Tuning solid emission by salts (see abstract)

Article abstract of DOI:10.1039/b107113k

The synthesis and charcterization of 4′-(2?-methylphenyl)-2,2′:6′,2″-terpyridine[4 ′-(o-CH₃Ph)trpy] and 4′-(2?-trifluoromethylphenyl)-2,2′:6′,2″-terpyrid ine[4′-(o-CF₃-Ph)trpy] were presented. The roles of the ortho-substituents (CH<

Full text of DOI:10.1039/b107113k

View Article Online / Journal Homepage / Table of Contents for this issue
Tuning solid emission by salts of the [Pt{4Ј-(o-CH3–Ph)trpy}Cl]؉
and [Pt{4Ј-(o-CF3–Ph)trpy}Cl]؉ luminophores: crystal structures
of [Pt{4Ј-(o-CH3–Ph)trpy}Cl]A (A ؍ BF4 or SbF6) and [Pt{4Ј-
(o-CF3–Ph)trpy}Cl]SbF6 (trpy ؍ 2,2Ј:6Ј,2Љ-terpyridine)†
John S. Field,a Raymond J. Haines,a David R. McMillinb and Grant C. Summertona
a School of Chemical and Physical Sciences, University of Natal, Private Bag X01,
Pietermaritzburg, South Africa 3201. E-mail: field@nu.ac.za
b Department of Chemistry, Purdue University, West Lafayette, Indiana 47907-1393, USA
Received 6th August 2001, Accepted 18th December 2001
First published as an Advance Article on the web 6th March 2002
The synthesis and characterisation of 4Ј-(2ٞ-methylphenyl)-2,2Ј:6Ј,2Љ-terpyridine [4Ј-(o-CH3–Ph)trpy] and
4Ј-(2ٞ-trifluoromethylphenyl)-2,2Ј:6Ј,2Љ-terpyridine [4Ј-(o-CF3–Ph)trpy] are described. Reaction of these ligands
with [Pt(PhCN)2Cl2] in the presence of the appropriate silver salt afforded [Pt{4Ј-(o-CH3–Ph)trpy}Cl]A and [Pt{4Ј-
(o-CF3–Ph)trpy}Cl]A (A = BF4 or SbF6). The crystal structure of the SbF6Ϫ salt of the [Pt{4Ј-(o-CH3–Ph)trpy}Cl]ϩ
cation consists of columns of cations and anions with the cations stacked parallel and head-to-tail and with a constant
Pt ؒ ؒ ؒ Pt distance [3.368(1) Å] and interplanar spacing (3.36 Å) along the stack. Crystals of the BF4Ϫ salt also contain
columnsofcationsandanionswiththecationsstackedparallelandhead-to-tail.However,theplatinumatomsofsuccessive
cation pairs are offset to different extents with respect to a line drawn perpendicular to the stack. As a result the Pt ؒ ؒ ؒ Pt
distances along the stack alternate between 3.573(1) and 3.827(1) Å while the interplanar spacing is within experi-
mental error, constant at an average value of 3.42 Å. The [Pt{4Ј-(o-CF3–Ph)trpy}Cl]SbF6 salt exhibits the same
packing motif but with alternating Pt ؒ ؒ ؒ Pt distances of 3.629(1) and 3.685(1) Å and an average interplanar spacing
of 3.46 Å. The variable temperature emission spectra recorded on microcrystalline samples of these compounds
reflect the environments of the cations in the crystals. The red [Pt{4Ј-(o-CH3–Ph)trpy}Cl]SbF6 salt exhibits 3MMLCT
(MMLCT = metal–metal–ligand-charge-transfer) emission [λ(em)max = 616 nm] that is distinguished by a large red-
shift of ca. 60 nm in the emission maximum when the sample is cooled from room temperature to 80 K; such a
2
2
red-shift is associated with strong platinum dz –dz orbital interactions perpendicular to the cation stack. The other
three salts are yellow with broad featureless emission spectra that do not change position when the temperature is
lowered but which are red-shifted when compared to the low temperature dilute glass spectrum recorded for [Pt{4Ј-
(o-CH3–Ph)trpy}Cl]SbF6 in a dimethylformamide/methanol/ethanol (1 : 5 : 5) mixture; this is interpreted in terms of
excimeric emission consistent with the presence of π–π interactions in the solid. The roles of the ortho-substituents
(CH3 and CF3) and the anions (SbF6Ϫ and BF4Ϫ) in determining the precise arrangements of the cations in the
crystals and hence the photoluminescence properties of these materials are briefly discussed.
ligand-charge transfer).7 Both emission types typically exhibit
Introduction
red-shifts in their emission maxima compared to that observed
The solid state photoluminescence properties of mononuclear
when the emission is from the excited state of an isolated
2,2Ј-bipyridyl (bpy) and 2,2Ј:6Ј,2Љ-terpyridyl (trpy) ligand com-
luminophore. A further distinguishing feature of 3MMLCT
plexes of platinum() depend on the environment in which the
emission is a red-shift in the emission maximum when the
material is cooled, as evidenced by measurements over a range
luminophore finds itself in the crystalline state.1–10 Indeed, there
are few examples of compounds of this type that exhibit emis-
of temperatures of the emission spectra of the red form of
sion in the solid state that is not perturbed by the crystalline
[Pt(bpy)Cl2],5,8 the orange form of [Pt(trpy)Cl]CF3SO3,6 and the
environment in some way, the perchlorate salt of [Pt(bpy)(en)]2ϩ
red form of [Pt{4Ј-(Ph)trpy}]BF4.10 This occurs when cooling
being a rare exception.2 Of particular interest are compounds
of the crystal causes a shortening of the relevant Pt ؒ ؒ ؒ Pt
distances along the stack and a concomitant decrease in the
HOMO/LUMO energy gap. Crystal structure determinations
where the planar luminophores stack parallel to each other in
the crystal with significant electronic interactions between
them. Two such interactions have been identified: π–π inter-
of the red form of [Pt(bipy)Cl2] at 20 and 294 K substantiate
actions between the π-systems of adjacent ligands and platinum
these conclusions.8
dz –dz orbital interactions along the stack.1 The former gives
2
2
Of interest here are the solid state luminescence properties of
rise to emission from excited state aggregates as exemplified by
the solid state emission exhibited by [Pt(phen)2]Cl2ؒ3H2O (phen
= 1,10-phenanthroline). Strong platinum dz –dz orbital inter-
actions result in emission from 3[dσ*,π*] states previously
labelled as 3MMLCT emission (MMLCT = metal–metal–
Ϫ
Ϫ
the BF4 and SbF6 salts of the [Pt{4Ј-(o-CH3–Ph)trpy}Cl]ϩ
(1ϩ) and [Pt{4Ј-(o-CF3–Ph)trpy}Cl]ϩ (2ϩ) cations. The platinum
atoms are bound by a chloro group and by a tridentate ter-
pyridyl ligand substituted in the 4Ј-position by a phenyl group
itself substituted in an ortho-position by a bulky group, specifi-
cally by a methyl or trifluoromethyl group. Ortho-substitution
of the 4Ј-phenyl ring was chosen since steric interactions
between the bulky group and the 3Ј(5Ј)-proton on the central
pyridine ring are expected to force the 4Ј-substituent to rotate
around the interannular bond i.e., the cationic luminophore will
2
2
2
† Electronic supplementary information (ESI) available: Fig. S1–3: unit
cell contents of [1]SbF6 viewed down the [c]-axis and [1]BF4 and [2]SbF6
viewed down the [a]-axis. See http://www.rsc.org/suppdata/dt/b1/
b107113k/
DOI: 10.1039/b107113k
J. Chem. Soc., Dalton Trans., 2002, 1369–1376
This journal is © The Royal Society of Chemistry 2002
1369

View Article Online
Table 1 Selected interatomic distances (Å) and angles (Њ)
[Pt{4Ј-(o-CH3–Ph)trpy}Cl]SbF6, [1]SbF6
be rendered non-planar. We wished to establish what effect this
would have on the crystal structure and hence the emission,
noting that the trifluoromethyl group has a larger van der Waals
radius than the methyl group. A second variable is the size of
the counterion, and in this work we have chosen two essentially
spherical anions, one (BF4Ϫ) that is much smaller than the other
(SbF6Ϫ). In order to explain the changes in the solid state emis-
sion brought about by changing these two variables, the crystal
structures of [Pt{4Ј-(o-CH3–Ph)trpy}Cl]SbF6 ([1]SbF6), [Pt{4Ј-
(o-CH3–Ph)trpy}Cl]BF4 ([1]BF4) and [Pt{4Ј-(o-CF3–Ph)trpy}-
Cl]SbF6 ([2]SbF6) have been determined.
Pt–Cl
Pt–N(2)
C(8)–C(16)
2.296(3)
1.924(9)
1.50(2)
Pt–N(1)
Pt–N(3)
2.019(9)
2.027(11)
Cl–Pt–N(1)
Cl–Pt–N(3)
N(1)–Pt–N(3)
99.2(3)
99.0(2)
161.7(3)
Cl–Pt–N(2)
N(1)–Pt–N(2)
N(2)–Pt–N(3)
176.0(2)
81.2(3)
80.6(4)
[Pt{4Ј-(o-CH3–Ph)trpy}Cl]BF4, [1]BF4
Pt–Cl
Pt–N(2)
C(8)–C(16)
2.293(3)
1.935(9)
1.505(14)
Pt–N(1)
Pt–N(3)
2.025(11)
2.025(11)
Cl–Pt–N(1)
Cl–Pt–N(3)
N(1)–Pt–N(3)
98.8(3)
98.8(3)
162.3(4)
Cl–Pt–N(2)
N(1)–Pt–N(2)
N(2)–Pt–N(3)
179.2(3)
80.6(4)
81.7(4)
[Pt{4Ј-(o-CF3–Ph)trpy}Cl]SbF6, [2]SbF6
Pt–Cl
Pt–N(2)
C(8)–C(16)
C(22)–F(2)
2.286(3)
1.931(8)
1.516(14)
1.337(17)
Pt–N(1)
Pt–N(3)
C(22)–F(1)
C(22)–F(3)
2.004(10)
2.014(10)
1.306(17)
1.332(15)
Cl–Pt–N(1)
Cl–Pt–N(3)
N(1)–Pt–N(3)
99.0(3)
98.7(3)
162.3(3)
Cl–Pt–N(2)
N(1)–Pt–N(2)
N(2)–Pt–N(3)
179.9(3)
81.1(4)
81.2(4)
Results and discussion
The method of Kröhnke was employed for the synthesis
of 4Ј-(2ٞ-methylphenyl)-2,2Ј:6Ј,2Љ-terpyridine [4Ј-(o-CH3–Ph)-
trpy] and 4Ј-(2ٞ-trifluoromethylphenyl)-2,2Ј:6Ј,2Љ-terpyridine
[4Ј-(o-CF3–Ph)trpy] and is summarised in Scheme 1.11 Full
Fig. 1 ORTEP view of the cation (1ϩ) in [Pt{4Ј-(o-CH3–Ph)trpy}Cl]-
SbF6. Thermal ellipsoids are drawn at the 50% probability level except
for the hydrogen atoms that are shown as spheres of arbitrary radius.
and angles is given in Table 1. The coordination geometry of
the platinum atom is irregular square-planar, deviations from
the idealised geometry being evident in the N(1)–Pt–N(2) and
N(2)–Pt–N(3) angles of 81.2(3) and 80.6(4)Њ respectively. This is
due to the geometric constraints imposed by the tridentate
ligand and is typical of terpyridyl ligand complexes of
platinum().6,7,9,10,12 Also characteristic of a coordinated ter-
pyridyl ligand is that the platinum to bridgehead nitrogen [N(2)]
distance of 1.924(9) Å is significantly shorter than the two
platinum to outer nitrogen distances of 2.019(9) [N(1)] and
2.027(11) Å [N(3)] respectively.6,7,9,10,12 As expected the ter-
pyridyl moiety is essentially planar with typical values for the
internal bond lengths and angles.6,7,9,10,12 However, the cation as
a whole is distinctly non-planar, because the o-tolyl moiety is
twisted about the C(8)–C(16) bond, as reflected in a dihedral
angle between the mean plane through the carbon atoms of the
4Ј-substituent and those of the terpyridyl moiety of 65.2Њ. The
corresponding angle measured for free 4Ј-phenyl-2,2Ј:6Ј,2Љ-
terpyridine in the solid state is 10.9Њ,13 and for the same ligand
Scheme 1
details of the synthetic procedures as well as the characteris-
ation data for both the chalcone intermediates and the ligands
themselves are given in the Experimental section. Treatment of
a refluxing acetonitrile solution of [Pt(PhCN)2Cl2] with one
equivalent of AgBF4 or AgSbF6 afforded a yellow solution and
a white precipitate of silver chloride. After removal of the silver
chloride one equivalent of the ligand, [4Ј-(o-CH3–Ph)trpy] or
[4Ј-(o-CF3–Ph)trpy], was added to afford the desired product as
an analytically pure, air-stable crystalline solid. Full character-
isation data are given in the Experimental section. The colours
of the salts are as follows: [Pt{4Ј-(o-CH3–Ph)trpy}Cl]SbF6 is
red while [Pt{4Ј-(o-CH3–Ph)trpy}Cl]BF4, [Pt{4Ј-(o-CF3–Ph)-
trpy}Cl]BF4 and [Pt{4Ј-(o-CF3–Ph)trpy}Cl]SbF6 are yellow.
Crystal structure of [Pt{4Ј-(o-CH3–Ph)trpy}Cl]SbF6 ([1]SbF6)
Fig. 1 gives a perspective view of the cation as well as the atomic
labelling scheme. A list of the important interatomic distances
1370
J. Chem. Soc., Dalton Trans., 2002, 1369–1376

View Article Online
coordinated to platinum in [Pt{4Ј-(Ph)trpy}Cl]BF4ؒCH3CN
was determined as 33.4 and 33.3Њ for each cation in the asym-
metric unit.10 Clearly, substitution of a methyl group in the
ortho-position of the 4Ј-phenyl ring causes a larger rotation
about the interannular bond because of steric interactions
between the methyl group and a hydrogen atom of the central
pyridine ring that is also ortho with respect to the interannular
bond. The Pt–Cl distance of 2.296(3) Å is similar to those
reported for [Pt(trpy)Cl]CF3SO3 (2.307 Å),6 [Pt(trpy)Cl]ClO4
(2.302 Å)7 and [Pt{4Ј-(Ph)trpy}Cl]BF4ؒCH3CN (2.296 and
2.313 Å).10
The crystal structure of [Pt{4Ј-(o-CH3–Ph)trpy}Cl]SbF6 con-
sists of separate columns of cations and anions stacked parallel
to the [c]-axis of the unit cell (Fig. S1†). The cations stack in
such a way that the mean planes through the platinum, chlorine
and terpyridyl moiety atoms (i.e., excluding the atoms of
the 4Ј-group) are parallel to each other and perpendicular to
the [c]-axis. Fig. 2 illustrates one cation pair viewed perpendicu-
Pt ؒ ؒ ؒ Pt separation of 3.33 Å and an inter-tetramer separation
of 3.62 Å.10
Crystal structure of [Pt{4Ј-(o-CH3–Ph)trpy}Cl]BF4 ([1]BF4)
The cation atomic labelling scheme is the same as that illus-
trated in Fig. 1. A list of the important interatomic distances
and angles is given in Table 1. There are no significant
differences between these values and those listed in Table 1
Ϫ
for the SbF6 salt and the same comments therefore apply to
both sets of geometric parameters (vide supra). However, one
intramolecular parameter that is particularly sensitive to
packing effects and so possibly influenced by a change in the
size of the anion, is the dihedral angle between the mean
plane through the non-hydrogen atoms of the 4Ј-substituent
and the mean plane through the non-hydrogen atoms of the
terpyridyl moiety. This angle is 69.9Њ i.e., about 5Њ larger than
the corresponding angle of 65.2Њ measured for the SbF6Ϫ salt.
Ϫ
The BF4 salt also has the cations stacked parallel and
head-to-tail with the anions occupying channels between the
cation stacks (Fig. S2†). However, there are important differ-
Ϫ
ences of detail compared to those observed for the SbF6 salt.
First, the mean plane through the platinum, chlorine and ter-
pyridyl moiety atoms is within experimental error perpendicu-
lar to the [a]-axis with successive cations related by a centre of
inversion rather than a two-fold screw axis. Second, the
Pt ؒ ؒ ؒ Pt distances along the stack are not uniformly equal but
alternate between values of 3.573(1) and 3.827(1) Å. On the
other hand, the mean planes are within experimental error
equally spaced, with an averaged interplanar spacing of 3.42 Å.
The packing motif of [Pt{4Ј-(o-CH3–Ph)trpy}Cl]BF4 is there-
fore very similar to that reported for the triflate and per-
chlorate salts of the [Pt(trpy)Cl]ϩ cation.6,7 Fig. 3 gives views
perpendicular to the mean planes through the platinum, chlor-
ine and non-hydrogen atoms of the terpyridyl moiety for each
of the two different cation–cation interactions along the stack.
The extent of slippage of adjacent cations with respect to a line
perpendicular to the stack determines the Pt ؒ ؒ ؒ Pt distance for
Fig. 2 ORTEP diagram showing a cation–cation interaction in [Pt{4Ј-
(o-CH3–Ph)trpy}Cl]SbF6 perpendicular to the mean planes through the
platinum, chlorine and terpyridyl moiety atoms.
lar to the stack. As required by the presence of a two-fold screw
axis along [c] the platinum atoms along the stack are equally
spaced, an arrangement that is usually described as a linear
chain structure.1,8 The Pt ؒ ؒ ؒ Pt distance is 3.368(1) Å while
the Pt ؒ ؒ ؒ Pt ؒ ؒ ؒ Pt angle of 162Њ indicates that successive
platinum atoms are almost eclipsed when viewed down the
stacking axis (Fig. 2). Given that the Pt ؒ ؒ ؒ Pt distance is simi-
lar to values normally taken to indicate a strong σ-interaction
a pair of cations. The shorter distance of 3.573(1) Å is too long
14
2
to support effective overlap between the platinum dz -orbitals,
especially as the orbitals are not lined-up in the same direction
because of the lateral offset of adjacent cations. Clearly the
same is true for the pair of cations with the longer Pt ؒ ؒ ؒ Pt
distance of 3.827(1) Å. In so far as π–π interactions along the
stack are concerned the following comments apply. The
interplanar spacing of 3.42 Å is short enough to support π–π
interactions being well within the upper distance limit of 3.8 Å
for π-interactions between organic species.15 However, in the
absence of detailed calculations on the nature of the π-inter-
actions in stacked terpyridines, we are uncertain of the opti-
mum geometry for a favourable π-interaction. What is certain is
that even large lateral offsets in stacked porphyrins allow for a
relatively strong electrostatic attraction between the π-systems
of adjacent ligands.15 Also, the emission data obtained on
microcrystalline samples (vide infra) suggest that intermolecular
π–π interactions are important in determining the photo-
luminescence behaviour of [Pt{4Ј-(o-CH3–Ph)trpy}Cl]BF4.
between overlapping platinum dz -orbitals,1,14 this implies direct
2
and strong overlap of the platinum dz2-orbitals along the stack.
As described below this has a dramatic effect on the emission
properties of the solid. The perpendicular distance between the
mean planes through the platinum, chlorine and terpyridyl
moiety atoms of adjacent cations (the interplanar spacing) is
3.33 Å (= c/2). This is considerably less than the value of ca. 3.8
Å usually taken as the upper distance limit for π–π interactions
in organic species,15 and could be taken to imply that π–π
interactions are important in determining the photophysical
properties of the salt. However, as noted below, it is the strong
2
2
platinum dz –dz orbital interactions that are largely responsible
for the photoluminescence behaviour of this salt.
The structure of [Pt{4Ј-(o-CH3–Ph)trpy}Cl]SbF6 is unusual
in that terpyridyl ligand complexes of platinum() either do not
exhibit stacked structures or if a stacked structure is observed
the Pt ؒ ؒ ؒ Pt distances are not uniformly equal along the
stacking direction. This contrasts the situation for α-diimine
complexes of platinum() for which there are several well-
documented examples with linear chain structures.16 Other
examples of terpyridyl ligand complexes of platinum() with
stacked structures are the triflate and perchlorate salts of
the [Pt(trpy)Cl]ϩ cation;6,7 both consist of discrete Pt2 units
arranged along an infinite trpy-π stack. Uniform spacing of the
platinum atoms is also not observed for the low temperature
crystal structure of [Pt{4Ј-(Ph)trpy}Cl]BF4ؒCH3CN where the
stacking motif comprises a series of tetramers with an internal
Ϫ
The question arises as to why, by replacing the larger SbF6
Ϫ
with the smaller BF4 anion while retaining the same [Pt{4Ј-
(o-CH3–Ph)trpy}Cl]ϩ cation, the crystal structure changes from
a linear chain with a uniform Pt ؒ ؒ ؒ Pt distance along the stack
to one where the Pt ؒ ؒ ؒ Pt distances alternate along an infinite
π-stack? We first note that the structural change is brought
about by the parallel planes of cations slipping past each other
in such a way that for successive pairs of cations the extent of
slippage is not the same. Also significant is that slippage of this
2
2
kind removes the possibility of extended platinum dz –dz
orbital interactions along the stack; yet these interactions are
regarded as the most important in so far as stabilising a linear
chain structure is concerned.16 Other factors that are expected
J. Chem. Soc., Dalton Trans., 2002, 1369–1376
1371

View Article Online
Fig. 4 ORTEP view of the cation (2+) in [Pt{4Ј-(o-CF3–Ph)trpy}Cl]-
SbF6. Thermal ellipsoids are drawn at the 50% probability level except
for the hydrogen atoms that are shown as spheres of arbitrary radius.
The overall packing motif exhibited by the cations and
anions in [Pt{4Ј-(o-CF3–Ph)trpy}Cl]SbF6 is very similar to that
for [Pt{4Ј-(o-CH3–Ph)trpy}Cl]BF4, the only differences being in
the precise dimensions for the cation stack. Thus the cations
stack parallel and head-to-tail with the anions occupying
channels between the cation stacks (Fig. S3†); the mean planes
through the platinum, chlorine and terpyridyl moiety atoms are
perpendicular to the [a]-axis and are within experimental error
equally spaced; and the Pt ؒ ؒ ؒ Pt distances alternate along the
stack. The Pt ؒ ؒ ؒ Pt distances are 3.629(1) and 3.685(1) Å and
the interplanar spacing averages at 3.46 Å. Noteworthy is that
the Pt ؒ ؒ ؒ Pt distances are more nearly equal than those deter-
mined for the tetrafluoroborate salt. Fig. 5 illustrates the two
different cation–cation interactions along the stack taking a
view perpendicular to the mean planes through the platinum,
Fig. 3 ORTEP diagrams showing the two types of cation–cation
interactions along the [Pt{4Ј-(o-CH3–Ph)trpy}Cl]+ stacks in [Pt{4Ј-(o-
CH3–Ph)trpy}Cl]BF4 viewed perpendicular to the mean planes through
the platinum, chlorine and non-hydrogen atoms of the terpyridyl
moiety: A, Pt ؒ ؒ ؒ Pt distance = 3.573 ; B, Pt ؒ ؒ ؒ Pt distance = 3.827.
The perpendicular distances between the mean planes are nearly
identical for interaction A (3.40 ) and interaction B (3.44 ). Atoms are
shown as 40% probability ellipsoids. Hydrogen atoms are omitted.
to play a role in determining the dimensions of the stack
are steric repulsions and intermolecular forces, both attractive
and repulsive.16 The former should not be responsible for
initiating the structural change given that the cation remains the
same, in particular with regard to the methyl substituent in
the ortho-position of the 4Ј-phenyl ring. As far as inter-
molecular forces are concerned it is possible that a strengthen-
ing of attractive intermolecular forces of the π–π type helps
drive the structural change but, without detailed lattice energy
calculations, this must be regarded as a tentative conclusion.
What is clear however, is that the structural change brings
about an improvement in the packing efficiency. This is evident
from a comparison of the densities of the two solids: for the
Ϫ
Ϫ
SbF6 salt it is 2.21 g cmϪ3 while for the BF4 salt it is 1.97 g
cmϪ3. Were BF4 anions to replace the SbF6 anions in the
Ϫ
Ϫ
Ϫ
crystal structure of the SbF6 salt the density would be only
1.79 g cmϪ3
.
Crystal structure of [Pt{4Ј-(o-CF3–Ph)trpy}Cl]SbF6 ([2]SbF6)
Fig. 4 gives a perspective view of the cation as well as of the
atomic numbering scheme. A list of selected interatomic dis-
tances and angles is given in Table 1. The coordination geom-
etry of the platinum atom and the bond lengths and angles
internal to the substituted terpyridyl ligand are typical of those
discussed above for [Pt{4Ј-(o-CH3–Ph)trpy}Cl]SbF6 and for
terpyridyl ligand complexes of platinum() in general.6,7,9,10,12
Of particular interest is the dihedral angle between the mean
planes through the carbon atoms of the 4Ј-substituent on one
hand and those of the terpyridyl moiety on the other. This is
62.1Њ, a smaller value than those of 65.2 and 69.9Њ obtained for
the SbF6 and BF4 salts of the [Pt{4Ј-(o-CH3–Ph)trpy}Cl]ϩ
cation respectively, an unexpected result given that the van der
Waals radius of a trifluoromethyl group is larger than that of a
methyl group and that steric repulsions involving the 3Ј(5Ј)-
hydrogen of the central pyridine ring will therefore be greater
with the trifluoromethyl group than with the methyl group.
Presumably, the unexpectedly smaller angle is a consequence of
packing effects as discussed below.
Ϫ
Ϫ
Fig. 5 ORTEP diagrams showing the two types of cation–cation
interactions along the [Pt{4Ј-(o-CF3–Ph)trpy}Cl]+ stacks in [Pt{4Ј-
(o-CF3–Ph)trpy}Cl]SbF6 viewed perpendicular to the mean plane
through the platinum, chlorine and non-hydrogen atoms of the
terpyridyl moiety: A, Pt ؒ ؒ ؒ Pt distance = 3.629; B, Pt ؒ ؒ ؒ Pt distance =
3.685 . The perpendicular distances between the mean planes are nearly
identical for interaction A (3.45 ) and interaction B (3.47 ). Atoms are
shown as 40% probability ellipsoids. Hydrogen atoms are omitted.
1372
J. Chem. Soc., Dalton Trans., 2002, 1369–1376

View Article Online
chlorine and non-hydrogen atoms of the terpyridyl moiety. The
The emission spectra measured over a range of temperatures
lateral offset of the platinum atoms and the long Pt ؒ ؒ ؒ Pt dis-
tances preclude any significant overlap of platinum dz -orbitals
on a microcrystalline sample of the red [Pt{4Ј-(o-CH3–Ph)-
trpy}Cl]SbF6 salt are shown in Fig. 6. Also shown in Fig. 6 is
2
perpendicular to the stack. The interplanar spacing falls within
the upper distance limit of 3.8 Å for π-interactions between
organic species.15 Following the same line of reasoning given
above for [Pt{4Ј-(o-CH3–Ph)trpy}Cl]BF4 we tentatively con-
clude that π–π interactions play a role in determining the crystal
structure and photophysical properties of the salt (vide infra).
The question arises as to why, by replacing a methyl with a
trifluoromethyl group as the ortho-substituent on the 4Ј-ring
Ϫ
while retaining the same SbF6 anion, the crystal structure
changes from linear chain with a uniform Pt ؒ ؒ ؒ Pt distance
along the stack to one where the Pt ؒ ؒ ؒ Pt distances alternate
along an infinite π-stack? Again the structural change is
brought about by the parallel planes of cations slipping past
each other in such a way that for successive pairs of cations the
extent of slippage is not the same; at the same time, extended
Fig.
6
Dilute (0.01 mM) dimethylformamide/methanol/ethanol
2
2
platinum dz –dz orbital interactions along the stack are lost.
Steric factors are probably important here as the methyl group
with a van der Waals radius of 2.0 Å is somewhat smaller than a
trifluoromethyl group with a van der Waals radius of 2.3 Å.17
Evidence for this comes from consideration of the non-bonded
contact distance between the carbon atom of the trifluoro-
methyl group [C(22)] and the chlorine atom of the cation in the
next layer (Fig. 5). This distance is 4.27 Å, a value that is close
to the sum of the van der Waals radii for a trifluoromethyl
group and a chlorine atom of 4.2 Å.17,18 It would appear that
the C(22) ؒ ؒ ؒ ClЈ contact prevents the cations from sliding
(1 : 5 : 5) glass emission spectrum recorded at 77 K ( ؒ ؒ ؒ ) and solid
state emission spectra recorded at 40 K intervals over the range 80 (a) to
280 K (b) (—) of [Pt{4Ј-(o-CH3–Ph)trpy}Cl]SbF6. λex = 330 nm.
the emission spectrum of the same salt recorded in a dilute
(0.01 mM) dimethylformamide/methanol/ethanol (1 : 5 : 5)
glass at 77 K. The dilute glass spectrum reflects emission by the
[Pt{4Ј-(o-CH3–Ph)trpy}Cl]ϩ luminophores when they are in a
monomeric environment,19c and thus comparison of the glass
spectrum with that recorded in the solid state can provide
information on the effect of solid state interactions on the elec-
tronic structure of the luminophore.2 The vibrationally struc-
tured glass spectrum is very similar to that recorded for
[Pt(trpy)Cl]CF3SO3 in a dilute butyronitrile glass at 77 K.3 On
this basis we make the same assignment which is that the glass
emission for [1]SbF6 derives from an essentially intraligand
3(π–π*) excited state.6 Turning to the emission recorded for the
solid (Fig. 6) we first note that there is a substantial red-shift
compared to that observed for the glass, an observation that
suggests that solid state interactions are important in determin-
ing the emission properties of the salt. The 280 K spectrum
consists of a broad asymmetric peak centred at 616 nm that is
devoid of any vibrational structure. On cooling the sample to
80 K the peak increases in intensity, narrows and shifts dramat-
ically to the red, specifically from a peak maximum of 616 nm
at 280 K to 673 nm at 80 K. These data and, in particular the
red-shift observed on cooling, are strongly suggestive of
2
into a position that allows for directed platinum dz -orbital
interactions perpendicular to the stack. Interestingly, the
C(22) ؒ ؒ ؒ ClЈ distance is linked to the angle of rotation of
the 4Ј-group about the interannular bond; the smaller the angle
the larger the distance. As noted above, the dihedral angle of
62.1Њ observed for [2]SbF6 is smaller than those of 65.2 and
69.9Њ observed for [1]SbF6 and [1]BF4 respectively, despite the
fact that the trifluoromethyl group is sterically more demanding
than the methyl group.
Photophysical properties
The absorption spectra measured in acetonitrile are nearly
identical for all four of the salts and, moreover, are very similar
to those recorded in acetonitrile for salts of the [Pt(trpy)Cl]ϩ 6,7
and [Pt(4Ј-Ph–trpy)Cl]ϩ 6,10 luminophores. Accordingly, the
same assignments are made. The relatively sharp band at ca.
280 nm and the vibrationally structured band with maxima in
the range 300–350 nm are attributed to 1(π–π) transitions of the
substituted terpyridyl ligands, while the two less intense and
poorly resolved peaks at wavelengths greater than 350 nm are
assigned to Pt(5d) trpy(π*) (1MLCT) transitions. Data are
given in the Experimental section. None of the salts gives rise to
detectable emission when dissolved in degassed acetonitrile and
irradiated with light of 330 nm wavelength at room temper-
ature. Salts of the [Pt(trpy)Cl]ϩ 6,7 and [Pt(4Ј-Ph–trpy)Cl]ϩ 6,10
luminophores also do not display emissions in degassed
acetonitrile at room temperature; on the other hand salts of
the [Pt{4Ј-(p-R–Ph)trpy}Cl]ϩ (R = CH3, OCH3, Br, or CN)
luminophores do give rise to emission in degassed acetonitrile.6
When fluid emission by monomers of platinum() is not
observed one explanation is that low-lying d–d excited states
provide facile and non-radiative deactivation pathways via
molecular distortions,19a while another is that axial interactions
involving the donor solvent quench emissions by the excited
state monomer.19b Unfortunately, because of solubility prob-
lems, it was not possible to measure emission spectra in a
non-coordinating solvent such as dichloromethane and we
are therefore uncertain as to the reason for the lack of fluid
emission by salts of the [Pt{4Ј-(o-R–Ph)trpy}Cl]ϩ (R = CH3 or
CF3) chromophores.
3
emission from a [dσ*,π*] (3MMLCT) state. Also consistent
3
with emission from a [dσ*,π*] state is the narrowing of the
band on cooling.1,9,10 Furthermore, the solid state structure of
the compound shows a Pt ؒ ؒ ؒ Pt distance along a uniform
stack of luminophores that is short enough [3.368(1) Å] to
2
2
support strong platinum dz –dz orbital interactions along the
stack; a requirement for 3MMLCT emission. Similar solid state
luminescence behaviour by the orange triflate salt of the
[Pt(trpy)Cl]ϩ luminophore,6 as well as by the red form of [Pt-
(bipy)Cl2]8 has also been attributed to emissions from
3MMLCT states. Noteworthy, is the systematic increase in
intensity of the emission on lowering the temperature. Linked
to this trend is a ten-fold increase in the emission lifetime from
155 ns at 295 K to 1680 ns at 77 K. The solid state emission data
recorded for [Pt{4Ј-(o-CH3–Ph)trpy}Cl]SbF6 as well as for the
other salts studied in this work are summarised in Table 2.
The variable temperature emission spectra measured on a
microcrystalline sample of the yellow [Pt{4Ј-(o-CH3–Ph)trpy}-
Cl]BF4 salt are shown in Fig. 7. The 280 K spectrum consists of
a relatively broad (full-width-at-half-maximum = 2508 cmϪ1
)
asymmetric band that is devoid of vibrational structure. On
lowering the temperature of the sample there is a systematic
increase in the intensity, slight narrowing of the band but no
red-shift in the emission maximum. In assigning the origin of
the emission we note that there is a substantial red-shift in
J. Chem. Soc., Dalton Trans., 2002, 1369–1376
1373

View Article Online
Table 2 Emission dataa
λ(em)max/nmb
band. Applying the same arguments used to interpret the
emission spectra measured for [Pt{4Ј-(o-CH3–Ph)trpy}Cl]BF4
leads to the conclusion that [Pt{4Ј-(o-CF3–Ph)trpy}Cl]SbF6
and [Pt{4Ј-(o-CF3–Ph)trpy}Cl]BF4 also exhibit solid emission
that is largely excimeric in origin. Support for this conclusion
comes from the results of the crystal structure determination
of [Pt{4Ј-(o-CF3–Ph)trpy}Cl]SbF6 that suggest the existence
of intermolecular trpy π–π interactions in the solid state
(vide supra).
fwhm/cmϪ1 b, c
τ/nsd
[1](SbF6)
[1](BF4)
[2](SbF6)
[2](BF4)
616[673]
1635[1119]
2508[2109]
2735[2596]
3100[2450]
155[1680]
348[2880]
325[2620]
not measured
564[566]
589[589]
571[571]
a Measured on microcrystalline samples. b At 280 K. Data recorded at
80 K are given in square brackets. c fwhm = full-width-at-half-
maximum. d At 295 K. Data recorded at 77 K are given in square brack-
ets.
Concluding remarks
The crystal structures of [1]SbF6, [1]BF4 and [2]SbF6 all consist
of separate and parallel columns of cations and anions. Also
common to all three structures is that the cations stack parallel
to each other in the head-to-tail orientation that allows the
out-of-plane 4Ј-group to point towards a “hole” in the adjacent
cation layer; indeed, the head-to-tail arrangement appears to be
a necessary consequence of the out-of-plane twisting of the
4Ј-group. These structural features are in accord with the prin-
ciple that free volume in crystals is energetically unfavourable.20
Two variables have been used to alter the geometry of a cation
stack. The first is the choice of anion. Changing the anion from
Ϫ
Ϫ
SbF6 to BF4 while retaining the same cation disrupts the
uniform stacking of the platinum atoms, as evidenced by differ-
ent lateral offsets of the platinum atoms in successive pairs of
cations with respect to a line perpendicular to the stacking
planes. The second is the choice of R-group in the ortho-
position of the 4Ј-phenyl ring. Changing a methyl group for a
Fig. 7 Solid state emission spectra of [Pt{4Ј-(o-CH3–Ph)trpy}Cl]BF4
recorded at 40 K intervals over the range 80 (ka) to 280 K (b). λex
=
330 nm.
Ϫ
the band position compared to that recorded for the [Pt{4Ј-
(o-CH3–Ph)trpy}Cl]ϩ cation in a dilute glass (Fig. 6); also that
the vibrational structure present in the glass spectrum and
characteristic of monomeric intraligand π–π* emission is no
longer present. These observations lead to the conclusion that
solid state interactions are important in determining the emis-
sion properties of the solid and that monomer emission in
the solid is unlikely. In view of the long Pt ؒ ؒ ؒ Pt distances in
trifluoromethyl group while retaining the same SbF6 anion,
disrupts the uniform stacking of the platinum atoms in the
same way but for different reasons. We have also learnt from
this work that the choice of the fourth ligand bonded to
platinum, in particular its size and/or ability to bond to the
R-group, will play a role in determining the precise dimensions
of the stack. Thus, by suitable choice of R-group, anion and the
fourth ligand bonded to platinum it is possible to fine-tune the
dimensions of the cation stack and hence the photolumin-
escence properties of the salt. Of particular interest are the
structural requirements for 3MMLCT emission by stacked
terpyridyl ligand complexes of platinum(). This work as well
as that by others,4–8,16 confirms that the stack must include a
2
2
the crystal (vide supra) platinum dz –dz orbital interactions can
be ruled out. Moreover the emission behaviour, especially with
regard to the lack of any dependence of the wavelength of the
emission maximum on temperature, bears no resemblance to
that expected for 3MMLCT emission. On the other hand there
is evidence from the crystal structure analysis for extended π–π
interactions in the solid (vide supra). On this basis emission
from a π-dimer that forms when the compound is irradiated
must be considered as a possibility.1,2 Indeed, the presence of
closely stacked layers of cations in [1]BF4 appears to mandate
the formation of excited state dimers (excimers) but almost
certainly not with the geometry that is ideal for the excited
state. The inevitable mismatch between the ground and excited
state potential energy surfaces will in turn promote structural
relaxation in the excited state and loss of vibrational structure
in the emission spectrum. This explains the broad featureless
appearance of the emission band that is generally regarded as
2
2
Pt ؒ ؒ ؒ Pt distance that is short enough to support dz –dz
orbital interactions perpendicular to the stack, the upper dis-
tance limit being about 3.5 Å. A second structural requirement
is that the closely-spaced platinum atoms must be positioned
2
such that their dz -orbitals point towards each other. Note that
these two requirements do not place restrictions on the pattern
of Pt ؒ ؒ ؒ Pt distances along the stack. For example, a linear
chain structure where the Pt ؒ ؒ ؒ Pt distances are all the same is
3
not a pre-condition for MMLCT emission; nor is it a pre-
condition for a red-shift in the emission maximum when the
material is cooled as evidenced by the temperature dependence
6
of the 3MMLCT emission exhibited by [Pt(trpy)Cl]CF3SO3
and [Pt(4Ј-Ph-trpy)Cl]BF4ؒCH3CN10 neither of which has a
linear chain structure. We finally note that in terms of the
Peierls theorem one dimensional structures with equal separ-
ations of the repeating units are not intrinsically stable;21 as the
temperature of the material is lowered distortions are antici-
pated that will lead to the formation of molecular units along
the stack. Based on the limited structural data available so far
for terpyridyl ligand complexes of platinum() these may be Pt2
dimers6,7 or Pt4 tetramers.10
typical of excimer emission; for example, excimer emission by
Ϫ
[Pt(phen)2]Cl2ؒ3H2O2 as well as by the orange ClϪ, ClO4
,
CF3SO3 and yellow PF6 salts of the [Pt(trpy)Cl]ϩ cation7 is
also broad and featureless. Also typical of excimer emission is a
red-shift in the emission maximum compared to that observed
for the isolated chromophore.1,2 Taken together the evidence
suggests that the solid state emission exhibited by [1]BF4 is
largely excimeric in origin.19d
Ϫ
Ϫ
Ϫ
Ϫ
Both the SbF6 and BF4 salts of the [Pt{4Ј-(o-CF3–Ph)-
trpy}Cl]ϩ luminophore are yellow and, moreover, both exhibit
luminescence behaviour in the solid state very similar to that
measured for [Pt{4Ј-(o-CH3–Ph)trpy}Cl]BF4; the emission data
are summarised in Table 2. The emission bands are asymmetric,
unstructured, relatively broad and red-shifted with respect to
the dilute glass emission shown in Fig. 6; cooling the com-
pounds to 80 K has no effect on the position of the emission
Experimental
Materials
Reagents for the ligand syntheses were obtained and used as
follows: The 2-acetylpyridine was obtained commercially
1374
J. Chem. Soc., Dalton Trans., 2002, 1369–1376

View Article Online
(Fluka) and used without further purification; the 2-tolu-
aldehyde (Merck) and α,α,α-trifluoro-2-tolualdehyde (Aldrich)
were also obtained commercially both being further purified
by distillation; the N-{1-(2Ј-pyridyl)-1-oxo-2-ethyl}pyridinium
iodide was synthesised by the method of Kröhnke.11 All
manipulations for the complex salt syntheses were performed
under an atmosphere of nitrogen using standard Schlenk tube
techniques. The acetonitrile used as the solvent and for the pur-
poses of crystal growth was purified by the method of Carlsen
et al.22 The dichlorobis(benzonitrile)platinum() (Strem) and
the silver salts AgSbF6 and AgBF4 (Fluka) were used without
further purification.
11.0%). MS(EI) m/z: R = CH3 (323 Mϩ); R = CF3 (377, Mϩ). 1H
NMR (CDCl3): R = CH3 [δ 8.71 (m, 2 H, H6,6Љ); 8.67 (m, 2 H,
H3,3Љ); 8.47 (s, 2 H, H3Ј,5Ј); 7.87 (m, 2 H, H4Ј,4Љ); 7.32 (m, 4 H,
C6H4); 7.30 (m, 2 H, H5,5Љ); 2.37 (s, 3 H, CH3)]. R = CF3 [δ 8.72
(m, 2 H, H6,6Љ); 8.70 (m, 2 H, H3,3Љ); 8.54 (s, 2 H, H3Ј,5Ј); 7.84 (m,
2 H, H4,4Љ); 7.54 (m, 4 H, C6H4); 7.35 (m, 2 H, H5,5Љ)]. UV/vis
(CH3CN): λmax/nm (ε/MϪ1 cmϪ1): R = CH3 [303(sh, 1.6 × 104);
277(3.0 × 104); 250(3.5 × 104)]. R = CF3 [303(sh, 1.3 × 104);
277(2.9 × 104); 239 (3.4 × 104); 208 (3.6 × 104)].
[Pt{4Ј-(o-R–Ph)trpy}Cl]A (R ؍ CH3 or CF3; A ؍ SbF6 or
BF4). A suspension of [Pt(PhCN)2Cl2] (0.10 g, 0.21 mmol) in
acetonitrile (10 mL) was treated with an equimolar amount of
AgA (0.073 g for A = SbF6; 0.041 g for A = BF4) dissolved in
acetonitrile (5 mL). The reaction mixture was heated under
reflux for 16 h, the AgCl precipitate removed by filtration and
one equivalent of 4Ј-(C6H4R-o)-2,2Ј:6Ј,2Љ-terpyridine (0.060 g
for R = CH3; 0.080 g for R = CF3) added to the filtrate. The
reaction mixture was heated under reflux for an additional 24 h
after which the volume was reduced in vacuo, the solution
cooled to room temperature and allowed to stand for a further
24 h. This resulted in the precipitation of [Pt{4Ј-(o-R–Ph)-
trpy}Cl]A. The precipitate was washed with cold acetonitrile
(ca. 5 mL) and diethyl ether (ca. 10 mL) and dried in vacuo to
afford an analytically pure microcrystalline product. Yields and
colours: R = CH3, A = SbF6 (0.14 g, 84%, red); R = CH3, A =
BF4 (0.10 g, 78%, yellow); R = CF3, A = SbF6 (0.14 g, 76%,
yellow); R = CF3, A = BF4 (0.13 g, 89%, yellow). Anal. R = CH3,
A = SbF6 (Calcd. for C22H17ClF6N3PtSb: C 33.4; H 2.3; N 5.3.
Found: C 33.1; H 1.9; N 5.2%). R = CH3, A = BF4 (Calcd. for
C22H17BClF4N3Pt: C 41.8; H 2.9; N 6.7. Found: C 41.6; H 2.9;
N 6.6%). R = CF3, A = SbF6 (Calcd. for C22H14ClF9N3PtSb: C
31.3; H 1.8; N 5.0. Found: C 31.3; H 1.4; N 4.9%). R = CF3,
A = BF4 (Calcd. for C22H14BClF7N3Pt: C 38.0; H 2.2; N 6.0.
Physical measurements and instrumentation
Microanalyses for %C, H and N were performed by the micro-
analytical laboratory at the University of Natal, Pietermaritz-
burg and by Galbraith Laboratories Inc., Knoxville, Tennessee,
USA. Melting points were recorded on a Kofler hot stage
1
apparatus and are uncorrected. H NMR (200 MHz) spectra
were recorded on a Varian Gemini 200 spectrometer at 25 ЊC
with chemical shifts referenced to SiMe4. Infrared spectra were
recorded as KBr discs on a Shimadzu FTIR-4300 spectrometer.
Mass spectra were obtained on a Hewlett Packard GCMS using
electron impact (EI) ionisation. UV/vis absorption spectra were
recorded at 22 ЊC using a Shimadzu UV-2101PC scanning
spectrophotometer. Emission spectra were recorded on a SLM-
Amico SPF 500C fluorometer at 22 ЊC unless otherwise stated.
Deoxygenated spectroscopic grade solvents were used for both
the absorption and emission measurements. Solid state emis-
sion spectra were recorded on microcrystalline samples. For the
variable temperature emission measurements the cryostat was
an Oxford Instruments DN1704 liquid-nitrogen-cooled system
complete with an Oxford Instruments temperature controller.
The excitation wavelength was 330 nm, with the scattered light
removed by a 400 nm long-wave-pass filter. The 337 nm line
from a nitrogen laser served as the excitation source for the
lifetime measurements, with a 337 nm band pass filter used to
remove stray light from the beam. Lifetime data were analysed
as described previously.23
1
Found: C 38.0; H 2.0; N 5.9%). H NMR (CDCl3): R = CH3,
A = SbF6 [δ 2.43 (s, 3 H, CH3); 7.45 (m, 4 H, aromatic CH ); 7.74
(m, 2 H, aromatic CH ); 8.25 (m, 6 H, aromatic CH ); 8.84 (m,
2 H, aromatic CH )]. R = CF3, A = SbF6 [δ 7.6–8.2 (m, 6 H,
aromatic CH ); 8.45 (m, 2 H, aromatic CH ); 8.64 (m, 2 H,
aromatic CH ); 8.79 (m, 4 H, aromatic CH )]. IR (KBr, cmϪ1): R
= CH3, A = SbF6 [ν{4Ј-(o-CH3–Ph)trpy} 1618s, 1557m, 1477m,
1419m, 1035m, 888m; ν(SbF6Ϫ) 659vs]. R = CH3, A = BF4
[ν(BF4Ϫ) 1064vs]. R = CF3, A = SbF6 [ν{4Ј-(o-CF3–Ph)trpy}
1611s, 1478m, 1422m, 1319m, 1179m; ν(SbF6Ϫ) 661vs]. R =
Syntheses
2-R-1-{3-(2-pyridyl)-3-oxopropenyl}benzene (R ؍ CH3 or
CF3). A solution of 2-R-benzaldehyde (20 mmol) in absolute
ethanol (100 mL) was cooled to 0 ЊC and 2-acetylpyridine
(2.42 g, 20 mmol) added. Aqueous sodium hydroxide (20 mL,
1.0 M) was then added dropwise and the reaction mixture
stirred at 0 ЊC for 3 h. This resulted in the separation of a
light yellow precipitate that was collected by filtration, washed
with ethanol and dried in vacuo. Yields: R = CH3 (4.30 g,
96%); R = CF3 (5.39 g, 97%). Mp’s: R = CH3 (68 ЊC); R = CF3
(63 ЊC). Anal. R = CH3 (Calcd. for C15H13NO: C 80.7; H 5.9;
N 6.3. Found: C 80.8; H 5.8; N 6.2%). R = CF3 (Calcd. for
C15H10F3NO: C 65.0; H 3.6; N 5.1. Found: C 64.8; H 3.6;
N 5.1%). MS(EI) m/z: R = CH3 (223, Mϩ); R = CF3 (277, Mϩ).
IR (KBr, cmϪ1): R = CH3 [ν(CO) 1670s]; R = CF3 [ν(CO) 1680s].
CF3, A = BF4 [ν(BF4Ϫ) 1061vs]. UV/vis (CH3CN): λmax
/
nm (ε/MϪ1 cmϪ1): R = CH3, A = SbF6 [283(3.1 × 104); 307(1.7 ×
104); 316(1.7 × 104); 332(1.9 × 104); 347(7.6 × 103); 380(4.2 ×
103); 399(4.5 × 103)]. R = CF3, A = SbF6 [285(3.8 × 104);
306(1.3 × 104); 319(1.2 × 104); 332(1.6 × 104); 349(7.6 × 103);
381(3.4 × 103); 396(3.4 × 103)].
Crystal structure determinations
Needle-shaped crystals of [1]SbF6 (red), [1]BF4 (yellow) and
[2]SbF6 (yellow) were grown by slow evaporation at room
temperature of a saturated solution of the compound in
acetonitrile. Crystal data and details of the crystallographic
studies are reported in Table 3. Intensity data were obtained on
an Enraf-Nonius CAD4 diffractometer, using graphite mono-
chromated MoKα radiation and the ω–2θ scan technique. Unit
cell parameters were obtained by least squares fitting of 25
reflections monitored in the range 3Њ < θ < 12Њ while the diffrac-
tion data were collected in the range 2Њ < θ < 23Њ. Corrections
for Lorentz, polarisation, and absorption (χ scans of 9 reflec-
tions) effects were applied. The intensities of three standard
reflections showed no variations greater than those predicted by
counting statistics. The structures were solved by Patterson and
Fourier methods and refined by full-matrix least-squares using
SHELXS-97.24 For [1]SbF6 anisotropic displacement param-
eters were only assigned to the Pt, Sb, Cl and F atoms, the
4Ј-(C6H4R-o)-2,2Ј:6Ј,2Љ-terpyridine (R ؍ CH3 or CF3). N-{1-
(2Ј-pyridyl)-1-oxo-2-ethyl}pyridinium iodide (0.68 g, 2.2 mmol)
and ammonium acetate (10 g, excess) were added to a suspen-
sion of 2-R-{3-(2-pyridyl)-3-oxopropenyl}benzene (2.0 mmol)
in absolute ethanol (8 mL) and the mixture heated at reflux for
40 min. An off-white solid precipitated on cooling. This was
collected by filtration, washed with 50% aqueous ethanol and
dried in vacuo. Recrystallisation from ethanol afforded colour-
less crystals of the desired ligands. Yields: R = CH3 (0.31 g,
47%); R = CF3 (0.41 g, 54%). Mp’s: R = CH3 (156 ЊC); R = CF3
(148 ЊC). Anal. R = CH3 (Calcd. for C22H17N3: C 81.7; H 5.3;
N 13.0. Found: C 81.6; H 5.0; N 13.0%). R = CF3 (Calcd. for
C22H14F3N3: C 70.0; H 3.7; N 11.1. Found: C 69.9; H 3.9; N
J. Chem. Soc., Dalton Trans., 2002, 1369–1376
1375

View Article Online
Table 3 Crystal and refinement data
[1]SbF6
[1]BF4
[2]SbF6
Formula
Fw
Crystal system
Space group
a/Å
b/Å
c/Å
C22H17ClF6N3PtSb
789.67
orthorhombic
Pna21
19.64(2)
18.166(5)
6.653 (2)
90
2373(3)
4
2.210
293(2)
7.20
1480
C22H17BClF4N3Pt
640.73
monoclinic
P21/n
C22H14ClF9N3PtSb
843.64
monoclinic
P21/n
6.87(2)
6.93(2)
19.489(5)
16.500(2)
101.923(5)
2161(8)
4
1.969
293(2)
20.114(5)
17.979(2)
100.750(5)
2462(8)
β/Њ
V/Å3
Z
4
Dc/g cmϪ3
T /K
2.276
293(2)
6.97
1576
µ/mmϪ1
F(000)
6.66
1224
Total no. of reflns. measured
2450
3589
4055
Independent reflns.
Reflections observed [I > 2σ(I )]
Refinement type
2146 [Rint = 0.0158]
2996 [Rint = 0.0219]
3423 [Rint = 0.0375]
1727
2208
2504
F 2
F 2
F 2
R1 a, wR2 b [I > 2σ(I )]
R1 a, wR2 b [all data]
No. of refined parameters
(shift/esd.)max
0.0358, 0.1013
0.0533, 0.1142
182
0.001
0.993, Ϫ0.608
0.0431, 0.1233
0.0713, 0.1395
281
0.001
0.882, Ϫ1.133
0.0449, 0.1160
0.0745, 0.1317
335
0.000
2.035, Ϫ1.075
max, min ∆ρ/e ÅϪ3
2
2
a R1 = Σ||Fo| Ϫ |Fc||/Σ|Fo|. b wR2 = [Σ[w(Fo2 Ϫ Fc2)2]/Σ[w(Fo )2]]1/2 where w = 1/[σ2(Fo ) ϩ (0.1P)2] and P = (Fo2 ϩ 2Fc2)/3.
7 J. A. Bailey, M. G. Hill, R. E. Marsh, V. M. Miskowski, W. P.
Schaefer and H. B. Gray, Inorg. Chem., 1995, 34, 4591.
8 W. B. Connick, L. M. Henling, R. E. Marsh and H. B. Gray,
Inorg. Chem., 1996, 35, 6261.
9 R. Büchner, J. S. Field, R. J. Haines, C. T. Cunningham and
D. R. McMillin, Inorg. Chem., 1997, 36, 3952.
10 R. Büchner, C. T. Cunningham, J. S. Field, R. J. Haines,
D. R. McMillin and G. C. Summerton, J. Chem. Soc., Dalton Trans.,
1999, 711.
remaining non-hydrogen atoms being assigned individual
isotropic displacement parameters and the hydrogen atoms (in
calculated positions) a single overall isotropic displacement
parameter. Attempts to carry out a full anisotropic refinement
of the non-hydrogen atoms in [1]SbF6 led to problems with
atoms N(2) and C(5) being assigned as “non-positive definite”.
Inverting the absolute structure of [1]SbF6 and re-refining gave
a fractionally higher Flack parameter. For [1]BF4 the seven
carbon atoms [C(16)–C(22)] of the o-tolyl substituent in the
4Ј-postion of the terpyridyl ring were constrained to lie in
the same plane using the SHELXS-97 command “FLAT”.24
This was necessitated because of the high thermal vibration of
some of these atoms and the ensuing difficulty experienced in
refining them to sensible positions in the absence of the con-
straint. All the non-hydrogen atoms in [1]BF4 and [2]SbF6 were
refined anisotropically while the hydrogen atoms (in calculated
positions) were assigned a single overall isotropic displacement
parameter. The crystal structure diagrams were produced by the
ORTEP program.25
11 F. Kröhnke, Synthesis, 1976, 1.
12 K. W. Jennette, J. T. Gill, J. A. Sadownick and S. J. Lippard, J. Am.
Chem. Soc., 1976, 98, 6159; J. A. Bailey, V. M. Miskowski and
H. B. Gray, Acta Crystallogr., Sect. C., 1992, 48, 1420.
13 E. C. Constable, J. Lewis, M. C. Liptrot and P. R. Raithby,
Inorg. Chim. Acta, 1990, 178, 47.
14 D. S. Martin, in Extended Interactions between Metal Ions, ed. L. V.
Interannte, ACS Symp. Ser. 5, American Chemical Society,
Washington, DC, 1974, p. 254.
15 C. A. Hunter and J. K. M. Sanders, J. Am. Chem. Soc., 1990, 112, 5525.
16 W. B. Connick, R. E. Marsh, W. P. Schaefer and H. B. Gray,
Inorg. Chem., 1997, 36, 913.
17 Ref. 18 gives the van der Waals radius of a methyl group as 2.0 Å.
The van der Waals radius of 2.3 Å for a trifluoromethyl group was
obtained by increasing the van der Waals radius of a methyl group
by a distance equal to the difference in the covalent radius for a
hydrogen atom and that for a fluorine atom.
CCDC reference numbers 168557–168559.
See http://www.rsc.org/suppdata/dt/b1/b107113k/ for crystal-
lographic data in CIF or other electronic format.
18 J. E. Huheey, E. A. Keitner and R. L. Keitner, Inorganic Chemistry,
HarperCollins, New York, 4th edn., 1993.
Acknowledgements
19 (a) T. K. Aldridge, E. M. Stacey and D. R. McMillin, Inorg. Chem.,
1994, 33, 722; (b) D. K. Crites Tears and D. R. McMillin, Coord.
Chem. Rev., 2001, 211, 195; (c) the possibility that the structured
emission from the dilute glass is due to the presence of free ligand
can be discounted because the highest vibronic component in the
emission of the free 4Ј-phenyl-2,2Ј:6Ј,2Љ-terpyridyl ligand is at 435
nm i.e., at a much shorter wavelength than the signal in Fig. 6. See:
J. F. Michalec, S. A. Bejune, D. G. Cuttell, G. C. Summerton,
J. A. Gertenbach, J. S. Field, R. J. Haines and D. R. McMillin, Inorg.
Chem., 2001, 40, 2193; (d ) another possibility is that the broad
emission shown in Fig. 7 arises from a d–d state. Certainly,
Miskowski and co-workers have assigned similar broad emisions as
d–d bands in a number of related compounds; see ref. 5.
20 A. I. Kitaigorodskii, Organic Chemical Crystallography,
Consultants Bureau, New York, 1961.
21 R. E. Peierls, Quantum Theory of Solids, Oxford, London, 1953.
22 L. Carlsen, H. Egsgaard and J. R. Anderson, Anal. Chem., 1979, 51,
1593.
23 F. Liu, K. L. Cunningham, W. Uphues, G. W. Fink, J. Schmolt and
D. R. McMillin, Inorg. Chem., 1995, 34, 2015.
24 G. M. Sheldrick, SHELXS-97, A program for crystal structure
determination and refinement, University of Göttingen, 1997.
25 L. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
We acknowledge financial support from the University of
Natal and the South African National Research Foundation.
Thanks also go to the United States National Research
Foundation for funding through grant CHE 97–26435. We
extend our appreciation to Dr Orde Munro for helpful
discussions.
References and notes
1 V. H. Houlding and V. M. Miskowski, Coord. Chem. Rev., 1991, 111,
145.
2 V. M. Miskowski and V. H. Houlding, Inorg. Chem., 1989, 28, 1529.
3 M. Weiser-Wallfahrer and G. Gliemann, Z. Naturforsch., Teil B,
1990, 45, 652.
4 V. M. Miskowski and V. H. Houlding, Inorg. Chem., 1991, 30,
4446.
5 V. M. Miskowski, V. H. Houlding, C.-M. Che and Y. Wang,
Inorg. Chem., 1993, 32, 2518.
6 H.-K. Yip, L.-K. Cheng, K.-K. Cheung and C.-M. Che, J. Chem.
Soc., Dalton Trans., 1993, 2933.
1376
J. Chem. Soc., Dalton Trans., 2002, 1369–1376