Mono- and Dinuclear Complexes of a Flexible Schiff Base Ligand - Crystal Structures of a Bishelicate and Two Acentric Monohelicates

Article abstract of DOI:10.1002/ejic.200300248

Neutral Ni^II, Pd^II, Cu^II, Ag^I, Zn^II, Cd^II, and Pb^II complexes of N,N'-bis(2-tosylaminobenzylidene)-1,4-diaminobutane (H₂SB) have been prepared. The Schiff base seems t

Full text of DOI:10.1002/ejic.200300248

FULL PAPER
Mono- and Dinuclear Complexes of a Flexible Schiff Base Ligand ؊ Crystal
Structures of a Bishelicate and Two Acentric Monohelicates
[a]
[a]
[a]
[a]
´
´
´
Jesus Sanmartın,* Ana M. Garcıa-Deibe, Manuel R. Bermejo, Fernando Novio,
Debora Navarro,[a] and Matilde Fondo[a]
Keywords: Helical structures / N ligands / Schiff bases / Supramolecular chemistry / Transition metals
Neutral NiII, PdII, CuII, AgI, ZnII, CdII, and PbII complexes
of N,NЈ-bis(2-tosylaminobenzylidene)-1,4-diaminobutane
(H2SB) have been prepared. The Schiff base seems to exist
in a helically coiled state around nickel, copper and zinc ions,
geometries involving the four N atoms of the dianionic li-
gand. X-ray structural characterisation also shows that this
ligand behaves as a N2+N2 donor in the [4+4] bishelicate
Ni2(SB)2·5MeCN. Secondary interactions between metal
but not in Pd(SB)·3H2O·MeCN, Ag2(SB)·2H2O, Cd(SB)·3H2O, centres and one of the O atoms of each tosyl group are de-
and Pb(SB)·3H2O, the metal ions of which seem to show
planar geometries. The determination of the absolute struc-
tures of Cu(SB)·MeCN and Zn(SB)·2MeCN, which crystallise
as non-centrosymmetric racemic compounds, shows that
these metal(II) ions assume distorted tetrahedral coordination
tected in these three crystal structures, being especially in-
tense in the dimeric one.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2003)
Introduction
similarities between these two symmetric N-tosyl-substi-
tuted Schiff bases, which contain (CH2)3 and
Among other reasons, helical metal complexes[1Ϫ4] have CH2CH(OH)CH2 groups, respectively, as spacers
attracted physicists’ and chemists’ attention due to their po- (Scheme 1), their behaviour in their corresponding helical
tential optical activity.[5] As achiral ligands are unable to ZnII complexes is either binucleating (N2ϩN2 donor) or
induce predetermined chirality, helicates crystallise mononucleating (N4 donor), respectively. Another ligand
as racemic compounds or as mechanical mixtures of with a three-membered spacer, although a more rigid one,
enantiopure crystals (racemic mixtures or conglo- in the form of H2SB3 (Scheme 1), only yielded racemic
merates).[6Ϫ7] Of these two types of crystalline racemates, monohelicates.[11]
conglomerates are far easier to resolve, but much rarer.
In an effort to understand the behaviour of this type of
The obtaining of non-centrosymmetric inorganic materi- helical thread, we wished to explore how the higher flexi-
als with nonlinear optical properties is undoubtedly a chal- bility of a (CH2)4 spacer might influence the coordination
lenge in chemistry.[8] Since the inverse of a left-handed helix behaviour and spatial arrangement of the potential helicand
is a right-handed one, a helical compound may crystallise H2SB (‘‘S’’ arranged, for a classic bisbidentate conduct, in
with an inversion centre if both types of handedness appear Scheme 1) in several main group and transition (first and
in a crystal. Thus, with achiral ligands, the heterochiral in- second row) metal complexes.
teraction required to form racemic compounds takes place
in centrosymmetric space groups, the overwhelming ma-
Results and Discussion
jority, whereas the minority homochiral crystallisation re-
quired to obtain conglomerates exclusively occurs in non-
centrosymmetric space groups.
Mass and IR Spectroscopic Data
We have recently reported the crystal structures of the
two compounds Zn2(SB1)2·1.5H2O·MeCN[9] (a typical ra-
cemic compound) and Λ-Zn(SB2)·H2O,[10] which homochir-
ally crystallises as a conglomerate. Despite the intrinsic
The mass spectra of these complexes exhibit medium- to
high-intensity signals, with satisfactory isotopic profiles, at-
tributable to [M(SB) ϩ H]ϩ species, as expected for mono-
nuclear and monomeric compounds. This kind of species is
more in accordance with a solvated nature than with
coordinating behaviour of the water molecules present in
these complexes. With respect to the ESI MS spectrum of
the AgI complex, a group of peaks assignable to [Ag2(SB)
ϩ H]ϩ species could be notable, since it seems to corrobor-
ate the Ag2(SB)·2H2O formula found by elemental analysis.
[a]
´
´
´
Dpto. de Quımica Inorganica, Facultade de Quımica,
Universidade de Santiago de Compostela,
Campus Sur, Santiago de Compostela, 15782 Spain
Fax: (internat.) ϩ34-(0)981-597525
E-mail: qisuso@usc.es
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
Eur. J. Inorg. Chem. 2003, 3905Ϫ3913
DOI: 10.1002/ejic.200300248
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3905

´
J. Sanmartın et al.
FULL PAPER
Single-Crystal X-ray Diffraction Studies
Cu(SB)·MeCN and Zn(SB)·2MeCN
Slow evaporation of saturated acetonitrile solutions of
Cu(SB)·4H2O and Zn(SB)·H2O yielded brown and yellow
X-ray-quality
crystals
of
Cu(SB)·MeCN
and
Zn(SB)·2MeCN, respectively. The crystallographic analyses
reveal that in both asymmetric units, a solvated acetonitrile
molecule (with 0.5 and 1 occupancy sites, respectively) co-
exists with half a molecule of the symmetric helical complex
M(SB) (Mϭ Cu or Zn), the metal ion of which is situated
on a twofold axis positioned along z. The whole complex
molecules can be generated by the same symmetry oper-
ation (Ϫx, Ϫy, z). All the atoms so generated have been lab-
elled with #1
.
Although both helical enantiomers (Λ and ∆) are present
in the unit cell, only the molecular structure of the single-
stranded helix corresponding to Λ-Cu(SB) is represented in
Figure 1, whereas ∆-Zn(SB) is shown in Figure 2, with the
atom labelling schemes used in Table 1. This table lists the
most significant bond lengths and angles.
Scheme 1. Schematic representation of the N-tosyl-substituted
Schiff bases H2SB1, H2SB2, H2SB3, and H2SB (with labelling
1
scheme for H NMR studies)
It may also be worth mentioning that no peaks with inten-
sities higher than 4% can be detected in the spectra of these
complexes in those regions in which dimeric species could
appear, even in the case of the dimeric NiII complex. This
prevents the use of the ESI-MS technique to determine
whether these complexes are of the type M(SB) or M2(SB)2.
The IR spectra of these complexes contain absorptions
consistent with the presence of the N4 donor ligand. The
CN stretching frequencies show shifts to lower wavenumber
values with respect to those observed for the free ligand.[12]
The slight shifts of the ν(CϭN) bands to lower wavenumber
(in the 1Ϫ15 cmϪ1 range) contrasts with the substantial
Figure 1. ORTEP view of Λ-Cu(SB), contained in the
Cu(SB)·MeCN unit cell; symmetry transformations used to gener-
#1
ate equivalent atoms:
Ϫx, Ϫy, z
The ions in both complexes are symmetrically tetra-coor-
shifts observed for the ν(CϪN) stretching frequencies dinated to the four N atoms of the Schiff base, with bond
(30Ϫ45 cmϪ1). These facts, together with the absence of the lengths in the ranges typical for this sort of complex. Both
ν(NϪH) band, are in agreement with double deprotonation coordination environments can be regarded as distorted
of the ligand, as well as with the participation of both the tetrahedra, especially that corresponding to the CuII ion, in
amide and imine N atoms in the coordination to the metal that the dihedral angles between the two NϪMϪN ter-
centres.[9Ϫ12]
minal planes (θ) are 65.79(14)° and 73.31(7)° for the CuII
Other bands attributable to νas(SO2) and νs(SO2) vi- and ZnII complexes, respectively. Moreover, one of the
bration modes also experience significant decreases in their O atoms of each tosyl group [O(2) and O(2)#1] is weakly
frequencies. This behaviour could be a sign of interaction interacting with the CuII and ZnII ions, although these
˚
between the metal centre and the O atoms belonging to the M···O distances (ca. 2.8 A) are too long to be viewed as
tosyl groups.
true coordinated bonds. The S(1)···S(1)#1 distance (about
3906
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 3905Ϫ3913

Mono- and Dinuclear Complexes of a Flexible Schiff Base Ligand
FULL PAPER
which means that the unit-cell origin may be arbitrarily
placed along this axis. In these cases, the values found for
the Flack parameter[14] [0.03(19) and 0.021(11) for the CuII
and ZnII complexes, respectively] determine a correct abso-
lute structure, as well as a closely coincident orientation of
the structure with respect to the polar axis, more than the
enantiopurity or racemic twinning evaluation related with
homochiral crystals.
Both crystal packings, which we can see in Figure 3 for
Zn(SB)·2MeCN, are based on feeble CϪH···O, CϪH/π and
π-π stacking interactions. They clearly show both types of
helical molecules, with right- and left-handedness (∆ and Λ
enantiomers, respectively) arranged on infinite sheets paral-
lel to their C faces, indicative of their non-linearity. This
heterochiral crystallisation satisfactorily explains the lack of
global optical activity exhibited by crystalline samples of
these racemic compounds in acetonitrile solution.
Ni2(SB)2·5MeCN
This crystal structure shows the double helical assembly
of this neutral dinuclear NiII complex (Figure 4), in which
the ligand assumes a N2ϩN2 bisbidentate coordinating be-
haviour. As is the case for many other homotopic double-
stranded helicates, this one crystallises as a racemate in the
most frequently occurring space group for metallohelicates:
the centrosymmetric C2/c.[3] The X-ray structural charac-
terisation reveals that its asymmetric unit contains half the
bishelicate, with a crystallographically imposed twofold
symmetry, which allows the other half molecule to be gener-
ated by use of the symmetry operation Ϫx, y, Ϫz ϩ 1/2. All
Figure 2. ORTEP view of ∆-Zn(SB), contained in the Zn(SB)·
2MeCN unit cell; symmetry transformations used to generate
#1
equivalent atoms:
Ϫx, Ϫy, z
˚
4.34 and 4.62 A for the copper and zinc complex, respec-
tively) could in both cases be regarded as a rough estimate
of the pitch.[3] As also occurs in other related
monohelicates,[9Ϫ12] if we consider the whole ligand, the
wrapping angle would exceed 360°.
The chemical species concerned here, Cu(SB)·MeCN and
Zn(SB)·2MeCN, are intrinsically chiral due to their helicity,
crystallising in the non-centrosymmetric space group Fdd2
of the orthorhombic system, belonging to the mm2 crystal
class. Although this class is not qualified as chiral or enan-
tiomorphic, it does present a polar axis direction (z),[13]
the atoms related in this way have been denoted with #1
.
Additionally, two and a half molecules of solvated aceto-
nitrile are also present, with some disorder, in the asymmet-
ric unit.
Table 1. Selected bond lengths and angles for Ni2(SB)2·5MeCN, Cu(SB)·MeCN, and Zn(SB)·2MeCN
Ni2(SB)2·5MeCN[a]
Cu(SB)·MeCN[b]
Zn(SB)·2MeCN[b]
Bond lengths
Ni(1)ϪN(21)
Ni(1)ϪN(11)
Ni(1)ϪN(12)
Ni(1)ϪN(22)
N(11)ϪC(108)
N(11)ϪS(1)
N(12)ϪC(114)
N(21)ϪS(2)
N(21)ϪC(208)
N(22)ϪC(214)
Bond angles
1.965(6) Cu(1)ϪN(1)
1.965(3)
1.965(3)
1.961(4)
1.961(4)
1.393(5)
1.612(3)
1.273(6)
1.455(6)
1.436(3)
1.447(3)
Zn(1)ϪN(1)
Zn(1)ϪN(1)#1b
Zn(1)ϪN(2)
Zn(1)ϪN(2)#1b
N(1)ϪC(8)
1.978(2)
1.978(2)
2.009(2)
2.009(2)
1.405(3)
1.605(2)
1.274(4)
1.475(4)
1.444(2)
1.444(2)
1.973(8) Cu(1)ϪN(1)#1b
1.987(6) Cu(1)ϪN(2)
2.030(5) Cu(1)ϪN(2)#1b
1.418(11) N(1)ϪC(8)
1.588(9) N(1)ϪS(1)
1.316(9) N(2)ϪC(14)
1.605(5) N(2)ϪC(15)
1.408(9) S(1)ϪO(1)
1.267(9) S(1)ϪO(2)
N(1)ϪS(1)
N(2)ϪC(14)
N(2)ϪC(15)
S(1)ϪO(1)
S(1)ϪO(2)
N(11)ϪNi(1)ϪN(22)
N(11)ϪNi(1)ϪN(12)
N(11)ϪNi(1)ϪN(21)
N(12)ϪNi(1)ϪN(21)
Torsion angles
104.0(3) N(1)ϪCu(1)ϪN(1)#1b
108.2(2)
N(1)ϪZn(1)ϪN(1)#1b
119.02(13)
94.47(9)
124.55(9)
101.07(15)
91.3(3)
N(1)ϪCu(1)ϪN(2)
93.90(14) N(1)ϪZn(1)ϪN(2)
151.9(3) N(1)ϪCu(1)ϪN(2)#1b
107.0(2) N(2)ϪCu(1)ϪN(2)#1b
132.99(14) N(1)ϪZn(1)ϪN(2)#1b
100.5(2)
N(2)ϪZn(1)ϪN(2)#1b
N(12)ϪC(115)ϪC(116)ϪC(16)#1a 45.7(6)
N(22)ϪC(215)ϪC(216)ϪC(216)#1a 75.7(6)
N(2)ϪC(15)ϪC(16)ϪC(16)#1b Ϫ77.8(7)
N(2)ϪC(15)ϪC(16)ϪC(16)#1b Ϫ78.3(4)
[a]
Symmetry transformations used to generate equivalent atoms: #1a: Ϫx, y, Ϫz ϩ 1/2. [b] #1b: Ϫx, Ϫy, z.
Eur. J. Inorg. Chem. 2003, 3905Ϫ3913
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3907

´
J. Sanmartın et al.
FULL PAPER
Figure 4. ORTEP view of ∆,∆-Ni2(SB)2 contained in the
Ni2(SB)2·5MeCN unit cell, with the secondary Ni···O interactions
Figure 3. View normal to the C face of the crystal packing of
Zn(SB)·2MeCN, showing the lamination of both types of helical
enantiomers (∆ and Λ); acetonitrile molecules have been omitted
for clarity
shown as dotted lines; symmetry transformations used to generate
#1
equivalent atoms:
Ϫx, y, Ϫz ϩ 1/2
The [4 ϩ4] bishelical nature of Ni2(SB)2·5MeCN is not ligand,[9Ϫ12,16] it could be worthy of mention that these in-
common for dinuclear NiII complexes with bisbidentate li- teractions seem to be significantly shorter when the ligand
[9,16]
˚
gands, which mostly give rise to [6 ϩ 6] triple-stranded behaves as N2ϩN2-bisbidentate (about 2.5 A),
than as
helicates.[3Ϫ4,15] However, so far the X-ray evidence avail- merely N4-tetradentate (usually about 2.8 A),[9Ϫ12] as in the
˚
able indicates, this sort of N-tosyl-substituted ligand only preceding CuII and ZnII complexes (vide supra).
yields four coordinate complexes, both in double-stranded
The bishelical arrangement of Ni2(SB)2·5MeCN with me-
helicates[9,16] and monohelicates.[9Ϫ12] This even occurs with tal coordination environments that could be also regarded
metal ions that characteristically display pseudooctahedral as severely distorted N4O2 pseudooctahedral (Figure 4),
geometries, such as MnII or FeII, in that they exhibit unfam- therefore seems to be more appropriate than simple mono-
iliar tetrahedral geometries with H2SB3.[11]
nucleating N4 donor behaviour for satisfying the stereochem-
The explanation for this low coordination number may ical requirements of NiII (preferably octahedral). This tend-
lie in the bulky tosyl groups. On one hand, they appear to ency of NiII to achieve a pseudooctahedral coordination
exert such substantial steric hindrance as apparently to through secondary Ni···O contacts had also been indicated
avoid the presence of exogenous coordinating groups, sol- for some mesohelicates containing bisbidentate ligands de-
vent molecules or additional ligand units suitable for signed to provide tetrahedral coordination environments.[17]
achieving an actual hexa-coordination. On the other hand, Something similar also occurred in the case of
this uncommon tetra-coordination of the NiII ions may be Ni2(SB1)2·MeCN,[16] whereas the rigid H2SB3 (Scheme 1),
easier to understand if we also consider that significant sec- designed to be N4-mononucleating, is able to impose a dis-
II
˚
ondary Ni···O interactions (ca. 2.5 A) exist between the Ni
torted tetrahedral geometry, though also exhibiting Ni···O
[11]
˚
ion and an O atom of each adjacent tosyl group [O(2) and interactions (2.5 and 2.8 A).
In contrast, a simple N4-
O(4)]. As a result, each N4 pseudotetrahedral NiII environ- compartment can adequately fulfil the tetrahedral coordi-
ment (Figure 4) suffers a substantial seesaw-shaped distor- nation preferences of CuII ions. Unlike CuII and NiII, the
tion that the value of θ (ca. 80.15°) does not reveal at first spherical ZnII ions (d10) allow both monohelicates to be ob-
sight, but the NϪNiϪN angles do (Table 1). Although tained, with H2SB, H2SB2
or H2SB3, and also a bis-
[10]
these minor M···O contacts are typical of metal complexes helicate,[9] when interacting with H2SB1 (Scheme 1). Hence,
containing this type of N-tosyl-substituted Schiff-base while the spacer length has a significant influence on the
3908
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 3905Ϫ3913

Mono- and Dinuclear Complexes of a Flexible Schiff Base Ligand
stereochemistry of some types of helicates,[18] it does not
FULL PAPER
Finally, the CϪN distances found confirm those ob-
appear to be a crucial factor for providing single- or served in the IR spectroscopy section. Thus, the CϭNimine
double-stranded helicates with N-tosyl-substituted Schiff and C-Nimine distances show values similar to those re-
bases.
ported for the free ligand,[12] whereas the C-Namide distances
Ideally, homotopic helicates should give rise to palin- are significantly longer. The MϪN distances (Table 1) are
dromic helices; this means with constant pitch,[3] as occurs in the usual range reported for NiII, CuII, and ZnII com-
in Ni2(SB1)2·MeCN,[16] in which both (SB1)2Ϫ threads are plexes containing Namide and Nimine donor atoms.[9Ϫ12,16,20]
similarly arranged, the centroids of their benzylidene As was also suggested by the IR techniques, M···O interac-
˚
groups being at about 10 A. In principle, the ligand array tions seem to affect the SϪO bonds, since these are slightly
in Ni2(SB)2·5MeCN could appear similar to that, even also longer than those found for the free ligand.[12]
showing a minor and a major groove. However, a thorough
1H NMR Studies
inspection allows it to be established that the higher flexi-
bility of SB2Ϫ seems to tolerate a clear disparity of both
helical strands, with substantially different arrangements
Zn(SB)·H2O
(Figure 5). Thus, we can see that the C101ϪC101#1 ligand
In order to investigate the solution behaviour of
strand (shaded in Figure 5) seems more stretched than the
Zn(SB)·H2O, we have studied its 1H NMR spectra in
C201ϪC201#1 one, which shows a pronounced folding.
CDCl3 at temperatures ranging between 323 and 218 K (de-
Hence, the theoretical pitch must be longer for the first one.
posited as Electronic Supporting Information, see footnote
In fact, the distances between their benzylidene centroids
on the first page of this article). From these results, and in
view of the ligand symmetry, the observation of two signals
each for Hj and Hk (Scheme 1), must be a consequence of
the diastereotopic natures of the geminal protons of each
methylene group. This is consistent with a helical arrange-
ment,[21] as also occurred in the solid state. Since both
methylene protons are non-equivalent even at 323 K,
Zn(SB)·H2O seems to preserve its helical conformation in
chloroform solution on the NMR timescale, which is also
indicative of a strong coordination of the N atoms to the
metal centre. On the basis of the kinetic stability of this
complex, the results obtained at room temperature can be
used as a diagnostic probe for helicity.
˚
are about 10.9 and 8.6 A, respectively. Although the two
coilings are fairly different, their assembly, lacking signifi-
cant intra- or intermolecular π-π or CϪH/π interactions,
produces a symmetric, even crystallographically, homotopic
double-helix.[3] Consequently, their also different wrapping
angles could compensate for their differences, and result in
˚
an intramolecular Ni···Ni distance of about 6.8 A. This is
only slightly longer than that reported for Ni2(SB1)2·MeCN
(6.2 A).[16] Similitude in the intramolecular M···M distances
˚
with (CH2)2 and (CH2)3 bridging groups has also been ob-
served for some [6ϩ6] FeII bishelicates.[19]
Both COSY and NOESY experiments at room tempera-
ture in chloroform, acetonitrile or dimethyl sulfoxide solu-
1
tions allow a full assignment of the H NMR signals (see
Experim. Section). They clearly show AB spin behaviour of
the methylene protons on the NMR timescale, since those
protons situated near to equatorial positions with respect
to the C(15)N(2)C(14) or C(15#1)N(2#1)C(14#1) planes,
Hjeq and Hkeq, are observed at higher field than those situ-
ated close to axial positions (Hjax and Hkax).
A NOESY experiment (Figure 6) shows coupling of Hb
with Hc and Hjeq (but not with Hjax), as well as between Hg
and Hf. Consequently, these intramolecular H···H distances
˚
should be lower than ca. 3.1 A in solution. The X-ray crys-
tal studies determined that the Hb···Hc, Hb···Hjeq and
˚
Hg···Hf distances in the solid state are in the 2.2Ϫ2.7 A
˚
range, whereas Hb···Hjax is higher (3.1 A). These facts seem
to indicate that the spatial arrangement of the helical ligand
in the zinc complex is analogous, both in the solid state and
in solution (even in a coordinating solvent such as dmso).
Pb(SB)·3H2O
A NOESY experiment was a key element in the full as-
signment of the proton signals (deposited as Electronic
Supporting Information). Since the Hb and Hi signals are
two clear references in the spectrum, the HiϪHh, HbϪHc
and HbϪHj couplings allow the unequivocal identification
Figure 5. Two opposite views of a compact ball scheme for
Ni2(SB)2·5MeCN with the helical axis parallel to left margin; the
more stretched ligand strand (C101ϪC101#1) is shaded; the intra-
molecular distances between those ring centroids signed by arrows
˚
˚
are: 8.6 A (top) and 10.9 A (bottom)
Eur. J. Inorg. Chem. 2003, 3905Ϫ3913
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3909

´
J. Sanmartın et al.
FULL PAPER
cal arrangement of SB2Ϫ in this complex. Its anomalous
stoichiometry and the ligand versatility allows the postu-
lation that the ligand could behave as bisbinucleating, so it
could act as a bridge between the two AgI ions.
Diffuse Reflectance and Magnetism
The diffuse reflectance spectrum of Ni2(SB)2·2H2O is
characterised by a medium intensity broad band around
680 nm, which is attributed to a d-d transition expected for
d8 ions in tetrahedral fields.[22] In the spectrum of
Cu(SB)·4H2O, the only absorption attributable to a d-d
transition is observed at about 460 nm, which is also con-
sistent with a tetrahedral geometry.[23]
The magnetic moment for the NiII dimer at room tem-
perature (3.0 BM per mol of Ni2ϩ) is quite low for a tetra-
hedral compound. This could be caused by a considerable
distortion from the ideal geometry, resulting in a fall in the
orbital contribution to this value.[24] In fact, we have found
a
similar value (3.3 BM per mol of Ni2ϩ
) for
Ni2(SB1)2·MeCN,[16] in which the ligand, helical arrange-
˚
ment and the Ni···Ni distance (ca. 6.2 A) are not very differ-
Figure 6. NOESY experiment for Zn(SB)·H2O in CDCl3 at 300 K
ent from those observed for Ni2(SB)2·2H2O. No coupling
interactions between the two Ni2ϩ ions are to be expected,
˚
of Hh, Hc, and Hj, respectively. Similarly, the HcϪHd,
HdϪHe, and HeϪHf couplings provide evidence of the Hd,
He, Hf, and Hg positions, whilst the absence of coupling
between Hg and Hf seems to indicate that the spatial ar-
rangement of the tosyl groups, with respect to benzylidene
rings, is not equivalent to that observed in Zn(SB)·H2O.
The experiment mentioned also shows the enantiotopic
nature of the geminal protons in each methylene group on
the NMR timescale at room temperature. This could rule
out a helical arrangement of SB2Ϫ in this complex. Al-
though the observation of one signal each for Hj and Hk
(in [D6]DMSO or Cl3CD at room temperature) could be
due to a fast exchange between helical and non-helical ar-
rangements of SB2Ϫ in the PbII complex, a series of 1H
NMR spectra in the 323Ϫ218 K range reveals the kinetic
stability of this complex in chloroform solution, even at
high temperatures.
because these are quite far apart (ca. 6.8 A) and the ligands
seem incapable of transmitting any interaction. Finally, a µ
value at room temperature (2.1 BM) in the narrow range
expected for a magnetically dilute tetrahedral compound
arises from the magnetic susceptibility measurements for
Cu(SB)·4H2O.
Conclusions
The studied N4 donor ligand displays a helical arrange-
ment around those ions corresponding to the fourth period
Ϫ NiII, CuII, and ZnII Ϫ but it appears planar around PdII,
AgI, CdII, and PbII ions, belonging to the fifth and sixth
periods. Ionic radii thus seem to play a relevant role in the
stereochemistry. Similarly, and in view of the secondary
M···O interactions, a predilection for pseudooctahedral
rather than distorted tetrahedral environments could be re-
Pd(SB)·3H2O·MeCN, Ag2(SB)·2H2O and Cd(SB)·3H2O
The diamagnetism of the PdII complex attests to the elec- sponsible for the obtaining of double-stranded NiII hel-
2
tronic pairing in the dxz, dyz, dz , and dxy orbitals, which is icates, instead of monohelicates, with N-tosyl-substituted
consistent with a square planar symmetry for a d8 ion. Both Schiff bases containing a pure alkyl chain of three or four
the spectra of the PdII and CdII complexes display only one members as spacer. Both CuII and ZnII ions assume dis-
signal each for Hj and Hk, as would be expected for a torted tetrahedral coordination geometries involving the
planar arrangement of SB2Ϫ. NOESY experiments in four N atoms of the dianionic ligand in their monohelicates,
[D6]DMSO at room temperature allow a rigorous assign- whereas the ligand behaves as a N2ϩN2 donor in
1
ment of the H NMR signals for these complexes (see Ex- Ag2(SB)·2H2O and Ni2(SB)2·5MeCN.
perim. Section).
The studied (CH2)4 spacer between the two 2-tosylami-
The diamagnetism shown by Ag2(SB)·2H2O attests to the nobenzylidene units has not given rise to homochiral crys-
presence of AgI ions and consequently the inability of the tals
of
Ni2(SB)2·5MeCN,
Cu(SB)·MeCN,
or
ligand to stabilise AgII even under strongly oxidising con- Zn(SB)·2MeCN. However the last two crystallise in an
ditions (10 V). As in previous cases, the observation of only acentric space group with a polar axis, forming alternative
one signal each for Hj and Hk at room temperature (Elec- sheets of the two helical enantiomers (Λ and ∆), which is
tronic Supplementary Material) is indicative of a non-heli- unusual for helical racemic compounds.
3910
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 3905Ϫ3913

Mono- and Dinuclear Complexes of a Flexible Schiff Base Ligand
FULL PAPER
Ag2(SB)·2H2O: Grey solid. Yield 76%. M.p. 250 °C.
C32H36Ag2N4O6S2 (852.52): calcd. C 45.1, H 4.3, N 6.6, S 7.5;
Experimental Section
General: Elemental analyses were performed on a CarloϪErba EA
1108 analyser. NMR spectra were recorded on a Bruker DRX 500
spectrometer in CDCl3, CD3CN, or [D6]DMSO as solvents. Infra-
red spectra were recorded as KBr pellets on a Bio-Rad FTS 135
spectrophotometer in the 4000Ϫ600 cmϪ1 range. Electrospray mass
spectra were recorded on a LC/MSD HewlettϪPackard 1100 spec-
trometer with DMSO and methanol with formic acid (2%) as sol-
vent. The FAB mass spectrum was recorded on a Micromass Auto-
spec spectrometer, with m-nitrobenzyl alcohol as matrix. Diffuse
reflectance spectra were recorded on a PerkinϪElmer Lambda 9
spectrophotometer. Magnetic susceptibility measurements were
performed on pulverised samples at room temperature with a Sher-
wood Scientific magnetic susceptibility balance.
found C 45.2, H 4.1, N 7.0, S 7.3. IR (KBr): ν
˜ ϭ 3453 (br., OϪH),
1633 (s, CϭN), 1293 (s, CϪN), 1261, 1126 (s, SO2) cmϪ1. ESI-MS
(150 V): m/z (%) ϭ 603.3 (100) [H2SB ϩ H]ϩ, 711.2 (25) [Ag(HSB)
ϩ
H]ϩ, 817.0 (5) [Ag2(SB) H]ϩ. 1H NMR (500 MHz,
ϩ
[D6]DMSO): δ ϭ 1.85 (m, 4 H, Hk), 2.29 (s, 6 H, Hi), 3.62 (m, 4
H, Hj), 6.70 (t, 2 H, Hd), 7.05 (t, 2 H, He), 7.16 (d, 4 H, Hh), 7.22
(d, 2 H, Hc), 7.33 (d, 2 H, Hf), 7.67 (d, 4 H, Hg), 8.47 (s, 2 H,
Hb) ppm.
Zn(SB)·H2O: Beige solid. Yield 70%. M.p. 210 °C.
C32H34N4O5S2Zn (684.15): calcd. C 56.2, H 5.0, N 8.2, S 9.4; found
C 56.8, H 5.0, N 8.6, S 8.9. IR (KBr): ν˜ ϭ 3442 (br., OϪH), 1634
(s, CϭN), 1299 (s, CϪN), 1261, 1140 (s, SO2) cmϪ1. FAB-MS:
m/z (%) ϭ 665.3 (72) [Zn(SB) ϩ H]ϩ. 1H NMR (500 MHz, Cl3CD):
δ ϭ 1.76 (m, 2 H, Hkeq), 2.35 (s, 6 H, Hi), 2.39 (m, 2 H, Hkax), 3.76
(m, 2 H, Hjeq), 4.19 (m, 2 H, Hjax), 6.82 (t, 2 H, Hd), 7.12 (d, 4 H,
Hh), 7.18 (t, 2 H, He), 7.25 (d, 2 H, Hc), 7.40 (d, 2 H, Hf), 7.75 (d,
4 H, Hg), 8.32 (s, 2 H, Hb) ppm.
Syntheses: Chemicals of the highest commercial grade available
(Aldrich) were used as received. N,NЈ-Bis(2-tosylaminobenzylid-
ene)-1,4-diaminobutane (H2SB) was prepared[12] by condensation
of 2-(tosylamino)benzaldehyde[25] with 1,4-diaminobutane. Com-
plexes, except for Pd(SB)·3H2O·MeCN, were obtained by an elec-
trochemical method,[9Ϫ12,22] in which a metal anode was oxidised in
an acetonitrile solution of H2SB. The preparation of Zn(SB)·H2O is
outlined below. The cell can be summarised as:
Cd(SB)·3H2O: White solid. Yield 70%. M.p. 24
5
°C.
C32H38CdN4O7S2 (767.21): calcd. C 50.1, H 5.0, N 4.9, S 8.3; found
C 49.6, H 5.4, N 4.7, S 8.0. IR (KBr): ν˜ ϭ 3443 (br., OϪH), 1628
(s, CϭN), 1293(s, CϪN), 1261, 1128 (s, SO2) cmϪ1. ESI-MS (150
V): m/z (%) ϭ 715.1 (24) [Cd(SB) ϩH]ϩ. 1H NMR (500 MHz,
[D6]DMSO): δ ϭ 2.27 (m, 4 H, Hk), 2.31 (s, 6 H, Hi), 3.80 (m, 4
H, Hj), 6.84 (t, 2 H, Hd), 7.20 (m, 4 H, HeϩHc), 7.24 (d, 4 H, Hh),
7.37 (d, 2 H, Hf), 7.75 (d, 4 H, Hg), 8.56 (s, 2 H, Hb) ppm.
Zn(ϩ)|H2SB ϩ NMe4ClO4(MeCN)|Pt(Ϫ)
.
Caution: Although no problem has been encountered in this work, all
perchlorate compounds are potentially explosive, and should be
handled in small quantities and with great care!
An acetonitrile solution (80 cm3) of H2SB (100 mg, 0.166 mmol),
containing about 30 mg of tetramethylammonium perchlorate, was
electrolysed for 1 h 45Ј, with use of a 5 mA current intensity and an
initial voltage of 16 V. Filtration and concentration of the resulting
solution yielded a solid that was washed with diethyl ether and
dried in vacuo. The ligand amount, current intensity and electroly-
sis time were identical for all the complexes. Initial voltages were
in the 8Ϫ16 V range.
Pb(SB)·3H2O: White solid. Yield 80%. M.p. 320 °C.
C32H38N4O7PbS2 (862.00): calcd. C 40.8, H 4.1, N 5.6, S 6.8; found
C 41.7, H 4.9, N 5.6, S 6.8. IR (KBr): ν˜ ϭ 3442 (br., OϪH), 1635
(s, CϭN), 1283(s, CϪN), 1259, 1131 (s, SO2) cmϪ1. ESI-MS (150
V): m/z (%) ϭ 809.2 (15) [Pb(SB) ϩH]ϩ. 1H NMR (500 MHz,
Cl3CD) δ ϭ 1.92 (m, 4 H, Hk) 7.69 (d, 4 H, Hg), 7.60 (d, 2 H, Hf),
7.25 (m, 4 H, HeϩHc), 7.07 (d, 4 H, Hh), 7.00 (t, 2 H, Hd), 3.75
(m, 4 H, Hj), 2.20 (s, 6 H, Hi), 8.31 (s, 2 H, Hb) ppm.
Pd(SB)·3H2O·MeCN was obtained by heating of a methanol solu-
tion containing H2SB, NaOH, and Pd(AcO)2 in 1:2:1 molar ratio
at reflux for 6 hours. Filtration of the resulting suspension yielded
a solid that was washed with diethyl ether and dried in vacuo. The
reaction end was determined by TLC.
Single-Crystal X-ray Diffraction Studies
Cu(SB)·MeCN and Zn(SB)·2MeCN: Diffraction intensity data
were collected at room temperature on a Bruker SMART CCD-
1000 diffractometer with use of graphite-monochromatised Mo-Kα
˚
radiation (λ ϭ 0.71073 A). Absorption correction was carried out
Ni2(SB)2·2H2O: Green solid. Yield 81%. M.p.
Ͼ 320 °C.
from equivalents with the aid of the SADABS program.[26] The
structures were solved by direct methods and refined by full-matrix,
least-squares based on F2, with the aid of SHELX-97 software.[27]
All non-hydrogen atoms were anisotropically refined. Hydrogen
atoms were included at geometrically calculated positions with
thermal parameters derived from the parent atoms. Table 2 pro-
vides a summary of crystal data, data collection and refinement
parameters. Molecular graphics are represented by Ortep-3 for
Windows, RasWin, and Platon.[28]
C64H68N8Ni2O10S4 (1354.91): calcd. C 56.7, H 5.1, N 8.3, S, 9.5;
found C 56.7, H 5.3, N 8.5, S 9.2. IR (KBr): ν˜ ϭ 3440 (br., OϪH),
1626 (s, CϭN), 1301 (s, CϪN), 1260, 1134 (s, SO2) cmϪ1. FAB-
MS: m/z (%) ϭ 659.0 (55%) [Ni(SB) ϩ H]ϩ; µ (300 K): 6.0 B.M.
3
UV/Vis (solid): λmax (nm) ϭ 680 [m, T1(P)ǟ3T1(F)].
Pd(SB)·3H2O·MeCN: Brown solid. Yield 82%. M.p. 260 °C.
C34H41N5O7PdS2 (802.27): calcd. C 51.9, H 4.9, N 7.5, S 8.6; found
C 51.6, H 4.7, N 7.7, S 8.3. IR (KBr): ν˜ϭ 3441 (br., OϪH), 1632
(s, CϭN), 1294 (s, CϪN), 1242, 1143 (s, SO2) cmϪ1. ESI-MS (100
V): m/z (%) ϭ 707.0 (100) [Pb(SB) ϩ H]ϩ. 1H NMR (500 MHz,
[D6]DMSO): δ ϭ 1.58 (m, 4 H, Hk), 1.88 (m, 4 H, Hj), 2.07 (s, 6
H, Hi), 7.10 (m, 14 H, HcϩHdϩHeϩHgϩHh), 7.20 (d, 2 H, Hf),
7.65 (s, 2 H, Hb) ppm.
Ni2(SB)2·5MeCN: Several green needle crystals from different
recrystallisation attempts of Ni2(SB)2·2H2O were measured at
120 K, showing poor diffracting behaviour. This could be the result
of extensive solvation and disorder,[3] as several low-quality refine-
ments indicated. For the best data set that could be obtained, the
measured crystal proved to be a two-component twin [twin ratio:
0.719(5)], arising from a rotation of Ϫ179.94° around the vector
normal to (1 0 0) in the reciprocal space. The twin law was used to
process the data with the aid of the GEMINI program.[29] The
Cu(SB)·4H2O: Brown solid. Yield 87%. M.p. 250 °C.
C32H40CuN4O8S2 (736.36): calcd. C 52.2, H 5.5, N 7.6, S, 8.7;
found C 52.7, H 5.3, N 7.8, S 8.7. IR (KBr): ν˜ ϭ 3442 (br., OϪH),
1621 (s, CϭN), 1295 (s, CϪN), 1263, 1138 (s, SO2) cmϪ1. FAB-
MS: m/z (%) ϭ 664.0 (60) [Cu(SB) ϩH]ϩ. µ (293 K): 2.1 B.M. UV/ structure was solved with the help of DIRDIF[30] and then treated
2
Vis (solid) λmax (nm) ϭ 460 (m, Eǟ2T).
as in the two previous cases. The C atoms of one of the methylene
3911
Eur. J. Inorg. Chem. 2003, 3905Ϫ3913
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

´
J. Sanmartın et al.
FULL PAPER
Table 2. Crystal and structure refinement data for Ni2(SB)2·5MeCN, Cu(SB)·MeCN, and Zn(SB)·2MeCN
Ni2(SB)2·5MeCN
Cu(SB)·MeCN
Zn(SB)·2MeCN
Empirical formula
M
C74H79N13Ni2O8S4
1524.16
monoclinic
C2/c (No. 15)
34.212(12)
12.443(4)
18.114(6)
93.234(7)
7699(5)
120(2)
C34H35CuN5O4S2
705.33
C36H38N6O4S2Zn
748.21
orthorhombic
Fdd2 (No. 43)
12.3928(15)
47.606(6)
12.1526(15)
90
7169.7(15)
298(2)
Crystal system
Space group
orthorhombic
Fdd2 (No. 43)
47.029(11)
12.258(3)
12.297(3)
90
7089(3
293(2)
8
˚
a [A]
˚
b [A]
˚
c [A]
β [°]
3
˚
U [A ]
T [K]
Z
4
8
µ(Mo-Kα) [mmϪ1
]
0.659
0.777
0.849
No. ref. col./ no. ref. ind./Rint
Data/restraints/parameters
R1, wR2 [I Ͼ 2σ(I)]
12223/12228/0.00
12228/6/456
0.0831, 0.1987
0.1241, 0.2199
8095/3112/0.0257
3112/1/223
0.0397, 0.1073
0.0565, 0.1167
0.034(19)
0.409, Ϫ0.300
3732/3531/0.00
3531/1/224
0.0304, 0.0690
0.0423, 0.0747
0.021(11)
R1, wR2 (all data)
Abs. struct. Flack parameter
Ϫ3
˚
Residuals [e·A
]
0.769, Ϫ0.590
0.241, Ϫ0.214
[12]
´
´
M. R. Bermejo, J. Sanmartın, A. M. Garcıa-Deibe, M. Fondo,
F. Novio, D. Navarro, Inorg. Chim. Acta 2003, 347, 53Ϫ60.
P. G. Jones, Acta Crystallogr., Sect. A 1986, 42, 57.
chains could be located in two disordered positions
[C(115)ϪC(116) and C(15Ј)ϪC16Ј) with 51.6 and 48.4% occu-
pation sites, respectively], and they were isotropically treated. Some
of the solvent molecules also show some disorder, and they were
isotropically treated and some restraints applied.
[13]
[14]
H. D. Flack, Acta Crystallogr., Sect. A 1983, 39, 876Ϫ881.
M. J. Hannon, C. L. Painting, A. Jackson, J. Hamblin, W.
Errington, Chem. Commun. 1997, 1807Ϫ1808.
[15] [15a]
[15b]
H. Cheng,
CCDC-189324, -197776, and -197777 contain the supplementary
crystallographic data for this paper. These data can be obtained
free of charge at www.ccdc.cam.ac.uk/conts/retrieving.html [or
from the Cambridge Crystallographic Data Centre, 12, Union
Road, Cambridge CB2 1EZ, UK; Fax: (internat.) ϩ44-1223-336-
033; E-mail: deposit@ccdc.cam.ac.uk].
D. Chun-Ying, F. Chen-Jie, M. Qing-Jin, J. Chem. Soc., Dalton
Trans. 2000, 2419Ϫ2424.
[16]
´
´
M. Vazquez, M. R. Bermejo, M. Fondo, A. M. Gonzalez, J.
´
Mahıa, L. Sorace, D. Gatteschi, Eur. J. Inorg. Chem. 2001,
1863Ϫ1868.
[17] [17a]
A. Bilyk, M. M. Harding, P. Turner, T. W. Hambley, J.
[17b]
Chem. Soc., Dalton Trans. 1994, 2783Ϫ2790.
A. Bilyk, M.
M. Harding, P. Turner, T. W. Hambley, J. Chem. Soc., Dalton
Trans. 1995, 2549Ϫ2553.
[18] [18a] M. J. Hannon, S. Bunce, A. J. Clark, N. W. Alcock, Angew.
[18b]
[1]
Chem. Int. Ed. 1999, 38, 1277Ϫ1278.
M. Albrecht, Chem.
Eur. J. 2000, 6, 3485Ϫ3489 and references cited therein.
J. M. Lehn, Supramolecular Chemistry: Concepts and Perspec-
tives, VCH, Weinheim, Germany, 1995.
E. C. Constable, in: Comprehensive Supramolecular Chemistry,
Vol. 9 (Eds.: J. P. Sauvage, M. W. Hosseini), Pergamon, Oxford,
1996, pp. 213Ϫ252.
C. Piguet, G. Bernardinelli, G. Hopfgartner, Chem. Rev. 1997,
97, 2005Ϫ2062.
[19] [19a]
[2]
S. L. Larson, S. M. Hendrickson, S. Ferrere, D. L. Derr,
[19b]
C. M. Elliott, J. Am. Chem. Soc. 1995, 117, 5881Ϫ5882.
B. R. Serr, K. A. Anderson, C. M. Elliott, O. P. Anderson,
[3]
Inorg. Chem. 1988, 27, 4499Ϫ4504.
[20] [20a]
T. Koike, E. Kimura, I. Nakamura, Y. Hashimoto, M.
Shiro, J. Am. Chem. Soc. 1992, 114, 7338Ϫ7345. [20b] A. Jäntti,
K. Rissanen, J. Valkonen, Acta Chem. Scand. 1998, 52,
1010Ϫ1016.
[4]
M. Albrecht, Chem. Rev. 2001, 101, 3457Ϫ3497.
A. von Zelewsky, O. Mamula, J. Chem. Soc., Dalton Trans.
[5]
2000, 219Ϫ231.
[21] [21a]
[6] [6a]
T. Adatia, N. Beynek, B. P. Murphy, Polyhedron 1995, 14,
A. Collet, M.-J. Brienne, J. Jacques, Chem. Rev. 1980, 80,
[21b]
215Ϫ230. [6b] J. J. Jacques, A. Collet, S. H. Wilen, Enantiomers,
Racemates and Resolutions, John Wiley & Sons, New York,
1981.
335Ϫ338.
D. Zurita, P. Baret, J.-L. Pierre, New. J. Chem.
1994, 18, 1143Ϫ1146. [21c] G. Mugesh, H. B. Singh, R. P. Patel,
[21d]
R. J. Butcher, Inorg. Chem. 1998, 37, 2663Ϫ2669.
gesh, H. B. Singh, R. J. Butcher, Eur. J. Inorg. Chem. 1999,
G. Mu-
[7]
I. Katsuki, Y. Motoda, Y. Sunatsuki, N. Matsumoto, T. Naka-
1229Ϫ1236.
shima, M. Kojima, J. Am. Chem. Soc. 2002, 124, 629Ϫ640.
[22]
[8] [8a]
´
´
J. Sanmartın, M. R. Bermejo, A. M. Garcıa-Deibe, M.
Maneiro, C. Lage, A. J. Costa-Filho, Polyhedron 2000, 19,
185Ϫ192.
S. R. Marder, J. E. Sohn, G. D. Stucky, Materials for Non-
linear Optics: Chemical Perspectives, ACS Symposium Series
455, American Chemical Society, Washington DC, 1991. [8b] P.
A. Maggard, C. L. Stern, K. R. Poeppelmeier, J. Am. Chem.
[23] [23a]
N. Yoshida, H. Oshio, T. Ito, J. Chem. Soc., Perkin Trans.
[23b]
2 1999, 975Ϫ983.
N. Yoshida, H. Oshio, T. Ito, J. Chem.
Soc. 2001, 123, 7742Ϫ7743.
[9]
´
´
Soc., Perkin Trans. 2 2001, 1674Ϫ1678.
M. Vazquez, M. R. Bermejo, M. Fondo, A. M. Garcıa-Deibe,
´
A. M. Gonzalez, R. Pedrido, Eur. J. Inorg. Chem. 2002,
[24] [24a] A. Bencini, F. Mani, L. Sacconi, in: Comprehensive Coordi-
465Ϫ472.
nation Chemistry, Vol. 5 (Eds.: G. Wilkinson, R. D. Gillard and
[10]
[11]
[24b]
´
´
´
M. R. Bermejo, M. Vazquez, J. Sanmartın, A. M. Garcıa-
Deibe, M. Fondo, C. Lodeiro, New J. Chem. 2002, 1365Ϫ1370.
J. McCleverty), Pergamon Press, Oxford, 1987.
M. Ger-
loch, D. A. Cruse, J. Chem. Soc., Dalton Trans. 1977, 152Ϫ159.
[24c]
´
´
M. Vazquez, M. R. Bermejo, M. Fondo, A. M. Garcıa-Deibe,
M. Gerloch, D. A. Cruse, J. Chem. Soc., Dalton Trans.
´
J. Sanmartın, M. Fondo, Eur. J. Inorg. Chem. 2003, 1128Ϫ1135.
1977, 1613Ϫ1617.
3912
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 3905Ϫ3913

Mono- and Dinuclear Complexes of a Flexible Schiff Base Ligand
FULL PAPER
A. L. Spek, PLATON, A Multipurpose Crystallographic
[25]
[28c]
´
´
J. Mahıa, M. Maestro, M. Vazquez, M. R. Bermejo, A. M.
´
Gonzalez, M. Maneiro, Acta Crystallogr., Sect. C 1999, 55,
2158Ϫ2160.
Area-Detector Absorption Correction. Siemens Industrial Auto-
mation Inc., Madison, WI, 1996.
Tool, Utrecht University, Utrecht, The Netherlands, 1998.
R. Sparks, GEMINI, Bruker-AXS Inc., Madison, WI, 1999.
P. T. Beurskens, G. Beurskens, R. de Gelder, S. Garcia-Granda,
R. O. Gould, R. Israel, J. M. M. Smits, DIRDIF 99.2 for Win-
dows. Programs for Crystal Structure Determination, Crystal-
lography Laboratory, University of Nijmegen, The Nether-
lands, 1999.
[29]
[30]
[26]
[27]
G. M. Sheldrick, SHELX-97 (SHELXS 97 and SHELXL 97),
Programs for Crystal Structure Analyses, University of
Göttingen, Germany, 1998.
[28] [28a]
[28b]
L. J. Farrugia, J. Appl. Cryst. 1997, 30, 565.
R. Sayle,
Received April 29, 2003
RasWin Molecular Graphics, Windows version 2.6-ucb, 1995.
Eur. J. Inorg. Chem. 2003, 3905Ϫ3913
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3913