Photoconductive Nile red cyclopalladated metallomesogens

Article abstract of DOI:10.1039/c2jm34946a

The synthesis and the electrochemical, photophysical and photoconductive properties of two new cyclometallated metallomesogens are described. These two complexes were respectively obtained by covalent metallation of Nile red, or a polyalkylated Nile red derivative, to a palladium(ii) ion and by using a polyalkylated Schiff base as an ancillary ligand to complete the coordination sphere of the metal centre. The introduction of long alkyl chains on either one or both ligands leads to disk-shaped metallomesogens with ordered room temperature columnar organisation induced, in both cases, by the formation of dimeric pairs. As shown by powder X-Ray diffraction studies, both complexes exhibit a columnar rectangular mesophase over a wide temperature range. These complexes show a redox solution behaviour characterised by one reversible single-electron reduction and two consecutive reversible oxidation waves, which give them an electrochemical amphoteric character. Energy levels and distributions of frontier orbitals have been estimated. Photoconductivity of the mesophase at room temperature has been observed from UV-Vis to near IR wavelength. The high thermal, electrochemical and mesophase stabilities of the studied Nile red complexes confirm that cyclometallated Pd(ii) complexes form a new class of photoconductors with features suitable for molecular semiconductor applications. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c2jm34946a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 23617
www.rsc.org/materials
PAPER
Photoconductive Nile red cyclopalladated metallomesogens†
ab
Andreea Ionescu,ab Nicolas Godbert,
and Mauro Ghediniabd
Alessandra Crispini,ac Roberto Termine,ad Attilio Golemmeabd
*
Received 25th July 2012, Accepted 20th September 2012
DOI: 10.1039/c2jm34946a
The synthesis and the electrochemical, photophysical and photoconductive properties of two new
cyclometallated metallomesogens are described. These two complexes were respectively obtained by
covalent metallation of Nile red, or a polyalkylated Nile red derivative, to a palladium(II) ion and by
using a polyalkylated Schiff base as an ancillary ligand to complete the coordination sphere of the metal
centre. The introduction of long alkyl chains on either one or both ligands leads to disk-shaped
metallomesogens with ordered room temperature columnar organisation induced, in both cases, by the
formation of dimeric pairs. As shown by powder X-Ray diffraction studies, both complexes exhibit a
columnar rectangular mesophase over a wide temperature range. These complexes show a redox
solution behaviour characterised by one reversible single-electron reduction and two consecutive
reversible oxidation waves, which give them an electrochemical amphoteric character. Energy levels
and distributions of frontier orbitals have been estimated. Photoconductivity of the mesophase at room
temperature has been observed from UV-Vis to near IR wavelength. The high thermal, electrochemical
and mesophase stabilities of the studied Nile red complexes confirm that cyclometallated Pd(II)
complexes form a new class of photoconductors with features suitable for molecular semiconductor
applications.
their columnar organisation which enhances charge mobility.6
Introduction
Indeed, along the column stacks of discotics, charge mobilities
several orders of magnitude higher than in other directions have
been measured, and a hole mobility as high as 1.4 cm2 Vꢀ1 sꢀ1
has recently been reported for tri-indole-based discotic liquid
crystals. This particularly high charge mobility was obtained by
reducing the stacking distance between molecules and preserving
a high intramolecular order along the columns.7
In recent years, a rising interest in low-cost, flexible and easy to
process electronic devices motivated the development of organic
electronic technologies.1 Applications of organic semiconducting
materials include photodetectors, sensors, thin film transistors,
lasers, and light emitting and photovoltaic devices.2 Semi-
conductors, where charge carrier generation is promoted in situ
by light absorption, are photoconductors.3 The photogeneration
process at the origin of photoconduction can be either extrinsic4
(when assisted by injection at electrodes or defect sites, or
induced by doping) or intrinsic5 (when a charge-pair state is
formed and after its dissociation the resulting holes/electrons can
drift through the bulk material). The best photoconductors
present high photogeneration efficiency as well as high charge
mobilities. Over the last decade, discotic liquid crystals have
received particular attention as semiconducting materials due to
For most organic compounds, the photoconduction is limited
by the low photogeneration efficiency. To overcome this limita-
tion, organic devices must often be sensitized by incorporation of
pigments such as phthalocyanines, squaraines, perylenes or azo-
pigment derivatives, or by means of a double layer architecture
where the photosensitizing layer is in contact with a transport
layer.3,8 Another approach, which is yet to be developed but
whose preliminary results are encouraging, is the use of photo-
conductive metallomesogens. Metallomesogens combine the
properties of transition-metal centres with the self-organisation
of the liquid crystalline state, hence high absorption properties at
higher wavelength are expected, in contrast to fully organic
liquid crystals. In this context we recently reported on a new class
of photoconductive cyclometallated complexes. These complexes
feature a 5-membered Pd(II) or Pt(II) metallocycle, obtained by
using a H(C^N) cyclometalled ligand such as azobenzene H(Az),
benzo[h]quinoline H(Bq) or 2-phenylpyridine H(Pp), and a 6-
membered metallocycle formed by a Schiff base H(O^N)
ancillary ligand. Some of these complexes have shown excellent
aCentro di Eccelenza CEMIF.CAL, LASCAMM CR-INSTM, Unitaꢀ
INSTM della Calabria, I-87036 Arcavacata di Rende, CS, Italy. E-mail:
nicolas.godbert@unical.it; Fax: +39 0984 492066; Tel: +39 0984 492881
bDipartimento di Chimica – Universitaꢀ della Calabria, I-87036 Arcavacata
di Rende, CS, Italy
cDipartimento di Scienze Farmaceutiche – Universitaꢀ della Calabria, I-
87036 Arcavacata di Rende, CS, Italy
dCNR-IPCF UOS di Cosenza c/o Dipartimento di Fisica, Universitaꢀ della
Calabria, I-87036 Arcavacata di Rende, CS, Italy
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c2jm34946a
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 23617–23626 | 23617

View Article Online
photorefractive properties either dispersed into a polymeric
matrix9 or in their bulk amorphous state10 and their excellent
photogeneration–photoconduction properties were recently
correlated with their chemical structure.11 In fact, the HOMO
and LUMO distributions are separated by the metal centre, with
the HOMO mainly localised onto the Schiff base moiety while
the LUMO is mainly distributed onto the cyclometallated ligand.
Moreover, theoretical computations have shown that the first
singlet and triplet excited states exhibit a distorted geometry
around the metal centre when compared to the ground state.
Such distortion may further delay the return to the ground state
of excited molecules, hence enhancing charge separation effi-
ciency, leading to the excellent photoconductive behaviour
observed for this new class of complexes (Chart 1).
As illustrated in Chart 2 (complexes I and II), by incorporating
long alkyl chains at the periphery of both ligands, discotic
mesophases were obtained.12 Furthermore, by changing the
chemical nature of the linker between the cyclometallated core
and the alkyl chains (ester for I vs. ether for II), it was possible to
tune the absorption properties of the resulting complexes and to
obtain cyclopalladated liquid crystal complexes photoconductive
at room temperature from UV-Vis to near IR (from 300 to
800 nm).12 Due to their high viscosity, low mesophase tempera-
ture range (from room temperature to ca. 50 ꢁC) and low degree
of order of the columnar organisation, the photoconduction of
these complexes could be, however, increased by only one order
of magnitude upon thermal annealing, when compared to the
same property in non-annealed samples.
Chart 2 Photoconductive discotics (I, II) and luminescent discotic (III).
liquid crystals, obtained by using either Nile red or a poly-
alkylated Nile red derivative as cyclopalladated ligands and a
polyalkylated Schiff base as an ancillary ligand. The incorpora-
tion of Nile red as a cyclometallated ligand has proven to
maintain the physical separation of the frontier orbitals on their
respective fragment, as shown by comparative cyclic voltamme-
try studies, leading to efficient photoconduction over a wide
wavelength window. The Nile red fragment also induces, via
formation of hydrogen bonded interactions through paired
molecules, a wide temperature range of liquid-crystallinity with a
high degree of order.
In general, in order to improve both the stability and the
degree of order of discotic mesophases, several approaches have
been developed and are commonly used, such as (i) increasing the
size of the discoid core,13 (ii) the introduction of secondary
interactions such as hydrogen bonds,14 and (iii) the generation of
helicoidal conformation that induce a higher degree of order.15
Taking advantage of such tools, a red luminescent cyclo-
palladated complex (Chart 2, complex III) exhibꢁiting a columnar
phase over a broad temperature range (35–173 C) was recently
obtained.16 This discotic mesogen was synthesized by cyclo-
palladation of the Nile red dye using a polyalkylated derivative
of curcumin as an ancillary ligand.
Results and discussion
Synthesis
The synthesis of both complexes IV and V (Scheme 1) was
achieved in two steps through the formation of their corre-
sponding palladium acetato-bridged intermediates (1 and 10
respectively) obtained by cyclometallation of either Nile red
H(NR) or the Nile red derivative H(NR)1 with palladium acetate,
as previously described for the cyclometallation of H(NR).17 The
bridge cleavage of 1 and 10 was performed using the Schiff base
In the present work we report the synthesis and character-
isation of two new room temperature photoconductive discotic
Scheme 1 Experimental conditions: (i) acetic acid, 55 ꢁC, 4 h; (ii)
Chart 1 Photoconductive cyclometallated complexes.
23618 | J. Mater. Chem., 2012, 22, 23617–23626
ethanol/dichloromethane, r.t., 5 days.
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Table 1 Thermodynamic data for all thermal transitions observed by
DSC for complex IV during 3 heating–cooling cycles (Colr: columnar
rectangular, I: isotropic liquid)
Transitional properties, T (ꢁC),
Complex
DSC scan
[DH in kJ molꢀ1
]
IV
1st heating
1st cooling
Waxy powder – 177 [72.6] – I
I – 139 [29.5] – Colr0 – 62 [11.0] –
Colr – 37 [3.6] – frozen
2
nd heating
Frozen – 33 [4.0] – Colr0 – 77 [11.6] –
Colr – 160 [24.4] – I
2
nd cooling
Identical data of the 1st cooling
Identical data of the 2nd cycle
3rd heating cycle
heating/cooling cycles are identical, indicative of the high
stability and reversibility of the mesophases. The mesophases of
complex IV, observed by POM upon slow cooling from the
isotropic phase, show typical birefringent textures of liquid
crystals but their identification was not possible for the absence
of perceptible specific motifs.
Scheme 2 Experimental conditions: (i) HCl, H2O, 0–5 ꢁC, 4 h; (ii) DMF,
Unequivocally, the mesophase determination was achieved by
powder X-ray diffraction studies (PXRD) at variable tempera-
tures. Both complexes IV and V present a columnar rectangular
organisation (Colr). The indexations of the PXRD patterns and
the cell parameters of complexes IV and V are presented in Table
2, together with the previously reported data of complexes I–III,
for direct comparison.
70 ꢁC, 8 h; (iii) K2CO3, KI cat., DMF, 130 ꢁC, 8 h.
H(O^N)1, prepared according to the previously reported
procedure.12
H(NR) and its hydroxy-derivative 2 (Scheme 2) were prepared
in two steps by slight modifications of previously reported
syntheses.18–20 Nitrosation of 3-diethylaminophenol gave a
mixture of 5-diethylamino-2-nitrosophenol 3a and its hydro-
chloride salt 3b in a 1 : 2 ratio, as determined from the 1H-NMR
spectrum of the mixture. Since both 3a20 and 3b19 have been
successfully used in cycling reactions for obtaining Nile red
derivatives, the mixture was used in the second step without any
further purification. H(NR) and 2 were then obtained by reaction
of 3a,b with naphthol and 1,7-dihydroxynaphthalene, respec-
tively. H(NR)1 was finally obtained by etherification of 2 with
3,4,5-tris(dodeciloxy)bromomethylbenzene 4.12
While the substituted azobenzene cyclometallated fragment
induced a hexagonal columnar mesophase for complexes I and II
(Chart 1) over a very short range of temperature (from r.t. to ca.
50 ꢁC), complexes III–V present a rectangular mesophase over a
ꢁ
wide range of temperature up to 160–170 C. Furthermore, the
cell parameters and hence the cross-sections (Sr) of their meso-
phases are much larger than the average radius of the molecules
ꢁ
ꢁ
(estimated at ca. 32 A for IV, ca. 53 A for V), as previously
observed also for complex III, for which it was hypothesised on
this basis the formation of a possible dimer constituting the
mesophase.16 To further verify this hypothesis, the number of
molecules by cell unit (z) for complexes II–V has been estimated
according to eqn (1),21
Complex (NR)Pd(BSC6), analogous to the photoconductive
complexes (C^N)Pd(BSC6) (Chart 1), has been synthesised with
the purpose of carrying out a complete electrochemical study and
of determining the influence of the NR fragment onto the frontier
orbitals of the resulting NR cyclometallated complexes. (NR)
Pd(BSC6) has been obtained through the rupture of the bridge
intermediate 1 by the Schiff base H(BSC6) following the pathway
previously described for complexes IV and V.
r N ASrh
M
Z ¼
(1)
where r is the density of the liquid crystal phase, NA is Avoga-
dro’s constant, Sr is the columnar cross-section, h is the height of
the columnar unit cell and M is the molecular mass of the liquid
crystal. Sr can be easily calculated from the cell parameters
keeping in mind that its expression is dependent on the cell’s
geometry (see Table 3). The height of the columnar unit cell can
be extracted from the PXRD pattern as the mean distance given
by the halo in the wide angle attributed to intracolumnar
stacking distance (h ¼ h0). For the liquid crystal density some
hypothesis has to be made. It is customary to consider a standard
density of 1 g cmꢀ3, a typical value for most organic phases.22
However, for discotic liquid crystals, density values are expected
to be slightly higher. As an example, density values of ca. 1.1 g
cmꢀ3 have been reported for columnar hexaalkyloxy-
triphenylenes23 and values up to 1.2 g cmꢀ3 were recently
Mesomorphic properties
At room temperature, the new cyclopalladated complexes IV and
V are dark blue waxy solids. Their thermal behaviour was
investigated by using polarized optical microscopy (POM),
thermogravimetry (TGA) and differential scanning calorimetry
(DSC). Both compounds, as observed by TGA, are thermally
stable until 180 ꢁC. While complex IV shows typical DSC
diagrams of liquid crystalline phases through repetitive scans of
heating–cooling cycles (Table 1, DSC traces are reported in the
ESI†), for complex V no thermal transitions were observed by
DSC, probably due to its high viscosity and a clearing point close
to its decomposition temperature. It is noteworthy that for
complex IV the DSC traces obtained on the third and successive
adopted for some potassium salts of columnar crown ethers.24
A
higher density than those encountered in common organic
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 23617–23626 | 23619

View Article Online
Table 2 Indexations of PXRD spectra of complexes I–V
Mesophase and
cell parametersa
Molecules number
in unit cell (z)
ꢁ
dobs (A)
ꢁ
dcalcd (A)
Complex
Temp.
r.t.
hk
Density r (g cmꢀ3
)
I12
32.83
19.32
4.47
31.07
18.67
16.24
12.24
4.36
10
11
hch
10
11
20
21
hch
h0
20
11
31
02
13
33
hch
h0
20
02
32
23
14
44
54
45
hch
h0
32.83
18.95
Colh, p6mm
ꢁ
N.C.b
0.98
ah ¼ 37.9 A
2
ꢁ
Sr ¼ 1245 A
II12
r.t.
31.07
17.93
15.53
11.74
—
Colh, p6mm
1.2
ꢁ
ah ¼ 35.9 A
2
ꢁ
Sr ¼ 1115 A
3.44
—
III16
140 ꢁ
C
46.46
38.57
24.53
21.78
14.07
12.83
4.56
46.46
38.57
25.01
21.20
13.97
12.86
—
Colr, c2mm
ꢁ
1.2
4.24
ar ¼ 92.9 A
ꢁ
On second heating cycle
br ¼ 42.4 A
2
ꢁ
Sr ¼ 3939 A
3.39
—
IV
125 ꢁC
32.52
18.62
14.14
11.42
9.17
8.03
7.66
6.70
4.27
32.52
18.62
14.14
11.61
9.22
8.09
7.60
6.77
—
Colr, p2gg
1.2
4.08
ꢁ
ar ¼ 65.0 A
ꢁ
br ¼ 37.2 A
On cooling from 160 ꢁC
ar/br ¼ O3
2
ꢁ
Sr ¼ 2421 A
3.85
—
V
r.t
On cooling from 160 ꢁC
35.81
21.19
14.36
12.53
7.28
4.42
4.03
36.74
14.42
20
02
13
33
55
hch
h0
35.81
21.19
13.85
12.15
7.29
—
Colr, c2mm
ar ¼ 71.62 A
1.2
3.84
ꢁ
ꢁ
br ¼ 42.38 A
2
ꢁ
Sr ¼ 3035 A
—
70 ꢁC
On second heating cycle
Highly disordered LC
a
b
Calculated with ah ¼ 2d10/O3 (Colh); and ar ¼ 2d20, br ¼ 2d02 (Colr). Not calculated in the absence of the h0 signal.
discotics is expected for complexes I–V, which contain a palla-
dium metal centre. For this reason, a density value of r ¼ 1.2 g
cmꢀ3 has been chosen for all complexes. Under these conditions,
the unit cell of the Colh mesophase of complex II turns out to be
constituted of only one molecule. For complex I the number of
molecules by cell unit cannot be determined, due to the higher
disorder of its mesophase, resulting in the absence of the halo peak
h0 in the wide angle region of its PXRD pattern. For all Nile red
metallomesogens, complexes III–V, Colr mesophases have been
obtained with a number of molecules by a unit cell of z ¼ 4.
Considering the estimated indexations of their PXRD patterns
(Table 2) together with the extinction rules,15c,25 the space groups
of the mesophases have been determined to be c2mm for III and
V and p2gg for IV. As evidenced in Table 3, for both these space
groups the number of diskoids present in the cross-section is
Zdisk ¼ 2, which unequivocally implies that each disk is formed
by two molecules, forming a dimer-like diskoid. In order to gain
a deeper insight into the possible structure of these dimers, we
have performed theoretical calculations on H(NR) molecules by
molecular mechanics (Experimental section). The energy-mini-
mised conformation of two drawn up H(NR) molecules
converged to the geometrically highly stable optimized structure
of a dimer presenting a 6-hydrogen-bond network, established
between two side-by-side molecules placed in a head-to-tail
arrangement, which involves oxygen atoms and aromatic
hydrogen atoms of the H(NR) molecules, as well as hydrogen
atoms of the amino-alkyl chains, as shown in Fig. 1a. Conse-
quently, the mesophases of the new complexes IV and V are
constituted of full-disk shaped dimer structures, as represented in
Fig. 1b. The formation of this dimer structure can explain the
high thermal stability of the Colr mesophase, which is maintained
over a wide temperature range, as well as its higher degree of
order when compared to the Colh mesophases formed by
complexes I and II (Chart 2).
Another important feature evidenced by the PXRD pattern is
the higher order of the columnar mesophase of complex IV even
at high temperatures, when compared with the rather disordered
mesophase displayed by complex V, for which the indexation of
the PXRD pattern was only possible at room temperature, in a
frozen state. This drastic increment of disorder can easily be
attributed to the additional alkyl-chains on the NR fragment,
which probably sterically hinder the p–p stacking of adjacent
disks that would promote a long-range order within the columns.
It is noteworthy that this steric hindrance does not prevent the
formation of the highly stable dimeric form of the complex
through the hydrogen bonded network.
23620 | J. Mater. Chem., 2012, 22, 23617–23626
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Table 3 Lattice structures of hexagonal and rectangular columnar liquid crystals with the corresponding cross-section area and the number of diskoids
Zdisk (disk or ellipsoid) in the cross-section
No. of diskoids
in cross-section
Type
Notation
Colh
Space group
Schema
Cross-section
pffiffiffi
3
Columnar hexagonal
p6mm
Zdisk ¼ 1
2
Sh ¼ ah
2
p2mm
c2mm
Zdisk ¼ 1
Zdisk ¼ 2
p2gg
Zdisk ¼ 2
Columnar rectangular
Colr
Sr ¼ arbr
p2mg
Zdisk ¼ 4
with the data of parent complexes (C^N)Pd(BSC6), C^N ¼ Az,
Bq, Pp (Chart 1) for direct comparisons.
Within the (C^N)Pd(BSC6) series, it can easily be noticed that
regardless of the nature of the cyclometallated ligand (H(Az),
H(Bq), H(Pp) or H(NR)), the HOMO energy value is maintained
at ca. 5.4 eV. In contrast, the LUMO energy is highly dependent
on the cyclometallated ligand, varying from 2.6 to 3.6 eV. This
observation is consistent with the separation between HOMO
and LUMO, the HOMO being localised onto the Schiff base
fragment while the LUMO is mainly localised onto the cyclo-
metallated fragment, as previously described and confirmed in
the case of complexes (Az)Pd(O^N), (Bq)Pd(O^N) and (Pp)
Pd(O^N) by theoretical calculations.11 The incorporation of Nile
red as a cyclometallated ligand therefore seems to maintain this
separation. The cyclic voltammograms of complexes IV and V
are reported in Fig. 2. Both complexes show a one electron
reduction wave and two consecutive one-electron oxidation
waves. Complexes IV and V show an identical reduction poten-
tial as their analogue (NR)Pd(BSC6) but present a lower oxida-
tion potential which is slightly decreased (by ca. 300 mV). This
suggests that the addition of supplementary alkyl chains onto the
Schiff base ligand allows a fine tuning of the HOMO energy level,
lowering it, while the introduction of alkyl chains through an
ether linkage onto the NR fragment does not modify the LUMO
energy. Furthermore, while the first oxidation for complexes IV
and V occurs at the same potential (+300 mV vs. Fc/Fc+),
the second oxidation for complex V occurs at a lower
potential, reducing the gap between the two oxidation waves
Fig. 1 (a) Geometrically optimised structure of H(NR) dimer. Hydrogen
ꢁ
bonds are represented by dashed lines (from 2.33 to 2.37 A). (b) Scheme
of resulting ellipsoid dimers IV and V.
Solution electrochemistry
Solution electrochemical data for complexes IV and V, obtained
by cyclic voltammetry (CV), are collected in Table 4, together
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 23617–23626 | 23621

View Article Online
Table 4 Cyclic voltammetry dataa
Compound
E
red (V) vs. Fc/Fc+
E
ox (V) vs. Fc/Fc
Estimated LUMO (eV)
Estimated HOMO (eV)
(Az)Pd(BSC6)11
(Bq)Pd(BSC6)11
(Pp)Pd(BSC6)11
(NR)Pd(BSC6)
IV
ꢀ1.48 (Rev.)
ꢀ2.23 (Irr.)
ꢀ2.43 (Irr.)
ꢀ1.20 (Rev.)
ꢀ1.20 (Rev.)
ꢀ1.20 (Rev.)
+0.78 (Irr.)
+0.57 (Irr.)
+0.61 (Irr.)
+0.60 (Irr.)
+0.30 (Rev.)
+0.30 (Rev.)
ꢀ3.3
ꢀ2.6
ꢀ2.4
ꢀ3.6
ꢀ3.6
ꢀ3.6
ꢀ5.6
ꢀ5.4
ꢀ5.4
ꢀ5.4
ꢀ5.1
ꢀ5.1
V
a
Potentials are given versus ferrocene/ferrocenium (Fc/Fc+) (Rev.: reversible wave, Irr.: irreversible wave).
Table 5 Absorption maxima recorded in dichloromethane solution at
room temperature
Complex l/nm (log 3)
IV
V
259 (3.79), 306 (3.45), 334*, 373*, 437 (3.01), 570*, 608 (3.59)
254 (3.81), 302*, 352 (3.39), 377*, 573*, 609 (3.44)
irradiation is shown in Fig. 3 for complex IV, where the light was
turned on at t ¼ 2.5 s and turned off at t ¼ 17.5 s.
Fig. 4 shows the photoconductivity of complexes IV and V
versus the applied electric field E, normalized over the absorption
coefficient a and the light intensity I. For complex V, it was not
Fig. 3 Current measured on a sample of IV exposed to light at l ¼
Fig. 2 Cyclic voltammograms of complexes IV (a) and V (b) in the
presence of ca. 1 eq. of ferrocene as an internal reference (Fc/Fc+). Scan
rate: 100 mV sꢀ1, DCM, Bu4NPF6 (0.1 M).
560 nm under an applied electric field E ¼ 23 V mmꢀ1
.
ox
ox
ox
ox
(E2 ꢀ E1 ¼ 300 mV for IV, E2 ꢀ E1 ¼ 200 mV for V). This
oxidation behaviour suggests that the HOMO + 1 might be
localised onto the NR fragment and that its energy is stabilised by
the ether-linked alkyl chains. It is noteworthy that all redox
processes for complexes IV and V are fully reversible, indicative
of their high electrochemical stabilities.
UV/Vis spectroscopy and photoconduction
UV-Vis spectra of metallomesogens IV and V were obtained in
dichloromethane solution, and data are reported in Table 5. The
absorption properties of both complexes IV and V are spread on
a wide wavelength window. In contrast to complex III,26 no
emission was observed, and both complexes do not show the
solvatochromic properties of H(NR).27
Fig. 4 Photoconductivities normalized over the absorption coefficient a
and the light intensity I are reported for IV and V. The experiments were
performed at l ¼ 520 nm and I ¼ 5 ꢂ 10ꢀ3 W cmꢀ2 for IV and at l ¼
633 nm and I ¼ 2 W cmꢀ2 for V, while the absorption coefficients were
5213 cmꢀ1 for IV and 112 cmꢀ1 for V. The inset shows the photocon-
ductivity of IV normalized over light intensity only.
Both complexes IV and V, however, show photoconductive
properties, i.e. an increase in electrical conductivity when
exposed to light.
A typical current increase under light
23622 | J. Mater. Chem., 2012, 22, 23617–23626
This journal is ª The Royal Society of Chemistry 2012

View Article Online
possible to acquire data with an applied field higher than 7–8 V
mmꢀ1 because of dielectric breakdown at higher fields. It is clear
that the photoconductivity of IV is about 1 order of magnitude
higher than in V. This difference could be explained by consid-
ꢁ
ering that the stacking distance of IV is shorter than in V (3.85 A
ꢁ
vs. 4.03 A). In addition, and perhaps more importantly, IV
exhibits a higher long range order within the columns. It is well
known that in discotic liquid crystals both a long range order and
a short stacking distance increase the p-orbitals overlap between
neighbouring molecules in the columns, favouring one-direc-
tional charge transport. This can increase charge mobility and
photoconductivity even by several orders of magnitude.28
The data reported in the inset of Fig. 4 show that the s/I value
for IV is in the order of 3 ꢂ 10ꢀ8 S cm Wꢀ1 at 25 V mmꢀ1, which
is a very high value when compared to the photoconductivity of
other organic photoconductors,29 especially because this result is
obtained without addition of dopants and without any annealing
process to increase the order. This result represents a tremendous
step forward in the development of this class of photoconductors
since in previously analyzed cyclopalladated complexes, at the
same applied electric field, the s/I value was of the order of 10ꢀ12
S cm Wꢀ1 for annealed cyclopalladated discotics (complexes I
and II),12 while it was of ca. 10ꢀ9 S cm Wꢀ1 for crystalline
samples of the (C^N)Pd(O^N) series reported in Chart 1.11
The wavelength dependence of the normalized photocurrent
and of the absorption coefficient of complex IV is reported in
Fig. 5. For photoconduction measurements, a particular cell
geometry (Experimental section) has been used, allowing the
observation of the photocurrent even for high values of
absorption coefficients and for wavelengths well below the
absorption limit of the ITO covered glass substrates. The
observed photocurrent is proportional to the absorption of
the material. In particular, it is noteworthy that how even a little
decrease in the absorption between 300 nm and 330 nm results in
a corresponding decrease in the photocurrent. Such behaviour
was also observed for complex V. These new complexes show,
therefore, photoconductive properties on a quite wide range of
wavelength from below 200 nm up to 800 nm.
Fig. 6 Photocurrent versus light intensity for IV. Measurements were
performed at l ¼ 520 nm with an applied field E ¼ 16 V mmꢀ1
.
intensity. This can be explained by assuming that the carrier
kinetics is controlled by a bi-molecular recombination process.30
The extended absorption window coupled with their photo-
conductive properties render such Nile red metallomesogens
potential candidates for many optoelectronic applications,
particularly in photovoltaics where such features are important,
a field in which organic discotics have recently proven to boost
the power conversion efficiencies of devices.31
Conclusions
Two Nile red H(NR) cyclopalladated complexes exhibiting a
columnar rectangular mesophase in a wide temperature range
were synthesized. The mesophases were obtained by the self-
assembly of ellipsoidal dimeric pairs kept together by a hydrogen
bonding network. This particular dimer interaction lends a high
stability to the mesophase and, in the case of complex IV, also a
high order. The additional introduction of alkyl chains onto the
NR fragment (complex V) does not prevent the formation of
these dimeric pairs but dramatically increases the columnar
disorder by steric hindrance. Both metallomesogens have shown
an amphoteric redox behavior with fully reversible oxidation and
reduction processes, showing their high electrochemical stabili-
ties. Through the electrochemical estimation of the energy levels
of their frontier orbitals, we have shown that HOMO and
LUMO are physically separated by the palladium metal centre,
the HOMO being mainly distributed onto the Schiff base ancil-
lary ligand while the LUMO is mainly localized onto the cyclo-
metallated Nile red fragment. A band-gap of ca. 1.5 eV imparts
to these complexes a wide range window of absorption from the
UV-Vis up to the near IR region, while the HOMO/LUMO
physical separation probably originates their excellent photo-
conduction properties. Photoconductivity measurements, per-
formed with a cell geometry that allows the observation of
photocurrent even at low wavelengths, have shown the correla-
tion between the absorption of the complexes and its photocur-
Photoconductivity was also measured as a function of the
intensity of light. As shown in Fig. 6, it can clearly be deduced
that the photocurrent varies with the square root of the light
rent, with normalized photocurrent values reaching 3 ꢂ 10ꢀ8
S
cm Wꢀ1 at an applied electric field E ¼ 25 V mmꢀ1. All together,
the features displayed by these new metallomesogens constitute
highly significant improvements towards the use of such
complexes in optoelectronic devices.
Fig. 5 Normalized photocurrent and absorption, measured on film,
versus the wavelength for compound IV. Photoconductivity measure-
ments were performed with an applied field E ¼ 16 V mmꢀ1
.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 23617–23626 | 23623

View Article Online
Synthesis of H(NR). 1-Hydroxynaphthalene (1.00 g, 6.9 mmol)
and the mixture 3a,b (1.5 g, ca. 6.9 mmol) were dissolved in ca.
40 mL of dimethylformamide and the resulting solution heated at
70 ꢁC for 8 h. The solvent was evaporated under reduced pressure
and the crude solid purified by column chromatography (silica
gel, dichloromethane/diethylether: 2/1).
H(NR): dark red solid; yielꢁd 23% (505 mg); m. p. 211 ꢁC; 1H
NMR (300 MHz, CDCl3, 25 C, TMS): d ¼ 8.66 (d, J ¼ 7.69,
1H), 8.32 (d, J ¼ 7.26, 1H), 7.74–7.59 (m, 3H), 6.66 (dd, J ¼ 5.5
Hz, J ¼ 2.15 Hz, 1H), 6.46–6.44 (m, 1H), 6.38 (s, 1H), 3.46 (q, J ¼
6.1 Hz, 4H), 1.28 (t, J ¼ 11.9 Hz, 6H); FT-IR (KBr, cmꢀ1): 2964,
2935, 1624, 1584, 1555, 1403, 1306, 1255, 1111, 1020, 843, 778;
elemental analysis: calcd (%) for C20H18N2O2 (318.14): C 75.45,
H 5.70, N 8.80; found C 75.42, H 5.73, N 8.80.
Experimental
General methods and procedures
All commercially available chemicals were purchased from
Aldrich Chemical Co. and were used without further purifica-
tion. I.R. spectra (KBr pellets) were recorded on a Perkin-Elmer
Spectrum One FT-IR spectrometer. 1H-NMR spectra were
recorded on a Bruker WH-300 spectrometer in deuterated
solvents with TMS as an internal standard. Elemental analyses
were performed with a Perkin-Elmer 2400 analyzer CHNS/O.
The powder X-Ray diffraction patterns were obtained by using a
Bruker AXS General Area Detector Diffraction System (D8
Discover with GADDS) with Cu-Ka radiation; the area detector
was placed at a distance of 20 cm from the sample and at an angle
2qD of 14ꢁ. Measurements were performed by charging samples
in Lindemann capillary tubes with inner diameters of 0.5 mm. A
CalCTec (Italy)ꢀh1eating stage was used to heat the samples at a
rate of 5ꢁC min to the appropriate temperature.
Synthesis of 2. 1,7-Dihydroxynaphthalene (1.243 g, 7.5 mmol)
and the mixture of 3a,b (1.580 g, ca. 7.1 mmol) were solubilized in
ca. 40 mL of dimethylformamide and the resulting solution
heated at 70 ꢁC for 8 h. The solvent was evaporated under
reduced pressure and the crude solid purified by column chro-
matography (silica gel, ethyl acetate/ethanol: 1/1).
Electrochemical studies
2: green solid; yield 53% (1.327 g); m. p. 250 ꢁC; 1H NMR (300
MHz, DMSO, 25 ꢁC, TMS): d ¼ 10.38 (s, 1H), 8.57 (d, J ¼ 9 Hz,
1H), 8.40 (d, J ¼ 8.79 Hz, 1H), 7.43 (d, J ¼ 2.58 Hz, 1H), 7.21 (d,
J ¼ 2.55 Hz, 1H), 7.18 (d, J ¼ 2.58 Hz, 1H), 6.51 (d, J ¼ 2.8 Hz,
1H), 6.02 (s, 1H), 3.44 (q, J ¼ 7 Hz, 4H), 1.28 (t, J ¼ 7.2 Hz, 6H);
FT-IR (KBr, cmꢀ1): 3145 (br), 2971, 2920, 1635, 1584, 1559,
1328, 1114, 832, 518; elemental analysis: calcd (%) for
C20H18N2O3 (334.13): C 71.84, H 5.43, N 8.38; found C 71.85, H
5.43, N 8.37.
All potentials were measured with IR compensation using an
Epsilon electrochemical analyser. Voltammetry experiments
were performed in a 3 mL cell of dry, freshly distilled, and
degassed (N2) dichloromethane solution using tetrabutylammo-
nium hexafluorophosphate (0.1 M) as a supporting electrolyte, a
Pt disk working electrode, a Pt wire counter-electrode and an Ag
wire as a pseudo-reference electrode. Voltammograms were
recorded at a 100 mV sꢀ1 scan rate from ca. 10ꢀ3 M complex
solution. Redox potentials are given relative to a ferrocene/
ferrocenium (Fc/Fc+) redox couple used as an internal reference.
Estimation of HOMO/LUMO energy values was performed
taking into account ꢀ4.8 eV for Fc/Fc+.32
Synthesis of H(NR)1. 2 (276 mg, 0.83 mmol) was solubilized in
30 mL of DMF. To the obtained solution were added 1.5 eq. of
K2CO3 (171 mg, 1.25 mmol) and a catalytic amount of KI. The
reaction mixture was stirred for 10 min and 1 eq. of 4 (598 mg,
0.83 mmol) was added. The solution was heated at 130 ꢁC for 8 h.
The solvent was removed under reduced pressure and the
product purified by column chromatography (silica gel, petro-
Synthesis
5-Diethylamino-2-nitrosophenol (3a) and 5-diethylamino-2-
nitrosophenol hydrochloride (3b). 3-Diethylaminophenol (12.02 g,
72.6 mmol) was dissolved in a mixture of 26 mL of HCl and 16
mL of distilled water. The obtained solution was cooled to 0 ꢁC
under stirring and an aqueous solution of NaNO2 (5.01 g, 72.6
mmol) was added. The reaction mixture was stirred for 4 h at 0–
5 ꢁC. The precipitated solid was filtered, dried under vacuum and
dissolved in hot ethanol (200 mL). After addition of diethylether
(200 mL) and 24 h in the freezer, a mixture of 3a and 3b (in 1 : 2
molar ratio) was obtained as a brown-yellow crystalline powder
(11.79 g), identified by comparison with NMR samples of pure
compounds.19,20 Since either compounds have been already
successfully used in cycling reactions for obtaining Nile red
derivatives,19,20 the product was used without any further
purification.
3a + 3b: 1H NMR (300 MHz, DMSO, 25 ꢁC, TMS): d ¼ 7.53
(d, J ¼ 10.65 Hz, 1H, 3b), 7.26 (d, J ¼ 10.3 Hz, 0.5H, 3a), 7.19
(dd, J ¼ 9.18 Hz, J ¼ 2.2 Hz, 1H, 3b), 6.83 (dd, J ¼ 10.1 Hz, J ¼
2.4 Hz, 0.5H, 3a), 6.54 (d, J ¼ 2.6 Hz, 1H, 3b), 5.7 (d, J ¼ 2.6 Hz,
0.5H, 3a), 3.7–3.3 (m, 6H, 3a + 3b), 1.25 (t, J ¼ 7.3 Hz, 6H, 3b),
1.16 (t, J ¼ 7.4 Hz, 3H, 3a); FT-IR (KBr, cmꢀ1): 3424 (br), 2979,
2755, 1626, 1545, 1491, 1146, 1040, 857, 749.
ꢁ
leum ether (b.p. 40–60 C)/diethylether: 1/1).
H(NR)1: red solid; yꢁield 53% (428 mg); m.p. 113 ꢁC; 1H NMR
(300 MHz, CDCl3, 25 C, TMS): d ¼ 8.60 (d, J ¼ 8.8 Hz, 1H),
7.85 (d, J ¼ 2.5 Hz, 1H), 7.62 (d, J ¼ 9.1 Hz, 1H), 7.36 (dd, J ¼
8.8 Hz, J ¼ 2.55 Hz, 1H), 6.71 (d, J ¼ 2.55 Hz, 1H), 6.67 (d, J ¼
4.4 Hz, 2H), 6.51 (d, J ¼ 2.6 Hz, 1H), 6.42 (s, 1H), 5.11 (s, 2H),
3.97 (m, 6H), 3.48 (q, J ¼ 7 Hz, 4H), 1.82–1.47 (m, 60H), 1.20 (m,
6H), 0.9–0.85 (m, 9H); FT-IR (KBr, cmꢀ1): 2928, 2848, 1624,
1588, 1559, 1332, 1259, 1122, 844; elemental analysis: calcd (%)
for C63H96N2O6 (976.73): C 77.41, H 9. 09, N 2.87; found C
77.40, H 9.91, N 2.90.
Synthesis of acetato-bridged intermediate complexes 10. The
intermediate complex 10 was synthesized according to the
previously reported procedure for 1 (ref. 16) with the following
modifications: H(NR)1 (187 mg, 0.2 mmol) was solubilized in
60 mL of acetic acid and 10 mL of dichloromethane. The
resulting solution was heated at 55 ꢁC for 6 days and filtered on
Celite. The obtained solid was washed with acetonitrile.
10: dark-blue solid; yield 75% (161 mg); m.p. 154 ꢁC; 1H NMR
(300 MHz, CDCl3, 25 ꢁC, TMS): d ¼ 7.74 (d, J ¼ 9.5 Hz, 1H), 7.0
23624 | J. Mater. Chem., 2012, 22, 23617–23626
This journal is ª The Royal Society of Chemistry 2012

View Article Online
(d, J ¼ 2.5 Hz, 1H), 6.50 (s, 2H), 6.40 (d, J ¼ 1.8 Hz, 1H), 6.35
(dd, J ¼ 9.5 Hz, J ¼ 2.9 Hz, 1H), 6.21 (d, J ¼ 2.2 Hz, 1H), 6.03 (s,
1H), 4.69 (s, 2H), 4.1–3.8 (m, 6H), 3.3–3.1 (m, 4H), 2.31 (s, 3H),
1.9–1.6 (m, 6H), 1.5–1.4 (m, 6H), 1.4–1.1 (m, 54H), 0.9–0.8 (m,
9H); FT-IR (KBr, cmꢀ1): 2920, 2853, 1636, 1575, 1234, 1117,
1075, 720; elemental analysis: calcd (%) for C130H198N4O16Pd2
(2283.29): C 68.31, H 8.73, N 2.45 found C 68.34, H 8.70, N 2.45.
Fig. 7 Side view of a sample (not in scale).
stripes. Patterned glasses were again covered with positive
photoresist and each ITO stripe was exposed to light through a
2 mm long slit with an opening of 5 mm or 10 mm. By choosing the
right process parameters, it was possible to remove the ITO from
the glass in the exposed region, obtaining two ITO electrodes
with sharp edges, as shown in Fig. 7. The span between the
electrodes can be modulated by changing the slit width and the
exposition time. Electrode separations in the 5–40 mm range were
obtained. The last step of the process was the spin-coating of the
material, dissolved in chloroform with a typical concentration of
10 mg mLꢀ1. The thickness of the materials and the spacing
between the electrodes were measured by using a Dektak 8
profilometer from Veeco.
Synthesis of complex IV. Intermediate 1 (100 mg, 0.10 mmol)
and 1.5 eq. of H(O^N)1 (ref. 12) (190 mg, 0.15 mmol) were
solubilized in dichloromethane and ethanol and the reaction
mixture was stirred at room temperature for 5 d. The solvent was
removed under reduced pressure, and the resulting solid was
washed with acetonitrile and precipitated from ethanol/
dichloromethane, filtered and reprecipitated from diethylether/
dichloromethane, filtered, washed with diethylether and with hot
ethanol to yield a dark-blue solid.
1
IV: dark-blue waxy solid; yield 83% (211 mg); H NMR (300
MHz, CDCl3, 25 ꢁC, TMS): d ¼ 9.35 (d, J ¼ 9.51 Hz, 1H), 7.98 (s,
1H), 7.66 (d, J ¼ 4.65 Hz, 1H), 7.41–7.39 (m, 1H), 7.38 (d, J ¼
8.07 Hz, 1H), 6.89–6.87 (m, 1H), 6.58 (s, 2H), 6.51 (d, J ¼ 2.94
Hz, 1H), 6.36 (s, 1H), 6.28 (s, 1H), 6.23 (d, J ¼ 8.07 Hz, 1H), 4.90
(s, 1H), 4.0–3.9 (m, 10H), 3.52 (q, J ¼ 6.72 Hz, 4H), 1.9–1.7 (m,
10H), 1.6–1.2 (m, 96H), 0.95–0.81 (m, 15H); FT-IR (KBr, cmꢀ1):
2920, 2853, 1635, 1624, 1580, 1234, 1118, 1075, 720; elemental
analysis: calcd (%) for C100H153N3O9Pd (1646.06): C 72.89, H
9.36, N 2.55; found C 72.90, H 9.38, N 2.56.
Samples were then used to perform photoconductivity
measurement. Note that by using the geometry shown in Fig. 7
and irradiating the sample from the top, light reaches the
material under investigation directly, making it possible to
perform photoconductivity experiments even at wavelengths
below 350 nm, where ITO starts to absorb.
Photoconductivity experiments were performed by using
either light from a Xe-Lamp/Monochromator system or from a
He–Ne laser. To change the light intensity, a set of neutral
density filters with different transmittance were used. The
photocurrent iphoto was obtained from the difference between the
dark and the light currents, measured using an electrometer
(6517A from Keithley), also used to apply the electric field.
Synthesis of complex V. Intermediate 10 (161 mg, 0.07 mmol)
and 2 eq. of H(O^N)1 (ref. 12) (176 mg, 0.14 mmol) were dis-
solved in 30 mL of dichloromethane and 10 mL of ethanol. The
reaction mixture was stirred at r.t. for 5 days. The solvent was
removed under reduced pressure and the resulting solid was
filtered and washed with hot ethanol, immediately filtered on a
cotton filter and dissolved in dichloromethane.
1
Acknowledgements
V: dark-blue waxy solid; yield 67% (222 mg); H NMR (300
MHz, CDCl3, 25 ꢁC, TMS): d ¼ 9.28 (d, J ¼ 9.54 Hz, 1H), 7.94 (s,
1H), 7.55–7.39 (m, 2H), 7.04–6.78 (m, 4H), 6.7–6.4 (m, 6H), 6.36
(s, 1H), 6.26 (s, 1H), 5.92 (s, 1H), 4.85 (s, 2H), 4.47 (s, 2H), 4.12–
3.71 (m, 16H), 3.5–3.6 (m, 4H), 2.2–1 (m, 190H), 0.1–0.8 (m,
15H); FT-IR (KBr, cmꢀ1): 2913, 2852, 1624, 1581, 1234, 1117,
Financial support from the Ministero dell’Istruzione,
ꢀ
dell’Universita e della Ricerca (MiUR) through the Centro di
Eccellenza CEMIF.CAL (CLAB01TYEF) and from the
ꢀ
Doctorate School Bernardino Telesio of the Universita della
Calabria (A. Ionescu) is gratefully acknowledged.
720; elemental analysis: calcd (%) for
C143H231N3O13Pd
(2304.65): C 74.64, H 10.09, N 1.82; found C 74.61, H 10.12, N
1.62.
Notes and references
1 S. R. Forrest, Nature, 2004, 428, 911.
2 Special Issue: Organic Electronics and Optoelectronics, Chem. Rev.,
2007, 107(4), 923–1386 and papers therein.
Molecular modelling
The H(NR) dimer structure was modelled and optimised in the
gas phase using the Polak–Ribiere conjugate gradient method of
the Hyperchem 7.51 programme (Hyperchem, Hypercube Inc.,
1115 NW 4th street, Gainsville FL 32601, USA) following stan-
dard techniques.33
3 D. S. Weiss and M. Abkowitz, Chem. Rev., 2010, 110, 479.
4 C. Im, E. V. Emelianova, H. Bassler, H. Spreitzer and H. Beker, J.
Chem. Phys., 2002, 117, 2961.
5 N. Geacintov and M. Pope, J. Chem. Phys., 1967, 47, 1194.
6 S. Sergeyev, W. Pisula and Y. H. Geerts, Chem. Soc. Rev., 2007, 36,
1902.
7 E. M. Garcia-Frutos, U. K. Pandey, R. Termine, A. Omenat,
J. Barbera, J. L. Serrano, A. Golemme and B. Gomez-Lor, Angew.
Chem., Int. Ed., 2011, 50, 7399.
Photoconductivity experiments
8 K. Y. Law, Chem. Rev., 1993, 93, 449.
9 (a) I. Aiello, D. Dattilo, M. Ghedini and A. Golemme, J. Am. Chem.
Soc., 2001, 123, 5598; (b) R. Termine, D. Dattilo, I. Aiello,
M. Ghedini and A. Golemme, Adv. Mater., 2003, 15, 723.
10 I. Aiello, D. Dattilo, M. Ghedini, A. Bruno, R. Termine and
A. Golemme, Adv. Mater., 2002, 14, 1233.
The substrates used to prepare the samples were obtained from
Indium Tin Oxide (ITO) covered glass sheets (UNAXIS, 110 nm
ITO thickness, sheet resistivity 12 U ,ꢀ1). A photolithographic
process was applied in order to pattern the ITO in 2 mm wide
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 23617–23626 | 23625

View Article Online
11 M. Ghedini, A. Golemme, I. Aiello, N. Godbert, R. Termine,
A. Crispini, M. La Deda, F. Lelj, M. Amati and S. Belviso, J.
Mater. Chem., 2011, 21, 13434.
21 M. Lehmann, C. Koln, H. Meier, S. Renker and A. Oehlhol, J. Mater.
Chem., 2006, 16, 441.
ꢀ
22 M. Kaller, S. Tussetschlager, P. Fisher, C. Deck, A. Baro,
12 N. Godbert, D. Dattilo, R. Termine, I. Aiello, A. Bellusci, A. Crispini,
A. Golemme and M. Ghedini, Chem.–Asian J., 2009, 4, 1141.
13 (a) H. Eichhorn, J. Porphyrins Phthalocyanines, 2000, 4, 88; (b)
M. G. Debije, J. Piris, M. P. de Haas, J. M. Warman, Z. Tomovic,
F. Giesselmann and S. Laschat, Chem.–Eur. J., 2009, 15, 9530.
23 V. A. Gunyakov, N. P. Shestakov and S. M. Shibli, Liq. Cryst., 2003,
30, 871.
24 M. Kaller, C. Deck, A. Meister, G. Hause, A. Baro and S. Laschat,
Chem.–Eur. J., 2010, 16, 6326.
€
C. D. Simpson, M. D. Watson and K. Mullen, J. Am. Chem. Soc.,
2004, 126, 4641; (c) A. M. van de Craats and J. M. Warman, Adv.
Mater., 2001, 13, 130.
25 K. Venkatesan, P. H. J. Kouwer, S. Yagi, P. Muller and
T. M. Swager, J. Mater. Chem., 2008, 18, 400.
14 (a) T. Q. Nguyen, M. L. Bushey, L. E. Brus and C. J. Nuckolls, J. Am.
Chem. Soc., 2002, 124, 15051; (b) T. Kato, N. Mizoshita and
K. Kishimoto, Angew. Chem., Int. Ed., 2006, 45, 38.
15 (a) J. J. van Gorp, J. A. J. M. Vekemans and E. W. Meijer, J. Am.
Chem. Soc., 2002, 124, 14759; (b) M. L. Bushey, A. Hwang,
P. W. Stephens and C. Nuckolls, Angew. Chem., Int. Ed., 2002, 41,
2828; (c) S. Laschat, A. Baro, N. Steinke, F. Giesselmann,
C. Hagele, G. Scalia, R. Judele, E. Kapatsina, S. Sauer,
A. Schreivogel and M. Tosoni, Angew. Chem., Int. Ed., 2007, 46,
4832.
16 M. Ghedini, D. Pucci, A. Crispini, A. Bellusci, M. La Deda, I. Aiello
and T. Pugliese, Inorg. Chem. Commun., 2007, 10, 243.
17 M. La Deda, M. Ghedini, I. Aiello, T. Pugliese, F. Barigelletti and
G. Accorsi, J. Organomet. Chem., 2005, 690, 857.
18 M. S. J. Briggs, I. Bruce, J. N. Miller, C. J. Moody, A. C. Simmonds
and E. Swann, J. Chem. Soc., Perkin Trans. 1, 1997, 1051.
19 S. A. Swanson, G. M. Wallraff, J. P. Chen, W. Zhang, L. D. Bozano,
K. R. Carter, J. R. Salem, R. Villa and J. Campbell Scott, Chem.
Mater., 2003, 15, 2305.
26 A. Ionescu, N. Godbert, I. Aiello, A. Crispini and M. Ghedini, Mol.
Cryst. Liq. Cryst., 2012, 558, 84.
27 C.M.Golini,B.W.WilliamsandJ.B.Foresman,J. Fluoresc., 1998, 8, 395.
28 E. M. Garcia-Frutos, U. K. Pandey, R. Termine, A. Omenat,
J. Barbera, J. L. Serrano, A. Golemme and B. Gomez-Lor, Angew.
Chem., Int. Ed., 2011, 50, 7399.
29 N. A. Davidenko, N. A. Derevyanko, A. A. Ishchenko,
N. G. Kuvshinsky, A. V. Kulinich, O. Y. Neiland and
M. V. Plotniece, Russ. Chem. Bull., 2004, 53, 1674.
30 J. Kalinowski and K. Szybowska, Org. Electron., 2008, 9, 1032.
31 (a) H. C. Hesse, J. Weickert, M. Al-Hussein, L. Dossel, X. Feng,
K. Mullen and L. Schmidt-Mende, Sol. Energy Mater. Sol. Cells,
2010, 94, 560; (b) Q. Zheng, G. Fang, W. Bai, N. Sun, P. Qin,
X. Fan, F. Cheng, L. Yuan and X. Zhao, Sol. Energy Mater. Sol.
Cells, 2011, 95, 2200.
32 J. Pommerehne, H. Vestweber, W. Guss, R. F. Mahrt, H. Bassler,
M. Porsch and J. Daub, Adv. Mater., 1995, 7, 555.
33 (a) Y. S. Gerasymchuk, V. Y. Chernii, L. A. Tomachynskii,
M. Kowalska, J. Legendziewics and St. Radzki, Opt. Mater., 2010,
32, 1193; (b) M. Gervaldo, M. Funes, J. Durantini, L. Fernandez,
F. Fungo and L. Otero, Electrochim. Acta, 2010, 55, 1948.
20 S. A. Martin-Brown, Y. Fu, G. Saroja, M. M. Collinson and
D. A. Higgins, Anal. Chem., 2005, 77, 486.
23626 | J. Mater. Chem., 2012, 22, 23617–23626
This journal is ª The Royal Society of Chemistry 2012