A butterfly-shaped water tetramer in a Cu4 complex supported by a hydrazone ligand: Synthesis, crystal structure, magnetic properties, and quantum chemical study

Article abstract of DOI:10.1071/CH09192

A potentially tetradentate NOOO donor hydrazone ligand, LH₂ (condensation product of benzhydrazide with O-vanillin) generates a tetranuclear Cu^II complex [Cu₄(L)₄]·4H₂O (1), whose void spaces are occu

Full text of DOI:10.1071/CH09192

Full Paper
CSIRO PUBLISHING
Aust. J. Chem. 2009, 62, 1614–1621
www.publish.csiro.au/journals/ajc
A ‘Butterfly’-Shaped Water Tetramer in a Cu4 Complex
Supported by a Hydrazone Ligand: Synthesis, Crystal
Structure, Magnetic Properties, and Quantum
Chemical Study
Sambuddha Banerjee,A Soma Sen,A Joy Chakraborty,A Ray J. Butcher,B
Carlos J. Gómez García,C Ralph Puchta,D,E and Samiran MitraA,F
ADepartment of Chemistry, Jadavpur University, Kolkata-700 032, India.
BDepartment of Chemistry, Howard University, 2400 Sixth Street, N.W., Washington,
DC, 20059, USA.
CInstituto de Ciencia Molecular (ICMol). Parque Científico. Universidad de Valencia,
46980 Paterna (Valencia) Spain.
DInorganic Chemistry, Department of Chemistry and Pharmacy, University of Erlangen-Nürnberg,
Egerlandstrasse 1, 91058 Erlangen, Germany.
EComputer Chemistry Center, Department of Chemistry and Pharmacy, University of
Erlangen-Nürnberg, Nägelsbachstr 25, 91052 Erlangen, Germany.
FCorresponding author. Email: smitra_2002@yahoo.com
A potentially tetradentate NOOO donor hydrazone ligand, LH2 (condensation product of benzhydrazide with O-vanillin)
generates a tetranuclear CuII complex [Cu4(L)4]·4H2O (1), whose void spaces are occupied by water tetramers presenting
a ‘butterfly’ conformation with the highest dihedral angle reported to date, as revealed by its X-ray crystal structure.
1 has also been characterized using various spectroscopic techniques, including IR, UV-vis, and elemental analysis.
Variable temperature magnetic susceptibility measurements reveal the presence of moderate antiferromagnetic intra-
tetramer coupling between the four CuII centres connected through simple oxo groups of the hydrazone ligand with two
different coupling constants (J1 = −61.7(3) cm−1 and J2 = −92(1) cm−1) corresponding to the two different CuII tetramers
identified in the X-ray structure. We also report a quantum chemical study (MP2(full)/6–311+G(3df,2p)//B3LYP/6–
311+G(3df,2p)) to calculate the stability of the water tetramers.
Manuscript received: 6 April 2009.
Manuscript accepted: 10 July 2009.
Introduction
complex. As reported earlier,[7a] in phenoxo-bridged CuII com-
plexes, the J value depends on (i) the Cu–O(phenolate)–Cu
angle, (ii) the Cu–Cu distance, (iii) the geometry around
the phenolate oxygen atom, and (iv) the copper coordina-
tion geometry. A very comprehensive overview on tetranuclear
CuII complexes is also available.[7b] Variable temperature mag-
netic (VTM) susceptibility measurements of the title complex
reinforced these points.
Another research area enjoying increased interest, is the
study of small water clusters that act as bridges between sin-
gle water molecules and liquid water or ice. The aim of such
research is many fold: from understanding numerous ‘anoma-
lous’ behaviours of bulk water, to probe its role in the stabi-
lization and action of bio-molecules; from unravelling the roles
in chemical processes, to the design of new materials and the
exploration of their possible structures and stability in differ-
ent environments. H-bonding interactions and their fluctuations
determine the properties of water in bulk, with associated high
directionality leading to a reduced number of neighbours (∼4)
around each water molecule, in contrast with other solvents
The synthesis and characterization of polynuclear CuII coordina-
tion complexes with different nuclearties has been, and remains,
one of the most important areas in coordination chemistry, due to
thehugenumberofdifferentstructures, functionalities, andmag-
netic properties observed in CuII coordination complexes.[1,2]
From the structural point of view, there are thousands of char-
acterized CuII complexes with one, two, and three dimensions;
presenting weak, moderate, or strong magnetic couplings (either
ferro or antiferromagnetic). If we restrict our search to tetra-
nuclear CuII complexes, we can find several hundreds, of which
only 24 present exclusively simple oxo-bridges.
In this area, we have recently reported a CuII tetranuclear
complex with the same hydrazone ligand having a distorted
cubane structure with a µ3-phenoxo bridge.[3] In our ongoing
research on phenoxo-bridged transition metal complexes,[4–6]
here we describe the synthesis of a completely different metal-
ligandarrangement, showingµ2-phenoxobridges, obtainedwith
the same metal and ligand but with a slight modification of the
experimental set up which yielded a very different tetranuclear
© CSIRO 2009
10.1071/CH09192
0004-9425/09/121614

‘Butterfly’-Shaped Water Tetramer in a Cu4 Complex Supported by a Hydrazone Ligand
1615
C(13A)
C(12A)
C(14A)
C(15A)
C(11A)
C(10A)
C(9A)
O(3A)
Cu(1)
N(2A)
N(1A)
C(8A)
C(3B)
O(1A)
O(2B)
O(1B)
C(8B)
C(2B)
C(1B)
C(7A)
Cu(2)
C(1A)
C(4B)
C(5B)
O(3B)
C(6A)
C(7B)
C(15B)
C(12B)
C(14B)
C(13B)
C(6B)
N(1B)
O(2A)
N(2B)
C(2A)
C(9B)
C(10B)
C(11B)
C(5A)
C(4A)
C(3A)
Fig. 1. ORTEP diagram of [Cu4(L)4] cluster of 1 showing the Cu(1) and Cu(2) centres with atom numbering scheme.
H-atoms are omitted for clarity.
where this number is 10–11.[8] Theoretical and experimental
studies of water clusters in restricted environments have been
carried out and their structures have been predicted by theo-
retical ab initio calculations. In order to account for the heat
capacity of liquid water (which largely exceeds that of the gas
phase), examination of small and long-lived water clusters has
become a major concern in recent studies. Water tetramers play a
crucial role in such studies as they are taken as a standard model
for theoretical calculations of the heat capacity of bulk water.[8]
Moreover, it has been possible to calculate different configura-
tions of water tetramers embedded in different matrices, with
some even being characterized by far infrared vibration rotation
tunnellingspectroscopy.[9,10] Theoreticalcalculationsofthelow-
est energy conformation for small water clusters show that water
trimers, tetramers, and pentamers prefer cyclic and quasi-planar
structures. In the present work we report a cyclic water tetramer,
having a ‘butterfly’ conformation, stabilized by a tetranuclear
CuII hydrazone complex [Cu4(L)4]·4H2O (1). The calculations
show that the observed conformation of the ‘butterfly’ shaped
tetrameric water clusters aquire some of their stability from the
confinement of the hydrazone ligands of 1.
its magnetic properties that show a moderate antiferromagnetic
coupling with two distinct coupling constants, corresponding to
the two different CuII tetramers present in the structure of 1.
Results and Discussion
Crystal Structure of Complex 1
The structure of 1 consists of two crystallographically inde-
pendent CuII units where each tetranuclear unit lies on a crys-
tallographic symmetry operation thus giving only two unique
CuII centres per unit and are designated as Cu(1)–Cu(2) and
Cu(3)–Cu(4) (Fig. 1). All the CuII ions in the Cu4 core are five-
coordinated with a CuO4N coordination environment. The bond
lengths and angles in both CuII tetramers are listed inTable 1.The
CuII ions are linked in the tetranuclear cluster by µ2-phenoxo
bridges of the hydrazone ligand H2L.
The four crystallographically independent CuII ions are
present in distorted square pyramidal geometries, as confirmed
by the trigonality Addison indices (τ) of 0.0695, 0.1111,
0.1032, and 0.0231 for Cu(1) to Cu(4) atoms, respectively
(τ = |β − α|/60, where α and β are the two greatest basal angles
when the polyhedron is viewed as a square-pyramid. For a per-
fectly square-pyramidal geometry, τ is equal to zero and it
becomes unity for a perfect trigonal bipyramidal geometry).[11]
Here we report the synthesis, X-ray crystal structure, spectral
properties, and quantum chemical study showing the stability of
a water tetramer in a [Cu4(L)4]·4H2O (1) complex along with

1616
S. Banerjee et al.
Table 1. Selected bond lengths [Å] and bond angles [◦] around Cu(1), Cu(2), Cu(3), and Cu(4) ions in 1
Bond lengths [Å]
Cu(1)
Cu(2)
Cu(3)
Cu(4)
Cu(1)–N(1A)
Cu(1)–O(3A)
Cu(1)–O(1A)
Cu(1)–O(2B)#1
Cu(1)–O(1B)#
1.913(3)
Cu(2)–N(1B) 1.917(3) Cu(3)–N(1C)
1.922(3) Cu(4)–N(1D) 1.922(3)
1.917(2) Cu(4)–O(3D) 1.909(2)
1.969(2) Cu(4)–O(1C) 2.005(2)
1.9316(19) Cu(2)–O(3B) 1.939(3) Cu(3)–O(3C)
1.9623(19) Cu(2)–O(1A) 1.991(2) Cu(3)–O(1C)
2.289(2)
1.983(2)
Cu(2)–O(2A) 2.287(2) Cu(3)–O(2D)#2 2.322(2) Cu(4)–O(2C) 2.336(2)
Cu(2)–O(1B) 1.959(2) Cu(3)–O(1D)#2 1.989(2) Cu(4)–O(1D) 1.922(2)
Bond angles [◦]
N(1A)–Cu(1)–O(3A)
N(1A)–Cu(1)–O(1A)
O(3A)–Cu(1)–O(1A)
81.78(10)
92.36(10)
171.33(10)
N(1A)–Cu(1)–O(1B)#1 175.50(12)
O(3B)–Cu(2)–O(1B)
N(1B)–Cu(2)–O(1A)
O(1B)–Cu(2)–O(2A)
O(1A)–Cu(2)–O(2A)
O(3C)–Cu(3)–N(1C)
O(3C)–Cu(3)–O(1C)
N(1C)–Cu(3)–O(1C)
171.38(9)
178.18(10)
99.58(8)
77.13(9)
81.13(11)
170.03(10)
92.25(11)
N(1C)–Cu(3)–O(1D)#2 176.22(10)
N(1C)–Cu(3)–O(2D)#2 108.00(11)
O(3D)–Cu(4)–O(1D)
N(1D)–Cu(4)–O(1C)
O(3D)–Cu(4)–O(1D)
N(1D)–Cu(4)–O(1C)
O(1D)–Cu(4)–O(1C)
O(3D)–Cu(4)–O(2C)
173.87(10)
172.48(11)
173.87(10)
172.48(11)
88.80(9)
84.58(9)
c
b
a
Fig. 2. H-bonding diagram of 1.
The water molecules are present in the ‘exo-pockets’ of the
tetrameric CuII units holding the individual tetranuclear metal-
clusters to form 1D polynuclear chains (Fig. 2) along the c-axis.
The formation of a small water cluster can be antici-
pated following the vibrational spectra of the complex. The
intermolecular O–H bending peak was observed ∼80 cm−1
,
blue-shifted with respect to a mononuclear lattice water, whereas
the appearance of a sharp band at nearly 500 cm−1, red-shifted
from the standard band for lattice water, indicated the probable
presence of bridging H-atoms.[8b]

‘Butterfly’-Shaped Water Tetramer in a Cu4 Complex Supported by a Hydrazone Ligand
1617
Table 2. The inter-water separations [Å] in O(1W)–O(3W)
Table 3. Hydrogen bonds for 1 [Å and ◦]
and O(2W)–O(4W) cluster
Symmetry transformations used to generate equivalent atoms: #1 −x + 1/2,
y, −z + 1/2; #2 −x + 1/2, y, −z + 3/2; #3 x, y − 1, z; #4 −x + 1/2, y + 1,
−z + 1/2; #5 −x + 1/2, y + 1, −z + 3/2
O(1W)–O(3W)
O(3W)–O(1W)
O(2W)–O(4W)
O(4W)–O(2W)
2.830
2.779
2.849
2.858
D–H· · ·A
d(D–H) d(H· · ·A) d(D· · ·A) <(DHA)
O(1W)–H(1W1)· · ·N(2A)
O(3W)–H(3W1)· · ·N(2B)
O(4W)–H(4W1)· · ·N(2C)
O(2W)–H(2W1)· · ·N(2D)
0.812(18) 2.039(18) 2.849(4) 176(4)
0.792(18) 2.059(19) 2.847(4) 174(5)
0.818(19) 2.08(2)
2.887(4) 170(5)
There are in total four crystallographically independent
water molecules in the void space of the tetranuclear CuII
cores whose oxygen atoms are designated as O(1W), O(2W),
O(3W), and O(4W) (refer to Table 2 for inter-water distances).
These four water molecules are arranged in two different
water tetramers (first as O(1W)–O(3W)–O(1W)#4–O(3W)#3,
referred to as the O(1W)–O(3W) cluster; and second as O(2W)–
O(4W)–O(2W)#5–O(4W)#3, referred to as the O(2W)–O(4W)
cluster). Each water molecule in the cluster forms one H-
bond with its immediate neighbouring water molecule and thus,
for each water, one remaining hydrogen does not participate
in an H-bonding interaction within the water cluster. These
‘non-interacting’ hydrogens are involved in H-bonding inter-
actions with the N(2A) and N(2B) of the Cu(1)–Cu(2) core,
and with the N(2C) and N(2D) of the Cu(3)–Cu(4) core. The
geometrical arrangements of these latter hydrogen atoms, with
respect to the ones involved in H-bonding interactions within the
water clusters, are also important and will be treated in detail
in the quantum chemical study result section. In both water
tetramers (O(1W)–O(3W) and O(2W)–O(4W)), the hydrogen
atoms forming H-bonds with the ligand N atoms, are arranged
in an up-down-up-down fashion, leading to the formation of 1D
H-bonded polynuclear clusters formed by Cu4 cores, linked by
cyclic water oligomers (the O(1W)–O(3W) cluster links the
Cu(1)–Cu(2)tetramers, whereastheO(2W)–O(4W)clusterlinks
the Cu(3)–Cu(4) core).
0.809(18) 2.181(19) 2.989(4) 176(5)
O(1W)–H(1W2)· · ·O(3W)#3 0.802(17) 2.04(2)
2.830(4) 169(4)
O(2W)–H(2W2)· · ·O(4W)#3 0.784(17) 2.076(18) 2.849(4) 169(4)
O(3W)–H(3W2)· · ·O(1W)#4 0.816(18) 1.97(2)
2.779(4) 171(5)
2.858(6) 152(5)
O(4W)–H(4W2)· · ·O(2W)#5 0.808(19) 2.12(3)
O
O
O
1.78 Å
O
Fig. 3. Calculated water tetramer (d(O· · ·O) = 2.75Å).
the lengthening of the covalent O–H bond and shortening of
the O· · ·O separation. That the O· · ·O distance within the water
clusters, as predicted from this theory, falls in the range reported
for these water tetramers, indirectly indicates the cooperative CT
effect in H-bonding interaction.[8b]
DFT Study on the Stability of the Water Tetramer
To learn about the stability of this ‘butterfly’ tetrameric
water cluster, we have calculated the structures of some
cyclic water oligomers (H2O)n (n = 3–6)[13] applying B3LYP/6–
311+G(3df,2p),[14] i.e. using an extended basis set as in G2
theory[15] and evaluated the energy per H-bond (MP2(full)/
6–311+G(3df,2p)//B3LYP/6–311+G(3df,2p) + ZPE(B3LYP/
6–311+G(3df,2p))). Comparison of the cyclic water clusters
shows that from the trimeric to the tetrameric cyclic clus-
ter the stability is improved to −22.2 kJ mol−1 per hydro-
The two tetrameric water clusters differ in their geometry and
symmetry and thus, are discussed separately.
The O(1W)–O(3W) water cluster is more planar with a di-
hedral angle of 17.10◦[10] between the mean planes, whereas for
the O(2W)–O(4W) water cluster this angle is 32.63◦. According
to our literature survey, an inter-planer angle of 32.55◦ between
the mean planes of water tetramers is yet to be reported, and this
is therefore, the first water tetramer with a ‘butterfly’conforma-
tion with such a large dihedral angle. Each water molecule in the
tetrameric water cluster is tri-coordinated, linking the CuII metal
clusters by H-bonds with uncoordinated hydrazide nitrogen
atoms of the ligand fragment (N(2A) and N(2B) for the O(1W)–
O(3W) cluster and N(2C) and N(2D) for the O(2W)–O(4W)
cluster).
The H-bond distances are, as expected, quite different for the
two water tetramers (Table 3 and Fig. 3). The O· · ·O distances
in both clusters are quite different (2.858 and 2.849 Å in the
O(2W)–O(4W) cluster compared with 2.779 and 2.830 Å in the
O(1W)–O(3W) cluster), but they are all in the range of those
observed in solid phase water tetramers.[8b,12]
Other researchers have already shown that the major factor
contributing to the stabilization of H-bonds is the charge transfer
(CT) interaction, and not the electrostatic interaction for many
small cluster systems, where the latter is cancelled by exchange
interactions. The CT interaction was described as the stabiliza-
tion of a O–H antibonding level by acceptance of a lone pair
from oxygen. This leads to both enthalpic and entropic stabi-
lization too. The other consequences of such CT interaction is
gen bond, as opposed to −13.8 kJ mol−1 [16]
;
and the H-bond
donor-acceptor lengths between the water molecules are short-
ened (for (H2O)3, d(OH· · ·O) = 1.91 Å, whereas for (H2O)4,
d(OH· · ·O) = 1.78 Å). Going to the pentameric and hexameric
cyclic water clusters shows no further improvement of the sta-
bilization energy per hydrogen bond and the hydrogen bond
length does not change significantly (for (H2O)5 the energy
per H-bond is −23.1 kJ mol−1 [12] and d(OH· · ·O) = 1.75 Å and
for (H2O)6 the energy per H-bond is −24.3 kJ mol−1 and
d(OH· · ·O) = 1.74 Å) (Table S1 and Fig. S1 in the Accessory
Publication).
This findings are in reasonable agreement with ear-
lier calculations and experimental work on water trimers,
tetramers[16a,b] and pentamers which prefer cyclic and quasi-
planar structures.[16]
The cyclic water tetramers (H2O)4, where every water
molecule accepts and donates one hydrogen bond within the
cluster, keeping the other free which can present four differ-
ent conformations.[17] Thus, the hydrogen atoms which are not
involved in the hydrogen bonding system within the cluster can

1618
S. Banerjee et al.
(a)
(b)
O
O
O
O
O
O
O
O
(c)
(d)
O
O
O
O
O
O
O
O
Fig. 4. Different possible orientations of the H atoms not involved in H-bonding within the cluster (a) up-up-up-up, (b) up-up-
up-down, (c) up-up-down-down, and (d) up-down-up-down.
0.009
be arranged in 4-up (up-up-up-up) conformation, 3-up, 1-down
(up-up-up-down), neighbouring 2-up, 2-down (up-up-down-
down) and, finally, alternating 2-up, 2-down (up-down-up-down)
conformation, as in our butterfly structure (Figs 3 and 4).
Calculations show that our butterfly-shaped up-down-up-
down tetrameric conformer is the most stable, and that the
up-up-up-up conformation rearranges during the calculation to
the most stable up-down-up-down one. The up-up-down-down
and up-up-up-down conformers are ∼3.3 kJ mol−1 higher in
energy than the alternate up-down-up-down conformer, i.e., they
are nearly iso-energetic. Given the small energy gap of less than
4.2 kJ mol−1 we can deduce that the conformation of the water
tetramer in 1 is mainly originated by the hydrogen bonds to the
ligands (N(2A) and N(2B) for the O(1W)–O(3W) water cluster
and N(2C) and N(2D) for the O(2W)–O(4W) water cluster) of
the two CuII tetrameric complexes.
Therefore, from theoretical calculations, we can conclude
that for a water tetramer the possible conformations that can
be adopted are 3-up, 1-down (up-up-up-down), neighbouring 2-
up, 2-down (up-up-down-down) and, finally, alternating 2-up,
2-down (up-down-up-down). In our case, the particular up-
down-up-down conformation observed gets its stability from
H-bonding interactions with the N atoms of the coordinated
hydrazone ligands (N(2A) and N(2B) for the O(1W)–O(3W)
water cluster and N(2C) and N(2D) for the O(2W)–O(4W) water
cluster) and so from the tetrameric CuII cluster.
0.008
0.007
0.006
0.005
0.004
0.003
0.002
0.001
1.2
1.0
0.8
0.6
0.4
0.2
0.0
0
50
100 150 200 250 300
T [K]
0
50
100
150
200
250
300
T [K]
Fig. 5. Thermal variation of the molar magnetic susceptibility per CuII
tetramer for compound 1. Inset shows the thermal variation of the χMT
product. Solid and dashed lines are the best fit to the models (see text).
this antiferromagnetic coupling is provided by the thermal varia-
tion of χM that shows a rounded maximum at ∼75 K (Fig. 5).The
divergence observed in the χM plot at low temperatures is due
to the presence of a small amount of paramagnetic impurity (c).
Since the structure of 1 consists of two similar CuII tetramers,
initially the magnetic properties must be appropriated to a model
of an S = 1/2 tetramer with a regular square exchange pathway
(we assume that the four exchange pathways inside the square
are identical (J), as shown in Fig. 6):[18]
Variable Temperature Magnetic Susceptibility Measurement
Ng2β2 6ex + 12e2x + 30e3x
3kT 1 + 3ex + 7e2x + 5e3x
Ng2β2
4kT
The thermal variation of the product of molar magnetic suscepti-
bility and temperature per CuII tetramer (χMT) for 1 is presented
in the inset of Fig. 5. The χMT at room temperature was found to
be ∼1.10 emu K−1 mol−1, which is below the spin only value for
four non-interacting CuII, S = 1/2 ions (1.5 emu K−1 mol−1). On
lowering the temperature, the χMT product shows a progressive
decrease, already present at high temperatures, to reach a value
of ∼0.02 emu K−1 mol−1 at 2 K.This behaviour clearly indicates
that compound 1 presents an antiferromagnetic coupling, as
deduced form the decrease in the χMT product. Confirmation of
χtetram = (1 − c)
+ c
with x = J/kT
(1)
This model satisfactorily reproduces the magnetic data
over the entire temperature range with the following para-
meters (dashed line in Fig. 5): g = 2.333(4), J = −104.5(4)
K = −72.6 cm−1, and c = 0.67% (the Hamiltonian is written as
H = −JSiSi+1). Although this simple model is able to reproduce
the general shape of the χM plot, there are some differences with

‘Butterfly’-Shaped Water Tetramer in a Cu4 Complex Supported by a Hydrazone Ligand
1619
Cu
Cu
J
J
1.970
O(1C)
Cu(1)
Cu(3)
1.966
J
J
1.987
O(1C)
1.985
O(1A)
109.9Њ
114.3Њ O(1A)
Cu
Cu
2.001
Cu(1)
Cu(3)
O(1B)
1.957
O(1D)
1.923
1.990
112.7Њ
110.7Њ
Cu(4)
Cu(2)
Cu(2)
Cu(4)
O(1B)
O(1D)
Fig. 6. Coordination environments of the two CuII tetramers in compound 1 showing the bridging bond distances [Å] and angles
[◦]. The magnetic exchange scheme used to fit the magnetic data is located in the center (colour code: Cu = green, O = pink).
the experimental points, especially in the height of the maximum
and in the high temperature region. These differences suggest
that the model is not exact enough and therefore, we have utilized
a second model that considers two different CuII tetramers (each
one with a different exchange coupling constant, J1 and J2). In
fact, if the bridging bond distances and angles in both tetramers
are carefully examined (Fig. 6), we can see that the tetramer
containing Cu(1) and Cu(2) atoms (tetramer 12) presents big-
ger Cu–O–Cu bond angles than the tetramer containing Cu(3)
and Cu(4) atoms (tetramer 34). As can be seen in Fig. 6, this
model better reproduces the magnetic susceptibility data over
the entire temperature range with the following parameters (solid
compound 1 should be approximately −580 cm−1 in tetramer 12
and approximately −530 cm−1 in tetramer 34 (α = 112.7◦ and
114.3◦ in tetramer 12; and α = 109.9◦ and 110.7◦ in tetramer
34). This overestimation has already been observed in a simi-
lar tetranuclear CuII complex (although with alkoxo- instead of
phenoxo-bridges)[20] and was attributed to the lack of planarity
of the CuII coordination environments, because the magneto-
structural correlations established for double oxo bridged CuII
dimers imply a high planarity around the central Cu2O2 unit.
Thus, the torsion angles of the CuII coordination environments
in compound 1 (66.4◦ and 67.3◦ in tetramer 12 and 66.8◦ and
66.6◦) clearly show that both CuII tetramers are far from being
planar, resulting in an important decrease of the overlap of the
magnetic orbitals and, therefore, in a significant decrease of
the antiferromagnetic coupling.[19] A very recent and complete
study confirms that when this torsion angle is close to 90◦, the
magnetic orbitals are almost orthogonal and the magnetic cou-
line in Fig. 5): g = 2.373(4), J1 = −88.8(5) K = −61.7(3) cm−1
,
J2 = −133(1) K = −92(1) cm−1, and c = 0.64(1)% (the Hamil-
tonian is written as H = −JSiSi+1).
As expected from the antiferromagnetic coupling in com-
pound 1, the isothermal magnetization data at 2 K show a very
weak saturation value close to 0.03 µB, due to the small amount
of S = 1/2 paramagnetic impurity (not shown) (the paramagnetic
impurity is present at a very low level and can be attributed to
the presence of any CuII monomer arising from the decomposi-
tion of the tetranuclear clusters or from clusters presenting CuII
vacancies).
pling is expected to be ferromagnetic (between ∼5 and 10 cm−1
,
depending on the bridging Cu–O bond distance).[18,21–27]
Further proof of the dependence of the magnetic coupling on
the CuII coordination environment torsion angle is provided by
theonlythreestructurallyandmagneticallycharacterizedsimilar
CuII tetramers presenting a single oxo bridge.[28–30] These com-
plexes show J values of −87.8, −103, and −130 cm−1, close to
the values observed in compound 1.
Finally, it would be interesting to attempt to assign the two
different coupling constants to the two alternate CuII tetramers
present in compound 1. Although it is not straightforward,
because the torsion angles are very similar (see above), we can
assign the higher coupling constant (−92 cm−1) to tetramer 12,
because it presents higher Cu–O–Cu bond angles (see above).
As already mentioned above, even if both CuII tetramers
present quite similar structures, there are slight differences that
justify the use of a two-square model instead of a simple one-
square model.Thus, besides the Cu–O–Cu bond angles (112.72◦
and 114.38◦ in tetramer 12 versus 109.84◦ and 110.68◦ in
tetramer 34), there are also differences in the Cu–O bond dis-
tances, which are 1.966 and 1.985 Å around Cu(1); 1.957 and
1.990 Å around Cu(2); 1.970 and 1.987 Å around Cu(3); and
1.923 and 2.001 Å around Cu(4). These differences affect the
Cu–Cu distances, which are longer in tetramer 12 (3.322 Å
and 3.282 Å through O(1)A and O(1)B, respectively) than in
tetramer 34 (3.252 Å and 3.217 Å through O(1)C and O(1)D,
respectively). Of course, a more exact model might include a
rectangularmodelforbothtetramerswithtwodifferentexchange
coupling constants in each square (one for each different oxo
bridge), albeit, this model would include too many parameters
(four J values, g and c) and would not improve significantly
the magnetic fit because the model already reproduces very
satisfactorily the magnetic data.
Conclusions
The crystal structure of 1 showed the presence of two symmetri-
cally different types of water tetramers in the exo-pockets of the
Cu4 complexes, assuming an unusual ‘butterfly’ conformation.
In these water tetramers, each water molecule participates in
one H-bonding interaction with a neighbouring water molecule,
leaving one remaining hydrogen atom in a non-interacting posi-
tion within the cluster. The quantum mechanical calculations
showed that the stability of the ‘butterfly’ conformation of the
water tetramers, was the highest among all other conformers
and moreover, that this particular conformation was stabilized
by H-bonds with the ligand fragments of the complex molecule.
1showedantiferromagneticcouplingbetweentheCuII unitswith
From previous magneto-structural correlations established
between J and the Cu–O–Cu bond angles (α) in double
phenoxo-bridged CuII complexes (−J = 31.95α − 2463),[19] we
can deduce that the coupling constants per phenoxo bridge in

1620
S. Banerjee et al.
Table 4. Crystallographic and refinement data of 1
two different J values, proving the dependence of the coupling
constants on the Cu–O–Cu bridging angles.This dependence has
led us to assign the higher coupling constant (J2 = −92 cm−1
)
Complex 1
1
to the CuII tetramer 12, where the Cu–O–Cu bond angles are
higher.
Elemental formula
Formula weight
Crystal system
Space group
a [Å]
C60H56Cu4N8O16
1399.29
Monoclinic
P 1 2/n 1
21.1297(7)
13.4375(4)
21.6104(9)
90
Experimental
Materials
b [Å]
Benzhydrazide, O-vanillin, and copper(ii) perchlorate hexa-
hydrate were purchased from Sigma-Aldrich and were used as
received without further purification. All the solvents were of
reagent grade.
c [Å]
α [◦]
β [◦]
104.767(4)
90
5933.2(4)
γ [◦]
V [Å3]
Z
4
Physical Methods
D [Mg m−3
]
1.566
1.491
2864
Elemental analyses were carried out using a Perkin–Elmer 2400
II elemental analyzer. The infrared spectra were recorded on a
Perkin–Elmer RX I FT-IR spectrophotometer with a KBr disc.
The electronic spectra were recorded in acetonitrile on a Perkin–
Elmer Lambda 40 (UV-Vis) spectrophotometer.
µ [mm−1
F(000)
Crystal size [mm3]
θ range for data collection [◦]
No. of reflections collected
No. of independent reflections (Rint
Goodness of fit on F2
Final R indices [I > 2σ(I)]
R indices (all data)
Largest diff. peak and hole [e Å−3
]
0.2925 × 0.0947 × 0.0511
4.50 to 25.18◦
57648
Structures were fully optimized at the B3LYP/6–311+
G(3df,2p) level[14a–c] and characterized by computation of vibra-
tional frequencies. Gas phase stabilization energies were eval-
uated by MP2(ful)/6–311+G(3df,2p) single point calculations
(all electrons were included in the correlation treatment). The
Gaussian 03 suite of programs was used throughout.[14d]
The magnetic susceptibility measurements were carried
out in the temperature range 2–300 K with an applied mag-
netic field of 0.1T on a polycrystalline sample of compound
1 (mass = 16.20 mg) with a Quantum Design MPMS-XL-5
SQUID magnetometer. The isothermal magnetization was per-
formed on the same sample at 2 K with magnetic fields up to
5T.The susceptibility data were corrected for the sample holders
previously measured using the same conditions and for the dia-
magnetic contributions of the salt as deduced by using Pascal’s
constant tables (χdia = −760.8 × 10−6 emu mol−1).
)
19627 (Rint = 0.1498)
0.740
R1 = 0.0502, wR2 = 0.0476
R1 = 0.2314, wR2 = 0.0678
0.646 and −0.685
]
(λ = 0.71073 Å) source in the ϕ and ω scan modes at 200 K.
Data were processed using the CrysAlis-CCD programs.[31]
The structure was determined by direct methods procedures
in SHELXS[32a] and refined by full-matrix least-squares meth-
ods, on F2’s, in SHELXL.[32b] The crystallographic data and the
refinement results are listed in Table 4.
Supplementary Information
CCDC number for 1 is CCDC 673139. These data can
be obtained free of charge at www.ccdc.cam.ac.uk (or from
Cambridge Crystallographic Data Centre, 12, Union Road,
Cambridge CB2 1EZ, UK; email: deposit@ccdc.cam.ac.uk).
Synthesis of the Ligand
The hydrazone ligand was obtained by condensing benzhy-
drazide with O-vanillin in a 1:1 molar ratio in methanol for
4.5 h.[3] The crystalline ligand having an off-white colour sepa-
rated out on standing over night. The ligand was re-crystallized
from methanol and was used for the synthesis of the complex.
Caution! Perchlorate salts are potentially explosive, should
only be prepared in small quantities and handled with care.
Accessory Publication
The Gaussian archive entries of the frequency calculations are
provided in the Accessory Publication on the Journal’s website.
Synthesis of Metal Complex
Acknowledgement
The complex was synthesized using 1 mmol of Cu(ClO4)2·6H2O
(0.370 g) in methanol-water (1:1) solution containing the lig-
and (1 mmol, 0.270 g) at room temperature. Crystals were
obtained on standing this solution at room temperature for
1 week. Yield: 65%. Anal. Calcd. for C60H56Cu4N8O16 [%]:
C, 51.50; H, 4.03; N, 8.01. Found: C, 51.53; H, 4.09; N, 8.06. IR
data (cm−1): 3433(w), 1617(m), 1498(m), 1438(m), 1361(m),
1256(m), 1247(m), 1218(m), 1170(s), 1069(s), 972(s), 848(s),
709(m), 528(s), 448(s), 344(s).
The authors sincerely acknowledge DRDO andAICTE New Delhi, the Euro-
peanUnion(MAGMANetnetworkofexcellence)andtheSpanishMinisterio
de Educación y Ciencia (Projects MAT2007–61584 and Consolider-Ingenio
2010 CSD 2007–00010 in Molecular Nanoscience) for funding this project.
We also thank Professor Tim Clark, of University of Erlangen-Nürnberg,
Germany, for hosting this work in the CCC and the Regionales Rechenzen-
trum Erlangen (RRZE) for a generous allotment of computer time. We also
thank the Generalitat Valenciana (Project PROMETEO/2009/095).
References
UV-Vis data [nm]: 460–340 (n–π*), 300–230 (π–π*), 310–
360 (LMCT), 535–560 (d–d).
[1] (a) P. Comba, M. Kerscher, Coord. Chem. Rev. 2009, 253, 564.
doi:10.1016/J.CCR.2008.05.019
(b) B. F. Gherman, C. J. Cramer, Coord. Chem. Rev. 2009, 253, 723.
doi:10.1016/J.CCR.2007.11.018
X-ray Data Collection and Structure Determination
Data collection for 1 was made using an Oxford Diffrac-
tion Gemini diffractometer equipped with a Mo Kα radiation
(c) R. J. Deeth, A. Anastasi, C. Diedrich, K. Randell, Coord. Chem.
Rev. 2009, 253, 795. doi:10.1016/J.CCR.2008.06.018

‘Butterfly’-Shaped Water Tetramer in a Cu4 Complex Supported by a Hydrazone Ligand
1621
[2] (a) H. Kozlowski, T. Kowalik-Jankowska, M. Jezowska-Bojczuk,
Coord. Chem. Rev. 2005, 249, 2323. doi:10.1016/J.CCR.2005.04.027
(b) I. P. Beletskaya, A. V. Cheprakov, Coord. Chem. Rev. 2004, 248,
2337. doi:10.1016/J.CCR.2004.09.014
[16] (a) J. F. Pérez, C. Z. Hadad, A. Restrepo, Int. J. Quantum Chem. 2008,
108, 1653. doi:10.1002/QUA.21615
(b) L. J. Yuan, G. X. Huang, S.-P. Chen, Z. Peng, J. Coord. Chem.
2008, 61, 3575. doi:10.1080/00958970802087060
(c) G. Ambrosi, M. Formica, V. Fusi, L. Giorgi, M. Micheloni, Coord.
Chem. Rev. 2008, 252, 1121. doi:10.1016/J.CCR.2007.09.027
[3] J. Chakraborty, S. Thakurta, G. Pilet, D. Luneau, S. Mitra, Polyhedron
2009, 28, 819. doi:10.1016/J.POLY.2008.12.018
[4] A. Datta, C. R. Choudhury, P. Talukder, S. Mitra, L. Dahlenburg,
T. Matsushita, J. Chem. Res. (S) 2003, 642.
[5] P. Talukder, A. Datta, S. Mitra, G. Rosair, M. S. El Fallah, J. Ribas,
Dalton Trans. 2004, 4161. doi:10.1039/B413084G
[6] C. R. Choudhury, S. K. Dey, R. Karmakar, C.-D.Wu, C.-Z. Lu, M. S. El
Fallah, S. Mitra, N. J. Chem. 2003, 27, 1360. doi:10.1039/B300217A
[7] (a) D. J. Hodgson, Prog. Inorg. Chem. 1975, 19, 173. doi:10.1002/
9780470166208.CH4
(b)K. S. Murray,Adv. Inorg. Chem. 1995, 43, 261. doi:10.1016/S0898-
8838(08)60119-1
[8] (a) S. W. Benson, E. D. Siebert, J. Am. Chem. Soc. 1992, 114, 4269.
doi:10.1021/JA00037A034
(b) R. Ludwig, Angew. Chem. Int. Ed. 2001, 40, 1808. doi:10.1002/
1521-3773(20010518)40:10<1808::AID-ANIE1808>3.0.CO;2-1
[9] G. Jiang, J. Bai, H. Xing, Y. Li, X. You, Cryst. Growth Des. 2006, 6,
1264. doi:10.1021/CG0600389
[10] M. Zuhayra, W. U. Kampen, E. Henze, Z. Soti, L. Zsolnai, G. Huttner,
F. Oberdorfer, J. Am. Chem. Soc. 2006, 128, 424. doi:10.1021/
JA057142J
(c) K. Müller-Dethlefs, P. Habza, Chem. Rev. 2000, 100, 143 and
literature cited there.
[17] W. Schiessl, R. Puchta, Ž. D. Bugarcˇic´, F. W. Heinemann, R. van Eldik,
Eur. J. Inorg. Chem. 2007, 1390. doi:10.1002/EJIC.200601028
[18] S. Roy, T. N. Mandal, A. K. Barik, S. Gupta, R. J. Butcher,
M. S. El Fallah, J. Tercero, S. K. Kar, Polyhedron 2008, 27, 105.
doi:10.1016/J.POLY.2007.08.040
[19] F. Tuna, L. Patron, Y. Journaux, M. Andruh, W. Plass, J. C. Trombe,
J. Chem. Soc., Dalton Trans. 1999, 539. doi:10.1039/A806338I
[20] A. Burkhardt, A. Buchholz, H. Görls, W. Plass, Eur. J. Inorg. Chem.
2006, 3400. doi:10.1002/EJIC.200600404
[21] E. Ruiz, C. De Graaf, P. Alemany, S. Alvarez, J. Phys. Chem. A 2002,
106, 4938. doi:10.1021/JP015565B
[22] C. J. Matthews, K. Avery, Z. Xu, L. K. Thompson, L. Zhao,
D. O. Miller, K. Biradha, K. Poirier, M. J. Zaworotko, C. Wilson,
A. E. Goeta, J. A. K. Howard, Inorg. Chem. 1999, 38, 5266.
doi:10.1021/IC990598F
[23] Z. Xu, L. K. Thompson, D. O. Miller, Dalton Trans. 2002, 2462.
[24] L. N. Dawe,T. S. M.Abedin,T. L. Kelly, L. K.Thompson, D. O. Miller,
L. Zhao, C. Wilson, M. A. Leech, J. A. K. Howard, J. Mater. Chem.
2006, 16, 2645. doi:10.1039/B602595A
[25] S. Roy, T. N. Mandal, A. K. Barik, S. Pal, R. J. Butcher,
M. S. El Fallah, J. Tercero, S. K. Kar, Dalton Trans. 2007, 1229.
doi:10.1039/B617587B
[26] N. Saha, K. M. Dutta, J. Inorg. Nucl. Chem. 1981, 43, 1405.
doi:10.1016/0022-1902(81)80055-3
[27] W. H. Gu, X. Y. Chen, L. H. Yin, A. Yu, X. Q. Fu, P. Cheng, Inorg.
Chim. Acta 2004, 357, 4085. doi:10.1016/J.ICA.2004.06.006
[28] A. Fragoso, M. L. Kahn, A. Castineiras, J.-P. Sutter, O. Kahn, R. Cao,
Chem. Commun. 2000, 1547. doi:10.1039/B002360O
[11] A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn, G. C. Verschoor, J.
Chem. Soc., Dalton Trans. 1984, 1349. doi:10.1039/DT9840001349
[12] J.-Q. Sun, J. Zang, Z.-F. Ju, G.-Y. Yang, Aust. J. Chem. 2005, 58, 572.
doi:10.1071/CH05110
[13] For n = 4 and 6 the water molecules have been arranged in the
alternating ‘butterfly’ structure up-down-up-down (n = 4) and up-
down-up-down-up-down (n = 6).
[14] (a)A. D. Becke, J. Chem. Phys. 1993, 98, 5648. doi:10.1063/1.464913
(b) C. Lee, W. Yang, R. G. Parr, Phys. Rev. B: Condens. Matter Phys.
1988, 37, 785.
[29] G. Mislin, E. Graf, M. W. Hosseini, A. Bilyk, A. K. Hall,
J. M. Harrowfield, B. W. Skelton, A. H. White, Chem. Commun. 1999,
373. doi:10.1039/A809184F
(c) P. J. Stephens, F. J. Devlin, C. F. Chabalowski, M. J. Frisch, J. Phys.
Chem. 1994, 98, 11623. doi:10.1021/J100096A001
[30] J. P. Laurent, J. J. Bonnet, F. Nepveu, H.Astheimer, L. Walz, W. Haase,
J. Chem. Soc., DaltonTrans. 1982, 2433. doi:10.1039/DT9820002433
[31] Programs CrysAlis-CCD and -RED 2005 (Oxford Diffraction Ltd.:
Abingdon, UK).
[32] G. M. Sheldrick, SHELX97 1997 – Programs for crystal structure
analysis: (a) structure determination (SHELXS) and (b) refinement
(SHELXL) (University of Göttingen: Germany).
(d) M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr, T. Vreven, et al.,
Gaussian 03, Revision B.03 2004 (Gaussian Inc.: Wallingford, CT).
[15] L. A. Curtiss, K. Raghavachari, G. W. Trucks, J. A. Pople, J. Chem.
Phys. 1991, 94, 7221. doi:10.1063/1.460205