Isomerism of coordination modes and numbers in pentanuclear organozinc hydroxylamides: An exercise in subtle substituent size effects

Article abstract of DOI:10.1002/ejic.200600625

The first organometallic zinc hydroxylamides [Zn(RZn)₄- (ONMe₂)₆] (1: R = Me; 2: R = iPr) and [Zn(EtZn) ₄(ONEt₂)₆] (3) have been prepared by alkane elimination from dialkylzinc solutions upon treatment with N,N- dialkylhydroxylamines. The molecular structures of 1 and 2 reveal that a subtle change in the constitution of the aggregates can make a striking difference. By the simple exchange of the MeZn⁺ groups with iPrZn⁺ groups, the coordination number of the central Zn²⁺ nucleus increases from four (1) to six (2). The two cluster types are unprecedented in both hydroxylamine and organometallic chemistry. Compounds 1 and 3 adopt a doubly bridged fenestrane-like Zn₅N₆O₆ core with two dangling NR₂ moieties, whereas iPrZn species 2 comprises an octahedroid backbone with all donor atoms attached to the five Zn nuclei with the maintenance of the same heteroatom composition as 1 and 3. Variable-temperature ¹³C NMR experiments of all-ethyl cluster 3 reveal that the R₂NO^- ligands rapidly exchange between the RZn⁺ and Zn²⁺ cations. This process is rationalised by a plausible model which also points out that the core structure of 3 could be an intermediate motif in the observed heteroatom connectivity exchange. In line with this, computational investigations reveal that both skeletons are very similar in energy. In sum, the coordination flexibility of the hydroxylamide ligand makes this group of organometallic zinc complexes a highly dynamic family of cluster compounds. Wiley-VCH Verlag GmbH & Co. KGaA, 2006.

Full text of DOI:10.1002/ejic.200600625

SHORT COMMUNICATION
DOI: 10.1002/ejic.200600625
“Isomerism” of Coordination Modes and Numbers in Pentanuclear Organozinc
Hydroxylamides: An Exercise in Subtle Substituent Size Effects
Matthias Ullrich,[a] Raphael J. F. Berger,[a] Christian Lustig,[a] Roland Fröhlich,[b] and
Norbert W. Mitzel*[a]
Dedicated to Prof. Gerhard Erker on the occasion of his 60th birthday
Keywords: Zinc / Hydroxylamines / Organometallic / Coordination modes / Crystal structure
The first organometallic zinc hydroxylamides [Zn(RZn)4-
(ONMe2)6] (1: R = Me; 2: R = iPr) and [Zn(EtZn)4(ONEt2)6] (3)
have been prepared by alkane elimination from dialkylzinc
solutions upon treatment with N,N-dialkylhydroxylamines.
The molecular structures of 1 and 2 reveal that a subtle
change in the constitution of the aggregates can make a
striking difference. By the simple exchange of the MeZn+
groups with iPrZn+ groups, the coordination number of the
central Zn2+ nucleus increases from four (1) to six (2). The
two cluster types are unprecedented in both hydroxylamine
and organometallic chemistry. Compounds 1 and 3 adopt a
five Zn nuclei with the maintenance of the same heteroatom
composition as 1 and 3. Variable-temperature 13C NMR ex-
periments of all-ethyl cluster 3 reveal that the R2NO– ligands
rapidly exchange between the RZn+ and Zn2+ cations. This
process is rationalised by a plausible model which also points
out that the core structure of 3 could be an intermediate motif
in the observed heteroatom connectivity exchange. In line
with this, computational investigations reveal that both skel-
etons are very similar in energy. In sum, the coordination
flexibility of the hydroxylamide ligand makes this group of
organometallic zinc complexes a highly dynamic family of
doubly bridged fenestrane-like Zn5N6O6 core with two dan- cluster compounds.
gling NR2 moieties, whereas iPrZn species 2 comprises an
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
octahedroid backbone with all donor atoms attached to the
Germany, 2006)
The high coordination flexibility of the hydroxylamide li-
gands that is demonstrated in Al and Ga organometallics
(e.g. in the readily fluctuating tetracyclic Al3 cluster
[{(Me2Al)[ON(Me)]2CH2}2(AlMe)][6]) can be expected by
hypothesis to allow for the construction of highly dynamic
Zn-based multinuclear cluster compounds. Zinc, on its own
accord, has a very rich alkoxide chemistry.[1,3] However, hy-
droxylamide derivatives of divalent metals are rather poorly
understood and encompass merely inorganic coordination
compounds of zinc[11] {e.g. [(H2NOH)2ZnCl2], Crismer’s
salt[12]}, but organometallic examples are completely un-
known.
We report herein the initial evidence to support the above
hypothesis. The reaction of N,N-dialkylhydroxylamines
Me2NOH or Et2NOH with binary zinc alkyls affords three
pentanuclear organozinc cluster compounds: [Zn(RZn)4-
(ONMe2)6] (1: R = Me; 2: R = iPr) and [Zn(EtZn)4-
(ONEt2)6] (3, vide infra).
Introduction
Compared to organometallics of the trivalent group 13
elements, zinc requires further contact to a donor atom to
reach coordination number (CN) four because of its di-
valent nature. In metal alkoxide/aryloxide chemistry,[1] an
increase in aggregate dimensionality is already well estab-
lished [compare the Al2O2 ring {(Me2Al)[O(2,6-iPr2-
[2]
C6H3)]}2 with the “double cube” [(MeZn)3(OMe)4]2Zn[3]
(CN = 4, 6)].[4] The highly diverse chemistry of the O–N
double-donor array (hydroxylamines or oximes) provides a
plethora of aggregation motifs, for example in group 13
complexes,[5–9] and ought to be transferable to zinc for the
exploration of similar extensions {compare the six-mem-
bered ring dimer [(Me2Al)(ON=CMe2)]2
rahedroid oximate [(MeZn)(ON=CMe2)]4
[5a,5b]
with the tet-
(CN = 4)}.
[10]
[a] Institut für Anorganische und Analytische Chemie, Westfälis-
che Wilhelms-Universität Münster,
Corrensstrasse 30, 48149 Münster, Germany
Fax: +49-251-8336007
E-mail: mitzel@uni-muenster.de
Results and Discussion
[b] Organisch-Chemisches Institut, Westfälische Wilhelms-Uni-
versität Münster,
With the focus on compounds 1 and 2 (Scheme 1), it is
elegantly revealed that the highly flexible coordinating
O–N ligands allow Zn atoms to adopt a CN Ͼ 4. This
Corrensstrasse 40, 48149 Münster, Germany
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
Eur. J. Inorg. Chem. 2006, 4219–4224
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4219

M. Ullrich, R. J. F. Berger, C. Lustig, R. Fröhlich, N. W. Mitzel
SHORT COMMUNICATION
stands in contrast to the isoelectronically related organozinc M–C bonds. The aggregation motif of 1 (all Zn atoms: CN
hydrazides (N–N double-donor), where zinc is exclusively = 4) is unprecedented in both hydroxylamine and organo-
tetracoordinate.[13]
metallic chemistry.
With a subtle size change of the alkyl group at zinc (from
Me to iPr), the core structure of the aggregate completely
shifts as revealed by the crystal structure of ether solvate
2·OEt2 (Figure 2).[14] In this cluster, the ether molecule fills
a void but does not contact any of the Zn atoms. The hy-
droxylamide cluster [Zn(ONMe2)6(ZniPr)4] (2) adopts an
octahedral-based structure with all six O–N ligands O-
bound to the central Zn2+ (CN = 6). Charge compensation
of this [Zn(ONMe2)6]4– fragment is achieved by four iPrZn+
groups, which bind to the six O–N units either in an N,O,O-
or in an N,N,O mode. Because all of the O and N atoms
are now coordinated to Zn, this contrasts the situation in 1
which has two dangling NMe2 groups – one key feature
of the observed coordination “isomerism”. Consequently,
cluster 2 is one of the limited examples of Zn species that
exhibit mixed CNs, whether in organometallic[1,3] or in in-
organic chemistry.[16]
Scheme 1.
Though the formulae of 1 and 2 (apart from the alkyl
groups on Zn) are identical, their three-dimensional coordi-
nation frameworks differ markedly. The all-methyl cluster
[Zn(MeZn)4(ONMe2)6] (1) adopts a fenestrane-like motif,
as determined by X-ray diffraction analysis (Figure 1).[14]
Four O–N moieties coordinate around the central “inor-
ganic” Zn2+ ion by their oxygen atoms and are themselves
interconnected by RZn+ units via O–Zn–N linkages. The
resulting dicationic heterofenestrane {[Zn(ONMe2)4-
(ZnMe)4]2+} is then charge-compensated by two anionic µ2-
O-bridging Me2NO– ligands. If the latter are neglected, the
core of 1 is reminiscent of a handful of known complexes
(e.g. in bioinorganics),[15] but those complexes exhibit solely
square-planar coordinated central metal ions and have no
Figure 2. Solid-state structure of 2 in a crystal of the composition
2·OEt2 (thermal ellipsoids at 50% probability; hydrogen atoms and
the Et2O solvate molecule are omitted; symmetry transformations
used to generate equivalent atoms: –x+1, –y+2, –z). Selected dis-
tances and angles [Å/°]: Zn2–O4Ј 1.974(2), Zn1–O2 2.205(2), Zn3–
N4 2.135(2), O4–N4 1.433(2), Zn2–C201 1.987(3), N3–C32
1.467(3); O2–Zn1–O3 78.0(1), O2–Zn1–O3Ј 102.0(1), O4Ј–Zn2–
N12 85.2(1), O3–Zn3–N4 106.0(1), Zn1–O12–Zn3 102.1(1), Zn2–
O3–Zn3 119.3(1), Zn3–O3–N3 108.5(1), Zn2–O3–N3 122.8(1),
Zn3–N4–O4 112.2(2), Zn3–N4–C41 109.0(2), O3–N3–C33
106.9(2), C41–N4–C43 110.8(3).
Although the structures of 1 and 2 are not the same, their
core compositions are, and thus, the question arises whether
a pathway for the interconversion between the two cluster
types exists. VT-NMR spectroscopy sheds some light onto
this and onto the molecular dynamics of the aggregates. Be-
tween 300 and 200 K, the 1H NMR spectra of 1 show some
signal broadening but no substantial change to indicate
structural variations; the 13C NMR spectra give a similar
picture. A drawback to these measurements is the limited
solubility of 1 especially at low temperatures, which is most
probably due to its minimal alkyl periphery.
Figure 1. Crystal structure of 1 (thermal ellipsoids at 50% prob-
ability; hydrogen atoms are omitted; symmetry transformations
used to generate equivalent atoms: –x+1, y, –z+3/2). Selected dis-
tances and angles [Å/°]: Zn1–O3 1.923(3), Zn2–O1 2.100(2), Zn3–
N1 2.175(3), O3–N3 1.435(4), Zn2–C2 1.966(4), N3–C32 1.464(5);
O1–Zn2–O2 86.8(1), O1–Zn1–O1Ј 132.5(1), O3–Zn3–N1 85.8(1),
O2–Zn2–N3Ј 103.5(1), Zn1–O1–Zn2 100.0(1), Zn2–O2–Zn3
121.9(1), Zn1–O1–N1 107.9(2), Zn3–O2–N2 123.8(2), Zn3–N1–O1
107.3(2), Zn2–N3Ј–O3Ј 111.6(2), Zn2–N3Ј–C32Ј 110.2(3), O2–N2–
C21 107.2(3), C21–N2–C22 109.8(4).
4220
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 4219–4224

Pentanuclear Organozinc Hydroxylamides – Coordination Modes and Numbers
SHORT COMMUNICATION
Fortunately, the all-ethyl analogue [Zn(EtZn)4(ONEt2)6] be expected. The overall packing in the aggregation motif
(3, Scheme 2) is much more soluble and represents an ex- found for 2 is more compact than the one found for 1 and
quisite VT-NMR substitute for all-methyl cluster 1 because 3.
it adopts the same core motif (Figure 3). Furthermore, the
The 13C NMR spectrum of 3 at 298 K exhibits broad
fact that 1 and 3 exhibit the same coordination framework features for the nitrogen-bound ethyl substituents in the re-
underlines that the structural differences between 1 and 2 gions between ca. 50 and 60 ppm and between ca. 10 and
cannot be accounted for merely by means of the steric con- 15 ppm (Figure 4). At 213 K however, all five resonances
gestion of the “organic sphere” about the Zn5N6O6 hetero- are well-resolved and observable. This is in accordance with
atom core. The number of carbon atoms in that sphere a C2 molecular symmetry with freely rotating O–N bonds
doubles from 16 (compound 1) to 32 (compound 3) without to the dangling NEt2 groups. The ethyl groups at zinc give
any effect on the aggregation motif. Exchange of the two sets of sharp resonances at both low and ambient tem-
MeZn+ groups (compound 1) by iPrZn+ groups (compound peratures. Obviously, the R2NO– units in compounds 1 and
2) increases that number only by 8 (from 16 to 24). A simple 3 exchange between the five Zn nuclei in a highly fluxional
focus on group sizes – though obvious when only differ- manner.
ences between 1 and 2 are examined based on the literature
findings of EtZn boryloxides[17] or Stephan’s bulky Me2Al
aryloxide[2] (compared to “normal” Al–O organyls with CN
Ͼ 4[1]) – is not a definite reason for the varied CN of the
present organozinc hydroxylamides. In fact, if bulkiness of
the group had an influence on the CN of the central “inor-
ganic” Zn2+ nucleus, the opposite trend in the CN would
Scheme 2.
Figure 4. Details of the 13C NMR spectra of [Zn(ZnEt)4(ONEt2)6]
(3) at 298 K (in [D6]benzene) and at 213 K (in [D8]toluene, DEPT
135 mode).
The 1H NMR spectrum of 2 (Figure 5) displays one
prominent resonance for the Me2NO– units at 300 K, which
indicates an exchange on the NMR time scale comparable
in magnitude to that found for the R2NO– units of 1 and
3. At temperatures below 270 K, this peak splits into a com-
plex set of signals of rather low resolution. Thus, a drop in
the symmetry of the molecule below that of the crystalline
state (C2) can be assumed since the latter would require six
chemically different N–Me groups. From the complexity of
the NMR spectrum, the presence of more than one isomeric
species in solution is reasonable. The two motifs found in
the solid state for 1 and 2 are two likely possibilities.
The observed highly dynamic behaviour is consistent
with rapidly isomerising aggregates. A plausible model is
illustrated in Scheme 3a with the heteroatom core of 1. The
Figure 3. Crystal structure of 3 (thermal ellipsoids at 50% prob-
process by which this core is transformed into itself can
ability; hydrogen atoms and the disorder of the ethyl group at Zn2
be reasoned by the division of the process into four single
are omitted). Selected distances and angles [Å/°]: Zn1–O11
1.944(7), Zn2–O3 2.041(7), Zn4–N11 2.186(9), O2–N2 1.444(11),
steps.[18,19] The first two steps involve the bond formation
Zn5–C501 1.972(12), N2–C23 1.461(15); O5–Zn5–O11 88.7(3),
O11–Zn1–O12 139.7(3), O2–Zn2–N12 86.2(3), O3–Zn3–N4
107.8(3), Zn1–O11–Zn5 102.3(3), Zn4–O5–Zn5 117.7(3), Zn1–
O12–N12 109.7(5), Zn4–O5–N5 121.9(6), Zn4–N11–O11 104.5(5),
Zn5–N2–O2 112.9(6), Zn5–N2–C21 112.6(7), O2–N2–C23
106.0(8), C21–N2–C23 110.8(10).
between the peripheral noncoordinated N atom and an
RZn+ group [colour code: green (1)]; in addition, there is a
change in connectivity of the O atom of the hydroxylamide
from a peripheral Zn atom to the central Zn atom [colour
code: orange (2)]. The central Zn atom then becomes hexa-
Eur. J. Inorg. Chem. 2006, 4219–4224
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
4221

M. Ullrich, R. J. F. Berger, C. Lustig, R. Fröhlich, N. W. Mitzel
SHORT COMMUNICATION
ferred by 5.4 kcalmol–1. At the RI-BP/SV(P) level, the re-
verse was found; the structure of compound 2 was preferred
with an energy difference of 1.2 kcalmol–1.[20]
The small differences in the theoretical calculations along
with the temperature-dependent NMR studies reveal both
aggregation motifs to be very comparable in energy. This
explains the sensitivity of the observed aggregation and co-
ordination modes on small substituent effects and the
highly dynamic behaviour of the species. Thus, pentanuclear
organozinc hydroxylamides 1–3 disclose for the first time
that a metal can be flexibly coordinated by O–N ligands
not only in terms of coordination modes but also numbers.
On the basis of the experimental findings and the unique
ability of the hydroxylamide ligand to be flexibly coordinat-
ing, this group of organometallic zinc complexes forms a
highly dynamic family of cluster compounds.
Experimental Section
CAUTION: Pure zinc organyls may spontaneously ignite upon
contact to air or moisture!
1
Figure 5. Variable-temperature H NMR spectra of 2 (in [D8]tolu-
ene) in the chemical shift region of the ONMe2 moieties. For clar-
ity, the highlighted temperature range around the coalescence point
is provided in steps of 10 K.
Materials and Methods: All manipulations were carried out under
a dry nitrogen atmosphere with standard Schlenk and high-vacuum
techniques with double manifolds or in a glove box (MBraun Lab
Master 130) operated under an argon atmosphere. All solvents
were purified by standard procedures immediately prior to use.
Me2Zn was obtained in its pure form by the fractional distillation
of a commercial solution (2.0  in toluene, Aldrich). Et2Zn
(Crompton GmbH, Bergkamen, Germany) was transferred into
glassware in the glove box and used without further purification.
iPr2Zn was synthesised according to the literature[21] and further
purified by fractional condensation (removal of trace Et2O from
synthesis). It was stored in the dark at –78 °C because it decom-
poses quite quickly upon exposure to light or ambient temperature.
Me2NOH was obtained from its hydrochloride salt (Aldrich) by
treatment with excess NH3 at –78 °C and separated from the
NH4Cl residue by distillation.[22] Et2NOH (Aldrich) was purified
by fractional distillation.
coordinate, and this finalises the transformation of the core
structure from the type adopted by compound 1 to that of
2 in the crystalline state.
Scheme 3. a) Plausible model for the fluctuation of the O–N li-
gands. The overall C2 symmetry compels each step to occur twice.
b) The isomeric heteroatom cluster core as found for 2 in the solid
state – an “intermediate” of the periodical fluctuation.
General Synthetic Protocol: R2Zn (R = Me, Et or iPr) was con-
densed into a round-bottomed flask and then diluted with n-hexane
to generate a ca. 1  solution. With heavy stirring, a solution of
Me2NOH or Et2NOH in THF (ca. 1 ) was slowly added at 0 °C.
The ice bath was removed, and the reaction mixture was stirred
until gas evolution had ceased. All volatiles were evaporated under
reduced pressure, and the colourless to light yellow viscous solid
was extracted with n-hexane/Et2O and filtered. The resulting solu-
tion was carefully concentrated under reduced pressure and stored
at –26 °C (1 and 3) or –45 °C (2) for several days to yield colourless
crystals.
Two further steps that are reverse in nature to the preced-
ing two allow the ligands to rearrange about the RZn+ and
Zn2+ atoms to reconstruct the heteroatom connectivity
pattern as found in the solid state of 1 {the O atom of one
O–N unit moves from the central Zn2+ nucleus to a periph-
eral RZn+ group [colour code: blue (3)], and the N atom of
the same O–N unit disconnects from the organozinc residue
to become the dangling NR2 group [colour code: red (4)]}.
In order to gain more detailed insight into the energetics
of these systems, we calculated the energies of the two clus-
ter types represented by 1 and 2 in the crystalline state for
the most simple cases, all-methyl compound 1 and the
MeZn+ analogue of 2, because both are unaffected by sub-
stituent conformations. The computations were carried out
at the SCF/SV(P) and RI-BP/SV(P) levels of theory and
reproduce the core structure parameters of the experimen-
tally determined geometries well. At the SCF/SV(P) level,
1: From Me2Zn (15.8 mmol, 1.51 g) in n-hexane (15 mL), and
Me2NOH (19.0 mmol, 1.16 g, 1.30 mL) in THF (20 mL). Colour-
1
less cubes. Yield 1.35 g (1.80 mmol, 57%). H NMR (200.1 MHz,
C6D5CD3): δ = –0.68 to –0.15 (m, 12 H, 4×ZnCH3), 2.20–3.00
[m, very br., 36 H, 6×N(CH3)2] ppm. 13C{1H} NMR (50.3 MHz,
C6D5CD3): δ = 3.4, 5.7 [2×(2 C, 2×ZnCH3)], 51–61 [very br., 12
C, 6 ×N(CH3)2] ppm. C16H48N6O6Zn5 (747.47): calcd. C 25.71, H
6.47, N 11.24; found C 25.34, H 6.32, N 11.12.
2: From iPr2Zn (17.4 mmol, 2.64 g) in n-hexane (20 mL), and
Me2NOH (21.0 mmol, 1.28 g, 1.44 mL) in THF (20 mL). Colour-
1
the aggregation adopted by compound 1 is slightly pre- less blocks. Yield 2.70 g (2.89 mmol, 83%). H NMR (200.1 MHz,
4222
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 4219–4224

Pentanuclear Organozinc Hydroxylamides – Coordination Modes and Numbers
SHORT COMMUNICATION
CDCl3): δ = 0.53 [sept, 3JH,H = 7.5 Hz, 4 H, 4×ZnCH(CH3)2], 1.25
[d, JH,H = 7.5 Hz, 24 H, 4×ZnCH(CH3)2], 2.70 [s, 36 H,
[6]
[7]
N. W. Mitzel, C. Lustig, Angew. Chem. 2001, 113, 4521; Angew.
Chem. Int. Ed. 2001, 40, 4390.
a) N. W. Mitzel, C. Lustig, M. Woski, Dalton Trans. 2004, 397;
b) H.-D. Hausen, G. Schmöger, W. Schwarz, J. Organomet.
Chem. 1978, 153, 271.
M. Ullrich, N. W. Mitzel, K. Bergander, R. Fröhlich, Dalton
Trans. 2006, 714.
N. Voiculescu, Appl. Organomet. Chem. 2002, 16, 569.
N. A. Bell, H. M. M. Shearer, C. B. Spencer, Acta Crystallogr.,
Sect. C 1984, 40, 613.
a) V. A. Kuksa, S. M. S. V. Wardell, P. K. T. Lin, R. A. Howie,
Acta Crystallogr., Sect. E 2002, 58, m68; b) R. Reichenbach-
Klinke, M. Zabel, B. König, Dalton Trans. 2003, 141.
J. E. Walker, D. M. Howell, Inorg. Synth. 1967, 9, 2, and refer-
ences therein.
S. Jana, R. Fröhlich, N. W. Mitzel, Chem. Eur. J. 2006, 12, 592.
Crystallographic details: Single-crystals of compounds 1–3
were mounted under inert perfluoro-polyether at the tip of a
glass fibre and cooled in the cryostream of the diffractometer.
The crystallographic data set of 1 was collected at 173(2) K
with a Nonius Turbo CAD4 diffractometer; the data sets of 2
and 3 were measured at 198(2) K with a Nonius KappaCCD
(both operated with graphite-monochromated Mo-Kα radia-
tion; λ = 0.71073 Å). Data collection and reduction was per-
formed with Express[23a] and MolEN[23b] (1), or with Col-
lect[23c] and Denzo-SMN[23d] (2 and 3). The structures were
solved by direct methods and refined against F2 by full-matrix
least-squares with SHELXS-97 (1)[23e] and SHELXL-97 (2 and
3).[23f] In the case of 2, absorption correction was performed
with SORTAV.[23g] CCDC-609256 (1), CCDC-609255 (2) and
CCDC-617869 (3) contain the supplementary crystallographic
data for this paper. These data can be obtained free of charge
from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif. 1: (C16H48N6O6Zn5): Mr
= 747.47, monoclinic, space group C2/c, a = 12.910(1), b =
11.851(2), c = 20.324(2) Å, β = 90.02(1)°, V = 3109.2(7) Å3, Z
= 4, ρcalcd. = 1.597 gcm–3, F(000) = 1536, µ(Mo-Kα) =
3.837 mm–1. A total of 4837 reflections were measured in the
3
6×N(CH3)2] ppm. 13C{1H} NMR (50.3 MHz, CDCl3): δ = 11.6
[4 C, 4×ZnCH(CH3)2], 24.6 [8 C, 4×ZnCH(CH3)2], 52.9 [12 C,
6×N(CH3)2] ppm. MS (EI = 70 eV): m/z (%) = 648 (100) {M +
H –(iPr) –[(iPr)ZnONMe2]}+. C24H64N6O6Zn5 (859.66): calcd. C
33.53, H 7.50, N 9.78; found C 33.24, H 7.43, N 9.65.
[8]
[9]
[10]
3: From Et2Zn (19.0 mmol, 2.35 g) in n-hexane (20 mL), and Et2-
NOH (22.9 mmol, 2.04 g, 2.35 mL) in THF (20 mL). Colourless
blocks. Yield 3.36 g (3.46 mmol, 91%). 1H NMR (599.8 MHz, [11]
3
298 K, C6D5CD3): δ = 0.37, 0.37 [2×(q, JH,H = 8.1 Hz, 4 H,
2×ZnCH2CH3)], ca. 0.85–1.25 [m, very br., 24 H,
4×Ncoord(CH2CH3)2], 1.25 [t, 3JH,H
=
7.2 Hz, 12 H,
[12]
3
2×Nuncoord(CH2CH3)2], 1.51, 1.53 [2×(t, JH,H = 8.1 Hz, 6 H,
2×ZnCH2CH3)], ca. 2.70–3.20 [m, very br., 24 H, 6×N(CH2-
[13]
[14]
1
CH3)2] ppm. H NMR (599.8 MHz, 213 K, C6D5CD3): δ = 0.40–
0.57 (m, 8 H, 4×ZnCH2CH3), 0.75, 0.86 (2×{t, br., 2×[3 H,
2×Nuncoord(CH2CH3)2]}), 1.18, 1.21, 1.27, 1.28 (4×{t, br., 2×[3
3
H, 4×Ncoord(CH2CH3)2]}), 1.67, 1.69 [2×(t, br., JH,H = 8.3 Hz, 6
H, 2×ZnCH2CH3)], 2.27 (q, br., 3JH,H = 6.1 Hz), 2.64 (q, br.), 2.69
3
3
(q, br., JH,H = 6.3 Hz), 2.81 (q, br., JH,H = 6.9 Hz), 2.83 (q, br.,
3JH,H = 6.9 Hz), 2.88 (q, br., JH,H = 6.4 Hz), 2.97 (q, br., JH,H
=
3
3
6.2 Hz), 3.08 (q, br.), 3.80 (m, br.) {8×(2 H, NcoordCH2CH3);
1×[4×2 H, 2×Nuncoord(CH2CH3)2]} ppm. 13C{1H} NMR
(50.3 MHz, 298 K, C6D6): δ = 3.4, 5.7 [2×(2 C, 2×ZnCH2CH3)],
ca. 11–17 [very br.,
8 C, 4×Ncoord(CH2CH3)2], 13.1 [4 C,
2×Nuncoord(CH2CH3)2], 14.1, 14.4 [2×(2 C, 2×ZnCH2CH3)], ca.
51–61 {very br., 12 C, 4×[2 C, 2×Ncoord(CH2CH3)2]; 1×[4 C,
2×Nuncoord(CH2CH3)2]} ppm. 13C{1H} NMR (150.8 MHz, 213 K,
C6D5CD3): δ = 2.4, 5.4 [2×(2 C, 2×ZnCH2CH3)], 12.3, 13.0 [2×(2
C, 2 ×ZnCH2CH3)], 13.6, 13.8, 13.9, 14.2, 14.7 {4×[2 C,
2×Ncoord(CH2CH3)2]; 1×[4 C, 2×Nuncoord(CH2CH3)2]}, 51.8,
53.8, 55.1, 56.4, 61.2 {4×[2 C, 2×Ncoord(CH2CH3)2]; 1×[4 C,
2×Nuncoord(CH2CH3)2]} ppm. C32H80N6O6Zn5 (971.87): calcd. C
39.55, H 8.30, N 8.65; found C 38.93, H 8.07, N 8.41.
range 4.00° Յ 2θ Յ 53.88°, of which 3101 were unique (Rint
=
0.041). Final R indices: R1 = 0.034 [I Ͼ 2σ(I)], wR2 = 0.084
(all data); max./min. residual electron density 0.89/–0.37 e·Å–3.
2: (C24H64N6O6Zn5·C4H10O): Mr = 933.78, orthorhombic,
Acknowledgments
space group Pccn,
a = 19.870(1), b = 12.360(1), c =
Financial support from the Deutsche Forschungsgemeinschaft is
gratefully acknowledged. The authors are indebted to the NRW
Graduate School of Chemistry at Münster, GSC-MS (Ph.D. sti-
pend grant for M. U.), and to Dr. K. Bergander and Dr. A. Hepp
for conducting the NMR measurements.
17.052(1) Å, V = 4187.9(5) Å3, Z = 4, ρcalcd. = 1.481 gcm–3,
F(000) = 1960, µ(Mo-Kα) = 2.867 mm–1. A total of 21700 re-
flections were measured in the range 3.88° Յ 2θ Յ 56.52°, of
which 5139 were unique (Rint = 0.028). Final R indices: R1 =
0.031 [I Ͼ 2σ(I)], wR2 = 0.074 (all data); max./min. residual
electron density 0.64/–0.54 e·Å–3. 3: (C32H80N6O6Zn5): Mr =
971.87, monoclinic, space group Cc, a = 16.480(1), b =
11.595(1), c = 25.073(1) Å, β = 108.54(1)°, V = 4542.4(5) Å3,
Z = 4, ρcalcd. = 1.421 gcm–3, F(000) = 2048, µ(Mo-Kα) =
2.644 mm–1. A total of 9839 reflections were measured in the
[1] D. C. Bradley, R. C. Mehrotra, I. P. Rothwell, A. Singh, Alkoxo
and Aryloxo Derivatives of Metals, Academic Press, London,
San Diego, 2001.
[2] A. V. Firth, J. C. Stewart, A. J. Hoskin, D. W. Stephan, J. Or-
ganomet. Chem. 1999, 591, 185.
range 6.04° Յ 2θ Յ 55.88°, of which 9386 were unique (Rint
=
0.071). Final R indices: R1 = 0.071 [I Ͼ 2σ(I)], wR2 = 0.199
(all data); max./min. residual electron density 1.83/–1.12 e·Å–3.
This research area mainly encompasses pentanuclear Cu2+ spe-
cies; see a) J. J. Bodwin, V. L. Pecoraro, Inorg. Chem. 2000, 39,
3434; b) J. A. Halfen, J. J. Bodwin, V. L. Pecoraro, Inorg. Chem.
1998, 37, 5416; c) B. Kurzak, E. Farkas, T. Glowiak, H.
Kozlowski, J. Chem. Soc., Dalton Trans. 1991, 163, and refer-
ences cited therein.
[3] a) J. Lewin´ski, W. Marciniak, J. Lipkowski, I. Justyniak, J. Am.
Chem. Soc. 2003, 125, 12698; b) M. L. von Ziegler, J. Weiss,
Angew. Chem. 1970, 82, 931; Angew. Chem. Int. Ed. 1970, 9,
905.
[4] Note that within the metal alkoxides of group 13 (especially
those of Al), a strong tendency of the metal/metal complex
fragment to be bound by at least one additional donor unit is
common and gives the metal a CN Ͼ4 (see ref.[1]). This is only
suppressed if ligands of elevated steric demand are incorpo-
rated (see ref.[2]). Nevertheless, this tendency is mainly due to
the Lewis-acidity and/or oxophilicity of the group 13 elements
and cannot be accounted for by means of core shell saturation
demands, as it is the case with Zn.
[15]
[16]
[17]
For example, see N. Lalioti, C. P. Raptopoulou, A. Terzis, A. E.
Aliev, S. P. Perlepes, I. P. Gerothanassis, E. Manessi-Zoupa,
Chem. Commun. 1998, 1513.
3
Compare the heterocubane [µ -(9-BBN-9-O)(ZnEt)]4: S. Lulin´-
ski, I. Madura, J. Serwatowski, J. Zachara, Inorg. Chem. 1999,
38, 4937 with [µ2-(Mes2B–O)(ZnEt)]2: R. Anulewicz-Ostrow-
ska, S. Lulin´ski, E. Pindelska, J. Serwatowski, Inorg. Chem.
2002, 41, 2525.
[5] a) N. W. Mitzel, C. Lustig, M. Woski, Z. Naturforsch. 2003,
58b, 363; b) J. R. Jennings, K. Wade, J. Chem. Soc. A 1967,
1333; c) I. Pattison, K. Wade, J. Chem. Soc. A 1968, 2618.
Eur. J. Inorg. Chem. 2006, 4219–4224
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
4223

M. Ullrich, R. J. F. Berger, C. Lustig, R. Fröhlich, N. W. Mitzel
SHORT COMMUNICATION
[18] For clarity, the single steps of isomerisation are discussed as if
they occurred one after the other. By logic, a nearly simulta-
neous course of bond formation and disconnection is most
likely to exist.
[19] In principle, the whole process of isomerisation of the hetero-
atom array into itself would consist of six such periods (if the
single atoms of a same element were labelled distinguishable).
After one period, a molecule equal by connectivity, but of in-
verted stereochemistry would be furnished. Thus, after three
periods, the starting molecule would be reattained in its enan-
tiomeric form. Conversion of this ent-molecule back to its orig-
inal would then call for another three periods.
Flachsmann, L. Micouin, M. Oestreich, P. Knochel, Chem.
Eur. J. 2000, 6, 2748; c) H. Soroos, M. Morgana, J. Am. Chem.
Soc. 1944, 66, 893.
[22] N. W. Mitzel, B. A. Smart, S. Parsons, H. E. Robertson,
D. W. H. Rankin, J. Chem. Soc., Perkin Trans. 2 1996, 2727.
[23] a) Nonius B. V., 1994; b) K. Fair, Enraf Nonius B. V., 1990;
c) Nonius B. V., 1998; d) Z. Otwinowski, W. Minor, Methods
Enzymol. 1997, 276, 307; e) G. M. Sheldrick, Acta Crystallogr.,
Sect Sect. A 1990, 46, 467; f) G. M. Sheldrick, Universität
Göttingen, 1997; g) R. H. Blessing, Acta Crystallogr., Sect
Sect. A 1995, 51, 33; R. H. Blessing, J. Appl. Crystallogr. 1997,
30, 421.
[20] Supporting Information for computational details is available
(see footnote on the first page of this article).
Received: May 30, 2006
Published Online: September 14, 2006
[21] a) I. D. Gridnev, J. M. Serafimov, H. Quiney, J. M. Brown, Org.
Biomol. Chem. 2003, 1, 3811; b) A. Boudier, C. Darcel, F.
4224
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 4219–4224