NCN pincer palladium complexes: Their preparation via a ligand introduction route and their catalytic properties

Article abstract of DOI:10.1021/ja052780n

A wide range of NCN pincer palladium complexes, [4-tert-butyl-2,6-bis(N- alkylimino)phenyl]-chloropalladium (alkyl = n-butyl, benzyl, cyclohexyl, tert-butyl, adamantyl, phenyl, 4-methoxyphenyl), were readily prepared from trans-(4-tert-butyl-2,6-diformylphenyl)chlorobis(triphenylphosphine)palladium via dehydrative introduction of the corresponding alkylimino ligand groups (ligand introduction route) in excellent yields (71-98%). NMR studies on this route for forming pincer complexes revealed the intermediacy of [4-tert-butyl-2,6-bis(N-alkylimino)phenyl]chlorobis(triphenylphosphine) palladium which is in equilibrium with the corresponding NCN pincer complexes via coordination/dissociation of the intramolecular imino groups and triphenylphosphine ligands. A series of chiral NCN pincer complexes bearing pyrroloimidazolone units as the trans-chelating donor groups, [4-tert-butyl-2,6-bis{(3R,7aS)-2-phenylhexahydro-1H-pyrrolo[1,2-c] -imidazol-1-on-3-yl}phenyl]chloropalladium, were also prepared from the same precursor via condensation with proline anilides in high yields. The catalytic properties of the NCN imino and the NCN pyrroloimidazolone pincer palladium complexes were examined in the Heck reaction and the asymmetric Michael reaction to demonstrate their high catalytic activity and high enantioselectivity.

Full text of DOI:10.1021/ja052780n

Published on Web 08/13/2005
NCN Pincer Palladium Complexes: Their Preparation via a
Ligand Introduction Route and Their Catalytic Properties
Kazuhiro Takenaka, Maki Minakawa, and Yasuhiro Uozumi*
Contribution from the Institute for Molecular Science (IMS), The Graduate UniVersity for
AdVanced Studies, and CREST, Higashiyama 5-1, Myodaiji, Okazaki 444-8787, Japan
Received April 29, 2005; E-mail: uo@ims.ac.jp
Abstract: A wide range of NCN pincer palladium complexes, [4-tert-butyl-2,6-bis(N-alkylimino)phenyl]-
chloropalladium (alkyl ) n-butyl, benzyl, cyclohexyl, tert-butyl, adamantyl, phenyl, 4-methoxyphenyl), were
readily prepared from trans-(4-tert-butyl-2,6-diformylphenyl)chlorobis(triphenylphosphine)palladium via
dehydrative introduction of the corresponding alkylimino ligand groups (ligand introduction route) in excellent
yields (71-98%). NMR studies on this route for forming pincer complexes revealed the intermediacy of
[4-tert-butyl-2,6-bis(N-alkylimino)phenyl]chlorobis(triphenylphosphine)palladium which is in equilibrium with
the corresponding NCN pincer complexes via coordination/dissociation of the intramolecular imino groups
and triphenylphosphine ligands. A series of chiral NCN pincer complexes bearing pyrroloimidazolone units
as the trans-chelating donor groups, [4-tert-butyl-2,6-bis{(3R,7aS)-2-phenylhexahydro-1H-pyrrolo[1,2-c]-
imidazol-1-on-3-yl}phenyl]chloropalladium, were also prepared from the same precursor via condensation
with proline anilides in high yields. The catalytic properties of the NCN imino and the NCN pyrroloimidazolone
pincer palladium complexes were examined in the Heck reaction and the asymmetric Michael reaction to
demonstrate their high catalytic activity and high enantioselectivity.
Scheme 1. Synthetic Strategy for Pincer Complexes
Introduction
Organometallic pincer complexes containing terdentate
monoanionic ligands composed of an anionic aryl carbon atom
and two mutually trans-chelating donor sites at the 2,6-positions
of the aromatic ring have been attracting widespread interest in
catalysis and material science.1 Various methods for the
preparation of such compounds have been developed, and they
can be mainly divided into four strategies: (1) direct cyclom-
etalation, (2) oxidative addition, (3) transmetalation, and (4)
transcyclometalation.1a All of these methods must involve a
metalation reaction of the corresponding pincer ligands creating
a new metal-carbon σ bond in the final step (hereafter, we
will refer to these methods as “metal introduction routes,”
Scheme 1a). While the metal introduction route is the most
straightforward process, a synthetic limitation has emerged as
a serious problem. First, it appears difficult for the pincer ligands
having sterically demanding groups as coordination sites to react
with transition metals: Chung and co-workers reported that the
N-benzyl aminal-type ligand was inert to palladation due to its
bulkiness.2 Another problem is that chemically unstable func-
tional groups, such as imines, are not suitable for metalation,
and substantial decomposition of the ligands under reaction
conditions was observed.3a Furthermore, in the direct cyclom-
etalation protocol via C-H activation, the regioselectivity of
Chart 1
metalation is not controlled: in particular, palladation would
prefer the 3,5-positions to the desired 1-position (Chart 1).4
These problems therefore impose a restriction on research for
pincer complexes bearing imine-donating groups; i.e., there are
few articles describing the Rh,5 Pd,6 and Pt3,7 complexes in
contrast to the many reports for those with other coordinating
groups.1a
(1) For reviews on pincer complexes, see: (a) Albrecht, K.; van Koten, G.
Angew. Chem., Int. Ed. 2001, 40, 3750. (b) van der Boom, M. E.; Milstein,
D. Chem. ReV. 2003, 103, 1759. (c) Singleton, J. T. Tetrahedron 2003, 59,
1837. (d) Dupont, J.; Consorti, C. S.; Spencer, J. Chem. ReV. 2005, 105,
2527.
(2) Jung, I. G.; Son, S. U.; Park, K. H.; Chung, K.-C.; Lee, J. W.; Chung, Y.
K. Organometallics, 2003, 22, 4715.
(3) (a) Fossey, J. S.; Richards, C. J. Organometallics, 2002, 21, 5259. (b)
Fossey, J. S.; Richards, C. J. Tetrahedron Lett. 2003, 44, 8773.
(4) (a) Steenwinkel, P.; Gossage, R. A.; Maunula, T.; Grove, D. M.; van Koten,
G. Chem.sEur. J. 1998, 4, 763. (b) Fossey, J. S.; Richards, C. J.
Organometallics, 2004, 23, 367.
9
10.1021/ja052780n CCC: $30.25 © 2005 American Chemical Society
J. AM. CHEM. SOC. 2005, 127, 12273-12281
12273

A R T I C L E S
Scheme 2
Takenaka et al.
To overcome these problems, we planned to develop a new
synthetic strategy based on a metalation-ligand-construction
sequence, referred to as “ligand introduction route” (Scheme
1b). The site-controlled metalation of an aromatic ring before
the introduction of the ligand moieties can overcome the
inhibition of formation of the M-C bond by the bulky
coordinating groups. The subsequent ligand construction at the
2,6-positions makes it easy to use sensitive donor groups.
Here we report the preparation of NCN pincer palladium
complexes having imine or chiral pyrroloimidazolone moieties
via the ligand introduction route. The complexation pathway
of this new protocol was also investigated by variable-
temperature NMR spectroscopy. In addition, the pincer com-
plexes obtained here exhibited high catalytic activity and high
stereoselectivity in the Heck reaction and in the asymmetric
Michael reaction, respectively.
Figure 1. ORTEP drawing of the complex 1 with thermal ellipsoids at
50% probability levels. The solvated CH3CN molecule and the hydrogen
atoms are omitted for clarity. Selected bond distances (Å) and angles
(deg): Pd-C(1) ) 1.999(4), Pd-P(1) ) 2.322(2), Pd-P(2) ) 2.317(2),
Pd-Cl ) 2.378(2); P(1)-Pd-P(2) ) 178.80(4), Cl-Pd-C(1) ) 171.25-
(11).
Scheme 3
Results and Discussion
1. Preparation of N-Alkylimine Pincer Complexes -
Ligand Introduction Route 1. Metalation of a functionalized
aromatic ring prior to construction of coordinating groups is a
prerequisite for our strategy. As the key precursor, the trans-
(4-tert-butyl-2,6-diformylphenyl)chlorobis(triphenylphosphine)-
palladium (1), in which the two formyl groups at the 2,6-
positions of the anionic aromatic ligand can be converted into
imino groups, was designed. The complex 1 was readily
prepared in high yield starting with 4-tert-butyl-2,6-diformylphe-
nyl trifluoromethanesulfonate (3) via oxidative complexation
with Pd0 in the presence of PPh3, as shown in Scheme 2. The
X-ray structure of 1 unambiguously shows that the aromatic
ring is directly connected to the palladium atom, where the two
formyl groups are located at both ortho positions (Figure 1).
With the precursor 1 in hand, we examined whether the ligand
introduction route would indeed work. We were consequently
pleased to find that a pincer palladium complex having imine
functionalities was obtained using this strategy. Thus, the
reaction of 1 with 5 equiv of cyclohexylamine in acetonitrile at
reflux temperature under an O2 atmosphere afforded the complex
2a in 92% yield (Scheme 3). Since it was reported that a similar
complex 4 was isolated in less than 15% yield,6 this ligand
introduction route would provide an effective synthetic method
for the pincer complexes having imine donors. The X-ray
structure of 2a with selected bond lengths and angles is
presented in Figure 2. The PdsC(1) distance of 1.927(3) Å is
significantly shorter than that of the precursor 1 (1.999(4) Å),
which reflects the strong binding nature of the pincer ligand
caused by two additional intramolecular imine coordinations.
Although the ClsPdsC(1) angle is 176.21(7)°, the N(1)sPds
N(2) angle is 156.74(8)°, which indicates the distortion from
the ideal square planar geometry due to the two fused five-
membered rings. In the IR spectrum of 2a, the band at 1601
cm-1 diagnostic of the coordinated CdN stretch6 was observed
instead of the carbonyl stretch at 1672 cm-1 of 1. The iminyl
protons and carbons appeared as singlets at 8.00 and 169.3 ppm
in the 1H and 13C NMR spectra, respectively. The signal
responsible for the ipso carbon connected to palladium was
found at 179.2 ppm, which was shifted toward high frequency
as compared to the starting material 1 (172.4 ppm). The absence
of PPh3 ligands was established by the lack of any resonance
in the 31P NMR spectrum.
Other primary amines were used in this ligand introduction
route. Under the same reaction conditions, 1 reacted with
benzylamine and butylamine to produce the pincer complexes
2b (86% yield) and 2c (98% yield), respectively (Scheme 4).
The X-ray structures of both complexes are depicted in Scheme
4. For each complex, all the spectroscopic data and the elemental
analysis were consistent with the expected structure (see
Supporting Information).
(5) Hoogervorst, W. J.; Goubitz, K.; Fraanje, J.; Lutz, M.; Spek, A. L.; Ernsting,
J. M.; Elsevier, C. J. Organometallics, 2004, 23, 4550.
(6) Vila, J. M.; Gayoso, M.; Pereira, M. T.; Torres, M. L.; Ferna´ndez, J. J.;
Ferna´ndez, A.; Ortigueira, J. M. J. Organomet. Chem. 1996, 506, 165.
(7) (a)Hoogervorst, W. J.; Elsevier, C. J.; Lutz, M.; Spek, A. L. Organometallics
2001, 20, 4437. (b) Hoogervorst, W. J.; Koster, A. L.; Lutz, M.; Spek, A.
L.; Elsevier, C. J. Organometallics 2004, 23, 1161.
9
12274 J. AM. CHEM. SOC. VOL. 127, NO. 35, 2005

NCN Pincer Pd Complexes via Ligand Introduction Route
A R T I C L E S
Scheme 6
Figure 2. ORTEP drawing of one of the two independent crystal structures
of the complex 2a with thermal ellipsoids at 50% probability levels. One
molecule of two independent molecules in a unit cell is presented. The
solvated CH3CN molecule and the hydrogen atoms are omitted for clarity.
Selected bond distances (Å) and angles (deg): Pd-C(1) ) 1.927(3), Pd-
N(1) ) 2.117(2), Pd-N(2) ) 2.087(2), Pd-Cl ) 2.4112(8); N(1)-Pd-
N(2) ) 156.74(8), Cl-Pd-C(1) ) 176.21(7).
Scheme 7. Proposed Pathway
Scheme 4
Complex 2e having more bulky 1-adamantyl groups at both
nitrogen atoms was obtained in 91% yield after 50 h (Scheme
6). The relatively long reaction time could be attributed to the
steric hindrance of 1-adamantamine. The X-ray structure of 2e
elucidates the coordination of the N-(1-adamantyl)imino func-
tionalities to the palladium center (Scheme 6). The structure of
2e was also identified by NMR, IR, FAB-MS, and elemental
analysis.
Scheme 5
From these results, the ligand introduction route should be
recognized as being an alternative synthetic protocol for pincer
palladium complexes since it overcomes the problems of the
known pincer complex formation (introduction of sterically
demanding substituents onto the ligand sites, utilization of
sensitive donor groups, and regioselectivity of metalation).
2. Variable-Temperature NMR Study. In the ligand
introduction route, it was presumed that the pincer complexes
were formed through condensation of the 2,6-formyl groups of
1 with primary amines and subsequent ligand exchange by the
resulting imine functionalities constructing the Pd-N bonds
(Scheme 7). Irreversible trapping of PPh3 ligands liberated from
a possible intermediate 2-P by oxidation under an O2 atmosphere
might be an important step, since van Koten and co-workers
reported quantitative replacement of the amine ligands in the
Using this method, we also built up a pair of N-tert-butylimino
moieties on the aromatic ring of 1 to give the complex 2d in
96% yield as shown in Scheme 5. The conventional synthetic
route via oxidative addition to the corresponding ligand 5, which
was prepared from 3 and tert-butylamine, furnished only up to
48% of 2d even with prolonged reaction time and higher
temperatures (Scheme 5).
9
J. AM. CHEM. SOC. VOL. 127, NO. 35, 2005 12275

A R T I C L E S
Chart 2
Takenaka et al.
for a broad singlet which represents a free PPh3 at -3.0 ppm.
At 60 °C a new signal appears at 22.9 ppm. Upon cooling, the
new signal increases gradually, whereas the PPh3 signal
decreases. At -40 °C the former exceeds the latter in intensity.
Reheating to 80 °C restored the original spectrum, indicating
the reversibility of this behavior. During these experiments a
monophosphine-adduct 2b-P′ was not perceived at all, although
the half-pincer Pd complexes 6 (R ) (CH2)3Si(OEt)3, X ) Br;9a
R ) Ph, X ) OCOCF39b) which possess one PPh3 ligand at the
trans position to the coordinated imine moiety were prepared
and structurally characterized (Chart 2). The steric hindrance
between the phosphine ligand and the noncoordinated imine
moiety seemed to be the reason for the instability. These
observations imply that the proposed equilibrium exists between
2b and the corresponding bisphosphine complex 2b-P (Figure
3). Fortunately, we were able to obtain single crystals of 2b-P
suitable for X-ray diffraction analysis by layering a hexane
solution of an excess amount of PPh3 onto a CHCl3 solution of
2b. The molecular structure of 2b-P given in Figure 4 definitely
shows coordination of two PPh3 molecules to Pd dangling two
N-benzylimino groups on the aromatic ring. NMR spectra of
the single crystals measured at -40 °C indicated that the
structure found in the solid-state was retained in solution (see
Supporting Information). It is of great importance to note that
the solution of 2b-P exhibited the same reversible temperature-
dependent dynamic behavior as shown in Figure 3. Thus, 2b-P
was an observable species at only lower temperatures, while
the pincer complex 2b and the free PPh3 were generated at
higher temperature. These results are compatible with our
hypothesis that 2b and 2b-P are in equilibrium with each
NCN pincer complex by PPh3.8 Therefore, recoordination of
the phosphine ligands to Pd (dissociation-association equilib-
rium of PPh3) would likely take place at the ligand exchange
step.
To confirm our hypothesis, we carried out variable-temper-
ature 1H and 31P NMR measurements for the isolated complexes
2b and 2c in the presence of 2 mol equiv of PPh3. Figure 3a
and 3b show the H and 31P NMR spectra of 2b ranging from
-40 to 80 °C, respectively. In the 1H NMR spectrum at 80 °C,
only a broad peak at 5.01 ppm ascribed to the benzyl protons
of the complex 2b is observed. Upon cooling, the resonance
becomes broader. At 30 °C a new signal appears around 4.50
ppm, which increases gradually as the initial peak of 2b
decreases below that temperature. Finally, at -40 °C the new
peak (at 4.42 ppm) becomes the major peak, accompanied by
a tiny peak of 2b (at 4.96 ppm). This behavior is reversible.
From the two-dimensional 1H EXSY NMR spectrum, it becomes
obvious that these two peaks are interchangeable. In the 31P
NMR spectrum measured at 80 °C, there are no peaks except
1
1
Figure 3. Variable-temperature (a) H and (b) 31P NMR spectra of complex 2b in the presence of 2 mol equiv of PPh3.
9
12276 J. AM. CHEM. SOC. VOL. 127, NO. 35, 2005

NCN Pincer Pd Complexes via Ligand Introduction Route
A R T I C L E S
Scheme 8
Figure 4. ORTEP drawing of complex 2b-P with thermal ellipsoids at
50% probability levels. The carbon atoms of PPh3 ligands except for the
ipso-carbons and the hydrogen atoms are omitted for clarity. Selected bond
distances (Å) and angles (deg): Pd-C(1) ) 2.023(2), Pd-P(1) ) 2.3344-
(7), Pd-P(2) ) 2.3258(7), Pd-Cl ) 2.4007(6); P(1)-Pd-P(2) ) 174.16-
(2), Cl-Pd-C(1) ) 177.15(5).
other: the van’t Hoff plots afforded the thermodynamic
parameters (∆H° ) -90.5 ( 2.9 kJ mol-1, ∆S° ) -231 ( 11
J K-1 mol-1). Complex 2c displayed similar fluxional behavior
(Figure 5): ∆H° ) -63.0 ( 2.1 kJ mol-1 and ∆S° ) -167 (
8.2 J K-1 mol-1. While the large negative entropy apparently
reflects the convergent reaction of the three molecules, the
negative enthalpy value is large enough to compensate for this
disadvantage (Figures 3a and 5a, forward reaction). Hence, this
relationship makes the postulated equilibrium during the reaction
possible.
turned out to be an efficient preparation method for pincer
complexes bearing N-alkylimine coordinating groups, aromatic
amines were studied to develop the utility of this method.
However, when aniline and para-anisidine were used as the
amine, the desired pincer complexes were not produced but
instead the corresponding PPh3 adducts (Scheme 8). Thus, under
the same conditions mentioned above, the reaction of 1 with
aniline and para-anisidine gave trans-[4-tert-butyl-2,6-bis(N-
phenylimino)phenyl]chlorobis(triphenylphosphine)palladium (2f-
P) and trans-[4-tert-butyl-2,6-bis{N-(p-methoxyphenyl)imino}-
phenyl]chlorobis(triphenylphosphine)palladium (2g-P) in 83%
and 91% yield, respectively. Unequivocal confirmation of the
3. Preparation of N-Arylimine Pincer Complexes - Ligand
Introduction Route 2. Because the ligand introduction route
1
Figure 5. Variable-temperature (a) H and (b) 31P NMR spectra of complex 2c in the presence of 2 mol equiv of PPh3.
9
J. AM. CHEM. SOC. VOL. 127, NO. 35, 2005 12277

A R T I C L E S
Scheme 9
Takenaka et al.
Scheme 10
reported that compound 7 did not oxidize the coordinated
phosphines but instead the uncoordinated phosphines,10 the PPh3
ligand was oxidized after dissociation from the metal center at
the ligand exchange stage. No detectable change was observed
during the reaction of the formyl complex 1 with 7, suggesting
that the intramolecular interaction of the imines to the palladium
center assisted the liberation of PPh3. Consequently, the pincer
complexes 2 were obtained by way of the imino phosphine
complexes 2-P in the ligand introduction route, where both the
complexes were in equilibrium and the ortho imino groups
played a key role (Scheme 7).
4. Catalytic Activities of the Imine Pincer Complexes -
the Heck Reaction. A variety of pincer palladium complexes
have been emerging as a new class of efficient catalysts for
several coupling reactions such as the Heck reaction.2,11 To test
the catalytic abilities of the imino pincer palladium complexes
2 to promote the Heck reaction, we initially examined the
reaction of iodobenzene (8a) with methyl acrylate (9a) (Table
1). The reaction proceeded smoothly when a combination of
1-methyl-2-piperidone (NMP) as the solvent and tributylamine
(Bu3N) as the base was used. Thus, the reaction of 8a with 9a
was carried out in NMP in the presence of 1.4 mol equiv of
Bu3N and 1.0 mol % of the pincer complex 2a at 100 °C for 2
h to give methyl trans-cinnamate (10) in 90% yield (entry 1).
The other complexes 2b-f exhibited similar catalytic activities
under otherwise similar conditions (entries 2-6).
The reaction of 8a and butyl acrylate (9b) was found to
proceed very efficiently with only 1 mol ppm of the pincer
catalyst 2c by employing an NMP/H2O mixture (7:3) as a
solvent to give 90% isolated yield of butyl trans-cinnamate 11a
(entry 7). Using the reaction conditions (NMP + H2O/Bu3N/
140 °C) identified above, various aryl halides were examined
for the Heck reaction with acrylates in the presence of 1 mol
ppm of the pincer palladium complex 2c. Thus, the reactions
of ortho-, meta-, and para-iodotoluenes (8b-d) with 1.4 equiv
of 9b were catalyzed by 1.0 mol ppm of the pincer palladium
complex 2c at 140 °C in NMP to give 83%, 88%, and 91%
isolated yields of the Heck products 11b, 11c, and 11d, where
the turnover numbers (TONs) of the palladium catalyst were
proposed connectivity pattern of these complexes was obtained
from X-ray structure analyses (Scheme 8: The carbon atoms
of PPh3 ligands except for the ipso-carbons are omitted for
clarity). Prolonged reaction, however, did not convert these
phosphine complexes into the desired pincer complexes. Pre-
sumably the lowered nucleophilicity of the resulting imines
derived from the aromatic amines rather than those derived from
the aliphatic amines keeps the PPh3 ligands bonded to the Pd
atom. Such a marked difference in reactivity is well correlated
with the order of pKa values of the conjugated acids of the
amines: 9.33-10.77 for aliphatic amines, 4.63 for aniline, and
5.34 for para-anisidine. It is noteworthy that the structures of
2f-P and 2g-P are reminiscent of the intermediate 2-P in the
proposed reaction pathway of the ligand introduction route. As
discussed above, the ligand introduction route seems to include
equilibria between the pincer complexes 2 and the corresponding
PPh3 adducts 2-P (Scheme 7). It follows then that the complexes
2f-P and 2g-P might be equilibrated with the desired pincer
complexes; however gaseous oxygen was too weak to oxidize
the trace amount of PPh3 released from 2f-P and 2g-P. If a
more powerful oxidizing agent had been employed in these
reactions, the pincer complexes might have been obtained by
shifting the equilibrium to the product side.
After screening several oxidizing agents, the urea hydrogen
peroxide addition compound, H2NCONH2‚H2O2 (7), was found
to be a suitable oxidant to produce the pincer complexes bearing
the N-arylimine coordination groups. Thus, condensation of 1
with aniline and subsequent treatment of the resulting reaction
mixture with 7 at 50 °C gave the complex 2f in 71% yield
(Scheme 9). Similarly, the pincer complex 2g was obtained by
using para-anisidine as the primary amine in 80% yield (Scheme
9). The X-ray structure of 2f illustrated in Scheme 9 elucidates
the expected η3-N,C,N terdentate bonding mode of the ligand.
As shown in Scheme 10, the formation of 2f from the isolated
complex 2f-P by treatment with the oxidant 7 indicates that
such phosphine complexes are actual intermediates in the ligand
introduction route. The reversed transformation starting from
the pincer complex is easy to predict.8 NMR experiments,
indeed, revealed that addition of 2 equiv of PPh3 to 2f led to a
quantitative formation of 2f-P (Scheme 10). Since it was
(9) (a) Bedford, R. B.; Cazin, C. S. J.; Hursthouse, M. B.; Light, M. E.; Pike,
K. J.; Wimperis, S. J. Organomet. Chem. 2001, 633, 173. (b) Bedford, R.
B.; Cazin, C. S. J.; Coles, S. J.; Gelbrich, T.; Hursthouse, M. B.; Scordia,
V. J. M. J. Chem. Soc., Dalton. Trans. 2003, 3350.
(10) Dieleman, C. B.; Marsol, C.; Matt, D.; Kyritsakas, N.; Harriman, A.;
Kintzinger, J.-P. J. Chem. Soc., Dalton Trans. 1999, 4139.
(11) For recent studies on the Heck reaction with pincer catalysts, see: (a) Ohff,
M.; Ohff, A.; van der Boom, M. E.; Milstein, D. J. Am. Chem. Soc. 1997,
119, 11687. (b) Bergbreiter, D. E.; Osburn, P. L.; Liu, Y.-S. J. Am. Chem.
Soc. 1999, 121, 9531. (c) Miyazaki, F.; Yamaguchi, K.; Shibasaki, M.
Tetrahedron Lett. 1999, 40, 7379. (d) Morales-Morales, D.; Redo´n, R.;
Yung, C.; Jensen, C. M. Chem. Commun. 2000, 1619. (e) Gruber, A. S.;
Zim, D.; Ebeling, G.; Monteiro, A. L.; Dupont, J. Org. Lett. 2000, 2, 1287.
(f) Morales-Morales, D.; Cramer, R. E.; Jensen, C. M. J. Organomet. Chem.
2002, 654, 44. (g) Sjo¨vall, S.; Wendt, O. F.; Andersson, C. J. Chem. Soc.,
Dalton Trans. 2002, 1396. (h) D´ıez-Barra, E.; Guerra, J.; Hornillos, V.;
Merino, S.; Tejeda, J. Organometallics 2003, 22, 4610. (i) Huang, M.-H.;
Liang, L.-C. Organometallics 2004, 23, 2813. (j) Yao, Q.; Kinney, E. P.;
Zheng, C. Org. Lett. 2004, 6, 2997. (k) Consorti, C. S.; Ebeling, G.; Flores,
F. R.; Rominger, F.; Dupont, J. AdV. Synth. Catal. 2004, 346, 617.
(8) (a) Albrecht, M.; Dani, P.; Lutz, M.; Spek, A. L.; van Koten, G. J. Am.
Chem. Soc. 2000, 122, 11822. (b) Rodriguez, G.; Albrecht, M.; Schoen-
maker, J.; Ford, A.; Lutz, M.; Spek, A. L.; van Koten, G. J. Am. Chem.
Soc. 2002, 124, 5127.
9
12278 J. AM. CHEM. SOC. VOL. 127, NO. 35, 2005

NCN Pincer Pd Complexes via Ligand Introduction Route
A R T I C L E S
Table 1. Heck Reaction of Aryl Iodides and Acrylates Catalyzed by Imino Pincer Complexesa
entry
aryl iodide
olefin
catalyst (mol %)
T (°
C)
time (h)
product
yieldb (%)
1
2
3
4
8a (R1 ) H)
9a
9a
9a
9a
9a
9a
9b
9b
9b
9b
9b
9b
9b
9b
2a (1.0)
2b (1.0)
2c (1.0)
2d (1.0)
2e (1.0)
2f (1.0)
2c (0.0001)
2c (0.0001)
2c (0.0001)
2c (0.0001)
2c (0.0001)
2c (0.0001)
2c (0.0001)
2c (0.0001)
100
100
100
100
100
100
140
140
140
140
140
140
140
140
2
2
2
2
2
10
10
10
10
10
10
11a
11b
11c
11d
11e
11f
11g
11h
90
89
91
86
86
89
90
83
88
91
91
87
80
78
8a
8a
8a
8a
8a
5
6
2
7c
8a
21
21
21
21
48
48
48
48
8c
8b (R1 ) 2-CH3)
8c (R1 ) 3-CH3)
8d (R1 ) 4-CH3)
8e (R1 ) 4-OCH3)
8f (R1 ) 4-Cl)
8 g (R1 ) 4-COCH3)
8h (R1 ) 4-CF3)
9c
10c
11c,d
12c,d
13c,d
14c,d
a All reactions were carried out in the presence of the pincer palladium complex and Bu3N (1.4 equiv). b Isolated yield. c A mixture of NMP and H2O
(7:3) was used as a solvent. d Addition of 1 equiv of tetrabutylammonium bromide (TBAB).
830 × 103, 880 × 103, and 910 × 103, respectively (entries
Scheme 11
8-10). Excellent TONs were also achieved in the reaction of
the aryl iodides having electron-donating as well as electron-
withdrawing substituents by adding 1equiv of tetrabutylammo-
nium bromide (TBAB). The observed TONs in the reactions
of para-iodoanisole (8e), para-chloroiodobenzene (8f), para-
iodoacetophenone (8g), and para-(trifluoromethyl)iodobenzene
(8h) affording 11e, 11f, 11g, and 11h were 910 × 103, 870 ×
103, 800 × 103, and 780 × 103, respectively, demonstrating
the wide substituent tolerance of this reaction system (entries
11-14).
The roles of H2O and TBAB are not yet clear, but it has
been reported that the combination of Pd(OAc)2 and both the
additives can be used as an effective catalyst for the Suzuki-
Miyaura coupling of aryl chlorides in which the true active
catalysts are Pd colloids stabilized by the ammonium salt.12
Moreover, it has been proposed that the Pd0 nanocluster species
generated in situ from PCP13 or SCS14 pincer palladium
complexes is likely to be the true active catalyst for the Heck
reaction. Although the detailed mechanism of the present Heck
reaction with the NCN pincer 2 is still unclear, an induction
period was observed in the preliminary kinetic study of the
reaction of 8a and 9b in good agreement with the profile of the
pincer palladium catalysts.
5. Preparation of Chiral Pincer Complexes - Development
of Ligand Introduction Route. We have recently reported that
hexahydro-1H-pyrrolo[1,2-c]imidazolone serves as an effective
chiral auxiliary.15 These findings prompted us to prepare chiral
pincer complexes having pyrroloimidazolone coordinating
groups because the framework consists of a cyclic amino acid,
a primary amine, and an aldehyde. In fact, the chiral pincer
palladium complexes 12-Cl and 13-Cl were also prepared by
the ligand introduction route (Scheme 11).16 Thus, the arylpal-
ladium complex 1 was treated with 5 equiv of the proline anilide
16 to give 98% yield of [4-tert-butyl-2,6-bis{(3R,7aS)-2-
phenylhexahydro-1H-pyrrolo[1,2-c]imidazol-1-on-3-yl}phenyl]-
chloropalladium (12-Cl). The structure of 12-Cl was un-
equivocally established by X-ray diffraction study (Figure 6).
Similarly, the analogue 13-Cl was obtained in 87% yield by
the reaction of 1 with the anilide 17 derived from trans-4-
hydroxy-L-proline. As shown in Scheme 12, the corresponding
pincer ligands showed little reactivity to palladium due to the
steric bulkiness of the pyrroloimidazolone groups, similar to
the preparation of complex 2d (see above). The chloride ligands
(12) Bedford, R. B.; Blake, M. E.; Butts, C. P.; Holder, D. Chem. Commun.
2003, 466.
(13) Eberhard, M. R. Org. Lett. 2004, 6, 2125.
(14) (a) Yu, K.; Sommer, W.; Weck, M.; Jones, C. W. J. Catal. 2004, 226,
101. (b) Yu, K.; Sommer, W.; Richardson, J. M.; Weck, M.; Jones, C. W.
AdV. Synth. Catal. 2005, 347, 161.
(15) (a) Uozumi, Y.; Shibatomi, K. J. Am. Chem. Soc. 2001, 123, 2919. (b)
Uozumi, Y.; Yasoshima, K.; Miyachi, T.; Nagai, S. Tetrahedron Lett. 2001,
42, 411. (c) Shibatomi, K.; Uozumi, Y. Tetrahedron: Asymmetry 2002,
13, 1769. (d) Uozumi, Y.; Tanaka, H.; Shibatomi, K. Org. Lett. 2004, 6,
281.
(16) Preliminary results have been published, see: (a) Takenaka, K.; Uozumi,
Y Org. Lett. 2004, 6, 1833. (b) Takenaka, K.; Uozumi, Y AdV. Synth. Catal.
2004, 346, 1693.
9
J. AM. CHEM. SOC. VOL. 127, NO. 35, 2005 12279

A R T I C L E S
Takenaka et al.
Table 2. Asymmetric Michael Addition of R-Cyanoesters to Vinyl
Ketones Using Chiral Pincer Complexesa
entry
substrates
cat
time (h)
product
yieldb (%)
% eec
1
2
3
4
5
6
7d
8
18/19a
18/19a
18/19a
18/19a
18/19a
18/19b
18/19c
21/19b
12-OTf
13-OTf
14-OTf
15-OTf
12-Cl
13-OTf
13-OTf
13-OTf
3
4
3
4
144
4
20a
20a
20a
20a
20a
20b
20c
22
95
89
93
97
<2
90
93
91
8
81
6
9
Figure 6. ORTEP drawing of complex 12-Cl with thermal ellipsoids at
50% probability levels. The solvated toluene molecule and the hydrogen
atoms are omitted for clarity. Selected bond distances (Å) and angles
(deg): Pd-C(1) ) 1.912(3), Pd-N(1) ) 2.110(3), Pd-N(3) ) 2.111(3),
Pd-Cl ) 2.437(1); N(1)-Pd-N(3) ) 162.96(11), Cl-Pd-C(1) ) 176.37-
(10).
80
80
83
24
4
a All reactions were carried out in the presence of 0.5 mol % of the
pincer palladium complexes and 0.1 equiv of i-Pr2EtN at 25 °C in benzene
or toluene unless otherwise noted. b Isolated yield. c Determined by GC
analysis (Cyclodex CB). d 1.0 mol % of 13-OTf was used.
Scheme 12
The pincer complexes 12-OTf, 14-OTf, and 15-OTf,
which lacked hydroxyl groups, were much less stereoselective
and only gave 8% ee, 6% ee, and 9% ee of 20a, respectively
(entries 1, 3, and 4). It is also interesting to note that the chemical
yield of the product 20a is strongly affected by the anionic ligand
of the pincer complexes. Thus, the Michael addition did not
take place with the complex 13-Cl even after a reaction time
of 144 h (entry 5), whereas, with 13-OTf, the reaction gave a
high yield of the product in 4 h. Isopropyl ester 19b and
diisopropylmethyl ester 19c were subjected to the Michael
addition under similar conditions to give 80% ee (S) of both
20b and 20c in 90% and 93% yields, respectively (entries 6
and 7). The highest stereoselectivity was obtained when the
reaction was carried out with ethyl vinyl ketone (21) and 19b
in the presence of the chiral pincer catalyst 13-OTf to give
91% yield of the heptanoate (S)-22 with 83% enantiomeric
purity (entry 8).
The structure of the pincer complex 13-OTf which has
hydroxyl groups on their pyrrole rings could not be determined
by X-ray analysis because of the difficulty in obtaining adequate
single crystals suitable for X-ray diffraction. Molecular modeling
study of 13-OTf indicates that the hydroxyl substituents on
the pyrrole rings would play an essential role in the asymmetric
induction. Thus, as can be seen from the schematic structure of
13-OTf shown in Figure 7a, the hydroxyl groups on the pyrrole
rings are situated in close proximity to the metal species in the
regions of the second and fourth quadrants (from the viewpoint
of metal side) where they would fix the enolate of the cyanoester
and the vinyl ketone via hydrogen bonds to induce high
enantioselectivity (Figure 7b).17-19
of 12-Cl and 13-Cl were replaced with the more labile triflate
ligand by treatment with silver triflate to give 12-OTf and 13-
OTf in 95% and 94% yields, respectively. The pincer palladium
complexes 14-OTf and 15-OTf having methoxy and silyloxy
groups on their pyrrole rings were also prepared from 13-Cl
in 76% and 83% yields, respectively, via etherification followed
by treatment with silver triflate.
To explore the enantiocontrolling potential of the chiral pincer
palladium complexes, we elected to study the catalytic asym-
metric Michael reaction of vinyl ketones and R-cyanocarboxy-
lates as nucleophiles17 which has attracted increasing attention
since the products bear a quaternary carbon center with various
functionalities.18 The reaction of methyl vinyl ketone (18) with
methyl 2-cyanopropionate (19a) was performed in benzene at
25 °C in the presence of 0.5 mol % of the chiral pincer palladium
complex and 0.1 equiv of diisopropylethylamine to give the
desired Michael adduct 20a. Representative results are sum-
marized in Table 2. Among the chiral pincer catalysts 12-15,
13-OTf bearing hydroxyl groups on the pyrrole rings turned
out to be the best catalyst, giving 20a with high enantioselec-
tivity. Thus, the asymmetric Michael addition catalyzed by 13-
OTf afforded 81% ee (S) of the adduct ethyl 2-cyano-2-methyl-
5-oxohexanoate (20a) in 89% yield (entry 2).
Conclusion
(17) Naota, T.; Taki, H.; Mizuno, M.; Murahashi, S.-I. J. Am. Chem. Soc. 1989,
111, 5954.
The ligand introduction route, in which a metal atom is
introduced onto the aromatic ring prior to the construction of
(18) For asymmetric Michael addition of R-cyanoesters with chiral transition
metal catalysts, see: Rh complexes: (a) Sawamura, M.; Hamashima, H.;
Ito, Y. J. Am. Chem. Soc. 1992, 114, 8295. (b) Sawamura, M.; Hamashima,
H.; Ito, Y. Tetrahedron 1994, 50, 4439. (c) Inagaki, K.; Nozaki, K.; Takaya,
H. Synlett 1997, 119. (d) Motoyama, Y.; Koga, Y.; Kobayashi, K.; Aoki,
K.; Nishiyama, H. Chem.sEur. J. 2002, 8, 2968. Pt complexes: (e) Blacker,
A. J.; Clarke, M. L.; Loft, M. S.; Mahon, M. F.; Williams, J. M. J.
Organometallics 1999, 18, 2867. Pd complexes: (f) Stark, M. A.; Jones,
G.; Richards, C. J. Organometallics 2000, 19, 1282.
(19) Mechanistic studies on the catalytic Michael addition, see: (a) Murahashi,
S.-I.; Naota, T.; Taki, H.; Mizuno, M.; Takaya, H.; Komiya, S.; Mizuho,
Y.; Oyasato, N.; Hiraoka, M.; Hirano, M.; Fukuoka, A. J. Am. Chem. Soc.
1995, 117, 12436. (b) Naota, T.; Tannna, A.; Murahashi, S.-I. Chem.
Commun. 2001, 63 and references therein.
9
12280 J. AM. CHEM. SOC. VOL. 127, NO. 35, 2005

NCN Pincer Pd Complexes via Ligand Introduction Route
A R T I C L E S
removed under reduced pressure, and the crude product was purified
by column chromatography on silica gel, yielding the imino pincer
palladium complex as a yellow solid.
General Procedure for the Preparation of Imino Pincer Pal-
ladium Complexes (Ligand Introduction Route 2). The palladium
complex 1 and the primary amine were suspended in MeCN. The
suspension was refluxed under a N2 atmosphere to give a yellow
solution, which was cooled to room temperature. The urea hydrogen
peroxide addition compound (7) was added, and the reaction mixture
was stirred at 50 °C. The solvent was removed under reduced pressure,
and the resulting mixture was purified by column chromatography on
silica gel to afford the imino pincer palladium complex as a yellow
solid.
Equilibria Studies by NMR. The complex (2b; 10.5 mg, 20 µmol
or 2c; 9.0 mg, 20 µmol) and PPh3 (10.6 mg, 40 µmol) were dissolved
in 0.60 mL of tetrachloroethane-d2. The 1H and 31P{1H} NMR spectra
were measured from -40 to 80 °C after reaching an equilibrium. At
each temperature (for 2b, -40 to 30 °C; for 2c, -40 to 20 °C), the
equilibrium constant, Keq, was determined by the integration of the
resonances assigned to the methylene protons attached to the N atom.
Thermodynamic parameters were obtained from a plot of ln Keq vs
1/T, where the slope ) -∆H°/R and the intercept ) ∆S°/R, using
Microsoft Excel.
Figure 7. (a) Schematic structure of the complex 13-OTf and (b) the
plausible transition state of the Michael addition.
the ligand moieties, has been developed as a new synthetic
methodology for pincer complexes. This concept has been
applied to the preparation of NCN pincer palladium complexes
bearing imine functionalities as coordinating sites. The three
major problems found in conventional pincer formation, namely,
introduction of sterically demanding donors, utilization of
sensitive coordination groups, and regioselectivity of metalation,
are no longer impediments using this strategy. The VT NMR
experiments established the existence of an equilibrium between
the pincer complexes and the corresponding PPh3 adducts which
are possible intermediates in the course of the reaction. It was
also shown that the desired pincer complexes could be obtained
by oxidative removal of the liberated PPh3, thereby displacing
the equilibrium. The ligand introduction route was also found
to be an efficient method of preparation for the chiral pincer
palladium complexes having pyrroloimidazolone coordination
groups.
The imine pincer complexes obtained here displayed high
catalytic activity in the Heck reaction of aryl iodides and butyl
acrylate. Furthermore, the asymmetric Michael addition of
R-cyanocarboxylates to vinyl ketones was catalyzed by the chiral
pyrroloimidazolone pincer complexes with high enantioselec-
tivity.
The application of the ligand introduction route to other
transition metals as well as various functional groups is currently
underway and will be reported in due course.
General Procedure for the Heck Reaction. To a solution of Bu3N
(1.4 equiv) in a NMP/H2O mixture (0.7 mL/0.3 mL) were added the
aryl halide (1.0 equiv), butyl acrylate (1.4 equiv), and the catalyst 2
(as a 0.1 mM solution in NMP). The reaction mixture was heated at
140 °C for a specified period of time and allowed to cool to room
temperature. The reaction mixture was then diluted with water, and
the product was extracted three times with ether. The combined extracts
were dried over Na2SO4. The organic phase was concentrated under
reduced pressure, and the crude product was purified by silica gel
column chromatography (eluent: hexane/EtOAc ) 100/1), giving the
desired product. CAS Registry numbers of the Heck products: 10, 1754-
62-7; 11a, 52392-64-0; 11b, 163977-61-5; 11c, 173593-27-6; 11d,
123248-21-5; 11e, 121725-19-7; 11f, 123248-22-6; 11g, 173464-57-
8; 11h, 220466-27-3.
General Procedure for the Michael Reaction. To a solution of
the catalyst (0.005 equiv) in either toluene or benzene (2.0 mL) were
added the cyano ester (1.0 equiv), the Michael acceptor (1.5 equiv),
and finally Hu¨nig’s base (0.1 equiv). The reaction mixture was stirred
at 25 °C for an appropriate time. The solvent was removed, and the
residue was purified by Kugelrohr distillation to give the desired
product.
Acknowledgment. This work was supported by the CREST
program sponsored by JST. We also thank the JSPS (Creative
Scientific Research, No. 13GS0024; Grant-in-Aid for Scientific
Research, No. 15205015) and the MEXT (Scientific Research
on Priority Areas, Nos. 412 and 420) for partial financial support
of this work.
Experimental Section
General Procedure for the Preparation of Imino Pincer Pal-
ladium Complexes (Ligand Introduction Route 1).20 The palladium
complex 1 and the primary amine were suspended in MeCN. The
suspension was refluxed under an O2 atmosphere to give a clear yellow
solution, which was then cooled to room temperature. The solvent was
Supporting Information Available: Full experimental details,
spectral and analytical data for the products, and X-ray
crystallographic data for 1, 2a, 2b, 2c, 2e, 2f, 2b-P, 2f-P, 2g-
P, and 12-Cl in CIF format. This material is available free of
charge via the Internet at http://pubs.acs.org.
(20) General experimental conditions, detailed procedures, and characterization
of the products are given in the Supporting Information.
JA052780N
9
J. AM. CHEM. SOC. VOL. 127, NO. 35, 2005 12281