Reactions of the unsaturated ditungsten anion [W2Cp2(μ-PCy2)(μ-CO)2]- with C- and P-based electrophiles

Article abstract of DOI:10.1021/om501202j

The Na⁺ salt of the title anion reacted with MeI to give a mixture of isomeric methoxycarbyne [W₂Cp₂(μ-COMe)(μ-PCy₂)(μ-CO)] (major) and methyl-bridged [W₂Cp₂(μ-CH₃)(μ-PCy₂

Full text of DOI:10.1021/om501202j

This is an open access article published under an ACS AuthorChoice License, which permits
copying and redistribution of the article or any adaptations for non-commercial purposes.
Article
pubs.acs.org/Organometallics
Reactions of the Unsaturated Ditungsten Anion [W2Cp2(μ-PCy2)(μ-
CO)2]− with C- and P‑Based Electrophiles
M. Angeles Alvarez, M. Esther García, Daniel García-Vivo,* Miguel A. Ruiz,* and M. Fernanda Vega
́
́ ́
Departamento de Química Organica e Inorganica/IUQOEM, Universidad de Oviedo, E-33071 Oviedo, Spain
S
* Supporting Information
ABSTRACT: The Na+ salt of the title anion reacted with MeI
to give a mixture of isomeric methoxycarbyne [W2Cp2(μ-
COMe)(μ-PCy2)(μ-CO)] (major) and methyl-bridged
[W2Cp2(μ-CH3)(μ-PCy2)(CO)2] (minor) derivatives, follow-
ing from competitive methylation at the O atoms or at the
ditungsten center, respectively. In contrast, its reaction with
ClCH2Ph gave exclusively the benzyl-bridged complex
[W2Cp2(μ-CH2Ph)(μ-PCy2)(CO)2], which in solution dis-
plays a medium-strength agostic W−H−C interaction, as
suggested by the reduced C−H coupling of 90 Hz for the
atoms involved. This complex could be dehydrogenated
photochemically to give the 30-electron benzylidyne derivative [W2Cp2(μ-CPh)(μ-PCy2)(μ-CO)] in a selective way. The
t
title anion reacted rapidly with several chlorophosphines ClPR2 (R = Bu, Et, Cy) to give also two types of isomers: the
phosphinoxycarbyne complexes [W2Cp2(μ-COPR2)(μ-PCy2)(μ-CO)] and the mixed-phosphide derivatives [W2Cp2(μ-PCy2)(μ-
t
PR2)(CO)2], with the former being obtained selectively when R was the bulky Bu group, whereas the latter was the major
product for the smaller Et group. The phosphinoxycarbyne complexes were quite unstable species that could not be isolated as
pure materials, but when R = Et, Cy, they underwent thermal rearrangement to give the corresponding mixed-phosphide isomers,
among other processes. In contrast, the reaction with ClP(O)(OPh)2 gave a more stable phosphatecarbyne complex, [W2Cp2{μ-
COP(O)(OPh)2}(μ-PCy2)(μ-CO)], which could be isolated and fully characterized (W−W = 2.5034(4) Å). The title anion also
reacted with P4 via its ditungsten center to give the Na+ salt of the diphosphorus-bridged anionic complex [W2Cp2(μ-κ2:κ2-P2)(μ-
PCy2)(CO)2]−, following from a symmetrical cleavage of the P4 molecule. The latter anion reacted rapidly with MeI to give the
new methyldiphosphenyl-bridged complex [W2Cp2(μ-κ2:κ2-P2Me)(μ-PCy2)(CO)2], which could be isolated in good yield.
of unsaturated derivatives such as the hydride [Mo2Cp2(μ-
H)(μ-PCy2)(CO)2], the methoxycarbyne [Mo2Cp2(μ-COMe)-
(μ-PCy2)(μ-CO)], or the agostic alkyls [Mo2Cp2(μ-CH2R)(μ-
PCy2)(CO)2] (R = H, Ph), all of them displaying a wide
reactivity on their own, as shown by independent studies.5−7
Encouraged by these remarkable results and to test the
influence of the metal (W instead of Mo) on all this chemistry,
more recently we developed an efficient synthetic route for the
Na+ salt of the related ditungsten anion [W2Cp2(μ-PCy2)(μ-
CO)2]− (1) (Chart 1).8 Indeed our initial exploration on the
reactivity of this anion revealed significant differences when
compared with its dimolybdenum analogue. Notably, the
INTRODUCTION
■
Binuclear cyclopentadienyl complexes having metal−metal
triple bonds are a well-established family of compounds within
the organometallic field. The coordinative unsaturation
inherent to these systems provides them with a high reactivity
toward an immense variety of molecules and, therefore, with a
wide synthetic potential.1 When combined with a neat negative
charge, these species exhibit enhanced nucleophilicity, which
then can be exploited to create new molecular complexity via
direct reaction with different electrophiles, this rendering
products having new M−E bonds (E = H, C, p- or d-block
element, etc.). However, the number of systems combining a
metal−metal triple bond and a negative charge is quite limited
so far; in fact, we can only quote the molybdenum
pentachloride [Mo2Cp2(μ-Cl)Cl4]− reported by Rheingold
and co-workers2 and the organophosphide-bridged anions
[Mo2Cp2(μ-PR2)(μ-CO)2]− (R = Cy, Ph, OEt) prepared
some time ago in our lab.3 Among all these anions, only the
PCy2-bridged one has been the subject of a systematic study of
reactivity, which revealed it as a highly active and versatile
intermediate with two distinct nucleophilic sites located at the
metal and oxygen atoms.3,4 Moreover this anion turned out to
be a key synthetic intermediate for the preparation of a variety
Chart 1
Received: November 28, 2014
Published: February 19, 2015
© 2015 American Chemical Society
870
DOI: 10.1021/om501202j
Organometallics 2015, 34, 870−878

Organometallics
Article
ditungsten anion displayed an enhanced nucleophilicity at the
O atoms of the bridging carbonyls, this enabling it to react at
this site with simple C- and P-based electrophiles such as MeI
or ClPtBu2 to give new methoxycarbyne- (μ-COMe) or
phosphinoxycarbyne-bridged (μ-COPtBu2) complexes,
although the latter compound was rather unstable.8 In contrast,
similar reactions of the Mo2 analogue of 1 had been shown
previously to take place at the dimetal site, to lead instead to
methyl-bridged [Mo2Cp2(μ-CH3)(μ-PCy2)(CO)2] or mixed-
phosphide [Mo2Cp2(μ-PCy2)(μ-PR2)(CO)2] derivatives, re-
spectively, whereas the formation of the methoxycarbyne
complex [Mo2Cp2(μ-COMe)(μ-PCy2)(μ-CO)] required the
use of harder electrophilic reagents such as [Me3O][BF4] or
Me2SO4.3,4 A further point of interest in the above results is the
fact that there seems to be no precedents in the literature for
phosphinoxycarbyne (C−O−PR2) complexes, thus opening an
opportunity to explore the chemical behavior of this unusual
ligand. Actually, only relatively few carbyne complexes having a
P-containing substituent have been reported at all, with all of
them displaying C−P bonds.9−11 Building on these results then
it was of interest to expand our initial studies on the reactivity
of 1 by including different C- and P-based electrophiles, which
is the main topic of this paper, while the behavior of 1 toward
different H+ sources and metal-based electrophiles has been
analyzed in a recent paper.12 Herein we report full details on
the reactions of the Na+ salt of anion 1 (1-Na) with some
haloalkanes such as MeI and ClCH2Ph, which now have
allowed us to fully characterize the corresponding alkyl-bridged
derivatives and to explore their photochemical dehydrogenation
to render new carbyne-bridged complexes. We have also
expanded significantly the pool of P electrophiles under
examination, this now including different chlorophosphines,
chlorophosphates, and also white phosphorus (P4). As it will be
shown below, the above organophosphorus reagents added
preferentially at the O-sites of the anion, but only the P(V)
derivative had enough stability, this enabling us to characterize
crystallographically for the first time a complex displaying a
phosphate carbyne [C−O−P(O)(OR)2] ligand. In contrast, the
reaction with P4 leads to an anionic diphosphorus-bridged
complex following from the formal addition of a P2 unit to the
dimetal center of the anion 1 and displaying P-based
nucleophilicity.
Chart 2
the corresponding methyl-bridged product, as noted above. As
expected, the use of stronger methylating agents such as
Me2SO4 or MeSO3CF3 further favors methylation at the O-site
of the anion 1, to the point that it is then difficult to avoid a
second methylation that yields small amounts of the bis-
(alkoxycarbyne) cation [W2Cp2(μ-COMe)2(μ-PCy2)]+, a spe-
cies that we have not further pursued, but likely is analogous to
the dimolybdenum complex [Mo2Cp2(μ-COMe)2(μ-PCy2)]+
previously characterized by us.6a In contrast, the reaction of 1-
Na with ClCH2Ph matches that of its Mo2 analogue,3,5b
because in this case the unique product formed is the benzyl-
bridged dicarbonyl complex [W2Cp2(μ-CH2Ph)(μ-PCy2)-
(CO)2] (4), which in solution displays a medium-strength
agostic interaction of the benzyl group (Chart 2).
We have shown previously that, upon photolysis, the Mo2
analogue of 4 can be conveniently transformed into the 30-
electron benzylidyne derivative [Mo2Cp2(μ-CPh)(μ-PCy2)(μ-
CO)], a molecule itself displaying a remarkable chemical
behavior.13 Then, it was of interest to check whether a similar
process could be induced for the ditungsten compound.
Indeed, irradiation of toluene solutions of 4 with visible−UV
light promotes fast decarbonylation and dehydrogenation of the
alkyl ligand, to give in good yields the corresponding
benzylidyne derivative [W2Cp2(μ-CPh)(μ-PCy2)(μ-CO)] (5),
with no intermediates being detected along this (necessarily)
multistep process.
Solution Structure of Agostic Alkyl Complexes 3 and
4. The spectroscopic data available for complex 4 (Table 1 and
Experimental Section) reveal a strong relationship with its
dimolybdenum analogue [Mo2Cp2(μ-CH2Ph)(μ-PCy2)-
(CO)2], a species previously characterized by us through
spectroscopic, DFT, and X-ray diffraction methods.5b There-
fore, a similar geometry is assumed for this molecule, with the
bridging benzyl ligand being σ-bound to one tungsten atom and
displaying an α-agostic interaction with the second metal atom.
A first indication in this direction comes from the fact that 4
displays a relatively low 31P chemical shift (73.3 ppm), which
falls within the region expected for 32-electron W2 complexes
(cf. 89.3 ppm for [W2Cp2(μ-I)(μ-PCy2)(CO)2]) and far away
from the significantly higher chemical shifts typically observed
for 30-electron complexes (cf. 152.7 ppm for 1 or 167.2 ppm
for the hydride-bridged isomer [W2Cp2(μ-H)(μ-PCy2)-
(CO)2]).8 This suggests the presence of an agostic, three-
electron-like coordination of the bridging benzyl group akin to
that of the Mo2 complex, which is more directly corroborated
RESULTS AND DISCUSSION
■
Reactions of Complex 1 with Haloalkanes. Freshly
prepared suspensions of the salt 1-Na in THF react slowly over
the course of 4 days with excess MeI to give the
methoxycarbyne complex [W2Cp2(μ-COMe)(μ-PCy2)(μ-
CO)] (2) as the major product, along with small amounts of
the methyl-bridged isomer [W2Cp2(μ-CH3)(μ-PCy2)(CO)2]
(3) (Chart 2). The latter product could not be isolated from
these mixtures, but it was identified on the basis of
spectroscopic analogies with the related dimolybdenum
complex.3,5b We noticed that the above reaction could be
accelerated significantly (complete conversion in ca. 5 min)
upon addition of a few drops of water to the reaction solvent,
an action that increases dramatically the solubility of the Na+
salt of 1 and likely also modifies the solvation of the Na+ cation
as well as the anion−cation interactions, but neither the
products obtained nor their relative ratio was modified
significantly under these conditions. In any case, we note that
these results are in stark contrast with the behavior of the Mo2
analogue of 1, which reacts with MeI to give almost exclusively
through the H and 13C NMR data. First we note that two
1
distinct resonances are observed for the methylenic protons in
1
the H NMR spectra, a doublet at 2.37 (2JHH = 15 Hz) and a
doublet of doublets at −2.59 ppm (2JHH = 15 Hz, JHP = 1.7),
3
871
DOI: 10.1021/om501202j
Organometallics 2015, 34, 870−878

Organometallics
Article
a
b
Table 1. Selected IR and 31P{1H} NMR Data for New
Compounds
Scheme 1. Fluxional Process Proposed for Compound 4 in
Solution (P = PCy2)
compound
ν(CO)
δP (JPW) [JPP
]
c
c
Na[W2Cp2(μ-PCy2)(μ-CO)2] 1546 (s)
(1-Na)
152.7 (378)
[W2Cp2(μ-COMe)(μ-PCy2)
(μ-CO)] (2)
1625 (s)
172.5 (358)
d
[W2Cp2(μ-CH3)(μ-PCy2)
(CO)2] (3)
80.4
e
[W2Cp2(μ-CH2Ph)(μ-PCy2)
(CO)2] (4)
1858 (w, sh), 1819 (s)
73.3 (300)
on lowering the temperature, so that the nonagostic resonance
shifts by ca. +0.4 ppm while the agostic one shifts by ca. −0.6
ppm when going from 298 to 213 K. Similar thermal shifts of
the methylenic resonances were observed for the Mo2 analogue
of 45b and for the heterometallic complex [ReW{μ-CH2(p-
tol)}(μ-CO)(CO)6(μ-Ph2PCH2PPh2)].14 We have explained
these changes in the dimolybdenum complex by assuming the
presence in solution of small amounts of a nonagostic benzyl-
bridged structure coexisting in solution with the major, agostic
isomer, with both species being in fast equilibrium on the NMR
time scale at all temperatures studied, a proposal also supported
by DFT calculations.5b Presumably, a similar situation holds for
the ditungsten complex 4. In any case, the spectroscopic
features of the methylenic 13C NMR resonance of the benzyl
group in 4 (δ −8.6 ppm, displaying one normal (130 Hz) and
one reduced C−H coupling (90 Hz)) are in full agreement with
the agostic coordination of this bridging group via one of the
methylenic hydrogens.5b We note that the value of coupling
involving the agostic hydrogen falls within the range typically
observed for agostic interactions (70−100 Hz)15 and is
significantly smaller than the figure measured for the analogous
Mo2 complex (100 Hz), this suggesting a stronger agostic C−
H−M interaction for the ditungsten compound 4.
e
[W2Cp2(μ-CPh)(μ-PCy2)(μ- 1637 (s)
CO)] (5)
172.8 (352)
f
f
f
g
[W2Cp2(μ-COPtBu2)(μ-
PCy2)(μ-CO)] (6a)
1660 (s)
165.7 (362)
g
135.4
f
[W2Cp2(μ-COPCy2)(μ-
PCy2)(μ-CO)] (6b)
1659 (s)
166.1 (361)
f
152.4 (14)
[W2Cp2(μ-COPEt2)(μ-PCy2) 1660 (s)
(μ-CO)] (6c)
[W2Cp2(μ-PCy2)2(CO)2]
(7b)
31.8 (278)
e
[W2Cp2(μ-PCy2)(μ-PEt2)
(CO)2] (7c)
1857 (w, sh), 1822 (vs) 40.0 (277) [4]
e
10.4 (288) [4]
f
[W2Cp2{μ-COP(O)(OPh)2} 1670 (s)
(μ-PCy2)(μ-CO)] (8)
173.2 (351)
−22.6
Na[W2Cp2(μ-κ2:κ2-P2)(μ-
PCy2)(CO)2] (9-Na)
[W2Cp2(μ-κ2:κ2-P2Me)(μ-
PCy2)(CO)2] (10)
1852 (s), 1823 (vs),
70.7 (296) [16,
f
fh
16] ,
1775 (w), 1695 (m)
f
1860 (s), 1837 (vs)
59.1 (259, 238)
[20, 6]
−166.0 (255,
235) [467, 20]
−390.1 (25)
[467, 6]
a
Recorded in dichloromethane solution, unless otherwise stated, with
b
C−O stretching bands [ν(CO)] in cm−1. Recorded in CD2Cl2 at
The available spectroscopic data for the methyl complex 3
are rather limited, because it could not be separated from the
corresponding mixtures, as noted above. Yet, these data (Table
1 and Experimental Section) are informative enough to trace a
clear structural analogy with the Mo2 complex [Mo2Cp2(μ-
CH3)(μ-PCy2)(CO)2], a molecule previously shown to display
a weak C−H−Mo agostic interaction both in solution and in
the solid state.5 Of particular relevance in this respect is the
121.50 MHz and 295 K, with 31P−183W (JPW) and 31P−31P [JPP]
c
coupling constants in Hz, unless otherwise stated. Data taken from ref
d
8, with IR recorded in THF and 31P NMR in acetone-d6. The P−W
coupling constant could not be measured due to the broadness of the
e
f
signal. Recorded at 162.01 MHz. Recorded in THF solution.
Recorded in toluene-d8 solution at 162.01 MHz. The resonance of
the P2 ligand could not be identified in this spectrum (see text).
g
h
1
with the latter being safely assigned to the hydrogen atom
directly involved in the agostic C−H···W interaction.5b
However, other spectroscopic data are incompatible with a
static asymmetric coordination of the benzyl group and
therefore are suggestive of the occurrence of dynamic processes
in solution. For instance, the 31P resonance of 4 exhibits just
one set of 183W satellites (300 Hz), while the room-temperature
1H and 13C NMR spectra display single resonances for pairs of
Cp and CO ligands, which should be inequivalent in a static
agostic structure. However, on lowering the temperature, two
appearance of a poorly shielded H NMR signal at −1.18 ppm
corresponding to the bridging methyl group in 3, to be
compared with a chemical shift of −0.77 ppm for its Mo2
analogue. In addition, compound 3 displays a single resonance
for the strictly inequivalent Cp ligands, then pointing to the
operation of a fast fluxional process (not studied) exchanging
the environments of the hydrogen atoms within the Me group,
otherwise similar to that studied in detail previously for the
Mo2 complex.5b Finally, the 31P chemical shift of compound 3
(80.4 ppm) is quite similar to that of 4, also in support of the
presence of an agostic (three-electron-like) coordination of the
methyl ligand to the ditungsten center of this complex.
1
spectroscopic modifications were observed in the H NMR
spectra: First, the Cp resonance broadened and then split into
two well-resolved signals of equal intensity at 253 K, now in
agreement with the static asymmetric structure proposed for
this compound. To account for this observation, we propose
the operation of a fluxional process analogous to that described
for the molybdenum complex,5b involving a 180° rotation of
the benzyl group around the C−M vector (Scheme 1). Such a
process generates an apparent C2 axis relating the pairs of Cp,
CO, and cyclohexyl groups, but not the methylenic protons,
which remain inequivalent.
Solution Structure of the Carbyne Complexes 2 and
5. Spectroscopic data in solution for compounds 2 and 5 are
comparable to each other, indicating that both molecules share
the same general structural features, in turn consistent with the
solid-state structures of related dimolybdenum complexes
previously characterized by us: the ethoxycarbyne complex
[Mo2Cp2(μ-COEt)(μ-PCy2)(μ-CO)]3b and the benzylidyne
complex [Mo2Cp2(μ-CPh)(μ-PCy2)(μ-CO)].13 Therefore, a
detailed spectroscopic discussion is not needed. We just note
that the presence of a bridging CO ligand in these compounds
is clearly denoted by the appearance of the characteristic low-
The second spectroscopic modification involves a progressive
shift of the methylenic benzyl resonances away from each other
872
DOI: 10.1021/om501202j
Organometallics 2015, 34, 870−878

Organometallics
Article
frequency band in their IR spectra [1625 (2) and 1637 (5)
cm−1] and by a quite deshielded resonance in the 13C spectra
(ca. 300 ppm). The observation of a more energetic C−O
stretch in the benzylidyne complex might seem counterintuitive
if we recall the lower electronegativity of a Ph group, compared
with a OMe one. We have previously observed the same trend
for a number of pairs of alkoxycarbyne/benzylidyne dimolyb-
denum complexes of formula [Mo2Cp2(μ-CX)(μ-PCy2)(CO)n]
(X = Ph, OMe; n = 1, 2), which can be attributed to the
presence of a significant π(C−O) bonding interaction in the
alkoxycarbyne complexes (not possible for a benzylidyne
group), which increases, comparatively, the electron density
at the metal site (Figure 1), an effect also substantiated by DFT
Less sterically demanding chlorophosphines such as ClPCy2
and ClPEt2 yielded even less stable carbyne complexes.
Reaction of salt 1-Na with ClPCy2 in tetrahydrofuran proceeds
quickly and quite selectively at 253 K to give the
phosphinoxycarbyne derivative [W2Cp2(μ-COPCy2)(μ-PCy2)-
(μ-CO)] (6b) as major product, as judged from the
corresponding IR and 31P NMR spectra, to be discussed later
(see the SI). Interestingly, a small amount of another binuclear
product was present in the initial mixture, characterized by a
relatively shielded 31P NMR resonance at 31.8 ppm with
medium P−W coupling (1JPW = 278 Hz), it being tentatively
identified as the bis(phosphide) isomer [W2Cp2(μ-
PCy2)2(CO)2] (7b) by analogy with the major product in
the ClPEt2 reaction (see below). The carbyne complex 6b,
however, is rather unstable, and upon 1 h stirring in
tetrahydrofuran solution at room temperature it completely
disappears to leave an increased amount of isomer 7b and the
hydride [W2Cp2(H)(μ-PCy2)(CO)2], along with smaller
amounts of other uncharacterized species. This suggests that
6b undergoes, inter alia, two competitive processes: (a)
hydrolytic reaction with trace amounts of water present in
the solvent, to give the ditungsten hydride and PH(O)Cy2 (also
the hydrolysis product of excess ClPCy2), and (b) thermal
rearrangement to give the bis(phosphide) dicarbonyl isomer 7b
(Scheme 2). Independent experiments supported this view:
Figure 1. Canonical forms in alkoxycarbyne-bridged binuclear
complexes.
calculations.6b,9b On the other hand, both 2 and 5 display
highly deshielded 31P NMR resonances (δP ca. 172 ppm)
retaining large W−P couplings (ca. 350 Hz), with values not far
from those of anion 1 (Table 1). These shifts are in agreement
with the presence of bridging PCy2 ligands in 30-electron W2
complexes, while the large W−P couplings are indicative of
metal atoms in a low coordination environment.16 Finally, the
presence of bridging carbyne ligands in both compounds is
firmly established by the appearance of strongly deshielded
resonances in their 13C{1H} NMR spectra, with chemical shifts
(342.9 and 363.9 ppm for 2 and 5, respectively) comparable to
those of their Mo2 analogues.
Scheme 2
Reactions of 1 with Organophosphorus Electrophiles.
As mentioned in the Introduction section, we found in a
preliminary study that reaction of the Na+ salt of anion 1 with
ClPtBu2 led to the formation of an unprecedented phosphinox-
ycarbyne complex, [W2Cp2(μ-COPtBu2)(μ-PCy2)(μ-CO)]
(6a) (Chart 3), following from addition of the P-based
Chart 3
first, addition of a drop of water to a freshly prepared solution
of 6b at 253 K caused its immediate decomposition to give
increased amounts of the ditungsten hydride. Second, the
reaction of 1 with ClPCy2 at 343 K yielded a complex mixture
containing 7b as the major product. Unfortunately, we were
unable to isolate the latter product for full spectroscopic
characterization. Noticeably, compound 7b cannot be prepared
through the more conventional synthetic routes previously
developed by us for bis(phosphide) complexes of the type
[M2Cp2(μ-PR2)(μ-PR′R″)(CO)2].17 We should finally note
that our data do not exclude that the phosphide complex 7b
might be also formed to some extent via direct reaction at the
dimetal site of the anion 1 (Scheme 2).
The transformation 6b/7b is worth noting after considering
that the analogous transformation of the related methox-
ycarbyne complex into its methyl-bridged isomer (i.e., a
hypothetical isomerization 2/3) does not take place even in
refluxing toluene or upon photochemical activation. It is
sensible to propose that such a rearrangement is facilitated by
the presence of the lone electron pair at phosphorus in the
phosphinoxycarbyne complexes (a feature absent in the related
alkoxycarbyne complexes), this enabling the formation of less
electrophile at one of the O atoms of the bridging carbonyls,
thus resembling the formation of the methoxycarbyne complex
2 in the reaction of 1-Na with MeI. However, although 6a is
formed selectively in the above reaction (see the Supporting
Information (SI)) and we could complete its spectroscopic
characterization, we have been unable to isolate this air-
sensitive compound as a pure material because of its
progressive decomposition upon manipulation.
873
DOI: 10.1021/om501202j
Organometallics 2015, 34, 870−878

Organometallics
Article
unsaturated intermediates A prior to the cleavage of the O−P
bond, eventually leading to the bis(phosphide) complexes 7
(Scheme 3). Expectedly, such an intermediate would be more
Scheme 3
Figure 2. ORTEP diagram (30% probability) of compound 8 with H
atoms and Cy groups (except their C1 atoms) omitted for clarity.
Table 2. Selected Bond Lengths (Å) and Angles (deg) for
Compound 8
t
disfavored when R = Bu on steric grounds. Conversely,
intermediate A would be more easily formed for smaller
substituents. Indeed, reaction of 1-Na with ClPEt2 takes place
rapidly at 253 K to yield the mixed-phosphide derivative
[W2Cp2(μ-PEt2)(μ-PCy2)(CO)2] (7c) as major product, along
with smaller amounts of the hydride [W2Cp2(H)(μ-PCy2)-
(CO)2]. IR monitoring of this reaction revealed the presence of
significant amounts of the corresponding phosphinoxycarbyne
complex [W2Cp2(μ-COPEt2)(μ-PCy2)(μ-CO)] (6c) in the
initial moments (see the SI), but it rapidly disappeared upon
further stirring of the solution even at 253 K, so we could not
even identify the corresponding 31P NMR resonances. These
results are consistent with a fast thermal rearrangement 6c/7c,
while the appearance of the ditungsten hydride would follow
from a competitive hydrolytic process, as observed for the PCy2
analogue.
In order to prepare more stable carbyne complexes, we
decided to test if a P(V)-electrophile might be more resistant to
the undesired hydrolytic reactions observed for complexes 6,
while the thermal rearrangement leaving the P-electrophile at
the dimetal site would now be inhibited by the absence of a
lone electron pair at phosphorus, here replaced by a poorly
coordinating PO function. Indeed, reaction of 1-Na with
ClP(O)(OPh)2 led almost instantaneously and selectively to
the phosphatecarbyne complex [W2Cp2{μ-COP(O)(OPh)2}-
(μ-PCy2)(μ-CO)] (8) (Chart 2), a species with conventional
stability that could be isolated and fully characterized both in
solution and in the solid state, as next discussed. We note,
however, that no phosphatecarbyne complex appears to have
been reported previously at all.
Structural Characterization of Compounds 6, 7, and 8.
The structure of 8 in the crystal lattice (Figure 2 and Table 2)
resembles that of the isoelectronic carbyne complexes
[Mo2Cp2(μ-COEt)(μ-PCy2)(μ-CO)],3b [Mo2Cp2(μ-CPh)(μ-
PCy2)(μ-CO)],13 and [W2Cp2(μ-COMe)(μ-PPh2)2]+;18 there-
fore a detailed analysis is not required. The molecule is built
from two WCp fragments bridged symmetrically by three
ligands: PCy2, CO, and a phosphatecarbyne group, [C−O−
P(O)(OPh)2]. The short intermetallic separation of 2.5034(4)
Å is consistent with the formulation of a metal−metal triple
bond in this 30-electron complex and is comparable to those
measured for the isoelectronic carbyne complexes mentioned
above (in the range 2.46−2.53 Å) or to that in the
bis(phosphide) complex [Mo2Cp2(μ-PPh2)2(μ-CO)]
W1−W2
W1−P1
W2−P1
W1−C1
W2−C1
W1−C2
W2−C2
2.5034(4)
2.398(2)
2.397(2)
2.104(7)
2.086(8)
1.975(6)
1.983(6)
P2−O2
P2−O3
P2−O4
P2−O5
C1−O1
C2−O2
C3−O4
C9−O5
1.580(4)
1.447(4)
1.596(4)
1.572(4)
1.199(7)
1.414(7)
1.402(7)
1.422(7)
101.9(2)
89.6(2)
C2−O2−P2
O2−P2−O3
O2−P2−O4
122.7(4)
117.7(3)
100.9(2)
O2−P2−O5
P1−W1−C1
P1−W1−C2
87.4(2)
[2.515(2) Å].19 The W−C distances follow the expected
trend, with those involving the carbyne ligand being ca. 0.12 Å
shorter than those involving the carbonyl group, in agreement
with the respective formal bond orders (1.5 vs 1). Noticeably,
the C2−O2 bond length of 1.414(7) Å within the
phosphatecarbyne ligand is somewhat longer than typical
values measured for the corresponding bond in different
alkoxycarbyne complexes (cf. 1.332(5) Å for the mentioned
ethoxycarbyne complex) and actually is comparable to the
C(sp2)−O single bonds of the OPh groups (1.402(7) and
1.422(7) Å). Therefore, it can be concluded that, in contrast to
alkoxycarbyne ligands, the degree of π-bonding interaction in
the C−O bond of the phosphatecarbyne ligand is negligible,
perhaps due to the strong electron-withdrawing power of the
P(V) group.
Spectroscopic data for compounds 6a−c and 8 are quite
similar to each other and fully consistent with the solid-state
structure of 8. All these compounds display one low-frequency
C−O stretch (1659−1670 cm−1) in the IR spectra and a quite
deshielded 13C resonance (ca. 300 ppm), as expected for
compounds having a bridging carbonyl ligand and also found
for the isostructural carbynes 2 and 5. We note that the
corresponding C−O stretching frequencies follow the order 8
(1970) > 6 (1960) > 5 (1937) > 2 (1925 cm−1), in full
agreement with the absence of any π(C−O) bonding
interaction involving the P−O−C linkage in 8 discussed
above. Moreover, from these figures we should conclude that
the same holds for the phosphinoxycarbyne ligands in
complexes 6. In addition, the presence of μ-COX ligands is
firmly established by the observation of characteristic highly
874
DOI: 10.1021/om501202j
Organometallics 2015, 34, 870−878

Organometallics
Article
deshielded resonances in the 13C NMR spectra (δC ca. 360
ppm), with shift comparable to those typically observed for
bridging alkoxy- and hydroxycarbyne ligands in related
complexes such as 2, [W2Cp2(μ-COR)(μ-PPh2)2]BF4 (R =
H, Me),17 [Mo2Cp2(μ-COMe)(μ-PEt2)2]BF4,6a and
[Mo2Cp2(μ-COR)(μ-PCy2)(μ-CO)] (R = Me, Et).3 As
observed for the carbynes 2 and 5, complexes 6 and 8 also
exhibit highly deshielded 31P resonances for their bridging PCy2
ligands, retaining large one-bond P−W coupling in the range
351−362 Hz, while the OPR2 and OP(O)(OR)2 groups give
Solution Structure of Complexes 9 and 10. The
available spectroscopic data for complexes 9 and 10 are similar
to those of the corresponding Mo2 analogues; therefore, similar
structures are proposed for these molecules, with transoid
arrangement of the WCp(CO) fragments and a bridging κ2:κ2
coordination of the diphosphorus fragment, leaving one P atom
close to the W2P(Cy2) plane. The Na+ salt of anion 9 displays
four C−O stretching bands in the IR spectrum, which is
consistent with the presence of a fast equilibrium between tight
(1824 and 1695 cm−1) and solvent-separated (1824 and 1757
cm−1) ion pairs in solution, a situation studied in detail for the
Mo2 anion.23b Unfortunately the low solubility of this salt in
typical organic solvents prevented us from recording the
corresponding low-temperature 31P NMR spectra, and at room
temperature we could identify only the average resonance for
its bridging PCy2 ligand (70.7 ppm), while resonances for the
P2 unit were not observed, probably because of their broadness
(we note that the P2 ligand in the Mo2 analogue of 9 gives rise
to a very broad resonance at room temperature, −166.4 ppm,
which eventually splits to give separated peaks with shifts of
−80.0 and −273.0 ppm at 163 K). Yet the PCy2 coupling of 16
Hz with both atoms of the P2 ligand in 9 indicates a dynamic
behavior comparable to that observed for its Mo2 analogue,
whereby the two inequivalent sites of the P2 ligand are mutually
exchanged.23b
3
rise to a second resonance with low (6b, JPW = 14 Hz) or
negligible coupling to tungsten.
As for the bis(phosphide) complex 7c, the available
spectroscopic data (Table 1 and Experimental Section) are
very similar to those of the large family of complexes of type
trans-[M2Cp2(μ-PR2)(μ-PR′2)(CO)2] (M = Mo, W; R, R′ =
Ph, Cy, Et, H, etc.) prepared some time ago by our group and
others,17,19 and then a detailed analysis is not needed. The most
characteristic spectroscopic features of these molecules are their
relatively low 31P chemical shifts with medium W−P couplings
and very small P−P coupling and a C−O stretching pattern
(two bands with weak and strong intensity, in order of
decreasing frequency) as expected for transoid M2(CO)2
oscillators.20 The 31P NMR resonance of 7b displays a chemical
shift and P−W coupling comparable to those of the PCy2
resonance of the mixed-phosphide complex 7c (Table 1) and
most likely has the same transoid structure.
Reaction of 1 with White Phosphorus. Activation of the
P4 molecule by transition metal complexes is a matter of
current interest as an alternative, environmentally more friendly
route to the functionalization of white phosphorus, to
eventually build organophosphorus compounds for many
different uses.21 Such activation generally relies on the
electrophilic character of the metal complex, which enables
the formation of P4 complexes, thus initiating activation. In
contrast, compounds of the p-block elements can activate white
phosphorus either through electrophilic or, more commonly,
through nucleophilic attack on the P4 molecule.22 Given the
high nucleophilicity of anion 1, it might be anticipated that it
should be able to cleave P−P bonds in the P4 tetrahedron
under mild conditions, especially after considering that its
dimolybdenum analogue has been shown previously to induce a
symmetrical cleavage of this molecule under mild conditions to
give a novel anionic diphosphorus-bridged complex.23 Indeed,
suspensions of 1-Na in THF react at 333 K with one equivalent
of P4 to give the Na+ salt of the related diphosphorus-bridged
anion [W2Cp2(μ-κ2:κ2-P2)(μ-PCy2)(CO)2]− (9) (Chart 4).
As for the methyldiphosphenyl complex 10, its IR spectrum
displays two C−O stretching bands with a pattern (strong and
very strong, in order of decreasing frequency) consistent with a
distorted transoid arrangement of the M2(CO)2 oscillator,20 as
determined crystallographically for its Mo2 analogue. The 31P
NMR resonances of the P2Me group expectedly appear as
strongly shielded multiplets [−166.0 (PMe) and −390.1 ppm
(P)], with a large mutual coupling of 467 Hz, which is
characteristic of directly bound phosphorus atoms retaining
some π-interaction, as commonly observed for a variety of κ1-
diphosphene complexes,24 and also found for its Mo2 analogue
(1JPP = 503 Hz), which in turn is consistent with the relatively
short P−P bond distance found in the solid-state structure of
the latter complex (2.085(1) Å).23b We finally note that the
PCy2 ligand in 10 displays two sets of 183W satellites (1JPW
=
259, 238 Hz), in agreement with the inequivalence of the metal
centers, while their values (lower than the figures of ca. 280 Hz
for the 32-electron complexes of type 6) reflect a somewhat
increased coordination number at the metals, due to the κ2:κ2
coordination of the diphosphenyl ligand. The PMe atom of the
latter ligand displays comparable P−W couplings, as expected
(1JPW = 255, 235 Hz). However, the P atom of this ligand
unexpectedly displays an extremely low coupling of 25 Hz to
each of the W atoms (Table 1). We have found a similar
spectroscopic anomaly in the agostic-like diphosphenyl
complex [Mo2Cp2(μ-κ2:κ1,η2-HP2)(μ-PCy2)(CO)2], which dis-
plays an anomalously low P−H coupling of 4 Hz attributed to a
negative contribution to this coupling of the lone pair located at
the P atom,23b and a similar effect might be operative for the P-
PMe resonance in complex 10.
Chart 4
CONCLUSION
■
This air-sensitive Na+ salt (9-Na) could not be isolated, but was
trapped easily through reaction with MeI at 273 K, rendering
the corresponding methyldiphosphenyl-bridged complex
[W2Cp2(μ-κ2:κ2-P2Me)(μ-PCy2)(CO)2] (10), a more stable
species that could be isolated in good yield by conventional
methods.
Compared to its molybdenum analogue, the Na+ salt of the
ditungsten anion [W2Cp2(μ-PCy2)(μ-CO)2]− (1-Na) behaves
similarly when faced with C-based electrophiles, but with an
increased nucleophilicity at the O atoms of the bridging
carbonyls favoring electrophile attachment at this site.
Incorporation of the electrophile at the metal site is dominant
875
DOI: 10.1021/om501202j
Organometallics 2015, 34, 870−878

Organometallics
Article
under vacuum, and the residue was extracted with dichloromethane/
petroleum ether (1:3) and chromatographed through alumina at 288
K. An orange fraction was eluted using dichloromethane. Removal of
solvent from this fraction gave compound 2 as an orange solid (0.016
g, 53%). Compound 3 could not be isolated from this mixture;
therefore the corresponding NMR data were obtained from the crude
reaction mixtures. We also note that this reaction can be completed in
5 min if “wet” THF is used as solvent, but the ratio of products thus
obtained was essentially the same. Anal. Calcd for compound
C25H35O2PW2 (2): C, 39.19; H, 4.60. Found: C, 39.03; H, 4.41.
for benzyl chloride to give the benzyl-bridged complex
[W2Cp2(μ-CH2Ph)(μ-PCy2)(CO)2], but this is a minor
pathway in the MeI reaction, which instead results in the
incorporation of the electrophile at the O sites of the anion to
give the methoxycarbyne complex [W2Cp2(μ-COMe)(μ-
PCy2)(μ-CO)]. This O-based nucleophilicity of anion 1
induces the initial formation of related phosphinoxycarbyne
complexes [W2Cp2(μ-COPR2)(μ-PCy2)(μ-CO)] upon reac-
tion with the corresponding chlorophosphines ClPR2 (R = tBu,
Cy, Et), a behavior not observed for the analogous Mo2 anion.
These products, however, are quite unstable and either
rearrange thermally to give the bis(phosphide) isomers
[W2Cp2(μ-PCy2)(μ-PR2)(CO)2] when R = Et, Cy, a process
facilitated by the presence of a lone electron pair at phosphorus,
or evolve via hydrolysis (to give the hydride [W2Cp2(H)(μ-
PCy2)(CO)2]), or undergo more complex processes. The use
of P(V) electrophiles, in contrast, yields related but more stable
products, as shown by the reaction of 1-Na with ClP(O)-
(OPh)2 to give [W2Cp2{μ-COP(O)(OR)2}(μ-PCy2)(μ-CO)],
which stands as the first phosphatecarbyne complex to ever be
reported. Spectroscopic and crystallographic data in the above
carbyne complexes reveal the presence of a negligible π-
bonding component in the corresponding C−O−P linkages,
which should make the dimetal center more electrophilic in the
phosphinoxycarbyne and phosphatecarbyne complexes than in
related alkoxycarbyne complexes. In contrast to the above
differences, no significant effect of the metal (W instead of Mo)
was observed in the reaction of 1-Na with P4, which follows the
same route found for its Mo2 analogue to yield the new
diphosphorus-bridged complex [W2Cp2(μ-κ2:κ2-P2)(μ-PCy2)-
(CO)2]−, an anion with P2-centered nucleophilicity, as probed
by its reaction with MeI to give the methyldiphosphenyl
derivative [W2Cp2(μ-κ2:κ2-P2Me)(μ-PCy2)(CO)2].
1
Spectroscopic data for 2: H NMR: δ 6.00 (s, 10H, Cp), 3.60 (s, 3H,
OMe), 2.00−0.50 (m, 22H, Cy). 13C{1H} NMR (100.63 MHz): δ
2
342.9 (d, JCP = 9, μ-COMe), 302.3 (s, μ-CO), 93.7 (s, Cp), 66.8 (s,
1
OMe), 43.2 [d, JCP = 25, C1(Cy)], 41.9 [d, 1JCP = 23, C1(Cy)], 34.4,
3
34.3 [2s, C2(Cy)], 27.5 [d, JCP = 15, 2C3(Cy)], 26.5, 26.3 [2s,
1
C4(Cy)]. Spectroscopic data for 3: H NMR: δ 5.28 (s, 10H, Cp),
2.20−0.30 (m, 22H, Cy), −1.18 (d, JHP = 2, 3H, μ-Me).
Preparation of [W2Cp2(μ-CH2Ph)(μ-PCy2)(CO)2] (4). A suspen-
sion containing ca. 0.037 mmol of compound 1-Na in tetrahydrofuran
(5 mL) was stirred with an excess of benzyl chloride (40 μL, 0.35
mmol) for 2 days. Solvent was then removed under vacuum, the
residue was extracted with dichloromethane (2 × 6 mL), and the
extracts were filtered through diatomaceous earth. Solvent was
removed again from the filtrate, and the residue was extracted with
dichloromethane/petroleum ether (1:6) and chromatographed
through alumina at 288 K. Elution with the same solvent mixture
gave a brown fraction yielding, after removal of solvents, compound 4
as a brown solid (0.026 g, 82%). Anal. Calcd for C31H39O2PW2: C,
1
44.20; H, 4.67. Found: C, 44.04; H, 4.44. H NMR (400.13 MHz): δ
2
7.30−6.70 (m, 5H, C6H5), 5.28 (s, 10H, Cp), 2.37 (d, JHH = 15, 1H,
2
3
μ-CH2), 2.50−1.00 (m, 22H, Cy), −2.59 (dd, JHH = 15, JHP = 1.7,
1H, μ-CH2). 1H NMR (400.13 MHz, 273 K): δ 5.30 (s, br, 10H, Cp),
2.60 (d, 2JHH = 15, 1H, μ-CH2), −2.92 (dd, 2JHH = 15, 3JHP = 1.7, 1H,
μ-CH2). 1H NMR (400.13 MHz, 253 K): δ 5.53, 5.13 (2s, br, 2 × 5H,
Cp), 2.68 (d, JHH = 15, 1H, μ-CH2), −3.04 (dd, 2JHH = 15, 3JHP = 1.7,
1H, μ-CH2). 1H NMR (400.13 MHz, 233 K): δ 5.56, 5.14 (2s, 2 × 5H,
2
2
Cp), 2.73 (d, JHH = 15, 1H, μ-CH2), −3.11 (d, br, JHH = 15, 1H, μ-
CH2). 1H NMR (400.13 MHz, 213 K): δ 5.58, 5.16 (2s, 2 × 5H, Cp),
2
2
EXPERIMENTAL SECTION
2.76 (d, br, JHH = 15, 1H, μ-CH2), −3.14 (d, br, JHH = 15, 1H, μ-
CH2). 13C{1H} NMR (100.62 MHz): δ 235.5 (s, br, CO), 155.5 [s,
C1(Ph)], 127.6, 127.5 [2s, C2,3(Ph)], 122.3 [s, C4(Ph)], 87.6 (s, Cp),
■
General Procedures and Starting Materials. All manipulations
and reactions were carried out under a nitrogen (99.995%)
atmosphere using standard Schlenk techniques. Solvents were purified
according to literature procedures and distilled prior to use.25
Tetrahydrofuran suspensions of Na[W2Cp2(μ-PCy2)(μ-CO)2] (1-
Na)8 were prepared as described previously, and 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 338−343 K. Filtrations were carried out
through diatomaceous earth unless otherwise indicated. Chromato-
graphic separations were carried out using jacketed columns
refrigerated by tap water (ca. 288 K) or by a closed 2-propanol
circuit kept at the desired temperature with a cryostat. Commercial
aluminum oxide (activity I, 150 mesh) was degassed under vacuum
prior to use. The latter was mixed afterward under nitrogen with the
appropriate amount of water to reach the activity desired (activity IV,
unless otherwise stated). IR stretching frequencies of CO ligands were
measured in solution (using CaF2 windows), in Nujol mulls (using
NaCl windows), and are referred to as ν(CO) (solvent) and ν(CO)
(Nujol), respectively. Nuclear magnetic resonance spectra were
routinely recorded at 300.13 (1H), 121.50 (31P{1H}), and 75.47
MHz (13C{1H}) at 295 K in CD2Cl2 solution unless otherwise stated.
Chemical shifts (δ) are given in ppm, relative to internal
tetramethylsilane (1H, 13C) or external 85% aqueous H3PO4 (31P).
Coupling constants (J) are given in hertz.
49.5 [d, JCP = 25, C1(Cy)], 34.2 [d, JCP = 4, C2(Cy)], 33.3 [s,
C2(Cy)], 28.4 [d, 3JCP = 12, C3(Cy)], 28.2 [d, 3JCP = 11, C3(Cy)], 26.6
[s, C4(Cy)], −8.7 (s, 1JCW = 31, μ-CH2). 13C NMR (100.62 MHz): δ
1
2
1
−8.7 (dd, JCH = 130, 90, μ-CH2).
Preparation of [W2Cp2(μ-CPh)(μ-PCy2)(μ-CO)] (5). A toluene
solution (6 mL) of complex 4 (0.020 g, 0.024 mmol) was irradiated
with visible−UV light in a Pyrex Schlenk flask at 288 K for 15 min with
a gentle N2 purge to give an orange solution. Solvent was then
removed under vacuum, the residue was extracted with dichloro-
methane/petroleum ether (1:1), and the extracts were chromato-
graphed through alumina at 288 K. Elution with dichloromethane gave
a yellowish-green fraction, yielding, after removal of solvents,
compound 5 as a yellow solid (0.017 g, 86%). Anal. Calcd for
1
C30H37OPW2: C, 44.36; H, 4.59. Found: C, 44.19; H, 4.54. H NMR
(400.13 MHz): δ 7.13 [m, 2H, H3(Ph)], 6.97 [tt, JHH = 7, 1, 1H,
H4(Ph)], 6.55 [m, 2H, H2(Ph)], 6.03 (s, 10H, Cp), 2.00 (m, 1H, Cy),
1.78−1.68 (m, 5H, Cy), 1.59 (m, 2H, Cy), 1.25, 1.18 (2m, 2 × 4H,
Cy), 1.01 (m, 2H, Cy), 0.57 (m, 4H, Cy). 13C{1H} NMR (100.63
MHz): δ 363.9 (d, JCP = 8, μ-CPh), 300.3 (s, μ-CO), 169.9 [s,
C1(Ph)], 127.7 [s, C3(Ph)], 124.0 [s, C4(Ph)], 120.9 [s, C2(Ph)], 95.7
(s, Cp), 44.5 [d, 1JCP = 26, C1(Cy)], 42.5 [d, 1JCP = 23, C1(Cy)], 34.3,
3
34.1 [2s, C2(Cy)], 27.4, 27.3 [2d, JCP = 14, C3(Cy)], 26.4, 26.2 [2s,
C4(Cy)].
Preparation of [W2Cp2(μ-COPtBu2)(μ-PCy2)(μ-CO)] (6a). A
suspension containing ca. 0.037 mmol of compound 1-Na in
tetrahydrofuran (5 mL) was stirred with excess ClPtBu2 (20 μL,
0.116 mmol) to give almost instantaneously an orange solution
containing compound 6a as the major product. Solvent was then
removed under vacuum, the residue was extracted with toluene (3 × 3
Reaction of Complex 1 with MeI. A suspension containing ca.
0.037 mmol of compound 1-Na in tetrahydrofuran (5 mL) was stirred
with an excess MeI (100 μL, 1.6 mmol) for 4 days at room
temperature to give an orange solution containing a mixture of
compounds [W2Cp2(μ-COMe)(μ-PCy2)(μ-CO)] (2) and [W2Cp2(μ-
CH3)(μ-PCy2)(CO)2] (3) in a 3:1 ratio. Solvent was then removed
876
DOI: 10.1021/om501202j
Organometallics 2015, 34, 870−878

Organometallics
Article
mL), and the extracts were filtered using a canula. Removal of solvent
under vacuum yielded an orange solid containing 6a as major
component. However, pure samples of this air-sensitive product could
not be obtained due to its progressive decomposition upon
turned out to be highly air-sensitive, and all attempts to isolate it as a
pure solid led to its progressive decomposition.
Preparation of [W2Cp2(μ-PCy2)(μ-κ2:κ2-P2Me)(CO)2] (10).
Excess MeI (20 μL, 0.32 mmol) was added to a freshly prepared
solution of 9-Na (ca. 0.05 mmol) in THF (5 mL), and the mixture was
stirred for 5 min at 253 K to yield an orangish-red solution. Solvent
was then removed under vacuum, the residue was extracted with
dichloromethane/petroleum ether (1:4), and the extracts were
chromatographed through alumina at 288 K. A red fraction was
eluted using dichloromethane/petroleum ether (1:2), which gave,
upon removal of solvents, compound 10 as an air-sensitive red solid
(0.021 g, 50%). Anal. Calcd for C25H35O2P3W2: C, 36.26; H, 4.26.
Found: C, 35.95; H, 4.05. 1H NMR: δ 5.72 (s, 5H, Cp), 5.14 (dd, JPH
= 2, 1, 5H, Cp), 2.07 (ddd, JPH = 12, 3, 1, 3H, Me), 1.90−0.90 (m,
22H, Cy).
1
manipulation. H NMR (400.13 MHz, tol-d8): δ 5.97 (s, 10H, Cp),
2.50−0.50 (m, 22H, Cy), 0.97 (d, 3JHP = 11, 18H, tBu). 13C{1H} NMR
(100.63 MHz, tol-d8): δ 342.7 (dd, JCP = 9, 7, μ-COP), 291.6 (s, μ-
1
1
CO), 94.2 (s, Cp), 43.9 [d, JCP = 26, C1(Cy)], 43.6 [d, JCP = 25,
C1(Cy)], 35.6 [d, 1JCP = 28, C1(tBu)], 34.4, 34.0 [2s, C2(Cy)], 29.6 [d,
3
2JCP = 16, C2(tBu)], 27.6 [d, JCP = 15, 2C3(Cy)], 25.6 [s, 2C4(Cy)].
Preparation of Solutions of [W2Cp2(μ-COPCy2)(μ-PCy2)(μ-
CO)] (6b). A suspension containing ca. 0.037 mmol of compound 1-
Na in tetrahydrofuran (5 mL) was stirred with ClPCy2 (8 μL, 0.037
mmol) at 253 K for 3 min to give almost instantaneously an orange
solution, shown by 31P NMR to contain compound 6b as the major
product. This species, however, was unstable and evolved within 1 h in
solution at room temperature to yield a mixture of the hydride
[W2Cp2(H)(μ-PCy2)(CO)2], the phosphide 7b, and other products
that could not be isolated.
Reaction of Complex 1 with ClPEt2. A suspension containing ca.
0.037 mmol of compound 1-Na in tetrahydrofuran (5 mL) was stirred
with excess ClPEt2 (8 μL, 0.07 mmol) to give almost instantaneously a
brown solution containing compound [W2Cp2(μ-PCy2)(μ-PEt2)-
(CO)2] (7c) as major product, along with smaller and variable
amounts of [W2Cp2(μ-COPEt2)(μ-PCy2)(μ-CO)] (6c) and the
known complex [W2Cp2(H)(μ-PCy2)(CO)2]. Solvent was then
removed under vacuum, the residue was extracted with dichloro-
methane (3 × 5 mL), and the extracts were filtered through
diatomaceous earth. Solvent was removed again from the filtrate, the
residue was extracted with dichloromethane/petroleum ether (1:8),
and the extracts were chromatographed through alumina at 253 K.
Elution with the same solvent mixture gave a green fraction, yielding,
after removal of solvents, compound 7c as a green solid (0.019 g,
61%). Complex 6c could not be isolated nor fully characterized due to
its fast decomposition in solution. Anal. Calcd for C28H42O2P2W2
(7c): C, 40.02; H, 5.04. Found: C, 40.17; H, 4.99. Spectroscopic data for
7c: 1H NMR (400.54 MHz): δ 5.35 (s, 10H, Cp), 2.79, 2.39 [2m, 2 ×
2H, CH2(Et)], 1.90−0.80 [m, 28H, Cy and CH3].
X-ray Crystal Structure Determination for Compound 8.
Diffraction data were collected on a Kappa-Apex-II Bruker
diffractometer using graphite-monochromated Mo Kα radiation at
100 K. The software APEX26 was used for collecting frames with the
ω/ϕ scan measurement method. The SAINT software was used for
data reduction,27 and a multiscan absorption correction was applied
with SADABS.28 Using the program suite WinGX,29 the structure was
solved by Patterson interpretation and phase expansion using
SHELXL14 and refined with full-matrix least-squares on F2 using
SHELXL14.30 One of the cyclohexyl groups of the molecule was found
to be disordered over two positions and satisfactorily modeled with
0.6/04 occupancies and some restraints in the C−C distances. All non-
H atoms were refined anisotropically except for the carbon atoms
involved in disorder, which were refined isotropically to prevent their
temperature factors from becoming nonpositive definite. All hydrogen
atoms were geometrically placed and refined using a riding model.
ASSOCIATED CONTENT
* Supporting Information
■
S
A PDF file containing a table with crystal data for compound 8,
figures displaying 31P NMR and IR spectra of compounds 6a−
c, and a CIF file containing full crystallographic data for 8
(CCDC 1033926). This material is available free of charge via
the Internet at http://pubs.acs.org.
Preparation of [W2Cp2{μ-COP(O)(OPh)2}(μ-PCy2)(μ-CO)] (8). A
suspension containing ca. 0.062 mmol of compound 1-Na and
ClP(O)(OPh)2 (12 μL, 0.058 mmol) in tetrahydrofuran (5 mL) was
stirred in a Schlenk flask equipped with a Young’s valve for 40 min to
give a brown solution. Solvent was then removed under vacuum, the
residue was extracted with toluene (3 × 5 mL), and the extracts were
filtered through diatomaceous earth. Solvent was removed again from
the filtrate, and the residue was extracted with dichloromethane and
chromatographed through alumina at 263 K. Elution with the same
solvent gave an orange fraction yielding, after removal of solvents,
compound 8 as an orange solid (0.053 g, 87%). The crystals used in
the X-ray diffraction study were grown by the slow diffusion of a layer
of petroleum ether into a concentrated toluene solution of the
complex at room temperature. Anal. Calcd for C36H42O5P2W2: C,
43.93; H, 4.30. Found: C, 44.26; H, 4.35. ν(CO) (Nujol): 1657 (vs)
cm−1. 1H NMR: δ 7.35 [m, 4H, H3(Ph)], 7.21 [m, 2H, H4(Ph)], 7.24
[m, 4H, H2(Ph)], 6.05 (s, 10H, Cp), 2.20−0.03 (m, 22H, Cy).
13C{1H} NMR: δ 330.0 (dd, JCP = 14, 7, μ-COP), 297.4 (s, μ-CO),
AUTHOR INFORMATION
Corresponding Authors
*E-mail: garciavdaniel@uniovi.es (D. Garcıa-Vivo)
*E-mail: mara@uniovi.es (M. A. Ruiz).
■
́
́
.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the DGI of Spain for a grant (to M.F.V.) and
financial support (Projects CTQ2009-09444 and CTQ2012-
33187). We also thank the Universidad de Santiago de
Compostela (X-ray unit) for collection of diffraction data.
2
149.8 [d, JCP = 8, C1(Ph)], 129.0 [s, C3(Ph)], 124.7 [s, C4(Ph)],
REFERENCES
■
119.3 [d, 3JCP = 5, C2(Ph)], 94.7 (s, Cp), 42.0 [d, 1JCP = 26, C1(Cy)],
41.1 [d, 1JCP = 24, C1(Cy)], 33.1, 32.8 [2s, C2(Cy)], 26.2 [d, 3JCP = 13,
2C3(Cy)], 25.0 [s, 2C4(Cy)].
(1) (a) García-Vivo,
́
D.; Ramos, A.; Ruiz, M. A. Coord. Chem. Rev.
2013, 257, 2143. (b) Winter, M. J. Adv. Organomet. Chem. 1989, 29,
101. (c) Curtis, M. D. Polyhedron 1987, 6, 759.
(2) Poli, R.; Rheingold, A. L. J. Chem. Soc., Chem. Commun. 1990,
552.
Preparation of Tetrahydrofuran Solutions of Na[W2Cp2(μ-
PCy2)(μ-κ2:κ2-P2)(CO)2] (9-Na). A solution of P4 in toluene (250 μL
of a solution ca. 0.25 M, 0.062 mmol) was placed into a reaction vessel
equipped with a Young’s valve, and the solvent was removed under
vacuum to give a white residue of solid P4. A freshly prepared
suspension of 1-Na (ca. 0.05 mmol) in THF (5 mL) was then
transferred using a canula to the reaction vessel, and the mixture was
stirred for 40 min at 333 K to give an orange solution containing some
excess P4 and the Na+ salt of anion 9 ready for further use. This salt
(3) (a) García, M. E.; Melon
2009, 8171. (b) García, M. E.; Melon
M. A.; Belletti, D.; Graiff, C.; Tiripicchio, A. Organometallics 2003, 22,
1983.
́
, S.; Ramos, A.; Ruiz, M. A. Dalton Trans.
́
, S.; Ramos, A.; Riera, V.; Ruiz,
(4) (a) Alvarez, M. A.; García, M. E.; Ramos, A.; Ruiz, M. A. J.
Organomet. Chem. 2010, 695, 36. (b) Alvarez, M. A.; García, M. E.;
Ramos, A.; Ruiz, M. A. J. Organomet. Chem. 2009, 694, 3864.
877
DOI: 10.1021/om501202j
Organometallics 2015, 34, 870−878

Organometallics
Article
(5) (a) Alvarez, C. M.; Alvarez, M. A.; García, M. E.; Ramos, A.; Ruiz,
M. A.; Lanfranchi, M.; Tiripicchio, A. Organometallics 2005, 24, 7.
Chem. Rev. 2010, 110, 4178. (c) Peruzzini, M.; Gonsalvi, L.; Romerosa,
A. Chem. Soc. Rev. 2005, 34, 1038. (d) Peruzzini, M.; Abdreimova, R.
R.; Budnikova, Y.; Romerosa, A.; Scherer, O. J.; Sitzmann, H. J.
Organomet. Chem. 2004, 689, 4319.
(b) García, M. E.; Ramos, A.; Ruiz, M. A.; Lanfranchi, M.; Marchio,
Organometallics 2007, 26, 6197 and references therein.
(6) (a) García, M. E.; García-Vivo, D.; Ruiz, M. A.; Alvarez, S.;
Aullon, G. Organometallics 2007, 26, 4930. (b) García, M. E.; García-
Vivo, D.; Ruiz, M. A. Organometallics 2010, 29, 2157 and references
therein.
(7) (a) Alvarez, M. A.; García-Vivo,
E.; Ramos, A.; Ruiz, M. A. Organometallics 2008, 27, 1973. (b) Alvarez,
M. A.; García, M. E.; García-Vivo, D.; Martínez, M. E.; Ramos, A.;
Ruiz, M. A. Dalton Trans. 2014, 43, 7780 and references therein.
(8) Alvarez, M. A.; García, M. E.; García-Vivo, D.; Ruiz, M. A.; Vega,
M. F. Organometallics 2010, 29, 512.
̀
L.
(22) (a) Giffin, N. A.; Masuda, J. D. Coord. Chem. Rev. 2011, 255,
́
1342. (b) Scheer, M.; Balaz
(23) (a) Alvarez, M. A.; García, M. E.; García-Vivo,
Ruiz, M. A. Inorg. Chem. 2011, 50, 2064. (b) Alvarez, M. A.; García, M.
E.; García-Vivo, D.; Ramos, A.; Ruiz, M. A. Inorg. Chem. 2012, 51,
11061.
́
s, G.; Seitz, A. Chem. Rev. 2010, 110, 4236.
́
́
D.; Ramos, A.;
́
́
́
D.; García, M. E.; Martínez, M.
(24) Weber, L. Chem. Rev. 1992, 92, 1839.
́
(25) Armarego, W. L. F.; Chai, C. Purification of Laboratory
Chemicals, 5th ed.; Butterworth-Heinemann: Oxford, UK, 2003.
(26) APEX 2, version 2.0-1; Bruker AXS Inc: Madison, WI, 2005.
(27) SMART & SAINT Software Reference Manuals, Version 5.051
(Windows NT Version); Bruker Analytical X-ray Instruments:
Madison WI, 1998.
́
(9) For some mononuclear complexes see: (a) Colebatch, A. L.; Hill,
A. F.; Sharma, M. Organometallics 2015, 34 (in press, DOI: 10.1021/
om500833n). (b) Cordiner, R. L.; Gugger, P. A.; Hill, A. F.; Willis, A.
C. Organometallics 2009, 28, 6632. (c) Li, X.; Wang, A.; Wang, L.; Sun,
H.; Harms, K.; Sundermeyer, J. Organometallics 2007, 26, 1411 and
references therein. (d) Greco, J. B.; Peters, J. C.; Baker, T. A.; Davis,
W. M.; Cummins, C. C.; Wu, G. J. Am. Chem. Soc. 2001, 123, 5003.
(e) Weber, L.; Dembeck, G.; Loenneke, P.; Stammlee, H. G.;
Neumann, B. Organometallics 2001, 20, 2288 and references therein.
(f) Jamison, G. M.; White, P. S.; Templeton, J. L. Organometallics
1991, 10, 1954.
(28) Sheldrick, G. M. SADABS, Program for Empirical Absorption
Correction; University of Gottingen: Gottingen, Germany, 1996.
̈
̈
(29) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837.
(30) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112.
(10) There are only a few examples of carbyne-bridged (μ2) binuclear
complexes: (a) Schroder, F. G.; Lichtenberg, C.; Elfferdin, M.;
̈
Sundermeyer, J. Organometallics 2013, 32, 5082. (b) Hill, A. F.;
Sharma, M.; Willis, A. C. Organometallics 2012, 31, 2538. (c) Hughes,
K. A.; Dopico, P. G.; Sabat, M.; Finn, M. G. Angew. Chem., Int. Ed.
Engl. 1993, 32, 554. (d) Schmidbaur, H.; Pichl, R.; Muller, G. Angew.
̈
Chem., Int. Ed. Engl. 1986, 25, 574. (e) Rice, G. W.; Ansell, G. B.;
Modrick, M. A.; Zentz, S. Organometallics 1983, 2, 154.
(11) For some polynuclear complexes with μ3-carbyne ligands, see:
(a) Wong, W. Y. Molecules 2005, 10, 640. (b) Braunstein, P.; Graiff,
C.; Morise, X.; Tiripicchio, A. J. Organomet. Chem. 1997, 541, 417.
(c) Choi, Y. Y.; Wong, W. Y.; Wong, W. T. J. Organomet. Chem. 1996,
518, 227 and references therein. (d) Heineke, D.; Bohle, D. S.;
Vahrenkamp, H. Chem. Ber. 1993, 126, 355 and references therein.
(e) Johnson, B. F. G.; Lahoz, F. J.; Lewis, J.; Prior, N. D.; Raithby, P.
R.; Wong, W. T. J. Chem. Soc., Dalton Trans. 1992, 1701. (f) Ching, S.;
Sabat, M.; Shriver, D. F. Organometallics 1989, 8, 1047. (g) Schmid-
baur, H.; Scherbaum, F.; Huber, B.; Muller, G. Angew. Chem. 1988,
̈
100, 441. (h) Seyferth, D.; Hallgreen, J. E.; Hung, P. L. K. J.
Organomet. Chem. 1973, 50, 265.
(12) Alvarez, M. A.; García, M. E.; García-Vivo,
M. F. Dalton Trans. 2014, 43, 16044.
(13) (a) Alvarez, M. A.; García, M. E.; Martínez, M. E.; Menen
Ruiz, M. A. Organometallics 2010, 29, 710. (b) Alvarez, M. A.; García,
M. E.; García-Vivo, D.; Martínez, M. E.; Ruiz, M. A. Organometallics
2011, 30, 2189.
́
D.; Ruiz, M. A.; Vega,
́
dez, S.;
́
(14) Jeffery, J. C.; Orpen, A. G.; Stone, F. G. A.; Went, M. J. J. Chem.
Soc., Dalton Trans. 1986, 173.
(15) Brookhart, M.; Green, M. L. H. J. Organomet. Chem. 1983, 250,
395.
(16) Jameson, C. J. In Phosphorus-31 NMR Spectroscopy in
Stereochemical Analysis; Verkade, J. G., Quin, L. D., Eds.; VCH:
Deerfield Beach, FL, 1987; Chapter 6.
(17) García, M. E.; Riera, V.; Ruiz, M. A.; Rueda, M. T.; Sae
́
z, D.
Organometallics 2002, 21, 5515.
(18) García, M. E.; Riera, V.; Rueda, M. T.; Ruiz, M. A.; Halut, S. J.
Am. Chem. Soc. 1999, 121, 1960.
(19) Adatia, T.; McPartlin, M.; Mays, M. J.; Morris, M. J.; Raithby, P.
R. J. Chem. Soc., Dalton Trans. 1989, 1555.
(20) Braterman, P. S. Metal Carbonyl Spectra; Academic Press:
London, U.K., 1975.
(21) For some reviews on transition-metal-mediated activation of P4,
see: (a) Cossairt, B. M.; Piro, N. A.; Cummins, C. C. Chem. Rev. 2010,
110, 4164. (b) Caporali, M.; Gonsalvi, L.; Rossin, A.; Peruzzini, M.
878
DOI: 10.1021/om501202j
Organometallics 2015, 34, 870−878