Comparative studies of photochemical cross-linking methods for stabilizing the bulk hetero-junction morphology in polymer solar cells

Article abstract of DOI:10.1039/c2jm34284g

We are here presenting a comparative study between four different types of functionalities for cross-linking. With relatively simple means bromine, azide, vinyl and oxetane could be incorporated into the side chains of the low band-gap polymer TQ1. Cross-linking of the polymers was achieved by UV-light illumination to give solvent resistant films and reduced phase separation and growth of PCBM crystallites in polymer:PCBM films. The stability of solar cells based on the cross-linked polymers was tested under various conditions. This study showed that cross-linking can improve morphological stability but that it has little influence on the photochemical stability which is also decisive for stable device operation under constant illumination conditions. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c2jm34284g

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 24417
www.rsc.org/materials
PAPER
Comparative studies of photochemical cross-linking methods for stabilizing
the bulk hetero-junction morphology in polymer solar cells†
ꢀ
Jon Eggert Carle, Birgitta Andreasen, Thomas Tromholt, Morten V. Madsen, Kion Norrman, Mikkel Jørgensen
and Frederik C. Krebs
*
Received 3rd July 2012, Accepted 14th August 2012
DOI: 10.1039/c2jm34284g
We are here presenting a comparative study between four different types of functionalities for cross-
linking. With relatively simple means bromine, azide, vinyl and oxetane could be incorporated into the
side chains of the low band-gap polymer TQ1. Cross-linking of the polymers was achieved by UV-light
illumination to give solvent resistant films and reduced phase separation and growth of PCBM
crystallites in polymer:PCBM films. The stability of solar cells based on the cross-linked polymers was
tested under various conditions. This study showed that cross-linking can improve morphological
stability but that it has little influence on the photochemical stability which is also decisive for stable
device operation under constant illumination conditions.
groups that can be cleaved off by a thermal treatment of the
Introduction
processed films. The residual carboxylic acid groups then form
hydrogen bonds resulting in a very stiff matrix that also immo-
bilizes the acceptor part.9 Yet another approach that is also
explored in this work is to incorporate cross-linkable groups in
some of the polymer side chains. Several different photo-curable
groups have been used for this purpose such as oxetane groups,10
alkyl-bromide,11,12 azide13–15 and vinyl.16,17 The idea is once again
that the cross-linking immobilizes the structure inhibiting further
growth of domains.
Research on polymer solar cells (PSC) has reportedly delivered an
increase in the power conversion efficiency to some 10% by opti-
mization of each component and especially in the case of the active
layer composed of a light harvesting polymer and a molecular
acceptor.1,2 Another aspect that needs to be addressed is the
stability of these devices, which has also improved by several
orders of magnitude during the last decade.3 Modern PSC rely on
a so-called bulk hetero-junction where the polymer–acceptor
mixture in the active layer is micro-phase segregated to form a bi-
continuous structure with channels for both electron and hole
transport. A key issue is that the excitons formed upon irradiation
with light have a limited diffusion length in these materials of
10–20 nm, which means that this is also the optimal physical
dimension of the domains in the hetero-junction. Unfortunately,
this is not usually the thermodynamic equilibrium (i.e. the mate-
rial is metastable), which is manifested in a growth of PCBM
acceptor crystallites that erodes the optimal morphology.4,5
Several strategies have been developed to mitigate this
problem. One of the first proposed was to combine the donor
polymer with the acceptor part to create block-copolymers that
form stable bi-continuous networks by supra-molecular
forces.6–8 This approach has not been so successful, presumably
due to the formidable synthetic challenges. Another strategy is to
cross-link the active layer after it has been deposited by a cross-
linking reaction. One possibility that has been explored is to use
side chains that are attached to the polymer with tertiary ester
Previous studies have each focused on one specific cross-linking
reaction only. This study compares four different types of func-
tionalities for cross-linking attached to a low band-gap polymer
TQ1.18,19 Furthermore, experiments have been carried out in an
inert atmosphere and with hot dark storage between measure-
ments to enhance the thermally induced morphological insta-
bility. Finally, different experimental conditions aimed for
degradation were compared in order to investigate the importance
of morphological stability compared to photochemical stability.
Experimental
Synthesis
2,5-Bis-(trimethylstannyl)-thiophene,20
3,30-(5,8-dibromoqui-
noxaline-2,3-diyl)diphenol (1) and TQ1 were synthesized
according to the procedures described in the literature.18
5,8-Dibromo-2,3-bis(3-(8-bromooctyloxy)phenyl)quinoxaline (2b).
Compound 1 (1 g, 2.118 mmol), 1,8-dibromooctane (5.76 g, 21.18
mmol) and potassium carbonate (1.464 g, 10.59 mmol) ꢀwere dis-
solved in DMSO (20 ml). The mixture was stirred at 50 C under
argon overnight. Water was added and the organic phase was
Department of Energy Conversion and Storage, Technical University of
Denmark, Frederiksborgvej 399, DK-4000 Roskilde, Denmark. E-mail:
frkr@dtu.dk
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c2jm34284g
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 24417–24423 | 24417

View Online
extracted with ethyl acetate. The organic phase was washed three
times with water and dried over MgSO4. The crude product was
added to a silica column and eluted with heptane–ethyl acetate to
give the product as a solid. Yield 32.1% (580 mg, 0.679 mmol). 1H
NMR (500 MHz, CDCl3) d 7.92 (s, 2H), 7.30–7.20 (m, 4H), 7.16 (m,
2H), 6.99–6.85 (m, 2H), 3.87 (t, J ¼ 6.5 Hz, 4H), 3.42 (t, J ¼ 6.8 Hz,
Polymer TQ-Vinyl. Monomers 2c (50 mg, 0.064 mmol), 2a
(404 mg, 0.579 mmol) and 2,5-bis(trimethylstannyl)thiophene
(264 mg, 0.644 mmol). Yield 370 mg (91%).
Polymer TQ-Oxetane. Monomers 2d (50 mg, 0.058 mmol), 2a
(361 mg, 0.518 mmol) and 2,5-bis(trimethylstannyl)thiophene
(236 mg, 0.576 mmol). Yield 330 mg (90%).
4H), 1.93–1.79 (m, 4H), 1.79–1.64 (m, 4H), 1.50–1.27 (m, 16H). 13
C
NMR (126 MHz, CDCl3) d 159.05, 154.00, 139.32, 139.17, 133.11,
129.32, 123.72, 122.60, 116.54, 115.77, 77.27, 77.02, 76.77, 68.02,
33.95, 32.80, 29.17, 29.08, 28.72, 28.12, 25.94.
Polymer TQ-N3. TQ-Br (300 mg, 0.483 mmol) was dissolved in
toluene (100 ml) at 100 ꢀC and sodium azide (314 mg, 4.83 mmol)
in DMF (100 ml) was added slowly. The mixture was stirred at
100 ꢀC under argon for 48 hours. The solvents were removed
under reduced pressure and the polymer redissolved in chloro-
form and precipitated in methanol. The polymer was purified by
Soxhlet extraction first with methanol then with chloroform and
finally precipitated in methanol. Yield: 290 mg (95%).
5,8-Dibromo-2,3-bis(3-(undec-10-enyloxy)phenyl)quinoxaline (2c).
Prepared as for 2b: compound 1 (500 mg, 1.059 mmol), 11-bro-
moundec-1-ene (617 mg, 2.65 mmol) and potassium carbonate
(1464 mg, 10.59 mmol) were dissolved in DMSO (10 ml). Yield 87%
(715 mg, 0.921 mmol). 1H NMR (500 MHz, CDCl3) d 7.93 (s, 2H),
7.29–7.23 (m, 4H), 7.23–7.17 (m, 2H), 6.96 (m, 2H), 5.84 (dd, J ¼
17.0, 10.3 Hz, 2H), 5.06–4.90 (m, 4H), 3.88 (t, J ¼ 6.6 Hz, 4H), 2.07
(dd, J ¼ 14.5, 6.8 Hz, 4H), 1.83–1.68 (m, 4H), 1.50–1.21 (m, 26H).
13C NMR (126 MHz, CDCl3) d 159.09, 154.03, 139.33, 139.18,
133.06, 129.30, 123.74, 122.56, 116.61, 115.82, 114.13, 68.15, 33.79,
29.54, 29.44, 29.35, 29.13, 28.95, 26.02.
Device fabrication
The polymers TQ1, TQ-Br, TQ-Vinyl, TQ-Oxetane or TQ-N3
and [60]PCBM (Solenne b.v., The Netherlands) were dissolved
separately in chlorobenzene (20 mg mlꢁ1) and stirred overnight
at 50 ꢀC. The polymer and PCBM solutions were mixed and
further stirred at 50 ꢀC and then filtered (1 mm pore size). To the
TQ-Oxetane blend was added 5% (by weight) of the photoacid
generator (bis(4-tert-butylphenyl)iodonium p-toluenesulfonate)
(Sigma-Aldrich). The prefabricated ITO coated glass substrates
were first ultrasonically cleaned in water and then in 2-propanol.
Zinc oxide nanoparticles (ZnO), prepared according to the
literature,21 were spin-coated from water onto the ITO covered
substrate at 1000 rpm and annealed at 140 ꢀC for 10 minutes. The
active layer, composed of the polymer:PCBM solution, was spin-
coated at 700 rpm onto the ZnO layer followed by UV-irradia-
tion at 254 nm with a laboratory lamp (commonly employed for
thin layer chromatography) for 10 minutes in a glove box to
cross-link the polymer. A PEDOT:PSS (Agfa EL-P 5010) solu-
tion was then spin-coated on top at 2800 rpm followed by
5,8-Dibromo-2,3-bis(3-(6-((3-ethyloxetan-3-yl)methoxy)hex-
yloxy)phenyl)quinoxaline (2d). Prepared as for 2b: compound 1
(500 mg, 1059 mmol), 3-((6-bromohexyloxy)methyl)-3-ethyl-
oxetane (739 mg, 2.65 mmol) and potassium carbonate (1464 mg,
10.59 mmol) were dissolved in DMSO (10 ml). Yield 80% (740
1
mg, 0.852 mmol). H NMR (500 MHz, CDCl3) d 7.91 (s, 2H),
7.24 (m, 4H), 7.19–7.11 (m, 2H), 7.00–6.84 (m, 2H), 4.45 (d, J ¼
5.8 Hz, 4H), 4.37 (d, J ¼ 5.8 Hz, 4H), 3.88 (t, J ¼ 6.5 Hz, 4H),
3.53 (s, 4H), 3.47 (t, J ¼ 6.6 Hz, 4H), 1.74 (q, J ¼ 7.4 Hz, 8H),
1.66–1.58 (m, 4H), 1.50–1.35 (m, 8H), 0.89 (t, J ¼ 7.5 Hz, 6H).
13C NMR (126 MHz, CDCl3) d 159.09, 153.98, 139.30, 139.20,
133.08, 129.28, 123.73, 122.61, 116.53, 115.82, 78.57, 73.55,
71.51, 68.02, 43.50, 29.54, 29.12, 26.80, 25.97, 25.89, 8.15.
ꢀ
annealing at 110 C for 2 minutes. The devices were transferred
to a vacuum chamber where silver electrodes were applied by
thermal evaporation at a pressure below 10ꢁ6 mbar. The active
area of the devices was 0.25 cm2.
General procedure for the Stille cross-coupling polymerization
Optical microscopy
Monomer 2a and one of the monomers 2b, 2c or 2d (9 : 1 molar
ratio) and 2,5-bis(trimethylstannyl)thiophene were mixed in
degassed toluene to give a 0.04 M solution. To this was added a
catalyst mix of 2 mol% tris(dibenzylideneacetone)dipalladium(0)
and 8 mol% tri-o-tolylphosphine. The solution was stirred at
100 ꢀC for at least 48 hours to complete the polymerization. The
crude polymer was then precipitated by adding the reaction
mixture to a large volume of methanol. The polymers were
purified by Soxhlet extraction, first with methanol, then with
hexane and finally with chloroform. The chloroform fraction was
then precipitated by pouring it into 10 times the volume of
methanol. The precipitate was filtered off and dried in vacuum to
give the purified polymer.
Blends (1 : 1 by weight) of the polymers and [60]PCBM in
chlorobenzene (20 mg mlꢁ1) were spin-coated on glass slides at
700 rpm. The samples were then treated with UV-irradiation
(254 nm) for 10 minutes in a glove box using a hand held lamp.
The samples were then annealed in ambient air for 13 hours at
150 ꢀC. Optical micrographs of the samples were acquired before
and after the annealing procedure.
Photochemical degradation studies
Photochemical stabilities were evaluated using a fully automated,
high capacity degradation setup with an AM1.5G spectrum in
the ambient atmosphere at 1000 W mꢁ2 described elsewhere.22
Each polymer was spin-coated on glass substrates from a chlo-
robenzene solution. The spin coating parameters were adjusted
in order to obtain a film thickness of around 60 nm.
Polymer TQ-Br. Monomers 2b (100 mg, 0.117 mmol), 2a
(734 mg, 1.053 mmol) and 2,5-bis(trimethylstannyl)thiophene
(480 mg, 1.170 mmol). Yield 620 mg (83%).
24418 | J. Mater. Chem., 2012, 22, 24417–24423
This journal is ª The Royal Society of Chemistry 2012

View Online
Quantification of the degradation rate was based on the evolu-
tion of the gradual decrease of UV-visible absorbance, which was
recorded at 20 min intervals.
TQ-Vinyl and TQ-Oxetane were performed through a Stille
coupling either between pure 2a and 2,5-bis(trimethylstannyl)-
thiophene or using a mixture of 2a and one of the monomers 2b–d.
A monomer feed ratio of 2b–d to 2a was chosen to be 1 : 9 which
should secure several cross-linkable groups per polymer chain.
The synthesis of TQ-N3 was carried out by treating TQ-Br with
sodium azide in hot toluene–DMF, replacing bromine with an
azide group. This transformation is clearly observed by 1H NMR
of the polymers. The polymers were characterized by size exclu-
sion chromatography (SEC) with THF as the eluent, using poly-
styrene as the standard (see Table 1), and also by 1H NMR (see
ESI Fig. S1–5†).
Results and discussion
Synthesis of cross-linkable versions of TQ1
We selected the low band gap polymer TQ1 for creating cross-
linkable versions. The typical octyloxy side chains on the
diphenyl-quinoxaline monomer can easily be substituted with
alkyl groups adorned with azide, bromine, vinyl or oxetane
functionalities that can be used for photocross-linking reactions.
These types of cross-linking functionalities have previously been
investigated with the purpose of stabilizing the morphology in
other types of PSC.10–16
Cross-linking experiments
The syntheses of the monomers and the polymerization of the
five polymers studied in this work are outlined in Scheme 1.
One of the reasons for choosing the TQ1 system for this work
was the relatively straightforward preparation of the functional-
ized monomers. The key was the synthesis of the starting material
quinoxaline 3,30-(5,8-dibromoquinoxaline-2,3-diyl)diphenol (1)
that is common to all the monomers, which is prepared from the
known 5,8-dibromo-2,3-bis(3-methoxyphenyl)quinoxaline18 by
cleaving of the methyl groups with concentrated hydrobromic
acid. The different quinoxaline monomers were then prepared by
alkylation at the phenolate functions with either 1-bromooctane
or a substituted alkyl bromide as shown in Scheme 1. Polymeri-
zation reactions to give either TQ1 or the co-polymers TQ-Br,
The four different photoactive groups used in this study are
expected to cross-link via different chemical reaction mechanisms
when initiated by UV exposure (254 nm). The bromo-alkyl group
is presumably cleaved homolytically to give an alkyl radical and
a bromine atom23 while the alkyl azide group splits off molecular
nitrogen (N2) leaving an alkyl nitrene.24,25 The highly reactive
nitrene can then react with either the polymer or the fullerene in
the bulk through an addition reaction to double bonds.26 The
reactions of the vinyl and the oxetane groups are slightly more
speculative. The oxetane ring undergoes cross-linking initiated
by a photoacid generator (PAG). The propagation probably
involves ring-opening of the oxetane through an attack of
another oxetane group, as shown in Scheme 2.27
Scheme 1 Synthesis of the monomers 2a–d and subsequent polymerization with 2,5-bis-(trimethylstannyl)-thiophene to give the polymers with different
functionalities in the side chains. TQ1, TQ-Br, TQ-N3, TQ-Vinyl and TQ-Oxetane. i: K2CO3, DMSO; ii: Pd(PPh3)4, toluene; and iii: NaN3, toluene,
DMF.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 24417–24423 | 24419

View Online
Table 1 Molecular weight and optical data for the five polymers
lmax (nm)
cross-linked
Eogpt (eV)
cross-linked
Polymer
Mn (kDa)
Mw (kDa)
PD
lmax (nm)
Eogpt (eV)
TQ1
TQ-Br
TQ-N3
TQ-Vinyl
TQ-Oxetane
37.7
20.7
22.7
23.7
22.1
174.8
67.0
92.2
86.8
90.1
4.6
3.23
4.1
3.66
4.1
364/623
364/624
360/621
361/623
361/623
1.7
1.7
1.7
1.7
1.7
364/622
363/618
363/620
364/620
364/622
1.7
1.7
1.7
1.7
1.7
TQ-Oxetane showed partial solvent resistance. PCBM has a
strong absorption in the 254 nm range and therefore absorbs part
of the UV light used for cross-linking. This could be a reason for
the lower solvent resistance.
Absorption spectra of each polymer were acquired as thin
films before and after cross-linking by illumination of the films
with UV light. Incorporation of the functional groups does not
change the absorption spectra of the polymers when compared to
TQ1 (Fig. 1). The spectra of the polymers all have similar
features with a p / p* transition at ca. 360 nm and a charge
transfer transition at ꢂ620 nm (band gap at 1.7 eV). The
absorption spectra of the polymers were essentially unchanged
after photocross-linking, showing the same transitions and band
gaps as before UV-illumination, which suggests that none of the
four different cross-linking reactions damaged the conjugated
polymer backbone.
Scheme 2 A possible mechanism for cross-linking of oxetane through a
ring opening polymerization. Et ¼ ethyl, R ¼ polymer and PAG ¼
photoacid generator.
In all cases reactive intermediates are formed that can react
fast with neighboring groups in the film. These may be on other
polymer strands giving rise to cross-linking of the polymer
matrix and/or with PCBM cross-linking the donor–acceptor
domains.
Stability investigations
Each of the five polymers was spin-coated onto glass
substrates from chloroform solution and dried to give thin films.
The films were then irradiated at 254 nm using a mercury lamp
UV-lamp for 10 minutes to induce photochemical cross-linking,
which was then investigated by a solvent resistance test. The glass
substrates with the polymer films were immersed in hot 1,2-
dichlorobenzene where only the TQ1 film could be dissolved
proving that cross-linking had taken place for the TQ-Br, TQ-
N3, TQ-Vinyl and TQ-Oxetane films. Thin films prepared from
blends of the polymers with PCBM in a 1 : 1 ratio (by weight)
were tested in the same way. The TQ1:PCBM film was fully
soluble while the thin films with TQ-Br, TQ-N3, TQ-Vinyl and
Photochemical stability. The photochemical stability of
conjugated polymers is dependent on several different parame-
ters such as oxygen concentration, temperature, and the molec-
ular structure.28 Incorporation of different functionalities into
the polymer could change its stability. Photochemical stabilities
of the five polymers were therefore evaluated using a fully
automated, high capacity degradation setup with an AM1.5G
spectrum in the ambient atmosphere at 1000 W mꢁ2 at ambient
temperature.22 The gradual decrease of UV-vis absorption of the
ꢂ60 nm polymer thin films was spectroscopically monitored
during ageing. The normalized absorption versus irradiation time
Fig. 1 UV-vis spectra of TQ1, TQ-Br, TQ-N3, TQ-Vinyl and TQ-Oxetane films before (A) and after (B) cross-linking by UV irradiation for 10 minutes
at 254 nm.
24420 | J. Mater. Chem., 2012, 22, 24417–24423
This journal is ª The Royal Society of Chemistry 2012

View Online
Dark thermal degradation in an ambient atmosphere. One of the
tests that has been used to prove the increased stability of devices
with cross-linked active layers is thermal annealing with inter-
mittent testing under illumination. This test brings out the
degradation due to thermally induced morphology changes such
as growth of PCBM domains. Devices with an inverted type
geometry (ITO/ZnO/active layer/PEDOT/Ag) were prepared for
each of the five polymers and subjected to thermal annealing at
100 ꢀC for 50 hours. At intervals the IV curves of the devices were
measured under illumination (AM1.5G, 1000 W mꢁ2). As seen in
Fig. 3 the devices with TQ-N3 and TQ-Oxetane degraded to a
lesser degree and stabilized at a higher power conversion level
than TQ1, TQ-Vinyl and TQ-Br.
The active layer of these devices was further investigated by
optical microscopy and atomic force microscopy (AFM) (the
AFM images are available in the ESI†). Each blend was imaged
before and after annealing at 150 ꢀC for 13 hours (see Fig. 4). As
expected, large PCBM crystallites formed in the TQ1:PCBM film
similar to what has been observed for annealing of
P3HT:PCBM.4,5 Blends containing TQ-Vinyl, TQ-N3 and TQ-
Oxetane showed either none or only very little phase segregation
while the blend containing TQ-Br showed some phase segrega-
tion but not to the extent seen for TQ1:PCBM. This confirms
that the cross-linking has taken place for all the polymers with
incorporated functional groups and that the cross-linking
stabilizes the morphology of the BHJ layer towards thermal
annealing as has been reported earlier.11–16
Fig. 2 Photochemical stability of the five polymers as thin films.
Normalized absorption versus time under constant illumination
(AM1.5G, 1000 W mꢁ2).
Constant illumination in ambient atmosphere versus inert
atmosphere. Dark thermal degradation reduces the effect of
illumination so it is also obvious to investigate device degrada-
tion under constant illumination. The results from a study
carried out in the ambient atmosphere are shown in Fig. 5a where
a similar exponential decay over 17 hours from a maximum to
almost no residual efficiency is observed. The cross-linking does
not seem to infer any added stability in this case indicating that
the dominant degradation mode in this test is photochemical.
This observation is consistent with the outcome of the photo-
chemical degradation experiments performed on the pure poly-
mer films that were also unaffected by the cross-linking.
Fig. 3 Efficiencies of devices containing TQ1, TQ-Br, TQ-N3, TQ-Vinyl
or TQ-Oxetane during thermal annealing at 100 ꢀC in the dark in an
ambient atmosphere (normalized).
for the five polymers is shown in Fig. 2. The degradation rates of
the five polymers are almost identical and total bleaching of the
films is reached after about 70 hours of illumination. This
suggests that incorporation of the functionalities does not affect
the photochemical stability of these polymers.
When this study was repeated under inert atmosphere
(Fig. 5b), however, the overall rate of degradation was retarded
and some differences between the polymers became apparent. A
fast initial decay in the performance was once again observed,
Fig. 4 Optical micrographs (195 ꢃ 260 mm2) of spin-coated thin films of the five polymers blended with PCBM (1 : 1 wt) before (0 hours) and after
annealing at 150 ꢀC for 13 hours. All the thin films have been irradiated with UV light at 254 nm for 10 minutes before recording images. The dark areas
correspond to PCBM rich domains.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 24417–24423 | 24421

View Online
Fig. 5 Deterioration of the power conversion efficiencies of the devices under constant illumination in (A) ambient atmosphere, and in (B) inert
atmosphere (normalized).
which could be due to the fact that the devices were fabricated
Acknowledgements
under ambient conditions, which means that they contained a
This work has been supported by the Danish Strategic Research
residual amount of oxygen–water. After about 5 hours the decay
Council (2104-07-0022), EUDP (j.no. 64009-0050 and 64011-
rate slowed and clear differences in the performance between the
0002), the Danish National Research Foundation, and from
different polymers were observed in the order: TQ-Br > TQ-N3 >
PVERA-NET (project acronym POLYSTAR).
TQ1 > TQ-Vinyl.
Notes and references
Conclusions
ꢀ
1 T. Chu, J. Lu, S. Beaupre, Y. Zhang, J. Pouliot, S. Wakim, J. Zhou,
M. Leclerc, Z. Li, J. Ding and Y. Tao, J. Am. Chem. Soc., 2011, 133, 4250.
2 H. Chen, J. Hou, S. Zhang, Y. Liang, G. Yang, Y. Yang, L. Yu,
Y. Wu and G. Li, Nat. Photonics, 2009, 3, 649.
3 M. Jørgensen, K. Norrman, S. A. Gevorgyan, T. Tromholt,
B. Andreasen and F. C. Krebs, Adv. Mater., 2012, 24, 580.
4 S. Bertho, G. Janssen, T. J. Cleij, B. Conings, W. Moons, A. Gadisa,
J. D’Haen, E. Goovaerts, L. Lutsen, J. Manca and D. Vanderzande,
Sol. Energy Mater. Sol. Cells, 2008, 92, 753.
5 M. Drees, H. Hoppe, C. Winder, H. Neugebauer, N. S. Sariciftci,
€
W. Schwinger, F. Schaffler, C. Topf, M. C. Scharber, Z. Zhu and
R. Gaudiana, J. Mater. Chem., 2005, 15, 5158.
6 P. D. Topham, A. J. Parnell and R. C. Hiorns, J. Polym. Sci., Part B:
Polym. Phys., 2011, 49, 1131.
This study has shown that different types of cross-linking
moieties can be incorporated into the side chains of the low band
gap TQ1 type polymer by relatively simple means. The cross-
linking reaction could be achieved by UV-irradiation of the pure
polymer films to give insoluble products. When PCBM was
included the cross-linking was less efficient presumably due to
the optical absorption band of PCBM. Some solvent resistance
was however observed in this case indicating some degree of
cross-linking.
The UV-vis spectra of the polymer films were not affected
proving that the cross-linking reaction did not damage the
conjugated backbone of the polymers even though widely
different reaction types are expected for the four different types
of side chain functionalities: bromide, azide, vinyl and oxetane.
Cross-linking was shown to inhibit excessive phase separation
and growth of PCBM crystallites in polymer:PCBM films during
dark thermal annealing as shown by optical microscopy. This
resulted in improved solar cell device stability under the condi-
tions in question. It did, however, not improve the device
stability under constant illumination in an ambient atmosphere,
which is probably dominated by photochemical degradation
rather than by thermal mechanisms. When oxygen–water was
excluded by employing an inert atmosphere the stability
increased somewhat and more importantly, some differences in
stability became apparent between the polymers with TQ-Br and
TQ-N3 giving the most stable devices. At present no explanation
is provided for this observed difference, but it could be ascribed
to different cross-linking mechanisms and also to different
reaction rates.
7 G. Ren, P.-T. Wu and S. A. Jenekhe, Chem. Mater., 2010, 22, 2020.
8 E. J. W. Crossland, M. Kamperman, M. Nedelcu, C. Ducati,
U. Wiesner, D. M. Smilgies, G. E. S. Toombes, M. A. Hillmyer,
S. Ludwigs, U. Steiner and H. J. Snaith, Nano Lett., 2009, 9, 2807.
9 J. Liu, E. N. Kadnikova, Y. Liu, M. D. McGehee and
ꢀ
J. M. J. Frechet, J. Am. Chem. Soc., 2004, 126, 9486.
ꢀ
10 J. Farinhas, Q. Ferriera, R. E. Di Paolo, L. Alcacer, J. Morgado and
A. Charas, J. Mater. Chem., 2011, 21, 12511.
11 B. J. Kim, Y. Miyamoto, B. Ma and J. M. J. Frechet, Adv. Funct.
ꢀ
Mater., 2009, 19, 2273.
12 G. Griffini, J. D. Douglas, C. Piliego, T. W. Holcombe, S. Turri,
ꢀ
J. M. J. Frechet and J. L. Mynar, Adv. Mater., 2011, 23, 1660.
13 B. Gholamkhass and S. Holdcroft, Chem. Mater., 2010, 22, 5371.
14 H. J. Kim, A. R. Han, C. H. Cho, H. Kang, H. H. Cho, M. Y. Lee,
ꢀ
J. M. J. Frechet, J. H. Oh and B. J. Kim, Chem. Mater., 2012, 24, 215.
15 C. Y. Nam, Y. Qin, Y. S. Park, H. Hlaing, X. Lu, B. M. Ocko,
C. T. Black and R. B. Grubbs, Macromolecules, 2012, 45, 2338.
16 C. H. Hsieh, Y. J. Cheng, P. J. Li, C. H. Chen, M. Dubosc,
R. M. Liang and C. S. Hsu, J. Am. Chem. Soc., 2010, 132, 4887.
17 U. R. Lee, T. W. Lee, M. H. Hoang, N. S. Kang, J. W. Yu,
K. H. Kim, K. G. Lim, T. W. Lee, J. I. Jin and D. H. Choi, Org.
Electron., 2011, 12, 269.
18 T. Yamamoto, B. L. Lee, H. Kokubo, H. Kishida, K. Hirota,
T. Wakabayashi and H. Okamoto, Macromol. Rapid Commun.,
2003, 24, 440.
This study shows that cross-linking can improve morpholog-
ical stability, but that other factors such as photochemical
degradation might be more important for device stability under
constant illumination conditions.
€
€
19 E. Wang, L. Hou, Z. Wang, S. Hellstrom, F. Zhang, O. Inganas and
M. R. Andersson, Adv. Mater., 2010, 22, 5240.
20 M. Helgesen, S. A. Gevorgyan, F. C. Krebs and R. A. J. Janssen,
Chem. Mater., 2009, 21, 4669.
24422 | J. Mater. Chem., 2012, 22, 24417–24423
This journal is ª The Royal Society of Chemistry 2012

View Online
21 W. J. E. Beek, M. M. Wienk, M. Kemerink, X. Yang and
R. A. J. Janssen, J. Phys. Chem. B, 2005, 109, 9505.
25 K. A. Murray, A. B. Holmes, S. C. Moratti and G. Rumbles, J.
Mater. Chem., 1999, 9, 2109.
ꢀ
22 T. Tromholt, M. V. Madsen, J. E. Carle, M. Helgesen and
26 N. Yasuda, S. Yamamoto, Y. Wada and S. Yanagida, J. Polym. Sci.,
Part A: Polym. Chem., 2001, 39, 4196.
27 S. Feser and K. Meerholz, Chem. Mater., 2011, 23, 5001.
F. C. Krebs, J. Mater. Chem., 2012, 22, 7592.
23 K. S. Lee, K. Y. Yeon, K. H. Jung and S. K. Kim, J. Phys. Chem. A,
2008, 112, 9312.
ꢀ
28 M. Manceau, E. Bundgaard, J. E. Carle, O. Hagemann, M. Helgesen,
ꢀ
24 R. F. Klima and A. D. Gudmundsdottir, J. Photochem. Photobiol., A,
2004, 162, 239.
R. Søndergaard, M. Jørgensen and F. C. Krebs, J. Mater. Chem.,
2011, 21, 4132.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 24417–24423 | 24423