[1,2]-rearrangement of imino-N-heterocyclic carbenes - Synthesis and structures of chelating iminoimidazole Pd and Ni complexes

Article abstract of DOI:10.1002/ejic.200400070

Imidazolium and benzimidazolium salts with an N-iminoyl and an N′-alkyl group are potential precursors for new bidentate [N,C] ligands which are of interest as novel steering ligands for applications of their metal complexes in homogeneous catalysis. Deprotonation of these azolium salts with potassium hydride does indeed occur, however, the initially formed imino-N-heterocyclic carbenes rearrange spontaneously under migration of the N-iminoyl group from the nitrogen to the former carbene carbon with the formation of 2-iminoyl(benz)imidazoles. These rearranged compounds are new [N,N] ligand systems, which can also be synthesized by aluminum-assisted condensation of anilines with the corresponding 2-acylimidazoles. Palladium and nickel [N,N] complexes have been prepared, characterized by single-crystal X-ray analysis, and briefly evaluated for their catalytic performance in ethylene polymerizations and in Suzuki cross-coupling reactions. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejic.200400070

FULL PAPER
[1,2]-Rearrangement of Imino-N-heterocyclic Carbenes ؊ Synthesis and
Structures of Chelating Iminoimidazole Pd and Ni Complexes
Georg Steiner,[a] Alexander Krajete,[a] Holger Kopacka,[a] Karl-Hans Ongania,[b]
Klaus Wurst,[a] Peter Preishuber-Pflügl,[c] and Benno Bildstein*[a]
Keywords: N ligands / Carbenes / Chelates / Transition metals / Rearrangement
Imidazolium and benzimidazolium salts with an N-iminoyl
and an NЈ-alkyl group are potential precursors for new bi-
dentate [N,C] ligands which are of interest as novel steering
new [N,N] ligand systems, which can also be synthesized by
aluminum-assisted condensation of anilines with the corres-
ponding 2-acylimidazoles. Palladium and nickel [N,N] com-
ligands for applications of their metal complexes in homo- plexes have been prepared, characterized by single-crystal
geneous catalysis. Deprotonation of these azolium salts with
potassium hydride does indeed occur, however, the initially
formed imino-N-heterocyclic carbenes rearrange spontan-
eously under migration of the N-iminoyl group from the ni-
trogen to the former carbene carbon with the formation of 2-
iminoyl(benz)imidazoles. These rearranged compounds are
X-ray analysis, and briefly evaluated for their catalytic per-
formance in ethylene polymerizations and in Suzuki cross-
coupling reactions.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
Introduction
[C,N,C] metal complexes. In addition to the NHC donor, in
all reported [N,C] ligand systems the N-donor is part of a
planar heterocycle, i.e. pyridine and oxazoline, respectively.
To the best of our knowledge, no other N-donor functional-
ities have been investigated as components of heterotopic
[N,C] ligands for potential catalytic applications of their
metal complexes, except for one very recent example,[10]
published during the preparation of this manuscript.
Soon after Arduengo and co-workers[1] isolated the first
stable imino-N-heterocyclic carbene (NHC) in 1991, the
chemistry of these nucleophilic species attracted the interest
of many chemists due to their very rich coordination chem-
istry with almost all elements of the periodic table, and due
to the possibility of employing modified and optimized
NHC’s as steering ligands in metal complexes with catalytic
applications, i.e. for carbon-carbon cross-coupling reac-
tions, olefin metathesis, and other reactions.[2] From a his-
toric perspective,[3] metal complexes of simple (in situ gen-
erated) NHC’s have been studied for over 40 years by the
groups of Wanzlick,[4] Lappert,[5] Raubenheimer,[6] Öfele
and Herrmann.[2b][2c]
The most common N-heterocyclic carbenes are imidazol-
2-ylenes containing various N-substituents, usually simple
alkyl or aryl groups.[2] In recent years tailor-made hetero-
topic chelating NHCs containing an additional donor
functionality gained more and more importance, due to the
general focus on applications of NHC metal complexes in
catalysis.[2b] Chart 1 gives an overview of selected, recent
examples from the current literature[7Ϫ9] of [N,C] and
Conceptually, our motivation for the work reported here
is based on the following considerations: We have a general
interest in the development of new and improved olefin
polymerization catalysts.[11] Up to now, however, there are
no NHC metal complexes as olefin polymerization precata-
lysts, only oligmerization catalysts have been reported.[12]
In contrast, ring opening olefin metathesis polymerization
(ROMP) is one of the most prominent uses for NHC ru-
thenium complexes.[13,14] The NHC ligands reported in the
literature thus far are not suitable as steering ligands for
late transition metal olefin polymerization catalysts based
on iron or nickel,[15] because the planar N-donor (e.g. pyri-
dine, compare Scheme 1) prevents the axial shielding re-
sulting in suppressed chain termination pathways at the ac-
tive metal site. The most active non-cyclopentadienyl late
transition metal polymerization catalysts are [N,N,N] bis-
(imino)pyridine iron and [N,N] bisimine nickel complexes
with peripheral sterically demanding N-aryl substituents
(Scheme 2).[15] We therefore anticipate that [N,C] and
[N,C,N] ligand systems where one N-donor is replaced by
an NHC moiety will be very promising novel ligands for
[a]
Institute of General, Inorganic and Theoretical Chemistry,
University of Innsbruck,
Innrain 52a, 6020 Innsbruck, Austria
[b]
Institute of Organic Chemistry, University of Innsbruck,
Innrain 52a, 6020 Innsbruck, Austria
[c]
BASF Aktiengesellschaft,
67056 Ludwigshafen, Germany
Eur. J. Inorg. Chem. 2004, 2827Ϫ2836
DOI: 10.1002/ejic.200400070
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2827

B. Bildstein et al.
FULL PAPER
new olefin polymerization precatalysts. In this contribution,
In other work[11b] we found that iminoyl chlorides 1[17]
we summarize our results on the chemistry of such imino- are very convenient and useful electrophilic reagents which
NHCs and of their metal complexes.
react quantitatively with N-H nucleophiles. They also con-
vert imidazole and benzimidazole to the corresponding N-
iminoyl heterocycles 2a and 2b in very good yield. Both
compounds are yellow air-stable organic materials with
properties in line with their structure (for details see Exp.
Sect.). Spectroscopically, the imine group is clearly evident
from a strong stretching vibration in the IR spectrum (2a:
νCϭN ϭ 1652 cmϪ1; 2b: νCϭN ϭ 1651 cmϪ1) and from a
low-field 13C chemical shift of the imine carbon (2a: δ ϭ
148.6 ppm; 2b: δ ϭ 149.9 ppm).
Monoiminoylazoles 2a and 2b already contain one steri-
cally bulky N-arylated donor Ϫ in this case 2,6-diisopro-
pylphenyl Ϫ in agreement with the design concept outlined
in the introduction. To obtain access to azolium precursors
for [N,C] and [N,C,N] ligands according to Scheme 2, either
a simple alkylation step or a second acylation with iminoyl
chloride is all that is formally needed.
The attempt at forming N,NЈ-bisiminoylazolium salts by
reaction of 2a and 2b with a second equivalent of iminoyl
chloride 1 gave inconclusive results: a reaction did occur
but the very air-sensitive product could not be characterized
and its follow-up chemistry ruled out the formation of an
azolium precursor to a [N,C,N] ligand. After half a year of
trying to improve reaction conditions and workup pro-
cedures we finally concluded that such tridentate ligand sys-
tems are unfortunately not accessible due to fragmentation
of the precursor imidazolium species.
On the other hand, the use of azolium salts as the re-
quired starting materials for [N,C] ligands produces no
major problems: alkylation at the NЈ-position of 2a and 2b
is indeed possible using methyl triflate as a highly electro-
philic reagent, affording N-iminoyl-NЈ-methyl-azolium salts
3a and 3b, respectively, in high yield (3a: 100%; 3b: 77%).
Other, more common alkylating agents like iodomethane
yielded no reaction at all under a variety of experimental
conditions, including forced conditions like sealed-tube re-
actions using excess methyl iodide as solvent and reagent.
Both azolium salts are yellow, slightly air-sensitive com-
pounds of surprising stability. Key spectroscopic data in-
clude the observation of molecular ions of the cations in
the positive mode FAB mass spectra (3a: m/z ϭ 346; 3b:
m/z ϭ 396), detection of the low-field NMR signal of the
azolium hydrogen in the 2-position of the heterocycle [3a:
δ(1H) ϭ 9.09 ppm; 3b: δ(1H) ϭ 9.50 ppm], and observation
of the N-methyl NMR signals [3a: δ(1H) ϭ 4.08 ppm,
δ(13C) ϭ 37.6 ppm; 3b: δ(1H) ϭ 4.25 ppm, δ(13C) ϭ
34.8 ppm]. Interestingly, the 13C chemical shifts of the imine
carbon are almost identical to those of the non-cationic
non-methylated starting materials 2a and 2b suggesting that
the positive charge predominantly located at the N-methyl
nitrogen. In the case of 3b suitable crystals for an X-ray
structure analysis were obtained (Table 1, Figure 1). In the
solid state, the cation of 3b has a conformation where the
imine functionality is oriented away from the 5-ring N-het-
erocycle. Due to steric crowding, the aryl substituents are
tilted with respect to the plane of the benzimidazole-imine
Scheme 1. Recent examples[7Ϫ9] of [N,C] chelating NHC metal
complexes
Results and Discussion
The most versatile and common route to imino-N-hetero-
cyclic carbenes consists of deprotonation of their corre-
sponding (benz)imidazolium salts by suitable strong
bases.[2] Simple azolium salts are easily accessible by alky-
lation or arylation of the parent heterocycles, whereas intro-
duction of an additional N-donor with sterically demanding
substituents is more challenging.
In a first approach, one might assume that condensation
of known N,NЈ-dibenzoylbenzimidazolium salts[16] with
anilines might give access to tridentate [N,C,N] ligand pre-
cursors (compare Scheme 2). However, it proved impossible
to convert mono- or bis-acylated azolium salts into the cor-
responding iminoyl derivatives under a variety of exper-
imental conditions. Therefore we chose an alternate syn-
thetic strategy which introduces the imino functionality at
an early stage in the synthetic sequence (Scheme 3).
Scheme 2. Design concept for new [N,C] and [N,C,N] ligands
2828
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2827Ϫ2836

[1,2]-Rearrangement of Imino-N-heterocyclic Carbenes
FULL PAPER
Scheme 3. Synthesis of compounds 2؊10 (ImH ϭ imidazole, BzImH ϭ benzimidazole, iPr ϭ isopropyl, Me ϭ methyl, Ph ϭ phenyl)
moiety by 39.8° (phenyl group) and 76.6° (2,6-diisopro- suggesting that this 1,2-rearrangement is an unavoidable
pylphenyl group), respectively. The unsymmetric charge dis- process under normal experimental conditions, thereby
tribution indicated by the NMR spectroscopic data in solu- more or less ruling out the existence of iminocarbenes 4a
tion is also evident in the crystalline state: The bond length and 4b in the condensed phase. The 1,2-shift of the iminoyl
of C(14)ϪN(2) equals 133.8(4) pm, slightly longer than the moiety of carbenes 4a or 4b is reminiscent of the Stevens
value of C(14)ϪN(3) with 132.0(4) pm, suggesting more rearrangement[18] of acceptor-substituted ammonium salts,
double-bond character to this latter bond and more positive but to the best of our knowledge no such migration has
charge on the nitrogen N(3) of the N-methyl group.
ever been described for ylides derived from (benz)imidazol-
Both azolium salts 3a and 3b serve as the key precursors ium salts.
for the corresponding [N,C] carbenes 4a and 4b, respectively
In addition we report that the attempted preparation of
(Scheme 3). Reaction of THF solutions of 3a or 3b with a Pd complex of 4a by Raubenheimer’s methodology[6] was
potassium hydride as base led to the evolution of H2 gas, equally unsuccessful: deprotonation of 2a followed by com-
indicating deprotonation at the imidazole C(2) carbon. plexation with PdCl2 and subsequent methylation with
Subsequent reaction of these deprotonated (benz)imidazol- methyl triflate did not afford the desired [N,C]PdCl2 com-
ium species with PdCl2 afforded stable palladium com- plex, instead a regular non-chelating imidazoleϪPd com-
plexes, as anticipated, but their properties and structures plex was formed.[19]
(vide infra) were not consistent with [N,C]PdCl2 complexes.
The rearrangement of [N,C] carbenes 4a and 4b to [N,N]
It soon became clear that the initial carbenes 4a and 4b ligands 5a and 5b prompted us to look for a more versatile
must have rearranged by a 1,2-iminoyl shift to C(2)-substi- synthetic route for this potentially useful ligand framework.
tuted iminoyl(benz)imidazole 5a and 5b, and with ‘‘regular’’ To this end, benzoyl chloride 6 was reacted with methyl-
[N,N] ligands easily formed the [N,N]PdCl2 complexes 10a imidazole 7 to afford 2-benzoyl-N-methylimidazole (8) by a
and 10b, respectively. Low temperature reactions aimed at published procedure (Scheme 3).[20] Introduction of the im-
trapping the initial carbenes 4a or 4b failed in all cases, ine functionality was achieved with aluminum-metalated
Eur. J. Inorg. Chem. 2004, 2827Ϫ2836
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2829

B. Bildstein et al.
FULL PAPER
Table 1. Crystallographic data for 3b, 10a, 10b, 10c
Compound
3b
10a
10b
10c
Molecular formula
Molecular mass
Crystal system
Space group
a, pm
b, pm
c, pm
α, deg
C28H30F3N3O3S
545.61
monoclinic
P21/c (No. 14)
853.60(2)
2896.60(9)
1185.39(5)
90
C23H27Cl2N3Pd
522.78
C27H29Cl2N3Pd
572.83
C19H19Cl2N3Pd
466.67
monoclinic
P21/c (No. 14)
1110.34(2)
1269.48(3)
1649.36(3)
90
monoclinic
P21/c (No. 14)
1122.59(5)
1391.70(5)
1622.83(6)
90
monoclinic
P21/c (No. 14)
1314.09(4)
966.81(2)
1518.84(4)
90
β, deg
γ, deg
103.800(2)
90
94.566(1)
90
91.865(2)
90
94.852(1)
90
Z
4
4
4
4
V, nm3
2.84632(16)
1.273
233(2)
0.166
1144
2.31748(8)
1.498
233(2)
1.045
1064
2.53402(17)
1.502
293(2)
0.963
1168
1.92273(9)
1.612
233(2)
1.249
936
ρcalcd, Mg·mϪ3
Temp, K
Absorption coefficient (mmϪ1
F(000)
)
Color, habit
Crystal size
θ range for data collection
Index ranges
colorless prism
0.3 ϫ 0.2 ϫ 0.12 mm
1.90° to 21.00°
0 ϭ h ϭ 8; Ϫ29 ϭ k ϭ 29;
Ϫ11 ϭ l ϭ 11
11078
3042 (Rint ϭ 0.0469)
2291
none
Full-matrix
least-squares on F2
3042/0/345
1.083
R1 ϭ 0.0534; wR2 ϭ 0.1406 R1 ϭ 0.0239; wR2 ϭ 0.0590
R1 ϭ 0.0742;
orange prism
0.3 ϫ 0.25 ϫ 0.16 mm
2.03° to 25.00°
yellow prism
0.08 ϫ 0.06 ϫ 0.05 mm
1.93° to 22.00°
orange plate
0.35 ϫ 0.3 ϫ 0.15 mm
1.97° to 26.00°
0 ϭ h ϭ 13; Ϫ15 ϭ k ϭ 15; 0 ϭ h ϭ 11; Ϫ14 ϭ k ϭ 14; 0 ϭ h ϭ 16; Ϫ11 ϭ k ϭ 11;
Ϫ19 ϭ l ϭ 19
13548
4070 (Rint ϭ 0.0243)
3631
none
Full-matrix
least-squares on F2
4070/0/264
1.057
Ϫ17 ϭ l ϭ 17
11746
3107 (Rint ϭ 0.0591)
2434
none
Full-matrix
least-squares on F2
3107/0/300
Ϫ18 ϭ l ϭ 18
11762
3768 (Rint ϭ 0.0285)
3235
none
Full-matrix
least-squares on F2
3768/0/231
Reflections collected
Independent reflections
Reflections with I Ͼ 2σ(I)
Absorption correction
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R int [I Ͼ 2σ(I)]
R int (all data)
1.022
1.062
R1 ϭ 0.0348; wR2 ϭ 0.0693
R1 ϭ 0.0539;
wR2 ϭ 0.0745
0.00053(19)
R1 ϭ 0.0261; wR2 ϭ 0.0627
R1 ϭ 0.0337;
wR2 ϭ 0.0652
0.0038(3)
R1 ϭ 0.0291;
wR2 ϭ 0.1518
0.0043(10)
343 and Ϫ282 e nmϪ3
wR2 ϭ 0.0608
0.0007(2)
405 and Ϫ356 e nmϪ3
Extinction coefficient
Largest diff. peak and hole
316 and Ϫ260 e nmϪ3
367 and Ϫ339 e nmϪ3
anilines 9a, 9b, and 9c in yields of 68%-85%, thereby al-
lowing steric fine-tuning of these [N,N] ligand systems. The
aluminum-assisted condensation of carbonyl functionalities
with primary amines is a very convenient method,[11a,21]
which is especially useful for electronically deactivated and
sterically crowded systems. Mechanistically it involves a me-
talation of aniline with trimethylaluminum and sequential
formation of the hemiaminal and aminal of the carbonyl
moiety, driven by the high oxophilicity of aluminum and by
the formation of the thermodynamically very stable
AlϪOϪAl bond of tetramethylaluminoxane.[11a,21] This
aluminum-assisted formation of 5aϪ5d is a necessary syn-
thetic protocol for the chemistry reported in this paper, be-
cause standard condensation procedures of 8 with anilines
(e.g. acid-catalyzed condensation with aceotropic removal
of water) gave only partial conversion into the product in
non-useful preparative yields. Compounds 5a, 5c, and 5d
are air-stable viscous oils with spectroscopic properties in
accordance with their structure (for details see Exp. Sect.).
During the course of this work an analogous [N,N] ligand
has been claimed in a patent,[22] although without a phenyl
substituent on the imine carbon and without a NЈ-alkyl
group. Otherwise the only comparable bidentate N-hetero-
Figure 1. Molecular structure of 3b; hydrogen atoms are omitted
for clarity; selected bond lengths (pm): C(1)ϪN(1) ϭ 126.1(4),
C(1)ϪN(2) ϭ 145.4(4), N(2)ϪC(14) ϭ 133.8(4), C(14)ϪN(3) ϭ
132.0(4); selected angles (°): N(2)ϪC(14)ϪN(3) ϭ 110.6(3); tilt an-
gle phenyl group of C(1) versus plane N(1)ϪC(1)ϪN(2): 39.8(3);
tilt angle 2,6-diisopropylphenyl group of N(1) versus plane
N(1)ϪC(1)ϪN(2): 76.6(3)
2830
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2827Ϫ2836

[1,2]-Rearrangement of Imino-N-heterocyclic Carbenes
FULL PAPER
cyclic imine ligands reported in the literature are pyridinyl-
imines published three years ago.[23]
Reaction of 5a, 5c and 5d with either PdCl2 or NiBr2
afforded the [N,N]MX2 complexes 10a, 10cϪ10g, respec-
tively, in isolated yields of 65Ϫ99% (Scheme 3). Complex
10b with a benzimidazole backbone has been prepared in
85% yield by the one-pot sequence 3b-4b-5b-10b (vide su-
pra). All four palladium compounds 10aϪ10d are orange,
air-stable and diamagnetic complexes, indicating the ex-
pected regular square-planar coordination geometry at the
palladium center. In contrast, all three nickel complexes
10eϪ10f are green-brown, highly air-sensitive and strongly
paramagnetic compounds, suggesting a tetracoordinate
tetrahedrally distorted geometry in the case of monomeric
complexes, or a bromine-bridged pentacoordinate coordi-
nation sphere in the case of dimeric complexes. The para-
magnetism of these nickel compounds prevented NMR
spectroscopic characterization, on the other hand it is clear
evidence for the ‘‘designed’’ steric bulk of the [N,N] ligands
5a, 5c and 5d.
Figure 2. Molecular structure of 10a; hydrogen atoms are omitted
for clarity; selected bond lengths (pm): Pd(1)ϪCl(1) ϭ 228.36(6),
Pd(1)ϪCl(2)
ϭ
228.15(6), Pd(1)ϪN(1)
ϭ
207.75(17),
The different substitution patterns of the N-aryl groups
of palladium complexes 10aϪ10d is obviously responsible
for varying solubility in this series of complexes: com-
pounds 10a and 10c are soluble in common organic sol-
vents, allowing their routine characterization by NMR
Pd(1)ϪN(2) ϭ 199.20(19), C(1)ϪN(1) ϭ 130.3(3), C(1)ϪC(14) ϭ
145.4(3), N(2)ϪC(14) ϭ 132.7(3), C(14)ϪN(3) ϭ 136.1(3); selected
angles (°): Cl(1)ϪPd(1)ϪCl(2) ϭ 91.60(2); Cl(1)ϪPd(1)ϪN(1) ϭ
97.40(5); Cl(2)ϪPd(1)ϪN(2) ϭ 91.25(6); N(1)ϪPd(1)ϪN(2) ϭ
79.73(7); N(1)ϪC(1)ϪC(14) ϭ 113.86(19); N(2)ϪC(14)ϪC(1) ϭ
118.23(19); tilt angle phenyl group of C(1) versus plane
spectroscopic analysis, but complexes 10b and 10d showed N(1)ϪC(1)ϪC(14): 71.9(1); tilt angle 2,6-diisopropylphenyl group
only insufficient solubilities even in polar solvents like
of N(1) versus plane N(1)ϪC(1)ϪC(14): 74.9(2)
acetonitrile, acetone and dimethyl sulfoxide, preventing de-
tection of their NMR signals. For compounds 10a and 10c
the 13C signal of the imine carbon (10a: δCϭN ϭ 165.7 ppm;
10c: δCϭN ϭ 167.5 ppm) is shifted to lower field by approxi-
mately 7 ppm in comparison with the non-complexed free
ligands 5a and 5c.
Single crystals for X-ray analyses were obtained for com-
plexes 10aϪ10c (Table 1, Figures 2Ϫ4). Overall, all three
compounds show the expected square-planar coordination
geometry at the palladium centers, deviations from the ideal
90° angle are most pronounced for the structurally enforced
[N,N] chelate (angle NϪPdϪN; 10a: 79.7°, 10b: 79.8°, 10c:
79.6°) whereas all other angles are, more or less, close to
the ideal value (10a: 91.3°-97.4°, 10b: 89.1°-95.8°, 10c:
92.2°-95.7°). On the other hand, due to steric crowding, the
phenyl groups and the peripheral N-aryl groups show large
tilt angles of 71.9° to 87.4° in relation to the coordination
Figure 3. Molecular structure of 10b; hydrogen atoms are omitted
plane. Clearly visible in Figures 2Ϫ4 is the different steric
for clarity; selected bond lengths (pm): Pd(1)ϪCl(1) ϭ 227.93(12),
shielding of the palladium metal centers by the differently
substituted N-aryl groups, indicating ‘‘steric control’’ in
these precatalysts by employment of different building
blocks. The lengths of the PdϪCl bonds (226.9Ϫ228.4 pm)
and of the PdϪN bonds (199.1Ϫ207.8 pm) are unexcep-
tional. The bond lengths of the coordinated imine moieties
are distinctly shorter (129.7Ϫ130.3 pm) in comparison with
those of the coordinated NϭC double bonds (132.7Ϫ133.4
pm) of the (benz)imidazole rings.
Pd(1)ϪCl(2) ϭ 226.91(13), Pd(1)ϪN(1) ϭ 204.4(3), Pd(1)ϪN(2) ϭ
202.6(3), C(1)ϪN(1)
N(2)ϪC(14) ϭ 133.1(5), C(14)ϪN(3) ϭ 136.7(5); selected angles
89.10(5); Cl(1)ϪPd(1)ϪN(1) ϭ
ϭ 129.7(5), C(1)ϪC(14) ϭ 147.0(6),
(°): Cl(1)ϪPd(1)ϪCl(2)
ϭ
95.52(10); Cl(2)ϪPd(1)ϪN(2) ϭ 95.79(11); N(1)ϪPd(1)ϪN(2) ϭ
79.77(14); N(1)ϪC(1)ϪC(14) ϭ 113.8(4); N(2)ϪC(14)ϪC(1) ϭ
117.6(4); tilt angle phenyl group of C(1) versus plane
N(1)ϪC(1)ϪC(14): 86.2(2); tilt angle 2,6-diisopropylphenyl group
of N(1) versus plane N(1)ϪC(1)ϪC(14): 79.5(3)
geneous solution (70 °C, 20 bar ethylene pressure) showed
Some preliminary catalytic screenings were performed only very low activities, indicating the predominance of
with these new palladium and nickel precatalysts. Ethylene chain termination processes. This result is not too unexpec-
polymerization experiments with methylaluminoxane ted, because axial shielding of the active metal site in these
(MAO) activated nickel complexes 10eϪ10g in homo- nickel complexes is insufficient due to the sterically ‘‘open’’
Eur. J. Inorg. Chem. 2004, 2827Ϫ2836
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2831

B. Bildstein et al.
FULL PAPER
C(2) carbon using potassium hydride as base is successful,
however, the initially formed imino-NHC rearranges by a
Stevens 1,2-shift to 2-iminoyl(benz)imidazoles which are
new [N,N] ligand systems. These are also synthesized by an
alternate synthetic route, which introduces the N-aryl group
by an aluminum-assisted condensation of anilines with 2-
benzoyl-1-methylimidazole. The [N,N] palladium complexes
derived from these ligands are regular square-planar, dia-
magnetic complexes, as shown by single-crystal structure
analysis of three representatives. In contrast, the [N,N]
nickel complexes are highly air-sensitive and strongly para-
magnetic complexes, due to ligand induced distortions. Pre-
liminary catalytic tests of olefin polymerizations and of
CϪC cross-coupling reactions are performed, showing that
the palladium complexes are useful catalysts for Suzuki re-
actions, but the nickel complexes are only poor catalysts for
MAO-cocatalyzed ethylene polymerizations.
Figure 4. Molecular structure of 10c; hydrogen atoms are omitted
for clarity; selected bond lengths (pm): Pd(1)ϪCl(1) ϭ 228.11(7),
Pd(1)ϪCl(2) ϭ 227.73(7), Pd(1)ϪN(1) ϭ 205.7(19), Pd(1)ϪN(2) ϭ
Experimental Section
199.1(2), C(1)ϪN(1)
ϭ
130.2(3), C(1)ϪC(8)
ϭ
146.0(3),
General: Commercially available starting materials were used as ob-
tained. Solvents were dried, deoxygenated and saturated with argon
according to standard procedures in organometallic chemistry. Re-
actions of air-sensitive materials were performed in Schlenk glass-
ware in an atmosphere of argon by techniques common in inor-
ganic/organometallic chemistry. New compounds were charac-
terized by MS and IR and NMR spectroscopy, etc by current state
of the art procedures, details of instrumentation have been pub-
lished previously.[24]
N(2)ϪC(8) ϭ 133.4(3), C(8)ϪN(3) ϭ 135.1(3); selected angles (°):
Cl(1)ϪPd(1)ϪCl(2) ϭ 92.76(3); Cl(1)ϪPd(1)ϪN(1) ϭ 95.65(5);
Cl(2)ϪPd(1)ϪN(2) ϭ 92.16(6); N(1)ϪPd(1)ϪN(2) ϭ 79.60(7);
N(1)ϪC(1)ϪC(8) ϭ 113.1(2); N(2)ϪC(8)ϪC(1) ϭ 117.5(2); tilt an-
gle phenyl group of C(1) versus plane N(1)ϪC(1)ϪC(8): 76.6(2);
tilt angle 2,6-dimethylphenyl group of N(1) versus plane
N(1)ϪC(1)ϪC(8): 87.4(2)
imidazole side of the bidentate iminoylimidazole ligands.
On the other hand, palladium complexes 10a, 10c, 10d are
efficient catalysts for Suzuki carbon-carbon cross-coupling
reactions: aromatic boronic acids are quantitatively coupled
with aryl bromides without formation of undesired homo-
coupled products. However, only bromo electrophiles are
sufficiently activated for this reaction, attempts to cross-co-
uple boronic acids with the industrially more important
chloro aromatics gave no conversion at all.
Imidazole 2a: A Schlenk tube was charged with imidazole (1.76 g,
25.8 mmol) and dry dichloromethane (40 mL). N-(2,6-diisopro-
pylphenyl)benziminoyl chloride[11b,16] 1 (12.3 mmol) dissolved in
dichloromethane (30 mL) was added to this solution. Imidazolium
chloride immediately precipitated, indicating fast formation of the
product. Stirring at ambient temperature was continued for 3
hours. Workup: 150 mL of water was added to the mixture to dis-
solve and remove the imidazolium chloride, the organic product
was extracted with three portions of dichloromethane, the com-
bined organic layers were washed with a brine solution, the di-
chloromethane solution of the product was dried with Na2SO4 and
concentrated in vacuo to a volume of approximately 5 mL. After
an equal volume of n-hexane was added, 3.30 g 2a crystallized in
the form of a yellow, microcrystalline solid in 81% yield, m.p:
112Ϫ115 °C. IR (KBr): ν˜ ϭ 2969 (w), 2962 (m), 1652 (s), 1602 (w),
1590 (m), 1515 (m), 1469 (s), 1439 (s), 1376 (s), 1364 (m), 1333 (m),
1314 (s), 1297 (s), 1283 (m), 1239 (s), 1160 (s), 1059 (s), 1023 (s),
Conclusion
The work reported in this contribution is motivated by
the search for novel bidentate [N,C] ligand systems as steer-
ing ligands for late transition metal complexes with appli-
cations in polymerization catalysis and fine chemical syn-
thesis. Conceptually, the new ligands should contain one
imine donor with a sterically demanding N-aryl substituent
and a N-heterocyclic carbene (NHC) as the second donor
site. Synthetically, imidazole and benzimidazole are N-deri-
vatized with iminoyl chloride followed by NЈ-alkylation
940 (m), 918 (s), 835 (s), 706 (s), 700 (s) cmϪ1. MS (FAB): m/z ϭ
3
331 [M]ϩ. 1H NMR (300 MHz, CDCl3): δ ϭ 0.89 [d, JH,H
ϭ
6.6 Hz, 6 H, CH(CH3)2], 1.07 [d, 3JH,H ϭ 6.6 Hz, 6 H, CH(CH3)2],
2.82 [2 H, sept., 2 ϫ CH(CH3)2], 6.65Ϫ7.71 (m, 10 H, C6H5, C6H3,
C3H3N2), 7.91 (s, 1 H, C3H3N2) ppm. 13C NMR (75.432 MHz,
with methyl triflate, affording novel iminoyl(benz)imidazol- CDCl3): δ ϭ 21.7, 23.8, 28.4 [CH(CH3)2], 118.0, 122.8, 124.1,
128.3, 128.9, 129.9, 130.2, 130.7, 136.6, 137.3, 141.9 (C6H5, C6H3,
C3H3N2), 148.6 (CϭN) ppm. C22H25N3 (331.46): calcd. C 79.72,
H 7.60, N 12.68; found C 80.00, H 7.63, N 12.63.
ium salts of surprising stability. An X-ray structure and
NMR spectroscopic data of benzimidazolium triflate indi-
cate that the positive charge resides predominantly at the
NЈ-methyl side, suggesting that the N-iminoyl moiety is in
a ‘‘normal’’ chemical environment, thereby explaining the
unusual stability of this azolium salt. Deprotonation of
Benzimidazole 2b: A Schlenk tube was charged with benzimidazole
(761 mg, 6.44 mmol) and dry dichloromethane (70 mL). N-(2,6-
diisopropylphenyl)benziminoyl chloride[11b,17] 1 (3.07 mmol) dis-
these (benz)imidazolium triflates at the (benz)imidazole solved in dichloromethane (8 mL) was added to this solution.
2832  2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2827Ϫ2836

[1,2]-Rearrangement of Imino-N-heterocyclic Carbenes
FULL PAPER
Benzimidazolium chloride immediately precipitated, indicating fast
formation of the product. Stirring at ambient temperature was con-
tinued for 4 hours. Workup: 150 mL of water was added to the
mixture to dissolve and remove the benzimidazolium chloride, the
organic product was extracted with three portions of dichlorometh-
ane, the combined organic layers were washed with a brine solu-
tion, the dichloromethane solution of the product was dried with
Na2SO4 and concentrated in vacuo to a volume of approximately
5 mL. After an equal volume of n-hexane was added, 1.12 g 2b
crystallized in the form of a yellow, microcrystalline solid in 96%
yield. m.p.: 123Ϫ127 °C. IR (KBr): ν˜ ϭ 2963 (w), 2867 (m), 1651
(s), 1609 (s), 1588 (s), 1493 (s), 1451 (s), 1369 (s), 1290 (m), 1217
(s), 1161 (m), 1057 (m), 931 (s), 800 (s), 764 (s), 708 (s), 698 (s)
(545.62): calcd. C 61.64, H 5.54, N 7.70; found C 61.53, H 5.57, N
7.66. Single crystal X-ray structure: Table 1, Figure 1.
Ketone 8:[20] A round-bottom flask was charged (without protection
from air) with 1-methylimidazole 7 (10 mL, 0.125 mol) and aceto-
nitrile (125 mL). The stirred solution was cooled to 0 °C and ben-
zoyl chloride 6 (14.6 mL) was added from a dropping funnel within
a period of 15 minutes, followed by addition of triethylamine (17.5
mL, 0.125 mol). The mixture was stirred overnight at ambient tem-
perature, resulting in a yellow solution and a white precipitate of
triethylammonium chloride. The ammonium salt was filtered off,
solvents and other volatile materials were removed on a rotary
evaporator, affording crude 8 as a yellow oil. Distillation in an oil
pump vacuum gave 17.5 g (75%) pure 8 as a yellow viscous liquid,
1
cmϪ1. MS (FAB): m/z ϭ 381 [M]ϩ. H NMR (300 MHz, CDCl3):
1
b.p. 120Ϫ125 °C. MS (EI): m/z ϭ 186 [M]ϩ. H NMR (300 MHz,
δ ϭ 0.95Ϫ1.27 [m, 12 H, CH(CH3)2], 2.92Ϫ3.19 [m, 2 H, 2 ϫ
CD2Cl2): δ ϭ 4.03 (s, 3 H, CH3), 7.13Ϫ7.18 (d, 2 H, C3H2N2),
7.45Ϫ7.58 (m, 3 H, C6H5), 8.26Ϫ8.29 (d, 2 H, C6H5) ppm. 13C
NMR (75.432 MHz, CD2Cl2): δ ϭ 36.6 (NϪCH3), 127.3, 128.2,
129.3, 131.1, 132.8, 137.8, 143.5 (C6H5, C3H2N2), 184.1 (CϭO)
ppm. 1H NMR spectroscopic data and the boiling point concur
with published data.[20]
CH(CH3)2], 7.02Ϫ8.07 (m, 12 H, C6H5, C6H3, C7H5N2) ppm. 13C
NMR (75.432 MHz, CDCl3):
δ ϭ 21.8, 24.2, 28.4, 28.8
[CH(CH3)2], 115.3, 120.4, 122.9, 123.5, 124.0, 124.7, 127.2, 128.6,
128.9, 130.8, 131.7, 136.7, 142.2, 143.0, 146.4 (C6H5, C6H3,
C7H5N2), 149.9 (CϭN) ppm. C26H27N3 (381.52): calcd. C 81.85,
H 7.13, N 11.01; found C 81.98, H 7.15, N 10.97.
Imidazole 5a: A Schlenk tube was charged with 2,6-diisopropylani-
line 9a (2 mL, 10.6 mmol), toluene (40 mL), and 5.3 mL of a 2.0
molar toluene solution of trimethylaluminum (10.6 mmol). The
stirred solution was heated to 80 °C. After 90 minutes methane
evolution was completed and the solution was cooled to ambient
temperature. Under protection from air, (1-methyl-2-imidazolyl)-
phenylketone (8) (986 mg, 5.30 mmol) was added in one portion,
and an immediate color change from yellow to burgundy red was
observed. The mixture was stirred overnight at ambient tempera-
ture and heated to 80 °C for a further three hours, affording a
bright yellow solution. Workup: the solution was cooled to 0 °C,
the mixture was carefully hydrolyzed with crushed ice, the organic
product was extracted with three portions of diethyl ether, the com-
bined organic layers were washed with one portion of a 5% aque-
ous NaOH solution, the diethyl ether solution was washed with
two portions of water, the organic phase was dried with Na2SO4,
and all volatile materials were removed in vacuo on a rotary evap-
orator, affording the crude product together with 2,6-diisopro-
pylaniline. Chromatography on alumina with diethyl ether/n-hex-
ane (v/v, 1:1) as eluent afforded 1.45 g pure 5a in 79% yield as a
yellow oil. MS (EI): m/z ϭ 345 [M]ϩ. 1H NMR (300 MHz, CDCl3):
Imidazolium Triflate 3a: A Schlenk tube was charged with 2a
(150 mg, 0.453 mmol) and dry dichloromethane (40 mL). The solu-
tion was cooled to Ϫ80 °C and methyltrifluoromethansulfonate (50
µL, 0.453 mmol) was added via a syringe. After 20 minutes stirring
at Ϫ80 °C, the cooling bath was removed and the mixture was
stirred for a further 2 hours, resulting in a clear bright yellow solu-
tion. Thin layer chromatography (TLC) indicated complete reac-
tion, all starting material had been consumed with the formation
of a polar non-eluting product. Workup: all volatile materials were
removed on a vacuum line, the solid yellow residue was washed
with two portions of dry diethyl ether and dried in vacuo, affording
224 mg 3a as a yellow, moderately air-sensitive powder in 100%
1
yield. MS (FAB): m/z ϭ 346 [M of cation]ϩ. H NMR (300 MHz,
3
CD2Cl2): δ ϭ 0.93 (d, JH,H ϭ 6.9 Hz, 6 H, CH(CH3)2], 1.12 [d,
3JH,H ϭ 6.6 Hz, 6 H, CH(CH3)2], 2.81 (m, 2 H, 2 ϫ CH(CH3)2],
4.08 (s, 3 H, CH3), 7.07Ϫ7.87 (m, 10 H, C6H5, C6H3, C3H3N2),
9.09 (s, 1 H, C3H3N2) ppm. 13C NMR (75.432 MHz, CD2Cl2): δ ϭ
21.9, 24.1, 28.8 [CH(CH3)2], 37.6 (NϪCH3), 118.9, 120.5, 123.7,
124.9, 125.8, 129.4, 129.5, 132.7, 136.9, 137.0, 140.8 (C6H5, C6H3,
C3H3N2), 147.3 (CϭN) ppm. C24H28F3N3O3S (495.56): calcd. C
58.17, H 5.70, N 8.48; found C 58.08, H 5.72, N 8.43.
3
3
δ ϭ 0.89 [d, JH,H ϭ 6.9 Hz, 6 H, CH(CH3)2], 1.10 [d, JH,H
6.6 Hz, 6 H, CH(CH3)2], 2.87 [m, 2 H, 2 ϫ CH(CH3)2], 4.08 (s, 3
H, NϪCH3), 6.98Ϫ7.17 (m, 10 H, C6H5, C6H3, C3H2N2) ppm. 13
ϭ
Benzimidazolium Triflate 3b: A Schlenk tube was charged with 2b
(214 mg, 0.561 mmol) and dry dichloromethane (40 mL). The solu-
tion was cooled to Ϫ80 °C and methyltrifluoromethansulfonate (62
µL, 0.561 mmol) was added via a syringe. After 20 minutes stirring
at Ϫ80 °C, the cooling bath was removed and the mixture was
stirred for a further 2 hours, resulting in a clear bright yellow solu-
tion. Thin layer chromatography (TLC) indicated complete reac-
tion, all starting material had been consumed with the formation
C
NMR (75.432 MHz, CDCl3): δ ϭ 21.6, 24.3, 28.2 [CH(CH3)2], 36.8
(NϪCH3), 122.6, 123.5, 125.4, 127.3, 128.4, 129.0, 129.2, 134.8,
135.5, 144.6, 144.9 (C6H5, C6H3, C3H2N2), 158.5 (CϭN) ppm.
C23H27N3 (345.49): calcd. C 79.96, H 7.88, N 12.16; found C 80.05,
H 7.86, N 12.12.
Imidazole 5c: A Schlenk tube was charged with 2,6-dimethylaniline
of a polar non-eluting product. Workup: all volatile materials were (9b) (2.5 mL, 20.3 mmol), toluene (40 mL) and 10.2 mL of a 2.0
removed on a vacuum line, the solid yellow residue was washed
with two portions of dry diethyl ether and dried in vacuo, affording
molar toluene solution of trimethylaluminum (20.3 mmol). The
stirred solution was heated to 80 °C. After 90 minutes methane
236 mg 3b as a yellow, moderately air-sensitive powder in 77% evolution was completed and the solution was cooled to ambient
1
yield. MS (FAB): m/z ϭ 396 [M of cation]ϩ. H NMR (300 MHz,
temperature. Under protection from air, (1-methyl-2-imidazolyl)-
3
CD2Cl2): δ ϭ 0.95 (d, JH,H ϭ 6.9 Hz, 6 H, CH(CH3)2], 1.14 [d, phenylketone (8) (1.89 g, 10.2 mmol) was added in one portion,
3JH,H ϭ 6.6 Hz, 6 H, CH(CH3)2], 2.92 [m, 2 H, 2 ϫ CH(CH3)2], and an immediate color change from yellow to burgundy red was
4.25 (s, 3 H, CH3), 7.11Ϫ8.09 (m, 12 H, C6H5, C6H3, C7H5N2), observed. The mixture was stirred over the weekend at ambient
9.50 (s, 1 H, C7H5N2) ppm. 13C NMR (75.432 MHz, CD2Cl2): δ ϭ temperature, affording a bright yellow solution. Workup: the solu-
21.9, 24.4, 28.8 [CH(CH3)2], 34.8 (NϪCH3), 113.7, 117.4, 123.7, tion was cooled to 0 °C, the mixture was carefully hydrolyzed with
125.8, 128.5, 128.9, 129.6, 129.7, 130.7, 132.7, 136.9, 141.1, 143.2
crushed ice, the organic product was extracted with three portions
(C6H5, C6H3, C3H3N2), 149.1 (CϭN) ppm. C28H30F3N3O3S of diethyl ether, the combined organic layers were washed with one
Eur. J. Inorg. Chem. 2004, 2827Ϫ2836
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2833

B. Bildstein et al.
FULL PAPER
portion of a 5% aqueous NaOH solution, the diethyl ether solution
was washed with two portions of water, the organic phase was dried
with Na2SO4, and all volatile materials were removed in vacuo on
a rotary evaporator, affording the crude product together with 2,6-
dimethylaniline. Chromatography on alumina with diethyl ether/n-
hexane (v/v, 1:1) as eluent afforded 2.0 g pure 5c in 68% yield as a
yellow oil. MS (EI): m/z ϭ 289 [M]ϩ. 1H NMR (300 MHz,
CD2Cl2): δ ϭ 2.09 (s, 6 H, CH3), 4.15 (s, 3 H, NϪCH3), 6.82Ϫ7.24
(m, 10 H, C6H5, C6H3, C3H2N2) ppm. 13C NMR (75.432 MHz,
CD2Cl2): δ ϭ 18.8 (CH3), 37.6 (NϪCH3), 123.2, 125.9, 126.1,
127.6, 127.9, 128.5, 128.9, 129.3, 136.7, 144.8, 148.4 (C6H5, C6H3,
C3H2N2), 159.8 (CϭN) ppm. C19H19N3 (289.38): calcd. C 78.86,
H 6.62, N 14.52; found C 78.99, H 6.62, N 14.48.
acetonitrile as eluent, the orange fraction was collected, solvents
were removed on a rotary evaporator, and the product was dried
in vacuo, yielding 800 mg (84%) 10a as an orange powder.
Data for 10a: MS (FAB): m/z ϭ 522 [M]ϩ was not observed, due
to clustering signals of higher masses were detected. 1H NMR
(300 MHz, CD2Cl2): δ ϭ 1.05 [d, 3JH,H ϭ 6.6 Hz, 6 H, CH(CH3)2],
3
1.46 (d, JH,H ϭ 6.9 Hz, 6 H, CH(CH3)2], 3.21 [m, 5 H, 2 ϫ
CH(CH3)2, NϪCH3], 6.98Ϫ7.93 (m, 10 H, C6H5, C6H3, C3H2N2)
ppm. 13C NMR (75.432 MHz, CD2Cl2): δ ϭ 23.3, 24.7, 29.1
[CH(CH3)2], 37.5 (NϪCH3), 123.6, 127.7, 127.9, 128.4, 128.8,
129.5, 129.7, 132.3, 140.3, 140.8, 146.3 (C6H5, C6H3, C3H2N2),
165.7 (CϭN) ppm. C23H27Cl2N3Pd (522.81): calcd. C 52.84, H
5.21, N 8.04; found C 52.76, H 5.20, N 8.10. Single-crystal X-ray
structure: Table 1, Figure 2.
Imidazole 5d: A Schlenk tube was charged with 2-aminobiphenyl
(9c) (1.9 g, 11.2 mmol), toluene (40 mL), and 5.6 mL of a 2.0 molar
toluene solution of trimethylaluminum (11.2 mmol). The stirred
solution was heated to 80 °C. After 90 minutes methane evolution
was completed and the solution was cooled to ambient tempera-
ture. Under protection from air, (1-methyl-2-imidazolyl)phenylke-
tone (8) (1.05 g, 5.61 mmol) was added in one portion, and an im-
mediate color change from yellow to burgundy red was observed.
The mixture was stirred for 48 hours at ambient temperature, af-
fording a bright yellow solution. Workup: the solution was cooled
to 0 °C, the mixture was carefully hydrolyzed with crushed ice, the
organic product was extracted with three portions of diethyl ether,
the combined organic layers were washed with one portion of a 5%
aqueous NaOH solution, the ether solution was washed with two
portions of water, the organic phase was dried with Na2SO4, and
all volatile materials were removed in vacuo on a rotary evaporator,
affording the crude product together with 2-aminobiphenyl. Chro-
matography on alumina with diethyl ether/n-hexane (v/v, 1:1) as
eluent afforded 1.61 g pure 5d in 85% yield as a yellow oil. MS
(EI): m/z ϭ 337 [M]ϩ. 1H NMR (300 MHz, CDCl3): δ ϭ 2.69, 4.15
(s, 3 H, NϪCH3, E/Z isomers in 1:2 ratio), 6.59Ϫ7.72 (m, 16 H,
C6H5, C12H9, C3H2N2) ppm. 13C NMR (75.432 MHz, CDCl3): δ ϭ
33.3, 37.2 (NϪCH3, E/Z isomers in 1:2 ratio), 120.8, 121.2, 124.2,
125.1, 125.9, 127.0, 127.8, 128.2, 128.8, 128.9, 129.2, 129.4, 129.6,
129.8, 130.4, 131.4, 132.9, 133.5, 135.4, 138.3, 140.1, 140.4, 142.2,
144.8, 147.7, 148.2, 158.3 (C6H5, C10H9, C3H2N2), 160.2 (CϭN)
ppm. C23H19N3 (337.42): calcd. C 81.87, H 5.68, N 12.45; found C
81.96, H 5.71, N 12.39.
Palladium(II) Complex 10b: A Schlenk tube was charged with 3b
(165 mg, 0.302 mmol), of THF (40 mL) and an excess of potassium
hydride (25 mg, 0.62 mmol). H2 evolution commenced immediately
and the mixture was stirred for 3 hours at ambient temperature,
after that period no more gas evolution was observed. The slightly
yellow suspension was filtered under protection from air through a
short plug of glass wool to remove excess KH and most of the
potassium trifluoromethanesulfonate. PdCl2 (65 mg, 0.363 mmol)
was added to the resulting slightly turbid solution and the mixture
was stirred at room temperature for 24 hours, resulting in a brown
suspension. Workup: the volatiles were removed on a rotary evap-
orator, the brown residue was chromatographed on alumina with
dichloromethane and acetonitrile as eluent, the brown fraction was
decolorized with activated charcoal, and the solvents were removed
in vacuo, affording 147 mg of 10b as a yellow solid in 85% yield.
IR (KBr): ν˜ ϭ 2961 (m), 2869 (m), 1588 (m), 1487 (s), 1410 (m),
1385 (m), 1346 (s), 1011 (s), 789 (m), 746 (s), 713 (s) cmϪ1. MS
(FAB): m/z ϭ 573 [M]ϩ was not observed, due to clustering signals
of higher masses were detected. 1H and 13C NMR spectroscopy
were attempted in deuterated acetonitrile, acetone, and dimethyl
sulfoxide, respectively, but the solubility of 10b was insufficient to
obtain meaningful NMR spectroscopic data. C27H29Cl2N3Pd
(572.87): calcd. C 56.61, H 5.10, N 7.33; found C 56.48, H 5.07, N
7.38. Single crystal X-ray structure: Table 1, Figure 3.
Palladium(II) Complex 10c: A Schlenk tube was charged with 5c
(710 mg, 2.45 mmol), acetonitrile (30 mL), and PdCl2 (435 mg,
2.45 mmol). The mixture was stirred overnight at ambient tempera-
ture to afford an orange solution. Workup: the solution was filtered
through celite and chromatographed on silica with dichlorometh-
ane and acetonitrile as eluent, the orange fraction was collected,
solvents were removed on a rotary evaporator, the product was
dried in vacuo, and recrystallized from acetonitrile, yielding
1.03 mg (90%) 10c as orange crystals. MS (FAB): m/z ϭ 467 [M]ϩ
was not observed, due to clustering signals of higher masses were
Palladium(II) Complex 10a. Method A: A Schlenk tube was charged
with 3a (145 mg, 0.293 mmol), THF (40 mL), and an excess of
potassium hydride (25 mg, 0.62 mmol). H2 evolution commenced
immediately and the mixture was stirred for 3 hours at ambient
temperature, after this time no more gas evolution was observed.
The slightly yellow suspension was filtered under protection from
air through a short plug of glass wool to remove excess KH and
most of the potassium trifluoromethanesulfonate. Palladium di-
chloride (52 mg, 0.293 mmol) was added to the resulting slightly
turbid solution and the mixture was stirred at room temperature
for 20 hours, resulting in a brown suspension. Workup: the volatiles
were removed on a rotary evaporator, the brown residue was chro-
matographed on alumina with dichloromethane and acetonitrile as
eluent, the brown fraction was decolorized with activated charcoal,
and the solvents were removed in vacuo, affording 100 mg of 10a
as a tan solid in 65% yield.
1
detected. H NMR (300 MHz, CD3CN): δ ϭ 2.22, 2.34 (6 H, 2 ϫ
s, CH3), 3.11 (s, 3 H, NϪCH3), 6.89Ϫ7.47 (m, 10 H, C6H5, C6H3,
C3H2N2) ppm. 13C NMR (75.432 MHz, CD3CN): δ ϭ 19.4 (CH3),
37.3 (NϪCH3), 127.5, 127.9, 128.2, 128.5, 128.7, 129.8, 130.9,
131.9, 132.5, 144.1, 147.4 (C6H5, C6H3, C3H2N2), 167.5 (CϭN)
ppm. C19H19Cl2N3Pd (466.71): calcd. C 48.90, H 4.10, N 9.00;
found C 49.10, H 4.15, N 8.94. Single crystal X-ray structure:
Table 1, Figure 4.
Method B:
A
Schlenk tube was charged with 5a (630 mg,
Palladium(II) Complex 10d: A Schlenk tube was charged with 5d
1.82 mmol), acetonitrile (30 mL) and PdCl2 (323 mg, 1.82 mmol).
(545 mg, 1.62 mmol), acetonitrile (30 mL) and PdCl2 (286 mg,
The mixture was stirred overnight at ambient temperature to afford 1.62 mmol). The mixture was stirred overnight at ambient tempera-
an orange solution. Workup: the solution was filtered through ce-
lite and chromatographed on silica with dichloromethane and
ture to afford an orange solution. Workup: the solution was filtered
through celite and chromatographed on silica with dichlorometh-
2834
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2827Ϫ2836

[1,2]-Rearrangement of Imino-N-heterocyclic Carbenes
FULL PAPER
ane and acetonitrile as eluent, the orange fraction was collected,
solvents were removed on a rotary evaporator, the product was
dried in vacuo, and recrystallized from a mixture of dichlorometh-
ane and acetonitrile, yielding 835 mg (98%) 10d as orange crystals.
MS (FAB): m/z ϭ 515 [M]ϩ was not observed, due to clustering
signals of higher masses were detected. 1H and 13C NMR spec-
troscopy were attempted, but the solubility of 10d was insufficient
to obtain meaningful NMR spectroscopic data. C23H19Cl2N3Pd
(514.47): calcd. C 53.67, H 3.72, N 8.16; found C 53.37, H 3.75,
N 8.21.
Acknowledgments
We thank BASF Aktiengesellschaft, Ludwigshafen, Germany for
financial support and for preliminary ethene polymerizations as
well as for investigations of CϪC cross-coupling Suzuki reactions.
[1]
A. J. Arduengo, R. L. Harlow, M. Kline, J. Am. Chem. Soc.
1991, 113, 361.
[2]
[2a]
Reviews:
Carbene Chemistry (Ed.: G. Bertrand), Marcel
Dekker, Inc., New York, 2002. [2b]W.A. Herrmann, Angew.
Chem. 2002, 114, 1343; Angew. Chem. Int. Ed. 2002, 41, 1290.
Nickel(II) Complex 10e: A Schlenk tube was charged with 5a
(787 mg, 2.28 mmol), acetonitrile (30 mL) and NiBr2 (550 mg,
2.52 mmol). The mixture was stirred overnight, giving a green sus-
pension. Workup: acetonitrile was removed on a vacuum line, the
residue was dissolved in dry dichloromethane, the solution was fil-
tered through a Schlenk frit to remove excess NiBr2, and the solu-
tion was evaporated to dryness on a vacuum line, affording 1.22 g
(95%) 10e as a green-brown, paramagnetic, air-sensitive powder. 1H
NMR in deuterated dimethyl sulfoxide showed only very broad
signals. C23H27Br2N3Ni (563.98): calcd. C 48.98, H 4.83, N 7.45;
found C 48.70, H 4.89, N 7.58.
[2c]
W. A. Herrmann, T. Weskamp, V. P. W. Böhm, Adv. Or-
[2d]
ganomet. Chem. 2001, 48, 1.
C. J. Carmalt, A. H. Cowley,
Adv. Inorg. Chem. 2000, 50, 1. [2e] D. Bourissou, O. Guerret, F.
[2f]
P. Gabbai, G. Bertrand, Chem. Rev. 2000, 100, 39.
A. J.
Arduengo, Acc. Chem. Res. 1999, 32, 913. [2g] W. A. Herrmann,
C. Köcher, Angew. Chem. 1997, 109, 2257; Angew. Chem. Int.
Ed. Engl. 1997, 36, 2163.
A. J. Arduengo, R. Krafczyk, Chemie in unserer Zeit 1998,
32, 6.
[3]
[4]
[5]
[6]
H.-W. Wanzlick, E. Schikora, Angew. Chem. 1962, 74, 129.
M. F. Lappert, J. Organomet. Chem. 1988, 358, 185.
H. G. Raubenheimer, S. Cronje, J. Organomet. Chem. 2001,
617Ϫ618, 170.
Nickel(II) Complex 10f: A Schlenk tube was charged with 5c
(510 mg, 1.76 mmol), dichloromethane (40 mL) and NiBr2·DME
(544 mg, 1.76 mmol). The mixture was stirred for 48 hours at room
temperature, giving a brown solution. Workup: dichloromethane
was removed on a vacuum line, affording 814 mg (91%) 10f as a
brown, paramagnetic, air-sensitive powder. 1H NMR spectra, in
deuterated dimethyl sulfoxide, showed only very broad signals.
C19H19Br2N3Ni (507.88): calcd. C 44.93, H 3.77, N 8.27; found C
44.54, H 3.67, N 7.96.
[7] [7a]
[7b]
R. H. Crabtree, Pure Appl. Chem. 2003, 75, 435.
M.
Poyatos, J. A. Mata, E. Falomir, R. H. Crabtree, E. Peris, Or-
[7c]
ganometallics 2003, 22, 1110.
cevic, M. Albrecht, J. W. Faller, R. H. Crabtree, J. Am. Chem.
S. Gründemann, A. Kova-
[7d]
Soc. 2002, 124, 10473.
J. A. Loch, M. Albrecht, E. Peris, J.
Mata, J. W. Faller, R. H. Crabtree, Organometallics 2002, 21,
[7e]
700.
S. Gründemann, M. Albrecht, J. A. Loch, J. W. Faller,
R. H. Crabtree, Organometallics 2001, 20, 5485.
[8] [8a]
A. A. Danopoulos, A. A. D. Tulloch, S. Winston, G. Eas-
[8b]
Nickel(II) Complex 10g: A Schlenk tube was charged with 5d
(480 mg, 1.42 mmol), dichloromethane (40 mL) and NiBr2·DME
(439 mg, 1.42 mmol). The mixture was stirred for 48 hours at room
temperature, giving a green solution. Workup: dichloromethane
was removed on a vacuum line, affording 790 mg (99%) 10g as a
green, paramagnetic, air-sensitive powder. 1H NMR spectra, in de-
uterated dimethyl sulfoxide, showed only very broad signals.
C23H19Br2N3Ni (555.92): calcd. C 49.69, H 3.44, N 7.56; found C
49.21, H 3.51, N 7.42.
tham, M. B. Hursthouse, Dalton Trans. 2003, 1009.
A. A.
D. Tulloch, S. Winston, A. A. Danopoulos, G. Eastham, M.
B. Hursthouse, Dalton Trans. 2003, 699. [8c] A. A. Danopoulos,
S. Winston, W. B. Motherwell, Chem. Commun. 2002, 1376. [8d]
A. A. D. Tulloch, A. A. Danopoulos, G. J. Tizzard, S. J. Coles,
M. B. Hursthouse, R. S. Hay-Motherwell, W. B. Motherwell,
Chem. Commun. 2001, 1270. [8e] A. A. D. Tulloch, A. A. Dano-
poulos, S. Kleinhenz, M. E. Light, M. B. Hursthouse, G. Eas-
tham, Organometallics 2001, 20, 2027.
[9]
W. A. Herrmann, L. J. Goossen, M. Spiegler, Organometallics
1998, 17, 2162.
[10]
X-ray Crystallography: Single-crystal X-ray measurements and
structure determinations of 3b, 10a, 10b, 10c (Table 1, Figures
1Ϫ4): The data collection was performed on a Nonius Kappa CCD
equipped with graphite-monochromatized Mo-Kα-radiation (λ ϭ
L. G. Bonnet, R. E. Douthwaite, B. M. Kariuki, Organometall-
ics 2003, 22, 4187.
[11] [11a]
G. A. Luinstra, J. Queisser, B. Bildstein, H.-H. Görtz, C.
Amort, M. Malaun, A. Krajete, G. Werne, M. O. Kristen, N.
Huber, C. Gernert, Highly Active Ethene Polymerization Cata-
lysts with Unusual Imine Ligands, in Late Transition Metal
Polymerization Catalysis (Eds.: B. Rieger, L. S. Baugh, S.
Kacker, S. Striegler), Wiley-VCH, Weinheim, 2003, chapter 3,
p. 59. [11b] A. Krajete, G. Steiner, H. Kopacka, K.-H. Ongania,
K. Wurst, M. O. Kristen, P. Preishuber-Pflügl, B. Bildstein,
Eur. J. Inorg. Chem. 2004, 1740.
O. Kuhnert, M. Malaun, H. Kopacka, B. Bildstein, Z. Anorg.
˚
0.71073 A) and a nominal crystal to area detector distance of
36 mm. Intensities were integrated using DENZO and scaled with
SCALEPACK.[25] Several scans in the ϕ and ω directions were
made to increase the number of redundant reflections, which were
averaged in the refinement cycles. This procedure replaces an em-
pirical absorption correction. The structures were solved with di-
rect methods (SHELXS-86) and refined against F2 (SHELX-97).[26]
Hydrogen atoms at carbon atoms were added geometrically and
refined using a riding model. All non-hydrogen atoms were refined
with anisotropic displacement parameters.
[11c]
U. Siemeling, T.-C. Auch,
[11d]
Allg. Chem. 2003, 629, 1334.
Krajete, H. Kopacka, K. Wurst, M. Christ, D. Lilge, M. O.
C. Amort, M. Malaun, A.
Kristen, B. Bildstein, Appl. Organomet. Chem. 2002, 16, 506.
[11e]
U. Siemeling, O. Kuhnert, B. Neumann, A. Stammler, H.-
G. Stammler, B. Bildstein, M. Malaun, M. O. Kristen, P.
Zanello, Eur. J. Inorg. Chem. 2001, 913.
CCDC-216464 (for 3b), -216465 (for 10a), -216466 (for 10b),
-216467 (for 10c) contain the crystallographic data for this paper.
These data can be obtained free of charge at www.ccdc.cam.ac.uk/
conts/retrieving.html [or from the Cambridge Crystallographic
Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; Fax: (in-
ternat.) ϩ44-1223-336-033; E-mail: deposit@ccdc.cam.ac.uk].
[12]
D. S. McGuinness, V. C. Gibson, D. F. Wass, J. W. Steed, J.
Am. Chem. Soc. 2003, 125, 12716.
[13] [13a]
T. Weskamp, F. J. Kohl, W. A. Herrmann, J. Organomet.
[13b]
Chem. 1999, 582, 362.
M. Spiegler, W. A. Herrmann, Angew. Chem. 1998, 110, 2631;
T. Weskamp, W. C. Schattenmann,
Eur. J. Inorg. Chem. 2004, 2827Ϫ2836
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2835

B. Bildstein et al.
FULL PAPER
[13c]
Angew. Chem. Int. Ed. 1998, 37, 2490.
H. Grubbs, Angew. Chem. 2000, 112, 3025; Angew. Chem. Int.
C. W. Bielawski, R.
and Structure, 5th ed., (Eds.: M. B. Smith, J. March), John
Wiley & Sons, Inc.; New York, 2001, chapter 16, p. 1419.
G. Steiner, B. Bildstein, K. Wurst, unpublished.
E. Regel, K. H. Büchel, Justus Liebigs Ann. Chem. 1977, 145.
B. Bildstein, P. Denifl, Synthesis 1994, 158.
J. A. Ponasik Jr., J. P. McDevitt, C. M. Killian, P. B. Mackenzie,
L. S. Moody (Eastman Chemical Company) US 6372682 B2,
example 92, p. 43.
T. V. Laine, U. Piironen, K. Lappalainen, M. Klinga, E. Aitola,
M. Leskelä, J. Organomet. Chem. 2000, 606, 112.
B. Bildstein, M. Malaun, H. Kopacka, K. Wurst, M. Mit-
terböck, K.-H. Ongania, G. Opromolla, P. Zanello, Organome-
tallics 1999, 18, 4325.
Processing of X-ray diffraction data collected in oscillation
mode: Z. Otwinowski, W. Minor, Methods Enzymol. 1997,
276, 307.
[19]
[20]
[21]
[22]
Ed. 2000, 39, 2903.
[14]
Review: M. R. Buchmeiser, Well-Defined Transition Metal
Catalysts for Metathesis Polymerization, in Late Transition Me-
tal Polymerization Catalysis (Eds.: B. Rieger, L. S. Baugh, S.
Kacker, S. Striegler), Wiley-VCH, Weinheim, 2003, chapter 5,
p. 155.
Recent reviews:
[15]
[15a]
[23]
[24]
Late Transition Metal Polymerization Ca-
talysis (Eds.: B. Rieger, L. S. Baugh, S. Kacker, S. Striegler),
Wiley-VCH, Weinheim, 2003. [15b]V. C. Gibson, S. K. Spitz-
[15c]
messer, Chem. Rev. 2003, 103, 283.
S. Mecking, Angew.
Chem. 2001, 113, 550; Angew. Chem. Int. Ed. 2001, 40, 534.
[15d] S. D. Ittel, L. K. Johnson, M. Brookhart, Chem. Rev. 2000,
[25]
[26]
[15e]
100, 1169.
S. Mecking, Coord. Chem. Rev. 2000, 203, 325.
[15f]
G. J. P. Britovsek, V. C. Gibson, D. F. Wass, Angew. Chem.
1999, 111, 448; Angew. Chem. Int. Ed. 1999, 38, 428.
H. W. Gibson, H. R. Jois, J. Org. Chem. 1991, 56, 865.
R. T. Boere, V. Klassen, G. Wolmershäuser, J. Chem. Soc., Dal-
ton Trans. 1998, 4147.
G. M. Sheldrick, Program package SHELXTL V.5.1, Bruker
Analytical X-ray Instruments Inc, Madison, USA 1997.
Received January 28, 2004
[16]
[17]
Early View Article
Published Online May 13, 2004
[18]
March’s Advanved Organic Chemistry: Reactions, Mechanism,
2836
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2827Ϫ2836