Tailored donor-acceptor polymers with an A-D1-A-D2 structure: Controlling intermolecular interactions to enable enhanced polymer photovoltaic devices

Article abstract of DOI:10.1021/ja500935d

Extensive efforts have been made to develop novel conjugated polymers that give improved performance in organic photovoltaic devices. The use of polymers based on alternating electron-donating and electron-accepting units not only allows the frontier molecular orbitals to be tuned to maximize the open-circuit voltage of the devices but also controls the optical band gap to increase the number of photons absorbed and thus modifies the other critical device parameter-the short circuit current. In fact, varying the nonchromophoric components of a polymer is often secondary to the efforts to adjust the intermolecular aggregates and improve the charge-carrier mobility. Here, we introduce an approach to polymer synthesis that facilitates simultaneous control over both the structural and electronic properties of the polymers. Through the use of a tailored multicomponent acceptor-donor-acceptor (A-D-A) intermediate, polymers with the unique structure A-D1-A-D2 can be prepared. This approach enables variations in the donor fragment substituents such that control over both the polymer regiochemistry and solubility is possible. This control results in improved intermolecular π-stacking interactions and therefore enhanced charge-carrier mobility. Solar cells using the A-D1-A-D2 structural polymer show short-circuit current densities that are twice that of the simple, random analogue while still maintaining an identical open-circuit voltage. The key finding of this work is that polymers with an A-D1-A-D2 structure offer significant performance benefits over both regioregular and random A-D polymers. The chemical synthesis approach that enables the preparation of A-D1-A-D2 polymers therefore represents a promising new route to materials for high-efficiency organic photovoltaic devices.

Full text of DOI:10.1021/ja500935d

Article
pubs.acs.org/JACS
Tailored Donor−Acceptor Polymers with an A−D1−A−D2 Structure:
Controlling Intermolecular Interactions to Enable Enhanced Polymer
Photovoltaic Devices
Tianshi Qin,*,† Wojciech Zajaczkowski,‡ Wojciech Pisula,‡ Martin Baumgarten,‡ Ming Chen,† Mei Gao,†
Gerry Wilson,† Christopher D. Easton,† Klaus Mullen,*,‡ and Scott E. Watkins*,†
̈
†Ian Wark Laboratory, CSIRO Materials Science & Engineering, Clayton South, Victoria 3169 Australia
‡Max Planck Institute for Polymer Research, Ackermannweg 10, D-55128 Mainz, Germany
S
* Supporting Information
ABSTRACT: Extensive efforts have been made to develop
novel conjugated polymers that give improved performance in
organic photovoltaic devices. The use of polymers based on
alternating electron-donating and electron-accepting units not
only allows the frontier molecular orbitals to be tuned to
maximize the open-circuit voltage of the devices but also
controls the optical band gap to increase the number of
photons absorbed and thus modifies the other critical device
parameterthe short circuit current. In fact, varying the
nonchromophoric components of a polymer is often secondary
to the efforts to adjust the intermolecular aggregates and improve the charge-carrier mobility. Here, we introduce an approach to
polymer synthesis that facilitates simultaneous control over both the structural and electronic properties of the polymers.
Through the use of a tailored multicomponent acceptor−donor−acceptor (A−D−A) intermediate, polymers with the unique
structure A−D1−A−D2 can be prepared. This approach enables variations in the donor fragment substituents such that control
over both the polymer regiochemistry and solubility is possible. This control results in improved intermolecular π-stacking
interactions and therefore enhanced charge-carrier mobility. Solar cells using the A−D1−A−D2 structural polymer show short-
circuit current densities that are twice that of the simple, random analogue while still maintaining an identical open-circuit
voltage. The key finding of this work is that polymers with an A−D1−A−D2 structure offer significant performance benefits over
both regioregular and random A−D polymers. The chemical synthesis approach that enables the preparation of A−D1−A−D2
polymers therefore represents a promising new route to materials for high-efficiency organic photovoltaic devices.
combination allows the molecular orbitals to be tuned so as to
enable intramolecular charge transfer (ICT).11 The open circuit
voltage (Voc) of OPVs is directly related to the gap between the
highest occupied molecular orbital (HOMO) of the polymer
and lowest unoccupied molecular orbital (LUMO) of the
fullerene. Tuning of the polymer molecular orbitals therefore
enables the Voc to be maximized. While straightforward in
design and practice, the use of simple, D−A polymers has a
number of drawbacks. First, conventional polycondensation
polymerizations do not allow regioselectivity when asymmetric
monomers are employed. As such, there is no control over the
orientation of functional groups with respect to each other.
Regioregular polymers have a regular, alternating structure,
whereas random polymers have substituents that are randomly
oriented along the chain. It is well-known that variations in
molecular order reduce intermolecular π-stacking interactions
that in turn can limit charge transport through polymer films.12
A second disadvantage of simple D−A polymers is that there
INTRODUCTION
■
Bulk heterojunction (BHJ) organic photovoltaics (OPVs),
based on blends of conjugated polymers and soluble fullerene
derivatives, have attracted attention due to their unique
advantages of solution processability, flexibility, lightweight,
large area, and low cost production.1 Over the past few years,
organic chemists and material scientists have been working
toward developing novel molecular structures and device
architectures to improve the performance of photovoltaic
devices.2 To maximize the performance of OPV devices,3
attention has shifted beyond just the molecular formula to
other considerations such as molecular weight characteristics,4
chemical and thermal stabilities,5 end-group contributions,6 and
extrinsic impurities7 as well as intra- and intermolecular charge
carrier mobilities.8 In addition, strategies such as the use of
annealing processes or additives9 as well as the use of interfacial
materials10 have been employed.
Progress due to advances in the chemical structure of
materials has mainly come through the use of a donor−
acceptor (D−A) approach; polymers based on alternating
fragments of electron-rich and electron-poor groups. This
Received: February 2, 2014
Published: April 4, 2014
© 2014 American Chemical Society
6049
dx.doi.org/10.1021/ja500935d | J. Am. Chem. Soc. 2014, 136, 6049−6055

Journal of the American Chemical Society
Article
are only two variables, the solubilizing groups on the acceptor
and donor fragments. This means that the solubility of the
polymer can only be altered by changing all of the solubilizing
groups on a donor or acceptor fragment at once. Such a change
could dramatically alter the properties of the polymer. A recent
perspective article emphasized how the size and topology of
side chains as well as the orientation of repeating units can
influence the macroscopic properties of polymers.13
could satisfy both requirements. Compared with other well-
known donor units, such as silafluorene18 or dithienosilole,19 in
which two out-of-plane side chains have to be located on the
fused heteroatom between the aromatic rings, the DT-BDT
unit can provide multiple positions for solubilizing substituents
with varied out-of-plane dihedral angles. For the acceptor
fragment, monofluoro substituted benzothiadiazole (FBT) is
one of only two reported acceptor units (the other is
pyridalthiadiazole20), which demonstrates asymmetric reactivity
between ortho- and meta-positions.
The design strategy conceived in this work is illustrated in
Scheme 1. We present a stepwise synthetic strategy, based on
RESULTS AND DISCUSSION
Scheme 1. Design Strategy Based on an A−D−A
■
a
Intermediate
Synthesis. To begin, we prepared two DT-BDT monomers
(D1 and D2, Figure S3) as donor units with various substituted
alkyl-chains as well as the acceptor monomer 3,6-dibromo-5-
fluoro-2,1,3-benzothiadiazole (FBT) (A, Figure S3). By using
direct or stepwise synthesis methods, four novel DT-BDT-FBT
based conjugated polymers with either random (BFR1 and
BFR2) or regioregular (BFS3 and BFS4) orientations in the
polymer backbone were achieved, Scheme 2. The random and
Scheme 2. Repeat Units of the DT-BDT-FBT-Based
a
Polymers
a
The different coloured ellipses on D1 and D2 represent different
substituents.
an intermediate A−D−A intermediate, which provides
simultaneous control over both the polymer regiochemistry
and solubility. In this terminology, “A” represents an electron-
accepting group, and “D” represents an electron-donating
group. This route provides access to a series of π-conjugated
polymers based on both A−D1−A−D1 and A−D1−A−D2
structures. In the latter, two different donor fragments (D1 and
D2) are used, and the structure of the polymer is controlled. In
side-by-side comparisons we demonstrate that by controlling
the orientation of substituents and the amount of non-
chromophoric components in the polymer, the macromolecular
interactions can be enhanced which in turn lead to significant
improvements in the short-circuit current densities of the
OPVs.
As outlined above, our strategy enables two different donor
fragments to be used while still maintaining an alternating D−A
structure. Janssen et al. have recently shown that combining the
properties of multiple D−A copolymer systems can have
advantageous benefits.14 Patil et al. have described an A1−D−
A2−D polymer for use in OFETs where optimized chain
packing and the formation of large crystalline domains in the
solid state dramatically enhanced charge-carrier mobility.15 Our
design strategy is distinct in that it extends on these ideas by
also providing a means to control the regiochemistry of the
polymer through the fixed orientation of the acceptor
fragments. To demonstrate the strategy we sought donor and
acceptor fragments that would (i) enable variation in the
positioning of side chains and (ii) enable controllable reactivity
in an asymmetric monomer. After a review of the literature
dithienyl-benzo[1,2-b:4,5-b′]dithiophene (DT-BDT)16 and 5-
fluoro-2,1,3-benzothiadiazole (FBT)17 were selected as they
a
Random F-atom arranged BFR1 and BFR2, regioregular BFS3 and
BFS4. The coloured ellipses highlight the different solubilising groups,
and the red arrows emphasize the regiochemical control.
regioregular labels refer to the orientation of the fluorine atom
on the FBT fragment with respect to the donor fragment. The
synthetic routes are shown in Figure S3. The branched 2-
ethylhexyl chain was first used on the DTBDT units in an
attempt to ensure adequate solubility of the target polymers.
Unfortunately, the first random copolymer BFR1 demonstrated
limited solubility and could not be processed using common
solvents. Accordingly, an extra linear hexyl chain was added to
the design, resulting in the second random copolymer BFR2
with high solubility in chloroform, chlorobenzene and
dichlorobenzene. Polymer BFR2 is a control for comparison
with the regioregular polymers BFS3 and BFS4.
Polymers BFS3 and BFS4 were prepared in a two-step route
via an A−D1−A intermediate. Control over the structure of the
intermediate was achieved due to the preferential reaction of
the Pd-catalyzed Stille coupling at the position meta- to the
fluorine atom on the FBT group. This asymmetric reactivity
gave the A−D1−A intermediate in yields as high as 90%. The
proposed regiochemical structure was confirmed by 2D NMR
spectroscopy (Figure S4). No proton peaks were observed
6050
dx.doi.org/10.1021/ja500935d | J. Am. Chem. Soc. 2014, 136, 6049−6055

Journal of the American Chemical Society
Article
from A or D1 starting materials or from their polymeric side
product. The A−D1−A intermediate was then used to prepare
the two regioregular, conjugated polymers BFS3 and BFS4.
Coupling A−D1−A with monomer D1 and D2, respectively,
varied the amounts of pendant side chains while keeping the
basic conjugated polymer backbone unchanged. BFR2 and
BFS3 are structural isomers differing only in the fact that in the
latter the position of the fluorine atom is controlled, while in
the former it is random. Polymer BFS4 has a lower proportion
of nonchromophoric groups while still maintaining high
solubility in common solvents.
Table 1. Optical, Optoelectronic, And Frontier Molecular
Orbital Level Data
a
b
solution
film
λmax
λmax λedge
λedge
polymer (nm) (nm) (nm) (nm) Egopt(eV)
HOMO LUMO
d
c
(eV)
(eV)
BFR2
BFS3
BFS4
656
661
665
733
744
749
667
668
678
778
783
793
1.59
1.58
1.56
5.20
5.20
5.15
3.61
3.62
3.59
a
b
c
10−6 M in chloroform solution. 100 nm thick film. Calculated from
d
λedge of film. Estimated by PESA.
Thermal, Optical, and Optoelectrical Properties. High-
temperature gel permeation chromatography (HT-GPC) was
employed to evaluate the molecular weight distribution of the
conjugated polymers. Although the random polymer BFR1
showed poor solubility and could not be analyzed, the other
three polymers were soluble in 1,2,4-trichlorobenzene which is
used as the eluent in the HT-GPC system. The regioregular
polymer BFS4 was prepared with a high molecular weight with
Mn = 46 kDa and PDI = 2.8. For comparison, batches of the
two regioregular polymers BFS3 and BFS4 were synthesized
with similar molecular weights and polydispersity as the
random counterpart BFR2 (Figure S5).
The molar extinction coefficients of the three, soluble,
conjugated polymers in chloroform solutions with a concen-
tration of 10−6 M and the absolute absorbance spectra of neat,
100 nm thick films are shown in Figure 1, and the data listed in
Table 1. In dilute solutions, the main absorption band of
polymer BFS4 is bathochromically shifted and broadened to
some extent in comparison to that of its analogue BFS3 and
random counterpart BFR2. This difference can be attributed to
enhanced intermolecular π−π interactions resulting from the
reduced size of the pendant side groups and the regioregular
arrangement of the polymer main chain.21 In the solid-state
absorption spectra these differences were even more
pronounced. The maximum absorption edge of BFR2 and
BFS3 was at 778 and 783 nm, respectively, and 793 nm for
BFS4. Furthermore, the absorption intensity of BFS4 as a thin
film was higher than the other two polymers, an observation
that is consistent with the reduction of the proportion of
nonchromophoric alkyl chains in the polymer.22 The onset of
the film absorbance spectra was used to estimate the HOMO−
LUMO gaps of the three polymers, BFR2, BFS3, and BFS4.
Photoelectron spectroscopy in air (PESA) was applied to
estimate the HOMO of these conjugated polymers in thin films
(Figure S6).23 The experimentally determined HOMO values
and the calculated LUMO levels are summarized in Table 1. It
is clear that the variations to the polymers described here do
not have a significant effect on the energy levels of the
materials.
Intermolecular Interactions. The organization of the
regioregular polymers BFR2, BFS3, and BFS4 in the bulk solid
state was investigated by using two-dimensional, wide-angle X-
ray scattering (2D-WAXS) measurements (Figure 2a−c).24
Figure 2. 2D-WAXS patterns of filament extruded polymers of (a)
BFR2, (b) BFS3, and (c) BFS4 at 30 °C and GIWAXS patterns of (d)
BFR2/PC61BM, (e) BFS3/PC61BM, and (f) BFS4/PC61BM thin
films.
Fibers of the neat polymers were macroscopically aligned by
extrusion, and no further thermal pre- or postannealing
processes were applied in order to ensure the same thermal
treatment as during the device fabrication. At 30 °C the
polymers revealed a characteristic pattern with reflections being
distributed in the equatorial and meridional planes.25 However,
significant differences in intensity and position of the reflections
were evident for both systems. The positions of the reflections
of BFR2 and BFS3 in the equatorial small- and wide-angle
Figure 1. UV−vis absorption spectra of DT-BDT-FBT polymers. (a)
□
○
△
Molar absorptivities (ε) of BFR2 ( ), BFS3 ( ), and BFS4 ( ) in
10−6 M chloroform solutions. (b) Absolute absorbance spectra of the
corresponding polymers in 100 nm films.
6051
dx.doi.org/10.1021/ja500935d | J. Am. Chem. Soc. 2014, 136, 6049−6055

Journal of the American Chemical Society
Article
region are similar (Figures 2a,b and S7). However, BFS3
displayed only a faint 2D pattern due to its poor scattering. For
both polymers, the equatorial small-angle reflections were
attributed to a chain-to-chain distance of 19.65 Å between the
lamellar structures; the broad wide-angle scattering intensity on
the same plane was ascribed to a quite large stacking distance of
4.18 Å, and the meridional, middle-angle position scattering
reflection, corresponding to a d-spacing of 12.0 Å for BFR2 and
12.65 Å for BFS3, was assigned to the length of the single DT-
BDT repeat unit of polymer chain.19 By contrast, BFS4
exhibited sharper, more distinct and a higher number of
reflections with a chain-to-chain distance of 24.51 Å, much
smaller stacking distance of 3.66 Å and an unchanged repeating
spacing between DT-BDT units of 12.8 Å (Figures 2c and S7).
Thus, the significantly closer packing and higher crystallinity for
BFS4 in the bulk are in agreement with the absorption data and
could be related to the polymer structure.
To gain further information about the organization in the
solar cell, grazing incidence WAXS (GIWAXS) on thin films
was performed (Figure 2d−f). The thin films were prepared in
the same way as for the devices including the PEDOT:PSS
surface and the blending with PC61BM and 1,8-diiodooctane. In
both cases, the order was lower than in the bulk as indicated by
the lack of higher order reflections. This could be attributed to
the spin-coating process. Nevertheless, a significant distinction
in surface organization between the polymers could be derived.
The meridional position of the wide-angle scattering intensity
at qxy = 0 Å−1 and qz = 1.72 Å−1 in the pattern for BFS4
(Figures 2f and S8), which was related to the stacking distance
of 3.66 Å, was characteristic for a face-on arrangement of the
backbones toward the surface. In accordance with this
alignment, the chain-to-chain reflections appeared in the
small-angle equatorial plane. Additional meridional small-
angle reflections were due to an edge-on orientation of a
certain misaligned polymer fraction leading to a mixed
organization with a coexistent face- and edge-on arrangement
in the film.26 It is expected that a face-on polymer arrangement
favors the charge carrier transport perpendicular to the surface
as is the case in a solar cell. By contrast, BFR2 and BFS3 reveal
only a meridional small-angle scattering intensity and no π-
stacking reflection indicating low order in edge-on arranged
polymer lamellae (Figures 2d,e and S8). Therefore, with
respect to both factors, better packing and face-on orientation,
the data suggest that BFS4 is more optimized for solar cell
applications than BFS3.
Figure 3. Computational representations. (a) B3LYP/6-31+G(d,p)
calculation of D−A dimer fragments. (b) 3D simulation (semi-
empirical AM1 model) of repeating unit A−D1−A−D2 in BFS4. (c)
Schematic edge-view illustration of the possible supramolecular
organization in BFS3 and BFS4.
Computational analysis of the D−A dimer fragment in the
two regioregular polymers (BFS3 and BFS4) by density
functional theory (DFT)27 at the B3LYP/6-31+G(d,p) level of
theory estimated that the dihedral angle between donor and
acceptor units is only 5−7°, meaning that the polymer
backbones should be planar (Figure 3a). The pendant
thiophene unit on the BDT unit displayed a calculated dihedral
angle around 51−55°. This means that the alkyl substituents
may hinder close packing of the polymer chains. We then used
a semiempirical AM1 method28 to calculate the volume of
space occupied by the various alkyl chains. For the bis-
substituted donor fragment the combination of the para-
substituted ethylhexyl and meta-substituted hexyl chains
occupies a sphere with a maximum diameter of 4.8 Å (Figure
3b). By contrast, for the monosubstituted donor fragment, with
only a single para-substituted ethylhexyl group, the maximum
size of a sphere occupied by the alkyl chains is significantly
smaller, around 2.2 Å. As noted previously, the 2D-WAXS
analysis showed that stacking distance reduced by around 0.52
Å from 4.18 Å for BFS3 to 3.66 Å for BFS4 for unannealed
bulk organization. For BFS3, every donor fragment features the
large (4.8 Å) bis-substituted group. Two adjacent bis-
substituted groups when aligned directly would fill a space
that is larger than the measured π−π stacking distance. It is
therefore likely that, in BFS3, the donor fragments in adjacent
chains are offset with respect to each other (Figure 3c). By
contrast, the presence of the smaller, monosubstituted donor
fragment in BFS4 should not hinder the close approach of
donor units on adjacent chains as the sum of the calculated
maximum distance occupied by a large and small donor unit
[(4.8 + 2.2 Å)/2] = 3.5 Å is less than the measured stacking
distance of 3.66 Å. It has previously been shown that D−A
polymers are known to prefer an arrangement, where D and A
groups on adjacent chains align with the same groups above
and below in a columnar fashion.29 By reducing the steric bulk
6052
dx.doi.org/10.1021/ja500935d | J. Am. Chem. Soc. 2014, 136, 6049−6055

Journal of the American Chemical Society
Article
of the substituents in BFS4 we postulate that an arrangement
closer to an aligned, columnar structure (Figure 3c) is possible
and is indeed consistent with the WAXS data. Such an outcome
is directly attributable to the structural control provided by our
synthetic method.
Device Performances. In order to investigate the influence
of the intermolecular interactions of the conjugated polymers
on their electronic properties, single charge carrier devices were
prepared. The measured current density/voltage (J−V)
characteristics were then fitted by using the space-charge-
limited current (SCLC) model.30 At a typical electric field of
105 V cm−1 (corresponding to an applied voltage of 1 V to a
100 nm neat polymer device), the observed occurrence of a
SCLC enabled a direct estimate of the hole mobility of 2.3 ×
10−5, 2.0 × 10−4, and 1.5 × 10−3 cm2 V−1 s−1 for BFR2, BFS3,
and BFS4, respectively (Figure S9). At a high voltage of 5 V,
BFS4 also exhibited a field-dependent electron mobility31 of 7.5
× 10−4 cm2 V−1 s−1 (at 5 × 105 V cm−1, Figure S10). Notably,
neither of the other two polymers showed any measurable
electron mobility. The observation that BFS4 has both the
highest hole mobility and a measurable electron mobility is
consistent with the highest order and close π-stacking distance
as derived from the structural analysis, and the structure
postulated above whereby the reduced steric bulk permits an
arrangement that is closer to a columnar structure. The high
charge mobility for BFS4 is also desirable with respect to
increasing the short circuit current density (JSC) in OPV
devices.
Figure 4. OPV performance of DT-BDT-FBT polymers. (a)
□
Photocurrent density−voltage (J−V) curves of BFR2 ( ), BFS3
−2
○
△
( ) and BFS4 ( ) under illumination of AM 1.5G, 100 mW cm .
OPV devices were fabricated using these three polymers
(BFR2, BFS3, and BFS4), in turn, as the electron donor and
PC61BM as the electron acceptor. The device structure was
ITO/PEDOT:PSS/polymer:PC61BM/PFN/Al. PFN is a poly-
electrolyte that is used as an interlayer to assist with charge
extraction.32 The optimized weight ratio of polymer to PC61BM
is 1:1.4. About 2% (1,8-diiodooctane (DIO)/o-dichlorobenzene
(o-DCB), v/v) of DIO was added as an additive to enable
differential solubility.33 No thermal pre- or postannealing
processes were applied during the device fabrication. Device J−
V characteristics are shown in Figure 4a, and the parameters are
listed in Table 2. From the J−V curves, a clear improvement of
photovoltaic properties was observed in going from the random
to the regioregular polymers. The VOC and fill factor (FF)
values were in the same range for the three polymers, a finding
that is consistent with their similar HOMO levels, equal D/A
ratio, and identical device structures. However, the JSC values
for devices based on BFS4 were significantly increased in
comparison with those of BFS3 and BFR2, even with similar
molecular weights. Notably, a JSC as high as 14.20 mA/cm2 was
obtained for devices based on BFS4 and PC61BM. Combined
with its high VOC and FF, a high PCE of 7.38 0.4% (average
from 40 devices, simulated sunlight, AM 1.5G, 100 mW cm−2)
was achieved, the highest measured PCE being 7.80%. To the
best of our knowledge, the only reports of higher PCEs from
polymer OPV devices make use of the higher cost C70
derivative.
(b) IPCE curves of the corresponding OPVs.
Table 2. Photovoltaic Properties of the OPVs Based on DT-
BDT-FBT Polymers/PC61BM
PCE (%)
Mn
(kDa)
VOC
(V)
JSC
FF
a
polymer
PDI
(mA/cm2)
(%)
best
ave
BFR2
BFS3
BFS4
30
34
32
46
2.2
2.2
2.3
2.8
0.89
0.90
0.90
0.90
−7.25
−10.82
−13.19
−14.20
b
60.56
60.81
60.33
61.05
3.91
5.92
7.16
7.80
3.74
5.67
7.02
7.38
b
BFS4
a
Averaged from at least 40 devices. Results are shown for two
different batches of BFS4.
17−32% to 36−54% and up to 60−68% for BFR2, BFS3, and
BFS4, respectively. The integrated IPCE curves give calculated
Jscs of 7.5, 10.7, and 14.0 mA/cm2 for BFR2, BFS3, and BFS4,
respectively, confirming the accuracy of our photovoltaic
measurements. These results are again consistent with the
conclusion that the increased structural order and a reduction
in the proportion of nonchromophoric side chains are
responsible for the significant improvements in device perform-
ance.
The advantageous effect of the A−D1−A−D2 structure on
the PCE is most likely also related to the microstructure of the
bulk heterojunction blends that are obtained during film
deposition. AFM phase images of the photoactive layers reveal
clear phase separation arising from aggregated polymer chains
(Figure 5). BFR2/PC61BM and BFS3/PC61BM phase images
are dominated by areas of dark phase (i.e., a low phase value),
with needle-shaped bright phase regions ranging in length from
approximately 10 to 200 nm. For BFS4/PC61BM, the needle-
shaped regions are more densely packed, leading to larger
To further explore the properties of the devices, the incident
photon-to-current efficiency (IPCE) of the devices based on
the three polymers was measured. The IPCE curves are shown
in Figure 4b. All the devices exhibited a good response to short-
wavelength sunlight in the range 300−450 nm, attributable to
the contribution of PC61BM.34 Their photo conversion
efficiencies in the long-wavelength range, 450−750 nm, which
correspond to the absorption of the polymers,35 varied from
6053
dx.doi.org/10.1021/ja500935d | J. Am. Chem. Soc. 2014, 136, 6049−6055

Journal of the American Chemical Society
Article
ACKNOWLEDGMENTS
■
This research was funded through the Flexible Electronics
Theme of the CSIRO Future Manufacturing Flagship and was
also supported by the Victorian Organic Solar Cell Consortium
(Victorian Department of Primary Industries, Victorian
Department of Business and Innovation and the Australian
Renewable Energy Agency (ARENA)) and the Australian
Centre for Advanced Photovoltaics (ACAP). W.P., M.B., and
K.M. acknowledge financial support from the ERC grant
NANOGRAPH, European Community EC-ITN-SUPERIOR
(GA-2009-238177) and the Nano Sci-E SENSORS program.
T.Q. acknowledges a Postdoctoral Fellowship from the CSIRO
Office of the Chief Executive and a Postdoctoral Fellowship
from ARENA. T.Q. thanks Dr. Xiwen Chen for supply of the
interfacial PFN material, Dr. Noel W. Duffy for IPCE data
collection, and Dr. Roger Mulder for 2D-NMR measurements.
Figure 5. AFM phase images of the photoactive layers: (a) BFR2/
PC61BM, (b) BFS3/PC61BM, and (c) BFS4/PC61BM thin films.
islands of bright phase and a greater ratio of bright to dark
phase for this blend. Thus, the bright phases mostly likely
correspond to the donor material as we expect closer packing
for BFS4. The superior device performance of BFS4 is
therefore likely the result of the packing density and frequency
of these regions observed herein.
REFERENCES
■
(1) Li, G.; Zhu, R.; Yang, Y. Nat. Photonics 2012, 6, 153−161.
(2) (a) Li, Y. F. Acc. Chem. Res. 2012, 45, 723−733. (b) Sun, Y. M.;
Welch, G. C.; Leong, W. L.; Takacs, C. J.; Bazan, G. C.; Heeger, A. J.
Nat. Mater. 2012, 11, 44−48. (c) He, Z. C.; Zhong, C. M.; Su, S. J.;
Xu, M.; Wu, H. B.; Cao, Y. Nat. Photonics 2012, 6, 591−595. (d) Dou,
L. T.; You, J. B.; Yang, J.; Chen, C. C.; He, Y. J.; Murase, S.; Moriarty,
T.; Emery, K.; Li, G.; Yang, Y. Nat. Photonics 2012, 6, 180−185.
(3) Janssen, R. A. J.; Nelson, J. Adv. Mater. 2013, 25, 1847−1858.
(4) Chu, T. Y.; Lu, J. P.; Beaupre, S.; Zhang, Y. G.; Pouliot, J. R.;
Zhou, J. Y.; Najari, A.; Leclerc, M.; Tao, Y. Adv. Funct. Mater. 2012, 22,
2345−2351.
CONCLUSION
■
To conclude, we have demonstrated a chemical design
approach to improving the performance of organic solar cells.
The JSC of the OPV devices was increased through the use of
tailored A−D−A intermediates and a stepwise approach to the
synthesis of conjugated polymers. We have shown that the use
of A−D1−A−D2 repeating units, as opposed to a typical A−D
structure, enables control of the supramolecular interactions
within a polymer film. Specifically, our method enables control
over both the regiochemistry of the polymer backbone and the
amount of nonchromophoric components in the polymer.
Through molecular simulations backed by 2D-GIWAXS
measurements we have presented compelling evidence that
the intermolecular π-stacking interactions have been enhanced.
In turn, these conclusions were supported by AFM, charge
mobility, and photovoltaic device measurements that show
significant improvements in the performance of the OPVs.
Using the BFS4 polymer in a BHJ OPV with PC61BM a PCE
higher than 7% has been reported. More generally, the
approach we have presented also provides an opportunity to
further vary the optoelectronic properties of conjugated
polymers through the use of two different electron-donor
units. For example, the synthesis of A−D1−A−D2 type
structures is now possible where D1 and D2 are electronically
and not just structurally different. Such work would also enable
a focus on increasing the VOC and therefore even further
improve device performance.
(5) Jorgensen, M.; Norrman, K.; Gevorgyan, S. A.; Tromholt, T.;
Andreasen, B.; Krebs, F. C. Adv. Mater. 2012, 24, 580−612.
(6) Park, J. K.; Jo, J.; Seo, J. H.; Moon, J. S.; Park, Y. D.; Lee, K.;
Heeger, A. J.; Bazan, G. C. Adv. Mater. 2011, 23, 2430−2435.
(7) Ashraf, R. S.; Schroeder, B. C.; Bronstein, H. A.; Huang, Z. G.;
Thomas, S.; Kline, R. J.; Brabec, C. J.; Rannou, P.; Anthopoulos, T. D.;
Durrant, J. R.; McCulloch, I. Adv. Mater. 2013, 25, 2029−2034.
(8) Johnson, K.; Huang, Y. S.; Huettner, S.; Sommer, M.; Brinkmann,
M.; Mulherin, R.; Niedzialek, D.; Beljonne, D.; Clark, J.; Huck, W. T.
S.; Friend, R. H. J. Am. Chem. Soc. 2013, 135, 5074−5083.
(9) (a) Chen, L. M.; Hong, Z. R.; Li, G.; Yang, Y. Adv. Mater. 2009,
21, 1434−1449. (b) Peet, J.; Kim, J. Y.; Coates, N. E.; Ma, W. L.;
Moses, D.; Heeger, A. J.; Bazan, G. C. Nat. Mater. 2007, 6, 497−500.
(10) (a) Tan, Z. A.; Zhang, W. Q.; Zhang, Z. G.; Qian, D. P.; Huang,
Y.; Hou, J. H.; Li, Y. F. Adv. Mater. 2012, 24, 1476−1481. (b) He, Z.
C.; Zhong, C. M.; Huang, X.; Wong, W. Y.; Wu, H. B.; Chen, L. W.;
Su, S. J.; Cao, Y. Adv. Mater. 2011, 23, 4636−4643.
(11) Li, Y. W.; Guo, Q.; Li, Z. F.; Pei, J. N.; Tian, W. J. Energy
Environ. Sci. 2010, 3, 1427−1436.
(12) (a) Kim, Y.; Cook, S.; Tuladhar, S. M.; Choulis, S. A.; Nelson, J.
Nat. Mater. 2006, 5, 197−203. (b) Li, G.; Shrotriya, V.; Huang, J. S.;
Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y. Nat. Mater. 2005, 4, 864−
868.
ASSOCIATED CONTENT
■
S
* Supporting Information
(13) (a) Piliego, C.; Holcombe, T. W.; Douglas, J. D.; Woo, C. H.;
Beaujuge, P. M.; Frechet, J. M. J. J. Am. Chem. Soc. 2010, 132, 7595−
́
Experimental details of synthesis and additional character-
ization; figures and tables as described in the text; description of
the material included. This material is available free of charge
via the Internet at http://pubs.acs.org.
7597. (b) Rivnay1, J.; Toney, M. F.; Zheng, Y.; Kauvar, I. V.; Chen, Z.
H.; Wagner, V.; Facchetti, A.; Salleo, A. Adv. Mater. 2010, 22, 4359−
4363. (c) Mei, J. G.; Kim, D. H.; Ayzner, A. L.; Toney, M. F.; Bao, Z.
H. J. Am. Chem. Soc. 2011, 133, 20130−20133. (d) Beaujuge, P. M.
AUTHOR INFORMATION
́
Frechet, J. M. J. J. Am. Chem. Soc. 133, 20009−20029. (e) Mei, J. G.;
■
Bao, Z. N. Chem. Mater. 2013, 26, 604−615. (f) Henson, Z. B.;
Corresponding Authors
tianshi.qin@csiro.au
Mullen, K.; Bazan, G. C. Nat. Chem. 2012, 4, 699−704.
̈
(14) Hendriks, K. H.; Heintges, G. H. L.; Gevaerts, V. S.; Wienk, M.
M.; Janssen, R. A. J. Angew. Chem., Int. Ed. 2013, 52, 8341−8344.
(15) Kanimozhi, C.; Yaacobi-Gross, N.; Chou, K. W.; Amassian, A.;
Anthopoulos, T. D.; Patil, S. J. Am. Chem. Soc. 2012, 134, 16532−
16535.
scott.watkins@csiro.au
muellen@mpip-mainz.mpg.de
Notes
The authors declare no competing financial interest.
6054
dx.doi.org/10.1021/ja500935d | J. Am. Chem. Soc. 2014, 136, 6049−6055

Journal of the American Chemical Society
Article
(16) (a) Huo, L. J.; Zhang, S. Q.; Guo, X.; Xu, F.; Li, Y. F.; Hou, J. H.
Angew. Chem., Int. Ed. 2011, 50, 9697−9702. (b) Huo, L. J.; Hou, J.
H.; Zhang, S. Q.; Chen, H. Y.; Yang, Y. Angew. Chem., Int. Ed. 2010,
49, 1500−1503. (c) Wang, M.; Hu, X. W.; Liu, P.; Li, W.; Gong, X.;
Huang, F.; Cao, Y. J. Am. Chem. Soc. 2011, 133, 9638−9641. (d) Dou,
L. T.; Gao, J.; Richard, E.; You, J. B.; Chen, C. C.; Cha, K. C.; He, Y. J.;
Li, G.; Yang, Y. J. Am. Chem. Soc. 2012, 134, 10071−10079. (e) Yang,
T. B.; Wang, M.; Duan, C. H.; Hu, X. W.; Huang, L.; Peng, J. B.;
Huang, F.; Gong, X. . Energy Environ. Sci. 2012, 5, 8208−8214.
(f) Zhou, J. Y.; Zuo, Y.; Wan, X. J.; Long, G. K.; Zhang, Q.; Ni, W.;
Liu, Y. S.; Li, Z.; He, G. R.; Li, C. X.; Kan, B.; Li, M. M.; Chen, Y. S. J.
Am. Chem. Soc. 2013, 135, 8484−8487.
(33) Lee, J. K.; Ma, W. L.; Brabec, C. J.; Yuen, J.; Moon, J. S.; Kim, J.
Y.; Lee, K.; Bazan, G. C.; Heeger, A. J. J. Am. Chem. Soc. 2008, 130,
3619−3623.
(34) He, Y. J.; Chen, H. Y.; Hou, J. H.; Li, Y. F. J. Am. Chem. Soc.
2010, 132, 1377−1382.
(35) Roncali, J. Acc. Chem. Res. 2009, 42, 1719−1730.
(17) (a) Thomas, S. P.; Love, J. A.; Nguyen, T. Q.; Bazan, G. C. Adv.
Mater. 2012, 24, 3646−3649. (b) Zhang, Y.; Chien, S. C.; Chen, K. S.;
Yip, H. L.; Sun, Y.; Davies, J. A.; Chen, F. C.; Jen, A. K. Y. Chem.
Commun. 2011, 47, 11026−11028. (c) Albrecht, S.; Janietz, S.;
Schindler, W.; Frisch, J.; Kurpiers, J.; Kniepert, J.; Inal, S.; Pingel, P.;
Fostiropoulos, K.; Koch, N.; Neher, D. J. Am. Chem. Soc. 2012, 134,
14932−14944. (d) Li, Y. X.; Zou, J. Y.; Yip, H. L.; Li, C. Z.; Zhang, Y.;
Chueh, C. C.; Intemann, J.; Xu, Y. X.; Liang, P. W.; Chen, Y.; Jen, A.
K. Y. Macromolecules 2013, 46, 5497−5503. (e) Xiao, L.; Liu, B.; Chen,
X. W.; Li, Y. F.; Tang, W. J.; Zou, Y. P. RSC Adv. 2013, 3, 11869−
11876.
(18) Duan, C. H.; Cai, W. Z.; Huang, F.; Zhang, J.; Wang, M.; Yang,
T. B.; Zhong, C. M.; Gong, X.; Cao, Y. Macromolecules 2010, 43,
5262−5268.
(19) Beaujuge, P. M.; Tsao, H. N.; Hansen, M. R.; Amb, C. M.;
Risko, C.; Subbiah, J.; Choudhury, K. R.; Mavrinskiy, A.; Pisula, W.;
Bredas, J. L.; So, F.; Mullen, K.; Reynolds, J. R. J. Am. Chem. Soc. 2012,
́
̈
134, 8944−8957.
(20) (a) Henson, Z. B.; Welch, G. C.; van der Poll, T.; Bazan, G. C. J.
Am. Chem. Soc. 2012, 134, 3766−3779. (b) Ying, L.; Hsu, B. Y.; Zhan,
H. M.; Welch, G. C.; Zalar, P.; Perez, L. A.; Kramer, E. J.; Nguyen, T.
Q.; Heeger, A. J.; Wong, W. Y.; Bazan, G. C. J. Am. Chem. Soc. 2011,
133, 18538−18541.
(21) (a) Szarko, J. M.; Guo, J. C.; Liang, Y. Y.; Lee, B.; Rolczynski, B.
S.; Strzalka, J.; Xu, T.; Loser, S.; Marks, T. J.; Tu, L. P.; Chen, L. X.
Adv. Mater. 2010, 22, 5468−5472. (b) Yiu, A. T.; Beaujuge, P. M.; Lee,
O. P.; Woo, C. H.; Toney, M. F.; Frec
134, 2180−2185.
́
het, J. M. J. Am. Chem. Soc. 2012,
(22) Yue, W.; Zhao, Y.; Shao, S. Y.; Tian, H. K.; Xie, Z. Y.; Geng, Y.
H.; Wang, F. S. J. Mater. Chem. 2009, 19, 2199−2206.
(23) Davis, R. J.; Lloyd, M. T.; Ferreira, S. R.; Bruzek, M. J.; Watkins,
S. E.; Lindell, L.; Sehati, P.; Fahlman, M.; Anthony, J. E.; Hsua, J. W. P.
J. Mater. Chem. 2011, 21, 1721−1729.
(24) Zhu, L.; Cheng, S. Z. D.; Calhoun, B. H.; Ge, Q.; Quirk, R. P.;
Thomas, E. L.; Hsiao, B. S.; Yeh, F. J.; Lotz, B. J. Am. Chem. Soc. 2000,
122, 5957−5967.
(25) Zhang, M.; Yang, C.; Mishra, A. K.; Pisula, W.; Zhou, G.;
Schmaltz, B.; Baumgarten, M.; Mullen, K. Chem. Commun. 2007, 17,
̈
1704−1706.
(26) Mei, J. G.; Kim, D. H.; Ayzner, A. L.; Toney, M. F.; Bao, Z. N. J.
Am. Chem. Soc. 2011, 133, 20130−20133.
(27) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(28) Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J. P. J.
Am. Chem. Soc. 1985, 107, 3902−3909.
(29) (a) Tsao, H. N.; Cho, D. M.; Park, I.; Hansen, M. R.;
Mavrinskiy, A.; Yoon, D. Y.; Graf, R.; Pisula, W.; Spiess, H. W.;
Mullen, K. J. Am. Chem. Soc. 2011, 133, 2605−2612. (b) Osaka, I.;
̈
́
Zhang, R.; Sauve, G.; Smilgies, D. M.; Kowalewski, T.; McCullough, R.
D. J. Am. Chem. Soc. 2009, 131, 2521−2529.
(30) Rose, A. Phys. Rev. 1955, 97, 1538−1544.
(31) (a) Kreouzis, T.; Baldwin, R. J.; Shkunov, M.; McCulloch, I.;
Heeney, M.; Zhang, W. Appl. Phys. Lett. 2005, 87, 172110.
(b) Murgatroyd, P. N. J. Phys. D: Appl. Phys. 1970, 3, 151−156.
(32) Huang, F.; Wu, H. B.; Wang, D. L.; Yang, W.; Cao, Y. Chem.
Mater. 2004, 16, 708−716.
6055
dx.doi.org/10.1021/ja500935d | J. Am. Chem. Soc. 2014, 136, 6049−6055