Trinucleating copper: Synthesis and magnetostructural characterization of complexes supported by a hexapyridyl 1,3,5-triarylbenzene ligand

Article abstract of DOI:10.1002/anie.201005232

Copper threesome: A hexapyridyl ligand based upon a 1,3,5-triphenylbenzene framework coordinates three metal centers in a constrained environment (see picture). The tricopper(I) complex reduces dioxygen to form a tricopper(II) cluster. The capping anions affect the magnetism and EPR spectra of these species and reveal a linear dependence between the antiferromagnetic exchange parameter and the Cu-O-Cu angles.

Full text of DOI:10.1002/anie.201005232

Communications
DOI: 10.1002/anie.201005232
Multinuclear Complexes
Trinucleating Copper: Synthesis and Magnetostructural
Characterization of Complexes Supported by a Hexapyridyl
1,3,5-Triarylbenzene Ligand**
Emily Y. Tsui, Michael W. Day, and Theodor Agapie*
Multimetallic active sites are common in enzymes responsible
for the catalysis of multielectron chemical reactions such as
O2 activation, H2O oxidation, CO2 reduction, and N2 reduc-
tion.[1] In the context of reproducing the activity of such
enzymes and understanding the electronic structure and
reactivity of the active sites, the design of ligand architectures
capable of nucleating several metal centers is of interest. The
active sites of oxidases (laccase, ascorbate oxidase) and
oxygenases (particulate methane monooxygenase), which are
known or proposed to contain multicopper centers, have been
appealing synthetic targets due to the potential applications in
fuel cell technology, functionalization of organic substrates,
and liquefaction of methane.[2] Treatment of mononuclear
copper diamines with O2 generates a self-assembled Cu3O2
moiety; reduction, however, causes the loss of the trinuclear
core.[3] Multinucleating ligands represent an appealing alter-
nate approach for accessing multinuclear copper complexes.[4]
Of the variety of trinucleating ligands designed and inves-
tigated,[5] most are rather flexible and do not enforce
cooperative trinuclear reactivity or a well-defined arrange-
ment of the metal centers. Macrocyclic frameworks have
shown promise for forming tricopper(II) complexes, but
reactivity with O2 has yet to be reported for most of these
systems.[6]
We report here the design of a trinucleating ligand based
upon a 1,3,5-triphenylbenzene core that can closely constrain
Scheme 1. Synthesis and reactions of 1 and 2.
three copper centers. In addition to three bidentate dipyridyl
copper binding sites, the ligand variant discussed here (H3L,
Scheme 1) contains three hydroxy moieties that may serve as
shuttles for the protons necessary for dioxygen reduction to
water. The synthesis, oxygen reactivity, and spectroscopic
characterization of trinuclear copper(I) and copper(II) com-
plexes based upon this framework are reported.
The framework H3L was prepared by lithium–halogen
exchange of 1,3,5-tri(2’-bromophenyl)benzene[7] followed by
addition of three equivalents of di(2-pyridyl)ketone. Varia-
ble-temperature 1H NMR studies were conducted on a
solution of H3L in CD2Cl2 (see Supporting Information). At
room temperature, the 1H NMR spectroscopy signals are
broad, indicating site exchange due to fluxional behavior. The
spectrum at ꢀ508C has 24 signals corresponding to an isomer
of H3L with one dipyridyl moiety located on the opposite face
of the central aryl ring from the other two. At lower
temperatures other exchange processes are observed, possi-
bly related to rotation about the aryl–alkyl bonds.
[*] E. Y. Tsui, Dr. M. W. Day, Prof. T. Agapie
Division of Chemistry and Chemical Engineering
California Institute of Technology, Pasadena, CA 91125 (USA)
E-mail: agapie@caltech.edu
[**] We thank Lawrence M. Henling for crystallographic assistance,
David Vandervelde for help with NMR experiments, and Harry Gray,
Jay Winkler and Liviu Mirica for stimulating discussions. We are
grateful to Caltech and NSF GRFP (E.Y.T.) for funding. The Bruker
KAPPA APEXII X-ray diffractometer was purchased through an NSF
CRIF:MU award to Caltech, CHE-0639094. SQUID data were
collected at the MMRC of the Beckman Institute of the California
Institute of Technology.
Addition of three equivalents of a copper(I) salt such as
[Cu(CH3CN)4]OTf or [Cu(CH3CN)4]BF4 to H3L (Scheme 1)
forms the yellow compounds [Cu3(H3L)]·3X (X = OTf, BF4;
1·3X). The copper centers are each expected to bind two
pyridyl nitrogens and solvent molecules or the counter-
anion.[8] Variable-temperature 1H NMR spectra of 1·3BF4
indicate fluxional processes in solution similar to those of
H3L. The three copper centers are not constrained to remain
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201005232.
1668
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 1668 –1672

together and likely have rotational freedom about both aryl–
aryl and aryl–alkyl bonds. At low temperatures, 1·3OTf likely
adopts the same conformation as free H3L, with one copper
center located on the opposite face of the central aryl ring
from the other two copper ions. Addition of excess dioxygen
to a propionitrile solution of 1·3OTf at ꢀ788C does not result
in a reaction within 24 h. However, the reaction of 1·3OTf
with dioxygen in acetonitrile over 15 h at room temperature
forms a green compound [Cu3L]·3OTf (2·3OTf) quantita-
tively, as determined by 1H NMR spectroscopy using an
internal standard.
each copper(II) center is further coordinated by two pyridyl
ꢀ
nitrogens from neighboring dipyridyl moieties. The Cu
II
ꢀ
O(alkoxide) and Cu N distances are typical of other Cu
ꢀ
compounds (1.91–2.01 ꢀ), although the Cu O(anion) distan-
ces are longer (2.4 and 2.2 ꢀ, respectively), indicating weaker
coordination of the triflate or phosphate.
NMR spectroscopy data suggest that the [Cu3O3N6]
geometry observed in the solid-state structures remains
intact in solution allowing for three potentially accessible
coordination sites on the same face of the tricopper unit. The
1H NMR spectra of both 2·3OTf and 2·PO4 contain thirteen
signals (see Supporting Information), indicating that the
molecules are C3-symmetric in solution; a higher-symmetry
structure would give rise to nine peaks. The peaks are very
well resolved, indicating the presence of low-lying excited
states of the copper(II) complex that allow for fast electronic
relaxation.[6b,11] The signals were assigned by comparing
longitudinal relaxation times to distances in the X-ray
diffraction structure, as well as using gCOSY experiments
and NMR spectra of complexes with a deuterium-labeled
ligand (Supporting Information).
In the reaction with O2, complex 1·3OTf formally acts as a
source of both protons and electrons for the reduction of
between 0.75 and 1 equivalent of O2 (Toepler pump mea-
surement). The product derived from O2 in the oxidation of
1·3OTf has not been identified to date, but a peroxotitanyl
test indicates that H2O2 is not present in solution when the
conversion to 2·3OTf is complete.[9] In the absence of
dioxygen, oxidation of 1·3OTf in acetonitrile with three
equivalents of silver(I) triflate in the presence of triethyl-
amine cleanly forms 2·3OTf. Addition of fewer equivalents of
silver(I) does not form a mixed valence compound but rather
a mixture of 1·3OTf and 2·3OTf in the expected ratio as
To test the lability of the coordinated anions, a solution of
2·3OTf in CD3CN was titrated with nBu4NX (X = Br, I)
(Scheme 2). Addition of less than one equivalent of halide
leads to gradual changes in the 1H NMR spectra of the
1
judged by H NMR spectroscopy. For the reverse reaction,
2·3OTf can be converted to 1·3OTf upon addition of three
equivalents of cobaltocene in the presence of an acid such as
triethylamine hydrochloride or upon treatment with benzoin.
Compound 2·3OTf was independently prepared by
deprotonation of H3L with three equivalents of base followed
by addition of three equivalents of Cu(OTf)2. The solid-state
structures of 2·3OTf and of a phosphate analogue, 2·PO4,
show that the trinuclear complexes are pseudo-C3 symmetric
alkoxo-bridged trinuclear copper clusters capped by a triflate
or phosphate anion (Figure 1).[10] The copper(II) centers and
the alkoxo oxygens are linked in a chair-like [Cu3O3] core and
Scheme 2. Synthesis of 3·X·2OTf.
Figure 1. a) Solid-state structure of 2·3OTf as 50% thermal ellipsoids. Outer-sphere triflate anions and solvent molecules omitted for clarity.
Truncated views of solid-state structures of b) 2·3OTf, c) 2·PO4, d) 3·Br·2OTf, and e) 3·I·2OTf. Selected bond lengths [ꢀ]: b) Cu–Cu 3.4426(3)–
3.4624(3), Cu–O(alkoxide) 1.9087(9)–1.9544(10), Cu–N 1.9350(9)–1.9864(12), Cu–O(OTf) 2.4251(10)–2.5080(10). c) Cu–Cu 3.2948(5)–3.3309(5),
Cu–O(alkoxide) 1.9307(15)–1.9944(16), Cu–N 1.972(2)–2.014(2), Cu–O(PO4) 2.1945(17)–2.2147(16). d) Cu–Cu 3.2959(5)–3.3217(5), Cu–O-
(alkoxide) 1.9072(18)–1.9752(19), Cu–N 1.965(2)–2.000(2), Cu–Br 2.8243(5)–2.8851(5). e) Cu–Cu 3.3192(5)–3.3612(5), Cu–O(alkoxide)
1.9086(19)–1.9688(19), Cu–N 1.959(2)–1.993(2), Cu–I 3.0442(5)–3.1308(4).
Angew. Chem. Int. Ed. 2011, 50, 1668 –1672
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 1669

Communications
mixture. After the addition of one equivalent of nBu4NX, the
below 0.4 cm3 molꢀ1 K, indicating intermolecular antiferro-
magnetic interactions.[6d,12c,13b]
spectra of the mixture remain unchanged even upon adding
three equivalents of halide. This behavior suggests that only
one halide ion can bind to the [Cu3O3] cluster. X-ray
diffraction studies of single crystals of 3·Br·2OTf and
3·I·2OTf confirm that a single halide atom bridges the three
copper centers of each cluster, with the triflate anions
remaining outer-sphere (Figure 1). Based on the 1H NMR
spectra, it is likely that these halide-capped structures are
preserved in acetonitrile solution. Similar halide-bridged
clusters have previously been observed in tricopper(II)
complexes of other ligand systems.[12]
The intramolecular exchange coupling was modeled for a
triangle of S = 1/2 centers, and the energies of the spin states
of the systems were calculated using the isotropic spin
Hamiltonian [Eq. (1)]. The fits were not significantly
improved by varying two coupling constants. As a result, it
was assumed that J = J13 in an approximately equilateral
arrangement of the copper(II) centers. Application of the Van
Vleck equation yields the magnetic susceptibility equation
[Eq. (2)].[17]
H ¼ ꢀ2 J½ðS1 S2Þ þ ðS2 S3Þꢁꢀ2 J13ðS3 S1Þ
ð1Þ
ð2Þ
Alkoxo-bridged tricopper(II) clusters with similar
[Cu3O3] motifs are known, but generally form by self-
assembly of mononuclear copper(II) moieties.[12a,13] The
magnetic behavior of these complexes has been studied
because spin-frustrated triangular complexes may exhibit
interesting ground states. It has been shown that varying the
anion bridges of linear[14] and triangular[15] trinuclear cop-
per(II) complexes changes the magnetic coupling, but no
systematic study has been reported of these effects on m-
alkoxo-bridged trimers lacking a strongly m3-coordinated
ligand.[12a,16] Since the synthesis of anion variants of the
tricopper(II) complexes supported by the trinucleating ligand
L described above is facile, a magnetostructural investigation
of the effect of the capping anion the [Cu3O3N6] core was
performed.
The variable-temperature magnetic susceptibilities of
compounds 2 and 3 were studied, and plots of cM T vs. T are
shown in Figure 2. For 2·PO4, the cM T value at 300 K is close
to 1.2 cm3 molꢀ1 K as expected for three nearly independent
S = 1/2 centers. The room temperature cM T values for the
other complexes are below 1.2 cm3 molꢀ1 K due to antiferro-
magnetic exchange between the copper centers. The slope of
these curves decreases at low temperatures, with most
approaching a plateau around cM T= 0.4 cm3 molꢀ1 K, which
agrees with the spin-only value for an S = 1/2 system. When
cooled below 20 K, 2·3OTf and 2·3BF4 have cM T values
ꢀ
ꢁ
Ng2b2
5 þ expðꢀ3J=kTÞ
cM
¼
4kðT ꢀ qÞ 1 þ expðꢀ3J=kTÞ
The fitted magnetic susceptibility parameters for each
complex are shown in Table 1. Except for 2·3BF4,[18] the fits
[19]
(R ꢂ 10ꢀ4
)
were obtained by varying only J, g, and the
Curie–Weiss parameter q.[20]
Table 1: Magnetic susceptibility and structural parameters.
Compound
Cu-O-Cu angle [8]
J [cmꢀ1
]
g
q [K]
2·PO4
2·3OTf
2·3BF4
114.6–118.5
125.9–126.5
–
ꢀ2.7
ꢀ52.0
ꢀ52.2
ꢀ22.0
ꢀ7.3
2.09
2.10
2.10
1.95
1.98
1.7
ꢀ0.6
ꢀ1.6
1.0
3·I·2OTf
3·Br·2OTf
116.7–120.8
116.1–118.5
0.8
The above data show that the antiferromagnetic exchange
between spins varies considerably with the character of the
coordinated anion. Triangular copper(II) complexes that
couple through m3-hydroxy or m3-oxo groups display the
strongest antiferromagnetic interactions in structures with
large Cu-O-Cu angles, where the bridged clusters are more
planar.[15a,b,16,21] Since compounds 2 and 3 contain more
weakly coordinated capping anions, exchange between cop-
per(II) centers is expected to occur primarily through the m-
alkoxo moieties. This is consistent with the fact that com-
plexes 3·Br·2OTf and 3·I·2OTf display antiferromagnetic
exchange and have longer Cu–halide bonds (by ca. 0.4 ꢀ)
compared to trinuclear copper complexes with m3-halide
ligands in which ferromagnetic exchange is proposed to occur
through the m3-halide.[15b] Furthermore, the complexes con-
taining the least coordinating anions in the series, 2·3OTf and
2·3BF4, show the largest exchange.
The facially coordinated capping ligand distorts the
geometry of the [Cu3O3] core and changes the Cu-O-Cu
angles (Table 1). As the capping ligand binds more strongly,
the [Cu3O3] core distorts from a planar geometry by decreas-
ing the Cu-O-Cu angles; this correlates with a weakening of
antiferromagnetic exchange. This behavior is consistent with
the solid- and solution-state EPR spectra of 2·PO4 at temper-
atures below 20 K, which show the expected signal at g ꢂ 2
corresponding to the S = 1/2 state as well as a broad lower
field signal proposed to arise from a S = 3/2 state (Supporting
Figure 2. Plots of cM T vs. T for compounds 2 and 3 at an applied field
of 0.5 T. Solid lines represent the best fits obtained.
1670
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 1668 –1672

Information).[22] The populations of the two states vary with
temperature, indicating a spin equilibrium with the doublet
being the ground state. Compound 2·3OTf, with a more
negative J and thus a larger doublet–quartet energy splitting,
does not display an EPR signal arising from the S = 3/2 state
at low temperatures.
A linear dependence of the exchange interactions of
dihydroxo-bridged copper(II) dimers with the Cu-O-Cu
angles has been reported.[23] Similar trends extended to
more complex systems have been complicated by the
presence of multiple exchange pathways. Nevertheless, a
non-linear trend has been reported for the Cu-(m3-X)-Cu
angle for m3-bridged trimers.[15b,c] The present compounds
allow for a systematic magnetostructural study. When the
exchange parameters of compounds 2 and 3 are plotted
against the average Cu-O-Cu angle in the respective struc-
ture, the data indeed follow an approximately linear trend
(Figure 3). This dependence also holds when the available
upon reaction with O2, leads to alcohol deprotonation and
oxidation of copper. The C3-symmetric trinuclear core is
robust by spectroscopy and chemical reactivity. Changing the
capping anion facially coordinated to the tricopper(II) cluster
alters the Cu-O-Cu angle to lead to changes of the antiferro-
magnetic exchange coupling between neighboring copper
ions, providing a strategy for tuning the magnetism of the
trinuclear core. The observed linear trend between the
antiferromagnetic exchange coupling and the Cu-O-Cu
angle holds for a variety of previously reported trinuclear
species providing an extension to more complex systems of
the classical dependence observed for dihydroxo-bridged
copper(II) dimers. Complexes of the present ligand provide a
versatile framework for mechanistic studies related to O2 and
other small-molecule reactivity at trinuclear centers and
current studies are targeted toward further exploring Cu3–O2
chemistry as well as a variety of first-row transition metal
trinuclear clusters.
Received: August 21, 2010
Revised: October 24, 2010
Published online: January 11, 2011
Keywords: copper complexes · dioxygen · magnetic properties ·
.
multinucleating ligands · trinuclear clusters
[1] a) J. P. McEvoy, G. W. Brudvig, Chem. Rev. 2006, 106, 4455;
b) R. H. Holm, P. Kennepohl, E. I. Solomon, Chem. Rev. 1996,
96, 2239.
[2] a) E. I. Solomon, A. J. Augustine, J. Yoon, Dalton Trans. 2008,
3921; b) E. I. Solomon, P. Chen, M. Metz, S. K. Lee, A. E.
Palmer, Angew. Chem. 2001, 113, 4702; Angew. Chem. Int. Ed.
2001, 40, 4570; c) R. Balasubramanian, A. C. Rosenzweig, Acc.
Chem. Res. 2007, 40, 573; d) R. L. Lieberman, A. C. Rosenzweig,
Nature 2005, 434, 177; e) S. I. Chan, S. S. F. Yu, Acc. Chem. Res.
2008, 41, 969; f) L. M. Mirica, X. Ottenwaelder, T. D. P. Stack,
Chem. Rev. 2004, 104, 1013; g) R. Balasubramanian, S. M. Smith,
S. Rawat, L. A. Yatsunyk, T. L. Stemmler, A. C. Rosenzweig,
Nature 2010, 465, 115.
Figure 3. J vs. bridge angle (f) for a series of alkoxo- or hydroxo-
&
^
bridged copper(II) trimers. ( ) This work. ( ) a) Ref. [12a];
b) Ref. [6d]; c) Ref. [13b]; d) Ref. [24]. Straight line was calculated from
J=ꢀ3.41 f+386 cmꢀ1
.
[3] A. P. Cole, D. E. Root, P. Mukherjee, E. I. Solomon, T. D. P.
Stack, Science 1996, 273, 1848.
[4] P. P. Y. Chen, R. B. G. Yang, J. C. M. Lee, S. I. Chan, Proc. Natl.
Acad. Sci. USA 2007, 104, 14570.
structural and magnetic data from other m-alkoxo and m-
hydroxo copper(II) trimers of pseudo-C3 symmetry are
included.[6d,12a,13b,24] This behavior reflects the exchange
through the m2-O donors rather than the weakly coordinated
capping ligands. The exchange parameters for tricopper
systems with m3-hydroxy or m3-oxo ligands do not lie along
this trend. It is noteworthy that the linear trend holds despite
significant variations in the nature of the non-bridging ligands
(pyridines, alkyl amines, pyrrolides, and alkoxides). This
magnetostructural behavior suggests that, as in hydroxo-
bridged dimers, there is a Cu-O-Cu angle at which the copper
centers switch from antiferromagnetic to ferromagnetic
coupling. Such distorted structures may be difficult to
access, however.
[5] a) H. Ohi, Y. Tachi, S. Itoh, Inorg. Chem. 2006, 45, 10825;
b) E. C. Brown, B. Johnson, S. Palavicini, B. E. Kucera, L.
Casella, W. B. Tolman, Dalton Trans. 2007, 3035; c) S. T. Frey,
H. H. J. Sun, N. N. Murthy, K. D. Karlin, Inorg. Chim. Acta 1996,
242, 329; d) K. D. Karlin, Q. F. Gan, A. Farooq, S. C. Liu, J.
Zubieta, Inorg. Chem. 1990, 29, 2549; e) K. D. Karlin, Q. F. Gan,
Z. Tyeklar, Chem. Commun. 1999, 2295; f) D. Maiti, J. S.
Woertink, R. A. Ghiladi, E. I. Solomon, K. D. Karlin, Inorg.
Chem. 2009, 48, 8342.
[6] a) H. Adams, N. A. Bailey, M. J. S. Dwyer, D. E. Fenton, P. C.
Hellier, P. D. Hempstead, J. M. Latour, J. Chem. Soc. Dalton
Trans. 1993, 1207; b) A. Gonzꢁlez-Alvarez, I. Alfonso, J. Cano, P.
Diaz, V. Gotor, V. Gotor-Fernandez, E. Garcia-Espana, S.
Garcia-Granda, H. R. Jimenez, F. Lloret, Angew. Chem. 2009,
121, 6171; Angew. Chem. Int. Ed. 2009, 48, 6055; c) M. P. Suh,
M. Y. Han, J. H. Lee, K. S. Min, C. Hyeon, J. Am. Chem. Soc.
1998, 120, 3819; d) M. Inoue, C. Ikeda, Y. Kawata, S. Venkatra-
man, K. Furukawa, A. Osuka, Angew. Chem. 2007, 119, 2356;
Angew. Chem. Int. Ed. 2007, 46, 2306.
In summary, a trinucleating ligand containing six pyridine
and three alcohol donors on a 1,3,5-triphenylbenzene plat-
form has been used to assemble a tricopper(I) complex which,
Angew. Chem. Int. Ed. 2011, 50, 1668 –1672
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1671

Communications
[7] X. L. Feng, J. S. Wu, V. Enkelmann, K. Mullen, Org. Lett. 2006, 8,
Angaridis, P. Baran, R. Boca, F. Cervantes-Lee, W. Haase, G.
Mezei, R. G. Raptis, R. Werner, Inorg. Chem. 2002, 41, 2219;
e) M. Sinha Ray, S. Chattopadhyay, M. G. B. Drew, A. Figuerola,
J. Ribas, C. Diaz, A. Ghosh, Eur. J. Inorg. Chem. 2005, 2005,
4562.
1145.
[8] Three structures of CuI complexes supported by (dipyridyl)-
(aryl)methanol variants have been obtained to date; all show
coordination of two pyridyl nitrogens per copper center. The
alcohol groups do not coordinate and point away from the metal
center.
[16] P. Mukherjee, M. G. B. Drew, M. Estrader, C. Diaz, A. Ghosh,
Inorg. Chim. Acta 2008, 361, 161.
[9] G. M. Eisenberg, Ind. Eng. Chem. Anal. Ed. 1943, 15, 327.
[10] CCDC 755941 (2·PO4), 797194 (2·3OTf), 797195 (3·I·2OTf), and
797196 (3·Br·2OTf) contain the supplementary crystallographic
data for this paper. These data can be obtained free of charge
from The Cambridge Crystallographic Data Centre via www.
ccdc.cam.ac.uk/data_request/cif.
[17] a) K. Kambe, J. Phys. Soc. Jpn. 1950, 5, 48; b) E. Sinn, Coord.
Chem. Rev. 1970, 5, 313.
[18] An improved fit for the magnetism susceptibility of compound
2·3BF4 was calculated by including a temperature-independent
magnetism of ꢀ500 ꢄ 10ꢀ6 cmꢀ1. This discrepancy may be due to
diamagnetic impurities such as an unknown number of solvent
molecules or other salts.
[11] N. N. Murthy, K. D. Karlin, I. Bertini, C. Luchinat, J. Am. Chem.
Soc. 1997, 119, 2156.
[19] The data were fit by minimizing:
P
P
2
2
[12] a) M. Kodera, Y. Tachi, T. Kita, H. Kobushi, Y. Sumi, K. Kano,
M. Shiro, M. Koikawa, T. Tokii, M. Ohba, H. Okawa, Inorg.
Chem. 2000, 39, 226; b) T. Afrati, C. Dendrinou-Samara, C.
Raptopoulou, A. Terzis, V. Tangoulis, D. P. Kessissoglou, Dalton
Trans. 2007, 5156; c) Y.-B. Jiang, H.-Z. Kou, R.-J. Wang, A.-L.
Cui, J. Ribas, Inorg. Chem. 2005, 44, 709.
[13] a) I. Gautier-Luneau, D. Phanon, C. Duboc, D. Luneau, J. L.
Pierre, Dalton Trans. 2005, 3795; b) H. Lꢂpez-Sandoval, R.
Contreras, A. Escuer, R. Vicente, S. Bernes, H. Noth, G. J. Leigh,
N. Barba-Behrens, J. Chem. Soc. Dalton Trans. 2002, 2648.
[14] a) S. Gehring, P. Fleischhauer, H. Paulus, W. Haase, Inorg.
Chem. 1993, 32, 54; b) O. Das, E. Zangrando, T. K. Paine, Inorg.
Chim. Acta 2009, 362, 3617; c) S. Gehring, H. Astheimer, W.
Haase, J. Chem. Soc. Faraday Trans. 2 1987, 83, 347.
[15] a) B. Sarkar, M. Ray, Y. Z. Li, Y. Song, A. Figuerola, E. Ruiz, J.
Cirera, J. Cano, A. Ghosh, Chem. Eur. J. 2007, 13, 9297; b) R.
R ¼ jðcMTÞobsꢀðcMTÞcalcdj = ðcMTÞobs
.
[20] The parameter q was included in the magnetic susceptibility
equation to account for antiferromagnetic (q < 0) or ferromag-
netic (q > 0) intermolecular interactions. See Ref. [15e], [23a]
for examples.
[21] R. J. Butcher, C. J. OꢅConnor, E. Sinn, Inorg. Chem. 1981, 20,
537.
[22] a) P. Fleischhauer, S. Gehring, C. Saal, W. Haase, Z. Tomkowicz,
C. Zanchini, D. Gatteschi, D. Davidov, A. L. Barra, J. Magn.
Magn. Mater. 1996, 159, 166; b) I. A. Koval, H. Akhideno, S.
Tanase, C. Belle, C. Duboc, E. Saint-Aman, P. Gamez, D. M.
Tooke, A. L. Spek, J.-L. Pierre, J. Reedijk, New J. Chem. 2007,
31, 512; c) L. Gutierrez, G. Alzuet, J. A. Real, J. Cano, J. Borras,
A. Castineiras, Inorg. Chem. 2000, 39, 3608; d) L. Banci, A.
Bencini, D. Gatteschi, Inorg. Chem. 1983, 22, 2681.
[23] a) V. H. Crawford, H. W. Richardson, J. R. Wasson, D. J. Hodg-
son, W. E. Hatfield, Inorg. Chem. 1976, 15, 2107; b) E. Ruiz, P.
Alemany, S. Alvarez, J. Cano, J. Am. Chem. Soc. 1997, 119, 1297.
[24] L. M. Mirica, T. D. P. Stack, Inorg. Chem. 2005, 44, 2131.
ˇ
ˇ
Boca, L. Dlhꢁn, G. Mezei, T. Ortiz-Pꢃrez, R. G. Raptis, J. Telser,
Inorg. Chem. 2003, 42, 5801; c) L.-L. Wang, Y.-M. Sun, Z.-Y. Yu,
Z.-N. Qi, C.-B. Liu, J. Phys. Chem. A 2009, 113, 10534; d) P. A.
1672
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 1668 –1672