Hierarchical self-assembly of heteronuclear co-ordination networks

Article abstract of DOI:10.1039/b926264d

Two ligands L¹ and L² have been prepared which contain a chelating pyrazolyl-pyridine group with a pendant aromatic nitrile (in L¹, a benzonitrile; in L², a naphthonitrile). These ligands react with Ag(i) salts to give a range of infinite coordination networks or dimeric 'boxes' in which the pyrazolyl-pyridine chelates and the aromatic nitrile groups both participate in coordination to Ag(i) ions. In contrast, L¹ and L² form simple mononuclear tris-chelates [ML ₃]²⁺ with first-row transition metal dications (M = Co, Ni, Zn) in which the aromatic nitrile groups are pendant such that the complexes can be used as 'complex ligands'. The crystal structures of [M(L ²)₃](BF₄)₂ are based on solely the mer tris-chelate geometry although in solution ¹H NMR spectroscopy reveals a mixture of both fac and mer isomers of the tris-chelates. Reaction of these with Ag(i) ions allows the interaction of the pendant nitrile groups with Ag(i) ions to generate coordination networks based on [ML₃] ²⁺ cations being crosslinked by Ag(i) ions. In these networks the [ML₃]²⁺ cations have solely the fac geometry and lie on threefold rotation axes with all three pendant nitrile groups coordinated to Ag(i) ions which are three-coordinate. {[AgM(L²)₃][BF ₄]₃}_∞ (M = Co, Ni) consist of two interpenetrated (10,3)a nets which have opposite chirality at the [M(L ²)₃]²⁺ centres but are not strictly enantiomorphic as the two nets are not crystallographically equivalent. {[AgNi(L¹)₃](BF₄)₃} _∞ in contrast contains two-dimensional sheets which have a (6,3) net structure of hexagonal rings of alternative Ni(ii) and Ag(i) centres; although not interpenetrating, two such adjacent (and enantiomorphic) sheets interact with each other via numerous CH...π interactions between aromatic ligands. Formation of these structures shows that the differential reactivity of the two binding sites in L¹ and L² (pyrazolyl-pyridine, and nitrile) can be used to generate mixed-metal coordination networks in a hierarchical, stepwise manner.

Full text of DOI:10.1039/b926264d

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Hierarchical self-assembly of heteronuclear co-ordination networks†
Hazel Fenton, Ian S. Tidmarsh* and Michael D. Ward*
Received 14th December 2009, Accepted 4th February 2010
First published as an Advance Article on the web 5th March 2010
DOI: 10.1039/b926264d
Two ligands L1 and L2 have been prepared which contain a chelating pyrazolyl-pyridine group with a
pendant aromatic nitrile (in L1, a benzonitrile; in L2, a naphthonitrile). These ligands react with Ag(I)
salts to give a range of infinite coordination networks or dimeric ‘boxes’ in which the pyrazolyl-pyridine
chelates and the aromatic nitrile groups both participate in coordination to Ag(I) ions. In contrast, L1
and L2 form simple mononuclear tris-chelates [ML3]2+ with first-row transition metal dications (M =
Co, Ni, Zn) in which the aromatic nitrile groups are pendant such that the complexes can be used as
‘complex ligands’. The crystal structures of [M(L2)3](BF4)2 are based on solely the mer tris-chelate
geometry although in solution 1H NMR spectroscopy reveals a mixture of both fac and mer isomers of
the tris-chelates. Reaction of these with Ag(I) ions allows the interaction of the pendant nitrile groups
with Ag(I) ions to generate coordination networks based on [ML3]2+ cations being crosslinked by Ag(I)
ions. In these networks the [ML3]2+ cations have solely the fac geometry and lie on threefold rotation
axes with all three pendant nitrile groups coordinated to Ag(I) ions which are three-coordinate.
{[AgM(L2)3][BF4]3}• (M = Co, Ni) consist of two interpenetrated (10,3)a nets which have opposite
chirality at the [M(L2)3]2+ centres but are not strictly enantiomorphic as the two nets are not
crystallographically equivalent. {[AgNi(L1)3](BF4)3}• in contrast contains two-dimensional sheets
which have a (6,3) net structure of hexagonal rings of alternative Ni(II) and Ag(I) centres; although not
interpenetrating, two such adjacent (and enantiomorphic) sheets interact with each other via numerous
CH ◊ ◊ ◊ p interactions between aromatic ligands. Formation of these structures shows that the
differential reactivity of the two binding sites in L1 and L2 (pyrazolyl-pyridine, and nitrile) can be used
to generate mixed-metal coordination networks in a hierarchical, stepwise manner.
being the series of d/f dinuclear triple helicates which contain
Introduction
a lanthanide(III) ion in a nine-coordinate compartment and
a transition metal cation in a six-coordinate compartment.4a
Similarly, ditopic ligands have been used to prepare mixed-metal
catenates and cages by exploiting e.g. the different stereoelectronic
preferences of Cu(I) and Pd(II).5 An alternative but very well-
known manifestation of assemblies containing two different types
of metal ion is the formation of heterometallic networks based on
cyanometallates, where a cyanometallate [M1(CN)n]x- is combined
with cations of a different metal M2 to give a network based on M1–
CN–M2 bridges,6 with M1 and M2 being attached respectively
to the C-terminus and N-terminus of the cyanide ligands. This
results in properties such as bulk magnetism6a and photoinduced
energy transfer6a due to interactions between different metal
centres.
In this paper we describe the coordination chemistry of some
simple bifunctional ligands which combine bidentate (pyrazolyl-
pyridine) and monodentate (nitrile) binding sites (Scheme 1). We
show how these ligands can be used to prepare in two steps
heteronuclear coordination networks which combine octahedral
tris-chelate M(II) centres with three-coordinate nitrile-ligated
Ag(I) centres. Interesting features of this work include how the
geometric isomerisation of a labile tris-chelate metal centre (fac vs.
mer) plays an important role in the assembly of the heterometal-
lic network, and how relatively small variations in the ligand
structure change the structure of the resulting mixed-metal nets
from a (10,3)a (three-dimensional) to a (6,3) (two-dimensional)
topology.
The ability to organise functional molecules into three-
dimensional arrays with well defined spatial relationships between
the components is a major goal of supramolecular chemistry, both
in solution and in the solid state.1 From the point of view of
materials science this is of fundamental importance for many
bulk properties of materials in areas ranging from molecular
magnetism to non-linear optics. In some cases the overall function
of the material may rely on the spatial relationship between the
components (e.g. non-linear optical properties of a coordination
network):2 in other cases the function may be associated with
a specific molecule, but an ordered array is desirable to allow
addressing of individual molecules, for example reading/writing
data to a single-molecule magnet.3
A simple type of assembly containing two different types of
functionality in an ordered array is the class of heteronuclear
complexes or coordination networks containing two different
types of metal ion, of which numerous examples are known. For
example mixed-metal helicates containing two types of metal ion
in binding sites with different coordination numbers have been
described by several groups,4 with the most notable examples
Department of Chemistry, University of Sheffield, Sheffield, UK S3 7HF.
E-mail: m.d.ward@sheffield.ac.uk
† CCDC reference numbers 758085, 758086, 758087, 758088, 758089,
758090, 758091, 758092, 758093, 758094, 758095, 758096. For crystallo-
graphic data in CIF or other electronic format see DOI: 10.1039/b926264d
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 3805–3815 | 3805
©

View Article Online
infinite networks but the complexes of L2 being discrete dimeric
molecular species.
Reaction of L1 with AgNO3 in MeCN in a 1 : 1 ratio, followed
by slow diffusion of diethyl ether vapour into the mixture, afforded
crystals of [Ag(L1)(NO3)]• which proved to be a one-dimensional
polymeric chain (Fig. 3, 4). The polymer is propagated not by L1
acting as a bridging ligand, but by the nitrate anions which connect
+
discrete {Ag(L1)} units (Fig. 3). Each L1 ligand coordinates to
an Ag(I) ion solely via the bidentate pyrazolyl-pyridine unit with
the aromatic nitrile group not being used as a donor but being
pendant. The nitrate ligand is monodentate to both Ag(I) ions
[via O(1) and O(3)], with Ag(1)–O(1) and Ag(1)–O(3) distances
Scheme 1
˚
being similar at 2.472(2) and 2.357(2) A respectively. In contrast
the remaining oxygen atom O(2) is separated from the two Ag(I)
Results and discussion
˚
ions on either side of it by 2.70 and 2.99 A, indicating a weak
bridging interaction. Given the plasticity of the coordination
environment in Ag(I) complexes in general the distinction between
a ‘true’ coordinate bond and a weaker non-bonding interaction is
arbitrary, but if we ignore the long interaction to O(2) then each
Ag(I) centre is in a typical irregular four coordinate environment
with an N2O2 donor set from a bidentate pyrazolyl-pyridine and
two monodentate nitrate ions. Fig. 4a shows a view of the whole
chain emphasising the alternating sequence of Ag+ cations and
nitrate anions which form the backbone of the chain. The pendant
nitrile groups are alternately arranged ‘up’ and ‘down’ with respect
to the chain axis. There is no aromatic stacking between ligands
within a chain, but there is such stacking between chains with
the pendant phenyl ring from one chain stacking between two
pyrazolyl-pyridine chains of the next chain and vice versa such
that there is an alternating, interdigitated sequence of phenyl
Ligand syntheses
The two ligands L1 and L2 (Scheme 1) were prepared by reac-
tion of deprotonated 3-(2-pyridyl)pyrazole with 4-bromomethyl-
benzonitrile or 4-bromomethyl-naphthonitrile respectively; the
necessary bromomethyl-substituted compounds were made by
bromination of the parent methyl compounds with NBS. Both
ligands contain a chelating bidentate pyrazolyl-pyridine unit
which we know from extensive previous work to have a high affinity
for transition metal dications,7 and a pendant monodentate
nitrile unit which will have a high affinity for softer metal cations
such as Ag(I). The two ligands have been structurally characterised
and the crystal structures are shown in Fig. 1 and 2.
˚
and pyrazolyl-pyridine units (average separation, 3.50 A) between
adjacent chains (Fig. 4b).
Fig. 1 ORTEP view of L1 (with thermal ellipsoids at the 40% probability
level) showing the atomic labelling scheme.
Fig. 2 ORTEP view of L2 (with thermal ellipsoids at the 40% probability
level) showing the atomic labelling scheme.
Fig. 3 ORTEP view of part of the one-dimensional chain structure of
[Ag(L1)(NO3)]• (with thermal ellipsoids at the 40% probability level) show-
˚
ing the atomic labelling scheme. Selected bond distances, A: Ag(1)–N(11),
2.337(2); Ag(1)–N(22), 2.3401(19); Ag(1)–O(1), 2.472(2); Ag(1)–O(3),
2.357(2).
Complexes with Ag(I)
The monometallic complexes of L1 and L2 with Ag(I) ions alone
show some interesting features, with the complexes of L1 being
Use of AgBF4 instead of AgNO3 afforded the quite different
coordination polymer [Ag(L1)(BF4)]• in which, because the anion
3806 | Dalton Trans., 2010, 39, 3805–3815
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Fig.
4 Two views of the one-dimensional chain structure of
[Ag(L1)(NO3)]•. (a) A single nitrate-bridged chain, showing the alternating
up/down arrangement of ligands, with one ligand highlighted in green. (b)
A view of two adjacent chains showing the p-stacking of alternating phenyl
and pyrazolyl rings resulting from interdigitation of ligand fragments; the
overlapping fragments are highlighted in green.
is non-coordinating, both coordination of the nitrile groups
and Ag ◊ ◊ ◊ Ag interactions are required to provide coordinative
saturation around the Ag(I) ions, resulting in a two-dimensional
polymeric sheet. Each ligand L1 spans two Ag(I) ions, presenting
a pyrazolyl-pyridine chelating site to one metal ion and the nitrile
group to the next, forming a zig-zag chain in which each Ag(I) ion
is N3 coordinate with an approximately planar geometry (Fig. 5).
Fig. 6 Three views of the structure of [Ag(L1)(BF4)]•. (a) A view
emphasising the Ag–Ag contacts between separate chains. (b) A view
showing how the chains (coloured separate for clarity) are associated
into a two-dimensional sheet via the Ag–Ag interactions. (c) The same
arrangement of chains as in (b) but viewed so that the sheet is face-on
rather than edge-on.
approximately trigonal pyramidal geometry, with the Ag ◊ ◊ ◊ Ag
interaction occupying the axial site. The interconnections to give
an infinite two-dimensional sheet arise because each Ag/L1 chain
is linked to both neighbouring chains, one on each side, via an
alternating sequence of Ag ◊ ◊ ◊ Ag contacts between each chain
and its neighbours on either side. Moving along a given chain the
Ag ◊ ◊ ◊ Ag interactions alternate between the chains on the ‘left’
and ‘right’. This is illustrated in Fig. 6. A consequence of this
arrangement of ligands is the presence of approximately square
voids in the structure which may be considered as having an
Ag ◊ ◊ ◊ Ag pair of metal ions at each corner. Alignment of these
voids through the crystal creates channels which are occupied by
the counter-ions. Perhaps surprisingly there are no inter-ligand
aromatic stacking interactions, either between ligands within a
chain, or between separate chains in a sheet, or between sheets.
Replacement of a phenyl ring by a naphthyl group in L2 results
in formation of quite different homoleptic complexes with Ag(I).
[Ag2(L2)2]X2 (X = BF4, NO3) are both discrete dimeric ‘boxes’8 in
which the ligands L2 coordinate to both metal ions; in these cases
the nature of the anion has a much less significant effect as both
complexes are of the same basic type. Fig. 7 shows the structure
of [Ag2(L2)2(NO3)2]. Each Ag(I) ion is bound by a pyrazolyl-
pyridine unit from one ligand and a nitrile group from the other
in a box-like structure which lies astride an inversion centre; in
addition a nitrate ligand binds to each Ag(I). The nitrate is shown
Fig. 5 ORTEP view (with thermal ellipsoids at the 40% proba-
bility level) of part of the one-dimensional chain of [Ag(L1)(BF4)]•
˚
showing the atom labelling scheme. Selected bond distances, A:
Ag(1)–N(211), 2.103(4); Ag(1)–N(111), 2.243(4); Ag(1)–N(122), 2.311(4);
Ag(1) ◊ ◊ ◊ Ag(1A), 3.2102(7).
These chains are connected in two two-dimensional sheets
by Ag ◊ ◊ ◊ Ag argentophilic interactions between Ag(I) ions in
adjacent chains (Fig. 6a). The Ag ◊ ◊ ◊ Ag distance is in every case
˚
3.210 A. This makes each Ag(I) ion four coordinate with an
This journal is The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 3805–3815 | 3807
©

View Article Online
Fig.
7 A
view of the centrosymmetric dimer [Ag2(L2)2](NO3)2
˚
showing the atomic labelling scheme. Selected bond distances, A:
Ag(1)–N(22), 2.3187(18); Ag(1)–N(11), 2.3641(19); Ag(1)–N(38A),
2.273(2); Ag(1)–O(3), 2.4106(16).
˚
as monodentate [Ag(1)–O(3), 2.411 A] although there is also a
Fig. 9 ORTEP view (with thermal ellipsoids at the 40% probability
level) of the complex cation of [Co(L2)3](BF4)2 showing the atomic
˚
weak interaction with O(2) [Ag(1)–O(2), 2.781 A] such that the
˚
nitrate could be considered as an asymmetric bidentate chelate,
making the Ag(I) ion either four-coordinate or five-coordinate.
labelling scheme. Selected bond distances, A: Co(1)–N(111), 2.153(3);
Co(1)–N(122), 2.190(3); Co(1)–N(211), 2.148(3); Co(1)–N(222), 2.192(3);
Co(1)–N(311), 2.180(3); Co(1)–N(322), 2.177(3).
˚
The Ag ◊ ◊ ◊ Ag separation across the ‘box’ is 8.12 A. There is
no aromatic stacking between ligands within each dimer, but
extensive stacking between dimers in the crystal. Fig. 8 shows
the structure of [Ag2(L2)2(BF4)2] which is essentially identical,
except that the site previously occupied by a nitrate ligand is
now occupied by a tetrafluoroborate anion which forms two weak
˚
Ag ◊ ◊ ◊ F interactions with the Ag(I) ion [Ag(1) ◊ ◊ ◊ F(13), 2.797 A;
˚
Ag(1) ◊ ◊ ◊ F(14), 2.848 A] that are shown by dotted lines in Fig. 8.
˚
The Ag ◊ ◊ ◊ Ag distance within the dimer is 8.18 A.
Fig. 10 Alternative view of the complex cation of [Co(L2)3](BF4)2, with
each ligand coloured separately, emphasising the inter-ligand stacking
interactions within the complex.
this arrangement of ligands allows the pendant cyanonaphthyl
groups to arrange themselves so that they lie parallel to and
stacked with the coordinated pyrazolyl-pyridine chelating groups.
Thus, the naphthyl groups of the ligands coloured blue and green
(Fig. 10) lie either side of the pyrazolyl-pyridine for the red ligand,
giving a three-layer donor/acceptor/donor stack (on the basis that
the naphthyl groups are more electron rich than the pyrazolyl-
pyridines which are coordinated to a 2+ metal centre); and the
naphthyl group from the red ligand is stacked with the coordinated
pyrazolyl-pyridine of the blue ligand. There are additional similar
stacking interactions between molecules in the crystal resulting in
infinite sequences of parallel, stacked aromatic groups. The Co–
N bond distances are unremarkable for high-spin Co(II), lying in
Fig. 8 A view of the centrosymmetric dimer [Ag2(L2)2](BF4)2 showing
the atomic labelling scheme. Selected bond distances, A: Ag(1)–N(22),
2.3106(13); Ag(1)–N(11), 2.3252(14); Ag(1)–N(38A), 2.1693(15).
˚
Complexes with first-row transition metal dications
Reaction of L1 or L2 with Co(II) or Ni(II) salts affords
simple mononuclear tris-chelate complexes as exemplified by
[Co(L2)3][BF4]2 (Fig. 9, 10) in which L2 coordinates via the
pyrazolyl-pyridine sites only with the nitrile groups pendant. The
crystal structure reveals a meridional tris-chelate geometry in the
solid state, which is the sterically preferred arrangement for a
tris-chelate complex in which the bidentate ligands do not have
twofold symmetry. In addition it will be apparent from Fig. 10 that
˚
the range 2.1–2.2 A. The analogous Ni(II) and Zn(II) complexes
are isostructural. Mass spectrometric and analytical data also
confirmed formation of [M(L1)3](BF4)2 although these did not
form X-ray quality crystals.
Although [M(L2)3][BF4]2 (M = Co, Ni, Zn) crystallised exclu-
sively as the mer isomer, the 1H NMR spectra of redissolved
3808 | Dalton Trans., 2010, 39, 3805–3815
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
crystals indicates the presence of a mixture of fac and mer isomers
in solution (Fig. 11). The mer isomer has only C1 symmetry,
with all three ligands inequivalent, and should therefore exhibit
42 proton signals in a chiral tris-chelate environment which will
make the methylene protons diastereotopic. The fac isomer in
contrast has C3 symmetry and should therefore exhibit 14 proton
signals. The 1H NMR spectrum of [Co(L2)3][BF4]2 - whose signals
are characteristically shifted over the range -80 to +90 ppm
by the paramagnetism of high-spin Co(II)9 - shows (at least)
54 clearly resolved signals, labelled with asterisks in the figure,
with others possibly obscured under the water/protonated solvent
peaks or too broad to see, i.e. consistent with the presence of 56
environments arising from a mixture of both fac and mer isomers
(Fig. 11a).
doublets with a smaller coupling constant which are naphthyl
H2/H3 protons of each ligand, labelled A–D in Fig. 11b. This
confirms the presence of four different ligand environments, three
associated with the mer isomer and one with the fac isomer, in
an approximately 3 : 1 ratio such that all signals have the same
intensity.
Formation of heteronuclear Co(II)/Ag(I) and Ni(II)/Ag(I)
coordination networks
The mononuclear complexes [M(L2)3](BF4)2 (M = Co, Ni),
bearing three pendant nitrile groups, were then reacted with
Ag(I) salts to see if mixed-metal coordination networks could
be generated by attachment of the pendant nitrile groups to
Ag(I). Both afforded in good yield X-ray quality crystals of what
proved to be {[AgM(L2)3][BF4]3}• (M = Co, Ni), an infinite three-
dimensional network in which the nitrile groups of the [M(L2)3]2+
cations all coordinate to Ag(I) ions. The description and figures
below apply to the Co(II)/Ag(I) network; the Ni(II)/Ag(I) structure
is isostructural and isomorphous.
The structure (Fig. 12–14) consists of two crystallographically
independent but structurally very similar Co(II)/Ag(I) nets which
interpenetrate, one containing Co(1) and Ag(1) ions, the other
containing Co(2)/Ag(2). It is notable that the [Co(L2)3]2+ cations
now all have a fac tris-chelate geometry, as opposed to the mer
tris-chelate geometry observed in the crystal structure of the
monomers, and in fact each Co(II) ion lies on a threefold rotation
axis such that all ligands attached to a given Co(II) ions are
crystallographically equivalent. The result of this is that each
pendant naphthyl group stacks with a coordinated pyrazolyl-
pyridine group from another ligand, i.e. there are three equivalent
stacking interactions within each complex cation (Fig. 12). All
three pendant nitrile groups from each Co(II) centre coordinate to
a Ag(I) ion. In turn each Ag(I) ion is bound by three equivalent
nitrile ligands, each from a different [Co(L2)3]2+ cation, such that
the Ag(I) ion also lies on a threefold rotation axis. The coordination
environment is not strictly planar however so the three N–Ag–
N bond angles are fractionally less than 120◦, although they
are 119.9◦ or greater so they are effectively trigonal planar. The
Fig. 11 1H NMR spectra of (a) [Co(L2)3](BF4)2 (250 MHz, CD3NO2):
and (b) [Zn(L2)3](BF4)2 (500 MHz, CD3NO2). The signals labelled a–d
(lower case) are the methylene protons, in four diastereotopic pairs; the
signals labelled A–D are the coupled pairs of naphthyl H2/H3 protons in
each of the four ligand environments (see main text).
Similar behaviour is clear from the 1H NMR spectrum of
[Zn(L2)3][BF4]2 (Fig. 11b) Although these signals are concentrated
in a much narrower range (5.1–8.3 ppm) and cannot be indi-
vidually resolved, even at 500 MHz, the number of signals and
their integrals is greater than can be accounted for by the mer
isomer alone, but is consistent with an approximately 3 : 1 mix
of mer and fac isomers, giving 56 signals of approximately equal
intensity. From a COSY spectrum we can identify four pairs of
diastereotopic methylene protons (labelled a–d in Fig. 11b) in the
form of eight doublets with a characteristically large coupling
constant of ca. 18 Hz; similarly we can identify four pairs of
Fig. 12 Part of the structure of {[AgCo(L2)3][BF4]3}• showing the
atomic labelling scheme. Note the fac arrangement of ligands around
the tris-chelate Co(II) centre, the presence of a threefold rotational axis
through Co(11), and the coordination of each of the three pendant
˚
nitrile groups to an Ag(I) ion. Selected bond distances, A: Co(11)–N(111),
2.172(6); Co(11)–N(122), 2.162(11); Ag(11)–N(152), 2.200(12). In the in-
dependent network: Co(21)–N(211), 2.129(6); Co(21)–N(222), 2.158(12);
Ag(21)–N(252), 2.229(12).
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 3805–3815 | 3809
©

View Article Online
Fig. 13 Two views of one of the two independent (10,3)a networks
[containing Co(11) and Ag(11)] in the structure of {[AgCo(L2)3][BF4]3}•.
(a) A view of the whole network with one [Co(L2)3]2+ unit highlighted
in red for clarity. (b) A view showing only the Co(II) ions (blue) and the
Ag(I) ions (orange) with solid lines denoting the connections formed by
the bridging ligands; two of the fourfold helical chains are highlighted in
red and one of the eightfold helical chains in green.
Fig. 14 Two views of the entire crystal structure of {[AgCo(L2)3][BF4]3}•
showing the two independent, interpenetrating networks. (a) A view
showing all atoms (apart from counter-ions and H atoms; the two
independent (10,3)a networks are highlighted in red and blue. (b) A
view similar to that in 13(b) showing only the Co(II) and Ag(I) ions; the
two independent, networks are highlighted with red and green Co–Ag
connections.
fact that each Co(II) centre has three pendant nitriles, and each
Ag(I) unit binds three nitrile ligands, results in a 1 : 1 Co : Ag
stoichiometry: each node, be it a Co(II) centre or a Ag(I) centre, is
therefore three-connected.
identical view in which identical larger and smaller helical chains
are also aligned orthogonal to those in Fig. 13b along all three
directions as required by the cubic symmetry (space group P213).
The shortest circular path which allows one to return to the same
node involves 10 metal atoms, five of each type, and in fact the
structural type is that of a (10,3)a net. This is inherently chiral
and a fairly well-known type of three-dimensional network with
several recent examples being described.10,11
Fig. 13 shows the structure of one of the two independent
Co(II)/Ag(I) nets. All atoms (apart from H atoms and counter
ions) are included in Fig. 13a, with one Co(II) tris-chelate unit
picked out in red for clarity. Fig. 13b shows only the metal atoms,
with links between Co(II) and Ag(I) centres that are connected
by bridging ligands L2 in the crystal structure. All such Co ◊ ◊ ◊ Ag
The second independent net has an essentially identical struc-
˚
distances are 9.67 A in this network. Alternating Co(II) and Ag(I)
ture in which the Co ◊ ◊ ◊ Ag separations between metal centres
˚
centres form helical chains which, in the view in Fig. 13b, are
oriented along the crystallographic a axis and shown in red; in
this view the chain describes a clockwise path into the page with a
connected by a bridging ligand are separated by 9.47 A. The
pitch length in the helical chains is however the same as in the
˚
other one, viz. 23.05 A. Fig. 14 shows how the two independent
˚
pitch length in the helical chain of 23.05 A. The set of bonds
nets interpenetrate, with the smaller helical chains of one net
occupying the void described by the larger helical chains of the
other, and vice versa. The Co(II) tris-chelate centres [Co(2)] in this
net have the opposite chirality to those in the other independent
net based on Co(1), but the lack of a symmetry operation relating
highlighted in green also results in a helical chain descending
‘into the page’ down the a-axis, but this one has an anticlockwise
sense; the pitch length is necessarily the same. Rotating this view
by 90◦ so that the view is along the b- or the c-axis reveals an
3810 | Dalton Trans., 2010, 39, 3805–3815
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Co(1) and Co(2) means that the crystal is chiral. Thus the crystal
structure is based on two interpenetrating nets which are not quite
enantiomorphic. This is a rather unusual case for (10,3)a nets, with
the more usual situation being that the crystal is achiral because the
nets are exactly enantiomorphic - a situation leading to formation
of a ‘three-dimensional racemate’.11a-c In this case however the
crystallographic inequivalence of the two substructures leads to
an overall chiral assembly even though the two independent nets
have opposite optical configurations.
A
similar experiment involving cocrystallisation of
[Ni(L1)3](BF4)2 with AgBF4 resulted in formation of {[AgNi(L1)3]-
(BF4)3}•, a quite different type of network. Like the previous
examples the [Ni(L1)3]2+ units in this structure all have a fac
tris-chelate geometry, with this arrangement of ligands resulting
in each pendant phenyl ring stacking with a coordinated
pyrazolyl-pyridine unit from a separate ligand (Fig. 15a). Each
fac-[Ni(L1)3]2+ unit is connected to three different Ag(I) ions, one
via each pendant nitrile group (Fig. 15). Each Ag(I) ion in turn
is connected to three different Ni(II) units in a trigonal array,
although the AgN3 unit is slightly pyramidal rather than planar,
with N–Ag–N angles of 114.7◦. All Ni(II) ions and Ag(I) ions lie
on threefold axes. The two types of three-connecting node are
however connected in a simpler arrangement than we saw above
for {[AgM(L2)3][BF4]3}•, with the basic structure being that of a
(6,3) net consisting of six-membered rings of alternating Ni(II)
and Ag(I) units (Fig. 15b) which form an infinite two-dimensional
sheet.12 Connecting the Ni(II) and Ag(I) ions forms a regular
hexagonal array with 120◦ angles at each node and Ag ◊ ◊ ◊ Ni
˚
distances along each edge of 10.28 A. This (6,3) net is chiral with
all Ni(II) tris-chelate centres having the same chirality.
An interesting additional feature of the structure is that a
second crystallographically equivalent net of opposite chirality
is associated with the first one in such a way that the bulky
[Ni(L1)3]2+ unit of one net fits in the hexagonal cavity of the
other, and vice versa (Fig. 16a). Thus the structure is a genuine
racemate based on two equivalent but enantiomorphic network
components. This is not an interpenetration; the two nets just lie
‘side by side’. This results in inter-network interactions in which
the pyrazolyl proton H(24) from a Ni(II) unit in one network lies
over the centre of a phenyl ring from a Ni(II) unit in the other
Fig. 15 Two views of one of the (6,3) nets in the structure of
{[AgNi(L1)3](BF4)3}•. (a) A view of a fac [Ni(L1)3]2+ unit with each ligand
shown in a different colour for clarity; the atomic labelling scheme is
shown. Note the fac arrangement of ligands around the tris-chelate Ni(II)
centre, the presence of a threefold rotational axis through the Ni(II) ion,
and the coordination of each of the three pendant nitrile groups to an
Ag(I) ion, all of which mirror the behaviour of {[AgM(L2)3][BF4]3}• (M =
Co, Ni), cf. Fig. 12. (b) A wider view showing the (6,3) net. Selected bond
˚
network with a distance of 3.02 A from H(24) to the mean plane
˚
distances, A: Ni(1)–N(11), 2.107(4); Ni(1)–N(22), 2.107(4); Ag(1)–N(42),
2.239(7).
of the phenyl ring [the distance from the pyrazolyl carbon atom
˚
C(24) to the same phenyl mean plane is 3.61 A]. The pyrazolyl and
phenyl rings concerned are nearly perpendicular (86.2◦ between
mean planes) such that this is an edge-to-face attractive CH–
p interaction;13 the high symmetry of the structure means that
each [Ni(L1)3]2+ unit is involved in six such interactions, with
all three pyrazolyl rings acting as the CH donors and all three
pendant phenyl rings acting as the acceptor units (Fig. 16b). A
consequence of this arrangement is that the Ag(I) ions in each
network are brought into proximity such that there are Ag ◊ ◊ ◊ Ag
Conclusions
The bifunctional ligands L1 and L2, with a chelating pyrazolyl-
pyridine unit and pendant nitrile, show interesting coordination
behaviour which allows heterometallic networks to be built up
in a stepwise manner. The homometallic complexes with Ag(I)
ions exhibit a range of structures such as dinuclear M2L2 boxes
and infinite two-dimensional sheets in which, in most cases, both
pyrazolyl-pyridine and nitrile groups are involved in coordination
to Ag(I).
˚
distances of 3.67 A between the two nets. This is however too
long to be considered an argentophilic interaction (sum of van der
˚
Waals radii, 3.44 A), and the fact that the pyramidal distortion
around the Ag(I) ions is tending to separate them rather than bring
them together (Fig. 16c) suggests that any Ag ◊ ◊ ◊ Ag attractive
interaction is negligible, and it is the CH–p interactions shown
in Fig. 16b that constitute the main supramolecular interaction
between the two independent (6,3) nets.
In contrast, with the first-row transition metal dications Co(II),
Ni(II) and Zn(II) simple mononuclear tris-chelates form in which
only the pyrazolyl-pyridine units are involved in coordination
to the pseudo-octahedral metal centres and the nitrile groups
are pendant; the crystal structures show adoption of the mer
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 3805–3815 | 3811
©

View Article Online
of two interpenetrated (10,3)a nets which have opposite chirality
at the tris-chelate metal centres but are not crystallographically
equivalent, so the crystal structures as a whole is chiral; there
are extensive p–p stacking interactions within and between the
two separate nets. The latter structure consists of a pair of
strictly enantiomorphic (6,3) nets associated via CH–p interactions
between ligands.
Thus, formation of first-row transition metal complexes via
attachment of the pyrazolyl-pyridine units of L1 and L2 to the
metal centres, followed by attachment of the pendant nitrile groups
to Ag(I) ions in a separate step, has allowed stepwise, hierarchical
assembly of heteronuclear coordination networks.
Experimental
General details
Electrospray mass spectra were recorded using a low cone
voltage (typically 5V) on a Micromass LCT instrument. 1H
NMR spectra were recorded on Bruker AV1-250, AV3-400 or
DRX-500 spectrometers. Organic reagents and metal salts were
purchased from Aldrich and used as received. The following
compounds were prepared according to the published methods:
3-(2-pyridyl)pyrazole,14 4-(bromomethyl)naphthonitrile,15 and 4-
(bromomethyl)benzonitrile.15b
Synthesis of L1
A mixture of 4-(bromomethyl)benzonitrile (1.00 g, 5.1 mmol)
and 3-(2-pyridyl)pyrazole (0.74 g, 5.1 mmol) was added to THF
(100 cm3) with stirring. A solution of NaOH (2.04 g, 0.051 mol) in
water (10 cm3) was added dropwise to the reaction, which was then
◦
heated to 70 C and left to reflux for 20 h. The organic (upper)
layer was then extracted and filtered through MgSO4 before being
evaporated to yield an oily solid. This was purified by column
chromatography on alumina with 99 : 1 CH2Cl2–MeOH as the
eluant. Following the removal of unwanted by-products from the
column (the first two narrow bands which eluted), the product L1
was isolated as a white powder (0.94 g, 71%). Colourless X-ray
quality crystals were grown by diffusion of diethyl ether into a
dichloromethane solution of L1 over a period of days. ESMS:
Fig. 16 Three views of the structure of {[AgNi(L1)3](BF4)3}• showing
the interactions between two adjacent (6,3) nets. (a) A view showing
the arrangement of the two 2-D sheets with each net highlighted in a
different colour. (b) An expanded view showing the six crystallographically
equivalent CH ◊ ◊ ◊ p interactions (black lines) between H atoms H(24) of a
[Ni(L1)3]2+ unit in one (6,3) net and the pendant phenyl ring from another
[Ni(L1)3]2+ unit in the alternate (6,3) net. (c) A view showing the bending
away of two Ag(I) ions in alternate nets from each other, which results in
an Ag–Ag contact too long to be an argentophilic interaction (see main
text).
1
m/z 261 (M + H)+. H NMR (250 MHz, CD2Cl2): d 8.66 (1H,
dd; pyridyl H6), 7.94 (1H, d; pyridyl H3), 7.74 (1H, td; pyridyl
H4), 7.64 (2H, d; phenyl), 7.50 (1H, d; pyrazolyl H5), 7.31 (2H, d;
phenyl), 7.23 (1H, m; pyridyl H5), 6.98 (1H, d; pyrazolyl H4), 5.47
(2H, s; CH2). Anal. Calcd for C16H12N4: C, 73.8; H, 4.7; N, 21.5%.
Found: C, 73.4; H, 4.7; N, 21.0%.
Synthesis of L2
1
tris-chelate geometry although H NMR spectra indicate that a
4-(Bromomethyl)naphthonitrile (1.00 g, 4.1 mmol) and 3-(2-
pyridyl)pyrazole (0.59 g, 4.1 mmol) were dissolved in THF
(100 cm3) to which aqueous NaOH (1.64 g, 0.041 mol) in water
(10 cm3) was added. The resulting mixture was heated to reflux
for 20 h, before being allowed to cool to room temperature. The
organic layer was extracted and filtered through MgSO4 before
being evaporated to yield an oily solid. Diethyl ether was added
to precipitate the pure ligand L2 (0.99 g, 79%). Colourless X-ray
quality crystals were grown by diffusion of diethyl ether into a
dichloromethane solution of L2 over a period of days. ESMS:
mixture of mer and fac isomers is present in solution. Reaction
of these ‘complex ligands’ with Ag(I) results in formation of
heteronuclear coordination networks via attachment of the pen-
dant nitrile groups to Ag(I) ions. In these mixed-metal networks,
which all have the general formula {[AgMIIL3](BF4)3} (M = Co
or Ni) the tris-chelate [ML3]2+ units now invariably have a fac tris-
chelate arrangement with the M2+ cation on a threefold axis; this
facilitates formation of the (10,3)a nets {[AgM(L2)3][BF4]3}• and
the (6,3) net {[AgNi(L1)3](BF4)3}•. The former structures consist
3812 | Dalton Trans., 2010, 39, 3805–3815
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
m/z 311 (M + H)+, 333 (M + Na)+. 1H NMR (500 MHz, CD3NO2):
d 8.53 (1H, d; pyridyl H6), 8.33 (1H, d; naphthyl H5 or H8), 8.25
(1H, d; naphthyl H8 or H5), 7.96 (2H, m; pyridyl H3 and naphthyl
H2 or H3), 7.74–7.82 (3H, m; pyridyl H4, and naphthyl H6 and H7),
7.71 (1H, d; pyrazolyl H4 or H5), 7.29 (1H, d; naphthyl H3 or H2),
7.26 (1H, ddd; pyridyl H5), 6.94 (1H, d; pyrazolyl H5 or H4), 5.99
(2H, s; CH2). Anal. Calcd for C20H14N4·0.5H2O : C, 75.2; H, 4.7;
N, 17.5%. Found: C, 75.0; H, 4.6; N, 17.3%.
C34H24AgB2F8N8 : C, 43.7; H, 2.6; N, 12.0. Found: C, 43.9; H,
2.5; N, 12.0%.
[Ni(L1)3][BF4]2
A Teflon-lined autoclave was charged with Ni(BF4)2.xH2O
(0.011 g, 0.030 mmol), L1 (0.025 g, 0.090 mmol) and methanol
◦
(9 cm3). Heating to 100 C for 12 h followed by slow cooling to
room temperature yielded a blue microcrystalline solid directly
from the cooled reaction mixture which was separated from the
reaction mixture by decantation and then dried (0.091 g, 68%).
Syntheses of metal complexes
+
+
ESMS: m/z 926.1, {[Ni(L1)3](BF4)} ; 665.0, {[Ni(L1)2](BF4)} ;
The Co, Ni and Zn complexes could be prepared by either
solvothermal or conventional synthetic methods, with solvother-
mal syntheses providing the benefits of higher yields and, often,
X-ray quality crystals directly from the cooled reaction mixture.
The Ag complexes and all heteronuclear complexes could only be
obtained by synthesis under ambient conditions.
+
2+
2+
597.0, {Ni(L1)2F} ; 419.0, {Ni(L1)3} ; 289.0, {Ni(L1)} . Anal.
Calcd. for C48H36B2F8N12Ni : C, 56.9; H, 3.6; N, 16.6. Found: C,
56.7; H, 3.6; N, 16.4.
{[AgNi(L1)3][BF4]3}•
A mixture of [Ni(L1)3](BF4)2 (0.025 g, 0.025 mmol), Ag(ClO4)
(0.010 g, 0.049 mmol) and NaBF4 (0.0054 g, 0.049 mmol)
was stirred in nitromethane (2 cm3) for 24 h. Crystals grew
on diffusion of diethyl ether into this solution over a period
of 2-3 weeks. The crystals were separated from the reaction
mixture by decantation and dried (0.004 g, 15%). Anal. Calcd.
for C48H36AgB3F12N12Ni·2H2O : C, 46.3; H, 3.2; N, 13.5. Found:
C, 46.2; H, 3.2; N, 13.5.
[Ag(L1)(NO3)]•
A solution of AgNO3 (0.033 g, 0.192 mmol) in MeCN (5 cm3)
was combined with a solution of L1 (0.050 g, 0.192 mmol) in
chloroform (5 cm3). The resulting solution was stirred vigorously
in the dark for 24 h. The solvent was subsequently removed under
reduced pressure, and the resulting crude powder was washed
with methanol and then chloroform to remove any unreacted
starting materials. Diethyl ether was allowed to diffuse slowly
into a solution of the dried white powder in acetonitrile; X-
ray quality colourless block crystals grew on standing within
[M(L2)3](BF4)2 (M = Co, Ni, Zn). The following preparation
for M = Co is typical. A Teflon-lined autoclave was charged with
Co(BF4)2.xH2O (0.036 g, 0.110 mmol), L2 (0.050 g, 0.071 mmol)
and methanol (9 cm3). Heating to 100 ◦C for 12 h followed by
slow cooling to room temperature yielded large orange prismatic
crystals directly from the cooled reaction mixture. The crystals
were separated from the reaction mixture by decantation and dried
(0.091 g, 74%).
one week (0.044 g, 53%). ESMS: m/z 800.3, {Ag2(L1)2(NO3)} ;
+
628.5, {Ag(L1)2} ; 538.0, {Ag2(L1)(NO3)} . Anal. Calcd. for
C16H12AgN5O3 : C, 44.7; H, 2.8; N, 16.3. Found: C, 44.3; H, 2.7;
N, 16.0%.
+
+
[Ag(L1)(BF4)]•
Data for M = Co. Yield: 0.026 g, 82%. ESMS: m/z
+
+
1076.2, {[Co(L2)3](BF4)} ; 766.1, {[Co(L2)2](BF4)} ; 698.1,
This was prepared in the same way as [Ag(L1)(NO3)]• (above)
using AgBF4 (0.037 g, 0.192 mmol) and L1 (0.050 g, 0.192 mmol).
Colourless crystals were isolated in a yield of 0.028 g (33%).
{Co(L2)2F} ; 494.7, {Co(L2)3} ; 339.5, {Co(L2)} . Anal. Calcd
for C60H42B2CoF8N12·2H2O : C, 60.1; H, 3.9; N, 14.0. Found: C,
60.3; H, 3.5; N, 14.0%.
+
2+
2+
ESMS: m/z 823.1, {Ag2(L1)2](BF4)} ; 627.1, {Ag(L1)2} ; 367.0,
+
+
+
{Ag(L1)} . Anal. Calcd. for C16H12AgN5O3 : C, 42.2; H, 2.7; N,
Data for M = Ni. Yield: 0.021 g, 66%. ESMS: m/z 1075.3,
+
+
2+
{[Ni(L2)3](BF4)} ; 765.1, {[Ni(L2)2](BF4)} ; 494.1, {Ni(L2)3} ;
12.3. Found: C, 42.1; H, 2.4; N, 12.3%.
2+
339.0, {Ni(L2)2} . Anal. Calcd for [Ni(L2)3](BF4)2·2H2O : C, 60.1;
H 3.9; N, 14.0. Found: C, 60.3; H, 3.5; N, 14.0%.
[Ag2(L2)2](NO3)2
Data for M = Zn. Yield: 0.023 g, 73%. ESMS: m/z
Ag(NO3) (0.027 g, 0.16 mmol) and L2 (0.050 g, 0.16 mmol) were
stirred in nitromethane (2 cm3) for 24 h. The resultant pale yellow
solution was filtered. Colourless crystals grew on diffusion of
diethyl ether into this solution over a period of days (0.077 g,
+
+
1081.2, {[Zn(L2)3](BF4)} ; 771.1, {[Zn(L2)2](BF4)} ; 703.1,
{Zn(L2)2F} ; 497.1, {Zn(L2)3} ; 342.1, {Zn(L2)2} . Anal. Calcd
for [Zn(L2)3](BF4)2:C, 61.6; H 3.6; N, 14.4. Found: C, 61.6; H, 4.0;
N, 14.1%.
+
2+
2+
100%). ESMS: m/z 898.0, {Ag2(L2)2(NO3)} ; 729.6, {Ag(L2)2} ;
+
+
673.0, {Ag2(L2)(NO3)2Na} ; 418.0, {Ag2(L2)2} . Anal. Calcd. for
C40H28Ag2N10O6 : C, 50.0; H, 2.9; N, 14.6. Found: C, 49.5; H, 2.7;
N, 14.4%.
+
2+
{[AgM(L2)3][BF4]3}• (M = Co, Ni)
A mixture of [M(L2)3](BF4)2 (0.025 g, 0.023 mmol), Ag(NO3)
(0.007 g, 0.040 mmol) and NaBF4 (0.0044 g, 0.040 mmol) was
stirred in nitromethane (2 cm3) for 24 h. Crystals grew on diffusion
of diethyl ether into this solution over a period of a week. The yield
was typically 20%.
[Ag2(L2)2](BF4)2
This was prepared in exactly the same way as [Ag2(L2)2](NO3)2
detailed above but using AgBF4 as starting material. Colourless
crystals were isolated in a yield of 0.054 g, 66%. ESMS: m/z
Data
for
M
=
Co.
Anal.
Calcd.
for
C60H42AgB3CoF12N12·CH3NO2·1.5H2O : C, 50.7; H, 3.3; N,
923.1, {Ag2(L2)2(BF4)} ; 418.2, {Ag2(L2)2} . Anal. Calcd. for
12.6. Found: C, 50.8; H, 3.4; N, 12.4%
+
2+
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 3805–3815 | 3813
©

View Article Online
Data
for
M
=
Ni.
Anal.
Calcd.
For
C60H42AgB3F12N12Ni·1.5CH3NO2·2H2O : C, 49.7; H, 3.4; N,
12.7. Found: C, 49.6; H, 3.3; N, 12.5%.
X-ray crystallography
In each case a suitable crystal was mounted in a stream of cold N2
on a Bruker APEX-2 CCD diffractometer equipped with graphite-
monochromated Mo-Ka radiation from a sealed-tube source.
Details of the crystal, data collection and refinement parameters
are summarized in Table 1; selected structural parameters are
in the figure captions. Data were corrected for absorption using
empirical methods (SADABS)16 based upon symmetry-equivalent
reflections combined with measurements at different azimuthal
angles. The structures were solved by direct methods and refined
by full-matrix least squares on weighted F2 values for all reflections
using the SHELX suite of programs.17 Non-hydrogen atoms were
refined anisotropically. Hydrogen atoms were placed in calculated
positions, refined using idealized geometries (riding model) and
assigned fixed isotropic displacement parameters. Several of the
structures of metal complexes presented problems during data
collection and refinement due to rapid solvent loss and/or anion
disorder which resulted in weak scattering. The treatments of
these issues are discussed in detail in the individual CIFs. The
mixed-metal networks in particular showed weak scattering and
not all of the counter-ions could be located; in addition disordered
solvent molecules that could not be modelled satisfactorily were
eliminated using the ‘Squeeze’ function in PLATON.18
Acknowledgements
We thank EPSRC for financial support and Mr. Harry Adams for
assistance with the X-ray crystallography.
References
1 J. W. Steed and J. L. Atwood, Supramolecular Chemistry (2nd edition),
Wiley, 2009.
2 (a) L. Zhang, Y.-Y. Qin, Z.-J. Li, Q.-P. Lin, J.-K. Cheng, J. Zhang and
Y. - G. Ya o , Inorg. Chem., 2008, 47, 8286; (b) Y. Li, Z.-X. Zhang, K.-C.
Li, W.-D. Song and q.-S. Li, Inorg. Chem. Commun., 2007, 10, 1557;
(c) P. Ayyappan, G. Sirokman, O. R. Evans, T. H. Warren and W.-B.
Lin, Inorg. Chim. Acta, 2004, 357, 3999; (d) Y. Liu, W.-M. Xuan, H.
Zhang and Y. Cui, Inorg. Chem., 2009, 48, 10018; (e) M. J. Prakash
and T. P. Radhakrishnan, Inorg. Chem., 2006, 45, 9758; (f) S.-L. Li,
J.-Y. Wu, Y.-P. Tian, H. Ming, P. Wang, M.-H. Jiang and H.-K. Fun,
Eur. J. Inorg. Chem., 2006, 2900.
3 M. Mannini, F. Pineider, P. Sainctavit, C. Danieli, E. Otero, C.
Sciancalepore, A. M. Talarico, M. A. Arrio, A. Cornia, D. Gatteschi
and R. Sessoli, Nat. Mater., 2009, 8, 194.
4 (a) D. Imbert, M. Cantuel, J.-C. G. Bu¨nzli, G. Bernardinelli and C.
Piguet, J. Am. Chem. Soc., 2003, 125, 15698; (b) E. C. Constable, A. J.
Edwards, P. R. Raithby and J. V. Walker, Angew. Chem., Int. Ed. Engl.,
1993, 32, 1465; (c) S. P. Argent, H. Adams, L. P. Harding, T. Riis-
Johannessen, J. C. Jeffery and M. D. Ward, New J. Chem., 2005, 29,
904.
5 (a) C. Dietrich-Buchecker, B. Colasson, M. Fujita, A. Hori, N. Geum,
S. Sakamoto, K. Yamaguchi and J.-P. Sauvage, J. Am. Chem. Soc.,
2003, 125, 5717; (b) S. Hiraoka, Y. Sakata and M. Shionoya, J. Am.
Chem. Soc., 2008, 130, 10058; (c) F. Gruppi, F. Boccini, L. Elviri and
E. Dalcanale, Tetrahedron, 2009, 65, 7289.
6 (a) R. Lescouezec, L. M. Toma, J. Vaissermann, M. Verdaguer, F. S.
Delgado, C. Ruiz-Perez, F. Lloret and M. Julve, Coord. Chem. Rev.,
2005, 249, 2691; (b) G. M. Davies, S. J. A. Pope, H. Adams, S. Faulkner
and M. D. Ward, Inorg. Chem., 2005, 44, 4656.
3814 | Dalton Trans., 2010, 39, 3805–3815
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
7 M. D. Ward, Chem. Commun., 2009, 4487.
12 (a) L.-P. Zhang, J.-F. Ma, J. Yang, Y.-Y. Liu and G.-H. Wei, Cryst.
Growth Des., 2009, 9, 4660; (b) F.-Y. Cui, K.-L. Huang, Y.-Q. Xu, Z.-G.
Han, X. Liu, Y.-N. Chi and C.-W. Hu, CrystEngComm, 2009, 11, 2757;
(c) Q.-Y. Yang, S.-R. Zheng, R. Yang, M. Pan, R. Cao and C.-Y. Su,
CrystEngComm, 2009, 11, 680; (d) A. A. Massoud and M. A. M. Abu-
Youssef; (e) Y.-B. Go, X.-Q. Wang and A. J. Jacobson, Inorg. Chem.,
2007, 46, 6594; (f) J.-R. Li, R.-H. Zhang and X.-H. Bu, J. Chem.
Crystallogr., 2004, 34, 501; (g) J. Li, L. Song and S. Du, Inorg. Chem.
Commun., 2007, 10, 358; (h) G. A. Farnum, M. R. Montney, R. M.
Supkowski and R. L. LaDuca, Z. Anorg. Allg. Chem., 2009, 635, 1549.
13 C. A. Hunter and J. K. M. Sanders, J. Am. Chem. Soc., 1990, 112, 5525.
14 (a) A. J. Amoroso, A. M. W. Cargill Thompson, J. C. Jeffery, P. L. Jones,
J. A. McCleverty and M. D. Ward, J. Chem. Soc., Chem. Commun.,
1994, 2751; (b) H. Brunner and T. Scheck, Chem. Ber., 1992, 125, 701;
(c) Y. Lin and S. A. Lang, J. Heterocycl. Chem., 1977, 14, 345.
15 J. J. McCullough, W. K. MacInnis, C. J. L. Lock and R. Faggiani, J. Am.
Chem. Soc., 1982, 104, 4644.
8 M. J. Hannon, C. L. Painting and W. Errington, Chem. Commun., 1997,
1805; M. J. Hannon, C. L. Painting and W. Errington, Chem. Commun.,
1997, 307.
9 (a) I. S. Tidmarsh, B. F. Taylor, M. J. Hardie, L. Russo, W. Clegg
and M. D. Ward, New J. Chem., 2009, 33, 366; (b) H. Amouri,
L. Mimassi, M. N. Rager, B. E. Mann, C. Guyard-Duhayon and L.
Raehm, Angew. Chem., Int. Ed., 2005, 44, 4543; (c) H. E. Fenton,
I. S. Tidmarsh and M. D. Ward, Dalton Trans., 2009, 4199; (d) E. C.
Constable, C. E. Housecroft, T. Kulke, C. Lazzarini, E. R. Schofield
and Y. Zimmermann, J. Chem. Soc., Dalton Trans., 2001, 2864.
10 S. R. Batten and R. Robson, Angew. Chem., Int. Ed., 1998, 37,
1460.
11 (a) X.-H. Bu, K. Biradha, T. Yamaguchi, M. Nishimura, T. Ito, K.
Tanaka and M. Shionioya, Chem. Commun., 2000, 1953; (b) B. F.
Abrahams, S. R. Batten, H. Hamit, B. F. Hoskins and R. Robson,
Chem. Commun., 1996, 1313; (c) L. Carlucci, G. Ciani, D. M. Proserpio
and A. Sironi, Chem. Commun., 1996, 1393; (d) X.-H. Bu, W. Chen, M.
Du, K. Biradha, W.-Z. Wang and R.-H. Zhang, Inorg. Chem., 2002, 41,
437; (e) L. Carlucci, G. Ciani, D. M. Proserpio and A. Sironi, J. Am.
Chem. Soc., 1995, 117, 12861; (f) B. F. Abrahams, P. A. Jackson and R.
Robson, Angew. Chem., Int. Ed., 1998, 37, 2656.
16 G. M. Sheldrick, SADABS: A program for absorption correction with
the Siemens SMART system, University of Go¨ttingen, Germany, 1996.
17 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008,
64, 112.
18 A. L. Spek, J. Appl. Crystallogr., 2003, 36, 7.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 3805–3815 | 3815
©