Synthesis and decarbonylation reactions of the triiron phosphinidene complex [Fe3Cp3(μ-H)(μ3-PPh)(CO) 4]: Easy cleavage and formation of P-H and Fe-Fe bonds

Article abstract of DOI:10.1021/ic201491v

The binuclear phosphine complex [Fe₂Cp₂(μ-CO) ₂(CO)(PH₂Ph)] (Cp = η⁵-C₅H ₅) reacted with the acetonitrile adduct [Fe₂Cp ₂(μ-CO)₂(CO)(NCMe)] in di

Full text of DOI:10.1021/ic201491v

ARTICLE
pubs.acs.org/IC
Synthesis and Decarbonylation Reactions of the Triiron
Phosphinidene Complex [Fe3Cp3(μ-H)(μ3-PPh)(CO)4]: Easy Cleavage
and Formation of PꢀH and FeꢀFe Bonds
Celedonio M. Alvarez, M. Angeles Alvarez, M. Esther García, Rocío Gonzꢀalez, Alberto Ramos, and
Miguel A. Ruiz*
Departamento de Química Orgꢀanica e Inorgꢀanica/IUQOEM, Universidad de Oviedo, E-33071 Oviedo, Spain
S Supporting Information
b
ABSTRACT: The binuclear phosphine complex [Fe2Cp2(μ-CO)2-
(CO)(PH2Ph)] (Cp = η5-C5H5) reacted with the acetonitrile
adduct [Fe2Cp2(μ-CO)2(CO)(NCMe)] in dichloromethane at
293 K to give the trinuclear hydrideꢀphosphinidene derivative
[Fe3Cp3(μ-H)(μ3-PPh)(CO)4] as a mixture of cis,anti and trans
isomers (FeꢀFe = 2.7217(6) Å for the cis,anti isomer). In contrast, photochemical treatment of the phosphine complex with
[Fe2Cp2(CO)4] gave the phosphide-bridged complex trans-[Fe3Cp3(μ-PHPh)(μ-CO)2(CO)3] as the major product, along with small
amounts of the binuclear hydrideꢀphosphide complexes [Fe2Cp2(μ-H)(μ-PHPh)(CO)2] (cis and trans isomers), which are more
selectively prepared from [Fe2Cp2(CO)4] and PH2Ph at 388 K. The photochemical decarbonylation of either of the mentioned triiron
compounds led reversibly to three different products depending on the reaction conditions: (a) the 48-electron phosphinidene cluster
[Fe3Cp3(μ-H)(μ3-PPh)(μ-CO)2] (FeꢀFe = 2.592(2)ꢀ2.718(2) Å); (b) the 50-electron complex [Fe3Cp3(μ-H)(μ3-PPh)(μ-CO)-
(CO)2], also having carbonyl- and hydride-bridged metalꢀmetal bonds (FeꢀFe = 2.6177(3) and 2.7611(4) Å, respectively); and (c) the
48-electron phosphide cluster [Fe3Cp3(μ-PHPh)(μ3-CO)(μ-CO)2], an isomer of the latter phosphinidene complex now having three
intermetallic bonds (FeꢀFe = 2.5332(8)ꢀ2.6158(8) Å).
’ INTRODUCTION
complexes structurally characterized is rather limited,4,5 even fewer
of them also display bridging hydride ligands, and moreover none of
them has cyclopentadienyl groups. We thus considered that com-
pound 2 might be an interesting precursor for new triiron hydride
phosphinidene complexes and clusters via its decarbonylation
reactions.
In this paper we report our studies to develop an efficient
synthetic route to the triiron phosphinidene complex 2 and a
detailed analysis of its photochemical decarbonylation reactions.
As it will be shown, this expectedly leads to reversible formation
of new metalꢀmetal bonds, but some unexpected processes are
also observed, such as the photochemically induced reductive
elimination of a PꢀH bond to yield a phosphide-bridged
(instead of phosphinidene-bridged) product.
We have recently shown that the phosphinidene complexes
[Fe2Cp2(μ-PR)(μ-CO)(CO)2] (Cp = η5-C5H5; R = Cy, Ph,
t
2,4,6-C6H2Me3, 2,4,6-C6H2 Bu3) can be readily prepared in high
yields through a two-step procedure starting from the corresponding
phosphine complexes [Fe2Cp2(μ-CO)2(CO)(PH2R)].1 During
the corresponding experimental studies we found that preparation
of the phenylphosphine complex [Fe2Cp2(μ-CO)2(CO)(PH2Ph)]
(1), which is made by reacting [Fe2Cp2(μ-CO)2(CO)(NCMe)]
with PH2Ph in dichloromethane, occasionally yielded small amounts
of a minor product. This product could be eventually isolated and
fully identified as the hydride- and phosphinidene-bridged complex
[Fe3Cp3(μ-H)(μ3-PPh)(CO)4] (2, Chart 1). Formation of 2 is
surprising if we consider the mild conditions under which this takes
place (room temperature, dichloromethane solution), because two
PꢀH bonds in the coordinated phosphine have to be cleaved along
the way. Moreover, we noticed that because this 52-electron complex
bears a FeCp(CO)2 fragment, it should be therefore susceptible to
decarbonylation to yield clusters with a phosphinidene ligand
bridging three iron centers. Indeed, we have shown recently that
related heterometallic species derived from the PCy complex
[Fe2Cp2(μ-PCy)(μ-CO)(CO)2] can be decarbonylated photoche-
mically, thus triggering formation of new metalꢀmetal bonds to give
new heterometallic clusters.2 This synthetic strategy constitutes a
new and rational approach to prepare μ3-PR clusters to be added to
other methods previously reported in the literature, often less
selective.2,3 Currently, the number of triiron phosphinidene
’ RESULTS AND DISCUSSION
Thermal and Photochemical Attempts To Prepare Com-
pound 2. We have now found that although the phosphine
complex 1 is itself prepared from the acetonitrile adduct
[Fe2Cp2(μ-CO)2(CO)(NCMe)] and PH2Ph, it is still able to
react slowly with a second molecule of this adduct at room
temperature to give almost quantitatively the triiron phosphini-
dene derivative [Fe3Cp3(μ-H)(μ3-PPh)(CO)4] (2). Inspection
Received: July 13, 2011
Published: October 07, 2011
r
2011 American Chemical Society
10937
dx.doi.org/10.1021/ic201491v Inorg. Chem. 2011, 50, 10937–10948
|

Inorganic Chemistry
ARTICLE
Chart 1
Table 1. Selected Bond Lengths (Å) and Angles (deg) for cis,
anti-2
Fe1ꢀFe2
Fe1ꢀP1
Fe2ꢀP1
Fe3ꢀP1
Fe1ꢀC1
Fe2ꢀC2
Fe3ꢀC3
Fe3ꢀC4
Fe1ꢀH1
Fe2ꢀH1
2.7217(6)
2.2181(8)
2.2174(8)
2.3363(8)
1.734(3)
1.740(3)
1.758(3)
1.765(3)
1.66(4)
Fe1ꢀP1ꢀFe2
Fe1ꢀP1ꢀFe3
Fe2ꢀP1ꢀFe3
C1ꢀFe1ꢀFe2
C2ꢀFe2ꢀFe1
C3ꢀFe3ꢀC4
C1ꢀFe1ꢀP1
C2ꢀFe2ꢀP1
C3ꢀFe3ꢀP1
C4ꢀFe3ꢀP1
H1ꢀFe1ꢀP1
H1ꢀFe2ꢀP1
75.7(1)
124.1(1)
124.4(1)
95.5(1)
96.5(1)
95.0(1)
91.6(1)
95.7(1)
91.2(1)
87.9(1)
1.68(4)
87(1)
86(1)
contrast with the behavior of the binuclear hydrideꢀphosphide
complex [Fe2Cp2(μ-H)(μ-PPh2)(CO)2], which has been shown
to undergo photochemically induced trans to cis isomerization.7 In
the case of 2, decarbonylation takes place rapidly under these
conditions instead, as it will be discussed below.
To obtain additional information on the elemental steps leading
to 2, we irradiated a mixture of [Fe2Cp2(CO)4] and 1 with
visibleꢀUV light in toluene at 288 K and using Pyrex glassware.
Surprisingly, this only led to formation of small amounts of 2.
Instead, the major products were the binuclear hydride [Fe2Cp2-
(μ-H)(μ-PHPh)(CO)2] (3a) and the trinuclear complex trans-
[Fe3Cp3(μ-PHPh)(μ-CO)2(CO)3] (4). The hydride 3a follows
from decarbonylation of compound 1 (eq 2), and it can be
more conveniently prepared from [Fe2Cp2(CO)4] and PH2Ph.
On the other hand, formation of compound 4 can be described
through eq 3, and the elemental steps there involved will be
discussed later on.
Figure 1. ORTEP diagram (30% probability) of compound cis,anti-2
with H atoms (except the hydride ligand) omitted for clarity.
½Fe2Cp2ðμ-COÞ2ðCOÞðPH2PhÞꢁð1Þ
f ½Fe2Cp2ðμ-HÞðμ-PHPhÞðCOÞ2ꢁð3aÞ þ CO
ð2Þ
(by means of IR spectroscopy) of the reaction mixture allowed
detection of significant amounts of the mononuclear hydride
[FeCpH(CO)2], but this byproduct could not be isolated from
the mixture, presumably because of its easy degradation to
[Fe2Cp2(CO)4] during the workup.6 Thus, formation of 2 seems
to follow the stoichiometric balance implied by eq 1, as will be
discussed later on.
½Fe2Cp2ðμ-COÞ2ðCOÞðPH2PhÞꢁð1Þ
þ ½Fe2Cp2ðCOÞ4ꢁ f ½FeCpHðCOÞ ꢁ
þ ½Fe3Cp ðμ-PHPhÞðμ-COÞ ðCOÞ 2ꢁð4Þ
ð3Þ
3
2
3
Structural Characterization of Compound 2. The structure
of 2 was confirmed through a single-crystal study of the major cis,
anti isomer (Figure 1 and Table 1). The molecule displays two
FeCp(CO) fragments symmetrically bridged by hydride and PPh
ligands (FeꢀH ca. 1.67 Å and FeꢀP ca. 2.22 Å), which define
an almost flat Fe2HP plane (angle between Fe2H and Fe2P
planes ca. 15°), with the CO ligands placed almost parallel
to each other and perpendicular to the intermetallic vector
(FeꢀFeꢀC angles ca. 96°). The intermetallic distance of
2.7217(6) Å is consistent with the formulation of a single,
H-bridged FeꢀFe bond, and it is comparable to the values of ca.
2.70 Å measured for the binuclear hydrideꢀphosphide complexes
[Fe2Cp2(μ-H)(μ-PPh2)(CO)2]8 and [Fe2Cp2(μ-H)(μ-PHMes)-
(CO)2].9 The phosphinidene ligand is also attached to a third
metal fragment (FeCp(CO)2), but the connection with this
fragment is weaker, as deduced from the longer Fe3ꢀP separation
(2.3363(8) Å). This length is comparable to those measured in
the trimetallaphosphonium cation [{FeCp(CO)2}3(μ3-PH)]+
(ca. 2.33 Å)10 and certainly longer than the values of ca. 2.13ꢀ2.21 Å
½Fe2Cp2ðμ-COÞ2ðCOÞðPH2PhÞꢁð1Þ
þ ½Fe2Cp2ðμ-COÞ2ðCOÞðNCMeÞꢁ f ½FeCpHðCOÞ2ꢁ
þ ½Fe3Cp3ðμ-HÞðμ3-PPhÞðCOÞ4ꢁð2Þ þ NCMe
ð1Þ
We should note that there are three different isomers for
complex 2 (Chart 1), depending on the relative arrangement of
the FeCp(CO) fragments (cis and trans) and that of the latter Cp
ligands with respect to the phenyl group on phosphorus (syn and
anti). However, only two of these isomers are actually formed in
the above reaction, these being identified as the cis,anti and trans
isomers, with a relative ratio of ca. 11/2, according to the 31P{1H}
NMR spectra of the crude reaction mixture. The third possible
isomer (cis,syn) is notformedinthis reaction, most likely because it
would be the most disfavored one on steric grounds. We finally
note that the cis and trans isomers of 2 do not interconvert in
solution, even under irradiation with visibleꢀUV light. This is in
10938
dx.doi.org/10.1021/ic201491v |Inorg. Chem. 2011, 50, 10937–10948

Inorganic Chemistry
ARTICLE
Table 2. Selected IR and NMR Data for New Compounds
c
δH (JHP
)
compound
ν(CO)a
δP
μ-H
PꢀH
b
[Fe3Cp3(μ-H)(μ3-PPh)(CO)4] (trans-2)
[Fe3Cp3(μ-H)(μ3-PPh)(CO)4] (cis,anti-2)
[Fe2Cp2(μ-H)(μ-PHPh)(CO)2] (trans-3a)
[Fe2Cp2(μ-H)(μ-PHPh)(CO)2] (cis,anti-3a)
[Fe2Cp2(μ-H)(μ-PHPh)(CO)2] (cis,syn-3a)
[Fe2Cp2(μ-H)(μ-PHCy)(CO)2] (trans-3b)
[Fe2Cp2(μ-H)(μ-PHCy)(CO)2] (cis-3b)
[Fe3Cp3(μ-PHPh)(μ-CO)2(CO)3] (4)
[Fe3Cp3(μ-H)(μ3-PPh)(μ-CO)(CO)2] (trans-5)
[Fe3Cp3(μ-H)(μ3-PPh)(μ-CO)(CO)2] (cis-5)
[Fe3Cp3(μ-H)(μ3-PPh)(μ-CO)2] (6)
2010 (s), 1964 (s), 1918 (m, sh), 1905 (vs)d
2005 (m), 1956 (s), 1938 (vs), 1899 (w)
1913
246.1
258.7e
132.3e,f
133.1e
137.6e
166.0
174.3
32.9
ꢀ17.70 (33)
ꢀ18.51 (34)e
ꢀ19.24 (43)g
ꢀ19.67 (43)e
ꢀ20.25 (45)e
6.00 (341)g
5.26 (360)e
6.91 (321)e
4.86 (316)e,g
5.74 (300)e,g
3.19 (298)
1949 (vs), 1910 (w)
1908
ꢀ19.15 (40)e,g
ꢀ20.23 (43)e,g
1942 (vs), 1902 (w)
2023 (s), 1971 (s), 1924 (s), 1762 (w), 1721 (vs)
1939 (m), 1921 (vs), 1787 (m)d
1973 (vs), 1949 (m), 1787 (m)d
1769 (vs), 1727 (w)
481.4
491.3
506.5f
194.0f
ꢀ26.37 (41)
ꢀ25.63 (40)
ꢀ34.82 (45)g
[Fe3Cp3(μ-PHPh)(μ3-CO)(μ-CO)2] (7)
1777 (vs), 1738 (m), 1644 (s)d
6.66 (350)g
a Recorded in dichloromethane solution, with CꢀO stretching bands (ν(CO)) in cmꢀ1. b Recorded in CD2Cl2 solutions at 290 K and 121.52 MHz
unless otherwise stated, with δ in ppm relative to external 85% aqueous H3PO4. c Recorded in CD2Cl2 solutions at 290 K and 300.13 MHz unless
e
otherwise stated, with δ in ppm relative to internal SiMe4 and PꢀH coupling constants (JHP) in Hertz. d In toluene solution. In CDCl3 solution.
f Recorded at 162.00 MHz. g Recorded at 400.13 MHz.
for different complexes of the type [Fe2Cp2(μ-CO)2(CO)(L)]
having classical two-electron donor phosphines and phos-
phites (L)7a,11 and hence dative PfFe bonds. This is consistent
with the prediction, based on simple electron counts, that the
phosphinidene ligand in this molecule should formally provide the
Fe2(μ-H) center with three electrons (PR2-like coordination) and
the FeCp(CO)2 center with just one electron. This situation is
comparable to that recently found by us in the photochemically
generated complexes [MFe2Cp2(μ3-PCy)(μ-CO)(CO)n] having
an MFe(μ-CO) core and a dangling FeCp(CO)2 group (M = Cr,
Fe)2 and thus confirms a general trend in these complexes whereby
MꢀP bonds to 17-electron metal fragments are significantly longer
than MꢀP bonds to 16-electron metal fragments.
trans gives rise to three different Cp and four different CO
resonances. In contrast, the presence of a symmetry plane
relating the FeCp(CO) fragments in the isomer cis,anti leads
to the observation of just two sets of CO and Cp resonances. We
finally note that the bridging hydride ligand in both isomers gives
rise to a doublet resonance (JPH ca. 33 Hz) at ca. ꢀ18 ppm, these
being spectroscopic parameters comparable to those measured
for compounds of the type [Fe2Cp2(μ-H)(PRR0)(CO)L] pre-
viously reported.7
Synthesis and Structural Characterization of Compounds
3. Although the hydride 3a is obtained as a byproduct in the
photochemical reaction between 1 and [Fe2Cp2(CO)4], it can be
more conveniently prepared by reacting [Fe2Cp2(CO)4] and
PH2Ph in refluxing toluene. A similar reaction takes place with
PH2Cy to give the corresponding hydride derivative [Fe2Cp2-
(μ-H)(μ-PHCy)(CO)2] (3b) in high yield. This synthetic approach
has been previously used by us and others to prepare related
complexes of the type [Fe2Cp2(μ-H)(μ-PR2)(CO)2],7,13 and we
note that intermediates comparable to 1 were also detected by IR
spectroscopy in the course of the formation of compounds 3,
suggesting the operation of a reaction pathway similar to that
proposed for reaction of [Fe2Cp2(CO)4] with PHPh2.7
As discussed for 2, there are also three possible isomers for
compounds 3, depending on the relative disposition of the CO
ligands (cis and trans) and that of the Cp groups with respect to
the R substituent at the P atom (syn and anti). Actually all the
possible isomers are formed for the phenyl compound 3a,
whereas only two of them could be detected for 3b, these being
the trans isomer and one of the cis isomers, which we assume it to
be the cis,anti isomer, less disfavored on steric grounds (Chart 2).
We note that the cis and trans isomers of these compounds do
not interconvert in solution at room temperature and that they
could be isolated separately in a conventional way. In contrast,
the cis,syn and cis,anti isomers of 3a could not be separated
chromatographically, even if no interconversion between them
seems to take place in solution either.
The spectroscopic data for the isomer cis,anti-2 in solution are
in good agreement with the solid-state structure just described
(Table 2). Its IR spectrum in the CꢀO stretching region can be
described as derived from the superimposition of the bands
originated by relatively independent FeCp(CO)2 and Fe2(CO)2
oscillators.12 This allows assignment of the bands at 1938 (vs)
and 1899 (w) cmꢀ1 to the dimetal oscillator, with their relative
intensities indicating an almost parallel arrangement of the
Fe2(CO)2 carbonyls, as found in the crystal. The IR spectrum
of trans-2 can be interpreted along the same lines, but the relative
intensities of the two less energetic bands (1918 (m, sh) and
1905 (vs) cmꢀ1) now clearly denote an antiparallel arrangement
of carbonyls in the Fe2(CO)2 fragment (Chart 1).
The 31P{1H} NMR spectra of these compounds exhibit reso-
nances at ca. 250 ppm, a position comparable to that of the
photochemically generated compound [Fe3Cp2(μ3-PCy)(μ-CO)-
(CO)6] having a dangling FeCp(CO)2 group (283.7 ppm) but
substantially shielded with respect to that of the isomer [Fe3Cp2-
(μ3-PCy)(μ-CO)(CO)6] having a dangling Fe(CO)4 group
(421.5 ppm).2 Thus, it seems that the chemical shift of the μ3-PR
ligand might be sensitive to the fine details of the electron distribu-
tion among the three metal atoms in these 52-electron complexes
with just one metalꢀmetal bond.
The different symmetry of the isomers of 2 is clearly reflected
in their 1H and 13C NMR spectra. Thus, the asymmetric isomer
The IR spectra of compounds 3 in solution show a band at ca.
1910 cmꢀ1 for the trans isomers and two bands at ca. 1945 (vs)
10939
dx.doi.org/10.1021/ic201491v |Inorg. Chem. 2011, 50, 10937–10948

Inorganic Chemistry
ARTICLE
Chart 2
Scheme 1. Possible Pathway for Formation of 2
Chart 3
known to decompose progressively at the same temperature and
solvent to give the tetracarbonyl [Fe2Cp2(CO)4] as the only
carbonylic product. Therefore, we might guess that, to some extent,
this adduct undergoes acetonitrile dissociation even in the absence
of a good donor to give unstable intermediates of formula
[Fe2Cp2(CO)3]. The latter species are actually known (if unstable)
transient species that can be photochemically generated from
[Fe2Cp2(CO)4],16,17 and they have been experimentally shown
to undergo oxidative addition of the HꢀSn bond of HSnBu3 to
eventually give a mixture of [FeCp(CO)2H] and [FeCp(CO)2-
(SnBu3)].18 In the same line, we propose that compound 1 would
easily add to the [Fe2Cp2(CO)3] intermediate resulting from
acetonitrile dissociation, this triggering the oxidative addition of
one of the PꢀH bonds of the coordinated phosphine, to give the
hydrideꢀphosphide intermediate A (Scheme 1). Actually, this is
not very different from the mechanism proposed previously for
formation of [Fe2Cp2(μ-H)(μ-PPh2)(CO)2] upon decarbonyla-
tion of [Fe2Cp2(μ-CO)2(CO)(PHPh2)].7b Intermediate A would
then rapidly degrade by releasing [FeCpH(CO)2] (observed
spectroscopically), this leaving a trimetallic intermediate B having
a 16-electron center that would trigger a second oxidative addition
of the remaining PꢀH bond to give the hydrideꢀphosphinidene
intermediate C, eventually yielding 2 after some rearrangement.
The transformation of a μ-PHR ligand into a hydrideꢀphosphinidene
derivative is a rare event, but we can quote at least a few examples
and 1905 (w) cmꢀ1 for the cis isomers, as expected for this type
of M2(CO)2 oscillators.12 The trans isomers are also identified by
the chemical inequivalence of their Cp ligands, and they give rise
to 31P NMR resonances, at 132.3 (3a) and 166.0 (3b) ppm,
comparable to those measured for related PHR-bridged
complexes.13 The cis isomers have equivalent Cp groups and
give rise to 31P resonances slightly more deshielded. We finally
note that, as expected for molecules having a Fe2(μ-H)(μ-PHR)
central core, all these isomers display strongly coupled HꢀP
1
resonances (δH ca. 5ꢀ7 ppm, with JPH ca. 300ꢀ360 Hz) and
highly shielded hydride resonances (ca. ꢀ20 ppm, with 2JPH ca.
43 Hz).
Structural Characterization of Compound 4. The structure
proposed for 4 can be derived from that of 1 by just replacing one
of the P-bound H atoms with a FeCp(CO)2 group, so that the
terminal phosphine ligand in 1 becomes a bridging phosphide
group in 4 (Chart 3). Since there is no metalꢀmetal bond
connecting the corresponding iron atoms, it is expected that the
P atom becomes considerably shielded,14 and this is consistent
with the very low chemical shift of the complex (δP 32.9 ppm).
The IR spectrum of 4 can be interpreted considering relatively
independent oscillators: the two bands at lower frequencies are
not very different from the corresponding bands in 1 and are
assigned to the CꢀO stretches of a flat M2(μ-CO)2 oscillator, as
denoted by their relative intensities; the two strong bands at higher
frequencies are comparable to the corresponding bands in 2 and
therefore can be attributed to the FeCp(CO)2 moiety; finally, the
band at 1924 cmꢀ1 must necessarily correspond to the CꢀO
stretch of the FeCp(CO) fragment, and its relatively low frequency
must be taken as an indication of a transoid arrangement of the P
and CO ligands around the central Fe2(μ-CO)2 core.7,15
19
taking place at Os3 or FeCoRu20 centers to give μ3-PR
derivatives. More recently, we have shown that two metal centers
are enough to promote this transformation under mild condi-
tions if electronic and coordinative unsaturation are present, as
proved by the formation at room temperature of the cation
[Mo2Cp2(μ-H)(μ-PR)(CO)4]+ upon proton-induced dehydro-
genation of [Mo2Cp2(μ-H)(μ-PHR)(CO)4], when R is the
t
bulky substituent 2,4,6-C6H2 Bu3.21
As stated above, photolysis of [Fe2Cp2(CO)4] in the presence
of 1 led only to formation of small amounts of 2, while the major
products were the binuclear hydride complex 3a and the tri-
nuclear pentacarbonyl 4. The hydride 3a obviously follows from
the photochemically induced decarbonylation of 1, a process not
very different from the thermal formation of this complex (vide
supra). To explain the formation of 4 we must take into account
that photolysis of [Fe2Cp2(CO)4] actually generates large
amounts of the 17-electron radical [FeCp(CO)2] following
Reaction Pathways in Formation of Compounds 2 and 4.
As discussed above, compound 2 is obtained almost quantitatively
by reacting 1 with the adduct [Fe2Cp2(μ-CO)2(CO)(NCMe)] in
dichloromethane at room temperature. The latter complex is
10940
dx.doi.org/10.1021/ic201491v |Inorg. Chem. 2011, 50, 10937–10948

Inorganic Chemistry
ARTICLE
Scheme 2. Possible Pathway for the Formation of 4
Chart 4
mixture of the corresponding cis and trans isomers (Chart 4)
with their ratio being ca. 1/10 in C6D6 and 1/5 in CD2Cl2. If,
however, a flow of N2 is gently bubbled through solution under
the same conditions (to facilitate removal of CO) then the
reaction does not stop at the tricarbonyl step but proceeds with
evolution of a second molecule of CO to eventually give the
dicarbonyl cluster [Fe3Cp3(μ-H)(μ3-PPh)(μ-CO)2] (6) after 2 h.
The same result is achieved (but in shorter reaction times) if
quartz glassware is used in the latter reaction, and a separate
experiment confirmed that 5 is quantitatively transformed into 6
under any of the above photolytic conditions. The latter process
is reversible, and compound 5 can be regenerated in 30 min by
reaction of 6 with CO (1 atm) in toluene at 293 K. We finally
note that the same results, and in comparable reaction times,
were obtained in the above photolytic experiments when using
the pentacarbonyl 4 instead of the tetracarbonyl 2, although no
detectable amounts of 2 were identified spectroscopically in the
course of the corresponding reactions. This indicates that
compounds 2 and 4 are decarbonylated to give 5 through
independent reaction pathways, a matter to be discussed later on.
Photolysis at 288 K of either 2 or 5 using quartz glassware and
without the N2 purge leads to an inseparable 2/1 mixture of the
dicarbonyl 6 and the tricarbonyl phosphide-bridged cluster
[Fe3Cp3(μ-PHPh)(μ3-CO)(μ-CO)2] (7). Fortunately, the
new complex was inert toward CO, and the 6 + 7 mixture could
be converted into a mixture of 5 and 7 by reaction with CO,
allowing isolation of the latter. We note that photolysis of the
pentacarbonyl 4 using quartz glassware and no N2 purge also
yielded a mixture of compounds 6 and 7 in 30 min at 288 K. Since
5 and 7 are isomeric species, we also attempted to obtain
selectively 7 from 5 with no success. For instance, irradiation
of 5 using quartz glassware and a CO atmosphere (to prevent
formation of 6) led to no neat transformation after 3 h. Likewise,
no transformation of 5 could be detected in refluxing toluene
after 1 h.
Scheme 3. Photolytic Reactions of Compounds 2 and 4
homolysis of the FeꢀFe bond, a process competing with the
CO dissociation pathways leading to tricarbonyl intermediates
[Fe2Cp2(CO)3].16ꢀ18 Then we might conceive that the iron radical
would initiate cleavage of a HꢀP bond by an atom transfer
mechanism. This is a well-documented reaction of 17-electron
organometallic complexes MLn toward different XꢀY bonds that
can lead to both XꢀMLn and YꢀMLn derivatives.18,22 In our case,
the corresponding derivatives would be [FeCpH(CO)2] and 4, and
a radical intermediate D would be involved (Scheme 2). However,
we must note here that we found no examples of related reactions
involving the PꢀH bond of a coordinated phosphine.
Photochemical Decarbonylation of Compounds 2 and 4.
Removal of CO from 2 or 4 could be accomplished photoche-
mically, triggering stepwise and reversible formation of new
FeꢀFe bonds to yield different clusters depending on the precise
experimental conditions (Scheme 3). Thus, irradiation of toluene
solutions of the tetracarbonyl 2 with visibleꢀUV light at 288 K
and using Pyrex glassware gave the tricarbonyl derivative
[Fe3Cp3(μ-H)(μ3-PPh)(μ-CO)(CO)2] (5) in quantitative
yields. This compound exists in solution as an equilibrium
Structural Characterization of Compound 5. The structure
of the trans isomer of 5 is built from three metal fragments
bridged by a phosphinidene ligand and defining a V-shaped metal
core (Figure 2 and Table 3). The external FeCp(CO) groups are
arranged in a mutually transoid disposition, and the central FeCp
moiety is connected to the other two by intermetallic bonds with
carbonyl and hydride bridges over them, the latter being posi-
tioned opposite the PPh ligand, with respect to the intermetallic
10941
dx.doi.org/10.1021/ic201491v |Inorg. Chem. 2011, 50, 10937–10948

Inorganic Chemistry
ARTICLE
Figure 3. ORTEP diagram (30% probability) for one of the molecules
in the asymmetric unit of 6, with H atoms and the Ph group (except the
C1 atom) omitted for clarity.
Figure 2. ORTEP diagram (30% probability) of trans-5 with H atoms
(except the hydride ligand) omitted for clarity.
Table 4. Selected Bond Lengths (Å) and Angles (deg) for 6
Table 3. Selected Bond Lengths (Å) and Angles (deg) for
Fe1ꢀFe2
Fe1ꢀFe3
Fe2ꢀFe3
Fe1ꢀP1
Fe2ꢀP1
Fe3ꢀP1
Fe1ꢀC1
Fe1ꢀC2
Fe2ꢀC2
Fe3ꢀC1
2.629(2)
2.591(2)
2.718(2)
2.144(3)
2.100(3)
2.122(3)
1.99(1)
1.97(1)
1.87(1)
1.89(1)
Fe1ꢀFe2ꢀFe3
Fe1ꢀFe3ꢀFe2
Fe2ꢀFe1ꢀFe3
C1ꢀFe1ꢀC2
Fe1ꢀP1ꢀFe2
Fe1ꢀP1ꢀFe3
Fe2ꢀP1ꢀFe3
C1ꢀFe1ꢀP1
C1ꢀFe3ꢀP1
C2ꢀFe1ꢀP1
C2ꢀFe2ꢀP1
57.9(1)
59.3(1)
62.8(1)
79.4(4)
76.5(1)
74.8(1)
80.2(1)
98.2(3)
102.3(4)
96.0(4)
100.8(3)
trans-5
Fe1ꢀFe2
Fe2ꢀFe3
Fe1ꢀP1
Fe2ꢀP1
Fe3ꢀP1
Fe1ꢀC1
Fe1ꢀC2
Fe2ꢀC2
Fe3ꢀC3
Fe2ꢀH1
Fe3ꢀH1
2.6177(3)
2.7611(4)
2.2010(5)
2.1593(5)
2.1787(5)
1.739(2)
2.079(2)
1.840(2)
1.749(2)
1.66(3)
Fe1ꢀFe2ꢀFe3
Fe1ꢀP1ꢀFe2
Fe1ꢀP1ꢀFe3
Fe2ꢀP1ꢀFe3
Fe1ꢀP1ꢀC19
Fe2ꢀP1ꢀC19
Fe3ꢀP1ꢀC19
C1ꢀFe1ꢀP1
C2ꢀFe1ꢀP1
C2ꢀFe2ꢀP1
C3ꢀFe3ꢀP1
89.9(1)
73.8(1)
120.5(1)
79.1(7)
116.1(1)
129.6(1)
122.1(1)
92.6(1)
88.5(1)
96.3(1)
91.9(1)
(m) and 1921 (vs) cmꢀ1 and is therefore identified as the trans
isomer, while these relative intensities are reversed in the minor
species (1973 (vs) and 1949 (m) cmꢀ1), which is then identified
as the cis isomer.12 Both isomers give rise to strongly deshielded
31P NMR resonances (ca. 490 ppm), some 240 ppm above that
of 2, and to strongly shielded resonances for the bridging hydride
(at ca. ꢀ26 ppm, with JHP ca. 40 Hz), in agreement with the
retention of the bridging coordination of these ligands in solution.
Structural Characterization of Compound 6. Although the
quality of our diffraction data for 6 was not optimal and the
hydride ligand could not be located, all other structural features
of the molecule are well defined (Figure 3 and Table 4). The
cluster is built from three FeCp fragments defining a triangle
bridged by a μ3-PPh ligand displaying quite short PꢀFe lengths
(ca. 2.10ꢀ2.14 Å). The shortest FeꢀFe bonds (ca. 2.60 Å) are
bridged by a carbonyl ligand each, and the third edge of the
triangle is significantly longer (2.718(2) Å), presumably because
of the presence of the bridging hydride over that edge. That
intermetallic separation is now comparable to the corresponding
ones in the isoelectronic complexes [Fe3(μ-H)2(μ3-PR)(CO)9].4a,c
As found for 5, the bridging carbonyls are placed opposite the
phosphinidene ligand, relative to the intermetallic plane, a feature
also observed for the isoelectronic Fe2Mn cluster [Fe2MnCp2Cp0-
(μ3-PCy)(μ-CO)3] recently reported by us.2
1.61(3)
plane. The respective FeꢀFe lengths of 2.6177(3) and 2.7611(4)
Å are within the range of values previously measured for related
48- and 50-electron Fe3(μ3-PR) clusters,4,5 with the tricentric
Fe2(μ-H) bond being expectedly longer. The observed differ-
ence in the FeꢀFe lengths of 5 (ca. 0.15 Å) is, however, larger
than anticipated (cf. 0.03 Å for [Fe3(μ-H)2(μ3-PPh)(CO)9]4a or
[Fe3(μ-H)(μ-PMeiPr)(μ3-PPh)(CO)9],23 with the latter being
the only other 50-electron triiron hydride phosphinidene com-
plex structurally characterized previously). In the case of 5, the
large difference in the FeꢀFe lengths might be attributed to the
favorable combination of two opposite effects: a lengthening in
one bond, due to the presence of the hydride ligand, and a
contraction on the other bond, due to the presence of the
bridging carbonyl. The phosphinidene ligand in 5 displays
comparable and short FeꢀP lengths (ca. 2.16ꢀ2.20 Å), with
an environment around P quite distorted from the tetrahedral
geometry, with the P, C19, Fe1, and Fe3 atoms almost in the
same plane (Σ XꢀPꢀY ca. 359°; X, Y = C19, Fe1, Fe3).
This environment seems to be characteristic of 50-electron
Fe3(μ3-PR) and related clusters.5,3
Although the trans and cis isomers of 5 can be separated by
chromatographic techniques, they quickly isomerize (ca. 10 min
at 293 K) to reach a solvent-dependent equilibrium. The IR
spectra of the mixture of isomers in the terminal CꢀO stretching
region allow us to identify the relative disposition of the FeCp-
(CO) in these molecules: the major isomer gives bands at 1939
The presence of a bridging hydride ligand in 6 is clearly
denoted by the appearance of a strongly shielded resonance in its
1H NMR spectrum (δH ꢀ35.82 ppm, JHP = 45 Hz), while its IR
spectrum in solution exhibits two CꢀO stretching bands in the
10942
dx.doi.org/10.1021/ic201491v |Inorg. Chem. 2011, 50, 10937–10948

Inorganic Chemistry
ARTICLE
Scheme 4. Possible Pathway for the Formation of 5 from 2
[Fe = FeCp(CO)]
and the third edge of the triangle (ca. 2.62 Å) is bridged by a
phenylphosphide ligand, with its Ph group pointing away from
the intermetallic plane, presumably to minimize the steric
repulsions with the Cp ligands. As found for 6, the edge-bridging
ligands are placed opposite the face-bridging ligand, relative to
the intermetallic plane. The FeꢀC bonds of that μ3-CO ligand
(av. 1.99 Å) are somewhat longer than those of the edge-bridging
ones (av. 1.92 A) as expected, and have comparable lengths to
those measured in the tetrahedral cluster [Fe4Cp4(μ3-CO)4].24
In the phosphinidene-bridged complex [Fe3(μ3-PtBu)(μ3-CO)-
(CO)6(η6-C6H5Me)] (a related molecule also having a 48-
electron count) the FeꢀCO lengths of the face-bridging ligand
are much longer (ca. 2.21 Å),25 a difference that can be attributed
to the lower oxidation state of iron in that case.
The spectroscopic data in solution for 7 are consistent with the
retention of its solid state structure. Its IR spectrum exhibits a
band at very low frequency (1644 cmꢀ1) assigned to the CꢀO
stretch of its face-bridging carbonyl, while the bands at 1777 (vs)
and 1738 (m) cmꢀ1 denote the presence of two edge-bridging
carbonyls in a cisoid arrangement. These assignments were con-
firmed by the presence of resonances at 297.0 and 272.7 ppm (1:2
intensity) in the corresponding 13C{1H} NMR spectrum with the
face-bridging carbonyl giving the most deshielded resonance as
expected.26 The 31P{1H} NMR spectrum of 7 exhibits a relatively
shielded resonance (δP 194.0 ppm) with a shift comparable to that
of the cation [Fe2Cp2(μ-PPhH)(μ-CO)(CO)2]+,1a,c which is
indicative of the retention of the PHPh ligand in solution, also
denoted by the appearance of a strongly coupled PH resonance at
6.66 ppm (1JHP = 350 Hz) in the 1H NMR spectrum.
Reaction Pathways in Formation of Compounds 5ꢀ7.
Formation of the tricarbonyl 5 from the tetracarbonyl 2 most
likely involves loss of a terminal CO from the unique dicarbonyl
fragment in the molecule (FeCp(CO)2) to give the intermediate
E, with a 16-electron FeCp(CO) center (Scheme 4). This would
trigger formation of a new FeꢀFe bond with a simultaneous
rearrangement of a terminal CO to a bridging position to give
compound 5. The process would be completely parallel to that
recently proposed to explain formation of the isoelectronic com-
plex [Fe2MnCp2Cp0(μ3-PCy)(μ-CO)2(CO)2] from its 52-elec-
tron precursor [Fe2MnCp2Cp0(μ3-PCy)(μ-CO)(CO)4].2
In contrast, formation of 5 from 4 has to be more complex. It is
sensible to assume that the reaction also involves first loss of a
CO molecule from the unique FeCp(CO)2 fragment of the
complex (Scheme 5). This would give the same intermediate B
that is presumably involved in thermal formation of 2
(Scheme 1). Since we have not detected the intermediacy of 2
in the way from 4 to 5, we must assume that, under photolytic
conditions, intermediate B does not decay thermally to 2 but
instead undergoes a photochemical rearrangement, perhaps with
formation of a new FeꢀFe bond to give a 50-electron inter-
mediate E with a new FeCp(CO)2 fragment. The photochemical
decarbonylation of the latter fragment would now yield an
Figure 4. ORTEP diagram (30% probability) of compound 7 with H
atoms (except the PꢀH) and Ph group (except the C1 atom) omitted for
clarity.
Table 5. Selected Bond Lengths (Å) and Angles (deg) for 7
Fe1ꢀFe2
Fe1ꢀFe3
Fe2ꢀFe3
Fe1ꢀC1
Fe1ꢀC2
Fe1ꢀC3
Fe2ꢀC1
Fe2ꢀC3
Fe3ꢀC2
Fe3ꢀC3
Fe2ꢀP1
Fe3ꢀP1
P1ꢀH1
2.5332(8)
2.5334(7)
2.6158(8)
1.914(4)
1.897(4)
1.937(4)
1.924(4)
2.022(4)
1.943(4)
2.009(4)
2.167(1)
2.147(1)
1.34(4)
Fe1ꢀFe2ꢀFe3
Fe1ꢀFe3ꢀFe2
Fe2ꢀFe1ꢀFe3
C1ꢀFe1ꢀC3
C1ꢀFe2ꢀC3
C2ꢀFe1ꢀC3
C2ꢀFe3ꢀC3
C3ꢀFe2ꢀP1
C3ꢀFe3ꢀP1
C2ꢀFe3ꢀP1
C1ꢀFe2ꢀP1
C1ꢀFe1ꢀC2
58.9(1)
58.9(1)
62.2(1)
100.6(2)
97.2(2)
100.6(2)
96.5(2)
101.6(1)
102.8(1)
86.8(1)
85.6(1)
89.7(2)
region of the bridging ligands, with relative intensities indicative
of a cisoid arrangement, as found in the crystal. It should be
noticed the progressive shielding of the μ2-H ligand as we
proceed from 2 (ca. ꢀ20 ppm) to 5 (ca. ꢀ25 ppm) and then
to 6 (ca. ꢀ35 ppm). This hardly can be attributed to significant
variations in the geometrical parameters within each Fe2(μ-H)
unit. Instead, we trust that these shielding differences actually
reflect the relative proximity of the third metal atom to the
hydride ligand. In 2, this metal center is too far away (ca. 4.74 Å)
to cause a significant shielding on the hydride nucleus, hence the
similitude of its chemical shift to those of the binuclear hydrides 3
(Table 2). In 5, the third metal atom is significantly closer to the
hydride ligand (3.51 Å), thus explaining the observed shielding of
ca. 5 ppm relative to the former compounds. In 6, it can be
estimated that this distance is reduced to ca. 3.2 Å, thus
explaining the large magnetic shielding of the hydride nucleus.
We finally note that the phosphinidene ligand in 6 gives rise to a
strongly deshielded 31P NMR resonance (506.5 ppm). This shift
is comparable to that of the 50 electron complex 5 but much
higher than those observed for 2 and related heterometallic 52-
electron PR-bridged complexes.2
Structural Characterization of Compound 7. This cluster is
built from three FeCp fragments defining an isosceles triangle
bridged by a μ3-CO ligand (Figure 4 and Table 5). The shortest
FeꢀFe edges (ca. 2.53 Å) are bridged by a carbonyl ligand each,
10943
dx.doi.org/10.1021/ic201491v |Inorg. Chem. 2011, 50, 10937–10948

Inorganic Chemistry
ARTICLE
Scheme 5. Possible Pathway for Formation of 5 from 4
consistent with the absence of 7 when a N2 purge is used in
the experiment (suppression of the H/7 step) and the absence of
6 and 7 when photolysis is carried out under a CO atmosphere
(suppression of the 5/G step).
We should stress that the reductive formation of a PꢀH bond
between μ-H and μ3-PPh ligands leading to the PHPh-bridged
cluster 7 is unexpected, especially under photolytic conditions
(this usually would trigger the opposite process, as discussed in
this work). Interestingly, we quote a couple of examples were reductive
formation of a PꢀH bond is favored for a hydrideꢀphosphinidene
substrate, both of them involving anionic complexes containing
pyramidal phosphinidene bridges (2-electron donors): this is the case
of the cluster [Os3(μ-H)(μ2-PPh)(CO)10]ꢀ 27 and the dimanganese
anion [Mn2(μ-H)(μ-PCy)(CO)8]ꢀ.28
’ CONCLUDING REMARKS
The room-temperature reaction of the phenylphosphine
complex [Fe2Cp2(μ-CO)2(CO)(PH2Ph)] with the acetonitrile
adduct [Fe2Cp2(μ-CO)2(CO)(NCMe)] triggers spontaneous
oxidative addition of two PꢀH bonds to eventually give the
hydrideꢀphosphinidene derivative [Fe3Cp3(μ-H)(μ3-PPh)(CO)4]
(2) in excellent yield, whereas its photochemical reaction with
[Fe2Cp2(CO)4], using visible light, induces cleavage of a single
PꢀH bond to give the phosphide derivative [Fe3Cp3(μ3-PHPh)-
(μ-CO)2(CO)3] (4). Photochemical decarbonylation of com-
pounds 2 and 4 (using visibleꢀUV light) can be tuned to give
either 50-electron or 48-electron phosphinidene clusters hav-
ing hydride ligands, following from stepwise formation of new
FeꢀFe bonds. A phosphide-bridged cluster can be also formed
under the right conditions, following from the unexpected
reductive elimination of a PꢀH bond between hydride and
μ3-phosphinidene ligands.
Scheme 6. Possible Pathway for Formation of Compounds 6
and 7 from 5 [Fe = FeCp]
’ EXPERIMENTAL SECTION
General Procedures and Starting Materials. All manipula-
tions 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.29 Petroleum
ether refers to that fraction distilling in the range 338ꢀ343 K.
Compounds 11a,c and [Fe2Cp2(μ-CO)2(CO)(NCMe)]30 were synthe-
sized according to literature procedures. All other reagents were
obtained from the usual commercial suppliers and used as received.
Photochemical experiments were performed using jacketed quartz or
Pyrex Schlenk tubes, cooled by tap water (ca. 288 K). A 400 W mercury
lamp placed ca. 1 cm away from the Schlenk tube was used for the
experiments with visibleꢀUV light. Chromatographic separations were
carried out using jacketed columns cooled 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 under
nitrogen with the appropriate amount of water to reach the activity
desired. IR CꢀO stretching frequencies were measured in solution and
are referred to as ν (CO). Nuclear magnetic resonance (NMR) spectra
were routinely recorded at 300.13 (1H), 121.52 (31P{1H}), and 75.48
MHz (13C{1H}) at 290 K in CD2Cl2 solutions unless otherwise stated.
Chemical shifts (δ) are given in ppm, relative to internal tetramethylsi-
lane (1H, 13C) or external 85% aqueous H3PO4 (31P). Coupling
constants (J) are given in Hertz.
unsaturated intermediate F with a 16-electron FeCp(CO) center
that would then trigger oxidative addition of the PꢀH bond of
the phosphide ligand to finally give the phosphinidene 5.
To explain formation of compounds 6 and 7, we assume that
decarbonylation of 5 should trigger formation of a new FeꢀFe
bond, thus yielding a 48-electron intermediate G, rendering
cluster 6 after a terminal to bridging rearrangement of the
remaining carbonyl (Scheme 6). These two steps should be
reversible, because 6 gives back 5 upon carbonylation. To
account for formation of 7, we must assume an alternative
evolution of G, only operative under irradiation, via reductive
elimination of μ-H and μ3-PPh ligands to give an unsaturated
μ2-PHPh intermediate H, the latter reacting rapidly with CO to
give the electron-precise cluster 7. This proposal is fully
Preparation of [Fe3Cp3(μ-H)(μ3-PPh)(CO)4] (2). A dichloromethane
solution (5 mL) of freshly prepared [Fe2Cp2(μ-CO)2(CO)(NCMe)]
(ca. 2.260 mmol) was added to a dichloromethane solution (5 mL) of
10944
dx.doi.org/10.1021/ic201491v |Inorg. Chem. 2011, 50, 10937–10948

Inorganic Chemistry
ARTICLE
compound 1 (0.500 g, 1.147 mmol), and the mixture was stirred for 16 h
to give a red solution. The solvent was then removed under vacuum, and
the residue was dissolved in a minimum amount of dichloromethane/
petroleum ether (1/3) and chromatographed on alumina (activity II) at
288 K. Elution with the same solvent mixture gave a red fraction that
gave, upon removal of solvents under vacuum, compound trans-2 as an
orange-reddish microcrystalline solid (0.067 g, 11%). Elution with
dichloromethane/petroleum ether (1/2) gave a second red fraction
giving analogously compound cis,anti-2 as a red microcrystalline solid
(0.569 g, 85%). X-ray-quality crystals of the latter compound were
grown by slow diffusion of a layer of petroleum ether into a dichlor-
omethane solution of the complex at 253 K. Data for compound trans-2.
Anal. Calcd for C25H21Fe3O4P: C, 51.42; H, 3.62. Found: C, 51.29; H,
3.48. 1H NMR: δ 7.91 (m, 2H, Ph), 7.25 (m, 2H, Ph), 7.15 (m, 1H, Ph),
visibleꢀUV light for 1 h in a Pyrex Schlenk tube refrigerated with tap
water while gently bubbling N2 through the solution to give a brown-
reddish solution. The solvent was then removed under vacuum, and the
residue was dissolved in a minimum amount of dichloromethane/
petroleum ether (1/2) and chromatographed on alumina (activity II)
at 253 K. Elution with the same solvent mixture gave three different
fractions: the first one gave, after removal of solvents, compound trans-
3a as an orange solid (0.009 g, 10%), the second one, also orange, gave
analogously compound cis-3a (0.023 g, 25%) as a mixture of the syn and anti
isomers in a ratio 1/2, whereas the third one (red) contained some
[Fe2Cp2(CO)4]. Finally, elution with dichloromethane/petroleum
ether (1/1) gave another red fraction yielding analogously compound cis-2
(0.013 g, 10%) and then a green-purple fraction giving trans-[Fe3Cp3-
(μ-PHPh)(μ-CO)2(CO)3] (4) as a purple solid (0.072 g, 52%). Data for
compound 4. Anal. Calcd for C26H21Fe3O5P: C, 51.03; H, 3.46. Found: C,
50.72; H, 3.18. 1HNMR:δ 7.40ꢀ7.28(m,5H,Ph),5.07(d,JHP =2,5H,Cp),
4.67 (s, 5H, Cp), 4.45 (d, JHP = 2, 5H, Cp), 3.19 (d, JHP = 298, 1H, PꢀH).
Preparation of [Fe3Cp3(μ-H)(μ3-PPh)(μ-CO)(CO)2] (5). A toluene
solution (4 mL) of compound cis-2 (0.050 g, 0.086 mmol) was irradiated
for 40 min with visibleꢀUV light in a Pyrex Schlenk tube refrigerated
with tap water while gently bubbling N2 through the solution to give a
brown mixture. The solvent was then removed under vacuum, and the
residue was dissolved in a minimum amount of dichloromethane/
petroleum ether (1/2) and chromatographed on alumina (activity
IV) at 288 K. Elution with the same solvent mixture gave a green-
brownish fraction containing trans-5, which isomerizes in solution
outside the column to reach the equilibrium ratio with the correspond-
ing isomer cis-5 in ca. 10 min. Elution with dichloromethane/petroleum
ether (1/1) gave a green fraction containing the isomer cis-5, which also
isomerizes outside the column to reach the equilibrium ratio with the
corresponding trans isomer. The fractions were mixed to give, after
removal of solvents, compound 5 as a brown solid (0.039 g, 81%). The
cis/trans equilibrium ratio in solution was measured (by NMR) to be ca.
1/10 in C6D6 and 1/5 in CD2Cl2. X-ray-quality crystals of trans-5 were
grown by slow diffusion of a layer of petroleum ether into a toluene
solution of the mixture of isomers at 253 K. Using the same experimental
procedure, irradiation of a toluene solution (5 mL) of trans-2 (0.050 g,
0.086 mmol) for 30 min also gave compound 5 after similar workup
(0.040 g, 84%). Anal. Calcd for C24H21Fe3O3P: C, 51.85; H, 3.81.
Found: C, 51.69; H, 3.68. Spectroscopic data for trans-5. 1H NMR: δ 8.44
(m, 2H, Ph), 7.75ꢀ7.57 (m, 3H, Ph), 4.57, 4.48, 4.36 (3s, 3 ꢂ 5H, Cp),
ꢀ26.37 (d, JHP = 41, 1H, μ-H). 1H NMR (400.13 MHz, C6D6): δ 8.49
(m, 2H, Ph), 7.43 (m, 2H, Ph), 7.33 (m, 1H, Ph), 4.42, 4.25, 4.08 (3s,
3 ꢂ 5H, Cp), ꢀ26.05 (d, JHP = 41, 1H, μ-H). 13C{1H} NMR (100.62
MHz): δ 260.5 (d, JCP = 6, μ-CO), 219.3, 214.9 (2d, JCP = 24, FeCO),
153.1 (d, JCP = 11, C1ꢀPh), 133.1 (d, JCP = 8, C2ꢀPh), 129.0 (s,
C4ꢀPh), 128.2 (d, JCP = 9, C3ꢀPh), 84.0, 83.9, 80.8 (3s, Cp). Spectro-
scopic data for cis-5. 1H NMR: δ 8.38 (m, 2H, Ph), 7.72 (m, 2H, Ph), 7.17
(m, 1H, Ph), 4.69, 4.34, 4.21 (3s, 3 ꢂ 5H, Cp), ꢀ25.63 (d, JHP = 40, 1H,
μ-H). 1H NMR (400.13 MHz, C6D6): δ 8.44 (m, 2H, Ph), 8.21 (m, 2H,
4.67 (d, JHP = 1, 5H, Cp), 4.64, 4.43 (2s, 2 ꢂ 5H, Cp), ꢀ17.70 (d, JHP
33, 1H, μ-H). 13C{1H} NMR: δ 218.6 (d, JCP = 16, CO), 216.9 (d, JCP
17, CO), 215.8 (d, JCP = 20, CO), 214.5 (d, JCP = 14, CO), 158.1 [d, JCP
=
=
=
14, C1ꢀPh], 132.4 (d, JCP = 6, C2ꢀPh), 126.9 (s, C3ꢀPh), 126.8
(s, C4ꢀPh), 88.1, 82.2, 81.9 (3s, Cp). Data for compound cis,anti-2. Anal.
Calcd for C25H21Fe3O4P: C, 51.42; H, 3.62. Found: C, 51.26; H, 3.42.
1H NMR (CDCl3): δ 7.70 (m, 2H, Ph), 7.16 (m, 2H, Ph), 7.05 (m, 1H,
Ph), 4.68 (s, 15H, Cp), ꢀ18.51 (d, JHP = 34, 1H, μ-H). 13C{1H} NMR
(CDCl3): δ 215.8 (d, JCP = 15, 2CO), 215.7 (d, JCP = 16, 2CO), 156.3
(d, JCP = 3, C1ꢀPh), 131.5 (s, C4ꢀPh), 126.5 (s, C2 and C3ꢀPh), 88.6
(s, Cp), 80.9 (s, 2Cp).
Preparation of [Fe2Cp2(μ-H)(μ-PHPh)(CO)2] (3a). A toluene solu-
tion (20 mL) containing [Fe2Cp2(CO)4] (0.200 g, 0.565 mmol) and
PH2Ph (64 μL, 0.582 mmol) was refluxed for 2.5 h to give a red solution.
The solvent was then removed under vacuum, and the solid residue thus
obtained was dissolved in a minimum amount of dichloromethane/
petroleum ether (1/8) and chromatographed on alumina (activity II) at
253 K. Elution with dichloromethane/petroleum ether (1/6) gave an
orange fraction giving, after removal of solvents, compound trans-3a as
an orange microcrystalline solid (0.160 g, 69%). Elution with dichlor-
omethane/petroleum ether (1/4) gave another orange fraction yielding
analogously compound cis-3a (0.048 g, 21%) as a mixture of the syn and
anti isomers in a relative ratio of ca. 1/2. Data for compound trans-3a.
Anal. Calcd for C18H17Fe2O2P: C, 52.99; H, 4.20. Found: C, 52.76; H,
4.16. 1H NMR (CDCl3): δ 7.78 (m, 2H, Ph), 7.47ꢀ7.27 (m, 3H, Ph),
6.00 (d, JHP = 341, 1H, PꢀH), 4.57, 4.41 (2s, 2 ꢂ 5H, Cp), ꢀ19.24
(d, JHP = 43, 1H, μ-H). Data for compound cis-3a. Anal. Calcd for
C18H17Fe2O2P: C, 52.99; H, 4.20. Found: C, 52.68; H, 4.09. Spectro-
scopic data for isomer cis,anti-3a. 1H NMR (CDCl3): δ 7.68ꢀ7.25
(m, 5H, Ph), 5.26 (d, JHP = 360, 1H, PꢀH), 4.54 (s, 10H, Cp), ꢀ19.67
(d, JHP = 43, 1H, μ-H). Spectroscopic data for isomer cis,syn-3a. 1H NMR
(CDCl3): δ 7.68ꢀ7.25 (m, 5H, Ph), 6.91 (d, JHP = 321, 1H, PꢀH),
4.57 (s, 10H, Cp), ꢀ20.25 (d, JHP = 45, 1H, μ-H).
Preparation of [Fe2Cp2(μ-H)(μ-PHCy)(CO)2] (3b). The procedure is
completely analogous to that just described for 3a but using PH2Cy
(76 μL, 0.572 mmol) and a reaction time of 3 h instead. In this way the
compounds trans-3b (0.164 g, 70%) and cis-3b (0.040 g, 17%) were
obtained as orange solids. Data for compound trans-3b. Anal. Calcd for
C18H23Fe2O2P: C, 52.22; H, 5.60. Found: C, 52.12; H, 5.25. 1H NMR
(400.13 MHz, CDCl3): δ 4.86 (ddd, JHP = 316, JHH = 9, 2, 1H, PꢀH),
4.55, 4.47 (2d, JHP = 1, 2 ꢂ 5H, Cp), 2.45 (m, 1H, Cy), 2.32 (m, 1H,
Cy), 1.95ꢀ1.27 (m, 9H, Cy), ꢀ19.15 (dd, JHP = 40, JHH = 2, 1H, μ-H).
Data for compound cis-3b. Anal. Calcd for C18H23Fe2O2P: C, 52.22; H,
5.60. Found: C, 52.07; H, 5.39. 1H NMR (400.13 MHz, CDCl3): δ 5.74
(ddd, JHP = 300, JHH = 10, 1, 1H, PꢀH), 4.50 (d, JHP = 1, 10H, Cp),
2.24ꢀ2.22 (m, 3H, Cy), 1.93ꢀ1.10 (m, 8H, Cy), ꢀ20.23 (dd, JHP = 43,
JHH = 1, 1H, μ-H).
Ph), 7.82 (m, 1H, Ph), 4.51, 3.98, 3.86 (3s, 3 ꢂ 5H, Cp), ꢀ25.51 (d, JHP
=
40, 1H, μ-H). 13C{1H} NMR (100.62 MHz): δ 263.2 (d, JCP = 5,
μ-CO), 216.3 (d, JCP = 16, FeCO), 216.2 (d, JCP = 17, FeCO), 154.7
(d, JCP = 14, C1ꢀPh), 131.3 (d, JCP = 6, C2ꢀPh), 128.7 (s, C4ꢀPh),
128.6 (s, C3ꢀPh), 84.4, 83.0, 81.7 (3s, Cp).
Preparation of [Fe3Cp3(μ-H)(μ3-PPh)(μ-CO)2] (6). A toluene solu-
tion (10 mL) of cis-2 (0.075 g, 0.128 mmol) was irradiated for 1 h with
visibleꢀUV light in a quartz Schlenk tube refrigerated with tap water
while gently bubbling N2 through the solution to give a red-brown
solution. The solvent was then removed under vacuum, and the residue
was dissolved in a minimum amount of dichloromethane/petroleum
ether (1/2) andchromatographedon alumina(activityII) at 288K. Elution
with dichloromethane/petroleum ether (1/3) gave a brown-reddish fraction
giving, upon removal of solvents, compound 6 as a brown microcrystalline
Photochemical Reaction of Compound 1 with [Fe2Cp2(CO)4]. A
toluene solution (8 mL) containing compound 1 (0.100 g, 0.229 mmol)
and [Fe2Cp2(CO)4] (0.082 g, 0.232 mmol) was irradiated with
10945
dx.doi.org/10.1021/ic201491v |Inorg. Chem. 2011, 50, 10937–10948

Inorganic Chemistry
ARTICLE
Table 6. Crystal Data for New Compounds
2
5
6
7.
mol formula
mol wt
C
25H21Fe3O4P
C24H21Fe3O3P
555.93
C23H20Fe3O2P
C24H21Fe3O3P
555.93
583.94
526.91
cryst syst
monoclinic
triclinic
orthorhombic
monoclinic
C2/c
space group
radiation (λ, Å)
a, Å
P21/c
Pꢀ1
0.71073
Pna21
0.71073
0.71073
0.71073
28.592(2)
7.7225(5)
18.428(1)
90
10.8336(3)
8.6127(3)
9.3554(3)
14.2511(4)
78.704(2)
75.595(2)
73.497(2)
1056.63(6)
2
30.1320(8)
b, Å
13.5644(4)
9.3911(2)
c, Å
16.2488(7)
14.1378(4)
α, deg
90
90
β, deg
109.481(1)
90
91.669(2)
90
γ, deg
V, Å3
90
90
2251.1(1)
4000.6(2)
4067.3(5)
8
Z
4
8
calcd density, g cmꢀ3
absorp coeff, mmꢀ1
temperature, K
θ range (deg)
index ranges (h, k, l)
no. of reflns collected
no. of indep reflns (Rint
reflns with I > 2σ(I)
1.723
1.747
1.750
1.816
2.010
2.133
2.244
2.216
100(2)
100(2)
100(2)
100(2)
1.99ꢀ25.24
ꢀ12, 12; ꢀ16, 0; ꢀ19, 11
15 522
2.29ꢀ25.33
0, 10; ꢀ10, 11; ꢀ16, 17
15 999
1.98ꢀ25.34
ꢀ36, 0; 0, 11; ꢀ16, 17
27 007
2.21ꢀ26.39
ꢀ35, 35; 0, 9; 0, 23
15 091
)
3987 (0.0285)
3828 (0.0229)
7244 (0.0519)
4116 (0.0525)
3532
3640
6695
3044
R indexes [data with I > 2σ(I)]a
R indexes (all data)a
GOF
R1 = 0.0311, wR2 = 0.0847b
R1 = 0.0375, wR2 = 0.1032b
1.186
R1 = 0.0218, wR2 = 0.0553c
R1 = 0.0232, wR2 = 0.0561c
1.114
R1 = 0.0861, wR2 = 0.2122d
R1 = 0.0953, wR2 = 0.2237d
1.093
R1 = 0.0397, wR2 = 0.0821e
R1 = 0.0685, wR2 = 0.0935e
1.047
no. of restraints/params
ΔF(max, min), e Åꢀ3
0/302
0/284
115/509
0/284
0.634, ꢀ0.657
0.386, ꢀ0.286
5.655, ꢀ2.061
0.805, ꢀ0.617
a R = Σ||Fo| ꢀ |Fc||/Σ|Fo|. wR = [Σw(|Fo|2 ꢀ |Fc|2)2/Σw|Fo|2]1/2. w = 1/[σ2(Fo ) + (aP)2 + bP], where P = (Fo2 + 2Fc2)/3. b a = 0.0515, b = 2.4279.
2
c a = 0.0246, b = 0.7989. d a = 0.1692, b = 11.2139 e a = 0.0337, b = 14.9859.
solid (0.061 g, 90%). X-ray-quality crystals of 6 were grown by slow diffusion
of layers of petroleum ether and toluene into a dichloromethane solution of
the complex at 253 K. Anal. Calcd for C23H21Fe3O2P: C, 52.33; H, 4.01.
Found: C, 52.18; H, 3.95. 1H NMR (400.13 MHz): δ 8.85 (m, 2H, Ph),
130.0 (s, C4ꢀPh), 128.3 (d, JCP = 11, C3ꢀPh), 94.3 (s, Cp), 90.3
(s, 2Cp).
X-ray Structure Determination for Compounds 2, 5, and 6. The
X-ray intensity data were collected at 100 K on a Nonius KappaCCD
single-crystal diffractometer using graphite-monochromated Mo Kα
radiation. Images were collected at a fixed crystalꢀdetector distance
(29 mm for 2, 30 mm for 5, and 45 mm for 6) using the oscillation
method with 1° (2 and 5) or 0.7° (6) oscillations and 40 s exposure time
per image. Data collection strategies were calculated with the program
Collect.31 Data reduction and cell refinement were performed with the
programs HKL Denzo and Scalepack.32 A semiempirical absorption
correction was applied using the program SORTAV.33 Using the
program suite WinGX,34 the structures were solved by Patterson
interpretation and phase expansion using SHELXL9735 and refined
with full-matrix least-squares on F2 using SHELXL97. For compounds 2
and 5, all non-hydrogen atoms were refined anisotropically and all
hydrogen atoms were fixed at calculated positions except for the hydride
H(1) atoms, which were located in the Fourier map and refined; all them
were given an overall isotropic thermal parameter. For compound 6 two
molecules of the complex were found to be present in the asymmetric
unit. Moreover, due to the low quality of the diffraction data (some
twinning was found to occur in the crystal but the twin law could not be
determined), not all the positional parameters and anisotropic tempera-
ture factors of all the non-H atoms could be refined anisotropically. A
significant number of atoms had to be refined anisotropically in
combination with the instructions DELU and SIMU, and 3 atoms were
finally refined isotropically to prevent their temperature factors from
becoming nonpositive definite. Besides, restrictions in the CꢀC lengths
7.90ꢀ7.82(m,3H, Ph),4.28(s,10H, Cp),4.14(s, 5H,Cp),ꢀ34.82 (d, JHP
=
45, 1H, μ-H). 13C{1H} NMR (100.62 MHz): δ 273.8 (s, μ-CO), 145.3
(d, JCP = 9, C1ꢀPh), 134.5 (d, JCP = 9, C2ꢀPh), 131.9 (s, C4ꢀPh), 129.6
(d, JCP = 11, C3ꢀPh), 86.5 (s, Cp), 81.2 (s, 2Cp).
Preparation of [Fe3Cp3(μ-PHPh)(μ3-CO)(μ-CO)2] (7). A toluene
solution (6 mL) of cis-2 (0.060 g, 0.103 mmol) was irradiated for 2 h
with visibleꢀUV light in a quartz Schlenk tube refrigerated with tap
water to give a brown solution. The mixture was then exposed to a CO
atmosphere and further stirred for 10 min at room temperature to give a
green solution. The solvent was then removed under vacuum, and the
residue was dissolved in a minimum amount of dichloromethane/
petroleum ether (1/1) and chromatographed on alumina (activity
IV) at 288 K. Elution with dichloromethane/petroleum ether (1/2)
gave a green-brownish fraction giving, upon removal of solvents,
compound 5 (0.034 g, 59%). Elution with dichloromethane/petroleum
ether (2/1) gave a dark-green fraction yielding analogously compound 7
as a green microcrystalline solid (0.020 g, 35%). X-ray-quality crystals of
7 were grown by slow diffusion of layers of petroleum ether and toluene
into a dichloromethane solution of the compound at 253 K. Anal.
Calcd for C24H21Fe3O3P: C, 51.85; H, 3.81. Found: C, 51.72; H, 3.77.
1H NMR (400.13 MHz): δ 7.65 (m, 2H, Ph), 7.40 (m, 1H, Ph), 7.32
(m, 2H, Ph), 6.66 (d, JHP = 350, 1H, PꢀH), 4.71 (d, JHP = 1, 10H, Cp),
4.69 (s, 5H, Cp). 13C{1H} NMR (100.62 MHz): δ 297.0 (s, μ3-CO),
272.7 (s, μ-CO), 147.6 (d, JCP = 14, C1ꢀPh), 134.3 (d, JCP = 7, C2ꢀPh),
10946
dx.doi.org/10.1021/ic201491v |Inorg. Chem. 2011, 50, 10937–10948

Inorganic Chemistry
ARTICLE
of the cyclopentadienyl rings had to be imposed for a convenient
convergence. All hydrogen atoms were geometrically placed and refined
using a riding model, except for the hydride ligand, which could not be
located in the Fourier map nor modeled due to the low quality of the
crystal. Crystallographic data and structure refinement details for
compounds 2, 5, and 6 are collected in Table 6.
X-ray Structure Determination for Compound 7. X-ray intensity data
were collected on a Kappa-Appex-II Bruker diffractometer using gra-
phite-monochromated Mo Kα radiation at 100 K. The software APEX36
was used for collecting frames with the ω/ϕ scans measurement method.
The Bruker SAINT software was used for data reduction,37 and a
multiscan absorption correction was applied with SADABS.38 Using
the program suite WinGX,34 the structure was solved by direct methods
using SHELXL9735 and refined with full-matrix least-squares on F2 using
SHELXL97. All positional parameters and anisotropic temperature
factors of all the non-H atoms were refined anisotropically, and all
hydrogen atoms were geometrically placed and refined using a riding
model except for the P-bound H(1), which was located in the Fourier
maps and refined isotropically. Crystallographic data and structure
refinement details for 7 are collected in Table 6.
Schauer, C. K. Inorg. Chem. 1993, 32, 5429. (h) Borg-Breen, C. C.;
Bautista, M. T.; Schauer, C. K.; White, P. S. J. Am. Chem. Soc. 2000,
122, 3952.
(6) Shackleton, T. A.; Mackie, S, C.; Fergusson, S. B.; Johnston, L. J.;
Baird, M. C. Organometallics 1990, 9, 2248.
(7) (a) Alvarez, C. M.; Galꢀan, B.; García, M. E.; Riera, V.; Ruiz, M. A.;
Vaissermann, J. Organometallics 2003, 22, 5504. (b) Alvarez, C. M.;
García, M. E.; Ruiz, M. A.; Connelly, N. G. Organometallics 2004,
23, 4750.
(8) Decaen, A.; Craig, C. D.; Bottomley, F. Acta Crystallogr., Sect. E
2004, 60, 1284.
(9) Sugiura, J.; Kakizawa, T.; Hashimoto, H.; Tobita, H.; Ogino, H.
Organometallics 2005, 24, 1099.
(10) Pohl, W.; Lorenz, I. P.; Noeth, H.; Schmidt, M. Z. Naturforsch. B
1995, 50, 1485.
(11) (a) Cotton, F. A.; Frenz, B. A.; White, A. J. Inorg. Chem. 1974,
13, 1407. (b) Klasen, C.; Lorenz, I. P.; Schmid, S.; Beuter, G. J. Organomet.
Chem. 1992, 428, 363.
(12) Braterman, P. S. Metal Carbonyl Spectra; Academic Press:
London, 1975.
(13) (a) Brunner, H.; Peter, H. J. Organomet. Chem. 1990, 393, 411.
(b) Brunner, H.; R€otzer, M. J. Organomet. Chem. 1992, 425, 119.
(14) Carty, A. J.; MacLaughlin, S. A.; Nucciarone, D. In Phosphorus-
31 NMR Spectroscopy in Stereochemical Analysis; Verkade, J. G., Quin,
L. D., Eds.; VCH: Deerfield Beach, FL, 1987; Chapter 16.
(15) (a) Haines, R. J.; du Preez, A. L. Inorg. Chem. 1969, 8, 1459. (b)
Klasen, C.; Lorenz, I. P.; Schmid, S.; Beuter, G. J. Organomet. Chem.
1992, 428, 363. (c) Zhang, S.; Brown, T. L. Organometallics 1992, 11,
4166.
’ ASSOCIATED CONTENT
S
Supporting Information. CIF file giving crystallographic
b
data for structural analysis of compounds 2, 5, 6, and 7. This
material is available free of charge via the Internet at http://pubs.
acs.org.
(16) Bitterwolf, T. E. Coord. Chem. Rev. 2000, 206ꢀ207, 419.
(17) (a) Zhang, S.; Brown, T. L. Organometallics 1992, 11, 4166. (b)
Zhang, S.; Brown, T. L. J. Am. Chem. Soc. 1993, 113, 1779.
(18) Zhang, S.; Brown, T. L. Organometallics 1992, 11, 2122.
(19) (a) Ebsworth, E. A. V.; McIntosh, A. P.; Schr€oder, M.
J. Organomet. Chem. 1986, 312, C41. (b) Arce, A. J.; Machado, R.; De
Sanctis, Y.; Gonzꢀalez, T.; Atencio, R.; Deeming, A. J. Inorg. Chim. Acta
2003, 344, 123.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: mara@uniovi.es.
’ ACKNOWLEDGMENT
(20) Mani, D.; Vahrenkamp, H. Chem. Ber. 1986, 119, 3639.
(21) Alvarez, C. M.; Alvarez, M. A.; García-Vivꢀo, D.; García, M. E.; Ruiz,
M. A.; Sꢀaez, D.; Falvello, L. R.; Soler, T.; Herson, P. Dalton Trans. 2004, 4168.
(22) For reviews on the reactivity of paramagnetic organometallic
compounds see, for example: (a) In Paramagnetic Organometallic Species
in Activation Selectivity, Catalysis; Chanon, M., Julliard, M., Poite, J. C.,
Eds.; Kluwer Academic Publishers: Dordrecht, 1989. (b) In Organome-
tallic Radical Processes; Trogler, W. C., Ed.; Elsevier: Amsterdam, 1990.
(c) Astruc, D. Electron Transfer and Radical Processes in Transition-Metal
Chemistry; VCH: New York, 1995. (d) Sun, S.; Sweigart, D. A. Adv.
Organomet. Chem. 1996, 40, 171. (e) Poli, R. Chem. Rev. 1996, 96, 2135.
(f) Hoff, C. D. Coord. Chem. Rev. 2000, 206, 451. (g) Torraca, K. E.;
McElwee-White, L. Coord. Chem. Rev. 2000, 206, 469. (h) Smith, K. M.
Organometallics 2005, 24, 778. (i) Geiger, W. E. Organometallics 2007,
26, 5738.
(23) Brauer, D. J.; Hietkamp, S.; Sommer, H.; Stelzer, O.; Muller, G.;
Kruger, C. J. Organomet. Chem. 1985, 288, 35.
(24) Neuman, M. A.; Tohan, T.; Dahl, L. F. J. Am. Chem. Soc. 1972,
94, 3383.
(25) Wadepohl, H.; Gebert, S.; Pritzkow, H. J. Organomet. Chem.
2000, 614, 158.
(26) Yeh, W.-Y.; Liu, Y.-C.; Peng, S.-M.; Lee, G.-H. J. Organomet.
Chem. 2004, 689, 1014.
(27) (a) Colbran, S. B.; Johnson, B. F. G.; Lewis, J.; Sorrell, R. M.
J. Organomet. Chem. 1985, 296, C1. (b) Deeming, A. J.; Doherty, S.; Day,
M. W.; Hardcastle, K. I.; Minassian, H. J. Chem. Soc., Dalton Trans.
1991, 1273.
(28) Haupt, H. J.; Schwefer, M.; Egold, H.; Fl€orke, U. Inorg. Chem.
1995, 34, 5461.
(29) Armarego, W. L. F.; Chai, C. Purification of Laboratory Chemicals,
5th ed.; Butterworth-Heinemann: Oxford, U.K., 2003.
We thank the DGI of Spain (Projects CTQ2006-01207 and
CTQ2009-09444) and the COST action CM0802 ‘‘PhoSciNet’’ for
supporting this work and the MEC of Spain for a grant (to R.G.).
’ REFERENCES
(1) (a) Alvarez, C. M.; Alvarez, M. A.; García, M. E.; Gonzꢀalez, R.;
Ruiz, M. A.; Hamidov, H.; Jeffery, J. C. Organometallics 2005, 24, 5503.
(b) Alvarez, M. A.; García, M. E.; Gonzꢀalez, R.; Ramos, A.; Ruiz, M. A.
Organometallics 2010, 29, 1875. (c) Alvarez, M. A.; García, M. E.;
Gonzꢀalez, R.; Ramos, A.; Ruiz, M. A. Organometallics 2011, 30, 1102.
(2) Alvarez, M. A.; García, M. E.; Gonzꢀalez, R.; Ramos, A.; Ruiz,
M. A. Inorg. Chem. 2011, 50, 7894.
(3) Huttner, G.; Knoll, K. Angew. Chem., Int. Ed. Engl. 1987, 26, 743.
(4) For representative 48-electron clusters see:(a) Huttner, G.;
Schneider, J.; Mohr, G.; von Seyerl, J. J. Organomet. Chem. 1980,
191, 161. (b) Knoll, K.; Huttner, G.; Zsolnai, L.; Jibril, I.; Wasiucionek,
M. J. Organomet. Chem. 1985, 294, 91. (c) Sunick, D. L.; White, P. S.;
Schauer, C. K. Organometallics 1993, 12, 245. (d) Bitterer, F.; Kucken, S.;
Langhans, K. P.; Stelzer, O.; Sheldrick, W. S. Z. Naturforsh. B 1994,
49, 1223. (e) Wang, D. Z.; Wu, W. F.; Liu, S. T.; Gong, P. J. Chin. J. Inorg.
Chem. 2003, 19, 1090.
(5) For representative 50-electron complexes, see: (a) Winter, A.;
Zsolnai, L.; Huttner, G. J. Organomet. Chem. 1982, 234, 337. (b) Cook,
S. L.; Evans, J.; Gray, L. R.; Webster, M. J. Organomet. Chem. 1982,
236, 367. (c) Williams, G. D.; Geoffroy, G. L.; Whittle, R. R.; Rheingold,
A. L. J. Am. Chem. Soc. 1985, 107, 729. (d) Sherer, O. J.; Gobel, K.; Kaub,
J. Angew. Chem., Int. Ed. Engl. 1987, 26, 59. (e) Barlett, R. A.; Dias,
H. V. R.; Flynn, K. M.; Olmstead, M. M.; Power, P. P. J. Am. Chem. Soc.
1987, 109, 5699. (f) Song, J. S.; Geoffroy, G. L.; Rheingold, A. L. Inorg.
Chem. 1992, 31, 1505. (g) Bautista, M. T.; Jordan, M. R.; White, P. S.;
10947
dx.doi.org/10.1021/ic201491v |Inorg. Chem. 2011, 50, 10937–10948

Inorganic Chemistry
ARTICLE
(30) Labinger, J. A.; Madhavan, S. J. Organomet. Chem. 1977, 134, 381.
(31) Collect; Nonius BV: Delft, The Netherlands, 1997ꢀ2004.
(32) Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307.
(33) Blessing, R. H. Acta Crystallogr., Sect. A 1995, 51, 33.
(34) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837.
(35) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112.
(36) APEX 2, version 2.0-1; Bruker AXS Inc.: Madison, WI, 2005.
(37) SMART & SAINT Software Reference Manuals, version 5.051
(Windows NT version); Bruker Analytical X-ray Instruments: Madison,
WI, 1998.
(38) Sheldrick, G. M. SADABS, Program for Empirical Absorption
Correction; University of G€ottingen: G€ottingen, Germany, 1996.
10948
dx.doi.org/10.1021/ic201491v |Inorg. Chem. 2011, 50, 10937–10948