High photovoltage achieved in low band gap polymer solar cells by adjusting energy levels of a polymer with the LUMOs of fullerene derivatives

Article abstract of DOI:10.1039/b811957k

Solar cells based on organic molecules or conjugated polymers attract great attention due to their unique advantages, such as low cost, and their use in flexible devices, but are still limited by their low power conversion efficiency (PCE). To improve the PCEs of polymer solar cells, more efforts have been made to increase short-circuit current (J_sc) or open-circuit voltage (V_oc). However, the trade-off between J_sc and V _oc in bulk heterojunctions solar cells makes it tricky to find a polymer with a low band gap to efficiently absorb photons in the visible and near infrared region of the solar spectrum, while maintaining a high V _oc in solar cells. Therefore, it is crucial to design and synthesize polymers with energy levels aligning with the LUMO (lowest unoccupied molecular orbital) of an electron acceptor to minimize the LUMO level difference between donor and acceptor to keep enough driving force for charge generation, thereby maximizing photovoltage in solar cells. Here a novel copolymer APFO-Green 9 was synthesized. Polymer solar cells based on APFO-Green 9 blended with a derivative of fullerene demonstrate high photovoltage by fine tuning the HOMO and LUMO level of APFO-Green 9. Solar cells based on APFO-Green 9 and [6,6]-phenyl-C71-butyric acid methyl ester ([70]PCBM) present a photoresponse extended to 900 nm with J_sc of 6.5 mA cm^-2, V_oc of 0.81 V and PCE of 2.3% under illumination of AM1.5 with light intensity of 100 mW cm^-2. As a low band gap polymer with a V_oc bigger than 0.8 V, APFO-Green 9 is a promising candidate for efficient tandem solar cells.

Full text of DOI:10.1039/b811957k

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
High photovoltage achieved in low band gap polymer solar cells
by adjusting energy levels of a polymer with the LUMOs
of fullerene derivatives
ab
ab
*
¨
Fengling Zhang,
and Mats R. Andersson
Johan Bijleveld,c Erik Perzon,c Kristofer Tvingstedt,ab Sophie Barrau,ab Olle Inganas
bc
*
Received 14th July 2008, Accepted 10th September 2008
First published as an Advance Article on the web 13th October 2008
DOI: 10.1039/b811957k
Solar cells based on organic molecules or conjugated polymers attract great attention due to their
unique advantages, such as low cost, and their use in flexible devices, but are still limited by their low
power conversion efficiency (PCE). To improve the PCEs of polymer solar cells, more efforts have been
made to increase short-circuit current (Jsc) or open-circuit voltage (Voc). However, the trade-off
between Jsc and Voc in bulk heterojunctions solar cells makes it tricky to find a polymer with a low band
gap to efficiently absorb photons in the visible and near infrared region of the solar spectrum, while
maintaining a high Voc in solar cells. Therefore, it is crucial to design and synthesize polymers with
energy levels aligning with the LUMO (lowest unoccupied molecular orbital) of an electron acceptor to
minimize the LUMO level difference between donor and acceptor to keep enough driving force for
charge generation, thereby maximizing photovoltage in solar cells. Here a novel copolymer
APFO-Green 9 was synthesized. Polymer solar cells based on APFO-Green 9 blended with a derivative
of fullerene demonstrate high photovoltage by fine tuning the HOMO and LUMO level of
APFO-Green 9. Solar cells based on APFO-Green 9 and [6,6]-phenyl-C71-butyric acid methyl ester
([70]PCBM) present a photoresponse extended to 900 nm with Jsc of 6.5 mA cmꢀ2, Voc of 0.81 V
and PCE of 2.3% under illumination of AM1.5 with light intensity of 100 mW cmꢀ2. As a low band
gap polymer with a Voc bigger than 0.8 V, APFO-Green 9 is a promising candidate for efficient tandem
solar cells.
which may be split into free charge carriers under strong electric
field or at the interface between two components with different
electron affinities and ionization potentials, thereby forming free
charge carriers.5 Therefore, the most efficient polymer solar cells
are based on heterojunctions with a large interface for facilitating
exciton dissociation or charge generation. The two components
used in polymer solar cells are polymers acting as electron donors
(D) and derivatives of fullerene acting as electron acceptors (A),
such as [6,6]-phenyl-C61-butyric acid methyl ester (PCBM)6 or
[6,6]-phenyl-C71-butyric acid methyl ester ([70]PCBM).1 The
LUMO level of an acceptor divides the band gap of a polymer
into two parts, LUMO(D)–LUMO(A) and LUMO(A)–HOMO
(highest occupied molecular orbital) (D) where LUMO(D)–
LUMO(A) determines the driving force for exciton dissociation
Introduction
Impressive progress has been made in the field of polymer solar
cells during the last decade. PCE of 5–6% has been demon-
strated.1,2 To make polymer solar cells viable, more effort is
needed to improve the PCE, which is proportional to photocur-
rent and photovoltage under illumination. To enhance photo-
current, many low band gap (Eg < 2 eV) polymers were
synthesized to match solar spectrum, which generally resulted in
reducing Voc because the band gap of a material is the up-limit
photogenerated potential in a solar cell.3 It is challenging to
fabricate a single junction solar cell using a low band gap material
to match solar spectrum, while maintaining a high Voc due to the
intrinsic conflict between maximum Jsc and maximum Voc.4
Unlike inorganic solar cells where the charge carriers are
directly generated upon absorbing photons, the situation is more
complicated in polymer solar cells because photocurrent is
generated via excitons (bound electron and hole pairs). Upon
absorbing photons in polymer solar cells, excitons are generated,
and
LUMO(A)–HOMO(D)
determines
photovoltage.
LUMO(D)–LUMO(A) should be at least larger than the exciton
binding energy of polymers (in the range of 0.1–0.4 eV) to
dissociate excitons.7 Therefore, the photocurrent of a polymer
solar cell relies on absorbance, electron affinity difference
between two components, charge carrier mobility, morphology8
and thickness9 of active layer. Contrary to photocurrent, pho-
tovoltage of a polymer solar cell is fairly stable, which is mainly
determined by LUMO(A)–HOMO(D) and the work function
difference between two electrodes.4,10 Consequently, alignment
of LUMO and HOMO levels of donor and acceptor plays an
essential role in determining the performance of polymer solar
cells. Small LUMO(D)–LUMO(A) benefits photovoltage, but
aBiomolecular and organic electronics, Department of Physics, Chemistry
and Biology, Linko¨ping University, SE-58183, Linko¨ping, Sweden
bCenter of Organic Electronics (COE), Linko¨ping University, Linko¨ping,
Sweden. E-mail: fenzh@ifm.liu.se; Fax: +46 13288969; Tel: +46 13281257
cPolymer Technology, Department of Chemical and Biological
Engineering, Chalmers University of Technology, SE-41296, Go¨teborg,
Sweden. E-mail: mats.andersson@chalmers.se; Fax: +46 317723418;
Tel: +46 317723401
5468 | J. Mater. Chem., 2008, 18, 5468–5474
This journal is ª The Royal Society of Chemistry 2008

View Article Online
may result in inefficient exciton dissociation or small photocur-
rent. On the other hand, large LUMO(D)–LUMO(A) benefits
photocurrent, but may unnecessarily reduce Voc, for instance,
poly(3-hexylthiophene)(P3HT) has a similar band gap with
APFO3 (about 2 eV), but very different Voc in solar cells.11,3b The
Voc of the solar cells based on P3HT:PCBM is smaller than
700 mV and the Voc of the solar cells based on APFO3:PCBM is
1 V due to smaller LUMO(D)–LUMO(A) in APFO3:PCBM
than in P3HT:PCBM.11 Therefore, from a materials point of
view, the synthesis of polymers with low band gap absorbing
visible and near infrared photons, with LUMO level aligning
with that of an electron acceptor to retain a high photovoltage is
a big challenge because there are only two ways to narrow the
band gaps of polymers, either by moving the HOMO level
toward vacuum level or the LUMO level away from vacuum
level. The former will result in a reduction of photovoltage and
the latter will attenuate the driving force for exciton dissociation
at the interface. PCBM is commonly used as the most efficient
electron acceptor both for charge generation and charge trans-
port.12 Therefore, the LUMO level of the polymer is generally
limited by the LUMO of PCBM, and otherwise new acceptors
are needed.13
Here we demonstrate that the goal of minimizing LUMO(D)–
LUMO(A) for exciton dissociation at the interface between
polymer and PCBM to maximum Voc in solar cells was achieved
in a novel synthesized low band gap alternating polyfluorene
copolymer APFO-Green 9 by fine tuning its HOMO and LUMO
levels to match the LUMO level of PCBM. In this polymer we
have chosen to introduce a pyrazino[2,3-g]quinoxaline unit in-
between two thiophene rings. The pyrazino[2,3-g]quinoxaline
acts as an electron-accepting part and the thiophenes as electron
donating, giving a partial charge transfer within the segment,
which results in a relatively low band gap of the final polymer
(Eg ꢁ 1.8 eV). Solar cells composed of APFO-Green 9 and either
PCBM or [70]PCBM (molecular structures shown in Fig. 1)
present a photoresponse extended to 900 nm and a Voc bigger
than 0.8 V. The latter case demonstrated a Jsc of 6.5 mA cmꢀ2
and a PCE of 2.3% under illumination of AM1.5 (100 mW cmꢀ2).
Scheme 1 Synthetic route to APFO-Green 9.
described procedure in ref. 15 actually worked, but only after
adding a few drops of water into the reaction mixture. The
formed tetraamine was immediately condensed with benzil and
compound 1 was formed. This compound was dibrominated
with N-bromosuccinimide yielding dibromide 2. This compound
was analyzed with MALDI-TOF (without addition of matrix)
and the resulting mass spectrum and isotope pattern was in
excellent agreement with the structure of 2. Compound 2 was
polymerized together with 2,7-bis(4,4,5,5-tetramethyl-1,3,2-
dioxaborolane-2-yl)-9,9-dioctylfluorene,16 the color of the final
polymer was green and the molecular weight was determined to
Mn ¼ 13 000 g molꢀ1 and Mw ¼ 30 000 g molꢀ1, relative to
polystyrene standards.
Results and discussion
Energy levels and optical properties
Synthesis of APFO-Green 9
To evaluate the HOMO and LUMO values of polymer, cyclic
voltammetry was used. The LUMO and HOMO values of
APFO-Green 9 were estimated using a cyclic voltammogram
(CV) to be 3.9 and 5.7 eV, respectively, with a HOMO–LUMO
gap of 1.8 eV, which corresponds to the peak of absorption
spectrum.
The synthetic route of the polymer is shown in Scheme 1. We
have previously described a modified procedure of preparing the
starting material.14 Attempts to reduce the starting material
following a procedure described in the literature15 failed and we
looked for other procedures for reducing the starting material.
After testing several different alternatives we found that the
Normalized optical absorbance of APFO-Green 9, PCBM,
[70]PCBM and blends of APFO-Green 9 with PCBM, and
[70]PCBM with the ratio of 1 : 4 (by weight) are shown in Fig. 2.
The absorption spectrum of pure APFO-Green 9 displays two
peaks at 400 and 710 nm, which extend to 900 nm. There is one
peak at 335 nm in pure PCBM film and three peaks at 320, 380
and 475 nm in pure [70]PCBM film. The peak of optical
absorption of APFO-Green 9 at 710 nm (1.75 eV) is a measure of
the band gap in a 1D semiconductor, and corresponds to the gap
from electrochemical study (vide infra) (1.8 eV). The long
wavelength absorption peak of APFO-Green 9 is greatly
Fig.
1 The chemical structures of APFO-Green 9, PCBM and
[70]PCBM.
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 5468–5474 | 5469

View Article Online
Fig. 2 The normalized absorptions of APFO-Green 9 (squares), PCBM
(filled circles), [70]PCBM (filled triangles) and the blends of APFO-Green
9 with PCBM (open-circles) and [70]PCBM (open triangles) in the ratio
of 1 : 4 (by weight).
attenuated in blends (and in devices) by adding 80% PCBMs.
Absorption of [70]PCBM complements the valley in the
absorption spectrum of APFO-Green 9, which gives a contribu-
tion to photocurrent in the devices.
Neat APFO-Green 9 film presents broad photoluminescence
(PL) peaking at 900 nm when excited at 450 or 650 nm. The
photoluminescence (PL) of APFO-Green 9 was significantly
quenched, but not completely in blend films with 80% PCBM or
[70]PCBM and 20% APFO-Green 9, which indicates incom-
pletely electron transfer from APFO-Green 9 to PCBMs.
However, the PL quenching in APFO-Green 9:[70]PCBM blend
is more pronounced than that in APFO-Green 9:PCBM. The
incomplete PL quenching may be partly due to small driving
forces at the interfaces between the two components. The offset
of LUMOs between APFO-Green 9 and PCBMs is only about
0.1 eV (LUMO of PCBM ꢁ ꢀ4.0 eV). It was observed that PL
quenching depends on pumping wavelength, and more quench-
ing appears when the film was excited by long wavelength
photons rather than short wavelength ones. On the other hand,
the emission from both PCBM (720 nm) and [70]PCBM (710 nm)
were quenched in blend films, which indicates efficient hole
transfer from PCBMs to APFO-Green 9. Hole transfer from
[70]PCBM to APFO-Green 9 was also confirmed by the fact that
the photocurrents of the devices do come from both components
with very pronounced contribution from [70]PCBM in EQE, due
to the high absorption in the visible range, compared to PCBM.
Fig. 3 (a) EQEs of the cells under illumination of monochromatic light,
(b) J–V characteristics under AM1.5 (100 mW cmꢀ2) where 1 : 2
(squares), 1 : 3 (circles), 1 : 4 (triangles), filled symbols for cells spin-cast
at 1000 rpm, open symbols at 2000 rpm.
Table 1 Summary of the performances of all the diodes
Ratio and thickness
of active layer
Jsc/mA cmꢀ2
Voc/V
FF
PCE (%)
1 : 2 (70 nm)
1 : 2 (90 nm)
1 : 3 (75 nm)
1 : 3 (140 nm)
1 : 4 (80 nm)
1 : 4 (110 nm)
3.9
4.2
4.4
4.9
4.6
4.8
0.85
0.83
0.84
0.83
0.82
0.82
0.39
0.38
0.43
0.41
0.44
0.42
1.3
1.3
1.6
1.7
1.6
1.7
parameters of the performance of these diodes are summarized in
Table 1.
Larger Jsc and lower FF were achieved in the diodes with
thicker active layer than thinner ones in all three stoichiometries.
Voc decreases with increasing amount of PCBM in the active
layers, which results in a maximum power conversion efficiency
(PCE) of 1.7% in the diodes with both stoichiometries of 1 : 3
and 1 : 4. Voc of all diodes, despite the thickness or stoichiome-
tries, are above 0.8 V, which is a remarkable number for low
band gap polymer solar cells.
Device characterization
First, solar cells based on APFO-Green 9 and PCBM with
different thicknesses (from 70 to 140 nm) and stoichiometries
(1 : 2, 1 : 3 and 1 : 4) were fabricated and characterized.
External quantum efficiencies (EQEs) of all cells under mono-
chromatic light are presented in Fig. 3a and the J–V character-
istics under illumination of AM1.5 are shown in Fig. 3b. Almost
identical profiles between the EQEs of the cells and the absor-
bance of APFO-Green 9 (Fig. 2) indicates that excitons are
mainly generated in polymer phase. It was observed that
contributions from long wavelength photons are pronounced in
thick active layers and contributions from short wavelength
photons are enhanced in thin active layers (Fig.3a). The
To compensate the valley around 500 nm in the absorbance of
APFO-Green 9 and to enhance Jsc, [70]PCBM was used to
replace PCBM as electron acceptor to harvest the photons in that
region. The EQE of the diode based on APFO-Green
9:[70]PCBM (1 : 3) is presented in Fig. 4a. The pronounced
contribution of [70]PCBM to EQE in visible range makes the
5470 | J. Mater. Chem., 2008, 18, 5468–5474
This journal is ª The Royal Society of Chemistry 2008

View Article Online
Fig. 5 The EQEs of the diodes of APFO-Green 9:PCBM (circles) and
APFO-Green 9:[70]PCBM (squares) with (open symbols) and without
(filled symbols) applied electric field.
optimized when synthesizing new polymers. The transport
properties of APFO-Green 9 are under investigation.
Morphology studies
Photocurrent of polymer solar cells is sensitive to morphology.
We imaged the surface of the active layer to know whether the
incomplete PL quenching in blend films is due to phase separa-
tion hindering electron transfer from APFO-Green 9 to PCBMs,
in addition to the small driving force at the interface between
APFO-Green 9 and PCBM. The morphology of the blends
APFO-Green9:[70]PCBM with the ratio of 1 : 3 spin-coated
from pure chloroform solution (CF) or from co-solvent (chlo-
roform (CF) mixed with dichlorobenzene (DCB) in ratio of
10 : 1 by volume) were recorded using atomic force microscopy
(AFM). The topography images of these two films are shown in
Fig. 6. The images indicate a strong dependence of the structure
on the nature of the solvent. Some small phase-separated
domains are present in the film spin-coated from CF solution
(rms roughness of 0.54 nm) while the one elaborated from the
CF:DCB solution presents a very smooth surface with an rms
roughness of 0.39 nm. The morphological difference is explained
by the higher solubility of [70]PCBM in DCB compared to CF,
resulting in the formation of smoother films.
Fig. 4 (a) EQEs of the cells based on APFO-Green 9:[70]PCBM
(squares) and PCBM (circles) with stoichiometry of 1 : 3 under illumi-
nation of monochromatic light, (b) I–V characteristics of the cells
based on APFO-Green 9:[70]PCBM under illumination of AM1.5
(100 mW cmꢀ2).
peak of APFO-Green 9 in EQE at 700 nm invisible. For
comparison, the EQE of the diode based on APFO-Green
9:PCBM (1 : 3) is also depicted in the same figure. By optimizing
the thickness and stoichiometry of the active layers, Jsc of 6.5 mA
cmꢀ2, Voc of 0.81 V, FF of 0.44 and PCE of 2.3% were achieved
for the diodes with 1 : 3 and thickness of 70 ꢂ 10 nm (see
Fig. 4b). As shown in Fig. 4a, the [70]PCBM indeed makes
a great contribution for harvesting the photons from 400 nm to
650 nm, which results in an enhancement of Jsc by 33% compared
to PCBM under illumination of AM1.5.
To know the limitation of the performance of the solar cells,
EQEs of the cells based on APFO-Green 9:[70]PCBM and
PCBM (1 : 3) were recorded under applied reverse electric field
(Fig. 5). The fact that EQE was increased by applied a reverse
bias of 1 V indicates that strong electric field was needed to
extract photoinduced charge carriers. Therefore one of the most
important loss mechanisms of the diodes was charge recombi-
nation due to less efficient extraction of charges from the active
layers after they were generated. The small FF and the field
dependence of the photocurrent also suggest that the PCE is
limited by transport, even though the charge generation is also
less efficient in this combination, due to smaller driving force at
the interface between two components as indicated by incom-
plete PL. Therefore, not only absorption coefficient and HOMO,
LUMO values, but also the transport property should be
Fig. 6 AFM topography of APFO-Green9:[70]PCBM (1 : 3) films spin-
coated from CF solution (a) and from a CF–DCB solvent mixture
solution with a ratio of 10 : 1 in volume (b). The scan size is 2 ꢃ 2 mm and
the Z-range is 10 nm.
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 5468–5474 | 5471

View Article Online
most solar cells based on low band gap polymers present lower
Voc and smaller Jsc than those of high band gap polymers, which
either limits the Jsc in series connection tandem cells or limit
the Voc in parallel connection tandem cells.18 APFO-Green 9
combined with [70]PCBM demonstrates decent Jsc and remark-
able Voc, which are crucial requirements for low band gap
polymers employed in efficient tandem solar cells. A high PCE
of 6.5 % was achieved in the stacked tandem solar cells composed
of P3HT and PCPDTBT with a Voc of 1.24 V.1 We have
demonstrated that Voc ¼ 1.8 V in folded reflective tandem solar
cells based on APFO-Green 9 and APFO3.18 As a low band gap
polymer with a Voc bigger than 0.8 V, APFO-Green 9 is
a promising polymer for tandem solar cells.
Experimental
Synthetic procedure
2,3,6,7-Tetraphenyl-9,10-dithien-2-ylpyrazino[2,3-g]quinoxaline.
A suspension of zinc dust (498 mg, 7.6 mmol) and 1,4-dithien-
2-yl-2,3-dinitrobenzo[2,1,3]thiadiazole14 (100 mg, 0.25 mmol) in
acetic acid (10 ml) and a few drops of water was kept at 60 ꢄC for
one hour and the reaction mixture turned white–pink. The
mixture was filtered and benzil (214 mg, 1.0 mmol) was added to
the pink solution. The solution was stirred for 48 hours at room
temperature, after which it was filtered and the residue was
washed with acetic acid to remove the excess benzil. The product,
80 mg (49%) was collected as a brown–black solid.
Fig. 7 (a) PL of blend films coated from CF solution (thick line) and
from CF:DCB solution (thin line). (b) the I–V characteristics of the solar
cells fabricated from CF solution (squares) and from CF:DCB solution
(circles).
1H NMR: (d, ppm) 8.48 (2H, dd, J 3.9 Hz), 7.82 (8H, m), 7.76
(2H, dd, J 5.1 Hz), 7.39 (14H, m)
2,3,6,7-Tetraphenyl-9,10-di(10-bromothien-2-yl)pyrazino[2,3-
g]quinoxaline. To a solution of 2,3,6,7-tetraphenyl-9,10-dithien-
2-ylquinazoline (1) (80 mg, 0.12 mmol) in chloroform (25 ml) and
of acetic acid (25 ml), N-bromosuccimide (44 mg, 0.48 mmol) was
added. This mixture was stirred for 3 hours at room temperature
in darkness after which TLC proved that the reaction was
finished. The reaction mixture was filtered and the precipitate
was collected. This yielded 70 mg (70%) of a black solid.
1H NMR: (d, ppm) 8.48 (2H, d, J 4.2 Hz), 7.82 (8H, m),
7.43 (12H, m), 7.29 (2H, d, J 4.5 Hz), MS (MALDI), m/z
805.8 (M+).
The PL quenching of these two films and the corresponding
device performance were compared. Similar PL quenching was
observed and shown in Fig. 7a. The I–V characteristics of the two
devices (Fig. 7b) present comparable performance, irrespective of
morphology difference, which indicates that domain size in the
blends has insignificant influence on exciton dissociation. The
main limitation for electron transfer from APFO-Green 9 to
PCBMs is the small offset of LUMO values between APFO-
Green 9 and PCBMs. Accordingly, the maximum Voc of the solar
cells made from the low band gap APFO-Greens was reached. A
further moving of the LUMO level of the polymer away from
vacuum level (towards the LUMO of PCBM) will result in loss of
photocurrent. Thus, with APFO-Green 9 optimized HOMO and
LUMO levels were realized to balance Voc and Jsc and to match
the LUMO level of PCBM.
APFO-Green 9
To a degassed toluene (5 ml) mixture of 2,3,6,7-tetraphenyl-
9,10-di(1-bromothien-2-yl)quinazoline (2) (70 mg, 0.086 mmol),
9,9-dioctyl-2,7-di(40,40,50,50-tetramethyldioxyboralane)fluorene
(58 mg, 0.91 mmol) and tetrakis(triphenylphosphine)palladium
(5 mg, 4 mmol) under nitrogen a tetraethylammonium hydroxide
solution (0.37ml, 20 wt% in water) was added. This mixture
was boiled under reflux for 3 hours after which bromobenzene
(20 mg, 0.12 mmol) was added. This was again refluxed for one
hour, then phenylboronic acid (15 mg, 0.12 mmol) was added.
After one additional hour of refluxing the mixture was left
stirring overnight. The reaction mixture was precipitated in
methanol (150 ml) and filtered. The crude polymer was dissolved
in boiling chloroform (50 ml) and extracted with NH4OH solu-
tion (50 ml, 28%). This extraction was repeated four times, but
the last time pure water was used. The organic layer was again
Potential application
It is challenging to reach 10% power conversion efficiency with
a single material due to the trade-off between Jsc and Voc. In
order to increase PCE of polymer solar cells, more efforts have
been made to build tandem diodes composed of one polymer or
two polymers, but few of them show bigger PCE than single cells
because the increase of Voc cannot usually compensate for the
loss of Jsc in series connection of tandem cells.1,17 To harvest
photons in a broad range of the solar spectrum, using polymers
with complementary absorbance are preferred, that is,
combining low band gap and high band gap polymer.1 However,
5472 | J. Mater. Chem., 2008, 18, 5468–5474
This journal is ª The Royal Society of Chemistry 2008

View Article Online
precipitated in methanol (150 ml), and filtered. The resulting
solid was soxhlet extracted overnight with diethyl ether, after
which the ether soaked thimble was allowed to dry in ambient
conditions. The thimble was then soxhlet extracted with chlo-
roform. The chloroform solution was reduced in volume and
the polymer was precipitated with methanol (70 ml) and collected
by filtration. The solid was dried in a vacuum oven overnight
(40 ꢄC). This yielded 57 mg (64 %) of a black–green solid.
1H NMR: (d, ppm) 8.73 (2H, m), 7.95 (8H, m), 7.84 (6H, m),
7.71 (2H, m), 7.48 (12H, m), 2.18 (4H, m), 1.13 (24H, m), 0.76
(6H, m).
Acknowledgements
This work was supported by the Center of Organic Electronics
(COE) at Linko¨ping University, financed by the Strategic
Research Foundation SSF of Sweden. Financial support from
the Knut and Alice Wallenberg Foundation for equipment is
greatly acknowledged.
References
1 J. Y. Kim, K. Lee, N. E. Coates, D. Moses, T. Q. Nguyen, M. Dante
and A. J. Heeger, Science, 2007, 317, 222.
2 (a) M. Green, K. Emery, Y. Hishikawa and W. Warta, Prog.
Photovolt.: Res. Appl., 2008, 16, 61; (b) E. G. Wang, L. Wang,
L. F. Lan, C. Luo, W. L. Zhuang, J. B. Peng and Y. Cao, Appl.
Phys. Lett., 2008, 92, 033307.
Physical property characterization
Absorption spectra were recorded using a spectrophotometer
(Lambda 950) and the PL spectra were characterized with an
Andor spectrometer (Shamrock sr-303i-B, coupled to a Newton
EMCCD detector). The samples for absorption and photo-
luminescence measurements were prepared on glass substrates.
Electrochemical experiments were done as reported in the liter-
ature.19 The film surface morphology was investigated using
atomic force microscopy (AFM) with a Dimension 3100 system
(Digital Instruments/Veeco) operating in tapping mode. Silicon
3 (a) M. Svensson, F. L. Zhang, S. C. Veenstra, W. J. H. Verhees,
J. C. Hummelen, J. M. Kroon, O. Inganas and M. R. Andersson,
Adv. Mater., 2003, 15, 988; (b) O. Inganas, M. Svensson, F. Zhang,
A. Gadisa, N. K. Persson, X. Wang and M. R. Andersson, Appl.
Phys. A: Mater. Sci. Process., 2004, 79, 31; (c) X. J. Wang,
E. Perzon, J. L. Delgado, P. de la Cruz, F. L. Zhang, F. Langa,
M. Andersson and O. Inganas, Appl. Phys. Lett., 2004, 85, 5081; (d)
F. L. Zhang, W. Mammo, L. M. Andersson, S. Admassie,
M. R. Andersson, L. Inganas and O. Ingands, Adv. Mater., 2006,
18, 2169; (e) A. Gadisa, W. Mammo, L. M. Andersson,
S. Admassie, F. Zhang, R. MatsAndersson and O. Ingana¨s,
Adv. Funct. Mater., 2007, 17, 3836; (f) M. M. Wienk, M. Turbiez,
J. Gilot and R. A. J. Janssen, Adv. Mater., 2008, 20, 2556; (g)
L. H. Slooff, S. C. Veenstra, J. M. Kroon, D. J. D. Moet,
J. Sweelssen and M. M. Koetse, Appl. Phys. Lett., 2007, 90, 143506.
4 M. C. Scharber, D. Wuhlbacher, M. Koppe, P. Denk, C. Waldauf,
A. J. Heeger and C. L. Brabec, Adv. Mater., 2006, 18, 789.
5 P. W. M. Blom, V. D. Mihailetchi, L. J. A. Koster and D. E. Markov,
Adv. Mater., 2007, 19, 1551.
6 (a) W. Ma, J. Y. Kim, K. Lee and A. J. Heeger, Macromol. Rapid
Commun., 2007, 28, 1776; (b) G. Li, V. Shrotriya, Y. Yao,
J. S. Huang and Y. Yang, J. Mater. Chem., 2007, 17, 3126; (c)
Y. Yao, C. J. Shi, G. Li, V. Shrotriya, Q. B. Pei and Y. Yang, Appl.
Phys. Lett., 2006, 89, 153507; (d) M. M. Wienk, M. G. R. Turbiez,
M. P. Struijk, M. Fonrodona and R. A. J. Janssen, Appl. Phys.
Lett., 2006, 88, 153511.
7 (a) D. Moses, J. Wang, A. J. Heeger, N. Kirova and S. Brazovski,
Proc. Nat. Acad. Sci. U. S. A., 2001, 98, 13496; (b) S. F. Alvarado,
P. F. Seidler, D. G. Lidzey and D. D. C. Bradley, Appl. Phys. Lett.,
1998, 81, 1082; (c) E. M. Conwell, Synth. Met., 1996, 83, 101; (d)
J. L. Bredas, J. Cornil and A. J. Heeger, Adv. Mater., 1996, 8, 447.
8 (a) F. L. Zhang, K. G. Jespersen, C. Bjorstrom, M. Svensson,
M. R. Andersson, V. Sundstrom, K. Magnusson, E. Moons,
A. Yartsev and O. Inganas, Adv. Funct. Mater., 2006, 16, 667; (b)
D. Chirvase, J. Parisi, J. C. Hummelen and V. Dyakonov,
Nanotechnology, 2004, 15, 1317; (c) T. Roder, H. S. Kitzerow and
J. C. Hummelen, Synth. Met., 2004, 141, 271.
cantilevers (NSG10) with a force constant of 5.5–22.5 N mꢀ1
,
a resonance frequency of 190–325 kHz and a tip curvature radius
of 10 nm were used. For observations, the films were prepared
under the same conditions as the devices.
Diode fabrication and characterization
The solar cells are fabricated by spin-coating. First a polymer
anode of PEDOT-PSS (EL grade) was spin-coated on ITO
coated on glass substrate following annealing at 120 ꢄC for 5
minutes. The active layer of mixed APFO-Green 9 and PCBM
was spin-coated from chloroform on top of PEDOT-PSS. LiF
(1nm) and Al (60nm) as top electrode were deposited under
vacuum on the active layer. The thickness of the active layer is
about 60–140 nm. The size of the diode is defined by a mask when
depositing Al under vacuum, and is about 5 mm2.
The spectral response was recorded by a Keithley 485
picoammeter under short circuit condition when devices were
illuminated through the ITO side with a monochromatic light
from a halogen lamp. Current–voltage characteristics were
recorded using a Keithley 2400 Source Meter under illumination
of AM 1.5 with intensity of 100 mW cmꢀ2 from a solar simulator
(Model SS-50A, Photo Emission Tech., Inc.). All fabrication and
characterizations were performed in an ambient environment.
9 D. W. Sievers, V. Shrotriya and Y. Yang, J. Appl. Phys., 2006, 100,
114509.
10 A. Gadisa, M. Svensson, M. R. Andersson and O. Inganas, Appl.
Phys. Lett., 2004, 84, 1609.
11 (a) G. Li, V. Shrotriya, Y. Yao, J. Huang and Y. Yang, J. Mater.
Chem., 2007, 17, 3126; (b) W. L. Ma, C. Y. Yang, X. Gong, K. Lee
and A. J. Heeger, Adv. Funct. Mater., 2005, 15, 1617.
12 (a) A. Gadisa, X. J. Wang, S. Admassie, E. Perzon, F. Oswald,
F. Langa, M. R. Andersson and O. Inganas, Org. Electron., 2006,
7, 195; (b) L. M. Andersson and O. Inganas, Appl. Phys. Lett.,
2006, 88, 082103; (c) L. M. Andersson, F. L. Zhang and
O. Inganas, Appl. Phys. Lett., 2006, 89, 142111; (d) K. G. Jespersen,
F. L. Zhang, A. Gadisa, V. Sundstrom, A. Yartsev and O. Inganas,
Org. Electron., 2006, 7, 235.
13 (a) X. J. Wang, E. Perzon, F. Oswald, F. Langa, S. Admassie,
M. R. Andersson and O. Inganas, Adv. Funct. Mater., 2005, 15,
1665; (b) X. J. Wang, E. Perzon, W. Mammo, F. Oswald,
S. Admassie, N. K. Persson, F. Langa, M. R. Andersson and
O. Inganas, Thin Solid Films, 2006, 511, 576; (c) F. B. Kooistra,
Conclusions
A low band gap polymer APFO-Green 9 was synthesized. Solar
cells based on APFO-Green 9 combined with PCBM or
[70]PCBM were fabricated and characterized. Decent Jsc (6.5 mA
cmꢀ2), impressive Voc (0.81 V) and PCE (2.3%) were obtained
under AM1.5 (100 mW cmꢀ2) by aligning the HOMO and
LUMO levels of the polymer to match the LUMO of PCBM,
extending absorption to harvest photons in near infrared region
and maintain Voc. The Voc (0.8 V) is among the largest value for
low band gap polymer solar cells, which makes the APFO-Green
9 a very competitive candidate for tandem solar cells.
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 5468–5474 | 5473

View Article Online
J. Knol, F. Kastenberg, L. M. Popescu, W. J. H. Verhees,
J. M. Kroon and J. C. Hummelen, Org. Lett., 2006, 9, 551.
14 E. Perzon, X. J. Wang, S. Admassie, O. Inganas and
M. R. Andersson, Polymer, 2006, 47, 4261.
15 C. Kitamura, S. Tanaka and Y. Yamashita, Chem. Mater., 1996, 8,
570.
16 M. J. Edelmann, J. M. Raimundo, N. F. Utesch, F. Diederich,
C. Boudon, J. P. Gisselbrecht and M. Gross, Helv. Chim. Acta,
2002, 85, 2195.
A. Hadipour, B. de Boer, J. Wildeman, F. B. Kooistra,
J. C. Hummelen, M. G. R. Turbiez, M. M. Wienk, R. A. J. Janssen
and P. W. M. Blom, Adv. Funct. Mater., 2006, 16, 1897; (c)
K. Kawano, N. Ito, T. Nishimori and J. Sakai, Appl. Phys. Lett.,
2006, 88, 073514; (d) J. Gilot, M. M. Wienk and R. A. J. Janssen,
Appl. Phys. Lett., 2007, 90, 143512; (e) V. Shrotriya, E. H. E. Wu,
G. Li, Y. Yao and Y. Yang, Appl. Phys. Lett., 2006, 88, 064104.
18 K. Tvingstedt, V. Andersson, F. Zhang and O. Inganas, Appl. Phys.
Lett., 2007, 91, 123514.
17 (a) G. Dennler, H. J. Prall, R. Koeppe, M. Egginger, R. Autengruber
and N. S. Sariciftci, Appl. Phys. Lett., 2006, 89, 073502; (b)
19 W. Mammo, S. Admassie, A. Gadisa, F. Zhang, O. Inganas and
M. R. Andersson, Sol. Energy Mater. Sol. Cells, 2007, 91, 1010.
5474 | J. Mater. Chem., 2008, 18, 5468–5474
This journal is ª The Royal Society of Chemistry 2008