Processing additives for improved efficiency from bulk heterojunction solar cells.
ABSTRACT Two criteria for processing additives introduced to control the morphology of bulk heterojunction (BHJ) materials for use in solar cells have been identified: (i) selective (differential) solubility of the fullerene component and (ii) higher boiling point than the host solvent. Using these criteria, we have investigated the class of 1,8-di(R)octanes with various functional groups (R) as processing additives for BHJ solar cells. Control of the BHJ morphology by selective solubility of the fullerene component is demonstrated using these high boiling point processing additives. The best results are obtained with R = Iodine (I). Using 1,8-diiodooctane as the processing additive, the efficiency of the BHJ solar cells was improved from 3.4% (for the reference device) to 5.1%.
- The Journal of Physical Chemistry C 08/2014; 118(32):18424-18435. · 4.84 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: In this work, we investigate the morphology and microstructure of the aggregates, and the gelation behaviour of Poly(3-hexylthiophene) (P3HT) conjugated polymer in xylene solution as functions of P3HT concentration and aging time by the means of ageing time test, wide angle X-ray diffraction (WAXD), scanning electron microscopy (SEM), UV-visible absorption (UV-vis) and photoluminescence (PL) spectra. The result reveals that the gelation time of P3HT/xylene solution decreases markedly with increasing P3HT concentration. The photophysical properties of the P3HT aggregates in P3HT/xylene solution increase as P3HT concentration and ageing time are raised. It indicates that the well soluble P3HT polymer chains in xylene solution present microphase separation and self-assemble into stiff sheetlike structure, which associates by rodlike nanowhiskers of P3HT polymers during aging. Upon prolonged aging, the sheetlike structure of P3HT aggregates to from the three-dimension network that improves the electronic particle mobility in the organic solar cell.Applied Mechanics and Materials. 12/2013; 479-480:115-120.
- The Journal of Physical Chemistry C 09/2014; 118(37):21377-21384. · 4.84 Impact Factor
Processing Additives for Improved Efficiency from Bulk
Heterojunction Solar Cells
Jae Kwan Lee,†Wan Li Ma,†Christoph J. Brabec,‡Jonathan Yuen,†Ji Sun Moon,†
Jin Young Kim,†Kwanghee Lee,†Guillermo C. Bazan,†and Alan J. Heeger*,†
Center for Polymers and Organic Solids, UniVersity of California at Santa Barbara,
Santa Barbara, California 93106, and Konarka Technologies Austria, Altenbergerstrasse 69,
A-4-4- Linz, Austria
Received November 13, 2007; E-mail: firstname.lastname@example.org
Abstract: Two criteria for processing additives introduced to control the morphology of bulk heterojunction
(BHJ) materials for use in solar cells have been identified: (i) selective (differential) solubility of the fullerene
component and (ii) higher boiling point than the host solvent. Using these criteria, we have investigated
the class of 1,8-di(R)octanes with various functional groups (R) as processing additives for BHJ solar cells.
Control of the BHJ morphology by selective solubility of the fullerene component is demonstrated using
these high boiling point processing additives. The best results are obtained with R ) Iodine (I). Using
1,8-diiodooctane as the processing additive, the efficiency of the BHJ solar cells was improved from 3.4%
(for the reference device) to 5.1%.
Polymer solar cells based on bulk heterojunction (BHJ)
materials comprising π-conjugated (semiconducting) polymers
and fullerene derivatives have demonstrated promising per-
formance.1-9In order to understand the BHJ materials and to
optimize the efficiency of BHJ photovoltaic cells, the nano-
structure has been studied using transmission electron micros-
copy (TEM), X-ray diffraction (XRD), nuclear magnetic
resonance (NMR), and atomic force microscopy (AFM); the
achievement of high efficiency BHJ polymer solar cells requires
the existence of interpenetrating channel-like domains which
separate the polymer and fullerene phases.7-10,12-16
After annealing at elevated temperatures, relatively high
efficiency solar cells have been demonstrated using regioregular
poly(3-hexylthiophene) (rrP3HT) with [6,6]-phenyl-C61-butyric
acid methyl ester (C61-PCBM) and poly[2-methoxy-5-(3,7-
dimethyloctyloxy)]-1,4-phenylene-vinylene (MDMO-PPV) with
C61-PCBM; the annealed BHJ films fabricated with these
materials show well-defined bicontinuous interpenetrating net-
works. The importance of the solvent in determining the BHJ
morphology has also been established, with finer phase separa-
tion in films cast from chlorobenzene than in films cast from
The addition of alkanedithiols into the solvent was recently
shown to improve the performance of BHJ solar cells.18By
incorporating a few volume percent of alkanedithiol into the
chlorobenzene from which BHJ films comprising the low band
gap polymer [2,6-(4,4-bis(2-ethylhexyl)-4H-cyclopenta[2,1-
DTBT) and C71-PCBM are cast, the power conversion efficiency
(AM 1.5G conditions) was increased from 2.8% to 5.5%.17,18
Although evidence of the formation of phase separated BHJ
materials was clearly observed, there was no alkanedithiol in
thoroughly dried films. Thus, although the alkanedithiols
functioned as “processing additives” for improving the morphol-
ogy of the BHJ material, the mechanism by which this occurred
Herein, we clarify the mechanism by which processing
additives control the morphology, and using the insight provided
†University of California at Santa Barbara.
‡Konarka Technologies Austria.
(1) Sacriciftci, N. S.; Smilowitz, L.; Heeger, A. J.; Wudl, F. Science 1992,
(2) Yu, G.; Gao, J.; Hummelen, J. C.; Wudl, F.; Heeger, A. J. Science 1995,
(3) Shaheen, S. E.; Brabec, C. J.; Sariciftci, N. S.; Padinger, F.; Fromherz, T.;
Hummelen, J. C. Appl. Phys. Lett. 2001, 78, 841-843.
(4) Padinger, F.; Rittberger, R. S.; Sariciftci, N. S. AdV. Funct. Mater. 2003,
(5) Al-lbrahim, M.; Ambacher, O.; Sensfuss, S.; Gobsch, G. Appl. Phys. Lett.
2005, 86, 201120-3.
(6) Kim, Y.; Choulis, S. A.; Nelson, J.; Bradley, D. D. C.; Cook, S.; Durrant,
J. R. Appl. Phys. Lett. 2005, 86, 063502-3.
(7) Hoppe, H.; Saricitfci, N. S. J. Mater. Chem. 2006, 16, 45-61.
(8) Savenije, T. J.; Kroeze, J. E.; Yang, X.; Loos, J. AdV. Funct. Mater. 2005,
(9) Yang, C.; Hu, J. G.; Heeger, A. J. J. Am. Soc. Chem. 2006, 128, 12007-
(10) Ma, W.; Yang, C.; Gong, X.; Lee, K.; Heeger, A. J. AdV. Funct. Mater.
2005, 15, 1617-1622.
(11) Hoppe, M.; Niggemann, M.; Winder, C.; Kraut, J.; Hiesgen, R.; Hinch,
A.; Meissner, D.; Sariciftci, N. S. AdV. Funct. Mater. 2004, 14, 1005-
(12) Kline, R. J.; Mcgehee, M. D.; Toney, M. F. Nat. Mater. 2006, 5, 222-
(13) Sivula, K.; Ball, Z. T.; Watanabe, N.; Frechet, J. M. J. AdV. Mater. 2006,
(14) Reyes-Reyes, M.; Kim, K.; Dewald, J.; Lopez-Sandoval, R.; Avadhanula,
A.; Curran, S.; Carroll, D. Org. Lett. 2005, 7, 5749-5752.
(15) Zhokhavets, U.; Erb, T.; Hoppe, H.; Gobsch, G.; Sariciftci, N. S. Thin Solid
Film 2006, 496, 679-682.
(16) Camaioni, N.; Ridolfi, G.; Casalbore-Miceli, G.; Possamai, G.; Maggini,
M. AdV. Mater. 2002, 14, 1735-1738.
(17) Peet, J.; Soci, C.; Coffin, R. C.; Nguyen, T. Q.; Mikhailovsky, A.; Moses,
D.; Bazan, G. C. Appl. Phys. Lett. 2006, 89, 252105-252107.
(18) 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.
Published on Web 02/21/2008
10.1021/ja710079w CCC: $40.75 © 2008 American Chemical Society
J. AM. CHEM. SOC. 2008, 130, 3619-3623 9 3619
by the mechanism, we identify an entire class of processing
additives which yields still higher efficiency solar cells than
those fabricated with 1,8-octanedithiol as the processing
The alkanedithiols do not react with either the polymer or
the fullerene; they function as processing additives. As indicated
by Fourier transform infrared (FTIR) spectroscopy, Raman
spectroscopy, and X-ray photoelectron spectroscopy (XPS), the
alkanedithiol is removed while drying the film under high
vacuum.18We find that the alkanedithiol selectively dissolves
the PCBM; C61-PCBM and C71-PCBM are readily dissolved in
alkanedithiol, but P3HT and PCPDTBT are not soluble in
alkanedithiol. Because the fullerenes are selectively dissolved
in alkanedithiol, three separate phases are formed during the
process of liquid-liquid phase separation and drying:
fullerene-alkanedithiol phase, a polymer aggregate phase, and
a polymer-fullerene phase. In addition, because the alkanedithi-
ol has a higher boiling point than the chlorobenzene host solvent,
the PCBM tends to remain in solution (during drying) longer
than the semiconducting polymer, thereby enabling control of
the phase separation and the resulting morphology of the BHJ
material; see Scheme 1.
Figure 1 shows the UV-visible absorption spectrum of
PCPDTBT/C71-PCBM BHJ films spin-cast from a blend solution
containing 2.5% 1,8-octanedithiol in chlorobenzene (black) and
the UV-vis spectrum of a PCPDTBT network film which has
been soaked in alkanedithiol to remove the C71-PCBM (red).
The absorption spectrum obtained from a pristine PCPDTBT
film is shown for comparison (green). The data confirm the
selective removal of C71-PCBM; the absorption features of C71-
PCBM are not detected in the BHJ film after soaking in
alkanedithiol. We note that the red-shifted π-π* absorption
band with vibronic side bands of BHJ films processed with 1,8-
octanedithiol is not changed after removal of C71-PCBM,
indicating that the improved local structure within the PCPDTBT
is maintained even after removal of the PCBM. The clear
observation of the morphology of the bare polymer network
confirms that the selective solubility of the processing additive;
the implied control of the phase separation affects both the
polymer and the fullerene networks.
Thus, two criteria for processing additives for use in the
fabrication of BHJ solar cells have been identified: (i) selective
(differential) solubility of the fullerene component of the BHJ
material and (ii) higher boiling point than the host solvent. Using
these criteria, we have investigated the class of 1,8-di(R)octanes
with various functional groups (R) as processing additives for
BHJ solar cells and obtained the best results with 1,8-
diiodooctane. The 1,8-dibromooctane functions almost as well;
both 1,8-diiodooctane and 1,8-dibromooctane give better results
aSchematic depiction of the role of the processing additive in the self-assembly of bulk heterojunction blend materials (a) and structures of PCPDTBT,
C71-PCBM, and additives (b).
Figure 1. UV-visible absorption spectra of PCPDTBT/C71-PCBM films
processed with 1,8-octanedithiol: before removal of C71-PCBM with
alkanedithiol (black); after removal of C71-PCBM with alkanedithiol
(red) compared to the absorption spectrum of pristine PCPDTBT film
A R T I C L E S
Lee et al.
3620 J. AM. CHEM. SOC.9VOL. 130, NO. 11, 2008
Results and Discussion
Figure 2 shows the UV-visible absorption spectra of
PCPDTBT/C71-PCBM films: (a) cast from pure cholorobenzene
(black curve), (b) cast from cholorobenzene containing 2.5%
1,8-octanedithiol, (c) cast from cholorobenzene containing 2.5%
1,8-diiodooctane, and (d) cast from cholorobenzene containing
2.5% 1,8-dibromooctane. Note that these films were spin-cast
from solution under identical conditions with the same spin
speed (2000 rpm); the absorption spectra are not normalized.
As shown in Figure 2, the peak in the absorption band of
PCPDTBM in the PCPDTBT/C71-PCBM composite deposited
from solution with the processing additives is red-shifted by
41 nm (to 800 nm) compared with that from film cast from
solution without any processing additive. The π-π* absorption
band intensities of the BHJ films processed using the 1,8-
dibromooctane and 1,8-octanedithiol processing additives are
nearly the same, but the band intensity of the BHJ film processed
using 1,8-diiodooctane is higher than that of the films cast using
the other additives.
We find that photovoltaic cells based on PCPDTBT/C71-
PCBM processed with 1,8-diiodooctane as the processing
additive exhibited the highest short circuit current (Jsc), the
highest fill factor (FF), and an efficiency that is approximately
10% higher than that obtained using 1,8-octanedithiol.18Figure
3 shows the current (J)-voltage (V) curves under AM 1.5
conditions (100 mW per cm2) of PCPDTBT/C71-PCBM BHJ
solar cells made using the various processing additives. The
data obtained from Figure 3 are summarized in Table 1.
Figure 2. UV-vis spectra of (a) PCPDTBT/C71-PCBM films cast from
cholorobenzene and PCPDTBT/C71-PCBM films cast from cholorobenzene
containing 2.5% of the following: (b) 1,8-octanedithiol (red), (c) 1,8-
dibromooctane (green), and (d) 1,8-diiodooctane (blue).
Figure 3. Current (J)-voltage (V) characteristics of PCPDTBT/C71-PCBM
composite films with various additives:
octanedithiol (red), (c) 1,8-dicholorooctane (green), (d) 1,8-dibromooctane
(blue), (e) 1,8-diiodooctane (cyan), (f) 1,8-dicyanooctane (magenta), and
(g) 1,8-octanediacetate (yellow).
(a) none (black), (b) 1,8-
Table 1. Photovoltaic Performances of the BHJ Polymer Solar
Cells Composed PCPDTBT/C71-PCBM Fabricated with Various
Voc (V)fill factor efficiency (%)
aThe optimum devices were fabricated with PCPDTBT/C71-PCBM
(1:3.6) films which were spin-cast at 2.000 rpm from pristine chlorobenzene
and chlorobenzene containing 2.5 vol % of additives. The performances
are determined under simulated 100 mW/cm2AM 1.5G illumination. For
light intensity, calibrated standard silicon solar cells with a proactive window
made from KG5 filter glass traced to the National Renewable Energy
Laboratory (NREL) were used. The active area of the device is 4.5 mm2.
Figure 4. AFM topography of films cast from PCPCTBT/C71-PCBM with
additives: (a) 1,8-octanedithiol, (b) 1,8-cicholorooctane, (c) 1,8-dibromooc-
tane, (d) 1,8-diiodooctane, (e) 1,8-dicyanooctane, and (f) 1,8-octanediacetate.
Improved Efficiency from Bulk Heterojunction Solar Cells
A R T I C L E S
J. AM. CHEM. SOC. 9 VOL. 130, NO. 11, 2008 3621
Performance optimization involved over 2000 devices made
from over 400 independently prepared PCPDTBT/C71-PCBM
devices. The most efficient photovoltaic cells were fabricated
using a polymer/fullerene ratio of approximately 1:3.5, a spin
speed between 2000 and 2500 rpm, a polymer concentration in
solution between 0.7 and 1.0 wt %, and a processing additive
concentration between 2.5 and 3.0 vol % in chlorobenzene. The
optimum thickness of the BHJ films obtained under these
conditions was approximately 200 nm. As shown in Figure 2
and Table 1, the devices made using the 1,8-octanedithiol, 1,8-
dibromooctane, and 1,8-diiodooctane exhibited enhanced Jsc
compared to films made without any processing additive, but
the devices made with 1,8-dichlorooctane, 1,8-dicyanooctane,
and 1,8-octanediacetate exhibited reduced Jsc. All the processing
additives (except for 1,8-dichlorooctane) caused the fill factor
to increase. The most efficient devices, obtained using 1,8-
diiodooctane, had an average power-conversion efficiency of
5.1% under 100 mW/cm2, with short-circuit current Jsc ) 15.7
mA/cm2, fill factor FF ) 0.53, and open-circuit voltage Voc )
0.61 V, i.e., an ∼10% higher efficiency than that obtained with
the use of 1,8-octanedithiol. Note that because of batch-to-batch
variation in the quality of the PCPDTBT, the efficiency obtained
using 1,8-octanedithiol was only 4.5% compared with the 5.5%
reported earlier.18Thus, solar cells processed with 1,8-diio-
dooctane are expected to be capable of efficiencies above 6%.
Figure 4 shows the surface topography, measured by AFM,
of films cast from PCPCTBT/C71-PCBM with the various
processing additives. The 1,8-octanedithiol (a), 1,8-dibromooc-
tane (c), and 1,8-diiodooctane (d) gave phase-segregated mor-
phologies having finer domain sizes than those obtained with
1,8-dichlorooctane (b), 1,8-dicyanooctane (e), and 1,8-octanedi-
acetate (f). The morphology of films processed with 1,8-
diiodooctane showed more elongated domains than those
processed with 1,8-octanedithiol and 1,8-dibromooctane. The
1,8-di(R)octanes with SH, Br, and I, which gave finer domain
sizes, exhibited more efficient device performances than those
with R ) Cl, CN, and CO2CH3. The AFM images of the BHJ
films processed using 1,8-di(R)octanes with R ) Cl, CN, and
CO2CH3showed large scale phase separation with round-shape
domains and no indication of bicontinuous networks.
TEM images are shown in Figure 5. The morphology of BHJ
film processed with 1,8-octanedithiol (b) and 1,8-diiodooctane
(c) showed larger scale phase separation than that in the BHJ
film made without using any processing additive (a). Whereas
fibril-like domains are observed over the entire area of the TEM
images of the BHJ films processed with 1,8-diiodooctane, these
fibril-like domains are observed only in some areas of the BHJ
film processed with 1,8-octanedithiol.
We have characterized the polymer networks after selectively
dissolving out the fullerene from BHJ composite films using
alkanedithiol. With this approach, we can directly probe the
exposed polymer networks (rather than indirectly infer the
regions of PCPDTBT and C71-PCBM in the BHJ films using
conventional AFM and TEM tools) resulting in the clear
interpretation of improved charge transport and enhanced
performance by processing with the additive. Figure 6 shows
AFM images of BHJ films cast from PCPCTBT/C71-PCBM
without and with 1,8-octanedithiol and AFM and TEM images
of exposed PCPDTBT networks after selective removal of C71-
PCBM from the BHJ film. AFM images of the BHJ films
Figure 5. TEM image of films cast from PCPCTBT/C71-PCBM with
additives: (a) none, (b) 1,8-octanedithiol, and (c) 1,8-diiodooctane.
Figure 6. AFM and TEM images of BHJ films cast from PCPCTBT/C71-
PCBM without and with 1,8-octanedithiol and exposed PCPDTBT networks
after removal of C71-PCBM in BHJ film; AFM image of BHJ film (a)
without and (b) with 1,8-octanedithiol; AFM image of exposed polymer
networks (c) without and (d) with 1,8-octanedithiol; and TEM image of
exposed polymer networks (c) without and (d) with 1,8-octanedithiol.
A R T I C L E S
Lee et al.
3622 J. AM. CHEM. SOC.9VOL. 130, NO. 11, 2008
processed with the additive exhibit larger interconnected regions
of PCPDTBT and larger porous domains (the C71-PCBM regions
prior to selective removal) compared with images of the BHJ
film cast without using the processing additive. The TEM
images are consistent with the AFM results. The polymer
networks and porous regions are imaged in the TEM as darker
and lighter colored, respectively. The sharp, high contrast TEM
images reveal that while both films contain well-connected
PCPDTBT networks, the BHJ film initially cast from solution
containing the processing additive showed larger areas of
connectivity with larger connective cross sections. Since the
short circuit current and fill factor are strongly dependent on
the transport properties of the networks in the BHJ film, it is
clear that, at least in part, the improved device performance
results from improved hole transport as a result of the formation
of the larger domains with wider connective cross sections
within the PCPDTBT network.
AFM and TEM images are shown in Figures 4 and 5. After
selective removal of C71-PCBM from the BHJ film processed
with 1,8-diiodooctane, the exposed PCPDTBT network exhibits
a finer porous and fibril-like structure. After selective removal
of C71-PCBM from the BHJ film processed with 1,8-dichlo-
rooctane, the exposed PCPDTBT network exhibits larger hole-
like “craters” than those observed from 1,8-octanedithiol (see
also Figure S3 in the Supporting Information). From these
results, it becomes clear that the exposed PCPDTBT networks
are closely related to the morphology of BHJ films processed
with different additives. These data provide a better understand-
ing of correlation between the polymer morphology and solar
We have demonstrated the utility of a class of processing
additives introduced to control the morphology of BHJ materials
for use in solar cells. Two criteria for processing additives for
use in the fabrication of BHJ solar cells have been identified:
(i) selective (differential) solubility of the fullerene component
and (ii) higher boiling point than the host solvent. The control
of the BHJ morphology by selective solubility of the fullerene
component in the BHJ blend was demonstrated. Using these
criteria, we have investigated the class of 1,8-di(R)octanes with
various functional groups (R) as processing additives for BHJ
solar cells and obtained the best results with R ) I or Br. Using
1,8-di-iodooctane as the processing additive, the efficiency of
the BHJ solar cells was improved from 3.4% (for the reference
device) to 5.1%.
The PCPDTBT and the soluble fullerene derivatives were obtained
from Konarka Technologies, Inc. BHJ films were prepared under
optimized conditions according to the following procedure reported
previously: The indium tin oxide (ITO)-coated glass substrate was first
cleaned with detergent, ultrasonicated in acetone and isopropyl alcohol,
and subsequently dried overnight in an oven. Poly(3,4-ethylene-
dioxythiophene)/poly(styrensulfonate) PEDOT/PSS (Baytron P) was
spin-cast from aqueous solution to form a film of thickness of ∼40
nm. The substrate was dried for 10 min at 140 °C in air and then
transferred into a glovebox to spin-cast the charge separation layer. A
solution containing a mixture of PCPDTBT/C71-PCBM (1:3.6) in
chlorobezene with or without 2.5 vol % additives was then spin-cast
on top of the PEDOT/PSS layer. A PCPDTBT network film was
prepared by rinsing the PCPDTBT/C71-PCBM BHJ film with al-
kanedithiol for 5 s. The AFM images were obtained with a Dimension
3100 atomic force microscope. TEM specimens were prepared by
detaching a BHJ film from the substrate onto the surface of deionized
water and picking it up with a copper grid. TEM images were obtained
with a FEI T20 transmission electron microscope operated at 200 kV
under proper defocus conditions.
Acknowledgment. We thank Konarka Technologies Inc. for
providing the PCPDTBT and fullerene derivatives. The research
was supported by Konarka Technologies and the Air Force
Office of Scientific Research (Charles Lee, Program Officer).
J.K.L. was partially supported by a fellowship (KRF-2006-352-
Supporting Information Available: UV-vis data, postan-
nealing effects, AMF images, and boiling point of additives.
This material is available free of charge via the Internet at
Improved Efficiency from Bulk Heterojunction Solar Cells
A R T I C L E S
J. AM. CHEM. SOC. 9 VOL. 130, NO. 11, 2008 3623