The electronic, optical, and morphological properties of molecularly p-doped polyfuran (PF) films were investigated over a wide range of doping ratio in order to explore the impact of doping in photovoltaic applications. We find evidence for integer-charge transfer between PF and the prototypical molecular p-dopant tetrafluoro-tetracyanoquinodimethane (F4TCNQ) and employed the doped polymer in bilayer organic solar cells using fullerene as acceptor. The conductivity increase in the PF films at dopant loadings ≤2% significantly enhances the short-circuit current of photovoltaic devices. For higher doping ratios, however, F4TCNQ is found to precipitate at the heterojunction between the doped donor polymer and the fullerene acceptor. Ultraviolet photoelectron spectroscopy reveals that its presence acts beneficial to the energy-level alignment by doubling the open-circuit voltage of solar cells from 0.2 V to ca. 0.4 V, as compared to pristine PF.

For several decades, polymer based organic photovoltaic cells (OPVCs) have been intensively investigated due to their high potential for low-cost, simple, and large-scale processability, in particular, from solution.1–9 Key material parameters that impact both the short-circuit current density (JSC) and the open-circuit voltage (VOC) in OPVCs10 are the hole and electron mobilities of the donor and acceptor materials, respectively. On the one hand, too high mobilities can be detrimental to device performance, which is reflected in a decrease in Voc.11 On the other hand, Jsc is limited by the mobility of the slowest carriers, if the photocurrent is space-charge limited.12,13 Consequently, achieving balanced mobilities for both holes and electrons is a goal in ongoing research and a prerequisite for achieving high efficiency OPVCs. In particular, polymer/fullerene based OPVCs suffer from the low hole mobility in the polymer compared to the typically higher electron mobility in the fullerene promoting, e.g., parasitic charge recombination at the polymer/fullerene interface.14–17 In this context, molecular electrical p-doping,18 which is typically done by admixing strong electron acceptors like, e.g., 2,3,5,6-tetrafluoro-7,7,8,8-tetracyanoquinodimethane (F4TCNQ) to the organic semiconductor (OSC) host,19,20 emerged as valuable approach for enhancing the hole mobility (in the low doping concentration regime) as well as the conductivity through increased carrier concentration in the donor polymer. Concomitantly, electron/hole recombination via charge transfer excitons at the polymer/fullerene interface was reported to be significantly reduced upon doping the polymer host, which favors the formation of separated charges.17 In a recent study on the benchmark polymer poly(3-hexylthiophene) (P3HT) at doping ratios (given as percentage of F4TCNQ molecules per thiophene repeat unit) around 1%, Pingel and Neher reported both conductivity and mobility to increase superlinearly through molecular doping.21 Interestingly, however, while essentially all dopants undergo integer-charge transfer (ICT) with the polythiophene, only 5% contribute mobile holes for conduction. The rest of the hole/dopant-anion pairs remains bound after the ground state ICT,21 which rationalizes, to a certain extent, the high dopant concentrations employed in organic electronics, as compared to their inorganic counterparts. Recently, such F4TCNQ-doped P3HT layers have been successfully employed in OPVCs showing a doping-related increase in power conversion efficiency.22,23

While numerous fundamental and application-related studies exist on thiophene-based OSCs,24 the corresponding oligo- and polyfurans (PFs) have yet attracted much less attention.25 This is insofar surprising, as furan-based compounds are biodegradable and can be directly obtained from biomass rendering them ideal for low-cost, large-scale applications. In particular, oligofurans show promising optical properties while they exhibit better solution processability than their thiophene counterparts, and, most importantly, they comparably perform as active materials in, e.g., field-effect transistor applications.26,27

In this article, we explore the applicability of poly(4′,3″-dihexyl-2,2′:5′,2″:5″,2‴-quaterfuran) (PF) in OPVCs and the impact of molecularly p-type doping PF on the solar-cell performance. By investigating doping-induced changes in the optical absorption characteristics, the electronic properties, the conductivity, as well as the morphology of PF films with different F4TCNQ dopant loading, we aim at relating the microscopic processes underlying molecular polymer doping to device performance. In ultraviolet/visual/near-infrared (UV-VIS-NIR) absorption spectroscopy on the doped polymer, we observe evidence for integer-charge transfer between PF and F4TCNQ. Employed in bilayer heterojunction OPVCs with fullerene (C60) as acceptor, we find JSC to be significantly increased at low dopant concentrations (≤2%, given as percentage of F4TCNQ molecules per furan monomer unit). Unexpectedly, however, for (very) high doping ratios (up to 40%, i.e., in overdoped systems), VOC is significantly increased, while JSC was only slightly higher than in the undoped system. The increase in VOC is explained by an increased photovoltaic gap (Epvg), evidenced by ultraviolet photoelectron spectroscopy (UPS). Our results distinguish the two effects of doping, and highlight the beneficial impact of F4TCNQ presence on the energy-level alignment at the PF/C60 heterojunction in overdoped systems.

The common perception of molecular electrical p-doping assumes ICT between the OSC and the molecular acceptor. For molecularly p-doped P3HT, ICT was shown to occur in solution and in the solid state.21 For oligomer OSCs, however, evidence for the formation of ground-state charge transfer complexes (CPXs) was provided instead. CPXs were identified by the characteristic optical transitions between supramolecular hybrid states of the CPX and the simultaneous absence of spectral features characteristic for the respective OSC and dopant ions.28–30 Noteworthy, recently, evidence for the simultaneous occurrence of ICT and CPX formation in thiophene-based copolymers was observed.31 It was shown that the degree of electronic coupling between dopant and host is decisive for the doping mechanism (i.e., ICT vs. CPX formation) and significantly depends on the local microstructure.28–30 As PF is, in contrast to P3HT, planar in solution,32 an increased electronic coupling favoring CPX formation over ICT cannot be ruled out.

To that end, we carried out UV-VIS-NIR absorption spectroscopy on PF films increasingly doped with F4TCNQ; the results are depicted in Fig. 1, in comparison to that of pristine PF (blue curve) and F4TCNQ (red curve) reference samples. From the absorption onset of pure PF, its optical gap is determined to 2.30 eV, which is close to the theoretically predicted value of 2.38 eV.33 Upon doping, pronounced sub-bandgap absorptions arise in addition to the still visible pristine PF features, which clearly evidences ICT instead of CPX formation for F4TCNQ-doped PF. A comparison of the NIR region to published spectra of ionized F4TCNQ34,35 allows unambiguously assigning the prominent peaks at 1.43 eV and 1.62 eV to be due to F4TCNQ anions (vertical red lines in Fig. 1). Furthermore, in analogy to F4TCNQ doped P3HT,21 the broad absorption between 1.3 eV and 1.8 eV, as well as that below 1.1 eV are assigned to PF polarons. For higher doping concentrations (≥10%), however, the region above 3 eV shows enhanced absorption, which is explained by an increasingly present portion of neutral F4TCNQ pointing towards efficient ICT at low doping concentrations only.

FIG. 1.

UV-VIS-NIR spectra of pristine PF (blue), pristine F4TCNQ (red), and increasingly F4TCNQ doped PF films. Vertical red lines indicate the positions of F4TCNQ anion related absorptions; the chemical structures of a PF monomer unit and of F4TCNQ are shown in the inset. All spectra are normalized to their maximum optical density.

FIG. 1.

UV-VIS-NIR spectra of pristine PF (blue), pristine F4TCNQ (red), and increasingly F4TCNQ doped PF films. Vertical red lines indicate the positions of F4TCNQ anion related absorptions; the chemical structures of a PF monomer unit and of F4TCNQ are shown in the inset. All spectra are normalized to their maximum optical density.

Close modal

To gain insight into the film morphology and doping-related changes thereof, we carried out Atomic Force Microscopy (AFM) (Fig. 2). The topography of pristine and 1% F4TCNQ doped PF (Figs. 2(a) and 2(b)) appears equally smooth and homogenous, with an rms-roughness of 3 nm. In contrast, the 10% doped films, which already showed a pristine F4TCNQ portion in UV/VIS/NIR (shoulder around 3 eV in Fig. 1), exhibit a significantly rougher surface morphology (Fig. 2(c), rms: 13.6 nm), which we explain by phase separation between (doped) PF and neutral F4TCNQ forming 3D islands. This is more clearly seen in the highly overdoped system (40%), where an even rougher morphology comprising needle-like structures is observed, which we assign, again, to pure F4TCNQ (Fig. 2(d)), as this morphology is typical for solution processed F4TCNQ.36 

FIG. 2.

AFM micrographs (range: 5 × 5 μm) of spin coated undoped (a) and F4TCNQ doped PF films [(b) 1%, (c) 10%, and (d) 40% dopant ratio] on PEDOT:PSS coated ITO.

FIG. 2.

AFM micrographs (range: 5 × 5 μm) of spin coated undoped (a) and F4TCNQ doped PF films [(b) 1%, (c) 10%, and (d) 40% dopant ratio] on PEDOT:PSS coated ITO.

Close modal

Hence, above 10% doping level, F4TCNQ appears to have reached its solubility limit in PF for the solid phase. Pristine F4TCNQ then nucleates at the polymer surface, as two-component systems of a composition unstable to small concentration fluctuations (e.g., due to limited solubility in the respective solvent)37 tend to spontaneously phase separate.38–41 Consequently, F4TCNQ nucleation at the polymer surface is enhanced for higher dopant concentrations.

The conductivity of the doped PF films increases rapidly with the doping concentration at low ratio (Fig. 3(a)), which is in line with what is reported for F4TCNQ doped P3HT. There, the rise in conductivity was shown to be due to a doping induced increase in hole density, and, after initial trap filling,10–15,42,43 due to an increase in mobile hole mobility at dopant concentrations ≤1%.21,44 For PF beyond a doping ratio of 2%, this trend, however, reverses and the conductivity decreases, which we attribute to the morphological changes evidenced by AFM (Fig. 2), as areas of pristine F4TCNQ are expected to be detrimental to charge transport in the doped film; note that PF films are reported to be only weakly ordered, or even amorphous.45,46 Hence, a potential decrease in crystallinity upon doping cannot be held to fully explain the dramatic decrease in conductivity observed in the high doping regime.

FIG. 3.

(a) Conductivity of PF films as a function of F4TCNQ ratio; inset: schematic of the experimental setup employing intercalated ITO contacts. (b) Current density versus voltage curves for OPVCs upon doping.

FIG. 3.

(a) Conductivity of PF films as a function of F4TCNQ ratio; inset: schematic of the experimental setup employing intercalated ITO contacts. (b) Current density versus voltage curves for OPVCs upon doping.

Close modal

Bilayer heterojunction OPVCs were fabricated using C60 as top acceptor layer both for doped and undoped PF films, the corresponding I–V curves are depicted in Fig. 3(b). For OPVCs comprising pure PF, Jsc is low compared to today's benchmark polymer OPVCs.47 Importantly, however, only 1% of F4TCNQ doping increases Jsc by a factor of two due to the improved PF conductivity (cf. Fig. 3(a)), while Voc stays unchanged at 0.2 V. Higher dopant concentrations again lead to lower Jsc, only slightly higher than that of the pristine film, which agrees well with the trend in conductivity. Unexpectedly, however, Voc now increases up to the double of the value measured for the pristine film (0.39 V at 40% dopant loading).

To understand the underlying mechanism giving rise to this increase in Voc, the energy-level alignment at the respective heterojunctions was investigated by UPS to determine Epvg as upper limit for Voc. In this context, Epvg is defined as the difference in energy between the onsets of the valance band (VB) of the donor polymer and the lowest unoccupied molecular orbital (LUMO) level of the acceptor (e.g., fullerene) at the donor/acceptor interface.48,49 As illustrated by the schematic energy-level diagrams obtained from UPS experiments36 (Fig. 4), vacuum level (VL) alignment between PF and C60 is established for the undoped system (Fig. 4(a)). The PF VB-onset is separated from the Fermi level (EF) by only 0.40 eV, which points towards energy-level pinning of the PF VB to the substrate EF. The VB-onset of F4TCNQ doped PF is shifted by 0.20 eV to higher binding energy (BE), which suggests an increased density of occupied states (DOS) in the gap and energy disorder due to doping (Figs. 4(b)–4(d)). Consequently, the tail of the polymer VB is broadened and the pinning level is shifted further into the band gap.21,44 Note that the actual pinning level in the VB tail is typically not accessible by UPS due to its DOS being below the sensitivity of the technique in standard configuration.50–52 As obvious in Fig. 4(b), the VL rigidly shifts with the VB resulting in a work function (Φ) of the 1% doped PF film (4.20 eV) that is lower than the electron affinity of C60 (4.45 eV).53 Therefore, for the C60 overlayer, pinning of the LUMO at EF is observed with a concomitant upward VL shift.50 Most importantly, Epvg is essentially unaffected by the altered energy-level alignment upon doping, which explains the observation of identical Voc values for pristine PF and for 1% dopant ratio, as seen in Fig. 3 (cf. Table I). In marked contrast, Epvg is significantly increased for the high doping ratios of 10% and 40%, which is rationalized by the increase in Φ of the doped PF films pushing the LUMO of C60 away from EF (Figs. 4(c) and 4(d)). Note that, in contrast to the 1% case, the increase in Φ at high dopant concentrations can no longer be explained by ICT upon doping itself, but rather by the formation of a pristine F4TCNQ interlayer between the (overdoped) polymer film and C60: Because of the high electron affinity of F4TCNQ (5.08 eV),54 which exceeds Φ of the doped PF film (4.45–4.75 eV) in this regime, electrons are transferred to the pristine F4TCNQ film portion in order to establish electronic equilibrium. These negative charges therefore induce a local VL shift at the heterojunction with the C60 layer.55 Consequently, the higher the doping ratio in the overdoped regime, the more pristine F4TCNQ precipitates from the doped PF layer and (to a certain extent) nucleates thereon. Hence, this F4TCNQ interlayer increases Φ for subsequently deposited C60, as described above, and, therefore, increases Epvg. All values of Epvg and Voc are listed in Table I; note that Epvg and Voc do not fully increase in parallel, which indicates enhanced charge recombination for the highly doped OPVCs.

FIG. 4.

Schematic energy-level diagrams deduced from UPS for (a) the undoped PF/C60 and the F4TCNQ-doped PF/C60 heterostructures with (b) 1%, (c) 10%, and (d) 40% dopant concentration. The red areas between the PF and C60 layers symbolize the increasing amount of pristine F4TCNQ domains at the PF/C60 interface, a portion of which is charged due to the high dopant electron affinity inducing a local VL shift; HOMO denotes the highest occupied molecular orbital of C60, CB is the conduction band of PF.

FIG. 4.

Schematic energy-level diagrams deduced from UPS for (a) the undoped PF/C60 and the F4TCNQ-doped PF/C60 heterostructures with (b) 1%, (c) 10%, and (d) 40% dopant concentration. The red areas between the PF and C60 layers symbolize the increasing amount of pristine F4TCNQ domains at the PF/C60 interface, a portion of which is charged due to the high dopant electron affinity inducing a local VL shift; HOMO denotes the highest occupied molecular orbital of C60, CB is the conduction band of PF.

Close modal
TABLE I.

Epvg and Voc for PF films of different F4TCNQ doping ratios.

Doping ratioPristine PF1%10%40%
Epvg (eV) 0.50 0.60 0.80 0.90 
Voc (eV) 0.20 0.20 0.27 0.39 
Doping ratioPristine PF1%10%40%
Epvg (eV) 0.50 0.60 0.80 0.90 
Voc (eV) 0.20 0.20 0.27 0.39 

In summary, heterostructures of F4TCNQ-doped PF and C60 were employed in OPVCs comparing various doping concentrations. Integer charge transfer occurs upon molecular p-doping PF despite the planarity of the polymer backbone, which leads to tenfold increase in conductivity at moderate doping ratios between 1% and 2%. This process increases Jsc in the OPVCs while retaining a constant Epvg. For high doping ratios beyond 10%, however, Epvg is significantly increased, which translates into an increased Voc in such devices. The morphology of the highly doped PF films is found to be severely interrupted by pristine F4TCNQ domains reducing both conductivity and Jsc in the solar cell. The present study highlights the key role of microstructure and the precise knowledge thereof for understanding and improving organic electronic devices. In addition to the expected conductivity increase upon doping, our data demonstrate that molecular doping allows further deliberately increasing the photovoltaic gap in OPVCs due to a work function increase of the doped polymer films. Overall, our results show that molecular electrical doping not only favorably acts on conductivity but also enables tuning the energy-level alignment in OPVCs, thus paving the way for achieving enhanced functionality in organic electronic devices.

In memoriam of Professor M. Bendikov who passed away in 2013. We thank Paul Zybarth (HUB) for absorption measurements, Stefanie Winkler (HZB) for fruitful discussions, and gratefully acknowledge financial support from the Helmholtz Energy Allianz “Hybrid Photovoltaics” and the Sfb951 (DFG).

1.
C. W.
Tang
,
Appl. Phys. Lett.
48
,
183
(
1986
).
2.
C. W.
Tang
and
A. C.
Albrecht
,
J. Chem. Phys.
62
(
6
),
2139
(
1975
).
3.
C. M.
Ramsdale
,
J. A.
Barker
,
A. C.
Arias
,
J. D.
MacKenzie
,
R. H.
Friend
, and
N. C.
Greenham
,
J. Appl. Phys.
92
(
8
),
4266
(
2002
).
4.
L. J. A.
Koster
,
V. D.
Mihailetchi
,
R.
Ramaker
, and
P. W. M.
Blom
,
Appl. Phys. Lett.
86
(
12
),
123509
(
2005
).
5.
H. J.
Snaith
,
L.
Schmidt-Mende
,
M.
Gratzel
, and
M.
Chiesa
,
Phys. Rev. B
74
(
4
),
045306
(
2006
).
6.
A.
Geiser
,
B.
Fan
,
H.
Benmansour
,
F.
Castro
,
J.
Heier
,
B.
Keller
,
K.
Emanuel Mayerhofer
,
F.
Nüesch
, and
R.
Hany
,
Sol. Energy Mater. Sol. Cells
92
(
4
),
464
(
2008
).
7.
R. R.
Søndergaard
,
M.
Hösel
, and
F. C.
Krebs
,
J. Polym. Sci., Part B: Polym. Phys.
51
(
1
),
16
(
2013
).
8.
J.
You
,
C. C.
Chen
,
Z.
Hong
,
K.
Yoshimura
,
K.
Ohya
,
R.
Xu
,
S.
Ye
,
J.
Gao
,
G.
Li
, and
Y.
Yang
,
Adv. Mater.
25
(
29
),
3973
(
2013
).
9.
J.
You
,
L.
Dou
,
K.
Yoshimura
,
T.
Kato
,
K.
Ohya
,
T.
Moriarty
,
K.
Emery
,
C. C.
Chen
,
J.
Gao
,
G.
Li
, and
Y.
Yang
,
Nat. Commun.
4
,
1446
(
2013
).
10.
G.
Dennler
,
M. C.
Scharber
, and
C. J.
Brabec
,
Adv. Mater.
21
(
13
),
1323
(
2009
).
11.
Y. X.
Wang
,
S. R.
Tseng
,
H. F.
Meng
,
K. C.
Lee
,
C. H.
Liu
, and
S. F.
Horng
,
Appl. Phys. Lett.
93
(
13
),
133501
(
2008
).
12.
J. T.
Shieh
,
C. H.
Liu
,
H. F.
Meng
,
S. R.
Tseng
,
Y. C.
Chao
, and
S. F.
Horng
,
J. Appl. Phys.
107
(
8
),
084503
(
2010
).
13.
A. M.
Goodman
,
J. Appl. Phys.
42
(
7
),
2823
(
1971
).
14.
A. V.
Tunc
,
A.
De Sio
,
D.
Riedel
,
F.
Deschler
,
E.
Da Como
,
J.
Parisi
, and
E.
von Hauff
,
Org. Electron.
13
(
2
),
290
(
2012
).
15.
C.
Melzer
,
E. J.
Koop
,
V. D.
Mihailetchi
, and
P. W. M.
Blom
,
Adv. Funct. Mater.
14
(
9
),
865
(
2004
).
16.
L.
Jan Anton Koster
,
V. D.
Mihailetchi
,
M.
Lenes
, and
P. W. M.
Blom
, in
Organic Photovoltaics
(
Wiley-VCH Verlag Gmbh & Co. KGaA
,
Weinheim, Germany
,
2009
).
17.
F.
Deschler
,
E.
Da Como
,
T.
Limmer
,
R.
Tautz
,
T.
Godde
,
M.
Bayer
,
E.
von Hauff
,
S.
Yilmaz
,
S.
Allard
,
U.
Scherf
, and
J.
Feldmann
,
Phys. Rev. Lett.
107
(
12
),
127402
(
2011
).
18.
V. A.
Trukhanov
,
V. V.
Bruevich
, and
D. Y.
Paraschuk
,
Phys. Rev. B
84
(
20
),
205318
(
2011
).
19.
P.
Pingel
,
R.
Schwarzl
, and
D.
Neher
,
Appl. Phys. Lett.
100
(
14
),
143303
(
2012
).
20.
K. H.
Yim
,
G. L.
Whiting
,
C. E.
Murphy
,
J. J. M.
Halls
,
J. H.
Burroughes
,
R. H.
Friend
, and
J. S.
Kim
,
Adv. Mater.
20
(
17
),
3319
(
2008
).
21.
P.
Pingel
and
D.
Neher
,
Phys. Rev. B
87
(
11
),
115209
(
2013
).
22.
X. F.
Lei
,
F. T.
Zhang
,
T.
Song
, and
B. Q.
Sun
,
Appl. Phys. Lett.
99
(
23
),
233305
(
2011
).
23.
X. Y.
Han
,
Z. W.
Wu
, and
B. Q.
Sun
,
Org. Electron.
14
(
4
),
1116
(
2013
).
24.
D. F.
Perepichka
and
I. F.
Perepichka
,
Handbook of Thiophene-Based Materials
(
John Wiley & Sons Ltd.
,
Chichester, United Kingdom
,
2009
).
25.
U. H. F.
Bunz
,
Angew. Chem. Int. Ed.
49
(
30
),
5037
(
2010
).
26.
O.
Gidron
,
A.
Dadvand
,
Y.
Sheynin
,
M.
Bendikov
, and
D. F.
Perepichka
,
Chem. Commun.
47
(
7
),
1976
(
2011
).
27.
O.
Gidron
,
A.
Dadvand
,
E. W. H.
Sun
,
I.
Chung
,
L. J. W.
Shimon
,
M.
Bendikov
, and
D. F.
Perepichka
,
J. Mater. Chem. C
1
(
28
),
4358
(
2013
).
28.
H.
Méndez
,
G.
Heimel
,
A.
Opitz
,
K.
Sauer
,
P.
Barkowski
,
M.
Oehzelt
,
J.
Soeda
,
T.
Okamoto
,
J.
Takeya
,
J.-B.
Arlin
,
J.-Y.
Balandier
,
Y.
Geerts
,
N.
Koch
, and
I.
Salzmann
,
Angew. Chem. Int. Ed.
52
(
30
),
7751
(
2013
).
29.
I.
Salzmann
,
G.
Heimel
,
S.
Duhm
,
M.
Oehzelt
,
P.
Pingel
,
B. M.
George
,
A.
Schnegg
,
K.
Lips
,
R. P.
Blum
,
A.
Vollmer
, and
N.
Koch
,
Phys. Rev. Lett.
108
(
3
),
035502
(
2012
).
30.
G.
Heimel
,
I.
Salzmann
, and
N.
Koch
,
AIP Conf. Proc.
1456
(
1
),
148
(
2012
).
31.
F.
Ghani
,
A.
Opitz
,
P.
Pingel
,
G.
Heimel
,
I.
Salzmann
,
J.
Frisch
,
D.
Neher
,
A.
Tsami
,
U.
Scherf
, and
N.
Koch
,
J. Polym. Sci., Part B: Polym. Phys.
53
(
1
),
58
(
2015
).
32.
J. K.
Politis
,
J. C.
Nemes
, and
M. D.
Curtis
,
J. Am. Chem. Soc.
123
(
11
),
2537
(
2001
).
33.
S. S.
Zade
and
M.
Bendikov
,
Org. Lett.
8
(
23
),
5243
(
2006
).
34.
J. B.
Torrance
,
J. J.
Mayerle
,
K.
Bechgaard
,
B. D.
Silverman
, and
Y.
Tomkiewicz
,
Phys. Rev. B
22
(
10
),
4960
(
1980
).
35.
D. A.
Dixon
,
J. C.
Calabrese
, and
J. S.
Miller
,
J. Chem. Phys.
93
(
6
),
2284
(
1989
).
36.
See supplementary material at http://dx.doi.org/10.1063/1.4921484 for Experimental Details, AFM on pure F4TCNQ, and UPS spectra.
37.
S.
Nilsson
,
A.
Bernasik
,
A.
Budkowski
, and
E.
Moons
,
Macromolecules
40
(
23
),
8291
(
2007
).
38.
N. P.
Balsara
,
C.
Lin
, and
B.
Hammouda
,
Phys. Rev. Lett.
77
(
18
),
3847
(
1996
).
39.
J. W.
Cahn
and
J. E.
Hilliard
,
J. Chem. Phys.
28
(
2
),
258
(
1958
).
40.
M.
Geoghegan
and
G.
Krausch
,
Prog. Polym. Sci.
28
(
2
),
261
(
2003
).
41.
J.
Smith
,
R.
Hamilton
,
I.
McCulloch
,
N.
Stingelin-Stutzmann
,
M.
Heeney
,
D. D. C.
Bradley
, and
T. D.
Anthopoulos
,
J. Mater. Chem.
20
(
13
),
2562
(
2010
).
42.
Y.
Zhang
,
B.
de Boer
, and
P. W. M.
Blom
,
Adv. Funct. Mater.
19
(
12
),
1901
(
2009
).
43.
W.
Gao
and
A.
Kahn
,
J. Appl. Phys.
94
(
1
),
359
(
2003
).
44.
V. I.
Arkhipov
,
P.
Heremans
,
E. V.
Emelianova
, and
H.
Bassler
,
Phys. Rev. B
71
(
4
),
045214
(
2005
).
45.
A.
Gok
,
H. K.
Can
,
B.
Sari
, and
M.
Talu
,
Mater. Lett.
59
(
1
),
80
(
2005
).
46.
S.
Sen
,
B.
Bardakci
,
A. G.
Yavuz
, and
A. U.
Gok
,
Eur. Polym. J.
44
(
8
),
2708
(
2008
).
47.
48.
A.
Wilke
,
J.
Endres
,
U.
Hormann
,
J.
Niederhausen
,
R.
Schlesinger
,
J.
Frisch
,
P.
Amsalem
,
J.
Wagner
,
M.
Gruber
,
A.
Opitz
,
A.
Vollmer
,
W.
Brutting
,
A.
Kahn
, and
N.
Koch
,
Appl. Phys. Lett.
101
(
23
),
233301
(
2012
).
49.
K.
Vandewal
,
K.
Tvingstedt
,
A.
Gadisa
,
O.
Inganas
, and
J. V.
Manca
,
Nat. Mater.
8
(
11
),
904
(
2009
).
50.
J.
Frisch
,
A.
Vollmer
,
J. P.
Rabe
, and
N.
Koch
,
Org. Electron.
12
(
6
),
916
(
2011
).
51.
J.
Hwang
,
A.
Wan
, and
A.
Kahn
,
Mater. Sci. Eng., R
64
(
1-2
),
1
(
2009
).
52.
T.
Sueyoshi
,
H.
Fukagawa
,
M.
Ono
,
S.
Kera
, and
N.
Ueno
,
Appl. Phys. Lett.
95
(
18
),
183303
(
2009
).
53.
K.
Akaike
,
K.
Kanai
,
H.
Yoshida
,
J.
Tsutsumi
,
T.
Nishi
,
N.
Sato
,
Y.
Ouchi
, and
K.
Seki
,
J. Appl. Phys.
104
(
2
),
023710
(
2008
).
54.
J.
Niederhausen
,
P.
Amsalem
,
A.
Wilke
,
R.
Schlesinger
,
S.
Winkler
,
A.
Vollmer
,
J. P.
Rabe
, and
N.
Koch
,
Phys. Rev. B
86
(
8
),
081411(R)
(
2012
).
55.
A.
Kahn
,
N.
Koch
, and
W. Y.
Gao
,
J. Polym. Sci., Part B: Polym. Phys.
41
(
21
),
2529
(
2003
).

Supplementary Material