C, 52.97; H,
6.25; N 12.22%.
Assignments were arrived at from 1Hꢀ1H correlation
/
1
working from uniquely identifiable H H16, methylene
6CH2 (identified by HSQC correlation with broad 13C
signal showing adjacent 59Co C6) and 13C C9 identified
Fig. 3. Numbering for tricyclo[4.3.12,5]-8-en-8-ethylcobaloxim-7-de-
canone.

132
M.P. Coogan et al. / Journal of Organometallic Chemistry 677 (2003) 125ꢀ
/132
(2003) 1;
(b) H. Greenfield, H.W. Sternburg, R.A. Friedel, J.H. Wotiz, R.
Markby, I. Wender, J. Am. Chem. Soc. 78 (1956) 120;
(c) R.S. Dickson, P.F. Fraser, Adv. Organomet. Chem. 12 (1974)
323.
[2] C. Moreno, M.-L. Marcos, G. Dominguez, A. Arnanz, D.H.
Farrar, R. Teeple, A. Lough, J. Gonzalez-Velasco, S. Delago, J.
Organomet. Chem. 631 (2001) 19.
[3] (a) C.S. Chin, G. Won, D.S. Chong, Acc. Chem. Res. 35 (2002)
218;
(b) P.L. Pauson, Tetrahedron 41 (1985) 5855;
(c) S. Shambayati, W.E. Crow, S.L. Schreiber, Tetrahedron Lett.
31 (1990) 5289;
Fig. 4. Numbering for tricyclo[4.3.12,5]-8-en-8-ethylcobaloxim-7-de-
canone.
(d) S.E. Gibson, A. Stevenazzi, Angew. Chem. Int. Ed. Eng. 42
(2003) 1800.
[4] K.M. Nicholas, Acc. Chem. Res. 20 (1987) 207.
[5] (a) P.G. Lenhert, D.C. Hodgin, Nature 192 (1961) 937;
(b) B.T. Golding, Chem. Br. 26 (1990) 950;
(c) G.N. Schrauzer, Acc. Chem. Res. 1 (1968) 97.
[6] K.G. Suddaby, D.R. Maloney, D.M. Haddleton, Macromole-
cules 30 (1997) 702.
5. Supplementary material
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallography
Data Centre, CCDC No. 194766 for compound 1 and
CCDC No. 194765 for compound 2. Copies of this
information may be obtained free of charge from The
Director, CCDC, 12 Union Road, Cambridge CB2 1EZ,
[7] (a) M.E. Welker, Curr. Org. Chem. 5 (2001) 785;
(b) D.C. Harrowven, G. Pattenden, Tetrahedron Lett. 32 (1991)
234;
(c) L.M. Grubb, B.P. Branchaud, J. Org. Chem. 62 (1997) 242;
(d) K.M. McCauley, S.R. Wilson, W.A. van der Donk, Inorg.
Chem. 41 (2002) 393.
UK (Fax: ꢁ44-1223-336033; e-mail: deposit@ccdc.
/
cam.ac.uk or www: http://www.ccdc.cam.ac.uk).
[8] (a) N.C. Harden, E.R. Humphrey, J.C. Jeffery, S.-M. Lee, M.
Marcaccio, J.A. McCleverty, L.H. Rees, M.D. Ward, J. Chem.
Soc. Dalton Trans. (1999) 2417;
Acknowledgements
(b) P.J. Low, K.A. Udachin, G.D. Enright, A.J. Carty, J.
Organomet. Chem. 578 (1999) 103;
(c) D.T. Rutherford, S.D.R. Christie, Tetrahedron Lett. 39 (1998)
9805;
We thank the EPSRC National Mass Spectrometry
Centre for mass spectra, Dr. Alan Kenwright and
Catherine Heffernan (University of Durham) for 500
MHz NMR and helpful discussions on NMR issues, the
ERASMUS program (T.W.) and the EPSRC GR/
R06458 for the Nonius-Kappa CCD X ray-crystal-
lography facility on which the data sets were collected.
(d) P.J. Low, M.I. Bruce, Adv. Organomet. Chem. 48 (2002) 71.
[9] J.L. Gage, B.P. Branchaud, J. Org. Chem. 61 (1996) 831.
[10] J.P. Collman, J.N. Cawse, J.W. Kang, Inorg. Chem. 8 (1969)
2574.
[11] A. Furstner, K. Langmann, Synthesis 7 (1997) 792.
[12] W. Stolle, A.E. Frissen, A.T.M. Marcelis, H.C. van der Plas, Y.
Wang, L. Haming, C.H. Stam, J. Org. Chem. 56 (1991) 2411.
[13] J.H. Rigby, C.R. Heap, N.C. Warshakoon, M.J. Heeg, Org. Lett.
1 (1991) 507.
References
[14] X.-N. Chen, J. Zhang, Y.-Q. Yin, X.-Y. Huang, Organometallics
18 (1999) 3164.
[15] G.M. Sheldrick, SHELXS-97, SHELXL-97, University of Gottingen,
[1] (a) For a special issue addressed to such complexes, their
properties and applications, see J. Organomet. Chem. 670
¨
Lower Saxony, Germany, 1997.