Photocurrent enhancement from diketopyrrolopyrrole polymer solar cells through alkyl-chain branching point manipulation

Article abstract of DOI:10.1021/ja406934j

Systematically moving the alkyl-chain branching position away from the polymer backbone afforded two new thieno[3,2-b]thiophene-diketopyrrolopyrrole (DPPTT-T) polymers. When used as donor materials in polymer:fullerene solar cells, efficiencies exceeding 7% were achieved without the use of processing additives. The effect of the position of the alkyl-chain branching point on the thin-film morphology was investigated using X-ray scattering techniques and the effects on the photovoltaic and charge-transport properties were also studied. For both solar cell and transistor devices, moving the branching point further from the backbone was beneficial. This is the first time that this effect has been shown to improve solar cell performance. Strong evidence is presented for changes in microstructure across the series, which is most likely the cause for the photocurrent enhancement.

Full text of DOI:10.1021/ja406934j

Communication
pubs.acs.org/JACS
Photocurrent Enhancement from Diketopyrrolopyrrole Polymer
Solar Cells through Alkyl-Chain Branching Point Manipulation
Iain Meager,*,† Raja Shahid Ashraf,† Sonya Mollinger,§ Bob C. Schroeder,† Hugo Bronstein,∥
Daniel Beatrup,† Michelle S. Vezie,‡ Thomas Kirchartz,‡ Alberto Salleo,§ Jenny Nelson,‡
and Iain McCulloch†
†Department of Chemistry and Centre for Plastic Electronics, Imperial College London, London SW7 2AZ, U.K.
‡Department of Physics and Centre for Plastic Electronics, Imperial College London, London SW7 2AZ, U.K.
§Department of Materials Science and Engineering, Stanford University, Stanford, California 94305, United States
∥Department of Chemistry, University College London, London WC1H 0AJT, U.K.
S
* Supporting Information
demonstrating their importance.7−13 Differences between
ABSTRACT: Systematically moving the alkyl-chain
branching position away from the polymer backbone
afforded two new thieno[3,2-b]thiophene−diketopyrrolo-
pyrrole (DPPTT-T) polymers. When used as donor
materials in polymer:fullerene solar cells, efficiencies
exceeding 7% were achieved without the use of processing
additives. The effect of the position of the alkyl-chain
branching point on the thin-film morphology was
investigated using X-ray scattering techniques and the
effects on the photovoltaic and charge-transport properties
were also studied. For both solar cell and transistor
devices, moving the branching point further from the
backbone was beneficial. This is the first time that this
effect has been shown to improve solar cell performance.
Strong evidence is presented for changes in microstructure
across the series, which is most likely the cause for the
photocurrent enhancement.
branched and linear alkyl chains and the resultant influence on
the photovoltaic performance have been demonstrated.14−17
Further studies have found that the specific effect of the alkyl-
chain branching position can be critical, with related work on
transistor devices showing improved charge carrier mobili-
ties.18,19 However, the effect of this on solar cell performance has
yet to be examined. We previously demonstrated the promising
performance of DPPTT-T in solar cell and OFET devices. The
effect of the side-chain branching point on the performance and
processability of solar cells made with this polymer is therefore of
interest.20 By moving the branching point from its regular
position relatively close to the polymer backbone, as in the
original DPPTT-T polymer (C1), to positions further from the
backbone, we aimed to determine whether the π−π stacking
distance could be influenced and to evaluate the effect that this
would have on the crystallinity, morphology, and photovoltaic
properties.
The three DPPTT-T polymers C1, C2, and C3 (Figure 1),
where Cn refers to the number of linear carbon atoms between
the alkylated nitrogen and the alkyl-chain branching point, were
synthesized by copolymerization of their respective monomeric
units with thiophene. The monomers were synthesized by
alkylation of the DPPTT core with the corresponding alkyl
iodide. All of the alkyl chains were synthesized from
he desirable optical and electrical properties exhibited by
Tconjugated semiconducting polymers make them promis-
ing candidate materials for use in the next generation of solar cells
and organic field-effect transistors (OFETs). Their solution
processability provides the potential for large-scale inexpensive
manufacture, and their lightweight and flexible nature leads to a
large number of potential applications currently beyond the
scope of conventional inorganic cells. To date, one of the most
promising classes of polymer materials are diketopyrrolopyrrole
(DPP)-based polymers, which have shown some of the highest
solar cell efficiencies reported to date and have been used to
fabricate transistor devices with good ambipolarity and mobilities
that routinely surpass 1 cm2 V−1 s−1.1−6 The DPP unit has an
electron-withdrawing core, which is typically flanked by an
electron-rich unit. Copolymerization with an electron-rich
comonomer can afford a range of extended push−pull-type
conjugated polymers. The backbone planarity promotes strong
interchain π−π interactions that in turn can facilitate good charge
transport. In the design of semiconducting polymers, a large
amount of effort is focused on band-gap engineering via control
over the backbone electron distribution. Thus, the effect of the
solubilizing alkyl chains has been less studied, despite literature
Figure 1. Chemical structures of C1−C3.
Received: July 8, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja406934j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Table 1. Properties of C1−C3
λmax (nm)
EHOMO (eV)
ELUMO (eV)
f
a
a
a
b
c
d
e
d
e
band gap (eV)
polymer
Mn (kDa)
Mw (kDa)
PDI
film
solution
PESA
CV
PESA
CV
C1
C2
C3
24
45
80
89
83
3.7
1.8
1.9
790
810
803
764
812
804
−5.1
−5.1
−5.1
−5.1
−5.1
−5.1
−3.7
−3.7
−3.7
−3.6
−3.4
−3.5
1.4
1.4
1.4
154
a
Number-average and weight-average molecular weights (Mn and Mw, respectively) and weight-average polydispersity indexes (PDI = Mw/Mn)
b
determined by gel-permeation chromatography using low-PDI (<1.10) polystyrene standards and chlorobenzene as the eluent. Spin-coated from 5
mg/mL chlorobenzene solution. Measured in dilute chlorobenzene solution. HOMO energies (EHOMO) measured by PESA and LUMO energies
(ELUMO) estimated by adding the UV−vis absorption onset to EHOMO. Measured by cyclic voltammetry Estimated as the difference between the
experimentally determined EHOMO and the optically estimated ELUMO
c
d
e
f
.
commercially available 2-octyl-1-dodecanol. To move the
branching position two carbons from the polymer backbone in
C2, the alkyl chain was obtained by Grignard reaction of the
corresponding alkyl bromide with paraformaldehyde followed by
iodination to afford 3-octyl-1-tridecyl iodide. The alkyl chain
with the branching point three carbons from the backbone in C3
was synthesized by formation of the corresponding malonic ester
from the alkyl bromide, hydrolysis to give the acid, and
subsequent reduction to the alcohol followed by iodination to
afford 4-octyl-1-tetradecyl iodide. C1−C3 were synthesized
under Pd-catalyzed Stille coupling conditions using microwave
irradiation. Full synthetic details are included in the Supporting
Information (SI). After polymerization, the polymers were
precipitated into methanol and purified by Soxhlet extraction in
acetone and hexane (24 h each) to remove catalytic impurities
and lower-molecular-weight oligomers. The purified polymers
were removed from the Soxhlet thimble with chloroform and
finally chlorobenzene. Of the polymers, C3 with the largest alkyl
chain was the most soluble, allowing us to achieve a very high
number-average molecular weight (Mn) of 80 kDa, whereas in
the case of C2 not all of the material could be solubilized in
chlorobenzene, thus leading to a slightly lower Mn of 45 kDa. The
solubility of C3 was noticeably improved relative to that of the
original polymer C1, with all of the polymeric material dissolving
in chloroform. Both C2 and C3 had improved solubility relative
to C1, which is reflected in their higher molecular weights and
narrower polydispersities (Table 1).
which allows stronger intermolecular π−π stacking of the
aromatic rings, ultimately leading to enhanced molecular
aggregation in solution and similar absorption features in
solution and the solid state. For C1, however, the proximity of
the alkyl-chain branching point to the DPP core possibly hinders
the π−π stacking and aggregation in solution; this is overcome by
intermolecular forces in the solid state, planarizing the backbone
and thus causing a significant red shift in the solid-state UV−vis
absorption bands relative to the solution spectrum. The frontier
molecular orbital energies (EHOMO and ELUMO) were determined
using photoelectron spectroscopy in air (PESA) and the
absorption onsets from the UV−vis spectra and were further
investigated by cyclic voltammetry (Table 1); it appears that
manipulating the alkyl-chain branching point has little effect on
E
HOMO and ELUMO
Bulk heterojunction (BHJ) solar cells were fabricated using
.
C1, C2, or C3 as the donor material in the active layer by spin-
coating of a 1:2 polymer/PC71BM mixture from a 4:1
chloroform:o-dichlorobenzene solvent mixture. Cells were
prepared and tested under simulated 100 mW/cm AM1.5G
illumination. Device fabrication details are given in the SI. The
current density−voltage (J−V) curves averaged over the pixels of
the best-performing substrate and representative external
quantum efficiency (EQE) spectra are shown in Figure 3. The
open-circuit voltage (Voc), short-circuit current (Jsc), fill factor
(FF), and power conversion efficiency (PCE) of the best
performing pixel of each type are summarized in Table 2. Both
Figure 3. (left) J−V curves for C1−C3 averaged over the pixels on the
best-performing substrates and (right) EQE spectra for typical blend
devices using C1−C3.
Figure 2. UV−vis absorption spectra of C1−C3 in (left) chlorobenzene
solution and (right) thin films spun from chlorobenzene.
UV−vis absorption spectra of C1−C3 are shown in Figure 2.
The absorption maxima of both C2 and C3 are red-shifted in
solution and thin films relative to those of C1. C2 and C3 show
narrow absorption profiles with observable shoulders at lower
wavelengths, whereas the absorption band of C1 shows a far less
pronounced vibronic structure. Interestingly, both of the longer-
alkyl-chain polymers show negligible red shifts in going from
solution to the solid state, in contrast to C1, which shows a
significant red shift of ∼30 nm. In the case of C2 and C3, the
alkyl-chain branching point is further away from the DPP core,
Table 2. Solar Cell Device Characterstics of C1−C3 Taken
from the Best-Performing Pixel of Each Type
polymer
Jsc (mA/cm2)
Voc (V)
FF
PCE (%)
C1
C2
C3
16.6
18.6
18.7
0.59
0.61
0.60
0.60
0.64
0.62
5.9
7.3
6.9
B
dx.doi.org/10.1021/ja406934j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Table 3. OFET Properties of C1−C3 in Devices with a
Bottom-Gate Top-Contact Architecture
a
b
c
c
polymer
dπ−π (Å−1
)
μhole (cm2 V−1 s−1
)
Vth (V)
Ion/Ioff
C1
C2
C3
1.75
1.78
1.78
0.014
0.052
0.066
−20
−17
−15
∼5 × 102
∼1 × 103
∼1 × 103
b
a
Polymer π−π stacking distances as measured by 2D GIXS. Highest
c
effective hole mobilities measured in the saturation regime. Threshold
voltages (Vth) and on/off ratios (Ion/Ioff) extracted from the linear
regime (Vd = −5 V).
C2 and C3 show Voc values comparable to that of C1, which is in
good agreement with the similar EHOMO values measured by
PESA and validates the idea that the improved photocurrent
arising from increased branching point distance is not a result of
changes in the polymer energy levels. The most dramatic
improvement is seen in the Jsc of C2 and C3 (18.6 and 18.7 mA/
cm2 respectively). The very high Jsc values together with good FF
and Voc values resulted in PCEs of 7.3% for C2 and 6.9% for C3,
up from 5.9% for C1. The EQE spectra confirmed the higher
photocurrent generation efficiencies for the devices made with
C2 and C3 relative to C1. Bottom-gate top-contact OFETs with
gold source and drain electrodes were fabricated to directly
compare the trend in the hole transport properties of C1−C3
(Table 3); further details and representative transfer and output
Figure 5. (a−c) 2D GIXS patterns of Cn/PC[71]BM blend films using
(a) C1, (b) C2, and (c) C3. (d, e) Line cuts in the (d) qz and (e) Qxy
directions.
scattering vector parallel to the substrate plane is given by Qxy,
and qz is the component perpendicular to the substrate plane. In
general, shifting the branching point away from the backbone
seemed to increase the crystallinity of the cast films of C2 and C3
relative to C1. Furthermore, in the neat polymer films, the (200)
peak shifts to slightly lower q with increasing branching point
distance: for C1−C3, the (200) peak is located at qz = 0.62, 0.60,
and 0.59 Å−1, respectively (Figure 4a−c; Figure 4d displays line
cuts showing relative peak distances). The location of the π-
stacking peak is also weakly dependent on the polymer and is
located at 1.78 Å−1 for C2 and C3 and 1.75 Å−1 for C1 (Figure
4e). However, the texture of the neat polymer film is strongly
dependent on the location of the branching point. It is interesting
to note that C1 exhibits a pronounced face-on character, while
both C2 and C3 have significant edge-on character. Thus, the
more distant branching points in C2 and C3 seem to encourage
an edge-on morphology in the neat polymer films. Similar
variations have previously been noticed with DPP polymers, and
more detailed studies of these differences are ongoing.21 In the
blend films (Figure 5), the GIXS patterns contain peaks
associated with the pure polymers in addition to a diffraction
halo peaked near 1.33 Å−1 associated with the PC[71]BM
amorphous phase. The diffraction patterns of the blends also
show a set of alkyl-stacking peaks, located at 0.65 Å−1 for C1, 0.62
Å−1 for C2, and 0.63 Å−1 for C3 (Figure 5d). These are shifted
slightly toward higher q from their positions in the pure films,
indicating that in the blends the lamellar spacing is narrower than
in the pure films. The blend films also show a diffraction feature
located near 1.79 Å−1, roughly the same location as the π-stacking
peak found in the pure films. For C2 (Figure 5b) the widths of
this peak in the qxy direction are similar [0.08 Å−1 for the neat film
(Figure 4e) and 0.11 Å−1 for the blend (Figure 5e)]; for C3 they
are 0.14 Å−1 in the neat film and 0.16 Å−1 in the blend. Thus, the
coherence length in the π-stacking direction is not reduced by
blending. In the blend with C2 in particular, the polymer assumes
a nearly isotropic texture, which is more favorable for out-of-
plane transport in solar cells. Analysis of the (200) lamellar peak
intensities in the qz direction for the polymer blend films with
PC[71]BM shows a substantial increase in intensity upon going
from the C1 blend to the C2 blend. Though a truly quantitative
comparison cannot be made for different materials, this
qualitative comparison of the intensities suggests that at a larger
branching point distance, the crystallinity of the film increases in
curves are included in the SI. The recorded hole mobilities (μhole
)
were significantly improved, showing a trend similar to the
improvements observed in the solar cell properties with
increasing distance between the nitrogen and the branching
position in C2 and C3. It should be noted that the mobilities
recorded for these polymers are noticeably lower than those
previously published for C1 devices. The initial testing for the
present series of polymers was performed with a bottom-gate
top-contact device architecture, whereas the previous high-
performance transistor mobilities for C1 were observed with a
top-gate bottom-contact architecture. Although the μhole values
are lower than those reported previously for C1, the promising
trend in μhole with branching point position indicates that higher
mobilities will be realized for polymers C2 and C3 in optimized
FET device structures.
We used 2D grazing-incidence X-ray scattering (GIXS) to
study films of the pure polymers (Figure 4) and 1:2 (w/w) Cn/
PC[71]BM blend films (Figure 5). The component of the
Figure 4. (a−c) 2D GIXS patterns of (a) C1, (b) C2, and (c) C3 films.
(d, e) Line cuts in the (d) qz and (e) Qxy directions.
C
dx.doi.org/10.1021/ja406934j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(14) Bronstein, H.; Leem, D. S.; Hamilton, R.; Woebkenberg, P.; King,
S.; Zhang, W.; Ashraf, R. S.; Heeney, M.; Anthopoulos, T. D.; de Mello,
J.; McCulloch, I. Macromolecules 2011, 44, 6649.
the blends similarly to what was observed in the neat polymer
films. The same trend is preserved for the π-stacking peak.
In summary, we have prepared two new DPPTT-T based
copolymers by systematically moving the alkyl-chain branching
point. These polymers exhibit high solar cell performance, giving
efficiencies in excess of 7% without the use of processing
additives. X-ray scattering analysis showed that the location of
the branching point can influence the orientation of the
conjugated backbone plane and that the improvement in solar
cell and OFET performance upon moving the branching point
further from the backbone is most likely a result of improved
crystallinity of the polymer in the neat and blend films.
(15) Piliego, C.; Holcombe, T. W.; Douglas, J. D.; Woo, C. H.;
Beaujuge, P. M.; Frec
(16) Yiu, A. T.; Beaujuge, P. M.; Lee, O. P.; Woo, C. H.; Toney, M. F.;
Frechet, J. M. J. J. Am. Chem. Soc. 2012, 134, 2180.
́
het, J. M. J. J. Am. Chem. Soc. 2010, 132, 7595.
́
(17) Biniek, L.; Fall, S.; Chochos, C. L.; Anokhin, D. V.; Ivanov, D. A.;
Leclerc, N.; Leveque, P.; Heiser, T. Macromolecules 2010, 43, 9779.
(18) Zhang, F.; Hu, Y.; Schuettfort, T.; Di, C.; Gao, X.; McNeill, C. R.;
Thomsen, L.; Mannsfeld, S. C. B.; Yuan, W.; Sirringhaus, H.; Zhu, D. J.
Am. Chem. Soc. 2013, 135, 2338.
(19) Lei, T.; Dou, J. H.; Pei, J. Adv. Mater. 2012, 24, 6457.
(20) Bronstein, H.; Chen, Z.; Ashraf, R. S.; Zhang, W.; Du, J.; Durrant,
J. R.; Tuladhar, P. S.; Song, K.; Watkins, S. E.; Geerts, Y.; Wienk, M. M.;
Janssen, R. A. J.; Anthopoulos, T. D.; Sirringhaus, H.; Heeney, M.;
McCulloch, I. J. Am. Chem. Soc. 2011, 133, 3272.
(21) Zhang, X.; Richter, L. J.; DeLongchamp, D. M.; Kline, R. J.;
Hammond, M. R.; McCulloch, I.; Heeney, M.; Ashraf, R. S.; Smith, J. N.;
Anthopoulos, T. D.; Schroeder, B.; Geerts, Y. H.; Fischer, D. A.; Toney,
M. F. J. Am. Chem. Soc. 2011, 133, 15073.
ASSOCIATED CONTENT
■
S
* Supporting Information
Synthesis of monomers and copolymers, solar cell and OFET
characteristics and fabrication, GIXD measurements, and cyclic
voltamograms. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
i.meager11@imperial.ac.uk
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was carried out primarily with funding and support
from The Leventis Foundation, the National Research Fund of
Luxembourg, and the X10D Project (EP 287818). H.B. and T.K.
acknowledge support via Imperial College Junior Research
Fellowships. M.S.V. and J.N. acknowledge support from the
Engineering and Physical Sciences Research Council.
REFERENCES
■
(1) Nielsen, C. B.; Turbiez, M.; McCulloch, I. Adv. Mater. 2013, 25,
1859.
(2) Qu, S.; Tian, H. Chem. Commun. 2012, 48, 3039.
(3) Kronemeijer, A. J.; Gili, E.; Shahid, M.; Rivnay, J.; Salleo, A.;
Heeney, M.; Sirringhaus, H. Adv. Mater. 2012, 24, 1558.
(4) Dou, L.; Gao, J.; Richard, E.; You, J.; Chen, C.-C.; Cha, K. C.; He,
Y.; Li, G.; Yang, Y. J. Am. Chem. Soc. 2012, 134, 10071.
(5) Lee, J.; Han, A. R.; Kim, J.; Kim, Y.; Oh, J. H.; Yang, C. J. Am. Chem.
Soc. 2012, 134, 20713.
(6) Kanimozhi, C.; Yaacobi-Gross, N.; Chou, K. W.; Amassian, A.;
Anthopoulos, T. D.; Patil, S. J. Am. Chem. Soc. 2012, 134, 16532.
(7) Tsao, H. N.; Cho, D. M.; Park, I.; Hansen, M. R.; Mavrinkskiy, A.;
Yoon, D. Y.; Graf, R.; Pisula, W.; Spiess, H. W.; Mullen, K. J. Am. Chem.
̈
Soc. 2011, 133, 2605.
(8) Kline, R. J.; Delongchamp, D. M.; Fischer, D. A.; Lin, E. K.; Richter,
L. J.; Chabinyc, M.; Toney, M. F.; Heeney, M.; McCulloch, I.
Macromolecules 2007, 40, 7960.
(9) Osaka, I.; Abe, T.; Shinamuru, S.; Miyazaki, E.; Takimiya, K. J. Am.
Chem. Soc. 2010, 132, 5000.
(10) Savage, R. C.; Orgiu, E.; Mativetsky, J. M.; Pisula, W.; Schnitzler,
T.; Eversloh, C. L.; Li, C.; Mullen, K.; Samori, P. Nanoscale 2012, 4,
̈
2387.
(11) Cabanetos, C.; El Labban, A.; Bartelt, J. A.; Douglas, J. D.;
Mateker, W. R.; Frechet, J. M. J.; McGehee, M. D.; Beaujuge, P. M. J. Am.
́
Chem. Soc. 2013, 135, 4656.
(12) Thompson, B. C.; Kim, B. J.; Kavulak, D. F.; Sivula, K.; Mauldin,
́
C.; Frechet, J. M. J. Macromolecules 2007, 40, 7425.
(13) Liang, Y.; Feng, D.; Wu, Y.; Tsai, S.-T.; Li, G.; Ray, C.; Yu, L. J. Am.
Chem. Soc. 2009, 131, 7792.
D
dx.doi.org/10.1021/ja406934j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX