Synthesis of a chloro protected iridium nitrido complex

Article abstract of DOI:10.1039/c1dt10886g

Intramolecular activation processes of vulnerable ligand C-H bonds frequently limit the thermal stability and accessibility of late transition metal complexes with terminal metal nitrido units. In this study chloro substitution of the 2,6-ketimine N-aryl substituents (2,6-C₆H ₃R₂, R = Cl) of the pyridine, diimine ligand is probed to increase the stability of square-planar iridium nitrido compounds. The thermal stability of iridium azido precursor and nitrido compounds was studied by a combination of thermoanalytical methods (DTG/MS and DSC) and were compared to the results for the related complexes with 2,6-dialkyl substituted N-aryl groups (R = Me, iPr). The investigations were complemented by DFT calculations, which allowed us to unravel details of the thermal decomposition pathways and provided mechanistic insights of further conversion steps and fluctional processes. The DTG/MS and DSC measurements revealed two different types of thermolysis pathways for the azido compounds. For the complexes with R = Cl and iPr substituents, two well-separated exothermic processes were observed. The first moderately exothermic loss of N₂ is followed by a second, strongly exothermic transformation. This contrasts the experimental results for the compound with 2,6-dimethyl substituents (R = Me), where both steps proceed concurrently in the same temperature range. The separation of the two thermal steps in the 2,6-dichloro substituted derivative allowed us to develop a protocol for the isolation of the highly insoluble nitrido complex, which was characterized by UV/vis, IR-spectroscopy and elemental analysis. Its constitution was further confirmed by reaction with silanes, which gave the corresponding silyl amido complexes.

Full text of DOI:10.1039/c1dt10886g

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2011, 40, 9512
www.rsc.org/dalton
PAPER
Synthesis of a chloro protected iridium nitrido complex†
Daniel Sieh, Julia Scho¨ffel and Peter Burger*
Received 10th May 2011, Accepted 15th July 2011
DOI: 10.1039/c1dt10886g
Intramolecular activation processes of vulnerable ligand C–H bonds frequently limit the thermal
stability and accessibility of late transition metal complexes with terminal metal nitrido units. In this
study chloro substitution of the 2,6-ketimine N-aryl substituents (2,6-C6H3R2, R = Cl) of the pyridine,
diimine ligand is probed to increase the stability of square-planar iridium nitrido compounds. The
thermal stability of iridium azido precursor and nitrido compounds was studied by a combination of
thermoanalytical methods (DTG/MS and DSC) and were compared to the results for the related
complexes with 2,6-dialkyl substituted N-aryl groups (R = Me, iPr). The investigations were
complemented by DFT calculations, which allowed us to unravel details of the thermal decomposition
pathways and provided mechanistic insights of further conversion steps and fluctional processes. The
DTG/MS and DSC measurements revealed two different types of thermolysis pathways for the azido
compounds. For the complexes with R = Cl and iPr substituents, two well-separated exothermic
processes were observed. The first moderately exothermic loss of N2 is followed by a second, strongly
exothermic transformation. This contrasts the experimental results for the compound with 2,6-dimethyl
substituents (R = Me), where both steps proceed concurrently in the same temperature range. The
separation of the two thermal steps in the 2,6-dichloro substituted derivative allowed us to develop a
protocol for the isolation of the highly insoluble nitrido complex, which was characterized by UV/vis,
IR-spectroscopy and elemental analysis. Its constitution was further confirmed by reaction with silanes,
which gave the corresponding silyl amido complexes.
Complexes of this type show interesting reactivity patterns that
Introduction
were explored only in the last few years. The dimerization to
bridged dinitrogen complexes according to 2 LnM N → LnM–
N2–MLn is a well-established reaction pathway,5 which can be
prevented6 by the use of sterically demanding ligands shielding
the terminal metal nitrido unit. Further prominent examples
include the intermolecular formation of ammonia by H atom
abstraction,7 the activation of main group hydrides8 and molecular
hydrogen.4 Frequently, intramolecular deactivation steps of highly
reactive putative or isolated metal imido9 and nitrido10 units by
insertion into benzylic and arylic C–H bonds is observed. For
the aforementioned isolated iridium nitrido complex, we recently
reported the thermal intramolecular activation of a benzylic C–H
bond leading to the corresponding “tuck-in” amido complex.11
Warren et al. showed that these undesirable intramolecular steps
can be prevented by chloro substitution of the vulnerable C–H
bonds.9 Taking up this concept, we herein report the synthesis
and thermolysis of the 2,6-chloro aryl protected pyridine, diimine
iridium azido complex 4-Cl. Its thermal stability will be compared
to the 2,6-methyl substituted analogue 4-Me, which has a similar
sterical demand of the ligand.
In low valent late transition metal complexes with a termi-
nal nitrido unit, LnM N, 4 electron two orbital destabilizing
interactions between the strong p-donor nitrido ligand and
occupied metal orbitals are anticipated to lead to highly reactive
intermediates.1 Caulton et al. reported the structural character-
ization of a d4-configured Ru nitrido complex in 2005,2 which
was succeeded by the isolation of iron complexes with the
same electron configuration.3 We have recently published the
synthesis, X-ray crystallographic characterization and reactivity
of a square-planar pyridine, diimine iridium nitrido complex.
Based on experimental XAS and XPS measurements and detailed
theoretical investigations including multi-reference calculations
we rationalized a d6-electron configuration for this compound with
a terminal nitrido unit.4
Institut fu¨r Anorganische und Angewandte Chemie Universita¨t
Hamburg, Martin-Luther-King-Platz 6, 20146 Hamburg, Germany.
E-mail: burger@chemie.uni-hamburg.de; Fax: (+49) 40 42838 6097;
Tel: (+49) 40 42838 3662
† Electronic supplementary information (ESI) available: Further details of
the thermoanalytical experiments, selected UV/vis, IR and NMR spectra
of isolated and indistinct complexes and the DFT optimized geometries
including total energies. CCDC reference numbers 825841–825845. For
ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/c1dt10886g
Results and discussion
Syntheses
The synthesis of the chloro substituted ligand 1 was initially
attempted by a previously reported route of Qian et al.12 Following
9512 | Dalton Trans., 2011, 40, 9512–9524
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
their procedure, however, we had significant problems to isolate
the product from the resulting brown oil. Applying a small
modification, i.e. using benzene instead of toluene as solvent,
allowed us to isolate ligand 1 in 29% yield after crystallization
from acetone. This also provided single crystals suitable for X-
ray crystallography, which was performed to analyze changes
within the ligand upon coordination. The molecular structure of
ligand 1 is shown in Fig. 1, selected bond lengths are given in the
figure caption. Further details of the crystal data and structure
refinement for compound 1 and the other crystal structures
presented in this work are summarized in Table 4 and in the X-
ray crystallography section. The solid state structure of ligand 1
shows E-configurations for both imine double bonds with regard
to the aryl and pyridine groups. This is in agreement with previous
findings for pyridine, diimine ligands.12,13 The pyridine, diimine
N-donors belong to the well-established class of non-innocent
ligands. They undergo significant changes of the bond distances
of the exocyclic pyridine C–C and imine C–N double bonds in their
transition from the innocent to non-innocent state. The averaged
˚
bond lengths of 1.28 A for the C–N double bonds of the imine
˚
units (N1–C2, N3–C8) and the averaged bond distances of 1.50 A
for the exocyclic C–C bonds of the pyridine ring are therefore
relevant benchmarks for the innocent behaviour of this ligand (cf.
below).
Using a standard protocol,14–16 the rhodium and iridium chlo-
rido compounds 2-Rh and 2-Ir were obtained in analytically
pure form and in moderate to good yields (Scheme 1). The
1H and 13C NMR data of the chlorido compounds 2-Rh,Ir
were consistent with the anticipated pseudo square-planar C2v-
symmetrical geometries for d8-systems, which was unambiguously
confirmed by the sum of angles of the Rh center of 359.99◦
observed in the X-ray crystal structure of complex 2-Rh. The
molecular structure of 2-Rh is displayed in Fig. 2, details of the
data collection and structural solution are summarized in Table
4, selected geometrical parameters in Table 1. The effects of the
oxidation state of the metal center and the donor properties of the
additional ligands on the geometrical parameters of the pyridine,
diimine donor in 2-Rh will be discussed below in context with the
molecular structures obtained in this study.
Fig. 1 Ortep representation of ligand 1 with anisotropic displacement
parameters shown at 50% probability level. Hydrogen atoms are omitted
˚
for clarity. Selected bond lengths [A] with esd’s in parentheses: N1–C2
1.281(2), N1–C10 1.415(2), N3–C8 1.277(2), N3–C16 1.413(2), N2–C3
1.342(2), N2–C7 1.348(2), Cl1–C11 1.739(1), Cl2–C15 1.742(1), Cl3–C17
1.742(2), Cl4–C21 1.743(2), C1–C2 1.505(2), C2–C3 1.500(2), C3–C4
1.396(2), C4–C5 1.389(2), C5–C6 1.388(2), C6–C7 1.394(2), C7–C8
1.497(2), C8–C9 1.500(2), C10–C11 1.404(2), C10–C15 1.395(2), C16–C17
1.400(2), C16–C21 1.395(2).
It should be noted that compared to analogous complexes
with 2,6-dialkyl substituents, the dichloro derivatives 2–5 show
a substantially lower solubility in organic solvents, e.g. toluene,
Et2O or THF. The chlorido complexes 2-Rh,Ir, for example, are
Scheme 1 Synthesis and reactivity of the rhodium and iridium complexes 2–5.
The Royal Society of Chemistry 2011 Dalton Trans., 2011, 40, 9512–9524 | 9513
This journal is
©

View Article Online
tion was previously also observed for pyridine, diimine Rh(I)
complexes.14 Upon warming the solution to 55 ◦C the reaction
goes to completion to give 5-Rh and 5-Ir in 88% and 95% isolated
1
yields. In the aromatic range of the H NMR spectrum of 5-
Rh and 5-Ir, two well separated regions for the pyridine and
phenyl rings centered at ca. 8 ppm and 7.4 were detected. For
the three protons of the 2,6-dichloro phenyl groups three sets of
signals in a 1 : 1 : 1 integration ratio were observed. This suggested
a diastereotopic situation for the protons in the 3 and 5 positions,
which is caused by the absence of a mirror plane or C2-axis
located within the framework of the pyridine, diimine N-donors
and the iridium center. On the other hand, the observation of a
singlet for the (enantiotopic) ketimine methyl resonances implied
a mirror plane perpendicular to the ligand framework. Based
on these results Cs-symmetry was anticipated for 5-Rh and 5-
Ir, which required placement of the chloromethyl group (CH2Cl)
in the apical position of the octahedron with a cis-configuration
of the remaining two chlorido ligands. This assignment was
unambiguously confirmed by the results of the X-ray crystal
structure analyses of complexes 5-Rh and 5-Ir, which evidenced
a distorted octahedral coordination around the rhodium and
iridium metal centers. The isostructural molecular structures of
the complexes 5-Rh and 5-Ir are presented in Fig. 3, details of the
data collection, structure solution and refinement are tabulated in
Table 4, selected geometrical parameters in Table 2. The bonding
parameters are essentially identical for both the iridium and the
rhodium system. Since the quality of the crystal structure 5-Rh is
significantly higher, the data will be discussed only for the latter.
By comparison with the bonding parameters in the free ligand 1,
only negligible differences of the C–N and C–C distances of the
diimine group and the exocyclic C–C bond of the pyridine group
clearly speak in favour of an innocent pyridine, diimine ligand
Fig. 2 Ortep plot (50% probability level) for the molecular structure of
2-Rh. Hydrogen atoms and the co-crystallized THF molecule are omitted
for clarity.
◦
a
˚
Table 1 Selected bond lengths (A) and angles ( ) of 2-Rh
Rh1–N1
Rh1–N3
N1–C2
2.0168(14)
2.0323(14)
1.315(2)
Rh1–N2
Rh1–Cl1
N3–C8
1.8915(14)
2.3211(4)
1.316(2)
C2–C3
1.457(2)
158.34(6)
79.03(6)
C7–C8
1.457(2)
178.77(5)
99.74(4)
N1–Rh1–N3
N1–Rh1–N2
N2–Rh1–N3
N2–Rh1–Cl1
N1–Rh1–Cl1
N3–Rh1–Cl1
79.32(6)
101.90(4)
a Esd’s in parentheses.
only slightly soluble in THF and display reasonable solubility only
in dichloromethane.
In the course of the NMR spectroscopic characterization we
noted that the rhodium and iridium chlorido compounds 2-
Rh,Ir react at room temperature within several hours with the
dichloromethane solvent to give the corresponding octahedral
complex 5-Rh,Ir under oxidative addition. This type of reac-
Fig. 3 Front and side view of the Ortep representation of the octahedral complexes 5-Rh and 5-Ir with anisotropic displacement parameters shown at
50% probability level. The hydrogen atoms and a co-crystallized CH2Cl2 molecule are omitted for clarity.
◦
a
˚
Table 2 Selected bond lengths (A) and angles ( ) of 5-Rh and 5-Ir
M
Rh1
M
Ir1
M
Rh1
M
Ir1
M–N1
M–N3
M–Cl2
N1–C2
2.0649(14)
2.0782(15)
2.4892(4)
1.302(2)
2.067(10)
2.068(10)
2.465(4)
1.278(14)
1.508(16)
1.825(12)
158.9(4)
178.0(4)
79.3(4)
M–N2
M–Cl1
M–C10
N3–C8
1.9221(15)
2.3559(4)
2.0424(18)
1.307(2)
1.856(10)
2.357(4)
1.997(12)
1.287(15)
1.510(17)
3.070(10)
177.3(3)
117.5(7)
100.4(3)
100.6(3)
C2–C3
1.482(2)
C7–C8
1.480(3)
C10–Cl3
N1–M–N3
Cl2–M–C10
N1–M–N2
N2–M–N3
1.8141(18)
159.21(6)
177.45(5)
79.77(6)
N2 ◊ ◊ ◊ Cl3
N2–M–Cl1
M–C10–Cl3
N1–M–Cl1
N3–M–Cl1
3.0548(15)
177.91(4)
114.73(9)
100.39(4)
100.11(4)
79.88(6)
79.8(4)
a Esd’s in parentheses.
9514 | Dalton Trans., 2011, 40, 9512–9524
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
in complexes 5-Rh and 5-Ir (cf. below). The rhodium chlorido
˚
bond length of 2.3559(4) A (Rh1–Cl1) within the plane of the
terdentate N-donor is comparable to the corresponding distance
in the rhodium(I) chlorido complex 2-Rh (Rh1–Cl1, 2.3211(4)). It
˚
is notably shorter than the Rh–Cl bond (Rh1–Cl2, 2.4892(4) A)
to the apical chlorido ligand, which is attributed to the larger
trans influence of the chloromethyl group. A similar alternation of
the bond lengths in related octahedral pyridine, diimine complexes
was previously reported by us and others for methyl bis-triflato Rh
complexes.15,16 Particularly noteworthy is the distortion from the
ideal octahedral geometry, which becomes immediately apparent
from inspection of the side view of complex 5-Rh shown in
Fig. 3 and which is also present in the iridium congener 5-Ir.
Similar deviations are found in related octahedral rhodium(III)
complexes previously studied by us16 and might be attributed
to steric repulsion of the apical groups and the ligands in the
equatorial plane.
Following a previously established route by our group,4,16 the
desired methoxido iridium precursor compound 3 was obtained
from 2-Ir by metathesis with sodium methoxide and was then
further converted to the azido complex 4-Cl by reaction with
trimethylsilyl azide (Scheme 1). The constitutions of complexes
1
3 and 4-Cl were confirmed by H, 13C, and IR spectroscopy, as
well as elemental analysis for complex 3. The IR spectrum of the
azido complex 4-Cl shows a strong absorption at 2035 cm-1, which
could be unambiguously assigned to the asymmetrical stretching
vibration of the azido group by comparison of the IR spectrum of
4-Cl with its single labeled 15N-isotopologue, 4-Cl-15N, Ir-15N N2
(Ir–N
N
15N).
Fig. 4 Above: DSC traces for complexes 4-Me (gray) and 4-Cl (black);
below: DTA and TG/MS curves for 4-Me (black: TG, gray: DTA, dotted:
mass traces).
Thermoanalytical studies
The thermolyses of the 2,6-dichloro and 2,6-dimethyl aryl sub-
stituted azido complexes 4-Cl and 4-Me were studied by dif-
ferential scanning calorimetry (DSC) and combined differential
thermogravimetry/mass spectrometry (DTG-MS). The recorded
DSC curves for complexes 4-Cl and 4-Me are presented in Fig. 4.
The inspection of the DSC traces shown in Fig. 4 self-evidently
reveals the disparities of the thermal properties of the chloro and
methyl aryl substituted derivatives 4-Me and 4-Cl. While there
is only one exothermic process notable for complex 4-Me, there
are two clearly separated exothermic steps in the DSC trace of
the chloro aryl compound 4-Cl. The DSC experiment of the
2,6-dimethyl aryl substituted azido complex 4-Me revealed an
exothermic peak at 112 ◦C with an enthalpy of -35.2 kcal mol-1.
The complementary DTG/MS experiment for 4-Me evidenced
a mass loss of 6.89% (calculated for N2 loss: 4.64%) at this
+
temperature with concurrent signals at m/z = 28 and 14 for N2
2+
and N2 in the mass spectrometric trace of the volatiles. The
observed experimental enthalpy of -35.2 kcal mol-1 for 4-Me
is considerably higher than the previously determined value of
-6 kcal mol-1 for the formation of the (isolated) iridium nitrido
complex from its azido precursor in the 2,6-diisopropyl phenyl
substituted derivative.4 For the conversion of the azido compound
to the putative nitrido complex ([Ir]–N3 → [Ir]∫N + N2; 4-Me →
6-Me) values of -0.4, respectively -5.3 kcal mol-1 were calculated
by DFT methods with the BP-86 and B3LYP functionals.
Scheme 2 Thermolysis of the azido complex 4-Me and a potential
product.
For the C–H activation step of one of the benzylic methyl
groups (Scheme 2) by the putative nitrido complex 6-Me a strongly
energetically
favoured
situation
(-29.9(BP-86)
and
-29.3(B3LYP) kcal mol-1) was estimated by DFT calculations. It
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 9512–9524 | 9515
©

View Article Online
should be noted that an experimental value of DH = -21 kcal mol-1
was previously established for C–H activation of the benzylic
C–H bond in the corresponding 2,6-diisopropyl substituted
derivative in DSC measurements.11 These combined experimental
and theoretical findings indicate that the extrusion of dinitrogen
and the tentative subsequent C–H activation step (“tuck-in”-
formation) are not kinetically separated for the methyl substituted
azido complex 4-Me (Scheme 2). Attempts to fully characterize
the final thermolysis product(s) were so far unsuccessful. The 1H
NMR spectrum of a sample in THF-d8 recorded at RT revealed
a large number of peaks, which could not be unambiguously
assigned, suggesting the presence of several products rather
◦
1
than a single product. The H NMR spectra recorded at -80 C
of a sample prepared at low temperatures by condensation of
THF-d8 to the solid thermolyzed residue of 4-Me at -196 ◦C
showed prominent broad down-field shifted peaks at d = 25.5 and
16.8 ppm besides further signals in the range from 0–10 ppm (see
ESI†). Most notably, we did not detect peaks in the anticipated
high-field range of -10 to -40 ppm, indicative of hydride ligands.
The broad resonances at d = 25.5 and 16.8 ppm disappeared upon
allowing the sample to warm up to RT, giving the aforementioned
1H NMR spectrum of a sample prepared at RT. Although, we can
only speculate on the identity of the thermolyzed material, based
on the resonances d = 25.5 and 16.8 ppm, we suspect that a para-
magnetic intermediate is among it, which converts further at RT.
The inspection of the DSC curve for the chloro substituted azido
complex 4-Cl revealed an entirely different picture. In contrast to
the results for complex 4-Me, two well separated signals with
maxima at 114 and 196 ◦C were detected in the temperature
range from 70–270 ◦C. Based on the recorded mass loss of 4.63%
Fig. 5 DTA and TG/MS curves for 4-Cl (black: TG, gray: DTA, dotted:
mass traces).
196 ◦C with an integrated enthalpy of -22.5 kcal mol-1. DTG-MS
analysis for this step revealed a mass loss of 2.8%, but elemental
analysis of the product obtained by batch thermolysis at 150 ◦C for
19 h showed no significant deviation from the product obtained
from thermolysis at 90 ◦C, suggesting decomposition setting in
above 200 ◦C (cf. ref. 11).
The product obtained in the second step of the DSC mea-
surement by heating 4-Cl over 150 ◦C proved to be rather
insoluble in all tested organic solvents thus thwarting further
NMR characterization. Since the IR and UV/vis spectra recorded
as KBr pellets were not instructive, we can only speculate on
the product(s). DFT calculations were performed for the two
potential products 8a and 8b obtained by activation of the aryl C–
Cl bond (Scheme 3), which are both energetically downhill from
6-Cl. As anticipated, the formation of the less likely amido N–Cl
compound 8a is significantly less favoured than the corresponding
chlorido, imido complex 8b (6-Cl → 8a: -18 kcal mol-1; 6-Cl →
8b: -32 kcal mol-1). The transition state for C–Cl activation could
be located in the DFT calculations and is presented in Fig. 6.
The corresponding barrier of 28 kcal mol-1 for this process is
consistent with a set-in temperature of 150 ◦C observed in the
DSC measurements. Based on the analysis of the frontier orbitals
+
(calculated: 4.09%) and a concomitant signature of 28 m/z (N2 )
of the volatiles, initiated at 70 ◦C in the DTG-MS analysis (Fig. 5),
the first step is assigned to the release of dinitrogen from the azido
complex 4-Cl giving the corresponding nitrido compound 6-Cl.
The signal integrates to an enthalpy of -11.5 kcal mol-1, which is
in the range derived for the isolated iridium nitrido complex with
2,6-diisopropyl phenyl groups.4 The signal for dinitrogen release
passes on to the second exothermic peak, which has a maximum at
Scheme 3 Thermolysis of the azido complex 4-Cl.
This journal is
9516 | Dalton Trans., 2011, 40, 9512–9524
The Royal Society of Chemistry 2011
©

View Article Online
Fig. 6 Transition state for intramolecular C–Cl activation with selected
◦
˚
distances and angles in [A] and [ ].
and considering the rather electron-deficient nature of the 2,6-
dichloro substituted aryl group, it is deemed that this step proceeds
by nucleophilic attack of the nitrido group at the aromatic carbon
atom. At first glance, inspection of Fig. 6 might suggest formation
of a Meisenheimer intermediate with close resemblance to the
transition state structure. However, the IRC calculations provided
strong arguments against this view and showed that the reaction
mode proceeds in fact by release of a chloride ion to give an
intermediate 6-I in the product direction (Scheme 3).
Fig. 8 DSC measurement of the azido complex 4-Cl in cyclic mode
between 50–90 ◦C and between 0–350 ◦C after complete N2 extrusion.
As shown in Fig. 8, thermolysis of complex 4-Cl at 90 ◦C yields
only the product of the N2 release, suppressing the second thermal
step.
Characterization of the thermolysis product 6-Cl
The Fukui function for nucleophilic attack f+ of 6-I presented
in Fig. 7 revealed two preferential sites for addition of the chloride
ion in a consecutive step. The preferential sites for nucleophilic
attack are a) the nitrogen atom of the former nitrido unit and b) the
iridium center, which would lead to either 8a and 8b upon addition
of the chloride ion. As noted above, the DFT calculations showed
that the iridium(III) complex 8b is 14 kcal mol-1 more stable than
the Ir(I) compound 8a.
The compound obtained by the reaction of the azido complex 4-
Cl at 90 ◦C was sparingly soluble in all tested organic solvents,
thwarting NMR measurements in solution.17 The elemental
analysis of the thermolyzed product was consistent with N2 loss,
which was further supported through the absence of an azido
stretching vibration in the range of 2030 cm-1 in the IR spectrum.
Furthermore, a new IR band was detected at 955 cm-1, which red-
shifted by ca. 30 cm-1 for the thermolyzed 15N labeled isotopologue
of 4-Cl-15N (Fig. 9). Together with the good agreement of the
n(Ir N) stretching frequency of the aforementioned isolated
terminal Ir nitrido complex with 2,6-diisopropyl phenyl groups,
6-iPr located at 958 cm-1,4 we assigned this band to a n(Ir N)
vibration.
Fig. 7 Fukui f+ function for the nucleophilic attack in 6-I. The arrows
indicate sites for preferential attack.
In order to optimize the conditions for the preparation of the
tentative nitrido complex 6-Cl, DSC measurements of the azido
complex 4-Cl were performed in cyclic mode between 50 and 90 ◦C
(Fig. 8).
Fig. 9 Section of the IR spectra (KBr) of the nitrido complexes 6-Cl
(black) and 6-Cl-15N (gray).
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 9512–9524 | 9517
©

View Article Online
The solid state UV/vis spectrum measured as KBr pellet is also
comparable to the structural characterized Ir nitrido complex 6-
iPr, exhibiting a strong maximum at 593 nm in the visible range
with two further broad maxima at 766 and 830 nm in the near
infrared. Based on this combined analytical data, assignment of
the thermolyzed material to the terminal nitrido complex 6-Cl
seemed justified.
Further, compelling evidence was provided by reaction of 6-Cl
with silanes. Although the thermolyzed material 6-Cl is almost
insoluble in organic solvents, addition of a THF or toluene
solution of either triethyl- or triphenylsilane to a solid sample of
6-Cl, resulted in a quick dissolution of most of the solid material
to give brown/green colored solutions (Scheme 4).
Fig. 10 Ortep representation of complex 10 with anisotropic displace-
ment parameters shown at the 50% probability level. Hydrogen atoms and
a co-crystallized Et2O molecule are omitted for clarity.
◦
a
˚
Table 3 Selected bond lengths (A) and angles ( ) of 10
Ir1–N1
Ir1–N3
N4–H4n
N1–C2
C2–C3
N1–Ir1–N3
N1–Ir1–N2
N2–Ir1–N3
Ir1–N4–Si1
2.042(6)
2.006(6)
0.95(12)
1.350(9)
1.416(11)
157.2(2)
78.7(2)
Ir1–N2
Ir1–N4
N4–Si1
N3–C8
1.911(6)
1.998(5)
1.694(6)
1.336(9)
1.431(10)
173.4(3)
107.9(2)
94.7(2)
Scheme 4 Reaction of the nitrido complex 6-Cl with silanes.
C7–C8
N2–Ir1–N4
N1–Ir1–N4
N3–Ir1–N4
1
Inspection of the H NMR spectra of the products 9 and 10
78.8(2)
144.5(4)
obtained from reaction of 6-Cl with triethyl- or triphenylsilane
revealed the formation of C2v-symmetrical complexes, indicated by
singlets for the ketimine methyl groups and doublets and triplets
in 2 : 1 integration ratio for the meta and para pyridine protons.
Mostly diagnostical was the observation of broad singlets at d =
8.06 (Et3SiH; C6D6) and 7.22 (Ph3SiH; THF-d8) ppm, which were
assigned to amido protons based on 1H-15N correlated NMR
spectroscopy (HSQC). Compared to the estimated yields of up
to 50% and up to 90% for 9 and 10 derived from NMR tube
reactions of 6-Cl with the silanes by integration of appropriate 1H
NMR signals against the resonance of ferrocene used as internal
standard (cf. ESI†), the isolated yields are significantly diminished.
In fact only the triphenyl silylamido complex 10 could be obtained
in 8% yield as a reasonably pure product, which was further
characterized by 13C NMR spectroscopy. For complex 10, we
also obtained single crystals suitable for X-ray structure analysis,
which allowed to establish the anticipated amido structure. The
molecular structure of complex 10 is given in Fig. 10, selected
bond lengths and angles are given in Table 3.
a Esd’s in parentheses.
structural feature is the bonding situation of the amido nitrogen
atom N4. While the Ir1–N4 bond length of 1.998(5) A is in the
expected range, it displays a short N–Si bond of 1.694(6) A. It
˚
˚
is also bonded to hydrogen atom H4n, which was located in the
difference Fourier map and was isotropically refined. The sum
angles around N4 of 357.5◦ and the position of atom Si1 within
the plane of the terdentate ligand framework and the iridium
center suggests sp2-hybridzation of the amido nitrogen atom N4.
Mostly noticeable are the large Ir1–N4–Si1 angle of 144.5◦ and the
asymmetrical bonding angles N4–Ir1–N1 (107.9(2)◦) and N4–Ir1–
N3 (94.7(2)◦) between the ketimine (N1,N3) and amido nitrogen
atoms. The larger value of the latter two angles is found on the side
of the bulky triphenylsilane group suggesting that steric repulsion
is responsible for widening up both the Ir1–N4–Si1 and N4–
Ir1–N1 angles. This view is supported by DFT calculations of
the hydrogen substituted model complex shown◦below (Ar = 2,6-
The sum of angles of 360.1◦ around the iridium center revealed a
pseudo square-planar geometry. In comparison with the rhodium
chlorido complex 2-Rh the ketimine C–N bonds are slightly longer
˚
C6H3Cl2), bond lengths and angles in (A) and ( )) and the DFT
˚
˚
(N1–C2: 1.350(9) A, N3–C8: 1.336(9) A), while the exocylic C–C
bonds to the pyridine ring are shorter (C2–C3: 1.416(11), C7–
C8: 1.431(10)). This is indicative of more electron transfer to the
pyridine, diimine ligand system, which is consistent with the strong
p-donor propensity of the amido group. The most prominent
optimized geometry of the full structure, which revealed a smaller
Ir–Namido–Si angle of 130.3◦.
Since the idealized molecular structure observed in the solid
state corresponds to Cs-symmetry, the C2v-symmetrical signature
in the 1H NMR spectrum of 10 at RT suggests a fast site
9518 | Dalton Trans., 2011, 40, 9512–9524
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
discussed by us for the corresponding methoxido complexes, for
which a slightly smaller barrier of 12 kcal mol-1 was calculated.16 It
deserves a special mention that IRC-calculations for this process
show that it requires only little movement of the triphenylsilyl
group while the hydrogen atom H4 is essentially responsible for
the left-right site exchange.
Experimental
General
All manipulations of the complexes were performed under stan-
dard Schlenk techniques or in a dinitrogen filled glovebox. Pen-
tane, benzene, toluene, diethylether (Et2O) and tetrahydrofuran
(THF) were distilled over sodium benzophenone ketyl prior to
use. CH2Cl2 was distilled over P2O5 followed by CaH2, MeOH
was distilled over Mg prior to use. Trimethylsilyl azide was
purchased from Merck or Aldrich and was vacuum transferred
prior to use. Isotopically labelled NaN3 (terminal 15N, 98%) for
the preparation of Me3Si15NN2/N215N was used as purchased
from Cambridge Isotope Laboratories. All other reagents were
purchased from commercial sources and were used as received.
The dimeric Ir18 and Rh19 precursors [M(C2H4)2Cl]2 and the
methoxido complex [Ir(Me4N3)OMe]16 were prepared as described
in literature. 1H, 13C, 15N and 29Si NMR spectra were recorded on
Bruker Avance 400, Bruker DRX 500 and Varian Gemini 2000 BB
NMR spectrometers. The 1H and 13C NMR spectra are referenced
to the residual resonances of the solvent. The 15N NMR shifts
are referenced to NH3, the internal standard of the instrument
(d(NH3)(l) = -380 ppm vs. MeNO2(l)). The 29Si NMR shifts are
referenced to Me4Si. The assignments of the 13C NMR spectra
were carried out by combined analyses of DEPT and 13C–1H
correlated spectra (HSQC and HMBC). The UV/vis spectra were
recorded on a Cary 50 spectrometer in 1 cm quartz cuvettes or as
KBr pellets. Elemental analyses were measured by: Zentrale Ele-
mentaranalyse, Universita¨t Hamburg. IR spectra were measured
on a Bruker Vertex 70 IR spectrometer. DSC measurements were
performed on a Netzsch DSC 204 F1 with a heating rate of 10
K min-1 in broached aluminium pans in a nitrogen atmosphere.
The instrument was calibrated against the melting enthalpy of
indium. DTA/TG-MS measurements were recorded on a Netzsch
STA/TG 409C/CD coupled with a Balzer MID quadrupole mass
exchange of the SiPh3 group and the amido hydrogen atom on
the NMR time scale. When a sample of complex 10 was cooled
◦
1
to -100 C, significant changes of the H NMR spectrum were
indeed noticed. For the resonances of the meta hydrogen atoms
of the pyridine ring, the singlet for the ketimine methyl groups
as well as all aromatic resonances a significant broadening was
now observed, suggesting loss of the mirror plane perpendicular
to the N3 ligand framework. This is consistent with the Cs-
symmetrical structure observed in the solid state. Unfortunately,
with coalescence expected slightly below -100 ◦C, we were not able
to perform a lineshape NMR analysis.
We analyzed this process by DFT methods and located the
transition state presented in Fig. 11. The transition state also
displays idealized Cs-symmetry, the mirror plane is now however
perpendicular to the plane spanned by the pyridine, diimine donor
plane and contains the hydrogen and Si atom of the amido ligand.
The calculated barrier amounts to 15 kcal mol-1, which is in
1
the range anticipated for fluctional processes detectable by H
NMR spectroscopy. The barrier can be rationalized by loss of a
stabilizing push pull interaction (4e- 3 orbital) in the transition
state between (i) the lone pair (pz) of the sp2-hybridized amido
p-donor atom N4, (ii) the occupied dxz orbital of the iridium
center and (iii) an acceptor orbital located on the pyridine part
of the terdentate ligand. This type of stabilization was previously
◦
˚
Fig. 11 Schematic, side and front view of the transition state for the left-right site exchange. Selected distances [A] and angles [ ]: Ir–N4: 2.06, Ir–N2:
1.92., N4–Si1: 1.71, N2–Ir–N4: 170.3, N1–Ir–N4: 101.8, N3–Ir–N4: 100.2, Ir–N4–H4: 105.4, Ir–N4–Si1: 142.9, H4–N4–Si1:111.7.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 9512–9524 | 9519
©

View Article Online
spectrometer or a Netzsch STA/TG 449F3 coupled with a Netzsch
QMS 403 C Ae¨olos quadrupole mass spectrometer with a heating
rate of 10 K min-1 in an argon atmosphere.
data collection and processing are given in Table 4 and in the
supplemental CIF files†.
Syntheses
X-ray crystallography
Synthesis of the ligand (Cl4N3) (1). A 250 ml round bottom
flask, equipped with a dean stark trap and a reflux condenser
was charged with 6.54 g (40.1 mmol) 2,6-diacetylpyridine and
15.0 g (92.6 mmol) 2,6-dichloroaniline. 175 ml of dry benzene and
a spatula point (~20 mg) of p-toluenesulfonic acid monohydrate
were added and the atmosphere was changed to nitrogen by three
evacuation/refill cycles. The reaction solution was then refluxed
for 17 h under nitrogen, whereas the colour of the solution changed
to brown-yellow. The solvent was removed under reduced pressure
and the resulting brown oil was dissolved in 400 ml of Et2O,
which was then extracted with three portions of 100 ml distilled
H2O. The Et2O phase was dried over sodium sulphate, filtrated
and removed under reduced pressure. The residue was dried in
vacuum. Addition of 20 ml of methanol resulted in a beige solid,
which was collected by filtration, washed four times with 10 ml
methanol and dried in vacuum. Recrystallization from refluxing
acetone afforded 3.4 g of pale yellow crystals (suitable for single
crystal X-ray diffraction), which were washed with two portions
of 10 ml acetone and dried in vacuum. More product can be
obtained by successive recrystallization from the mother liquor
Crystals of 1, 2-Rh, 5-Ir, 5-Rh and 10 suitable for X-ray structural
determinations were mounted in polybutene oil on a glass fibre and
transferred on the goniometer head to the precooled instrument.
Crystallographic measurements were carried out with Mo Ka
radiation from a Incoatec Microfocus source on a Bruker APEX
single crystal diffractometer operating at 100(2) K. The structures
were solved by direct methods and refined against F2 by full
matrix least squares (SHELX97)20 using all unique data. All
non-hydrogen atoms were anisotropically refined unless otherwise
reported; the carbon bound hydrogen atoms were placed in
geometrically calculated positions and refined using a riding
model. Molecular graphics were prepared using ORTEP3.21 The
asymmetric unit of complex 2-Rh contains one co-crystallized
THF molecule. The thermal ellipsoid of the THF oxygen atom
suggests a small degree of disorder. The asymmetric unit of
complex 5-Ir contains one co-crystallized CH2Cl2 molecule. After
anisotropical refinement, several atoms (N1–N3, C2, C3, C7, C8,
C10, C12, C16, C17, C20) were non-positive definite, which is
assumed to be due to the low overall quality of the collected data.
ISOR was used to adjust the atomic displacements to realistic
values (0.008: C2, C3, C7, C12, C16; 0.007: C8, C17; 0.005: C10;
0.004: N1–N3). The asymmetric unit of complex 5-Rh contains
one co-crystallized CH2Cl2 molecule. The asymmetric unit of
complex 10 contains one co-crystallized diethylether molecule.
After anisotropical refinement the amido nitrogen atom (N4)
was non-positive definite. Refinement as a nitrogen atom seems
justified, since complex 10 is neutral and diamagnetic and displays
1J(1H-15N) coupling in the aforementioned HSQC experiment.
ISOR (0.0015) was used to adjust the atomic displacements
to realistic values. The amido hydrogen atom (H4n) could be
located in the Fourier electron density maps and was refined freely
with isotropic displacement parameters. Further details of the
1
giving a combined yield of 5.20 g (11.5 mmol, 29%). H NMR
3
(400 MHz, CD2Cl2): d = 8.51 (d, 2H, J = 7.8 Hz, CpyH(3,5));
3
3
7.99 (t, 1H, J = 7.8 Hz, CpyH(4)); 7.40 (d, 4H, J = 8.1 Hz,
3
CaromH(3,5)); 7.04 (t, 2H, J = 8.1 Hz, CaromH(4)); 2.36 (s, 6H,
NC-CH3) ppm. 13C{1H} NMR (100 MHz, CD2Cl2): d = 171.8
(C N); 155.0 (Cpy(2,6)); 146.2 (Carom(1)); 137.9 (Cpy(4)); 128.8
(Carom(3,5)); 125.0 (Carom(4)); 124.7 (Carom(2,6)); 124.0 (Cpy(3,5));
17.9 (NC-CH3) ppm. Anal. calcd. for C21H15Cl4N3: C, 55.90; H,
3.35; N, 9.31. Found: C, 56.00; H, 3.44; N, 9.3.
Synthesis of [Ir(Cl4N3)Cl] (2-Ir). In a 100 ml round-bottomed
Schlenk flask 2.12 g (4.69 mmol) of the ligand (Cl4N3) (1) was
dissolved in 15 ml of THF under N2. This solution was then
Table 4 Crystallographic data
Compound reference
Chemical formula
Formula mass
1
2-Rh·THF
C25H23Cl5N3ORh
661.62
Monoclinic
9.7883(2)
10.4967(2)
26.4832(5)
90.00
98.0900(10)
90.00
2693.93(9)
100(2)
P21/n
4
50469
6193
0.0252
0.0221
0.0512
0.0249
5-Rh·CH2Cl2
C23H19Cl9N3Rh
759.37
Monoclinic
11.5360(2)
19.3875(4)
13.7023(3)
90.00
106.1570(10)
90.00
2943.53(10)
100(2)
P21/n
4
59216
10337
0.0386
0.0310
0.0593
0.0488
5-Ir·CH2Cl2
C23H19Cl9IrN3
848.66
Monoclinic
11.557(9)
19.363(15)
13.696(11)
90.00
105.860(13)
90.00
2948(4)
103(2)
P21/n
4
15898
5183
0.1327
0.0553
0.0870
0.1217
10·Et2O
C21H15Cl4N3
451.16
Monoclinic
9.3466(5)
14.2468(7)
15.4849(8)
90.00
C43H41Cl4IrN4OSi
991.58
Crystal system
Triclinic
˚
a/A
10.8214(5)
12.3189(5)
15.9789(7)
81.104(2)
78.966(2)
86.105(3)
2063.97(16)
100(2)
˚
b/A
˚
c/A
a/◦
b/◦
g /◦
99.4560(10)
90.00
2033.94(18)
100(2)
P21/n
4
3
˚
Unit cell volume/A
Temperature/K
Space group
¯
P1
No. of formula units per unit cell, Z
No. of reflections measured
No. of independent reflections
Rint
Final R1 values (I > 2s(I))
Final wR(F2) values (I > 2s(I))
Final R1 values (all data)
Final wR(F2) values (all data)
Goodness of fit on F2
2
49083
5901
23008
8803
0.0558
0.0352
0.0822
0.0520
0.0873
0.972
0.0458
0.0517
0.1288
0.0626
0.1354
1.160
0.0534
1.049
0.0658
1.043
0.0999
0.715
9520 | Dalton Trans., 2011, 40, 9512–9524
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
treated with a red solution of 1.20 g (2.11 mmol) [Ir(C2H4)2Cl]2 in
30 ml Et2O, which resulted in an immediate colour change to dark
green. The solution was stirred for an additional hour, whereas
the green product precipitated from the solution. The solvent was
removed in vacuum and the dried raw product was washed in
~10 ml portions of a total of 60 ml of a THF/Et2O mixture (1/1)
and 5 ml of pentane. After drying the residue for several hours in
vacuum, the analytically pure product was obtained in 42% yield
temperature. The solution was then transferred to a Schlenk tube
which was sealed with a J.◦Young teflon valve. While stirred, the
solution was heated at 55 C for 16 h. The solvent was removed
in vacuum and residual solvents were co-evaporated with 3 ml
pentane. After drying in vacuum, 48 mg (71 mmol; 88%) of
the orange complex 5-Rh were obtained. Single crystals suitable
for X-ray diffraction analysis were obtained from a saturated
solution of complex 2-Rh in CH2Cl2 after several days at room
temperature. Because of the very low solubility of 5-Rh, 13C
NMR spectroscopic characterization was not attempted. 1H NMR
(400 MHz, CD2Cl2): d = 8.32 (m, 1H, CpyH(4)); 8.15 (m, 2H,
CpyH(3,5)); 7.47 (dd, 2H, 3J = 8.0 Hz, 4J = 1.4 Hz, CaromH(5)); 7.44
(dd, 2H, 3J = 8.1 Hz, 4J = 1.4 Hz, CaromH(3)); 7.29 (dd, 2H, 3J = 8.1
Hz, 3J = 8.1 Hz, CaromH(4)); 4.43 (d, 2H, 2JRh,H = 3.2 Hz, CH2Cl);
2.63 (s, 6H, NC-CH3) ppm. Anal. calcd. for C22H17Cl7RhN3: C,
39.18; H, 2.54; N, 6.23. Found: C, 38.94; H, 2.61; N, 5.99.
1
(1.35 g; 1.99 mmol). H NMR (400 MHz, CD2Cl2): d = 8.87 (t,
3
3
1H, J = 8.0 Hz, CpyH(4)); 8.09 (d, 2H, J = 8.0 Hz, CpyH(3,5));
3
3
7.53 (d, 4H, J = 8.3 Hz, CaromH(3,5)); 7.27 (t, 2H, J = 8.1 Hz,
CaromH(4)); 0.87 (s, 6H, NC-CH3) ppm. 13C{1H} NMR (100 MHz,
CD2Cl2): d = 176.6 (C N); 164.1 (Cpy(2,6)); 149 (Carom(1)); 129.5
(Carom(2,6)); 128.7 (Carom(3,5)); 128.4 (Carom(4)); 125.2 (Cpy(3,5));
124.3 (Cpy(4)); 20.6 (NC-CH3) ppm. IR (KBr): nmax [cm-1] = 3085,
1934, 1563, 1485, 1436(s), 1411, 1375, 1327, 1294, 1233, 1148,
790(m), 773, 729, 680. Anal. calcd. for C21H15Cl5IrN3: C, 37.15;
H, 2.23; N, 6.19. Found: C, 36.76; H, 2.35; N, 6.00.
Synthesis of [Ir(Cl4N3)OMe] (3). A 250 ml round bottom flask
was charged with 744 mg (1.10 mmol) of the complex [Ir(Cl4N3)Cl]
(2-Ir) and 179 mg (3.31 mmol) sodium methoxide. 15 ml THF
and 120 ml of methanol were added and the suspension was
stirred for 16 h. The solvents were removed in vacuum and the
residue was dried for six hours in vacuum. Residual solvents
were co-evaporated with two portions of 10 ml pentane each.
After additional nine hours of drying in vacuum, the product was
extracted from the residue into toluene. The toluene phase was
removed in vacuum, which gave 640 mg of the green complex 3
Synthesis of [Rh(Cl4N3)Cl] (2-Rh). A solution of 179 mg
(0.460 mmol) of the dimeric complex [Rh(C2H4)2Cl]2 in 8 ml
of toluene was added dropwise to a stirred solution of 400 mg
(0.887 mmol) of ligand 1 in 11 ml of toluene. The solution was
stirred for 11 h, whereas the green product precipitated from the
solution. The solution was decanted from the solid residue, which
was then washed with three portions of 5 ml THF/Et2O (1/2) and
dried in vacuum. 499 mg (0.846 mmol; 95%) of the green complex
[Rh(Cl4N3)Cl] (2-Rh) were obtained. Single crystals suitable for X-
ray diffraction analysis were obtained ◦from a concentrated THF
1
(0.95 mmol, 86%). H NMR (400 MHz, THF-d8): d = 8.61 (d,
3
3
2H, J = 7.9 Hz, CpyH(3,5)); 8.27 (t, 1H, J = 7.9 Hz, CpyH(4));
1
solution layered with hexane at -35 C. H NMR (400 MHz,
3
3
7.47 (d, 4H, J = 8.1 Hz, CaromH(3,5)); 7.27 (t, 2H, J = 8.1 Hz,
CaromH(4)); 5.06 (s, 3H, O-CH3); 0.54 (s, 6H, NC-CH3) ppm.
13C{1H} NMR (100 MHz, THF-d8): d = 167.4 (C N); 161.3
(Cpy(2,6)); 152.4 (Carom(1)); 130.4 (Carom(2,6)); 129.2 (Carom(3,5));
128.4 (Carom(4)); 124.4 (Cpy(3,5)); 117.6 (Cpy(4)); 68.0 (O-CH3);
21.0 (NC-CH3) ppm. IR (KBr): nmax [cm-1] = 3058, 2866, 2787,
1958, 1649, 1562, 1434(s), 1389(m), 1347, 1302(m), 1234, 1193,
1147, 1062, 1042, 996, 886, 790(m), 738, 725, 681, 567, 523. Anal.
calcd. for C22H18Cl4IrN3O: C, 39.18; H, 2.69; N, 6.23. Found: C,
39.29; H, 2.97; N, 6.10.
3
CD2Cl2): d = 8.59 (t, 1H, J = 8.0 Hz, CpyH(4)); 7.83 (d, 2H,
3J = 8.0 Hz, CpyH(3,5)); 7.47 (m, 4H, CaromH(3,5)); 7.27 (m,
2H, CaromH(4)); 1.60 (s, 6H, NC-CH3) ppm. 13C{1H} NMR
(100 MHz, CD2Cl2): d = 171 (C N); 156.5 (Cpy(2,6)); 146.0
(Carom(1)); 129.1 (Carom(2,6)); 128.9 (Carom(3,5)); 127.8 (Carom(4));
127.1 (Cpy(3,5)); 125.0 (Cpy(4)); 18 (NC-CH3) ppm. Anal. calcd.
for C21H15Cl5N3Rh: C, 42.78; H, 2.56; N, 7.13. Found: C, 42.89;
H, 2.64; N, 6.93.
Synthesis of [Ir(Cl4N3)(Cl)2(CH2Cl)] (5-Ir). A solution of
70 mg (100 mmol) of the complex [Ir(Cl4N3)Cl] (2-Ir) in 15 ml
CH2Cl2 was placed in a J. Young teflon valve tapped Schlenk
tube. While stirred, the solution was heated at 55 ◦C for 63 h,
upon which the color of the solution changed to orange. The
solution was allowed to cool to room temperature, which gave
single crystals suitable for X-ray diffraction. The solvent was
removed in vacuum and the residue was washed in portions
with 15 ml of pentane. After five hours drying in vacuum, the
analytically pure orange product was obtained in 95% yield
(72 mg, 94 mmol). Because of the very low solubility of 5-Ir, 13C
NMR spectroscopic characterization was not attempted. 1H NMR
(400 MHz, CD2Cl2): d = 8.03–7.95 (m, 3H, CpyH(3,4,5)); 7.50 (dd,
Synthesis of [Ir(Cl4N3)N3] (4-Cl). A solution of 237 mg
(351 mmol) of complex [Ir(Cl4N3)OMe] (3) in 7 ml of THF and a
solution of 85 ml (82 mg, 710 mmol) trimethylsilylazide in◦2 ml of
THF were cooled separately in scintillation vials to -35 C. The
cold azide solution was then added dropwise to the cold, stirred
solution of complex 3. The resulting reaction solution was stored
at -35 ◦C for 2.5 h. The solvent was then removed in vacuum and
the remaining volatiles were co-evaporated with 5 ml of pentane.
After drying the solid in vacuum for 4 h, the green product was
obtained in 91% yield (220 mg, 321 mmol). Because of the thermal
instability of 4-Cl, elemental analysis was not attempted. 1H NMR
(400 MHz, THF-d8): d = 8.74 (t, 1H, 3J = 7.9 Hz, CpyH(4));
3
4
3
2H, J = 8.1 Hz, J = 1.4 Hz, CaromH(5)); 7.46 (dd, 2H, J = 8.2
3
3
8.37 (d, 2H, J = 7.9 Hz, CpyH(3,5)); 7.53 (d, 4H, J = 8.2 Hz,
4
3
3
Hz, J = 1.4 Hz, CaromH(3)); 7.29 (dd, 2H, J = 8.2 Hz, J = 8.2
Hz, CaromH(4)); 4.14 (s, 2H, CH2Cl); 2.98 (s, 6H, NC-CH3) ppm.
Anal. calcd. for C22H17Cl7IrN3: C, 34.60; H, 2.24; N, 5.50. Found:
C, 34.46; H, 2.42; N, 5.19.
3
CaromH(3,5)); 7.26 (t, 2H, J = 8.1 Hz, CaromH(4)); 0.69 (s, 6H,
NC-CH3) ppm. 13C{1H} NMR (100 MHz, THF-d8): d = 173.9
(C N); 162.8 (Cpy(2,6)); 151.1 (Carom(1)); 129.8 (Carom(2,6)); 129.6
(Carom(3,5)); 129.0 (Carom(4)); 125.3 (Cpy(3,5)); 122.5 (Cpy(4)); 20.5
(NC-CH3) ppm. IR (KBr): nmax [cm-1] = 3067(w), 2920(w), 2872,
2034(vs) (nasym(N3)), 1638, 1562, 1476, 1436(m), 1408, 1379, 1349,
Synthesis of [Rh(Cl4N3)(Cl)2(CH2Cl)] (5-Rh). 48 mg (81 mmol)
2-Rh were dissolved in CH2Cl2 and stirred for 24 h at room
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 9512–9524 | 9521
©

View Article Online
1317, 1295, 1272, 1234, 1143, 1044, 790, 773, 728, 680. UV/vis
(KBr): lmax [nm] = 435, 492, 523, 820, 981.
Thermolysis of [Ir(Cl4N3)N] (6-Cl). A solid sample of 30 mg
(46 mmol) of complex [Ir(Cl4N3)N] (6-Cl) was placed in a Schlenk
tube equipped with a J. Young Teflon valve. The sample was then
placed under vacuum, the Schlenk tube was closed and heated to
150 ◦C for 19 h. Anal. calcd. for C21H15Cl4IrN: C, 38.37; H, 2.30;
N, 8.52. Found: C, 38.72; H, 2.44; N, 8.43.
Synthesis of [Ir(Cl4N3)15N N3/N2 15N] (4-Cl-15N). The syn-
thesis of the 15N labeled azido isotopologue (4-Cl-15N) was
performed analogous to the unlabeled complex (4-Cl) with in
situ prepared (CH3)3Si15N N2/N2 15N (50/50) as previously
described.4 IR (KBr): nmax [cm-1] = 3067(w), 2920(w), 2027(vs)
(nasym(14N215N)), 2017(vs) (nasym(15N14N2)), 1562, 1477, 1435(m),
1407, 1381, 1350, 1333, 1296, 1251, 1234, 1143, 1070, 1044, 995,
790, 774, 727, 680.
NMR scale reaction of [Ir(Cl4N3)N] (6-Cl) with Ph3SiH or
Et3SiH. A solution of either Ph3SiH or Et3SiH (ca. 0.23 mmol)
in 0.5 ml of THF-d8 was added to a solid sample of the
complex [Ir(Cl4N3)N] (ca. 15 mg, 0.023 mmol) containing ca.
5 mg (0.027 mmol) of ferrocene as an internal standard. The
resulting suspension was stirred for 2–3 min and then filtered. The
residue was extracted with 0.2 ml of THF-d8, the combined THF-
d8 solutions were placed in a NMR tube and a 1H NMR spectrum
Synthesis of [Ir(Me4N3)N3] (4-Me). A solution of 221 mg
(373 mmol) of the complex [Ir(Me4N3)OMe] in 8 ml of THF and a
solution of 99 ml (86 mg, 0.75 mmol) trimethylsilylazide in 3 ml of
◦
1
THF were cooled separately in scintillation vials to -35 C. The
was recorded. From integration of the H NMR product signals
cold azide solution was then added dropwise to the cold, stirred
solution of the iridium methoxido complex. The resulting reaction
solution was stored at -35 ◦C for 3 h. The solvent was then removed
in vacuum and the remaining volatiles were co-evaporated 3 times
with 3 ml of pentane. After drying the solid in vacuum for
several hours, the brown/green product was obtained in 90% yield
(203 mg, 336 mmol). Because of the thermal instability of 4-Me,
elemental analysis was not attempted. 1H NMR (200 MHz, THF-
d8): d = 8.61 (t, 1H, 3J = 7.9 Hz, CpyH(4)); 8.08 (d, 2H, 3J = 7.9 Hz,
CpyH(3,5)); 7.17–7.01 (m, 6H, CaromH(3,4,5)); 2.04 (s, 12H, CH3);
0.97 (s, 6H, NC-CH3) ppm. 13C{1H} NMR (100 MHz, THF-
d8): d = 171.7 (C N); 162.7 (Cpy(2,6)); 153.6 (Carom(1)); 131.0
(Carom(2,6)); 129.1 (Carom(3,5)); 127.1 (Carom(4)); 123.8 (Cpy(3,5));
121.9 (Cpy(4)); 18.9 (CH3); 18.6 (NC-CH3) ppm. IR (KBr): nmax
[cm-1] = 3064, 3025, 2954w, 2918w, 2038vs (nas(N3)), 1591, 1469 m,
1443w, 1380 m, 1352, 1314 m, 1294, 1218, 1174, 1165, 1165, 1092,
1047, 991, 841w, 776m. UV/vis (THF): lmax [nm] = 334, 430, 493,
589, 650, 800.
against the resonance of ferrocene, yields of 40–90% and 30–50%
for Ph3SiH and Et3SiH were estimated.
Reaction of [Ir(Cl4N3)N] (6-Cl) with Et3SiH. A solution of
95 ml (69 mg, 0.59 mmol) of Et3SiH in 10 ml of THF was added
to 41 mg (61 mmol) of the solid complex [Ir(Cl4N3)N] (6-Cl). This
resulted in a brown solution, which was stirred for 10 min. The
solvent was removed in vacuum and the residue was extracted
into Et2O. The Et2O phase was removed in vacuum, the remaining
volatiles were co-evaporated with 1 ml of pentane and the solid was
dried for ca. 5 h in vacuum. 29 mg of a brown solid were obtained,
1
which according to H NMR analysis were mainly consistent of
complex 9. All attempts to re-crystallize the complex were so far
unsuccessful or resulted in decomposition of the complex. 1H
NMR (400 MHz, C6D6): d = 8.20 (d, 2H, 3J = 7.9 Hz, CpyH(3,5));
8.06 (s, 1H, NH); 7.64 (t, 1H, 3J = 7.0 Hz, CpyH(4)); 7.11 (d, 4H,
3
3J = 8.1 Hz, CaromH(3,5)); 6.63 (t, 2H, J = 8.1 Hz, CaromH(4));
1.03 (t, 9H, 3J = 7.9 Hz, Si(CH2CH3)3); 0.41 (q, 6H, 3J = 7.9 Hz,
1
Si(CH2CH3)3); 0.19 (s, 6H, NC-CH3) ppm. H-15N HSQC (400
MHz/41 MHz, C6D6): d = 127 (NH) ppm.
DSC assisted synthesis of [Ir(Cl4N3)N] (6-Cl). Between 4 and
9 mg (6–13 mmol) of the complex [Ir(Cl4N3)N3] (4-Cl) were placed
in aluminum pans for DSC measurements and the pans were
sealed under a nitrogen atmosphere. The pans were opened with a
needle and the samples were then directly measured with a Netzsch
DSC 204 F1 instrument in cycles between 50 and 90 ◦C. The
measurement was stopped when no additional reaction enthalpy
could be detected. IR (KBr): nmax [cm-1] = 3070, 2975, 2870, 2036,
1951, 1636, 1561, 1434(m), 1366, 1345, 1286(s), 1243, 1135, 1094,
1046, 993, 955 (IrN), 891, 788, 737, 685, 552. UV/vis (KBr): lmax
[nm] = 466, 593, 770, 836. Anal. calcd. for C21H15Cl4IrN: C, 38.37;
H, 2.30; N, 8.52. Found: C, 38.94; H, 2.38; N, 8.30.
Synthesis of [Ir(Cl4N3)N(H)SiPh3] (10). A solution of 34 mg
(131 mmol) of triphenylsilane in 6 ml of toluene was added to a
solid sample of 77 mg (117 mmol) of the complex [Ir(Cl4N3)N] (6-
Cl). The resulting suspension was stirred for 25 min, upon which
most of the solid material dissolved, leading to a brown–green
colour of the supernatant. The suspension was filtered and the
residue was extracted with 1.5 ml of toluene. The solvent was
removed in vacuum and the resulting solid was co-evaporated
three times with 1.5 ml of pentane. The raw product was then
washed in 1.5 ml portions with a total of 20 ml of pentane.
The residue was then extracted into Et2O and filtered. The Et2O
was removed in vacuum and the washing/extracting procedure
was repeated. The product was recrystallized twice from an Et2O
Synthesis of [Ir(Cl4N3)N] (6-Cl), batch. In a Schlenk tube,
fitted with a J. Young teflon valve, 85 mg (126 mmol) 4-Cl were
placed under vacuum. The Schlenk tube was sealed and heated up
to 90 ◦C for one hour.
◦
solution layered with pentane at -35 C, yielding 8 mg (9 mmol;
8%) of the green complex 10. Single crystals suitable for X-ray
diffraction analysis were obtained from an Et2O solution at -35 ◦C
1
Synthesis of [Ir(Cl4N3)14N/15N] (6-Cl-15N). The synthesis of
the 15N labeled nitrido complex (6-Cl-15N) was performed analo-
gous to the unlabeled complex 6-Cl with a Netzsch DSC 204 F1
instrument in cycles between 50 and 90 ◦C. IR (KBr): nmax [cm-1] =
3070, 2914, 2026, 1951, 1636, 1561, 1435(m), 1345, 1285(m), 1242,
1135, 1094, 1070, 1022, 992, 955 (Ir14N), 925 (Ir15N), 891, 788, 738,
685, 552.
after several month. H NMR (400 MHz, THF-d8): d = 8.68 (d,
3
3
2H, J = 7.9 Hz, CpyH(3,5)); 8.25 (t, 1H, J = 7.9 Hz, CpyH(4));
7.22 (s(br), 1H, NH); 7.33–7.29 (m, 6H, CphenylH(2,6)); 7.18–713
(m, 3H, CphenylH(4)); 7.11–7.06 (m, 6H, CphenylH(3,5)); 6.89 (m, 4H,
CaromH(3,5)); 6.74 (m, 4H, CaromH(4)); 0.47 (s, 6H, NC-CH3) ppm.
1
13C{ H} NMR (100 MHz, THF-d8): d = 167.2 (C N); 159.3
(Cpy(2,6)); 150.8 (Carom(1)); 142.7 (Cphenyl(1)); 136.9 (Cphenyl(2,6));
9522 | Dalton Trans., 2011, 40, 9512–9524
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
130.1 (Carom(2,6)); 129.1, 128.9, 128.4, 127.7 (Carom(3,5), Carom(4),
phenyl(3,5), Cphenyl(4)); 124.0 (Cpy(3,5)); 118.2 (Cpy(4)); 20.8 (NC-
assigned structure with a terminal nitrido unit. Currently, we are in
the process to synthesize related complexes with chloro protected
ligands of higher solubility and will report their chemistry in due
course. Mechanistic aspects of silane addition to iridium nitrido
units to produce amido complexes will be discussed in detail in a
separate forthcoming publication.
C
CH3) ppm. 1H-15N HSQC (400 MHz/41 MHz, THF-d8) d [ppm]:
104 (IrNH-SiPh3). 1H-29Si HMBC (400/79 MHz, THF-d8) d
[ppm]: -16 (IrNH-SiPh3). Anal. calcd. for C39H31Cl4IrN4Si: C,
51.04; H, 3.40; N, 6.10. Found: C, 50.31; H, 3.42; N, 5.98.
VT NMR measurement of [Ir(Cl4N3)N(H)SiPh3] (10). 3 mg
(3 mmol) of complex 10 were dissolved in 0.7 ml of THF-d8.
The solution was transferred to a NMR tube fitted to a Cajoin
ultravacuum adapter attached to a J. Young high vacuum tap. The
NMR tube was then transferred to a high vacuum manifold, the
solution was frozen in liquid nitrogen and the NMR tube was
flame sealed under vacu◦um. The solution was thawed by shaking
the NMR tube at -100 C in an EtOH/N2 coldbath. The NMR
Acknowledgements
Funding by Deutsche Forschungsgemeinschaft is gratefully ac-
knowledged. The authors are indebted to Prof. Ulrich Behrens for
assistance at the final stage of the crystal structure analysis and
Dr Thomas Hackl for helpful discussions concerning lineshape
analysis. The authors would like to thank Uta Sazama and Sandra
Maracke for DSC and DTG/MS measurements and Eike Anheier,
Jan-Moritz Adam and Matthias Schreyer for assistance with the
syntheses.
1
tube was then transferred to the pretempered NMR probe. H
NMR spectra were recorded from -100 ◦C to -50 ◦C in 10 ◦C
steps.
Notes and references
Computational Studies
1 J. F. Berry, Comments Inorg. Chem., 2009, 30, 28–66.
2 A. Walstrom, M. Pink, X. Yang, J. Tomaszewski, M.-H. Baik and K.
G. Caulton, J. Am. Chem. Soc., 2005, 127, 5330–5331.
3 (a) C. Vogel, F. Heinemann, J. Sutter, C. Anthon and K. Meyer, Angew.
Chem., Int. Ed., 2008, 47, 2681–2684; (b) J. J. Scepaniak, M. D. Fulton,
R. P. Bontchev, E. N. Duesler, M. L. Kirk and J. M. Smith, J. Am.
Chem. Soc., 2008, 130, 10515–10517.
4 J. Scho¨ffel, A. Yu Rogachev, S. DeBeer George and P. Burger, Angew.
Chem., Int. Ed., 2009, 48, 4734–4738.
5 T. A. Betley and J. C. Peters, J. Am. Chem. Soc., 2004, 126, 6252–6254.
6 J.-U. Rohde, T. A. Betley, T. A. Jackson, C. T. Saouma, J. C. Peters and
L. Que, Inorg. Chem., 2007, 46, 5720–5726.
The DFT calculations were carried out with the parallelized
versions 6.2 of Turbomole22 on our 192 and 80 core Intel Harper-
town and Nehalem clusters. The geometries of the ground and
transition states were optimized without symmetry restrictions
(C1-symmetry). The complexes described herein were probed for
singlet and triplet states S = 0 and 1 using unrestricted calculations.
They were characterized as stationary points by the calculation of
analytical second derivatives through the absence (ground states)
or presence of one imaginary frequency (transition states). The
transition states were further analyzed by IRC calculations, which
assured the correctness of the reaction paths. For the latter, the
Gaussian 03 geometry optimizer23 was used in combination with a
perl script (Gau_external written by Karin Wichmann) interfacing
to Turbomole. Additionally a perl script graciously provided by Dr
Armin Hellweg from COSMOlogic GmbH & Co. KG was utilized
for the calculation of IRC paths.
The Becke-Perdew (BP-86) functional24 was used for the
geometry optimizations employing the resolution of identity
(RI) methodology25 available in Turbomole (ridft, rdgrad). The
modified grids, m3 and m4, available in Turbomole were employed
for the DFT calculations.
The dependence of the energetics (thermodynamics and bar-
riers) on the DFT functionals and the basis set dependence
was previously probed for related systems4,11 and evidenced only
negligible deviations. We therefore refrained from the assessment
of the performance of the DFT functionals and basis sets for
the systems described herein. For the iridium center a Stuttgart-
Dresden pseudopotential (ECP-28-MWB- and ECP-60-MWB)
was used throughout.26 The employed DFT functional and basis
sets are provided along with the optimized geometries in the ESI†.
7 J. Scepaniak, J. Young, R. Bontchev and J. Smith, Angew. Chem., Int.
Ed., 2009, 48, 3158–3160.
8 (a) W.-L. Man, W. W. Y. Lam, H.-K. Kwong, S.-M. Peng, W.-T. Wong
and T.-C. Lau, Inorg. Chem., 2010, 49, 73–81; (b) X.-Y. Yi, H.-Y. Ng,
I. D. Williams and W.-H. Leung, Inorg. Chem., 2011, 50, 1161–1163.
9 Y. Badiei, A. Dinescu, X. Dai, R. Palomino, F. Heinemann, T. Cundari
and T. Warren, Angew. Chem., Int. Ed., 2008, 47, 9961–9964.
10 (a) C. Ni, J. C. Fettinger and P. P. Power, Organometallics, 2010, 29,
269–272; (b) A. K. Musch Long, R. P. Yu, G. H. Timmer and J. F. Berry,
J. Am. Chem. Soc., 2010, 132, 12228–12230; (c) C. C. H. Atienza, A. C.
Bowman, E. Lobkovsky and P. J. Chirik, J. Am. Chem. Soc., 2010, 132,
16343–16345.
ˇ
11 J. Scho¨ffel, N. Susˇnjar, S. Nu¨ckel, D. Sieh and P. Burger, Eur. J. Inorg.
Chem., 2010, 4911–4915.
12 Y. Chen, R. Chen, C. Qian, X. Dong and J. Sun, Organometallics, 2003,
22, 4312–4321.
13 (a) A. Mentes, J. Fawcett and R. D. W. Kemmitt, Acta Crystallogr.,
Sect. E: Struct. Rep. Online, 2001, 57, o424–o425; (b) Y.-B. Huang,
X.-L. Ma, S.-N. Zheng, J.-X. Chen and C.-X. Wei, Acta Crystallogr.,
Sect. E: Struct. Rep. Online, 2006, 62, o3044–o3045.
14 H. F. Haarman, J. M. Ernsting, M. Kranenburg, H. Kooijman, N.
Veldman, A. L. Spek, P. W. N. M. van Leeuwen and K. Vrieze,
Organometallics, 1997, 16, 887–900.
15 E. L. Dias, M. Brookhart and P. S. White, Organometallics, 2000, 19,
4995–5004.
16 S. Nu¨ckel and P. Burger, Organometallics, 2001, 20, 4345–4359.
17 The residues of some of the batch reactions dissolved partially in THF.
However, the UV/vis and IR spectra of the THF extracted material
are different to the spectra of the solid samples from which they have
been obtained (cf. ESI†). This suggests that the extracted compounds
are minor impurities or decomposition products.
Conclusions
The chloride protection of potential ligand sites amenable to C–
H activation allowed us isolate a terminal nitrido complex 6-
Cl with sterically less demanding groups. The low solubility in
organic solvents thwarted the full characterization of the nitrido
compound. Its reaction with silanes to give the corresponding
silyl amido complexes provided however good evidence for the
18 M. A. Arthurs, J. Bickerton, S. R. Stobart and J. Wang, Organometallics,
1998, 17, 2743–2750.
19 R. R. Cramer, J. A. McCleverty and J. Bray, Inorg. Synth., 1990, 28,
86–88.
20 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2007,
64, 112.
21 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 568.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 9512–9524 | 9523
©

View Article Online
22 TURBOMOLE V6.2 2010, a development of University of Karl-
sruhe and TURBOMOLE GmbH since 2007, available from
http://www.turbomole.com.
Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K.
Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q.
Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith,
M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P.
M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A.
Pople, Gaussian, Inc., Wallingford CT, 2004.
23 Gaussian 03, Revision C.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr.,
T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross,
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G.
24 (a) A. D. Becke, Phys. Rev. A: At., Mol., Opt. Phys., 1988, 38, 3098;
(b) J. P. Perdew, Phys. Rev. B, 1986, 33, 8822.
25 O. Treutler and R. Ahlrichs, J. Chem. Phys., 1995, 102,
346.
26 D. Andrae, U. Haeussermann, M. Dolg, H. Stoll and H. Preuss, Theor.
Chim. Acta, 1990, 77, 1431.
9524 | Dalton Trans., 2011, 40, 9512–9524
This journal is
The Royal Society of Chemistry 2011
©