P-H activation of secondary phosphanes on a parent amido diiridium complex

Article abstract of DOI:10.1039/c3dt52911h

Selected secondary phosphanes (H-PR₂; R = Ph, Cy, ⁱPr) smoothly react with a parent amido-bridged diiridium cyclooctadiene complex affording mixed amido/(bis)phosphido dinuclear species. A careful investigation of the reaction profile, carried out by experimental and theoretical tools, revealed that, after an initial amido/phosphido exchange, at low temperatures a second molecule of secondary phosphane adds to the dinuclear system through an oxidative addition process leading to a hydrido amido/bis(phosphido) mixed-valence complex [Ir^III/Ir^I]. These species rearrange above -10 °C into the most stable isomer that arises from a migration of the hydrido moiety to one of the CH fragments of a coordinated cod molecule, a transformation facilitated by the formation of an intermetallic bond. Further heating of these species reductively eliminates ammonia affording bis(phosphido)-metal-metal bonded complexes. The Royal Society of Chemistry.

Full text of DOI:10.1039/c3dt52911h

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
P–H activation of secondary phosphanes on a
parent amido diiridium complex†‡
Cite this: Dalton Trans., 2014, 43,
1609
Inmaculada Mena,a Miguel A. Casado,*a Víctor Polo,b Pilar García-Orduña,a
Fernando J. Lahoza and Luis A. Oro*a,c
i
Selected secondary phosphanes (H–PR2; R = Ph, Cy, Pr) smoothly react with a parent amido-bridged
diiridium cyclooctadiene complex affording mixed amido/(bis)phosphido dinuclear species. A careful investi-
gation of the reaction profile, carried out by experimental and theoretical tools, revealed that, after an
initial amido/phosphido exchange, at low temperatures a second molecule of secondary phosphane adds
to the dinuclear system through an oxidative addition process leading to a hydrido amido/bis(phosphido)
mixed-valence complex [IrIII/IrI]. These species rearrange above −10 °C into the most stable isomer that
arises from a migration of the hydrido moiety to one of the vCH fragments of a coordinated cod mole-
cule, a transformation facilitated by the formation of an intermetallic bond. Further heating of these
species reductively eliminates ammonia affording bis(phosphido)-metal–metal bonded complexes.
Received 15th October 2013,
Accepted 18th October 2013
DOI: 10.1039/c3dt52911h
www.rsc.org/dalton
dinuclear assemblies by bridging both metals quite tightly.6
Late transition metal parent–amido complexes are extremely
Introduction
Metal–metal interactions on dinuclear complexes often rely on rare; however, they are emerging as quite relevant species in
the architecture of the bridging ligands that support the met- the most recent organometallic chemistry. Their interest
allic assemblies, that is, their size and flexibility.1 While the emerges from two distinct perspectives: (i) the activation of
size of a given bridging ligand establishes the proximity of the ammonia as a source of amido ligands;7,8 under this idea, new
metal centres, its flexibility is a key factor determining the imaginative methodologies that utilize raw ammonia directly
possibility of approach between the metals and the establish- to prepare amido complexes have been reported, an approach
ment of intermetallic interactions, so that cooperative reactiv- that may eventually drive metal-mediated functionalization of
ity may in principle be an issue to consider.2 In this context, ammonia;9 and (ii) the non-innocent behaviour of the amido
the ability of –SR,3 –OR4 and –NHR5 moieties to efficiently ligands, a phenomenon that is beginning to be disclosed and
behave as bridging ligands towards late metals is known, as influences the reactivity and catalytic properties of amido com-
they are highly flexible to allow a wide range of intermetallic plexes.10 In this line, a significant contribution has been
separations, a situation that opens the possibility of metal– recently reported dealing with the role of [Ir–NH2–Ir] linkages
metal bonding.
in catalytic transfer hydrogenation, indicating a novel binuc-
Among the series of these heteroatom-based anions, the lear, outer-sphere Noyori-like mechanism, which illustrates the
simplest nitrogen-based one is the parent-amido (–NH2) key role of NH2 groups in alcohol dehydrogenation and imido
ligand, which has drawn a great deal of attention in the last formation.11
few years and is able to stabilize a number of different
Herein we report on the activation of P–H bonds of selected
secondary phosphanes by a parent-amido bridged diiridium
cyclooctadiene complex, which leads to rare parent amido/
phosphido-bridged complexes and ultimately to bis(phos-
phido) bridged metal–metal bonded complexes under thermal
conditions, on extrusion of ammonia.
aInstituto de Síntesis Química y Catálisis Homogénea ISQCH, Universidad de
Zaragoza-CSIC, C/Pedro Cerbuna, 12, 50009 Zaragoza, Spain.
E-mail: mcasado@unizar.es, oro@unizar.es; Fax: +34 976 761 187;
Tel: +34 976 761 143
bDepartamento de Química Física e Instituto de Biocomputación y Física de Sistemas
Complejos (BIFI), Universidad de Zaragoza, 50009 Zaragoza, Spain
cCenter for Refining & Petrochemicals King Fahd University of Petroleum & Minerals,
Results and discussion
Dhahran, 31261, Saudi Arabia
†Dedicated to the memory of Professor María Pilar García Clemente.
Treatment of the parent amido complex [{Ir(µ-NH2)(cod)}2] (1)
with two molar equiv. of diphenylphosphane afforded the
mixed bis(phosphido) amido-bridged diiridium complex
‡Electronic supplementary information (ESI) available: Additional experimental
and full computational details (PDF). CCDC 951633–951635. For ESI and crystallo-
graphic data in CIF or other electronic format see DOI: 10.1039/c3dt52911h
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1609–1619 | 1609

View Article Online
Paper
Dalton Transactions
[(1-κ-4,5-η2-C8H13)Ir(µ-NH2)(µ-PPh2)2Ir(1,2,5,6-η4-cod)] (3) (cod =
1,5-cyclooctadiene) in an excellent yield as a deep red, air-
sensitive microcrystalline solid. Complex 3 can be formally
described as the result of an exchange of an amido group by a
phosphido ligand in 1 plus the net addition of diphenylphos-
phane to a postulated mixed phosphido/amido-bridged di-
iridium system. Accordingly, with this approach, other selected
secondary phosphanes H–PR2 reacted in a similar manner
to 1, affording complexes [(1-κ-4,5-η2-C8H13)Ir(µ-NH2)(µ-PR2)2Ir-
i
(1,2,5,6-η4-cod)] (R = Cy (4), Pr (5)) as red solids in acceptable
yields, which were characterized in solution by multinuclear
NMR techniques, including 1H–15N HMQC spectroscopy and
mass spectrometry.
Fig. 1 Crystal structure of complex 2. Only hydrogen atoms for the
hydride and the amido ligand have been shown, and the phenyl rings of
the phosphanes have been omitted for clarity. Selected bond lengths [Å]
and angles [°]: Ir(1)–P(1) 2.3764(10), Ir(1)–P(2) 2.3669(10), Ir(2)–P(1)
2.4190(10), Ir(2)–P(2) 2.3937(10), Ir(1)–N 2.175(3), Ir(2)–N 2.131(3), mean
Ir–Ct 2.060(2), Ir(1)–H 1.608(19), P(1)–Ir(1)–P(2) 75.56(4), P(1)–Ir(2)–P(2)
74.28(3), P(1)–Ir(1)–N 72.87(10), P(2)–Ir(1)–N 75.32(10), P(1)–Ir(2)–N
72.73(10), P(2)–Ir(2)–N 75.53(10), P(1)–Ir(1)–Ct(2) 172.67(13), P(2)–Ir(1)–
Ct(1) 177.25(11), P(1)–Ir(2)–Ct(3) 136.50(13), P(2)–Ir(2)–Ct(4) 106.21(13),
N–Ir(2)–Ct(4) 177.64(16), N–Ir(1)–H 156.5(18). Ct(1), Ct(2), Ct(3) and
Ct(4) represent the midpoints of the olefinic C(1)–C(2), C(5)–C(6), C(9)–
C(10) and C(13)–C(14) bonds, respectively.
In order to gain insight into the mechanism of the for-
mation of 3–5, we monitored the reaction of 1 with diphenyl-
phosphane at various temperatures by NMR techniques. We
found that below −40 °C there is a clean and quantitative
formation of new hydrido phosphido/amido diiridium species
[(η4-cod)(H)Ir(µ-NH2)(µ-PPh2)2Ir(η4-cod)] (2) along with release
of one molar equiv. of ammonia. Above −10 °C, complex 2 is
gradually transformed into 3 (Scheme 1).
Both dinuclear complexes 2 and 3 were fully characterized
in solution by multinuclear NMR techniques and in the solid
state by X-ray diffraction studies (see below). While deep-red
crystals of 3 were obtained from concentrated solutions of the
complex in toluene and hexane mixtures, the isolation and
measurement of an orange monocrystal of 2 required careful
manipulation at low temperatures (below −40 °C) throughout
the process.
Fig. 1 and 2 show the molecular structures of complexes 2
and 3, both having dinuclear cores, with the metals bridged by
two diphenylphosphido ligands located at opposite sides of
the iridium–iridium axes, and an amido ligand. Although both
dinuclear species share the same chemical composition, there
are remarkable structural differences between them. While
complex 2 displays a clearly non-bonding intermetallic separ-
ation (Ir1⋯Ir2 3.2928(3) Å), 3 shows a very short Ir–Ir bond
Fig. 2 Crystal structure of complex 3. Only hydrogen atoms for the
hydride and the amido ligand have been shown, and the phenyl rings of
the phosphanes have been omitted for clarity. Selected bond lengths [Å]
and angles [°]: Ir(1)–P(1) 2.3321(15), Ir(2)–P(1) 2.3459(15), Ir(1)–P(2)
2.3067(15), Ir(2)–P(2) 2.3841(15), Ir(1)–N 2.180(5), Ir(2)–N 2.125(5), Ir(1)–
(Ir1–Ir2 2.6795(3) Å), favoured by the flexibility of the phos- Ct(1) 1.993(6), Ir(2)–Ct(2) 1.993(6), Ir(1)–C(6A) 2.098(9), Ir(2)–Ct(3)
2.089(9), Ir(1)–Ir(2) 2.6795(3), P(1)–Ir(1)–P(2) 109.59(5), P(1)–Ir(2)–P(2)
106.51(5), P(1)–Ir(1)–N 74.86(15), P(1)–Ir(2)–N 70.58(14), P(1)–Ir(1)–Ct(1)
phido and amido ligands in these dinuclear systems. This pli-
ability is clearly seen in the central Ir2P2N cores. Differences in
119.0(2), P(1)–Ir(2)–Ct(2) 131.8(2), P(1)–Ir(2)–Ct(3) 101.8(2), P(2)–Ir(1)–N
76.05(14), P(2)–Ir(2)–N 75.42(16), P(2)–Ir(1)–C(6A) 102.6(3), P(2)–Ir(2)–
Ct(3) 106.4(2), N–Ir(1)–C(6A) 174.8(3), N–Ir(2)–Ct(3) 177.2(3), Ct(1)–
Ir(1)–C(6A) 83.1(3), Ct(2)–Ir(2)–Ct(3) 83.7(3). Ct(1), Ct(2) and Ct(3)
represent the midpoints of the olefinic C(1)–C(2), C(9)–C(10) and
C(13A)–C(14A) bonds, respectively.
the coordination of the diphenylphosphido bridges not only
affect its asymmetry (more pronounced in 3), but also the
bond length (longer in 2) and angle values (mean P–Ir–P
angle values: 74.92(2) and 108.05(3)° in 2 and 3, respectively).
The two iridium and the two phosphorus atoms display a butter-
fly disposition, with a [P1Ir1P2]–[P1Ir2P2] dihedral angle of
59.51(3)° in 2 and 26.41(5)° in 3. The metal coordination geo-
metries reflect the distinct formal oxidation states: thus, the
Ir1 atom shows
a
distorted octahedral environment,
Scheme 1 Formation of complexes 2 and 3.
1610 | Dalton Trans., 2014, 43, 1609–1619
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
surrounded by a cod ligand, two bridging phosphorus atoms While complex 2 gave two well-resolved multiplets centred at
and one nitrogen atom, and the hydride, trans to the latter, −3.61 ppm (δ(15N) −100.3 ppm) at room temperature, those for
while Ir2 exhibits a trigonal bipyramid geometry, with the 4 and 5 were observed as broad multiplets at −3.71 and
phosphorus atoms and one olefinic bond of the cod ligand −3.83 ppm at −40 °C, respectively, which correlated with
defining the trigonal plane. This situation differs from that nitrogen resonances at −103.2 (4) and −104.6 (5) in their
shown in 3, in which both Ir geometries can be described as 15N–1H HMBC spectra.
severely distorted octahedra with the two phosphorus atoms in
The 31P{1H} NMR spectra of 4–6 were virtually identical, all
apical positions. Both complexes also differ in the coordi- of them showing an AB pattern at very low fields (δ(31P):
nation of the carbocyclic ligands; while in 2 the diolefins are 91.2 (3); 113.3 (4); 125.6 (5)). The marked difference in chemi-
coordinated in their usual η4-CvC mode, but are rotated 90° cal shift for the phosphane bridging ligands in 2 and 3 (Δδ =
with each other (due to the presence of the hydride ligand), in 199 Hz) should have its origin on the oxidation states of the
3 the initial diolefin attached to Ir1 has become a cyclooctenyl metals (and therefore on their geometry) in both complexes.13
ligand κ-C metallated trans to the amido nitrogen atom and From this point of view, complex 2 can be described as a
η2-CvC coordinated to iridium.
mixed-valence species IrIII/IrI (where the IrIII centre is labelled
The multinuclear NMR spectra of intermediate 2 showed as Ir1 in Fig. 1) without metal–metal bonds. In contrast,
some peculiarities derived from a non-rigidity phenomenon complex 3 displays a short iridium–iridium bond, a situation
operative within the diiridium system. More precisely, while that could be formally associated with an IrII–IrII metal–metal
the cod attached to the octahedral iridium atom Ir1 in 2 gave bonded system or an IrIII ← IrI dinuclear complex in which the
two signals both in the 1H and 13C{1H} NMR spectra, as IrI atom displays a dative bond to IrIII (Ir2 and Ir1 respectively,
expected for the Cs symmetry observed in the solid state, the in Fig. 2).14 A geometrical analysis of 3 reveals general trends
other cod molecule was found to undergo a fluxional motion, similar to those observed in 2: the Ir2–P bond distances are
more likely described as a propeller-like rotation that makes longer than Ir1–P ones, while the Ir2–N bond length is shorter
both vCH and CH2 protons and carbons of this carbocyclic than Ir1–N, a situation that seems to be in favour of the possi-
ligand chemically equivalent at all temperatures, respectively. ble existence of an intermetallic dative bond connecting the
The VT NMR spectra in CD2Cl2 did not show any decoales- two metals and reducing significantly their electronic differ-
cence at the lowest temperature available (−90 °C), indicating ences. An analysis of the NBO atomic charges of DFT calcu-
an unexpected low barrier for cod rotation. Its origin is most lated structures 2 and 3 has been performed in order to gain
likely related to the intrinsic stereochemical non-rigidity insight into the oxidation state of Ir centers. In complex 2, a
usually associated with pentacoordinated low-valent late metal difference of 0.2374 electron can be observed between the Ir1
complexes.12 On the other hand, the hydrido ligand was (qIr1 = −0.6219) and Ir2 (qIr2 = −0.3845) centers. On the other
located at δ −13.28 ppm as a triplet, and showed a strong NOE hand, in complex 3 both Ir centers present very similar
effect with the vCH protons pointing downwards of the charges qIr1 = −0.4570 and qIr2 = −0.4268. These results
amido ligand at Ir1 (those attached to C1 and C6 in Fig. 1), suggest a different electronic structure between Ir centers in
and the 31P{1H} NMR spectrum reflected the Cs symmetry complex 2 while a more similar electronic environment is
observed in the molecular structure of 2, since it showed a sole expected in complex 3. It is interesting to point out that the
signal at −108.1 ppm. The equivalent amido protons emerged Ir2P2N core is flexible enough to allow the structural changes
as a broad triplet at high field (δ(1H) −0.71 ppm; δ(15N) that encompass the transformation
−174.6 ppm). resembled in the geometry of the iridium centers, always held
The NMR spectra of complex 3 reflected the lack of sym- in close proximity by the triply-bridged system.
2 → 3, which are
metry observed in the molecular structure at various tempera-
Remarkably, complex 2 results from a formal oxidative
tures. A set of 2D NMR experiments allowed unambiguously addition of a P–H bond to a low-valent metal complex.15 The
assigning each resonance in both carbocyclic olefins (see overall formation of 2 implies two P–H bond activation pro-
ESI‡). More specifically, the metallated Ir–CH group was cesses of different nature: (i) a heterolytic split leads to an
observed at 2.55 and 9.2 ppm in the 1H and 13C{1H} NMR NH2/PPh2 exchange, and (ii) an iridium atom from the result-
spectra, respectively, while six well-separated resonances were ing mixed amido/phosphido species induces a homolytic scis-
clearly identified for each vCH carbon of both carbocyclic sion of the P–H bond of a second molecule of diphenyl-
molecules. It is worth mentioning the unusual shift to low phosphane giving formally a one-centre net cis addition of
field and multiplicity observed for the methylenic carbon H–PPh2. The former process should lead to a dinuclear inter-
(labelled as C5A in Fig. 2) in the 13C{1H} NMR spectra (δ 50.0, mediate [(cod)Ir(μ-NH2)(μ-PPh2)Ir(cod)] (Scheme 1) as supported
3JC–P = 9 Hz), which corresponds precisely to the vCH carbon by theoretical calculations (see ESI‡), with ammonia being
initially coordinated to iridium in complex 2. These character- formed in the process; as a matter of fact, the rhodium ana-
istic resonances were reflected in the related complexes 4 and logue [Rh2(µ-NH2)(μ-PPh2)(cod)2] was readily observed in situ by
5, which showed their respective methylenic carbons at 51.2 NMR spectroscopy on the addition of H–PPh2 to the rhodium
and 50.5 ppm in their 13C{1H} NMR spectra, respectively. A sig- bis(amido) complex [{Rh(µ-NH2)(cod)}2], along with the for-
nificant feature of complexes 3–5 was the presence of signals mation of the known bis(phosphido) species [{Rh(µ-PPh2)-
at high field that corresponded to the bridging amido protons. (cod)}2]16 and free ammonia (see ESI‡). In complex 2 the
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1609–1619 | 1611

View Article Online
Paper
Dalton Transactions
hydrido ligand is coplanar with one of the CvC double conjunction with complex 3, slow incorporation of deuterium
bonds of the neighbouring coordinated cod, a situation that at that methylenic carbon was observed in the 2H NMR
usually leads to a facile hydride transfer from the metal to the spectra.
olefin, affording in the present case unsaturated species 3, a
process that has already been observed for some mononuclear explained taking into consideration the microreversibility
late metal systems.17 status that may be established between complexes 2 and 3.18
These H/D exchange processes could in principle be
The fate of the protons attached to phosphorus in diphenyl- Although we have not been able to observe complex 2 on dis-
phosphane on interaction with 1 was clearly determined by solution of complex 3 at low temperatures by NMR techniques,
making use of isotope labelling techniques. The reaction of 1 species 2 should be present in the medium albeit in undetect-
2
with two molar equiv. of D–PPh2 was followed using H NMR able concentrations. From this perspective, it may be argued
spectroscopy, and this allowed observing the release of the that the H/D exchange occurs actually not on 3, but instead on
ammonia isotopomer NDH2 (δ −0.32 ppm) and a sole signal at complex 2 specifically at the hydrido site. In this context, there
δ 1.94 ppm, which specifically corresponded to one of the are some reports dealing with hydrogen exchange processes at
protons attached at C5A (Scheme 2, 3-d1). No deuterium hydrido late metal complexes with weak proton donors that
scrambling was observed in solutions of 3-d1 for a prolonged revealed that protonation at the hydrido moiety does indeed
time at room temperature. The net incorporation of deuterium lead to hydrogen exchange.19 From this perspective, we
at that methylenic carbon was reflected both in the multi- propose that it is the mixed-valence hydrido complex 2 that is
plicity of the phosphorus resonance by coupling with deuter- actually responsible for the H/D exchange processes observed
ium and in an upfield shift of 0.2 ppm in the 31P{1H} NMR experimentally. In this way, the hydrido ligand in 2 may
spectrum of 3-d1. In turn, treatment of the N-deuterated undergo H/D exchange with protic deuterated reagents
complex [{Ir(µ-ND2)(cod)}2] with diphenylphosphane allowed forming the deuterido analogue [(η4-cod)(D)Ir(µ-NH2)-
2
observing by H spectroscopy the amido deuterium atoms as (µ-PPh2)2Ir(η4-cod)], which then experiences a transfer process
the only iridium species (δ −3.6 ppm), along with release of to the coordinated cod molecule selectively to the C5A carbon
the ND2H isotopomer (δ −0.21 ppm), an observation that affording 3-d1 as the only observable product (Scheme 2).
evidenced that protons of one of the amido ligands do not
Indirect evidence of the microreversibility status established
participate in the formation of 2 and 3.
between species 2 and 3 (and therefore of the presence of
Some observations made from the above experiments complex 2 in solution) came from the observed catalytic
allowed us to observe H/D exchange processes promoted by activity in olefin isomerization. For instance, when solutions
complex 3 in solution at room temperature. As a matter of fact, of 3 in [D6]benzene were treated with 100-fold excess of 2-allyl-
treatment of 3 with D–PPh2 (δ(31P) −42.6 ppm) showed, after phenol at room temperature for 24 h, quantitative isomeriza-
ten minutes, the formation of free H–PPh2 (δ(31P) −41.3 ppm) tion to a 40 : 60 mixture of the Z and E isomers, respectively,
and the net incorporation of deuterium at C5A as a conse- was observed. Allyl benzene required 60 °C to isomerize to a
quence of an intermolecular H/D exchange. Interestingly, 35 : 65 mixture of the Z and E isomers in 75%. Since catalytic
ammonia was also found to participate in such intermolecular isomerization of olefins is often promoted by hydrido metal
exchange events. When a solution of 3 in toluene was treated complexes,20 it is reasonable to believe that the actual iso-
with ND3 at atmospheric pressure in a Young NMR tube, we merization catalyst should be the hydrido complex 2, which
observed selective H/D exchange and net incorporation of deu- should be present in equilibrium with 3 to a certain degree, as
terium specifically at the C5A carbon within 30 minutes along stated above.
with concomitant formation of ND2H, as confirmed by 2H,
1H and 31P{1H} NMR spectroscopy. In the same lines, when Young NMR pressure tube at 120 °C allowed observing conver-
On the other hand, heating toluene solutions of 3 in a
D2O or CD3OD was used as
a
deuterated reagent in sion to the known bis(phosphido) complex [{Ir(μ-PPh2)-
(cod)}2]16 after 12 h, along with released ammonia. Full conver-
sion of the related complexes 4–5 to the corresponding novel
phosphido-bridged derivatives [{Ir(μ-PR2)(cod)}2] (R = Cy (6),
iPr (7)) was achieved under milder conditions (see ESI‡). As a
i
matter of fact, the Pr complex 5 slowly transformed at room
temperature for 24 hours in toluene solutions to the corres-
ponding phosphido-bridged species 7.
Dinuclear phosphido complex 6 displayed two distinct
signals for the vCH fragments from different cod molecules
1
both in the H and 13C{1H} NMR spectra C2v symmetry, a situ-
ation that indicated the selective formation of one of the three
possible isomers, specifically that containing a square-planar
iridium centre (16 e−) and a tetrahedral iridium (18 e−)
connected by a metal–metal bond. The 31P{1H} NMR singlet
Scheme 2 Deuterium labelling studies carried out on complex 1.
1612 | Dalton Trans., 2014, 43, 1609–1619
observed at δ 26.0 ppm supported this proposal, since it
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
is comparable to other structurally related dinuclear bis- (µ-PPh2)}2] (2.551(1) Å),23 [{Ir(CO)2(µ-PtBu2)}2] (2.545(1) Å),24
(phosphido) late metal complexes found in the literature.21 In [{Ir(PEt3)2(µ-η1-2,4-dimethylphosphapentadienyl)}2] (2.587(1) Å),25a
contrast, complex 7 was isolated as a 3 : 1 mixture of isomers. and [{Ir(PEt3)2(µ-η1-phosphapentadienyl)}2] (2.576(1) Å),25b all
The most abundant one showed equivalent vCH cod of them examples considered to feature IrvIr double bonds.
1
fragments both in the H and the 13C{1H} NMR spectra and a Under this consideration, the electron count of 18 e− for each
significant signal at δ 230.0 ppm in the 31P{1H} NMR spec- iridium in 7 is reflected in the tetrahedral geometry around
trum, quite far from the resonance observed for the second both metals.26
isomer (δ 39.0 ppm). The low-field chemical shift observed for
The thermal reactions that lead to phosphido-bridged com-
the major isomer nicely fits those observed in some related plexes are sound examples of reductive elimination of
complexes which contain both metals around tetrahedral geo- ammonia, a rare phenomenon (the reverse process of oxidative
metries; furthermore, it also fits the C2h symmetry observed. addition of ammonia) seldom observed in late metal com-
On the other hand, the minor isomer corresponded to the plexes.27,10b In the mixed amido-phosphido bridged complexes,
square-planar–tetrahedral species: the pattern of the cod reso- the NH2 fragment is tightly bound to both metals, as observed
nances was virtually similar to that shown by 6, displaying a both experimentally and theoretically specifically in complex 3
C2v symmetry.
(see below), remaining unaltered throughout the sequential
X-ray quality crystals of derivative 7 were obtained and sub- transformation: 1 → 2 → 3. Along this manifold, the reductive
jected to an X-ray diffraction analysis. Fig. 3 shows the mole- elimination of ammonia should take place from hydrido
cular structure of 7, together with main bond lengths and complex 2. On the other hand, the temperature required to
angles. Complex 7 is dinuclear, with two phosphido ligands achieve elimination of ammonia from complex 2 has been
bridging both metals, and corresponds to the major C2h found to be higher than that needed for derivatives 4 and 5
isomer. The central Ir2P2 core, very different from those (PiPr2 and PCy2 phosphido bridges, respectively), indicating
observed in 2 and 3, is almost planar; the dihedral angle that basicity injected in the bimetallic system by the phos-
between the [P1Ir1P2] and [P1Ir2P2] planes is 0.08(3)°. More- phido ligands facilitates the NH3 reductive elimination process.
over, Ir–P bond lengths are significantly shorter than those
The reactions shown in this work have been examined by
previously reported for these amido-bridged complexes. As DFT calculations at the B3LYP level. Hence, the reaction
expected from other late metal phosphido-bridged systems, energy profile for the transformation of 1 (structure A) into 2
both metals display a strong intermetallic interaction illus- (structure C) and 3 (structure D) is presented in Fig. 4. The
trated by a very short Ir–Ir distance (Ir1vIr2 2.60679(18) Å). replacement of the amido bridge in A by the phosphido-bridge
This short intermetallic distance, which is shorter than that in B on addition of diphenylphosphane and release of
observed in [{Ir(µ-PCy2)(Cl)(CO)(PEt3)}2] (Ir–Ir 2.762(1) Å), for- ammonia is an exothermic reaction (−7.1 kcal mol−1) with an
mally an IrII–IrII single metal–metal bonded complex,22 energy barrier of 8.3 kcal mol−1. The energetic profile of this
matches well with those found in structurally related phos- process is shown in the ESI (Fig. S1‡) and it takes place via
phido-bridged diiridium complexes, such as in [{Ir(CO)(PPh3)- phosphide coordination to the metal, followed by amido-
bridge decoordination from one metallic centre and hydrogen
transfer from the phosphide group to the lone pair of the NH2
moiety. Addition of a second phosphane molecule to B follows
a different process as shown in Fig. 4. The oxidative addition
of the P–H bond to the Ir occurs through the transition struc-
ture TSB/C with an activation barrier of 21.4 kcal mol−1. The
lone pair of the Ir–PPh2 moiety can coordinate to the other Ir
centre yielding complex C, which corresponds to complex 2.
The migration of the hydride to the cod ligand takes place via
the transition structure TSC/D with an activation energy of
21.5 kcal mol−1, yielding structure D (corresponding to
complex 3).
The energetic profile for the reductive elimination of
ammonia from 2 to yield complex 6 is presented in Fig. 5. The
energy required to de-coordinate the amido-bridge from one
metal yielding the int4 structure is 13.1 kcal mol−1 and reduc-
tive elimination of NH3 occurs through the TSC/E structure.
Fig. 3 Crystal structure of complex 7. Selected bond lengths [Å] and
angles [°]: Ir(1)vIr(2) 2.60679(18), Ir(1)–P(1) 2.2808(7), Ir(1)–P(2) 2.2783(7), The energetic barrier for this process relative to the resting
state D is 39.2 kcal mol−1 and therefore high temperatures are
needed (it is experimentally observed at 120 °C). The outcome
Ir(2)–P(1) 2.2989(8), Ir(2)–P(2) 2.2994(7), mean Ir–Ct 2.061(1), P(1)–Ir(1)–
P(2) 111.30(3), P(1)–Ir(2)–P(2) 109.88(2), P(1)–Ir(1)–Ct(1) 117.72(8), P(1)–
Ir(2)–Ct(3) 128.46(8), P(2)–Ir(1)–Ct(2) 116.56(9), P(2)–Ir(2)–Ct(4)
128.77(8), Ct(1)–Ir(1)–Ct(2) 83.63(11), Ct(3)–Ir(2)–Ct(4) 83.83(11). Ct(1),
of this process is the ammonia adduct of the phosphido-
bridged complex E. Geometry optimization of the complex
E⋯NH3 on removal of ammonia leads to the observed complex
Ct(2), Ct(3) and Ct(4) represent the midpoints of the olefinic C(1)–C(2),
C(5)–C(6), C(9)–C(10) and C(13)–C(14) bonds, respectively.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1609–1619 | 1613

View Article Online
Paper
Dalton Transactions
Fig. 4 Computed DFT potential-energy profile (ΔE in kcal mol−1) for the formation of complex 3 from 1 and two molecules of diphenylphosphane.
Fig. 5 Computed DFT potential-energy profile (ΔE in kcal mol−1) for the reductive elimination of ammonia from complex 2.
E′ (corresponding to the known complex [{Ir(µ-PPh2)(cod)}2]). in the formation of ammonia by undergoing respectively
In summary, this DFT investigation shows that the transform- elementary oxidative addition (P–H) and reductive elimination
ation 1 → 2 → 3 is thermodynamically favored with activation (N–H) processes, where this metal experiences variations in its
barriers affordable at low temperatures, while the reductive oxidation state. We believe that this chemistry is partly fuelled
elimination of NH3 to yield 6 can only take place at high by the presence of a neighboring iridium basic center, which,
temperatures.
due to the high flexibility of the triply-bridged ligand system,
is indeed able to approach the other metal establishing an
intermetallic bond that helps in stabilizing these species.
Conclusions
We have shown a parent amido diiridium complex able to acti-
vate the P–H bond of a variety of secondary phosphanes under
Experimental
mild conditions, rendering unusual triply-bridged amido/ All manipulations were performed under a dry argon atmos-
phosphido species quite sensitive to temperature. It is note- phere using Schlenk-tube techniques. Solvents were obtained
worthy that throughout the chemical transformations from a solvent purification system (Innovative Technologies) or
described within the dinuclear system, only one iridium atom were dried by standard procedures and distilled under argon
seems to be involved both in the activation of P–H bonds (and prior to use. HPPh2, HPCy2 and HPiPr2 were purchased from
the subsequent olefin migratory insertion to a Ir–H bond) and Aldrich and gaseous NH3 and ND3 were commercially obtained
1614 | Dalton Trans., 2014, 43, 1609–1619
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
and com- 2.45 (m, 1H, vCH9), 2.40 (m, 1H, H7), 2.34 (m, 2H, H7′ + H3),
28
and used without further purification. D–PPh2
plexes 1 and [{Rh(µ-NH2)(cod)}2] were prepared as previously 2.26 (m, 1H, H11), 2.05 (m, 1H, H11′), 1.97 (m, 1H, H12),
described in the literature.8a All the other chemicals used in 1.93 (m, 1H, H16), 1.79 (m, 2H, H5 + H5′), 1.73 (m, 1H, H12′),
this work have been purchased from Aldrich Chemicals and 1.67 (m, 1H, H3′), 1.54 (m, 1H, H16′), 1.44 (m, 2H, H4 + H4′)
used as received. Carbon, hydrogen and nitrogen analyses (CH2) (cod), −3.45 (m, 1H), −3.77 (m, 1H) (NH2). 31P{1H} NMR
were performed using a Perkin-Elmer 2400 CHNS/O microana- (121 MHz, C6D6): δ 91.2 (AB system). 13C{1H} RMN (75 MHz,
lyzer. Mass spectra were recorded on a VG Autospec double- C6D6): δ 138. 6 (m, Cipso), 136.4, 132.7, 131.6 (m, Co), 129.7 (m,
3
focusing mass spectrometer operating in the FAB+ mode. Ions Cipso), 128.9 (d, JC–P = 5 Hz, Cm), 128.1, 127.9, 127.8 (Cm + Cp)
were produced with the standard Cs+ gun at ca. 30 kV; 3-nitro- (PPh2), 78.5 (s, vC14H), 75.3 (s, vC13H), 56.7 (d, 2JC–P = 10 Hz,
benzyl alcohol (NBA) was used as a matrix. ESI-MS were vC10H), 56.0 (dd, 2JC–P = 14 Hz, 3JC–P = 3 Hz, vC9H), 54.3 (dd,
3
2
recorded on a Bruker Micro-Tof-Q by using sodium formate as 2JC–P = 15 Hz, JC–P = 4 Hz, C2), 50.0 (t, JC–P = 9 Hz, C5), 42.2
a reference. 1H, 31P{1H} and 13C{1H} NMR spectra were (d, JC–P = 12 Hz, C1), 38.3 (s, C15), 37.1 (s, C4), 37.0 (s, C11),
2
recorded on a Varian UNITY, Bruker ARX 300, and Bruker 33.0 (s, C16), 30.6 (s, C3), 30.1 (s, C7), 29.9 (s, C8), 29.6 (s, C12),
Avance 400 spectrometers operating at 299.95, 121.42 and 9.2 (t, JC–P = 3 Hz, Ir–C6). 15N–1H HMQC (40 MHz, d8-toluene,
2
75.47 MHz, 300.13, 121.49 and 75.47 MHz, and 400.13, 161.99 −40 °C): δ −100.3. MS (ESI+): m/z 987.0 [M+ − H], 971.2
and 100.00 MHz, respectively. Chemical shifts are reported in [M+ − NH2]. Anal. Calcd (%) for C40H47Ir2NP2 (988.19):
ppm and referenced to Me4Si using the residual signal of the C 48.62, H 4.79, N 1.42; found: C 48.03, H 4.75, N 1.71.
deuterated solvent (1H and 13C), H3PO4 as an external refer-
[(1-κ-4,5-η2-C8H13)Ir(µ-NH2)(µ-PCy2)2Ir(1,2,5,6-η4-cod)] (4)
ence (31P) and liquid NH3 (1H–15N HMQC).
Pure dicyclohexylphosphane (0.09 g, 0.47 mmol) was slowly
added via a syringe to a solution of 1 (0.15 g, 0.24 mmol) in
[(η4-cod)(H)Ir(µ-NH2)(µ-PPh2)2Ir(η4-cod)] (2)
Pure diphenylphosphane (47 mg, 44 µL, 0.25 mmol) was toluene (6 mL), giving rise to a dark red mixture which was
slowly added via a microsyringe to a pre-cooled solution of 1 stirred for 1 hour. The solvent was then removed under
(80 mg, 0.13 mmol) in dichloromethane (8 mL) at −50 °C, vacuum leaving an oil that was subsequently washed with
layered with cold hexanes, and then the mixture was left for hexanes, affording a red solid which was isolated via cannula
5 days at −38 °C. A small crop of orange crystals identified as and dried under vacuum (0.22 g, 92%). 1H NMR (400 MHz,
complex 2 was separated very carefully, and a mono-crystal was d8-toluene, −40 °C): δ 4.26 (m, 1H, vCH), 3.64 (m, 2H, vCH +
subjected to X-ray analysis. The in situ 1H NMR monitoring CH2), 3.21 (m, 1H, vCH), 3.12 (m, 1H, vCH), 2.86 (m, 3H,
of the reaction of diphenylphosphane with 1 under these CH2), 2.80 (m, 4H, Ir–CH + CH2), 2.54 (m, 12H), 2.30 (m, 6H),
conditions allowed observing quantitative formation of 2.02 (m, 12H), 1.76 (m, 12H), 1.35 (m, 12H) (CH2 cod + PCy2),
complex 2 after 10 minutes. 1H NMR (400 MHz, CD2Cl2, −3.71 (br m, 2H, NH2). 31P{1H} NMR (121 MHz, d8-toluene,
−50 °C): δ 7.86 (m, 4H, Ho), 7.61 (m, 4H, Ho), 7.31 (m, 4H, Hm), −40 °C):
δ
113.3 (AB system). 13C{1H} RMN (75 MHz,
7.24 (m, 2H, Hm), 7.05 (m, 6H, Ho + Hm) (PPh2), 4.39 (m, d8-toluene, −40 °C): δ 64.8 (s, vCH), 62.8 (s, vCH), 51.0
2H, vCH), 3.11 (m, 4H, vCH), 2.42 (m, 8H, vCH + CH2), 2.08 (m, CH2), 48.0 (m, vCH), 47.0 (m, vCH), 46.9 (m, vCH), 41.2
(m, 2H), 1.87 (m, 6H), 1.63 (m, 2H) (CH2) (cod), −0.71 (t, 2H, (s, CH2) (cod), 40.8, 40.3 (m, Cipso PCy2), 48.9, 35.6 (m, vCH
2
3JH–P = 15.9 Hz, NH2), −13.28 (br t, 1H, JH–P = 15.0 Hz, Ir–H). cod), 32.2, 31.9, 31.5, 31.2, 29.6, 28.4, 28.2, 27.6, 27.0, 26.5
31P{1H} NMR (162 MHz, CD2Cl2, −50 °C): δ −108.1 (s). (set of s; CH2 cod + PCy2), 3.0 (s, Ir–CH). 15N–1H HMQC
13C{1H} RMN (100 MHz, CD2Cl2, −50 °C): δ 136.1 (s, Co), 135.8 (40 MHz, d8-toluene, −40 °C): δ −103.2. MS (micro-TOF): m/z
1
1
(d, JC–P = 20 Hz, Cipso), 135.7 (d, JC–P = 20 Hz, Cipso), 995.4057 [M+ − NH2]. Anal. Calcd (%) for C40H71Ir2NP2:
133.8 (s, Co), 127.7 (s, Cm), 127.3 (s, Cm), 126.9 (s, Cp), 126.1 C 47.46, H 7.07, N 1.38; found: C 47.03, H 4.75, N 1.71.
(s, Cp) (PPh2), 83.6, 75.6, 54.8 (all s, vCH), 34.0, 34.2, 28.4
[(1-κ-4,5-η2-C8H13)Ir(µ-NH2)(µ-PiPr2)2Ir(1,2,5,6-η4-cod)] (5)
(all s, CH2) (cod). 15N–1H HMQC (40 MHz, d8-toluene, −40 °C):
δ −174.6.
A solution of diisopropylphosphane (10% in hexane, 0.81 g,
0.67 mmol) was slowly added via a syringe to a solution of 1
(0.20 g, 0.33 mmol) in toluene (8 mL), giving rise to a dark red
[(1-κ-4,5-η2-C8H13)Ir(µ-NH2)(µ-PPh2)2Ir(1,2,5,6-η4-cod)] (3)
Pure diphenylphosphane (0.09 g, 82 µL, 0.47 mmol) was slowly solution. On stirring the mixture for 1 hour, the solvent was
added via a syringe to a red solution of 1 (0.15 g, 0.24 mmol) dried under vacuum and the oil obtained was washed with
in toluene (8 mL), giving rise to a dark red solution which was hexane, giving rise to a dark suspension. The solvent was
stirred for 45 min. The solvent was then removed via cannula removed via cannula and the solid was washed repeatedly with
1
and the remaining red solid was washed repeatedly with hexanes and then dried under vacuum (0.13 g, 41%). H NMR
hexanes and then dried under vacuum. Yield: 0.21 g (89%). (300 MHz, d8-toluene, −40 °C): δ 3.84 (m, 1H, vCH), 3.74
1H NMR (300 MHz, C6D6): δ 8.24 (m, 4H, Ho), 7.63 (m, 2H, Ho), (m, 1H, CH2), 3.38 (m, 1H, vCH), 3.24 (m, 1H, vCH), 3.04
i
7.27–6.73 (m, 14H, Ho + Hm + Hp) (PPh2), 3.89 (m, 1H, vCH1), (m, 1H, vCH), 2.70 (m, 4H, CH Pr + CH2 + vCH cod), 2.54
3.65 (m, 1H, vCH2), 3.58 (m, 1H, vCH13), 3.51 (m, 1H,vCH14), (m, 1H, Ir–CH), 2.43 (m, 4H, vCH + CH2), 2.33 (m, 2H, CH2),
3.37 (m, 1H, H8), 3.14 (m, 1H, vCH10), 2.83 (m, 1H, H8′), 2.75 2.15 (m, 4H, CH2), 1.95 (m, 2H, CH2), 1.76 (m, 2H, CH2), 1.56
(m, 1H, H15), 2.68 (m, 1H, H15′) (CH2), 2.55 (m, 1H, Ir–CH6), (m, 8H, CH2 cod + CH, CH3 iPr), 1.45 (m, 9H), 1.37 (m, 3H),
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1609–1619 | 1615

View Article Online
Paper
Dalton Transactions
1.22 (m, 3H), 1.01 (m, 3H), 0.79 (m, 3H) (CH3 iPr), −3.83 (br m,
Crystal structure determination for complexes 2, 3 and 7.
2H, NH2). 31P{1H} NMR (100 MHz, d8-toluene, −40 °C): δ 125.6 Single crystals of suitable size of 2, 3 and 7 were selected
(AB system). 13C{1H} RMN (75 MHz, d8-toluene, −40 °C): and mounted onto the end of a thin glass fibre using Fomblin
δ = 64.6 (s, vCH), 62.2 (s, vCH), 51.3 (m, vCH), 50.5 perfluoropolyether oil. Manipulation of sample 2 required
2
2
(t, JC–P = 8 Hz, CH2), 48.2 (d, JC–P = 24 Hz, vCH), 47.4 special care, as it is extremely temperature-sensitive (see
(d, 2JC–P = 21 Hz, vCH), 41.7 (m, CH2), 37.4 (m, CH2 cod), 35.2 below). X-ray diffraction data were collected at 100(2) K on a
(d, 2JC–P = 15 Hz, vCH), 31.5 (s, CH2), 30.0 (s, CH2) (cod), 28.9, Bruker APEX II (complexes 2 and 7) and on a Bruker SMART
27.9, 24.7, 24.0 (m, CH Pr), 22.9, 21.6, 21.5, 21.0 (s, CH3 iPr), APEX (compound 3) area detector diffractometers with graph-
i
2.9 (s, Ir–C). 15N–1H HMQC (40 MHz, d8-toluene, −40 °C): ite-monochromated MoKα radiation (λ = 0.71073 Å) by using
δ −104.6. MS (micro-TOF+): m/z 835.2822 (M+ − NH2).
narrow ω rotations (0.3°). Intensities were integrated and cor-
rected for absorption effect with SAINT+33 and SADABS34 pro-
grams, included in the APEX2 package. The structures were
solved by direct methods with SHELXS-97.35 Refinement,
by full-matrix least squares on F2, was performed with
SHELXL-97.36 Hydrogen atoms bound to olefinic carbon atoms
of the cod fragment in 2 and 3 were included in the model in
observed positions, and their isotropic thermal parameters
were fixed to 1.2 times the U value of the carbon atoms they
are linked to. Particular details of the presence of a solvent
and specific refinement details are listed below.
Crystal handling of compound 2. Due to the sensitivity of 2,
the sample in its mother liquor was stored in an ordinary
Schlenk flask whose crystallization temperature was main-
tained using a cooling bath. Single crystals were picked out
from the mother solution in a small spoon with Fomblin oil.
They were quickly immersed in inert oil on a microscope slide
with a cavity surrounded by dry-ice blocks. The generated cool
inert gas atmosphere maintained the crystal quality.
[{Ir(µ-PCy2)(cod)}2] (6)
A Young NMR tube was charged with 4 (15 mg), dissolved in
[D8]toluene (0.4 mL) and then sealed under argon. On heating
the mixture at 80 °C for 3 hours, complete conversion to 6 was
achieved. 1H NMR (300 MHz, C6D6): δ 4.17 (m, 4H), 3.88
(m, 4H) (vCH cod), 2.31 (m, 4H), 2.17 (m, 4H) (CH2 cod), 2.00
(m, 12H, CH2 cod + Cm Cy), 1.65 (m, 12H, CH2 cod + Co Cy),
1.35 (m, 4H, Cipso Cy), 1.04 (m, 4H, Cp Cy). 31P{1H} NMR
(121 MHz, C6D6): δ 26.0 (s). 13C{1H} RMN (75 MHz, C6D6):
2
2
δ 74.9 (d, JC–P = 12 Hz), 54.6 (s) (vCH cod), 38.8 (d, JC–P
=
14 Hz, Cipso Cy), 36.8 (s, CH2 cod), 35.1 (d, 3JC–P = 3 Hz, Cm Cy),
31.1 (s, CH2 cod), 28.7 (d, JC–P = 10 Hz, Co Cy), 26.5 (s, Cp Cy).
2
MS (ESI+): m/z 995.2 (M+
+ 1H). Anal. Calcd (%) for
C40H68Ir2P2: C 48.27, H 6.89; found: C 48.13, H 6.57.
[{Ir(µ-PiPr2)(cod)}2] (7)
A Young NMR tube was charged with 5 (15 mg), dissolved in
[D8]toluene (0.4 mL) and then sealed under argon. On heating
the mixture at 40 °C for 12 hours, complete conversion to 7
Crystal data for complex 2. C40H47Ir2NP2; M = 988.21;
orange prism; 0.148 × 0.141 × 0.134 mm3; monoclinic; P21/n;
a = 10.8643(12), b = 29.946(3), c = 11.1359(13) Å; β = 107.334(2)°;
1
was achieved. Td-Pc isomer: H NMR (300 MHz, C6D6): δ 4.03
Z = 4; V = 3458.5(7) Å3; ρcalc = 1.898 g cm−3; μ = 7.810 mm−1
;
(m, 4H), 3.80 (m, 4H) (vCH cod), 2.93 (m, 2H), 2.32 (m, 2H)
(CH iPr), 2.30 (m, 4H), 2.10 (m, 4H), 1.96 (m, 4H), 1.69
(m, 4H), (CH2 cod), 1.28 (m, 6H), 0.95 (m, 6H) (CH3 iPr).
31P{1H} NMR (121 MHz, C6D6): δ 39.0 (s). 13C{1H} RMN
(75 MHz, C6D6): δ 79.7 (d, 1JC–Rh = 11 Hz), 59.6 (m) (vCH cod),
min. and max. transmission factors: 0.317 and 0.461; 2θmax
59°; 44 776 reflections collected, 9093 unique [Rint = 0.0450];
the number of data/restraints/parameters: 9093/2/442; final
=
GoF: 1.035, R1 = 0.0282 [7517 reflections, I > 2σ(I)], wR2
=
.
0.0606 for all data; the largest difference peak: 1.838 e Å−3
2
1
i
39.8 (d, JC–Rh = 4 Hz, CH2), 35.7 (d, JC–P = 8 Hz, CH Pr), 35.6
A hydride atom has been observed in the Fourier difference
map. It has been included in the model at this observed posi-
tion and refined with a restraint on the Ir–H distance. At the
end of the refinement, nine residual density peaks higher than
1 e Å−3 were found. All of them are close to the metal atoms.
They have no chemical sense.
Crystal data for complex 3. C40H47Ir2NP2·C6D6; M = 1072.35;
red prism; 0.258 × 0.111 × 0.110 mm3; orthorhombic; Pbca; a =
21.7209(10), b = 10.3274(5), c = 33.4874(16) Å; Z = 8; V = 7511.9(6)
Å3; ρcalc = 1.896 g cm−3; μ = 7.199 mm−1; min. and max.
transmission factors: 0.242 and 0.446; 2θmax = 57.08°; 85 822
1
i
(s, CH2 cod), 31.9 (d, JC–P = 14 Hz, CH Pr), 30.8 (s), 25.7 (s)
1
(CH3 iPr). Td–Td isomer: H NMR (300 MHz, C6D6): δ 3.50 (m,
i
8H; vCH cod), 2.93 (m, 4H; CH Pr), 2.42 (8H, m), 1.96 (8H,
m) (CH2 cod), 0.95 (m, 12H, CH3 iPr). 31P{1H} NMR (121 MHz,
C6D6): δ 230.0 (s). 13C{1H} RMN (75 MHz, C6D6): δ 59.5
1
(m, vCH), 39.8 (s, CH2) (cod), 36.0 (d, JC–P = 8 Hz, CH), 25.7
(s, CH3) (iPr). MS (ESI+): 837.1 (M+ + 1H). Anal. Calcd (%) for
C28H52Ir2P2: C 40.27, H 6.28; found: C 40.19, H 6.12.
Computational details
The geometry of all structures has been optimized with the reflections collected, 9153 unique [Rint = 0.0456]; the number
G09 program package29 at the DFT level using the B3LYP of data/restraints/parameters: 9153/19/431; final GoF: 1.214,
approximation30 in combination with the 6-31G(d) basis set R1 = 0.0419 [8274 reflections, I > 2σ(I)], wR2 = 0.0871 for all
for H, C, N, and P atoms31 and the SDD pseudo-potential32 for data; the largest difference peak: 1.674 e Å−3. A disorder has
Ir. The nature of the stationary points has been confirmed by been observed in four carbon atoms of both cod-based ligands
frequency analysis and intrinsic reaction paths have been (C8H12 and C8H13 fragments). They have been included in the
traced connecting the transition structures with the respective model in two sets of positions with complementary occupancy
minima.
factors (0.7/0.3). They have been refined with geometrical
1616 | Dalton Trans., 2014, 43, 1609–1619
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
restraints. Hydrogen atoms of the NH2 bridging group have
been included in the model in observed positions and refined
with a restraint on N–H distances. They are very large solvent
accessible voids in the structure and a zone where residual
density peaks are observed. The solvent is highly disordered
and could not be properly modelled. Therefore, SQUEEZE37
corrections have been applied. The total potential solvent
accessible void volume and the electron count may agree with
the presence of eight C6D6 molecules.
Organometallics, 2000, 19, 2420; (h) D. Ristic-Petrovic,
J. R. Torkelson, R. W. Hilts, R. McDonald and M. Cowie,
Organometallics, 2000, 19, 4432; (i) D. A. Vicic and
W. D. Jones, Organometallics, 1999, 18, 134;
( j) J. R. Torkelson, F. H. Antwi-Nsiah, R. McDonald,
M. Cowie, J. G. Pruis, K. J. Jalkanen and R. L. DeKock, J.
Am. Chem. Soc., 1999, 121, 3666; (k) H. Suzuki, H. Omori,
D. H. Lee, Y. Yoshida, M. Fukushima, M. Tanaka and
Y. Moro-oka, Organometallics, 1994, 13, 1129; (l) D. A. Vicic
and W. D. Jones, Organometallics, 1997, 16, 1912;
(m) W. D. McGhee, F. J. Hollander and R. G. Bergman, J.
Am. Chem. Soc., 1988, 110, 8428.
3 See for example: (a) V. Miranda-Soto, J. J. Pérez-Torrente,
L. A. Oro, F. J. Lahoz, M. L. Martín, M. Parra-Hake and
D. B. Grotjahn, Organometallics, 2006, 25, 4374;
(b) S. Matsukawa, S. Kuwata and M. Hidai, Inorg. Chem.,
2000, 39, 791; (c) M. Nishio, H. Matsuzaka, Y. Mizobe and
M. Hidai, Inorg. Chim. Acta, 1997, 263, 119; (d) F. A. Cotton,
P. Lahuerta, J. Latorre, M. Sanati, I. Solaria and
W. Schwotze, Inorg. Chem., 1988, 27, 2131; (e) C. Claver,
J. Fis, P. Kalck and J. Jaud, Inorg. Chem., 1987, 26, 3479;
(f) P. Kalck and J. J. Bonnet, Organometallics, 1982, 1,
1211.
4 (a) P. García-Orduña, M. A. Casado, I. Mena and
F. J. Lahoz, Acta Crystallogr., Sect. C: Cryst. Struct. Commun.,
2012, 68, m113; (b) S. Takemoto, T. Honma and
H. Matsuzaka, Organometallics, 2011, 30, 1013; (c) C. Tejel,
M. A. Ciriano, M. Bordonaba, J. A. López, F. J. Lahoz and
L. A. Oro, Inorg. Chem., 2002, 41, 2348; (d) L. A. Oro,
D. Carmona, M. P. Lamata, M. C. Apreda, C. Foces-Foces,
F. H. Cano and P. M. Maitlis, J. Chem. Soc., Dalton Trans.,
1984, 1823; (e) M. P. García, M. V. Jimenez, T. Luengo and
L. A. Oro, J. Organomet. Chem., 1996, 510, 189.
5 (a) C. Tejel, M. A. Ciriano, J. A. López, S. Jiménez,
M. Bordonaba and L. A. Oro, Chem.–Eur. J., 2007, 13, 2044;
(b) C. Tejel, M. A. Ciriano, M. Bordonaba, J. A. López,
F. J. Lahoz and L. A. Oro, Chem.–Eur. J., 2002, 8, 3128;
(c) S. Kannan, A. J. James and P. R. Sharp, Inorg. Chim.
Acta, 2000, 19, 155; (d) J. Brunet, J.-C. Daran, D. Neibecker
and L. Rosenberg, J. Organomet. Chem., 1997, 538, 251;
(e) M. K. Kolel-Veetil, A. L. Rheingold and K. J. Ahmed,
Organometallics, 1993, 12, 3439; (f) M. K. Kolel-Veetil,
M. Rahim, A. J. Edwards, A. L. Rheingold and K. J. Ahmed,
Inorg. Chem., 1992, 31, 3877.
Crystal data for complex 7. C28H52Ir2P2; M = 835.11; red
prism; 0.281 × 0.114 × 0.111 mm3; monoclinic; P21/c; a =
12.0438(6), b = 13.3654 (6), c = 17.5773(8) Å; β = 95.2757(6)°;
Z = 4; V = 2817.4(2) Å3; ρcalc = 1.968 g cm−3; μ = 9.559 mm−1
;
min. and max. transmission factors: 0.185 and 0.331; 2θmax
54.96°; 27 939 reflections collected, 6455 unique [Rint
=
=
0.0277]; the number of data/restraints/parameters: 6455/1/409;
final GoF: 1.032, R1 = 0.0173 [5340 reflections, I > 2σ(I)], wR2 =
0.0400 for all data; the largest difference peak: 0.974 e Å−3
.
Hydrogen atoms of methyl groups have been included in the
model at calculated positions and refined with a riding model.
The rest of the hydrogen atoms have been included in the
model at observed positions and freely refined. Only one
restraint on a C–H distance has been included in the
refinement.
Acknowledgements
Financial support from the project CTQ2012-35665 and the
CONSOLIDER INGENIO-2010 program under the projects
MULTICAT (CSD2009-00050) and Factoría de Cristalización
(CSD2006-0015), and DGA/FSE 07, is acknowledged. P.G.O.
acknowledges the CSIC “JAE-Doc” program, cofunded by the
ESF and Factoría de Cristalización projects, for financial
support. The author V.P. thankfully acknowledges the
resources from the supercomputer “Terminus”, and the
technical expertise and assistance provided by the Institute
for Biocomputation and Physics of Complex Systems (BIFI) –
Universidad de Zaragoza.
Notes and references
1 (a) A. L. Gavrilovaa and B. Bosnich, Chem. Rev., 2004, 104,
349; (b) L. A. Oro, M. A. Ciriano and C. Tejel, Pure Appl.
Chem., 1998, 70, 779.
6 (a) I. Mena, E. A. Jaseer, M. A. Casado, P. García-Orduña,
F. J. Lahoz and L. A. Oro, Chem.–Eur. J., 2013, 19, 5665;
(b) P. Kläring, S. Pahl, T. Braun and A. Penner, Dalton
Trans., 2011, 40, 6785; (c) M. A. Salomon, A.-K. Jungton and
T. Braun, Dalton Trans., 2009, 7669; (d) T. Kimura, H. Arita,
K. Ishiwata, S. Kuwata and T. Ikariya, Dalton Trans., 2009,
2912; (e) R. Koelliker and D. Milstein, Angew. Chem., Int.
Ed., 1991, 30, 907; (f) A. L. Casalnuovo, J. C. Calabrese and
D. Milstein, Inorg. Chem., 1987, 26, 971.
2 For selected examples on dinuclear cooperative activation,
see: (a) T. S. Teets, T. R. Cook, B. D. McCarthy and
D. G. Nocera, Inorg. Chem., 2011, 50, 5223; (b) Y. Takahashi,
M. Nonogawa, K. Fujita and R. Yamaguchi, Dalton Trans.,
2008, 3546; (c) T. Shima and H. Suzuki, Organometallics,
2005, 24, 3939; (d) K. Fujita, H. Nakaguma, T. Hamada
and R. Yamaguchi, J. Am. Chem. Soc., 2003, 125, 12368;
(e) Y. Yuan, M. V. Jiménez, E. Sola, F. J. Lahoz and L. A. Oro,
J. Am. Chem. Soc., 2002, 124, 752; (f) M. V. Jiménez, E. Sola,
J. Caballero, F. J. Lahoz and L. A. Oro, Angew. Chem., Int.
Ed., 2002, 41, 1208; (g) T. Shima and H. Suzuki,
7 For homolytic splitting of ammonia leading to parent
amido complexes, see: (a) E. Morgan, D. F. MacLean,
R. McDonald and L. Turculet, J. Am. Chem. Soc., 2009, 131,
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1609–1619 | 1617

View Article Online
Paper
Dalton Transactions
14234; (b) C. M. Fafard, D. Adhikari, B. M. Foxman,
D. J. Mindiola and O. V. Ozerov, J. Am. Chem. Soc., 2007,
E. S. Shubina, E. Collange and R. Poli, J. Am. Chem. Soc.,
2003, 125, 11106.
129, 10318; (c) J. Zhao, A. S. Goldman and J. F. Hartwig, 20 For catalytic isomerizations promoted by hydrido late
Science, 2005, 307, 1080.
metal complexes, see for example: (a) S. Biswas, Z. Huang,
Y. Choliy, D. Y. Wang, M. Brookhart, K. Krogh-Jespersen
and A. S. Goldman, J. Am. Chem. Soc., 2012, 134, 13276;
(b) A. Scarso, M. Colladon, P. Sgarbossa, C. Santo,
R. A. Michelin and G. Strukul, Organometallics, 2010, 29,
1847; (c) R. H. Crabtree, in The Organometallic Chemistry of
the Transition Metals, John Wiley & Sons, Hoboken, NJ, 5th
edn, 2009, pp. 229–231; (d) P. W. N. M. van Leeuwen,
in Homogeneous Catalysis, Kluwer Academic Publishers,
Dordrecht, 2004, pp. 101–105.
8 For heterolytic splitting of ammonia leading to parent
amido complexes, see: (a) I. Mena, M. A. Casado, P. García-
Orduña, V. Polo, F. J. Lahoz, A. Fazal and L. A. Oro, Angew.
Chem., Int. Ed., 2011, 50, 11735; (b) T. Kimura, N. Koiso,
K. Ishiwata, S. Kuwata and T. Ikariya, J. Am. Chem. Soc.,
2011, 133, 8880; (c) T. Khaskin, M. A. Iron, L. J. W. Shimon,
J. Zhang and D. Milstein, J. Am. Chem. Soc., 2010, 132,
8542; (d) C. Ni, H. Lei and P. P. Power, Organometallics,
2010, 29, 1988; (e) J. J. Li, W. Li, A. J. James, T. Holbert,
T. P. Sharp and P. R. Sharp, Inorg. Chem., 1999, 38, 21 (a) T. S. Targos, G. L. Geoffroy and A. L. Rheingold, Organo-
1563.
metallics, 1986, 5, 12; (b) E. W. Burkhardt, W. C. Mercer
and G. L. Geoffroy, Inorg. Chem., 1984, 23, 1779;
(c) R. A. Jones, N. C. Norman, M. H. Seeberger, J. L. Atwood
and W. E. Hunter, Organometallics, 1983, 2, 1629;
(d) R. A. Jones, T. C. Wright, J. L. Atwood and W. E. Hunter,
Organometallics, 1983, 2, 470; (e) D. W. Meek, P. E. Kreter
and G. G. Christoph, J. Organomet. Chem., 1982, 231, C53;
(f) R. A. Jones and T. C. Wright, Organometallics, 1983, 2,
1842.
9 (a) J. L. Klinkenberg and J. F. Hartwig, Angew. Chem., Int.
Ed., 2011, 50, 86; (b) T. Braun, Angew. Chem., Int. Ed., 2005,
44, 5012; (c) J. I. van der Vlugt, Chem. Soc. Rev., 2010, 39,
2302; (d) Y. Aubin, C. Fischmeister, C. M. Thomas and
J. L. Renaud, Chem. Soc. Rev., 2010, 39, 4130.
10 (a) J. Zhou and J. F. Hartwig, Angew. Chem., Int. Ed., 2008,
47, 5783; (b) D. Conner, K. N. Jayaprakash, T. R. Cundari
and T. B. Gunnoe, Organometallics, 2004, 23, 2724;
(c) D. J. Fox and R. G. Bergman, J. Am. Chem. Soc., 22 H. A. Jenkins, S. J. Loeb and N. T. NcManus, Inorg. Chim.
2003, 125, 8984; (d) K. N. Jayaprakash, D. Conner and
T. B. Gunnoe, Organometallics, 2001, 20, 5254.
11 I. Mena, M. A. Casado, V. Polo, P. García-Orduña,
F. J. Lahoz and L. A. Oro, Angew. Chem., Int. Ed., 2012, 51,
8259.
Acta, 1989, 159, 83–86.
23 (a) P. L. Bellon, C. Benedicenti, G. Caglio and M. Manassero,
J. Chem. Soc., Chem. Commun., 1973, 946; (b) R. Mason,
I. Søtofte, S. D. Robinson and M. F. Uttley, J. Organomet.
Chem., 1972, 46, C61.
12 G. W. Bushnell, M. A. Casado and S. R. Stobart, Organo- 24 A. M. Arif, D. E. Heaton, R. A. Jones, K. B. Kidd,
metallics, 2001, 20, 601.
T. C. Wright, B. R. Whittlesey, J. L. Atwood, W. E. Hunter
13 O. Kühl, Phosphorus–31 NMR Spectroscopy: A Concise Intro-
and H. Zhang, Inorg. Chem., 1987, 26, 4065.
duction for the Synthetic Organic and Organmometallic 25 (a) J. B. Bleeke, A. M. Rohde and K. D. Robinson, Organo-
Chemist, Springer, Germany, 2008.
14 L. A. Oro, E. Sola, J. A. López, F. Torres, A. Elduque and
F. J. Lahoz, Inorg. Chem. Commun., 1998, 1, 64.
metallics, 1995, 14, 1674; (b) J. B. Bleeke, A. M. Rohde and
K. D. Robinson, Organometallics, 1994, 13, 401.
26 S.-K. Kang, T. A. Albright, T. C. Wright and R. A. Jones,
Organometallics, 1985, 4, 666.
15 (a) T. Shima, Y. Sugimura and H. Suzuki, Organometallics,
2009, 28, 871; (b) G. W. Bushnell, S. R. Stobart, R. Vefghi 27 Reported examples of reductive elimination of ammonia
and M. J. Zaworoto, J. Chem. Soc., Chem. Commun., 1984,
282.
16 P. E. Kreter and D. W. Meek, Inorg. Chem., 1983, 22, 319.
17 M. Martín, E. Sola, O. Torres, P. Plou and L. A. Oro, Organo-
metallics, 2003, 22, 5406.
on late metal complexes: (a) M. Morgan, D. F. MacLean,
R. McDonald and L. Turculet, J. Am. Chem. Soc., 2009, 131,
14234; (b) J. Zhao, A. S. Goldman and J. F. Hartwig, Science,
2005, 307, 1080; (c) D. Conner, K. N. Jayaprakash,
T. R. Cundari and T. B. Gunnoe, Organometallics, 2004, 23,
2724; (d) M. Kanzelberger, X. Zhang, T. J. Emge,
A. S. Goldman, J. Zhao, C. Incarvito and J. F. Hartwig,
J. Am. Chem. Soc., 2003, 125, 13644; (e) S. Park,
D. M. Roundhill and A. L. Rheingold, Inorg. Chem., 1987,
26, 3972; only one example of reductive elimination of
primary arylamines from palladium parent amido
complexes has been reported, see: J. L. Klinkenberg and
J. F. Hartwig, J. Am. Chem. Soc., 2010, 132, 11830.
18 The complex [Co(η4-cod)2(H)] has been found to be in
labile equilibrium with the product resulting from a
hydrido migratory insertion [Co(η4-cod)(1-κ-4,5-η2-C8H13)],
see: S. Otsuka and M. Rossi, J. Chem. Soc. A, 1968, 2530.
19 (a) G. A. Silantyev, O. A. Filippov, P. M. Tolstoy,
N. V. Belkova, L. M. Epstein, K. Weisz and E. S. Shubina,
Inorg. Chem., 2013, 52, 1787; (b) O. A. Filippov,
N. V. Belkova, L. M. Epstein, A. Lledós and E. S. Shubina,
Comput. Theor. Chem., 2012, 998, 129; (c) M. Baya, 28 A. Wawer and I. Wawer, J. Chem. Soc., Perkin Trans. 2, 1990,
O. Maresca, R. Poli, Y. Coppel, F. Maseras, A. Lledós, 2045.
N. V. Belkova, P. A. Dub, L. M. Epstein and E. S. Shubina, 29 M. J. Frisch, et al., Gaussian 09, revision A.1, Gaussian, Inc.,
Inorg. Chem., 2006, 45, 10248; (d) N. V. Belkova, P. O. Revin,
Wallingford, CT, 2004. The full reference is given in the
L. M. Epstein, E. V. Vorontsov, V. I. Bakhmutov,
ESI.‡
1618 | Dalton Trans., 2014, 43, 1609–1619
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
30 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 1372; (b) C. Lee, 33 SAINT+, 6.01, Bruker AXS, Inc., Madison, USA, 2000.
W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter, 34 G. M. Sheldrick, SADABS program, University of Göttingen,
1988, 37, 785; (c) A. D. Becke, J. Chem. Phys., 1993, 98,
5648.
31 (a) W. J. Hehre, R. Ditchfield and J. A. Pople, J. Chem. Phys.,
1972, 56, 2257; (b) P. C. Hariharan and J. A. Pople, Theor.
Chim. Acta, 1973, 28, 213.
Göttingen, Germany, 1999.
35 (a) G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crys-
tallogr., 1990, 46, 467; (b) G. M. Sheldrick, Methods
Enzymol., 1997, 276, 628.
36 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystallogr.,
2008, 64, 112.
32 (a) D. Andrae, U. Haussermann, M. Dolg, H. Stoll and
H. Preuss, Theor. Chim. Acta, 1990, 77, 123; (b) M. Dolg, H. Stoll, 37 P. V. der Luis and A. L. Spek, Acta Crystallogr., Sect. A:
H. Preuss and R. M. Pitzer, J. Phys. Chem., 1993, 97, 5852.
Fundam. Crystallogr., 1990, 46, 194.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1609–1619 | 1619