Identification of a highly efficient alkylated pincer thioimido- palladium(II) complex as the active catalyst in negishi coupling

Article abstract of DOI:10.1002/chem.200802238

The induction period of Negishi coupling catalyzed by pincer thio-amide-palladium complex 1 was investigated. A heterogeneous mechanism was excluded by kinetic studies and comparison with Negishi coupling reactions promoted by Pd(OAc)₂/Bu₄

Full text of DOI:10.1002/chem.200802238

FULL PAPER
DOI: 10.1002/chem.200802238
Identification of a Highly Efficient Alkylated Pincer Thioimido–
Palladium(II) Complex as the Active Catalyst in Negishi Coupling
AHCTUNGTRENNUNG
[
a, b]
[a]
[a]
[a]
[a]
[a]
Jing Liu,
Haibo Wang, Heng Zhang, Xiaojun Wu, Hua Zhang, Yi Deng,
[b] [a]
Zhen Yang,* and Aiwen Lei*
1
Abstract: The induction period of Ne-
gishi coupling catalyzed by pincer thio-
amide–palladium complex 1 was inves-
tigated. A heterogeneous mechanism
was excluded by kinetic studies and
comparison with Negishi coupling reac-
was observed by in situ IR, H NMR,
low temperatures (08C or À208C). To
evaluate the influence of the catalyst
structure upon the induction period,
complex 9 was prepared, in which the
nBu groups of 1 were displaced by
1
3
and C NMR spectroscopy for the first
time. The reaction of 7 with methyl 2-
iodobenzoate afforded 74% of the
cross-coupled product, methyl 2-meth-
II
ylbenzoate, together with 60% of Pd
more
bulky
1,3,5-trimethylphenyl
tions promoted by Pd
A
H
U
G
R
N
U
G
complex 2. Furthermore, the catalyst,
as an electron-rich Pd species, effi-
ciently promoted the Negishi coupling
of aryl iodides and alkylzinc reagents
without an induction period, even at
groups. Trimer 10 was isolated from
the reaction of complex 9 and basic or-
ganometallic reagents such as CyZnCl
or CyMgCl (Cy: cyclohexyl); this is
consistent with the result obtained with
complex 1. The rate in the induction
period of the model reaction catalyzed
by 9 was faster than that with 1. Plausi-
ble catalytic cycles for the reaction,
based upon the experimental results,
are discussed.
2
4
II
(
a
palladium–nanoparticle system).
Tetramer 2 was isolated from the reac-
tion of 1 and organozinc reagents. Dis-
sociation of complex 2 by PPh3 was
achieved, and the structure of resultant
complex 8 was confirmed by X-ray dif-
fraction analysis. A novel alkylated
Keywords:
ate
complexes
·
coupling reactions · homogeneous
II
pincer thioimido–Pd complex, 7, gen-
catalysis
·
palladium
·
reaction
erated from catalyst precursor 1 and
basic organometallic reagents (RM),
mechanisms
0
Introduction
responding Pd complexes, have the potential to be active
catalysts as well and might offer new opportunities in both
[3–11]
Palladium is among the most common transition-metal cata-
mechanistic studies and application aspects.
For exam-
[1,2]
II
II
IV
lysts in coupling reactions today,
and a majority of Pd-
ple, an active Pd complex might initiate a Pd –Pd catalyt-
ic cycle, which could address some unsolved challenges, such
as the deleterious b-hydride elimination in Csp3-involved
carbon–carbon bond-formation reactions,
reductive elimination of Pd species has been reported to
catalyzed reactions have been well established or accepted
as initiating from Pd species. On the other hand, Pd com-
plexes, more electron deficient yet more stable than the cor-
0
II
[12–17]
because the
IV
IV
be facile due to the electron deficiency of the Pd
[
a] Dr. J. Liu, H. Wang, Dr. H. Zhang, X. Wu, H. Zhang, Y. Deng,
Prof. A. Lei
[18–20]
center.
College of Chemistry and Molecular Sciences
Wuhan University
Wuhan, Hubei, 430072 (China)
Fax : (+86)27-6875-4067
Theoretically, if there were enough and appropriate elec-
tron-donating surrounding ligands, the reaction of aryl hal-
ides and Pd species might be realized. Amid the palladium
II
E-mail: aiwenlei@whu.edu.cn
catalysts utilized in cross-coupling reactions so far, palladi-
um pincer complexes, which have mainly been employed in
Heck and Suzuki processes, have been proposed to involve
[
b] Dr. J. Liu, Prof. Z. Yang
College of Chemistry and Molecular Engineering
Peking University
IV
[10,11]
Pd intermediacy in some cases.
However, to date, con-
Beijing 100871 (China)
Fax : (+86)10-6275-9105
E-mail: zyang@pku.edu.cn
vincing evidence is still required. Proximal thiolate ligands
IV
are known to stabilize high-valency Fe –heme cation radi-
cals through electron donation in biologically significant
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200802238.
[21–25]
metalloenzymatic oxygenase systems.
In stoichiometric
Chem. Eur. J. 2009, 15, 4437 – 4445
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4437

IV
studies, Bennett et al. isolated a Pd complex stabilized by
a tridentate thioether ligand together with alkyl ligands,
which provided further evidence of the stabilizing ability of
IV
sulfur-type ligands towards the electron-deficient Pd
[26]
center. These characteristics inspired us to design and syn-
[27,28]
thesize a pincer thioamide–palladium complex, 1,
the
two NH groups of which were crucial because they could be
deprotonated by organozinc reagents to give the thioimido
[29,30]
complex 2 (Scheme 1).
The thioimido ligand provided
Scheme 1. Pincer thioamide–palladium complex 1 and pincer thioimido–
palladium complex 2. THF: tetrahydrofuran.
one neutral nitrogen atom, one neutral sulfur atom and two
II
sulfur anions as electron donors for the Pd center and pre-
0
vented reduction to Pd species by the organozinc reagents.
The catalytic capabilities of the two sulfur-containing palla-
dium species 1 and 2 were explored in Negishi reactions in-
volving primary and secondary alkylzinc reagents; the reac-
tions occurred readily under mild conditions, substrates with
b-hydrogen atoms were tolerated, and the catalysis was per-
[
31]
formed with unusually high efficiency.
scribe our mechanistic studies of sulfur-containing complex
in the Negishi reaction, which revealed base-induced for-
Herein, we de-
1
II
mation of an alkylated Pd complex, and the subsequent re-
action with aryl iodides.
Figure 1. Conversion of ethyl 2-iodobenzoate (3) into cross-coupled prod-
uct 5, as monitored by the ReactIR system (see the Experimental Sec-
tion), with [3]=0.29m, [4]=0.65m, and [Pd]=1.6 mm (~: [1]=1.6 mm,
2
2
58C; !: [1]=1.6 mm, 08C;
58C; *: Pd
(OAc) ]=1.6 mm, [Bu
4
&
: [Pd
2 4
ACHTUNGTRENNGU( OAc) ]=1.6 mm, [Bu NBr]=0.29m,
Results and Discussion
A
H
U
G
R
N
U
G
2
NBr]=0.29m, 08C).
Testing of the kinetic behavior of the Negishi coupling cata-
lyzed by complex 1: To test the catalytic behavior of com-
plex 1, the coupling between ethyl 2-iodobenzoate (3) and
cyclohexylzinc chloride (4) was chosen as the model reac-
tion, with [3]=0.29m, [4]=0.65m, and [1]=1.6 mm. Al-
The reaction was finished within 2 min without an induction
period at 258C, and the kinetic plot was similar at 08C (Fig-
ure 1B). The induction periods of these two reactions were
not as sensitive to temperature as those involving 1 as the
catalyst precursor (Figure 1).
0
though complex 1 could hardly be reduced to a Pd species
by 4, it promoted the reaction effectively. The reaction cata-
lyzed by 1 was completed within 5 min with 97% yield (as
measured by GC analysis) at room temperature after a
short induction period (5 min), and it also progressed
smoothly at 08C after a long induction period (ꢀ60 min),
Ligand inhibition experiments have been proven to be an
efficient method to clarify whether a reaction was catalyzed
[34,43]
by a homogenous catalyst or NPs.
Figure 2 illustrates
our experimental results. When PPh (0.7 equiv relative to
3
Pd) was added to the mixture of 3, 4, and catalyst 1
(2 mol%), the reaction was completed within 3 min with
[32]
with 95% yield (Figure 1A and B).
9
6% yield (Figure 2). However, PPh (0.5 equiv relative to
3
Exclusion of a heterogeneous mechanism in the catalytic
system of complex 1: The existence of induction periods and
sigmoidal-shaped kinetic profiles in the catalytic system of a
pincer palladium complex pointed to the possibility of the
formation of palladium nanoparticles (PdNPs) as the true
catalytic species. For comparison, Negishi couplings promot-
Pd) significantly inhibited the reaction when Pd ACHTUNGTRENNUNG( OAc)
2
(2 mol%) was used as the catalyst precursor. Within 30 min,
30% of 3 was converted, of which only 14% was converted
into the cross-coupled product 5 (as detected by GC analy-
sis); 16% of the dehalogenated product, ethyl benzoate, was
[44]
formed. Comparison of the two reactions suggested that
[33–42]
ed by ligand-free Pd
A
H
U
G
R
N
N
(OAc) /Bu NBr were examined.
the active catalyst formed from 1 was not a PdNP system.
2
4
4438
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 4437 – 4445

Thioimido–Palladium(II) Complexes in the Negishi Coupling
FULL PAPER
[
50]
Scheme 2. Proposal for ate complex 7 as the active catalyst.
ence of excess alkylzinc halide, complex 6 might be further
alkylated to yield complex 7, which is an anionic complex
Figure 2. Quantitative poisoning experiments: conversion of ethyl 2-iodo-
benzoate (3), as monitored by the ReactIR system (see the Experimental
with two S anions, one N atom, and one Csp anion capable
3
of donating electrons towards the palladium center; hence, 7
Section), with [3]=0.29m and [4]=0.65m (*: [1]=6.4 mm, [PPh
]=
]=
3
4
3
.5 mm, 258C;
.2 mm, 258C).
&
:
[Pd
A
H
U
G
R
N
N
(OAc)
2
]=6.4 mm, [Bu
4
NBr]=0.29 m, [PPh
3
is possibly electronically rich enough to react with aryl io-
[51]
dides.
draw the equilibrium from 2 toward 6. Yet the alkylation of
by the organozinc reagent was a minor process (see
Consumption of 6 by alkylation to form 7 could
6
Furthermore, kinetic measurements were conducted for
the reaction of 3 with excess 4 in the presence of different
amounts of catalyst 1 in THF at 258C, with [3]=0.29m, [4]=
Figure 4 below), probably due to the low reactivity and nu-
cleophilicity of the zinc reagents, and aryl iodide was re-
quired to react with 7 to enable the alkylation to proceed.
0
0
.65m, and [1]=0.16–0.98 mm. A linear plot of the observed
Anionic Pd adducts as precursors for oxidative addition in
[
52–54]
rate (kobs) versus [1] showed the reaction to be first order
with respect to catalyst 1 (see Figure 4 in the Supporting In-
formation). It has been reported that the reaction rate is not
first order with respect to the concentration of the catalyst
catalytic systems have been proposed,
but correspond-
II
[55–57]
ing Pd species have rarely been reported.
Identification of ate complex 7: To explore the feasibility of
[42,45–49]
when NPs are the true catalytic species.
Therefore,
the dissociation process (Scheme 2), 2 was treated with PPh
3
[56,57]
these data further supported the conclusion that 1 was ho-
mogenous rather than heterogeneous in this catalytic
system.
(1 equiv relative to Pd).
The tetrameric structure was
dissociated, and complex 8, the structure of which was con-
Proposal for the induction period: formation of an electron-
rich ate complex as the active catalyst: To probe the induc-
tion period, we investigated stoichiometric reactions of com-
plex 1 and the two reactants 3 and 4. No obvious reaction
was observed when 1 and 3 were mixed together in THF at
between 08C and 608C. However, when zinc reagent 4
(
10 equiv relative to 1) was added in to the system, the reac-
tion progressed smoothly and 74% of the cross-coupled
product 5 was obtained; this indicated the involvement of
cyclohexylzinc chloride (4) in the active-catalyst formation
process.
As mentioned above, direct reaction of 1 with 4 gave the
II
more electron-rich Pd complex 2 instead of reducing 1 to
0
the Pd species (Scheme 1). Yet a long induction period ex-
isted when complex 2 was used as the catalyst precursor
(
see Figure 7 in the Supporting Information), and the stoi-
Figure 3. Single-crystal structure of complex 8.
chiometric reaction of 2 and 3 could only occur in the pres-
ence of excess 4. Therefore, a further reaction of 2 with cy-
clohexylzinc chloride (4) was necessary to generate the
active catalyst.
By comparing the structures of complexes 1 and 2, and on
the basis of the experiments above, we could reasonably
speculate that the alkylzinc reagent should first deprotonate
the NH groups of 1 to generate 6, as shown in Scheme 2.
Aggregation of 6 resulted in the formation of 2. In the pres-
3
Scheme 3. Dissociation of 2 by PPh .
Chem. Eur. J. 2009, 15, 4437 – 4445
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4439

A. Lei, Z. Yang et al.
firmed by X-ray diffraction analysis (Figure 3), was obtained
in quantitative yield (Scheme 3).
In situ monitoring of the reaction of 1 and 4 (10 equiv) re-
tion was a fast process and little or no alkylation to produce
[58]
complex 7 occurred in this system.
1
3
C NMR detection for the reaction of tetramer 2 with
vealed the formation of only one new component in 1 min
MeZnCl (2 equiv) revealed that 2 remained unchanged,
(
1
Figure 4A) with absorbances at n˜ =1606, 1571, and
562 cm (Figure 4B), which tallied with those for complex
which further confirmed the above results (see Figure 10 in
the Supporting Information).
À1
[59]
2
(Figure 5). Thus, the intermediate was assigned as 6 or 2
Grignard reagents are usually more electronegative and
reactive than the corresponding zinc reagents, so we envis-
aged that the formation of 7 would be easier with Grignard
reagents. Hence, the stoichiometric reaction of complex 1
and CyMgCl was monitored by in situ IR spectroscopy, and
two components were observed (A and B in Figure 6A).
because both compounds possess similar functional groups
for IR absorbance. The results indicated that the deprotona-
Figure 4. Reaction of 1 (24.3 mg, 0.05 mmol) with 4 (0.8 mL, 0.5 mmol) in
THF at 258C, as monitored by the ReactIR system (see the Experimental
Section). A) The absorbance of the newly formed component versus
time; B) the ConcIRT spectrum of the newly formed component.
Figure 6. Reaction of 1 (24.3 mg, 0.05 mmol) with CyMgCl (0.2 mL,
0
.28 mmol) in THF, as monitored by the ReactIR system (see the Experi-
mental Section). A) The concentration of components A and B versus
time; B) the ConcIRT spectra of components A and B.
The ConcIRT spectra of component A (Figure 6B) tallied
with the IR spectrum in Figure 4B, which suggested that the
kinetic profile of component A (Figure 6A) represented a
change of complex 6 or 2. The spectrum of the newly
formed component B (Figure 6B) shifted to lower wave-
À1
À1
numbers than that of A (1541 cm instead of 1560 cm ).
This phenomenon was rationalized as arising from the elec-
tron-density increase of the palladium center upon alkyla-
tion to form 7. When CyMgCl was added to the mixture of
complex 1 and CyZnCl, the spectrum of the newly generat-
ed complex was the same as that of component B (see
Figure 9 in the Supporting Information).
Figure 5. The ConcIRT spectrum of tetramer 2.
4440
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 4437 – 4445

Thioimido–Palladium(II) Complexes in the Negishi Coupling
FULL PAPER
To validate the aforementioned assumption, 1 was treated
methyl 2-iodobenzoate (10 equiv) at room temperature and
1
3
with MeMgCl (5 equiv) in THF and monitored by NMR
monitored by C NMR spectroscopy (Figure 8). The peak
at d=À21.7 ppm disappeared as expected; 74% of methyl
2-methylbenzoate, the cross-coupled product, was obtained
13
spectroscopy. C NMR spectroscopy revealed the appear-
ance of a peak at d=À21.7 ppm and supported the exis-
tence of a PdÀMe bond (see the Supporting Information).
1
(as determined by H NMR spectroscopy), together with
II
Importantly, the reaction between 1 and a smaller amount
60% of the Pd complex tetramer 2 (Scheme 4). This
1
3
of MeMgCl (3.2 equiv) afforded a similar C NMR spectra
proved that complex 7 was active toward the aryl iodide.
with a peak at d=À21.7 ppm (see Figure 8A and the Sup-
1
porting Information). In the H NMR spectrum (Figure 7),
Scheme 4. Stoichiometric reaction of complex 7.
Addition of 2,2,6,6-tetramethyl-1-piperidinoxyl (TEMPO;
1
equiv relative to Pd) to the reaction mixture of 1 and
MeMgCl (3.2 equiv) before addition of the aryl iodide had
little influence on the reaction yield. Thus, radical pathways
for the reaction of complex 7 with methyl 2-iodobenzoate
[60]
were disfavored.
To test the catalytic activity of complex 7, 1 (0.5 mol%)
was treated with CyMgBr (2.5 mol%) at 08C for 2 min; this
was followed by immediate addition of CyZnCl (4) and aryl
iodide 3. To our great pleasure, no induction period was ob-
served, just as expected, and the reaction took only 8 min to
reach 100% conversion (98% yield of the cross-coupled
product 5 as monitored by GC analysis; Figure 9). It is note-
worthy that the reaction of 3 and 4 also occurred smoothly
at À208C after catalyst 1 had been pretreated with CyMgCl
at À208C and no induction period existed under these con-
ditions either (Figure 9).
Evaluation of the influence of the catalyst structure upon
the induction period: To explore the influence of the cata-
lyst structure upon the induction period, we prepared com-
plex 9 in which the nBu groups were displaced by more
bulky 1,3,5-trimethylphenyl groups (Scheme 5). A trimer
was isolated from the reaction of complex 9 and basic or-
ganometallic reagents, such as CyZnCl or CyMgCl
1
Figure 7. Selected regions of the H NMR spectrum (600 MHz) following
the reaction of 1 with MeMgCl (3.2 equiv) in THF.
the peak corresponding to 3H atoms at d=0.060 ppm was
assigned to the methyl group attached to the palladium
center. In the lowfield region, a
duplet peak corresponding to
2
H atoms at d=8.028 ppm and
a triplet peak equivalent to 1H
atom at d=7.729 ppm indicated
that the resultant complex was
symmetrical. All of this infor-
mation was consistent with the
proposed structure of the alky-
lated complex 7 (R: Me) shown
in Scheme 2.
Testing of the activity of ate
complex 7: The reaction mix-
ture of complex 1 and MeMgCl
13
Figure 8. A) Selected region of the C NMR spectrum following the reaction of 1 with MeMgCl (3.2 equiv) in
13
THF. B) Selected region of C NMR spectrum of the mixture in I after addition of methyl 2-iodobenzoate
(
3.2 equiv) was treated with (10 equiv).
Chem. Eur. J. 2009, 15, 4437 – 4445
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4441

A. Lei, Z. Yang et al.
faster than that of the latter, whereas the rate in the period
afterwards was slower than that of the latter. Theoretically,
the steric hindrance of the 1,3,5-trimethylphenyl groups
would accelerate the dissociation process of the trimer (see
above) and decelerate the alkylation process to form the ate
complex. Thus, the rate increase in the induction period was
probably due to the acceleration of the dissociation process,
and the rate decrease after the induction period might result
from the lower reactivity of the ate complex because of the
increased steric hindrance and the decreased electron densi-
II
ty of the Pd center.
Discussion about the induction period and the subsequent
steps of the Negishi coupling catalyzed by pincer thioamide–
palladium or pincer thioimido–
Figure 9. Kinetic profiles of the reaction of 3 and 4 at 08C (
À208C (*) catalyzed by complex 7 (R: Cy).
&
) and
palladium complexes: In the
process of investigating the re-
activities of monoorganozinc
and diorganozinc reagents in
the Negishi coupling catalyzed
by pincer thioamide–palladium
Scheme 5. Deprotonation of complex 9.
or pincer thioimido–palladium
(
Scheme 5 and Figure 10), which is consistent with the result
for complex 1 (Scheme 1). The lengths of the PdÀS bonds
broken in the dissociation process were similar (2.314 ꢁ for
2
versus 2.317 ꢁ for 10), but the SÀPdÀS angles were quite
different. The angle in 2 was 112.218, whereas that in 10 was
8.98, which indicates that the ring strain of the latter was
9
greater than that of the former. Also, p–p stacking observed
in complex 2 could decelerate the dissociation of the com-
plex by a stabilization effect, whereas the greater steric hin-
drance around the PdÀS ring in 10 might accelerate the dis-
[61]
sociation process.
Monitoring the model reaction cata-
lyzed by 9 at 158C revealed a kinetic plot different from
that for the reaction catalyzed by complex 1 (Figure 11).
The rate in the induction period of the former reaction was
Figure 11. Conversion of ethyl 2-iodobenzoate (3), as monitored by the
ReactIR system (see the Experimental Section), at 158C with initial con-
centrations of [3]=0.16m, [4]=0.39m, and [catalyst]=1.6 mm.
complexes, two plausible pathways were proposed to ration-
alize the differences between the two type of zinc reagents
[31]
(
Scheme 6). Deprotonation of the catalyst precursor I re-
sulted in monomer III. III was inclined to aggregate to form
complex II and to stay out of the catalytic cycle. However,
under certain conditions, II could dissociate to release III.
III could be alkylated by organometallic reagents, such as
organozinc or Grignard reagents, to afford ate complex IV,
which was proven to be active toward aryl iodides. Reduc-
tive elimination of V afforded the cross-coupled product
ArR and returned to complex III (path A). Alternatively,
complex III might be further alkylated to produce VI, which
[62]
upon reductive elimination would release IV (path B).
The deprotonation process was displayed clearly by the
Figure 10. Single-crystal structure of complex 10.
formation of complexes 2 and 10. Stoichiometric and cata-
4442
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 4437 – 4445

Thioimido–Palladium(II) Complexes in the Negishi Coupling
FULL PAPER
cross-coupled products and dis-
played high efficiency even at
[63]
À208C. The induction period
of the reaction could be influ-
enced by both the alkylation re-
agents and the catalyst struc-
tures. Plausible catalytic cycles
for the reaction were discussed
based upon the experimental
results, and further studies to
clarify the subsequent mecha-
nistic steps are ongoing in our
laboratory and will be reported
in due course.
Scheme 6. Speculative mechanism for the Negishi coupling catalyzed by pincer thioamide–palladium or pincer
thioimido–palladium complexes.
Experimental Section
Toluene and THF were dried and dis-
tilled from sodium/benzophenone im-
mediately prior to use under a nitro-
lytic reactions manifested that II was not the active catalyst
and that dissociation was required to enable the complex to
enter the catalytic cycle. The structure of ate complex IV
was established by spectroscopic data. The reactivity of IV
toward aryl iodides was proven stoichiometrically and cata-
lytically. Radical pathways were disfavored because little in-
fluence of TEMPO was observed in the stoichiometric reac-
tion of the ate complex with aryl iodide. Therefore, Pd inter-
mediate V was proposed.
If V went through direct reductive elimination (path A),
III would be regenerated and the dissociation of II would be
responsible for the induction period. On the other hand, if
V underwent further alkylation to produce VI, IV would be
regenerated after reductive elimination and both the dissoci-
ation of II and the alkylation of III would be responsible for
the induction period.
As shown above, utilization of a more reactive Grignard
reagent (to affect the alkylation process) and the more hin-
dered complex 9 (to affect the dissociation process) both in-
fluenced the induction period. Therefore, path B possibly
occurs in the catalytic process. However, the participation of
path A could not be excluded because the joint operation of
both pathways would be consistent with the experimental
results too.
gen atmosphere. CH
3
CN was dried over anhydrous sodium sulfate
(
Na SO ). Analytical grade solvents and commercially available reagents
2
4
were used as received, unless otherwise stated.
Glass 0.25 mm silica gel plates were employed for thin-layer chromatog-
raphy (TLC). Flash-chromatography columns were packed with 200–
3
00 mesh silica gel or neutral alumina with 200–300 mesh in petroleum
(boiling point=60–908C). Gradient flash chromatography was conducted
by eluting with a continuous gradient from petroleum to the indicated
solvent (v/v ratios are given).
1
13
H and C NMR data were recorded with a Varian Mercury VX300
300 MHz) or a Varian Mercury VX600 (600 MHz) spectrometer. All
(
1
H NMR chemical shifts (d in ppm) are reported relative to the internal
or external standard. Coupling constants (J) are reported in Hertz (Hz).
High-resolution mass spectra (HRMS) were measured with a Waters Mi-
cromass GCT instrument, and accurate masses are reported for the mo-
+
lecular ion [M ]. GC yields were recorded with a Varian GC 3900 gas
chromatography instrument with an FID detector. For the ReactIR ki-
netic experiments, the reaction spectra were recorded by using a IC 10
apparatus from Mettler-Toledo AutoChem fitted with a diamond-tipped
probe. Data manipulation was carried out by using the iC IR (ver-
[
65]
sion 1.05) software.
Crystal diffraction intensity data were collected on a Bruker CCD 4K dif-
fractometer with graphite-monochromatized Mo Ka radiation (l=
0
.71073 ꢁ). Lattice determination and data collection were carried out
[
66]
by using SMART (version 5.625) software. Data reduction and absorp-
tion corrections were performed by using SAINT (version 6.45)
SADABS (version 2.03)
were performed by using the SHELXTL (version 6.14) software pack-
age.
[67]
and
[
68]
software. Structure solution and refinement
[
69]
Representative procedure for the catalytic reaction of 3 and 4 monitored
by ReactIR: The spectra were acquired in 33 scans at a gain of 1 and a
resolution of 2 by using the ReactIR (version 1.05) software. The reaction
was carried out as follows. 1 (2.4 mg, 0.54 mol%) was added to an oven-
dried three-necked reaction vessel fitted with a magnetic stirring bar. The
IR probe was then inserted through an adapter into the middle neck; the
other two necks were capped by rubber septums. The reaction vessel was
evacuated and flushed with nitrogen 3 times, put into a 258C bath, and
charged with cyclohexylzinc chloride (4; 3.0 mL THF solution,
1.95 mmol). The data collection was then started. Ethyl 2-iodobenzoate
Conclusion
In conclusion, a heterogeneous mechanism for the Negishi
coupling catalyzed by pincer thioamide–palladium complex
was excluded by kinetic investigations. A novel alkylated
1
II
pincer thioimido–Pd complex, 7, generated from catalyst
precursor 1 and basic metal reagents, was observed by in
1
13
situ IR, H NMR, and C NMR spectroscopy for the first
time, and it was proven to be the active catalyst by stoichio-
metric and catalytic reactions. The catalyst, as an electron-
(
3; 254 mg, 0.92 mmol) was added to initiate the reaction, and data col-
lection was continued until the reaction was finished. Cross-coupled
[64]
1
product 5 (ethyl 2-cyclohexylbenzoate) was characterized by H NMR
II
13
1
rich Pd species, could react with aryl iodides to afford
and C NMR spectroscopy: H NMR (300 MHz, CDCl
3
): d=7.71 (dd,
Chem. Eur. J. 2009, 15, 4437 – 4445
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4443

A. Lei, Z. Yang et al.
J=7.8, 1.2 Hz, 2H), 7.46–7.36 (m, 2H), 7.23–7.18 (m, 1H), 3.70 (q, J=
[3] M. Catellani, G. P. Chiusoli, C. Castagnoli, J. Organomet. Chem.
7
8
1
.2 Hz, 2H), 3.32–3.25 (m, 1H), 1.89–1.74 (m, 5H), 1.45–1.37 ppm (m,
1991, 407, C30–C33.
[4] G. Bocelli, M. Catellani, S. Ghelli, J. Organomet. Chem. 1993, 458,
1
3
H); C NMR (75.4 MHz, CDCl
3
): d=168.53, 148.16, 131.31, 130.39,
29.58, 126.63, 125.25, 60.73, 40.15, 34.24, 26.85, 26.13, 14.17 ppm.
C12–C15.
[
[
[
5] M. Ohff, A. Ohff, M. E. van der Boom, D. Milstein, J. Am. Chem.
Soc. 1997, 80, 11687–11688.
6] D. Kalyani, N. R. Deprez, L. V. Desai, M. S. Sanford, J. Am. Chem.
Soc. 2005, 127, 7330–7331.
7] O. Daugulis, V. G. Zaitsev, Angew. Chem. 2005, 117, 4114–4116;
Angew. Chem. Int. Ed. 2005, 44, 4046–4048.
Synthesis of Pd complex 8: Complex 2 (41.3 mg, 0.1 mmol Pd) and PPh
26.2 mg, 0.1 mmol) were added to a dry Schlenk tube under N . Purified
THF (1 mL) was then injected, and a red solution was formed. The solu-
tion was stirred overnight and a yellow solid precipitated. The solvent
was removed by rotary evaporation, and the residue was dried under
vacuum for 2 h to afford product 8 as a yellow solid (68.0 mg, 99%). Re-
3
(
2
crystallization from CH
that were suitable for X-ray diffraction: H NMR (300 MHz, CDCl
.11 (d, J=7.5 Hz, 2H), 7.87–7.60 (m, 7H), 7.59–7.29 (m, 9H), 3.55 (t,
J=6.9 Hz, 4H), 1.77–1.64 (m, 4H), 1.52–1.36 (m, 4H), 0.94 ppm (t, J=
2
Cl
2
/hexane gave red block-shaped crystals of 8
[8] R. Giri, N. Maugel, J.-J. Li, D.-H. Wang, S. P. Breazzano, L. B. Saun-
ders, J.-Q. Yu, J. Am. Chem. Soc. 2007, 129, 3510–3511.
[9] R. Giri, J. Liang, J.-G. Lei, J.-J. Li, D.-H. Wang, X. Chen, I. C.
Naggar, C. Guo, B. M. Foxman, J.-Q. Yu, Angew. Chem. 2005, 117,
7586–7590; Angew. Chem. Int. Ed. 2005, 44, 7420–7424.
[10] M. E. van der Boom, D. Milstein, Chem. Rev. 2003, 103, 1759–1792.
[11] M. Albrecht, G. van Koten, Angew. Chem. 2001, 113, 3866–3898;
Angew. Chem. Int. Ed. 2001, 40, 3750–3781.
1
3
): d=
8
1
3
7
1
2
3
.4 Hz, 6H); C NMR (75 MHz, CDCl ): d=169.40, 159.80, 136.56,
34.50, 134.34, 131.17, 130.20, 129.49, 128.46, 128.31, 123.83, 53.05, 32.65,
0.87, 14.05 ppm; HRMS (MALDI; 2,5-dihydroxybenzoic acid): m/z
+
calcd for C33
H
36
N
3
PPdS
2
[M+H] : 675.1123; found: 675.1121.
[
[
12] J. F. Hartwig, Inorg. Chem. 2007, 46, 1936–1947.
13] T.-Y. Luh, M.-k. Leung, K.-T. Wong, Chem. Rev. 2000, 100, 3187–
Formation of ate complex 7 (R: Me): 1 (24.3 mg, 0.05 mmol) was put
into an NMR spectroscopy tube, and the tube was placed in a glove box.
Dry THF (0.1 mL) was added to the tube, followed by dropwise addition
of a THF solution of MeMgCl (80 mL, 0.25 mmol). After the reaction
mixture stopped bubbling, a clear reddish-orange solution was obtained.
3
204.
14] F. Ozawa, T. Ito, Y. Nakamura, A. Yamamoto, Bull. Chem. Soc. Jpn.
981, 54, 1868–1880.
[
1
[
[
15] F. O. Arp, G. C. Fu, J. Am. Chem. Soc. 2005, 127, 10482–10483.
16] M. R. Netherton, G. C. Fu, Adv. Synth. Catal. 2004, 346, 1525–1532.
CDCl
3
sealed in a capillary was put into the tube as an external standard,
the tube was then capped and removed from the glove box, and the con-
1
3
13
[17] J. Zhou, G. C. Fu, J. Am. Chem. Soc. 2003, 125, 14726–14727.
tents of the tube were characterized by C NMR spectroscopy: C NMR
150 MHz, THF): d=176.94, 157.80, 134.05, 121.16, 53.36, 33.44, 21.17,
3.97, À17.204, À21.68 ppm.
Reaction of complex
.05 mmol) was put into an NMR spectroscopy tube, and the tube was
[
[
18] C. Duecker-Benfer, R. van Eldik, A. J. Canty, Organometallics 1994,
3, 2412–2414.
(
1
1
19] B. A. Markies, A. J. Canty, J. Boersma, G. van Koten, Organometal-
lics 1994, 13, 2053–2058.
20] A. Moravskiy, J. K. Stille, J. Am. Chem. Soc. 1981, 103, 4182–4186.
21] Cytochrome P450: Structure, Mechanism, and Biochemistry, 2nd ed.
7
with methyl 2-iodobenzoate: 1 (24.3 mg,
0
[
[
placed in a glove box. Dry THF (0.1 mL) was added to the tube, followed
by dropwise addition of a THF solution of MeMgCl (51 mL, 0.16 mmol).
After the reaction mixture stopped bubbling, a clear reddish-orange solu-
tion was obtained. Methyl 2-iodobenzoate (75 mL, 0.5 mmol) was added,
(
Ed.: P. R. Ortiz de Montellano), Springer, Heidelberg, 1995.
[
[
22] J. H. Dawson, Science 1988, 240, 433–439.
23] T. Higuchi, S. Uzu, M. Hirobe, J. Am. Chem. Soc. 1990, 112, 7051–
3
and the color of the solution became darker. CDCl sealed in capillary
7
053.
was put into the tube as an external standard, the tube was then capped
and removed from the glove box, and the contents of the tube were char-
[
[
[
[
[
[
[
24] T. Higuchi, K. Shimada, N. Maruyama, M. Hirobe, J. Am. Chem.
Soc. 1993, 115, 7551–7552.
25] H.-A. Wagenknecht, W.-D. Woggon, Angew. Chem. 1997, 109, 404–
1
3
acterized by C NMR spectroscopy. The peak observed at d=À21.7 ppm
for complex 7 had disappeared. The resultant solution was then trans-
ferred to a round-bottomed flask. The NMR tube was rinsed three times
4
07; Angew. Chem. Int. Ed. Engl. 1997, 36, 390–392.
26] M. A. Bennett, A. J. Canty, J. K. Felixberger, L. M. Rendina, C. Sun-
derland, A. C. Willis, Inorg. Chem. 1993, 32, 1951–1958.
27] R. A. Begum, D. Powell, K. Bowman-James, Inorg. Chem. 2006, 45,
with CH
Al column chromatography (2% ethyl acetate in petroleum ether to
give the mixture of organic compounds, followed by 2% CH OH in
CH Cl to give complex 2). The yields of the cross-coupled product
methyl 2-methylbenzoate and complex 2 were determined to be 74%
2 2
Cl . The organic phases were combined and subjected to neutral
2
O
3
3
9
64–966.
2
2
28] M. A. Hossain, S. Lucarini, D. Powell, K. Bowman-James, Inorg.
Chem. 2004, 43, 7275–7277.
29] The yield of 2 was improved by optimization of the workup proce-
dures of the reaction; see the Supporting Information.
30] The deprotonation process was solvent dependent and was much
slower in diethyl ether. The difference was speculated to arise from
the solvent-dependent Schlenk equilibrium of the alkylzinc reagents;
see the Supporting Information for a detailed discussion.
31] H. Wang, J. Liu, Y. Deng, T. Min, G. Yu, X. Wu, Z. Yang, A. Lei,
Chem. Eur. J. 2009, 15, 1499-1507.
32] The model reaction in diethyl ether was much slower than that in
THF and the induction period was much longer, probably because
of the solvent-dependent Schlenk equilibrium of the alkylzinc re-
agents. See reference [30] and the Supporting Information.
33] A. H. M. De Vries, F. J. Parlevliet, L. Schmieder-Van De Vonder-
voort, J. H. M. Mommers, H. J. W. Henderickx, M. A. M. Walet, J. G.
De Vries, Adv. Synth. Catal. 2002, 344, 996–1002.
1
and 60% respectively by H NMR spectroscopy with CH
2
Br
2
as the inter-
nal standard.
[
70]
The cross-coupled product methyl 2-methylbenzoate
obtained after
1
13
further purification was characterized by H NMR and C NMR spec-
1
3
troscopy: H NMR (600 MHz, CDCl ): d=7.90 (d, J=7.8 Hz, 1H), 7.30–
1
3
7
.20 (m, 2H), 3.87 (s, 3H), 2.59 ppm (s, 3H); C NMR (150 MHz,
CDCl
3
): d=168.0, 140.1, 131.9, 131.6, 130.5, 129.4, 125.6, 51.7, 21.7 ppm.
[
[
Acknowledgements
[
This work was supported by the National Natural Science Foundation of
China (grant nos.: 20772093, 20502020, and 20832003), the Excellent
Youth Foundation of the Hubei Scientific Committee, a postdoctoral re-
search fund (grant no.: 203-180549), and a startup fund from Wuhan Uni-
versity.
[34] J. A. Widegren, R. G. Finke, Mol. Catal. A 2003, 198, 317–341.
[35] A. Alimardanov, L. Schmieder-van de Vondervoort, A. H. M. de V-
ries, J. G. de Vries, Adv. Synth. Catal. 2004, 346, 1812–1817.
[
[
36] M. R. Eberhard, Org. Lett. 2004, 6, 2125–2128.
37] K. Yu, W. Sommer, M. Weck, C. W. Jones, J. Catal. 2004, 226, 101–
[
[
1] F. Zeng, E.-i. Negishi, Org. Lett. 2002, 4, 703–706.
2] A. de Meijere, F. Diederich, Metal-catalyzed cross-coupling reac-
tions, 2nd ed., Wiley-VCH, Weinheim, 2004.
110.
[38] D. E. Bergbreiter, P. L. Osburn, J. D. Frels, Adv. Synth. Catal. 2005,
347, 172–184.
4444
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 4437 – 4445

Thioimido–Palladium(II) Complexes in the Negishi Coupling
FULL PAPER
[
39] S. J. Broadwater, D. T. McQuade, J. Org. Chem. 2006, 71, 2131–
134.
40] N. T. S. Phan, M. Van Der Sluys, C. W. Jones, Adv. Synth. Catal.
006, 348, 609–679.
slower than that with 1. The kinetic behavior of this complex is not
discussed here because PPh (1 equiv relative to Pd) in the system
3
would compete for the coordination site on the Pd center during the
whole catalytic process and would make the kinetic analysis more
complicated.
2
[
2
[
[
[
[
41] D. Astruc, Inorg. Chem. 2007, 46, 1884–1894.
42] M. T. Reetz, J. G. de Vries, Chem. Commun. 2004, 1559–1563.
43] Y. Lin, R. G. Finke, Inorg. Chem. 1994, 33, 4891–4910.
44] Hg drop tests did not succeed in either reaction, probably because
the amalgamation was much slower than the reaction rate under the
above mild conditions, which are different from the usual high-tem-
perature conditions for pincer palladium catalytic systems. Also, the
excess mercury in the reaction system might cause decomposition of
the zinc reagent or interact with the ligand as a sulfurphilic metal,
which would destroy the catalyst.
[59] MeZnCl was used for simplicity of the NMR spectra.
[60] H.-B. Kraatz, M. E. van der Boom, Y. Ben-David, D. Milstein, Isr. J.
Chem. 2001, 41, 163–171.
[61] CCDC-721678 (2), 721679 (8), and 721680 (10) contains the supple-
mentary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
[62] A third cycle was also possible, which consists of alkylation from III
II
to IV, an alkyl shift from the Pd species to the S anion to result in
0
0
[
45] G. P. F. Van Strijdonck, M. D. K. Boele, P. C. J. Kamer, J. G. De V-
ries, P. W. N. M. Van Leeuwen, Eur. J. Inorg. Chem. 1999, 1073–
the Pd species, oxidative addition of the aryl iodide to the Pd spe-
II
cies, and an alkyl shift from the S anion to the Pd species to release
1
076.
the product and III. See Scheme 1 in the Supporting Information.
II
IV
[
[
[
46] A. H. M. de Vries, J. M. C. A. Mulders, J. H. M. Mommers, H. J. W.
Henderickx, J. G. de Vries, Org. Lett. 2003, 5, 3285–3288.
[63] The catalytic cycle might be a Pd –Pd catalytic cycle, although
IV
there is not enough data to support the Pd oxidation state.
47] I. J. S. Fairlamb, S. Grant, P. McCormack, J. Whittall, Dalton Trans.
[64] A. E. Jensen, W. Dohle, P. Knochel, Tetrahedron 2000, 56, 4197–
4201.
[65] iC IR, version 1.05, Reaction Analysis Software for Chemical Devel-
opment, AutoChem, Mettler-Toledo (USA).
2
007, 859–865.
48] I. J. S. Fairlamb, R. J. K. Taylor, J. L. Serrano, G. Sanchez, New J.
Chem. 2006, 30, 1695–1704.
[
[
[
49] J. G. De Vries, Dalton Trans. 2006, 421–429.
[66] SMART, version 5.625, Area Detector Software Package and SAX
Area Detector Integration Program, Brucker Analytical, Madison,
WI (USA), 1997.
[67] SAINT, version 6.45, Area Detector Software Package and SAX
Area Detector Integration Program, Brucker Analytical, Madison,
WI (USA), 1997.
50] The vacant site of 6 might be coordinated by solvent THF.
51] Low conversion (<5%) was observed within 90 min at 08C with the
0
pincer Pd complex generated in situ from Pd(trans,trans-dibenzyli-
2
deneacetone) and the pincer ligand.
[
52] C. Amatore, A. Jutand, Acc. Chem. Res. 2000, 33, 314–321.
[
53] L. M. Alcazar-Roman, J. F. Hartwig, J. Am. Chem. Soc. 2001, 123,
[68] SADABS, version 2.03, Area Detector Software Package and SAX
Area Detector Integration Program, Brucker Analytical, Madison,
WI (USA), 1997.
[69] G. M. Sheldrick, SHELXTL, version 6.14, Crystal Structure Analysis
Package, Brucker Analytical, Madison WI (USA), 1997.
[70] F. Macaev, G. Rusu, S. Pogrebnoi, A. Gudima, E. Stingaci, L. Vlad,
N. Shvets, F. Kandemirli, A. Dimoglo, R. Reynolds, Bioorg. Med.
Chem. 2005, 13, 4842–4850.
1
2905–12906.
[
54] E. Poverenov, M. Gandelman, L. J. W. Shimon, H. Rozenberg, Y.
Ben-David, D. Milstein, Chem. Eur. J. 2004, 10, 4673–4684.
55] B. L. Shaw, Chem. Commun. 1998, 1361–1362.
[
[
56] D. Sellmann, C. Allmann, F. Heinemann, F. Knoch, J. Sutter, J. Or-
ganomet. Chem. 1997, 541, 291–305.
[
[
57] D. Sellmann, F. Geipel, F. W. Heinemann, Eur. J. Inorg. Chem. 2000,
2
71–279.
Received: October 28, 2008
Revised: January 14, 2009
Published online: March 13, 2009
58] Complex 8 could catalyze the model reaction too, with a short in-
duction period. The rate of the reaction catalyzed by 8 at 258C was
Chem. Eur. J. 2009, 15, 4437 – 4445
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4445