Synthesis of Fe(II) and Cu(II) building blocks for metal-organic frameworks

Article abstract of DOI:10.1016/j.ica.2008.09.052

In situ reaction of the aminobenzoic acids 2-aminobenzoic acid and 3,5-diaminobenzoic acid with salicylaldehyde provide easy access to the ligands 2-[{(2-hydroxyphenyl)methylene}amino]benzoic acid (L1) and 3,5-bis[{(2-hydroxyphenyl)methylene}amino]benzoic

Full text of DOI:10.1016/j.ica.2008.09.052

Inorganica Chimica Acta 362 (2009) 2205–2212
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Synthesis of Fe(II) and Cu(II) building blocks for metal–organic frameworks
Rebecca H. Laye b, E. Carolina Sañudo a,
*
a Departament de Química Inorgànica, Universitat de Barcelona, Diagonal 647, 08028 Barcelona, Spain
b The University of Manchester, Oxford Road, M13 9PL Manchester, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
In situ reaction of the aminobenzoic acids 2-aminobenzoic acid and 3,5-diaminobenzoic acid with salicyl-
aldehyde provide easy access to the ligands 2-[{(2-hydroxyphenyl)methylene}amino]benzoic acid (L1)
and 3,5-bis[{(2-hydroxyphenyl)methylene}amino]benzoic acid (L2). Addition of a Fe(II) or Cu(II) salt to
the solution of the ligand yields the corresponding Fe and Cu complexes. The species synthesized have
been structurally characterized by single-crystal X-ray diffraction. The Fe(II) complex [Fe(L1)(MeOH)3]
Received 10 July 2008
Received in revised form
18 September 2008
Accepted 23 September 2008
Available online 8 October 2008
ꢀ
(1) crystallizes in the triclinic space group P1. The Cu(II) complex [Cu(L1)] (2) is a one-dimensional chain
and crystallizes in the monoclinic space group P21. The Cu(II) complex [Et3NH]2[Cu2(L2)2] (3) crystallizes
in the monoclinic space group P21/n. The magnetic properties of 1, 2 and 3 have been studied, showing
that the Cu(II) ions of 2 and 3 are ferromagnetically coupled. Complexes 1 and 3 have strong potential as
metal-bearing building blocks for the synthesis of metal–organic frameworks.
Ó 2008 Elsevier B.V. All rights reserved.
Keywords:
Ferromagnetic coupling
Building blocks
Imine-based ligands
Coordination complexes
1. Introduction
synthesis is straightforward, and can be done in situ without need-
ing to isolate and purify the ligand beforehand. The second and
The design and synthesis of new magnetic materials is a hot
field of research nowadays. Possible applications like high density
information storage and molecular spintronics have attracted the
interest of the scientific community in the last few years [1]. The
combination of appropriate ligands with transition metals that
possess unpaired electrons has afforded a large number of species
with interesting magnetic properties, the structural diversity of
which ranges from single-molecule magnets [2] to extended sys-
tems [3]. Bridging ligands that can mediate ferromagnetic coupling
between the metals are particularly sought after, since this ensures
a large spin ground state. Among these, the most studied are azide
[4] in the end-on bridging mode [5] and the alkoxides and oxides
with bridging angles around 90°. However, in most cases the bridg-
ing mode of the azide cannot be predicted and the very interesting
clusters synthesized have been obtained by so-called ‘serendipi-
tous assembly’ [6]. This is even more so with alkoxides and oxides,
which are usually provided for by the solvent (an alcohol for the
alkoxides or the adventitious water of an organic solvent for the
oxides) [7]. A group of ligands that can be designed and offer rigid
binding modes are the aromatic imine-based ligands, these have
been widely used in the synthesis of new coordination complexes,
some of them with interesting magnetic properties [8]. They are
synthesized by the condensation reaction between an aldehyde
and a primary amine. This type of ligands offers a unique combina-
tion of possibilities for coordination chemists. On the one hand, the
greatest asset of imine ligands is the possibility of functionaliza-
tion. The two reagents needed are an aldehyde and a primary
amine, which can be functionalized practically at will: a large num-
ber of aromatic aldehydes are available commercially, as well as
primary amines. Additionally, the fact that the two ‘building
blocks’ of the ligand can be functionalized independently leads to
the possibility of preparing libraries of ligands. In this work, given
our interest in magnetic materials, functional groups such as
carboxylates and alkoxides were sought after. Thus, 2-hydroxy-
benzaldehyde – also known as salicylaldehyde – was chosen as
the aldehyde reagent, and 2-aminobenzoic acid – also known as
anthranillic acid – and 3,5-diaminobenzoic acid as the amine do-
nors. In this way, ligands that contain one or two alkoxides and a
carboxylate group, as well as the imino N donor are obtained. Reac-
tion of the ligand, prepared in situ, with metal salts leads to the for-
mation of the Cu(II) and Fe(II) complexes reported here.
2. Experimental
All chemicals were purchased from Aldrich and were used with-
out further purifications. The imine ligand was prepared in situ for
each reaction following standard procedures. Magnetic data were
collected at the Unitat de Mesures Magnètiques at the Universitat
de Barcelona using crushed crystals of the sample on a Quantum
Design MPMS-XL SQUID magnetometer equipped with a 5 T mag-
net. The data were corrected for TIP and the diamagnetic correc-
tions were calculated using Pascal’s constants; an experimental
correction for the sample holder was applied.
* Corresponding author. Tel.: +34 93 4039144; fax: +34 93 4907725.
E-mail address: esanudo@ub.edu (E.C. Sañudo).
0020-1693/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2008.09.052

2206
R.H. Laye, E.C. Sañudo / Inorganica Chimica Acta 362 (2009) 2205–2212
2.1. [Fe(L1)(MeOH)3] (1)
Cu(BF4)2 (0.59 g, 2.5 mmol) is added to the ligand solution in
MeOH. The colour turns brown and a green precipitate forms.
The precipitate is filtered off and the solution left undisturbed.
Dark green crystals of 3 form after seven days in 28% yield. Elemen-
tal Anal. Calc.: C, 62.05; H, 5.59; N, 8.04. Found: C, 61.77; H, 5.71; N,
7.99%.
2-Aminobenzoic acid (0.27 g, 2.0 mmol) is dissolved in 50 mL
of MeOH in a Schlenck flask. 2-hydroxybenzaldehyde (0.21 mL,
2.0 mmol) and Et3N (0.55 mL, 4.0 mmol) are added to the metha-
nolic solution of 2-aminobenzoic acid. The resulting orange solu-
tion is purged under N2 for 1 h. Solid FeCl2 ꢀ 4H2O (0.39 g,
2.0 mmol) is added with stirring. The solution turns dark red. Stir-
ring is maintained for 1 h and the solution is left undisturbed, un-
der N2, for five days. Crystals of the product can be filtered off in
53% yield. The sample is stored under N2. Elemental Anal. Calc.
for 1ꢀH2O: C, 49.86; H, 5.66; N, 3.42. Found: C, 50.11; H, 5.39; N,
3.63%.
3. Crystallography
Crystal data for complexes 1 and 2 were collected at the Univer-
sity of Manchester on an Oxford Diffraction Excalibur 2 diffractom-
eter. Crystal data for complex 3 were collected on a Bruker ApexII
diffractometer at the Unidade de Raios X at the University of San-
tiago de Compostela. Hydrogen atoms were placed geometrically
and refined isotropically using a riding model. The structures were
2.2. [Cu(L1)]n (2)
solved by direct methods (SHELXS-97), and refined on F2
(SHELXL-97).
2-Aminobenzoic acid (0.27 g, 2.0 mmol) is dissolved in 50 mL of
MeOH. 2-hydroxybenzaldehyde (0.21 mL, 2.0 mmol) and Et3N
(0.55 mL, 4.0 mmol) are added to the methanolic solution of 2-ami-
nobenzoic acid. To the resulting orange solution is added Cu(BF4)2
(0.47 g, 2.0 mmol) with stirring. The solution turns dark green with
a light green precipitate. The precipitate is filtered off and the dark
green solution is left undisturbed. After seven days dark green
crystals of 2 can be isolated by filtration in 33% yield. Elemental
Anal. Calc.: C, 55.63; H, 3.00; N, 4.63. Found: C, 55.49; H, 2.95; N,
4.42%.
Collection parameters can be found in Table 1. The hydrogen atoms
on the protonated triethylamine cations of 3 were found in the dif-
ference map and refined isotropically.
4. Results and discussion
4.1. Syntheses
The ligand L1 was synthesized in situ in the reaction pot by the
condensation reaction of 2-hydroxybenzaldehyde and 2-amino-
benzoic acid, in basic MeOH, as shown in Fig. 1. Enough base was
used to keep the ligand deprotonated to avoid precipitation and fa-
vour coordination upon addition of the transition metal salt. In
fact, the addition of FeCl2 to the solution of the ligand under N2
atmosphere caused a sudden change of colour, from yellow to deep
purple, indicating the coordination of the Fe(II) ions to the imino
ligand. The solution was left undisturbed and kept under N2 to
2.3. [Et3NH]2[Cu2(L2)2] (3)
3,5-Diaminobenzoic acid (0.19 g, 1.2 mmol) is dissolved in
50 mL of MeOH. To this solution are added 2-hydroxybenzalde-
hyde (0.26 mL, 2.5 mmol) and Et3N (0.69 mL, 5.0 mmol). The colour
changes to orange, indicating the formation of the di-imine L2.
Table 1
Crystallographic data collection and refinement parameters for complexes 1, 2 and 3.
[Fe(L1)(MeOH)3] (1)
[Cu(L1)] ꢀ MeOH (2)
[Et3NH]2[Cu2(L2)2] ꢀ 4MeOH (3)
Temperature (K)
Wavelength (nm)
Crystal system
Space group
a (Å)
100 (3)
0.71073
triclinic
100 (3)
0.71073
296 (1)
0.71073
monoclinic
P21
9.5867 (2)
7.0487 (2)
9.8425 (2)
90
98.256 (2)
90
2
monoclinic
P21/n
15.0478 (3)
12.0824 (3)
16.0325 (3)
90
99.4910 (10)
90
2
ꢀ
P1
9.4113 (8)
9.9020 (9)
10.0051 (7)
71.879 (7)
84.018 (6)
75.980 (7)
2
859.32 (12)
0.911
408
b (Å)
c (Å)
a
(°)
b (°)
c
(°)
Z
Volume (Å3)
658.20 (3)
1.674
342
2875.02 (11)
0.81
1236
Absorption coefficient (mmꢁ1
F(000)
)
Crystal size (mm)
h Range for data collection (°)
Index ranges
0.20 ꢂ 0.08 ꢂ 0.04
2.59–32.13
ꢁ13 < h < 13
ꢁ10 < k < 10,
ꢁ14 < l < 14
9064
0.37 ꢂ 0.20 ꢂ 0.03
2.77–31.44
ꢁ13 < h < 13
ꢁ10 < k < 10
ꢁ14 < l < 14
9124
0.22 ꢂ 0.18 ꢂ 0.07
1.72–26.39
ꢁ18 < h < 18
ꢁ155 < k < 15
ꢁ19 < l < 19
Reflections collected
Independent reflections (Rint
Completeness (h) (%)
Absorption correction
Maximum and minimum transmission
Data/restraints/parameters
Goodness-of-fit
)
5396 (0.0533)
89.5 (32.13°)
Analytical
0.9645 and 0.8388
5396/0/229
1.044
3884 (0.039)
94.7 (31.44°)
Multi-scan
0.799 and 1.000
3884/1/185
1.05
5885 (0.046)
99.8 (26.39°)
Numerical
0.843 and 0.946
5885/0/360
1.04
Final R indices [I > 2
r
(I)]
R1 = 0.0459
wR2 = 0.1284
R1 = 0.0667
wR2 = 0.1419
0.774 and ꢁ0.765
R1 = 0.037
wR2 = 0.103
R1 = 0.041
wR2 = 0.107
0.95 and ꢁ0.70
R1 = 0.0279
wR2 = 0.0695
R1 = 0.0348
wR2 = 0.0729
0.452 and ꢁ0.335
R indices (all data)
Largest difference peak and hole (e Å3)
Refinement method: Full-matrix least-squares on F2.

R.H. Laye, E.C. Sañudo / Inorganica Chimica Acta 362 (2009) 2205–2212
2207
H
+
N
NH2
OH
O
OH
HO
O
HO
L1
O
OH
HO
H2N
NH2
N
N
H
2
+
O
OH
O
OH
O
OH
L2
Fig. 1. The ligands L1 and L2.
avoid the oxidation of the Fe(II) to Fe(III). Crystals of the new com-
plex [Fe(L1)(MeOH)3] (1) started to grow after a few hours. The
crystals were collected after five days and were of good quality
for single-crystal X-ray diffraction, which allowed the structural
determination of the complex.
The synthesis of the Cu(II) polymer [Cu(L1)]n (2) is similar. In
this case, upon addition of the Cu(II) salt to the solution of the li-
gand in MeOH, the colour changed from yellow to dark green
and after a few minutes a green precipitate formed. The precipitate
was filtered off and the dark green solution was left undisturbed to
slowly evaporate. After a few days, dark green crystals could be
isolated by filtration and single-crystal X-ray diffraction analysis
revealed the structure of complex 2. The precipitate obtained
was analyzed by infra-red spectroscopy: comparison with the in-
fra-red spectra of the crystals of 2 showed that both were the same
material.
brown-green solution of the ligand L2 in its deprotonated form.
Addition of hydrated Cu(BF4)2 to the solution of L2 in basic MeOH
afforded a very dark green solution. A precipitate was filtered off.
After a few days undisturbed, well formed crystals of the new spe-
cies [Et3NH]2[Cu2(L2)2] (3) were isolated by filtration from the
solution. The crystals were suitable for single-crystal X-ray diffrac-
tion and the crystal structure was determined. The precipitate iso-
lated was identified as compound 3 by comparison of the infra-red
spectrum of the powder with that of the crystals. In the formation
of compound 3 the ligand L2 is not using all of its possible binding
positions. In fact, the carboxylate group of the central aromatic
moiety is not coordinated. This opens up the possibility of using
the complex anion 3 as a metal-bearing building block for polyme-
tallic compounds and metal–organic frameworks. This possibility
is currently being explored.
A variety of binding modes can be predicted for the imino li-
gand L1, and two different binding modes are actually found in
the complexes reported herein. In both complexes 1 and 2, the li-
gand takes three coordination positions around the metal in a
meridional (mer) fashion, taking three neighbour positions in the
equatorial plane of an octahedral complex such as 1, or three posi-
tions in a distorted square-planar complex such as 2. Both reac-
tions are performed in MeOH, a strongly coordinating solvent. In
the formation of complex 1, given the preference of Fe(II) for hex-
acoordination, the three free coordination positions left by the li-
gand are occupied by neutral MeOH ligands, affording a neutral
complex that crystallizes easily from the polar MeOH solvent.
Cu(II), on the other hand, is usually found to be tetracoordinated.
In this case, the tridentate ligand forces a distorted square-planar
geometry around the Cu(II) ion. The fourth coordination site is ta-
ken by an oxygen atom from the neighbouring [Cu(L1)] unit linking
the monomers into a zigzag one-dimensional polymer, which ex-
plains the low solubility of 2.
To prepare polimine ligands, 3,5-diaminobenzoic acid was cho-
sen due the 1,3,5 pattern of substituents in the aromatic ring. The
two amines, meta to each other, will result in a meta-substituted
aromatic ring with two imine donor groups, which will favour fer-
romagnetic coupling in the complexes synthesized [9]. Addition-
ally, the carboxylate might extend the coordination network
further. The synthetic procedure to prepare L2 was similar to the
one followed to prepared L1 in situ. In both cases, the ligand is pre-
pared in situ, to the best of our knowledge, these organic molecules
have not been reported per se in the literature, although L1 has
been prepared in situ and used to prepare coordination complexes
in the past [11]. Reaction of 3,5-diaminobenzoic acid with two
equivalents of 2-hydroxybenzaldehyde in basic MeOH afforded a
4.2. Description of crystal structures
ꢀ
[Fe(L1)(MeOH)3] (1) crystallizes in the triclinic space group P1.
Data collection and structural parameters are found in Table 1. Se-
lected bond distances and angles are found in Table 2. The asym-
metric unit consists of a full molecule of complex 1, shown in
Fig. 2 The Fe centre is hexacoordinated in a axially elongated octa-
hedral fashion, with oxidation state Fe(II), as shown by close exam-
ination of the structural parameters and bond-valence sum
calculations [10]. One imino ligand is coordinated to the Fe(II)
ion taking up three equatorial positions, the Fe–N(1) bond distance
is 2.122 Å, Fe–O(1) is 2.003 Å and Fe–O(2) is 2.056 Å, where N(1) is
Table 2
Selected bond distances and angles for [Fe(L1)(MeOH)3] (1).
Distance (Å)
Angle (°)
Fe1–O1
Fe1–O2
Fe1–O6
Fe1–N1
Fe1–O5
Fe1–O4
N1–C1
2.0033(17)
2.0557(16)
2.0914(16)
2.1220(19)
2.1586(17)
2.3027(17)
1.290(3)
O1–Fe1–O2
O1–Fe1–O6
O2–Fe1–O6
O1–Fe1–N1
O2–Fe1–N1
O2–Fe1–N1
O6–Fe1–N1
O1–Fe1–O5
O2–Fe1–O5
O6–Fe1–O5
N1–Fe1–O5
O1–Fe1–O4
O2–Fe1–O4
O6–Fe1–O4
N1–Fe1–O4
O5–Fe1–O4
165.90(7)
93.18(7)
91.59(6)
87.68(7)
87.60(7)
87.60(7)
179.13(7)
94.58(7)
99.01(7)
85.68(6)
94.14(7)
88.86(6)
78.62(6)
82.43(6)
97.70(7)
167.80(6)

2208
R.H. Laye, E.C. Sañudo / Inorganica Chimica Acta 362 (2009) 2205–2212
Fig. 2. Crystal structure of complex 1, with atom labels. The hydrogen atoms have been omitted for clarity (top). Packing diagram for complex 1 along the a axis of the unit
cell. The intermolecular H-bonds are shown as dashed lines (bottom).
the nitrogen atom of the imino group, O(1) is one of the oxygen
atoms of the carboxylate and O(2) is the phenoxide oxygen. The li-
gand is not planar, in fact the O(1)–Fe–O(2) angle is 165.90°, where
O(1) and O(2) are the oxygen atoms of the ligand bound to the Fe
centre trans to each other. Occupying the fourth position of the
equatorial plane trans to the imino N–Fe bond is one MeOH ligand,
with Fe(1)–O(6) of 2.091 Å. The two axial positions are taken by
two MeOH ligands, with Fe–O bond distances of 2.159 Å to O(5)
and 2.303 Å to O(4), clearly elongated from the normal Fe(II) bond
distance to an oxygen atom. The monomers of 1 are organized into
linear chains by short hydrogen bond interactions along the a axis
of the unit cell, as shown in Fig. 2, with O(1)–O(5)0 contacts of
2.652 Å, O(3)–O(6)0 at 2.587 Å and O(2)–O(4)0 at 2.798 Å.
O(3)) and the nitrogen (N(1)) of the ligand, as shown in Fig. 2.
The Cu(II) ions in 2 are tetracoordinated, in a distorted square-pla-
nar fashion, in fact, the O(1)–Cu–O(3) angle is 130.40°. The fourth
coordination position is occupied by an oxygen atom (O(2)) from
the next [Cu(L1)] unit, linking the [Cu(L1)]n into a one-dimensional
zigzag chain along the b axis of the unit cell, as shown in Figs. 2 and
3.
[Et3NH]2[Cu2(L2)2] (3) crystallizes in the monoclinic space
group P21/n. Data collection and structural parameters can be
found in Table 1. Selected bond distances and angles can be found
Table 3
Selected bond distances and angles for [Cu(L1)]n ꢀ MeOH (2).
The crystal structure of complex 2 at 293 K has been reported
previously by Wang et al. [11], but its magnetic properties had
not been studied. New data have been collected at 100 K. Data col-
lection and structural parameters are found in Table 1. Selected
bond distances and angles are found in Table 3. Complex 2 crystal-
lizes in the monoclinic space group P21. The structure of 2 consists
of linear 1-D chains of the repeating asymmetric unit [Cu(L1)].
Each Cu(II) ion is coordinated to two oxygen atoms (O(1) and
Distance (Å)
Angle (°)
Cu1–O1
Cu1–O3
Cu1–N1
Cu1–O2(a)
N1–C1
1.890(2)
1.933(2)
1.933(2)
1.945(2)
1.285(4)
O1–Cu1–O3
O1–Cu1–N1
O3–Cu1–N1
O1–Cu1–O2(a)
O3–Cu1–O2(a)
N1–Cu1–O2(a)
151.58(10)
95.17(10)
93.23(10)
86.41(9)
93.38(9)
162.88(10)
(a) refers to atoms generated by a symmetry operation.

R.H. Laye, E.C. Sañudo / Inorganica Chimica Acta 362 (2009) 2205–2212
2209
Fig. 3. Crystal structure of the asymmetric unit of 2. The hydrogen atoms have been omitted for clarity (top). Packing diagram of 2, showing the zigzag chains of the repeating
unit along the b axis of the unit cell (bottom).
in Table 4. The asymmetric unit consists of one Cu(II) ion and one
ligand L2, along with two MeOH molecules and one protonated tri-
ethylamine. The full complex anion is shown in Fig. 4. The Cu(II)
ion is tetracoordinated to two N atoms N(1) and N(2) (Cu–N(1) dis-
tance is 1.974 Å and Cu–N(2) is 1.971 Å) and two oxygen atoms
from the alkoxide group of the di-imine ligand, O(1) and O(2), with
Cu–O distances of 1.896 Å and 1.904 Å, respectively. The coordina-
tion geometry of the Cu centre is very distorted from ideal square-
planar: the L–Cu–L angles of two atoms coordinated in cis positions
with respect to the Cu centre range from 91.75° to 100.22° (for an
ideal square-planar complex these angles would be 90°) and the L–
Cu–L angles for two atoms bound trans to each other are 147.46°
and 144.97°, respectively. For an ideal square-planar complex
these angles should be 180°, and the deviation from this value indi-
cates a distortion to a non-planar species. A tetrahedral Cu com-
plex would require all the L–Cu–L angles to be 109°, thus, the
Cu(II) in 3 is in a coordination environment between square-planar
and tetrahedral. The two ligands in complex 3 use the imino and
alkoxo N and O atoms to coordinate to the Cu(II) ions, but the car-
boxylato group of the central aromatic residue remains non-coor-
dinated, as can be clearly seen in Fig. 4. In the crystal packing,
the free carboxylate group is hydrogen bonding to one of the solva-
tion methanol molecules (the O(3)–O(1s) distance is 2.641 Å) and
to the protonated triethylamine cation (O(4)–N(3) distance is
2.648 Å). The triethylamine molecule of the asymmetric unit is
protonated, and the proton was found and its position refined.
The Cu–Cu distance in complex 3 is 7.437 Å. The two ligands are
Table 4
Selected bond distances and angles for [Et3NH]2[Cu2(L2)2] ꢀ 4MeOH (3).
Distance (Å)
Angle (°)
Cu1–O2
Cu1–O1
Cu1–N1
Cu1–N2
N1–C14
N2–C1
1.8962(11)
1.9034(11)
1.9710(13)
1.9742(13)
1.300(2)
O2–Cu1–O1
O2–Cu1–N1
O1–Cu1–N1
O2–Cu1–N2
O1–Cu1–N2
N1–Cu1–N2
91.75(5)
144.96(5)
93.27(5)
93.80(5)
147.47(5)
100.22(5)
p–p
stacking, with the two central aromatic groups at an average
1.300(2)
distance of 3.38 Å.

2210
R.H. Laye, E.C. Sañudo / Inorganica Chimica Acta 362 (2009) 2205–2212
Fig. 4. Crystal structure of the anion complex of 3. Hydrogen atoms have been removed for clarity.
4.3. Magnetic properties
g = 2.11, D = 0.01 cmꢁ1 and
H
= ꢁ4.63 K. The negative Curie–Weiss
constant indicates antiferromagnetic interactions between the
monomers of 1. This fitting is in disagreement with the magnetiza-
tion, measured at 2 K (Fig. 5). The magnetization tends to satura-
tion at a value of 2.2, which would be the expected for an
intermediate spin Fe(II) ion with S = 1. Spin crossover to the inter-
mediate S = 1 spin state could explain magnetization data, as well
as the sudden drop observed in the susceptibility curve. Hexacoor-
dinated imine complexes of iron have been claimed to show spin
crossover before [12].
Dc magnetic susceptibility data were collected for complex 1 at
an applied field of 0.5 T in the 300 to 2 K temperature range. The
v
MT value at 300 K was 3.40 cm3 K molꢁ1, in agreement with ex-
pected value of 3.0 cm3 K molꢁ1 for one Fe(II) ion with S = 2 and
g = 2.0. As temperature decreases, the MT value remains nearly
constant, until a drop to a
MT value of 1.1 cm3 K molꢁ1 is observed
v
v
at temperatures below 50 K. This drop can be due to zero-field
splitting of the Fe(II) ion, to intermolecular antiferromagnetic
interactions, or a partial spin crossover to an intermediate (S = 1)
or low spin (S = 0) state. In order to model the magnetic data, the
zero-field splitting of an S = 2 ion and the intermolecular interac-
tions (as a Curie–Weiss constant) are included in the Hamiltonian,
which is solved using the Van Vleck formula. The best fitting ob-
tained is shown in Fig. 5 as a solid line, the fitting parameters were
Magnetic susceptibility data for a sample of complex 2 were
collected in the 2–300 K temperature range with an applied dc
field of 1.0 T, shown in Fig. 6. The
vMT product per Cu(II) ion has
a value of 0.39 cm3 K molꢁ1 at 300 K, slightly larger than the ex-
0.8
1.2
4
1.0
0.7
0.8
3
2.5
0.6
0.6
2.0
0.4
1.5
2
0.2
0.5
0.0
1.0
0
10000
20000
30000
40000
50000
Field (Oe)
0.5
1
0.4
0.3
0.0
0
10000
20000
Field (Oe)
30000
40000
50000
0
0
50
100
150
200
250
300
0
50
100
150
200
250
300
T (K)
T (K)
Fig. 5.
vT vs. T plot for complex 1. The solid line is the best fit to the experimental
Fig. 6. vT vs. T plot for complex 2. The solid line is the best fit to the experimental
data (see text for fitting parameters). Inset: field dependence of the magnetization
data (see text for fitting parameters). Inset: field dependence of the magnetization
of complex 1 at 2 K, the solid line is the Brillouin function for S = 1 and g = 2.0.
of complex 2 at 2 K, the solid line is the Brillouin function for S = 1/2 and g = 2.0.

R.H. Laye, E.C. Sañudo / Inorganica Chimica Acta 362 (2009) 2205–2212
2211
pected 0.375 cm3 K molꢁ1 for an isolated Cu(II) ion with g = 2.0 and
S = 1/2. As temperature decreases, the MT product is practically
J = 0.66 cmꢁ1, and the calculated susceptibility is shown in Fig. 7
as a solid line. The calculated ferromagnetic constant results in
an S = 1 spin ground state for compound 3. As can be seen in
Fig. 4, the two imine donors on the ligand are meta with respect
to each other in the aromatic ring. Thus, the magnetic orbital of
v
constant, until below 30 K it starts to rise, up to a value of
0.64 cm3 K molꢁ1 at 2 K. This rise is not field dependent, and is
indicative of ferromagnetic interaction between the Cu(II) ions in
2. The experimental data were modelled using the equation de-
rived by Baker [13] for chains of S = 1/2 and only one exchange
constant between the spins. The best fit was obtained for g = 2.05
and J = 1.3 cmꢁ1, it is shown in Fig. 6 as a solid line. The field
dependence of the magnetization was studied for 2 at 2 K, a
M/Nb versus Field plot is shown in Fig. 6. The dots are the experi-
mental data, while the solid line is the calculated magnetization at
2 K using the Brillouin function for S = 1/2 and g = 2.0. As can be
clearly seen, the magnetization of 2 is at all times larger than the
calculated for an isolated S = 1/2 with g = 2.0, as expected due to
the ferromagnetic interactions between the Cu(II) ions in the chain.
The material does not order in the long range, this has been proved
by the lack of an out-of-phase ac signal, as well as the coincidence
of the measurements at low temperature at different fields. Exam-
ining the solid-state structure of compound 2, the very distorted
square-planar Cu(II) ions are bridged by syn,anti-carboxylate
groups from the ligand. In this bridging mode, relatively uncom-
mon for carboxylate ligands, there is poor overlap between the
orbitals on the bridge with those bearing the unpaired electrons
on each copper ion, and as predicted by Alemany et al., the
coupling should be weakly ferromagnetic, [14] as observed for
compound 2.
the Cu(II) ion will cause spin polarization of the appropriate
p-
symmetry orbitals on the imine nitrogen atom and due to the
meta-substitution on the aromatic ring, weak ferromagnetic cou-
pling results [11].
5. Conclusion
Imine ligands have been used to obtain metal-bearing building
blocks of Fe(II) and Cu(II). The complex [Fe(L1)(MeOH)3] (1), with
three labile MeOH ligands, is a promising material for the design
of Fe(II) metal–organic frameworks. The magnetic properties of
the coordination polymer [Cu(L1)]n (2) have been studied and it
has been found to display ferromagnetic coupling. We have de-
signed a ligand with three distinct metal-binding sites, 3,5-
bis[{(2-hydroxyphenyl)methylene}amino]benzoic acid (L2), the
three bonding sites are in the 1,3,5 positions of an aromatic ring,
thus facilitating ferromagnetic coupling between the metal ions
coordinated to the ligand. This is the case of the compound
[Et3NH]2[Cu2(L2)2] (3), which possesses a S = 1 spin ground state.
Additionally, compound 3 has a free and accessible carboxylate
arm and its properties as good building block for the preparation
of metal–organic frameworks are currently being investigated.
Magnetic susceptibility data for a crushed crystalline sample of
complex 3 were collected in the 30–300 K temperature range at an
Acknowledgements
applied field of 1.0 T and below that temperature at 500 G. The vT
value at 300 K is 0.80 cm3 K molꢁ1, in agreement with the expected
value for two Cu(II) ions (0.750 cm3 K molꢁ1, 2S = 1/2 and g = 2.0).
E.C.S. acknowledges the financial support from Spanish Govern-
ment, (Grant CTQ2006/03949BQU and Juan de la Cierva fellow-
ship) and the Unidade de Raios X (RIAIDT. University of Santiago
de Compostela, Spain) for the crystallography of complex 3.
As temperature decreases, the
vT value remains practically con-
stant, but below 30 K it starts to increase, up to a value of
0.91 cm3 K molꢁ1 at 2 K, indicating weak ferromagnetic interaction
between the Cu(II) ions of complex 3 (Fig. 7). The magnetization
versus field data supports this conclusion. At 2 K the magnetization
tends to a saturation value of 2.0, and follows the Brillouin law for
an S = 1 with g = 2.0, in agreement with a ferromagnetic interaction
between the two Cu(II) ions that form the anion complex 3. A Van
Vleck equation can be easily derived to model the experimental
susceptibility data. The fitting parameters were g = 2.09 and
Appendix A. Supplementary material
CCDC 681123, 681121 and 681122 contain the supplementary
crystallographic data for 1, 2 and 3. These data can be obtained free
of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif. Supplementary data associ-
ated with this article can be found, in the online version, at
doi:10.1016/j.ica.2008.09.052.
1.2
1.0
References
[1] L. Bogani, W. Wernsdorfer, Nat. Mater. 7 (2008) 179.
[2] (a) G. Aromí, E.K. Brechin, Struct. Bond. 122 (2006) 1;
(b) R. Sessoli, H.L. Tsai, A.R. Schake, S. Wang, J.B. Vincent, K. Folting, D.
Gatteschi, G. Christou, D.N. Hendrickson, J. Am. Chem. Soc. 115 (1993)
1804;
0.8
(c) E.C. Sañudo, W. Wernsdorfer, K.A. Abboud, G. Christou, Inorg. Chem. 43
(2004) 4137;
2.5
(d) E.K. Brechin, E.C. Sañudo, W. Wernsdorfer, C. Boskovic, J. Yoo, D.N.
Hendrickson, A. Yamaguchi, H. Ishimoto, T.E. Concolino, A.L. Rheingold, G.
Christou, Inorg. Chem. 44 (2005) 502.
0.6
2.0
1.5
[3] (a) M. Ohba, H. Okawa, Coord. Chem. Rev. 198 (2000) 313;
(b) J.R. Li, Q. Yu, E.C. Sañudo, Y. Tao, W.C. Song, X.H. Bu, Chem. Mater. 20 (2008)
1218;
0.4
1.0
0.5
(c) S. Neogi, E.C. Sañudo, P.K. Bharadwaj, Eur. J. Inorg. Chem. (2007) 5426.
[4] (a) A. Escuer, G. Aromí, Eur. J. Inorg. Chem. 23 (2006) 4721;
(b) F.C. Liu, Y.F. Zeng, J.P. Zhao, B.W. Hu, E.C. Sañudo, J. Ribas, X.H. Bu, Inorg.
Chem. 46 (2007) 7698.
[5] G.S. Papaefstathiou, S.P. Perlepes, A. Escuer, R. Vicente, M. Font-Bardia, X.
Solans, Angew. Chem., Int. Ed. 40 (2001) 884.
[6] R.E.P. Winpenny, J. Chem. Soc., Dalton Trans. (2002) 1.
[7] E.C. Sañudo, A.A. Smith, P.V. Mason, M. Helliwell, G. Aromí, R.E.P. Winpenny,
Dalton Trans. (2006) 1981.
0.2
0.0
0
10000 20000 30000 40000 50000
Field (Oe)
0.0
0
50
100
150
200
250
300
T (K)
Fig. 7.
vT vs. T plot for complex 3. The solid line is the best fit to the experimental
[8] Some examples can be found here: (a) H. Oshio, N. Hoshino, T. Ito, J. Am. Chem.
Soc. 122 (2000) 12602;
data (see text for fitting parameters). Inset: field dependence of the magnetization
of complex 3 at 2 K, the solid line is the Brillouin function for S = 1 and g = 2.0.

2212
R.H. Laye, E.C. Sañudo / Inorganica Chimica Acta 362 (2009) 2205–2212
(b) C. Boskovic, H.U. Güdel, G. Labat, A. Neels, W. Wernsdorfer, B. Moubaraki,
K.S. Murray, Inorg. Chem. 44 (2005) 3181.
[9] A.R. Paital, T. Mitra, D. Ray, W. Tak Wong, J. Ribas-Ariño, J.J. Novoa, J. Ribas, G.
Aromı, Chem. Commun. (2005) 5172.
[10] I.D. Brown, D. Altermat, Acta Crystallogr., Sect. B 41 (1985) 244.
[11] J. Wang, F.L. Bei, X.J. Yang, L.D. Lu, X. Wang, J. Mol. Struct. 643 (2002) 129.
[12] (a) G. Lemercier, N. Bréfuel, S. Shova, J.A. Wolny, F. Dahan, M. Verelst, H.
Paulsen, A.X. Trautwein, J.P. Tuchagues, Chem. Eur. J. 12 (2006) 7421;
(b) M.A. Halcrow, Chem. Soc. Rev. 37 (2008) 278.
[13] G.A. Baker, G.S. Rushbrooke, Phys. Rev. 135 (1964) 1272.
[14] A. Rodríguez-Fortea, P. Alemany, S. Alvarez, E. Ruiz, Chem. Eur. J. 7 (2001)
627.