Synthesis, structural characterization and the first electroluminescent properties of tris- and bis-cycloiridiated complexes of sterically hindered electron-poor 2-(3,5-bis(trifluoromethyl)phenyl)-4-trifluoromethylpyridine

Article abstract of DOI:10.1039/b514797b

An application of the new sterically hindered electron-poor 2-(3,5-bis(trifluoromethyl)phenyl)-4-trifluoromethylpyridine [HCN] (1) in the one-step high temperature cyclometalation by Ir^IIICl₃ in the presence of Ag^IOC(O)CF<

Full text of DOI:10.1039/b514797b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Synthesis, structural characterization and the first electroluminescent
properties of tris- and bis-cycloiridiated complexes of sterically hindered
electron-poor 2-(3,5-bis(trifluoromethyl)phenyl)-4-trifluoromethylpyridine†
Alex S. Ionkin,* William J. Marshall, D. Christopher Roe and Ying Wang
Received 20th October 2005, Accepted 10th February 2006
First published as an Advance Article on the web 22nd February 2006
DOI: 10.1039/b514797b
An application of the new sterically hindered electron-poor 2-(3,5-bis(trifluoromethyl)phenyl)-
4-trifluoromethylpyridine [HC∧N] (1) in the one-step high temperature cyclometalation by IrIIICl3 in the
presence of AgIOC(O)CF3 resulted in the synthesis of tris-cyclometalated complexes [C∧N]2Ir[C∧C] (3)
and [C∧N]3Ir (5). A neutral silver cluster with a repeating unit of hexa-silver groups in an infinite chain
of (2) was isolated from the above reaction as well. When this cyclometalation was carried out in
trimethylphosphate at lower temperature, bis-cyclometalated derivatives [C∧N]2Ir(l-Cl)2Ir[C∧N]2 (6),
2
[C∧N]2Ir[g -(O(C(tBu))2CH] (7), and [C∧N]2Ir(l-O-P(OMe)2-O)2Ir[C∧N]2 (8) were synthesized.
According to X-ray analyses complex (3), while trivalent, contains four cyclometalated single Ir–C
bonds. One of the Ir–C bonds, next to the nitrogen atom of the C∧C pyridinium ligand, was found to be
˚
the shortest to date (1.977(4) A) for a single bond between iridium and carbon atoms. The coordination
of the C∧C ligand in (3) to iridium has a decidedly interesting bonding pattern and can be explained by
various formulations. The first one is considering this ligand as a monoanionic chelating ligand, in
which the second coordination site arises from a carbene or azomethine ylide. Overall the best single
picture may be a dianionic ligand making two “normal” Ir–C bonds, in which the ligand just happens
to contain a pyridinium function that compensates for one negative charge on the iridium. LEDs
constructed with compounds (7) and (8) give blue-green emission with peak electroluminescent
efficiency of 15 and 2 cd A−1, respectively. An LED constructed with compound (5) gives a yellowish
emission with peak electroluminescent efficiency of 5.5 cd A−1.
iridium atom inthe phenylmoiety of the cyclometaledligand(Type
Introduction
B in Scheme 1) was found to hypochromically shift the emission
to the blue part of the spectrum.4 The blue emissive materials are
among the most sought after components for an emerging OLED
technology.
Tennant discovered the element iridium over 200 years ago in the
black residues remaining from the treatment of platinum ores.1
Since then iridium has been linked to phenomena ranging from
the disappearance of dinosaurs2 to organic light emitting diodes
(OLEDs).3 Particularly, bis- and tris-cyclometalated (N∧C) com-
plexes of IrIII with 2-phenylpyridine ligands (Type A in Scheme 1)
have been used widely as one of the promising electroluminescent
materials for OLED technology. The emission colors from these
complexes depends on the choice of cyclometalated ligand,
ranging from blue-green, to green and to red. The introduction
of trifluoromethyl groups in the para- and ortho-positions to the
In our quest for blue phosphorescent materials, we decided
to synthesize bis- and tris-cyclometalated (N∧C) complexes of
IrIII with 2-phenylpyridine ligands bearing three trifluoromethyl
groups: two in the phenyl moiety (like in Type B in Scheme 1) and
one additional group in the pyridine moiety para to the nitrogen
(Type C in Scheme 1). There are two effects of the trifluoromethyl
group in the Ir-cyclometalated complexes. The first one is the
strong electron withdrawing effect on the ligand p-system, and the
second is to provide steric protection around the metal, which
is an important parameter for increasing the intensity of the
emission.4
In this report we describe the isolation of several luminescent
bis- and tris-cyclometalated (N∧C) Ir-complexes based on a new
sterically hindered fluorinated 2-(3,5-bis(trifluoromethyl)phenyl)-
4-trifluoromethylpyridine 1. The high temperature cyclometala-
tion of 1 by IrIIICl3 in the presence of AgIOC(O)CF3 afforded
the neutral silver cluster 2 containing six silver atoms, the first
mono-nuclear trivalent iridium complex 3 with four cyclometa-
lated iridium–carbon bonds with a decidedly interesting bonding
pattern, bipyridine 4 and target tris-cyclometalated product 5.
Bis-cyclometalated derivatives 6–8 were prepared through the
cyclometalation of 1 in trimethylphosphate as a medium.
Scheme 1
DuPont Central Research & Development, Experimental Station, Wilming-
ton, DE, 19880-0328, USA. E-mail: alex.s.ionkin@usa.dupont.com
† This is DuPont contribution #8572.
2468 | Dalton Trans., 2006, 2468–2478
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
was 18.0 g (91.0%) after distillation as a colorless liquid with bp
84–86 ◦C/0.1 mmHg. The 19F NMR spectrum of 1 contains
two chemical shifts at d = −62.37, which corresponds to the
trifluoromethyl groups of the phenyl moiety, and at d = −64.30,
which corresponds to the trifluoromethyl groups of the pyridine
moiety, with the expected ratio between them as two to one. The
cyclometalation experiments with compound 1 are described in
the following sections.
Results and discussion
Synthesis of 2-(3,5-bis(trifluoromethyl)phenyl)-
4-trifluoromethylpyridine 1
The palladium-catalysed Suzuki cross-coupling reaction between
2-chloro-4-trifluoromethylpyridine and 3,5-bis(trifluoromethyl)-
phenylboronic acid was used to prepare 2-(3,5-bis(trifluoromethyl)-
phenyl)-4-trifluoromethylpyridine 1 (Scheme 2).
Synthesis, structural and bonding studies of the tris-cyclometalated
derivatives (3 and 5)
One-step synthesis leading directly to tris-cyclometalated IrIII
complexes was reported by the reaction of IrIIICl3 with 2-
phenylpyridines in the presence of AgI trifluoroacetate.4 AgI
trifluoroacetate was used to facilitate tris-cyclometalation by
generating in situ IrIII trifluoroacetate, which is a more active tris-
cycloiridation agent than IrIIICl3.4 We decided to synthesize the
tris-cyclometalated complex by the above one-pot method from 2-
(3,5-bis(trifluoromethyl)phenyl)-4-trifluoromethylpyridine 1 and
IrIIICl3/AgI trifluoroacetate (Scheme 3). Unexpectedly, several
compounds in this reaction were isolated with unusual bonding
patterns: the neutral silver cluster 2 containing six silver atoms,
the first mono-nuclear trivalent iridium complex 3 with four
cyclometalated iridium–carbon bonds with a decidedly interesting
bonding pattern, bipyridine 4 and the target tris-cyclometalated
product 5. The isolated yields of compounds 2–5 were rather low,
varying from 3.4% for 4 to 9.8% for 2.
Scheme 2
The synthetic protocol involves Pd2dba3/di-tert-butyl-
trimethylsilylanylmethylphosphine as the catalyst in the presence
of caesium fluoride in 1,4-dioxane as the solvent. The yield of
2-(3,5-bis-trifluoromethyl-phenyl)-4-trifluoromethylpyridine 1
Scheme 3 Cyclometalation of 1 by IrIIICl3 in the presence of AgOC(O)CF3. There is one trifluoroacetate ligand per silver atom in 2. The silver and
oxygen atoms in brackets are from neighboring repeating units.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 2468–2478 | 2469
©

View Article Online
The 2-phenylpyridine ligands 1 behave in 2 as blocking groups
or terminators to prevent further aggregation of silver trifluoroac-
etate moieties in two dimensions: above and below the infinite
chain of silver atoms (Scheme 3 and Fig. 1).5
atom grouping in silver clusters.5d The repeating units in the cluster
are bridged together by six-coordinate silver atoms forming Ag–
O–Ag four-membered rings. Complex 3 was isolated as dark gold
crystals with orange photoluminescence. The diamagnetic nature
of 3 (judging from sharp lines in the NMR spectra of 3) suggests
a trivalent status. In a four-valent state, complex 3 should be
paramagnetic, consistent with a d5 iridium. According to X-ray
analysis, the nitrogen atoms coordinated to iridium are in cis-
positions (N1 and N15 atoms in Fig. 2).
Fig. 2 An ORTEP drawing of 3. The shortest Ir–C(30) bond length is
˚
1.977(4) A. Thermal ellipsoids drawn to the 30% probability level. All
hydrogen atoms and pentane solvent molecule are removed for clarity.
Fig. 1 An ORTEP drawing of centrosymmetric dimer 2. The phenyl
moieties of the 2-phenylpyridine ligands 1 are assembled in face-to-face
mode of p–p stacking. The centroid–centroid contacts between the phenyl
The bond between iridium and carbon next to the nitrogen
atom of the C∧C pyridinium ligand was found to be among the
˚
shortest for cyclometalated Ir–C bonds (1.977(4) A, Table 1). The
˚
groups in 2 are 3.96 A. The mean planes of the two phenyl groups are tilted
recent account6 of Ir–C bonds of tris-cyclometalated structures
by 11.2◦. The hydrogen and fluorine atoms are omitted for clarity. Thermal
ellipsoids drawn to the 20% probability level. Symmetry operation codes
are designated on atom labels with A = (x, y, z), B = (1 − x, −y, −z),
C = (1 + x, y, z) and D = (2 − x, −y, −z).
˚
gives a range from 2.024(6) to 2.151(9) A for the 2-phenylpyridine
derivatives. The second Ir–C bond length of this C∧C pyridinium
˚
ligand (2.065(4) A) is within the range found for other Ir–C bonds
in this report with the same cyclometalated N∧C ligand 1 (Table 1,
structure 5) and within the literature data.6 Both of these Ir–C
bonds are under the same trans-influence of the nitrogen atoms
of the pyridine rings. The bonding pattern of the C∧C pyridinium
ligand to iridium is not a trivial issue, but it can be rationalized
(formulated) by the structures depicted in Scheme 4.
Cluster 2 contains two kinds of silver atoms: the two peripheral
three-coordinate silver atoms each connected with two nitrogen
atoms of two ligands 1; and a four-atom flat core with two six-
coordinate and two seven-coordinate silver atoms in the center of
the cluster (Scheme 3). Within the four atoms, the Ag–Ag distances
˚
˚
are 3.00 A, which are less than 3.4 A (less than twice the van der
Waals radius of a silver atom). The central cage is perfectly planar,
in contrast to the butterfly geometry previously observed for four-
The experimental values from the X-ray analyses for a single
bond length between iridium and N-heterocyclic carbenes vary
7
˚
˚
from 2.019 A to 2.055 A. Our corresponding value for the Ir–C
˚
Table 1 Selected bond lengths (A) for 3, 5, 7 and 8
3
5
7
8
Ir–C
Ir–C
Ir–C
Ir–C
Ir–N
Ir–N
Ir–N
1.977(4), Ir–C30
2.136(4), Ir–C11
2.119(4), Ir–C24
2.065(4), Ir–C45
2.125(3)
2.123(3)
N/A
N/A
N/A
N/A
2.156(7), Ir–C11
2.122(7), Ir–C39
2.061(6), Ir–C25
2.123(5)
2.051(5)
2.048(5)
2.039(5), Ir–C11
2.030(5), Ir–C25
N/A
2.032(4)
2.031(4)
2.022(8), Ir–C13
2.013(8), Ir–C28
N/A
2.037(6)
2.026(6)
N/A
N/A
2470 | Dalton Trans., 2006, 2468–2478
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Scheme 4 Three plausible modes of bonding in complex 3.
˚
radicals in the presence of trifluoacetate anions can lead to the
formation of the intermediate pyridinium salt, parent to the C∧C
ligand in 3, as in Scheme 2. This intermediate pyridinium salt could
be incorporated in to complex 3 upon subsequent cyclometalation.
The pyridinyl radical can attack the tris-cyclometalated complex
5 with subsequent rearrangements leading to 3. The studies
of the thermal stability of AgI co◦mplexes with pyridines show
decomposition occurs above 130 C.13 This simple test of the
stability of silver cluster 2 reveals that it deposits a silver mirror
upon heating to 200 ◦C. However, it is hard to carry out ESR
studies on the radical species formed because these temperatures
lie outside the range of our ESR instrumentation. Further studies
of the mechanism of formation of compounds 2 and 3 have not
been pursued at this point. Besides thermal decomposition of silver
clusters leading to pyridine radicals, several plausible routes can
be envisioned for the formation of the observed products. Silver
catalyzed oxidation of starting ligand 1 to biphenyl derivatives
with subsequent cyclometalation, or thermal decomposition of
iridium cyclometalated complexes leading to products 4 and 3,
are just two of the additional possibilities.
bond in complex 3 is even shorter at 1.977(4) A. However, this
˚
=
distance is longer than the Ir C double bond (1.904 A) observed
in a known alkylidene complex [Cp*(PiPr3)Ir CPh2]. This C–Ir
distance in 3 is shorter than typical single Ir–C bonds (2.1 A).
=
8
˚
The resonance of the carbene moiety is a very characteristic
feature for iridium–carbene complexes in the 13C NMR spectum.
It appears downfield between d = 155 and 181 ppm.7 The most
downfield signal in the 13C NMR spectra of 3 has a chemical shift
at d = 187.24 ppm (Table 2), which is in agreement with the carbene
formulation. In the reference compounds, the chemical shifts for
the Ir–C carbon atoms in tris-cyclometalated complex 5 appear
upfield with respect to 3.
1-(Pyrididin-2-yl)-pyridinium perchlorates resembling the core
of the C∧C cyclometalated ligand of 3 are known.9 The most
downfield signals in the 13C NMR spectrum of reported 1-
(pyrididin-2-yl)-pyridinium perchlorates were observed between
d = 155 and 158 ppm, which corresponds to the carbon atoms
connected to the pyridinium nitrogen. Compound 3 exhibits
chemical shifts close to that range, but more deshielded.
2,2ꢀ- Or 2,4ꢀ-bipyridine ligands with one nitrogen atom quat-
ernized are known to undergo mono-C-cyclometalation to form
PtII and PdII complexes with an internal pyridinium fragment.10
Our studies did not begin with the pyridinium salt of the C∧C
cyclometalated ligand of 3, but rather this ligand structure had to
be generated in situ. The mechanism of the formation of this ligand
and ultimately complex 3 itself seems to involve highly reactive
intermediates, like pyridine-2 radicals. The pyridine radicals gener-
ated in situ by the different methods are known to convert to 4,4ꢀ- or
2,2ꢀ-bipyridines.11 The pyridine radicals generated in the presence
of other aromatic substrates form various isomeric heteroaromatic
products including the pyridinium salts.12 The cyclometalation of
ligand 1 by IrCl3 was carried out at high temperatures from 180 to
190 ◦C for 58 hours. Under these harsh conditions the silver cluster
2 could generate pyridinyl radicals, which in turn are responsible
for the formation of 3 and 4. The dimerization of the pyridinyl
The ORTEP drawing of tris-cyclometalated complex 5 is shown
in Fig. 3. In contrast to the previously reported sterically less bulky
fluorinated analogs, 5 exists as a meridional isomer, not the facial
isomer.4 When the complex assumes a meridional conformation,
sterical repulsion between the C∧N ligands is reduced.6
Synthesis and structural studies of the bis-cyclometalated
derivatives (6–8)
The bis-cyclometalation of 2-(3,5-bis(trifluoromethyl)phenyl)-4-
trifluoromethylpyridine 1 was carried out in trimethylphosphate.
Electrophilic reactions, like cyclometalation, in trimethylphos-
phate take place at lower temperature in many difficult cases.14
The IrIII-bridged di-chloride dimer 6 was precipitated from
the trimethylphosphate solution by dilution with diethyl ether
Table 2 13C NMR data for carbon atoms attached to iridium in complexes 3 and 5
3
5
Cyclometalated
Ligand
d (ppm)
C∧N
1
169.06
C∧N
1
168.23
C∧C
Pyridinium
187.24, 154.65
C∧N
1
172.14
C∧N
1
167.36
C∧N
1
152.07
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 2468–2478 | 2471
©

View Article Online
Scheme 5 The cyclometalation of 2-(3,5-bis(trifluoromethyl)phenyl)-4-trifluoromethylpyridine 1 in trimethylphosphate and related transformations.
Fig. 4 ORTEP drawing of complex 7. This is a trans-N,N-isomer. The
Fig. 3 ORTEP drawing of tris-cyclometalated complex 5. This is a
˚
shortest Ir–C(25) bond length is 2.030(5) A. Thermal ellipsoids drawn to
˚
meridional isomer. The shortest Ir–C(25) bond length is 2.061(6) A.
the 50% probability level. All hydrogen atoms are removed for clarity.
Thermal ellipsoids drawn to the 50% probability level. Only one of two
molecules in the asymmetric unit is depicted and all hydrogen atoms are
removed for clarity.
observed to be the trans-N,N isomer. The eight-membered ring
formed by the bridging iridium dimethoxyphosphate groups has a
chair conformation. There is an averaging between the phosphoryl
(Scheme 5). Complex 6 reacted with 2,2,6,6-tetramethyl-heptane-
3,5-dione and tetrabutylammonium hydroxide in THF to yield
bis-cyclometalated complex 7.
According to the results of X-ray analysis, complex 7 was
found to be a trans-N,N-isomer (Fig. 4). Prolonged heating of
complex 6 in trimethylphosphate was found to form complex 8
with two bridged dimethoxyphosphate moieties (Scheme 4). The
structure of 8 was confirmed by X-ray analysis (Fig. 5). It was
=
(P O) and single phosphorus–oxygen (P–O) bonds in 8. The
averaged P–O bond in the ring system of 8 is 1.483(5) A, which
˚
is shorter than the average single bond between phosphorus and
˚
the exo methoxy group in complex 8 (1.5745(5) A). The shortest
˚
Ir–C(8) bond length is 2.013(8) A in 8, which is positioned trans
to the oxygen atom connected to phosphorus.
2472 | Dalton Trans., 2006, 2468–2478
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Fig. 6 Thin film absorption (+), excitation (᭹), and luminescence spectra
(᭺) of compound 5, thickness 120 nm. Excitation spectrum was monitored
at 500 nm. Luminescence spectrum was excited at 350 nm.
Fig. 5 ORTEP drawing of centrosymmetric dimer 8. The shortest Ir–C(8)
˚
bond length is 2.013(8) A. Thermal ellipsoids drawn to the 30% probability
level. Hydrogen atoms, disordered CF3 positions and dimethylphosphate
are removed for clarity. Symmetry operation codes are designated on atom
labels with A = (2 − x, −y, −z).
The order of the trans influence on the Ir–C bond length in
complexes 3, 5, 7, 8 bearing the same (N∧C) cyclometalated
2-(3,5-bis(trifluoromethyl)phenyl)-4-trifluoromethylpyridine 1 lig-
and can be summarized as follows: C > N > O–(C) > O–(P).
The longest Ir–C(sp2) bonds have been found to be in trans
positions to carbon, then in trans positions to nitrogen, then in
trans positions to oxygen connected to carbon and the shortest Ir–
C(sp2) bonds have been found when the carbon lies trans to oxygen
connected to phosphorus. This sequence correlates perfectly with
the electronegativities of these elements in the Periodic Table of
elements. One exception in this series is the Ir–C bond formed by
“carbene” carbon in complex 3, which is trans to nitrogen (Table 1,
second column for C∧C cyclometalated ligand).
Fig. 7 Thin film absorption (+), excitation (᭹), and luminescence spectra
(᭺) of compound 7, thickness 150 nm. Excitation spectrum was monitored
at 490 nm. Luminescence spectrum was excited at 350 nm.
Photoluminescent and electroluminescent properties of
cyclometalated complexes
Thin film absorption, luminescence, and excitation spectra of
cyclometalated Ir compounds 5, 7, and 8 are shown in Fig. 6–8.
Compound 3 showed thermal decomposition upon sublimation
and was not studied for these properties. Luminescent wavelengths
for all compounds studied are in the blue-green region. In all cases,
the excitation spectra agree with the absorption spectra (Fig. 6–8).
The peak luminescent wavelengths are summarized in Table 3. The
similarity between the spectra indicates that the luminance mainly
originates from the fluorinated phenylpyridine ligand perturbed by
the charge transfer interaction with Ir. The presence of the third
ligand has only a minor effect on the photo-luminescent spectrum.
Compound 5 shows less vibronic structure than compounds 7 and
Fig. 8 Thin film absorption (+), excitation (᭹), and luminescence spectra
(᭺) of compound 8, thickness 150 nm. Excitation spectrum was monitored
at 490 nm. Luminescence spectrum was excited at 350 nm.
Table 3 Summary of the photoluminescent and electroluminescent data for cyclometalated Ir compounds
Peak electroluminescence efficiency/cd A−1
Electroluminescence peak wavelength/nm
Photoluminescence, peak wavelength/nm
Compound 5
Compound 7
Compound 8
5.5 at 15 V
15 at 12 V
2 at 17 V
597
492
500
497
492
493
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 2468–2478 | 2473
©

View Article Online
8, which could indicate the existence of a stronger metal–
ligand charge transfer interaction in that compound.
Preliminary device work was done using compounds 5, 7, and
8 as the electroluminescent emitters. The device configuration
consists of ITO as the anode, MPMP (30 nm) as the hole transport
material, Ir emitters (40 nm), DPA (10 nm) as the electron
transport material, AlQ (30 nm) as the electron injection material,
and LiF/Al as the cathode. Molecular structures of the materials
used are shown in Scheme 6.
Fig. 11 (a) Electroluminescent efficiency vs voltage for compound 8;
(b) comparision of photoluminescent and electroluminescent spectra of
compound 8.
In spite of the similarity of their photoluminescent spectra, the
electroluminescent spectra show more deviations from each other.
In particular, compound 5 gives a yellowish emission which is
red-shifted by about 100 nm from its photoluminescent spectrum.
There is little doubt that this is either excimer or exciplex emission
due to molecular interaction in the solid state. Since the same
hole transport (MPMP) and electron transport (DPA) layers are
used for all three OLED devices and all three Ir compounds have
essentially the same lumiphore (phenylpyridine ligand), but only
compound 5 shows the dramatic red-shifted electroluminescence,
exciplex formation between the Ir compound and the neighboring
layer can be ruled out as the origin of the red-shifted luminescence.
The most likely mechanism is due to excimer formation between
the Ir complexes. The difference between compounds 5, 7, and
8 may be attributed to different solid state packing induced
by the presence of the third ligand which is different for each
compound. Therefore although the solid state photoluminescent
spectra of all three compounds are similar, originating from the
same fluorinated phenylpyridine ligand, the presence of the third
ligand does affect solid state packing and the electroluminescent
spectra. Since the effect of the third ligand shows up in the
electroluminescence spectra rather than the photoluminescence
spectra, it suggests that the transport of electrons and holes
are affected by the presence of the third ligand. The detailed
mechanism for this difference still needs to be understood with
further studies.
Scheme 6
Compound 7 gives blue-green electroluminescence with a peak
located at 492 nm. The maximal electroluminescence efficiency is
15 cd A−1 at 12 V. Compound 8 gives green electroluminescence
with a peak located at 500 nm and a maximal electroluminescence
efficiency of 2 cd A−1 at 17 V. Compound 5 gives yellowish
electroluminescence at 597 nm with a peak efficiency of 5.5 cd A−1
at 15 V. These results are displayed in Fig. 9–11 and summarized
in Table 3.
Conclusion
Fig. 9 (a) Electroluminescent efficiency vs voltage for compound 5;
(b) comparison of photoluminescent and electroluminescent spectra of
compound 5.
The application of sterically hindered electron poor 2-
phenylpyridine 1 in the tris-cyclometalation by IrCl3 resulted in
the isolation of the first mono-nuclear trivalent iridium complex
3 with four cyclometalated Ir–C bonds. One of the Ir–C bonds of
the C∧C ligand in 3 was found to be the shortest to date (1.977(4)
˚
A) for a single bond between iridium and carbon atoms. The mode
of coordination of the C∧C ligand in 3 to iridium can be explained
by various formulations. The first one is considering this ligand as
a monoanionic chelating ligand, in which the second coordination
site arises from a carbene or azomethine ylide. The azomethine
ylide/carbene question is largerly one of semantics because these
are just resonance structures. Singlet carbenes coordinated to
metals show small valence angles at carbon and the carbon–metal
interaction is often weak and long.7 As was mentioned above 3
contains the shortest single Ir–C bond, this is why it may not
be completely fair to formulate this C∧C ligand as containing
Fig. 10 (a) Electroluminescent efficiency vs voltage for compound 7;
(b) comparision of photoluminescent and electroluminescent spectra of
compound 7.
2474 | Dalton Trans., 2006, 2468–2478
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
a carbene—although one can, if one considers that the carbene
center is incorporated in a six-membered ring and not in the, more
typical, five-membered ring systems, for which the major structural
data are available. Overall the best single picture may be a dianionic
ligand making two “normal” Ir–C bonds, in which the ligand just
happens to contain a pyridinium function that compensates for
one negative charge on the iridium.
nium hydroxide as 55–60% solution in water were purchased from
Aldrich. Iridium(+3) chloride trihydrate was purchased from Alfa
Aesar. 2-Chloro-4-trifluoromethyl-pyridine was purchased from
Matrix Scientific.
Syntheses
2-(3,5-Bis(trifluoromethyl)phenyl)-4-trifluoromethylpyridine (1).
15.0 g (0.058 mol) of (3,5-bis(trifluoromethyl)phenyl)boronic acid,
10.00 g (0.055 mol) of 2-chloro-4-trifluoromethylpyridine, 2.66 g
(0.0029 mol) of tris(dibenzylideneacetone)dipalladium(0), 1.62 g
(0.00698 mol) of di-tert-butyl-trimethylsilylmethylphosphine,
10.60 g (0.0698 mol) of caesium fluoride and 150 ml of dioxane
were stirred at room temperature for 24 hours. The resulting
mixture was poured into 200 ml of water and extracted twice
with 200 ml of methylene chloride. The organic phase was
dried on magnesium sulfate overnight and filtered. The solvent
was removed on a rotovapor and the residue was purified by
vacuum distillation. Yield of 2-(3,5-bis(trifluoromethyl)phenyl)-4-
trifluoromethylpyridine was 18.0 g (91.0%) as a colorless liquid
with bp 84–86 ◦C/0.1 mmHg. 19F NMR (377 MHz, CD2Cl2)
d = −62.37 (s, 6F, CF3), −64.30 (s, 3F, CF3). Anal. Calcd. for
C14H6F9N (mol. wt.: 359.19): C, 46.81; H, 1.68; F, 47.60; N, 3.90.
Found: C, 46.89; H, 1.92; F, 47.50; N, 4.15%. LC/MS calculated
for C14H6F9N: 359.04. Found: 359.04.
The thermally stable to the sublimation tris- and bis-
cyclometalated derivatives 5, 7, and 8 were used to construct
LED devices. Preliminary results show that they emit yellow, blue-
green and green colors correspondingly with moderate efficiencies.
Further work is required to optimize the device efficiency.
Experimental
General procedures
All air-sensitive compounds were prepared and handled un-
der a N2/Ar atmosphere using standard Schlenk and inert-
atmosphere box techniques. Anhydrous solvents were used in
the reactions. Solvents were distilled from drying agents or
passed through columns under an argon or nitrogen atmo-
sphere. 3,5-Bis(trifluoromethyl)phenylboronic acid, caesium flu-
oride, tris(dibenzylideneacetone) dipalladium(0), trimethylphos-
phate, 2,2,6,6-tetramethyl-heptane-3,5-dione, tetrabutylammo-
1
Two-dimensional H and 13C NMR correlation data for com-
pound 1 (Tables 4–6), numbering of atoms in 1 is shown in
Table 4 COSY data for compound 1. 1-H Chemical shifts in F1
Scheme 7.
F2
H3
H5
H6
H9
H11
Cyclometalation of 2-(3,5-bis(trifluoromethyl)phenyl)-4-trifluoro-
methylpyridine (1) by IrIIICl3 in the presence of AgOC(O)CF3.
13.04 g (0.0363 mol) of 2-(3,5-bis(trifluoromethyl)phenyl)-4-
trifluoromethylpyridine, 2.13 g (0.00604 mol) of iridium(III) chlo-
ride trihydrate, 4.80 g (0.0217 mol) of silver trifluoroacetate and
40 ml of water were stirred at room temperature under a flow
F2_av
8.048 7.587 8.919 8.552 7.991 F1_av
F1
8.047 8.047 8.047
7.587 7.588 7.586
8.917 8.918 8.918
8.047
7.587
8.918
H3
H5
H6
H9
H11
8.551 8.551 8.551
7.990 7.991 7.990
Table 5 HSQC data for compound 1 (single-bond correlation)
F2
H3
H5
7.580
119.771
H6
8.911
H9
8.545
H11
7.983
F2_ppm
8.040
116.655
13C
F1
116.655
119.771
151.787
127.680
123.874
C3
C5
C6
C9
C11
151.787
127.680
123.874
H11
Table 6 HMBC data for compound 1 (multiple-bond correlation)
F2
H3
8.042
H5
7.582
H6
8.914
H9
F2_av
8.547
155.774
116.655
7.985
155.769
13C
F1
C2
C3
C4
C5
C6
C7
C8
155.763
116.635
140.144
119.751
151.758
123.435
140.547
155.763
116.632
140.127
119.745
151.763
123.454
140.581
155.775
116.635
140.187
119.767
151.771
123.461
155.769
116.639
140.153
119.754
151.764
123.450
140.564
127.642
132.796
123.846
124.018
140.562
127.639
132.796
123.848
124.016
140.565
127.645
132.795
123.845
124.020
C9
C10
C11
C14
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 2468–2478 | 2475
©

View Article Online
no mp until 200 ◦C. 1H NMR (500 MHz, CD2Cl2, TMS) d 6.90 (m,
2H, arom-H), 7.30 (s, 2H, arom-H), 7.52 (s, 1H, arom-H), 7.70–
7.90 (m, 3H, arom-H), 8.10–8.40 (m, 7H, arom-H). 19F NMR
(377 MHz, CD2Cl2) d −59.20 to −65.91 (m, 27F, CF3). Anal.
Calcd. for C42H15F27IrN3 (mol. wt.: 1266.76): C, 39.82; H, 1.19;
N, 3.32. Found: C, 40.07; H, 1.38; N, 3.50%. The structure was
proven by X-ray analysis.
Di-l-chlorotetrakis[4,6-bis(trifluoromethyl)-2-(4-trifluoromethyl-
Scheme 7 Numbering of atoms in 1 for 1H and 13C NMR correlation
2-pyridinyl-jN)phenyl-jC]diiridium (6). 5.0
g (0.0139 mol)
of 2-(3,5-bis(trifluoromethyl)phenyl)-4-trifluoromethylpyridine,
2.21 g (0.00627 mol) of iridium(III) chloride trihydrate, and 40 ml
of trimethylphosphate was stirred at 90 ◦C for 6 hours under
the flow of nitrogen. The resulting solution was poured into
200 ml of diethyl ether. The formed precipitate was filtered off
and dried under 1.0 mmHg vacuum. The yield of the dimer was
5.83 g (88.7%) as a yellow powder. The crude chlorodimer was
used “as is” in the next steps according to established practice in
cyclometalated iridium research.3
experiments.
of nitrogen for 12 hours. Then, the reaction mixture was placed
into a heating bat◦h and the temperature of the bath was gradually
increased to 180 C during 10 hours. The formed trifluoroacetic
acid and water were removed from the reaction mixture by the
constant flow of nitrogen. Heating was continued for the next
48 hours at the above conditions. Then, the reaction mixture
was cooled to ambient temperature and dissolved in 200 ml of
methylene chloride. The concentrated filtrate yielded silver cluster
2 upon standing for 12 hours at room temperature. The yield of
silver cluster 2 was 2.47 g (9.8%) as colorless crystals with no mp
until 200 ◦C. 1H NMR (500 MHz, CD2Cl2, TMS) d 7.50–9.00 (m,
12H, arom-H). 19F NMR (377 MHz, CD2Cl2) d −63.72 (s, 12F,
CF3), −65.67 (s, 6F, CF3), −73.60 (s, 9F, CF3). Anal. Calcd. for
C34H12Ag3F27N2O6: (mol. wt.: 1381.03): C, 29.57; H, 0.88; N, 2.03.
Found: C, 29.74; H, 1.03; N, 2.33%. The structure was proven by
X-ray analysis. The residue from the crystallization was purified by
chromatography on silica gel with petroleum ether (bp 35–60 ◦C)–
ethyl ether 10 : 1 as eluent.
Bis[4,6-bis(trifluoromethyl)-2-(4-trifluoromethyl-2-pyridinyl-
jN)phenyl-jC](2,2,6,6-tetramethyl-3,5-heptanedionato-jO,jOꢀ )-
iridium (7). 2.68 g (0.00142 mol) of 6, 3.0 g (0.0163) of 2,2,6,6-
tetramethyl-heptane-3,5-dione, 4.0 g of a 55–60% solution of
tetrabutylammonium hydroxide in water and 20 ml of THF were
refluxed for 2 hours under argon atmosphere. The reaction mixture
was poured into 200 ml of water and extracted twice with 200 ml
of diethyl ether. The extracts were dried over magnesium sulfate
overnight. The solvent was removed in a rotavapor and the residue
was purified by chromatography on silica gel with petroleum ether
◦
(bp 35–60 C)–ethyl ether 10 : 0.5 as eluent. The yield of 7 was
◦
1
Bis[4,6-bis(trifluoromethyl)-2-(4-trifluoromethyl-2-pyridinyl-
jN)phenyl-jC][bisphenylpyridine]iridium (3). The yield of 3
(from the above cyclometalation of 1) was 0.74 g (4.8%) as dark
2.41 g (77.7%) as a yellow solid with no mp until 200 C. H
NMR (500 MHz, CD2Cl2, TMS) d 0.80 (s, 18H, t-Bu), 5.45 (s,
=
1H, H–C ), 7.20 (m, 2H, arom-H), 7.55 (m, 2H, arom-H), 8.15
◦
1
(m, 6H, arom-H). 19F NMR (377 MHz, CD2Cl2) d −60.44 (s, 6F,
CF3), −63.07 (s, 6F, CF3), −65.74 (s, 6F, CF3). Anal. Calcd. for
C39H29F18IrN2O2 (mol. wt.: 1091.85): C, 42.90; H, 2.68; N, 2.57.
Found: C, 43.25; H, 2.87; N, 2.71%. The structure was proven by
X-ray analysis (Fig. 4).
gold crystals from pentane with no mp until 200 C. H NMR
(500 MHz, CD2Cl2) d 0.90–1.12 (br, 12H, pentane), 7.30 (m, 2H,
arom-H), 7.59 (m, 3H, arom-H), 7.80 (m, 2H, arom-H), 8.05 (m,
4H, arom-H), 8.20 (m, 2H, arom-H), 8.40 (m, 3H, arom-H), 8.63
(m, 2H, arom-H), 8.90 (m, 1H, arom-H). 19F NMR (377 MHz,
CD2Cl2) d −57.68 to −66.64 (m, 36F, CF3). Anal. Calcd. for the
pentane adduct C61H31F36IrN4 (mol. wt.: 1696.09): C, 43.20; H,
1.84; N, 3.30. Found: C, 43.41; H, 1.74; N, 3.59%. The structure
was proven by X-ray analysis.
Di-l-(dimethylphosphato-jO,jOꢀꢀ )tetrakis[4,6-bis(trifluoro-
methyl)2-(4-trifluoromethyl-2-pyridinyl-jN)phenyl-jC]diiridium
(8). 2.50 g (0.00132 mol) of 6, and 70 ml of trimethylphosphate
was stirred at 90 ◦C for 7 days under a flow of nitrogen. The
formed precipitate was filtered off and dried under 1.0 mmHg
vacuum. Yield of 8 was 2.50 g (91.3%) as a bright yellow solid
with no mp until 200 ◦C. 1H NMR (500 MHz, CDCl3) d 3.70 (br,
12H, MeO), 7.40 (m, 4H, arom-H), 7.50 (m, 4H, arom-H), 8.10
(m, 4H, arom-H), 8.20 (m, 4H, arom-H), 9.00 (m, 4H, arom-H).
19P NMR (377 MHz, CDCl3) d 6.16 (s, 2P, OP(O)(OMe)2). Anal.
Calcd. for C60H32F36Ir2N4O8P2 (mol. wt.: 2067.24): C, 34.86; H,
1.56; N, 2.71. Found: C, 34.90; H, 1.83; N, 2.76%. The structure
was proven by X-ray analysis (Fig. 5).
2-(3,5-Bis(trifluoromethyl)phenyl)-6-[2,4-bis(trifluoromethyl)-
6-(4-trifluoromethylpyridin-2-yl)phenyl]-4-trifluoromethylpyridine
(4). The yield of 4 (from the above cyclometalation of 1) was
0.46 g (3.5%) as white crystals from pentane with mp 65.2 ◦C.
1H NMR (500 MHz, CD2Cl2, TMS) d 7.20 (s, 2H, arom-H), 7.51
(s, 1H, arom-H), 7.90 (m, 2H, arom-H), 8.23 (m, 2H, arom-H),
8.35 (s, 2H, arom-H), 8.63 (s, 1H, arom-H). 19F NMR (377 MHz,
CD2Cl2) d −57.70 (s, 3F, CF3), −63.70 (s, 3F, CF3), −63.78 (s,
6F, CF3), −65.69 (s, 3F, CF3), −65.98 (s, 3F, CF3). Anal. Calcd.
for C28H10F18N2 (mol. wt.: 716.36): C, 46.95; H, 1.41; N, 3.91.
Found: C, 47.03; H, 1,52; N, 4.27%. LC/MS calculated for
C28H10F18N2: 716,06. Found: 716,06. The structure was proven by
X-ray analysis.
X-Ray crystallography
Crystallographic data for compounds 2–4 and 5, 7, 8 are given in
Tables 7 and 8.
Tris[4,6-bis(trifluoromethyl)-2-(4-trifluoromethyl-2-pyridinyl-
jN)phenyl-jC]iridium (5). The yield of 5 (from the above
cyclometalation of 1) was 1.24 g (8.1%) as colorless crystals with
CCDC reference numbers 257681–257686.
For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b514797b
2476 | Dalton Trans., 2006, 2468–2478
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Table 7 Summary of crystal data, data collection, and structural refinement parameters of 2–4
2
3
4
Empirical formula
FW
C34H12Ag3F27N2O6
1381.07
Colourless plate
Triclinic
¯
P1
8.2935(15)
12.797(2)
19.913(3)
82.984(3)
83.660(3)
83.647(3)
2074.6(6)
2
C61H31F36IrN4
1696.1
Gold block
Triclinic
¯
P1
12.9220(13)
15.9360(16)
17.7587(18)
67.8845(17)
72.2741(17)
69.862(2)
3115.9(5)
2
C28H10F18N2
716.38
Colourless block
Triclinic
¯
P1
10.4708(12)
11.8225(14)
12.6383(15)
104.597(2)
105.654(2)
105.656(2)
1358.6(3)
2
Crystal color, form
Crystal system
Space group
˚
a/A
˚
b/A
˚
c/A
◦
a/
b/
◦
◦
c /
3
˚
V/A
Z
q/g cm−3
2.211
1.572
1.808
2.295
1.751
0.191
l/mm−1
F(000)
1324
1652
708
Crystal size/mm
0.42 × 0.42 × 0.05
0.28 × 0.20 × 0.12
0.50 × 0.36 × 0.35
◦
T/ C
−100
−100
−100
Scan mode
x
x
x
Detector
Bruker-CCD
28.28
Bruker-CCD
28.3
Bruker-CCD
28.31
◦
hmax
/
No. obsrvd. refs
No. uniq. refs
Rmerge
13789
9463
0.014
699
18319
13836
0.020
921
9024
6203
0.013
453
No. params
Sb
1.041
1.041
1.048
R indices [I > 2r(I)]a
wR2 = 0.070, R1 = 0.028
wR2 = 0.073, R1 = 0.034
0.621, −0.665
wR2 = 0.094, R1 = 0.038
wR2 = 0.101, R1 = 0.048
2.819, −0.668
wR2 = 0.131, R1 = 0.050
wR2 = 0.146, R1 = 0.065
0.418, −0.456
R indices (all data)a
−3
˚
Max diff peak, hole/e A
)
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
a R1 = ꢁFo| − |Fcꢁ/ |Fo|, wR2 = { [w(Fo − Fc2)2]/ [w(Fo2)2]} (sometimes denoted as Rw2). b GooF = S = { [w(Fo − Fc2)2 ]/(n − p)}
,
2
1/2
2
1/2
where n is the number of reflections, and p is the total number of refined parameters.
Table 8 Summary of crystal data, data collection, and structural refinement parameters of 5, 7 and 8
5
7
8
Empirical formula
FW
C84H30F54Ir2N6
2533.54
Gold prism
Monoclinic
C2
41.862(4)
9.2252(9)
22.004(2)
C39H29F18IrN2O2
1091.84
Gold plate
Orthorhombic
Pbca
20.852(4)
16.227(3)
24.092(4)
90
90
90
8152(3)
C34H30F18IrN2O12P3
1285.71
Yellow plate
Triclinic
¯
P1
10.923(6)
14.898(8)
15.863(8)
64.147(8)
85.432(9)
85.384(9)
2313(2)
2
Crystalcolor, form
Crystal system
Space group
˚
a/A
˚
b/A
˚
c/A
◦
a/
b/
90
◦
99.2780(18)
90
8386.5(14)
4
2.007
3.343
4864
◦
c /
3
˚
V/A
Z
8
q/g cm−3
l/mm−1
F(000)
)
1.779
3.395
4256
1.846
3.122
1256
Crystal size/mm
0.28 × 0.28 × 0.10
0.24 × 0.14 × 0.05
0.14 × 0.06 × 0.01
◦
T/ C
−100
−100
−100
Scan mode
x
x
x
Detector
Bruker-CCD
28.28
Bruker-CCD
28.27
Bruker-CCD
28.39
◦
hmax
/
No. obsrvd. refs
No. uniq. refs
Rmerge
22126
11872
0.029
1316
50219
9831
0.074
565
28327
10851
0.0809
666
No. params
Sb
1.026
1.01
1.021
R indices [I > 2r(I)]a
R indices (all data)a
wR2 = 0.072, R1 = 0.032
wR2 = 0.075, R1 = 0.037
2.596, −0.793
wR2 = 0.092, R1 = 0.041
wR2 = 0.110, R1 = 0.081
2.967, −1.397
wR2 = 0.129, R1 = 0.058
wR2 = 0.161, R1 = 0.118
2.995, −1.587
−3
˚
Max diff peak, hole/e A
)
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
a R1 = ꢁFo| − |Fcꢁ/ |Fo|, wR2 = { [w(Fo − Fc2)2]/ [w(Fo2)2]} (sometimes denoted as Rw2). b GooF = S = { [w(Fo − Fc2)2 ]/(n − p)}
,
2
1/2
2
1/2
where n is the number of reflections, and p is the total number of refined parameters.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 2468–2478 | 2477
©

View Article Online
E. Herdtweck and W. A. Herrmann, Organometallics, 2000, 19, 1692;
(d) A. T. Termaten, M. Schakel, A. W. Ehlers, M. Lutz, A. L. Spek and
K. Lammertsma, Chem. Eur. J., 2003, 9, 3577.
8 (a) D. A. Ortmann, B. Weberndorfer, J. Schoneboom and H. Werner,
Organometallics, 1999, 18, 952; (b) P. Burger and R. G. Bergman, J. Am.
Chem. Soc., 1993, 115, 10462.
Acknowledgements
The authors wish to thank Kurt Adams for funding the X-ray
research, and Anthony Arduengo for valuable suggestions while
reading the manuscript.
9 (a) R. Kubik, S. Bohn, I. Ruppertova and J. Kuthan, Collect. Czech.
Chem. Commun., 1996, 61, 126; (b) R. Pohl, S. Bohn, I. Ruppertova
and J. Kuthan, Collect. Czech. Chem. Commun., 1999, 64, 1274.
10 (a) S. Wimmer and F. L. Wimmer, J. Chem. Soc., Dalton Trans., 1994,
879; (b) P. S. Braterman, J.-I. Song, F. L. Wimmer, S. Wimmer, W. Kaim,
A. Klein and R. D. Peacock, Inorg. Chem., 1992, 31, 5084; (c) P. Castan,
F. Dahan, S. Wimmer and F. L. Wimmer, J. Chem. Soc., Dalton Trans.,
1990, 2971.
11 (a) K. Luender and H. Fuellbier, Z. Chem. (Stuttgart, Ger.), 1990, 30,
370; (b) V. Kalyanaraman, C. N. R. Rao and M. V. George, J. Chem.
Soc. B, 1971, 2406; (c) D. Lavorato, J. K. Terlouw, T. K. Dargel, W.
Koch, G. A. McGibbon and H. Schwarz, J. Am. Chem. Soc., 1996, 118,
11898; (d) D. Lavorato, J. K. Terlouw, G. A. McGibbon, T. K. Dargel,
W. Koch and H. Schwarz, Int. J. Mass Spectrom., 1998, 179/180, 7.
12 (a) S. Vidal, J. Court and J. M. Bonnier, Tetrahedron Lett., 1976, 24,
2023; (b) Y. M. Yutilov and N. N. Smolyar, Russ. J. Org. Chem. (Transl.
Of Zh. Org. Khim.), 1996, 32, 1364; (c) A. K. Sheinkman, V. A. Ivanov,
N. A. Klyuev and G. A. Maltseva, Russ. J. Org. Chem. (Transl. Of
Zh. Org. Khim.), 1973, 9, 2550; (d) L. E. Ramfrez-Arizmendi, J. L.
Heidbrink, L. P. Guler and H. I. Kenttamaa, J. Am. Chem. Soc., 2003,
125, 2272.
References
1 S. Tennant, Philos. Trans. R. Soc. London, 1804, 94, 411.
2 (a) M. Zhao and J. L. Bada, Nature, 1989, 339, 463; (b) S. Bajpai and
G. V. R. Prasad, J. Geol. Soc. (London, UK), 2000, 157, 257; (c) B. F.
Bohor, D. M. Triplehorn, D. J. Nichols and H. T. Millard, Geology,
1987, 15, 896.
3 (a) S. Lamansky, P. Djurovich, D. Murphy, F. Abdel-Razzaq, H. E. Lee,
C. Adachi, P. E. Burrows, S. R. Forrest and M. E. Thompson, J. Am.
Chem. Soc., 2001, 123, 4304; (b) M. A. Baldo, D. F. O’Brien, Y. You,
A. Shoustikov, S. Sibley, M. E. Thompson and S. R. Forrest, Nature,
1998, 395, 151.
4 V. V. Grushin, N. Herron, D. D. LeCloux, W. J. Marshall, V. A. Petrov
and Y. Wang, Chem. Commun., 2001, 1494.
5 (a) Q.-M. Wang and T. C. W. Mak, Angew. Chem., Int. Ed., 2001,
40, 1130; (b) P. Reiss, F. Weigend, R. Ahlrichs and D. Fenske, Angew.
Chem., Int. Ed., 2000, 39, 3925; (c) Q.-M. Wang and T. C. W. Mak,
Angew. Chem., Int. Ed., 2002, 41, 4135; (d) P. Wei, T. C. W. Mak and
D. A. Atwood, Inorg. Chem., 1998, 37, 2605; (e) J. Feldman and J. C.
Calabrese, Inorg. Chem., 1994, 33, 5955.
6 A. B. Tamayo, B. D. Alleyne, P. I. Djurovich, S. Lamansky, I. Tsyba,
N. N. Ho, R. Bau and M. E. Thompson, J. Am. Chem. Soc., 2003, 125,
7377.
13 (a) S. Sabir, A. U. Malik and K. Iftikhar, Indian J. Chem., Sect. A:
Inorg., Bio-inorg., Phys., 1998, 37, 834; (b) P. S. Gomm, A. E. Underhill
and R. W. A. Oliver, J. Inorg. Nucl. Chem., 1974, 36, 3699.
14 (a) A. H. Cowley, N. C. Norman and M. Pakulski, Inorg. Synth., 1990,
27, 235; (b) A. S. Ionkin, W. J. Marshall and Y. Wang, Organometallics,
2005, 24, 619; (c) A. S. Ionkin and W. J. Marshall, Organometallics,
2004, 23, 6031.
7 (a) A. R. Chianese, X. Li, M. C. Janzen, J. W. Faller and R. H. Crabtree,
Organometallics, 2003, 22, 1663; (b) F. Hanasaka, K.-i. Fujita and R.
Yamaguchi, Organometallics, 2004, 23, 1490; (c) M. Prinz, M. Grosche,
2478 | Dalton Trans., 2006, 2468–2478
This journal is
The Royal Society of Chemistry 2006
©