Perylenediimide-based donor-acceptor dyads and triads: Impact of molecular architecture on self-assembling properties

Article abstract of DOI:10.1021/ja4129108

Perylenediimide-based donor-acceptor co-oligomers are particularly attractive in plastic electronics because of their unique electro-active properties that can be tuned by proper chemical engineering. Herein, a new class of co-oligomers has been synthesized with a dyad structure (AD) or a triad structure (DAD and ADA) in order to understand the correlations between the co-oligomer molecular architecture and the structures formed by self-assembly in thin films. The acceptor block A is a perylene tetracarboxyl diimide (PDI), whereas the donor block D is made of a combination of thiophene, fluorene, and 2,1,3-benzothiadiazole derivatives. D and A blocks are linked by a short and flexible ethylene spacer to ease self-assembling in thin films. Structural studies using small and wide X-ray diffraction and transmission electron microscopy demonstrate that AD and ADA lamellae are made of a double layer of co-oligomers with overlapping and strongly π-stacked PDI units because the sectional area of the PDI is about half that of the donor block. These structural models allow rationalizing the absence of organization for the DAD co-oligomer and therefore to draw general rules for the design of PDI-based dyads and triads with proper self-assembling properties of use in organic electronics.

Full text of DOI:10.1021/ja4129108

Article
pubs.acs.org/JACS
Perylenediimide-Based Donor−Acceptor Dyads and Triads: Impact of
Molecular Architecture on Self-Assembling Properties
Pierre-Olivier Schwartz,† Laure Biniek,‡ Elena Zaborova,§ Benoît Heinrich,† Martin Brinkmann,‡
Nicolas Leclerc,§ and Stephane Mery*,†
́
́
†Institut de Physique et de Chimie des Mater
43, 67034 Strasbourg, Cedex 2, France
‡Institut Charles Sadron, UPR22 CNRS, 23 rue du Loess, BP 84047, 67034 Strasbourg Cedex 2, France
§Institut de Chimie et Proced
́
es pour l'Energie, l'Environnement et la Sante, UMR 7515, CNRS, Universite de Strasbourg, ECPM, 25
́ ́
iaux de Strasbourg, UMR 7504, CNRS, Universite de Strasbourg, 23 rue du Loess, BP
́
́
́
rue Becquerel, 67087 Strasbourg Cedex 2, France
S
* Supporting Information
ABSTRACT: Perylenediimide-based donor−acceptor co-
oligomers are particularly attractive in plastic electronics because
of their unique electro-active properties that can be tuned by
proper chemical engineering. Herein, a new class of co-oligomers
has been synthesized with a dyad structure (AD) or a triad
structure (DAD and ADA) in order to understand the correlations
between the co-oligomer molecular architecture and the structures
formed by self-assembly in thin films. The acceptor block A is a
perylene tetracarboxyl diimide (PDI), whereas the donor block D
is made of a combination of thiophene, fluorene, and 2,1,3-
benzothiadiazole derivatives. D and A blocks are linked by a short
and flexible ethylene spacer to ease self-assembling in thin films.
Structural studies using small and wide X-ray diffraction and
transmission electron microscopy demonstrate that AD and ADA lamellae are made of a double layer of co-oligomers with
overlapping and strongly π-stacked PDI units because the sectional area of the PDI is about half that of the donor block. These
structural models allow rationalizing the absence of organization for the DAD co-oligomer and therefore to draw general rules for
the design of PDI-based dyads and triads with proper self-assembling properties of use in organic electronics.
(macro)molecular systems e.g., block copolymers8 or co-
1. INTRODUCTION
oligomers9,10 (including low-molecular weight com-
Nanostructured 3D morphologies composed of electronically
pounds)11−13 and (ii) physicochemical methods that use
active constituents have gained interest in the field of organic
electronic.1−3 In particular organic solar cells with a well-
specific thin film processing methods e.g. nanoimprint
lithography14 or template-based methods.15 While the latter
ordered phase-separated donor and acceptor morphology, i.e. a
so-called nanostructured D−A bulk heterojunction has been
proposed as an “ideal structure”, leading to OPV devices with
methods seem very promising in terms of cost, reproducibility
and large-scale integration, exploiting the self-assembling of
enhanced efficiency and stability.4,5 Generating sub-10 nm-
sized donor and acceptor domains with a large interfacial area is
expected to ease exciton separation, whereas a bicontinuous
“percolation” network of the two components may improve
charge-carrier conduction and extraction from the active layer
to the electrodes.6 Nanometer-scaled domains are also required
because of the limited exciton diffusion length in organic thin
films.7 Ordered D and A arrays are also promising candidates
for organic field effect transistors (OFETs) devices, since they
could intrinsically exhibit an ambipolar charge transport
through a bicontinuous p-type and n-type network. Various
strategies were proposed to prepare thin films that fulfill all
these requirements. They can be classified into two major
categories: (i) methods based on chemical engineering that
exploit the self-assembling properties of donor−acceptor
precisely designed donor−acceptor block co-oligomers is of
particular interest as the preparation methods may be facilitated
by the intrinsic self-assembling properties of the co-oligomers.
Several teams have investigated the donor−acceptor block
(D−A) copolymer approach implying the synthesis of a
macromolecule where an electron donor and an electron
acceptor block are chemically bound.16 For many copolymers,
the acceptor block was made of pendant fullerene C60.
However, in the past decade, perylenediimide (PDI) appeared
as an interesting alternative acceptor unit to C60 in OPV
devices.8d−h Many macromolecular architectures using PDI
were synthesized and interesting properties such as n-type
Received: January 7, 2014
Published: March 27, 2014
© 2014 American Chemical Society
5981
dx.doi.org/10.1021/ja4129108 | J. Am. Chem. Soc. 2014, 136, 5981−5992

Journal of the American Chemical Society
Article
Figure 1. Chemical structure of the donor−acceptor block co-oligomers categorized into the three molecular architectures: ADn, DnADn and ADA.
transport and OPV activity were demonstrated.17 In particular,
D−A copolymers based on perylenediimides have recently
gained high interest because of the combination of remarkable
self-assembling properties along with a strong n-type charge
transport.8g,h More generally, D−A copolymers have been used
as active layers as well as stabilizers of donor/acceptor blends,
but lead to rather limited charge transport and photovoltaic
properties in thin films.18
molecular weight systems. The aim of this contribution is to
investigate the structure−property relationship of these
materials with particular emphasis on spectroscopic, thermal,
and structural properties. A careful look will be brought to the
molecular architecture dependence of the material organization,
by means of X-ray diffraction studies (powder and GIWAXS)
and electron microscopy investigation (TEM).
One possible way to improve the block copolymer approach
consists in the design of co-oligomers with a precise
composition and molecular architecture avoiding, in particular,
the issues of polydispersity and chemical purity inherent to
copolymer. Recently, Y. Geng et al. demonstrated interesting
OPV properties in thin film devices prepared from PDI-based
oligomers forming lamellar mesophases with a perfect phase
separation between donor and acceptor groups.10c,d The quality
of the organization was found to be significantly improved by
increasing the donor block length and by using solvent-vapor
annealing to improve the film morphology. In these reports,
however, only a series of oligomers of dyad architecture (AD)
was investigated. Derived from the block co-oligomer approach,
one has to mention the large number of low-molecular weight
systems reported in the literature which are made of a “small”
donor conjugated unit (<800 Da) covalently linked to a C60,11
peryleneimide12,13 or other acceptor unit,19 into dyad or triad
architectures. Only a few systems, however, could successfully
lead to microphase separation of the blocks clearly leading to
self-assembling into 3-D supramolecular D/A arrays.11b−d,13,18
Liquid crystals seem to constitute a promising approach in that
direction.10c,d,11b,c,13a,g With low-molecular weight systems,
however, the small size of the electro-active species does
represent a challenge to build up well-defined D and A channels
over large-length scales to ease ambipolar charge trans-
port.11b,c,12f,13
2. RESULTS AND DISCUSSION
Molecular Design and Synthesis. We have designed new
molecular structures based on the approach of monodisperse
block co-oligomers including both electron-acceptor (A) and
electron donor blocks (D). Various molecular architectures e.g.
AD, DAD and ADA (Figure 1) are prepared by assembling
properly these two blocks.
The studied co-oligomers are made of the association of
several chemical units: perylene tetracarboxyl diimides (PDI),
thiophene, fluorene, and 2,1,3-benzothiadiazole derivatives
(Figure 1). PDI presents an n-type character (A), while the
other units compose the p-type block (D). PDI is made of a
large aromatic macrocycle providing strong π−π interactions
that is the primary force to drive the formation of stacked
structures.20 The thienofluorene building block can be seen as a
rigid conjugated rod, which incidentally showed its ability to
produce mesophases when substituted by appropriate alkyl
chains.21 Its association with PDI was therefore expected to
promote the formation of a lamellar organization with
segregated D and A domains. This assumption was recently
demonstrated in the case of the AD molecular architecture.10c,d
The electron-deficient thiophene−benzothiadiazole−thiophene
(TBzT) moiety has been included in the donor block with the
intent of lowering the energy bandgap of the donor block by
decreasing its LUMO level and red-shifting its absorption
spectrum.22 Finally, the covalent bonding of the donor block by
N-substitution of the PDI core ensures an electronic
disconnection between both blocks. In our systems, a
nonconjugated ethylene linker is inserted between the D and
A blocks to impede the direct charge recombination.13c It is
also expected to bring some flexibility between both blocks to
Herein we report the syntheses and the physicochemical
properties of a series of donor−acceptor block co-oligomers of
different molecular architectures within dyad (AD) and triad
(ADA and DAD) systems. The donor block length has been
only slightly varied to remain in the range of moderate
5982
dx.doi.org/10.1021/ja4129108 | J. Am. Chem. Soc. 2014, 136, 5981−5992

Journal of the American Chemical Society
Article
Scheme 1. Synthetic Pathway for the Preparation of the Amine-Ended Donor Blocks (D0, D1, and D) with Their Subsequent
Coupling to the PDI unit, To Provide the Donor−Acceptor Block-co-Oligomers of Different Molecular Architectures (ADn,
DnADn and ADA)
favor their respective packing. Moreover, this linker is expected
to be short enough to prevent the D and A backfolding, already
noticed for other systems with longer linkers,23 which indeed is
detrimental for the formation of nanostructures with segregated
D and A domains.
connected to the carboxylic anhydride functionality of the PDI
acceptor block, via an imidation reaction, to prepare the final
dyads and triads. The synthetic routes are outlined in Scheme
1, while all detailed procedures are given in Supporting
Information (SI).
Finally the choice of the solubilizing side chains, necessary to
process these materials from solution, is also extremely
influential for the material organization and consequently the
optoelectronic properties.24 In our case, we decided to use
linear octyl chains on the fluorene moieties and 2-ethylhexyl-
ramified chains on the PDI cores (when asymmetrized) and on
thiophene units adjacent to the benzothiadiazole. Ethylhexyl
side chains were chosen because of their higher solubilizing
capability with respect to that of their linear counterparts.
All oligomers were synthesized using the same two-step
strategy (Scheme 1). First, the whole donor blocks bearing an
amine group at one end (for the AD and DAD architectures) or
at the two ends (for the ADA architecture) were synthesized by
using a convergent approach. These donor blocks were then
For the AD and DAD architecture, the donor block length
was slightly varied by adding a thiophene−di(octyl)fluorene−
thiophene segment to the starting compounds AD0 and D0AD0,
as to obtain the AD1 and D1AD1 co-oligomers, respectively.
The lengthening of the donor block was not attempted on the
ADA triad because of its poor solubility and processability due
to the presence of two highly aggregating PDI units.
The electron-acceptor blocks 3 of the ADA triad and ADn
dyads are similar and required a previous asymmetrization step
which was carried out according to the work of Bock et al.25 In
the case of the DnADn triad, the commercially available
perylene-3,4,9,10-tetracarboxylic dianhydride (PTCDA) 1 was
directly used.
5983
dx.doi.org/10.1021/ja4129108 | J. Am. Chem. Soc. 2014, 136, 5981−5992

Journal of the American Chemical Society
Article
Scheme 2. Chemical Structure of the Dyad AD0T
Figure 2. Normalized UV−vis absorption spectra of the co-oligomers in CHCl3 (dashed line), in thin film (plain black line), and in post-annealed
film (gray line).
The synthesis of electron-donor blocks is identical for that of
the DnADn and ADn series (Scheme 1). It starts with the
synthesis of the electron-deficient TBzT derivative 15 bearing
at one end a phthalimide moiety, precursor to the amine
functionality and at the other end, a bromine atom for
subsequent adjunction of a thienofluorene moiety. This
intermediate compound (15) was obtained from successive
reaction steps, involving a Stille cross-coupling reaction
between N-bromophenethyl(phthalimide) 12 and the stanny-
lated TBzT derivative 13,23 and then a bromination using NBS.
All reaction steps were achieved in rather high yield except the
Stille cross-coupling reaction (∼60% yield), most likely due to
the steric hindrance caused by the ethylhexyl chain adjacent to
the stannylated position in 13. In a next step, 15 was cross-
coupled with the thienofluorene derivative 7 or 11 to produce
the phthalimide-ended donor blocks 16 or 18 of different
length (n = 0 or 1, respectively) in 78−90% yield. The latter
were obtained from multiple reaction steps (85−95% yield
each), involving trimethylstannylation (via either a H/Li or Br/
Li exchange), bromination (NBS), and Stille cross-coupling
reactions. The whole amino-ended donor blocks 17 (D0) and
19 (D1) were finally obtained after cleavage of the phthalimide
protecting group by hydrazinolysis in quantitative yield. These
donor blocks were used without further purification for the last
imidation step with the perylene carboxylic anhydride 1 or
dianhydride 3 to provide the targeted dyads (AD0 and AD1) or
triads (D0AD0 and D1AD1), respectively. These final reactions
were carried out in quinoline at 200 °C in the presence of zinc
acetate. Several purification steps by column chromatography
on silica gel (petroleum ether/CH2Cl2) were required to afford
pure oligomers. They were obtained in good yields (around
80%) as purple solids and characterized by NMR and Maldi-
TOF analyses.
In the case of the ADA triad, we have synthesized a specific
donor block composed of two amino-functionalized TBzT
moieties 22 (D). This donor block synthesis was performed by
using a Stille cross-coupling reaction between the TBzT
derivative 15 and the bis-stannylated thienofluorene 20,26
followed by a cleavage (hydrazinolysis) of the phthalimide
protecting groups. The final ADA triad was obtained from a
double imidation reaction between the bis-amino derivative 22
and the perylene dicarboxylic anhydride 1 with a limited yield
of 26%. This low yield is ascribed to the increased amount of
byproducts and the poor solubility of the final compound.
Proper purification by several column chromatographies on
5984
dx.doi.org/10.1021/ja4129108 | J. Am. Chem. Soc. 2014, 136, 5981−5992

Journal of the American Chemical Society
Article
Table 1. Optical, Electrochemical and Thermal Characterization Data of the Co-oligomers
E
Eox Onset D
Ered Onset A
HOMO D
LUMO A
(eV)
Tg
T
degd
g opat
b
b
b
c
c
Oligomers
(eV)
(V)
(V)
(eV)
(°C) Tiso /°C (ΔH/J·g−1
)
Tc /°C (ΔH/J·g−1
)
(°C)
AD0
1.97
1.96
1.96
1.97
1.96
0.79
0.77
0.94
0.73
0.74
−0.50
−0.50
−0.85
−0.50
−0.50
−5.54
−5.52
−5.69
−5.48
−5.49
−4.25
−4.25
−3.90
−4.25
−4.25
−
−
−
223 (17.97)
205 (11.42)
269 (17.25)
−
209 (18.91)
194 (11.84)
240 (15.61)
−
405
350
320
330
400
AD1
ADA
D0AD0
D1AD1
38
53
−
−
a
b
Optical band gap estimated from the absorption onset of thin films. Oxidative (Eox and HOMO) and reductive (Ered and LUMO) contribution
from the donor and acceptor block, respectively. All measurements were made in solution (CH2Cl2), except for ADA (solid film). Onset
isotropization and crystallization temperatures. Calculated from the inflection point measurement.
c
d
silica gel (chloroform/methanol) provided the purple solid of
ADA of good purity.
rise to a typical emission band centered at about 650 nm. More
remarkable is the almost complete fluorescence quenching
observed whatever the excitation wavelength selected. Thus,
upon excitation at 530 and 570 nm, an efficient quenching of
fluorescence was observed (Figures S1−S4 in SI). At such
wavelengths, only the electron-deficient benzothiadiazole-based
part of the donor block and the PDI acceptor block is
absorbing. These results point at an efficient exciton
dissociation in those block co-oligomers, even in diluted
solution.
In order to evaluate the role of the thienofluorene unit on the
molecular organization, another dyad (AD0T) was also
synthesized in which the thienofluorene was replaced by a
terthiophene unit (Scheme 2). Both units similarly contain two
linear octyl side chains. The synthetic pathway is reported in SI
(Schemes S6 and S7). On the basis of the thin film morphology
analysis performed by TEM, the structural organization of
AD0T will be compared with the one of AD0.
Spectroscopic and Redox Properties. The UV−visible
absorption spectra of the co-oligomers recorded in solution (in
chloroform) and in thin films (doctor-bladed from solution at
45 °C) before and after post-annealing treatment (close to the
melting) are presented in Figure 2. In solution, the absorption
spectra of all co-oligomers are dominated by the characteristic
vibronic structure of the PDI core with components at 465,
495, and 530 nm. The donor contribution is represented by
two bands, one in the range 325−425 nm, whose position is
correlated to the donor length conjugation, and the second one,
centered at about 450 nm which extends beyond the absorption
range of PDI, to 600 nm. This latter contribution can be
correlated to the electron-deficient benzothiadiazole-based part
of the donor group (see Figure S1 in SI). In thin films, the
contribution of the PDI block broadens as a consequence of the
excitonic coupling in the solid state with an additional red-
shifted contribution seen as a shoulder at λ ≈ 600 nm for AD0.
After annealing of the films at a temperature close to the
isotropization temperature, a clear variation of the fine structure
of the absorption of the PDI block is observed for all the
oligomers, except for D0AD0 which remains unchanged
whatever the thermal post-treatment applied. This is in
agreement with the DSC results showing a clear first-order
phase transition for AD0, AD1, and ADA upon cooling from the
melt, whereas such transition is observed neither for D0AD0 nor
for D1AD1 (vide infra).
As a first investigation of the excited states, the emission
spectra in solution (10−6 mol L−1) were recorded for one
representative material, the dyad AD1 (Figure S2 in SI). Upon
excitation of AD1 at 400 nm, a quasi-quantitative quenching of
fluorescence is observed. This implies the exciton dissociation
via a charge transfer from the donor to the acceptor block and
no radiative recombination of separated charges. This implies
also that the (−(CH2)2−) linker does not prevent the
intramolecular charge transfer, contrary to previous observa-
tions in another series of block co-oligomers made of hexa-peri-
hexabenzocoronene and PDI units.13c Similar behavior was
observed for all our synthesized co-oligomers (Figures S1−S4
in SI). By comparison, excitation of the isolated donor block
(not connected to A) at the same wavelength (400 nm), gives
The electrochemical properties of the co-oligomers were
probed by cyclic voltammetry. Ferrocene was used as internal
standard to convert the values obtained with Ag/Ag+ reference
to the saturated calomel electrode scale (SCE). In order to have
an insight into the intrinsic electrochemical properties of our
D−A block co-oligomers, the analyses were carried out in
solution (methylene chloride) for all materials, except for ADA
molecule. Due to its poor solubility in most organic solvents,
ADA was only analyzed in the solid state, using acetonitrile as
electrolytic support. Cyclic voltammograms are presented in
the SI (Figures S5, S6, and S7), while the values are reported in
Table 1.
All materials exhibited well-reversible oxidative and reductive
processes. In solution, they displayed two successive reversible
monoelectronic reduction waves around −0.50 V and −0.70 V,
respectively, that are classically found for PDI (Figures S5 and
S6 in SI). In the case of ADA in the solid state, the two-electron
reduction occurs in a single wave at −0.85 V (Figure S7 in SI).
It is not a particular feature since all our block co-oligomers
show similar behaviors in the solid state. By using the NHE
formal potential of −4.75 eV, LUMO levels of PDI blocks were
calculated at −4.25 eV for ADn dyads and DnADn triads in
solution, and at −3.90 eV for ADA triad in solid state. No
oxidation of the PDI core could be observed in the range of
measured voltages. Regarding the electron-donor blocks, we
were able to determine the oxidative and reductive processes
for all molecules. Similarly to the LUMO level of PDI, minor
differences were observed in the HOMO levels of all molecules
in solution. They all exhibited two oxidation waves in the range
0.7−1.1 V, corresponding to extracted HOMO levels ranging
from −5.48 to −5.54 eV. The cathodic current increase
observed for DnADn triads in comparison with ADn dyads
confirms the bielectronic nature of those oxidative waves. This
feature is the result of the presence of two decoupled electron
donor blocks in one single DnADn molecule. For all molecules,
we were also able to measure a reduction process around −1.2
V, corresponding to the electron-donor blocks D0 and D1.
Electrochemical band gaps of 1.9−2.0 eV were thus obtained
for both donor blocks in solution, in good agreement with the
band gaps measured in solid state for the pure electron-donor
5985
dx.doi.org/10.1021/ja4129108 | J. Am. Chem. Soc. 2014, 136, 5981−5992

Journal of the American Chemical Society
Article
blocks. Regarding the electron-donor block of the ADA triad in
solid state, similar voltammograms could be recorded with two
oxidation waves and one reduction wave (Figure S7 in SI). Due
to a shift of the first oxidation wave toward the higher
potentials, a higher electrochemical band gap of 2.1 eV has
been calculated. Although the block co-oligomers presented
structural variations, the HOMO and LUMO levels were not
significantly changed which implies that the imide group
together with the flexible (−(CH2)2−) linker between donor
and acceptor blocks ensures adequate decoupling of electronic
states of the blocks. This is in agreement with the absence of
charge transfer (CT) band for co-oligomers in UV−vis spectra
(diluted solution and thin films).
Thermal and Structural Properties. The thermal proper-
ties of the molecules have been characterized by thermogravi-
metric analysis (TGA), differential scanning calorimetry
(DSC), and polarized optical microscopy (POM). First, all
materials show a good thermal stability with a temperature of
degradation occurring at Tdeg ≥ 320 °C (see Figures S8−S12 in
SI and Table 1). As shown in Figure 3, the DSC thermograms
amorphous-like character of D0AD0 and D1AD1 is confirmed by
POM observation from the absence of birefringence in the
whole temperature range explored from 20 to 300 °C.
In the pristine state, the five compounds give rise to small-
angle X-ray scattering patterns characteristic of a segregated
structure (see Figure S13 in SI). More exactly, the patterns
contain a broad wide-angle scattering maximum at about 1.5
Å−1 (∼4.5 Å) that overlaps a semi-broad scattering (∼3.5 Å)
due to distances between stacked PDI groups. The four
compounds (ADn and DnADn series) exhibit, in the small and
medium angle region, several broad reflections from an
embryonic structure resulting from the microsegregation in
separate domains of the lateral chains, the donor rigid parts and
the PDI unit. For the triad ADA, the segregation is enhanced
with respect to the dyads and leads to positional long-range
order as indicated by a set of sharp reflections.
Most importantly, thermal annealing the oligomers is found
to improve substantially the ordering of AD and ADA. Figure 4
Figure 3. DSC thermograms of the co-oligomers (heating and cooling
rates of 10 and 5 °C·min−1, respectively).
Figure 4. X-ray diffraction patterns of the series of co-oligomers
obtained at room temperature after thermal annealing above 170 °C.
show a first transition peak upon heating and cooling only for
AD0, AD1, and ADA compounds, while no transition is
detected for D0AD0 or D1AD1 besides a glass transition
temperature (Tg). The peaks observed upon heating corre-
spond to the transition to the isotropic liquid state (Tiso). These
Tiso values are close for AD0 and AD1 (223 and 205 °C,
respectively), while ADA exhibits the highest temperature (Tiso
= 269 °C). The weak supercooling effect (Tiso − Tc < 30 °C)
associated with the low enthalpy value of the transitions (<20 J·
g−1) suggested that the low-temperature phase in ADn and
ADA is a mesophase rather than a pure crystalline phase. This
assumption is corroborated by POM observations from the
pasty character of the birefringent texture found for ADn and
ADA materials. The mesomorphic nature of these compounds
is further confirmed using X-rays analyses. In contrast, the
shows the characteristic powder X-ray diffractograms of the five
oligomers (AD0, AD1, D0AD0, D1AD1, and ADA) recorded at
room temperature after a melting or heating above 170 °C.
Except for the DAD-based triads, all patterns exhibit the same
sequence of (0 0 l) reflections at low angle corresponding to
lamellar mesophases with long-range order. Lamellar periods
for AD0, AD1, and ADA after annealing are 70.8, 96.5, and 58.8
Å, respectively. Further evidence of lamellar mesophases after
melt-cooling is obtained by TEM (Figure 5 and Figure 7). In
addition, all three co-oligomers (AD0, AD1, and ADA) show a
set of at least five well-defined peaks with almost identical
positions and intensities in the range 0.28 Å−1 < q < 1.85 Å−1.
This sequence of peaks characterizes the molecular packing of
co-oligomers within the lamellae. More specifically, the quite
broad peak at 1.80−1.88 Å−1 corresponds to the π-stacking of
5986
dx.doi.org/10.1021/ja4129108 | J. Am. Chem. Soc. 2014, 136, 5981−5992

Journal of the American Chemical Society
Article
Figure 5. ED patterns (a and d), two-dimensional grazing incidence X-ray diffraction (GIWAXS) intensity maps (b and e), and TEM-BF images (c
and f) of AD0 (a,b,c) and AD1 (d,e,f) thin films after annealing above the melting temperature. The insets represent the dominant orientation of the
lamellae on the SiO2 substrate. The inset in (c) highlights a set of standing lamellae also shown in an enlarged view in the lower left corner.
the PDI blocks, hPDI (maximum at 3.5 Å).27 The broad
scattering maximum at 1.5 Å−1 comes from lateral distances
between molten alkyl side chains (hch) and likely between rigid
moieties of donor blocks segments (hDr). Additional peaks
correspond to the reflections of the three-dimensional structure
emerging from the segregation of alkyl chains and donor
blocks.
Contrarily to ADn and ADA compounds, both DAD
derivatives exhibit, even after annealing, a poorly defined
pattern (Figure 4). The latter is dominated by a slightly
broadened peak corresponding to lamellar periods of 56 and 76
Å for D0AD0 and D1AD1, respectively, besides two broader
peaks at 0.31 and 1.5 Å−1 coming from lateral distances
between chains and rigid moieties. The absence of a clear hPDI
peak combined to the persistence of a quite sharp lamellar
period (DLong) shows that the layers of parallel packed
molecules are preserved, but without the regular PDI stacks
and only over intermediate correlation ranges (see Table 2).
Finally, the lack of long-range order in D0AD0 and D1AD1
results in POM observation of textures, typical of an isotropic
liquid, i.e. without birefringence in the liquefied-cooled thin
films. In contrast to DAD-based triads, the ADA triad forms a
lamellar mesophase already in the pristine state. However, this
pristine lamellar structure differs from that obtained after melt-
cooling (see Figure S14 in SI and Table 2). In particular, the
lamellar period is 30% smaller in the pristine state. Above 100
°C the pristine structure gradually rearranges toward the stable
lamellar mesophase, which is also the structure obtained in
melt-cooled thin films. The sequence of peaks in the range 0.28
Å−1 < q < 1.85 Å−1 observed for ADA after annealing is almost
identical to that observed for the AD0 and AD1 dyads,
indicating a strong similarity in the molecular packing of the
ADA triad and the ADn compound series.
5987
dx.doi.org/10.1021/ja4129108 | J. Am. Chem. Soc. 2014, 136, 5981−5992

Journal of the American Chemical Society
Article
a
Table 2. Mesomorphism and Structural Parameters of the Compound Series Annealed above 170 °C for about 1 h
molec areas
d
e
cmpd name
molecular length (Å)
phase and lattice parameters
ρ (g·cm−3
)
AD
APDI
b
orthorhombic phase
a ; b ; c (Å)
V (Å)3 (Z)
AD0
50
66
86
86
35.5 ; 7.43 ; 70.6
35.4 ; 7.44 ; 96.3
35.2 ; 7.45 ; 117.8
29.4 ; 12.7 ; 81.2
18600 (Z = 8)
25400 (Z = 8)
30800 (Z = 8)
1.14
1.13
1.21
1.23
65.9
65.8
65.4
94
33.0
32.9
32.7
47
AD1
ADA
ADA (prist)
30400 (Z = 8)
c
isotropic-like phase
D1at (ξ) (Å)
21 (30)
DLong(ξ) (Å)
56 (160)
D0AD0
D1AD1
84
1.17
1.14
66
71
66
71
116
18 (50)
76 (400)
a
b
The structural parameters for both dyads and ADA triad have been determined considering the model proposed in Figure 8b and d. a, b, c: lattice
parameters; V: lattice volume; Z: number of molecules per lattice; ρ: density extracted from the unit cell parameters or calculated from reference
c
density measurements (in italics). DLong: lamellar period; D1at: periodicity from lateral packing; ξ: correlation length in Å extracted from the Scherrer
d
formula. AD: molecular area per donor group (i.e., the area of the layer portion covered by a single group), as obtained from the ratio of molecular
e
volume and either the parameter c or DLong. APDI: molecular area per PDI group.
Thin Films Morphology and Structure. TEM inves-
tigation of AD0 and AD1 dyads as-cast films do not give any
indication for long-range order or structure. The situation
changes upon melt-cooling of the samples. Figure 5 depicts the
morphology and the ED and GIWAXS for thin films of AD0
and AD1 dyads after melt-cooling. The films of AD0 consist of a
majority of flat-lying lamellae with a typical terraced
morphology. However, a small fraction of standing lamellae
are also observed, and these areas give rise to a periodic fringed
pattern characteristic of the lamellar mesophase with a period l
= 70.6 Å. For the analogous AD0T dyad, shown in Scheme 2, a
similar terraced morphology was also evidenced by TEM and
ED after melt-cooling (Figure 6). In strong contrast, the films
0 0) reflections with a weak (0 0 3) contribution for AD0.
Instead, for AD1, the (0 0 l) reflections are dominant. Along qz,
the situation is reversed: for AD0, the presence of both (h 0 0)
and (0 0 l) reflections along qz confirm the coexistence of both
flat-on and edge-on oriented lamellae. For AD1, the (h 0 0)
reflections are very strong without any contribution from the (0
0 l) reflections along qz.
The difference in molecular orientation in dyad films
observed as a function of the donor block length has important
implications regarding potential devices properties. Accord-
ingly, in melt-cooled films, one can expect edge-on oriented
lamellae with lying molecules of AD1 to be more favorable for
OPV applications with respect to AD0 and AD0T films
consisting of terraced domains with standing molecules.
As previously demonstrated by XRD on pristine powders
(Figure S14 in SI), the ADA triad presents enhanced ordering
with a well-defined lamellar mesophase. This is confirmed by
TEM observations on as-cast films prepared by doctor-blading.
It is worth mentioning that for ADA triad, doctor-blading has
been realized at 180 °C for solubility reasons. In this condition,
a clear segregation of donor and acceptor blocks leads to a
lamellar mesophase (Figure 7a). However, as seen in Figure 7c,
after melt-cooling, the dimension of the domains is substantially
increased and the contrast/sharpness of the lamellar morphol-
ogy is strongly enhanced. The extension of long-range lamellar
order is also evidenced by a strong increase of the intensity of
the (0 0 l) reflections in the corresponding ED pattern (see
Figure 7b and d).
Figure 6. (a) ED pattern and (b) terrace-like morphology observed by
BF-TEM of AD0T thin film after annealing above the melting state.
of AD1 consist essentially of standing lamellae as indicated by
the observation of a periodic fringed pattern. The contrast in
the TEM-BF images (Figure 5f) is attributed to the difference
between the packing density of PDI and of the donor blocks.
The sharp and dark lines are attributed to dense layers of PDI
blocks, whereas the brighter zones correspond to the layers of
donor blocks. The difference in lamellar orientation for AD0
(and AD0T) and AD1 indicates that, for short donor blocks, the
dyad molecules can stand up on the substrate upon annealing,
whereas for AD1 the molecular long axis remains in the plane of
the substrate. These differences in morphology are confirmed
by ED. Indeed, the characteristic (0 0 l) reflections are only
seen in the case of AD1, whereas for AD0 (and AD0T) the ED
pattern shows exclusively the (h 0 0) reflections. The same
analysis can be conducted for the GIWAXS patterns. If we
consider the in-plane reflections (along qx), one observes the (h
In strong contrast to ADn and ADA compounds, TEM
investigations on D0AD0 and D1AD1 films did not show any
evidence of long-range-ordering independently of the thermal
annealing (see Figure S15 in SI).
Correlation between Molecular Packing and Molec-
ular Architecture. The strong similarity in the XRD patterns
of dyads based on different donor blocks (different lengths and
chemical compositions) indicates that the self-organization of
dyads into lamellar mesophases is mainly determined by the
packing of the PDI blocks. Let us consider first the case of AD0.
For AD0, the lamellar period lies between 1 and 2 times the
length of the extended molecule. This suggests strongly that the
lamellae involve two overlapping dyad molecules. This overlap
may concern either the PDI block or the donor block. To
discriminate between these two possibilities, we have to
5988
dx.doi.org/10.1021/ja4129108 | J. Am. Chem. Soc. 2014, 136, 5981−5992

Journal of the American Chemical Society
Article
Figure 7. Comparison of TEM-BF images and low-angle ED patterns
of as-cast (a and b) and annealed (c and d) ADA thin films.
consider the sectional areas of both blocks. The partial volumes
of the donor and acceptor segments can be deduced from
reference data and complementary measurements. They show
that the sectional area of the donor groups including the molten
lateral chains is at least 55 Å2. This section largely exceeds the
one of the stacked PDI of ∼28.5 Å2 as obtained from reference
crystalline structures,28 and it increases to roughly 30−32 Å2 in
mesophases.29 In other words, the sectional area of the donor
group is about twice that of the PDI block. This supports
strongly the idea that the PDI blocks of dyad molecules are
overlapping, but not the donor blocks. This overlapping of PDI
blocks of two dyads in a lamella generates very dense planes of
perfectly π-stacked PDI units with a short 3.5 Å stacking period
as illustrated in Figure 8a,b. These dense planes of PDI blocks
explain further the very strong contrast in the TEM-BF of the
lamellae, the sharp and dark lines corresponding to the planes
of densely stacked PDI units. The same reasoning holds also for
the organization of the ADA triad with overlapping layers of
PDI groups on either side of the donor block (see Figure 8c,d).
Very likely, it is the presence of the flexible spacer between A
and D units that gives sufficient freedom to the entire system to
comply with the dense packing of PDI layers whatever the
detailed structure of the D group (D0, D1, or D0T). Such a
flexible spacer allows to adjust the tilting of the D group within
the unit cell. The net result of this is that both tilting and
coiling of the lateral chains lead to a molecular area per PDI
group of 33 Å2 and thus to a molecular area per donor group
around 66 Å2 for all three investigated compounds (see Table
2).
Figure 8. Schematic view of the two proposed models of organization
of the AD and ADA based co-oligomers. A doubling of the a
parameter is implied for models (b) and (d). The packing of DAD
molecules, shown in (e), following the proposed models of ADA triad,
is not possible because of the steric hindrance generated by the
ethylhexyl side chains. PDI blocks are in red, donor blocks in blue, and
alkyl side chains are in green. Donor blocks shown in dark blue are in
an upper layer, whereas light-blue ones correspond to the layer
beneath. l corresponds to the lamellar period.
Bridging PDI monolayers freeze the positions of neighboring
co-oligomer molecules and donor moieties and then arrange
laterally. Fluorene or thiophene moieties can give rise to a
chevron packing, rather than face-to-face packing, which is
moreover hindered by the bulky lateral chains. These segments
are present in the three dyads and impose the chevron packing
to the whole donor blocks and thus the b parameter of 7.4 Å.
This value logically exceeds the natural face-to-face packing
5989
dx.doi.org/10.1021/ja4129108 | J. Am. Chem. Soc. 2014, 136, 5981−5992

Journal of the American Chemical Society
Article
distance of PDI groups (∼3.5 Å deduced from the hPDI
scattering), and the discrepancy needs to be compensated by
small tilts of the stacked PDI groups. GIWAXS patterns
confirm these tilts, since the π-stacking scattering maxima
appear tilted about 15−20° from the b vector direction (see
Figures 5b and e). Furthermore, the lamellae of donor moieties
are interrupted by perpendicular layers of molten aliphatic
chains (see Figure 8). The lateral periodicity coinciding with
the parameter a = 35−35.5 Å involves the chains and rigid
moieties of two donor blocks and therefore four PDI groups
shared by two lamellae. The side-by-side packing of the PDI
stacks then imposes a spacing of 8.8 Å per stack, close to values
in crystalline phases (8.3 Å28). Except for the different c
parameters related to the donor block lengths, the three dyads
and the ADA triad just show the same annealed structures and
similar geometrical parameters. As mentioned above, the ADA
triad moreover shows a mesophase and long-range order
already in the pristine state. The main specificities of the
pristine geometry consist in a 40% reduced lamellar spacing and
a 70% expansion along b, revealing an average 45° tilt of
molecules from the b direction. As confirmed by the absence of
all crossed reflections, the regular microsegregated layer
alternation along a of the orthorhombic phase does not form
natively, forcing rigid moieties to tilt. With the softening of the
sample on first heat, this segregated structure develops,
bringing rigid moieties back to the a × c plane and expanding
the parameter a to the size of a regular PDI monolayer, in
consistency with the gradual geometry change observed (see
Figure S14 in SI). For all cases, the few visible reflections and
molecular volumes are compatible with an orthorhombic phase
of similar nature. The structural parameters have been reported
in Table 2.
Putting together all structural information gathered in this
study, one can propose two different packing schemes of AD
dyads and the ADA triad, illustrated in Figure 8. As a
convention, the a, b, and c axes of the unit cell are oriented
along the alkyl side chains of the donor group, the π-stacking of
the PDI, and the long molecular axis of the AD molecules,
respectively. Considering the first model in Figure 8a, the AD
molecules are π-stacked into pairs with a strong π-overlap of
both the PDI and the donor groups. Donor blocks in successive
layers along the a axis are pointing alternatively along c and −c.
Within such AD pairs, the solubilizing chains of the two donor
blocks are rejected on either side of the conjugated skeleton,
and all alkyl chains are grouped together. Figure 8b shows an
alternative packing for which the AD molecules are π-stacked
into pairs of two molecules with a strong overlap of the PDI
units only, whereas donor blocks are arranged side-by-side with
the alkyl side chains rejected on each side of the donor pair.
However, as for the first model, dense planes of π-stacked PDI
blocks are formed also for this molecular arrangement. It is
likely that the stacking of PDI units involves a slight lateral
offset (arrangement already observed in other studies27b,30).
However, the PDI long axes are still colinear. These two
packing schemes can apply for AD0, AD0T, and AD1 dyads. The
two different packing schemes in Figure 8a,b can also apply for
ADA as illustrated in Figure 8c,d. The two terminal PDI groups
on each side of the donor block in ADA can π-stack efficiently
without perturbing the packing of donor blocks and can respect
the different sectional areas of the two blocks. However, the
major difference between ADn dyads and the ADA triad is the
strong interdigitation of the layers of ADA molecules. It can be
anticipated that this strong interdigitation is responsible for the
enhanced long-range lamellar ordering in ADA films even in
the absence of melt-cooling. As a matter of fact, the strong
interdigitation of PDI groups on both sides of the donor could
partly explain the higher density (1.21 g·cm−3) for the ADA
triad compared to the ADn analogues (1.13−1.14 g·cm−3).
Regarding the DnADn triads, the large sectional area of donor
blocks imposes a high molecular area per PDI group which is
incompatible with the generation of dense PDI layers. In some
sense, the DAD self-assembling into a lamellar structure at
long-range order is prevented by the presence of solubilizing
ethylhexyl side chains on the donor block that are responsible
for the high sectional area of this block as compared to PDI
(Figure 8e). The marked discrepancy between molecular areas
of D and A segments could similarly be held as responsible for
the lack of formation of well-defined D/A segregated structure
in other peryleneimide-donor systems.13c,31,32 Note that the
few published examples of well-defined lamellar mesophases in
DAD triads (with A = PDI) should be attributed to the
presence of a high density of chains (aliphatic13a or siloxane13g
)
at the triad extremities, which generates microphase separation.
To conclude, a packing of DAD molecules following a
scheme similar to that given in Figure 8 for the AD dyads or the
ADA triad is virtually impossible. A more precise structure
determination is needed for discriminating one or the other
packing behavior of the molecules. Further studies are being
undertaken in order to provide a more complete assessment of
the molecular packing of AD and ADA molecules.
3. CONCLUSIONS
We have successfully synthesized and characterized a new series
of donor−acceptor block co-oligomers with different molecular
architectures (AD, DAD, and ADA) with the aim of correlating
molecular architecture and self-assembling properties, which is
essential to improve optoelectronic devices such as OPV cells
or OFET. First, short ethylene linkers have been systematically
inserted between the A and D blocks to favor their respective
packing. Second, we have verified that these co-oligomers
comply with favorable optoelectronic properties, e.g. broad
range of absorption, appropriate HOMO and LUMO levels,
decoupling of A and D electronic states. This was possible by a
careful chemical design of the donor block made of a
combination of thienofluorene segments of two different
lengths (D0 or D1) with the electron-deficient benzothiadiazole
unit. Efficient optoelectronic devices require ordered structures
with well-segregated acceptor and donor domains that are
indeed determined by the molecular architecture. Intensive
characterization of the co-oligomers by X-ray diffraction and
TEM gave evidence for lamellar mesophases with long-range
order after melt-cooling for all systems except for DAD triads.
Regardless of the length and chemical nature of the donor
block (D0, D0T, and D1), all the AD dyads and the ADA triad
show very similar packing of the co-oligomers within lamellae.
The major difference between these two systems is the
interdigitated character of the lamellae for ADA, which is absent
for AD, and results in enhanced long-range ordering in both
pristine and melt-cooled ADA films. In the end, AD and ADA
superstructures contain extended rows of PDI-stacks within
lamellae, leading to three-dimensional mesophases, while DAD
systems show short-range lamellar correlations without marked
PDI-stacks. The difference in this behavior is explained by the
discrepancy of the molecular cross sections between the PDI
and the donor blocks. Placing the PDI unit at the end of the
molecules allows to compensate its smaller cross-section by the
5990
dx.doi.org/10.1021/ja4129108 | J. Am. Chem. Soc. 2014, 136, 5981−5992

Journal of the American Chemical Society
Article
Chem. 2010, 20, 10788−10797. (f) Zhang, Q.; Cirpan, A.; Russel, T.
P.; Emrick, T. Macromolecules 2009, 42, 1079−1082. (g) Sommer, M.;
Lindner, S. M.; Thelakkat, M. Adv. Funct. Mater. 2007, 17, 1493−1500.
(h) Lohwasser, R. H.; Gupta, G.; Kohn, P.; Sommer, M.; Lang, A. S.;
Thurn-Albrecht, T.; Thelakkat, M. Macromolecules 2013, 46, 4403−
4410.
formation of interdigitated rows of PDI-stacks, ultimately
leading to the cohesion of the lamellae. In contrast, with the
DAD architecture, the large cross-section of the donor blocks
prevents the central PDI unit from stacking.
Another important aspect of the preparation of optoelec-
tronic devices is the control of the orientation of the
nanostructure. In this work, we have evidenced the influence
of the donor block length on the orientation of the lamellar
domains on a SiO2 substrate after melt-cooling. A majority of
lying lamellae has been obtained for the short donor blocks
(AD0 and AD0T), whereas AD1 molecules led to edge-on
lamellae. In consequence, the modulation of the block length
might be an efficient tool to control the orientation of the
oligomers on a substrate, depending on the optoelectronic
application (OPV vs OFETs).
Taken all together, these results are important as they allow
us to set some basic principles for the molecular design of
efficient self-assembling PDI-based donor−acceptor co-
oligomers to be used in monocomponent optoelectronic
devices.
(9) (a) Nierengarten, J.-F. Sol. Energy Mater. Sol. Cells 2004, 83, 187−
199. (b) Roncali, J. Chem. Soc. Rev. 2005, 34, 483−495. (c) Segura, J.
L.; Martin, N.; Guldi, D. M. Chem. Soc. Rev. 2005, 34, 31−47.
(10) (a) Wurthner, F.; Chen, Z.; Hoeben, F. J. M.; Osswald, P.; You,
̈
C.-C.; Jonkheijm, P.; van Herrikhuyzen, J.; Schenning, A. P. H. J.; van
der Schoot, P. P. A. M.; Meijer, E. W.; Beckers, E. H. A.; Meskers, S. C.
J.; Janssen, R. A. J. J. Am. Chem. Soc. 2004, 126, 10611−10618.
(b) Ramos, A. M.; Meskers, S. C. J.; Beckers, E. H. A.; Prince, R. B.;
Brunsveld, L.; Janssen, R. A. J. J. Am. Chem. Soc. 2004, 126, 9630−
9644. (c) Bu, L.; Guo, X.; Yu, B.; Qu, Y.; Xie, Z.; Yan, D.; Geng, Y.;
Wang, F. J. Am. Chem. Soc. 2009, 131, 13242−13243. (d) Bu, l.; Guo,
X.; Yu, B.; Fu, Y.; Qu, Y.; Xie, Z.; Yan, D.; Geng, Y.; Wang, F. Polymer
2011, 52, 4253−4260.
(11) (a) Nierengarten, J.-F.; Eckert, J.-F.; Nicoud, J.-F.; Ouali, L.;
Krasnikov, V.; Hadziioannou, G. Chem. Commun. 1999, 617−618.
(b) Li, W. S.; Yamamoto, Y.; Fukushima, T.; Saeki, A.; Seki, S.;
Tagawa, S.; Masunaga, H.; Sasaki, S.; Takata, M.; Aida, T. J. Am. Chem.
Soc. 2008, 130, 8886−8887. (c) Nishizawa, T.; Tajima, K.; Hashimoto,
K. J. Mater. Chem. 2007, 17, 2440−2445. (d) Nishizawa, T.; Tajima,
K.; Hashimoto, K. Nanotechnology 2008, 19, 424017.
ASSOCIATED CONTENT
■
S
* Supporting Information
1
Detailed synthetic procedures, H NMR, 13C NMR, TGA, and
(12) (a) Chen, L. X.; Xiao, S.; Yu, L. J. Phys. Chem. B 2006, 110,
111730−11738. (b) Cremer, J.; Mena-Osteritz, E.; Pschirere, N. G.;
MS spectra of final co-oligomers, UV−vis absorption and
fluorescence spectra in solution, cyclic voltammetry traces,
XRD patterns of ADA as a function of the temperature, ED-
patterns and BF-TEM images of D0AD0. This material is
available free of charge via the Internet at http://pubs.acs.org.
Mullen, K.; Bauerle, P. Org. Biomol. Chem. 2005, 3, 985−995.
̈
̈
(c) Cremer, J.; Bauerle, P. Eur. J. Org. Chem. 2005, 3715−3723.
̈
(d) Cremer, J.; Bauerle, P. J. Mater. Chem. 2006, 16, 874−884.
̈
(e) Bullock, J. E.; Carmieli, R.; Mickley, S. M.; Vura-Weis, J.;
Wasielewski, M. J. Am. Chem. Soc. 2009, 131, 11919−11929. (f) Balaji,
G.; Kale, T. S.; Keerty, A.; Della Pelle, A. M.; Thayumanavan, S.;
Valliyaveettil, S. Org. Lett. 2011, 13, 18−21.
AUTHOR INFORMATION
Corresponding Author
stephane.mery@ipcms.unistra.fr
■
(13) (a) Peeters, E.; van Hal, P. A.; Meskers, S. C. J.; Janssen, R. J. A.;
Meijer, E. W. Chem.Eur. J. 2002, 8, 4470−4474. (b) van der Boom,
T.; Hayes, R. T.; Zhao, Y.; Bushard, P. J.; Weiss, E. A.; Wasielewski, M.
Notes
The authors declare no competing financial interest.
R. J. Am. Chem. Soc. 2002, 124, 9582−9590. (c) Dossel, L. F.; Kamm,
̈
V.; Hoiward, I. A.; Laquai, F.; Pisula, W.; Feng, X.; Li, C.; Takase, M.;
ACKNOWLEDGMENTS
Kudernac, T.; de Feyter, S.; Mullen, K. J. Am. Chem. Soc. 2012, 134,
̈
■
5876−5886. (d) Kim, M. H.; Cho, M. J.; Kim, K. H.; Hoang, M. H.;
Lee, T. W.; Jin, J.-I.; Kang, N. S.; Yu, J.-W.; Choi, D. H. Org. Electron.
2009, 10, 1429−1441. (e) Jonkheijm, P.; Stutzmann, N.; Chen, Z.; de
This work has been supported by the French National Research
Agency (ANR PICASSO Project) and by the European
Community via the Interreg IV-A program (C25, Rhin-
Solar). We thank Pohang Accelerator Laboratory (PAL) for
giving us the opportunity to perform the GIWAXS measure-
ments, MEST and POSTECH for supporting these experi-
ments, Dr. Tae Joo Shin for adjustments and help, and other
people from 9A U-SAXS beamline for assistance.
Leeuw, D. M.; Meijer, E. W.; Schenning, A. P. H. J.; Wurthner, F. J.
̈
Am. Chem. Soc. 2006, 128, 9535−9540. (f) Mativestsky, J. M.; Kastler,
M.; Savage, R. C.; Gentilini, D.; Palma, M.; Pisula, W.; Mullen, K.;
̈
Samori, P. Adv. Funct. Mater. 2009, 19, 2486−2494. (g) Roland, T.;
́ ́
Leonard, J.; Hernandez Ramirez, G.; Mery, S.; Yurchenko, O.;
Ludwigs, S.; Haacke, S. Phys. Chem. Chem. Phys. 2012, 14, 273−279.
(14) Kim, M. S.; Kim, J. S.; Cho, J. C.; Shtein, M.; Guo, L. J.; Kim, J.
Appl. Phys. Lett. 2007, 90, 123113.
REFERENCES
■
(15) (a) Fisslthaler, E.; Blumel, A.; Landfester, K.; Scherf, U.; List, E J
̈
(1) Chen, J. T.; Hsu, C.-S. Polym. Chem. 2011, 2, 2707−2722.
(2) Liu, H.; Xu, J.; Li, Y.; Li, Y. Acc. Chem. Res. 2010, 43, 1496−1508.
(3) Hoeben, F. J. M.; Jonkhejim; Meijer, E. W.; Schenning, A. P. H. J.
Chem. Rev. 2005, 105, 1491−1546.
(4) Coakley, K. M.; McGehee, M. D. Chem. Mater. 2004, 16, 4533−
4542.
W. Soft Matter 2008, 4, 2448−2453. (b) Kim, H.-C.; Park, S.-M.;
Hinsberg, W. D. Chem. Rev. 2010, 110, 146−177. (c) He, X.; Gao, F.;
Tu, G.; Hasko, D. G.; Huttner, S.; Greenham, N. C.; Steiner, U.;
̈
Friend, R. H.; Huck, W. T. S. Adv. Funct. Mater. 2011, 21, 139−146.
(16) (a) Mishra, A.; MA, C.-Q.; Bauerle, P. Chem. Rev. 2009, 109,
̈
1141−1276. (b) Yassar, A.; Miozzo, L.; Gironda, R.; Horowitz, G.
Prog. Polym. Sci. 2013, 38, 791−844. (c) Kozma, E.; Catellani, M. Dyes
Pigments 2013, 98, 160−179. (d) Sommer, M.; Huettner, S.;
Thelakkat, M. Adv. Polym. Sci. 2010, 228, 123−153.
(5) Hoppe, H.; Sariciftci, N. S. Adv. Polym. Sci. 2008, 214, 1−86.
(6) Yang, X.; Loos, J. Macromolecules 2007, 40, 1353−1362.
(7) Thompson, B. C.; Frec
47, 58−77.
(8) (a) De Boer, B.; Stalmach, U.; Melzer, C.; Krasnikov, V. V.;
Hadziioannou, G. Polymer 2001, 43, 9097−9109. (b) Zhang, F.;
Svensson, M.; Andersson, M. R.; Maggini, M.; Bucella, S.; Menna, E.;
́
het, J. M. J. Angew. Chem., Int. Ed. 2008,
(17) (a) Wurthner, F. Chem. Commun. 2004, 1564−1579. (b) Zhan,
̈
X.; Facchetti, A.; Barlow, S.; Marks, T. J.; Ratner, M. A.; Wasielewski,
M. R.; Marder, S. R. Adv. Mater. 2011, 23, 268−284. (c) Li, C.;
Wonneberger, H. Adv. Mater. 2012, 24, 613−636. (d) Kozma, E.;
Catellani, M. Dyes Pigments 2013, 98, 160−179.
Inganas, O. Adv. Mater. 2001, 13, 1871−1874. (c) Miyanishi, S.;
̈
Zhang, Y.; Tajima, K.; Hashimoto, K. Chem. Commun. 2010, 46,
6723−6725. (d) Lindner, S. M.; Thelakkat, M. Macromolecules 2004,
37, 8832−8835. (e) Sommer, M.; Huettner, S.; Thelakkat, M. J. Mater.
(18) (a) Richard, F.; Brochon, C.; Leclerc, N.; Eckhardt, D.; Heiser,
T.; Hadziioannou, G. Macromol. Rapid Commun. 2008, 29, 885−891.
5991
dx.doi.org/10.1021/ja4129108 | J. Am. Chem. Soc. 2014, 136, 5981−5992

Journal of the American Chemical Society
Article
(b) Yassar, A.; Miozzo, L.; Gironda, R.; Horowitz, G. Prog. Polym. Sci.
2013, 38, 791−844.
(19) (a) Samori, P.; Yin, X.; Tchebotareva, N.; Wang, Z.; Pakula, T.;
Jackel, F.; Watson, M. D.; Venturini, A.; Mullen, K.; Rabe, J. R. J. Am.
̈
̈
Chem. Soc. 2004, 126, 3567−3575. (b) Yamamoto, Y.; Fukushima, T.;
Suna, Y.; Ishii, N.; Saeki, A.; Seki, S.; Tagawa, S.; Taniguchi, M.; Kawai,
T.; Aida, T. Science 2006, 314, 1761−1764.
(20) (a) Langhals, H. Helv. Chim. Acta 2005, 88, 1309−1343.
(b) Vura-Weis, J.; Ratner, M. A.; Wasielewski, M. R. J. Am. Chem. Soc.
2010, 132, 1738−1739.
(21) (a) Grell, M.; Bradley, D. D. C.; Inbasekaran, M.; Woo, E. P.
Adv. Mater. 1997, 9, 798. (b) Sherf, U.; List, E. J. W. Adv. Mater. 2002,
14, 477−487. Zhang, X.; Qu, Y.; Bu, L.; Tian, H.; Zhang, J.; Wang, L.;
Geng, Y.; Wang, F. Chem.Eur. J. 2007, 13, 5238−6248.
́ ̂
(22) (a) Biniek, L.; Fall, S.; Chochos, C. L.; Leclerc, N.; Leveque, P.;
Heiser, T. Org. Electron. 2012, 13, 114−120. (b) Schwartz, P. O.;
Zaborova, E.; Bechara, R.; Leveque, P.; Heiser, T.; Mery, S.; Leclerc,
N. New J. Chem. 2013, 37, 2317−2323.
́
̂
́
(23) Ramos, A. M.; Beckers, E. H. A.; Offermans, T.; Meskers, S. C.
J.; Janssen, R. A. J. J. Phys. Chem. A 2004, 108, 8201−8211.
(24) (a) Biniek, L.; Fall, S.; Chochos, C. L.; Anokhin, D. V.; Ivanov,
D.; Leclerc, N.; Lev
9779−9786. (b) Chu, T.-Y.; Lu, J.; Beaupre,
R.; Zhou, J.; Najari, A.; Leclerc, M.; Tao, Y. Adv. Funct. Mater. 2012,
́
eq
̂
ue, P.; Heiser, T. Macromolecules 2010, 43,
S.; Zhang, Y.; Pouliot, P.-
́
22, 2345−2351.
(25) Kelber, J.; Bock, H.; Thiebaut, O.; Grelet, E.; Langhals, H. Eur. J.
Org. Chem. 2011, 707−712.
(26) Baek, M.-J.; Jang, W.; Lee, S.-H.; Lee, Y.-S. Synth. Met. 2012,
161, 2785−2791.
(27) (a) Hansen, M. R.; Graf, R.; Sekharan, S.; Sebastiani, D. J. Am.
Chem. Soc. 2009, 131, 5251−5256. (b) Struijk, C. W.; Sieval, A. B.;
Dackorst, J. E. J.; van Dijk, M.; Kimbes, P.; Koehorst, R. B. M.;
Donker, H.; Schaafsma, T. J.; Picken, S. J.; van de Craats, A. M.;
Warman, J. M.; Zuilhof, H.; Sudholter, E. J. R. J. Am. Chem. Soc. 2000,
122, 11057−11066.
(28) For instance in the structure DICLIK (N,N′-diethylperylene-
3,4:9,10-bis(dicarboximide)) from the Cambridge Structural Database.
(29) From the estimated partial volume expansion with respect to
crystal.
(30) Kazmaier, P. M.; Hoffmann, R. J. Am. Chem. Soc. 1994, 116,
9684−9691.
(31) Neuteboom, E. E.; Meskers, S. C. J.; van Hal, P. A.; van Duren,
J. K. J.; Meijer, E. W.; Janssen, R. A. J.; Dupin, H.; Pourois, G.; Cornil,
J.; Lazzaroni, R.; Bred
125, 8625−8638.
(32) Pron, A.; Reghu, R. R.; Rybakiewicz, R.; Cybulski, H.; Djurado,
D.; Grazulevicius, J. V.; Zagorska, M.; Kulszewicz-Bajer, I.; Verilhac, J.-
M. J. Phys. Chem. C 2011, 115, 15008−15017.
́
as, J.-L.; Beljonne, D. J. Am. Chem. Soc. 2003,
5992
dx.doi.org/10.1021/ja4129108 | J. Am. Chem. Soc. 2014, 136, 5981−5992