Palladation of diimidazolium salts at the C4 position: Access to remarkably electron-rich palladium(ii) centers

Article abstract of DOI:10.1039/b812405a

Palladation of C2-protected diimidazolium salts with Pd(OAc)₂ afforded complexes comprising C4-bound N-heterocyclic dicarbene ligands. The reactivity of these complexes towards Lewis acids (AgBF₄, AgOAc) and Bronsted acids (H₂SO₄, H₃PO₄, HOAc) revealed that abnormal C4 bonding of the carbenes markedly increases the nucleophilicity of the coordinated palladium center as compared to C2 bonding. Despite its formal +2 charge, the palladium center in these complexes is best described as a Lewis base. The abnormal carbene bonding mode induces new reaction patterns such as the formation of a Pd-Ag adduct. Based on metallation studies including the palladation of a dissymmetric diimidazolium salt, a rationale for the selective activation of the C4-H bond in the diimidazolium precursor salts is proposed. The Royal Society of Chemistry 2008.

Full text of DOI:10.1039/b812405a

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Palladation of diimidazolium salts at the C4 position: access to remarkably
electron-rich palladium(II) centers†
a
a
b
a
Marion Heckenroth, Evelyne Kluser, Antonia Neels and Martin Albrecht*
Received 21st July 2008, Accepted 21st August 2008
First published as an Advance Article on the web 1st October 2008
DOI: 10.1039/b812405a
Palladation of C2-protected diimidazolium salts with Pd(OAc)
C4-bound N-heterocyclic dicarbene ligands. The reactivity of these complexes towards Lewis acids
AgBF , AgOAc) and Brønsted acids (H SO , H PO , HOAc) revealed that abnormal C4 bonding of
2
afforded complexes comprising
(
4
2
4
3
4
the carbenes markedly increases the nucleophilicity of the coordinated palladium center as compared to
C2 bonding. Despite its formal +2 charge, the palladium center in these complexes is best described as a
Lewis base. The abnormal carbene bonding mode induces new reaction patterns such as the formation
of a Pd–Ag adduct. Based on metallation studies including the palladation of a dissymmetric
diimidazolium salt, a rationale for the selective activation of the C4–H bond in the diimidazolium
precursor salts is proposed.
Introduction
The organometallic chemistry of N-heterocyclic carbenes
1
(
NHCs) has provided an entirely new twist on the development
2
of active catalysts and materials for optical and electronic
applications. The success of NHCs in these areas is quite generally
ascribed to the relative stability of the M–CNHC bond and the
3
4
strong donor ability of this type of ligand. While most studies
have focused on NHC ligands resulting from metallation of
imidazolium and related salts at the C2 position (A, Fig. 1), a
variety of new NHC systems have recently been developed in which
the carbene and at least one of the stabilizing heteroatoms are
5
mutually more remote. In such a ligand framework, the inductive
Fig. 1 a) Generic C2 and C4 bonding of imidazolium derived NHCs (A
and B, respectively); b), palladium dicarbene complexes featuring normal
and mesomeric influence of the heteroatom is significantly altered.
As a consequence and also due to the associated steric differences,
the bonding of these less stabilized carbene ligands to the metal,
and likewise the reactivity patterns of the coordinated metal center
vary distinctly from those of the classical imidazolium-derived
NHC systems A.
(
1) and abnormal bonding (2).
7,13
a broad range of transition metals. While crucial mechanistic
details have been disclosed for the iridium- and osmium-mediated
12
activation of the imidazolium C4–H bond, little is known for the
6
Stimulated by its serendipitous discovery, we have become
corresponding palladation reaction.
Following our preliminary communication, we report here in
interested in exploiting the potential of abnormal imidazolylidenes
14
7
B (E = NR, Fig. 1) as ligands to transition metals. Different
full on the preparation and the unique reactivity of palladium(II)
complexes comprising chelating C4-bound dicarbene ligands with
the bonding motif B (2, Fig. 1). Since complexes 2 are active
hydrogenation catalysts but their normal congeners 1 featuring
the C2 bonding motif A are not, a special focus has been directed
towards a detailed comparison of the structure and reactivity of
complexes 1 and 2 (Fig. 1).
rational procedures for installing the metal at the imidazolium C4
8
position have been developed in the last few years, including
9
10
inter alia C–H oxidative addition, C–X oxidative addition,
I
11
and transmetallation from Ag precursors. Most conveniently,
metallation of the imidazolium C4 position can be accomplished
via direct C–H bond activation, provided the C2 carbon is
appropriately protected. This method circumvents tedious pre-
functionalization of the C4 position and may be applicable for
12
Experimental
a
Department of Chemistry, University of Fribourg, Chemin du Musee 9, CH-
700, Fribourg, Switzerland. E-mail: martin.albrecht@unifr.ch; Fax: +41
63009738; Tel: +41 263008786
General comments
1
2
The synthesis and spectroscopic data of the palladium complexes
b
CSEM Centre Suisse d’Electronique et de Microtechnique, Rue Jaquet-Droz
14
2a–c, 5, and 6 have been reported previously. All other reagents
1
, CH-2002, Neuch aˆ tel, Switzerland
are commercially available and were used as received. Unless
specified otherwise, NMR spectra were recorded at 25 C on
Bruker spectrometers and chemical shifts (d in ppm, coupling
†
1
1
CCDC reference numbers 603584, 695648–695651 for 2a, 2c, 2d, 7, and
1. For crystallographic data in CIF or other electronic format see DOI:
0.1039/b812405a
◦
6
242 | Dalton Trans., 2008, 6242–6249
This journal is © The Royal Society of Chemistry 2008

View Online
1
13
constants J in Hz) were referenced to external SiMe
4
( H, C).
d 144.6 (Cimi–Ph), 132.9, 130.4, 129.6 (3 ¥ Caryl), 122.8, 122.2 (2 ¥
Assignments are based on homo- and heteronuclear shift corre-
lation spectroscopy. Elemental analyses were performed by the
Microanalytical Laboratory of Ilse Beetz (Kronach, Germany).
C
(CH
imi), 119.6 (Caryl), 57.9 (NCH
2
N), 48.1 (NCH
2
), 30.7 (CH
2
), 18.5
2
), 13.0 (CH
3
). Elem. anal. calcd. for C27
H
34
I
2
N
4
(668.39) ¥
MeOH: C 48.01, H 5.47, N 8.00; found: C: 47.88, H 5.24, N 7.98.
Synthesis of 2d. To a solution of 4d (0.67 g, 1.0 mmol)
Synthesis
in DMSO (5 mL) was added Pd(OAc) (0.22 g, 1.0 mmol).
2
◦
The reaction mixture was stirred at 50 C for 2 h and then
General procedure for the synthesis of the diimidazolium salts
◦
at 120 C for 3 h. After cooling to RT, CH
2
2
Cl (20 mL) and
4
a–d. To a solution of the 1,2-substituted imidazole 3 in toluene
was added CH (0.5 mol equiv.). The reaction mixture was stirred
at reflux temperature for 18 h. The formed precipitate was filtrated
Et
and washed repeatedly by adding CH
precipitation with Et O (80 mL). After drying in vacuo, complex
d was obtained as an off-white powder (0.42 g, 55%). An
analytically pure sample was obtained by recrystallization from
2
O (80 mL) were added. The formed precipitate was collected
2 2
I
2
2
Cl (20 mL) followed by
off and washed with toluene (3 ¥ 30 mL) and Et O (3 ¥ 30 mL),
and dried in vacuo.
2
2
2
Synthesis of 4a. From 2-methyl-N-methyl imidazole 3a (4.81 g,
0 mmol) and CH (6.7 g, 25 mmol) according to the general
1
acetone/pentane. H NMR (360 MHz, DMSO-d ): d 7.59 (m,
6
5
I
2 2
2
H, Haryl), 7.5–7.44 (m, 8H, Haryl), 7.40 (s, 2H, Himi), 5.64 (s, 2H,
procedure. The diimidazolium salt 4a was obtained as an off-white
solid (6.97 g, 61%). An analytically pure sample was obtained by
recrystallization from MeOH/Et
3
3
NCH
.3 Hz, 4H, CH
2
N), 3.91 (t, JHH = 7.3 Hz, 4H, NCH
2
), 1.55 (quint, JHH
=
3
7
2
), 1.12 (sextet, JHH = 7.3 Hz, 4H, CH
2
), 0.71 (t,
1
3
13
1
2
O. H NMR (360 MHz, DMSO-
J
HH = 7.3, 6H, CH
41.5 (C–Ph), 131.7, 130.3, 129.2 (3 ¥ Caryl), 129.2 (Cimi–H), 121.7
C–Pd), 60.3 (NCH N), 47.0 (NCH ), 31.5 (CH ), 18.7 (CH ),
), quaternary Caryl not resolved. Elem. anal. calcd. for
Pd (772.80): C 41.96, H 4.17, N 7.25; found: C: 41.75,
3
). C{ H} NMR (125 MHz, DMSO-d
6
): d
d
6
): d 7.92, 7.76 (2 ¥ s, 2H, Himi), 6.65 (s, 2H, NCH
2
N), 3.78 (s, 3H,
). C{ H} NMR (125 MHz, DMSO-
): d 146.4 (Cimi–Me), 123.0 (Cimi), 120.9 (Cimi), 56.7 (NCH N),
5.1 (NCH ), 10.3 (Cimi–CH ). Elem. anal. calcd. for C11
1
(
1
C
1
3
1
NCH
d
3
3
), 2.75 (s, 3H, Cimi–CH
3
2
2
2
2
6
2
3.1 (CH
3
3
3
H
18
I
2
N
4
27
H
32
I
2
N
4
(460.10): C 28.72, H 3.94, N 12.18; found: C: 29.01, H 3.97, N
H 4.51, N 7.31.
11.92.
Synthesis of 7. Complex 2b (0.40 g, 0.64 mmol) and silver
Synthesis of 4b. From 2-methyl-N-isopropyl imidazole 3b
acetate (0.22 g, 1.28 mmol) were stirred in MeCN (10 mL) in the
◦
(
3.07 g, 24.07 mmol) and CH
2
I (3.32 g, 12.4 mmol) according
2
dark at 60 C for 2 d. The solution was filtered over Celite, dried
to the general procedure. Compound 4b was obtained as an off-
over Na
2
SO , and the solvent was removed in vacuo, thus yielding
4
white solid (4.35 g, 68%), which was recrystallized from MeOH
crude 7 as a waxy yellow solid (0.27 g, 87%). Colorless crystals were
grown by slow diffusion of pentane into an acetone solution of 7.
◦
1
at -30 C. H NMR (500 MHz, DMSO-d ): d 8.00, 7.97 (2 ¥ s,
3
6
2
H, Himi), 6.56 (s, 2H, NCH
CHMe
CH(CH
Me), 121.9, 118.7 (2 ¥ Cimi), 56.5 (NCH
2
N), 4.73 (sept, JHH = 6.8 Hz, 2H,
1
H NMR (360 MHz, CDCl
3
): d 6.35 (s, 4H Himi), 6.1, 5.9 (2 ¥ br AB
3
2
), 2.82 (s, 6H, Cimi–CH
3
), 1.44 (d, JHH = 6.8 Hz, 12H,
doublet, 2H, NCH
N), 4.38 (2 ¥ sept, JHH = 6.5 Hz, 4H, CHMe
), 2.10, 1.93 (2 ¥ s, 6H, CH COO),
). C{ H} NMR (125 MHz, DMSO-d ):
3
),
2
2
1
3
1
3
)
2
). C{ H} NMR (125 MHz, DMSO-d
N), 50.6 (CHMe
). Elem. anal. calcd. for C15
516.20): C 34.90, H 5.08, N 10.85; found: C: 34.93, H 5.01, N
6
): d 145.14 (Cimi
–
2
1
.77 (br s, 12H, Cimi–CH
3
13
3
1
2
2
), 21.8
.3 (m, 24H, CH(CH
)
3
2
6
(
(
CH(CH
3
)
2
), 10.4 (Cimi–CH
3
H
26
I
2
N
4
d 178.85 (MeCOO), 172.75 (MeCOO), 140.69 (C–Me), 131.87
(
(
C–Pd), 116.65 (Cimi–H), 59.12 (NCH
CH COO), 24.29 (CH COO), 21.93 (CH(CH
). Elem. anal. calcd. for C38 Pd (969.77) ¥ 6 H
C 42.34, H 6.73, N 10.40; found: C 42.27, H 6.63, N 10.36.
2
N), 48.95 (CHMe ), 30.66
2
10.86.
3
3
3
)
2
), 9.53 (Cimi
–
CH
3
H
60
N
8
O
8
2
2
O:
Synthesis of 4c. Reaction of 2-methyl-N-mesityl imidazole 3c
(
1.11 g, 5.5 mmol) and CH
2
I (0.74 g, 2.75 mmol) according to
2
the general procedure gave 4c as an off-white solid (1.32 g, 35%).
An analytically pure sample was obtained by recrystallization from
Acidolysis of 2. Reactions of 2 with acids were carried out in
an NMR tube in CD CN or DMSO-d solutions using benzene
3
6
1
MeOH/Et
2
2
O. H NMR (500 MHz, DMSO-d
6
): d 8.24, 8.06 (2 ¥ s,
N), 2.62 (s, 6H, Cimi
as an internal standard. In a typical experiment, H SO (40 mg,
2
4
H, Himi), 7.20 (s, 4H, Hmes), 6.80 (s, 2H, NCH
2
–
0.41 mmol) was added to a solution of 2b (67 mg, 0.11 mmol) and
1
3
1
CH
3
), 2.36 (s, 6H, Cmes–CH
3
), 1.99 (s, 12H, Cmes–CH
3
). C{ H}
benzene (8 mg, 0.10 mmol) in CD CN (0.8 mL). The reaction was
3
1
NMR (125 MHz, DMSO-d
29.9, 129.6 (4 ¥ Cmes), 123.8, 122.4 (2 ¥ Cimi), 58.2 (NCH
), 16.9 (Cmes–CH ), 10.2 (Cimi–CH ). Elem. anal. calcd.
(668.39): C 48.52, H 5.13, N 8.38; found: C: 48.50,
H 5.14, N 8.32.
6
): d 146.8 (Cimi–Me), 140.9, 134.6,
monitored by H NMR and reached 89% conversion after 15 min
1
1
2
N), 20.6
and quantitative formation of 8 after 3 h. H NMR (360 MHz,
3
3
(
Cmes–CH
3
3
3
CD CN): d 7.52 (d, 1H, J = 2.2 Hz, H ), 7.20 (d, 1H, J
=
3
HH
imi
HH
for C27
H
34
I
2
N
4
2.2 Hz, H ), 7.02 (s, 1H, H ), 6.30 (s, 2H, NCH N), 4.67 (sept,
imi
imi
2
3
3
JHH = 6.6 Hz, 1H, CHMe ), 4.47 (sept, J = 6.6 Hz, 1H,
2
HH
CHMe
J
CH(CH
2
), 2.83 (s, 3H, Cimi–CH
3
), 2.55 (s, 3H, Cimi–CH
3
), 1.44 (d,
Synthesis of 4d. Addition of CH
2
2
I (13.4 g, 50 mmol) to 2-
3
3
HH = 6.6 Hz, 6H, CH(CH
3
)
2
), 1.39 (d, JHH = 6.6 Hz, 6H,
phenyl-N-butyl imidazole 3d (20.0 g, 100 mmol) according to
the general procedure afforded 4d as an off-white solid (18.70 g,
1
3
1
3
)
2
). C{ H} NMR (100 MHz, DMSO-d
6
): d 145.1, 143.9
(
2 ¥ Cimi–Me), 121.3 (Cimi–H), 119.9 (C–Pd), 119.0, 118.6 (2 ¥
5
6%). Analytically pure material was obtained by recrystallization
C
imi–H), 59.0 (NCH
2
N), 50.4, 50.0 (2 ¥ CHMe
).
2
), 21.9, 21.6 (2 ¥
1
from MeOH/Et
2
O. H NMR (500 MHz, DMSO-d
6
): d 8.07, 7.95
CH(CH
3
)
2
), 11.1, 9.8 (2 ¥ Cimi–CH
3
(
2 ¥ s, 2H, Himi), 7.72 (m, 2H, Haryl), 7.60 (m, 8H, Haryl), 6.30 (s, 2H,
3
3
NCH
2
N), 3.92 (t, JHH = 7.4 Hz, 4H, NCH
2
), 1.57 (quint, JHH
=
Synthesis of 9. A mixture of 1-isopropyl-2-methylimidazole
(1.00 g, 8.0 mmol), 1-isopropylimidazole (0.91 g, 8.0 mmol) and
3
7
.4 Hz, 4H, CH
2
), 1.08 (sextet, JHH = 7.4 Hz, 4H, CH
2
), 0.70 (t,
3
13
1
J
HH = 7.4 Hz, 6H, CH
3
). C{ H} NMR (125 MHz, DMSO-d
6
):
CH
2
I
2
(1.08 g, 4.0 mmol) was stirred in toluene (20 mL) at reflux
This journal is © The Royal Society of Chemistry 2008
Dalton Trans., 2008, 6242–6249 | 6243

View Online
for 16 h. The formed precipitate was filtered off, washed with
pentane and recrystallized from hot MeOH at-30 C to give 9
binuclear complex 7 crystallized with two acetate anions (three
positions, two of them are half occupied), one pentane (half
occupied) and one water molecule (occupancy 0.25, hydrogen
atoms not localized) in the asymmetric unit. The two half-occupied
acetate anions are disordered, and therefore their bond distances
and thermal values were constrained. Crystals of complex 11
contain two palladium complexes and one molecule of MeCN
in the asymmetric unit. In each of the two independent complex
molecules, one isopropyl group was found to be disordered over
two positions (occupancies 0.5). Further details on data collection
and refinement parameters are collected in Table 1.
◦
1
as colorless crystals (0.57 g, 27%). H NMR (DMSO-d
6
): d 9.60,
N), 4.71
), 2.83 (s, 3H, Cimi–CH ),
). C{ H} NMR
8
(
.06, 8.03, 7.99, 7.98 (5 ¥ s, 1H, Himi), 6.58 (s, 2H, NCH
2
3
2 ¥ sept, JHH = 6.4 Hz, 2H, CHMe
2
3
3
13
1
1
.50, 1.43 (2 ¥ d, JHH = 6.4 Hz, 6H, CH(CH
)
3 2
(
DMSO-d
6
): d 146.21, 137.03, 123.25, 122.47, 122.12, 119.60 (6 ¥
imi), 57.90 (NCH N), 53.64 (CHMe ), 51.39 (CHMe ), 22.93
), 22.60 (CH(CH ), 11.29 (Cimi–CH ). Elem. anal.
calcd. for C14 (502.18): C, 33.48; H, 4.82; N, 11.16. Found:
C
2
2
2
(
CH(CH
3
)
2
3
)
2
3
H
24
N
I
4 2
C, 33.57; H, 4.72; N, 11.02.
Synthesis of 11. To a solution of the diimidazolium salt 9
Results and discussion
(
(
0.50 g, 1.0 mmol) in DMSO (10 mL) was added [Pd(OAc)
0.22 g, 1.0 mmol) and the mixture was stirred at 50 C for 2 h
2
]
◦
◦
Synthesis of complexes
and at 110 C for another 3 h. After cooling, CH
2
Cl
O (100 mL). The formed precipitate
was collected by centrifugation and washed by repetitive addition
of CH Cl and subsequent precipitation with Et O to afford 11
as a yellow powder (0.42 g, 63%). An analytically pure sample
2
(20 mL)
was added followed by Et
2
The 2,2¢-disubstituted imidazolium salts 4a–d were prepared by
quaternization of the corresponding N-functionalized 2-methyl
2
2
2
or 2-phenyl imdazoles 3a–d with CH
with Pd(OAc)
I
. Subsequent palladation
2
2
14,17
2
according to established procedures
induced
1
was obtained by recrystallization from MeCN/Et
500 MHz, DMSO-d ): d 7.93 (s, 2H, Himi), 7.80, 7.73 (2 ¥ s, 1H,
imi), 6.46 (s, 2H, NCH N), 5.30 (br, 1H, CHMe ), 4.74 (sept,
), 3.1 (br s, 3H, Cimi–CH ), 1.44, 1.40
). C{ H} NMR (125 MHz,
2
O. H NMR
C–H bond activation and afforded the palladium complexes 2a–
d containing C4-bound dicarbene ligands as off-white air-stable
(
6
18
H
2
2
solids (Scheme 1). Yields are generally excellent (>90%), indicating
that competing side reactions are negligible. In particular, no
C–H bond activation of the methyl group in C2 position has
been observed with ligands 4a–c, thus contrasting the reactivity
of similar imidazolium salts towards silver(I) and iridium(III)
3
J
HH = 6.6 Hz, 2H, CHMe
2
3
3
13
1
(
2 ¥ d, JHH = 6.6 Hz, 6H, CH(CH
)
3 2
DMSO-d
imi), 118.02 (C–Pd), 59.65 (NCH
CHMe ), 21.98 (CH(CH ), 21.06 (CH(CH
). Elem. anal. calcd. for C14 Pd (734.49) ¥ 0.5 MeCN:
6
): d 144.34 (Cimi–Me), 125.30, 121.33, 119.28, 118.77 (4 ¥
N), 53.24 (br, CHMe ), 50.29
), 12.30 (br, Cimi
C
(
2
2
)
2
3
)
2
–
19
2
3
precursors. Since the reaction conditions used for preparing
CH
3
H
23
I
3
N
4
the dicarbene complexes 1 and 2 are very similar, it seems that
C, 23.86; H, 3.27; N, 8.35. Found: C, 24.35; H, 3.56; N, 8.17.
the pK values of the C2-bound proton in the precursor of 1
a
and of C4–H in 4 may be less separated than the theoretically
Crystal structure determinations†
20
predicted 8 pK
a
units (calculated pK 24 and 32, respectively).
a
Clearly however, functionalization of the imidazolium C2 position
is mandatory for directing the regioselectivity of C–H bond
activation to the C4 position.
Suitable single crystals were mounted on a Stoe Mark II-Imaging
Plate Diffractometer System (Stoe & Cie, 2002) equipped with
a graphite-monochromator. Data collection was performed at
17,21
◦
˚
-
100 C using Mo-Ka radiation (l = 0.71073 A) with a nominal
crystal to detector distance of 70 mm (for 2d, 7, and 11) and
35 mm (for 2a and 2c), respectively. All structures were solved
by direct methods using the program SHELXS-97 and refined
1
2
15
by full matrix least squares on F with SHELXL-97. The
hydrogen atoms were included in calculated positions and treated
as riding atoms using SHELXL-97 default parameters. All non-
hydrogen atoms were refined anisotropically, except for partically
occupied solvent atoms, which were refined isotropically. A semi-
empirical absorption correction was applied using MULscanABS
1
6
as implemented in PLATON03.
2 2 2
Scheme 1 Reagents and conditions: a) CH I , toluene, reflux; b) Pd(OAc) ,
DMSO, 50 C to 120 C.
◦
◦
The complex molecule in crystals of 2a adopts two different ori-
entations. As a consequence, all atom positions are half occupied.
In addition, one disordered acetonitrile molecule co-crystallized
with an occupancy factor for all atoms of 0.25. Complex 2d
crystallized with two complex molecules in the asymmetric unit.
Two butyl groups of the ligand in one of the complex molecules
are disordered. The involved carbon atoms were refined with
occupancies of 0.5 and their thermal values were constrained to be
equal. The co-crystallized pentane molecule is strongly disordered.
Successful cyclopalladation is indicated by the pertinent NMR
spectroscopic data of the complexes 2. Notably, the methylene
group interlinking the two metallated heterocycles appears as a
1
singlet in the H NMR spectrum (d
H
6.1). The magnetic equiv-
alence of the two CH
of the metallacycle. In the C NMR spectrum, the metal-bound
carbon appears around d 120–124, which is virtually at the same
2
protons indicates a flexible conformation
1
3
C
The SQUEEZE option in PLATON03 was used to allocate the
frequency as in the imidazolium precursors. This comparable
substituent effect of a hydrogen and a palladium(II) center on
the C4 chemical shift points to a relatively low carbene contri-
bution to the Pd–C bond in complexes 2. In the corresponding
3
˚
electron density corresponding to the pentane molecule (201 A
containing and about 32 electrons; one pentane molecule requires
4
2 electrons), which was included in all further calculations. The
6
244 | Dalton Trans., 2008, 6242–6249
This journal is © The Royal Society of Chemistry 2008

View Online
Table 1 Crystallographic data for complexes 2a, 2c, 2d, 7, and 11
2
a
2c
2d
7
11
Color, shape
Size/mm
Empirical formula
Orange rod
Light brown block
Orange rod
Yellow block
0.45 ¥ 0.40 ¥ 0.35
Orange needle
0.45 ¥ 0.25 ¥ 0.15
0.25 ¥ 0.10 ¥ 0.10
0.25 ¥ 0.20 ¥ 0.15
0.35 ¥ 0.30 ¥ 0.25
C
11
H
16
I
2
N
4
Pd·
C
27
H
32
I
2
N
4
Pd
C
0.25C
27
H
32
I
H
2
N
12
4
Pd·
C
0.5C
1010.32
173(2)
38
H
5
60
N
8
O
8
Pd
2
·
C
14
H
23
I
3
N
4
I
3
Pd·
0
.5CH
3
CN
5
H
12·0.25H
2
O
0.5CH
754.99
173(2)
3
CN
Fw
T/K
585.01
173(2)
Orthorhombic
772.77
173(2)
790.80
173(2)
Crystal system
Space group
Unit cell
a/A˚
b/A˚
c/A˚
Triclinic
–1 (No. 2)
Triclinic
–1 (No. 2)
Triclinic
–1 (No. 2)
Monoclinic
P2 /a (No. 14)
Pbam (No. 55)
P
P
P
1
13.7424(13)
18.4993(17)
6.9570(7)
90
90
90
1768.6(3)
4
2.197
4.536
20143, 1725
0.0539
9.8664(11)
12.8204(15)
15.0692(17)
16.8311(19)
94.572(13)
107.296(13)
104.040(13)
2971.1(6)
4
1.768
2.726
23096, 10901
0.0633
13.1967(10)
13.7943(11)
16.4100(12)
68.202(8)
69.115(9)
86.363(9)
2583.2(4)
2
1.299
0.747
20450, 9375
0.0425
13.9765(12)
9.9155(6)
33.885(3)
90
97.039(10)
90
4660.6(6)
8
2.152
12.3939(15)
12.8016(13)
66.710(8)
83.725(9)
78.530(9)
1408.3(3)
2
1.822
2.873
15991, 5004
0.0385
a/deg
b/deg
g /deg
V/A˚
3
Z
-3
D
calc/g cm
-
1
m/mm (Mo K
No. of total, unique reflns
a
)
4.776
31627, 9055
0.0450
R
int
Transmission range
No. paras, restr.
0.458–0.582
157, 2
0.0421, 0.1153
0.0481, 0.1178
1.247
0.619–0.678
315, 0
0.0262, 0.0570
0.0401, 0.0598
0.936
0.437–0.476
552, 12
0.0499, 0.1135
0.1108, 0.1261
0.748
0.776–0.825
529, 15
0.0496, 0.1470
0.0738, 0.1574
0.984
0.252–0.296
408, 8
0.0327, 0.0819
0.0530, 0.0989
0.909
a
a
b
b
R
R
1
wR
wR
2
for I > 2s(I)
for all I
1
2
GOF
Min/max resid density/e A˚ -
3
-0.713, 1.693
-0.592, 0.777
-1.423, 1.280
-0.826, 1.107
-1.861, 1.623
ꢀꢀF
| - |F
ꢀ/ꢀ
|F
|. wR
= [ꢀ
w(F
- F
) / )]
ꢀ(w(F
.
a
b
2
2
2
4
1/2
R
1
=
o
c
o
2
o
c
o
C2-bound dicarbene complexes 1 (d
induces typically a lowfield shift of around 20 ppm for the metal-
C
around 160), palladation
Table 2 Selected bond lengths ( A˚ ) and angles (deg) of complexes 2a, 2c,
and 2d
bound carbon.
2
d
2
a
2c
1.990(4)
Molecule 1
Molecule 2
Solid state structures
Pd1–C1
Pd1–C7
Pd1–I1
Pd1–I2
N2–C1
N3–C7
C1–C2
C6–C7
C1–Pd1–C7
I1–Pd1–I2
1.994(17)
2.014(17)
2.6558(12)
2.6835(12)
1.40(2)
1.39(2)
1.37(3)
1.35(2)
87.3(8)
91.81(4)
1.966(10)
1.990(9)
2.6747(10)
2.6619(11)
1.403(12)
1.407(12)
1.376(13)
1.366(12)
87.1(4)
1.966(10)
1.995(9)
2.6832(11)
2.6569(10)
1.393(11)
1.398(12)
1.368(12)
1.337(13)
86.4(4)
Further structural information was obtained by crystallographic
analysis of the C4-bound dicarbene palladium complexes. The
molecular structures of the complexes 2a, 2c, and 2d are depicted
in Fig. 2 and reveal several common features. The palladium center
is generally located in a distorted square-planar environment with
1.994(4)
2.6522(5)
2.6998(5)
1.401(5)
1.411(5)
1.358(5)
1.358(5)
86.68(15)
92.59(15)
28.3(3)
(
pseudo-)C2v symmetry. The bite angle of the chelating dicarbene
◦
is slightly smaller than 90 and hence well-suited for stabilizing
square-planar or octahedral coordination geometries.
93.83(4)
30.0(7)
28.4(7)
93.75(3)
37.2(7)
32.7(8)
N2–C1–Pd–C7 29.7(16)
N3–C7–Pd–C1 30.3(15)
29.4(3)
Table 2 lists the pertinent bond distances and angles of the new
structures of 2a, 2c, and 2d. It is particularly instructive to compare
2
2
these geometrical data with those available for the known C2-
bound dicarbene analogues 1. For example, the Pd–C bond length
is identical within standard deviations in all complexes, averaging
steric effects compensate a potentially stronger trans influence of
the C4-bound carbene. Such a hypothesis is further supported
by the different torsion of the heterocycles with respect to the
palladium square plane. In the C2-bound carbene complexes 1, the
˚
˚
to 1.98(2) A in C4-bound dicarbenes and 1.99(1) A in C2-bound
systems. Apparently, the Pd–C bond length is encoded by the size
of, and the hybridization within the metallacycle. Similarly the Pd–
I bond distances are statistically equal, averaging to 2.673(15) A
in complexes 2 and to 2.658(11) A in the C2-bound complexes 1.
This Pd–I bond length analysis may hence point to a virtually
identical trans influence of the two NHC isomers, although steric
factors need to be considered as well. Repulsion of the halide from
the metal coordination sphere due to the presence of substituents
R at nitrogen (Scheme 1) is expected to be more pronounced in the
◦
22
heterocycles are twisted 45–62 out of the coordination plane. In
the C4-bound carbene complexes, which comprise only a hydrogen
rather than an alkyl group at the a-positioned atom, these dihedral
˚
◦
˚
angles are significantly smaller (N–C–Pd–C between 28 and
◦
41 ). A further difference concerns the heterocyclic C–C bond.
In the C4-bound carbene complexes 2, the average of this bond is
˚
1.368(18) A and represents a conjugated double bond, while in the
C2-bound ligands, the short 1.336(11) distance rather indicates
a localized olefinic bond. A similar trend has been observed in
22f
7a,11
C2-bound carbene than in the C4-bound ligand. Possibly, these
related iridium and rhodium chemistry
and suggests a different
This journal is © The Royal Society of Chemistry 2008
Dalton Trans., 2008, 6242–6249 | 6245

View Online
Scheme 2 Reagents and conditions: a) AgBF
4
, MeCN; b) DMSO or
MeOH; c) AgOAc, DMSO.
24
as ligand to silver presumably originates from the strong donor
ability of the C4-bound dicarbene ligand.
Silver decoordination and formation of the bissolvento complex
6
has been achieved by dissolving complex 5 in DMSO or
MeOH. These solvents appear to be nucleophilic enough to
successfully compete with the palladium center for silver coor-
dination, while MeCN is not. Accordingly, the direct synthesis
of the bissolvento complex 6 from the neutral complex 2 is best
performed by using 3 mol equiv. AgBF
4
followed by a work-up
that includes a MeOH/Et O precipitation sequence.
2
The different reactivity of the palladium complexes 1 and 2
+
towards Ag coordination provides a qualitative probe for the
donor ability of the palladium center relative to various solvents,
increasing in the order Pd(C2-dicarbene) < MeCN < Pd(C4-
dicarbene) < MeOH, DMSO. From this sequence, it is obvious
that the C4-bonding mode increases the electron density at the
metal center as compared to C2-bound carbene ligands.
Fig. 2 ORTEP representation of complexes 2a (a; 50% probability),
c (b; 50% probability), and 2d (c; 30% probability). All hydrogen
atoms, co-crystallized solvent molecules, and the second molecule in the
asymmetric unit of 2d have been omitted for clarity. Only one of the two
disordered orientations of 2a is shown.
2
When using different silver(I) precursors, charge compensation
at the palladium may occur along other pathways. For example
with 2 mol equiv. Ag(OAc), formation of the dimeric species 7 was
observed. Crystallographic analysis of complex 7 shows an open
bonding situation within the isomeric heterocycles in complexes 1
and 2. This conclusion corroborates the NMR analyses in solution
˚
book type structure (Fig. 3). The Pd ◊ ◊ ◊ Pd distance is 2.9441(6) A
and hence comparable to related acetate-bridged palladium(II)
(
see above).
25
dimers, though significantly shorter than the sum of the van
26
˚
der Waals radii (3.30 A). Notably under identical reaction
conditions, the analogous C2-bound dicarbene complexes provide
Impact of C4 bonding on reactivity
1
27
In the presence of silver(I) ions, the C4-bound carbene complexes
monomeric complexes containing two k -bound acetates. Anion
2
exhibit unique reactivity patterns. Most metal-halide complexes,
exchange was recently reported to yield a dimeric structure that is
28
including the C2-bound dicarbene complexes 1, undergo ligand
exchange and yield the corresponding cationic solvento complexes.
closely related to that of 7. However, the Pd ◊ ◊ ◊ Pd distance in the
˚
C2-bound analogue (3.1683(6) A) is substantially larger than that
In contrast, exposure of complex 2 in MeCN to AgBF
4
does
of 7.
not only induce the expected halide abstraction from the metal
coordination sphere, but simultaneously, the adduct 5 is formed
Complex 7 is only sparingly soluble in most organic solvents. Its
1
H NMR spectrum in DMSO-d
6
solution shows unusually broad
◦
(
Scheme 2). Crystallographic analysis unequivocally revealed the
features which do not sharpen upon heating up to 100 C. This
may point to a geometrically constrained structure in solution and
hence to a conservation of the rigid geometry as determined by
solid state analysis. This is further supported by the observation of
formation of the five-coordinate palladium complex 5 comprising
2
3
˚
unusually short Ag ◊ ◊ ◊ Pd contacts of 2.8701(6) A. The molecular
14
structure of this adduct has been preliminarily reported. Bonding
analysis identifies the palladium center as a Lewis basic site,
despite its formally dicationic charge. The behavior of palladium
two acetate signals in 1:1 molar ratio at d 1.93 and 2.10. The low
C
shift of the latter resonance is indicative for cationic palladium
6
246 | Dalton Trans., 2008, 6242–6249
This journal is © The Royal Society of Chemistry 2008

View Online
of the diimidazolium salt with Pd(OAc)
kinetically controlled.
2
, is irreversible and hence
Acidolysis of the Pd–C bond in complex 2 may involve forma-
tion of an intermediate palladium-hydrogen adduct in analogy
to the reaction of 2 with AgBF (cf. adduct 5, Scheme 2). A
4
subsequent hydrogen shift from palladium to carbon would induce
Pd–C bond cleavage. Attempts to detect the surmised metal-
bound hydrogen by in situ NMR spectroscopy have failed thus
far. Participation of the metal center in the acidolysis reaction as
opposed to direct attack at the metal-bound carbon is supported,
however, by the fact that only one metal–carbene bond of 2
is broken, while the monocarbene complex 8 is acid-stable.
Obviously, forthcoming studies need to address also other factors,
such as the electrostatic protection of the metal center in 8 by the
dangeling cationic imidazolium residue.
Notably, the bissolvento complex 6 undergoes similar acid-
mediated Pd–C bond cleavage and forms a monocarbene complex
related to 8, while the C2-bound dicarbene complexes 1 are
Fig. 3 Thermal ellipsoid plot (50% probability) of the dimeric complex
7
2
2
. Selected bond lengths ( A˚ ): Pd1–C1 1.951(5), Pd1–C7 1.947(6), Pd1–O1
completely stable towards strong acids (H
2
SO ), even upon heating
4
.098(4), Pd1–O3 2.087(4), Pd2–C16 1.955(5), Pd1–C7 1.932(6), Pd2–O2
.089(4), Pd2–O4 2.098(3), Pd1 ◊ ◊ ◊ Pd2 2.9441(6); selected bond angles
◦
to 80 C for prolonged time. This different stability properties
indicate that the carbene bonding mode influences the reactivity of
the coordinated metal center considerably more than the ancillary
ligands do, in particular when using polar solvents. The reactivity
(
deg): C1–Pd1–C7 87.4(2), O1–Pd1–O3 90.68(17), C16–Pd2–C22 87.4(2),
O2–Pd2–O4 90.68(16).
7,11,14
of complex 2 but not 1 with acids lends further
support to
2
8,29
13
acetate complexes.
The C NMR signals are better resolved
132.
the notion that C4-bound carbenes are ligands with exceptionally
strong donor ability.
and display a resonance for the metal-bound carbon at d
C
This frequency corresponds to a downfield shift as compared to
the neutral diiodide precursor 2b and may be attributed in parts
-
-
to the different donor power of OAc and I .
Mechanistic aspects of C4 metallation
+
The ability of complexes 2a–d to bind Ag ions reflects the
Lewis basic character of the palladium center. Likewise, these
complexes also display Brønsted basic properties. Addition of
The stability of complexes 2 and 8 towards HOAc (see above)
implies that the palladium-carbene bond formation is an irre-
30
versible processes. This resistence towards Pd–C bond cleavage
excludes chelation as a driving force for the first metallation,
i.e. the formation of a product similar to 8. In this light, the
regioselectivity of the palladation is remarkable, also because
of the similar acidity of the C4–H and C5–H protons in the
imidazolium ligand precursor salt 4. Hence a pre-equilibrium
must exist, in which the metal center is directed towards the
C4–H bond. Steric arguments would fall short, given the high
yield of cyclopalladated product is independent of the wingtip
group R (e.g. R = Me, Mes). Herrmann and coworkers observed
H
2
SO (5 mol equiv.) to complex 2 induced rapid palladium–
4
carbene bond cleavage. Spectroscopic analysis of the product
indicates the cleavage of one Pd–C bond only and formation of
the monocarbene complex 8 (eqn (1), X presumably HSO ). Three
4
different proton signals are visible in the 7–7.5 ppm range, two of
3
which are mutually coupled ( JHH = 2.2. Hz) and were therefore
assigned to the imidazolium residue of 8. Likewise, two sets of
methyl and isopropyl groups appeared in a 1:1 ratio.
2
-
the presence of the dianionic palladate [PdI
2
(OAc)
2
]
prior to
22b
palladation of related diimidazolium salts at the C2 position.
Similar palladate formation and concomitant ion pairing with
the diimidazolium dication (C, Fig. 4) may provide a rationale
for the selective activation of the (inner) C4–H rather than the
(
1)
31
more remote C5–H bond from C. The relevance of a putative
ion pair complex is further strengthened by the fact that all our
-
palladation attempts using diimidazolium salts containing BF
4
In MeCN or DMSO, acidolysis is fast at RT (<15 min). No
counteranions were unsuccessful. C–H bond activation from the
ion pair C, perhaps mediated by a bridging coordination mode
of the acetate ligands, affords the monocarbene intermediate E,
or D if the C2 position is not substituted (Fig. 4). An analog of
further reaction was observed, even when keeping complex 8 in
◦
the presence of excess H
the H SO
equiv. H SO
2
SO
4
at 80 C for several hours. Lowering
2
4
concentrations led to partial acidolysis only, e.g. 2.2 mol
22b
2
4
gave an approximate 2:1 mixture of 8 and 2. With
D has been isolated and structurally characterized (R = tBu).
Cyclometallation from D or E may then occur along established
◦
weaker acids such as H
3
PO
4
, heating to 85 C is required in order
32
to initiate Pd–C bond cleavage. No reaction occurs with very weak
acids (HOAc), even at elevated temperature. The stability of the
Pd–Ccarbene bonds in complexes 2 and 8 towards HOAc suggests
that the opposite process of the acidolysis, that is the metallation
pathways.
A potential driving force for the cyclometallation step has been
deduced from palladation experiments using the dissymmetric
33
diimidazolium salt 9 (Scheme 3). Metallation of 9 under the
This journal is © The Royal Society of Chemistry 2008
Dalton Trans., 2008, 6242–6249 | 6247

View Online
Fig. 4 Proposed intermediates for the palladation of imidazolium cations
via C4–H bond activation: ion pair complex C preceding the first C–Pd
bond formation with high regioselectivity for C4–H bond activation;
macrocyclic complexes D and E for inducing cyclometallation at the C2
and the C4 position, respectively.
Fig. 5 Thermal ellipsoid plot (50% probability) of the monocarbene
complex 11 (50% probability, only one of the two crystallographically in-
dependent molecules shown). Selected bond lengths ( A˚ ): Pd1–C1 1.968(6),
Pd1–I1 2.6302(7), Pd1–I2 2.6751(7), Pd1–I3 2.6162(7), C2–C3 1.328(10),
C5–C6 1.340(10); selected bond angles (deg): C1–Pd1–I1 87.89(18),
C1–Pd1–I2 177.37(17), C1–Pd1–I3 85.59(18), I1–Pd1–I3 172.33(2).
Scheme 3 Reagents and conditions: a) CH
2
I
2
, toluene, reflux, then
◦
◦
recrystallization; b) Pd(OAc)
2
, DMSO, 50 C to 120 C; c) NaI, acetone.
erties have been characterized. All evidence suggests that the
palladium center is a stronger Lewis (and probably also Brønsted)
base when coordinated by C4-bound dicarbene ligands as opposed
to C2-bound analogues. This may be attributed to a more
pronounced zwitterionic bonding of C4-carbenes, comprising a
anionic vinyl-type fragment for metal coordination and a cationic
NCN amidinium residue for intramolecular charge compensation.
Irrespective of the exact bond description, the substantially higher
donor power of C4-bound carbenes is expected to have a strong
impact on the catalytic activity of the coordinated metal center
conditions applied previously for the palladation of 4 gave the
monocarbene complex 10 as the major product. Upon work-up,
anion metathesis took place, leading to the formation of complex
11 comprising three iodide ligands at palladium. Complex 11 was
unambiguously characterized by single crystal X-ray diffraction
1
(
Fig. 5). In the H NMR spectrum, the presence of three different
aromatic signals at d 7.93, 7.80, and 7.73 (2:1:1 ratio) was
H
diagnostic. No cyclometallation could be induced by heating either
complex 10 or the in situ formed complex 11 for prolonged periods
of time, and instead only slow decomposition was observed.
Apparently, C4–H bond activation is considerably more difficult
in complex 10 and 11 than in the related intermediate E.
and may disclose new applications beyond H activation and
2
hydrogenation. Studies along these lines are currently in progress
in our laboratories.
The bonding mode of the carbene is the most notable difference
between the stable complex 10 and the proposed intermediate E.
Hence, cyclometallation via C–H bond activation from monocar-
bene complexes seems to be crucially depending on i) the electron
density at palladium and ii) the acidity of the C–H proton. Thus,
Acknowledgements
We thank Mireille Elsen for synthetic work, Felix Fehr for
NMR assistance, and the Swiss National Science Foundation for
financial support. M. A. gratefully acknowledges an Assistant
Professorship of the Alfred Werner Foundation.
20
activation of a less acidic C4–H bond requires an increased
basicity of the acetate ligand, which is apparently generated
in a C4-bound (cf. E) but not in a C2-bound monocarbene
intermediate such as 10. In contrast C2–H bond activation occurs
also with less basic acetates as in the monocarbene intermediate
D. Such a model is in line with the concept of C4-bound carbenes
as substantially stronger donors than C2-bound carbenes. In
addition, it may rationalize the similar conditions required for
the preparation of the cyclopalladated complexes 2 and 1. Clearly,
further studies are warranted to verify and refine this model.
Notes and references
1
(a) D. Bourissou, O. Guerret, F. Gabbai and G. Bertrand, Chem. Rev.,
000, 100, 39; (b) V. Cesar, S. Bellemin-Laponnaz and L. H. Gade,
2
Chem. Soc. Rev., 2004, 33, 619; (c) F. E. Hahn and M. C. Jahnke,
Angew. Chem. Int. Ed., 2008, 47, 3122 For most recent highlights in
this area, see; (d) O. Kaufhold and F. E. Hahn, Angew. Chem. Int. Ed.,
2
008, 47, 4057.
2
3
(a) W. A. Herrmann, Angew. Chem. Int. Ed., 2002, 41, 1290; (b) N-
heterocyclic carbenes in synthesis, ed. S. P. NolanWiley-VCH, Wein-
heim, 2006; (c) A. T. Normand and K. J. Cavell, Eur. J. Inorg. Chem.,
Conclusions
2
008, 2781.
Palladium complexes comprising C4-bound imidazolium-derived
dicarbene ligands have been prepared and their reactivity prop-
(a) T. Sajoto, P. I. Djurovich, A. Tamayo, M. Yousufuddin, R. Bau,
M. E. Thompson, R. J. Holmes and S. R. Forrest, Inorg. Chem., 2005,
6
248 | Dalton Trans., 2008, 6242–6249
This journal is © The Royal Society of Chemistry 2008

View Online
4
4, 7992; (b) D. M. Khramov, A. J. Boydston and C. W. Bielawski,
B. W. Skelton, A. H. White and C. C. Williams, J. Chem. Soc. Dalton
Trans., 2001, 111; (c) J. A. Loch, M. Albrecht, E. Peris, J. Mata, J. W.
Faller and R. H. Crabtree, Organometallics, 2002, 21, 700.
Angew. Chem. Int. Ed., 2006, 45, 6186; (c) L. Mercs, A. Neels and M.
Albrecht, Dalton Trans., 2008, DOI: 10.1039/b809721f.
4
5
(a) A. A. Danopoulos, N. Tsoureas, J. C. Green and M. B. Hursthouse,
Chem. Commun., 2003, 756 Notably, the metal-carbene bond in NHC
complexes may participate in reactions and may not be innocent. For
examples, see; (b) K. J. Cavell and D. S. McGuinness, Coord. Chem.
Rev., 2004, 248, 671; (c) S. Burling, M. F. Mahon, R. E. Powell, M. K.
Whittlesey and J. M. Williams, J. Am. Chem. Soc., 2006, 128, 13702;
18 The metal-bound carbons in complex 2a should formally be labeled as
C5 carbons. For reasons of consistency, however, we use here only the
atom labeling scheme depicted in Scheme 1.
19 (a) A. R. Chianese, B. M. Zeglis and R. H. Crabtree, Chem. Commun.,
2004, 2176; (b) M. Viciano, M. Feliz, R. Corberan, J. A. Mata, E. Clot
and E. Peris, Organometallics, 2007, 26, 5304.
(
d) B. K. Shaw, T. Steinke and M. D. Fryzuk, XXIII International
20 (a) A. M. Magill and B. F. Yates, Aust. J. Chem., 2004, 57, 1205; (b) R. W.
Alder, P. R. Allen and S. J. Williams, J. Chem. Soc. Chem. Commun.,
Conference on Organometallic Chemistry, abstract P756.
(a) R. W. Alder, M. E. Blake, C. Bortolotti, S. Bufali, C. P. Butts, E.
Linehan, J. M. Oliva, A. G. Orpen and M. J. Quayle, Chem. Commun.,
a
1995, 1267 For experimentally determined pK of the C2–H proton,
see; (c) T. L. Amyes, S. T. Diver, J. P. Richard, F. M. Rivas and K. Toth,
J. Am. Chem. Soc., 2004, 126, 4366; (d) A. M. Magill, K. J. Cavell and
B. F. Yates, J. Am. Chem. Soc., 2004, 126, 8717.
1
999, 241; (b) S. Gr u¨ ndemann, A. Kovacevic, M. Albrecht, J. W.
Faller and R. H. Crabtree, Chem. Commun., 2001, 2274; (c) C. C.
Scarborough, M. J. W. Grady, I. A. Guzei, B. A. Gandhi, E. E. Bunel
and S. S. Stahl, Angew. Chem. Int. Ed., 2005, 44, 5269; (d) V. Lavallo, Y.
Canac, C. Pr a¨ sang, B. Donnadieu and G. Bertrand, Angew. Chem. Int.
Ed., 2005, 44, 5705; (e) V. Lavallo, Y. Canac, A. DeHope, B. Donnadieu
and G. Bertrand, Angew. Chem. Int. Ed., 2005, 44, 7236; (f) S. Gomez-
Bujedo, M. Alcarazo, C. Pichon, E. Alvarez, R. Fernandez and J. M.
Lassaletta, Chem. Commun., 2007, 1180; (g) Y. Han, H. V. Huynh and
G. K. Tan, Organometallics, 2007, 26, 6581; (h) H. G. Raubenheimer
and S. J. Cronje, Dalton Trans., 2008, 1265; (i) M. A. Esteruelas, F. J.
Fernandez-Alvarez and E. Onate, J. Am. Chem. Soc., 2006, 128, 13044;
21 (a) D. S. McGuinness, K. J. Cavell, B. F. Yates, B. W. Skelton and A. H.
White, J. Am. Chem. Soc., 2001, 123, 8317 The same conclusion has
been drawn for palladation via C–H bond oxidative addition: see ref. 9
and; (b) S. Gr u¨ ndemann, M. Albrecht, A. Kovacevic, J. W. Faller and
R. H. Crabtree, J. Chem. Soc., Dalton Trans., 2002, 2163.
22 (a) W. A. Herrmann, C.-P. Reisinger and M. Spiegler, J. Organomet.
Chem., 1998, 557, 93; (b) W. A. Herrmann, J. Schwarz and M. G.
Gardiner, Organometallics, 1999, 18, 4082; (c) F. E. Hahn and M. Foth,
J. Organomet. Chem., 1999, 585, 241; (d) J. Schwarz, V. P. W. B o¨ hm,
M. G. Gardiner, M. Grosche, W. A. Herrmann, W. Hieringer and G.
Raudaschl-Sieber, Chem. Eur. J., 2000, 6, 1773; (e) S. Ahrens, A. Zeller,
M. Taige and T. Strassner, Organometallics, 2006, 25, 5409; (f) M.
Heckenroth, A. Neels, H. Stoeckli-Evans and M. Albrecht, Inorg. Chim.
Acta, 2006, 359, 1929; (g) M. Nonnenmacher, D. Kunz, F. Rominger
and T. Oeser, J. Organomet. Chem., 2007, 692, 2554.
(
j) E. Alvarez, S. Conejero, M. Paneque, A. Petronilho, M. L. Poveda,
O. Serrano and E. Carmona, J. Am. Chem. Soc., 2006, 128, 13060;
k) C. E. Ellul, M. F. Mahon, O. Saker and M. K. Whittlesey, Angew.
(
Chem. Int. Ed., 2007, 46, 6343; (l) M. Iglesias, D. J. Beetstra, A. Stasch,
P. N. Horton, M. B. Hursthouse, S. J. Coles, K. J. Cavell, A. Dervisi and
I. A. Fallis, Organometallics, 2007, 26, 4800For different engineering
of the basicity of carbenes, see:; (m) D. M. Khramov, V. M. Lynch and
C. W. Bielwaski, Organometallics, 2007, 26, 6042; (n) A. F u¨ rstner, M.
Alcarazo, H. Krause and C. W. Lehmann, J. Am. Chem. Soc., 2007,
23 (a) Y. Liu, K. H. Lee, J. J. Vittal and T. S. A. Hor, J. Chem. Soc., Dalton
Trans., 2002, 2747; (b) J. Forni e´ s and A. Martin, in Metal Clusters in
Chemistry, ed. P. Braunstein, L. A. Oro and P. R. Raithby, Wiley-VCH,
Weinheim, 1999, vol. 1, p. 417.
1
29, 12676.
24 For a review on the complex-as-ligand concept, see E. Pardo, R. Ruiz-
Garc ´ı a, J. Cano, X. Ottenwaelder, R. Lescou e¨ zec, Y. Journaux, F. Lloret
and M. Julve, Dalton Trans., 2008, 2780.
6
7
S. Gr u¨ ndemann, A. Kovacevic, M. Albrecht, J. W. Faller and R. H.
Crabtree, Chem. Commun., 2001, 2274.
(a) L. Yang, A. Kr u¨ ger, A. Neels and M. Albrecht, Organometallics,
25 A CCDC search on acetate-bridged palladium dimers returned 130 hits
2
008, 27, 3161; (b) M. Albrecht, Chem. Commun., 2008, 3601.
with an average Pd ◊ ◊ ◊ Pd distance of 2.991(11) A˚ . About 50% of the
8
9
P. L. Arnold and S. Pearson, Coord. Chem. Rev., 2007, 251, 596.
D. Bacciu, K. J. Cavell, I. A. Fallis and L.-I. Ooi, Angew. Chem. Int.
Ed., 2005, 44, 5282.
found complexes display a Pd ◊ ◊ ◊ Pd distance in the 2.82–2.96 A˚ range.
26 T. W. Hambley, Inorg. Chem., 1998, 37, 3767.
27 W. A. Herrmann, J. Schwarz, M. G. Gardiner and M. Spiegler,
J. Organomet. Chem., 1999, 575, 80.
1
1
0 E. Kluser, A. Neels and M. Albrecht, Chem. Commun., 2006, 4495.
1 (a) A. R. Chianese, A. Kovacevic, B. M. Zeglis, J. W. Faller and R. H.
Crabtree, Organometallics, 2004, 23, 2461; (b) G. Song, Y. Zhang and
X. Li, Organometallics, 2008, 27, 1936.
28 J. C. Slootweg and P. Chen, Organometallics, 2006, 25, 5863.
29 N. Thirupathi, D. Amoroso, A. Bell and J. D. Protasiewicz,
Organometallics, 2005, 24, 4099.
1
2 (a) S. Gr u¨ ndemann, A. Kovacevic, M. Albrecht, J. W. Faller and R. H.
Crabtree, J. Am. Chem. Soc., 2002, 124, 10473 In some cases, C2
protection is not necessary, see; (b) L. N. Appelhans, D. Zuccaccia,
A. Kovacevic, A. R. Chianese, J. R. Miecznikowski, A. Macchioni,
E. Clot, O. Eisenstein and R. H. Crabtree, J. Am. Chem. Soc., 2005,
30 Similar conclusions have been drawn from different experiments, see
ref. 22f and C. K o¨ cher and W. A. Herrmann, J. Organomet. Chem.,
1997, 532, 261.
31 (a) J. Dehand and M. Pfeffer, Coord. Chem. Rev., 1976, 18, 327 Based on
our experimental data, we cannot rule out a transient dearomatization
of the imidazolium heterocycle and nitrogen coordination to the metal
center as an initial process for carbene metallation. Heteroatom-
assisted C–H bond activation is well-known, see for examples; (b) A. D.
Ryabov, Chem. Rev., 1990, 90, 403; (c) M. Albrecht and G. van Koten,
Angew. Chem. Int. Ed., 2001, 40, 3750; (d) V. V. Dunina and O. N.
Gorunova, Russ. Chem. Rev., 2004, 73, 309.
1
27, 16299; (c) M. Baya, B. Eguillor, M. A. Esteruelas, M. Loivan
and E. Onate, Organometallics, 2007, 26, 6556; (d) B. Eguillor, M. A.
Esteruelas, M. Olivan and M. Puerta, Organometallics, 2008, 27,
4
45.
3 (a) G. Song, X. Wang, Y. Li and X. Li, Organometallics, 2008, 27, 1187;
b) H. Lebel, M. K. Janes, A. B. Charette and S. P. Nolan, J. Am. Chem.
1
(
Soc., 2004, 126, 5046; (c) L. Campeau, P. Thansandote and K. Fagnou,
Org. Lett., 2005, 7, 1857; (d) M. Alcarazo, S. J. Roseblade, A. R. Cowley,
R. Fernandez, J. M. Brown and J. M. Lassaletta, J. Am. Chem. Soc.,
32 (a) D. L. Davies, S. M. A. Donald and S. A. Macgregor, J. Am. Chem.
Soc., 2005, 127, 13754; (b) M. Albrecht, in Palladacycles, ed. J. Dupont
and M. Pfeffer, Wiley-VCH, Weinheim, 2008, p.13.
2
005, 127, 3290.
33 The ligand precursor 9 was prepared by quaternizing a 1:1 mixture
1
4 M. Heckenroth, E. Kluser, A. Neels and M. Albrecht, Angew. Chem.
Int. Ed., 2007, 46, 6293.
5 G. M. Sheldrick, Acta Crystallogr. A, 2008, 64, 112.
6 A. L. Spek, J. Appl. Crystallogr., 2003, 36, 7.
7 (a) M. G. Gardiner, W. A. Herrmann, C.-P. Reisinger, J. Schwarz and
M. Spiegler, J. Organomet. Chem., 1999, 572, 239; (b) M. V. Baker,
2 2
of N-isopropyl imidazole and the imidazole 1b with CH I . This
reaction produced, according to NMR integration, a mixture of the
two homocoupled salts and the dissymmetric diimidazolium salt 9 in
the statistically expected 1:1:2 ratio. After crystallization from MeOH,
pure 9 was isolated in 25% yield, while a 1:1:1 mixture remained in
solution.
1
1
1
This journal is © The Royal Society of Chemistry 2008
Dalton Trans., 2008, 6242–6249 | 6249