Red and blue luminescent metallo-supramolecular coordination polymers assembled through π-π interactions

Article abstract of DOI:10.1039/b000871k

The use of π-stacking interactions to control the aggregation of photo-active metal centres is explored through the design of bis(2,2′;6′,2″-terpyridyl) metal complexes functionalised with biphenyl 'tails'. Aryl-aryl interactions control the aggregation of the metal complexes into polymetallic arrays in the solid state. Cobalt(n), ruthenium(n), nickel(n), copper(n), zinc(n) and cadmium(n) bis-ligand complexes and a mixed ligand ruthenium(n) complex have been structurally characterised. The solid-state structures are dependent on which units dominate the π-stacking. For cobalt, ruthenium, nickel and copper, biphenylene-biphenylene interactions lead to linear rod-like arrays, while for the group 12 d¹⁰ ions zinc and cadmium, biphenylene-pyridyl interactions lead to two-dimensional sheets. The addition of the biphenylene tail has favourable effects on the photophysical-properties of the complexes which exhibit room temperature red (ruthenium) or blue (zinc and cadmium) luminescence, both in solution and the solid state. The Royal Society of Chemistry 2000.

Full text of DOI:10.1039/b000871k

View Article Online / Journal Homepage / Table of Contents for this issue
Red and blue luminescent metallo-supramolecular coordination
polymers assembled through ꢀ–ꢀ interactions†
Nathaniel W. Alcock,a Philip R. Barker,a Johanna M. Haider,b‡ Michael J. Hannon,*a
Claire L. Painting,a Zoe Pikramenou,*‡b Edward A. Plummer,a Kari Rissanenc and
Pauli Saarenketoc
a Centre for Supramolecular and Macromolecular Chemistry, Department of Chemistry,
University of Warwick, Coventry, UK CV4 7AL. E-mail: M.J.Hannon@warwick.ac.uk
b Department of Chemistry, University of Edinburgh, King’s Buildings, West Mains Road,
Edinburgh, UK EH9 3JJ
c Department of Chemistry, University of Jyväskyla, FIN-40351 Jyväskyla, Finland
Received 1st January 2000, Accepted 13th March 2000
The use of π-stacking interactions to control the aggregation of photo-active metal centres is explored through the
design of bis(2,2Ј;6Ј,2Љ-terpyridyl) metal complexes functionalised with biphenyl ‘tails’. Aryl–aryl interactions
control the aggregation of the metal complexes into polymetallic arrays in the solid state. Cobalt(), ruthenium(),
nickel(), copper(), zinc() and cadmium() bis-ligand complexes and a mixed ligand ruthenium() complex have
been structurally characterised. The solid-state structures are dependent on which units dominate the π-stacking.
For cobalt, ruthenium, nickel and copper, biphenylene–biphenylene interactions lead to linear rod-like arrays, while
for the group 12 d10 ions zinc and cadmium, biphenylene–pyridyl interactions lead to two-dimensional sheets. The
addition of the biphenylene tail has favourable effects on the photophysical-properties of the complexes which exhibit
room temperature red (ruthenium) or blue (zinc and cadmium) luminescence, both in solution and the solid state.
crystal-engineered structures,15 their combination with poly-
Introduction
pyridyl metal complexes has received scant attention.16 Impor-
The design of linear and branched (dendritic) polynuclear
tantly, the introduction of aromatic residues onto the outside
coordination arrays has attracted considerable recent interest.
of polypyridyl complexes should not be detrimental to their
In view of the exciting electrochemical and photophysical
redox and photophysical properties. Indeed introduction of a
tolyl group confers both increased extinction coefficients and
properties of polypyridyl metal complexes,1 the construction
of multi-metallic arrays containing polypyridyl centres has
extended emission lifetimes on ruthenium() and platinum()
been
a particular focus. Construction approaches have
terpyridyl complexes.4,17
focused mainly on covalently-linked systems through which
discrete oligomers of controlled nuclearity2–6 or infinite
polymers7,8 may be generated. Such discrete polynuclear
arrays are proposed as potential supramolecular devices,
exhibiting properties such as light-harvesting and energy-
funneling,9 while polymeric systems have been used as novel
polyelectrolytes which can be incorporated into devices.7
These covalent systems generally require multi-step (often
low-yielding) syntheses. Non-covalent (supramolecular)
approaches to generate multimetallic arrays have received less
attention and have centred primarily on the use of hydro-
gen bonding.10–14 Heterobimetallic systems have been
prepared by linking metal complexes through nucleic
acids10–13 and polymetallic arrays prepared through carboxylic
acid dimerisation.14 A drawback to the use of hydrogen-
bonding interactions is the poor solubility often associated with
the precursors. This is especially true for nucleic acids.
Our design strategy, therefore, was to add poly-aryl ‘tails’ to
the outside of metal complexes and to use the interactions
between these tails to control the aggregation of the metal
centres into polymetallic arrays. To investigate this approach we
chose biphenylene units as our poly-aryl ‘tails’. This extended
aryl system is capable of acting as a self-recognition motif
(forming aryl–aryl interactions with another biphenylene unit)
without introducing the solubility problems associated with
larger aryl systems, such as pyrene. While biphenyl–biphenyl
interactions should not be strong enough to cause complex
aggregation in solution (except at very high concentrations),
they should be adequate to achieve directionally-controlled
aggregation in the solid state. This property is desirable, as it
permits the discrete units from which the solid-state polymer is
comprised to be fully characterised by the usual array of solu-
tion techniques. In contrast, traditional crystal engineering is
solely dependent on solid-state techniques (most usually X-ray
crystallography) for characterisation. We have focused our
initial studies on biphenyl-substituted complexes of the simple
polypyridyl ligand 2,2Ј;6Ј,2Љ-terpyridine (tpy), since the
coordination chemistry of the unsubstituted tpy ligand with
metal dications is well established18,19 and tpy-based poly-
pyridyl ligands have been used extensively in metallo-
supramolecular design. As enunciated by Constable,3 in
contrast to 2,2Ј-bipyridine, tpy complex formation does not
introduce complications of chirality. We believe that the
strategy developed herein may be extended to a wide variety of
alternative metallo-supramolecular arrays.
We reasoned that aryl–aryl (π-stacking) non-covalent inter-
actions might offer an alternative means of controlling the
assembly of photo-active metal centres. While such interactions
have proved a powerful tool for linking organic molecules into
† Electronic supplementary information (ESI) available: rotatable 3-D
crystal structure diagrams in CHIME format, structure and packing
diagrams for the Co, Cd, Zn and Ru complexes. See http://
www.rsc.org/suppdata/dt/b0/b000871k
‡ Current address: School of Chemistry, The University of
Birmingham, Edgbaston, Birmingham, UK B15 2TT. E-mail:
Z.Pikramenou@bham.ac.uk
DOI: 10.1039/b000871k
J. Chem. Soc., Dalton Trans., 2000, 1447–1461
This journal is © The Royal Society of Chemistry 2000
1447

View Article Online
Table 1 250 MHz 1H NMR chemical shift (δ) data for biptpy and other relevant ligands in CDCl3 solution at 298 K
H6
H5
H4
H3
H3Ј
H4Ј
Ha/b
Ho/m
Hp
biptpy
phtpya
tpya
8.73
8.71
8.70
7.35
7.35
7.34
7.89
7.89
7.86
8.69
8.68
8.63
8.78
8.75
8.46
8.01, 7.74
7.67, 7.47
7.91, 7.48
7.47
7.48
7.97
a Ref. 21.
ammonium hexafluorophosphate afforded the mixed ligand
complex [Ru(tpy)(biptpy)][PF6]2 in 59% yield. Reaction of
biptpy with ruthenium() trichloride in ethanol afforded
[Ru(biptpy)Cl3] as a brown solid, which was reacted crude with
a further equivalent of biptpy in methanol in the presence of
4-ethylmorpholine. Treatment of this solution with ammonium
hexafluorophosphate afforded [Ru(biptpy)2][PF6]2 as a red solid
in 62% yield. Both of these ruthenium() complexes may also
be isolated as their chloride salts by treatment of the reaction
mixtures with lithium chloride or tetrafluoroborate salts
through treatment with ammonium tatrafluoroborate.
The IR spectra of all these transition metal complexes exhibit
peaks corresponding to the coordinated ligand and to the
hexafluorophosphate counter-ion and partial microanalytical
data for the complexes are consistent with the proposed formu-
lations. The FAB mass spectral data for the bis-ligand com-
plexes are also consistent with the proposed [M(biptpy)2][PF6]2
formulation; all the complexes exhibit strong peaks for the
parent ion [M(biptpy)2(PF6)]ϩ and, in most cases, peaks for the
fragments [M(biptpy)2]ϩ and [M(biptpy)]ϩ are also observed. In
the case of the mixed ligand complex [Ru(biptpy)(tpy)][PF6]2,
peaks corresponding to [Ru(biptpy)(tpy)(PF6)]ϩ, [Ru(biptpy)-
(tpy)]ϩ, [Ru(biptpy)]ϩ and [Ru(tpy)]ϩ are observed.
Results and discussion
Ligand preparation
The ligand biptpy was prepared using the Kröhnke approach.20
Reaction of an equimolar amount of 2-acetylpyridine with
4-biphenylcarboxaldehyde in aqueous ethanolic base at 0 ЊC
yielded a yellow precipitate of 1-(2Ј-pyridyl)-3-(4Љ-biphenyl)-
2-propen-1-one in 95% yield. Reaction of the enone with one
equivalent of N-{2-(2Ј-pyridyl)-2-oxoethyl}pyridinium iodide20
in ethanol in the presence of ammonium acetate afforded a
deep green precipitate, which was purified by recrystallisation
from ethanol in the presence of activated charcoal to give
biptpy as an off-white solid in 42% yield.
Solid-state structures
The IR spectrum of biptpy reveals strong absorptions in the
region 1600–1500 cmϪ1, corresponding to aromatic stretches,
and the lack of carbonyl peaks confirms the absence of starting
materials. The FAB mass spectrum of biptpy shows a single
peak corresponding to [M ϩ 1] and microanalytical data is
Background. Given the nature of our molecular design and
the goal of designing non-covalently linked coordination poly-
mers it is pertinent to consider how substituted and unsubsti-
tuted metal bis-terpyridyl complexes are normally arranged in
the solid state. A survey of the Cambridge Crystallographic
Database revealed a common packing motif for unsubstituted
metal terpyridyl complexes [M(tpy)2]nϩ. The cations pack
together through short face–face and edge–face aromatic–
aromatic interactions between the terminal terpyridyl rings
(centroid–centroid distances: face–face 3.5–4.1 Å; face–edge
4.9–5.3 Å. Face–edge interactions are often also termed aro-
matic CH ؒ ؒ ؒ π interactions). This gives rise to sheets of inter-
locked cations (Fig. 1). Layers of anions are located between
the sheets. Since an almost identical database survey has
recently been reported in detail by Dance et al.22 for unsubsti-
tuted terpyridyl complexes, further extensive description is
unnecessary.
The introduction of substituents can disturb this packing
motif; the interlocked sheet motif is maintained when small
substituents (such as a catechol group)23 are introduced at the
4Ј-position on the back of the terpyridyl central ring, but dis-
rupted by large substituents. When a crown ether is attached
to this 4Ј-position, the bis(terpyridyl)ruthenium() structure
reveals only a one-dimensional chain containing such π–π con-
tacts.24 These tpy–tpy π–π interactions are completely absent
from the structure of a bis(terpyridyl)cobalt() complex bear-
ing bulky 4Ј-biphenylphosphine oxide substituents.25 π–π
interactions between substituents on the terpyridine are rare,
although a tolyl–tolyl interaction has been described in a planar
platinum() complex of 4Ј-tolyl-2,2Ј;6Ј,2Љ-terpyridine.26
1
consistent with the proposed formulation. The H NMR spec-
trum of biptpy in CDCl3 solution has been recorded and the
chemical shift data is tabulated in Table 1, together with
the data for the related ligands tpy and 4Ј-phenyl-2,2Ј:6Ј,2Љ-
terpyridine (phtpy). The 1H NMR spectrum of biptpy indicates
that the ligand is symmetrical on the NMR time-scale and is
readily assigned from the distinctive splitting patterns of the
pyridyl resonances and the splitting patterns, coupling con-
stants and roofing patterns of the biphenylene protons; the
spectral assignment has been confirmed by a COSY experi-
ment. Comparison with the data for the unsubstituted tpy
ligand21 reveals that the presence of the substituent causes
the proton adjacent to the site of substitution (H3Ј) to shift
downfield by 0.32 ppm, but the other pyridyl resonances are
essentially unaffected. This downfield shift is anticipated for an
adjacent aryl substituent and a very similar shift is observed in
the spectrum of phtpy.21
Transition metal complex formation
Heating methanolic solutions containing two equivalents of
biptpy with one equivalent of the acetate salts of cobalt(),
nickel(), zinc(), cadmium() or copper(), or with the sulfate
salt of iron(), produced the complex cations [M(biptpy)2]2ϩ
which were isolated as their hexafluorophosphate salts in
69–79% yield on treatment with ammonium hexafluoro-
phosphate. The corresponding tetrafluoroborate salts could be
obtained through treatment with ammonium tetrafluoroborate.
Two ruthenium() complexes were prepared. Reaction of
[Ru(tpy)Cl3] with biptpy in methanol in the presence of
4-ethylmorpholine, followed by treatment with methanolic
To investigate the effect of a biphenyl substituent on the tpy–
tpy packing motif and to introduce a new motif based on
biphenyl–biphenyl interactions, we have investigated the crys-
tal structures of seven of the complexes described herein.
Although we have obtained all of the complexes described in a
1448
J. Chem. Soc., Dalton Trans., 2000, 1447–1461

View Article Online
crystalline form, the remainder have not yielded X-ray quality
crystals.
Thus, in this complex, we observe that the introduction of
the biphenylene tails completely disrupts the tpy–tpy packing
motif, observed with unsubstituted tpy ligands, and replaces it
with tail–tail stacking interactions which aggregate the cations
into one dimensional wire-like chains.
Crystallographic investigations
Cobalt(II) complex. Red crystals of the complex [Co(biptpy)2]-
[PF6]2 were obtained from an acetonitrile solution by the slow
diffusion of a diethyl ether–thf mixture and proved suitable for
X-ray analysis.
The cobalt() occupies a distorted octahedral geometry,
coordinating to two terdentate ligands which each occupy three
mer coordination sites (Fig. 2). The cobalt–nitrogen bond
lengths to the central ring [1.873(11)–1.894(11) Å] are shorter
than those to the terminal rings [1.925(11)–1.971(13) Å]. This
arises from the constrained bite of this terdentate ligand:
a similar pattern is observed in complexes of 2,2Ј:6Ј,2Љ-
terpyridine with a range of different metals ions,18,19 although,
for this d7 metal ion, the possibility of a Jahn–Teller contribu-
tion to the distortion cannot be excluded. A number of crystal
structures of the unsubstituted [Co(tpy)2]2ϩ cation with differ-
ent counter-ions have been reported.28–33 Such cobalt() ter-
pyridyl systems exhibit temperature-dependent spin-crossover
behaviour and this influences the metal–ligand bond lengths.
The structure of [Co(biptpy)2][PF6]2 was obtained at low tem-
The structure confirms the anticipated bis-ligand formulation
of the complex cation. Intermolecular face–face aryl inter-
actions between the biphenyl tails link the cations into infinite
one-dimensional chains (Fig. 2). The two biphenyl rings and the
central pyridyl to which they are attached stack with the equiv-
alent three rings on an adjacent molecule, such that outer
biphenyl rings stack with central pyridyl rings (4.29 Å centroid–
centroid) and inner biphenyl rings stack with an adjacent inner
biphenyl ring (4.22 Å centroid–centroid). This type of stacking
is illustrated schematically in Fig. 3 as type A. Each pair of
rings is approximately coplanar and offset. The planes of the
rings are separated by ca. 3.7 Å. The one-dimensional chains
are packed together into a two-dimensional plane (Fig. 2),
however, no short aryl–aryl interactions (either face–face or
face–edge) are observed between the chains.27 The hexafluoro-
phosphate anions and tetrahydrofuran solvent molecules are
located between the planes containing the cation chains and
form short contacts to protons in the planes above and below
(ten H ؒ ؒ ؒ F contacts in the range 2.49–2.20 Å). The chains
stretch in the same direction in each plane and there are no
short contacts between the two cation planes.27
Fig.
1
Stacking motif observed in complexes of unsubstituted
2,2Ј;6Ј,2Љ-terpyridine; hydrogens omitted for clarity.
Fig. 3 Stacking motifs observed in the complexes of biptpy ligands.
Fig. 2 Structure of the cation chains in the complex [Co(biptpy)2][PF6]2; hydrogens omitted for clarity. Analogous cation chains are observed with
ruthenium(), nickel() and copper().
J. Chem. Soc., Dalton Trans., 2000, 1447–1461
1449

View Article Online
Fig. 4 Structure of the cation chains in the complex [Cd(biptpy)2][PF6]2: (a) cation 1; (b) cation 2; hydrogens omitted for clarity.
perature (180 K) and the observed Co–N bond lengths
[1.873(11)–1.971(13) Å] are significantly shorter than those of
[Co(tpy)2][NO3]2ؒ2H2O where the Co() centre is high spin28
[2.075(5)–2.189(3) Å]. The terpyridyl units are approximately
planar (torsion angles between rings 1.0–6.1Њ). The phenyl rings
are twisted with respect to the adjacent rings about the inter-
annular C–C bond by 23.5–43.3Њ; this is slightly larger than
observed in related systems34 and could be due to the crystal
packing. The twist between the tpy unit and the inner phenyl
ring (31.6, 43.3Њ) is greater than that between the two phenyl
rings (23.5, 33.6Њ).
The twist between the tpy unit and the inner phenyl ring (36.3,
37.1Њ) is again greater than that between the two phenyl rings
(27.2, 32.8Њ).
Copper(II) and nickel(II) complexes. Slow diffusion of diethyl
ether into an acetone solution of the nickel() complex [Ni-
(biptpy)2][PF6]2 yielded crystalline material, as did slow diffu-
sion of benzene into a nitromethane solution of the copper()
complex [Cu(biptpy)2][PF6]2. Although these two complexes are
not isomorphous with the cobalt or ruthenium complexes, they
have very similar crystal packing. In each case, the anions and
solvent molecules lie in channels between the cation planes and
(for nickel and copper) are highly disordered. As a result their
diffraction is very weak, which, with the difficulty of modelling
the contents of the channels, leads to high conventional
R-values. Despite this anion and solvent disorder, the
cation geometries are well defined and the overall pattern of the
packing is clear.
As indicated, the packing is analogous to that observed in
the cobalt() and ruthenium() complexes (chains of cations
linked by type A three-ring face–face stacking interactions
through the central pyridyl and the two biphenyl rings), con-
firming that formation of chains through tail–tail interactions
is a common motif for these d-block complexes. The face–face
interactions are comparable with those in the cobalt() and
ruthenium() complexes (copper complex: inner biphenyls
4.0–4.1 Å centroid–centroid; pyridyl/outer biphenyl 4.2–4.3 Å
centroid–centroid; planes of the rings separated by ca. 3.7 Å.
Nickel complex: inner biphenyls 3.9 Å centroid–centroid;
pyridyl/outer biphenyl 4.3 Å centroid–centroid; planes of the
rings separated by ca. 3.7 Å). The chains again pack into planes
between which the anions and solvent molecules reside.
Ruthenium(II) complex. Red crystals of the complex [Ru-
(biptpy)2][BF4]2 were obtained from a nitromethane solution by
the slow diffusion of diethyl ether. The packing of this ruthen-
ium bis-ligand complex is analogous to that in the cobalt()
complex with type A three-ring face–face stacking interactions
through the tail units leading to chains of cations. The outer
biphenyl rings again stack with central pyridyl rings (4.20 and
3.99 Å centroid–centroid) and inner biphenyl rings stack with
an adjacent inner biphenyl ring (3.82 Å centroid–centroid). The
π–π separations are comparable with, though slightly shorter
than, those observed in the cobalt complex. Again, the one-
dimensional chains are packed together into a two-dimensional
plane and the anions and solvent molecules reside between the
planes. Thus, in this ruthenium() complex, the biphenylene
tails again behave as effective assembly motifs, aggregating the
cations into one- dimensional wire-like non-covalent chains.
The ruthenium() occupies a distorted octahedral geometry,
coordinating to two terdentate ligands which each occupy three
mer coordination sites. The ruthenium–nitrogen bond lengths
to the central ring [1.978(3)–1.980(3) Å] are shorter than
those to the terminal rings [2.063(3)–2.081(3) Å] and the bond
lengths and angles are comparable with those observed in other
ruthenium() terpyridyl systems.21,24,25,35 The terpyridyl units
are approximately planar (torsion angles between rings 1.5–
7.5Њ) and the phenyl rings are twisted with respect to the
adjacent rings about the interannular C–C bond by 27.2–37.1Њ.
Cadmium(II) complex. Slow diffusion of benzene into a
turquoise nitromethane solution of the cadmium complex
[Cd(biptpy)2][PF6]2 afforded pale purple crystals. The structure
is illustrated in Fig. 4 and 5 and contains two crystallographic-
ally distinct cations (1 and 2). Both have the same basic
1450
J. Chem. Soc., Dalton Trans., 2000, 1447–1461

View Article Online
Fig. 5 Packing of the cation chains in the complex [Cd(biptpy)2][PF6]2; hydrogens omitted for clarity. Red cation 1, blue cation 2.
structure and differ essentially only in the twisting of their
biphenylene tails and the extent to which they form π-stacking
interactions with other cations within the crystal.
The arrangement of these cations in the crystal lattice is
quite different from that observed in the previous complexes
and no three-ring (type A) stacks are observed between
the biphenyl tails of the complex. Instead, the structure
is dominated by interactions of types B and C (Fig. 3) in which
a ring from the biphenyl tail stacks onto a terminal pyridyl
ring.
The two types of cation chains pack alongside each other to
give a sheet containing alternate chains of cations 1 and 2
(Fig. 5). The chains are linked through face–face π-interactions
between one pair of terminal pyridyl rings (centroid–centroid
3.84 Å: Fig. 3, type D interaction). This interaction is also
associated with one face–edge π–π interaction (centroid–
centroid 5.09 Å; H ؒ ؒ ؒ centroid 2.95 Å). The other pair of
terminal pyridyl rings, although coplanar, are distant (centroid–
centroid 4.51 Å). Each cation thus forms only one set of short
face–face interactions with one adjacent chain (i.e. these inter-
chain links occur between any two chains only at every second
cation). There are also face–edge links (through two rings)
between two of the biphenyl tail units (centroid–centroid 4.63
and 4.24 Å; H ؒ ؒ ؒ centroid 3.05 and 3.06 Å). The sheets also
contain one nitromethane and one benzene solvent molecule
for every two cations. The face of the benzene molecule stacks
onto the edge of a pyridyl ring (centroid–centroid 4.89 Å;
H ؒ ؒ ؒ centroid 3.19 Å) while one of the methyl protons on the
nitromethane forms a short contact with the faces of a pyridyl
ring (H ؒ ؒ ؒ centroid 3.12 Å). The hexafluorophosphate anions
are also packed around the sheets and there are multiple short
F ؒ ؒ ؒ H contacts in the range 2.3–2.7 Å.
Each cadmium centre is 6-coordinate, bound to two terden-
tate biptpy ligands arranged in the anticipated mer configur-
ation. The bond lengths to the central pyridyl ring [Cd–N
2.270(2)–2.290(3) Å] are shorter than those to the terminal
pyridyl rings [Cd–N 2.322(3)–2.355(3) Å] as anticipated from
the constrained tpy bite. Surprisingly, this appears to be the first
crystallographically characterised cadmium() bis-complex of
a terpyridyl ligand, however, these bond lengths are similar to
those observed for other cadmium polypyridyl complexes.36
The terpyridyl units are essentially planar (torsion angles 0.8–
10.3Њ) and, as in the previously described complexes, the
biphenyl tail rings are twisted (torsion angles: py–Ph 29.8–
43.9Њ; Ph-Ph 30.5–36.6Њ).
These interactions lead to the formation of chains of cations
and these are illustrated for the two types of cations in Fig. 4a
and b. For cation 1 (Fig. 4a), one biptpy ligand forms an
interaction of type B with another cation 1 resulting in
two face–face π-interactions (centroid–centroid 4.01 Å, e.g.
red–pink). These interactions lead to short contacts between
the edge of the biphenyl ring and the face of a terminal pyridyl
ring on the other coordinated biptpy ligand (centroid–centroid
5.13 Å; H ؒ ؒ ؒ centroid 3.30 Å) and this may contribute to
the interaction. At the other end of the complex, the
second biptpy ligand interacts in a type C fashion with another
cation 1 (centroid–centroid 3.99 Å, e.g. orange–pink). These
two sets of interactions at either end of the cation lead to the
chain structure. The chains containing the 2 cations are
arranged in a similar fashion, although, as is evident in Fig. 4b,
the face–face stacking is less effective and there are some
face–edge (CH ؒ ؒ ؒ π) contributions. For the type B interaction
(centroid–centroid 4.28 Å, e.g. light blue–dark green), the rings
are not coplanar but turned to give an H ؒ ؒ ؒ centroid distance
of 3.01 Å. This interaction also leads to the edge of the outer
biphenyl ring stacking onto the face of a pyridyl ring on the
other ligand (centroid–centroid 4.91 Å; H ؒ ؒ ؒ centroid 2.87 Å).
At the other end of the complex, there is a face–face type C
interaction (centroid–centroid 4.16 Å, e.g. dark blue–dark
green).
J. Chem. Soc., Dalton Trans., 2000, 1447–1461
1451

View Article Online
Fig. 6 Structure of the complex cation [Zn(biptpy)2]2ϩ; hydrogens omitted for clarity.
Zinc(II) complex. Recrystallisation of the zinc() complex
tetrafluoroborate counter-ions are packed around the sheets.
[Zn(biptpy)2][BF4]2 from acetonitrile solution by slow diffusion
of diethyl ether yielded colourless crystals. The structure reveals
the anticipated [Zn(biptpy)2]2ϩ cations with zinc in a pseudo-
octahedral coordination geometry encapsulated by two mer-
coordinated biptpy ligands. The packing of these cations is
again different from that of the cobalt, ruthenium, nickel and
copper complexes, with no significant π-stacking interactions
of type A between the biphenyl tails being observed. The
structure shows similarities to the cadmium() complex but is
slightly simpler, containing only type B and D (and no type C)
interactions (Fig. 6).
Both biptpy ligands participate in type B interactions,
forming two face–face π-stacking interactions with an adjacent
molecule through a terminal pyridyl and the inner biphenyl ring
(centroid–centroid 3.70 and 3.75 Å; e.g. yellow–green). These
interactions are also associated with face–edge interactions
(centroid–centroid 4.82 and 4.92 Å; H ؒ ؒ ؒ centroid 2.76 and
3.09 Å) from the edge of biphenyl rings to the face of terminal
pyridyl rings. These type B interactions lead to the formation of
chains. Perpendicular to these chains, the terminal pyridyl rings
of one set of biptpy ligands face–face π-stack with the terminal
pyridyl rings on the adjacent chains (type D interaction;
centroid–centroid 3.87 Å; e.g. green–green) aggregating the
chains into a two dimensional sheet (Fig. 6). This face–face
interaction is also associated with face–edge interactions
(centroid–centroid 5.08 and 5.36 Å; H ؒ ؒ ؒ centroid 3.06 and
3.17 Å) from the edge of one of the pyridyl rings in the face–
face stack to a terminal pyridyl ring on the other biptpy ligand
on the adjacent complex.
Some short face–edge contacts are observed between the sheets
(centroid–centroid 4.96 and 5.41 Å; H ؒ ؒ ؒ centroid 3.39 and
3.13 Å).
As anticipated, the zinc–nitrogen bond lengths to the central
ring [2.066(3) and 2.077(3) Å] are shorter than those to the
terminal rings [2.174(3)–2.194(3) Å]. The bond lengths and
angles [75.51(11)–75.91(91)Њ] are unremarkable and compar-
able with those previously reported for bis-ligand zinc()
complexes of terpyridyls bearing 3,4-dihydroxyphenyl and 3,4-
dimethoxyphenyl groups in the 4Ј-position.23 The terpyridyl
units are approximately planar (Py–Py torsion angles 4.4–8.7Њ)
as is expected and the interannular twists in the biphenyl tails
(torsion angles; py–Ph 2.5 and 16.9Њ; Ph–Ph 17.5 and 28.1Њ)
are smaller than those observed in the other structures.
The observed crystal structures of these bis-ligand metal
dication complexes are thus divided into two classes. For
cobalt(), ruthenium(), nickel() and copper() the biphenyl–
biphenyl interaction envisaged in the design (type A) dominates
and results in the formation of one-dimensional wire-like
coordination polymers. The tpy–tpy (type D) interactions seen
in the complexes of tpy ligands with small or no substituents22
are not present. However, for the d10 metal ions zinc() and
cadmium() this D-type interaction, coupled with tpy–
biphenyl interactions (types B and C), leads to alternative
structures containing two-dimensional sheet-like polymeric
networks of metal centres. These two different classes of
structure must be similar energetically and we speculate that
more extended aryl substituents may drive the structure into
a wire-like polymer for all metal ions. Clearly the electronic
nature of the metal centre will influence the energy of the
σ- and π-orbitals of the bound terpyridyl ligand (and hence
the energetics of any tpy–tpy interactions), however, this is
likely to be a subtle effect and difficult to quantify (although
In this manner, two dimensional sheets are produced in
which, as in the cadmium() structure, the biphenyl tails of
the cations are all oriented along the same axis in the crystals.
Solvent molecules (two acetonitriles per cation) and the
1452
J. Chem. Soc., Dalton Trans., 2000, 1447–1461

View Article Online
Fig. 7 Structure of the complex cation [Ru(biptpy)(tpy)]2ϩ; hydrogens omitted for clarity.
Fig. 8 Packing of the complex cation [Ru(biptpy)(tpy)]2ϩ illustrating the formation of a network; hydrogens omitted for clarity.
it might explain why the second class of structures were
observed with d10 metal centres).
however, the structure clearly reveals the cations and their solid-
state packing.
The extended type A, face–face π-stacking between biphenyl-
ene tails is not observed for this complex. Instead, the structure
is indeed a three-dimensional network dominated by pyridyl–
pyridyl and pyridyl–biphenyl interactions (Fig. 7 and 8).
Perhaps unsurprisingly, the biptpy ligand engages in more
extensive π-stacking than the tpy ligand. The terminal pyridyl
rings of the biptpy face–face π-stack (3.7 Å centroid–centroid)
onto terminal pyridyl rings of the biptpy ligands of adjacent
complexes in a type D interaction (Fig. 7; see, for example, the
red cations). This leads to a one-dimensional chain. Each of
these face–face stacking interactions is also associated with two
face–edge π-interactions from the edge of the terminal pyridyl
rings of the biptpy ligands to the face of a terminal pyridyl of
Ruthenium(II) mixed ligand complex. The structure adopted
by the mixed ligand complex [Ru(tpy)(biptpy)]2ϩ is of interest,
since it contains only one biphenyl tail unit per cation. Given
the similar energetics of the two possible classes of structure
when two biphenyl units are attached to each complex, for this
mixed ligand cation bearing only one biphenyl substituent per
complex, we would expect a structure similar to the second type
(with type B/C and D interactions) to be favoured.
Red crystals of the mixed ligand complex [Ru(tpy)(biptpy)]-
Cl2 were obtained by slow evaporation of a methanol solution
of the complex. The crystals were of poor quality, giving weak
diffraction. There is considerable anion and solvent disorder,
J. Chem. Soc., Dalton Trans., 2000, 1447–1461
1453

View Article Online
1
Table 2 250 MHz H NMR chemical shift (δ) data for [M(biptpy)2][PF6]2 (M = Fe, Ru, Zn and Cd) in CD3CN solution at 298 K and related
complexes
H6
H5
H4
H3
H3Ј
H4Ј
Ha/b
Ho/m
Hp
[Fe(biptpy)2][PF6]2
7.23
7.06
7.19
7.46
7.35
7.43
7.35
7.43
7.85
8.10
7.11
7.06
7.09
7.20
7.17
7.17
7.15
7.18
7.42
7.55
7.91
7.88
7.91
7.97
7.96
7.92
7.90
7.95
8.19
8.23
8.65
8.46
8.61
8.70
8.66
8.50
8.50
8.64
8.75
8.80
9.26
8.91
9.12
9.09
9.06
8.75
8.75
9.01
9.05
9.01
8.43, 8.12
7.91, 7.53
7.53
a
[Fe(tpy)2][BF4]2
[Fe(phtpy)2][PF6]2
8.40
a
8.32, 7.82
7.86, 7.52
7.86, 7.59
7.74
7.52
7.50
[Ru(biptpy)2][PF6]2
[Ru(biptpy)(tpy)][PF6]2
8.32, 8.05
8.32, 8.06
8.41
8.40
a
[Ru(tpy)2][PF6]2
b
[Ru(phtpy)2][PF6]2
[Zn(biptpy)2][PF6]2
[Cd(biptpy)2][PF6]2
8.21, 7.75
7.85, 7.61
7.85, 7.55
7.75
7.61
7.55
8.33, 8.05
8.28, 8.03
a Ref. 37. b Ref. 36.
1
Table 3 400 MHz H NMR chemical shift (δ) data for [M(biptpy)2]-
[PF6]2 (M = Co or Ni) and related complexes in CD3CN solution at
298 K
an unsubstituted terpyridyl ligand on the adjacent complex
(4.9, 5.0 Å centroid–centroid). The complexes in the chain
below are rotated through 180Њ permitting a series of inter-
chain face–face π-stacking interactions (3.75 Å centroid–
centroid; e.g. between blue and red cations) of type B between
terminal pyridyl rings of the biptpy and inner rings of the
biphenylene tails. This therefore leads to linked pairs of cation
chains (Fig. 8; see for example blue and red cations). From
these pairs, outer biphenyl rings face–face π-stack with terminal
rings on the unsubstituted terpyridyl ligands (3.6 Å centroid–
centroid) of adjacent pairs (Fig. 8) in a pseudo-type C inter-
action, and a three-dimensional network results.
Pyridyl
Biphenyl
[Ni(biptpy)2][PF6]2
[Ni(tpy)2][PF6]2
135
135
92
74
67
71
42
47
44
44
32
34
13
14
14
9
11–0
10–7
a
75
54
57
19
22
[Co(biptpy)2][PF6]2
b
[Co(tpy)2][PF6]2
99
a Ref. 38. b Ref. 39.
The stacking is less extensive for the unsubstituted tpy ligand,
with no pyridyl–pyridyl face–face interactions being observed.
In addition to the aforementioned face–face interaction with
the outer ring of a biphenylene tail and face–edge interactions
with the edges of biptpy pyridyl rings, one additional type of
interaction is observed; a face–edge interaction from the edge
of the inner ring of a biphenylene and the face of a tpy terminal
pyridyl (5.15 Å centroid–centroid).
The individual cations have the anticipated structure, with
both terdentate ligands coordinated in a mer fashion to a
pseudo-octahedral ruthenium() centre. The aryl rings in the
biphenylene tail are less twisted with respect to the pyridines
than in most of the other structures reported herein. The inner
ring is almost coplanar with the pyridine to which it is attached
(torsion angle 7Њ), while the outer ring is twisted through a
further 29Њ with respect to the inner. Although the smaller
spatial requirements of the chloride anion (cf. hexafluoro-
phosphate and tetrafluoroborate in the other structures des-
cribed herein) may well influence the structure adopted, we have
been unable to obtain suitable crystals with these larger anions.
The terminal pyridyl rings of each ligand in the [M(biptpy)2]2ϩ
cations are clearly in very similar environments to the terminal
rings in the corresponding [M(tpy)2]2ϩ cations. The biphenyl
substituent causes the proton on the central pyridyl ring (H3Ј)
to shift with respect to the tpy complex, however, this shift is
closely mimicked in phtpy complexes. For [Cd(biptpy)2]2ϩ, the
H6 resonance is significantly broadened. This probably arises
from coupling to the spin active cadmium; no distinct cadmium
satellites were resolved but may be hidden under the adjacent
1
resonances. The H NMR spectrum of [Ru(biptpy)(tpy)][PF6]2
is more complicated due to the presence of the two differ-
ent ligands, but is readily assigned by COSY and NOE
experiments. The proton resonances are at comparable shifts to
those in [Ru(biptpy)2][PF6] and [Ru(tpy)2][PF6]2. Serial NMR
dilutions were performed on the complex [Fe(biptpy)2][PF6]
in CD3OD solution, starting from a saturated methanolic
solution. The positions and appearances of the proton reson-
ances were unaltered, implying that the biphenyl tails do not
cause solution aggregation at milli- and submillimolar
concentrations.
The complexes [M(biptpy)2]2ϩ (M = Ni or Co) are para-
Solution behaviour
1
magnetic. Nevertheless, the 400 MHz H NMR spectra of the
The solid-state structures illustrate that the biphenyl group
has indeed been effective at aggregating the discrete metal
terpyridyl centres into polymeric solid-state arrays either with a
one-dimensional wire-like structure or a two-dimensional sheet
structure. The introduction of a conjugated biphenyl unit might
also be expected to cause perturbations to the properties of the
discrete metal complexes and to investigate this, the solution
chemistry has been examined.
complexes at 298 K in CD3CN solutions were recorded. The
1H resonances are paramagnetically shifted and considerably
broadened (coupling data is thus lost), but reasonably well
resolved. The chemical shift (δ) data for the nickel() and
cobalt() complexes of biptpy are tabulated, together with the
data for the corresponding tpy complexes,38,39 in Table 3. There
are 5 resonances due to the terpyridyl unit. Comparison with
1
the H NMR spectra of the tpy complexes shows, as expected,
that the resonances are at similar chemical shifts, as would be
anticipated for a similar solution structure. It has not been pos-
sible to fully assign the spectra, but the number of environ-
ments suggests that the solution species are symmetrical on
the NMR time-scale. Comparison of the spectral data of the
complexes of tpy and biptpy reveals the absence of one proton
in the complexes of biptpy (H4Ј) and allows us to assign the
multiplets in the 0–11 range as arising from the biphenyl
protons. The small shifts of these biphenyl resonances is com-
patible with the position of these protons being remote from
the paramagnetic centre.
1H NMR. The complexes [M(biptpy)2]2ϩ (M = Fe, Ru, Zn and
Cd) are diamagnetic. The 250 MHz H NMR spectra of these
1
complexes in CD3CN solutions at 298 K have been recorded
and are assigned and tabulated in Table 2, together with the
corresponding data for the iron() and ruthenium() complexes
of tpy and phtpy.37
The spectra are sharp and well resolved and illustrate the
high symmetry of the complexes on the 1H NMR time-scale. As
expected, on complexation, H6 shows the most dramatic shift
with respect to the free ligand as it is closest to the metal ion.
1454
J. Chem. Soc., Dalton Trans., 2000, 1447–1461

View Article Online
Table 4 Electrochemical data for [M(biptpy)2][PF6]2 in acetonitrile
solution ([NBu4][PF6] supporting electrolyte). Potentials (V) quoted vs.
Fc/Fcϩ
Ligand based
reduction(s)
M(/)
Ϫ0.16
M(/)
Ϫ1.16
[Co(biptpy)2][PF6]2
Ϫ1.98
a
[Co(phtpy)2][PF6]2
Ϫ0.16
Ϫ0.13
0.89
0.90
0.92
0.90
0.69
0.67
0.77
Ϫ1.17
Ϫ1.18
Ϫ1.99
Ϫ2.06
Ϫ1.58
Ϫ1.58
Ϫ1.67
Ϫ1.66
Ϫ1.63
Ϫ1.62
Ϫ1.64
a
[Co(tpy)2][PF6]2
[Ru(biptpy)2][PF6]2
Ϫ1.79
Ϫ1.87
Ϫ1.92
Ϫ1.92
Ϫ1.76
Ϫ1.74
Ϫ1.82
[Ru(biptpy)(tpy)][PF6]2
b
[Ru(tpy)2][PF6]2
c
[Ru(phtpy)2][PF6]2
[Fe(biptpy)2][PF6]2
a
[Fe(phtpy)2][PF6]2
[Fe(tpy)2][PF6]2
b
a Ref. 40. b Ref. 41. c Ref. 21.
Table 5 Selected electronic absorption spectroscopic data for the
complexes [M(biptpy)2][PF6]2 (M = Ru, Fe and Co) in acetonitrile
solutions unless otherwise indicated
Fig. 9 Emission spectra of [Zn(biptpy)2][PF6]2 in the solid state using
a powder sample (solid line) and in methanol solution (intermittent
line). The samples were excited at 250 nm.
λmax/nm (ε/cmϪ1 molϪ1
)
[Ru(biptpy)2][PF6]2
494 (45 000)
484 (17 000)
474 (10 400)
487 (26 200)
[Ru(biptpy)(tpy)][PF6]2
a
[Ru(tpy)2][PF6]2
a
[Ru(phtpy)2][PF6]2
[Fe(biptpy)2][PF6]2
571 (23 000)
565 (22 600)
552 (11 900)
b
[Fe(phtpy)2][PF6]2
b
[Fe(tpy)2][PF6]2
[Co(biptpy)2][PF6]2
445 (3400), 518 (3000)
415 (1500), 443 (1500), 503 (1200)
253 (74 000), 284 (13 000)
c
[Co(tpy)2][PF6]2
biptpyd
[Cd(biptpy)2][PF6]2 d
[Cd(biptpy)2][PF6]2d
[Zn(biptpy)2][PF6]2
236 (60 000), 285 (59 000), 326 (70 000)
233 (37 000), 286 (46 000), 322 (36 000)
238 (67 000), 285 (63 000), 336 (83 000)
a Ref. 42. b Ref. 40. c Ref. 43. d In methanol solution.
Electrochemistry. The electrochemical behaviour of some of
the salts [M(biptpy)2][PF6]2 (M = Co, Ru, Fe, Cu and Ni) in
acetonitrile solution was investigated by cyclic voltammetry.
The results are summarised in Table 4, together with the corres-
ponding data for the cobalt(), ruthenium() and iron() tpy
and phtpy complexes.21,40,41 As would be expected, the biphenyl
substituent has little effect on the metal-based redox chemistry,
however all the ligand-based reductions are slightly lower in
energy, due to the increase in conjugation within biptpy in
comparison to tpy.
Cyclic voltammetry on the nickel() and copper() com-
plexes is less informative: In the cyclic voltammetric spectrum
of [Ni(biptpy)2][PF6]2, an irreversible reduction process at
Ϫ0.85 V and an irreversible terpyridyl-centred reduction at
Ϫ1.85 V are observed. For [Cu(biptpy)2][PF6]2, an irreversible
reduction process is observed at Ϫ0.44 V with an associated
absorption spike at Ϫ0.67 V. This is due to copper(0) plating on
the electrode. An irreversible terpyridyl-centred reduction is
also observed at Ϫ1.87 V. In both cases, the complexes of the
unsubstituted tpy ligand behave in a similar fashion.
Fig. 10 Emission spectra of [Cd(biptpy)2][PF6]2 in the solid state using
a powder sample (solid line) and in methanol solution (intermittent
line). The samples were excited at 250 nm.
two distinct MLCT bands might be expected, however, as a
result of the broad nature of the bands and their relatively small
wavelength separation, a single broad peak is observed centred
between the expected positions of the two bands.
Serial dilutions over the 10Ϫ3–10Ϫ5 M concentration range
were performed on the complex [Fe(biptpy)2][PF6] in methanol
solution and revealed no changes in peak maxima, shape or
extinction coefficients which is consistent with an absence of
solution aggregation and in agreement with the NMR
observations.
Luminescence studies. The cadmium and zinc complexes show
strong blue emission with high quantum yields when excited
with UV light. The luminescence of the complexes in solution
and the solid state are shown in Fig. 9 and 10. While the free
ligand emits in the UV region of the spectrum (390 nm in
MeOH and 395 nm in solid state), the emission of the com-
plexes is considerably red-shifted and consequently, emission
occurs in the visible blue region of the spectrum. The red-shift
in one of the ligand-based absorption bands observed upon
coordination is consistent with this. Coordination also
enhances the quantum yield; in methanol solution the zinc
complex emits with a quantum yield of 0.80, while the quantum
yield for the cadmium complex is 0.71. These values are both
higher than that for the ligand (0.55) under the same excitation
Absorption spectra.The absorption spectra were recorded
for acetonitrile solutions of the complexes [M(biptpy)2][PF6]2
(M = Co, Ru, and Fe). Selected wavelengths of the peaks and
absorption coefficients are given in Table 5. Data for [M(tpy)2]-
[PF6]2 and [M(phtpy)2][PF6]2 complexes are also shown for
comparison.40,42,43
As observed in the electrochemistry, the energy of the ligand
π* is lowered with respect to terpyridine in the more conjugated
biptpy ligand. Consequently, the MLCT bands of these biptpy
complexes are shifted to higher wavelength when compared to
the corresponding tpy complexes. For [Ru(biptpy)(tpy)][PF6]2,
J. Chem. Soc., Dalton Trans., 2000, 1447–1461
1455

View Article Online
Table 6 Emission spectroscopic data for the complexes [Ru(biptpy)2]-
[PF6]2, [Ru(biptpy)(tpy)][PF6]2 and related complexes
λem(max)/nm
298 K
77 K in CH3CN
Solvent
[Ru(biptpy)2][PF6]2
[Ru(biptpy)2]Cl2
640
640
—
—
640
634
MeCN
H2O
MeCN
MeCN
MeCN
MeCN
[Ru(biptpy)(tpy)][PF6]2
630
598
628
622
a
[Ru(tpy)2][PF6]2
a
[Ru(ttpy)2][PF6]2
b
[Ru(tpy)(ttpy)][PF6]2
a Ref. 4. b Ref. 49.
those for related compounds for comparison, are listed in
Table 6.
The complexes of this biptpy ligand therefore represent an
example of classes of systems in which there is potential not
only to create blue luminescent materials of controlled solid-
state architecture but, through selection of the metal ion, to
tune the luminescent properties of the material to achieve red
luminescence also.
Fig. 11 Emission spectra of [Ru(biptpy)2][PF6]2 in the solid state at
room temperature (solid line) λexc = 480 nm and in degassed acetonitrile
at 77 K (intermittent line) λexc = 490 nm.
Conclusions
conditions. The same emission wavelength is observed on
excitation at 250 nm or in the lower energy absorption band
at 330 nm. Blue luminescence has previously been reported
from zinc complexes of other pyridyl-containing ligands and
it is unsurprising that the zinc complex of the biphenyl
terpyridine ligand employed in our studies exhibits similar
behaviour.44–46 The luminescence is usually attributed to intra-
ligand ππ* transitions, with no reported shifts upon coordin-
ation, however, recently, involvement of charge transfer states
in the case of aryl tpy ligands has been proposed.46 Given
that zinc() polypyridyl complexes are known emitters, it is
perhaps surprising that emission from cadmium() poly-
pyridyl complexes has not, to the best of our knowledge, been
previously examined. For this biptpy ligand, the luminescence
behaviour of the cadmium complex is similar to that of the
zinc analogue.
The ability to design blue luminescent materials is desirable
for display applications and to this end Che et al. have reported
blue luminescence of solutions of some zinc pyridylamine
complexes.45,47 For the biptpy complexes, the high quantum
yields of luminescence in solution (which are enhanced with
respect to the free ligand) and the strong luminescence in the
solid state, coupled with the potential to use supramolecular
design to control and vary the solid-state structure, suggest that
materials such as these merit exploration for application as blue
light emitting diodes.
We have shown that addition of a biphenylene tail to a simple
terpyridyl ligand permits the synthesis of complexes which
exhibit red and blue luminescence in both solution and the
solid state. The solid-state structures are dependent on which
types of π-stacking interactions predominate. Biphenylene–
biphenylene interactions lead to rod-like wire structures
while biphenylene–pyridyl interactions lead to two-dimensional
sheets. The strategy developed herein is currently being
extended to other metallo-supramolecular architectures (such
as boxes49) and the structural and photophysical effects of
more extended aryl tails are being examined. Aside from their
self-recognition properties, the aryl tails may also potentially be
captured by cyclodextrin hosts and these complexes are being
explored as part of ongoing studies aimed at employing
cyclodextrin binding to construct non-covalent photo-active
arrays.50
Experimental
General details
IR spectra were recorded on a Perkin-Elmer Paragon 1000
FT spectrophotometer at a resolution of 4 cmϪ1, with the
samples in the form of compressed KBr discs. Absorption
spectra were recorded on Unicam 8700 and Jasco V550
spectrometers. 1H NMR spectra were recorded on Brüker
ACF250, DPX300 and ACP400 spectrometers. 2D sequences
used standard Brüker software.63 Electron impact (EI) mass
While ruthenium() complexes of bipyridyl ligands exhibit
3
characteristic red luminescence from the MLCT state, for the
ruthenium() complex of tpy, this state is quenched by a
low-lying 3MC state and room temperature emission is
not observed. Addition of simple donor and acceptor sub-
stituents at the 4Ј-position ligand alters the relative energies
of these states and appropriate combinations can achieve
luminescent behaviour in a predictable fashion.42 For aryl
substituents the situation is complicated by the potential for
inter-aryl twisting in the ground and excited states. Nitrile
solutions of bis(tolylterpyridine) complexes [but not bis(phenyl-
terpyridyl) complexes] of ruthenium() exhibit weak room
temperature luminescence.4 Similarly, we observe that aceto-
nitrile or aqueous solutions of the complex [Ru(biptpy)2]2ϩ
prepared herein, exhibit weak red luminescence. In the solid
state (powder) and in acetonitrile solution at low temperature
the complex shows strong red luminescence on irradiation in
the visible MLCT absorption band (Fig. 11).48 The emission
spectroscopic data for the ruthenium complexes, along with
spectra were recorded on a Micromass Autospec spec-
trometer. Fast atom bombardment (FAB) mass spectra were
recorded using 3-nitrobenzyl alcohol as matrix on Micromass
Autospec spectrometers either at Warwick or by the EPSRC
National Mass Spectrometry Service Centre, Swansea.
Microanalyses were conducted on a Leeman Labs CE44
CHN analyser by the University of Warwick Analytical
Service. Electrochemical measurements were performed using
an ecochimie Autolab electrochemical workstation using
standard GPES software.64 A conventional three-electrode
configuration was used, with platinum working and auxiliary
electrodes and a Ag–Agϩ reference. The measurements were
conducted in acetonitrile solution with 0.1 M [Bun N][PF6]
4
base electrolyte. Potentials are quoted vs. ferrocene/
ferrocenium couple (Fc/Fcϩ = 0.0 V), and all potentials were
referenced to internal ferrocene added at the end of each
experiment.
1456
J. Chem. Soc., Dalton Trans., 2000, 1447–1461

View Article Online
The luminescence studies were performed on a Quanta-
Master QM-1 steady-state emission spectrometer from Photon
Technology Instruments equipped with a 75 W xenon arc lamp
and a model 810 photon counting detection system with a red-
sensitive R928 photomultiplier tube. Single-grating mono-
chromators are used for wavelength selection of the excitation
light and the luminescence. The emission monochromator
grating is blazed at 550 nm. The data were collected and
analysed with Felix software.65 The luminescence quantum
yields of methanolic solutions of the complexes were measured
using quinine sulfate in 0.01 N H2SO4 as a standard with
quantum yield 0.54.51 All the samples for the quantum yield
measurements were excited at 250 nm to minimise excitation
wavelength errors. An error value of 10% is associated with the
quantum yield measurements.
saturated methanolic ammonium hexafluorophosphate. On
cooling, a purple solid separated and was isolated by filtration
(0.023 g, 78%). IR (KBr): ν 1616m, 1403m, 1122m, 844s cmϪ1
.
MS (ϩve FAB): m/z 441 [Fe(biptpy)], 826 [Fe(biptpy)2], 971
[Fe(biptpy)2(PF6)]. Found: C, 56.9; H, 3.4; N, 7.3. Calc. for
FeC54H38N6ؒ1.5H2O: C, 56.7; H, 3.6; N, 7.3%. UV–Vis (MeCN):
λmax (ε) 239 (36 000), 285 (48 000), 324 (61 000), 571 nm (23 000
1
cmϪ1 molϪ1). H NMR (CD3CN, 250 MHz, 298 K): δ 9.26
(2H, s, H3Ј), 8.65 (2H, d, J = 7.8 Hz, H3), 8.43 (2H, d, J = 8.7
Hz, Ha/b), 8.12 (2H, d, J = 8.4 Hz, Ha/b), 7.91 (4H, m, Ho/m,
4), 7.53 (3H, m, Ho/m,p), 7.23 (2H, d, J = 4.9 Hz, H6), 7.11 (2H,
ddd, J = 8.7, 5.5, 1.2 Hz, H5). 1H NMR (CD3OD, 250 MHz, 298
K): δ 9.56 (2H, s, H3Ј), 8.93 (2H, d, J = 7.9 Hz, Ha/b), 8.56 (2H,
d, J = 8.6 Hz, H3), 8.12 (2H, d, J = 8.7 Hz, Ha/b), 8.04 (2H, td,
J = 8.4, 1.4 Hz, H4), 7.90 (1H, dd, J = 8.4, 1.5 Hz, Ho), 7.59
(2H, t, J = 7.2 Hz, Hm), 7.49 (1H, t, J = 7.2 Hz, Hp), 7.38 (2H,
d, J = 4.6 Hz, H6), 7.27 (2H, ddd, J = 8.1, 5.4, 1.2 Hz, H5).
The tetrafluoroborate was prepared in 73% yield in an
analogous fashion, precipitating by addition of a saturated
methanolic solution of ammonium tetrafluoroborate. The
chloride salt was prepared in 84% yield from iron() chloride.
Precious metal salts (Johnson Matthey) and other chemicals
(Aldrich) were used as supplied. N-{2-(2Ј-Pyridyl)-2-oxo-
ethyl}pyridinium iodide20 and [Ru(tpy)Cl3]21 were prepared
according to literature methods.
1-(2Ј-Pyridyl)-3-(4Љ-biphenyl)-2-propen-1-one
A solution of biphenyl-4-carboxaldehyde (1.50 g, 8.2 mmol) in
ethanol (18 cm3) was treated with a solution of sodium hydrox-
ide (1 g, excess) in water (13 cm3), and 2-acetylpyridine (0.9 cm3,
8.2 mmol). The yellow mixture was stirred at 0 ЊC for 1 h. The
yellow solid that precipitated was collected by vacuum filtration
and washed with ice cold methanol (1.87 g, 59%). IR (KBr):
ν 1670s, 1600s, 1484w, 1450m, 1334s, 1218s, 1033s, 998s, 836m,
759s, 674s cmϪ1. MS (ϩve FAB): m/z 286 [M ϩ 1]. Found: C,
82.6; H, 5.2; N, 5.1. Calc. for C20H15NOؒ0.25H2O: C, 82.9; H,
5.4; N, 4.8%. 1H NMR (CDCl3, 250 MHz, 298 K): δ 8.76 (1H,
d, J = 4.6 Hz, H6), 8.33 (1H, d, J = 16.3 Hz, Hc/d), 8.19 (1H, d,
J = 7.0 Hz, H3), 7.98 (1H, d, J = 16.0 Hz, Hc/d), 7.88 (1H, td,
J = 7.9, 1.8 Hz, H4), 7.80 (2H, d, J = 8.4 Hz, Ho/m/a/b), 7.64
(4H, dd, J = 8.4, 6.1 Hz, Ho/m/a/b), 7.50 (4H, m, Hp,5,o/m/a/b).
[Co(biptpy)2][PF6]2
Biptpy (0.0201 g, 0.05 mmol) and cobalt() acetate (0.0064 g,
0.026 mmol) were heated to reflux in methanol (15 cm3) for
12 h. The resulting red-coloured solution was cooled and
treated with saturated methanolic ammonium hexafluorophos-
phate. A red precipitate separated and was isolated by filtration
(0.020 g, 71%). The red precipitate was recrystallised from
acetonitrile by the slow diffusion of diethyl ether containing
tetrahydrofuran to afford red crystals of [Co(biptpy)2][PF6]2.
IR (KBr): ν 1616s, 836s, 767w, 559m, 466m cmϪ1. MS (ϩve
FAB): m/z 444 [Co(biptpy)], 829 [Co(biptpy)2], 974 [Co-
(biptpy)2(PF6)]. Found: C, 57.4; H, 3.5; N, 7.3. Calc. for
CoC54H38N6P2F12ؒ0.5H2O: C, 57.5; H, 3.5; N, 7.4%. UV–Vis
(MeCN): λmax (ε) 236 (43 000), 284 (46 000), 324 (50 000), 445
(3400), 518 nm (3000 cmϪ1 molϪ1). 1H NMR (CD3CN, 400 MHz,
298 K): δ 92, 54, 42, 32, 14, 10–7.
The tetrafluoroborate salt was prepared in 83% yield in an
analogous fashion, precipitating by addition of a saturated
methanolic solution of ammonium tetrafluoroborate.
[Ni(biptpy)2][PF6]2
Biptpy (0.0202 g, 0.05 mmol) and nickel() acetate (0.0065 g,
0.026 mmol) were heated to reflux in methanol (15 cm3) for
12 h. The resulting green solution was cooled and treated with
saturated methanolic ammonium hexafluorophosphate. On
cooling, a green solid separated and was isolated by filtration
(0.020 g, 69%). IR (KBr): ν 1616s, 1569w, 1546w, 1473s, 1427s,
1400m, 1384w, 1303w, 1245m, 1164w, 1018m, 833s, 794m, 767s,
732m, 701m, 659w, 559s cmϪ1. MS (ϩve FAB): m/z 443 [Ni(bip-
tpy)], 973 [Ni(biptpy)2(PF6)]. Found: C, 56.3; H, 3.4; N, 7.0.
4Ј(4-Biphenyl)-2,2Ј:6Ј,2Љ-terpyridine (biptpy)
1-(2Ј-Pyridyl)-3-(4Љ-biphenyl)-2-propen-1-one (2.000 g, 7.0
mmol), N-{2-(2Ј-pyridyl)-2-oxoethyl}pyridinium iodide (2.282
g, 7.0 mmol) and ammonium acetate (5 g, excess) in ethanol
(80 cm3) were heated under reflux for 12 h. On cooling, a green
solid precipitated and was collected by vacuum filtration and
recrystallised from ethanol in the presence of activated charcoal
to yield biptpy as an off-white solid (1.132 g, 42%). IR (KBr):
ν 1591s, 1582s, 1538w, 1569m, 1442w, 1388s, 790s, 763s, 721m,
682m cmϪ1. MS (ϩve FAB): m/z 386 [M ϩ 1]. Found: C, 83.3;
H, 4.9; N, 10.8. Calc. for C27H19N3ؒ0.25H2O: C, 83.2; H, 5.0; N,
10.8%. UV–Vis (MeOH): λmax (ε) 253 (74 000), 284 (13 000 cmϪ1
1
Calc. for NiC54H38N6P2F12ؒ2H2O: C, 56.1; H, 3.7; N, 7.3%. H
NMR (CD3CN, 400 MHz, 298 K): δ 135, 74, 67, 44, 13, 11–0.
The tetrafluoroborate salt was also prepared, in 66% yield, in
an analogous fashion by precipitating via addition of a satur-
ated methanolic solution of ammonium tetrafluoroborate.
1
molϪ1). H NMR (CDCl3, 250 MHz, 298 K): δ 8.78 (2H, s,
[Cu(biptpy)2][PF6]2
H3Ј), 8.73 (2H, d, J = 4.7 Hz, H6), 8.69 (2H, d, J = 8.1 Hz, H3),
8.01 (2H, d, J = 8.4 Hz, Ha/b), 7.89 (2H, td, J = 7.6, 1.8 Hz,
H4), 7.74 (2H, d, J = 8.7 Hz, Ha/b), 7.67 (2H, d, J = 7.0 Hz, Ho/
m), 7.47(3H, m, Hp,o/m), 7.35(2H, ddd, J = 7.6, 4.9, 1.2Hz, H5).
Biptpy (0.0203 g, 0.05 mmol) and copper() acetate (0.0052 g,
0.026 mmol) were heated to reflux in methanol (15 cm3) for
12 h. The resulting green solution was cooled and treated with
saturated methanolic ammonium hexafluorophosphate. On
cooling, a green solid separated and was isolated by filtration
(0.023 g, 79%). IR (KBr): ν 1619s, 1573m, 1477s, 1427m, 1400w,
1384w, 1249m, 1164w, 1022m, 848s, 794w, 767m, 728w, 690w,
859w, 559s cmϪ1. MS (ϩve FAB): m/z 448 [Cu(biptpy)], 833
[Cu(biptpy)2], 978 [Cu(biptpy)2(PF6)]. Found: C, 56.8; H, 3.4;
[Fe(biptpy)2][PF6]
Biptpy (0.0205 g, 0.05 mmol) and iron() sulfate (0.0074 g,
0.026 mmol) were heated to reflux in methanol (15 cm3) for
12 h. The resulting purple solution was cooled and treated with
J. Chem. Soc., Dalton Trans., 2000, 1447–1461
1457

View Article Online
N, 7.0. Calc. for CuC54H38N6P2F12ؒ1.25H2O: C, 56.6; H, 3.6; N,
7.3%.
The tetrafluoroborate salt was prepared in 77% yield in an
analogous fashion, precipitating by addition of a saturated
methanolic solution of ammonium tetrafluoroborate.
b,6), 7.46 (1H, m, Hp), 7.29 (2H, ddd, J = 7.6, 5.8, 1.5 Hz,
H5).
The chloride salt was prepared in 54% yield in an analogous
fashion, precipitating by addition of a saturated methanolic
solution of lithium chloride, and the tetrafluoroborate salt
prepared in 58% yield by addition of a saturated methanolic
solution of ammonium tetrafluoroborate.
[Ru(biptpy)(tpy)][PF6]2
Biptpy (0.0201 g, 0.05 mmol) and [Ru(tpy)Cl3] (0.0229 g, 0.05
mmol) were heated to reflux in methanol (15 cm3) containing a
few drops of 4-ethylmorpholine for 4 h. The resulting red solu-
tion was cooled and treated with saturated methanolic ammo-
nium hexafluorophosphate. On cooling, a red solid separated
and was isolated by filtration (0.031 g, 59%). IR (KBr): ν 1604s,
1450s, 1388m, 1307w, 1245m, 1164w, 840s, 790w, 767s, 732m,
698m, 559s cmϪ1. MS (ϩve FAB): m/z 334 [Ru(tpy)], 486
[Ru(biptpy)], 719 [Ru(tpy)(biptpy)], 865 [Ru(tpy)(biptpy)(PF6)].
Found: C, 48.3; H, 3.0; N, 7.9. Calc. for RuC42H30N6P2F12ؒ
2H2O: C, 48.2; H, 3.3; N, 8.0%. UV–Vis (MeCN): λmax (ε)
231 (28 000), 272 (35 000), 283 (32 000), 308 (54 000), 329
[Zn(biptpy)2][PF6]2
Biptpy (0.0200 g, 0.05 mmol) and zinc() acetate (0.0057 g,
0.026 mmol) were heated to reflux in methanol (15 cm3) for
12 h. The resulting clear solution was cooled and treated with
saturated methanolic ammonium hexafluorophosphate. On
cooling, a cream solid separated and was isolated by filtration
(0.0184 g, 63%). IR (KBr): ν 1604s, 1573s, 1477s, 1427s, 836s,
794s, 767s, 559s cmϪ1. MS (ϩve FAB): m/z 834 [Zn(biptpy)2],
979 [Zn(biptpy)2(PF6)]. Found: C, 55.9; H, 3.4; N, 7.1. Calc. for
1
ZnC54H38N6P2F12ؒ2H2O: C, 55.8; H, 3.6; N, 7.2%. H NMR
(CD3CN, 250 MHz, 298 K): δ 9.05 (2H, s, H3Ј), 8.75 (2H, d,
J = 8.1 Hz, H3), 8.33 (2H, d, J = 8.8 Hz, Ha/b), 8.19 (2H, td,
J = 7.5, 1.5 Hz, H4), 8.05 (2H, d, J = 8.7 Hz, Ha/b), 7.85
(4H, m, H6, o/m), 7.61 (3H, m, H,o/m), 7.42 (2H, ddd, J = 7.6,
5.2, 1.2 Hz, H5). UV–Vis (MeOH): λmax (ε) 238 (67 000), 285
(63 000), 336 nm (83 000 cmϪ1 molϪ1).
The tetrafluoroborate salt was prepared in 69% yield in
an analogous fashion, precipitating by addition of a saturated
ethanolic solution of ammonium tetrafluoroborate.
1
(35 000), 484 nm (17 000 cmϪ1 molϪ1). H NMR (CD3CN, 250
MHz, 298 K): δ 9.06 (2H, s, H3Ј), 8.75 (2H, d, J = 8.1 Hz,
H3Јtpy), 8.66 (2H, d, J = 7.8 Hz, H3), 8.50 (2H, d, J = 7.1 Hz,
H3tpy), 8.41 (1H, t, J = 8.0 Hz, H4Јtpy), 8.32 (2H, d, J = 8.7
Hz, Ha), 8.06 (2H, d, J = 8.7 Hz, Hb), 7.97–7.89 (4H, m,
H4,4tpy), 7.86 (2H, d, J = 7.0 Hz, Ho), 7.59 (2H, d, J = 7.0 Hz,
Hm), 7.50 (1H, t, J = 7.0 Hz, Hp), 7.43 (2H, d, J = 5.0 Hz,
H6tpy), 7.35 (2H, d, J = 4.7 Hz, H6), 7.17 (4H, m, H5,5tpy). 1H
NMR (CD3OD, 250 MHz, 298 K): δ 9.38 (2H, s, H3Ј), 9.02
(2H, d, J = 8.2 Hz, H3Јtpy), 8.97 (2H, d, J = 7.9 Hz, H3/3tpy),
8.77 (2H, d, J = 7.9 Hz, H3/3tpy), 8.35 (1H, t, J = 8.1 Hz,
H4Јtpy), 8.47 (2H, d, J = 8.7 Hz, Ha/b), 8.04 (6H, m, Ha/
b,4,4tpy), 7.87 (2H, dd, J = 8.4, 1.5 Hz, Ho/m), 7.58 (7H, m, Ho/
m,p,6,6tpy), 7.32 (4H, t, J = 6.7 Hz, H5,5tpy).
[Cd(biptpy)2][PF6]2
Biptpy (0.020 g, 0.05 mmol) cadmium() acetate (0.007 g, 0.026
mmol) were heated to reflux in methanol (15 cm3) for 12 h. The
resulting colourless solution was cooled and treated with
saturated methanolic ammonium hexafluorophosphate. On
cooling, a cream solid precipitated and was isolated by fil-
tration (0.042 g, 72%). IR (KBr): ν 1600m, 1477m, 840s, 794m,
767m, 559m cmϪ1. MS (ϩve FAB): m/z 882 [Cd(biptpy)2], 1029
[Cd(biptpy)2(PF6)]. Found: C, 53.1; H, 3.3; N, 7.1. Calc. for
The chloride salt was prepared in 53% yield in an analogous
fashion, precipitating by addition of a saturated methanolic
solution of lithium chloride.
1
[Ru(biptpy)Cl3]
CdC54H38N6P2F12ؒ2.5H2O: C, 53.2; H, 3.6; N, 6.9%. H NMR
(CD3CN, 250 MHz, 298 K): δ 9.01 (2H, s, H3Ј), 8.80 (2H, d,
J = 8.0 Hz, H3), 8.28 (2H, d, J = 8.7 Hz, Ha/b), 8.23 (2H, td,
J = 7.9, 1.5 Hz, H4), 8.10 (2H, d, J = 4.7 Hz, H6), 8.03 (2H, d,
J = 8.7 Hz, Ha/b), 7.85 (2H, dd, J = 8.4, 1.5 Hz, Ho/m), 7.55
(5H, m, H5, p,o/m). UV–Vis (MeCN): λmax (ε) 236 (60 000),
285 (59 000), 326 nm (70 000 cmϪ1 molϪ1). UV–Vis (MeOH): λmax
(ε) 233 (37 000), 286 (46 000), 322 nm (36 000 cmϪ1 molϪ1).
The tetrafluoroborate salt was prepared in 73% in an
analogous fashion, precipitating by addition of a saturated
methanolic solution of ammonium tetrafluoroborate.
Biptpy (0.1008 g, 0.026 mmol) and ruthenium() trichloride
(0.0543 g, 0.026 mmol) were heated to reflux in ethanol (25
cm3) for 1 h to give a red solution. On cooling a red solid
of [Ru(biptpy)Cl3] separated and was isolated by filtration. It
was washed with ice-cold ethanol and taken onto the next step
without further purification (0.1427 g, 92%).
[Ru(biptpy)2][PF6]2
Biptpy (0.0200 g, 0.05 mmol) and [Ru(biptpy)Cl3] (0.0307 g,
0.05 mmol) were heated to reflux in methanol (15 cm3) contain-
ing a few drops of 4-ethylmorpholine for 4 h. The resulting red
solution was cooled and treated with saturated methanolic
solution of ammonium hexafluorophosphate. On cooling, a
brown solid separated and was isolated by filtration (0.0373 g,
62%). The red solid was recrystallised from acetonitrile by the
slow diffusion of benzene. IR (KBr): ν 2985m, 2939m, 2726m,
2692m, 2596w, 1616m, 1605m, 1469m, 1430w, 1110s, 921w,
833s, 771w, 559s cmϪ1. MS (ϩve FAB): m/z 486 [Ru(biptpy)],
872 [Ru(biptpy)2], 1017 [Ru(biptpy)2(PF6)]. Found: C, 56.4;
H, 3.5; N, 6.9. Calc. for RuC54H38N6P2F12ؒ0.25C6H6: C, 56.4;
H, 3.4; N, 7.1%. UV–Vis (MeCN): λmax (ε) 238 (66 000), 279
X-Ray crystallography
The crystallographic data for all the compounds examined are
collected in Table 7.
[Co(biptpy)2][PF6]2 and [Ni(biptpy)2][PF6]2. Data were meas-
ured on a Siemens SMART52 three-circle system with CCD
area detector using the oil-mounting method at 180(2) K (main-
tained with the Oxford Cryosystems Cryostream Cooler).53
Absorption correction by Psi-scan. The structures were solved
by direct methods using SHELXS54 (TREF).
1
(100 000), 314 (110 000), 494 (45 000 cmϪ1 molϪ1). H NMR
(CD3CN, 250 MHz, 298 K): δ 9.09 (2H, s, H3Ј), 8.70 (2H, d,
J = 8.2 Hz, H3), 8.32 (2H, d, J = 8.7 Hz, Ha/b), 8.05 (2H, d,
J = 8.4 Hz, Ha/b), 7.97 (2H, td, J = 1.5, 7.9 Hz, H4), 7.86 (2H,
d, J = 6.9 Hz, Ho/m), 7.52 (3H, m, Ho/m,p), 7.46 (2H, d, J = 4.1
[Co(biptpy)2][PF6]2. Crystal character: red blocks. The
crystals of the cobalt() complex were severely twinned, but the
selected crystal had one dominant component. The data were,
however, significantly contaminated by spurious reflections,
which undoubtedly explains the relatively high R-value
obtained. Systematic absences indicated either space group Cc
or C2/c. The former was chosen for structure solution and
shown to be correct by successful refinement. Inspection of the
1
Hz, H6), 7.20 (2H, ddd, J = 7.9, 5.8, 1.5 Hz, H5). H NMR
(CD3OD, 250 MHz, 298 K): δ 9.35 (2H, s, H3Ј), 8.92 (2H, d,
J = 7.8 Hz, H3), 8.42 (2H, d, J = 8.7 Hz, Ha/b), 8.04 (4H, m,
H4,a/b), 7.82 (2H, dd, J = 1.5, 8.4 Hz, Ho/m), 7.58 (4H, m, Ha/
1458
J. Chem. Soc., Dalton Trans., 2000, 1447–1461

View Article Online
J. Chem. Soc., Dalton Trans., 2000, 1447–1461
1459

View Article Online
final structure did not reveal any additional symmetry. The
asymmetric unit includes one molecule of tetrahydrofuran. The
absolute structure of the individual crystal chosen was checked
by refinement of a delta-fЉ multiplier. Absolute structure par-
ameter x = 0.10(4). This gives a rather weak indication that the
absolute structure is correct. Floating origin constraints were
generated automatically. Refinement used SHELXL96.55
were refined as oxygen atoms with population parameters
chosen to create reasonable temperature factors (<0.2).
CCDC reference number 186/1899.
See http://www.rsc.org/suppdata/dt/b0/b000871k/ for crystal-
lographic files in .cif format.
Acknowledgements
[Ni(biptpy)2][PF6]2. Crystal character: yellow-orange plates.
The crystals tend to be twinned, but the crystal used was pri-
marily a single component. Systematic absences indicated either
space group Pbam or Pba2. The second was chosen initially,
and the structure solved. Inspection revealed the presence of
mirror symmetry and the structure was transformed into Pbam.
The Ni, both P atoms and several F atoms lie on 2-fold axes.
The anions show indications of severe disorder and two low-
occupancy F atoms and a number of partial occupancy solvent
atoms were added to compensate for this and for a disordered
solvent molecule (possibly ether). The P–F and F ؒ ؒ ؒ F dis-
tances were constrained to standard values. The relatively
high final R-value is readily understandable in the light of the
partial twinning and the severe disorder. Refinement used
SHELXTL.56
We thank the EPSRC (C. L. P.; P. R. B.; J. H. GR/L91788)
and the Finnish Ministry of Education (P. S.) for support,
the Royal Society for the purchase of the electrochemical
workstation (M. J. H.) and the EPSRC and Siemens Anal-
ytical Instruments for grants in support of a diffractometer
(N. W. A.). We are grateful to the EPSRC National Mass
Spectrometry Service Centre, Swansea for recording some of
the mass spectra and acknowledge the use of the EPSRC’s
Chemical Database Service at Daresbury.62
References and notes
1 K. Kalyasundaran, Photochemistry of Polypyridine and Porphyrin
Complexes, Academic Press, London, 1992; A. Juris, V. Balzani,
F. Barigeletti, S. Campagna, P. Belser and A. Von Zelewsky, Coord.
Chem. Rev., 1988, 84, 85; J. E. Moser, P. Bonnôte and M. Gratzel,
Coord. Chem. Rev., 1998, 171, 245.
[Cd(biptpy)2][PF6]2, [Zn(biptpy)2][BF4]2, [Ru(biptpy)2][BF4]2,
[Cu(biptpy)2][PF6]2 and [Ru(tpy)(biptpy)]Cl2. Data were meas-
ured using a Nonius Kappa CCD diffractometer. Data were
collected and processed with the HKL package of programs.57
The cadmium and Ru(biptpy)2 complexes were solved with
SHELXS-9758 and the zinc, copper and Ru(tpy)(biptpy) com-
plexes with SIR92.59 The structures were refined by full-matrix
least squares on F2 with the SHELXL-97 programs.58 The
SIR92 and SHELX-97 programs were used within the WinGX
suite of programs.60 Absorption correction for the cadmium,
zinc and Ru(biptpy)2 complexes was done using SORTAV61
but the HKL package produced better data in these structures.
The HKL package was used in final refinements of all the
structures.
The cadmium complex [Cd(biptpy)2][PF6]2 crystallised with
benzene and nitromethane solvent molecules and the zinc com-
plex [Zn(biptpy)2][BF4]2 with acetonitrile and water. In the case
of the cadmium complex the nitromethane solvent molecule
was disordered over two sites; it was refined isotropically with
geometrical restraints. In the zinc complex both BF4 anions
were also disordered over two sites. Geometrical restraints were
used for the BF4 tetrahedra. In the zinc complex, a water
molecule was located (hydrogen atoms not located).
The [Ru(biptpy)2][BF4]2 complex crystallised with nitro-
methane and water solvent molecules. Only the solvent mole-
cules were disordered. One nitromethane is on a mirror plane.
The second and third nitromethane molecules are disordered
over two sites (refined isotropically with geometrical restraints).
One water molecule is disordered over four sites.
The copper complex [Cu(biptpy)2][PF6]2 crystallised with
nitromethane, methanol and benzene solvent molecules. The
crystal quality was poor and the poor final R-value is attributed
to the weak reflection data (Rint = 0.125, Rsig = 0.23) and dis-
order of anions and solvent molecules. One of the PF6 anions
was disordered over two sites. In the other PF6 anion one
fluorine atom was divided over two sites. Geometrical restraints
were used to keep P–F distances chemically reasonable. The
nitromethane molecule situated on a mirror plane was dis-
ordered over two sites. A benzene molecule was found on a
symmetry element. The methanol molecule was found and
refined with population parameter of 0.5.
2 V. Balzani, S. Campagna, G. Denti, A. Juris, S. Serroni and
M. Venturi, Acc. Chem. Res., 1998, 31, 26.
3 E. C. Constable, Chem. Commun., 1997, 1073.
4 J.-P. Sauvage, J.-P. Collin, J.-C. Chambron, S. Guillerez and
C. Coudret, Chem. Rev., 1994, 94, 993.
5 A. Von Zelewsky and P. Belser, Chimia, 1998, 52, 620; P. Belser,
S. Berhard, C. Blum, A. Beyeler, L. DeCola and V. Balzani, Coord.
Chem. Rev., 1999, 192, 155.
6 M. Hissler, A. El-ghayoury, A. Harriman and R. Ziessel, Angew.
Chem., Int. Ed., 1998, 37, 1717; A. Harriman and R. Ziessel, Coord.
Chem. Rev., 1998, 171, 331.
7 M. Schutte, D. G. Kurth, M. R. Linford, H. Cölfen and H.
Möhwald, Angew. Chem., Int. Ed., 1998, 37, 2891.
8 S. Kelch and M. Rehahn, Macromolecules, 1999, 32, 5818.
9 M. D. Ward, Chem. Soc. Rev., 1995, 121; V. Balzani and F. Scandola,
Supramolecular Photochemistry, Ellis Horwood, Chichester, UK,
1991; V. Balzani, A. Juris, M. Venturi, S. Campagna and S. Serroni,
Chem. Rev., 1996, 96, 759.
10 E. C. Constable and R.-A. Fallahpour, J. Chem. Soc., Dalton Trans.,
1996, 2389.
11 N. Armaroli, F. Barigelletti, G. Calogero, L. Flamigni, C. M.
White and M. D. Ward, Chem. Commun., 1997, 2181; M. D. Ward,
C. M. White, F. Barigelletti, N. Armaroli, G. Calogero and
L. Flamigni, Coord. Chem. Rev., 1998, 171, 481.
12 D. J. Hurley and Y. Tor, J. Am. Chem. Soc., 1998, 120, 2194.
13 J. L. Sessler, C. T. Brown, R. Z. Wong and T. Hirose, Inorg. Chim.
Acta, 1996, 251, 135.
14 P. G. Desmartin, A. F. Williams and G. Bernardinelli, New J. Chem.,
1995, 19, 1109.
15 G. R. Desiraju, Crystal Engineering; The Design of Organic Solids,
Elsevier, New York, 1989; G. R. Desiraju and A. Gavezotti, J. Chem.
Soc., Chem. Commun., 1989, 621; G. R. Desiraju, Chem. Commun.,
1997, 1475.
16 E. Ishow, A. Gourdon and J.-P. Launay, Chem. Commun., 1998,
1909.
17 H. K. Yip, L. K. Cheng, K. K. Cheung and C. M. Che, J. Chem.
Soc., Dalton Trans., 1993, 2933.
18 W. R. McWhinnie and J. D. Miller, Adv. Inorg. Chem. Radiochem.,
1969, 12, 135.
19 E. C. Constable, Adv. Inorg. Chem. Radiochem., 1986, 30, 69.
20 F. Kröhnke, Synthesis, 1976, 1.
21 E. C. Constable, A. M. W. Thompson, T. Cargill, D. A. Tocher and
M. A. M. Daniels, New J. Chem., 1992, 16, 855.
22 M. L. Scudder, H. A. Godwin and I. G. Dance, New J. Chem., 1999,
23, 695.
23 B. Whittle, E. L. Horwood, L. H. Rees, S. R. Batten, J. C. Jeffery and
M. D. Ward, Polyhedron, 1998, 17, 373.
24 B. Whittle, E. L. Horwood, S. R. Batten, J. C. Jeffrey, L. H. Rees and
M. D. Ward, J. Chem. Soc., Dalton Trans., 1996, 4269.
25 E. C. Constable, C. E. Housecroft, M. Neuberger, A. G. Schneider
and M. Zehnder, J. Chem. Soc., Dalton Trans., 1997, 2427.
26 M.-C. Tse, K.-K. Cheung, M. C.-W. Chan and C.-M. Che, Chem.
Commun., 1998, 2295.
In the ruthenium mixed ligand complex, [Ru(tpy)(biptpy)]-
Cl2, the crystal quality was also poor and the solvent molecules
extremely disordered. Solvent electron density was spread
around the Ru() complex and could not be identified as separ-
ate molecules, having many short contacts. All solvent peaks
1460
J. Chem. Soc., Dalton Trans., 2000, 1447–1461

View Article Online
27 Only long face–edge interactions (H ؒ ؒ ؒ centroid > 3.5 Å) are
observed between the chains and between the planes.
28 F. Takusagawa, P. G. Yohannes and K. Mertes, Inorg. Chim. Acta,
1986, 114, 165.
29 W. Henke and S. Kremer, Inorg. Chim. Acta, 1982, 65, L115.
30 E. N. Maslen, C. L. Raston and A. H. White, J. Chem. Soc., Dalton
Trans., 1974, 1803.
44 G. Albano, V. Balzani, E. C. Constable, M. Mastri and D. R. Smith,
Inorg. Chim. Acta, 1998, 277, 225.
45 K.-Y. Ho, W.-Y. Yu, K.-K. Cheung and C.-M. Che, J. Chem. Soc.,
Dalton Trans., 1999, 1581.
46 J.-P. Collin, I. M. Dixon, J.-P. Sauvage, J. A. G. Williams,
F. Barigelletti and L. Flamigni, J. Am. Chem. Soc., 1999, 121,
5009.
31 C. L. Raston and A. H. White, J. Chem. Soc., Dalton Trans., 1976, 7.
32 B. N. Figgis, E. S. Kucharski and A. H. White, Aust. J. Chem., 1983,
36, 1527.
33 B. N. Figgis, E. S. Kucharski and A. H. White, Aust. J. Chem., 1983,
36, 1537.
34 E. C. Constable, J. Lewis, M. C. Liptrot and P. R. Raithby, Inorg.
Chim. Acta, 1990, 178, 47.
47 K.-Y. Ho, W.-Y. Yu, K.-K. Cheung and C.-M. Che, Chem.
Commun., 1998, 2101.
48 The spike on the solid state luminescence spectrum is scattered
light.
49 M. J. Hannon, C. L. Painting and W. Errington, Chem. Commun.,
1997, 307; M. J. Hannon, C. L. Painting and W. Errington, Chem.
Commun., 1997, 1805.
35 See for example K. Lashgari, M. Kritikos, R. Norrestam and
T. Norrby, Acta Crystallogr., Sect. C, 1999, 55, 64; M.
Beley, J.-P. Collin, R. Louis, B. Metz and J.-P. Sauvage, J. Am.
Chem. Soc., 1991, 113, 8521; K. L. Bushell, S. M. Couchman,
J. C. Jeffery, L. H. Rees and M. D. Ward, J. Chem. Soc., Dalton
Trans., 1998, 3397; M. Ziegler, V. Monney, H. Stoeckli-Evans,
A. Von Zelewsky, I. Sasaki, G. Dupic, J.-C. Daran and G. A.
Balavoine, J. Chem. Soc., Dalton Trans., 1999, 667.
36 The structure of a dimetallic cadmium() double-helical complex
of 2,2Ј;6Ј,2Љ;6Ј,2ٞ;6ٞ,2٣;6٣,2Љٞ-sexipyridine has been reported:
E. C. Constable, M. D. Ward and D. A. Tocher, J. Chem. Soc.,
Dalton Trans., 1991, 1675. A mono-terpyridyl complex [Cd(tpy)-
{Mn(CO)5}2] has been characterised: W. Clegg and P. J. Wheatley,
J. Chem. Soc., Dalton Trans., 1973, 190. We have recently crystal-
lographically characterised a cadmium complex of the tris-pyridyl
ligand 4-thiomethyl-6-(3Љ-pyridyl)-2,2Ј-bipyridine: M. J. Hannon,
C. L. Painting and W. Errington, Chem. Commun., 1997, 307.
37 E. C. Constable, M. Neuberger, D. R. Smith and M. Zehnder, Inorg.
Chim. Acta, 1998, 275–6, 359.
50 S. Weidner and Z. Pikramenou, Chem. Commun., 1998, 1473;
M. Chavarot and Z. Pikramenou, Tetrahedron Lett., 1999, 40,
6865.
51 J. N. Demas and G. A. Crosby, J. Phys. Chem., 1971, 75, 991.
52 SMART User’s manual, Siemens Industrial Automation Inc,
Madison, WI, 1994.
53 J. Cosier and A. M. Glazer, J. Appl. Crystallogr., 1986, 19, 105.
54 G. M. Sheldrick, Acta Crystallogr., Sect. A, 1990, 46, 467.
55 G. M. Sheldrick, SHELX-96 (beta-test) (including SHELXS and
SHELXL), University of Göttingen, 1996.
56 G. M. Sheldrick, SHELXTL Ver. 5.1, Bruker Analytical X-Ray
Systems, Madison, WI, 1997.
57 Z. Otwinowski and W. Minor, Methods Enzymol., 1997, 276,
307.
58 G. M. Sheldrick, SHELX-97: Programs for Crystal Structure
Analysis (Release 97-2), Institüt für Anorganische Chemie der
Universität, Tammanstrasse 4, D-3400 Göttingen, Germany, 1998.
59 A. Altomare, G. Cascarano, C. Giacovazzo and A. Guagliardi,
J. Appl. Crystallogr., 1993, 26, 343.
38 E. C. Constable, R. Martinez-Manez, A. M. W. Cargill Thompson
and J. V. Walker, J. Chem. Soc., Dalton Trans., 1994, 1585.
39 E. C. Constable, P. Haverson, D. R. Smith and L. Whall,
Polyhedron, 1997, 16, 3615.
40 J. M. Rao, D. J. Macero and M. C. Hughes, Inorg. Chim. Acta, 1980,
41, 221.
41 E. C. Constable and A. M. W. Cargill Thompson, J. Chem. Soc.,
Dalton. Trans., 1994, 1409.
42 M. Maestri, N. Armaroli, V. Balzani, E. C. Constable and A. M. W.
Cargill Thompson, Inorg. Chem., 1995, 34, 2759.
60 L. J. Farrugia, WinGX: A Windows Program for Crystal Structure
Analysis, University of Glasgow, 1998.
61 R. H. Blessing, Acta Crystallogr., Sect. A, 1995, 51, 33; R. H.
Blessing, J. Appl. Crystallogr., 1997, 30, 421.
62 D. A. Fletcher, R. F. McMeeking and D. Parkin, J. Chem. Inf.
Comput. Sci., 1996, 36, 746; F. H. Allen and O. Kennard, Chem. Des.
Automat. News, 1993, 8, 1 and 31.
63 NMR software manual, Bruker analytik GmbH, 1997.
64 GPES for windows, version 4.2, Eco Chemie B.V., Utrecht, The
Netherlands, 1995.
43 D. K. Liu, B. S. Brunschwig, C. Creutz and N. Sutin, J. Am. Chem.
Soc., 1986, 108, 1750.
65 Felix Software, version 1.1, Photon Technology International Inc.,
Brunswick, USA, 1996.
J. Chem. Soc., Dalton Trans., 2000, 1447–1461
1461