Facile synthesis and structural variation of novel heterobimetallic alkali metal-zinc-alkoxide and -siloxide clusters

Article abstract of DOI:10.1039/b302585c

The novel alkali metal zinc-alkoxide and the first -siloxide aggregates [(thf)M(MeZn)(O^tBu)₂]₂ 1a (M = Li), 1b (M = Na), [(thf)₂K(MeZn)(OSiMe₃)₂]₂ 2and [(tmeda)KZn(OSiMe₃)₃]₂ 3 are easily accessible from the reaction of Me₂Zn with MOR (molar ratio 1 : 1; M = Li, Na, K; R = ^tBu, SiMe₃) in boiling thf and tmeda, respectively. While 1a, 1b and 2 possess distorted M₂Zn₂O₄ heterocubane frameworks, compound 3 consists of a K₂Zn₂O₆ core of a strongly distorted, face-fused double-heterocubane with two missing corners. In contrast, heating a mixture of Me₂Zn and KO^tBu in the molar ratio of 1 : 1 in toluene affords the donor solvent-free K-Zn-O cluster [K(MeZn)₃(O^tBu)₄] 4 which crystallizes as a polymer of strongly distorted [KZn₃O₄] heterocubanes via intermolecular agostic K ... MeZn interactions. The formation of the clusters may be rationalized in terms of alkali metal ion- and donor solvent-dependent ligand exchange reactions of methyl(alkoxide)- and methyl(siloxide)-zincates as initial products. Some of the initial products have been detected by means of electro spray ionisation (ESI) mass spectrometry.

Full text of DOI:10.1039/b302585c

View Article Online / Journal Homepage / Table of Contents for this issue
Facile synthesis and structural variation of novel heterobimetallic
alkali metal–zinc-alkoxide and -siloxide clusters†
Klaus Merz, Stefan Block, Robert Schoenen and Matthias Driess
Lehrstuhl für Anorganische Chemie 1: Cluster- und Koordinationschemie,
Ruhr-Universität Bochum, Universitätsstrasse 150, D-44801 Bochum, German.
E-mail: matthias.driess@ruhr-uni-bochum.de
Received 6th March 2003, Accepted 2nd July 2003
First published as an Advance Article on the web 28th July 2003
The novel alkali metal zinc-alkoxide and the first -siloxide aggregates [(thf )M(MeZn)(OtBu)2]2 1a (M = Li), 1b
(M = Na), [(thf )2K(MeZn)(OSiMe3)2]2 2 and [(tmeda)KZn(OSiMe3)3]2 3 are easily accessible from the reaction of
Me2Zn with MOR (molar ratio 1 : 1; M = Li, Na, K; R = tBu, SiMe3) in boiling thf and tmeda, respectively. While 1a,
1b and 2 possess distorted M2Zn2O4 heterocubane frameworks, compound 3 consists of a K2Zn2O6 core of a strongly
distorted, face-fused double-heterocubane with two missing corners. In contrast, heating a mixture of Me2Zn and
KOtBu in the molar ratio of 1 : 1 in toluene affords the donor solvent-free K–Zn–O cluster [K(MeZn)3(OtBu)4] 4
which crystallizes as a polymer of strongly distorted [KZn3O4] heterocubanes via intermolecular agostic K ؒ ؒ ؒ MeZn
interactions. The formation of the clusters may be rationalized in terms of alkali metal ion- and donor solvent-
dependent ligand exchange reactions of methyl(alkoxide)- and methyl(siloxide)-zincates as initial products. Some of
the initial products have been detected by means of electro spray ionisation (ESI) mass spectrometry.
zinc alkoxides is only scarcely developed, although they have a
Introduction
long tradition as reactive species in both organic and inorganic
The rich structural variety of discrete metal–oxygen clusters is
synthesis. For example, it is well known that dialkylzinc
of importance for the development of new multi-component
compounds can be highly activated as alkyl transfer reagents
materials and catalysts based on metal oxides.1 Especially the
towards unsaturated organic substrates in the presence of alkali
development of novel single-source precursors which allow the
metal alkoxides,7 and diorganozinc compounds react with
synthesis of size- and shape-selected, homo- and hetero-metal
CO to give acyloins but only in the presence of alkali metal
oxide particles is one of the actual challenges in inorganic
alkoxides as promotors.8
chemistry with regard to material science. Volatile organo-
metallic oxide presursors offer several advantages for the syn-
thesis of nano-scaled metal oxides in different environments
because they can be used in chemical-vapour-synthesis (CVS),
A survey of the literature on alkali metal zincates shows
that the structural chemistry is mostly devoted to lithium tri-
organo- and tetraorgano-zincates,9 while little is known about
alkoxyzincates.10 Recently, remarkable one-pot syntheses of
in solution (non-hydrolytic and hydrolytic sol–gel-synthesis) or
mixed organo-alkoxyzincates have been reported by reactions
in solid state synthesis (synthesis of nanoscaled solids). We are
of ZnMe2 with secondary amines or amidine ligands and LitBu
currently using organozinc-alkoxide2 and -siloxide clusters3 as
in the presence of molecular oxygen but this method seems
molecular models and volatile multiple precursors for nano-
rather limited to lithium derivatives.11,12 The latter method gave
scaled ZnO, Zn and ortho-zinc silicate (Zn2SiO4) particles.4
access to the structurally characterized Li–Zn–O clusters
ZnO is one of the most important substrates in heterogeneous
[((thf )Li)2(MeZn)2(OtBu)4] A and [Li(MeZn)3(OtBu)4] B having
catalysis (e.g., industrial methanol synthesis5) but its (micro)-
a heterocubane core. Related bimetallic organozinc siloxides
structure–reactivity relationship is not yet well understood even
with larger cluster frameworks have also been reported.13,14
after more than 30 years of intensive research. We have shown
that organozinc siloxide clusters3 can be used as remarkable
versatile precursors for the synthesis of highly active nanocrys-
talline ZnO and heterometal ZnO supports for the conversion
of CO2 and H2 into methanol and H2O.5 Remarkably, alkali
metal ion-modified ZnO supports show uncommonly strong
one- and two-electron donor properties and are unusual super-
basic heterogeneous catalysts.6 Since the catalytic activity of
ZnO supports can be drastically enhanced in the presence of
several promotor components (e.g., Cu particles, alkali metal
ions, etc.) it is furthermore highly desirable to gain access to
appropriate single-source precursors for such multi-component
While previous investigations on the reaction of diorgano-
ZnO substrates which are far less developed or hitherto
zinc compounds (R2Zn) with alkali metal alkoxides (MORЈ) in
unknown. Therefore we are also currently focusing on the
benzene were reported to give only organozincates with R2Zn :
synthesis of structurally defined heterometal zinc-alkoxide and
MORЈ ratios of 1 : 1 or 2 : 1,15 we have found that ligand
-siloxide clusters as molecular single-source presursors for
exchange reactions between dimethylzinc and alkali metal-
alkali metal-modified ZnO systems. We learned that the
alkoxides and -siloxides can also lead to a variety of novel
structural chemistry of seemingly simple mixed alkali metal
heterobimetallic ZnO clusters. We present here a simple and
productive method for the synthesis of novel alkali metal–
methylzinc-alkoxide and -siloxide clusters, starting from di-
methylzinc and the respective alkali metal salts in coordinating
and non-coordinating solvents at elevated temperature.
† Electronic supplementary information (ESI) available: the ESI spec-
tra of the reaction mixtures tBuOLi + ZnMe2, Me3SiOK + ZnMe2 and
tBuOK + ZnMe2. See http://www.rsc.org/suppdata/dt/b3/b302585c/
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
D a l t o n T r a n s . , 2 0 0 3 , 3 3 6 5 – 3 3 6 9
3365

View Article Online
Results and discussion
Reaction of a mixture of the respective alkali metal tert-but-
oxides and trimethylsiloxides with ZnMe2 in the molar ratio of
1 : 1 in thf, tmeda or toluene affords the bimetallic M–Zn–O
clusters 1a (M = Li), 1b (M = Na), 2, 3, and 4 (M = K), respect-
ively, which were isolated in the form of colorless crystals in
62–74% yield (Scheme 1).
Scheme 1 Reagents and conditions to synthesize the compounds 1a,
1b, 2, 3, and 4.
Scheme 2 Proposed mechanism for the formation of 1a, 1b, 2, and 3
Since the driving force for the formation of the mixed aggre-
gates is attributed to ligand exchange and effective solvation of
the alkali metal ions during formation of different contact ion
pairs, one can expect that the polarity and donor ability of the
solvent have a large influence on the nature of the products.
Surprisingly, replacement of thf by tmeda affords neither the
structural analogue of 1a, 1b nor that of 2. Instead the reaction
of a thf/tmeda solution of KOtBu with ZnMe2 affords a solu-
tion from which the crystalline compound 3 deposits. This
signals that the donor ability of the solvent plays a major role
during the required ligand exchange processes. Although the
mechanism is still unknown, it is reasonable to assume that
the initial step of the reaction of MOR with ZnMe2 occurs via
several multinuclear zincate species as outlined in Scheme 2.
Previous results by Richey et al. suggested that the dimeric
zincate species C is one of the initial products.15 Our attempts to
detect C and the other anionic intermediates D, E and F
(Schemes 2 and 3) in reaction mixtures were only of partial
success. We applied ESI (electro spray ionisation) mass spec-
trometry for the detection of negatively charged intermediates
in the reaction mixtures. Thus, reaction mixtures for the
via C, D, and E.
synthesis of 1a contain mostly the cluster anion {[C ϩ Liϩ
ϩ
(LiOtBu)2 ϩ 4 thf]} (see ESI†). The formation of C is also
evident by detection of [C ϩ Kϩ] in the mixture for the
synthesis of 2. Additionally, the spectrum for the synthesis of
2 shows the anion [D ϩ Kϩ] besides other not as yet identified
anions (see ESI†).
Scheme 3 Proposed mechanism for the formation of 4 via D, F, and G.
However, alkali metal-coordinated or “free” E could not be
detected by ESI mass spectrometry of reaction mixtures in the
synthesis of 3. Apparently, coordination of D to thf-solvated Li
and K ions readily leads to 1a, 1b, 2, while coordination of E to
tmeda-solvated potassium ions furnishes 3. The formation of
4 can be simply explained on the basis of another equilibrium
where D dismutates, under liberation of ORϪ, to the pseudo-
cubic chelate ligand F and G as depicted in Scheme 3. While G
could not be detected, the mixed anion [(MeZn)3(OtBu)4]Ϫ (F) is
visible in ESI mass spectrometry from reaction mixtures in the
synthesis of 4.
Due to fast ligand exchange processes at ambient temper-
ature on the NMR time scale and the scarce solubility of the
1
zincates below room temperature, neither H nor 13C NMR
D a l t o n T r a n s . , 2 0 0 3 , 3 3 6 5 – 3 3 6 9
3366

View Article Online
spectroscopic measurements were appropriate for the assign-
ment of the products in the reaction mixtures. Therefore the
structural characterization of 1a, 1b, 2, 3 and 4 is based on
single-crystal X-ray diffraction.
The lithium salt 1a is identical with A and adopts a pseudo-
cubane structure which has already been reported.16 X-Ray
analysis of the sodium analogue 1b revealed the presence of the
isostructural Na2Zn2O4 cubane core but due to the moderate
crystal quality the structural model could not be refined accur-
ately. In the case of the siloxide cluster 2 the molecular structure
consists of a pseudo-cubic K2Zn2O4 cluster core where each
potassium ion bears two thf ligands (Fig. 1). Apparently, the
large potassium ion radius and the presence of the thf donor
molecules affects a remarkably strong distortion of the hetero-
cubane core. The nearly planar Zn2O2-moiety reveals mean
Zn–O distances of 2.033(4) and 2.063(4) Å, which are slightly
shorter than the values observed in [MeZn(OSiEt3)]4 (Zn–O
average 2.089(4) Å) and related systems.3c,17 However, the
Zn1–O3 and Zn2–O4 distances in 2 are even shorter (averaged
1.995(3) and 1.984(3) Å) due to the higher negative partial
charge of the O3 and O4 atoms. In contrast to the almost
planar Zn2O2 partial structure, the four-membered K2O2
moiety is puckered with a dihedral angle between the two KO2
planes of 36.6Њ. In line with that, the endocyclic O–K–OZn
angles (average 69Њ) deviate from the ideal value of 90Њ for a
cubic system.
Fig. 2 Molecular structure of dimeric tmeda solvate [(tmeda)KZn-
(OSiMe3)3]2 3; hydrogen atoms omitted for clarity. Selected distances
[Å] and angles [Њ]: Zn(1)–O(1) 2.025(2), Zn(1)–O(2) 1.889(3), K(1)–O(1)
2.807(4), K(1)–O(2) 2.698(4); O(1)–Zn(1)–O(2) 104.5(1), Zn(1)–O(1)–
Zn(1) 94.0(1), O(2)–K(1)–O(1) 68.44(8).
refined with 100% occupancy. The result of the refinement of 4
shows significant differences in M–O bond lengths and angles.
The average Zn–O and K–O distances of 2.055(7) and 2.687(7)
Å as well as the intramolecular Zn ؒ ؒ ؒ Zn (3.050(2) Å) and
Zn ؒ ؒ ؒ K (3.575(3) Å) distances reflect a strong distortion of
the pseudo-cubic KZn3O4 core. Similar to 2 and 3, the O–K–O
angles are strongly distorted (66.4(2), 67.3(2) and 67.6(2)Њ) and
differ from the expected value of 90Њ for an ideal cube. Interest-
ingly, the KZn3O4 cubes in 4 are connected to a zig-zag polymer
via agostic intermolecular K ؒ ؒ ؒ H3CZn coordination
(K ؒ ؒ ؒ H–C interactions) of the coordinatively unsaturated
potassium ion with one close Me group of an adjacent cluster
(Fig. 3b). The K1 ؒ ؒ ؒ C2a–Zn2a angle of the bridge is nearly
linear (174.0(6)Њ) and the intermolecular K1 ؒ ؒ ؒ C2a distance
(3.20(1) Å) indicates K ؒ ؒ ؒ H–C interactions. The respective
K ؒ ؒ ؒ H distances (hydrogen atoms of Zn–Me could not be
located) were calculated to range from 2.9 to 3.1 Å. There is
no sign of significant coordinating interactions between the
toluene molecules in the voids of the crystal and the potassium
ion.
Fig. 1 Molecular structure of the K2Zn2O4 heterocubane [(thf )2-
K(MeZn)(OSiMe3)2]2 2; hydrogen atoms omitted for clarity. Selected
distances [Å] and angles [Њ]: Zn(1)–O(1) 2.033(4), Zn(1)–O(2) 2.063(4),
Zn(2)–O(1) 2.061(4), Zn(2)–O(2) 2.081(4), Zn(1)–O(3) 1.995(3), Zn(2)–
O(4) 1.984(3), K(1)–O(1) 2.682(4), K(1)–O(4) 2.692(4), K(1)–O(3)
2.716(3), K(2)–O(3) 2.672(4), K(2)–O(2) 2.696(4), K(2)–O(4) 2.698(4);
O(1)–K(1)–O(4) 68.5(1), O(1)–K(1)–O(3) 68.7(1), O(4)–K(2)–O(2)
70.1(1), O(2)–K(2)–O(4) 69.9(1).
Conclusion
We have shown that the ligand exchange reactions between
dimethylzinc and alkali metal tert-butoxides and trimethyl-
siloxides in coordinating (thf, tmeda) or non-coordinating
solvents (toluene) offer a simple and efficient access to novel
mixed alkali metal–zinc-alkoxide and -siloxide clusters. Appar-
ently the coordination ability of the solvent molecules towards
the alkali metal ions has a more delicate influence on ligand
exchange and formation of higher mixed aggregates than
previously proposed by Richey et al. and Wheatley et al.11,12
Further work is necessary to elucidate the mechanism for the
nucleation process and to probe whether one could also use
this method to synthesize other molecular heterobimetallic
transition-metal/zinc oxo clusters.
X-Ray structure analysis of [(tmeda)KZn(OSiMe3)3]2
3
revealed a centrosymmetric structure of a novel hetero-
bimetallic alkali metal zincate cluster framework. As shown in
Fig. 2, the crystal is composed of units which formally are
dimers of mixed K(tmeda)OR/Zn(OR)2 aggregates (R = SiMe3)
in which all metal centers share bridging OSiMe3 groups. The
molecular structure of the mixed K2Zn2O6 cluster core consists
of a strongly distorted face-fused double heterocubane with
two missing corners. The planar Zn2O2 moiety has uniform
Zn–O distances which resemble the values observed for
related Zn2O2 dimers.3c,17 While each Zn atom exhibits a slightly
distorted tetrahedral ligand surrounding, the K ions are five-
fold-coordinated.
The structural analysis of 4 revealed that the heterocubane
cluster crystallizes as a toluene solvate and is polymeric. The
framework-constituents of the new heterobimetallic polymer
are strongly distorted cubes of KZn3O4 units (Fig. 3). In
contrast to the previously reported crystal structure analysis
of LiZn3O4 B,11 the metal centers in the KZn3O4 core can be
Experimental
Methods and materials
General remarks. All manipulations were carried out under
anaerobic conditions in dry argon using standard Schlenk
techniques. Solvents were refluxed over an appropriate drying
agent, and distilled and degassed prior to use. All other chem-
ical reagents were used as received from Aldrich or Lancaster
D a l t o n T r a n s . , 2 0 0 3 , 3 3 6 5 – 3 3 6 9
3367

View Article Online
Fig. 3 a) Molecular structure of 4ؒC7H8 and b) crystal packing of the polymeric [K(MeZnOtBu)4]ؒC7H8 (4ؒC7H8); hydrogen atoms have been
omitted for clarity. Selected distances [Å] and angles [Њ]:.Zn(1)–O(1) 2.055(7), Zn(1)–O(2) 2.093(7), Zn(1)–O(3) 2.015(7), Zn(2)–O(1) 2.017(7),
Zn(2)–O(2) 2.087(7), Zn(2)–O(4) 2.005(7), Zn(3)–O(2) 2.104(7), Zn(3)–O(3) 2.047(7), Zn(3)–O(4) 2.036(7), K(1)–O(1) 2.696(7), K(1)–O(3) 2.667(7),
K(1)–O(4) 2.698(7), Zn(2)–C(2) 2.00(1), K(1) ؒ ؒ ؒ C(2a) 3.20(1); O(1)–K(1)–O(4) 66.4(2), O(1)–K(1)–O(3) 67.3(2), O(3)–K(1)–O(4) 67.6(2).
without further purification. NMR spectra were recorded on a
Bruker Avance250 spectrometer at ambient temperature oper-
ating at 250.1 MHz (1H). Chemical shifts are referenced to
SiMe4 at 0.00 ppm (1H). ESI mass spectra were measured on a
Bruker Daltonik Esquire 3000 plus (4 kV spray voltages) in
negative mode, controlled by Esquirecontrol 5.0.
[(tmeda)KZn(OSiMe3)3]2 3. A solution of ZnMe2 (1.44 g,
15.2 mmol, 2 solution in toluene (Aldrich)) was added at Ϫ78
ЊC to a solution of 1.95 g (15.2 mmol) KOSiMe3 in 80 ml thf
and 20 ml tmeda and stirred at this temperature for 1 . After
being allowed to warm to room temperature the mixture was
refluxed for 1 h. At room temperature all volatiles were removed
in vacuo (10Ϫ3 Torr). Colorless crystals were isolated after
recrystallization of the residue from a little thf at Ϫ25 ЊC. Yield:
4.51 g (4.71 mmol, 62%); mp = 178–182 ЊC (decomp.). 1H NMR
(C6D6): δ = 2.2 (s, 8 H; tmeda), 2.4 (s, 24 H; tmeda), 0.60 (s, 36
H; SiMe3), 0.58 (s, 18H; SiMe3). C30H68N4O6K2Si6Zn2 (958.36):
calc. C 37.60, H 7.15; found C 37.15, H 7.12%.
Synthesis and characterization. [(thf)M(MeZn)(OtBu)2]2
1a: M = Li; 1b: M = Na. A solution of ZnMe2 (1.44 g, 15.2
mmol, 2 M solution in toluene (Aldrich)) at Ϫ78 ЊC in ca. 100
ml thf was slowly treated with (15.2 mmol) MOtBu in 10 ml thf.
After being allowed to warm to room temperature the mixture
was refluxed for 1 h. At room temperature all volatile com-
ponents were removed in vacuo (10Ϫ3 Torr). Colorless plates
were isolated after recrystallization of the residue from a little
toluene at Ϫ25 ЊC.
1a: M = Li. Yield: 3.43 g (5.63 mmol, 74%); the compound
was characterized according to the literature procedure.16
1b: M = Na. Yield: 3.25 g (5.17 mmol, 68%); mp = 183–188 ЊC
(decomp.). 1H NMR (C6D6): δ = 3.61 (m, 8H; thf ), 1.79 (m, 8H;
thf ), 1.22 (s, 18H; tBu), 1.17 (s, 18H; tBu), Ϫ0.75 (s, 6H; ZnMe);
C26H48Na2O2Zn2 (643.5): calc. C 48.53, H 9.09; found C 48.01,
H 9.06%.
[K(MeZn)3(OtBu)4]ؒC7H8 4ؒC7H8. A solution of ZnMe2
(1.44 g, 15.2 mmol, 2 M solution in toluene (Aldrich)) at
Ϫ78 ЊC in ca. 100 ml toluene was slowly treated with 1.70 g
(15.2 mmol) KOtBu in 100 ml toluene. The reaction mixture was
allowed to warm to room temperature within 8 h and then
stirred for 1 d at 70 ЊC. The solvent was removed under reduced
pressure affording a solid residue. The latter was redissolved in
ca. 20 ml toluene from which the desired product crystallizes
in a few days in the form of colorless crystals. Yield: 2.19 g
1
(3.29 mmol, 65%); mp > 330 ЊC (decomp.). H NMR (C6D6):
t
δ = 7.12–7.83 (m, 8 H), 1.25 (s, 9 H; BuOK), 1.27 (s, 27 H;
[(thf)2K(MeZn)(OSiMe3)2]2 2.
A
solution of ZnMe2
tBuOZn), Ϫ0.61 (s, 3H; MeZn). C26H53O4KZn3 (664.89): calc.
(1.44 g, 15.2 mmol, 2 M solution in toluene (Aldrich)) was
added at Ϫ78 ЊC to a solution of 1.95 g (15.2 mmol) KOSiMe3
in 100 ml thf and stirred at this temperature for 1 h. After being
allowed to warm to room temperature the mixture was refluxed
for 2 h. After removal of the solvent and other volatile com-
ponents in vacuo (10Ϫ3 Torr), the resulting residue was taken up
in ca. 20 ml toluene. The desired product crystallizes at Ϫ25 ЊC
after several days in the form of colorless needles. Yield: 5.09 g
(5.78 mmol, 76%); mp = 181–185 ЊC (decomp.). 1H NMR
(C6D6): δ = 3.63 (m, 16H; thf ), 1.80 (m, 16 H; thf ), 0.62 (s, 18 H;
SiMe3), 0.61 (s, 18 H; SiMe3), Ϫ0.71 (s, 6 H; ZnMe).
C30H72O8K2Si4Zn2 (882.2): calc. C 40.84, H 8.13; found C 39.62,
H 8.10%.
C 46.96, H 8.03; found C 46.51, H 7.99%.
X-Ray crystallography
The intensities were measured with a Bruker-axs-SMART
diffractometer (MoKα radiation, λ = 0.71707 Å, ω-scan, T = 213
K), the structures were solved by direct methods.18 Refinements
were carried out with the SHELXL-97 package.19 All non
hydrogen atoms were refined with anisotropic temperature
factors. The hydrogen atoms were placed in calculated positions
and refined isotropically in riding mode. All refinements were
made by full-matrix least-squares on F 2. For 3, the tmeda
solvent molecule and one of the two independent trimethylsilyl
D a l t o n T r a n s . , 2 0 0 3 , 3 3 6 5 – 3 3 6 9
3368

View Article Online
2 Review: N. Y. Turova, E. P. Turevskaya, V. G. Kessler and
groups was found to be disordered and therefore modeled
in two equally-occupied orientations. The contribution of the
disordered thf solvent molecules in 3 to the structure factors
was taken into account by back-Fourier transformation using
PLATON/SQUEEZE.20
M. I. Yanovskaya, The overview of metal alkoxide derivatives along
the groups of the periodic table, in The Chemistry of Metal
Alkoxides, Kluwer Academic Publishers, Dordrecht, 2002, ch. 12,
p. 217.
3 (a) F. Schindler, H. Schmidbaur and U. Krüger, Angew. Chem., 1965,
77, 865; (b) F. Schindler and H. Schmidbaur, Angew. Chem., 1967,
79, 697; (c) M. Driess, K. Merz and S. Rell, Eur. J. Inorg. Chem.,
2000, 2517; (d ) K. Merz, H.-M. Hu, S. Rell and M. Driess, Eur. J.
Inorg. Chem., 2003, 51.
4 (a) K. Merz, R. Schoenen and M. Driess, J. Phys. IV, 2001, 11, 467;
(b) M. Driess, K. Merz, R. Schoenen, R. Rabe, F. E. Kruis, A. Roy
and A. Birkner, C. R. Chimie, 2003, in the press; (c) J. Hambrock,
S. Rabe, K. Merz, A. Wohlfarth, A. Birkner, R. A. Fischer and
M. Driess, J. Mater. Chem., 2003, 13, 1731; (d ) S. Polarz, S. Rabe,
A. Roy, E. Kruis, M. Driess, Chem. Eur. J., submitted.
5 (a) J. B. Hansen, in Handbook of Heterogeneous Catalysis, ed.
G. Ertl, H. Knözinger and J. Weitkamp, Wiley-VCH, Weinheim
1997, p. 1856; (b) B. A. Peppley, J. C. Amphlett, L. M. Kearns and
R. F. Mann, Appl. Catal. A, 1999, 179, 21; (c) H. Wilmer, M. Kurtz,
K. V. Klementiev, O. P. Tkachenko, W. Grünert, O. Hinrichsen,
A. Birkner, S. Rabe, K. Merz, M. Driess, C. Wöll and M. Muhler,
Phys. Chem. Chem. Phys., 2003, 5, DOI: 10.1039/b304425d.
6 P. Winiarek and J. Kijenski, J. Chem. Soc., Faraday Trans., 1998, 94,
167.
Compound 2. Orthorhombic, space group Pna21, formula:
C30H74K2O8Si4Zn2, a = 24.379(4), b = 13.394(3), c = 15.094(3) Å,
V = 4929(2) Å3, Z = 4, µ(Mo-Kα) = 1.28 mmϪ1, 416 parameters,
26363 reflections measured, 8568 unique (Rint = 0.0484), 6255
independent reflections (I > 2σ(I )), R1 = 0.0446 (observed
reflections), wR2 = 0.1095 (all data).
Compound 3. Orthorhombic, space group Pnnm, formula:
C30H86K2N4O6Si6Zn2, a = 19.208(7), b = 10.995(4), c = 13.369(5)
Å, V = 2823(2) Å3, Z = 2, µ(Mo-Kα) = 1.16 mmϪ1, 191 param-
eters, 14611 reflections measured, 2595 unique, (Rint = 0.0632),
1767 independent reflections (I > 2σ(I )), R1 = 0.0534 (observed
reflections), wR2 = 0.1545 (all data).
Compound 4. Orthorhombic, space group P212121, formula:
C26H53KO4Zn3, a = 10.748(5), b = 13.214(6), c = 23.706(9) Å,
V = 3367(2) Å3, Z = 4, µ(Mo-Kα) = 2.27 mmϪ1, 295 parameters,
7099 reflections measured, 4215 unique, (Rint = 0.0941), 3067
independent reflections (I > 2σ(I )), R1 = 0.0625 (observed
reflections), wR2 = 0.1259 (all data).
CCDC reference numbers 205578–205580.
See http://www.rsc.org/suppdata/dt/b3/b302585c/ for crystal-
lographic data in CIF or other electronic format.
7 (a) C. A. Musser and H. G. Richey, Jr., J. Org. Chem., 2000, 65, 7750
and references therein; (b) D. J. Ager, I. Fleming and S. K. Patel,
J. Chem. Soc., Perkin Trans. 1, 1981, 2520.
8 M. W. Rathke and H. Yu, J. Org. Chem., 1972, 37, 1732.
9 (a) E. Weiss and H. Plass, J. Organomet. Chem., 1968, 14, 21; (b)
D. Linton, P. Schooler and A. E. H. Wheatley, Coord. Chem. Rev.,
2001, 223, 53 and references therein.
10 A. P. Purdy and C. F. George, Polyhedron, 1994, 13, 709 and
references therein.
11 A. D. Bond, D. J. Linton, P. Schooler and A. E. H. Wheatley,
J. Chem. Soc., Dalton Trans., 2001, 3173.
12 S. R. Boss, R. Haigh, D. J. Linton and A. E. H. Wheatley, J. Chem.
Soc., Dalton Trans., 2002, 3129.
13 G. Anantharaman, H. W. Roesky and J. Magull, Angew. Chem., Int.
Ed., 2002, 41, 1226.
14 G. Anantharaman, N. D. Reddy and H. W. Roesky, Organometallics,
2001, 20, 5777.
Acknowledgements
We are grateful to the Deutsche Forschungsgemeinschaft (SFB
558 “Metall-Substratwechselwirkungen in der heterogenen
Katalyse”), the Fonds der Chemischen Industrie and the
Ministry of Science and Research of Nordrhein-Westfalen for
financial support.
15 R. M. Fabicon and H. G. Richey, Jr., J. Chem. Soc., Dalton Trans,
2001, 783.
16 R. P. Davis, D. J. Linton, P. Schoolwer, R. Snaith and A. E. H.
Wheatley, Eur. J. Inorg. Chem., 2001, 7, 3696.
17 M. M. Olmstead, P. P. Power and S. C. Shoner, J. Am. Chem. Soc.,
1991, 113, 3379.
18 G. M. Sheldrick, Acta Crytallogr., Sect. A., 1990, 46, 467.
19 G. M. Sheldrick, SHELXL-97, Program for Crystal Structure
Refinement, University of Göttingen, Germany, 1997.
20 P. v. d. Sluis and A. L. Spek, PLATON/SQUEEZE, Acta
Crystallogr., Sect. A, 1990, 46, 194.
References
1 (a) M. Veith, J. Chem. Soc., Dalton Trans, 2002, 2405; (b)
M. Marciniec and H. Maciejewski, Coord. Chem. Rev., 2001, 223,
301; (c) J. S. Matthews, O. Just, B. Obi-Johnson and W. S. Rees, Jr.,
Chem. Vap. Deposit., 2000, 6, 129; (d ) J. W. Kriesel and T. D. Tilley,
J. Mater. Chem., 2001, 11, 1081; (e) K. L. Fujdala and T. D. Tilley,
Chem. Mater., 2001, 13, 1817.
D a l t o n T r a n s . , 2 0 0 3 , 3 3 6 5 – 3 3 6 9
3369