Bimetallic η6,η1 NCN-pincer ruthenium palladium complexes with η6-rucp coordination: synthesis, X-ray structures, and catalytic properties

Article abstract of DOI:10.1021/om801200c

The synthesis, structure, and catalytic activity of a series of bimetallic η⁶,η¹-NCN-pincer ruthenium palladium complexes, [3]⁺-[5]⁺, have been studied, in which η⁶-coordination of [Ru(C₅R<

Full text of DOI:10.1021/om801200c

Organometallics 2009, 28, 2325–2333
2325
Bimetallic η6,η1 NCN-Pincer Ruthenium Palladium Complexes with
η6-RuCp Coordination: Synthesis, X-ray Structures, and Catalytic
Properties
Sylvestre Bonnet,† Joop H. van Lenthe,‡ Maxime A. Siegler,§ Anthony L. Spek,§
Gerard van Koten,† and Robertus J. M. Klein Gebbink*,†
Chemical Biology & Organic Chemistry, Debye Institute for Nanomaterials Science, Faculty of Science,
Utrecht UniVersity, Padualaan 8, 3584 CH Utrecht, The Netherlands, Theoretical Chemistry Group, Debye
Institute for Nanomaterials Science, Faculty of Science, Utrecht UniVersity,
Padualaan 8, 3584 CH Utrecht, The Netherlands, and Crystal and Structural Chemistry,
BijVoet Center for Biomolecular Research, Faculty of Science, Utrecht UniVersity,
Padualaan 8, 3584 CH Utrecht, The Netherlands
ReceiVed December 19, 2008
The synthesis, structure, and catalytic activity of a series of bimetallic η6,η1-NCN-pincer ruthenium
palladium complexes, [3]+-[5]+, have been studied, in which η6-coordination of [Ru(C5R5)]+ (R ) H or
Me) is realized either directly to the central arene ring of the NCN-pincer palladium complex ([3]+ and
[4]+) or to the para-phenyl substituent ([5]+). The X-ray structures of [4]+ and [5]+, as well as the
regioselectivity observed in the synthesis of [5]+, provide clear evidence of intramolecular steric interactions
in [4]+ between the amine arms of the pincer moiety and the methyl groups of the cyclopentadienyl
ligand. Cyclovoltammetry data point to strong electronic interactions between the two metal fragments
in [3]+ and [4]+ and only limited interactions in [5]+. Catalytic studies in the cross-coupling reaction
between trans-phenylvinylboronic acid and vinyl epoxide show that the catalytic activity of the palladium
center is enhanced by η6-coordination of the ruthenium fragment. These modifications of the catalytic
activity of palladium are not correlated to the decreased electron density on palladium, as confirmed by
DFT calculations. We hence propose a mechanism in which transmetalation is the rate-determining step.
pincer ligand offers some special features.3,11-13 First, the
terdentate framework of the NCN-pincer ligand stabilizes the
carbon-to-palladium bond. In addition, by occupying three out
of the four coordination sites on the metal center, the number
of catalytic intermediates is restricted, which can improve
selectivity.14 These peculiar properties of the NCN-pincer
framework have opened the door to new, selective organic
reactions of allyl halides,15 stannanes,16 and boronic acids.17-19
Introduction
Palladium complexes have been extensively used as catalysts
in a wide range of carbon-carbon bond-forming reactions,1,2
not only in homogeneous media,3,4 but also on a solid
support.5-10 Among the many different ligands that have been
studied for these reactions in combination with Pd, the NCN-
Up to now, two different approaches have been developed
to tune the catalytic activity and selectivity of NCN-pincer
palladium complexes: (1) substitution by electron-donating or
electron-withdrawing groups in para-position to the carbon-to-
palladium bond, which modifies the electron density on
palladium,13,20,21 and (2) changing the size and geometry of the
organic substituents borne by the nitrogen atoms, notably to
induce stereoselectivity.22-25
Dedicated to the late Professor Yoshihiko Ito (deceased Dec 23rd,
2006), in commemoration of a long friendship and fine collaboration. We
would like to honor his outstanding contribution in the fields of synthetic
chemistry and organometallic chemistry.
* Corresponding author. E-mail: r.j.m.kleingebbink@uu.nl. Fax: +31 30
2523615.
† Chemical Biology & Organic Chemistry, Debye Institute for Nano-
materials Science.
‡ Theoretical Chemistry Group, Debye Institute for Nanomaterials
Science.
§ Bijvoet Center for Biomolecular Research.
(1) Tsuji, J. Palladium Reagents and Catalysts: New PerspectiVes for
the 21st Century; Wiley: Weinheim, Germany, 2004.
(2) Tsuji, J. Palladium in Organic Synthesis; Springer: Berlin, Germany,
2005.
(3) Singleton, J. T. Tetrahedron 2003, 59, 1837.
(4) Albrecht, M.; van Koten, G. Angew. Chem., Int. Ed. 2001, 40, 3750.
(5) Song, C. E.; Lee, S. G. Chem. ReV. 2002, 102, 3495.
(6) Leadbeater, N. E.; Marco, M. Chem. ReV. 2002, 102, 3217.
(7) Choplin, A; Quignard, F. Coord. Chem. ReV. 1998, 178-180, 1679.
(8) Brunel, D.; Bellocq, N.; Sutra, P.; Cauvel, A.; Lasperas, M.; Moreau,
P.; Di Renzo, F.; Galarneau, A.; Fajula, F. Coord. Chem. ReV. 1998, 180,
1085.
(9) Mehendale, N. C.; Sietsma, J. R. A.; de Jong, K. P.; van Walree,
C. A.; Klein Gebbink, R. J. M.; van Koten, G. AdV. Synth. Catal. 2007,
349, 2619.
(10) Mehendale, N. C.; Bezemer, C.; van Walree, C. A.; Klein Gebbink,
R. J. M.; van Koten, G. J. Mol. Catal. A: Chem. 2006, 257, 167.
(11) Morales-Morales, D. The Chemistry of Pincer Compounds; Elsevier
B. V.: Amsterdam, The Netherlands, 2007.
(12) Szabo´, K. Synlett 2006, 6, 811.
(13) Slagt, M. Q.; Rodriguez, G.; Grutters, M. M. P.; Klein Gebbink,
R. J. M.; Klopper, W.; Jenneskens, L. W.; Lutz, M.; Spek, A. L.; van Koten,
G. Chem.-Eur. J 2004, 10, 1331.
(14) Solin, N.; Kjellgren, J.; Szabo´, K. J. Angew. Chem., Int. Ed. 2003,
42, 3656.
(15) Solin, N.; Kjellgren, J.; Szabo´, K. J. J. Am. Chem. Soc. 2004, 126,
7026.
(16) Wallner, O. A.; Szabo´, K. J. Org. Lett. 2004, 6, 1829.
(17) Olsson, V. J.; Sebelius, S.; Selander, N.; Szabo´, K. J. J. Am. Chem.
Soc. 2006, 128, 4588.
(18) Solin, N.; Wallner, O. A.; Szabo´, K. J. Org. Lett. 2005, 7, 698.
(19) Kjellgren, J.; Aydin, J.; Wallner, O. A.; Saltanova, I. V.; Szabo´,
K. J. Chem.-Eur. J. 2005, 11, 5260.
10.1021/om801200c CCC: $40.75
2009 American Chemical Society
Publication on Web 03/11/2009

2326 Organometallics, Vol. 28, No. 7, 2009
Bonnet et al.
Scheme 2. Synthesis of Complex [5](BF4)
Scheme 1. NCN-Pincer Palladium and Bimetallic Palladium
Ruthenium Complexes Used in This Study
Results
Synthesis and X-ray Structure of [5](BF4). The synthetic
route toward complex [5](BF4) is shown in Scheme 2. The Ph-
NCN-pincer arene ligand 7 was synthesized by a Suzuki cross-
coupling reaction of I-NCN-pincer ligand 6 with phenylboronic
acid.27 Then, lithiation of 7 at -78 °C in pentane with n-BuLi,
followed by reaction with PdCl2(SEt2)2 in diethyl ether, afforded
the para-phenyl-substituted NCN-pincer palladium complex 8.20
Subsequent reaction of 8 with [Ru(C5Me5)]+ using [Ru(C5Me5)-
(MeCN)3](BF4) in dichloromethane at room temperature af-
forded [5](BF4) in 50% yield (based on 7). 1H NMR spectra of
the product clearly showed that the reaction is regioselective:
η6-coordination at the η1-palladated arene ring had not taken
place (see Scheme 2), but instead selective η6-coordination to
the para-phenyl group had occurred. For complex [5]+ indeed,
the protons of the η6-coordinated arene ring appear at low field
(three signals between 6.1 and 6.6 ppm in acetone-d6), whereas
the singlet of the meta protons on the η1-coordinated ring
remains around 7.3 ppm (in acetone-d6).28 Unlike in [3]+ and
[4]+, where facial differentiation of the pincer moiety induced
by the ruthenium fragment leads to two different (diastereotopic)
methyl groups and two different (diastereotopic) methylene
protons,26 in [5]+ there is only one methyl group and one
methylene proton for the NCN-pincer arms, which shows that
rotation around the single, central C4-C1′ bond of the η6,η1-
coordinated biphenyl ligand is fast at room temperature, i.e.,
there is no facial differentiation of the pincer moiety induced
by the η6-coordinated ruthenium fragment in [5]+. Much like
[3](PF6) and [4](BF4), complex [5](BF4) has good solubilities
in organic solvents such as acetone and dichloromethane, and
gives a white solid upon precipitation from pentane.
Recently,26 we explored a third approach for influencing the
reactivity of the metal center of NCN-pincer metal compounds.
We showed that η6-coordination of the organometallic fragment
[Ru(C5R5)]+ (R ) H or Me) to the central arene ring of a
platinated or, notably, a palladated NCN-pincer ligand can be
realized in a direct electrophilic reaction affording heterobime-
tallic products, e.g., [3]+ and [4]+ in Scheme 1, in high yield.
In these η6,η1-heterobimetallic architectures the σ- and π-elec-
trons of one unique phenyl anion arrange the metal-to-carbon
interactions in an orthogonal fashion. As shown in our initial
communication, the redistribution of electrons in these systems
is large, which results in an electron-poor NCN-pincer metal
center. However, the difference in charge between monometallic
complex 1 and cationic complexes of type [3]+ and [4]+ hampers
a direct comparison of their electrochemical and/or chemical
properties.
For that reason, η6,η1-NCN-pincer bimetallic complexes of
type [5]+ (Scheme 1) were considered, in which η6-coordination
occurs at the para-phenyl substituent, and not at the arene ring
of the pincer moiety. In this paper, we describe the synthesis
of complex [5]+, as well as the application of the palladium-
containing complexes [3]+-[5]+ in the catalytic cross-coupling
reaction between trans-phenylvinylboronic acid and vinyl
epoxide.19 It appeared that η6-coordination made the resulting
pincer ruthenium palladium compounds faster catalysts. Elec-
trochemical measurements and DFT calculations were carried
out to substantiate the electron richness of palladium and link
the catalytic performance of these unique mixed Ru-Pd
compounds to the most plausible mechanism.
Colorless crystals of [5](BF4) suitable for X-ray analysis were
obtained by slow vapor diffusion of hexane into a dichlo-
romethane solution of [5](BF4) (see Experimental Part). In the
solid state, [5](BF4) was found to be isomorphous to its platinum
analogue.29
(20) Dijkstra, H. P.; Slagt, M. Q.; McDonald, A.; Kruithof, C. A.; Kreiter,
R.; Mills, A. M.; Lutz, M.; Spek, A. L.; Klopper, W.; van Klink, G. P. M.;
van Koten, G. Eur. J. Inorg. Chem. 2003, 830.
(21) Koecher, S.; Walfort, B.; Mills, A. M.; Spek, A. L.; van Klink,
G. P. M.; van Koten, G.; Lang, H. J. Organomet. Chem. 2008, 693, 1991.
(22) Gosiewska, S.; Herreras, S. M.; Lutz, M.; Spek, A. L.; Havenith,
R. W. A.; van Klink, G. P. M.; van Koten, G.; Klein Gebbink, R. J. M.
Organometallics 2008, 27, 2549.
(23) Gosiewska, S.; Martinez, S. H.; Lutz, M.; Spek, A. L.; van Koten,
G.; Klein Gebbink, R. J. M. Eur. J. Inorg. Chem. 2006, 4600.
(24) Gosiewska, S.; Veld, M. F. I.; de Pater, J. J. M.; Bruijnincx, P. C. A.;
Lutz, M.; Spek, A. L.; van Koten, G.; Klein Gebbink, R. J. M. Tetrahedron:
Asymmetry 2006, 17, 674.
(25) Takenaka, K.; Minakawa, M.; Uozumi, Y. J. Am. Chem. Soc. 2005,
127, 12273.
There are two crystallographically independent molecules in
the asymmetric unit and, after least-squares refinements, the
atomic displacement ellipsoids of the two independent molecular
cations were satisfactory (see Figure 1). Table 1 includes
selected distances and angles for the molecular structure of
(27) Rodriguez, G.; Albrecht, M.; Schoenmaker, J.; Ford, A.; Lutz, M.;
Spek, A. L.; van Koten, G. J. Am. Chem. Soc. 2002, 124, 5127.
(28) In Ph-NCNPdCl (8), all aromatic protons appear at lower fields, i.
e., between 7.0 and 8.0 ppm (in CD2Cl2).
(26) Bonnet, S.; Lutz, M.; Spek, A. L.; van Koten, G.; Klein Gebbink,
R. J. M. Organometallics 2008, 27, 159.
(29) Bonnet, S.; van Lenthe, J. H.; Siegler, M. A.; Lutz, M.; Spek, A.
L.; van Koten, G.; Klein Gebbink, R. J. M., submitted.

η6,η1 NCN-Pincer Ruthenium Palladium Complexes
Organometallics, Vol. 28, No. 7, 2009 2327
Catalytic Properties. The catalytic performances of [3]+,
[4]+, and [5]+ in the cross-coupling reaction of trans-phenylvi-
nylboronic acid and vinyl epoxide were tested (see Scheme 3)19
and compared with those of the neutral complexes 1, 2, and 8
(see Table 3 and Figure 2). All NCN-pincer palladium com-
plexes are active homogeneous catalysts and lead to good yields
at full conversion. In order to monitor the reaction in time, the
catalyst loading was lowered to 1 and 0.25 mol %, respectively
(cf. the use of 2.5 mol % in the original publication),19 while
the amount of water was substantially increased (+25%).
Figures 2 and 3 depict the reaction profiles analyzed by gas
chromatography, i.e., the total yields of the reaction products
(linear + branched) versus time and nature of the catalyst, and
the linear/branched ratio versus time and nature of the catalyst,
respectively. When 1 mol % catalyst was used, the best catalysts
(2 and [5]+) are able to lead the reaction to completion within
2 h (which regrettably was too fast to obtain satisfactory kinetic
data using the available ChemSpeed apparatus); with 0.25 mol
% catalyst the reaction required 6-24 h to go to completion,
leading to satisfactory kinetic data (Figure 2). In all cases, an
initial induction period was observed, which resulted for the
most active catalysts in a time profile that has a sigmoidal shape.
Table 3 gives GC yields after 2 h using 1 mol % catalyst and
the time (t50) necessary to reach 50% conversion using 0.25
mol % catalyst. Interestingly, for all catalysts the linear/branched
ratio gradually increased during the course of the reaction until
it had reached the value observed for [PdBr(H-NCN)] (l/b )
2.33, see Figure 3).19
Figure 1. Displacement ellipsoid plot (drawn at 50% probability
level) of one of the two crystallographically independent cations
-
in the asymmetric unit of [5](BF4) at 150 K. BF4 counteranions
and H-atoms have been omitted for clarity.
cation [5]+, as well as those for the neutral complex 2,30 the
related bimetallic cationic complex [4]+,26 and the ruthenium
complex [Ru(C5Me5)(η6-1,3-(Me2NCH2)2C6H4)](BF4), denoted
[9]+.31
The molecular structure of [5]+ clearly shows that the
palladium center is η1-coordinated to Cipso of the NCN-pincer
ligand, whereas the ruthenium center is η6-coordinated to the
para-phenyl substituent. Unlike in [4]+, there is no indication
for steric constraints in [5]+ between the C5Me5 ligand and the
N-methyl groups. The Ru-C distances are very similar, and
the planes defined by the η6-arene ring and the Cp* ligand are
nearly parallel. The approximate C2 symmetry of the pincer
fragment observed for this family of compounds is also present
in [5]+.
DFT Calculations. In order to investigate whether the
observed trends in catalysis can be explained in terms of
electronic effects, we performed DFT calculations based on the
B3LYP/LANL2DZ method as implemented in the GAMESS-
UK program.35 In a recent publication, the mechanism proposed
for the SeCSe-pincer palladium-catalyzed reaction comprised
two steps: (i) transmetalation between the boronic acid and the
palladium complex, yielding a trans-phenylvinyl SeCSe-pincer
palladium intermediate, and (ii) the nucleophilic attack of this
transmetalated intermediate to vinyl epoxide in an SN2-type
mechanism.19
Electrochemical Properties. In order to get information
about the electron density on the Pd center in the Pd-Ru
complexes, cyclic voltammograms of the neutral palladium (1,
2, 8) and cationic palladium-ruthenium ([3]+, [4]+, and [5]+)
complexes were recorded in acetonitrile (see Table 2). For all
NCN-pincer palladium complexes, an irreversible oxidation
wave was observed, which corresponds to the PdII/PdIV oxida-
tion.32 For 2, an additional reversible reduction wave was
observed at E1/2 ) -1.64 V vs Fc/Fc+, which corresponds to
We minimized by DFT the five intermediates [ISn] (n ) 1-5)
obtained by transmetalation of trans-phenylvinylboronic acid
with the NCN-pincer palladium catalysts 1, 2, [3]+, [4]+, and
[5]+, respectively (see Scheme 4). In [ISn], the chloride anion
coordinated to palladium has been replaced by a trans-
phenylvinyl organic ligand. In order to be able to compare
charged and neutral species, all cationic complexes were
minimized in the presence of an additional chloride counter-
anion. In all five minimized geometries, the plane of the PhC2H2
ligand coordinated in a η1-fashion to Pd was found perpendicular
to the plane of the arene ring, making an angle of ∼80° with
the coordination plane of Pd (see Figure 4). Table 4 gives for
33
•-
the formation of the para-NO2 radical anion. Reduction of
the ruthenium center in [Ru(η6-arene)(η5-C5R5)]+ complexes is
known to occur at very low potentials34 and could not be
observed for the bimetallic cations [3]+-[5]+. Overall, our
results show that η6-coordination of the ruthenium fragment
directly to the arene ring of the pincer framework as in [3]+
and [4]+ induces a very large shift of the PdII/PdIV oxidation
potential (∆Eox ) +240 mV compared to complex 1). This shift
is even larger than that induced by a para-NO2 substituent (∆Eox
) +120 mV) in the parent pincer palladium complex 2. When
η6-coordination takes place to a para-phenyl substituent as in
[5]+, the shift of the oxidation potential is much weaker (∆Eox
) 20 mV compared to 8).
36
[ISn] (n ) 1-5) the natural charges (Q) on Pd, Cipso, and Cvinyl
calculated according to natural bond orbital (NBO) analysis,37
the bond index (BI)38 of the Pd-Cvinyl bond, and the energy
level Eσ of the highest filled molecular orbital of σ character
(30) Slagt, M. Q.; Dijkstra, H. P.; McDonald, A.; Klein Gebbink,
R. J. M.; Lutz, M.; Ellis, D. D.; Mills, A. M.; Spek, A. L.; van Koten, G.
Organometallics 2003, 22, 27.
(31) Siegler, M. A.; Bonnet, S.; Schreurs, A. M. M.; Klein Gebbink,
R. J. M.; van Koten, G.; Spek, A. L, submitted.
(32) Gagliardo, M.; Rodriguez, G.; Dam, H. H.; Lutz, M.; Spek, A. L.;
Havenith, R. W. A.; Coppo, P.; De Cola, L.; Hartl, F.; van Klink, G. P. M.;
van Koten, G. Inorg. Chem. 2006, 45, 2143.
(33) Chauhan, B. G.; Fawcett, W. R.; Lasia, A. J. Phys. Chem. 1977,
81, 1476.
(34) Gusev, O. V.; Ievlev, M. A.; Peterleitner, M. G.; Peregudova, S. M.;
Denisovich, L. I.; Petrovskii, P. V.; Ustynyuk, N. A. Russ. Chem. Bull.
1996, 45, 1601.
(35) Guest, M. F.; Bush, I. J.; van Dam, H. J. J.; Sherwood, P.; Thomas,
J. M. H.; van Lenthe, J. H.; Havenith, R. W. A.; Kendrick, J. Mol. Phys.
2005, 103, 719.
(36) Cipso and Cvinyl are the two carbon atoms bound to Pd and belonging
to, respectively, the arene ring of the NCN-pincer ligand or the double bond
of the trans-phenylvinyl ligand, respectively.
(37) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. ReV. 1988, 88,
899.
(38) Stradella, O. G.; Villar, H. O.; Castro, E. A. Theor. Chim. Act.
1986, 70, 67.

2328 Organometallics, Vol. 28, No. 7, 2009
Bonnet et al.
Table 1. Selected Distances and Angles in the Crystal Structures of Monometallic Complexes [Ru(C5Me5)(η6-1,3-(Me2NCH2)2C6H4)](BF4)
{[9](BF4)} and 2 and Bimetallic Complexes [4](BF4) and [5](BF4)
distance (Å) and
angles (deg)
[9](BF4)a
2b
[4](BF4)c
[5](BF4)d
d(Pd-C1)
1.914(2)
2.4280(17)
2.102(2)
2.101(2)
163.53(9)
172.78(9)
6.8(2)
1.9122(14)
2.4236(4)
2.1104(3)
2.1097(12)
164.05(5)
175.86(4)
11.45(6)
1.911(8)
2.411(2)
2.115(7)
2.113(7)
162.5(3)
179.7(2)
12.0(11)
1.907(8)
2.405(2)
2.123(7)
2.106(7)
163.0(3)
178.1(3)
11.2(11)
d(Pd-Cl)
d(Pd-N1)
d(Pd-N2)
θ(N1-Pd-N2)
θ(Cl-Pd-C1)
θ(PdN1N2C1-C1C2C6)
d(Ru-arene)
- to average plane
- to carbon (shortest)
- to carbon (longest)
d(Ru-C5Me5)
θ(arene-C5Me5)
d(Ru-Pd)
1.6997(16)
2.199(7)
2.222(7)
1.8096(6)
1.1(5)
1.72304(12)
2.1869(15)
2.2847(14)
1.81396(12)
7.35(8)
1.7014(7)
2.185(8)
2.226(9)
1.8009(7)
3.4(5)
1.7057(8)
2.178(9)
2.231(8)
1.8008(7)
3.3(5)
3.9150(2)
7.6945(9)
13.2(13)
7.5233(10)
15.2(13)
biphenyl angle
a Four very similar but distinct geometries are found in the unit cell of the low-T structure of [9]+ (150 K); only the geometry of one cation (that
containing Ru1) is shown here; see ref 31. b See ref 30. c See ref 26. d This work.
Table 2. PdII/PdIV Oxidation Potentials As Measured by Cyclic
Voltametrya
complex
Eox(PdII/PdIV), V
1
+0.68
+0.80
+0.92
+0.93
+0.68
+0.66
2b
[3]+
[4]+
[5]+
8
a Conditions: Pt electrode, scan rate 100 mV s-1, 0.1 M Bu4NPF6 in
acetonitrile, Fc/Fc+ as internal reference. All PdII/PdIV oxidations are
b
electrochemically irreversible. In addition to the irreversible PdII/PdIV
oxidation, the reversible one-electron reduction of the para-NO2 group
is observed at E1/2 ) -1.64 V (see text).
Scheme 3. Cross-Coupling Reaction between trans-Phenylvinyl-
boronic Acid and Vinyl Epoxide Catalyzed by NCN-Pincer
Palladium Complexes
Figure 2. Monitoring the cross-coupling reaction between trans-
phenylvinylboronic acid and vinyl epoxide, depending on the nature
of the catalyst. Total yields (linear + branched products) as a
function of time. Conditions: 1.6 mmol of boronic acid, 1.2 equiv
of epoxide, 2 equiv of Cs2CO3, 0.25 mol % in Pd catalyst, 2 equiv
of Cs2CO3, 25 °C, THF/H2O 7.5:1, di(n-hexyl) ether (internal
standard).
Table 3. Catalytic Properties of NCN-Palladium Pincer Complexes
in the Cross-Coupling Reaction between trans-Phenylvinylboronic
Acid and Vinyl Epoxidea
catalyst
GC yield (%) after 2 hb
t50 (h)c
1
2
8
29
98
22
77
63
93
9.7
3.3
7.0
4.5
8.5
3.0
According to these calculations, when following the series
[IS1], [IS5], [IS2], [IS4], [IS3] the following trends are observed
(see Table 4): (1) the positive charge on palladium increases
consistently with the evolution of the PdII/PdIV oxidation
potentials as measured by cyclic voltammetry (see Table 3);
(2) the energy level Eσ and the charge Q(Cvinyl) decrease, which
corresponds to a decreasing nucleophilicity of [ISn].
[3](PF6)
[4](BF4)
[5](BF4)
a t50 is defined as the time necessary to reach 50% conversion.
Conditions: 1.6 mmol of boronic acid, 1.2 eq. of epoxide, 2 equiv of
Cs2CO3, 25°C, THF/H2O 7.5:1. GC yields were determined using
di(n-hexyl) ether as internal standard. For comparison, under Szabo´’s
conditions (0.16 mmol of epoxide, 1.2 equiv of boronic acid, 2 equiv of
Cs2CO3, 2.5 mol % [PdBr(H-NCN)], 20 °C, THF/H2O 10:1) the isolated
yield after 4 h is 95% and the lin/br ratio is 2.3.19 b Using 1 mol %
catalyst. c Using 0.25 mol % catalyst.
Discussion
The synthesis of the bimetallic ruthenium-palladium com-
plexes [3]+-[5]+ demonstrates the generality of the applied
organometallic synthetic route. η6-Coordination of the
[Ru(C5Me5)]+ fragment to one of the arene rings of an NCN-
corresponding to the Pd-Cvinyl bond (HOMO-1 for [IS1-2],
HOMO-5 for [IS3-5]).

η6,η1 NCN-Pincer Ruthenium Palladium Complexes
Organometallics, Vol. 28, No. 7, 2009 2329
Table 4. Electronic Parameters Calculated by DFT for
Palladium-Transmetalated Intermediates [ISn] (n ) 1-5)a
IS1
IS5
IS2
IS4
IS3
Q(Pd)
Q(Cvinyl
Q(Cipso
0.610
-0.549
-0.253
0.774
0.618
-0.547
-0.244
0.789
0.624
-0.549
-0.224
0.793
0.629
-0.539
-0.265
0.808
0.642
-0.533
-0.262
0.816
)
)
BI(Pd-Cvinyl
)
Eσ (eV)b
-5.23
-5.69
-5.76
-5.95
-5.90
a The table is ordered according to increasing Q(Pd) values. b Energy
of the highest filled molecular orbital of σ character corresponding to
the Pd-Cvinyl bond (HOMO-1 for [IS1-2], HOMO-5 for [IS3-5]).
Figure 3. Monitoring the cross-coupling reaction between trans-
phenylvinylboronic acid and vinyl epoxide, depending on the nature
of the catalyst (same notations as in Figure 2). Linear/branched
ratio as a function of time. Conditions: 1.6 mmol of boronic acid,
1.2 equiv of epoxide, 2 equiv of Cs2CO3, 0.25 mol % in Pd catalyst,
2 equiv of Cs2CO3, 25 °C, THF/H2O 7.5:1, di(n-hexyl) ether
(internal standard).
Figure 5. Side view of the X-ray structures of complexes [9](BF4),
[5](BF4), and [4](BF4). BF4- counteranions and H-atoms have been
omitted for clarity.
Scheme 4. Intermediate Species [ISn] (n ) 1-5) Derived from
Catalysts 1, 2, [3]+, [4]+, and [5]+, Respectively, by Substitution
of a Chloride Ligand by a trans-Phenylvinyl Organic Groupa
steric constraints between the N-methyl groups of the NCN-
pincer fragment and the Me groups of the Cp* ligand borne by
ruthenium.29 However, this steric hindrance is small enough not
to destabilize η6-coordination directly to the arene ring of the
NCN-pincer metal complex itself as in [4]+, in which an
alternate binding site is absent.26 For the reaction of the less
hindered [RuCp]+ fragment with 8 such regioselectivity is not
observed, and the reaction yields a mixture of isomers that
appeared difficult to separate.
The effect of steric constraints in [4]+ is also apparent by
comparing the X-ray structures of [4]+, [5]+, and [9]+ (see
Figure 5). In compound [9]+ there is no palladium coordi-
nated to the NCN-pincer ligand, and both free amine ligands
are located at the opposite side of the arene ring, compared
to the ruthenium fragment. In [5]+, the NCN-pincer palladium
pincer fragment induces a weak steric bulk with the Cp*
ligand, but the N-methyl groups are still far away from the
Cp* methyl groups (cf. the long Pd-Ru distance). In both
compounds [5]+ and [9]+ the dihedral angle between the plane
of the arene ring and that of the cyclopentadienyl ring is
small, and the distance between the ruthenium center and
the average plane of the arene ring is short (see Table 1). In
[4]+ the ruthenium fragment is directly coordinated to the
arene ring of the NCN-pincer palladium moiety, which
minimizes the distance between the N-methyl and Cp* methyl
groups, hence maximizing steric congestion. Consequently,
the arene and cyclopentadienyl ring are no longer parallel,
and the Ru-arene distance is elongated (1.72304(12) Å in
[4]+ vs 1.7014(7) and 1.7057(8) Å in [5]+).
a
The chloride counteranion has been added to ensure neutral species.
The present catalytic studies show that bimetallic ruthe-
nium NCN-pincer complexes [3]+-[5]+ catalyze the cross-
coupling reaction between trans-phenylvinylboronic acid and
vinyl epoxide, yielding a mixture of linear and branched
alcohols. Earlier work showed that SeCSe- and NCN-pincer
palladium complexes, but also [Pd2(dba)3], are good catalysts
for this reaction.19 Reactivity studies detailed in Szabo´’s
original article showed that the mechanism with SeCSe-pincer
palladium catalysts is different from that with [Pd2(dba)3],
which notably results in different linear-to-branched ratios
at full conversion.
Figure 4. DFT-minimized structures of intermediates [ISn] (n )
1-5).
pincer palladium complex was realized by a direct reaction of
[Ru(C5Me5)(MeCN)3]+ with the organopalladium complex 1 or
8; that is, under the chosen reaction conditions η6-coordination
takes place without affecting the carbon-to-palladium bond. The
reaction of [Ru(C5Me5)(MeCN)3]+ with 8 occurs regioselectively
at the para-phenyl ring. This selectivity most likely results from
Due to the mild conditions of the reaction (room temperature,
weak base), decomposition of the ECE-pincer catalyst, thus
forming palladium(0), is not likely to occur in the reaction shown

2330 Organometallics, Vol. 28, No. 7, 2009
Bonnet et al.
in Scheme 3.39 When NCN- or SeCSe-pincer palladium catalysts
were used, the linear-to-branched ratios observed at full conver-
sion were different from that obtained with [Pd2(dba)3]. In
contrast, it has been shown that under severe reaction conditions
(60-110 °C, weak to strong bases) palladium(0) nanoparticles
are the real catalytic species when SCS- and PCP-pincer
palladium complexes are used as precatalysts in, for example,
Heck or Suzuki cross-coupling reactions.40,41
Scheme 5. Proposed Mechanism for the Cross-Coupling
Reaction between trans-Phenylvinylboronic Acid and Vinyl
Epoxide Catalyzed by NCN-Pincer Palladium Complexes
Significant differences are observed between NCN-, SeCSe-,
and phosphinite PCP-pincer palladium catalysts. Notably, phos-
phinite PCP-pincer palladium catalysts are inactive in this
reaction, which has been attributed to a reduced electron density
on palladium when coordinated to the good π-acceptor and poor
σ-donor P atoms.19 In contrast to this lack of reactivity, both
NCN-pincer and SeCSe-pincer complexes are efficient catalysts,
however, with some marked differences. Experimentally, the
NCN-pincer catalysts are less active and less selective (toward
the linear isomer) than the SeCSe-pincer catalysts.19 First, the
better σ-donating properties of N (compared to Se) might
significantly enhance the rate of the nucleophilic attack of [IS1]
(Scheme 4) on the epoxide and also decrease the rate of
transmetalation. The latter is indeed known to occur faster with
electron-poor PCP-pincer palladium catalysts.15 Second, the
steric bulk around the palladium center in each of the pincer-
palladium units differs because of the different sizes of the NR2
versus SeR donor groups.42
Our initial assumption, which partly initiated the present
study, was that the electronic effects induced by η6-
coordination of a [Ru(C5R5)]+ cation (R ) H or Me) at the
arene ring of NCN-pincer palladium complexes might change
the catalytic properties of the resulting complexes.20,43,44 In
the mechanism proposed for SeCSe-pincer catalysts, the
nucleophilic attack of the Pd-coordinated phenylvinyl anion
on vinyl epoxide is rate determining, and there is a correlation
between the nucleophilicity of the transmetalated species and
the rate of the reaction. If a similar trend was supposed to
exist for NCN-pincer catalysts, electron-poor complexes such
as 2, [3]+, or [4]+ should have a decreased catalytic activity,
compared to 1. In contrast, as observed in Figure 2, all
electron-poor complexes are more active than 1. As a result,
the nucleophilic attack of [ISn] on the epoxide might not be
the rate-determining step of the reaction with such catalysts.
This conclusion is confirmed by comparing the catalytic
activity of 1, 8, and [5]+: although the electron-richness of
the palladium center in these three complexes is comparable
(see Table 2), they appear to have significantly different
catalytic activities, with t50 being 2 to 3 times lower for [5]+
than for 8 or 1, respectively (see Table 3).
This hypothesis would explain the surprisingly lower selectivity
observed with the more sterically demanding NCN-pincer
catalysts, compared to SeCSe-pincer catalysts:19 if the rate-
determining step is transmetalation, attack of the η1-palladated
trans-phenylvinyl ligand is fast at both sites of the epoxide,
resulting in lower selectivity.
In the last decade, studies have shown that transmetalation is
not a single step but a combination of elementary steps.45-49 In
basic media, boronates are recognized as the reactive species for
transmetalation of boronic acids to palladium.50,51 Recently, an
intermediate has been experimentally observed, in which the
boronate is bound to palladium via an oxygen atom.46 On the basis
of the above consideration and considering the induction period
experimentally observed in our catalytic experiments (see Figure
2), we propose the reaction mechanism shown in Scheme 5.
In the first place, slow displacement of the halide by transmeta-
lation takes place to form [I1], which allows the pincer complex
to enter the catalytic cycle. The first step of the catalytic cycle thus
consists of the nucleophilic attack of [I1] to vinyl epoxide, which
Considering the electron-richness of NCN-pincer palladium
centers and the above results, we assume that the nucleophilic
attack is fast and the transmetalation much slower with these
catalysts, as suggested earlier by Szabo´ et al.15 As a conse-
quence, the transmetalation step might become rate determining.
(45) Braga, A. A. C.; Morgon, N. H.; Ujaque, G.; Lledos, A.; Maseras,
F. J. Organomet. Chem. 2006, 691, 4459.
(46) Sicre, C.; Braga, A. A. C.; Maseras, F.; Cid, M. M. Tetrahedron
2008, 64, 7437.
(39) No palladium black formation was observed in our experiments.
(40) Yu, K. Q.; Sommer, W.; Weck, M.; Jones, C. W. J. Catal. 2004,
226, 101.
(41) Sommer, W.; Yu, K.; Sears, H. S.; Ji, Y.; Zheng, X.; Davis, R. J.;
Sherrill, D.; Jones, C. W.; Weck, M. Organometallics 2005, 24, 4351.
(42) van Beek, J. A. M.; van Koten, G.; Ramp, M. J.; Coenjaarts, N. C.;
Grove, D. M.; Goubitz, K.; Zoutberg, M. C.; Stam, C. H.; Smeets, W. J. J.;
Spek, A. L. Inorg. Chem. 1991, 30, 3059.
(43) Gottker-Schnetmann, I.; White, P. S.; Brookhart, M. Organome-
tallics 2004, 23, 1766.
(44) van de Kuil, L. A.; Luitjes, H.; Grove, D. M.; Zwikker, J. W.; van
der Linden, J. G. M.; Roelofsen, A. M.; Jenneskens, L. W.; Drenth, W.;
van Koten, G. Organometallics 1994, 13, 468.
(47) Piechaczyk, O.; Cantat, T.; Mezailles, N.; Le Floch, P. J. Org.
Chem. 2007, 72, 4228.
(48) Nova, A.; Ujaque, G.; Maseras, F.; Lledos, A.; Espinet, P. J. Am.
Chem. Soc. 2006, 128, 14571.
(49) Alvarez, R.; Faza, O. N.; de Lera, A. R.; Cardenas, D. J. AdV. Synth.
Catal. 2007, 349, 887.
(50) Braga, A. A. C.; Morgon, N. H.; Ujaque, G.; Maseras, F. J. Am.
Chem. Soc. 2005, 127, 9298.
(51) Goossen, L. J.; Koley, D.; Hermann, H. L.; Thiel, W. J. Am. Chem.
Soc. 2005, 127, 11102.

η6,η1 NCN-Pincer Ruthenium Palladium Complexes
Organometallics, Vol. 28, No. 7, 2009 2331
affords the palladium-alcoholate species [I2]. In the presence of
an excess of water, quick protonation and ligand exchange yields
the free alcohol and the aquo palladium species [I3] (step 2). [I3]
subsequently reacts with a boronate molecule (step 3) to yield the
palladium-bound boronate complex [I4] proposed by Sicre et al.46
Step 4 consists of a monomolecular transmetalation, yielding
B(OH)3 and re-forming [I1].
Late transition metal-oxygen bonds are known to be kineti-
cally labile.52 The only known oxygenated ligands bound to
NCN-pincer palladium complexes in a monodentate fashion are
water and phenolates.53 For that reason we propose that
protonation by water of the bound alcoholate in [I2] and
exchange of the alcohol for water, i.e., step 2 in Scheme 5, are
fast processes. As already discussed above, NCN-pincer pal-
ladium species are electron-rich, so that step 1 is also fast. The
rate-determining step of the reaction is hence the transmetalation,
which is composed of steps 3 and 4.
Considering that the boronate species is an anionic
nucleophile, it will be sensitive to the electron density on
palladium. Removing electron density from the palladium
center, either by para-NO2 substitution or by η6-coordination
of [Ru(C5R5)]+, would increase the rate of step 3, hence the
rate of the reaction. This is what was experimentally
observed: electron-poor NCN-pincer palladium catalysts 2
and [3]+ and [4]+ are more active than 1. The mere additional
positive charge brought about by the cationic ruthenium
fragment will facilitate ion pair formation with the anionic
boronate species, which would explain why cationic catalysts
[3]+-[5]+ are overall more active than 1. In step 4 steric
congestion might also play an important role. On the basis
of the work of Sicre et al.,46 it is likely that the transition
state of this monomolecular elementary step involves a
hindered five-coordinate species with an η2-coordinated
phenylvinyl ligand, which ultimately evolves into η1-binding
of the phenylvinyl organic group in [I1]. Our catalytic results
suggest that releasing steric congestion around the palladium
center increases the catalytic activity of the NCN-pincer
catalysts: both [3]+ and [5]+ are more active than complex
[4]+. In the former the Cp ligand is less hindering than Cp*;
in the latter the [Ru(C5Me5)]+ fragment is much farther from
palladium (cf. the Pd-Ru distance in Table 1). A less
hindered catalyst might lead to step 4 being faster, hence to
a faster reaction rate.
Overall, complex [5]+ shows a surprisingly high catalytic
activity, which might arise from (1) the presence of an additional
positive charge on the catalyst and (2) a reduced steric
congestion. In addition, both faces of the catalyst are available
for the approach of the nucleophile with [5]+, whereas with
[3]+ and [4]+ one face is sterically hindered due to the
η6-coordinated [Ru(C5R5)]+ fragment. This example highlights
the limited impact of intramolecular electronic effects in our
catalytic results; thus, “through-bond” interactions are not
affecting too much the catalytic activity of NCN-pincer pal-
ladium catalysts. On the contrary, steric hindrance and geometry
seem to play a major role here, as well as more unusual effects
related to the presence of a positive charge that modify ion pair
formation (“through-space” interactions).
(from 0 to 2.3) during the course of the reaction, with the
branched product being formed faster than the linear one (see
Figure 3). At high conversions, the linear-to-branched ratio
converged toward the same value of about 2.3, independent of
the nature of the NCN-pincer palladium catalyst. The occurrence
of a process involving interconversion between the linear and
branched alcohols seems unlikely, as it would imply reversible
C-C bond cleavage/bond formation. The present results suggest,
however, that the disappearance of the epoxide and/or the
formation of the products might influence the selectivity of the
reaction. Alcohols are known to bind reversibly to boronic
acids.54,55 Lewis bases such as epoxides might reversibly bind
to boronic acids, but also to the palladium center.56 Several
equilibria between vinyl epoxide, the linear and branched
alcohols, and the Lewis acids present in solution might be
considered, which would play a role in the rate of formation of
the linear and/or branched products when the concentrations of
the epoxide and alcohols vary. Additional mechanistic studies
must be undertaken to provide a better picture of the mechanism
shown in Scheme 5.
Conclusion
In this report, we present a series of ruthenium-palladium
η6,η1-bimetallic complexes of the NCN-pincer type, in which
the organic ligand is at the same time η6- and η1-coordinated
to two different metal centers. Comparison of the X-ray crystal
structures of [4](BF4),26 [5](BF4), and [9](BF4)31 shows that
steric hindrance between the N-methyl groups of the NCN-
pincer ligand and the bulky Cp* ligand increases in the series
[9](BF4) < [5](BF4) < [4](BF4). We tested the catalytic activity
of the ruthenium-modified NCN-pincer palladium complexes
[3]+-[5]+ in the cross-coupling reaction between trans-phe-
nylvinylboronic acid and vinyl epoxide: they have a better
catalytic activity than their monometallic palladium precursor
1. DFT calculations performed on the transmetalated species
[ISn] and palladium oxidation potentials measured in solution
agree with an opposite reactivity trend compared to previously
studied SeCSe-pincer palladium catalysts: electron-poor NCN-
pincer palladium complexes have a higher catalytic activity than
electron-rich ones. On the basis of this observation we propose
a reaction mechanism where transmetalation is rate determining
(Scheme 5). According to our analysis, the positive charge
induced by the η6-coordinated ruthenium fragment facilitates
the nucleophilic attack of trans-phenylvinylboronate on pal-
ladium (step 3). Meanwhile, the diminished steric congestion
in [5]+ explains (step 4) its surprisingly better catalytic
properties. Finally, kinetic studies have brought intriguing
information on the evolution of the linear/branched ratio during
the course of the reaction, which could not be observed by
looking only at the final state of the catalytic system.
Experimental Part
Synthesis. All reactions using sensitive reagents were performed
under an atmosphere of dinitrogen using Schlenk techniques. THF
and Et2O were dried over Na/benzophenone. MeCN and CH2Cl2
were dried over CaH2, and all solvents were freshly distilled under
1
nitrogen prior to use. H (300.1/400.0 MHz) and 13C (75.5/100.6
It might be noted that the evolution of the linear-to-branched
ratio with reaction progress showed that selectivity changed
MHz) NMR spectra were recorded on a Varian INOVA 300 or
(52) Bryndza, H. E.; Fong, L. K.; Paciello, R. A.; Tam, W.; Bercaw,
J. E. J. Am. Chem. Soc. 1987, 109, 1444.
(53) Alsters, P. L.; Baesjou, P. J.; Janssen, M. D.; Kooijman, H.;
Sichererroetman, A.; Spek, A. L.; van Koten, G. Organometallics 1992,
11, 4124.
(54) Fujita, N; Shinkai, S; James, T. D. Chem.-Asian J. 2008, 3, 1076.
(55) Selander, N.; Kipke, A.; Sebelius, S.; Szabo, K. J. J. Am. Chem.
Soc. 2007, 129, 13723.
(56) Solin, N.; Kjellgren, J.; Szabo, K. J. J. Am. Chem. Soc. 2004, 126,
7026.

2332 Organometallics, Vol. 28, No. 7, 2009
Bonnet et al.
400 MHz spectrometer; 31P{1H} (121.5 MHz) and 19F (376 MHz)
NMR spectra were recorded on a Varian INOVA 400 MHz
spectrometer. Chemical shift values are reported in ppm (δ) relative
to Me4Si (1H and 13C NMR). MALDI-TOF measurements were
carried out on an Applied Biosystems Voyager DE-STR MALDI-
TOF MS. Elemental analyses were performed by H. Kolbe
Microanalysis Laboratories, Mu¨lheim, Germany. GC analyses were
performed on a Perkin-Elmer Clarus-500 gas chromatograph. GC-
MS measurements were measured on a Perkin-Elmer Autosys-
temXL gas chromatograph with an attached Perkin-Elmer Turbo-
mass Upgrade mass spectrometer. For column chromatography,
Merck silica gel 60 (230-400 mesh) was used. All standard
reagents were purchased from Acros Organics or Sigma-Aldrich,
and used as received. para-Phenyl NCN-pincer ligand 7,57 pincer
metal complexes 158 and 2,13 and ruthenated derivatives [9]+,29
[3]+, and [4]+ 26 were prepared according to literature procedures.
DIRDIF9963 and was refined on F2 with SHELXL-97.64 The
temperature of the data collection was controlled using the Oxford
Cryostream 600 system (manufactured by Oxford Cryosystems).
The H-atoms were placed at calculated positions (AFIX 23, AFIX
43, or AFIX 137) with isotropic displacement parameters having
values 1.2 or 1.5 times Ueq of the attached C-atom. Illustrations
and structure validations were done with PLATON.65
Crystal Growth. A dichloromethane solution of the complex
(5 mg/mL) was put in an open 2 mL vial, which was contained in
a closed 20 mL vial. The larger vial contained about 6 mL of
hexane. As vapor diffusion took place, crystal growth occurred at
room temperature without any additional effort, and crystals
appeared at the bottom of the smaller vial after one to two weeks.
X-ray structure data for [5]+: C28H38ClN2PdRu · BF4, Mr ) 732.33,
3
j
colorless plate, 0.46 × 0.18 × 0.06 mm , triclinic, P1 (no. 2), a )
14.1334(5) Å, b ) 14.7569(5) Å, c ) 16.1141(6) Å, R )
78.019(2)°, ꢀ ) 77.718(2)°, γ ) 63.285(2)°, V ) 2909.14(18) Å3,
Z ) 4, Dx ) 1.672 g cm-3, µ ) 1.27 mm-1; 41 951 reflections
were measured at 150(2) K after the crystal had been flash-cooled
from room temperature. Multiscan empirical absorption corrections
8: A 1.69 g amount of ligand 7 (5.78 mmol) was weighed in a
flame-dried Schlenk flask and put under a nitrogen atmosphere.
Then 40 mL of dry pentane was added, and the solution was cooled
to -78 °C. A 3.5 mL amount of n-BuLi (as 1.6 M solution in
hexanes, 5.6 mmol) was added dropwise to give an orange solution,
which was stirred at -78 °C for 30 min. The cooling bath was
then removed, and the solution warmed to room temperature
overnight (18 h) to yield a yellowish suspension. Pentane was
evaporated, replaced by 100 mL of dry diethyl ether, the suspension
was cooled to -78 °C, and 1.96 g of PdCl2(SEt2)2 (5.49 mmol)
was slowly added under a flow of dinitrogen. The suspension was
stirred 1 h at -78 °C, then allowed to warm to room temperature
overnight (24 h) under nitrogen. Diethyl ether was evaporated, and
the crude material was dissolved in 200 mL of dichloromethane,
washed with water and brine, dried over MgSO4, and filtered over
Celite. Addition of 100 mL of hexane and slow evaporation of
CH2Cl2 afforded 1.48 g of crystalline product (62%). Characteriza-
tion was identical to the published data.20
[5]+: A 159 mg amount of 8 (388 µmol) was weighed in a dry
Schlenk flask and put under inert atmosphere. A solution of
[RuCp*(MeCN)3](BF4)59 (174 mg, 389 µmol) in 5 mL of dry
dichloromethane was added, and the mixture stirred under nitrogen
at room temperature for 3 days. The crude solution was put on top
of an 80 mL silica gel column and eluted with CH2Cl2 containing
1-3% methanol to remove unreacted starting materials. Yield: 229
mg of [5](BF4) as a whitish powder (81%). 1H NMR δ (400 MHz,
ppm in acetone-d6): 7.31 (s, 2H, m), 6.54 (d, 2H, o′, J 5.9), 6.16 (t,
2H, m′, J 6.2), 6.11 (t, 1H, p′, J 5.7), 4.20 (s, 4H, CH2N), 2.92 (s,
12H, NMe2), 1.89 (s, 15H, C5Me5). 13C NMR δ (100 MHz, ppm
in acetone-d6): 162.2 (i), 147.6 (o), 129.5 (p), 119.3 (m), 103.5
(i′), 97.1 (CIV5R5), 88.4, 88.2, 85.4 (o′, m′, p′), 75.3 (CH2N), 53.2
(NMe2), 10.3 (C5Me5). 19F NMR δ (376 MHz, ppm in acetone-d6):
-152.1 (s). MALDI-TOF m/z (calc): 645.11 (645.08, [M - BF4]+).
C, H, N: 45.92/5.23/3.83 (calc); 45.79/5.20/3.78 (found).
were applied to the data using the program TWINABS:66 Tmin
0.710 and Tmax ) 0.930; 12 040 reflections were unique (Rint
)
)
0.0583), of which 9157 were observed with the criterion of I >
2σ(I); 784 parameters were refined with 236 least-squares restraints.
R1/wR2[I > 2σ(I)]: 0.060/0.110. R1/wR2[all reflns]: 0.096/0.128. S
) 1.15. The residual electron density was found between -1.25
and 1.49 e Å-3. Crystals were nonmerohedrally twinned with a
2-fold axis along the [-1 1 0] direction as the twin operation. The
minor twin fraction refined to 0.467(1). Occupancy factors for the
major components of the disordered counteranions were 0.65(3)
and 0.57(1).
Catalytic Studies. Eight experiments were run simultaneously
using a ChemSpeed automated apparatus. In each reaction vessel,
1.04 g (3.2 mmol) of Cs2CO3, 1.5 mL of a 1.06 M solution of
trans-phenylvinylboronic acid in tetrahydrofuran (1.6 mmol), 400
µL of water, 100 µL of dihexyl ether (internal reference), and 160
µL (1.2 equiv) of vinyl epoxide were added one after another.
Subsequently, 1 or 0.25 mol % palladium catalyst was added as a
solid, and the vial containing the catalyst was rinced with 1.5 mL
of tetrahydrofuran that was added to the vessel. After preparation
of the 8 vessels, the vessel holder was mounted on the ChemSpeed
system thermostated at 25 °C, and t ) 0 was set upon starting
stirring at 1000 rpm. At regular intervals, stirring was stopped, the
biphasic mixture left untouched during 10 s, and 50 µL of the upper
layer (THF) taken via a syringe in each of the 8 vessels and poured
into 3.0 mL of dichloromethane. Stirring of the vessels was then
switched on before workup of the 8 samples was started. Workup
consisted in adding 1.0 mL of 1 M NaOH to the dichloromethane
phase and extracting it three times to remove unreacted boronic
acid from the organic layer. The dichloromethane phase was
analyzed by GC using dihexyl ether as internal reference.
DFT Calculation. All DFT calculations were performed using
B3LYP/LANL2DZ as implemented in the GAMESS-UK program35
and using MOLDEN67 to visualize the structures and edit Z-
matrices. The geometry found in the X-ray structures of complexes
[4]+ and [5]+ was used as a starting point, using for [5]+ the
geometry containing Pd1 and Ru1. The chloride ligand was replaced
X-ray Crystallography. All reflection intensities were measured
using a Nonius KappaCCD diffractometer (rotating anode) with
graphite-monochromated Mo KR radiation (λ ) 0.71073 Å) using
the program COLLECT.60 The program PEAKREF61 was used to
refine the cell dimensions. Data reduction was done using the
program EVALCCD.62 The structure was solved with the program
(63) Beurskens, P. T.; Admiraal, G.; Beurskens, G.; Bosman, W. P.;
Carcia-Granda, S.; Gould, R. O.; Smits, J. M. M.; Smykalla, C. The
DIRDIFF99 Program System; Technical Report of the Crystallography
Laboratory, University of Nijmegen; University of Nijmegen: Nijmegen,
The Netherlands, 1999.
(64) Sheldrick, G. M. SHELXL97; University of Go¨ttingen: Go¨ttingen,
Germany, 1997.
(57) Steenwinkel, P.; James, S. L.; Grove, D. M.; Veldman, N.; Spek,
A. L.; van Koten, G. Chem.-Eur. J. 1996, 2, 1440.
(58) Terheijden, J.; van Koten, G.; Mul, W. P.; Stufkens, D. J.; Muller,
F.; Stam, C. H. Organometallics 1986, 5, 519.
(59) Fagan, P. J.; Ward, M. D.; Calabrese, J. C. J. Am. Chem. Soc. 1989,
111, 1698.
(60) Nonius. COLLECT; Nonius B. V.: Delft, The Netherlands1998.
(61) Schreurs, A. M. M. PEAKREF; Utrecht University: Utrecht, The
Netherlands, 2005.
(62) Duisenberg, A. J. M.; Kroon-Batenburg, L. M. J.; Schreurs,
A. M. M. J. Appl. Crystallogr. 2003, 36, 220.
(65) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7.
(66) Sheldrick, G. M. TWINABS; University of Go¨ttingen: Go¨ttingen,
Germany, 2003.
(67) Schaftenaar, G.; Noordik, J. H. J. Comput.-Aided Mol. Des. 2000,
14, 123.

η6,η1 NCN-Pincer Ruthenium Palladium Complexes
Organometallics, Vol. 28, No. 7, 2009 2333
by a trans-phenylvinyl ligand parallel to the arene ring, and the
structures were minimized without symmetry constraint (conver-
gence threshold: 10-4). A chloride counteranion was then added
on the side of the ruthenium-arene complex, and the geometries
were minimized again, to give the neutral geometries shown in
Figure 4.
Scientific Research (CW-NWO) and in part by the Stichting
Nationale Computerfaciliteiten (National Computing Facili-
ties Foundation, NCF) for the use of supercomputer facilities.
Supporting Information Available: Crystallographic data for
[5](BF4) in CIF format, as well as the Cartesian coordinates for
the DFT-minimized structures shown in Scheme 4, are available
free of charge via the Internet at http://pubs.acs.org.
Acknowledgment. We gratefully acknowledge the support
of this research by Utrecht University. This work was
supported in part (M.S., A.L.S.) by the Council for the
Chemical Sciences of The Netherlands Organization for
OM801200C