An unusual half-open cubane-like tetranuclear copper(II) complex supported by both μ-alkoxo and μ3-hydroxo bridges: Structure, magnetic properties, EPR and DFT studies

Article abstract of DOI:10.1016/j.poly.2013.01.044

The compound 4-methyl-2,6-bis[(2-(pyridin-2-yl)hydrazinylidene)methyl] phenol (H-PHMP) was prepared by the direct condensation of 2,6-diformyl-4- methylphenol (PCDA) and 2-hydrazinopyridine. It reacts with copper(II) nitrate to yield a self-assembled, both alkoxo- and hydroxo-bridged, tetranuclear homoleptic copper(II) complex, [Cu₄(PHMP) ₂(μ₃-OH)₂(NO₃) ₂](NO₃)₂·9H₂O (1), with a [Cu₄(μ-O)₂(μ₃-OH)₂] ⁴⁺ half-open cubane core. Compound 1 is composed of two dinuclear [Cu₂(PHMP)] units linked by the two μ-bridging alkoxide oxygen atoms of the PCDA backbone. Variable temperature magnetic susceptibility measurements indicate the existence of antiferromagnetic exchange interactions (J₁ = -304.0 cm^-1, J₂ = -127.3 cm^-1 and g = 2.05) in complex 1. The exchange pathway parameters were also evaluated from density functional theoretical calculations and fully agree with the experimental findings.

Full text of DOI:10.1016/j.poly.2013.01.044

Polyhedron 53 (2013) 269–277
Contents lists available at SciVerse ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
An unusual half-open cubane-like tetranuclear copper(II) complex supported
by both
l-alkoxo and l3-hydroxo bridges: Structure, magnetic properties,
EPR and DFT studies
b
c,
c
a
Somnath Roy a, , Ray J. Butcher , Mohamed Salah El Fallah , Javier Tercero , João Costa Pessoa
⇑
⇑
a Centro de Química Estrutural, Instituto Superior Técnico, Universidade Técnica de Lisboa, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal
b Department of Chemistry, Howard University, 2400 Sixth Street, N. W., Washington, DC 200 59, USA
c Departament de Quimica Inorganica and Institut de Nanociencia i Nanotecnologia, Universitat de Barcelona, Marti i Franques 1-11, 08028 Barcelona, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
The compound 4-methyl-2,6-bis[(2-(pyridin-2-yl)hydrazinylidene)methyl]phenol (H-PHMP) was pre-
pared by the direct condensation of 2,6-diformyl-4-methylphenol (PCDA) and 2-hydrazinopyridine. It
reacts with copper(II) nitrate to yield a self-assembled, both alkoxo- and hydroxo-bridged, tetranuclear
Received 14 August 2012
Accepted 5 January 2013
Available online 9 February 2013
homoleptic copper(II) complex, [Cu4(PHMP)2(
OH)2]4+ half-open cubane core. Compound 1 is composed of two dinuclear [Cu2(PHMP)] units linked
by the two -bridging alkoxide oxygen atoms of the PCDA backbone. Variable temperature magnetic sus-
l3-OH)2(NO3)2](NO3)2ꢀ9H2O (1), with a [Cu4(
l-O)2(l3-
Keywords:
Synthesis
Self-assembly
Heterocylic ligand
Cu(II) compound
IR, UV–Vis, NMR, EPR spectroscopy
Magnetic property
DFT study
l
ceptibility measurements indicate the existence of antiferromagnetic exchange interactions (J1
=
ꢁ304.0 cmꢁ1, J2 = ꢁ127.3 cmꢁ1 and g = 2.05) in complex 1. The exchange pathway parameters were
also evaluated from density functional theoretical calculations and fully agree with the experimental
findings.
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
grown in the design and in the magnetic properties of polynuclear
molecules and molecule based coordination architectures with
Supramolecular chemistry is an actively pursued area of re-
search which involves the relation between organic ligands and
metal ions to achieve the result of multinuclear aesthetical frame-
works in the nanometer range [1–5]. Hence, the use of suitable
bridging units and well-designed polydentate ligands can afford
an exciting array of polynuclear coordination complexes with
new aesthetical building blocks [3,4]. High nuclearity transition
metal complexes are studied enthusiastically because of their rel-
evance for the multi-metal active sites of metalloproteins as well
as for their importance in the field of molecular magnetism [6,7].
Copper(II) complexes are of particular interest for both structural
and functional modes of performance [8,9].
Over the years particular emphasis has been placed on copper
because of its central role in biology and its diverse role and plas-
ticity effect in coordination chemistry [10,11]. Moreover, from a
magneto-structural point of view, multi-copper complexes have
been extensively studied by both experimental and theoretical ap-
proaches [8,9]. In addition, they find use in the development of
new functional molecule based materials [12–15]. Interest has also
mono and multi-dimensional frameworks. The main inspiration
of such studies stems from the understanding of the fundamental
science of magnetic interactions and magneto-structural correla-
tions in molecular systems [12–15].
In this context, tetranuclear cubane-like copper(II) complexes
have acquired unique attention because of their scarce findings
in nature [16–18]. These complexes are particularly interesting
for their intermediate size between the simplest binuclear com-
plexes and bulk materials, which may develop completely new
magnetic properties, thus providing good models of nanometer
sized magnetic particles [2,5]. Nevertheless, although many chiral
or achiral polydentate ligands containing hydroxyl groups as ter-
minal coordinating atoms have been employed for the synthesis
of cubane-like copper(II) complexes [16–18], examples of PCDA
derived ligands to form half-open cubane-like copper(II) com-
plexes have not yet been reported. Hence, detailed new examples
of different systems with well-designed structural features and
correlated physical properties are needed to further understand
the magnetic exchange between bridged metals.
The present work explores the synthesis and characterization of
an unusual half-open cubane-like tetranuclear copper(II) complex
of a PCDA derived polydentate ligand. Since these particular types
of open half-cubane are rare [18], we also report an experimental
⇑
Corresponding authors. Tel.: +351 927721417; fax: +351 218419239 (S. Roy).
E-mail addresses: somnath.roy@ist.utl.pt (S. Roy), salah.elfallah@qi.ub.es
(M.S. El Fallah).
0277-5387/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.poly.2013.01.044

270
S. Roy et al. / Polyhedron 53 (2013) 269–277
study of its magnetic properties and carry out DFT calculations to
determine the relevant exchange pathway parameters.
spectrometer was operated at ꢂ9.51 GHz with a frequency modu-
lation of 100 kHz. While keeping the resolution at 2048 points, the
microwave power was adjusted to 20 dB attenuations and the re-
ceiver gain was set to either 3.2 ꢃ 105 or 4.0 ꢃ 105. To improve
the signal to noise ratio, 10 scans of each sample were accumu-
lated. The spectral acquisition parameters were constant for each
batch of experiments. All measurements were done using 3 mm
quartz tubes (Wilmad 707-SQ-250M). The EPR spectra were simu-
lated using an EPR simulation program [21].
2. Experimental
2.1. Materials
2-Hydrazino pyridine was purchased from Aldrich and was
used as received. Other commercially available chemicals and sol-
vents were used and purified by standard procedures [19].
2.3. Synthesis and characterization of the compounds
2.2. Physical measurements
2.3.1. Preparation of PCDA
Elemental analyses (carbon, hydrogen and nitrogen) of H-PHMP
and metal complex 1 were carried out at the Laboratório de Anális-
es do Instituto Superior Técnico (Lisbon, Portugal) with a Perkin El-
mer CHNS/O analyzer 2400. The electronic spectrum of 1 in
purified CH3OH solution was recorded on a Hitachi model U-
3501 spectrophotometer. The 1H NMR (in d6-DMSO) spectrum of
the ligand was recorded with a Bruker AM 300L (300 MHz) super-
conducting FT-NMR. IR spectra (KBr pellet, 4000–500 cmꢁ1) were
recorded on a Perkin Elmer model 883 infrared spectrophotometer.
The mass spectrum of the ligand was recorded on a JEOLJMS-AX
500 mass spectrometer. Simple single point room temperature
magnetic susceptibility was measured with a PAR 155 vibrating
sample magnetometer. Variable temperature magnetic susceptibil-
ity measurements for the compound were carried out on polycrys-
talline samples at the Servei de Magnetoquímica of the Universitat
de Barcelona, with a Quantum Design SQUID MPMS-XL susceptom-
eter apparatus in the range 2–300 K under magnetic fields of
approximately 500 G (2–30 K) and 1000 G (35–300 K). Diamag-
netic corrections were estimated by using the appropriate Pascal
constant values [20]. The X-band EPR spectra were measured
either at room temperature (RT) or at liquid nitrogen (LN) temper-
ature (77 K) on a Bruker ESP 300E spectrometer. The ESP 300E
PCDA was prepared following an establish method [22].
2.3.2. Synthesis of H-PHMP
A methanolic solution (20 mL) of 2-hydrazino pyridine (0.218 g,
2 mmol) was added dropwise to a methanolic solution (15 mL) of
2,6-diformyl-4-methylphenol (0.164 g, 1 mmol) with constant stir-
ring (Scheme 1). The stirring was continued for 30 min, then the
mixture was refluxed for 5 h and finally cooled to room tempera-
ture and filtered. The volume of the filtrate was reduced to
15 mL with a rotary evaporator. A yellow microcrystalline solid
was obtained, which was filtered, washed thoroughly with cold
methanol and dried in vacuo over fused CaCl2. Yield: 0.305 g, 80
%. M.P. 245 °C (decomp.). MS (m/z): 346 (M+, 100%). IR/cmꢁ1
3369 N–H); 1661(s), 1493 C@C), 1597(s) CO/CN), 1051(s)
N–N), 1023(s) m(py) Elemental Anal. Calc. for C19H18N6O
:
(m
(
m
(m
(m
.
(MM = 346.39): C, 65.89; H, 5.20; N, 24.27. Found: C, 65.7; H,
5.3; N, 24.2%.
The ESI-mass spectrum of H-PHMP is shown in Fig. SI-1. The
molecular ion peak was found at m/z 369 (M+Na+) with 100% prob-
ability for the formation of H-PHMP.
CH3
CH3
MeOH
N
N
OH
2
+
HN
NH
Reflux
NH2
N
NH
OHC
CHO
N
N
OH
H-PHMP
2+
CH3
CH3
Cu(NO3)2.6H2O
N
N
OH
HN
NH
N
O
N
N
MeOH
Cu
NH
Cu
NH
N
N
O
H
N
Scheme 1. Schematic representation and preparation of H-PHMP and one perspective of its binding mode in the complex isolated.

S. Roy et al. / Polyhedron 53 (2013) 269–277
271
Table 1
2.3.3. Synthesis of [Cu4(PHMP)2(
l
3-OH)2(NO3)2](NO3)2ꢀ9H2O (1)
Experimental crystallographic data for complex 1.
H-PHMP (0.346 g, 1 mmol) was added to a hot solution of
Cu(NO3)2ꢀ6H2O (0.591 g, 2.0 mmol) in purified CH3OH (30 mL).
The suspension was stirred for an additional 1 h until complete dis-
solution of the compound. The resulting deep green solution was
filtered to remove any solid and left at room temperature. Rectan-
gular dark green crystals were isolated after standing for several
Empirical formula
Formula weight
Temperature (K)
Wavelength (Å)
Crystal system
Space group
Unit cell dimensions
a (Å)
C38H54Cu4N16O25
1389.13
200(2)
0.71073
monoclinic
P21/n
days. Yield: 0.749 g, 80%. IR/cmꢁ1: 3401, 3385 (
1539(s) ( CO/CN); 1084(s) ( N–N); 1019(s) ( py). UV–Vis (kmax/nm):
453, 417, 375, 345, 248. lRT 2.95 B.M. Elemental Anal. Calc. for
[Cu4(PHMP)2(
mNH/H2O); 1625(s),
16.5486(8)
17.0167(5)
20.2529(9)
90
m
m
m
b (Å)
c (Å)
l
3-OH)2(NO3)2](NO3)2ꢀ9H2O (1) (MM = 1387.12): C,
a
(°)
32.87; H, 3.75; N, 16.15. Found: C, 32.8; H, 3.8; N, 16.1%.
b (°)
107.944(5)
c
(°)
90
5425.9(4)
Volume (Å3)
2.3.4. Single crystal X-ray diffraction study of 1
z
4
Densitycalc (Mg mꢁ3
)
1.698
1.645
2840
Selected crystal data of 1 are given in Table 1 and selected met-
rical parameters are also listed in Table 2. Data collection of 1 was
made using an Oxford Diffraction Gemini diffractometer equipped
Absorption coefficient (mmꢁ1
F(000)
)
Crystal size (mm3)
h range (°) for data collection
Index ranges
0.52 ꢃ 0.47 ꢃ 0.38
with a Mo Ka radiation (k = 0.71073 Å) source in the x scan mode
4.68–32.64
at 200(2) K. Data collection, cell parameters and data reductions
were carried out using the CRYSALIS RED (Oxford Diffraction Ltd., ver-
sion 1.171.32.15) [23]. The structure was solved by conventional
direct methods and refined by full-matrix least squares methods
using F2 data. SHELXS-97 and SHELXL-97 programs [24] were used
for the structure solution and refinement, respectively. Positions
of the hydrogen atoms were calculated from the geometry of the
molecular skeleton. For the water molecules some H’s could be lo-
cated in reasonable positions in the difference Fourier maps. Sub-
sequently they were idealized and allowed to refine as a group.
For some water molecules no reasonable positions could be located
in the difference Fourier maps so these H’s were not included in the
refinement. The hydrogen atoms on O2(A/B) were located in differ-
ence Fourier maps and then idealized. Refinement of F2 was done
against all reflections.
ꢁ25 6 h 6 24
ꢁ24 6 k 6 17
ꢁ28 6 l 6 28
0.915
Goodness-of-fit on F2
Completeness to theta = 25.00°
Independent reflections
Absorption correction
Max. and min. transmission
Refinement method
Data/restraints/parameters
Reflections collected
98.6%
17688 [R(int) = 0.0575]
Semi-empirical from equivalents
1.00000 and 0.84022
Full-matrix least-squares on F2
17688/10/758
39493
R1 = 0.0625, wR = 0.1676
R1 = 0.1394, wR2 = 0.1866
1.865 and ꢁ0.983
Final R indices [I > 2
R indices (all data)
r
(I)]
Largest diff. peak and hole (e Åꢁ3
)
Table 2
Selected bond distances (Å) and angles (°) for 1.
2.3.5. DFT computational details
Bond distances (Å)
The computational strategy to calculate the exchange coupling
constants in transition metal complexes has been described previ-
ously. [25,26] For the calculation of the exchange coupling con-
stants for any polynuclear complex with n different exchange
constants, at least the energy of n + 1 spin configurations must
be calculated. For 1 the energy corresponding to eight different
spin distributions was calculated to obtain (by least-squares fit-
ting) the six exchange coupling constants, with their standard
deviations indicated in parentheses. The eight spin distributions
calculated correspond to the high spin solution, S = 2 [+,+,+,+], four
S = 1, [ꢁ,+,+,+], [+,ꢁ,+,+], [+,+,ꢁ,+], [+,+,+,ꢁ] and three S = 0,
[ꢁ,ꢁ,+,+], [ꢁ,+,+,ꢁ], [ꢁ,+,+,ꢁ]. The calculations were carried out
with GAUSSIAN03 [27] using guess functions generated with Jaguar
6.0 [28]. The hybrid B3LYP functional [29] was used in all calcula-
tions. A triple-n all electron basis set with a two ‘p’ polarization
function was employed for copper atoms [30] and a double-n all
electron basis set for the other elements, as proposed by Schaefer
et al. [31]. All energy calculations were done including 10ꢁ8 den-
sity-based convergence criterion.
Cu(1)–O(2A)
Cu(1)–N(2A)
Cu(1)–O(1A)
Cu(1)–N(1A)
Cu(1)–O(11N)
Cu(1)–O(1B)
Cu(1)–Cu(2)
Cu(2)–O(2A)
Cu(2)–N(3A)
Cu(2)–O(1A)
Cu(2)–N(4A)
Cu(2)–O(2B)
Cu(2)–O(12N)
Cu(3)–N(2B)
Cu(3)–O(2B)
Cu(3)–O(1B)
Cu(3)–N(1B)
1.924(3)
1.940(4)
1.944(3)
1.973(3)
2.404(3)
2.882(3)
2.930(7)
1.914(3)
1.944(3)
1.947(3)
1.964(3)
2.397(3)
3.265(5)
1.938(3)
1.944(3)
1.968(3)
1.970(4)
Cu(3)–O(2A)
2.487(3)
2.534(3)
2.898(7)
1.931(4)
1.932(3)
1.955(3)
1.956(4)
2.339(3)
3.126(3)
1.312(6)
1.342(5)
1.284(5)
1.351(5)
1.294(5)
1.365(5)
1.275(6)
1.376(5)
Cu(3)–O(21N)
Cu(3)–Cu(4)
Cu(4)–N(3B)
Cu(4)–O(2B)
Cu(4)–O(1B)
Cu(4)–N(4B)
Cu(4)–O(22N)
Cu(4)–O(1A)
N(2A)–C(6A)
N(2A)–N(5A)
N(3A)–C(14A)
N(3A)–N(6A)
N(2B)–C(6B)
N(2B)–N(5B)
N(3B)–C(14B)
N(3B)–N(6B)
Bond angles (°)
O(2A)–Cu(1)–O(1A)
O(2A)–Cu(1)–O(11N)
O(1A)–Cu(1)–O(11N)
O(2A)–Cu(1)–O(1B)
O(1A)–Cu(1)–O(1B)
O(11N)–Cu(1)–O(1B)
O(2B)–Cu(3)–O(1B)
O(2B)–Cu(3)–O(2A)
O(1B)–Cu(3)–O(2A)
O(2B)–Cu(3)–O(21N)
O(1B)–Cu(3)–O(21N)
O(2A)–Cu(3)–O(21N)
Cu(1)–O(1A)–Cu(2)
Cu(1)–O(1A)–Cu(4)
Cu(2)–O(2A)–Cu(3)
80.87(11)
97.15(12)
103.59(12)
75.71(10)
82.49(10)
169.95(11)
81.23(12)
81.77(10)
85.39(10)
89.62(12)
85.90(12)
168.59(10)
97.73(13)
96.70(10)
94.89(10)
O(2A)–Cu(2)–O(1A)
O(2A)–Cu(2)–O(2B)
O(1A)–Cu(2)–O(2B)
O(2A)–Cu(2)–O(12N)
O(1A)–Cu(2)–O(12N)
O(2B)–Cu(2)–O(12N)
O(2B)–Cu(4)–O(1B)
O(2B)–Cu(4)–O(22N)
O(1B)–Cu(4)–O(22N)
O(2B)–Cu(4)–O(1A)
O(1B)–Cu(4)–O(1A)
O(22N)–Cu(4)–O(1A)
Cu(1)–O(2A)–Cu(3)
Cu(3)–O(1B)–Cu(1)
Cu(4)–O(2B)–Cu(3)
81.04(11)
84.82(10)
91.29(11)
73.89(10)
78.46(11)
157.50(9)
81.86(12)
100.06(13)
91.03(12)
71.86(10)
75.93(10)
165.34(10)
106.14(11)
91.83(10)
96.81(12)
3. Results and discussion
Using the PCDA precursor, the polytopic ligand H-PHMP was
prepared by a one step synthesis, as depicted in Scheme 1. The pre-
pared H-PHMP was characterized by IR, NMR and ESI-MS studies,
as well as elemental analysis. Upon dissolving and mixing H-PHMP
and Cu(NO3)2 in methanol, after several days crystals of 1, suitable
for X-ray diffraction studies, were obtained.

272
S. Roy et al. / Polyhedron 53 (2013) 269–277
copper atoms are bridged with two by two by two alkoxo oxygen
atoms (O1A/B) from two anionic ligands PHMP and three by
three by two hydroxo groups (O2A/B). The molecular structure of
[Cu4(PHMP)2(l3-OH)2(NO3)2](NO3)2ꢀ9H2O (1) is shown in Fig. 1
and the corresponding tetranuclear core is shown in Fig. 2. Com-
plex 1 crystallizes in the monoclinic space group P121/n1; it con-
tains nine H2O molecules of crystallization, this being consistent
with the analytical results. Out of the four copper(II) centers, three
of them, namely Cu1, Cu2 and Cu4, are penta-coordinated with a
N2O3 binding set, while the remaining copper(II) Cu3 center is
hexa-coordinated with a N2O4 binding set.
The coordination environment of Cu1, Cu2 and Cu4 are best de-
scribed as square-pyramidal (
s = 0.83 for Cu1, 0.38 for Cu2 and
0.50 for Cu4, where = 0 and 1 are for the perfect square pyramidal
s
and trigonal bipyramidal geometries, respectively) [32] with the
basal plane established by the O2A, N2A, N1A, O1A atoms for
Cu1; O2A, N3A, N4A, O1A atoms for Cu2 and O2B, N3B, N4B, O1B
atoms for Cu4 of the PCDA units of the two Schiff base ligands.
The apical positions of Cu1 and Cu4 are occupied by the two nitro-
gen atoms (O11N and O22N) of the two NO3ꢁ ligands at rather long
donor distances, while the apical position of Cu2 is satisfied by O2B
from the hydroxo group. The donor atoms of the basal coordination
planes of Cu1 and Cu4 centers lie in almost same plane and the re-
lated copper(II) centers are slightly displaced (0.118 Å for the Cu1
center and 0.132 Å for the Cu4 center) towards the apical positions
(O11N and O22N). The donor atoms of all basal coordination planes
of Cu2 and Cu3 centers deviate by 0.013 and 0.006 Å, respectively
(for Cu1 and Cu4, it is 0 Å), from the corresponding mean planes,
and the related copper(II) centers are only slightly displaced
(0.118 Å for Cu1, 0.051 Å for Cu2, 0.026 Å for Cu3 and 0.132 Å for
Cu4) towards the apical position (O11N for Cu1, O2B for Cu2,
O2A for Cu3 and O22N for Cu4).
Fig. 1. Representation of the molecular structure of compound 1.
By a self-assembly process, two [Cu2(PHMP)] building blocks
form the tetranuclear complex 1 with a half-open cubane-like
[Cu4(l-O)2(l
3-OH)2]4+ core. The vertices of the cube are alterna-
tively occupied by four copper(II) and four oxygen atoms (O1A/B
and O2A/B), leading to a interlocked half-open cubane structure.
Consequently, the two copper(II) centers are linked in a l3-bridg-
ing fashion by two O2A/B atoms of the hydroxo groups. The
C13(A/B) alkoxide oxygen atoms O1A/B are deprotonated, afford-
ing a monoanionic ligand fragment.
ꢁ
Out of the four NO3 anions, two are bonded to copper(II)
centers and other two serve as counter ions. Each PHMP acts as a
hexadentate ligand, binding two copper(II) ions, leading to two
five-membered and two six-membered chelate rings with bite
angles of ꢂ82° and ꢂ90°, respectively.
According to the Cu–O bond lengths in the [Cu4(l-O)2(l3-
OH)2]4+ cores, the structure of 1 should be described as consisting
Fig. 2. Representation and atom numbering scheme of the tetranuclear core in 1.
of two dimeric [Cu2(l-O)(
l
3-OH)]2+ units with short (ꢂ1.91 Å) and
long (ꢂ1.95 Å) Cu–O bonds within the dinuclear units. The two
resulting Cu1Cu2 and Cu3Cu4 units are constituted by the respec-
tive copper-ligand moieties. These short and long Cu–O contacts
The 1H NMR spectrum of H-PHMP in d6-DMSO is shown in
Fig. SI-2. The 1H NMR (dH, ppm) spectrum displays a singlet at d
2.25 (3H) assignable to C4–CH3 (benzene ring). Singlets at d 7.36
(2H) and d 8.26 (2H) are due to C3/5–H (benzene ring) and two
imine –C–H, respectively, whilst multiplets at d 6.75 (2H) and at
d 7.62 (2H) are ascribed to C4–H (pyridine rings) and C5–H (pyri-
dine rings), respectively. A doublet at d 7.02 (2H) and a triplet at
d 8.10 (2H) are ascribed to C3–H (pyridine rings) and C6–H (pyri-
dine rings), respectively. A singlet at d 10.97 (2H) is assignable to
the two –NH protons of PHMP. A singlet at d 11.64 (1H) confirms
the presence of the phenolic –OH.
lead to the non-planarity of the [Cu2(l-O)(l
3-OH)]2+ core frag-
ments of the dinuclear units. The dihedral angles between the
CuO2 planes of the resulting butterfly-type structures of the
Cu1Cu2 and Cu3Cu4 units are 171° and 158°, respectively. Hence,
this leads to Cu–O–Cu bond angles at the bridging oxygen atoms
within the dimeric units Cu1Cu2 and Cu3Cu4 in the range between
98° and 96° (Table 2), which are relatively small angles for alkoxo-
bridged compounds with a Cu2O2 core fragment.
It is interesting to note that the Cu1Cu2 pair can be virtually
transformed into the analogous dimer found in complex 1 by sim-
ply rotating through an angle of 90° in the plane of the paper
around the C2 axis. Complex 1 also has an inversion center. The
Cu1ꢀ ꢀ ꢀO1B and Cu4ꢀ ꢀ ꢀO1A axial contacts (2.882 and 3.126 Å) are
very weak when compared with the Cuꢀ ꢀ ꢀO axial distances usually
3.1. Molecular structure of 1
From the X-ray diffraction data, it was found that 1 consists of a
half-open ‘cubic’ unit, [Cu4(l-O)2(l
3-OH)2]4+, in which the four

S. Roy et al. / Polyhedron 53 (2013) 269–277
273
from the point of view of the skeleton atom arrangements in the
cubane core.
3.2. Further characterization of 1
Polycrystalline samples of 1 gave satisfactory C, H and N analy-
ses. Spectral properties of the complexes are briefly described
below.
3.2.1. IR spectra
The IR spectrum of 1 was recorded in the region 4000–
500 cmꢁ1. The stretching frequency of the imine (–CH@N) group
in the ligand was observed near 1597 cmꢁ1; upon complexation
it is shifted by 58 cmꢁ1 to ꢂ1539 cmꢁ1, suggesting coordination
via the imine group [33–35]. In metal complexes this vibration is
often coupled to vibrations in the nearby aromatic rings [33,34].
Fig. 3. Plot of vMT ms T for 1 (molar value referred to four Cu atoms). The solid line
corresponds to the best fit (see text).
The increase in m
(N-N) from 1051 to 1084 cmꢁ1 in 1 is due to the in-
crease in the double bond character of N–N, offsetting the loss of
electron density via donation to the metal ion. In addition 1 shows
a broad band around 3400 cmꢁ1 for m(N-H) of the hydrazino moie-
ties and m(O-H) of hydrogen bonded H2O; the m(C-O) stretching also
decreases from 1625 cmꢁ1 (in H-PHMP) to 1661 cmꢁ1 (in 1).
reported for octahedral arrangements around copper [16–18]. Also
the Cu2ꢀ ꢀ ꢀO12N contact (3.265 Å) is much longer than the common
coordination distances [16–18]. Therefore, the coordination
arrangements around the Cu1, Cu2 and Cu4 centers should be re-
garded as predominantly square pyramidal, and consequently the
[Cu4(l-O)2(l
3-OH)2]4+ unit should be considered as an ‘open cub-
3.2.2. Electronic spectra
The UV–Vis spectrum of 1 was recorded at room temperature in
methanol. It depicts six bands at 248, 345, 375, 417, 453 and
650 nm. The low intensity broad band at ca. 650 nm may be attrib-
uted to the d–d transitions of the copper(II) ion [36]. The intense
bands in the 300–453 nm range can be assigned to both the
n ? p⁄ transitions of the imine groups and to LMCT bands
[37,34]. The higher energy bands below 300 nm may be attributed
ane-like core’. In the crystal structure of 1, an additional nine water
molecules of crystallization are present, located at fully occupied
positions in the asymmetric unit.
A space filling packing diagram of 1 is shown in Fig. SI-3, and it
is interesting to note that complex 1 forms a grid-like cage
throughout the whole lattice, thus acquiring stability. To the best
of our knowledge, the title compound represents the first structur-
ally characterized example of a tetranuclear copper(II) complex in
which two hydroxo groups and two N4O-pentadentate ligands
(involving an alkoxo oxygen, two imine nitrogens as central donors
and two terminal pyridine nitrogen atoms) are coordinated to give
to the
complex 1 does not maintain its integrity as [Cu4(PHMP)2(l3
OH)2(NO3)2]2+
p ?
p⁄ ring transitions. As emphasized below, in solution
-
.
a copper(II) complex with a half-open single cubane core, [Cu4(
l-
3.3. Variable temperature magnetic susceptibility studies and DFT
calculations
O)2( -O)(
l
3-OH)2]4+. In 1, the two dinuclear [Cu2( 3-OH)]2+ units
l
l
are puckered to each other, the angle between the two planes
being ꢂ4°. The only known example of a copper(II) complex con-
taining a half-open Cu4O4 cubane-like arrangement is [NEt4]3
The variable temperature magnetic data for crystals of 1 were
recorded between 300 and 2 K, and the plot of MT vs T is shown
in Fig. 3. At 300 K, a
MT value of 1.103 cm3 molꢁ1 K is observed
v
[WOS3(CuI)3(
ton of the anionic cluster is composed of one W, three
Cu and one 2-I atom. So, the title compound is totally different
l
2-I)]ꢀH2O, reported by Shu Shi et al. [18]. The skele-
v
l3-S, three
(for the tetranuclear unit), smaller than that expected for four
uncoupled S = 1/2 ions with g = 2.0 (1.5 cm3 K molꢁ1). On cooling,
l
Model 1: dinuclear [Cu1-Cu2]
O
Cu2
d1
τ
O
Cu1
0º
OH bridges
OR bridges
O
Cu4
α
d2
O
Cu3
Model 2: dinuclear [Cu3-Cu4]
(a)
(b)
(c)
Fig. 4. (a) Exchange coupling constants in 1: J4 (interaction between Cu1–Cu3) and J5 (interaction between Cu2–Cu4) are not represented for clarity. (b) Schematic view of the
two dinuclear models for Cu2O2 used in the B3LYP calculation. (c) Structure of the [Cu3–Cu4] core showing
angles.
s (out of plane hydrogen displacement) and c (hinge distortion)

274
S. Roy et al. / Polyhedron 53 (2013) 269–277
Fig. 6. Spin density distribution for 1 corresponding to the calculated S = 0 ground
Fig. 5. Antiferromagnetic interaction, J,
ms the out-of-plane hydrogen atom
state B3LYP solution. Positive and negative values are represented as white and
dark surfaces, respectively. The isodensity surface represented corresponds to a
displacement,
The open symbols are the JB3LYP values for the dinuclear models. The solid ones
correspond to J values of the tetranuclear compound calculated for = 20°.
s. J1 and J2 are represented with circles and squares, respectively.
value of 0.02 e-bohrꢁ3
.
s
v
MT decreases quickly, indicating a very strong antiferromagnetic
this approach, DFT calculations were carried out to evaluate the
sign and the magnitude order of the six possible exchange path-
ways by using the experimental atomic coordinates of the com-
pound. Based on the well known correlation established by Ruiz
et al. [39] between the magnetic interaction and the out of plane
coupling, reaching a value close to zero at 30 K, indicating diamag-
netic behaviour.
Based on the topology and the structural parameters of the
compound, which is a tetramer built of two consociated dimers,
we count six exchange pathways in the compound, grouped into
six averaged different exchange parameters, J1–J6, corresponding
to the different alkoxo bridges between the copper atoms.
The analysis of the experimental susceptibility data was carried
out using the CLUMAG program [38], resolving the following
Hamiltonian:
hydrogen displacement (
s angle) in dinuclear hydroxo bridged
compounds (Fig. 4 b and c), as the hydrogen atoms of O2A and
O2B were not detected by X-ray diffraction, we anticipate this
may make the determination of the correct ‘J’ values more difficult
as their position can influence the exchange pathways between the
copper atoms.
To overcome these difficulties, two steps of calculation were
done:
Step 1: The two interactions J1 and J2 between Cu1–Cu2 and
Cu3–Cu4 atoms, respectively, were calculated separately using
two dinuclear models (1 and 2) as is shown in Fig. 4b, by varying
H ¼ ꢁJ1ðS1 ꢀ S2Þ ꢁ J2ðS3 ꢀ S4Þ ꢁ J3ðS2 ꢀ S3Þ ꢁ J4ðS1 ꢀ S3Þ ꢁ J5ðS2
ꢀ S4Þ ꢁ J6ðS1 ꢀ S4Þ
ð1Þ
Here the numbering of the spins follows the numbering of the
copper atoms in Fig. 4a.
the
s angle in the range [10–40°]. These calculations showed a lin-
In the fitting attempts of the experimental data we obtained dif-
ferent low J values for J3–J6, with large errors depending of the
starting parameters, while the J1, J2 and ‘g’ parameters always con-
verged to values of ca. ꢁ300 cmꢁ1, ꢁ125 cmꢁ1 and 2.05, respec-
tively. Taking into account the expected relatively low J3–J6
values in contrast with J1 and J2, we can consider that in the com-
pound the magnetic coupling is mainly dominated by those stron-
gest interactions, which ‘reduces’ the system to two dinuclear units
magnetically isolated in the compound. To test this approximation,
the experimental magnetic data were fitted again to the following
simplified Hamiltonian:
ear tendency in the two models with the same slope: the antiferro-
magnetic interaction and absolute values increase when the
s
angle decreases, by approximately 3.2 cmꢁ1 for each degree
(Fig. 5). It seems therefore that the calculated J1 = ꢁ309.4 cmꢁ1
and J2 = ꢁ137.9 cmꢁ1 values agree with the above experimental
fit when the
s angle is equal to 20°.
Step 2: The results of step 1 were used to evaluate the magni-
tude and the sign of the six J1–J6 values between the Cu(II) atoms
in the tetranuclear compound (see computational details). The cal-
culation yielded the following parameters: J1 = ꢁ255.7(1) cmꢁ1, J2 =
ꢁ113.3(1) cmꢁ1 J3 = +0.1(1), J4 = +3.4(1), J5 = +5.0(1) cmꢁ1 and
J6 = ꢁ0.3(1) cmꢁ1
.
H ¼ ꢁJ1ðS1 ꢀ S2Þ ꢁ J2ðS3 ꢀ S4Þ
ð2Þ
It is clearly observed that the interactions corresponding to J1
and J2 are found to be strongly antiferromagnetic through
and 2-OR bridging ligands, with J1 > J2 due to the larger Cu–O–
Cu angle (a few degrees but enough to vary the magnitude of the
l2-OH
The best fit parameters found in this case were: J1 =
ꢁ304.0 cmꢁ1, J2 = ꢁ127.3 cmꢁ1 and g = 2.05. These values are con-
sistent with the values found in the first fit. To check the validity of
l
Table 3
Main geometrical (average) parameters and exchange coupling constants for 1.
JB3LYP/cmꢁ1 dinuclear models
JB3LYPb/cmꢁ1 full structure
Jexp/cmꢁ1
a
Cuij
a
(Cu–O–Cu)/°
d1(Cu–O)/Å
d2(Cuꢀ ꢀ ꢀCu)/Å
Cu12
Cu34
Cu23
Cu13
Cu24
Cu14
97.7/99.6
95.2/96.8
94.9/97.0
91.8/106.1
85.9/110.3
96.7/104.6
1.919/1.946
1.938/1.962
1.929/2.442
1.923/[2.487–2.882]
1.932/[2.397–3.126]
1.950/3.004
2.930
2.898
3.265
3.542
3.562
3.869
ꢁ309.4
ꢁ255.7(1)
ꢁ113.3(1)
+0.1(1)
+3.4(1)
+5.0(1)
ꢁ304.0
ꢁ137.9
ꢁ127.3
–
–
–
–
–
–
–
–
ꢁ0.3(1)
a
Cuij, i and j indicates the copper number considered in the exchange coupling in each cube face.
b
The J values are calculated for
s = 20° and the number in parentheses indicates the standard deviation of the least-squares fitting (see computational details).

S. Roy et al. / Polyhedron 53 (2013) 269–277
275
Fig. 7. First derivative EPR spectra of a polycrystalline sample of 1 at room temperature (a) and at 77 K (b).
due to the presence of neighbouring Cu(II) cations with opposite
spin density. This spin density is probably an artefact due to the
single-determinant wave function considered. The spin population
on the copper atoms is around 0.66 eꢁ and the remaining spin den-
sity, relative to one unpaired electron, appears mainly delocalized
over the oxygen and nitrogen atoms of the ligands.
3.4. EPR spectroscopy
The X-band EPR spectra of a polycrystalline sample of 1 at room
temperature (Fig. 7a) shows an isotropic band centered at g = 2.192
(3109 G for
MS = 1. No clearly detectable zero field splitting or half-filled
signals were observed. The spectrum at 77 K (Fig. 7b) shows a band
centered at g = 2.15 (3164 G for = 9.52(23) GHz) which also corre-
sponds to the transition MS = 1. By simulation of the spectrum
m = 9.52(96) GHz) which corresponds to the transition
D
m
D
Fig. 8. Experimental and stimulated first derivative EPR spectrum of 1 in solution
[21,42], the following gyromagnetic factors were obtained:
g1 = 2.049, g2 = 2.174 and g3 = 2.222. The geometric parameter G,
which is a measure of the exchange interaction between the cop-
per centers in the polycrystalline compound, is calculated using
the equation: for rhombic spectra G = (g3 ꢁ 2.003)/(g\ ꢁ 2.003),
where g\ = (g1 + g2)/2. If G < 4.0, a considerable exchange interac-
tion is indicated in the solid complexes [42]. The G value for com-
pound 1 in the polycrystalline state at 77 K is 2.018. For complexes
of this type, the parameter R can be indicative of the predominance
(methanol) at 77 K.
coupling, see Table 3), and to the greater hinge distortion (Fig. 4c)
in the dinuclear fragment of model 2, which is 11° versus 5° for the
dinuclear fragment of model 1. These two situations confirm the
well-known established assumption that the increase of the Cu–
O–Cu angle and the increment of the hinge distortion in the
[Cu2O2] core reduce the antiferromagnetic behaviour [40].
On the other hand, the difference observed in the J1 and J2 val-
ues between the dinuclear and the tetranuclear models (ꢁ309.4/
137.9 cmꢁ1 and ꢁ255/ꢁ113.3 cmꢁ1) can be understood consider-
ing the conclusion recently reported by Tercero et al. [41] for 2+4
cubane compounds. The structure of this type of compound corre-
sponds to two dinuclear entities with long Cu–O bond distances,
d1, (see Table 3) which causes an increase in the antiferromagnetic
coupling between the dinuclear units. Therefore the antiferromag-
netic interaction is higher in dinuclear entities than in tetranucle-
ars ones.
of the dz2 or dx2
orbital in the ground state, where R = (gz ꢁ gy)/
ꢁy2
(gy ꢁ gx). If R > 1, the greater contribution to the ground state arises
from the dz2 orbital and the structure is closer to a trigonal bipyr-
amid than to a square pyramid. Instead, if R < 1, a greater contribu-
tion to the ground state arises from the dx2
orbital and the
ꢁy2
structure is closer to a square pyramid than to a trigonal bipyramid
[43–45]. In the present case, R = 0.384; this is in agreement with
the X-ray crystal structure, magnetochemistry and the DFT results
of the studied complex.
The EPR spectrum measured for a frozen solution in methanol
at 77 K is depicted in Fig. 8. It is clear that the predominant EPR-ac-
tive species in solution is not the same as that obtained in the solid
state. The complex depicts a major signal from a ‘mononuclear spe-
cies’ of S = 1/2. The following gyromagnetic factors are obtained: g\
ꢂ2.084, g|| ꢂ2.429 [gav ꢂ2.199], A|| ꢂ121 ꢃ 10ꢁ4 cmꢁ1 and A\
ꢂ12 ꢃ 10ꢁ4 cmꢁ1 by simulation of the spectrum [42].
The J3–J6 interactions were found to be very low (J3 and J6 are
anti- or ferromagnetic close to zero, while J4 and J5 are ferromag-
netic). Again these values are in agreement with the previously re-
ported results [41], the J values usually being weak ferromagnetic
or sometimes antiferromagnetic.
The spin distribution is localized mainly in dx2
orbitals at the
ꢁy2
paramagnetic centers, as is observed in the spin density map rep-
resented in Fig. 6, which corresponds to the S = 0 ground state of
the tetranuclear compound. In the bridging oxygen atoms there
are two lobes with spin densities of different sign that appear
A weaker signal is also discernible. It is probably due to a spe-
cies with strong copper(II)–copper(II) interactions. As the powder
spectra are very broad and not resolved, it is not clear whether
the paramagnetic species present in the solid state is the same as

276
S. Roy et al. / Polyhedron 53 (2013) 269–277
in solution or not. For this weaker signal, the separation between
the two Z-components directly gives the value of 2D, which works
out to be 303 G (0.0287 cmꢁ1). The distance between the metal
centres can be approximately evaluated by the use of the following
equation: R = (0.650 gz2/D)1/3 [44]. Following this relationship, the
copper–copper distance in the species in solution under discussion
is estimated as 6.08 Å [45].
The main EPR-active complex detected in methanolic solution
has g|| > g\ > 2.0023, revealing that the Cu(II) centers lie in a dis-
torted octahedral coordination geometry with a predominant
Appendix A. Supplementary data
CCDC 895171 contains the supplementary crystallographic data
for compound 1. These data can be obtained free of charge via
http://www.ccdc.cam.ac.uk/conts/retrieving.html, or from the
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336 033; or e-mail: depos-
it@ccdc.cam.ac.uk. Supplementary data associated with this article
can be found, in the online version, at http://dx.doi.org/10.1016/
j.poly.2013.01.044.
dx2
ground state [46–48]. The representation of (g||, A||) in the
ꢁy2
Peisach–Blumberg plot [49] lies between the N4 and the O4 lines,
this being consistent with a N4O or N3O binding set, suggesting
for the Cu complex present in methanolic solution a N3O set for
the ‘equatorial’ donor atoms. The g||/A|| value is ꢂ200, is consistent
with significant distortion from a planar geometry [50].
References
[1] M.S. Alam, S. Stromsdorfer, V. Dremov, P. Muller, J. Kortus, M. Ruben, J.-M.
Lehn, Angew. Chem., Int. Ed. 44 (2005) 7896.
[2] M.S. Alam, V. Dremov, P. Muller, A.V. Postnikov, S.S. Mal, F. Hussain, U. Kortz,
Inorg. Chem. 45 (2006) 2866.
[3] R.W. Saalfrank, A. Scheurer, I. Bernt, F.W. Heinemann, A.V. Postnikov, V.
Schunemann, A.X. Trautwein, M.S. Alam, H. Rupp, P.J. Muller, Dalton Trans.
(2006) 2865.
[4] M. Ruben, J.-M. Lehn, P. Muller, Chem. Soc. Rev. 35 (2006) 1056. and references
therein.
4. Conclusions
Over the past two decades, PCDA-derived polynuclear, hetero-
nuclear and macrocycle containing metal complexes have been
extensively studied. In this work the self-assembly of two building
blocks of [Cu2(PHMP)], with PHMPꢁ being a polydentate ligand de-
rived from PCDA, led to the half-open cubane-like copper(II) tetra-
[5] M. Ruben, J. Rojo, F.J. Romero-Salguero, L.H. Uppadine, J.-M. Lehn, Angew.
Chem., Int. Ed. 43 (2004) 3644. and references therein.
[6] O. Kahn, Molecular Magnetism, VCH, Weinheim, 1993.
[7] R.E.P. Winpenny, Adv. Inorg. Chem. 52 (2001) 1.
[8] 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.
[9] S. Roy, T.N. Mandal, A.K. Barik, S. Gupta, R.J. Butcher, M.S. El Fallah, J. Tercero,
S.K. Kar, Polyhedron 27 (2008) 105.
[10] E.I. Solomon, U.M. Sundaram, T.E. Machonkin, Chem. Rev. 96 (1996) 2563.
[11] A. Messerschmidt, R. Huber, T. Poulos, K. Wieghardt (Eds.), Handbook of
Metalloproteins, vol. 2, John Wiley & Sons, Chichester, 2001.
[12] R. Neumann, M. Dahan, Nature 388 (1997) 353.
[13] M. Fujita, Y.J. Kwon, S. Washizu, K.J. Ogura, J. Am. Chem. Soc. 116 (1994) 1151.
[14] J.-M. Lehn, Supramolecular Chemistry. Concept and Perspectives, VCH, New
York, 1995. p. 200.
[15] J.-M. Lehn, Science 295 (2002) 2400.
[16] A. Mukherjee, R. Raghunathan, M.K. Saha, M. Nethaji, S. Ramasesha, A.R.
Chakravarty, Chem. Eur. J. 11 (2005) 3087.
[17] J. Tercero, E. Ruiz, S. Alvarez, A. Rodríguez-Fortea, P.J. Alemany, Mater. Chem.
16 (2006) 2729.
[18] H. Hou, B. Liang, X. Xin, K. Yu, P. Ge, W. Ji, S. Shi, Faraday Trans. 13 (1996) 2343.
[19] D.D. Perrin, W.L.F. Armarigo, D.R. Perrin, Purification of Laboratory Chemicals,
second ed., Pergamon Press, Oxford, 1980.
[20] O. Kahn, Molecular Magnetism, Wiley-VCH, New York, 1993.
[21] A. Rockenbauer, L. Korecz, Appl. Magn. Reson. 10 (1996) 29.
[22] R.R. Gagn, C.L. Spiro, T.J. Smith, C.A. Hamann, W.R. Thies, A.K. Shiemke, J. Am.
Chem. Soc. 103 (1981) 4073.
[23] Oxford Diffraction, CRYSALIS CCD and CRYSALIS RED, Versions 1.171.32, Oxford 490
Diffraction Ltd., Abingdon, Oxfordshire, England, 2007, p. 491.
[24] G.M. Sheldrick, SHELXS-97 and SHELXL-9, University of Göttingen, Göttingen,
1997. p. 494.
nuclear complex [Cu4(PHMP)2(
l
3-OH)2(NO3)2](NO3)2ꢀ9H2O (1).
Compound 1 consists of a half-open cubic-like unit which is sub-
stantially distorted; this distortion is due to the geometric prereq-
uisites given by the PCDA-based ligand (H-PHMP). The two
constituting dinuclear [Cu2(PHMP)] units, with a [Cu2(l-O)(l3-
OH)]2+ core, are stabilized by forming a half-open cubane-like
building block with another dinuclear unit bridging through the
l3-hydroxo group supplied from the mother liquor. From the space
filling packing diagram of 1 it is observed that 1 forms a grid-like
cage throughout the whole lattice, thus acquiring stability.
The variable temperature magnetic study revealed that 1 shows
a net antiferromagnetic interaction. Super exchange via the l-oxy-
gen atoms affords an unusual antiferromagnetic interaction which
is attributed to the large Cu–O–Cu bridging angles. The large Cu–
O–Cu angles and the long axial Cu–O distances are responsible
for the strong antiferromagnetic interaction in 1. Besides being
the first example of both an alkoxo and hydroxo bridged tetranu-
clear copper(II) complex with a [Cu4(l-O)2(l
3-OH)2]4+ core, having
an anti-ferromagnetic signature, complex 1 is also a rare example
for the class of molecules having an half-open cubane structure.
The EPR spectroscopic data indicates that in methanolic
solution complex 1 does not maintain its integrity as [Cu4(PHMP)2
[25] E. Ruiz, Struct. Bond. 113 (2004) 71.
[26] J. Cano, R. Costa, S. Alvarez, E. Ruiz, J. Chem. Theory Comput. 3 (2007) 782.
[27] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
J.A. Montgomery, T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar, J.
Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson,
H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, H. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P. Hratchian,
J.B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J.
Austin, R. Cammi, C. Pomelli, J. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P.
Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C.
Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V.
Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-
Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W.
Chen, M.W. Wong, C. Gonzalez, J.A. Pople, GAUSSIAN 03 (Revision B.4), Pittsburgh,
PA, 2003.
(l
3-OH)2(NO3)2]2+; in fact the main EPR signal recorded is charac-
teristic of a mononuclear Cu-species.
Acknowledgments
The authors would like to thank FCT (Fundação para a Ciencia e
Tecnologia, Portugal), the programs FEDER and POCI, the BPD post-
doctoral fellowship SFRH/BPD/74390/2010 and PEst-OE/QUI/
UI0100/2011 for financial support. S. Roy also thanks Prof. S.K.
Kar, University of Calcutta, Kolkata. This research was also sup-
ported by the Spanish MEC (Grants CTQ 2009-07264 and CTQ
2011-23682-C02-01 financed by FEDER funds) and the Generalitat
de Catalunya (Grants 2009SGR1454 and 2009SGR-1459). The
authors gratefully acknowledge the computer resources, technical
expertise and assistance provided by the Centre de Supercompu-
tació de Catalunya, as well as the the Portuguese NMR Network
(IST-UTL Center).
[28] JAGUAR 6.0, Schrodinger Inc., Portland, 2005.
[29] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
[30] A. Schaefer, C. Huber, R. Ahlrichs, J. Chem. Phys. 100 (1994) 5829.
[31] A. Schaefer, H. Horn, R. Ahlrichs, J. Chem. Phys. 97 (1992) 2571.
[32] A.W. Addison, T.N. Rao, J. Reedijk, J. van Rijn, G.C. Verschoor, J. Chem. Soc.,
Dalton Trans. (1984) 1349.
[33] J.C. Pessoa, I. Cavaco, I. Correia, D. Costa, R.T. Henriques, R.D. Gillard, Inorg.
Chim. Acta 305 (2000) 7.
[34] J. Costa Pessoa, M.J. Calhorda, I. Cavaco, I. Correia, M.T. Duarte, V. Felix, R.T.
Henriques, M.F.M. Piedade, I. Tomaz, J. Chem. Soc., Dalton Trans. (2002) 4407.
[35] P. Bindu, M.R.P. Kurup, T.R. Satyakeerthy, Polyhedron 18 (1999) 321.
[36] P.V. Bernhardt, E.G. Moore, M.J. Riley, Inorg. Chem. 40 (2001) 5799.

S. Roy et al. / Polyhedron 53 (2013) 269–277
277
[37] I. Cavaco, J. Costa Pessoa, M.T. Duarte, R.T. Henriques, P.M. Matias, R.D. Gillard,
[43] E. Garribba, G. Micera, J. Chem. Educ. 83 (2006) 1229.
J. Chem. Soc., Dalton Trans. (1996) 1989.
[44] J. Reedijk, D. Knetsch, B. Nieuwenhuijse, Inorg. Chim. Acta 54 (1971) 568.
[38] The series of calculations were made using the computer program CLUMAG
which uses the irreducible tensor operator formalism (ITO): D. Gatteschi, L.
Pardi, Gazz. Chim. Ital. 123 (1993) 231.
[45] R.N. Patel, K.B. Pandeya, Synth. React. Inorg. Met. Org. Chem. 28 (1998) 23.
[46] C. Kallay, K. Varnagy, G. Malandrinos, N. Hadjiliadis, D. Sanna, I. Sovago, Dalton
Trans. (2006) 4545.
[39] E. Ruiz, P. Alemany, S. Alvarez, J. Cano, J. Am. Chem. Soc. 119 (1997) 1297.
[40] E. Ruiz, P. Alemany, S. Alvarez, J. Cano, Inorg. Chem. 36 (1997) 3683.
[41] J. Tercero, E. Ruiz, S. Alvarez, A. Rodríguez-Fortea, P. Alemany, J. Mater. Chem.
16 (2006) 2729. and references therein.
[47] B.J. Hathaway, D.E. Billing, Coord. Chem. Rev. 5 (1970) 143.
[48] B.J. Hathaway, Struct. Bond. 57 (1984) 55.
[49] J. Peisach, W.E. Blumberg, Arch. Biochem. Biophys. 165 (1974) 691.
[50] E. Garribba, G. Micera, D. Sanna, L. Strinna-Erre, Inorg. Chim. Acta 299 (2000)
253.
[42] E.B. Seena, M.R.P. Kurup, Polyhedron 26 (2007) 829.