Activation of H-H and H-O bonds at phosphorus with diiron complexes bearing pyramidal phosphinidene ligands

Article abstract of DOI:10.1021/ic2026945

The complex [Fe₂Cp₂(μ-PMes*)(μ-CO)(CO) ₂] (Mes* = 2,4,6-C₆H₂^tBu ₃), which in the solid state displays a pyramidal phosphinidene bridge, reacted at room temperature with H₂ (ca. 4 atm) to give the known phosphine complex [Fe₂Cp₂(μ-CO) ₂(CO)(PH₂Mes*)] as the major product, along with small amounts of other byproducts arising from the thermal degradation of the starting material, such as the phosphindole complex [Fe₂Cp ₂(μ-CO)₂(CO){PH(CH₂CMe₂)C ₆H₂^tBu₂}], the dimer [Fe ₂Cp₂(CO)₄], and free phosphine PH ₂Mes*. During the course of the reaction, trace amounts of the mononuclear phosphide complex [FeCp(CO)₂(PHMes*)] were also detected, a compound later found to be the major product in the carbonylation of the parent phosphinidene complex, with this reaction also yielding the dimer [Fe₂Cp₂(CO)₄] and the known diphosphene Mes*P=PMes*. The outcome of the carbonylation reactions of the title complex could be rationalized by assuming the formation of an unstable tetracarbonyl intermediate [Fe₂Cp₂(μ-PMes*)(CO) ₄] (undetected) that would undergo a fast homolytic cleavage of a Fe-P bond, this being followed by subsequent evolution of the radical species so generated through either dimerization or reaction with trace amounts of water present in the reaction media. A more rational synthetic procedure for the phosphide complex was accomplished through deprotonation of the phosphine compound [FeCp(CO)₂(PH₂Mes*)](BF₄) with Na(OH), the latter in turn being prepared via oxidation of [Fe ₂Cp₂(CO)₄] with [FeCp₂](BF ₄) in the presence of PH₂Mes*. To account for the hydrogenation of the parent phosphinidene complex it was assumed that, in solution, small amounts of an isomer displaying a terminal phosphinidene ligand would coexist with the more stable bridged form, a proposal supported by density functional theory (DFT) calculations of both isomers, with the latter also revealing that the frontier orbitals of the terminal isomer (only 5.7 kJ mol^-1 above of the bridged isomer, in toluene solution) have the right shapes to interact with the H₂ molecule. In contrast to the above behavior, the cyclohexylphosphinidene complex [Fe₂Cp ₂(μ-PCy)(μ-CO)(CO)₂] failed to react with H ₂ under conditions comparable to those of its PMes* analogue. Instead, it slowly reacted with HOR (R = H, Et) to give the corresponding phosphinous acid (or ethyl phosphinite) complexes [Fe₂Cp ₂(μ-CO)₂(CO){PH(OR)Mes*}], a behavior not observed for the PMes* complex. The presence of BEt₃ increased significantly the rate of the above reaction, thus pointing to a pathway initiated with deprotonation of an O-H bond of the reagent by the basic P center of the phosphinidene complex, this being followed by the nucleophilic attack of the OR^- anion at the P site of the transient cationic phosphide thus formed. The solid-state structure of the cis isomer of the ethanol derivative was determined through a single crystal X-ray diffraction study (Fe-Fe = 2.5112(8) A, Fe-P = 2.149(1) A).

Full text of DOI:10.1021/ic2026945

Article
pubs.acs.org/IC
Activation of H−H and H−O Bonds at Phosphorus with Diiron
Complexes Bearing Pyramidal Phosphinidene Ligands
M. Angeles Alvarez, M. Esther García, Daniel García-Vivo, Alberto Ramos,* and Miguel A. Ruiz*
́
Departamento de Química Organ
́
ica e Inorgan
́
ica/IUQOEM, Universidad de Oviedo, E-33071 Oviedo, Spain
S
* Supporting Information
t
ABSTRACT: The complex [Fe2Cp2(μ-PMes*)(μ-CO)(CO)2] (Mes* = 2,4,6-C6H2 Bu3),
which in the solid state displays a pyramidal phosphinidene bridge, reacted at room
temperature with H2 (ca. 4 atm) to give the known phosphine complex [Fe2Cp2(μ-
CO)2(CO)(PH2Mes*)] as the major product, along with small amounts of other
byproducts arising from the thermal degradation of the starting material, such as the
t
phosphindole complex [Fe2Cp2(μ-CO)2(CO){PH(CH2CMe2)C6H2 Bu2}], the dimer
[Fe2Cp2(CO)4], and free phosphine PH2Mes*. During the course of the reaction, trace
amounts of the mononuclear phosphide complex [FeCp(CO)2(PHMes*)] were also
detected, a compound later found to be the major product in the carbonylation of the
parent phosphinidene complex, with this reaction also yielding the dimer [Fe2Cp2(CO)4]
and the known diphosphene Mes*PPMes*. The outcome of the carbonylation reactions
of the title complex could be rationalized by assuming the formation of an unstable
tetracarbonyl intermediate [Fe2Cp2(μ-PMes*)(CO)4] (undetected) that would undergo a
fast homolytic cleavage of a Fe−P bond, this being followed by subsequent evolution of the radical species so generated through
either dimerization or reaction with trace amounts of water present in the reaction media. A more rational synthetic procedure
for the phosphide complex was accomplished through deprotonation of the phosphine compound [FeCp(CO)2(PH2Mes*)]-
(BF4) with Na(OH), the latter in turn being prepared via oxidation of [Fe2Cp2(CO)4] with [FeCp2](BF4) in the presence of
PH2Mes*. To account for the hydrogenation of the parent phosphinidene complex it was assumed that, in solution, small
amounts of an isomer displaying a terminal phosphinidene ligand would coexist with the more stable bridged form, a proposal
supported by density functional theory (DFT) calculations of both isomers, with the latter also revealing that the frontier orbitals
of the terminal isomer (only 5.7 kJ mol−1 above of the bridged isomer, in toluene solution) have the right shapes to interact with
the H2 molecule. In contrast to the above behavior, the cyclohexylphosphinidene complex [Fe2Cp2(μ-PCy)(μ-CO)(CO)2] failed
to react with H2 under conditions comparable to those of its PMes* analogue. Instead, it slowly reacted with HOR (R = H, Et) to
give the corresponding phosphinous acid (or ethyl phosphinite) complexes [Fe2Cp2(μ-CO)2(CO){PH(OR)Mes*}], a behavior
not observed for the PMes* complex. The presence of BEt3 increased significantly the rate of the above reaction, thus pointing to
a pathway initiated with deprotonation of an O−H bond of the reagent by the basic P center of the phosphinidene complex, this
being followed by the nucleophilic attack of the OR− anion at the P site of the transient cationic phosphide thus formed. The
solid-state structure of the cis isomer of the ethanol derivative was determined through a single crystal X-ray diffraction study
(Fe−Fe = 2.5112(8) Å, Fe−P = 2.149(1) Å).
Chart 1
INTRODUCTION
■
The chemistry of phosphinidene (PR) complexes has been a
very active area of research during the past three decades.1
Most of this work has been focused on terminal complexes
having bent-phosphinidene ligands, which are usually compared
to carbenes and can also be classified as electrophilic or
nucleophilic. In contrast, the chemistry of binuclear complexes
bearing bridging phosphinidene ligands has been comparatively
less developed. Recently we implemented a high-yield route to
the stable diiron compounds [Fe2Cp2(μ-PR)(μ-CO)(CO)2]
(R= Cy (1a), Ph (1b), Chart 1), having a phosphinidene bridge
with a pyramidal environment and displaying a strong
nucleophilic behavior.2 We noticed at the time that there
were only a few diiron phosphinidene complexes previously
described in the literature, with most of them being either
thermally unstable (as the cyclopentadienyl complex
[Fe2Cp2(μ-PPh)(CO)4]),3 or transient species generated
from suitable precursors, such as [Fe2{μ-P(NiPr2)}2(CO)6],4
[Fe2(μ-PtBu)(CO)6],5 or [Fe2(μ-PPh)2(CO)6]2−.6 We can
quote also a couple of isolable diiron phosphinidene complexes,
Received: December 14, 2011
Published: March 1, 2012
© 2012 American Chemical Society
3698
dx.doi.org/10.1021/ic2026945 | Inorg. Chem. 2012, 51, 3698−3706

Inorganic Chemistry
Article
[Fe2{μ-P(OR)}2(CO)6]7 and [Fe2(μ-PPh)2(L2)2],8 both of
account the known degradation processes of 1c in solution.
Thus, we have previously shown that 1c slowly decomposes at
room temperature in toluene solution (ca. 80% conversion after
4 days) to give [Fe2Cp2(μ-CO)2(CO){PH(CH2CMe2)-
them having bulky protecting groups (R = 4,2,6-C6H2MetBu2,
i
L2 = HC(CMeNC6H3 Pr2)2) and a trigonal geometry at
phosphorus, although no reactivity appears to have been
developed around these complexes. We then started a detailed
study of the reactivity of our diiron complexes and found that
the cyclohexylphosphinidene complex 1a was quite reactive
toward a great variety of reagents, including alkyl halides,
chalcogens, alkynes, diazoalkanes, organic azides, and metal
carbonyl complexes, to give a variety of organophosphorus
derivatives and heterometallic clusters.2,9
t
C6H2 Bu2}], following from the intramolecular addition of a
C−H bond of an o-tBu group at the P center, and also some
[Fe2Cp2(CO)4].10b In addition, the phosphine complex
[Fe2Cp2(μ-CO)2(CO)(PH2Mes*)] is also known to decom-
pose slowly in solution to give free phosphine (PH2Mes*) and
more [Fe2Cp2(CO)4].10b Finally, the phosphinidene oxide
complex obviously arises from the reaction of 1c with trace
amounts of O2 present in the reaction media.10a
To examine the influence of steric factors on all this rich
chemistry we later prepared the complex [Fe2Cp2(μ-PMes*)-
(μ-CO)(CO)2] (1c), with Mes* being the bulky group 2,4,6-
To detect possible intermediates, we also carried out the
hydrogenation of 1c (C6D6 solution) within an NMR tube
equipped with a Young's valve and pressurized with about 3
atm of H2, to monitor the reaction by NMR spectroscopy. A
blank experiment was also carried out with another aliquot of
1c in C6D6 solution, under a N2 atmosphere. The hydro-
genation reaction was significantly slower under these
conditions, probably because of the slower incorporation of
dihydrogen into the solution in the absence of stirring, coupled
to the progressive loss of H2 pressure within the NMR tube
(overall reaction time of ca. 4 weeks). As a result, the relative
amount of byproducts increased significantly, especially those
derived from the thermal degradation of 1c: the dimer
[Fe2Cp2(CO)4] (30%) and the phosphindole complex
t
C6H2 Bu3, and found that the steric pressure introduced by the
Mes* group had pronounced effects concerning its synthesis,
structure, thermal stability, and general reactivity.10 Particularly
intriguing was the remarkable ability of this complex to react
with dihydrogen at room temperature, to give the phosphine
complex [Fe2Cp2(μ-CO)2(CO)(PH2Mes*)] as the major
product.10a This behavior was quite noticeable since no other
phosphinidene complex of any type had been reported
previously to be hydrogenated under similar conditions, while
phosphines and related organophosphorus molecules (also
having an electron pair and the ability to increase the
coordination number at phosphorus) equally fail to react easily
with hydrogen (except under Frustrated Lewis Pair con-
ditions).11 Noticeably, Bertrand et al. had recently reported the
first example of hydrogenation of a free carbene molecule.12 In
this paper we report a detailed study of the hydrogenation
reaction of compounds 1a,c including the analysis of some side
processes such as the reaction of 1c with CO. We have also
examined the ability of these compounds to activate the H−O
bonds of water and ethanol. As it will be shown, the pathways
involved in the activation of these two types of bonds (H−H vs
H−O) seem to be completely different, with the hydrogenation
of 1c following from the ability of this crowded molecule to
rearrange into an isomer having a terminal bent-phosphinidene
ligand, an hypothesis supported by density functional theory
(DFT) calculations on both 1c and its phenyl analogue 1b.
Interestingly, while this paper was being finished, Mathey et al.
reported that the transient bent-phosphinidene complexes
[Mo(PR)(CO)5] (R = Me, Ph) were hydrogenated under more
forcing conditions (20 bar, T > 393 K) to give the phosphine
derivatives [Mo(CO)5(PH2R)] via the corresponding diphos-
phine complexes [Mo2(μ-PhHPPHPh)(CO)10].13
t
[Fe2Cp2(μ-CO)2(CO){PH(CH2CMe2)C6H2 Bu2}] (25%),
whereas the yield of the hydrogenation product [Fe2Cp2(μ-
CO)2(CO)(PH2Mes*)] decreased considerably (12%). Un-
expectedly, at early stages in both experiments (H2 or N2
atmosphere) we detected the presence of small amounts of a
species that we have later identified as the mononuclear
phosphide complex [FeCp(CO)2(PHMes*)] (3) (ca. 5% by
1H NMR analysis), a degradation product of 1c which had
previously gone undetected (Chart 2).10 In the hydrogenation
Chart 2
experiment, however, complex 3 disappeared from the reaction
mixture once the starting material was totally consumed,
whereas it remained steady in the blank experiment as the
amount of the other products coming from the thermal
degradation of 1b increased. The final product distribution in
the blank experiment was [Fe2Cp2(μ-CO)2(CO){PH-
RESULTS AND DISCUSSION
■
Hydrogenation of Compound 1c. The reactivity of
compounds 1a,c toward H2 was tested under mild conditions
(room temperature, ca. 4 atm). While the cyclohexylphosphi-
nidene complex 1a failed to react after 6 d in toluene solution,
we found that the more congested compound 1c did react with
H2 under the same conditions, the reaction being completed in
about 16 h, to give the phosphine complex [Fe2Cp2(μ-
CO)2(CO)(PH2Mes*)] (actually, the synthetic precursor of
t
(CH2CMe2)C6H2 Bu2}] (55%), [Fe2Cp2(CO)4] (15%), com-
plex 3 (5%), PH2Mes* (traces), and other unidentified
products (15%).
Carbonylation of Compound 1c. While searching for
alternative degradation processes of 1c that could account for
the formation of the phosphide complex 3, we found that this
complex could be obtained in good yield through the reaction
of 1c with CO. Different carbonylation experiments were
carried out for 1c in different solvents (CH2Cl2 or toluene) and
reaction times (from 15 to 20 h), usually at a pressure of about
4 atm of CO, to give essentially the same product distribution:
the phosphide complex 3 (55%), the dimer [Fe2Cp2(CO)4]
1c)10 as the major compound (ca. 50% by H NMR). Other
1
products detected in the reaction mixture were free phosphine
PH2Mes* (20%), the dimer [Fe2Cp2(CO)4] (15%), the
oxophosphinidene complex [Fe2Cp2(CO)3{P(O)Mes*}]10a
(10%), and the phosphindole derivative [Fe2Cp2(μ-
10
t
CO)2(CO){PH(CH2CMe2)C6H2 Bu2}] (5%). The presence
of all these minor products can be rationalized by taking into
3699
dx.doi.org/10.1021/ic2026945 | Inorg. Chem. 2012, 51, 3698−3706

Inorganic Chemistry
Article
(40%), and the diphosphene Mes*PPMes* (5%), the latter
being identified by its 31P NMR resonance in C6D6 at 492
ppm.14 As discussed later on, the hydrogen needed to form 3
seems to arise from traces of water present in the solvent and
gas used.
complex formed after 3 d in ethanol solution at room
temperature was lower than 5%. Clearly, some kind of
interaction between 1c and the EtOH molecules must take
place to suppress the intramolecular C−H cleavage leading to
the phosphindole derivative. A comparable suppression of a
related C−H cleavage has been previously observed in the
photochemical reactions of the trigonal phosphinidene complex
[Mo2Cp2(μ-PMes*)(CO)4] when carried out in the presence
of acetonitrile.19
As opposed to 1c, the less crowded and more basic PCy-
bridged complex 1a did react slowly with both substrates. The
reaction with H2O (ca. 500 equiv.) in toluene was completed in
about 5 h at room temperature to give the phosphinous acid
complex [Fe2Cp2(μ-CO)2(CO){PHCy(OH)}] (4), a rather
unstable species decomposing easily upon manipulation (Chart
3). Interestingly, this reaction proceeded more rapidly in the
Rational Synthesis of Compound 3. The phosphide
complex 3 could be more selectively prepared through a two-
step procedure starting from [Fe2Cp2(CO)4]. First, the
phosphine complex [FeCp(CO)2(PH2Mes*)][BF4] (2) was
prepared in 56% yield through the oxidation of [Fe2Cp2(CO)4]
with [FeCp2]BF4 (2 equiv) in the presence of PH2Mes* (2
equiv), a procedure previously implemented by Schumann to
prepare the related PPh3 complex.15 Then, compound 2 was
deprotonated with Na(OH) to give compound 3 in 60% yield.
The spectroscopic data in solution for compounds 2 and 3 are
completely analogous to those of related primary phosphine
and phosphide complexes reported by Malisch and co-
workers,16 and therefore deserve no detailed comments, except
to note that the IR spectra of these complexes in dichloro-
methane solution show two strong C−O stretches (Table 1), as
Chart 3
Table 1. Selected IR and 31P NMR Data for New
Compounds
a
b
compound
ν(CO)
δP(JHP)
[FeCp(CO)2(PH2Mes*)]BF4 (2) 2065 (vs), 2021 (s)
−49.2 (392)
−91.8 (179)
171.3 (345)
192.4 (337)
189.9 (333)
presence of 1 equiv of BEt3, then being completed in only 45
min when using just 20 equiv of H2O. A similar effect took
place for the reaction of 1a with ethanol, this being used both as
solvent and reagent, then leading to the more stable
phosphinite complex [Fe2Cp2(μ-CO)2(CO){PHCy(OEt)}]
(5). In the absence of BEt3, only 50% conversion was obtained
after 3 days at room temperature, whereas full conversion was
achieved in the same reaction time after addition of 1 equiv of
BEt3. In the latter case, however, variable amounts of
compound 4 (up to 25%, measured by 31P NMR) were also
generated, probably arising from the reaction with traces of
H2O present in the solvent. We finally note that the above
reactions of the diiron complex 1a are reminiscent of the
reactions reported by Mathey and co-workers between
transient electrophilic phosphinidenes of the type [M-
(CO)n(PPh)] (M = Cr, Mo, W; n = 5;20a,b M = Fe, n =
420c) with water or simple alcohols to give the corresponding
phosphinite derivatives [M(CO)n{PHPh(OR)}] (R = H, alkyl
group).
Structural Characterization of Compounds 4 and 5.
The solid-state structure of cis-5 was determined through a
single-crystal X-ray diffraction analysis (Figure 1 and Table 2).
The structure of cis-5 is built from two CpFe fragments in a
cisoid arrangement bridged by two carbonyl ligands in an
almost symmetric fashion. The coordination sphere of each
metal center is completed by a carbonyl and a phosphinite
ligand, respectively. The bridging CO ligands are almost
coplanar, with the angle between the corresponding Fe2C
planes being about 160°, in the range found for related
complexes of the formula cis-[Fe2Cp2(μ-CO)2(CO)L] (L =
P(OPh)3,21 PPh2(CH2Ph),22 P(Ph)C4H2Me223). The interme-
tallic length of 2.5112(8) Å in cis-5 is consistent with the
formulation of a single metal−metal bond, and it is only
marginally shorter than the average distance measured for the
mentioned complexes (ca. 2.54 Å), while the Fe2−P1 distance
of 2.149(1) Å is also comparable to the iron-phosphite or iron-
phosphine separations in these species.
[FeCp(CO)2(PHMes*)] (3)
2007 (vs), 1955 (s)
c
[Fe2Cp2(μ-CO)2(CO)
{PHCy(OH)}] (4)
1941 (s), 1769 (w),
1720 (vs)
[Fe2Cp2(μ-CO)2(CO)
{PHCy(OEt)}] (trans-5)
1935 (s), 1767 (w),
1729 (vs)
[Fe2Cp2(μ-CO)2(CO)
{PHCy(OEt)}] (cis-5)
a
Recorded in dichloromethane solution, with C−O stretching bands
b
(ν(CO)) in cm−1. Recorded in CD2Cl2 solutions at 290 K and 121.50
MHz unless otherwise stated; δ in ppm relative to external 85%
aqueous H3PO4 and coupling constants (J) in hertz. Recorded in
CH2Cl2 with a D2O insert.
c
anticipated for cisoid M(CO)2 oscillators in a pseudooctahedral
coordination environment.17 The bands of the cationic 2 are
shifted significantly to higher frequencies (by ca. 60 cm−1) with
respect to those of the neutral 3, as expected. Yet, the most
informative spectroscopic features are found in the 31P and H
1
NMR resonances corresponding to the PHx moieties of these
compounds. These appear in the case of 2 (x = 2) as a triplet at
−49.2 ppm (31P) and a doublet at 6.71 ppm (1H) with a large
coupling constant (1JPH) of 392 Hz, whereas for compound 3
(x = 1) these resonances appear as two more shielded doublets,
at −91.8 ppm (31P) and 5.02 ppm (1H) respectively, with a
largely reduced coupling constant (1JPH) of 179 Hz, a diagnostic
effect of the presence of the lone electron pair at phosphorus in
the neutral phosphide complex.18
Reactions of Compounds 1a,c with H2O and EtOH.
The behavior of our phosphinidene complexes was completely
reversed when faced to the polar single O−H bonds of water or
alcohols. Thus, the PMes* complex 1c failed to react with these
reagents, even when using H2O as cosolvent (H2O/THF ratio
= 1/5) or ethanol as solvent. Only a progressive thermal
degradation of 1c to give the phosphindole complex
t
[Fe2Cp2(μ-CO)2(CO){PH(CH2CMe2)C6H2 Bu}] was ob-
served in these experiments. We noticed, however, that the
rate of this degradation was much lower than that observed in
toluene solution. For instance, the amount of phosphindole
3700
dx.doi.org/10.1021/ic2026945 | Inorg. Chem. 2012, 51, 3698−3706

Inorganic Chemistry
Article
for the mononuclear complex [Fe(CO)4{PHPh(OEt)}] (153.3
ppm).20c
The phosphinous acid complex 4 could only be characterized
through the IR and 31P NMR spectra of the crude reaction
mixture, because of its low stability. All attempts to isolate it as
a pure material resulted in its progressive decomposition. Its IR
spectrum in solution is comparable to that of compound 5
(Table 1), thus suggesting that the trans isomer is also the
major (or unique) species in solution. Its 31P NMR spectrum
exhibited only a broad doublet resonance at 171.3 ppm (1JPH ca.
345 Hz), not far from that of 5. The broadness of this
resonance, however, would prevent the detection of a second
isomer, if only present in small amounts.
Pathways for the Reactions of 1a with H2O and EtOH.
To account for the formation of the PHCy(OR) complexes 4
and 5 we can propose a first step in which the nucleophilic P
atom of the phosphinidene ligand in 1a would deprotonate a
molecule of H2O or EtOH to give the corresponding cationic
phosphide intermediates [Fe2Cp2(μ-CO)(μ-PCyH)(CO)2]-
[OR] (A in Scheme 1). These species were not detected
Figure 1. ORTEP diagram of compound cis-5, with H atoms (except
H1) omitted for clarity.
Table 2. Selected Bond Lengths (Å) and Angles (deg) for
Compound cis-5
Scheme 1
Fe(1)−Fe(2)
Fe(2)−P(1)
Fe(1)−C(1)
Fe(1)−C(2)
Fe(1)−C(3)
Fe(2)−C(2)
Fe(2)−C(3)
2.5112(8)
2.149(1)
1.744(4)
1.952(3)
1.947(3)
1.866(3)
1.887(3)
Fe(2)−C(2)−Fe(1)
Fe(2)−C(3)−Fe(1)
P(1)−Fe(2)−Fe(1)
C(1)−Fe(1)−Fe(2)
82.2(1)
81.8(1)
100.06(4)
103.1(1)
The spectroscopic data for compound 5 reveal the presence
of a solvent-dependent equilibrium mixture of cis and trans
isomers in solution (Chart 4 and Experimental Section), a
Chart 4
common feature for this type of compound. The IR spectrum
exhibits three C−O stretching bands with the expected pattern,
with the two bridging CO ligands in a transoid arrangement
giving rise to bands at 1769 (w) and 1729 (vs) cm−1, and the
terminal CO ligand yielding a strong band at 1935 cm−1. The
IR spectra of different compounds of the formula [Fe2Cp2(μ-
CO)2(CO)L] (L = phosphine,16 phosphite,16,17 κ1-
(EtO)2POP(EtO)218) exhibit a terminal C−O stretching
band in the range 1973−1951 cm−1 for the cis isomers, and
1950−1929 cm−1 for the trans isomers. On this basis we
propose a transoid arrangement of the FeCp(CO) fragments
for the major isomer of 5, as opposed to the conformation
found in the solid state (cis). The 31P NMR spectrum of 5
spectroscopically in the reaction mixtures, but the correspond-
ing cation has been previously isolated as the [BF4]− salt
(actually, that salt is the synthetic precursor of 1a).10b In the
case of salts A, however, the P atom in the cation would
undergo a fast nucleophilic attack of the alkoxide (or
hydroxide) counterion to give the corresponding PHCy(OR)-
bridged derivates [Fe2Cp2(μ-CO){μ-PCyH(OR)}(CO)2] (B),
these in turn rapidly rearranging into the more stable terminal
coordination mode, thus yielding the complexes 4 and 5. We
note that this pathway is different from that proposed for the
reactions of water or alcohols with transient electrophilic
phosphinidenes of Cr, Mo, W, and Fe mentioned above,20 the
latter involving first a nucleophilic attack of the OR− anion
1
exhibits two doublets, at 192.4 ppm (trans isomer, JPH = 337
1
Hz) and 189.9 ppm (cis isomer, JPH = 334 Hz) respectively.
These chemical shifts are reasonable for a phosphinite ligand
coordinated to an iron center, and not far from that measured
3701
dx.doi.org/10.1021/ic2026945 | Inorg. Chem. 2012, 51, 3698−3706

Inorganic Chemistry
Article
followed by the protonation of the (undetected) resulting
anionic phosphide intermediate. This difference of course
follows from the strong nucleophilic nature of the phosphini-
dene ligand in 1a. We finally note that a concerted mechanism
from A to 4/5 (i.e., not involving the intermediacy of B) cannot
be ruled out in any case.
Scheme 2
The limiting step in the reactions of 1a with H2O or EtOH
surely is the proton transfer step, because of the low acidity of
the H atoms involved, and a stronger acid would probably react
more easily. To check this hypothesis we tested the reaction
with HSPh. Indeed, just stoichiometric amounts of the thiol
were enough to react instantaneously with 1a at 273 K to give a
deep green solution containing the expected thiolophosphine
complex [Fe2Cp2(μ-CO)2(CO){PHCy(SPh)}], as deduced
from its IR (ν(CO) (CH2Cl2): 1938 (m), 1775 (w), 1735
(vs)) and 31P NMR spectra (δP 113.5 ppm, JHP = 348 Hz, in
CD2Cl2). This compound, however, was a rather unstable
species that could not be isolated or fully characterized because
of its progressive decomposition upon manipulation. A second
piece of evidence in favor of deprotonation as the limiting step
in the reactions of 1a under discussion is the observed increase
in the reaction rate when BEt3 is present. Presumably, this
effect would follow from the formation of small amounts of the
corresponding adducts R(H)O−BEt3, these in turn having O-
bound hydrogens of increased acidity. Although these adducts
appear to have been never reported in the literature, the
methanol−BEt3 adduct has been proposed as an intermediate
species in the methanolysis of BEt3 to give Et2B(OMe).27 In
fact, in the course of the reactions of 1a with excess H2O or
EtOH in the presence of BEt3, the alcoholysis or hydrolysis
products Et2B(OR) and EtB(OR)2 were detected by 11B NMR,
these giving rise to diagnostic resonances at about 55 ppm and
30 ppm respectively, which compare well to the values reported
in the literature for these or analogous compounds.27,28
We are aware of only one other report involving a borane-
mediated reaction of water with a phosphinidene complex, even
if the results are different. Thus, the ruthenium complexes
[Ru(η6-Ar)(PCy3)(PMes*)] (Ar = p-cymene, benzene) have
been reported to react with H2O in the presence of BPh3 to
give the corresponding phosphide species [Ru(η6-Ar)(PCy3)-
(PHMes*)][BPh3OH].29 For comparison purposes, we also
carried out the reaction of 1a with water in the presence of
BPh3. This led to a complex mixture containing the cation
[Fe2Cp2(μ-CO)(μ-PHCy)(CO)2]+ and other uncharacterized
products, as deduced from the analysis of the IR and 31P NMR
spectra of the reaction mixture. It thus seems that the
hydroxoborate ion [(OH)BPh3]− presumably formed in these
reactions would be a nucleophile weaker than its triethyl
counterpart, not strong enough to attack the cationic species A
formed in the first step of these reactions.
be thermally unstable.3 Intermediate C could then undergo an
homolytic fission of one of its Fe−P bonds, to give the
mononuclear radicals [FeCp(CO)2] and [FeCp-
(CO)2(PMes*)] (D), with the first of them rapidly dimerizing
to render [Fe2Cp2(CO)4]. In contrast, the phosphinidene
radical D, perhaps unable to dimerize rapidly because of steric
hindrance, might follow two mutually competitive reaction
pathways: (a) A thermal degradation to give the radical
[FeCp(CO)2] and free phosphinidene “PMes*” (path A in
Scheme 2), with these species rapidly dimerizing to yield more
[Fe2Cp2(CO)4] and the diphosphene Mes*PPMes* respec-
tively. (b) An H-atom abstraction (presumably at its P atom)
from traces of water present in the solvent (and in the gas
used) to give the phosphide complex 3 (path B in Scheme 2).
We note that H-atom abstraction is a typical reaction of
organometallic radicals,30 while the formation of the mentioned
diphosphene upon dimerization of the transient phosphinidene
PMes* has been also recognized previously.31 From the
observed product distribution it is obvious that the second
route is faster, accounting for about 85% of the evolution of the
radical D, a conclusion not surprising in view of the high
reactivity of organometallic radicals.30
The Role of a Terminal Isomer. Pathways in the
Hydrogenation Reaction of 1c. To account for the
formation of the phosphine derivative [Fe2Cp2(μ-CO)2(CO)-
(PH2Mes*)] upon hydrogenation of 1c we propose for the
latter complex the presence in solution of small amounts of a
terminal isomer T-1c in equilibrium with the dominant,
PMes*-bridged isomer B-1c (Scheme 3). This would be
analogous to the observed equilibrium in the corresponding
oxophosphinidene complex [Fe2Cp2(CO)3{P(O)Mes*}],10a
and perhaps it is also behind the thermal degradation of 1c
to give the phosphindole complex [Fe2Cp2(μ-CO)2(CO){PH-
Pathways in the Carbonylation Reaction of 1c. As
mentioned above, three products (the phosphide complex 3,
the dimer [Fe2Cp2(CO)4], and the diphosphene Mes*P
PMes*) were obtained in the room temperature carbonylation
of 1c, with little dependence on the reaction conditions
(solvent, reaction time). To account for the formation of all of
them we assume that the first step in this reaction would
involve the addition of a CO molecule to the dimetal center, to
give a transient (undetected) tetracarbonyl derivative
[Fe2Cp2(μ-PMes*)(CO)4] with no direct Fe−Fe interaction
(C in Scheme 2). Interestingly, the isostructural phenyl-
phosphinidene complex [Fe2Cp2(μ-PPh)(CO)4] has been
previously reported by Lorenz and co-workers, and found to
(CH2CMe2)C6H2 Bu2}].10b The isomer T-1c is expected to be
t
more reactive than its bridged isomer because of both the lower
steric hindrance around the P atom and its more favorable
electronic structure (see below), then becoming able to induce
the required homolytic cleavage of H2 at the P site. We have
3702
dx.doi.org/10.1021/ic2026945 | Inorg. Chem. 2012, 51, 3698−3706

Inorganic Chemistry
Article
Scheme 3
Table 3. Selected Bond Lengths (Å) and Angles (deg) for B-
1c and T-1c
parameter
B-1c
T-1c
Fe(1)−Fe(2)
Fe(1)−P(3)
Fe(1)−C(7)
Fe(1)−C(8)
Fe(1)−C(9)
Fe(2)−P(3)
Fe(2)−C(8)
Fe(2)−C(9)
Fe(2)−P(3)−C(Ar)
2.586
2.320
1.755
2.578
1.757
1.902
1.920
2.129
1.938
1.929
111.9
1.912
2.332
1.758
1.903
116.1
orbital (Figure 3). In agreement with this, an analysis of the
topology of the electron density (ρ) of the molecule under the
carried out DFT32 calculations for both isomers of 1c and also
for the phenylphosphinidene complex 1b, for comparative
purposes (see the Experimental Section and the Supporting
Information for further details).
The gas phase Gibbs energy for the isomerization B-1c to T-
1c was computed to be only 0.97 kcal mol−1 (Keq (B→T) about
0.19), this rising to 1.37 kcal mol−1 when the solvation
correction for toluene was applied (Keq ca. 0.10). In contrast,
the terminal isomer of the phenylphosphinidene complex 1b
was computed to be much higher in energy than its bridged
isomer in the gas phase (ΔGgas = 21.1 kcal mol−1; Keq (B→T) =
3.4 × 10−16). From this we conclude that the presence of small
amounts of the terminal isomer in the solutions of 1c is feasible
and mainly derived from the steric congestion induced by the
bulky Mes* group. Although we failed to detect the presence of
such a terminal isomer in the low-temperature 31P{1H} NMR
spectra of 1c we note that the anomalous upfield shift (ca. 20
ppm) of the resonance from 298 to 193 K would be consistent
with the above calculations, these predicting higher prevalence
of the bridged isomer (with much lower 31P chemical shift than
the terminal isomer) at lower temperatures, because of its lower
entropy content (ΔSgas = 9 cal mol−1 K−1).
Figure 3. DFT-computed HOMO (−5.15 eV, left), LUMO (−2.43
eV, right), and HOMO−4 (−6.33 eV, down) orbitals in T-1c.
Atoms in Molecules scheme (AIM)33 revealed a substantially
increased value of ρ at the Mo−P bond critical point (0.72 eÅ
−3
−3, to be compared with an average value of 0.50 eÅ in the
bridged isomer).
Recent work by Power and other researchers has proved the
ability of different types of main-group compounds (ER2, E2R2,
etc., with E = Si−Pb) to activate the single bonds of hydrogen
and related molecules.34 This transition-metal-like reactivity
relies on the presence in the main-group compounds of suitable
frontier orbitals, an empty one able to interact with the σ-
orbital of H2, and an occupied one able to provide back-
donation to the σ* orbital of H2. In the case of T-1c, this role
clearly would be played respectively by the lowest unoccupied
molecular orbital (LUMO) (largely located at P, with π*(Fe−
P) character) and the highest occupied molecular orbital
(HOMO) (also largely located at P, with dominant lone pair
character), this accounting for its hydrogenation to give the
phosphine derivative [Fe2Cp2(μ-CO)2(CO)(PH2Mes*)] in a
straightforward way. Interestingly, a recent report has shown
the ability of the mononuclear phosphinidene complex
[FeCp(CO)2(PNiPr2)][AlCl4] to activate the Si−H bond of
different silanes under mild conditions, to give the correspond-
ing silylphosphine derivatives.35
The DFT-optimized geometries for B-1c and T-1c (Figure
2) display unremarkable geometric parameters (Table 3) except
for the Fe−P distance in the terminal isomer, about 0.2 Å
shorter than the Fe−P separations in the bridged isomer,
therefore indicative of a significant multiplicity in that bond.
Indeed the presence of a small π component in the Fe−P bond
of T-1c can be grasped through the nature of its HOMO−4
CONCLUDING REMARKS
■
The ability of the supermesitylphosphinidene complex 1c to
react at room temperature with H2 at a single P center, a unique
behavior to date, is attributed to the presence in solution of
Figure 2. DFT optimized geometries of B-1c (left) and T-1c (right).
3703
dx.doi.org/10.1021/ic2026945 | Inorg. Chem. 2012, 51, 3698−3706

Inorganic Chemistry
Article
product distribution was reached when using dichloromethane as
solvent.
small amounts of an isomer bearing a terminal phosphinidene
ligand, estimated to be in toluene solution only 1.4 kcal mol−1
less stable than the major, PMes*-bridged isomer, according to
DFT calculations. This is ultimately a steric effect, since the
analogous phenylphosphinidene complex 1b (inert toward H2)
exists only in the bridged form. The presence of CO induces
the cleavage of the Fe−Fe bond of 1c, to give the mononuclear
phosphide complex [FeCp(CO)2(PHMes*)] as major product,
this following from an H-atom abstraction by the radical
[FeCp(CO)2(PMes*)] presumably generated in the process. In
contrast to the above behavior, the cyclohexylphosphinidene
complex 1a, which fails to react with H2 presumably because of
the absence of significant amounts of the terminal isomer in
solution, is able to activate the polar O−H bonds of H2O or
EtOH to give the corresponding derivatives [Fe2Cp2(μ-
CO)2(CO){PH(OR)Mes*}]. This, however, takes place via a
two step mechanism initiated with the deprotonation of the O-
bound hydrogen by the basic P atom of the phosphinidene
ligand, followed by a nucleophilic attack of the OR− anion at
the P atom in the cationic phosphide complex first formed.
Preparation of [FeCp(CO)2(PH2Mes*)](BF4) (2). A mixture of
[Fe2Cp2(CO)4] (0.035 g, 0.099 mmol), [FeCp2]BF4 (0.055 g, 0.202
mmol) and PH2Mes* (0.055 g, 0.198 mmol) in dichloromethane (7
mL) was stirred for 4 h at room temperature to give a green solution.
The solvent was then removed under vacuum and the green residue
was washed with diethyl ether (4 × 6 mL) to give compound 2 as a
green microcrystalline solid (0.060 g, 56%). Anal. Calcd for
1
C25H36BF4FeO2P: C, 55.38; H, 6.69. Found: C, 55.26; H, 6.88. H
NMR: δ 7.54 (d, JHP = 4, 2H, C6H2), 6.71 (d, JHP = 392, 2H, PH2),
5.09 (d, JHP = 1, 5H, Cp), 1.57 (s, 18H, o-tBu), 1.31 (s, 9H, p-tBu).
Preparation of [FeCp(CO)2(PHMes*)] (3). Solid NaOH (ca. 0.1
g, excess) was added to a solution of compound 2 (0.060 g, 0.111
mmol) in dichloromethane (7 mL), and the mixture was stirred for 5
min to give a dark orange solution. The solvent was then removed
under vacuum, and the residue was dissolved in the minimum amount
of petroleum ether and chromatographed on alumina (activity IV) at
288 K. Elution with a dichloromethane/petroleum ether (1/9) mixture
gave an orange fraction yielding, upon removal of solvents under
vacuum, compound 3 as a dark orange microcrystalline solid (0.030 g,
60%). Anal. Calcd for C25H35FeO2P: C, 66.09; H, 7.76. Found: C,
66.35; H, 7.90. 1H NMR (C6D6): δ 7.43 (s, 2H, C6H2), 5.02 (d, JHP
=
t
179, 1H, PH), 3.87 (s, 5H, Cp), 1.88, 1.73, 1.30 (3s, 3 × 9H, Bu).
Preparation of [Fe2Cp2(μ-CO)2(CO){PHCy(OH)}] (4). A 1 M
solution of BEt3 in hexane (300 μL, 0.3 mmol) and degassed water
(100 μL, 5.55 mmol) were added to a solution of compound 1a (ca.
0.28 mmol) in dichloromethane (10 mL). The reaction mixture was
stirred for 45 min to give a brown solution containing compound 4 as
the major compound. All the attempts to isolate it as a pure species led
invariably to its progressive decomposition.
EXPERIMENTAL SECTION
■
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.36
Petroleum ether refers to that fraction distilling in the range 338−
343 K. Compounds [FeCp2]BF4,37 and PH2Mes*,38 were synthesized
according to literature procedures. Solutions of the phosphinidene
complexes [Fe2Cp2(μ-PR)(μ-CO)(CO)2] (R = Cy (1a), Mes* (1c))
were prepared in situ as described previously.10b All other reagents
were obtained from the usual commercial suppliers and used as
received. Chromatographic separations were carried out using jacketed
columns cooled by tap water (ca. 288 K). Commercial aluminum
oxide (activity I, 150 mesh) was degassed under vacuum prior to use.
The latter was afterward mixed under nitrogen with the appropriate
amount of water to reach the activity desired. IR C−O 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), 121.50 (31P{1H}), 75.48 (13C{1H}) or 121.52 MHz
(11B) at 290 K in CD2Cl2 solutions unless otherwise stated. Chemical
shifts (δ) are given in parts per million (ppm), relative to internal
tetramethylsilane (1H, 13C), external 85% aqueous H3PO4 (31P) or
external BF3·OEt2 (11B). Coupling constants (J) are given in hertz
(Hz).
Preparation of [Fe2Cp2(μ-CO)2(CO){PHCy(OEt)}] (5). A 1 M
solution of BEt3 in hexane (200 μL, 0.2 mmol) was added to an
ethanol suspension (7 mL) of compound 1a (ca. 0.2 mmol), and the
mixture was stirred for 3 d at room temperature to give a brown
solution containing compounds 5 and 4 (ratio 5/4: 3/1, as determined
from the 31P{1H} NMR spectrum of the mixture). After removal of the
solvent under vacuum, the residue was dissolved in dichloromethane/
petroleum ether (1/9), and the extract was chromatographed on
alumina (activity IV) at 288 K. Elution with dichloromethane/
petroleum ether (1/2) gave a brown fraction yielding, upon removal of
solvents under vacuum, compound 5 as a red-brownish microcrystal-
line solid (0.040 g, 41%). This compound exists in solution as an
equilibrium mixture of the corresponding cis and trans isomers, with
the trans/cis ratio being 10/1 in CD2Cl2 and 13/1 in C6D6, as
determined from the 31P{1H} NMR spectra. Brown crystals suitable
for an X-ray analysis of the cis isomer were grown from a concentrated
solution of the mixture of isomers in petroleum ether at 253 K. Anal.
Calcd for C21H27Fe2O4P: C, 51.89; H, 5.60. Found: C, 51.86; H, 5.93.
Spectroscopic data for trans-5: 1H NMR: δ 4.82 (d, JHP = 337, 1H, PH),
4.72, 4.52 (2s, 2 × 5H, Cp), 3.94 (dq, JHH = 15, 7, 1H, CH2), 3.74 (dq,
JHH = 15, 7, 1H, CH2), 2.2−1.0 (m, 11H, Cy), 1.22 (t, JHH = 7, 3H,
Me). The proton resonances of the minor isomer cis-5 were masked
by those of the major isomer.
X-ray Structure Determination of Compound 5. The X-ray
intensity data for cis-5 were collected on a Smart-CCD-1000 Bruker
diffractometer using graphite-monochromated MoKα radiation at 120
K. Cell dimensions and orientation matrices were initially determined
from least-squares refinements on reflections measured in 3 sets of 30
exposures collected in 3 different ω regions and eventually refined
against all reflections. The software SMART39 was used for collecting
frames of data, indexing reflections, and determining lattice
parameters. The collected frames were then processed for integration
by the software SAINT,39 and a multiscan absorption correction was
applied with SADABS.40 Using the program suite WinGX,41 the
structure was solved by Patterson interpretation and phase expansion
using SHELXL97,42 and refined with full-matrix least-squares on F2
using SHELXL97. All non-hydrogen atoms were refined anisotropi-
cally. All hydrogen atoms were geometrically placed and refined using
a riding model except for H(1), which was located in the Fourier Maps
Hydrogenation of Compound 1c. In a typical experiment, a
Schlenk tube equipped with a Young's valve containing a solution of
compound 1c (ca. 0.05 mmol) in toluene (5 mL) was frozen with
liquid N2. The tube was degassed under vacuum, and then refilled with
H2. The valve was then closed, and the mixture was allowed to reach
room temperature (PH2 ca. 4 atm). The solution was then stirred for
16 h to yield a brown-greenish solution containing [Fe2Cp2(μ-
CO)2(CO)(PH2Mes*)],10b as an equilibrium mixture of the cis and
trans isomers (50%), PH2Mes* (20%), [Fe2Cp2(CO)4] (15%),
[Fe2Cp2(μ-CO)2(CO){P(O)Mes*}] (10%),10a and [Fe2Cp2(μ-
10b
t
CO)2(CO){PH(CH2CMe2)C6H2 Bu2}] (5%).
The relative ratio
of the above compounds in the reaction mixture was determined
1
through H NMR measurements in C6D6.
Carbonylation of Compound 1c. In a typical experiment, a
Schlenk tube equipped with a Young's valve containing a solution of
compound 1c (ca. 0.05 mmol) in toluene (5 mL) was frozen with
liquid N2. The tube was degassed under vacuum, and then refilled with
CO. The valve was then closed, and the mixture was allowed to reach
room temperature (PCO ca. 4 atm). The solution was then stirred for
16 h to yield an orange solution containing [FeCp(CO)2(PHMes*)]
(3) (55%), [Fe2Cp2(CO)4] (40%), and Mes*PPMes* (5%), as
1
determined through the H NMR spectra of the mixture. The same
3704
dx.doi.org/10.1021/ic2026945 | Inorg. Chem. 2012, 51, 3698−3706

Inorganic Chemistry
Article
and refined isotropically. Crystallographic data and structure refine-
ment details for cis-5 are collected in Table 4.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: ara_12_79@hotmail.com (A.R.), mara@uniovi.es
Table 4. Crystal Data for Compound cis-5
(M.A.R.).
cis-5
Notes
mol formula
mol wt
C21H27Fe2O4P
The authors declare no competing financial interest.
486.1
cryst syst
orthorhombic
ACKNOWLEDGMENTS
■
space group
radiation (λ, Å)
a, Å
Pbca
We thank the DGI of Spain (Project CTQ2009-09444) and the
COST action CM0802 “PhoSciNet” for supporting this work.
0.71073
15.293(4)
b, Å
11.795(3)
c, Å
23.293(6)
REFERENCES
■
α, deg
90
(1) Reviews: (a) Aktas, H.; Slootweg, J. C.; Lammerstma, K. Angew.
Chem., Int. Ed. 2010, 49, 2102. (b) Waterman, R. Dalton Trans. 2009,
18. (c) Mathey, F. Dalton Trans. 2007, 1861. (d) Lammertsma, K. Top.
Curr. Chem. 2003, 229, 95. (e) Streubel, R. Top. Curr. Chem. 2003,
223, 91. (f) Mathey, F. Angew. Chem., Int. Ed. 2003, 42, 1578.
(g) Lammertsma, K.; Vlaar, M. J. M. Eur. J. Org. Chem. 2002, 1127.
(h) Mathey, F.; Tran-Huy, N. H.; Marinetti, A. Helv. Chim. Acta 2001,
84, 2938. (i) Stephan, D. W. Angew. Chem., Int. Ed. 2000, 39, 314.
(j) Shah, S.; Protasiewicz, J. D. Coord. Chem. Rev. 2000, 210, 181.
(k) Schrock, R. R. Acc. Chem. Res. 1997, 30, 9. (l) Cowley, A. H. Acc.
Chem. Res. 1997, 30, 445. (m) Cowley, A. H.; Barron, A. R. Acc. Chem.
Res. 1988, 21, 81. (n) Huttner, G.; Knoll, K. Angew. Chem., Int. Ed.
Engl. 1987, 26, 743. (o) Huttner, G.; Evertz, K. Acc. Chem. Res. 1986,
19, 406.
β, deg
90
γ, deg
90
V, Å3
4202(2)
Z
8
calcd density, g cm−3
absorp coeff, mm−1
temperature, K
θ range (deg)
index ranges (h, k, l)
no. of reflns collected
no. of indep reflns (Rint)
reflns with I > 2σ(I)
R indexes [data with I > 2σ(I)]
1.537
1.481
120(2)
1.75−26.37
0, 19; 0, 14; 0, 29
35866
4290 (0.0699)
3062
a
b
R1 = 0.0368; wR2 = 0.0761
́
(2) Alvarez, C. M.; Alvarez, M. A.; García, M. E.; Gonzalez, R.; Ruiz,
M. A.; Hamidov, H.; Jeffery, J. C. Organometallics 2005, 24, 5503.
(3) (a) Klasen, C.; Effinger, G.; Schmid, S.; Lorenz, I.-P. Z.
a
b
R indexes (all data)
R1 = 0.073; wR2 = 0.0943
GOF
1.121
no. of restraints/params
Δρ(max., min.), e Å−3
0/258
Naturforsch B 1993, 48, 705. (b) Lorenz, I.-P.; Pohl, W.; Noth, H.;
Schmidt, M. J. Organomet. Chem. 1994, 475, 211. (c) Lorenz, I.-P.;
̈
0.497, −0.44
a
Murschel, P.; Pohl, W.; Polborn, K. Chem. Ber. 1995, 128, 413.
̈
2
R = ∑||Fo| − |Fc||/∑|Fo|. wR = [∑w(Fo − Fc2)2/∑|Fo|2]1/2. w = 1/
b
(4) (a) King, R. B. J. Organomet. Chem. 1998, 557, 29. (b) King, R.
B.; Bitterwolf, T. E. Coord. Chem. Rev. 2000, 206−207, 563.
(5) Lang, H.; Zsolnai, L.; Huttner, G. Chem. Ber. 1985, 118, 4426.
(6) Treichel, P. M.; Douglas, W. M.; Dean, W. K. Inorg. Chem. 1972,
11, 1615.
2
2
[σ2(Fo ) + (aP)2 + bP] where P = (Fo + 2Fc2)/3. a = 0.0276, b =
10.7231.
DFT Analysis of Phosphinidene Complexes. All computations
described in this work were carried out with the GAUSSIAN03
package,43 in which the hybrid method B3LYP was applied with the
Becke three parameters exchange functional44 and the Lee−Yang−
Parr correlation functional.45 Effective core potentials (ECP) and their
associated double-ζ LANL2DZ basis set were used for the iron
centers.46 The light elements (O, C, P, and H) were described with 6-
31G* basis.47 Geometry optimizations were performed under no
symmetry restrictions, using initial coordinates derived from X-ray data
of the same or comparable complexes, and frequency analyses were
performed to ensure that a minimum structure with no imaginary
frequencies was achieved. For interpretation purposes, Mulliken
charges were computed as usual,48 and natural population analysis
(NPA) charges were derived from the natural bond order (NBO)
analysis of the data.49 Molecular orbitals and vibrational modes were
visualized using the Molekel program.50 The topological analysis of ρ
was carried out with the Xaim routine.51 Solvent effects (toluene, ε =
2.3741) were modeled using the polarized-continuum-model
(PCM),52 and the B3LYP optimized structures.
(7) (a) Flynn, K. M.; Bartlett, R. A.; Olmstead, M. M.; Power, P. P.
Organometallics 1986, 5, 813. (b) Bartlett, R. A.; Dias, H. V. R.; Flynn,
K. M.; Olmstead, M. M.; Power, P. P. J. Am. Chem. Soc. 1987, 109,
5699.
(8) Bai, G.; Wei, P.; Das, A. K.; Stephan, D. W. Dalton Trans. 2006,
1141.
(9) (a) Alvarez, M. A.; García, M. E.; Gonzal
Organometallics 2008, 27, 1037. (b) Alvarez, M. A.; García, M. E.;
Gonzalez, R.; Ruiz, M. A. Organometallics 2010, 29, 5140. (c) Alvarez,
M. A.; García, M. E.; Gonzalez, R.; Ramos, A.; Ruiz, M. A. Inorg. Chem.
2011, 50, 7894.
(10) (a) Alvarez, M. A.; García, M. E.; Gonzal
M. A. Organometallics 2010, 29, 1875. (b) Alvarez, M. A.; García, M.
E.; Gonzalez, R.; Ramos, A.; Ruiz, M. A. Organometallics 2011, 30,
1102.
́
ez, R.; Ruiz, M. A.
́
́
́
ez, R.; Ramos, A.; Ruiz,
́
(11) (a) Stephan, D. W. Org. Biomol. Chem. 2008, 6, 1535.
(b) Stephan, D. W. Dalton Trans. 2009, 3129. (c) Stephan, D. W.
Chem. Commun. 2010, 8526. (d) Stephan, D. W.; Erker, G. Angew.
Chem., Int. Ed. 2010, 49, 46.
(12) Frey, G. D.; Lavallo, V.; Donnadieu, B.; Schoeller, W. W.;
Bertrand, G. Science 2007, 316, 439.
(13) Duffy, M. P.; Ting, L. Y.; Nicholls, L.; Li, Y.; Ganguly, R.;
Mathey, F. Organometallics 2012 (in press, DOI: 10.1021/
om2009974).
ASSOCIATED CONTENT
■
S
* Supporting Information
(14) (a) Yoshifuji, M.; Shima, I.; Inamoto, N.; Hirotsu, K.; Higuchi,
T. J. Am. Chem. Soc. 1981, 103, 4587. (b) Yoshifuji, M.; Shima, I.;
Inamoto, N.; Hirotsu, K.; Higuchi, T. J. Am. Chem. Soc. 1982, 104,
6167.
Figures and tables of data from the DFT calculations (pdf), and
a CIF file giving the crystallographic data for the structural
analysis of compound cis-5. This material is available free of
charge via the Internet at http://pubs.acs.org.
(15) Schumann, H. J. Organomet. Chem. 1985, 290, C34.
3705
dx.doi.org/10.1021/ic2026945 | Inorg. Chem. 2012, 51, 3698−3706

Inorganic Chemistry
Article
(16) See for example: (a) Malisch, W.; Thirase, K.; Reising, J. Z.
Naturforsch B 1998, 53, 1084. (b) Malisch, W.; Gunzelmann, N.;
Thirase, K.; Neumayer, M. J. Organomet. Chem. 1998, 571, 215.
(c) Malisch, W.; Thirase, K.; Reising, J. J. Organomet. Chem. 1998, 568,
247. (d) Malisch, W.; Thirase, K.; Rehmann, F.-J.; Reising, J.;
Gunzelmann, N. Eur. J. Inorg. Chem. 1998, 1589.
(17) Braterman, P. S. Metal Carbonyl Spectra; Academic Press:
London, U. K., 1975.
(18) Jameson, C. J. In Phosphorus-31 NMR Spectroscopy in
Stereochemical Analysis; Verkade, J. G., Quin, L. D., Eds.; VCH: New
York, 1987; Chapter 6.
Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;
Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.;
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J.
B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R.
E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;
Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J.
J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.;
Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,
J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; and Pople, J. A. Gaussian 03, Revision
C.02; Gaussian, Inc.: Wallingford, CT, 2004.
(19) Amor, I.; García, M. E.; Ruiz, M. A.; Sae
Jeffery, J. C. Organometallics 2006, 25, 4857.
́
z, D.; Hamidov, H.;
(20) (a) Marinetti, A.; Mathey, F.; Fischer, J.; Mitschler, A. J. Am.
Chem. Soc. 1982, 104, 4484. (b) Marinetti, A.; Mathey, F.
Organometallics 1982, 1, 1488. (c) Luo, W.; Ciric, A.; Tian, R.;
Mathey, F. Organometallics 2010, 29, 1862.
(21) Cotton, F. A.; Frenz, B. A.; White, A. J. Inorg. Chem. 1974, 13,
1407.
(22) Klasen, C.; Lorenz, I.-P.; Schmid, S.; Beuter, G. J. Organomet.
Chem. 1992, 428, 363.
(23) Bitta, J.; Fassbender, S.; Reiss, G.; Ganter, C. Organometallics
(44) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(45) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
(46) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299.
(47) (a) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28,
213. (b) Petersson, G. A.; Al-Laham, M. A. J. Chem. Phys. 1991, 94,
6081. (c) Petersson, G. A.; Bennett, A.; Tensfeldt, T. G.; Al-Laham, M.
A.; Shirley, W. A.; Mantzaris, J. J. Chem. Phys. 1988, 89, 2193.
(48) Mulliken, R. S. J. Chem. Phys. 1955, 23, 1833.
(49) Reed, A. E.; Curtis, L. A.; Weinhold, F. Chem. Rev. 1988, 88,
899.
(50) MOLEKEL: An Interactive Molecular Graphics Tool:
2006, 25, 2394.
(24) Zhang, S.; Brown, T. L. Organometallics 1992, 11, 4166.
(25) Haines, R. J.; Du Preez, A. L. Inorg. Chem. 1969, 8, 1459.
(26) Alvarez, C. M.; Galan, B.; García, M. E.; Riera, V.; Ruiz, M. A.;
́
Bois, C. Organometallics 2003, 22, 3039.
(27) Chen, K.-M.; Gunderson, K. G.; Hardtmann, G. E.; Prasad, K.;
Repic, O.; Shapiro, M. J. Chem. Lett. 1987, 1923.
Portmann, S.; Luthi, H. P. Chimia 2000, 54, 766.
̈
(51) Ortiz, J. C.; Bo, C. Xaim; Departamento de Quimica Fisica e
́
Inorganica, Universidad Rovira i Virgili: Tarragona, Spain, 1998.
(52) (a) Cossi, M.; Barone, V.; Cammi, R.; Tomasi, J. Chem. Phys.
Lett. 1996, 255, 327. (b) Fortunelli, A.; Tomasi, J. Chem. Phys. Lett.
1994, 231, 34. (c) Tomasi, J.; Persico, M. Chem. Rev. 1994, 94, 2027.
(d) Floris, F.; Tomasi, J. J. Comput. Chem. 1989, 10, 616. (e) Pascual-
Ahuir, J. L.; Silla, E.; Tomasi, J.; Bonaccorsi, R. J. J. Comput. Chem.
1987, 8, 778. (f) Mieritus, S.; Tomasi, J. J. Chem. Phys. 1982, 65, 239.
(g) Mieritus, S.; Scrocco, E.; Tomasi, J. J. Chem. Phys. 1981, 55, 117.
(28) (a) Fenzl, W.; Koster, R. Inorg. Synth. 1983, 22, 193. (b) Koster,
R.; Idelmann, P. Inorg. Synth. 1986, 24, 83.
(29) Menye-Biyogo, R.; Delpech, F.; Castel, A.; Pimienta, V.;
̈
̈
Gornitzka, H.; Riviere, P. Organometallics 2007, 26, 5091.
̀
(30) (a) Paramagnetic Organometallic Species in Activation Selectivity,
Catalysis; Chanon, M., Julliard, M., Poite, J. C., Eds.; Kluwer Academic
Publishers: Dordrecht, The Netherlands, 1989. (b) Organometallic
Radical Processes; Trogler, W. C., Ed.; Elsevier: Amsterdam, The
Netherlands, 1990. (c) Astruc, D. Electron Transfer and Radical
Processes in Transition-Metal Chemistry; VCH: New York, 1995.
(31) Smith, R. C.; Shah, S.; Protasiewicz, J. D. J. Organomet. Chem.
2002, 646, 255 , and references therein.
(32) (a) Koch, W.; Holthausen, M. C. A Chemist’s Guide to Density
Functional Theory, 2nd ed.; Wiley-VCH: Weinheim, Germany, 2002.
(b) Ziegler, T. Chem. Rev. 1991, 91, 651. (c) Foresman, J. B.; Frisch,
Æ. Exploring Chemistry with Electronic Structure Methods, 2nd ed.;
Gaussian, Inc.: Pittsburg, PA, 1996.
(33) Bader, R. F. W. Atoms in molecules-A Quantum Theory; Oxford
University Press: Oxford, U.K., 1990.
(34) (a) Power, P. P. Acc. Chem. Res. 2011, 44, 627. (b) Power, P. P.
Nature 2010, 463, 171. (c) Li, J.; Schenk, C.; Goedecke, C.; Frenking,
G.; Jones, C. J. Am. Chem. Soc. 2011, 133, 18622 , and references
therein.
(35) Vaheesar, K.; Bolton, T. M.; East, A. L. L.; Sterenberg, B. T.
Organometallics 2010, 29, 484.
(36) Armarego, W. L. F.; Chai, C. Purification of Laboratory
Chemicals, 5th ed.; Butterworth-Heinemann: Oxford, U.K., 2003.
(37) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877.
(38) Cowley, A. H.; Norman, N. C.; Pakulski, M. Inorg. Synth. 1990,
27, 235.
(39) SMART & SAINT Software Reference Manuals, Version 5.051
(Windows NT Version); Bruker Analytical X-ray Instruments:
Madison, WI, 1998.
(40) Sheldrick, G. M. SADABS, Program for Empirical Absorption
Correction; University of Gottingen: Gottingen, Germany, 1996.
̈
̈
(41) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837.
(42) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112.
(43) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, Jr., J. A.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
3706
dx.doi.org/10.1021/ic2026945 | Inorg. Chem. 2012, 51, 3698−3706