Enhanced electron injection and efficiency in blended-layer organic light emitting diodes with aluminium cathodes: New 2,5-diaryl-1,3,4-oxadiazole- fluorene hybrids incorporating pyridine units

Article abstract of DOI:10.1039/b510003h

This work focuses on the first 2,5-diaryl-1,3,4-oxadiazole-fluorene hybrids which incorporate pyridine units within the π-electron system, viz. 2,7-bis{5-[2-(4-dodecyloxyphenyl)-1,3,4-oxadiazol-5-yl]-2-pyridyl}-9, 9-dihexylfluorene (6) and 2,7-bis{5-[2-(4-dodecyloxyphenyl)-1,3,4-oxadiazol-5- yl]-2-pyridyl}spirobifluorene (7). The thiophene analogue 2,7-bis{5-[5-(4-tert- butylphenyl)-1,3,4-oxadiazol-2-yl]-thien-2-yl}-9,9-dihexylfluorene 11 was also synthesised and its X-ray crystal structure was obtained. There is a progressive red shift in the UV-Vis absorption and photoluminescence spectra on replacing benzene (8) with pyridine (6) and thiophene (11) consistent with increased planarity of the π-system and reduced HOMO-LUMO gap along the series. Calculations at the DFT (density functional theory) level establish that inclusion of the pyridyl rings in 6 and 7 considerably enhances the electron affinity of the system, compared to phenyl analogues. Single-layer organic light-emitting diodes (OLEDs) have been fabricated by spin-coating blends of poly[2-(2-ethylhexyloxy)-5-methoxy-1,4-phenylenevinylene] (MEH-PPV) as the emissive material with added electron transport compounds 6 or 7 to enhance electron injection. The external quantum efficiencies of the devices were greatly enhanced compared to pure MEH-PPV reference devices. ITO/PEDOT : PSS/MEH-PPV : 7 (30 : 70% by weight)/Al devices exhibited an external quantum efficiency (EQE) of 0.5% and a luminous efficiency of 0.93 cd A^-1 at 9.5 V and a luminance of 100 cd m^-2. The modest increase in efficiency for the same device when Al was replaced by a Ca/Al cathode (EQE 0.6% and 1.2 cd A^-1 at 10.5 V) suggests that the two methods of enhancing electron injection into the MEH-PPV emitter are mutually exclusive. Utilising blended layers is an attractive alternative to using Ca electrodes, which are highly reactive and are unstable in the atmosphere. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b510003h

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Enhanced electron injection and efficiency in blended-layer organic light
emitting diodes with aluminium cathodes: new 2,5-diaryl-1,3,4-oxadiazole–
fluorene hybrids incorporating pyridine units{
Stephen Oyston,a Changsheng Wang,a Igor F. Perepichka,{a Andrei S. Batsanov,a Martin R. Bryce,*a
Jin H. Ahnb and Michael C. Petty*b
Received 13th July 2005, Accepted 23rd September 2005
First published as an Advance Article on the web 26th October 2005
DOI: 10.1039/b510003h
This work focuses on the first 2,5-diaryl-1,3,4-oxadiazole–fluorene hybrids which
incorporate pyridine units within the p-electron system, viz. 2,7-bis{5-[2-(4-dodecyloxyphenyl)-
1,3,4-oxadiazol-5-yl]-2-pyridyl}-9,9-dihexylfluorene (6) and 2,7-bis{5-[2-(4-dodecyl-
oxyphenyl)-1,3,4-oxadiazol-5-yl]-2-pyridyl}spirobifluorene (7). The thiophene analogue
2,7-bis{5-[5-(4-tert-butylphenyl)-1,3,4-oxadiazol-2-yl]-thien-2-yl}-9,9-dihexylfluorene 11 was also
synthesised and its X-ray crystal structure was obtained. There is a progressive red shift in the
UV–Vis absorption and photoluminescence spectra on replacing benzene (8) with pyridine (6) and
thiophene (11) consistent with increased planarity of the p-system and reduced HOMO–LUMO
gap along the series. Calculations at the DFT (density functional theory) level establish that
inclusion of the pyridyl rings in 6 and 7 considerably enhances the electron affinity of the system,
compared to phenyl analogues. Single-layer organic light-emitting diodes (OLEDs) have been
fabricated by spin-coating blends of poly[2-(2-ethylhexyloxy)-5-methoxy-1,4-phenylenevinylene]
(MEH–PPV) as the emissive material with added electron transport compounds 6 or 7 to enhance
electron injection. The external quantum efficiencies of the devices were greatly enhanced
compared to pure MEH–PPV reference devices. ITO/PEDOT : PSS/MEH–PPV : 7 (30 : 70% by
weight)/Al devices exhibited an external quantum efficiency (EQE) of 0.5% and a luminous
efficiency of 0.93 cd A21 at 9.5 V and a luminance of 100 cd m22. The modest increase in
efficiency for the same device when Al was replaced by a Ca/Al cathode (EQE 0.6% and 1.2 cd A21
at 10.5 V) suggests that the two methods of enhancing electron injection into the MEH–PPV
emitter are mutually exclusive. Utilising blended layers is an attractive alternative to using Ca
electrodes, which are highly reactive and are unstable in the atmosphere.
electron affinities. In a device structure this creates an
Introduction
imbalance of electron injection (from the low work-function
Intense interest in the synthesis and properties of new materials
cathode) and hole injection (from the high work-function
for incorporation into organic light emitting diodes (OLEDs)
anode) which leads to charge recombination occurring near the
for display applications1 has been sustained since the discovery
polymer/cathode interface and a lowering of EL efficiency due
of electroluminescence (EL) in low molecular weight organic
to quenching of excitons by the metal electrode. A strategy
molecules2 and in conjugated polymers.3 The most widely
which has proved successful in mitigating this problem is
studied emissive polymers are derivatives of poly(phenylene-
to incorporate electron-deficient (n-type) polymeric or low
vinylene) and poly(fluorene). These are predominantly hole-
molecular weight materials into the devices, often as an
transporting (i.e. p-dopable) materials which possess low
additional electron-transporting hole-blocking (ETHB) layer,
or in some cases as a blend with the emissive polymer.4
aDepartment of Chemistry and Centre for Molecular and Nanoscale
Electronics, University of Durham, Durham, UK DH1 3LE.
E-mail: m.r.bryce@durham.ac.uk
2,5-Diaryl-1,3,4-oxadiazole (OXD) derivatives are electron-
deficient systems which possess good thermal and chemical
stabilities and high photoluminescence quantum yields,5 and
this combination of properties has led to their use as ETHB
materials in OLEDs.6 We have recently reported that OXD
derivatives enhance electron injection in single-layer devices as
blends with poly[2-(2-ethylhexyloxy)-5-methoxy-1,4-phenyl-
enevinylene] (MEH–PPV) as the emissive material.7 The
present study is focused on new OXD analogues 6 and 7 in
which pyridine rings are incorporated into the conjugated
p-system. Pyridine is an electron-deficient heterocycle, so its
presence is expected to increase the electron affinity compared
bSchool of Engineering and Centre for Molecular and Nanoscale
Electronics, University of Durham, Durham, UK DH1 3LE.
E-mail: m.c.petty@durham.ac.uk
{ Electronic supplementary information (ESI) available: B3LYP/6-
31G(d) optimised geometries (figures and tables of coordinates);
orbital energy level diagrams and frontier orbital localisation for
compounds 6a, 7a, 6a-iso and 7a-iso; X-ray crystallographic data for
compound 11 and a diagram and discussion of its crystal structure. See
DOI: 10.1039/b510003h
{ On leave from the L. M. Litvinenko Institute of Physical Organic
and Coal Chemistry, National Academy of Sciences of Ukraine,
R. Luxemburg Street 70, Donetsk 83114, Ukraine.
5164 | J. Mater. Chem., 2005, 15, 5164–5173
This journal is ß The Royal Society of Chemistry 2005

View Article Online
to OXD.8 Thiophene analogue 11 was synthesised to probe the
effect of heterocycle substitution on the UV–Vis absorption
and photoluminescence properties. Density functional theory
(DFT) calculations (see below) established that compound 6a,
with the pyridyl nitrogen adjacent to the fluorene, is a stronger
electron acceptor than its isomer 6a-iso which has the pyridyl
nitrogen adjacent to the oxadiazole ring. Therefore, we chose
the former system as our synthetic target.
is shown in Scheme 1. 6-Bromonicotinic acid 1 was synthesised
from 2-bromo-5-methylpyridine following the literature pro-
cedure.11 The reaction of compound 1 with thionyl chloride at
reflux gave 2-chloropyridine-4-carbonyl chloride which was
reacted directly with 4-dodecyloxybenzoic acid hydrazide 2 in
pyridine to give the intermediate dihydrazide which was not
purified. In situ dehydrative cyclisation in refluxing phos-
phorus oxychloride,12 followed by reaction with HBr in glacial
acetic acid (to convert the 2-chloro substituent to 2-bromo)
gave the functionalised 2-phenyl-5-pyridyl-1,3,4-oxadiazole
reagent 3 in 37% yield from 1. Two-fold reactions of 3
with 9,9-dihexylfluorene-2,7-diboronic acid 413 or the dipina-
colboronate reagent 514 under palladium-catalysed Suzuki–
Miyaura conditions gave the target compounds 6 and 7,
respectively, in 35% and 56% yields. The spirobifluorene
core of 7 was chosen because it is known that it can be
advantageous in eliminating the formation of fluorenone
defects which occur in some fluorene derivatives during
device operation; such defects act as charge traps and impair
device performance.15 The dodecyloxy chain was attached to
ensure good solubility in organic solvents at all stages of the
synthesis.
Our strategy was to combine the strong electron-transport-
ing ability of the 2-phenyl-5-pyridyl-1,3,4-oxadiazole units
with a 9,9-dihexylfluorene core unit to provide blue emission9
and to utilise energy- or charge-transfer processes within
blended layer films to improve further the efficiency of the
OLEDs. The attraction of single-layer blended devices of this
type is that their manufacture requires only a single spin-
coating process.10 Comparisons of 6 and 7 with 8 and 9
show that the pyridyl rings of 6 and 7 have a significant and
beneficial effect on the optoelectronic properties of the devices.
Comparisons with compound 11 show that the absorption and
photoluminescence properties can be tuned by heterocycle
substitution in the backbone.
Results and discussion
The thiophene analogue 11 was similarly obtained by two-
fold reaction of 4 with compound 1016 (Scheme 2).
Synthesis
The X-ray molecular structure of 11 was determined.§
The most notable feature is that the interplanar angles
between the outer 6-membered rings of the fluorene moiety
and the adjacent thiophene rings vary from 4.6 to 36.4u, but on
The 2-phenyl-5-pyridyl-1,3,4-oxadiazole ring system has not
been reported previously. The route to compound 3 and its
subsequent conversion into the target ETHB materials 6 and 7
Scheme 1 Reagents and reaction conditions: (i) 1, SOCl2, reflux; then add 2, pyridine, reflux; then POCl3, reflux; then DCM, HBr (33% in glacial
acetic acid) 37% yield; (ii) 4, Pd[PPh3]4, Na2CO3, THF, reflux, 48 h, 35% yield; (iii) 5, Pd[PPh3]4, tri-tert-butylphosphine, K2CO3, toluene, 80 uC,
48 h, 56% yield.
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 5164–5173 | 5165

View Article Online
Scheme 2 Reagents and reaction conditions: (i) 4, Pd(PPh3)2Cl2, tri-tert-butylphosphine, Na2CO3, THF, reflux, 96 h, 42% yield.
average are smaller than the corresponding angles in 8,7b which
has benzene rings instead of thiophene. For a more detailed
discussion see the ESI.
Theoretical calculations
DFT calculations were performed to elucidate the geometry
and the electronic state of the new pyridyl-containing deriva-
tives 6 and 7 in comparison with previously studied ET
compounds 8 and 9.7b To decrease the computational time,
calculations were performed on molecules 6a and 7a (where
index ‘‘a’’ means that C12H25O was replaced by CH3O, and
C6H13 was replaced by C2H5; see ESI{). Calculations were also
performed on the geometrical isomers of compounds 6a and
7a, with different substitution positions at the pyridine rings
(compounds 6a-iso and 7a-iso).
§ The single-crystal X-ray diffraction experiment was carried out on a
Bruker SMART 3-circle diffractometer with an APEX CCD area
detector, using graphite monochromated Mo–Ka radiation (l =
˚
0.71073 A) from a 60 W Mo-target microfocus Bede Microsource1
X-ray generator with glass polycapillary X-ray optics and
a
Cryostream-Plus open-flow N cryostat. The structure was solved by
2
direct methods and refined by full-matrix least squares against F2 of all
reflections, using SHELXTL 6.14 software (Bruker-Nonius AXS,
Madison, WI, USA, 2003). Crystal data: C57H62N4O2S2, M = 899.23,
The optimised geometries of compounds 6a-iso and 7a-iso
are generally similar to the previously reported benzene
analogues 8a and 9a, whereas the reduced steric hindrance
between the pyridine and fluorene rings in 6a and 7a results in
substantial planarisation of the system and consequently an
increase in the conjugation (see Fig. S4 in ESI{). Thus, the
¯
T = 120 K, triclinic, space group P1 (no. 2), a = 16.095(2), b =
˚
16.647(2), c = 19.390(2) A, a = 82.61(1), b = 87.70(1), c = 72.94(1)u, U =
4925(1) A , Z = 4, Dc = 1.213 g cm23, m = 0.15 mm21, 47 733
3
˚
reflections (2h ( 50u), 17 322 unique, Rint = 0.093, R = 0.048 [6940
data with F2 > 2s(F2)], wR(F2) = 0.147 (all data). CCDC reference
number 282827. For crystallographic data in CIF or other electronic
format see DOI: 10.1039/b510003h
5166 | J. Mater. Chem., 2005, 15, 5164–5173
This journal is ß The Royal Society of Chemistry 2005

View Article Online
Fig. 1 B3LYP/6-311G(2d,p)//B3LYP6-31G(d) frontier orbitals of compounds 6a and 6a-iso.
dihedral angles between the pyridine and fluorene rings in
compounds 6a and 7a are 15.5u and 16.3u, respectively,
whereas these angles in 6a-iso and 7a-iso are 36.4u and 36.5u,
respectively, which are similar to those between the fluorene
and adjacent benzene rings in 8a and 9a, viz. 35.9u and 36.3u,
respectively. All six structures show planarity at the oxadiazole
site; the dihedral angles between the oxadiazole moiety and the
adjacent benzene and pyridine rings are less then 1u.
The localisation of the HOMO and LUMO orbital coeffi-
cients for compounds with different substituents at position 9
of the fluorene moiety is quite similar for both series, i.e. 6a/7a
and 6a-iso/7a-iso. The main population of the HOMO is on the
central aromatic fluorene moiety with some extension onto the
pyridine and oxadiazole moieties, and quinoidal character of
the LUMO orbital (see Fig. S5 and S6 in ESI{). Comparison
of the two isomers, 6a and 6a-iso, demonstrates some
differences in the occupancy of their HOMOs. In the case of
6a, better planarity and consequently increased conjugation
between the pyridine and fluorene rings, results in increased
population on the central fluorene ring, whereas for the
isomer 6a-iso the HOMO is more delocalised over the molecule
even extending onto the terminal alkoxyphenyl moieties
(Fig. 1).
The most interesting feature of the electronic structure of the
new compounds 6 and 7, compared to their benzene analogues
8 and 9, is the difference in the frontier orbital energies (Fig. 2).
Whereas the HOMO energies for both series of compounds are
almost the same (the difference is only #0.03–0.04 eV), a
pronounced decrease in the LUMO energies by 0.28 eV is
observed when the benzene rings in 8a and 9a are replaced
by pyridine rings (6a and 7a) (Fig. 2). Substitution of the
phenyl rings in 8a and 9a by isomeric pyridine rings
(compounds 6a-iso and 7a-iso) results in a decrease in
HOMO energy levels by 0.14 eV (in contrast to 6a and 7a)
whereas the decrease in their LUMO energies is less pro-
nounced (0.19 eV), so the isomers 6a-iso and 7a-iso are
expected to be weaker electron acceptors than 6a and 7a
(although still stronger acceptors compared to 8 and 97b).
Similar results were obtained from the comparative orbital
analysis of compounds 6a, 7a, 8a and 9a at the B3LYP/6-
31G(d) level of theory (very similar HOMO orbital energies
and a decrease in LUMO energies upon replacing the benzene
rings by pyridine rings; see Fig. S3 in ESI{). This pronounced
decrease in LUMO orbital energies in compounds 6a and 7a
originates from the electron-deficient character of the pyridine
rings and their orientation with respect to the fluorene moiety;
Fig. 2 B3LYP/6-311G(2d,p)//B3LYP/6-31G(d) orbital energy level diagrams for compounds 6a, 7a and their comparison with isomers 6a-iso and
7a-iso, and benzene analogues 8a and 9a.
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 5164–5173 | 5167

View Article Online
Fig. 3 Normalised UV–Vis absorption and photoluminescence
spectra of compounds 8, 6 and 11 in dichloromethane.
this makes them stronger electron acceptors and on this basis
increased ET properties and improved performance of OLEDs
would be expected from compounds 6 and 7.
Fig. 5 The external quantum efficiencies of MEH–PPV polymer
blend OLEDs incorporating 70% by weight of 6, 7, 8 and 9. Device
configuration: ITO/MEH–PPV : ETHB/Al.
Optoelectronic properties and device performance
The UV–Vis absorption and photoluminescence spectra in
dichloromethane (Fig. 3) show an incremental red shift
along the series 8, 6, 11 (Abs: 356 A 372 A 394 nm; PL:
401 A 411 A 439 nm), i.e. as the benzene rings are replaced
by pyridines and thiophenes, consistent with a consecutive
lowering of the HOMO–LUMO gap.
The current versus voltage (I–V) and light output versus
voltage (L–V) characteristics of single-layer polymer blend
devices using 6, 7, 8 and 9 are shown in Fig. 4 (positive bias
applied to the ITO electrode). The device configuration
was ITO/MEH–PPV : ETHB/Al and the polymer blends all
contained 70% of the electron transport materials by weight.
While the current density was similar for all the devices
investigated, the electroluminescent output varied consider-
ably. The light outputs from the new derivatives 6 and 7
were significantly higher than for the previously studied com-
pounds 8 and 9. For example, at 10 6 105 V cm21, the light
emission from blended devices containing 6 and 7 was about
2.5 times higher and 10 times higher, respectively, than from
the device incorporating 8.
Fig. 5 shows the external quantum efficiencies of the OLEDs
(data calculated from the optoelectronic characteristics of
Fig. 4). The efficiency of devices based on 70% blends of 6 and
7 was ca. 0.08% and 0.25%, respectively. In comparison, the
efficiency of blend devices incorporating 70% of 8 and 9 was
ca. 0.05%, while that of a pure MEH–PPV reference device
was 1023%, and for OLEDs based on pure 7 less than 1023%.
The increase of the external quantum efficiency of an OLED
can result from either an increase in the light emission or a
decrease in the current (or both). From Fig. 4, it is evident that
the improvement in the efficiencies of the devices containing 6
and 7 is due to increased light emission. This can almost
certainly be attributed to an enhanced electron (minority
carrier) injection from the top aluminium cathode.
Fig. 4 Current density versus electric field and light output versus
electric field characteristics for MEH–PPV polymer blend OLEDs
incorporating 6, 7, 8 and 9. The polymer blends each contain 70%
by weight of the electron transport materials. Device configuration:
ITO/MEH–PPV : ETHB/Al.
The variation of quantum efficiency with the blend
composition is depicted in Fig. 6. While the efficiency for
5168 | J. Mater. Chem., 2005, 15, 5164–5173
This journal is ß The Royal Society of Chemistry 2005

View Article Online
Fig. 7 The EL spectra of a pure MEH–PPV OLED and blended-
layer devices incorporating compounds 50% 8, 70% 6, and 50% 7 by
weight.
Fig. 6 External quantum efficiency of blended MEH–PPV OLEDs
incorporating 6, 7 and 8. Data are shown for blend devices with 50%,
70%, and 95% by weight.
observed for our blends. This suggests that the two methods
of enhancing the electron injection into the emissive polymer
are mutually exclusive, i.e. one can either use a low work
function metal such as Ca, or exploit a blended-layer structure
containing an electron transporting component. For the latter
devices, it appears that there is not much further gain in then
adding Ca as a top electrode. Indeed, for the systems studied in
our work the use of blended layers with an Al cathode offers a
distinct advantage in terms of environmental stability and ease
of handling as Ca electrodes are highly reactive and unstable
in the atmosphere. Previous workers noted no significant
improvement in EL efficiency by replacing Al with Ca for an
emissive poly(p-phenylenevinylene) derivative with a pendant
1,3,4-oxadiazole substituent.18 Prior to our work it was not
clear that blended layer devices comprising materials with very
different molecular weights would achieve the same result.
Fig. 7 contrasts the EL spectrum of a pure MEH–PPV
device with the spectra from 50% 8, 50% 7 and 70% 6 blended-
layer OLEDs. The emission from all these devices was from
MEH–PPV. No EL could be detected from the ET materials,
which are all blue emitters. The pure MEH–PPV device
exhibited a main peak at 570 nm while the spectra of the
blended-layer devices were red-shifted, with the main peaks
located at ca. 590 nm. Devices incorporating 9 also showed
a similar red-shifted EL spectrum (data not shown). All
these materials possess C12H25O groups at both ends of the
molecules. It is believed that these long terminal groups are
responsible for the red shift, as the EL spectrum from a
blended-layer device incorporating a compound with the same
molecular structure as 8, but no long hydrocarbon end groups,
did not show the red-shift.7b It is known that the emission of
conjugated polymers such as MEH–PPV can be changed by
varying its chain conformation. This, in turn, is affected by the
polarity of the solvent or by the spin casting speed.19,20
8-containing OLEDs increased with concentration over the
range of composition investigated, the efficiencies of devices
incorporating compounds 6 and 7 appeared to saturate (in the
case of 6) or even reduce (7) for devices containing high
concentrations of the ET materials. The diminished efficiency
of OLEDs containing compound 7 is probably related to a
decrease in the light emission at high concentrations (95%) as
the hole current is expected to reduce for blends consisting
predominantly of the ET compound. The maximum external
quantum efficiency that could be achieved from the blended-
layer devices was 0.24% for the OLED containing 70% of
compound 7. This efficiency is over 200 times higher than
that for pure MEH–PPV devices and about twice the highest
values achieved with blended-layer devices incorporating
2,5-diphenyl-1,3,4-oxadiazole–fluorene hybrids 8 and 9.6 It
is also significant that the 7 blended-layer devices were more
efficient than OLEDs incorporating compound 6 throughout
the concentration range investigated. There are other examples
in the literature where the spiro-bifluorene system enhances
efficiency compared to the 9,9-dialkylfluorene analogues which
can yield fluorenone defects which quench the emission.15
The external quantum efficiency of our OLEDs could be
increased further by using a layer of poly(3,4-ethylenedioxy-
thiophene) doped with poly(styrenesulfonate) (PEDOT : PSS)
between the ITO and the blended-layer film.17 This resulted in
an efficiency increase of 2–3 times, regardless of the composi-
tion and material. A 70% compound 7-containing blend device
with PEDOT exhibited an external quantum efficiency of
0.5% and a luminous efficiency of 0.93 cd A21 at 9.5 V and a
luminance of 100 cd m22. When the Al cathode was replaced
by Ca/Al (thermally evaporated), the efficiency values for the
70% 7-based device increased further to 0.6% and 1.2 cd A21
at 10.5 V.
The implication of this work and of our previous studies7 is
that the components of the blended layers (polymer and ET
material) are intimately mixed. The fact that no direct EL from
the ET materials could be measured, even at high concentra-
tion (up to 95%), implies that the molecules of these materials
are well-distributed among the MEH–PPV polymer chains.
The above result is interesting as the use of Ca increases
significantly the efficiency with pure MEH–PPV (at 5 V and
100 cd m22 the EQE and the luminous efficiency for an ITO/
PEDOT/MEH–PPV/Ca/Al device were 0.42% and 0.72 cd A21
,
respectively). However, only a very modest increase was
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 5164–5173 | 5169

View Article Online
Fig. 8 AFM images of the 70% 6 (a) and 50% 7 (b) blend films spin-cast from the blend solution using chloroform as the solvent; 50% 6 (c) and
50% 7 (d) blend films from the blend solutions using the mixture of chloroform and p-xylene (3 : 1 v/v).
Under certain conditions, however, we found that phase
separation could be observed for blended-layer devices based
on 6 and 7; for example, by using films that were spin-cast
from solutions of a mixed solvent. Fig. 8 shows atomic force
microscope images of 70% 6 (a) and 50% 7 (b) blended-layer
films spin-cast from the mixed solution using chloroform as
the solvent; and 50% 6 (c) and 50% 7 (d) films formed using a
mixed solvent system, viz. chloroform and p-xylene (3 : 1 v/v).
There is no notable morphology in the blend films spin-coated
from the pure chloroform solution (the micron-size particles
evident in Figs. 8a and 8b are thought to originate from
environmental contamination during processing). However, a
needle-like phase is clearly visible in both the blended-layer
films formed from the mixed solvent. The phase separation
was particularly acute in the films containing compound 7.
This is probably related to the presence of the bulky spiro unit
at the centre of the molecule, which inhibits mixing with the
chains of the MEH–PPV. (This solvent effect was not observed
in comparable experiments using ET materials 8 and 9). The
microstructure and crystallinity of MEH–PPV films are known
to vary depending on the solvents from which the film has
been cast.21
compound 7. These peaks coincide with the emission from
devices based on the pure ET compounds 6 and 7. The phase
separation also affected the external quantum efficiencies of
the devices. The efficiency of all these devices decreased as the
blend composition exceeded 50% of the ET material and was
less than 0.1%. The interface between the separated phase and
the matrix can act as a quenching site where non-radiative
singlet-exciton decay takes place, consequently reducing the
light emission and the efficiency. Another reason for the
efficiency decrease with the concentration can be that more
light will be emitted from clusters of 7.
In contrast to the results presented earlier (Fig. 7), EL
from the ET compounds was evident in OLEDs incorporating
phase-separated layers. Fig. 9 compares the EL spectra of a
70% 7 blended-layer device formed from pure chloroform with
70% 6 and 70% 7 blended-layer devices formed using the mixed
solvent. Small shoulders in the 420–520 nm region can be
seen in the spectra for the OLEDs fabricated using the mixed
solvent; this is more evident for the device containing
Fig. 9 The EL spectra of 70% 7 devices spin-cast from chloroform
solution; 70% 6 and 7 device from the solution using a mixture of
chloroform (CF) and p-xylene (Xy) solvent.
5170 | J. Mater. Chem., 2005, 15, 5164–5173
This journal is ß The Royal Society of Chemistry 2005

View Article Online
Electrical measurements were undertaken in a vacuum
chamber (1021 mbar). The d.c. bias was applied and the
current measured by a Keithley 2400 Source Meter. The light
emitted from the device was collected by a large area
photodiode (1.5 cm diameter) connected to a Keithley 485
Digital Picoammeter. For external quantum efficiency mea-
surements, the light power was calculated using the photo-
current and the conversion factor (wavelength dependent)
of the photodiode (ampere/watt). Electroluminescence (EL)
spectra were measured using an Ocean Optics USB2000
Miniature Fibre Optic Spectrometer. The surface morpho-
logies of the blend films were observed using a Digital
Instrument NanoScope E atomic force microscope.
Conclusions
We have synthesised the first 2,5-diaryl-1,3,4-oxadiazole–
fluorene hybrids which incorporate pyridine units within the
linearly-extended p-electron system, viz. 2,7-bis{5-[2-(4-dodecyl-
oxyphenyl)-1,3,4-oxadiazol-5-yl]-2-pyridyl}-9,9-dihexylfluorene
(6) and 2,7-bis{5-[2-(4-dodecyloxyphenyl)-1,3,4-oxadiazol-5-yl]-
2-pyridyl}spirobifluorene (7). Calculations at the DFT level
establish that inclusion of the pyridyl rings in 6 and 7
considerably enhances the electron affinity of the system,
compared to regioisomers and phenyl analogues. Single-layer
OLEDs were fabricated by spin-coating blends of MEH–
PPV as the emissive material with added ET compounds 6
or 7. The external quantum efficiencies of the devices were
greatly enhanced compared to pure MEH–PPV reference
devices. The EQE for the device ITO/PEDOT : PSS/MEH–
PPV-7 (30 : 70% by weight)/Al reached 0.5% and a luminous
efficiency of 1.1 cd A21 at 11 V and a luminance of
100 cd m22. Replacing the Al cathode with Ca/Al gave only
a modest increase in efficiency (EQE 0.6% and 1.3 cd A21 at
9.3 V) unlike the pure MEH–PPV reference device where a
far greater increase in efficiency occurs. This leads to the
important conclusion that electron injection into the MEH–
PPV emitter can be enhanced either by using a low work
function metal such as Ca, or by exploiting a blended-layer
structure. There are important practical implications in this
discovery, as if comparable device performance can be
achieved, it is preferable to use Al electrodes rather than Ca
electrodes: the latter are highly reactive and are unstable in
the atmosphere. Further work which exploits the advantages
of blended-layer devices is in progress.
Computational procedures
The ab initio computations for compounds 6a, 7a, 6a-iso, and
7a-iso were carried out with the Gaussian 0322 package of
programs at density-functional theory (DFT) level. The
geometries were optimised using Pople’s 6-31G split valence
basis set supplemented by d-polarisation functions on heavy
atoms. In the DFT calculations, Becke’s three-parameter
hybrid exchange functional23 with Lee–Yang–Parr gradient-
corrected correlation functional (B3LYP)24 was employed. No
symmetry restrictions and no constraints of bonds/angles/
dihedral angles were applied and all atoms were free to
optimise. For the B3LYP/6-31G(d) optimised geometries,
electronic structures were then calculated for single points
using a triple split-valence 6-311 basis set supplemented by two
d-polarisation functions on heavy atoms and p-polarisation
functions on hydrogens [B3LYP/6-311G(2d,p)]. Contours of
HOMO and LUMO orbitals were visualised using the Molekel
v.4.3 program.25
Experimental
6-Bromonicotinic acid (1)
General
2-Bromo-5-methylpyridine (17.2 g, 100 mmol) and phase
transfer agent Aliquat-336 (0.5 cm3) in water (350 cm3) were
heated at 70 uC. Powdered KMnO4 (44 g, 0.28 mol) was added
carefully in small potions over a period of 3 h. The mixture was
then refluxed for a further 1.5 h. After cooling, the solid was
filtered off, washed with hot water and the combined filtrates
were concentrated to ca. 150 cm3, then acidified with HBr
(48%) and cooled in an ice bath. The precipitate was then
filtered off, washed with water, recrystallised from water and
dried to yield 1 as colourless plates (10.1 g, 50%), mp 194.5–
195.0 uC (lit.11 195 uC). Anal. Calcd. for C6H4BrNO2: C, 35.67;
H, 2.00; N, 6.93. Found: C, 35.61; H, 1.99; N, 6.95%. dH
(DMSO-d6, 400 MHz) 8.87 (d, J 2, 1H), 8.16 (dd, Jab 8.2, Jac 2,
1H), 7.80 (d, J 8.2, 1H). dC (CDCl3, 100 MHz) 165.51, 151.12,
145.54, 139.86, 128.23, 126.42.
The procedures and equipment used for the synthesis and
characterisation of materials are the same as those reported
recently.7
Device fabrication and measurement
MEH–PPV was purchased from Aldrich whereas the ET
materials were synthesised as described below. Indium–tin–
oxide (ITO) coated glass from Merck with sheet resistance
21
of 9 V %
was used as the anode. This was cleaned by
ultrasonication in acetone and isopropyl alcohol for 30 min
each and dried with a nitrogen gun. The polymer and ET
materials were dissolved in chloroform or in a mixture of
chloroform and p-xylene (3 : 1 v/v) to provide the blend
solution, which was spin-coated onto the ITO. The concentra-
tion of ET material was changed from 50% to 95% of the total
weight. Following the spin-coating, Al or Ca/Al top electrodes,
in the form of dots (radius 1 mm; thickness 150 nm) were
thermally evaporated at a pressure of about 1026 mbar. In
some cases PEDOT : PSS, purchased from Bayer AG, was
spin-coated onto the ITO prior to the deposition of the
polymer blend. These PEDOT layers (40 nm in thickness) were
dried for 12 h in nitrogen at room temperature to remove
residual solvent.
2[5-(2-Bromopyridyl)]-5-(4-dodecyloxyphenyl)-1,3,4-oxadiazole
(3)
6-Bromonicotinic acid 1 (6.6 g, 32.7 mmol) was refluxed in
thionyl chloride (40 cm3) for 12 h under nitrogen. The excess
thionyl chloride was then removed by vacuum distillation. Dry
toluene (30 cm3) was then added and distilled off at reduced
pressure to remove any remaining thionyl chloride. After
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 5164–5173 | 5171

View Article Online
cooling to rt, 4-dodecyloxybenzoic acid hydrazide 226(10.58 g,
33 mmol) dissolved in pyridine (50 cm3) was added via a
syringe. The solution was stirred for 1 h at rt and then for a
further 2 h at 110 uC. Pyridine was removed by vacuum
distillation. After cooling to rt, methanol (50 cm3) was then
added and the suspension was heated at reflux for 15 min,
cooled and a solid obtained by vacuum filtration. The crude
intermediate product was dried for 12 h under high vacuum
and then refluxed with phosphorus oxychloride (40 cm3) for
12 h after which time the POCl3 was removed by vacuum
distillation yielding a cream solid. Methanol (50 cm3) was then
added and the suspension was heated at reflux for 15 min,
cooled and the solid obtained by vacuum filtration. The crude
product was dried for 12 h under high vacuum and then
dissolved in a minimum amount of dichloromethane. HBr (in
glacial acetic acid, 33%) (20 cm3) was then added and the
suspension was stirred at rt for 96 h. The suspension was then
diluted with H2O (20 cm3) and neutralised with Na2CO3 (2 M).
The mixture was extracted with dichloromethane, washed
with NaOH and then H2O and filtered through celite before
drying with MgSO4 and filtering. The solution was concen-
trated in vacuo and the crude product was then purified
by column chromatography (silica, eluent: CHCl3–EtOAc
19 : 1 v/v) and crystallisation (ethanol) to yield white crystals
of compound 3 (5.9 g, 37%), mp 144–144.5 uC. MS (EI) m/z 487
(M+, 81Br), 485 (M+, 79Br). Anal. Calcd for C25H32BrN3O2: C,
61.73; H, 6.63; N, 8.64. Found: C, 61.58; H, 6.63; N, 8.43%. dH
(CDCl3, 400 MHz) 9.08 (d, J 2.4, 1H), 8.28 (dd, Jab 8.4, Jab 2.4,
1H), 8.07 (d, J 8.8, 2H), 7.69 (d, J 8.4, 1H), 7.04 (d, J 8.8, 2H),
4.05 (t, J 6.8, 2H), 1.83 (m, 2H), 1.28 (m, 18H), 0.89 (t, J 6,
3H), dC (CDCl3, 100 MHz) 165.30, 162.40, 160.00, 154.21,
147.68, 136.50, 128.89, 128.54, 119.63, 115.50, 115.14, 68.38,
31.91, 29.64, 29.62, 29.58, 29.55, 29.35, 29.33, 29.11, 25.99,
22.67, 14.09.
(0.073 g, 35%), mp 166.7–167.5 uC. MS (MALDI-TOF) m/z
Calcd. for C75H96N6O4: 1145.76. Found 1145.73 (M+); Anal.
Calcd. for C75H96N6O4: C, 78.63; H, 8.45; N, 7.34. Found: C,
78.99; H, 8.51; N, 6.94%. dH (CDCl3, 400 MHz) 9.44 (d, J 2,
2H), 8.51 (dd, Jab 8, J 2, 2H), 8.14 (m, 4H), 8.12 (d, J 8.8,
ac
4H), 8.02 (d, J 8, 2H), 7.9 (d, J 8, 2H), 7.06 (d, J 8.8, 4H), 4.06
(t, J 6.4, 4H), 2.17 (m, 4H), 1.84 (m, 4H), 1.5–1.3 (m, 36H),
1.1 (m, 12H), 0.9 (t, J 6.4, 6H), 0.74 (m, 10H). dC (CDCl3,
100 MHz) 165.02, 162.25, 159.81, 152.33, 147.66, 142.40,
134.80, 134.78, 128.84, 126.43, 121.59, 120.64, 120.44, 120.41,
118.62, 115.86, 115.12, 68.37, 55.70, 53.39, 40.38, 31.91, 31.50,
29.65, 29.63, 29.59, 29.56, 29.37, 29.34, 29.13, 26.00, 23.88,
22.68, 22.56, 14.10, 13.95. UV–Vis (DCM) lmax 372 nm, PL
(DCM) lmax 411, 431 nm.
2,7-Bis{5-[2-(4-dodecyloxyphenyl)-1,3,4-oxadiazol-5-yl]-2-
pyridyl}spirobifluorene (7)
Compound 3 (0.486 g, 1 mmol), compound 5 (0.284 g,
0.5 mmol), tetrakis(triphenylphosphino)palladium (58 mg,
0.05 mmol), tri-tert-butylphosphine (0.02 g, 0.1 mmol), toluene
(20 cm3) and K2CO3 (2 M, 2.5 cm3); reaction time 48 h;
extracted with chloroform. Chromatography eluent: CHCl3–
EtOAc (17 : 3 v/v) followed by recrystallisation from a
toluene–ethanol mixture gave compound 7 as a yellow solid
(0.32 g, 56%), mp 203.0–203.4 uC. MS (MALDI-TOF) m/z
Calcd. for C75H78N6O4: 1127.62. Found 1127.65 (M+); Anal.
calcd. for C75H78N6O4: C, 79.90; H, 6.97; N, 7.45. Found: C,
79.70; H, 7.00; N, 7.26%. dH (CDCl3, 400 MHz) 9.27 (s, 2H),
8.33 (d, J 8.4, 2H), 8.23 (d, J 8, 2H), 8.04 (m, 6H), 7.94 (d, J 8,
2H), 7.72 (d, J 8.4, 2H), 7.43 (m, 4H), 7.15 (t, J 7.6, 2H), 7.02
(d, J 8.4, 4H), 6.84 (d, J 7.6, 2H), 4.04 (t, J 6, 4H), 1.82 (m,
4H), 1.5–1.3 (m, 36H), 0.89 (t, J 6.4, 6H). dC (CDCl3,
100 MHz) 164.95, 162.19, 162.09, 159.13, 150.4, 147.98,
147.45, 142.85, 142.0, 138.30, 134.57, 128.81, 128.10, 127.48,
127.17, 124.25, 122.72, 121.00, 120.43, 120.30, 118.62, 115.76,
115.06, 68.33, 66.15, 31.92, 29.66, 29.64, 29.59, 29.57, 29.39,
29.37, 29.12, 25.99, 22.70, 14.13. UV–Vis (DCM) lmax 371 nm,
PL (DCM) lmax 407, 430 nm.
General procedure for the cross-coupling reactions
The diboronic acid or boronate reagent, the halide and
the catalyst [tetrakis(triphenylphosphino)palladium] (5 mol %
relative to the boronic acid) were added sequentially to
degassed THF or toluene and the mixture was stirred at
20 uC for 0.5 h. Degassed aqueous Na2CO3 or K2CO3 solution
was added and the reaction mixture was heated under argon
(oil bath temperature 80 uC) until TLC monitoring showed
that the reaction was complete. Solvent was evaporated
in vacuo and the crude products were extracted into organic
solvent. The organic layer was washed with H2O, separated
and dried over MgSO4. Products were purified by column
chromatography.
2,7-Bis{5-[5-(4-tert-butylphenyl)-1,3,4-oxadiazol-2-yl]-thien-2-
yl}-9,9-dihexylfluorene (11)
By analogy with the synthesis of 6, compound 10 (0.3 g,
0.73 mmol), compound 4 (0.17 g, 0.40 mmol), Pd(PPh3)2Cl2
(29 mg, 0.04 mmol), tri-tert-butylphosphine (0.01 cm3,
0.1 mmol), THF (50 cm3) and Na2CO3 (2 M, 1.0 cm3);
reaction time 96 h; extracted with toluene. Chromatography
eluent: DCM–EtOAc (32.3 : 1 v/v) followed by DCM–EtOAc
(9 : 1 v/v) and recrystallisation from toluene–ethanol gave
yellow needles of 11 (0.15 g, 42%), mp 252.6–253.0 uC. MS (EI)
m/z 898 (M+, 100%). Anal. calcd. for C57H62N4O2S2: C, 76.13;
H, 6.95; N, 6.23. Found: C, 75.73; H, 6.89; N, 5.91%. dH
(CDCl3, 400 MHz) 8.08 (d, J 8.4, 4H), 7.84 (d, J 4, 2H), 7.78
(d, J 8.0, 2H), 6.70 (d, J 8.0, 2H), 7.65 (s, 2H), 7.57 (d, J 8.4,
4H), 7.47 (d, J 4, 2H), 2.07 (m, 4H), 1.40 (s, 18H), 1.11 (m,
12H), 0.77 (m, 10H). dC (CDCl3, 100 MHz) 164.12, 160.93,
155.43, 152.16, 149.75, 141.11, 132.39, 130.54, 128.81, 126.08,
125.34, 123.85, 123.75, 120.94, 120.61, 120.41, 55.53, 40.35,
2,7-Bis{5-[2-(4-dodecyloxyphenyl)-1,3,4-oxadiazol-5-yl]-2-
pyridyl}-9,9-dihexylfluorene (6)
Following the general procedure, compound 3 (0.196 g,
0.4 mmol), compound 4 (0.077 g, 0.18 mmol), tetrakis(tri-
phenylphosphino)palladium (23 mg, 0.02 mmol), THF (25 cm3)
and Na2CO3 (2 M, 0.6 cm3); reaction time 48 h; extracted with
DCM. Chromatography eluent: DCM–EtOAc (19 : 1 v/v)
followed by a further column, eluent: DCM–EtOAc (9 : 1 v/v)
and recrystallisation from ethanol gave 6 as a white solid
5172 | J. Mater. Chem., 2005, 15, 5164–5173
This journal is ß The Royal Society of Chemistry 2005

View Article Online
13 G. Hughes, C. Wang, A. S. Batsanov, M. Fearn, S. Frank,
M. R. Bryce, I. F. Perepichka, A. P. Monkman and B. P. Lyons,
Org. Biomol. Chem., 2003, 1, 3069.
35.12, 31.43, 31.13, 29.62, 23.79, 22.54, 13.96. UV–Vis (DCM)
lmax 394 nm, PL (DCM) lmax 439, 466 nm.
14 K.-T. Wong, Y.-T. Chien, R.-T. Chen, C.-F. Wang, Y.-T. Lin,
H.-H. Chiang, P.-Y. Hsieh, C.-C. Wu, C. H. Chou, Y. O. Su,
G.-H. Lee and S.-M. Peng, J. Am. Chem. Soc., 2002, 124, 11576.
15 (a) L. Romaner, A. Pogantsch, P. Scandiucci de Freitas, U. Scherf,
M. Gaal, E. Zojer and E. J. W. List, Adv. Funct. Mater., 2003,
13, 3069; (b) M. R. Craig, M. M. de Kok, J. W. Hofstraat,
A. P. H. L. Schenning and E. W. Meijer, J. Mater. Chem., 2003, 13,
2861.
Acknowledgements
This work was funded by EPSRC, Durham County Council
under the Science and Technology for Business and Enterprise
Programme SP/082, and the Postdoctoral Fellowship
Programme of Korea Science and Engineering Programme
(KOSEF) (for J. H. A.).
16 G. Hughes, D. Kreher, C. Wang, A. S. Batsanov and M. R. Bryce,
Org. Biomol. Chem., 2004, 2, 3363.
17 For a review on the applications of PEDOT : PSS see:
S. Kirchmeyer and K. Reuter, J. Mater. Chem., 2005, 15, 2077.
18 J. H. Kim, J. H. Park and H. Lee, Chem. Mater., 2003, 15,
3414.
19 T. Q. Nguyen, V. Doan and B. J. Schwartz, J. Chem. Phys., 1999,
110, 4068.
References
1 (a) S. M. Kelly, Flat Panel Displays, Royal Society of Chemistry,
Cambridge, 2000; (b) Organic Electroluminescence, ed. Z. H. Kakafi,
CRC Press, Boca Raton, 2005.
20 Y. Shi, J. Liu and Y. Yang, J. Appl. Phys., 2000, 87, 4254.
21 C. Y. Yang, F. Hide, M. A. Diaz-Garcia, A. J. Heeger and Y. Cao,
Polymer, 1998, 39, 2299.
2 C. W. Tang and S. A. VanSlyke, Appl. Phys. Lett., 1987, 51, 913.
3 J. H. Burroughes, D. D. C. Bradley, A. R. Brown, R. N. Marks,
K. McKay, R. H. Friend, P. L. Burn and A. B. Holmes, Nature,
1990, 347, 539.
22 Gaussian 03, Revision B.04, M. J. Frisch, G. W. Trucks,
H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman,
J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C. Burant,
J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci,
M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji,
M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene,
X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken,
C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala,
K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain,
O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford,
J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham,
C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill,
B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople,
Gaussian, Inc., Wallingford CT, 2004.
4 Reviews: (a) A. Kraft, A. C. Grimsdale and A. B. Holmes, Angew.
Chem., Int. Ed., 1998, 37, 402; (b) U. Mitschke and P. Ba¨uerle,
J. Mater. Chem., 2000, 10, 1471; (c) A. P. Kulkarni, C. J. Tonzola,
A. Babel and S. A. Jenecke, Chem. Mater., 2004, 16, 4556; (d)
G. Hughes and M. R. Bryce, J. Mater. Chem., 2005, 14, 94.
5 B. Schultz, M. Bruma and L. Brehmer, Adv. Mater., 1997, 9, 601.
6 (a) M. Thelakkat and H.-W. Schmidt, Polym. Adv. Technol., 1998,
9, 429; (b) C. Wang, G.-Y. Jung, A. S. Batsanov, M. R. Bryce and
M. C. Petty, J. Mater. Chem., 2002, 12, 173; (c) S. W. Cha,
S.-H. Choi, K. Kim and J.-I. Jin, J. Mater. Chem., 2003, 13, 1900;
(d) S. W. Cha, S.-H. Choi, K. Kim and J.-I. Jin, J. Mater. Chem.,
2002, 13, 1900; (e) F. Liang, Q. Zhou, X. Cheng, L. Wang, D. Ma,
X. Jing and F. Wang, Chem. Mater., 2003, 15, 1935.
7 (a) J. H. Ahn, C. Wang, C. Pearson, M. R. Bryce and M. C. Petty,
Appl. Phys. Lett., 2004, 85, 1283; (b) S. Oyston, C. Wang, G. Hughes,
A. S. Batsanov, I. F. Perepichka, M. R. Bryce, J. H. Ahn, C. Pearson
and M. C. Petty, J. Mater. Chem., 2005, 15, 194.
8 C. Wang, G.-Y. Jung, Y. Hua, C. Pearson, M. R. Bryce,
M. C. Petty, A. S. Batsanov, A. E. Goeta and J. A. K. Howard,
Chem. Mater., 2001, 13, 1167.
23 A. D. Becke, Phys. Rev. A, 1988, 38, 3098; A. D. Becke, J. Chem.
Phys., 1993, 98, 5648.
24 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B, 1988, 37, 785.
25 (a) P. Flu¨kiger, H. P. Lu¨thi, S. Portmann and J. Weber, Molekel,
Version 4.3, Swiss Center for Scientific Computing, Manno,
(Switzerland), 2002; http://www.cscs.ch/molekel/; (b) S. Portmann
and H. P. Lu¨thi, Chimia, 2000, 54, 766.
9 M. T. Bernius, M. Inbasekaran, J. O’Brien and W. Wu, Adv.
Mater., 2000, 12, 1737.
10 Y. Cao, I. D. Parker, G. Yu, C. Zhang and A. J. Heeger, Nature,
1999, 397, 414.
11 P.-M. Windschief and F. Vo¨gtle, Synthesis, 1994, 87.
12 (a) F. N. Hayes, B. S. Rogers and D. J. Ott, J. Am. Chem. Soc.,
1955, 77, 1850; (b) A. Kraft, Chem. Commun., 1996, 77.
26 X. Zhao, X.-Z. Wang, X.-K. Jiang, Y.-Q. Chen, Z.-T. Li and
G.-J. Chen, J. Am. Chem. Soc., 2003, 125, 15128.
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 5164–5173 | 5173