Significant improved performance of photovoltaic cells made from a partially fluorinated cyclopentadithiophene/benzothiadiazole conjugated polymer

Article abstract of DOI:10.1021/ma3009178

A partially fluorinated low bandgap polymer, poly[2,6-(4,4-bis(2- ethylhexyl)-4H-cyclopenta [2,1-b;3,4-b′]dithiophene)-alt-4,7-(5-fluoro-[2, 1,3]-benzothiadiazole)] (PCPDTFBT) was synthesized through a microwave-assisted Stille polymerization. It was found that PCPDTFBT has better π-π stacking in solution than its nonfluorinated analogue, poly[2,6-(4,4-bis(2-ethylhexyl)- 4H-cyclopenta[2,1-b;3,4-b′]dithiophene)-alt-4,7-([2,1,3]-benzothiadiazole) ] (PCPDTBT), resulting in 2 times higher hole mobility. Power conversion efficiency (PCE) of the device using PCPDTFBT/PC₇₁BM as active layer (5.51%) is much higher than the device using PCPDTBT/PC₇₁BM (2.75%) that was fabricated under the same condition without using any solvent additive to modify the morphology. The significantly enhanced PCE is the result of improved open circuit voltage and short circuit current coming from the lower lying HOMO energy level and the appropriate morphology of PCPDTFBT. In addition, the device with PCPDTFBT/PC₇₁BM could also be processed from nonchlorinated organic solvents such as o-xylene to obtain high PCE of 5.32% (which is the highest value for PCPDTBT type polymers processed without using chlorinated solvents). Further device optimization by inserting a thin layer of fullerene-containing surfactant between the active layer and Ag cathode resulted in even higher PCE of 5.81%. These encouraging results showed that PCPDTFBT has the potential to be used as a low bandgap polymer to provide complementary absorption in tandem solar cells.

Full text of DOI:10.1021/ma3009178

Article
pubs.acs.org/Macromolecules
Significant Improved Performance of Photovoltaic Cells Made from a
Partially Fluorinated Cyclopentadithiophene/Benzothiadiazole
Conjugated Polymer
Yong Zhang, Jingyu Zou, Chu-Chen Cheuh, Hin-Lap Yip, and Alex K.-Y. Jen*
Department of Materials Science and Engineering, University of Washington, Seattle, Washington, 98195, United States
*
S Supporting Information
ABSTRACT: A partially fluorinated low bandgap polymer, poly-
2,6-(4,4-bis(2-ethylhexyl)-4H-cyclopenta [2,1-b;3,4-b′]-
dithiophene)-alt-4,7-(5-fluoro-[2,1,3]-benzothiadiazole)]
PCPDTFBT) was synthesized through a microwave-assisted Stille
[
(
polymerization. It was found that PCPDTFBT has better π−π
stacking in solution than its nonfluorinated analogue, poly[2,6-(4,4-
bis(2-ethylhexyl)-4H-cyclopenta[2,1-b;3,4-b′]dithiophene)-alt-4,7-
(
[2,1,3]-benzothiadiazole)] (PCPDTBT), resulting in 2 times
higher hole mobility. Power conversion efficiency (PCE) of the
device using PCPDTFBT/PC BM as active layer (5.51%) is much
7
1
higher than the device using PCPDTBT/PC BM (2.75%) that was
7
1
fabricated under the same condition without using any solvent
additive to modify the morphology. The significantly enhanced PCE
is the result of improved open circuit voltage and short circuit current coming from the lower lying HOMO energy level and the
appropriate morphology of PCPDTFBT. In addition, the device with PCPDTFBT/PC BM could also be processed from
71
nonchlorinated organic solvents such as o-xylene to obtain high PCE of 5.32% (which is the highest value for PCPDTBT type
polymers processed without using chlorinated solvents). Further device optimization by inserting a thin layer of fullerene-
containing surfactant between the active layer and Ag cathode resulted in even higher PCE of 5.81%. These encouraging results
showed that PCPDTFBT has the potential to be used as a low bandgap polymer to provide complementary absorption in
tandem solar cells.
1
. INTRODUCTION
maximize the photovoltaic performance, various methods, such
as thermal annealing, vapor annealing, and solvent additives
have been used to control the morphology for achieving
Bulk heterojunction (BHJ) based polymer solar cells (PSCs), in
which conjugated polymers are blended with fullerene
derivatives (i.e., PCBM), are very promising for realizing the
2
8,29
optimal phase separation.
Bazan and Heeger et al. have
1
−4
shown that the morphology of the active blend can be
effectively controlled by adding a small amount of high boiling
point solvent such as 1,8-dithioloctane (DTO) or 1,8-
diiodooctane (DIO) into the polymer solution. For example,
the devices made from poly[2,6-(4,4-bis(2-ethylhexyl)-4H-
cyclopenta[2,1-b;3,4-b′]dithiophene)-alt-4,7-(2,1,3-benzothia-
diazole)] (PCPDTBT) could reach much higher PCE by
adding a small amount of DTO or DIO into the PCPDTBT/
goal of achieving low-cost and scalable renewable energy.
Over the past decade, the power conversion efficiency (PCE) of
PSCs has been steadily increased to above 8% through various
optimizations of conjugated polymers, material/electrode
interfaces, and device architectures.
important developments for conjugated polymers is the rational
design of narrow band gap polymers to better match the solar
5
−13
One of the most
1
spectrum. In general, these polymers are copolymers based on
an electron-rich donor (D) and an electron-deficient acceptor
2
8,30
PC BM solution in chlorobenzene.
The optimized
71
PCPDTBT/PC BM devices showed a J of ∼15−16 mA
(
A) on the polymer backbone to facilitate the intramolecular
71
sc
−2
cm , a V of 0.37−0.6 V, a fill factor (FF) of 40−55%, and a
PCE of between 3.5 to 5.4%.
charge transfer between D and A. The molecular units such as
carbazole, benzodithiophene, and cyclopentadithiophene, etc.,
have been used as the donor, and benzothiadiazole,
thienothiophene, and thienopyrroledione, etc., have been
oc
2
8,30,31
However, without solvent
additive, the PCPDTBT/PCBM-based devices only gave a PCE
−
2
of 2.6−3.2% and a low J of 8−11 mA cm because of the
sc
2
8,32
5
,6,14−25
formation of unfavorable morphology.
Another well-
commonly used as the acceptor.
studied polymer, poly[(4,4′-bis(2-ethylhexyl)dithieno[3,2-
The morphological control of the BHJ active layer in PSC
plays a key role in charge generation, separation and transport
2
6,27
within the device.
The ideal morphology contains a phase
Received: May 6, 2012
Revised: June 12, 2012
Published: June 22, 2012
separation between polymer donor and fullerene acceptor to
form an interpenetrating network of ∼10 nm length. To
©
2012 American Chemical Society
5427
dx.doi.org/10.1021/ma3009178 | Macromolecules 2012, 45, 5427−5435

Macromolecules
Article
Scheme 1. Synthesis of Monomers and Polymer
b:2′,3′-d]silole)-2,6-diyl-alt-((5-octylthieno[3,4-c]pyrrole-4,6-
dione)-1,3-diyl)] (PDTSTPD), also showed significant mor-
phological changes upon the addition of processing additive.
Without DIO in processing solvent, the device performance of
PDTSTPD only showed a PCE of less than 1%. After adding
PCE of 5.51% with higher Voc (0.75 V) compared to its
PCPDTBT/PC BM analogue (<0.6 V). Most importantly, this
71
performance was achieved without using any solvent additive.
We have also demonstrated that PCPDTFBT can be processed
from a nonchlorinated solvent, o-xylene to achieve a high PCE
of 5.32%, which is the best performance reported so far for
devices processed from nonchlorinated solvents. Furthermore,
an even higher PCE of 5.81% could also be achieved after
inserting a thin layer of fullerene-containing surfactant between
the active layer and Ag cathode to facilitate the extraction of
electrons.
2
% DIO, the PCE of PDTSTPD device dramatically increased
33
to ∼7% due to the improved morphology. These results
showed that proper control of the morphology is very critical to
improve the overall performance of device. However, these
small amounts (0.2−3%) of high boiling point solvent additives
are difficult to control and remove afterward. In addition, the
tedious processing conditions are also unfavorable for large-area
inkjet or roll-to-roll printing due to possible residual solvent in
2
. RESULTS AND DISCUSSIONS
3
4,35
the device.
Therefore, it would be ideal if a conjugated
2.1. Synthesis. Scheme 1 shows the synthetic route of
polymer/PCBM blend can be processed from single solvent
PCPDTFBT. Compounds 1 and M1 were synthesized from
32
system to afford optimal morphology for fabricating high-
literature reported methods. The synthesis of M2 involves a
multistep synthesis starting from 2,5-dibromo-3-fluorobenzene
3
4,35
performance PSC.
36
Recently, the introduction of fluorine (F) atom onto
as shown in our previous paper. It is difficult to achieve high
molecular weight polymer under conventional oil-bath heating
because the M1 monomer is difficult to purify by column
conjugated polymer backbone has been proven to be an
7
,8,36
effective way to enhance the overall performance of PSCs.
31
The F atom plays two important roles: (1) the electron-
withdrawing property of F atom can lower the highest occupied
molecular orbital (HOMO) energy level, therefore, results in an
chromatography or recrystallization. Hence, PCPDTFBT was
synthesized through a microwave-assisted Stille polymerization,
which has been demonstrated to be effective in increasing the
31
increased V in the corresponding device; (2) F atom can form
molecular weight. Purifications were performed by reprecipi-
tation and Soxhlet extraction of the crude polymers with
acetone and hexane to remove oligomers and residual catalyst.
Because of the asymmetrical FBT monomer, it is worthy to
note that PCPDTFBT is a regiorandom polymer. The number-
oc
F−H, F−F bonding through inter- or intramolecular
interactions, which may affect π−π stacking of polymer to
fine-tune its morphology with fullerene.
Herein, we report a fluorinated polymer, PCPDTFBT
(
Scheme 1), synthesized via Stille polymerization of mono-
average molecular weight (M ) of PCPDTFBT was determined
n
fluoro-substituted benzothiadiazole and distannylcyclopentadi-
thiophene. Better π−π stacking in PCPDTFBT enables
PCPDTFBT/PC BM blend based PSC to have a promising
to be 23.4 kDa with a PDI of 1.54. PCPDTBT was prepared
also under the same condition with a M of 21.3 kDa and a PDI
n
of 1.78. The solubility of PCPDTFBT is strongly varied with
71
5
428
dx.doi.org/10.1021/ma3009178 | Macromolecules 2012, 45, 5427−5435

Macromolecules
Article
Figure 1. The UV−vis spectra of PCPDTFBT and PCPDTBT in CB and o-DCB solutions (a), PCPDTFBT in CB solution under different
temperatures (b) and in thin films (c) of both polymers.
different solvents. It was found that PCPDTFBT possesses
good solubility in o-dichlorobenzene (DCB), CB and o-xylene,
but has limited solubility in chloroform. However, PCPDTBT
is soluble in all above-mentioned solvents although they have
similar molecular weights, which indicates that PCPDTFBT has
stronger packing in solution than PCPDTBT.
spectrum. As shown in Figure 1b, the packing peak at 776 nm is
almost disappeared due to the breakup of polymer π−π
stacking (aggregation) upon heating. When the solution was
cooled to ∼30 °C, however, an intense packing peak
reappeared (Figure 1b). Further cooling of the solution to
room temperature (“cold” solution), the packing induced peak
became even more evident. In “warm” solution, the thermal
history was removed and the PCPDTFBT chains were fully
dissolved and freely extended and rotated compared to those in
“cold” solution. Under slow cooling, these polymer chains
aggregate due to stronger intermolecular interaction, therefore,
a stronger stacking peak was observed in the “cold” solution.
However, the absorption of PCPDTBT did not show such
2
.2. Optical Properties. The UV−vis absorption spectra of
PCPDTFBT and PCPDTBT were investigated in both CB and
DCB solutions and in thin film (Figure 1). As shown in Figure
1
a, PCPDTBT in CB and DCB solutions showed a
characteristic peak at the low energy absorption band, which
is attributed to the intramolecular charge transfer (ICT) band
1
6,32
between CPDT and BT.
A slightly red-shifted ICT peak
31
(
732 nm) was found in DCB solution compared to that was
changes, only blue-shifted during the temperature cycles.
observed in CB solution (722 nm). This could be due to the
better solubility of PCPDTBT in DCB than in CB solution.
However, it is interesting to find a double peak at the low
energy absorption band for PCPDTFBT in CB solution
These results clearly showed that PCPDTFBT has stronger
stacking than PCPDTBT due to subtle F−H and/or F−F
7
interactions. Figure 1c showed the absorption spectra of
PCPDTFBT and PCPDTBT in thin films. Both polymers
possess almost the same peaks and onset due to similar
polymer backbone. The introduction of F atom onto BT unit
seems to have little effect on the absorption spectra, regardless
(
Figure 1a). The peak at 722 nm is similar to that of
PCPDTBT; however, the stronger intensity peak at 776 nm is
believed to be the result of π−π stacking of PCPDTFBT
polymer chains. In DCB solution, PCPDTFBT still showed a
certain degree of packing which is evident by the shoulder peak
at ∼770 nm compared to the main ICT peak at 730 nm (Figure
3
6
the polymer stacking.
The absorption maxima for
PCPDTFBT and PCPDTBT are at 776 nm, which is similar
to the stacking peak observed for PCPDTFBT in CB solution.
The onset of thin film absorption for both polymers is 860 nm,
1a). The absorption onsets of PCPDTFBT in both CB and
28
DCB solutions are almost the same at 850 nm. The lower
intensity of the π−π stacking peak in DCB compared to that in
CB is attributed to the better solubility of polymer in DCB than
in CB.
To further verify the packing behavior, the CB solution of
PCPDTFBT was heated to 60 °C to measure its UV−vis
which corresponds to an optical bandgap of 1.44 eV.
2.3. Electrochemical Properties. The CV curves of
PCPDTFBT and PCPDTBT in thin films are shown in Figure
2. The HOMO energy level was determined from the onset of
oxidative peak. The LUMO energy level was calculated from
the difference between the HOMO energy level and the optical
5
429
dx.doi.org/10.1021/ma3009178 | Macromolecules 2012, 45, 5427−5435

Macromolecules
Article
levels of polymer and PCBM should enable efficient charge
separation in PSC.
2
.4. Charge Transporting Properties of Polymers. The
mobility of polymer is one of the key parameters for achieving
37−39
efficient polymer solar cell performance.
The polymer
field-effect transistors (FETs) were fabricated with top-contact
and bottom-gate geometry. SiO was used as the gate dielectric
2
with divinyltetramethyldisiloxane-bis(benzocyclobutene)
(
BCB) as the buffer layer. Detailed fabrication procedures
can be found in the Experimental Section. Figure 3 shows the
output and transfer characteristics of PCPDTFBT for positive
and negative. It is interesting to note that both polymers
showed ambipolar behavior, which may be beneficial for the
performance of solar cells. The charge carrier mobilities (μ) of
both devices were extracted from the following equation:
I ds = 1 μC(V − V )
W
2 L
2
i
G
T
Figure 2. CV curves of PCPDTFBT and PCPDTBT films.
Here W is the channel width, L is the channel length, C is the
i
bandgap. As shown in Figure 2, PCPDTFBT and PCPDTBT
showed similar oxidative behavior with good reversible peaks.
The HOMO of PCPDTFBT was determined to be −5.15 eV
compared to −5.02 eV for PCPDTBT, which is similar to the
gate dielectric capacitance per unit area, and V is the gate
G
voltage. The hole mobility of the PCPDTFBT and PCPDTBT
−2
2
−1 −1
devices in the saturated region are 1.4 × 10 cm V
6
S
and
.8 × 10− cm V S , respectively.
3
2
−1 −1
2
8,30
value reported in literature.
A deeper HOMO energy level
Since there is only a small variation of the polymer backbone,
(
0.13 eV) was found for PCPDTFBT compared to that of
the enhanced hole mobility of PCPDTFBT should be the result
of stronger π−π stacking which facilitates charge transport. The
electron mobility of PCPDTFBT and PCPDTBT are 1.2 ×
PCPDTBT after introducing a F atom onto the BT unit.
Therefore, a higher Voc in PCPDTFBT-based PSCs can be
expected due to deeper HOMO level of PCPDTFBT than
PCPDTBT. The LUMO levels of PCPDTFBT and PCPDTBT
are calculated to be −3.71 eV and −3.58 eV, respectively. The
reasonably large difference (>0.4 eV) between the LUMO
−
3
2
−1 −1
−3
2
−1 −1
10 cm V
S
and 1.8 × 10 cm V S , respectively
which are almost at the same order with hole mobility. These
combined ambioplar property and higher hole mobility of
Figure 3. Typical output characteristics (a) and the typical transfer characteristics (b, c) of PCPDTFBT.
5
430
dx.doi.org/10.1021/ma3009178 | Macromolecules 2012, 45, 5427−5435

Macromolecules
Article
Figure 4. J−V (a) and EQE (b) curves of PCPDTFBT and PCPDTBT devices processed from chlorobenzene solutions.
PCPDTFBT are beneficial to the photovoltaic performance,
especially to current density and fill factor.
PCPDTFBT/PC BM device is mainly attributed to higher V
71 oc
and Jsc.
2
.5. Bulk Heterojunction Photovoltaic Device Per-
It is known that the Voc of PSC is proportional to the
difference between the HOMO level of polymer and the
LUMO level of PC BM, regardless of the cathode
formance. 2.5.1. Performance of Devices Processed from
Chlorobenzene Solutions. The photovoltaic devices of
PCPDTFBT were investigated in the configuration of ITO/
PEDOT:PSS/polymer:PC BM/Ca/Al. PCPDTBT-based de-
7
1
modifications. Therefore, the lower lying HOMO level
(−5.15 eV) of PCPDTFBT compared to PCPDTBT (−5.02
eV) contributes to the increased V . In general, the J of a PSC
7
1
vices were also made under the same condition for direct
comparison. Detailed fabrication and characterization are
shown in the Experimental Section. The J−V curves of
oc
sc
is relevant to the charge mobility and morphology of a BHJ
blend. Higher hole mobility measured in PCPDTFBT should
also contribute to the higher Jsc and FF observed in
PCPDTFBT/PC BM device. It is well-known that proper
PCPDTFBT/PC BM and PCPDTBT/PC BM devices are
71
71
shown in Figure 4a. The optimized polymer to PC BM ratio is
7
1
71
1
:2. The active layers were processed from their pure CB
solutions without any additive. The device performance is
shown in Table 1. The PCPDTFBT/PC BM device showed a
control of nanoscale morphology of the active layer is also very
crucial to the performance of PSCs. Too large or too small
phase separation (domains) is not favorable for efficient charge
separation, which may lead to severe charge recombination.
Figure 5 and 6 showed the TEM and AFM images of the thin
film blends of PCPDTBT and PCPDTFBT with PC BM,
7
1
Table 1. Device Performance of PCPDTFBT and PCPDTBT
Devices
7
1
which were prepared from their CB solutions. As shown in
Voc
Jsc
FF
PCE
(%)
active layer (solvent)
PCPDTBT/PC BM
cathode
Ca/Al
(V) (mA cm−2) (%)
Figure 5, amorphous film was found in PCPDTBT blend film,
2
8,30
which is similar to that observed previously.
The smaller
0.65
0.75
0.77
0.76
10.1
15.0
14.4
15.0
42
49
48
51
2.75
5.51
5.32
5.81
71
a
(CB)
size of phase separation in PCPDTBT blend film increased the
PCPDTFBT/PC BM
Ca/Al
Ca/Al
possibility for charge recombination, therefore, resulted in
71
b
(
CB)
18
lower J and FF. Under the same condition, the larger phase
sc
PCPDTFBT/PC BM
71
c
separation was observed in PCPDTFBT blend film leading to
more efficient charge separation and transport than that of
PCPDTBT blend. The AFM images in Figure 6 further provide
the evidence for larger phase separation in PCPDTFBT/
PC BM compared to that of PCPDTBT/PC BM. In the
(
o-xylene)
PCPDTFBT/PC BM
Bis-C60/
Ag
71
b
(CB)
a
b
c
∼
95 nm. ∼120 nm. ∼110 nm.
71
71
PCPDTBT/PC BM case, it showed a smaller rms of 0.504 nm
71
compared to 1.205 nm for PCPDTFBT/PC BM, which may
increase the possibility for charge recombination.
promising PCE of 5.51% with a V of 0.75 V, a J of 15.0 mA
71
oc
sc
−2
cm , and a FF of 49%. However, the PCPDTBT/PC BM
71
As a result, higher J and FF values were obtained in the
device only gave lower values with a V of 0.65 V, a J of 10.1
sc
oc
sc
−2
PCPDTFBT-based device. By combining higher V , J , and
mA cm , and a FF of 42% which results in a PCE of 2.75%.
oc sc
FF, a higher PCE (5.51%) could be reached in PCPDTFBT-
based device compared to 2.75% for the PCPDTBT-based
device. It is worthy to note that these values were obtained
from the pure CB solution of PCPDTFBT/PC71BM, without
adding any additive. The attempt to add 2−3% of DIO did not
further increase the performance of the device. However, the
PCPDTBT device with 2% DIO showed an increased PCE
(3.69%) with similar Jsc but lower Voc (0.59 V).
The theoretical Voc is calculated to be 0.75 and 0.62 V under
donor
PCMB
the equation of Voc = 1/e(|E
| − |E
|) − 0.3 V
HOMO
LUMO
16,40
considering the LUMO level of PCBM to be −4.10 eV.
The external quantum efficiency (EQE) curves of devices are
shown in Figure 4b. It can be seen that the devices showed
broad response over the range 350−900 nm. The EQE value
between 400 to 800 nm for PCPDTFBT devices is more than
5
0% compared to ∼35% for PCPDTBT devices, which is
consistent with the result of measured J . The integrated J
2.5.2. Performance of Device Processed from o-Xylene
Solution. As discussed above, processing solvent is a critical
parameter in determining the performance of polymer solar
sc
sc
−
2
from EQE is 10.4 and 14.8 mA cm , respectively for
PCPDTBT and PCPDTFBT devices, showing the accuracy
of measurements. The significantly improved performance in
41−43
cells.
In literature, most of the efficient BHJ solar cells
5
431
dx.doi.org/10.1021/ma3009178 | Macromolecules 2012, 45, 5427−5435

Macromolecules
Article
Figure 5. TEM images of PCPDTBT/PC BM and PCPDTFBT/PC BM films processed from chlorobenzene solutions.
71
71
Figure 6. Tapping mode AFM images of PCPDTFBT/PC BM and PCPDTBT/PC BM films processed from chlorobenzene solutions.
71
71
were processed using chlorinated solvents, such as o-DCB and
CB, to afford better morphology due to better solubility and
There is very little information about D/A polymer based
device that can be processed from o-xylene solution to show
good efficiency. Since PCPDTFBT has good solubility in o-
xylene, the active layer of its photovoltaic device can also be
processed from its o-xylene solution. Figure 7a shows the J−V
curve of PCPDTFBT/PC BM device processed from o-xylene
9
,20,28,29,38,39
viscosity compared to other solvents.
However,
chlorinated solvents are not suitable for large scale production
due to their high cost, toxicity, and environmental issues.
Therefore, it is highly desirable to find an alternative solvent
that can be used to afford appropriate morphology during thin
film processing to lead to highly efficient PSCs. o-Xylene is a
good choice because of the similar boiling point and viscosity
compared to CB. The most studied polymer using o-xylene as
71
solution. The device showed a V of 0.77 V, a J of 14.4 mA
oc
sc
−2
cm , and a FF of 48%, which results in an overall PCE of
.32% (Table 1). This is comparable to device processed from
5
CB solution. To our knowledge, it is one of the highest values
reported in literature for device processed from nonchlorinated
solvents. This performance is attributed to the appropriate
44−46
solvent is P3HT because of its good solubility.
5
432
dx.doi.org/10.1021/ma3009178 | Macromolecules 2012, 45, 5427−5435

Macromolecules
Article
Figure 8. J−V curve of PCPDTFBT-based device with Bis-C
surfactant as an electron-collecting layer and Ag as cathode.
6
0
PCPDTBT/PC BM based device that was fabricated under
71
the same condition. The increased Voc is believed to be the
result of lower lying HOMO in PCPDTFBT. The larger phase
separation on the order of excition diffusion length and the
balanced ambipolar charge mobility contributed to the
Figure 7. (a) J−V curves of PCPDTFBT/PC BM devices processed
7
1
from chlorobenzene and o-xylene solutions. (b) AFM images of
PCPDTFBT/PC BM films processed from o-xylene solution.
enhanced J . These results showed that the introduction of F
sc
71
atom provides an effective way to simultaneously improve Voc
and morphology. Most importantly, there is no need to add any
solvent additive to control the morphology of active layer.
Furthermore, PCPDTFBT could also be processed from o-
xylene and its PSC showed a PCE of 5.32%, which is among the
best performance for donor−acceptor type polymers processed
from nonchlorinated solvents. In addition, the PCE of
PCPDTFBT device could also be increased to reach 5.81%
by inserting a thin layer of fullerene-containing surfactant
between the active layer and Ag cathode. These encouraging
results showed that PCPDTFBT has the potential to be used as
a low bandgap polymer to provide complementary absorption
in tandem solar cells.
morphology achieved for the blend (Figure 7b). The
PCPDTFBT/PC BM film processed from o-xylene solution
71
showed the similar phase separation as the film processed from
CB solution (Figure 6).
2.5.3. Device Performance with Hybrid Electron-Collecting
Layer/Ag Cathode. Cathode modification for organic elec-
tronics has received much attention because the interface
between active layer and electrode is critical for achieving
10,47−52
optimal device performance.
Water-/alcohol-soluble
polymers or polyelectrolytes have been employed as effective
interfacial modification materials for improving the perform-
4
8,49,51,53
ance of PSCs.
Recently, we have reported a new
alcohol-soluble, bis-adduct fullerene surfactant and its function
as an efficient electron selective material when inserted as a thin
layer between the active material and high work function
cathode, such as Al or Ag. We have adapted this approach
here to improve the performance of PCPDTFBT-based
devices. The device were fabricated with the configuration of
4
. EXPERIMENTAL SECTION
General Characterization Methods. UV−vis spectra were
54
measured using a Perkin-Elmer Lambda-9 spectrophotometer. The
1
H NMR spectra were collected on a Bruker AV 500 spectrameter
operating at 500 MHz in deuterated chloroform solution with TMS as
reference. Cyclic voltammetry of polymer film was conducted in
acetonitrile with 0.1 M of tetrabutylammonium hexafluorophosphate
using a scan rate of 100 mV s . ITO, Ag/AgCl, and Pt mesh were
used as working electrode, reference electrode and counter electrode,
respectively.
ITO/PEDOT:PSS/PCPDTFBT:PC BM/Bis-C /Ag. The
7
1
60
−1
Bis-C60 was spin-coated from its methanol solution. Ag was
selected as cathode because it is air stability and good
reflectivity. Figure 8 shows the J-V curve of PCPDTFBT/
PC BM device with Bis-C surfactant as the electron-
Device Fabrication. Polymer FETs were fabricated through the
7
1
60
top-contact and bottom-gate geometry. A thermally grown 300 nm
thickness of SiO was purchased from Montco Silicon Technologies
collecting layer. The PCE was further improved to be 5.81%
−2
2
(
a V of 0.76 V, a J of 15.0 mA cm , and a FF of 51%, Table
oc sc
INC and used as the gate dielectric. The source/drain regions were
defined by a 50 nm thickness of Au through a shadow mask, and the
channel length (L) and width (W) was 30 and 1000 μm, respectively.
Before gilding the devices, the oxide layer was passivated with a thin
divinyltetramethyldisiloxane−bis(benzocyclobutene) (BCB) buffer
layer. BCB precursor solution in toluene was spun onto the silicon
oxide and subsequently annealed at 250 °C overnight. The total
capacitance density measured from parallel-plate capacitors was 10.6
nF/cm . Polymer solution in CB was first filtered through 0.2 μm
syringe filters, spin-coated onto the BCB-modified substrate, and then
annealed at 110 °C for 10 min under nitrogen. Output and transfer
characteristics of the transistor devices were performed in a N -filled
1
), which is even higher than the device with Ca/Al as cathode.
This enhancement may be attributed to the efficient charge
collection and as an optical buffer between active layer and
cathode.
5
4,55
3. CONCLUSION
We have designed and synthesized a partially fluorinated low
bandgap polymer, PCPDTFBT, through microwave-assisted
Stille polymerization between CPDT and FBT units. By
introducing F atom onto the BT unit, PCPDTFBT exhibited
better π−π stacking in solution than PCPDTBT. The PSC
processed from the solution of PCPDTFBT/PC BM in CB
2
2
glovebox using an Agilent 4155B semiconductor parameter S6
analyzer. The field-effect mobility was calculated in the saturation
regime from the linear fit of (Ids)1 vs V . The threshold voltage (V )
7
1
/2
showed a high PCE of 5.51% compared to 2.75% for
gs
t
5
433
dx.doi.org/10.1021/ma3009178 | Macromolecules 2012, 45, 5427−5435

Macromolecules
Article
was estimated as the x intercept of the linear section of the plot of
After filtering, the purple solid was collected and dried overnight under
1/2
1
(
I ) vs Vgs.
vacuum (140 mg). H NMR (o-DCB (o-C D Cl ), ppm): 8.55−8.35
ds
6
4
2
PSCs were fabricated using ITO-coated glass substrates (15 Ω/sq),
(d, br, 2H), 7.66 (br, 1H), 2.23 (br, 2H), 1.40−1.18 (d, br, 32H).
which were cleaned with detergent, deionized water, acetone, and
isopropyl alcohol. A thin layer (ca. 30 nm) of PEDOT:PSS (Baytron P
VP AI 4083, filtered at 0.45 μm) was first spin-coated on the
precleaned ITO-coated glass substrates at 5,000 rpm and baked at 140
Molecular weight: M = 23.4 kDa, PDI = 1.54.
n
ASSOCIATED CONTENT
■
*
S
Supporting Information
H NMR spectra of PCPDTFBT. This material is available free
°
C for 10 min under ambient conditions. The substrates were then
transferred into a nitrogen-filled glovebox. Subsequently, the
1
polymer:PC BM active layer was spin-coated onto the PEDOT:PSS
of charge via the Internet at http://pubs.acs.org.
71
layer. For devices, the solution was prepared by dissolving the polymer
and fullerene at a 1:2 weight ratio in chlorobenzene or in o-xylene
overnight and filtered through a 0.2 μm PTFE filter, and the substrates
were annealed at 110 °C for 10 min prior to electrode deposition. For
the device with surfactant, the Bis-C60 surfactant in methanol was spin-
coated onto the active layer. At the final stage, the substrates were
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: ajen@u.washington.edu.
−6
Notes
pumped under high vacuum (<2 × 10 Torr), and calcium (20 nm)
topped with aluminum (100 nm) or silver (100 nm) was thermally
evaporated onto the active layer. Shadow masks were used to define
The authors declare no competing financial interest.
−2
2
the active area (10.08 × 10 cm ) of the devices.
ACKNOWLEDGMENTS
■
Device Characterization. The current−voltage (J−V) character-
istics of unencapsulated photovoltaic devices were measured under
ambient using a Keithley 2400 source-measurement unit. An Oriel
xenon lamp (450 W) with an AM1.5 G filter was used as the solar
simulator. A Hamamatsu silicon solar cell with a KG5 color filter,
which is traced to the National Renewable Energy Laboratory
The authors thank the support from NSF (DMR-0120967),
DOE (DEFC3608GO18024/A000), AFOSR (FA9550-09-1-
0426), ONR (N00014-11-1-0300), AOARD (FA2386-11-1-
4072). C.-C.C. thanks the National Science Council, Taiwan
(NSC98-2917-I-564-031) for support. The authors thank Mr.
Xi Yang for helping with the TEM measurement.
(
NREL), was used as the reference cell. To calibrate the light
intensity of the solar simulator, the power of the xenon lamp was
adjusted to make the short-circuit current (ISC) of the reference cell
under simulated sun light as high as it was under the calibration
condition. The spectral mismatches resulting from the test cells, the
reference cell, the solar simulator, and the AM1.5 were calibrated with
mismatch factors (M). According to Shrotriya et al. the mismatch
factor is defined as
REFERENCES
■
(1) Chen, J. W.; Cao, Y. Acc. Chem. Res. 2009, 42, 1709−1718.
(2) Cheng, Y. J.; Yang, S. H.; Hsu, C. S. Chem. Rev. 2009, 109, 5868−
5923.
(3) Krebs, F. C. Sol. Energy Mater. Sol. Cells 2009, 93, 394−412.
4) Thompson, B. C.; Frechet, J. M. J. Angew. Chem., Int. Ed. 2008,
(
λ2
λ2
∫λ ERef (λ)S (λ) dλ ∫ E (λ)S (λ) dλ
47, 58−77.
R
S
T
1
λ2
λ1
λ2
M =
(5) Chen, H. Y.; Hou, J. H.; Zhang, S. Q.; Liang, Y. Y.; Yang, G. W.;
Yang, Y.; Yu, L. P.; Wu, Y.; Li, G. Nature Photon. 2009, 3, 649−653.
∫λ ERef (λ)S (λ) dλ ∫ E (λ)S (λ) dλ
T
S
R
1
λ1
(
6) Liang, Y. Y.; Xu, Z.; Xia, J. B.; Tsai, S. T.; Wu, Y.; Li, G.; Ray, C.;
where E (λ) is the reference spectral irradiance (AM1.5), E (λ) is the
Yu, L. P. Adv. Mater. 2010, 22, E135−E138.
ref
S
source spectral irradiance, S (λ) is the spectral responsivity of the
(7) Zhou, H. X.; Yang, L. Q.; Stuart, A. C.; Price, S. C.; Liu, S. B.;
You, W. Angew. Chem., Int. Ed. 2011, 50, 2995−2998.
(8) Price, S., C.; Stuart, A. C.; Yang, L. Q.; Zhou, H. X.; You, W. J.
Am. Chem. Soc. 2011, 133, 4625−4631.
R
reference cell, and S (λ) is the spectral responsivity of the test cell,
T
each as a function of wavelength (λ). The spectral responsivities of the
test cells and the reference cell were calculated from the corresponding
external quantum efficiencies (EQE) by the relationship
(9) Amb, C. M.; Chen, S.; Graham, K. R.; Subbiah, J.; Small, C. E.;
So, F.; Reynolds, J. R. J. Am. Chem. Soc. 2011, 133, 10062−10065.
(10) He, Z.; Zhong, C.; Huang, X.; Wong, W.-Y.; Wu, H.; Chen, L.;
Su, S.; Cao, Y. Adv. Mater. 2011, 23, 4636−4643.
S(λ) = qh λc EQE(λ)
5
(11) Beaujuge, P. M.; Frec
0009−20029.
12) Zhang, Y.; Zou, J.; Yip, H.-L.; Chen, K.-S.; Zeigler, D. F.; Sun,
Y.; Jen, A. K. Y. Chem. Mater. 2011, 23, 2289−2291.
13) Wang, E.; Ma, Z.; Zhang, Z.; Vandewal, K.; Henriksson, P.;
Inganas, O.; Zhang, F.; Andersson, M. R. J. Am. Chem. Soc. 2011, 133,
́
het, J. M. J. J. Am. Chem. Soc. 2011, 133,
where the constant term q/hc equals 8.07 × 10 for wavelength in
−1
2
units of meters and S(λ) in units of AW . The Hamamatsu solar cell
was also used as the detector for determining the spectral irradiance of
the solar simulator. To minimize the spectral transformation, the
irradiance spectrum has been calibrated with the spectral responsively
of the Hamamatsu cell and the grating efficiency curve of the
monochromator (Oriel Cornerstone 130).
(
(
̈
14244−14247.
Materials Synthesis. All chemicals, unless otherwise specified,
were purchased from Aldrich and used as received. The monomer
FBT was synthesized by following the literature method as shown in
(14) Zhang, Y.; Hau, S. K.; Yip, H.-L.; Sun, Y.; Acton, O.; Jen, A. K.
Y. Chem. Mater. 2010, 22, 2696−2698.
(15) Blouin, N.; Michaud, A.; Leclerc, M. Adv. Mater. 2007, 19,
2295−2300.
36
32
Scheme S1. Compound 1 was prepared as reported previously.
The polymer PCPDTBT was synthesized from compound 1 and 4,7-
(16) Muhlbacher, D.; Scharber, M.; Morana, M.; Zhu, Z. G.; Waller,
D.; Gaudiana, R.; Brabec, C. Adv. Mater. 2006, 18, 2884−2889.
(17) Zou, Y. P.; Najari, A.; Berrouard, P.; Beaupre, S.; Aich, B. R.;
Tao, Y.; Leclerc, M. J. Am. Chem. Soc. 2010, 132, 5330−5331.
(18) Zhang, Y.; Zou, J. Y.; Yip, H. L.; Sun, Y.; Davies, J. A.; Chen, K.
S.; Acton, O.; Jen, A. K. Y. J. Mater. Chem. 2011, 21, 3895−3902.
(19) Chu, T. Y.; Lu, J. P.; Beaupre, S.; Zhang, Y. G.; Pouliot, J. R.;
Wakim, S.; Zhou, J. Y.; Leclerc, M.; Li, Z.; Ding, J. F.; Tao, Y. J. Am.
Chem. Soc. 2011, 133, 4250−4253.
3
1,32
dibromobenzothiadiazole by following the literature method.
Synthesis of PCPDTFBT. In a 10 mL tube, compound 1 (320 mg,
0
.44 mmol), FBT (125 mg, 0.40 mmol), Pd (dba) (7 mg) and P(o-
2
3
tol) (18 mg) were added consequently. After purging for three times
3
with nitrogen, chlorobenzene (3 mL) was added into the mixture.
Then, the tube was heated up to 140 °C for 1 h under microwave
heating. After cooling to room temperature, the resulted mixture was
poured into hexane and stirred for 1 h. The collected precipitate was
then dissolved into a small amount of chlorobenzene, which was
poured into hexane, and the process was repeated one more time.
(20) Piliego, C.; Holcombe, T. W.; Douglas, J. D.; Woo, C. H.;
Beaujuge, P. M.; Frechet, J. M. J. J. Am. Chem. Soc. 2010, 132, 7595.
5
434
dx.doi.org/10.1021/ma3009178 | Macromolecules 2012, 45, 5427−5435

Macromolecules
Article
(
2
(
21) Zhang, G. B.; Fu, Y. Y.; Zhang, Q.; Xie, Z. Y. Chem. Commun.
010, 46, 4997−4999.
22) Su, M.-S.; Kuo, C.-Y.; Yuan, M.-C.; Jeng, U. S.; Su, C.-J.; Wei,
K.-H. Adv. Mater. 2011, 23, 3315−3319.
23) Liang, Y. Y.; Feng, D. Q.; Wu, Y.; Tsai, S. T.; Li, G.; Ray, C.; Yu,
L. P. J. Am. Chem. Soc. 2009, 131, 7792−7799.
24) Liang, Y. Y.; Wu, Y.; Feng, D. Q.; Tsai, S. T.; Son, H. J.; Li, G.;
Yu, L. P. J. Am. Chem. Soc. 2009, 131, 56−57.
25) Huo, L. J.; Hou, J. H.; Zhang, S. Q.; Chen, H. Y.; Yang, Y.
Angew. Chem. Int. Edit 2010, 49, 1500−1503.
26) van Bavel, S.; Veenstra, S.; Loos, J. Macromol. Rapid Commun.
(52) Li, C.-Z.; Yip, H.-L.; Jen, A. K. Y. J. Mater. Chem. 2012, 22,
4161−4177.
(53) Ma, H.; Yip, H.-L.; Huang, F.; Jen, A. K. Y. Adv. Funct. Mater.
2010, 20, 1371−1388.
(
(54) O’Malley, K. M.; Li, C.-Z.; Yip, H.-L.; Jen, A. K. Y. Adv. Energy
Mater. 2012, 2, 82−86.
(55) Li, C. Z.; Chueh, C. C.; Yip, H. L.; O’Malley, K. M.; Chen, W.
C.; Jen, A. K. Y. J. Mater. Chem. 2012, 22, 8754−8578.
(
(
(
2
(
(
010, 31, 1835−1845.
27) Yang, X.; Loos, J. Macromolecules 2007, 40, 1353−1362.
28) Peet, J.; Kim, J. Y.; Coates, N. E.; Ma, W. L.; Moses, D.; Heeger,
A. J.; Bazan, G. C. Nat. Mater. 2007, 6, 497−500.
29) Li, G.; Shrotriya, V.; Huang, J. S.; Yao, Y.; Moriarty, T.; Emery,
K.; Yang, Y. Nat. Mater. 2005, 4, 864−868.
30) Lee, J. K.; Ma, W. L.; Brabec, C. J.; Yuen, J.; Moon, J. S.; Kim, J.
Y.; Lee, K.; Bazan, G. C.; Heeger, A. J. J. Am. Chem. Soc. 2008, 130,
619−3623.
31) Coffin, R. C.; Peet, J.; Rogers, J.; Bazan, G. C. Nature Chem
009, 1, 657−661.
32) Zhu, Z.; Waller, D.; Gaudiana, R.; Morana, M.; Muhlbacher, D.;
Scharber, M.; Brabec, C. Macromolecules 2007, 40, 1981−1986.
33) Chu, T.-Y.; Lu, J.; Beaupre, S.; Zhang, Y.; Pouliot, J.-R. m.;
(
(
3
(
2
(
(
́
Wakim, S.; Zhou, J.; Leclerc, M.; Li, Z.; Ding, J.; Tao, Y. J. Am. Chem.
Soc. 2011, 133, 4250−4253.
(
(
34) Krebs, F. C. Sol. Energy Mater. Sol. Cells 2009, 93, 1636−1641.
35) Zhang, Y.; Zou, J.; Yip, H.-L.; Chen, K.-S.; Davies, J. A.; Sun, Y.;
Jen, A. K. Y. Macromolecules 2011, 44, 4752−4758.
36) Zhang, Y.; Chien, S.-C.; Chen, K.-S.; Yip, H.-L.; Sun, Y.; Davies,
J. A.; Chen, F.-C.; Jen, A. K. Y. Chem. Commun. 2011, 47, 11026−
1028.
37) Qin, R. P.; Li, W. W.; Li, C. H.; Du, C.; Veit, C.; Schleiermacher,
(
1
(
H. F.; Andersson, M.; Bo, Z. S.; Liu, Z. P.; Inganas, O.; Wuerfel, U.;
Zhang, F. L. J. Am. Chem. Soc. 2009, 131, 14612−+.
(
38) Liang, Y. Y.; Feng, D. Q.; Wu, Y.; Tsai, S. T.; Li, G.; Ray, C.; Yu,
L. P. J. Am. Chem. Soc. 2009, 131, 7792.
39) Liang, Y. Y.; Wu, Y.; Feng, D. Q.; Tsai, S. T.; Son, H. J.; Li, G.;
Yu, L. P. J. Am. Chem. Soc. 2009, 131, 56.
40) Scharber, M. C.; Wuhlbacher, D.; Koppe, M.; Denk, P.;
Waldauf, C.; Heeger, A. J.; Brabec, C. L. Adv. Mater. 2006, 18, 789−
94.
41) Kawano, K.; Sakai, J.; Yahiro, M.; Adachi, C. Sol. Energ. Mat. Sol.
C 2009, 93, 514−518.
42) Cai, W.; Wang, M.; Zhang, J.; Wang, E.; Yang, T.; He, C.;
Moon, J. S.; Wu, H.; Gong, X.; Cao, Y. J. Phys. Chem. C 2011, 115,
314−2319.
43) Alem, S.; Chu, T.-Y.; Tse, S. C.; Wakim, S.; Lu, J.; Movileanu,
R.; Tao, Y.; Belanger, F.; Desilets, D.; Beaupre, S.; Leclerc, M.;
Rodman, S.; Waller, D.; Gaudiana, R. Org. Electron. 2011, 12, 1788−
793.
44) Park, C.-D.; Fleetham, T. A.; Li, J.; Vogt, B. D. Org. Electron.
011, 12, 1465−1470.
45) Hoth, C. N.; Steim, R.; Schilinsky, P.; Choulis, S. A.; Tedde, S.
F.; Hayden, O.; Brabec, C. J. Org. Electron. 2009, 10, 587−593.
46) Zhao, J.; Swinnen, A.; Van Assche, G.; Manca, J.; Vanderzande,
D.; Mele, B. V. J. Phys. Chem. B 2009, 113, 1587−1591.
47) He, C.; Zhong, C.; Wu, H.; Yang, R.; Yang, W.; Huang, F.;
Bazan, G. C.; Cao, Y. J. Mater. Chem. 2010, 20, 2617−2622.
48) Seo, J. H.; Gutacker, A.; Sun, Y.; Wu, H.; Huang, F.; Cao, Y.;
Scherf, U.; Heeger, A. J.; Bazan, G. C. J. Am. Chem. Soc. 2011, 133,
416−8419.
49) He, Z.; Zhang, C.; Xu, X.; Zhang, L.; Huang, L.; Chen, J.; Wu,
H.; Cao, Y. Adv. Mater. 2011, 23, 3086−3089.
50) Zou, J.; Yip, H.-L.; Zhang, Y.; Gao, Y.; Chien, S.-C.; O’Malley,
(
(
7
(
(
2
(
́
́
́
1
(
2
(
(
(
(
8
(
(
K.; Chueh, C.-C.; Chen, H.; Jen, A. K. Y. Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.20110293.
(
51) Yip, H.-L.; Jen, A. K. Y. Energ. Environ. Sci. 2012, 5, 5994−6011.
5
435
dx.doi.org/10.1021/ma3009178 | Macromolecules 2012, 45, 5427−5435