Synthesis of the triply-bonded dimolybdenum anions [Mo2(η 5-C5H5)2(μ-PA2)(μ- CO)2]- (A = Cy, Et, Ph, OEt): Unsaturated hydride and carbyne derivatives

Article abstract of DOI:10.1039/b909493h

Tetrahydrofuran solutions of the 30-electron anions [Mo₂Cp ₂(μ-PA₂)(μ-CO)₂]^- (A = Cy, Et, Ph, OEt) are conveniently prepared through a two-step approach. In the first step, [Mo₂Cp₂

Full text of DOI:10.1039/b909493h

View Article Online / Journal Homepage / Table of Contents for this issue
Dalton
Transactions
An international journal of inorganic chemistry
www.rsc.org/dalton
Number 39 | 21 October 2009 | Pages 8129–8444
ISSN 1477-9226
ARTICLE TYPE
ARTICLE TYPE
Ruiz et al.
Calhorda and Costa
Synthesis of the dimolybdenum anions Expanding the role of oxo-
[Mo (ŋ5-C5H5)2(μ-PA )(μ-CO)2]–
(A =2Cy, Et, Ph, OEt): 2unsaturated
hydride and carbyne derivatives
molybdenum(vi) catalysts: a DFT
interpretation of X–H activation
leading to reduction or oxidation
1477-9226(2009)39;1-T

View Article Online
PAPER
www.rsc.org/dalton | Dalton Transactions
Synthesis of the triply-bonded dimolybdenum anions
[Mo2(g5-C5H5)2(l-PA2)(l-CO)2]- (A = Cy, Et, Ph, OEt): unsaturated
hydride and carbyne derivatives
M. Esther Garc´ıa, Sonia Melo´n, Alberto Ramos and Miguel A. Ruiz*
Received 13th May 2009, Accepted 23rd June 2009
First published as an Advance Article on the web 31st July 2009
DOI: 10.1039/b909493h
Tetrahydrofuran solutions of the 30-electron anions [Mo2Cp2(m-PA2)(m-CO)2]- (A = Cy, Et, Ph, OEt)
are conveniently prepared through a two-step approach. In the first step, [Mo2Cp2(CO)6] is treated with
the chlorophosphines ClPR2 (R = Cy, Et, Ph) or the chlorophosphite ClP(OEt)2, in refluxing toluene or
diglyme respectively, to give the corresponding 32-electron chloro-complexes [Mo2Cp2(m-Cl)(m-PA2)-
(CO)2] as major products. In the second step, these air-sensitive intermediates are treated in
tetrahydrofuran solution at room temperature with one of several reducing agents such as Li[BHEt3],
Li(Hg), Na(Hg) or K[BHsBu3] to give red solutions of the corresponding alkali-metal salts of the
anions, which display significant ion pairing involving one or both oxygen atoms of the bridging
carbonyl ligands, depending on the cation. All these triply bonded species are quite air-sensitive and
could not be isolated as pure solids, but they can be easily protonated using a weak acid such as
[NH4]PF6 to give with good yield the corresponding unsaturated hydrides [Mo2Cp2(m-H)(m-PA2)(CO)2],
which are species of low to moderate sensitiveness to air, and also formally containing an intermetallic
triple bond. The reactivity of the dicyclohexylphosphide-bridged anion (mainly as its Li+ salt) towards
different hydrocarbon halides RX was studied in detail. These reactions were found to be rather
complex, critically depending on the reagent used, and generally resulting in the formation of several
products, of which four types were identified: (a) the known agostic products [Mo2Cp2(m-PCy2)(m-R)-
(CO)2] (R = Me, CH2Ph), (b) the new alkoxycarbyne products [Mo2Cp2(m-COR)(m-PCy2)(m-CO)] [R =
Me, Et, C(O)Ph, iPr, Cy], which could be conveniently isolated as pure solids, (c) the iodoxycarbyne
complex [Mo2Cp2(m-COI)(m-PCy2)(m-CO)], a very unstable species formed in the reaction with EtI, and
(d) the halide complexes [Mo2Cp2(m-PCy2)(m-X)(CO)2] [X = Cl, Br, I], which were more conveniently
prepared by the direct reaction of the anion with the pertinent halogen (X = Br, I). The analysis of the
above results suggests that at least three primary reaction pathways are in operation: (a) nucleophilic
attack of the anion through its dimetal centre, (b) nucleophilic attack of the anion through the oxygen
atoms of its bridging carbonyls and (c) electron-transfer with the reagent, this being the main path to
the halo-complexes [Mo2Cp2(m-PCy2)(m-X)(CO)2].
the latter two cases. A few years ago we discovered an efficient
Introduction
synthetic route leading to the 30-electron dimolybdenum anions
of general formula [Mo2Cp2(m-PA2)(m-CO)2]-, (A = Cy, Ph, OEt)
Transition-metal carbonyl anions are classic reagents of enormous
utility in the organometallic synthesis.1 Although a large number
of mononuclear and polynuclear anions have been described, the
number of binuclear species is significantly lower. Moreover, if
we only consider those complexes having multiple metal–metal
bonds, then just about half a dozen complexes have been reported
in the literature so far. Within this reduced group we can quote one
33-electron paramagnetic complex [Fe2(m-PtBu2)2(CO)5]-,2 and a
few 32-electron anions such as the dihydrides [M2(m-H)2(CO)8]2-
(M = Cr, Mo, W)3 and [Mn2(m-PPh2)(m-H)2(CO)6]-,4 the di-
iron complex [Fe2(m-PPh2)(CO)6]-,5 and the dimanganese one
[Mn2(CO)6(m-Ph2PCH2PPh2)]2-.6 For different reasons (synthetic
difficulty, easy degradation to mononuclear species, etc.) the
reactivity of these complexes has been only studied in detail in
via the corresponding chloro-complexes.7 These compounds were
almost the first carbonyl anions having metal–metal triple bonds
to be reported. Actually, to our knowledge the only other 30-
electron binuclear anion previously described appears to be [Re2(m-
H)3(CO)6]-, a species reported a long time ago, but never studied in
detail.8 More interestingly, our preliminary experiments revealed
that the dicyclohexylphosphide-bridged anion was highly reactive
and exhibited two distinct nucleophilic sites located at the Mo
and O atoms, thus enabling the preparation of novel unsaturated
derivatives.7 Some of these derivatives proved afterwards to
display a remarkable reactivity and synthetic potential, as it
is the case of the hydride [Mo2Cp2(m-H)(m-PCy2)(CO)2],9 the
methoxycarbyne [Mo2Cp2(m-COMe)(m-PCy2)(m-CO)],10 and the
agostic alkyl [Mo2Cp2(m-CH2R)(m-PCy2)(CO)2] derivatives (R =
Me, Ph).9f ,11 We then can safely state that the unsaturated anions
of type [Mo2Cp2(m-PA2)(m-CO)2]- are indeed the key synthetic
Departamento de Qu´ımica Orga´nica e Inorga´nica/IUQOEM, Universidad
de Oviedo, E-33071, Oviedo, Spain. E-mail: mara@uniovi.es
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 8171–8182 | 8171
©

View Article Online
entry to a whole chemistry to be developed around unsaturated
dimolybdenum carbonyl complexes having very different function-
alities. In this paper we give full details of the preparation of these
30-electron anions (including the diethylphosphide-bridged ana-
logue) with several counter-ions and their hydride derivatives. We
also analyze in detail the reactions of the dicyclohexylphosphide-
bridged anion with different C-based electrophiles, these revealing
the occurrence of at least three different reaction pathways: (a)
nucleophilic attack of the anion through the metal centre to
give alkyl-bridged derivatives, (b) nucleophilic attack of the anion
through the oxygen atom of the carbonyl ligand to give carbyne-
bridged derivatives of the general formula [Mo2Cp2(m-COR)(m-
PCy2)(m-CO)], and (c) electron transfer to give halide-bridged
derivatives. The electronic structure of the dicyclohexyl-bridged
anion [Mo2Cp2(m-PCy2)(m-CO)2]- as well as those of its hydride
and methoxycarbyne derivatives have been theoretically studied
using DFT methods.9f ,10a
small amounts of several intermediate species which were not
investigated, as well as the formation of a few minor side-products.
Likely intermediates in these reactions would be at least the
tetracarbonyl complexes [Mo2Cp2(Cl)(m-PA2)(CO)4], analogous
to the phenylphosphide complex [Mo2Cp2(Cl)(m-PHPh)(CO)4]
recently characterized by us, although prepared using a milder
synthetic route.12
The available spectroscopic data for compounds 1, although
limited, indicate that they all share the same structure. Thus, their
IR spectra (Table 1) exhibit two C–O stretching bands with the
relative intensities (weak and strong, in order of decreasing fre-
quency) characteristic of transoid M2(CO)2 oscillators,13 and with
frequencies a bit higher than those measured for the isoelectronic
bis(phosphide) complexes of the type trans-[Mo2Cp2(m-PR2)(m-
PR2¢)(CO)2],14 as expected. The chemical shifts of the 31P nuclei in
compounds 1 however, are some 50 ppm higher than the values
measured for comparable bis(phosphide) complexes, a fact that
canbe ascribed tothe higher electronegativityofthe chlorineatom,
when compared to phosphorus. This effect does not hold for the
diethoxyphosphide complex 1d, which displays a chemical shift (dP
347.4 ppm) almost identical to that of the bis(phosphide) complex
[Mo2Cp2(m-PCy2){m-P(OEt)2}(CO)2] (dP 351.1 ppm).7 Thus, it
seems that the presence of the ethoxy substituents dominates
the P shielding of the diethoxyphosphide ligand and minimizes
the influence of factors otherwise relevant (presence and order
of metal–metal bonds, other ligands, etc.), as it will be also seen
later on. We have noticed previously similar insensitivity of the
31P chemical shifts of bridging diethoxyphosphide ligands in other
Mn2, Mo2 and Fe2 substrates.15
Results and discussion
Synthesis and characterization of the chloro-complexes 1
The title compounds can be conveniently prepared from commer-
cial reagents through the thermal reaction of [Mo2Cp2(CO)6] with
the chlorophosphines ClPR2 (R = Cy, Et, Ph) or the chlorophos-
phite ClP(OEt)2, in refluxing toluene or diglyme respectively. This
gives dark green solutions containing the corresponding chloro-
complexes [Mo2Cp2(m-Cl)(m-PA2)(CO)2], [A = Cy (1a), Et (1b),
Ph (1c), OEt (1d)], as major species (Scheme 1). Unfortunately,
compounds 1 are quite air sensitive and they could not be
isolated as pure materials in the usual way (crystallization,
chromatography, etc.). In spite of this, the crude reaction products
were pure enough to be used in the preparation of the unsaturated
anions 2.
Synthesis of the unsaturated anions 2
The chloro-complexes 1a–d react readily at room temperature
in tetrahydrofuran solution with one of several reducing agents
such as Li[BHEt3], Na(Hg) or K[BHsBu3] to give red solutions
of the corresponding alkaline metal salts of the anions 2a–d. The
solubility of these salts change noticeably with the cation, the
sodium salts being only moderately soluble in tetrahydrofuran.
Although these complexes could not be isolated, IR and 31P
NMR data of the corresponding reaction mixtures indicated the
presence of a single major species in each case (Table 1), so these
crude solutions could be used for further reactivity studies. These
solutions obviously contain also the corresponding alkaline salt
and, in the case of the Li+ and K+ salts, different boron species
too, including any excess of reagent used. In some cases it was
desirable to get a solution of these anions free from these secondary
species, and some alternative procedures were developed for
this purpose (see Experimental section). For instance, LiCl–
free solutions of the salt Li-2a can be generated by treating
the hydride 3a with Li[BHEt3] or with Li(Hg), the latter also
yielding a boron-free solution (methods B and C). Alternatively,
a boron-free lithium salt can be also generated by reducing the
pertinent chloro-complex with a Li amalgam. In the case of the
dicyclohexylphosphide-bridged anion 2a we also accomplished a
change in the counterion by treating its lithium salt with (PPN)Cl
in tetrahydrofuran. The resulting PPN-2a salt, however, was not
much more stable either, and we were not successful in crystallizing
it for a more precise structural characterization of all these anionic
species.
Scheme 1 Synthetic route to the anions 2a–d and their hydride derivatives.
The formation of compounds 1a–d requires the oxidative
addition of the P–Cl bond of the added ligand to the dimetal
centre and the removal of four CO molecules, and obviously it
must involve several intermediate species. Indeed, IR monitoring
of the corresponding reaction mixtures allows the detection of
8172 | Dalton Trans., 2009, 8171–8182
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 1 Selected IRa and NMRb spectroscopic data for new compounds
Compound
n(CO)/cm-1
dP
dH m-H [JHP
]
dC m-C [JCP]
[Mo2Cp2(m-Cl)(m-PCy2)(CO)2] (1a)
[Mo2Cp2(m-Cl)(m-PEt2)(CO)2] (1b)
1880m, sh, 1844sc
1885m, sh, 1849sc
1897m, sh, 1863sc
1890m, sh, 1867sc
1657w, 1568vse
1580vse
155.6d
135.0d
135.2
347.4
206.5e
[Mo2Cp2(m-Cl)(m-PPh2)(CO)2] (1c)
[Mo2Cp2(m-Cl){m-P(OEt)2}(CO)2] (1d)
Li[Mo2Cp2(m-PCy2)(m-CO)2] (Li-2a)
Na[Mo2Cp2(m-PCy2)(m-CO)2] (Na-2a)
K[Mo2Cp2(m-PCy2)(m-CO)2] (K-2a)
PPN[Mo2Cp2(m-PCy2)(m-CO)2] (PPN-2a)
Li[Mo2Cp2(m-PEt2)(m-CO)2] (Li-2b)
Na[Mo2Cp2(m-PPh2)(m-CO)2] (Na-2c)
Li[Mo2Cp2{m-P(OEt)2}(m-CO)2] (Li-2d)
Na[Mo2Cp2{m-P(OEt)2}(m-CO)2] (Na-2d)
K[Mo2Cp2{m-P(OEt)2}(m-CO)2] (K-2d)
[Mo2Cp2(m-H)(m-PCy2)(CO)2] (3a)f
[Mo2Cp2(m-H)(m-PEt2)(CO)2] (3b)
1605m, 1580vse
1643m, sh, 1624vse
1660m, 1572vs, 1556 m, she
1664m, 1604vs, 1593 s, she
1671s, 1584vs, 1568 s, she
1628m, sh, 1609vse
1628m, sh, 1615vse
1858m, sh, 1829vs
1868m, sh, 1833vs
1894m, sh, 1853vsc
1899m, sh, 1862vsc
1674s
202.5e
195.6e
334.8e
232.3
203.2
186.5
322.9
228.5
227.8
228.5
227.8
227.8
222.8e
159.6g
165.7
196.4
-6.94 [11]
-6.61 [8]
-6.85 [8]
-5.26 [11]
[Mo2Cp2(m-H)(m-PPh2)(CO)2] (3c)
[Mo2Cp2(m-H){m-P(OEt)2}(CO)2] (3d)
[Mo2Cp2(m-COMe)(m-PCy2)(m-CO)] (4e)
[Mo2Cp2(m-COEt)(m-PCy2)(m-CO)] (4f)
[Mo2Cp2{m-COC(O)Ph}(m-PCy2)(m-CO)] (4g)
[Mo2Cp2(m-COiPr)(m-PCy2)(m-CO)] (4h)
[Mo2Cp2(m-COCy)(m-PCy2)(m-CO)] (4i)
[Mo2Cp2(m-COI)(m-PCy2)(m-CO)] (5)
[Mo2Cp2(m-Br)(m-PCy2)(CO)2] (6j)
[Mo2Cp2(m-I)(m-PCy2)(CO)2] (6k)
352.0 [15]
351.0 [15]
342.4 [17]
1673s
1738m, 1684s
1669s
1672s
1698se
1874m, sh, 1834vs
1869m, sh, 1832vs
1887s
cis-[Mo2Cp2(m-COMe)(m-PCy2)(O)(CO)] (7)
346.8 [7]
1
1
a Recorded in dichloromethane solution, unless otherwise stated. b Recorded at 300.13 (1H), 121.50 (31P{ H}) or 75.47 MHz (13C{ H}) at 290 K in CD2Cl2
solution. d In C6D6 solution. e In THF solution. f Data taken from ref. 9f . g In CDCl3 solution.
solutions unless otherwise stated, d in ppm relative to internal TMS (1H, 13C{ H}) or external 85% aqueous H3PO4 (31P{ H}), J in Hz. c In toluene
1
1
Solution structure of the unsaturated anions 2: ion pairing effects
The IR spectra of the different salts of the anions 2 prepared
by us clearly reveal the presence of only bridging carbonyls in
all cases. These spectra are quite sensitive to the nature of the
alkaline counterion, an indication of the presence of significant
ion pairing effects, as usually found for transition-metal carbonyl
anions.16
The best image of the unperturbed anion is expected to be given
by the IR spectrum of the PPN+ salts (or related large cations),
since cation–anion interactions are very weak in these cases. In
this sense, the IR spectrum of PPN-2a displays C–O stretching
bands at 1643 (m, sh) and 1624 (vs) cm-1. The very low frequency
of these bands is of course indicative of the bridging position of
the carbonyl ligands in an anionic species, while their relative
intensities suggest that both carbonyl ligands define an angle
Fig. 1 DFT-optimized structure of the anion 2a, with H atoms omitted
for clarity.10a Selected bond distances (A): Mo–Mo 2.499; Mo1–CO 2.120,
˚
intermediate between 90 and 180◦, as expected for two CpMo
moieties bridged by a phosphide and two carbonyl ligands. This
is in excellent agreement with the experimental data recorded for
the isoelectronic (although neutral) and structurally characterized
phosphinidene complex [Mo2Cp2(m-PR*)(m-CO)2], [R* = 2,4,6-
2.089; Mo2–CO 2.091, 2.118; Mo–P 2.437, 2.439.
The formal replacement of the “non-interacting” PPN+ cation
by the strongly polarizing alkaline cations causes the following
changes in the IR spectra of these anions (see Table 1): (a) The
frequency of both C–O bond stretches is reduced substantially
and by a similar amount (40–60 cm-1) in each case, although
in some of the salts the weaker band (symmetric C–O stretch)
disappears into the strong band of the asymmetric C–O stretch.
(b) The frequencies are lower for the more polarizing cations, the
ordering being Li+ < Na+ < K+. (c) For the same cation, the
frequencies follow the expected ordering considering the electron-
releasing properties of the substituents at the phosphorus ligands,
t
C6H2 Bu3; n(CO) 1741 (m), 1709 (vs)],17 and with the computed
C–O stretches in the DFT-optimized structure of the anion 2a in
the gas phase (Fig. 1, n(CO) 1770 and 1756 cm-1, with 34 : 100
relative intensity).10a Moreover, we note that the above theoretical
calculation gives support to the triple intermetallic bond that is
proposed for these 30-electron anions on the basis of the EAN rule.
More precisely, the intermetallic bond in 2a is better described as
composed of one s and two d components, the latter having some
p backbonding (Mo to CO) character.10a
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 8171–8182 | 8173
©

View Article Online
with that ordering being Cy < Et < Ph < OEt. (d) The Li+ salts
exhibit at least an additional, high-frequency band at ca. 1660–
1670 cm-1, with its relative intensity increasing in the sequence
2a < 2b < 2d. The intensity of this band seems to correlate with
the appearance of a low-frequency band, barely observed as a
shoulder of the most intense of all C–O stretching bands present
in each spectrum. A similar effect can be appreciated for the Na+
salt of the anion 2c, but not for the other Na+ salts prepared
by us.
the relative abundance of single-carbonyl interactions of type B in
solution, and we thus conclude that single-carbonyl interactions
appear to be more favoured for the complexes having the less-
donor phosphide ligands (PPh2 and P(OEt)2). We finally note that
even when the IR spectra of these salts denote the presence of one
1
or several species, the corresponding 31P{ H} NMR spectra exhibit
a single resonance in each case, thus implying rapid exchange
among the different ion pairs on the NMR timescale.
The above features in the IR spectra of the alkaline salts of
the anions 2a–d can be explained by taking into account the
presence of several cation-oxygen interactions in the tetrahydro-
furan solutions of these salts. Although most of the studies on
this type of interactions have been carried out on mononuclear
carbonyl anions,16 a recent X-ray diffraction study on two
Synthesis and structural characterization of the hydride
derivatives 3
As expected, the anions 2a–d can be easily protonated to give
hydride derivatives. This takes place readily using a weak acid
such as [NH4]PF6 and gives with good yield the corresponding
hydrides [Mo2Cp2(m-H)(m-PA2)(CO)2] (3a–d), which are species of
low to moderate sensitiveness to air. The above reactions imply
the incorporation of the entering proton to the intermetallic
region and rearrangement of the carbonyl ligands from bridging
to terminal coordination modes. We examined with special detail
(changes in temperature, use of other acids, etc.) the protonation
of the dicyclohexylphosphide-bridged anion 2a in search for
possible intermediates in the formation of the above hydrides, i.e.
hydroxycarbyne complexes derived from initial protonation at the
oxygen atom of the carbonyl ligands,10a,19 but found no evidence
for the presence of any intermediate species.
The structure of the dicyclohexylphosphide-bridged hydride
complex 3a has been previously studied in detail by us in solution
and in the solid state (Fig. 2), both experimentally and using
DFT methods, and then needs no further discussion.9a,f Since
the spectroscopic data for the hydrides 3b–d are very similar
to those of 3a (see Table 1 and Experimental section), then
we can safely assume that all these species are isostructural.
According to the EAN formalism, a triple intermetallic bond
must be formulated for these 30-electron hydride complexes and,
by following the results of the mentioned DFT calculations on
3a, this intermetallic binding can be more precisely described
as composed of a tricentric (Mo2H) plus two bicentric (Mo2)
interactions, the latter being of s and p types.9f The hydride ligand
is thus placed on the nodal plane of the pMM bonding orbital
and then would experience a strong deshielding effect due to the
+
different Li(THF)n salts of the electron-precise binuclear anion
[Fe2(CO)8]2- is particularly revealing of the solution structure of
the anions 2.18 This study has identified two types of Li ◊ ◊ ◊ O
interactions involving the bridging carbonyl ligands in the solid
state, one involving a single carbonyl, with the lithium cation
completing its coordination sphere with three THF molecules, and
one involving two carbonyls, with the lithium cation completing
its coordination sphere with two THF molecules. In the case of the
anions 2, the solution data can be explained by assuming that the
second type of interaction prevails in the solutions of these salts,
effectively creating sorts of polymeric chains of anions connected
by M(THF)n+ cations through the oxygen atoms of both carbonyl
ligands of each anion (A in Chart 1; n = 2 for Li). Since M+ ◊ ◊ ◊ OC
interactions are well known to decrease the stretching frequency
of the involved carbonyl ligands, and to be stronger for the lighter
alkaline cations,16 then an interaction of type A would explain the
major features of the spectra of the alkaline salts of anions 2, that
is, a generalized frequency decrease of both C–O stretching bands
when compared to the PPN+ salt, and a stronger decrease for the
lighter cations (effects a and b, see above). To explain the minor
features in the IR spectra of the Li+ salts (and in Na-2c) we assume
that the smaller cations are able, to some extent, to be involved
in interactions with a single carbonyl ligand (B in Chart 1; n =
3 for Li). This would lead to a decrease in the C–O stretching
frequency of the involved carbonyl as above, but to an increase in
the C–O stretching frequency of the non interacting carbonyl, thus
explaining the appearance of the high frequency bands (effect d,
see above). Indeed, the alkylation at the oxygen atom of the anions
2 would be an extreme version of this interaction, and this leads
to stable alkoxycarbyne complexes exhibiting C–O stretches only
marginally more energetic (compounds 4 in Table 1), as it will
be discussed later on. Under this view, the relative intensity of the
high-frequency band in the IR spectra of our Li+ salts would reflect
Fig. 2 ORTEP diagram (25% probability) of compound 3a, with
H atoms (except the hydride ligand) and Cy rings (except their C1
atoms) omitted for clarity.9a Selected bond lengths (A) and angles
˚
(◦): Mo(1)–Mo(2) 2.528(2), Mo(1)–P(1) 2.382(3), Mo(2)–P(1) 2.386(3),
Mo(1)–C(1) 1.918(14), Mo(2)–C(2) 1.901(7); C(1)–Mo(1)–Mo(2) 79.7(4),
C(2)–Mo(2)–Mo(1) 80.1(2).
Chart 1
8174 | Dalton Trans., 2009, 8171–8182
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 2 Product distribution in the reactions of Li-2a with different
magnetic anisotropy of such a bond, assumed to be qualitatively
20
C-based electrophilesa
analogous to that of the well studied double C+ C bonds.
=
This explains the abnormally high (for a bridging ligand) hydride
chemical shifts of the unsaturated compounds 3, in the range -5.3
to -6.9 ppm, to be compared with the more usual values of -11
to -13 ppm measured for the related electron-precise complexes
[Mo2Cp2(m-H)(m-PRR¢)(CO)4].12,14,21 As for the 31P resonances,
the dialkyl or diarylphosphide hydrides 3a–c exhibit resonances
much more deshielded than those of the corresponding chloro-
complexes 1a–c, in spite of the much lower electronegativity of the
hydrogen atom, compared to chlorine. In any case, this follows a
persistent experimental trend found by us when comparing pairs
of 30- and 32-electron complexes of the types [Mo2Cp2(m-PR2)(m-
X)(m-CO)]/trans-[Mo2Cp2(m-PR2)(m-X)(CO)2] (X = 3-electron
donor ligand),14,22 and trans-[Mo2Cp2(m-PR2)(m-X)(CO)2]/trans-
[Mo2Cp2(m-PR2)(m-Y)(CO)2] (Y = one-electron donor ligand).9f In
contrast, the diethoxyphosphide ligand does not follow the above
spectroscopic trend, and in fact the 31P chemical shift of the hydride
3d is somewhat lower than that of the chloro-complex 1d. This is
just the ordering expected on the basis of the electronegativity
differences in the bridging groups if the nuclear shielding were to
be dominated by the changes in the diamagnetic contribution, a
circumstance generally not granted for the phosphorus nuclei.23
The compounds 3a–d are still rare examples of transition-metal
carbonyl complexes displaying a bridging H ligand over a formally
triple M–M bond. Precedents for such 30-electron compounds are
restricted to the monocarbonyls [M2Cp*2(m-H)(m-CO)] (Cp* =
Product distribution (%)
Reagent
m-R
m-COR
m-COI
m-X
Other
Cl–CH2Ph9f
Cl–CH2C(O)Me
Cl–CH2CN
Cl–CH2SPh
Cl–CPh3
90
10
95
90
10
100
5
10
90b
Cl–C(O)Ph
90
10
Br–CH2Me
Br–CHMe2
3-Br–cyclohexene
(60)
(20)
(40)
(80)
20
80
I–CH3
I–CH2Me
I–Cy
90 (85)
10 (0)
25 (50)
100 (75)
100
0 (15)
0 (30)
0 (10)
15 (15)
0 (15)
60 (5)
I–CH2I
Me2SO4
Et2SO4
75 (80)
90
25 (20)
10
a Relative amounts of products estimated from the 31P{ H} NMR spectra
of the corresponding reaction mixtures. The values in parentheses cor-
respond to the results of the reaction when using the K-2a salt. b The
major species formed was identified as [Mo2Cp2(m-PCy2)(m-SPh)(CO)2]
(see ref. 21).
1
Cl, Br or I and R ranging from simple alkyl groups (Me, Et, Cy,
etc) to other ones including different functional groups, such as
CH2Y (Y = CN, SPh, etc). We will also discuss the reactions with
a few stronger electrophiles such as benzoyl chloride and sulfate
esters (Me2SO4 and Et2SO4). Besides this, in a few selected cases
we have carried out the same reaction not only using the usual
Li-2a salt but also the salt K-2a, so as to examine the influence of
the alkaline counterion in the product distribution.
The results of the above reactions are collected in Table 2, along
with an approximate product distribution in each case, as deduced
from the 31P{ H} NMR spectra of the corresponding reaction
5
h -C5Me5; M = Ru,24a,b Os24c), the cations [W2Cp2(m-H)(CO)2(m-
R2PCH2PR2)]+ (R = Me, Ph),19 and the anion [Re2(m-H)3(CO)6]-,8
but none of these species has been studied in detail. In contrast,
we have been able to develop an extensive chemistry around the
dicyclohexylphosphide-bridged hydride 3a, as mentioned above.9
Reactions of the dicyclohexylphosphide-bridged anion 2a with
C-based electrophiles
1
In our preliminary report we found out that the anion 2a
seemed to exhibit two distinct nucleophilic sites located at the
Mo and O atoms, since the room-temperature reaction with
two different sources of the methyl cation, MeI and [Me3O]BF4,
yielded two isomers, these being the agostic methyl complex
[Mo2Cp2(m-Me)(m-PCy2)(CO)2] and the methoxycarbyne com-
plex [Mo2Cp2(m-COMe)(m-PCy2)(m-CO)] (4e), respectively.7 Sub-
sequent independent experiments revealed that neither of these
complexes transformed into each other, even in refluxing toluene
or under photochemical activation. Therefore it was clear that
both the metal atoms and the oxygen atoms of the carbonyl
ligands in the anion 2a were genuine primary nucleophilic sites
of the molecule. The critical influence on the course of these
reactions of the electrophilic reagent used was further confirmed
by taking into account that Et2SO4 reacted with 2a to yield
the ethoxycarbyne derivative [Mo2Cp2(m-COEt)(m-PCy2)(m-CO)]
(4f),7 whereas benzyl chloride gave the agostic benzyl complex
[Mo2Cp2(m-CH2Ph)(m-PCy2)(CO)2].9f We then decided to examine
with some more detail the reactivity of the anion 2a (usually as the
Li+ salt) towards a wider set of C-based electrophiles, so as to find
out the dominant factors favouring the electrophilic attack at the
Mo and O sites of the anion. In this paper we report our results on
the reactions of 2a with different alkyl halides RX, with X being
mixtures. Some of the reactions were quite selective, but others
led to complex mixtures of products. Overall, we have been able
to identify four types of products (Scheme 2): (a) the mentioned
agostic products [Mo2Cp2(m-R)(m-PCy2)(CO)2] (R = Me, CH2Ph),
derived from the incorporation of the hydrocarbon fragment to
the dimetal centre, (b) the alkoxycarbyne products [Mo2Cp2(m-
COR)(m-PCy2)(m-CO)] [R = Me (4e), Et (4f), C(O)Ph (4g), iPr (4h),
Cy (4i)], derived from the binding of the hydrocarbon fragment
to the oxygen atom of a carbonyl ligand, (c) the iodoxycarbyne
complex [Mo2Cp2(m-COI)(m-PCy2)(m-CO)] (5), derived from the
binding of a I+ cation to the oxygen atom of a carbonyl ligand,
and (d) the halide complexes [Mo2Cp2(m-PCy2)(m-X)(CO)2] [X =
Cl (1a), Br (6j), I (6k)], that might be formed through two different
pathways, as it will be discussed later on. As expected, the halide
complexes are more conveniently prepared by direct reaction of
the anion 2a with the elemental halogens. All other unidentified
products detected in the corresponding reaction mixtures are
grouped in Table 2 under the entry “others”, this including
the thiolate complex [Mo2Cp2(m-SPh)(m-PCy2)(CO)2], previously
prepared by us using a different synthetic route.21
All the alkoxycarbyne complexes 4e–i could be conveniently
purified by column chromatography. They were found to be
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 8171–8182 | 8175
©

View Article Online
changes in the hydrocarbon group has dramatic consequences;
thus while isopropyl bromide gives no detectable amounts of the
bromo-complex 6j, 3-bromocyclohexene gives 6j as major product.
(d) The reactions with alkyl iodides are the most complex ones; in
all cases low to high amounts of the iodo-derivative 6k are formed
(100% when using diiodomethane), but the hydrocarbon group
has a strong influence on the formation of other products. Thus,
the reaction with MeI gives only the methyl derivative, whereas
the very similar EtI gives a small amount of the ethoxycarbyne
4f and the iodoxycarbyne complex 5 as major product, with no
ethyl derivative being formed in detectable amounts. Complex 5 is
rather unstable, and it slowly rearranges in solution (rapidly upon
any simple manipulation such as removal of the solvent) to give the
iodo-complex 6k, thus making difficult to extract any mechanistic
conclusions from all our reactions involving alkyl iodides. We
should note that, to our knowledge, no other haloxycarbyne
(COX) complex appears to have been previously reported in the
literature. (f) Concerning the influence on the alkaline counterion,
it is generally observed that K-2a reacts faster than Li-2a, and has a
beneficial influence in the formation of alkoxycarbyne derivatives.
Structural characterization of the carbyne derivatives 4 and 5
Scheme 2 Reactivity of the anion 2a towards alkyl halides.
The solid-state structure of the ethoxycarbyne complex 4f was
determined in our preliminary study by X-ray diffraction methods
and is depicted in Fig. 3 along with a few selected bond
distances and angles.7 A separated DFT calculation on the gas-
phase structure of the methoxycarbyne analogue 4e gave excellent
agreement with the above experimental solid-state structure, and
allowed a precise analysis of the M–M and M–C interactions in
these molecules.10a According to the EAN formalism, a triple inter-
metallic bond must be formulated for these 30-electron complexes
and, by following the results of the mentioned DFT calculation,
this intermetallic binding can be more precisely described as
composed of a s and two d M–M interactions, the latter being
involved in p bonding with the carbyne ligand and, to a lesser
extent, in p backbonding with the bridging carbonyl. All of this is
in agreement with the very short intermetallic separation measured
moderately air-sensitive, and we have fully characterized the main
product formed by the action of air on the dicyclohexylphosphide-
bridged compound as the oxo-complex cis-[Mo2Cp2(m-COMe)(m-
PCy2)(O)(CO)] (7) (Chart 2). We finally note that compounds 4
are the first reported and yet unique neutral complexes displaying
alkoxycarbyne bridges along a triple intermetallic bond. In fact,
the only other 30-electron complexes previously reported in the
literature are the cations [M2Cp2(m-COR)(m-PR¢2)2]+ (M = Mo,
W; R = H, Me; R¢ = Ph, Et), also prepared in our laboratory.10a.25
˚
in 4f (2.477(2) and 2.479(2) A in the two independent molecules),
Chart 2
The analysis of the product distribution of all the above reac-
tions indicates that there are many subtle factors changing dramat-
ically their overall course, thus making very difficult the rational-
ization of all observations. Yet a few general trends can be appre-
ciated: (a) In the absence of halogen at the reagent (reactions with
Me2SO4 and Et2SO4), nucleophilic attack of the anion through its
oxygen atoms prevails, to give the corresponding alkoxycarbyne
derivative. (b) Among the hydrocarbon halides studied, the chlo-
rides exhibit the largest tendency to give the corresponding halide
complex, with one of the most selective reactions being that with
ClCPh3 to give the chloro-complex 1a as the unique product. In-
terestingly, separated experiments revealed that 2 equiv. of reagent
were needed to achieve the complete disappearance of the anion,
an observation of relevance to propose a likely reaction pathway,
as it will be discussed later on. (c) The formation of halo-complexes
seems to be minimized when using alkyl bromides, but very small
Fig. 3 ORTEP diagram (25% probability, only one of the two in-
dependent molecules shown) of compound 4f, with H atoms and Cy
rings (except their C1 atoms) omitted for clarity.7 Selected bond lengths
◦
˚
(A) and angles ( ): Mo(1a)–Mo(2a) 2.479(2), Mo(1a)–P(1a) 2.394(2),
Mo(2a)–P(1a) 2.393(2), Mo(1a)–C(1a) 2.086(5), Mo(2a)–C(1a) 2.080(5);
Mo(1a)–C(2a) 1.976(5); Mo(2a)–C(2a) 2.014(5); C(2a)–O(2a) 1.329(5);
O(2a)–C(3a) 1.445(5); C(3a)–C(4a) 1.460(6); C(2a)–O(2a)–C(3a) 117.9(4).
8176 | Dalton Trans., 2009, 8171–8182
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
with values comparable to or even shorter than those measured
in other triply-bonded complexes lacking carbyne bridges, such
donor–acceptor interaction with the solvent (m-CO–I ◊ ◊ ◊ OC4H8),
similar to the interaction leading to the well-known charge-
transfer complexes of the iodine molecule.30 This would explain the
rapid decomposition of this molecule when the solvent is removed
in vacuum and the experimental finding that this complex can be
only detected in significant amounts when using iodide-free and
boron-free solutions of the salt Li-2a, since the presence of any
potential donor (I-, excess [BHEt3]-) would remove the mentioned
weak solvent–iodine interaction. At present, however, we cannot
determine whether the rearrangement of the iodoxycarbyne
complex 5 into its isomer 6k is an intramolecular process or it
is assisted by the presence of external reagents. We note, however,
that this rearrangement is completely analogous to that observed
previously by us on some 32- and 30-electron hydroxycarbyne-
bridged complexes to give the corresponding hydridocarbonyl
derivatives.10a,19 It is to be seen from future studies whether, by
using the appropriate electrophiles, other p-block-element based
groups (E) different from hydrocarbons can be attached to the
oxygen atom of the anion 2a, to give neutral derivatives having
novel (m-C–O–E) carbyne ligands.
26a
˚
as [Mo2Cp2(CO)4] (2.4477(12) A), [Mo2Cp2(m-PPh2)2(m-CO)]
26b
17
˚
˚
(2.515(2) A), and [Mo2Cp2(m-PR*)(m-CO)2] (2.5322(3) A).
As for the alkoxycarbyne ligand, the Mo–C(carbonyl) lengths
˚
(ca. 2.08 A) are longer than the Mo–C(carbyne) distances (ca.
˚
2.00 A), in agreement with the change in the formal order of
the corresponding bonds, which increase from 1 in the bridging
carbonyl to 1.5 in the bridging ethoxycarbyne ligand. The latter
exhibits a slightly asymmetric coordination, being closer to the
metal centre placed away from the ethyl group (Mo1a in Fig. 3),
which is possibly a genuine steric effect. Besides this, the relatively
short value of the carbyne C–O length (ca. 1.33 A), somewhat
shorter than those found for comparable single bonds, such as the
2
C(sp )-O bonds in organic molecules (ca. 1.35 A),27 can be taken as
˚
indicative of the presence of some multiplicity in this bond. This
is in turn consistent with the conformation of the COEt ligand
(arrangement in the Mo2C plane, with a C–O–C angle close to
120 ◦), and is also in agreement with the theoretical analysis of the
bonding in 4e mentioned above.
The similarity of the spectroscopic data in solution for all com-
pounds 4e–i suggests that they all have the same structure, compa-
rable to that found in the solid state for 4f. The most characteristic
spectroscopic features are the presence in the corresponding IR
spectra of a band at ca. 1675 cm-1 due to the C–O stretch of the
bridging carbonyl, and a 31P resonance at ca. 228 ppm, a chemical
shift comparable to those of the dicyclohexylphosphide ligand in
the isoelectronic complexes [Mo2Cp2(m-PCy2)(m-PR¢2)(m-CO)],14
while the bridging alkoxycarbyne ligands give rise to a strongly
deshielded 13C resonance at ca. 350 ppm, a chemical shift not very
different from the values reported for electron-precise compounds
having alkoxycarbyne bridges.28 Except for the cyclohexyloxy-
carbyne complex 4i, all these compounds were found to exhibit
dynamic behaviour in solution, as deduced in each case from the
Structure of the oxo-complex 7
The reaction of 4e with oxygen reproduces those of the iso-
electronic complexes of type [M2Cp2(m-PPh2)2(m-CO)] (M =
Mo,26b W),25 to give the corresponding oxo-derivatives [M2Cp2(m-
PPh2)2(O)(CO)], mainly as their cis isomers. The spectroscopic
data of compound 5 share many common features with those
of the above oxo-complexes, such as the presence of a single
C–O stretching band in the region of terminal carbonyls, and
two well separated 13C resonances (differing by some 10 ppm)
for the inequivalent Cp groups as a result of the substantial
difference in the environments around both metal atoms. The
presence of the oxygen atom has no great influence, however, on
the 13C carbyne resonance of 7, which is only 5 ppm shielded with
respect to that one in the parent complex. Another spectroscopic
feature shared with the mentioned oxo-complexes is the presence
1
presence in the 1H and 13C{ H} NMR spectra of just one resonance
for both cyclopentadienyl rings, which are inequivalent in the static
structure (Fig. 3). This fluxional behaviour can be attributed to the
fast rotation, on the NMR timescale, of the OR groups around
the O–C(carbyne) bond, and it has been previously studied in
detail for other alkoxycarbyne complexes,19,29 therefore this was
not investigated further. Apparently, the cyclohexyl group is large
enough to impose some restrictions to the mentioned rotation so
as to significantly decrease its rate, thus rendering distinguishable
1H NMR resonances for the inequivalent cyclopentadienyl groups
at room temperature (see Experimental section).
of a quite deshielded resonance in the 31P{ H} NMR spectrum,
1
only some 30 ppm below that of its triply-bonded precursor.
This would be indicative of a metal–metal bond order close
to 1, according to the general trends observed for all these
binuclear dialkyl or diarylphosphide-bridged complexes,14 since
the doubly-bonded 32-electron complexes usually give rise to
much more shielded resonances. All of this can be conciliated
by assuming that the terminal oxo ligand in these complexes is
effectively behaving as a four-electron donor, rather than a two-
electron donor, thus rendering an electron-precise complex with
a single metal–metal bond (Chart 2). Indeed, the intermetallic
distances measured in the crystal for the oxo-complexes trans-
1
The IR and 31P{ H} NMR spectra of compound 5 (Table 1)
are very similar to those of the alkoxycarbyne complexes 4. Thus
the C–O stretching band of 5 (1698 cm-1) appears just in the
same position than that of the minor alkylation product of the
reaction with EtI (i.e. the ethoxycarbyne 4f, nCO 1698 cm-1 in
THF solution), but its 31P resonance (dP 222.8 ppm) appears
a bit more shielded than that of 4f (dP 223.3 ppm in THF
solution). We therefore trust that the structure of compound 5 is
essentially identical to that of the above alkoxycarbyne complexes
4, although its stability is much lower. As mentioned above, this
complex is only moderately stable, since it slowly rearranges in
tetrahydrofuran solution (rapidly upon any simple manipulation
such as removal of the solvent) to give the iodo-complex 6k. It is
conceivable that the O–I binding in 5 could be stabilized through a
26b
˚
[Mo2Cp2(m-PPh2)2(O)(CO)] (2.942(1) A) and trans-[Mo2Cp2(m-
1
2
h :h -CHCHPh)(m-PPh2)(O)(CO)] (2.885(1) A)31 are substantially
˚
˚
longer than the values of ca. 2.70 A expected for double Mo–Mo
or W–W bonds, and then give substantial support to this view.
Finally, the identification of compound 7 as the isomer displaying
a cisoid arrangement of the Cp ligands can be made taking into
account that the treatment with air of the dicarbonyl carbyne
complex trans-[Mo2Cp2(m-COMe)(m-PCy2)(CO)2]10b gives mainly
an isomer of 7, but this exhibiting a C–O stretching band at
lower frequency (1871 cm-1 in dichloromethane solution). This
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 8171–8182 | 8177
©

View Article Online
reproduces the behaviour of the diphenylphosphide complex trans-
[Mo2Cp2(m-PPh2)2(CO)2], which reacts slowly with air to give
mainly trans-[Mo2Cp2(m-PPh2)2(O)(CO)], a compound displaying
a C–O stretching band at lower frequency than its cis-isomer
(1826 cm-1 vs. 1859 cm-1 in dichloromethane solution).26b
O sites to give carbyne derivatives. This explains the dominant
formation of carbyne derivatives when using alkyl sulfates or
benzoyl chloride. For the other alkyl halides, this is a minor or
non-operative pathway except for the ethyl halides, with EtBr
giving the ethoxycarbyne derivative and EtI giving a mixture of the
ethoxycarbyne and the iodoxycarbyne derivatives. The latter might
be perhaps explained by recalling that carbon and iodine have not
very different electronegativities, so the nucleophilic displacement
could possibly take place in both directions. Indeed, the formation
of iodo-complexes from mononuclear carbonylates has been
reported when using organic halides having strong electron-
withdrawing groups, such as perfluoroaryl iodides,32 although this
would not be an expected pathway for ethyl iodide. Alternatively,
the formation of 5 might perhaps follow from an electron-transfer
process, as discussed below.
For most of the alkyl halides used, we really should expect
the reaction to be orbital-controlled and therefore bound to give
the pertinent alkyl-bridged derivatives. However, this only occurs
in the reactions with benzyl chloride and methyl iodide. The
reactions of 2a with all other reagents give the corresponding
halide derivatives as major products. The fact that one of the most
selective reactions in this group (that with ClCPh3 to give the
chloro-complex 1a quantitatively) was experimentally verified to
require 2 equiv. of reagent suggests to us that an electron-transfer
process might be responsible for the formation of most of the
halo-complexes in these reactions. This could be accomplished
in two steps involving radical species, with the first step being an
electron transfer yielding a 29-electron neutral radical [Mo2Cp2(m-
PCy2)(CO)2] and (XR)-, the latter evolving as X- and R∑ (eqn
(1)). In a second step the formed organometallic radical would
react with a second molecule of XR by trapping preferentially
the halogen atom (eqn (2)). Atom-transfer processes are one
of the characteristic reactions of mononuclear organometallic
radicals,33 and we have previously shown that this is also the
case of some unsaturated binuclear radicals such as the 33-
electron diphenylphosphide complex [Mo2Cp2(m-PPh2)(CO)4],34
and the isoelectronic diphosphine-bridged cations [Mo2Cp2(m-
CO)2(CO)2(m-R2PCH2PR2)]+ (R = Me, Ph).35 Interestingly, these
paramagnetic complexes react with thiols and organic disulfides to
give the corresponding thiolate-bridged derivatives. Analogously,
the radical initially formed after oxidation of 2a could react with
ClCH2SPh to give the corresponding thiolate derivative (eqn (3)).
In this case, the cleavage of the C–S bond would prevail over that
of the C–Cl bond because of its lower bond strength.
Halogenation reactions of the anion 2a
Although the bromo- and iodo-complexes 6j,k are obtained in
variable amounts in the reactions of 2a with the hydrocarbon
halides discussed above, they are best prepared by the direct action
of the corresponding elemental halogens, this reaction taking place
instantaneously at room temperature and giving complexes6 as the
unique organometallic products. As it is the case of the analogous
chloro-complex 1a, these compounds are quite air-sensitive, and
only the iodo-complex 6k could be isolated as a pure material after
chromatographic workup. In any case, the spectroscopic data for
compounds 6j,k are very similar to those of the chloro-complex
1a and need not to be discussed in detail again. We just note that
the observed frequencies of the C–O stretching bands follow the
expected trends considering the relative electronegativities of the
halogen atoms, this yielding the order 1a > 6j > 6k, but the 31P
chemical shifts follow the opposite order, with a stepwise increase
of ca. 5 ppm when going from the chloro- to the iodo-complex.
This is not surprising if we take into account that the chemical
shifts of phosphorus are not dominated by the diamagnetic, but
rather by the paramagnetic contribution, and therefore bear no
linear relationship with the electron density at the corresponding
nuclei.23
Pathways in the reactions of the anion 2a with C-based
electrophiles: concluding remarks
From the synthetic data discussed in the preceding sections it is
clear that the reactions of 2a under study are quite complex and
that several competitive reaction pathways are operative, with the
observed products resulting from a delicate balance between the
chemical properties of the anion and those of the hydrocarbon
electrophile in each case. From the nature and distribution of
the observed products (Scheme 2 and Table 2) we conclude that
there are at least three primary reaction pathways: (a) electron
transfer resulting in the formation of the halide derivatives, (b)
nucleophilic substitution through the metal site of the anion to give
hydrocarbyl-bridged products, and (c) nucleophilic substitution
through the oxygen sites of the anion. The latter mechanism has
in turn two possibilities: (c1) displacement of the halide ion to
give alkoxycarbyne derivatives, and (c2) at least in the case of the
ethyl iodide, displacement of the hydrocarbyl anion to give the
iodoxycarbyne derivative.
The pathways (b) and (c) are the most straightforward ones if we
consider the electronic structure computed for the anion 2a, this
revealing that the highest occupied molecular orbitals are metal-
based orbitals, while the largest negative charge is accumulated
at the oxygen atom of the carbonyl ligands.10a Thus we could
predict that an electrophilic fragment should be incorporated at
the dimetal site under conditions of orbital control, but instead
should bind the oxygen atom under conditions of charge control.
As a result, we should expect the hydrocarbon reagents bearing
the highest positive charge at carbon to react with 2a at the
[Mo2Cp2(m-PCy2)(CO)2]- + RX →
[Mo2Cp2(m-PCy2)(CO)2]∑ + R∑ + X-
(1)
(2)
(3)
(4)
[Mo2Cp2(m-PCy2)(CO)2]∑ + RX →
[Mo2Cp2(m-X)(m-PCy2)(CO)2] + R∑
[Mo2Cp2(m-PCy2)(CO)2]∑ + ClCH2SPh →
∑
[Mo2Cp2(m-SPh)(m-PCy2)(CO)2] + ClCH2
[Mo2Cp2(m-PCy2)(CO)2]∑ + IEt →
[Mo2Cp2(m-COI)(m-PCy2)(CO)2] + Et∑
On the other hand, since 5 transforms readily into the iodo-
complex 6k, it is not possible from our present data to conclude
whether 6k in all our reactions is formed directly (eqn (2))
8178 | Dalton Trans., 2009, 8171–8182
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
or via its iodoxycarbyne isomer 5. Even more, we cannot
exclude at present that electron-transfer processes (rather than
nucleophilic attack by the anion) could be even responsible
for the formation of the iodoxycarbyne complex 5 in the re-
action of the anion 2a with ethyl iodide. Actually, we have
previously shown that some binuclear radicals such as the 31-
electron cation [W2Cp2(m-CO)(CO)2(m-Ph2PCH2PPh2)]+ and the
33-electron cations [Mo2Cp2(m-CO)2(CO)2(m-R2PCH2PR2)]+ (R =
Me, Ph) can abstract hydrogen atoms at their O(carbonyl) atoms
to give hydroxycarbyne derivatives.36 Thus we cannot exclude that
the radical initially formed after oxidation of 2a could alternatively
react with EtI to give the iodoxycarbyne complex 5 by trapping
a iodine atom at the O(carbonyl) site (eqn (4)). Further studies
will have to be carried out in order to better define the relative
relevance of the electron-transfer and acid–base processes in the
general reactivity of the unsaturated anion 2a.
Preparation of compounds [Mo2Cp2(l-Cl)(l-PA2)(CO)2] (A =
Et, Ph, OEt) (1b-d). In a typical experiment, a solution of the
complex [Mo2Cp2(CO)6] (0.100 g, 0.204 mmol) and a slight excess
of ClPA2 (ca. 0.25 mmol) in 10 ml of toluene (A = Et, Ph) or
diglyme (A = OEt) were refluxed for 30 min or 10 min respectively
to give a dark green solution shown (by IR and 31P NMR) to
contain compounds 1b–d as major species. All attempts to isolate
or further purify these highly air-sensitive complexes resulted in
their progressive decomposition, but removal of solvent from the
reaction mixture gave crude residues ready for further use. dH
(200.13 MHz; C6D6, 1b) 4.94 (10H, s, Cp).
Preparation of THF solutions of [Mo2Cp2(l-PA2)(l-CO)2]- (2a–
d). Method A: In a typical experiment, the crude products
obtained as described above (containing ca. 0.2 mmol of com-
plexes 1a–d) were dissolved in THF (10 ml) and then stirred with
Li[BHEt3] (0.5 ml of a 1.0 M solution in THF, 0.5 mmol; reaction
time: ca. 3 h) or Na(Hg) (ca. 1 ml of a 0.5% amalgam, excess;
reaction time: ca. 15 min), or K[BsBu3H] (0.5 ml of a 1.0 M
solution in THF, 0.5 mmol; reaction time: ca. 30 min) to give dark
red solutions shown (by IR and 31P NMR) to contain the anions
2a–d as major species, as the corresponding Li+, Na+ or K+ salts.
All attempts to isolate these highly air-sensitive complexes resulted
in their progressive decomposition. These solutions, however, were
ready for further use. In the case of the sodium salts, the solutions
can be separated from the amalgam by filtration using a cannula.
Method B.: A solution containing 0.060 g of the hydride complex
3a (0.104 mmol) in THF (10 ml) was stirred with Li[BHEt3] (140 ml
of a 1.0 M solution in THF, 0.14 mmol; reaction time: ca. 15 min)
or K[BsBu3H] (0.5 ml of a 1.0 M solution in THF, 0.5 mmol;
reaction time: ca. 24 h) to give dark-red solutions of the salts
Li-2a and K-2a respectively. Method C.: A solution containing
0.030 g of the hydride complex 3a (0.052 mmol) in THF (10 ml)
was vigorously stirred with Li(Hg) (ca. 1 ml of a 0.3% amalgam,
excess) for 10 min, to give a dark-red solution of the salt Li-2a that
can be separated from the amalgam using a cannula and is free of
any other salt. Method D.: A solution containing ca. 0.2 mmol of
the chloro-complex 2d in THF (10 ml) was vigorously stirred with
Li(Hg) (ca. 1 ml of a 0.3% amalgam, excess) for 15 min, to give a
dark-red solution of the salt Li-2d that can be separated from the
amalgam by using a cannula.
Experimental
General procedures and starting materials
All manipulations and reactions were carried out under a nitro-
gen (99.995%) atmosphere using standard Schlenk techniques.
Solvents were purified according to literature procedures, and
distilled prior to use.37 The hydride complex 3a was prepared as
described previously.9f All other reagents were obtained from the
usual commercial suppliers and used as received, unless otherwise
stated. Petroleum ether refers to that fraction distilling in the
range 338–345 K, diglyme refers to bis(2-methoxyethyl) ether
and (PPN)+ refers to the bis(triphenylphosphine)iminium cation.
Filtrations were usually carried out through diatomaceous earth.
Low-temperature chromatographic separations were carried out
using jacketed columns refrigerated by a closed 2-propanol circuit
kept at the desired temperature with a cryostat, or by tap water.
Commercial aluminium oxide (activity I, 150 mesh) was degassed
under vacuum prior to use. The latter was mixed afterwards under
nitrogen with the appropriate amount of water to reach the activity
desired. IR stretching frequencies of CO ligands were measured
either in solution (using CaF2 windows) or in Nujol mulls (using
NaCl windows), and are referred to as n(CO). Nuclear magnetic
resonance (NMR) spectra were routinely recorded at 300.13 (1H),
1
1
121.50 (31P{ H}) or 75.47 MHz (13C{ H}) at 290 K in CD2Cl2
solutions unless otherwise stated. Chemical shifts (d) are given
in ppm, relative to internal tetramethylsilane (1H, 13C) or external
85% aqueous H3PO4 solutions (31P). Coupling constants (J) are
given in Hertz.
Preparation of THF solutions of (PPN)[Mo2Cp2(l-PCy2)(l-
CO)2] (PPN-2a). Solid (PPN)Cl (0.032 g, 0.075 mmol) was
added to a THF solution containing ca. 0.05 mmol of the salt
Li-2a, prepared as described above (method C), and the mixture
was stirred for 2 min to give a deep red solution shown (by IR and
31P NMR) to contain the (PPN)+ salt of the anion 2a.
Preparation of [Mo2Cp2(l-Cl)(l-PCy2)(CO)2] (1a). Complex
[Mo2Cp2(CO)6] (0.500 g, 1.02 mmol) and ClPCy2 (300 ml,
1.22 mmol) were refluxed in toluene (15 ml) for 30 min with
vigorous stirring to give a dark green solution shown (by IR and
31P NMR) to contain compound 1a as major species. The solvent
was then removed under vacuum and the residue was washed
with petroleum ether (2 ¥ 10 ml) to give complex 1a as a black–
green microcrystalline solid. Due to its high sensitivity to air, this
material was used without further purification and assumed to be
formed quantitatively. dH (200.13 MHz; C6D6) 5.03 (10H, s, Cp),
2.50–0.85 (22H, m, Cy).
Preparation of [Mo2Cp2(l-H)(l-PEt2)(CO)2] (3b). Solid
[NH4]PF6 (0.040 g, 0.245 mmol) was added to a THF solution
of Li-2a, prepared from [Mo2Cp2(CO)6] (0.100 g, 0.204 mmol) as
described above, and the mixture was stirred for 10 min to give
a dark-brown solution. Solvent was then removed under vacuum
and the residue then chromatographed on alumina (activity IV) at
253 K. Elution with dichloromethane–petroleum ether (1 : 3) gave
a brown fraction which yielded, after removal of solvents under
vacuum, compound 3b (0.057 g, 60% overall yield) as a brown, air-
sensitive powder (Found: C, 41.9; H, 4.0. Calc. for C16H17Mo2O2P:
C, 41.5; H, 3.7%). dH (200.13 MHz) 5.04 (10H, s, Cp), 2.57, 2.24
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 8171–8182 | 8179
©

View Article Online
(2 ¥ 2H, 2 ¥ m, CH2), 1.31 (6H, dt, JPH 16, JHH 7, CH3), -6.61 (1H,
d, JPH 8, m-H).
0.374 mmol) for 20 min without noticeable change in colour.
Workup of the mixture as described for 4e (method A) gave
compound 4f (0.027 g, 45%) as a red solid (Found: C, 51.4; H,
6.0. Calc. for C26H37Mo2O2P: C, 51.7; H, 6.2%). dH 5.72 (10H, s,
Cp), 3.92 (2H, q, JHH 7, OCH2), 1.28 (3H, t, JHH 7, CH3), 1.90–
0.30 (22H, m, Cy). dC (100.63 MHz) 351.0 (d, JCP 15, m-COEt),
303.5 (d, JCP 9, m-CO), 93.6 (s, Cp), 76.3 (s, OCH2), 41.7 (d, JCP
18, C1-Cy), 41.1 (d, JCP 18, C1-Cy), 33.6, 33.4 (2 ¥ s, C2-Cy), 27.5
(d, JCP 12, 2 ¥ C3-Cy), 26.4, 26.3 (2 ¥ s, C4-Cy), 15.1 (s, CH3).
Preparation of [Mo2Cp2(l-H)(l-PPh2)(CO)2] (3c). Solid
[NH4]PF6 (0.050 g, 0.30 mmol) was added to a THF solution
of Na-2c, prepared from [Mo2Cp2(CO)6] (0.050 g, 0.102 mmol) as
described above, and the mixture was stirred at 273 K for 10 min to
give a brown-greenish solution. Solvent was then removed under
vacuum, the residue was extracted with toluene (2 ¥ 10 ml) and the
extracts were then filtered. Removal of the solvent under vacuum
and washing of the residue with petroleum ether (5 ml) yielded
essentially pure compound 3c as a brown–green, quite air-sensitive
powder (0.048 g, 85% overall yield). All attempts to further purify
this material resulted in its progressive decomposition. dH 7.59–
7.31 (10H, m, Ph), 5.13 (10H, s, Cp), -6.85 (1H, d, JPH 8, m-H).
Preparation of [Mo2Cp2{l-COC(O)Ph}(l-PCy2)(l-CO)] (4g).
A THF solution of Li-2a, prepared from [Mo2Cp2(CO)6] (0.050 g,
0.102 mmol) as described above, was stirred with PhC(O)Cl (30 ml,
0.26 mmol) for 5 min to give a red-rose solution. Workup of
the mixture as described for 4e (method B) gave compound 4g
(0.033 g, 47%) as a red solid (Found: C, 54.4; H, 5.7. Calc. for
C31H37Mo2O3P: C, 54.7; H, 5.5%). dH 8.04 (2H, m, H2-Ph,), 7.62
(1H, tt, JHH 7, 1.3, H4-Ph), 7.49 (2H, false t, JHH + JHH 14, H3-Ph),
5.87 (s, Cp, 10H), 2.00–0.40 (m, Cy, 22H). dC 342.4 (d, JCP 17, m-
COR), 300.7 (d, JCP 9, m-CO), 160.1 [s, C(O)Ph], 133.7 (s, C4-Ph),
130.3 (s, C2-Ph), 130.0 (s, C1-Ph), 128.9 (s, C3-Ph), 95.6 (s, Cp),
41.6 (d, JCP 17, C1-Cy), 41.2 (d, JCP 18, C1-Cy), 33.6, 33.2 (2 ¥ s,
C2-Cy), 27.5 (d, JCP 13, C3-Cy), 27.4 (d, JCP 12, C3-Cy), 26.3 (d,
JCP 1.5, C4-Cy), 26.2 (d, JCP 1.5, C4-Cy).
Preparation of [Mo2Cp2(l-H){l-P(OEt)2}(CO)2] (3d). Solid
[NH4]PF6 (0.040 g, 0.245 mmol) was added to a THF solution
of Li-2d, prepared from [Mo2Cp2(CO)6] (0.100 g, 0.204 mmol) as
described above, and the mixture was stirred for 10 min to give
a brown solution. Solvent was then removed under vacuum, the
residue was extracted with toluene (2 ¥ 10 ml) and the extracts were
then filtered. Removal of the solvent under vacuum and washing
of the residue with petroleum ether (5 ml) yielded a powder
containing compound 3d as a major species. All attempts to further
purify this air-sensitive complex led to complete decomposition.
dH -5.26 (1H, d, JPH 11, m-H); all other resonances could not be
unequivocally assigned due to considerable decomposition of the
complex during acquisition of the NMR data.
Preparation of [Mo2Cp2(l-COiPr)(l-PCy2)(l-CO)] (4h).
A
THF solution of K-2a, prepared from [Mo2Cp2(CO)6] (0.050 g,
i
0.102 mmol) as described above, was stirred with PrBr (0.5 ml,
5.3 mmol) for 5 h to give a red-brown solution shown (by 31P NMR)
to be a mixture of the carbyne complex 4h (minor) along with
several unidentified species. Workup of the mixture as described
for 4e (method B) gave compound 4h as an orange solid (0.012 g,
19%). dH 5.70 (10H, s, Cp), 4.04 (1H, sept, JHH 6, OCH), 2.20–0.40
(22H, m, Cy), 1.22 (6H, d, JHH 6, Me).
Preparation of [Mo2Cp2(l-COMe)(l-PCy2)(l-CO)] (4e).
Method A:
A THF solution of Li-2a, prepared from
[Mo2Cp2(CO)6] (0.050 g, 0.102 mmol) as described above,
was stirred with solid [Me3O]BF4 (0.030 g, 0.202 mmol) for
10 min without noticeable changes in colour. Solvent was
then removed under vacuum, and the residue was extracted
with toluene and filtered. Removal of solvent from the filtrate
gave a residue which was then chromatographed on alumina
(activity IV) at 253 K. Elution with dichloromethane gave a
red-rose fraction which yielded, after removal of solvents under
vacuum, compound 4e as a red-rose air-sensitive solid (0.026 g,
45%). Method B: Neat Me2SO4 (0.2 ml, 2.1 mmol, excess), was
added slowly to a well stirred THF solution of Li-2a, prepared
from [Mo2Cp2(CO)6] (0.500 g, 1.02 mmol) as described above,
and the mixture was further stirred for 1 h. The solvent was
then removed under vacuum, the residue was extracted with
dichloromethane–petroleum ether (1 : 4) and the extracts were
chromatographed on alumina (activity IV) at 285 K. Elution with
dichloromethane–petroleum ether (2 : 1) gave a red fraction which
yielded, after removal of solvents under vacuum, compound 4e
(0.360 g, 60%) as a red-rose solid (Found: C, 50.5; H, 5.9. Calc.
for C25H35Mo2O2P: C, 50.9; H, 6.0%). dH 5.75 (10H, s, Cp), 3.74
(3H, s, OMe), 2.00–0.30 (22H, m, Cy). dC (100.63 MHz; 213 K)
352.0 (d, JCP 15, m-COMe), 305 (d, JCP 9, m-CO), 93.5 (s, Cp),
66.4 (s, OMe), 40.2 (d, JCP 18, C1-Cy), 39.5 (d, JCP 19, C1-Cy),
33.5, 33.2 (2 ¥ s, C2-Cy), 27.4 (d, JCP 11, 2 ¥ C3-Cy), 26.3, 26.1 (2
¥ s, C4-Cy).
Preparation of [Mo2Cp2(l-COCy)(l-PCy2)(l-CO)] (4i).
A
THF solution of K-2a, prepared from [Mo2Cp2(CO)6] (0.050 g,
0.102 mmol) as described above, was stirred with CyI (200 ml,
1.51 mmol) for 4 h to give a green solution shown (by 31P NMR)
to be a mixture of the iodo-complex 6k (major) and the carbyne
complex 4i (minor). Workup of the mixture as described for 4e
(method B) gave compound 4i (0.010 g, 15%) as a red solid. dH
5.73, 5.72 (2 ¥ 5H, 2 ¥ s, Cp), 3.61 (1H, m, OCy), 2.10–0.40 (32H,
m, Cy).
Preparation of solutions of [Mo2Cp2(l-COI)(l-PCy2)(l-CO)]
(5). A THF solution of Li-2a, prepared from 0.030 g of the
hydride complex 3a (0.052 mmol) as described above (method C),
was stirred with EtI (100 ml, 1.25 mmol) for 1.5 h to give a red
solution shown (by IR and 31P NMR) to contain the carbyne
complex 5 as major species, along with the iodo-complex 6k and a
small amount of the ethoxycarbyne complex 4f. Any manipulation
of this solution (even removal of the solvent under vacuum) caused
the complete transformation of the main product into the iodo-
complex 6k.
Preparation of [Mo2Cp2(l-Br)(l-PCy2)(CO)2] (6j). A THF
solution of Li-2a, prepared from 0.030 g of the hydride complex 3a
(0.052 mmol) as described above (method C), was stirred with Br2
(25 ml of a 0.347 g ml-1 solution in CH2Cl2, 0.05 mmol) for 1 min
to give a green solution. After removal of solvents under vacuum
Preparation of [Mo2Cp2(l-COEt)(l-PCy2)(l-CO)] (4f).
A
THF solution of K-2a, prepared from [Mo2Cp2(CO)6] (0.050 g,
0.102 mmol) as described above, was stirred with Et2SO4 (50 ml,
8180 | Dalton Trans., 2009, 8171–8182
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
the residue was extracted with toluene (2 ¥ 6 ml) and the extracts
filtered. Removal of the solvent from the filtrate gave compound
6j as a dark-green, very air-sensitive powder. dH (200.13 MHz,
CDCl3) 5.03 (10H, d, JHP 0.5, Cp), 2.50–1.10 (22H, m, Cy).
Organometallics, 2006, 25, 5374; (c) C. M. Alvarez, M. A. Alvarez,
M. E. Garc´ıa, A. Ramos, M. A. Ruiz, C. Graiff and A. Tiripicchio,
Organometallics, 2007, 26, 321; (d) M. A. Alvarez, M. E. Garc´ıa, A.
Ramos and M. A. Ruiz, Organometallics, 2007, 26, 1461; (e) M. A.
Alvarez, M. E. Garcıa, A. Ramos, M. A. Ruiz, M. Lanfranchi and
A. Tiripicchio, Organometallics, 2007, 26, 5454; (f) M. E. Garc´ıa, A.
Ramos, M. A. Ruiz, M. Lanfranchi and L. Marchio, Organometallics,
2007, 26, 6197.
´
Preparation of [Mo2Cp2(l-I)(l-PCy2)(CO)2] (6k). Solid I2
(0.040 g, 0.16 mmol) was added to a THF solution of K-2a,
prepared from [Mo2Cp2(CO)6] (0.025 g, 0.051 mmol) as described
above, and the mixture was stirred for 1 min to give a green
solution. Solvent was then removed under vacuum, and the residue
extracted with toluene and filtered. Removal of solvent from
the filtrate gave a residue which was then chromatographed on
alumina (activity IV) at 285 K. Elution with dichloromethane–
petroleum ether (1 : 3) gave a green fraction which yielded, after
removal of solvents under vacuum, compound 6k as a dark-green,
air-sensitive solid (0.028 g, 83%). dH 5.26 (10H, s, Cp), 2.50–1.00
(22H, m, Cy).
10 (a) M. E. Garc´ıa, D. Garc´ıa-Vivo´, M. A. Ruiz, S. Alvarez and G. Aullo´n,
Organometallics, 2007, 26, 4930; (b) M. E. Garc´ıa, D. Garc´ıa-Vivo´,
M. A. Ruiz, S. Alvarez and G. Aullo´n, Organometallics, 2007, 26, 5912;
(c) M. E. Garc´ıa, D. Garc´ıa-Vivo´ and M. A. Ruiz, Organometallics,
2008, 27, 169.
11 M. A. Alvarez, D. Garc´ıa-Vivo´, M. E. Garc´ıa, M. E. Mart´ınez,
A. Ramos and M. A. Ruiz, Organometallics, 2008, 27, 1973.
12 C. M. Alvarez, M. A. Alvarez, D. Garc´ıa-Vivo´, M. E. Garc´ıa, M. A.
Ruiz, D. Sa´ez, L. R. Falvello, T. Soler and P. Herson, Dalton Trans.,
2004, 4168.
13 P. S. Braterman, Metal Carbonyl Spectra, Academic Press, London,
UK, 1975.
14 M. E. Garc´ıa, V. Riera, M. A. Ruiz, M. T. Rueda and D. Sa´ez,
Organometallics, 2002, 21, 5515.
Preparation of cis-[Mo2Cp2(l-COMe)(l-PCy2)(O)(CO)] (7).
A solution of compound 4e (0.041 g, 0.08 mmol) in THF (10 ml)
was placed in a flask filled with air, and it was stirred for 24 h to
give a black mixture. The solvent was then removed under vacuum
and the residue was chromatographed on alumina (activity IV) at
285 K. Elution with dichloromethane–petroleum ether (4 : 1) gave
an orange fraction which yielded, after removal of solvents under
vacuum, compound 7 (0.030 g, 61%) as an orange solid (Found: C,
49.4; H, 6.0. Calc. for C25H35Mo2O3P: C, 49.5; H, 5.8%). dH 5.43,
5.40 (2 ¥ 5H, 2 ¥ s, Cp), 4.57 (3H, s, OMe), 2.60–0.80 (22H, m, Cy).
dC (CDCl3) 346.8 (d, JCP 7, m-COMe), 236.4 (d, JCP 10, MoCO),
98.6, 88.4 (2 ¥ s, Cp), 68.2 (s, OMe), 53.1 (d, JCP 13, C1-Cy), 41.8
(d, JCP 16, C1-Cy), 36.1, 34.9 (2 ¥ s, C2-Cy), 33.7 (d, JCP 6, C2-Cy),
33.4 (s, C2-Cy), 28.3–28.0 (4 ¥ d, C3-Cy), 26.4 (s, C4-Cy), 26.0 (d,
JCP 1.4, C4-Cy).
15 (a) X. Y. Liu, V. Riera, M. A. Ruiz, A. Tiripicchio and M. Tiripicchio-
Camellini, Organometallics, 1996, 15, 974; (b) M. A. Alvarez, C.
Alvarez, M. E. Garc´ıa, V. Riera, M. A. Ruiz and C. Bois, Organo-
metallics, 1997, 16, 2581; (c) C. M. Alvarez, B. Gala´n, M. E. Garc´ıa, V.
Riera, M. A. Ruiz and C. Bois, Organometallics, 2003, 22, 3039.
16 (a) M. Y. Darensbourg, Prog. Inorg. Chem., 1985, 33, 221; (b) M. Y.
Darensbourg and C. E. Ash, Adv. Organomet. Chem., 1987,
27, 1.
17 I. Amor, M. E. Garc´ıa, M. A. Ruiz, D. Sa´ez, H. Hamidov and J. C.
Jeffery, Organometallics, 2006, 25, 4857.
18 B. Neumu¨ller and W. Petz, Organometallics, 2001, 20, 163.
19 M. A. Alvarez, M. E. Garc´ıa, V. Riera and M. A. Ruiz, Organometallics,
1999, 18, 634.
20 H. Gu¨nther, NMR Spectroscopy, John Wiley, Chichester, UK, 1980,
p 70.
21 C. M. Alvarez, M. A. Alvarez, M. Alonso, M. E. Garc´ıa and M. A.
Ruiz, Inorg. Chem., 2006, 45, 9593.
22 G. Garc´ıa, M. E. Garc´ıa, S. Melo´n, V. Riera, M. A. Ruiz and
F. Villafan˜e, Organometallics, 1997, 16, 624.
23 A. J. Carty, S. A. MacLaughlin and D. Nucciarone, in Phosphorus-31
NMR Spectroscopy in Stereochemical Analysis, ed. J. G. Verkade and
L. D. Quin, VCH, Deerfield Beach, FL, 1987, ch. 16.
Acknowledgements
24 (a) N. J. Forrow and S. A. R. Knox, J. Chem. Soc., Chem. Commun.,
1984, 679; (b) B. S. Kang, U. Koelle and U. Thewalt, Organometallics,
1991, 10, 2569; (c) J. K. Hoyano and W. A. G. Graham, J. Am. Chem.
Soc., 1982, 104, 3722.
25 M. E. Garc´ıa, V. Riera, M. T. Rueda, M. A. Ruiz and S. Halut, J. Am.
Chem. Soc., 1999, 121, 1960.
We thank the MEC of Spain for a grant (to A. R.) and financial
support (projects BQU2003–05471 and CTQ2006-01207/BQU).
We also thank Dr. Daniel Garc´ıa-Vivo´ for the scaling-up in the
preparation of compound 4e (method B).
26 (a) R. J. Klinger, W. M. Butler and M. D. Curtis, J. Am. Chem. Soc.,
1978, 100, 5034; (b) T. Adatia, M. McPartlin, M. J. Mays, M. J. Morris
and P. R. Raithby, J. Chem. Soc., Dalton Trans., 1989, 1555.
27 F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, G. Orpen and
R. Taylor, J. Chem. Soc., Perkin Trans. 2, 1987, S1.
References
1 (a) Reviews: F. A. Cotton and G. Wilkinson, Advanced Inorganic
Chemistry, Wiley, New York, 5th edn, 1988, ch. 22; (b) J. E. Ellis,
Adv. Organomet. Chem., 1990, 31, 1, and references therein.
2 J. G. M. Van Der Linden, J. Heck, B. Walter and H. C. Bottcher, Inorg.
Chim. Acta, 1995, 217, 29.
3 J. T. Lin, G. P. Hagen and J. E. Ellis, J. Am. Chem. Soc., 1983, 105,
2296, and references therein.
4 K. Henrick, J. A. Iggo, M. J. Mays and P. R. Raithby, J. Chem. Soc.,
Chem. Commun., 1984, 209.
5 D. Seyferth, K. S. Brewer, T. G. Wood, M. Cowie and R. W. Hilts,
Organometallics, 1992, 11, 2570.
6 (a) X. Y. Liu, V. Riera, M. A. Ruiz, M. Lanfranchi and A. Tiripicchio,
Organometallics, 2003, 22, 4500; (b) X. Y. Liu, V. Riera, M. A. Ruiz
and C. Bois, Organometallics, 2001, 20, 3007; (c) X. Y. Liu, V. Riera
and M. A. Ruiz, Organometallics, 1994, 13, 2925.
7 M. E. Garc´ıa, S. Melo´n, A. Ramos, V. Riera, M. A. Ruiz, D. Belletti,
C. Graiff and A. Tiripicchio, Organometallics, 2003, 22, 1983.
8 (a) A. P. Ginsberg and M. J. Hawkes, J. Am. Chem. Soc., 1968, 90, 5930;
(b) M. J. Hawkes and A. P. Ginsberg, Inorg. Chem., 1969, 8, 2189.
9 (a) C. M. Alvarez, M. A. Alvarez, M. E. Garc´ıa, A. Ramos, M. A.
Ruiz, M. Lanfranchi and A. Tiripicchio, Organometallics, 2005, 24,
7; (b) M. A. Alvarez, M. E. Garc´ıa, A. Ramos and M. A. Ruiz,
28 (a) W. H. Hersh and R. H. Fong, Organometallics, 2005, 24, 4179;
(b) J. C. Peters, A. L. Odon and C. C. Cummins, Chem. Commun., 1997,
1995; (c) B. S. Bronk, J. D. Protasiewicz, L. E. Pence and S. J. Lippard,
Organometallics, 1995, 14, 2177; (d) D. Seyferth, D. P. Ruschke and
W. H. Davis, Organometallics, 1994, 13, 4695; (e) Y. Chi, S. H. Chuang,
B. F. Chen, S. M. Peng and G. H. Lee, J. Chem. Soc., Dalton Trans.,
1990, 3033; (f) A. E. Friedman and P. C. Ford, J. Am. Chem. Soc.,
1989, 111, 551; (g) J. B. Keister, Polyhedron, 1988, 7, 847; (h) A. E.
Friedman and P. C. Ford, J. Am. Chem. Soc., 1986, 108, 7851; (i) L. J.
Farrugia, A. D. Miles and F. G. A. Stone, J. Chem. Soc., Dalton Trans.,
1985, 2437; (j) L. R. Beanan and J. B. Keister, Organometallics, 1985,
4, 1713; (k) D. Nuel, F. Daha and R. Mathieu, J. Am. Chem. Soc.,
1985, 107, 1658; (l) J. R. Shapley, W. Y. Yeh, M. R. Churchill and Y. J.
Li, Organometallics, 1985, 4, 1898; (m) M. Green, K. A. Mead, R. M.
Mills, I. D. Salter, F. G. A. Stone and P. J. Woodward, J. Chem. Soc.,
Chem. Commun., 1982, 51.
29 (a) L. M. Bavaro and J. B. Keister, J. Organomet. Chem., 1985, 287, 357;
(b) J. B. Keister, M. W. Payne and M. J. Muscatella, Organometallics,
1983, 2, 219; (c) B. F. G. Johnson, J. Lewis, A. G. Orpen, P. R. Raithby
and G. Su¨ss, J. Organomet. Chem., 1979, 173, 187; (d) J. B. Keister,
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 8171–8182 | 8181
©

View Article Online
J. Chem. Soc., Chem. Commun., 1979, 214; (e) P. D. Gavens and M. J.
Mays, J. Organomet. Chem., 1978, 162, 389.
30 N. N. Greenwood and A. Earnshaw, Chemistry of the Elements,
Butterworth-Heinemann, Oxford, UK, 2nd edn, 1997, ch. 17,
p 806.
31 K. Endrich, R. Korswagen, T. Zahn and M. L. Ziegler, Angew. Chem.,
Int. Ed. Engl., 1982, 21, 919.
32 R. B. King, Acc. Chem. Res., 1970, 3, 417.
33 (a) Paramagnetic Organometallic Species in Activation Selectivity,
Catalysis, ed. M. Chanon, M. Julliard and J. C. Poite, Kluwer Academic
Publishers, Dordrecht, 1989; (b) Organometallic Radical Processes, ed.
W. C. Trogler, Elsevier, Amsterdam, 1990; (c) D. Astruc, Electron
Transfer and Radical Processes in Transition-Metal Chemistry, VCH,
New York, 1995.
34 M. E. Garc´ıa, V. Riera, M. T. Rueda, A. Tiripicchio and M. Lanfranchi,
J. Am. Chem. Soc., 1999, 121, 4060.
35 M. A. Alvarez, Y. Anaya, M. E. Garc´ıa and M. A. Ruiz,
Organometallics, 2004, 23, 3950.
36 (a) M. A. Alvarez, Y. Anaya, M. E. Garc´ıa, V. Riera, M. A. Ruiz and
J. Vaissermann, Organometallics, 2003, 22, 456; (b) M. A. Alvarez, Y.
Anaya, M. E. Garc´ıa, M. A. Ruiz and J. Vaissermann, Organometallics,
2005, 24, 2452.
37 W. L. F. Armarego and C. Chai, Purification of Laboratory Chemicals,
Butterworth-Heinemann, Oxford, UK, 5th edn, 2003.
8182 | Dalton Trans., 2009, 8171–8182
This journal is
The Royal Society of Chemistry 2009
©