Synthesis of planar chiral iridacycles by cationic metal π-coordination: Facial selectivity, and conformational and stereochemical consequences

Article abstract of DOI:10.1002/chem.201103577

Facial selectivity during the π-coordination of pseudo-tetrahedral iridacycles by neutral (Cr(CO)₃), monocationic (Cp*Ru ⁺), and biscationic (Cp*Ir²⁺) metal centers was directly influenced by the coulombic imbalance in the

Full text of DOI:10.1002/chem.201103577

FULL PAPER
DOI: 10.1002/chem.201103577
Synthesis of Planar Chiral Iridacycles by Cationic Metal p-Coordination:
Facial Selectivity, and Conformational and Stereochemical Consequences
Jean-Pierre Djukic,*[a] Wissam Iali,[a] Michel Pfeffer,[a] and Xavier-Frꢀdꢀric Le Goff[b]
On the occasion of the retirement of FranÅoise Rose-Munch and Eric Rose
Abstract: Facial selectivity during the
p-coordination of pseudo-tetrahedral
iridacycles by neutral (Cr(CO)3),
monocationic (Cp*Ru+), and biscation-
ic (Cp*Ir2+) metal centers was directly
influenced by the coulombic imbalance
in the coordination sphere of the che-
lated Ir center. We also showed by
using theoretical calculations that the
feasibility of the related metallacycles
that displayed metallocenic planar chi-
rality was dependent to the presence of
an electron-donating group, such as
NMe2, which contributed to the overall
stability of the complexes. When the p-
bonded moiety was the strongly elec-
tron-withdrawing Cp*Ir2+ group, the
electron donation from NMe2 resulted
in major conformational changes, with
a barrier to rotation of about 17 kcal
molꢀ1 for this group that became spec-
troscopically diastereotopic (high-field
1H NMR spectroscopy). This peculiar
property is proposed as a means to in-
troduce a new type of constitutional
chirality at the nitrogen center: planar
chirality at tertiary aromatic amines.
Keywords: density functional calcu-
lations · iridium · metallacycles · pi
interactions · planar compounds
Introduction
planar chiral metallacycles, around 80% involved square-
planar (SP-4) palladacycles.[5] Only a few cases of racemic or
enantioenriched planar chiral metallacycles with octahedral
(OC-6)[6] or pseudo-tetrahedral (pseudo-T-4)[7] chelated
metal centers have been reported.
Planar chirality has been used in ligands and catalysts with
the aim of promoting larger degrees of chiral induction in
systems in which the site of a given catalytic event was
located at or near to the planar chiral moiety.[1] Because
metallacycles[2] are known to have an aptitude for catalysis,[3]
the study of planar chiral analogues has fostered a wide
range of studies for more than two decades.
Various methods have been used to design molecules with
such geometric peculiarities, among which direct metalation
by C H bond activation and the so-called transmetalation
The challenge for OC-6 and pseudo-T-4 metallacycles is
to produce planar chiral analogues in a stereocontrolled
way, so as to provide compounds that would open new hori-
zons of applications. Enantiomeric enrichment is the under-
lying issue, and it can be addressed by resolving cationic
planar chiral metallacycles by using enantioenriched anions,
such as TRISPHAT.[8] Yet, the question to be answered is
how such cationic racemic planar chiral metallacycles can be
made stereospecifically by the p-coordination of a preformed
planar prochiral metallacyclic precursor (playing the role of
a formal Lewis base) with a Lewis-acidic electron-poor
p-bonding metal (Scheme 1)? The answer to this question is
in the fine-tuning of the electronic interactions, that is, cova-
lent and non-covalent, to ensure the stability of the p-com-
plex. Herein,[9] we report our investigations into the p-coor-
ꢀ
of pre-formed complexes are well represented in the litera-
ture.[4] Because the ultimate goal is the design of enantio-
pure (or enantioenriched) compounds, efficient and stereo-
specific routes to afford planar chiral compounds are highly
desirable.
In a recent review of the topic,[4] it was shown that,
among the reports that have dealt with the synthesis of
[a] Dr. J.-P. Djukic, W. Iali, Dr. M. Pfeffer
Institut de Chimie, CNRS
Universitꢀ de Strasbourg
4 rue Blaise Pascal, 67008 Strasbourg Cedex (France)
Fax : (+33)368851523
E-mail: djukic@unistra.fr
[b] Dr. X.-F. Le Goff
Laboratoire Hꢀtꢀroꢀlꢀments et Coordination, CNRS
Ecole Polytechnique, Route de Saclay
91128 PALAISEAU CEDEX (France)
Scheme 1. Can the p-coordination of a chiral metallacycle be face-selec-
tive? Facial discrimination may occur by tuning the electronic properties
of the ligands around M1.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201103577.
Chem. Eur. J. 2012, 18, 6063 – 6078
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6063

dination of prototypical pseudo-T-4 iridacycles that con-
tained a formal stereogenic center at the iridium atom by
[Cp*IrACHTUNGTRENNUNG ACHTUGNTRENNUGN
(acetone)3]2+ and [Cp*Ru(MeCN)3]+. The charge
born by the incoming metal moiety influenced the facial
specificity of the p-coordination process, which had collater-
al effects on the donating substituents of the chelating
ligand, such as NMe2. These latter substituents displayed un-
usual restrictions to rotation that generated a new class of
a dynamic stereogenic centers at the tertiary amine nitrogen
atoms.
Scheme 3. Structures of iridacycles 2a–2d used in this study.
Results and Discussion
The reactions of compounds 2a, 2b, and 2d with freshly
ꢀ
prepared solutions of [Cp*Ir
N
Synthesis of biscationic and cationic planar chiral irida-
cycles: It is well-established that the efficiency of the synthe-
sis of (h6-arene)tricarbonylchromium complexes by the ther-
molytic treatment of Cr(CO)6 with arenes directly depends
on the nature of the substituents on the arene.[10] Electron-
donors, in the widest sense, as inductive and/or mesomeric
donors, promote the p-coordination of the Cr(CO)3 frag-
ment, whereas electron-withdrawing groups disfavor p-coor-
dination and are typically responsible for poor yields. These
rules apply to all other types of p-coordination reactions in
which an electron-deficient metal-containing moiety binds
to an arene through its p-system.[11] In a recent report,[12] it
was shown that these electronic prerequisites in the synthe-
sis of a triscationic planar-chiral ruthenacycle were condi-
tional for the p-coordination of a [(h6-benzene)Ru]2+ moiety
to a ruthenacycle. The presence of an NMe2 substituent was
necessary for the formation of the triscationic homobi-
nuclear product.
tempted in acetone at room temperature over 24 h.
Herein, the prototypical iridacyclic arene “ligands”, i.e.,
compounds 2a and 2b–2d, were obtained in high yields
(80% for compounds 2b–2d) by the cycloiridation of 2-aryl-
pyridines (Scheme 2) with [(Cp*IrCl2)2][13] according to the
Under these conditions, compound 2a remained unreact-
ed, whereas compounds 2b and 2d were converted into
unique products, i.e., compounds 3bACTHUNRTGENNUG[PF6]2 and 3dACHTUNGTRENNUNG[PF6]2, in
78 and 20% yield, respectively (reaction (1)). The time re-
quired for complete conversion could be shortened to
15 min when the reaction of compound 2b with [Cp*Ir-
[15]
ACHUTNRGENU(NG acetone)3]ACHTUNGTRENN(GUN PF6)2 was carried out at low concentrations in
acetone. The reactions of ligands 2a–2c with [Cp*Ru-
[16]
AHCTUNGTRENNUNG
Scheme 2. 2-Arylpyridines used in this study and their substitution
patterns.
ACHTUNGTRENNUNG
30, 75, and 75% yield, respectively (reaction (2)). A long re-
action time was required to ensure the full conversion of the
substrates. However, we observed that a conversion of
about 80% could be obtained in approximately 7 h for com-
method reported by Davies et al.[14] These iridacycles
(Scheme 3) were chiral, that is, they possessed a stereogenic
center at the Ir atom, which sat at the center of a quasi-tet-
rahedron, if the pentamethylcyclopentadienyl ligand was as-
similated into a single substituent at the central metal
center. In these complexes, the halogenido ligand and the
bulky Cp* ligand were part of the coordination sphere of
a pseudo-T-4 iridium (III) center together with the aryl and
pyridyl moieties of the [C,N] chelate.
pound 4bACTHNUGRTENUNG[PF6].
For the purposes of comparison, we also undertook the
synthesis of the Cr(CO)3-containing complexes 5a and 5b
by the cycloiridation of ligands 1 f and 1g (Scheme 4). The
latter reaction afforded compounds 5a and 5b in 75 and
70% yield, respectively. Alternatively, compound 5a was
also synthesized by the quantitative ligand-exchange reac-
tion of tricarbonyl(h6-naphthalene)chromium[17] with com-
6064
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 6063 – 6078

Synthesis of Planar Chiral Iridacycles
FULL PAPER
pound 2b. The determination
pounds 4b, 4c+, and 5b, the
NMe2 moiety was not perfectly
planar but slightly pyramidal,
with the nitrogen atom lying
about 0.11 ꢂ above the mean
plane (defined by the Cipso
atom, the N atom, and the two
carbon atoms of the methyl
of the relative configuration of
each of these cationic com-
plexes was addressed by struc-
tural analysis.
Structural analysis and spectro-
Scheme 5. Geometrical param-
eters used in this study.
scopic
shows ORTEP-like structures
of compounds 1g, 2b, 3b[PF6]2,
4b[PF6], 4c[PF6], and 5b; the
acquisition and refinement data
for all of the structures discussed herein are given in
Table 1. The structures of compounds 2b, 3b[PF6]2, 4b[PF6],
[PF6], and 5b confirmed that the CAr NMe2 bond
properties:
Figure 1
Scheme 4. Structure of ligands
1 f and 1g.
ꢀ
groups). The Ir Cl distance
T
(dCl, Table 2, Scheme 5) was
A
U
also sensitive to the electronic
changes that occurred upon p-coordination. The Ir Cl in-
teratomic distance in compound 2b (2.4122(6) ꢂ) shortened
ꢀ
A
U
significantly in compounds 5b, 4b[PF6], 4c
3b[PF6]2 (Table 2).
ACHTUNGTERN[NUNG PF6], and
ꢀ
4c
A
ACHTUNGTRENNUNG
ꢀ
(Scheme 5, dN) was very sensitive to the nature of the
p-bonded metal atom. This interatomic distance shortened
slightly (by about 0.01 ꢂ) upon coordination of the Cr(CO)3
and [Cp*Ru]+ moieties to the iridacycle (assuming that the
tBu group on the pyridyl fragment had little influence),
whereas the coordination of [Cp*Ir]2+ caused a “contraction”
Similar effects were observed for the CAr Ir bond, which
shortened significantly on going from compound 2b to com-
pounds 5b, 4bACTHUNTRGENN[UG PF6], and 3bACHTUGNTRENN[UNG PF6]2 (Table 2). All of these
structural changes were related to the electron-withdrawing
character of the p-bonded metal moiety. Interestingly,
1H NMR spectroscopy provided further information that
was not available from X-ray diffraction: the constitutional-
ly diastereotopic methyl groups of the NMe2 substituent in
ꢀ
of about 0.05 ꢂ. This shortening of the C NMe2 bond did
not alter its single-bond character significantly. In com-
compounds 5b and 4b-cACHTNUTRGNEUGN[PF6] did not resonate as two re-
solved singlets but rather as a unique singlet. Cooling the
samples to ꢀ708C only resulted in a slight broadening of
the signal.
This observation may have two explanations: 1) the two
expected signals for the NMe2 resonance coincided acciden-
tally, or 2) the electron-withdrawing effect of the Cr(CO)3
and [Cp*Ru] moieties was not strong enough to induce
1
a strained rotation on the H NMR timescale of the spec-
trometers used (300 to 500 MHz); in other words, the two
methyl groups were conformationally homotopic. The most-
likely interpretation is the second one (see below). Numer-
ous planar chiral tricarbonylchromium complexes of N,N-di-
alkylaniline have been reported[18] and, to the best of our
knowledge, only very few complexes have displayed re-
solved 1H NMR signals for the diastereotopic methyl
groups.[19]
1
In contrast, the H NMR (CD2Cl2) spectrum of compound
3bACTHUNGRTNEUNG[PF6]2 displayed two distinct singlets that could be as-
signed to the NMe2 methyl groups at d=3.59 and 3.46 ppm
(Dd=0.15 ppm). The separation (Dd) of the two methyl sin-
glets depended on the nature of the anions in solution and
therefore on the tightness of the ion pair; the same depend-
ence applied to the aromatic region. For instance, for com-
pound 3b
that of compound 3b
addition of excess NaBPh4 to a solution of compound
3b[PF6]2 in CD2Cl2 drastically decreased the Dd value to
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
0.06 ppm.
Figure 1. ORTEP-type structures of: a) compound 1g; b) 2b; c) 3b
d) 4b[PF6], light-gray ellipsoids indicate alternative positions for the
C atoms of the Cp* ligand as a result of 50% positional disorder; e)
4c[PF6], light-gray ellipsoids indicate alternative positions for the
C atoms of the tBu group as a result of 50% positional disorder; f) 5b.
Ellipsoids set at 30% probability; hydrogen atoms, counteranions, and
solvent molecules have been omitted for clarity.
ACHTUGNTERN[NUNG PF6]2;
ACHTUNGTRENNUNG
Another interesting feature in the structures of p-com-
plexes 1g, 3bACHTGNUTERN[NGUN PF6]2, 4bACHTUNTREGNN[GUN PF6], 4cACHTNUGRTNEUN[GN PF6], and 5b was the distor-
tion of the hexahapto bonding mode of the p-bonded metal
fragments: the longest interatomic distance between the
p-bonded metal (Ir, Ru, Cr) and the arene carbon atoms
ACHTUNGTRENNUNG
Chem. Eur. J. 2012, 18, 6063 – 6078
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6065

J.-P. Djukic et al.
Table 1. Acquisition and refinement parameters from the X-ray diffraction of compounds 1g, 2b, 3b
[6b·OH2·A(PF6)3].
G
G
ACHTUGTNRENUNG[PF6], 4cHCATUNGTRENN[GUN PF6], 5b, and
CHTUNGTRENNUNG
Compound
1g
2b
3b
G
ACHTUNGTRENNUNG
formula
Mr
crystal shape/color
crystal size [mm]
crystal system
space group
a [ꢂ]
b [ꢂ]
c [ꢂ]
C20H22CrN2O3·CH2Cl2
475.32
orange plates
0.22ꢃ0.06ꢃ0.03
monoclinic
P21
9.822(1)
7.533(1)
14.900(1)
90.00
C23H28ClIrN2
560.12
yellow blocks
0.38ꢃ0.30ꢃ0.20
monoclinic
P21/c
16.5371 (6)
7.5131 (3)
20.9885 (6)
90.00
C33H43ClIr2N2·C3H6O·2PF6
1235.56
orange needles
0.28ꢃ0.12ꢃ0.10
monoclinic
P21/c
13.019(1)
20.090(1)
17.132(1)
90.00
C34H45ClIr2N2·C3H6O·2PF6
1249.59
yellow plates
0.40ꢃ0.08ꢃ0.01
monoclinic
P21/c
13.079(1)
20.186(1)
17.124(1)
90.00
a [8]
b [8]
90.094(1)
90.00
1102.4(2)
2
1.432
492
128.903 (2)
90.00
2029.35 (12)
4
1.833
1096
108.824(1)
90.00
4241.2(5)
4
1.935
2384
109.316(1)
90.00
4266.5(5)
4
1.945
2416
g [8]
V [A3]
Z
1 [gcmꢀ3
F (000)
T [K]
]
150.0(1)
27.48
173.0(1)
32.06
150.0(1)
30.02
150.0(1)
27.48
qmax [8]
h, k, l range
ꢀ12 12,
ꢀ24 20,
ꢀ10 11,
ꢀ30 31
20707
7014,
6324
ꢀ15 18,
ꢀ24 28,
ꢀ24 22
ꢀ14 16,
ꢀ25 26,
ꢀ22 21
ꢀ9 9,
ꢀ16 19
9982
9982,
4348
267
total reflns
unique reflns,
reflns with I>2s(I)
parameters
Rint
30593
30821
12327,
8865
519
0.0464
9657,
8232
528
0.0458
251
0.017
0.0449
R1
0.0589
0.023
0.0440
0.0484
wR2
0.1166
0.052
0.1312
0.1224
x (Flack)
GoF
0.06(3)
1.094
1.136
0.990
1.056
Compound
formula
Mr
crystal shape/color
crystal size [mm]
crystal system
space group
a [ꢂ]
4b
[PF6]
4c
[PF6]
5b
ACHUTNGTREN[NUG 6b·OH2·ACHTUGNTERN(NUGN PF6)3]
C33H43ClIrN2Ru·PF6
941.38
orange plates
0.24ꢃ0.18ꢃ0.12
monoclinic
P21/c
8.671(1)
13.989(1)
28.862(1)
90.00
2(C37H51ClIrN2Ru)·PF6
2152.91
yellow blocks
0.18ꢃ0.14ꢃ0.12
orthorhombic
Pba2
18.995(1)
19.887(1)
12.638(1)
90.00
C30H36ClCrIrN2O3·C3H6O
810.34
orange needles
0.34ꢃ0.04ꢃ0.02
monoclinic
P21/c
10.649(1)
15.812(1)
21.131(1)
90.00
C33H45Ir2N2O·2C3H6O·3PF6
1508.29
yellow blocks
0.30ꢃ0.16ꢃ0.16
monoclinic
P21/n
10.805(1)
23.863(1)
21.248(1)
90.00
b [ꢂ]
c [ꢂ]
a [8]
b [8]
105.194(2)
90.00
3378.5(5)
4
1.851
1848
90.00
90.00
4774.0(5)
2
1.498
2150
150.0(1)
30.03
107.397(3)
90.00
3395.3(4)
4
1.585
1616
93.916(1)
90.00
5465.8(6)
4
1.833
2952
g [8]
V [A3]
Z
1 [gcmꢀ3
F (000)
T [K]
]
150.0(1)
30.02
150.0(1)
29.95
150.0(1)
27.48
qmax [8]
h, k, l range
ꢀ12 12,
ꢀ19 18,
ꢀ31 40
29565
9531,
7714
ꢀ24 26,
ꢀ28 20,
ꢀ16 17
41776
13510,
12138
ꢀ12 14,
ꢀ22 20,
ꢀ24 29
26792
9667,
6935
ꢀ14 13,
ꢀ30 30,
ꢀ27 20
total reflns
unique reflns,
reflns with I>2s(I)
parameters
Rint
38423
11868,
10972
694
0.0391
454
0.0298
428
0.0358
391
0.0494
R1
0.0314
0.0442
0.0495
0.0510
wR2
0.0704
0.1241
0.0624
0.1157
x (flack)
GoF
0.034(6)
1.062
1.023
1.127
1.126
6066
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 6063 – 6078

Synthesis of Planar Chiral Iridacycles
FULL PAPER
Table 2. Selected interatomic distances [ꢂ] and angles [8].[a]
endo-IIIdCp*2+, respectively, where the endo prefix indicated
the position of the chlorido ligand with respect to the
p-bonded metal center. By applying the energy-decomposi-
tion analysis (EDA) of Morokuma[22] and Ziegler and co-
workers,[23] we extracted the interaction energy component
(DEint) that corresponded to the release of stabilizing energy
upon the interaction of two geometrically prepared frag-
ments, that is, the prepared iridacycle and [Cp*Ir]2+.
Figure 2 shows the minimum-energy singlet-ground-state ge-
Compound
dCl
dN
dipso
dM
a
b
1g
2b
–
1.355(5)
–
2.365(4) 116.2(4)
–
2.4122(6) 1.379(5) 2.037(3)
–
118.0(3) 77.94(12)
3b
4b
4cACHTUNGTRENNUNG
N
2.373(2) 1.326(7) 2.021(6) 2.444(5) 114.6(5) 78.7(2)
2.387(5) 1.378(4) 2.058(3) 2.325(3) 117.5(3) 77.7(1)
2.392(2) 1.35(1) 2.087(5) 2.338(7) 115.1(6) 77.9(2)
2.402(1) 1.366(6) 2.046(4) 2.359(5) 117.2(4) 78.1(2)
[a] Geometric parameters defined in Scheme 5.
was that to the carbon that contained the NMe2 group (dM;
Scheme 5). This phenomenon was already well-known for
arenetricarbonylchromium complexes.[20] Hunter et al. inter-
preted this phenomenon as the consequence of repulsive
electronic interactions between the metal and the lone pair
of the electrons at the nitrogen center, which could induce
a significant folding of the arene ring in some extreme
cases.[11b] Gadre and co-workers later established a direct
correlation between the structural distortions and the elec-
tronic properties of the substituent on the arene.[21] Further
theoretical investigations by using DFT calculations support-
ed this assertion (see below). Such folding existed in most of
the structures shown in Figure 1. For instance, in complex
2+
Figure 2. Singlet-ground-state gas-phase geometries of endo-IIIaCp*
,
2+
2+
endo-IIIbCp*
,
and endo-IIIdCp*
at the (ZORA) PBE/all-electron
TZP(Ir), all-electron DZP(remaining elements) level. Bonds to Cp* and
aryl compounds have been omitted for clarity. Intrinsic interaction ener-
gies DEint (kcalmolꢀ1) between prepared [Cp*Ir]2+ and the related irida-
cycle are reported below the structures.
3bACHTUNGTRENNUNG[PF6]2, the folding of the arene ligand at atoms C2 and C4
was about 48. Table 3 lists the main geometrical features of
the models discussed herein.
ometries of the three models and their associated DEint
values. It was interesting to note a difference of around
18 kcalmolꢀ1 between the computed values for endo-
Table 3. Selected interatomic distances [ꢂ] and angles [8] for the com-
puted geometries of the model compounds.[a]
Model
Method dCl
dN
dipso
dM
a
b
2+
IIIbCp* and endo-IIIaCp*2+, which was in good agreement
[b]
IIbCp
IIbCp
IIbCp*
endo-IIIaCp*
endo-IIIbCp*
endo-IIIbCp*
endo-IIIbCp
endo-IIIbCp
exo-IIIbCp
endo-IIIdCp*
endo-IVbCp
endo-IVbCp
2.401 1.387 2.025
2.395 1.385 2.019
2.404 1.382 2.018
–
–
–
118.4
118.4
118.4
78.6
78.7
78.3
with the data for triscationic bisruthenium planar chiral
compounds, for which the difference in interaction energy
was 22 kcalmolꢀ1 in favor of the NMe2-substituted metalla-
cycle.[12] A difference of around 6 kcalmolꢀ1 separated endo-
[c]
[d]
[d]
[d]
[e]
[b]
[f]
2+
2+
2.401
–
1.962 2.238 119.8* 78.5
2.404 1.343 1.975 2.472 114.1
2.420 1.337 1.989 2.476 114.1
2.388 1.335 1.985 2.524 112.2
2.383 1.330 1.981 2.534 112.3
2.399 1.327 1.998 2.586 110.6
78.5
78.3
79.0
79.0
78.9
2+
2+
IIIaCp* from endo-IIIdCp*2+, which was somewhat consistent
2+
2+
with the low yield obtained for the methoxy-substituted
complex 3dACHTUNTRGNEUNG[PF6]2.
2+
[b]
[d]
[b]
[f]
2+
2.420
–
1.989 2.476 114.1* 78.3
The central role of the NMe2 substituent was studied by
using the Mittoraj–Michalak–Ziegler[24] combination of the
extended-transition-state method with the natural orbitals
for chemical valence theory (the so-called “ETS-NOCV
scheme”) to investigate the p-interaction between the
closed-shell [Cp*Ir]2+ fragment and the closed-shell arene in
endo-IIIbCp*2+. In principle, this analytical scheme was capa-
ble of providing a fine deconvolution into the different com-
ponents (s, p, d-type orbitals) of a bonding interaction be-
tween two fragments and their associated deformation den-
sities (D1orb). Figure 3 shows the main deformation densities
that were associated with the strongest orbital-interaction
energies (DEorb) that had some s- (NOCV 1–3, donation to
the arene-bonded metal) and p-character (NOCV 4, back-
donation from the p-bonded metal). The remaining residual
orbital-stabilization energy was about 30 kcalmolꢀ1. The de-
formation-density plots that were associated with NOCV1
and NOCV2 showed that the NMe2 group, the chelated Ir
center, and the pyridyl moiety all contributed in transferring
+
2.398 1.371 2.018 2.387 116.1
2.390 1.366 2.013 2.392 116.3
2.396 1.362 2.021 2.422 115.6
2.407 1.375 2.025 2.391 117.3
2.398 1.370 2.021 2.381 117.5
78.7
78.6
78.8
78.7
78.8
–
+
+
[b]
[b]
[f]
exo-IVbCp
exo-Vb
exo-Vb
DMA[g]
DMA
[b]
[h]
–
–
1.398
1.401
–
–
–
–
117.4
117.5
–
[a] Geometric parameters defined in Scheme 5; [b] (ZORA) B-P/AE-
TZP; [c] PBE-D3/TZP(Ir.4d) DZP(C.1 s, N.1 s, Cl.2p, O.1 s);
[d] (ZORA) PBE/AE-TZP(Ir) AE-DZP(H,C,Cl,Cl,N,O); [e] COSMO
(acetone) (ZORA) PBE/AE-TZP(Ir) AE-DZP(H,C,Cl,Cl,N,O);
AHCTUNGTRENNUNG
[f] (ZORA) B-P/AE-TZP (metal) AE-DZP(H,C,Cl,Cl,N,O); [g] N,N-
dimethylaniline; [h] TPSS/AE-TZP.
Structure and bonding: DFT methods were used to further
investigate the energetics in the p-bonding of [Cp*Ir]2+. This
investigation was performed by using gas-phase optimized
2+
models of the fictitious structure 3a2+, i.e., endo-IIIaCp*
,
2+
and of structures 3b2+ and 3d2+, i.e., endo-IIIbCp* and
Chem. Eur. J. 2012, 18, 6063 – 6078
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6067

J.-P. Djukic et al.
Figure 3. NOCV deformation densities (D1orb
) and their stabilization
energies (DEorb, in kcalmolꢀ1) for endo-IIIbCp*2+; NOCV complementary
pairs have been omitted for clarity. Light-gray and dark-gray areas show
isosurfaces where charge-density depletion and charge build-up occur, re-
spectively; that is, electron-density transfer operates from light-gray to
dark-gray areas upon bonding. The contour of the deformation-density
isosurface was set to 0.005 e/bohr3.
Figure 5. NOCV deformation densities (D1orb
)
and their stabilization
energies (DEorb, in kcalmolꢀ1) for the interaction between the closed-
shell metallacycle and the p-bonded metal; NOCV complementary pairs
are omitted for clarity. Light-gray and dark-gray areas show isosurfaces
where charge-density depletion and charge build-up occur, respectively.
The contour of the deformation-density isosurface was set to 0.005
e/bohr3.
electron density into the bonding interaction with the
[Cp*Ir] moiety.
A comparative analysis was applied to models of com-
pounds 3b2+, 4b+, 4c+, and 5b, in which the Cp* group was
replaced by Cp and the tBu group by H to reduce computa-
tion time. The corresponding model compounds, i.e., endo-
IIIbCp2+, endo-IVbCp+, and exo-VaCp (Figure 4), were opti-
equilibrium between density transfer from the arene to the
metal and the back-donation from the metal to arene ligand
than in the [Ir2] system. Furthermore, electron-density trans-
fer from the pyridyl p-system seemed to be nonexistent in
2+
exo-VaCp and rather common in endo-IIIbCp and endo-
IVbCp+, thus suggesting that the [CpRu]+ moiety exerted
a slightly larger electron-withdrawing effect than Cr(CO)3.
Therefore, the following ranking by electron-withdrawing
character was proposed: [CpIr]2+ @[CpRu]+ >Cr(CO)3.
On the stereospecificity of the formation of endo-3b2+ and
endo-4b+: In a previous report on the stereoselective synthe-
sis of other analogues of exo-5b, it was shown that coulom-
bic repulsive interactions greatly disfavored the formation
of isomers in which the chlorido ligand would be endo with
respect to the Cr(CO)3 group.[25] In the cationic compounds
discussed herein, intuitively, under thermodynamic control,
a molecular pocket of low charge density that encapsulates
a chlorido ligand with high charge density, such as in com-
pounds 3b2+ and 4b–c2+, should be a more-stable construc-
tion than an alternative structure in which areas of high-
and low charge density would be separated from each other,
like in a hypothetical exo isomer. Close scrutiny of the natu-
ral charges computed within the NBO scheme provided
Figure 4. Singlet-ground-state gas-phase geometry of exo-VaCp at the
B-P/AE-TZP level.
mized at the (ZORA) B-P/all-electron (AE) TZP level and
analyzed by using the ETS-NOCV scheme. Figure 5 shows
the deformation-density (D1orb) plots for the NOCVs that
had the largest orbital-stabilization energy. The D1orb plots
2+
for endo-IIIbCp matched well with those with a different
level of theory (Figure 3). Lower steric interactions, which
resulted from the replacement of Cp* by Cp, increases the
stabilization of NOCV1–2 by about 15 kcalmolꢀ1, whereas
NOCV3–4 remained mostly unchanged. The p-arene–metal
interaction energies (DEint) for the three models ((ZORA)
B-P/AE-TZP) were: endo-IIIbCp2+, ꢀ222.2 kcalmolꢀ1; endo-
IVbCp+, ꢀ113.6 kcalmolꢀ1; and exo-VaCp, ꢀ66.3 kcalmolꢀ1.
2+
some support for this intuitive picture. In both endo-IIIbCp
+
(qp-Ir =+0.58) and endo-IVbCp (qp-Ru =+0.29), the natural
charge at the p-bound metal was significantly positive,
which favored stabilizing coulombic interactions with the vi-
cinal Ir-bound chlorido ligand (qCl ꢁꢀ0.44). Computations
suggested a bias in favor of the endo isomer; the exo iso-
mers of the [Ir,Ru] and the [Ir2] models in the gas phase at
the (ZORA) B-P/AE-TZP level were higher in energy than
the endo isomers by 6.6 and 3.2 kcalmolꢀ1, respectively.
+
The D1orb plots for models endo-IVbCp and exo-VaCp sug-
gested a much-lower contribution of the NMe2 group in
terms of electron-density donation and a more-balanced
6068
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 6063 – 6078

Synthesis of Planar Chiral Iridacycles
FULL PAPER
EDA provided additional information on the effects that fa-
vored the formation of the endo isomers.
than that of the endo isomer, the a angle was 1108 and the
distance (dM) was about 0.15 ꢂ longer than in the endo
isomer. The difference in deformation energy was less than
1 kcalmolꢀ1 in the [Ir,Ru] system. However, the bias in
favor of the endo system was ensured by the slightly larger
interaction energy.
EDA was carried out on the exo and endo isomers of
2+
+
IIIbCp and IVbCp (Figure 6) by considering the interac-
tions of the [CpIr]2+ and [CpRu]+ cations with iridacycle
IIbCp. The deformation and interaction energies are given in
Replacement of Cp by Cp* showed an even more marked
difference between the endo and exo isomers (see the Sup-
porting Information). This result was clearly observed for
2+
endo- and exo-IIIbCp* at the (ZORA) PBE/all-electron
TZP(Ir), all-electron DZP(remaining elements) level. In this
case, the deformation energy for the interaction between
structure IIbCp* and [Cp*Ir]2+ was 10 kcalmolꢀ1 larger for
the exo system (+27.3 kcalmolꢀ1) than the endo system
(+17.4 kcalmolꢀ1); the interaction energies of the prepared
fragments were basically the same for the exo (ꢀ182.6 kcal
molꢀ1) and endo systems (ꢀ181.6 kcalmolꢀ1).
Deconvolution of the formation energy (DEform) for the
2+
reaction IIbCp*+ACTHNGUTERNNUG
Figure 6. Singlet-ground-state gas-phase geometries of endo-IIIbCp2+, exo-
IIIbCp
level.
endo paths into orbital- (DEorb), repulsive Pauli- (DEPauli),
and attractive electrostatic-energy terms (DEES) indicated
that, in spite of a largely more-favorable DEorb term for the
exo system, the large Pauli repulsive term, which was insuffi-
ciently compensated for by a low electrostatic term, contrib-
2+
+
,
endo-IVbCp+, and exo-IVbCp at the (ZORA) B-P/AE-TZP
2+
Figure 7. In the [Ir2] system, the difference in deformation
energy between the endo and exo systems (about 6 kcal
uted to make the exo-IIIbCp* isomer less-stable than the
endo isomer by 8.9 kcalmolꢀ1. In other words, steric interac-
tions greatly disfavored the exo isomer.
The possible formation of exo isomers of either com-
pounds 3b2+ or 4b–4c+ as transient products in the early
stage of the p-coordination reaction could not be fully ruled
out. An exo–endo isomerization was conceivable in the case
of two competing p-coordination pathways (Scheme 6).
Polar solvents, such as MeOH, may promote the dissociation
ꢀ
of the Ir Cl bond at room temperature and allow, for exam-
ple, the reactions with alkynes.[26] To evaluate the possibility
of forming the exo analogue of compound 3b2+, we turned
to triscation 6b3+ as the most reasonable substrate to at-
tempt the controlled formation of compound exo-3b2+ by
“re-chloridation”.
Complex 6b3+ was readily formed by abstraction of the
chlorido ligand with AgPF6 (reaction (3)). The reaction re-
quired the use of a fresh sample of anhydrous silver salt in
completely anhydrous acetone. However, complex 6bACHTUNGTRENNUNG[PF6]3
could not be isolated as a pure product. Every attempt to
precipitate it led to rapid decomposition and the formation
Figure 7. a) Energy-decomposition analysis with the corresponding de-
formation and interaction energies (in kcalmolꢀ1). b) Geometries of the
“relaxed” singlet-ground-state fragments IIbCp, [CpIr]2+, and [CpRu]+ op-
timized at the B-P/AE-TZP level.
molꢀ1) indicated greater steric
constraints in the exo system. In
other words, the exo system re-
quired more structural changes
to accommodate the positioning
of the Cp ligand that was
bound to the chelated Ir in an
endo
fashion
to
the
p-bonded metal moiety: for in-
stance, the geometry of exo-
2+
IIIbCp
was more distorted Scheme 6. Possible path for the formation of endo-3b2+
.
Chem. Eur. J. 2012, 18, 6063 – 6078
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6069

J.-P. Djukic et al.
of an insoluble black material. We suspected that this insta-
bility was related to the increased reactivity of the chelated
and weakly solvated Ir center. However, slow diffusion of
Figure 9. 1H NMR spectroscopy of the reaction of salt 6a3+ with
[nBu4N]Cl in [D6]acetone at room temperature. The addition of the
chloride salt was done without stirring to limit the rate of the reaction to
the rate of diffusion. At t=3 min, a spectrum was recorded; the NMR
sample tube was subsequently shaken and a new spectrum was recorded
after t=18 min, which showed almost complete consumption of com-
a solution of compound 6bACHTNUGTRNEUGN[PF6]3 in acetone into a solution
pound 6a3+ and the formation of endo-3a2+
.
solvents, which tended to promote its isomerization into the
endo isomer.
Another explanation of the peculiar facial selectivity ob-
served for the reactions of the electrophilic [Cp*Ir-
AHCTUNGTRENNUNG
(acetone)3]2+ could be the assistance of the chlorido ligand
that was bonded to the chelated Ir center. In that case, the
formation of the m-chlorido bridged intermediates would
constitute a key step in the p-coordination reaction. Compu-
tational investigations were carried out by using the disper-
sion-corrected DFT-D3[27] method PBE-D3 to evaluate the
pertinence of such a pathway that involved the direct assis-
tance of the p-coordination of a model CpIr2+ moiety by the
chlorido ligand of compound IIbCp (see the Supporting In-
formation). The resulting gas-phase reaction profile showed
that, even though the elementary steps that implied solvent
de-coordination did not possess high barriers of activation,
the whole process suffered from significant endothermic
character (large enthalpy of reaction) that did not match
with experimental observations.
Figure 8. ORTEP-type structure of [6b·OH2·ACHTNUTRGNE(UNG PF6)3]; ellipsoids set at 30%
ꢀ
probability; disordered PF6 molecules and solvent molecules were omit-
ted for clarity. Selected distances [ꢂ] and angles [8]: Ir(1)–O(1W)
2.217(4), N(2)–C(3) 1.33(1), Ir(1)–N(1) 2.105(6), Ir(1)–C(1) 2.058(6),
Ir(2)–C(3) 2.439(7); C(1)-Ir(1)-N(1) 79.0(3).
of n-heptane at ꢀ208C over 5 months led to the structure of
aquo-complex [6b·OH2·ACHTNUTRGNE(NUG PF6)3] (Figure 8), wherein a mole-
cule of water was bonded to the chelated Ir center in an
endo fashion with respect to the p-bound Cp*Ir moiety.
The treatment of a fresh solution of compound 6b3+ with
a slight excess of a solution of nBu4NCl in deuterated ace-
The dimethylamino group, a reservoir of electron density:
The NMe2 group is a mesomer-inductive electron-donating
group. It is typical to represent this donation by resonance
structures in which the donation is represented by mesomer-
ism, i.e., by the conjugation of the nitrogen lone pair with
the neighboring aromatic system.
This simplified and somewhat intuitive Lewis-type repre-
sentation could not account for the apparent lack of diaster-
eotopicity of the Me groups that was observed for planar
chiral N,N-dimethylaniline complexes and for complexes
1
tone led, according to H NMR analysis, to the formation of
a minor species that was suspected to be exo-3b2+, which
disappeared within seconds upon vigorous shaking of the
NMR sample tube to produce a new mixture that contained
endo-3b2+. To increase the lifetime of exo-3b2+, a similar ex-
periment was carried out in a less-polar solvent, i.e.,
CH2Cl2. Therefore, a solution of compound 6b3+ was treated
with an equimolar amount of nBu4NCl at ꢀ788C; acetone
was removed under vacuum at low temperature and the res-
idue was swiftly re-dissolved in dry deuterated CH2Cl2. In
4b–4cACTHNUTRGNENU[G PF6] and 5a–5b, because the situations at the zwitter-
ionic-limit forms are unrealistic (Scheme 7). Natural popula-
tion analysis within the NBO scheme provides so-called
“natural charges” that are directly correlated to the charge
density around a given atom. Table 4 lists the natural charg-
es on the N atom of the NMe2 group (qN). The qN value de-
creased in absolute terms by 0.06 units and the Wiberg bond
1
this case, H NMR spectroscopy indicated that the mixture
contained 60% of endo-3b2+ and approximately 40% of
a labile compound that was assumed to be exo-3b2+, which
vanished rapidly overnight to be replaced by endo-3b2+
(Figure 9). These experiments demonstrated that if exo-3b2+
could be formed its existence was compromised by polar
ꢀ
index for the CAr
(0.2 units) on going from IIbCp to endo-IIIbCp (values for
N bond (wbiN) increased slightly
2+
6070
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 6063 – 6078

Synthesis of Planar Chiral Iridacycles
FULL PAPER
Scheme 7. Resonance forms for N,N-dimethylaniline.
ꢀ
Table 4. Wiberg bond indices (wbiN) for the CAr N bond, natural charges
(qN) at the nitrogen atom, and the natural valence populations (zN) at the
lone pair (LP) of the amine nitrogen atom for selected model structures.
Compound
wbiN
qN
zN (LP)
DMA[a,b]
IIbCp
exo-VbCp
endo-IVbCp
endo-IIIbCp
1.13
1.15
1.18
1.19
1.35
ꢀ0.311
ꢀ0.398
ꢀ0.394
ꢀ0.390
ꢀ0.339
1.706
1.665
1.667
1.654
1.538
[c]
[c]
+[c]
2+[c]
Figure 10. Plots of the deformation density (D1orb) for the formation of
[a] DMA: N,N-dimethylaniline; [b] singlet-ground-state gas-phase B-P/
AE-TZP; [c] singlet-ground-state gas-phase (ZORA) B-P/AE-TZP-
ACHTUNGTRENNUNG(metals), AE-DZP(H,C,N,Cl,O).
the CAr NMe2 bond in endo-IIIbCp*2+, endo-IVbCp+, and exo-VaCp; isosur-
ꢀ
face contours were set to 0.005 for D1orb and 0.002 e/bohr3 for D1orb
.
p
s
Light-gray and dark-gray areas show charge-density depletion and charge
build-up on the isosurfaces, respectively.
N,N-dimethylaniline are shown for comparison). The corre-
lation between the natural charge that was assigned to the
N atom and the Wiberg bond index (wbiN) within the series
of models considered herein is evident. The underlying fate
of the lone pair on the nitrogen atom is even more obvious
when considering the electron population at the nitrogen
lone pair (zN), which was computed for the Lewis structure
that was determined by NBO analysis. From these data, it
appears that the most-direct response to a depletion of elec-
tron density at the arene ligand is the transfer of electron
density from the nitrogen lone pair.
creased in the following order: IIbCp (8.0% of DEorbtot)<
+
2+
exo-VaCp
(8.4%)<endo-IVbCp
(9.0%)<endo-IIIbCp
2+
(10.5%)ꢁendo-IIIbCp* (9.6%).
This relatively small contribution of the p-bond in the
model of structure 3b2+ was obviously sufficient to slacken
ꢀ
the rotation of the NMe2 group around the CAr N axis.
Line-shape analysis[28] of the rotational exchange between
1
the Mea and Meb groups was carried out by using H NMR
spectroscopy at various temperatures with 9.3 mm solutions
ꢀ
ETS-NOCV analysis of the bonding of the CAr NMe2
group to the aryl moiety provided a precise picture of the
of compound 3bACHTUNGTERNU[NG PF6]2 in [D6]acetone; spectra were acquired
at every 10 K in the range 303–343 K (Figure 11). Varying
channels through which the transfer of density occurred.
the ionic strength from I=27 mm (native solution of com-
ꢀ
The decomposition scheme for the analysis of the CAr
bond entailed the interaction between open-shell fragments,
N
pound 3b
amounts of an electrolyte, i.e., [nBu4N]
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
C
C
that is, Ar and NMe2 radicals (with excess a-spin occupa-
the slope of the Eyring curve (Figure 11). This result sug-
gested that the ion pair was dissociated under these experi-
mental conditions. The resulting Gibbs barrier to rotation at
C
C
tion assigned to Ar and excess b-spin occupation to NMe2),
to form the CAr N bond. Full ETS-NOCV analysis of
N,N-dimethylaniline (noted DMA) at the TPSS/AE-TZP
level showed that the donation from the nitrogen p orbital
ꢀ
¼
¼
298 K, i.e., DG rotꢀ298, was +16.7(3) kcalmolꢀ1, with DH
rot
¼
and DS
values of 12.9(4) kcalmolꢀ1 and ꢀ12(1)
rot
occurred mostly by transfer of the density towards a C N
calmolꢀ1 K, respectively (Figure 11). This value of DG
¼
ꢀ
rotꢀ298
p-bond (see the Supporting Information); this result corre-
was compared to computed values for models IIbCp, endo-
sponds to an orbital-stabilization energy (DEorb
)
p
of
IIIbCp2+, endo-IVbCp+, and exo-VaCp, as well as for endo-
ꢀ30.0 kcalmolꢀ1, which is about 7.7% of the total sum of
IIIbCp*
for which solvation was accounted for with
2+
the orbital-stabilization energies (SDEorb =ꢀ389.6 kcal
COSMO. The value of +16.8 kcalmolꢀ1, which was obtained
by averaging the values of DG
molꢀ1). The CAr N s-bond contribution to the stabilization
and DG
¼
¼
ꢀ
rot1ꢀ298 rot1ꢀ298
energy (DEorbs) is ꢀ341.2 kcalmolꢀ1, which is about 87.5%
(Table 5), is in excellent accord with the experimental data.
¼
¼
ꢀ
of the total stabilization energy for the CAr NMe2 bond in
Rotational barriers DG
and DG
were com-
rot1ꢀ298
rot2ꢀ298
DMA.
puted by determining the energy difference between the
ground-state geometry, in which the NMe2 group was almost
co-planar with the arene ring, and the two possible rotation-
al transition states, for example, rot-TS1 and rot-TS2
(Figure 12), in which the NMe2 group had been rotated by
about 908. We considered two rotational transition states be-
cause of the nearly equal probability for the NMe2 group to
The same analysis was performed on models IIbCp, endo-
2+
+
2+
IIIbCp
,
endo-IVbCp
,
exo-VaCp, and endo-IIIbCp* .
Figure 10 shows plots of D1orb and D1orb for the CAr NMe2
ꢀ
s
p
bond of models endo-IIIbCp*2+, endo-IVbCp+, and exo-VaCp,
which were very similar to those computed for DMA. The
p-contribution to the C N bond as determined by D1
orb
p
ꢀ
in-
Chem. Eur. J. 2012, 18, 6063 – 6078
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6071

J.-P. Djukic et al.
(Figure 12). Figure 13 shows the optimized structures of se-
lected rotational transition states in which the pyramidaliza-
tion of the NMe2 group, which was slight in the related
ground states, was complete in the rotational transition
Figure 11. Solution 1H NMR spectroscopy (500 MHz) line-shape analysis
(DNMR3[29]) of the mutual-exchange (equipopulation) rotation dynam-
ics[28] of the NMe2 substituent in compound 3a
ACTHUNGTRNEUNG[PF6]2 (9.3 mm in
[D6]acetone). Top: Eyring curve that shows the measurements at three
different ionic strengths (I=27, 37, and 121 mm) in the presence of
known amounts of [nBu4N]ACHTNUGTRNEUNG[PF6]. Bottom: overlays of experimental
(I=27 mm) and simulated Lorentzian lines with the corresponding abso-
lute temperatures and exchange rates (ꢃ indicates solvent impurity).
Figure 13. Gas-phase rotational transition states at the (ZORA)
B-P/AE-TZPACTHUNRTGNEUNG(metal), AE-DZP level.
state. The main electronic con-
sequence of the 908 rotation of
the NMe2 group was an in-
crease in zN, a decrease in wbiN,
Table 5. Rotational Gibbs free rotational barriers for the NMe2 group, natural population analysis (NPA), and
ꢀ
Wiberg bond indices for the CAr NMe2 bond.
rot-TS1
rot-TS2
NPA for rot-TS1
zN (LP)
¼
¼
Model
DG
[kcalmolꢀ1
]
DG rot2ꢀ298 [kcalmolꢀ1
]
qN
wbiN
rot1ꢀ298
and
(Table 5).
a
net increase in qN
DMA[a]
IIbCp
endo-IIIbCp
endo-IVbCp
exo-VaCp
+5.7
+6.6
+21.2
+11.8
+10.6
+17.8
–
ꢀ0.405
ꢀ0.498
ꢀ0.498
ꢀ0.493
ꢀ0.494
ꢀ0.496
1.87
1.86
1.85
1.85
1.85
1.85
0.99
1.02
1.02
0.99
0.99
1.00
[b]
+6.6
2+[b]
2+[b]
+20.9
+11.5
+12.1
+15.8
Planar chirality at a tertiary
amine? Constitutional and dy-
namic aspects: The rotational
[b]
2+[c]
endo-IIIbCp*
[a] Singlet ground state and rotational transition states at the TPSS/AE-TZP level; [b] singlet ground state and
rotational transition states at the (ZORA) B-P/AE-TZP(metal),DZP(all non-metal elements) level; [c] singlet
barrier for compound 3bACHTUNGTRENNUNG[PF6]2
AHCTUNGTRENNUNG
in acetone was of the same
order as that of the NMe2
group in DMF.[31] In that case,
ground state and rotational transition at the (COSMO) ZORA-PBE/AE-TZP(Ir),AE-DZP(all non-metal
elements) level.
¼
a rotational barrier (DGrot ) of
about 17 kcalmolꢀ1 was due to p-conjugation with the neigh-
boring carbonyl group and to Pauli-type repulsive interac-
tions between the lone pair of the pyramidalized nitrogen
atom and the carbonyl backbone. ETS-NOCV analysis sug-
gested that the p-contribution remained surprisingly low,
but not negligible, because a reasonable amount of electron
population was transferred from the lone pair to the rest of
the molecule on relaxing the rotational transition state to its
ground state.
These observations led us to evaluate the possibility of
creating a configurationally stable stereogenic center at the
nitrogen atom. Reaching a persistent and stable configura-
tion entailed a long-enough life-time to allow for the physi-
cal separation of two possible diastereomers without epimer-
ization taking place. Because the pyramidalization of nitro-
gen was negligible or inexistent from inspection of the three
Figure 12. Newman-type projections of the conformational transition
states rot-TS1 and rot-TS2 and the ground states (GS).
rehybridize into a simili sp3 configuration (pyramidaliza-
tion)[30] by sending the lone pair of electrons on the N atom
towards either the chelated Ir center or outwards
6072
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 6063 – 6078

Synthesis of Planar Chiral Iridacycles
FULL PAPER
structures by X-ray diffraction
of compounds 3b[PF6]2, 3e-
[PF6]2, and 6b[PF6]3, we pro-
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
posed that Schlçglꢄs nomencla-
ture for planar chiral metallo-
cenes applied here for this pe-
culiar case of the tertiary aro-
matic amine (Scheme 8).
Scheme 10. Formation of structures 3e-aACTHNUTRGENN[UG PF6]2 and 3e-bAHCUTNGTREN[NUGN PF6]2.
In the case of compound
3a
[PF6]2, the quarter-lifetime
the dynamic diastereomers, that is, a singlet at d=3.50 (3e-
[PF6]2) and 3.70 ppm (3e-a[PF6]2). The ethyl group gave
rise to a triplet at d=1.44 (CH3CH2N, 3e-b[PF6]2) and
multiplet at d=
(t1/4), which corresponds to the
time required for a quarter of
a rotamer to be converted into
the other rotamer (in a fictitious
situation wherein the methyl
groups were differentiated by
labeling, i.e., Mea and Meb) is
a small fraction of a second,
b
G
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
Scheme 8. Schlçglꢄs formalism
applied to the determination
of the configuration of the
planar chiral tertiary aromatic
amine.
1.40 ppm (CH3CH2N, 3e-a
3.97 ppm (CH3CH2N,
ACHTUGNETNR[NUNG PF6]2), a
3e-a
ACHTUNGTRENNUNG
(CH3-CH2-N, 3e-bAHCTUNGTRENNUNG
As expected, all attempts to separate these diastereomers by
fractional recrystallization on large amounts of mixture
failed. However, slow diffusion of a concentrated solution of
the mixture in n-heptane led to the formation of a set of
crystals, among which, we revealed the exclusive formation
which is long enough to observe diastereotopism in the
1H NMR spectrum. However, the rotation of the NMe2
group is still effective. An increase of the rotational barrier,
which is necessary to create a conformationally persistent
chiral amino group, could be achieved only by increasing
the electron donation from the nitrogen atom and by intro-
ducing larger steric constraints. This result was achievable
with large and bulky alkyl substituents.
of isomer 3e-aACTHNUTRGNEUNG[PF6]2 (Figure 14), with no sensible positional
disorder at the N(Et)(Me) group.
In
a preliminary approach, we first considered the
N(Me)(Et) group as a readily accessible expedient to the
necessary constitutional asymmetry. Ligand 1e was synthe-
sized from the reaction of 4-N-ethyl,N-methyl aminophenyl
lithium (prepared from 4-N-ethyl,N-methyl bromobenzene)
with pyridine and the subsequent thermolysis of the result-
ing adduct into the expected ligand and LiH (Scheme 9).
Figure 14. ORTEP-type structure of complex (SIr*,pR*,pSN*)-3e-aACHTUNGTRENNUNG[PF6]2;
ellipsoids set at 30% probability; H atoms, counteranions, and solvent
molecules have been omitted for clarity. Selected distances [ꢂ] and
ꢀ
ꢀ
ꢀ
ꢀ
angles [8]: N1 Ir1 2.087(6), Ir1 Cl1 2.370(2), Ir1 C1 2.006(7), C3 N2
ꢀ
1.33(1), Ir2 C3 2.461(7); C2-C3-C4 116.2(6), N1-Ir1-C1 78.2(3).
Scheme 9. Synthesis of compound 2e; i: cyclohexane, reflux; ii:
[(Cp*IrCl2)2], NaOAc, CH2Cl2, room temperature.
DFT computations were carried out on models of
2+
[PF6]2, i.e., IIIe-aCp* and IIIe-bCp*2+, re-
3e-aACTHNUTGRENN[UG PF6]2 and 3e-bHCATUNGTRENNGUN
Complex 2e was prepared by using the method reported
spectively (Figure 15). By accounting for solvation in ace-
tone (COSMO), these computations indicated that, for the
by Davies et al.[14] The reaction of compound 2e with
ꢀ
2+
[Cp*Ir
N
conformational equilibrium IIIe-bCp* $IIIe-aCp*2+, the com-
complexes, 3e-a[PF6]2 and 3e-b
E
a
3:2 ratio
puted value for DG298K (ꢀ0.89 kcalmolꢀ1) was due to a bias
2+
(Scheme 10). In this case, at room temperature in deuterat-
ed acetone, the signals owing to the Me and Et groups of
N(Me)(Et) were distinctly resolved. In the 1H NMR spec-
trum of the aromatic region, most of the signals of the two
species were unresolved, except those signals that were as-
signed to the p-bound arene protons. The Me group at the
nitrogen center gave rise to two different sets of signals for
in favor of IIIe-aCp* thanks to a favorable entropic contri-
bution (DH298K =ꢀ0.5 kcalmolꢀ1, DS298K =1.3 calmolꢀ1 K). It
was tempting to propose that the isolation of pure
3e-aACTHNUTGRNEUGN[PF6]2 was due to a displacement of the above-men-
tioned equilibrium by crystal formation. This displacement
would mean that the dissolution of a crystal of pure
3e-aACTHNUTGRNEUGN[PF6]2 would result in a rapid re-equilibration, thereby
Chem. Eur. J. 2012, 18, 6063 – 6078
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6073

J.-P. Djukic et al.
Figure 15. Selected singlet-ground-state geometries of IIIe-a/b2+ and rota-
tional transition states IIIe-TS1/22+ at the (COSMO) ZORA-PBE/AE-
TZP(Ir), AE-DZP(remaining elements) level.
leading to an approximate 3:2 mixture of the two diastereo-
mers. DFT computations provided conclusive support of this
assumption.
The rotation of the N(Me)(Et) groups may lead to four
rotational transition states of similar energy. Only two states
were considered for the determination of the rotational bar-
Figure 16. Line-shape analysis of the non-mutual chemical exchange in
ꢀ
solution between the double PF6 salts of 3e-a2+ and 3e-b2+ at T=293–
343 K: a) line shape of the signals of the protons of the Me groups in
structures 3e-a2+ (peak with the largest intensity) and 3e-b2+, measured
by 1H NMR spectroscopy at 500 MHz with a 1.05 mm solution of com-
¼
riers (DGrot ; Figure 15): IIIe-TS1 and IIIe-TS2. The com-
puted rotational barriers were calculated for the ascension
pound 3eACHTNUGTRNEGNU[PF6]2; upper and lower lines corresponded to simulated and ex-
¼
perimental spectra, with an average T2 relaxation constant of 0.5 s for
both Me groups. Corrections to the population ratios in structures 3e-a2+
and 3e-b2+ were taken into consideration in the computation of the ex-
change rate krot(3e-b2+!3e-a2+) because the ratio varied sensibly from
60.7:39.3 (293 K) to 59.4:40.6 (303–323 K) and 57.0:42.9 (333 and 343 K).
from IIIe-a to IIIe-TS1(2): DGrotꢀ298K (IIIe-TS1)=19.7 kcal
¼
molꢀ1, DGrotꢀ298K (IIIe-TS2)=19.4 kcalmolꢀ1. These values
indicated that the substitution of one Me group for an Et
group induced an increase of about 3 kcalmolꢀ1 in the barri-
er to rotation.
b) Eyring curve for the conversion of 3e-b2+ into 3e-a2+
, thereby
¼
¼
leading to DH =12.9(5) kcalmolꢀ1 and DS =ꢀ12.1(8) calmolꢀ1 K, and
1
Figure 16 shows the line-shapes of the H NMR resonan-
DG 298 =16.5(5) kcalmolꢀ1 after a linear fit (R2 =99.5%).
¼
ces of the Me groups of diastereomers 3e–b2+ and 3e–a2+ at
different temperatures with the associated Eyring plot. The
experimentally determined rotational enthalpy barriers were
very close to that determined for compound 3b2+. If discrim-
ination between the two diastereomers of compound 3e2+
compensating for the depletion of electron density that re-
sulted from p-bonding. Theoretical investigations showed
that this compensation was operated by the transfer of elec-
tron density from the so-called lone pair on the nitrogen
atom to the arene ligand through two channels, for example,
1
was possible on the H NMR timescale, the prospect of pro-
longing the lifetime of either of these species to make them
isolable still remains far off: for instance, steric cluttering
may impede p-coordination. Nonetheless, these results lay
the foundations for further investigation of the peculiar con-
formational features that were introduced by the p-bound
IrCp* moiety, for which the slight depletion of electron den-
sity at the arene caused by this electron-avid moiety could
induce physically sensible and somewhat unusual perturba-
tions of the conformational behavior of a NR2 group.
ꢀ
ꢀ
the s N CAr bond and the weak p N CAr bond. Conforma-
tional diastereotopicity of the NMe2 group, i.e., the
1H NMR-timescale-detectable discrimination of the two Me
groups owing to slackened rotation, was the main conse-
quence of the electronic compensation by this donor group.
This entirely electronic effect, which was first raised by
Maitlis and co-workers[11a] for the related complexes of ani-
line, lays the foundation for further research towards a new
class of chiral compounds with persistent planar chiral ter-
tiary amines. We have also shown that the p-coordination of
cationic metal moieties of the MCp*-type occurred with
strong facial discrimination. The origin of this facial selectiv-
ity lay in coulombic interactions and steric relaxation that
tended to favor p-coordination on the side that displayed
the largest charge density. These findings will be essential
for further development in the chemistry of planar chiral
metallacycles. The underlying new rational emerging here is
important for the stereoselective synthesis of non-square-
Conclusion
Herein, we have provided evidence of the capability of
cycloiridated 2-phenylpyridines to undergo p-coordination
with cationic metal moieties that contained one or two posi-
tive charges. The syntheses were relatively efficient provided
that a donor substituent, such as NMe2, was present on the
phenylene group of the iridacycle. This electron-donating
group contributed to consolidating the p–metal bond by
6074
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 6063 – 6078

Synthesis of Planar Chiral Iridacycles
FULL PAPER
spectra were acquired with a FTIR Bruker alpha spectrometer by using
planar [C,N]–metal heterochelates; it is based on the opti-
mal engineering of intramolecular coulombic interactions.
[15]
an ATR solid-state sample cell. [Cp*Ir
(CH3COCH3)3]
E
and
[Cp*RuACHTNUGTRNEUNG
(NCCH3)3] [PF6][16] were prepared according to literature proce-
dures. The former was dissolved in acetone shortly after its preparation
from a reaction of [(Cp*IrCl2)2][13] with AgPF6 and used as a solution. Tri-
carbonyl(h6-naphthalene)chromium was either purchased from Aldrich
or synthesized according to the procedure proposed by Uemura et al.[17]
Experimental Section
Synthesis of 4-tert-butyl,2-(4-N,N-dimethylaminophenyl)pyridine (1c): A
solution of 4-lithio,N,N-dimethylaniline (57.5 mmol, prepared from the
reaction of 4-bromo,N,N-dimethylaniline with excess Li metal) in Et2O
(125 mL) was added to dry 4-tert-butylpyridine (7.75 g, 57.5 mmol) at RT
and the solution was left to stir for about 7 h. The resulting mixture was
evaporated to dryness and the residue was suspended in dry cyclohexane
(70 mL) and boiled overnight. The resulting suspension was hydrolyzed
with water and the mixture was extracted with CH2Cl2 according to a con-
ventional workup procedure. The organic phase was dried over MgSO4,
filtered through celite, and the solvent was removed in vacuo. The resi-
due was purified by column chromatography on silica gel (CH2Cl2/n-pen-
tane, 30:70). Compound 1c was recovered as off-white crystals upon con-
centration of the eluate and crystallization from pentane (8.7 g, 60%).
1H NMR (300 MHz, CDCl3): d=1.35 (s, 9H; tBu), 3.02 (s, 6H; NMe2),
6.80 (d, 2H, 3J=8.9 Hz), 7.11 (dd, 1H, 4J=1.9, 3J=5.3 Hz), 7.62 (d, 1H,
4J=2.1 Hz), 7.90 (d, 2H, 3J=8.9 Hz), 8.52 ppm (d, 1H, 3J=5.3 Hz);
13C NMR (75 MHz, CDCl3): d=159.9, 157.2, 150.5, 148.8, 127.4, 117.6,
115.9, 111.8, 40.0, 34.4, 30.2 ppm; elemental analysis calcd (%) for
C17H22N2: C 80.27, H 8.72, N 11.02; found: C 80.08, H 8.42, N 11.00.
Computational methods: Computations were performed by using DFT
methods using the Becke[32]–Perdew[33] (B-P) and the Perdew–Burke–
Ernzerhof[34] (PBE) GGA functionals, as well as the Tao–Perdew–Staro-
verov–Scuseria[35] (TPSS) meta-GGA functional that was implemented in
the Amsterdam Density Functional package[23,36] (ADF2010[37] version).
Within the B-P scheme, electron correlation was treated within the local
density approximation (LDA) in the Vosko–Wilk–Nusair parameteriza-
tion.[38] Within the PBE scheme, electron correlation was treated within
the local density approximation (LDA) in the PW92 parameterization.[39]
Scalar relativistic computations with the zeroth order regular approxima-
tion[40] were carried out by using ad hoc all-electron (AE) basis sets:[41]
polarized triple-z (TZP) and double-z (DZP) Slater-type orbitals were
used in this study; the choice of basis sets was determined by the purpose
of the calculations (for details, see the Tables and Figures). Owing to
technical limitations, NBO analysis was carried out on geometries of
models that were relaxed at either the (ZORA) B-P or (ZORA) PBE
levels by using a combination of AE-TZP basis sets for the metal atoms
and AE-DZP basis sets for the non-metal atoms. Geometry optimizations
by energy-gradient minimization were carried out in all cases with an in-
tegration grid of between 4.5 and 6.5, an energy gradient convergence
criterion of 10ꢀ3, and a tight SCF-convergence criterion. Every transition-
state (TS) search was preceded with an extensive scan of the potential
surface to identify the most-reasonable “candidate TS geometry”; this
task was achieved by using the so-called linear-transit procedure, which
forced stepwise incremental variations of the interatomic distances in the
direction of the expected change of atomic coordinates during the reac-
tion and relaxing each stationary geometry. For transition-state searches,
the integration grid was constrained between 5.5 and 6.5 with maximum
radial step values that were comprised of between 10ꢀ5 and 10ꢀ6 ꢂ. Com-
putation of the gas-phase reaction profile was carried out by taking ad-
vantage of the DFT-D3 approach reported by Grimme et al.[27a] All
ground-state- and transition-state geometries were computed by using
a “zero-damped” PBE-D3[27a] (ZORA) functional associated with frozen
core TZP basis sets for all elements except H. The orbital descriptor fol-
lowing the elementꢄs name indicates the upper orbital of the considered
“frozen core” approximation: C 1s, N 1s, Cl 2p, Ir 4d, O 1s.[41] The
COSMO treatment of solvation (with acetone) was applied by using the
procedure that was implemented within the ADF package with Klaamtꢄs
adjustment of the van der Waals radii of the atoms.[42] Counterpoise cor-
rection for basis set superposition error (BSSE) was neglected through-
out this study.[41] ETS-NOCV[24] analysis, as well as calculations of the vi-
brational modes (analytical second-derivative frequencies and two-point
numerical integration for COSMO geometries), were performed with op-
timized geometries by using ADF2010 subroutines. In all cases, the vibra-
tional modes were computed at 298.15 K to verify that the optimized ge-
ometries were either related to energy minima or to transition states:
statistical thermodynamic data were extracted for the further determina-
tion of enthalpies and variations in the Gibbs free energies by using con-
ventional methods. Natural atomic orbital (NAO) and natural bond orbi-
tal (NBO) analysis, as well as Wiberg indices determination, were per-
formed with the GENNBO 5.0 module of ADF. Representations of the
molecular structures and orbitals were drawn by using ADFview v10.
Synthesis of 2-(4-N-ethyl,N-methylaminophenyl)pyridine (1e): 4-
Lithio,N-ethyl,N-methylaniline (20.8 mmol), which was prepared in
a way similar to compound 1c, that is, from the reaction of 4-bromo,
N-ethyl,N-methylaniline (4.3 g, 20.8 mmol) with excess Li metal in anhy-
drous Et2O, and the subsequent addition to dry pyridine (1.58 g,
20.0 mmol). 1H NMR (300 MHz, CDCl3): d=1.16 (t, 3H, 3J=7.08 Hz),
2.97 (s, 3H; Me), 3.46 (q, 2H, 3J=7.1 Hz) 6.78 (d, 2H, 3J=8.7 Hz), 7.09
(ddd, 1H, 4J=1.9, 3J=4.8, 3J=6.6 Hz), 7.65 (m, 2H), 7.9 (d, 2H, 3J=
8.8 Hz), 8.61 ppm (dd, 1H, 4J=1.4, 3J=4.9 Hz); 13C NMR (75 MHz,
CDCl3): d=157.1, 149.9, 148.6, 137.4, 128.1, 120.5, 119.3, 112.1, 46.7, 37.5,
11.3 ppm; HRMS (ESI+): m/z calcd for C14H17N2: 213.1386; found:
213.1373.
General procedure for the cycloiridation of 2-phenylpyridines: A mixture
of [(Cp*IrCl2)2], NaOAc, and a 2-arylpyridine was stirred in CH2Cl2
(15 mL) at RT for 24 h. The product was typically recrystallized from
a mixture of CH2Cl2 and pentane after filtration of the reaction mixture.
Synthesis of complex 2b: The reaction of [(Cp*IrCl2)2] (500 mg,
0.6 mmol), NaOAc (307 mg, 3.70 mmol), and compound 1b (251 mg,
1.20 mmol) afforded cyclometalated compound 2b as a yellow–orange
solid in 80% yield upon recrystallization (559 mg). 1H NMR (500 MHz,
CDCl3): d=1.68 (s, 15H; C5Me5), 3.08 (s, 6H; NMe2), 6.45 (dd, 1H, 4J=
3
4
3
3
2.60, J=8.6 Hz), 6.86 (ddd,1H, J=1.5, J=5.7, J=7.2 Hz), 7.16 (d, 1H,
4J=2.5 Hz), 7.50 (m, 1H, 4J=1.3, 4J=2.6, 3J=7.5 Hz), 7.53 (d, 1H, 3J=
8.5 Hz), 7.58 (dd, 1H, 4J=1.41, 3J=8.12 Hz), 8.55 ppm (d, 1H, 3J=
5.8 Hz); 13C NMR (125 MHz, CDCl3): d=167.4, 164.7, 152.0, 150.7, 136.2,
132.9, 124.8, 119.6, 118.1, 117.3, 106.9, 88.0, 40.29, 8.8 ppm; elemental
analysis calcd (%) for C23H28ClIrN2·1/4CH2Cl2: C 48.03, H 4.94, N 4.82;
found: C 48.04, H 4.743, N 4.717.
Synthesis of complex 2c: The reaction of [(Cp*IrCl2)2] (796.0 mg,
1.0 mmol), NaOAc (197 mg, 2.4 mmol), and compound 1c (510.0 mg,
2.0 mmol) afforded cyclometalated compound 2c as a yellow solid in
80% yield upon recrystallization (986 mg). 1H NMR (500 MHz, CD2Cl2):
d=1.23 (s, 9H; tBu), 1.52 (s, 15H; C5Me5), 2.92 (s, 6H; NMe2), 6.30 (dd,
1H, 4J=2.6, 3J=8.6 Hz), 6.87 (dd, 1H, 4J=2.1, 3J=6.2 Hz), 6.99 (d, 1H,
4J=2.6 Hz), 7.44 (d, 1H, 3J=8.6 Hz), 7.50 (d, 1H, 4J=2.1 Hz), 8.33 ppm
(d, 1H, 3J=6.2 Hz); 13C NMR (125 MHz, CD2Cl2): d=167.17, 164.9,
160.9, 152.1, 150.8, 134.0, 124.8, 118.7, 118.0, 114.0, 106.8, 88.2, 40.2, 30.3,
8.7 ppm; elemental analysis calcd (%) for C27H36ClIrN2: C 52.6, H 5.89,
N 4.55; found: C 52.52, H 5.81, N 4.43.
General: All experiments were carried out under a dry argon atmosphere
using standard Schlenk techniques or in an argon-filled glove-box when
necessary. Anhydrous THF was distilled from purple solutions of Na/ben-
zophenone under an argon atmosphere. All other solvents were distilled
over sodium or CaH2 under an argon atmosphere. Deuterated solvents
were dried over sodium or CaH2 and purified by trap-to-trap techniques,
degassed by freeze-pump-thaw cycles, and stored under an argon atmos-
1
phere. H and 13C NMR spectra were obtained on Bruker DPX 300, 400,
Synthesis of complex 2d: The reaction of [(Cp*IrCl2)2] (325.13 mg,
0.40 mmol), NaOAc (197 mg, 2.40 mmol), and compound 1d (150 mg,
0.80 mmol) afforded cyclometalated compound 2d as a yellow solid in
or Avance 500 spectrometers. Chemical shifts (expressed in parts per mil-
lion) were referenced against solvent peaks or external references. IR
Chem. Eur. J. 2012, 18, 6063 – 6078
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6075

J.-P. Djukic et al.
80% yield upon recrystallization (350 mg). 1H NMR (500 MHz, CDCl3):
d=1.58 (s, 15H; C5Me5), 3.0 (s, 3H; OMe), 6.52 (dd, 1H, 4J=2.5, 3J=
8.5 Hz), 6.89 (ddd, 1H, 4J=1.4, 3J=5.8, 3J=7.2 Hz), 7.26 (d, 1H, 4J=
2.5 Hz), 7.51 (m, 2H, 4J=1.3, 3J=8.3 Hz), 7.58 (d, 1H, 3J=8.15 Hz),
8.52 ppm (d, 1H, 3J=5.8 Hz); 13C NMR (125 MHz, CDCl3): d=166.9,
165.4, 161.3, 151.1, 137.4, 136.8, 125.2, 119.9, 118.2, 108.6, 88.4, 55.1,
8.9 ppm; elemental analysis calcd (%) for C22H25IrClNO·1/2CH2Cl2:
C 45.84, H 4.44, N 2.38; found: C 45.55, H 4.49, N 2.4.
48.2, 46.8, 37.6 (C3eꢀa), 37.4 (C3eꢀb), 11.0 (C3eꢀb), 10.9 (C3eꢀa), 8.7 ppm; ele-
mental analysis calcd (%) for C34H45ClF12Ir2N2P2·CH2Cl2: C 32.93, H 3.71,
N 2.19; found: C 32.91, H 3.92, N 2.36.
General procedure for the synthesis of [Ir,Ru] complexes: A mixture of
iridacycle and [Cp*RuACHTUNRGTNE(UNG NCCH3)3]PF6 in THF was stirred at RT for 24 h
under an argon atmosphere. After a flash filtration of the solution
through celite, the filtrate was concentrated under vacuum and the result-
ing precipitate was recrystallized from pentane. The precipitate was
washed three times with pentane and evaporated to dryness under re-
duced pressure.
Synthesis of complex 2e: The reaction of [(Cp*IrCl2)2] (525 mg,
0.60 mmol), NaOAc (324 mg, 3.90 mmol), and compound 1e (280 mg,
1.30 mmol) afforded compound 2e as a yellow solid in 80% yield upon
recrystallization (605 mg). 1H NMR (500 MHz, CDCl3): d=1.19 (t, 3H,
3J=7.1 Hz), 1.68 (s, 15H; C5Me5), 3.04 (s, 3H; NMe2), 3.42 (m, 1H, 3J=
Synthesis of complex 4b
0.178 mmol) and [Cp*Ru
(10 mL) afforded compound 4bAHCTNUGTRENNNUG
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
3
4
3
7.3 Hz), 3.58 (m, 1H, J=7.3 Hz), 6.43 (dd, 1H, J=2.5, J=8.7 Hz), 6.84
yield (127 mg). 1H NMR (500 MHz, CDCl3): d=1.50 (s, 15H; C5Me5),
1.68 (s, 15H; C5Me5), 3.06 (s, 6H; NMe2), 5.34 (dd, 1H, 4J=2.0, 3J=
6.4 Hz), 5.73 (d, 1H, 4J=1.9 Hz), 6.41 (d, 1H, 3J=6.5 Hz), 7.26 (m, 1H,
4J=1.4, 3J=4.2 Hz), 7.83 (d, 1H, 3J=7.9 Hz), 7.9 (dt, 1H, 4J=1.6, 3J=
7.7 Hz), 8.54 ppm (d, 1H, 3J=5.5 Hz); 13C NMR (125 MHz, CDCl3): d=
163.6, 152.4, 151.0, 124.6, 119.3, 117.7, 97.1, 93.7, 88.7, 80.4, 69.2, 39.9,
(ddd, 1H, J=1.5, 3J=5.8, 3J=7.1 Hz), 7.13 (d, 1H, 4J=2.4 Hz), 7.53 (m,
4
3H, 4J=1.8, 3J=7.8, 3J=4.3 Hz), 8.54 ppm (dd, 1H, 4J=1.5, 3J=5.7 Hz);
13C NMR (125 MHz, CDCl3): d=167.5, 164.9, 150.8, 136.6, 132.6, 125.2,
119.6, 118.0, 117.3, 106.8, 88.1, 46.8, 37.5, 31.0, 12.0, 8.9; elemental analy-
sis calcd (%) for C24H30IrClN2·3/4C3H6O: C 51.04, H 5.33, N 4.83; found:
C 51.04, H 5.30, N 5.15.
31.0,
11.4,
8.7 ppm;
elemental
analysis
calcd
(%)
for
General procedure for the synthesis of [Ir2] complexes: A solution of the
iridacycle in acetone was added to a freshly prepared solution of [Cp*Ir-
ACHTUNGTRENNUNG(CH3COCH3)3]ACHTUNGTRENNUNG(PF6)2 and the resulting mixture was left to stir for 24 h at
RT. Upon filtration of the reaction mixture, the filtrate was concentrated
under vacuum and the resulting precipitate was washed with pentane
(15 mL).
C33H43ClF6IrN2PRu·5/2CHCl3: C 34.39, H 4.01, N 2.27; found: C 34.39,
H 3.70, N 2.26.
Synthesis of complex 4c
0.162 mmol) and [Cp*Ru
(10 mL) afforded compound 4cAHCTNUGTRENNNUG
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
1
after recrystallization (121 mg). H NMR (400 MHz, CDCl3): d=1.421 (s,
9H; tBu), 1.514 (s, 15H; C5Me5), 1.705 (s, 15H; C5Me5), 3.070 (s, 6H;
NMe2), 5.57 (dd, 1H, 4J=1.9, 3J=6.4 Hz), 5.71 (d, 1H, 4J=1.9 Hz), 6.67
(d, 1H, 3J=6.4 Hz), 7.23 (td, 1H, 4J=1.9, 3J=6.0 Hz), 7.80 (d, 1H, 4J=
2.0 Hz), 8.41 ppm (d, 1H, 3J=6.2 Hz); 13C NMR (100 MHz, CDCl3): d=
164.0, 163.3, 151.6, 124.5, 121.3, 117.3, 97.4, 93.5, 89.5, 80.9, 76.3, 69.7,
53.5, 40.0, 35.5, 30.6, 11.4, 8.7 ppm; elemental analysis calcd (%) for
C37H45IrRuN2ClPF6·CHCl3: C 34.35, H 3.80, N 1.95; found: C 34.40,
H 3.98, N 2.29.
Synthesis of complex 3bACHTNUGTRNEG[UN PF6]2: The reaction of a solution of compound
2b (230 mg, 0.41 mmol) in acetone (8 mL), [(Cp*IrCl2)2] (200 mg,
0.25 mmol), and AgPF6 (253 mg, 1 mmol) in acetone (8 mL) afforded
compound 3bACHTUNGTRENNUNG[PF6]2, which was isolated as an orange solid in 78% yield
after recrystallization (337.8 mg). 1H NMR (500 MHz, [D6]acetone): d=
1.7 (s, 15H; C5Me5), 2.14 (s, 15H; C5Me5), 3.57 (s, 3H; NMe), 3.73 (s,
3H; NMe), 6.68 (dd, 1H, 4J=2.4, 3J=7.1 Hz), 7.2 (d, 1H, 4J=2.3 Hz),
3
4
3
3
7.58 (d, 1H, J=7.0 Hz), 7.81 (ddd, 1H, J=2.3, J=5.7, J=6.6 Hz), 8.27
(m, 2H, 4J=1.7, 3J=5.4, 3J=7.0 Hz), 8.88 ppm (d, 1H, 3J=5.5 Hz);
13C NMR (125 MHz, [D6]acetone): d=157.6, 154.7, 151.3, 141.1, 140.7,
127.7, 120.8, 103.4, 101.9, 92.6, 86.7, 74.0, 68.8, 39.6, 39.3, 10.3, 7.9 ppm;
elemental analysis calcd (%) for C33H43ClF12Ir2N2P2: C 33.66, H 3.69,
N 2.38; found: C 33.82, H 4.10, N 2.11.
Synthesis of complex 4a
0.178 mmol) and [Cp*Ru
(10 mL) afforded compound 4aAHCTNUGTRENNNUG
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
lowing precipitation from CHCl3/pentane (47 mg). 1H NMR (400 MHz,
[D4]methanol/[D6]acetone 9:1): d=1.53 (s, 15H; C5Me5), 1.795 (s, 15H;
C5Me5), 5.65 (dd, 1H, 3J=5.4, 3J=6.0 Hz), 5.75 (t, 1H, 3J=5.7 Hz), 6.24
Synthesis of complex 3dACHTNUGTRNEG[UN PF6]2: The reaction of a solution of compound
2d (105 mg, 0.18 mmol) in acetone (8 mL), [(Cp*IrCl2)2] (90 mg,
0.11 mmol), and AgPF6 (114.4 mg, 0.45 mmol) in acetone (8 mL) afford-
3
3
4
3
(d, 1H, J=5.7 Hz), 6.45 (d, 1H, J=5.8 Hz), 7.55 (ddd, 1H, J=1.5, J=
5.7, 3J=7.4 Hz), 7.95 (dd, 1H, 4J=1.3, 3J=8.0 Hz), 8.06 (ddd, 1H, 4J=
1.6, 3J=7.4, 3J=8.1 Hz), 8.73 ppm (d, 1H, 3J=6.2 Hz); 13C NMR
(100 MHz, [D4]methanol/[D6]acetone 9:1): 164.2, 154.9, 140.3, 134.6,
126.6, 120.8, 103.1, 97.0, 92.6, 91.5, 88.3, 86.4, 82.2, 11.3, 8.6 ppm; elemen-
tal analysis calcd (%) for C31H38IrRuNClPF6·3CHCl3: C 33.60, H 3.45,
N 1.47; found: C 33.44, H 3.38, N 1.15.
ed compound 3dACHTUNGTRENNUNG[PF6]2, which was isolated as an orange solid in 20%
yield after recrystallization (41 mg). 1H NMR (500 MHz, [D6]acetone):
d=1.7 (s, 15H; C5Me5), 2.189 (s, 15H; C5Me5), 2.189 (s, 3H; OMe), 7.3
4
3
4
4
(dd, 1H, J=2.1, J=6.7 Hz), 7.79 (d, 1H, J=2.1 Hz), 7.89 (dd, 1H, J=
1.7, 3J=3.8 Hz), 7.91 (d, 1H, 3J=6.8 Hz), 8.34 (m, 2H, 4J=1.6, 3J=
8.0 Hz), 8.94 ppm (d, 1H, J=5.7 Hz); 13C NMR (125 MHz, [D6]acetone):
3
General procedure for the synthesis of Cr(CO)3-bound ligands: A mix-
ture of a 2-phenylpyridine and Cr(CO)6 was dissolved in a mixture of
n-Bu2O (150 mL) and THF (10 mL) under an argon atmosphere. The sus-
pension was gently heated to reflux for 7 days under an argon atmos-
phere. The resulting orange-yellow solution was cooled to RT and filtered
through celite. The filtrate was evaporated under reduced pressure; the
resulting oil was dissolved in CH2Cl2, and silica gel was added. After
evaporation of the solvent under reduced pressure, the coated silica gel
was loaded onto the top of a silica gel column and packed in with a pen-
tane/acetone (95:5) mixture. Complex 1c was eluted with pentane/ace-
tone (85:15). The solvents were removed under vacuum to afford
a canary-yellow solid.
d=155.5, 153.8, 141.6, 129.1, 122.2, 108.2, 104.2, 93.9, 89.0, 88.9, 86.7,
79.0, 59.0, 10.6, 8.3 ppm; elemental analysis calcd (%) for
C32H40Ir2ClONP2F12·1/4C3H6O: C 33.71, H 3.63, N 1.17; found: C 33.72,
H 3.76, N 1.19.
Synthesis of complex 3eACHTNUGTRNEG[UN PF6]2: The reaction of a solution of compound
2e (150 mg, 0.26 mmol) in acetone (8 mL), [(Cp*IrCl2)2] (114.6 mg,
0.14 mmol), and AgPF6 (145.7 mg, 0.57 mmol) in acetone (8 mL) afford-
ed a 6:4 mixture of compounds 3e-aACHTNURTGENNU[G PF6]2 and 3e-bAHCUTNTGERN[NUGN PF6]2, which were
isolated as an orange solid in 80% yield following recrystallization
(247 mg). 1H NMR (500 MHz, [D6]acetone): d=1.40 (t, 3H, 3J=7.2 Hz;
NCH2Me3eꢀa), 1.44 (t, 3H, 3J=7.0 Hz, NCH2Me3eꢀb), 1.67 (s, 15H;
C5Me5), 2.12 (d, 15H; C5Me5), 3.5 (s, 3H; NMe3eꢀb), 3.7 (s, 3H;
NMe3eꢀa), 3.87 (sextet, 1H, 3J=6.6 Hz; NCH2Me3eꢀb), 3.90 (m, 2H;
Synthesis of complex 1 f: The reaction of compound 1b (1.335 g,
6.74 mmol) and Cr(CO)6 (1.63 g, 7.42 mmol) in nBu2O (150 mL) and
THF (10 mL) afforded the product in 76% yield (1.7 g). 1H NMR
3
3
NEt3eꢀa), 4.33 (sextet, 1H, J=6.6 Hz; NCH2Me3eꢀb), 6.63 (d, 1H3eꢀb, J=
7.1 Hz), 6.76 (d, 1H3eꢀa
3J=7.1 Hz), 7.13 (d, 1H3eꢀa 4J=2.1 Hz), 7.23 (d,
1H3eꢀb
4J=2.1 Hz), 7.57 (dd, 1H, 3J=6.93 Hz), 7.82 (dd, 1H, 3J=5.14,
,
,
3
(500 MHz, CDCl3): d=2.94 (s, 6H; NMe2), 4.95 (d, 2H, J=7.4 Hz), 6.44
,
3
3
3
3
3J=8.0 Hz). 8.27 (m, 2H), 8.88 ppm (d, 1H, 3J=5.62 Hz); 13C NMR
(125 MHz, [D6]acetone): d=158.4, 155.5, 151.7, 143.0, 142.0, 142.6, 128.5,
121.6, 121.7, 102.6 (C3eꢀb), 102.5 (C3eꢀa), 93.4 (C3eꢀa), 93.3 (C3eꢀb), 87.7
(C3eꢀa), 87.6 (C3eꢀb), 74.4 (C3eꢀb), 73.7 (C3eꢀa), 69.5 (C3eꢀb), 68.9 (C3eꢀa),
(d, 2H, J=7.3 Hz), 7.17 (dd, 1H, J=4.89, J=7.4 Hz), 7.51 (d, 1H, J=
8.1 Hz), 7.67 (dd, 1H, 4J=1.66, 3J=6 Hz), 8.54 ppm (d, 1H, 3J=5.0 Hz);
13C NMR (125 MHz, CDCl3): d 233.9, 154.0, 136.8, 135.2, 122.6, 119.2,
96.6, 95.7, 73.9, 39.9 ppm; IR (ATR): n˜ =1936ACTHNUTRGNEUNG
(s, C=O), 1835 cmꢀ1 (vs, C=
6076
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 6063 – 6078

Synthesis of Planar Chiral Iridacycles
FULL PAPER
O); elemental analysis calcd (%) for C16H14CrN2O3: C 57.47, H 4.22,
N 8.38; found: C 57.47, H 4.19, N 8.25.
X-ray diffraction: The acquisition and processing parameters are given in
Table 1. Reflections were collected with a Nonius KappaCCD diffractom-
eter by using MoKa graphite-monochromated radiation (l=0.71073 ꢂ).
The crystal structures were solved by using SIR 97[43] and refined with
SHELXL-97.[44] The refinement and all further calculations were carried
out by using SHELXL-97. The H-atoms were included at their calculated
positions and were treated as riding atoms by using the SHELXL default
parameters. The non-H atoms were refined anisotropically, by using
weighted full-matrix least-squares on F2. In cases where highly distorted
solvent molecules were localized in the lattice, they were treated by the
SQUEEZE method to enable further refinement of the data. Such cases
are commented on (see the Supporting Information).
Synthesis of complex 1g: The reaction of compound 1e (2 g, 7.84 mmol)
and Cr(CO)6 (1.9 g, 8.62 mmol) in nBu2O (150 mL) and THF (10 mL) af-
1
forded the product in 72% yield (2.2 g). H NMR (500 MHz, CDCl3): d=
1.35 (s, 9H; tBu), 2.94 (s, 6H; NMe2), 4.96 (d, 2H, 3J=7.3 Hz), 6.43 (d,
2H, 3J=7.3 Hz), 7.19 (dd, 1H, 4J=1.8, 3J=5.3 Hz), 7.53 (d, 1H, 4J=
1.9 Hz), 8.44 ppm (d, 1H, 3J=5.3 Hz); 13C NMR (125 MHz, CDCl3): d=
234.1, 160.8, 154.0, 149.0, 13, 1, 116.6, 97.5, 96, 7, 39.8, 34.8, 30.5 ppm; IR
(ATR) n˜ =1936 (C=O), 1840 cmꢀ1 (C=O); elemental analysis calcd (%)
for C20H22CrN2O3: C 61.51, H 5.68, N 7.18; found: C 61.35, H 5.82,
N 7.04.
CCDC-805113 (2b), CCDC-805114 (3b
CCDC-805116 (5b), CCDC-853686 (1g), CCDC-853687 (3e
CCDC-853688 (4c[PF6]2), and CCDC-853689 ([6b·H2O][PF6]3) contain
the supplementary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
N
ACHTUNGTRENNUNG[PF6]2),
ACHTUNGTRENNUNG
General procedure for the synthesis of the [Ir,Cr] complexes: A mixture
of [(Cp*IrCl2)2], [(h6-arene)Cr(CO)3], and NaOAc·3H2O was dissolved
in CH2Cl2 (20 mL). The resulting mixture was stirred at RT for 24 h
under an argon atmosphere. The resulting red solution was filtered
through celite and the filtrate was evaporated to dryness under reduced
pressure to afford an orange solid, which was recrystallized from a mix-
ture of pentane (15 mL) and CH2Cl2 (5 mL) and dried under reduced
pressure overnight.
G
ACHTUNGTRENNUNG
Synthesis of complex 5a: The reaction of [(Cp*IrCl2)2] (0.498 mmol), 1 f
(336 mg, 1.0 mmol) and NaOAc·3H2O (324 mg, 3.95 mmol) in CH2Cl2
(20 mL) afforded the product as an orange powder in 75% yield
(520 mg). 1H NMR (500 MHz, CDCl3): d=1.77 (s, 15H; C5Me5), 2.96 (s,
6H; NMe2), 4.82 (dd, 1H, 4J=2.4, 3J=7.1 Hz), 5.61 (d, 1H, 4J=2.4 Hz),
6.23 (d, 1H, J=7.1 Hz), 7.07 (ddd, 1H, J=1.4, J=5.8, J=7.5 Hz), 7.49
(d, 1H, 3J=8.2 Hz), 7.68 (dt, 1H, 4J=1.5, 3J=7.8 Hz), 8.59 ppm (d, 1H,
3J=5.8 Hz); 13C NMR (125 MHz, CDCl3): d=235.9, 165.9, 150.6, 137.4,
135.1, 131.1, 122.5, 119.3, 103.4, 93.9, 89.6, 86.8, 72.4, 40.1, 8.9 ppm; IR
(ATR) n˜ =1917 (C=O), 1829 cmꢀ1 (C=O); elemental analysis calcd (%)
for C26H28ClCrIrN2O3·1/3CH2Cl2: C 43.77, H 3.99, N 3.88; found: C 43.76,
H 4.16, N 3.55.
Acknowledgements
The authors gratefully acknowledge the support provided by the CNRS
and the University of Strasbourg (PhD funding to W.I.). Lydia Brelot is
thanked for the resolution of the structure of compound 2b.
3
4
3
3
[1] a) C. Bolm, K. Muniz, Chem. Soc. Rev. 1999, 28, 51–59; b) R. Gꢅ-
mez Arrayꢆs, J. Adrio, J. C. Carretero, Angew. Chem. 2006, 118,
7836–7878; Angew. Chem. Int. Ed. 2006, 45, 7674–7715.
[2] M. Albrecht, Chem. Rev. 2010, 110, 576–623.
[3] a) J.-P. Djukic, J.-B. Sortais, L. Barloy, M. Pfeffer, Eur. J. Inorg.
Chem. 2009, 817–853; b) J.-B. Sortais, V. Ritleng, A. Voelklin, A.
Holuigue, H. Smail, L. Barloy, C. Sirlin, G. K. M. Verzijl, J. A. F. Bo-
ogers, A. H. M. de Vries, J. G. de Vries, M. Pfeffer, Org. Lett. 2005,
7, 1247–1250; c) J.-B. Sortais, L. Barloy, C. Sirlin, A. H. M. de Vries,
J. G. de Vries, M. Pfeffer, Pure Appl. Chem. 2006, 78, 457–462;
d) M. Albrecht, B. M. Kocks, A. L. Spek, G. van Koten, J. Organo-
met. Chem. 2001, 624, 271–286; e) P. Dani, T. Karlen, R. A. Gos-
sage, S. Gladiali, G. van Koten, Angew. Chem. 2000, 112, 759–761;
Angew. Chem. Int. Ed. 2000, 39, 743–745; f) D. Amoroso, A. Jabri,
P. A. Yap, D. G. Gusev, E. N. dos Santos, D. E. Fogg, Organometal-
lics 2004, 23, 4047–4054; g) W. Baratta, P. da Ros, A. del Zotto, A.
Sechi, E. Zangrando, P. Rigo, Angew. Chem. 2004, 116, 3668–3672;
Angew. Chem. Int. Ed. 2004, 43, 3584–3588.
Synthesis of complex 5b: The reaction of compound 1g (250 mg,
0.639 mmol), [(Cp*IrCl2)2] (254 mg, 0.32 mmol), and NaOAc·3H2O
(209.59 mg, 2.55 mmol) in CH2Cl2 (15 mL) afforded the product as
a yellow powder in 70% yield (336 mg). 1H NMR (500 MHz, CDCl3):
d=1.34 (s, 9H; tBu), 1.78 (s, 15H; C5Me5), 2.96 (s, 6H; NMe2), 4.82 (d,
1H, 3J=7.1 Hz), 5.61 (d, 1H, 4J=1.7 Hz), 6.23 (d, 1H, 3J=7.2 Hz), 7.06
(d, 1H, 3J=5.7 Hz), 7.4 (s, 1H), 8.46 ppm (d, 1H, 3J=6.3 Hz); 13C NMR
(125 MHz, CDCl3): d=235.9, 165.2, 161.8, 149.9, 135.1, 131.4, 120.5,
115.9, 104.0, 93.8, 89.4, 86.8, 72.3, 40.1, 35.2, 30.5, 8.9 ppm; IR n˜ =1917
(C=O), 1829 cmꢀ1 (C=O); elemental analysis calcd (%) for
C30H36ClCrIrN2O3·3/5CH2Cl2: C 45.88, H 4.89, N 3.09; found: C 45.76,
H 4.67, N 3.49.
An alternative synthesis of compound 5a: Compound 2e (159 mg,
0.28 mmol)
and
tricarbonyl(h6-naphthalene)chromium
(108 mg,
[4] J.-P. Djukic, A. Hijazi, H. D. Flack, G. Bernardinelli, Chem. Soc.
Rev. 2008, 37, 406–425.
0.41 mmol) were dissolved in dry and degassed THF (15 mL). The result-
ing mixture was stirred at RT for 24 h under an argon atmosphere. The
resulting solution was filtered through celite, the filtrate was concentrat-
ed to about 5 mL, and silica gel was added. The solvent was evaporated
under reduced pressure, and the coated silica gel was loaded onto the top
of a silica gel column that was packed with pentane/CH2Cl2 (50:50) at
58C. Elution with CH2Cl2/pentane (75:25) afforded the product in quanti-
tative yield (195 mg).
[5] A. Berger, J.-P. Djukic, M. Pfeffer, J. Lacour, L. Vial, A. De Cian,
N. Kyritsakas-Gruber, Organometallics 2003, 22, 5243–5260.
[6] A. Hijazi, J.-P. Djukic, L. Allouche, A. de Cian, M. Pfeffer, X.-F.
Le Goff, L. Ricard, Organometallics 2007, 26, 4180–4196.
[7] J.-P. Djukic, A. Berger, M. Duquenne, M. Pfeffer, A. de Cian, N.
Kyritsakas-Gruber, J. Vachon, J. Lacour, Organometallics 2004, 23,
5757–5767.
Preparation of tris-cationic compound 6b
ACHTUGNTRENUN[GN PF6]3: A mixture of compound
[8] a) S. Bonnet, M. Lutz, A. L. Spek, G. van Koten, R. J. M. Klein Geb-
bink, Organometallics 2008, 27, 159–162; b) S. Bonnet, J. H. van
Lenthe, M. A. Siegler, A. L. Spek, G. van Koten, R. J. M. Klein
Gebbink, Organometallics 2009, 28, 2325–2333; c) S. Bonnet, M.
Lutz, A. L. Spek, G. van Koten, R. J. M. Klein Gebbink, Organome-
tallics 2010, 29, 1157–1167.
3b[PF6]2 (102 mg, 0.084 mmol) and AgPF6 (22 mg, 0.084 mmol) in [D6]-
ACHTUNGTRENNUNG
acetone (10 mL) was stirred at RT for 1 h under an argon atmosphere.
The resulting solution was filtered through celite to remove the AgCl by-
product. The reaction proceeded quantitatively. Compound 6bACHTNUGTRNEG[UN PF6]3,
albeit stable when dissolved in acetone, was prone to fast decomposition
upon removal of the solvent under reduced pressure. 1H NMR
(500 MHz, [D6]acetone): d=1.69 (s, 15H; C5Me5), 2.15 (s, 15H; C5Me5),
[9] J.-P. Djukic, W. Iali, M. Pfeffer, X.-F. Le Goff, Chem. Commun.
2011, 47, 3631–3633.
4
3
3.60 (s, 3H; NMe), 3.80 (s, 3H; NMe), 6.58 (dd, 1H, J=2.4, J=7.1 Hz),
[10] a) B. Nicholls, M. C. Whiting, Proc. Chem. Soc. (London) 1958, 152;
b) B. Nicholls, M. C. Whiting, J. Chem. Soc. 1959, 551–556; c) W. R.
Jackson, B. Nicholls, M. C. Whiting, J. Chem. Soc. 1960, 469–470;
d) R. Dabard, A. Meyer, C. R. Hebd. Seances Acad. Sci. 1967, 264,
903–906; e) C. A. L. Mahaffy, Synth. React. Inorg. Met.-Org. Chem.
1984, 14, 895–903; f) C. A. L. Mahaffy, Synth. React. Inorg. Met.-
4
3
4
7.43 (d, 1H, J=2.3 Hz), 7.54 (d, 1H, J=7.0 Hz), 7.98 (ddd, 1H, J=2.3,
3J=5.7, 3J=6.6 Hz), 8.38 (m, 2H, 4J=1.2, 3J=4.0, 3J=7.0 Hz), 9.33 ppm
(d, 1H, 3J=5.5 Hz); 13C NMR (125 MHz, [D6]acetone): d=158.4, 155.7,
146.7, 142.5, 142.7, 129.3, 122.7, 103.3, 102.7, 92.3, 88.9, 74.5, 70.2, 40.2,
10.2, 9.0 ppm.
Chem. Eur. J. 2012, 18, 6063 – 6078
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6077

J.-P. Djukic et al.
Org. Chem. 1985, 15, 301–307; g) C. A. L. Mahaffy, J. Hamilton,
Synth. React. Inorg. Met.-Org. Chem. 1987, 17, 849–855; h) C. A. L.
Mahaffy, P. L. Pauson, Inorg. Synth. 1990, 28, 136–140.
[11] a) P. Espinet, P. M. Bailey, R. F. Downey, P. M. Maitlis, J. Chem. Soc.
Dalton Trans. 1980, 1048–1054; b) A. D. Hunter, L. Shilliday, W. S.
Furey, M. J. Zaworotko, Organometallics 1992, 11, 1550–1560;
c) S. P. Nolan, K. L. Martin, E. D. Stevens, P. J. Fagan, Organometal-
lics 1992, 11, 3947–3953.
[12] J.-P. Djukic, L. Fetzer, A. Czysz, W. Iali, C. Sirlin, M. Pfeffer, Orga-
nometallics 2010, 29, 1675–1679.
[13] J. W. Kang, K. Moseley, P. M. Maitlis, J. Am. Chem. Soc. 1969, 91,
5970–5977.
[30] a) D. Lauvergnat, P. C. Hiberty, J. Am. Chem. Soc. 1997, 119, 9478–
9482; b) N. G. Vassilev, V. S. Dimitrov, J. Mol. Struct. 1999, 484, 39–
47; c) C. R. Kemnitz, M. J. Loewen, J. Am. Chem. Soc. 2007, 129,
2521–2528.
[31] a) F. P. Gasparro, N. H. Kolodny, J. Chem. Educ. 1977, 54, 258–260;
b) M. Feigel, J. Phys. Chem. 1983, 87, 3054–3058.
[32] A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100.
[33] a) J. P. Perdew, Phys. Rev. B 1986, 34, 7406; b) J. P. Perdew, Phys.
Rev. B 1986, 33, 8822–8824.
[34] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77,
3865–3868.
[35] J. Tao, J. P. Perdew, V. N. Staroverov, G. E. Scuseria, Phys. Rev. Lett.
2003, 91, 146401.
[36] in Amsterdam Density Functional (Ed. SCM), Department of Theo-
retical Chemistry, Vrije Universiteit Amsterdam,The Netherlands,
2003.
[14] D. L. Davies, O. Al-Duaij, J. Fawcett, M. Giardiello, S. T. Hilton,
D. R. Russell, Dalton Trans. 2003, 4132–4138.
[15] I. I. Rangel Salas, M. A. Paz-Sandoval, H. Nçth, Organometallics
2002, 21, 4696–4710.
[16] M. D. Mbaye, B. Demerseman, J. L. Renaud, L. Toupet, C. Bruneau,
Adv. Synth. Catal. 2004, 346, 835–841.
[17] M. Uemura, T. Minami, K. Hirotsu, Y. Hayashi, J. Org. Chem. 1989,
54, 469–477.
[18] R. Meyer, P. L. Wessels, P. H. van Rooyen, S. Lotz, Inorg. Chim.
Acta 1999, 284, 127–132.
[19] J. Hamilton, C. A. L. Mahaffy, J. Rawlings, Inorg. Chim. Acta 1996,
244, 265–267.
[20] J.-P. Djukic, F. Rose-Munch, E. Rose, J. Vaissermann, Eur. J. Inorg.
Chem. 2000, 1295–1306.
[21] C. H. Suresh, N. Koga, S. R. Gadre, Organometallics 2000, 19, 3008–
3015.
[22] K. Morokuma, J. Chem. Phys. 1971, 55, 1236–1244.
[23] G. Te Velde, F. M. Bickelhaupt, S. J. A. van Gisbergen, C. Fonseca-
Guerra, E. J. Baerends, J. G. Snijders, T. Ziegler, J. Comput. Chem.
2001, 22, 931–967.
[24] M. P. Mitoraj, A. Michalak, T. Ziegler, J. Chem. Theory Comput.
2009, 5, 962–975.
[25] a) J.-P. Djukic, C. Boulho, D. Sredojevic, C. Scheeren, S. Zaric, L.
Ricard, M. Pfeffer, Chem. Eur. J. 2009, 15, 10830–10842; b) C.
Scheeren, F. Maasarani, A. Hijazi, J.-P. Djukic, M. Pfeffer, S. D.
Zaric, X.-F. Le Goff, L. Ricard, Organometallics 2007, 26, 3336–
3345.
[26] L. Li, Y. Jiao, W. W. Brennessel, W. D. Jones, Organometallics 2010,
29, 4593–4605.
[27] a) S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010,
132, 154104; b) S. Grimme, J. Comput. Chem. 2006, 27, 1787–1799;
c) S. Grimme, R. Huenerbein, S. Ehrlich, ChemPhysChem 2011, 12,
1258–1261.
[37] E. J. Baerends, J. Autschbach, A. Bꢀrces, F. M. Bickelhaupt, C. Bo,
P. M. Boerrigter, L. Cavallo, D. P. Chong, L. Deng, R. M. Dickson,
D. E. Ellis, M. van Faassen, L. Fan, T. H. Fischer, C. F. Guerra,
S. J. A. van Gisbergen, A. W. Gçtz, J. A. Groeneveld, O. V. Gritsen-
ko, M. Grꢇning, F. E. Harris, P. van den Hoek, C. R. Jacob, H. Ja-
cobsen, L. Jensen, G. van Kessel, F. Kootstra, M. V. Krykunov, E.
van Lenthe, D. A. McCormack, A. Michalak, J. Neugebauer, V. P.
Nicu, V. P. Osinga, S. Patchkovskii, P. H. T. Philipsen, D. Post, C. C.
Pye, W. Ravenek, J. I. Rodrꢈguez, P. Ros, P. R. T. Schipper, G.
Schreckenbach, J. G. Snijders, M. Solꢉ, M. Swart, D. Swerhone, G. te
Velde, P. Vernooijs, L. Versluis, L. Visscher, O. Visser, F. Wang,
T. A. Wesolowski, E. M. van Wezenbeek, G. Wiesenekker, S. K.
Wolff, T. K. Woo, A. L. Yakovlev and T. Ziegler in Amsterdam Den-
sity Functional (Ed. SCM), Department of Theoretical Chemistry,
Vrije Universiteit Amsterdam, The Netherlands, 2010.
[38] S. H. Vosko, L. Wilk, M. Nusair, Can. J. Phys. 1980, 58, 1200–1211.
[39] J. P. Perdew, Y. Wang, Phys. Rev. B 1992, B 45, 13244–13249.
[40] E. van Lenthe, A. E. Ehlers, E. J. Baerends, J. Chem. Phys. 1999,
110, 8943–8953.
[41] E. van Lenthe, E. J. Baerends, J. Comput. Chem. 2003, 24, 1142–
1156.
[42] a) A. Klamt, G. Schꢇꢇrmann, J. Chem. Soc. Perkin Trans. 2 1993,
799–805; b) A. Klamt, V. Jones, J. Chem. Phys. 1996, 105, 9972–
9981; c) A. Klamt, J. Phys. Chem. 1995, 99, 2224–2235.
[43] A. Altomare, M. C. Burla, M. Camalli, G. Cascarano, C. Giacovaz-
zo, A. Guagliardi, A. G. G. Moliterni, G. Polidori, R. Spagna in
SIR97, an integrated package of computer programs for the solution
and refinement of crystal structures using single-crystal data 1999.
[44] M. Sheldrick in SHELXL-97, Universitꢊt Gçttingen, Gçttingen,
Germany, 1998.
[28] J. Sandstrçm, Dynamic NMR spectroscopy, Academic Press,
London, 1982.
[29] B. W. Tattershall, in DNM3RUN and N3PLOT for Windows, New-
castle University, Newcastle, England, 2007.
Received: November 14, 2011
Published online: March 30, 2012
6078
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 6063 – 6078