Pd(II) complexes of a sterically bulky, benzannulated N-heterocyclic carbene and their catalytic activities in the Mizoroki-Heck reaction

Article abstract of DOI:10.1016/j.jorganchem.2007.04.037

The bis(N,N′-diisopropylbenzimidazolin-2-ylidene)Pd(II) complexes trans-[PdBr₂(ⁱPr₂-bimy)₂] (trans-1) and trans-[PdI₂(ⁱPr₂-bimy)₂] (trans-2) have been prepared in good

Full text of DOI:10.1016/j.jorganchem.2007.04.037

Journal of Organometallic Chemistry 692 (2007) 3606–3613
www.elsevier.com/locate/jorganchem
Pd(II) complexes of a sterically bulky, benzannulated N-heterocyclic
carbene and their catalytic activities in the Mizoroki–Heck reaction
*
Yuan Han, Han Vinh Huynh , Lip Lin Koh
Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543, Singapore
Received 18 February 2007; received in revised form 16 April 2007; accepted 24 April 2007
Available online 1 May 2007
Abstract
The bis(N,N0-diisopropylbenzimidazolin-2-ylidene)Pd(II) complexes trans-[PdBr2(iPr2-bimy)2] (trans-1) and trans-[PdI2(iPr2-bimy)2]
(trans-2) have been prepared in good yields by in situ deprotonation of the corresponding N,N0-diisopropylbenzimidazolium salt
(iPr2-bimyH+Xꢀ) (A: X = Br, B: X = I) with Pd(OAc)2 in DMSO at elevated temperature. Salt metathesis of trans-1 or trans-2 with
AgO2CCF3 in refluxing CH3CN afforded the novel mixed carbene–carboxylato complex cis-[Pd(O2CCF3)2(iPr2-bimy)2] (cis-3). This
halo/trifluorocarboxylato ligand substitution can be regarded as a selective method for the synthesis of cis-configured bis(carbene) com-
plexes. All compounds have been fully characterized by multinuclei NMR spectroscopies and ESI mass spectrometry. X-ray diffraction
studies on single crystals of trans-1, trans-2 and cis-3 revealed a square planar geometry and a fixed orientation of the N-isopropyl sub-
stituents with the C–H protons pointing to the metal center to maximize rare C–Hꢁ ꢁ ꢁPd preagostic interactions. These interactions are
1
also retained in solution as indicated by the large downfield shift of the isopropyl C–H protons in the H NMR spectrum compared to
those in precursor salts A or B. A preliminary catalytic study revealed that all complexes are highly active in the Mizoroki–Heck coupling
of aryl bromides and chlorides. However, these complexes gave slower conversions as compared to catalysts with less bulky benzimidaz-
olin-2-ylidenes. This is most likely due to the steric bulk of the ligands, which hamper a fast reductive formation of catalytically active
Pd(0) species.
ꢀ 2007 Elsevier B.V. All rights reserved.
Keywords: N-Heterocyclic carbene; Palladium; C–C coupling; Homogeneous catalysis; Preagostic interaction
1. Introduction
derived from benzimidazole as a 3rd class of NHCs, on
the other hand, have received less attention. This is in par-
ticular surprising, since benzannulated NHCs exhibit inter-
esting properties due to their intermediate position between
saturated and unsaturated analogues [3]. Only recently, we
and others reported the catalytic activities of Pd(II) benz-
imidazolin-2-ylidene complexes [4]. Most of these examples
contain non-bulky carbenes as ancillary ligands. As the ste-
ric bulk is believed to promote the reductive elimination
step occurring in the catalytic cycle of Heck-type reactions,
we became interested in benzimidazolin-2-ylidenes bearing
bulky N-substituents. Recently, we reported the synthesis
and catalytic activities of monocarbene–palladium(II)
complexes with the bulky N,N0-diisopropylbenzimidazo-
lin-2-ylidene ligand (iPr2-bimy), which also exhibited rare
preagostic C–Hꢁ ꢁ ꢁPd and Ccarbene–Br interactions [4b]. To
N-Heterocyclic carbenes (NHCs) and their transition
metal complexes are currently the focus of intense research
in organometallic chemistry and homogeneous catalysis [1].
In particular, palladium(II) carbene complexes derived
from imidazole and imidazoline precursors have been suc-
cessfully developed as highly active precatalysts for a wide
range of Heck-type C–C coupling reactions and CO-olefin
co-polymerization. Such complexes offer the distinctive
advantage of greater stability over the classical Pd/phos-
phine systems as the latter usually suffer from sensitivity
to air and moisture [2]. Catalytic applications of carbenes
*
Corresponding author. Tel.: +65 6516 2670; fax: +65 6779 1691.
E-mail address: chmhhv@nus.edu.sg (H.V. Huynh).
0022-328X/$ - see front matter ꢀ 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2007.04.037

Y. Han et al. / Journal of Organometallic Chemistry 692 (2007) 3606–3613
3607
expand the scope of benzimidazolin-2-ylidene complexes in
catalysis, we herein describe the synthesis and structural
characterization of palladium(II) bis(carbene) complexes
of this unique ligand as well as their catalytic activities in
the Mizoroki–Heck reaction.
complexes. Upon prolonged standing of the filtrate, a sec-
ond crop of precipitated trans-1 can be isolated, which has
most likely formed by isomerization of its cis-isomer. This
observation is in agreement with our recent report on a sol-
vent-controlled cis–trans isomerization equilibrium of
related Pd(II) complexes [4d].
1
2. Results and discussion
The formation of trans-1 was confirmed by H NMR
spectroscopy, which shows the absence of the NCHN
proton. In addition, the isopropyl C–H resonance is signif-
icantly shifted downfield upon coordination from 5.21 ppm
2.1. Preparation and characterization of the complexes
In general, Pd(II) bis(carbene) complexes can be pre-
pared by in situ deprotonation of azolium salts with
Pd(OAc)2 in a 2:1 ratio at moderate temperatures. As
we reported recently, however, initial attempts to synthe-
size a bis(N,N0-diisopropylbenzimidazolin-2-ylidene)Pd(II)
complex by reacting Pd(OAc)2 with N,N0-diisopropylbenz-
imidazolium bromide (A) in various organic solvents
(THF, CH3CN and DMSO) and at various temperatures
(30–90 ꢁC) were unsuccessful. Instead, these reactions
afforded complicated mixtures with the dimeric monocar-
bene complex [PdBr2(iPr2-bimy)]2 as a major component
[4b]. Similar binuclear complexes have been reported to
be intermediate species to the formation of bis(carbene)
complexes [5]. Apparently, the deprotonation of salt A
proves more difficult and would require more drastic reac-
tion conditions due to the +I-effect and the steric bulk of
the N-isopropyl substituents as compared to benzimidazo-
lium salts bearing N-methyl or N-methylene groups.
Indeed, the targeted dicarbene complex trans-[PdBr2(iPr2-
bimy)2] (trans-1) was obtained in 60% yield when the reac-
tion was carried out in DMSO at 120 ꢁC (Scheme 1). We
have also observed traces of Pd black that forms due to
the harsh reaction conditions. The choice of the solvent
used is crucial for the isolation of complex trans-1, which
is soluble in halogenated solvents, but only sparingly solu-
ble in more polar solvents such as DMSO and DMF. Con-
sequently, by employing DMSO as the reaction media pure
trans-1 precipitates from the initially clear yellow solution
and can be isolated by a simple filtration-step. The filtrate,
on the other hand, is a mixture as indicated by three major
sets of signals in the 1H NMR spectrum. These signals were
assigned to the dimeric complex [PdBr2(iPr2-bimy)]2
[4b], trans-1 and its cis-isomer based on comparison with
1H NMR spectra of the pure samples of the first two
in the precursor salt
A
to 6.25 ppm in trans-1
(DdH = 1.04 ppm). The large chemical shift of these pro-
tons is presumably caused by some type of C–Hꢁ ꢁ ꢁPd inter-
actions. In literature, a few metal-hydrogen interactions
(X–Hꢁ ꢁ ꢁM, X = C, N) including agostic [6] and preagostic
interactions [7] as well as hydrogen bonding [6c,8] have
been described. Although there is no clear cut between
these types of interactions, each of them has signature spec-
troscopic and geometric properties, which allow them to be
generally distinguished from each other. Generally, an ago-
stic interaction is described as 3-center-2-electron bond,
which leads to a highfield shift of the corresponding hydro-
gen. Typical metal–hydrogen bonds on the other hand are
linear 3-center-4-electron interactions leading to a down-
field shift. In comparison, the significant downfield chemi-
cal shift of the isopropyl C–H protons together with the
non-linear C–Hꢁ ꢁ ꢁPd geometry (vide infra) observed for
complex trans-1 best fit the definition of a C–Hꢁ ꢁ ꢁPd preag-
ostic interaction. We have observed a similar, but slightly
more pronounced downfield shift of the isopropyl C–H
protons in Pd(II) monocarbene complexes of the same
ligand [4b]. The smaller shift in bis(carbene) complex
trans-1 is most likely due to the competition of 4 compared
to only 2 C–H protons in monocarbene complexes. Such an
interaction has recently also been observed in Rh(I) com-
plexes of NHC ligands [9]. However, the origin of such
interactions is still under debate and may involve filled
dz2 or dxz/yz metal orbitals [6c,7c,7d,7e]. Finally, the 13C
NMR spectrum of trans-1 shows the carbene carbon reso-
nance at 180.0 ppm, which falls well within the reported
range for trans-configured benzimidazolin-2-ylidene Pd(II)
complexes [4c,10].
Single crystals of trans-1 suitable for X-ray diffraction
studies were obtained by evaporation of a CHCl3 solution
N
N
X
Pd
X
0.5 eq. Pd(OAc)2
DMSO
2 AgO2CCF3
N
N
N
N
N
N
O2CCF3
O2CCF3
Pd
N
CH3CN
X
N
3 h
120 C,
°
,
100 C overnight
°
A: X = Br
B: X = I
trans-1: X = Br
trans-2: X = I
cis-3
Scheme 1. Synthesis of bis(carbene) Pd(II) complexes (trans-1, trans-2, cis-3).

3608
Y. Han et al. / Journal of Organometallic Chemistry 692 (2007) 3606–3613
Table 2
Selected crystallographic data for complexes trans-1,trans-2 and cis-3
trans-1
trans-2
cis-3
Formula
Formula weight 670.81
C
26H36Br2N4Pd C26H36I2N4Pd
C30H36F6N4O4Pd
737.03
764.79
Color, habit
Pale yellow,
block
0.50 · 0.40 · 0.34 0.34 · 0.30 · 0.16 0.20 · 0.06 · 0.06
Yellow, block
Colorless, block
Crystal size
(mm)
Temperature
(K)
295(2)
223(2)
223(2)
Crystal system Orthorhombic
Orthorhombic
Pbca
Monoclinic
C2/c
21.057(3)
10.773(2)
14.149(2)
90
101.415
90
3146.2(8)
4
Space group
a (A)
Pbca
Fig. 1. Molecular structure of the complex trans-1.
˚
17.2150(9)
9.4878(5)
17.8106(9)
90
90
90
17.774(2)
9.402(1)
18.014(2)
90
90
90
3010.3(6)
4
1.687
˚
b (A)
˚
c (A)
at ambient temperature, and its molecular structure is
shown in Fig. 1. Selected bond parameters are summarized
in Table 1 and crystallographic data are listed in Table 2.
The palladium center is coordinated by two carbene and
two bromo ligands in a square-planar fashion. As found
by 13C NMR spectroscopy in solution, the two carbene
ligands are arranged trans to each other with an ideal angle
of 180ꢁ due to symmetry. Both carbene ring planes are ori-
ented almost perpendicular to the PdC2Br2 plane with a
torsion angle of 88.0ꢁ. More importantly, the Pd–C bonds
a (ꢁ)
b (ꢁ)
c (ꢁ)
3
˚
V (A )
2909.0(3)
4
Z
Dc (g cmꢀ3
)
1.532
1.556
Radiation used Mo Ka
Mo Ka
2.687
2.26–27.50
19972
Mo Ka
0.667
1.97–27.49
11028
l (mmꢀ1
)
3.405
h Range (ꢁ)
Unique data
Maximum,
minimum
2.29–27.50
33420
0.3907, 0.2809
0.6731, 0.4619
0.9611, 0.8782
˚
˚
in trans-1 (2.017(2) A) are longer than those (1.947(3) A) in
the dimeric [PdBr2(iPr2-bimy)]2 complex. This is in line
with the increasing steric bulk as well as the less Lewis
acidic palladium center resulting from the coordination
of an additional carbene ligand. It is also noteworthy, that
the C–H protons of the isopropyl groups are all oriented
towards the metal center resulting in relatively short C–
transmission
Final R indices R1 = 0.0272,
R1 = 0.0297,
WR2 = 0.0719
R1 = 0.0367,
R1 = 0.0571,
WR2 = 0.1268
R1=0.0625,
[I > 2r(I)]
R indices (all
data)
WR2 = 0.0691
R1 = 0.0344,
WR2 = 0.0729
Goodness-of-fit 1.060
on F2
WR2 = 0.0754
1.053
WR2 = 0.1293
1.257
˚
Hꢁ ꢁ ꢁPd distances of 2.6856(1) and 2.7143(1) A, respec-
Largest
difference in
peak/hole
1.040/ꢀ0.800
1.210/ꢀ0.665
1.373/ꢀ2.080
ꢀ3
˚
Table 1
[e A
]
˚
Selected bond lengths (A) and angles (ꢁ) for trans-1, trans-2 and cis-3
trans-1
trans-2
cis-3
tively. These structural properties support the aforemen-
tioned C–Hꢁ ꢁ ꢁPd preagostic interactions as indicated by
1H NMR spectroscopy. A comparison of structural and
spectroscopic data supporting these interactions for all
complexes discussed here is shown in Table 3.
In order to investigate the effects of different halo
ligands on the Mizoroki–Heck coupling reaction as well
as to have a more direct comparison of the catalytic activities
with the previously reported complex trans-[PdI2(Me2-
bimy)2] [4d], we also prepared the iodo analogue trans-
Pd1–C1
Pd1–Br1
Pd1–I1
Pd1–O1
N1–C1
N1–C2
N1–C8
N2–C1
2.017(2)
2.4260(3)
–
2.018(3)
–
2.5906(3)
–
1.348(3)
1.400(3)
1.469(4)
1.350(3)
1.397(4)
1.480(3)
1.391(4)
–
1.975(4)
–
–
–
2.067(3)
1.352(5)
1.397(5)
1.481(5)
1.349(5)
1.392(5)
1.483(5)
1.391(6)
–
1.344(3)
1.392(3)
1.477(3)
1.347(3)
1.395(3)
1.472(3)
1.392(3)
90.63(6)
89.37(6)
–
N2–C7
N2–C11
C2–C7
C1–Pd1–Br1
C1–Pd1–Br1A
C1–Pd1–I1
C1–Pd1–I1A
C1–Pd1–C1A
O1–Pd1–O1A
C1–Pd1–O1
C1–Pd1–O1A
C1–N1–C2
C1–N2–C7
N1–C1–N2
PdC2X2/carbene dihedral angle
(X = Br, I or O)
–
–
–
–
89.15(7)
90.85(7)
–
–
–
–
–
–
–
Table 3
95.2(2)
82.1(2)
91.4(1)
91.4(1)
109.7(3)
110.0(3)
107.4(3)
60.7ꢁ
Comparison of selected structural and spectroscopic data for complexes
trans-1, trans-2 and cis-3 supporting C–Hꢁ ꢁ ꢁPd preagostic interactions
˚
Complex d(C–Hꢁ ꢁ ꢁPd) (A)
h(C–Hꢁ ꢁ ꢁPd) (ꢁ)
dH (DdH)a (ppm)
–
–
trans-1
trans-2
cis-3
a
2.6856(1), 2.7143(1) 124.0(2), 124.1(2) 6.25 (1.04)
2.6972(2), 2.7182(2) 123.1(2), 123.3(2) 6.00 (0.79)
2.7112(4), 2.8355(4) 119.4(2), 120.0(2) 5.97 (0.76)
110.4(2)
110.2(2)
107.2(2)
88.0ꢁ
110.1(2)
110.0(2)
107.4(2)
89.3ꢁ
DdH = dH (CHMe2 in complex) ꢀ dH (CHMe2 in salt A) measured in
CDCl3.

Y. Han et al. / Journal of Organometallic Chemistry 692 (2007) 3606–3613
3609
[PdI2(iPr2-bimy)2] (trans-2) using Pd(OAc)2 and two equiv-
alents of salt B (Scheme 1). In contrast to the preparation
of trans-1 there is no evidence of Pd black formation in this
reaction. We believe that this is due to the stabilizing effect
of the softer iodo ligands. The replacement of bromo with
iodo ligands does not affect the NMR spectra to a great
extent. The only notable difference is an upfield shift of
the isopropyl C–H resonance from 6.25 ppm in trans-1 to
6.00 ppm in trans-2, which may be due to weaker C–
Hꢁ ꢁ ꢁPd preagostic interactions caused by the bigger size
of the iodo ligands. Finally, an X-ray diffraction study on
single crystals confirmed that trans-2 is isostructural to
trans-1 with bond parameters in the expected range (Table
1). Therefore, no further comments are required.
imidazolin-2-ylidene ligand. Therefore, this halo/trifluoro-
carboxylato ligand substitution can be regarded as a
selective method for the synthesis of cis-configured bis(car-
bene) complexes.
Single crystals of cis-3 suitable for X-ray diffraction
studies were grown from a concentrated CH3CN solution
and its molecular structure is depicted in Fig. 2. As indi-
1
cated by H and 13C NMR spectroscopy in solution, a
cis configuration is revealed with the palladium center
coordinated by two carbene and two trifluoroacetato
ligands in a distorted square-planar geometry. The two tri-
fluoroacetato ligands coordinate in a monodentate fashion
with their pendent oxygen atoms found in anti conforma-
tion. More importantly, the Pd–C bonds in cis-3 (1.975
˚
Complexes trans-1 and trans-2 offer a convenient entry
to bis(carbene) complexes with more labile co-ligands.
We anticipated that such catalyst precursors would offer
the advantage of a faster activation step, since labile co-
ligands can be more easily replaced by reducing agents to
form catalytically active Pd(0) species. Therefore, trans-1
(or trans-2) was reacted with two equivalents of AgO2CCF3
in refluxing CH3CN (Scheme 1) according to a reported
method [4c]. This reaction afforded the mixed di(trifluoro-
acetato)-bis(carbene) complex cis-[Pd(O2CCF3)2(iPr2-
bimy)2] (cis-3) in 75% yield. Complex cis-3 is well soluble
in halogenated solvents, acetone, CH3CN, DMSO and
DMF, but insoluble in nonpolar solvents such as diethyl
A) are shorter than those found in trans-configured ana-
˚
logues (ꢂ2.04 A) [12], reflecting a strong trans influence
of the carbene ligands and a preference for the cis arrange-
ment. Remarkably, the dihedral angle between the carbene
ring plane and the PdC2O2-coordination plane has signifi-
cantly decreased from 88.0ꢁ in the precursor trans-1 to
60.7ꢁ in cis-3 in order to minimize ligand–ligand repulsion.
2.2. Catalytic studies
In a recent study, we have reported the catalytic activity
of Pd(II) benzimidazolin-2-ylidene complexes with non-
bulky N-methyl substituents in the Mizoroki–Heck reac-
tion [4c,4d]. Similarly, the dicarbene complexes trans-1,
trans-2 and cis-3 have been subjected to a catalytic investi-
gation in order to determine the influence of more bulky
N-substituents in benzimidazolin-2-ylidene supported cata-
lysts. In addition, the previously reported mixed carbene–
phosphine complex cis-[PdBr2(iPr2-bimy)(PPh3)] (cis-4)
[4b] was also included in this study for comparison. The
coupling of aryl bromides and chlorides with tert-butyl
acrylate in DMF under air with 1 mol% catalyst loading
and a reaction time of 24 h was chosen as a standard test
reaction and the results are summarized in Table 4. Entries
1–4 show that all four complexes can couple activated 4-
bromobenzaldehyde in quantitative yield at 120 ꢁC. How-
ever, the coupling reactions of deactivated 4-bromoanisole
1
ether, hexane and toluene. The H NMR spectrum of cis-
3 in CDCl3 shows two doublets of equal intensity at 1.72
and 1.33 ppm suggesting two inequivalent Me-groups. Cor-
respondingly, two singlets at 21.4 and 20.3 ppm for the Me-
groups are found in the 13C NMR spectrum. These obser-
vations are in agreement with a sterically hindered rotation
of the Pd–Ccarbene bond in the more congested cis configu-
ration. The resonance for the isopropyl C–H protons are
found at 5.97 ppm indicating slightly weaker C–Hꢁ ꢁ ꢁPd
preagostic interactions in cis-3 as compared to in trans-1
again due to sterical reasons. Furthermore, the 13C NMR
spectrum shows a significant upfield shift of the carbene
carbon resonance from 180.0 ppm in trans-1 (or
177.9 ppm in trans-2) to 164.8 ppm in cis-3 due to the
shielding effect of fluorocarboxylato ligands [11]. The 13C
2
NMR signals for the CO groups (162.3 ppm, q, J(C,F)
1
= 36.0 Hz) and CF3 groups (116.0, q, J(C,F) = 289.1 Hz)
of the trifluoroacetato ligands show the expected splitting
pattern due to 13C–19F heteronuclear couplings, which fall
in the range of 2J(C,F) = 35–37 Hz and 1J(C,F) = 264–
290 Hz. The formation of cis-3 was further confirmed by
ESI mass spectrometry. The positive mode mass spectrum
shows a base peak for the [MꢀO2CCF3]+ fragment origi-
nating from loss of one trifluoroacetato ligand. In agree-
ment with the previous results in our group [4c], it is
noteworthy that the formation of cis-3 from trans-1 (or
trans-2) supports the proposal, that the breaking of the
Pd-halo bond and the subsequent coordination of the more
labile trifluoroacetato ligand thermodynamically favor a cis
arrangement due to the strong trans influence of the benz-
Fig. 2. Molecular structure of the complex cis-3.

3610
Y. Han et al. / Journal of Organometallic Chemistry 692 (2007) 3606–3613
Table 4
The Mizoroki–Heck reactions catalyzed by trans-1, trans-2, cis-3 and cis-4a
O
Cat. 1 mol%
O
O
+
R
X
DMF, NaOAc,
-HX
O
R
X=Br,Cl
R=CHO,CH3CO, OCH3
Entry
Catalyst
Aryl halide
Tempertaure (ꢁC)
Yield (%)b
1
2
3
trans-1
trans-2
cis-3
4-Bromobenzaldehyde
4-Bromobenzaldehyde
4-Bromobenzaldehyde
4-Bromobenzaldehyde
4-Bromoanisole
4-Bromoanisole
4-Bromoanisole
4-Bromoanisole
4-Chlorobenzaldehyde
4-Chlorobenzaldehyde
4-Chlorobenzaldehyde
4-Chlorobenzaldehyde
4-Chloroacetophenone
4-Chloroacetophenone
4-Chloroacetophenone
4-Chloroacetophenone
120
120
120
120
140
140
140
140
140
140
140
140
140
140
140
140
100
100
100
100
72
79
71
75
90
93
91
88
82
83
69
51
4
cis-4
5c
trans-1
trans-2
cis-3
6c
7c
8c
cis-4
9c
trans-1
trans-2
cis-3
10c
11c
12c
13c
14c
15c
16c
a
cis-4
trans-1
trans-2
cis-3
cis-4
Reaction conditions generally not optimized.
Yields were determined by 1H NMR spectroscopy for an average of at least two runs.
With addition of 1.5 equivalents of [N(n-C4H9)4]Br.
b
c
and activated aryl chlorides are more difficult and require
the addition of [N(n-C4H9)4]Br and a higher temperature
of 140 ꢁC to afford moderate to very good conversions
(Entries 5–16). Reactions catalyzed by complex trans-2
showed overall the best yields. The slight superiority of
trans-2 over trans-1 can be traced back to its higher stabil-
ity (vide supra), which may lead to a higher concentration
of active catalyst. Furthermore, it is noteworthy, that the
mixed carbene–phosphine complex cis-4, in contrast to
the results reported by Herrmann et al. on a imidazolin-
2-ylidene analogue [13], did not exhibit a better catalytic
activity than the dicarbene complexes trans-1, trans-2 and
cis-3 under aerobic conditions. Instead, cis-4 afforded the
lowest yields in the coupling of aryl chlorides (Entries
9–16). This may be due to its lower stability compared to
those of dicarbene complexes under the aerobic and rela-
tively harsh reaction conditions, and thus leading to a more
rapid decomposition of the catalyst or its precursor.
In an attempt to establish a reaction profile, the coupling
of 4-bromobenzaldehyde with tert-butyl acrylate with
1 mol% trans-2 has also been monitored. Surprisingly, this
study revealed an unexpectedly slow conversion with only
6% yield after 4 h. This observation is in drastic contrast to
the same reaction catalyzed by trans-[PdI2(Me2-bimy)2],
where a complete conversion was achieved within 90 min
[4d]. The two complexes only differ in their N-substituents,
which in general are believed to have a limited effect on the
donor ability of NHCs derived from imidazole and imidazo-
lines [14]. Electronic effects of N-substituents on benzimid-
azole-based NHCs, on the other hand, have not been
investigated in detail yet. We have recently proposed that
the formation of catalytically active Pd(0) species occurs
more efficiently from cis-configured bis(carbene) complexes.
This is because the strong trans effect of the NHC in such
complexes facilitates dissociation of the halides, therefore
allowing easier attack of the reducing agent. On the other
hand, a Pd–halide bond cleavage is less favored in trans-
bis(carbene) complexes and may account for an induction
period in which trans–cis isomerization step might occur to
facilitate the Pd–halide bond cleavage prior to the reduction
step [4d]. Apparently, such a process is strongly hampered by
the steric bulk of the isopropyl substituents in trans-2 and
may explain the observed slower conversion compared to
that observed for reactions catalyzed by trans-[PdI2(Me2-
bimy)2]. To find further support for this proposal, the behav-
ior of cis-4 was investigated under the same conditions. Here
a higher, but yet low yield of 14% was obtained, indicating
that the steric bulk around the metal center also makes the
approach of the reducing agent difficult. With addition of
sodium formate as reducing agent the reactions can be accel-
erated for all 4 complexes (Entries 3–6, Table 5). Overall, the
results summarized in Table 5 show that the reactions cata-
lyzed by cis-configured complexes (Entry 5/6) are generally
faster than those catalyzed by trans-complexes (Entry 3/4).

Y. Han et al. / Journal of Organometallic Chemistry 692 (2007) 3606–3613
3611
Table 5
The effects of sodium formate addition and trans/cis configuration on the Mizoroki–Heck coupling reactions
O
H
O
Cat. 1 mol%
O
+
Br
H
DMF, NaOAc,
O
O
120
C, 4 h
°
O
Entry
Catalyst
Yield (%)a
1
2
trans-1
cis-4
trans-1
trans-2
cis-3
6
14
18
16
58
50
3b
4b
5b
6b
a
cis-4
Yields were determined by 1H NMR spectroscopy.
With addition of 2 mol% of sodium formate.
b
This result is in line with the easier dissociation of the halo or
carboxylato ligands in cis-configured complexes due to the
strong trans effect of the carbene or phosphine ligand and
therefore facilitating the reduction of Pd(II) to Pd(0) [4d].
N,N0-diisopropylbenzimidazolium bromide A and cis-4
were prepared according to a literature procedure [4b].
1H, 13C and 19F NMR spectra were recorded on a Bruker
ACF 300 spectrometer and the chemical shifts (d) were
internally referenced by the residual solvent signals relative
to tetramethylsilane (1H, 13C) or externally to CF3CO2H
(19F). Mass spectra were measured using a Finnigan
MAT LCQ (ESI) spectrometer. Elemental analyses were
performed on a Perkin–Elmer PE 2400 elemental analyzer
at the Department of Chemistry, National University of
Singapore.
3. Conclusion
In summary, we have synthesized and structurally char-
acterized Pd(II) bis(carbene) complexes of the sterically
bulky N,N0-diisoproylbenzimidazolin-2-ylidene ligand.
The complexes show rare intramolecular C–Hꢁ ꢁ ꢁPd preag-
ostic interactions as indicated by significant downfield
shifts of the isopropyl C–H resonances and a common fixed
orientation of the isopropyl substituents in their solid state
structures. A preliminary catalytic study revealed that the
complexes are highly active in the Mizoroki–Heck coupling
of aryl bromides and chlorides. However, the steric bulk of
the ligands is believed to hamper a fast reductive formation
of catalytically active Pd(0) species and thus lead to a
slower conversion as compared to catalysts with less bulky
benzimidazolin-2-ylidenes. Electronic contributions origi-
nating from the +I-effect of the isopropyl groups resulting
in a slower conversion are less likely, but cannot be fully
excluded. Investigations are currently underway to gain a
better understanding about electronic effects of N-substitu-
ents on catalytic activities.
4.2. Synthesis of N,N0-diisopropylbenzimidazolium iodide
(B)
Salt B was prepared in analogy to a literature procedure
[4b] from benzimidazole (354 mg, 3 mmol), K2CO3
(456 mg, 3.3 mmol) and isopropyl iodide (1.8 ml, 18 mmol).
Yield: 894 mg, 2.7 mmol, 90%. 1H NMR (300 MHz, CDCl3,
25 ꢁC): d = 10.79 (s, 1 H, NCHN), 7.83 (dd, 2 H, Ar–H), 7.64
(dd, 2 H, Ar–H), 5.21 (m, 3J(H,H) = 6.7 Hz, 2 H,
NCH(CH3)2), 1.86 (d, 3J(H,H) = 6.7 Hz, 12 H, CH3).
13C{1H} NMR (75.47 MHz, CDCl3, 25 ꢁC): d = 139.5
(s, NCHN), 130.9, 127.2, 114.1 (s, Ar–C), 52.5
(s, NCH(CH3)2), 22.3 (s, CH3). Anal. Calc. for C13H19IN2:
C, 47.29; H, 5.80; N, 8.48. Found: C, 47.46; H, 6.02; N,
8.49%. MS (ESI): m/z = 203 [MꢀI]+.
4. Experimental
4.3. Synthesis of trans-dibromo-bis(N,N0-diisopropyl-
benzimidazolin-2-ylidene)palladium(II) (trans-1)
4.1. General considerations
Unless otherwise noted all operations were performed
without taking precautions to exclude air and moisture.
N,N-Dimethylformamide used for the Mizoroki–Heck
reaction was purchased from J.T. Baker (‘‘Baker analyzed’’
ACS reagent). All solvents were used as received. Pd(OAc)2
was purchased from Alfa Aesar. Silver trifluoroacetate was
obtained from Fluka. All chemicals were used as received
without any further treatment if not noted otherwise.
A mixture of A (283 mg, 1 mmol) and Pd(OAc)2
(112 mg, 0.5 mmol) was dissolved in wet DMSO (6 ml)
and stirred at 120 ꢁC for 3 h and then at 100 ꢁC overnight.
The yellow precipitate obtained was filtered off and washed
with small portions of DMSO and diethyl ether. It was
then dried in vacuo to give the product as a yellow powder.
Upon stirring the DMSO-filtrate at 100 ꢁC overnight a sec-
ond crop of the product can be obtained giving an overall

3612
Y. Han et al. / Journal of Organometallic Chemistry 692 (2007) 3606–3613
1
yield of 60% (200 mg, 0.30 mmol). H NMR (300 MHz,
CDCl3, 25 ꢁC): d = 7.59 (dd, 4 H, Ar–H), 7.22 (dd, 4 H,
Ar–H), 6.25 (m, 3J(H,H) = 7.1 Hz, 4 H, NCH(CH3)2),
1.87 (d, 3J(H,H) = 7.1 Hz, 24 H, CH3). 13C{1H} NMR
(75.47 MHz, CDCl3, 25 ꢁC): d = 180.0 (s, NCN), 133.8,
122.1, 112.7 (s, Ar–C), 54.1 (s, NCH(CH3)2), 21.3 (s,
CH3). Anal. Calc. for C26H36Br2N4Pd: C, 46.55; H, 5.41;
N, 8.35. Found: C, 46.63; H, 5.59; N, 8.32%. MS (ESI):
m/z = 591 [MꢀBr]+.
The reaction mixture was vigorously stirred at the appro-
priate temperature. After the desired reaction time, the
solution was allowed to cool. Ten milliliters of dichloro-
methane was added to the reaction mixture and the organic
phase was extracted with water (6 · 5 ml) and dried over
MgSO4. The solvent was removed by evaporation to give
a crude product, which was analyzed by 1H NMR
spectroscopy.
4.7. X-ray diffraction studies
4.4. Synthesis of trans-diiodo-bis(N,N0-diisopropyl-
benzimidazolin-2-ylidene)palladium(II) (trans-2)
Diffraction data for trans-1, trans-2 and cis-3 were col-
lected with a Bruker AXS APEX CCD diffractometer
equipped with a rotation anode at 223(2) or 296(2) K using
Complex 2 was prepared in analogy to 1 from B
(528 mg, 1.6 mmol) and Pd(OAc)2 (180 mg, 0.8 mmol).
Yield: 390 mg, 0.51 mmol, 64%. 1H NMR (300 MHz,
CDCl3, 25 ꢁC): d = 7.57 (dd, 4 H, Ar–H), 7.20 (dd, 4 H,
Ar–H), 6.00 (m, 3J(H,H) = 7.1 Hz, 4 H, NCH(CH3)2),
1.80 (d, 3J(H,H) = 7.1 Hz, 24 H, CH3). 13C{1H} NMR
(75.47 MHz, CDCl3, 25 ꢁC): d = 177.9 (s, NCN), 134.0,
122.0, 112.8 (s, Ar–C), 54.0 (s, NCH(CH3)2), 20.6 (s,
CH3). Anal. Calc. for C26H36I2N4Pd: C, 40.83; H, 4.74;
N, 7.33. Found: C, 40.74; H, 4.63; N, 7.20%. MS (ESI):
m/z = 637 [MꢀI]+.
˚
graphite monochromated Mo Ka radiation (k = 0.71073 A).
Data were collected over the full sphere and were corrected
for absorption. Structure solutions were found by the
Patterson method. Structure refinement was carried out
by full-matrix least squares on F2 using SHELXL-97 [15] with
first isotropic and later anisotropic displacement parame-
ters for all non-hydrogen atoms. A summary of the most
important crystallographic data is given in Table 2.
5. Supplementary material
4.5. Synthesis of cis-di(trifluoroacetato)-bis(N,N0-
diisopropylbenzimidazolin-2-ylidene)palladium(II) (cis-3)
CCDC 632107, 632108 and 632106 contain the supple-
mentary crystallographic data for trans-1, trans-2 and cis-
3. These data can be obtained free of charge via http://
www.ccdc.cam.ac.uk/conts/retrieving.html, or from the
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: (+44) 1223-336-033; or e-
mail: deposit@ccdc.cam.ac.uk.
A mixture of complex 1 (112 mg, 0.17 mmol) and
AgO2CCF3 (81 mg, 0.37 mmol) was suspended in acetoni-
trile (15 ml) and refluxed overnight shielded from light.
The resulting suspension was filtered over celite and the
solvent was removed in vacuo to give the crude product
as yellowish powder. Slow evaporation at ambient temper-
ature of a concentrated acetonitrile solution afforded the
product as colorless crystals (94 mg, 0.128 mmol, 75%).
1H NMR (300 MHz, CDCl3, 25 ꢁC): d = 7.61 (dd, 4 H,
Acknowledgement
We thank the National University of Singapore for
financial support and technical assistance from our
department.
3
Ar–H), 7.28 (dd, 4 H, Ar–H), 5.97 (m, J(H,H) = 7.0 Hz,
4 H, NCH(CH3)2), 1.72 (d, 3J(H,H) = 7.0 Hz, 12 H,
CH3), 1.33 (d, 3J(H,H) = 7.0 Hz, 12 H, CH3). 13C{1H}
NMR (75.47 MHz, CDCl3, 25 ꢁC): d = 164.8 (s, NCN),
References
2
162.3 (q, J(C,F) = 36.0 Hz, CF3CO), 132.8, 123.5 (s, Ar–
[1] (a) F.E. Hahn, Angew. Chem., Int. Ed. 45 (2006) 1348;
(b) K.J. Cavell, D.S. McGuinness, Coord. Chem. Rev. 248 (2004)
671;
1
C), 116.0 (q, J(C,F) = 289.1 Hz, CF3), 113.5 (s, Ar–C),
54.6 (s, NCH(CH3)2), 21.4, 20.3 (s, CH3). 19F{1H} NMR
(282.38 MHz, CDCl3, 25 ꢁC): 1.83 (s, CF3). Anal. Calc.
for C30H36F6N4O4Pd: C, 48.89; H, 4.92; N, 7.60. Found:
C, 48.57; H, 5.16; N, 7.31%. MS (ESI): m/z = 623
[MꢀO2CCF3]+.
´
(c) V. Cesar, S. Bellemin-Laponnaz, L.H. Gade, Chem. Soc. Rev. 33
(2004) 619;
(d) W.A. Herrmann, Angew. Chem., Int. Ed. 41 (2002) 1290;
(e) A.C. Hillier, G.A. Grasa, M.S. Viciu, H.M. Lee, C. Yang, S.P.
Nolan, J. Organomet. Chem. 653 (2002) 69;
(f) D. Bourissou, O. Guerret, F.P. Gabba¨ı, G. Bertrand, Chem. Rev.
100 (2000) 39;
(g) M. Tamm, F.E. Hahn, Coord. Chem. Rev. 182 (1999) 175;
(h) W.A. Herrmann, C. Ko¨cher, Angew. Chem., Int. Ed. Engl. 36
(1997) 2162.
4.6. General procedure for the Mizoroki–Heck coupling
reaction
[2] W.A. Herrmann, M. Elison, J. Fisher, C. Ko¨cher, G.J. Artus,
Angew. Chem., Int. Ed. Engl. 34 (1995) 2371.
[3] (a) F.E. Hahn, L. Wittenbecher, D. Le Van, R. Fro¨hlich, Angew.
Chem., Int. Ed. 39 (2000) 541;
In a typical run, a reaction vessel was charged with a
mixture of aryl halide (1.0 mmol), tert-butyl acrylate
(1.5 mmol), anhydrous sodium acetate (1.5 mmol), catalyst
(0.01 mmol) and [N(n-C4H9)4]Br (1.5 mmol) (for Entries 5–
16, Table 4). To the mixture was then added DMF (3 mL).
(b) F.E. Hahn, L. Wittenbecher, R. Boese, D. Bla¨ser, Chem. Eur. J. 5
(1999) 1931.

Y. Han et al. / Journal of Organometallic Chemistry 692 (2007) 3606–3613
3613
[4] (a) F.E. Hahn, M.C. Jahnke, T. Pape, Organometallics 25 (2006)
5927;
(d) J.C. Lewis, J. Wu, R.G. Bergman, J. Ellman, Organometallics 24
(2005) 5737;
(b) H.V. Huynh, Y. Han, J.H.H. Ho, G.K. Tan, Organometallics 25
(2006) 3267;
(c) H.V. Huynh, T.C. Neo, G.K. Tan, Organometallics 25 (2006) 1298;
(d) H.V. Huynh, J.H.H. Ho, T.C. Neo, L.L. Koh, J. Organomet.
Chem. 690 (2005) 3854;
(e) F.E. Hahn, M.C. Jahnke, V. Gomez-Benitez, D. Morales-Morales,
T. Pape, Organometallics 24 (2005) 6458;
(f) G. Zou, W. Huang, Y. Xiao, J. Tang, New J. Chem. 30 (2006) 803;
(g) C. Metallinos, F.B. Barret, J.L. Chaytor, M.E.A. Heska, Org. Lett.
6 (2004) 3641;
(e) L. Brammer, Dalton Trans. (2003) 3145.
[8] (a) L. Brammer, J.M. Charnock, P.L. Goggin, R.J. Goodfellow,
T.F. Koetzle, A.J. Orpen, J. Chem. Soc., Chem. Commun. (1987)
443;
(b) J.C. Lee, A.L. Rheingold, B. Muller, P.S. Pregosin, R.H.
Crabtree, J. Chem. Soc., Chem. Commun. (1994) 1021;
(c) L. Brammer, M.C. McCann, R.M. Bullock, R.K. McMullan, P.
Sherwood, Organometallics 11 (1992) 2339;
(d) D. Braga, F. Grepioni, E. Tedesco, Organometallics 16 (1997)
1846.
(h) R. Jackstell, A. Frisch, M. Beller, D. Ro¨ttger, M. Malaun, B.
Bildstein, J. Mol. Cat. A: Chem. 185 (2002) 105;
(i) F.E. Hahn, M.C. Jahnke, T. Pape, Organometallics 26 (2007) 150.
[5] (a) L.J. Xu, W.P. Chen, J.L. Xiao, Organometallics 19 (2000) 1123;
(b) L.J. Xu, W.P. Chen, J.F. Bickley, A. Steiner, J.L. Xiao, J.
Organomet. Chem. 598 (2000) 409.
[9] (a) P. Bazinet, G.P.A. Yap, D.S. Richeson, J. Am. Chem. Soc. 125
(2003) 13314;
´
´
(b) S. Gomez-Bujedo, M. Alcarazo, C. Pichon, E. Alvarez, R.
´
Fernandez, J.M. Lassaletta, Chem. Commun. (2007) 1180.
[10] H.M.J. Wang, I.J.B. Lin, Organometallics 17 (1998) 972.
[11] H.V. Huynh, D. Le Van, F.E. Hahn, T.S.A. Hor, J. Organomet.
Chem. 689 (2004) 1766.
[12] (a) M.M. Konnick, I.A. Guzei, S.S. Stahl, J. Am. Chem. Soc. 126
(2004) 10212;
[6] (a) M. Brookhart, M.L.H. Green, J. Organomet. Chem. 250 (1983)
395;
(b) R.H. Crabtree, Angew. Chem., Int. Ed. Engl. 32 (1993) 789;
(c) W. Yao, O. Eisenstein, R.H. Crabtree, Inorg. Chim. Acta 254
(1997) 105.
(b) N. Kuhn, C. Maichle-Mo¨c¸mer, E. Niquet, I. Walker, Z.
Naturforsch. 57b (2002) 47.
[7] (a) M. Bortolin, U. Bucher, H. Reugger, L.M. Venanzi, A. Albinati,
F. Lianza, Organometallics 11 (1992) 2514;
[13] W.A. Herrmann, V.P.W. Bo¨hm, C.W.K. Gsto¨ttmayr, M. Grosche,
C.-P. Reisinger, T. Weskamp, J. Organomet. Chem. 617–618 (2001)
616.
´
(b) M. Cano, J.V. Heras, M. Maeso, M. Alvaro, R. Fernandez, E.
Pinilla, J.A. Campo, A. Monge, J. Organomet. Chem. 534 (1997) 159;
(c) Y. Zhang, J.C. Lewis, R.G. Bergman, J.A. Ellman, E. Oldfield,
Organometallics 25 (2006) 3515;
[14] R. Dorta, E.D. Stevens, N.M. Scott, C. Costabile, L. Cavallo, C.D.
Hoff, S.P. Nolan, J. Am. Chem. Soc. 127 (2005) 2485.
[15] G.M. Sheldrick, SHELXL-97, Universita¨t Go¨ttingen, Germany, 1997.