Molecular engineering to improve the charge carrier balance in single-layer silole-based OLEDs

Article abstract of DOI:10.1039/b900780f

We report a molecular engineering study on optical, structural and electrical properties of seven silole derivates aimed at enhancing the charge carrier balance in single-layer devices. By functionalizing two hole-transporting groups, dipyridylamine and anthracene, on the silole ring, we have investigated the influence of both substituents on the hole current. We have concluded that in contrast to dipyridylamine groups, anthracene groups decrease the charge carrier balance since the latter groups not only increase the hole current but also electron contribution. Mixing these hole-transporting groups and doubling their number lead to a novel silole becoming a very efficient emissive layer exhibiting threshold voltage below 3 V and luminous efficiency L_e = 0.8 cd A^-1 at 7 V. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009.

Full text of DOI:10.1039/b900780f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Molecular engineering to improve the charge carrier balance
in single-layer silole-based OLEDsw
Laurent Aubouy,a Nolwenn Huby,b Lionel Hirsch,*b Arie van der Leec
and Philippe Gerbier*a
Received (in Montpellier, France) 15th January 2009, Accepted 31st March 2009
First published as an Advance Article on the web 27th April 2009
DOI: 10.1039/b900780f
We report a molecular engineering study on optical, structural and electrical properties of seven
silole derivates aimed at enhancing the charge carrier balance in single-layer devices. By
functionalizing two hole-transporting groups, dipyridylamine and anthracene, on the silole ring, we
have investigated the influence of both substituents on the hole current. We have concluded that in
contrast to dipyridylamine groups, anthracene groups decrease the charge carrier balance since the
latter groups not only increase the hole current but also electron contribution. Mixing these
hole-transporting groups and doubling their number lead to a novel silole becoming a very efficient
emissive layer exhibiting threshold voltage below 3 V and luminous efficiency Le = 0.8 cd Aꢀ1
at 7 V.
basically consists of a p-doped hole transport layer (p-HTL),
Introduction
an intrinsic electron blocking layer (EBL), emission layer
(EML), hole blocking layer (HBL) and an n-doped electron
transport layer (n-ETL). The p-doping and n-doping result in
high conductivity and Fermi level shift. This leads to high
current injection from both electrodes into the organic
layers.11 Nevertheless, this approach suffers some drawbacks
due to a large number of interfaces and/or the apparition of
segregation phase.
Organic light-emitting diodes (OLEDs) using small molecules
or polymers have been intensively pursued after the initial
work by Tang, Van Slyke and Burroughes and co-workers1,2
because of their enormous potential in flat, flexible panels for
lighting and displays. The search for efficient and stable, new
emitting materials with appropriate emission spectra remains
one of the most active areas of these studies. Different
strategies have been developed to enhance the efficiency of
the devices such as assisted singlet–triplet internal conversion
and the balance of charge carriers in the emissive zone. Among
them, the approach involving the incorporation of heavy
metal complexes has attracted great attention since it provides
both very high efficiencies and white emission.3–8 On the other
hand, the balance of charge carriers in the emissive zone has
attracted much less attention due to the development of
multilayer structures as a response to this issue.9,10 Indeed,
in organic semiconductors, one of the two charge carriers
presents a higher mobility compared to the other one. This
leads to several drawbacks such as, for instance, the location
of the recombination zone close to an electrode, leading to a
huge quenching of excitons. It is possible to overcome
this problem by using PIN OLED structures. This structure
The aim of this work was to design a fluorescent molecule
able to transport both charge carriers. In this way, we focussed
on silole derivatives since they appear to possess all the
requirements to achieve single-layer OLEDs. The siloles12–15
or silacyclopentadienes are
a group of five-membered
silacycles that possess s*–p* conjugation arising from the
interaction between the s* orbital of two exocyclic s bonds
on the silicon atom and the p* orbital of the butadiene
moiety.16 As a consequence, the calculated LUMO level of a
silole ring is lower than those of other heterocyclopentadienes,
such as pyrrole, furan, and thiophene. Moreover, thanks to its
non-aromatic character the p-system of the silole ring is more
prone to allow electron delocalization when compared with its
thiophene cousins.17,18 From a structural point of view,
because of the non-coplanar structure of 2,3,4,5-tetraarylsiloles,
the distances between silole cores of any two adjacent mole-
cules, even in the solid state, are far from the normal p–p
interaction distance (ca. 3–4 A).19 This gives rise to a very
interesting photophysical property called aggregation-induced
photoluminescence (PL) emission (AIE).20,21 Because of the
AIE characteristics, 2,3,4,5-tetraphenylsiloles can show
extremely high PL quantum yields (up to 100%), even in a
crystalline form.22,23 Therefore, 2,3,4,5-tetraphenylsiloles are
excellent emitters in the fabrication of electroluminescence
(EL) devices, an external quantum efficiency (ZEL) up to 8%,
close to the theoretical limit for a singlet emitter, being realized
with such derivatives in the emissive layer.24,25 Finally, siloles
exhibit very high electron mobilities, exceeding those for the
a
´
Universite Montpellier 2, Institut Charles Gerhardt de
Montpellier—UMR 5253, CC 007, Place Eugene Bataillon,
`
34095 Montpellier cedex 5, France. E-mail: gerbier@univ-montp2.fr;
Fax: +33 (0)4 67 14 38 52
b
´
´
´
Universite Bordeaux 1, Laboratoire d’Integration du Materiau au
`
`
Systeme (IMS)—UMR 5218, Ecole Nationale Superieure de Chimie
et de Physique de Bordeaux, 16, Avenue Pey Berland, 33607 Pessac
Cedex, France. E-mail: lionel.hirsch@ims-bordeaux.fr;
Fax: +33 (0)5 40 00 66 31
c
´
Universite Montpellier 2, Institut Europeen des Membranes—UMR
5635, CC 047, Place Eugene Bataillon, 34095 Montpellier cedex 5,
´
`
France. E-mail: avderlee@univ-montp2.fr;
Fax: +33 (0)4 67 14 38 52
w CCDC reference number 725599. For crystallographic data in CIF
or other electronic format see DOI: 10.1039/b900780f
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
1290 | New J. Chem., 2009, 33, 1290–1300

View Article Online
amino functionalities grafted on each side, which act as
hole-transporting groups.29,30 By associating these two func-
tionalities, we have achieved a sufficient balance of charge to
make light from single-layer OLEDs. However, investigations
of the temperature dependence and the electron injection
barrier dependence have highlighted the weak hole contribu-
tion in hole-only devices that is three orders of magnitude
lower than the electron one.27 Since it may be expected that a
better balance of charge should improve greatly the efficiency
of the devices, we have designed the siloles shown in Schemes 1
and 3 to increase their ambipolar character and, consequently,
their hole-transporting properties.31 Firstly, the dipyridyl-
amino hole-transporting groups were substituted by anthracenyl
groups, which can also be considered as good hole-transporters
in spite of their displaying an ambipolar character (siloles B
and B0).32,33 Secondly, we varied the ratio of hole carriers to
electron carriers, 1 : 1 for silole E, 2 : 1 for siloles A, A0, B, B0
and 4 : 1 for siloles C and D. The effect of the conjugation
between electron- and hole-transporting moieties was also
studied by inserting a disrupting ether bridge between the
two (siloles A vs. A0 and B vs. B0). Optical and structural
properties are systematically correlated to the device perfor-
mances in order to highlight the influence of the number of
hole-transporting groups on the balance of charge carriers.
Scheme 1 Tamao’s synthetic route to 2,5-difunctionalized siloles:
(i) 4 equiv. LiNp, (ii) 4 equiv. ZnCl2ꢂTMEDA, (iii) 2 equiv. ArBr,
PdCl2(PPh3)2.
well-known tris(8-hydroxyquinoline) aluminium (Alq3), and
have been utilized as the electron-transporting layer for EL
devices.26–28
The results presented in this paper follow previous reports
concerning the silole A (Scheme 1).25,27 This molecule is based
on a silacyclopentadiene core, which acts both as the emissive
and the electron-transporting component, and two dipyridyl-
Results and discussion
Syntheses
The siloles A–D were conveniently prepared by the method
described by Tamao and Yamaguchi and co-workers34 involving
the one-pot reductive intramolecular cyclization of bis(phenyl-
ethynyl)silane and the subsequent Pd(0)-catalysed cross-
coupling reaction with the desired arylbromide (Scheme 1). The
synthesis of 9-(4-bromophenyl)anthracene 3 and 9-[4-(4-bromo-
phenoxy)phenyl]anthracene
4 (Scheme 2) was achieved
starting from anthrone by using the procedure described by
Murphy et al.35 3,5-Bis(2,20-dipyridylamino)bromobenzene
5 and [4-(4-bromo-phenoxy)phenyl]dipyridyn-2-yl-amine 6
(Scheme 2) were synthesized through a modification of the
original Ullman reaction.36,37 The preparation of the asymmetri-
cally 9,10-diarylanthracene 10 (Scheme 3) was achieved
Scheme 2 Syntheses of the hole-transporting functionalities: (i) 1 equiv.
n-BuLi, (ii) MeOH, HCl (6 M), (iii) K2CO3, CuSO4ꢂ5H2O.
Scheme 3 Synthesis of 10: (i) n-BuLi, (ii) 2 equiv. 1-bromo-4-lithio-
benzene, (iii) NaH2PO4, KI, glacial acetic acid.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 1290–1300 | 1291

View Article Online
through an adaptation of the procedure described by Smet
et al.38,39 It involves firstly the lithiation of compound 725,40
followed by the addition of the resulting lithio derivative to
anthraquinone to afford the monoadduct 8 in 43% yield. To
this compound, a two-fold excess of 4-bromophenyllithium
was added yielding the diol 9. The excess of 4-bromophenyl-
lithium was necessary to react with the OH group present in
the monoadduct 8. Reduction of the latter using NaH2PO2
and KI in refluxing acetic acid afforded 10 as a light yellow
solid in 20% yield.
Scheme 5 Location of the torsion angles reported in Table 1.
The synthesis of silole E (Scheme 4) involves firstly the
preparation of the bis-silole derivative 11 by the Tamao–
Yamaguchi reaction between the dizincic intermediate 2 and
1.5 equiv. of 1,4-dibromobenzene. The bis(bromophenyl)silole
12 which is formed along with 11 is easily isolated by column
chromatography and serves as starting material for other
syntheses. The subsequent Suzuki coupling between 11 and
the boronic acid derivative 1340 afforded the expected bis-silole
E in good yield.
Fig.
1 DFT-optimized (B3LYP-6/31G) molecular structures of
siloles A (top) and B0 (bottom).
their electronic behaviour. Since crystals suitable for a X-ray
structure determination could only be obtained for A,41 B,42
and 11, we turned to density functional theory (DFT) calcula-
tions with the B3LYP functional to obtain information about
the molecular conformations for the other siloles.43 Due to the
size of the molecules, geometry optimizations without symmetry
constrains were performed with the 6-31G basis set to the standard
convergence criteria as implemented in Gaussian98.44 Such
calculations were followed by single point runs using a
6-31+G* basis to obtain accurate energies. Structurally
characterized siloles, dipyridylamines, and diphenylethers
served as benchmarks to test how well the experimentally
determined geometry is reproduced by the calculations, and
some relevant torsion angles are collected in Table 1.19 As
exemplified with A (Fig. 1), all compounds have a propeller-
like arrangement of the four phenyl rings, as found in
the crystal structures, while the two methyl substituents on the
silicon atom are nearly perpendicular to the mean plane of the
SiC4 ring. The torsion angles of the substituted phenyl rings at
the 2- and 5-positions of the central silole ring (j1) are in
Scheme 4 Synthesis of E: (i) PdCl2(PPh3)2, (ii) 1 equiv. n-BuLi,
B(OMe)3, (iii) H2O, NH4Cl, (iv) Pd(PPh3)4, K2CO3.
Geometries of siloles
The determination of the conformational preferences of these
molecules is of utmost importance for the understanding of
Table 1 Torsion angles [1] in silolesa
Ab
Bb
C
Dbd
E
c
j1c
43.4 (44.7)
45.4 (34.3)
—
49.1 (58.4)
—
77.0 (68.5)
38.6
46.8
—
47.8
50.8 (50.4)
71.6 (74.2)
43.5 (45.1)
55.5
—
j2c
j3
a
b
Average values. Values from crystal structures (see text) are in
Fig. 2 X-Ray structure of silole 11. The CH2Cl2 crystallization
c
parentheses. See Scheme 5. See ref. 45.
d
molecule has been removed for the sake of clarity.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
1292 | New J. Chem., 2009, 33, 1290–1300

View Article Online
Table 2 Main values of the optical properties (UV-visible and fluorescence spectra) in solution and in thin filmsa
Silole
Absorptionb lmax/nm
PLb lmax/nm
Quantum yieldc
EL lmax/nm
HOMO and (LUMO) levelsd/eV
A
B
388 (403)
389 (394)
379 (385)
404 (437)
419 (ꢀ)
526 (542)
428 (507)
503 (511)
455 (503)
538 (557)
505 (505)
407 (507)
0.040
0.015
0.038
0.040
0.002
0.006
0.020
545
521
520
518
586
504
517
ꢀ5.20 (ꢀ2.04)
ꢀ5.39 (ꢀ2.00)
ꢀ5.32 (ꢀ1.79)
ꢀ5.09 (ꢀ1.85)
ꢀ5.11 (ꢀ1.89)
ꢀ5.00 (ꢀ2.01)
ꢀ5.16 (ꢀ2.32)
C
D
E
A0
B0
378 (375)
388 (392)
a
b
c
The values recorded with thin solid films are in parentheses. Measured in CH2Cl2. Measured in solution, quantum yield relative to perylene
d
(fem = 0.94). From B3LYP-6/31G* DFT calculations.
lies about a two-fold axis that passes through the middle of the
central phenyl ring. The torsion angles j1 between this ring
and the two adjacent siloles have a value of 45.141. As a result,
the two silole rings are nearly perpendicular, which contrasts
very strongly with the thiophene analogues that are nearly
planar. This situation is also encountered in the optimized
geometry of silole E.
Optical and electronic properties
The UV-visible absorption, and photoluminescence (PL)
spectra have been measured both in solution and thin films.
Electroluminescence (EL) spectra were obtained from single
layer devices with the structure: ITO/PEDOT:PSS/silole
(50 nm)/Ca. The most relevant data obtained from these
spectra are collected in Table 2. Fig. 3 and 4 are representative
of the two behaviours that are encountered in this series of
molecules. As it is seen in Fig. 3, compounds A, A0, C and
E display a broad absorption band in the range of 368 nm to
389 nm which is characteristic of the p - p* transition in the
silole ring.34 However, as seen in Fig. 4, in the compounds B,
B0 and D, this transition is overlapped by the readily recogniz-
able pattern of phenylanthracenes. The comparison of Fig. 3
and 4 reveals that the siloles without anthracene side-groups
behave differently from those bearing them. In the first family
(siloles A, A0, C and E) all the emission spectra are nearly
superimposable whatever the excitation mode or the physical
state (solution vs. thin film). The most important deviation is
found with silole C in which a shift of ca. 9 nm is found
between the PL and the EL spectra (Table 2). In the second
family (siloles B, B0 and D), the PL and EL spectra show
differences both in the position of their emission maxima and
in their shape, as exemplified in Fig. 4 with silole D. In
solution, the anthracene moieties appear to be mainly respon-
sible of the emission, as attested by the vibronic coupling seen
on the curves. Moreover, it is worth noting that the Stokes
shift that is observed with this second family (ca. 20–50 nm) is
substantially smaller than with the first one (ca. 120–140 nm).
As usually observed for 2,3,4,5-tetraphenylsiloles, the
quantum yields in solution are rather low (Table 2), the lowest
value being found with E in which two silole rings are present
in the structure. This behaviour likely must originate from a
resonant photon absorption, since there is a substantial overlap
between the absorption and the emission spectra. Moreover
this observation, which indicates that the two siloles rings
behave independently, is in good agreement with their perpendi-
cular arrangement in the molecular structure (see above).
Fig. 3 Normalized UV-visible (&), photoluminescence (in solution
J and thin film $) and electroluminescence (TT) spectra of silole A.
Fig. 4 Normalized UV-visible (&), photoluminescence (in solution:
J and thin film: $) and electroluminescence (TT) spectra of silole D.
the range of what is usually observed with tetraarylsiloles
(ca. 30–601).
The torsion angles between the anthracene main plane and
the adjacent phenyl ring (j3) fall well in the range of what is
usually observed with related molecules (ca. 701). This is
expected to induce a strong reduction of the conjugation
between the electron-transporting silole ring and the lateral
hole-transporting groups. The same is expected when an ether
bridge is inserted between the two electroactive components
since the plain planes of the phenyl ring on both sides of the
oxygen atom are nearly perpendicular (see Fig. 1). Though no
crystal suitable for X-ray diffraction was obtained for the
bis-silole E, we were able to solve the structure of its precursor
11. This compound crystallizes along with one CH2Cl2 mole-
cule in the C2/c space group. As seen in Fig. 2, the molecule
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 1290–1300 | 1293

View Article Online
Fig. 5 B3LYP/6-31G*-calculated highest occupied (HOMO) and
lowest unoccupied (LUMO) molecular orbitals for siloles A and B.
Interestingly, the presence of either 9-phenylanthracene or
9,10-diphenylanthracene highly fluorescent subunits (see
below) in the molecular structures of siloles B, B0 and D has
no positive effect on their quantum yields. Along with what is
observed in the fluorescence spectra, this indicates that a large
amount of energy is transferred from the anthracene chromo-
phores to the silole and then released via non-radiative pro-
cesses. Finally, semi-quantitative measurements of the
fluorescence quantum yields on thin films have also been
performed. The following sequence was found: B E B0 4
D 4 A 4 A0 4 C 4 DPA 4 perylene E E where DPA
(9,10-diphenylanthracene, fem (solution) = 1.00) and P
perylene (fem (solution) = 0.94)46 are given for comparison.
On account of the AIE phenomenon,20,21 the siloles display a
very strong fluorescence in the solid state that exceeds both
DPA and perylene which possess nearly quantitative quantum
yields in solution.
Fig. 7 EL spectra from devices ITO/PEDOT:PSS/silole (50 nm)/Ca.
examination of the wavefunctions calculated for siloles A0
and B0 (Fig. 6 for B0) shows a nearly identical orbital distribu-
tion on the tetraphenylsilole core. However, in contrast to
their parents A and B, very little electron probability density is
found on either the dipyridylamino or the phenylanthracene
moieties. This illustrates the expected disruption of the conju-
gation brought about by the diphenylether bridges, and the
reason why the emission maximum of these two molecules is
blue-shifted by 20 nm compared to their parents A and B.
To better understand the optical data, we now turn to a
description of the main characteristics of the HOMO and
LUMO levels as calculated at the DFT level. The analysis of
the HOMO and LUMO wavefunctions shown for siloles A and
B in Fig. 5 shows the typical pattern of tetraphenylsiloles.47,48
The HOMO wavefunctions show a very similar spatial
distribution with an antibonding character between the silole
ring and the phenyl rings located at the 2,5-positions. The
same similarity is found with the LUMO wavefunctions in
which bonding character is observed between the silole ring
and the adjacent phenyl rings. The energies of the HOMO and
LUMO orbitals (Table 2), which do not vary to any great
extent upon modification of the substituents, are in the
range of what is usually reported for tetraarylsiloles. The
Electroluminescence properties and balance of charge carriers
In order to study the EL properties, single layer devices were
investigated, with the following structure: ITO/PEDOT:PSS/
silole (50 nm)/Ca. The electroluminescent spectra are shown in
Fig. 7. As observed during the photoluminescence studies,
only the silole ring contributes to the emission and the device
based on molecule A is 20 nm red-shifted compared to the
others. All the emissions correspond to the yellow–green
domain in the chromatic diagram of the Commission Inter-
nationale de l’Eclairage. Their corresponding current density–
voltage and luminance–voltage are presented in Fig. 8(a) and
8(b), respectively. When compared with the calculated HOMO
and LUMO energy levels of the siloles (see Table 2), the high
work function of the anode ITO/PEDOT:PSS (ꢀ5.2 eV) and
the low work function of the calcium cathode (ꢀ2.9 eV)
should favour the injection of both charge carriers in the
active layer. As a consequence, threshold voltage values
generally below 4V are necessary for the detection of lumi-
nance of the device (Table 3). Moreover, with the exception of
siloles C and E, all the molecules presented here display quite
good luminous efficiencies (Le) for single-layer devices. In
terms of performances, D exhibits the best values with a
threshold voltage below 3 V and a luminous efficiency Le of
0.75 cd Aꢀ1 at 7 V. In contrast, C and E are weakly electro-
luminescent. Indeed, these molecules need higher applied
voltages to reach the same order of current density than with
A, B or D.
Fig. 6 B3LYP/6-31G*-calculated HOMO and LUMO molecular
orbitals for silole B0.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
1294 | New J. Chem., 2009, 33, 1290–1300

View Article Online
Fig. 8 (a) Current density–voltage characteristics for devices based on /PEDOT:PSS/silole (50 nm)/Ca, and (b) corresponding luminance–voltage
characteristics.
that defines the efficiency of an OLED on which we can play
from a molecular engineering point of view is the balance of
the charge carrier (g).
Table 3 The luminance (L), luminous efficiency (Le) and energetic
efficiency (Re) of siloles operating in ITO/PEDOT:PSS/silole/Ca
OLEDsa
L/cd mꢀ2
Le/cd Aꢀ1
Re/lm Wꢀ1
To analyse the result of molecular engineering on the silole
core in terms of the balance of charge carriers, one has still to
take into account both the charge transport processes and the
quantum yield of fluorescence in the solid state of each
molecule. Concerning the first issue, at least two parameters
have to be taken into consideration: the orbital energy levels
and the organization of the molecules in the thin film.27,42,51,52
In previous work we have studied the transport properties of
siloles A and B.42 Actually, they are very close in behaviour on
account of their similarities, both in terms of molecular
organization (they form amorphous films) and in terms of
the energy levels and orbital distribution (see above).27,42,51
Therefore, it seems reasonable to set the factor Zrecomb in
eqn (1) to the same arbitrary value for all the series of
molecules studied here. In this way, the comparison of the
luminous efficiencies corrected by the relative solid-state FPL
value should allow an estimation of the effect of molecular
engineering on the balance of charge carriers.
Silole
Vth/V
A
B
3.5
3.1
9
2.9
4.5
18
26 (370)b
25 (350)b
8 (74)c
0.17 (0.20)b
0.16 (0.18)b
0.05 (0.036)c
0.52 (0.75)c
ꢀ(0.09)c
0.095 (0.06)b
0.100 (0.09)b
0.012 (0.015)c
0.320 (0.35)c
—
C
D
E
A0
B0
80 (1550)c
ꢀ(86)c
5 (5)b
0.03 (0.03)b
0.14 (0;19)c
0.003 (0.003)b
0.060 (0.060)c
4.2
21 (290)c
a
b
Values measured at a current density of 20 mA cmꢀ2
. Value
Value in parentheses
in parentheses measured at 100 mA cmꢀ2
measured at 200 mA cmꢀ2
.
c
.
To better understand the origin of this different behaviour
and to try to outline the relationships that may exist between
the molecular structure and the balance of charge carriers in
this series of molecules, one has to consider the factors that
determine the efficiency of an OLED. Actually, this may
approximately be calculated by the following equation:49,50
From semi-quantitative measurements we have found the
following sequence for the solid-state photoluminescence
quantum yield FPL: B E B0 4 D 4 A 4 A0 4 C 4 E. By
using the procedure of normalization described in the
Zexternal = g ꢃ Zrecomb ꢃ ZST ꢃ Zoptical ꢃ FPL
(1)
where Zexternal is the total power efficiency of the device, g is the
balance of charge carrier, Zrecomb represents the recombination
probability of injected holes and electrons, ZST is the ratio of
singlet and triplet excitons contributing to the radiative
recombination, Zoptical is the efficiency of the optical out-
coupling from the device, and FPL is the quantum yield of
fluorescence of the emissive material. By spin statistics,
Table 4 The luminous efficiency (Le), relative solid-state PL quantum
yield (FPL) and the hole carrier to the electron carrier formal ratio
(h+ to eꢀ) of siloles operating in ITO/PEDOT:PSS/silole/Ca OLEDs
Silole Lea/cd Aꢀ1 Relative solid-state FPL Le/FPL/cd Aꢀ1 h+ : eꢀb
Z
ST, which is the ratio of singlet to triplet excitons, should
A
B
0.17
0.16
0.05
0.44
1.00
0.39
0.16
41.00
0.78
—
2
2
4
4
1
2
2
be ZST = 0.25, since parallel spin pairs will recombine to triplet
excitons while antiparallel spin pairs will recombine to singlet
and triplet excitons. Thus, for fluorescent emitters, we find
C
D
E
A0
B0
o0.05
—
0.52
0.67
—
Z
ST = 0.25, which is a severe limitation of quantum efficiency
0.03
0.14
0.36
1.00
0.08
0.14
of an OLED. Concerning the optical outcoupling efficiency, a
simple estimation regarding the OLED as a classical optics
device shows that a flat device with typical refractive index of
the organic layers of 1.7, deposited on ITO/glass, achieves
approximately 20% outcoupling. Therefore, the first factor
a
b
Values measured at a current density of 20 mA cmꢀ2
.
A silole ring
accounts for 1 eꢀ whereas either an anthracenyl or a dipyridylamino
side-group accounts for 1 h+
.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 1290–1300 | 1295

View Article Online
experimental section, we have found that the photolumines-
cence intensity of B is ca. 1.5 times higher than that of
D, ca. 2.3 times higher than that of A, ca. 2.8 times higher
than that of A0 and more than 20 times higher than that of C
(the value for E is not given since it is of the same order as the
error for the measurement). Therefore, the ratio Le over FPL
should give a good indication as to the correlation between the
balance of charge carriers in the device and the hole carrier
moieties (h+: dipyridylamino or anthracenyl side-groups) over
the electron carrier moieties (eꢀ: silole ring) present in the
molecular structure (Table 4).
ambipolar molecules, in which the silole central core acts as an
electron-transporting unit, whereas either dipyridylamino or
anthracene groups act as hole-transporting groups. The siloles
were conveniently prepared by the method described by
Tamao and Yamaguchi. The studies have clearly outlined
some trends.
The presence of either dipyridylamino or anthracene groups
in the molecular structure of the siloles brings about a
significant improvement of the balance of charge in the
devices.
The dipyridylamino groups appear to be more efficient than
the anthracenyl groups since they are capable of ambipolar
transport.
From an examination of Table 4, it appears that the major
trend is that the higher the h+ to eꢀ ratio, the more the balance
of charges appears to be improved. This result is in good
agreement with the fact that the silole ring possesses an
exceptional electron carrier ability26–28 greatly exceeding the
hole carrier ability of the organic groups grafted to it. As a
consequence a large number of hole-transporting groups are
needed to correct the balance of charge. Therefore, the silole D
in which the h+ to eꢀ ratio is equal to 4 displays the best
luminous efficiency. This is not so surprising since silole D
may also be considered as silole B bearing two additional hole-
transporting side-groups. In the case of the silole C, in spite of
a similar ratio, the performances are disappointing since
luminance and efficiencies (see Table 2) are one order of
magnitude lower than A and B at 20 mA cmꢀ2. Moreover
the current density is considerably lower than that observed
with the other molecules at a given applied voltage. This
phenomenon may be attributed to two main reasons: (i) this
silole possesses a weak solid-state FPL when compared to D,
and (ii) the four dypirydilamino groups generate strong steric
hindrance which disfavours the electron transfer between the
silole rings in the device.
The ratio of hole-transporting groups to siloles effectively
improves the balance of charge, but the overall effect is not
easy to forecast since intermolecular parameters are also
involved in charge transport.
The conjugation between electron- and hole-transporting
groups appears to be important for achieving high efficiency.
This indicates that there is synergy between the various
components of the molecule.
As a consequence, we have synthesized the silole D, in which
two anthracene groups and two dipyridylamino groups are
connected to the central silole core through a conjugated
backbone. This silole, operating in a single-layer OLED,
exhibited an excellent performance with a threshold voltage
below 3 V and luminous efficiency Le = 0.8 cd Aꢀ1 at 7 V.
Additional studies are in progress to better understand the
interplay between the charge-transporting groups in these
ambipolar molecules.
Experimental
The comparison of siloles A and B in which the h+ to eꢀ
ratio is equal to 2 allows us to estimate the relative ability of
the side-group to transport holes. They are both equivalently
efficient in OLEDs but A is characterized by an Le to FPL ratio
of 0.39 whereas that for B is 0.16. In other words, the
dipyridylamino groups appear to be more efficient than
anthracenyl groups as hole carriers in correcting the balance
of charges in silole-based devices. This may originate from the
fact that anthracene entities not only enhance the hole current
compared to the dipyridylamine ones, but also increase the
electron current, leading to the smallest correction of the
balance of charge carriers. The comparison of siloles A, A0,
B and B0 allows us now to evaluate the importance of the
conjugation between both charge carriers since the presence of
the diphenyl bridge has been shown to isolate both moieties
from an electronic point of view (see above). The disruption of
the conjugation in silole A0 is accompanied by a marked
decrease in efficiency when compared to A, while the solid-
state FPL of both are close enough. In contrast to that, the
same modification only weakly affects the efficiency of the
devices based on siloles B and B0.
General methods and device performance measurements
Solvents were distilled prior to use. THF and ether were dried
over sodium–benzophenone, and distilled under argon. All
reactions were carried out under an argon atmosphere. 1H, 13
C
and 29Si NMR spectra were recorded on a Bruker Advance
200 DPX spectrometer, the FT-IR spectra on a Thermo
Nicolet Avatar 320 spectrometer, the UV-visible spectra on
a Secomam Anthelie instrument and the MS spectra on a Jeol
JMS-DX 300 spectrometer. Fluorescence spectra in thin films
were recorded with an Edinburgh Instruments Ltd spectro-
fluorimeter. Absorption spectra in thin film were realized
with an UV-visible SAFAS Monaco 190 DES spectrometer.
Current–voltage (I–V) characteristics were recorded using a
Keithley 4200 Semiconductor analyser, and luminance–
voltage (L–V) with a photodiode calibrated with a Minolta
CS-100 luminancemeter. Electroluminescence (EL) spectra
were measured using an Ocean Optics HR2000 CCD spectro-
meter. All electroluminescent devices were fabricated and
characterized in a glove box under nitrogen with [O2] and
[H2O] less than 0.1 ppm.
The semi-quantitative solid-state quantum yields have been
measured on evaporated films of the same thickness and
corrected by the corresponding absorption coefficient at the
excitation wavelength (380 nm). In order to compare mate-
rials, care was taken concerning the experimental conditions.
Conclusions
With the goal to improve the balance of charge carrier in single
layer silole-based OLEDs, we have synthesized a series of
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
1296 | New J. Chem., 2009, 33, 1290–1300

View Article Online
Spectrofluorimeter parameters were kept unchanged from one
sample to another, thickness of the thin films were the same
and the fluorescence spectrum of each molecule was corrected
by the corresponding absorption coefficient at the excitation
wavelength. An excitation wavelength at 380 nm was selected,
since all maxima absorption bands are localized in this
domain.
under stirring for 8 h while allowing the temperature to slowly
reach room temperature. An aqueous solution of HCl (1 M)
was then added to the reaction mixture until a pH of 4–5 was
reached and extracted with Et2O. After the usual processing,
the resulting residue was subjected to a silica gel column
chromatography (CH2Cl2–THF gradient 97 : 3 to 90 : 10) to
give 8 as a light yellow powder (yield: 43%). Mp: 265–267 1C.
1H NMR (CDCl3, d, ppm): 8.35–8.09 (m, 4H), 7.77
(dd, 3J(H,H) = 8, 3J(H,H) = 1 Hz, 2H), 7.62 (td, 3J(H,H) = 7,
Syntheses
3J(H,H)
= 2 Hz, 2H ), 7.58–7.48 (m, 4H), 7.37
9-[4-(4-Bromophenoxy)phenyl]anthracene (4). A solution of
n-BuLi (2.5 M) in hexane (7.7 mL, 19 mmol) was added to
an ethereal solution (70 mL) of 4,40-dibromodiphenylether
(6.25 g, 19 mmol) at ꢀ78 1C. The reaction mixture was stirred
for 0.5 h at this temperature and anthrone was added in small
portions (3 g, 15 mmol). This mixture was left under stirring
for 3 h at ꢀ78 1C and the temperature was allowed to slowly
reach room temperature. An aqueous solution of HCl (0.5 M)
was then added to the reaction mixture until a pH of 4–5 was
reached and extracted with Et2O. After the usual processing,
the resulting residue was subjected to silica gel column chromato-
graphy (CH2Cl2–pentane 10 : 90) to give 4 as a white–yellow
(d, 3J(H,H) = 9 Hz, 2H), 7.10 (d, 3J(H,H) = 9 Hz, 2H),
7.00–6.89 (m, 4H), 2.93 (s, 1H).13C NMR (CDCl3, d, ppm):
182.32, 151.40, 148.71, 147.94, 138.45, 134.53, 130.75, 130.36,
128.71, 128.60, 127.37, 127.21, 126.65, 118.83, 117.04, 116.57,
73.5. HRMS (FAB+, m-nitrobenzyl alcohol matrix) m/z:
calcd for [M + H]+ C30H22N3O2: 455.5227; found: 456.1704.
4-[10-(4-Bromophenyl)anthracen-9-yl]phenyldipyridin-2-yl-amine
(10). A solution of n-BuLi (2.5 M) in hexane (10 mL, 25 mmol)
was added to a THF solution (50 mL) of 1,4-dibromobenzene
(5.92 g, 25 mmol) at ꢀ78 1C. The reaction mixture was stirred
for 1 h at this temperature and added via a cannula to a THF
solution (100 mL) of 8 (3.26 g, 7.1 mmol) also at ꢀ78 1C. This
mixture was stirred for 8 h while allowing the temperature to
slowly reach room temperature, then treated with aqueous
HCl (1 M) to pH 4–5 and extracted with ether. The combined
organic layers are dried over MgSO4 and evaporated under
vacuum to afford compound 9 as a viscous oil. The residue
was then dissolved in glacial acetic acid (60 mL), treated with
NaH2PO4 (8.44 g, 95 mmol) and KI (4.22 g, 25 mmol), and
heated to reflux for 20 min. After cooling, the reaction mixture
was treated with cold water (300 mL) and extracted with
CH2Cl2. After the usual processing, the resulting residue was
1
solid (yield: 50%). Mp: 149 1C. H NMR (CDCl3, d, ppm):
8.54 (s, 1H), 8.09 (d, 3J(H,H) = 8, 2H), 7.75 (d, 3J(H,H) = 8 Hz,
3
1H), 7.60–7.37 (m, 8H), 7.23 (d, J(H,H) = 9 Hz, 2H), 7.12
3
(d, J(H,H) = 9 Hz, 2H).13C NMR (CDCl3, d, ppm): 158.81,
156.71, 136.56, 134.35, 133.27, 133.15, 131.79, 130.79, 128.83,
127.13, 127.07, 125.86, 125.55, 121.33, 119.01, 116.45.
HRMS (FAB+, m-nitrobenzyl alcohol matrix) m/z: calcd
for [M + H]+ C26H17BrO: 424.0463; found: 424.0456.
[4-(4-Bromophenoxy)phenyl]dipyridyn-2-yl-amine (6).
A
mixture of 4,40-dibromodiphenylether (7.14 g, 21.7 mmol),
di-2-pyridylamine (1.50 g, 8.70 mmol), K2CO3 (1.40 g, 10.4 mmol)
and CuSO4ꢂ5H2O (0.217 g, 0.87 mmol) in water (20 mL) and
CH2Cl2 (100 mL) was stirred well and evaporated to dryness
under vacuum. The mixture was ground in a mortar and 3–5
drops of CH2Cl2 were added to this mixture. The mixture was
heated in a Schlenk tube at 210 1C for 6 h. After being cooled
at room temperature, the mixture was dissolved in CH2Cl2
(100 mL) and water (100 mL) and extracted. After evapora-
tion of the solvent, the residue was subjected to column
chromatography CH2Cl2–THF (95 : 5) to afford compound
6 as a white solid (yield: 80%). Mp: 102 1C. 1H NMR (CDCl3,
subjected
to
silica
gel
column
chromatography
(CH2Cl2–MeOH 99 : 1) to give 10 as a light yellow powder
(yield: 20%). Mp: 267 1C. 1H NMR (CDCl3, d, ppm): 8.42 (dd,
3J(H,H) = 6, 3J(H,H) = 2 Hz, 2H), 7.93–7.62 (m, 8H),
7.53–7.39 (m, 10H), 7.21 (d, 3J(H,H) = 7 Hz, 2H), 7.08
3
3
(td, J(H,H) = 7, J(H,H) = 2 Hz, 2H). 13C NMR (CDCl3,
d, ppm): 158.66, 148.85, 144.99, 138.46, 138.05, 136.04,
135.86, 133.49, 132.73, 132.06, 130.31, 130.16, 127.39,
127.07, 126.95, 125.73, 125.59, 122.05, 118.86, 117.81,
116.41. HRMS (FAB+, m-nitrobenzyl alcohol matrix) m/z:
calcd for [M + H]+ C36H24N3Br: 577.5097; found: 578.1204.
3
3
d, ppm): 8.33 (dd, J(H,H) = 6, J(H,H) = 2 Hz, 2H), 7.62
3
3
3
(td, J(H,H) = 7, J(H,H) = 2 Hz, 2H ), 7.47 (d, J(H,H) =
9 Hz, 2H), 7.21 (d, 3J(H,H) = 6 Hz, 2H), 7.00–6.81 (m,
8H).13C NMR (CDCl3, d, ppm): 157.99, 156.17, 154.50,
148.50, 148.47, 140.33, 137.63, 128.92, 120.82, 119.84,
118.15, 116.71, 115.95. HRMS (FAB+, m-nitrobenzyl alcohol
matrix) m/z: calcd for [M + H]+ C22H16BrN3O: 418.0550;
found: 418.0547.
2-[4-(5-(4-Bromophenyl)-1,1-dimethyl-3,4-diphenyl)silol-2-yl)-
phenyl)-5-(4-bromophenyl)]-1,1-dimethyl-3,4-diphenylsilole (11).
A mixture of lithium (0.055 g, 8 mmol) and naphthalene
(1.03 g, 8 mmol) in THF (15 mL) was stirred at room
temperature under argon for 5 h to form a deep green solution
of lithium naphthalenide. To this mixture was added bis-
(phenylethynyl)dimethylsilane 1 (0.50 g, 2 mmol) in THF
(10 mL). After stirring for 10 min, the reaction mixture was
cooled to 0 1C and [ZnCl2(tmen)] (tmen = N,N,N0,N0-tetra-
methylenediamine) (2.01 g, 8 mmol) was added, followed by
an addition of THF (20 mL). After stirring for an hour at
room temperature, a solution of 1,4-dibromobenzene (0.80 g,
34 mmol) in THF (20 mL) and [PdCl2(PPh3)2] (0.10 g,
0.13 mmol) were successively added. The mixture was heated
[4-(Dipyridin-2-yl-amino)phenyl]-10-hydroxyanthracen-9-one
(8). A solution of n-BuLi (2.5 M) in hexane (10.5 mL,
26 mmol) was added to a THF solution (100 mL) of
4-(2,20-dipyridylamino)bromobenzene 725,40 (7 g, 21 mmol)
at ꢀ78 1C. The reaction mixture was stirred for 1 h at this
temperature. A THF solution (150 mL) of anthraquinone
(8.73 g, 42 mmol) was then added to this mixture and left
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 1290–1300 | 1297

View Article Online
under reflux and stirred for 20 h. After hydrolysis by water, the
mixture was extracted with Et2O. After evaporation of the
solvents, the resulting residue was subjected to silica gel
column chromatography (pentane–CH2Cl2 95 : 5) to yield 11
and the silole 12 as separate solids. The siloles were each
recrystallized from a hexane–CH2Cl2 mixture to give 11 as
dark yellow crystals (yield: 30%) and 12 as light yellow
crystals (yield: 20%). Characterization of silole 11: Mp:
310 1C. UV-visible (lmax, nm, loge): 255 (5.46), 399 (5.20).
1H NMR (CDCl3, d, ppm): 7.15 (d, 3J(H,H) = 9 Hz, 4H),
6.96–6.86 (m, 12H), 6.74–6.64 (m, 12H ), 6.58 (s, 4H), 0.35
(s, 1H).13C NMR (CDCl3, d, ppm): 155.14, 153.86, 142.14,
140.60, 139.26, 139.15, 138.84, 137.38, 131.46, 130.82, 130.32,
130.25, 128.93, 127.92, 127.82, 126.77, 126.63, 119.72, ꢀ3.31.
29Si NMR (CDCl3, d, ppm): 8.09. HRMS (FAB+, m-nitro-
benzyl alcohol matrix) m/z: calcd for [M + H]+ C54H44Br2Si2:
906.1348; found: 906.1330. Characterization of silole 11: the
synthesis of this compound was previously described starting
from 1-bromo-4-iodobenzene.53 Mp: 224 1C. UV-visible
(lmax, nm, loge): 250 (5.54), 361 (5.23). 1H NMR (CDCl3,
of the solvents, the resulting residue was subjected to silica
gel column chromatography (pentane–CH2Cl2 85 : 15) and
recrystallized from a hexane–CH2Cl2 mixture to give B as
a yellow crystalline powder (yield: 35%). Mp: 333 1C.
UV-visible (lmax, nm, loge): 254 (6.37), 352 (5.22), 368
(5.43), 386 (5.50). 1H NMR (CDCl3, d, ppm): 8.51 (s, 2H),
8.07 (d, 3J(H,H) = 8 Hz, 4H), 7.72 (d, 3J(H,H) = 8 Hz, 4H),
7.53–7.37 (m, 8H), 7.28–7.00 (m, 18H), 0.77 (s, 6H).13C NMR
(CDCl3, d, ppm): 155.02, 146.18, 142.73, 139.58, 139.39,
137.53, 135.22, 133.80, 131.31, 130.62, 130.55, 129.18,
128.72, 127.90, 127.33, 126.81, 125.61, 125.47, ꢀ3.07.
29Si (CDCl3, d, ppm): 2.80. MS (FAB+, m-nitrobenzyl
alcohol matrix) m/z: 766 [M]+. Analysis calcd for C58H42Si:
90.82 %C, 5.52 %H; found: 90.54 %C, 5.62 %H.
Silole B0. Same procedure as for silole B, using a solution of
4 (2.04 g, 4.8 mmol) in THF (20 mL). After evaporation of the
solvents, the resulting residue was firstly subjected to silica gel
column chromatography (pentane–CH2Cl2 80 : 20) and
recrystallized from hexane–CH2Cl2 to give B0 as a yellow
powder (yield: 89%). Mp: 264 1C. UV-visible (lmax, nm, loge):
257 (6.47), 350 (4.95), 362 (5.13), 385 (5.10). 1H NMR (CDCl3,
d, ppm): 8.54 (s, 2H), 8.09 (d, 3J(H,H) = 8 Hz, 4H), 7.76
(d, 3J(H,H) = 8 Hz, 4H), 7.55–7.38 (m, 12H), 7.24 (d, 3J(H,H) =
8 Hz, 4H), 7.12–76.90 (m, 18H), 062 (s, 6H).13C NMR
(CDCl3, d, ppm): 157.18, 155.30, 154.27, 140.97, 139.39,
136.84, 135.57, 133.75, 132.95, 131.81, 130.83, 130.77,
130.44, 128.79, 128.00, 127.21, 127.01, 126.73, 125.79,
125.53, 119.21, 118.94, ꢀ3.10. 29Si NMR (CDCl3, d, ppm):
7.99. MS (FAB+, m-nitrobenzyl alcohol matrix) m/z: 951
[M + H]+. Analysis calcd for C70H50O2Si: 88.39 %C,
5.30 %H; found: 88.03 %C, 5.40 %H.
3
d, ppm): 7.28 (d, J(H,H) = 9 Hz, 4H), 7.10–7.02 (m, 6H),
6.81–6.75 (m, 8H), 0.47 (s, 1H).13C NMR (CDCl3, d, ppm):
154.90, 141.17, 139.04, 138.57, 131.56, 130.78, 130.24, 128.02,
126.94, 119.95, ꢀ3.52. 29Si NMR (CDCl3, d, ppm): 8.23.
HRMS (FAB+, m-nitrobenzyl alcohol matrix) m/z: calcd
for [M + H]+ C30H18Br2Si: 571.9996; found: 571.9985.
Silole A0. Same procedure as for silole B, using a solution of
p-2,20-dipyridylaminophenyl-4-bromophenylether 651 (1.87 g,
4.4 mmol) in THF (20 mL). After evaporation of the solvents,
the resulting residue was firstly subjected to silica gel column
chromatography (MeOH–CH2Cl2 3 : 97) followed by an
alumina column (THF–CH2Cl2 5 : 95) to give A0 as a bright
yellow powder (yield: 42%). Mp: 128 1C. UV-visible
(lmax, nm, loge): 277 (5.70), 377 (5.20). 1H NMR (CDCl3,
Silole C. Same procedure as for silole B, using a solution of
3,5-bis(2,20-dipyridylamino)bromobenzene37 5 (1.43 g, 4.4 mmol)
in THF (20 mL). After evaporation of the solvents, the
resulting residue was subjected to silica gel column chromato-
graphy (MeOH–CH2Cl2 10 : 90) and recrystallized from a
hexane–CH2Cl2 mixture to give C as a bright yellow powder
(yield: 28%). Mp: 135–137 1C. UV-visible (lmax, nm, loge): 282
(5.80), 303 (5.79), 379 (5.10). 1H NMR (CDCl3, d, ppm): 8.31
(dd, 3J(H,H) = 6, 3J(H,H) = 2 Hz, 8H), 7.53 (td, 3J(H,H) = 7,
3J(H,H) = 2 Hz, 8H), 6.92–6.82 (m, 22H), 6.72–6.66 (m, 6H),
6.51 (d, 3J(H,H) = 2 Hz, 4H), 0.27 (s, 6H).13C NMR
(CDCl3, d, ppm): 157.56, 154.46, 148.20, 145.25, 142.38,
141.19, 138.25, 137.76, 120.51, 127.41, 126.16, 123.49,
121.87, 118.33, 117.24, ꢀ4.11. 29Si NMR (CDCl3, d, ppm):
8.67. MS (FAB+, m-nitrobenzyl alcohol matrix) m/z: 1091
[M + H]+. Analysis calcd for C70H54N12Si: 77.03 %C,
5.16 %H, 15.40 %N; found: 76.56 %C, 5.37 %H, 15.99 %N.
3
3
d, ppm): 8.35 (dd, J(H,H) = 6, J(H,H) = 2 Hz, 4H), 7.58
3
3
3
(td, J(H,H) = 7, J(H,H) = 2 Hz, 4H), 7.16 (d, J(H,H) =
9 Hz, 4H), 7.14–6.86 (m, 30H), 0.52 (s, 6H).13C NMR (CDCl3,
d, ppm): 158.13, 155.05, 154.67, 153.77, 148.47, 140.48, 139.84,
138.88, 137.50, 135.11, 132.16, 132.06, 128.86, 127.53, 126.28,
119.69, 118.68, 117.98, 116.66, ꢀ3.60. 29Si NMR (CDCl3, d,
ppm): 7.87. MS (FAB+, m-nitrobenzyl alcohol matrix) m/z:
937 [M + H]+. Analysis calcd for C62H48N6O2Si: 79.46 %C,
5.16 %H, 8.97 %N; found: 79.01 %C, 5.29 %H, 8.82 %N.
Silole B. A mixture of lithium (0.055 g, 8 mmol) and
naphthalene (1.03 g, 8 mmol) in THF (15 mL) was stirred at
room temperature under argon for 5 h to form a deep green
solution of lithium naphthalenide. To this mixture was added
bis(phenylethynyl)dimethylsilane 1 (0.50 g, 2 mmol) in THF
(10 mL). After stirring for 10 min, the reaction mixture was
cooled to 0 1C and [ZnCl2(tmen)] (tmen = N,N,N0,N0-tetra-
methylenediamine) (2.01 g, 8 mmol) was added, followed by
an addition of THF (20 mL). After stirring for an hour at
room temperature, a solution of 9-(4-bromophenyl)anthracene
335 (1.59 g, 4.8 mmol) in THF (20 mL) and [PdCl2(PPh3)2]
(0.10 g, 0.13 mmol) were successively added. The mixture was
heated under reflux and stirred for 20 h. After hydrolysis by
water, the mixture was extracted with Et2O. After evaporation
Silole D. Same procedure as for silole B, using a solution of
10 (2.76 g, 4.8 mmol) in THF (20 mL). The residue was
purified by silica gel column chromatography (CH2Cl2–THF
gradient 80 : 20 to 70 : 30) and crystallized from a hexane–
CH2Cl2 mixture to afford D as a yellow solid (yield: 15%).
Mp: 376 1C. UV-visible (lmax, nm, loge): 268 (6.30), 389 (5.57),
407 (5.63). 1H NMR (CDCl3, d, ppm): 8.45 (dd, 3J(H,H) = 6,
3J(H,H) = 2 Hz, 4H), 7.91–7.86 (m, 4H), 7.77–7.86 (m, 8H),
7.50–7.39 (m, 16H), 7.28–7.19 (m, 10H), 7.19–07.08 (m, 8H),
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
1298 | New J. Chem., 2009, 33, 1290–1300

View Article Online
7.05–7.02 (m, 8H), 0.88 (s, 6H). 13C NMR (CDCl3, d, ppm):
159.81, 159.33, 155.16, 148.90, 144.84, 141.53, 139.74, 139.29,
138.57, 137.56, 136.75, 136.29, 135.81, 133.48, 130.80, 130.42,
130.33, 129.50, 128.82, 127.85, 127.23, 127.03, 125.26, 125.14,
119.73, 118.37, 117.15, ꢀ4.54. 29Si NMR (CDCl3, d, ppm):
8.27. MS (FAB+, m-nitrobenzyl alcohol matrix) m/z: 1257
[M + H]+. Analysis calcd for C90H64N6Si: 85.95 %C,
5.09 %H, 6.68 %N; found: 85.41 %C, 5.25 %H, 6.57 %N.
for refinement with 307 parameters and two restraints.
R = 0.0391. wR = 0.0383.
Device fabrication
Devices of ca. 10 mm2 were fabricated on ITO-coated
glass substrates (Merck, thickness E 115 nm, sheet resistance
r E 17 O/&). After the cleaning process in trichloroethylene,
ethanol and deionized water,
a UV–ozone treatment
was performed for 15 min. Then, a 50 nm thick layer of
PEDOT-PSS was spin coated at 5000 rpm on top of ITO and
baked at 80 1C for about 1 hour. PEDOT-PSS is a conducting
polymer, acting as a buffer, in reducing short circuit problems
induced by the ITO roughness. It weakly increases the work
function of the anode and acts as a barrier to oxygen and
indium diffusion from ITO.56 On the PEDOT-PSS layer, the
organic compounds as well as the cathodes were thermally
evaporated under secondary vacuum (10ꢀ6 mbar). The deposi-
tion rate of the silole layer was set at about 1 nm sꢀ1 with a
thickness of 50 nm measured in situ using a quartz balance and
ex situ using a Tencor AS-IQ profilometer. Finally, an 80 nm
thick calcium layer capped by 100 nm thick aluminium layer
was evaporated through a shadow mask on top of the silole
derivative. Each preparation and characterization step took
place in glove box under an inert atmosphere.
Silole E. A mixture of bis-silole 11 (0.40 g, 0.44 mmol),
Pd(PPh3)4 (0.05 g, 0.044 mmol), and toluene (50 mL) was
stirred for 10 min. The boronic acid 1340 (0.76 g, 26 mmol) in
20 mL of EtOH and NaOH (0.18 g) in 20 mL of H2O were
subsequently added. The mixture was stirred and refluxed for
72 h and allowed to cool to room temperature. The water layer
was separated and extracted with CH2Cl2. The combined
organic layers were dried over MgSO4, and the solvents were
evaporated under reduced pressure. Purification of the crude
product was carried out by silica gel column chromatography
(CH2Cl2–THF 85 : 15) followed by the recrystallization of the
solid from a CH2Cl2–pentane mixture to afford E as a dark
yellow solid in 86% yield. Mp: 334 1C. UV-visible (lmax, nm,
1
loge): 284 (5.77), 311 (5.79), 419 (5.65). H NMR (CDCl3, d,
ppm): 8.41 (dd, 3J(H,H) = 6, 3J(H,H) = 2 Hz, 4H), 7.66–7.57
3
3
(m, 8H), 7.39 (d, J(H,H) = 8.5 Hz, 4H), 7.24 (d, J(H,H) =
8.5 Hz, 4H), 7.07–6.96 (m, 24H), 6.88–6.83 (m, 4H), 6.72
(s, 4H), 0.53 (s, 12H). 13C NMR (CDCl3, d, ppm): 157.62,
154.31, 153.68, 148.18, 141.58, 140.80, 139.04, 139.01, 138.96,
138.01, 137.14, 137.05, 129.98, 129.96, 129.37, 128.56, 128.13,
127.51, 127.41, 127.36, 126.36, 126.23, 126.15, 118.30, 116.95,
ꢀ3.45. 29Si NMR (CDCl3, d, ppm): 7.95. MS (FAB+,
m-nitrobenzyl alcohol matrix) m/z: 1241 [M + H]+. Analysis
calcd for C86H68N6Si2: 83.18 %C, 5.51 %H, 6.78 %N; found:
82.87 %C, 5.67 %H, 6.80 %N.
Acknowledgements
We thank the French ANR (PSICO project no. ANR-07-
BLAN-0281-01) and the Region Languedoc-Roussillon for
´
their financial support and for the award of a BDI thesis. We
also thank Dr William Douglas for scientific discussions and
technical assistance.
References
1 C. W. Tang and S. A. VanSlyke, Appl. Phys. Lett., 1987, 51, 913.
2 J. H. Burroughes, D. D. C. Bradley, A. R. Brown, R. N. Marks,
K. Mackay, R. H. Friend, P. L. Burns and A. B. Holmes, Nature,
1990, 347, 539.
3 D. F. O’Brien, M. A. Baldo, M. E. Thompson and S. R. Forrest,
Appl. Phys. Lett., 1999, 74, 442.
X-Ray diffraction
The diffraction intensities for silole 11 were collected at the
joint X-ray Scattering Service of the Institut Charles Gerhardt
and the Institut Europeen des Membranes of the University of
´
4 M. A. Baldo, S. Lamansky, P. E. Burrows, M. E. Thompson and
S. R. Forrest, Appl. Phys. Lett., 1999, 75, 4.
Montpellier II, France, at 175 K using an Oxford Diffraction
Xcalibur-I diffractometer. The structure was solved by
ab initio (charge-flipping) methods using SUPERFLIP54 and
refined by least-squares methods on F using CRYSTALS,55
against |F| on data having I 4 2s(I); R-factors are based on
these data. Hydrogen atoms were located from difference
Fourier synthesis. The CH2Cl2 crystallization molecule was
found to be heavily disordered. Two carbon positions and
seven chlorine positions were found. The total site occupancy
of the carbon positions was strongly restrained to 1 and that of
the chlorine positions to 2. Restraints were also put on the Uiso
parameters of carbon and chlorine in order to have them
approximately equal. No attempts were made to place the
proton sites for the two solvent carbon atoms. Basic structural
data for silole 11 C54H44Br2Si2ꢂ2(CH2Cl2); two solvent
hydrogens not placed: a = 13.8694(8) A, b = 14.3213(8) A,
c = 26.3741(14) A, a = 901, b = 98.256(4)1, g = 901,
V = 5184.4(5) A3, space group C21/c, 50 775 reflections
measured, 3634 independent reflections with I 4 2s(I) used
5 S. Lamansky, P. Djurovich, D. Murphy, F. Abdel-Razzaq,
H. E. Lee, C. Adachi, P. E. Burrows, S. R. Forrest and
M. E. Thompson, J. Am. Chem. Soc., 2001, 123, 4304.
6 S. Welter, K. Brunner, J. W. Hofstraat and L. De Cola, Nature,
2003, 421, 54.
7 C. Borek, K. Hanson, P. I. Djurovich, M. E. Thompson,
K. Aznavour, R. Bau, Y. Sun, S. R. Forrest, J. Brooks,
L. Michalski and J. Brown, Angew. Chem., Int. Ed., 2007, 46, 1109.
8 M. Cocchi, J. Kalinowski, D. Virgili, V. Fattori, S. Develay and
J. A. Williams, J. Appl. Phys., 2007, 90, 163508.
´
9 A. Fischer, S. Chenais, S. Forget, M.-C. Castex, D. Ades, A. Siove,
C. Denis, P. Maisse and B. Geffroy, J. Phys. D: Appl. Phys., 2006,
39, 917.
10 G. Gu, G. Parthasarathy and S. R. Forrest, Appl. Phys. Lett.,
1999, 74, 305.
11 M. Pfeiffer, K. Leo, X. Zhou, J. S. Huang, M. Hofmann,
A. Werner and J. Blochwitz-Nimoth, Org. Electron., 2003, 4, 89.
12 J. Dubac, C. Guerin and P. Meunier, in The Chemistry of Organic
Silicon Compounds, ed. Z. Rappoport and Y. Apeloig, Wiley,
Chichester, 1998, vol. 2, pp. 1961–2036.
13 J. Dubac, A. Laporterie and G. Manuel, Chem. Rev., 1990, 90, 215.
14 S. Yamaguchi and K. Tamao, J. Chem. Soc., Dalton Trans., 1998,
3693–3702.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 1290–1300 | 1299

View Article Online
15 J. Chen and Y. Cao, Macromol. Rapid Commun., 2007, 28,
1714.
16 S. Yamaguchi and K. Tamao, Bull. Chem. Soc. Jpn., 1996, 69,
2327.
17 N. Roques, P. Gerbier, U. Schatzschneider, J.-P. Sutter,
P. Guionneau, J. Vidal-Gancedo, J. Veciana, E. Rentschler and
37 W.-L. Jai, D.-R. Bai, T. McCormick, Q.-D. Liu, M. Motala,
R.-Y. Wang, C. Seward, Y. Tao and S. Wang, Chem.–Eur. J.,
2004, 10, 994.
38 M. Smet and W. Dehaen, Molecules, 2000, 5, 620.
39 M. Smet, J. Van Dijk and W. Dehaen, Tetrahedron, 1999, 55, 7859.
40 W.-L. Jia, D. Song and S. Wang, J. Org. Chem., 2003, 68, 701.
41 J. Lee, Q.-D. Liu, M. Motala, J. Dane, J. Gao, Y. Kang and
S. Wang, Chem. Mater., 2004, 116, 1869–1877.
42 N. Huby, L. Hirsch, L. Aubouy, P. Gerbier and A. van der Lee,
Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 75, 115416.
43 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter
Mater. Phys., 1988, 37, 785.
44 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski,
J. A. Montgomery, Jr., R. E. Stratmann, J. C. Burant,
S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin,
M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi,
R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford,
J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma,
D. K. Malick, A. D. Rabuck, K. Ragha-vachari, J. B. Foresman,
J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Ste-fanov, G. Liu,
A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Mar-tin,
D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara,
M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong,
J. L. Andres, C. Gonzalez, M. Head-Gordon, E. S. Replogle and
J. A. Pople, R. A. Gaussian 98, Gaussian, Inc., Pittsburgh, PA, 1998.
45 L. Aubouy, N. Huby, L. Hirsch, A. Van der Lee and P. Gerbier, to
be published.
C. Gue
18 N. Roques, P. Gerbier, J.-P. Sutter, P. Guionneau, D. Luneau and
C. Guerin, Organometallics, 2003, 22, 4833.
´
rin, Chem.–Eur. J., 2006, 12, 5547.
´
19 Cambridge Structural Database (CSD), 2003.
20 J. Chen, C. C. W. Law, J. W. Y. Lam, Y. Dong, S. M. F. Lo,
I. D. Williams, D. Zhu and B. Z. Tang, Chem. Mater., 2003, 15,
1535–1546.
21 Y. Ren, Y. Dong, J. W. Y. Lam, B. Z. Tang and K. S. Wong,
Chem. Phys. Lett., 2005, 402, 468.
22 H. Murata, Z. H. Kafafi and M. Uchida, Appl. Phys. Lett., 2002,
80, 189.
23 J. Chen, B. Xu, K. Yang, Y. Cao, H. H. Y. Sung, I. D. Williams
and B. Z. Tang, J. Phys. Chem. B, 2005, 109, 17086.
24 H. Y. Chen, W. Y. Lam, J. D. Luo, Y. L. Ho, B. Z. Tang,
D. B. Zhu, M. Wong and H. S. Kwok, Appl. Phys. Lett., 2002, 81,
574.
25 L. Aubouy, P. Gerbier, N. Huby, G. Wantz, L. Vignau, L. Hirsch
and J.-M. Janot, New J. Chem., 2004, 28, 1086.
26 H. Murata, G. G. Malliaras, M. Uchida, Y. Shen and Z. H. Kafafi,
Chem. Phys. Lett., 2001, 339, 161–166.
27 N. Huby, G. Wantz, L. Hirsch, L. Vignau, A. S. Barriere,
L. Aubouy and P. Gerbier, J. Appl. Phys., 2006, 99, 084907.
28 G. Yu, S. Yin, Y. Liu, J. Chen, X. Xu, X. Sun, D. Ma, X. Zhan,
Q. Peng, Z. Shuai, B. Tang, D. Zhu, W. Fang and Y. Luo, J. Am.
Chem. Soc., 2005, 127, 6335.
29 J. Pang, Y. Tao, S. Freiberg, X.-P. Yang, M. D0Iorio and S. Wang,
J. Mater. Chem., 2002, 12, 206.
46 H. Du, R. A. Fuh, J. Li, A. Corkan and J. S. Lindsey, Photochem.
Photobiol., 1998, 68, 141.
47 S. Yamaguchi and K. Tamao, Bull. Chem. Soc. Jpn., 1996, 69,
2327.
´
48 C. Risko, G. P. Kushto, Z. H. Kafafi and J. L. Bredas, J. Chem.
Phys., 2004, 121, 9031.
30 J. Lee, Q.-D. Liu, D.-R. Bai, Y. Kang, Y. Tao and S. Wang,
Organometallics, 2004, 23, 6205.
31 Z. H. Li, M. S. Wong, Y. Tao and H. Fukutani, Org. Lett., 2007, 9,
3659.
49 Y. Shirota and H. Kageyama, Chem. Rev., 2007, 107, 953.
50 K. Walzer, B. Maennig, M. Pefeiffer and K. Leo, Chem. Rev., 2007,
107, 1233.
32 K. Danel, T. H. Huang, J. T. Lin, Y. T. Tao and C. H. Chuen,
Chem. Mater., 2002, 14, 3860.
51 L. Aubouy, P. Gerbier, C. Guerin, N. Huby, L. Hirsch and
´
L. Vignau, Synth. Met., 2007, 157, 91.
33 P. Raghunath, M. A. Reddy, C. Gouri, K. Bhanuprakash and
V. J. Rao, J. Phys. Chem. A, 2006, 110, 1152–1162.
34 S. Yamaguchi, T. Endo, M. Uchida, T. Izumizawa, K. Furukawa
and K. Tamao, Chem.–Eur. J., 2000, 6, 1683–1692.
35 S. Murphy, X. Yang and G. B. Schuster, J. Org. Chem., 1995, 60,
2411.
36 F. Kong, X. L. Wu, G. S. Huang, R. K. Yuan, C. Z. Yang,
P. K. Chu and G. G. Siu, Appl. Phys. A: Mater. Sci. Process., 2006,
84, 203–206.
52 V. Coropceanu, J. Cornil, D. A. da Silva Filho, Y. Olivier,
R. Silbey and L.-L. Bredas, Chem. Rev., 2007, 107, 926.
53 F. Wang, J. Luo, J. Chen, F. Huang and Y. Cao, Polymer, 2005,
46, 8422.
54 L. Palatinus and G. Chapuis, J. Appl. Crystallogr., 2007, 40, 786.
55 P. W. Betteridge, J. R. Carruthers, R. I. Cooper, K. Prout and
D. J. Watkin, J. Appl. Crystallogr., 2003, 36, 1487.
56 A. V. Dijken, A. Perro, E. A. Meulenkamp and K. Brunner, Org.
Electron., 2003, 4, 131–141.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
1300 | New J. Chem., 2009, 33, 1290–1300