Synthesis, structural and magnetic properties of a series of copper(ii) complexes containing a monocarboxylated perchlorotriphenylmethyl radical as a coordinating open-shell ligand

Article abstract of DOI:10.1039/b316458f

A series of complexes of copper(n)-containing a perchlorotriphenylmethyl radical functionalized with a carboxylic group as a new ligand is reported. The compounds [Cu(PTMMC)₂(L)₃] (PTMMC = (tetradecachloro-4- carboxy-triphenyl)methyl radical; L = (1) H₂O, (2) pyrimidine and ethanol or (3) pyridine), [Cu₂(PTMMC)₂(MeCOO) ₂(H₂O)₂] (4) and [Cu(HPTMMC) ₂(L)₃] (HPTMMC = α-H-(tetradecachlorotriphenyl) methane-4-carboxylic acid; L = pyridine) (5) were structurally characterized. In complexes 1, 2, 3 and 5, the copper(II) ion is coordinated to two PTMMC (or HPTMMC) units in a slightly distorted square planar surrounding, while 4 shows a paddle-wheel copper(II) dimer structure, where each Cu metal ion has four O atoms of different carboxylate groups, two of them belonging to two PTMMC radicals. The copper(II)-radical exchange couplings are antiferromagnetic for complexes 1, 2 and 3. A linear three-spin model was applied to complexes 1,2 and 3 to give J/k_B = -24.9, -15.0 and -20.7 K, respectively. Magnetic properties of 4 show that it is one of the scarce examples of a spin-frustrated system composed of organic radicals and metal ions. In this case, experimental data were fitted to a magnetic model based on a symmetrical butterfly arrangement to give a copper(II)-copper(II) exchange coupling of J/k_B = -350.0 K and a copper(II)-radical exchange coupling of J/k_B = -21.3 K, similar to that observed for the copper(II)-radical interactions in complexes 1, 2 and 3.

Full text of DOI:10.1039/b316458f

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis, structural and magnetic properties of a series of
copper(II) complexes containing a monocarboxylated
perchlorotriphenylmethyl radical as a coordinating
open-shell ligand
D. Maspoch,a D. Ruiz-Molina,a K. Wurst,b J. Vidal-Gancedo,a C. Roviraa and J. Veciana*a
a Institut de Ciència de Materials de Barcelona (CSIC), Campus Universitari de Bellaterra,
08193 Cerdanyola, Spain
b Institut für Allgemeine Anorganische und Theoretische Chemie, Universität Innsbruck, A-6020,
Innrain 52a, Austria
Received 16th December 2003, Accepted 24th February 2004
First published as an Advance Article on the web 8th March 2004
A series of complexes of copper()-containing a perchlorotriphenylmethyl radical functionalized with a carboxylic
group as a new ligand is reported. The compounds [Cu(PTMMC)2(L)3] (PTMMC = (tetradecachloro-4-carboxy-
triphenyl)methyl radical; L = (1) H2O, (2) pyrimidine and ethanol or (3) pyridine), [Cu2(PTMMC)2(MeCOO)2(H2O)2]
(4) and [Cu(HPTMMC)2(L)3] (HPTMMC = α-H-(tetradecachlorotriphenyl)methane-4-carboxylic acid; L = pyridine)
(5) were structurally characterized. In complexes 1, 2, 3 and 5, the copper() ion is coordinated to two PTMMC (or
HPTMMC) units in a slightly distorted square planar surrounding, while 4 shows a paddle-wheel copper() dimer
structure, where each Cu metal ion has four O atoms of different carboxylate groups, two of them belonging to two
PTMMC radicals. The copper()–radical exchange couplings are antiferromagnetic for complexes 1, 2 and 3. A
linear three-spin model was applied to complexes 1, 2 and 3 to give J/kB = Ϫ24.9, Ϫ15.0 and Ϫ20.7 K, respectively.
Magnetic properties of 4 show that it is one of the scarce examples of a spin-frustrated system composed of organic
radicals and metal ions. In this case, experimental data were fitted to a magnetic model based on a symmetrical
butterfly arrangement to give a copper()–copper() exchange coupling of J/kB = Ϫ350.0 K and a copper()–radical
exchange coupling of J/kB = Ϫ21.3 K, similar to that observed for the copper()–radical interactions in complexes 1,
2 and 3.
chemical stability and their stereochemical characteristics,
with a helical (chiral) surrounding of the radical center by
Introduction
The design and synthesis of molecular-based magnetic
materials is a major focus of molecular materials research. One
of the more promising strategies is the so-called metal–radical
approach that combines paramagnetic metal ions and pure
organic radicals as ligating sites.1 The electronic open-shell
character of such organic ligands is particularly appealing
since they are expected to interact with transition metal ions
enhancing the strength of magnetic interactions and increasing
the magnetic dimensionality of the molecular material in
comparison with systems made up from paramagnetic metal
ions and diamagnetic coordinating ligands. However, even
though a large number of metal–radical systems have been
studied, the variety of radical-based ligands used up to know is
fairly limited. Among them, one of the most extensively used
families is that of nitroxide-based radicals, namely nitroxides
and α-nitronyl or imino nitroxide radicals.2–5 Besides nitroxide-
based radicals, other examples of open-shell ligands used
up to now are: (1) derivatives of the verdazyl radical family,
which turned out to be very attractive because of their abund-
ance of donor atoms and efficiency in transmitting magnetic
interactions;6 (2) o-quinone ligands that can be found in
different accessible oxidation states, including the radical
semiquinones,7 (3) TCNE and TCNQ radical anions8 and (4)
diphenylcarbenes substituted with chemical functionalities able
to coordinate with metal ions.9
very bulky substituents. Up to now, it has been demonstrated
that the supramolecular self-assembly of PTMMC in the
solid state forms hydrogen-bonded dimeric motifs that promote
the presence of ferromagnetic interactions.10 Herein, we present
the controlled synthesis, crystal structure analysis, spectro-
scopic characterization and magnetic properties of a new
family of Cu() complexes using the perchlorinated tri-
phenylmethyl radical PTMMC as a new ligand.11 More
specifically, we report a monomeric Cu() complex [Cu-
(PTMMC)2(H2O)3]ؒ6H2Oؒ2EtOH, 1, which in its magnetic
properties shows a relatively strong exchange coupling con-
stant between the metal ion and the radical ligand. This result
confirms that carboxylic-based PTM radicals could act as
excellent ligands for the design of new magnetic complexes
following the metal–radical approach. With the simple
modification of the auxiliary ligands of 1, two new monomeric
complexes, [Cu(PTMMC)2(pyrimidine)2(EtOH)]ؒTHFؒ4EtOH,
2, and [Cu(PTMMC)2(py)3]ؒ2.5n-hexane, 3, have been syn-
thesized. Structurally, the bulkiness of these ligands has
significant implications for their microporous crystal packing.
Furthermore, the simple modification of the stoichiometry
of the initial reaction for obtaining 1 gives a paddle-wheel
Cu() complex [Cu2(PTMMC)2(MeCOO)2(H2O)2]ؒ4EtOH, 4.
Its magnetic behavior can be understood as a spin-frustrated
system. Finally, the possibility to work with the non-radical
analogues of PTMMC, the hydrogenated form HPTMMC,
allows us to synthesize the non-radical Cu() complex
[Cu(HPTMMC)2(py)3]ؒ4THF, 5, for comparison purposes. Its
condition of isostructurality with respect on 3 highlights the
magnetic interactions between radical PTMMC and Cu()
ions.
Following this metal–radical approach, our group has
recently initiated a scientific project focussed in the use of
perchlorinated triphenylmethyl (PTM) radicals functionalized
with one carboxylic group, PTMMC, as new ligating sites to
build magnetic molecular materials. Among others, the main
advantages of these radicals are their astonishing thermal and
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
D a l t o n T r a n s . , 2 0 0 4 , 1 0 7 3 – 1 0 8 2
1073

View Article Online
Scheme 1 Schematic synthesis of PTMMC-based complexes. Reagents and conditions: (i) PTMMC/Cu2(O2CMe)4ؒH2O (4 : 1) in EtOH–H2O; (ii)
PTMMC/Cu2(O2CMe)4ؒH2O (2 : 1) in EtOH–H2O; (iii) 1/excess of pyrimidine (pyrim) in EtOH–n-hexane and THF; (iv) 1/excess of pyridine (py) in
EtOH–n-hexane and THF.
prompted us to control the stoichiometry of the metal–radical
complex by modification of the molar ratio of PTMMC in
its reaction with Cu2(O2CMe)4ؒ2H2O since this may lead to
the formation of a mixed dimeric complex with PTMMC and
acetate groups. Indeed, slow addition of PTMMC radical to an
excess of Cu2(O2CMe)4ؒ2H2O in EtOH–H2O at room temper-
ature yielded a crystalline sample of complex 4. Finally, crystals
of 5 were obtained in a similar synthetic methodology as for the
monomeric PTM-based complexes. Thus, an n-hexane solution
of pyridine was diffused into a THF solution of the solid
isolated by filtration from the reaction of copper acetate
Chart 1
monohydrate with an ethanolic solution of HPTMMC.
Results and discussion
B. Spectroscopic properties
A. Synthesis
The IR spectra of 1–4 exhibit the typical vibrational modes
Complex 1 was synthesized in an initial reaction of copper
acetate monohydrate and PTMMC in a solution of ethanol–
water in high yields (89%). High-quality crystals of 1 suitable
for single crystal X-ray diffraction analysis, were obtained by
slow evaporation of the reaction solvent. Afterwards, slow dif-
fusion of an excess of pyrimidine (or pyridine) in a solution of
ethanol–n-hexane into a solution of 2 in THF allowed the
preparation of complexes 2 and 3. In both complexes the
auxiliary water ligands of 1 were replaced by pyrimidine and
ethanol in 2 and by pyridine in 3 (Scheme 1). Crystals suitable
for single crystal X-ray diffraction were isolated directly
from the diffusion method. Surprisingly, the crystal structure
analysis showed that 1–3 do not adopt the paddle-wheel motif
with four bidentate carboxylate ligands joining two Cu() ions,
characteristic of most of copper acetate clusters. The tendency
to form mononuclear Cu() clusters instead of binuclear ones is
attributed to the extreme steric demand of the molecules of
the PTMMC radical.12 Indeed, due to the great steric hindrance
of the chlorine atoms located at the ortho positions of the
carboxylate group, the rotation angles of the polychlorinated
phenyl rings relative to the COOϪ moieties in 1–3 are in the
range of 79 to 89Њ. Such a structural arrangement is expected to
disrupt the formation of a binuclear complex with a paddle-
wheel disposition of the four molecules of PTMMC around the
two Cu() units, simply due to the direct confrontation of bulky
chlorine atoms of neighboring (syn) radicals. This result
attributed to PTM radicals. Indeed, in such PTM-based
complexes, the benzenoic peaks appear at 1509 and 1380–1300
cmϪ1, due to the highly chlorinated character of PTMMC units.
Furthermore, the typical peak corresponding to the PTM
moieties is found around 730 cm .
Ϫ1 13 More information of these
complexes can be extracted from the stretching peaks produced
by the carboxylate groups. Thus, a detailed analysis of the IR
spectra of 1–3 indicates the presence of the antisymmetric COO
stretching frequency around 1600–1620 cmϪ1, while the sym-
metric COO stretching frequency appears at 1410–1400 cmϪ1
.
Thus, a relatively large splitting of the COO stretching
frequencies (200–225 cmϪ1) is observed, being an indication of
monodentate coordination.14 By contrast, the IR spectrum
of the paddle-wheel complex 4 shows a lower value of this
splitting since the antisymmetric COO stretching frequency
appears at 1607 cmϪ1 and the symmetric one at 1417 cmϪ1
.
The electronic absorption spectra recorded in the UV-visible
region in pure THF also exhibits the absorptions corresponding
to PTM radicals. Thus, the electronic absorption spectrum of
1–3 show the typical four bands, denoted as A, B, C and D,
around 222 (ε ∼160000), 290 (ε ∼17000), 385 (ε ∼65000) and 500
(ε∼2500 MϪ1 cmϪ1) nm (Fig. 1), respectively.15 Among them, C
and D bands are associated with the radical character of
PTMMC ligand, while the other two are the primary and
secondary (1Lb) bands characteristic of substituted benzene
units of such a ligand. No additional bands, that could be
D a l t o n T r a n s . , 2 0 0 4 , 1 0 7 3 – 1 0 8 2
1074

View Article Online
PTMMC ligands, where most of the spin density is localized.
The two central carbon atoms of PTMMC ligands are
separated by 16.56 Å.
The self-assembly of metal–radical clusters in 1, probably
steered by chlorine–chlorine contacts, generates large micro-
pores along the [010] direction. Thus, each metal–radical cluster
interacts through 16 different chlorine–chlorine contacts
(3.60 and 3.25 Å) with its neighbouring clusters, building a
three-dimensional network of such supramolecular inter-
actions. Two facing metal–radical clusters create a rectangular
synthon, which is partially divided in two nearly square
micropores, which are separated by the theoretical wall formed
by the water ligands O5 of both complexes (Fig. 4). The Cu–Cu
distance in this configuration is 9.33 Å and the dimensions
of the micropores are 8.4 and 8.1 Å, with an effective size of
5.1 × 4.6 Å, when van der Waals radii are considered. The
cavities are filled with water and ethanol molecules, with all of
them bonded through hydrogen bonds between themselves and
between water ligands and the PTMMC carboxylate groups.
The void volume of the cavities is 2028 Å3, which equals 28%
of the unit cell volume.16
Fig. 1 Electronic absorption spectra of the PTMMC radical (dashed
line) and complex 3 (continuous line).
assigned to a radical-to-metal charge-transfer phenomenon,
were observed.
(ii) [Cu(PTMMC)2(pyrimidine)2(EtOH)]ؒTHFؒ4EtOH, 2. In
As occurs in the above-described radical-based complexes,
the non-radical complex 5 shows identical IR and electronic
spectrum as its precursor HPTMMC. Thus, the characteristic
¯
complex 2, molecules crystallize in the P1 triclinic space group
with the cell parameters given in Table 1. An ORTEP drawing
of 2 is shown in Fig. 3. The coordination geometry of metal–
radical clusters around the Cu() ion is distorted square
pyramidal in which the Cu() is coordinated to the two oxygen
atoms of two PTMMC, two nitrogen atoms of pyrimidine
molecules and one oxygen atom of an ethanol molecule. As
occurs in 1, two molecules of PTMMC are coordinated to
one Cu() ion through one oxygen atom of their carboxylate
group, which are twisted by angles of 80 and 81Њ with respect
to the phenyl plane to which they are bonded. The distances
Cu–O1 (carboxylate) and Cu–O3 (carboxylate) are 2.068(8)
and 1.926(6) Å and the O1–Cu–O3 angle is 169.7(3)Њ.
Moreover, the distances between the Cu() ion and the central
carbon of PTMMC are 8.48 and 8.33 Å and the distance
between the two central carbon atoms of bonded PTMMC
units are 16.72 Å.
peak of non-radical PTM derivatives appears at 804 cmϪ1 10
.
The electronic spectrum shows the presence of the primary
band characteristic of substituted benzenes around 223 nm
(ε ∼265000 MϪ1 cmϪ1).
C. X-Ray crystal structures
The molecular and crystal structures of 1–5 were investigated
by single crystal X-ray analysis. Crystallographic data and
experimental parameters for all complexes are summarized in
Table 1. Selected bond lengths and angles for metal complexes
are given in Table 2. Structural disorder analysis of complexes
1–3 and 5 is described in the Experimental Section.
(i) [Cu(PTMMC)2(H2O)3]ؒ6H2Oؒ2EtOH, 1. Complex 1 crys-
tallizes in the P21/c monoclinic space group with the cell
parameters reported in Table 1. An ORTEP drawing for the
metal–radical cluster, formed by two PTMMC molecules and
one Cu() ion, is shown in Fig. 2. The Cu() ion displays a
square pyramidal coordination geometry by binding to the two
oxygen atoms of two distinct PTMMC ligands and the three
oxygen atoms of water molecules acting as auxiliary ligands.
The Cu–O1 (carboxylate) and Cu–O3 (carboxylate) distances
are 1.967(14) and 1.99(2) Å, respectively, so that the PTMMC
anion is unidentate with the copper() ion. Due to the great
steric hindrance of the chlorine atoms in ortho positions with
respect the carboxylate group in PTMMC, the latter groups are
twisted by angles of 87 and 89Њ with respect to the phenyl plane
to which they are bonded. Furthermore, the O1–Cu–O3 angle is
175.4(9)Њ, which leads to a linear system of the three electronic
open-shell units (S = 1/2) with distances of 8.43 and 8.34 Å
between the copper() ion and the central carbon of both
Fig. 3 ORTEP drawing of metal–radical clusters of 2 at the 50%
probability level.
Crystal packing of the metal–radical clusters of 2 also leads
to micropores stabilized through 16 chlorine–chlorine contacts
in the range 3.47–3.33 Å. In this case, four molecules of 2 form
repetitive rectangular micropores, where molecules of ethanol
are not confronted in comparison with water molecules of 1.
Due to the bulky nature of ethanolic and pyrimidine molecules,
compared with water molecules, 2 can not adopt the same top-
ology as 1. Indeed, while the PTMMC moieties are confronted
in 1, these units undergo a displacement along the [0Ϫ11] direc-
tion in 2. To illustrate this displacement, the Cu–Cu distance
between opposite walls, which forms such micropores, is
15.33 Å. The differences of the self-assembly of 1 and 2 are
shown in Fig. 4. This new arrangement creates rectangular
micropores with dimensions of 10.7 and 5.9 Å, with an effective
size of 7.3 × 3.0 Å. Furthermore, such micropores are stabilized
through additional π–π interactions between the chlorinated
Fig. 2 ORTEP drawing of metal–radical clusters of 1 at the 50%
probability level.
D a l t o n T r a n s . , 2 0 0 4 , 1 0 7 3 – 1 0 8 2
1075

D
Table 1 Crystal data and structure refinement for complexes 1–5.
1
2
3
4
5
Empirical formula
Formula weight
T /K
λ(Mo-Kα)/Å
Crystal system
Space group
a/Å
b/Å
c/Å
α/Њ
β/Њ
CuC44Cl28O15H30
1854.82
223(2)
0.71069
Monoclinic
P21/c
25.644(4)
15.118(3)
18.580(3)
90
92.18(1)
90
CuC62Cl28O10N4H46
2063.17
293(2)
CuC70Cl28O4N3H50
2053.27
223(2)
Cu2C52Cl28O14H34
2002.47
223(2)
0.71069
Monoclinic
P21/c
26.296(2)
16.9040(9)
8.5648(6)
90.00
93.760(3)
90.00
CuC71Cl28O8N3H49
2128.27
293(2)
0.71069
0.71069
0.71069
Triclinic
Triclinic
Triclinic
¯
¯
¯
P1
P1
P1
16.2920(1)
16.5722(8)
17.9520(1)
97.986(4)
102.953(4)
115.831(4)
4094.1(4)
2
8.774(1)
14.332(1)
17.379(2)
84.411(6)
79.880(4)
87.933(6)
2140.8(4)
1
8.8122(3)
14.3777(6)
17.4037(7)
93.852(2)
100.012(2)
90.004(2)
2166.4(2)
1
γ/Њ
V/Å3
7198.(2)
4
1.712
3798.9(4)
2
1.751
Z
Dc/g cmϪ3
1.674
1.236
1.593
1.177
1.631
1.169
µ/mmϪ1
1.399
1.601
F(000)
3676
2062
1028
1984
1065
Crystal size/mm
θ Range/Њ
hkl Range
Reflections collected
Independent reflections
Parameters
GOF
0.30 × 0.25 × 0.03
1.56–19.00
Ϫ24/0, Ϫ14/13, Ϫ16/17
9744
4617
754
1.049
7.07
16.78
561, Ϫ350
0.31 × 0.24 × 0.11
1.21–22.00
0/17, Ϫ17/15, Ϫ18/18
17838
9965
894
1.049
6.68
18.96
921, Ϫ637
0.29 × 0.16 × 0.14
1.20–22.00
0/9, Ϫ15/15, Ϫ18/19
8337
4972
503
1.044
4.84
11.54
321, Ϫ452
0.20 × 0.15 × 0.02
1.43–19.75
0/24, Ϫ16/16, Ϫ8/8
12387
3413
444
1.218
5.95
11.47
298, Ϫ325
0.30 × 0.15 × 0.09
1.19–22.49
0/9, Ϫ15/15, Ϫ18/18
10847
5663
506
1.104
7.72
21.36
517, Ϫ542
Ra (I > 2σ(I ))
Rw b (I > 2σ(I )
Largest diff. peak/e nmϪ3
a R = Σ|Fo| Ϫ |Fc||/Σ|Fo|. b Rw = [(Σw(|Fo| Ϫ |Fc|)2/ΣwFo )]1/2
.
2

View Article Online
Table 2 Selected bond lengths (Å) and angles (Њ) for complexes 1–5.
1
2
3
Cu(1)–O(1)
Cu(1)–O(3)
Cu(1)–O(5)
Cu(1)–O(6)
Cu(1)–O(7)
1.967(14)
1.99(2)
2.16(12)
2.05(7)
Cu(1)–O(1)
Cu(1)–O(3)
Cu(1)–N(1)
Cu(1)–N(3)
Cu(1)–O(5)
2.068(8)
1.926(6)
2.080(11)
1.981(8)
2.236(15)
Cu(1)–O(1)
Cu(1)–O(2a)
Cu(1)–N(1)
Cu(1)–N(2)
Cu(1)–N(1a)
1.976(5)
2.147(5)
2.184(10)
2.334(11)
2.087(9)
2.27(9)
O(1)–Cu(1)–O(3)
O(1)–Cu(1)–O(5)
O(3)–Cu(1)–O(5)
O(1)–Cu(1)–O(6)
O(3)–Cu(1)–O(6)
O(5)–Cu(1)–O(6)
O(1)–Cu(1)–O(7)
O(3)–Cu(1)–O(7)
O(5)–Cu(1)–O(7)
O(6)–Cu(1)–O(7)
175.4(9)
92.3(8)
92.2(8)
88.9(7)
89.8(7)
95.1(12)
89.2(6)
91.1(7)
97.7(11)
167.1(11)
O(1)–Cu(1)–O(3)
O(1)–Cu(1)–N(1)
O(3)–Cu(1)–N(1)
O(1)–Cu(1)–N(3)
O(3)–Cu(1)–N(3)
N(1)–Cu(1)–N(3)
O(1)–Cu(1)–O(5)
O(3)–Cu(1)–O(5)
N(1)–Cu(1)–O(5)
N(3)–Cu(1)–O(5)
169.7(3)
88.8(3)
86.7(3)
93.4(3)
94.3(3)
157.9(5)
87.8(5)
83.7(5)
98.3(6)
103.8(6)
O(1)–Cu(1)–O(2a)
O(1)–Cu(1)–N(1)
O(2a)–Cu(1)–N(1)
O(1)–Cu(1)–N(2)
O(2a)–Cu(1)–N(2)
N(1)–Cu(1)–N(2)
O(1)–Cu(1)–N(1a)
O(2a)–Cu(1)–N(1a)
N(1)–Cu(1)–N(1a)
N(2)–Cu(1)–N(1a)
158.2(2)
89.1(2)
91.1(2)
80.0(3)
79.8(3)
112.7(5)
97.0(2)
97.0(3)
139.4(2)
107.9(5)
4
5
Cu(1)–O(1)
Cu(1)–O(2)
Cu(1)–O(3)
Cu(1)–O(4)
Cu(1)–O(5)
1.971(7)
1.964(7)
1.954(7)
1.950(7)
2.102(8)
Cu(1)–O(1)
Cu(1)–O(2a)
Cu(1)–N(1)
Cu(1)–N(2)
Cu(1)–N(1a)
2.288(11)
1.978(7)
2.110(17)
2.340(16)
2.070(12)
O(1)–Cu(1)–O(2)
O(1)–Cu(1)–O(3)
O(1)–Cu(1)–O(4)
O(1)–Cu(1)–O(5)
O(2)–Cu(1)–O(3)
O(2)–Cu(1)–O(4)
O(2)–Cu(1)–O(5)
O(3)–Cu(1)–O(4)
O(3)–Cu(1)–O(5)
O(4)–Cu(1)–O(5)
168.1(3)
90.1(3)
88.2(3)
97.2(4)
91.5(3)
87.7(3)
94.3(4)
167.6(3)
98.2(3)
94.2(3)
O(1)–Cu(1)–O(2a)
O(1)–Cu(1)–N(1)
O(2a)–Cu(1)–N(1)
O(1)–Cu(1)–N(2)
O(2a)–Cu(1)–N(2)
N(1)–Cu(1)–N(2)
O(1)–Cu(1)–N(1a)
O(2a)–Cu(1)–N(1a)
N(1)–Cu(1)–N(1a)
N(2)–Cu(1)–N(1a)
152.4(4)
98.4(4)
93.9(4)
69.7(5)
83.1(5)
107.7(7)
92.1(4)
90.4(2)
147.9(2)
104.4(8)
(iii) [Cu(PTMMC)2(py)3]ؒ2.5hexane,
3
and [Cu(HPTM-
MC)2(py)3]ؒ4THF, 5. The molecular structures of 3 and 5 are
similar to that of 2, but with three pyridine molecules instead of
the two pyrimidine molecules and one molecule of ethanol.
ORTEP drawings of metal–radical clusters present in both
structures are shown in Fig. 5. As for 2, the Cu() ion displays a
distorted square pyramidal coordination geometry by binding
to the two oxygen atoms of two molecules of PTMMC (or
HPTMMC) and the three nitrogen atoms of pyridine mole-
Fig. 4 Crystal packing of 1 (top) and 2 (bottom). Water and ethanol
for 1 and THF and ethanol solvent molecules for 2 have been omitted
for clarity.
rings, which form their walls. The cavities are filled with ethanol
and THF molecules. The void volume of the cavities is 1142 Å3,
which equals 28% of the unit cell volume.16
Fig. 5 ORTEP drawing of metal–radical clusters of 3 (top) and 5
(bottom) at the 30% probability level.
D a l t o n T r a n s . , 2 0 0 4 , 1 0 7 3 – 1 0 8 2
1077

View Article Online
Table 3 Details of the hydrogen bonding distances (Å) and angles (Њ)
of complex 4
cules. The carboxylate groups of PTMMC and HPTMMC
moieties are twisted by angles of 79Њ in 3 and 86Њ in 5 with
respect to the phenyl plane to which they are bonded. In
complex 3 the distance between the Cu() ion and the central
carbon atom of the PTMMC moieties are 8.47 and 8.33 Å and
the distance between the two units of PTMMC are 16.72 Å,
values that are close to those found for complex 2.
D–H ؒ ؒ ؒ A
H ؒ ؒ ؒ A
1.94
D ؒ ؒ ؒ A
2.73
D–H ؒ ؒ ؒ A
154
O(5)–H(5A) ؒ ؒ ؒ O(7)
O(5)–H(5B) ؒ ؒ ؒ O(6)
O(6)–H(6) ؒ ؒ ؒ O(4)
O(7)–H(7) ؒ ؒ ؒ O(4)
1.89
2.18
2.74
2.95
175
153
2.47
3.09
132
The self-assembly of metal–radical clusters of 3 and 5 is also
similar to that of 2, forming rectangular pores stabilized
through 12 and 16 chlorine–chlorine contacts, respectively, with
distances in the range 3.46–3.32 Å, occurring along the three
directions. Such channels are filled with n-hexane in 3 and THF
molecules in 5. The void volume of the cavities is 632 Å3, which
equals 30% of the unit cell volume.16
(Table 3). Such hydrogen bonds generate a primary hydrogen-
bonded cyclic motif between two molecules of ethanol and two
molecules (Fig. 7). In a parallel way, four of these cyclic
motifs are hydrogen-bonded to create a bigger cyclic pattern
composed of four ethanol solvent molecules and four molecules
of 4, which are extended along the ab plane generating a
two-dimensional hydrogen-bonded network (Fig. 7).
(iv) [Cu2(PTMMC)2(MeCOO)2(H2O)2]ؒ4EtOH, 4. Complex
4 crystallizes in the P21/c monoclinic space group with the cell
parameters reported in Table 1. The crystal structure of 4 shows
a paddle-wheel Cu() dimer structure, where each Cu metal ion
has four O atoms of different carboxylate groups in the
equatorial positions, with distances in the range of 1.950–1.971
Å, and a water molecule at 2.102 Å at the apex, completing the
square pyramidal coordination geometry (Fig. 6). Two of the O
atoms situated at the opposite sites in the base of the square
pyramid belong to two different PTMMC ligands, while the
other two belong to two distinct acetate groups. The end-to-end
distance of such metal–radical paddle-wheel supramolecules is
26 Å.
C. Electrochemical properties
Cyclic voltammograms of the radical PTMMC and complexes
1–3 were performed in THF at room temperature with
nBu4NPF6 (0.1 M) as supporting electrolyte (vs. Ag/AgCl) and
a Pt wire as a working electrode. While the radical PTMMC
showed only one reversible wave at ϩ0.09 V, assigned to the
one-electron reduction of the triphenylmethyl unit, all these
complexes revealed two one-electron reversible waves. One of
such waves corresponds to the reduction of the perchloro-
triphenylmethyl radical ligands (∼0 V) and the other to the
reduction of the Cu() ion (∼0.1 V) (Fig. 8). The reversible
waves were observed at Ϫ0.02 and 0.12 V for 1, Ϫ0.03 and 0.13
V for 2, and at Ϫ0.03 and 0.10 V for 3 (vs. Ag/AgCl).
D. Magnetic properties
The molar paramagnetic susceptibility data of crystalline
samples of 1–5 were obtained with a SQUID magneto/
susceptometer in the temperature range 2–300 K at a constant
field of 0.1 T using Pascal’s constants to calculate the
diamagnetic contributions. The results are displayed in the
form of the product of the molar paramagnetic susceptibility
(χmol) with temperature (T ) vs. the temperature. The principal
components of the g-tensors of PTM-based complexes
1–5 (g1, g2, g3) were determined by simulation of EPR spectra
of powdered, crystalline samples (see Experimental section),
and then the averaged g values [gav = (g1 ϩ g2 ϩ g3)/3] were used
in the theoretical calculation, as fixed values.
(i) [Cu(PTMMC)2(H2O)3]ؒ6H2Oؒ2EtOH, 1, [Cu(PTMMC)2
(pyrimidine)2(EtOH)]ؒTHFؒ4EtOH,
2 and [Cu(PTMMC)2-
(py)3]ؒ2.5n-hexane, 3. Similar magnetic behavior was observed
for complexes 1–3 (Fig. 9). A χT value of 1.12 emu K molϪ1 was
obtained at 300 K, which is close to the theoretical one (χT =
0.12505ؒgؒ2[3S(S ϩ 1)] = 0.12505ؒ2ؒ2[3ؒ½(½ ϩ 1)] = 1.13 emu
K molϪ1) expected for three isolated S = ½ spins in terms of the
spin-only equation. As the temperature was decreased, the
Fig. 6 ORTEP plot at the 50% probability level of metal–radical
supermolecule (top) and a view of the crystal packing of 4 along the c
axes (bottom). Ethanol solvent molecules have been omitted for clarity.
χT value also decreased until a value of 0.32 emu K molϪ1
,
corresponding to a S = 1/2 doublet fundamental state. The
resulting χT vs. T curves for 1–3 were analyzed quantitatively
on the basis of a linear three-spin model (Scheme 2), using
the following effective Hamiltonian, H = Ϫ2J(SR1SM ϩ SMSR2),
modified to take into account the presence of intermolecular
interactions (θ) in the molecular field approximation.17 This
model is in accordance with the X-ray molecular and crystal
structure analysis previously described. Eqn. (1) for a system
with SR1 = SR2 = SM = 1/2 was applied and fitted to the observed
χT vs. T plots by means of a least-square method.
The crystal packing is built up by the self-assembly of such
supramolecules in such a way as to form microchannels. A view
down in the c direction revels an arrangement of square micro-
channels stabilized by 12 chlorine–chlorine contacts (Fig. 6).
Thus, two confronted supramolecules along the c direction
create a rectangular synthon, which is partially divided in two
nearly square micropores, which are separated by a wall formed
by the water ligands of both molecules. The dimensions of
the micropores are 7.5 and 8.0 Å, with an effective size of
4.2 × 4.5 Å when van der Waals radius are considered. The void
volume is 917 Å3, which equals 24% of the unit cell volume.16
The cavities are filled with ethanol molecules, strongly bonded
through hydrogen bonds with the water and acetate ligands of 4
(1)
D a l t o n T r a n s . , 2 0 0 4 , 1 0 7 3 – 1 0 8 2
1078

View Article Online
Fig. 7 Hydrogen-bonded cyclic pattern formed by two molecules of 4 and two ethanol solvent molecules (left). Two-dimensional hydrogen-bonded
layer along the bc plane (right). Chlorine atoms (left) and chlorine and carbon atoms (right) have been omitted for clarity.
Fig. 8 Cyclic voltammograms of PTMMC radical (top) and 1
(bottom) in THF.
Fig. 9 Temperature dependence of the product of the magnetic
susceptibility and the temperature for complexes 3 (top) and 5
(bottom).
Scheme 2 Magnetic exchange interactions in the linear three-spin
model (a) and in the symmetrical four-spin butterfly model (b).
(ii) [Cu(HPTMMC)2(py)3]ؒ4THF, 5. Magnetic measurements
of a microcrystalline sample of 5 shows a quasi-ideal para-
magnetic behavior (Fig. 9). Thus, the χT value in the range of
300 to 10 K is 0.38 emu K molϪ1, which is close to the theoreti-
cal value expected for one isolated S = ½ spin. Upon further
cooling, χT gradually decreased to a value of 0.33 emu K molϪ1
at 2 K. The experimental χT vs. T curve for 5 was analyzed
quantitatively using the Curie–Weiss equation. From these
data, a molar Curie constant of 0.379 cm3 K molϪ1 and a Weiss
constant of Ϫ0.25 0.01 K were obtained. The latter value is
associated to the existence of very weak antiferromagnetic
interactions between the supramolecules.
Fixing the evaluated g values, the resulting parameters for the
best fits were J/kB = Ϫ24.9 0.3 K and θ = Ϫ0.68 0.05 K for 1,
J/kB = Ϫ15.0 0.1 K and θ = Ϫ0.45 0.02 K for 2, and J/kB =
Ϫ20.7 0.1 K and θ = Ϫ1.70 0.03 K for 3. Thus, the values of
the copper()–radical exchange coupling constants for all com-
plexes were very similar (J/kB ≈ Ϫ20 K) reflecting the presence
of relatively strong antiferromagnetic interactions between
the Cu() ion and the two coordinated radical PTMMC
molecules.
D a l t o n T r a n s . , 2 0 0 4 , 1 0 7 3 – 1 0 8 2
1079

View Article Online
(iii) [Cu2(PTMMC)2(MeCOO)2(H2O)2]ؒ4EtOH, 4. As can be
seen in Fig. 10, the χT value at 300 K is 0.9 emu K molϪ1, which
is still far from that expected for a system with four non-
interacting electrons (1.9 emu K molϪ1). Such a divergence is
consistent with the presence of strong antiferromagnetic inter-
actions among the two Cu() ions and between the Cu() ions
and the PTMMC coordinated radicals. Upon cooling χT
gradually decreases down to 70 K, where it reaches a plateau at
a value of 0.7 emu K molϪ1. Below 30 K, the χT value decreases
abruptly to a χT value of 0.4 emu K molϪ1 at 2 K, due to the
presence of weak intermolecular interactions.
Conclusions
In summary, it has been demonstrated that the carboxylic
substituted perchlorotriphenylmethyl radical is an excellent
coordinating ligand to obtain new metal complexes with differ-
ent magnetic properties following the metal–radical approach.
The reaction of the PTMMC radical with Cu() metal ion gave
a series of metal–radical complexes, showing a relatively strong
antiferromagnetic exchange coupling between the copper() ion
and the PTMMC radicals of J/kB ≈ Ϫ20 K. Furthermore, the
use of different molar ratios yield different structures for the
resulting metal–radical clusters. In particular, a metal–radical
paddle-wheel dimer exhibiting a spin frustration phenomenon
was obtained. Further studies to expand the range of structural
motifs, the structural dimensionality and magnetic behaviours
of metal–radical complexes based on polychlorinated triphenyl-
methyl radicals, either mono-, bi- or trifunctionalised with
carboxylic groups, are currently underway.
Experimental
Physical measurements
Microanalyses were performed at the Servei d’Analisi of the
Universitat Autònoma de Barcelona. The FT-IR spectra were
measured on a Perkin-Elmer Spectrum One spectrometer. The
UV-visible spectra were obtained on a VARIAN Cary 5
instrument. Electrochemical experiments were performed
with potentiostat Galvanostat 263a de EG i PAR, using a
platinum wire as working electrode and a Ag/AgCl electrode as
reference electrode. Anhydrous THF was freshly distilled over
sodium/benzophenone under nitrogen. Commercial tetrabutyl-
ammonium hexafluorophosphate (Fluka, electrochemical
grade) was used as the supporting electrolyte. The magnetic
susceptibility were measured on the bulk materials in the 2–300
K temperature range with a Quantum Design MPMS super-
conducting SQUID magnetometer operating at a field strength
of 0.1 T. The data were corrected for diamagnetism of the con-
stituent atoms using Pascal constants. The EPR spectra have
been recorded on X-band Bruker spectrometer (ESP-300E).
Fig. 10 Temperature dependence of the product of the magnetic
susceptibility and the temperature for complex 4.
The data was nicely fitted to a magnetic model based on a
symmetrical butterfly arrangement of the two metal ions and
two organic radicals (Scheme 2). The spin Hamiltonian, H,
for this butterfly system is given by eqn. (2)18 where J is the
wing-tip/body exchange interaction between the Cu() ions and
the organic radicals and J13 is the body/body interaction
between the two Cu() ions (Scheme 2).
ˆ
ˆ
ˆ
ˆ
ˆ
ˆ
ˆ
ˆ
ˆ
ˆ ˆ
H = Ϫ2J(S1ؒS2 ϩ S2ؒS3 ϩ S3ؒS4 ϩ S4ؒS1) Ϫ2J13S1ؒS3 (2)
The eigenvalues E deduced from eqn. (2), using the Kambe
Materials
method,19 are:
Solvents were distilled before use. In particular, THF was dried
over sodium/benzophenone, and distilled under argon. All the
reagents were used as received and they were purchased from
Aldrich and Panreac. Ligands 1 and 5 were synthesized accord-
ing to the procedure previously described.7b All reactions were
carried out in dark.
E = ϪJؒ[ST(ST ϩ 1) Ϫ S13(S13 ϩ 1) Ϫ
S24(S24 ϩ 1)] Ϫ J13[S13(S13 ϩ 1)] (3)
where S13 = S1 ϩ S3, S24 = S2 ϩ S4 and ST = S13 ϩ S24. A
theoretical expression for the χT vs. T behavior was derived
from the use of eqn. (3) and the van Vleck equation. The result-
ing expression, modified to take into account intermolecular
interactions (θ) in the molecular field approximation, was then
used to least-squares fit the experimental data of 4. Fixing the
evaluated g = 2.01, the best fit was obtained for J = Ϫ21.3 K,
J13 = Ϫ350.0 K and θ = Ϫ1.7 K. The value of the body/body
[copper()–copper()] interaction, J13 = Ϫ350 K, is within the
range of those previously reported for the same kind of inter-
action in other paddle-wheel Cu() dimers,20 whereas the value
Preparation of complexes
(i) [Cu(PTMMC)2(H2O)3]ؒ6H2Oؒ2EtOH, 1. A solution of
Cu2(O2CMe)4ؒ2H2O (0.032 g, 0.080 mmol) in 8 mL of water
was added dropwise to a solution of PTMMC (0.250 g, 0.320
mmol) in 14 mL of ethanol and stirred for 15 min at room
temperature. A red microcrystalline solid (0.246 g) was isolated
through filtering. The remaining solution was allowed to stand
in air for slow evaporation. After 5 days, 0.011 g of red plate
crystals of 1 were obtained, suitable for X-ray analysis. The
total yield of the reaction was 97%. Elemental analysis (%):
Calc. for C40H6O7Cl28Cu: C 29.04, H 0.36; Found: C 29. 41,
H 0.47. FT-IR (KBr, cmϪ1): 3393 (m), 1707 (w), 1607 (m), 1407
(m), 1335 (s), 1324 (s), 1260 (m), 1038 (w), 819 (w), 768 (w),
734 (m), 718 (m), 697 (w), 675 (w), 642 (m), 617 (w), 522 (m).
UV-Vis (THF, λ/nm, ε/MϪ1 cmϪ1): 223 (153800), 385 (62150),
508 (2412), 564 (2347). EPR (RT, powder crystalline sample):
g-factors: g1 = 2.040, g2 = 2.060, g3 = 2.090.
of the wing-tip/body [copper()–radical interaction], J13
=
Ϫ21.3 K, is similar to that previously reported for the Cu()–
PTMMC radical interaction in complexes 1–3. The fact that
both exchange coupling parameters, J and J13, are negative
indicates that any two neighboring spins tend to align anti-
ferromagnetically. However, the topological arrangement of
metal ions and organic radicals in the butterfly structure results
in a spin-frustration due to the presence of competing anti-
ferromagnetic interactions. The ground state of 4 is therefore
degenerated, with two different states that, in the format
(ST,S13,S24), are: the triplet state (1,0,1) and the singlet state
(0,0,0). This is one of the scarce examples of a spin-frustrated
system composed of organic radicals and metal ions.21
(ii) [Cu(PTMMC)2(pyrimidine)2(EtOH)]ؒTHFؒ4EtOH, 2. A
solution containing 15.7 µL of pyrimidine in 20 mL of ethanol
was layered onto a solution of 1 (0.100 g, 0.060 mmol) in 20 mL
D a l t o n T r a n s . , 2 0 0 4 , 1 0 7 3 – 1 0 8 2
1080

View Article Online
of THF. Slow diffusion over 19 days yielded 0.041 g (37%) of
red prism crystals of 2. Elemental analysis (%): Calc. for
C50H14O5Cl28N4Cu: C 33.24, H 0.77, N 3.10; Found: C 32.87,
H 0.47, N 2.73. FT-IR (KBr, cmϪ1): 2970 (w), 1737 (w), 1611
(m), 1594 (m), 1564 (w), 1506 (w), 1470 (w), 1334 (s), 1324 (s),
1259 (m), 1226 (w), 1173 (w), 1038 (w), 819 (w), 770 (w), 734
(w), 715 (m), 674 (w), 642 (m), 616 (w), 577 (w), 540 (w), 522
(m). UV-Vis (THF, λ/nm, ε/MϪ1 cmϪ1): 222 (176200), 385
(67340), 512 (1770), 565 (1802). EPR (RT, powder crystalline
sample): g-factors: g1 = 2.018, g2 = 2.032, g3 = 2.090.
structures were solved and refined using the SHELXTL
software.22
For 4, the refinement was run in a normal way, all non-
hydrogen atoms were refined with anisotropic displacement
parameters, all hydrogen atoms were calculated or found on
their positions and refined isotropically. In the refinement of the
other structures the hydrogen atoms of the water, ethanol,
hexane and THF molecules were omitted, because of poorer
crystal data owing to disorder problems.
Two kinds of disorder were observed around the environ-
ment of the well ordered PTMMC radicals, one for 1 and 2, the
other for 3 and 5. In 1, there is a 5 : 1 positional disorder of the
Cu(H2O)3-unit. The Cu(1) atom of the major part coordinates
to O1 and O3 (as shown in Fig. 2) and in the minor part the
Cu(1a) atom coordinates to O2 and O4 with an apparent
Cu(1)–Cu(1a) distance of 2.60 Å. The oxygen atoms of the
minor part and the solvent water and ethanol were refined with
isotropic displacement parameters, all other non-hydrogen
atoms were refined anisotropically. A similar disorder of 3 : 1
occurs for the Cu(pyrimidine)2(EtOH)-unit of 2 with an
apparent Cu(1)–Cu(1a) distance of 1.34 Å. This short distance
between the disordered Cu atoms leads to nearly overlying pyr-
imidine rings with highly distorted anisotropic displacement
parameters and separate ethanol groups in opposite directions.
The ethanol molecule of the minor part and the solvent ethanol
and THF were refined with isotropic displacement parameters,
all other non-hydrogen atoms were refined anisotropically. In 3
and 5 there is only a half molecule in the asymmetric unit and
the Cu(py)3-unit lies nearby a symmetry centre, which produces
a second Cu(py)3-unit in a 1 : 1 disorder with an apparent
Cu(1)–Cu(1a) distance of 1.48 Å for 3 and 1.15 Å for 5, respect-
ively. Similarly to 2, there are two nearly overlying pyridine
rings with distorted displacement parameters and one separate
pyridine ring for each Cu atom. In 3 all non-hydrogen atoms
were refined with anisotropic displacement parameters, except
the solvent hexane, which was refined isotropically. In 5 there
additionally exists a 1 : 1 positional disorder of the methane
group of the HPTMMC molecule. The solvent THF were
refined isotropically and also the carbon atoms of the separate
pyridine ring, because of a nearly overlying with a THF mole-
cule from the disordered part, all other atoms were refined with
anisotropic thermal parameters.
Interestingly, 2, 3 and 5 have nearly the same packing motifs
in the crystal lattice, although the cell volume of 2 is twice as
large. An examination of the intensities of 2 shows that all
reflections with h ϩ k = 2n ϩ 1 are systematically weak suggest-
ive of a C-centred unit cell, which is not defined in the triclinic
system. The relationship to 3 and 5 is, that the 1 : 1 disorder of
the Cu-units leads to a complete extinction of these weak reflec-
tions and formally to a C-centred unit cell. The transformation
matrix from the C-centered to the correct primitive P cell is
0.5 0.5 0; Ϫ0.5 0.5 0; 0.5 0.5 1 (row by row). If the unit cell of
2 is transformed in this manner, the cell constants are compar-
able to those of 3 and 5 with cell constants a = 8.729, b = 13.923,
c = 17.076 Å, α = 86.52, β = 81.22 and γ = 88.91Њ. This shows that
a simple 1 : 1 disorder from a small part of the molecule can
generate a new unit cell.
(iii) [Cu(PTMMC)2(py)3]ؒ2.5hexane, 3. A solution containing
3 mL of pyridine in 15 mL of hexane was layered onto a solu-
tion of 1 (0.095 g, 0.057 mmol) in 20 mL of THF. Slow
diffusion over 4 days yielded 0.040 g (39%) of red prism crystals
of 3. Elemental analysis (%) Calc. for C57H22O4Cl28N3Cu:
C 37.03, H 1.17, N 2.23; Found: C 36.93, H 1.85, N 1.70. FT-IR
(KBr, cmϪ1): 2951 (w), 2927 (w), 2859 (w), 1617 (m), 1610 (m),
1450 (m), 1397 (m), 1334 (s), 1324 (s), 1258 (m), 1073 (m), 1048
(w), 1037 (w), 917 (w), 876 (w), 820 (w), 812 (w), 769 (m), 733
(w), 718 (w), 694 (m), 674 (w), 645 (w), 617 (w), 579 (w), 537
(w), 522 (m). UV-Vis (THF, λ/nm, ε/MϪ1 cmϪ1): 222 (163800),
385 (67530), 512 (2615), 565 (2715). EPR (RT, powder crystal-
line sample): g-factors: g1 = 2.019, g2 = 2.028, g3 = 2.088.
(iv) [Cu2(PTMMC)2(MeCOO)2(H2O)2]ؒ4EtOH, 4. A solu-
tion of PTMMC (0.075 g, 0.098 mmol) in 40 mL of ethanol
was added dropwise to a solution of Cu2(O2CMe)4ؒ2H2O (0.051
g, 0.127 mmol) in 9 mL of water and 1 mL of ethanol during
2 h and stirred for 5 min at room temperature. The reaction
mixture was filtered and the remaining solution was allowed to
stand in air for slow evaporation. After 15 days, 0.009 g (14%)
of red prism crystals of complex 4 were obtained. Elemental
analysis (%): Calc. for C44H10O10Cl28Cu2: C 31.19, H 1.70;
Found: C 31.26, H 1.64. FT-IR (KBr): 1624 (m), 1607 (m), 1417
(m), 1335 (s), 1324 (s), 1259 (m), 1039 (w), 819 (w), 767 (w), 734
(m), 717 (W), 675 (w), 642 (m), 616 (w), 522 (m). EPR (110 K,
powder crystalline sample): g-factors: g1 = 2.008, g2 = 2.009,
g3 = 2.018.
(v) [Cu(HPTMMC)2(py)3]ؒ4THF, 5. A solution of Cu2-
(O2CMe)4ؒ2H2O (0.026 g, 0.130 mmol) in 7 mL of water was
added dropwise to a solution of 13 mL of ethanol and 0.5 mL
of ether of HPTMMC (0.200 g, 0.260 mmol) and stirred for
15 min at room temperature. A green solid was isolated through
filtration and 100 g of this was dissolved in 20 mL of THF. A
solution of 30 mL of hexane containing 3 mL of pyridine was
layered onto a solution of THF. Slow diffusion over 6 days
yielded 0.033 g of blue prism crystals of 6. Elemental analysis
(%): Calc. for C55H17O4Cl28N3Cu: C 35.90, H 0.92, N 2.28;
Found: C 36.46, H 1.13, N 1.82. FT-IR (KBr, cmϪ1): 2950 (w),
2926 (w), 1742 (w), 1628 (m), 1618 (m), 1609 (m), 1556 (w),
1534 (w), 1489 (w), 1469 (w), 1450 (m), 1438 (w), 1405 (m), 1371
(m), 1334 (s), 1314 (s), 1300 (s), 1240 (m), 1218 (w), 1152 (w),
1123 (w), 1048 (w), 912 (w), 872 (w), 804 (mw), 763 (m), 719
(w), 694 (m), 676 (m), 652 (w), 634 (w), 619 (w), 579 (w), 543
(w), 521 (m). UV-Vis (THF, λ/nm, ε/MϪ1 cmϪ1): 223 (264900).
EPR (RT, powder crystalline sample): g-factors: g1 = 2.052,
g2 = 2.070, g3 = 2.094.
CCDC reference numbers 186141 (1), 226899 (2), 226900 (3),
186142 (4) and 226898 (5).
See http://www.rsc.org/suppdata/dt/b3/b316458f/ for crystal-
lographic data in CIF or other electronic format.
X-Ray data collection and structure determination
X-Ray single-crystal diffraction data were collected on a Non-
ius KappaCCD diffractometer with graphite-monochromized
Mo-Kα radiation (λ = 0.7106 Å) and a nominal crystal to area
detector distance of 36 mm. Intensities were integrated using
DENZO and scaled with SCALEPACK. Several scans in ꢀ and
ω directions were made to increase the number of redundant
reflections, which were averaged in the refinement cycles. This
procedure replaces an empirical absorption correction. The
Acknowledgements
This work was supported by the Dirección General de
Investigación (Spain), under project MAT2003-04699 and the
3MD Network of the TMR program of the EU (contract
ERBFMRX CT980181).D. R.-M. thanks the Programa
Ramón y Cajal for financial support. D. M. is grateful to the
Generalitat de Catalunya for a predoctoral grant. D. M. is
D a l t o n T r a n s . , 2 0 0 4 , 1 0 7 3 – 1 0 8 2
1081

View Article Online
6 R. G. Hicks, M. T. Lemaire, L. K. Thompson and T. M. Barclay,
J. Am. Chem. Soc., 2000, 122, 8077.
7 (a) A. Caneschi, A. Dei, H. Lee, D. A. Shultz and L. Sorace, Inorg.
Chem., 2001, 40, 408; (b) D. Ruiz-Molina, J. Veciana, K. Wurst,
D. N. Hendrickson and C. Rovira, Inorg. Chem., 2000, 39, 617.
8 K. I. Pokhodnya, N. Petersen and J. S. Miller, Inorg. Chem., 2002, 41,
1996.
enrolled in the PhD program of the Universitat Autònoma de
Barcelona. We warmly thank Dr Carlos Gómez-García and
J. M. Martínez (Universitat de Valencia) for the magnetic
susceptibility measurements.
9 S. Karasawa, H. Kumada, N. Koga and H. Iwamura, J. Am. Chem.
Soc., 2001, 123, 9685.
10 (a) D. Maspoch, L. Catala, P. Gerbier, D. Ruiz-Molina,
J. Vidal-Ganzedo, K. Wurst, C. Rovira and J. Veciana, Chem. Eur. J.,
2002, 8, 3635; (b) M. Ballester, J. Castañer, J. Riera, A. Ibáñez and
J. Pujadas, J. Org. Chem., 1982, 47, 259.
11 A preliminary communication has already been reported; see:
D. Maspoch, D. Ruiz-Molina, K. Wurst, C. Rovira and J. Veciana,
Chem. Commun., 2002, 2958.
12 A similar tendency to form mononuclear complexes have already
been shown in transition metal complexes with highly sterically
hindered carboxylate ligands encapsulated by arene groups:
F. A. Chavez, L. Que and W. B. Tolman, Chem. Commun., 2001, 111.
13 M. Ballester, J. Riera, J. Castañer, C. Badia and J. M. Monsó,
J. Am. Chem. Soc., 1971, 93, 2215.
References
1 (a) A. Caneschi, D. Gatteschi and P. Rey, Prog. Inorg. Chem., 1991,
39, 331; (b) H. Iwamura, K. Inoue and T. Hayamizu, Pure Appl.
Chem., 1996, 68, 243.
2 For a review of nitroxide-based complexes, see: H. Iwamura and
K. Inoue, Magnetism: Molecules to Materials II, ed. J. S. Miller and
M. Drillon, Wiley-VCH, Weinheim, 2001.
3 For examples of complexes with nitroxide radicals, see: (a) L. M.
Field, P. M. Lahti and F. Palacio, Chem. Commun., 2002, 636;
(b) F. Mathevet and D. Luneau, J. Am. Chem. Soc., 2001, 123, 7465;
(c) P. Rabu, M. Drillon, H. Iwamura, G. Görlitz, T. Itoh,
K. Matsuda, N. Koga and K. Inoue, Eur. J. Inorg. Chem., 2000, 211;
(d ) H. Kumada, A. Sakane, N. Koga and H. Iwamura, J. Chem.
Soc., Dalton Trans., 2000, 911.
4 For examples of complexes with α-nitronyl nitroxide radicals, see:
(a) A. Caneschi, D. Gatteschi, N. Lalioti, C. Sangregorio, R. Sessoli,
G. Venturi, A. Vindigni, A. Rettori, M. G. Pini and M. A. Novak,
Angew. Chem., Int. Ed., 2001, 40, 1760; (b) M. Minguet, D. Luneau,
E. Lhotel, Vincent Villar, C. Paulsen, D. B. Amabilino and
J. Veciana, Angew. Chem., Int. Ed., 2002, 41, 586; (c) G. Stroh,
P. Turek, P. Rabu and R. Ziessel, Inorg. Chem., 2001, 21, 5334;
(d ) C. Rancurel, J.-P. Sutter, T. Le Hoerff, L. Ouahab and O. Kahn,
New. J. Chem., 1999, 1333; (e) M. G. F. Vaz, L. M. M. Pinehiro,
H. O. Stumpf, A. F. C. Alcantara, S. Goleen, L. Ouahab, O. Cador,
C. Mathonière and O. Kahn, Chem. Eur. J., 1999, 5, 1486; ( f )
K. Fey, D. Luneau, T. Ohm, C. Paulsen and P. Rey, Angew. Chem.,
1998, 110, 1331; K. Fey, D. Luneau, T. Ohm, C. Paulsen and P. Rey,
Angew. Chem., Int. Ed., 1998, 37, 1270.
5 For examples of complexes with imino nitroxide radicals, see:
(a) K. E. Vostrikova, D. Luneau, W. Wernsdorfer, P. Rey and
M. Verdaguer, J. Am. Chem. Soc., 2000, 122, 718; (b) H. Oshio and
T. Ito, Coord. Chem. Rev., 2000, 198, 329; (c) A. Marvilliers, Y. Pei,
J. Cano, K. E. Vostrikova, C. Paulsen, E. Rivière, J.-P. Audière and
T. Mallah, Chem. Commun., 1999, 1951; (d ) H. Oshio, T. Wanabe,
A. Ohto and T. Ito, Inorg. Chem., 1997, 36, 1608.
14 R. C. Mehrotra, Metal Carboxylates, Academic Press Inc.,
New York, 1983.
15 M. Ballester, J. Riera, J. Castañer, A. Rodríguez, C. Rovira and
J. Veciana, J. Org. Chem., 1982, 47, 4498.
16 A. L. Spek, A.M.C.T. PLATON, Utrecht University, Utrecht, The
Netherlands, 1998.
17 Y. Ishimaru, M. Kitano, H. Kumada, N. Koga and H. Iwamura,
Inorg. Chem., 1998, 37, 2273.
18 J. B. Vincent, C. Christmas, H.-R. Chang, Q. Li, P. D. Boyd,
J. C. Huffman, D. N. Hendrickson and G. Christou, J. Am. Chem.
Soc., 1989, 111, 2086.
19 K. Kambe, J. Phys. Soc., Jpn., 1950, 5, 48.
20 O. W. Steward, B. S. Johnston, S.-C. Chang, A. Harada, S. Ohba,
T. Tokii and M. Kato, Bull. Chem. Soc. Jpn., 1996, 69, 3123.
21 (a) D. Ruiz-Molina, C. Sporer, K. Wurst, P. Jaitner and J. Veciana,
Angew. Chem., Int. Ed., 2000, 39, 3688; (b) M. Tanaka, K. Matsuda,
T. Itoh and H. Iwamura, Angew. Chem., 1998, 110, 866; M. Tanaka,
K. Matsuda, T. Itoh and H. Iwamura, Angew. Chem., Int. Ed., 1998,
37, 810.
22 G. M. Sheldrick: SHELXL-97, Program for Crystal Structure
refinement, University of Göttingen, Germany 1997.
D a l t o n T r a n s . , 2 0 0 4 , 1 0 7 3 – 1 0 8 2
1082