Mononuclear Cu and dinuclear Cu-Ln complexes of benzimidazole based ligands including N and O donors: Syntheses, characterization, X-ray molecular structures and magnetic properties

Article abstract of DOI:10.1016/j.poly.2010.04.010

The synthesis of two mononuclear precursor copper complexes, [(HL ²)₂Cu], 1, and [(HL³)₂Cu] ·H₂O,2, and three dinuclear Cu-Ln complexes, [(HL ¹)₂Cu(CH₃CN)₂G

Full text of DOI:10.1016/j.poly.2010.04.010

Polyhedron 29 (2010) 2111–2119
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Mononuclear Cu and dinuclear Cu–Ln complexes of benzimidazole based
ligands including N and O donors: Syntheses, characterization, X-ray
molecular structures and magnetic properties
Fatima Zohra Chiboub Fellah a,b,c, Jean-Pierre Costes a,b, , Carine Duhayon a,b, Jean-Claude Daran a,b
,
*
Jean-Pierre Tuchagues a,b,
**
a CNRS, LCC (Laboratoire de Chimie de Coordination), 205, Route de Narbonne, F-31077 Toulouse, France
b Université de Toulouse, UPS, INPT, LCC, F-31077 Toulouse, France
c Université Aboubaker Belkaid, Faculté des Sciences, Département de Chimie, BP 119, 13000 Tlemcen, Algeria
a r t i c l e i n f o
a b s t r a c t
The synthesis of two mononuclear precursor copper complexes, [(HL2)2Cu], 1, and [(HL3)2Cu]ꢀH2O, 2, and
three dinuclear Cu–Ln complexes, [(HL1)2Cu(CH3CN)2Gd(NO3)3], 3, [(HL3)2CuGd(NO3)3]ꢀ2(H2O), 4, and
[(HL3)2CuTb(NO3)3]ꢀ2(H2O), 5, based on the ligands H2L1 (4-bromo-2-[1-(5-bromo-2-hydroxy-3-
methoxybenzyl)-1H-benzimidazol-2-yl]-6-methoxyphenol), H2L2 (2-(1H-benzimidazol-2-yl)-4-bromo-
6-methoxyphenol) and H2L3 (2-(1H-benzimidazol-2-yl)-6-methoxyphenol) are described in this
contribution. The X-ray crystal structures of H2L2, 1, 3, 4, and 5 have been solved. The novel ligand
H2L2 crystallizes with two independent molecules in the asymmetric unit; several intermolecular hydro-
gen contacts connect alternate independent H2L2 molecules into chains developing along c. In complex 1,
two (HL2)ꢁ ligands chelate the copper ion through their imidazolyl nitrogen and phenoxo oxygen atoms,
in a relative head to tail arrangement. The molecular structure of 3 is similar to those of the previously
reported Cu–Ln complexes of H2L1. In the isostructural complexes 4 and 5, two HL3 ligands sandwich one
Cu2+ ion through their N,O sites and one Ln3+ ion through their O2 site, implying a relative head to head
arrangement, at variance with the relative head to tail arrangement of HL2 in the mononuclear copper
precursor 1. The magnetic properties of 1, 3, 4, and 5 have been investigated. Extended intermolecular
antiferromagnetic interactions operate in complex 1 ((JChain = ꢁ0.8(1) cmꢁ1). Ferromagnetic interactions
between Gd (S = 7/2) and Cu (S = 1/2) centers operate in complexes 3 and 4, leading to an S = 4 ground
state (JCuGd = 7.2(2) cmꢁ1 for 3 and JCuGd = 6.5(2) cmꢁ1 for 4). Depopulation of the Tb Stark levels, preclude
obtaining reliable information on the presence and sign of the Cu–Tb interaction in 5. These new com-
plexes are complementary to those previously reported: the Cu–O2–Gd core is planar while deformations
are borne by the ligands at variance with previous examples where the constraints were located at the
Cu–O2–Gd core. The presence of two independent ligands in the Cu,Gd coordination spheres confers a
degree of freedom greater than that allowed by a unique tetradentate ligand. As a result, the strength
of the magnetic interaction is not solely related to the dihedral angle between the CuOO and GdOO planes
in the central core.
Article history:
Received 10 March 2010
Accepted 8 April 2010
Available online 24 April 2010
Keywords:
Benzimidazole based ligands
Mononuclear copper complexes
Copper–lanthanide complexes
X-ray molecular structures
Magnetic properties
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
zero-dimensional (0D) or extended one-dimensional to three-
dimensional (1–3D) high-spin materials. The material obtained
A large majority of strictly dinuclear Cu–Gd complexes exhibit a
ferromagnetic behavior [1–3]. It is then of interest to link such fer-
romagnetic Cu–Gd units with the aim of achieving either finite
may be a single molecule magnet (SMM) [4–12], in the former
case, or a single chain magnet (SCM) in the case of a 1D material
[13], or a genuine ferromagnet in the latter cases [14,15]. To this
end, the dinucleating ligand must include an additional free coor-
dination site located in such a way that each copper ion is not only
intra-molecularly, but also inter-molecularly, bridged to Gd ions.
This strategy has allowed synthesizing tetranuclear [Cu–Gd]2 spe-
cies by using unsymmetrical ligands including an amide function
[4,6,7], and an infinite chain of such tetranuclear components
{[Cu–Gd]2}n by using ligands including amide, alcohol and phenol
* Corresponding author at. CNRS, LCC (Laboratoire de Chimie de Coordination),
205, Route de Narbonne, F-31077 Toulouse, France.
** Corresponding author at. CNRS, LCC (Laboratoire de Chimie de Coordination),
205, Route de Narbonne, F-31077 Toulouse, France. Tel.: +3 561 333154; fax: +33
561 553003.
E-mail addresses: jean-pierre.costes@lcc-toulouse.fr, jean-pierre.tuchagues@
lcc-toulouse.fr (J.-P. Tuchagues).
0277-5387/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2010.04.010

2112
F.Z. Chiboub Fellah et al. / Polyhedron 29 (2010) 2111–2119
functions [13]. In both types of materials, all Cu–Gd interactions,
including those mediated by the amido bridges, are ferromagnetic,
thus achieving the goal pursued. However, the Cu–Gd ferromag-
netic interactions mediated by the amido bridges are weaker than
those involving phenoxo bridges. In our quest for dinucleating li-
gands promoting both intra-molecular and inter-molecular Cu–
Gd bridges, we now consider the potential of (benz)imidazole
function(s) for bridging metal centers upon deprotonation
[16,17]. In this contribution, we describe the synthesis of two novel
ligands, H2L2 and H2L3, two mononuclear precursor copper com-
plexes, [(HL2)2Cu], 1, and [(HL3)2Cu]ꢀH2O, 2, and three dinuclear
Cu–Ln complexes, [(HL1)2Cu(CH3CN)2Gd(NO3)3], 3, [(HL3)2CuGd
(NO3)3]ꢀ2(H2O), 4, and [(HL3)2CuTb(NO3)3]ꢀ2(H2O), 5, based on
the H2L1, H2L2 and H2L3 ligands depicted in Scheme 1. The X-ray
crystal and molecular structures of H2L2, 1, 3, 4, and 5 are reported,
as well as the magnetic properties of 1, 3, 4, and 5.
9.36 (1 H, s, OH, H19); 10.41 (1 H, s, OH, H11). 13C NMR: 43.2,
C17; 56.7, C24; 56.8, C16; 110.2, C22; 110.3, C14; 111.3, C8;
114.5, C21; 116.6, C13; 119.1, C10; 119.5, C5; 121.9, C23; 122.6,
C7; 123.2, C6; 124.7, C15; 125.3, C18; 135.6, C9; 142.6, C4;
143.7, C19; 145.5, C11; 148.9, C20; 149.6, C12; 151.0, C2.
3.2.2. H2L2
2-(1H-benzimidazol-2-yl)-4-bromo-6-methoxyphenol. Follow-
ing filtration of the H2L1 precipitate, the filtrate was refluxed for
8 h and left to cool down overnight. The resulting off-white micro-
crystalline precipitate was filtered off, washed with ethanol and
diethyl ether. Yield: 1.4 g (44%). Anal. Calc. for C14H11BrN2O2
(319.16): C, 52.7; H, 3.5; N, 8.8. Found: C, 52.6; H, 3.3; N, 8.6%.
Characteristic IR absorptions (KBr, cmꢁ1): 3361, 1491, 1480,
1463, 1443, 1399, 1385, 1248, 1050, 825, 734, 708, 601. Mass spec-
trum: (EI): m/z = 318 (1 0 0), [H2L2]+.. 1H NMR, dmso-d6: 3.84 (3 H,
s, OCH3, H16); 7.20 (1 H, d, J = 2.0 Hz, CH, H15); 7.34 (2 H, m, CH,
H6, H7); 7.35 (1 H, m, CH, H8); 7.70 (1 H, m, CH, H5); 7.86 (1 H, d,
J = 2.0 Hz, CH, H13); 13.29 (1 H, s, OH, H11); 13.41 (1 H, s, NH, H1).
13C NMR: 56.3, C16; 109.9, C14; 111.8, C8; 113.8, C5; 116.5, C13;
118.3, C10; 119.7, C15; 122.8, C7; 123.7, C6; 137.9, C4, C9; 147.8,
C11; 149.9, C12; 150.7, C2. Recrystallization from chloroform
yielded off-white X-ray quality crystals.
2. Experimental section
3.1. Materials
All chemicals and solvents, obtained from Aldrich, were of re-
agent grade and were used for the syntheses without further
purification.
3.2.3. H2L3
2-(1H-benzimidazol-2-yl)-6-methoxyphenol.
A
solution of
3.2. Ligands
o-phenylenediamine (1.08 g, 0.01 mol) and o-vanillin (3.04 g,
0.02 mol) in ethanol (80 mL) was refluxed for 20 h and left to cool
down overnight. The resulting off-white microcrystalline precipi-
tate was filtered off, washed with ethanol and diethyl ether. Yield:
1.25 g (52%). Anal. Calc. C14H12N2O2 (240.26): C, 70.0; H, 5.0; N,
11.7. Found: C, 69.6; H, 4.8; N, 11.6%. Characteristic IR absorptions
(KBr, cmꢁ1): 3331, 1495, 1475, 1462, 1449, 1421, 1388, 1253, 1236,
1058, 831, 787, 739, 715, 602. Mass spectrum: (EI): m/z = 318
(1 0 0), [H2L3]+.. 1H NMR, dmso-d6: 3.84 (3 H, s, OCH3, H16); 6.96
(1 H, t, J = 2.0 Hz, CH, H14); 7.10 (1 H, d, J = 2.0 Hz, CH, H13);
7.30 (2 H, m, CH, H6, H7); 7.65 (2 H, m, CH, H5, H8); 7.64 (1 H,
d, J = 2.0 Hz, CH, H15); 13.29 (1 H, s, OH, H11); 13.41 (1 H, s, NH,
H1). 13C NMR: 56.2, C16; 111.8, C5, C8; 113.0 C10; 114.4, C13;
118.0, C14; 119.5, C15; 123.3, C6, C7; 142.1, C4, C9; 148.8, C11;
149.0, C12; 152.4, C2.
3.2.1. H2L1
4-Bromo-2-[1-(5-bromo-2-hydroxy-3-methoxybenzyl)-1H-ben-
zimidazol-2-yl]-6-methoxyphenol, was prepared according to the
reported experimental procedure [18]. The recrystallized material
was obtained as colorless crystals. Yield 1.6 g (29%). Characteristic
IR absorptions (KBr, cmꢁ1): 3401, 2934, 1609, 1487, 1277, 1258,
1074, 758, 747. 1H NMR, dmso-d6: 3.79 (3 H, s, OCH3, H24); 3.89
(3 H, s, OCH3, H16); 5.33 (2 H, s, CH2, H17); 6.21 (1 H, d,
J = 2.0 Hz, CH, H23); 7.00 (1 H, d, J = 2.0 Hz, CH, H21); 7.09 (1 H,
d, J = 2.1 Hz, CH, H15); 7.25 (2 H, m, CH, H6, H7); 7.27 (1 H, d,
J = 2.1 Hz, CH, H13); 7.45 (1 H, m, CH, H8); 7.72 (1 H, m, CH, H5);
3.3. Complexes
3.3.1. [(HL2)2Cu], 1
A mixture of H2L2 (0.32 g, 0.001 mol) and Cu(CH3COO)2ꢀH2O
(0.10 g, 0.0005 mol) in methanol (10 mL) was stirred for 1 h, and
the magenta precipitate was filtered off and washed with metha-
nol and diethyl ether. Yield: 0.32 g (91%). Anal. Calc. C28H20Br2Cu-
N4O4 (699.84): C, 48.1; H, 2.9; N, 8.0. Found: C, 48.4; H, 2.7; N,
8.1%. Characteristic IR absorptions (KBr, cmꢁ1): 3236, 1537, 1489,
1471, 1459, 1384, 1254, 1229, 1196, 1183, 1069, 841, 799, 747,
738, 711, 625. Slow evaporation of a DMF solution of the precipi-
tate yielded brown crystals of X-ray quality.
3.3.2. [(HL3)2Cu]ꢀH2O, 2
This complex, prepared in a way similar to that used for 1, was
obtained as a maroon precipitate. Yield: 0.29 g (91%). Anal. Calc.
C28H22CuN4O4 (560.07): C, 60.0; H, 4.3; N, 10.0. Found: C, 59.5;
H, 4.1; N, 9.7%. Characteristic IR absorptions (KBr, cmꢁ1): 3236,
1536, 1489, 1473, 1451, 1436, 1260, 1236, 1200, 1067, 1033,
853, 736, 731, 723.
3.3.3. [(HL1)2Cu(CH3CN)2Gd(NO3)3], 3
This dinuclear complex was prepared according to the experi-
mental procedure reported for the parent Cu–Eu compound [18].
Scheme 1. Experimental route to the H2L1, H2L2 and H2L3 ligands.

F.Z. Chiboub Fellah et al. / Polyhedron 29 (2010) 2111–2119
2113
Yield: 0.098 g (58%). Anal. Calc. C48H40Br4CuGdN9O17 (1555.31): C,
3.5. Crystallographic data collection and structure determination for
37.1; H, 2.6; N, 8.1. Found: C, 36.8; H, 2.6; N, 8.1%. Characteristic IR
absorptions (KBr, cmꢁ1): 3360, 1478, 1384, 1279, 1237, 812, 743.
Slow diffusion of diethyl ether in an acetonitrile solution of the
powder complex yielded X-ray quality green crystals of
3ꢀ(H2O)ꢀ(C2H5OH)ꢀ(C4H10O).
H2L2, [(HL2)2Cu], 1, [(HL1)2Cu(CH3CN)2Gd(NO3)3], 3, [(HL3)2CuGd
(NO3)3], 4, and [(HL3)2CuTb(NO3)3], 5
Crystals of H2L2, 1, 3, 4, and 5 were kept in the mother liquor
until they were dipped into oil. The selected crystals, sticked on
a Mitegen micromount and quickly cooled down to 180 K, were
mounted on a Stoe Imaging Plate Diffractometer System (IPDS)
(H2L2, off-white, 0.40 ꢂ 0.20 ꢂ 0.05 mm3; 1, brown, 0.15 ꢂ 0.1 ꢂ
0.1 mm3; 4, reddish-green, 0.25 ꢂ 0.25 ꢂ 0.2 mm3; 5, reddish-
green, 0.3 ꢂ 0.25 ꢂ 0.125 mm3) or an Oxford-Diffraction XCALIBUR
(3, green, 0.22 ꢂ 0.20 ꢂ 0.18 mm3) using a graphite monochroma-
tor (k = 0.71073 Å) and equipped with an Oxford Cryosystems
cooler device, and a Cryojet cooler device from Oxford Instru-
ments, respectively. The data were collected at 180 K. The unit cell
determination and data integration were carried out using the
Xred [22] and CrysAlis RED [23] packages for the data recorded
on the IPDS and Xcalibur diffractometers, respectively. 9776
reflections were collected for H2L2, of which 4504 were indepen-
dent (Rint = 0.071), 15040 reflections for 1, of which 5549 were
independent (Rint = 0.3036), 68765 reflections for 3, of which
20213 were independent (Rint = 0.057), 21685 reflections for 4,
of which 7952 were independent (Rint = 0.1088) and 21687 reflec-
tions for 5, of which 8020 were independent (Rint = 0.0767).
Absorption corrections were applied using Multiscan for H2L2
and complexes 1 and 3 [24]. The structures have been solved by
Direct Methods using SIR92 [25a] and SIR97 [25b], and refined
by least-squares procedures on F2 using the program SHELXL97
[26] included in the software package WinGX version 1.63 [27]
for 1, 4, and 5, and with the program CRYSTALS [28] for H2L2 and
3. The Atomic Scattering Factors were taken from International ta-
bles for X-ray crystallography [29]. Hydrogens atoms were intro-
duced in idealized positions and refined by using a riding model.
Non-hydrogens atoms were anisotropically refined, and in the last
cycles of refinement a weighting scheme was used. Although sev-
eral data sets were recorded on selected crystals of 1, we could get
3.3.4. [(HL3)2CuGd(NO3)3]ꢀ2(H2O), 4ꢀ2(H2O)
A
mixture of H2L3CuꢀH2O (0.15 g, 2.7 ꢂ 10ꢁ4 mol) and
Gd(NO3)3ꢀ5H2O (0.13 g, 2.7 ꢂ 10ꢁ4 mol) in acetone (10 mL) was
stirred for thirty minutes, yielding a bronze precipitate that was fil-
tered off, washed with acetone and diethyl ether and dried. Yield:
0.15 g (62%). Anal. Calc. C28H26CuGdN7O15 (921.35): C, 36.5; H, 2.8;
N, 10.6. Found: C, 36.1; H, 2.7; N, 10.1%. Characteristic IR absorp-
tions (KBr): 3217, 1493, 1481, 1452, 1440, 1286, 1247, 1203,
1053, 995, 760, 748, 739 cmꢁ1. Slow diffusion of diethyl ether in
a 1:1 acetonitrile/ethanol solution of the powder complex yielded
X-ray quality reddish-green crystals of 4ꢀsolvent.
3.3.5. [(HL3)2CuTb(NO3)3]ꢀ2(H2O), 5ꢀ2(H2O)
Yield: 0.14 g (56%). Anal. Calc. C28H26CuTbN7O15 (923.03): C,
36.4; H, 2.8; N, 10.6. Found: C, 36.1; H, 2.7; N, 10.1%. Characteristic
IR absorptions (KBr): 3206, 1492, 1481, 1451, 1439, 1285, 1246,
1201, 1053, 995, 849, 759, 732 cmꢁ1. Slow diffusion of diethyl
ether in a 1:1 acetonitrile/ethanol solution of the powder complex
yielded X-ray quality reddish-green crystals of 5ꢀsolvent.
Alternately, compounds 4 and 5 may be prepared through one
pot reactions: in the first step, H2L3 and Cu(CH3COO)2.H2O are dis-
solved in a 1:1 acetonitrile/ethanol mixture under stirring and re-
fluxed for 30 min. In the second step, the appropriate lanthanide
nitrate, dissolved in the minimum amount of ethanol, is poured
into the reaction mixture, which is further refluxed for 2 h. Upon
cooling down, compound 4 (5) precipitates.
only 1058 independent reflections with I > 2r(I) over the 5549
independent ones; therefore, among non-hydrogen atoms, only
the copper center and the nitrogen and oxygen donor atoms were
anisotropically refined in order to get reliable information about
the coordination sphere in spite of the poor refinement. Although
better than those recorded for 1, the relatively low quality of all
data sets recorded on selected crystals of 4 and 5 precluded
obtaining good enough refinements. In both crystals, the unit cell
contains a certain amount of water molecules. However, these
water molecules appear to be highly disordered and it was diffi-
cult to model their positions and distribution reliably. Therefore,
the SQUEEZE function of PLATON [30] was used to eliminate the contri-
bution of the electron density in the solvent region from the inten-
sity data, and the solvent-free model was employed from the final
refinement. Due to the omission of the water molecules from the
model, it was not possible to analyse the hydrogen-bonding inter-
actions. There is one cavity of 820 A3 per unit cell of 5. PLATON esti-
mated that the cavity contains 41 electrons which may
correspond to roughly four water molecules within the cell. This
is contradictory with the size of the cavity in which we could ex-
pect to introduce roughly 41 water molecules. It is well known
that the SQUEEZE procedure is very dependent on the low-angle
reflections and that the electron count may be underestimated if
those reflections are missing which could be the case when data
are collected on a CCD type machine. The data for complex 4 were
of very poor quality resulting in rather high R and wR2 values and
large residual electron density. Crystal data collection and refine-
ment parameters are summarized in Table 1. Drawings of mole-
cules are performed with the program DIAMOND [31] and 30%
probability displacement ellipsoids for non-hydrogen atoms.
3.4. Physical measurements
C, H, and N elemental analyses were carried out on a Perkin–
Elmer 2400 series II device at the microanalytical Laboratory of
the Laboratoire de Chimie de Coordination in Toulouse, France.
Infrared spectra were recorded at room temperature using a
9800 FTIR spectrometer (Perkin–Elmer) with samples as KBr pel-
lets (%T). 1H and 13C NMR spectra were recorded at ambient tem-
perature (295 K) with a Bruker WM250 spectrometer. 2D 1H
COSY experiments using standard programs and 2D pulse-field
gradient HMQC 1H–13C correlation using the PFG-HMQC standard
program were performed on a Bruker AMX400 spectrometer. All
chemical shifts (1H and 13C) are given in ppm versus TMS using
dmso d6 as solvent. Mass spectra (EI) were recorded on a Nermag
R10-10 spectrometer using acetone as solvent. Magnetic suscepti-
bilities were measured in the 2–300 K temperature range, at a
sweeping rate of 2 K minꢁ1 under an applied magnetic field of
0.1 T, and isothermal magnetization measurements were per-
formed up to 5 T at 2 K, using an MPMS5 Quantum Design SQUID
susceptometer. The apparatus was calibrated with palladium me-
tal. All samples were 3 mm diameter pellets molded from ground
crystalline samples. Diamagnetic corrections were applied by using
Pascal’s constants [19]. The magnetic susceptibilities have been
computed by exact calculations of the energy levels associated to
the spin Hamiltonian through diagonalization of the full matrix
with a general program for axial symmetry [20]. Least-squares fit-
tings were accomplished with an adapted version of the function-
minimization program MINUIT [21].

2114
F.Z. Chiboub Fellah et al. / Polyhedron 29 (2010) 2111–2119
Table 1
Crystallographic data for H2L2 and complexes 1, 3, 4 and 5.
H2L2
1ꢀdmf
3ꢀH2OꢀEtOHꢀ(Et)2O
4
5
Formula
Fw
C14H11BrN2O2
319.16
C31H27Br2CuN5O5
772.94
C54H56Br4CuGdN9O20
1691.5
C28H20CuGd N7O13
883.30
C28H22CuTbN7O13
886.99
Temperature (K)
k (Å)
180(2)
180(2)
0.71073
180(2)
0.71073
monoclinic
P21/c
16.267(5)
180(2)
0.71073
180(2)
0.71073
0.71073
triclinic
Crystal system
Space group
triclinic
triclinic
triclinic
ꢀ
ꢀ
ꢀ
ꢀ
P1
P1
P1
P1
0
7.2960(15)
11.990(2)
14.599(3)
11.150(5)
11.211(5)
13.914(5)
11.475(5)
12.987(5)
15.332(5)
11.463(5)
13.032(5)
15.365(5)
a (AÅ)
0
21.911(8)
21.231(8)
b (AÅ)
0
c (AÅ)
a
b (°)
(°)
75.97(3)
89.81(2)
86.92(3)
1237.2(5)
4
113.526(5)
104.907(5)
94.783(5)
1506.7(11)
2
90
92.454(3)
90
7560(5)
4
1.486
98.744(5)
102.940(5)
93.618(5)
2189.7(15)
2
99.016(5)
102.931(5)
93.547(5)
2198.2(15)
2
c
(°)
V (Åꢁ3
)
Z
qcalc (g cmꢁ3
)
1.711
3.321
1.704
3.427
1.340
2.044
1.340
2.136
l
(Mo K
a
) (mmꢁ1
)
3.331
F(0 0 0)
638
774
3332
868
874
Theta range for data collection (°)
Goodness-of-fit
1.98–25.98
1.046 (F)
0.0812, 0.0940
0.1148, 0.1195
2.49–26.16
0.646 (F2)
0.0653, 0.1195
0.3058, 0.2073
2.67–29.08
1.129 (F2)
0.0875*, 0.0993*
0.1720, 0.1197
2.02–25.95
1.059 (F2)
0.0867, 0.2504
0.1278, 0.2635
2.28–26.03
0.848 (F2)
0.0469, 0.1072
0.0779, 0.1209
R1, wR2 [I > 2r(I)]
R1, wR2 (all data)
aR =
[I > 3
R
||Fo| ꢁ |Fc||/
R
|Fo|. bwR2 = [
R
w(|F2o | ꢁ |F2c |)2/
R .
w|F2o |2]1/2
*
r
(I)].
Crystallographic data have been deposited with the Cambridge
Crystallographic Data Centre, data see Appendix 1.
in the legend to the figure. In both molecules, the fused benzene
and imidazole rings are nearly planar. The dihedral angles between
the benzimidazole moiety and the substituted aryl ring are nearly
equal for the two independent molecules, 4.9(3)° and 4.4(3)°,
respectively. The N1C7C8C9 (N3C21C22C23) and N2C7C8C13
(N4C21C22C27) torsion angles are almost similar, 5.4(2)°
(4.8(2)°) and 3.6(2)° (3.0(2)°), respectively: the molecule is slightly
twisted about the C7–C8 (C21–C22) bonds.
4. Results
4.1. Syntheses
The reaction of o-phenylenediamine with 5-bromo-2-hydroxy-
3-methoxybenzaldehyde in ethanol has been previously de-
scribed [18]. Following the same experimental procedure, we
successively obtained the Schiff base and the N-substituted benz-
imidazole ligand H2L1. Once this ligand was isolated by filtration,
the resulting filtrate was refluxed for 8 h, yielding a new precip-
itate. This off-white solid was identified as the N–H benzimid-
Although the phenolic hydrogen atom H(O1) could not be lo-
cated, intra-molecular NꢀꢀꢀH–O hydrogen contacts involving the
0
imine and phenol functions (N2ꢀꢀꢀO1 = 2.567(9) ÅA, N4ꢀꢀꢀO3 =
0
2.555(12) ÅA) operate inside both independent molecules: together
with the intermolecular contacts described below, they may be
responsible for the reduced twist angles reported above. Several
weak intermolecular hydrogen contacts connect alternate inde-
pendent H2L2 molecules into chains developing along c (N1ꢀꢀꢀO3,
N1ꢀꢀꢀO4, C9ꢀꢀꢀO4, N3ꢀꢀꢀO10, N3ꢀꢀꢀO20 and C23ꢀꢀꢀO20, Fig. 1). Addi-
tional weak hydrogen contacts and van der Waals interactions
interconnect these chains into the 3D crystal structure of H2L2.
azole ligand H2L2 depicted in Scheme 1.
(COSY, HMQC) of H2L1 and H2L2 allowed
A
a
2D NMR study
straightforward
assignment of the entire set of 1H and 13C signals, thanks to long
range CꢀꢀꢀH correlations. A similar reaction involving o-vanillin
and o-phenylenediamine yielded the non substituted benzimid-
azole ligand exclusively.
4.3. X-ray single crystal structure of [(HL2)2Cu]ꢀdmf, 1ꢀdmf
The monomeric copper complexes [(HL2)2Cu], 1, and
[(HL3)2Cu]ꢀH2O, 2, were obtained by reacting H2L2 and H2L3 with
copper acetate in methanol, two ligands chelating the copper ion
in a head to tail arrangement, as shown by the structural determi-
nation of 1. Cu–Ln complexes (Ln = Gd, Tb) have been obtained
from H2L1 and H2L3, the H2L2 ligand yielding only the copper com-
plex 2. Complex 3, [(HL1)2Cu(CH3CN)2Gd(NO3)3], is analogous to
the previously reported Cu–Ln complexes of H2L1 (Ln = Eu, Tb, Er,
Yb) [18]. Complexes 4, [(HL3)2CuGd(NO3)3], and 5, [(HL3)2CuTb
(NO3)3], were obtained either through reaction of the mononuclear
copper precursor 1 with the appropriate Ln nitrate, or through one
pot reactions involving H2L3, copper acetate and the appropriate Ln
nitrate.
Although, for the reasons explained in the experimental section,
the refinement achieved for 1 was average, it is of good enough
quality to reliably evidence the molecular composition and main
geometrical features of the [(HL2)2Cu] complex molecule of 1. This
is important in order to evaluate the changes in the relative
arrangement of the HL2 ligands around the copper center on going
from the mononuclear precursor 1 to the dinuclear Cu–Ln com-
plexes 4 and 5. The structure of 1 is shown in Fig. 2 and selected
geometric parameters are collated in the legend to the figure.
Two HL2 ligands chelate the copper ion through their imidazolyl
nitrogen and phenoxo oxygen atoms, in a relative arrangement
close to head to tail. The copper cation adopts a distorted four-
coordinate environment. The dihedral angle of 43.1(5)° between
the two coordinating ligands (as defined by the Cu–N–O planes)
denotes a substantial distortion from a square planar geometry
(0°) towards a tetrahedral geometry (90°). This distortion is con-
firmed by the values of the angles N1CuN3 and O1CuO3 which
deviate largely from 180°: 153.6(5) and 145.2(5)°, respectively.
4.2. X-ray single crystal structure of H2L2
The ligand H2L2 crystallizes with two independent molecules in
the asymmetric unit; they are shown together with an additional
molecule in Fig. 1, and selected geometric parameters are collated

F.Z. Chiboub Fellah et al. / Polyhedron 29 (2010) 2111–2119
2115
Fig. 1. DIAMOND plot of three molecules of H2L2. Thermal ellipsoids are drawn at the 30% probability level. Except those involved in donorꢀꢀꢀacceptor contacts, H atoms have
been omitted for clarity. Selected bond lengths (Å) and angles (°): O1–C13 1.358(10), O2–C12 1.368(10), Br1–C10 1.890(9), C7–C8 1.458(11), N1–C1 1.379(11), N1–C7
1.375(10), N2–C6 1.393(10), N2–C7 1.340(11), O3–C27 1.359(10), O4–C26 1.373(10), Br2–C24 1.889(9), C21–C22 1.458(11), N3–C15 1.380(11), N3–C21 1.372(10), N4–C20
1.405(11), N4–C21 1.317(10); C1–N1–C7 106.7(7), C6–N2–C7 105.9(7), C7–C8–C13 118.4(7), C12–O2–C14 116.8(7), C15–N3–C21 107.1(7), C20–N4–C21 105.8(7), C21–C22–
C27 119.1(7), C26–O4–C28 116.3(6); hydrogen contacts N1ꢀꢀꢀO3 3.243, H1ꢀꢀꢀO3 2.61, N1–H1ꢀꢀꢀO3 132; N1ꢀꢀꢀO4 3.290(12), H1ꢀꢀꢀO4 2.46, N1–H1ꢀꢀꢀO4 162; C9ꢀꢀꢀO4 3.544, H6ꢀꢀꢀO4
2.69, C9–H6ꢀꢀꢀO4 153; N3ꢀꢀꢀO10 3.206(12), H31ꢀꢀꢀO10 2.53, N3–H31ꢀꢀꢀO10 136; N3ꢀꢀꢀO20 3.456, H31ꢀꢀꢀO20 2.64, N3–H31ꢀꢀꢀO20 159; C23ꢀꢀꢀO20 3.498, H36ꢀꢀꢀO20 2.63, C23–H36ꢀꢀꢀO20
0
00
156; O3ꢀꢀꢀN400 2.555(12), H38ꢀꢀꢀN400 1.84, O3–H38ꢀꢀꢀN4 142. Symmetry operations: = x, y, z ꢁ 1; = x, y, 1 + z.
4.4. X-ray single crystal structure of [(HL1)2Cu(CH3CN)2Gd(NO3)3], 3
The structure of the asymmetric unit of 3, together with atom
labeling scheme is displayed in Fig. 4 and selected geometric
parameters are collated in the legend to the figure. The crystal
belongs to the monoclinic system, P21/c space group. The molec-
ular structure of 3 is similar to those of the previously reported
Cu–Ln complexes of H2L1 (Ln = Eu, Tb, Er, Yb) [18]. The copper
ion is in a pseudo-octahedral coordination environment including
two imidazolyl nitrogen and two phenoxo oxygen atoms pro-
vided by two HL1 ligands and two apical nitrogen atoms from
acetonitrile molecules. The phenoxo oxygen donors bridge the
0
metal ions with a CuꢀꢀꢀGd separation of 3.460(3) AÅ. The equatorial
N2O2 donor set deviates from planarity: the dihedral angle be-
tween the CuN2O3 and CuN4O7 intra-ligand planes is 19.9(4)°.
0
The copper cation is very close (0.02 ÅA) to the mean N2O2 plane.
The gadolinium cation is surrounded by ten oxygen donors, six of
them being provided by three bidentate nitrate anions and the
remaining four by the {(HL1)2Cu} constituting unit. The dihedral
angle between the CuO3O7 and GdO3O7 planes is minute:
0.4(4)°. The overall structure has a C2 axis running through
O(14)–Gd–Cu. Two among the four Br substituents are disor-
dered: they were satisfactorily modeled over two sites with occu-
pancy ratios of 0.5.
Fig. 2. DIAMOND plot of [(HL2)2Cu], 1. Thermal ellipsoids are drawn at the 30%
probability level. H atoms have been omitted for clarity. Selected bond lengths (Å)
and angles (°): Cu–N1 1.913(14), Cu–O1 1.901(11), Cu–N3 1.899(13), Cu–O3
1.941(13), N1–Cu–O1 92.3(5), O1–Cu–N3 95.6(5), N3–Cu–O3 93.1(5), O3–Cu–N1
94.7(5), N1–Cu–N3 153.6(5), O1–Cu–O3 145.2(5).
4.5. X-ray single crystal structures of [(HL3)2CuGd(NO3)3], 4, and
[(HL3)2CuTb(NO3)3], 5
The angle of 65.7(3)° between the (HL2)Cu planes indicates that
they are almost perpendicular to each other. The dihedral angles
between the benzimidazole moieties and the substituted aryl rings
are now considerably larger than in the free ligand, 8.6(4)°
(16.5(5)°) versus 4.9(3)° (4.4(3)°).
X-ray crystallographic studies have shown that 4 and 5 are iso-
structural, for which reason we describe only [(HL3)2CuTb(NO3)3],
5. A perspective view of this dinuclear molecule with the corre-
sponding labeling scheme is depicted in Fig. 5 while selected bond
distances and angles are listed in the legend to the figure. Two HL3
ligands sandwich one Cu2+ ion through their N,O sites and one Tb3+
ion through their O2 site, implying a relative arrangement close to
head to head, at variance with the relative arrangement close to
The [(HL2)2Cu] molecules of 1 are interconnected through
N2H2ꢀꢀꢀO3, N2H2ꢀꢀꢀO4, N4H4ꢀꢀꢀO1 and N4H4ꢀꢀꢀO2 contacts into
chains developing along c as shown in Fig. 3. The chains of this
1D structure are stacked along a and b through van der Vaals
interactions.

2116
F.Z. Chiboub Fellah et al. / Polyhedron 29 (2010) 2111–2119
Fig. 3. DIAMOND plot showing one chain (fragment of 5 molecules) of [(HL2)2Cu], 1, developing along c. Interatomic NꢀꢀꢀO distances (Å) characterizing the relevant contacts:
N2(H2)ꢀꢀꢀO3, 3.069(18); N2(H2)ꢀꢀꢀO4, 2.844(19); N4(H4)ꢀꢀꢀO1, 3.147(17); N4(H4)ꢀꢀꢀO2, 2.950(19).
Fig. 4. DIAMOND plot of [(HL1)2Cu(CH3CN)2Gd(NO3)3], 3. Thermal ellipsoids are drawn at the 50% probability level. H atoms have been omitted for clarity. Selected bond lengths
(Å) and angles (°): Cu–N2 1.949(11), Cu–N4 1.977(12), Cu–O3 1.954(9), Cu–O7 1.930(10), Cu–N5 2.64(2), 2.554(14), Gd–O3 2.370(9), 2.610(8), Gd–O7 2.362(10), Gd–O8
2.607(9), Gd–O9 2.465(10), Gd–O10 2.480(10), Gd–O12 2.519(14), Gd–O13 2.470(13), Gd–O15 2.483(12), Gd–O16 2.457(11), CuꢀꢀꢀGd 3.460(3), N2–Cu–O3 89.6(4), O3–Cu–O7
82.0(3), O7–Cu–N4 91.3(5), N4–Cu–N2 100.6(4), N2–Cu–O7 162.5(5), N4–Cu–O3 163.6(5), N5–Cu–N6 176.0(5), N5–Cu–N2 83.8(5), N5–Cu–N4 99.1(5), N5–Cu–O3 94.7(5),
N5–Cu–O7 81.6(5), N6–Cu–N2 97.0(5), N6–Cu–N4 84.6(5), N6–Cu–O3 81.4(4), N6–Cu–O7 97.0(4), O3–Gd–O7 65.2(3), O3–Gd–O9 79.4(3), O3–Gd–O12 147.2(4), O4–Gd–O3
62.5(3), O4–Gd–O7 114.3(3), O4–Gd–O8 176.7(3), O7–Gd–O8 62.6(3), O7–Gd–O9 92.6(3), O7–Gd–O10 130.8(3), O9–Gd–O10 51.7(3), O12–Gd–O13 50.0(4), O15–Gd–O16
51.0(4).
head to tail of their HL2 analogs in the mononuclear copper precur-
sor, [(HL2)2Cu]. The dihedral angles between the benzimidazolyl
groups and the aryl rings are also considerably larger than that
in the free ligand H2L2 (9.7(3)° (23.6(3)°) versus 4.9(3)° (4.4(3)°)).
The Cu2+ ion has square-planar geometry; however, N1, O1, O3
and N3 deviate significantly from their mean plane: the angle be-
tween the CuO3N3 and CuN1O1 planes is 26.1(2)°. The terbium
ion is ten-coordinate. In addition to the four oxygen atoms from
the {(HL3)2Cu} constituting unit, six oxygen atoms from three
bidentate nitrato anions complete its coordination sphere. The
two metal ions are doubly bridged to one another through two
phenoxo oxygen atoms0 belonging to the HL3 ligand with a CuꢀꢀꢀTb
separation of 3.458(2) ÅA. The dihedral angle between the O1TbO3
and O1CuO3 planes equals 0.8(2)°.

F.Z. Chiboub Fellah et al. / Polyhedron 29 (2010) 2111–2119
2117
ing chain interactions. As shown in Fig. 6, the fits obtained when
computing the magnetic susceptibility with the assumption of an
Heisenberg chain of S = 1/2 spins [32] were fairly good for the
parameter values g = 2.03, J (cmꢁ1) = ꢁ0.76, Par (paramagnetic
P
contribution) = 0.0%, with an agreement factor R ¼ ½ð
v
MTÞobs
ꢁ
P
2
2
ðv
MTÞcalcꢃ = ½ð
v
MTÞobsꢃ = 2.3 ꢂ 10ꢁ4
.
At 300 K, the
v
MT product for complex
4
equals
8.27 cm3 molꢁ1 K, which is very close to the value expected for
one copper and one gadolinium ions without magnetic interactions
(8.25 cm3 molꢁ1 K). Lowering the temperature results initially in a
smooth increase of
v
MT (8.51 cm3 molꢁ1 K at 100 K) and then in a
steeper increase, up to 9.73 cm3 molꢁ1 K at 8 K, followed by a
smooth decrease to 9.70 cm3 molꢁ1 K at 2 K (Fig. 7). The maximum
value compares well with that expected for an S = 4 spin-state
resulting from ferromagnetic interaction between Gd (S = 7/2)
and Cu (S = 1/2) centers with gCu = gGd = 2. Similar experimental
data have been obtained for complex 3, [(HL1)2Cu(CH3CN)2-
Gd(NO3)3] (Fig. S1).
Fig. 5. DIAMOND plot of [(HL3)2CuTb(NO3)3], 5. The thermal ellipsoids are drawn
at the 30% probability level. H atoms have been omitted for clarity. Selected bond
lengths (Å) and angles (°):Cu–N1 = 1.925(5), Cu–N3 = 1.934(5), Cu–O1 = 1.934(4),
A quantitative analysis has been performed on the basis of an
expression derived from the spin-only Hamiltonian H ¼ ꢁJCuGd
ðSCuSGdÞ. Least squares fitting to the experimental data obtained for
4 leads to JCuGd = 6.5(2) cmꢁ1, g = 1.99(1) with a good agreement fac-
Cu–O3 = 1.914(4),
Tb–O1 = 2.337(4),
Tb–O2 = 2.687(4),
Tb–O3 = 2.361(4),
Tb–O4 = 2.609(4),
Tb–O5 = 2.448(5),
Tb–O6 = 2.465(5),
Tb–O8 = 2.470(5),
Tb–O9 = 2.479(4), Tb–O11 = 2.468(5), Tb–O12 = 2.471(5), CuꢀꢀꢀTb = 3.458(2), N1–
Cu–O1 = 92.10(18), O1–Cu–O3 = 80.67(16), O3–Cu–N3 = 91.80(19), N3–Cu–
N1 = 101.5(2), O1–Cu–N3 = 159.3(2), N1–Cu–O3 = 157.6(2), Cu–O1–Tb 107.75(17),
Cu–O3–Tb 107.53(17), O1–Tb–O2 61.51(13), O1–Tb–O3 = 64.05(14), O1–Tb–O4
118.41(13), O1–Tb–O5 80.85(16), O1–Tb–O8 153.26(18), O1–Tb–O9 130.85(15),
O2–Tb–O4 179.64(16), O5–Tb–O6 52.02(15), O6–Tb–O8 74.91(19), O9–Tb–O4
110.71(15), O9–Tb–O11 72.48(17), O12–Tb–O5 164.47(15), O12–Tb–O6
130.48(16).
P
P
2
2
tor, R = 1 ꢂ 10ꢁ4 (R ¼ ½ð
v
MTÞobs ꢁ ð
v
MTÞcalcꢃ = ½ð
v
MTÞobsꢃ ). In or-
der to take into account the slight
v
MT decrease at very low
temperature, a Weiss constant h has been introduced in the theoret-
ical expression, leading to the minute ꢁ0.006 K value (Fig. 7).
4.6. Magnetic properties
The magnetic susceptibilities of the CuII precursor 1 and Cu–Ln
complexes 3–5 have been measured in the 2–300 K temperature
range in an applied magnetic field of 0.1 T. The thermal variation
of the
v
MT product for complexes 1, [(HL2)2Cu], and 4,
[(HL3)2CuGd(NO3)3], are shown in Figs. 6 and 7, respectively, vM
being the molar magnetic susceptibility corrected for the diamag-
netism of the ligands. At 50 K, the
vMT product for complex 1
equals 0.37 cm3 molꢁ1 K, which is very close to the value expected
for an isolated CuII ion (0.374 cm3 molꢁ1 K for g = 2. The
vMT prod-
uct remains practically constant down to 12 K (0.36 cm3 molꢁ1 K)
and then decreases to 0.21 cm3 molꢁ1 K at 2 K. Considering the
1D chains formed through the N2H2ꢀꢀꢀO3, N2H2ꢀꢀꢀO4, N4H4ꢀꢀꢀO1
and N4H4ꢀꢀꢀO2 contacts, we fitted the magnetic data by consider-
Fig. 7. Thermal variation of the magnetic susceptibility,
vM, and vMT product for
complex 4, [(HL3)2CuGd(NO3)3]ꢀ2(H2O). The solid line corresponds to the best fit of
the experimental data (see text).
Fig. 6. Thermal variation of the magnetic susceptibility,
vM, and vMT product for
complex 1, [(HL2)2Cu]. The solid line corresponds to the best fit of the experimental
Fig. 8. Thermal variation of the magnetic susceptibility,
complex 5, [(HL3)2CuTb(NO3)3]ꢀ2(H2O).
vM, and vMT product for
data (see text).

2118
F.Z. Chiboub Fellah et al. / Polyhedron 29 (2010) 2111–2119
A
quantitative analysis performed for complex 3,
of 7.2 and 6.5 cmꢁ1 have been found, respectively. We have previ-
ously shown that a correlation between the interaction parameter J
and the above mentioned dihedral angle does exist: lowering the
bending of the CuO2Gd core causes an increase of the ferromag-
netic interaction [3]. The J values obtained for complexes 3 and 4
do not seem to agree with our previous data: J values around
10 cmꢁ1 would be expected. A closer look at the molecular struc-
tures of 3 and 4 evidences the main difference. In the present
examples the central CuO2Gd core is the only part of the structure
to be planar: the two independent (HL3)ꢁ ligands linked to the Cu
and Gd ions depart to a large extent from planarity. This situation
is at variance with the previous examples where the Cu and Gd
ions were coordinated to a unique ligand and where the deforma-
tions were located at the CuO2Gd core. It becomes clear that the
dihedral angle is not the right parameter to gauge the interaction
value, here. But it is clear that the orbitals of the bridging phenoxo
oxygen atoms are not correctly oriented to yield the larger interac-
tion, thus explaining that J values lower than 10 cmꢁ1 are obtained
for complexes 3 and 4.
[(HL1)2Cu(CH3CN)2Gd(NO3)3], on the basis of the same spin-only
Hamiltonian leads to very similar parameter values, JCuGd
7.2(2) cmꢁ1, g = 2.01(2) and R = 1 ꢂ 10ꢁ4 (Fig. S1).
=
Although a quantitative analysis of complex 5, [(HL3)2CuTb
(NO3)3], is not possible, due to the orbital momentum of the
terbium ion [33,34], its magnetic behavior is shown in Fig. 8. At
300 K, the
non interacting copper and terbium ions. Upon lowering the tem-
perature,
MT decreases steadily to 10.3 cm3 molꢁ1 K at 20 K before
reaching a value of 9.3 cm3 molꢁ1 K at 3 K. The decrease of
MT
with temperature in the high temperature range is most probably
governed by depopulation of the Tb Stark levels, precluding obtain-
ing reliable information on the presence and sign of the Cu–Tb
interaction, these two magnetic phenomena being competitive.
v
MT product equals 12.2 cm3 molꢁ1 K, as expected for
v
v
5. Discussion
The unsubstituted parent ligand of H2L2 and H2L3, 2-(1H-ben-
zimidazol-2-yl)phenol [35,36], and substituted derivatives [37],
have been previously described. To the best of our knowledge how-
ever, the synthetic route to H2L2 and H2L3, is novel as compared to
that used in the previous reports [35–37] (condensation between
salicylic acid (or its substituted derivatives) and o-phenylenedi-
amine). These organic molecules have been studied as laser dyes
[38–40], while their coordination chemistry has been studied in
conjunction with the electro- [41] and photoluminescence [42]
properties of their complexes. The FeIII chemistry of this type of li-
gand has also been explored owing to the interest of their mononu-
clear and dinuclear oxo (hydroxo) bridged FeIII complexes as
models of non-heme iron metalloproteins with mono and dinucle-
ar active sites [38,43,44]. However to date, the potential dinucleat-
ing ability of their 6-methoxy(phenol) derivatives has not been
explored.
In previous work [33,45], we have described several examples
of strictly dinuclear Cu–Gd complexes that exhibit ferromagnetic
behavior. Once we know that the Cu–Gd interaction is ferromag-
netic in a large majority of simple dinuclear complexes, it is inter-
esting to link these units in order to aim at high-spin entities. In
this contribution, we have considered ligands including simulta-
neously one 6-methoxyphenol and one benzimidazole moieties
in view of the potential dinucleating and bridging abilities offered
by their 6-methoxyphenol and imidazole parts, respectively. In-
deed, the bridging ability of such ligands may result from deproto-
nation of their benzimidazole ring to yield an imidazolate bridge. A
look at the molecular structure of the Cu–Ln complexes 4 and 5
shows that the mean planes of the two HL3 ligands are far from
being aligned. Deprotonation of the benzimidazole functions
would release two supplementary coordination sites: their
involvement as bridges between dinuclear Cu–Ln units would sig-
nificantly modify the relative arrangement of the two ligands.
Unfortunately, attempts to further deprotonate (HL2)ꢁ and (HL3)ꢁ
did not yield the desired products: we could not isolate crystals
from the ill-defined powder compounds obtained. However, the
magnetic data collected on samples of these compounds indicate
that the ferromagnetic Cu–Gd interaction is lost. We have observed
that Ln complexation implies a rearrangement of the two (HL2)ꢁ or
(HL3)ꢁ ligands chelated to the copper ion, cf. the molecular struc-
tures of 4 and 5 versus that of 1. It is then quite plausible that
the presence of an additional nitrogen donor at each ligand is able
to promote a novel rearrangement: indeed benzimidazole type li-
gands are good chelating ligands toward Ln ions [46,47].
6. Conclusion
We have shown that benzimidazole ligands 2-substituted by
phenyl rings bearing phenol and methoxy functions can yield hete-
rodinuclear Cu–Ln complexes. These 2-(20-hydroxy-30-methoxy-
phenyl)benzimidazole derivatives coordinate copper ions in a
head to tail arrangement while introduction of a lanthanide ion in-
duces a rearrangement of the ligands in head to head relative posi-
tions, allowing formation of a new O2O2 coordination site available
for the incoming Ln ion. This situation has been encountered for
bidentate 1-NH and 1-N substituted benzimidazole ligands. The
Cu–Gd interactions are ferromagnetic, as observed in a large
majority of Cu–Gd complexes. Nevertheless, these new complexes
are complementary to the previous ones for the Cu–O2–Gd core is
kept planar while deformations are borne by the ligands, at vari-
ance with previous examples in which the constraints were located
at the Cu–O2–Gd core. The main difference originates from the
presence of two independent ligands in the coordination spheres
conferring a degree of freedom greater than that allowed by a un-
ique tetradentate ligand. Consequently, at variance with our previ-
ous suggestion, the strength of the magnetic interaction is not
related solely to the dihedral angle between the CuOO and GdOO
planes in the central core. Furthermore, these benzimidazole li-
gands do possess an NH function not involved in the coordination
sites of the heterodinuclear complex: through deprotonation, hete-
rodinuclear units like 4 and 5 are thus good candidates for self-
assembling into larger entities that should be high-spin species.
We have not yet reached this goal for the behavior of these hetero-
dinuclear units, upon deprotonation, is more complex than ex-
pected, each bidentate benzimidazole ligand being able to
promote a novel rearrangement. By using a unique ligand includ-
ing both benzimidazole moieties, we expect better results: work
in this direction is in progress.
Acknowledgements
F.Z.C.F. acknowledges the Algerian Ministry for Education and
Research and Université Paul Sabatier (Toulouse III) for short term
postdoctoral grants.
Appendix A. Supplementary data
In the two structurally characterized Cu–Gd complexes 3 and 4,
the dihedral angle defined by the OCuO and OGdO planes is close
to 0° (0.4(4) and 0.3(4)°), respectively) and magnetic interactions
CCDC 767999, 768000, 768001, 768002 and 768003 contain the
supplementary crystallographic data for the ligand H2L2 and com-
plexes 1, 3, 4, and 5, respectively. These data can be obtained free

F.Z. Chiboub Fellah et al. / Polyhedron 29 (2010) 2111–2119
[24] R.H. Blessing, Acta Crystallogr., Sect. A 51 (1995) 33.
2119
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: de-
posit@ccdc.cam.ac.uk. Supplementary data associated with this
article can be found, in the online version, at doi:10.1016/
j.poly.2010.04.010.
[25] (a) A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, SIR92, A program
for crystal structure solution, J. Appl. Crystallogr. 26 (1993) 343–350.;
(b) A. Altomare, M.C. Burla, M. Camalli, G.L. Cascarano, C. Giacovazzo, A.
Guagliardi, A.G.G. Moliterni, G. Polidori, R. Spagn, SIR97,
A program for
automatic solution of crystal structures by direct methods, J. Appl. Crystallogr.
32 (1999) 115–119.
[26] G.M. Sheldrick, SHELX97 [Includes SHELXS97, SHELXL97, CIFTAB], Programs for
crystal structure analysis (release 97-2), Institüt für anorganische chemie der
universität, Tammanstrasse 4, D-3400 Göttingen, Germany, 1998.
References
[27] L. Farrugia, WINGX, 1.63 Integrated system of windows programs for the
solution, Refinement and analysis of single crystal X-ray diffraction data, J.
Appl. Crystallogr. 32 (1999) 837–840.
[28] P.W. Betteridge, J.R. Carruthers, R.I. Cooper, K. Prout, D.J. Watkin, J. Appl.
Crystallogr. 36 (2003) 1487.
[29] International tables for X-ray crystallography, vol IV, Kynoch press,
Birmingham, England, 1974.
[30] P.V.D. Sluis, A.L. Spek, BYPASS, An effective method for the refinement of
crystal structures containing disordered solvent regions. Acta Crystallogr.,
Sect. A, 46 (1990) 194–201.
[1] C. Benelli, D. Gatteschi, Chem. Rev. 102 (2002) 2369.
[2] M. Sakamoto, K. Manseki, H. Okawa, Coord. Chem. Rev. 219–221 (2001) 379.
[3] J.P. Costes, F. Dahan, A. Dupuis, J.P. Laurent, Inorg. Chem. 39 (2000) 165.
[4] S. Osa, T. Kido, N. Matsumoto, N. Re, A. Pochaba, J. Mrozinski, J. Am. Chem. Soc.
126 (2004) 420.
[5] J.P. Costes, F. Dahan, W. Wernsdorfer, Inorg. Chem. 45 (2006) 5.
[6] J.P. Costes, M. Auchel, F. Dahan, V. Peyrou, S. Shova, W. Wernsdorfer, Inorg.
Chem. 45 (2006) 1924.
[7] J.P. Costes, S. Shova, W. Wernsdorfer, Dalton Trans. (2008) 1843–1849.
[8] F. He, M.L. Tong, X.M. Chen, Inorg. Chem. 44 (2005) 8285.
[9] T. Kajiwara, M. Nakano, S. Takaishi, M. Yamashita, Inorg. Chem. 47 (2008)
8604.
[10] T. Hamamatsu, K. Yabe, M. Towatari, S. Osa, N. Matsumoto, N. Re, A. Pochaba, J.
Mrozinski, J.L. Gallani, A. Barla, P. Imperia, C. Paulsen, J.P. Kappler, Inorg. Chem.
46 (2007) 4458.
[11] G. Novitchi, W. Wernsdorfer, L.F. Chibotaru, J.P. Costes, C.E. Anson, A.K. Powell,
Angew. Chem., Int. Ed. 48 (2009) 1614.
[12] C. Aronica, G. Pilet, G. Chastanet, W. Wernsdorfer, J.F. Jacquot, D. Luneau,
Angew. Chem., Int. Ed. 45 (2006) 4659.
[13] J.P. Costes, J.M. Clemente-Juan, F. Dahan, J. Milon, Inorg. Chem. 43 (2004) 8200.
[14] J. Milon, M.-C. Daniel, A. Kaiba, P. Guionneau, S. Brandès, J.-P. Sutter, J. Am.
Chem. Soc. 129 (2007) 13872–13878.
[15] S. Tanase, F. Tuna, P. Guionneau, T. Maris, G. Rombaut, C. Mathonière, M.
Andruh, O. Kahn, J.P. Sutter, Inorg. Chem. 42 (2003) 1625.
[16] M.A. Martinez Lorente, F. Dahan, V. Petrouleas, A. Bousseksou, J.-P. Tuchagues,
Inorg. Chem. 34 (1995) 5346.
[31] K. Brandenburg, H. Putz, DIAMOND
Germany, 2004.
, Version 3.1, Crystal Impact GbR, Bonn,
[32] K. Konig, V.P. Desai, B. Kanellakopulos, R. Klenze, Chem. Phys. 54 (1980) 109.
[33] J.P. Costes, F. Dahan, A. Dupuis, J.P. Laurent, Chem. Eur. J. 4 (1998) 1616.
[34] M.L. Kahn, C. Mathonière, C.O. Kahn, Inorg. Chem. 38 (1999) 3692.
[35] A.W. Addison, P.J. Burke, J. Heterocycl. Chem. 18 (1981) 803.
[36] A. Pla-Dalmau, J. Org. Chem. 60 (1995) 5468.
[37] C.G. Wahlgren, A.W. Addison, S. Burman, L.K. Thompson, E. Sinn, T.M. Rowe,
Inorg. Chim. Acta 166 (1989) 59.
[38] A. Costela, I. Garcia-Moreno, Chem. Phys. 206 (1996) 383.
[39] C.E. Fellows, U. Täuber, C.E.M. Carvalho, C.G. Carvalhaes, Braz. J. Phys. 35
(2005) 933.
[40] C.E. Fellows, U. Täuber, C.C. Rodegheri, C.E.M. Carvalho, D.F. Acevedo, S.G.
Bertolotti, C. Barbero, Opt. Mater. 27 (2004) 499.
[41] Y. Hamada, T. Sano, H. Fujii, Y. Nishio, H. Takahashi, K. Shibata, Jpn. J. Appl.
Phys. 35 (1996) L1339.
[42] Y.-P. Tong, S.-L. Zheng, X.-M. Chen, Eur. J. Inorg. Chem. (2005) 3734–3741.
[43] R.K. Dubey, U.K. Dubey, C.M. Mishra, Transition. Met. Chem. 31 (2006) 849–
855.
[44] A.K. Boudalis, J.M. Clemente-Juan, B. Donnadieu, F. Dahan, V. Psycharis, C.P.
Raptopoulou, Y. Sanakis, J.-P. Tuchagues, Inorg. Chem. 47 (2008) 11314.
[45] F.Z. Chiboub Fellah, J.P. Costes, F. Dahan, C. Duhayon, G. Novitchi, J.P.
Tuchagues, L. Vendier, Inorg. Chem. 47 (2008) 6444.
[17] N. Matsumoto, Y. Mizuguchi, G. Mago, S. Eguchi, H. Miyasaka, T. Nakashima, J.-
P. Tuchagues, Angew. Chem., Int. Ed. 36 (1997) 1860.
[18] X. Yang, R.A. Jones, R.J. Lai, A. Waheed, M.M. Oyie, A.L. Holmes, Polyhedron 25
(2006) 881–887.
[19] P. Pascal, Ann. Chim. Phys. 19 (1910) 5.
[20] A.K. Boudalis, J.-M. Clemente-Juan, F. Dahan, J.-P. Tuchagues, Inorg. Chem 43
(2004) 1574.
[21] F. James, M. Roos, Phys. Commun. 10 (1975) 343.
[22] STOE: IPDS manual, Version 2.75, Stoe & Cie, Darmstadt, Germany, 1996.
[23] CrysAlis RED, Oxford diffraction Ltd., Version 1.171.32.5, 2007.
[46] G. Canard, S. Koeller, G. Bernardinelli, C. Piguet, J. Am. Chem. Soc. 130 (2008)
1025.
[47] C. Piguet, G. Bernardinelli, G. Hopfgartner, Chem. Rev. 97 (1997) 2005.