Pyrenoxy-based NCN-pincer palladium(II) molecular tweezers: Synthesis, properties in solution and catalysis

Article abstract of DOI:10.1002/ejoc.200210601

A highly flexible molecular tweezer, (aqua) (2,6-bis[(dimethylamino)methyl]-4-{3-[(1-pyrenyloxy)methyl]-5- (trimethylsilyl)benzyloxy}phenyl)palladium(II) tetrafluoroborate (1a), constructed from a para-hydroxy-functionalized NCN-pincer palladium(II) complex {NCN-OH = [2,6-(Me₂NCH₂)₂ C₆H₂-4-OH]}, a pyrenoxy unit and a xylyl spacer, was synthesized. X-ray analysis of the structure of 1-[3-({3,5-bis[(dimethylamino)methyl]-4-(trimethylsilyl) phenyloxy}methyl)-5-bromobenzyloxy]pyrene (10), a precursor ligand of la, showed a completely flattened conformation in the solid state, imposed by favorable intramolecular π-stacking interactions. The molecular structure of la in solution proved to be fluxional on the NMR timescale. UV/Vis and (time-resolved) fluorescence spectroscopy revealed that, in spite of the high flexibility, the catalyst and pyrenoxy units are in close proximity because of cation-π interactions. This was further supported by conformer searches using molecular mechanics (MM2). The binding affinity of picric acid towards the pyrenoxy moiety was studied by ¹H NMR titrations, revealing association constants K_a in the order of 10¹ to 10² M^-1. Tweezer la shows small rate enhancements in aldol condensations of aromatic aldehydes with methyl isocyanoacetate compared to (aqua)(2,6-bis[(dimethylamino)methyl]4-{[3-(trimethylsilyl) benzyloxy}phenyl)palladium(II) tetrafluoroborate (2), which lacks the pyrenoxy unit. The tweezer was immobilized on a G2-carbosilane dendrimer, generating the dendritic (aqua)(G2-carbosilane)palladium complex 1b, showing a decreased catalytic activity compared to 1a. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003.

Full text of DOI:10.1002/ejoc.200210601

FULL PAPER
Pyrenoxy-Based NCN-Pincer Palladium(II) Molecular Tweezers: Synthesis,
Properties in Solution and Catalysis
Martijn Q. Slagt,[a] Johann T. B. H. Jastrzebski,[a] Robertus J. M. Klein Gebbink,[a]
Hendrik J. van Ramesdonk,[b] Jan W. Verhoeven,[b] Dianne D. Ellis,[c] Anthony L. Spek,[c] and
Gerard van Koten*[a]
Keywords: Homogeneous catalysis / Palladium / Fluorescence spectroscopy / Ligand design / Pi interactions
A highly flexible molecular tweezer, (aqua)(2,6-bis[(dimethyl-
amino)methyl]-4-{3-[(1-pyrenyloxy)methyl]-5-(trimethyl-
close proximity because of cation−π interactions. This was
further supported by conformer searches using molecular
mechanics (MM2). The binding affinity of picric acid towards
silyl)benzyloxy}phenyl)palladium(II) tetrafluoroborate (1a),
constructed from a para-hydroxy-functionalized NCN-pincer
palladium(II) complex {NCN−OH = [2,6-(Me2NCH2)2C6H2-4-
OH]}, a pyrenoxy unit and a xylyl spacer, was synthesized.
X-ray analysis of the structure of 1-[3-({3,5-bis[(dimethyl-
amino)methyl]-4-(trimethylsilyl)phenyloxy}methyl)-5-
the pyrenoxy moiety was studied by 1H NMR titrations, re-
−1
vealing association constants Ka in the order of 101 to 102
M .
Tweezer 1a shows small rate enhancements in aldol con-
densations of aromatic aldehydes with methyl isocyanoacet-
ate compared to (aqua)(2,6-bis[(dimethylamino)methyl]-
bromobenzyloxy]pyrene (10),
showed a completely flattened conformation in the solid
state, imposed by favorable intramolecular π-stacking inter- was immobilized on a G2-carbosilane dendrimer, generating
a
precursor ligand of 1a,
4-{[3-(trimethylsilyl)benzyloxy}phenyl)palladium(II
)
tetra-
fluoroborate (2), which lacks the pyrenoxy unit. The tweezer
actions. The molecular structure of 1a in solution proved to
be fluxional on the NMR timescale. UV/Vis and (time-re-
solved) fluorescence spectroscopy revealed that, in spite of
the dendritic (aqua)(G2-carbosilane)palladium complex 1b,
showing a decreased catalytic activity compared to 1a.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
the high flexibility, the catalyst and pyrenoxy units are in Germany, 2003)
Introduction
features are represented in a system published earlier by our
group. Catalytic NCN-pincer palladium() cationic sites
were encapsulated by strategic placement of diphenylsilane
moieties and subsequently immobilized on a dendritic
backbone.[4] This rather rigid encapsulation of the catalytic
site resulted in enhanced reaction rates, but had little influ-
ence on the selectivity of the studied aldol reaction cata-
lyzed by the cationic palladium() sites. This approach,
where functional groups are introduced in the vicinity of
the metal site to modify its (catalytic) properties by second-
ary interactions, has motivated us to construct the new mo-
lecular tweezers 1a and 1b based on a pincer palladium()
complex and a pyrenoxy moiety, as shown in Scheme 1.
In this paper we report the synthesis of 1a, 1b and their
relevant variations 2Ϫ4 (Scheme 2), according to an ap-
proach in which the separately prepared building blocks are
assembled on the xylyl spacer group by highly flexible ether
linkages. This spacer has an additional reactive site, which
allows immobilization on suitable (dendritic) support sys-
tems. In 1a, a trimethylsilyl substituent is attached at this
anchoring point, serving as a mimic for dimethylsilyl-func-
tionalized carbosilane dendrimers. The dendrimer-immobil-
ized multisite catalyst 1b contains 36 pyrenoxy-based NCN
palladium moieties attached to an inert carbosilane den-
The use of NCN-pincer complexes {NCN ϭ 2,6-bis[(di-
methylamino)methyl]phenyl anion} as building blocks in
the construction of multisite or dendritic catalysts, of supra-
molecular assemblies and of sensor materials has attracted
considerable attention.[1] Synthetic flexibility for the intro-
duction of functional groups, combined with the variety of
metallation procedures available, are important issues in
their construction.[2] Whereas ligand substitution allows im-
mobilization of transition metal complexes on various sup-
port systems,[3] the introduction of functional groups in the
vicinity of a metal site can alter its catalytic properties. Both
[a]
Debye Institute, Department of Metal-Mediated Synthesis,
Utrecht University,
Padualaan 8, 3584 CH Utrecht, The Netherlands
Fax: (internat.) ϩ 31-30/252-3615
E-mail: g.vankoten@chem.uu.nl
[b]
Institute of Molecular Chemistry, University of Amsterdam,
Nieuwe Achtergracht 129, 1018 WS Amsterdam, The
Netherlands
[c]
Bijvoet Center for Biomolecular Research, Department of
Crystal and Structural Chemistry,
Padualaan 8, 3584 CH Utrecht, The Netherlands
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
1692
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejoc.200200601
Eur. J. Org. Chem. 2003, 1692Ϫ1703

Pyrenoxy-Based NCN-Pincer Palladium(II) Molecular Tweezers
FULL PAPER
Scheme 1. (Dendritic) NCN palladium() complexes 1a and 1b
functionalized with pyrenoxy units
Scheme 3. Synthesis of precursor ligands: i) K2CO3, 18-crown-6; ii)
Bu4NF; iii) 7, K2CO3, 18-crown-6; iv) tBuLi, Me3SiCl; v) tBuLi,
Si[(CH2)3Si{(CH2)3Si[(CH2)3SiMe2Cl]3}3]4; vi) Bu4NF, 9, K2CO3,
18-crown-6; vii) 6, K2CO3, 18-crown-6
(Scheme 3). An excess of tribromide 5 was treated with pyr-
enoxy unit 6 to afford a 1:2:1 mixture of unchanged 5,
monofunctionalized 7 and difunctionalized 8. Selective extrac-
tion of 5 from the reaction mixture with refluxing pentane
afforded a 2:1 mixture of pyrenyl ethers 7 and 8. This unre-
solved mixture was used in the subsequent coupling with
3,5-bis[(dimethylamino)methyl]-4-(trimethylsilyl)phenoxide,
which was prepared in situ from 9 by deprotection with
tetrabutylammonium fluoride. Purification of the coupling
products by careful precipitation of the bis(pyrenyl) ether 8
from dichloromethane/pentane afforded pure 10 in the
mother liquor. Precursor ligand 10 was subsequently at-
tached to either a trimethylsilyl substituent to prepare 11a,
or to a G2-carbosilane dendrimer to produce 11b. Reaction
of 10 with tBuLi followed by treatment with trimethylsilyl
chloride or a chlorosilane-functionalized G2-carbosilane
Scheme 2. NCN palladium() complex 2 and pyrenyl ethers 3 and 4
drimer. Anchoring these systems to dendrimers allows cata-
lyst recycling by means of nanofiltration techniques.[5] The
molecular flexibility of 1a in solution is studied by spectro-
scopic techniques using the pyrenoxy group as fluorescent
probe, and by molecular modeling studies. Furthermore,
the influence of the pyrenoxy moiety on the application of
1a as homogeneous catalysts in aldol reactions is described,
together with its binding properties to aromatic molecules.
Results
Synthesis
Precursor ligand 10 was synthesized by connecting 1-pyr- dendrimer afforded ligands 11a and 11b, respectively
enol (6) and the NCN-pincer ligand 9 to the linker 3,5- (Scheme 3). Similarly, lithiation of aromatic bromide 8 fol-
bis(bromomethyl)phenyl bromide (5) by ether syntheses lowed by quenching with Me3SiCl afforded bis(pyrenyl)
Eur. J. Org. Chem. 2003, 1692Ϫ1703
www.eurjoc.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1693

G. van Koten et al.
FULL PAPER
ether 4. Ligand 14, the trimethylsilyl precursor of complex
2, and pyrenyl ether 3 were obtained by similar coupling
and lithiation procedures, starting from 3-(bromometh-
yl)phenyl bromide as anchoring point for either NCN li-
gand 9 or pyrenoxy unit 6 (Scheme 3).
Earlier work by Valk et al.[6] and Steenwinkel et al.[7] had
demonstrated that in R2NCH2-substituted aromatic sys-
tems replacement of an ortho-aryl-H by a trimethylsilyl sub-
stituent can steer the regioselectivity of an electrophilic pal-
ladation reaction. Thus, treatment of 11a, 11b and 14 with
Pd(OAc)2 and subsequent addition of LiCl led to the selec-
tive formation of the corresponding (chloro)palladium()
complexes 12a, 12b and 15, respectively (Scheme 4). The
quantitative metallation of the dendritic polyligand 11b by
this mild and selective method is noteworthy. The cationic
palladium() complexes 1a and 1b were prepared from the
palladium halide complexes 12a and 12b, respectively. Cat-
ionic complex 2 was prepared from 14 in a similar manner.
This approach offers the advantage that a wide number of
chemical reactions, e.g. lithiation at the linker and sub-
sequent immobilization, can be performed prior to a site-
selective metallation step. The (chloro)palladium() com-
plexes 12a, 12b and 15, and their corresponding cationic
complexes 1a, 1b and 2, can be handled and stored under
aerobic and/or moist conditions. However, in concentrated
solutions slow decomposition is observed for all complexes,
i.e. formation of Pd0. For this reason, the cationic com-
plexes were freshly prepared for use in catalysis.
Figure 1. ORTEP representation of 10 (displacement ellipsoids
with 30% probability); hydrogen atoms are omitted for clarity
Figure 2. Packing plot of 10 with unit cell, showing the π-stacked
layered structure
forded suitable crystals for structure determination. The
molecular and crystal structure, together with selected bond
lengths and angles, are given in Figures 1 and 2, and
Table 1.
The structure of 10 shows both the pincer ligand and the
pyrenoxy unit connected to the linker. Interesting aspects
of the structure include the silicon center which is σ-bonded
to three methyl groups, to Cipso [C(1)] of the ligand, and in
addition has an interaction with one of the ortho-Me2NCH2
˚
groups. The Si(1)···N(1) distance is 3.264(2) A, which is sub-
stantially smaller than the sum of the van der Waals radii
˚
of nitrogen and silicon (3.65 A). As a consequence of this
hypervalency of silicon,[8] the methyl groups on the silicon
atom are distorted from a true tetrahedral geometry, with
the Si(1)ϪC(13) bond being significantly longer than the
Si(1)ϪC(14) and Si(1)ϪC(15) bonds. Steenwinkel et al. re-
˚
ported an even shorter N···Si contact of 3.040 A for Me3Si-
substituted NCN-pincer ligands.[7b]
In the crystal lattice, the molecule is completely flattened
as a result of crystal packing (Figure 2). The bulky substitu-
ents on the ligand, i.e. the Me3Si and ortho-Me2NCH2 sub-
stituents, are placed in the crystal lattice in such a way that
optimal stacking between the aromatic parts can take place.
The requirements for good FFCE (face-to-face, center-to-
edge) stacking[9] are fulfilled for several intermolecular
ringϪring interactions. For example, the interaction be-
Scheme 4. Synthesis of 1a and 1b by site-selective palladation and
subsequent halide abstraction: i) Pd(OAc)2, LiCl; ii) AgBF4
Crystal Structure of 10
All attempts to obtain suitable crystals for X-ray analysis tween the aromatic ring containing C(17)ϪC(22) and the
of cationic complex 1a or its (chloro)palladium() analogue ring C(34)ϪC(39) of the neighboring molecule, has the
˚
˚
12a failed. Crystallization of the precursor ligand 10, af- stacking parameters α ϭ 2.9°, x ϭ 1.8 A and z ϭ 3.5 A,
1694
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 1692Ϫ1703

Pyrenoxy-Based NCN-Pincer Palladium(II) Molecular Tweezers
FULL PAPER
˚
Table 1. Selected bond lengths [A], bond angles [°] and dihedral angles [°] of 10; the numbering refers to Figure 1
Bond lengths
Br(1)ϪC(19)
N(1)-C(7)
N(1)-C(8)
1.903(2)
1.460(3)
1.463(3)
1.455(3)
1.467(3)
1.458(4)
1.467(4)
1.371(3)
O(1)ϪC(16)
O(2)ϪC(23)
O(2)ϪC(24)
Si(1)ϪC(1)
Si(1)ϪC(13)
Si(1)ϪC(14)
Si(1)ϪC(15)
Si(1)ϪN(1)
1.371(3)
1.419(3)
1.372(3)
1.909(2)
1.886(2)
1.870(3)
1.879(3)
3.264(2)
N(1)-C(9)
N(2)ϪC(10)
N(2)ϪC(11)
N(2)ϪC(12)
O(1)ϪC(4)
Bond angles
C(4)ϪO(1)ϪC(16)
C(7)ϪN(1)ϪC(8)
C(7)ϪN(1)ϪC(9)
C(8)ϪN(1)ϪC(9)
C(10)ϪN(2)ϪC(11)
C(10)ϪN(2)ϪC(12)
C(11)ϪN(2)ϪC(12)
C(13)ϪSi(1)ϪC(1)
117.9(2)
C(13)ϪSi(1)ϪC(14)
C(13)ϪSi(1)ϪC(15)
C(14)ϪSi(1)ϪC(1)
C(14)ϪSi(1)ϪC(15)
C(15)ϪSi(1)ϪC(1)
C(21)ϪC(23)ϪO(2)
C(23)ϪO(2)ϪC(24)
O(1)ϪC(16)ϪC(17)
110.0(1)
110.3(2)
111.5(2)
109.6(2)
110.2(2)
109.1(2)
110.9(2)
108.7(2)
101.1(1)
112.4(1)
109.1(1)
114.9(1)
110.2(2)
117.1(2)
108.1(2)
Dihedral angles
C(3)ϪC(2)ϪC(10)ϪN(2)
C(4)ϪO(1)ϪC(16)ϪC(17)
C(5)ϪC(6)ϪC(7)ϪN(1)
C(16)ϪO(1)ϪC(4)ϪC(5)
46.0(3)
Ϫ173.4(2)
Ϫ123.6(2)
171.6(2)
C(22)ϪC(21)ϪC(23)ϪO(2)
C(23)ϪO(2)ϪC(24)ϪC(37)
C(24)ϪO(2)ϪC(23)ϪC(21)
O(1)ϪC(16)ϪC(17)ϪC(18)
Ϫ175.3(2)
189.1(2)
179.9(2)
177.0(2)
which all lie within the boundaries for good FFCE stack- yield by a factor of 10.[10] Fluorescence-lifetime measure-
ing.
ments revealed that the decay curves of 1a and 11a were
both due to one-component systems. In dichloromethane,
the fluorescence lifetime for 1a is 1.8 ns while the lifetime
of its non-metallated analogue 11a is 10.6 ns (Figure 3). Im-
mobilization of the system on a G2-carbosilane dendrimer
(11b and 1b) almost completely quenched the fluorescence
of the pyrenoxy fragment.
1
UV/Vis, Fluorescence, and H NMR Spectroscopic Studies
As a result of relatively unhindered rotation around the
ether linkages, a wide range of conformations can be ad-
opted by complex 1a in solution. To obtain an indication
of the preferred conformation(s) in solution, 1a was sub-
jected to UV/Vis, fluorescence and variable-temperature 1H
NMR studies. The highly fluorescent pyrenoxy moiety of-
fers an excellent probe for studying the conformational be-
havior of this cationic palladium complex. Conformational
molecular mechanics studies were performed to obtain
more insight.
1
Variable-temperature H NMR studies were performed
on a solution of 1a in CDCl3 between Ϫ60 and ϩ40 °C. At
room temperature, the signals corresponding to the benzylic
protons of the central linker were broadened. Lowering of
the temperature led, apart from broadening of all signals
due to an increase in solvent viscosity, to severe broadening
of the resonances of these benzylic protons. However, no
distinct conformation could be frozen out at Ϫ60 °C. At
higher temperatures all signals sharpened, but no other sig-
Figure 3. Fluorescence decay curves for 1a (τ ϭ 1.8 ns) and 11a
(τ ϭ 10.6 ns)
1
nificant changes in the H NMR spectrum were observed.
The electronic absorption spectrum of cationic pal-
ladium() complex 1a shows small bathochromic shifts for
the longer wavelength bands and significant hypochroism
To gain more insight into the flexibility of the 1,3-bis(hy-
compared to its non-metallated analogue 11a. Besides this, droxymethyl)benzene ethers, bis(pyrenoxy) ether 8 was sub-
the fluorescence bands of 1a show a bathochromic shift of jected to time-resolved fluorescence spectroscopy. A new
approximately 5 nm and a decrease of the relative quantum emission band (λmax ϭ 450 nm) appeared, apart from the
Eur. J. Org. Chem. 2003, 1692Ϫ1703
www.eurjoc.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1695

G. van Koten et al.
FULL PAPER
monomer emission, in the steady-state fluorescence spec- Molecular Mechanics
trum of 8 (Figure 4, a). This band was attributed to excimer
The preferred conformations of complex 1a were calcu-
lated by conformer searches, allowing stepwise rotation
around the six bonds connecting the NCN ligand and the
pyrenoxy unit to the linker followed by a molecular mech-
anics optimization of the structure. All optimized conform-
emission, owing to the proximity of two pyrenoxy frag-
ments. The formation of this intramolecular excimer was
studied by multiwavelength time-resolved fluorescence spec-
troscopy using a streak camera in solvents varying in po-
larity and viscosity.
˚
ers of 1a showed a close proximity (Ͻ 5 A) of the cationic
palladium ion and the pyrenoxy unit. In similar conformer
searches of 11a, proximity between the NCN ligand (Me3Si
substituent) and the pyrenoxy moiety was not observed as
a major conformation. A representation of the structure of
the functionalized dendrimer 11b (26 kD) was also calcu-
lated by molecular mechanics, showing a densely crowded
dendrimer with an average diameter of 5.3 nm.
Binding Studies with Picric Acid
The binding affinities of 1a, 3 and the bis(pyrenyl) ether
4 towards aromatic molecules were studied by 1H NMR
titrations with picric acid (2,4,6-trinitrophenol) as guest
molecule. Picric acid is known for its ability to form π-
stacking complexes with a wide variety of aromatic mol-
ecules. Furthermore, the diagnostic 1H resonances of picric
acid are situated in an isolated region of the spectrum. The
titration experiments were performed in CDCl3 at constant
guest concentrations in the order of 0.5 KaϪ1 to obtain opti-
mal results.[11] In these experiments the upfield shift of the
aromatic proton resonance of picric acid was monitored as
a function of host/guest ratio. The titration curves were fit-
ted to obtain the association constants and complexation-
induced shift values (Table 2). The stoichiometry of the
stacking complexes was determined by Job’s method of
continuous variation for species in rapid exchange.[12] Job
plots of ∆δ·XG versus XG (∆δ: induced shift; XG: mole frac-
tion of guest) for the supramolecular complexes of picric
acid with pyrenoxy-based NCN palladium() complex 1a,
monopyrenoxy compound 3 and bis(pyrenoxy) compound
4 are shown in Figure 5. The maxima in the curves for 1a
and 3 are found at XG ϭ 0.5 revealing the formation of 1:1
complexes. The maximum of the curve in the Job plot of 4
is found at XG ϭ 0.7 indicating the formation of a 2:1 com-
plex with picric acid.
Figure 4. a) Steady-state fluorescence spectrum in hexane and b)
fluorescence decay curve of 8 (τ1 ϭ 10.2 ns, τ2 ϭ 1.8 ns) in cyclo-
hexane
The fluorescence trace of 8 in cyclohexane could be fitted
to a biexponential decay with lifetimes of τ1 ϭ 10.2 ns and
τ2 ϭ 1.8 ns attributed to the local (monomer) and the ex-
cimer emission, respectively (Figure 4, b). Formation of the
excimer from the excited monomer could not be observed,
however. In more polar benzonitrile, similar results were
obtained (τ1 ϭ 13.5 ns and τ2 ϭ 3.8 ns), albeit with a lower
relative intensity of the excimer band. Since no growth of
monomer into excimer fluorescence could be observed,
possibly as a result of the speed of excimer formation with
respect to detection limits, the solvent was changed to the
highly viscous sucrose octaacetate (SOA). Measurements in
SOA in the temperature range from 30 to 150 °C showed a
decrease in excimer emission intensity compared to the lo-
cal fluorescence at higher temperatures. Furthermore, the
fine structure in the local fluorescence band was lost. The
lifetimes of local and excimer fluorescence were found to be
in the same order as in nonviscous solvents (τ1 ϭ 20.0 ns
and τ2 ϭ 2.3 ns at 30 °C to τ1 ϭ 15.5 ns and τ2 ϭ 4.8 ns at
Table 2. Association constants and complexation induced shift val-
ues (CIS) of supramolecular complexes between picric acid and
hosts 1a, 3, and 4
Ka [Ϫ1 [a]
]
CIS [ppm]
1a
3
4
94
Ϫ0.67
Ϫ1.08
Ϫ0.64
44
81[b]
[a]
[b]
Estimated error of 10%. Ka1 ϭ Ka2 ϭ 81 Ϫ1
.
Catalysis ؊ Aldol Condensations
The cationic NCN palladium() complexes 1a, 1b and 2
110 °C) and as in the previous cases no growth of monomer can be applied as Lewis acidic catalysts. These types of cat-
into excimer fluorescence was observed. ionic palladium() salts, amongst several other late tran-
1696
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 1692Ϫ1703

Pyrenoxy-Based NCN-Pincer Palladium(II) Molecular Tweezers
FULL PAPER
Table 3. Results from aldol condensations between methyl isocy-
anoacetate and (nitro-substituted) aldehydes and cinnemaldehydes
in dichloromethane in the presence of Hünig’s base at ambient tem-
perature
Substrate
Time [h] Conversion[a] (%)
cis/trans[b]
1a
2
2 ϩ 3 1b
PhCHO
6
24
3
24
5
24
3
24
48
95
42
25
91
41
27 11
92 76
43 n.d. 1.98
1.82
4-O2NC6H4CHO
PhCHϭCHCHO
2-O2NC6H4CHϭCHCHO
Ͼ 99 Ͼ 99 Ͼ 99
50
Ͼ 99
49
34
95
38
97
30 n.d. 1.40
96
38 n.d. 1.43
96
Ͼ 99
Figure 5. Job plots for complexes between picric acid and pyrenoxy
hosts 1a, 3, and 4
[a]
[b]
Determined by 1H NMR spectroscopy.
The cis/trans ratios
were identical for catalysts 1a, 1b, 2, and 2ϩ3; conditions: 1.2 mmol
of substrate, 1 mol % of catalyst, 10 mol % of Hünig’s base, 5 mL
of CH2Cl2, room temperature.
sition metal salts, have been successfully applied as catalysts
in aldol reactions between aldehydes and methyl isocy-
anoacetate (Scheme 5).[13]
attachment point on the xylyl linker, which in this study has
been used for immobilization purposes, can be used for the
selective introduction of a third functional moiety on the
molecule.
Knowledge of the favored conformations of 1a in solu-
tion is helpful in assessing the influence of the pyrenoxy
unit on the catalytic properties of the cationic complex. The
polycyclic flat aromatic surface of a pyrenoxy moiety can
exhibit interactions with other π-systems. The πϪπ stacking
interactions observed in the crystal packing of 10 are a clear
example of this (see Figure 2). This crystal structure also
shows that the NCN-pincer moiety itself can be involved
in stacking interactions. For the catalysis experiments it is
important to know whether the πϪπ stacking found in the
solid state is also present in dilute solution. In the case of
pure 1a, 1H NMR dilution experiments showed no evidence
for significant formation or break-up of di- or oligomeric
intermolecularly assembled species at the concentrations
used in catalysis (10Ϫ3 ). Complexation experiments in-
volving 1a and 3 with picric acid show binding of the aro-
matic guest molecule to the pyrenoxy moiety by πϪπ stack-
ing interactions in a 1:1 ratio, with typical association con-
stants of 101Ϫ102 Ϫ1. The bis(pyrenoxy) compound 4 can
be expected to bind picric acid in two distinct ways. Com-
plexation of one picric acid molecule in a tweezer-type man-
Scheme 5. Aldol condensation of methyl isocyanoacetate with aro-
matic aldehydes or cinnemaldehydes
Complexes 1a and 2 were tested as catalysts in this aldol
condensation with several aromatic substrates. All aldol re-
actions were also performed with a mixture of equimolar
amounts of 2 and the pyrenyl ether 3. Furthermore, multis-
ite catalyst 1b was applied in the aldol condensation of ben-
zaldehyde to study the effect of immobilization on catalytic
performance. The results from the catalytic experiments are
collected in Table 3.
Discussion
The approach in which distinct building blocks are as- ner, interacting with both pyrenoxy moieties at the same
sembled on a multisite linker has been shown to be success- time, results in a 1:1 complex. Alternatively, picric acid mol-
ful for the synthesis of complex 1a. The strategic posi- ecules can complex to each of the pyrenoxy groups, re-
tioning of the virtually inert trimethylsilyl substituent on sulting in the formation of a 2:1 complex. The titration
Cipso of the NCN-pincer ligand allows regioselective metal- curve and Job plots clearly show that 2:1 complexation oc-
lation of the ligand at a later stage of the synthesis. Further curs, which indicates that there is no significant synergic
modifications, such as immobilization on a carbosilane den- binding of picric acid by both pyrenoxy units of 4.
drimer (see 1b) appeared to be feasible even after introduc-
The flexibility of complex 1a was probed by low-tempera-
tion of the ligand, which makes ligand 9 a versatile building ture NMR studies, revealing that rotations around the
block for the construction of complex (multimetallic) struc- benzyl ether linkages are fast on the NMR timescale, and
tures. With this in mind a synthetic route is available which that preferred conformations could not be frozen out. How-
allows the placement of various other functional groups in ever, the broadening of these benzylic proton resonances at
the vicinity of the cationic NCN-pincer palladium() site, low temperatures point at hindered rotation. This broaden-
apart from the pyrenoxy fluorescent probe in 1a. The third ing could result from an attractive interaction between the
Eur. J. Org. Chem. 2003, 1692Ϫ1703
www.eurjoc.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1697

G. van Koten et al.
FULL PAPER
NCNϪpalladium moiety and the pyrenoxy unit, such as a apart a rearrangement is necessary to form the excimer
cationϪπ interaction, bringing the substituents on the xylyl after excitation. This rearrangement process is apparently
linker into a tweezer-type arrangement. Immobilization of too slow to occur within the lifetime of the monomer fluor-
36 of these units on a G2-carbosilane dendrimer results in escence (10Ϫ20 ns). The relative intensity of the excimer
a dendrimer with a sterically congested periphery, undoubt- emission compared to the local emission, which decreases
edly influencing the flexibility and conformation of the indi- with higher solvent polarity, indicates that in less-polar sol-
vidual units. Interestingly, severe broadening of all proton vents 8 does indeed possess a more closed conformation,
signals was observed even at room temperature in the poly- with the pyrenoxy surfaces close together. The structural
functionalized dendrimers 11b and 12b. This can be attri- relationship between 8 and 1a, i.e. the tweezer-type arrange-
buted to the crowded structure of the dendrimer with sub- ment of substituents, can allow an extension of these results
stituents in various conformations, with ample πϪπ stack- to the conformation of 1a in solvents of different polarity.
ing interactions. This densely crowded periphery and the
Catalytic Behavior in Aldol Condensations
formation of various intramolecular (interbranch) πϪπ
stacking contacts was also observed in molecular-mech-
anics-optimized structures of 11b, and further supported by and aromatic aldehydes or cinnemaldehydes is especially
the quenching of fluorescence in both 11b and 12b. useful for studying the effect of modifications of a Lewis
The aldol condensation between methyl isocyanoacetate
Optical spectroscopic methods offer the possibility of acidic catalyst on its performance. The formation of a mix-
studying the flexibility and conformation of the system on ture of two pairs of diastereoisomers allows monitoring of
a smaller timescale. Close proximity between the pyrenoxy both the (stereo)selectivity and the activity of the catalyst.
unit and the cationic palladium() site, resulting from These condensations are assumed to occur via coordination
cationϪπ interactions, would influence the optical proper- of the isocyanoacetate to the metal ion upon replacement
ties of the fluorescent pyrenoxy unit in 1a compared to 11a, of the weakly coordinating water ligand, followed by depro-
which lacks the palladium cation. The NCNϪpalladium tonation of the coordinated isocyanoacetate to its corre-
complex and the pyrenoxy moiety are electronically isolated sponding enolate. The enolate subsequently reacts with the
from each other by the linker, hence the influence of the aldehyde in the rate-determining step to form the product
palladium ion on the electronic spectra can only arise from oxazolines as a mixture of the cis and trans isomers.[13e,14]
a through-space interaction either inter- or intramolecu-
Enhancements of up to two times in the initial reaction
larly. Since the fluorescence studies were performed at high rates are found for complex 1a compared to complex 2. In
dilution, we concluded that significant intermolecular inter- addition, the presence of pyrenyl ether 3 does not affect the
actions were absent. The bathochromic shifts in combi- reaction rate of the aldol reaction catalyzed by 2, ruling out
nation with the observed hypochroism in UV/Vis spectra of any independent effect of the pyrenoxy unit on the catalysis.
1a compared to 11a imply that the cationic palladium() In the case of 4-nitrobenzaldehyde (Table 3, Entry 2) no
ion is in close proximity to the pyrenoxy site. The presence influence of the pyrenoxy unit on catalytic performance is
of this interaction is further supported by fluorescence spec- observed. The electron-withdrawing nitro substituent acti-
troscopy. The fluorescence decay curves of 1a and 11a differ vates the aldehyde for nucleophilic attack of the palladium-
markedly. Not only is the relative quantum yield decreased coordinated isocyanoacetate enolate. A change in the rate-
by a factor 10 but also its lifetime is significantly reduced. determining step resulting from this substrate activation can
Interactions between the pyrenoxy π-system and the pal- make the presence or absence of the pyrenoxy unit irrel-
ladium() cation result in (partial) quenching of the pyr- evant for catalytic performance. The small, but reproduc-
enoxy fluorescence. The influence of cationϪπ interactions ible, rate enhancements observed in the aldol reactions of
on the adopted structure of 1a was further supported by the other substrates with complex 1a, may be explained by
molecular mechanics conformation searches. While for 1a several aspects of this tweezer. First, the cationϪπ interac-
all conformations showed a close proximity of the pal- tion activates the catalyst by allowing faster ligand ex-
ladium() cation and the π-system, no such preferred con- change on the metal ion, i.e. exchange of water for methyl
formations were found for the trimethylsilyl analogue 11a. isocyanoacetate. Secondly, the presence of the pyrenoxy
The NCNϪpalladium() complex and the pyrenoxy moiety unit stabilizes one or more of the transition states or inter-
connected to the xylyl linker thus preferentially adopt a mediates in the catalytic cycle. Stabilization of the pro-
tweezer-type arrangement. The behavior of bis(pyrenoxy) tonated ammonium cocatalyst (Hünig’s base) by cationϪπ
compound 8 in time-resolved fluorescence spectroscopy was interactions can be envisaged. Thirdly, complexation of the
indicative of a relation between the solvent polarity and the aromatic substrate molecules to the pyrenoxy unit by πϪπ
preferential formation of either an ‘‘open’’ or ‘‘closed’’ stacking interactions would increase their local concen-
(tweezer) conformation. While both local (monomer) and tration near the active site. The increased local concen-
excimer bands are observed in the fluorescence spectrum, tration would facilitate the rate-determining step, resulting
no interconversion of the excited monomer in the excimer in an increased reaction rate. Dendritic multisite complex
is observed. This behavior can be explained by considering 1b shows a dramatic fourfold drop in the rate in comparison
two distinct conformations. First, the pyrenoxy fragments with complex 1a. This decreased activity can be attributed
can be in close proximity, resulting in excimer fluorescence to the steric crowding, making various catalytic sites inac-
upon excitation. Alternatively, when the groups are further cessible for substrate molecules. The selectivity, e.g. the cis/
1698
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 1692Ϫ1703

Pyrenoxy-Based NCN-Pincer Palladium(II) Molecular Tweezers
FULL PAPER
ϫ 100 mL), yielding 4.23 g of a 2:1 mixture of the mono- and di-
substituted products 7 and 8 according to 1H NMR integration.
This mixture was used in the subsequent step without further puri-
fication.
trans ratio of the product oxazolines, was found to be inde-
pendent of the catalyst applied. The pyrenoxy unit does not
influence the orientation of the substrate in a (stereo)spec-
ific manner towards the coordinated enolate.
3,5-Bis[(dimethylamino)methyl]-4-(trimethylsilyl)phenyl tert-Butyldi-
methylsilyl Ether (9): nBuLi (10.4 mL, 1.6  in hexanes, 16.5 mmol)
was added dropwise over a period of 5 min to a cooled (Ϫ78 °C)
solution of 3,5-bis[(dimethylamino)methyl]phenyl tert-butyldimeth-
ylsilyl ether (5.16 g, 16 mmol) in hexane (20 mL). The reaction mix-
ture was allowed to warm to room temperature and was stirred for
an additional 4 h. The reaction mixture was cooled to 0 °C, and a
solution of trimethylsilyl trifluoromethylsulfonate (3.71 mL,
19.2 mmol) in THF (20 mL) was added dropwise over a period of
10 min and stirring was continued for 1 h at room temperature. The
volatiles were evaporated in vacuo and the residue was subjected
to flash chromatography (basic Al2O3; CH2Cl2) to afford 9 as a
yellow oil. Yield: 5.64 g (89%). 1H NMR (300 MHz, C6D6, 25 °C):
δ ϭ 7.14 (s, 2 H, ArH), 3.45 (s, 4 H, CH2N), 2.01 (s, 12 H, NMe2),
0.99 (s, 9 H, tBuSi), 0.46 (s, 9 H, SiMe3), 0.18 (s, 6 H, SiMe2) ppm.
13C NMR (75 MHz, C6D6, 25 °C): δ ϭ 156.4, 148.9, 130.5, 120.7
(Ar), 65.8 (CH2N), 45.0 (NMe2), 25.9, Ϫ0.4 (tBuSi), Ϫ3.6 (SiMe3),
Ϫ4.2 (SiMe2) ppm. C21H42N2OSi2 (394.7): calcd. C 63.90, H 10.72,
N 7.10; found C 63.70, H 10.80, N 6.98.
Conclusion
The availability of mild site-selective palladation pro-
cedures for NCN-pincer ligands, in combination with their
facile para functionalization, allowed us to use them as
building block in the construction of organometallic
tweezer 1a. The reaction sequence applied in assembling the
pyrenoxy, NCN-pincer and xylyl spacer building blocks was
flexible enough to allow immobilization of the tweezer on
a G2-carbosilane dendrimer, followed by site-selective met-
allation to produce 1b. The non-immobilized tweezer 1a is
highly flexible and can bind to aromatic substrates (picric
acid) by πϪπ stacking interactions. It preferentially adopts
a tweezer-type conformation in solution as a result of
cationϪπ interactions between the palladium() ion and the
pyrenoxy unit. The abundance of this conformation, com-
pared to a more open one with no interaction between the
pyrenoxy and the NCNϪpalladium unit, appears to be
higher in apolar solvents. The minimalistic design ap-
proach, with the building blocks simply combined on a flex-
ible linker, leads to observable effects in catalysis. The pres-
ence of the pyrenoxy unit in 1a was beneficial for its per-
formance as catalyst in the aldol reaction between methyl
isocyanoacetate and aromatic aldehydes. Complex 2, which
is devoid of a pyrenoxy unit shows lower activities in the
aldol condensations, both in the presence and the absence
of pyrenyl ether 3. It seems apparent that the close proxim-
ity of the pyrenoxy unit and the catalytic palladium() site
is responsible for the observed rate enhancements.
1-[3-({3,5-Bis[(dimethylamino)methyl]-4-(trimethylsilyl)phenyl-
oxy}methyl)-5-bromobenzyloxy]pyrene (10): NBu4F (6 mL, 1  in
THF, 6 mmol) was added to a stirred solution of 9 (2.22 g, 6 mmol)
in THF (30 mL). After 30 min, K2CO3 (5.0 g, 30 mmol), 18-crown-
6 (0.15 g, 0.6 mmol), and the above-mentioned 2:1 mixture of 7 and
8 (4.23 g, contains 5.4 mmol of 7) were added and the reaction
mixture was stirred for an additional 18 h. The volatiles were re-
moved in vacuo and subjected to flash chromatography (basic
Al2O3; CH2Cl2) resulting in a yellow solid (5.61 g). The product
mixture was redissolved in CH2Cl2 and careful precipitation of 8
with pentane followed by filtration afforded pure 10 in the mother
liquor. Evaporation of the volatiles in vacuo yielded 1.34 g 37%,
based on 7 of 10 as a white solid. 1H NMR (300 MHz, CDCl3,
3
25 °C): δ ϭ 8.51 (d, JH,H ϭ 9.2 Hz, 1 H, pyrenyl), 8.15Ϫ7.85 (m,
7 H, pyrenyl), 7.69 (s, 1 H, Ar), 7.63 (s, 1 H, Ar), 7.58 (s, 1 H, Ar),
3
7.53 (d, JH,H ϭ 8.2 Hz, 1 H, pyrenyl), 7.05 (s, 2 H, Ar), 5.37 (s, 2
H, CH2O), 5.09 (s, 2 H, CH2O), 3.51 (s, 4 H, CH2N), 2.15 (s, 12
H, NMe2), 0.37 (s, 9 H, SiMe3) ppm. 13C NMR (75 MHz, CDCl3,
25 °C): δ ϭ 158.5, 151.9, 148.3, 139.6, 139.2, 131.4, 131.3, 129.8,
129.6, 129.1, 127.0, 126.3, 125.9, 125.5, 125.3, 125.1, 125.0, 124.6,
124.3, 124.2, 124.1, 122.6, 120.9, 120.1, 114.3, 108.9 (Ar), 69.2, 68.2
(CH2O) 65.2 (CH2N), 45.0 (NMe2), 3.4 (SiMe3) ppm.
C39H43BrN2O2Si (679.8): calcd. C 68.91, H 6.38, N 4.12, Si 4.13;
found C 68.83, H 6.46, N 3.98, Si 4.23. ES-MS: calculated for
C39H43BrN2O2Si [Mϩ ϩ H] 679.2; found 680.2.
Experimental Section
General Remarks: All experiments were conducted under dry nitro-
gen using standard Schlenk techniques, when using air- and/or
water-sensitive reagents. Solvents were dried with appropriate
agents and distilled prior to use. The reagent 3,5-bis[(dimethyl-
amino)methyl]phenyl tert-butyldimethylsilyl ether was prepared
according to literature procedures.[15] Elemental analyses were per-
formed by H. Kolbe, Mikroanalytisches Laboratorium, Mülheim,
1
Germany. H and 13C NMR spectra were recorded at 298 K with
1-[3-({3,5-Bis[(dimethylamino)methyl]-4-(trimethylsilyl)phenyl-
a Varian Inova 300 or Mercury 200 spectrometer. UV/Vis spectra
were recorded with a Varian Cary I spectrophotometer.
oxy}methyl)-5-(trimethylsilyl)benzyloxy]pyrene
(11a):
tBuLi
(1.66 mL, 1.5  in pentane, 2.50 mmol) was added dropwise to a
precooled (Ϫ100 °C) solution of 10 (0.84 g, 1.25 mmol) in THF
(10 mL). The reaction mixture was stirred at Ϫ100 °C for 1 h and
trimethylsilyl chloride (0.32 mL, 2.55 mmol) was added in one por-
1-[3-Bromo-5-(bromomethyl)benzyloxy]pyrene (7): K2CO3 (10 g,
130 mmol) was added to a stirred solution of 1-pyrenol (6) (3.0 g,
14 mmol), 3,5-bis(bromomethyl)phenyl bromide (5) (5.30 g,
15 mmol) and 18-crown-6 (0.3 g, 1.3 mmol) in THF (75 mL) and tion. The reaction mixture was allowed to warm to room tempera-
the mixture was stirred at room temperature for 18 h. The volatiles
were evaporated in vacuo and the residue was extracted with di-
chloromethane (3 ϫ 50 mL). The combined dichloromethane ex-
tracts were concentrated in vacuo, yielding a brownish solid which
contained unchanged 5, the desired product 7 and the disubstituted
compound 8 in a ratio of approximately 1:2:1. This mixture was
extracted with pentane until all unchanged 5 was removed (ca. 10
ture and all volatiles were evaporated in vacuo. After purification
by column chromatography (basic Al2O3; hexane/Et2O, 3:1, 3%
TEA) pure 11a was obtained as an off-white solid (0.81 g, 96%).
3
1H NMR (300 MHz, CDCl3, 25 °C): δ ϭ 8.53 (d, JH,H ϭ 9.2 Hz,
1 H, pyrenyl), 8.11Ϫ8.01 (m, 4 H, pyrenyl), 7.96Ϫ7.84 (m, 3 H,
pyrenyl), 7.70 (s, 1 H, Ar), 7.69 (s, 1 H, Ar), 7.62 (s, 1 H, Ar), 7.56
3
(d, JH,H ϭ 8.3 Hz, 1 H, pyrenyl), 7.08 (s, 2 H, Ar), 5.39 (s, 2 H,
Eur. J. Org. Chem. 2003, 1692Ϫ1703
www.eurjoc.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1699

G. van Koten et al.
FULL PAPER
CH2O), 5.14 (s, 2 H, CH2O), 3.51 (s, 4 H, CH2N), 2.14 (s, 12 H,
the residue was extracted with CH2Cl2 (3 ϫ 10 mL). The combined
NMe2), 0.38 (s, 9 H, SiMe3), 0.33 (s, 9 H, SiMe3) ppm. 13C NMR CH2Cl2 extracts were filtered through Celite, and evaporation of
(75 MHz, CDCl3, 25 °C): δ ϭ 159.0, 152.8, 148.4, 141.3, 136.8, the solvent afforded 60 mg (2.09 µmol, 80%) of 12b as a yellow
136.6, 132.3, 132.0, 131.6, 129.6, 127.4, 127.2, 126.4, 126.0, 125.8,
125.5, 125.4, 125.1, 124.9, 124.3, 124.2, 121.3, 120.6, 114.5, 109.6 8.20Ϫ7.65 (m, 216 H, pyrenyl), 7.65Ϫ7.20 (m, 180 H, pyrenyl), 6.32
(Ar), 71.0 (CH2O), 69.6 (CH2O), 65.4 (CH2N), 45.1 (NMe2), 3.4 (br. s, 72 H, Ar), 5.21 (s, 72 H, CH2O), 4.83 (s, 72 H, CH2O),
(SiMe3), Ϫ1.1 (SiMe3) ppm. C42H52N2O2Si2 (673.1): calcd. C 3.65 (s, 144 H, CH2N), 2.77 (s, 432 H, NMe2), 1.50Ϫ1.20 (br. m,
solid. 1H NMR (300 MHz, CDCl3, 25 °C): δ ϭ 8.37 (br., 36 H),
74.95, H 7.79, N 4.16, Si 8.35; found C 74.86, H 7.88, N 4.20,
Si, 8.26.
dendrimer), 1.00Ϫ0.75 (br. m, dendrimer) 0.75Ϫ0.45 (br. m, den-
drimer) 0.45Ϫ0.00 (br. m, dendrimer) 0.08 (s, 324 H, SiMe3) ppm.
13C NMR (75 MHz, CDCl3, 25 °C): δ ϭ 157.4, 152.8, 145.7, 140.8,
137.0, 136.8, 132.4, 132.2, 131.8, 130.6, 127.4, 126.7, 126.5, 126.0,
125.7, 125.4, 125.0, 124.7, 124.5, 121.3, 120.7, 109.7, 107.2 (Ar),
74.8 (CH2N), 71.1 (CH2O), 70.8 (CH2O), 53.2 (NMe2), 23.5, 23.1,
21.8, 21.5, 20.8, 18.9, 18.3, 17.7, 1.3, Ϫ2.5 ppm.
C1524H1752Cl36N72O72Pd36Si53 (28826.3): calcd. C 63.50, H 6.13, N
3.50; found C 63.10, H 6.34, N 3.38.
G2-Carbosilane Dendrimer Ligand 11b: tBuLi (0.57 mL, 1.5  in
pentane, 0.85 mmol) was added dropwise to a precooled (Ϫ100 °C)
solution of 10 (0.30 g, 0.45 mmol) in THF (10 mL). The reaction
mixture was stirred at Ϫ100 °C for 1 h and a solution of G2-(Si-
Me2Cl)36 (68 mg, 11.2 µmol) in THF (5 mL) was added in one por-
tion. The reaction mixture was allowed to warm to room tempera-
ture, a few drops of water were added and all volatiles were evapo-
rated in vacuo. The solid was dissolved in dichloromethane (20 mL)
and washed with water (2 ϫ 20 mL). The solution was dried with
MgSO4 and purified by dialysis against neat dichloromethane.
Dendrimer 11b was obtained as a yellowish solid (0.30 g) in 96%
(Aqua)(2,6-bis[(dimethylamino)methyl]-4-{3-[(1-pyrenyloxy)methyl]-
5-(trimethylsilyl)benzyloxy}phenyl)palladium(II
)
Tetrafluoroborate
(1a): AgBF4 (45.4 mg, 0.23 mmol) was added to a solution of 12a
(0.16 g, 0.22 mmol) in wet acetone, resulting in the immediate for-
mation of a white precipitate. The reaction mixture was stirred at
room temperature for 1 h and filtered through Celite. Evaporation
of the acetone in vacuo afforded pure 1a as an off-white solid
(0.18 g, Ͼ 99%). 1H NMR (200 MHz, CDCl3, 25 °C): δ ϭ 8.50 (d,
3JH,H ϭ 9.2 Hz, 1 H, pyrenyl), 8.16Ϫ7.85 (m, 7 H, pyrenyl), 7.69
1
yield. H NMR (300 MHz, CDCl3, 25 °C): δ ϭ 8.50Ϫ8.30 (br., 36
H), 8.20Ϫ7.70 (br. m, 216 H) 7.70Ϫ7.20 (br. m, 180 H), 7.07 (br.
s, 72 H, Ar), 5.30 (br. s, 72 H, CH2O), 5.10 (br., 72 H, CH2O), 3.49
(br. s, 144 H, CH2N), 2.11 (s, 432 H, NMe2), 1.50Ϫ1.20 (br. m,
dendrimer), 1.00Ϫ0.75 (br. m, dendrimer) 0.75Ϫ0.45 (br. m, den-
drimer) 0.45Ϫ0.00 (br. m, dendrimer) 0.12 (s, 324 H, SiMe3) ppm.
13C NMR (75 MHz, CDCl3, 25 °C): δ ϭ 159.4, 153.0, 148.1, 140.8,
136.9, 136.8, 132.8, 132.3, 131.9, 130.0, 127.8, 127.4, 126.6, 126.3,
126.0, 125.6, 125.3, 125.1, 124.4, 121.6, 120.7, 114.9, 109.7 (Ar),
71.1 (CH2O), 70.9 (CH2O), 65.4 (CH2N), 45.2 (NMe2), 23.5, 23.1,
21.8, 21.5, 20.8, 18.9, 18.3, 17.7, 3.7 (SiMe3), 1.3, Ϫ2.5 ppm.
3
(s, 1 H, Ar), 7.64 (s, 1 H, Ar), 7.62 (d, JH,H ϭ 8.3 Hz, 1 H, pyr-
enyl), 7.53 (s, 1 H, Ar), 6.46 (s, 2 H, Ar), 5.46 (s, 2 H, CH2O), 5.01
(s, 2 H, CH2O), 3.87 (s, 4 H, CH2N), 2.85 (s, 12 H, NMe2), 0.31
(s, 9 H, SiMe3) ppm. 13C NMR (50 MHz, CDCl3, 25 °C): δ ϭ
157.6, 152.6, 145.2, 143.6, 141.3, 136.6, 136.5, 132.1, 132.0, 131.6,
131.5, 127.2, 127.1, 126.3, 126.0, 125.7, 125.4, 125.3, 125.0, 124.7,
124.2, 124.1, 121.2, 120.5, 109.7, 107.5 (Ar), 73.8 (CH2N), 70.9,
C1632H2076N72O72Si89 (26354.4): calcd. C 74.38, H 7.94, N 3.83, Si
9.48; found C 74.49, H 8.06, N 3.75, Si 9.39.
70.5
(CH2O),
52.7
(NMe2),
Ϫ1.1
(SiMe3)
ppm.
(2,6-Bis[(dimethylamino)methyl]-4-{3-[(1-pyrenyloxy)methyl]-5-
C39H45BF4N2O3PdSi (811.1): calcd. C 57.75, H 5.59, N 3.45, Si
3.46; found C 57.71, H 5.62, N 3.35, Si 3.53. FAB-MS: calculated
for C39H45BF4N2O3PdSi [Mϩ Ϫ BF4 Ϫ H2O] 705.2; found 705.1.
(trimethylsilyl)benzyloxy}phenyl)palladium(II
)
Chloride
(12a):
Pd(OAc)2 (52 mg, 0.23 mmol) was added to a suspension of 11a
(0.15 g, 0.22 mmol) in freshly distilled methanol (5 mL). The reac-
tion mixture was stirred for 3 h at room temperature after which
an excess of LiCl (42 mg, 1 mmol) was added and stirring was con-
tinued for 1 h. After evaporation of all volatiles in vacuo, the resi-
due was extracted with CH2Cl2 (3 ϫ 10 mL). The combined
CH2Cl2 extracts were filtered through Celite, and evaporation of
the solvent afforded 0.16 g (99%) of pure 12a as an off-white solid.
G2-Carbosilane Dendrimer (Aqua)palladium Complex 1b: The den-
dritic (aqua)palladium complex 1b was prepared in situ for use in
the catalytic experiments by treatment of 12b (6.12 mg, 0.22 µmol)
with an amount of AgBF4 (1.56 mg, 8 µmol) equimolar with re-
spect to the amount of palladium chloride sites. The complex was
not isolated.
3
1H NMR (300 MHz, CDCl3, 25 °C): δ ϭ 8.51 (d, JH,H ϭ 9.2 Hz,
1 H, pyrenyl), 8.14Ϫ8.04 (m, 4 H, pyrenyl), 8.00Ϫ7.85 (m, 3 H,
5-(3-Bromobenzyloxy)-1,3-bis[(dimethylamino)methyl]-2-(trimethyl-
silyl)benzene (13): Bu4NF (1  in THF, 1.6 mL, 1.6 mmol) was ad-
ded in a single portion to a stirred solution of 9 (0.64 g, 1.6 mmol)
in 15 mL of THF. After 30 min, K2CO3 (2.0 g, 14.5 mmol), 18-
crown-6 (50 mg, 0.2 mmol) and 3-bromobenzyl bromide (0.40 g,
1.6 mmol) were added and the reaction mixture was stirred for an
additional 18 h. The volatiles were evaporated in vacuo and the
residue was extracted with hexane (3 ϫ 20 mL). Purification by
chromatography (basic Al2O3; hexane/Et2O, 3:1, 3% TEA) afforded
13 as a colorless oil (0.41 g, 60%). 1H NMR (200 MHz, CDCl3, 25
3
pyrenyl), 7.70 (s, 1 H, Ar), 7.65 (s, 1 H, Ar), 7.63 (d, JH,H
ϭ
8.3 Hz, 1 H, pyrenyl), 7.56 (s, 1 H, Ar), 6.44 (s, 2 H, Ar), 5.41 (s,
2 H, CH2O), 5.00 (s, 2 H, CH2O), 3.82 (s, 4 H, CH2N), 2.87 (s, 12
H, NMe2), 0.34 (s, 9 H, SiMe3) ppm. 13C NMR (75 MHz, CDCl3,
25 °C): δ ϭ 157.0, 152.5, 145.2, 141.2, 138.6, 136.6, 136.5, 132.0,
131.9, 131.5, 131.4, 127.1, 127.0, 126.3, 126.0, 125.6, 125.3, 125.2,
124.9, 124.6, 124.2, 124.0, 121.0, 120.3, 113.3, 109.4, 106.9 (Ar),
74.4 (CH2N), 70.7 (CH2O), 70.4 (CH2O), 52.8 (NMe2), Ϫ1.2
(SiMe3) ppm. C39H43ClN2O2PdSi (741.7): calcd. C 63.15, H 4.78,
N 3.78; found C 62.97, H 4.90, N 3.72. ES-MS: calculated for
C39H43ClN2O2PdSi [Mϩ Ϫ Cl] 705.2; found 705.3.
3
°C): δ ϭ 7.63 (s, 1 H, Ar), 7.44 (d, JH,H ϭ 8.3 Hz, 1 H), 7.36 (d,
3JH,H ϭ 7.3 Hz, 1 H), 7.23 (dd, 3JH,H ϭ 8.3, 3JH,H ϭ 7.3 Hz, 1 H),
7.02 (s, 2 H, Ar), 5.05 (s, 2 H, CH2O), 3.50 (s, 4 H, CH2N), 2.14
(s, 12 H, NMe2), 0.37 (s, 9 H, SiMe3) ppm. 13C NMR (50 MHz,
G2-Carbosilane Dendrimer Palladium Chloride Complex 12b: A
solution of Pd(OAc)2 (26.4 mg, 117 µmol) in methanol (4 mL) was
added to a solution of 11b (70 mg, 2.61 µmol) in dichloromethane CDCl3, 25 °C): δ ϭ 158.6, 148.4, 139.6, 130.8, 130.4, 130.0, 129.9,
(10 mL). The reaction mixture was stirred for 3 h at room tempera-
ture, an excess of LiCl (20 mg, 0.5 mmol) was added and stirring
was continued for 1 h. After evaporation of all volatiles in vacuo,
125.9, 122.5, 114.3 (Ar), 68.6 (CH2O), 65.3 (CH2N), 45.1 (NMe2),
3.5 (SiMe3) ppm. C22H33BrN2OSi (449.5): calcd. C 58.78, H 7.40,
N 6.23, Si 6.25; found C 59.02, H 7.34, N 6.21, Si 6.33.
1700
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 1692Ϫ1703

Pyrenoxy-Based NCN-Pincer Palladium(II) Molecular Tweezers
1,3-Bis[(dimethylamino)methyl]-2-(trimethylsilyl)-5-[3-(trimethyl- pyrenyl), 7.30 (dd, 1 H, Ar), 5.40 (s, 2 H, CH2O) ppm. 13C NMR
FULL PAPER
silyl)benzyloxy]benzene (14): tBuLi (1.21 mL, 1.5  in pentane,
1.82 mmol) was added dropwise to a precooled (Ϫ100 °C) solution
(50 MHz, CDCl3, 25 °C): δ ϭ 152.0, 139.2, 131.5, 131.4, 130.8,
130.0, 129.9, 127.2, 127.0, 126.4, 126.0, 125.5, 125.5, 125.1, 125.0,
of 13 (0.41 g, 0.91 mmol) in diethyl ether (15 mL), and the reaction 124.6, 124.2, 124.1, 122.5, 120.9, 120.2, 109.1 (Ar), 69.4 (CH2O)
mixture was stirred for 1 h at Ϫ100 °C. Trimethylsilyl chloride
ppm. C23H15BrO (387.3): calcd. C 71.33, H 3.90; found C 71.39,
(0.23 mL, 1.82 mmol) was added dropwise to the yellow solution H 3.92. ES-MS: calculated for C23H15BrO [Mϩ] 386.0; found 386.6.
and the reaction mixture was allowed to warm to room tempera-
1-[3-(Trimethylsilyl)benzyloxy]pyrene (3): tBuLi (1.5  in pentane,
1.65 mL, 2.48 mmol) was added dropwise to a precooled (Ϫ100 °C)
solution of 16 (0.48 g, 1.24 mmol) in THF (15 mL) and the reaction
mixture was stirred for 1 h at Ϫ100 °C. Trimethylsilyl chloride
(0.30 mL, 2.50 mmol) was added dropwise to the dark-red solution
and the reaction mixture was allowed to warm to room tempera-
ture. The volatiles were evaporated in vacuo and the residue was
extracted with dichloromethane and filtered through Celite. Purifi-
cation by chromatography (basic Al2O3; hexane, 2% TEA and
Et2O, 2% TEA) afforded 0.22 g (55%) of pure 14 as a colorless oil.
1H NMR (200 MHz, CDCl3, 25 °C): δ ϭ 7.64 (s, 1 H, Ar),
7.57Ϫ7.37 (m, 3 H, Ar),7.09 (s, 2 H, Ar), 5.13 (s, 2 H, CH2O), 3.55
ture. The volatiles were evaporated in vacuo and the residue was
(s, 4 H, CH2N), 2.18 (s, 12 H, NMe2), 0.42 (s, 9 H, SiMe3), 0.33
extracted with dichloromethane and filtered through Celite. Evap-
oration of the dichloromethane in vacuo afforded 0.44 g (93%) of
(s, 9 H, SiMe3) ppm. 13C NMR (50 MHz, CDCl3, 25 °C): δ ϭ
159.0, 148.3, 140.6, 136.2, 132.8, 132.6, 129.5, 128.2, 127.8, 114.5
3 as a yellow oil. 1H NMR (200 MHz, CDCl3, 25 °C): δ ϭ 8.52 (d,
(Ar), 69.8 (CH2O), 65.4 (CH2N), 45.5 (NMe2), 3.4 (SiMe3), Ϫ1.1
3JH,H ϭ 9.2 Hz, 1 H, pyrenyl), 8.08Ϫ7.80 (m, 7 H, pyrenyl, Ar),
(SiMe3) ppm. C25H42N2OSi2 (442.8): calcd. C 67.81, H 9.56, N
6.33, Si 12.69; found C 67.75, H 9.49, N 6.28, Si 12.48.
7.74 (s, 1 H, pyrenyl) 7.58Ϫ7.38 (m, 4 H, Ar), 5.33 (s, 2 H, CH2O),
0.33 (s, 9 H, SiMe3) ppm. 13C (50 MHz, CDCl3, 25 °C): δ ϭ 152.6,
{2,6-Bis[(dimethylamino)methyl]-4-[3-(trimethylsilyl)benzyloxy]-
phenyl}palladium(II) Chloride (15): Pd(OAc)2 (0.12 g, 0.52 mmol)
was added to a suspension of 14 (0.22 g, 0.5 mmol) in freshly dis-
tilled methanol. The reaction mixture was stirred for 45 min, during
which a yellow precipitate formed. LiCl (0.21 g, 5.0 mmol) was ad-
ded to the reaction mixture in a single portion and stirring was
continued for an additional 30 min. Evaporation of the volatiles in
vacuo followed by extraction with dichloromethane (3 ϫ 50 mL)
and filtration through Celite afforded 15 as a yellow solid in 90%
yield (0.23 g). 1H NMR (200 MHz, CDCl3, 25 °C): δ ϭ 7.54Ϫ7.32
(m, 4 H, Ar), 6.50 (s, 2 H, Ar), 4.96 (s, 2 H, CH2O), 3.97 (s, 4 H,
CH2N), 2.92 (s, 12 H, NMe2), 0.27 (s, 9 H, SiMe3) ppm. 13C NMR
(50 MHz, CDCl3, 25 °C): δ ϭ 156.9, 146.5, 145.1, 140.4, 135.8,
132.6, 132.1, 127.9, 127.8, 127.6, 106.7 (Ar), 74.3 (CH2N), 70.4
(CH2O), 52.7 (NMe2), Ϫ1.4 (SiMe3) ppm. C22H33ClN2OPdSi
(511.5): calcd. C 51.66, H 6.50, N 5.48; found C 51.70, H 6.57, N
5.52. ES-MS: calculated for C22H33ClN2OPdSi [Mϩ Ϫ Cl] 475.1;
found 474.8.
140.7, 132.9, 132.3, 131.6, 131.5, 130.9, 128.0, 127.9, 127.7, 127.1,
126.3, 125.9, 125.7, 125.3, 124.9, 124.8, 124.1, 124.0, 121.2, 120.4,
109.3 (Ar), 70.9 (CH2O) ppm. C26H24OSi (380.6): calcd. C 82.06,
H 6.36, Si 7.38; found C 81.89, H 6.38, Si 7.26. ES-MS: calculated
for C26H24OSi [Mϩ] 380.2; found 380.7.
1,3-Bis[(1-pyrenyloxy)methyl]-5-(trimethylsilyl)benzene (4): tBuLi
(0.41 mL, 1.5  in pentane, 0.62 mmol) was added dropwise to a
precooled (Ϫ100 °C) solution of 8 (0.20 g, 0.32 mmol) in THF
(5 mL) and the reaction mixture was stirred at Ϫ100 °C for 30 min.
Excess trimethylsilyl chloride (0.30 mL, 2.50 mmol) was added
dropwise to the dark-red solution and the reaction mixture was
allowed to warm to room temperature. The volatiles were evapo-
rated in vacuo and the residue was extracted with dichloromethane
and filtered through Celite. Evaporation of the solvent in vacuo
afforded 0.18 g (92%) of 4 as a yellowish solid. 1H NMR
3
(300 MHz, CDCl3, 25 °C): δ ϭ 8.58 (d, JH,H ϭ 9.6 Hz, 2 H, pyr-
enyl), 8.25Ϫ7.91 (m, 14 H, pyrenyl), 7.89 (s, 1 H, Ar), 7.83 (s, 2 H,
3
Ar), 7.59 (d, JH,H ϭ 8.1 Hz, 1 H, pyrenyl), 5.47 (s, 4 H, CH2O),
0.48 (s, 9 H, SiMe3) ppm. 13C (75 MHz, CDCl3, 25 °C): δ ϭ 141.8,
138.0, 132.3, 132.0, 131.9, 127.7, 127.5, 127.4, 126.8, 126.4, 126.1,
125.8, 125.7, 125.4, 125.2, 124.6, 124.5, 121.6, 121.0, 109.9 (Ar),
71.2 (CH2O), Ϫ0.7 (SiMe3) ppm. C436H34O2Si (5331.1): calcd. C
84.55, H 5.61; found 84.67, H 5.70. ES-MS: calculated for
C43H34O2Si [Mϩ ϩ H] 611.2; found 611.1.
(Aqua){2,6-Bis[(dimethylamino)methyl]-4-[3-(trimethylsilyl)benzyl-
oxy]phenyl}palladium(II) Tetrafluoroborate (2): AgBF4 (42 mg,
0.21 mmol) was added to a solution of 15 (0.10 g, 0.20 mmol) in
wet acetone (5 mL). A white precipitate formed instantaneously.
The reaction mixture was stirred at room temperature for 1 h and
filtered through Celite. Evaporation of the acetone in vacuo af-
forded 0.12 g (Ͼ 99%) of pure 2 as a yellow solid. 1H NMR
(200 MHz, CDCl3, 25 °C): δ ϭ 7.55Ϫ7.36 (m, 4 H, Ar), 6.51 (s, 2
H, Ar), 4.96 (s, 2 H, CH2O), 3.98 (s, 4 H, CH2N), 2.92 (s, 12 H,
NMe2), 0.27 (s, 9 H, SiMe3) ppm. 13C NMR (50 MHz, CDCl3, 25
°C): δ ϭ 157.9, 149.5, 145.2, 140.9, 135.9, 133.0, 132.4, 128.1,
127.9, 127.8, 107.7 (Ar), 73.7 (CH2N), 70.7 (CH2O), 52.7 (NMe2),
Ϫ1.1 (SiMe3) ppm. C22H35BF4N2O2PdSi (580.8): calcd. C 45.49,
H 6.07, N 4.82; found 45.38, H 6.16, N 4.75.
Catalyst Screening: The (substituted) benzaldehyde or cinnamal-
dehyde (0.8 mmol), (diisopropyl)ethylamine (10 mol %, 80 µmol)
and methyl isocyanoacetate (0.8 mmol) were added to a solution
of the catalyst (1 mol %, 8 µmol) in 2.5 mL of CH2Cl2. The mixture
was stirred in a closed reaction vessel under ambient conditions.
1
Samples (100 µl) for H NMR (200 MHz, CDCl3, 25 °C) analysis
concentrated in a stream of nitrogen gas, and redissolved in CDCl3
(0.5 mL). Conversions (methyl isocyanoacetate consumptions) were
calculated from the intensity of the CNCH2COOMe protons versus
the combined methyl ester signals.
1-(3-Bromobenzyloxy)pyrene (16): K2CO3 (3.45 g, 25 mmol) was
added to a stirred solution of 1-pyrenol (6) (1.09 g, 5 mmol), 3-
bromobenzyl bromide (1.22 g, 4.9 mmol) and 18-crown-6 (0.13 g,
0.5 mmol) in THF (30 mL), and the mixture was stirred at room Fluorescence Spectroscopy: Steady-state fluorescence measurements
temperature for 36 h. The volatiles were evaporated in vacuo and
were carried out with a SPEX Fluorolog 3-22 using a 450-W Xe
the residue was extracted with dichloromethane (3 ϫ 50 mL). The
excitation source and a Peltier-cooled R636Ϫ10 photomultiplier by
combined dichloromethane extracts were concentrated in vacuo. Hamamatsu, operating in photon-counting mode. For time-re-
Purification by flash chromatography (basic Al2O3; CH2Cl2) af-
solved emission measurements we used a Chromex IS250 spectro-
forded 1.70 g (90%) of pure 16 as a yellow oil. 1H NMR (200 MHz, graph coupled to a Hamamatsu streak camera (C5680-21 with
3
CDCl3, 25 °C): δ ϭ 8.51 (d, JH,H ϭ 9.2 Hz, 1 H, pyrenyl), M5677 Low-Speed Single-Sweep Unit and C4742-95 CCD cam-
8.12Ϫ7.87 (m, 7 H, pyrenyl), 7.77 (s, 1 H, Ar), 7.57Ϫ7.47 (m, 3 H,
www.eurjoc.org
era). For excitation an LTB MSG400 nitrogen laser (337 nm,
1701
Eur. J. Org. Chem. 2003, 1692Ϫ1703
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

G. van Koten et al.
FULL PAPER
Rietveld, D. M. Grove, G. van Koten, New J. Chem. 1997, 21,
fwhm ഠ 0.6 ns) was used. Lifetimes were determined from monoex-
ponential (Figure 3) or biexponential (Figure 4) fits to the decay
curves in the time domain beyond the laser pulse.
[1e]
751.
G. van Koten, Pure Appl. Chem. 1989, 61, 1681.
[2] [2a]
´
G. Rodrıguez, M. Albrecht, J. Schoenmaker, A. Ford, M.
Lutz, A. L. Spek, G. van Koten, J. Am. Chem. Soc. 2002, 124,
5127Ϫ5138. [2b] M. Q. Slagt, R. J. M. Klein Gebbink, M. Lutz,
A. L. Spek, G. van Koten, J. Chem. Soc., Dalton Trans. 2002,
252591Ϫ252592.
Computational Details: Results were obtained utilizing the multiple
conformer search as implemented in the SPARTAN 5.1.1 (UNIX)
package, with a MMFF94 mechanics force field.[16] The input data
used in the conformer searches of both 1 and 11 were imported into
the program by using the builder routine in the molecular modeling
package followed by a geometry optimization with the MMFF94
mechanics force field. The square-planar geometry of the pal-
ladium complexes, known from single-crystal X-ray structures and
DFT calculations, is well reproduced by the mechanics force field.
Special treatment of the palladium atom in the calculations was
therefore omitted. Conformer searches were performed using
Monte Carlo and systematic stepwise rotation around the six bonds
connecting the NCN ligand and the pyrenoxy unit to the linker,
both leading to similar results.
[3]
For overviews on catalytically active metallodendritic com-
[3a]
plexes, see:
R. Heerbeek, P. C. J. Kamer, P. W. N. M. van
Leeuwen, J. N. H. Reek, Chem. Rev. 2002, 102, 3717Ϫ3756. [3b]
R. Kreiter, A. W. Kleij, R. J. M. Klein Gebbink, G. van Koten,
Top. Curr. Chem. 2001, 217, 163Ϫ199.
[3c]
G. E. Oosterom, J.
H. N. Reek, P. C. J. Kamer, P. W. N. M. van Leeuwen, Angew.
Chem. Int. Ed. 2001, 40, 1828Ϫ1849. [3d] D. Astruc, F. Chardac,
Chem. Rev. 2001, 101, 2991Ϫ2023.
[4]
´
G. Rodrıguez, M. Lutz, A. L. Spek, G. van Koten, Chem. Eur.
J. 2002, 8, 45Ϫ57.
[5] [5a]
H. P. Dijkstra, G. P. M. van Klink, G. van Koten, Acc.
Chem. Res. 2002, 35, 798Ϫ810. [5b] A. W. Kleij, R. A. Gossage,
J. T. B. H. Jastrzebski, J. Boersma, G. van Koten, Angew. Chem.
2000, 39, 176Ϫ178. [5c] P. Wijkens, J. T. B. H. Jastrzebski, P. A.
X-ray Crystallographic Study: Crystal data and data collection for
10. Empirical formula C39H43BrN2O2Si·CH2Cl2, Mr ϭ 764.68, tri-
van der Schaaf, R. Kolly, A. Hafner, G. van Koten, Org. Lett.
[5d]
¯
2000, 11, 1621Ϫ1624.
U. Kragl, C. Dreisbach, Angew.
clinic, space group P1 (no. 2), a ϭ 9.9388(2), b ϭ 13.2651(3), c ϭ
Chem. Int. Ed. Engl. 1996, 35, 342. [5e] U. Kragl, C. Dreisbach,
C. Wandrey, ‘‘Applied Homogeneous Catalysis with Or-
ganometallic Compounds’’ in Membrane Reactors in Homo-
geneous Catalysis (Eds.: B. Cornils, W. A. Herrmann), Wiley-
VCH, Weinheim, 1996, p. 833Ϫ843.
˚
14.7898 A, α ϭ 84.512(1)°, β ϭ 80.614(1)° γ ϭ 74.891(1)°, V ϭ
3
1854.38(7) A , F(000) ϭ 796, Z ϭ 2, ρcalcd. ϭ 1.370 Mg/m3, µ(Mo-
˚
Kα) ϭ 1.322 mmϪ1, 2θmax ϭ 55.0°. 19551 reflections measured,
8432 independent, Rint ϭ 0.0349. Refinement converged at a wR2
value of 0.1054 (S ϭ 1.023), R ϭ 0.0408 [for 7064 reflections with
[6] [6a]
J.-M. Valk, R. van Belzen, J. Boersma, A. L. Spek, G. van
Koten, J. Chem. Soc., Dalton Trans. 1994, 2293. [6b] J.-M. Valk,
J. Boersma, G. van Koten, J. Orgmet. Chem. 1994, 483, 213.
3
˚
Fo Ͼ 4σ(Fo)]. Maximum residual electron density 0.44 e/A . A pale-
yellow plate crystal (0.36 ϫ 0.21 ϫ 0.09 mm) was mounted under
a nitrogen stream (150 K) on a Nonius Kappa CCD area detector
[7] [7a]
P. Steenwinkel, R. A. Gossage, G. van Koten, Chem. Eur.
J. 1998, 4, 759Ϫ762. [7b] P. Steenwinkel, J. T. B. H. Jastrzebski,
B.-J. Deelman, D. M. Grove, H. Kooijman, N. Veldman, W. J.
J. Smeets, A. L. Spek, G. van Koten, Organometallics 1997,
16, 5486Ϫ5498.
˚
diffractometer with a graphite monochromator (λ ϭ 0.71073 A),
and a rotating anode source. Lorentz, polarization, and absorption
corrections (the latter using multiple measured equivalent reflec-
tions) were applied (PLATON/MULABS[17a]). The structure was
solved with the DIRDIF97[17c] program and refined with
SHELXL-97[17b] using full-matrix least squares on F2. All non-hy-
drogen atoms were refined with anisotropic displacement param-
eters, hydrogen atoms were located at calculated positions and re-
fined riding on their carrier atoms. The dichloromethane molecule
is equally disordered across two positions. CCDC-165285 contains
the supplementary crystallographic data for this paper. These data
can be obtained free of charge at www.ccdc.cam.ac.uk/conts/
retrieving.html [or from the Cambridge Crystallographic Data
Centre, 12 Union Road, Cambridge CB2 1EZ, UK; Fax: (internat.)
ϩ 44-1223/336-033; E-mail: deposit@ccdc.cam.ac.uk].
[8]
Overviews and studies involving hypervalent/pentacoordinate
[8a]
silicon atoms:
J. M. Anglada, C. Bo, J. M. Bofill, R. Cre-
[8b]
huet, J. M. Poblet, Organometallics 1999, 18, 5584Ϫ5593.
M. Weinmann, A. Gehring, B. Schiemenz, G. Huttner, B.
Nuber, G. Rheinwald, H. Lang, J. Orgmet. Chem. 1998, 563,
[8c]
61Ϫ72.
W. Ziche, B. Ziemer, P. John, J. Weis, N. Auner, J.
[8d]
Orgmet. Chem. 1997, 531, 245.
R. R. Holmz, Chem. Rev.
1996, 96, 927Ϫ950. [8e] C. Chuit, R. J. P. Corriu, C. Reye, J. C.
[8f]
Young, Chem. Rev. 1993, 93, 1371Ϫ1448.
H. Handwerker,
C. Leis, R. Probst, P. Bissenger, A. Grohmann, P. Kiprof, E.
Herdtureck, J. Blümel, N. Auner, C. Zybill, Organometallics
1993, 12, 2162.
[9]
Good FFCE (face-to-face, center-to-edge) stacking is present
in cases where the stacking parameters fulfil the following re-
˚
˚
˚
quirements: α Ͻ 45°, 1.4 A Ͻ x Ͻ 2.7 A and 3.3 A Ͻ z Ͻ 4.0
˚
A (α ϭ angle between planes, x ϭ horizontal displacement,
z ϭ vertical displacement). See also: C. B. Martin, H. R. Mulla,
P. G. Willis, A. Cammers-Goodwin, J. Org. Chem. 1999, 64,
7802Ϫ7806.
For fluorescence and adsorption spectra of 1a and 11a, see
Supporting Information (see also the footnote on the first page
of this article).
Acknowledgments
The authors wish to thank Mr. C. Versluis from the Analytical
Chemical Department (Utrecht University) for measuring the
FAB- and ES-MS spectra. This work was supported by the Council
for Chemical Sciences of The Netherlands’ Organisation for Scien-
tific Research (CW/NWO), the Dutch Technology Foundation
(STW), and the Netherlands Research School Combination Cataly-
sis (NRSC-C).
[10]
[11]
[12]
J. Granot, J. Magn. Res. 1983, 55, 216.
K. A. Connors, Binding Constants, The Measurement of Molec-
ular Complex Stability, Wiley-Interscience, New York, 1987,
p. 24Ϫ28.
[13]
[13a]
Examples of PdII- and PtII-catalyzed aldol condensations:
H. P. Dijkstra, M. D. Meijer, J. Patel, R. Kreiter, G. P. M.
[1]
[1a]
For overviews on NCN-pincer chemistry see:
M. Albrecht,
van Klink, M. Lutz, A. L. Spek, A. J. Canty, G. van Koten,
G. van Koten, Angew. Chem. Int. Ed. 2001, 40, 3750Ϫ3781. [1b]
P. Steenwinkel, R. A. Gossage, G. van Koten, Chem. Eur. J.
R. A. Gossage, L. A. van de Kuil, G.
van Koten, Acc. Chem. Res. 1998, 31, 423Ϫ431.
Organometallics 2001, 20, 3159Ϫ3168.
M. A. Stark, C. J.
[13b]
[13c]
Richards, Tetrahedron Lett. 1997, 38, 5881Ϫ5884.
J. M.
[1c]
1998, 4, 759Ϫ763.
Longmire, X. Zhang, Organometallics 1998, 17, 4374Ϫ4379.
[1d]
[13d]
M. H. P.
O. Fujimura, J. Am. Chem. Soc. 1998, 120, 10032Ϫ10039.
1702
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 1692Ϫ1703

Pyrenoxy-Based NCN-Pincer Palladium(II) Molecular Tweezers
FULL PAPER
SPARTAN SGI, version 5.1.1, Wavefunction, Inc., 18401 Von
[13e]
[16]
F. Gorla, A. Togni, L. M. Venanzi, A. Albinati, F. Lianza,
Organometallics 1994, 13, 1607Ϫ1616. [13f] R. Nesper, P. S. Pre-
gosin, K. Püntener, M. Wörle, Helv. Chim. Acta 1993, 76, 2239;
see also ref.[4]
Karman Ave., Ste. 370, Irvine, CA 92612, USA.
[17] [17a]
A. L. Spek, PLATON, A multipurpose crystallographic
[17b]
tool, Utrecht University, The Netherlands, 2000.
G. M.
[14] [14a]
T. Hayashi, M. Sawamura, Y. Ito, Tetrahedron 1992, 48,
Sheldrick, SHELXL-97, Program for crystal structure
[14b]
[17c]
1999Ϫ2012.
V. A. Soloshonok, T. Hayashi, Tetrahedron
refinement, University of Göttingen, Germany, 1997.
P. T.
[14c]
Lett. 1994, 35, 2713Ϫ2716.
A. Togni, S. D. Pastor, J. Org.
Beurskens, G. Admiraal, G. Beurskens, W. P. Bosman,
S. Garcia-Granda, R. O. Gould, J. M. M. Smits, C. Smykalla,
DIRDIF97 Program System, Technical Report. Crystallogra-
phy Laboratory, University of Nijmegen, The Netherlands,
1997.
Chem. 1990, 55, 1649.
[15] [15a]
P. J. Davies, D. M. Grove, G. van Koten, Organometallics
[15b]
1997, 16, 800Ϫ802.
B. L. Kropman, D. H. A. Blank, E. G. Keim, M. M. A. Smi-
W. T. S. Huck, F. C. J. M. van Veggel,
thers, D. N. Reinhoudt, J. Am. Chem. Soc. 1995, 117, 8293.
Received October 30, 2002
Eur. J. Org. Chem. 2003, 1692Ϫ1703
www.eurjoc.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1703