Merocyanine-like chromophores as ligands for catalytic active metals of group VIII

Article abstract of DOI:10.1002/ejic.200600787

Nickel and palladium complexes that contain merocyanine-like 4H-imidazoles as ligands have been synthesized in good yields. The complexes were characterized by elemental analyses, mass spectrometry, ¹H NMR, ¹³C NMR spectroscopy and in the solid state by single crystal diffraction analyses. The new palladium complexes of type 4 can be regarded as new functional dyes as well as deeply coloured and redox active metallacycles that display catalytic activity. In contrast to parent compounds 2, the palladacycles of type 4 show two well-separated reductions that are fully reversible. The values of semiquinone formation constants of derivatives 4 (between 10¹¹ and 10¹²) are only somewhat lower than those reported for similar boracycles. Wiley-VCH Verlag GmbH & Co. KGaA, 2007.

Full text of DOI:10.1002/ejic.200600787

FULL PAPER
DOI: 10.1002/ejic.200600787
Merocyanine-Like Chromophores as Ligands for Catalytic Active Metals of
Group VIII
Jörg Blumhoff,[a] Rainer Beckert,*[a] Dirk Walther,[b] Sven Rau,[b] Manfred Rudolph,[b]
Helmar Görls,[b] and Winfried Plass[b]
Keywords: N ligands / Chromophores / Palladium / Nickel / Electrochemistry
Nickel and palladium complexes that contain merocyanine-
like 4H-imidazoles as ligands have been synthesized in good
yields. The complexes were characterized by elemental
analyses, mass spectrometry, 1H NMR, 13C NMR spec-
troscopy and in the solid state by single crystal diffraction
analyses. The new palladium complexes of type 4 can be re-
garded as new functional dyes as well as deeply coloured
and redox active metallacycles that display catalytic activity.
In contrast to parent compounds 2, the palladacycles of type
4 show two well-separated reductions that are fully revers-
ible. The values of semiquinone formation constants of deriv-
atives 4 (between 1011 and 1012) are only somewhat lower
than those reported for similar boracycles.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
Introduction
Nitrogen-containing ligands based on oxalic amidines 1
have recently attracted increasing interest.[1–3] These bifunc-
tional amidines can act as diamide or diamidinate ligands
after deprotonation and the ligands can be easily modified
in their electronic and steric effects by variation of the sub-
stituents on the nitrogen atoms. Moreover, they can adopt
several possible coordination modes in their metal com-
plexes. During the last decade we focussed our interests on
the development of new cyclic versions of oxalic amidines.
Thus, amidines without hydrogen in the α-position, such as
benzamidines, can easily be cyclized with bis(imidoyl chlo-
rides) derived from oxalic acid in order to yield 4H-imid-
azoles 2 (Figure 1).[4] A new class of chromophores/fluoro-
phores was developed employing these diazafulvene
imines.[5,6] In addition, the 4H-imidazoles 2 are reversible
two-electron redox systems which have been optimized to
powerful electron-accepting derivatives in the form of bo-
ratetraazapentalenes.[7] Their radical anions are charac-
terized by an unusual thermodynamic stability as can be
seen from semiquinone formation constants KSEM in the
magnitude of 1015.
Figure 1. Oxalic amidines 1 and their cyclic versions 4H-imidazoles
2.
The 4H-imidazoles 2 can therefore be regarded as multi-
functional molecules in which the properties can easily be
tuned by variation of the aryl substituents. Compared with
derivatives of type 1, they possess the same substructure
which may serve as a donor site for metal complexation.
Because of the fact that open-chained oxalic amidines 1
have been successfully applied as ligands for cross-coupling
methods,[8] their cyclic versions “cycloamidines” 4H-imid-
azoles 2, should also be promising candidates. Zinc com-
plexes of 4H-imidazoles have previously been reported.[9]
Attempts to deprotonate derivatives 2 by lithium hydride[10]
yielded unusual lithium complexes that surprisingly contain
one molecule of water in their first complexation sphere.
Apart from porphyrins, phthalocyanines and related sys-
tems metal complexes in which the ligand is identical with
an efficient chromophore are quite rare. Our aim was there-
fore to combine the redox active chromophore of 4H-imid-
azoles 2 with the catalytic activity of selected d-metals.
[a] Institut für Organische und Makromolekulare Chemie der
Friedrich-Schiller-Universität,
Humboldtstr. 10, 07743 Jena, Germany
Fax: +49-3641-948212
E-mail: c6bera@uni-jena.de
[b] Institut für Anorganische und Analytische Chemie der Fried-
rich-Schiller-Universität,
Lessingstr. 8, 07743 Jena, Germany
Eur. J. Inorg. Chem. 2007, 481–486
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
481

R. Beckert et al.
FULL PAPER
Results and Discussion
Synthesis and Characterization
First we attempted to synthesize NiII complexes starting
from nickelII-(acac)2. This diketonato complex even reacts
at room temperature with 2a, the red solution changing to
a deep purple. In order to complete the reaction, the mix-
ture has to be heated under reflux for approx. 4 h (monitor-
ing by TLC). It is remarkable that the complex formation
takes place without addition of bases; most likely the acetyl-
acetonato moiety functioned as a base for the deproton-
ation of the NH function. The X-ray structural analysis ob-
tained from a single crystal of complex 3a revealed a molec-
ular structure as illustrated in Figure 2.
Scheme 1. Synthesis of complex 3a.
Because of the fact that palladium plays a vital role as a
catalytically active metal in cross-coupling methods, further
investigations were directed towards the synthesis of analo-
gous Pd complexes. As a starting metal complex, the easily
accessible allyl palladium(II) chloride dimer was applied.
Because of its tendency to dissociate in solution into two
coordinative unsaturated fragments, it enables easy com-
plexation of other ligands at the palladium(II) centre.
Furthermore, the allyl palladium-fragment preferably forms
complexes with an almost square-planar geometry showing
diamagnetism that allows a detailed structural elucidation
by NMR spectroscopy.
Even at room temperature the complex formation is vis-
ible by a colour change from red to purple. After heating
under reflux conditions, for 2–4 h, chromatographic purifi-
cation allowed the isolation of new complexes of type 4.
Elemental analysis, NMR spectroscopic data and MS data
confirmed the composition of a 1:1 complex in 4a. Accord-
ing to Scheme 2, the dimeric palladium species reacted with
one molecule of the ligands 2a–d to yield palladacycles 4a–
d. A high symmetry of the molecule was indicated by the
1H NMR spectrum of 4a, which showed only five signals
for the aromatic H-atoms in addition to the characteristic
coupling pattern of the allyl substructure. Moreover, the
13C-NMR spectra confirm the symmetry of the molecule
by a single set of signals. The signal for C-atom C4 is, com-
pared to the ligand (δ = 190 ppm), shifted to a lower field
in complex 4a (δ = 199 ppm). This down-field shift is com-
parable to that in the corresponding boratetraazapental-
enes,[7] thus indicating the change from the merocyanine to
the cyanine chromophore.
Figure 2. Molecular structure of complex 3a (H atoms have been
omitted for clarity). Selected bond lengths [Å]: Ni1–N1A 2.100(4),
Ni1–N2A 2.197(4), Ni1–N1B 2.072(4), Ni1–N2B 2.201(4), Ni1–O1
2.009(3), Ni1–O2 2.042(3).
In the new chelate complex 3a (Scheme 1) two 4H-imid-
azoles coordinate the nickelII centre, whereby the two re-
maining coordination sites are occupied by an acetylace-
tonato ligand. While one 4H-imidazole is coordinated in
the deprotonated form, the other is bound as a neutral li-
gand. This fact could be verified by determining the Ni–N
distances of the coordinating nitrogen atoms. This molecule
is in agreement with the stoichiometry of the reaction. The
octahedral nickel centre in 3a exhibits paramagnetism and
therefore, no NMR-data could be obtained. Characteriza-
tion of complex 3a via mass spectroscopy confirmed the
results of the X-ray analysis. A high-resolution mass spec-
trum verified the chemical composition of compound 3a
(C55H30F24N8NiO2). In addition, the complex 3a shows a
characteristic fragmentation due to the split-up of the li-
gands.
Scheme 2. Synthesis of complexes 4a–d.
482
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 481–486

Merocyanine-Like Chromophores as Ligands
FULL PAPER
A single-crystal X-ray structure analysis of 4a allowed an Electrochemical Measurements of Complexes 4
unambiguous assignment of these compounds, as shown in
The electrochemical reduction of the palladacycles 4 oc-
Figure 3. According to our presumptions, one 4H-imid-
curs in two one-electron steps. A typical series of cyclic
azole is coordinated at the palladium(II) centre. Being a sec-
square-wave voltammograms measured with square-wave
ond ligand, the allyl-anion is η3-bonded and causes square-
frequencies from 25 to 800 Hz is shown in Figure 5. The
planar coordination geometry in view to the palladium(II)
half-wave potentials extracted from these measurements are
centre and shows C2 symmetry. In analogy to the synthesis
summarized in Table 1.
of the bis(3,5-trifluoromethyl) derivative 4a, complexes of
4H-imidazoles could be prepared that possess electron-do-
nating substituents. The various aryl substituents have no
significant influence on the complexation ability of the cor-
responding 4H-imidazole. Interestingly, all new synthesized
Pd complexes of type 4 show high stabilities as solids as
well as in solution. They can be handled under normal at-
mospheric conditions and tolerate slightly acidic and basic
conditions. In comparison with the 4H-imidazoles the UV/
Vis maxima of the palladium complexes are shifted to
higher wavelengths (2a: λmax = 460, 515 nm; 4a: λmax = 549,
590 nm) as shown in Figure 4. Furthermore, complex 4a
shows a weak fluorescence at 623 nm (THF).
Figure 5. Cyclic square-wave voltammograms of the reduction of
1.5 m of 4b in DMF/tetrabutylammonium hexafluorophosphate
at a mercury drop electrode using square-wave frequencies of 800,
400, 200, 100, 50 and 25 Hz. Uref is the potential of the ferrocen-
ium/ferrocene couple in the same electrolyte.
Table 1. Half-wave potentials and semiquinone formation con-
stants of complexes 4a–c.
1/2
1/2
Complex
U1
U2
Ksem
4b
4a
4c
–1.460 V
–1.015 V
–1.455 V
–2.175 V
–1.670 V
–2.175 V
1.26·1012
1.26·1011
1.58·1012
Figure 3. Molecular structure of complex 4a. Selected bond lengths
[Å]: Pd–N1 2.128(4), Pd–N1A 2.128(4), Pd–C2 2.120(5), Pd–C2A
2.120(5), Pd–C1 2.109(9).
If only the first reduction process is measured (by revers-
ing the potential scan before the second reduction takes
place), the experimental curves agree very well with those
simulated for a fast (fully reversible) charge-transfer process
followed by a slow first-order decomposition reaction. This
means, as far as the charge transfer steps are concerned, a
deviation from Nernstian behaviour cannot be detected up
to square-wave frequencies of approx. 1000 Hz. However,
the one-fold reduced species is chemically not fully stable in
the solvent system. A second, (much faster) follow-up reac-
tion is induced upon scanning over a potential range in-
cluding the second reduction step. Interestingly, this follow-
up reaction has a much larger effect on the re-oxidation
wave associated with the second charge transfer reaction.
The ratio between cathodic and anodic peak current re-
mains close to 1 for the first charge transfer step even when
using relatively low square-wave frequencies such as shown
in Figure 6 (f = 25 Hz). In the case of 4b an almost perfect
agreement between experimental and simulated curves was
obtained on the basis of the reaction shown in Scheme 3.
Figure 4. UV/Vis spectra of 2a (···) and 4a (__) in THF.
Eur. J. Inorg. Chem. 2007, 481–486
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
483

R. Beckert et al.
FULL PAPER
heating under reflux, 60% of the substrate was converted
into cinnamic derivative 5. Prolonged reaction times led to
a nearly quantitative conversion.
Figure 6. Cyclic square-wave voltammogram of the reduction of
1.5 m of 4b in DMF/tetrabutylammonium hexafluorophosphate
at a mercury drop electrode using a square-wave frequency of
25 Hz. Uref is the potential of the ferrocenium/ferrocene couple in
the same electrolyte.
Scheme 4. The Heck coupling as a model reaction for the determi-
nation of the catalytic activity of 4a.
On the basis of these experimental findings the subject
of further research will be the synthesis of homo- and het-
erobimetallic complexes of 4H-imidazoles and studies di-
rected to the electronic communication between both metal
centres. Applications of molecular architectures similar to
those described above could prove to be light-driven reac-
tions and catalyses.
Experimental Section
General Comments: All reactions were monitored by TLC, carried
out on 0.25-mm Merck TLC aluminium sheets (aluminium oxide
150 F254 neutral) using UV light. H and 13C NMR spectra were
1
Scheme 3. Reaction scheme for an almost perfect agreement be-
tween experimental and simulated curves.
recorded with a Bruker DRX 400 (400 MHz) or a Bruker AC 250
(250 MHz) spectrometer. UV/Vis spectra were recorded with a Per-
kin–Elmer Lambda 19 spectrophotometer. MS spectra were taken
from measurements on a Finnigan MAT SAQ 710 mass spectrome-
ter. Elemental analyses were carried out in-house with an auto-
matic analyzer LECO CHNS 932.
We assume that the twofold reduced species reacts to give
an intermediate from which the onefold reduced species is
formed in a subsequent reaction; the data determined for
4c are virtually identical. In the case of 4a, the first re-
duction process is accompanied by a pre-wave. Since the
latter could not be reproduced by digital simulation, only
the half-wave potentials are reported in Table 1.
The lack of success of the Sonogashira and Suzuki reac-
tions with these cross-conjugated heterocycles can now be
explained due to the fact that the 4H-imidazoles act as very
efficient chelating ligands for palladium. This means that
required catalytic amounts of palladium were immobilized
immediately by complexation and consequently no catalytic
activity was observed. However, the palladacycles them-
selves are able to catalyze other types of cross-coupling re-
actions as shown by the Heck reaction.
Synthesis of the Nickel(II) Complex 3a: In a 100-mL one-necked
round flask, equipped with a magnetic stirrer, 2a (1 mmol) was dis-
solved in THF (50 mL). Then, bis(acetylacetonato)nickel(II)
(1 mmol) dissolved in THF (10 mL) was added dropwise at room
temperature. A darkening of the reaction mixture indicated a fast
complexation reaction. The reaction mixture was heated to comple-
tion under reflux for 3–5 h. HRMS of 3a: C55H31F24N8NiO2: M
= 1349.15401113 gmol–1 (calculated), M = 1349.15494000 gmol–1
(found).
Synthesis of the Palladium(II) Complexes 4a–d: In a 100-mL one-
necked round flask, equipped with a magnetic stirrer, 1 mmol of
the corresponding 4H-imidazole 2a was dissolved in THF (50 mL).
Then, allylpalladium(II) chloride dimer (0.5 mmol) dissolved in
THF (10 mL) was added dropwise at room temperature. The reac-
tion mixture was heated to completion under reflux for 3–5 h and
the progress of the reaction was monitored by TLC. For complex
4a, the solvent was removed in vacuo and the residue was sus-
pended in acetone. The resulting precipitate was filtered off and
dried in vacuo. The complexes 4b–d were purified by column
chromatography (alumina, activity V, toluene).
In an initial experiment we used the substrates 4-bromo-
acetophenone and acrylic acid n-butylester in dimethylacet-
amide as the solvent at 120 °C, yielding the 4-substituted
cinnamic acidester 5 (Scheme 4) to evaluate the catalytic ac-
tivity of palladium complex 4a. This complex belongs to
the very rare group of Pd-catalysts with N,N-chelate ligands
1
Complex 4a: Yield: 587 mg (79%). H NMR (CD2Cl2, 250 MHz):
that are active in this C–C-coupling method. After 2 h of δ = 8.34 (m, 2 H, aryl-H), 8.16 (s, 4 H, aryl-H), 7.59 (m, 3 H, aryl-
484 Eur. J. Inorg. Chem. 2007, 481–486
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Merocyanine-Like Chromophores as Ligands
FULL PAPER
H), 7.45 (m, 2 H, aryl-H), 5.50 (m, 1 H, allyl C-H), 3.55 (d, J =
7.5 Hz, 2 H, allyl CH2), 3.03 (d, J = 12.5 Hz, 2 H, allyl CH2) ppm.
15.8755(2), b = 20.4987(3), c = 32.8158(4) Å, β = 101.482(1)°, V =
10465.4(2) Å3, T = –90 °C, Z = 2, ρcalcd. = 1.475 gcm–3, µ (Mo-Kα)
13C NMR (CD2Cl2, 100 MHz): δ = 199.0, 178.8, 149.1, 134.9, = 10.77 cm–1, F(000) = 4714, 60125 reflections in h(–18/20), k(–26/
132.3, 131.5 [q, 2J(C,F) = 33.2 Hz], 130.9, 128.6, 125.8, 123.3 [q,
1J(C,F) = 272.7 Hz], 119.2, 116.1, 62.3 ppm. FAB-MS (NBA): m/z pleteness Θmax = 99.3%, 23821 independent reflections, Rint
(%) = 743 [M]+, 702 [M – allyl]+, 597 [M – allyl – Pd + H]+. UV/
0.030, 18727 reflections with Fo Ͼ 4σ(Fo), 1272 parameters, 0 re-
Vis (THF): λmax (lgε) = 549 (4.0), 590 nm (4.0). C28H16F12N4Pd straints, R1obs = 0.087, wR2obs = 0.271, R1all = 0.111, wR2all = 0.289,
26), l(–42/36), measured in the range 1.18° Յ Θ Յ 27.47°, com-
=
(742.86): calcd. C 45.27, H 2.17, N 7.54; found C 45.25, H 2.16, N
7.45.
GOOF = 2.184, largest difference peak and hole: 3.707/–1.610
e Å–3.
Crystal Data for 4a:[15] C28H16F12N4OPd·H2O, Mr
=
1
Complex 4b: Yield: 274 mg (55%). H NMR (CD2Cl2, 250 MHz):
759.86 gmol–1, red-brown prism, size 0.04ϫ0.04ϫ0.03 mm, mo-
noclinic, space group C2/m, a = 18.7198(8), b = 18.9056(8), c =
9.8808(2) Å, β = 120.749(2)°, V = 3005.29(19) Å3, T = –90 °C, Z
= 4, ρcalcd. = 1.679 gcm–3, µ (Mo-Kα) = 7.22 cm–1, F(000) = 1500,
10750 reflections in h(–24/24), k(–22/24), l(–12/12), measured in the
range 1.66° Յ Θ Յ 27.50°, completeness Θmax = 99.4%, 3540 inde-
pendent reflections, Rint = 0.039, 2996 reflections with Fo Ͼ 4σ(Fo),
δ = 8.33 (m, 2 H, aryl-H), 7.51 (m, 7 H, aryl-H), 7.21 (m, 4 H,
aryl-H), 5.63 (m, 1 H, allyl-CH), 3.62 (d, J = 6.9 Hz, 2 H, allyl-
CH2), 3.16 (d, J = 12.4 Hz, 2 H, allyl-CH2), 2.40 (s, 6 H, CH3) ppm.
Micro-ESI-MS (methanol): m/z (%) = 499 [M + H]+. UV/Vis
(THF): λmax (lgε) = 560 (4.0), 600 nm (4.0). C26H24N4Pd (498.92):
calcd. C 62.59, H 4.85, N 11.23; found C 62.55, H 4.80, N 11.31.
1
Complex 4c: Yield: 291 mg (50%). H NMR (CD2Cl2, 250 MHz):
226 parameters, 0 restraints, R1obs = 0.064, wR2obs = 0.186, R1all
=
δ = 8.35 (m, 2 H, aryl-H), 7.57 (m, 5 H, aryl-H), 7.43 (m, 6 H,
aryl-H), 5.60 (m, 1 H, allyl-CH), 3.64 (d, J = 6.9 Hz, 2 H, allyl-
CH2), 3.15 (d, J = 12.4 Hz, 2 H, allyl-CH2), 1.38 (s, 18 H,
CH3) ppm. Micro-ESI-MS (methanol): m/z (%) = 582 [M + H]+.
UV/Vis (THF): λmax (lgε) = 563 (4.1), 604 nm (4.1). C32H36N4Pd
(583.08): calcd. C 65.92, H 6.22, N 9.61; found C 65.90, H 6.18, N
9.55.
0.077, wR2all = 0.203, GOOF = 1.055, largest difference peak and
hole: 2.860/–0.943 e Å–3.
Cyclovoltammetric Instrumentation and Procedures: Cyclic square-
wave voltammetry was conducted with the 3-electrode technique
using a home-built computer-controlled instrument based on the
PCI 6110 data acquisition board (National Instruments). The ex-
periments were performed in DMF containing 0.25- tetrabu-
tylammonium hexafluorophosphate (Merck, Darmstadt) under a
blanket of solvent-saturated argon. The DMF was purified by stir-
ring with NaH, distilled under reduced pressure and stored over
molsieves 4A under argon.
1
Complex 4d: Yield: 362 mg (65%). H NMR (CD2Cl2, 250 MHz):
δ = 8.22 (m, 2 H, aryl-H), 7.61 (m, 4 H, aryl-H), 7.39 (m, 3 H,
aryl-H), 6.64 (m, 4 H, aryl-H), 5.53 (m, 1 H, allyl-CH), 3.56 (d, J
= 6.9 Hz, 2 H, allyl-CH2), 3.09 (d, J = 12.4 Hz, 2 H, allyl-CH2),
2.91 [s, 12 H, N(CH3)2] ppm. Micro-ESI-MS (methanol): m/z (%)
= 556 [M]+. UV/Vis (THF): λmax = 630, 673 nm. C28H30N6Pd
(557.01): calcd. C 60.38, H 5.43, N 15.09; found C 60.40, H 5.48,
N 15.05.
The Ohmic resistance that had to be compensated was determined
by measuring the impedance of the system at potentials where the
Faradayic current was negligibly small. Background correction was
accomplished by subtracting the current curves of the blank elec-
trolyte (containing the same concentration of supporting electro-
lyte) from the experimental curves.
Heck Reaction Catalyzed by Complex 4a: In a typical experiment,
4-bromoacetophenone (6.25 mmol) and anhydrous sodium acetate
(7 mmol) were placed in a 25-mL two-necked flask equipped with
a stirring bar, reflux condenser and septum. The flask was degassed
under vacuum and purged with argon to ensure an inert reaction
atmosphere. Then, N,N-dimethylacetamide (10 mL) as the solvent
and n-butyl acrylate (0.5 g) were added through the septum. The
reaction mixture was thoroughly stirred and heated to the appro-
priate reaction temperature at which it was held for 5 min followed
by treatment with the catalyst solution containing complex 4a
(1.25 µmol in 0.5 mL of THF). After the appropriate time intervals
samples (0.5 mL) of the reaction mixture were hydrolyzed with 2 
HCl (0.5 mL) and extracted with CH2Cl2 (0.5 mL). The organic
phase was dried with K2CO3 and stored at –21 °C until GC analy-
sis for determination of the yield.
Working electrode: hanging mercury drop (mHg–drop = 3.95–
4.00 mg) produced by the CGME instrument (Bioanalytical Sys-
tems Inc., West Lafayette, USA). A platinum wire served as the
counter electrode. Internal reference electrode: Ag/AgCl electrode
in acetonitrile containing 0.25  tetrabutylammonium chloride.
However, for convenience all potentials reported in this paper
refer to the ferrocenium/ferrocene couple. The latter was
measured at the end of each experiment using a 1.5-mm Pt disk
electrode. The evaluation of the experiments was accomplished by
subjecting the experimental voltammograms to a nonlinear re-
gression procedure using the freely available DigiElch software
(htpp://www.DigiElch.de).
Crystal Structure Determination: The intensity data for the com-
pounds were collected with a Nonius KappaCCD diffractometer,
using graphite-monochromated Mo-Kα radiation. Data were cor-
rected for Lorentz and polarization effects, but not for absorp-
tion.[11,12] The structures were solved by direct methods
(SHELXS[13]) and refined by full-matrix least-squares techniques
against Fo2 (SHELXL-97[14]). The hydrogen atoms of the structures
were included at calculated positions with fixed thermal param-
eters. All non-hydrogen atoms were refined anisotropically.[14] XP
(Siemens Analytical X-ray Instruments, Inc.) was used for structure
representations.
Acknowledgments
This work was supported by Deutsche Forschungsgemeinschaft
(SFB 436).
[1] a) M. Döhring, R. Beckert, H. Görls, Z. Anorg. Allg. Chem.
1994, 620, 551–560; b) P. Fehling, M. Döring, F. Knoch, R.
Beckert, H. Görls, Chem. Ber. 1995, 128, 405–412; c) M.
Döring, P. Fehling, H. Görls, W. Imhof, R. Beckert, D. Lin-
dauer, J. Prakt. Chem./Chem.-Ztg. 1999, 341, 748–756.
[2] a) M. Ruben, S. Rau, A. Skirl, K. Krause, H. Görls, D. Wal-
ther, J. G. Vos, Inorg. Chim. Acta 2000, 303, 206–214; b) T.
Döhler, H. Görls, D. Walther, Chem. Commun. 2000, 945–946;
c) M. Ruben, D. Walther, R. Knake, H. Görls, R. Beckert, Eur.
Crystal Data for 3a:[15] 2C55H31F24N8NiO2·C50H82Ni8O30·
3CH2Cl2·H2O, Mr
0.05ϫ0.05ϫ0.04 mm, monoclinic, space group P21/c,
=
4649.27 gmol–1, red-brown prism, size
a
=
Eur. J. Inorg. Chem. 2007, 481–486
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
485

R. Beckert et al.
FULL PAPER
J. Inorg. Chem. 2000, 1055–1060; d) D. Walther, M. Stollenz,
L. Böttcher, H. Görls, Z. Anorg. Allg. Chem. 2001, 627, 1560–
1570; e) D. Walther, M. Stollenz, H. Görls, Organometallics
2001, 20, 4221–4229; f) M. Stollenz, M. Rudolph, H. Görls, D.
Walther, J. Organomet. Chem. 2003, 687, 153–160.
[3] M. Deng, Y. Yao, Y. Zhang, Q. Shen, Chem. Commun. 2004,
2742–2743.
[4] J. Atzrodt, J. Brandenburg, C. Käpplinger, R. Beckert, W.
Günther, H. Görls, J. Fabian, J. Prakt. Chem./Chem.-Ztg. 1997,
339, 729–734.
[10] J. Atzrodt, R. Beckert, K. Nordhoff, M. Bräuer, E. Anders, H.
Görls, Eur. J. Org. Chem. 1998, 2557–2563.
[11] COLLECT, Data Collection Software, Nonius B. V., Nether-
lands, 1998.
[12] Z. Otwinowski, W. Minor, “Processing of X-ray Diffraction
Data Collected in Oscillation Mode”, in Methods in Enzy-
mology, vol. 276, Macromolecular Crystallography, Part A
(Eds.: C. W. Carter, R. M. Sweet), Academic Press, San Diego,
1997, pp. 307–326.
[13] G. M. Sheldrick, Acta Crystallogr., Sect. A 1990, 46, 467–473.
[14] G. M. Sheldrick, SHELXL-97, University of Göttingen, Ger-
many, 1993.
[5] J. Fabian, H. Görls, R. Beckert, J. Atzrodt, J. Prakt. Chem./
Chem.-Ztg. 1997, 339, 735–741.
[6] J. Atzrodt, R. Beckert, W. Günther, H. Görls, Eur. J. Org.
Chem. 2000, 1661–1668.
[7] T. Gebauer, R. Beckert, D. Weiß, K. Knop, C. Käpplinger, H.
Görls, Chem. Commun. 2004, 1860–1861.
[8] S. Rau, K. Lamm, H. Görls, J. Schöffel, D. Walther, J. Or-
ganomet. Chem. 2004, 689, 3582–3592.
[15] CCDC-614437 (for 3a) -614438 (for 4a) contain the supple-
mentary crystallographic data for this paper. These data can
be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Received: August 25, 2006
Published Online: November 27, 2006
[9] J. Atzrodt, R. Beckert, H. Görls, Heterocycles 1991, 51, 763–
783.
486
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 481–486