(Schiff-base)CuII complexes with phenolate-oxygen/chloro or -oxygen/azido mixed bridges - Syntheses, crystal structures and magnetic properties

Article abstract of DOI:10.1002/ejic.200400449

Four new Cu^II complexes of formulae [Cu₃(SE) ₂(H₂O)₂-Cl₃]Cl·6H ₂O (1), Cu₃(SE)₂(N₃) ₄(DMF)₂ (2), Cu(SE)N₃·DMF (3

Full text of DOI:10.1002/ejic.200400449

FULL PAPER
(Schiff-base)CuII Complexes with Phenolate-Oxygen/Chloro or -Oxygen/Azido
Mixed Bridges ؊ Syntheses, Crystal Structures and Magnetic Properties
Yun-Bo Jiang,[a] Hui-Zhong Kou,*[a] Ru-Ji Wang,[a] and Ai-Li Cui[a]
Keywords: Chloro bridges / Azido bridges / Magnetic properties / CuII ions / Crystal structures
Four new CuII complexes of formulae [Cu3(SE)2(H2O)2-
Cl3]Cl·6H2O (1), Cu3(SE)2(N3)4(DMF)2 (2), Cu(SE)N3·DMF (3)
and Cu(SE)N3·Cu(SE)N3 (4) with tridentate Schiff-base li-
1,1) azido ligands. Uniformly spaced 1D chains with end-to-
end (µ-1,3) azido bridges are found in 3 and 4. As expected,
complexes 1 and 2 exhibit antiferromagnetic and ferromag-
gands SE−, derived from condensation of N,N-dimethylethyl- netic exchange, respectively. Compounds 3 and 4 show un-
enediamine with salicylaldehyde, have been synthesised and
usual ferromagnetic coupling though the 1,3-azido path-
characterised by IR spectroscopy, X-ray structural analyses ways. Theoretical simulations have been carried out in order
and variable-temperature magnetic susceptibility measure- to evaluate the magnitude of the magnetic exchange and the
ments. Complex 1 has an alternating phenoxy- and chloro-
bridged 1D chain structure. Complex 2 possesses trinuclear
units in which the CuII ions are bridged by single end-on (µ-
magneto-structural correlation has been investigated.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
Introduction
of ClO4Ϫ, the reaction between CuCl2 and SEϪ in a 1:1
molar ratio did not produce the expected 1:1 complex. In-
stead, a chain-like complex with the empirical formula Cu3-
(SE)2(H2O)8Cl4 (1) was obtained which can be regarded as
a mixture of Cu(H2O)2Cl2 and Cu(SE)Cl (molar ratio ϭ
1:2). It is worth noting that this complex exhibits an inter-
esting and very rare structure, namely a 1D polymeric chain
containing mixed bridges and hexa- and pentacoordinate
copper atoms. The replacement of the bridging chloro li-
gands by azido ligands yields complexes with different
structural types depending upon the solvents used (see
Scheme 1). One example is the end-on azido-bridged trinu-
clear structure 2 [Cu3(SE)2(N3)4(DMF)2]. Cu(SE)N3·DMF
(3) possesses a uniformly spaced end-to-end azido-bridged
copper() chain while Cu(SE)N3·Cu(SE)N3 (4) displays a
chain structure similar to that of 3 with a cocrystallised
mononuclear Cu(SE)N3 unit. Herein, we report the synth-
eses, structural characterisation and magnetic properties of
the four complexes.
During the past decade, particular interest has been di-
rected towards the investigation of copper() complexes
with halogen bridging ligands as well as pseudohalogens,
due to their correlative structures and ability to transmit
magnetic interactions.[1Ϫ24] Among the ligands examined,
the chloride anion has the capability of acting as either a
bridging or a terminal ligand, occupying either axial or
equatorial coordination positions. As a result, mono- or po-
lynuclear complexes as well as 1D chain-like compounds
have been obtained.[1Ϫ6] In addition, the azide ion is a ver-
satile and popular pseudohalogen ligand due to its func-
tionality as a bridging bi-, tri- and tetradentate ligand and
its rich coordination behaviour, e.g. end-on (µ-1,1 and µ-
1,1,1) or end-to-end (µ-1,3). These bridging ligands have
yielded plenty of metal complexes with a wide variety of
coordination geometries including tetrahedral or square-
planar, trigonal-bipyramidal or square-pyramidal and octa-
hedral, the geometry being dependent upon the coligands
and the counterions.[8Ϫ24]
Results and Discussion
It has been reported that {N-[2-(dimethylamino)ethyl]-
salicylaldiminato}copper() chloride, Cu(SE)Cl, can be
synthesised in an anhydrous methyl/isopropyl alcohol
Synthesis and General Characterisation
The trinuclear complex [Cu3(OH)(SE)3](ClO4)2(H2O)0.5
was always found to be the product from the reaction of
Cu(ClO4)2·6H2O with SEϪ (molar ratio ϭ 1:1).[6] We have
observed that the employment of CuCl2·2H2O instead of
Cu(ClO4)2·6H2O results in the formation of the 1D chain
complex Cu3(SE)2(H2O)8Cl4 (Cu/SEϪ, 3:2). The loss of the
SEϪ ligand around a CuII ion may be due to the partial
hydrolysis of the Schiff base while in anhydrous solvent en-
vironments 1:1 complexes were formed, as previously re-
medium.[7]
A
trinuclear complex [Cu3(µ3-OH)(SE)3]-
(ClO4)2(H2O)0.5 containing a partially cubic Cu3O4 core
was obtained from the reaction of Cu(ClO4)2 with SEϪ in
a methanol/water solution.[6] We have found, in the absence
[a]
Department of Chemistry, Tsinghua University,
Beijing 100084, China
Fax: (internat.) ϩ 86-10-62795477
E-mail: kouhz@mail.tsinghua.edu.cn
4608
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejic.200400449
Eur. J. Inorg. Chem. 2004, 4608Ϫ4615

(Schiff-base)CuII Complexes with Mixed Bridges
FULL PAPER
Scheme 1. Formation of CuII compounds
ported.[6,7] This observation indicates that the structures of buted to the ν(CϭO) of DMF. The appearance of a broad
CuII-SEϪ complexes vary with the synthetic conditions and strong absorption at 3434 cmϪ1 in 1 indicates the presence
counteranions used. Several complexes were obtained from of water molecules, consistent with the single-crystal X-ray
the reaction of complex 1 with NaN3.
diffraction results. Peaks at 2070 and 2031 cmϪ1 in the IR
The IR spectra of the complexes are similar and show spectrum of 2 are due to the stretching of the µ-1,1 azide
peaks for the ν(CϭN) bands at 1647 cmϪ1 in 1 and 1635 while the appearance of absorptions at 2074 and 2047 cmϪ1
cmϪ1 in 2, 3 and 4, suggesting the presence of a Schiff base. in 3 and 4 is due to the asymmetric stretching of the µ-1,3
The peaks at 1655 cmϪ1 in 2 and 3 can be reasonably attri- azide anions.
Figure 1. ORTEP representation of the cationic polymeric chain for 1 Ϫ 2.5H2O with the atom labelling scheme showing 30% probability
thermal ellipsoids; hydrogen atoms, solvent molecules and counterions have been omitted for clarity
Figure 2. ORTEP representation of the trinuclear unit for 2 with the atom labelling scheme showing 30% probability thermal ellipsoids;
for clarity hydrogen atoms have been omitted
Eur. J. Inorg. Chem. 2004, 4608Ϫ4615
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4609

Y.-B. Jiang, H.-Z. Kou, R.-J. Wang, A.-L. Cui
FULL PAPER
Crystal Structures
The structures of 1Ϫ4 are displayed in Figures 1Ϫ4,
respectively. Selected bond lengths and angles for complexes
1 Ϫ 2.5H2O, 2, 3 and 4 are listed in Tables 1Ϫ4, respec-
tively.
Figure 4. ORTEP representations of the structure of 4: (a) the 1D
chain, (b) the monomer with ellipsoids drawn at 30% probability
level; for clarity hydrogen atoms have been omitted
Two of the copper atoms are pentacoordinate while the
other is six-coordinate in the independent trinuclear frag-
ment. Cu(1) and Cu(3) have slightly distorted square-pyr-
amidal geometries as ascertained by the Reedijk factors τ
of 0.113 and 0.132, respectively.[30] Cu(2) is in an elongated
octahedral environment, being coordinated to two bridged
chloro atoms [Cl(1) and Cl(2)] and four oxygen atoms, two
of these being from water molecules [O(3) and O(4)] and
two from the SEϪ ligands [O(1) and O(2)]. For Cu(2), the
˚
CuϪCl bonds [2.7176(15) and 2.7890(15) A] are longer
than CuϪO bonds [2.010(3), 1.984(4), 1.981(3) and 1.966(3)
˚
A]. The elongated octahedron shares two equator-to-apex
edges with the bases of two other square pyramids. The
neighbouring pyramids share an apex occupied by a
bridged chloro atom, yielding a 1D chain. Their basal
planes, defined by N, N, O and Cl, are not coplanar and
exhibit a dihedral angle of 21.1°, while the angles between
these planes and the octahedral equatorial plane
[O(1),O(2),O(3),O(4)] are 100° and 99.2°, respectively.
Figure 3. a) ORTEP representation of the 1D polymeric chain for
3 with the atom labelling scheme showing 30% probability thermal
ellipsoids; for clarity hydrogen atoms and solvent DMF molecules
have been omitted; b) view of the packing of the 1D chains in
complex 3 (DMF was displayed in space-filling diagram)
˚
Within the chain, the Cu(1)ϪCu(2) distance is 3.257(7) A,
˚
The structure of 1 Ϫ 2.5H2O consists of trinuclear cationic similar to the Cu(2)ϪCu(3) distance of 3.223(7) A, while
nϩ
˚
[Cu3(SE)2(H2O)2Cl3]n chains along the a axis formed by the Cu(3)ϪCu(1A) separation [5.161(14) A] is very large.
the chloro and phenolate oxygen bridges of SEϪ (Figure 1),
Complex 2 contains a hexacoordinate and two pentaco-
discrete ClϪ counteranions and solvated water molecules. ordinate CuII ions which are bridged by the two end-on
4610
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 4608Ϫ4615

(Schiff-base)CuII Complexes with Mixed Bridges
FULL PAPER
˚
˚
Table 1. Selected bond lengths [A] and angles [°] for 1 Ϫ 2.5H2O
Table 4. Selected bond lengths [A] and angles [°] for 4
Cu(1)ϪN(5)
Cu(1)ϪN(3)
Cu(1)ϪCl(3)
Cu(2)ϪO(2)
Cu(2)ϪO(1)
Cu(2)ϪCl(1)
Cu(3)ϪO(2)
Cu(3)ϪCl(1)
N(5)ϪCu(1)ϪO(1)
1.954(4) Cu(1)ϪO(1)
2.057(4) Cu(1)ϪCl(2)
2.5961(14) Cu(2)ϪO(4)
1.981(3) Cu(2)ϪO(3)
2.010(3) Cu(2)ϪCl(2)
2.7890(15) Cu(3)ϪN(2)
1.976(3) Cu(3)ϪN(1)
2.2577(15) Cu(3)ϪCl(3a)
90.13(15) N(5)ϪCu(1)ϪN(3)
1.994(3)
2.2524(14)
1.966(3)
1.984(4)
2.7176(15)
1.958(4)
2.071(4)
2.5974(14)
84.44(17)
Cu(1)ϪO(1)
Cu(1)ϪN(5)
CuϪN(7a)
Cu(2)ϪN(3)
Cu(2)ϪN(4)
O(1)ϪCu(1)ϪN(1)
N(1)ϪCu(1)ϪN(5)
1.905(4) Cu(1)ϪN(1)
1.972(7) Cu(1)ϪN(2)
2.604(6) Cu(2)ϪO(2)
1.926(6) Cu(2)ϪN(8)
2.072(7)
1.955(6)
2.068(6)
1.891(6)
1.946(9)
93.4(2)
O(1)ϪCu(1)ϪN(5
92.4(2)
164.8(3)
O(1)ϪCu(1)ϪN(2) 177.2(2)
N(6)ϪN(5)ϪCu(1) 121.0(5)
O(2)ϪCu(2)ϪN(8) 92.0(4)
O(2)ϪCu(2)ϪN(4) 176.3(3)
N(8)ϪCu(2)ϪN(4) 90.7(4)
N(10)ϪN(9)ϪN(8) 177.0(12)
N(6a)ϪN(7a)ϪCu(1) 124.4(3)
O(2)ϪCu(2)ϪN(3)
N(3)ϪCu(2)ϪN(8)
N(3)ϪCu(2)ϪN(4)
N(7)ϪN(6)ϪN(5)
93.2(3)
173.2(5)
84.3(3)
O(1)ϪCu(1)ϪN(3) 170.15(15) N(5)ϪCu(1)ϪCl(2) 163.20(14)
O(1)ϪCu(1)ϪCl(2)
N(5)ϪCu(1)ϪCl(3) 96.26(14) O(1)ϪCu(1)ϪCl(3)
87.38(9)
N(3)ϪCu(1)ϪCl(2)
95.45(12)
94.05(10)
177.1(8)
N(3)ϪCu(1)ϪCl(3) 94.70(12) Cl(2)ϪCu(1)ϪCl(3) 100.49(5)
O(4)ϪCu(2)ϪO(2) 166.03(15) O(4)ϪCu(2)ϪO(3)
89.75(16)
92.55(15)
Symmetry operation: a: 0.5 Ϫ x, y, z Ϫ 0.5.
O(2)ϪCu(2)ϪO(3)
O(2)ϪCu(2)ϪO(1)
O(4)ϪCu(2)ϪCl(2)
O(3)ϪCu(2)ϪCl(2)
N(2)ϪCu(3)ϪO(2)
92.04(15) O(4)ϪCu(2)ϪO(1)
88.57(13) O(3)ϪCu(2)ϪO(1) 167.99(15)
91.50(12) O(2)ϪCu(2)ϪCl(2) 102.24(10)
92.94(12) O(1)ϪCu(2)ϪCl(2)
89.65(14) N(2)ϪCu(3)ϪN(1)
75.23(9)
84.65(16)
azido ligands (Figure 2). The central copper atom has an
elongated octahedral coordination geometry with four az-
ido nitrogen atoms (two terminal ligands, two end-on dico-
ordinated bridging ligands) in equatorial positions [CuϪN
O(2)ϪCu(3)ϪN(1) 170.86(14) N(2)ϪCu(3)ϪCl(1) 163.06(13)
O(2)ϪCu(3)ϪCl(1 88.47(10) N(1)ϪCu(3)ϪCl(1) 94.90(12)
N(2)ϪCu(3)ϪCl(3a) 92.80(13) O(2)ϪCu(3)ϪCl(3a) 92.79(10)
N(1)ϪCu(3)ϪCl(3a) 94.62(11) Cl(1)ϪCu(3)ϪCl(3a) 104.10(5)
Cu(2)ϪCl(1)ϪCu(3) 78.60(13) Cu(1)ϪCl(2)ϪCu(2) 81.38(4)
Cu(1)ϪO(1)ϪCu(2) 108.99(13) Cu(2)ϪO(2)ϪCu(3) 109.05(14)
˚
1.976(3)Ϫ2.042(3) A] and two phenoxy oxygen atoms from
the SEϪ ligands in the axial positions (CuϪO 2.469 A).
˚
Alternatively, the trinuclear structure of 2 is composed of
two [Cu(SE)(DMF)]ϩ cations connected by a [Cu(N3)4]2Ϫ
anion through two azido ligands. Similar to complex 1, the
elongated octahedron shares two equator-to-vertex edges
with the bases of two other square pyramids. With a τ fac-
tor of 0.048, the basal plane of the slightly distorted square
pyramid is defined by N(2)ϪN(3)ϪN(4)-O(1) and the cop-
Symmetry operation: a: 1 Ϫ x, Ϫy, 1 Ϫ z.
˚
Table 2. Selected bond lengths [A] and angles [°] for 2
˚
per atom Cu(1) is 0.1570(5) A above the basal plane
Cu(1)ϪO(1)
Cu(1)ϪN(4)
1.927(2) Cu(1)ϪN(2)
2.001(3) Cu(1)ϪN(3)
1.941(3)
2.050(3)
1.976(3)
2.042(3)
2.468(3)
towards the apical oxygen atom O(5) of the DMF molecule.
The CuϪNϪCu angle is 102.34(13)°. It is evident that the
basal planes of the square pyramids are perfectly coplanar,
while the angle between these planes and the equatorial
plane of the central copper ion is 70°. The intramolecular
Cu(1)ϪO(2)
Cu(2)ϪN(7)
2.326(3) Cu(2)ϪN(7a)
1.976(3) Cu(2)ϪN(4)
Cu(2)ϪN(4a)
2.042(3) Cu(2)ϪO(1)
Cu(2)ϪO(1a)
2.468(3)
O(1)ϪCu(1)ϪN(2)
N(2)ϪCu(1)ϪN(4
N(2)ϪCu(1)ϪN(3)
O(1)ϪCu(1)ϪO(2)
N(4)ϪCu(1)ϪO(2)
N(7a)ϪCu(2)ϪN(4)
N(7)ϪCu(2)ϪN(4a)
N(6)ϪN(5)ϪN(4)
92.33(11) O(1)ϪCu(1)ϪN(4)
169.19(13) O(1)ϪCu(1)ϪN(3)
84.33(13) N(4)ϪCu(1)ϪN(3
95.29(11) N(2)ϪCu(1)ϪO(2)
96.20(12) N(3)ϪCu(1)ϪO(2)
91.60(15) N(7)ϪCu(2)ϪN(4)
86.62(11)
172.15(12)
95.33(13)
94.61(12)
92.06(12)
88.40(15)
˚
Cu(1)ϪCu(2) distance is 3.150(5) A.
The structure of 3 consists of end-to-end azido-bridged
1D infinite uniformly spaced chains with DMF molecules
situated in the spaces between the chains (Figure 3). In the
chain, each copper atom is coordinated by an SEϪ ligand
and two azido ligands in a slightly distorted square-pyrami-
dal arrangement (τ ϭ 0.110). The basal plane is defined by
N(1)ϪN(2)ϪN(3A)ϪO(1) and the copper atom is 0.1482(4)
91.60(15) Cu(1)ϪN(4)ϪCu(2) 102.34(13)
178.5(6)
Cu(2)ϪO(1)ϪCu(1)
90.70(15)
Symmetry operation: a: Ϫ1 ϩ x, y, z.
˚
A above the basal plane towards N(5) in the apical position.
The angles between the copper ions and the azido ligands
are
119.1(2)°
[CuϪN(5)ϪN(4)]
and
127.0(2)°
˚
[CuϪN(3A)ϪN(4A)] and the CuϪN3ϪCu torsion angle is
81.5°. The intrachain CuϪCu distance is 5.530(4) A, while
the closest interchain CuϪCu distance is 7.671(5) A. Inter-
stitial DMF molecules occupy interchain spaces and con-
nect neighbouring 1D chains through van der Waals inter-
actions and hydrogen bonds.
Table 3. Selected bond lengths [A] and angles [°] for 3
˚
˚
CuϪO(1)
CuϪN(3a)
1.9126(19)
1.974(3)
CuϪN(1)
CuϪN(2)
1.948(2)
2.067(2)
CuϪN(5)
2.474(2)
O(1)ϪCuϪN(1)
N(1)ϪCuϪN(3a)
N(1)ϪCuϪN(2)
O(1)ϪCuϪN(5)
N(3a)ϪCuϪN(5)
N(5)ϪN(4)ϪN(3)
92.76(9)
O(1)ϪCuϪN(3a)
O(1)ϪCuϪN(2)
N(3a) ϪCuϪN(2)
N(1)ϪCuϪN(5)
N(2)ϪCuϪN(5)
92.42(10)
173.71(8)
89.63(11)
88.81(9)
167.08(10)
84.05(10)
94.72(9)
102.55(10)
176.4(3)
Complex 4 contains 1D chains similar to complex 3 with
discrete Cu(SE)N3 monomers beside the chains (Figure 4).
In the chains, each copper atom is in a slightly distorted
square-pyramidal arrangement with a τ factor of 0.200. The
basal plane of the square pyramid is defined by
90.64(9)
Symmetry operation: a: 0.5 ϩ x, 1.5 Ϫ y, 2 Ϫ z.
Eur. J. Inorg. Chem. 2004, 4608Ϫ4615
N(1)ϪN(2)ϪN(5)ϪO(1) and Cu2ϩ is 0.1264(8) A above the
˚
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4611

Y.-B. Jiang, H.-Z. Kou, R.-J. Wang, A.-L. Cui
FULL PAPER
basal plane towards the bridging azido nitrogen atom
N(7A) in the apical position. The bond angles between
copper ions and azido ligands are 124.4(5)°
[Cu(1)ϪN(7A)ϪN(6A)] and 121.0(5)° [Cu(1)ϪN(5A)Ϫ
N(6A)] and the torsion angle (CuϪN3ϪCu) is 80.2°. The
˚
intrachain CuϪCu distance is 5.392(2) A. The copper atom
in the monomer is in a square-planar arrangement and is
tetracoordinated by SEϪ and one terminal azido ligand.
The chains and monomers form a complicated network by
hydrogen bonds and van der Waals interactions.
Magnetic Properties
The χMT value at 300 K for 1 is 1.229 emu·K·molϪ1 and
this decreases slowly with decreasing temperature. The
Weiss constant derived from the χϪm1 vs. T plot is equal to
Ϫ12.8 K with the Curie constant of 1.26 emu·K·molϪ1 be-
ing close to the calculated value of 1.125 emu·K·molϪ1 (g ϭ
2.0) per Cu3 unit. The overall behaviour of 1 corresponds
to a weak antiferromagnetically coupled system. To evalu-
ate the strength of the intermetallic exchange constants of
1, we used an approximate approach similar to that pre-
viously used for 1D and quasi-2D complexes.[32] On the
basis of the crystal data, for simplicity, the two coupling
constants between Cu(1) and Cu(2) and between Cu(3) and
Cu(2) which are bridged by the phenoxy and chloro bridges
were assumed to be equal. The single axis-to-axis
CuϪClϪCu coupling constants were considerably small
compared with the coupling through the phenolate oxygen
and the equator-to-vertex chloro bridges. Therefore, the 1D
chain can be treated as being composed of weakly inter-
acting trimers. The magnetic susceptibility data were fitted
using the following expansion [Equation (1)] in which in-
tertrimer and interchain exchanges were included in the
parameter θ:
Figure 5. Plot of χM vs. T (open triangles) and χMT vs. T (open
circles) for 1; the solid lines represent the theoretical values based
on the parameters described in the text
gen-bridging copper orbital planes. The delocalisation is
poor due to the small overlap and, consequently, the low
superexchange can be understood.
The temperature dependence of the magnetic suscepti-
bility of complex 2 illustrated in Figure 6 suggests the pres-
ence of ferromagnetic coupling between the adjacent CuII
ions. The Weiss constant derived from the χϪm1 vs. T plot is
equal to ϩ12.1 K and the Curie constant of 1.31
emu·K·molϪ1 is close to the calculated value of 1.125
emu·K·molϪ1 (g ϭ 2.0) per trimer. According to the crystal
structure data, complex 2 has a symmetrical trinuclear en-
tity which requires only one coupling constant. Similar to
complex 1, the magnetic susceptibility data can be success-
fully fitted using Equation (1) while intertrimeric exchange
was included in the parameter θ. The best fit results, shown
as the solid lines in Figure 6, are J ϭ ϩ32.6 cmϪ1, g ϭ 2.07
and θ ϭ Ϫ0.44 K with R ϭ 1.6 ϫ 10Ϫ4
.
The results of the best fit, shown as the solid lines in
Figure 5, are J ϭ Ϫ4.89 cmϪ1, g ϭ 2.02 and θ ϭ Ϫ0.69 K
with the agreement factor R ϭ Σ[(χMT)obsd. Ϫ (χMT)calcd.]2/
Σ[(χMT)2obsd.] of 6.7 ϫ 10Ϫ4
.
Generally speaking, oxygen-bridging exchange interac-
tions are much stronger than those involving halogens. On
the basis of the structural data of 1, the chloro-bridging
conformation involves equator-to apex linkages [Cl(1) and
Cl(2)] and axis-to-axis linkages [Cl(3)] while oxygen atoms
occupy equator-to-equator positions. As a result, strong
magnetic coupling with a large value (|J| Ͼ 100 cmϪ1) can
be expected via the oxygen bridge. In our case, however, the
observed superexchange interaction (J ϭ Ϫ4.89 cmϪ1) is
smaller than expected. We have ascribed the low superex-
Figure 6. Plot of χM vs. T (open triangles) and χMT vs. T (open
circles) for 2; the solid lines represent the theoretical values based
on the parameters described in the text.
change to the small overlap of neighbouring CuII dx Ϫy
Most of the µ1,1-azido bridged complexes are ferromag-
2
2
2
2
magnetic orbitals. If the dx Ϫy orbitals interact in the same netic as determined by the spin distribution in the spin trip-
plane, strong coupling can be observed. In complex 1, there let ground states according to the spin polarisation mecha-
are large dihedral angles of 100° and 99.2° between the oxy- nism.[25,26] It has been widely cited that the superexchange
4612
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 4608Ϫ4615

(Schiff-base)CuII Complexes with Mixed Bridges
FULL PAPER
interactions via the end-on azido bridges become antiferro-
magnetic at higher CuϪNϪCu angles.[8Ϫ13] From the den-
sity functional study by Ruiz et al., the critical angle is
104°.[8] Also, Thompson et al. suggested that the realm of
ferromagnetic coupling ceases for angles greater than
108°.[9] However, an unprecedented low CuϪNϪCu angle
of 89.1° was recently reported with an actual antiferromag-
netic coupling.[13] This has been assumed to be due to mini-
mal orbital overlap in a basal-axial fashion and a CuϪN
˚
˚
distance (2.505 A) longer than the cutoff distance (2.05 A)
for ferromagnetic interactions. In our case, the ferromag-
netic coupling is in agreement with a CuϪNϪCu angle of
102.4° which is lower than the critical angle of 104° and the
˚
CuϪN distances [2.001(3) and 2.042(3) A] are shorter than
2.05 A. Generally speaking, the basal-to-basal coordination
Figure 8. Plot of χM vs. T (open triangles) and χMT vs. T (open
circles) for 4; the solid lines represent the theoretical values based
on the parameters described in the text
˚
suggests a strong exchange coupling parameter and the ac-
tual value of ϩ32.6 cmϪ1 for 2 can be understood in spite
of the moderately large dihedral angle (70°) between the
2
2
neighbouring Cu basal dx Ϫy orbital planes. It should be
mentioned that we have not considered the possible interac-
tion through phenolate oxygen groups. Obviously, the
superexchange interaction is negligible due to its basal-to-
netic CuII 1D system. However, a high-temperature series
expansion, valid for both positive and negative J values, has
been proposed.[27,28]
This expansion is:
˚
axial orientation and a long CuϪO distance [2.469(3) A].
The gradual increase of χMT values with decreasing T
for compounds 3 and 4 clearly indicates the existence of
intrachain ferromagnetic interactions (Figures 7 and 8). The
magnetic susceptibilities above 20.0 K obey the
CurieϪWeiss law with positive Weiss constants (θ) of
ϩ9.68 K for 3 and ϩ6.3 K for 4, suggesting the existence
of ferromagnetic coupling between the adjacent CuII ions.
The Curie constant (C) is equal to 0.355 emu·K·molϪ1 for
3 which is close to the expected value of 0.375 emu K molϪ1
and g ϭ 2.0, while the Curie constant of 0.747 emu K
molϪ1 for 4 is in good agreement with the calculated value
of 0.75 emu·K·molϪ1 (g ϭ 2.0). On the basis of the struc-
tural data for complex 3, the magnetic susceptibility data
were fitted using the 1D S ϭ 1/2 expression based on the
with A ϭ 1 ϩ 5.7979916 ϫ x ϩ 16.902653 ϫ x2
ϩ
29.376885 ϫ x3 ϩ 29.832959 ϫ x4 ϩ 14.036918 ϫ x5; B ϭ
1 ϩ 2.7979916 ϫ x ϩ 7.008678 ϫ x2 ϩ 8.6538644 ϫ x3 ϩ
4.5743114 ϫ x4 and x ϭ J/2kT.
With the consideration of the interchain interactions:[29]
the best-fit results were obtained as J ϭ ϩ6.7 cmϪ1, g ϭ
2.0, zJЈ ϭ Ϫ1.3 cmϪ1 with R ϭ 1.5 ϫ 10Ϫ3
.
Because complex 4 has a 1D chain similar to that in com-
plex 3, the magnetic susceptibilities of 4 were similarly
simulated based on the chain model. The magnetic contri-
bution of the discrete Cu(SE)N3 monomer was added sep-
arately. All the non-intrachain interactions were included in
the parameter zJЈ using the molecular-field approxi-
mation.[29] This expansion is:
ˆ
Hamiltonian H ϭ ϪJΣSiSiϩ1. According to Kahn, no ana-
lytical expression has yet been proposed for the ferromag-
The best fit results, as shown in Figure 8, are J ϭ ϩ4.64
cmϪ1, g ϭ 2.01 and zJЈϭ Ϫ0.30 cmϪ1 with R ϭ 2.6 ϫ 10Ϫ4
.
For the end-to-end bridging mode of the azide, superex-
change is generally antiferromagnetic in nature.[14,15,21Ϫ26]
Recently, a few compounds have been reported to show a
weak ferromagnetic interaction through the 1,3-bridging
azido pathway.[16Ϫ20] It was found that in these unusual
Figure 7. Plot of χM vs. T (open triangles) and χMT vs. T (open
circles) for 3; the solid lines represent the theoretical values based
on the parameters described in the text
Eur. J. Inorg. Chem. 2004, 4608Ϫ4615
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4613

Y.-B. Jiang, H.-Z. Kou, R.-J. Wang, A.-L. Cui
FULL PAPER
CuII complexes, the bridging azido ligand involves an equa-
torial linkage to one CuII ion and an axial coordination to
another CuII ion in the square-pyramidal or elongate octa-
hedral geometry.[16] It is known that for the copper() ions,
of the materials should be prepared and they should be handled
with care.
Synthesis of 1: N,N-Dimethylethylenediamine (5.5 mL, 50 mmol) in
methanol (10 mL) was mixed with salicylaldehyde (7.605 g,
50 mmol) in methanol (10 mL). CuCl2·2H2O (8.525 g, 50 mmol) in
water (50 mL) was then added slowly and the mixture was heated
to reflux for 0.5 h. The resultant dark green solution was reduced
to 50 mL by rotary evaporation. Dark green needle-shaped single
crystals suitable for X-ray analysis separated after several days.
Yield: 4.44 g (31%). C22H46Cl4Cu3N4O10 (828.03): calcd. C 30.76,
H 5.40, N, 6.52; found C 30.98, H 5.26, N 6.63. IR (KBr): ν˜ ϭ
3434 (OϪH, H2O), 1647 (CϭN) cmϪ1. The water content of 1 was
confirmed by thermogravimetric analysis.
2
2
the unpaired electrons occupy the dx Ϫy orbitals while the
2
dz orbitals are filled with the paired electrons. Since the
2
azido σ- and π-orbitals mainly overlap the full dz orbitals
via the axial interaction, there is almost no coupling be-
tween them and thus magnetic coupling is expected to be
weak. Furthermore, the 1,3-azido pathway superexchange
coupling mainly depends on the CuϪNϪN angle α and the
CuϪNϪNϪNϪCu torsion angle θ. Large α and θ values
may strongly reduce the antiferromagnetic component of
superexchange. In our examples, the effects of the
CuϪNϪN angles were not drastic (below 104°), so the ef-
fect of torsion angles should be dominant. The antiferro-
magnetic component decreases with increasing θ and be-
comes zero when θ ϭ 90° due to accidental orthogonality.
The antiferromagnetic interactions then increase until
Syntheses of 2, 3 and 4: A slight excess of sodium azide (57 mg,
0.88 mmol), dissolved in a minimum amount of water, was added
to an aqueous solution (10 mL) of 1 (170.2 mg, 0.2 mmol) with
continuous stirring. This gave an immediate yellowish-brown pre-
cipitate which was collected by filtration and used for the following
synthesis. Dark green platelet crystals of 4 were obtained by con-
centration of the filtrate at room temperature. Yield: 33.5 mg
(30%). C11H15CuN5O (593.64): calcd. C 44.51, H 5.09, N 23.59;
found C 44.26, H 5.07, N 24.20. IR (KBr): ν˜ ϭ 2074, 2048 (N3),
1635(CϭN) cmϪ1. The yellow brown precipitate was directly dis-
reaching the maximum when
θ
ϭ
180°.[31] The
CuϪNϪNϪNϪCu torsion angles of 81.5° for 3 and 80.2°
for 4 (close to 90°) nullify the antiferromagnetic component
and, as a result, net weak ferromagnetic interactions can be solved in DMF (10 mL). Black block-shaped crystals suitable for
X-ray analysis were formed by slow concentration at room tem-
perature over several weeks. Pure 2 was obtained by manual separ-
ation under a microscope. Yield: 32.6 mg (19%). C28H44Cu3N18O4
(887.43): calcd. C 37.90, H 5.00, N 28.41; found C 37.12, H 4.94,
N 27.82. IR (KBr): ν˜ ϭ 2070, 2031 (N3), 1655 (CϭO, DMF), 1636
(CϭN) cmϪ1. The yellow-brown precipitate was at first dissolved
in MeCN (10 mL). After concentration to dryness, the resultant
observed and explained on the basis of the above argu-
ments.
Conclusion
The present report describes the syntheses, X-ray crystal microcrystalline solid was collected and dissolved in DMF (10 mL).
Slow concentration of the DMF solution gave bluish-green crystals
structures and magneto-structural correlations of four novel
phenoxy/chloro- and phenoxy/azido-bridged CuII com-
plexes. The syntheses of the four compounds are intriguing
and the final products are strongly dependent on the syn-
thetic conditions. The χMT data for 1 and 2 were simulated
using a trinuclear model, while the data for 3 and 4 were
treated using a 1D infinite chain model for S ϭ 1/2. The
unusual ferromagnetic behaviour of the two µ1,3-N3-bridged
1D complexes has been rationalised qualitatively on the
basis of the axial-equatorial bridging mode as well as the
CuϪN3ϪCu torsion angles.
of suitable for X-ray analysis. Yield: 11.9 mg (8%).
3
C14H22CuN6O2 (369.92): calcd. C 45.46, H 5.99, N 22.72; found C
44.16, H 5.73, N 23.16. IR (KBr): ν˜ ϭ 2074, 2047 (N3), 1655
(CϭO, DMF), 1634 (CϭN) cmϪ1
.
X-ray Crystallographic Study: The data collections for 1 Ϫ 2.5H2O,
2, 3 and 4 were made with a Bruker Smart CCD (293 K) and a
Siemens Bruker P4 diffractometer, respectively. Intensity data were
corrected for Lorentz and polarisation factors. Absorption correc-
tions were applied using SADABS (Bruker 2000) (complexes 1 Ϫ
2.5H2O, 3 and 4) and ψ-scan (complex 2). The structures were
solved by direct methods (SHELXS-97) and refined by full-matrix
least squares (SHELXL-97) on F2. Hydrogen atoms attached to the
C and N atoms were added geometrically and refined using the
riding model. In complex 1 Ϫ 2.5H2O, considerable disorder of the
interstitial water molecules was observed and a split model was
employed during the refinement. The crystal data are summarised
in Table 5. CCDC-239759 to -239762 contain the supplementary
crystallographic data for this paper. These data can be obtained
free of charge at www.ccdc.cam.ac.uk/conts/retrieving.html [or
from the Cambridge Crystallographic Data Centre, 12 Union
Road, Cambridge CB2 1EZ, UK; Fax: ϩ 44-1223-336-033; E-mail:
deposit@ccdc.cam.ac.uk].
Experimental Section
General Remarks: Elemental analyses (C, H, N) were carried out
with an Elementar Vario EL instrument. Infrared spectroscopy was
performed with
a Magna-IR 750 spectrophotometer in the
4000Ϫ400 cmϪ1 region. The TGA measurement of 1 was per-
formed in the temperature range 20Ϫ300 °C under nitrogen with
a Universal V2.6D TA instrument. Magnetic measurements were
performed with a few single crystals (ca. 46 mg for 1, 32.6 mg for
2, 11.9 mg for 3 and 33.5 mg for 4) using a MagLab 2000 magnet-
ometer in a field of 10 kOe (1, 2 and 4) or 20 kOe (3) from 2
to 300 K. The experimental susceptibilities were corrected for the
diamagnetism of the constituent atoms (Pascal’s Tables).
Acknowledgments
This work was supported by the National Natural Science Foun-
dation of China (project nos. 20201008 and 50272034). We are
grateful to Prof. Song Gao for the magnetic measurements.
Caution! Although not encountered in our experiments, azido com-
plexes of metal ions are potentially explosive. Only small amounts
4614
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 4608Ϫ4615

(Schiff-base)CuII Complexes with Mixed Bridges
FULL PAPER
Table 5. Crystal data for complexes 1 Ϫ 2.5H2O, 2, 3 and 4
1 Ϫ 2.5H2O
2
3
4
Empirical formula
Formula mass
Space group
C23H43Cl4Cu3N4O7.5
828.03
P21/n
9.154(2)
19.589(5)
18.371(5)
94.677(6)
3283.2(15)
4
C28H44Cu3N18O4
887.43
P21/c
10.7026(13)
17.4369(15)
11.1640(15)
111.572(8)
1937.5(4)
2
C14H22CuN6O2
369.92
P212121
9.3270(19)
12.049(2)
15.321(3)
90
1721.8(6)
4
1.427
C22H30Cu2N10O2
593.64
Pca21
21.783(3)
11.927(3)
10.207(2)
90
2651.8(8)
4
˚
a [A]
˚
b [A]
˚
c [A]
β [°]
3
˚
V [A ]
Z
ρcalcd. [g cmϪ3
]
1.675
2.298
1.521
1.690
1.487
1.643
µ(Mo-Kα) [mmϪ1
]
1.286
Data/restraint/parameters
GOF
R1 [I Ͼ 2σ(I)]
wR2 (all data)
5760/1/393
1.064
0.0458
3412/0/241
1.006
0.0388
6317/0/208
1.158
0.0552
4193/1/325
0.992
0.0468
0.1183
0.1135
0.0981
0.1582
[1]
472Ϫ474. [18d] T. K. Maji, P. S. Mukherjee, G. Mostafa, T. Mal-
S. Haddad, R. D. Willett, Inorg. Chem. 2001, 40, 2457Ϫ2460.
M. Y. Wei, R. D. Willett, Inorg. Chem. 1996, 35, 6381Ϫ6385.
R. Bhattacharya, A. Ghosh, M. S. Ray, L. Righi, G. Bocelli,
S. Chaudhuri, R. D. Willett, J. M. Clemente-Juan, E. Coron-
lah, J. Cano-Boquera, N. R. Chaudhuri, Chem. Commun. 2001,
[2]
[3]
[18e]
1012Ϫ1013.
P. S. Mukherjee, S. Dalai, G. Mostafa, T.-H.
Lu, E. Rentschler, N. R. Chaudhuri, New J. Chem. 2001, 25,
1203Ϫ1207.
´
ado, C. J. Gomez-Garicia, Eur. J. Inorg. Chem. 2002,
[19]
[20]
[21]
L.-C. Li, D.-Z. Liao, Z.-H. Jiang, S.-P. Yan, Inorg. Chem. 2002,
41, 1019Ϫ1021.
B. Woodard, R. D. Willett, S. Haddad, B. Twamley, C. J. Go-
mez-Garcia, E. Coronado, Inorg. Chem. 2004, 43, 1822Ϫ1824.
C. R. Choudhuri, S. K. Dey, R. Karmakar, C.-D. Wu, C.-Z.
Lu, M. S. El Fallah, S. Mitra, New J. Chem. 2003, 27,
1360Ϫ1366.
S. K. Dey, N. Mondal, M. S. El Fallh, R. Vicente, A. Escuer, X.
Solans, M. Font-Bardfa, T. Matsushita, V. Gramlich, S. Mitra,
Inorg. Chem., in press.
S. S. Tandon, L. K. Thompson, D. O. Miller, J. Chem. Soc.,
Chem. Commun. 1995, 1907Ϫ1908.
L. K. Thompson, S. S. Tandon, F. Lloret, J. Cano, M. Julve,
Inorg. Chem. 1997, 36, 3301Ϫ3306.
J. Comarmond, P. Plumere, J.-M. Lehn, Y. Agnus, R. Louis, R.
Weiss, O. Kahn, I. Morgenstern-Badarau, J. Am. Chem. Soc.
1982, 104, 6330Ϫ6340.
M.-F. Charlot, O. Kahn, M. Chaillet, C. Larried, J. Am. Chem.
Soc. 1986, 108, 2574Ϫ2581.
O. Kahn, Molecular Magnetism, VCH, Weinheim, 1993, chap-
ter 6.
4253Ϫ4259.
[4]
[5]
[6]
´
´
´
M. Hernandez-Molina, J. Gonzalez-Platas, C. Ruiz-Perez, F.
Lloret, M. Julve, Inorg. Chim. Acta 1999, 284, 258Ϫ265.
A. M. Schuitema, A. F. Stassen, W. L. Driessen, J. Reedijk,
Inorg. Chim. Acta 2002, 337, 48Ϫ52.
H.-D. Bian, J.-Y. Xu, W. Gu, S.-P. Yan, P. Cheng, D.-Z. Liao,
Z.-H. Jiang, Polyhedron 2003, 22, 2927Ϫ2932.
E. W. Hatfield, F. L. Bunger, Inorg. Chem. 1968, 8, 1194Ϫ1195.
E. Ruiz, J. Cano, S. Alvarez, P. Alemany, J. Am. Chem. Soc.
1998, 120, 11122Ϫ11129.
S. S. Tandon, L. K. Thompson, M. E. Manuel, I. N. Bridson,
Inorg. Chem. 1994, 33, 5555Ϫ5570.
L. K. Thompson, S. S. Tandon, M. E. Manuel, Inorg. Chem.
1995, 34, 2356Ϫ2366.
P. S. Mukherjee, T. K. Maji, G. Mostafa, T. Mallah, N. R.
Chaudhuri, Inorg. Chem. 2000, 39, 5147Ϫ5150.
E.-Q. Gao, S.-Q. Bai, C.-F. Wang, Y.-F. Yue, C.-H. Yan, Inorg.
Chem. 2003, 42, 8456Ϫ8464.
S. Koner, S. Saha, T. Mallah, K.-I. Okamoto, Inorg. Chem.
2004, 43, 840Ϫ842.
C. G. Pierpont, D. N. Hendrickson, D. M. Duggan, F. Wagner,
E. K. Barefield, Inorg. Chem. 1975, 14, 604Ϫ610.
A. Escuer, R. Vicente, J. Ribas, M. S. El Fallah, X. Solans,
Inorg. Chem. 1993, 32, 1033Ϫ1035.
C. S. Hong, Y. Do, Angew. Chem. Int. Ed. 1999, 38, 193Ϫ195.
P. S. Mukherjee, S. Dalai, E. Zangrando, F. Lloret, N. R.
Chaudhuri, Chem. Commun. 2001, 1444Ϫ1445.
[22]
[7]
[8]
[23]
[24]
[25]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[26]
[27]
[28]
G. A. Baker, G. S. Rushbrooke, H. E. Gilbert, Phys. Rev. 1964,
135, A1272.
[29]
[30]
C. J. OЈConner, Prog. Inorg. Chem. 1982, 29, 203.
A. W. Addison, T. N. Rao, J. Reedijk, J. V. Rijin, G. C. Ver-
schoor, J. Chem. Soc., Dalton Trans. 1984, 1349.
A. Escuer, R. Vicente, J. Ribas, M. S. El Fallah, X. Solans, M.
Font-Bardia, Inorg. Chem. 1993, 32, 3727Ϫ3732.
[16]
[17]
[31]
[18] [18a] P. S. Mukherjee, T. K. Maji, A. Escuer, R. Vicente, J. Ribas,
G. Rosair, F. A. Mautner, N. R. Chaudhuri, Eur. J. Inorg.
Chem. 2002, 943Ϫ949. [18b] S. Dalai, P. S. Mukherjee, M. G. B.
[32] [32a] S. G. Kang, K. Ryu, S.-K. Jung, J. Kim, Inorg. Chim. Acta
[32b]
1999, 293, 140.
B. Chiari, A. Cinti, O. Piovesana, P. F.
Zanazzi, Inorg. Chem. 1995, 34, 2652Ϫ2657.
Drew, T.-H. Lu, N. R. Chaudhuri, Inorg. Chim. Acta 2002, 335,
Received May 282004
[18c]
85Ϫ90.
S. Dalai, P. S. Mukherjee, T. Mallah, M. G. B.
Early View Article
Drew, N. R. Chaudhuri, Inorg. Chem. Commun. 2002, 5,
Published Online October 7, 2004
Eur. J. Inorg. Chem. 2004, 4608Ϫ4615
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4615