Indole-substituted nickel dithiolene complexes in electronic and optoelectronic devices

Article abstract of DOI:10.1039/c1jm12466h

The synthesis and full characterisation of a novel indole-substituted nickel dithiolene [Ni(mi-5edt)₂] (3) is reported, and compared to its alkyl-substituted analogue [Ni(mi-5hdt)₂] (4) that has been previously communicated [Dalgleish et al., Chem. Commun., 2009, 5826] [mi-5edt = 1-(N-methylindol-5-yl)-ethene-1,2-dithiolate; mi-5hdt = 1-(N-methylindol-5-yl)- hex-1-ene-1,2-dithiolate)]. Both complexes are shown to undergo oxidative electropolymerisation, yielding polymer films that retain the redox and optical properties of the monomer. The more soluble analogue 4 is shown to form high quality thin films by spin coating, which have been utilised to fabricate field-effect transistors (FETs) and bulk heterojunction photovoltaic devices (BHJ-PVs). From FET studies, the material shows ambipolar charge transport behaviour, with a maximum carrier mobility of ～10^-6 cm ² V^-1 s^-1 for electrons. By using 4 simultaneously as the electron acceptor as well as a NIR sensitiser in BHJ-PVs, the complex is shown to contribute to the photocurrent, extending light harvesting into the NIR region.

Full text of DOI:10.1039/c1jm12466h

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2011, 21, 15422
www.rsc.org/materials
PAPER
Indole-substituted nickel dithiolene complexes in electronic and optoelectronic
devices†
Simon Dalgleish,a John G. Labram,b Zhe Li,c Jianpu Wang,c Christopher R. McNeill,c
b
Thomas D. Anthopoulos, Neil C. Greenham and Neil Robertson
c
a
*
Received 1st June 2011, Accepted 27th July 2011
DOI: 10.1039/c1jm12466h
The synthesis and full characterisation of a novel indole-substituted nickel dithiolene [Ni(mi-5edt)2] (3)
is reported, and compared to its alkyl-substituted analogue [Ni(mi-5hdt)2] (4) that has been previously
communicated [Dalgleish et al., Chem. Commun., 2009, 5826] [mi-5edt ¼ 1-(N-methylindol-5-yl)-
ethene-1,2-dithiolate; mi-5hdt ¼ 1-(N-methylindol-5-yl)-hex-1-ene-1,2-dithiolate)]. Both complexes are
shown to undergo oxidative electropolymerisation, yielding polymer films that retain the redox and
optical properties of the monomer. The more soluble analogue 4 is shown to form high quality thin
films by spin coating, which have been utilised to fabricate field-effect transistors (FETs) and bulk
heterojunction photovoltaic devices (BHJ-PVs). From FET studies, the material shows ambipolar
charge transport behaviour, with a maximum carrier mobility of ꢀ10ꢁ6 cm2 Vꢁ1 sꢁ1 for electrons. By
using 4 simultaneously as the electron acceptor as well as a NIR sensitiser in BHJ-PVs, the complex is
shown to contribute to the photocurrent, extending light harvesting into the NIR region.
charge-transport properties, transition metal complexes can also
1. Introduction
display tuneable visible or NIR absorption, leading to applications
in optoelectronic devices, such as photovoltaic cells.
Organic field-effect transistors have enjoyed considerable attention
over the past decade, with confirmed reports of mobilities compa-
rable to those of amorphous silicon.1 Whilst the initial research into
organic semiconductors focussed on purely organic systems, in
recent years focus has broadened to look at transition metal
complexes.2–4 Transition metal complexes often display multiple
stable redox states, corresponding to electron transfer processes
based on the metal or spread over the ligand, and the frontier
orbitals are often relatively non-bonding, leading to low reorgan-
isation energies upon electron transfer and, as such, are attractive
targetsforstablechargetransportinorganicdevices.5 Inadditionto
Improvements in device efficiencies for organic photovoltaic
devices have been realised by increasing the interface between
electron donor and acceptor materials.6 One common method
for achieving this is to form a bulk heterojunction architecture.
Bulk heterojunction photovoltaic cells (BHJ-PVs) are most
commonly prepared by blending a polymeric electron donor with
a substituted fullerene as electron acceptor. To achieve high
photocurrents, light absorption over a large part of the solar
spectrum is required. This has been achieved for binary systems
by replacing PCBM with PC71BM, which has a significantly
broader absorption in the visible range than PCBM,7 or by
developing ternary systems with a sensitiser molecule incorpo-
rated to effect light capture at longer wavelengths.8–11
aSchool of Chemistry and EaStCHEM Research School, University of
Edinburgh, West Mains Road, Edinburgh, EH9 3JJ, UK. E-mail: neil.
robertson@ed.ac.uk
Nickel-bis-1,2-dithiolenes often show non-innocent character,
with multiple redox processes occurring at low potentials due to
a low lying LUMO orbital, and efficient delocalisation of the
electron density over the ligand.12 In addition, the narrow
HOMO–LUMO gap leads to intense low energy absorption.
Despite intensive study over nearly five decades, only a few
studies are reported of their incorporation in field-effect tran-
sistors. Anthropoulos et al. demonstrated ambipolar behaviour
for bis-(4-dimethylaminodithiobenzyl)-nickel, with typical hole
and electron mobilities (m) of 10ꢁ4 cm2 Vꢁ1 sꢁ1,3 which so far
bDepartment of Physics and Centre for Plastic Electronics, Blackett
Laboratory, Imperial College London, London, SW7 2BW, UK
cUniversity of Cambridge, Cavendish Laboratory, JJ Thomson Avenue,
Cambridge, CB3 0HE, UK
† Electronic supplementary information (ESI) available: Table S1:
Selected bond lengths and angles of complexes 3 and 4 from geometry
optimised structures, compared to the crystal structures of 4 and ligand
precursor 1; Table S2: Calculated energies, intensities and contributions
to the NIR absorptions of 3 and 4; Table S3: Redox processes of 3 and
4; Fig. S1: Comparison of 1H NMR of 3 and 1; Fig. S2: Frontier
molecular orbitals of 3 and 4; Fig. S3: Cyclic voltammetry of 4;
Fig. S4: Electropolymerisation of 3 and 4; Fig. S5: Cyclic voltammetry
of poly 4 on Pt electrode, compared to 4 in solution; Fig. S6:
Absorption spectra of P3HT/4 and PPV/4 thin films; Fig. S7:
Photoluminescence spectra of P3HT/4 and PPV/4 thin films. See DOI:
10.1039/c1jm12466h
represents the highest reported for a metal dithiolene.2,4
A
limiting consideration remains the ability of these species to be
readily processed into continuous thin films.
15422 | J. Mater. Chem., 2011, 21, 15422–15430
This journal is ª The Royal Society of Chemistry 2011

View Article Online
We previously communicated the preparation of the nickel bis-
1,2-dithiolene complex (4) (Scheme 1) and its electro-
polymerisation to give a NIR electrochromic film.13 Due to the
propensity of 4 to form uniform thin films by drop or spin
coating, it presents an ideal example to further study the appli-
cation of nickel dithiolene complexes to FETs, and also, for the
first time, to explore their potential as the electron acceptor in
BHJ-PVs. We report the full synthesis, structural character-
isation and study of 4 and the analogous 3, and their application
in (opto)electronic materials and devices through both electro-
polymerisation and solution-coating approaches.
Like most reported examples of unsymmetrically substituted
nickel dithiolenes,19–23 4 crystallised in the trans geometry (Fig. 1
(upper)). Initially, crystallisation failed under a myriad of
growing conditions, favouring instead the formation of contin-
uous films over the vessels (plastic, glass, and hexamethyldisilane
(HMDS) treated glass). However, optimal growing conditions
were found to be elevated temperatures of 25 ꢂC in a wide-based
conical flask from a EtOH/DCM mixture. This was attributed to
the minimised exposure of fresh air/liquid/glass interface upon
evaporation, thus favouring homogeneous crystallisation, rather
than film deposition. Note however that the facile film formation
using 4 forms the basis of the subsequent studies on FET and
BHJ-PV devices.
2. Results and discussion
The packing structure of 4 showed the complex to crystallise in
a columnar stacking arrangement, with an interplanar distance
We previously reported the synthesis of 4, using the route
proposed by Gareau et al. involving synthesis of the dithol-2-one
ligand precursor, via the alkyne.14 To prepare 3 containing
a terminal alkyne, the Corey–Fuchs procedure provides a cheap,
catalyst-free alternative to the Sonogashira reaction used for 2.15
5-bromo-1-methylindole was converted to the aldehyde by
a microwave assisted Grignard reaction, in near quantitative
yield, by modification of the route described by Mutule et al.16
This was converted under a modified Corey–Fuchs procedure to
the alkyne in good yield (57%).17 By this method, a cheap, quick
and relatively simple procedure has been developed toward
a novel dithiolene ligand precursor. Both of these alkyne species
could then be converted to the desired dithiol-2-one (Scheme 1).
The nickel TBA or TMA salts were synthesised by a route similar
to that of Anjos et al.,18 and oxidation to the neutral complex
could be effected using elemental iodine.
ꢀ
of 3.822 A, approaching the van der Waals limit for efficient
sulfur-nickel overlap (Fig. 1 (lower)). Close contact between the
stacks was achieved through the indolyl pyrrole unit (C2) and the
dithiolene S (Fig. 1 (upper)). As observed for other pendent
aromatic systems,24,25 the indolyl groups were twisted out of the
molecular plane by some 60ꢂ, thus prohibiting a closer packing of
the dithiolene cores. This was interesting as the incorporation of
a flexible butyl group was included to allow the in-plane twist
of the indolyl group, and thus suggested that the origin of this
twist perhaps did not lie solely with sterics.26 Compared to the
crystal structure of the ligand precursor 2, which showed classical
dithiolate bond lengths for the C–S and C]C bonds (Table S1,
ESI†), the complexed ligand showed a shortening of the C–S
Scheme 1 Synthetic routes to 3 and 4. (i) Pd(PPh3)2Cl2, CuI, NEt3, 1-
hexyne, DMF; (ii) diisopropyl xanthogen disulfide, AIBN, benzene; (iii)
Mg, THF, microwave, DMF; (iv) (Ph3P+CH2Br)Brꢁ, tBuOK, THF, ꢁ78
ꢂC; (v) [TMA][OH], NiCl2$6H2O, MeOH, THF, or NaOMe/MeOH
NiCl2$6H2O, TBABr; (vi) I2, MeCN, or [O].
Fig. 1 Single crystal X-ray data for 4 showing (upper) packing in a plan
view of the unit cell with close contacts between the molecules through
ꢀ
the indolyl pyrrole unit (C2) and the dithiolene S (3.930 A); (lower)
stacking of the dithiolene molecules down the a-axis with short inter-
ꢀ
planar distance of 3.822 A.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 15422–15430 | 15423

View Article Online
bonds, and a lengthening of the C]C bond, consistent with
a more delocalised aromatic p system over the NiSCCS ring.
No crystals suitable for XRD could be obtained for 3,
however, the aromatic nature of the dithiolene core could be
[TBA][4], however this might reflect the increased conforma-
tional freedom of the indolyl groups in [TBA][3] having a greater
effect on the frontier orbital energies.
Hybrid DFT calculations of the electronic structure of 4 were
reported previously. To support the experimental results of the
electronic absorption behaviour, this has been extended in this
work to include the TD-DFT calculated absorption spectra for 4
and 3 (Fig. 2). The calculations were carried out using a polar-
isable continuum model of DCM (3 ¼ 8.93).29 TD-DFT calcu-
lations showed excellent agreement with the experimentally
determined absorption spectra, and showed the low energy NIR
absorption (910 nm (10,989 cmꢁ1) and 840 nm (11,905 cmꢁ1) for 3
and 4, respectively) to be due to a mixed HOMO and HOMO-2
to LUMO transition. Since the HOMO-2 orbital was predomi-
nantly ligand based (pictures of relevant frontier orbitals, ener-
gies and contribution to the low-energy absorption are given in
Fig. S2, Table S2, ESI†), this suggests extensive indole contri-
bution to the electronic properties of the complex.30 The lower
energy NIR transition for 3 is suggested to be due to a more
extensive contribution from the HOMO-2 orbital, reflecting its
increased contribution to the frontier molecular orbitals, due to
its more in-plane conformation. For nickel dithiolenes, electron
donating groups raise the energy of the HOMO more than that
of the LUMO, resulting in a smaller HOMO/LUMO gap, and
thus a lower energy absorption in the NIR. Since –Bu is weakly
donating, through induction, compared to –H, this suggests that
the dominant effect of the –Bu group in 4 is to sterically disfavour
a more planar geometry, and thus reduce the extent of donation
through resonance from the indole group, compared to 3.
Cyclic voltammetry of 3 and 4 in DCM showed two reversible
one-electron processes (at E11/2 ¼ ꢁ0.76 V and ꢁ0.90 V, and E21/2
¼ +0.07 V and ꢁ0.03 V, for 3 and 4, respectively), as well as two
irreversible oxidation processes (at E3ox ¼ +0.92 V and +1.12 V,
and E4ox ¼ +1.46 V and +1.46 V, for 3 and 4, respectively) (Fig. 3;
Fig S3, Table S3, ESI†). The first two processes correspond to
sequential reduction of the neutral complex to a mono (E21/2) and
a dianion (E11/2), similar to previously reported neutral metal bis-
dithiolene complexes.19 The reduction processes of 3 occurred at
more positive potentials than for 4. This suggests that the LUMO
orbital, into which the electrons are added, is lower in energy for
1
observed by H NMR by the large shift to higher frequency for
the SCCS ring –H upon complexation, from 6.73 ppm in the
ligand precursor (1) to 9.74 ppm in the complex (3) (Fig. S1,
ESI†). The increase in the aromaticity of this hydrogen has led
some groups to investigate the reactivity of such dithiolene
complexes,24,27 and could be a possible future step for simple
tuning of the physical or electronic properties of 3 without the
need for a bottom-up synthetic approach.
UV/Vis/NIR spectroscopy of 3 and 4 in DCM show an intense
peak in the UV region, and a single peak in the NIR region [910
nm (10,989 cmꢁ1) and 840 nm (11,905 cmꢁ1) for 3 and 4 respec-
tively] (Fig. 2). Neutral nickel dithiolenes characteristically show
an intense NIR absorption due to a p / p* transition between
the HOMO and LUMO orbitals.12 For 3 this transition occurs at
ꢀ1,000 cmꢁ1 lower energy than for 4, indicating a smaller
HOMO/LUMO gap, which is in agreement with conclusions
drawn from electrochemical studies, and supported by DFT
calculations (vide infra). For the monoanionic TBA salts, the
NIR absorption is preserved as a transition is still possible
between the HOMO and the half filled LUMO.28 However, the
absorption is shifted to lower energy [960 nm (10,417 cmꢁ1) and
848 nm (11,792 cmꢁ1) for [TBA][3] and [TBA][4] respectively]. It
is not clear why the shift for [TBA][3] is 5 times greater than for
Fig. 2 Experimental and calculated UV/Vis/NIR absorbance spectra:
(upper) 3 in DCM; (lower) 4 in DCM.
Fig. 3 Cyclic voltammetry of 3 in 0.3 M TBABF4/DCM, starting at 0.5
V, with separate scans to negative and to positive potentials.
15424 | J. Mater. Chem., 2011, 21, 15422–15430
This journal is ª The Royal Society of Chemistry 2011

View Article Online
3 than 4. The first oxidation process (E3ox) occurs at less positive
potentials for 3 than 4, consistent with a greater donor influence
from a more in-plane indolyl group in 3 raising the energy of the
HOMO more than for 4. These results are in agreement with the
UV/Vis/NIR studies, which showed a NIR absorption at lower
energies for 3 than 4, consistent with a smaller HOMO/LUMO
separation in 3. Additionally, the decreased separation of the
reduction processes for 3 than 4 (DE ¼ +0.83 V and +0.87 V,
respectively) suggest that 3 is better able to stabilise two negative
charges developed on the complex, possibly due to increased
delocalisation onto the indolyl groups. Such a process is
consistent with a more in-plane conformation of the indolyl
groups, as suggested by calculations.
Therefore, it seems that the dithiolene radical, formed upon
oxidation, resides, as least partially, on the indolyl group, and
results in the oxidative coupling of the indolyl groups. Cycling of
poly-3 and poly-4 after transferring to monomer-free electrolyte
solution (of the same electrolyte and solvent, as for their
formation) showed two chemically reversible processes, corre-
sponding to the sequential reduction of the polymer films (Fig. 4
(middle, lower)), as well as an irreversible oxidation process at
a similar potential to the monomers (not shown). For both 3 and
4, films formed at faster scan rates (up to v ¼ 1.0 V sꢁ1) appeared
to show improved film redox behaviour. This was attributed to
a shorter residence time at strongly oxidising potentials.
For poly-3, both reduction processes were chemically revers-
ible (Fig. 4 (lower)), however, the second process diminished at
faster scan rates (v > 0.5 V sꢁ1). Additionally, the peak currents of
the two reduction processes were rather unbalanced, with the
reduction to the monoanion producing a larger current response
than the reduction to the dianion. Both these observations
suggest that a large steric barrier exists for the incorporation of
two molar equivalents of [TBA]+ into the film. However, both
reduction processes could clearly be observed, with the redox
processes for poly-3 at very similar potentials to 3 in solution,
suggesting that, contrary to previous results from thienyl con-
taining systems,31,32 the dithiolene redox processes were largely
unaffected by incorporation into a polymer film.
2.1 Electropolymerisation studies
For more concentrated solutions of 3 in DCM, cycling the
potential into the third (irreversible) process (between ꢁ0.3 and
+1.0 V) resulted in the deposition of a conducting film on the
electrode (Fig. 4 (upper)), analogous to the results previously
communicated for 4.13 The increase in peak current indicates the
increasing quantities of redox-active material deposited on the
electrode. The insolubility of the film in DCM, compared to
the monomer, suggested the film to be polymeric in nature.
For poly-4, the redox processes also appeared largely
unchanged upon incorporation into a polymer film (Fig. S5,
ESI†).13 However, the current response of the two reduction
processes appeared to be more balanced, suggesting improved ion
transfer through the film, compared to poly-3. The more balanced
reduction processes in poly-4 may be due to the formation of a less
dense polymer network, facilitating improved ion transport
through the film. This is possibly caused by the steric bulk of the
–Bu group on the dithiolene ring, disfavouring close packing of
the polymer chains. This is supported by the smaller peak currents
for poly-4, than for poly-3, as well as a visibly more coloured film
for poly-3, suggesting a smaller dithiolene content in the film of
poly-4. To further characterise the polymers, films were grown on
a fluorine-doped tin oxide (FTO) coated glass electrodes and were
investigated by UV/Vis/NIR spectroscopy (Fig. 5). For poly-3,
the absorption spectrum of the formed film showed a broadening
of the UV absorption. This is consistent with the linking of the
indolyl groups upon polymerisation, as an increase in conjugation
would shift the indole-based transitions to lower energy. The
absorption spectrum also showed the NIR absorption to be
shifted by 1,522 cmꢁ1 to a longer wavelength of 1055 nm. This shift
for poly-3 might be due to a number of factors: longer conjugated
chain length in the ligand might increase the delocalisation in the
polymer backbone, thus reducing the HOMO/LUMO separation;
increased planarity upon polymerisation might also raise the
HOMO level, resulting in a reduced HOMO/LUMO gap. A third
possibility might be that, due to the increased film density (or
thickness), some dithiolene species might remain as the mono-
anion in the film, resulting in a lower energy absorption being
observed (vide supra).
For both poly-3 and poly-4, the shift in the NIR absorption
was less than that observed for the absorption spectra of the
thiophene-substituted complex, poly-Ni(b-2ted)2,31 which
showed the NIR absorption to shift by 1896 cmꢁ1 further into the
Fig. 4 Electropolymerisation of 3 to form poly-3 (upper). Cyclic vol-
tammetry of poly-3 coated platinum electrode in monomer free electro-
lyte solution (lower), compared to the redox behaviour of the monomer
(middle).
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 15422–15430 | 15425

View Article Online
Fig. 6 Transfer characteristics of example bottom-gate, bottom-contact
(BGBC) organic field-effect transistor (OFET) fabricated from 4, with (a)
negative and (b) positive applied drain voltages (VD). The channel length
and width are in this case L ¼ 2 mm and W ¼ 10 mm respectively. Inset
(a): energy levels of 4 relative to Fermi level of gold. Inset (b) architecture
of BGBC OFET employed here.
Fig. 5 UV/Vis/NIR spectra of poly-3 grown by potential cycling on an
FTO coated glass electrodes, where the absorption intensity is given as
the molar extinction coefficient (3), and the polymer absorptions are
normalised relative to the NIR absorption of the monomer.
SCE).2,4,38 Thus the observation of stable n-channel, and ambi-
polar charge transport for 4, can be understood in terms of the
extensive delocalisation of the frontier orbitals leading to readily
accessible redoxprocesses, combined with the easeof formation of
a continuous film of 4 from solution. It is notable that 3 did not
form continuous films using similar solution processing methods,
emphasising the crucial role of molecular substituent groups in
determining utility in thin-film devices.
NIR upon polymerisation. It also contrasts with the electro-
polymerised dithiolenes based on a fused thiophene backbone
[Ni(b-2tted)2],32 for which broad redox processes, and a broad
absorption across the whole Vis/NIR range were observed,
arising from the extensive polyterthiophene component which, in
some cases, masked the dithiolene characteristics altogether.
2.3 Photovoltaic devices
2.2 Solution coated film and FET studies
Complex 4 shows strong absorption in the NIR, a region poorly
covered by other molecular semiconductors. Within the field of
organic photovoltaic devices, extending the absorption further
into the NIR region is an active area of interest as it would allow
for greater harvesting of the solar spectrum. Despite showing
a relatively low n-type mobility, the absorption properties of 4
are suited for incorporation into such devices.
Deposition of a thin film of 4 on a glass substrate by spin coating
led to facile formation of a continuous film, with a shift in the
NIR absorption to lower energy [lmax ¼ 915 nm (1.36 eV),
absorption onset ¼ 1323 nm (0.94 eV)]. This suggests some
degree of interaction in the solid state, required for good charge
transport in a FET. Bottom-gate, bottom-contact (BGBC)
organic field-effect transistors were fabricated with gold source
and drain electrodes. The Fermi level of gold is approximately
4.8 eV,33 and lies between the energy levels of the frontier orbitals
of 4 (estimated from electrochemical studies),34 giving a barrier
to injection of less than 1 eV for both charge carriers (see inset
Fig. 6 (a)). Such electrodes are therefore suitable for investigating
ambipolar charge transport behaviour in 4. BGBC OFETs were
prepared by spin-casting 4 from chloroform solution.
The structure of the devices prepared involved a binary bulk
heterojunction architecture, with 4 as the electron transport
material, and either P3HT or MDMO-PPV (hereafter PPV) as
the hole transport material, since the photophysics of both these
polymer systems is well documented.39 In the first instance the
optical properties of the blended films were investigated to ensure
both components had been incorporated into the film.
The UV/Vis absorption for both P3HT/4 and PPV/4 blends
showed incorporation of both species into the film, with the
dithiolene component enhancing the absorption in the UV, and
extending the absorption properties of the film to lower energies
by a discrete absorption in the NIR (Fig. S6, ESI†). Upon
blending, the extent of red-shifting of the dithiolene NIR
absorption (compared to the solution spectrum) was reduced for
both P3HT and PPV blends, when compared to spin coated films
of pure 4 (a red-shift of 113 and 278 cmꢁ1 for blends of P3HT and
PPV, respectively, compared to a red-shift of 990 cmꢁ1 for pure
4). This suggests less ordering in the dithiolene phase, due to the
intermolecular interactions of the dithiolene being broken up by
the polymer component.
Fig. 6 shows transfer characteristics (drain current ID, as
a function of gate voltage VG) of an example BGBC OFET using
4 as the active layer. The channel length and width were in this
case L ¼ 2 mm and W ¼ 10 mm respectively. Using standard
semiconductor models,35 the field-effect mobility of holes and
electrons were extracted from Fig. 6 (a) and (b) in the saturation
regime to be 10ꢁ7 cm2 Vꢁ1 sꢁ1 and 10ꢁ6 cm2 Vꢁ1 sꢁ1 respectively.
Whilst hole accumulation (p-type behaviour) has been readily
observed in many organic semiconductors, electron accumulation
(n-type) has been observed only rarely.36 One reason for this is the
poorer environmental stability of the reduced species upon
n-doping, leading to rapid carrier trapping at the air/semi-
conductor, or substrate/semiconductor interface.37 Several
studies have suggested the limit for ambient stability of n-type
semiconductors to be dictated by the oxidation processes with
atmospheric water, taking place at E ¼ ꢁ0.658 V (versus
Film thicknesses of between 60–130 nm were observed, with
thicker films formed at slower spin speeds. As devices formed
from P3HT have been shown to benefit from an annealing step,40
due to the reorganisation of the polymer to form more crystalline
15426 | J. Mater. Chem., 2011, 21, 15422–15430
This journal is ª The Royal Society of Chemistry 2011

View Article Online
domains, the effect of annealing on the UV/Vis absorption was
also investigated, however these showed negligible change in the
intensity or wavelength of maximum absorption (lmax) for the
polymer, and no change in the dithiolene absorption. No
significant reduction in film thickness (within experimental
error), was observed upon annealing, which was consistent with
negligible reorganisation taking place.
Photoluminescence (PL) measurements were performed on the
deposited films with emission wavelengths between 550 and 950
nm, and excitation at 500 nm. For P3HT, the incorporation of
dithiolene (at a ratio of 5 : 1 of P3HT : 4) resulted in a reduction
of the intensity of the emission by 95% (at lmax ¼ 718 nm). The
high degree of quenching observed, even at low dithiolene
concentration, suggests that efficient charge transfer had
occurred. Alternatively, due to the overlap of the P3HT emission
and dithiolene absorption bands, energy transfer cannot be dis-
counted. Further increases in dithiolene concentration (for a 2 : 1
and 1 : 1 P3HT : 4) resulted in less marked reductions in the
emission intensity, suggesting that, already at a 5 : 1 (P3HT :
dithiolene) ratio, the majority of excitons had been separated, or
transferred. For PPV, quenching of the emission at a 5 : 1
(PPV : 4) ratio was far greater, with quenching of over 99% (at
lmax ¼ 591 nm), again, with little increase in the quenching at
higher dithiolene loadings (Fig. S7, ESI†). It therefore seems that
even at low dithiolene loadings, efficient quenching of the poly-
mer excited state of both P3HT and PPV takes place, consistent
with good mixing of the materials upon blending, and efficient
electron or energy transfer between the polymer and the
dithiolene.
Fig. 7 External quantum efficiencies (EQE) for a 1 : 1 P3HT : 4 blend
(annealed and unannealed), as well as a 1 : 1 PPV : 4 blend (unannealed).
for the low EQE values is not known, with either poor electron
transport, or rapid charge recombination being the possible
reasons. The large structural variety of known Ni-dithiolene
complexes, and the observation of higher electron mobility in
other examples, suggests these limitations might be overcome in
future studies. As expected based on the low EQE values, the
devices yielded poor cell efficiencies, with the efficiency of the
best performing device (unannealed P3HT : 4) at <0.1%.
Photovoltaic devices were constructed by a standard proce-
dure, using 1 : 1 weight ratios of polymer : dithiolene, in chlo-
robenzene solution. The current response under device
illumination was measured, with diode-like behaviour for all
devices in the dark, suggesting that the devices were all oper-
ating correctly. From the external quantum efficiency (EQE) of
each device as a function of wavelength (Fig. 7), it is clear that,
for the P3HT devices, the dithiolene contributed to the overall
efficiency of the cell, due to a photocurrent generated from an
absorption centred around 860 nm, as well as a polymer
contribution in the visible range. These data are consistent with
charge transfer from 4 to P3HT, however, given the low EQE
values, it is unclear whether charge transfer occurs between
P3HT and 4, or whether energy transfer occurs, followed by
subsequent charge transfer of holes from 4 to P3HT. For the
P3HT devices, annealing increased the efficiency of both the
dithiolene and polymer contributions. For the PPV devices,
negligible photocurrent was generated by the dithiolene, sug-
gesting that no such charge transfer occurs at NIR wavelengths,
and the efficiency for the polymer component was far lower
than for the P3HT devices. The EQE was low for all regions of
the spectra, for all the devices tested (compared with >20% for
a reference device of P3HT:PCBM), with maximum external
quantum efficiencies for the best performing device of 0.15%
and 0.95% for the dithiolene and polymer absorptions,
respectively.
3. Conclusions
We have synthesised two analogous Ni-dithiolene complexes
containing indolyl moieties, differing in the presence or absence
of a short alkyl chain. The synthetic routes described for each are
applicable to a range of related ligand structures with a combi-
nation of an aryl and an alkyl (or H) unit attached to each end of
the dithiolene core. The X-ray structure of 4 showed stacked
metal complexes, which can potentially lead to intermolecular
interactions. Indeed upon film formation, intermolecular inter-
actions are evidenced by broadening of the NIR absorption,
which may arise due to similar stacking. Complex 3 was shown to
undergo similar electropolymerisation to that reported previ-
ously for 4, however a significantly more extensive NIR band
resulted which, along with poorer counterion penetration upon
reduction, suggests a denser film for poly-3, with enhanced
interactions between molecules. Extending the NIR absorption
to 1400 nm through molecular tuning is highly significant given
the technological importance of NIR dyes for applications such
as optical communications, and much previous synthetic Ni-
dithiolene work has sought to address this aim.12
Due to the excellent film-forming properties of the butyl-
substituted 4, field-effect transistors could be fabricated with the
best performing devices showing ambipolar behaviour, with
maximum charge carrier mobility of around 10ꢁ6 cm2 Vꢁ1 sꢁ1 for
electrons, and 10ꢁ7 cm2 Vꢁ1 sꢁ1 for holes. BHJ-PV devices were
fabricated using for the first time a nickel dithiolene (4) as both
an electron transport material and a NIR sensitiser. For binary
systems with P3HT, we demonstrated the contribution of
Whilst these values were very small, the concept of using
a dithiolene moiety as a NIR absorbing species for simultaneous
light harvesting and electron transport, was shown for the first
time to work in a bulk-heterojunction solar cell. The exact reason
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 15422–15430 | 15427

View Article Online
a dithiolene complex to the efficiency of the cell. Although at
present the overall cell efficiencies remain poor, future optimi-
sation of charge mobility or blend morphology could enable this
strategy to greatly enhance the NIR light harvesting in BHJ-PVs.
a saturated solution of NH4Cl (50 ml), and extracted with Et2O
(4 ꢃ 20 ml). The organics were combined and dried over MgSO4
and solvent removed under reduced pressure. The product was
purified by recrystallisation from hexane. (2.32 g, 97%). dH (250
MHz; CDCl3) 3.84 (s, 3H, -NMe), 6.65 (dd, 1H, J ¼ 3.1, 0.8 Hz,
Hb), 7.15 (d, 1H, J ¼ 3.1 Hz, Ha), 7.40 (d, 1H, J ¼ 8.5 Hz, He)
7.80 (dd, 1H, J ¼ 8.5, 1.5 Hz, Hd), 8.15 (dd, 1H, J ¼ 1.5, 0.8 Hz,
Hc), 10.03 (s, 1H, C(O)H) (Scheme 2).
4. Experimental
Electrochemical experiments were performed in dry DCM (stored
over KOH, distilled over P2O5), or dry MeCN (Sigma-Aldrich),
and solutions were degassed with N2 before use. TBABF4 was
used as supporting electrolyte, which was prepared from tetra-
fluoroboric acid and tetrabutyl ammonium hydroxide, and was
recrystallised 5 times from water and dried under vacuum at 60 ꢂC
for 2 weeks. The working electrode was a 3 mm2 Pt disc. The
reference electrode was Ag/AgCl (sat. KCl), calibrated at +0.55 V
against ferrocene/ferrocenium in the background electrolyte, and
the counter electrode was a Pt rod. Electrochemical polymerisa-
tion was performed at room temperature on a ca. 10 mM mono-
mer solution using the same supporting electrolyte and electrode
setup as for the electrochemical experiments.
Scheme 2 NMR assignment guide for 5-substituted N-methyl indoles
synthesised in this work.
5-Ethynyl-1-methylindole
UV/Vis/NIR measurements were recorded in solution using
a quartz cell of path length 10 mm, or as thin films on quartz or
FTO substrates fixed perpendicular to the beam on a Perkin-
Elmer Lambda 9 spectrophotometer, controlled by a datalink
PC, running UV/Winlab software.
To a stirred suspension of bromomethyltriphenylphosphonium
bromide (6.07 g, 13.9 mmol) in dry THF (40 ml) at ꢁ78 ꢂC was
added portion-wise potassium tert-butoxiꢂde (3.12 g, 27.8 mmol).
The reaction mixture was stirred at ꢁ78 C for 30 mins, during
which time the yellow ylide formed. A solution of 1-methylindole-
5-carboxldehyde (2.21g, 13.9 mmol) was added in THF (10 ml),
and the reaction was stirred for a further 30 mins at ꢁ78 ꢂC before
being warmed to room temperature overnight. After 12 h, the
reaction was shown to be complete by the complete consumption
of the aldehyde, observed by thin layer chromatography (TLC)
and the reaction quenched by addition of NH4Cl solution
(100 ml). The product was extracted with EtOAc (4 ꢃ 100 ml) and
the organics washed with brine and dried over MgSO4. The crude
product was purified by flash chromatography (Hex : EtOAc,
6 : 1), yielding the target material as a pale yellow oil (1.22 g, 57%).
dH (250 MHz; CDCl3) 3.00 (s, 1H, CCH), 3.76 (s, 3H, -NMe), 6.47
(dd, 1H, J ¼ 3.1, 0.8 Hz, Hb), 7.06 (d, 1H, J ¼ 3.1 Hz, Ha), 7.24 (d,
1H, J ¼ 8.5 Hz, He) 7.35 (dd, 1H, J ¼ 8.5, 1.5 Hz, Hd), 7.81 (dd, 1H,
J ¼ 1.5, 0.8 Hz, Hc).
Geometry optimisations for isolated molecular units of 3 and 4
were carried out using the Gaussian 03 program,41 utilising the
X-ray crystallographic coordinates of 4 for the starting structures
(modified for 3 using ArgusLab 4.0). The wavefunction was
expanded using the Pople 6-31G* basis set for all atoms,42,43
coupled to the Becke three parameters hybrid exchange and the
Perdew-Wang 1991 correlation functionals (B3PW91).44,45
Optimised structures were subsequently verified as minima on
the potential energy surfaces by the absence of negative values in
the frequency calculations. The molecular orbital isosurfaces
were generated using the cubegen utility in Gaussian 03 and
visualised using ArgusLab 4.0. TD-DFT calculations were
carried out in the presence of the Tomassi polarisable continuum
model (PCM) in a DCM solvation field,46 with the first 70 singlet
transitions calculated. Simulated spectra were generated using
Gausssum 2.1 freeware program, using a full-width half-
4-(1-Methyl-5-indol-5-yl)-[1,3]dithiol-2-one (1)
maximum value of 3000 cmꢁ1
.
All reagents were used as received, unless otherwise stated.
Complex 4, and pro-ligand 2 were synthesised as previously
reported.13
To a stirred solution of 5-ethynyl-1-methylindole (1.16 g, 7.47
mmol) in benzene (75 ml) was added AIBN (551 mg, 3.36 mmol)
and diisopropylxanthogen disulfide (2.22 g, 8.22 mmol), and the
reaction mixture stirred at reflux under a flow of nitrogen for 24
h. The solvent was removed under reduced pressure and the
crude product purified by trituration in warm methanol (6 h),
yielding pure product after filtration (1.26 g, 68%). dH (360 MHz;
CDCl3) 3.82 (s, 3H, -NMe), 6.52 (d, 1H, J ¼ 3.1 Hz, Hb), 6.73 (s,
1H, C(S)H), 7.11 (d, 1H, J ¼ 3.1 Hz, Ha), 7.30 (dd, 1H, J ¼ 8.5,
1.5 Hz, Hd) 7.34 (d, 1H, J ¼ 8.5 Hz, He), 7.67 (bd, 1H, Hc).
1-Methylindole-5-carboxaldehyde
To an oven dried microwave vial (25 ml), equipped with a Teflon-
coated stirrer bar, was added freshly recrystallised 5-bromo-1-
methylindole (3.15 g, 15.0 mmol) and Mg turnings (401 mg,
16.5 mmol) and the flask flushed with nitrogen before being
sealed using an aluminium open-top seal with a polytetrafluoro-
ethylene (PTFE) faced septum. Dry THF (15 ml) was charged via
syringe and the reaction mixture microwave-irradiated for 30
mins at 120 ꢂC, whereupon the flask was cooled to 0 ꢂC, and dry
DMF (1.16 ml, 15.0 mmol) was added via syringe. The reaction
[TBA][Ni(mi-5edt)2] ([TBA][3])
To a stirred solution of freshly cut sodium (117 mg, 5.10 mmol)
in dry degassed MeOH (35 ml) under nitrogen was added 1
(600 mg, 2.43 mmol), and the solution stirred with gentle
warming for 1 h until complete dissolution. [TBA]Br was added
ꢂ
mixture was stirred for 30 mins at 0 C before being warmed to
room temperature. The reaction mixture was poured into
15428 | J. Mater. Chem., 2011, 21, 15422–15430
This journal is ª The Royal Society of Chemistry 2011

View Article Online
3
ꢀ
and allowed to dissolve, whereafter a solution of NiCl2$6H2O
(289 mg, 1.22 mmol) in MeOH (25 ml) was added dropwise over
45 mins. The flask was opened to air and stirring was continued
for one hour before cooling in the freezer for one hour. The crude
product was filtered, and washed with cold MeOH and purified
by recrystallisation from DCM/EtOH, affording the title
compound as a brown precipitate (480 mg, 53%). Anal calcd for
C38H54N3S4Ni: C, 61.69; H, 7.36; N, 5.68; S, 17.34; found C,
61.73; H, 7.43; N, 5.57; S, 17.45, lmax ¼ 960 nm.
V ¼ 1103.74(10) A , Dc ¼ 1.488 Mg mꢁ3, m ¼ 4.168 mmꢁ1, no. of
reflections for cell ¼ 1937, Qmax ¼ 72.6537ꢂ, Z ¼ 4, 3096
reflections collected, unique [Rint] ¼ 2124 [0.0350], Tmin/Tmax
¼
0.3299/0.7900, parameters 146, R1 [F > 4s(F)] ¼ 0.0674, wR ¼
0.1751.
[Ni(mi-5hdt)2] (4). Dark green plate (dimensions 0.25 ꢃ 0.19 ꢃ
0.06 mm3) C30H34N2Ni1S4, T ¼ 100 K, space group P21/c, Mr ¼
ꢀ
ꢀ
609.59, monoclinic, a ¼ 4.31010(13) A, b ¼ 15.6910(4) A, c ¼
ꢂ
3
ꢀ
ꢀ
21.4795(7) A, b ¼ 91.526(3) , V ¼ 1452.14(8) A , Dc ¼ 1.394 Mg
mꢁ3, m ¼ 0.978 mmꢁ1, no. of reflections for cell ¼ 5598, Qmax
¼
[Ni(mi-5edt)2] (3)
28.478ꢂ, Z ¼ 2, 8971 reflections collected, unique [Rint] ¼ 3166
[0.046], Tmin/Tmax ¼ 0.62/0.94, parameters 169, R1 [F > 4s(F)] ¼
Route 1. Complex [TBA][3] (300 mg, 0.405 mmol) was dis-
solved in MeCN (20 ml) and I2 (206 mg, 0.810 mmol) was added
in MeCN (10 ml). A green precipitate formed immediately, and
was filtered and washed with EtOH. The crude product was
recrystallised from DCM/EtOH to yield 3 as a dark green
precipitate (121 mg, 60%). dH (500 MHz; CDCl3) 3.83 (s, 6H,
-NMe), 6.60 (d, 2H, J ¼ 3.1 Hz, Hb), 7.11 (d, 2H, J ¼ 3.1 Hz, Ha),
7.37 (d, 2H, J ¼ 8.7 Hz, He) 7.90 (d, 2H, J ¼ 8.7, 1.7 Hz, Hd), 8.33
(d, 2H, J ¼ 1.7 Hz, Hc), 9.75 (s, 2H, C(S)H).
ꢁ3
ꢁ3
ꢀ
ꢀ
0.0458, wR ¼ 0.1299, Drmax/e A ¼ 0.67, Drmin/e A ¼ ꢁ1.01.
Device fabrication and testing
For organic field-effect transistor (OFET) measurements,
a bottom-contact, bottom-gate architecture was employed. A
heavily doped silicon wafer served as both the substrate and
global gate electrode, with a 200 nm thermally grown SiO2 layer
(capacitance (Ci) of 17 nF cmꢁ2) as the gate dielectric. Gold
source and drain electrodes were defined using standard photo-
lithographic techniques. The SiO2 surface was then passivated
with a layer of hexamethyldisilane (HMDS). Material 4 was then
deposited onto the samples via spin-casting from solution, using
the solvent chloroform. Finally the OFETs were annealed at 50
ꢂC for 30 min to evaporate any residual solvent. A schematic
representation of the OFET structure is shown in the inset to
Fig. 6(b).
Route 2. To a stirred solution of 1 (594 mg, 2.40 mmol) in THF
(24 ml) was added [TMA][OH]$5H2O (958 mg, 5.29 mmol) in
MeOH (2 ml). After 5 mins, NiCl2$6H2O (286 mg, 0.515 mmol)
in MeOH (2 ml) was added and the reaction mixture stirred at
room temperature overnight. [TMA][3] was filtered off as
a brown precipitate. [TMA][3] was dissolved in MeCN (40 ml),
and I2 (974 mg, 3.84 mmol) was added in MeCN (20 ml). A green
precipitate formed immediately, and was filtered and washed
with EtOH. Recrystallisation in DCM/Et2O yielded pure 3 as
a dark green precipitate (736 mg, 63%). dH (500 MHz; CDCl3)
3.83 (s, 6H, -NMe), 6.60 (d, 2H, J ¼ 3.1 Hz, Hb), 7.11 (d, 2H, J ¼
3.1 Hz, Ha), 7.37 (d, 2H, J ¼ 8.7 Hz, He) 7.90 (d, 2H, J ¼ 8.7, 1.7
Hz, Hd), 8.33 (d, 2H, J ¼ 1.7 Hz, Hc), 9.75 (s, 2H, C(S)H). Anal
calcd for C22H18N2S4Ni: C, 53.13; H, 3.65; N, 5.63; found C,
53.06; H, 3.53; N, 5.56, lmax ¼ 910 nm (3 ¼ 25.6 ꢃ 103 Mꢁ1cmꢁ1).
For blended films of polymer/4, these were spin coated at
a range of speeds onto quartz substrates, and the UV/Vis spectra
measured. For P3HT, a 1 : 1 weight ratio of dithiolene to poly-
mer was dissolved in chlorobenzene (at 70 ꢂC), giving a total
dissolved content of 40 mg mlꢁ1. This blend was spin coated at
2000, 2500, 3000 and 3500 rpm. For PPV, due to its reduced
solubility in chlorobenzene compared to P3HT, a 1 : 1 weight
ratio of dithiolene to polymer, with a total dissolved content of
10 mg mlꢁ1, was used. The blend was spun at 1500, 2000, 2500
and 3000 rpm. The thickness of these films was analysed using
a Dektak surface profiler.
[TBA][Ni(mi-5hdt)2] ([TBA][4])
A solution of freshly cut sodium (68.5 mg, 2.98 mmol) in dry
degassed MeOH (20 ml) was added under nitrogen to 5 (430 mg,
1.42 mmol), and the solution stirred for 1 h until homogeneous.
[TBA]Br (914 mg, 2.84 mmol) was added and allowed to dissolve,
whereafter a solution of NiCl2$6H2O (168 mg, 0.709 mmol) in
MeOH (10 ml) was added dropwise over 45 mins. The flask was
opened to air and stirring was continued for one hour before
cooling in the freezer for one hour. The crude product was
filtered and washed with cold MeOH, and purified by recrystal-
lisation from DCM/EtOH, affording the title compound as
EQE measurements were performed using a 100 W tungsten
halogen lamp dispersed through a monochromator as the light
source and a Keithley 237 source meter unit (SMU) to measure
the short circuit current at various wavelengths. Incident light
intensity was continuously monitored during measurement by
a reference photodiode calibrated by a Hamamatsu S8746-01
calibrated photodiode.
For photovoltaic devices, the substrates used were ITO coated
glass, cleaned in an ultrasonic bath (10 mins IPA, 10 mins
acetone), and oxygen plasma etched (250 W, 10 mins) before use.
An anodic buffer layer of poly(3,4-ethylene-dioxythiophene):
poly(styrene sulfonate) (PEDOT:PSS)47 was first spin coated
(6000 rpm, 60 s), and annealed (250 ꢂC, 10 mins) under N2.
Under air free conditions, the active layers were then spin coated
at 2000 rpm for 60 s for both polymers, affording films of
thickness ꢀ100 nm. In order to assess whether an annealing step
resulted in better performance for the P3HT devices, half of the
P3HT substrates were annealed at 70 ꢂC for 10 mins. Aluminium
a
brown precipitate (297 mg, 49%). Anal calcd for
C46H70N3S4Ni: C, 64.84; H, 8.28; N, 4.93; found C, 64.96; H,
8.15; N, 4.84, lmax ¼ 848 nm.
X-Ray crystallography
4-(1-Methyl-5-indol-5yl)-[1,3]dithiol-2-one (1). Colourless
shard (dimensions 0.17 ꢃ 0.14 ꢃ 0.06 mm3) C12H9NOS2, T ¼
100 K, space group P21/c, Mr ¼ 247.32, monoclinic, a ¼ 6.6997
ꢂ
ꢀ
ꢀ
ꢀ
(3) A, b ¼ 7.3685(4) A, c ¼ 22.4087(12) A, b ¼ 93.852(5) ,
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 15422–15430 | 15429

View Article Online
23 J. M. Tunney, A. J. Blake, E. S. Davies, J. McMaster, C. Wilson and
C. D. Garner, Polyhedron, 2006, 25, 591–598.
24 A. Sugimori, N. Tachiya, M. Kajitani and T. Akiyama,
Organometallics, 1996, 15, 5664–5668.
25 C. A. S. Hill, A. Charlton, A. E. Underhill, K. M. A. Malik,
M. B. Hursthouse, A. I. Karaulov, S. N. Oliver and S. V. Kershaw,
J. Chem. Soc., Dalton Trans., 1995, 587–594.
cathodes were thermally evaporated through a shadow mask
defining 8 pixels per device of area 0.1 cm2. The electrodes were
attached, and devices were sealed with a glass cover slip by an
epoxy resin. All devices were made in duplicate, yielding 16
testable units for each device architecture.
26 S. Dalgleish and N. Robertson, Coord. Chem. Rev., 2010, 254, 1549–
1558.
Acknowledgements
27 M. Kajitani, G. Hagino, M. Tamada, T. Fujita, M. Sakurada,
T. Akiyama and A. Sugimori, J. Am. Chem. Soc., 1996, 118, 489–490.
28 R. K. Szilagyi, B. S. Lim, T. Glaser, R. H. Holm, B. Hedman,
K. O. Hodgson and E. I. Solomon, J. Am. Chem. Soc., 2003, 125,
9158–9169.
Fraser White (University of Edinburgh chemistry crystallog-
raphy service) is gratefully acknowledged for crystallographic
data collection. We thank the EPSRC for funding, including
collaboration within the EPSRC Supergen Excitonic Solar Cells
Consortium. This work has made use of the resources provided
by the EaStChem Research Computing Facility (http://www.
eastchem.ac.uk/rcf). This facility is partially supported by the
eDIKT initiative (http://www.edikt.org).
29 Gaussian, Online Manual, 2003.
30 G. Bruno, M. Almeida, F. Artizzu, J. C. Dias, M. L. Mercuri, L. Pilia,
C. Rovira, X. Ribas, A. Serpe and P. Deplano, Dalton Trans., 2010,
39, 4566–4574.
31 C. L. Kean and P. G. Pickup, Chem. Commun., 2001, 815–816.
32 C. Pozo-Gonzalo, R. Berridge, P. J. Skabara, E. Cerrada, M. Laguna,
S. J. Coles and M. B. Hursthouse, Chem. Commun., 2002, 2408–2409.
33 B. de Boer, A. Hadipour, M. M. Mandoc, T. van Woudenbergh and
P. W. M. Blom, Adv. Mater., 2005, 17, 621–625.
34 B. W. D’Andrade, S. Datta, S. R. Forrest, P. Djurovich,
E. Polikarpov and M. E. Thompson, Org. Electron., 2005, 6, 11–20.
35 J. Zaumseil and H. Sirringhaus, Chem. Rev., 2007, 107, 1296–1323.
36 L.-L. Chua, J. Zaumseil, J.-F. Chang, E. C.-W. Ou, P. K.-H. Ho,
H. Sirringhaus and R. H. Friend, Nature, 2005, 434, 194–
199.
References
1 C. D. Dimitrakopoulos and P. R. L. Malenfant, Adv. Mater., 2002,
14, 99–117.
2 H. Wada, T. Taguchi, B. Noda, T. Kambayashi, T. Mori,
K. Ishikawa and H. Takezoe, Org. Electron., 2007, 8, 759–766.
€
3 T. D. Anthopoulos, S. Setayesh, E. Smits, M. Colle, E. Cantatore,
37 A. Dodabalapur, Nature, 2005, 434, 151–152.
38 D. M. de Leeuw, M. M. J. Simenon, A. R. Brown and
R. E. F. Einerhand, Synth. Met., 1997, 87, 53–59.
39 N. C. Greenham, I. D. W. Samuel, G. R. Hayes, R. T. Philips,
Y. A. R. R. Kessener, S. C. Moratti, A. B. Holmes and
R. H. Friend, Chem. Phys. Lett., 1995, 241, 89–96.
B. de Boer, P. W. M. Blom and D. M. de Leeuw, Adv. Mater.,
2006, 18, 1900–1904.
4 T. D. Anthopoulos, G. C. Anyfantis, G. C. Papavassiliou and
D. M. de Leeuw, Appl. Phys. Lett., 2007, 90, 122105.
5 J.-Y. Cho, B. Domercq, S. C. Jones, J. Yu, X. Zhang, Z. An,
M. Bishop, S. Barlow, S. R. Marder and B. Kippelen, J. Mater.
Chem., 2007, 17, 2642–2647.
40 Y.-C. Huang, Y.-C. Liao, S.-S. Li, M.-C. Wu, C.-W. Chen and
W.-F. Su, Sol. Energy Mater. Sol. Cells, 2009, 93, 888–892.
41 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. 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. G. Johnson, W. Chen,
M. W. Wong, C. Gonzalez and J. A. Pople, GAUSSIAN 03
(Revision E.01), Gaussian, Inc., Wallingford, CT, 2004.
42 R. Ditchfield, W. J. Hehre and J. A. Pople, J. Chem. Phys., 1971, 54,
724–728.
6 A. C. Mayer, S. R. Scully, B. E. Hardin, M. W. Rowell and
M. D. McGehee, Mater. Today, 2007, 10, 28–33.
7 M. M. Wienk, J. M. Kroon, W. J. H. Verhees, J. Knol,
J. C. Hummelen, P. A. van Hal and R. A. J. Janssen, J. Am. Chem.
Soc., 2003, 42, 3371–3375.
8 J. Peet, A. B. Tamayo, X.-D. Dang, J. H. Seo and T.-Q. Nguyen,
Appl. Phys. Lett., 2008, 93, 163306–163308.
€
9 E. M. J. Johansson, A. Yartsev, H. Rensmo and V. Sundstrom,
J. Phys. Chem. C, 2009, 113, 3014–3020.
10 M. Koppe, H.-J. Egelhaaf, G. Dennler, M. C. Scharber, C. J. Brabec,
P. Schilinsky and C. N. Hoth, Adv. Funct. Mater., 2010, 20, 338–346.
11 Y. Ismail, T. Soga and T. Jimbo, Sol. Energy Mater. Sol. Cells, 2009,
93, 1582–1586.
12 U. T. Mueller-Westerhoff, B. Vance and D. I. Yoon, Tetrahedron,
1991, 47, 909–932.
13 S. Dalgleish and N. Robertson, Chem. Commun., 2009, 5826–5828.
14 Y. Gareau and A. Beauchemin, Heterocycles, 1998, 48, 2003–2017.
ꢁ
ꢁ
15 H. Henon, F. Anizon, R. M. Golsteyn, S. Leonce, R. Hofmann,
B. Pfeiffer and M. Prudhomme, Bioorg. Med. Chem., 2006, 14,
3825–3834.
16 I. Mutule and E. Suna, Tetrahedron, 2005, 61, 11168–11176.
17 M. Matsumoto and K. Kuroda, Tetrahedron Lett., 1980, 21, 4021–
4024.
18 T. Anjos, S. J. Roberts-Bleming, A. Charlton, N. Robertson,
A. R. Mount, S. J. Coles, M. B. Hursthouse, M. Kalaji and
P. J. Murphy, J. Mater. Chem., 2008, 18, 475–483.
19 U. T. Mueller-Westerhoff and B. Vance, Compr. Coordin. Chem.,
Pergamon, Oxford, 1987, vol. 2, pp. 595–631.
43 V. A. Rassolov, M. A. Ratner, J. A. Pople, P. C. Redfern and
L. A. Curtiss, J. Comput. Chem., 2001, 22, 976–984.
44 J. P. Perdew, J. A. Chevary, S. H. Vosko, K. A. Jackson,
M. R. Pederson, D. J. Singh and C. Fiolhais, Phys. Rev. B:
Condens. Matter, 1992, 46, 6671–6687.
45 J. P. Perdew, K. Burke and Y. Wang, Phys. Rev. B: Condens. Matter,
1996, 54, 16533–16539.
ꢁ
20 M. Formigue and J. N. Bertran, Chem. Commun., 2000, 2111–2112.
ꢂ
ꢂ
46 T. Cances, B. Mennucci and J. Tomasi, J. Chem. Phys., 1997, 107,
3032–3041.
ꢁ
21 O. Jeannin, J. Delaunay, F. Barriere and M. Formigue, Inorg. Chem.,
2005, 44, 9763–9770.
47 Y. Zhou, Y. Yuan, J. Lian, J. Zhang, H. Pang, L. Cao and X. Zhou,
Chem. Phys. Lett., 2006, 427, 394–398.
22 S. A. Baudron, N. Avarvari and P. Batail, Inorg. Chem., 2005, 44,
3380–3382.
15430 | J. Mater. Chem., 2011, 21, 15422–15430
This journal is ª The Royal Society of Chemistry 2011