Polynuclear nickel(II) complexes: Preparation and magnetic properties of NiII4, NiII5, and Ni II6 species with ligands containing O∩ X∩O (X = S, Se or N) donor atoms

Article abstract of DOI:10.1002/ejic.200300164

A family of bisphenol-containing ligands with XO₂ donor atoms, where X = S, Se, or N, yields [Ni^II]₄ (1) and (2), [NiII]₅ (3), and [NiII]₆ (4) complexes. Complex 2, 3, and 4 have been structurally characterised by X-ray diffraction. The compounds were characterised spectroscopically and by variable-temperature (2-295K), variable-field (1, 4, and 7 T) magnetic susceptibility measurements. Analysis of the magnetic data shows predominant ferromagnetic interactions between the nickel centres (S_Ni = 1) indicating high-spin ground states for 1, 2, and 3, whereas complex 4 shows antiferromagnetic interactions. A plot of J vs. Ni-O-Ni angles for all structurally characterised Ni ₄O₄ cubane cores, including 2, irrespective of their symmetry, shows a large variation of J values within a small range of Ni-O-Ni angles. A simple two-J model was found to be adequate to describe the magnetic properties of the hexanuclear complex, 4. ( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003).

Full text of DOI:10.1002/ejic.200300164

FULL PAPER
Polynuclear Nickel(II) Complexes: Preparation and Magnetic Properties of
II
II
II
ʝ ʝ
Ni , Ni , and Ni Species with Ligands containing O X O
4
5
6
(
X ؍ S, Se or N) Donor Atoms
[a]
[a]
[a]
[a]
Tapan K. Paine, Eva Rentschler, Thomas Weyhermüller, and Phalguni Chaudhuri*
Keywords: Magnetic properties / Nickel / Selenium / Sulfur / Tripodal ligands
A family of bisphenol-containing ligands with XO
2
donor
(1) and (2), [NiII]
(4) complexes. Complex 2, 3, and 4 have been
structurally characterised by X-ray diffraction. The com- 2, irrespective of their symmetry, shows a large variation of
ground states for 1, 2, and 3, whereas complex 4 shows anti-
ferromagnetic interactions. A plot of J vs. Ni−O−Ni angles for
all structurally characterised Ni O cubane cores, including
4 4
II
atoms, where X = S, Se, or N, yields [Ni ]
4
5
II
(
3), and [Ni ]
6
pounds were characterised spectroscopically and by vari-
able-temperature (2−295 K), variable-field (1, 4, and 7 T)
magnetic susceptibility measurements. Analysis of the mag-
netic data shows predominant ferromagnetic interactions be-
J values within a small range of Ni−O−Ni angles. A simple
two-J model was found to be adequate to describe the mag-
netic properties of the hexanuclear complex, 4.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
tween the nickel centres (SNi = 1) indicating ‘‘high-spin’’ Germany, 2003)
Introduction
of obtaining ‘‘high spin’’ species. Although polynuclear
nickel() complexes containing up to four metal ions are
There are several intriguing features associated with po-
lynuclear transition metal complexes. Firstly, these com-
plexes can have unusual electronic properties and may serve
as sources of fundamental information about exchange
coupling in multinuclear assemblies. A second general rea-
son to study polynuclear metal complexes is that they may
be building blocks for molecularly based magnetic materi-
als. In another field of coordination chemistry, polynuclear
complexes are important for the advancement of bioinor-
ganic chemistry, both in their roles in advancing under-
[8Ϫ9]
[10]
[11]
[12]
not rare,
assemblies with five,
six,
and more
nickel() ions still remain a matter of curiosity. In this paper
we report on the extension of our studies with the ligands
containing the donor atoms (O, X, O), where X ϭ S, Se, or
N, and include the preparation and magnetochemical
characterisation of
[
[
[
NiII
4
LS
2
(µ
3
-OMe)
-OMe)
-OH) (µ
4
(MeOH)
(MeOH)
-OH (EtOH)
2
4
(MeCN)
(MeCN)
] (3),
2
] (1),
II Se
Ni
Ni
4
L
L
2
(µ
3
4
4
2
] (2),
II
N
5
4
(µ
3
2
2
2
)
2
[1]
standing of the chemistry of metallobiomolecules and in
the production of useful new biomimetic and bioinspired
reactions. The magnetic molecular assemblies are therefore [NiII LN (µ -OHMe)(µ -OMe) (µ -OH) (MeOH) (MeCN) ] (4),
and
6
4
2
3
2
3
2
2
2
of interest for research areas as different as biochemistry
and molecular magnetism.[ These potential applications where H L is 2,2Ј-thiobis(4,6-di-tert-butylphenol), H L
2]
S
Se
2
2
N
and issues have sparked and fuelled the interest of coordi- is 2,2Ј-selenobis(4,6-di-tert-butylphenol), and H L
is
2
nation chemists in synthesising molecules with large num- methylamino-N,N-bis(2-methylene-4,6-dimethylphenol).
bers of unpaired electrons in ground states. Consequently,
the search for ligands capable of giving rise to molecules of
high nuclearity is an important theme in modern coordi-
nation chemistry.
We have been exploring the feasibility of using phenol-
[
3Ϫ6]
containing ligands
to influence the nuclearity and top-
ology of metal complexes. Since nickel() is known to have
a large single-ion zero-field splitting and often gives rise
to ferromagnetic coupling, we have especially focussed our
attention on polynuclear nickel() complexes[ with the aim
Results and Discussion
7]
Depending on the solvents and the bases used during the
preparations, complexes 1Ϫ4 were isolated in good yields
[
a]
Max-Planck-Institut für Bioanorganische Chemie,
Stiftstrasse 34Ϫ36, 45470 Mülheim an der Ruhr, Germany
E-mail:Chaudh@mpi-muelheim.mpg.de
(
Scheme 1). The sulfur- and selenium-containing ligands,
S
Se
H L and H L , reacted with NiCl ·6H O at room tem-
2
2
2
2
Eur. J. Inorg. Chem. 2003, 3167Ϫ3178
DOI: 10.1002/ejic.200300164
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3167

T. K. Paine, E. Rentschler, T. Weyhermüller, P. Chaudhuri
FULL PAPER
shown by the reactions’ insensitivity to the ligand/NiII ra-
tios, which vary between 2:1 to 0.5:1.
Selected IR data for complexes 1Ϫ4 are given in the Exp.
Ϫ1
Sect. The sharp ν(OH) band around 3400 cm for the free
ligands is replaced by a broad band, indicating that the phe-
nol character of the ligand has been lost on complexation.
Ϫ1
There are several strong bands in the 2960Ϫ2820 cm re-
gion, due to the ν(CϪH) stretchings of the tert-butyl
Ϫ1
groups. The weak bands in the 2300Ϫ2250 cm region
correspond to the ν(CN) vibrations of the CH CN mol-
ecule. The strong sharp band at 1047 cm for both 1 and
3
Scheme 1
Ϫ1
2
is attributed to the ν(OCH ) group of the bridging me-
3
[
19]
Ϫ1
thoxide ligand.
The corresponding band at 1057 cm
perature in a solvent mixture (5:1) of methanol and aceto-
for 4 appears as a medium band. The other ν(CϪH), ν(Cϭ
C), and ν(CϪO) vibrations are found in the normal ranges
for these types of linkages.
nitrile containing LiOCH as the base to deprotonate the
3
phenolic ϪOH groups, from which complexes 1 and 2,
respectively, were isolated. On the other hand, treatment of
N
Mass spectrometry in the EI and ESI mode has not pro-
ved to be a useful analytical tool for characterisation of the
the nitrogen-containing ligand H L with NiCl ·6H O in
2
2
2
ethanol in the presence of the base LiOC H resulted in
2
5
II
polynuclear Ni complexes 1Ϫ4, presumably due to their
a yellow-green microcrystalline solid, which on subsequent
recrystallisation from a solvent mixture (1:2) of CH Cl and
decomposition at the temperatures applied in mass spec-
trometry.
2
2
C H OH yielded pure pentanuclear material 3 in a good
2
5
Ϫ
isolated yield of 75%. The incorporation of OH and H O
2
Single-Crystal X-ray Diffraction Studies
in 3 presumably implies the involvement of water from the
starting material. Interestingly, the isolation of the green 4,
The analytical, spectroscopic and magnetic data for 1 and
II
a Ni species, from a solvent mixture of methanol and 2 showed their isostructural characters, so X-ray analysis
6
acetonitrile was particularly sensitive to the use of dry sol- only of 2 was undertaken, to remove the doubts regarding
vents, although the starting material was a hydrated metal connectivity.
salt. This indicates that complex 4 appears in the reaction
mixture only at water concentrations considerably lower Molecular Structure of 2
than present in the case of complex 3, and hence the ab-
The crystal structure of 2 is shown in Figure 1, with selec-
sence of water as ligands in the final product 4. That the ted bond lengths and angles provided in Table 1. The solid-
isolated complexes 1Ϫ4 are thermodynamic products is state structure of 2·3.5MeCN·0.5MeOH contains discrete
Figure 1. Molecular structure of [Ni
4
LSe
2
(µ
3
-OMe)
4
(MeOH)
4
(MeCN)
2
] (2)
www.eurjic.org
3168
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2003, 3167Ϫ3178

Polynuclear Nickel() Complexes
FULL PAPER
˚
LSe
nickel atoms and one of four µ -oxygen atoms originating
from bridging methoxide groups. Each Ni is in a distorted
Table 1. Selected bond lengths [A] and angles [deg] for [Ni
OMe) (MeOH) (MeCN) ]·3.5MeCN·0.5MeOH (2)
4
2
(µ
3
-
3
II
4
4
2
octahedral environment and bound to three µ -OCH li-
3
3
Ni(1)ϪO(1)
2.039(2)
O(1)ϪNi(1)ϪO(2)
96.49(10)
gands. For Ni(1) and Ni(2), two facially coordinated
ʝ
ʝ
Se 2Ϫ
O Se O donor sets of the deprotonated ligands [L ]
complete the sixfold coordination. Each Ni(3) and Ni(4) is
coordinated to three bridging methoxide oxygens and two
terminal MeOH ligands. The sixth coordination site in each
nickel centre is occupied by a N-atom of a terminal MeCN
molecule. Although it was not possible to locate all the hy-
drogen atoms in the structure, the intramolecular contacts
between the oxygen atoms of the methanol molecules and
the oxygen atoms of the phenolate groups of the ancillary
Ni(1)ϪO(2)
Ni(1)ϪO(65)
Ni(1)ϪO(67)
Ni(1)ϪO(66)
Ni(1)ϪSe(1)
Ni(2)ϪO(3)
Ni(2)ϪO(64)
Ni(2)ϪO(4)
Ni(2)ϪO(66)
Ni(2)ϪO(67)
Ni(2)ϪSe(2)
Ni(3)ϪO(66)
Ni(3)ϪO(65)
Ni(3)ϪO(64)
Ni(3)ϪO(63)
Ni(3)ϪO(61)
Ni(3)ϪN(80)
Ni(4)ϪO(67)
Ni(4)ϪO(65)
Ni(4)ϪO(64)
Ni(4)ϪO(60)
Ni(4)ϪN(70)
Ni(4)ϪO(62)
2.047(2) O(1)ϪNi(1)ϪO(65)
2.056(2) O(2)ϪNi(1)ϪO(65)
2.073(2) O(1)ϪNi(1)ϪO(67) 170.61(9)
2.093(2) O(2)ϪNi(1)ϪO(67) 90.35(9)
2.4452(6) O(65)ϪNi(1)ϪO(67) 82.20 (9)
2.054(2) O(1)ϪNi(1)ϪO(66) 94.52(9)
2.058(2) O(2)ϪNi(1)ϪO(66) 168.72(9)
2.060(2) O(65)ϪNi(1)ϪO(66) 81.69(9)
2.074(2) O(67)ϪNi(1)ϪO(66) 78.44(9)
2.082(2) O(1)ϪNi(1)ϪSe(1)
2.4480(6) O(2)ϪNi(1)ϪSe(1)
2.011(2) O(65)ϪNi(1)ϪSe(1) 176.40(7)
2.036(2) O(67)ϪNi(1)ϪSe(1) 100.26(6)
2.043(2) O(66)ϪNi(1)ϪSe(1) 96.19(6)
2.058(3) O(3)ϪNi(2)ϪO(64)
2.073(3) O(3)ϪNi(2)ϪO(4)
2.074(3) O(64)ϪNi(2)ϪO(4)
2.031(2) O(3)ϪNi(2)ϪO(66) 169.84(10)
2.034(2) O(64)ϪNi(2)ϪO(66) 82.10(9)
2.043(2) O(4)ϪNi(2)ϪO(66)
2.054(3) O(3)ϪNi(2)ϪO(67)
90.69(9)
95.73(9)
86.57(7)
86.90(7)
Se 2Ϫ
˚
˚
[
L ]
ligand Ϫ of ca. 2.70 A [O(1)···O(63) 2.674 A;
˚
˚
O(2)···O(62) 2.718 A; O(3)···O(60) 2.702 A; O(4)···O(61)
2.720 A] Ϫ can be interpreted as hydrogen bonds. The intra-
˚
92.07(9)
97.97(10)
95.42(9)
molecular hydrogen bonds have the effect of lowering the
effective symmetry of the [Ni (OMe) ]-core from T to D .
4
4
d
2d
A similar pattern of intramolecular hydrogen bonding
and structural distortions have also been observed in other
II
90.88(10) alkoxide cubane complexes containing Ni and bound al-
92.34(10)
cohol. The NiϪOϪNi angles in 2 vary between 95.9(1) and
2.066(3) O(64)ϪNi(2)ϪO(67) 82.12(9)
2.076(3) O(4)ϪNi(2)ϪO(67) 169.48(10)
O(66)ϪNi(2)ϪO(67) 78.66(9)
1
01.5(1)°. The Ni···Ni distances on different cubic faces are
˚
also different, the shortest being Ni(3)···Ni(2) 3.046(2) A
and the longest Ni(1)···Ni(2) 3.217(2) A. Average NiϪO
bond lengths of 2.045 A (phenolate) and 2.075 A (methox-
ide) lie well within the range of reported
˚
O(3)ϪNi(2)ϪSe(2)
Ni(3)ϪO(64)ϪNi(4) 96.93(10) O(64)ϪNi(2)ϪSe(2) 177.83(7)
Ni(3)ϪO(64)ϪNi(2) 95.94(10) O(4)ϪNi(2)ϪSe(2) 86.40(7)
Ni(4)ϪO(64)ϪNi(2) 96.54(10) O(66)ϪNi(2)ϪSe(2) 99.07(6)
Ni(4)ϪO(65)ϪNi(3) 97.44(10) O(67)ϪNi(2)ϪSe(2) 96.30(6)
86.51(7)
˚
˚
[4,8]
values for the
corresponding bond lengths of the tetranuclear cubane-like
II
Ni . All phenyl rings of the ligands are planar and retain
Ni(4)ϪO(65)ϪNi(1) 96.40(9)
O(66)ϪNi(3)ϪO(65) 84.22(9)
Ni(3)ϪO(65)ϪNi(1) 96.48(10) O(66)ϪNi(3)ϪO(64) 84.06(9)
Ni(3)ϪO(66)ϪNi(2) 96.41(10) O(65)ϪNi(3)ϪO(64) 82.71(10)
their aromaticity after complexation. No substantial differ-
ences in bond lengths and angles are found between the two
crystallographically independent molecules.
Ni(3)ϪO(66)ϪNi(1) 96.09(9)
Ni(2)ϪO(66)ϪNi(1) 101.05(10) O(65)ϪNi(3)ϪO(63) 88.37(10)
Ni(4)ϪO(67)ϪNi(1) 95.93(9) O(64)ϪNi(3)ϪO(63) 169.29(10)
Ni(4)ϪO(67)ϪNi(2) 96.15(10) O(66)ϪNi(3)ϪO(61) 88.15(10) Molecular Structure of 3
O(66)ϪNi(3)ϪO(63) 89.18(10)
Ni(1)ϪO(67)ϪNi(2) 101.45(10) O(65)ϪNi(3)ϪO(61) 170.29(10)
O(64)ϪNi(3)ϪO(61) 90.55(10)
The molecular structure of 3 is shown in Figure 2. Selec-
ted interatomic distances and bond angles are listed in
Ni(1)···Ni(2)
Ni(1)···Ni(3)
Ni(1)···Ni(4)
Ni(2)···Ni(3)
Ni(2)···Ni(4)
Ni(3)···Ni(4)
3.217(2) O(63)ϪNi(3)ϪO(61) 97.54(11)
3.052(2) O(66)ϪNi(3)ϪN(80) 177.73(11)
3.049(2) O(65)ϪNi(3)ϪN(80) 93.72(12)
Table 2. The solid-state structure of 3·0.75CH Cl contains
2
2
N
discrete pentanuclear molecules of formula [Ni L (µ3-
5
4
3.046(2) O(64)ϪNi(3)ϪN(80) 96.63(12) OH) (µ -OH ) (EtOH) ]. The structure can be regarded as
2
2
2 2
2
3.061(2) O(63)ϪNi(3)ϪN(80) 89.82(12)
3.058(2) O(61)ϪNi(3)ϪN(80) 93.99(12)
O(67)ϪNi(4)ϪO(65) 83.80(9)
a corner-shared dicubane, in which one of the corners of
each cubane is missing. The pentanuclear assembly is com-
posed of a Ni ϪNiϪNi array; there is a virtually twofold
O(67)ϪNi(4)ϪO(64) 83.75(9)
2
2
O(65)ϪNi(4)ϪO(64) 82.75(10) axis passing through the central Ni(3). There are two differ-
O(67)ϪNi(4)ϪO(60) 89.06(10) ent types of peripheral Ni atoms: the five-coordinate Ni(2)
O(65)ϪNi(4)ϪO(60) 170.10(10)
and Ni(5) with a NO square-pyramidal environment and
the six-coordinate Ni(1) and Ni(4) in a distorted octahedral
4
O(64)ϪNi(4)ϪO(60) 89.69(10)
O(67)ϪNi(4)ϪN(70) 178.40(11)
NO5 environment. The peripheral metal atoms, Ni(1),
O(65)ϪNi(4)ϪN(70) 96.09(11)
O(64)ϪNi(4)ϪN(70) 94.65(11) Ni(2), Ni(4), and Ni(5), are bound equatorially to the de-
O(60)ϪNi(4)ϪN(70) 90.86(12)
O(67)ϪNi(4)ϪO(62) 88.87(10)
O(65)ϪNi(4)ϪO(62) 90.73(10)
O(64)ϪNi(4)ϪO(62) 170.63(10)
N 2Ϫ
protonated ancillary ligand [L ] . Two µ -hydroxide ions
3
[O(100) for Ni(2) and O(200) for Ni(5)] act as the axial do-
nor atoms of the square pyramids. That the donor atoms
O(60)ϪNi(4)ϪO(62) 95.99(11) O(101) and O(201) are constituents of water molecules is
N(70)ϪNi(4)ϪO(62) 92.73(12) discernible from the NiϪO bond lengths. Furthermore, the
hydrogen atoms of the water molecules and of hydroxy
groups were located from a difference map and are shown
tetrameric molecules of formula [Ni L (µ -OMe) - as circles of arbitrary radii in Figure 2. Two ethanol mol-
Se
4
2
3
4
(
MeOH) (MeCN) ], based on a Ni O cubane-type unit and ecules, O(300) and O(400), are bound as terminal ions to
4 2 4 4
consisting of two interpenetrating tetrahedra: one of four Ni(1) and Ni(4), respectively, trans to the µ -OH ligands.
2
2
Eur. J. Inorg. Chem. 2003, 3167Ϫ3178
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3169

T. K. Paine, E. Rentschler, T. Weyhermüller, P. Chaudhuri
FULL PAPER
5 4 3 2 2 2 2 2
Figure 2. ORTEP representation of [Ni (µ -OH) (µ -OH ) (EtOH) ] (3)
LN
The hydrogen atoms on the OH and EtOH ligands enter six-coordinate nickel centres have NO environments. The
2
5
N 2Ϫ
into hydrogen bonding with the phenolate oxygens O(1), ancillary ligand [L ] , O(15), N(8) and O(1), together with
O(41), O(35) and O(61), with O(1)···O(201) 2.459 A, the phenolate oxygen O(70) (which is a µ -bridging ligand)
O(41)···O(101) 2.448 A, O(61)···O(400) 2.563 A, and from a second ligand [L ] , form the equatorial plane for
O(35)···O(300) 2.590 A distances, thus presumably making the five-coordinate Ni(1). The six-coordinate Ni(2) is bound
˚
2
˚
˚
N 2Ϫ
˚
the cluster unsusceptible to disruption. The central Ni(3) to two µ -OH (O10 and O60), one µ -OMe (O70), one µ -
3
3
2
is six-coordinate, with oxygen donor atoms only. The long OHMe (O60), one µ -phenolate (O1) and an acetonitrile
2
Ni···Ni distances (Table 2) exclude the possibility that the (N50) ligand. On the other hand, Ni(3) is triply coordinated
N 2Ϫ
exchange coupling between the paramagnetic centres is to the ligand [L ] along with its monocoordination to a
transmitted directly through space. All NiϪO and NiϪN methanol (O80), a µ -hydroxide and a µ -methoxide group.
3
3
bond lengths are in conformity with the nickel ions in ϩII
oxidation state.
Magnetic Susceptibility Studies
Molecular Structure of 4
Magnetic susceptibility data for polycrystalline samples
X-ray analysis reveals that 4 contains a Ni unit and is of complexes 1Ϫ4 were collected in the 2Ϫ290 K tempera-
6
N
best
written
as
[Ni L (µ -OHMe)(µ -OMe) (µ - ture range in an applied magnetic field of 1 T to character-
6 4 2 3 2 3
OH) (MeOH) (MeCN) ], as shown in Figure 3(a). Fig- ise the nature and magnitude of the exchange interaction
2
2
2
ure 3(b) highlights the central hexanuclear core with the propagated by the bridging ligands. We use the Heisenberg
ˆ
ˆ
ˆ
atom connectivity in the coordination sphere. Selected spin-Hamiltonian in the form H ϭ Ϫ2J S ·S for an iso-
A B
metrical parameters are given in Table 3. The structure of tropic exchange coupling with SNi ϭ 1 for 1Ϫ4. The exper-
the Ni -cluster can be regarded as a face-shared distorted imentally measured magnetic data were simulated by use of
6
[
13]
dicubane in which one of the corners of each cubane is a least-squares fitting computer program with a full-ma-
missing. There is a twofold crystallographic axis passing trix diagonalisation of exchange coupling, Zeeman split-
through an oxygen atom (O60) of the bridging µ -methanol ting, and axial single-ion zero-field interactions (DS ), if
2
2
z
molecule. The nickel centres Ni(1) and Ni(1)# are five-coor- necessary. The susceptibility data were corrected for diam-
dinate, whereas Ni(2), Ni(3) and their symmetry related agnetism (Pascal corrections). No temperature-independent
nickel centres are six-coordinate. Five-coordinated nickel paramagnetism (TIP) or paramagnetic impurities (PI) were
centres possess square-pyramidal NO4 geometries, with used for the simulations. The experimental data are dis-
O(70) of the µ -OMe group as the axial ligand, whereas the played in Figure 4 as the effective magnetic moment (µeff)
3
3170
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 3167Ϫ3178

Polynuclear Nickel() Complexes
Table 2. Selected interatomic distances [A] and angles [deg] for
FULL PAPER
˚
versus temperature (T). The solid lines in the figure rep-
N
[Ni
5
L
4
(µ
3
-OH)
2
(µ
2
-OH
2
)
2
(EtOH)
2
]·0.75 CH
2
Cl
2
(3)
resent the simulations.
The magnetic moment µ /molecule for 1 and 2 of 6.54
eff
Ϫ1
Ni(1)ϪO(1)
1.999(3) O(1)ϪNi(1)ϪO(15)
2.047(3) O(1)ϪNi(1)ϪO(300)
2.081(3) O(15)ϪNi(1)ϪO(300) 91.05(13)
2.083(4) O(1)ϪNi(1)ϪN(8)
2.096(3) O(15)ϪNi(1)ϪN(8)
2.159(3) O(300)ϪNi(1)ϪN(8)
1.980(3) O(1)ϪNi(1)ϪO(100)
2.024(3) O(15)ϪNi(1)ϪO(100) 81.73(12)
2.031(3) O(300)ϪNi(1)ϪO(100) 88.70(13)
167.33(14)
97.52(13)
3
µ (χ ·T ϭ 5.332 cm ·K·mol ) and 6.52 µ (χ ·T ϭ 5.233
B
M
B
M
3
Ϫ1
cm ·K·mol ), respectively, at 290 K increases monoton-
Ni(1)ϪO(15)
Ni(1)ϪO(300)
Ni(1)ϪN(8)
ically with decreasing temperature, reaching a maximum at
3
Ϫ1
92.34(15) 20 K with µeff ϭ 7.54 µ (χ ·T ϭ 7.115 cm ·K·mol ) for
B M
3
Ni(1)ϪO(100)
Ni(1)ϪO(101)
Ni(2)ϪO(35)
Ni(2)ϪO(100)
Ni(2)ϪO(15)
Ni(2)ϪO(21)
Ni(2)ϪN(28)
Ni(3)ϪO(100)
Ni(3)ϪO(200)
Ni(3)ϪO(75)
Ni(3)ϪO(21)
Ni(3)ϪO(201)
Ni(3)ϪO(101)
Ni(4)ϪO(41)
Ni(4)ϪO(55)
Ni(4)ϪO(400)
Ni(4)ϪO(200)
Ni(4)ϪN(48)
Ni(4)ϪO(201)
Ni(5)ϪO(61)
Ni(5)ϪO(200)
Ni(5)ϪO(55)
Ni(5)ϪO(75)
Ni(5)ϪN(68)
96.20(14) 1 and µeff ϭ 7.15 µB (χ ·T ϭ 6.444 cm ·K·molϪ1) for 2.
M
95.07(14)
89.11(13)
Below 20 K, µeff starts to decrease, reaching a value of 6.01
3
Ϫ1
µ (χ ·T ϭ 4.60 cm ·K·mol ) for 1 and 4.68 µ (χ ·T ϭ
B
M
B
M
3
Ϫ1
3
.525 cm ·K·mol ) for 2 at 2 K (Figure 4); it is clear that
2.044(3) N(8)ϪNi(1)ϪO(100) 175.75(14) the magnetic properties of 1 and 2 are dominated by a fer-
2.100(4) O(1)ϪNi(1)ϪO(101)
2.009(3) O(15)ϪNi(1)ϪO(101) 77.10(12)
2.011(3) O(300)ϪNi(1)ϪO(101) 164.01(13)
92.62(12)
3
romagnetic exchange interaction between four A nickel()
2
ions as propagated by bridging methoxides. The structural
parameters of 2 strongly suggest a lower symmetry than Td
2.097(3) N(8)ϪNi(1)ϪO(101)
96.84(14)
2.102(3) O(100)ϪNi(1)ϪO(101) 79.10(12) (idealised D2d) for the molecule. We have assumed the same
2.126(3) O(35)ϪNi(2)ϪO(100) 93.63(13) symmetry for 1.
2.128(3) O(35)ϪNi(2)ϪO(15)
89.06(13)
The most important parameter in the magnetostructural
1.999(3) O(100)ϪNi(2)ϪO(15) 83.91(13)
2.059(3) O(35)ϪNi(2)ϪO(21) 172.21(14)
2.069(4) O(100)ϪNi(2)ϪO(21) 78.64(12)
2.085(3) O(15)ϪNi(2)ϪO(21)
2.087(4) O(35)ϪNi(2)ϪN(28)
2.187(3) O(100)ϪNi(2)ϪN(28) 112.41(14)
1.983(3) O(15)ϪNi(2)ϪN(28) 163.59(14)
2.030(3) O(21)ϪNi(2)ϪN(28)
correlation of the Ni O cubane cores has been reported in
4
4
[
4,9i,9h,9l,9p]
the literature
to be the averaged NiϪOϪNi an-
89.22(13) gle of a cubane face. In accordance with the three averaged
91.50(14)
NiϪOϪNi angles of 96.2°, 97.2°, and 101.3° observed for
, a three-J model based on the diagram shown in Scheme 2
was used to analyse the magnetic data for 1 and 2.
Very good fits were obtained by use of this model, the
2
92.29(14)
2.046(4) O(100)ϪNi(3)ϪO(200) 177.78(14)
Ϫ1
2.053(3) O(100)ϪNi(3)ϪO(75) 103.80(13) best fit parameter sets being J ϭ ϩ2.39 cm , J ϭ Ϫ5.77
1
2
2.109(5) O(200)ϪNi(3)ϪO(75) 77.70(13)
O(100)ϪNi(3)ϪO(21) 77.62(12)
Ϫ1
Ϫ1
cm , J ϭ ϩ6.9 cm , g ϭ 2.27 for 1 and J ϭ ϩ1.05
cm , J ϭ Ϫ4.1 cm , J ϭ ϩ5.5 cm , g ϭ 2.25 for 2.
3
1
Ϫ1
Ϫ1
Ϫ1
2
3
Ni(2)ϪO(15)ϪNi(1) 96.78(13) O(200)ϪNi(3)ϪO(21) 103.18(13)
Ni(2)ϪO(21)ϪNi(3) 99.91(13) O(75)ϪNi(3)ϪO(21) 117.96(13)
No zero-field splitting parameter D, TIP, or PI was needed
Ni(5)ϪO(55)ϪNi(4) 95.99(14) O(100)ϪNi(3)ϪO(201) 96.66(13) to obtain the good quality of fit shown in Figure 4 as solid
Ni(5)ϪO(75)ϪNi(3) 100.09(13) O(200)ϪNi(3)ϪO(201) 81.87(13)
lines. The differences and natures of the signs of the ex-
Ni(3)ϪO(100)ϪNi(2) 103.83(14) O(75)ϪNi(3)ϪO(201) 82.63(13)
change parameters are in full agreement with the
Ni(3)ϪO(100)ϪNi(1) 100.30(13) O(21)ϪNi(3)ϪO(201) 159.34(13)
NiϪOϪNi angle correlation and support the three-J model
Ni(2)ϪO(100)ϪNi(1) 95.43(13) O(100)ϪNi(3)ϪO(101) 81.81(13)
Ni(3)ϪO(101)ϪNi(1) 94.66(13) O(200)ϪNi(3)ϪO(101) 96.27(13) used, which takes account of the reduced symmetry of the
Ni(3)ϪO(200)ϪNi(5) 103.88(14) O(75)ϪNi(3)ϪO(101) 161.52(13) cubane core observed in the X-ray structure of 2. These
Ni(3)ϪO(200)ϪNi(4) 100.99(15) O(21)ϪNi(3)ϪO(101) 80.32(12)
results for 1 and 2 thus predict a switch from ferromagnetic
Ni(5)ϪO(200)ϪNi(4) 95.67(13) O(201)ϪNi(3)ϪO(101) 79.21(13)
to antiferromagnetic coupling only for NiϪOϪNi angles
Ni(3)ϪO(201)ϪNi(4) 94.22(13) O(41)ϪNi(4)ϪO(55) 166.40(13)
[
4]
above ca. 98°, as also observed by us earlier. We have also
Ni(1)···Ni(2)
Ni(1)···Ni(3)
Ni(2)···Ni(3)
Ni(3)···Ni(5)
Ni(3)···Ni(4)
Ni(4)···Ni(5)
3.049(2) O(41)ϪNi(4)ϪO(400) 96.86(14)
3.152(2) O(55)ϪNi(4)ϪO(400) 93.35(14) plotted the observed J values for 2 and other structurally
3.174(2) O(41)ϪNi(4)ϪO(200) 89.15(13)
3.181(2) O(55)ϪNi(4)ϪO(200) 82.01(13)
3.160(2) O(400)ϪNi(4)ϪO(200) 89.24(14)
4ϩ
characterised [Ni (OR) ]
cubanes against average
4
4
NiϪOϪNi angles (Figure 5 and Table 4). The solid line in
Figure 5, shown only as a guide for the eyes, allows us to
3.050(2) O(41)ϪNi(4)ϪN(48)
O(55)ϪNi(4)ϪN(48)
92.99(16)
95.41(16) see that the antiferromagnetic interactions (ϪJ) fall fairly
O(400)ϪNi(4)ϪN(48) 93.11(15) well on the correlated line with respect to NiϪOϪNi angles.
O(200)ϪNi(4)ϪN(48) 176.61(14)
O(41)ϪNi(4)ϪO(201) 92.13(13)
O(55)ϪNi(4)ϪO(201) 76.08(13)
O(400)ϪNi(4)ϪO(201) 164.92(15)
In contrast, the positive J values scatter appreciably.
The magnetic data indicate that 1 and 2 have compli-
cated, low-lying magnetic structures with non-diamagnetic
O(200)ϪNi(4)ϪO(201) 78.77(13) ground states smaller than the spin state S ϭ 4 expected for
t
N(48)ϪNi(4)ϪO(201) 98.52(14)
O(61)ϪNi(5)ϪO(200) 92.64(14)
a ferromagnetically coupled tetranuclear nickel() species
(SNi ϭ 1). We have performed magnetisation measurements
O(61)ϪNi(5)ϪO(55)
88.17(15)
at applied magnetic fields of 1, 4, and 7 T and collected
O(200)ϪNi(5)ϪO(55) 83.69(14)
O(61)ϪNi(5)ϪO(75) 170.92(15) most of the magnetisation data in the 2Ϫ10 K temperature
O(200)ϪNi(5)ϪO(75) 78.29(13) range (26 data point) and only 7 data points in the
O(55)ϪNi(5)ϪO(75)
O(61)ϪNi(5)ϪN(68)
90.77(13)
90.97(16)
12.3Ϫ260.0 K range, which are shown in Figure 6 as plots
of M/Ngβ vs. βH/kT for 1 and 2. It is clear from Figure 6
that the saturated magnetisation value reaches a value of
O(200)ϪNi(5)ϪN(68) 114.27(15)
O(55)ϪNi(5)ϪN(68) 162.04(15)
O(75)ϪNi(5)ϪN(68)
92.80(15) approximately 3.45 for 1 and 3.10 for 2 in the 2.0Ϫ2.8 K
temperature range at the highest field of 7 T. Noteworthy is
Eur. J. Inorg. Chem. 2003, 3167Ϫ3178
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3171

T. K. Paine, E. Rentschler, T. Weyhermüller, P. Chaudhuri
FULL PAPER
N
6 4 2 3 2 3 2 2 2
Figure 3. (a) Molecular structure of [Ni L (µ -OHMe)(µ -OMe) (µ -OH) (MeOH) (MeCN) ] (4); (b) a view of 4 highlighting the coordi-
nation spheres of the hexanuclear nickel core
a downward tendency of magnetisation at low temperatures was used to obtain the good quality of fit shown in Fig-
and low fields, presumably due to the population of the ure 4, suggesting that any zero-field splitting effects are
lower component of the multiplet split in the magnetic fi- small. As the value of J is very small, we checked its influ-
elds.
ence on the quality of the simulation by keeping J ϭ 0
The nature of the field dependence of the susceptibility (fixed). The good quality of the simulation shown in Fig-
in the present system is extreme. The first-order Zeeman ure 4 could not be reproduced, indicating the importance,
splitting of the ground state S ϭ 4 results in nine separate albeit very small, of J for the fit procedure. The spin ladder
energy levels. Thus, in the absence of zero-field splittings, calculated by the evaluated exchange parameters exhibits
the energy span of the ground state is 8gβH, so that at 1 T clearly that the energy separation between low-lying states
Ϫ1
the ground state Zeeman levels span several wave numbers. is very small (Ͻ 1 cm ), yielding an intermixing of the two
It is not possible to calculate the ground state in the form states with S ϭ 4 and S ϭ 5. This is also corroborated with
t
t
of an S value; S is not a good quantum number to describe the magnetisation measurements at 1, 4, and 7 T, shown in
t
the ground state, but rather M . However, the magnetic Figure 6. The saturation behaviour observed at 7 T with a
S
analysis of 1 and 2 unambiguously demonstrates the overall magnetisation value of ca. 4.5 unambiguously shows the
ferromagnetic nature of the exchange interactions yielding dominant nature of the ferromagnetic coupling within the
‘
‘high-spin’’ ground states of S Ͼ 3.0.
pentanuclear cluster, the nature of which is a mixture of
t
The plot of µeff vs. T (Figure 4) for 3 shows the exchange S ϭ 4 and S ϭ 5 states.
t
t
II
coupling within the pentanuclear Ni complex to be pre-
Variable-temperature magnetic data for 4 (Figure 4) exhi-
dominantly ferromagnetic in nature. The effective magnetic
moment µ shows a maximum value of 8.52 µ (χ ·T ϭ
.079 cm ·K·mol ) at 15 K and then decreases with
decreasing temperature, reaching a value of 7.98 µ
χ ·T ϭ 7.96 cm ·K·mol ) at 5 K. The maximum µeff
value of 8.52 µ clearly shows that both the five- and the
six-coordinate nickel centres present in 3 are paramagnetic
with SNi ϭ 1. The maximum value of 8.52 µ is significantly
smaller than the spin-only value of 10.95 µ expected for
five ferromagnetically coupled Ni ions. The data were fit-
ted by use of a two-J model based on the model shown in
Scheme 3, as two different ranges of NiϪOϪNi angles of
bit a steady decrease of µ from 7.65 µB (χ ·T ϭ 7.308
cm ·K·mol ) at 290 K to 2.90 µB (χ ·T ϭ 1.047
cm ·K·mol ) at 2 K, which is indicative of an overall anti-
ferromagnetic interaction within the hexanuclear nickel()
complex 4. That all five- and six-coordinate nickel()
centres are paramagnetic with SNi ϭ 1 is clearly attested to
by the observed magnetic moment of µeff ϭ 7.65 µB at
90 K. In accordance with the structural parameters, a
three-J model, shown in Scheme 4, can be envisaged for the
analysis of magnetic data.
eff
M
eff
3
B
M
3
Ϫ1
M
Ϫ1
9
3
Ϫ1
B
3
Ϫ1
(
M
B
B
2
B
II
The data were initially fit by use of g (fixed), J, JЈ, and
JЈЈ, which yielded a good fit (not shown) with J ϭ Ϫ1.16
cm , JЈ ϭ Ϫ9.11 cm , JЈЈ ϭ Ϫ6.13 cm , g ϭ 2.269
fixed). To avoid overparametrisation, we did not use the
9
4Ϫ96° and 100Ϫ103° prevail in 3.
Ϫ1
Ϫ1
Ϫ1
The spin-Hamiltonian used is
(
ˆ
ˆ ˆ ˆ ˆ ˆ ˆ ˆ ˆ ˆ ˆ ˆ ˆ
zero-field splitting parameter D for the simulations. The
H ϭ Ϫ2J(S ·S ϩ S ·S ϩ S ·S ϩ S ·S ) Ϫ2JЈ(S ·S ϩ S ·S ).
1 2 2 3 4 5 3 5 1 3 3 4
above solution was discarded, as the NiϪOϪNi angles re-
The best fit obtained with the parameter set JЈ ϭ ϩ6.5 lated to JЈЈ are smaller than 90° and a small ferromagnetic
Ϫ1
Ϫ1
cm , J ϭ Ϫ0.24 cm , g ϭ 2.32 is shown as a solid line coupling could be expected for the path. This suggested that
in Figure 4. No zero-field splitting parameter D, TIP or PI a better fitting model should be used, so we again fitted the
3172
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 3167Ϫ3178

Polynuclear Nickel() Complexes
FULL PAPER
˚
LN
Table 4. Selected bond lengths [A] and angles [deg] for [Ni
6
4
(µ
2
-OHMe)(µ
3
-OMe)
2
(µ
3
-OH)
2
(MeOH)
2
(MeCN)
2
] (4)
Ni(1)ϪO(15)
Ni(1)ϪO(35)#1
Ni(1)ϪN(8)
Ni(2)ϪO(70)
Ni(2)ϪO(10)#1
Ni(2)ϪO(60)#1
Ni(3)ϪO(21)
Ni(3)ϪO(80)
Ni(3)ϪN(28)
1.981(4)
2.030(4)
2.121(5)
2.031(4)
2.073(4)
2.224(9)
1.989(4)
2.075(5)
2.101(5)
Ni(1)ϪO(1)
Ni(1)ϪO(70)
Ni(2)ϪO(10)
Ni(2)ϪO(1)
Ni(2)ϪN(50)
Ni(2)ϪO(60)
Ni(3)ϪO(35)
Ni(3)ϪO(10)
2.017(4)
2.031(4)
2.028(4)
2.054(4)
2.086(6)
2.233(8)
2.038(4)
2.097(4)
2.140(4)
[
a]
Ni(3)ϪO(70)#1
O(15)ϪNi(1)ϪO(1)
O(1)ϪNi(1)ϪO(35)#1
O(1)ϪNi(1)ϪO(70)
O(15)ϪNi(1)ϪN(8)
O(35)#1ϪNi(1)ϪN(8)
177.20(16)
88.39(15)
81.38(16)
90.54(18)
165.41(18)
O(15)ϪNi(1)ϪO(35)#1
O(15)ϪNi(1)ϪO(70)
O(35)#1ϪNi(1)ϪO(70)
O(1)ϪNi(1)ϪN(8)
88.84(15)
97.86(17)
83.45(15)
92.25(17)
111.06(18)
O(70)ϪNi(1)ϪN(8)
O(10)ϪNi(2)ϪO(70)
O(70)ϪNi(2)ϪO(1)
O(70)ϪNi(2)ϪO(10)#1
O(10)ϪNi(2)ϪN(50)
O(1)ϪNi(2)ϪN(50)
167.39(16)
80.50(15)
84.75(15)
96.33(19)
101.39(19)
O(10)ϪNi(2)ϪO(1)
100.19(16)
82.67(18)
96.87(17)
95.86(19)
161.6(2)
O(10)ϪNi(2)ϪO(10)#1
O(1)ϪNi(2)ϪO(10)#1
O(70)ϪNi(2)ϪN(50)
O(10)#1ϪNi(2)ϪN(50)
O(10)ϪNi(2)ϪO(60)
O(1)ϪNi(2)ϪO(60)
N(50)ϪNi(2)ϪO(60)
86.0(2)
164.1(2)
92.4(3)
O(70)ϪNi(2)ϪO(60)
O(10)#1ϪNi(2)ϪO(60)
90.3(2)
69.2(2)
O(21)ϪNi(3)ϪO(35)
O(35)ϪNi(3)ϪO(80)
O(35)ϪNi(3)ϪO(10)
O(21)ϪNi(3)ϪN(28)
O(80)ϪNi(3)ϪN(28)
O(35)ϪNi(3)ϪO(70)#1
O(10)ϪNi(3)ϪO(70)#1
N(28)ϪNi(3)ϪO(10)
171.31(16)
87.87(18)
83.68(16)
94.36(18)
90.36(18)
80.60(15)
81.52(15)
98.46(17)
O(21)ϪNi(3)ϪO(80)
O(21)ϪNi(3)ϪO(10)
O(80)ϪNi(3)ϪO(10)
O(35)ϪNi(3)ϪN(28)
O(21)ϪNi(3)ϪO(70)#1
O(80)ϪNi(3)ϪO(70)#1
N(28)ϪNi(3)ϪO(70)#1
96.2(2)
91.09(18)
168.10(17)
93.27(18)
91.79(16)
88.86(16)
173.85(18)
Ni(1)ϪO(70)ϪNi(3)#1
Ni(2)ϪO(70)ϪNi(1)
Ni(1)#1ϪO(35)ϪNi(3)
Ni(2)#1ϪO(10)ϪNi(3)
Ni(2)ϪO(10)ϪNi(2)#1
94.93(16)
99.07(17)
98.20(17)
96.85(17)
86.98(16)
Ni(2)ϪO(70)ϪNi(3)#1
Ni(2)#1ϪO(60)ϪNi(2)
Ni(1)ϪO(1)ϪNi(2)
96.80(15)
78.6(3)
98.76(17)
144.5(2)
Ni(2)ϪO(10)ϪNi(3)
Ni(1)···Ni(2)
3.091(2)
3.929(2)
2.823(2)
3.119(2)
3.074(2)
Ni(2)···Ni(3)
Ni(2)···Ni(2)#1
Ni(2)···Ni(3)#1
Ni(1)···Ni(3)#1
[
a]
#
1 indicates the symmetry-related atom.
Ϫ1
data with use of three different exchange parameters, but tained with the parameter set g ϭ 2.295, J ϭ Ϫ1.12 cm
Ϫ1
with the following constraints: g, J, and JЈ free for reiter- and JЈ ϭ JЈЈ ϭ Ϫ9.77 cm is shown as a solid line in
ation but JЈЈ fixed between zero and a small positive value Figure 4. We note that the sign of the exchange parameter
of ϩ2.0. The quality of fit became poor, but the exchange JЈЈ is not in line with the commonly accepted NiϪOϪNi
Ϫ1
values vary negligibly between Ϫ1.79 and Ϫ1.95 cm for angle correlation; this might be due to three different com-
Ϫ1
J and between Ϫ9.0 and Ϫ8.93 cm for JЈ, indicating that petitive exchange couplings operating on Ni(2) and Ni(2)*
although JЈЈ has some influence on the quality of fit, J and making the correlation more complex. Irrespective of that,
JЈ are nearly independent of JЈЈ. We therefore started an- the two-J model suitably describes the magnetic properties
other simulation, making all four parameters free to move. of complex 4, which does not possess a well separated
The quality of fit increased dramatically, yielding the fol- ground state of S ϭ 0, as would be expected for an antifer-
t
Ϫ1
II
lowing parameters: g ϭ 2.295, J ϭ Ϫ1.15 cm , JЈ ϭ Ϫ9.87 romagnetically coupled hexanuclear Ni complex; the ex-
cm , and JЈЈ ϭ Ϫ9.66 cm . As the evaluated values for cited state of S ϭ 1 is energetically very close to the ground
Ϫ1
Ϫ1
t
JЈ and JЈЈ are nearly the same, we carried out the magnetic state, and so a non-zero magnetic moment is observed even
analysis with the constraint JЈ ϭ JЈЈ (i.e., a two-J model to at 2 K. This is also corroborated by the variable-tempera-
reduce the number of fit parameters). A very good fit ob- ture magnetisation curves at 1, 4, and 7 T (Figure 6).
Eur. J. Inorg. Chem. 2003, 3167Ϫ3178
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3173

T. K. Paine, E. Rentschler, T. Weyhermüller, P. Chaudhuri
FULL PAPER
Figure 4. Plots of µeff vs. T for solid 1Ϫ4; the solid lines represent the best fit of the data to the exchange coupling model
N
gand H L forms six-membered chelate rings and, presum-
2
N
ably as a consequence of the ring size, the ligand H L
2
coordinates meridionally. This different geometrical coordi-
nation mode seems to yield tetranuclear complexes with
S
Se
N
H L and H L , but H L results in penta- and hexanu-
2
2
2
clear complexes. Thus, our starting goal of preparing poly-
nuclear complexes with ferromagnetic exchange coupling has
been partly achieved, although the hexanuclear nickel()
complex does not belong to the class of molecules with
large spin ground states.
Scheme 2
Complexes 1 and 2 exhibit overall ferromagnetic interac-
tions, producing ground states with ‘‘large spins’’. The J
values obtained here are similar to those found for other
4ϩ
[9]
[
Ni (µ -OR) ]
cubanes reported in the literature
4
3
4
(Table 4). The coupling between the nickel() ions is very
close to the crossover point between antiferromagnetic and
ferromagnetic coupling. In all cases, bond angles of less
than ca. 98° produce ferromagnetic coupling, whilst those
that are larger give rise to antiferromagnetic coupling. This
correlation is similar to that in the bridged CuϪOϪCu sys-
[
14]
[15]
tems,
in which ferromagnetic pathways of the form
Figure 5. A plot of average NiϪOϪNi angles vs. the exchange in-
teractions (J); filled squares are literature data, triangles are from
e ЌOЌe are found to be most effective for CuϪOϪCu
ref. and open circles are from this work; the solid line is drawn angles smaller than 97.5°.
g
g
[
4]
only as a guide
Complex 2 contains the low-symmetry cubane core
4ϩ
[
Ni (µ -methoxide) ] with differing Ni···Ni distances and
4 3 4
NiϪOϪNi angles, resulting in three types of Ni O faces.
2
2
Concluding Remarks
The observation of three discrete exchange parameters for
is consistent with the lower symmetry of the cubane; the
2
The tridentate bisphenol ligands used in this work give symmetry of 2 is lower than that in the majority of the
rise to polynuclear nickel() complexes of different nu- nickel-cubanes reported in the literature. Figure 5 shows a
clearity: four, five or six, depending on the heterodonor plot of J vs. NiϪOϪNi angles for all structurally character-
atoms such as S, Se, or N. The S- and Se-containing ligands ised Ni O cubane cores, irrespective of their symmetry; it
4
4
S
Se
H L and H L coordinate facially with five-membered exhibits a large variation of J values within a small span of
2
2
chelation to a metal centre, whereas the N-containing li- NiϪOϪNi angles (90Ϫ103°). This figure also indicates that
www.eurjic.org Eur. J. Inorg. Chem. 2003, 3167Ϫ3178
3174
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Polynuclear Nickel() Complexes
FULL PAPER
4 3 4
Table 5. J as a function of the NiϪOϪNi angles for the [Ni (µ -OR) ] cubanes
Complex[a]
NiϪOϪNiav (deg)
J (cmϪ1
)
Reference
[
[
[
9a,9b]
9c,9h]
9e]
[Ni
[Ni
[Ni
4
4
4
(µ
3
(µ
3
(µ
3
-OMe)
-OH) (chta)
-OMe) (TMB)
4
(sal)
4
(EtOH)
(NO
(µ-O
4
]
97.73
99.0
93.0
7.46
Ϫ0.57
17.5
4
4
3
)
4
]
4
4
2 4
CMe) ]
100.9
[
[
[
9i]
9l]
9p]
[
[
[
Ni
Ni
Ni
4
4
4
(µ
(µ
(µ
3
3
3
-OH)
4
(tzdt)
4
(py)
4
]
95.89
03.2
17.5
2.2
1
-OMe)
4
(dbm)
4
(MeOH)
4
]
96.7
99.6
-OH)
2
(pypentO)(pym)-
(OH )]
89.9
92.9
00.5
15.0
6.7
Ϫ3.09
(
µ-OAc)
2
(NCS)
2
2
1
[
4]
[
Ni
4
4
(µ
3
3
-OMe)
-OMe)
2
4
(LH)
2
(OAc)
2
(MeOH)
2
]
95.94
8.0
1.0
Ϫ3.8
9
9
7.85
9.0
Se
[Ni
(µ
(L )
2
(MeOH)
2
(MeCN)
2
]
96.17
7.17
01.25
5.50
1.05
Ϫ4.1
this work
this work
this work
9
1
[
a]
salH ϭ salicylaldehyde; chta ϭ r-1-c-3-c-5 triaminocyclohexane; TMB ϭ 2,5-dimethyl-2,5-diisocyanohexane; dzdt ϭ 1,3-thiazolidine-
-thione; dbmH ϭ dibenzoylmethane; py ϭ pyridine; pypentO ϭ 1,5-bis[2-pyridylmethyl)amino]-3-pentanolate; pym ϭ 2-pyridylmethox-
2
ide; L ϭ a dinucleating Schiff-base ligand.
Figure 6. The field-dependent magnetisations of 1Ϫ4 as a function of temperature
the symmetry of the cubane core has a profound effect on surprising in view of the high tolerance attached in general
the magnetostructural correlation. In particular, the scat- to the positive J values (ferromagnetic interactions) evalu-
tering of the positive J values is noteworthy. This is not ated through simulation of the susceptibility data. The syn-
Eur. J. Inorg. Chem. 2003, 3167Ϫ3178
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3175

T. K. Paine, E. Rentschler, T. Weyhermüller, P. Chaudhuri
FULL PAPER
Experimental Section
Materials and Physical Measurements: Commercial grade chemi-
cals were used for synthetic purposes, and solvents were distilled
and dried before use. Fourier transform infrared spectroscopy on
KBr pellets was performed on a PerkinϪElmer 2000 FT-IR instru-
ment. Solution electronic spectra were measured on
PerkinϪElmer Lambda 19 spectrophotometer. Mass spectra were
recorded either in the EI or the ESI (in CH Cl ) modes with a
a
Scheme 3
2
2
Finnigan MAT 95 or 8200 spectrometer. Magnetic susceptibilities
of the polycrystalline samples were recorded on a SQUID magnet-
ometer (MPMS, Quantum Design) in the 2Ϫ290 K temperature
range, with an applied field of 1 T. Diamagnetic contributions were
estimated for each compound by use of Pascal’s constants.
X-ray Crystallographic Data Collection and Refinement of the Struc-
tures: Single crystals of 2, 3, and 4 were coated with perfluoropoly-
ether, picked up with glass fibres, and mounted on a Nonius
Kappa-CCD diffractometer equipped with a cryogenic nitrogen
cold stream operating at 100(2) K. Graphite monochromated Mo-
˚
Scheme 4
Kα radiation (λ ϭ 0.71073 A) was used. Intensity data were cor-
rected for Lorentz and polarisation effects. The intensity data set
of 2 and 3, were corrected for absorption by use of the program
thesis of more such distorted cubane-type NiII complexes Gaussian face-indexed, whereas that of 4 was not corrected. The
Siemens SHELXTL software package (G. M. Sheldrick, Univer-
sität Göttingen) was used for solution, refinement, and depiction
of the structure, and neutral atom scattering factors of the program
would therefore be warranted in order to resolve this prob-
lem of magnetostructural correlation.
In the literature, there are only two examples of pentanu-
were used. All structures were solved and refined by direct methods
II
clear Ni complexes in which all five metal centres are para-
and difference Fourier techniques. Non-hydrogen atoms were re-
magnetic with S ϭ 1. The first example, [Ni (en) (µ -
Ni
10g]
5
4
3
fined anisotropically, and hydrogen atoms were placed at calculated
positions and refined as riding atoms with isotropic displacement
parameters. Details of the data collection and structure refinements
[
OH) (NO ) ],
reported as early as 1981, contains a
2
2 8
planar, pentameric arrangement of nickel() ions, linked by
bridging hydroxide and nitrite ligands; unfortunately no are summarised in Table 5.
[
10a]
magnetic data were reported. In the second example,
II
CCDC-204445 (2), -204446 (3), and -204447 (4) contain the sup-
plementary crystallographic data for this paper. These data can be
obtained free of charge at www.ccdc.cam.ac.uk/conts/retriev-
ing.html [or from the Cambridge Crystallographic Data Centre, 12,
Union Road, Cambridge CB2 1EZ, UK; Fax: (internat.) ϩ44-1223/
four six-coordinate Ni ions are arranged at the apices of
a tetrahedron with the fifth nickel at the centre; the com-
pound is antiferromagnetically coupled with an admixed
ground state of S ϭ 0, 1. Complex 3 constitutes the first
example of a pentanuclear Ni complex exhibiting ferromag-
t
II
336-033; E-mail: deposit@ccdc.cam.ac.uk].
netic interactions with an intermixed ground state of S ϭ 5.0
t
Ligand Synthesis: The ligands H
thesised by the published procedures. The ligand H
2
LS [16] and H
2
LN [3,17] were syn-
and 4.0. Pentanuclear nickel() complexes with five SNi
centres still belong to the rare examples of coordination
complexes.
An apparently trivial modification of the reaction, use of a
ϭ
Se
2
L
was syn-
1
thesised by a modification of the method described in the litera-
ture.[
18]
Se
mixed solvent medium, resulted in the isolation of complex 4, 2,2Ј-Selenobis(4,6-di-tert-butylphenol) (H
2
L ): A suspension of
II
2,4-di-tert-butylphenol (61.89 g, 0.3 mol) and SeO2 (11.1 g, 0.1
mol) in conc. HCl (55 mL) was heated at 80 °C with constant stir-
a hexanuclear Ni complex, under reaction conditions other-
wise the same as those used for complex 3. The amount of
water present in the reaction mixture is decisive for the iso-
lation of complex 4. The central nickel centres, Ni(2) and
Ni(2)#, are primarily devoid of coordination to the ancillary
ring for 2 h. The mixture was then cooled, extracted with CHCl
3 ϫ 100 mL), washed with brine solution (2 ϫ 50 mL) and then
with water (100 mL), dried with Na SO and concentrated in va-
3
(
2
4
cuo. The crude brown product was then recrystallised from heptane
and MeOH (1:1) to obtain a pure white compound. Yield: 18.5 g
N 2Ϫ
N 2Ϫ
ligand [L ] . In complex 4, and also in 3, the ligand [L ]
also for the first time shows the bridging functionality
through one of its phenolate oxygens. The magnetic studies 68.69, H 8.65; found C, 68.7, H 8.7%. H NMR (CDCl ): δ ϭ 1.20
(
25%). Melting Point: 141 °C. C28
H
42SeO
2
(489.6 g/mol): calcd. C
1
3
1
3
exhibiting predominantly antiferromagnetic exchange be- (s, 18 H); 1.38 (s, 18 H); 6.25 (s, OH, 2 H); 7.24 (m, 4 H) ppm.
tween the metal centres are disappointing in view of the orig- NMR (CDCl ): δ ϭ 29.5 (s, tert-butyl); 31.6 (s, tert-butyl); 34.3 (s,
tert-butyl); 35.2 (s, tert-butyl); 117.0 (s), 125.4 (s); 129.6 (s); 135.7
s); 143.3 (s), and 151.6 (s) ppm. MS (EI): The desired molecular
C
3
inal expectation of large-spin ground states. A simple two-J
model shows good agreement with experimentally ascer-
tained low-temperature magnetic data. The sign of an ex-
(
ϩ
ion (M ϭ 490) was observed in the ratio calculated for the nat-
urally abundant isotope mixture of Se-74, 76, 77, 80, and 82.
Ϫ1
change coupling parameter JЈЈ (Ϫ9.77 cm ) cannot be cor-
related with the corresponding NiϪOϪNi angles, which are General Method of Synthesis for Complexes 1 and 2: The ligand
appreciably smaller than 90°. (1 mmol) was dissolved in a mixture of MeOH (12.5 mL) and
3176
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 3167Ϫ3178

Polynuclear Nickel() Complexes
FULL PAPER
Table 5. Crystallographic data for [Ni
4
LSe
2
(µ
-OMe)
3
-OMe)
(µ
4
(MeOH)
4
(MeCN)
2
]·3.5MeCN·0.5MeOH (2), [Ni
] (4)
5 4 3 2 2 2 2 2
LN
(µ -OH) (µ -OH ) (EtOH) ]·
N
0
.75CH
2
Cl
2
(3), and [Ni
6
L
4
(µ
2
-OHMe)(µ
3
2
3
-OH) (HOMe)
2
2
(MeCN)
2
NiII
2
NiII
3
NiII
4
4
5
6
Empirical formula
Molecular mass
Temperature
Wavelength (Mo-K
Crystal system
Space group
C
68
H
114
N
2
Ni
4
O
12Se
2
·3.5CH
3
CN·0.5CH
3
OH C80
H
110
N
4
Ni
5
O
14·0.75CH
2
Cl
2
C
85
H
118 6 6 15
N Ni O
1704.08
100(2) K
1708.96
100(2) K
1816.11
100(2) K
˚
˚
˚
0.71073 A
α
)
0.71073 A
Triclinic
0.71073 A
Monoclinic
P2 /c
Monoclinic
C2/c
a ϭ 20.6998(12) A
b ϭ 15.0379(9) A
c ϭ 28.599(2) A
β ϭ 110.98(1)°
¯
P
1
˚
˚
˚
˚
˚
Unit cell dimensions
a ϭ 15.4538(6) A
b ϭ 16.6106(8) A
c ϭ 18.2370(10) A
α ϭ 90.31(1)°
β ϭ 108.99(1)°
γ ϭ 105.71(1)°
4238.7(4); 2
a ϭ 20.6988(8) A
b ϭ 25.7220(8) A
c ϭ 17.8131(6) A
β ϭ 111.65(1)°
˚
˚
˚
˚
˚
3
Volume [A ]; Z
Density (calcd.) [Mg/m ]
8814.9(5); 4
1.288
1.153
8312.2(9); 4
1.451
1.399
3
1.335
Ϫ1
Absorption coefficient [mm ] 1.791
F(000)
1796
3606
3832
Crystal size [mm]
θ range for data collect.
Reflections collected
Independent reflect.
Absorption correction
Data/restraints/param.
0.26 ϫ 0.25 ϫ 0.11
4.09 to 28.50°
96465
21433 [R(int.) ϭ 0.0802]
Gaussian, face indexed
21321/150/873
0.19 ϫ 0.14 ϫ 0.13
2.92 to 23.00°
62882
12227 [R(int.) ϭ 0.0938]
Gaussian, face indexed
12227/36/1030
1.056
0.07 ϫ 0.04 ϫ 0.04
2.11 to 24.00°
20329
[6529 [R(int.) ϭ 0.0694]
not corrected
6529/0/519
1.016
2
Goodness-of-fit on F
1.031
Final R indices [I Ͼ 2σ(I)]
R indices (all data)
R
R
1
ϭ 0.0538; wR
ϭ 0.0851; wR
2
2
ϭ 0.1277
ϭ 0.1441
R
R
1
ϭ 0.0509; wR
ϭ 0.0710; wR
2
ϭ 0.1287
ϭ 0.1416
R
R
1
ϭ 0.0587; wR
ϭ 0.1069; wR
2
ϭ 0.1193
ϭ 0.1402
1
1
2
1
2
MeCN (2.5 mL). LiOMe (1  methanolic solution, 2 mL, 2 mmol)
was added, and the solution was stirred for 5 min. NiCl ·6H
0.24 g, 1 mol) was added to that yellow solution, and the resulting 3448 cm (br), 2918 (s), 1475 (s), 1266 (s), 1057, 802, 453.
green solution was stirred for 5 min and then kept for slow evapor-
0.11 g (49%). C85
118 6 6
H N O15Ni (1816.04): calcd. C 56.22, H 6.55,
2
2
O
N 4.63, Ni 19.39; found C 53.3, H 6.3, N 3.9, Ni 19.2. IR: ν˜ ϭ
Ϫ1
(
ation of solvent under air. A green crystalline compound sepa- Acknowledgments
rated overnight.
Financial support from the DFG, the Max-Planck Society and
Fonds der Chemischen Industrie is gratefully acknowledged.
(1450.54): calcd. C 56.31, H 7.92, S 4.42, N Thanks are due to Mrs. H. Schucht, Mrs. R. Wagner, and Mr. A.
Göbels for skilful technical assistance.
LS
[Ni
4
2 4 4 2
(µ-OMe) (MeOH) (MeCN) ] (1): Yield: 0.42 g (58%).
68 114 2 4 2
C H S O12Ni N
1
.93, Ni 16.18; found C 55.7, H 7.9, S 4.3, N 2.0, Ni 15.7%. IR:
Ϫ1
ν˜ ϭ 3423 cm (br), 2957Ϫ2818 (s), 2315Ϫ2253, 1432 (s), 1298 (s),
047 (s), 829, 731.
[
1]
For selected examples see: [1a] L. Que Jr., W. Tolman, Angew.
Chem. 2002, 114, 1160; Angew. Chem. Int. Ed. 2002, 41, 1114.
1
[
Ni (µ-OMe) (MeOH)
LSe
4
2
4
4
(MeCN)
2
]
(2): Yield: 0.34 g (44%).
[1b]
V. Mahadevan, R. J. M. Klein Gebbink, T. D. P. Stack, Cur.
C
68
H114Se
2
O12Ni
4
N
2
(1544.34): calcd. C 52.89, H 7.44, Ni 15.20;
[1c]
Opin. Chem. Biol. 2000, 4, 228.
J. Du Bois, T. J. Mizoguchi,
Ϫ1
[1d]
found C 49.6, H 7.3, Ni 14.9%. IR: ν˜ ϭ 3404 cm (br), 2954Ϫ2821
s), 2315Ϫ2253, 1429 (s), 12989 (s), 1047 (s), 828, 724.
S. J. Lippard, Coord. Chem. Rev. 2000, 200Ϫ202, 443.
Mul-
(
ticopper Oxidases (Ed.: A. Messerschmidt), World Scientific,
Singapore, 1997. [ Bioinorganic Chemistry of Copper (Eds.:
1e]
[
Ni (µ -OH) (µ-H O) (EtOH) ]·0.75CH Cl (3): The ligand
LN
5
4
3
2
2
2
2
2
2
K. D. Karlin, Z. Tyeklar), Chapman & Hall, New York, 1993.
[2]
[2a]
(1 mmol, 0.30 g) was dissolved in EtOH (15 mL). LiOEt (1  meth-
See for example:
‘‘Journal of Solid State Chemistry’’, Vol.
[2b]
anolic solution, 2 mL, 2 mmol) was added, and the solution was
stirred for 5 min. NiCl ·6H O (0.24 g, 1 mol) was added to the
yellow solution, and the resulting light green solution was stirred
under air. The separated yellowish green microcrystalline com-
pound was filtered and dried. X-ray-quality crystals were grown
159, No. 2, July 2001 Ϫ a tribute to Olivier Kahn.
Magnetic
Molecular Materials (Eds.: D. Gatteschi, O. Kahn, J. S. Miller,
F. Palacio), Kluwer, Dordrecht, The Netherlands, 1991.
T. Weyhermüller, T. K. Paine, E. Bothe, E. Bill, P. Chaudhuri,
Inorg. Chim. Acta 2002, 337, 344.
S. Mukherjee, T. Weyhermüller, E. Bothe, K. Wieghardt, P.
Chaudhuri, Eur. J. Inorg. Chem. 2003, 863.
S. Mukherjee, T. Weyhermüller, E. Bothe, P. Chaudhuri, Eur.
J. Inorg. Chem. 2003, 1956.
R. Siefert, T. Weyhermüller, P. Chaudhuri, J. Chem. Soc., Dal-
ton Trans. 2000, 4656.
V. V. Pavlishchuk, F. Birkelbach, T. Weyhermüller, K. Wiegh-
2
2
[
[
3]
4]
from a solvent mixture (1:2) of CH
75%). C80 ·0.75CH Cl
.58, N 3.28, Ni 17.17; found C 56.9, H 6.7, N 3.2, Ni 17.4. IR:
2
Cl
2
and EtOH. Yield: 0.32 g
(
H
140
N
4
O
14Ni
5
2
2
(1708.96): calcd. C 56.75, H
[5]
[6]
6
Ϫ1
ν˜ ϭ 3510 cm (br), 2915 (s), 1477 (s), 1263 (s), 1162, 801, 648, 451.
L N
[
N i
6
4
( µ - M e O H ) ( µ
3 2 3 2 2
- O M e ) ( µ - O H ) ( M e O H ) -
[
7]
(MeCN)
2
] (4): The ligand (0.5 mmol, 0.15 g) was treated with Na-
ardt, P. Chaudhuri, Inorg. Chem. 2002, 41, 4405.
OMe (2 mL, 0.5  methanolic solution) in a solvent mixture of
MeCN (3 mL) and MeOH (15 mL). NiCl ·6H O (0.12 g, 0.5 mmol)
was added, and the light green solution was kept at room tempera-
ture. A green crystalline complex separated out overnight. Yield:
[8]
[8a]
Selected examples:
K. S. Murray, Adv. Inorg. Chem. 1995,
[8b]
2
2
43, 261.
V. V. Pavlishchuk, S. V. Kolotilov, A. W. Addison,
M. J. Prushan, D. Schollmeyer, L. K. Thompson, E. A. Go-
reshnik, Angew. Chem. Int. Ed. 2001, 40, 4734. [ J. M. Cle-
8c]
Eur. J. Inorg. Chem. 2003, 3167Ϫ3178
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3177

T. K. Paine, E. Rentschler, T. Weyhermüller, P. Chaudhuri
FULL PAPER
Liu, J. Jubieta, Polyhedron 1988, 7, 1397. [10f] M. Knege, G.
mente-Juan, H. Andres, J. J. Borr a´ s-Almenar, E. Coronado, H.
U. Güdel, M. Eebersold, G. Kearly, H. Büttner, M. Zolliker, J.
Am. Chem. Soc. 1999, 121, 10021. [ A. Escuer, R. Vincente,
S. B. Kumar, F. A. Mautner, J. Chem. Soc., Dalton Trans. 1998,
Henkel, Z. Naturforsch., Teil B 1987, 42, 1121. [
10g]
A. J. Fin-
8d]
ney, M. A. Hitchman, C. L. Raston, G. L. Rowbottom, A. H.
F. Cecconi, C. A.
Ghilardi, S. Midollini, A. Orlandini, A. Vacca, Inorg. Chem.
White, Aust. J. Chem. 1981, 34, 2139. [
10h]
[
8e]
3
2
473.
H. Adams, S. Clunas, D. E. Fenton, Chem. Commun.
[
8f]
[10i]
002, 418.
M. Du, X.-H. Bu, Y.-M. Guo, L. Zhang, D.-Z.
Commun. 2000, 3, 276.
S.-M. Peng, Inorg. Chem. 2002, 41, 4096.
W. Zhang, Q. Wang, X. Gao, P. Lin, Chem. Commun. 1998,
C-Y. Yeh, Y. L. Chiang, G.-H. Lee,
[
8g]
[10j]
Liao, J. Ribas, Chem. Commun. 2002, 1478.
T. K. Kar-
T. Sheng, X. Wu,
makar, S. K. Chandra, J. Ribas, G. Mostafa, T. H. Lu, B. K.
10k]
Ghosh, Chem. Commun. 2002, 2364.
263. [
Chem. Soc., Dalton Trans. 2002, 3933.
[11] [11a]
H. Adams, S. Clunas, D. E. Fenton, D. N. Towers, J.
[
9] [9a]
[9b]
J. E. Andrew, A. B. Blake, J. Chem. Soc., A 1969, 1456.
J. A. Barnes, W. E. Hatfield, Inorg. Chem. 1971, 10, 2355.
[9c]
P. Woodward, L. F. Dahl, E. W. Abel, B. C. Crosse, J. Am.
B. Aurivillius, Acta Chem. Scand., Ser. A 1977, 31, 501. [ J.
A. Bertrand, C. Marabella, D. G. Vanderveer, Inorg. Chim.
Acta 1978, 26, 113. [ W. L. Gladfelter, M. W. Lynch, W. P.
Schaefer, D. N. Hendrickson, H. B. Grey, Inorg. Chem. 1981,
9d]
Chem. Soc. 1965, 87, 5251.
Inorg. Chem. 1969, 8, 1304. [
Inorg. Chem. 1983, 22, 2895.
[11b]
11c]
[11d]
F. A. Cotton, B. H. Winquist,
G. E. Lewis, C. S. Kraihanzel,
A. Cornia, A. C. Fabretti, D.
9e]
Gatteschi, G. Palyi, E. Rentschler, O. I. Shchegolikhina, A. A.
[
9f]
[11e]
20, 2390.
F. Paap, E. Bouwman, W. L. Driessen, R. A. G.
Zhadanov, Inorg. Chem. 1995, 34, 5383.
E. K. Brechin, W.
de Graaf, J. Reedijk, J. Chem. Soc., Dalton Trans. 1985, 737.
Clegg, M. Murrie, S. Parsons, S. J. Teat, R. E. P. Winpenny, J.
Am. Chem. Soc. 1998, 120, 7365.
[12] [12a]
[
9g]
K. Bizilj, S. G. Hardin, B. F. Hoskins, P. J. Oliver, E. R. T.
Tiekink, G. Winter, Aust. J. Chem. 1986, 39, 1035. [ P. D.
Boyd, R. L. Martin, G. Schwarzenbach, Aust. J. Chem. 1988,
9h]
R. E. P. Winpenny, Adv. Inorg. Chem. 2001, 52, 1 and refer-
ences cited therein. [
12b]
E. Diamantopoulou, C. P. Raptopou-
[
9i]
4
1, 1449.
L. Ballester, E. Coronado, A. Guti e´ rrez, A.
lou, A. Terzis, V. Tangoulis, S. P. Perlepes, Polyhedron 2002, 21,
2117. [
12c]
J. Faus, F. Lloret, M. Julve, J. M. Clemente-Juan, M.
Monge, M. F. Perpin a´ n, E. Pinilla, T. Rico, Inorg. Chem. 1992,
3
[
9j]
1, 2053.
A. J. Atkins, A. J. Blake, M. Schröder, J. Chem.
C. Munoz, X. Solans, M. Font-Bardia, Angew. Chem. Int. Ed.
[9k]
[12d]
Soc., Chem. Commun. 1993, 1662.
chin, A. Codron, R. O. Gould, C. M. Grant, S. Parsons, J. M.
A. J. Blake, E. K. Bre-
Engl. 1996, 35, 1485.
omb, Inorg. Chem. 1985, 24, 1201. [
Evans, H. U. Güdel, Angew. Chem. Int. Ed. 2001, 40, 1957.
[13] [13a]
I. G. Dance, M. L. Scudder, R. Sec-
12e]
M. Murrie, H. Stoeckli-
Rawson, R. E. P. Winpenny, J. Chem. Soc., Chem. Commun.
[
9l]
1
995, 1983.
Christou, Inorg. Chem. 1995, 34, 4167.
Rentschler, A. Caneschi, D. Gatteschi, Inorg. Chim. Acta 1996,
M. A. Halcrow, J.-S. Sun, J. C. Huffman, G.
F. Birkelbach, C. Krebs, V. Staemmler, unpublished,
Bochum, Germany, 1997.
Chim. Ital. 1993, 123, 231.
[9m]
[13b]
M. S. E. Fallah, E.
D. Gatteschi, L. Pardi, Gazz.
[
9n]
[14]
[15]
[16]
247, 231.
A. Escuer, M. Font-Bardi a´ , S. B. Kumar, X. Sol-
W. E. Hatfield, Comments, Inorg. Chem. 1981, 1, 105.
A. P. Ginsberg, Inorg. Chim. Acta Rev. 1971, 5, 45.
S. D. Pastor, J. D. Spivack, L. P. Steinhuebel, J. Heterocycl.
Chem. 1984, 21, 1285.
N. N. Timosheva, A. Chandrasekharan, R. O. Day, R. R.
Holmes, Inorg. Chem. 1998, 37, 4945.
[9o]
ans, R. Vicente, Polyhedron 1999, 18, 909.
M. L. Tong, H.
K. Lee, S.-L. Zheng, X.-M. Chen, Chem. Lett. 1999, 1087. [
J. M. Clemente-Juan, B. Chansou, B. Donnadieu, J.-P. Tuchag-
ues, Inorg. Chem. 2000, 39, 5515.
10] [10a]
9p]
[
[
17]
18]
[
V. Tangoulis, C. P. Raptopoulou, A. Terzis, E. G. Bakal-
bassis, E. Diamantopoulou, S. P. Perlepes, Inorg. Chem. 1998,
T. Thompson, S. D. Pastor, G. Rihs, Inorg. Chem. 1999, 38,
4163.
[
10b]
37, 3142.
S.-J. Shieh, C.-C. Chou, G.-H. Lee, C.-C. Wang,
[10c]
[19] [19a]
L. A. P. M. Hall, Polyhedron 1990, 9, 2575. [19b] V. Ta n g -
S.-M. Peng, Angew. Chem. Int. Ed. Engl. 1997, 36, 56.
S. Velazquez, T. F. Baumann, M. M. Olmsted, H. Hope, A. G.
C.
oulis, C. P. Raptopoulou, S. Paschalidou, A. E. Tsohos, E. G.
Bakalbassis, A. Terzis, S. P. Perlepes, Inorg. Chem. 1997, 36,
5270.
Barrett, B. M. Hoffman, J. Am. Chem. Soc. 1993, 115, 9997.
[
10d]
D. Fenske, H. Krautscheid, M. Müller, Angew. Chem. Int.
Ed. Engl. 1992, 31, 321. [
10e]
B. K. Koo, E. Block, H. Kang, S.
Received March 27, 2003
3178
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 3167Ϫ3178