Azido-Functionalized Thiophene as a Versatile Building Block to Cross-Link Low-Bandgap Polymers

Article abstract of DOI:10.1021/acs.macromol.5b02659

We unveil a concept for the design of cross-linkable semiconducting polymers that is based on a modular tercopolymerization which stands out by its low synthetic effort, easy accessibility, and its broad range of applications. 3-(6-Azidohexyl)thiophene wa

Full text of DOI:10.1021/acs.macromol.5b02659

Article
pubs.acs.org/Macromolecules
Azido-Functionalized Thiophene as a Versatile Building Block To
Cross-Link Low-Bandgap Polymers
Christian J. Mueller, Tobias Klein, Eliot Gann, Christopher R. McNeill, and Mukundan Thelakkat*
†
†
‡,§
§
,†
†
Applied Functional Polymers, Macromolecular Chemistry I, University of Bayreuth, 95440 Bayreuth, Germany
‡
Australian Synchrotron, 800 Blackburn Road, Clayton, VIC 3168, Australia
§
Department of Materials Science and Engineering, Monash University, Wellington Road, Clayton, Victoria 3800, Australia
*
S Supporting Information
ABSTRACT: We unveil a concept for the design of cross-
linkable semiconducting polymers that is based on a modular
tercopolymerization which stands out by its low synthetic effort,
easy accessibility, and its broad range of applications. 3-(6-
Azidohexyl)thiophene was used as a comonomer in the synthesis
of a variety of low-bandgap copolymers using different polymer-
ization techniques such as Suzuki−Miyaura cross-coupling and
Stille cross-coupling. We show that when only a small amount
(
5−10 mol %) of azide groups is introduced into the polymers,
the impact on absorption and electrochemical properties
HOMO/LUMO values) is negligible. The small amount of
(
azide functionality is however enough to obtain polymers that can
easily be cross-linked by UV illumination. Thermal stability of the
solid state packing and alignment is studied in neat polymer thin
films as well as in blends with [6,6]-phenyl-C -butyric acid methyl ester (PC BM) as a relevant model blend system. Solvent
7
1
70
resistivity of these polymer films is investigated by absorption and photoluminescence measurements. It is finally shown in
organic field effect transistors that the introduction of 10% azide-functionalized monomer does not considerably influence hole
2
−1 −1
transport mobility (0.20−0.45 cm V s ).
INTRODUCTION
into the side chains of their respective monomers. These
groups include alkyl halides, azides, oxetanes, or vinylenes that
■
Since the successful introduction of polymer bulk hetero-
junction solar cells in 1995, tremendous progress has led to
bulk heterojunction solar cell devices exceeding 10% power
conversion efficiency (PCE). Aggregation and morphology
control in the active layer of those bulk-heterojunction solar
cells has been identified as an important key in accessing high-
efficiency solar cells above 10% PCE. When moving from the
research level to commercial production, morphological
stability becomes indispensable not only for long lifetimes
but also for the particular processing methods. Roll-to-roll
7
−14
1
,2
can then be used for cross-linking, mostly by UV curing.
In
the case of cross-linking using azides, highly reactive nitrenes
are formed which can undergo a variety of cross-linking
reactions, such as C−H insertions, CC cycloadditions, and
3
12,14
other triplet reactions.
Taking into consideration that the
3
new generation of donor−acceptor alternating copolymers are
prepared from specifically functionalized monomers (e.g.,
borolanes, stannyl derivatives), this approach requires the
laborious modification of one of the comonomers used; i.e.,
several additional synthetic steps are necessary to render a
particular polymer system cross-linkable. This increases the
material costs and thus impairs commercialization. The second
concept relies on cross-linking using low molecular weight
(
R2R) printing with high throughput at appreciable web
speeds, for example, demands fast drying of printed active
layers, and the morphology must therefore be stable under
repeated curing conditions which are commonly above
15−18
4
additives such as bisazides or dithiols.
In this approach,
1
00 °C.
nonvolatile additives are introduced that might affect photo-
physics of charge generation as well as charge transport and
extraction in the final blends. Even though each cross-linkable
functional group has its advantages and disadvantages, we
concentrate in our work on azides to test the generality of our
Cross-linking of the polymer offers a viable route to stable5
morphologies as it allows for locking-in of the morphology
once an optimal condition has been attained, e.g., by tuning the
printing parameters or postprocessing, such as thermal or vapor
6
annealing. Two different concepts have been shown to offer
good cross-linking efficiency leading to thermally stable blends
that exhibit very good PCE retention after aging the solar cells.
In the first approach semiconducting polymers can be made
cross-linkable by synthetically introducing cross-linkable groups
Received: December 31, 2015
Revised: April 21, 2016
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.macromol.5b02659
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 1. Modular concept for cross-linkable tercopolymers using 3-(6-azidohexyl)thiophene as a building block for Stille or Suzuki
polycondensation.
Figure 2. Synthesis of the azide-functionalized building block 5. Reagents and conditions: (i) NaOMe, EtOH, 2 h, reflux; (ii) Mg, Et O,
2
Ni(dppp)Cl , 3 days, reflux; (iii) DCM, 1.5 h, reflux; (iv) THF, 24 h, rt; (v) DMSO, 2 h, rt.
2
novel approach for different polymerization reactions. Azides
are reported to cross-link via UV illumination due to the
formation of nitrenes, which can then undergo alkyl side-chain
can be obtained that exhibits repeating units y which are
identical to the regular strictly alternating copolymer and a
repeating unit x consisting of the azido-functionalized
thiophene 5 and the comonomer carrying the complementary
functional groups.
1
5
insertion reactions.
In this contribution, we unveil a third concept for cross-
linking semiconducting polymers that is based on a modular
tercopolymerization. We avoid the weaknesses of both above-
mentioned concepts, namely, the synthetic effort of modifying
monomers required for each particular polymer system and the
dilution of electroactive materials and introduction of an at best
electronically nonactive cross-linking additive. Instead, we show
that a simple azide-functionalized thiophene comonomer can
be used in small quantities in diverse polymerizations to give
various tercopolymers with an azide content that allows
efficient cross-linking but without altering any of the optical
and electronical properties of the final polymers. This paves the
way for effectively synthesizing various low-bandgap copoly-
mers with inherent cross-linkability without the need of any
additional synthetic steps to modify the respective electron-rich
or electron-deficient monomers. A detailed analysis of structure
formation and charge carrier transport is undertaken to evaluate
the influence of cross-linking.
The synthetic route of the azide-functionalized comonomer
is shown in Figure 2. Starting from 1,6-dibromohexane, we
2
1
followed a previously reported route to obtain 3-(p-
methoxyphenoxy)hexyl bromide 1. Conversion into the
Grignard compound and Kumada cross-coupling using 3-
bromothiophene yielded 3-(6-(4-methoxyphenoxy)hexyl)-
thiophene 2. The p-methoxyphenyl (PMP) protecting group
was then cleaved off by the use of boron tribromide and the
resulting intermediate simultaneously converted to the ω-alkyl
22
halide 3. Dibromination using N-bromosuccinimide almost
quantitatively gave 2,5-dibromo-3-(6-bromohexyl)thiophene 4,
and finally a Finkelstein reaction using sodium azide yielded 3-
(
6-azidohexyl)-2,5-dibromothiophene 5 which was used as the
comonomer in tercopolymerizations. We preferred this route
23
over a previously published one in order to avoid a mixture of
2
-(6-bromohexyl)thiophene and 3-(6-bromohexylthiophene)
due to the “halogen shuffling” reaction after lithiation. This
mixture is extremely hard to purify. The route shown here
results in the desired substitution product via Grignard
metathesis.
We chose four different polymer systems which were selected
due to their promising semiconductor properties and wide-
spread use in the literature: three from the diketopyrrolopyrrole
RESULTS AND DISCUSSION
■
Synthesis. The synthetic strategy pursued in our cross-
linkable tercopolymer approach is based on a versatile and
common building block that can be applied to different
semiconducting low-bandgap polymers without having too
much influence on their final properties. We hence chose
thiophene as an almost ubiquitous motif in conjugated
polymers. The functional groups for polymerization were
chosen to be bromines, not only in order to keep the synthesis
route as short as possible but also mostly as it can be employed
in many different polymerization techniques that are commonly
used to synthesize state-of-the-art low-bandgap polymers, such
as Stille, Suzuki, Yamamoto, Negishi, Kumada, and Heck cross-
coupling polymerizations. It has also been shown earlier that no
strictly alternating donor−acceptor sequence is required for
lowering the optical gap and random or statistical copoly-
(
PDPP) family and one from the carbazole−benzothiadiazole
(
PCDTBT) family. PDPP[T] -T (P1−P4) is a widely used and
2
investigated low-bandgap donor for use in organic solar cells,
24−26
also showing excellent field effect transistor mobilities.
PDPP[Py] -T (P5/P6) was selected due to its remarkable n-
type properties and promising bulk electron mobility.
PDPP[T] -Ph (P7/P8) is another easily accessible and well-
established donor polymer used in solar cells which has
gained a lot of interest due to the high power conversion
efficiencies (7.4%) obtained with this material. Finally, we
also investigate a PCDTBT system (P9/P10) as another
widespread donor material which is not based on diketo-
2
2
7
2
28
29
19,20
merization is a viable option.
When the azide-functionalized
30
comonomer 5 is added in small quantities (less than 5 wt %) to
AA/BB-type polycondensations, a cross-linkable tercopolymer
pyrrolopyrroles. Moreover, we verify the suitability of our
approach for Suzuki as well as Stille polycondensations.
B
DOI: 10.1021/acs.macromol.5b02659
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 3. Stille polycondensations of typical hole conducting low-bandgap polymers PDPP[T] -T (P1−P4) and typical electron conducting low-
2
bandgap polymers PDPP[Py] -T (P5, P6) with a varying degree of azide functionalization.
2
Figure 4. Suzuki polycondensation of a PDPP system PDPP[T] -Ph (P7, P8) and another low-bandgap reference system PCDTBT (P9, P10) with
2
and without azide functionalization.
For the cross-linkable PDPPs, polymerizations were
performed by using either Stille polycondensation (see Figure
Supporting Information (section 5.1). It also is evident from
the H NMR that the azide group is suitable for the different
1
3) or Suzuki polycondensation (see Figure 4), whereas the
polymerization conditions and survives in high yield under
cross-linkable PCDTBT was obtained via Suzuki polyconden-
sation (Figure 4). All polymers were purified by repeated
reprecipitations and Soxhlet extractions. The removal of
unreacted monomers was verified using thin layer chromatog-
raphy (TLC). Feed ratios (f , f , f ) of the three monomers a, b,
and c as well as resulting built-in ratios determined by H NMR
spectroscopy are summarized in Table 1. The mole fractions χ
of the repeating units x and y determined by NMR are very
similar to the calculated values expected from the feed ratio and
are well within the error of polymer NMR spectroscopy.
Detailed information on the calculation of the built-in ratio of
the azido-functionalized thiophene comonomer is given in the
different reaction conditions by comparing the methylene
protons next to the azide group (H ) and next to the thiophene
g
core (H ). As given in Table S1 (see Supporting Information),
b
the ratio is approximately 1 within the errors of polymer NMR
spectroscopy. The molecular weights and the degrees of
polymerization (DP) obtained from gel permeation chroma-
tography (GPC) as well as the expected average number of
azide moieties per polymer chain calculated from the feed ratio
are summarized in Table 1. The elugrams of the individual
polymers are shown in the Supporting Information. The azide-
functionalized tercopolymers show similar molecular weights
and degrees of polymerization compared to their non-
a
b
c
1
C
DOI: 10.1021/acs.macromol.5b02659
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Table 1. Molecular Characteristics of the Synthesized Polymers
a
a
b
b
c
d
e
f
g
h
polymer
fa
fb
χf(y) [χNMR(y)]
χf(x) [χNMR(x)]
Mn [kg/mol] Mw [kg/mol]
Đ
DP
n(N3)
wt %(N3) [%]
PDPP[T] -T system
2
P1
P2
P3
P4
0.00
0.05
0.10
0.25
0.50
0.45
0.40
0.25
1.00 [1.00]
0.90 [0.92]
0.80 [0.85]
0.50 [0.59]
0.00 [0.00]
0.10 [0.08]
0.20 [0.15]
0.50 [0.41]
36.9
31.6
22.2
12.9
146
254
3.94
8.03
3.82
2.13
44
41
31
23
0
4
0.0
2.7
84.6
6
5.7
27.5
11
18.5
PDPP[Py] -T system
2
P5
P6
0.00
0.05
0.50
0.45
1.00 [1.00]
0.90 [0.92]
0.00 [0.00]
0.10 [0.08]
19.6
18.3
54
2.76
2.89
24
24
0
2
0.0
2.7
52.7
PDPP[T] -Ph system
2
P7
0.00
0.05
0.50
0.45
1.00 [1.00]
0.90 [0.92]
0.00 [0.00]
0.10 [0.08]
27.9
31.6
46.5
74.1
1.67
2.35
34
41
0
4
0.0
2.7
P8
PCDTBT system
P9
0.00
0.05
0.50
0.45
1.00 [1.00]
0.90 [0.90]
0.00 [0.00]
0.10 [0.10]
16.3
36.7
15.3
2.25
2.15
23
10
0
1
0.0
3.0
P10
7.11
a
b
Feed ratio f of the respective comonomers a and b as identified in Figures 3 and 4; the feed ratio of c f was kept constant at 0.50. Molar fraction χ
c
1
of repeating unit y and x according to feed ratio (χ ) and calculated from H NMR (χNMR) in brackets; see Supporting Information section 5.1 for
f
c
details. Number-average molecular weight, determined by gel permeation chromatography at 120 °C using 1,2,4-trichlorobenzene as the eluent.
d
e
Weight-average molecular weight, determined by gel permeation chromatography at 120 °C using 1,2,4-trichlorobenzene as the eluent. Dispersity,
f
determined by gel permeation chromatography at 120 °C using 1,2,4-trichlorobenzene as the eluent. Degree of polymerization estimated from M
and the average molecular weight of the polymers repeating unit MRU (for details see Supporting Information, Table S2). Average number of azide
moieties per polymer chain estimated from DP and χRU(x). Weight percentage of azido functionalized thiophene in the final polymers calculated by
n
g
h
χf(x) × 207.29 g/mol / MRU, where 207.29 g/mol is the molecular weight of the azido functionalized thiophene moiety and MRU is the average
molecular weight of the polymer’s repeating unit given in Supporting Information Table S2.
Figure 5. Representative infrared spectroscopy of a Stille polycondensation product (PDPP[T] -T system, P1/P2) and a Suzuki polycondensation
2
product (PDPP[T] -Ph, P7/P8) of the tercopolymers with 5% azide repeating unit content (x = 0.1, red lines) and their respective
2
−1
nonfunctionalized reference copolymers (black lines). Insets show magnification of the characteristic azide vibration at 2100 cm . Infrared spectra of
the polymer pairs P5/P6 and P9/P10 are given in the Supporting Information, Figure S12.
functionalized reference copolymers for reasonably small
degrees of azide-functionalization. Assuming full incorporation
of the feed ratios of 5 to 25 mol %, the wt % of azide-
functionalized thiophene in the final polymers was in the range
of only 2.7% to 18.5% for P2 to P4.
reached the precipitation limit in pressurized chlorobenzene at
180 °C, higher molecular weights of tercopolymers with high
azide content are not accessible in this system. Moreover, the
minimum degree of azide functionalization for making the
polymers sufficiently insoluble was tested in the series P1−P4,
and it was found that a molar fraction of 5% content (x = 0.1)
was sufficient (vide infra). Therefore, only 5 mol %
functionalization was carried out in further systems (P5−
The PDPP[T] -T system (P1−P4) has the following
2
−1
number-average molecular weights: 36.9 kg mol for P1,
3
1.6 kg mol− for P2, 22.2 kg mol for P3, and 12.9 kg mol
1
−1
−1
for P4 corresponding to 0, 5, 10, and 25 mol % of
functionalized thiophene, respectively. The decreasing molec-
ular weight with increasing degree of functionalization can be
explained by the lower solubility of the tercopolymers with a
higher x fraction due to the small alkyl chains on 5 when
compared to the DPP comonomer. As all polymerizations
P10). The PDPP[Py] -T (P5/P6) system which was also
2
prepared using Stille polycondensation shows very comparable
−1
molecular weights of 19.6 kg mol for the nonfunctionalized
reference polymer P5 and 18.3 kg mol− for the polymer P6
containing 10% azido-functionalized repeating units with
almost identical dispersities. Concerning Suzuki polycondensa-
1
D
DOI: 10.1021/acs.macromol.5b02659
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 6. Absorption and photoluminescence spectra of cross-linkable tercopolymers compared to their respective nonfunctionalized reference
polymers: PDPP[T] -T system P1−P4 with varying monomer ratios in (a) solution and (b) thin film; PDPP[T] -Ph system P7/P8 in (c) solution
2
2
and (d) thin film. See Supporting Information Figure S13 for absorption and photoluminescence spectra of P5/P6 and P9/P10.
tion, the PDPP[T] -Ph system (P7/P8) also shows comparable
Table 2. Electrochemical Properties of Azide-Functionalized
Polymers Compared to the Nonfunctionalized Reference
Polymers
2
−1
values of 27.9 kg mol for the nonfunctionalized P7 and 31.6
kg mol− for the tercopolymer P8 containing 10% azido-
functionalized repeating units. Although the dispersity is a little
higher for the functionalized polymer (2.35) than for the
reference (1.67), no extreme broadening or tailing can be
observed in the GPC elugram; thus, there is no indication for
cross-linking under reaction and purification conditions. Only
the functionalized PCDTBT polymer P10 shows a considerably
1
a
b
c
d
e
IP
EA
ECV
λonset
Eopt
polymer
[eV]
[eV]
[eV]
[nm]
[eV]
PDPP[T] -T system
2
P1
P2
P3
P4
−5.75
−5.69
−3.69
−3.70
2.04
1.99
928
938
933
911
1.34
1.32
1.33
1.36
−1
lower molecular weight (7.11 kg mol ) than its reference
−1
polymer P9 (16.3 kg mol ), which is mainly attributed to two
different polymerization protocols (i.e., catalyst systems)
employed.
PDPP[Py] -T system
2
P5
P6
−6.08
−6.14
−3.78
−3.80
2.30
2.34
720
724
1.72
1.71
IR Spectroscopy. In addition to the NMR analysis, the
successful azide functionalization was monitored by infrared
PDPP[T] -Ph system
2
P7
−5.72
−5.77
−3.65
−3.67
2.07
2.10
812
802
1.53
1.55
(IR) spectroscopy of thin polymer films. Comparisons of
P8
nonfunctionalized reference polymers and tercopolymers with
an azide comonomer content of 10% are shown in Figure 5 and
Figure S12 (Supporting Information). A characteristic azide
vibration band emerges in all functionalized copolymers at
around 2100 cm− showing successful azide functionalization of
the polymers.
Optical Properties. The optical properties of all the
polymers were investigated with respect to absorption as well as
photoluminescence (PL) emission. All spectra are shown in
Figure 6, and the individual onset wavelengths as well as optical
gaps determined from solid state absorption onset are
PCDTBT system
P9
−5.75
−5.79
−3.50
−3.52
b
2.25
2.27
665
685
1.81
1.86
P10
a
Ionization potential (HOMO). Electron affinity (LUMO) deter-
1
c
mined by cyclic voltammetry in thin film. Bandgap determined by
d
cyclic voltammetry in thin film. Absorption onset in film determined
e
therefrom. Optical gap determined therefrom.
offset already at the lowest degree of functionalization with x =
0
.1 in P2 (5 mol % azide monomer). This shift only becomes
pronounced at a functionalization ratio of x = 0.5 in P4 (25 mol
% azide monomer) showing an onset blue-shift of 44 nm (68
meV) and a peak shift of 86 nm (170 meV).
summarized in Table 2. The PDPP[T] -T (P1−P4) polymer
2
system with the widest range of degree of functionalization
allows to investigate the influence of the tercopolymer ratio x:y
on the optical properties and hence differences in the optical
gap as well as aggregation properties. Whereas the spectral
nature itself does not change considerably, the solution spectra
For lower functionalization degrees containing 5−10 mol %
azide monomer with x = 0.1−0.2 only a very small onset shift of
7−15 nm appears. Whereas the functionalized polymers can
still be distinguished from the reference polymers in the
solution absorption spectra, the solid state spectra (Figure 6b)
(Figure 6a) show a small hypsochromic shift of absorption
E
DOI: 10.1021/acs.macromol.5b02659
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
taken of polymer thin films spin-coated on glass are almost
identical for low to moderate functionalization degrees. Only
P4 with x = 0.5 shows a distinct small hypsochromic onset shift
which is mainly caused by the missing aggregation shoulder. P2
and P3 with x = 0.1 and x = 0.2, respectively, show an identical
absorption spectrum compared to the unfunctionalized
reference polymer P1. This trend is also observable in the
photoluminescence of these four polymers. Only P4 shows a
distinctly different emission spectrum in solution and a small
hypsochromic shift in the solid state compared to P1−P3. For
this PDPP system it can therefore be concluded that
functionalization degrees up to x = 0.2 can be used without
altering any solid state optical properties and that an immense
amount of functionalized comonomer 5 has to be used in order
to alter the optical properties considerably. It should be
Figure 7. Comparison of energy levels of the polymers determined by
cyclic voltammetry in thin film. Individual cyclic voltammograms are
shown in the Supporting Information, Figure S11.
emphasized that the repeating unit of the PDPP[T] -T system
2
comprises three thiophene moieties without addition of the
functionalized comonomer 5 and that the addition of another
thiophene as third comonomer only partically reduces the DPP
content in the tercopolymers, which is the reason for the
negligible change in optical properties. We therefore also
investigated other systems, for example the system PDPP[T]2-
Ph (P7/P8, Figure 6c,d), where the comonomer is not a
thiophene but a phenyl moiety, thus decreasing the overall
thiophene content in the polymer repeating unit. Taking this
one step further, even switching to a different DPP core in the
electrochemical properties in thin film that are dependent on
the energy levels of the materials and will not change upon
functionalization in the tercopolymers compared to the
reference polymers, at least for a functionalization of 5−10
mol %.
Solvent Resistivity. In order to check the necessary degree
of functionalization to obtain solvent resistivity of the polymers,
−2
thin films were cross-linked by UV (254 nm, 1780 μW cm )
exposure and then submerged in chloroform for 5 min followed
by excessive rinsing with chloroform. The films were dried and
analyzed by means of UV/vis absorption and photolumines-
cence, and the peak intensities were compared to the spectra of
the untreated films as obtained after spin-coating. The relative
change after UV exposure and after chloroform submerging and
rinsing for the polymers P2, P3, and P8 is given in Figure 8.
Because of the shift of the absorption spectrum in the case of
the highly functionalized P4 polymer (see Figure 6a,b), we
chose to omit this polymer in solvent resistivity checks as the
application of this polymer is limited by its changed
semiconductor properties, even though it has the highest
degree of functionalization. Neither absorption nor photo-
luminescence intensity changes after the films have been
exposed to UV radiation, even after 20 min, indicating that the
films are stable against UV light and that cross-linking is not
negatively influencing the optical properties associated with the
conjugated π-system of the polymer backbone. After rinsing the
UV-exposed films with chloroform, a functionalization degree
pyridine flanked DPP copolymer system PDPP[Py] -T (P5/
2
P6, Supporting Information Figure S13a,b) did not lead to a
change in optical properties upon introduction of the third
monomer. Finally, another low-bandgap polymer system,
PCDTBT (P9/P10, Supporting Information, Figure S13c,d),
was shown not to be influenced by tercopolymerization with 5.
As it was found that sufficient cross-linking in the PDPP[T] -T
2
system (P1−P4) with up to 80% retention of the polymer film
(vide infra) is possible with as little as 5 mol % azide
functionalization, we chose to pursue only this degree of
functionalization for all other copolymer systems (P5−P10) in
order to minimize the influence of chemical changes on the
polymer backbone. The absorption as well as emission spectra
of the functionalized tercopolymers are compared to the
reference polymers in solution (Figure 6c,e,g) and thin film
(Figure 6d,f,h), and it can be concluded that a functionalization
degree of x = 0.1 has only negligible, if any, influence on the
spectral response in the wide range of low-bandgap polymers
investigated in this study.
of x = 10% (5 mol % azide monomer) in the PDPP[T] -T
2
system (P2) requires a minimum UV exposure time of 10−15
min to obtain solvent-resistant films, which retain about 80% of
their initial absorption. When the functionalization degree is
increased to x = 20% (10 mol % azide monomer), the films
exhibit onsetting solvent resistivity even after 5 min of UV
exposure, again retaining 80% of the absorption. A saturation in
solvent resistivity can be obtained after 5−10 min. In the
Electrochemical Properties. For each polymer system,
one functionalized polymer is compared to its unfunctionalized
reference polymer. The electrochemical behavior of pairs of
similar systems with and without azide functionalization (P1/
P3, P5/P6, P7/P8, P9/P10) was representatively compared
with respect to reversibility as well as redox potentials. The
individual cyclic voltammograms are given in the Supporting
Information (Figure S11), and the individual energy levels are
listed in Table 2 and plotted for comparison in Figure 7. The
energy levels were calculated from the redox potentials
calibrated against the ferrocene/ferrocenium redox couple
using a published procedure taking into account solvent
PDPP[T] -Ph system a functionalization degree of 5 mol % is
2
enough to obtain full chloroform resistivity after only 1 min of
UV exposure. It should be noted that even when saturation in
solvent resistivity is reached, the maximal obtainable absorption
level was at around 80% of the initial absorption for all cases.
Thus, the degree of functionalization influences mainly the
onset time required for cross-linking, but for appreciable
solvent resistivity 5−10 mol % functionalization is sufficient.
Pictures of a typical experiment outcome is shown in Figure
9, showing two polymer thin films of P3 spin-coated on a glass
substrate with and without prior UV-cross-linking. Interest-
3
1
effects. All redox potentials were obtained by cyclic
voltammetry on polymer thin films with indium tin oxide
(ITO) as the working electrode. Neither the reversibility, the
absolute reduction and oxidation potentials, nor the individual
voltammograms are influenced by the introduction of the azide-
functionalized comonomer. It can hence be concluded that the
ingly, neither P6 (PDPP[Py] -T) nor P10 (PCDTBT), both
2
F
DOI: 10.1021/acs.macromol.5b02659
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 8. Comparison of absorption (black symbols) and photoluminescence (green symbols) for P2, P3, and P8 films after UV exposure and
chloroform rinsing indicating solvent resistivity for cross-linked films.
Figure 9. Immersion of an as-cast and cross-linked P3 thin film on glass substrates: (a) immersion in chloroform (the green hue in the solution with
the cross-linked sample is due to refraction of the green polymer thin film in the glass of the Petri dish); (b) samples after rinsing with chloroform.
with x = 10% (5 mol % azide functionalization), could be cross-
linked to a degree where resistivity against chloroform is
obtained. This could be attributed to the high intrinsic
solubility of those polymers. The pyridine flanked PDPP-
Table 3. GIWAXS Peak Analysis of Neat Polymer Thin Films
for Alkyl, π−π, and Backbone Stacking
a
dalkyl [nm] ζalkyl , [nm] dπ−π [nm] ζπ−π , [nm]
cf
b
d
ef
polymer
T
P1
P1
P2
P2
P5
P5
P6
P6
P7
P7
P8
P8
as cast
250 °C
as cast
250 °C
as cast
250 °C
as cast
250 °C
as cast
250 °C
as cast
250 °C
1.98(4)
2.01(4)
1.89(2)
2.00(5)
1.98(3)
2.03(5)
1.89(0)
1.98(5)
1.94(3)
1.95(3)
1.89(4)
1.95(2)
6.3(8)
25.2(3)
4.5(2)
0.401(3)
0.402(1)
0.380(1)
0.395(2)
0.393(1)
0.391(0)
3.5(1)
4.27(6)
3.2(2)
4.0(3)
3.4(2)
3.9(1)
[
Py] -M system, for example, is even soluble in hexane as has
2 co
27
been previously reported.
It can therefore be expected that chloroform provides
enough solubility even for cross-linked systems. This also
indicates that the degree of cross-linking should be tuned
depending on the solubility properties of the concerned
polymer system.
Solid State Structure. In order to elucidate the influence
of cross-linking on the solid state structure of the polymers in
thin films, i.e., crystallinity and the alignment with the substrate,
we have measured grazing incidence wide-angle X-ray scattering
13.5(1)
3.36(7)
25.1(2)
3.94(0)
14.0(1)
8.32(3)
16.2(1)
5.85(6)
15.6(2)
0.393(0)
0.382(2)
0.406(5)
0.402(2)
0.412(0)
2.74(2)
3.3(2)
3.43(3)
2.86(7)
3.58(4)
a
b
c
(GIWAXS) for the three different systems (P1/P2, P5/P6, and
Annealing process. Alkyl stacking spacing. Alkyl stacking coherence
d
length (determined from (100) peak in out-of-plane direction). π−π
P7/P8) comparing the unfunctionalized reference polymers to
their functionalized, cross-linked tercopolymers. The individual
spacing d and coherence length ζ values for the different
crystalline directions (along the alkyl-stacked lamella and the
π−π stacked faces) are summarized in Table 3. As an example,
Figure 10 shows a comparison of unfunctionalized P1 to the
e
stacking spacing. π−π stacking coherence length (determined from
f
1
st order π−π stacking). The crystal coherence length ζ gives
information about the distance over which order is maintained and is
defined as ζ = 2π/fwhm, where fwhm is the full width at half-
maximum of the first-order lamellar stacking or π−π stacking peak,
respectively. It is related to the Scherrer equation, which connects the
width of a peak to the crystal size.
azido-functionalized P2 (PDPP[T] -T system) in both as-cast
2
as well as UV-exposed and annealed films. UV exposure was
performed under ambient conditions; annealing was carried out
at 250 °C under an inert atmosphere. Whereas in P1 a
significant increase in the degree of edge-on alignment can be
observed after thermal annealing, showing up to the fourth-
order reflection of the alkyl-stacking peak (labeled h00) and
very large coherence lengths (25.2 nm) in both the in-plane
and out-of-plane direction, annealing of P2 only leads to a
moderate increase of edge-on crystallinity.
While edge-on crystallinity is dominant in P1, there is a
significant population of face-on stacking with much larger
coherence lengths evident by the quite sharp alkyl-stacking
reflections in-plane. Comparisons for the other polymer
systems (P5/P6, P7/P8) are given in Supporting Information
Figures S14 and S15, respectively. Similar behavior can be
observed in the pyridine flanked PDPP[Py] -T system when
2
comparing unfunctionalized P5 to functionalized P6 (see
G
DOI: 10.1021/acs.macromol.5b02659
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 10. GIWAXS scatter patterns of neat unfunctionalized (P1) and azide-functionalized (P2) PDPP[T] -T polymer thin films. Both the
2
reference P1 and the P2 film were exposed to UV light before thermal annealing.
Table 4. GIWAXS Peak Analysis of Polymer:PC BM Blend Systems for Alkyl, π−π, and Backbone Stacking
7
0
a
b
ch
ζalkyl , [nm]
d
eh
ζπ−π , [nm]
f
gh
ζbb , [nm]
blend
T
dalkyl [nm]
dπ−π [nm]
dbb [nm]
P1:PC BM
as cast
200 °C
as cast
200 °C
as cast
200 °C
as cast
200 °C
as cast
200 °C
as cast
200 °C
1.97(3)
2.05(4)
1.95(2)
1.94(1)
1.89(7)
2.01(6)
1.84(6)
1.90(1)
1.95(3)
1.95(4)
1.91(1)
7.8(1)
20.0(2)
12(1)
0.380(2)
0.358(7)
0.381(1)
0.381(1)
0.359(1)
0.389(1)
0.360(2)
0.369(1)
0.393(1)
0.391(1)
0.388(2)
0.397(0)
1.7(1)
2.4(1)
1.6(1)
1.6(1)
3.0(2)
2.1(1)
2.9(3)
1.8(1)
2.2(1)
2.3(1)
1.6(1)
6(1)
1.56(1)
1.43(1)
1.53(5)
1.46(3)
5.3(5)
3.8(3)
3.5(6)
3.5(6)
70
P1:PC BM
70
P2:PC BM
70
P2:PC BM
7.47(9)
4.5(2)
15.5(1)
6.7(6)
5.06(0)
9.3(1)
9.1(1)
7.58(8)
7.2(1)
70
P5:PC BM
70
P5:PC BM
70
P6:PC BM
70
P6:PC BM
70
P7:PC BM
1.54(1)
1.50(1)
1.54(1)
1.47(2)
6(1)
5.8(6)
5.2(7)
70
P7:PC BM
70
P8:PC BM
70
P8:PC BM
1.91(1)
2.9(5)
70
a
b
c
d
Annealing process. Alkyl stacking spacing. Alkyl stacking coherence length (determined from (100) peak in out-of-plane direction). π−π
e
f
g
stacking spacing. π−π stacking coherence length (determined from first-order π−π stacking). Backbone stacking spacing. Backbone stacking
coherence length (determined from first-order backbone stacking in in-plane direction). The crystal coherence length ζ gives information about the
h
distance over which order is maintained and is defined as ζ = 2π/fwhm, where fwhm is the full width at half-maximum of the first-order lamellar
stacking or π−π stacking peak, respectively. It is related to the Scherrer equation, which connects the width of a peak to the crystal size.
Supporting Information Figure S14). P5 shows a considerable
increase in crystallinity and well-aligned films with several
orders of alkyl spacing present both out-of-plane and in-plane
after thermal annealing. Again, the P6 film is somewhat
influenced by annealing but to a much smaller extent, showing
first-order peaks and considerably smaller coherence lengths
alignment after thermal annealing, independent of the azide-
content and cross-linking. In fact, P8 shows a clearer edge-on
conformation than the reference P7 as indicated by the
complete absence of the alkyl spacing peaks in the in-plane
direction in P8 after annealing. This points to face on
orientation in these films being hindered by cross-linking.
(
Table 3). Finally, in the PDPP[T] -Ph system comparing
2
The backbone spacing d and the respective backbone spacing
bb
unfunctionalized P7 to functionalized P8 (see Supporting
Information, Figure S15) cross-linking of P8 has much less of
an effect on both crystallinity and degree of alignment. Both
polymers show a strongly enhanced crystallinity and edge-on
coherence length ζbb could only be observed in P7 after
annealing and was found to be 1.53(0) nm and 10.0(5) nm,
respectively.
H
DOI: 10.1021/acs.macromol.5b02659
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 11. GIWAXS scatter patterns of unfunctionalized (P1) and azide-functionalized (P2) PDPP[T] -T polymer:PC BM blends. Both the
2
70
reference P1 and the P2 film were exposed to UV light before thermal annealing. Comparisons for different polymer systems (P5/P6, P7/P8) are
given in Supporting Information Figures S16 and S17, respectively.
Blends. Since the importance of cross-linking a low-bandgap
polymer is usually discussed as a route to stabilize the
morphology of blends with fullerene derivatives as in solar
cell applications, we tested the potential of our concept in
similar blends as well. For this, we have the selected the
polymer pairs P1/P2, P5/P6, and P7/P8 and tested the cross-
linking conditions. From our experience with the non-
functionalized reference polymers P1, P5, and P7, the
polymer:PC BM blend morphology is much more favorable
for solar cell applications when DIO is used as a solvent
additive. Hence, we chose to add DIO to the blends
investigated in GIWAXS. Additionally, we have also performed
GIWAXS measurements on polymer/fullerene blends before
and after cross-linking/annealing. 2D GIWAXS patterns were
taken for the as cast films as well as for films that were exposed
to UV radiation followed by thermal annealing at 200 °C for 15
min in an inert atmosphere. The individual spacings d and
coherence lengths ζ along the different crystalline directions
thermal annealing, indicating a lower chain mobility which
hinders improved packing upon annealing. The same effect can
also be observed in the blends employing the PDPP[Py] -T
2
system (see Supporting Information, Figure S16). Again the
unfunctionalized P5 blend shows an increasing crystallinity and
exhibits distinct second-order peaks upon annealing which are
lacking in the cross-linked and annealed P6 blend. In fact, the
scattering of the annealed P6 blend is identical to the as-cast P6
blend and the as-cast P5 blend, indicating successful cross-
linking and chain immobilization in this system. Interestingly,
70
the PDPP[T] -Ph system does not show any differences after
2
thermal annealing in the unfunctionalized P7 blend as well as in
the tercopolymer P8 blend having 5 mol % azide functionaliza-
tion (see Supporting Information Figure S17).
We want to highlight a noticeable peculiarity apart from any
cross-linking questions, namely, the fact that for the thiophene
flanked PDPP[T] -T (P1/P2) and PDPP[T] -Ph (P7/P8)
2
2
systems we consistently observe backbone stacking peaks when
blended with PC BM (Table 4), which are not observable in
(
1
alkyl, π−π, and backbone) are summarized in Table 4. Figure
70
1 shows the 2D scattering patterns for the PDPP[T] -T
the corresponding neat polymer thin films (Table 3), especially
in P1, P2, and P8. This is suggesting a PC BM-influenced
2
system comparing unfunctionalized P1 with functionalized P2
x = 0.1) as a typical example. Whereas the as-cast blend films
70
(
packing in the blends.
show almost identical scattering in both in-plane and out-of-
plane direction, upon annealing the P1 blend shows a distinct
increase in crystallinity with second- and third-order lamellar
stacking peaks appearing in the out-of-plane direction. Also, the
in-plane scattering shows a higher intensity and a narrower
peak for the first-order lamellar stacking peak, indicating a
larger crystal coherence length in the annealed sample. These
annealing effects are absent in the cross-linked P2 blend after
OFET. The performance of the polymers in organic field
effect transistors (OFET) was representatively investigated to
study the influence of azide incorporation on the PDPP[T] -T
2
system using the unfunctionalized P1 as a reference and P3
exhibiting a functionalization degree of x = 0.2 (10 mol % azide
monomer). OFET devices were prepared in a bottom gate/
bottom contact (BGBC) configuration. In order to get a clear
picture about the influence of incorporation of a third
I
DOI: 10.1021/acs.macromol.5b02659
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 12. OFET p-type transfer and output I−V characteristics after annealing at 250 °C for 15 min of the nonfunctionalized strictly alternating
PDPP[T] -T P1 and the tercopolymer containing 10% azide-functionalized repeating units P3. Solid and dashed lines represent the forward and
2
reverse scans in all graphs, respectively. The voltages given in the legend of the output characteristics are the applied gate voltages.
comonomer into this copolymer system, neither of the films has
been exposed to UV radiation for cross-linking before
characterizing the transistors. The mobility was averaged over
at least three individual devices. Both polymers show an
ambipolar behavior, although the p-channel is much more
dominant. Transfer and output I−V characteristics for p-
channel operation are shown in Figure 12. In the as-cast films
CONCLUSION
■
In conclusion, we have shown a new and highly modular
concept for the synthesis of photo-cross-linkable semiconduct-
ing polymers to obtain both solvent-resistant films as well as
thermally stable solid state packing and morphologies in thin
films of neat polymers and in PC BM blends after photo-cross-
70
linking. The synthetic effort was kept low by the universal use
of a common azido-functionalized thiophene comonomer in
four different copolymer systems without influencing the
optical and electrochemical properties of the polymers.
Successful incorporation of azide moieties in the functionalized
tercopolymers was proven by infrared and NMR spectroscopy.
The introduction of the additional monomer containing the
azide moiety did not considerably alter the optical or electrical
properties in thin films and devices as demonstrated in an
organic field effect transistor for a degree of functionalization
up to x = 0.2 (10 mol % azide monomer). We anticipate that
this cross-linking approach will provide an easily accessible
route toward a variety of inherently cross-linkable semi-
conductor polymers using a single universal cross-linkable
unit without the necessity of the addition of low molecular
weight cross-linking agents.
2
−1 −1
the hole mobility was μ = 0.28 ± 0.17 cm V
s
for P1
in the case of the
functionalized derivative P3. Upon annealing, the mobilities
h
2
−1 −1
compared to μ = 0.04 ± 0.01 cm V
s
h
2
−1 −1
increased considerably to μ (P1) = 0.45 ± 0.02 cm V
s
h
2
−1 −1
and μ (P3) = 0.19 ± 0.03 cm V
s , respectively. Even
h
though the final mobility of the azide-functionalized tercopol-
ymer P3 is a little lower than that of the reference polymer P1
after annealing, it is important to note that mobilities in the
same order of magnitude are accessible using this tercopolyme-
rization approach and that even a moderate azide-functionaliza-
tion of 10 mol % with x = 0.2 does obviously not drastically
limit the thermal annealing options of non-UV-cross-linked
films, allowing good postprocessing opportunities via thermal
annealing which is common for organic semiconducting
materials. The same holds true for n-type operation. Transfer
and output characteristics for n-channel operation are shown in
the Supporting Information, Figure S18. The transistors exhibit
a significant threshold voltage of 50 V, and hence output in the
EXPERIMENTAL SECTION
■
Materials and Methods. All reagents were used without further
purification unless otherwise noted. Microwave reactions were
conducted in sealed containers using a Biotage Initiator Eight+
n-channel can only be observed for gate voltages as high as 60−
1
microwave. H NMR (300 MHz) spectra of monomers were recorded
2
8
0 V. The electron mobilities are μ (P1) = 0.14 ± 0.01 cm V
1
e
on a Bruker AC 300 spectrometer. All polymer H NMR spectra were
recorded on a Varian INOVA 300 spectrometer at 393 K in 1,1,2,2-
−
1
−1
2
−1 −1
s
and μ (P3) = 0.06 ± 0.01 cm V s .
e
J
DOI: 10.1021/acs.macromol.5b02659
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1
tetrachloroethane (TCE) as solvent. The H spectra were referenced
internally by using the residual solvent resonances. Deuterated
solvents were obtained from Deutero. Gel permeation chromatog-
raphy (GPC) analysis was carried out on an Agilent (Polymer
Laboratories Ltd.) PL-GPC 220 high temperature chromatographic
unit equipped with DP, RI, and LS (15° and 90°) detectors and three
linear mixed bed columns of PLgel 13 μm (Olexis) with a linear MW
modified version of the NIKA small-angle scattering analysis
package.
3
3
Organic Thin Film Transistors. Substrates in bottom gate/
bottom contact configuration were bought from Fraunhofer IPMS
(OFET Gen. 4). Heavily n-doped silicon (doping at wafer surface: n ∼
17
−3
3 × 10 cm ) was used as substrate and gate electrode. Thermally
grown silicon oxide (230 ± 10 nm) was used as the gate dielectric.
Gold electrodes (30 on 10 nm ITO as adhesion layer) were used as
source and drain contacts. The channel width was 10 mm for all
devices and channel lengths of 10 and 20 μm were used. The
substrates were cleaned subsequently in acetone and 2-propanol in an
ultrasonic bath for 10 min each. Treatment in an ozone oven at 50 °C
for 20 min was followed by immersion in a 1 wt % solution of
octadecyltrichlorosilane in toluene at 60 °C for 60 min. After rinsing
with toluene and 2-propanol the substrates were dried in a nitrogen
stream, and the polymer was spin-coated from a 6 mg/mL chloroform
solution at 2000 rpm under ambient conditions. Devices were
measured in a nitrogen atmosphere using an Agilent B1500
Semiconductor Parameter Analyzer. The devices were annealed in a
−
1
operating range: 500−15 000 000 g mol . GPC analysis was
performed at 150 °C using 1,2,4-trichlorobenzene as the mobile
phase. The samples were prepared by dissolving the polymer (0.1 wt
) in the mobile phase solvent in an external oven and the solutions
were run without filtration. The molecular weights of the samples were
referenced to linear polystyrene (M = 162−6 000 000 g mol , K =
2.100, and Alpha = 0.707) and were not corrected with K and Alpha
values for the measured sample. Cyclic voltammetry was performed
under moisture- and oxygen-free conditions using a 0.1 M tetra-n-
butylammonium hexafluorophosphate in acetonitrile electrolyte
solution. A standard three-electrode assembly connected to a
potentiostat (model 263A, EG&G Princeton Applied Research) was
used at a scanning rate of 100 mV s . The working electrode was a 10
Ω/□ ITO-coated glass substrate. The polymers were spin-coated onto
the ITO substrates from chloroform at 3 mg/mL and 1500 rpm to
obtain thicknesses of 10−20 nm. A platinum wire in acetonitrile was
used as counter electrode, and the quasi-reference electrode consisted
%
−1
w
1
−1
nitrogen atmosphere at a maximum of 0.9 ppm of O for 15 min
2
consecutively at the temperatures given in the main text. Mobilities
0
.5
were calculated from the slopes in the (I ) −V plots in the saturation
d
g
regime using eq 3, where I is the drain current, W the channel width,
d
L the channel length, C the capacitance, V the gate voltage, and V
i
g
T
of an Ag wire in an AgNO /acetonitrile solution (0.1 M). The
the threshold voltage.
3
measurements were calibrated with an external ferrocence/ferroce-
nium standard; IP and EA values were calculated considering the
W
2L
2
Id ≈
Cμ(V − V )
i
g
T
31
(3)
solvent effects as per a published procedure using eqs 1 and 2 where
+
the workfunction of Fc/Fc is taken to be −5.23 eV. The reduction
red
+
half-step potential E1 (vs Fc/Fc ) is negative whereas the oxidation
/2
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.macro-
mol.5b02659.
ox
/2
+
■
*
half-step potential E1 (vs Fc/Fc ) is positive.
S
red
+
EA ≈ −5.23 eV − E1 (vs Fc/Fc )
/2
(1)
(2)
ox
+
IP ≈ −5.23 eV − E1 (vs Fc/Fc )
/2
Monomer syntheses, characterization of monomers,
polymer syntheses, detailed characterization of polymers
including GPC, cyclic voltammetry, and additional
GIWAXS data (PDF)
Absorption measurements were carried out on a JASCO V-670
spectrophotometer. Photoluminescence measurements were carried
out on a JASCO FP-8600 spectrofluorometer; excitation wavelengths
were chosen at the maximum of the transition around 430 nm for all
samples. Optical properties in solution were measured in chloroform
at a concentration of 0.01 mg/mL and a path length of 10 mm; films
were spin-coated onto glass slides from a 7 mg/mL chloroform
solution at 1500 rpm. Solutions for spin-coating were prepared by
dissolving the polymer in chloroform at the given concentration and
stirring at 55 °C for 8 h. Cross-linking by UV exposure for the
resistivity experiments was performed using a benchtop UV lamp
AUTHOR INFORMATION
Corresponding Author
E-mail mukundan.thelakkat@uni-bayreuth.de (M.T.).
■
*
Notes
The authors declare no competing financial interest.
−2
ACKNOWLEDGMENTS
We thank Christina Saller (Universitat Bayreuth) for providing
operating at 254 nm with an intensity of 1780 μW cm ; cross-linking
for the GIWAXS experiments was performed using a commercial
system (Light Hammer 6, Fusion Systems Inc., 200−450 nm, 200 W
■
̈
the unfunctionalized PCDTBT as reference material. We
acknowledge financial support from the Bavarian State Ministry
of the Environment and Consumer Protection (Umwelt
NanoTech) and the Bavarian State Ministry of Education,
Science and the Arts (Solar technologies go hybrid). C.J.M.
thanks the Fonds der Chemischen Industrie for funding the
−2
cm on a LC6 benchtop conveyor operating at 3.5 cm/s).
GIWAXS. The GIWAXS measurements were conducted at the
32
SAXS/WAXS beamline of the Australian Synchrotron. Neat polymer
films were spin-coated from chloroform onto an octyltrichlorosilane/
−1
SiO modified silicon wafer from a 6 mg mL solution at 2000 rpm.
2
Blend samples were prepared by spin-coating polymer:PC BM blends
70
(
2
ratio 1:2) from chloroform at a total concentration of 18 mg/mL with
́
PhD with a Kekule scholarship. C.J.M. and T.K. kindly
vol % 1,8-diiodooctane (DIO) onto an MoO (15 nm) modified
acknowledge scholarships from the German National Academic
Foundation and support from the Elitenetzwerk Bayern (ENB),
Macromolecular Science. C.R.M. acknowledges support from
the Australian Research Council (DP130102616). This
research was undertaken in part on the SAXS/WAXS beamline
3
silicon wafer at 2000 rpm. After spin-coating the films were
−7
immediately put under ultrahigh vacuum (base pressure 10 mbar)
in order to evaporate remaining DIO. Highly collimated 9 keV X-rays
were calibrated to be at a tilt angle of 0 ± 0.01° when parallel to the
surface of each sample by use of a silicon crystal analyzer. A Dectris
Pilatus 1M detector collected 2D scattering patterns, including those
shown in Figures 10 and 11. Each scattering pattern was tiled together
from three 1 s images with the detector slightly moved between
exposures, such that the resulting image removes gaps between the
detector modules. The sample-to-detector distance was measured
using a silver behenate scattering standard. Data were analyzed using a
32
at the Australian Synchrotron, Victoria, Australia.
REFERENCES
■
(1) Halls, J. J. M.; Walsh, C. A.; Greenham, N.; Marseglia, E. A.;
Friend, R. H; Moratti, S. C.; Holmes, A. B. Efficient photodiodes from
interpenetrating polymer networks. Nature 1995, 376, 498.
K
DOI: 10.1021/acs.macromol.5b02659
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
(
2) Yu, G.; Gao, J.; Hummelen, J. C.; Wudl, F.; Heeger, A. J. Polymer
(21) Bau
̈
erle, P.; Wu
̈
rthner, F.; Heid, S. Einfache Synthese von 3-(ω-
Photovoltaic Cells: Enhanced Efficiencies via a Network of Internal
Donor-Acceptor Heterojunctions. Science 1995, 270, 1789.
̈
Halogenalkyl)thiophenen als Grundbausteinen fur funktionalisierte
Thiophene und Polythiophene. Angew. Chem. 1990, 102, 414.
(22) Buckel, F.; Persson, P.; Effenberger, F. Synthesis of Function-
alized Long-Chain Thiols and Thiophenols for the Formation of Self-
Assembled Monolayers on Gold. Synthesis 1999, 1999, 953.
(
3) Liu, Y.; Zhao, J.; Li, Z.; Mu, C.; Ma, W.; Hu, H.; Jiang, K.; Lin,
H.; Ade, H.; Yan, H. Aggregation and morphology control enables
multiple cases of high-efficiency polymer solar cells. Nat. Commun.
(23) Zhai, L.; Pilston, R. L.; Zaiger, K. L.; Stokes, K. K.; McCullough,
2
(
014, 5, 5293.
4) Hosel, M.; Søndergaard, R. R.; Jørgensen, M.; Krebs, F. C. Fast
R. D. A Simple Method to Generate Side-Chain Derivatives of
Regioregular Polythiophene via the GRIM Metathesis and Post-
polymerization Functionalization. Macromolecules 2003, 36, 61.
̈
Inline Roll-to-Roll Printing for Indium-Tin-Oxide-Free Polymer Solar
Cells Using Automatic Registration. Energy Technology 2013, 1, 102.
(
24) Bijleveld, J. C.; Zoombelt, A. P.; Mathijssen, S. G. J.; Wienk, M.
M.; Turbiez, M.; de Leeuw, D. M.; Janssen, R. A. J. Poly
diketopyrrolopyrrole-terthiophene) for Ambipolar Logic and Photo-
voltaics. J. Am. Chem. Soc. 2009, 131, 16616.
25) Bijleveld, J. C.; Verstrijden, R. A. M.; Wienk, M. M.; Janssen, R.
(
5) Griffini, G.; Douglas, J. D.; Piliego, C.; Holcombe, T. W.; Turri,
S.; Frechet, J. M.; Mynar, J. L. Long-term thermal stability of high-
efficiency polymer solar cells based on photocrosslinkable donor-
acceptor conjugated polymers. Adv. Mater. 2011, 23, 1660.
(
(
(
6) Rumer, J. W.; McCulloch, I. Organic photovoltaics: Crosslinking
for optimal morphology and stability. Mater. Today 2015, 18, 425.
7) Carle, J. E.; Andreasen, B.; Tromholt, T.; Madsen, M. V.;
A. J. Copolymers of diketopyrrolopyrrole and thienothiophene for
photovoltaic cells. J. Mater. Chem. 2011, 21, 9224.
(
(26) Venkatesan, S.; Adhikari, N.; Chen, J.; Ngo, E. C.; Dubey, A.;
Norrman, K.; Jorgensen, M.; Krebs, F. C. Comparative studies of
photochemical cross-linking methods for stabilizing the bulk hetero-
junction morphology in polymer solar cells. J. Mater. Chem. 2012, 22,
Galipeau, D. W.; Qiao, Q. Interplay of nanoscale domain purity and
size on charge transport and recombination dynamics in polymer solar
cells. Nanoscale 2014, 6, 1011.
2
(
4417.
8) Wantz, G.; Derue, L.; Dautel, O.; Rivaton, A.; Hudhomme, P.;
(27) Mueller, C. J.; Singh, C. R.; Fried, M.; Huettner, S.; Thelakkat,
M. High Bulk Electron Mobility Diketopyrrolopyrrole Copolymers
with Perfluorothiophene. Adv. Funct. Mater. 2015, 25, 2725.
(28) Bijleveld, J. C.; Gevaerts, V. S.; Di Nuzzo, D.; Turbiez, M.;
Mathijssen, S. G. J.; de Leeuw, D. M.; Wienk, M. M.; Janssen, R. A. J.
Efficient Solar Cells Based on an Easily Accessible Diketopyrrolo-
pyrrole Polymer. Adv. Mater. 2010, 22, E242.
(29) Li, W.; Hendriks, K. H.; Furlan, A.; Roelofs, W. S.; Meskers, S.
C.; Wienk, M. M.; Janssen, R. A. Effect of the fibrillar microstructure
on the efficiency of high molecular weight diketopyrrolopyrrole-based
polymer solar cells. Adv. Mater. 2014, 26, 1565.
Dagron-Lartigau, C. Stabilizing polymer-based bulk heterojunction
solar cells via crosslinking. Polym. Int. 2014, 63, 1346.
(
Lin, Y.-Z.; Bundgaard, E.; Krebs, F. C.; Zhan, X.-W.; Chen, H.-Z.
Solvent-resistant small molecule solar cells by roll-to-roll fabrication
via introduction of azide cross-linkable group. Synth. Met. 2014, 195,
9) Chen, M.-R.; Fan, C.-C.; Andersen, T. R.; Dam, H. F.; Fu, W.-F.;
2
(
99.
́
10) Kim, B. J.; Miyamoto, Y.; Ma, B.; Frechet, J. M. J.
Photocrosslinkable Polythiophenes for Efficient, Thermally Stable,
(30) Blouin, N.; Michaud, A.; Leclerc, M. A Low-Bandgap Poly(2,7-
Organic Photovoltaics. Adv. Funct. Mater. 2009, 19, 2273.
(
M. Y.; Frechet, J. M. J.; Oh, J. H.; Kim, B. J. Solvent-Resistant Organic
Transistors and Thermally Stable Organic Photovoltaics Based on
Cross-linkable Conjugated Polymers. Chem. Mater. 2012, 24, 215.
(
Black, C. T.; Grubbs, R. B. Photo-Cross-Linkable Azide-Functionalized
Polythiophene for Thermally Stable Bulk Heterojunction Solar Cells.
Macromolecules 2012, 45, 2338.
Carbazole) Derivative for Use in High-Performance Solar Cells. Adv.
Mater. 2007, 19, 2295.
11) Kim, H. J.; Han, A. R.; Cho, C.-H.; Kang, H.; Cho, H.-H.; Lee,
́
(
̈
31) Graf, K.; Rahim, M. A.; Das, S.; Thelakkat, M. Complementary
co-sensitization of an aggregating squaraine dye in solid-state dye-
sensitized solar cells. Dyes Pigm. 2013, 99, 1101.
12) Nam, C.-Y.; Qin, Y.; Park, Y. S.; Hlaing, H.; Lu, X.; Ocko, B. M.;
(32) Kirby, N. M.; Mudie, S. T.; Hawley, A. M.; Cookson, D. J.;
Mertens, H. D. T.; Cowieson, N.; Samardzic-Boban, V. A low-
background-intensity focusing small-angle X-ray scattering undulator
beamline. J. Appl. Crystallogr. 2013, 46, 1670.
(
13) Murray, K. A.; Holmes, A. B.; Moratti, S. C.; Friend, R. H. The
(33) Ilavsky, J. Nika: software for two-dimensional data reduction. J.
synthesis and crosslinking of substituted regioregular polythiophenes.
Synth. Met. 1996, 76, 161.
Appl. Crystallogr. 2012, 45, 324.
(
14) Murray, K. A.; Holmes, A. B.; Moratti, S. C. Rumbles, G.
Conformational changes in regioregular polythiophenes due to
crosslinking. J. Mater. Chem. 1999, 9, 2109.
(
15) Png, R. Q.; Chia, P. J.; Tang, J. C.; Liu, B.; Sivaramakrishnan, S.;
Zhou, M.; Khong, S. H.; Chan, H. S.; Burroughes, J. H.; Chua, L. L.;
Friend, R. H.; Ho, P. K. High-performance polymer semiconducting
heterostructure devices by nitrene-mediated photocrosslinking of alkyl
side chains. Nat. Mater. 2010, 9, 152.
(
16) Liu, B.; Png, R. Q.; Zhao, L. H.; Chua, L. L.; Friend, R. H.; Ho,
P. K. High internal quantum efficiency in fullerene solar cells based on
crosslinked polymer donor networks. Nat. Commun. 2012, 3, 1321.
(
17) Davis, A. R.; Carter, K. R. Controlling Optoelectronic Behavior
in Poly(fluorene) Networks Using Thiol−Ene Photo-Click Chemistry.
Macromolecules 2015, 48, 1711.
(
18) Tao, C.; Aljada, M.; Shaw, P. E.; Lee, K. H.; Cavaye, H.; Balfour,
M. N.; Borthwick, R. J.; James, M.; Burn, P. L.; Gentle, I. R.; Meredith,
P. Controlling Hierarchy in Solution-processed Polymer Solar Cells
Based on Crosslinked P3HT. Adv. Energy Mater. 2013, 3, 105.
(
19) Neubig, A.; Thelakkat, M. Random vs. alternating donor-
acceptor copolymers: A comparative study of absorption and field
effect mobility. Polymer 2014, 55, 2621.
(
20) Karolewski, A.; Neubig, A.; Thelakkat, M.; Kummel, S. Optical
absorption in donor-acceptor polymers–alternating vs. random. Phys.
Chem. Chem. Phys. 2013, 15, 20016.
L
DOI: 10.1021/acs.macromol.5b02659
Macromolecules XXXX, XXX, XXX−XXX