Photovoltaic performance of polymers based on dithienylthienopyrazines bearing thermocleavable benzoate esters

Article abstract of DOI:10.1021/ma9024812

Thermocleavable low-band-gap polymers based on dithienylthienopyrazines were prepared and copolymerized with different donor units like dialkoxybenzene, fluorene, thiophene, and cyclopentadithiophene (CPDT) using both Stille and Suzuki cross-coupling reac

Full text of DOI:10.1021/ma9024812

Macromolecules 2010, 43, 1253–1260 1253
DOI: 10.1021/ma9024812
Photovoltaic Performance of Polymers Based on Dithienylthienopyrazines
Bearing Thermocleavable Benzoate Esters
Martin Helgesen* and Frederik C. Krebs
Riso National Laboratory for Sustainable Energy, Technical University of Denmark, Frederiksborgvej 399,
e
DK-4000 Roskilde, Denmark
Received November 10, 2009; Revised Manuscript Received December 17, 2009
ABSTRACT: Thermocleavable low-band-gap polymers based on dithienylthienopyrazines were prepared
and copolymerized with different donor units like dialkoxybenzene, fluorene, thiophene, and cyclopenta-
dithiophene (CPDT) using both Stille and Suzuki cross-coupling reactions. In the solid state the band gaps
are in the range of 1.17-1.37 eV. The polymers were explored as donor materials in bulk heterojunction solar
cells together with PCBM as the acceptor material where they were shown to exhibit a photoresponse in the
full absorption range up to 900 nm and power conversion efficiencies of up to 1.21% under 1 sun irradiation.
A red shift of the absorption edge on going from solution to the solid film was observed for all the polymers.
Thermogravimetric analysis of the polymers in the temperature range from 25 to 500 °C showed a weight loss
at just above 200 °C, corresponding to loss of the tertiary ester groups, and a second weight loss above 400 °C,
corresponding to loss of CO2 and decomposition. Upon thermocleavage the power conversion efficiency
decreased for all the polymers while the polymer films became insoluble which was desired in the context of
multilayer film processing. Thermocleavable low-band-gap materials can potentially offer better light
harvesting, better operational stability, and a higher level of permissible processing conditions due to the
insolubility of thermocleaved films in all solvents.
Introduction
of why devices and materials break down.10 By using time-of-
flight secondary ion mass spectrometry (TOF-SIMS)11 and
Low-band-gap polymers for photovoltaics are designed to
match the solar emission spectrum better, which has a maximum
in photon flux near 700 nm and an appreciable tail stretching into
the infrared region.1,2 The extended absorption by low-band-gap
polymers can potentially increase the power conversion efficiency
by absorbing more photons. One approach to designing these
materials is by use of alternating electron-rich (donor) and
electron-poor (acceptor) units giving rise to a material with a
low-energy absorption band that is generally ascribed to a charge
transfer band. The absorption can be tuned by adjusting the
donor-acceptor strengths, or HOMO-LUMO levels, respec-
tively. For this purpose, polymers with alternating dithiophene
and thienopyrazine units have been explored by several groups3-9
who reports band gaps in the range 1.2-1.6 eV for this type of
polymer. In our earlier work,3 we explored the chemistry of the
thienopyrazine-type acceptor moiety to characterize the influence
of the substituents and extended π-system on the absorption
spectrum. Here we found that adding phenyl groups to the
dithienylthienopyrazine system caused a red shift of the lowest
energy absorption band with up to 50 nm, presumably due to the
more extended conjugation. In addition, polymers based on fused
aromatic thienopyrazine units can reduce the band gap even
further caused by a more planar backbone between repeating
units.3,5 Low-band-gap polymers based on dithienylthienopyra-
zine in blends with soluble methanofullerenes have shown high
power conversion efficiencies (2.2%)9 and an extended photo-
response up to 900 nm,7 indicating that these materials are
promising for photovoltaic applications.
isotopic labeling (18O2 and H218O), the main finding is that
oxygen and water diffuse into the various layers of the solar cell,
react with the bulk of the materials and the interfaces, and thus
degrade the solar cell and device performance.12-18 Moreover,
photodegradation studies of both MDMO-PPV and P3HT under
illumination in the presence and absence of oxygen19-22 have
shown that widely different mechanisms are in play. Illumination
of MDMO-PPV in the absence of oxygen suggests that absorp-
tion of UV-vis light by MDMO-PPV can induce the homolytic
scission of the O-CH2 bond. The generated radicals may react
with the vinylene groups, which lead to loss of conjugation, or
undergo photo-Fries rearrangement. Furthermore, different
photochemical mechanisms have been shown to be in play, the
photochemical instability of P3HT has been suggested to be
mainly due to the hexyl side chains, and it has been predicted that
the photochemical stability of native polythiophene should be
significantly longer. Taking the above-mentioned issues into
consideration, one could explore the many possibilities in em-
ploying a conjugated material that is reached either through a
precursor route or through a route where side chains are removed
post film formation. This can be realized with the use of thermo-
cleavable side chains. The side chains provide solubility in organic
solvents and allow film formation via solution processing. Sub-
sequently, they can be removed by heating in a postprocessing
step forming a harder insoluble material where diffusion phe-
nomena are slowed down and in addition the photochemical
reactions associated with the side chains are avoided. Ideally, the
thermocleavage of the side chains leads to a high-Tg material,
characterized by its high glass transition temperature, which has
been demonstrated to strongly suppress morphological changes
in high-Tg PPV:PCBM active layers that leads to high thermal
stability of the photovoltaic characteristics.23 Because of a high
There has been a recent interest in the operational stability of
polymer solar cells and more importantly on the understanding
*Corresponding author. E-mail: manp@risoe.dtu.dk.
r 2010 American Chemical Society
Published on Web 01/07/2010
pubs.acs.org/Macromolecules

1254 Macromolecules, Vol. 43, No. 3, 2010
Helgesen and Krebs
Figure 1. Low-band-gap polymers based on dithienylthienopyrazine alternating with different donor segments.
glass transition temperature, the active layer forms a more rigid
and stable morphology which limit the possible migration and
segregation of the PCBM molecules leading to a more stable
active layer and consequently to a more stable photovoltaic
behavior.23 Alternative routes to polymer materials for polymer
solar cells without side chains are precursor routes such as the
dithiocarbamate route.24-28 Lifetimes over 10000 h have been
reported for solar cells based on the thermocleavable polymer
poly-3-(2-methylhexan-2-yl)oxycarbonylbithiophene
Experimental Section
Synthetic procedures for synthesis of monomers and polymers
according to Schemes 1 and 2 and their characterization data
1
(including H NMR and 13C NMR) are described in detail in
the Supporting Information together with general experimental
details.
Polymer Solar Cell Fabrication and Analysis. Photovoltaic
devices were made by spin-coating PEDOT:PSS (Aldrich,
1.3 wt % aqueous solution) onto precleaned, patterned indium
tin oxide (ITO) substrates (9-15 Ω per square) (LumTec)
followed by annealing at 140 °C for 5 min. The active layer
was deposited, in a glovebox, by spin-coating a blend of
the polymer and [60]PCBM dissolved in o-dichlorobenzene
(40 mg/mL). After a thermal treatment (see Table 2) the counter
electrode of aluminum was deposited by vacuum evaporation at
(2-3) ꢀ 10-6 mbar. The active area of the cells was 0.5 cm2. I-V
characteristics were measured under AM1.5G corresponding to
74.3 mW/cm2 white light from a multiwavelength high-power
LED array using a Keithley 2400 source meter. IPCE spectra
were recorded on the same solar test platform with the LED-
based illumination system.
(P3MHOCT) and C60 after thermal elimination of the solubiliz-
ing groups which transforms P3MHOCT into the more rigid and
insoluble poly-3-carboxydithiophene (P3CT).17,29
In spite of the more complex synthetic chemistry and materials
handling requirements for thermocleavable materials, the moti-
vations for exploring those in the context of polymer solar cells
include improvement of morphological, interface, and photo-
chemical stability, improvement of the chromophore density in
the device film, and significant advantages in terms of processing
(solubility/insolubility switching). Thermocleavable materials
remain inferior to the current state of the art in terms of power
conversion efficiency while recent progress have shown power
conversion efficiencies approaching 2%. It is likely that thermo-
cleavable materials can be improved at least to the level of the
current state-of-the-art pending the same investment in optimiza-
tion as materials such as P3HT has received.
Herein we report our efforts in this direction through the
synthesis of a series of alternating thermocleavable low-band-
gap polymers and their photovoltaic performance in blends with
[6,6]-phenyl C61 butyric acid methyl ester (PCBM). The materials
are copolymers based on dithienylthienopyrazine, bearing thermo-
cleavable benzoate esters on the pyrazine ring, alternating with
different donor segments, i.e., dialkoxybenzene, fluorene, thio-
phene, and cyclopentadithiophene (CPDT) (Figure 1). The effects
of the different donor segments on the photovoltaic performance of
the polymers in blends with [60]PCBM with and without thermal
treatment are presented. The alkyl benzoate ester groups make the
polymer soluble in organic solvents and allow for film formation.
Subsequently, they can be removed by heating in a postprocessing
step forming the free acid and a volatile alkene.
Results and Discussion
Synthesis. The synthetic steps involved in the preparation
of the monomers 2a, 2b, 5, 7, and 10 are outlined in Scheme 1.
Monomer 1 was functionalized by NBS bromination and
by deprotonation using lithium diisopropylamine (LDA)
followed by treatment with trimethyltin chloride. This af-
forded 2a30 and 2b31 to be used in Suzuki- and Stille-type
copolymerizations. According to a literature procedure,32
monomer 5 can be synthesized in good yield starting with a
standard alkylation of hydroquinone (3) followed by bromi-
nation of 4 with NBS. The diboronic acid pinacol ester 7 is
prepared by lithiation of readily available 2,7-dibromo-9,9-
dioctylfluorene (6) followed by addition of 2-isopropoxy-
4,4,5,5-tetramethyl-1,3,2-dioxaborolane.33 The synthetic
route to 1034 initiates with a deprotonation of 4H-cyclopenta-
[2,1-b:3,4-b0]-dithiophene (8) and a subsequent alkyla-
tion which affords 9 in good yield. Deprotonation of 9 using

Article
Macromolecules, Vol. 43, No. 3, 2010 1255
Scheme 1. Synthetic Steps Involved in the Preparation of the Monomers
Scheme 2. Copolymerizations Leading to the Polymers P1-P4a
a Y = 5, 7, 10, and 2,5-bis(trimethylstannyl)thiophene. (i) Stille coupling using Pd2dba3 and tri-o-tolylphosphine. (ii) Suzuki coupling using Pd2dba3,
tri-o-tolylphosphine and Cs2CO3.
n-butyllithium followed by treatment with trimethyltin
chloride affords monomer 10.
Thermal Behavior. The thermal behavior of the thermo-
cleavable polymers was investigated by thermogravimetric
analysis (TGA). The sample holders were carefully weighed
and the samples introduced. TGA was then carried out using
heating rate of 10 °C min-1. TGA of P1-P4 are shown in
Figure 2 and indicates that the tertiary ester starts to
eliminate around 200 °C, in agreement with earlier results.30
The second loss peak at ∼400 °C that corresponds to loss of
CO230 (not prior to decomposition) can only be observed for
P3 because a greater weight loss for P1, P2, and P4 is
showing in the same temperature range. The observed value
for this loss peak is ∼20%, which corresponds to loss of the
alkyl chains on the donor units: dialkoxybenzene, fluorene,
and CPDT. The same precursor film prepared by standard
solution processing of P1-P4 can give access to two chemi-
cally different thin films, as shown in Figure 3.
Copolymerizations leading to the final polymers P1-P4
are presented in Scheme 2. Copolymerization of 2b via
Stille coupling, using the catalyst system Pd2dba3/tri-o-
tolylphosphine, with 5 gave polymer P1 in 77% yield as a
dark brown solid (Mw = 7 kg/mol, PDI = 1.9). Coupling of
2a with 7 was performed with a Suzuki-type copolymeriza-
tion reaction using Pd2dba3/tri-o-tolylphosphine as a cata-
lyst and caesium carbonate as a base. The polymer P2 was
afforded in 90% yield as a green solid with a molecular
weight (Mw) of 42.3 kg/mol and a polydispersity (PDI) of 3.
Using the same conditions as for the preparation of P1,
copolymerization of 2a via Stille coupling with 2,5-bis-
(trimethylstannyl)thiophene and the cyclopentadithiophene
10 gave polymer P3 and P4 as dark green solids in 92-93%
yield. All the polymers were isolated in good yields and are
soluble in organic solvents such as chloroform and toluene at
room temperature.
Optical Properties. The absorption spectra for the poly-
mers in chloroform solution are shown in Figure 4a. The
copolymers P1-P4 based on dithienylthienopyrazine does

1256 Macromolecules, Vol. 43, No. 3, 2010
Helgesen and Krebs
Figure 2. (a) TGA of P1, (b) TGA of P2, (c) TGA of P3, and (d) TGA of P4 in the temperature range 50-500 °C. The data were recorded at
10 °C min-1 under an argon atmosphere. A derivative weight loss curve has been included to tell the point at which weight loss is most apparent.
Figure 3. Possible chemical transitions of P1-P4.
indeed show a considerable spectral coverage of the solar
spectrum which is varied with the different donor units. The
optical band gaps, defined by the onset of absorption, are
ranging from 1.22 to 1.50 eV (Table 1), which is much smaller
than that of poly(3-hexylthiophene) (P3HT) homopolymer
(Eg ∼ 1.9-2.1 eV). This supports the idea that the internal
charge-transfer interaction between donor and acceptor
moieties in donor-acceptor copolymers is an efficient
method to lower the band gap of conjugated polymers. Partial
aggregation of P1 in solution gives it the lowest optical band
gap with an onset at 1015 nm. P2 has a somewhat higher band
gap of 1.5 eV because of the decreased donor strength of the
fluorene unit (high degree of aromaticity) which reduce con-
jugation in the polymer backbone.
Three alternating thiophene units provide P3 with a band
gap of 1.3 eV in solution. Further extending the thiophene
content by incorporating CPDT lowers the band gap to
1.27 eV (P4). Despite the improved donor character of the
CPDT unit, caused by its planarity and electron-donating
alkyl chains, P3 and P4 have rather similar band gaps though
the absorption maxima (λmax) of P3 is blue-shifted compared
to P4. The thin film absorption spectra for polymers P1-P4
are shown in Figure 4b. The optical band gaps are ranging
from 1.17 to 1.37 eV where only P2 shows a significant
decrease compared to in solution (Table 1). The polymers P1
and P2 have absorption maxima in the range from 665 to
745 nm in chloroform solution, and these are red-shifted
further to 710-845 nm when in a solid film (Table 1),
indicating significant interchain association in the solid state.
In addition, λmax for P2 is red-shifted with 50-95 nm
compared to corresponding polymers without the thermo-
cleavable side chains.4,9 P3 reveals a shoulder around 800 nm
in solution, and the same, but weaker, vibronic fine structure
remains in the solid state. P4 also reveals a shoulder in
solution around 830 nm, but in the solid state the absorption
band has broadened, caused by intermolecular interactions,
and the vibronic fine structure has disappeared. Upon
annealing the films only P3 and P4 shows a significant
change in the absorption spectra (Figure 5). Upon thermo-
cleavage of the films by heating them at 250 °C for 1 min a
color change from olive green to a more brownish color is
observed. The associated changes in the absorption spectrum
are a less intense low-energy absorption band and a smaller
band gap which is reduced to 1.23 eV for P3 and 1.18 eV for
P4. There may be several explanations for the lower absorp-
tion intensity. First, the associated change in film thickness,

Article
Macromolecules, Vol. 43, No. 3, 2010 1257
and secondly, the dielectric constant may lead to changes in
the reflection phenomena that also contribute to the inten-
sities in the observed absorption spectrum for a solid film in
transmission geometry. Thirdly, the intensity of absorption
quite often decreases as the band gap is lowered. After the
short thermal treatment the films maintained the optical
quality and were insoluble in organic solvents.
Photovoltaic Performance. Bulk heterojunction solar cells
with an active area of 0.5 cm2 were prepared on an indium tin
oxide (ITO) covered glass substrate, using conventional
device architecture. A thin layer of poly(3,4-ethylenedi-
oxythiophene)-poly(styrenesulfonate) (PEDOT-PSS) was
spin-coated on top of the ITO coating followed by spin-
coating of the active layer. The active layer contained a blend
of the respective polymer and [60]PCBM.
After spin-coating of the active layer the devices were
either processed directly into a solar cell by evaporation of
aluminum as back electrode or subjected to a thermal
treatment at the temperature of thermocleavage immediately
before evaporation of the back electrode. The obtained
current-voltage curves are presented in Figure 6 which
shows the current-voltage characteristics of the polymer:
PCBM solar cells measured under 74.3 mW/cm2 white light.
The unannealed devices based on P1, with the lowest band
gap (1.15 eV), and PCBM had low open-circuit voltages
(Voc) of 0.36 V, moderate fill factors (FF) of 0.40, and current
densities (Jsc) of 1.82 mA/cm2. This resulted in power con-
version efficiencies of up to 0.35% (Table 2). Devices based
on the fluorine-coupled polymer P2 and PCBM showed a
somewhat higher Voc up to 0.65 V (Figure 6a) as expected
from earlier reports9 with a similar system. P2 provides a
descent FF of 0.44, but the low current density (1.41 mA/
cm2) limits the performance to 0.54%. Changing the polymer
backbone to be a complete thiophene segment raises the
Jsc up to 2.22 mA/cm2 for P3:PCBM devices. The Voc was
0.5 V, and together with a FF of 0.38 the devices had a power
conversion efficiency up to 0.57%. Solar cells based on P4:
PCBM exhibits the best performance with the highest cur-
rent density of 3.20 mA/cm2 and a good fill factor of 0.51.
Together with an open-circuit voltage of 0.55 V, the power
conversion efficiency sets to 1.21%. The somewhat higher
Jsc obtained with P4 is also reflected in the incident photon to
current efficiency (IPCE) which reaches an average IPCE of
17% with a photoresponse up to 900 nm (Figure 7a). In
contrast, P1-P3 have an average IPCE in the range 7-8%
but also extends up to 900 nm, except P2 in agreement with
the absorption spectra (Figure 4b). Despite the extended
photoresponse of P1-P4, IPCE is inferior compared to
the state of the art system P3HT:PCBM (Table 2) which
may be explained by limited exciton dissociation at the
Figure 4. (a) UV-vis absorption spectra of the polymers P1-P4 in
chloroform solution and (b) in thin film. P3HT in chloroform solution
and in thin film is also shown for comparison.
Figure 5. (a) UV-vis absorption spectra of P3 and (b) P4 in thin film
before and after annealing for 1 min.
Table 1. GPC and Spectroscopic Data for Polymers P1-P4
solution
film
polymer
Mw (g/mol)
PDI
R(λmax) (L/(g cm))
λmax (nm)
λonset (nm)
Eg (eV)
λmax (nm)
λonset (nm)
Eg (eV)
P1
P2
P3
P4
7 000
42 300
39 400
1.9
3.0
1.9
4.8
16
18
23
29
745
665
770
868
1015
824
955
980
1.22
1.50
1.30
1.27
845
710
760
825
1057
906
955
1.17
1.37
1.30
1.24
363 000
1002

1258 Macromolecules, Vol. 43, No. 3, 2010
Helgesen and Krebs
polymer-PCBM interface. However for P4, the generated
charge carriers seem to be extracted relatively efficiently as is
indicated by a rather good fill factor of 0.51. J-V curves of
the polymer:PCBM solar cells after a thermal treatment are
shown in Figure 6b, and a general observation is that the
performance drops after the thermocleavage. Table 2 shows
a large drop in the current density for all polymers after
thermocleavage together with minor drops in the Voc and
FF. The drop in performance is also reflected in the IPCE
which is lower at all wavelengths compared to the un-
annealed devices (Figure 7b).
Morphology. The P3:PCBM and P4:PCBM device films
annealed at different temperatures, as measured by atomic
force microscopy (AFM), are shown in Figure 8. AFM
reveals changes in the surface topography of the films and
generally gives a good first insight into morphology of the
active layer.35 All films shows a significant roughness with a
peak-to-valley difference around 15-20 nm. Comparing the
films before and after thermocleavage (∼200 °C) reveals that
the domain sizes increases to features with dimensions larger
than 100 nm which indicate extensive phase segregation of
the polymer and PCBM upon annealing at high tempera-
tures. Moreover, Figure 8c,g indicates that phase segregation
commence prior to thermal cleavage of the tertiary esters.
The reduced current densities of the polymer:PCBM devices
after thermocleavage might be a direct consequence of the
changed morphology which is possibly limiting charge car-
rier generation (reduced number of exitons reach the inter-
face) and transport to the electrodes (insufficient percolating
pathways). The drop in the current density after thermo-
cleavage to the free carboxylic acid has been observed before
for polymers where the thermocleavable ester resides on a
thiophene unit. These polymers can undergo further trans-
formation into the native system by decarboxylation which
leads to a significant improvement in performance due to an
increase in mainly the current density.36,37 For the materials
reported here heating to 300 °C resulted in significantly
poorer performance. As measured by AFM, one possible
explanation is that the morphology changes undesirably for
this class of materials at the high temperatures, and further
work on understanding the complex interplay between the
Figure 6. (a) J-V characteristics of the P1:PCBM, P2:PCBM, P3:
PCBM, and P4:PCBM solar cells measured under 74.3 mW/cm2 white
light before and (b) after a thermal treatment (see Table 2).
Figure 7. (a) IPCE of polymer:PCBM solar cells before and (b) after a
thermal treatment.
Table 2. Photovoltaic Performance of Devices Based on Blends of Polymer and PCBM
polymer
polymer:PCBM (w/w ratio)
thermal treatment (°C)
Voc (V)
Jsc (mA/cm2)
FF
η (%)
P1
P1
P2
P2
P3
P3
P4
P4
1:2
1:2
1:3
1:3
1:4
1:4
1:3
1:3
1:1
0.36
0.36
0.65
0.60
0.50
0.44
0.55
0.50
0.62
1.82
1.16
1.41
1.18
2.22
1.66
3.20
2.13
7.69
0.40
0.47
0.44
0.35
0.38
0.36
0.51
0.45
0.48
0.35
0.27
0.54
0.33
0.57
0.35
1.21
0.64
2.3c
250a
250a
230a
225a
150b
P3HT
a Annealed for 30 s. b Annealed for 5 min. c Typical PCE reached at RisoDTU with commercially available regioregular P3HT in the same device
e
geometry.

Article
Macromolecules, Vol. 43, No. 3, 2010 1259
Figure 8. AFM topography images(2 μm ꢀ 2 μm) ofsolar cellsbased on blendsofPCBM and (a) P3 unannealed, heightscale is 20 nm; (b) P3 annealed
at 150 °C, height scale is 15 nm; (c) P3 annealed at 180 °C, height scale is 15 nm; (d) P3 annealed at 230 °C, height scale is 15 nm; (e) P4 unannealed,
height scale is 15 nm; (f) P4 annealed at 150 °C, height scale is 15 nm; (g) P4 annealed at 180 °C, height scale is 15 nm; (h) P4 annealed at 225 °C, height
scale is 15 nm.
changes in morphology as a result of thermocleavage is
warranted. This points to the importance of the difference
between the temperature where changes in morphology take
place and the temperature at which thermocleavage takes
place. It is likely that the few examples where similar or better
performance was obtained after thermocleavage of the film
represent cases where the morphology does not change
before thermocleavage.
Acknowledgment. This work was supported by the Danish
Strategic Research Council (DSF 2104-05-0052 and 2104-07-
0022).
Supporting Information Available: General procedures and
characterization data including NMR spectra; experimental
procedures for the synthesis of the monomers and polymers
according to Schemes 1 and 2. This material is available free of
charge via the Internet at http://pubs.acs.org.
Conclusion
References and Notes
A series of new thermocleavable low-band-gap polymers based
on dithienylthienopyrazine, bearing thermocleavable benzoate
esters on the pyrazine ring, alternating with different donor
segments (including dialkoxybenzene, fluorene, thiophene, and
CPDT) have been synthesized. The solubilizing benzoate ester
groups are thermocleavable around 200 °C where a volatile
alkene is eliminated, leaving the polymer component more rigid.
Furthermore, it was found that no decarboxylation takes place
prior to decomposition at ∼400 °C where a greater weight loss for
P1, P2, and P4 is observed in the same temperature range which
corresponds to loss of the alkyl chains on the donor units:
dialkoxybenzene, fluorene, and CPDT. The four polymers
optical properties and photovoltaic performance in blends
with PCBM have been investigated. In chloroform solution the
polymers had optical band gaps ranging from 1.22 to 1.50 eV.
The optical band gaps are lowered to 1.17-1.37 eV in thin film,
showing a considerable spectral coverage of the solar emission
spectrum. Furthermore, polymers P3 and P4 showed a less
intense low-energy absorption band and a smaller band gap after
annealing the film for 1 min. The best performing polymer in a
bulk heterojunction solar cell was P4 with Jsc = 3.20 mA/cm2,
Voc = 0.55 V, FF = 0.51, and η = 1.21%. Devices generally
performed worse after thermocleavage due to a drop in mainly
the current density giving power conversion efficiencies up to
0.64% for P4:PCBM solar cells. The drop in performance after
thermocleavage can be linked to extensive phase segregation of
the polymer and PCBM upon annealing as measured by AFM.
We finally conclude that the interplay between temperature,
morphology, and film chemistry needs to be understood before
efficient thermocleavable materials can be optimally designed.
(1) Bundgaard, E.; Krebs, F. C. Sol. Energy Mater. Sol. Cells 2007, 91,
954–985.
(2) Kroon, R.; Lenes, M.; Hummelen, J. C.; Blom, P. W. M.; de Boer,
B. Polym. Rev. 2008, 48, 531–582.
(3) Petersen, M. H.; Hagemann, O.; Nielsen, K. T.; Jorgensen, M.;
e
Krebs, F. C. Sol. Energy Mater. Sol. Cells 2007, 91, 996–1009.
(4) Zhang, F. L.; Perzon, E.; Wang, X. J.; Mammo, W.; Andersson,
M. R.; Inganas, O. Adv. Funct. Mater. 2005, 15, 745–750.
(5) Mondal, R.; Miyaki, N.; Becerril, H. A.; Norton, J. E.; Parmer, J.;
ꢀ
Mayer, A. C.; Tang, M. L.; Bredas, J.-L.; McGehee, M. D.; Bao, Z.
Chem. Mater. 2009, 21, 3618–3628.
(6) Campos, L. M.; Tontcheva, A.; Gunes, S.; Sonmez, G.; Neugebauer,
H.; Sariciftci, N. S.; Wudl, F. Chem. Mater. 2005, 17, 4031–4033.
(7) Zoombelt, A. P.; Gilot, J.; Wienk, M. M.; Janssen, R. A. J. Chem.
Mater. 2009, 21, 1663–1669.
(8) Wienk, M. M.; Turbiez, M. G. R.; Struijk, M. P.; Fonrodona, M.;
Janssen, R. A. J. Appl. Phys. Lett. 2006, 88, 153511.
(9) Zhang, F. L.; Mammo, W.; Andersson, L. M.; Admassie, S.;
Andersson, M. R.; Inganas, L.; Admassie, S.; Andersson, M. R.;
Ingands, O. Adv. Mater. 2006, 18, 2169–2173.
(10) Jorgensen, M.; Norrman, K.; Krebs, F. C. Sol. Energy Mater. Sol.
Cells 2008, 92, 686–714.
e
(11) Norrman, K.; Krebs, F. C. Surf. Interface Anal. 2004, 36, 1542–
1549.
(12) Norrman, K.; Krebs, F. C. Sol. Energy Mater. Sol. Cells 2006, 90,
213–227.
(13) Lira-Cantu, M.; Norrman, K.; Andreasen, J. W.; Krebs, F. C.
Chem. Mater. 2006, 18, 5684–5690.
(14) Norrman, K.; Larsen, N. B.; Krebs, F. C. Sol. Energy Mater. Sol.
Cells 2006, 90, 2793–2814.
(15) Alstrup, J.; Norrman, K.; Jorgensen, M.; Krebs, F. C. Sol. Energy
e
Mater. Sol. Cells 2006, 90, 2777–2792.
(16) Norrman, K.; Alstrup, J.; Krebs, F. C. Surf. Interface Anal. 2006,
38, 1302–1310.

1260 Macromolecules, Vol. 43, No. 3, 2010
Helgesen and Krebs
(17) Krebs, F. C.; Norrman, K. Prog. Photovoltaics 2007, 15, 697–712.
(18) Norrman, K.; Gevorgyan, S. A.; Krebs, F. C. ACS Appl. Mater.
Interfaces 2009, 1, 102–112.
(19) Chambon, S.; Rivaton, A.; Gardette, J. L.; Firon, M. Sol. Energy
Mater. Sol. Cells 2008, 92, 785–792.
(20) Chambon, S.; Manceau, M.; Firon, M.; Cros, S.; Rivaton, A.;
Gardette, J.-L. Polymer 2008, 49, 3288–3294.
(21) Manceau, M.; Rivaton, A.; Gardette, J.-L. Macromol. Rapid
(27) Banishoeib, F.; Henckens, A.; Fourier, S.; Vanhooyland, G.;
Breselge, M.; Manca, J.; Cleij, T. J.; Lutsen, L.; Vanderzande,
D.; Nguyen, L. H.; Neugebauer, H.; Sariciftci, N. S. Thin Solid
Films 2008, 516, 3978–3988.
(28) Girotto, C.; Cheyns, D.; Aernouts, T.; Banishoeib, F.; Lutsen, L.;
Cleij, T. J.; Vanderzande, D.; Genoe, J.; Poortmans, J.; Heremans,
P. Org. Electron. 2008, 9, 740–746.
(29) Krebs, F. C.; Spanggaard, H. Chem. Mater. 2005, 17, 5235–5237.
(30) Petersen, M. H.; Gevorgyan, S. A.; Krebs, F. C. Macromolecules
2008, 41, 8986–8994.
Commun. 2008, 29, 1823–1827.
(22) Manceau, M.; Rivaton, A.; Gardette, J.-L.; Guillerez, S.; Lemaı
ˆtre,
N. Polym. Degrad. Stab. 2009, 94, 898–907.
(31) Hagemann, O.; Bjerring, M.; Nielsen, N. C.; Krebs, F. C. Sol.
Energy Mater. Sol. Cells 2008, 92, 1327–1335.
(32) Aubert, P. H.; Knipper, M.; Groenendaal, L.; Lutsen, L.; Manca,
J.; Vanderzande, D. Macromolecules 2004, 37, 4087–4098.
(33) Ranger, M.; Rondeau, D.; Leclerc, M. Macromolecules 1997, 30,
7686–7691.
(23) Bertho, S.; Janssen, G.; Cleij, T. J.; Conings, B.; Moons, W.;
Gadisa, A.; D’Haen, J.; Goovaerts, E.; Lutsen, L.; Manca, J.;
Vanderzande, D. Sol. Energy Mater. Sol. Cells 2008, 92, 753–760.
(24) Henckens, A.; Colladet, K.; Fourier, S.; Cleij, T. J.; Lutsen, L.;
Gelan, J.; Vanderzande, D. Macromolecules 2005, 38, 19–26.
€
(25) Nguyen, L. H.; Gunes, S.; Neugebauer, H.; Sariciftci, N. S.;
Banishoeib, F.; Henckens, A.; Cleij, T. J.; Lutsen, L.; Vanderzande,
(34) Zhu, Z.; Waller, D.; Gaudiana, R.; Morana, M.; Muhlbacher, D.;
Scharber, M.; Brabec, C. Macromolecules 2007, 40, 1981–1986.
(35) Yang, X.; Loos, J. Macromolecules 2007, 40, 1353–1362.
(36) Gevorgyan, S. A.; Krebs, F. C. Chem. Mater. 2008, 20, 4386–4390.
(37) Helgesen, M.; Gevorgyan, S. A.; Krebs, F. C.; Janssen, R. A. J.
Chem. Mater. 2009, 21, 4669–4675.
D. Sol. Energy Mater. Sol. Cells 2006, 90, 2815–2828.
(26) Banishoeib, F.; Adriaensens, P.; Berson, S.; Guillerez, S.; Douheret,
O.; Manca, J.; Fourier, S.; Cleij, T. J.; Lutsen, L.; Vanderzande, D.
Sol. Energy Mater. Sol. Cells 2007, 91, 1026–1034.