Characterization and reactivity studies of dinuclear iridium hydride complexes prepared from iridium catalysts with N,P and C,N ligands under hydrogenation conditions

Article abstract of DOI:10.1021/om4007467

The dinuclear iridium hydride complexes [IrH(CH₃CN)(L1)(μ-H)] ₂(BAr_F)₂ (7; L1 = (S)-2-(2- ((diphenylphosphanyl)oxy)propan-2-yl)-4-isopropyl-4,5-dihydrooxazole, BAr _F = tetrakis[3,5-bis(trifluoromethyl)phenyl]borate), [IrH(CH ₂Cl₂)(L1)(μ-H)]₂(BAr_F) ₂ (8), [IrH(L2)(μ-H)]₂(BAr_F)₂ (9a; L2 = (S)-1-[2-(2-adamantan-2-yl-4,5-dihydrooxazol-4-yl)-ethyl]-3-(2,6- diisopropylphenyl)-1,2-dihydroimidazol-2-ylidene), and [IrH(L3)(μ-H)] ₂(BAr_F)₂ (9b; L3 = (S)-1-[2-(2-tert-butyl-4,5- dihydrooxazol-4-yl)-ethyl]-3-(2,6-diisopropylphenyl)-1,2-dihydroimidazol-2- ylidene) were prepared from the corresponding mononuclear [Ir(COD)(L)]BAr _F precursors by treatment with H₂ and characterized by 2D NMR spectroscopy and X-ray diffraction. Conversion to a trinuclear iridium hydride complex, which is usually observed for N,P iridium hydride complexes, is inhibited by addition of 0.5 equiv of [H(OEt₂)₂]BAr _F or acetonitrile. Reactions with acetonitrile or 6,6′-bi-2-picoline afforded the mononuclear iridium dihydride complexes [Ir(H)₂(CH₃CN)₂(L1)]BAr_F (5), [Ir(H)₂(CH₃CN)₂(L3)]BAr_F (10), or [Ir(H)₂(6,6′-bi-2-picoline)(L3)]BAr_F (11). The CH₃CN complexes 7 and 10 are inactive as hydrogenation catalysts. In contrast, the coordinatively unsaturated dinuclear complexes 9a and 9b are active catalysts for the hydrogenation of (E)-1,2-diphenyl-1-propene at 50 bar hydrogen pressure.

Full text of DOI:10.1021/om4007467

Article
pubs.acs.org/Organometallics
Characterization and Reactivity Studies of Dinuclear Iridium Hydride
Complexes Prepared from Iridium Catalysts with N,P and C,N Ligands
under Hydrogenation Conditions
Stefan Gruber,† Markus Neuburger,‡ and Andreas Pfaltz*,†
†Department of Chemistry, University of Basel, St. Johanns-Ring 19, 4056 Basel, Switzerland
‡Laboratory for Chemical Crystallography, University of Basel, Spitalstrasse 51, 4056 Basel, Switzerland
S
* Supporting Information
ABSTRACT: The dinuclear iridium hydride complexes
[IrH(CH3CN)(L1)(μ-H)]2(BArF)2 (7; L1 = (S)-2-(2-((di-
phenylphosphanyl)oxy)propan-2-yl)-4-isopropyl-4,5-dihy-
drooxazole, BArF = tetrakis[3,5-bis(trifluoromethyl)phenyl]-
borate), [IrH(CH2Cl2)(L1)(μ-H)]2(BArF)2 (8), [IrH(L2)(μ-
H)]2(BArF)2 (9a; L2 = (S)-1-[2-(2-adamantan-2-yl-4,5-
dihydrooxazol-4-yl)-ethyl]-3-(2,6-diisopropylphenyl)-1,2-dihy-
droimidazol-2-ylidene), and [IrH(L3)(μ-H)]2(BArF)2 (9b; L3
= (S)-1-[2-(2-tert-butyl-4,5-dihydrooxazol-4-yl)-ethyl]-3-(2,6-
diisopropylphenyl)-1,2-dihydroimidazol-2-ylidene) were prepared from the corresponding mononuclear [Ir(COD)(L)]BArF
precursors by treatment with H2 and characterized by 2D NMR spectroscopy and X-ray diffraction. Conversion to a trinuclear
iridium hydride complex, which is usually observed for N,P iridium hydride complexes, is inhibited by addition of 0.5 equiv of
[H(OEt2)2]BArF or acetonitrile. Reactions with acetonitrile or 6,6′-bi-2-picoline afforded the mononuclear iridium dihydride
complexes [Ir(H)2(CH3CN)2(L1)]BArF (5), [Ir(H)2(CH3CN)2(L3)]BArF (10), or [Ir(H)2(6,6′-bi-2-picoline)(L3)]BArF (11).
The CH3CN complexes 7 and 10 are inactive as hydrogenation catalysts. In contrast, the coordinatively unsaturated dinuclear
complexes 9a and 9b are active catalysts for the hydrogenation of (E)-1,2-diphenyl-1-propene at 50 bar hydrogen pressure.
of iridium hydride complexes makes it difficult to study their
chemistry and to characterize such species as intermediates in
catalytic reactions.
INTRODUCTION
■
Iridium catalysts based on chiral N,P or C,N ligands have
considerably enhanced the scope of asymmetric hydrogenation
of olefins. In contrast to rhodium- or ruthenium-diphosphine
catalysts, they do not require a coordinating group next to the
CC bond and, therefore, perform well with a wide range of
structurally diverse olefins.1 Despite the large number of
iridium catalysts and successful applications reported in the
literature, experimental data on the reaction mechanism are still
scarce.2
Chart 1. Selection of Iridium Hydrogenation Catalysts
On the basis of DFT calculations, the groups of
Andersson2d,3 and Burgess4 proposed a catalytic cycle involving
an Ir(III) olefin complex containing two hydride ligands and a
side-on coordinated dihydrogen molecule, which, after
migratory insertion and concomitant oxidative addition of
dihydrogen, forms an Ir(V) trihydride complex. These
computational studies indicated that an Ir(III)−Ir(V) cycle is
preferred over an Ir(I)−Ir(III) cycle going through an
iridium(III) dihydride complex, which reacts with the substrate
through migratory insertion, followed by reductive elimination,
to yield the hydrogenation product and an Ir(I) complex. On
the other hand, experimental studies of Dietiker and Chen2c
have shown that, at least in the gas phase, an Ir(I)−Ir(III) cycle
is possible.
Previous studies5 have shown that Crabtree’s catalyst,
[Ir(COD)(PCy3)(py)]PF6 (1), forms trimeric and tetrameric
iridium hydride clusters under hydrogenation conditions.
Brown and co-workers5d have suggested that trimerization is
a consequence of ligand lability, in particular due to dissociation
of the Ir−N bond. The chiral hydrogenation catalyst [Ir-
(COD)(PHOX)]PF6 (2) as well forms a trinuclear iridium
hydride complex upon treatment with hydrogen gas (see eq
Obviously, iridium hydrides play a central role as
intermediates in the catalytic cycle. However, the high reactivity
Received: July 26, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/om4007467 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
1).6 All of these iridium polyhydride clusters are inactive as
hydrogenation catalysts.7
Here, we report the identification, synthesis, and character-
ization of a series of new dinuclear iridium hydride complexes
that are formed under hydrogenation conditions. Furthermore,
we discuss the activity of these new complexes as catalysts for
the hydrogenation of (E)-1,2-diphenyl-1-propene.
RESULTS AND DISCUSSION
■
(SimplePHOX)Ir(III)(H)2(CH3CN)2 Complexes. On the
basis of initial experiments with several N,P iridium complexes,
we chose the SimplePHOX complex 4 for our studies. Catalyst
4 had been previously fully characterized (including X-ray
analysis) and successfully used for the hydrogenation of various
olefins.11 Treatment of iridium complex 4 with 1 bar of
hydrogen gas at room temperature in CD2Cl2 for 20 min,
followed by the addition of 30 equiv of MeCN and stirring for
30 min, cleanly afforded a mixture of two iridium species.12 2D
NMR analyses revealed the formation of two stereoisomeric
bis-acetonitrile cis-dihydride complexes 5a and 5b in a ratio of
5:1 (Scheme 1). Because of the stabilizing effect of the
coordinated acetonitrile molecules, the complexes proved to be
sufficiently stable, even at room temperature, to allow proper
characterization. The two isomers differ in the position of Ha
relative to the isopropyl group of the SimplePHOX ligand. In
The rate of conversion to trinuclear hydride complexes
depends on the anion. With hexafluorophosphate, iridium
cluster formation competes with hydrogenation, and therefore,
catalyst deactivation is a serious problem. In contrast, the
corresponding BArF salts (BArF = tetrakis[3,5-bis(trifluoro-
methyl)phenyl]borate) show a much lower tendency to form
inactive iridium clusters. Thus, much higher turnover numbers
of >5000 become possible with iridium complexes with BArF or
related very weakly coordinating anions.2a
The cationic C,N complex 3a, developed by Burgess and co-
workers,8 was also reported to form inactive iridium hydride
species when treated with dihydrogen in the absence of
substrate. However, the structures of the components in the
resulting complex product mixtures were not elucidated.9
In addition to a trinuclear hydride complex, three
mononuclear Ir(III) dihydride complexes derived from
precatalyst 2 have been characterized by low-temperature
NMR experiments in THF-D8.10 Treatment of 2 with H2 at
−40 °C led to a single stereoisomer of [Ir(H)2(COD)-
(PHOX)]BArF, which reacted at 0 °C to give two isomers of
[Ir(H)2(PHOX)]BArF and cyclooctane. At room temperature,
formation of trinuclear hydride complexes was observed. A
coordinating solvent, such as THF, had to be used in these
experiments, because it stabilizes these labile hydride
complexes. However, THF is not suited for hydrogenation as
it inhibits the catalyst. Analogous experiments in dichloro-
methane, commonly used as a hydrogenation solvent, led to
complex mixtures of unidentified hydride species. Moreover,
attempts to identify further intermediates in the reaction
sequence leading to trinuclear complexes or to detect catalytic
intermediates under hydrogenation conditions failed so far.
Thus, further studies of iridium hydride complexes will be
necessary for a better understanding of the mechanism of C
C bond hydrogenation and catalyst deactivation.
1
the H NMR spectrum, the two hydrides appeared as two
doublets of doublets at −19.64 and −21.41 ppm for the major
isomer and at −18.95 and −21.85 ppm for the minor isomer,
respectively. The 2J(H,P) values of 20−24 Hz for both isomers
confirmed that both hydrides were located cis to the
phosphorus atom.13 Similar results had been previously
obtained by treating iridium complex 2 with H2 in THF-
D8.10a However, the resulting dihydride complexes were not
stable above 0 °C and conversion remained incomplete. The
observed electronically favored formation of an Ir−H bond
trans to the coordinated N atom is consistent with previous
studies by Crabtree and co-workers14 and Mazet et al.10a
Identification and Characterization of New Dinuclear
N,P Iridium Hydride Complexes. Attempts to isolate the
two stereoisomeric iridium complexes 5a and 5b from the
reaction mixture by evaporation of the solvent afforded new
1
products. The H NMR spectrum of the redissolved crude
material in CD2Cl2 showed the appearance of several new
hydride signals. Attempts to purify the crude product by
Scheme 1. Formation of Dinuclear Iridium(III) Hydride Complexes under Substrate-Free Hydrogenation Conditions
B
dx.doi.org/10.1021/om4007467 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 1. ORTEP view of the cations of salts 6 (left) and 7 (right). Hydrogen atoms and solvent molecules are omitted for clarity, and thermal
ellipsoids are set to 30% probability. Selected bond lengths (Å) for cation 6: Ir1−Ir2, 2.82334(14); Ir1−Cl1, 2.4343(5); Ir2−Cl1, 2.4346(5); Ir1−
N3, 2.168(2); Ir2−N4, 2.168(2). Bond angle (deg) for Ir1−Cl−Ir2 in cation 6: 70.884(15). Selected bond lengths (Å) for cation 7: Ir1−Ir2,
2.67849(11); Ir1−N3, 2.164(4); Ir2−N4, 2.163(5).
crystallization from a dichloromethane/hexane solution
afforded a few single crystals suitable for X-ray analysis. X-ray
diffraction of these crystals showed that an unexpected hydride-
and chloro-bridged dinuclear Ir(III) complex 6 had formed as a
side product (Scheme 1). Figure 1 (left) shows the cation of
this complex with selected bond lengths in the caption. The
solid-state structure reveals a bridging μ-chloride atom trans to
the phosphorus atoms P1 and P2. The hydride ligands were not
located in the final difference density map, but two apparent
vacant sites in the coordination spheres of Ir1 and Ir2 trans to
N3 and N4 were presumed to be the positions for two terminal
hydrides, while a bridging hydride was tentatively positioned
trans to N1 and N2. To verify this result, we redissolved a single
does not react with dichloromethane. Nevertheless, upon
standing for 8 h at room temperature in CD2Cl2, partial
decomposition with concomitant back reaction to the isomeric
mixture of 5a and 5b was observed. Addition of acetonitrile to a
CD2Cl2 solution of complex 7 induced rapid conversion to 5a
and 5b within <5 min. Crystals of the dinuclear complex 7
suitable for X-ray diffraction were obtained from a hexane/
dichloromethane solution. The crystal structure of this complex
is shown in Figure 1 (right) with selected bond lengths in the
caption. The hydride ligands were not located in the final
difference density map, but two apparent vacant sites in the
coordination sphere trans to the nitrogen atoms N3 and N4
represent the positions of two terminal hydrides,. The Ir−Ir
distance of 2.68 Å is consistent with literature values for
dinuclear Ir(III)/Ir(III) complexes containing an Ir2(H)2(μ-
H)2 fragment.18 Comparison of solid-state structures 6 and 7
reveals that, in the former, the phosphorus atoms are oriented
syn, whereas, in the latter, they adopt an anti arrangement. In
contrast to complex 6, complex 7 does not possess C2
symmetry, consistent with the solution structure analyzed by
NMR spectroscopy. The two isopropyl groups of the
SimplePHOX ligand are on the same side of the plane defined
by these six atoms, Ir1, P1, N1, Ir2, P2, and N2. Consequently,
the terminal hydrides H1 and H2 are nonequivalent and appear
as separate signals at −20.04 and −20.24 ppm. The 31P{1H}
NMR spectrum as well shows two different phosphorus signals
at 71.1 and 67.8 ppm. A signal at −17.89 ppm is observed for
the two bridging hydrides. The 2J(H,P) values of 20 Hz for the
terminal hydrides and 55 Hz for the bridging hydrides fully
support the structure drawn in Scheme 1 (for full character-
ization, see the Experimental Section and the Supporting
Information).
1
crystal in CD2Cl2 and analyzed the compound by H NMR
spectroscopy. The spectrum showed two hydride signals, one as
a doublet of doublets at −23.68 ppm (terminal hydrides) and
the other as a multiplet at −28.70 ppm (bridging hydride) with
a ratio of 2:1. The 31P{1H} NMR spectrum exhibited only one
2
singlet at 82.8 ppm. The J(H,P) values of <20 Hz clearly
indicate that none of the three hydrides is located trans to a
phosphorus atom. The NMR and crystal structure data are in
agreement with the C2 symmetric structure of 6 shown in
Scheme 1. The bridging hydride and chloride atoms can be
viewed to form 3-center 2-electron and 3-center 4-electron
units, respectively. The Ir−Ir distance of 2.82 Å is consistent
with a reported structure of a dinuclear Ir(I)/Ir(III) complex
containing an Ir2(μ-H)(μ-Cl) fragment.15 The bridging μ-
chloride atom presumably arises from dichloromethane used
for crystallization. Oxidative addition of dichloromethane with a
mononuclear iridium(I) complex has been described by Tejel
and co-workers.16
To avoid formation of the chloro-bridged complex 6, we
performed the hydrogenation of precatalyst 4 in acetonitrile at
50 bar hydrogen pressure. Evaporation of the solvent and
The Ir2 core of the dinuclear complex possesses a 32-electron
configuration with the bridging hydrides forming two 3-center
2-electron units. The bond order between the two Ir centers in
hydride-bridged complexes of this type has been discussed
controversially.18 Structures with an IrIr double bond as well
as structures without a bond between the two metal centers
have been formulated.
1
subsequent H NMR analysis of the crude material showed
partial formation of a new dinuclear iridium(III) hydride
complex 7 (Scheme 1). The crude material was dried at 0.3
mbar at 60 °C for 24 h, and after washing with hexane/
dichloromethane, the dinuclear complex 7 was isolated in 84%
17
1
yield. H NMR studies in CD2Cl2 showed that complex 7
C
dx.doi.org/10.1021/om4007467 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Treatment of iridium complex 4 with hydrogen gas (1 bar) at
room temperature in the presence of 0.5 equiv of [H(OEt2)2]-
BArF19 in dichloromethane for 20 min, followed by evaporation
of the solvent and washing with hexane/dichloromethane, led
The 1H NMR spectrum of a dissolved crystal was identical to
that of the crude material. The dinuclear complex 8 is dynamic
1
in solution. The H NMR spectrum shows broadened signals
consistent with a time-averaged symmetrical structure. We
assume that the coordinated dichloromethane molecules
exchange rapidly with the solvent.
1
to a new hydride complex (see eq 2). The H NMR spectrum
When the reaction shown in eq 2 was repeated under the
1
same conditions, but in the absence of [H(OEt2)2]BArF, H
NMR analysis revealed that ca. 20% of a trinuclear complex
[Ir3(H)6(L1)3(μ3-H)]2+ had formed.21 In the reaction sequence
leading to the trinuclear complex, one proton is liberated (eq
1), which explains why addition of acid inhibits its formation.
Identification and Characterization of New Dinuclear
C,N Iridium Hydride Complexes. On the basis of the results
obtained with the N,P iridium complex 4, we carried out
analogous experiments with C,N iridium complexes 3a and 3b
developed by Burgess et al.8 When complex 3a was treated with
hydrogen gas at 1 bar in dichloromethane at room temperature
for 30 min, followed by evaporation of the solvent, a new
complex was obtained in >95% yield. On the basis of extended
2D NMR studies, elemental analysis, and ESI-MS data
structure, 9a was proposed.
of the crude product showed that complex 4 had been fully
consumed and converted almost quantitatively (>95%) to a
single species that displayed two hydride signals at −16.14 and
−26.45 ppm in a 1:1 ratio. The 31P{1H} NMR spectrum
showed a singlet at 72.0 ppm. Single crystals suitable for X-ray
diffraction were obtained from a hexane/dichloromethane
solution. The solid-state structure of 8 shows an anti
orientation of the phosphorus atoms similar to that found for
complex 7 (Figure 2). Two terminal and two bridging hydride
The ESI mass spectrum of complex 9a showed a signal at m/
z 654.3027 ([C60H86N6O2Ir2]2+ calcd 654.3031) consistent
with a dinuclear species charged +2, as evidenced by the m/z
1
difference of 0.5 between the isotopomers. The H NMR
spectrum displayed two singlets for the bridging and terminal
hydrides at −18.40 and −36.85 ppm, respectively. The very low
frequency shift of the terminal hydrides is consistent with
literature values22 reported for iridium complexes of this type
with a vacant coordination site or a C−H agostic interaction
trans to a terminal hydride. The 13C{1H} spectrum exhibits a
very characteristic low-frequency singlet at 133.4 ppm for the
iridium-bound C atom of the N-heterocyclic carbene (NHC)
unit.23 Figure 3 shows two segments of the 2D NOESY
spectrum of complex 9a at room temperature, in which the
contacts from the hydrides to the various protons of the C,N
ligands are visible. This measurement allows to distinguish
between the terminal and bridging hydrides Ht and Hμ. For
example, Hμ shows a strong cross-peak to H22, whereas there is
no interaction between Ht and H22.
Figure 2. ORTEP view of the cation of salt 8. Hydrogen atoms are
omitted for clarity except the hydrides and ellipsoids are set to 30%
probability. The dichloromethane molecule coordinated to Ir2 is
disordered over two sites with occupancy factors of approximately 50%
for each site. One of the disordered CH2Cl2 molecules (not shown) is
too far from Ir2 to form a bond. Selected bond lengths (Å): Ir1−Ir2,
2.6443(2); Ir1−Cl1, 2.6121(14); Ir2−Cl3, 2.519(6); Ir1−H1, 1.51;
Ir1−H2, 1.76; Ir1−H4, 2.09; Ir2−H2, 1.96; Ir2−H3, 1.50; Ir2−H4,
1.73.
ligands were located in the final difference density map. The
Ir−Ir distance of 2.64 Å is similar to that found for complex 7.
Each of the two Ir centers bears a coordinated dichloromethane
molecule. The CH2Cl2 molecule at Ir2 is disordered over two
sites (occupancy factors ∼ 50% for each site). One of the
disordered solvent molecules forms an axial Ir−Cl bond,
whereas the other is too distant to coordinate to the metal
center. Complex 8 represents a rare example of a solid-state
structure with an iridium bound dichloromethane molecule.20
The relatively long Ir−Cl bonds in complex 8 with 2.61 and
2.52 Å indicate that the dichloromethane molecules are only
weakly bound.
Figure 3. Section of the 2D NOESY spectrum for salt 9a, showing the
contacts from the hydrides (Hμ (blue) and Ht (red)) to the various
protons of the C,N ligands.
D
dx.doi.org/10.1021/om4007467 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Although we were unable to produce crystals of complex 9a
for X-ray analysis, a related crystalline complex 9b was obtained
in high yield by hydrogenation of the Bu analogue 3b (eq 3).
complex 9a is stabilized in a similar fashion by agostic
interactions with the adamantyl groups.
t
Both complexes 9a and 9b are stable for several hours in
CD2Cl2 at room temperature under a hydrogen atmosphere at
1 bar. However, at 50 bar hydrogen pressure, a complex
mixture of products is formed, showing a multitude of hydride
1
signals in the H NMR spectrum.
Apparently, the dinuclear hydride C,N complexes 9a and 9b
are stable under 1 bar of H2 in a noncoordinating environment,
in contrast to analogous N,P ligand complexes, which are
converted to trinuclear hydride complexes in the absence of
stabilizing additives, such as coordinating ligands or [H-
(OEt2)2]BArF. Consistent with these observations, attempts
to form a C,N-based trinuclear iridium hydride complex from
3a under the conditions previously used for iridium N,P
complexes (eq 1) failed.25 These findings may be explained by
the stronger σ-donation of NHC ligands resulting in lower
acidity of C,N compared to N,P iridium hydride complexes,26
rendering the deprotonation step required for the conversion to
a trinuclear complex (see eq 1) more difficult.
The 1H NMR spectrum showed broad resonances in
dichloromethane at room temperature, but the signals became
sufficiently sharp at −30 °C to allow determination of the
solution structure, which closely corresponds to that of 9a (see
the Experimental Section and the Supporting Information).
1
The low-temperature H 2D exchange spectrum of 9b reveals
two further minor components in equilibrium with structure
9b. Crystallization from hexane/dichloromethane led to crystals
that were used for structure determination by X-ray diffraction
(Figure 4). The hydride ligands were not located in the final
The presence of two vacant coordination sites in complex 9b
led us to examine its reactivity toward nitrogen-based ligands.
Stirring an acetonitrile solution of 9b for 1 h at room
temperature cleanly led to the mononuclear complex 10 in
1
almost quantitative yield (Scheme 2). The H NMR spectrum
Scheme 2. Synthesis of New Mononuclear Iridium Hydride
Complexes 10 and 11
Figure 4. ORTEP view of the cation of salt 9b. Hydrogen atoms are
omitted for clarity except H33 and H292. Only one orientation of the
disordered isopropyl group is shown, and thermal ellipsoids are set to
30% probability. Selected bond lengths (Å) and distances (Å): Ir1−
Ir2, 2.6012(2); Ir1−H33, 2.17; Ir2−H292, 2.13; Ir1−C3, 2.77; Ir2−
C29, 2.79.
showed two hydride signals as doublets at −21.93 and −26.12
ppm and the 13C{1H} spectrum a singlet at 154.2 ppm for the
iridium-bound NHC C atom. An analogous product 11 was
obtained by stirring a dichloromethane solution of complex 9b
in the presence of 6,6′-bi-2-picoline for 16 h in 90% yield
(Scheme 2). Complex 11 can be also prepared by treating a
dichloromethane solution of 3b and 6,6′-bi-2-picoline with
difference density map, but the coordination geometry of the
two Ir centers is in agreement with the structure deduced by
NMR spectroscopy in which the two terminal hydrides are
located on the same side of the Ir2(μ-H)2 plane (see eq 3). The
structure possesses C2 symmetry with the two NHC units
oriented anti. The Ir−Ir distance is 2.60 Å, somewhat shorter
than the Ir−Ir bond in the N,P complex 8 (2.64 Å) The two Ir
centers are coordinatively unsaturated with two vacant
coordination sites trans to the terminal hydrides. The 14-
electron configuration of the two Ir centers is stabilized by
agostic interactions with the tert-butyl methyl groups, as
evidenced by the short Ir(1)−C(3), Ir(1)−H(33), Ir(2)−
C(29), and Ir(2)−H(292) distances, which are in agreement
with literature values.22a,b,24 We assume that the dinuclear
1
hydrogen gas for 4 h. The H NMR spectrum of this complex
shows two hydride signals as doublets at −21.92 and −28.62
ppm. The 13C signal of the iridium-bound NHC C atom
appears at 158.2 ppm. It is noteworthy that the sterically
demanding 6,6′-bi-2-picoline is capable of coordination in the
presence of the bulky C,N iridium fragment. Structures 10 and
11 indicate that the formation of an Ir−H bond trans to the
NHC ligand is electronically disfavored27 and that the
stereoisomers with hydrides trans to the oxazoline and
acetonitrile/picoline N atoms are thermodynamically preferred.
E
dx.doi.org/10.1021/om4007467 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 1. Hydrogenation of (E)-1,2-Diphenyl-1-propene Catalyzed by Different N,P Iridium Complexes
a
b
a
entry
cat.
molarity (M)
pressure (bar)
13 (%)
ee 13 (%)
byproduct (%)
c
d
e
1
4
0.2
0.2
0.2
2.0
0.2
0.2
0.2
2.0
2.0
50
1
>99
85 (R)
n.d.
f
2
7
7
7
8
8
8
8
8
8
<1
4
f
3
50
50
1
n.d.
f
4
6
n.d.
f
5
<1
<1
<1
5
n.d.
24
42
61
13
11
13
f
6
50
100
1
n.d.
f
7
n.d.
f
8
n.d.
f
9
50
100
57
71
71 (R)
79 (R)
f
10
2.0
a
b
c
d
1
Yields were determined by H NMR. Enantioselectivities were determined by HPLC on a chiral stationary phase. Result reported in ref 11. 1
e
f
mol % cat. Yield was determined by GC analysis. Average values of at least two experiments.
Table 2. Hydrogenation of (E)-1,2-Diphenyl-1-propene Catalyzed by Different C,N Iridium Complexes
a
b
entry
cat.
cat. loading (mol %)
molarity (M)
pressure (bar)
13 (%)
ee 13 (%)
c
1
3a
3b
10
9a
9a
9a
9b
9b
9b
3a
3a
3b
3b
3a
3b
3a
0.6
0.6
0.6
0.3
0.3
0.3
0.3
0.3
0.3
0.6
0.6
0.6
0.6
0.6
0.6
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
0.2
0.2
2.0
50
50
50
1
>99
81
98 (S)
50 (S)
n.d
c
2
3
3
d
4
16
97 (S)
97 (S)
97 (S)
n.d.
d
5
50
90
1
95
d
6
96
d
7
d
<1
62
8
50
90
1
96 (S)
96 (S)
99 (S)
98 (S)
99 (S)
98 (S)
63 (S)
68 (S)
86 (S)
d
9
77
10
>99
>99
>99
>99
26
d
11
50
1
12
d
13
50
50
50
50
d
14
d
15
24
d
16
0.1
39
a
b
c
Yields were determined by GC analysis. Enantioselectivities were determined by HPLC on a chiral stationary phase. Results reported in ref 8b.
Average values of at least two experiments.
d
1
In summary, these results demonstrate that the NHC-based
12 (1 equiv) by H NMR spectroscopy showed no change in
the hydride region after 16 h. Next, we tested the dinuclear
iridium complex 8 prepared according to eq 2, in which the two
acetonitrile ligands were replaced by more weakly bound
dichloromethane molecules. Under standard conditions (0.2
M) at 1, 50, and 100 bar, no conversion to the alkane was
observed. However, significant amounts of byproducts were
formed, which resulted from acid-catalyzed dimerization of
alkene 12 (entries 5−7).28 These side reactions were likely
caused by proton transfer from acidic iridium hydride species26
present in the reaction mixture. When the substrate
concentration was increased to 2.0 M, the dinuclear iridium
complex 8 showed some weak activity at 1 bar. However, at 50
and 100 bar, it afforded the alkane in 57% and 71% yield with
71% and 79% ee, respectively (entries 8−10). At this
concentration, less than 13% byproducts were formed.
Apparently, the dinuclear iridium complex 8 was converted to
an active catalyst under these conditions. However, the
complexes 3a,3b are converted to dinuclear hydride-bridged
iridium complexes under hydrogenation conditions in the
absence of hydrogenation substrates. In contrast to their
counterparts derived from N,P ligands, they show no tendency
to form trinuclear complexes.
Hydrogenation Reactions. With these new iridium
hydride complexes in hand, we decided to examine their
reactivity as hydrogenation catalysts. Catalyst 4 has been
previously reported to hydrogenate (E)-1,2-diphenyl-1-propene
(12) with full conversion and an enantiomeric excess of 85%
(entry 1 of Table 1).11 The dinuclear iridium complex 7
showed almost no conversion to the corresponding alkane 13
under the same reaction conditions at 1 or 50 bar hydrogen
pressure and 0.2 and 2.0 M substrate concentrations (entries
2−4). This was expected, since acetonitrile is a relatively strong
ligand, and coordinating species are known to strongly inhibit
catalysts of this type.1c Monitoring CD2Cl2 solutions of
complex 7 under 1 bar hydrogen gas or in the presence of
F
dx.doi.org/10.1021/om4007467 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
conversion and enantioselectivity were lower than those in
hydrogenations with the mononuclear precatalyst 4.
The C,N dinuclear hydride complexes proved to be
remarkably stable despite their coordinatively unsaturated 28-
electron Ir2 core. In contrast to analogous N,P complexes, they
were cleanly formed in high yields upon treatment of the
Ir(COD) precursors 3a and 3b with H2 in pure CH2Cl2,
whereas the hydrogenation of the N,P analogue 4 gave
substantial amounts of a stable catalytically inactive trinuclear
hydride complex. The conversion to Ir3 hydride clusters is a
known deactivation pathway for N,P iridium catalysts,
presumably proceeding through dinuclear hydride complexes
of the type described in this work. Notable in this context is our
finding that formation of a trinuclear complex can be
suppressed by addition of the strong Brønsted acid [H-
(OEt2)2]BArF. In continuation of this work, we are currently
exploring the reactivity of iridium hydride complexes toward
olefins, with the aim of identifying potential hydrogenation
intermediates.
Burgess and co-workers8b reported full conversion and 98%
ee for this hydrogenation reaction using their catalyst 3a,
whereas complex 3b, with a tert-butyl instead of an adamantyl
substituent in the oxazoline ring, gave significantly lower
conversion and ee (Table 2, entries 1 and 2). As expected, the
bis-acetonitrile complex 10 showed almost no activity (entry
3). On the other hand, the dinuclear complex 9a with two
vacant coordination sites proved to be an active and highly
enantioselective catalyst. Whereas, at 1 bar hydrogen pressure,
only 16% conversion was observed after the standard reaction
time of 2 h, almost full conversion with the same ee of 97% was
achieved at 50 and 90 bar (entries 4−6). The analogous
complex 9b with a tert-butyl instead of an adamantyl group was
inactive at 1 bar hydrogen pressure, but led to 77% conversion
with 96% ee at 90 bar (entries 7−9). Surprisngly, the
enantioselectivity induced by the dinuclear complex 9b was
much higher than the reported value for the corresponding
mononuclear complex 3b (entries 2 and 8). This prompted us
to reinvestigate the hydrogenation of 12 with complexes 3a and
3b at 1 and 50 bar hydrogen pressure. In all reactions, we
consistently observed full conversion and >98% ee (entries 10−
13).29 Remarkably, the performance strongly depended on the
substrate concentration. A decrease in the alkene concentration
from 2 to 0.2 M resulted in a much lower conversion and
enantioselectivity (entries 14−15). Similar results were
obtained for the dinuclear iridium hydride complexes 9a and
9b (see the Supporting Information, Table S6). Lowering the
catalyst loading of complex 3a to 0.1 mol % also decreased the
conversion and enantiomeric excess (entry 16, Table 2).
EXPERIMENTAL SECTION
■
General Information. All reactions and manipulations with air- or
moisture-sensitive materials were carried out under an atmosphere of
argon using standard Schlenk techniques. Schlenk tubes were heated in
an oven (120 °C) and then dried under high vacuum. CH2Cl2 was
dried over CaH2 and then distilled under argon. CH3CN and pentane
were dried and degassed using a solvent purification system (PureSolv,
Innovation Technology Inc.). Dry hexane was purchased in septum-
sealed bottles from Aldrich and stored over 4 Å molecular sieves under
argon. CD2Cl2 was distilled (bulb-to-bulb) over CaH2, degassed by
three freeze−pump−thaw cycles, and stored over 4 Å molecular sieves
under argon. All commercially available starting materials were
purchased from commercial sources and used as received. [Ir(L1)-
(COD)]BArF (4),11 [Ir(L2)(COD)]BArF (3a),8b Ir(L3)(COD)]BArF
30
(3b),8b and [H(OEt2)2]BArF were synthesized according to
literature procedures.1H, 13C, 31P, and 2D NMR spectra were recorded
with a Bruker Advance 400 (400 MHz) and a Bruker Advance DRX-
500 (500 MHz) spectrometer. Chemical shifts (δ) are given in parts
CONCLUSIONS
■
Previous studies of iridium hydrides derived from phosphi-
nooxazoline complexes were carried out in THF,10 which
stabilizes these reactive species by coordination. However, for
catalytic hydrogenations, weakly coordinating solvents, such as
dichloromethane, have to be used, because coordinating
solvents or additives deactivate the catalyst. We have shown
here that iridium hydride complexes can be prepared in
dichloromethane under hydrogenation conditions in the
absence of substrate and characterized by NMR and X-ray
analysis. In contrast to the mononuclear dihydride complexes
obtained in THF, dinuclear complexes with two bridging
hydrides and two terminal hydride ligands are formed in
CH2Cl2. These complexes prepared from the N,P ligand
complex 4 and analogous NHC-based catalysts, developed by
Burgess et al.,8 are coordinatively unsaturated (28-electron
configuration of the Ir2 core) with a vacant coordination site at
each Ir center. In the case of the N,P ligand complexes, the two
vacant coordination sites are occupied by two weakly bound
CH2Cl2 molecules in the crystal structure, whereas the NHC-
based analogues are stabilized by agostic interactions between
these sites and adjacent alkyl groups of the C,N ligand.
Both the N,P and the C,N hydride complexes show catalytic
activity in the hydrogenation of (E)-1,2-diphenyl-1-propene,
with enantioselectivities similar to those recorded for the
corresponding mononuclear Ir(COD) precursors. While it
seems unlikely that the dinuclear hydride complexes are the
actual catalysts, our results indicate that they are converted to
an active (presumably mononuclear) species under the reaction
conditions.
per million and referenced relative to TMS for H and 13C NMR
1
spectra and H3PO4 for 31P NMR spectra. 1H 2D NOESY spectra were
acquired using a standard three-pulse sequence and a mixing time of
700 ms.
X-ray Analyses. Data sets were obtained using a Bruker Kappa
Apex2 diffractometer (graphite monochromated Mo Kα radiation, λ =
0.71073 Å) at 123 K. The structures were solved by direct methods
using the program SIR92.31 CRYSTALS32 was used for structure
refinement. The crystallographic data and R values of the full-matrix
least-squares refinements are given in Tables S7 (6), S8 (7), S9 (8),
and S10 (9b), respectively (see the Supporting Information). All
atoms except hydrogen atoms and atoms of disordered molecules were
refined with anisotropic displacement parameters. In general, H atoms
were placed at calculated positions based on stereochemical
considerations and refined according to the riding model. The
crystallographic data of complexes 6 (CCDC 943439), 7 (CCDC
943440), 8 (CCDC 943442), and 9b (CCDC 943441) can be
obtained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
Synthesis of [IrH(CH3CN)(L1)(μ-H)]2(BArF)2 (7). Iridium complex
4 (150 mg, 99 μmol) was added to a 4 mL vial and dissolved in
CH3CN (1.0 mL), and the vial was placed into an autoclave. The
autoclave was purged with hydrogen for 30 s, pressurized to 50 bar of
hydrogen, and sealed. The reaction mixture was stirred at 900 rpm for
2 h at room temperature. After release of hydrogen, the solution was
transferred into a Schlenk tube. The solvent was completely
evaporated under vacuum, and the residue was dried at 0.3 mbar for
1 h. The colorless residue was dissolved in CH2Cl2 (1 mL). The
solvent was then completely evaporated under vacuum, and the
residue was dried at 0.3 mbar at 333 K for 8 h. This procedure was
repeated two times. The resulting yellow residue was washed with a
mixture of CH2Cl2/hexane (3/1, 2 × 0.7 mL) and dried under vacuum
G
dx.doi.org/10.1021/om4007467 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
to afford complex 7 as a yellow solid (121 mg, 84%). A CH2Cl2
solution (0.5 mL) of this solid (10 mg) was layered with hexane (2
mL) and stored at 4 °C, to afford crystals of 7 suitable for X-ray
11), 124.3 (C-13), 117.9 (br s, C-BArF), 74.1 (C-2), 66.9 (C-3), 46.0
(C-5), 37.2 (C-21), 36.3 (C-22), 35.8 (C-24), 34.6 (C-4), 29.9 (C-18),
29.1 (C-15), 27.7 (C-23), 25.9 (C-19), 25.8 (C-16), 23.8 (C-20), 22.3
(C-17). Anal. Calcd for C124H110B2F48Ir2N6O2: C, 49.08; H, 3.65, N,
2.77. Found: C, 49.26; H, 3.72; N, 2.83. HR ESI-MS (CH2Cl2, 333 K)
Calcd m/z for ([C60H86N6O2Ir2]2+: 654.3031. Found: 654.3027.
Synthesis of [IrH(L3)(μ-H)]2(BArF)2 (9b). Iridium catalyst 3b (100
mg, 65 μmol) was added to a Schlenk tube and dissolved in CH2Cl2
(2.0 mL). The Schlenk tube was placed in liquid N2 until the yellow
solution froze. The Schlenk tube was then evacuated and purged with
hydrogen gas (balloon), and the solution was allowed to warm up to
room temperature. After stirring for 5 h at room temperature, the
solvent was completely evaporated under vacuum. The resulting
yellow foam was washed with hexane (3 × 1 mL) to afford a yellow
solid, which was dissolved in CH2Cl2 (1.5 mL) and layered with
hexane (5 mL). The yellow crystals formed during 48 h at −4 °C were
separated from the mother liquor and dried at 0.3 mbar for 16 h to
afford complex 9b (84 mg, 90%). These crystals were used for X-ray
1
diffraction. H NMR (500 MHz, CD2Cl2, 233 K): δ 7.70−7.60 (m,
20H, BArF-H + Harom.), 7.54−7.39 (m, 22H, BArF-H + Harom.), 7.29−
7.25 (m, 2H, Harom.), 4.57 (t, 1H, J = 10 Hz, C-5-Ha), 4.49−4.38 (m,
3H, C-5′-Ha, C-5-Hb + C-5′-Hb), 3.94−3.92 (m, 1H, C-1-H), 3.75−
3.73 (m, 1H, C-1′-H), 2.38−2.34 (m, 2H, C-2-H + C-2′-H), 1.91 (s,
3H, C-11-H), 1.78 (s, 3H, C-11′-H), 1.76 (s, 3H, C-9′-H), 1.73 (s, 3H,
C-9-H), 1.45 (s, 3H, C-8-H), 1.37 (s, 3H, C-8′-H), 0.77 (d, 3H, J = 6.5
Hz, C-3′-H), 0.68−0.66 (m, 6H, C-4′-H + C-4-H), 0.54 (d, 3H, J = 6.6
Hz, C-3-H), −17.77−(−18.00) (m, 2H, Ir-H′μ + Ir-Hμ), −20.04 (d,
1H, J = 18.7 Hz, Ir-H′t), −20.24 (dd, 1H, J = 19.7 Hz, J = 3.2 Hz, Ir-
3
Ht). 13C{1H} NMR (100 MHz, CD2Cl2, 295 K): δ 169.8 (d, JCP = 5
Hz, C-6′), 169.1(d, 3JCP = 5 Hz, C-6), 162.1 (q, 1JCB = 50 Hz, C-BArF),
136.9 (Carom.), 136.2 (Carom.), 135.8 (Carom.), 135.3 (Carom.), 135.2 (C-
2
BArF), 134.0−130.0 (9 × Carom.), 129.3 (q, JCF = 32 Hz, C-BArF),
129.2−128.6 (3 × Carom.), 125.0 (q, 1JCF = 272 Hz, C-BArF), 121.0 (d,
3
3JCP = 10 Hz, C-10), 120.8 (d, JCP = 10 Hz, C-10′), 117.9 (br s, C-
1
analysis. H NMR (500 MHz, CD2Cl2, 243 K): δ 7.71 (br s, 16H,
BArF), 80.7 (d, 2JCP = 7 Hz, C-7′), 80.6 (d, 2JCP = 7 Hz, C-7), 78.5 (C-
1′), 74.2 (C-1), 70.1 (C-5), 70.0 (C-5′), 29.6 (C-2), 28.5 (C-2′), 27,6
(d, 3JCP = 7 Hz, C-9′), 27.4 (d, 3JCP = 6 Hz, C-9), 26.4 (d, 3JCP = 3 Hz,
C-8), 26.3 (d, 3JCP = 3 Hz, C-8′), 18.8 (C-4′), 18.4 (C-4), 13.6 (C-3′),
12.7 (C-3), 3.0 (C-11), 2.3 (C-11′). 31P{1H} NMR (202 MHz,
CD2Cl2, 295 K): 71.1 (P-B), 67.8 (P-A). Anal. Calcd for
C110H86B2F48Ir2N4O4P2: C, 45.44; H, 2.98, N, 1.93. Found: C,
45.32; H, 2.95, N, 1.93. ESI-MS (CH2Cl2, 323 K): m/z 591 ([M − 2 ×
BArF]2+).
BArF-H), 7.55 (br s, 8H, BArF-H, 7.37 (t, 2H, J = 7.6 Hz, C-12-H),
7.23 (d, 2H, J = 7.6 Hz, C-13-H), 7.17 (d, 2H, J = 7.6 Hz, C-11-H),
6.97 (br s, 4H, C-6-H + C-7-H), 4.82 (dd, 2H, J = 10.0 Hz, J = 10.0
Hz, C-2-Ha), 4.50−4.45 (m, 2H, C-3-H), 4.40 (dd, 2H, J = 9.5 Hz, J =
6.1 Hz, C-2-Hb), 4.20 (dd, 2H, J = 15.0 Hz, J = 8.2 Hz, C-5-Hb), 3.69
(dd, 2H, J = 15.0 Hz, J = 10.0 Hz, C-5-Ha), 2.73−2.67 (m, 2H, C-4-
Ha), 2.37−2.31 (m, 2H, C-18-H), 2.08−2.03 (m, 2H, C-4-Hb), 1.78−
1.72 (m, 2H, C-15-H), 1.22 (d, 6H, J = 6.8 Hz, C-17-H), 1.19 (d, 6H, J
= 6.8 Hz, C-20-H), 1.01 (d, 6H, J = 6.8 Hz, C-19-H), 0.88 (d, 6H, J =
6.8 Hz, C-16-H), 0.80 (s, 18H, C-22-H), −18.90 (s, 2H, Ir-Hμ),
−34.45 (s, 2H, Ir-Ht). In addition, the spectrum shows weak signals of
minor isomers, which are in equilibrium with the major component
according to the 1H 2D exchange spectrum (see the Supporting
Information).13C{1H} NMR (125 MHz, CD2Cl2, 243 K): δ 181.4 (C-
1), 161.8 (q, 1JCB = 50 Hz, C-BArF), 147.3 (C-10), 145.0 (C-14), 136.9
(C-9), 134.8 (C-BArF), 132.2 (C-8), 130.3 (C-12), 128.8 (q, 2JCF = 32
Hz, C-BArF), 126.2 (C-6), 124.6 (q, 1JCF = 272 Hz, C-BArF), 124.3 (C-
13), 123.8 (C-7 + C-11), 117.6 (br s, C-BArF), 73.9 (C-2), 66.1 (C-3),
45.7 (C-5), 34.2 (C-21), 32.5 (C-4), 28.9 (C-18), 28.4 (C-15), 26.0 (C-
19), 25.5 (C-16), 23.4 (C-22), 22.3 (C-17), 22.0 (C-20). Anal. Calcd
for C112H98B2F48Ir2N6O2: C, 46.74; H, 3.43, N, 2.92. Found: C, 46.46;
H, 3.34, N, 3.03. ESI-MS (CH2Cl2, 323 K): m/z 576 ([M −
2×BArF]2+).
[IrH(CH2Cl2)(L1)(μ-H)]2(BArF)2 (8). Iridium complex 4 (15 mg, 9.9
μmol, 1.0 equiv) and [H(OEt2)2][BArF] (100 mg, 4.3 μmol, 0.5 equiv)
were added to a Schlenk tube and dissolved in CH2Cl2 (0.5 mL). The
Schlenk tube was placed in liquid N2 until the yellow solution froze.
The Schlenk tube was then evacuated and purged with hydrogen gas
(balloon), and the solution was allowed to warm up to room
temperature. After stirring for 20 min at that temperature, the solvent
was completely evaporated. The remaining yellow solid was washed
with a mixture of CH2Cl2/pentane (3/1, 3 × 0.5 mL) and dried at 0.3
mbar for 16 h. 1H NMR spectroscopy of this material indicates
formation of one major species (>95%). A CH2Cl2 solution (0.5 mL)
of this solid (10 mg) was layered with hexane (2 mL) and stored at 4
1
°C, to afford crystals of 7 suitable for X-ray diffraction. The H NMR
spectrum of a dissolved crystal in CD2Cl2 showed two hydride signals
at −16.15 and −26.46 ppm and the 31P{1H} NMR spectrum one
signal at 72.0 ppm (see the Supporting Information).
Synthesis of [Ir(H)2(CH3CN)2(L3]BArF (10). Iridium catalyst 9b
(30.0 mg, 10.4 μmol) was added to a Schlenk tube and dissolved in
CH3CN (1.0 mL). The resulting solution was stirred for 1 h at room
temperature, and the solvent was evaporated under vacuum. The
residue was dissolved in CH2Cl2 (1.0 mL), and the solvent was
completely removed at 0.3 mbar. This procedure was repeated two
times. The resulting residue was dried at 0.3 mbar for 16 h to afford
complex 10 as a pale yellow solid (31 mg, 95%). 1H NMR (500 MHz,
CD2Cl2, 263 K): δ 7.72 (br s, 8H, BArF-H), 7.56 (br s, 4H, BArF-H),
7.46 (t, 1H, J = 7.8 Hz, C-12-H), 7.26 (dd, 1H, J = 7.8 Hz, J = 1.2 Hz,
C-11-H), 7.24 (dd, 1H, J = 7.8 Hz, J = 1.2 Hz, C-13-H), 6.98 (d, 1H, J
= 1.9 Hz, C-6-H), 6.87 (d, 1H, J = 1.9 Hz, C-7-H), 4.53 (dt, 1H, J =
13.7 Hz, J = 3.3 Hz, C-5-Ha), 4.13−4.05 (m, 2H, C-2-Ha + C-3-H),
4.01 (br d, 1H, J = 7.5 Hz, C-2-Hb), 3.90 (dd, 1H, J = 13.7 Hz, J = 2.9
Hz, C-5-Hb), 2.29 (s, 3H, C-3′-H), 2.27−2.21 (m, 5H, C-15-H, C-18-
H + C-1′-H), 2.04−1.97 (m, 1H, C-4-Ha), 1.74−1.68 (m, 1H, C-4-
Hb), 1.35 (s, 9H, C-22-H), 1.24 (d, 3H, J = 6.9 Hz, C-17-H), 1.22 (d,
3H, J = 6.9 Hz, C-20-H), 1.03 (d, 3H, J = 6.9 Hz, C-16-H), 1.00 (d,
3H, J = 6.9 Hz, C-19-H), −21.93 (d, 1H, J = 7.3 Hz, Ir-Ha), −26.12 (d,
1H, J = 7.3 Hz, Ir-Hb). 13C{1H} NMR (125 MHz, CD2Cl2, 263 K): δ
Synthesis of [IrH(L2)(μ-H)]2(BArF)2 (9a). Iridium catalyst 3a (100
mg, 62 μmol) was added to a Schlenk tube and dissolved in CH2Cl2
(4.0 mL). The Schlenk tube was placed in liquid N2 until the yellow
solution froze. The Schlenk tube was then evacuated and purged with
hydrogen gas (balloon), and the solution was allowed to warm up to
room temperature. After stirring for 30 min at room temperature, the
solvent was completely evaporated under vacuum. The resulting
yellow foam was washed with hexane (3 × 1.5 mL) and dried at 0.3
mbar for 16 h to afford complex 9a as a yellow solid (93 mg, 95%). 1H
NMR (500 MHz, CD2Cl2, 295 K): δ 7.73 (br s, 16H, BArF-H), 7.57
(br s, 8H, BArF-H), 7.38 (t, 2H, J = 7.7 Hz, C-12-H), 7.28 (dd, 2H, J =
7.8 Hz, J = 1.2 Hz, C-13-H), 7.14 (dd, 2H, J = 7.8 Hz, J = 1.2 Hz, C-
11-H), 6.96 (d, 2H, J = 2.0 Hz, C-7-H), 6.95 (d, 2H, J = 2.0 Hz, C-6-
H), 4.76 (dd, 2H, J = 9.7 Hz, J = 9.7 Hz, C-2-Ha), 4.37−4.33 (m, 2H,
C-3-H), 4.30 (dd, 2H, J = 9.3 Hz, J = 5.8 Hz, C-2-Hb), 4.11 (dd, 2H, J
= 15.2 Hz, J = 8.2 Hz, C-5-Hb), 3.74 (dd, 2H, J = 15.2 Hz, J = 9.2 Hz,
C-5-Ha), 2.73−2.65 (m, 2H, C-18-H), 2.37−2.31 (m, 2H, C-4-Ha),
2.10−2.05 (m, 2H, C-4-Hb), 2.00 (br s, 6H, C-23-H), 1.95 (br d, 6H,
C-22-H), 1.83 (br d, 6H, C-24-H), 1.62 (br d, 6H, C-24-H), 1.52−
1.48 (m, 2H, C-15-H), 1.32 (d, 6H, J = 6.8 Hz, C-17-H), 1.30 (d, 6H, J
= 6.8 Hz, C-20-H), 1.10−1.07 (m, 12H, C-22-H + C-19-H), 0.92 (d,
6H, J = 6.8 Hz, C-16-H), −18.40 (s, 2H, Ir-Hμ), −36.85 (s, 2H, Ir-Ht).
13C{1H} NMR (100 MHz, CD2Cl2, 295 K): δ 181.6 (C-1), 162.2 (q,
1JCB = 50 Hz, C-BArF),147.5 (C-10), 145.5 (C-14), 137.6 (C-9), 135.2
(C-BArF), 133.4 (C-8), 130.3 (C-12), 129.3 (q, 2JCF = 32 Hz, C-BArF),
126.7 (C-6), 125.0 (q, 1JCF = 272 Hz, C-BArF), 124.6 (C-7), 124.4 (C-
1
179.4 (C-1), 161.9 (q, JCB = 50 Hz, C-BArF), 154.2 (C-8), 146.0 (C-
14), 145.6 (C-10), 138.0 (C-9), 134.9 (C-BArF), 129.7 (C-12), 128.9
(q, 2JCF = 32 Hz, C-BArF), 124.7 (q, 1JCF = 272 Hz, C-BArF), 124.1 (C-
13), 124.0 (C-11), 123.2 (C-7), 120.6 (C-6), 119.8 (C-4′), 118.1 (C-
2′), 117.7 (br s, C-BArF), 71.9 (C-2), 70.7 (C-3), 45.6 (C-5), 36.2 (C-
4), 34.9 (C-21), 28.7 (C-15), 28.6 (C-18), 27.9 (C-22), 25.3 (C-16),
25.1 (C-19), 22.1 (C-17), 21.9 (C-20), 4.6 (C-3′), 3.4 (C-1′). Anal.
Calcd for C60H55BF24IrN5O: C, 47.38; H, 3.64; N, 4.60. Found: C,
H
dx.doi.org/10.1021/om4007467 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
47.42; H, 3.66; N, 4.91. ESI-MS (CH2Cl2, 323 K): m/z 617 ([M −
BArF − CH3CN]+).
ABBREVIATIONS
■
L1 = (S)-1-[2-(2-adamantan-2-yl-4,5-dihydrooxazol-4-yl)-
ethyl]-3-(2,6-diisopropylphenyl)-1,2-dihydroimidazol-2-yli-
dene; L2 = (S)-1-[2-(2-adamantan-2-yl-4,5-dihydrooxazol-4-
yl)-ethyl]-3-(2,6-diisopropylphenyl)-1,2-dihydroimidazol-2-yli-
dene; L3 = (S)-1-[2-(2-tert-butyl-4,5-dihydrooxazol-4-yl)-
ethyl]-3-(2,6-diisopropylphenyl)-1,2-dihydroimidazol-2-yli-
dene; COD = 1,5-cyclooctadiene; Cy = cyclohexyl; py =
pyridine; BArF = tetrakis[3,5-bis(trifluoromethyl)phenyl]-
borate; NHC = N-heterocyclic carbene; n.d. = not determined
Synthesis of [Ir(H)2(L3)(6,6′-Bi-2-picoline)]BArF (11). Method
A. Iridium complex 3b (50.0 mg, 32.4 μmol, 1.0 equiv) and 6,6′-
dimethyl-2,2′-bipyridine (59.6 mg, 323.5 μmol, 10.0 equiv) were
added to a Schlenk tube and dissolved in CH2Cl2 (2.5 mL). The
Schlenk tube was placed in liquid N2 until the yellow solution froze.
The Schlenk tube was then evacuated and purged with hydrogen gas
(balloon), and the solution was allowed to warm up to room
temperature. After stirring for 4 h at that temperature, the solvent was
removed under vacuum. The residue was washed with hexane (3 × 6
mL) and dried at 0.3 mbar to afford complex 11 as an orange solid (46
mg, 87%).
Method B. Iridium complex 9b (10.0 mg, 3.5 μmol, 1.0 equiv) and
6,6′-dimethyl-2,2′-bipyridine (6.4 mg, 34.7 μmol, 10.0 equiv) were
added to a Schlenk tube and dissolved in CH2Cl2 (1.0 mL). After
stirring at room temperature for 16 h, the solvent was evaporated
under vacuum. The residue was washed with hexane (3 × 2 mL) and
dried at 0.3 mbar for 16 h to afford complex 11 as an orange solid (10
REFERENCES
■
(1) Reviews: (a) Cui, X.; Burgess, K. Chem. Rev. 2005, 105, 3272.
(b) Roseblade, S. J.; Pfaltz, A. Acc. Chem. Res. 2007, 40, 1402.
(c) Church, T. L.; Andersson, P. G. Coord. Chem. Rev. 2008, 252, 513.
(d) Woodmansee, D. H.; Pfaltz, A. Top. Organomet. Chem. 2011, 34,
31. (e) Woodmansee, D. H.; Pfaltz, A. Chem. Commun. 2011, 47,
7912. (f) Cadu, A.; Andersson, P. G. J. Organomet. Chem. 2012, 714, 3.
(2) (a) Smidt, S. P.; Zimmermann, N.; Studer, M.; Pfaltz, A. Chem.
Eur. J. 2004, 10, 4685. (b) Mazet, C.; Smidt, S. P.; Meuwly, M.; Pfaltz,
A. J. Am. Chem. Soc. 2004, 126, 14176. (c) Dietiker, R.; Chen, P.
Angew. Chem., Int. Ed. 2004, 43, 5513. (d) Brandt, P.; Hedberg, C.;
Andersson, P. G. Chem.Eur. J. 2003, 9, 339.
1
mg, (90%). H NMR (500 MHz, CD2Cl2, 295 K): δ 7.93−7.90 (m,
2H, C-5′-H + C-8′-H), 7.80−7.75 (m, 2H, C-4′-H + C-9′-H), 7.73 (br
s, 8H, BArF-H), 7.56 (br s, 4H, BArF-H), 7.43−7.42 (m, 2H, C-3′-H +
C-10′-H), 7.28 (t, 1H, J = 7.8 Hz, C-12-H), 7.13 (dd, 1H, J = 7.8 Hz, J
= 1.2 Hz, C-11-H), 7.10 (d, 1H, J = 1.9 Hz, C-6-H), 6.98 (dd, 1H, J =
7.8 Hz, J = 1.2 Hz, C-13-H), 6.91 (d, 1H, J = 1.9 Hz, C-7-H), 4.88−
4.80 (m, 2H, C-3-H + C-5-Ha), 4.19 (dd, 1H, J = 8.7 Hz, J = 7.2 Hz,
C-2-Ha), 4.09 (dd, 1H, J = 8.7 Hz, J = 1.2 Hz, C-2-Hb), 4.02 (ddd, 1H,
J = 13.6 Hz, J = 6.2 Hz, J = 3.0 Hz, C-5-Hb), 2.76 (s, 3H, C-1′-H), 2.65
(s, 3H, C-12′-H), 2.40−2.33 (m, 1H, C-18-H), 2.29−2.21 (m, 2H, C-
4-Ha + C-15-H), 1.82 (ddt, 1H, J = 13.6 Hz, J = 11.3 Hz, J = 2.8 Hz, C-
4-Hb), 1.32 (d, 3H, J = 6.9 Hz, C-17-H), 1.06 (d, 3H, J = 6.9 Hz, C-16-
H), 0.96 (d, 3H, J = 6.9 Hz, C-19-H), 0.59 (s, 9H, C-22-H), 0.34 (d,
3H, J = 6.9 Hz, C-20-H), −21.92 (d, 1H, J = 7.3 Hz, Ir-Ha), −28.62 (d,
1H, J = 7.3 Hz, Ir-Hb). 13C{1H} NMR (100 MHz, CD2Cl2, 295 K): δ:
180.7 (C-1), 162.0 (q, 1JCB = 50 Hz, C-BArF), 162.2 (C-2′), 161.5 (C-
11′), 159.6 (C-6′), 158.4 (C-7′), 158.2 (C-8), 146.4 (C-10), 145.3 (C-
14), 138.5 (C-9), 138.2 (C-4′), 137.7 (C-9′), 135.2 (C-BArF), 129.6
(3) (a) Hedberg, C.; Kallstrom, K.; Brandt, P.; Hansen, L. K.;
̈
̈
Andersson, P. G. J. Am. Chem. Soc. 2006, 128, 2995. (b) Church, T. L.;
Rasmussen, T.; Andersson, P. G. Organometallics 2010, 29, 6769.
(c) Mazuela, J.; Norrby, P.-O.; Andersson, P. G.; Pam
̀ ́
ies, O.; Dieguez,
M. J. Am. Chem. Soc. 2011, 133, 13634.
(4) (a) Fan, Y.; Cui, X.; Burgess, K.; Hall, M. B. J. Am. Chem. Soc.
2004, 126, 16688. (b) Cui, X.; Fan, Y.; Hall, M. B.; Burgess, K.
Chem.Eur. J. 2005, 11, 6859. See also: (c) Hopmann, K. H.; Bayer,
A. Organometallics 2011, 30, 2483.
(5) (a) Chodosh, D. F.; Crabtree, R. H.; Felkin, H.; Morris, G. E. J.
Organomet. Chem. 1978, 161, C67. (b) Wang, H. H.; Pignolet, L. H.
Inorg. Chem. 1980, 19, 1470. (c) Wang, H. H.; Casalnuovo, A. L.;
Johnson, B. J.; Mueting, A. M.; Pignolet, L. H. Inorg. Chem. 1988, 27,
325. (d) Xu, Y.; Mingos, D. M. P.; Brown, J. M. Chem. Commun. 2008,
199. (e) Xu, Y.; Celik, M. A.; Thompson, A. L.; Cai, H.; Yurtsever, M.;
Odell, B.; Green, J. C.; Mingos, D. M. P.; Brown, J. M. Angew. Chem.,
Int. Ed. 2009, 48, 582.
(6) Smidt, S. P.; Pfaltz, A.; Martínez-Viviente, E.; Pregosin, P. S.;
Albinati, A. Organometallics 2003, 22, 1000.
(7) For related studies of polynuclear rhodium polyhydride clusters,
see: Kohrt, C.; Baumann, W.; Spannenberg, A.; Drexler, H. J.;
Gridnev, I. D.; Heller, D. Chem.Eur. J. 2013, 19, 7443 and
references therein.
2
(C-12), 129.0 (q, JCF = 32 Hz, C-BArF), 127.3 (C-10′), 126.9 (C-3′),
125.0 (q, 1JCF = 272 Hz, C-BArF), 124.6 (C-11), 123.9 (C-7), 123.8 (C-
13), 121.5 (C-5′), 121.1 (C-6), 120.9 (C-8′), 117.9 (br s, C-BArF), 73.7
(C-3), 72.9 (C-2), 46.6 (C-5), 36.3 (C-4), 34.5 (C-21), 30.8 (C-1′),
28.9 (C-18), 28.9 (C-15), 27.3 (C-22), 26.6 (C-12′), 25.3 (C-16), 24.7
(C-19), 22.8 (C-17), 21.4 (C-20). Anal. Calcd for C68H61BF24IrN5O:
C, 50.32; H, 3.79; N, 4.31. Found: C, 50.22; H, 4.02; N, 4.24. ESI-MS
(CH2Cl2, 323 K): m/z 760 ([M − 2×BArF]2+).
Hydrogenation Reactions. The hydrogenation experiments were
performed as previously described for the corresponding catalysts 411
and 3a.8b
(8) (a) Powell, M. T.; Hou, D.-R.; Perry, M. C.; Cui, X.; Burgess, K. J.
Am. Chem. Soc. 2001, 123, 8878. (b) Perry, M. C.; Cui, X.; Powell, M.
T.; Hou, D.-R.; Reibenspies, J. H.; Burgess, K. J. Am. Chem. Soc. 2003,
125, 113.
ASSOCIATED CONTENT
■
S
* Supporting Information
Additional spectroscopic data and complete labeling of the
complexes is given. Crystallographic data for 6, 7, 8, and 9b in
CIF format. This material is available free of charge via the
Internet at http://pubs.acs.org.
(9) Cui, X.; Fan, Y.; Hall, M. B.; Burgess, K. Chem.Eur. J 2005, 11,
6859.
(10) (a) Mazet, C.; Smidt, S. P.; Meuwly, M.; Pfaltz, A. J. Am. Chem.
Soc. 2004, 126, 14176. (b) Schramm, Y.; Barrios-Landeros, F.; Pfaltz,
A. Chem. Sci. 2013, 4, 2760.
(11) Smidt, S. P.; Menges, F.; Pfaltz, A. Org. Lett. 2004, 6, 2023.
(12) After 20 min in CD2Cl2, various hydride signals, but no signals
of Ir3 clusters, were observed. Upon addition of MeCN, the complex
mixture of hydrides present in pure CD2Cl2 was cleanly converted to
5a and 5b.
(13) For an orientation trans to a phosphorus atom, a coupling
constant of 100−200 Hz would be expected; see: Pregosin, , P. S.
NMR in Organometallic Chemistry; Wiley-VCH: Weinheim, 2012,.
(14) Crabtree, R. H.; Felkin, H.; Fillebeen-Khan, T.; Morris, G. E. J.
Organomet. Chem. 1979, 168, 183.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: andreas.pfaltz@unibas.ch.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support by the Swiss National Science Foundation is
gratefully acknowledged.
(15) Hlatky, G. G.; Johnson, B. F. G.; Lewis, J.; Raithby, P. R. J.
Chem. Soc., Dalton Trans. 1985, 1277.
I
dx.doi.org/10.1021/om4007467 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
́
(16) (a) Tejel, C.; Ciriano, M. A.; Jimenez, S.; Oro, L. A.; Graiff, C.;
Tiripicchio, A. Organometallics 2005, 24, 1105. For dichloromethane
activation with a rhodium complex, see: (b) Blank, B.; Glatz, G.;
Kempe, R. Chem.Asian J. 2009, 4, 321.
(17) The complex shows low solubility in toluene-D8. In acetone-D6,
exchange of acetonitrile ligands with the solvent was observed.
(18) (a) Robertson, G.; Tucker, P. Aust. J. Chem. 1984, 37, 257.
(b) Bianchini, C.; Farnetti, E.; Graziani, M.; Kaspar, J.; Vizza, F. J. Am.
Chem. Soc. 1993, 115, 1753. (c) Schnabel, R. C.; Roddick, D. M.
Organometallics 1993, 12, 704. (d) Schnabel, R. C.; Carroll, P. S.;
Roddick, D. M. Organometallics 1996, 15, 655. (e) Feldman, J. D.;
Peters, J. C.; Tilley, T. D. Organometallics 2002, 21, 4050. (f) Choualeb,
A.; Lough, A. J.; Gusev, D. G. Organometallics 2007, 26, 5224.
(g) Peloso, R.; Pattacini, R.; Cazin, C. S. J.; Braunstein, P. Inorg. Chem.
2009, 48, 11415. (h) Franzke, A.; Voss, F.; Pfaltz, A. Tetrahedron 2011,
67, 4358. (i) Pontiggia, A. J.; Chaplin, A. B.; Weller, A. S. J. Organomet.
Chem. 2011, 696, 2870.
(19) Addition of [H(OEt2)2]BArF to a CD2Cl2 solution of complex 4
1
had no apparent effect on the H NMR spectrum.
(20) (a) [Ir(Cp)(CH3)(CH2Cl2)(P(CH3)3)]BArF: Arndtsen, B. A.;
Bergman, R. G. Science 1995, 270, 1970. (b) [Ir(H)2(OTf)(CH2Cl2)
(H -PhtroppPh) ]: Piras, E.; Lang, F.; Ruegger, H.; Stein, D.; Worle,
̈
̈
̈
2
2
M.; Grutzmacher, H. Chem.Eur. J. 2006, 12, 5849.
̈
(21) The 1H NMR spectrum of the product obtained after
evaporation of the solvent and washing with hexane showed three
additional hydride signals at −3.00, −17.43, and −21.06 ppm with a
ratio of 1:3:3. These chemical shifts are similar to the those observed
for [Ir3(H)6(PHOX)3(μ3-H)](PF6)2; see ref 6.
(22) (a) Selnau, H. E.; Merola, J. S. Organometallics 1993, 12, 3800.
(b) Scott, N. M.; Pons, V.; Stevens, E. D.; Heinekey, D. M.; Nolan, S.
P. Angew. Chem., Int. Ed. 2005, 44, 2512. (c) Castillo, M. R.; Martín,
M.; Fraile, J. M.; Mayoral, J. A.; Sola, E. Angew. Chem., Int. Ed. 2011,
50, 3240.
(23) For a similar low-frequency 13C NMR chemical shift of an
iridium-bound NHC C atom, see: Albrecht, M.; Miecznikowski, J. R.;
Samuel, A.; Faller, J. W.; Crabtree, R. H. Organometallics 2002, 21,
3596.
(24) (a) Cooper, A. C.; Streib, W. E.; Eisenstein, O.; Caulton, K. G. J.
́
Am. Chem. Soc. 1997, 119, 9069. (b) Clot, E.; Eisenstein, O.; Dube, T.;
Faller, J. W.; Crabtree, R. H. Organometallics 2002, 21, 575.
(25) The 1H NMR spectrum showed a complex mixture of products,
whose structures could not be elucidated.
(26) Zhu, Y.; Fan, Y.; Burgess, K. J. Am. Chem. Soc. 2010, 132, 6249.
(27) Similar findings have been reported for N,P-based iridium
dihydride complexes; see refs 10 and 14.
(28) On the basis of 1H and 13C NMR analyses and HR-EI-MS data,
these byproducts were identified as 1,3-dimethyl-1,2,3-triphenyl-
1,2,3,4-tetrahydronaphthalene and two diastereomers of 1-benzyl-3-
methyl-1,3-diphenyl-1,2,3,4-tetrahydronaphthalene. These structural
assignments were confirmed by acid-catalyzed dimerization of (E)-
1,2-diphenyl-1-propene 12, as described in the literature for stilbene
derivatives; see: (a) Aguirre, J. M.; Alesso, E. N.; Moltrasio Iglesias, G.
Y. J. Chem. Soc., Perkin Trans. 1 1999, 1353. (b) Li, X.-C.; Ferreira, D.
Tetrahedron 2003, 59, 1501. Treatment of 12 with 0.5 equiv of sulfuric
acid gave the same products as the hydrogenation experiments (see
the Supporting Information).
(29) Three different batches of complex 3b were used, and the
reactions were performed in different autoclaves.
(30) Brookhart, M.; Grant, B.; Volpe, A. F. Organometallics 1992, 11,
3920.
(31) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.;
Burla, M. C.; Polidori, G.; Camalli, M. J. Appl. Crystallogr. 1994, 27,
435.
(32) Betteridge, P. W.; Carruthers, J. R.; Cooper, R. I.; Prout, K.;
Watkin, D. J. J. Appl. Crystallogr. 2003, 36, 1487.
J
dx.doi.org/10.1021/om4007467 | Organometallics XXXX, XXX, XXX−XXX