Activation of alkynols on transition metal carbonyl clusters. Reactions of 1-ethynylcyclohexanol with Fe3(CO)12 and Co2(CO)8

Article abstract of DOI:10.1039/b101119g

The reactions of Co₂(CO)₈ with 1-ethynylcyclohexanol (HC≡CC₆H₁₀OH, ECY) in benzene led mainly to Co₂(CO)₆(ECY) 1 and to smaller amounts of the methylidyne complex Co₃(CO)₉[μ₃-CCH₂(C₆H ₁₀OH)] 2. The reactions of Fe₃(CO)₁₂ with ECY in the same solvent led to binuclear metallacyclic derivatives Fe₂(CO)₆(ECY)₂ (isomers 3a, 3b) or Fe₂(CO)₆[(ECY)₂ - H₂O] 3c as the main products; small yields of trinuclear complexes Fe₃(CO)₉(μ-CO)[μ-η²-1,2-HC≡C(C ₆H₁₀OH)] 4, Fe₃(CO)₉(μ-CO)[μ₃-η²-1,2-C =C(C₆H₁₀)] 5 and Fe₃(CO)₇-[μ₃-η⁷-(C₆ H₁₀OH)CCHCHC(C₆H₉)] 6 {containing respectively a parallel alkynol, an allenylidene and a dimeric metallacyclic ligand} were also obtained, together with 7, a thermal decomposition product of 6. Finally, when Co₂(CO)₈ was treated with ECY in benzene, and Fe₃(CO)₁₂ was added, the heterometallic complex Co₂Fe(CO)₆-(μ-CO)[μ₃-η⁷-(C ₆H₉)CC(H)C(H)C(H)(C₆H₁₀)] 8 was obtained in low yields. The complexes have been characterized by means of IR and ¹H NMR spectroscopies and by mass spectrometry. The structures of 2, 5, 8 have been determined by X-ray diffraction. Complex 2 contains an hydrogenated ECY ligand, 5 an allenylidene ligand formed upon dehydration of ECY (loss of the OH and of the terminal hydrogen), whereas 8 contains a ligand formed by tail-to-tail coupling of two ECY molecules, with loss of water and oxygen and shift of hydrogen. The elemental analysis of complex 8 gave a Co : Fe ratio of 2 : 1; on the basis of this analysis, the refinement of the diffraction data allowed a hypothesis on the distribution of the metal atoms in the cluster. Reaction pathways for the formation of these clusters are proposed and dehydration mechanisms for the ligand discussed.

Full text of DOI:10.1039/b101119g

View Article Online / Journal Homepage / Table of Contents for this issue
Activation of alkynols on transition metal carbonyl clusters.
Reactions of 1-ethynylcyclohexanol with Fe3(CO)12 and Co2(CO)8.
Crystal structures of Co3(CO)9[ꢀ3-CCH2(C6H10OH)], Fe3(CO)9-
(ꢀ-CO)[ꢀ -ꢁ2-C᎐C(C H )] and Co Fe(CO) (ꢀ-CO)[ꢀ -ꢁ7-(C H )-
᎐
3
6
10
2
6
3
6
9
CC(H)C(H)C(H)(C6H10)]
Elisabetta Gatto,a Giuliana Gervasio,a Domenica Marabello*a and Enrico Sappab
a Dipartimento di Chimica IFM, Università di Torino, Via Pietro Giuria 7, I-10125 Torino, Italy
b Dipartimento di Scienze e Tecnologie Avanzate, Università del Piemonte Orientale,
C.so Borsalino 54, I-15100 Alessandria, Italy
Received 2nd February 2001, Accepted 7th March 2001
First published as an Advance Article on the web 4th April 2001
᎐
The reactions of Co (CO) with 1-ethynylcyclohexanol (HC᎐CC H OH, ECY) in benzene led mainly to
᎐
2
8
6
10
Co2(CO)6(ECY) 1 and to smaller amounts of the methylidyne complex Co3(CO)9[µ3-CCH2(C6H10OH)] 2. The
reactions of Fe3(CO)12 with ECY in the same solvent led to binuclear metallacyclic derivatives Fe2(CO)6(ECY)2
(isomers 3a, 3b) or Fe2(CO)6[(ECY)2 Ϫ H2O] 3c as the main products; small yields of trinuclear complexes
2
2
᎐
Fe (CO) (µ-CO)[µ -η -1,2-HC᎐C(C H OH)] 4, Fe (CO) (µ-CO)[µ -η -1,2-C᎐C(C H )] 5 and Fe (CO) -
᎐
᎐
3
9
3
6
10
3
9
3
6
10
3
7
[µ3-η7-(C6H10OH)CCHCHC(C6H9)] 6 {containing respectively a parallel alkynol, an allenylidene and a dimeric
metallacyclic ligand} were also obtained, together with 7, a thermal decomposition product of 6. Finally, when
Co2(CO)8 was treated with ECY in benzene, and Fe3(CO)12 was added, the heterometallic complex Co2Fe(CO)6-
(µ-CO)[µ3-η7-(C6H9)CC(H)C(H)C(H)(C6H10)] 8 was obtained in low yields. The complexes have been characterized
by means of IR and 1H NMR spectroscopies and by mass spectrometry. The structures of 2, 5, 8 have been
determined by X-ray diffraction. Complex 2 contains an “hydrogenated” ECY ligand, 5 an allenylidene ligand
formed upon dehydration of ECY (loss of the OH and of the terminal hydrogen), whereas 8 contains a ligand
formed by tail-to-tail coupling of two ECY molecules, with loss of water and oxygen and shift of hydrogen. The
elemental analysis of complex 8 gave a Co : Fe ratio of 2 : 1; on the basis of this analysis, the refinement of the
diffraction data allowed a hypothesis on the distribution of the metal atoms in the cluster. Reaction pathways
for the formation of these clusters are proposed and dehydration mechanisms for the ligand discussed.
Chemicals) and 1-ethynylcyclohexanol (Lancaster Syntheses)
Introduction
were commercial products used as received. Solvents (benzene,
toluene, heptane) were dehydrated over sodium. All the
reactions were performed under a dry nitrogen atmosphere
in conventional three necked flasks equipped with gas inlet,
cooler, mercury check valve and magnetic stirring.
Propargyl (2-propynyl) alcohols have played a considerable role
in the synthesis of important chemical intermediates; isoprene,
for example, was obtained in good yields from acetone and
acetylene through the intermediacy of 2-methyl-3-butyn-2-ol.1
Other alkynols were used as synthons for commodity chem-
icals; among these, 1-ethynylcyclohexanol (ECY) is a precursor
of corrosion inhibitors, a stabilizer for chlorinated organics
and an intermediate in the synthesis of pharmaceutical and
perfumery materials.2 A recent example of the use of ECY in
organic syntheses has also been reported.3
Here we report on the reactions of Co2(CO)8 and/or
Fe3(CO)12 with ECY in hydrocarbon solvents. Binuclear
products and trinuclear clusters were obtained and fully charac-
terized using analytic, spectroscopic and mass spectrometry
techniques. On the basis of spectroscopic results, structures
have been proposed for complexes Fe2(CO)6(ECY)2, Fe2-
The reaction mixtures were filtered under N2, brought to
small volume under reduced pressure and separated on TLC
plates (Merck Kieselgel PF, eluent mixtures of hexane and
diethyl ether in variable v/v ratios, depending on the reaction
mixtures). The products were crystallized when possible4 and
analysed by means of a Bruker Equinox 55 IR spectro-
1
photometer (KBr cells); the H NMR spectra were obtained
on a JEOL JNM 270/89 instrument and mass spectra with
a Finnigan-Mat TSQ-700 mass spectrometer (Servizio di
Spettrometria di Massa, Dipartimento di Scienza e Tecnologia
del Farmaco, Università di Torino). The 1H NMR spectrum of
pure ECY in CDCl3 has been registered for comparison with
2
᎐
(CO)6[(ECY)2 Ϫ H2O], Fe (CO) (µ-CO)[µ -η -1,2-HC᎐C(C -
᎐
3
9
3
6
those of the (dehydrated or non-dehydrated) ligands bound
H10OH)], Fe3(CO)7[µ3-η7-(C6H10OH)CCHCHC(C6H9)] and its
thermal decomposition product. The structures of three com-
plexes have been determined with X-ray diffraction analyses.
All of the clusters, but one, contain “modified” ECY ligands.
This behaviour is discussed and reaction (and dehydration)
mechanisms are proposed.
᎐
to the clusters; δ 3.07 s (1H, HC᎐), 2.39 s (1H, OH), 1.80 m,
᎐
1.59 m, 1.44 m, 1.14 m (10 H, cyclohexanol ring).
Reaction of Co2(CO)8 with 1-ethynylcyclohexanol
In a typical reaction 1.0 g (ca. 2.9 mmol) of the cobalt carbonyl
was dissolved in benzene (50 cm3) and 0.5 g (ca. 4.0 mmol) of
ECY were added; the solution was refluxed for 4–5 min, then
allowed to cool and filtered under N2. It was then brought to
small volume under reduced pressure; on the TLC plates a red-
brown (complex 1, ca. 80%) and a purple band (complex 2, ca.
Experimental
General details, materials, analysis of the products
Dicobalt octacarbonyl, triiron dodecacarbonyl (Strem
DOI: 10.1039/b101119g
J. Chem. Soc., Dalton Trans., 2001, 1485–1491
This journal is © The Royal Society of Chemistry 2001
1485

View Article Online
5%) were observed together with some decomposition. Com-
and dark brown bands (8, ca. 5%), unchanged iron carbonyl
plex 1: C 43.0 (42.6), H 3.3 (3.1)% (Calc.values in parentheses);
(ca. 10%), deep red (1, ca. 10%), purple (2, ca. 20%), brown-
orange (3b, ca. 5%), and light green bands (not investigated, ca.
3%) and decomposition. Complex 8: Fe 9.51 (9.56), Co 20.10
(20.17)%; IR 2097 m-w, 2050 vs, 2043 s, 2035 s, 1975 m(br),
1887 m-w(br) cmϪ1; 1H NMR δ 7.37 s (1H), 6.03 s(br) (1H) [H
on C3 and C2 respectively]; 3.79 d-t [H on C1], 2.98 t [H on C6],
1.90 m, 1.36 m, 1.27 s, 0.89 t [18 H, cyclohexane rings]; EI mass
spectrum Pϩ at m/z 584 loss of 7 CO (intense peaks), isotopic
pattern Co2Fe. This reaction is particularly well reproducible
regardless of the reflux time before adding the iron carbonyl.
1
IR (νCO, C7H16) 2095 s, 2055 vs, 2033 vs, 2023 s(sh) cmϪ1; H
᎐
NMR (CDCl , r.t.): δ 6.0 s (vb) (1H, HC᎐), 2.26 s (1H, OH),
᎐
3
1.64 vb (10 H). Identified as Co2(CO)6(ECY). Complex 2: C
36.7 (36.8), H 2.5 (2.4)%; IR 2101 w, 2050 vs, 2035 vs, 2024
m(sh) cmϪ1; 1H NMR (CDCl3) δ 3.96 s (2H, CH2), 3.68 s (1H,
OH), 1.77–1.23 mm (10 H, C6H10); EI mass spectrum Pϩ at
m/z 554, loss of 9 CO followed by complex fragmentation.
Reaction of Fe3(CO)12 with 1-ethynylcyclohexanol and TMNO
In a typical experiment triiron dodecacarbonyl (2.0 g, ca. 4.0
mmol) was suspended in benzene (50 cm3) and solid ECY (1.0 g,
1.08 mmol) added, together with Me3NOؒ2H2O (TMNO). The
suspension was refluxed for 4–7 min. TLC purification showed
the presence of about 15% of unchanged parent carbonyl
and of the following products: brown (complex 4, ca. 1%),
purple-brown (5, ca. 5%), orange (3a, ca. 10%), yellow-brown
(3c, ca. 10%), yellow (7, ca. 10%) and of a band containing an
orange and a purple oily complex (ca. 15%) which were not
investigated. Decomposition products were also observed.
Complex 4: C 37.6 (37.8), H 2.2 (2.1)%; IR 2081 w, 2052
Reaction of Co2(CO)6(ECY) with Fe3(CO)12
Complex 1 (0.20 g) was dissolved in benzene (50 cm3) and 0.50 g
of triiron dodecacarbonyl added; the suspension was allowed
to reflux for 7 min. After filtering away the unchanged iron
carbonyl and decomposition products, TLC purification
showed the presence of brown (8, ca. 5%), orange (1, ca. 15%),
orange-brown (unidentified, ca. 5%), purple (2, ca. 20%) and
orange-red bands (unidentified, 5%), traces of complex 6 and
decomposition.
1
s(sh), 2041 vs, 2034 s(sh), 1970 m, 1854 m cmϪ1; H NMR
Reaction of Co2(CO)8 with ECY and Fe(CO)5
᎐
δ 6.93 s (1H, HC᎐), 2.21 s (1H, OH), 2.0–1.16 (10H); EI mass
᎐
To a suspension of the cobalt (1.0 g) and iron carbonyl (2.0 ml)
in benzene (50 cm3), ECY (0.5 g) was added; the suspension was
refluxed for 4 min. After TLC separation the following products
were obtained; Co2(CO)6(ECY) (1, about 40%) and complex 2
(10%) together with some decomposition.
spectrum Pϩ at m/z 572, loss of 10 CO followed by complex
fragmentation, and a peak of medium intensity at m/z 554 (loss
of water?; see complex 5). Complex 5: C 39.2 (39.0), H 1.9
(1.8)%; IR 2090 m, 2052 s, 2031 vs, 1980 m, 1888 m-w cmϪ1; 1H
NMR δ signals (multiplets) in the 2.0–1.0 region, correspond-
ing to C6H10; EI MS Pϩ at m/z 554, loss of 10 CO. Complex 3a:
IR (C7H16) 2073 vs, 2027 vs, 2008 vs, 1987 vs, 1972 m cmϪ1; 1H
Reaction of complex 5 with ECY
᎐
᎐
Complex 5 (50 mg) was dissolved in benzene (50 cm3) and
ECY (0.4 g) added; the suspension was refluxed for 4 min.
TLC purification showed the presence of some complex 6 (IR
identification), traces of an unidentified yellow compound and
decomposition.
᎐
᎐
NMR (CDCl , r.t.) δ 9.18 s (1H, HC᎐), 3.45 s (1H, HC᎐), 2.60
3
m, 2.45 m (2H, OH), 2.16–0.89 mm (20H, C6H10 rings); EI mass
spectrum Pϩ at m/z 528 (first intense peak 472, very intense
426). Complex 3c: IR 2078 s, 2030 vs, 2013 s, 1995 s, 1980 m(sh),
1
1970 m cmϪ1; H NMR δ 8.91 s (1H, HC᎐), 7.98 s (1H, HC᎐),
᎐
᎐
᎐
᎐
1.58 s (1H, OH), 1.27 m (20 H, C6H10 rings); EI mass spectrum
Pϩ at m/z 510. Complex 7: C 51.8, H 4.3%; IR 2067 m, 2031 vs,
Reaction of complex 6 with Co2(CO)8
1
1987 m-s cmϪ1; H NMR δ 6.37 d (1H, J = 2.0), 6.04 d (1H,
Complex 6 (200 mg) was dissolved in benzene (50 cm3) and 0.5 g
of cobalt carbonyl added. After 5 min reflux, TLC separation
yielded trace amounts of two unidentified complexes (dark
orange and yellow), about 30% of yellow complex 7, 30% of
J = 2.0 Hz), 5.52 br (1H), 2.21–0.87 mm; EI mass spectrum
decomposition. For the attribution see Discussion.
Reaction of Fe3(CO)12 with ECY in the absence of TMNO
unchanged
6 and decomposition. No cobalt-containing
Under the same conditions as above, but in the absence of
TMNO, the following products were obtained: brown (complex
5, ca. 5%), orange (3a, ca. 15%), yellow (7, ca. 20%), green (6,
ca. 25%) and yellow-brown (3b), two unidentified products in
trace amounts and decomposition. Complex 6: C 46.8 (46.5), H
3.8 (3.7)%, IR 2083 w, 2047 m-s, 2030 s(sh), 2016 vs, 1995 m,
derivatives could be obtained.
X-Ray analysis
Crystal data. Complex 2. C23H22Co3O7, M = 554.06, mono-
clinic, space group P21/c (no. 14), a = 8.717(3), b = 13.524(5),
c = 20.427(7) Å, β = 93.23(2)Њ, U = 2091(1) Å3, Z = 4, T =
293 K, µ = 2.403 mmϪ1. 8058 Reflections on a Siemens P4 dif-
fractometer, 6105 (Rint = 0.037) being unique; non-hydrogen
atoms anisotropically refined. The last Fourier-difference maps
showed the peaks corresponding to the H atoms of the ligands:
they were put in the positions derived from maps and refined.
For 3446 data with Fo > 4σ(Fo) the final R1 = 0.0474.
Complex 5. C18H10Fe3O10, M = 553.81, monoclinic, space
group P21/c (no. 14), a = 17.095(2), b = 10.366(2), c = 12.665(2)
Å, β = 109.84(1)Њ, U = 2111.1(6) Å3, Z = 4, T = 293 K, µ = 2.086
mmϪ1. 8022 Reflections collected on a Siemens P4 diffrac-
tometer, 6159 (Rint = 0.029) being unique; non-hydrogen atoms
anisotropically refined. The peaks of the last Fourier-difference
maps corresponding to the H atoms of the ligands were refined
with variable coordinates and fixed Uiso. For 3822 data with
Fo > 4σ(Fo) the final R1 = 0.0444.
1
1861 m, cmϪ1; H NMR δ 7.26–7.25 d (J = 2.2), 6.94–6.93 d
᎐
(J = 2.2) (1H, 1H, HC᎐), 5.60 t (1H, CH, C H ring) [expansion
᎐
6
9
shows that this signal is a triplet of triplets with J 1.46 and 4.03
Hz], 1.88 s (1H, OH), 1.64–1.35 mm, 1.16 m (18 H, C6H9, C6H10
rings); EI mass spectrum Pϩ at m/z 594, loss of 7 CO’s;
tentative attribution Fe3(CO)7[(ECY)2 Ϫ H2O]. Complex 3b: IR
1
2068 m, 2028 vs, 2000 m-s, 1986 m, 1980 m cmϪ1; H NMR
᎐
δ 6.39 s (2H, HC᎐), 1.73–1.57 mm (ca. 20H, OH, C H10 rings);
᎐
6
EI mass spectrum Pϩ at m/z 528, loss of 6 CO, see complex 3a.
Reaction of Co2(CO)8 and of Fe3(CO)12 with 1-ethynylcyclo-
hexanol
In a typical reaction dicobalt octacarbonyl (1.0 g, 2.9 mmol)
was dissolved in toluene (100 cm3) under N2 and ECY (1.0 g, ca.
8.0 mmol) added; the solution was brought to reflux. After 2
min, 1.0 g (2.0 mmol) of triiron dodecacarbonyl was added and
reflux was allowed for 6 min. During the reaction abundant
foaming was observed. TLC purification showed the presence
of trace amounts of complex 5 and of red-brown (3a, ca. 10%)
Complex 8. C17H13Co2FeO10, M = 584.12, monoclinic, space
group P21/n (no. 14), a = 10.415(4), b = 11.316(3), c = 17.759(6)
Å, β = 92.76(3)Њ, U = 2404(2) Å3, Z = 4, T = 293 K, µ =
1.999 mmϪ1. 5833 Reflections collected on a Siemens P4 dif-
1486
J. Chem. Soc., Dalton Trans., 2001, 1485–1491

View Article Online
fractometer, 4228 (Rint = 0.048) being unique. The H atoms of
the ligands were put in computed positions (rings) or in the
experimental positions (chain) with fixed coordinates and vari-
able Uiso. The non-hydrogen atoms were anisotropically refined.
According to the elemental analysis, where a Co : Fe ratio of
2 : 1 was found, three refinements were made with the iron atom
alternatively in position 1, 2 and 3 (Fig. 3). The best refinement
corresponded to iron in position 2 (R1 = 0.0594 with respect to
0.0623 and 0.0630), even if some uncertainty exists owing to the
difference of only one electron between Fe and Co. For 2789
data with Fo > 4σ(Fo) the final R1 = 0.0594.
unsaturated substituent interacting with one iron atom. Some
chemical evidence (e.g. the reaction of complex 5 with ECY)
supports this hypothesis. In addition, a structure closely com-
parable with that proposed for 6 (see Scheme 2, below) has been
In the three complexes acceptable values of the distances
were obtained for the hydrogen atoms, even if their location
must be considered with caution.
CCDC reference numbers 159811–159813.
See http://www.rsc.org/suppdata/dt/b101119g/ for crystallo-
graphic data in CIF or other electronic format.
Results and discussion
Spectroscopic characterization of new complexes
The reactions of Co2(CO)8 with ECY lead to high yields of
Co2(CO)6(ECY) 1, containing a non-dehydrated ligand; com-
plex 1 belongs to a well established family of dicobalt–alkyne
derivatives. Small yields of complex 2 are also obtained in these
reactions. It is known that, under thermal conditions, dicobalt–
alkyne complexes of type 1 may afford tricobalt methylidyne
clusters showing structures comparable with that of complex
2.5 The reactions of Fe3(CO)12 with ECY in the presence of
TMNO lead to the binuclear complexes 3a, 3c, 7 as the main
products and to small amounts of the trinuclear clusters 4 and
5; in the absence of TMNO the binuclear complexes 3a, 3b, 7
and the trinuclear complexes 5 and 6 were obtained. Analytical
and spectroscopic results indicate that 3a and 3b are isomeric,
non-dehydrated, Fe2(CO)6(ECY)2 ferrole complexes;6 3c has
been identified as a partly dehydrated ferrole derivative. Form-
ation of similar structures containing dehydrated alkynols has
been observed.7 Finally, chemical evidence shows that 7 is the
thermal decomposition product of 6; we cannot propose
an unequivocal structure for this complex, on the basis of the
spectroscopic results only. We suspect that it also belongs to the
ferrole family; however, attempts at crystallizing the complex
gave only powders unsuitable for X-rays. The structures and
isomerism proposed for complexes 3a,b,c are in Scheme 1.
Scheme 2
reported for iron;9 in this complex, a nitrogen-contaning alkyne
substituent interacts with one of the iron atoms, and two
alkynes are linked together to form a ferrole-like ring. Finally,
reactions of alkynols with Ru3(CO)12 lead, among others, to
trinuclear clusters of type 4 and to a complex showing struc-
tural features similar to those proposed for cluster 6 (crystal
structure).10 Repeated attempts at crystallizing 6 were made
but unfortunately only powders could be obtained; during
these attempts partial decomposition to Fe3(CO)12 occurred.
The structures proposed for complexes 4 and 6 are shown in
Scheme 2.
When treating Co2(CO)8 and Fe3(CO)12 with ECY the diiron
complexes 3a and 3b are still obtained; interestingly, however,
the yields of 2 are increased and the heterometallic com-
plex 8 is formed. The structures of complexes 2, 5 and 8 were
determined by X-ray crystallography and are discussed below.
Scheme
1
R = RЉ = C6H10(OH); RЈ = Rٞ = H 3a; R = RЈ = C6H10-
(OH); RЉ = Rٞ = H 3b; R = C6H10(OH); RЉ = C6H9; RЈ = Rٞ = H 3c.
X-Ray diffraction studies on complexes 2, 5 and 8
The trinuclear complex 4 could not be obtained as X-ray
grade crystals; it has been identified as Fe3(CO)9(µ-CO)[µ3-
Co3(CO)9[ꢀ3-CCH2(C6H10OH)] 2. The structure of the com-
plex is shown in Fig. 1 and relevant bond distances and angles
are in Table 1. This complex is formed by an isosceles triangle
of cobalt atoms [Co(1)–Co(2) 2.462(1), Co(1)–Co(3) 2.464(1),
Co(2)–Co(3) 2.476(1) Å] each bearing one axial and two
equatorial CO’s. A µ3-CCH2C6H10(OH) methylidyne ligand is
bound to all three cobalt atoms via three M–C bonds of differ-
ent lengths, Co(1)–C(1) 1.936(3), Co(2)–C(1) 1.911(3), Co(3)–
C(1) 1.890(3) Å. The C(1)–C(2) [1.504(4) Å] and C(2)–C(3)
[1.535(5) Å] distances are typical of C–C single bonds.
2
᎐
η -1,2-HC᎐C(C H OH)] containing a non-dehydrated ECY
᎐
6
10
ligand coordinated in parallel fashion. This type of structure
is not very common for iron;8a however, some examples of
similar complexes, obtained in the reactions of alkynols with
triiron dodecacarbonyl, have been reported previously.8b It is
likely that, under thermal conditions, complex 4 loses water to
form the allenylidene complex 5; mass spectrometric evidence
for this process has been obtained (see Experimental section).
The trinuclear complex 6 has been identified as Fe3(CO)7-
[(ECY)2 Ϫ H2O] and contains two ECY ligands, one of which
has been dehydrated; on the basis of the spectroscopic results,
we propose for 6 a structure containing a ferrole ring with an
This type of structure has long been known for cobalt,12 and
it is not common for iron.13 It is perhaps interesting that, under
the same reaction conditions, cobalt prefers the methylidyne
structure, whereas iron gives the allenylidenic complex 5.
J. Chem. Soc., Dalton Trans., 2001, 1485–1491
1487

View Article Online
Table 1 Selected bond lengths (Å) and angles (Њ) for Co3(CO)9-
[µ3-CCH2(C6H10OH)] (complex 2)
Table 2 Selected bond lengths (Å) and angles (Њ) for Fe3(CO)9(µ-CO)-
[µ3-η2-1,2-C᎐C(C6H10)] (complex 5)
᎐
Co(1)–C(1)
Co(1)–Co(2)
Co(1)–Co(3)
Co(2)–C(1)
Co(2)–Co(3)
Co(3)–C(1)
Co–C(CO)eq av.
Co–C(CO)ax av.
C–O(CO) av.
C(1)–C(2)
1.936(3)
2.4619(9)
2.4634(9)
1.911(3)
2.4759(9)
1.890(3)
1.790(4)
1.829(4)
1.127(5)
1.504(4)
1.535(5)
1.429(4)
1.518(6)
Co(2)–Co(1)–Co(3)
Co(1)–Co(2)–Co(3)
Co(1)–Co(3)–Co(2)
C(2)–C(1)–Co(1)
C(2)–C(1)–Co(2)
C(2)–C(1)–Co(3)
C(1)–C(2)–C(3)
O(1)–C(3)–C(2)
O(1)–C(3)–C(4)
O(1)–C(3)–C(8)
C(8)–C(3)–C(4)
60.36(3)
59.85(2)
59.79(2)
123.4(2)
136.3(3)
134.7(3)
121.2(3)
109.4(3)
110.7(3)
107.3(3)
109.5(3)
Fe(1)–Fe(2)
Fe(1)–Fe(3)
Fe(2)–Fe(3)
Fe(1)–C(1)
Fe(2)–C(1)
Fe(3)–C(1)
Fe(1)–C(2)
2.6077(6)
2.6031(7)
2.5503(6)
2.002(3)
1.914(3)
1.911(3)
2.135(3)
Fe(3)–Fe(1)–Fe(2)
Fe(3)–Fe(2)–Fe(1)
Fe(2)–Fe(3)–Fe(1)
C(1)–Fe(2)–C(44)
C(1)–Fe(3)–C(44)
Fe(2)–C(1)–Fe(1)
Fe(3)–C(1)–Fe(1)
Fe(3)–C(1)–Fe(2)
C(3)–C(2)–C(1)
C(2)–C(3)–C(8)
C(2)–C(3)–C(4)
O(44)–C(44)–Fe(2)
O(44)–C(44)–Fe(3)
Fe(2)–C(44)–Fe(3)
58.60(2)
60.61(2)
60.79(2)
99.3(1)
97.7(1)
83.5(1)
83.4(1)
83.6(1)
Fe–C(CO) av. 1.805(4)
Fe(2)–C(44)
Fe(3)–C(44)
C–O(CO) av.
C(44)–O(44)
C(1)–C(2)
C(2)–C(3)
C–C(ring) av.
1.972(3)
2.021(3)
1.130(4)
1.149(4)
1.349(4)
1.324(4)
1.513(5)
153.6(3)
120.7(3)
125.3(3)
141.9(3)
138.7(3)
79.4(1)
C(2)–C(3)
C(3)–O(1)
C–C(ring) av.
Fig. 1 An ORTEP11 plot of Co3(CO)9[µ3-CCH2(C6H10OH)] (2) with
30% thermal ellipsoids (as in all structures shown).
Fig. 2 ORTEP plot of Fe3(CO)9(µ-CO)[µ3-η2-1,2-C᎐C(C6H10)] (5).
In spite of the abundance of reports on tricobalt methylidyne
structures,12 complex 2 is, to our knowledge, the first derivative
containing a propargylic alcohol and showing the presence of
two hydrogens on C(2) as shown in Scheme 3. One of them
᎐
and angles are in Table 2. The complex is formed by an isosceles
triangle of iron atoms [Fe(1)–Fe(2) 2.6077(6), Fe(1)–Fe(3)
2.6031(7) and Fe(2)–Fe(3) 2.5503(6) Å, the shorter edge being
bridged by a CO group and by the organic ligand], each bearing
two equatorial and one axial CO group; the bridging CO is
slightly asymmetric. An allenylidene ligand, formed upon
dehydration of ECY (terminal hydrogen and alkynol OH), is
coordinated via C(1) to all three metals and via C(2) to the third
one; the coordination is asymmetric, Fe(1)–C(1) being shorter
than Fe(2,3)–C(1).
In recent times some structures of triiron allenylidene clus-
ters derived (or not) from alkynols have been reported;15 com-
plex 5 represents the fourth example of such an X-ray structural
study. A comparison with the other complexes16 shows that
these clusters are characterized by “rigid” bond distances and
angles, at least as far as the cluster core and the C(3) and C(2)
allenylidene atoms are concerned; i.e. the doubly bridged Fe–Fe
bond is the shortest metal–metal bond, the Fe(1)–C(1) bond is
the longest Fe–C, the C(1)–C(2) and C(2)–C(3) distances are
nearly equal and the C(1)–C(2)–C(3) angle is around 150Њ. The
effect of the substituents on the organic moiety seems to be of
less importance.
Scheme 3
could be the terminal hydrogen of the alkyne, but the other one
has to come either from the solvent (traces of moisture?)14 or
from another alkyne molecule. It is also worth noting that the
yields of 2 are increased when iron is present; this behaviour is
discussed below.
Complex 5 contains a dehydrated ECY ligand. Dehydration
pathways occurring when alkynols react with the iron triad
carbonyl clusters have recently been discussed.17 Two main pro-
cesses have been evidenced, that is: (i) loss of the terminal
hydrogen and of the alkynic OH (route A) to form allenylidene
ligands and (ii) loss of the alkynic OH and of a hydrogen from
a methyl group on the same carbon atom (route B) to
form vinylacetylide derivatives. The process observed here
Fe (CO) (ꢀ-CO)[ꢀ -ꢁ2-1,2-C᎐C(C H )] 5. The structure of
᎐
3
9
3
6
10
the complex is shown in Fig. 2 and relevant bonding distances
1488 J. Chem. Soc., Dalton Trans., 2001, 1485–1491

View Article Online
Table 3 Selected bond lengths (Å) and angles (Њ) for Co2Fe(CO)6-
(µ-CO)[µ3-η7-(C6H9)CC(H)C(H)C(H)(C6H10)] (complex 8)
other one [C(2) to C(18)] has undergone loss of oxygen and
probably transfer of a hydrogen atom from C(13) to C(1)
(Scheme 4).
Co(1)–Co(3)
Co(1)–Fe(2)
Fe(2)–Co(3)
Co(1)–C(1)
Co(1)–C(4)
Co(1)–C(13)
Co(3)–C(4)
Co(3)–C(5)
Co(3)–C(6)
Co(3)–C(23)
CO–C(CO)eq av.
CO–C(CO)ax av.
Fe(2)–C(2)
Fe(2)–C(3)
Fe(2)–C(4)
Fe(2)–C(23)
Fe–C(CO)eq av.
C–O(CO) av.
C(1)–C(2)
C(1)–C(13)
C(2)–C(3)
C(3)–C(4)
C(4)–C(5)
C(5)–C(6)
C(5)–C(10)
C(6)–C(7)
C(7)–C(8)
C(8)–C(9)
C(9)–C(10)
2.518(1)
2.547(1)
2.537(1)
2.150(6)
1.960(5)
2.164(6)
2.175(6)
2.184(6)
2.339(6)
1.939(7)
1.784(8)
1.791(8)
2.100(7)
2.076(6)
2.162(6)
1.901(7)
1.777(8)
1.139(9)
1.459(8)
1.396(8)
1.376(8)
1.446(8)
1.457(8)
1.360(8)
1.518(8)
1.495(8)
1.51(1)
Co(3)–Co(1)–Fe(2)
Co(3)–Fe(2)–Co(1)
Co(1)–Co(3)–Fe(2)
C(13)–C(1)–C(2)
C(3)–C(2)–C(1)
C(2)–C(3)–C(4)
C(3)–C(4)–C(5)
C(6)–C(5)–C(4)
C(6)–C(5)–C(10)
C(4)–C(5)–C(10)
C(5)–C(6)–C(7)
C(6)–C(7)–C(8)
C(9)–C(8)–C(7)
C(8)–C(9)–C(10)
C(9)–C(10)–C(5)
C(1)–C(13)–C(18)
C(1)–C(13)–C(14)
O(23)–C(23)–Fe(2)
O(23)–C(23)–Co(3)
60.12(4)
59.37(4)
60.51(4)
126.7(6)
119.9(5)
115.8(5)
122.1(5)
121.3(5)
119.5(5)
119.1(5)
124.5(6)
113.0(6)
111.2(6)
114.4(7)
113.2(6)
117.6(5)
121.5(5)
143.5(6)
133.8(6)
Scheme 4
The C(1)–C(6) chain is roughly planar (0.07 Å mean devi-
ation from planarity), whereas C(13) is clearly out of plane
(1 Å); hydrogen atoms, affected by relevant uncertainty, were
not considered in the computing of the mean plane. All angles
around atoms C(1)–C(6) are quite similar and their mean value
is 122Њ; the C–C distances in the C(13)–C(1)–C(6) chain cannot
be rationalized in terms of double or single bonds. It is notice-
able however that the significantly longest distances (1.46 Å av.)
are those corresponding to the triple bonds of the ECY reagent
[C(1)–C(2) and C(3)–C(4) and to C(4)–C(5)]. Interestingly the
distribution of long and short C–C distances in the ligand is
different from what could be expected by simply considering
the coupling of two alkynes. Apparently, the loss of the oxygen
atoms induces a rearrangement in the ligand. The coordination
to the metals also is likely to induce modifications in the organic
moiety. The C(13)–C(18) ring has a chair conformation, while
the other η2-C(5)–C(10) ring is flattened where coordinated to
Co(3), as expected.
1.46(1)
1.509(9)
C(13)–C(18)(ring) av. 1.512(8)
When considering the electron count around the metal
atoms, 14 electrons are provided by the seven carbonyl groups
and six by the metal–metal bonds; thus, according to the EAN
rule, the ligand must be a donor of eight electrons. Clusters
containing “linear” chains of four carbon atoms are not
uncommon; some have been obtained starting from diynes19
but a certain number has been obtained by coupling of alkynes
and/or acetylides, during or after the coordination of the
ligands on the metals.20 However, the bonding of the organic
ligand in cluster 8, which involves also carbon atoms of the
cyclohexanol rings, is to our knowledge unprecedented. Among
the M–C bonds Co(1)–C(4) has the shortest distance (1.960(5)
Å) and Co(3)–C(6) has the longest (2.339(6) Å). C(3), C(4),
C(5) and Co(1) lie roughly on a plane (0.012 Å mean devi-
ation), consistent with a Co(1)–C(4) σ interaction. The other
M–C distances are intermediate between these two values, 2.08
to 2.18 Å.
Fig. 3 ORTEP plot of Co2Fe(CO)6(µ-CO)[µ3-η7-(C6H9)CC(H)C(H)-
C(H)(C6H10)] (8).
corresponds to route A which, in the light of actual know-
ledge, is more common for iron (under “spontaneous” thermal
conditions). For ruthenium the above dehydration processes
require acidic catalysis or the intermediacy of inorganic
oxides.18 Besides the knowledge of the solid-state structure,
and dehydration process occurring, the interest in complex 5
lays in its behaviour as a reaction intermediate; experimental
evidence shows, indeed, that it is a precursor of 6 (see Scheme 5,
below).
Reaction pathways
The reaction of ECY with Co2(CO)8 leads to cobalt-containing
complexes of predictable structure;5,12 it is interesting, however,
that dehydrated cobalt-containing complexes could not be
obtained. Instead, small amounts of the “hydrogenated” com-
plex 2 were formed; better yields of 2 are obtained in the reac-
tions where also iron is present. Literature evidence (further
discussed below) indicates that iron favours dehydration and
deoxygenation reactions and presumably promotes the form-
ation of species which can transfer hydrogen on (free or
coordinated) ethynylcyclohexanol, thus giving origin to the new
ligand present on the tricobalt cluster.
The reactions of ECY with Fe3(CO)12 give the somewhat
expected binuclear ferrole complexes 3; these are considered the
final products of the reactions between triiron dodecacarbonyl
and alkynes. However, we could not obtain their “parents”,
i.e. the well known open clusters Fe3(CO)6(µ-CO)2(ECY)2;6
instead, we obtained complex 6 whose structure is closely
related to that of the above open clusters. We obtained very
small amounts of complex 4; this is presumably the parent of
the allenylidene complex 5, as shown by mass spectrometric
Co2Fe(CO)6(ꢀ-CO)[ꢀ3-ꢁ7-(C6H9)CC(H)C(H)C(H)(C6H10)] 8.
The structure of the complex is shown in Fig. 3 and relevant
bonding distances and angles are in Table 3. The attribution of
the nature of the metal atoms is described in the Experimental
section.
The complex is formed by a scalene triangle of metal atoms;
two terminal carbonyls are bound to each metal atom and a
symmetrically bridging carbonyl spans the Fe(2)–Co(3) edge
of the cluster [Fe(2)–C(23) 1.901(7), Co(3)–C(23) 1.939(7) Å].
A substituted organic ligand is coordinated to all the cluster
metals; this is formed by tail-to-tail coupling of two ECY
molecules, one of which [C(3) to C(10)] has lost water and the
J. Chem. Soc., Dalton Trans., 2001, 1485–1491
1489

View Article Online
(this work) and literature evidence.17 Interestingly, we also
obtained 6 containing two partially dehydrated and presumably
linked ECY ligands; again, chemical evidence shows that it is
formed through the intermediacy of the dehydrated allenyl-
idene 5 upon addition of a second (non-dehydrated) molecule
of ECY and shift of hydrogen from a cyclohexanol ring. This
behaviour has been observed, at least in part, in literature
reports.10 We therefore propose a reaction pathway leading
from complex 4 to 6; this would occur through the dehydration
and hydrogen shift processes shown in Scheme 5.
gradual substitution of iron for cobalt atoms until formation
of a diiron derivative was observed.21 Very recently, another
example of “isolobal replacement” of Co(CO)3 vertices by
Fe(CO)3 fragments has been reported for binuclear alkynol
cobalt complexes.22 The reverse process (e.g. replacement
of iron fragments by cobalt) has not been observed, to our
knowledge.
The formation of
8 also requires condensation and
modification of two ECY molecules; to our knowledge, the
condensation processes leading to complex 8 are unprece-
dented. Usually alkynes (and propargyl alcohols) undergo
cyclotrimerization on dicobalt complexes23 forming 1,2,4- or
1,3,5-substituted benzenes. Linear oligomerization of three
molecules of ECY in the presence of nickel has also been
reported:24 of these, two are bound tail-to-tail and the third
one head-to-tail. Interestingly, no dehydration, neither loss of
OH nor of “O”, has been observed in this process. However,
in some instances, partial dehydration occurs on nickel; for
example 2,5-dimethylhex-3-yne-2,5-diol undergoes linear/
cyclic oligomerization (5 alkyne molecules) with loss of water
and formation of ether linkages.25 Finally, the (carbonylative)
formation of oxygenated metallacycles and of lactones from
unsaturated diols in the presence of diiron nonacarbonyl has
been reported; in this process loss of hydrogen and/or of water
occurs.26
Concluding remarks
Reactions of propargyl alcohols with M3(CO)12 clusters
(M = Fe, Ru or Os) have been investigated17 and the alter-
native dehydration routes A and B established. For ruthenium,
dehydration reactions occur only under severe conditions; in
contrast, when M = Fe, route A is likely to occur easily, under
relatively mild conditions, to give allenylidene-substituted tri-
nuclear clusters. This process has been observed also in this
work (e.g. formation of cluster 5). “dehydroxylation” to form
allenylidene derivatives17,18 and “deoxygenation” reactions27
have also been observed in the presence of ruthenium or iron.
Some of these processes have been evidenced also during the
formation of the complexes discussed above, in particular when
8 is formed. The evidence obtained in this work and the liter-
ature reports mentioned above suggest that these processes are
likely to be promoted by the iron “fragments” involved in the
reactions.
The results obtained in this work, therefore, represent further
evidence for the role of iron fragments in inducing dehydra-
tion, dehydroxylation or deoxygenation reactions during the
coordination and/or oligomerization of acetylenic alcohols on
metal clusters. In particular, the formation of complexes 6 and
8 represent new examples of this behaviour, which is presum-
ably a general trend in the reactions of iron carbonyls with
acetylenic alcohols. Literature reports and experimental evi-
dence (this work) indicate indeed that dehydration and other
processes can occur, albeit to a lesser extent, on nickel, whereas
on cobalt this behaviour is not observed.
As pointed out above, it is interesting that cobalt and iron,
under comparable reaction conditions, apparently prefer dif-
ferent coordination modes for the ligand on clusters; thus,
with cobalt, the methylidyne complex 2 is obtained, whereas
for iron the (structurally related) allenylidene complex 5 is
formed.
New coupling pathways for ECY have also been evidenced,
especially in complex 8. Isomerization of butadiyne ligands
during the formation of hexa- and hepta-nuclear iron–cobalt
complexes has been reported;28 allenylidene ligands coordin-
ated to three iron atoms are formed, as well as a four-
carbon atom chain not dissimilar to that found in cluster
8. These results show, once again, that alkyne- and alkynol-
cluster chemistry is very versatile and far from being fully
explored.
Scheme 5
Complex 8, however, is not formed starting from the above
triiron clusters, upon metal fragment substitution (e.g. reaction
of 6 with dicobalt octacarbonyl). It is obtained only when both
parent carbonyls are treated with ECY. Chemical evidence also
shows that 8 is not formed when ECY is treated with the cobalt
carbonyl only or when the iron carbonyl only is used; it is how-
ever obtained when complex 1 is treated with Fe3(CO)12 (not
with Fe(CO)5, thus indicating the need for a polynuclear source
of iron fragments). Therefore, although the simultaneous
formation of several products in these reactions makes diffi-
cult the elucidation of reaction mechanisms, a reasonable
hypothesis for the formation of the metal core of complex 8 is
the condensation of dicobalt fragments with cluster-derived
mononuclear iron fragments. An alternative hypothesis could
be substitution of iron fragments for cobalt in trinuclear struc-
tures, such as 2. Isolobal substitution of cobalt- with iron-
containing “fragments” has indeed been evidenced. For
example, in a complex reaction sequence starting from clusters,
hydration of cluster-bound diethylacetylene led to a propargyl
alcohol coordinated to two cobalt atoms; in the presence of
Fe(CO)5 the complex underwent loss of OH and of H and
1490
J. Chem. Soc., Dalton Trans., 2001, 1485–1491

View Article Online
14 Examples of alkyne-substituted clusters containing an extra
Acknowledgements
Financial support to this work was obtained from MURST
(Cofin98) and from the Universities of Torino and Piemonte
Orientale.
hydrogen atom are: D. F. Jones, P. H. Dixneuf, A. Benoit and
J.-Y. Le Marouille, J. Chem. Soc., Chem. Commun., 1982, 1217;
E. Sappa, D. Belletti, A. Tiripicchio and M. Tiripicchio Camellini,
J. Organomet. Chem., 1989, 359, 419.
15 M. Iyoda, Y. Kuwatani and M. Oda, J. Chem. Soc., Chem.
Commun., 1992, 399; see also ref. 7.
16 G. Gervasio, D. Marabello and E. Sappa, J. Chem. Soc., Dalton
Trans., 1997, 1851.
17 The reactions of M3(CO)12 carbonyls (M = Fe, Ru or Os) with
propargylic alcohols are discussed in: S. Deabate, P. J. King and
E. Sappa, in Metal Clusters in Chemistry, eds. P. Braunstein,
L. A. Oro and P. R. Raithby, J. Wiley-VCH, Weinheim, 1999, vol. 2,
ch. 8, pp. 796–843.
References
1 J. W. Reppe, O. Schlichting and T. Toepel, Liebigs Ann. Chem., 1955,
596, 1; J. W. Reppe, O. Schlichting, K. Klager and T. Toepel, Liebigs
Ann. Chem., 1956, 601, 81.
2 R. J. Tedeschi, Acetylene-based Chemicals from Coal and Other
Natural Resources, Marcel Dekker, New York, 1982.
3 G. Zeni, A. Chieffi, R. L. O. R. Cunha, J. Zuckerman-Schpector,
H. A. Stefani and J. V. Comasseto, Organometallics, 1999, 18, 803.
4 In some instances oily products were obtained; these could not be
crystallized. Their IR and NMR signals were usually broad and did
not allow a full characterization of the complexes.
5 See for example: R. F. Heck, J. Am. Chem. Soc., 1963, 83, 657;
G. Palyi, F. Piacenti and L. Markò, Inorg. Chim. Acta Rev., 1970, 4,
109; G. Varadi, A. Vizi-Orosz, S. Vastag and G. Palyi, J. Organomet.
Chem., 1976, 108, 225; M. Arewgoda, B. H. Robinson and
J. Simpson, J. Am. Chem. Soc., 1983, 105, 1893; J. J. Bonnet
and R. Mathieu, Inorg. Chem., 1978, 17, 1973; S. M. Draper,
C. Long and B. M. Myers, J. Organomet. Chem., 1999, 588, 195.
6 See for example: E. Sappa, J. Organomet. Chem., 1999, 573, 139, and
references therein.
7 E. Sappa, G. Predieri, A. Tiripicchio and F. Ugozzoli, Gazz. Chim.
Ital., 1995, 125, 51.
8 Only one structural determination for iron is known up to now:
(a) D. Lentz and M. Reuter, Chem. Ber., 1991, 124, 773. See also:
(b) S. Brait, G. Gervasio, D. Marabello and E. Sappa, J. Chem. Soc.,
Dalton Trans., 2000, 989.
18 J. P. H. Charmant, P. Crawford, P. J. King, R. Quesada-Pato and
E. Sappa, J. Chem. Soc., Dalton Trans., 2000, 4390.
19 See for example: P. Blenkiron, G. D. Enright, N. J. Taylor and
A. J. Carty, Organometallics, 1996, 15, 2855; P. Blenkiron, G. D.
Enright, P. J. Low, J. F. Corrigan, N. J. Taylor, Y. Chi, J.-Y. Saillard
and A. J. Carty, Organometallics, 1998, 17, 2447; M. I. Bruce, B. W.
Skelton, A. H. White and N. N. Zaitseva, J. Organomet. Chem.,
1998, 558, 197.
20 M. I. Bruce, B. W. Skelton, A. H. White and N. N. Zaitseva,
J. Chem. Soc., Dalton Trans., 1999, 13; R. W. Eveland,
C. C. Raymond and D. F. Shriver, Organometallics, 1999, 18, 534;
A. J. Carty, G. Hogarth, G. Enright and G. Frapper, Chem.
Commun., 1997, 1883; C.-W. Shiu, Y. Chi, C. Chung, S.-M. Peng
and G.-H. Lee, Organometallics, 1998, 17, 2970; E. Delgado,
Y. Chi, W. Wang, G. Hogarth, P. J. Low, G. D. Enright,
S.-M. Peng, G.-H. Lee and A. J. Carty, Organometallics, 1998, 17,
2936.
21 S. Aime, L. Milone and D. Osella, J. Chem. Soc., Chem. Commun.,
1979, 704; S. Aime, L. Milone, D. Osella, A. Tiripicchio and
A. M. Manotti Lanfredi, Inorg. Chem., 1982, 21, 501; S. Aime,
D. Osella, L. Milone, A. M. Manotti Lanfredi and A. Tiripicchio,
Inorg. Chim. Acta, 1983, 71, 141; S. Aime, D. Osella, L. Milone and
A. Tiripicchio, Polyhedron, 1983, 2, 77.
9 E. Cabrera, J.-C. Daran, Y. Jeannin and O. Kristiansson,
J. Organomet. Chem., 1986, 310, 367.
10 C. S.-W. Lau and W.-T. Wong, J. Organomet. Chem., 1999, 589, 198.
See also: C. S.-W. Lau and W.-T. Wong, J. Chem. Soc., Dalton Trans.,
1999, 607.
22 J. A. Dunn, W. J. Hunks, R. Ruffolo, S. S. Rigby, M. A. Brook and
M. J. McGlinchey, Organometallics, 1999, 18, 3372.
11 C. K. Johnson, ORTEP II, Report ORNL-5138, Oak Ridge
National Laboratory, Oak Ridge, TN, 1976.
23 Recent examples of cyclotrimerization of alkynes on dicobalt
complexes are: R. J. Baxter, G. R. Knox, P. L. Pauson and
M. D. Spicer, Organometallics, 1999, 18, 197; R. J. Baxter,
G. R. Knox, J. H. Moir, P. L. Pauson and M. D. Spicer,
Organometallics, 1999, 18, 206; R. J. Baxter, J. R. Knox,
P. L. Pauson and M. D. Spicer, Organometallics, 1999, 18, 215.
24 P. Carusi, G. Cerichelli, A. Furlani, M. V. Russo and L. Suber, Appl.
Organomet. Chem., 1987, 1, 555.
25 A. Furlani, P. Bicev, S. Franzoso, A. Chiesi Villa, A. Gaetani
Manfredotti and C. Guastini, J. Chem. Soc., Perkin Trans. 2, 1977,
1011.
26 R. Schobert, H. Pfab, A. Mangold and F. Hampel, Inorg. Chim.
Acta, 1999, 291, 91.
27 G. Gervasio and E. Sappa, Organometallics, 1993, 12, 1458. See
also: S. Bernès, R. A. Toscano, A. C. Cano, O. Garcìa Mellado,
C. Alvarez-Toledano, H. Rudler and J.-C. Daran, J. Organomet.
Chem., 1995, 498, 15.
28 M.-C. Chung, A. Sakurai, M. Akita and Y. Moro-oka, Organo-
metallics, 1999, 18, 4684.
12 Some among the many examples available are: P. W. Sutton and
L. F. Dahl, J. Am. Chem. Soc., 1967, 89, 261; K. Bertl, R. Boese
and G. Schmid, J. Organomet. Chem., 1981, 206, 331; A. D.
Shaposhnikova, R. A. Stadnichenko, G. L. Kamalov, A. A.
Pasynskii, I. L. Eremenko, S. E. Nefedov, Yu. T. Struchkov and
A. I. Yanovsky, J. Organomet. Chem., 1993, 453, 279; M. D. Brice,
R. J. Dellaca, B. R. Penfold and J. L. Spencer, Chem. Commun.,
1971, 72; P. A. Dawson, B. H. Robinson and J. Simpson, J. Chem.
Soc., Dalton Trans., 1979, 1762; G. A. Acum, M. J. Mays, P. R.
Raithby, H. R. Powell and G. A. Solan, J. Chem. Soc., Dalton
Trans., 1997, 3427; K. Yang, J. M. Smith, S. G. Bott and M. G.
Richmond, Organometallics, 1993, 12, 4779; C. Renouard,
G. Rheinwald, H. Stoeckli-Evans, G. Suss-Fink, D. Braga and
F. Grepioni, J. Chem. Soc., Dalton Trans., 1996, 1875.
13 An example of a triiron methylidyne complex obtained from alkynes
is in: S. Aime, L. Milone, E. Sappa and A. Tiripicchio, J. Chem.
Soc., Dalton Trans., 1977, 227.
J. Chem. Soc., Dalton Trans., 2001, 1485–1491
1491