Hydrogen bonds as structural directive towards unusual polynuclear complexes: synthesis, structure, and magnetic properties of copper(II) and nickel(II) complexes with a 2-aminoglucose ligand

Article abstract of DOI:10.1002/chem.200800670

The reaction of benzyl 2-amino-4,6-O-benzylidene-2-deoxy-α-D- glucopyranoside (HL) with the metal salts Cu(ClO₄) ₂·6H₂O and Ni-(NO₃)₂· 6H₂O affords via self-assembly a tetranuclear μ₄- hydroxido bridged copper(II) complex [(μ₄-OH)Cu ₄(L)₄-(MeOH)₃(H₂O)](ClO ₄)₃ (1) and a trinuclear alcoholate bridged nickel(II) complex [Ni₃(L)₅(HL)]NO₃ (2), respectively. Both complexes crystallize in the acentric space group P2₁. The X-ray crystal structure reveals the rare (μ₄-OH)Cu₄O ₄ core for complex 1 which is μ₂-alcoholate bridged. The copper(II) ions possess a distorted square-pyramidal geometry with an [NO₄] donor set. The core is stabilized by hydrogen bonding between the coordinating amino group of the glucose backbone and the benzylidene protected oxygen atom O4 of a neighboring {Cu(L)} fragment as hydrogen-bond acceptor. For complex 2 an [N₄O₂] donor set is observed at the nickel(II) ions with a distorted octahedral geometry. The trinuclear isosceles Ni₃ core is bridged by μ₃-alcoholate O3 oxygen atoms of two glucose ligands. The two short edges are capped by μ₂-alcoholate O3 oxygen atoms of the two ligands coordinated at the nickel(II) ion at the vertex of these two edges. Along the elongated edge of the triangle a strong hydrogen bond (244 pm) between the O3 oxygen atoms of ligands coordinating at the two relevant nickel(II) ions is observed. The coordinating amino groups of the these two glucose ligands are involved in additional hydrogen bonds with O4 oxygen atoms of adjacent ligands further stabilizing the trinuclear core. The carbohydrate backbones in all cases adopt the stable ⁴C₁ chair conformation and exhibit the rare chitosan-like trans-2,3-chelation. Temperature dependent magnetic measurements indicate an overall antiferromagnetic behavior for complex 1 with J₁ = -260 and J₂ = -205 cm^-1 (g = 2.122). Compound 2 is the first ferromagnetically coupled trinuclear nickel(II) complex with J_A = 16.4 and J_B = 11.0 cm^-1 (g_1,2 = 2.183, g₃ = 2.247). For the high-spin nickel(II) centers a zero-field splitting of D_1,2 = 3.7 cm^-1 and D₃ = 1.8 cm^-1 is observed. The S = 3 ground state of complex 2 is consistent with magnetization measurements at low temperatures.

Full text of DOI:10.1002/chem.200800670

FULL PAPER
DOI: 10.1002/chem.200800670
Hydrogen Bonds as Structural Directive towards Unusual Polynuclear
Complexes: Synthesis, Structure, and Magnetic Properties of Copper(II) and
Nickel(II) Complexes with a 2-Aminoglucose Ligand
Anja Burkhardt, Eike T. Spielberg, Sascha Simon, Helmar Gçrls, Axel Buchholz, and
Winfried Plass*[a]
Abstract: The reaction of benzyl 2-
amino-4,6-O-benzylidene-2-deoxy-a-d-
glucopyranoside (HL) with the metal
ment as hydrogen-bond acceptor. For
complex 2 an [N4O2] donor set is ob-
served at the nickel(II) ions with a dis-
torted octahedral geometry. The trinu-
clear isosceles Ni3 core is bridged by
m3-alcoholate O3 oxygen atoms of two
glucose ligands. The two short edges
are capped by m2-alcoholate O3 oxygen
atoms of the two ligands coordinated
at the nickel(II) ion at the vertex of
these two edges. Along the elongated
edge of the triangle a strong hydrogen
bond (244 pm) between the O3 oxygen
atoms of ligands coordinating at the
two relevant nickel(II) ions is ob-
served. The coordinating amino groups
of the these two glucose ligands are in-
volved in additional hydrogen bonds
with O4 oxygen atoms of adjacent li-
gands further stabilizing the trinuclear
core. The carbohydrate backbones in
salts
(NO3)2·6H2O affords via self-assembly
tetranuclear m4-hydroxido bridged
copper(II) complex [(m4-OH)Cu4(L)4-
(MeOH)3A(H2O)](ClO4)3 (1) and a tri-
Cu(ClO4)2·6H2O
and
Ni-
4
E
all cases adopt the stable C1 chair con-
a
formation and exhibit the rare chito-
san-like trans-2,3-chelation. Tempera-
ture dependent magnetic measure-
ments indicate an overall antiferromag-
netic behavior for complex 1 with J1 =
ꢀ260 and J2 =ꢀ205 cmꢀ1 (g=2.122).
Compound 2 is the first ferromagneti-
cally coupled trinuclear nickel(II) com-
plex with JA =16.4 and JB =11.0 cmꢀ1
(g1,2 =2.183, g3 =2.247). For the high-
spin nickel(II) centers a zero-field split-
ting of D1,2 =3.7 cmꢀ1 and D3 =1.8 cmꢀ1
is observed. The S=3 ground state of
complex 2 is consistent with magnetiza-
tion measurements at low tempera-
tures.
A
N
ACHTUNGTRENNUNG
nuclear alcoholate bridged nickel(II)
complex [Ni3(L)5(HL)]NO3 (2), respec-
tively. Both complexes crystallize in the
acentric space group P21. The X-ray
crystal structure reveals the rare (m4-
OH)Cu4O4 core for complex 1 which is
m2-alcoholate bridged. The copper(II)
ions possess a distorted square-pyrami-
dal geometry with an [NO4] donor set.
The core is stabilized by hydrogen
bonding between the coordinating
amino group of the glucose backbone
and the benzylidene protected oxygen
atom O4 of a neighboring {Cu(L)} frag-
Keywords: carbohydrates · copper ·
hydrogen
bonds
·
magnetic properties · nickel
Introduction
lizing enzymes have been revealed to function with alkaline
earth and transition-metal ions in the active sites.[2] Al-
though the importance of sugar–metal interaction has been
known for many years,[3] information on the metal-binding
of saccharide moieties is still rather limited. Only for the
past two decades, sugars received a growing interest as
ligand components in coordination chemistry because of
their stable chiral scaffold, supramolecular arrangement via
hydrogen bonding, and polyfunctionality.[3,4] In particular
the latter aspect is of interest related to the field of high-nu-
clearity transition-metal complexes which have been attract-
ing continuous interest due to their relevance for multimetal
active sites of metalloproteins[5] as well as their importance
in the field of molecular magnetism.[6]
Carbohydrates are the most abundant biogenic class of com-
pounds involved in a wide range of functions in living organ-
isms, for example, monosaccharide fragments are participat-
ing in glycolipids and glycoproteins.[1] Many sugar-metabo-
[a] Dr. A. Burkhardt, E. T. Spielberg, S. Simon, Dr. H. Gçrls,
Dr. A. Buchholz, Prof. Dr. W. Plass
Institut fꢀr Anorganische und Analytische Chemie
Friedrich-Schiller-Universitꢁt Jena, Carl-Zeiss-Promenade 10
07745 Jena (Germany)
Fax : (+49)3641-948-132
E-mail: sekr.plass@uni-jena.de
Chem. Eur. J. 2009, 15, 1261 – 1271
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1261

From the coordination chemistry point of view, carbohy-
drates are suitable chelate ligands functionalized with a se-
quence of weak donor sites. The vicinal functional groups
feature several donor atoms capable of forming stable com-
plexes with transition metals[7] and offer several possibilities
to introduce additional donor groups into the sugar back-
bone. Well-known modification strategies are N-glycosyla-
tion of polyamines[8,9] and nucleophilic substitution of bro-
moethyl-O-glycosides.[10] TEMPO oxidation of the secon-
dary hydroxyl group at C6 results in the formation of uronic
acids. The corresponding carbohydrate-derived salicylidene
hydrazides are obtained via a three-step synthesis and form
noside (HL) was obtained in a three-step synthesis accord-
ing to published procedures starting from N-acetyl-a-d-glu-
copyranosamine as depicted in Scheme 1.[23] In the first two
steps the glycosidic hydroxyl group of the anomeric carbon
atom C1 and the hydroxyl groups at C4 and C6 are protect-
ed as benzyl ether and cyclic benzylidene acetal, respective-
ly. The following alkaline hydrolysis of the amide function
at C2 affords the bidentate 2-aminoglucose ligand HL. Due
to the hydrophobic character of the phenyl groups HL is
soluble in less polar solvents such as chloroform. The exclu-
sive presence of the a-anomer of the saccharide unit, associ-
ated with the cis position of the protons attached to C1 and
C2, is confirmed by the resonance of the proton on C1 of
+
mononuclear VO2 complexes.[11] On the other hand, con-
3
densation of amino saccharides and carbonyl components
readily leads to tridentate Schiff-base ligands.[12,13] These
Schiff-base ligands are based on either C1 or C6 amino
functionalized carbohydrate fragments.[13] In the case of
their copper(II) complexes varying nuclearities and architec-
tures are observed, where the structure of the central Cun
unit (n=2,3,4) can be influenced by the positions of the che-
lating donor atoms at the sugar backbone. Linear alcohol-
ate–acetate-bridged trinuclear copper(II) compounds were
obtained from glucopyranose derivatives carrying an imino-
or ketoenamine residue at C6, whereas 5-ketoenamine-sub-
stituted glucofuranoses as well as Schiff-base derivatives of
1- or 2-aminoglucopyranoses yield di- or tetranuclear alco-
holate-bridged complexes.[14–16]
the d-glucose moiety which leads to a doublet with J12 =
1
3.7 Hz in the H NMR spectrum of HL.
Scheme 1. Synthesis of ligand HL starting from N-acetyl-a-d-glucopyra-
nosamine.
In this context we started to explore the coordination
chemistry of 2-aminoglucopyranoses as chitosan-like chelate
ligands.[17–20] In particular, their copper(II) compounds show
a considerable tendency to self-aggregate affording oxoido-
bridged oligonuclear complexes with unexpected magnetic
properties induced by hydrogen bonding between the consti-
tuting glucose ligands.[17] The formation of oxygen bridged
polynuclear compounds is a general feature for transition-
metal complexes with carbohydrate derivatives as li-
gands,[21,8] which is usually prevented by the presence of ap-
propriate coligands to coordinatively saturate the metal ion.
Nevertheless very little is known about relevant carbohy-
drate-based polynuclear complexes.[22]
Herein we present the synthesis and characterization of
polynuclear copper(II) and nickel(II) complexes derived
from a 2-aminoglucose ligand with the rare chitosan-like
trans-2,3-chelation by the sugar backbone. The molecular
structures are directed by the possible hydrogen bonding
theme given by the ligand system. In the case of copper(II)
this leads to a tetranuclear complex with an unusual (m4-
OH)Cu4O4 core, whereas for nickel(II) the rare case of a
isosceles triangular Ni3 core is obtained. To our knowledge,
the obtained nickel(II) complex is the first trinuclear exam-
ple that exhibits ferromagnetic interactions.
For the synthesis of the copper(II) complex a methanol
solution of copper(II) perchlorate hexahydrate is treated
with a solution of the ligand HL in chloroform in a 1:1
molar ratio at room temperature. Slow evaporation of the
solvent leads to blue prismatic crystals [(m4-OH)Cu4(L)4-
AHCTUNGTRENUNG(N MeOH)3ACHTUNRTEGG(NNUN H2O)]CATHNUGTRNEN(GUN ClO4)3 (1) containing additional solvent
molecules of crystallization. This is confirmed by TGA
measurements of freshly isolated crystals. Variation of the
reaction conditions by addition of a stoichiometric amount
of triethylamine leads the same tetranuclear copper(II) com-
plex 1, but without significant increase in yield. Crystals of
better X-ray quality have been obtained by the base-free
method. The electronic absorption spectrum of complex 1 in
chloroform solution exhibits a charge-transfer band at
278 nm and a d–d transition band at 696 nm. To the best of
our knowledge complex 1 is the first tetranuclear copper(II)
complex with m4-hydroxido bridged Cu4 core without sup-
port of a macrocyclic ligand.[24]
The reaction of nickel(II) nitrate hexahydrate with two
equivalents of the 2-aminoglucose ligand HL in methanol
solution in the presence of two equivalents of aqueous po-
tassium hydroxide leads to the trinuclear complex
[Ni3(L)5(HL)]NO3 (2). Single crystals containing additional
solvent molecules of crystallization which are suitable for X-
ray structure analysis could be obtained by slow evaporation
of the solvent. This is consistent with the TGA measure-
ments of freshly isolated crystals. The electronic absorption
spectrum of complex 2 in chloroform solution shows two
Results and Discussion
Synthesis and characterization: The 2-aminoglucose ligand
benzyl 2-amino-4,6-O-benzylidene-2-deoxy-a-d-glucopyra-
1262
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1261 – 1271

Unusual Polynuclear Complexes
FULL PAPER
low energy bands at 643 and 1131 nm which are typical for
octahedral high-spin nickel(II) ions.[25,26] To our knowledge 2
is the first trinuclear nickel(II) complex based on a carbohy-
drate ligand system.
Structure description: Crystals of complexes 1 and 2 suitable
for X-ray crystallography have been grown by slow evapora-
tion of the employed solvent. For both complexes the ob-
tained crystals contain methanol and water molecules of
crystallization which are located on disordered positions.
For complexes 1 and 2 this leads to the crystal compositions
[(m4-OH)Cu4(L)4ACHTUNGTRENNUNG(MeOH)3ACHUTNRTGENG(NUN H2O)]ACTHUNGTREN(NUGN ClO4)3·2.7MeOH
(1·2.7MeOH) and [Ni3(L)5(HL)]NO3·3.25MeOH·0.75H2O
(2·3.25MeOH·0.75H2O), respectively. Both compounds crys-
tallize in the monoclinic space group P21 consistent with the
exclusive presence of one enantiomeric form. Details of the
crystal structure determination are summarized in the Ex-
perimental Section.
Copper(II) complex 1: The molecular structure of the cation
of compound 1 is depicted in Figure 1. Selected bond
lengths and angles as well as interatomic distances are listed
in Table 1. The tetranuclear copper(II) complex consists of
four {Cu(L)} building blocks which are linked by a central
m4-OH group. All four copper(II) ions possess [NO4] coordi-
nation environments with a slightly distorted square-pyrami-
dal geometry as indicated by the corresponding t values
which range from 0.15 to 0.32 (t=0 for a square pyramid,
t=1 for a trigonal bipyramid).[27] This is in accordance with
the small deviations of 17 to 25 pm observed for the cop-
per(II) ions from their basal coordination planes given by
the C2 amino nitrogen atom and the C3 alcoholate oxygen
atom of the sugar backbone, an additional C3 alcoholate
oxygen atom of the ligand of an adjacent {Cu(L)} moiety,
and the oxygen atom O1 of the central m4-OH bridge. The
Figure 1. Molecular
structure
of
the
cation
[(m4-OH)-
Cu4(L)4(MeOH)3(H2O)]3+ of complex 1. Solvent molecules, hydrogen
atoms, and perchlorate counterions are omitted for clarity. Disordered
phenyl rings are displayed in one position. Top: Cation with intramolecu-
lar hydrogen bonds indicated by dashed lines. Pertinent distances:
N1···O24 291, N2···O34 291, N3···O44 290, N4···O14 276 pm. Bottom:
Representation of the tetranuclear m4-OH bridged core of 1 with the co-
ordination environment of the copper(II) centers.
deprotonated oxygen atoms Oi3 of the chelating glucose
backbone are the m2-bridging links between two copper(II)
ions (i is the running number of the {Cu(L)} fragments). The
square planes of the four {Cu(L)} fragments are rotated ap-
proximately 908 with respect to their neighbors which are
linked by sharing two adjacent edges of the square planes.
This is supported by intramolecular hydrogen bonds be-
tween the amino nitrogen atom and the oxygen atom at C4
of neighboring {Cu(L)} building blocks. The relevant distan-
ces Ni···Oj4 vary from 276 to 291 pm (i and j are the running
numbers of the adjacent {Cu(L)} fragments). The glucose
backbones of all four {Cu(L)} building blocks adopt the gen-
Table 1. Selected bond lengths, interatomic distances [pm], and Cu-O-Cu
bridging angles [8] for 1.
Distances
ꢀ
ꢀ
Cu1 O1
211.8(4)
200.5(5)
194.0(4)
191.2(4)
223.3(4)
221.7(4)
201.4(4)
190.7(4)
189.8(4)
216.2(5)
291.68(10)
419.29(14)
290.41(10)
Cu2 O1
216.0(4)
199.6(5)
192.9(4)
190.5(4)
217.7(4)
211.8(4)
199.2(5)
191.3(4)
192.8(4)
221.9(5)
295.63(10)
407.67(12)
292.55(10)
ꢀ
ꢀ
Cu1 N1
Cu2 N2
ꢀ
ꢀ
Cu1 O13
Cu2 O23
ꢀ
ꢀ
Cu1 O23
Cu2 O33
ꢀ
ꢀ
Cu1 O1M
Cu2 O2M
ꢀ
ꢀ
Cu3 O1
Cu4 O1
ꢀ
ꢀ
Cu3 N3
Cu4 N4
ꢀ
ꢀ
Cu3 O33
Cu4 O13
ꢀ
ꢀ
4
Cu3 O43
Cu4 O43
erally more stable C1 chair conformation.
ꢀ
ꢀ
Cu3 O3W
Cu4 O4M
The slight variation in the coordination geometry of the
copper(II) ions in the four {Cu(L)} building blocks can be
attributed to differences in their axial coordination. The
apical position at Cu1, Cu2, and Cu4 is occupied by a meth-
anol molecule. Whereas in the case of Cu3 a water molecule
is axially bound to the metal center. All bond lengths and
angles of complex 1 are in the expected range. The overall
Cu1···Cu2
Cu1···Cu3
Cu1···Cu4
Cu2···Cu3
Cu2···Cu4
Cu3···Cu4
Cu-O-Cu bridging angles
Cu1-O1-Cu2
Cu1-O1-Cu3
Cu1-O1-Cu4
Cu1-O13-Cu4
Cu1-O23-Cu2
85.97(14)
150.5(2)
86.55(15)
97.84(17)
98.80(16)
Cu2-O1-Cu3
Cu2-O1-Cu4
Cu3-O1-Cu4
Cu2-O33-Cu3
Cu3-O43-Cu4
84.97(14)
144.7(2)
84.85(13)
101.69(17)
99.74(17)
charge of the complex cation [(m4-OH)Cu4(L)
4ACHTUNGTNER(NUNG MeOH)3-
AHCTUNGTRENNUNG
Chem. Eur. J. 2009, 15, 1261 – 1271
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1263

W. Plass et al.
ent. Except for Cl2P (see Figure 2), the perchlorate ions are
highly disordered at several crystallographic positions. Nev-
ertheless, the actual protonation of the central bridging
oxygen atom O1 and the axial oxygen ligands at the cop-
per(II) ions can not be directly determined by X-ray crystal-
lography. We therefore performed bond-valence sum (BVS)
calculations[28] to confirm both the overall charge of the
complex cation and the protonation state of the oxygen
atoms coordinated at the copper ions. The corresponding
data are given in Table 2.
non-protonated situation. In the case of the amine nitrogen
atoms Ni the inclusion of the attached hydrogen atoms at
their preset distances, as used in the crystal structure solu-
tion, gives values in good agreement with their trivalent
character. A similar situation is observed for the axially co-
ordinated oxygen atoms at the copper(II) ions, for which the
primarily obtained BVS values clearly indicate that an addi-
tional protonation is present. This is consistent with three
methanol (O1M, O2M, and O4M) and one water molecule
(O3W). Also for central m4-bridging oxygen atom the pri-
marily obtained BVS value clearly indicates the presence of
an additional proton leading to a hydroxido group at O1,
which is consistent with an electron density picks in the dif-
ference Fourier map, even though a hydrogen atom could
not be refinement at this position. The consequently penta-
coordinated oxygen atom O1 of the central m4-bridging hy-
droxido group is displaced by 59 pm from the mean plane
defined by the four copper(II) ions located at the corners of
the basal plane of the square-pyramidal coordination envi-
ronment.
The m4-OH bridge observed in complex 1 is a rare struc-
tural feature, which to the best of our knowledge thus far
has solely been reported for complexes with Robson-type
macrocyclic ligands.[24] The majority of m4-oxygen bridged
copper(II) complexes contain a central oxoido group with a
tetrahedral coordination sphere.[29] Nevertheless, the by far
more common structural motif for tetranuclear copper(II)
compounds is a cubane-like Cu4O4 core,[30] which thus far
has been exclusively observed in the case of complexes with
sugar-based ligands.[15–18]
Figure 2. Hydrogen bonding for 1·2.7MeOH. All atoms of the complex
cation except those involved in hydrogen bonding, coordinated to cop-
per(II), and the carbon atoms of the carbohydrate backbone chelate
rings are omitted for clarity. Hydrogen bonds are represented as dashed
lines. Pertinent distances: O1M···O3LA 268, O2M···O41P 282,
O3W···O2L 272, O3W···O33P 323, O3W···O34P 314, O4M···O23P 303,
N1···O24 291, N1···O12P 301, N2···O13P 317, N2···O24 291, N3···O1L 300,
N3···O44 290, N4···O11P 305, N4···O14 276, O1L···O11P 279, O2L···O23P
307, O2L···O24P 309, O3L···O31P 282, O3L···O32P 299 pm.
The crystal structure of 1·2.7MeOH is characterized by
extensive hydrogen bonding and intermolecular p–p interac-
tion of the benzyl substituents at the C1 position. The hy-
drogen bonding associated with the cation [(m4-OH)Cu4(L)4-
(H2O)]3+ of complex 1 is depicted in Figure 2. The
ACHUTNGERN(NUG MeOH)3ACHTUNGTRENNUGN
interactions of the axial ligands at the copper(II) ions to-
gether with two of the perchlorate counterions, one of which
is highly disordered (Cl3P) and solvent molecules lead to a
hydrogen-bonding network forming one-dimensional chains
with a two-fold screw symmetry along the crystallographic
[010] direction. The third perchlorate ion is located on two
disordered positions in the cavity below the Cu4 plane of the
complex cation, opposite to the central m4-OH group O1.
This cavity is surrounded by the benzyl groups of the glu-
cose fragments and provides additional space for a methanol
molecule (O1L) besides the disordered perchlorate ion.
Table 2. BVS values for selected atoms in the crystal structure of 1;
values including hydrogen atoms at preset distances used in the solution
of the crystal structure are given in parentheses.[a]
O1
1.12 (2.12)
0.27 (2.12)
1.18 (2.18)
1.10 (2.10)
1.23 (2.23)
1.95
N1
N2
N3
N4
Cu1
Cu2
Cu3
Cu4
1.35 (3.25)
1.31 (3.20)
1.36 (3.25)
1.38 (3.28)
1.90
1.94
1.93
1.94
O3W
O1M
O2M
O4M
O13
O23
O33
1.96
1.99
O43
1.96
[a] BVS values are calculated according to reference [28]. BVS=
ꢀiexp[(r0,iꢀri)/B] with B=37 pm, r0,Cu O =167.9 pm, r0,Cu N =161.0 pm,
Nickel(II) complex 2: Figure 3 shows the molecular struc-
ture of the cation [Ni3(L)5(HL)]+ of complex 2, which is
present in crystals of 2·3.25MeOH·0.75H2O. Selected bond
lengths and the Ni-O-Ni bridging angles are summarized in
Table 3. The [N2O4] donor set provided by the chelating 2-
aminoglucose ligands leads to a distorted octahedral coordi-
nation geometry for all three nickel(II) ions. The distortion
from the ideal octahedral geometry is best described by the
observed bite angles of the chelate ligands ranging from 81
to 858 and the trans angles at the nickel(II) centers of 154 to
ꢀ
ꢀ
r0,C O =139.0 pm, r0,C N =147.0 pm, r0,O H =95.0 pm, and r0,N H =90.0 pm.
ꢀ
ꢀ
ꢀ
ꢀ
The BVS values for the central copper(II) ions as well as
the m2-bridging alkoxido oxygen atoms Oi3 are in good
agreement with the expected value of two. As the hydrogen
atoms in the crystal structure of 1 could not be found and
refined, the primarily obtained BVS values correspond to a
1264
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1261 – 1271

Unusual Polynuclear Complexes
FULL PAPER
The nickel(II) ions of the trinuclear complex 2 are ar-
ranged as nearly isosceles triangle with two almost equiva-
lent Ni···Ni distances of 280 (Ni1···Ni2) and 283 pm
(Ni2···Ni3), and one elongated basal edge with 316 pm
(Ni1···Ni2). The isosceles triangle formed by the Ni3 core is
m3-bridged by the alcoholate O3 oxygen atoms of two 2-ami-
noglucose ligands coordinated to the nickel centers Ni1 and
Ni2. These m3-bridging oxygen atoms (O13 and O43) are lo-
cated about 127 pm above and below the Ni3 plane. Along
the equal sides of the isosceles triangle the nickel ions are
additionally linked by m2-bridging O3 oxygen atoms of two
ligand molecules coordinated to Ni3. This leads to a face
sharing of the central octahedron (Ni3) with the two termi-
nal octahedra (Ni1 and Ni2) and an edge sharing between
the two latter. Overall each nickel(II) ion is coordinated by
two bidentate ligand moieties through the C2 amino nitro-
gen atom and the C3 alcoholic oxygen atom of the glucose
backbone. Consequently, Ni1 and Ni2 are both coordinated
by an additional chelating ligand which is not involved in
any bridging of metal ions with the oxygen atoms O23 and
O33 terminally bound to Ni1 and Ni2, respectively. The car-
bohydrate backbones of all 2-aminoglucose ligands adopt
4
the stable C1 chair conformation.
Along the elongated basal edge of the isosceles triangle a
strong hydrogen bond with an O23···O33 separation of
244 pm is formed between the O3 oxygen atoms terminally
bound to Ni1 and Ni2 (see Figure 3). A somewhat larger dis-
tance is observed for a trinuclear nickel complex with a
comparable core structure supported by a methanol–pheno-
late interaction at a separation of 255 pm.[31] Nevertheless,
comparably short oxygen atom distances at about 244 pm
are also observed for hydrogen bonds between metal bound
alcohol and alcoholate groups (M-(R)O···H-O(R)-M) in
copper(II) complexes with amino alcohol ligands.[32] Addi-
tional intramolecular hydrogen bonding is observed for the
amino nitrogen atoms N2 and N3 of the two terminally
bound chelate ligands at Ni1 and Ni2, respectively. These ni-
trogen atoms form hydrogen bonds with the O4 oxygen
atoms of the remaining four 2-aminoglucose ligands with
distances in the range of 302 to 316 pm, further stabilizing
the trinuclear core (see Figure 3).
Trinuclear nickel(II) complexes with a Ni3 core motif sim-
ilar to that of complex 2 have been reported for phenolate
bridges between the nickel(II) atoms.[33] Moreover, this
structural motif has also been found for a hexanuclear nick-
el(II) complex with an amino alcohol ligand.[34] Although in
particular the coordination chemistry of N-glycosylamine li-
gands has been extensively studied in the past,[3,35] to our
knowledge complex 2 is the first trinuclear nickel(II) com-
pound based on a carbohydrate ligand system.
Figure 3. Molecular structure of the cation [Ni3(L)5(HL)]+ of complex 2.
Solvent molecules, hydrogen atoms, and the nitrate counterion are omit-
ted for clarity. Disordered phenyl rings are displayed in one position.
Top: Cation with intramolecular hydrogen bonds indicated by dashed
lines. Pertinent distances: N2···O44 302, N2···O64 307, N3···O14 297,
N3···O54 316, O23···O33 244 pm. Bottom: Representation of the trinu-
clear core of 2 with the coordination environment of the nickel(II) cen-
ters.
Table 3. Selected bond lengths, interatomic distances [pm], and Ni-O-Ni
bridging angles [8] for 2.
Distances
ꢀ
ꢀ
Ni1 N1
209.4(4)
212.1(3)
210.3(3)
209.1(3)
211.0(3)
212.6(2)
216.1(3)
214.1(3)
218.2(3)
280.41(7)
316.36(7)
Ni1 O23
205.2(3)
209.6(3)
201.9(3)
208.5(3)
208.0(2)
205.6(3)
222.2(3)
198.0(3)
199.1(3)
283.36(7)
ꢀ
ꢀ
Ni1 N2
Ni1 O43
ꢀ
ꢀ
Ni1 O13
Ni1 O63
ꢀ
ꢀ
Ni2 N3
Ni2 O33
ꢀ
ꢀ
Ni2 N4
Ni2 O43
ꢀ
ꢀ
Ni2 O13
Ni2 O53
ꢀ
ꢀ
Ni3 N5
Ni3 O43
ꢀ
ꢀ
Ni3 N6
Ni3 O53
ꢀ
ꢀ
Ni3 O13
Ni3 O63
Ni1···Ni3
Ni1···Ni2
Ni2···Ni3
Ni-O-Ni bridging angles
Ni1-O13-Ni2
Ni1-O13-Ni3
Ni1-O43-Ni2
Ni1-O43-Ni3
96.83(10)
81.73(8)
98.48(10)
80.92(8)
Ni1-O63-Ni3
88.74(9)
Ni2-O13-Ni3
Ni2-O43-Ni3
Ni2-O53-Ni3
82.25(10)
82.32(10)
89.16(11)
Structural directives in complexes with 2-aminoglucose li-
gands: For the copper(II) and nickel(II) complexes 1 and 2
unusual structures have been obtained. This can be attribut-
ed to the specific properties of the employed 2-aminoglu-
cose ligand HL, which leads to a controlled self-assembly of
the basic mononuclear building blocks {Cu(L)} and {Ni(L)2}.
1728. Overall the most significant distortion is observed for
the nickel center Ni3. Nevertheless, all bond lengths and
angles of complex 2 are within the expected range.
Chem. Eur. J. 2009, 15, 1261 – 1271
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1265

W. Plass et al.
It is well-known that in particular polynuclear copper(II)
complexes are governed by the presence of a Cu4O4 core.
Compounds with such a core structure have been classified
according their atomic distances within the core structure as
depicted in Figure 4.[30,36] Nearly all structures reported to
date fall into two classes, which are characterized according
their number of short and long Cu···Cu distances within the
Cu4O4 core as 2 + 4 class (or type I) and 4 + 2 class (or
type II) (see Figure 4), with the latter being by far the more
common of the two cases.
Figure 4. Tetranuclear cubane-like copper(II) complexes classified ac-
cording their Cu–O and Cu···Cu distances of the Cu4O4 core. Thick lines
represent short and dashed lines long Cu–O distances. Short Cu···Cu dis-
tances are indicated by a connecting line.
Utilizing the trans-2,3-chelating glucose backbone as che-
lating amino alcohol ligand to set up the mononuclear build-
ing blocks we have been able to selectively generate exam-
ples of either of the two major cubane classes as depicted in
Figure 5. The variation of the substitution at the C2 and C3
positions of the glucose backbone can be used to introduce
a specific coordinative and supramolecular presetting.
For a 2-aminoglucose imino functionalized at the C2 posi-
tion and C4/C6 protected by a cyclic benzylidene acetal a 4
+ 2 class core structure is obtained.[18] This can be rational-
ized by the possible weak coordination of the O4 oxygen
atom which is in an ideal position, due to the planarity of
the tridentate Schiff-base ligand and the rigidity of the glu-
cose backbone. This interaction is favored with respect to
the competing one with the m2-bridging C3 alkoxide oxygen
atom. Due to the lack of appropriate donors, the O4 oxygen
atom can not undergo any kind of hydrogen-bonding inter-
action.
On the other hand using the same ligand, but without pro-
tecting group at the O4 oxygen atom, this particular func-
tion becomes a potential hydrogen-bond donor. Conse-
quently, the resulting building block is capable of self-aggre-
gation through hydrogen bonding. Instead of the competing
situation as in the former case, the hydrogen-bonding inter-
actions of the O4 hydroxy groups synergistically enforce the
formation of the m3 bridges of the relevant C3 alkoxide
oxygen atoms. As a result a 2 + 4 class core structure is ob-
tained.[17]
Figure 5. Structural directives for copper(II) complexes with 2-aminoglu-
cose ligands. Variation of the 2-amino (free amine or Schiff base) and the
4-hydroxy (free or protected) functions at the backbone allows for the
specific design of the self-assembly of the tetranuclear core structure (see
text). Hydrogen bonds are indicated by dashed lines and weak coordina-
tive bonds at the axial positions by thin lines. In the structural representa-
tions the non-coordinating part of the carbohydrate moiety, except for
the chelate ring, is omitted for clarity.
ordinating ligand. An additional change is given by the de-
creased denticity of the chelate ligand. In the presence of an
appropriate coligand such bidentate amino alcohol ligands
are capable of supporting the common cubane-like Cu4O4
core structures.[30] However, also alternative motifs for the
assembly of a tetranuclear core with a central m4-bridge
have been reported for copper(II) building blocks. The most
prominent is the Cu4ACHTNUTRGNE(NUG m4-O) core with a tetrahedral assembly
of copper(II) ions around a central oxoido bridge.[29] The
second option is the rare case of a central m4-OH bridge
with a square-planar assembly of the copper(II) ions leading
to a (m4-OH)Cu4O4 core, which as yet has only been report-
ed for complexes supported by macrocyclic Schiff-base li-
gands.[24] The fact that ligand HL provides both a hydrogen-
bond acceptor (O4 group) and a hydrogen-bond donor
(amino group at C2) together with a rigid preset orientation
of these groups given by the glucose backbone, allows for
intramolecular hydrogen bonding between the {Cu(L)}
building blocks in a self-complementary fashion as depicted
in Figure 5. Consequently for complex 1 this leads to a (m4-
OH)Cu4O4 core.
If the protecting group at O4 is retained, but instead a
free 2-amino function is employed, the bidentate amino al-
cohol HL is obtained as chelate ligand. In this case the O4
group can only act as hydrogen-bond acceptor or in the ab-
sence of an appropriate hydrogen-bond donor as weakly co-
The specific hydrogen-bonding features of the 2-amino-
glucose ligand HL are ideally suited to support the rare case
1266
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1261 – 1271

Unusual Polynuclear Complexes
FULL PAPER
of a (m3-OR)2ACHTUNGTRENNUNG(m2-OR)2Ni3 core structure with an isosceles
Ni3 triangle (see Figure 6). The two ligands terminally che-
lating at the Ni1 and Ni2 centers seem to be of particular
*
*
Figure 7. Plots of cM
( ) and cMT ( ) vs. T for 1. The solid lines represent
the calculated values (see text). The spin topology of the employed two-J
model is depicted as inset.
perature values of cMT are in agreement with an S=0
ground state for complex 1, whereas the observed increase
of the cM values for low temperatures are indicative for the
presence of paramagnetic impurities.
Figure 6. Hydrogen bonding for 2 supporting the trinuclear (m3-OR)2(m2-
OR)2Ni3 core structure. Hydrogen bonds are indicated by dashed lines.
All atoms except the chelating part of the glucose backbone as well as
the C4 and O4 atoms are omitted for clarity.
According the rather similar intramolecular Cu···Cu dis-
tances ranging from 290 to 296 pm (see Table 1) a spin top-
ology with equal copper(II) ions can be assumed for com-
plex 1. We therefore tested a one-J model with a tetragonal
spin topology and equal interactions between the metal cen-
ters and a two-J model with additional diagonal interactions.
Neither model yielded an acceptable fit of the experimental
data. Based on the observed structural inequivalence of the
four copper(II) ions of the (m4-OH)Cu4O4 core of complex 1
an alternative spin topology is suggested. This inequivalence
is due to the variation of the apical ligands at the square-
pyramidal copper(II) centers. For Cu3 this is a water ligand,
whereas for the others a methanol molecule is coordinated
(see Figure 1). This leads to a kite topology for the tetranu-
clear core which can be expressed as a two-J model given in
Figure 7. The corresponding spin Hamiltonian with JA =J12 =
J14 and JB =J23 =J34 is given in Equation (1).
importance for the support of the core structure. Both
amino groups (N2 and N3) are oriented towards one of the
O4 oxygen atoms of the adjacent {Ni(L)2} building blocks,
that is, N2 (Ni1) is hydrogen bound to O44 (Ni2) and O64
(Ni3), whereas N3 (Ni2) is hydrogen bound to O14 (Ni1)
and O54 (Ni3) (for relevant distances see Figure 3). As a
consequence, the two O3 oxygen atoms (O23 and O33) of
these terminally bound chelate ligands are oriented towards
each other which makes the relevant M-(R)O···O(R)-M
fragment susceptible for protonation.[32] The resulting strong
hydrogen bond with a distance of 244 pm caps the elongated
basal edge of the isosceles triangle (see Table 3).
Magnetic properties: As yet neither for copper(II) com-
plexes with a (m4-OH)Cu4O4 core, as observed for complex
1, nor for nickel(II) complexes with a (m3-OR)2Ni3 core, as
observed for complex 2, temperature dependent magnetic
measurements have been reported in the literature. For
complexes 1 and 2 magnetic susceptibility measurements
were performed for air-dried powdered samples in the tem-
perature range from 2 to 300 K. The molar paramagnetic
susceptibilities cM are corrected for the diamagnetic contri-
butions.
The magnetic data for complex 1 are depicted in Figure 7
as temperature dependent plots of cM and cMT. At room
temperature a cMT value of 0.77 cm3 Kmolꢀ1 is observed.
This is far from the expected spin-only value for four inde-
pendent S=1/2 centers per molecule, which is indicative for
strong antiferromagnetic interactions within the (m4-
OH)Cu4O4 core of complex 1. The continuous decrease of
the cMT values with decreasing temperature further supports
the presence of antiferromagnetic interactions. The low-tem-
4
X
!
^
^ ^
^ ^
^ ^
^ ^
^
H ¼ ꢀJAðS1S2 þ S1S4ÞꢀJBðS2S3 þ S3S4Þ þ
gbSi H
ð1Þ
i¼1
A fit of the magnetic data of complex 1 based on the kite
topology yields the parameters g=2.122, J1 =ꢀ260 and J2 =
ꢀ205 cmꢀ1. The graph for the best fit of the data is repre-
sented in Figure 7. Although this spin topology leads to an
excellent agreement with the experimental data, the two
coupling constants J1 and J2 can not unambiguously be as-
signed to JA and JB based on structural features. Attempts to
introduce additional diagonal interactions (J13 and J24), re-
sulting in a four-J model, did not lead to an improvement of
the fit quality. In fact, fitting of the data with this extended
model indicated a dependence between the two new param-
eters.
Chem. Eur. J. 2009, 15, 1261 – 1271
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1267

W. Plass et al.
These results indicate predominate interactions along the
edges of the tetragonal (m4-OH)Cu4O4 core with four mixed
hydroxido- and alkoxido-bridged Cu2ACTHNUTRGENNGU(m-OH)ACHTUNTGREN(NUNG m-OR) frag-
ments. It is therefore tempting to interpret the data in terms
of the well-known magnetostructural correlations described
for dinuclear copper(II) complexes with either a m-hydroxi-
do or a m-alkoxido bridged Cu2O2 core.[37] Nevertheless,
taking the linear relationship established for m-hydroxido
bridges (J=7270–74.53a) and the very small Cui-O1-Cuj
bridging angles, ranging from 84.7 to 86.78, strong ferromag-
netic interactions would be predicted, which is in strong con-
tradiction to the observations for complex 1. Here it should
be mentioned that the underlying unusually small Cu-O-Cu
bridging angles are enforced by the m4-hydroxido bridged
core structure, which in turn also leads to an elongation of
the relevant Cu–OH distances. Moreover, a considerable
distortion of the square-planar coordination geometry at all
four copper centers is observed.
Instead assuming a predominate role of the m-alkoxido
bridges within the Cu2O2 cores, with Cui-Oj3-Cuj bridging
angles ranging from 98.2 to 101.68, strong antiferromagnetic
interactions in the order of ꢀ200 to ꢀ500 cmꢀ1 would be
predicted (J=7857–82.1a). Although this is the correct sign,
these values are still considerably too large. Nevertheless, it
is well-known from dinuclear systems that a hinge distortion
can considerably reduce the antiferromagnetic character of
the relevant copper-copper coupling constant.[38] In fact,
DFT calculations on dinuclear model systems indicate that a
hinge distortion with a dihedral angle of 1408 considerably
reduces the antiferromagnetic coupling (up to several hun-
dreds of wave numbers). Moreover, this effect is enhanced
at about the same order of magnitude by the so-called out-
of-plane distortion of the substituent on the bridging oxygen
atom at an angle of 358. Interestingly, for complex 1 similar-
ly large distortions are observed, with dihedral angles in the
order of approximately 1608 and out-of-plane distortions of
about 308. Although this gives a consistent qualitative pic-
*
*
Figure 8. Top: Plots of cM
( ) and cMT ( ) vs T for 2. The solid lines rep-
resent the calculated values (see text). The spin topology of the em-
ployed two-J model is depicted as inset. Bottom: Field-dependent mag-
netization for 2 at 2 K ( ). The Brillouin function for S=3 and the ap-
propriate g values is plotted as solid line (see text).
*
axial symmetry for the nickel centers Ni1 and Ni2, as both
possess
a similar local coordination environment (see
ture, the two bridges present in the Cu
G
G
Table 3).
fragments may nevertheless counterbalance their effects,
leading to a moderately antiferromagnetic coupling along
the edges.
^
^ ^
^ ^
^ ^
H ¼ ꢀJAðS1S3 þ S2S3ÞꢀJBS1S2
3
3
X
X
ð2Þ
1
3
!
gibSiH
2
^
^
The cM and cMT plots of the magnetic data obtained for
complex 2 are depicted in Figure 8. The molar susceptibility
cM continuously increases with decreasing temperature with-
out reaching a maximum in the investigated temperature
range, whereas the cMT curve goes through a maximum at
10 K upon decreasing the temperature, indicating a ferro-
magnetic interaction between the nickel(II) centers within
the trinuclear core. The decrease of the cMT values below
10 K can be attributed to zero-field splitting (ZFS) of the
nickel(II) ions.[19,26]
According to the (m3-OR)2Ni3 core structure of complex 2,
representing an isosceles triangle, a two-J model is em-
ployed as spin topology (see Figure 8). The corresponding
spin Hamiltonian with JA =J13 =J23 and JB =J12 is given in
Equation (2) and includes ZFS for the nickel(II) centers.
The latter was assumed to be equal with an approximate
þ
Di½Sz,iꢀ SiðSi þ 1Þꢁ
i¼1
i¼1
The best fit is obtained with the parameters g1,2 =2.183, g3 =
2.247, JA =16.4, JB =11.0 cmꢀ1, D1,2 =3.7, D3 =1.8 cmꢀ1 and
represented in Figure 8. The observed ferromagnetic cou-
pling within the Ni3 core is consistent with magnetization
measurements at 2 K depicted in Figure 8, which confirm a
ferromagnetic S=3 ground state for complex 2.
Although the effect of the ZFS is only detectable at very
low temperatures, preliminary simulations of W-band ESR
spectra of 2 recorded on powder samples at 10 K confirm
the magnitude of the obtained values of the ZFS parame-
ters. Moreover, these spectra even indicate a small rhombic-
ity of the ZFS contribution. A subsequent fit of the magnet-
ic susceptibility data including a rhombic ZFS leads to the
1268
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1261 – 1271

Unusual Polynuclear Complexes
FULL PAPER
1
2
ˆ
^
parameters E1,2/D1,2 =0.04, E3/D3 =0.09 (with HZFS =D[Sz ꢀ3
ing between the coordinating amino group of the glucose
backbone and the benzyl protected oxygen atom O4 of
neighboring building blocks. In the case of complex 2 a
strong hydrogen bond supported by the Ni-(R)O···H-O(R)-
Ni fragment is observed which is the result of an increased
basicity based on the preset steric orientation of the two un-
derlying terminal alkoxido groups.
For complex 1 antiferromagnetic interactions with a re-
sulting S=0 ground state are observed, whereas for complex
2 the observed interactions are all ferromagnetic leading to
an S=3 ground state. The nickel(II) complex 2 is the first
example of a ferromagnetically coupled trinuclear nickel(II)
complex. The magnetic interactions for both complexes can
be interpreted in terms of superexchange via the involved
oxygen bridges. Consequently, the steric presettings given by
the glucose backbone of the ligand have a strong influence
on the observed magnetic interactions.
2
2
^
^
S(S+1)+E/D (Sx ꢀSy )]). Nevertheless, this only leads to a
marginal improvement of the fit quality, which is a finger-
print of the herewith related problem of overparametriza-
tion when a full set of ZFS parameters is used for fitting
magnetic susceptibility data.
To the best of our knowledge complex 2 is the first trian-
gular nickel(II) complex with an overall ferromagnetic inter-
action within the Ni3 core. Nevertheless, for linear trinuclear
complexes ferromagnetic interactions between the adjacent
nickel(II) ions have been reported,[39] and date back to an
early paper of Ginsberg et al.,[40] which in most cases exhibit
an additional weak antiferromagnetic coupling between the
two terminal nickel(II) ions. Moreover, for triangular sys-
tems generally an antiferromagnetic interaction is observed,
independent of the molecular symmetry for isosceles[41] and
equilateral Ni3 triangles.[42]
However, for di-[43] and tetranuclear[44] nickel(II) com-
plexes magnetostructural correlations have been put for-
ward. These correlations indicate a Ni-O-Ni border line
angles of about 93.5 and 99.08 for the corresponding m2-O
and m3-O bridged di- and tetranuclear systems, respectively,
below which a ferromagnetic interaction is observed. Inter-
estingly, for complex 2 the Ni-(m2-O)-Ni angles are observed
at about 898, whereas the Ni-(m3-O)-Ni angles are found
within range from 82 to 998, which is consistent with the ob-
served overall ferromagnetic interactions within the isosce-
les Ni3 triangle.
Experimental Section
Physical measurements: Melting points are given uncorrected and were
determined via a VEB Analytik Dresden HMK 72/41555. Thermogravi-
metric analyses (TGA) for powdered samples were performed on a
NETZSCH STA409PC Luxx apparatus under a constant flow of nitrogen
ranging from room temperature up to 10008C with a heating rate of
18Cminꢀ1. Infrared and Raman spectra were recorded on a BRUKER
IFS 55 EQUINOX spectrometer. UV/Vis spectra were recorded on a
VARIAN CARY 5000 UV/Vis-NIR-spectrometer. Mass spectra were
measured on a Bruker MAT SSQ 710 spectrometer. Elemental analyses
were determined on a LECO CHNS/932 Analyser and a VARIO EL III.
Magnetic susceptibilities were obtained from a powdered sample of 1
and from a paraffin-triturated sample of 2 using a Quantum-Design
MPMSR-5S SQUID magnetometer equipped with a 5 Tesla magnet in
the range from 300 to 2 K with an applied magnetic field of 5000 Oe (for
details see ref. [45]). The experimental magnetic susceptibility data were
fitted using the program package julX version 1.3 which allows spin-
Hamiltonian simulations of the data by a full-matrix diagonalization ap-
proach and includes the treatment of paramagnetic impurities in the fit-
ting procedure.[46]
Conclusion
The 2-aminoglucose ligand benzyl 2-amino-4,6-O-benzyli-
dene-2-deoxy-a-d-glucopyranoside (HL) has been used to
generate appropriate {Cu(L)} and {Ni(L)} building blocks
which via self-assembly afford the tetranuclear m4-hydroxido
bridged copper(II) complex 1 and the trinuclear alcoholate
bridged nickel(II) complex 2, respectively. Complex 1 repre-
sents the first example of a m4-hydroxido bridged copper
complex not supported by a macrocyclic framework, where-
as complex 2 is a rare example of a trinuclear nickel com-
plex with a bis-m3-alkoxido bridged Ni3 core structure repre-
senting an isosceles triangle. The hydrogen-bonding proper-
ties of the employed ligand, providing both hydrogen-bond
donors and acceptors in a rigid preset orientation given by
the glucose backbone, allows for intramolecular hydrogen
bonding between the building blocks in a self-complementa-
ry fashion. In the case of copper(II) this leads to a rationale
for how specific tetranuclear core structures can be generat-
ed by the choice of the appropriate functionalization of the
2-aminoglucose backbone.
Syntheses: The ligand benzyl 2-amino-4,6-O-benzylidene-2-deoxy-a-d-
glucopyranoside (HL) was synthesized starting from N-acetyl-a-d-gluco-
pyranosamine according to the published procedures.[23] All chemicals
were purchased from commercial suppliers and used without further pu-
rification.
CAUTION: In general, perchlorate salts of metal complexes with organic
ligands are potentially explosive. While the present complex has not
proved to be shock sensitive, only small quantities should be prepared
and great care is recommended.
[(m4-OH)Cu4(L)4(MeOH)3(H2O)](ClO4)3 (1):
A
solution of Cu-
(ClO4)2·6H2O (519 mg, 1.40 mmol) in methanol (6.6 mL) was added to a
solution of the ligand HL (500 mg, 1.40 mmol) in chloroform (20 mL) at
room temperature. After 7 d of slow evaporation complex 1 was obtained
as blue prismatic crystals of 1·2.7MeOH which were suitable for X-ray
diffraction. The crystals were isolated, washed with a small amount of
methanol, and dried in air. Yield: 198 mg (28%). Decomposition inter-
ꢀ
val=191–2228C; IR (KBr): n˜ =3448 (br, n O H), 3326 and 3260 (m, n
Both polynuclear complexes 1 and 2 exhibit the rare coor-
dination of the trans-ee-configurated donor atoms N2 and
O3 of the glucose backbone which is assumed for chitosan-
based metal complexes. The carbohydrate backbone adopts
ꢀ
ꢀ
N H), 3169, 3067, and 3033 (w, n C H arom.), 2934 and 2911 (w, nas
CH2), 2874 (w, ns CH2), 1617 (w, n C=C), 1456 (m, d CH2), 1385 (m),
ꢀ
ꢀ
ꢀ
1200 (w, n C O), 1132, 1122, and 1093 (vs, n Cl O and C O), 1069, 1055,
ꢀ
and 1026 (vs, n C O), 990 (s), 927 (m), 764 (m), 737 (m), 701 (s), 668
(m), 624 (m) cmꢀ1; Raman (solid): n˜ =3066 (s, n C H arom.), 2943 (m,
ꢀ
4
for both cases the stable C1 chair conformation. The core
ꢀ
nas CH2), 2875 (m, ns CH2), 1606 (m, n C=C), 1028 (w, n C O), 1003 (s, n
C O), 930 (m) cmꢀ1; UV/Vis (CHCl3): lmax (e)=278 (11.100), 696 nm
ꢀ
structure of both complexes is stabilized by hydrogen bond-
Chem. Eur. J. 2009, 15, 1261 – 1271
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1269

W. Plass et al.
(264mꢀ1 cmꢀ1); MS (micro-ESI in EtOH): m/z (%): 380 (100) [L+Na]+,
777 (8) [{Cu(L)2}+H]+, 840 (16) [{Cu2(L)2}+H]+, 1195 (5) [{Cu2(L)3}]+;
elemental analysis calcd (%) for 1·2.7MeOH, C85.7H113.8Cl3Cu4N4O39.7
(2195.72): C 46.88, H 5.22, N 2.55; found: C 45.27, H 4.74, N 2.53.
Table 4. Crystallographic
2·3.25MeOH·0.75H2O.
data
for
1·2.7MeOH
and
formula
C
85.7H113.8Cl3Cu4N4O39.7
2195.72
183(2)
C123.25H147.5N7Ni3O37
2495.11
183(2)
Mr [gmolꢀ1
T [K]
]
[Ni3(L)5(HL)]NO3 (2): Under vigorous stirring
a
solution of
Ni(NO3)2·6H2O (82 mg, 0.28 mmol) in methanol (5 mL) was added to a
suspension of the ligand HL (200 mg, 0.56 mmol) in methanol (10 mL) at
room temperature. The addition of an aqueous potassium hydroxide solu-
tion (283 mL, 1.98m) led to a pale green solution. After three weeks of
crystal size [mm]
crystal system
space group
a [pm]
b [pm]
c [pm]
0.85ꢃ0.65ꢃ0.45
monoclinic
P21
1620.1(3)
1932.0(4)
1689.0(3)
112.61(3)
4.8803(17)
2
1.494
1.032
2.44–27.49
20946
15547
0.50ꢃ0.50ꢃ0.50
monoclinic
P21
1552.50(3)
1650.06(4)
2556.08(4)
105.8890(10)
6.2978(2)
2
1.316
0.526
2.78–27.49
24506
19301
slow evaporation complex
2 was obtained as green prisms of
2·3.25MeOH·0.75H2O which were suitable for X-ray crystallography.
The crystals were isolated, washed with a small amount of methanol, and
dried in air. Yield: 69 mg (31%). Decomposition interval=251–2648C;
b [8]
V [nm3]
Z
ꢀ
ꢀ
IR (KBr): n˜ =3435 (br, n O H), 3375, 3342, and 3284 (m, n N H), 3064
1calcd [gcmꢀ3
]
ꢀ
and 3033 (w, n C H arom.), 2929 and 2911 (w, nas CH2), 2862 (m, ns
m [mmꢀ1
]
CH2), 1608 and 1498 (w, n C=C), 1455 (m, d CH2), 1384 (s, n N=O), 1216
ꢀ
ꢀ
and 1191 (m, n C O), 1138 (s, n C O), 1090, 1072, 1024, and 984 (vs, n
qmin, qmax [8]
unique data
ꢀ1
ꢀ
C O), 918 (m), 876 (w), 756 (m), 734 (m), 698 (s), 678 (w), 659 (w) cm
;
UV/Vis (CHCl3): lmax (e)=381 (67), 643 (18), 1131 (26mꢀ1 cmꢀ1); MS
(micro-ESI in CHCl3/MeOH): m/z (%): 772 (100) [{Ni(L)2}+H]+, 1544
(12) [{Ni2(L)4}+H]+, 1958 (56) [{Ni3(L)5}]+, 2315 (2) [{Ni3(L)6}+H]+; el-
data with I>2s(I)
no. of parameters
goodness-of-fit on F2
wR2 (all data, F2)
R1 (I>2s(I))
1339
1.025
0.1234
0.0530
1560
1.019
0.1342
0.0530
emental
analysis
calcd
(%)
for
2·3.25MeOH·0.75H2O,
C123.25H147.5N7Ni3O37 (2495.11): C 59.33, H 5.87, N 3.93; found: C 59.58, H
5.68, N 3.89.
flack parameter
0.001(11)
ꢀ0.008(9)
X-ray structure determination: Single crystals were selected while still
covered with mother liquor under a polarizing microscope and fixed on
fine glass fibers. Single crystal X-ray measurements were carried out on a
Nonius Kappa CCD diffractometer using graphite monochromated MoKa
radiation (l=71.073 pm). The structures were solved by direct methods
with SHELXS-97 and were full-matrix least-squares refined against F2
using SHELXL-97.[47] The program XP (SIEMENS Analytical X-ray In-
struments, Inc.) was used for structure representations. Hydrogen atoms
were calculated and treated as riding atoms with fixed thermal parame-
ters. For 1·2.7MeOH anisotropic thermal parameters were used for all
non-hydrogen atoms except for the disordered benzyl substituents. The
perchlorate ion Cl1P, located in the cavity below the Cu4 plane which is
opposite to the m4-OH group, is disordered with an occupation factor of
0.5 at two crystallographic positions. This is consistent with the flexibility
of the hydrogen bonding interactions through the four amino N-H
donors within the cavity. For Cl2P no disorder is observed, whereas the
perchlorate ion Cl3P is highly disordered. This again also reflects in the
spacial flexibility of the crystallographic cavity given by possible hydro-
gen-bonding interactions. In fact this perchlorate ion is disordered at four
crystallographic positions (occupations: 0.4, 0.3, 0.15, 0.15). When the
two more probable of these positions are filled an additional methanol
molecule with an occupation factor of 0.7 is present in its vicinity. For
2·3.25MeOH·0.75H2O anisotropic thermal parameters were used for all
non-hydrogen atoms except for the disordered benzyl substituents, the
disordered water molecules, and some of the partially occupied methanol
molecules. Details of the data collection and refinement procedure are
summarized in Table 4.
[1] J. F. Kennedy, C. A. White, Bioactive Carbohydrates in Chemistry,
Biochemistry and Biology, Wiley, New York 1983.
[2] a) H. Hꢁusler, A. E. Stꢀtz, Top. Curr. Chem. 2001, 215, 77–114;
b) X. He, G. Agnihotri, H.-W. Liu, Chem. Rev. 2000, 100, 4615–
4661; c) S. H. Bhosale, M. B. Rao, V. V. Deshpande, Microbiol. Rev.
1996, 60, 280–300; d) R. W. Gracy, E. A. Noltmann, J. Biol. Chem.
1968, 243, 4109–4116; e) K. D. Hardman, R. C. Agarwal, M. J. Freis-
er, J. Mol. Biol. 1982, 157, 69–86; f) R. L. Root, J. R. Durrwachter,
C.-H. Wong, J. Am. Chem. Soc. 1985, 107, 2997–2999; g) G. K.
Farber, A. Glasfeld, G. Tiraby, D. Ringe, G. A. Petsko, Biochemistry
1989, 28, 7289–7297; h) J. Jenkins, J. Janin, F. Rey, M. Chiadmi, H.
van Tilbeurgh, I. Lasters, M. De Maeyer, D. Van Belle, S. J. Wodak,
M. Lauwereys, P. Stanssens, N. T. Mrabet, J. Snauwaert, G. Matthys-
sens, A.-M. Lambeir, Biochemistry 1992, 31, 5449–5458; i) Y.
Zhang, J.-Y. Liang, S. Huang, H. Ke, W. N. Lipscomb, Biochemistry
1993, 32, 1844–1857; j) A. Lavie, K. N. Allen, G. A. Petsko, D.
Ringe, Biochemistry 1994, 33, 5469–5480; k) K. K.-S. Ng, W. I. Weis,
Biochemistry 1997, 36, 979–988.
[3] S. Yano, Y. Mikata, Bull. Chem. Soc. Jpn. 2002, 75, 2097–2113.
[4] a) Y. E. Alexeev, I. S. Vasilchenko, B. I. Kharisov, L. M. Blanco,
A. D. Garnovskii, Y. A. Zhdanov, J. Coord. Chem. 2004, 57, 1447–
1517; b) D. Steinborn, H. Junicke, Chem. Rev. 2000, 100, 4283–4317;
c) B. Gyurcsik, L. Nagy, Coord. Chem. Rev. 2000, 203, 81–149; d) S.
Yano, Coord. Chem. Rev. 1988, 92, 113–156.
[5] R. H. Holm, P. Kennepohl, E. I. Solomon, Chem. Rev. 1996, 96,
2239–2314.
[6] a) O. Kahn, Molecular Magnetism, VCH, Weinheim 1993; b) R. E. P.
Winpenny, Adv. Inorg. Chem. 2001, 52, 1–111.
CCDC 682760, 682761 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.
[7] a) P. Klꢀfers, O. Krotz, M. Oßberger, Eur. J. Inorg. Chem. 2002,
1919–1923; b) P. Klꢀfers, T. Kunte, Eur. J. Inorg. Chem. 2002, 1285–
1289; c) P. Klꢀfers, T. Kunte, Angew. Chem. 2001, 113, 4356–4358;
Angew. Chem. Int. Ed. 2001, 40, 4210–4212; d) N. Habermann, G.
Jung, M. Klaassen, P. Klꢀfers, Chem. Ber. 1992, 125, 809–814.
[8] T. Tanase, H. Inukai, T. Onaka, M. Kato, S. Yano, S. J. Lippard,
Inorg. Chem. 2001, 40, 3943–3953.
[9] a) T. Tanase, Y. Yasuda, T. Onaka, S. Yano, J. Chem. Soc. Dalton
Trans. 1998, 345–352; b) T. Tanase, M. Nakagoshi, A. Teratani, M.
Kato, Y. Yamamoto, S. Yano, Inorg. Chem. 1994, 33, 5–6.
Acknowledgements
This work was financially supported by the “Deutsche Forschungsge-
meinschaft” (SFB 436 “Metal Mediated Reactions Modeled after
Nature” and SPP 1137 “Molecular Magnetism”). E.T.S. gratefully ac-
knowledges financial support by the Graduiertenfçrderung of the Frei-
staat Thꢀringen and the Carl-Zeiss Stiftung. We also thank the Marie-
Curie training site LAboratory for Molecular Magnetism (LAMM) in
Florence, Italy.
[10] a) Y. Mikata, Y. Shinohara, K. Yoneda, Y. Nakamura, K. Esaki, M.
Tanahashi, I. BrudziÇska, S. Hirohara, M. Yokoyama, K. Mogami, T.
Tanase, T. Kitayama, K. Takashiba, K. Nabeshima, R. Takagi, M.
1270
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1261 – 1271

Unusual Polynuclear Complexes
FULL PAPER
Takatani, T. Okamoto, I. Kinoshita, M. Doe, A. Hamazawa, M.
Morita, F. Nishida, T. Sakakibara, C. Orvig, S. Yano, J. Org. Chem.
2001, 66, 3783–3789; b) Y. Mikata, Y. Sugai, S. Yano, Inorg. Chem.
2004, 43, 4778–4780; c) T. Storr, Y. Sugai, C. A. Barta, Y. Mikata,
M. J. Adam, S. Yano, C. Orvig, Inorg. Chem. 2005, 44, 2698–2705;
d) T. Storr, M. Obata, C. L. Fisher, S. R. Bayly, D. E. Green, I. Brud-
ziÇska, Y. Mikata, B. O. Patrick, M. J. Adam, S. Yano, C. Orvig,
Chem. Eur. J. 2005, 11, 195–203; e) Y. Mikata, Y. Sugai, M. Obata,
M. Harada, S. Yano, Inorg. Chem. 2006, 45, 1543–1551.
[11] J. Becher, I. Seidel, W. Plass, D. Klemm, Tetrahedron 2006, 62,
5675–5681.
[27] W. Plass, H. P. Yozgatli, Z. Anorg. Allg. Chem. 2003, 14, 65–70.
[28] I. D. Brown, D. Altermatt, Acta Crystallogr. Sect. B 1985, 41, 244–
247.
[29] a) A. M. Kirillov, M. N. Kopylovich, M. V. Kirillova, M. Haukka,
M. F. C. G. da Silva, A. J. L. Pombeiro, Angew. Chem. 2005, 117,
4419–4423; Angew. Chem. Int. Ed. 2005, 44, 4345–4349; b) M.
Bera, W. T. Wong, G. Aromi, J. Ribas, D. Ray, Inorg. Chem. 2004,
43, 4787–4789; c) S. Mukherjee, T. Weyhermꢀller, E. Bothe, K. Wie-
ghardt, P. Chaudhuri, Eur. J. Inorg. Chem. 2003, 863–875; d) J.
Reim, R. Werner, W. Haase, B. Krebs, Chem. Eur. J. 1998, 4, 289–
298.
[12] a) M. E. Cucciolito, R. Del Litto, G. Roviello, F. Ruffo, J. Mol.
Catal. A 2005, 236, 176–181; b) J. C. Irvine, J. C. Earl, J. Chem. Soc.
1922, 121, 2376–2381.
[13] a) A. K. Sah, C. P. Rao, P. K. Saarenketo, E. K. Wegelius, E. Koleh-
mainen, K. Rissanen, Eur. J. Inorg. Chem. 2001, 2773–2781; b) J.
Zhao, X. Zhou, A. M. Santos, E. Herdtweck, C. C. Rom¼o, F. E.
Kꢀhn, Dalton Trans. 2003, 3736–3742.
[30] E. Ruiz, A. Rodrꢅguez-Fortea, P. Alemany, S. Alvarez, Polyhedron
2001, 20, 1323–1327.
[31] T. C. Higgs, C. J. Carrano, Inorg. Chem. 1997, 36, 298–306.
[32] W. Plass, A. Pohlmann, J. Rautengarten, Angew. Chem. 2001, 113,
4333–4336; Angew. Chem. Int. Ed. 2001, 40, 4207–4210.
[33] A. Yuchi, H. Murakami, M. Shiro, H. Wada, G. Nakagawa, Bull.
Chem. Soc. Jpn. 1992, 65, 3362–3373.
[14] a) R. Wegner, M. Gottschaldt, H. Gçrls, E.-G. Jꢁger, D. Klemm,
Angew. Chem. 2000, 112, 608–612; Angew. Chem. Int. Ed. 2000, 39,
595–599; b) R. Wegner, M. Gottschaldt, H. Gçrls, E.-G. Jꢁger, D.
Klemm, Chem. Eur. J. 2001, 7, 2143–2157; c) A. K. Sah, C. P. Rao,
P. K. Saarenketo, K. Rissanen, G. A. van Albada, J. Reedijk, Chem.
Lett. 2002, 31, 348–349; d) A. K. Sah, T. Tanase, Chem. Commun.
2005, 5980–5981; e) A. K. Sah, M. Kato, T. Tanase, Chem.
Commun. 2005, 675–677; f) A. Roth, J. Becher, C. Herrmann, H.
Gçrls, G. Vaughan, M. Reiher, D. Klemm, W. Plass, Inorg. Chem.
2006, 45, 10066–10076.
[15] A. Fragoso, M. L. Kahn, A. CastiÇeiras, J.-P. Sutter, O. Kahn, R.
Cao, Chem. Commun. 2000, 1547–1548.
[16] A. K. Sah, T. Tanase, M. Mikuriya, Inorg. Chem. 2006, 45, 2083–
2092.
[17] A. Burkhardt, E. T. Spielberg, H. Gçrls, W. Plass, Inorg. Chem.
2008, 47, 2485–2493.
[18] A. Burkhardt, A. Buchholz, H. Gçrls, W. Plass, Eur. J. Inorg. Chem.
2006, 3400–3406.
[19] A. Burkhardt, W. Plass, Inorg. Chem. Commun. 2008, 11, 303–306.
[20] A. Burkhardt, H. Gçrls, W. Plass, Carbohydr. Res. 2008, 343, 1266–
1277.
[21] a) S. Yano, M. Doi, S. Tamakoshi, W. Mori, M. Mikuriya, A. Ichi-
mura, I. Kinoshita, Y. Yamamoto, T. Tanase, Chem. Commun. 1997,
997–998; b) T. Tanase, S. Tamakoshi, M. Doi, M. Mikuriya, H. Sa-
kurai, S. Yano, Inorg. Chem. 2000, 39, 692–704.
[22] a) T. Tanase, K. Kurihara, S. Yano, K. Kobayashi, T. Sakurai, S.
Yoshikawa, J. Chem. Soc. Chem. Commun. 1985, 1562–1563; b) T.
Tanase, K. Kurihara, S. Yano, K. Kobayashi, T. Sakurai, S. Yoshika-
wa, M. Hidai, Inorg. Chem. 1987, 26, 3134–3139; c) J. Burger, C.
Gack, P. Klꢀfers, Angew. Chem. 1995, 107, 2950–2951; Angew.
Chem. Int. Ed. Engl. 1995, 34, 2647–2649; d) T. Tanase, T. Onaka,
M. Nakagoshi, I. Kinoshita, K. Shibata, M. Doe, J. Fujii, S. Yano,
Chem. Commun. 1997, 2115–2116; e) T. Tanase, T. Onaka, M. Na-
kagoshi, I. Kinoshita, K. Shibata, M. Doe, J. Fujii, S. Yano, Inorg.
Chem. 1999, 38, 3150–3159.
[23] a) P. H. Gross, R. W. Jeanloz, J. Org. Chem. 1967, 32, 2759–2763;
b) Z. Gyçrgydeꢄk, Liebigs Ann. Chem. 1991, 1291–1300; c) A. Bur-
khardt, A. Buchholz, H. Gçrls, W. Plass, Acta Crystallogr. Sect. E
2007, E63, o2527–o2529.
[24] a) V. McKee, S. S. Tandon, J. Chem. Soc. Dalton Trans. 1991, 221–
230; b) V. McKee, S. S. Tandon, J. Chem. Soc. Chem. Commun.
1988, 385–387.
[34] E. Ilina, V. G. Kessler, Polyhedron 2005, 24, 3052–3056.
[35] A. K. Sah, C. P. Rao, P. K. Saarenketo, K. Rissanen, Chem. Lett.
2001, 11, 1296–1297.
[36] a) L. Merz, W. Haase, J. Chem. Soc. Dalton Trans. 1978, 1594–1598;
b) R. Mergehenn, W. Haase, Acta Crystallogr. Sect. B 1977, 33,
2734–2739; c) J. Tercero, E. Ruiz, S. Alvarez, A. Rodrꢅguez-Fortea,
P. Alemany, J. Mater. Chem. 2006, 16, 2729–2735.
[37] F. Tuna, L. Patron, Y. Journaux, M. Andruh, W. Plass, J.-C. Trombe,
J. Chem. Soc. Dalton Trans. 1999, 539–545.
[38] a) M. F. Charlot, O. Kahn, S. Jeannin, Y. Jeannin, Inorg. Chem.
1980, 19, 1410–1411; b) E. Ruiz, P. Alemany, S. Alvarez, J. Cano,
Inorg. Chem. 1997, 36, 3683–3688; c) E. Ruiz, P. Alemany, S. Alvar-
ez, J. Cano, J. Am. Chem. Soc. 1997, 119, 1297–1303.
[39] A. K. Sharma, F. Lloret, R. Mukherjee, Inorg. Chem. 2007, 46,
5128–5130; and references therein.
[40] A. P. Ginsberg, R. L. Martin, R. C. Sherwood, Inorg. Chem. 1968, 7,
932–936.
[41] a) T. Weyhermꢀller, R. Wagner, S. Khanra, P. Chaudhuri, Dalton
Trans. 2005, 2539–2546; b) H. Adams, D. E. Fenton, L. R. Cum-
mings, P. E. McHugh, M. Ohba, H. Okawa, H. Sakiyama, T. Shiga,
Inorg. Chim. Acta 2004, 357, 3648–3656; c) A. Escuer, M. S. El
Fallah, S. B. Kumar, F. Mautner, R. Vicente, Polyhedron 1998, 8,
377–381; d) V. V. Pavlishchuk, S. V. Kolotilov, A. W. Addison, M. J.
Prushan, R. J. Butcher, L. K. Thompson, Inorg. Chem. 1999, 38,
1759–1766; e) A. Escuer, R. Vicente, S. B. Kumar, X. Solans, M.
Font-Bardia, A. Caneschi, Inorg. Chem. 1996, 35, 3094–3098.
[42] a) A. E. Ion, E. T. Spielberg, H. Gçrls, W. Plass, Inorg. Chim. Acta
2007, 360, 3925–3931; b) L. K. Thompson, V. Niel, H. Grove, D. O.
Miller, M. J. Newlands, P. H. Bird, W. A. Wickramasinghe, A. B. P.
Lever, Polyhedron 2004, 23, 1175–1184.
[43] X.-H. Bu, M. Du, L. Zhang, D.-Z. Liao, J.-K. Tang, R.-H. Zhang, M.
Shionoya, J. Chem. Soc. Dalton Trans. 2001, 593–598.
[44] a) M. A. Halcrow, J.-S. Sun, J. C. Huffman, G. Christou, Inorg.
Chem. 1995, 34, 4167–4177; b) J. M. Clemente-Juan, B. Chansou, B.
Donnadieu, J.-P. Tuchagues, Inorg. Chem. 2000, 39, 5515–5519.
[45] a) W. Plass, Angew. Chem. 1996, 108, 699–703; Angew. Chem. Int.
Ed. Engl. 1996, 35, 627–631; b) W. Plass, Inorg. Chem. 1997, 36,
2200–2205.
[46] E. Bill, Program package julX Version 1.3, MPI for Bioinorganic
Chemistry, Mꢀhlheim/Ruhr (Germany), 2007.
[47] G. M. Sheldrick, SHELXS-97 and SHELXL-97, University of Gçt-
tingen, Gçttingen (Germany), 1997.
[25] V. Lozan, A. Buchholz, W. Plass, B. Kersting, Chem. Eur. J. 2007,
13, 7305–7316.
[26] A. Roth, A. Buchholz, M. Rudolph, E. Schꢀtze, E. Kothe, W. Plass,
Chem. Eur. J. 2008, 14, 1571–1583.
Received: April 8, 2008
Revised: August 14, 2008
Published online: December 19, 2008
Chem. Eur. J. 2009, 15, 1261 – 1271
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1271